6756 4060

Durham E-Theses
Fire and light in the western triduum: their use at
tenebrae and at the paschal vigil
MacGregor, Alistair J
How to cite:
MacGregor, Alistair J (1989) Fire and light in the western triduum: their use at tenebrae and at the
paschal vigil, Durham theses, Durham University. Available at Durham E-Theses Online:
http://etheses.dur.ac.uk/6756/
Use policy
The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or
charge, for personal research or study, educational, or not-for-prot purposes provided that:
• a full bibliographic reference is made to the original source
• a link is made to the metadata record in Durham E-Theses
• the full-text is not changed in any way
The full-text must not be sold in any format or medium without the formal permission of the copyright holders.
Please consult the full Durham E-Theses policy for further details.
Academic Support Oce, Durham University, University Oce, Old Elvet, Durham DH1 3HP
e-mail: e-theses.admin@dur.ac.uk Tel: +44 0191 334 6107
http://etheses.dur.ac.uk
2
Alistair J. MacGregor
'Fire and Light in the Western Triduum : Their Use at Tenebrae
and at the Paschal Vigil'
Abstract of thesis submitted to the University of Durham for the
Degree of Ph.D.
October 1989
The stage-by-stage development of Tenebrae is described showing the

extension of light-loss at LaudF on uood Friday to the three night
offices of the Western Triduum. The emergence, deveJopment, and use
of the hearse at Tenebrae from the eleventh century onwards is explored,
together with the integration of that device into the liturgical-drama
that the service of Tenebrae represented. The varying number of lights
used and the extinction-points are shown to be derived from differing
liturgical traditions. The presence of other lights at the service is
discussed; and the extinguishing of lights is shown to have a
rememorative, not a utilitarian origin.
The new fire ceremonies of all the Western rites, which were of
Gallican origin, were deliberately adopted by the Church as part of her
.
missionary work. An in-depth survey of the ritual surrounding the
kindling of the fire and the subsequent procession with the fire into
church reveals a heritage of different cultural and liturgical traditions.
Not only was the threefold production of fire linked to the triple per-
formance of Tenebrae; the new fire ceremony was integrated into the
Paschal vigil liturgy because of the common theme of light; and to the
former was extended the Passover motif.
Not only are the geographical and liturgical origins of the Easter
candle considered; an historical analysis is presented of both the
Candle itself and of the ceremonial surrounding the blessing of the
Candle. This ceremonial, being largely of Gallican provenance, is ex-
amined in relation to the corresponding Milanese, Mozarabic, and Roman
Vigil liturgies, all of which are related to the Lucernarium of
Jerusalem. The study shows that the late medieval Paschal ceremony of
light was a synthesis of Roman and Gallican elements; and that a two-
fold tradition existed relating to the provision of li~ht at the Vigil.
FIRE and LIGHT
in the
WESTERN TRIDUUM
* * * * *
Their Use at Tenebrae and at the Paschal Vigil
A. J. MacGregor
The copyright of this thesis rests with the author.

No quotation from it should be published without
his prior written consent and information derived
from it should be acknowledged.
Thesis submitted to the University of

Durham for the Degree of Ph.D.
1989 ·_,·_,'_ ·~: . I,:.·.-

j-_·_-l:••·.'·_·
... .,,."""'
1~;j(*
I had originally intended that the present work should form the con=
eluding chapters of a survey tracing the use of light in Christian wor=
ship back to its origins in Judaism. It became clear, however, that
these chapters could not be adequately accommodated within the frame=
work of this treatise in view of the history and development over more
than a thousand years of the numerous services of the Triduum which
involve the use of light in both the Western and Eastern Churches.
Moreover, in order to do justice even to one aspect of the subject of
light at Eastertide, not only was it necessary for me to confine my-
self to the ceremonies of the western tradition; the limits imposed
upon a work such as this demanded that a study of some of the services
of the latter part of Holy Week, viz. the Commemoration of the Institu~
tion of the Lord's Supper, the Pedilavium, and the Good Friday Passion,
should be omitted; whilst the use of light at Baptism has been treated
somewhat cursorily in being reduced to an appendix.
There is no monograph in existence devoted specifically to a study

of light in the services of the Triduum. Prior to the 1950s most of the
information about this subject was confined to entries in encyclopedias
and handbooks on liturgy, with occasional references in histories of
liturgy. Following the 1951 liturgical experiments within the Roman
Catholic Church, which were an attempt to restore to their former pro-
minence the ceremonies of the Triduum, and the 1955 Decree of Maxima
Redemptionis, which officially permitted the performance of the revived
services, there appeared an abundance of literature treating of the re-
storation of the Paschal rites. Most of this literature, however, is
concerned primarily with the pastoral and theological aspects of the
Easter liturgy, and adds little to our knowledge of the history of the
Paschal ceremonies. The fullest treatment of this subject is the chapter
entitled ~ Ceremonies 2f ~ ~ ~n D.R.Dendy's The ~ ~ Lights ~
Christian Worship. Being a single chapter in a larger work, the subject
is perhaps inevitably treated rather cursorily; and the material is
presented somewhat indiscriminately.
The present work is written primarily from an historical point of

view. It identifies the cultural and liturgical milieux from which the
ceremonies of Tenebrae, the new fire, and the Easter candle emerged;
and in describing these three light-featuring rituals of the Triduum,
it traces their historical development, incorporating the theological
iii
significance and pastoral aspects of each 9 and relating them to the
liturgical changes of the 1950s and the reasons for those changeoo
This study alBo shows the close connection that existed between the
triple performance of Tenebrae and the threefold production of the new
fire, and examines the reasons for the adoption of the new fire cere~
mo~zy into the lit.urgy of the Church and its incorporation into the
Paschal vigil.
The area covered by the research includee th0 whole territorial ex=
tent of the medieval Latin West. Documentary evidence is plentiful for
England, Germ&Qy, and especially France, where the existence of a large
number of neo-Gallican missals reveals the survival of variant Vigil~
ceremonials from the early Middle Ages. In other parts of Europe the
domination of the Roman rite, as in Spain, or the imposition of a single
rite, as in Norway, or the competitive influence of the Orthodox Church,
inevitably limited the number of ~ocal rites, and so curtails our know-
ledge about non-Roman local variant ceremonies.
The term Gallican has been used with two different meanings. In
the period up to £•1000 it refers to the non-Roman rites of France and
Western Germany. It also describes the movement in France in the
eighteenth and nineteenth centuries for liturgical independence from
Rome, and the diocesan rites associated with that movement. Throughout
the present work'Vigil'indicates the Paschal vigil of Holy Saturday;
and(Candle'signifies the Easter candle.
Northwich, A. J. MacGregor
Cheshire.
October 1989
iv
INTRODUCTION
There is ~o firm ovidanco boforo the fourth century for the uao of
either liturgical or functional light at the Paschal vigil. However~
in view of the close association of Baptism with the Vigil and the
equating of Christian initiation with illumination, the centrality in
Christianity of Christ the Light of the World, as well as the importance
of the concept of light in the mystery religions 9 which may well have
influenced both Christian theology and liturgical practice 9 it was
almost inevitable that the ceremonies marking the climax of the Church 0 s
year and the annual commemoration of Christ's victory over the darkness
of this world should be held in an ambience of abundant light. The
1 2 2
fourth-century evidence from Constantinople, Jerusalem, and Spain
would suggest that the holding of the Paschal vigil in a milieu of light
was by then widespread throughout the Christian world.
However, by the end of the tenth century in the West and by the
beginning of the twelfth century in Jerusalem, 3 an alternative tradition
of holding the Vigil in semi-darkness emerged from a monastic milieu.
Both these traditions have been treated in Chapter 15 of Part IV.
The blessing of the Easter candle, the kindling of new fire, and the
gradual extinguishing of the lights at the three night offices of the
Triduum, later to be known as Tenebrae, were ceremonies peculiar to and
confined to the West. The first of these rituals formed the central
feature of a transformed Lucernnrium, the ancient office of the lighting
of the evening lamp, which itself was incorporated into the Paschal lit-
urgy. The production of Paschal fire was not unknown in the East. By
the ninth century the ceremony in Jerusalem was accompanied by scenes of
frenzied enthusiasm which have characterised the occasion ever since.
By contrast, the corresponding ceremonies of the western rites have en-
tailed the production of newly-kindled fire in circumstances less shroud-
ed in secrecy and mystery than those which obtain at Jerusalem. However.
the taking home of the new fire would have been no less meaningful to
the faithful of France, for example, than to the pilgrims in Jerusalem.
The antecedents of the Easter fire ceremonies of Jerusalem and Northern
1 'At Easter ••• Constantine changed the holy all-night vigil into the
light of day by arranging for wax candles of very great length to
be lit throughout the whole of the city.' Eusebius, ~ Constapt-
ini 4.22 (PG 20.1169).
2 Wilkinson p.1}8. See also Part IV Chapter 2 p.230.
3 As attested in Hagios Stauros ~· See Stevenson p.182.
v
Europe were differento For whilst the ritual at Jerusalem was intimate=
ly linked with the office of th0 Lucernarium, the new fire ceremonies of
Northern Europe existed as pre=Christian raligiou6 rites 9 which were
taken over and incorporated into the liturgy of the Church, in the same
way that the Easter egg and the Christmas tree have been accommodated
within the traditions of Christianityo
Tho officG of }latins/Lauds, known as Tenebrae~ emerged from a mon=

astic liturgical milieu, and, as we have argued in Part ! 9 developed as
a dramatic and liturgical representation of an historical event recorded
in the Gospels.
Unlike the ritual surrounding the Easter candle, which we have argued
had its provenance in Northern Italy, the new fire ceremony and the serv-
ice of Tenebrae emerged from the liturgically-active region of Northern
France and Western Germany according to the earliest surviving document-
ary evidence. In spite of the known influence of Milan in the region to
the north of the Alps, the absence of these two ceremonies from the
earlier Ambrosian rite, and indeed from the Roman rite, would suggest
that both rituals were indigenous Gallican liturgical developments.
The holding of the Vigil in anticipation was already established by

the eighth century, probably as a result of changing pastoral needs and
circumstances. For in earlier centuries the sacrament of baptism which
followed the blessing of the Easter candle and reading of the prophecies,
was administered primarily to adults. By the eighth century in regions
in which Christianity had been established for several centuries, the
perhaps inevitable fall in the number of adults who sought Baptism, the
incipient practice of administering this sacrament at other times during
the year, and the fact that what had become the traditional time for ad-
ministering Baptism, viz. a late hour on Holy Saturday or an early hour
on Easter Day, was hardly the most suitable for unweaned infants, result-
ed in the bringing forward of the time at which the Paschal vigil was
held. The anticipation of Tenebrae, though unaffected by these consid-
erations, came to pass, it is believed, through a desire within those
churches and cathedrals which were served by non-monastic clergy, to
sing the night office at a more convenient time.
By the end of the fifteenth century it had become the practice in

some parts of Western Europe to kindle the nwe fire at about 9.00 a.m.
on Holy Saturday, and to hold the Paschal vigil on the morning of that
vi
@y. TkAe ebo!ition by Popo UrbaEJl VIII of Holy Satlll"'day aG a public
holiday in Catholic countri0s colltributocl to the further decay of the
Tridu~; so that until. tho middl0 of tho present c~ntury it tlae tlide=
spread practice throughout tlastern Europe to hold the Vigil in tho
morning light of Holy Saturday.
In 1955 the ~fforta of the Liturgical Movement over a period of

fifty years bore fruit tlhOll 1 fo1lotl~ tho Decree of ~~J~iou Ro~omptiollie 9
it tlaB gfficially pormittod to hold tho PaGchol vigil at an hour tlhich tl&a
liturgically and commemor~tively ~ore meaningful ~d tlhich appro~imatod
to the time ai which the Vigil had been held in the early Middle Ages.
~ovioualy having burned in tho daylight hours of Holy Saturday 9 the
Easter candle reacquired its former ~portanco and significance not o~y
for the members of the cl.ergy 9 but for the laity also 9 who in previous
centuries ~cl not attended the Vigil either thro~h ignorance or
apathy or the demands of work; but who were notl once more able to
participate in the preparation for the ~~sa of Easter and to share
meaningfully in the Light of Christ.
vii
A Note on the Sourcea
~~~-
The principal sources for the period c.AD 700=900 are the ordines
Romani edited by M.Andrieu. The writer has generally accepted his
dating of them. Throughout the present work each of these has been
referred to simply as'Ordo' followed by an arabic numeral. They are to
be distinguished from the later ordines Momani, first published by
Mabillon and subsequently by J.P.Migne in Volume 78 of Patrologia
Latina. The latter group of documents are each indicated by 'Ordo 0
followed by a roman numeral. The writer accepts the view that the
ninth-century Pontifical of Boitiers relates to the Church in Poitiers,
in spite of the doubts of A.-G. Martimort and A. Martini. Neither
Mabillon nor Martene questioned the provenance of the pontifical. Both
may have had access to corroborating information which has since been
lost. Information relating to Gembloux is contained in the Customary
of Sigibert (Albers, Vol.2).
In 1984 the writer undertook a survey of the Paschal vigil as held

in the cathedrals of France at the present time. Information gained
from the 60% of the replies received is indicated by 'Survey of 1984•.
It had been intended to examine a number of other documents, mainly

French diocesan missals. Long after he had embarked upon the present
work, the writer discovered that these had been destroyed during the
Second World War. However, sufficient documents survived the bombing
so that the writer was able to proceed with his original plan.
Apart from the primary sources themselves, perhaps the most import-
ant quarry of information for the student of liturgiology is Edmond
Mart~ne's monumental E! Antiquis Ecclesiae Ritibus, originally published
between 1700 and 1702. This contains extracts from a very large number
of liturgical texts which he consulted in the monastic libraries of
France. Some of the manuscripts have subsequently disappeared.
A.-G. Martimort has demonstrated that the liturgical material recorded
by Mart~ne is not only reliable but is invaluable for the liturgiologist.
It therefore ranks on a par with other primary documentary evidence.
'Blackfriars Correspondence' relates to communications received from

Revd Fr R. Conrad of Blackfriars, Oxford.
viii
------------------------------------
Acknowledgements
The writer wishes to express his gratitude to the following persons

for their assistance in the writing of this thesis : to the Reverend
Dr A. Gelston, the writer's academic mentori for his highly-valued
advice and guidance, and for his continuous 0ncouragemont throughout
the years; to the Revd Dr DeG. Dragas of the Uepartment of Theology 9
University of Durham; to Dr J. Rhodes; to the Revd Dr M. Sharratt
and Revd Fr N. Paxton, librarians of St Cuthbert's College, Ushaw and
St Joseph's College, Upholland, respectively; to the Revd T.P. and
Dr Jo Wiersum; to the Revd Fr R. Conrad OP of Blackfriars, Oxford; to
the Revd H. Taylor; to the Revd Fr A. Griffiths of the Southampton
University Catholic Chaplaincy; to the Revd Dr K.W. Stevenson, formerly
of the Manchester University Anglican Chaplaincy; to the AbbeP. Grey,
Secretaire General of the Diocese of Amiens; and to a number of heads
of religious houses, and others, both lay and clergy, whose replies to
the writer's letters have been greatly appreciated.
Although the entire text of the thesis has been typed by the
writer, he wishes to thank profoundly the members of his family for
their much-appreciated assistance with the production and printing
of this work.
References
A reference in the text to a published work appears in a footnote

under the author's name. Where two or more of an author's works
have been consulted during the writing of this thesis, an abbreviated
form of the title of the relevant work follows the author's name.
Where an author has been referred to by name within the text, an
abbreviated form of the title of the relevant work only appears in
the footnote.
ABBREVIATIONS
AC Augustinian Canons ~ ~ Breviarii

AV Authorised Version .Q!:1 2!:!!2. Missalis
BCP Book of Common Prayer QB Q!:gg Romanus
c Century PJSM Pope John Sinday Missal
Cerem., Ceremonial PL Migne'e Patrologia Latina
ET English Translation PG Migne's Patrologia Graeca
GF Good Friday Ponto pontifical
HS Holy Saturday PV pars verna
l\1R f-ti.ocale Romo.num Rep. Repertorium
MT r-1aund.y Thursday §!!!·lli Semaine Sainte
nd no dato t.e. typical edition
np no pagination tpm title page missing
Alcuin Liber de divinis officiis, formerly attributed to Alcuin.

Bindo F Das Zeremonienbuch des Bindo Fesulani.
CA 1706 Die Sammlung !!!§ Codex Atigiion 1706.
~.Episc. Caeremoniale Episcoporum.
QMg Consuetudines Monasteriorum Germaniae.
~ Decreta Authentica Congregationis Sacrorum Rituum.
~ ~ Antiguis Ecclesiae Ritibus, Vo1.3.
DAMR ~ Antiguis Monachorum Ritibus, being Vol.4 of De
Antiquis Ecclesiae Ritibus.
Decrees Lanfranc's Decreta f£2 Ordine Sancti Benedicti.
!2! Pi!·
QU. ~ Divinis Officiis.
DHCR Dictionnaire historique des cultes religieux.
DMC ~ ~ Conditor.
~-~· Ephemerides Liturgicae.

Gav./Mer. Gavantus/Merati, being the Thesaurus Sacrorum Rituum of
B.Gavanti, revised by C.M.Merati.
The Gelasian Sacramentary of Angouleme.
The Gelasian Sacramentary of Gellone.
The Gelaaian Sacramentary of Autun (Phillipps' Gelasian).
The Gelasian Sacramentary of Prague.
The Gelasian Sacramentary (Gelasianum Vetus).
Henry Bradshaw Society Publications.
International Commission on English in the Liturgy.
Institutions Liturgiques 2! l'Egliae 22 Marseille.
Journal of Theological Studies.
Liber Censuum.
~ 4! Officiis Eccleaiis.
~. ~ ~ Ordine Antiphonarii.
4!, Q.£.g_. ~. :
Liturgies 2!' ~ :em.
Le Liber Politicua de Benoit.
Liturgies~ th§ Priiiiatial ~.
~Liturgy 2£ ~ Roman Church.
Liturgies 2! t~ Religious Orders.
Followed by a figure, it refers to a document, cited by
E.Martene and classified by A.-G.Martimort in ~ Document-
~ Liturgigue ~ ~ Edmond Martene.
Miss.
Gall.~. Missale Gallicanum ~·
(continued on page xi. )
X
ABBREVIATIONS (continued)
MVLA Monumenta Veteris Liturgiae Ambro~ianae.

Nom.Cist. Nomasticon Cisterciense.
OED Oxford English Dictionary.
Q!; Q£ge Officiorum Ecclesiae Lateranensis.
OLD Oxford Latin Dictionary.
-
ONE
OPA
Ordo Nidrosiensis Ecclesiae.
Ordo of Pierre Amiel.
~ Office ~ la Semaine &ainte.
PGD Le Pontifical S! Guillaume Durand.
Poi tiers Pontifical of Poitiers. ,
PR XII~ Le Pontifical Romain ga XIIe siecle •
PRC t8 POntifical 4! Ia Curie Romaine !!! ~ siecle·
PRG Pontificale Romano-Germanicum.
Reg.~. Regularis Concordia.
~ Ritus Ecclesiae Laudunenais Redivivi.
~ §ources of t.h! Modern~ LiturgY (S.J.P.Van Dijk).
Statutes Statutes of Lincoln Cathedral (Bradshaw and Wordsworth).
~ Die Zerem~enbucher der Romischen Kurie in Mittelalter
(Schimmelpfennig). - ~ ~
~.Vet. Sacramentarium Vetus
• • • • •
The Numeration of the Psalms
Throughout the present work the numbering of the psalms is that found
in the Greek Septuagint. For ease of reference both the Greek Septua-
gint and the Hebrew systems are given below.
Greek Septuagint Hebrew

1-8 1-8
9 9-10
10-112 11-113
113 114-115
114-115 116
116-145 117-146
146-147 147
148-150 148-150
xi
CONTENTS
PART I T.lilNEBRAE
Chapter
1e MATINS 1
2 THE DEVELOPMENT OF TENEBRAE
9 8
3. THE TIME OF TENEBRAE 20
4" THE PROVISION OF LIGHT AT TENEBRAE 22
(i) The Lighting of the Lamps and Candles 22
(ii) The Disposition of Lights 24
5. THE NUMBER OF LIGHTS AT TENEBRAE 34
(i) The Early Period 34
(ii) The later diversity of numbers 35
6. THE TENEBRAE HEARSE 46
(i) The Origin of the Hearse 46
(ii) The Development of the Hearse 48
(iii) The Hearse of Twenty-four Lights 50
(iv) The Origin of the Fifteen-candle Hearse 52
(v) The Significance of Fifteen Candles 57
(vi) The Position of the Hearse 58
(vii) The Construction of the Hearse 60
(viii) The Name hearse 62
7. THE EXTINGUISHING OF LIGHTS 64
(i) The order in which the lights were put out 64
(ii) The Extinction-points 68
(iii) The Candles 83
(iv) The Light before the Blessed Sacrament 83
8. THE CONCLUSION OF TENEBRAE 85
(i) The Tradition of the Benedictus without light 85
(ii) The Tradition of the Benedictus with light 86
(iii) The Last Candle 87
( i v) The Last Candle : Symbolism and Size 89
9. THE NAME AND ORIGIN OF TENEBRAE 91
(i) The Name Tenebrae 91
(ii) The Origin of Tenebrae 91
PARI' II : THE NEW FIRE CEREMONY
1 • THE DEVELOPMENT OF TID; CEREMONY 99
( i) The older Roman tradition 99
(ii) The Gallican tradition 101
(iii) The Spanish and Italian traditions 105
2. THE TRIPLE AND THE SINGLE PRODUCTION OF FIRE 1o8
(i) The triple production of fire 1o8
(ii) The single production of fire 112
3. THE BLESSING OF THE NEW FIRE 114
(i) Preliminary procession : psalms, litany, and reading 114
(ii) The formulas 116
(iii) Aspersion and incensation 127
(iv) The officiant 128
4. THE PRODUCTION OF NEW FIRE. (1) TIME 131
(i) Within the tradition of the threefold kindling 131
(ii) Within the tradition of the single kindling 135
xii
Chapter
5. THE PRODUCTION OF NEW FIRE. (2) LOCATION 1Lto
(i) 1~e early period 140
(ii) The later Middle Ages 145
6. THE PRODUCTION OF NEW FIRE. (3) ~lEANS 153
(i) Fire by friction 153
(ii) Fire by refraction of the sun's rays 161
7. THE NEvi FIRE AT ROME 170
(i) The eighth-century evidence of Pope Zachary 170
(ii) The ninth-century evidence 177
(iii) The use of a lens in the Homan rite 180
8. THE ANTECEDENTS OF THE NEtt/ FIRE CEID.:MONY 183
PART III : THE NEW FIRE PROCESSION
1. THE NEW FIRE AND THE PROCESSION 189

2. THE ARUNDO SERPENT INA 195
(i) Its description and use 195
(ii) The origin of the arundo serpentina 196
3. THE TRIPLE CANDLE 201
(i) Description and-construction 201
( ii) The origin of the triple candle 203
(iii) The lighting of the triple candle 206
(iv) The disposition and disposal of the processional candles 207
4. THE PSALMS, THE INVENTOR RUTILI, AND THE LUMEN CHRISTI 2o8
(i) The Psalms - 208
(ii) The Inventor rutili 209
(iii) The cry of ~ Christi 210
5. THE EASTER CANDLE AND THE PROCESSION 219
(i) The bearing of the Easter candle 1.Lit 219
(ii) The bearing of the Easter candle 2.Unlit 220
(iii) The evidence of PRG 220
PARr IV : THE EASTER CANDLE AND VIGIL
1. THE ORIGIN OF THE EASTER CANDLE 224

2. THE EASTER CANDLE : EARLIEST REFERENCES 228
3· THE PREPARATION, COMPOSITION, AND COST OF THE EASTER CANDLE 232
4. THE SIZE OF THE EASTER CANDLE 234
5. THE POSITION OF THE EASTER CANDLE 241
6. THE EASTER CANDLE AND THE SIGN OF THE CROSS 246
(i) The sign of the cross made with a gesture of the hand 246
(ii) The sign of the cross made by incision 248
(iii) The sign of the cross made with fire 250
(iv) The use of chrism for the sign of the cross 251
(v) The origin and significance of the chrismal cross 252
7· THE FIVE GRAINS OF INCENSE 255
(i) The development of the ceremony 255
(ii) The significance of the five grains of incense 260
(iii) The insertion of the grains of incense 264
(iv) The grains of incense at Milan 268
(v) The symbolism of the grains of incense 269
8. THE INCENSATION OF THE EASTER CANDLE 271
xiii
Chapter
9. THE INSIGNIA ON THE EASTER CANDLE 275
(i) Alpha and Omega 275
(ii) The Year 276
(iii) The Charta 279
(iv) Portraits and decorations 283
(v) Flowers 284
(vi) Crucifix 286
(vii) Branches 286
10o THE BLESSING OF THE EASTER CAI~DLE : THE FORMULA 287
(i) The Romano~Gallican A and Milanese traditions 287
(ii) The Romano-Gallioan B tradition 289
(iii) The Mozarabic tradition 291
(iv) The Exultet/Dialogue/Preface as a blessing 291
11 . THE LIGHTING OF THE EASTER CANDLE 293
(i) Before the Exultet 293
(ii) During the Preface following the Exultet 294
(iii) The Agent 295
(iv) The Means 297
12. THE EXTINGUISHING OF THE EASTER CANDLE 298
13. THE DISPOSAL OF THE EASTER CANDLE 302
14. THE SYMBOLISM OF THE EASTER CANDLE 306
15. THE PROVISION OF LIGHT AT THE PASCHAL VIGIL 309
(1) The Vigil in semi-darkness 310
(ii) The Vigil in the light 314
(iii) The rites of Lyon and Milan 320
16. THE ROMAN VIGIL AND THE GALLICAN RITE 322
(i) Outline description and the Vigil-candles 322
(ii) The Lucernarium at Jerusalem 324
(iii) The development of the Roman Vigil from the Lucernarium 324
(iv) The eighth-century Gallican rite : Mode A 331
(v) The later Gallican rite : Mode B 332
(vi) The Roman Vigil in Gaul 333
17. THE ROMANO-GALLICAN SYNTHESIS 335
(i) The diversity of practice 335
(ii) The Benedictine development 338
APPENDIXES 341
Bibliography 373
Index 391
xiv
PART I
Tenebrae
Chapter One
MATINS
The origins of the daily offices are obscure. The hymns of the hours by
Prudentiua suggest that the cycle of offices existed in some parte of
the West by the end of the fourth century. Paul Bradshaw has shown that
there wae a morning and evening service in Augustine 1 a church and in
1
Ambrose's time in Mil.ane Evidence for the offices at Jerusalem comes
from Egeria : by AD ,380 the office of Matins was said daily in the
church of the Anastasia in that city. 2 In the absence of any reference
to illumination at this service, we must assume that in addition to the
functional lights in that church, which would have been essential in the
early hours of the morning, any lit~gical illumination that there may
have been was not so unusual as to elicit a comment from Egeria.
Matins, known in the West as vigiliae, had developed within Christian

liturgical practice as a daily re-enactment of the vigil par excellence
viz. that held on Holy Saturday. Functional lighting there undoubtedly
was within the cathedral tradition. However 7 there is no firm evidence
for the use of liturgical light at the night office either in the cath-
edral or in the monastic tradition until the eighth century, unless the
testimony from Poitiers in the sixth century for the service being con-
ducted in total darkness can be classed paradoxically as an instance of
the negative use of liturgical light.3
In the very early church the real significance of the Easter mystery
involved the work of redemption in the death and resurrection of Jesus
Christ. When the commemoration of the institution of the Eucharist on
the day before Good Friday became attached to the paschal liturgical
drama, there emerged the notion of the Triduum, which came to be regard-
ed as a three-day liturgical unit. The incorporation of Maundy Thursday
inevitably resulted in the commencement of the Triduum with the celebra-
tion of the night office in the early hours of Thursday morning. By
AD 4oO the development of the Triduum was complete in the West at any
rate; for the liturgical idea of the sacred Triduum is found as early
as the time of Augustine and Ambrose.
4 From their writings it is
1 Daily Pr~er PP• 112-13. 2 Wilkinson PP• 66-69.

3 See PP·3- •
4 Ep.55.24 (PL 33.215) and Ep. 23.12-13 (P.L 16.1030) respectively.
1
evident that Maundy Thursday 9 Good Friday 9 and Holy Saturday were ro=
garded as a three~day liturgical unity. We may be confident that the
night office of each of these three days were similar in content in that
psalms and readings were characterised by a penitential and funereal but
'hope-touched solemnity'. In structure the night offices of Maundy
Thursday and Good Friday were similar to those of the other nights of
the year; but we shall see presently that Matins/Lauds of Holy Satur=
day was generally throughout the West quite different.
Pierre Batiffol, treating of Matins in the West in the early Middle

Ages, was almost certainly correct to claim that 'the office of these
days •••• was undoubtedly a purely Roman creation.• 1 As early as the
time of John Cassian (£.AD 415-435) Matins was divided by the monks into
three nocturns, 2 the structure still found in the twentieth century.
That St Benedict, whose rule was based on Roman liturgical practice and
was drawn up between AD 525 and AD 550, provided no offices of his own
for the Triduum suggests that Matins and Lauds at least existed in Rome
in the sixth century in much the same form as now. It seems unlikely
that Augustine's Tracts 2! the Psalmswhich came to be read during the
second nocturn of each of the days of the Triduum would have been in-
corporated into the night office during Augustine's lifetime. Its more-
or-leas final form, therefore, would have become fixed sometime between
AD 430 and £•AD 525, and in Rome,3 since after the Vandal conquest of
North Africa in AD429 liturgical ideas between Rome and North Africa
probably entailed difficulties of interchange. It is tempting to attri-
bute the fixing of the form of Matins to Pope Gelasius (AD 492-6), who,
though it has been shown that the Gelasian Sacramentary was wrongly
named after him, has been traditionally associated with liturgical re-
form. It may well account for the inclusion of Augustine's Tracts into
4
Matins by Gelasius who himself was an African.
The earliest mention of the night office of Matins in the West

occurs in the Life .2! §! Ambrose by his biographer Paulinus. It can be
1 History p.92.
2 De inst. Coen.3. 8-9. (PL 49.83Aand .144A).
3 It ISknownthat in the time of Leo the Great (£.AD 450) monks per-
formed the choral office of Lauds in the Roman basilicas. Under
their influence Lauds became closely associated with Matins. It was
at this time that Matins with its division into three nocturne was
introduced into Rome according to the evidence of John Cassian (~.).
4 Liber Pontificalis I p.255.
2
Hoc in tempore primum antiphonae, hymni, ac vigiliae in ecclesia
Mediolanensi celebrari coeperunt. Cuius celebritatia devotio
usque in hodernum diem non solum in eadem ecclesia, verum per
omnes pene Occidentia provincias manet.1
The vigiliae mentioned in this extract can hardly be the paschal vigil
which in Ambrose's day waG an observance of long standing. It must re~
fer rather to tho introduction of a daily performance of the night office
by Ambrose in the basilicas of Milan. This is further borne out by the
additional mention of the content viz. the antiphons and psalms (bympi).
Furthermore Paulinus is witness to the establishment of the night office
in almost all the provinces of the West (i.e.the western Roman empire)
at the end of the fourth century. However,it is not clear from his
Latin whether the Church of Milan introduced daily Matins before or
after the other provinces.
Evidence for the content and ceremonial of the night office for the
three days of the Triduum in the fourth and fifth centuries in the West
is lacking. Although we know that the night office of Holy Saturday
was an all-night vigil, 2 there is no reason to suppose that the night
offices of Maundy Thursday and Good Friday differed much structurally
from those performed on the other nights of the year• It is not until
~· AD 569 in Gaul and possibly a century and a half later in Rome that
we have evidence regarding illumination at this service, either function-
al or liturgical, or the lack of it; and it would be unwise to read
back into an earlier age the practice of a later century.
Sixth-century Gaul
The first indication we have for the liturgical use of light, or

rather in this instance the negative use of it, at the night office
comes in an incident, described by Gregory of Tours, which occurred in
the Abbey of the Holy Cross at Poitiers in Gaul and which can be dated
with some confidence to £· AD 569. The miracle of light which took
place in the monastery was seen to have been caused by the presence of
1 'At this time Matins with antiphons and hymns began to be celebrated
in the Church in Milan. Support for this service survives to this
day not only in that church but throughout nearly every province
in the West.• Vita sancti Ambrosii (PL 14-15.}1D).
2 Bradshaw p.1}2.
3
a fragment of the True Cross which Queen Radegonde~ consort of Clotairc I~
had brought back from Jerusalem in the above-mentioned year.
Sexta feria ante sanctum Pascha, cum in vigiliis sine lumine per=
nootarent, circa horam tertiam noctis apparuit ante altare lumen
parvulum in modum aointillae.1
Gregory supplies us with three items of information which are of eap=

ecial interest and relevance to our investigation
(i) The incident took place on Good Friday.
(ii) The vigil lasted throughout the night.
(iii) The service took place in darkness.
All-night vigils belonged to the older tradition of monasticism which
derived from the East. 2 It would appear,therefore,that Benedict's Rule,
which prescribed a shorter night office, had not yet been introduced
into the monastery at Poitiers. The apparently unassailable statement
that the night office of Good Friday was conducted in total darkness
raises a number of questions and merits further examination. (a) Was
Matins in the monastery at Poitiers in the sixth century conducted in
the dark on every night of the year? We cannot be absolutely certain;
but it does seem likely that some form of illumination was used, if not
for liturgical, then for utilitarian reasons especially for reading.
Moreovertthe very mention of~ lumine3 would imply that a service
held in total darkness was not normal. (b) Assuming that it was the
usual practice to have illumination at the night office, was Matins/Lauds
held sine lumine on either of the other two days of the Triduum? The
earliest documentary evidence for the night office of Good Friday acc-
ording to early Gallican practice is Ordo 17. Though written some 200
years later than the incident in question, it probably reflects some of
the usages of a much earlier age. This ordo does imply the use of lights
at Matins of Good Friday, which conflic~ith the evidence from the
4
monastery; but in view of its lateness and unknown provenance should
not be taken as corroborative of Gregory's information. (c) Is the
night office that the author has in mind that held in the early hours
of Friday morning, or that which followed Good Friday early on Saturday
morning? According to Gregory's account the vigiliae took place
1 cWhen the all-night vigil was being observed on Good Friday before
holy Easter, a small glint of light like a spark appeared before
the altar about the third hour of the night., Gregory of Tours,
Miraculorum ~ .!!!. Gloria MartyrUIJ! 1.5 (PL 71. 709B).
2 Aurelian, Regula ~ Monachos(PL 68"'.396).
3 The phrase must indicate total darkness otherwise the effect and
significance of the pin-point of light would have been considerably
4 See also Chapter 2, Stage 1 p. 9 • /weakened.
4
aexta f~~~ ,<!_n~ R_anCtll!t Paschae The usual interpretation of this re=
quires us to understand that the author is referring to the office held
in the early hours of Good l!~riday. However., if we study the account of
the miraculous occurrence more closely, it is possible to arrive at the
conclusion that the office which Gregory is referring to is actually
that of Holy Saturday morninge
Gregory continues with his narration to relate how this tiny spark
of light began to increase in size, and as it did so, scattered rays of
brilliance (~ fulgoris) in all directions. Then gradually it began
to rise into the air, lighting up the sky and bathing the earth in lum-
inescence. Little by little it began to fade; and eventually it dis=
appeared from view. 1
It is clear that the light represents Christ; and its emanation

from the fragment of the Cross kept in a casket under the altar, its
upward movement, and the illumination of the entire universe by its
brilliance symbolise respectively the Resurrection, the Ascension, and
the universality of Christ. We may leave to one side the question of
the incident's historicity, and rather concentrate on the significance
and time of its occurrence withi.D. the context of the Triduum; and con-
sider the possibility of the miracle having taken place at the night
office of Saturday rather than of Friday. It is of course not possible
to comment upon or rationalise about the timing of a miracle or strange
occurrence to the extent that one can say with authority that an incid-
ent should have occurred at a specific time to suit the circumstances.
Nevertheless,if we reflect upon the timing of the miracle in relation
to the predicament or situation in which the Pictavian monks found them-
selves, it may seem that the luminous phenomenon which was inseparably
linked to the fragment of the Cross can be more readily understood and
explained within the context of Friday night's devotions rather than
Thursday night's. Assuming that the Triduum in Poi tiers in the sixth
century bore the same funereal aspect that was prominently character-
istic of it in later centuries, then the significance of the inbursting
of light can be interpreted in two different ways according to whether
the miracle occurred at the vigiliae of Friday or Saturday. At the
night office of Friday it would be seen as an indication that the dark-
ness of the Passion which was about to be commemorated was not permanent,
and as an anticipation of the return of the divine light in the early
1 Miraculorum ~, ibidem.
5
hours of Easter morning. However this interpretation is strained 9 and
detracts from the sequence of events in the Passion narrative since a
foretaste, as it were, of the Resurrection is at this juncture in the
Triduum both awkward and out of place. On the other hand the appearance
of the light at the night office of Saturday is easier to explain both
liturgically and from a symbolical point of view : (i) This office fol-
lows the Veneration of the Cross. upon which tbe attention of the monks
had been fixed some hours before. (ii) Friday night/Saturday morning is
the time when the Church is in deepest mourning. The appearance of the
light, literally as a ray of hope, would have been much more meaningful and
significant at that time. (iii) The light is intrinsically associated
with the fragment of the True Cross which may well have figured in the
ceremony of Good Friday afternoon.
There are three additional considerations which support the view

that the service in question was the night office of Saturday. (i) The
first centres around the phrase ~ lumine. We have already referred
on page 4 to the probability that the night office was usually held with
some form of illumination, and that total darkness at this service was
exceptional. Since the office consisted of psalms, antiphons, and read-
ings from Scripture, the presence of lights must be presupposed. It is
true that in view of the frequency of repetition by the monks the whole
of the psalter was generally known by heart so that reference to the
written word could be dispensed with. At an all-night vigil conducted
in darkness, it seems unlikely that lessons were 'read'. The importance,
if not necessity, of light for the purpose of reading lessons is clear
1
from a rubric in Ordo 3QB.
A clue to the content of the office of Matins/Lauds at Poitiers in

the sixth century may be found in the former service of the night office
for Holy Saturday according to the Ambrosian rite. This office can be
traced back to the tenth century; 2 but the primitive form of the serv-
ice, especially its structure, suggests a ceremony of considerable anti-
quity. The combined office of Matins/Lauds comprised twenty-three psalms
and one canticle. Like the corresponding office in the Gallican and
Roman rites, it featured the Omissions;3 and the entire service took
1 See Chapter 2,
heart' figures
tlagt3 p.1 +. The idea of a lesson being read 'by
S Benedict's Rule. (Bradshaw p.145.)
2 Borella p.10~
3 Appendix 1 : The Omissions.
6
place in darkneso. r4oreove~Holy Saturday at Milan even in the reformed
rite is the only day when the night office is said without light. Ad=
mittedly, in trying to envisage or reconstruct the form of the night
office at Poitiers in the sixth century, one must use the Milanese evid=
ence with caution since it attests the practice of another church, and,
in spite of its likely antiquity, is valid strictly only as far back as
the tenth century; but as we shall see in the next chapter, the Milan=
ese form of ~~tins, in spite of its relatively late date, in all likeli=
hood preserves the primitive form of the night office for Holy Saturday
that was once widespread throughout the West.
(ii) We have already suggested that, although the Gallican Ordo 17

is to be dated to the latter part of the eighth century, it may well
describe the ceremonial and practice of a hundred or even two hundred
years prior to the period in which the ritual was committed to writing.
Since this ~ attests the absence of light in church on the night of
Good l!'riday (§ 98), it is reasonable to believe that no lights were lit
that night at the monastery in Poitiers.
(iii) Reference has already been made to the all-night monastic

vigils which were held during the night of Good Friday and the early
hours of Saturday morning. 1 In view of this it is almost certain that
it is this vigil to which Gregory of Tours is referring, and not to one
held on the previous day.
1 Bradshaw p.132.
?
Chapter Two
THE DEVELOPMENT OF TENEBRAE
\'le suggested in the previous chapter that the tenth-century Matins/Laude

of the Milanese rite may well preserve the primitive form of this office
in the West especially in view of the sixth-century evidence from Poitierso
Fortunately there are sufficient references to and descriptions of the
night offices of the Triduum in the ordines Romani and elsewhere to trace
the likely and logical development of Matins/Lauds from the vigiliae of
Holy Saturday to the -twice-repeated office of Tenebrae, and to confirm
our conclusions regarding the night· office of Holy Saturday at Milan and
at Poitiers. We propose to reconstruct the stages showing how the night
offices of Maun~ Thursday 9 Good Friday, and Holy Saturday may have dev-
eloped into the service of Tenebrae in the period from the sixth to the
1
eighth century. However,throughout this tentative reconstruction the
following preliminary observations should be borne in mind. (i) Although
each stage of the development is to be found in the sections of one or
more of the ordines which relate to the night offices, it does not nec-
essarily follow that a particular stage of development occurred for the
first time in that church with which the .2!:!!2, is associated. In view of
the fluidity of liturgical practice during this period, features of
ceremonial may have been borrowed from elsewhere. (ii) In the likeli-
hood of the interchange of liturgical ideas and customs, it is very
doubtful that every church passed through each of the six stages of dev-
elopment. (iii) This development was not chronologically parallel
throughout the churches of Gaul. For instance, since Ordo 26, which
attests the fully-developed service of Tenebrae on all three nights (i~.
Stage 2>, can be dated with some confidence to the period AD 750-775,
that stage antedates Stages z and ~by several years, and Stage ~ by as
much as sixty years.
It should also be borne in mind that the evidence from both Rome
and Gaul should be studied closely in conjunction, since the rites of
2
the Roman and the Gallican churches were mutually influential. More-
over, the evidence of the ordines is at times frustratingly fragmentary,
1 The practice of extinguishing gradually the lights at Matins/Lauds

emerged within the Gallican Church. It was unknown in the Mozarabic
rite; and in the Ambrosian tradition up to Vatican II corresponded
to the saying of Matins on Holy Saturday without lights (Borella p.402).
2 See Appendix 2 : !!e!!Hm and Gallican.
8
the descriptions of the services being by no means completo. Therefore»
the silence of an ~ does not necessarily signify the absencs of a part=
icular feature wherein it might have been expected to appear; and some
omissions of rubrical details may be fortuitous. At times the latter can
be inferred from the complementary evidence of other ordines.
Stage 1
The funereal aspect of the entire Triduum ia apparent from the

Omissions in the Roman Ordo 23 and in the Gallican Ordo 17. 1 Though
the former refere only to the Omissions at Matins/Lauds on Maundy Thurs-
day and the latter to those on Good Friday, it is reasonable to assume
that they formed a feature of the night office for these two days both
at Rome and in Gaul; and although there is no written evidence for the
Omissions on the Saturday, it is almost certain that they were also ob-
served on the third day. For not only would they have contributed to
the imposition of liturgical unity on the three night offices of the
Triduum; it would be especially at the night office of Holy Saturday
when one would most have expected the Omissions. 2
Of greater importance, however, for our study are the statements in

both the Romano-Gallican Ordo 16 and Ordo 17 that on Good Friday night,
no light was lit in church; but that it was hidden away out of the sight
of all until Holy Saturday.3 It follows that in the churches to which
the two ordines relate the night office of Holy Saturday was held in
total darkness. Although direct evidence from the eighth century is
wanting for illumination at this particular office in Rome, the absence
of light at this service can be inferred with confidence from the test-
4
imoey of Pope Zachary (AD 741-52). In a letter to Boniface of Mainz
he describes how fire is reserved on Maundy Thursday at the Lateran
Cathedral in three large lamps for consecration and use at the Vigil on
Holy Saturday. The concealment of the newly-kindled fire?in a remote
part of the church and the consignment of the unhallowed element to a
state of limbo, as it were, emphasized the unsuitability ot the secular
flame.,for liturgical use. It is true that the two honorific lights,which
1 OR 23.1 and OR 17.93 respectively. The former, though written in

the period AD 775-790, almost certainly reflects earlier liturgical
practice.
2 See Appendix 1 : ~ Omissions.
3 OR 16.36 and OR 17.93 respectively.
4 PL 89.951. For a fuller discussion of this evidence, and for the
text of the letter, see !S! New E!a ~ E2!!!!, pp.170-7.
5 Or more likely, an already-living flame.
9
accompanied the Pope when he walked from the Lateran Palace to tho
Church of s. Croce in Gerusalemme on Good Friday, were kindled from one
1
of the three reserved lamps. Howeve~ these torches or thick candles,
adopted by the papal court in the fourth century in imitation of those

borne before the Emperor, 2 were civic lights, and remained outside3 the
above-mentioned church on Good Friday because their flames were uncon-
secrated. In view of this it is hardly likely that lights for the night
office of Holy Saturday would have been kindled with a flame taken from
4
-one of the three lamps. Elsewhere we have suggested that the Lateran
Cathedral, in which the night office in question was held, remained un~
illuminated throughout the whole of Good Friday and for most of Holy
Saturday for both liturgical and symbolic reasons. There can be little
doubt, therefore, in view of the evidence from Poitiers, Milan, and Rome
and the attestation of the two ordines that the primitive night office
of Holy Saturday throughout a large part of the Western Church was con-
ducted in darkness. It is also likely that the content of the service
was similar to that of the Ambrosian night office, consisting primarily
of psalms with accompanying antiphons, and,as at Milan, lacked scriptural
readings because of the absence of light.
There is no evidence to suggest that the provision of light at the

night offices of Maundy Thursday and Good Friday differed from that
during the other nights of the year. According to the ordines Romani
illumination was of two kinds. The church lamps provided the functional
light, whilst the seven lamps that were placed before the altar supplied
the liturgical luminosity. Throughout the centuries in the Ambrosian
rite the lighting arrangements for the night offices of Maundy Thursday
and Good Friday remained unchanged.
Summary of evidence - Stage 1

wr5 Matins and Lauds normal lighting.
GF Matins and Lauds normal lighting.
HS Vigiliae : psalms sung in total darkness.
1 OR 23.10. The writer accepts Andrieu's suggestion that the phrase

~ unguario refers to one of the three reserved lamps (&!§ Ordines
Romani III pp.318-19).
2 Jungmann pp.132-33· See also Fulgentius Ferrandus, PL 67.884C.
3 Theodore, the Archdeacon of Rome, is quite explicit about the absence
of illumination in s. Croce during the Solemn Prayers and the Venera-
tion of the Cross : nullum lumen habetur lampadum ~ cereorum !a
ecclesia i!l Hierusalem (AD 832). (Amalarius, Lib •.Q! Ord.Ant. XLIV 2.)
4 See The Roman Vigil, P•325.
5 MT = Maundy' Thursday; GF = Good Friday; HS = Holy Saturday.
10
Stage 2
Evidence for the development of Stag0 1 comes from Rome~ and is re~
corded by Amalariua of Metz on the occasion of a visit by him to that

city in AD832. He had asked the Archdeacon Theodore about the use of
light at the night offices in Rome during the Triduum. The reply came
1
Soleo ease cum apostolico in Lateranis 9 quando officium celabratur
do caena Dominio Nihil autem ibi in eadem ~octe observatur de ex~
tinctione luminum. In feria sexta nullum lumen habetur lampadum
oive cereorum in ecclesia in Hieruaalem, quamdiu domnus apostolicus
ibi orationes solemnes facit, aut quamdiu crux salutatur, sed tamen
in ipsa die novus ignis aecenditur, de quo reservatur usque ad
nocturnale officium.2
At first sight Theodore appears to have ignored Amalarius' question con-

cerning the extinction of lights at the night office. This has led
scholars like H.A.P.Schmidt to believe that Amalarius and Theodore have
a different problem in mind. Schmidt goes on to say3 that Amalarius is
concerned with the extinction and production of fire and light at the
time of the Holy Triduum, Theodore with the use of light on Good Friday.
It is true that in ~1lapter XLIV Amalarius is indeed concerned with the
extinction and production of fire and light; but Schmidt has overlooked
the faot that Theodore has not only mentioned the extinction of lights,
but he has informed Amalarius about the production of new fire. Dendy,
assuming that Amalarius' question is primarily about the extinction of
4
light at the night office, writes that 'Theodore does not give a com-
plete direct answer'; and a superficial reading of the question and the
answer would suggest that this was so. Howeve~let us examine in greater
depth exactly what the question was which Amalarius asked, and especially
the information from which the question arose. Amalarius wrote5 :
1 Officium must be translated 'service', and not 'office' which would

almost certainly signify Matins, Vespers etc.; and the Dhrase ~
caena Domini refers to the institution of the Eucharist and not to
MaundJ Thursday generally. Otherwise Amalarius is almost certain
to have referred to that day as quinta feria.
2 'I am accustomed to be with the Pope in the Lateran Cathedral when the
service of the Lord's Supper is being held. That night in the cathe-
dral there is no ceremony involving the extinction of lights. On
Good Friday there is no light from any lamp or wax-candle in the
Church of S. Croce in Gerusalemme when the Pope recites the Solemn
Prayers there or during the Veneration of the Cross. However new
fire is kindled on that day, and this is reserved for use at the
night office.' Amalarius, ~·S! Ord.~. XLIV.2.
3 Hebdomada Sancta II p.811.
4 ~ Y!! 2J Lights p.135·
5 ~·!!! 2£!!.~. XLIV .1 and 2.
11
Mos eccleoia0 nostrae obtinet per treo noctes, id aat par fsriam
quintam, quae vacatur caena Domini, et per Qextam, quae vocatUX'
parascheve~ et per septimam, quae vocatur sabbatum sanctum, ut ex=
tinguantur luminaria ecclesiae in nocte. De more sanctae matris
nostrae Romanae ecclesiae interrogavi archidiaconum Theoqorum
memoratae ecclesiae, scilicet Romanae, qui respondit •• e.
Two facts concerning the custom of Amalarius' church emerge as being

important for our study. (1) The lights of the church were extinguished
at night. (ii) This happened on three successive nights. We have already
noted that in this chapter Amalarius is concerned with the extinction
and production of fire and light; and in recording these two facts he
wishes it to be known that in his church an extinction of light takes
place, and that it occurs on three separate occasions. It could be arg=
ued that hie mention of it having taken place at the night office is
almost incidental. We can only surmise as to the actual form of words
which Amalarius used when he asked Theodore about the custom of the
Roman Church; but in view of his concern about illumination, and from
Theodore's reply, it. is clear that he was interested primarily in the
extinction or rather in the possibility of the extinction of lights at
Rome, and not in the night office or the fact the loss of light took
place at night. This makes Theodore's reply both intelligible and satis-
factory. Theodore had been asked about the extinction of light; he
answered on that subject.
It is clear from the statement of Amalarius and from the reply of

Theodore that the development into Tenebrae of the three night offices
of the Triduum was almost complete within the church with which Amalarius
was familiar, but only in an embryonic stage in Rome. (Liturgiologiste
have often commented upon the conservative nature of the Roman Church;
and it is true that liturgical development took place much more quickly
in the regions to the north of the Alps. Howeve~ to refer to the purity
of the Roman rite, and to speak of the somewhat colourful developments
of transalpine liturgy as 'Gallioan corruptions' 2 betrays an approach to
scholarship which is marred by bias and lacking in impartialityJ
1 'During the last three nights of Holy Week, that is on Maundy Thurs-
day', Good Friday, and Holy Saturday, it is the custom of our Church
that the lights of the church building should be extinguished. Con-
cerning the custom of our holy mother Church of Rome, I asked the
Archdeacon Theodore, of the said Roman Church. He replied •••• •
2 Thus Dendy p.137. His prejudice is further borne out by a typically
vague assertion that Rome was 'for a considerable period large1y
successful' in repelling such corruptions. (The writer's italics.)
12
In view of th® reply of Theodore it must be assumed that in Rowe in
the firBt third of the ninth century there was nothing unusual about tho
provision of light at tho night offices of Maundy Thursday and Good
Friday. We saw that for the night office of Holy Saturdayi according to
Theodore's testimony, fire was kindled on Good Friday and then reserved
for use at that services How many lights were kindled at this office we
are not told; but a glance ahead at Ordo 30B, which we have adduced a6
evidence for Stage 3, may provide us with the answer. According to that
document a lamp was lit at the night office of Holy Saturday for the
purpose of reading (A 36). This, we suggest, was the result of the in=
elusion of lessons in the night office for that day. Prior to that time
the service was held in darkness, as Ordo 23 attests (Stage 1). It was
therefore sometime between the first half of the eighth century 1 and the
archidiaconate of Theodore that illumination was introduced into the
office in question. During the reign of Pope Leo III (AD 795-816)
Gallican influence resulted in the appearance in Rome of the .Minor Rog-
ation Days. His reign may well have witnessed other importations from
Gaul. Andrieu's dating of Ordo 30B to the end of the eighth century is
entirely consistent with a Leonine introduction of illumination into
the night office of Holy Saturday.
The testimony of Ordo 30B and the even earlier evidence of Ordo 26
conclusively demonstrate that the use of light at this service of
Matins/Lauds on Holy Saturday first appeared in Gaul and not in Rome.
We suggest that in Theodore's time only one lamp was used at the office.

MT Matins and Lauds normal lighting.
GF Matins and Lauds normal lighting.
HS Matins and Lauds one lamp (or candle) for reading.
Stage 3
In the evidence of Ordo 30B, which constitutes Stage 3 in the devel-

opment of Tenebrae, we encounter for the first time the phenomenon of
the gradual extinction of lights. Formerly known as §!;-~ from its
provenance at the Monaster,y of St-Amand-en-Puelle, this ~ has a
marked Roman appearance; but its provenance, date, and the distinctive
Gallicisms it contains indicate that it was used by a church in Gaul
1 Martini's dating. Vogel p.160.
13
which had adopted the Roman liturgy possibly as a result of tho illflu=
ence of Charlemagne 9 but which was still retentive of traditional cere=
monial and receptive to indigenous inf.lences. There are three reasons
for believing that Gallican influence is present in the night offices
of Good Friday and Holy Saturday as recorded in Ordo 30B. (i) There
was during this period, that is, the eighth and ninth centuries, a tend=
ency for the austere Roman ceremonial to yield to the more vigorous
Gallican ritual especially if the former had been divorced from ite
native Italian milieu. (ii) The ceremonial of the night office of Ordo
30B features a development in the use of light at Good Friday's service
which was unknown in Rome in AD 832. (iii) The evidence of Ordo 26,
attesting in Gaul the gradual loss of light during each of the night
offices of the Triduum, antedates Theodore's testimony by at least fifty
years. It is therefore difficult to believe that the ceremonial in Ordo
30B which relates to the llSe of light in the night offices of Good Fri-
day and Holy Saturday -- soon to be described -- could have had its pro-
venance in any other liturgical tradition but the Gallican.
The silence of Ordo 30B regarding nocturnal illumination at the

office of Matins/Lauds on Maund1 Thursday suggests that the service was
held with a display of lights normal for the night office; and there is
no reason to believe that it differed in this respect from the night
office of Maundy Thursd81 of Stage 2. Matins/Lauds of Holy Saturday is
likely to have been similar to the correspondi.Dg office of Stage 2 in :
respect of lights; for we are told that for the service :
tantum una lampada accendatur propter legendum (§ 36).
As suggested for Stage 2 it was almost certainly the addition of lessons

from Scripture to the night office of Holy Saturday that necessitated
the introduction of a light into a service previously held in the dark;
and it is reasonable to suppose that this was placed on or next to the
lectern.
Howeve~it is in the use of lights at the night office of Good

Friday that we find a significant development from and contrast with
the practice known to Theodore of Rome. Matins would appear to be con-
ducted with the usual illumination; it is at Lauds that we encounter
for the first time the gradual extinguishing of lights:
sed tantum inchoat ad matutinum antiphona in primo psalmo, tuta

lampada de parte dextra, in secundo psalmo de parte sinistra;
14
similiter par omnos psalmos usque VI aut VII, aut in finem
evangelii 9 reservetur abseonaa usquo in Sabbato sancto.1
There can be little doubt that the lampa 9 extinguished gradually as the
office progressed, were the seven sanctuary lamps attested in other
ordineso 2 The lamp, hidden at the Benedictus and reserved until Holy
Saturday, is to be identified with 'the light which was concealed on
Good Friday 0 o3 From this flame were lit the archdeacon°o two honorific
candles which remained burning during the readiug of St John°a Pasaion
on Good Friday (§3QJ) 9 tho lamp for the night office of Holy Saturday(§36) 0
4
and the two Vigil-candles for the Blessing of the Font (S37)e It must
be assumed that the seven lamps were alight during Matins also; but the
~ makes no mention of aQJ other form of illumination, functional or
liturgical, at either of these combined offices. It is likely that at

this stage in the development of Tenebrae church lights featured at Mat=
ins, and were extinguished before the start of Lauds. In view of the
absence of evidence to corroborate this, however, this suggestion must
remain suppositional.
It is a matter for speculation whether the lamp, removed at the con-

clusion of Lauds on Good Friday, was subsequently brought back into
church, its flame having been transferred to another lamp, and replaced
in its original position in readiness for the Vigil of Holy Saturday; or
whether it remained in the place of reservation until the following day.
It is unlikelJ, however 9 that this lamp provided the light for reading
at the night office of Holy Saturday. For the physical removal of the
sole source of fire would have exposed the flame of the lamp to the
possibility of wick-failure or some other mishap, to the inevitable
dismay and discomfiture of all.
1 'But as soon as the antiphon for the first psalm at Lauds begins,
a light is extinguished on the right-hand side and at (the antiphon
of) the second psalm one on the left. Similarly (they are exting-
uished from right to left) before all of the psalms i.e. until the
sixth or seventh psalm, or at the end of the Benedictus. Let the
last lamp be concealed and reserved for Holy Saturday.' OR 30B.28.
Note the solecisms in the Latin text.
2 See Appendix 3 : !!!!_ ~ Lamys.
3 'Lumen, quod feria sexta absconsum est.' This formula is found in
Ordines 17,23, and 30A, and in four Gelasian sacramentaries. See
especially p. 101.
4 Unlike the flame from Zachary's three lamps - see above p. 9 - this
reserved light was obtained from already-consecrated fire; and
could therefore be used for liturgical purposes.
Summary of evidence = Stage 3
MT Mati:os and Lauds g normal lighting.
GF Matins normal lighting.
Lauds gradual extinguishing of aeven lamps.
= last light removed.
HS Matins and Lauds : one lamp (or candle) for reading.
= extinguished at conclusion of service.
S.tage 4
Evidence for Stage 4 is provided by Ordo 30A, a document from north=

ern France, more or less contemporaneous with Ordo 30B. Section 1 of
this ~' which refers to the nine psalms, lessons, and responsories
of Matins on l4aun~ Thursday. makes no mention of illumination at this
service. However, at the same office on Good Friday it records (§ 5)
Deinde sequitur matutinum. Lucerne extinguuuntur.
Although the points at which the lights are extinguished are not given,
unlikeOrdo 30B, it is almost certain that a gradual loss of light is to
be understood here; for the presence of light at Matins and its loss
at Lauds corresponds with the arrangement described in Ordo 30B. It is
in the description of the night office for Holy Saturday that a diverg~
ence from the practice attested in Ordo 30B occurs (§ 12). We read
in lucernis accendendis vel extinguendis, sicut superius diximus

ita fiat. 1
This duplication of Good Friday's ritual on Holy Saturday is perhaps

the most significant stage in the development of Tenebrae. It is likely
that the repetition occurred both to signify the continuing period of
mourning within the Church and at the same time to commemorate the three
hours of darkness on Good Friday, a suggestion advanced by Rupert of
Deutz. 2 Though the number of lamps to be extinguished is not stated,
there is no reason to believe that it was other than seven. These were
lit on Friday night and extinguished one by one at Laude which followed.
Elsewhere in this ~ we learn that fire was hidden away on Good

Friday for use at the Rlechal vigil on Holy Saturday night. There can
1 'Let the arrangements for the kindling and extinguishing of the

lampe be the same as we have prescribed above (i.e. for Good Friday). 1
2 ~ ~.gu. V (PL 170.148B). He adds that the extinguishing of the
candles also signifies the darkness of the Jewish nation which
killed the prophets.
16
be little doubt that this fire was obtained f~orn the lamp ~eser~ed at
the conclusion of Lauds on Good F~id~. Since there was no need of a
second reserved flame at the end of Lauds on Holy Saturday 9 the last Qf
the seven lamps would have been extinguished at the conclusion of that
office.
Summary of evidence = Stage 4
MT MatiW3 and Lauds g normal lighting.

GF Matins normal lighting.
Lauds gradual extinguishing of seven lamps.
HS Matins normal lighting.
Stage 5
It was perhaps inevitable that the light-feature, which now disting-

uished the night offices of Good Friday and Holy Saturday, should be ex-
tended to include Matins/Lauds of Maundy Thursday in view of their
common funereal character; and it might have been expected that uniform-
ity would be achieved according to the following scheme :
MT)
Matins normal lighting.
GF)
) Lauds gradual extinguishing of seven lamps.
HS)
Howeve~no evidence has survived to show that development ever took
place along these lines. Before aqy changes occurred in the lighting
arrangements for Maundy Thursday's night office, further development
took place at Matins/Lauds of both Good Friday and Holy Saturday, in
which the gradual loss of light became a feature of Matins as well as
Lauds; and in view of the tripartite division of Matins into nocturne,
the complement of lights at this office was required to be divisible by
three. Gallican Ordo 28 is our sole witness for the employment of this
light-feature at Matins of Good Friday and Holy Saturday (§ 49). Accord-
ing to this same~ the night office of Ma~ Thursday began shortly
after midnight; and for its celebration it was prescribed that the
church should be fully illuminated (§ '1) : ,!!; !!.£2 ecclesia .2.!!!!!!. lumina
decoretur. 1 The silence of this ordo in respect of the loss of lights
on this day, in contrast with the clear directions to extinguish them
on Good Friday and Holy Saturday again indicates that normal illumination
1 'Then let the church be filled with light.'

17
obtained for the duration of the service. The use of the phrase ~
lumen suggeota that the illt!!Dina.tion of the church should 'bo brighter
than usual. In two ordines the phrase appearo in the rubric prior to
the start of Matins in which the gradual loss of lights occurs. 1 For
those participating in such a service the experience of passing from a
world of light into a darkness that symbolised death must have been
quite dramatic.
Although the offices of Matins and Lauds were sung concurrently,

the extinguishing of the lights, spread over the two services, had the
unifying effect of combining even further the two originally-separate
parts, and imposing upon them the single name of Tenebrae.

MT Matins and Lauds normal lighting.
GF Matins gradual extinguishing of all church lights.
HS Matins and Lauds : as for Good Friday
Stage 6
The final stage in the evolution of Tenebrae is attested in Ordo 26

and Ordo 29. 2 The practice of extinguishing the church lights at Matins
and the seven altar lamps at Lauds on all three days of the Triduum is
now established, and the shape of the service iB fundamentally the same
as that which has survived into the twentieth century. The attestation
of the twice-repeated service in the early Ordo 26 need not cause sur-
prise. For in the eighth and ninth centuries liturgical disorder reign-
ed throughout Gaul, and the development of ritual was sporadic and lack-
ed uniformity, as is apparent from a comparison of the rubrics of Ordo 26
with those of later ordines.
In view of the symbolism subsequently attached to the extinguishing

of the candles, and of the possible origins of that feature, and of the
significance of the Omissions, the performance of Tenebrae in the early
hours of Maune\1 Thursday could be said to be anachronistic. This was a
direct result of imposing liturgical uniformity on each of the services
of Matins/Lauds during the Triduum. On the other hand the funereal tone
1 OR 26.10 and OR 29.11.

2 OR 26.13-14 and 29.11-12.
18
of Tenebrae on t4aundy Thursday both foreohadowod tho subsequent ~vo~tG
of that day~ and anticipated the period of mourning on Good Friday and
Holy Saturdaya
This extraordinary service, known since the twelfth century and

possibly much earlier as Tenebrae, continued to be performed officially
on the last three days of Holy Week until the liturgical reforms follow=
1
ing tho Second Vatican Councilo Although the service underwent aomo
modifications in the use of light, its basic structure remained the same
and in ita eighth-century form it is still recognisably the same service
as that held in the twentieth century.
Ordo 26 contains the following directions for extinguishing the

lights ( § 13) : An unspecified number of lamps or candles is lit be-
fore the start of Matins, as are also the seven sanctuary lamps. The
former set of lights is gradually axtinguished throughout the course of
Matins, the first at the very beginning of the office. At the end of
the first nocturn.a third of the lights have been put out, and at the
end of the second nocturn another third. By the conclusion of the last
noeturn, which is the end of Matins, the remaining third has been ex-
tinguished, and only the seven altar lights remain burning in the church.
These are extinguished one by one during the course of Lauds. The
central light is put out last at the Benedictus.

MT Matins : gradual extinguishing of all church lights.
Lauds : gradual extinguishing of seven altar lamps.
GF Matins and Lauds as for Maundy Thursday.
HS Matins and Lauds as for Maundy Thursday.
1 It is still performed in some monastic houses e.g. at Solesmes, in

the London and Birmingham Oratories, and by the Society of St Pius X.
19
Chapter Three
THE TIME OF TENEBRAE
By the and of the eighth century the two traditions relating to the
time at which Tenebrae began were of long standing. Our sources at
this time attest the start of this service both at midnight and at the
1
oighth hour. In the cathedral churches the first office of the dey was
sung at the former time, which oignificantly was the beginning of the Qld
Roman civil day. Within the monasteries and those churches which foll-
owed monastic practice the later time was observed. In the sixth cent=
ury St Benedict had changed the time of rising for the monks under his
discipline so that, instead of rising at midnight, 'the brethren shall
rise at whatever time shall be calculated to be the eighth hour of the
2
night'. The change was made so as to allow more time for sleep.
Ordo 26 is our oldest authority attesting the start of Matins at the

eighth hour of the night, and Tenebrae on all three nights of the Trid-
uum;3 and we find that Ordo 29 in the ninth century, and Poitiers and
~in the tenth also prescribe monastic practice.
4
Howeve~from about
1200 the trend began in Benedictine houses to commence the night office
at the earlier time; and it is tempting to discern in this the begin-
nings of that process whereby Tenebrae became an anticipated office,
generally performed in the late afternoon of the previous day. Two later
ordines, 5 which describe papal ceremonial in the fourteenth century and
which were instrumental in shaping the liturgy of churches subject to
or in close contact with the Roman Church, state that Tenebrae began in
the late afternoon of the previous day. The Caeremoniale Episcoporum
explains that, according to the ancient Italian method of counting hours,
the service should begin at the twenty-first hour, that is, at 4.00 p.m.
in March and at 5.00 p.m. in April. 6 The former time is attested at
Limoges in the fifteenth century. 7 The change was brought about, it is
generally believed, partly through a wish to make the service more acc-
essible to both secular clergy and laity, and partly by the persistence
of the tradition inherited from Judaism in which a day is reckoned from
sunset to sunset; and this arrangement obtained generally until the
liturgical reforms following the Decree of Maxima redemptionis which is
1 For the equivalent of this time by modern reckoning, see Chap.9 pp.93-5·
2 Life of St Benedict of Aniane, PL 103. 872A; and Bradshaw p.14}.
3 OR 26.11 and .13. Tenebrae on MT implies Tenebrae on GF and HS.
4 OR 29.28; p.137; II p.56 A 212, respectively.
5 OR XIV.82 (PL 78.12o4B) and OR XV.62 (PL 78.13Q5D).
6 II.22 p.264. 7 Martene, DAER 4.22.1 p.81 (M 160).
20
dated 16 November 1955o Howeve~ in France certain cathedrals such as
1
Rouen and Langras, and the Collegiate Churches of St Victor~ and of
The Virgins in Paris 2 continued to observe the original time of eing=
ing Tenebrae as late as the eighteenth century; whilst at the Abbey of
St Germain at Auxerre, and in several Cluniac hospices, the office was
held very early in the morning up to the same period. 2
In the monasteries and houses of some of the religious orders 9 which

were not affected to any appreciable extent by the presence of the laity
at the offices, the primitive tradition of holding Tenebrae in the early
hours of the morning was perpetuated. In the seventeeth century some
monasteries such as St Maur and St Vanne in France reverted to holding
the service at the former time of 2.00 am.3 The Cistercians sang the
office at about 3.00 am until the time of the Second Vatican Council.
On the other hand among the Dominicans, the Franciscans, and some Bene-
dictine houses such as Quarr Abbey, Tenebrae was always an anticipated
office as in the Roman rite. Following the liturgical changes of 1955,
Matins and Lauds were no longer to be anticipated on the previous even=
ing; but were to be 'said in the morning (of the day itself) at the
appropriate hour•.
4 This has allowed a certain flexibility among the
religious orders and in the churches where the office of Tenebrae still
survives.
1 Mart~ne, ~ 4.22.1 p.81. 2 Grancolas p.296.

3 Schmitz Vol.6 pp.166-7. 4 Fortescue & O'Connell (11) p.281.
21
Chapter Four
THE PROVISION OF LIGHT AT TENEBRAE
In the history of this rite three main schemes are known to have existed
for the provision of light prior to the commencement of the officeo In
the period up to ~eAD 1000, all but one of our sources 1 mention the ex=
tinguishing of a certain number of (church) lights at Matins, and also
the putting out of the seven sanctuary lamps which stood before the
altar. From the eleventh century onwards in most places, the seven
lamps no longer feature in the ceremony, and the extinguishing of the
other lights is spread over Matins and Lauds. In the third phase of
development, which occurred probably in the fifteenth century, the ex-
tinguishing of the six altar candles is incorporated into the ceremonial
during the Benedict us.
(i) The Lighting of the Lamps and Candles
The time for lighting the lamps and candles could vary depending
upon which of the three days was involved, and upon whether the new fire
was kindled on MaunQy Thursday and subsequently reserved, or on each
separate day. According to the Customary of Fruttuaria2 the lights for
Tenebrae on Maundy Thursday were lit after Compline on Wednesday; and
it is very likely that this was the practice of other monasteries. How-
ever, since there was no obvious liturgical or ceremonial reason why they
should have been lit at this time, it is likely that in other places the
lights were kindled only shortly before the start of Matins. If the
seven lamps before the altar were perpetually alight, it would have been
necessary to kindle only the church lights prior to the commencement of
MaunQy Thursday's office.
The rekindling of the church lights and the seven lamps during the
daytime of Maundy Thursday is clearly described in most of our sources
for the period up to g.AD 1000.3 They all agree that illumination re-
turned to the church after the new fire had been brought in procession
into the building, at times ranging from 11.00 am (~) to 3.00 pm (OR 26);
and it is assumed that, until the arrival of the new fire, the lights
1 Viz. Ordinea 26,27,28,29,31; PRG; Poitiers; and~·

2 Albers IV p.49. -
3 Ordines 26.4; 28.25; 29.17; 31.29; ~ II p.58; QMg (Albers V p.32);
Customary of Farfa (Albers I p.48).
22
had remained extinguished since Tenebrae. For between the conclusion
of 1-la.tins/La.uds and the commencement of Hass in the late afternoon or
evening of Maundy Thursday~ no liturgical light would have been required
1
in church. Once lit the lights remained burning throughout the remain-
der of Maundy Thursday, 2 until they were once again extinguished at Tene~
brae of Good Friday. This pattern was repeated after the liturgy of
Good Friday afternoon in readiness for the final extinguishing of the
lights at Tenebrae of Holy Saturday. Long after the seven sanctuary
lamps ceased to be used at Tenebrae, this arrangement for the triple
provision of fire obtained in those churches and monasteries where the
new fire was brought into church on three successive days.3
In churches in which the kindling of new fire was confined to Holy

Saturday, the lighting of the candles for the ceremonies of Maundy
Thursday and Good Friday would have been effected by means of a flame
already in commission within the precincts of the church : either a braz-
ier or some other device for providing heat; or the fire of a kitchen
near a monastic church; or very likely, one suspects, the perpetual
flame that burned before the reserved sacrament.
Our sources provide few details regarding the lighting of the lamps
and candles for Tenebrae. At Poitiers in the tenth century, the thirty
lamps were lit by the sacristan using a candle. This done, he took up a
position in the doorway of the church where he extinguished the candle
4
as a signal to mark the start of Matins. The Ordinal of Barking re-
cords that the candles for Matins on Maundy Thursday were lit by an
official called the secretaria, but for the corresponding office on Good
Friday it was the duty of the sacristan.5 The Ordinal of Exeter merely
states that the candles were lit (just) before Matins. 6
1 As noted elsewhere (p.102), in churches where either of the primitive

Masses were still celebrated, liturgical light would have been ob-
tained from the lamp reserved for that purpose at the conclusion of
Tenebrae. The statement in Poitiers (p.138) that the clergy enter
church 'for all the Masses' (A9 missas ~) after the new fire
ceremonies relates to the one Mass of Maundy Thursday celebrated in
the several churches of Poitiers, and not to the celebration of more
than one Eucharist on that day.
2 'UsjBe !9 vigilias'. Thus OR 26.10; OR 27.11; OR 31.31; Poitiers
P• 1 ; ~ II P• 58 § 221 •
3 For example, at Barking and Canterbury (HBS 65 pp.91 ff. and HBS 23
pp.379 ff. respectively).
4 Poitiers p.193, but only recorded for Good Friday.
5 HBS 65 p.91 and p.97.
6 HBS 37 p.132.
23
( ii) Tho Dispoei tion of Lights
Scheme 1. Nave and choir lights extinguished during Matins; the seven
lamps before the altar extinguished during Lauds.
(a) The Seven Lamps. The disposition and arrangement of these aanctu=
ary lights is discussed in Appendix 3·
(b) The Nave and Choir Lights. It is not clear whether the twenty-four
lights mentioned by J\malo.rius, by ~' and by Alcuin, the t1:renty~seven
of Ordo 32, and the thirty=nine of Ordo 29 were originally different
from the functional illum.i.nationa of the churches in question; but the
relatively large numbers of lamps involved, and the positioning of them
both in the choir and in the nave would suggest that these were the nor-
mal church lights, even though in some churches they may have been
realigned in a more symmetrical arrangement to suit the liturgical re-
quirements of Tenebrae. HoweverJthe very act of extinguishing them one
by one during the course of divine service, and the fact that from at
least the time of Amalarius the number of lights featuring in the ritual
was given a s,ymbolic interpretation, endowed the lamps, perhaps inevit~
ably, with a liturgical significance.
The display and arrangement of the lights in the five sources ment-
ioned in the previous paragraph is unknown. However, a careful study of
the description of Tenebrae in the Pontifical of Foitiers makes it poss-
ible for us to reconstruct with some confidence the actual disposition
of lights not only at Poitiers, but in other Gallican churches where the
office of Tenebrae was held. On page 139 of this pontifical is the
following instruction :
On these three nights at the night office let thirty lamps be lit
which must be arranged _!!1 ~ ~ of equal spacing. 1
The silence of the source in respect of the direction in which the rows
of lamps ran presents us with the possibility that they may have run
parallel _to the main axis of the church in an east to west direction.
Alternatively the lamps may have been placed in rows which ran from one
side of the church to the other. Arguments may be advanced in favour
of either orientation.
1 (The writer's italics.) Aequo epacio trino ordine aptandae aunt.

The phrase aequo epacio must refer to the distances or intervals
between each of the rows, and not between the individual lamps.
24
Eaet~West Orientation
ThG possibility that the rows of lamps ran in this direction raises a
number of points. (i) If the length of the church was greater than its
width, as was usual, rows of lampe running in an east-to-west direction
would be more in keeping with the design and general appearance of the
building. (ii) In such an arrangement the intervals between the lamps
would provide a more satisfactory form of illumination; for three rows
of lampo extending across the church and parallel to the altar would
necessitate smaller intervals between each lamp and could result in con=
centrations of light over the areas immediately beneath those rows.
(iii) If the rows of lamps ran from north to south, were all three loc=
ated in the nave, or was one row positioned in the choir? The former
possibilit~ should not be dismissed on the grounds that there would
have been insufficient light in the choir for the lectors, since a cert-
ain amount of light for reading is like~to have emanated from the
1
seven sanctuary lamps during the whole of Matins.
It is not clear whether the thirt~ lamps extinguished during Tenebrae

at Poitiers were also used as the normal church lights; nor is it known
if a~ other functional lights, not lit on this occasion, existed else=
where in the church. However,the gerundive in the above-quoted rubric,
aptandae, 'fitted out' or 'arranged', may well indicate that these
thirt~ lights were set out in this order especiall~ for the night offices
of the Triduum. This in turn raises the interesting question of whether

the lights were pendant lamps or candles in free-standing candlesticks.
It is almost certain that the lamps in question would have been oil-lamps
and not candles, though the use of candles for functional illumination
should not be completely ruled out. However~ the difficult~ of suspend-
ing a single candle from a beam, and the even greater problem of placing
a candle upon a loft~ roof beam seem to preclude the use of this genre
of light at Poitiers. The objection might also be raised that the
limited and restricted use of these lamps hung in abnormal positions
within the church would hardly justify the possibly lengthy prepara-
tions involved in the fixing and suspending of thirt~ lamp-chains. The
objection, however, largely disappears if one supposes that in the beams
were hooks for the chains, which remained permanently in position from
year to ~ear.
1 There is no reason to believe that these lamps were of the hooded

variet~ similar to those which hang before St Peter's tomb in Rome,
and which provide little illumination.
25
In support of thG viow that the lightc for Tenebrae comprised
three rows of candles in their holders, which extended from the west
end of the church to the sanctuary~ it may be said that the portability
of the candelabra would permit a rapid disposition of the lights, and
allow the lamps in the same row to be spaced with whatever intervals
were required in a church unencumbered in those days by lines of chairs
or pews. The objection that the use of candleholders was impracticable
amid the jostling of the congregation cannot really be sustained; for
the night office was probably never attended anywhere by hordes of the
faithful. A much more serious objection arises over the height the
candleholders would need to have been. Both from tho point of view of
safety and to ensure that the lights provided maximum illumination, it
would have been necessary for the candlesticks to stand at least six
feet from the ground. Although it is not entirely beyond belief, it
does seem very unlikely that a church such as that at Poitiers should
have possessed a set of thirty very large candelabra for use at only
three relatively short services each year. Moreover,it is most unlikely
that such candleholders would have been used for the church's functional
illuminations. Oil-lamps almost certainly were used. It is, therefore,
difficult to escape the conclusion that the functional oil-lamps of the
church at Poitiers were used liturgically at Tenebrae either ~ !!i9 or
temporarily repositioned during the latter part of Holy Week.
North-south Orientation
The possibility of an arrangement of lights in rows running parallel

to the north-south axis of the church invites two comments. (i) Rows
of lights arranged according to this orientation are actually attested
in Scheme 2; but they were confined to one part of the church. The
1
position occupied by the sacristan before the start of the service
shows that some of the lights at least were burning in the nave. An
arrangement of one line of lights in the choir and two in the nave
seems unlikely; three rows in the nave would provide illumination in
that part of the church, but would leave the choir in gloom. (ii) A
later rubric from the same description of Tenebrae in Poitiers would,
on first reading, appear to favour a north-to-south orientation
lucernarum quae ab occidentali parte ecclesiae incipiantur

extingui.2
1 See p.65.
2 'The lamps which begin to be put out on the western side of the
church.' (Page 137.)
26
and to indicate that the Bac~istan proceeded to move along the western
row of lamps until all ton ware extinguishedj and then to put out next
those in the middle row. However,the evidence of this rubric is incon=
elusive, since the instruction is qualified by the words begin ~ 2!
extinguished, and gives no indication either of the direction in which
the sacristan then proeeeded 9 or of tho direction in which the rows of
lamps rano
Evidence to confirm that the rows of lamps ran in an east=to=west

direction is supplied by Alcuin and will be considered in Chapter 8.
The arrangement of the church lights in three rows for the service
of Tenebrae, as attested above by Poitiers, was probably typical of
ma~ Gallican churches in the period prior to AD 1000, regardless of
the number of lamps lit before the start of Matins. However,an inter-
esting but intriguingly concise statement by Martene reveals that the
display of lights in some churches was perhaps not as orderly as that
1
at Poitiers. For he records that at Corbie2 and at Monte Cassino2 in
the ninth and tenth centuries the lamps to be extinguished at the night
office were dispersedly arranged throughout the church. This does not
imply a random disposition of lights. It suggests that the normal
church illuminations were used rather than specially arranged rows of
lamps.
Scheme 2. Lights lit only in the choir and extinguished during Matins
and Lauds.
Writing in the first half of the eighteenth century Jean Grancolas

records3 that in some churches candles were placed upon
4 or lamps were
suspended from beams which spanned the entire width of the choir. It
is unfortunate for our enquiry that Grancolas does not specifY any
churches where this arrangement obtained or the period when these beams
were used; nor does he mention the number of candles or lamps or even
beams that were involved. His vagueness about these beams with their
lights is perhaps indicative of his own ignorance in this matter, and
1 ~ 3.13 p.122. 2 M 1145 and M 1139 respectively.

3 Commentarius p.296.
4 The notion of a temporary beam or beams transversely placed at a
height across the choir, from the top of which protruded metal spikes
to secure the candles, is not so incredible as it might at first seem.
We shall see that the wooden plank and the hearse with its spikes
may have developed from the choir beam(s). See Chapter 7.
27
may 6uggeBt that he is only recording tho hal£~remembered facts of an
informant about former practice. Howeve~ the use at Tenebrae of
candles placed upon or lamps hanging from beams in the choir does re=
present a significant transitional atage in the development of the cere=
monial between the earliest recorded arrangement whereby the lights of
the whole church were extinguished (Scheme 1) and the practice of exting=
uishing gradually only those candles which were placed on a candelabrum
or hearse near the altar (Scheme 5). It is significant that Grancolaa
employs the term hirpices when referring to the wooden beams with their
iron spikes for impaling wax-candles or for hanging lamps therefrom.
1
The same word in the singular, hirpex, is also used to describe the
hearse of candles which stood in the sanctuary during Tenebrae.
Three factors may have contributed to the abandonment of extinguish-

ing the nave lights during Tenebrae and to the restriction of their use
to the choir. (i) In those cathedrals and monastic churches where those
attending the night office could comfortably be accommodated within the
choir, the extinguishing of lights in an unused part of the building
may have seemed unnecessary, and may even have gone unnoticed especially
if a large choir-screen effectively isolated the choir from the nave.
One of two developments may then have occurred. Either the lights in
the nave ceased to be lit so that most of the church remained in dark-
ness; or, in churches such as Salisbury where the nave lights continued
to burn in the background, their presence was disregarded until almost
the end of Lauds when it was necessary to have the building in total
darkness. (ii) When Tenebrae became an anticipated service held in the
late afternoon, the illumination of the nave would in some places have
been perhaps unnecessary, and may have gone unnoticed especially if the
building was flooded with strong vernal sunshine towards the close of
the day. 2 (iii) The lighting of the lamps in the choir only and the
concentration of light in one particular area heighten the effect of the
proceedings both for those in the choir itself who are aware of the
darkness in the rest of the building, which is symbolic of the darkness
of the world, and especially for those in the darkness of the nave who
1 See also Chapter 6.

2 Augustus Hare quotes the vivid description from Anderson's Improv-
isatore of Tenebrae on Wednesday afternoon in the Sistine Chapel.
The service, it seems, was timed to end when 'the descending sun ••••
threw his last beams in through the uppermost window' and strongly
illuminated Christ and the Apostles. The sun set just as the last
psalm was ended. (Vol.2 p.29? of ~ in Rome.)
28
observe the lights and their extinction at a distance. The drama in
the presentation of the passion and death of Jesus was after all an
important element in the funereal content of Tenebrae.
Unless the supports for the lamps were the permanent tie~beams of
the roof itself, 1one can only speculate about the number and position
of temporary horizontal bars. It seems reasonable to conjecture that
they were placed at the sanctuary end of the choir so as to allow the
whole of the beam to bo used, thereby allowing greater intervals be~
tween each light, especially if twenty=four or twenty-five were placed

on each beam; and also to give the sacristan ready access to the lights.
Scheme 3. Candles lit on a stand near the altar and extinguished

during Matins and Lauds.
The next stage in the development of the use of light would find
the choir lamps of Scheme 2 replaced by a row or cluster of lights,
usually candles, placed and extinguished in the vicinity of the altar.
The display of lights in this position represents the final stage in
the transition from the use of functional lamps with symbolic associa-
tions to the use of liturgical lights with a minimal functional purpose.
It was perhaps inevitable that functional lights, once put to liturg-
ical use and interpreted symbolically, should subsequently be placed
in close proximity to the altar and mounted on a stand or on a candel-
abrum to underline their importance and to give them visual prominence.
In Chapter 7 we have shown how the desirability of concentrating the
lights in one place so as to enable those participating in Tenebrae to
observe better the decrease in the loss of light was the principal
factor in contributing to the location of these cultic lights within
the sanctuary.
Schemes 2 and .2. represent two successive increases in the concent-

ration of light, progressively in the direction of the altar : first in
the choir, then in the sanctuary. The area where the light ultimately
became concentrated viz. the vicinity of the altar, was also partly de-
termined by the need to have illumination for reading close by. It was
also perhaps inevitable that the last light to be extinguished should
be placed within the sanctuary; for this was usually identified with
1 It should be borne in mind that lofty ceilings and stone vaultings

belong to later medieval churches. Earlier cathedrals and monastic
churches were buildings of considerably more modest dimensions.
29
ChristG Within thia scheme should be included most of the documents
which attest the use at Tenebra~ of twonty=four or twenty=fivo lightB9
scant information concerning the use of other numbers 9 apart from fif-
teen, prevents our inclusion of them also within this scheme. Most of
our sources do not mention the use or even the presence of lights either
in the choir or nave vis-§-via those extinguished on the stand by the
altar. Howeve~tho evidence from Norwich, Roucn, Salisbury, and anum=
ber of other churches suggests that background illumination existed
withi~ the cathedral tradition at leasto 1
Scheme 4 Hearse=lights and lights ~ ~ 2f the altar.
(i) At the Monastery of Farfa in the eleventh century we encounter

the somewhat puzzling statements that thirty candles were lit before the
high altar and that fifteen candles were extinguished at Tenebrae. 2 In
Chapter ? we shall examine this information in greater detail, and offer
our own solution as to the manner in which the lights at Farfa and other
monasteries were displayed.
(ii) In the former cathedral church at Laon £.1090 eighteen lights
were placed on either side of the sanctuary.3 We are not told at what
stage these additional lights were extinguished or what the purpose was
in placing them in these positions. The number too is strange; thirty-
six (or 18 + 18) seems to have no special liturgical or symbolic sign-
ificance. Moreover, unlike the fifteen hearse-candles which burned be-
fore the altar and which were liturgically integrated into the office
of Tenebrae, these additional lights were presumably at some distance
from the altar. Their purpose may have been to provide light for the
readers; but it is not at all clear at what stage in the ceremoD1 they
were extinguished. There may have been a similar disposition of lights
4
at Coutances where forty-four candles were lit for Tenebrae. If a
hearse of twenty-four candles was used, the number mentioned by John of
Avranches,5 the remaining twenty lights may have been placed in two
groups of ten on either side of the altar or sanctuary. At Chartres-
en-Vallee6 in the twelfth century it is possible that twenty-four of the
thirty-four lights that were kindled at the night office burned on the
1 The use of lights in the nave and choir at Tenebrae is treated at

greater length in Chapter 8.
2 Albers I p.46 (= PL 150.119?B.) 3 Mart~ne, ~ 4.22.2 p.81 (M 156).
4 Heuser p.228. Unfortunately he gives no other details.
5 Lib.de Off.Eccl. 52 (PL 14?.48C). The dioceses of Coutances and
AV;~h~w~coterminous. Their rituals may have been similar.
6 Martene, ~ 4.22.2 p.81 (M ?6). This relates to the Church of St-Jean.
30
hearoa 9 while the remaining ten were perhaps disposed ue at Coutanceso
CD the other hand it is possible that the additional lights both at
Coutances and Chartres~an~Vall~e may have been placed in the choir tQ
provide functional illumination. Alternatively, the possibility of
candelabra holding forty-four or thirty-four candles respectively should
not be completely ruled out. Howeve~ the frequency of twenty-four=candle
and fifteen=candle hearses, and the silence of Sicardus and Durandua
and the lack of attestation from any other source make the existence of
candelabraw designed to hold large numbers of lights, seem very doubtfula
Mention should here be made of the continued use of the seven sanctuary
lamps at the Monastery of Monte Cassino 1 and at Chartres Cathedral. 2
Scheme 5(a). Hearse-lights and lights ~the altar.
The disposition of lights according to this scheme differs from that

of Scheme 5(b) only in the number of candles used. At Soissone in the
late twelfth century twenty=four candles burned on a candelabrum in
front of the altar and an unspecified number stood on the altar itself. 3
This is the first recorded instance of the use of altar-candles in con-
nection with Tenebrae, and, incidentally, one of the earliest references
'
to candles placed upon the altar. Martene's silence concerning these
candles suggests that the original ritual did not specifY a number. The
six candles attested at Tongres in the fifteenth century are almost
certain to have been placed on the altar. There only seven other cand-
4
les were lit on the hearse.
Scheme 5(b). Hearse-lights and lights upon the altar.
This is the disposition of lights officially prescribed by the

Church of Rome for those parts of the Western Church which owed alleg-
iance to her and which were required to adopt the new Roman service-
books in the sixteenth century.5
(i) 1568 - 1955. As at Soissons the Roman rite involved a tripartite
1 Martene, DAMR 3.13.11 p.123 (M1139). 2 Ordinary p.1o8.

3 Martene, DAER 4.2} p.137 (M 305). 4 Ordinary p.1.50.
5 The papal regulations, added to the decisions of the Council of
Trent whereby episcopal independence in liturgical matters was in
principle terminated, laid down that the Roman Missal and Roman
Breviary were to be adopted everywhere except in those dioceses
which had used their own service-books for more than 200 years.
The directions for Tenebrae are contained in the Roman Breviary of
1568 and in the Caeremoniale Episcoporum (II.22) of 1600.
31
uee of light viz. altar=candlee, hearse=lights~ and the functional
lamps of the church; but it appears to have been unique in its use of
the six altar~candles in that they were extinguished during the singing
of the Benedictus at Lauds, and not at intervals during the course of
the whole office as the, eeven sanctuary lamps had been. The origin of
this feature is obscure. The use of the six candles in this way did not
1
form part of the old Dominican rite which was based on the Roman rite
of the late twelfth century. In the liturgical reviRions of 1255=6
Dominican Tenebrae remained unaffected and survived unchanged until
twentieth century. 2 Neither do the Franciscan ordinea3 of 1243=4, which
were close~ based on the Roman practice of the papal court, mention the
six candles within this context; and Durandus writing c.1280 does not
4
refer to them. There are two likely periods when the practice of ex=
tinguishing the six altar-candles during the Benedictus may have been
introduced into the Roman liturgy. (a) During the residence of the
Popes at Avignon from 1309 to 1377 the papal liturgy was in direct con-
tact with the influences of the Galli.can Church. We have noted else-
where that Gallican ceremonial was somewhat less restrained than the
austere and sombre Roman ritual. If the use of the six candles was
introduced into Tenebrae during this period, it was very likely via the
papal court at Avignon, though it must be added that there is no con=
temporary direct evidence from the liturgy of that church to corroborate
this theory. In the chaotic state of the city of Rome at this time,
liturgical innovations there appear most unlikely. (b) The second
period to be considered is the reign of Fbpe Martin V (1417-31). This
pontiff inaugurated the restoration of the Roman Church after the Great
Schism, and was responsible for the refurbishment of many of Rome's
churches and the improvement of liturgical worship within that city.
According to Roman practice up to 1955 the six altar-candles were

lit for Tenebrae on each day of the Triduum. Under the former dispens-
ation each of these days was a Double Feast of the First Class ; and
though it was not usual to have six altar-candles lit at Matins/Lauds,
the funereal character of Tenebrae placed this office on a par with a
Requiem. Since it was customary to light the six candles at this latter
service, they were therefore kindled for Tenebrae.
(ii) 1955 to date. As a result of the reforms of 1955, which affected
1 Office of ~ Week p.82. 2 King, LRO p.337.

3 HBS 85 P:7~an:Dijk II p.84. 4 Rationa:Le VI.72 p.~31.
of the oervicos of the Triduum 9 whilst tho six nltar~candles ro=
ffiQDY
mained in use at Tonobrae on ~laundy Thursday~ alterations to the cere~
monial of Maundy Thursday itself and of Good Friday involved changes
which affected Tenebrae on both of those days. After the stripping of
the altar on Maundy Thursday, the altar remained bare until the Venera=
tion of the Cross on the following day. As a result there were no
candles on the altar during Tenebrae of Good Friday. During the Vener=
ation of the Cross, the cross to be venerated was brought into church
along with two candles; and after the Veneration these candles were
placed on the altar. Then the Sacrament was brought in accompanied by
two more candles; and these were also placed upon the altar. Thus at
Tenebrae of Holy Saturday four candles were lit, and then extinguished
during the Benedictus.1
Interestingly and somewhat paradoxically, the Dominican Order,which

before 1955 had not used the six candles at Tenebrae, borrowed from the
Roman rite the practice of lighting the six candles at about the same
time that the hearse ceased to be used in the Roman rite. These candles
continue to be featured at Dominican Matins/Lauds in those houses which
still observe Tenebrae on all three days of the Triduum.
1 Wuest p.263.
33
Chapter Five
THE NUMBER OF LIGHTS AT TENEBRAE
(i) ~ Early Period
In view of the ~vidence of Poitiers and Alcuin1 it is our contention

that in the period up to £..1000 the lamps and candles extinguished
during the course of ~~tins comprised the functional lights of the
church. There were two factors that determined the exact number of
lights to be used. As early as the time of Amalarius symbolic inter-
pretations were given to some numbers especially if they suggested a
biblical precedent. It was more important, however, that the number of
lights to be extinguished should be divisible by three in order to main-
tain liturgical symmetry within each of the three nocturns of Matins.
Thus we encounter in this period thirty-nine, 2 thirty,} twenty-seven, 4
and twenty-fa~ lights lit at the start of the night office. The
silence of Ordo 26 suggests that any suitable number could be used.
According to the ceremonial of the church with which Amalarius was

familiar twenty-four candles or lamps were lit on each day of the Trid-
uum. His mention of seventy-two lights represents the total for the
three days; and though he dwells on the significance of seventy-two
rather than twenty-four, twice he states that only twenty-four lights
were lit for each of the three celebrations of Tenebrae. 6 Amalarius'
commentary on the various features of the service is reminiscent of the
symbolic and allegorical interpretations of ceremonial and vestments by
later medieval writers such as Rupert of Deutz, John Beleth, and
Sicardus. His twofold and sometimes threefold interpretation of the
symbolism which he finds in the ceremonial is not always consistent;
but it does reveal the belief of that age that the ceremonial of the
Church reflected the teaching and theology of the Bible, and was there-
fore both of divine inspiration and of divine approbation.
For Amalarius each of the twenty-four lights represents an hour, or
1 See Chapters 2 p.24 and 7 p.65.

2 OR 29.12; but see pp.68-72. } Poitiers p.1}?. 4 OR}2.5.
5 Amalarius, Lib.Q!l. 4.22.1; PRG II p.56 321}; Alcuin, PL 101.120}B.
6 Amalarius, ibidem: Accenduntur per si.Dgulas noctes XXIII! lumina and
again in the same chapter per singulas noctes memoratarum feriarum
viginti quattuor lumina accenduntur ••• ~ ~ til, ill· ('Twenty-
four lights are lit each night' and 'on each night of the said days
twenty-four lights are lit ••• for:"this--happens three times.')
more precisely an hour of daylight; and collectively they s~boliae
Chri8t 'who illuminates his Church by day and by night 9 and who ua
~ ~ reate in the tomb duing the Triduum, mourned by his Church
and hidden from view in the same way that the oun is not visible during
an eclipseo For that reason and as a sign of sorrow the lights are ex=
1
tinguished.' Over the three-day period the seventy=two lights signify
the seventy=two hours which Jesus lay in the tomb. 2 In addition to the
temporal aspect of the symbolism, the extinguishing of seventy~two
lights over three days is interpreted somewhat freely as the desertion

of the seventy=two disciples. 3 Similar comparisons and analogies are
4
to bo found in Alcuin.
Evidence from the Romano-Germanic Pontifical shows that even by the

tenth century Tenebrae was not universally observed throughout northern
Gaul and northern Germany in spite of its attestation 200 years earlier:
accenduntur in quibusdam locis in hac nocte viginti quattuor lumina.5
The alternative interpretation that 'in some places' numbers of lights

other than twenty-four were lit is unlikely.
The importance of the symbolism attached to numbers may have been

responsible for the disappearance of, or at least the lack of evidence
for the use of twenty-seven, thirty, and thirty-nine lights at Tenebrae
in subsequent centuries; the evidence for the use of twenty-four lights
is plentiful and persists well into the twentieth century. 6 For none of
the three former figures have obvious associations with biblical
numbers or situations.
(ii) ~ !!!!! diversit: ~ numbers
(a) 72 lights. Both John Beleth and Sicardus, but not Durandus,
attest the practice of lighting seventy-two candles at Tenebrae. Beleth,
who like Sicardus states that the service of Tenebrae represents the
darkness of Christ's three hours on the cross,? repeats Amalarius'
analogy with the seventy-two disciples. Sicardus also refers to their
1 Lib.Off. 4.22.1.
2 The phrases .2!! ~third .2!z and after ~ days in reference to
Jesus' resurrection were frequently thought of as embracing three
who~e days. This was a result of the inclusive system of reckoning.
3 Luke 10:1, alternative reading. 4 Lib.de Div.Off. (PL 101.1203B).
5 'During this night 24 lights are lit'i.n-;ome piiees.' PRG II p.56
6 See Table 1. 7 PL 202.105A. I § 213.
35
1
deoertiono Beleth~ echoing Amalarius~ also likens the number of
candles to the hours Christ lay in the tomb, to the number of nationa 9
and b,y extension to the number of languages. Since both commentatoro
record that Tenebrae took place on each day of the Triduum and that it
was possible to light different numbers of candles at this service, of
which one was twenty=four, it seems almost certain that they are both
attesting the use of seventy-two candles on each of the three nights
and not referring to the total for the three services, as Amalarius did.
The arrangement in which they were disposed is tmknown; the very largo
number would suggest a fairly concentrated display in the region of the
altar rather than a sporadic disposition throughout the whole of the
church. Moreover, one can only conjecture that these lights were exting=
uished in groups of three at the same points in the service at which
twenty-four candles were put out. Durand's silence at the end of the
thirteenth century concerning this number may suggest that the practice
of lighting seventy-two candles at Tenebrae had everywhere fallen into
desuetude.
(b) 44 lights. On page 30 we referred to the forty-four candles at

Coutances and suggested a possible disposition for these lights. The
1499 Breviary of Coutances does not record the number of lights that
were lit for Tenebrae. Its silence on this score may indicate that a
not unusual number of candles were placed on the hearse. John of
Avranches, whose own diocese was adjacent to that of Coutances, clearly
states that twenty-four lights were to be lit for the night office. 2
The occurrence of four as the second digit both at Coutances and at
Chartres-en-Vall,e, where thirty-four lights were lit, may suggest the
presence of a twenty-four.candle hearse within the total number of lights.
(c) 39 lights. Dom Edmond Mart~ne records that in an ~ Romanus

the monks at Monte Cassino lit thirty-nine lights for Tenebrae which
they extinguished in more or less the same way as that described in
4
Ordo 26 or PRG.3 Elsewhere he mentions that the lights were dispersedly
arranged throughout the church. To a point Dendy is correct in stating
that 'the lights appear to be those of the whole church' ; but he is
wrong to assert that Mart$ne has 'misunderstood a MS. in giving the num-
ber thirty-nine 1 • '
Dendy, one suspects, has noticed that when Martene
1 Mitrale (PL 213.298D). 2 Lib.de Off.Eccl. 52 (PL 147.48C).

---~
3 ~ 3.12.3 p.123. The 2£92 Romanus which Martene is here referring
to is either OR 28 or OR 31. See Appendix 11.
4 ~ 3.12.2 p.122.
1
referred to tho position of the lamps, he stated that at Monte CuoGino
thoro were fifteen; and 9 perhaps understandably, Dendy has come to the
conclusion that Mart~no was in orror in respect of the thirty~nine
2
lights. However,on reflection it would appear from what Martene informs
us that it is not that he has misread or misunderstood his sources, but
that he has been puzzled by seemingly contradictory evidence from the
same monastery, and possibly also by some confusion in his original
notes. 1~is would account for
further statement that the number of
~
candles used at Monte Cassino is not clear. 3
A plausible explanation for the conflicting information emerges

from a closer examination of the sources which Martene consulted. Two
documents, separated by several centuries and attesting the practice of
two different periods, were examined by Martene. (i) We suggested
above that the 2£9e Romanus, which Mart$ne inspected for information
relating to Monte Cassino, had ita provenance within the same liturgical
tradition as Ordo 28 and Ordo 31.~ It also had elements in common with
Ordo 29. For instance, it attested tho use of thirty-nine lights at
Tenebrae. Moreove~ at Monte Cassino seven small lamps hung in front of
the apse;5 and, unlike other early ordines and pontificals, Ordo 29
was compiled primarily for use in a monastery. 6 (ii) The second docu-
ment relating to Monte Cassino, which Mart~ne consulted, was an ordinary
of c.1100. 7 This probably did not specify the number of lights to be
used at Tenebrae - hence his statement about the number not being clear.
Moreove~Martene did not know how many candles were lit for Tenebrae
during this period at Rome. Therefore,since he knew that the monks at
Monte Cassino followed the Roman rite at Tenebrae, he contented himself
by giving elsewhere the number which he knew the Roman Church subseque-
ntly used and which was correct at the time he was writing : fifteen.
Amongst the liturgical features shared by the monasteries of Corbie

and Monte Cassino was the 'scattering' of lights for Tenebrae through-
out the church building.3 The linking of the two monasteries in this
way and the affinity between Mart~ne's ~ Romanus and Ordines 28,29,
and 31 increase the likelihood that thirty-nine lights were also used
1 ~ 3.12.2 p.122. 2 The Use of Lights p.146.

3 ~ 3.12.3 p.123. 4 See especially Appendix 11.
5 Q£ge Casinensis p.101. These lamps are almost certainly a survival
of the seven primitive sanctuary lamps.
6 OR 29.29 and .45. ComRare OR 26.19 and .20; OR 28.26; OR 3QB.37;
OR 31.65; f§g II P•59 S 221; and Poitiers p.139.
7 M 1139.
37
at Corbie especially in view of Grancolo.s 1 tootimony. Admittedly 9 he
records that thirty=eight lights were lit at this monastery in northern
1
France; but this figure is otherwise unrecorded within the context of
Tenebrae; it has no relevant biblical significance; and it cannot be
divided by three into equal parts. A likely explanation for the occur~
renee of this number is to assume that Grancolaa or the COP1ist whose
manuscript Grancolas read mistook XXXVIII! for XXXVIIIo I~howeve~
this figure is correct, it would suggest that at Corbio for an unknown
reason only twelve lights were extinguished during one of the nocturne.
(d) 34 lights. We have suggested on page 30 how the thirty-four

candles at Chartres=en~Vall$e may have been displayed. No other instance
is known of the use of thirty-four lights at Tenebrae.
(e) 30 lights. Poitiers attests the use of thirty lights at the

night office. Their arrangement is discussed in Chapter 7. 2 The dis=
position of the thirty candles at Farfa is described in Chapter 6.3
(f) 27 lights. Our only evidence for the presence of twenty-seven

lights at Tenebrae comes from Ordo 32. 4 Although the information relat-
ing to the night office and to the Easter candles is assembled from two
different manuscripts, there is nothing to suggest that both do not re-
fer to the same ~. The Cambridge manuscript records the number of
lights; whilst the seven sanctuary lamps are mentioned by the Paris
manuscript. It was the opinion of Michel Andrieu that the ~ origin-
ated in northern France and possibly at Corbie. I~howeve~ thirty-nine
lights were used at this monastery, the Cambridge manuscript may well
have had its provenance in a neighbouring church in view of the discrep-
ancy in the numbers.
The fact that the numbers twenty-seven, thirty, and thirty-nine had
no obvious biblical associations may only be one of the reasons why they
did not survive in use at Tenebrae in the succeeding centuries. Perhaps
a more weighty factor was the influence of the Gallican liturgical com-
mentator Amalarius, the author of Alcuin, and later, John of Avranches,
for all of whom twenty-four lights was the norm; and this number is
known to have been used in the influential churches of Mainz, Rouen,
and York in the tenth and eleventh centuries. (See Tables 1 and~.)
1 Commentarius p.296. 2 Pages 64-66.

3 Pages 52-54. 4 OR 32.5.
(g) 26 lights. It is not known why twonty=six candlos wore lit at
1
Amiens and Fontanellas 9 2 and posoibly at Exeter at one time. 3 Since
this figure was not suggested by a precedent from Scripture, its exist=
ence may be accounted for in one of three ways. (i) In Paris and Reimo
there are instances of the use of thirteen candles& There may possibly
be a connection with the ceremonial of these two churches if it could
4
be shown that two candles were extinguished at a time as at Tongres.
On the other hand two separate hearses may once have been used 9 as we
have ~uggested elsewhere. 5 This number may have been one of tho results
of combining two hearses into one. (ii) The twenty~sixth candle may
originally have been the sacristan's light which provided light for
those entering a darkened church and was subsequently placed on or near
a hearse designed to hold twenty-five candles. (iii) It is just possible
that the lamp which burned in front of the Sacrament was included in
the twenty-six.
(h) 24 and 25 lights. The use of twenty=five lights at Tonebrae

is a variant or development of the twenty-four lights prescribed by
Amalariua and PRG. This is clear from the Breviary of York which
states that twenty-four candles are lit at Tenebrae with an additional
one in the middle (of the hearse) higher than the rest (p.375). More-
ove~the use of twenty-four or twenty-five candles is not confined ex-
clusively to either the cathedral or monastic traditions (Tables 1 and 2).
However, though the two numbers are found in both traditions, the addi=
tional candle almost certainly had ita origin in monastic practice. For
in the later days of medieval monasticism, it was the practice in mon-
asteries at the beginning of the day, as the bell was tolling summoning
the monks to Matins, for a junior brother to carry a lantern to show
6
them the way. Alternatively the use of twenty-five candles may have
been a survival from the earlier period when the seven lamps were ex-
tinguished during Lauds. The possible substitution of seven hearse-
candles for the seven sanctuary lamps presupposes that six candles were
extinguished during each of the three nocturne of Matins. Although the
1 Heuser p.228.
2 Spicilegium Fontanellenae, MS., 394, in Dendy p.146.
3 Feasey (1897) does not give his source for his reference to the
twenty-six hearse-candles mentioned at the Synod of Exeter in 1289
(p.394). No figure is given in the St~tes of Exeter, edited by
Bradshaw and Wordsworth; but we know from the Ordinal of Exeter
that in 1337 twenty-four lights were used in that cathedral(HBS 37
4 See p.79 and p.82. 5 See pp.56-7. I p.132.)
6 Crossley p.83.
39
Church/ Date Source
Commentator
Amalarius 832 Lib.Off. 4.22
.fB!! £e950 FJlg II p.56
Rouen £e1050 PL 147.168D
Alcuin c.1000 PL 101.1203B
Lanfranc £.1070 PL i50.458A
John/Avranches c .. 1070 PL 147.48C
Honorius/Autun £.114o PL 172.665
Gilbertines 1150+ HBS 59 p.30
John Beleth c.1180 PL 202.105D
Soissons - 1180-90 ~ 4.23 p.13?,M305
Sicardus c.1200 PL 213.298D
Salisbury c.1210 Ordinal p. 66
Nidaros - 13 c 2!'!! p.222
Durandus c.1280 Ra. tionale VI. 72
Canterbury - 13 c HBS 28 p.274
Trier £·1300 Ordinary p. 486
Exeter 1337 HBS 37 p.132
Senlis c.1390 ~ 4.22 p.81,M301
Barking - 14o4 HBS 65 p.91
Salisbury 1531 Breviary Fasc. 1
Rouen 18 c De Moleon p.206
Orle"ans 18 c De Moleon p. 206
Premonstatensians 1930 Breviary 'P. 386
Table 1. The Use of Twenty-four Lights at Tenebrae.
Church/ Source
Commentator
Fruttuaria £.1000 Albers IV P·39
York £.1050 Breviary, SS 71 p.375
Lanfranc c.1070 PL 1~.4,58A
Gembloux c.1oao Albers II p.90
Norwich £.1265 HBS 82 p.79
Hereford 13 c HBS 26 p.}o8
Fleury 13 c ~ 3.12 p.123,M1186
Bee c.1200 ~ 3.12 p.123.M1153
Lincoln £.1440 Statutes II p.303
Table 2. The Use of Twenty-five Lights at Tenebrae.

usc of eighteen candles at Matins is attested in later practice 9 there
is no GVidence for this number ~ the earlier period.
Another possibility to be considered is that the increase from

twenty-four to twenty-five was the result of the s,ymbolic association
1
attached to the latter number. For at Hereford, where the candles
were identified with the prophets and the apostles, the twenty=fifth
candle represented Christ~ an identification presumably also made at
York. This accounted for the candle's elevated position.
2
At Salisbury and Exeter3 the twenty=four candles also symbolised
the Old Testament prophets and the twelve apostles. The ordinal from
Exeter adds that the extinguishing of them signifies the cruelty of the
Jews who persecuted or murdered them. Dendy claims that the number
twenty-four was 'genera1ly taken to stand for the apostles and the
4
prophets', although only three of our twenty-three sources actually
mention this symbolism. However, there is no evidence at all for his
assertion that the twenty-four candles could include Christ, if Judas
was not reckoned in the number. Durandus, regarding the number in a
wholly New Testament context, substitutes apostolic !!! for prophets
and elaborates the idea by declaring that for twenty-four hours the
apostles and apostolic men serve Christ by day and the Church by night. 5
Durandus also states that the twenty-four candles are extinguished be-
cause the apostles hid for twenty-four hours; and he reiterates the
analogy drawn by Amalarius that Christ is the sun who gives light to
the world for twenty-four hours. On a different level Honorius of
Autun writes 6 that the candles indicate the number of Gloria's which
are omitted during the whole of Tenebrae. This negative kind of symbol-
ism is echoed by both Beleth7 and Durandus.5 The latter adds that the
Gloria's are not said because Christ is lying in the tomb. 8
(i) 23 lights. The twenty-three candles lit at Worcester9 may have

been placed on a hearse designed for twenty-four lights. A twenty-
fourth candle may have been the light which guided the monks into the
cathedral for the night office. If the twenty-three candles did stand
1 HBS 26 p.}08. 2 Ordinal p.146.

J HBS 37 p.132.
4 ~ Use~ Lights p.145. (The writer's italics.)
5 Rationale VI.72 p.331. 6 Gemma Animae (PL 172.665).
7 PL 202.105D. 8 g;;-;lso Appendix 1.
9 Antiphonary p.62.
41
for the twelve prophets and the apostles without Judas 9 that fact has
not been recordedo
(j) 15 lights. The use of the fifteen=candle hearse mainly in re=

ligious houses strongly suggests that this number emerged from a mon=
astic milieue The earliest evidence for this candelabrum comes from the
Monastery of St Benigna at Dijon in the eleventh century; but the use
of fifteen candles at Tenebrae is almost certainly much older. The num=
ber is attested in the Customary of Cluny which was compiled by Ulric
who died in 1o86. Cluny had been founded in AD 910 to counteract the
0
temporary relaxation of Benedictine discipline'; and the great emphasis
which Cluny placed on the choir office and on elaborate and well=executed
liturgical observance created an ambience conducive to changes in ritual.
From such a milieu the fifteen-candle hearse may first have emerged.
The success and rapid spread of Cluny's liturgical influence as well

as her political power were in part the result of her reforming zeal,
and in part the result of the number of monasteries which owed allegi-
1
ance to the mother house. The Monastery of Farfa was influenced by
Cluniac ideals during the time of Abbot Hugh (AD 997-1038); and the
long-established Benedictine houses, Corbie and Chezal-Benoit in France,
Monte Cassino, and St Paul's in Rome were all affected by Cluniac pract-
2
ice and all adopted the use of fifteen candles at Tenebrae. The
Grammontines with their sixty houses in France seem to have been influ-
enced by Cluny in this respect. It would therefore appear that the
three French cathedrals of Chalon-sur-Saone, Laon, and Uzes had adopted
monastic usage in the Middle Ages (Table 3). At Paris where Gallican-
influenced service-books continued to be used after the sixteenth cent-
ury, the use of fifteen candles may well antedate the Council of Trent.•
There is some uncertainty as to when the use of fifteen candles

first entered the Roman rite. Cluniac influence in this respect during
the pontificate of Pope Gregory VII at the end of the eleventh century
may be discounted; for we know that in the Cathedral of St John Lateran
in the middle of the twelfth century only twelve candles were exting-
uished at Tenebrae.3 Medieval Roman breviaries did not specify the
1 At the zenith of Cluny's influence in the mid-twelfth century there

were 314 dependent houses.
2 Cluniac reforms were also introduced at Subiaco and at Sta Maria on
the Aventine.
3 OEL pp. 45-6.
• Roman influence, however, does seem more likely.
42
~umber of candles to be used 9 possibly because ritual variants existed
within the Roman Church before a semblance of uniformity was achieved
largely through the labours of the Franciscan friars. They popularised
the ceremonial of the papal court~ which they both admired and adopted
as their own rite. Since Raymo's ~ Breviarii of the mid-thirteenth
century does not prescribe a specific number of candles, it is likely
that none was mentioned in the breviary of Pope Innocent III (1198~1216)~
'I
upon which Haymo 1 s service~book waa based. · The earliest documentary
evidence for the use of fifteen candles at Rome ie in fact the definit=
ive Breviarium Romanum of 1568, although the practice of lighting that
number of candles at Tenebrae was probably by then well-established.
Church/ Source
Commentator
Corbie c.1000 DAMR 3.12 p.122,M1145

St Benigne,Dijon - 11 c ~ 3.12 p.122,M1150
Cl~ 11 c PL 149.657B
Farfa 11 c PL 150.1197B
St Paul, Rome 11 c ~ 4.22 p.124,M1041
Grandmont c.1100 ~ 3.12 p.122,M1184
Beleth 'C.118o PL 202.105D
Chalon -sur-Saone - 1226+ ~ 4.22 p.81,M67
Durandus £-1280 Rationale VI.72 p.331
Laon 13 c" ~ 4.22.1 p.81,M156
Fleury 13 c ~ 3.12 p.122,M1186
Uzes 14 c ~ 4.22.1 p.81, M346
St Vincent, Laon 14 c ~ 3.12 p.122,M1164
St Mary, York c.14oo Ordinal, HBS 75 p.271
Rome - 1568 Breviary I p.445
Camaldoli 1634 Cerem. p.56
Besanion 1682 Cerem. p.263
Braga 1724 Breviary p.295
Ales 1758 Breviary p.272
Paris 1778 Breviary (fY) p.280
Coutances 1825 Cerem. p.311
Verdun 1832 Cerem. p.293
Table 3. The Use of Fifteen Lights at Tenebrae
(k) 14 lights. Mart~ne merely mentions that this number was used; but
2
he gives no details of time or place.
1 Van Dijk II p.41. On page 37 above we referred to the fact that

\
Martene was somewhat nonplussed as regards the number of lights
that were used in the Roman rite.
2 DAER 4.22.2 p.81.
1 2
(l) 13 lights. ~vidence for this number comoa from Tongras~ Paris~
3 4 5 /
Angers 9 Seville~ and Roime. D~ Moleon also mentions tho use of ~hir=
teen candles 9 but does not identify any churcheeo 6 This number rnlmoot
certainly would have suggested Christ and the Apostles, or Mary and the
Apostles as at Seville. 7 At Tongres the six altar-candles were included
in the thirteen, no distinction being made, apparently, between them and
the remaining s®vea within the context of Tenebrao.
(m) 12 lights. The testimony of Beloth 9 Sicardus 9 and Durandun

would suggest that the use of twelve candles at Tenebrae was more wide=
spread than the three known instances would allow us to believe. These
8
were at Rome, at Le Mans,5 and at St Vedast's Abbey, Arrase9 How they
were arranged is unknown. At Rome they burned before the Image. The
possibility that at Arras there were originally twenty-four candles on
two hearses, as at Farfa, and that the one referred to in the ordinal
is the survivor that continued to be used at Tenebrae, are interesting
and likely suggestions, but ones which in our present state of know-
ledge cannot be substantiated. Beleth states simply that the twelve
10
candles aymbolise the twelve apostles; whilst Sicardus likens the
11
extinguishing of the twelve candles to the scattering of the apostles;
12
as opposed to Durandus who refers to the apostles' loss of faith.
St Bruno of Segni's reference to the twelve apostles quorum doctrina
fugatae ~ tenebrae 13 and to the twelve candles which represent them
may be a reference to Tenebrae in the eleventh century. The context
is certainly in favour of this.
(n) 11 lights. The only known instance of this number occurs at

St James' Monastery in Li~ge. 14 Here the extinguishing of the candles
would seem to symbolise the desertion of the apostles after the arrest
of Jesus in Gethsemane.
(o) 9 lights. The lighting of nine candles for Tenebrae is attest-

5 10
ed only at Nevers, though its mention by Beleth may well suggest that
1 Ordinary pp.150-51. 2 1662 Cerem. p.337.

3 1731 Cerem. p.223. 4 Doblado p.284.
5 Heuser p. 228 • 6 Voyages p.298.
7 For the hearse, see p.61. 8 ~ p.45 ••
9 HBS 86 p.156. 10 PL 202.106A.
11 Mitrale (PL 213.298D). 12 Rationale VI.72 p.331.
13 'By whose teaching the darkness has been dispelled.' PL 165.1100B.
14 Ordinary p.20.
• These burned before the maior imago.
44
this number was use~ i~ other churches as wello According to hi@ the
1
numbor nine symbolised the human race 9 cut off by sin from tho nine
orders of angels and excluded from the true lighto One candle may have
been put out at the conclusion of each nocturn, and six during Lauds.
(p) 7 lights. The use of seven lights at Tenebrae is attested by

2
Beleth~ who compares the putting out of the lights to the gifts of the
Spirit which were almost extinguiohed in the hearts of the disciples 9
and by Mart~ne.* Both writers may well be referring to churches which
either had formerly used seven sanctuary lamps at Tenebrae in addition
to other lights, or continued to do soe Presumably all seven lights
were extinguished during Lauds, as in the Carmelite rite.
(q) 5 lights. In the Carmelite rite until 1955 five candles were
lit at the beginning of the night office. This practice had remained
unchanged since the fourteenth century. 3
In some churches the number of candles to be lit was not specified

and could in theory be limitless.
4 Rupert of Deutz, writing c.1111,
refers to 'numerosa luminaria' at the start of the night office. 5 These,
he declares, are the saints who foretold the coming of Christ and who
were murdered. Beleth and Durandus both mention that in some places
6
there is no fixed number. This, says Beleth, allows auy man or woman
to make an offering of a candle to be lit at Tenebrae. These candles
are the prophets and the saints.
It would seem that from the establishment of their order, the Cist-
ercians neither used a hearse at Tenebrae nor extinguished the lights
one by one. Instead a single lighted candle was placed on the first
step of the choir. This was in keeping with their strict form of mon-
asticism and the austere character of their liturgy. The lighting of
the single candle recalls the period when only one lamp was lit for
reading at the night office of Holy Saturday. 7
1 In another context Honorius of Autun identifies two separate groups

of those who comprise the human race: (i} patriarchs,prophets,apostles,
martyrs,confessors,monks,virgins,widows,marrieds; (ii) lay,doorkeepers,
readers,exorcists,acolytes,subdeacons,deacons,priesta,bishops.
(Speculum, PL 172.1182.)
2 PL 202.106A. ~ ~ 4.22.2 p.81.
3 Ordinary p. 162.
4 The 1492 Ordinary of Li~ge refers to plures (np).
5 De Div.Off. V (PL 170.148C).
6 PL202.106D and Rationale VI.72 p.331 respectively.
7 See pp.11-13.
Chapter Six
(i) The Origin £! the Hearse
The evidence for the earliest form of hearse is monastiee In the

1 2
eleventh century at Fruttuaria and Gembloux we find that twenty=five
candles were placed on a wooden stand (lignum) behind the altar; whilst
at Cluniac Farfa3 and St Paul's Monastery in Rome
4 fifteen candles were
lit in instrumento lignorum and placed in front of the altar. It is
not wrong to translate both lignum and instrumentum liguorwn as 'wooden
stand' as Dendy does (p.146); but it is unwise to overlook the poasib=
ility that the two terms refer to stands of differing construction. The
singularity of lignum and the plurality of lignorum do suggest that a
difference existed. Moreover,it is wrong to suppose that continental
monastic practice was everywhere uniform and to ignore variations in
three important aspects of the hearse viz. the type or shape of the de=
vice, its position, and the number of candles it held. Let us consider
whether auy differences can be detected.
(a) Lignum. The singular form of the word may give us some clue
about the design of the device. The Customary of Sigibert informs us
2
that the lignum was especially made for use at Tenebrae. Whilst more
than one piece of timber may have been used as supporting members in
its construction, the section that held the candles was probably one
length of wood and almost certainly horizontal. Contemporary evidence
from York5 seems to confirm this supposition. For we read that the
middle candle of the twenty-five lit for Tenebrae stood higher than the
others. This suggests that the bases of all the candles including the
central one rested on the same level. If we accept that the candles
were displayed in a line on a length of wood, it would give further
credence to our theory that the lignum was a development of the choir
beam (or beams) on which the lights had once been placed.
6 At Fruttuaria
and Gembloux the lignum was placed behind the altar. If the altar stood
next to the east wall, the device must have rested on a ratable, and of
1 Albers IV p.49 2 Albers II p.90. 3 Albers I p.46.

4 Mart~ne, ~ 4.22.8 p.124
5 Breviary P•375· 6 See Scheme 2, p.28.
46
1
necessity must have been elongated 9 as in Figure 1.
(b) Instrumentum lignorumG It must bn admitted that this phraoe

could equally refer to the wooden device for holding or
twenty~four
twenty-five candles which we referred to in the preceding paragraphv if

the use of the plural noun lignorum denotes both the supporting timbers
and tho horizontal length of wood that held the candles. If, however 9
the meaning of lignum is restricted to the one plank or length of timber
on which the candles rested 9 it would follow that the occurrence of tha
noun in the plural number denoted two or more such lengths of wood. The
twenty-four candles may have been mounted on three (or four 2 ) separate
and parallel lengths of wood on the same plane (Figure 2) or constructed
in tiers (Figure 3), as in contemporary frames for votive lights, with a
twenty-fifth candle somehow incorporated into the arrangement. Their
three-dimensional shape, as it were, would have required them to be
placed in front of or to one side of the altar. Regrettably for our en-
quiry the existence of these twenty-four-light candelabra, though likely,
must remain suppositional, since the two recorded instances of the ~
strumentum lignorum attest the use of fifteen candles.3
Fig.1
Fig.2
Fig.3
1 Altars began to be placed against the east wall of the church from
the sixth century onwards (KlausEr p.100). Even if the altar had
stood forward of the east wall at Fruttuaria and Gembloux, the cand-
elabrum used at Tenebrae would almost certainly have comprised a
single length of timber (Figure 1). Twenty-four candles behind the
altar, arranged in three parallel rows, seems less likely (Figure 2).
2 Depending upon whether a third of the candles were extinguished
during the three nocturne of Matins, or whether a quarter of them
were extinguished during Lauds.
3 For a discussion of the fifteen-candle hearse, see pp.52-57·
(ii) ~ Development of the Hearse
It is fortunate that thero has survived from the lost Customary

of St Benigne,Dijon, which was compiled in the eleventh century, a
reference not only to a fifteen~light Tenebrae candelabrum but also to
its shape and appearance :
XV candelae in modurn Pyramidl~ ante StiD.Ctt:un ta.ltartj a.ccend'Wltur.

1
The date of ita attestation (11th century), the number of lights, and
its position in relation to the altar strongly suggest that the device
upon which these candles were placed merited the description of instru=
mentum lignorum. This in turn raises the interesting possibility that
the candelabra at Farfa and St Paul's in Rome, which we referred to in
the previous section, were also triangular in shape. Exactly what is
meant by in ~ ?yramidis is not immediately clear; Martene does not
elaborate the phrase from the passage he has quoted. Given that the
frame was of triangular shape, the candles must have been displayed in
one of three ways in order to give the on-looker the impression of a
pyramidal arrangement (Figures ~,_2, and§).
Fig.6
~Though the candles in all three figures could be said to be disposed in

the form of a pyramid, one could claim with confidence that the success-
ive decrease in the number of candles in an upward direction, as exempl-
ified in Figure 4, corresponds more closely to the usual interpretation
of the phrase in ~ F7ramidis, if one were to observe the candles by
standing immediately in front of the candelabrum. However,practical
problems arise in the use of a two-dimensional candelabrum of triang-
ular construction. In the arrangement, depicted in Figure 4, where six
internal candles are mounted on the three lower cross-bars, there is a
danger that the flames of these candles will set alight the cross-bars
immediately above them, unless the frame is large enough to provide
1 Mart~ne, ~ 3.12.2 p.122 (M 1150): 'Fifteen candles are lit before

the sacred altar, arranged in the manner of a pyramid.'
48
sufficient clearance between the flame and the underside of the cross~
bar~ or unless the horizontal sections are made of metal. The practic~
ability of the lattor 9 however~ is doubtful; cross~bara made of metal

make unsuitable supports for candles if they are subjected to the heat
from naked flames directly below. We have no indication of the size of
these candelabra in the eleventh century; but we have noted that tho
frames were made of wood. However 9 if the candles were mounted on a
throe=dimensional frame (Figure 7) or a triangular=shaped devieo studd=
ed with nails to hold the eandles and sloping in a backward direction
(Figure 8), then not only ie the problem of damage to the frame by the
candles overcome, but the very structure of the candelabrum takes on
more of the shape and appearance of a pyramid.
Figure 7 Figure 8
The arrangement of lights, as shown in Figure 6, is illustrated on

page 265 of the first edition (1600) of the Caeremoniale Episcoporum,
which contains the official procedure for Tenebrae according to the
Roman rite. We believe that this disposition of lights represents both
a development and a simplification of the hearse which was used at Dijon
in the eleventh century, so modified that the device of three dimensions
has been reduced to two, and the candles have been placed on the two
angled sides, because a display, as shown in the hypothetical Figure 5,
would almost certainly have caused damage to both frame and candles, as
we have already suggested. The mounting of the candles on the outside
edges of the frame (Figure 6) was obviously done for reasons of safety
and to facilitate their arrangement. For if iron spikes, rather than
cups or candle-sockets, were used to affix the candles to the frame, a
straight vertical nail on the outside edge of the hearse would provide
a simpler candle-spike than a nail protruding horizontally from the
face of the hearse to clear the frame and then bent upwards like a
right-angled cup-hook to hold the candle.
It is not known when the triangular hearse was first used in the
49
Roman rite. On page 42 wo stated that there was some uncertai~ty
about the introduction of tho fifteen=~dlo hearse into Rome. Little

can be added to that statement othor than that its use may have entered
the papal ceremonial either through the influence or because of the
proximity of St Paul's Honastery in Rome.
We have so far argued in favour of the development of the triangular

hearse from tho pyramidal device as typified by that used at Dijon. Two
other possibilitiosi however 9 remain to be considered; though it must
be added that the lack of corroborative evidence prevents us from draw=
ing firm conclusions about either. (i) In view of the limited nature
of the evidence upon which we have based our theory, the converse of our
supposition may equally be true viz. that the pyramid-shaped candelabrum,
like that at Dijon, is an elaboration of the triangle; and it could be
argued that the pyramidal device did not survive, partly because of the
additional effort and workmanship required in its construction, and part=
ly because the sombre aspect of Tenebrae discouraged liturgical extra-
vagance. Alternatively,both types of hearse may have evolved in semi=
isolated liturgical milieux from a common ancestor. Figures 9-11
illustrate how the development of the triangular or pyramidal candel-
abrum from the horizontal hearse may have been induced by the superior
elevation of the central candle.
lllll'f'lllll
Fig.10 Fig.11
(ii) It is just possible that the great seven-branched candlesticks

1
which were popular in the cathedrals of Europe during the Middle Ages
may have provided inspiration for the shape of the much smaller Tene-
brae hearse on account of their quasi-triangular display of lights.
(iii) ~ Hearse of Twenty-~ Lights
The use of the twenty-four-light hearse at Tenebrae derives from

the liturgical tradition known to and recorded by Amalarius. We have
2
already noted that the hearse of twe~ty-five candles was a simple
1 Zarnecki p.134. 2 See p.39.
50
variation of the other candGlabrum an additional candle included
~dth
1
for practical or symbolic reasons. We have already suggested that
they were displayed iu linear formation and stood on one framco 2
Grancolas refers to the great candlestick with many branches which
stood near the altar in some churches and associates its use with the
extinguishing of lights at Tenebrae.3 Artistically~wrought metal cand=
elabra were to be found in the cathedrals and more prosperous monaster=
ies; and the evidence suggests that they were usually of the seven=
4
branchod variety, the menoraho However1 there is no reason to believe
that these ornate artefacts were designed to hold as many as twenty=four
candles, though their position to the right of the altar (as you face
it) may have influenced the siting of the twenty=four=light hearse; and
at Tongres the wooden hearse was even superimposed upon a bronze cand~
elabrum. 5 The possibility should also be considered that twenty-five
candles were mounted upon a triangular candelabrum in the same way as
fifteen : one at the apex and twelve down either side. Such a frame
would have been considerably larger than its fifteen-light counterpart.
With twenty-four lights there are problems of presentation. Either
there are two candles at the apex, or there is an asymmetrical displaye
Neither arrangement would have been visually satisfactory. Nevertheless,
6
according to the Gilbertine Ordinal the last of the twenty-four candles,
which was to be no larger than the rest, was placed 'at the top'.
Howeve~ there are two reasons for believing that the twenty-four
candles may not always have been displayed in this way. (i) In Chapter 5
we saw that it was important for the number of lights to be extinguished
to be divisible by three; 7 and in Chapter 4 we stated that at Poitiers
these lights were arranged in three rows.7 The suggestion is worthy of
consideration that when the Tenebrae-lights became confined to the sanct-
uary, the triple line of lamps was retained in the form of a frame,
Dimilar to those depicted in Figures ~ and 1• which stood in front or
to one side of the altar. If such a frame was used, it is more likely
to have been of the type shown in Figure 3, since this device provides
a better view of all the candles especially if it stands on the dais for
the altar. The variation from church to church in the number of candles
1 See PP• 46-7.

2 In some churches they may have stood in individual candlesticks, or
have been set upon a circular candelabrum of the kind similar to the
two candelabra in Great Budworth Church, Cheshire.
3 Commentarius p.296. 4 Zarnecki pp. 134-5.
5 Ordinary p.151. 6 HBS 59 p.31.
7 Pages 34 and 24 respectively.
51
used meant that the three rows misht each contain eight~ nine, thir~
teen 9 or as many as twenty~four lights. In churches where the full com-

plement of lights was extinguiohed during Lauds as well as £~tins 9 th®
division of candles into three groups was no longer relevant.
(ii) According to the thirteenth=century Customary of Fleury the

1
candles were arranged super pronas. The phrase is otherwise unknown.
One possible interpretation involves regarding fronas as the accuaative
@f the otherwisemunrecorded plural noun pronae, formed from the adject=
ive pronus -~ -~, 'inclined 8 or 'leaning forwards',and used as the
technical term for the banks or tiers of candles on a frame such as that
illustrated in Figure 3. For in a way the candles do incline and lean
forward from the altar. The existence of a stepped frame of this type
could easily have suggested the idea for the pyramid at Dijon; and, if
so, a frame of this construction would antedate the pyramidal device.
For the use of twenty-four lights at Tenebrae is anterior to that of
2
fifteen by a century and a half. It would.therefore,be not unreason-
able to conclude that the construction of the pronae at Fleury was aim=
ilar in appearance and shape to the pyramid used at Dijon.
(iv) ~ Origin 2f ~ Fifteen-candle Hearse
We have no clear evidence to explain why the fifteen-candle hearse

emerged in the tenth century from a monastic milieu, at a time when the
use of twenty-four lights was recommended by the influential Liber
Officialis of Amalarius, and prescribed by ~' which was used widely
in Gaul and Germany. A clue to its possible origin is to be found in
the eleventh-century Customary of Farfa, which gives these directives:
Coena domini quinta feria, ante Nocturnos quindecim in instrumento

lignorum, ante maius altare XXX sint accensae candelae quot psalmi
sund (!E:g) imponendi. Et remanentibus triginta psalmis, ante noct-
urnas dicatur quindecim su~ilentio post ternas orationes facta.
Signa sonet secretarius sicut solitus est. Quibus dictis sonant-
ibus signis, postquam dimissis fuerint aebdomadarius incipiat
antiphonam Zelus domus tuae. Et statim extinguatur una candela
huic, alia inde, sicque fit omnibus psalmis incipientibus.3
1 It is not clear how many lights were used at Fleury. In the passage
in question (~ 3.13 p.123), Martene mentions fifteen; elsewhere
(~·3.12.3 p.123), twenty-five. He is either referring to two
documents relating to different periods; or he has made a mistake.
The discrepancy,however, does not alter the fact the candles were
arranged super pronas.
2 Amalarius, AD 832: 24; Farfa, _£.1000 and Cluny, 11th C.: 12·
3 'On Maundy Thursday before Nocturne (Matins) let thirty(Cont.p.53.)
52
At th@ beginning of Psalm 148 Qll tho CQndloo have been oxtinguiohod
except one. tfuen the antiphon of tho Benedictus, tho Traditor, begins,
tho last candle is put out.
These rubrics from the above=mentioned customary may be summarised

as follows :
i. Thirty candles are lit, and fifteen are placed on a wooden
stand in f~out of the alta~.
ii. Thirty psalms are said, fifteen silently, fifteen aloud.
iii. A candle is extinguished befo~e each of the psalms of Matins.
iv. At the beginning of the twenty=ninth psalm, twenty=nino candlos
have been put out.
v. After the Traditor has been sung for the first time, there is
darkness in the church •
The rubrics also tell us that the sacristan, who presumably calculated
the length of time sufficient for the silent repetition of each psalm,
1
informed the monks by means of a noise at the conclusion of each period
2
that they should commence the next psalm. The sacristan, however, did
not indicate which that psalm was. The fifteen psalms said silently
after the triple prayer were the gradual psalms3 (Psalms 119-1}4) which
1 Quibus dictis sonantibus signis. The somewhat ambiguous Latin in-

dicates that a single noise was made at the end of each psalm rather
than several repeated sounds at the conclusion of all fifteen psalms.
The nature of the noise is not specified.
2 Mart~ne records an alternative method of regulating the silent re-
petition of the gradual psalms. According to the now-lost Customary
of St Benigna, Dijon, after the saying of the triple prayer, the
fifteen gradual psalms were said in silence; and as each monk fin-
ished his psalm, he leaned forward across his stall. When all the
monks had assumed this position, it was time to proceed with the
next psalm. (~ 3.12.2 p.122,M 1150.)
3 Graduales psalmi or cantica graduum. They are mentioned in Lanfranc's
Decrees, an ordinal of Corbie, the Use of Bee, and in the Customaries
of Cluny, St Benigna, Fleury, St Denis, Laon, St-Germain-des-Pres,
Lyre, and the Compendienses (Martene, DAMR 3.12.2 p.122); and also
in the Ordinal 2! ~ Mar:r 1s, ~ (HBS 75 p.271.).
Continued from p.52.

candles be lit in front of the high altar, one for each of the
psalms to be said - fifteen on a wooden frame. There being thirty
psalms to say, let fifteen be said in silence after the three
prayers and before the start of Matins. Let the sacristan make an
audible signal in the usual way after each psalm is said. When all
fifteen have been said, let the priest who is on duty for the week
begin the antiphon The zeal for your house. Let a candle be immedi-
ately extinguished 'Ono'ile"Side, and asecond (at the beginning of
the next antiphon) on the other side, and so on alternately at the
beginning of all the psalms.' (Albers I p.46.)
53
formed an introductory period of moditation between tho ontry of tho
monks into church and the start of Matins. The above~qnoted extract
from the Customary of Farfa leaves us with the impression that the cand=
les began to be extinguished only when the second group of fifteen
1
psalms, that is, those set for Matins and Lauds, began to be sung. Yet
at the Benedictus all thirty lights have been extinguished; and, as
only one candle io put out at the beginning of each psalm~ the other
fifteen in the first group must have been extinguished at some point
between the beginning and the end of the gradual psalms. Howover,
there is no indication from the text as to when this occurred.
Writing nearly a hundred years ago Hartmann Grisar, followed by

other commentators, suggested that the extinguishing of lights at Tene=
brae was a way of keeping track of the number of psalms which had been
2
sung. Ludwig Eisenhofer drew attention to the practice in the Circus
Maximus of indicating to the spectators the number of laps which had
been completed by the chariots.3 Later research has shown that the
method of counting
4 the number of laps in that stadium approximated more
closely than Eisenhofer realised to the way the progression of psalms
was noted at Farfa and almost certainly in other monasteries. Whilst
this method of counting at Farfa was obviously not borrowed directly
from the Circus Maximus, the notion of using a decreasing number of
markers for reckoning seems to have survived from the ancient world into
the Middle Ages. We therefore believe that the gradual extinguishing
of lights as a way of counting psalms was used, not at Matins and Lauds
as Grisar and others believed, but during the silent repetition of the
gradual psalms. For, whereas at Matins the antiphons, lections, and
divisions into nocturne marked the progress of the night office and the
various stages which had been reached, in the period of silence before
the service there was nothing to indicate which psalm was intended to
be repeated during a given period of silence in the event of a lapse in
concentration on the part of a monk through tiredness or inattentive-
ness; for the sound produced by the sacristan merely marked the con-
clusion of such a period. The number of candles that remained burning
would therefore indicate at a glance the correct psalm to be said at
1 Nine at Matins, and six at Lauds including the OT canticle.

2 Bugnini and -Brag& .p.123·
3 1Wie man bei den Zirkusrennen, urn dem Volke die Zahl der Umlaufe der
Wagen anzuzeigen, grosse holzerne Eier an dar Spitze der Spina nach
jedem Umlauf aufstellte.• (Handbuch I p.514.)
4 At one end of the central spine or barrier were seven marble eggs, at
the other a line of seven dolphins. At the end of each lap or half
lap an egg or a dolphin was removed. (Mannix p.12.)
that timeo The gradual loso of light during this period of preparation
adequately accounts for the extinguishing of the first group of candles
prior to the start of ~~tina.
Additional support for our argument can be gleaned from a closer

examination of that part of the text which states that fifteen candles
were placed upon a wooden stand (!g instrumento lignorum) and that
thirty were lit before the high altar. A superficial reading gives us
the impression that the candles on the wooden stand were intended for
Tenebrae and that the rest were placed somehow in front of the altar,
possibly in separate candleholders. On the other hand)if our theory is
correct that the first group of fifteen candles were extinguished one
by one as each of the gradual psalms was said, it would be necessary for
the candles to be arranged in such a way as to indicate to the monks at
a glance which psalm they should be repeating. It is obvious that the
candles would not be scattered at random in front of the altar; and one
might suppose that the instrumentum lignorum was used for this purpose.
However we have the impression that this device held the Tenebrae-candles.
I~ on the other hand)the wooden stand was used during the saying of the
gradual psalms, we must ask how the candles put out at Tenebrae were
then arranged. Crucial to the solution of this difficulty is the corr=
ect interpretation of ~ instrumento lignorum. If the phrase is trans-
lated 'on~ wooden stand', it follows that only one stand was used
and that the other candles were arranged in an unknown way on free-
standing candlesticks. However, if the phrase means 'on~ wooden stand',
the entire sentence becomes capable of quite a different interpretation.
For, bearing in mind that the Latin of the whole passage is at times
both grammatically and orthographically incorrect, it is possible that
'fifteen candles are lit on a wooden stand' should be interpreted 'a
wooden stand is used for lighting groups of fifteen candles' or, express-
in another way, 'fifteen candles are lit on each wooden stand'. It would
then be reasonable to assume, in view of the uncertainty regarding the
exact function of the wooden stand, that two wooden stands were used,
one for the gradual psalms, and one for the psalms of the night office~
1 It is not for us to suggest how the compiler of the customary should

have expressed more clearly the existence and use of two stands. For
the passage contains a number of linguistic errors, so that a careless
turn of phrase should not come as a surprise to us. Nor can we be
sure that the writer would have used either the phrase ~ utroque
instrumento or !e guogue inetrumento, either of which we maintain is
what the context demands. XXX candelae in duobus instrumentia could
imply uneven distribution; XXX candelae~n instrumentis lignorum
might suggest more than two stands.
55
It is our belief that at th0 beginning of the monastic day at Farfa
two candelabra~ each of fifteen lights~ stood in front of the altar.
That to the right of the altar may have held the candles extinguished
1
during Matins and Lauds. The desirability of positioning the lights
in a convenient and visible place for the purpose of keeping a tally of
the number of psalms said silently must surely have been an important
factor in locating the candelabrum near to the altar. For not all of
the church lights may have been in full view of those positioned in the
choir-stalls~ if the lights had been used for this purpose; and the
necessity of an upward glance in the event of the use of lights dis=
played on choir~beams was likely to have caused inconvenient distraction.
In the interest of a symmetrical display of light both candelabra are
likely to have been identical. Since one of these served to indicate
what stage the silent repetition of the gradual psalms had reached, an
arrowhead or pyramidal-shaped stand had obvious advantages over a frame
on which the lights were arranged horizontally in a row •
• • • • *
We will show in Chapter 7 that in the period before AD 1000 the

candles were extinguished at Matins in sympathy with the singing of the
psalms and the reading of the lessons. The greater the number of lights
extinguished, the greater was the scope for elaboration of ceremonial.
It is,therefore,somewhat surprising to find that in the ritual of Cluny,
with its emphasis on ceremonial splendour, there was a certain austerity
in the number and use of lights at Tenebrae. If, however, we accept
that at Cluny, where we believe the fifteen-candle hearse originated,
thirty lights were formerly used at Matins, as indeed they were at Farfa
and Poitiers, it is not difficult to see how, with the recognised cent-
rality of the psalms at the night office and the importance also of the
gradual psalms, one light may have come to symbolise and be assigned to
each psalm. With the convenient coincidence of fifteen gradual and fif-
teen Tenebrae psalms, it is likely that the thirty lights were divided
into two equal groups and placed on two separate candelabra. These
would almost certainly be identical since liturgy generally favours a
symmetrical arrangement.
A number of reasons may be advanced to explain why the arrangement
1 This may partly explain why in subsequent centuries the Tenebrae-

hearse was often located on the Epistle side of the altar.
we have suggeated is not atteoted either in othor ritoo or in subsequ=
ont centuries. (a) Not all monastorios employed this method of count-
ing the gradual psalms~ and perhaps oomo that did found it distracting.
(b) The extinguishing of the two groups of candles in close succession
would tend to attribute the same importance to the gradual psalms as
that which the office psalms held. Within the liturgical context in
question both groups served different functions, and were clearly of un-
equal status. (c) The repetition of the ritual act of extinguishing
fifteen lights would inevitably detract from the significance of that
act during Tenebrae. (d) Ther0 is no evidence to show that the gradual
psalms were said prior to the start of Matins in the cathedral tradition
or in a number of other religious rituals in which Tenebrae was observed
as an anticipated office. In this situation the use of two hearses
would have fallen into desuetude once the gradual psalms ceased to be
said, the candelabrum used at Tenebrae alone surviving.
( v) The Significance 2£ Fifteen Candles
In more recent times the fifteen candles have been understood to re~
present the wisdom of the centuries which was lost progressively from
1
the time of Moses onwards. In the Middle Ages they symbolised the
2
twelve apostles and the three Mary's. Durandus adds that the exting-
uishing of the lights symbalises the flight of the apostles and the fear
of the Mar,y's. 3 The same writer gives an alternative explanation accord-
ing to which fourteen of the candles signify the fourteen articles of
faith which were extinguished by the flight of the apostles, and the fif-
teenth stands for the death of Christ. The Ordinal of St Mary's, '!.2£!.,
whilst identifying twelve of the candles with the apostles, equates the
remaining three with the Law, the Prophets, and Christ.
4
It is somewhat strange why Sicardus, on whom Durandua drew for much

of his information, and who records the use of seventy-two, twenty-four,
and twelve lights at Tenebrae, does not mention the use of fifteen,
when, at the time that he was writing (~.1200), this number was well
established. Beleth writing some years before knew of it. Discounting
the unlikely possibility that Sicardus omitted any reference to it
1 Heuser p.227.
2 Beleth, PL 202.106A. It could be argued that this forced symbolism
is further evidence of the late and almost fortuitous origin of the
fifteen-candle hearse.
3 Rationale VI.72 p.331.
4 HBS 75 p.271.
57
because of personal disapproval, one can only guess either that he was
interested in Tenebrae primarily according to the cathedral tradition
(fifteen at this time being mainly a monastic number), or that he omitted
the information through carelessness 9 or that the omission is an error
on the part of a copyist.
(vi) The Posit:.l.on 2f the Hearse
(u) ~ tho right 2£ ~ ~. 1 According to the Roman rite th0

hearse was supposed to stand to the south of the altar on the Epistle
side. This is the position prescribed by the Caeremoniale Episcoporum2
of 1600; and the various editions of Fortescue and O'Connell contain
this regulation. 3 It is claimed that the hearse stood to the south of
4
the altar in imitation of the menorah in the Temple at Jerusalem.
Howeve~ the analogy was not strictly accurate in that the menorah stood
in the south-west corner of the hekal rather than to the south of the
-
altar. Surprisingly, there is no firm evidence before the end of the
sixteenth century for the placing of the hearse in this position; but
it is difficult to believe that on the publication of the Breviarium
Romanurn in 1568 the south-end position was a recent innovation. It may
have been placed there for centuries in some churches. Howeve~ with the
doubtful exception of St Vedast's Abbey at Arras, and the ambiguous
testimo~ from St Mary's, York, all the surviving evidence refers to
the position of the hearse behind, on top of, or in·front of the altar;
and all three positions occur in France up to the eve of the Revolution.
The Ordinal 2f §E Vedast's Abbey gives the following directions:5
Post completorium velum tollitur quod est inter duo altaria et

hercia ponitur cum duodecim candelis.6
It is not clear whether this veil served the same purpose as that at
St Mary's, York (p. 62); or whether it separated the altars from the
rest of the building, as was the custom in medieval churches in Lent.
1 From the point of view of the congregation.

2 P. 26'4. However. the line-drawing on the opposite page shows the hearse
on top of the altar apparently in a central position. It is just poss-
ible that because of the perspective the hearse only appears to be
perfectly central, but is actually slightly to the right of centre.
3 For example, 4th ed. (1932) p.303, and 11th ed. (1960) p.281. The
1682 Ceremonial of Besan~on places it at the Epistle side (p.263).
4 Heuser p.229. He does not enlarge upon the analogy.
5 HBS 86 p.156.
6 'After Compline the veil which is (hung) between the two altars is
raised and a hearse with twelve candles is put in position.'
Since both altars were exposed to view for the performance of the
officei it seems reasonable to suppose from the limited information
available to us that the rubric prescribes that the hearse should stand
either close to one altar or between the two altars.
The Ordinal 2! ~ Mary's, ~states that the hearse was placed

1
~ altari. It is true that tho English translation 'in the pre~
sence of 0 could be understood to mean~~ vicinity 2£ 9 and,therefore,

it would be possible to argue in this instance in favour of a position
to the right of the altar. On the other hand this interpretation is
perhaps forced, and a more likely rendering of ~ would be'in front
of'which conveys the usual meaning of the word 'face to face'.
(b) !g ~ ~ 2! ~ altar. The wooden stand (lignum) for the

twenty-five candles at Fruttuaria and Gembloux was placed behind the
altar (po46). Credence is given to our theory that the hearse at these
two monasteries was a wooden bar with one row of candles when it is re=
alised that from about the year 1000 it became the general rule for the
altar to be placed against the east wall of the church so that the priest
with his back to the congregation could celebrate Mass facing the East. 2
Any device placed on the gradine or ratable would of necessity be elong=
ated in design because of the restricted space. The placing of the
candles behind the altar is first firmly attested by John of Avranches,3
and the placing of them in this position survived in parts of France
until the Revolution, as is evidenced by the 1778 Breviary of Paris
which allows the choice of placing the hearse in front or to the rear
4
of the altar. The presence of candles on a gradine behind the altar
or on the altar itself virually precludes the use or even the presence
of the six altar-candles. Even when the hearse stood before the altar,
the six altar-candles, even if set out, were not lit. For in two Paris-
ian breviaries of the eighteenth centu~ and in the Romanised manual
6
for the Royal Chapel in Paris, the altar-candles are not mentioned.
(c) ~ the !!!!!· Evidence for the placing of the hearse upon
the altar is slender and vague. The above-mentioned illustration on
page 265 in the Caeremoniale Episcoporum would suggest that, in spite
of the directive to place the hearse on the Epistle side (p.264), the
practice of placing it upon the altar was not unknown. The statement
1 HBS 75 p.271. 2 Klauser pp.100-1.

3 ill·~ QU.~.52 (PL 147.48C).4 ~ ~ p. 280.
5 Of 1763 and 1778. 6 OSS p.211.
59
also by Grancolas that hearses were still placed super altaria in oomo
churches shows that the practico survived into the first half of the
1
eighteenth century.
(d) ~ ~ £! ~ ~· The earliest recorded instance of the

placing of the Tenebrae-candles in front of the altar occurs in the
eleventh=century Acta of the Cathedral Church of Rauen. 2 The frontal
position also feat~red in the monasteries at Dijon3 and Farfa4 in the
eleventh century 9 in Scissons Cathedral5and probably at St Mary 1 s 9 York6
in the twelfth century, and in the Gilbertine rite.? Sieur de Moleon 9
writing at the beginning of the eighteenth century, specifically ment=
ions the hearse at Rouen Cathedral as though its position was unusua1. 8
On the other hand he may simply be stating a fact about the ritual of
the cathedral; for Grancolas, writing several years later, observes
that the placing of the triangular candelabrum in front of the altar
was contemporary practice.* Earlier De Grassis had referred to this
central position; 9 but Bisso, writing a hundred years later, stated
10
that the hearse should not stand in front of the altar. The choice
of positions allowed at Paris (p.59) suggests that local usage in this
respect probably survived in other French dioceses which retained
their traditional Gallican ritual.
(vii) ~ Construction £1 ~ Hearse
No Tenebrae hearse in England is known to have survived from the

Middle Ages. Most of those used in parish churches would have been
11
constructed of timber to a simple design, as in more recent times.
Even without the iconoclasm of the Reformation, 12 the vast majority of
these wooden candelabra would almost certainly have not survived. Those
made for the cathedrals and more opulent monastic churches would almost
certainly have been of a more intricate and elaborate design, and very
likely fashioned of metal. Those in England in more recent times were
usually made of wood; and whilst the detail and ornamentation varied,
1 Commentari us p. 296. 2 PL 14?. 168C.

3 Martene, DAMR 3.12.2 p.122. 4 Albers I p.46.
5 Martene, DAER 4.23 p.137. 6 See p.62.
7 The Ordinal states that it should stand on the altar steps (HBS 59
8 Voyages p.298 9 .!2! Ceremoniis II fol.123./p.30).
10 Hterurgia I p.148. • See foot of p.61.
11 The writer is indebted to Mr D.Searle of Trowbridge for the descrip-
tion of the wooden hearse with its metal candle-holders and drip-pans
which was made by his father for use in his parish church in the
early years of the present century.
12 For the destruction of church furniture, see Peacock pp. 106,163,164.
60
the basic shapo remained a triangular frame of timber BUpported on a
tall staff and having fifteen spikes or socketD for tho candlesv ac was
officially proscribeclo
1
A ~riangular hoarse io also attested at Auxerre2
and in a Capuchin ceremonial.} Commenting on the shape Colti observed
that the hearse symbolised the Trinity and the single foot upon which it
4
rested expressed the unity of God. The reference by Dendy to the hearse
at Stanford=in=the=Vale, which was suspended by means of a rope 9 is sure=
ly inapposite. 5 The entry in the churchwardens' accounts relates to a
rope for the hearse. It is much more likely that it is the funeral
hearse (the wooden or iron frame holding the tapers, which was placed
on a coffin in church) that is here intended and, like the font covers
in some churches, could be raised by means of a to a pulley
~pe ~ttached
in the roof. Moreove~ the movement and sway of a Tenebrae-hearse, sus-

pended by a rope, not only might prove an unwanted distraction for wor-
shippers at this solemn service, but might make difficult the extinguish-
ing of the candles, an action which required careful eynchronisation.
The hearse at the Birmingham Oratory is heavily baroque and ornately

carved with the emblems of the Oratory. Perhaps the most elegantly-
designed hearse in England is that at Downside Abbey. It is fashioned of
wrought iron and stands about nine feet on a tripod base. It has a pro=
fusion of leafy and grape-like ornamentation around the central shaft to
which the seven pairs of corresponding candle-holders are linked by curv-
ed strips of iron reminiscent in a way of the threads in a spider's web.
From the outstretched arms there are suspended by chains, on the left,
the nails of the Crucifixion and, on the right, a crown of thorns in iron.
'Its fine and yet sombre design is most appropriate to the occasions on
which it is used. It is large but well proportioned to the height of the
6
Abbey Church.' At Seville a brass candlestick, 15' to 20' high, stood
between the altar and the choir. Triangular in shape it had twelve
candle-sockets, six on either arm of the candelabrum, next to each of
which was a figurine of an apostle. The Virgin Mary was represented at
its apex, adjoining the holder for the thirteenth candle. 7 The use of
1 Fortescue and O'Connell (11th edition) p.281 note 3.

2 1736 Breviary p.243. 3 1775 Ceremonial p.93.
4 Dictionarium II p.97.
5 Antiquary XVII (1888) p.119, cited in The ~ of Ligl1ts p.146.
6 DOm A.James, ~ Stocy 2!' Downside Abbey Church, Downside, 1961 p.71.
7 Doblado p.284.
From page 60
• It stood in this position at Trier (Ordinary p.486), at Auxerre (1736
Breviary p.243), and at Ales (1758 Breviary p.2?2).
61
ornately=executed candelabra is also attested by Bauldry in the
1
eighteenth century.
In some monasteries in the Middle Agos the hearse acquired embell=

ishments. It may have been in imitation of the drapery which surround=
ad the chapelle ardente or candle=bearing funeral hearse that tapestries
or embroidered cloths covered the lower part of the candelabrum at
St Mary 0 s~ York and at l'Jorcestor; 3 or it may hav® been done to match
2
the altar=frontal? especially if the hearse stood in front of the altar.

rfuatever the reason for its presence elsewhere, at St Mary's, York this
cloth was used as a screen to hide from view the acolyte or sacristan
whose duty it was to extinguish the candles.
4
At Tongres seven candles were affixed to the top of a spear. This
may have been the same device that held the triple candle; for this
shaft with its three lights was at times known as a hearse. (See p.63.)
(viii) ~ ~ hearse5
Further support for our theory that the Tenebrae-candles were at one
time arranged on wooden frames similar to those depicted in Figures 1, ~'
and l on page 47 comes from the traditional name for the candelabrum,
the hearse, a word derived, like the French ~~ from the late Latin
hercia6 which itself is a corruption of the classical Latin (~)irpex,
(~)irpicis meaning 'large rake' or 'harrow'. In addition to the occur-
rences mentioned below (note 6), the word is also found in the Ordinal
2! ~ Vedast's Abbey c.1300; 7 and it is to be supposed that where the
device is not referred to by name, it was generally known by this term. 8
It is generally held9 that the Tenebrae-hearse received its name because
of the similarity of the device, in both shape and appearance, to the
instrument of tillage which bears the same name; and we noted above on
page 28 that the beams (hirpices), referred to by Grancolas, received
1 Manuale p.167. 2 HBS 75 p.271.

3 Antiphonary p.62. 4 Ordinary pp.150-1.
5 Officially it was known as the candelabrum triangulare (Sacrarum
Caeremoniarum fol.155; Bisso I p. 148; Grancolas p.296).
6 The first recorded use of the word ia-Eag-l:aD.d·occurs in the statutes
of the Second Synod of Exeter in 1287 (Powicke and Cheney Vol.II pt 2
p.10o8), unless the reference in the thirteenth-century Customary of
St Augustine's Monastery, Canterbury (HBS 28 p.274) is older.
7 HBS 86 p.156.
8 Howeve~in the Statutes of Lincoln Cathedral (Bradshaw and Wordsworth
II p.131) it is called cratis tenebrarum, 'a harrow for Tenebrae'.
9 Fortescue and O'Connell (11th edition) p.281.
62
~heir names from the protruding spikes which caused the Gpars of wood
1
to ~esernblo harrows. The three=sidod candelabrum~ familiar from the
Roman was also referred to as a hearse, since harrowo of triang~
rite~
2
ular shape were also used in former times. In It~ly thoae hearses
were known by the clergy and sacristans as saette, 'bolts•, from the
representation of lightning by artists. 3
In English tho word hearse formerly had three other upplications 9

all within the context of the provision of light in church. (i) It re=
ferred to the wooden or iron frame placed over a corpso or coffin to
hold the funeral candles. Within the same context the word is now
applied to the vehicle which bears the deceased person. (ii) In an
inventcr.y of Christ Church, Canterbury dated 1563, mention is made of
a hearse of three lights for the carrying of the new fire on Holy Satur-
day.4 Here it signifies the triple candle.5 (iii) The English anti~
quarian Edward Peacock describes a triangular frame of wood, called a
hearse, which was suspended from the roof by a cord or chain. Across
the frame ran three bars; and at the points where the bars crossed
there were sockets for holding candles. 6 Its purpose, apart from pro-
viding light, is unknown. It is unlikely to have been a Tenebrae-hearse,
since in such a device the bars intersect each other only at three points.
1 That is, a harrow comprising a single wooden board with protruding

spikes.
2 In heraldry harrows are represented as triangular objects, having
three transverse bars into which tines are fixed.
3 Armellini p.354. Saetta also signifies 'arrow'.
4 OED under hearse. 5 See Part III Chapter 3.
6 Andrews pp.214-15.
Chapter 3even
THE EXTINGUISHING OF LIGHTS
(a) Before AD 1000. Seven of our ten sources for Tenebrae in the
earlier period stipulate that 9 regardless of the number used~ a third
of tho lightGJ of the church ohould bo extinguished during the course
of each nocturn of Matins = hence 9 ao we have seen, the numbers that
1
have come down to us are all divisible by three (p$34). Griaar cites
2
Durandus' atatement that in some churches the lights were put out in
three stages and understands 'three stages' to signify ~ groups of
lights. He believes that this method of extinguishing the lamps pre-
ceded that of quenching them gradually one by one. Durandus gives no
instance of a church where the former method obtained; and three3
early sources, including the earliest, Ordo 26, record that they were
4
extinguished 'paulatim'; whilst Poitiers statea that it was done
'singulatim'. Amalarius also attests the decrease of light by degrees.5
We have noted elsewhere
6 that Poitiers records that the lights begin to
be extinguished at the western end of the church so that, with the
lights arranged in three rows running in an east-to-west direction, the
last remaining lamps will be still burning in the choir, while the last
lessons of Matins are being read. We believe that the lights attested
in other ordines, and especially in Ordo 29, were arranged in a similar
way, so as to provide illumination for the lector at the end of Matins.
Poitiers also provides us with the valuable information that at

Matins on Good Friday, because of the cessation of bells which would
have been used to signal the start of the night office, the sacristan
stood in the doorway of the church holding a lighted candle in his hand,
which he extinguished to mark the start of Matins. 7 We do not know for
certain whether the compiler of this pontifical envisaged at which door
the sacristan would stand. It would almost certainly be the south8 or
1 Das Missale pp.100 ff. 2 Rationale VI.72 p.331.

3 OR 26.13; OR 31.13; PRG II pp.56-7 § 213.
4
6
Pontifical p.137.
Pages 26-27. --
5 Lib.Off. 4.22.1.
7 'Ilico antiphona primae nocturnae incipiatur a cantore, stante oust-

ode in ostio eclesiae et illuminatam candelam tenente, quam primam
extinguere debet.' (Page 193.)
8 The north door was opened for funerals, and to allow the departure
of exorcised devils during Baptism.
64
~est door; for the pontifical states that 'the lamps •••• bogin to be
put out on the western side of the church 1 (p.137). If the practice
observed at Poitiers corresponded to that described by AJcuin = see
below = the sacristan would have stood at the south door. Tho above=
mentioned rubric would appear to allow the possibility of the sacristan
moving in either an easterly or a northerly direction along a row of
lamps, were it not for the qualification of a subsequent rubricg
lucerna cuiusque ordinis •••• extinguatur. 1
This makes it quite clear that the three lights at the west end of the
church were put out first; but we cannot be sure of the order in which
the three lights (1,2, and 3) were extinguished (Figure 12). The evid=
ence of Alcuin, however, leaves us in no doubt about this order:
In initio primi psalmi, est custos paratus cum canna in loco

dexterae partis ecclesiae, et mox ut primam antiphonam audierit,
extinguit primam lucernam. In fine vero sequentis psalmi ex
parte sinistra, tutat aliam, in medio tertiam. Hoc ordine de
aliis prosequitur.2
Although we are not tdd where the sacristan initially took up his
position, it is safe to assume that this was near the south door.3
That the practice of extinguishing first the lamp nearest the door of
the building, also obtained in other churches is clear from the evidence
of Ordines 26, 27, 28, and ~,that the lights began to be extinguished
4
'at the entrance to the church'. Martene's statement5 that 'the monks
of Monte Cassino used to begin to extinguish the lights at the entrance
to the church' shows that the practice was also observed at that mon-
astery in the tenth century.
Figures 12-14 represent the amount of illumination in three diff-

erent churches at the conclusion of each nocturn of Matins. According
to the scheme attested by Alcuin the central light at the eastern end
of the lamps (shaded in Figure 13), would have been extinguished last.
The sacristan would then have moved to the sanctuary, causing a minimum
1 'Let a lamp from each row be extinguished.' (Ibidem p.137).

2 'At the start of the first psalm, the sacristan is ready with a reed
on the right-hand side of the church; and when he hears the first
antiphon, he extinguishes the first light. At the end of the foll-
owing psalm, a lamp is put out on the left, and then one in the
middle is extinguished. This order is repeated for the rest.•
(Lib.9! ~.Q!l. PL 101.1203C.)
3 Liturgical commentators traditionally referred to the sides of the
church, sanctuary, altar and so forth, from the point of view of a
priest facing the congregation.
4 Either the west or the south door. OR 26.13;0R 27.13;0R 28.30;PRG II
5 ~ }.12.3 p.123. 65 /p-5~213.
amount of distraction 9 in order to attend to the seven lamps. Visually
the display of thirty lamps at Boitiors (Figure 12) and the t~enty=four
of Alcuin (Figure 1~) ~ould not have been as satisfactory at the end of
the first and second nocturne as the twenty=seven lights of Ordo 32 9
assuming that the lights of that ~ were arranged in a similar way
(Figure 14). On the other hand the intervals between the nocturne
would not have been of such lengthy duration as to allow protracted
periods of asymmetrical pre~ent~tion.
Altar Altar Altar
c
0
0
0
0
0
0 c
0
0
•
0
0
()
c
0
0
0
0
0
0
0 0 0 0 0 0
0 0
0 0 0 0 ()
0 0 0
B B B 0 0 0
0 0 0
0 0 0
0 0
0 0 0 0 0
0 0 0
0 0 0 0 0 0 0 0
A A A
0 () 0 0 0 0 0 0 0
1 2 3
F~.12 Fi~.1;2 Fig.14
A section of lights extinguished after first nocturn.

B section of lights extinguished after second nocturn.
C section of lights extinguished after third nocturn.
The extinguishing last of those lamps nearest to the altar meant that
there was a sufficient light for the lectors, who would be standing in
1
the vicinity of those lamps, even at the conclusion of Matins.
Alcuin records that the seven sanctuary lamps were extinguished at

Lauds in the same order as the church lights at Matins. 2 Presumably
the outer lamp on the Gospel side of the altar was the first to be put
out, then the furthest on the Epistle side. Since it clearly states,
however, that the middle lamp was removed for reservation, it would
seem that the third light to be extinguished was another lamp on the
Gospel side, almost certainly that next to the lamp which had first
been extinguished. Interestingly, this order for extinguishing the
altar lights survived in the Roman rite. (See page 67.)
(b) After AD 1000.

(i) The hearse-lights. The only known description of the order in
1 See Section (ii) : ~extinction-points, Pp• 68 ff.

2
----
Lib.de Div.Off. PL 101.12030.
66
which the hearsa=lights ~ere extinguished relates to the Roman rite.
The first candle to be extinguished on the triangular candolabrum was
the lowest on the Gospel side~ then the lo~est on the Epistle side, and
so forth alternately in an upward direction. After the changes in 1955
the candle at the apex was not removed but continued to burn. 1
(ii) The altar lights. A distinctive feature of Tenebrae according

to the Roman rite was the extinguishing of the six altar=candles during
the Benedictus. The relationship between their uae and the extinguish=
ing of the seven sanctuary lamps at Lauds in the period before AD 1000
is not clear. It is likely that tho use of the latter fell into de-
suetude in many churches; though that they continued to feature at
Tenebrae after AD 1000 is attested at Monte Cassino in the twelfth cent~
2
ury and at Chartres in the thirteenth century.3 On the other hand in
some places the number of lights known to have been used may have in~
eluded the seven lamps; for the thirteen candles at Tongres included
4
the six altar lights which were used without distinction from the rest.
Neither the seven lamps nor the hearse-candles were used at Rome in the
twelfth century9 5 and Righetti is probably correct in concluding that
the custom at Rome of extinguishing the altar-candles during the
Benedictus is 'somewhat recent•. 6
According to the Roman rite prior to 1955 the six altar-candles

were put out during the last six verses of the Benedictus,7 beginning
8
with the outer candle on the Gospel side. Then the furthest on the
Epistle side was extinguished, and so on alternately. In practice,
however, in most places it was usual to extinguish them after every
second verse of the Song. The routine was carried out on all three days.
As a result of the reforms of 1955 this could only be performed at Lauds
of Maundy Thursday. On Good Friday there were now no candles on the
altar during Matins/Lauds, and on Holy Saturday only four. 9
1 Wuest p.259 and p.263; Fortescue and O'Connell (11th ed.) p.282.
2 Leuterman p.101. 3 Ordinary p.108.
4 Ordinary p.150. 5 ~ p.45. See also pp.44-45.
6 Manuale II p.158. 7 Caeremoniale Episcoporum II xxii 11.
8 In the Camaldolese rite the first candle to be extinguished was the
furthest on the Epistle side. (Ceremonial p.57.)
9 During the Good Friday liturgy two had accompanied the venerated
cross; and two had been brought in with the Blessed Sacrament.
67
( ;;)
~~ Th · t a1
__2 axt'~nc t'~on=po~n
(a) Poitiers and Ordo 29. Of all our sources for tho period up to
AD 1000 only Poitiere and Ordo 29 contaiu firm evidence for tho points
during Matins at which the lights were extinguished. Poitiers states :
Et singulatim a custode ecclesiae per singulao antiphonas 7 lectiones

et responsoria sod at versus trium nocturnarum, qui ante lectiones
proferuntur, cum harundine latenter quasi raptando extinguntuc. Qui
custos summopere observare debet 9 quatinua rnox ut praecentoris vox
elata in antiphona, responsorio 9 seu versu nee non et lectoris in
lectione fuerit 9 continuo aut in prima aut in secunda syllaba.2
The provision of light for the lector was important. Hence at the
beginning of the first lesson of the third nocturn, six lights were
still burning. It is important to bear in mind that these were the
functional lights of the church. In later Roman practice the gradual
loss of light affected the liturgical lights primarily, whilst the
church lights provided background illumination.
There might be some uncertainty about the precise moment when the
lamps were extinguished, particularly those put out at the antiphons,
were Poitiers the only document from this period which specified the
extinction-points. For it is not completely clear whether a light is
put out at the first antiphon before the psalm or at its repetition
after the psalm; though it must be admitted that the sense of the
passage and the absence of reference to the repetition of the antiphon
make the latter place moat unlikely. However, if we examine the
slightly earlier Ordo 29, which we believe envisaged a similar arrange-
ment of lights and contains similar ceremonial details, we see clearly
that the first lamp was extinguished before the first psalm. But not
only are we informed of the point at which the first light is exting-
uished; Ordo 29 is by far the moat important and valuable document for
our research, since we learn from it the precise moment at which all
thirteen lights were extinguished during any nocturn (8 12).
1 Throughout this section the reader is referred to Tables 4 and~

on pages 80 and 81, which summarise the evidence for the extinguish-
ing of lights in different churches at Matins and at Lauds.
2 'They are put out stealthily by the sacristan with a reed one by one
during the lessons, the antiphons, and the responsories, and also the
versicles of the three nocturne before the lessons. The sacristan
should be particularly alert for the moment the precentor's voice is
raised for the antiphon, the responsory, and the versicle, and for
when the reader begins a lesson. (The lamps are extinguished)
straigntaway at the first or second syllable.' (s 196 p.137.)
Light No. Point ~ Nocturn
1 start of tho first antiphou

2 end of the first psalm
3 start of the second antiphon
4 end of the second psalm
5 start of the third antiphon
6 end of the third psalm
7 0
while they pray'
8 after the first reading
9 start of the first responsory
10 during the second reading
11 at the second responsory
12 start of the third reading
13 start of the third responsory
The prayer being said when the seventh lamp was extinguished must be
the ~ noster. No other prayers are said during this part of a
nocturn. The ~ then states that after the thirteenth light has been
put out, a third of the lamps of the church have been extinguished. The
second nocturn follows the form of the first at the end of which, though
the 2£92 does not actually state it, two thirds of the lights have now
been extinguished. Ordo 29 then informs us somewhat to our surprise
that during the third nocturn only six lamps are extinguished; and
the extinction-points are as follows:
1 at the first antiphon

2 at the second antiphon
3 at the third antiphon
4 at the first reading
5 at the second reading
6 at the third reading
The ordo concludes that all the lights have now been put out except
the seven lamps in front of the altar.
The use of only six lights during the third nocturn is very puzzl-
ing especially as thirteen were used in each of the other two nocturne;
and, since two thirds of the church lights had been extinguished by the
end of the second nocturn, there should have been a third set of thir-
teen lights mentioned in the final nocturn, making a total of thirty-
1
nine lights in all. Instead only thirty-two seem to have been used.
1 13 + 13 + 6 at the three nocturne respectively.
69
If these figures are correct, we must conclude that to tho writer th®
concept of a third did not signify even in a mathematical sense a
strict numerical division into three equal parts~ rather it would
seem to imply any one of three divisions of a group~ none of which nee=
essarily contained two identical numbersc However~ again somewhat sur=
prisingly, we were informed in Section 11 of the ordo:
et tunc ecclesia omni lumine sit decorata 9 id est XXVIII

luminaribuso
If this figure is correct~ it follows that nine lights wero extinguish=

ed during the second nocturn; for we already know from the ~ that
thirteen and six lights were extinguished at the first and third noct=
urns respectively. Christian liturgy and ceremonial has always been
characterised by balance and symmetry, as the structure of Matins clear-
ly demonstrates. The combination of 13==9==6 seems most unlikelyo
Clearly something is wrong.
Since we know that the structure, though not the content 9 of each
nocturn was similar, let us re-examine the nocturns of Ordo 29. The
detailed analysis of the first nocturn and the clarity of the informa=
tion it contains leave us with little doubt that the information is gen-
uine and authentic. Nor is there any justification for doubting that
the structure of the second nocturn was identical with that of the first.
Therefore, in view of the statement of the ~ that the second nocturn
was like the first, it follows that thirteen lights were extinguished
during the second nocturn. Although we are told that there were only
six lamps extinguished during the third nocturn, it is difficult to be-
lieve that this information is correct and that there were not thirteen
lights as in the other two nocturne. We therefore believe that at the
start of the service thirty-nine and not twenty-eight lamps were lit,
even though to question the authenticity of the number twenty-eight on
the grounds that it is anembarrassmentand a potential stumbling block
to the argument may be viewed as an act of desperation. However,in this
instance we believe that there are very good reasons why this figure
should not be accepted as genuine. (i) The numbers specified in the
other documents viz. twe~ty-four (~), twenty-seven (Ordo 32),and
thirty (Poitiers) are all divisible by three; twenty-eight is not.
(ii) If the figure of twenty-eight is correct and six lights were ex-
tinguished during the third nocturn, we have to assume that nine lamps
were put out during the second nocturn. However, not only does the
compiler of Ordo 29 inform us that the lights in the second nocturn
70
~er@ extinguished in the same manner as those in the first nocturn;
bYt he omits any mention of the number of lamps used during the second
nocturn. He assumes that we will take it for granted that the number
of lamps extinguished in both nocturne was tho same. (iii) It is diff=
icult to believe that the compiler regarded the concept of a third so
imprecisely and in a way different to our understanding of the notion.
If thirteen lights were extinguished during the first nocturn = the
~vidence leaves us iu no doubt that this was so ~ and they comprised
one third of tho total number~ \10 can say \·lith oome confidence that a
total of thirty=nine lights t·Jere extinguished at IJJatins according to
Ordo 29; and that the figures which the ~ gives for the total viz.
twenty~eight, and for the third nocturn vize six, are both incorrect.
From the table below we can see how the arrangement for the exting=
uishing of lights during the third nocturn compares with that for the
first and second nocturne according to the information given in Ordo 29.
Nocturne 1 and 2 Nocturn 3
1 start of first antiphon 1

2
3 start of second antiphon 2
4
5 start of third antiphon 3
6
7 at the first reading 4
8
9 start of second reading 5
10
11
12 start of third reading 6
13
Lights 1,2 and 3 of Nocturn 3 correspond to Lights 1,3, and 5 of Noct-

urne 1 and 2 in that they are extinguished at the first singing of the
antiphons. Lights 4 and 5 come between the seventh and eighth, and
ninth and tenth lights respectively of Nocturne 1 and 2; and Light 6
of the third nocturn corresponds exactly with Light 12 of the first
two nocturne. The extinction-points in Nocturn 3 appear to be part of
a deliberate attempt on the part of the compiler of the ordo or a copy-
ist to harmonise them with the extinction-points in the first and second
nocturne in order to rectify a mistake which he wrongly believed to
71
existo It is our belief th&i whoro tho order for tho extinguishing of
lighto in tho third nocturfi now stands, the text originally stated that
the last nocturn was similar to the first and the second9 that a copy=
ist, knowing that it stated that thirty~nine lights were lit for Matins
(S 11), mistakenly assumed that this figure referred to the whole of the
combined office of Matins/Lauds, and thought that the seven lamps exting~
uished at Lauds \"Jere to be included in the thirty-nino; 1 and that real=
9
ising that the total number of lights extinguished at Tenobraa was forty=
six (13+13+13+7)~ deliberately reduced the number of lightD in the third
nocturn to six in order to make what he believed to be the total number
of lights for the entire service add up to thirty-nine. A possible ob=
jection to our theory has been dealt with above. For we have already
shown that the figure of twenty=eight, which this ~ records for the
number of lights lit at Matins (§ 11), is incorrect; and that thirty~
nine lamps were lit. It must be admitted that the alteration of thirty=
nine into twenty-eight cannot readily be explained. However, we may
well have here an instance of a copyist's double haplographic error in
which XXVIII has been read for XXXVIII!.
(b) The evidence of other documents. The three remaining document=

ary sources which specify the number of lights lit for the night office
during this period are the ~ Officialis of Amalarius, Ordo 32, and
PRG. We have already noted that Ordo 29 is the only document from this
period which explicitly states that the first lamp was extinguished at
the antiphon before the first' psalm. However, there is nothing in any
of the other sources to suggest that this was not the general practice.
Indeed Ordo 26, our earliest source, and Ordo 27, both from the eighth
century, state that the lights began to be extinguished ~ initio cantua
2
nocturnae, 'from the beginning of the night song', which almost certain-
ly indicates 'at the first antiphon'. The phrase is also found two cent-
uries later in ~. 3
There is no reason to believe that in the service
of Tenebrae described by Amalarius and set out in Ordo 32, practice
differed in this respect.
Having established that the antiphon before the first psalm was the
starting point for extinguishing the lights in this period, we are now
able to reconstruct with some confidence, using the three above-mention-
ed sources, the points in the service at which the rest of the lights
4
were put out. (i) According to Amalarius twenty-four lights are ex-
tinguished at the antiphons and responsories of Matins, eight during
1 The extinguishing of the seven lamps at Matins is unrecorded elsewhere.

2 § 13 in both ordines. 3 II p.56 §213. 4 Lib.Qll. 4.22.1.
72
each uoctur~. If the first light was extinguiohed at the antiphon
~hich introduced tha first psalmj it ooems reasonable to place the ex=
tinguishing of the oecoud and third lamps at the antiphons which pre=
ceded the second and third psalms respectively. The remaining five
lights clearly belong to the points in the nocturn where the five re=
sponsories are said. 1~e extinguishing of the two lamps at the respon=
sories after the third reading may seem strange9 but it ~eed not be
queotioneda For it is found in ~' at Norwich 9 and possibly at TongrGs.
Admittedly, the evidence of these later instances should not be adduced
in corroboration of the practice familiar to Amalarius; but it is im=
portant to remember that during the reading of the ninth lesson two
lights still burning, as opposed to one, would provide additional
illumination for the reader at the end of Matins.
(ii) According to ~ the twenty-four lights should be extinguished

'after each reading and responsory', and begin to be put out ·~initio
1
cantua nocturnae•. Taken literally, the instructions for extinguishing
the lamps present us with a very unusual scheme; for one light is put
out at the very start of each nocturn, and seven in the latter part of
the nocturn after the psalms have been sung. fSQ would appear to be
alone amongst all the documents in not featuring the loss of a light at
the antiphons (either before or after) of all the psalms; and it is
strange that a lamp should be put out before the first psalm but not be-
fore the second or third. It is true that Ordo 29 attests the exting-
uishing of seven lights at the readings and responsories in the latter
part of the nocturn; but this was done of necessity since thirteen
lamps were required to be extinguished during each nocturn, and only
2
sixteen points were available for the accommodation of each light.
Moreover, since ~ also states that the lights should be put out grad-
ually3 - a statement at variance with the literal interpreation of the
rubrics, we suspect either that a phrase relating to the antiphons has
been accidentally omitted from the text, or that the compiler of the
pontifical assumed that the reader would understand and take for grant=
ed a phrase such as•at each antiphon~ Nevertheless the possibility
must remain that the prima facie interpretation of the text of ~ is
correct; and in Table4. we have given a scheme for the loss of light
according to the literal evidence of the pontifical, as well as an
arrangement suggested by our interpretation of the text in conjunction
1 II p.56 § 213.
2 The psalms alone of the nineteen constituent elements of Matins
were never used as extinction-points. See Tables 4 and~
3 Lumen •••• incipiat paulatim tutari (ibidem).
73
with other documentary ovidence from this period.
(iii) Ordo 32 gives no indication of the extinction=points at

which nine lamps were put out during each nocturn. It io tempting to
reconstruct the scheme based on the detailed description in Ordo 29,
which we looked at above, especially as both ordines are of Gallican
provenance and are more or less contemporaneous. However, the details
concerning the Tciduum in Or-do 32 gGnerally ~how little correspondence
with those in Ordo 29. Evon Qt Tenebrae, where the 5tructure of the
office and the ceremonial were similar 9 affinity in respect of lights
should not be taken for granted. On the other hand the similarities
between Ordo 32 and ~ show that the latter document, compiled some
fifty or so years later, almost certainly derives from the same liturg=
ical milieu. The suggested scheme for the loss of light, therefore, is
based on the information contained in Ordo 29 and in ~·
As in Poitiers it is not immediately clear from Alcuin at which of

1
the two antiphons each light is extinguished. In view of the depend=
ence of this document on ~' we have suggested a scheme similar to
the latter.
Two of our n1ne2 principal sources for Tenebrae in the period be-
fore AD 1000 do not refer to the points at which the seven sanctuary
lamps were extinguished during Lauds. 3 They are Amalarius and Ordo 32.3
Ordo 31 and PRG mention only the first two lamps and the last. Howeve~
Ordo 29 and Pbitiers concur that six were extinguished at the antiphons
preceding and following the first three psalms of Lauds, and that the
seventh was put out at the Traditor, the antiphon before the Benedictus.
The extinguishing of the sixth lamp at the repeated antiphon of Psalm 66,
thus reducing illumination in the church for the chanting of the Canticle
and the Laudate Dominum to one light, would have made more urgent the
plea of the~ misereatur for God to'show the light of his countenance';
and in a way anticipated the total loss of light before the Benedictus.
It is almost certain that this arrangement regarding the extinguish-

ing of the seven lamps at Lauds also obtained in Alcuin in spite of the
1 That is, at the first chanting of the sentence or at its .repetitiono

2 OR 26; OR 28; OR 29; OR 31; OR 32; Amalarius; Poitiers; PRG; and
Alcuin.
3 Although Amalarius does not refer to Lauds, there is no reason to
believe that the lamps were not extinguished during the course of
the service that was familiar to him.
74
interest that the document shows in the ©rder in which the lamps wero
0xtinguiahed rather than in the oxtinction~points. Alcuin's reference
to the use of the last of the seven lamps on Naundy Thursday is not en~
tirely clear. For at the conclusion of Lauds on that day we are inform=
ed that the central lamp is removed and reserved for the lighting of
the Easter candle on Holy Saturday; yet later that same day we learn
that the sacristans prepare the seven lamps in front of the altar in
1
readiness for the evening liturgy. The difficulty is by no means in=
superable. Either the last lamp was returned extinguished to its cust=
omary place, once its flame had been used to kindle another lamp in the
place of reservation; or that lamp remained in the same place whither
it had been taken, and an eighth or spare sanctuary lamp was subsequent=
ly placed with the other six.
(c) Extinction-points after AD 1000. In the period after AD 1000

the fact that the seven sanctuary lamps are rarely known to have been
extinguished at Lauds strongly suggests that their use at this office
had generally fallen into desuetude. Their demise at Tenebrae seems to
have coincided with the emergence of the fifteen-candle hearse, and to
have been a consequence of the innovatory practice of extinguishing all
the lights, formerly lit for Matins, during Matins and Lauds. The spread
of the loss of the light from the same set of candles thus imposed a
uniformity on the whole service in respect of the use of illumination.
This is borne out at Tongres where the candles which formed two differ-
ent sets of lights were regarded without discrimination for the pur-
pose of Tenebrae. 2
A distinctive characteristic of Tenebrae before AD 1000 was the

practice of extinguishing a light at the antiphon preceding the psalm.
In the later period, however, it became customary to extinguish each
light at the repetition of the antiphon. The adoption of the latter
practice, which is probably linked with the introduction of the hearse
at Tenebrae, is first attested at York in the eleventh century,3 and
4
became a feature of the Roman rite. The extinguishing of a candle at
the antiphon preceding the psalm at Cluny and Farfa, where a hearse of
1 Lib.de ~.Off. 16 : PL 101.12030.

2 Ordinary p.150. See also p.44 above.
3 Breviary p.376.
4 Caeremoniale Episcoporum II.22 p.26 It is not known when this
change took place. In the Dominican rite, which was modelled on
the Roman, the candles were extinguished before the psalms i.D.
actual .,pnetiee.,: e'¥:9a.- ·though the Office of !!211 Week prescribed
'at the close of each psalm' (p.82).
75
fifteen lights was used 9 represents a transitional stage between the
practices of both periods. Tables!:. and .2 show that a number of dift'=
erent schemes for extinguishing the candles at Tenebrae are known to
have existed.
1. Fifteen candles extinguished before the psalms.
At Cluny and presumably at most of her 314 dependent houses the

fifteen candles were put out at tho antiphons before the psalms of
Matins and Lauds$ Further possible support is added to our theory that
at Farfa two hearses stood in front of the altar 1 since at this monast=
ery also the candles were extinguished before the psalms. For if the
candelabrum on the Gospel side of the altar was used to keep a tally of
the fifteen gradual psalms, the extinguishing of the first light on the
other candelabrum may have acted as the signal for the start of Matins
and for the singing of the antiphon before the first psalm. With the
first candle extinguished before the first psalm, the pattern was set;
and the remaining lights would also be put out before the psalms. At
St Mary's, York the candles may have been extinguished before the
psalms, but the phrase from its ordinal's rubric 'ad inceptionem ~=
cuiusque antiphone' is ambiguous and could equally refer to the anti-

phon or to its repetition. 2
2. Fifteen candles extinguished after the psalms.
It is not clear why the practice arose of extinguishing the lights

at the antiphons following the psalms. It might be thought that the
uttering of only the opening words of the antiphon which preceded the
psalm was too short a period of time to allow each candle to be exting-
uished with decorum; and that, in consequence of this, the candles
came to be extinguished after the psalm, as occurred in the Dominican
rite. However,in the Roman rite, in which the candles were put out at
the conclusion of a psalm, the antiphons were doubled,3 since each of
the three days of the Triduum was a Double Feast. Moreover, as we have
observed above, the Dominicans in practice extinguished the candles at
the abbreviated antiphon before the psalm. The evidence of local rites
4 5
such as Besan'ion, Coutances, and Verdun
6 would suggest that by the
eighteenth century Roman practice had become the norm.
1 See above, pp.52-56. 2 HBS 75 p.271.

3 i.e. said (or sung) in full before and after the psalm.
4 1707 Ceremonial p.235. 5 1825 Ceremonial p.311.
6 1832 Ceremonial p.294.
76
3· Twonty=four lights extinguished at the antiphons and responsories.
The use of twenty=four lights at Tenebrae was not only determined
by the symbolism readily suggested by that numberi 1 but also by the
very structure of the combined office of Matins and Lauds. For the
three psalms and the three lessons of each nocturn of Matins, and the
four psalms, OT canticle, and Benedictus of Lauds provided twenty=four
convenient points at which to extinguish the candles; and though vari=
ations are found at Norwich and Trier, these points are specified by
the majority of service=books and liturgical commentators.
From a number of our sources, however, it is not clear whether the

candles were extinguished at the antiphon preceding or following each
2 4
psalm. The Ordinals of Barking, Exeter, 3 and of the Gilbertines re=
fer only to the antiphons and responsories, as does Lanfranc;5 and
John of Avranches mentions only that the lights were put out 'secundum
psalmos ac lectiones•.
6
However1 Rupert of Deutz 7 and Honorius of Autun
8
both state that the lights were extinguished after the psalms; and this
is borne out by evidence from York. 9 In instances where the antiphon
is not specified, it would seem reasonable to assume that the light was
extinguished at its repetition. This not only accords well with the
firm evidence for the practice; it balances the loss of light ~
the reading of a lesson.
It is necessary to record a few observations about several churches:

10
Salisbury. The rubrics of the various Sarum breviariesstate that a
11
candle should be extinguished 'at the beginning of each antiphon•. It
is almost certain that the repetition of the antiphon is here intended.
Immediately following Psalm 148 is the rubric : 'While this psalm is
12
being sung, let a light be hidden where it cannot be seen.• This re-
lates to the twenty-third candle which was reproduced at the end of the
service. The twenty-fourth and last candle continued to burn, and was
extinguished at the conclusion of the Benedictua.
Nidaros. The Ordo of Nidaros prescribes twenty-four candles and informs
1 See above p.35. 2 HBS 65 p.91.

3 HBS 37 p.132. 4 HBS 59 p.30.
5 Decreta, PL 137.458A. 6 ~-~ Off.Eccl., PL 14?.480.
7 ~ ~.Off., PL 170.1480. 8 ~ Animae, PL 172.665D.
9 Breviary I P•376 10 Procter & Wordsworth col.dcclxxiii.
11 Ad inceptionem uniuscuiusgue Antiphonae.
12 ~ iste Psalmus canitur, ~ ~ videri nequeat abscondatur.
(Ibidem col.dcclxxxiiJ
77
us that as the service is about to start the sacristan is ready on thG
right=hand side of tho churcho It continues
At the first antiphon he puts out the first candle; then (the
others) !!, ~ ~ antiphons g.r at 'W ~ gi :Y!! psalm itself.
At Lauds they are extinguished in a similar way, one on one side,
one on the other, until the start of the Benedictus.1
A superficial reading of the above rubric leaves us with the impression

that a choice exists of extinguishing the candles before or after each
psalm; and that the first antiphon and the other antiphons mentioned
are those which preceded the psalms, in view of the alternative point
seemingly offered : at the end of the psalm. A more careful reading of
the rubric, however, reveals that another interpretation exists regard-
ing the choice of points. For if, as the present writer believes, the
antiphons mentioned in the rubric are those which follow the psalms, the
choice lies in extinguishing the candle either at the conclusion of the
psalm itself or during the antiphon which follows the psalm. Either
point accords with the above=mentioned testimoDY of Rupert of Deutz
and Honorius of Autun.
Norwich. From the Benedictine Customary of Norwich2 we learn that six

candles were extinguished during each nocturn in the following way
three at the antiphons (almost certainly after the psalms)

one at the versicle
one at the third responsory
one at the repetition of the third responsory
At the start of Lauds, of the seven candles that were still alight
five were put out at the antiphons of each psalm

one was put out at the Deus misereatur (Psalm 66)
one was put out at the Laudate Dominum (Psalm 150)
On the first reading it might appear that the antiphons at which the
candles were extinguished were those sung after the five psalms of
Lauds (including the OT canticle). However the extinguishing of the
sixth candle 'at the Deus misereatur' rules out this interpretation,
and indicates that the first five lights were put out before the start
of Psalm 62. In other words one candle was extinguished at each
1 ~ p.223. The rubrics have much in common with the directions

for conducting Tenebrae as prescribed by Alcuin. ~ adds the
words italicised by the present writer.
2 HBS 82 p.79.
78
antiphon and at its repetition as far ~s Psalm 62. Since the exting-
uishing of the sixth candle would not take place during the singing of
Psalm 66, it must be placed at the antiphon following that psalm. The
arrangement so far corresponds exactly with that found in Poitiers.
The information that 0 one is put out at Psalm 150' (i.e. the seventh
and last candle is put out at the repetition of the antiphon) is follow=
ed by a further statement that 9 when tho sacristan has taken a light in
a lantern outside the choir, 0
there will now be no light in the choir 0 •
The description then ceases to be lucid when we are subsequently inform=
1
ed that the last of the twenty-five candles is put out at the Traditor 9
the antiphon of the Benedictus; for we were previously told that the
last light was extinguished at Psalm 150. Since the last candle could
only be extinguished at one point, perhaps the moat satisfactory explan-
ation is to assume that this seventh light at Lauds was extinguished at
2
the very end of the repeated antiphon of Psalm 150, and that a later
hand added the statement that it was put out at the Traditor in view of
the fact that the Traditor immediately followed the quenching of the
flame. Alternatively the mention of the two points may be viewed as
evidence of a choice, even though such a choice is not specified.
~· The imprecise rubrics of the cathedral's thirteenth-century

ordinal3 allows two possible interpretations. Of the twenty-four cand-
les to be extinguished, the first is put out at the first antiphon, the
second after the first psalm, and the rest after the subsequent psalms
of Matins and Lauds. All are extinguished by the Traditor preceding the
Benedictus. These directions, however, taken literally account for
only sixteen candles. One solution is to assume that two lights were
4 -
extinguished after each psalm of Lauds; but this would still leave
two candles to be accounted for. Alternatively we must assume that a
light was also extinguished at each of the antiphons which preceded
the psalms, as we have suggested in Tables 4 and ..2.
Tongres. The scheme at Tongres was unusual in two respects. The
lights were extinguished two at a time, and apart from two at the end
of Matins they were extinguished during the course of Lauds. (Table$.)
1 The Traditor (or Traditor autem) was the antiphon of the Benedictus
on Maundy Thursday only. However,aa the structure of Lauds was the
same for each day, the name Traditor has been used to indicate this
antiphon on the other two days of the Triduum
2 Table ..5: Points 16 and 17. 3 13th century Ordinal p.486.
4 See Tongres below.
79
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15+ 16 17 18 19 20 21 22 23 24
0 0 A 0 0 0 p p A c F G T N s N T R B A A s c v
(For references1
r r m r r r 0 ~ I l a e r i a 0 0 0 e :n u e 0 G
see page 84)

d d a d d d i g .£ u r m i d l r n m 6 $ X v u E'
0 0 l 0 0 0 ! u n f b e a i w g e a e e i t o1
a i i y a l r r s i r n .lr' r l a 1l
26 28 r 29 31 32 e n e l
0 0
s
b c
s
r iS r El !:ll
i £ u u h 0 e e c
u .§. X r 11 e
s y s
No. of lights
per nocturn ? ? 8 13 ? 9 10 8 8 3 3 8 6 6 6 6 (2)* 3 3 ,; 3 3 3 3
1. Antiphon X X X X (x) X X X X X
2. Psalm (i}
3· Antiphon X X X X X X X X X X X
00 4. Antiphon X X X X (x) X X X X X
0
5· Psalm (ii)
6. Antiphon X X X X X X X X X X X
7• Antiphon X X X X (x) X X X X X
8. Psalm (iii) '
9· Antiphon X X X X X X X X X X X
10. Versicle X X X
11 • Responsory X X X
12. Pater noster X
13. Reading 1 X X 2£ X :lt X
14. Responsory X X X X X X X X
15. Reading 2 X X ~ X :It X
16. Responsory X X X X X X X X
17. Reading 3 X X ~ X
a
:K X
18. Responsory X X X X X X X X X X
19. Responsory X X X (x) X XX
+ York is identical. • 3rd nocturn only __

• ___ a Choice with Point 19. ~ Disputed.
Table 4. Extinction-points at each nocturn of Matins.

1 2 3 4 5 6 7 8 9 10 11 12 13 14+ 15 16 17 18 19 20 21 22 23 24
0 0 0 0 0 p p A c F c T N s N T R B A A A s c v
r r r r r 0 R 1 1 a a r i a 0 0 0 e II lil u e 0 e
(For references, d d d d d i G c u r r i d 1 r n m 8 g g X v u r
see page 84) 0 0 0 0 0 t u n f m e a i w g e a El e e i t Ol
i i y a e r r s i r n r r X' 1 & u.
26 28 ~8 29 31 1 b c e c r 1
a
-
e
r
n
- i
0
s u h s
•
0
6 8
e e
Ill
c
R'A
s t r n e
e y_ 8
No. of lights 6 6 6 6 11 6 6 13 4 6 4 6 6
at Lauds 7 7 7 7 7 7 7 ? 6 6 5
1. Antiphon 1 1 1 1 1 1 1 1 1 ( 1) ?1 1 1
2. 1st Psalm
3. Antiphon 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 ?1 1 1
00
~
4. Antiphon 1 1 1 1 (1) 1 ( 1) 1 1 ( 1) 1
5· 2nd Psalm
6. Antiphon 1 1 1 ( 1) 1 (1) 1 1 1 1 1 1 2 1 1 1 ?1 1 1
7. Antiphon 1 1 1 1 ( 1) 1 ( 1) 1 1 ( 1) 1
8. Psalm 62 ?1
9. Psalm 66
10. Antiphon 1 1 1 ( 1) 1 (1) 1 1 1 1 1 1 2 1 1 1 ?1 1 1
11. Antiphon 1 1 ( 1)
12. OT Canticle
13. Antiphon 1 1 1 1 2 1 1 1 1 1
14. Antiphon 1 1 ( 1)
15. Laudate 1 1
16. Antiphon 1 1 1 2 1 1 1 1 1
17. Traditor 1 1 1 1 1 1 1 1 1 1 1
18. Benedictus X X X X X X X X X X X X X X X 12 4
19. Traditor 1 1 1 1 1 1 1 1
+ York is identical. X = B§nedictue sung in darkness.
Table 5. Extinction-points at Lauds.

The latter feature was reminiscent of Carmelite practice. Moreover,the
thirteen candles comprised the six altar lights and the sovon atop a
1
spear.
Seville. At Seville, where a thirteen=light hearse was used, during

Lauds the candles were extinguished only after the first three psalms.
This matched the scheme for extinguishing the candles in each of the
thret?~ nocturna of Matins.
It is clear from the earlier evidence that the person responsible

2
for extinguishing the lights was the sacristan. In more recent times
the duty fell to a server or the master of ceremonies.3 At St Mary's,
York a server was concealed for this purpose behind the veil or curtain
4
which was suspended from the base of the hearse.
In addition to a snuffer, three different implements are known to

have been used for extinguishing the lights. (i) According to Poitiers,
~' and Alcuin5 the sacristan employed a reed. How it was used is clear
from the first of these documents which states that a lamp from each
6
row 'should be blown out with one puff of breath, if possible'. The
exhortation to emit only one blast of air from the reed was no doubt
intended to avoid embarassment. The use of a reed would suggest that
the lamps were positioned above the height of a man. (ii) Writing
about the churches of Italy John England records that in some places a
moist sponge was used to extinguish the candles. 7 (iii) Possibly of
greater antiquity than the preceding extinguisher was the use of a wax
hand, known as a 'Judas hand'. It is first attested by Beleth, who
comments that it may represent the hand described in Daniel 5:5; but
he does not elaborate.
8 Presumably the Judas hand at Tenebrae is assoc-
iated with the loss of light in the same way that the hand that wrote
upon the wall heralded the loss of Belshazzar's life. However.Beleth
admits that the wax hand is more likely to recall Jesus' prediction
1 See above, p.44. The spear for bearing the new fire in the Gilbert-
ine rite had five candles. (HBS 59 p.39.)
2 Poitiers p.137; OR 29.12; Alcuin (PL 101.1203B); ~ II p.57 I 214;
and York Breviary I p.376.
3 Fortescue & 0'Connell(4 ed.) p.3Q4. 4 HBS 75 p.271.
5 Page 137; II p.57 8 214; P.L 101.1203B, respectively. A reed is also
mentioned in OR 28.7 (1.33) and .30.
6 Uno, si poteet !1.!!:!• extinguatur !lm·
7 Ceremonies 2! Holy ~ p.50. 8 Rationale, PL 202.106B.
82
1
in Matthew 26:23, an opinion ahared by Durandus who adds that the
hand is made of wax bocauso it 1 bends towards ovil 0 • 2
(iii) The Candles
The practice of using yellow or unbleached candles for both the

hearse-lights and those on tho altar is probably of a venerable anti~
quity. The use of sombre-coloured wax aignifioa tho f~oreal aBpect of

Tanobruo. 3 In many places tho custom arose of placing a white candle
at the apex of the hearse
4 to indicate that that light symbolised
Christ. 5 Thurston would appear to regard this practice as the norm.
6
Howeve~Fortescue and O'Connell stated, 'There is no authority for using
a white candle in the centre'; 7 and in the Sistine Chapel all fifteen
8
candles were yellow.
It was recornmended9 that each candle should consist of one pound of

10
wax, a weight attested at Lincoln in the late Middle Ages. At Canter-
11
bur.y each of the twenty~four candles weighed three quarters of a pound.
(iv) The Light before ~ Blessed Sacrament
12
Although an altar-lamp is attested in Bergamo Cathedral as early
as AD 922, and although the presence of a light before the Blessed
Sacrament was common in parish churches by the thirteenth century, 13
there is little reference to this perpetual lamp at Tenebrae before
1600, largely, one suspects, because it remained unlit on Good Friday
and Holy Saturday with the removal of the Sanctissimum on Maundy Thurs-
day to a place of reservation. As regards most churches it must remain
a matter of speculation whether or not the sanctuary-lamps were exting-
uished at any point during Tenebrae of Maundy Thursday. However, it is
11
recorded at Canterbury in the thirteenth century that the lamp (bacinus)
before the high altar and those lights (bacini) which honoured the bodies
of the saints should be extinguished at Lauds during the singing of the
last psalm {Laudete).
1 'He who has dipped his hand into the dish with-me -will betray me.'
2 'Ad malum flexibilis' : .Rationale VI.72 p.331.
3 Candles of unbleached wax are recommended for funerals.
4 England p.5Q. If the highest candle was identified with Judas,
presumably it remained unbleached.
5 Heuser p.229. 6 Lent ~ Holy ~ pp.243ff.
7 4th edition p.303. 8 England p.50.
9 Caeremoniale Episcoporum II.22 p.264; Le Vavasseur p.363.
10 Bradshaw & Wordsworth II p.303. 11 HBS 23 p.380.
12 Ughelli IV pp.616-21. 13 King, Eucharistic Reservation p.129.
According to the Caeremoniale Episcoporum the sanctuary=light was
not to be extinguished at Tonebrae. 1 This direction is also enjoined
by La Vavasseur~ 2 and Forteocuo and 0'Connell.3 The Camnldoleso Cere-
monial states that the Sanctissimum should be transferred to another
tabernacle whose light was, presumably, unobtrusive (p.56). However,
Grancolas mentions that the last hearse~candle was hidden rather than
extinguished, and that the sanctuary=light was kindled from it as soon
4
as possible (~ ~).
There is no need to see at Tanebrae of Maundy Thursday a paradox in

the concealing or extinguishing of the last candle, which represented
Christ, whilst the sanctuary-lamp continued to burn in the darkness
which commemorated His passion and death. The celebration of Tenebrae
on l4aundy Thursday was an anticipation or foreshadowing of the events
of the following day, so that the use of light during that office was
in a sense detached from and did not relate to the events of Maundy
Thursday. Seen from a different perspective, on the other hand, the
removal of the last candle symbolised the presence, and specifically
the death, of Christ within an historical context; and the continuing
flame of the sanctuary-lamp represented his universality. At a higher
theological level the two lights visually portrayed within the liturgy
respectively His human and His divine natures.
1 I.xii.17 (typ.ed. 1886). 2 Cer~monial (1859) p.365.

3 4th edition p.3Q4. 4 Commentarius p.296.
Table4 references : (1) OR 26.1}; (2) OR 28.}0,.49; (}) ~.~.4.22.1;

(4) OR 29.12; (5) OR }1.13; (6) OR 32.5; (7) Page 137; (8) II p.56 8 213;
(9) PL 101.1203C; (10) Albers II p.16; (11) Albers I p.46; (12) Albers
II p.91; (1}) Ordinary p.486; (14) ~p.223; (15) Breviary col.dcclxxiii.
(York:Breviary I p.376); (16) HBS 82 pp.79-80; (17) Ordinary pp.150-1;
(18) Breviary I p.445; (19) Ceremonial p.253; (20) 1731 Ceremonial p.224;
(21) Breviary p.243; (22) Doblado p.284; (23) Ceremonial p.311;
(24) Ceremonial p.294.
Table 5references: (1) OR 26.13; (2) OR 28.7 (1.33 MS T); (3) OR 28.30;
(4) OR 29.12; (5) OR }1.13; (6) Page 137; (7) II p.57 i 214;
(8) PL 101.1203C; (9) Albers II p.16; (10) Albers I p.47; (11) Ordinary
p.162; (12) Ordinary p.486; (13) ~p.223; (14) Breviary col.dcclxxxii.
(York:Breviary I pp.376 and 382); (15) HBS 82 pp.79-80; (16) Ordinary
pp.150-1; (17) Breviary I ~.445; ~.Episc.II.xxii p.265;
(18) Ceremonial p.253; (19) 1734 Ceremonial p.260; (20) 1731 Ceremonial
pp.224-26; (21) Breviary p.243; (22) Doblado p.284; (23) Ceremonial
p.311; (24) Ceremonial p.294.
84
Chapter Eight
THE CONCLUSION OF TENEBRAE
(i) ~ Tradition £! ~ Benedictus without light
The practice of singing the Benedictus in darkness is well~attested

in the period before AD 1000. According to moat of the documents
1
the seventh sanctuary lamp was extinguished at the Traditor before the
Benedictua. 2 However, according to ~3 the choice was allowed of
either extinguishing the light at this point, or of withdrawing and
reserving it for use on Holy Saturday, a practice we have already en=
4
countered in Alcuin. The custom of singing the Benedictus without
light was perpetuated in both the cathedral and monastic traditions.
At Exeter? it was hidden during the chanting of the last psalm and ex-
tinguished, as at Cluny 6and presumably at most Cluniac houses, at its
repeated antiphon. At Worcester? it was extinguished, but at Rouen8
and Salisbury 9only hidden, at Psalm 148. According to the thirteenth-
century Franciscan ordo 10 the penultimate candle was hidden either at
11
the last psalm or its antiphon, and the last candle was extinguished
at the Traditor. The Regularis Concordia merely states 'at the ~~
12
dictus1; whilst at Nidaros and at Trier it is recorded that the
Benedictus was sung in the dark. 13 Elsewhere the final candle was put
14
out at the first rendering of the Traditor.
It is not difficult to see why the practice had developed of sing-

ing the Benedictus in darkness. The H1JDn of Thanksgiving related prim-
arily to the events that occurred prior to the birth of Jesus. In the
same way that the ministry of John the Baptist was preparatory to the
1 OR 26.13; OR 28.30; OR 29.12; OR 31.13; OR 32.5; Poitiers p.13?.

2 OR 30B.28, which attests Tenebrae only on Good Friday, offers the
choice of extinguishing either at the last psalm or at the Benedictus.
3 II p.57 8214 : tutatur media lampada, vel subtrahitur et servatur.
4 See above, p.75.
5 BBS 3? p.132. 6 Albers II p.16. ? Antiphonary p.62.
8 1480 Breviary np. 9 Breviary col. dcclxxxii.
10 HBS 85 P• ?6 and Van Dijk II p.84.
11 The Carmelites extinguished it at this point. (Ordinary p.163.)
12 PL 13?.490A.
13 -Breviary fol.lviii and Ordinary p.486, respectively.
14 At Farfa (PL 15Q.119?D); GembloUx, Fruttuaria, Chester (Albers: II
p.91; IV p.49; IV p.245 respectively); StMary's, York (HBS ?5 p.2?3);
St Paul's, Rome (Mart$ne,~ 4.22.8 p.124); Norwich (see pp.?S-?9).
Canterbury (HBS 23 p.380); among the Gilbertines (.HBS 59 p.31).
It is mentioned by John of Avranches (PL 14?.48C).
85
advent of the Light of the Worldg ao tho three days of the Triduum
recalled the temporary absence of that Light. As early ao the ninth
century Amalariue saw in the extinguishing of candles a commemoration
of Jesus' resting in the Tomb :
Quod lumen ecclesiae extinguitur in his noctibus, videtur nobis

aptari ipsi soli iustitiae 1 qui exstinctus est et sepultus tribus
diebus et tribus noctibus.l
The reference in the final veroo of tho Song to 'those that sit in
darkness and in the shadow of death' was therefore particularly appro=
priate to those who were chanting these verses from the Gospel in
total darkness.
(ii) The Tradition of ~ Benedictus ~ light
1. A single candle. In view of the above-quoted reference from the

Benedictus to the gift of light to those who exist in darkness, it is
perhaps not surprising that the practice arose of allowing the final
lamp or candle to remain lit until the conclusion of the Benedictus so
as to portray that Light emphatically in a symbolic way. The practice
is first attested at York in the eleventh century, 2 and was subsequent-
ly incorporated into the Roman rite.3 The practice is also found at
Seville, and in a number of French dioceses, such as Besan~on, 5
4
Coutances, 6 Paris, 7 and Verdun, 8 and in the Premonstratensian9 and

10
Camaldolese rites.
2. Altar lights. The extinguishing of the six altar-candles during the

Benedictus in the Roman rite has been discussed above on page 67. The
11
six candles were also put out in like manner at Paris, Besan~on,5
and Seville.
4 Somewhat surprisingly two manuals from Angers, separated
by only a three-year interval, record different schemes for extinguish-
ing the hearse-candles, not the altar lights, during the Benedictua.
12
According to the earlier one nine of the candles were extinguished
1 'The extinguishing of the church lights during these nights seems to

me to relate to the very Sun of Justice whose light was extinguish-
ed and buried for three days.' (~.2!£· 4.22.)
2 Breviary Vol.1 p.382. It is assigned to this century by J.M.Neale
(Christian Remembrancer XX (Oct.1850) p.285) cited in Breviary,
3 Caeremoniale Episcoporum II.22 p.266. /ibidem.
4 Doblado p.284. 5 1682 Ceremonial p.265.
6 1825 Ceremonial p.318. 7 1?78 Breviary, Pars ~· p.289.
8 1832 Ceremonial p.294. 9 1930 Breviary p:402.
10 1634 Ceremonial pp.56-58. 11 1662 Ceremonial p.34o.
12 1731 Ceremonial p.223.
86
during Matins, the remaining lights during the last four verses of the
1
Benedictus. The other manual, however, prescribes that one candle
should be extinguished at each of the twelve verses of the Benedictus,
the last one being removed at the repeated Traditor. (See also Table 5 .)
In the late fifteenth century the six lights above the chancel in the
Sistine Chapel were also extinguished simultaneously with the six
altar=candles. 2
(iii) The ~ Candle
Writing in the later part of the thirteenth century Durandus

states 3 that the last candle at Tenebrae was extinguished at the ~
dictus; but gives no indication of whether this was done at the anti-
4
phon before or after the Gospel. Since he makes no reference to the
Benedictus being sung in darkness, it is possible that his statement
was deliberately imprecise so as to allow a choice of points .for exting-
uishing the final candle. It is perhaps surprising that he makes no
overt reference to the practice, which had existed since the eighth
century5 and which was widespread in his day, of removing rather than
extinguishing the last candle, and subsequently restoring it before the
end of the service. 6
Regardless of whether the last candle had been removed before or

after the Benedictus, the service always ended in darkness. The gen-
eral practice then appears to have been to reproduce a light at the
conclusion of the office, originally to provide illumination for those
leaving a church in complete darkness. Thus, in those places where the
last candle had actually been extinguished, another light was produced. 7
8
At Gembloux the last candle was relit. In the Dominican rite the sacr-
istan's lamp, hidden during the fifth psalm, was reproduced.9 Where the
last candle had only been hidden, it was customary to bring it forth at
10 /
the end of the service. At St Agnan's, Orleans, where Lauds even in
the eighteenth century appears to have been held in darkness, several
1 1734 Diurnale p.260. 2 Dykmans II p.366.

3 Batiopele VI.72 p.322.
4 The Benedictus was formerly referred to as the Evangelium.
5 Ordo 30B.28.
6 The possibility should be considered that Durandus, who wrote
'extinguitur• of the last candle, was using the verb in the sense
of 'remove from view' in addition to 'extinguish'.
7 St Mary's, York (HBS 75 p.273); Haymo's ~ Breviarii (Van Dijk II
p.84); Worcester (Antiphonary p.63); York (Breviary I p.382).
8 Albers II p.91. 9 B~kfriars correspondence.
10 Rouen Breviary np; Premonstratensian Breviary p.4o2; Beleth PL 202.
/106D.
87
candles were relit at the very and of the offico. 1
According to the Roman rite official practice until 1955 was to re=
move the last candle at the repetition of the Traditor, to place it
momentarily on the altar, and then to hide it under the altar at the
Epistle corner; thence to bring it forth once more at the very end of
the service before extinguishing it after the departure of the congreg=
2
ation. An alternative (unofficial) practice waa to loav~ thG candle
burning until tho conclusion of thG service 9 and then to hide it. 3
After the reforms of 1955 this alternative practice was officially
4
adopted, except that the candle was not hidden but simply extinguished
when the church was empty.
The custom of bringing back the last candle arose for practical
reasons; and had its origin in the days when Tenebrae ended before
dawn in those churches where the Benedictus was sung by the light of
the candle which was subsequently hidden but not extinguished. Before
a symbolic interpretation became attached to this candle, its function5
was similar to the sacristan's lamp before the start of the service :
to provide illumination for the faithful as they left the church. Evi-
6
dence for this lamp is to be found in a number of places. At Norwich,
8
Farfa, 7 and St Paul's, Rome a candle was lit in a lantern before the
last light had been extinguished, and at Norwich it was removed from
the choir. At Worcester9 it had alrea~ been lit outside the church be-
fore the start of the service. After a signal from the under-sacristan
it was brought into the building to guide the monks back to their dorm-
itory. At St Mary's, York, where the last candle was put out at the
Benedictus, the sacristan, on hearing the knocks made by the abbot at
the end of the service, produced a candle lit from the copiosum ~
which he had kept hidden away in a lantern (sconsa). This was then
10
placed on the chancel step. In the Cistercian rite a candle was lit
in a lantern (sconsa) before the last light was extinguished, and
11
brought into the choir when the abbot began the~ noster.
1 De Moleon p.206.
2 Roman Breviary p.445; Fortescue and O'Connell, 4th edition, p.}05.
3 Thurston, Lent ~ ~ ~ p.243.
4 Fortescue and O'Connell, 11th ed., p.282; Wuest p.263.
5 Crossley p.83. 6 HBS 82 p.79.
7 Albers I p.46. 8 Martene, DAER 4.22.8 p.124
9 Antiphonary p.62. 10 HBS 75 p.273.
11 Nomasticon Cisterciense p.99.
88
( iv) The Last Candle : Symbolism ~ ~
In the same way that the medieval mind attached a symbolic inter=
pretation to other liturgical features of Tenobrae~ the last candle
especially came to be regarded with great importance, and to be under=
stood in a number of different ways. This was to a large extent due to
the darkneso which was to follow the extinction of ita flame. Accord=
ing to Hugh of St Victor the last caudle stood for Christ who was the
last prophet to be killed. However 9 sinco ho waa regarded as a prophet
while he was proclaiming the Gospel during his ministry, the candle was
1
to be of the same size as the rest. Hugh obviously understood this
light to represent only the human nature of Jesus. The same rationale
probably lies behind the stipulation in the Gilbertine Ordinal that the
2
candle at the top of the hearse should be no larger than the rest. On
the other hand the central candle of the twenty-five at York Minster
appears to have been larger than the rest, 3 aqopinion shared by Dom
4
AndreMocquereau. The identification of Christ with the last candle
is also found in the Ordinal 2J g Mary's, !2!!: 1 where Christ is com-
pared to the candela preeminens on the hearse. 5 Durandus, aware of the
two traditions regarding the size of the last candle, commented that
'according to some it is larger than the others because Christ was
greater than men. According to others it is the same size - Christ was
,6
one of the prophets.
John Beleth also compares the concealing of the last candle to the
physical death of Christ. Its reappearance at the conclusion of the
office for the kindling of the church lights anticipates the Resurrect-
ion.7 Pope Benedict XIV also likens the concealing of the last candle
behind the altar to the burial of Jesus. For him the fact that the
candle remains lit whilst hidden signifies the activity of the Lord in
the Underworld; and the restoration of the candle to its original place
8
on the hearse symbolises the coming Resurrection. At Angers the sign-
ificance and symbolism attached to the reappearance of the light at the
conclusion of Tenebrae was dramatically enacted by the senior choir boy.
He produced a lighted torch, which had been hidden behind the choir
stalls prior to the start of the service, and chanted ·~ Christi'. 9
1 Miscellanea, PL 177.889C. 2 HBS 59 p.31.

3 Breviary I p.375. 4 Paleographie Musicale XII p.6;
5 HBS 75 p.271. 6 Rationale VI.72 p.331. /Note G.
7 Rationale, PL 202.106D.
8 De Festis D.N.J.Cbr. P.1 8 126 in Eisenhofer, . Halld.lNch I p.514.
9 1?31 Ceremonial. PP: 226-7.
89
It is Durandus who records the wide range of symbolic interpreta=
tiona which tho last candlo had acquired by tho end of tho thirteenth
1
century. For it could represent
(i) the Blessed Virgin Mary in whom alone the faith remained.
(ii) Christ who was dead according to the flesh.
(iii) the faith which was hidden in the apostles.
(iv) the apostles' faith after their infidelity.
(v) the fire of the Holy Spirit which seemed exti~ct.
(vi) the renewal of Christ's light.

(vii) the commemoration of the Resurrection after the deaths of
the prophets.
It is not difficult to see why the light of the last candle came to be
identified with Christ or some aspect of the Godhead for those who
sang the last verse of the Benedictus in almost total darkness.
2
Christopher Wordsworth records a different symbolism which in
England became attached to the last candle on the hearse. This light
was sometimes known as the 'Judas candle' because the antiphon at which
that light was extinguished or hidden began 'Be that betrayed Him •••• '
(Traditor ~).3
1 Rationale VI.72 p.331. 2 Medieval Services p.168.

3 The name is found in medieval documents. (a) 'In j Judas de novo
facto adserviendum in choro per iij dies, videlicet diebus mercurii,
Jovis et Parasceven' (14o2-3). Memorials gJ Ripon III p.212, Surtees
Soc. Publ. 81. (b) 'In uno Judas de novo fact' p' oandel tenebrar'
deferend' festin' Pasch xxd.' R D.ymond, 'History of St Petrock's',
Transactions of~ Devonshire Association, Vol.XIV (1882), p.410.
90
Chapter Nine
THE NJU·iE AND ORIGIN OF TENEBRAE
(i) The ~ Tenebrae
The name by which the combined office of Matins and Lauds wae popularly
knot~ is first recorded by Peter Abelard in the first half of the
1 2
twelfth contury. John Beleth also uses the term a little later. But
the name is almost certainly much older. In the eighth-century Ordo 28
we read : septima •••• tenebratur candela;3 and the Regularis Concordia
of E•AD 970 states that the service is a commemoration of Jesus' three
hours on the Cross, and of the 'tenebrarum terror' which prevailed be~
cause of the Crucifixion.

4 However, in addition to the content of the
service, the time at which the office was sung and its conclusion in
darkness almost certainly contributed towards the name.
(ii) The Origin 2£ Tenebrae
(a) The traditional theories. It was formerly believed that Tene~

brae recalled the time when services were held in the catacombs5 or in
the darkness of a church building because of the fear of persecution;
6
or even that it commemorated the apostles' hiding in the upper room.
(b) The utilitarian theorz. According to this theory, which was

first propounded by Claude de Vert 7 at the beginning of the eighteenth
century, the practice of extinguishing the lights gradually at Tenebrae
emerged in the period when Matins and Lauds were sung in the very early
hours of the morning. With the gradual increase in daylight at dawn,
the church lights were put out one by one until, by the time the last
lamp was extinguished, there was sufficient natural light for the needs
of those at worship. The theory seems to have been accepted by tho re~
formers of the Holy Week liturgy in the 1950s; for, commenting on the
Decree of Maxima Redemptionis of 16 November 1955 which authorised the
liturgical changes, P.Jounel observed that Tenebrae had now been restored
1 Letter 1Q. (PL 178.340A): Atque ~ vulgo ~ dierum Vigiliae ,!!!;S-

cupantur Tenebrae. ('And so for this reason the night office of
these days is called Tenebrae.')
2 PL 202.105B.
3 'The seventh lamp is extinguished.' (OR 28.}0.)
4 PL 137.490B.
5 Amalarius, Lib.Off. 4.22 (PL 105.1201). 6 Houssaye p.46o.
7 Explication~ p.49 and p.292.
91
to the correct time of day 9 so that tho gradual extinguishing of the
1
lights coincided with the rising of the sun.
The theory, which is based largely on assumptions unsupported by

documentary evidence, must be challenged and exposed to further scrut=
iny. Dendy doubted the genuineness of the explanation; but, apart
from drawing attention to the phrase ~ ~ surgendum est which
~ ~
occurs in four of the earlier ordines,c he offered no theory of his o~Jn./

Any vindication of this theory depends on t~o major factors : (i) tho
time or times at which the office is known to have commenced and (ii)
the duration of the combined office of Matins and Lauds. We propose to
consider the length of the service first, and then to look at the time
at which it began, so as to calculate the hour at which the office ended.
1. The Duration of Tenebrae. Possibly the only reference in medieval

literature to the duration of Matins/Lauds is to be found in the Gemma
4
Animae of Honoriua of Autun. According to that writer Tenebrae lasts
three hours, the time that Jesus was on the Cross. A glance at the con-
tents of Matins and Lauds shows that the whole of the combined service
with ita fifteen psalms, nine lessons, Benedictua, Miserere, as well as
antiphons, responsories, versicles, kyrie's, and prayers can be perform=
ed in much less time than three hours. It is true that during the last
century in the Sistine Chapel in Rome Tenebrae, with its renowned sing-
ing of the Miserere and musical accompaniment,5 was known to last upwards
of two hours;
6 but in a medieval monastery without the same protracted
musical accompaniment and the contrived effects which made use of the
late afternoon sun and the frescos of Michelangelo, the entire office
must have been performed in rather less time. C.Butler was of the
opinion that the average length of time for Matins in the winter was an
hour and a half, and for Lauds between a half and three quarters of an
hour. 7 Honorius' three hours should be reckoned inclusively in much the
same way as Jesus' three days in the tomb, so that the service probably
began at the end of one hour and ended shortly after the start of the
third hour. Ninety minutes seems to be a more realistic figure for the
duration of Tenebrae, and at the same time does not contradict Honorius'
statement. At Seville Cathedral in the last century the service lasted
1 ~nouvel~ p.25. 2 Ordines 28.29; }OA.1; }OB.1; 31.11.

3 ~ ~ 2f Lights p.147. 4 PL 172.665.
5 Hare II p.297. 6 Newman p.14.
7 Benedictine ·Monachism pp.278-9. The present writer has in his poss-
ession a 1927 Dominican Office of Holy Week, in which a previous
owner of the book has inscribed in 1971 next to the rubric for Tene-
brae on Maundy Thursday, 'Tenebrae 9.30 am. Finish 10.15 am.'
92
1
one hour. When we come to examine tho time at which Tenebrac ended~
we shall allo~;oJ an upper limit of about t\•JO and a quarter hours dura=
tion for the purposes of our calculations.
2. 1h2 ~ fll; ~ Tenebrae began. t·Je havo already observed that in

the earlier period when Tenebrae was sung at the start of tho monastic
dayi the office began either at midnight or at the eighth hour of tho
.night (p.20). The pr1:1ctice of grCldi!Gilly exti:r.~guishing the lights at
Tenebrae emerged from a monastic milieu long after St Benedict had
changed the hour of rising for the monks from midnight to the eighth
hour of the night. Although there is evidence that the earlier time
continued to be observed in some monasteries, we may discount it for
our purposes with some justification. For it is inconceivable that
Tenebrae could have been such a protracted and prolonged office that it
lasted from midnight until dawn. Services in the West have never been
noted for their seemingly interminable duration. We must therefore
concern ourselves with the later time of the eighth hour.
The Benedictine injunction relating to the time of rising was ob=

served from the first day of November until Easter, regardless of the
date when that festival occurred. 2 Lauds, originally a separate office,
used to be sung at dawn, incipiente ~;3 but long before AD 700 it
had become customary to sing Lauds immediately after Matins in the early
hours of the morning.
4 We hope to show that by the eighth century, when
the service of Tenebrae is first attested, the office of Matins/Lauds
ended in darkness before sunrise; and that the utilitarian theory for
the gradual extinguishing of lights at that service cannot be upheld.
St Benedict adopted the traditional ecclesiastical system of divid-

ing the day and the night each into twelve hours of equal length in so
far as this was possible, given the chronometrical devices available at
the time - hence the qualification 'at whatever shall be calculated to
be the eighth hour•. Table 6 on the following page shows the earliest
1 Doblado p.284. 2 PL 103.872A. 3 PL 103.873A.

4 By the time of the compilation of Ordo 26 ~.AD 750, the union of
Matins and Lauds had already taken place. (See above, p.2 Note 3.)
The Regularis Concordia, which mentions that Lauds finished before
dawn and that Prime took place ~ facto (PL 137.490C), is too late
a document (c. AD 970) to be of much relevance for practices of the
eighth century, when the custom of extinguishing gradually the
lights at Tenebrae first became established. (Since Vatican II
Matins and Lauds have been separated.)
93
-----
<}
Da~_e Latitude SuMct Sunrise Minutes in 8th hour
~ .h2E£
18 March 40° 18.10 06.08 59.8 01.09-02.09
18 1'1arch 50° 18.09 06.08 59·9 01.09-02.09
18 March 60° 18.08 06.09 60.0 01.09~02.09
5 April 400 18.28 05.38 55.8 00.59~01.55
5 April 50° 18.38 05.29 53.4 00.59~01.54
5 April 60° 18.52 05.14 51.8 00.57~01.49
22 April 40° 18.45 05.13 52.3 00.51-01.43

22 April 50° 19.o4 o4.54 49.1 00.47~01.36
22 April 60° 19.35 o4.24 44.0 00.43-01.27
+ The times given in this column are ~nte meridiem, and correspond
to the hours approximately calculated in the Early Middle Ages.
Latitudes of various liturgical centres
Palermo 380"' Rouen 49°30' ..

Rome 42°* Mainz 50°
Poi tiers 46°30'"' Salisbury 51°1):
Paris 49°* York 54°"'
"' approximate
Table 6. The 8th hour by modern reckoning.
date (18 March) on which Maundy Thursday can fall, and also the latest
(22 April). The fifth day of April represents the mid-point between
the two extremes. It is quite clear that at the earliest date for
Maundy Thursday, one twelfth of the period of darkness (i.e. from sun-
et to sunrise) is, depending on the latitude, either one hour or almost
sixty minutes according to the modern system of measuring time; so that
everywhere between Rome and York on the 18 March a hour of darkness is
of virtually the same duration according to the former method of reckon-
ing time. After the vernal equinox the time differential begins to in-
crease in accordance with the degrees of latitude. The amount of day~
light increases in the northern hemisphere with a corresponding decrease

in darkness; so that the length of each hour of darkness grows smaller
more noticeably the further north one travels.
The tablo also shows the modern corresponding time for tho oighth
hour. This is arrived at by dividing the total period of darkness by
twelve and then calculating to what poriod during the night the eighth
division would correspond by modern reckoning. CeButler was right to
point out that ~ ~ eighth hour referred to the completion of the hour
and not to its inception; but when instancing the city of Rome~ where
on 25 March the sun rises and sets at six o'clock and there are exactly
twelve hours of darkness, he is incorrect in stating that the eighth
hour of the night lasted from 2.00 aemo to 3.00 a.m.9 and that~~
1
eighth hour indicates 3.00 a.m. For the first hour extends from 6.00
p.m. to 7.00 p.m., the second from 7e00 p.m. to 8.00 p.m., and so forth.
The eighth hour would therefore last from 1.00 aQm. to 2.00 a.m., and
2
this latter time would be the modern equivalent of at ~ eighth ~·
We are now in a position to calculate the varying times at which

Tenebrae finished, both geographical and seasonal differences being
taken into consideration. We have already indicated that we shall use
the longer time of two and a quarter hours for the duration of Tenebrae.
(i) 18 March. All three latitudes can be grouped together in view
of the almost-identical times of sunrise. Tenebrae starting at 2.09 a.m.
or thereabouts and lasting for two and a quarter hours would finish at
4.24 a.m. This is over one and a half hours before sunrise.
(ii) 5 April. Tenebrae would end :
0
at 4o one and a half hours before sunrise
at 50° one hour twenty minutes before sunrise
0
at 6o one hour and ten minutes before sunrise.
(iii) 22 April. Tenebrae would end :
at 40° an hour and a quarter before sunrise
at 50° just over an hour before sunrise
at 60° about forty minutes before sunrise.
Both the time allowed for Tenebrae and the figures given for 4o0
and 60° represent extremes. For the times relating to places on or
near the 50° line of latitude are of more relevance to our study in
view of the fact that this service with its gradual extinguishing of
lights had its origins in Northern Gaul or in Northern Germany through
both of which passes the fiftieth parallel.
1 Benedictine Monachism p.275.

2 It is known that the length of the hours was adjusted in N.Germany
throughout the year. The variation and adjustment is referred to
in the rubrics for the new fire ceremony in~ (II p.57). Seep. 133.
95
We have shown that on 18 March, the earliest possible date for
Tenebrae~ the entire office of ~mtins/Lauds terminated (nt places in
Gentral Germany and Northern Franco) over an hour and a half before
dat~ 9 and that even at the latest date of 22 April there was still
over an hour to pass before sunrise. The theory 9 therefore, that the
lights of Tenebrae were put out gradually as the amount of daylight in~
creased and the need for artificial illumination diminished c~nnot be

upheld. Moreover, it should be borne in mind that even on a cloudless
spring morning in March or April the interiors of somo churches in the
period under discussion would continue to have remained badly illumin=
ated in view of their architectural design and system of fenestration.
An explanation to account for the loss of light at the night office

must be sought elsewhere. Herbert Thurston in an uncharacteristic
lapse of scholarship wrongly thought that it grew lighter more quickly
in southern latitudes in late March and April, and supposed that north-
ern monks found it hard to get through the office of Matins without
1
better light. For more serious consideration is the possibility that
because of the funereal aspect of Tenebrae the practice of extinguish-
ing lights was inherited from a corresponding pagan custom according to
which a gradual decrease in the number of lights was regarded as a mark
of respect or honour for a deceased person. In view of the absence of
corroborative evidence, however, such a theory must remain conjectural.
Moreove~universal practice throughout Europe and the Near East involved
the lighting of a lamp or candle in honour of the dead rather than the
extinguishing of it. The theory, referred to above, 2 which explained
the Tenebrae-hearse as a device for counting the number of psalms sung
at the night office must also be discounted. For whilst the extinguish-
ing of candles may subsequently have been used for monitoring the pro-
gress of the gradual psalms, as we have argued, there is no reason why
this practice should originally have been adopted at Lauds any more than
at the other offices of the day when the psalms were sung aloud.
The most likely explanation for the emergence of this liturgical

feature at the night office is, in the opinion of the present writer,
to be found in the l1aundy Thursday rites in Jerusalem which were held
in and around the Garden of Gethsemane. Since the ceremonies held in
the original and historic locations could hardly be transferred out of
their Jerusalem milieu, it seems likely that in the gradual loss of
light at the night office an attempt has been made to commemorate the
1 ~ ~ Holy ~ p.262. 2 See above, pp.54-55.
96
events of that late Thursday evening within the context of & sa~vice
in church. Ferdinand Cabrol uaa the first to hint at the derivation

of Tenebrae from the Jerusalem rites; but he did not attempt to elab=
orate the connection between the western night office and tho historical
1
events in Gethsemane. He merely noted that the offices of Tenebrae on
Maundy Thursday and Good Friday may have been influenced by the local=
ised Jerusalem rite of Maundy Thursday evening. Cabrol was poseibly
closer to the truth than he realised. For in searching for the origin
of Tenebrae s gradual loss of light, it is significant to find that
0
the first recorded instance of this liturgical phenomenon occurs during

2
Lauds on Good Friday which was sung at the corresponding time, or not
long after the time, that Jesus was arrested in Gethsemane. Since the
lights, extinguished at Lauds and subsequently at Matins and Lauds, are
unlikely to represent the torches which we are told the chief priests'
soldiers carried (Jn 18:3), it is likel~ that they symbolise a non=
luminary element in the story. Later medieval commentators may have
been preserving an ancient tradition when they compared the extinguish=
ing of the lights with the flight and desertion of the apostles.3
Alternatively, the quenching of the lamps may have symbolised the heavi-
ness of the eyes of the disciples who were unable to stay awake and
keep watch with Jesus.
4
1 b!! Ori~ines p.181. 2 Stage 3 of Tenebrae pp.13=16.

3 See p.3 and p.41.
4 The commemoration of the three hours of darkness on Good Friday is
much more likely to be associated with Tenebrae of Holy Saturday
which follows the Crucifixion.
97
PART II
The New Fire Ceremony

Chapter One
THE DEV.t!ILOPMENT OF THE C~ONY
In order to observe the changes which occurred within the new fire
ceremony and to trace more easily the influences which contributed to
these changes~ we have classified the different known forms of the
ceremony as stages of development which we have entitled modes.
(i) The older ~ tradition
Mode A1
1
Elsewhere we have discussed the provision of new fire in mid-eighth
century Rome, attested in the letter of Pope Zachary to Boniface. We
suggested that the three large lamps were reserved during the consecra-
tion of the Host at Mass on Maundy Thursday. These remained in the
Cathedral of St John Lateran and continued to burn !9 loco secretiore,
until the fire was hallowed for use on Holy Saturday and used to light
the two Vigil-candles and kindle the other lamps of St John's.
Mode A2
In the eighty or so years which separate the pontificate of Zachary

from the visit of Amalarius to Rome in AD 832, a development occurred
within the Roman Church in the procedure for the provision of new fire
during the Triduum. It had been the practice at Rome in the time of
Zachary to obtain new fire on Holy Saturday from the flame of one of
1
the three lamps which had been hidden from view on Maundy Thursday.
However, according to the subsequent testimony of Archdeacon Theodore
new fire was kindled on Good Friday for use at the night office of Holy
Saturday. 2 The clarity of this statement is matched by the resulting
uncertainty regarding the provision of fire at other times during the
Triduum. For it raises two important and closely-connected questions
(a) Did the new fire for the ceremonies of Holy Saturday continue to be
taken from the three lamps reserved on Maundy Thursday? (b) Was the new
fire, kindled on Good Friday, destined only for use on Good Friday
night and subsequently extinguished at the conclusion of the night office
of Holy Saturday? Since Theodore is silent on this matter, there are
two possibilities to be considered. (i) If the practice of reserving
1 See pp.9-10, and especially Part II Ghapter 7(a) pp. 170-77.

2 'k ipsa s!!! (i.e. Good Friday)~ ignis accenditur, ~quo reserv-
~ usque ad nocturnale officium.' (Amalarius, Lib de Q!:l.Ant. XLIV .2)
99
the three large lamps on Maundy Thursday had been diacontinued 9 it
would follow that the new fire 1 kindled on Good Friday 9 was reserved
not only for the night office at tho end of that day, but also for the
Vigil during the evening of Holy Saturday. (ii) On the other hand the
continuing reservation of Maundy Thursday's fire would have restricted
the use of Friday's newly-kindled fire to the night office of Holy Sat=
urday, since a vecond reserved source of fire would have been super-
fluous. Theodore's silonco, however, regarding the three lamps is not
decisive; for the conversation between the archdeacon and Amalariue
had revolved around the loss of illumination at the night office, not
the provision of light for the Vigil of Holy Saturday.
The external evidence, which it is possible to adduce in support of

both the above views, is necessarily inconclusive. For the practice of
reserving fire on Maundy Thursday was well-established in the Gallican
1
Church; and featured in the Pontificale Romano-Germanicum, the service-
book which the Roman Church adopted in the tenth century. On the other
hand the Pontifical of Poitiers, which bears the marks of Roman influ-
2
ence, records that the Easter candle was lit with a flame kindled ~
on Good Friday (p.215); whilst the reservation of fire on Good Friday

is attested in three ordines and four sacramentaries. 3
It is safe to assume, from the evidence of Amalarius, that at Rome

the new fire was kindled shortly after the conclusion of the liturgy
on Good Friday afternoon :
In the Roman Church all fire is extinguished and (subsequently)

rekindled on Good Friday. By this action fire, which is fuelled
and maintained by stocks of firewood, imitates the principal
source of fire, that is, the sun in the sky, which hid itself from
human eyes from the sixth to the ninth hour at the time of our
Lord's passion; so that those who wickedly rejoiced at the shame
of their Lord and Creator might not enjoy its light. For this rea-
son that fire, which is obtained for our use, may be extinguished
on Good Friday at about the sixth hour and renewed at about the
ninth hour of the day.4
Howeve~we can only surmise that it took place somewhere in the vicinity
1 See below, Mode B2. 2 See especially Ft IV Chap.16 pp.327-8.

3 For these references see below, p.101.
4 'In Romana ecclesia extinguitur totua ignis in sexta feria et reacc-
enditur. In hoc facto imitatur ignis, fotus et conaervatus per con-
gesta lignorum, principalem ignem, id est solem corporeum, qui ab hum-
anis obtutibua se abscondit tempore passionis Domini a sexta ora usque
ad oram nonam, ne suo lumine fruerentur qui male gaudebant de ignom-
inia domini sui et creatoris. Hac ratione ignis iste, qui nostris
usibus procuratur, poteet extingui in sexta feria circa sextam oram
diei et renovari circa nonam horam diei.' (~.Off.4.22.2.)
100
of St John Lateran wherein thG night office of Holy Saturday would be
haldo Similarly we can but guess at the means employed in th®
production of that firo. (This evidence is treated more fully on p.178o)
( ii) ~ Gallic an tradi tiona
Mode B1
The use o£ fire 9 reserved on Good Friday, with which to light the
1
Easter candle is attested in three ordines and four sacramentarieoo
According to Ordo 23 it is used to light the two Vigil-candles which
were a feature of the old Roman tradition (§ 24). This light is to be
identified with that, described in Ordo }DB, which was reserved at the
conclusion of Lauds on Good Friday and which was also used for the pro-
vision of illumination at the night office of Holy Saturday. The pract-
ice of reserving Good Friday's fire must be of considerable antiquity,
and suggests a period when the liturgical situation was characterised
by a complete absence of light between the end of the night office of
Good Friday and the Vigil of Holy Saturday. It also dates from a time
anterior to that in which Tenebrae had reached its final stage of devel-
opment,2 since the loss of light at ~~undy Thursday's night office
caused the need for a fresh supply of fire the same day. The reserva-
tion of ~ fire is possibly a Gallican development of the Roman pract-
ice attested by Zachary {Mode A1), and took place at the conclusion of
the night office of Good Friday because of the necessity of illumination
at that service; whereas, in the circumstances familiar to Zachary,
the singing of Matins/Lauds of Good Friday in the Church of Sta Croce
in Gerusalemme, followed by the celebration of that day's liturgy in
the same church, enabled the fire to be reserved at St John Lateran in
circumstances undisturbed by any liturgical activity.
Mode B2
According to this alternative Gallican tradition the flame for

lighting the Easter candle was obtained from newly-kindled fire, and
not from an already-existing source of hallowed fire which had been re-
served for that purpose (Mode B1). The earliest evidence for the kind-
ling of new fire, as opposed to the reservation of ~ fire, comes from
Ordo 26 (I 3). The ceremony takes place on Maundy Thursday; and it is
not difficult to see why it occurs on that day. For within this Gallican
1 Viz. OR 17.103; OR 3QA.15; OR 30B.28; ~ (p.68); GePr (p.55);

~(p.52); ~ (pp.92-3).
2 Stage 6, pp.18-19.
101
tradition the production of new fire is closely linked to the perform=
anco of Tencbrao on all three days of the Triduum = a situation fittost0d
by this ~· In churchoa in which the development of Tenebrae had
reached one of Stages l' 4, or 5 9 the supply of fire for the night office
and Vigil of Holy Saturday was obtained from the lamp hidden and reserved
at the conclusion of Lauds on Good Friday. With the loss of fire at the
end of the night office of Maundy Thursday, however, it was now necessary
to ensure that a supply of fire was available not only for tho night
offieo of Good Friday, but aloo to kindlo the liturgical lights for the
1
Maae or Hasses of Maund;y Thursday. There is one recorded instance of
the reservation of 2!,g fire on Maundy Thursday; but according to the
majority of our sources the fire was kindled anew on that day.
Tho early ordines2 and Alcuin3 state that the fire was then reserved
for lighting the Easter candle on Holy Saturday; but they omit any re=
ference to the blessing of the fire on Maundy Thursday. However, accord-
ing to PRG, which attests the same ceremonial, the fire was first bless-
ed bcfo~being reserved.
4
Although the difficulty arising from the
subsequent hallowing of the fire on Holy Saturday precludes this docu-
ment from being regarded with complete confidence as corroborative
evidence - see Mode B3 - nevertheless it is very likely, in view of
other similarities of ritual between ~ and the above-mentioned ordines,
that a blessing of the new fire prior to its reservation should be under-
stood. According to this arrangement fire, required for any of the sub-
sequent services during the remainder of the Triduum, could be taken
from the reserved flame without the requirement of a preliminary act of
benediction.
Howeve~according to~ the Mass of the Chrism was celebrated at

the third hour of the day. 5 Since the new fire was kindled no earlier
1 The Gradual of St Gregory, cited by Macri in Hierolexicon (pp.141-2)

Deinde venit Archidiaconus ante altare accipiens lumen, quod quinta
feria absconditum fuit, faciensque crucem super cereum et illuminans
eum, ac benedicens, dicente ipso Lumen Christi, respondent omnes Deo
Gratias. (The writer's italics.) 'Then the Archdeacon comes in front
of the altar. He takes the light which was hidden away on Maundy
Thursday, and making the sign of the cross over the candle, he lights
the candle and blesses it. Then he acclaims "The Light of Christ"
and all reply "Thanks be to God". 1
The form of words used recalls the corresponding rubrics in the
ordines and sacramentaries referred-to above (p. 101). However, the
occurrence of quinta and the reference to ~ Christi suggest that
the Gradual is a later document.
2 OR 26.3; OR 27.6; OR 28.25; OR 31.29.
3 Q! ~.Q!!. (PL 101.1205C). 4 II p.57 6 216. 5 II p.59.
102
1
than the fifth hour according to the same pontifical, the flame for
the lights at thic Masa clearly could not be takon from tho new firo.
The rubric for the conclusion of Lauds on Maundy Thursday, however 9
states that, before the Benedictus is sung, the last lamp is either
extinguished, or withdrawn and reserved. 2 The writer believes that the
purpose in reserving the lamp at this service was to provide light for
the above-mentioned t4ass in those churches where it was still celebrated.
B,y the tenth cent~-y the Mass of the Chrism was no longer celebrated in
many churcheo; so that it was no longer necessary to reserve a flamo
at the conclusion of the night office of Maundy Thursday. The kindling
of the new fire later that same day took place before the late afternoon
Mass which commemorated the institution of the Lord's Supper.3
Mode B3
The new fire ceremonial attested in PRG bears a close similarity to

that described in the ordines of Mode B2. PRG differs from these docu-
ments, however, in that in the ritual of Holy Saturday there is a seem-
ingly-superfluous hallowing of the new fire which had previously been
blessed on Maun~ Thursday. It must be admitted that a second act of
4
benediction is not unknown in the ceremonies relating to the new fire;
but it is certainly unusual. This second blessing contained in ~ may
be a survival of an older Gallican tradition in which the new fire was
kindled and hallowed on Holy Saturday. Its continuing presence in the
tenth-century ~ reveals the composite character of that document, and
is in all likelihood the result of a synthesis of different traditions
relating to the new fire. This is particularly borne out by the
benediction-formulas used on Holy Saturday and especially in the use of
the ~ mundi Conditor. This prayer constitutes the formula for the
blessing of the Easter candle attested in the Gelasian sacramentaries,
and is not strictly a blessing for the new fire. 5 PRG also includes
Formula H for the blessing of Holy Saturday's fire. 6 This prayer differs
from the other formulas in that both the fire and the candle which is
lit from the fire are included in the words of sanctification.
This twofold blessing of the new fire is also found at Salzburg.?
1 See below, Chapter 4 p.133. 2 II p.57 § 214.

3 II p.58 § 220.
4 For instance, the so-called double blessing of the lamp and of the
candle in the Mozarabic rite. See p.~91.
5 See below, Chapter 3 p.126. and Part IV Chapter 10 pp.289-90.
6 Appendix 5 : The Benediction-Formulas.
7 1507 Missal fol.lxxxv (MT) and fol.xciii (HS).
103
~here the new fire was blessed on £4aundy Thursday with Formulas A9 Bv
1
and C, and on Holy Saturduy with the~~ Conditor and Formula H0
2
as in PRG on both days. From other rubrical similarities there can be
little doubt that the Niasal of Salzburg dorives directly from ~· For
even after the lapse of five and a half centuries the new fire ceremonial
of Salzburg is still recognisably the ritual prescribed by ~- (Mode Bj
is alluded to in the twelfth~century Pontificale Romanum.')
There is no evidence to suggest that the new fire, once kindled,

was reserved in an unhallowed state for the duration of the Triduum;
and that only fire taken from this source was blessed on each of these
three last days of Holy Week.
Mode B4
In Mode B2 we observed that the loss of fire occasioned by the

extinguishing of the lights at each of the three performances of Tene-
brae was made good by the kindling of the new fire on Maundy Thursday
and by its subsequent reservation for use during the remainder of the
Triduum. Within the same liturgical milieu, in what could be argued
was an obvious and logical development in the ritual for the provision
of new fire, there emerged the practice of kindling fire on each of the
4
days of the Triduum. It is first attested in Ordo 29 and became wide-
spread throughout the Western Church through ita adoption by a very
large number of monasteries. 5
It might have been expected that with a fresh supply of fire on each
of the three days the need to reserve Maundy Thursday's fire would dis-
appear. However, according to Ordo 29 there is not only a production
1 These two formulas for blessing the fire on Holy Saturday were also
used at Mainz, Ratisbon, and Abo. (See Table20).
2 For instance, the choice of times for kindling the new fire on
Maundy Thursday. (~ II p.56 8 215 and 1507 Missal fol.lxxxv.)
3 PR XII S xxxii.i.
4 OR 29.14, .28, and .45. H.A.P.Schmidt wrongly supposes that PRG
records the kindling of new fire on the three days (Hebdomada Sapcta
II pp.820-21). The pontifical actually states that on Good Friday :
~ deportatur, 'a light (= fire) is brought'(II p.86 § 304); and
on Holy Saturday : deportatur ~ quod quinta ~ fuerat excussum,
'fire is brought which had been kindled on Maundy Thursday' (II p.94
8 342). There is no mention of a production of new fire on either
of the last two days of Holy Week.
5 Lanfranc, Decrees.(PL 150.467B) and John of Avranches, ~-~ Q!I.
~· 53 (PL 147.49A). See Tables 8a and 8b.
1o4
of new fire on each day of the Triduum; Maundy Thuraday 0 s fire is also
reserved for lighting the Easter candle on Holy Saturday. Sinco the
Candle could have been lit with the fire kindled on Holy Saturday 9 the
reservation of Thursday's fire would appear to be a superfluous sur=
vival of a former practice. Ordo 29 is in fact a composite document.
The ceremonial of Holy Saturday is unique; for it combines the Roman
practice of lighting the two Vigil-candles prior to the reading of the
prophecies with the Gallican ritual of kindling the Easter candle. So
that the Vigil=candleo are lit with the fire struck on Holy Saturday 0
whilst the Easter candle is lit with the fire reserved from Maundy
Thursdayo In the Holy Saturday liturgy of Ordo 29 we have a synthesis
of both the Roman and the Gallican traditions.
Mode B5
As in Ordo 29 we find in the Pontifical of Poitiers elements deriv=

ing from both the Roman and the Gallican Vigil traditions. There is a
triple kindling of fire during the Triduum and the two Vigil-candles
1
are a feature of the liturgy of Holy Saturday. It differs from Ordo 29
in that the fire for lighting both the Easter candle and the Vigil-
candles is that kindled and reserved on Good Friday. It is possible
that this practice is derived from the Gallican tradition, attested in
Mode B1, of lighting the Easter candle with fire reserved on Good Friday.
Howeve~, in view of the presence of Roman elements in other aspects of
2
Poitiers' Holy Saturday liturgy, it is more likely that the fire is
kindled on Good Friday and subsequently reserved, in imitation of the
Roman practice described by Archdeacon Theodore and discussed in Mode A2.
(iii) The Spanish and Italian traditions
Mode C1. The Mozarabic, Ambrosian, and Beneventan rites.
In the three above-named traditions the absence of light during

4
Good Friday3 or the loss of light at the Passion on that day necessit-
ated the provision of newly-kindled fire on Holy Saturday for the light-
ing of the Easter candle, since in none of these three traditions5
1 Poitiers p.138 and p.215.

2 For instance, the use and significance of the Vigil-candles (p.215)
and the reference to the wax Agnus Dei's (p.218).
3 Antiphonary of Leon p .276. 4 Beroldus p.106.
5 It is very likely that Beneventum's Good Friday liturgy closely
followed that of Milan. (Borella p.109.)
105
was a flame rGservad from either Maundy Thursday or Good Friday. The
single production of new fire on Holy Saturday at Ripoll and at Vich
is almost certainly duo to Mozarabic influence~ and there is evidence
for this tradition within the Gallican Church. For one of mvo two
records a blessing of the new fire on Holy Satur~
1
principal manuscripts
day, so duplicating the hallowing of the fire on Maundy Thursday. It
strongly suggests that formerly in one tradition the fire had been
kindled on Holy Saturday. Tnis may have been prLuitive Gallican pract=
ice. On the other hand it may be argued that this was the result of
Milanese influence; for that rite is YJaown to have been used in South=
2
ern Germany as late as the eleventh century. The single kindling of
fire on Holy Saturday is also attested by Lanfranc, and may have feat=
ured in the pre-Conquest English Church. Table 10a lists the early
evidence for this mode (p.112).
Mode 02. The later Roman tradition.
When the Roman Church adopted the Romano-Germanic Pontifical (PRG)

in the tenth century, the arrangements for the production of the new
fire, as prescribed by that document, were not adopted into the Roman
rite in their entirety. For at Rome the new fire had previously been
kindled on Holy Saturday; 3 and, although~ directed that the new fire
should be produced on Maundy Thursday, this day remained unchanged after
Rome's adoption of that pontifical. This is clear from the rubric of
the twelfth-century Pontificale Romanum :
Hora autem quinta vel sexta, novus ignis, si non fuerit excussus
in caena domini, iuxta morem quarumdam ecclesiarum, excutiatur
hoc die extra ecclesiam de crystallo, vel etiam alio modo fiat. 4
The churches in which it was customary to kindle the new fire on Maundy
Thursday were those whose rites were regulated by ~ and other early
ordines. One of them was the Church of Salzburg whose new fire cere-
monial even in the sixteenth century was still fundamentally the same
as that prescribed in ~·
1 Manuscript C (II p.96 § 344). 2 Borella p.105.

3 See below, Chapter 7 p.179.
4 'Now on this day (Holy Saturday) at the fifth or sixth hour let the
new fire be produced from a lens outside the church - or it may be
done in any other way - if it has not already been kindled on
Maundy Thursday, according to the custom of some churches.'
PR XII xxxii. 1 p. 238 .
106
!t is not immediately clear why the practice of kindling the new
fire on Maundy Thursday and reserving it until Holy Saturday was not
adopted by the Roman Churchi especially since it had formerly been the
1
custom at Rome to reserve Maundy Thursday's fire. It may have been
felt more convenient not to kindle anew and reserve the fire on that
2
day, especially as the use of two churches for the ceremonies of tho
Triduum may have caused difficulties in the reservation or the trans=
portation of the new firo. On the other hand there may have been some
reluctance to change what had become a well=established practice, which
was also observed in the contemporary rites of Beneventum and Milan.
1 See pages 170ff.

2 The Cathedral of St John Lateran and Sta Croce in Gerusalemme.
107
--------- ------------------
Chapter Two
THE TRIPLE AND TH.ti: SINGLill PRODUCTION OF FIRE
(i) The triple production !Jl !!!:!,
There ~ere two main procedures for the provision of fire on each of the
days of the Triduum~ both of which have been described in the previous
ehaptero According to the arrangement outlined in Mode B2, tho new fire
was kindled on Maundy Thursday and reserved for use on the following
two days. Table 7 presents the evidence for this particular mode. This 9
the earlier of the two traditions involving a threefold production of
fire, was almost ever,ywhere replaced by the alternative procedure, out-
lined in Mode B4 (p.104). This arrangement involved a separate act of
kindling fire on each day of the Triduum to replace that lost at the
conclusion of each of the three night offices.
Church/ ~ Source
Document
, Ordo 26 750-75 OR 26.3

(Ordo 27) 750-800 OR 27.6
Ordo 28 £.800 OR 28.29
Ordo 31 880-900 OR 31.29
PRG £·950 Vol.II p.57 8215
Or do Corbeiensis 950-1000 ~ 3·13.34
p.126 M 1145
Alcuin .£_.1000 PL 101 .1205C
Avellana 11 c. PL 151.881B
Cassino c.1100 DAMR 3.15.34
p:-;i'41 M 1139
Salzburg 1507 Missal fol.lxxxv
(Feasey, !h! Paschal Candle, p.355, claimed that the practice

was attested in a missal of Auch, in a sacramentary of Albi,
and at Toulouse in 1555.)
Table 7. Evidence for the single kindling of new fire

and for its reservation on Maundy Thursday.
In addition to the implicit evidence of the early ordines and

other documents, the loss of fire on each day of the Triduum may be in-
1
ferred with confidence from Lanfranc's Decrees; but it is clearly
attested by Rupert of Deutz, who remarks, 'Having lost the fire (amisso
1 PL 150.467B.
1o8
igne) which was extinguished at Matins 1 for thos~ three days we resort
= 1
to a stone. 0
Durandus also records that firos wero extinguished on the
2
threo days. Centuries earlier Amalarius had explained that the fire
was rekindled during the Triduum, rather than being allowed to remain
extinguished for the whole of that period, because of human weakness
services could not be held in church without light;3 and it enabled
food to be cooked.
4
Churc!l/ Date Source

Commentator
Ordo 29 870-90 OR 29.15, .29,

and .45
Rupert of Deutz _£.1111 PL 170.149A
John of Avranchea _£.1070 PL 147.49A
St Mary's, York £.14oO HBS 75 p.275
Rouen £-1700 De Mol eon p. 299
(Grancolas stated c.1730 that there was still a threefold

kindling of fire ;t Reima and Cluny - Commentariua p.316.)
Table 8a. Firm evidence for a kindling of fire

on each day of the Triduum.
Table 8a presents the unequivocal evidence for a triple production

of fire by three separate acta of ignition. The rubrics of a number of
customaries and service-books are too imprecise to allow us to state
with complete confidence that there was a threefold kindling of new
fire during the Triduum. Some of these documents refer to a blessing
of the fire on each of the three days; whilst others describe a repet-
ition on Good Friday and on Holy Saturday of Maundy Thursday's new fire
procession. All the sources which do not specifically refer to the
kindling of fire on each of the three days have been assigned to Table 8b.
However, in each instance it is safe to infer that there was a triple
kindling of fire, and that the newly-kindled fire of Maundy Thursday
was not reserved on that day.
1 B! ~.Off.V (PL 170.149A). 2 Rationale VI.80 p.350.

3 Propter nostram infirmitatem reaccenditur, quia sine lumine non
possumus nostra officia peragere in Ecclesia (Lib.Off. 4.22).
4 ~·~ Q!:sl•.&ll• XLIV .6 : ut ex illo cibi coquantur:-
109
Church/ Date Sourc®
~·1onastocy
Regu!aria ConcQrdia £.-970 PL 137.491B &.494C

Fruttuaria c.1000 Albers IV p.54
Cluny c.1000 Albers II p.18
German Monasteries 0".1000 Albers V pp.32 & 38
Fleury C'.1ooo Albers V p.143
Farfa c.1000 Albers I ppo48 9 52,54
St Benigne,Dijon 11 c ~ 3.13.34 (M1150)
Lanfranc £-1070 PL 15Q.467B
St Paul's~ Rome 11 c ~ 4.22 p.124
Nidaros 13 c ONE p.232. I (M1041)
\'lorcester £-1250 Antiphonary p. 69
Evesham .£-1250 HBS 6 cola 80,88,90
Canterbury 13 c HBS 28 p.274
Norwich £-1265 HBS 82 PP• 81,88,91
Reims 14 c DAER 4.22.5 p.97 (M
Cl~ 1510 tussal fol.xlix./261)
Braga 1512 Missal np
St Martin d'Ainay, 1531 ~ 4.22 p.125
lzy'on (M 1?5)
Bayonne 1543 Missal pp. 42 &46
Valladolid 1568 King, &f§ p.214
Tibaes 16?4 King, ~ p.268
Table 8b. Additional evidence for the triple production

of new fire.
Regardless of whether the new fire was kindled and reserved on

Maundy Thursday or kindled on each of the three days, the schedule of
events relating to the provision of new fire during the Triduum would
assume the following pattern :
Maundy Thursday (i) Loss of fire at the night office.

(ii) Production of fire during the day for:
(a) Mass of the Lord's Supper
(b) Lotio
(c) Matins/Lauds of Good Friday
Good Friday (i) Loss of fire at the night office.

(ii) Production of fire during the day for:
(a) Mass of the Pre-sanctified
(b) Matins/Lauds of Holy Saturday
Holy Saturday (i) Loss of fire at the night office.

(ii) Production of fire for the Vigil.
In some of the earlier documents which record a threefold procession

of fire, we find that there was a gradation in the seniority of the per-
sonages who bore the fire on the second and third days. The assignation
of the function to the bishop or abbot on Holy Saturday indicates that
110
the fire ~hich lit the Easter candlo was considered to ba of groatar im=
portance and significance than the fire of the previous two days. Table 9
illustrates the gradation in rank of those who bore the new fire.
Thursday Friday Saturday
Ordo 26
PRG
:!?oiW:ers
Alcuin
1J s
M
Sacriota.n
Sacristan
Archdeacon
Prior
Junior Bishop
Abbot
Ordo 29 M Sacristru:~. Prior Abbot

Re~ularis
M Sacristan Deacon Prior
Concordia
ClUilY
Far fa
Fleury
Corbie M Sacristan Prior Abbot (or Bishop)
Dijon
Lanfranc
Reims
St Mary's, Abbot's principal
M Sacristan Prior
York chaplain
Salzburg s Sacristan ? Archdeacon
s = secular M::. monastic

(For references see Tables 7 & .§_)
Table 9. The bearers of the new fire.
Dendy's statement 1 that the evidence for the threefold blessing of

the new fire is primarily monastic is correct. Liturgical development
and the evolution of new forms of ceremonial are more likely to occur
in the conducive surroundings of a monastic institution. Nevertheless,
according to the eighth-century Ordo 26, the triple production of fire
was well-established in both the cathedral and the monastic traditions
by the middle of that century (Table 9); and in the absence of further
evidence the judgement should be withheld that the practice arose out
of a monastic milieu. The appearance in Ordo 26 of the cathedral
officials (Section 19) prior to their monastic counterparts (Section 20)
indicates that the ~ was compiled for use in a cathedral church in
the first instance.
1 The Use ~ Lights p.142.

111
(ii) ~ single production 2! !!!£
It is generally agreed that the ceremonial surrounding the product=

ion of new fire on Holy Saturday according to the Mozarabic rite has
much in common with the ritual attested by the documents illustrating
1
the former Jerusalem liturgy of Holy Saturday; and it is possible to
1
detect in the Milanese rite also the influence of Jerusalem. In our
discussion of the provision of new fire on Holy Saturday in the previous
chapter (Modes £j and gg), we suggested that the production of fire on
Holy Saturday at Rome may have been the result of both liturgical changes
within her rite and the utilisation of two church buildings for the ser=
vices of the Triduum (p.106). Also worthy of consideration is the poss=
ible influence of the ritual surrounding the 'miraculous fire' at the
Holy Sepulchre, which may have resulted following the renewal of contact
between Rome and the Holy Land and the establishment of the Latin Patri-
archate of Jerusalem in 1099.
As observed in our discussion of Mode C2, PR XII recognised that the

production of fire took place on Maundy Thursday 'in some churches'.
Subsequent documents relating to the Roman rite confine the ceremony to
Holy Saturday; and the influence of the Franciscans, who were instru-
mental in the popularisation of the Roman rite, ultimately ensured that
the Gallican-derived Benedictine practice of kindling new fire on three
days yielded to the Roman observance of a single production of new fire
on Holy Saturday (Tables 10a and 1Q2).
1. Leon 10 c 18. Apamea 1214

2. Beneventum 10/11 cc 19. Nidaros 1200-20
3· Ripoll 1038 20. Haymo 1243-4
4. Vich 10}8 21. St-Pierre-sur-Dive 1273
5. Lanfranc c.1070 22. Lesnes 13 c
6. Leofric (?) - 10/11 cc 23. Salisbury 13 c
7- Wulfstan 11 c 24. Hereford 13 c
8. Chester 11 c 25. St Vedast,Arras £·1300
9· Milan 12 c 26. Fontanelles £·1310
10. Cistercians 1119 27. Exeter 1337
11. Ordo XI c.1140 28. Strasbourg 1364
12. Rome £.1150 29. Westminster 1362-68
13. Gilbertines _£.1150 30. Durham 14 c
14. Ireland c.1150 31. Laon 14 c
15. Beleth C'.118o }2. Lyre 14 c
16. York 12 c 33· Toul,St Epvre 14 c
17. Sicardus o.1200
Table 10a. Early evidence for the single production of fire.
1 See Appendix 13.

112
1. Abo (1522) 18. Cosenza (1549) 35. Poitiers (1524)
2. Amiens (1555) 19. Dominicans (1482) 36. Ratisbon (1570)
3. Angers (1489) 20. Esztergom (1501) 37. Rennes (1523)
4. Aquileia (1519) 21. Freisingen (1487) 38. Rouen (1497)
5. Arbuthnott (1491) 22. Hamburg (1509) 39· St Malo (1503)
6. Arras (1508) 23. Hildesheim (1499) 40. Saragoasa (1552)
7. Barking (14o4) 24. Langres (1492) 41. Sene (1520)
8. Basel (1488) 25. Liege (1492) 42. Seville (1507)
9. Bremen (1511) 26. Lund (1514) 43. Spirea (1512)
10. Breslau (1483) 27. Mainz (1507) 44. Tongres (15 C.)
11. Burgos (1546) 28. Melk (1495) 45. Tournay (1540)
12. Bursfeld (1498) 29. Minden (1513) 46. Trier (1487)
13. Carmelites (1504) 30. Narbonne (1528) 47. Uzes (1495)
14. Cassino (1507) 31. Noyon (1541) 48. Valence (1504)
15. Chezal-Benoit(1531)32. Osma (1561) 49. Verdun (1481)
16. Cologne (1494) 33· Palencia (1568) 50. Wurzburg (1477)
17. Cordoba (1561) 34. Passau (1503)
Table 10b. Evidence for the single production of fire: 14oo-1570.

(References are given below)
References for Table 10a

(1) Antiphonary p.280. (2) Hesbert p.188. (3) Sacram. p.92. (4) Sacram.
pp.4=5· (5) PL 150.466-7. (6) Missal p.223. (7) HBS 56 p.536. (8)
Albers IV p.209. (9) Beroldus p.109. (10) Nom.Cist. p.1o4. (11)
PL 78.1o41C. (12) OEL p.61. (13) HBS 59 p.39. ~ Missal p.126.
(15) PL 202.110B. f16) Missal p.109. (17) PL 213.322D. (18) DAER 4.24
p.160, M25. (19) ONE p.232. (20) Van Dijk II p.245. (21) DAMR 3.13.35
p.127, M1156. (22~HBS 95 p.47. (23) HBS 91 p.19. (24) Missal p.105.
(25) HBS 86 p.160. (26) DAMR 3.15.5 p.141, M1187. (27) HBS 37 p.322.
(28) DAER 4.22 p.162, M3~29) HBS 5 col.574. (30) Missal p.185.
(31) ~ 3.13.35 p.127, M1164. (32) and (33) ~ 3.15.5 p.141,
M1154-5 and M1160.
References for Table 10b

(NB M.= Missal.) (1) Manual p.238. (2) Pontifical p.18. (3) M. np.
(4) M.fol.91. (5) M.p.150. (6) M.fol.lxiv. (7) HBS 65 p.101.
(8) M.fol.xci. (9) M.fol.lxxxv. (10) M.np. (11) M.fol.ciii.
(12) M.fol.lxxxviii. (13) M.fol.xcv. (14) M.fol.91. (15) DAMR 3.15
p.141, M1180. (16) M.fol.cxxii. (17) M.fol.ciiii. (18) M.fol.115.
(19) M.fol.69. (20) M.fol.lxxiii. (21) M.fol.ciii. (22) M.fol.xc.
(23) M.fol.xcvi. (24) M.fol.lxv. (25) Ordinary np. (26) M.fol.xc.
(2?) M.fol.xcii. (28) M.fol.lxvii. (29) M.fol.ciii. (30) M.fol.lxxxiv.
(31) M.fol.lix. (32) M.fol.lxxxix. (33) M.fol.c. (34) M.fol.lxxxiv.
(35) M.fol.lxix. (36) Cerem.np. (37) M.fol.lxxii. (38) M.np. (39)
M.np. (4o) M.fol.lxxii. (41) M.fol.c. (42) M.fol.lxxv. (43) M.fol.xciii.
(44) Ordinary p.164. (45) M.fol.lxix. (46) M.fol.cii. (47) M.fol.lxiii.
(48) M.fol.liiii. (49) M.fol.lxiii. (50) Ordinary np.
113
Chapter Three
THE BLESSING OF THE NEW FIRE
( i) Preliminary procession psalms, ;utaey, ~ reading.
In many of our sources mention is made of a procession of clergy and

1
people to the place where the n~w fire was to be kindled; and tho n0g~
ative characteristic of movement in silence is found in a number of the
2
early ordines. Included amongst these aro three ordines Romani, the
at Corbie mentioned by Mart~ne,
4
Pontificale Romano-Germanicum,3 the 2£9e
and the Customary of the German Monasteries, 5 all of which are of north=
ern Gallican provenance, and were compiled before AD 1000.
At a relatively early date and almost certainly originating in a

monastic milieu where liturgical experimentation and elaboration of cere-
monial were more likely to have occurred, a preliminary procession took
place either to the accompaniment of or followed by the chanting of the
first or all of the penitential psalms. The earliest recorded use of
6
Psalm 50 on this occasion comes from Farfa at the beginning of the
eleventh century; but the practice may have originated at Cluny in the
tenth century. Table 11 shows that it featured prominently within the
monastic tradition; and survived in places into the sixteenth century.
On the other hand the chanting of all seven psalms, either during the
procession to the new fire or at the place where the fire had been kind-
led, was a feature of the rites of a number of central European churches,
even though it is first attested in the French monastery of Corbie.
7 ... 8
(Tables 12 and~.) At Breslau and Wurzburg the procession moved
around the fire as the psalms were being chanted, nine circumambula-
tions being accomplished at the latter church; whilst at Breslau7 and
Ratisbon9 banners were borne in the procession. Both features seem
suspiciously to be survivals of pre-Christian ritual.
In a number of French diocesan rites a litany was sung during the
1 There was no procession in the Cistercian rite (~.Cist. p.104), or

in churches such as ~on Cathedral, where the new fire was kindled
at the altar.
2 OR 26.9; OR 29.17; OR 31.63. 3 ~II p.58 1220.
4 ~ 3.13.34 p.126 (M1145). 5 Albers V p.32.
6 It is just possible that the mention of this one psalm was in some
service-books an abbreviated rubric; and that the chanting of the
other six psalms was to be understood.
7 1519 Missal fol.lxxix. 8 1477 Ordinary np.
9 1570 Obsequiale np.
114
blessing of the fireo 1 Although all of the evidence is comparatively
2
late, the fact that this feature is attested over a wide area of France
would suggest that it belongs to Gallican practice during the Middle
Ages. At St Bertrand and Mende it was sung by six choristers who re=
4
mained in the choir.3 vfuilst the fire was being consecrated at Cahors
a senior cleric chanted the reading from the Second Book of Maccabees 5
which commemorated the discovery by NehemiQh of the sacred fireo
Church/ Date Source

Commentator
Cluny 10 c Albers II p.47

Far fa 11 c Albers I p.48
Fruttuaria* 11 c Albers II p.93
Gembloux" 11 c Albers IV p.53
St Benigna,Dijon 11 c DAMR 3.13.34 p.126
- (M 1150)
John of Avranches .s,.1070 PL 147.49A
Lanfranc _£.1070 PL 150.446D
Magd.Coll.Pont. + 12 c HBS 39 p.169
Avranches 12 c DAER 4.24.3 p.145
@
---- (M 339)
Evesham .£.1250 HBS 6 col.80
Norwich"' c.1265 HBS 82 p.81
St Vedast,Arras+ c.1300 BBS 86 p.160
Durham - 14 c Missal p.185
Reims 14 c Qm. 4.22.5 p.97
(M 261)
Camaldolese 1503 Missal fol.89
Vallombrosa 1503 Missal fol.xci
Carmelites 1504 Missal fol.xcv
• Psalm 42 was also sung.

+ The point at which it was sung is not stated.
@ Psalm 24 was also sung.
Table 11. Psalm 50 sung in procession.
Church ~ Source
Salisbury c.1486 Missal fol.lxxxiii

Burgos 1546 Missal fol.ciii
Palencia 1568 Missal fol.c
Table 12. Psalm 26 sung in procession.
1 At Breslau a litany was sung at the end of the blessing (op.cit.).

2 Paris (1666 Missal p.2}8); Sees (1742 Missal p.185); Carcassonne
(1749 Missal p.194); St Bertrand (1773 Missal p.209); Mende (1766
Missal p.199); Lulon (1828 Missal p.213); La Rochelle (1835 Missal
p.186); Autun (1845 Missal p.239).
3 1773 Missal p.209 and 1766 Missal p.199 respectively.
4 1760 Missal p.172. 5 2 Mace 1:18ff.
115
Church Date Source
Meiss en £o1500 Breviary np

Halberstadt .£·1505 Missal fol.lxx
t1ainz 1507 Missal fol. xcii
Salzburg 1507 Missal fol.xciii
Spires 1512 Agendi fol.xciii
Abo 1522 Manua p. 130
Table 13. Seven penitential paalms sung in processio~ to the new fire.
Church Source
Corbie 10 c PL 78.336D
Cologne 12 c ~ 4.24.3 p.145
(MS Gg 15)
Strasbourg 1364 ~ 4.24 p.162(M35)
Wurzburg 1477 Ordinary np
Freising 1487 Missal fol.ciii
Trier c.1487 Missal fol.cii
Prague - 1498 Missal fol.xci
Passau 1503 Missal fol.lxxxiv
Breslau 1519 Missal fol.lxxix
Rat is bon 1570 Ritual np
Table 14. Seven penitential psalms sung at the new fire.
( ii) The formulas "'
1
The forty or so surviving prayers or benediction-formulas for the
2
blessing of the newly-kindled fire belong to two main categories :
those which were specifically composed as benediction-formulas for the
hallowing of the new fire, and those, already in existence within re-
lated liturgical situations, which were adapted or reapplied to the cir-
cumstances appertaining to the new fire ceremony. The prayers belonging
to the former group were composed in response to the adoption by the
Church of the pagan practice of kindling new fire. Others, inherited
from well-established ceremonies, such as the Lucernarium, contain no
1 These all appear in Appendix 5.

2 Within the tradition in which the Easter candle was lit with fire
that had been reserved and not newly kindled (Mode B1, p.101), this
act of benediction was unnecessary since the fire was presumably
taken from an already-consecrated flame. In this situation it is
not clear whether a perpetual supply of consecrated fire was main-
tained from year to year, or whether fire was kindled as and when
it was required, and consecrated on each occasion with a minimum
of ceremony by a simple act of benediction •
• See References on p.130.
116
overt references to the production of the nev firop but their refer=
ences to light and illumination rendered them suitable for adoption
as blessings of the new fire.
(a) The use of one formula.
It subsequently became common practice to hallow the new fire uaing

two or more prayers. Earlier practice, however, was to pronounce only
one blessing, a feature which survived in a number of non=Roman rites
(Table 15), and since 1955 has been part of the Roman rite itself. The
use of a single prayer is found in the Mozarabic rite, the older Ambros-
ian rite, the Beneventan rite, and in the earlier English rite. Like=
wise one formula is attested in the eighth-century Sacramentary of
1 2
Prague and in ~·
(b) The use of two or more formulas.
The earliest evidence for the existence of more than one prayer for
the blessing of the new fire is to be found in ~· Although this pont-
ifical contains three benediction-formulas, it is almost certain that
only one was uttered each time that the new fire was blessed. The three
prayers of~ (viz. A,B1,C) are separated from each other by the rubri-
cal word !!!!, which indicates that a choice of formula existed for the
officiating priest. This choice is also found in the twelfth-century
Pontificale Romanum3 and in the Pontifical of the Roman Curia; 4 and in
the thirteenth-century Ritual of Evesham ~ appears between the two
prayers which that book recorda. In subsequent Roman documents and
other service-books which contain two or more benediction-formulas
(Tables 16-18), there is no indication that a choice existed. In some
rites this may suggest that all the prescribed prayers were said. How-
ever, in other rites, such as the Roman, in which the new fire could be
kindled with either a flint or a lens, presumably benediction-formulas
1 The prayer for the blessing of the fire in this sacramentary (8 96)
follows the Exultet (8 95). Its position here suggests that it was
a later insertion which made provision for the kindling of the fire.
~ is the only Gelasian sacramentary to record such a formula. Had
it been intended for the blessing of Good Friday's reserved fire
(8 94), it would surely have been placed in close proximity to the
relevant rubric.
2 For the difficulty presented by~' see below, p.126.
3 PR XII.xxxii p.238.
4 :PRC xliv p.470. Only Manuscripts C and E of Durand's Pontifical
(PGD) record alia (p.587).
117
1 Beneventum Co 1000 A 34 Poitiers 1524 B2a
2 Vallombrosa 11 c A 35 St Martin
3 Besanc;on 11 C A d 9 Ainay 'I ~on 1531 B2a
4 Rupert of Dautz c.1111 A 36 Milan 11 c B2b
5 Benedictin0s = 1481 A 37 Ripoll 1038 B2c
6 Uzes 1495 A 38 Braga 1512 B2e
7 Bursfeld 1498 A 39 Rouen 1497 B5a
8 Valence 1504 A 4o Coutances 1557 B5a
9 Soville 1507 A 41 Auch 1838 B5c
10 Narbonne 1528 A 42 1-ti.lan 1902 B6
11 Bayonne 1543 A 43 Leofric Coll. 11 c B7a
12 Besant;on 1766 A 44 St Epvre,Toul 14 c B7e.
13 Reims 1770 A 45 Rheinau 1114 B9
14 Meaux 1845 A 46 Milan 1981 B11
15 Rome 1955 A 47 Bee 11 c ?C
16 Rome 1970 A1 48 Lire c.1400 c
17 Mozarabic 10 c B 49 Tongres 15 c ?C
18 Wulfstan 11 c B1 50 Dominicans 1482 D
19 Palermo _£.1130 B1 51 Vich 1038 E
20 Lateran Missal 13 c B1 52 St Malo 1503 F
21 Tours 13 c B1 53 Sens 1520 F
22 St Denys _£.1273 B1 54 Troyes 1736 F
23 Cistercians 1487 B1 55 Beneventum c.1100 H
24 Hildesheim* 1499 B1 56 Sacr.Vetus 11 c J
2.5 Rennes 1523 B1 57 Cluny 11 c L
26 Fontevrault 1534 B1 58 St Florian 12 c L
2? Amiens 1555 B1 59 Cahors 12 c M
28 Prague Sacr. 8 c B2a 6o Boulogne 1780 N
29 Egbert Pont. 10C B2a 61 Toulouse 1490 p
30 St Benigne,Dijon 11 c B2a 62 Burgos 1546 p
31 Barking• 1404 B2a 63 Strasbourg 1742 R
32 Angers 1489 B2a 64 Beauvais 1783 R
33 Cluny 1510 B2a 65 Cambrai 1507 T
Table 15. One prayer for the blessing of the new fire.
~such as A and S, would have been omitted, when the latter means of
kindling fire was used, in view of the explicit reference in those
prayers to silex.
The influence will be noticed of ~' either directly or through its

adoption by the Roman Church, in the use of the benediction-formulas
A, B, and C in churches where three or more prayers were said at the
blessing of the new fire. The adoption of Roman practice is well ill-
ustrated in the Cistercian rite in which the single prayer of the 148?
Missal was replaced in 1669 by the three formulas from the Roman Missal,
and in the rite of Braga where in 1558 those same three prayers took
the place of Formula B2e of the 1512 Missal.
Within the Ambrosian rite an interesting development occurred.

According to the eleventh-century Manuale Ambrosianum 1 a single prayer
1 ~II p.199. • Excluding the Veniat quaesumus.

118
is used for the blessing of the fire. Howeve~~~uscript M of the
manuale records a tradition in which a triple blessing is used; 1 and
these three prayers were incorporated into the Am}osian Misoal of 1475·
The number was reduced to two in the 1560 Missal; increased to three in
the 1594 and 1669 Missals; and reduced to a single formula in 1902.
It is possible to explain the presence in a rite of two or more

prayers for tha blessing of the fire in a number of ways. (i) We have
already noted that the benodiction-formula for the new firo produced by
friction could not necessarily be pronounced over fire kindled by means
of a lens. An alternative prayer would be required. (ii) With the in~
corporation of the new fire ceremony into the liturgy of the Vigil it
was perhaps inevitable that the prayer with which the lamp was blessed
at the commencement of the Lucernarium should be reapplied to the new
fire, having been displaced as a benediction of the light by the Exultet
2
and Preface. (iii) Similarly, as a result of the universal adoption of
the Exultet/Preface formula for the blessing of the Easter candle and
the consequent displacement of the older blessing, Deus mundi Conditor,
this latter prayer and in particular its concluding pericope, which, as
the Veniat quaesumus, became a prayer in its own right, was reassigned
to a different function and, in the isolated instances where it survived,
became associated with the consecration of the new fire, and not the
Easter candle as formerly. (iv) The acquisition of additional formulas
for the blessing of the new fire may in some instances have been the re-
sult of borrowing from different liturgical traditions. (v) At St Aug-
4
ustine's abbey, Canterbury3 in the thirteenth century and at 'Reims
during the following century it was the practice to use a different
prayer for the blessing of the fire on each day of the Triduum. It is
possible that this arrangement obtained in other churches whose
service-books prescribed three prayers.
(c) A survey of the formulas and their contents.
(i) Benediction-formulas A, S, P, J. These four prayers appertain

specifically to the act of kindling fire. The first two relate to a
situation in which the fire is produced by means of a flint; and in J
we have a blessing of the fire kindled by the refraction of the sun's
rays through a lens. The reference to lapis in Formula P would suggest
fire by friction in view of the use of the verb prosilire to describe
1 MVLA II p.198. 2 See below, Part IV Chapter 10.

3 Missal pp.36,37,39. 4 ~ 4.22.5 p.97 (M 261).
119
1 Avellana"' 11 c A 1l1 16 Nox-bartines 1578 B1 c
2 Nantes 1503 A B1 17 Cologne 1514 B2a E
3 Mindel! 1513 A B1 18 Herefox-d 1502 B2d Gr
4 Cahors 1760 A B1 19 Fulda 10 c B2e K
5 Poitiers 1767 A B1 20 Cologne 1626 B3 E1
6 Perigueux 1782 A B1 21 Milan 1560 B6 0
7 Le Puy 1783 A B1 22 Tournay 1540 B10 E
8 York 12 c A c 23 Nidaros 13 c c B1
9 Camaldolese 1503 A c 24 Carmelites c.1312 c B1
,..
10 Passau 1503 ?C A 25 Bayeux = 1780 F \,1
11 Lis:i.eux 1752 A c 26 Trier 1488 H B1

12 Chartres 1782 A c 27 Arras 15o8 H B2e
13 Ireland Co1200 A D 28 Saragossa 1552 H A
14 Lyon 1510 A D 29 Cordoba 1561 p I
15 Uzes 1495 A I 30 Carmelites 1664 s B1
Table 16. Two prayers for the blessing of the new fire.
the manner in which the fire appeared. This, the sole prayer for the
blessing of the fire contained in the Missal of Burgos, is inapposite
for the kindling of fire by refraction; yet the Missal gives this means
as an alternative to fire by friction. However, the difficulty largely
disappears if lapis can also be interpreted 'gem' or 'precious stone'.
The word would then refer to a beryl or some other translucent stone
with which the fire was produced by refraction.
(ii) B-category benediction formulas (B- B10). Formulas B1 to B10

are all variants of a common original; and it is claimed 1 that their
Western archetype is the prayer for the first blessing of the lamp in
the Mozarabic rite, Formula B. All share a similar structure, and have
common themes and close linguistic affinities. God is addressed as both
the 'unfailing light' (~ indeficiens) and the creator and source of
all light. The petition for the blessing of the light is followed by a
request for a share in that light and for inward illumination. The re-
ference in the Spanish prayer (B) to 'the light •••• which we bear in our
hands' and the emphasis of the prayer on light as opposed to fire strong-
ly suggest that this benediction-formula, and by extension the variants
within this same group, was the prayer, or the development of the prayer,
formerly used in the daily (or weekly) office of the Lucernarium for the
blessing of the lamp; and there is no reason to disagree with Kenneth
Stevenson's view2 that in view of the similarity in structure and theme
with the blessing of the light in the Byzantine tradition this prayer
and its variant forms in other Western rites derives from a common
1 Bernal pp.1034ff.; Stevenson, The Ceremonies pp.181-2.

2 Stevenson, ibidem p.182. • Excluding the Veniat quaesumus.
120
-
1 Rome 12 c A B1 c
2 St=Germain=des=Pres 12 c A B1 c
3 Basel 1488 A B1 c
4 Augustinian Friars 1491 A B1 c
5 Melk 1495 A B1 c
6 Cassino 1507 A B1 c
7 Vallombrosa 1503 A B1 c
8 Bremen 1511 A B1 c
9 Braga 1558 A B1 c
10 Osma 1561 A B1 c
11 Rouen 1640 A B1 c
12 Cistercians 1669 A B1 c
13 Paris 1666 A B1 c
14 Evreux 174o A B1 c
15 Paris~ Royal Chapel 1741 A B1 c
16 sees 1742 A B1 c
17 Carcassonne 1749 A B1 c
18 Paris 1762 A B1 c
19 Mende 1766 A B1 c
20 St Bertrand 1773 A B1 c
21 Vienna 1782 A B1 c
22 Tours 1784 A :81 c
23 Lu~on 1828 A B1 c
24 Metz 1829 A B1 c
25 Limoges 1830 A B1 c
26 Toulouse 1832 A B1 c
27 La Rochelle 1835 A B1 c
28 Auch 1836 A B1 c
29 Nantes 1837 A B1 c
30 Autun 1845 A B1 c
31 Freising 1487 A B1 D
32 ~on 1771 A B1 D
33 Durham 14 c A B4 D
34 Westminster .£-1370 B1 D F
35 Evesham £-1250 B1 D G
36 Cosenza 1557 B2a A c
37 Fruttuaria 11 c B2a c G
38 Reims 14 c B2a B?b F
39 Camaldolese 1503 B3 A c
40 Salisbury 13 c B4 D G
41 Langres 1492 B5a D G
42 Milan 11 c + 1475 B6 0 E
43 Milan 1768 B6 0 C1
44 Canterbury, St Aug. 13 c F w B5d
45 Leofric c.1000 H B2a Q
46 Spires 1512 H B2a A
47 Auch c.1000 L B2a
- D
Table 17. Three prayers for the blessing of the new fire.*
~erusalem original. However, the claim that Formula B of the Mozarabic
rite is the Western archetype of the rest of the B-category benediction-
formulas must be challenged, since it has yet to be shown that Spanish
influence was a major factor in the development of the other Western
1
rites. Elsewhere, we have argued that the Mozarabic liturgy of Holy
Saturday assumed ita final form as a result of Gallican influence. In
1 See Appendix 13. • See also Table 20 on p.125.

121
1 ESQ. 13 c A B'i c H"'
2 Esztergom 1501 H u A I c
3 l-1ainz 1507 H 12!:!9 B2a A
4 Hamburg 1509 A B1 c H D l35a
5 ·Lund 1514 A B5b D H
6 Breslau 1519 A v D c
7 Abo + c.1522 ~ H A B1 c v
(8 :Braga = 1558 A B1 c I )
9 Palencia 1568 D A B1 c
* Found only in MS Vat.Lat.1154.
+ Formula I follows blessiug of incense.
Table 18o Four or more prayers for the blessing of the fire.•
view of the widespread incidence of B-category prayers, and of their

presence in the Milanese rite, and in the early Gallican rite as attest-
ed in the Prague Sacramentary~ it is more likely that they all derive
from an archetype which was the prayer from the Lucernarium for the
blessing of the lamp, inherited from the Jerusalem rite; and that the
differences of language and theme were the result of their use in a num-
ber of different liturgical traditions over several centuries.
We noted above that the theme of light was prominent in the Mozar-
abic benediction-formula B. This accords well with the suggestion that
the prayer originates in the office of the Lucernarium. Moreover, it
is significant that the prayer is used, not for the hallowing of the
new fire as in other Western traditions, but for the blessing of the
light of the lamp lit with the new fire. Further support for the ~
ernarium origin of the variant B-category benediction-formulas is pro-

vided by the fact that the petition in Formulas B1,B2,B3,B8,B9, and B10
is also for the blessing of the light rather than of the new fire. The
blessing of the latter is sought in Formulas B4,B5, and B6; whilst For-
mulas B7a and B7b both refer to the 'light of this new fire'. In view
of the likely origin of these prayers they must rank amongst the earli-
est surviving benediction-formulas for the blessing of the new fire.
(iii) ~ Mosaic motif. Unlike the Mozarabic benediction-formula B

whose pre-eminent theme is the permeation of God's light throughout
creation, the remaining B-category prayers (with the exception of B5)
include a reference to the pillar of fire which preceded Moses and the
Israelites as they departed from Egypt, and in this way link the pro-
vision and the blessing of the new fire closely with the Passover, the
1 • See also Table 20 on p.125.
122
dominant theme of the Holy Saturday Vigil in the Romano~Gallican trad=
1
ition. The column of firo in the Book of hlxodus not only foreshadowed
the flame of the Baster candle at the Christian Passover9 it was s~en
as a fiery manifestation of the Lord's presoncc 9 as wao His appearance

to Moses in the Burning Bush. In the same way, therefore, that a refer~
ence to God's use of this element was inserted into the above=mentioned
B=category prayers, thus forging a thematic link between tho new fire
ceremony and the Vigil itself 9 Hie appearance to l4oses in the Burning
Bush was also commemorated in a group of benediction=formulas F1 Y,
BB, and B9. For the Mosaic motif which ran through the Paschal vigil
included incidents from the life of the Old Testament lawgiver other
2
than his passing through the Red Sea. Formula B9 also makes reference
to Moses' mission to Pharaoh and to his ascent of Mt Sinai. Moreover,
it is significant that in the shorter Vigil-tradition3 there are a num-
ber of instances where two of the four prophecies were the Mosaic
4
readings, viz. Ex. 14:24-15:1 and Dt 31:22-30.
(iv) ~-concepts inherited~ paganism. (1) According to the

purificatory theory for the origin of fire-festivals, first propounded
by E.Westermarck and favoured by J.G.Frazer, 5 the new fire acted as a
cleansing agent and a disinfectant, and the purpose of its being kindled
was to destroy all harmful influences, such as witches and disease,
which threatened the survival of a community. This pagan belief is
echoed in Formula J. Closely linked to and almost certainly deriving
from this belief is the petition, expressed in Formulas O, T, and in
B-category prayers, for the eradication of sin within ourselves and for
spiritual purification. (2) The pre-Christian belief that the harness-
ing and control of fire gives power to the possessor is discussed in
Chapter 8. Formula C would appear to express this notion, and to warn
that power deriving from the possession of fire is important since
that element is also used by the Devil.
1 In the prayer following the second blessing of the lamp in the Moz-
arabic rite, there are references to the departure from Egypt and
to the pillar of fire. (Pinell, !&Benediccio p.116.)
2 See Part III Chapter 2 pp.196-200.
3 With four or five prophecies as opposed to twelve.
4 In ~' OR 28, and in the Missals of Salisbury, Mainz, Verdun, Passau,
Bressanone, Eichstadt,and Palencia amongst others. These two pro-
phecies were retained in the Roman rite when the number was reduced
from twelve to four in 1955; and the former was the only reading
to be made obligatory of the nine prescribed for the Vigil in the
revised Missal of 1970.
5 Golden Bough Vol.10 pp.}42ff.
123
(v) Tho ~ m~di Conditor. This pray~r waG originally a formula for
1
the blessing of the Easter candle, but as such tlas replaced univerBally
by the Exultet/Proface. It survived, however, as a prayer for the con=
secration of the new fire, the flame of the newly~hallowed candl® form~
ing a dominant themee In each of the instances in which it is recorded

as a benediction=formula for the new fire (Table 19), it is difficult
to know whether it had formerly served as a prayer for the blessing of
the ~star candle, or whether it was borrowod from another rite. In the
rite of Salzburg it is clearly derived from PRG. In all six instancos
listed below tho Exultet/Preface was used to bless the Easter candle.
Church/ Date Source

Document
~ .£·950 II p.95 8 343

Salzburg 1507 Missal fol. xciii
Mainz 1507 Missal fol.xcii
Aquileia 1519 Missal p.91
Abo £·1522 Manuale p.238
Ratisbon 1570 Ritual np
Table 19. The use of ~ as a new fire blessing.
(vi) The Veniat quaesumus. This prayer, which subsequently became

the formula for the blessing of the five grains of incense in a large
2
number of rites including the Roman, formed the final pericope of the
~ ~ Conditor, being a concluding petition for God's blessing on
the Easter candle which had been kindled shortly before. As such it is
found in the Gelasian Sacramentary3 and contains the phrase super ~
4
incensum, 'on this lighted candle'. A subsequent request in the prayer
invites God to 'intensify the splendour of this night' (~ nocturnum
splendorem intende). Detached from the~ and an independent prayer
for the blessing of the new fire, it is first encountered in ~~ and
then in the subsequent documents which are listed in Table 20. In the
former pontifical it follows the benediction-formulas A, B1, and C, and
1 See Part IV Chapter 10 'pp.289-90.

2 As such, it is found in PR XII (xxxii.5 p.239) and subsequent Roman
documents. The prayer is also found as a blessing of (i) the incense
in the thurible (p.272), (ii) the cereus minor (p.291), and (iii)
the Easter candle itself(p.290). -----
3 GeV p.69.
4 Incensum may be viewed in two ways. Either, it is the masculine form
of the perfect participle of the verb incendere, used as a noun to
mean 'a lighted candle' - compare the adjective cereus, which regul-
arly had the force of the noun 'candle' - or, it is the perfect part-
iciple used as an adjective with the noun cereus understood.
124
Church/ Date Formulas Source
:Uocument
~ £,·950 A B1 c v II Pe57 § 219

Avellana 11 c A B1 v PL 151.881A/B
St Germain- DAER 4.24 p.158
des-Pres 12 c A B1 c v 'C'M230)
Barking 1404 B2a v HBS 65 p.101
Verdun 1481 A v Missal fol.lxiii
Hildersh.eim 1499 B1 v Missal fol.xcvi
Breslau 1519 A v D c Missal fol.lxxix
Abo _£.1522 ~ H A l:J1 c v Manuale p.2~
·- . '
Liege 1540 B5e v Missal fol.lxxvii
Ra.tisbon 1570 12!1Q. H c v Ritual np
Table 20. The use of Veniat quaesumus as a new fire blessing.

in conjunction with other formulas. *(V "" Veniat 3!!.:)
forms a fourth prayer for the blessing of the new fire. The rubric
1
which immediately precedes it, oratio postquam incenditur, indicates
that the prayer was uttered after the small candle, used for bearing
the new fire, had been lit. It is now the small candle, and no longer
the Easter candle, over which the Veniat quaesumus is said; and the
emphasis of the blessing is on the fire and not the candle. This is
clear from the phrase in the prayer super ~ incensum, 'on this fire' ,
2
which has replaced super ~ incensum of the Gelasian Sacramentary.
The possibility of confusion arising over the use of incensum, which
had widely displaced the earlier term for incense, ~. was recognised
at least at Verdun. In the Veniat quaesumus of that church's 1481
Missal, ignem has replaced incenaum in the above-mentioned phrase.
(vii) General prayers ~ further observations. A number of prayers

are simple petitions for God's hallowing of the new fire. To this group
belong Formulas D, E, G, I, K, M, and 0. The first of these asks that
the new fire may benefit mankind. Formula L, which is attested at Cluny
and St Florian for the blessing of the new fire, is strictly a prayer
to be used at Candlemas. Formula Q, found as a blessing of the fire
1 Also found in the Pontifical of St-Germain-des-Pres (M230).

2 A later modification to the prayer centred around the above-mentioned
imperative intende, 'in~ensify', found in ~- In fB[, the Pontific-
al of St Germain-dee-Pres, and other documents this command has be-
come attende, 'pay heed to' 'observe' (~splendour~ this night).
When the prayer was subsequently changed into a blessing of the in-
cense, the verb in question underwent a further change and became
accende, 'kindle'. In 1955 when the Veniat quaesumus reverted to
its original function as a preliminary blessing of the Easter candle
- at Bourges it had continued to serve this purpose (see p. 291 ) -
the former request, intende, was restored to the text.
• Atchley (p.139) also cites the 1752 Missal of Lisieux and the 1778
Missal of Narbonne.
125
only in the Leofric Missal 9 is excerpted from the Preface for the bless~
ing of the Eaater candlo. At F~ttuaria the ~ noster preceded tho

blessing of the fir69 and at Seville Formula A was followed by the ~
qui divitiasv a prayer which was said in other rites prior to the read=
ing of the prophecies. At t·Jurzburg a short threefold litany was sung
1
V. ~ ignem ~ benedicere digneris; R. !! rogamus ~ ~· In
the romanised Sacramentary of Vich the use of Formula B1 for the bless~
ing of the light and Formula E for the new fire derived from Mozarabic
practice. Formula R~ attested at Beauvais and Strasbourg in the eight=
eenth century, is introduced in the respective rituals of those churches
as a blessing of the Easter candle on a day other than Holy Saturday.
It was prescribed for use in the event of the flame of the Easter candle
failing during the period it was supposed to burn continuously.
In some service-books, such as those of W~zburg and Aquileia, the

benediction=formulas are mentioned but not given. In the Missal of the
2
latter church they are said to be contained 'in the Pontifical'.
There are a number of instances where only the opening words of the
prayers are given. However,only at Tongres is there some doubt about
how the first three words were meant to be continued.
(viii) The Pontificale Romano-Germanicum. We have already touched

upon the seemingly-double blessing of the new fire on page 103, and have
suggested that this twofold hallowing represents a synthesis of two sep-
arate liturgical traditions. The ~~ pronounced on Holy Saturday, was
strictly an offering of the lighted candle, a blessing of the berakah-
type rather than an invocatory benediction. The use on Holy Saturday
of Formula H3 which follows the ~presents a greater problem, since
the blessing of the fire is quite clearly stated : benedicimus ~
ignem. The formula, however, continues : 'and we sanctify it (the fire)

4
together with the wax and all its component elements •••• • The pre-
sence and use of this prayer presents less of a difficulty if, in view
of the second clause, we regard the formula as a blessing of the small
candle, soon to be lit, which supplements the oblatory expressions of
the .!!!Q..
1 1477 Ordinary np. 2 1519 Missal p.91.

3 Although this prayer is recorded only in Manuscript C, its survival
together with the ~at Salzburg suggests that its use was not
confined to a minor tradition.
4 '···~~~£!£!~omnibus~ alimoniis sanctificamus ••• •
126
(iii) Aspersion ~ j.nce_I;1Sa;ti2!!
Although documentary evidence for the sprinkling with holy water

1
and the censing of the new fire exists only from the eleventh century 9
it io likely that this twofold ritual at the new fire ceremony goes
back much further in time. The earlier evidence for these two acts is
presented in Table 21. There is no reference to them in the twelfth~
century Pontif:icale Romanum 9 though their omission from this pontifical

should not necessarily be regarded as conclusive evidence for their ab~
sence from the Roman rite at that time. A number of later missals do
not contain rubrics relating to the two rituals; but one suspects that
in most of these instances they formed features of the rites, since the
evidence for the twofold act after 1500 is plentiful. Attestation of
their occurrence at Milan dates only from 1560.~
Churcb/ Date Source

Commentator
St.Benigne,Dijon 11 c ~ 3-13-34 p.126

John of Avranches £-1070 PL 147.49A
Evesham o·.1250 HBS 6 col.81
Norwich £.1265 HBS 82 p.81
Salisbury 13 c Missal pp.265-6
Rome 13 c ~ p.588 8 7
Durham 14 c Missal p.186
Reims 14 c ~ 4.22.5 p.97
(M 261)
Rome 1474 Missal p. 175
Table 21. Aspersion and incensation of the new fire.
2
According to Lanfranc's Decrees the new fire is only aspersed; and
it is significant that an act of incensation is prescribed in that same
work during the blessing of the Baster candle. A similar arrangement
is also found in the older Cistercian rite3 and in the fourteenth-century
4
Westminster Missal. It is unlikely, however, that there is any direct
connection between the omission of incensation at the new fire ceremony
and the censing of the Easter candle during the singing of the Preface,
since at Durham, Evesham, and Norwich the new fire and the Easter candle
were honoured with incense.
The Missale Romanum of 1474 does not state the number of times the
1 Customary of St Benigna's, Dijon (~ 3.13.34 p.126, M 1150).

2 PL 150.467A. 3 Guignard pp.116-17.
4 HBS 5 co1.576. + ~ 4.24 p.169.
127
1
new fire was to be aspersed and censedo Later Roma~ and other
2
diocosa~ missals~ as well as liturgical manualo~ 3 specify three timeso
At Lesnes the fire was aspersed and cenaed aftor it had beon brought
4
into church. Throughout the history of the now fire ceremony the order
has been invariably aspersion followed by incensation.
(iv) The officiant
The preparation and kindling of the new fire were preliminary and
functional duties comparable \'lith the lighting of the lamps and candles
before the start of Tenebrae, and could therefore be performed by a lay
person; and though we shall shortly consider instances in which it was
either obligatory or considered desirable that a person possessing sac-
erdotal authority should kindle the new fire, it seems very likely that
in many churches the production of fire was included amongst the duties
of the sacristan. The greater importance attached to the blessing of
the fire, as opposed to the kindling of the fire, may well account for
the fact the official responsible for the latter act is rarely mentioned
in our sources. At Nidaros, Spires, Auch, Besanlon, and Le Puy5 it was
the responsibility of the sacristan, and of the sacristan's assistant
at St B~nigne's, Dijon.
6
Amongst the Cistercians a servitor lit the
fire. 7 At Milan Cathedral in the eleventh century it was the duty of
the cicendelarius, the official responsible for church illuminations,
8
to prepare the new fire.
The different origins and subsequent development of the Mozarabic

rite and the unusual circumstances in which the fire was kindled made
it obligatory that -a priest should perform this ritual act. Indeed in
the cathedral churches it was the bishop himself who kindled the fire.9
The importance of the beryl in the production of fire at St Benigna's
Dijon perhaps made it inevitable that one of the monastery's dignitar-
10
ies should hold the lens. At Soissons in the late twelfth century
1 e.g. ~ 1570. 2 1762 Missal of Paris p.237.

3 e.g. Colti, Dictionarium IIp56. 4 HBS 95 p.4?.
5 ~ p.32; 1512 Agenda fol.xciii; 1836 Missal p.191; 1707 Ceremonial
p.315; and 1836 Ceremonial p.373, respectively.
6 This person officiated only if the new fire was kindled from a
stone (~ 3.13.34 p.126, M 1150).
7 1689 Ritual p.245. 8 MVLA II p.198.
9 Antiphonary of Lden p.280 and Ferot~ 86 for the older Mozarabic
rite; PL 85.437-A for the Hissale Mixtum.
10 ~ 3.13.34 p.126 (M 1150).
128
1
the task of providing tho new fire devolved upon a deacon. This may
have been the result of a dosire to make the kiudling of tho new firo
parallel and complementary to the archdeacon°s blessing of the Easter
At Li~ge the fire was lit by the Treasurer~
2
candle. and at Palencia
by the priest on duty for the week.3
\ii thin the Roman Church thri) official Caeremoniale Episcoporum of

1600 recommended that the new fire should be kindled by a bishop where-
ver possible.
4 However~ Roman service-books from EEQ to the Missal of
1970 have always expressed the action of kindling the fire in the pass-
ive voice of the verb 5 and have never specified an agent. Thus the per-
formance of this duty by a sacristan is not precluded. Both before and
after the Second Vatican Council the custom has been for a priest to
kindle the new fire. French missals of the eighteenth and nineteenth
centuries are as non-commital as the Roman in this respect; but it
seems likely that in those cathedrals in which the new fire is still
kindled by the bishop or by a priest the ancient custom is perpetuated.
In twenty-seven of thirty-eight cathedrals for which information is
available, 6 a bishop or a priest strikes the new fire; and in four
cathedrals in which this is performed by the sacristan, the kindling
may have been one of that official's traditional functions. In some
places, following the reforms of 1955 which were intended to increase
lay participation in the liturgy, the task of kindling the fire has been
given to lay persons. 6 This has happened at the Cathedrals of Agen,
Bayeux, Belley, Digne, and Vannes. At Reims it is always a young lay
person; and in the Cathedral of Troyes the duty of kindling the new
fire is performed by a nun.
Whilst the kindling of the new fire could be performed by cleric or

lay person alike, its commissioning for use in Christian worship requir-
ed a sacerdotal blessing. Table 22 lists instances where the new fire
is known to have been blessed by the bishop or the abbot in a cathedral
or monastic church (overleaf). At Rome, where in the later Middle Ages
the Pope was not always present at the ceremoDy, two other dignitaries
presided: a minor cardinal priest 7 or a minor cardinal deacon~ At
Salisbury the duty was performed by a priest, known as the 'executor•; 9
Whilst at Naples the blessing was pronounced by the cimiliarcha, the
1 ~ 4.24 p.161, M 305. 2 1492 Ordinary np.

3 1568 Missal fol.c. 4 1600 typical edition p.296.
5 See Table 38. 6 1984 Survey of France.
7 PR XII xxxii.1 p.2}8. 8 Ordo Albini p. 130.
9 Missal of c.1486 fol.lxxxiii.
129
1
cathedral treasurer. In the Milanese rite tho officiating priest
2
was formerly a cardinal7 but in more recent timos this official is
not specifiedo 3 Mart~no informs us that at Arlee4 and Soissons 5 the
duty was formerly carried out by a deacon.
By Abbot By Bishop
Regularis Concordia ~ (II PG95 s 342)

(PL 13?o491B)
Corbe (PL 78o336B)
Cistercians (Nom.Cist.p.104)
Toledo (Ferotin § 86)
Lesnes (HBS 95 p.4?)
Rouen (PL 14?.176A)
Evesham (HBS 6 col.80)
Nidrosa (~ p.232)
St Vedast, Arras
(HBS 86 p.160) Rome (~ IV.2 p.587)*
Laon (Rite p.813)
Cluny (1510 Missal fol.xlix)+
Vienne (De Mol{on p.23) 6
Beauvais (1756 Missal)?
• 'or another priest•. + 'or the celebrant'.
Table 22. The blessing of the new fire.
References
For the location of the benediction-formulas referred to in
Tables 15-18, the reader is requested to consult the
corresponding entry in the Key to the Bibliography on
pages 373-75.
1 Constitutions of J.Orsini (Mallardo p.33).

2 MS M in ~ II p.98; and 1475 Missal fol.lxx.x.
3 1902 Missal p.34 (Rep.); 1986 Missal (t.e.) p.242.
4 ~ 4.24.3 p.145 M 31/32. 5 ~ 4.24 p.161, M 305.
6 According to the 1519 Missal of Vienne the archbishop blessed the
new fire. However,the officiant is not mentioned in the 1782 Missal.
7 Cited by Jounel, ~ Semaine Sainte p.147.
130
Chaptl::lr Four
THE PRODUCTION OF NEl·J FIRE. ( 1 ) TIME
(i) Within lh! tradition of ~ threefold kindling
(a) Maundy Thursday
According to Ordo 26~

which is the earliest document to record a
time for this ceremony, the new fire was kindled at the ninth hour of
1
the day. This time became traditional for most churches which observed
the rite on Maundy Thursday. It is found in other early ordines 2 and
in the Ordo of Corbie, in the Regularis Concordia and in Lanfranc's
Decrees, and in several of the eleventh-century monastic customaries.
The Ritual of Evesham also records the same time. Since the Office of
None was sung at this hour, the phrase post nonam should not be inter-
preted in a strictly temporal sense with the meaning of 'at the end of
the ninth hour' that is, at the start of the tenth; it indicates simply
that the ceremony should take place at the conclusion of that office.
At St Mary's, York the fire was kindled during the singing of None. When
in the later Middle Ages the services of the Triduum came to be anticip-
ated, the new fire ceremony and the Office of None continued to take
place in conjunction with one another. However, by the sixteenth cent-
ury, when the new fire came to be kindled during the morning of Holy
Saturday, the link between the two ceremonies was generally no longer
maintained. At Bayonne i~that century the new fire was struck at
9.00 a.m. The rubric in the missal of that church strongly suggests
that E2!! ~ was now understood to refer to the modern system of
reckoning time .
In the period before the ceremony began to be held by anticipation

on Holy Saturday morning, two documents record exceptions to the time
of the ninth hour viz. Poitiers and ~.3 We propose to discuss Poitiers
presently. The latter pontifical, together with Alcuin, is unique in
1 OR 26.3. On 18 March, the earliest day on which Maundy Thursday can

fall, the ninth hour would have lasted from approximately 2.10 p.m.
to 3.10 p.m. On 22 April, the latest date for Maundy Thursday, the
same period would last from approximately 2.22 p.m. to 3.33 p.m. by
modern reckoning. The figures relate to the fiftieth parallel.
2 For references for these and subsequent documents, see Tables ~ &24.
3 Strictly there are three. Alcuin (PL 101.1205C) derives from~
and contains identical information to that found in the pontifical.
It cannot be regarded here as an independent source.
131
Document ~~~~
Hour of New Fi~e Source
Ic'IT GF HS
Ordo 26 (8 c ) 9 OR 26.3
Ordo 28 (_£.800) 9 9 OR 28.25 & .58
Ordo 29 (9 c ) 8 7 OR 29.29 & .45
Ordo 31 (9 c ) 9 8 OR 31.29 & .62
Poitiers 6 6 9 Pont. p. 138
.f.RG (c. 950) 9/5 5 7 !!.215,.304~.342
Reeaulario 22.B.,-
cordia (,£.970) 9 9 9 PL 137.491B/494C
Ordo of Corbie 9 DAMR 3.13.34
(10 c ) p:-126 (M 1145)
Table 2,2. The earlier evidence of times within the trad~

ition of the threefold production of new fire.
ChurcW' ~ 21 !:!!:! Ei!:! Source

Document
MT GF HS
Cluny (11 c ) 9 9 9 PL 149.658D,661C,

and 663A
Farfa (11 c ) 9 9 9 PL 150.1198D,1201C,
and 1203C
Avellana (11 C ) 9 9 PL 151.880D &.883A
St Benigne,Dijon 9 9 9 DAMR 3.13.34
( 11 c ) p.126 (M1150)
Gembloux (11 C ) 9 9 Albers II pp.96,99
German Monast
eries (11 C ) 9 Albers V p.32
Lanfranc (.£.1070) (9) 1 (9) 1 9 PL 150.466D
St Germain-des- ~ 4.23-4 p.135
Pres (12 c ) 5 5 7 and p.158 (M 230)
Evesham (.£.1250) 9 9 9 HBS 6 col.88
Norwich (.£.1265) 9 HBS 82 p.91
Canterbury (13 C ) 2
HBS 28 p.274
St Mary,York(c.1400) DN DH DN HBS 75 p.275
Salzburg ( 1507) 9/5 9 Missal fol.lxxxv
Cluny (1510) AN AN AN Missal fol.lxix
Braga (1512) AN AN AN Missal np
Bayonne ( 1543) 9.00 a.m. 9.00 a.m.9.00 a.m. Missal p.41
AN = After None DN = During None

1 Implied in PL 150.4678. 2 Almost certainly 9th hour on each day.
Table 24. The later evidence of times within the tradition
of the threefold production of new fire.
132
recognising that tho amount of daylight on Maundy Thursday depends on
when that day falls, and makes allowance for tho variation in tho
lengths of the days.
At the ninth hour when the days are longer or at the fifth hour
when they are shorter ••• 1
Any interpretation of this information seems fraught with difficulties.

On first examination the rubric appear6 to recogniao that on tha longer
days 9 that ie 9 from 5 to 22 April the now firo wao to bo otruck at the
ninth hour (between approximately 2.22 p.m. and 3.33 p.m.); and that
on the shorter days, between 18 March and 4 April, the ceremony was to
take place at some time between approximately 11.08 a.m. and 12.o8 p.m.,
which period is by former reckoning the fifth hour. Unfortunately the
pontifical does not define shorter and longer days so that the division
we have suggested, though a reasonable one, must remain tentative. On
the other hand it could be argued that the fifth hour and the ninth hour
respectively represent the times for the new fire on the earliest and
latest dates when Maundy Thursday can fall. This presupposes that an
adjustable scale of times was in operation for the kindling of the new
fire to accommodate the varying dates on which Maundy Thursday fell.
Thus, for example, on 5 April, which lies midway between the two extre~
mes, the new fire would be struck at the seventh hour, or by modern re-
ckoning between approximately 12.08 p.m. and 1.14 p.m. The medieval
churchmen of Mainz may not have used such a scale that featured the ex~
actitude of modern chronometry; but that they adjusted the time for
the ceremony between the two extremes of the fifth and the ninth hours
is not an unreasonable assumption. To the modern mind, however, a diff~
erential of four hours between the two extremes of times at which the
new fire ceremony took place seems excessive, since the length of day-
light on 18 March at the one extreme is approximately twelve hours by
modern reckoning, but on 22 April a little over fourteen hours. The
view cannot be sustained that the variation in time was prescribed so
as to facilitate the kindling of fire by the refraction of the sun's
rays. A lens may be used for this purpose both at an earlier and at a
later time of day than the two times prescribed by the pontifical.
Two suggestions may be advanced to explain the occurrence of this

variation in times, not attested independently elsewhere. 2 (a) It may
be argued that, since~ is a composite document, amongst the diverse
1 ~II p.57 8 215 = Alcuin (PL 101.1205C). 2 See p.131 n.3.
133
elements in its composition is a,choice of times fo~ the new fire
~hich come from diffe~i~ liturgical traditions; 1
and that in order to
accommodate o~ justify the retention of both times~ the fifth hour was
assigned to those days when Maundy Thursday fell before 5 April and the
ninth hour to those that followed that date. For in spite of the strong
evidence of the earlier ordines for the kindling of the now fire at the
ninth hour, the evidence of Poitiera, which prescribes the sixth hour
on ~~undy Thursday, shows that an earli8r time was not uuknown at this
periodo r~rcove~ it io significant that a twelfth~century pontifical
from St-Germain~des=Prea records the time of the fifth hour fo~ the new
fire. Howeve~ we cannot be sure whether this represents a survival of
the earlier and alternative tradition we suggest may have existed, or
2
whether that church in question originally observed the times enjoined
by ~' and subsequently opted for the fifth hour as their normal time
for the ceremony regardless of the date on which Maundy Thursday fell.
(b) The striking of the new fire at the earlier time may have been
the result of a concern about the weather. The compilers of the pontif-
ical were obviously not unaware of the possibility of having inclement
weather at the time that the new fire was kindled. It could be argued
that in mid to late r4arch the temperature is higher at the fifth hour
than at the ninth hour, or at least that the weather is more likely to
appear favourable in the late morning to a congregation assembled in the
open air for a ceremony which ideally should take place out of doors.
Although John of Avranchea and Rupert of Deutz both refer to the

blessing of the new fire on Maundy Thursday, 3 neither writer specifies
the time at which the ceremony was to take place. The omission of this
information may possibly indicate that in the rites of the two churches
with which these writers were familiar the times for this ceremony were
flexible. For it seems most unlikely that they both inadvertently
omitted to mention the time on this day, especially as both record the
times for the ceremony on Good Friday and on Holy Saturday.
1 Support for a time inherited from an alternative tradition might

come, it could be argued, from the fact that in the rubrics of PRG
for Good Friday the new fire is brought at the fifth hour, if it
could be proved that within this tradition Good Friday was the orig-
inal day on which the new fire was produced; and that, when the
ceremony was extended to the other two days of the Triduum, the need
for liturgical symmetry demanded the same hour on those days also.
The kindling of the uew fire at the seventh hour on Holy Saturday in
~' however, makes it difficult to theorise with confidence
concerning this alternative tradition.
2 Martimort suggested that the pontifical may have originated at Trier.
3 ~-~ Q!f.~ (PL 14?.49A) and~ ~.Q!!.V (PL 170.149A),respect-
/ively.
1~
(b) G-ood Friday
The evidence for the blessing of th® now fire on Good Friday shows
that a greater variety of times existed at which tho ceremony was per-
formed than on Maundy Thursday. This is perhaps not surprl.sJ.llg in
1
view of the other ceremonies held during the afternoon of Good Friday,
and the difficulty perhaps experienced by some churches in accommodat-
ing this ceremony at the usual time of the ninth hour. fBg and the
pontifical from St-Germain-des=Pr{s enJo~ the fifth hour of the day,
and Poitiere the sixth. The new fire ceremony, held at these times,
would have taken place before the Passion. According to Ordo 29 the
ritual took place at the eighth hour, possibly between the Veneration
and the Mass of the Presanctified. However, of the sources which re-
cord a time for Good Friday the majority prescribe the hour that was
observed in most churches on Maundy Thursday, viz. the ninth. It will
be noticed that the evidence for the ninth hour on Good Friday before
1500 is drawn entirely from the monastic tradition. Although a number
of monastic documents contain omissions of time, in view of the influe-
nce of the tenth-century Benedictine Regularis Concordia, which attests
all three times, the ninth hour may be inferred with confidence in all
such instances and on all three days.
(c) Holy Saturday

Some of the earlier sources severally record a variety of times.
Ordo 29, ~' and the pontifical from St~Germain=des-Pres prescribe the
seventh hour, Ordo 31 the eighth, and Ordo 28 and Poitiers the ninth
hour. This last-mentioned time is specified in the Regularis Concordia
and by Lanfranc the following century. John of Avranches also refers
2
to the same time. The kindling of the fire at StMary's, York during
the singing of None m2q suggest that None was no longer held at the
ninth hour and that the new fire ceremony took place earlier in the day.
(ii) Within~ tradition~ ~ single kindling
(a) The problem of interpreting the phrase post ~ ~ has been

alluded to above. The Office of None was formerly sung at the ninth
hour of the day; so that the new fire ceremony which followed that
office took place more or less in the middle of the afternoon. Subsequ-
ently when None came to be sung at midday and the new fire ceremony was
1 Viz. the Passion, the Solemn Prayers, the Veneration, and the Mass
of the Presanctified.
2 ~· ~ Q!f.~. (PL 147.52B).
135
Church/ Date Time Source
Order
~
(Lanfranc)"' _!.1070 After None PL 150.466D

Nidaros £·1210 II il
Q!1§ p.232
Salisbury 13 c II II
Missal p.265
Marseille 13 c fl
" ~ p.84
\·/estminst er * 1362-88 " II
HBS 5 col.574
Strasbourg 1364 II
DAER 4.24 p.162
" ~ (1'-135)
Durham* 14 c II
" 1\tl.ssal p. 185
Barking$ 14o4 HBS 65 p. 101
Cistercians$ 1487
"
II
"
II
Missal np
Basel 1488 II II
Missal fol.xci
Langres 1492 II II
Missal fol.lxv
Rouen 1497 II II
Missal np
Camaldolese• 1503 Before None Missal fol. 89
Vallombrosa• 1503 After None Missal fol.xci
Carmelites* 1504 II II
Missal fol.xcv
Seville 1507 II 11
Missal fol.lxxv
Braga 15.58 II II Missal fol.xcvi
Freising 1579 II II
Missal fol.85
Evrewc 1?4o II II
Missal p.186
Besanlon 1766 " 11
Missal p.206
Poi tiers 1767 II
" Missal p.244
Chartres 1782 II II
Missal p.171
Metz 1829 II
Missal p.158
Toulouse 1832 II
"
II Missal p.205
Auch 1836 II 11
Missal p.191
• monastic
Table 25. Time of kindling in relation to None.
also transferred to the earlier hour in a number of monastic and cath-

edral rites, the phrase ~ ~ lost its former temporal significance
in those instances. Moreove~at times None was sung together with the
other Little Hours before the principal Mass of the day; so that the
phrase~ nonam, 'after None', in the later Middle Ages and beyond
was capable of three interpretations, depending on when that office was
sung : (i) during the morning, (ii) at midday, and (iii) in monastic
churches, where the traditional times for the offices were still strict-
ly adhered to, during the middle of the afternoon. There is little
doubt that in most of the rites in Table 25 which date from the late
fifteenth century and beyond, where the time of the new fire ceremony
is given in relation to the singing of None, one of the two former
above-mentioned times obtained. The Missals of Vallombrosa and Poitiers
also add : hora competenti, 'at a convenient hour'. In the Ambrosian
rite the new fire was formerly kindled also after None~ According to
the revised Missal of 1902 the ritual took place during None (Rep. p.34).
1 1560 Missal fol.110; 1768 Missal p.110.

136
The interpretation of ~ ~ is further complicated by the fact
1
th&t in the fifteenth and sixteenth centuries two systems of calculat=
ing the time of day wero in use in different parto of the Western Church~
so that it is possible that in some of the instances listed in Table 25

the phrase should be interpreted 'after 9.00 a.m.' especially since in
places the new fire ceremony was performed during the morning of Holy
Saturday since at least from the close of the fiftoenth century.
Church Date Sourco
Verdun 1481 Missal fol.lxiii

Sens 1520 Missal fol.c
Osma 1561 Missal fol.lxxxix
Soissons* 1856 Ritual p.90
• before 9.00 a.m.
Table 26. New fire kindled at 9.00 a.m.
Church ~ Time Source
Esztergom 1501 sixth hour Missal fol.lxxiii

LUbeck 1505 .£.• sixth hour,
after None Missal fol.lxix
Mainz 1507 about midday Missal fol.xcii
Arras 15o8 midday Missal fol.lxiv
Table 2'Z· New fire kindled at midday.
Table 26 lists four instances about which we can be confident that

the new fire ceremony took place at 9.00 a.m. At Verdun and Osma the
modern method of telling the time is used; whereas at Sens the time is
2
expressed according to the older system. Churches where noon was spec-
ified as the time for the new fire are given in Table 27. These follow-
ed earlier Roman practice. A glance at Table 28 on the following page
shows that in a number of places no specific time was prescribed for
the new fire ceremony. This flexibility was also a feature of the later
Roman rite (MR 1570). These churches were therefore at liberty to begin
the liturgy at any suitable time during the morning of Holy Saturday.
At Cahors the new fire was kindled at the end of the Little Hours. 3
1 The old Roman system of dividing the hours of light and darkness
into twelve equal parts, and the modern method of reckoning time.
2 At Bayonne (Table 24) where the new fire was kindled at 9.00 a.m.,
the rubric of the Missal states : ~ tertia post ~ solis.
3 1760 Missal p.172.
137
Church Dato Time Sour co
Hildesheim 1499 conveniont hour l\iiasal fol.xcvi
Melk 1495 convenient hour Nissal fol.lxvii
Cassino 1507 after Sext Missal fol.91
Cordoba 1561 usual hour Missal fol.ciiii
Rat isbon 1570 convenient hour Missal np
Bayeux 1780 It II Holy Week p.493
Tours 1784 II II
Missal p.191
Meaux li II
Missal p.169
Table 28. New fire time unspecified.
An unusual feature of the Roman rite was the interval of time which
elapsed between the kindling and the blessing of the fire. There is no
evidence to suggest that this development occurred much before the
twelfth century. According to PRG the fire was both kindled and blessed
at the same ceremony. 1 Since the rubric of PR XII states that the new
fire should be kindled at the fifth or sixth hour, it is likely that
'the sixth hour', attested in the other twelfth-century Roman documents,*
also signifies 'midday'; but in later documents the new fire is kindled
after the Office of Sext, which may not necessarily have taken place at
noon, as previouly. lVhether the new fire was blessed at the ninth hour
during the twelfth century is not clear since the Ordo of the Lateran
Church and subsequent documents record that this ceremony took place
after the Office of None. However, since None on Holy Saturday in the
twelfth century was sung after the kindling of the fire, which had taken
place at midday, there is good reason to believe that the new fire was
blessed at some point during the middle of the afternoon in this period.
Moreove~ it is clear from PR XII that there was an interval of time
between the kindling and the blessing of the fire : 'Later at the hour
2
at which the Pope should enter the cathedral •••• '
Following the Roman reforms of 1955, which were intended to restore

some of the primitive character to the Paschal vigil, it became obligat-
ory for the liturgy of Holy Saturday to begin after sunset, 3 and the
acts of kindling and consecrating the new fire became complementary
aspects of the one ritual. Likewise in the revised Ambrosian rite the
4
ceremonies of Holy Saturday do not begin before nightfall.
1 II p.57 § 219.
2 Postea h2£! qua •••• pontifex intrare ~ ecclesiam.
3 Fortescue and O'Connell (11th edition) p.302.
4 Missale Ambrosianum (1981 t.e.) p.242.
For these and other Roman documents, see Table 29.
on the next page.
Document Date Time Source
Kindling Blessing
of fire of fire
PR XII 12 c 5th/6th 12ostea yp.238 xxxii.1
hour
~ c.1140 After ~p.60
None
OR XI c. 1140 6th hour PL 1041C
PRC
(:;;: OR X) 12 c 6th hour CMS r g p.470 xliv.1
None*) (PL 78. 1014B)
Ordo
Albini £o119Q 6th hour ~.~.II p.130
Ordo of 5th/6th ~ 4.24.3
Apamea 1214 hour p.145 (M 25)
~ _£.1295 6th hour After p.587 iv.1
After Sext None
Haymo's After After Van Dijk
~~· 1243 Sext None II p.245
Codex
A 1706 .£·1350 None* ZRKM p.213
Bindo F. 1377 Usual ~ p.276
hour
Missale 1474 After After HBS 17
Romanum Sext None Vol.1 p.174
Missale 1950 hora com-
--+
After t.e. p.186
Romanum petenti None
• The Pope says None while the new fire is being blessed.
+ 'at a convenient hour'.
Table 29. Times for the new fire at Rome.
139
Chapte~ Five
THE PRODUCTION OF NEW FIRE. (2) LOCATION
(i) ~ ear1y period
Our earliest sources for the new fire ceremony within the Romano=Gall=
ican tradition relate to the region of N.France and W.Germany, and stip~
ulate in the first instance that the fire should be struck from & stone
and that it should be performed out of doors. The external as opposed
to the internal location for the new fire almost certainly derived from
earlier pre-Christian practice, and was retained after the new fire
ritual had been adopted by the Church for liturgical, theological, and
1
cultural reasons. It need hardly be stated that an external location
was essential if the new fire was to be obtained by means of a lens.
2
All of our early sources, except Alcuin, record the additional
stipulation that the kindling of the new fire should take place in the
doorway of an oratorium, if one existed :
Si ibidem oratorium habuerint, super portam ibi excutiunt.3
Because of the variation in meaning of oratorium, this sentence is cap=

able of two different interpretations. In the early Middle Ages
oratorium could either describe a small chapel built over the tomb of
a martyr; or it could refer to a place of worship in a rural area
which facilitated the spiritual life of those living at a distance from
their parish church, being the forerunner of the later medieval chapel
of ease. Either interpretation is possible in the above-quoted rubric.
However, if oratorium is understood in the latter sense, the meaning of
ibidem becomes strained. For if oratorium indicates a chapel of prayer
in a small village or hamlet, it is difficult to see what ibidem is re-
ferring to. The only likely translation would be 'in the same area'
where 'area' refers to the diocese. However, it would follow from this
that the new fire rubrics related only to cathedral churches - they in
fact applied to monastic churches as well - and would imply that some
1 See p.185.
2 Alcuin records that a lamp is lit by the sacristan, and kept burning
until Holy Saturday (PL 101.1205C).
3 'If they have an oratorium in the same place, they strike (the fire)
there in the doorway.' OR 26.3; OR 28.25; OR 29.14; OR 31.29; ~
II p.57 8 217; Ordo of Corbie (~ 3.13.34 p.162, M 1145). In the
1764 edition of ~'~ ~ oratorium should read ~ibidem oratorium.
140
dioceses had no aratoria at all~ whilst others had only one. Much
more likely is the view that oratorium bearu th~ former moaning and ra=
fcrs to a shrine in close proximity to a cathedral or monastic church.
Ibidem would then have its usual maaning 9 'in the same place', and
would indicate the shrine's location within the vicinity of the church.
1
None of the six documents which refer to the oratorium offers any
explanation as to why the fire was kindled in the doorway of this chapel.
Since tho word doos not appear elsewh~re in connection with the new fire
ceremony, we can but make a few general observations and advance tenta=
tive theories to explain the use of the building on this occasion. If
an external location was an obligatory feature of this ceremony = the
rubrical phrase ~ ecclesiam implies that it was = it would perhaps
explain why the kindling took place in the doorway of the oratorium
rather than in the oratorium itself. In this period before AD 1000 our
sources attest that the old fire was extinguished on Maundy Thursday
2
and the new fire kindled the same day. In the older Roman tradition
fire was hidden on Maundy Thursday and continued to burn in !2£2 eccle-
siae secretiore3 until it was required for Baptism on Holy Saturday.
The reservation of fire was a feature of the Gallican rites also; but,
whereas three lamps were used in the Roman rite, according to Gallican
practice only one lamp was required. This is borne out by a number of
4
ordines which state,
Et de ipso igne continuo, in eadem ecclesia vel loco ubi accenditur,

lampada una servetur usque in sabbato sancto ad inluminandum cereum. 5
The phrase !a eadem ecclesia implies the existence and use of a locus
secretior; whilst ~ ~ accenditur may relate to the use of an
oratorium for the reservation of the new fire.
We believe that in both the Roman and Gallican traditions the con-
cealment and reservation of the fire
6 symbolised Jesus' seeming lack
of animation in the Tomb, and the remote part of the church building
in which the fire was reserved recalled the Sepulchre. In larger
churches the reservation of the fire would have occasioned little diff-
iculty if a crypt or sacristy could readily be utilised for that purpose.
1 See Note 3 on the previous page.

2 This is treated at length in Chapter 7.
3 'In a part of the church well removed from view.'
4 For instance, Ordo 26.5 and Ordo 28.27.
5 'Let one lamp be lit from the newly-kindled fire, and kept burning
in the same church or in the place where it was kindled until Holy
Saturday so that the Easter candle may be lit from it.'
6 The Gallican fire had been blessed; the Roman was in a state of limbo.
141
However,it is conceivable that problems may have arisen if the church
building contained no suitable side chapels or convenient niches for
concealing the lamp of reservation. It is our contontion 9 therefore,
that the desirability of having a conveniently=rernote place of safe=
keeping was one of the main reasons why the ordines recommended that an
oratorium~hould be used, if one was located near to the church!' For
such a building would provide protection both from the weather and from
tho gaze of hurnanityi and the sepulchral nature of tho building would
increase the significance of the reserved fire 9 and heighten the sym=
bolism associated therewith. As already mentioned, this alternative
place for the reserved fire is indicated by the phrase ~ ~ ~
accenditur; and since this is an alternative location to eadem ecclesia,
because of the necessity of protecting the flame for at least forty=
eight hours, it is almost certain that the writer has in mind the
oratorium in whose doorway the new fire was kindled.
Evidence from the sixth century shows that it had become customary
by then in parts of Gaul to keep a perpetual flame burning at the tomb
1
of a saint above which an oratorium had been built. Since our sources
state that the old fire was extinguished on Maun~ Thursday and rekind=
led the same day, as we have noted, it is our belief that an additional
reason for the reservation of fire in an oratorium was the possible
feeling of unease caused by leaving the saint's tomb unattended by a
light; for the loss of fire on Maundy Thursday was total. In later
centuries the light which burned before the Sacrament was also exting-
uished.2 The placing of the new fire on or near the saint's tomb,there-
fore, ensured that his presence continued to be honoured in this way.
Alternatively, it may have been believed that the placing of the fire
in an oratorium dimished the chances of its being extinguished, since
it enjoyed the protection of the saint. Then again, the placing of the
newly-kindled fire in the presence of such a holy person may have been
regarded as a means of enhancing in some way the essence or efficacy
of the fire. In the absence of contemporary corroborating evidence,
these suggestions must remain tentative. Nevertheless it is not com-
pletely inapposite to cite evidence from £.14oO in possible support of
these suggestions. It comes from St Mary's Abbey in York, and concerns
the disposition of the new fire at that monastery. After the kindling
of the new fire the rubric states,
1 Gregory of Tours, History 2! ~Franks 4.36 (P.L 71.291). The use

of light at a tomb is discussed more fully in Appendix 6.
2 See above, pp.83-84. In more recent times the sanctuary lamp has
remained lit at Tenebrae and at the new fire ceremony.
• For the use of the sacristy, see footnote at the bottom of p.145.
1~
accensas carbones de illo igne ponent super tumbas Abbatum in
capitulo dum cantatur hera Nonao1
The Ordinal does not comment upon or explain th® purpose of this prac-
tice; but it is clear from what we are told in this section that the
fire placed on these tombs was not reserved for the lighting of the
Easter candle on Holy Saturday; burning coals would only last for a
limited period of time. Two possible explanations for the use of the
coals in this way are : (i) The placing of the fire on the tombs meant
that the former abbots were the first to be honoured with the new fire.
(ii) The fire, which was about to be blessed for immediate liturgical
use, would by being so placed acquire some of the virtue of the former
abbots of the monastery.
As previously mentioned, Alcuin omits any reference to an oratorium,

though in other respects it closely follows the wording of the new fire
ceremony described in ~· It is generally thought that Alcuin was com-
piled considerably later than ~' although it incorporates many early
elements. It therefore seems reasonable to conclude that by the mid-
eleventh century the custom of utilising aratoria for the kindling and
reservation of the new fire had fallen into desuetude.
Immediately after the injunction in the ordines for the kindling of

the fire in the doorway of an oratorium, where an oratorium existed,
follows the directive for churches which did not possess this shrine
sin vero, in loco quo consideravit prior, ita ut ex eo possit

candela accendi.2
This concise and grammatically defective Latin sentence can be inter-

preted in a number of different ways, since it lacks a main verb and
the meaning of candela is not immediately clear. Since the verb
excutiunt occurs in the previous clause, one solution would be to supply
the same verb to complete the sense of this sentence. However, the dom-
inant idea of the fire's reservation, which this sentence contains,
makes the understanding of a verb denoting protection or concealment
equally probable. A third possibility would be to supply two main verbs,
one of kindling and one of reservation. Accordingly the sentence would
admit of three equally valid translations :
1 'Let them place lighted coals from that fire on the tombs of the
abbots in the chapter house, while None is being sung.' Ordinal
!!!! Customary 2! g Mary's, ~ (HBS 75 p.275).
2 For references, see Note 3 on p.140.
(i) But if there is no oratorium, they strike the fire, in a place
which the Prior1 has deemed suitable, in such a way that a
candela may be lit from it.
(ii) But if there is no oratorium, they reserve the fire, in a place
which the Prior has deemed suitable, in such a way that a
candela may be lit from it.
(iii) But if there is no oratorium, they strike the fire, in a placo
which the Prior has deemed suitable, and reserve it in such a
way that a candela may be lit from it.
It is clear that the next stage in interpreting this sentence corr=

ectly is to establish the meaning of candela; for the word may be
translated both 'lamp' and 'candle', and within the present context
may refer either to the lamp of reservation, or to the candle which was
carried into church on each of the three evenings of the Triduum, or to
the Easter candle. The first of the tentative translations is solely
concerned with the action of kindling the fire and with the manner in
which it is done, even though the manner is not explicitly stated. Here
candela would indicate the small candle or taper which was used to trans~
fer the new fire from the tinder to the lamp of reservation 2 and was
carried in procession into church on the evening of Maundy Thursday.
However, it is difficult to see what is the significance of ~~ 'in
such a way', and how the striking of the fire in this instance should
differ from that on any other occasion. Since the information it con~
veys appears to be gratuitously irrelevant, we conclude that the first

interpretation is incorrect.
The third interpretation is an expanded form of the second. The two

may be treated together since the dominant notion of reservation is
common to both. Moreover, though the latter version only states that
the Prior selected the location for the kindling, it may be safely inw
ferred that the decision regarding the location for the reserved fire
was also his. Of the three possible interpretations of candela, that
of 'lamp of reservation' may be eliminated, since it would have been in
that lamp that the fire burned from which the processional candle was
lit. It is also unlikely that the Easter candle is here intended; for
(a) it was kindled on Holy Saturday, not on Maundy Thursday, and (b)
the ordines use cereus to denote the Easter candle and not candela.
There remains the processional candle which was lit from the reserved
fire on each of the three evenings of the Triduum. This is surely the
most obvious interpretation and the one best suited to the context of
1 That is, the bishop or abbot.

2 ~fuilst the use of a wax or tallow candle should not be ruled out, an
oil-lamp with its more reliable flame seems more likely to have been
used for this purpose.
144
the reserved fire. That candela signifies candle and not reed is clear
from the fact it was placed at the end of a reed and carried in pro=
1
cession on high.
(ii) The later Middle ~
(a) The external location
The evidence for an external setting for the kindling of the new
fire beyond the period covered by the earlier ordines is set out in
Tables 30,~, and~' and like that of the ordines relates to churches
belonging to both the cathedral and the monastic traditions. The pract=
ice of kindling the fire in the open air was adopted by the Roman Church
in the eleventh or early twelfth century, probably as a result of
2
Cluniac influence. A number of mediev~l Roman documents do not specify
a location. The ~ Albini prescribes 3 'before the doors of the
Lateran'; and the open air location, enjoined by Durand's pontifical,
was a feature of the Pian Missal of 1570, and survives unchanged to this
4
day. The rubric of the Revised Order for Holy Week of 1955 specified
that the ceremony should take place at the door of the church; 5 and the
1970 Missale Romanum, with its emphasis on the participation of all the
faithful, enjoins that a large fire, visible to all, should be prepared
out of doors, with the proviso that in the event of inclement weather
or the difficulty of staging a fire outside the church, the ceremony
should be adapted accordingly. 6
As with other liturgical or ritual acts which had a utilitarian

origin, medieval writers explained the external location for the kind-
ling of new fire with reference to Scripture. Rupert of Deutz cites
Hebrews 13:11, itself an allusion to Leviticus 16:28, as the reason for
the production of fire out of doors. The allusion is to the Jewish
practice of burning 'outside the camp' the bodies of animals whose
blood had been shed to atone for the transgressions of men. Since,
therefore, the Jews led Jesus outside the city where his atoning
1 See Part III Chapter 1.

2 It seems likely that previously it had been kindled in church, a
theory we have argued in favour of in Section ( b ) on page 149.
3 Liber Censuum II p.13Q.
4 1lli, 1474 appears to be the exception. Its rubrics relate primarily
to the new fire at the Cathedral of St John Lateran. In any case
the striking of fire in the cloisters is a partial external kindling.
5 Schmidt I p.118. 6 1970 typical edition p.267.
(From p.142.) At Monte Cassino c.1100 the lamp of reservation was kept
burning in the sacristy. (~ 3715.5 p.141, M 1139).
sacrifice took place, our going outside the church to kindle the new
fire is dutifully obeying the injunction of the writer to the Hebrews
1
that we should 0
go forth to him outside the camp'. It need hardly be
said that this analogy is not apposite in every detail. Durandus is
content to record that the new fire is kindled outside in commemoration
of the Jewish place of crucifixion. 2 Dom Prosper Gueranger's comment
is both apt and relevanto He argues3 that the sparks of the n0w fire
symbolise tho Spirit of Jesus, and the stone from which they are struck
represents tho Sepulchre. Just as the Resurrection took pluce in a
4
tomb situated outside the city walls of Jerusalem, so the striking of
the fire should be performed outside the walls of the church building.
Churcb/ Dat,e Comment Source

Document
Farfa 11 c blessed in cloister PL 15Q.1199A

Gembloux 11 c or in cloister Albers II p.93
Rupert of
Deutz c.1111 PL 170.149A
E.Blli 12 c blessed !a ~ Ip. 2.38 xxxii. 1
Ireland c.1200 Missal p. 126
Sicardus c.1210 PL 213.323B
Apamea 1214 DAER 4.24 p.160
(M 25)
St Vedast,
Arras .s_.1300 on hospice step HBS 86 p.160
Strasbourg 1364 on the grass DAER 4.24 p.162
outside church <M"35)
Rome 1570 1950 Missal p.186
Laon 1662 blessed inside Rite p.813
Augustin.
Friars 1714 Ceremonial p.307
Ami ens 1752 Missal p.182
Cahors 1760 Missal p.173
Poi tiers 1767 Missal p.244
Auch 1836 Missal p.191
Table 30. An external location for the kindling of the new fire.
(Churches where a lena was used are not included.)
Table 31 lists the evidence attesting the kindling of the new fire
in the doorway or porch of the church. \~ether this practice owes its
origin to the procedure, found in the earlier ordines and in PRG, of
striking the fire in the doorway of an oratorium is difficult to say.
1 ~ ~.Qi!.V (PL 170.149C). This symbolism is also found in the

Mitrale of Sicardus (PL 213.323B).
2 Rationale VI.80 p.350. 3 Liturgical Year (PTHW) p.555.
4 The strong evidence in favour of an extramural location for the
tomb in which Jesus was laid has been forcefully summarised by
John Wenham in Easter Enigma pp.18-19
146
'fhe Regula:ds Concordia and John of Avranches' ~ ~ officiis eccl=
esiasticia were written close enough in time to th0 aforementioned
documents for there to be some direct link; but the instances from
French cathedral~ay possibly be the result of local development in
neo~Gallican times. The alternative location to the porch was used
presumably if the weather was fine.
Church/ Gom.ment Source

Document
Regularis
Concordia £• 970 PL 137.491B
John of
Avranches c .1070 PL 147.49A
Naples - 14 c Mallardo p.33
Tongres 15 c west door Ordinary p. 164
Uz'es 1495 p Missal fol.lxiii
Mainz 1507 in the atrium Missal fol.xcii
Rouen 1640 P/outside W.door Ritual p.305
Desideri 1739 P/before door Praxis p.143
Evreux 1740 P/outside at door Missal p.186
Maison
du Roy 1741 p Sem. Ste P·397
seez 1742 p Missal p.186
Carcassonne* 1749 p Missal p.194
Lisieux 1752 door of church Missal p.189
Mende* 1766 p Missal p.220
St Bertrand* 1773 p Missal p.209
Capuchins* 1775 P/outside at door Ceremonial p.124
Vienna• 1782 p Missal p.215
Chartres• p
, 1782 Missal p.171
Perigueux 1782 door of church Missal p.158
Bayeux 1790 at main door Missal p. 168
Poland 1819 P/threshold Manual II p.467
Coutances 1825 P/door of church Ceremonial p.329
Metz• 1829 p Missal p.158
Toulouse 1832 p Missal p.206
La Rochelle* 1835 p Missal p. 186
Le Puy 1836 P/door of church Ceremonial p.373
Nantes 1837 P/before door Missal p.198
Autun• 1845 p Missal p.239
• indicates that another location was allowed - see
other tables.
P = porch
Table 31. The kindling of fire near the church door.
With the exception of ~ 1474 the evidence for the kindling of

fire in the cloisters is entirely monastic (Table 32). For these
colonnaded walkways provided a convenient location for the ceremoQY in
that the striking of the fire could still be held to be performed out
of doors; and yet at the same time the measure of protection afforded
147
Churcb/ Comment Source
Document
Fleury 11 C at door of treasury Albers V p.143

Lanfranc c.1070 PL 150.466D
Gilbertines 12 C or in secreto ~ HBS 59 p.39
Bee c .1200 ~ p.127(M1153)
Haymo's
Q!:g.~. 1243 blessed before altar HBS 85 p.2o8
Worcester _£.1250 near refectory Antiphonary p.69
St=Denis .£o 1273
St-Epvre~
corner of cloisters ~·1 1158}~
3.13.
Toul 14 c M 1160 34
Lyre c.14oO M 1154 p.127
Rome 1474 blessed before altar HBS 17 p.174
Augustin.
Friars 1491 Missal np
Melk 1495 Missal fol.lxvii
Cassino 1507 in front of chap-
ter house Missal fol. 91
Cistercians 1689 Ritual p.245
Sens 1715 at door of treasury Missal p.238
Frejus 1745 De Rubeis p.327
Table 32. New fire kindled in the cloisters.
by the cloisters would largely overcome the difficulty experienced in

the lighting of a fire in the open air during particularly inclement
weather. As we have already observed, the injunction of~ 1lt74 must
relate specifically to the Cathedral of St John Lateran, since non-
monastic cathedrals and churches did not normally possess the luxury of
cloisters. Those at the Mother of Cathedrals were built sometime bet-
1
ween 1222 and 1230, a fact reflected in the~ Missalis of Haymo of
Faversham, whose service-books were modelled on the ceremonial of the
papal court. The Gilbertine rite also allowed the new fire to be
kindled in an unfrequented part of the church. This wish for a secluded
2
location is also found at Palencia.
Desideri's . recommendation3 that in country churches both the place

of kindling and the route leading to the altar should be strewn with
flowers is almost certainly a sanctioning of traditional practice. The
use of 'sweet-smelling flowers' either on the parvis outside the main
door, or in the entrance, or in the atrium was also prescribed in the
4
Capuchin Ceremonial of 1775.
1 Masson p.305. 2 1568 Missal fol.c.

3 Praxis p.141. 4 Memoriale p.124.
148
(b) The internal location
~vidence for the kindling of the new fire in the church itself
comes from both England and France. The practice is first attested by
1
John of Avranches in the eleventh century At Lyon, where the custom
has survived until the present day, the new fire is kindled behind the
altar. This location, which is peculiar to the rite of Lyon, was very
likely a feature of the earlier Roman rite; for it is generally agreed
that in a number of respects the ceremonial of Lyon preserves the prim=
itive form of the Roman rite. On the other hand in the absence of dir=
ect evidence for the practice at Rome, it can be argued that the kind=
ling of fire behind the altar was a local development confined to the
Church of Lyon. It may have been a survival of a pre-Christian ritual;
or there may have been some connection with the concealing of the last
candle at Tenebrae. Again, it is tempting to attribute its origin to
the Old Testament practice of maintaining a perpetual fire at the altar
2
of sacrifice. This may have come about in the wake of the Judaising
movement in the twelfth century which resulted in, amongst other things,
the introduction of the menorah into Christian churches. Possible sup=
port for this view might come from Narbonne, where according to former
practice the new fire was kindled ·~ ~ altaris'. Here it is not
difficult to see a close connection between the Jewish altar of sacrif=
ice with its four horns and the High Altar of the former Cathedral of
Narbonne whose corners are purposely termed ~·
It is significant that all three non-monastic English churches

whose service-books record the new fire ceremony enjoin the same loca-
tion for its kindling. The Exeter Ordinal stipulates 'near the south
column'; whilst the books of Salisbury and York agree that it should
take place between the two columns on the south side of the church, the
Missal of the latter church adding 'near the font'. Although it is gen-
erally agreed that the rite of Exeter was greatly influenced by that of
Sarum, the relationship between the rites of Salisbury and York in the
matter of the new fire is not at all clear. The ceremonies at both
churches may have derived from a common rite. Nor is it known why this
particular location was only to be found in England and only in the
secular or cathedral tradition. For according to the surviving evid-
ence for the monastic tradition in England the new fire was kindled in
the cloister. Since both ~eter and Salisbury Cathedrals possessed a
1 For this and other references below, see Table 33 on the next page.
2 Leviticus 6:12.
Church/ Comme~t Source
~
York* 12 c S.column,nr font Missal p. 109

Sarum 13 c between two columna HBS 91 p.20
Exeter 1337 on south side HBS 37 Po322
(ii) ~ ~~
John of
Avranches c.1070 (in the church) PL 147.49A
Cosenza - 1549 Missal fol. 115
Braga 1512 east end of nave Missal cited by
King ,Y§. p.223
(iii) M ~ step(s)
Cistercian 1119 only blessed ~.Cist. p.104
Bursfeld 12 c II II Mocquereau p.69
Dominican 1504 Missal fol.lxxxv
Premonstr~ 1578 Missal cited by
at ens ian King,J&Q. p.190
(iv) ~ the altar

Lyon 1392 behind altar Talaru Missal in
King, Jg§ p.59
Narbonne 1528 at horn of altar DAER 4.24.3
p.145 (M 203)
Saragossa 1552 at Epistle corner Missal fol.lxxii
(v) ~Easter candle

Scissons 1745 Missal p.163
* unless a lens was used for the new fire.
Table 33. An internal location for the kindling of the new fire
cloister, it is likely that this internal location was a feature of an

indigenous new fire tradition. Moreover the rationale underlying the
use of this place is unknown. We know of the close connection between
the light of the Easter candle and the blessing of the font at Baptism;
but a link between the kindling of the new fire and the font is
otherwise unknown.
~ the nineteenth century the new fire at St John Lateran was kind-
led in the sacristy, 1 the place where it had struck at Toledo, Llon,
and Milan since the early Middle Ages. The reason for the change of
location at Rome is unknown. It may have resulted from a desire to
have the ceremony held in a more convenient place. Milanese influence
1 Baggs p.98.
150
Church Qili Source
Leon 10 c Antiphonary p.280

Milan @ 12 c Beroldus p.109
St Mary's,York c.1400 HBS 75 p.292
Besan~on+ 1682 Ceremonial p.329
Angers 1731 Ceremonial p.258
Carcasson.ne>~> 1749 Missal Pe 194
Mende'l' 1766 Missal p.200
Reims 1770 Missal p.204
St Bertrand 0 1773 Misoal p.209
Vienna$ 1782 t4issal p.215
Chartres.;. 1782 Missal p.171
Metz* 1829 Missal p.158
Toulouse* 1832 Missal p.191
Verdun 1832 Ceremonial p.312
La Rochelle* 1835 Missal p.186
Meaux 1845 Missal p.169
Autun* 1845 Missal p.239
• see Table 31 for alternative locations.

+ fire blessed at Gospel corner of altar.
@ fire kindled in chapter house.
Table 34. New Fire kindled in the sacristy
may be discounted. The use of the sacristy in the eighteenth-century

and nineteenth-century instances listed in Table 34 was probably also
the result of ritual convenience caused by the decay of the ceremonies
of the Triduum during those centuries and the seemingly-futile process~
ion with the new fire in an empty church. However, in Spain and almost
certainly in Milan the practice of kindling the new fire in the sacristy
1
had antecedents in the Holy Saturday liturgy of Jerusalem. At Auxerre
in the sixteenth century the fire was kindled in one church and blessed
2
in another; and at Rouen by the eighteenth century the fire was kind-
led in the Church of St-Etienne, and then carried in procession to the
cathedral. 3
(c) The twentieth century
It had formerly been customary in many parts of France and Italy

for the lighting and blessing of the new fire to function as both a
religious and a civic ceremony if one of the doors of the church opened
onto the town square. The custom had largely fallen into desuetude
either because of the anticipation of the Paschal vigil during the
1 For the or1g1n of elements in these two rites, see Appendix 13.
2 Missal of 1537/8 (M 39) cited by Mart~ne, DAER 4.24.3 p.145.
3 De Mol eon p. 299 •
151
morning of Holy Saturday~ or in France because of its suppressio~ at
the time of the Revolution. It ~as not until after the liturgical re=
forms of 195~ which permitted the holding of the Vigil in the evening
of Holy Saturday, that the kindling of the new fire in the open air in
front of churches and cathedrals took place again on any widespread
scale; although at a small number of cathedrals, such as Lyon and Rodez,
the ceremony continued to be performed in the traditional place. In a
1
survey carried out in thirty~four French dioceses in1961 to ascertain
to ~hat extent the recently=revived ceremony had been popular and aucc=
essful, it was discovered that the main problem had been climatic. From
a number of replies it was found that on an inclement Holy Saturday
evening rain, wind, smoke, the difficulty in lighting the incense, and
the unexpected extinction of the flame of the Easter candle as a result
of a sudden gust of wind produced lack of concentration, and in places,
ridicule. Subsequently at Arras and at Vannes after 1978 the new fire
2
was kindled inside the cathedral. It is significant that the 1970
Roman Missal makes the alternative provision that if the weather is in-
temperate the blessing of the fire is adapted to circumstances. 3
Outside Square/Parvis Porch
Bayeux Carcassonne Pamiers

Coutances Angers Paris
Versailles Strasbourg Reims
Le Mans St Die Angers•
Belley Montpellier Autun
Arras1 St Flour St Brieuc
Vannes2 Le Mans*
Dijon Tulle
Mende
Bayeux•
Troyes
Outside
Main Door Cloister
Limoges Carcassonne• (W.End) Aix

Amiens3 Bayonne• (W.End) Bayonne
Nimes Agen (W.End) St Die*
orleans Orl€ans•
West Choir Transept Behind Altar
Nevers Nancy Lyon

Rodez
* If wet.
1 Now performed indoors. 2 Since 1978 performed indoors.
3 Since 1969 there has been no new fire ceremony in the cathedraL
Table 35. The new fire location in French cathedrals in 1984.
1 Morlot p.115. 2 1984 Survey.
3 Pope ~ Sundq Missal p.318. • Experimentation was permitted in 1951.
152
Chapter Six
THE PRODUCTION OF NE\1 FIRE. ( 3) MEANS
The production of new fire was achieved universally by one of the two
time-honoured methods derived from pre~Christian religious milieux :
the generation of fire by the friction of iron against stone or wood
against wood, and the concentration of tho sun's raya onto flammable
material by moano of a transluccn~ lens. Some churchoe proocribed tho
former means, some the latter, and a number permitted both. The means
favoured by one church waD probably determined by the method employed
in producing fire in pre-Christian times in the region in which that
church waD located. The choice offered by a number of service~books
or referred to by medieval writers indicates the validity of either

method in the eyes of the Church; and we shall discuss presently the
arguments put forward to justify the use of the flint and the lens, and
the symbolism associated with both the frictional and refractive methods.
(i) !!!! ~ friction
Evidence for the production of fire within a Christian liturgical

context by the friction of wood against wood relates to parts of Cent~
1
ral Germany. It is recorded that in Swabia the use of the fire-drill
was the only permissible means of kindling the new fire; and there is
no good reason for doubting that this was local practice. Unfortunate-
ly the 1555 Missal of Augsburg does not contain the new fire ceremony.
Elsewhere throughout the Western Church the frictional method of pro-
ducing fire involved the use of a stone.
The earlier Ordines Romani, our oldest sources for the new fire
ceremony, stipulate that the new fire should be kindled by the striking
of a stone. 2 There is no mention of an alternative method. Since the
object struck was considered to be more important than the implement
used for striking, it is the stone which is referred to in the rubric;
and this is true of all subsequent documents which refer to the new
fire kindled by friction. Of the medieval writers who refer to the
ceremony only John Beleth mentions that the spark is produced by strik-
ing the stone~ calibe ~ ~' 'with iron pyrites (!!§g) or iron'.3
1 J.G.Frazer, Golden ~uga 7.1 p.145.

2 Ordines 26.3; (27.6 ; 2 .25; 29.14; 31.29.
3 Rationale (PL 202.110B).
153
The ancients had long known the result of striking either of these min=
arals against flint. The use of the two=rock metho~ 9 that is 9 ~ith iron
pyrites and flint 9 when it had long been known that a spark is more
easily obtained using iron and flint, may suggest the survival from a
pagan religious ceremony, which antedated the discovery of iron, of a
very primitive means of producing fire. Its mention by Beleth implies
that in some places two stones were otruck together to obtain fire;
whilst elsewhere tho method of striking the flint with iron or steel had
been adopted. Since it is almost invariably only the flint or the
stone object struck that is referred to, we can have no idea how wide=
spread the more primitive method was.
One of the areas where this method may have been in use was South=
1
ern Italy; for according to the Beneventan rite the principal means
of generating a spark was '!! ignario'fi that is, from a fire=stone.
Ignarium is probably to be identified with the mineral copper pyrites
2
(CuFeS2). This oro, like iron pyrites, produces a spark when struck
with a flint. If this interpretation of ignarium is accepted, we may
have in the Beneventan ceremonial an instance of the two-rock metho~ of
obtaining fire. On the other hand it is possible to interpret ignarium
'flint', since lapis igniarius is found with this meaning~ The word
may be dialectal, or a synonym for silex. A similar confusion exists
in English regarding fire-stone; for the word may refer to the flint
or to the iron pyrites.
The survival into historical times of this primitive method of ob-

taining fire should cause no surprise. For it would seem that Christian
~ religiosus in Europe is just as traditional and conservative as
his pre-Christian forebears. For in an age when fire can be readily
obtained by matches, cigarette-lighters, or electrical current, in many
places the Church perpetuates the tradition of using flint.
4
1 Gradual and Missale Antiguum. Text in Hesbert, ~'Antiphonale p.188

2 Thus Du Gange. In classical Latin igniarium signified 'a stick for
making fire' (OLD), as is clear from Pliny, Natural History 16.207.
It is most unlikely that ignarium retains this meaning in the rub-
rics of the Beneventan service-books, since the Beneventan formula
for the blessing of the new fire (A) contains a reference to flint
(~). The discrepancy in spelling (ignarium and igniarium) is
not crucial. Both the Gradual and the Missale Antiquum contain
orthographical oddities.
3 Marcellus Empiricus 33.
4 In the survey of forty French cathedrals undertaken by the writer
in March 1984, a flint continued to be used as the sole means of
kindling the new fire in twenty-six churches. In a further four
it was alternative to matches.
In some documents and service=books ~hich record only the friction=
al production of fire~ the ~ord lapis is used to refer to the medium
employed in obtaining the fire. It is true that this word may indicate
any hard atone capable of causing iron or steel to produce a spark. Ho~=
ever, according to the majority of instances, including Ordo 26 our

earliest authority, the stone is specifically said to be silex, 'fire~
stone' otherwise 'flint'.(Si02.nH20). There are three main reasons for

ita alrnost~universal adoption as the means of creating a Bpark for the
new fire. (i) It is easily and cheaply obtained in most parts of Europe
and requires little attention to commission it for service. (ii) Where=
as a sunny day is a prerequisite for the production of fire by means of
a lens~ a flint may be used within or without a building and in any type
of weather. (iii) In addition to its use in the Milanese and Mozarabic
rites, and in many parts of Gaul, the adoption of the flint by the
Roman Church encouraged its use in those churches in whose rites the
new fire was kindled by means of a lens. (See Table 36 on page 157.)
The principle involved in the use of flint and steel is exactly the
same as that which lies behind the use of the tinder-box of more recent
times. The flint was struck with the steel, and the sparks thus gener-
1
ated fell into a patella or chafing=dish, if the fire was kindled
within the church. In this receptacle lay sarmentum, 'touchwood' or
'punk', twigs converted into an easily-ignitible consistency. 2 At
Freising, Vallombrosa, Cologne, and Prague dried twigs from vines were
4
used; 3 whilst a number of service-books mention simply ligna, 'twigs•.
Also in the dish, or close at hand, were pieces of charcoal, carbones,5
which would catch fire once the touchwood was ablaze. Some of the char~
coal would be transferred to the thurible in order that the new fire
might be censed. The rest would continue to burn in the vessel, if the
fire had been kindled in church, until the Easter candle was alight and
there was no danger that its flame would fail. This is attested in the
1 Nom.Ciat. p.104; Lesnes Missal (HBS 95 p.47); Dominican Missals :

1504-rol.lxxxvi and 1908 p.62. The Lateran Missal (Schmidt II p.110)
and ,!:!S 14?4 (HBS 17 p.174) use the term ~·
2 Colti II p.156. Also Sicardus, Mitrale (PL 213.322D) and MS Gg 15
(~ 4.24.3 p.145).
3 Respectively, 1487 Missal fol.ciii; 1503 Missal fol.xci; MS Gg 15
(As Note 2 above); 1498 Missal fol.xci.
4 1570 Ratisbon Ritual np; 1664 Carmelite Missal p.156. (continued p.157~
5 Their use is well-documented in a number of missals : Lesnes (HBS 95
p.47); Cistercian (1669 p.154); Evreux (1740 p.186); Sees (1?42
p.186); Amiens (1752 p.182); St Bertrand (1773 p.209); Perigueux
(1782 p.158); Le Puy (1783 p.159); Dominicans(1482 fol.69).
155
. .
Do mJ.nJ.can rJ.'t a. 1 If some form of bonfire had been prepared for the
neN fire outside the church 1 the touchwood would have been used to kind=
le the mound of wood. There
however 9 little documentary evidence
is~
2
for the construction of bonfires. The mention of strues at Spires
leaves open the possibility of a large mound of wood; and at the mon-
astery of St=Martin d 1 Ainay near Lyon the woodpile was sufficiently
large for the monks to warm themselves as they returned into church.3
In parts of Germa!JY uutil recent tim~~ bonflr~s wer~ lit at Easter from
4
the new fire. These, however, appear to have been comparable with the
Lent and Midsummer fires of Northern Europe, and in all probability
were kindled at a distance from the church after the completion of the
Paschal vigil by way of celebrating Easter. In more recent times wood=
fires observed by the writer and by those with whom the writer has com~
municated have been of very modest proportions.
The anticipation of the ceremonies of Holy Week was only one of

the causes which led to the decay of the Triduum. The abolition of the
three-day holiday by Pope Urban VIII in the seventeenth century must
have further reduced the tiny congregations of the faithful, who during
the morning of Holy Saturday could hear the deacon chant ·~ ~ ~·. 5
Moreover, one of the results of the increase in solemnity of the lit-
urgy, a sacerdotal domain in which the discharge of ceremonial came to
be regarded, increasingly after the Counter-Reformation, as the sole
prerogative of the priesthood, was the consequent exclusion and almost
irrelevant presence of the laity at any divine service other than the
Mass; and whereas in the Middle Ages the ritual that accompanied the
kindling of the new fire provided a liturgical pageantry which was cap-
able of winning the attention of the laity, to say nothing of the ex-
citement there must have been at the prospect of carrying some of the
new fire to their homes, 6 the ceremony of the new fire in later cent-
uries lost much of its former significance and importance. The anti-
cipated ritual on Holy Saturday morning was now of little liturgical
relevance to the laity; and those who did attend were present out of
either habit or curiosity.
The Holy Week reforms of 1955 within the Roman Catholic Church
were an attempt not only to restore the ceremonies of the Triduum to
1 1504 Missal fol.lxxxvi. 2 1512 Agenda fol.xciii.

3 Martene, ~ 4.22 p.125, M175·
4 Van Gennep 1.3 p.1259; Frazer, Golden Bough 7.1 p.141.
5 Beauduin p.6. 6 See Chapter 8 p.185.
156
Ear1y evidence 36 FloreE.l.ce c.1300
37 Frejus = 1754
1 Ordo 26 750=75 38 Halberstadt c.1505
2 Ordo 28 c.8oo 39 Hildesheim = 1499
3 Ordo 29 = 870-90 40 Irel.and c. 1200
4 ~ .£·950 41 Laon 1662
5 Alcuin c. 1000 42 La Rochelle 1835
43 L'on 10 c
Monaotic evidence 44 Le Puy 1783
45 Liego 1492
6 Austin Friart; 1491 46 Limoges 1830
7 Camaldolese 1503 47 Li.sieux 1752
8 Capuchins 1775 48 Lu9on 1828
9 Carmelites Co 1312 49 Lyon 1510
10 Cistercians = 1689 50 Mainz 1507
11 Cluny 1510 51 Meaux 1845
12 Dominicans 1504 52 Mende 1766
13 Franciscans 1243 53 Metz 1829
14 Melk 1495 54 Milan 1560
15 Monte Cassino 12 c 55 Nantes 1837
16 Regularis ~· .£· 970 56 Palencia 1568
17 Vallombrosa 1503 57 Paris 1662
58 Passau 1503
Secular evidence 59 Perigueux 1782
60 Poi tiers 1524
18 Amiens 1752 61 Reims 1770
19 Anderlecht 14 c 62 Rouen 1640
20 Aquileia 1519 63 St Bertrand 1773
21 Autun 1845 64 Salisbury 13 c
22 Basel 1488 65 Salzburg 1507
23 Bayonne 1543 66 Saragossa 1552
24 Beneventum c.1000 67 Sees 1742
25 Besan~on 1707 68 Seville 1507
26 Bourges 1741 69 Scissons 1745
27 Braga 1558 70 Toulouse 1832
28 Breslau 1483 71 Tours 1784
29 Cahors 1760 72 Trier 1488
30 Carcassonne 1749 73 Troyes 1736
31 Chartres 1782 74 Valence 1504
32 Cosenza 1549 75 Vienna 1782
33 Coutances 1825 76 Wiirzburg 1477
34 Esztergom 1501
35 Evreux 1740 (References are given on p.218.)
Table 36. Evidence for the use of flint and steel.

(The omission of any reference in a num-
ber of missals to the means of kindling
the fire may indicate that any method
was permissible.)
Continuation of Note 4 on page 155.
At Poitiers tow was also used (1524 Missal fol.lxix). The Carmelites
used small palm and olive branches (1664 Missal p.156); whilst the
Augustinian Friars placed flowers and fragrant herbs on top of the
olive twigs which they used (Ceremonial of 1714 p.307).
157
their original times of performance~ but to bring about a much gre~ter
lay participation in the Paschal liturgyo In the surge of enthusiasm

follo~ing those roformsi the presence of a large congregation in the
parvis of a church or cathedral encouraged the construction of large
bonfires. But the frustration, disappointment, and even ridicule ex~
perienced when the rain~sodden wood refused to catch firei and the wane
in interest in the ceremony in recent years have resulted in fires of
considerably less magnitudeo Other problems encountered in staging the
ceremony in the open air include the danger to life and limb risked by
the clergy attempting to read the prayers by the light of a small fire 9
and the inclemency of the weather, which makes difficult the lighting
of the incense, not to mention the new fire itself, and the possibility
1
of having the Easter candle extinguished by a sudden gust of wind. In
some churches, such as the Cathedrals of Annecy and Arras, the charcoals
which used to be brought into church in a small portable stove have
2
been replaced by a rag soaked in spirits. In 1983 at Tulle Cathedral
a quantity of methylated spirit burned in a small dish until the
2
Paschal vigil had ended.
The Roman Catholic Church never officially countenanced the use of

matches for the kindling of the new fire until 1970. However, in many
3
parishes in England and France and in some cathedrals matches have
been used for many years, mainly because of their reliability, and be~
cause of the speed with which fire can be obtained - the use of flint
and steel requires some skill. It is significant that most matches are
ignited by striking their heads against glass-paper, one of the constit-
uents of which is silicon dioxide (Si02 ), the basic compound of flint.
The 'matches' referred to in C.M.Merati's Ceremonies of ~~Church
were fire-sticks, the ligna sulphurata, described below on page 206.
The earliest evidence for the use of flint as the sole means of ob-
taining fire in the Roman rite is to be found in the Roman Missal of
4
1474. Although the possibility must not be ruled out that a lens may
have been used at Rome in the fifteenth century, the latest documentary
evidence for the use of a lens antedates this missal by about a century. 5
The use of a flint, prescribed in the Pian Missal of 1570,
6 was also
1 For the problems encountered following the revival of the ceremony

in the open air in France, see Morlot pp.114-22.
2 1984 Survey. /vey).
3 For instance, Aix, Arras, Digne, Laval, Reims, Strasbourg.(1984 Sur
4 Missale Romanum (HBS 17 p.174).
5 Ordo of P.Amiel (OPA), PL 78.1321c. 6 Missale Romanum (1950)p.186.
1
enjoined in the 1955 Revised Order of Holy Veek; but in the 1970
Nissale Homanum the historical and symbolic method of kindling the netJ
2
fire has been abandoned, and the rubric 1
a large fire is prepared'
allows of any convenient means of producing a flame. In ~,Holy ~'
~ Easter, the new service~book of the Church of England, the rubric

relating to the new fire contains no firm injunction, but merely states
that 'According to ancient custom, the light for the Easter candle was
taken from newly=kindled fire and not from an already existing source
of light 0 (p.226). It does, howeveri countenance the lighting of a
bonfire outside the building and a procession of the faithful into
church accompanying the new fire.3
The Symbolism of the Flint
Medieval commentators were not slow to detect a symbolic signific~
ance in the use of the flint, in the action of striking, and in the gen-
eration of a spark. In his identification of the stone with Christ,
4
Rupert of Deutz cites the authority of Psalm 117. The same author sees
in the striking of the stone a symbolic reminder of the Crucifixion;
and for him the production of the spark represents the release of the
Holy Spirit. 5 The notion of the symbolic representation of Christ by
7
the stone is mentioned by both John Beleth6 and Sicardus. The former
adds that the Church has been built on the rock which stands for Christ,
and from that rock comes the New Law. He thus implicitly draws a para~
llel with Moses and the tablets of stone in Chapter 32 of Exodus.

Durandus enlarges upon the symbolism, attached by these three writers
to the stone, by adding that the spark produced from the striking of
the stone represents the fire of God's love, and at the same time sym-
bolises the piercing of Jesus' side, out of which flowed the blood and
8
the water. In more recent times Gueranger has likened the spark leap-
ing from the flint to Christ rising from the rock-hewn sepulchre;9 and
Bouyer, expressing the notion more succinctly, writes that the new fire,
drawn from the flint, symbolises 'the divine spark that God himself
would cause to rise from the sepulchre of Golgotha to kindle the uni-
. 10
verse at the flame of Christ's own splendour'. As late as 1956
1 Schmidt I p.118. 2 Praeparatur rogue. (1970 t.e. p.267).

3 The Anglican rite provides no prayer for the blessing of the fire. •
4 De Div.Off. V (PL 170.149A) lapidem quem reprobaverunt
;;dific~es (verse 22).
5 Ibidem (PL 170.149B). 6 Rationale (PL 202.111B).
7 Mitrale (PL 213.323A). 8 Rationale VI.80 p.350.
9 Liturgical Year (Passion Time and Holy Week) p.554.
10 1!!!. Paschal Mystery p.267. * e.g. 1986 C.of E. and 1979 American BCP.
159
in a preamble to his edition of the revised services of Holy Heek, Dom
Godfrey Diekman described the kindling of fire from the flint as 'a
vivid image of Christ 0 s new presence among men : as the spark leaps
1
from the flint, so He arose from His rock tomb.' It is important to
note that there is a twofold symbolism in the identification of Christ
with the stone. For not only is He the lapis angularis of Psalm 117;
Christ is also the stone through the medium of which God's brightness
is brought to the faithful. The notion that it is the rock that ~tands
2
for Christ is also mentioned by Thurst@n.
With the simplification of the Paschal ceremonies in the liturgical

changes following the Second Vatican Council, much of the traditional
symbolism associated with the new fire ceremony was omitted from the
Missal of 1970. With the abandoning of the flint as the obligatory
means of obtaining the new fire, it was almost inevitable that the re-
ference to the flint in benediction-formula A should be removed from
that prayer.
It is perhaps not out of place at this juncture to refer to the be-

lief, recorded by Rupert of Deutz, that fire was inherent in certain
types of rock, for example, flint.3 John of Avranches implies that the
spark released from the stone had been confined within it prior to the
4
act of striking. The use of the verb elicere, 'to draw out', by anum-
ber of writers within this context reinforces this idea. It is beyond
the scope of this work to discuss ancient and medieval notions regard-
ing the nature and essence of fire. However, it is not difficult to
understand that the act of generating a spark from a stone was regarded
by the unscientific minds of primitive men both as a miracle and as a
mystery; so that the stone from which the fire leapt came to be held
in reverential awe, and the fire was believed to participate in the
divine nature and to be the visible presence of God in the world. This
belief is of course present in the Old Testament in the incidents of
the Burning Bush and the Pillar of Fire. Christian thought had inheri-
ted from Judaism the notion of conceiving the nature of God in terms of
fire and light; had evolved a theology of light from the discourses of
Jesus, as recorded in the Johannine writings; and to a certain extent
had adopted the pagan philosophical notion that fire was the underlying
principle behind creation.
1 ~Masses gi Holy Week p.127. 2 ~ ~ Ii2!;y; ~ p.411.
---
3 De Div.Off.V (PL 176:149A). 4 Y&.• <!! Q.U.~. (PL 147 .49A).
160
(ii) ~ ~ refraction 2f the sun's rays
Throughout this section 'lens 0 translates crystallum 9 the torm

found in our sources to signify the semi-precious stones, such as beryl
and rock-crystal, which were used to produce fire by the refraction of
the sun's rays. There is good reason to believe that these stones 9
1
tihich were widely used for this purpose in the ancient world, also
featured in pagan religious new firo ceremonies which were adopted by
the Church into her liturgy. Although the 1:1ord vitrum, 0
glass', is not
found within the context of the new fire rubrics, this material was
probably used as a substitute for translucent stones on occasions.
Since the use of a lens for the production of the new fire was de-
pendent upon the shining of the sun at the required moment, it may be
safely assumed that in those churches where the new fire was kindled by
this method, alternative means of obtaining fire must always have been
readily available in the event of a cloudy day. The oburnbration of the
sun is succinctly alluded to in the Missal of St-Martin d'Ainay. 2 Of
the thirty-two instances, compiled from documentary sources and listed
in Table 37: where a lens is stated to be a means of producing fire,
six, including St-Martin d 1 Ainay referred-to above, do not record an
alternative method. To this group belong the twelfth-century Ritual of
Soissons and the eleventh~century Sacramentary of Holy Trinity on Mons
Suavicinius. 3
In the Sacramentary Honorius of Autun refers only to Crystallum; but

4
in the Gemma Animae we find that a flint is alternative to a lens.
This would suggest that the lens was the principal but not the sole de-
vice for kindling the new fire in the liturgical milieu with which
Honorius was familiar. Hugh of St Victor refers only to the lens, and
also mentions the pieces of charcoal. In a somewhat forced analogy he
likens Christ to the mediatorial rays which shine through the lens and
bring back to life the 'dead' charcoals, which themselves symbolise the
souls of men in bondage to sin.
In the Customary of Cluny the lens is specifically stated to be a

beryl. As at St Benigna's, Dijon, it was kept in the custody of an
1 See especially F.Dolger, ~ Karsamstag Feuer pp.288-96.

2 '§!~poteet.' The phrase also indicates the importance
attached to this method. /formula J.
3 The use of a lens here is suggested by the use of benediction-
4 PL 172.?46A and PL 172.668A, respectively. • See p.165.
161
official called the apocrisar~ (sonior sacristan)~ 1:1ho carried H in
the procession for the blessing of the new fireo It was perhaps inevit-
able therefore~ in view of the importance attached to this translucent
stone at Cluny, that reference to an alternative fire=kindling device
was omitted from the rubrics relating to the new fire ceremony. In
addition to the two monasteries just mentioned, a beryl was also used
at fleury 9 Barking~ and York, and by the Gilbertineso
Unlike the identification of lfi ''{V llJ i.n the Old Testament~
1
there seems little doubt that the beryl referred to in our medieval
texts within the new fire context is to be equated with tho aluminium
silicate of beryllium (Be3Al2Si6018), the chemical composition of the
gemstone which is still known by that name. For in an eighth-century
(or ninth-century) description of the stone we read
Beryllus, nube aure togitur et sex angulos habet, tenentem

manu adurere dicitur.2
The vitreous and almost resinous quality of the beryl could well be
said to give the impression that the stone was surrounded by a haze;
and this particular silicate does indeed crystallise in the hexagonal
system. Its reputation for burning the hand, if held, almost certainly
arose, partly from its use in kindling the new fire, partly from ignor-
ance, and partly from its being endowed with wonderfully strange quali-
ties by the superstitious and unscientific mind of medieval man. For
the beryl, which was used for the new fire, almost certainly saw the
light of daypn only one day of the year3 and remained safely locked
away for the rest of the time. Beryls of various colours occur; but
the colourless or white variety is sufficiently translucent as to allow
the rays of the sun to pass through and to ignite combustible materials.
The view that the term beryllus within the context of the new fire mere-
ly refers to glass of high quality cannot be sustained. For in addition
to the evidence of Bede's Pseudographia, it must be stated that if the
lens had been made of glass or of rock-crystal, the writers who referred
1 In the Massoretic Text of Ex 28:20, 39:13; Ez 1:16, 10:9, 28:13; and

Dan 10:6, Tii "'lii lfl is rendered throughout in the AV 'beryl'. The
uncertainty of.the-LXX translators in their varying renditions of
this word is reflected in modern English versions, where the Hebrew
word is variously interpreted as 'chalcedony', 'chrysolite', 'a pre-
cious stone', as well as 'beryl'.
2 'Beryl is surrounded by a golden mist and has six facets. It is
said to burn the hand of anyone who holds it.' Baedae Pseudo-
graphia, PL 94.552A.
3 And presumably on three days of the year when the new fire came to
be kindled on each day of the Triduum.
162
to beryllus would almost certainly have used vitrum or crystallum re~
spectively~ and the importance attached to the lens in the procession

at Cluny and other places can more easily understood if that objec~
were a stone both rare and of considerable value rather than an easily=
obtainable piece of artificial glass.
The fact that no alternative means of kindling the new fire at

Tours is recorded by ~~rt~ne may well be explained by his desiro to
draw attention to the curious method involved in the uso of the lens.
For according to the thirteenth~century missal of that church the new
fire was produced
. "da •1
. t a 11o et aqua f r1g1
so l e e t cr1s
It is clear that the fire was kindled by means of a translucent stone

held ubove the tinderi but the mention of the third element, cold
water, seemingly essential for the production of the fire, is baffling.
This method involving water is also mentioned by John Beleth, who, when
describing how the new fire is kindled, writes
Nam si crystallus supponatur orificio phialae aqua plenae ad salem

sine mora ignis e crystallo illico excutuetur.2
Beleth gives no indication as to the shape or size of the bottle, al-

though the use of water as well as a lens in a device intended to re-
fract the rays of the sun strongly suggests that the bottle was made of
glass or some translucent substance.3 Without more information it is
difficult for us at this point in time to understand why the lens alone
was not sufficient to produce fire. One possibility is that the water-
filled bottle with a lens for a stopper provided a double refractive
surface and so, it was believed, enhanced the effectiveness of the de-
vice. The use of a translucent container for the production of the new
4
fire is recorded in the tenth-century Sacramentary of Corbie. It refers
1 'By sunlight, a lens, and cold water.' Mart~ne,~ 4.22.5 p.96 (M324).
2 'For if a lens is fitted in the mouth of a bottle filled with
water and tilted towards the sun, fire is very quickly produced
from the lens at that place.' PL 202.111B.
3 The use of rock-crystal for the manufacture of containers for
liquids is attested by Solinus (£21!.~ ~· 15, 29-31) :
cited by Dolger, ~ Karsamstag p.294.
4 PL 78.336B :~~~ampulla.!_~ illuminatum (~!..ill
~ excussum). Originally and by derivation an ampulla was a vessel
with two handles. Even in classical times the feature character-
istic of this container was a convex or bulbous shape. See, for ex-
ample, Plautus, ~.5.2.86; Persius 1.}.44; Cicero, ~ ~.4.12
and Atticus 1.14; Horace, ~ Poetica 97.
163
to an ampulla. l:Je are not told whether this vossol tJas filled tJith
~ater; but it must be presupposed that it ~as of a vitreous or cryst=
allino composition, and that part of its external surface wms convex in
shape. The need for a translucent container t-~hich tJould allow the rays
of the sun to pass through makes it extremely likely that the phiala,
referred to by Beleth, was also made of glass or crystal.
A choice of devices for refracting the sun's rays is att0stod in

the Pontifical of Poitiers :
ignis de christallo sive do amula sumitur 9 vel etiam de cote, si

neutrum horum fuerit, excutitur.1
The third choice provides for obtaining the new fire by frictional
means, should it not be possible to use either of the other two methods,
both of which involve the refraction of sunlight. There is no reason
to suppose that christallo indicat0s anything other than a lens of some
translucent stone. The presence of ~ between christallo and ~ ~
does not signify that this word relates to a method of obtaining fire
other than by refraction or friction; for it is difficult to see in
what other way fire could have been produced quickly given the circum=
stances in which it was obtained. Moreover, since~ amu1a and ~
christallo are dependent predicatively upon sumitur and linked to each

other by~ (as opposed to vel which contrasts the two main verbs),
it is almost certain that amula is an alternative refractive device.
The term is otherwise unknown within the context of the new fire. Most
attempts to connect the word with other documented nouns seem fraught
with difficulties and forced. Amula seems to have little in common with
(h)amulus 2 (a little hook), amulum3 (starch), and amuletum4 (amulet);
and the link with hamula5 (a small water-bucket used for extinguishing
fires) is surely one only in spelling. However, in view of the occur-
rence of ampulla in the Sacramentary of Corbie in the rubric relating
to the new fire, 6 the moat likely interpretation of amula is to regard
it either as a dialectal or regional variation of ampulla, or simply
as a wrongly-spelt form of this latter word.
The use of a lens, in preference to a flint, at Barking, Upsala,

and York shows that its use was not confined to the generally sunnier
1 'Fire is obtained from a lens or from amula, or, failing either of

these, it is struck from a stone' (p.1}8).
2 Plautus, ~.2.1.17. 3 Celsua 2.20.
4 Pliny, ~.ill:§1.29.4.19. 5 Columela 10.}87.
6 See Note 4 on the previous page.
164
Church/ ~ ~ Flint Source
Document
Poi tiers .£· 900 1 2 Poitiers p.138

PRG (MS C) f. .£· 950 2 1 II p.94 @ 342
Corbie 10 c 1 2 PL 78.336B (M106)
Cluny/ 11 c 1¢ PL 149,659A
St Benigna) 11 c 1"' 2 ~ 3.13.34
Dijon p.126 (M 1150)
1¢ ...,
Flew:·.t 11 c c. Albers V p.143
+Old Sacramentary11 c 1 PL 151 .846B
.Avellana 11 c 1 2 PL 151.880D
Ripoll (m) 1038 2 1 Sacramentaey p.92
Gilbertines 1150 2* 1 HBS 59 p.39
York 12 c 1* 2 Missal p.109 /(M305)
So is sons 1180-90 1 ~ 4o21 p.161
Tours 13 c 1 ~ 4e22 p.96{M324)
AJ.bi (m) 14 c 2 1 Feasey (1906) P•355
St ~1ary,York c.1400 1* 2 HBS 75 p.275
Barking 1404 1* 2 HBS 65 p.101
Prague 1498 1 2 Missal fol. xci
Spires 1512 1 2 Agenda fol.xciii
Upsala 1513 1 2 Missal fol.lxxvi
Narbonne (m) 1528 2 1 Missal fol.lxxxiii
St=Martin 1531 1 ~ 4.22 p.125
d'Ainay (M 175)
Burgos (m) 1546 2 1 Missal fol. ciii
Osma (m) 1561 2 1 Missal fol.lxxxix
Rat is bon 1570 1 2 Ritual np
Freising 1579 1 2 Missal fol.85
Sens 1715 1 2 Missal p.238
Commentator QW ~ ~ Source
Rupert (Deutz) 1111 2 1 PL 170.149A

Hugh (St Victor)£.1140 1 PL 177.8890
Honorius (Autun)c.1150 1 2 PL 172.668A
John Beleth c.T180 2 1 PL 202.111B
Sicardus c.1200 2 1 PL 213.3220
Bauldry 1762 2 1 Manuale p.189
+ used at the Church of the Holy Trinity, Mons Suavicinius.

* signifies the use of a beryl. 1. See p.180.
(m) indicates possible Mozarabic influence.
1
Table 37. Evidence for the production of fire by refraction.
(The figure 1 in the third and fourth columns in-
dicates that the rubrics give this method prior-
ity. The figure 2 indicates the alternative method.)
and warmer parts of Southern Europe. 2 A hitherto largely-unnoticed
1 In addition to the above there are two ordinaries cited in the 1856
Ritual of Soissons - see p.167- and the evidence of Bede, which is
discussed on the following page.
2 The writer has demonstrated by experiment that fire can be kindled
with the use of a lens at an early hour on both a late autumnal and
an early spring morning at a latitude of 540 N.
165
passage in Bede's ~ Tabernaculo may well be a reference to the use of
the lens, and if so 9 would antedate the first recorded instance of its
use (~) by over tuo hundred years.
Nullum offerri licet tabernaculo Dei, nisi quod de lignis olivarum

conficitur~ sicut nee ignis alius quam qui coelo descendit 9 vel in
lucernis sanctis vel in altari Dei debet accendi. Offerant ergo
filii Israel oleum ad lucernam Dei, non qualecumque, sed de lignis
olivarum, et hoc purissimum 9 piloque contusum.1
This reference to fire occurs in a passage containing regulations re-

lating to the use of oil in church lampse It is true that the passage
contains Old Testament allusions, and an attempt is made to justify the
liturgical practices of the day by appealing to biblical precedents;
but it is difficult to see any reference or allusion here to an instance
in the Old Testament of fire from heaven. For on those occasions when
fire is said to have originated \d. th the Lord, some (such as Lev 9:24
and 1 Ki 18:38) involve the consuming of the sacrifice upon the altar;
others (for example,Gen 19:24, Lev 10:2, Num 11:1) describe the punish=
ment of those who had sinned against the Lord. In the passage under
discussion neither of these above-mentioned aspects of fire seems to be
of relevance. It is more the fire which burns in the lamps for illumin=
ation or on the altar for symbolic reasons that is of concern to Bede,
and for this reason he is almost certainly referring to the source of
the flame for those lamps. Moreove~in the only instance in the Old
Testament where a downward movement of fire is recorded, viz. in the
contest on Mt Carmel (1 Ki 18:}8), the fire is said to have fallen
rather than to have descended ~ heaven;
although it must be admitted
2
that in that context there is little difference, if any, in meaning.
Since there was no tradition in Judaism of obtaining fire for liturgical
lights from any particular source, as opposed to either the flint or
lens in Christian ritual, it is tempting to see in the phrase ignis ••••
qui ~ descendit an implicit but unmistakable reference to the
kindling of the new fire by means of a lens.
1 'It is only permissible to offer for use in the temple of God oil
which is made from the fruit of the olive, just as the fire which
burns in the sacred lamps or upon the altar of God should only be
that which has come down from heaven. Accordingly let the oil which
the sons of Israel bring for the lamp of God be not any kind of oil,
but obtained from the fruit of the olive, of the best quality, and
beaten with a stick. 1 PL 91.463.. •
2 Neither in the original Hebrewi1Ji1~-lll?; ?·~.fll, nor in the Latin
translation cecidit ~ ignis Domini with which Bede would have
been familiar, does the phrase £!2m heaven occur.
166
The compilers of the 1856 Ritual of Soissons claimed that the
mention and use of crystallum in former times implied that the new fire
~as kindled by frictional means; and that the historian Jean Cabaret
uas guilty of a very serious mistake \..rhen he \'Jrote that at Soissons the
new fire was produced 'by a lens of crystal glass which they used to
expose to the rays of the sun' (p.308 note xi)e In support of their
assertion they cited the rubric from a:!l Ordinary of St=t~art:ial~ Limoges
ignc do cristallo vel silice noviter excusso,
in which the participle excusso qualifies both nouns, and part of the
corresponding instruction in the Ordinary of St=Pierre d'Orval
silice vel cristallo noviter cum calibe.
It is true that in the second instance~ calibe, 'with a steel',

appears to relate to both silice and cristallo; but the absence of a
verb makes difficult the interpretation of the rubric which, being in-
complete, is therefore inadmissible as corroborative evidence. Moreove~
the addition of~ calibe is rare in other rubrics similar to this.

The compilers would almost certainly have been familiar with the rubric
of the twelfth-century Ritual of Soissons which prescribed the use of
a lens (Table 37); but they may not have been able to reconcile this
information with the rubric of their 1745 Missal which referred to the
use of a flint (~, p.163).It would seem, therefore, that they
attempted to explain this inconsiotency of practice by maintaining that
the crystallus had formerly served the same purpose as the flint.
The compilers were apparently unaware of John Beleth's statement,

quoted on page 163, or Rupert of Deutz's reference to the use of a lens
1
'on a cloudless day', or the evidence of Sicardus. Moreover it would
seem that the verb excutere, whose perfect participle excussus is used
in the first of the above-quoted rubrics, and which formerly suggested
'shaking out' or 'driving out' with the use of rapid or violent move-
ment, had either extended its meaning to include other methods of kind-
ling fire, as Rubrics 1, a, 4, and 7 in Table }8 would suggest; or was
linking cristallo and silica in a sylleptic union. At first sight the
1
new fire rubric in the Gilbertine Ordinal would appear to support the
view of the nineteenth-century rubricians of Soissons :
a cilice (§i£) vel cristallo aut berillo excussus.
1 For references, see Table 37 on page 165.
167
In classical Latin the use of !!! would have indicated that silex and
cristallus ~ere variants within the same category, as opposed to
beryllus, which was contrasted with them by the use of vel. In medi-
eval Latin these former niceties of distinction no longer obtained.
Indeed, in the above-quoted instance and in the corresponding rubric in
1
the Pontifical of Poitiers the role and function of vel is reversed.
A number of observatioDB may be adduced in favour of the tradition-

al interpretation and understanding of the uae of cryatallus to show
that the liturgists of Soissons were themselves in error, and that the
word crystallus within the present context signified a lena used for
generating fire by the refraction of the sun's rays. (i) In earlier
documents it is clear that two different methods are involved from the
use of two separate verbs to express the actions of kindling fire by re-
fraction and by friction : in Poitiers sumere and excutere respectively,
1
and in the Customary of Fleur,y producere and excutere respectively.
(ii) The symbolic interpretation of the lens makes sense only if the
principle of refraction is under discussion. (iii) There are clear re=
1
ferences to the sun in the Sacramentary of Corbie and by John Beleth;
and the Ordinal of St Mary's, York refers to the possibility of cloud
covering. (iv) It is difficult to believe that a semi-precious stone
was subjected to the violent impact of a bar of iron. (v) Few semi-
precious stones, such as beryl, will produce a spark if struck with
iron. Moreover, their small size would have necessitated their being
held in a clamp or other similar device.
The writer is of the opinion that the placing in the rubric of the
lens before the flint indicates that within the rite priority was given
to the use of the lens. The inclusion of the alternative method of
fire by friction was a prudent precaution designed to obviate the poss-
ible dismay or frustration likely to be experienced in the event of a
cloudy day. In instances in which the rubric prescribes the use of the
flint first, followed by the alternative of a lens, it should not be
assumed that the frictional method of kindling fire had priority. In
many instances it is likely that a simple choice existed and that either
method was valid liturgically. One suspects that in a number of rites
the existence of a choice represents a synthesis of ceremonial elements
drawn from different cultural or religious traditions. The preference
of moat of the liturgical commentators for the frictional method of
1 For references, see Table 37·

kindling firs reflects the fact that the use of the flint was far more
common than that of the lens throughout the Western Church.
The Symbolism of the Lens
Whereas the action of striking a steel against a flint was thought

to result in the release of a spark contained in the flint, a lens was
believed to be the means of transferring fire from the sun to tho earth.
Medieval writers seized upon the potential symbolism inherent in this
method of obtaining fire, not only because of the obvious similarity in
function between a fire-producing lens and the mediatorial role of Christ,
but also because of the closeness of spelling and the identical pronunc-
iation of Christ with part of the Latin word for lens, cristallum.
Commenting upon the method of obtaining fire by the refraction of the
sun's rays, Rupert of Deutz compares the sun with God and the pieces of
charcoal with men in bondage to death. The lens is Christ who mediates
1
between God and men, and who brings life to the latter. Similarly the
fire which the lens brings into being and which ignites the charcoal is
the Holy Spirit which Jesus told his disciples he would send (Jn 14:26).
In another sense the lens brings the fire of God's love to men; for
without the mediation of Christ they would exist without it. 2 The sym-
bolic interpretation of the lens as Christ the Mediator is also mention-
4
ed by both Sicardus3 and Durandus. The notion of the fire representing
the Spirit of Christ spread amongst the faithful is found in the Gemma
Animae of Honorius of Autun. 5 John Beleth and Honorius both elaborate
the idea of the lens representing Christ; for they see the translucent
stone as symbolising the 'clear resurrection bo~ of Christ•. 6 Beleth
describes the resurrectional flesh of Christ as 'pure and pellucid', 7
a notion echoed in the Missal of Upsala •8
• • • • •
In some churches there was no new fire ceremony as such; instead,

the new fire was 'taken from elsewhere' (sumptus aliunde). This pract-
is discussed in Chapter 8 on page 186.
1 This notion is also found in the Miscellanea of Hugh of St Victor

(PL 177.889D).
2 !2! Div.QU. V (PL 170.149C). 3 Mitrale (PL 213.323A).
4 Rationale VI.80 p.350. Durandus says that the lens comes between
the sun and the moon. It is almost certain that he or a later
copyist should have written earth for the latter noun.
5 PL 172.668B. ~ Honorius, ibidem.
7 Rationale (PL 202.111B). 8 1513 Missal fol.lxxvi.
169
Chapter Seven
THE NE\~ FIRE AT :OOME
(i) The eighth-century evidence £! Pope Zachary
The earliest explicit reference to the new fire ceremony at Rome is

found i~ the following extract from a letter which Pope Zachary (741=52)
wrote in reply to an enquiry from Boniface, the Bishop of Mainz
De igne autem paschali quod inquisisti, a sanctis priacis patribus,

ex quo per Dei et Domini nostri Jesu Christi gratiam et pretioso
sanguine eius Ecclesia dedicata est, quinta feria Paschae, dum sacrum
consecretur, tree lampades magnae capacitatis, ex diversis candelis
ecclesiae oleo collecto, in secretiore ecclesiae loco ad figuram
interioris tabernaculi insistente indeficienter cum multa diligentia
ardebunt, ita ut oleum sufficere possit usque ad tertium diem. De
quibus candelis sabbato sancto pro sacro fontis baptismate sumptus
ignis per sacerdotem renovabitur. De crystallis autem, ut asseru-
isti, nullam habemus traditionem.1
Before we discuss the practice of reserving fire and comment on the

ceremonial details, we must first deal with a number of points of inter-
pretation which this passage raises. The custom of reserving the fire
at Rome on Maundy Thursday and renewing it on Holy Saturday was obvious-
ly of long standing. Zachary himself was writing in the 74os; and it
is reasonable to suppose that the practice went back at least to the
middle of the seventh century. The Pope was in no doubt that the cere-
mony was of very great antiquity, although precisely which era he is re-
ferring to in the above-quoted extract is debatable. For from a temp-
oral point of view the first three lines of the extract from the letter
are capable of two different interpretations. (i) The Latin text is
taken from the nineteenth-century edition of J.P.Migne. In the third
line of the Latin text the word for 'church' is printed with a capital
letter E. The editor clearly regards Zachary as here envisaging
1 'Now concerning the Paschal fire about which you enquired : since the
time of our saintly fathers of old, when the Church was established
(or when the church was dedicated) through the grace of God and our
Lord Jesus Christ and by His precious blood, on Thursday of Holy
Week while the Sacrament is being consecrated, three large lamps,
fuelled by a copious supply of oil which had been collected from the
various lamps in church, will be tended with great care and will
burn continuously in a remote part of the church, recalling the
flame of the inner tabernacle. There is sufficient oil to last
until the third day. From these lamps fire will be taken on Holy
Saturday for the sacred Baptism at the font, and will be rehallowed
by the Bishop (or by a priest). As you have made mention of lenses,
we have no tradition of using them.' (PL 89.951.)
170
the universal Church which came into being after the Resurrection; and
the sancti prisci patres are presumably the twelv~ apoatleso According
to this interpretation, therefore, it would appear that Zachary is ascr=
ibing a venerable age to the custom of reserving the Paschal fire, and
is claiming an antiquity for the practice co=eval with the Church itself.
(ii) Alternatively, if Ecclesia refers to the Cathedral of StJohn
Lateran and the sancti prisci patres are to be identified with the lead=
ers of the Church in Rome at the time of St John 9 s Constantinian fouuda=
tion, Zachary is attributing an age of about 4oo years to the customo
This latter explanation has two serious drawbackso First, the phrase
pretioso sanguine can only be accommodated with difficulty within a
fourth-century Roman context. Secondly, the identification of the
sancti prisci patres with a fourth-century Roman bishop and his pres-
byters is strained.
There is probably an element of credibility in both interpretations •

For on the one hand Zachary is specifically referring to the Church in
Rome, since his description of the way in which the fire is reserved re-
lates to Roman usage. Yet he is also claiming an apostolic descent and
approbation for the practice in order, perhaps, to assert the prior
claim of Rome to be guardian of the Church's traditions. For the Roman
Church had been established with the apostolic authority of St Peter;
Mainz and the other churches of Gaul and Germany had no comparable found-
ation. Yet at the same time one suspects that the Pope may have been
aware that the customs relating to the reservation of the Paschal fire
may have derived in part from pagan Roman ceremonial. The attribution
of the origin of the new fire ritual at Rome to the Apostolic Age may
have been a deliberate attempt by him, or more likely one of his pre-
decessors, to discountenance the suggestion or belief that the ritual
surrounding the reservation of fire at Rome was inherited from a pre-
1
Christian milieu. However, as we shall mention again, the custom of
renewing the sacred fire at Rome at this time of the year antedates the
Age of the Apostles by many centuries; for it was a feature of the
ancient Roman religion. Zachary's claim, therefore, that the reserva-
tion and renewal of fire during Holy Week is a practice of the Roman
Church dating from the first century is not entirely without truth. It
must be stressed, however, that the ceremonial surrounding the provision
of Paschal fire at Rome also contained elements drawn from the Lucerna-
~ and suggested by the events of the first Holy Week. These will be
discussed in Chapter 16 of Part IV.
1 See Chapter 8 for the Church's adoption of the new fire ritual.
171
It is clear from the ~ords innistente indeficienter •••• ardebunt
and from tho fact that they ~ere large tha~ the three lamp~ burned
simultaneously and not singly in succession as the supply of oil in
each ono became exhaustedo The size of these three lamps can only be
estimated by the amount of oil it was necessary for each to hold in its
reservoir. It is known that half a log of oil was required for keeping
alight the l"l.Jn
. .,. I].. (ner tam!d)
- in the Jewish Temple during a wintcr'o
1
night. If we ~ccept that a log wao equivalent to about one imperial
pint or half a litre and tha~or the sake of our argument a winter 0 s
2
night comprised twelve hours, each of the Roman lamps would have ro=
quired an oil reservoir with a capacity of at least two pints in order
to provide continuous illumination for the forty~eight hours between
the evenings of Maundy Thursday and Holy Saturday. This accords well
with the description of the lamps as being 'magnae capacitatis'. The
concealing of the lamps 'in a remote part• of the basilica prompted
Zachary to draw the analogy between the reserved fire of Rome and the
I")Jll
• T
I) in Jerusalem. The phrase cum multa diligentia reveals the
importance attached to keeping the fire alive and the concern felt that
it should not go out. The phrase would imply that the lamps were
constantly attended during this two-day period.
The phrase !a secretiore ecclesiae loco clearly indicates that the

lamps were placed in a less frequented part of the basilica. It is just
possible that the sacristy was used for this purpose; although one
might perhaps have expected Zachary to specify the sacristy if the fire
had been reserved in that room. However, the sepulchral aspect of the
occasion and the circumstances in which the fire was reserved demanded
that the place of concealment should be as far removed from the gaze of
men as possible; and although the Cathedral of St John Lateran remain-
ed unused for divine worship during the whole of Good Friday and for
most of Holy Saturday, it is likely that the sacristy of the basilica
received a number of visitors during that period. Moreover, the size of
that church afforded a much more suitable location for the reservation
of the fire than the sacristy.
It is true that the reservation of fire on Maundy Thursday was also

a feature of some of the Gallican rites. However, in those rites the
fire was reserved on Maundy Thursday in order to replace that lost at
Tenebrae on Maundy Thursday, Good Friday, and Holy Saturday. 3 At Rome
1 Babylonian Talmud, Shabbath 22b, p.96 note 13.

2 DeVaux, Ancient Israel pp.202-3. 3 See above, pp.101-2.
172
in the eighth century the purpose in reserving the fire wao partly
functional = to provide light at Baptism on Holy Saturday after the
Vigil = but its principal function was to convey through the symbolism
of fire something of the death and resurrection of Jesus. This was
facilitated by the use to which St John Lateran was put on Good Friday
and Holy Saturday. Commenting on the later practice of illuminating
the altar and church lights with a flame from the triple candle, Van
Gennep stated
'L 0 ensemble constitue una dramatisation viauelle du
~
1
schema deB~ de Passage: separation, marge, ot ronoissancaov
His triple-phase theory is equally pertinent to the ceremonial of the
Roman Church in the eighth century. For the fire is removed from view
together with all the oil in the church lamps; it is kept in a state
of limbo in its place of reservation; and it is subsequently reproduced
for use at Baptism on Holy Saturday. However, it is not at all clear to
what extent the new fire ritual of the Roman Church derived from pagan
antecedents,since the events of the latter part of Holy Week provided
a parallel historical setting for the evolution of a liturgical re-
enactment and 'visual dramatisation' of the death, the burial, and the
resurrection of Jesus, in which He was symbolically represented by the
fire. It is true that Zachary does not mention any symbolic signific-
ance attached to the reservation of the fire; but it is our belief that
by placing the old fire in a remote part of the church, concealed from
sight and the living world in the same way that the dead are hidden away
in a tomb, the old fire was in a sense held to have died; and that the
act of blessing performed by the Pope or priest accomplished the reviv-
ification of the dead fire, analogously to the way in which God had
raised Jesus from the dead. Thurston calls the Roman method of hiding
fire an image of the death of Christ and its reappearance 'a wonderful
2
type of Resurrection'.
There is some uncertainty as to whether the reserved fire was a

newly-kindled flame or whether it was taken from an already-existing
source of fire. The silence of Zachary in this matter leaves both poss-
ibilities open. It is true that in pagan Rome new fire had been kindled
annually in March in the Temple of Vesta.3 It is unlikely, however,
that there was any continuity of practice in Rome itself once that city
had closed its temples. Paganism in Rome was suppressed more ruthless-
ly than in most rural areas where pre-Christian rituals and traditions
were often either Christianised or adopted into the life and liturgy
1 Mgnuel 1.3 p.1257 n.2.

3 Ovid, ~ III.143.
173
of ths Churcho Possible support for the view that the fire was newly~
kindled comes from the Lucernarium hymn of Prudentius 9 which was sung
1
at that daily officeo According to lines 7 and 8 of the hymn the flamo
for the lamp which burned during that service was kindled anew each dayo
Since the bringing of the reserved fire for the lighting of the baptism~
al candles at Rome represented the survival of the old office of the

Lucernarium within the rite of that city's Churchi and 8ince it would
seem that the act~ if it took place, of kindling the new firQ at Roms
did not possess the aame liturgical importance that it did in the Gall=
ican and Spanish rites, it could be argued that reference to the source
of fire did not merit inclusion in any rubrics and was therefore omittede
From this it follows that there is no good reason to believe that the
flame for lighting the three large lamps was not obtained from newly-
2
kindled fire. The evidence of Prudentius in this matter, however,
should be treated with extreme caution. For not only does a period of
about 350 years stand between Prudentius and Zachary, we cannot be sure
that the lighting of the lamp for the Lucernarium was accomplished in
Spain in the same way as in Rome, assuming that, as a Spaniard, Prudent-
ius in his hymn was drawing on his experience of the Spanish form of
the service. 3
The final sentence of Zachary's letter provides weightier support

for the view that the three lamps were lit with newly-kindled fire.
Although we do not possess a copy of Boniface's letter and do not know
the precise form of the question which prompted Zachary in his reply to
mention the use of lenses, there can be little doubt that Boniface had
referred to their use in the kindling of the new fire for the Easter
4
vigil. In view of Zachary's denial of the Roman use of the lens it
1 Liber Cathemerinon V (Cunningham p.23) :

Incussu silicis lumina •••• monstras saxigeno semine, 'At a blow of
the flint you reveal the light with a rock-born spark.'
2 Zachary's silence concerning the source of the new fire is perhaps
comparable to the omission of reference to the water in a baptismal
font in the rubrics of service-books and manuals. Rarely is it
stated that the font should contain water; for its presence is
presupposed.
3 It is debatable whether in this instance a distinction should be
made between what Stevenson terms Vesper-light and Paschal-light
(~ Ceremonies p.178). For the Paschal themes from Exodus figure
prominently in Prudentius' Lucernarium hymn (11.37-88 and 89-104).
4 Boniface's letter appears to have contained a twofold request : for
information concerning the new fire procedure at Rome, and for papal
approval for the use of a lens in Christian ceremonial. The English-
man was almost certainly familiar with the method of kindling the
Paschal fire by friction - Ordo 26, which attests the practice, is
to be dated to the mid-eighth century - (continued !1 ~ foot 2f
~~~
would seem to follow that at Rome the Paschal fire was kindled by means
of a flint. Howeve~hia reply relates specifically to the use of a
lens as opposed to any other means of producing firei and his silence
concerning the use of a flint at Rome can equally be interpreted as in=
dicating that there was no production of new fire at Rome, but that the
lamps were lit from an already=existing source of fire.
In favour of the view, already refer.red~to above~ that the three

lamps were kindled with old fire is Zachary's silence regarding the
source of the fire. For in view of hia mention of the size and number
of the lamps, the source of the oil that fuelled them, the time at which
they were reserved, their location, and the care with which they were
tended, his failure to refer to the act of kindling anew the fire,
assuming that this act did occur, would seem somewhat surprising. A
more important consideration, however, centres around the interpreta-
tion of renovabitur, 'will be renewed', in the penultimate line of the
above-quoted passage. The verb obviously does not describe the kind-
ling of the fire either prior to or following its reservation since it
refers to the act of renewal which the already~burning flame undergoes
(~ candelis •••• sumptus ignis). It is true that renovare may be inter-
preted 'change' or 'alter' and that within the Roman context under dis-
cussion it would indicate that the newly-kindled and therefore uncon-
secrated fire in reservation was transformed by the Pope, or a priest
deputising for the Pope, on Holy Saturday into fire hallowed for liturg-
ical use. Howeve~it is difficult to see why Zachary did not write
benedicetur ('will be blessed') if he intended renovabitur to have the
sense of 'will be changed'. The usual interpretation of renovare is
'renew' or 'restore', and, given the circumstances in which the verb is
being used, it is unnecessary to translate the word otherwise. For it
is the view of the writer that the flame for kindling the three lamps
was taken from an already-existing source of consecrated fire and that,
since it symbolised the spirit of Jesus in this visual dramatisation
(Continuation 2f Note~~ !a! previous E!&!·)

and the guidance which he is seeking in the matter of the lens may
relate to the first occasion when that method of producing the new fire
for Easter was permitted by the Church. Interestingly, the use of a
lens for the new fire is allowed in ~~ which is thought to have had
its provenance in Mainz, the city of which Boniface was bishop.
Zachary's non-committalreply suggests a grudging approval. A policy of
toleration towards non-Roman ritual had characterised the pontificate
of St Gregory (Bede, !History 2f ~ En'lish Church~ People 1.27:
the reply to Augustine's second question •
175
of Eis death and burial 9 it was held to be 'dead 0 during the time of
its reservationo On Holy Saturday, when the fire was brought out of
its place of concealment 9 it was necessary to revivify the flame for
liturgical usc by means of a sacerdotal pronouncement of blessing. In
this way the old fire which had died was restored to life and thus
1
renewed.
The three lamps were lit during the Mas3 which commemorated the
Last Supper. Dendy, perhaps because of the superficial connection be=
tween the collecting of the oil for the lamps and the blessing of the
chrism on Maundy Thursday, states that the lamps were hidden during the
2
Mass of the Ho11 Oils. However, the fact that mention is made of the
consecration of the sacrum rather than the chrism would suggest that
Zachary is referring to the evening Mass of Maundy Thursday; and the
absence of Tenebrae on Maundy Thursday at Rome in the mid-eighth cent-
ury would indicate that in that city at that time the Triduum was held
to begin at the Mass of the Lord's Supper on Thursday evening,3 a time
more suitable for the lighting and reservation of the three large lamps
than one earlier in the day. Subsequently on Holy Saturday the reserved
fire was used to light the two Vigil-candles which were set close to
the font as it was being blessed- a fact alluded to by Zachary.
4
The Pope does not comment upon the reservation of three lamps as
opposed to one. It would be not unreasonable to believe that the simul-
taneous burning of three flames was designed to minimise greatly the
chances of losing the fire during this period. However, in the Gallican
tradition in which the reservation of fire also took place, only one
lamp was used; and since the flames were constantly attended by an
official of the Lateran Basilica, as we have already suggested, it is
clear that three lamps were lit not as a precautionary measure. A
symbolic significance would seem to be the most likely explanation for
the use of three lamps. Either they were lit in honour of the Trinity;
or more likely they symbolised the three-day repose of Jesus in the tomb.
The absence of illumination at the all-night vigil of Good Friday/
1 In Part IV Chapter 16 we will present additional corroborative

evidence for the sepulchral nature of the place in which the fire
was reserved.
2 ~ Y!! 2! Lights p.1JJ. According to the eighth-century Gregorian
3 Since the liturgical changes of 1955 this is now once more a feature
of the Roman rite.
4 For a description of the old Roman Vigil, see Part IV Chapter 16.
2 (cont.) Sacramentary there was only one Mass at Rome on MT.
176
Holy Saturday at Poitiers in the sixth century 9 and at Matins/Lauds of
1
Holy Saturday in Rome~ Milan 9 and parts of Gaul in the eighth century
leadsus to believe that in Rome before and including the time of Pope
Zachary no liturgical lights were lit on Good Friday at all. 2 This is
further borne out by the fact that the old fire was reserved during the
evening of Maundy Thursday and remained concealed until its renewal on
Holy Saturday. In any study of the Triduum at Rome in the early Middle
Ages, it is important to bear in mind that not all the ceremonies of the
last three days of Holy \'leek were held in the same church. In other
dioceses the ceremonies were necessarily confined to one building, at
least in the period of which we treat. At Rome the Good Friday cere-
monies viz. the Mass of the Catechumens, the Passion, the Solemn Prayers,
the Adoration, and the Mass of the Presanctified were held in the
Church of Sta Croce in Gerusalemme; whilst those of Maundy Thursday,
and the Vigil on Holy Saturday took place in the Cathedral of St John
Lateran. As a result no liturgical illumination was required in the
latter church on Good Friday since the Pope was officiating in Sta Croce.
The three lamps would therefore burn in the remote part of St John
Lateran from the evening of Maundy Thursday onwards, light not being
required for any purpose in the cathedral during the next forty-eight
hours. For the lamps used at the night office of Good Friday could be
lit from the flame of a candle used at the ~; and the night office
of Holy Saturday was sung in total darkness at this period, as we have
already noted.
(ii) ~ ninth-century evidence
(a) Theodore's evidence (AD 832)
In Chapter 1 we discussed Archdeacon Theodore's inconclusive evid-

ence relating to the provision of new fire at Rome during the Triduum. 3
·We observed that if, in addition to providing illumination at the night
office of Holy Saturday, the new fire of Good Friday was also reserved
to light the two Vigil-candles later that day, the ritual involving the
1 See Part I Chapter 1.

2 The absence of any liturgical light at Rome on Good Friday may be
assumed from the fact that the old fire was reserved in the three
large lamps during the whole of this day. Moreover, the Pope on that
day used unconsecrated fire for his honorific lights (pp.9-10).
3 (Pages 99-101). In ipsa die novus ignis accenditur de quo reservatur
usque ad nocturnale officium. 'On that same day (~Friday) new
fire is kindled and reserved for the night office. ' Amalarius ,
Liber de ordine antiphonarii XLIV.2.
177
reservation of the three large lamps would have lost its raison d'etre;
and presumably ~Gy would cease to have featured in the Roman rite. On
the other hand since the conversation between Amalariua and Theodoro
centred around the provision and the loss of light at the night offices
of the Triduum, and since the reserved fire could not be used for the
supply of illumination at the night office of Holy Saturday in view of
its unconsecrated state, there was no reason why Theodore should have
mentioned the three lamps~ assuming that they were burning all the while
in their state of reservation.
Although the evidence of Theodore by itself is inconclusive, the

1
statement by Amalarius that in the Roman Church all fire (totus ~)
was extinguished at midday and new fire was kindled at about the ninth
hour would imply that the old fire was no longer reserved in the three
large lamps, if totus here bears its all-inclusive meaning. However,
if, as we have suggested, the reserved fire was held to exist in a
state of limbo, as it were, it is unlikely that the fire considered to
be dead was included in the extinguishing of all the fires that took
place on Good Friday; and if the new fire was kindled during the after-
noon of that same day for the sole purpose of providing illumination at
the night office of Holy Saturday, we can but assume that the three
large lamps continued to be reserved as before.
In Chapter 5 we suggested2 that the kindling of the new fire at

Lyon behind the high altar may have been a feature of that church's
liturgy inherited from the old Roman rite which Leidrad of NUrnberg had
introduced at ~on in the early years of the ninth century. 3 We also
suggested on page 13 that the use of light at the night office of Holy
Saturday may have been introduced into the Roman rite during the ponti-
ficate of Leo III (AD 795-816). It would not be unreasonable to believe
that the kindling of the fire for that office became a part of Roman
ritual at the same time. Since Leidrad was responsible for the adoption
of the Roman rite by the Church of ~on, he presumably also introduced
the newrire ceremOQY which would Only recently have made its appearance
in Rome itself. From this it would follow that the new fire at Rome
was kindled behind the high altar of the Cathedral of St John Lateran,
in which church the night office of Holy Saturday was to be held.
1 For the full text of this statement, see above, p.100.

2 Section (ii) (b) p.149.
3 King, LPS p.11.
178
Some fifteen or twenty years after the date of Theodore 9 a testi-
monyi Pope Leo IV issued the following decree:
In sabbato Paschae extincto veteri novus ignis benedicatur, et per

populum dividatur, et aqua benedicta similitere 1
The uncertainty regarding the relationship of extincto veteri to the

rest of the first clause, and the absence of any reference to the kind=
ling of the new fire renders the first clause open to a number of diff-
erent interpretations. For instance the loss of fire may have taken
place and the new fire kindled on the previous day - the arrangement
familiar to Theodore - or the old fire may have been extinguished on
Good Friday and the new fire kindled shortly before it was blessed on
the fol.l.owing day. The most likely meaning, however, is that the loss
of fire, and the kindling and blessing of the new fire all took place
on Holy Saturday. If this interpretation is correct, it would indicate
that a significant development had occurred in the new fire ceremony of
the Roman Church between the archidiaconate of Theodore and the pontif=
icate of Leo IV.
In attempting to account for the change of day 2 on which the new

fire was kindled at Rome it is unnecessary to look for Milanese, Moz-
arabic, or direct Gallican influence. In view of Leo's keen interest
in liturgy it is not difficult to ascribe this development in Rome's
new fire ceremonial to the Pope. The reform survived the changes
brought about by the introduction into Rome of ~; for, as we have
already observed,3 the twelfth-century Pbntificale Romanum gave prior-
ity to the practice of kindling the new fire on Holy Saturday, reducing
E!!!'s performance of the ceremony on Maundy Thursday to an alternative.
There is no evidence from the middle of the ninth century to show

that the use of the Easter candle had been introduced into Rome by that
date. Attestation of its use in the Roman rite is first provided by
~ a century later. It would seem, therefore, that the new fire, kind-
led on Holy Saturday, provided the flame for lighting the two Vigil-
1 'On Holy Saturday let the old fire be extinguished and let the new
fire be blessed. Let it be distributed amongst the people in the
same way as the holy water.' PL 115.681/2. The exact year of the
decree is not known. Leo was Pope from AD 847 to AD 855.
2 That is, from Good Friday to Holy Saturday.
3 See above, page 106.
179
1
candles which presumably continued to form a feature of the Roman riteo
There is no firm evidence for the of a lens as the means of kind~
ling the new fire at Rome before the twelfth century. We have already
shown that the mid-eighth=century statement of Pope Zachary that 'we
have no tradition of using lenses 12 related to the production of fire
generally and not specifically to the ceremony for the kindling of the
Paschal fire; and, although in the evidence provided by Theodore and
Leo there is no indication of how the new fire was produced, Zachary's
denial of the use of lenses would suggest that the frictional method
was employed in ninth-century Rome. In assessing the evidence of PRG,
which became the service-book of the Roman Church, we must take into
account the fact that it is unknown which of the two traditions relat-
ing to the production of the new fire was adopted into the Roman rite. 3
The uncertainty is caused by the tradition, attested in Manuscript C,
in which the choice is offered between a flint and a lens for the kind-
ling of the new fire. This tradition vis-!-vis the practice of the
Roman Church invites the consideration of two possibilities. (i) Assum-
ing that prior to the introduction of ~ the new fire at Rome had been
kindled by means of a flint, did the existence of this choice of means,
attested in Manuscript C, so influence the practice of the Roman Church
that she abandoned the use of the fiint and opted for the lens? (ii)
Was the rubric, enjoining only the use of a flint, expanded to include
the use of a lens as an alternative in order to accommodate those
churches where the practice of using a lens was well-established? The
situation envisaged by the former suggestion is most unlikely. It re-
quires much more than the alternative directive of a rubric to change
radically a well-established practice. The latter possibility has more
to commend it. However, the uncertainty concerning the precise origin
of the double rubric, the fact that Manuscripts C and K relate to diff-
erent liturgical traditions, and doubt that a lens was used at Rome for
the new fire before the tenth century can lead in this instance only to
conclusions which are based on speculation.
1 For the use of these two candles, see Part IV Chapter 16.
2 PL 89.951. See also p.175.
3 PRG preserves two different traditions relating to the method by
Which the new fire was kindled. According to Manuscript K they ob-
tained fire 'excussum ~ lapide'; whilst Manuscript C records:
'excussum ~ silice vel christallo' (II p.94 I 342). The two
variants represent the practices of different liturgical milieux.
180
1
Rubric Date .Po_CJ!Plctnt
'L novus igniS.ooeXCUtiatur 12 c .PR XII
••• de cryatallo vel etiam 1214 Pontifical of
alio modo fiat. Apamea
2. ignem de cristallo vel c.1140 QrS.2 Ecclesiae
silice ••• noviter excussum La~eranensis
J. efficitur novus ignis a.1140 .Qr.gg Benedicti

4. ignis excutitur de 12 c Pl(C (..: Ordo X)
cristallo sive lapide
5· efficitur novus ignis £.1190 Ordo Albini
£_.1190 Ordo Cencii
6. ignis excuti tur de lapide 13 c Lateran Missal
7. ignis novus de cristallo £·1296 ~
vel silice excutitur
8. ••• de crystallo sive lapide .£·1310 ~ Caietani
9· extrahitur novus ignis de .£·1350 C.A.1760
cristalo sive lapide 1377 Bindo F.
10. extrahitur novue ignis de 14 c Ordo XV (= Ordo
crystallo vel alio lapide 1·Amelii)
11. efficitur novus ignis £·1451 Pontifical of
G.Barozzi
12. ignis excutitur de lapide 1474 Missale Rom' um
Table 3§. The Roman evidence (12 C- - 15 C ) for the

means by which the new fire was kindled.
On page 168 we suggested that,if in a new fire rubric a reference

to the lens preceded the mention of a flint, priority was to be given
to the production of fire b.f the refraction of the sun's rays. This is
well exemplified in the rubric of PR XII (Table 38, Rubric 1), which is
also the earliest evidence for the use of a lens at Rome. That this
method had priority at Rome is further intimated by the unspecified and
2
almost vague alternative which the rubric offere. ~~probably
implies the use of a stone; but the alternative device involving fric-
tion, viz. a wooden fire-drill, should not be completely ruled out. The
use of a lens is attested in eight documents which span a period of
some two centuries; and the absence of any reference to the means in
1 The documentary references are as follows : (1) I xxxii.1 p.2)8 and

DAER 4.24 p.160 (M 25). (2) ~ p.61. (3) ~ p.151. (4) II xliv.1
p.470 (= P.L 78.1014D). (5) Liber Ceneuum 1 p.296 and 2 p.130.
{6) Schmidt II 8 1o8 p.610.~III iv.1 p.587. {8) PL 78.1218A/B.
{9) ~ p.213 and p.276. (10) PL 78.13210. (11) ~p.373.
(12) HBS 17 p.1?4.
2 'Or it may also be done in any other way.'
181
Rubrics 3 9 5, and 11 (Table 38) leaves open the possibility that the
uso of the lena was allowed.
It might be argued that, although a lens may have been used at Rome
in the twelfth century, its use did not extend much beyond 1200, and
that mention was made of it in the fourteenth-century documents almost
as a matter of course, the rubric having assumed a set form of words
aud the device itself having by then become an anachronism. For it is
significant that the Lateran Missal makes reference only to the flint
(Rubric 6). An argument, however, based upon this assumption cannot be
sustained. To challenge the accuracy of these rubrics 1ia to call into
question the evidence of all other such rubrics which offer a choice of
action. Moreover, if the practice of using a translucent stone for the
new fire had fallen into desuetude, it is almost certain that any refer-
ence to the lens would have been omitted from the relevant rubric. The
apparent difficulty caused by the intrusion of the rubric from the Lat-
eran Missal can best be explained by the circumstances in which medieval
service-books were compiled. The compilation of a service-book, be it
~' pontifical, or missal, was the responsibility of an individual;
and the arrangement and content to some extent reflected the preference
and revealed the mind of its compiler. It is possible, therefore, that
the compiler of the iateran Missal omitted any reference to the lens
out of a personal preference for the use of the flint. 2
On page145 we referred to the influence of Cluniac practice upon

the Roman rite. In view of the prominence given to the beryl in the
new fire ceremo~ at that monastery (pp.161-2), it is not difficult to
believe that the Roman Church borrowed this liturgical feature from the
new fire ceremonial of Cl~; for it is known that Pope Gregory VII
and Pope Paschal II had close connections with that monaster,y.3 The
introduction of the lens into the new fire ceremo~ at Rome is probably
to be dated to the latter part of the eleventh century or to the early
years of the twelfth century.
1 That is, Rubrics 7, 8, 9, and 10 (Table 38).

2 It is just possible that 2! cristallo has disappeared from the
rubric in the transmission of the text.
3 The duration of their pontificates were respectively 1073 to 1o85,
and 1099 to 1118.
182
Chapter Eight
THE ANTECEDENTS OF THE NEW FIRE CEREMONY
The incorporation into the liturgy of Holy Saturday and the development of the new fire
ceremony were closely linked with the need for a supply of light at the Paschal vigil.
The antel:edents of this ceremony are to be found in two distinct liturgil:al traditions.
(1) According to the older tradition, which derives from the fourth-century
Jerusalem archetype, the flame, which provided the light at the Easter vigil, was
obtained from an existing source of fire which burned in a darkened place that repre-
sented the Sepulchre of Jesus. At Jerusalem a perpetual lamp burned within the Church
of the Anastasis at the very place where Jesus had been buried. A flame was taken from
this shrine in procession to the Church of the Martyrium, in which the Easter vigil at
Jerusalem was held. This two-church arrangement also obtained at Milan; and the use
of a darkened sacristy for the kindling of the new fire according to the Mozarabic rite is
almost certainly an attempt to reproduce conditions comparable to those found at
Jerusalem. 1 The use of the darkened west end of the nave of Salisbury Cathedral may
also ultimately derive from the Jerusalem-type setting for the production of the Paschal
light? It is possible that a similar arrangement obtained at Auxerre where the new fire
was kindled in one church and subsequently blessed in another. 3 However,there is some
uncertainty about the practice at Rouen where the new fire was kindled in the Church of
St-Etienne; for De Moleon informs us that the frre had previously been kindled in the
porch of the cathedral.4
(2) The Gallico-Germanic tradition was derived from the pre-Christian new fire
rituals of Northern Europe. Unlike the fue in the tradition which had its provenance in
Jerusalem, the flame for the provision of light at the Vigil was taken from newly-
kindled frre in a ceremony performed for that very purpose.
The study of the new fue rituals of pre-Christian Europe is beyond the scope of
this work:. 5 Suffice it to state that, in addition to the frres lit in honour of the sun or the
local deity, fue was also kindled for purificatory or for sympathetic reasons. The atti-
tude of the Church was at times hostile, if the use of frre posed a threat to the survival of
1
For the Milanese and Mozarabic rites, see Appendix 13.
2
Sarum Missal of ?1486 fol.lxxxiii.
3
Martene, DAER 4.24.3 p.145, M 39.
4
Voyages p.299.
5
J.G.Frazer's Golden Bough, researched in the earlier part of this century, still remains
the standard work on this aspect of pre-Christian European religion.
183
Christianity or was at odds with teaching of the Church. These were probably the
reasons for the denunciation of certain practices involving the use of frre at a synod held
c.AD 745 under the presidency of Boniface ofMainz. 1 Generally, however, the Church
seems to have been tolerant of pagan frre rituals especially if they could be transformed
and given a Christian orientation, or incorporated into the liturgy of the Church. To the
former category of rituals belong the Yule candles of England and Serbia, the bonfrres
lit on StJohn's Eve in Brittany2 and Spain,3 or those formerly lit in Belgium and North-
em France at the start of Lent. 4 Into the latter group may be placed the fonner lighting
of the candles at Candlemas in parts of the Cotswolds designed to strengthen the power
of the winter sun,5 and especially the lighting of frre on the eve of 1st May in Scot-
land,6 Ireland, 7 Wales and Scandinavia,8 and in other parts of Europe. 9 In view of the
need for the provision of newly-kindled frre for use at the Easter vigil, it would seem
that the production of frre for this pagan festival was transferred to an earlier, though
changeable, day before Easter. Thus the frres, formerly kindled in honour of the Celtic
sun-god Bel or Beal, 10 were henceforth to be lit in the worship of the Sun of Righteous-
ness.
The toleration afforded by the Church to pagan frre festivals and other allied rit-
uals depended on her control and monopoly of the use of frre. The Pythagorean belief
that frre was at the centre of the universe manifested itself in many of the religious
systems of Europe and the Middle East. It is found in the Rig Veda, in Zoroastrianism,
in the worship of Hephaestus amongst the Greeks and of Vesta at Rome, and amongst
the Celtic peoples of Northern Europe. In practical terms it often meant that the lighting
of a frre on a piece of grow1d entitled the kindler either to the possession of that ground
or to rights on that land. 11 It was 'a ritual proclamation of the ascendancy of the one
who lights it'. 12
1
Frazer, Golden Bough 10 p.270.
2
Le Braz p.lOl.
3
Herrera p.234.
4
Frazer, ibidem p.107.
5
Briggs p.l9.
~ossp.138.
7
Bury pp.104 ff.
8
Rees and Rees p.l93.
9
Frazer, ibidem p.159.
1
CXnown somewhat tautologically as 'Beltane frres'. The element tane signifies 'fire'.
11
Wade-Evans 16p.l0. The custom survived in parts of Wales up to the end of the
nineteenth century.
12
Rees and Rees p.157.
184
By adopting and incorporating this arumal ritual of kindling the new frre into her
own liturgy, the Church not only tacitly accepted the theological implications of this
understanding of frre, but appropriated to herself the rights and responsibilites that had
formerly belonged to those who had previously performed the new frre rites. The dis-
charge of this duty provided a means by which the Church was able to consolidate or
extend her authority both at a theological level and vis-a-vis each location in which
Christianity had been established. For the authority which she possessed related to the
places in which she had supervised the kindling of the new frre. This is clear from the
directive in the Pontifical of Poitiers which states that the new frre should be kindled
and blessed in the most recently-built church in the locality or in one at some distance
from the cathedral. 1 The performance of this ritual established an ontological rapport
between the church in which the new fire was kindled and the ground on which the
church stood. It also bestowed the benefits of God's protection on those who availed
themselves of the flre that had been kindled and blessed within the church.
The extinguishing of the old frre prior to the kindling of the new afforded the
Church an opportunity to reassert each year that authority which she possessed by virtue
of her role as administrator of that ritual, and to maintain a spiritual hegemony over the
lives of those who owed her their allegiance. For the conversion of communities to
Christianity did not result in the obliteration of existing pagan beliefs and rituals. These
age-old pre-Christian religious practices continued to thrive at times only just beneath
the surface of an outwardly Christian culture. The Church, therefore, did not discourage
the renewal of the frres, which were extinguished annually in the homes of the faithful,2
with a flame taken from the frre newly-kindled and hallowed by a Christian priest. For
in a sense the taking home of the new frre perpetuated the spiritual authority of the
Church and enabled the faithful to enjoy the benefit of a life-giving element upon which
God's blessing had been invoked.
Evidence for the taking of new frre to every home is plentiful throughout Western
Europe from the eighth century onwards. 3 The custom survived in parts of France well
into the present century.4
* * * * *
1
/n novissima sive forensi loci eclesia [sic]. Poitiers p.138.
2
Documentary evidence for this practice comes from Regensburg (Sacramentary p.l26),
Salzburg (1507 Missal folxcvii), and Auch (1836 Missal p.191).
3
0rdo 28.63; Ordo 32.21; Poitiers p.215; PRG ll p.99 §348; John of Avranches,
PL 147.49A.
4
Van Gennep p.1257.
185
Grancolas' statement that the new fue was not kindled in some churches but was
taken from another place' relates to churches which obtained their new fire from a cen-
tral supply either from the cathedral or from a nearby monastery. Although he himself
does not give any instance of this procedure, we know from another source that the new
fue in the churches of Evreux and its suburbs was obtained from Evreux Cathedial.2
* * * * *
The Symbolism of the Old and The New Fire
The extinguishing of all illumination during the afternoon liturgy of Good Friday
was, it would seem, a practice originally confmed to the Roman Church. Amalarius is
the earliest writer to record the absence of light during this period and the symbolic
interpretation attached to this Roman custom. For according to him the absence of light
during the Solemn Prayers and Veneration both commemorated and symbolised the
period from the sixth to the ninth hour on Good Friday when 'there was darkness over
all the earth'. 3 Moreover those three hours of darkness were held to foreshadow the
three days and nights when 'the creator and producer of light would cover himself with
darkness in the tomb' .4 Honorius of Autun also draws the same analogy between the
liturgical darkness of the Passion and the solar eclipse that took place during the Cruci-
ftxion.5
Commenting on the extinguishing of all fues prior to the start of the Paschal vigil
on Holy Saturday, Sicardus likens the old fue to the Law of the Old Covenant, now ful-
filled and superseded by Christ.6 John Beleth records a similar interpretation of the old
fue. 7
In a general sense the new fue is seen to represent different aspects of God's
nature and gifts. For John of Avranches it symbolises the lux deitatis which remained
concealed during the earthly life of our Lord, but which was revealed mystically to the
Church and which. shone forth in the hearts of the faithful during the Passion and after
. .
'Non eliciebatur sed aliunde surnebatur. Commentarius p.316.

2
1740 Missal p.187.
3
Liber de ord. ant. XLN.6.
4
Ut praedicaret orbi tribus diebus et tribus noctibus creatorem et operatorem obscuratu-
rurn se in sepulchro, (ibidem).
5
Gemma Animae, PL l72.667C.
6
Mitrale, PL 213.3220.
7
Rationale, PL 202.111B.
186
the Resurrection. 1 In the Speculum Ecclesiae Honorius of Autun likens the new fire to
the Holy Spirit who illuminates the souls of all the faithful; but in the Gemma Animae
he compares the new fire with the new Christian teaching. 2 Robert Paululus, echoing
Honorius, suggests that the new ftre also represents the new grace that results from the
Resurrection. 3 In his commentary on the ceremonies of the Roman Church,4 Philippo
Zazzera in more recent times compared the new fire with the life-sustaining flame with
which Jesus is continuously kindling us, a notion expressed some centuries earlier by
Bianco da Siena in his hymn Discendi Amor Santo,5 and also in a nineteenth-century
commentary from Langres. 6
In our discussion of the symbolic intetpretation of the kindling of ftre from flint
(p.160), we suggested that in view.ofthe medieval ignorance as to its physical nature,
fire was regarded as being essentially one with that aspect of God's nature which was
manifest in the Burning Bush and at the Transfiguration, though present and visible in
the world as perceived by mankind in an allotropic form, as it were. This is in contrast
with the understanding of fire purely as a symbol of life and lacking any essential
relationship with the divine nature, such as was held by the Gipsies and other peoples. 7
Using Johannine theological concepts, Durandus describes the new fire as the 'unfailing
light' of God,8 which comes into the world and illuminates our hearts and senses. It
brings us from darkness to light and eternal life.
1
Lib.de Off.Eccl., PL 147.49A.
2
PL 172.928C and PL 172.668A, respectively.
3
PL 172.452.
4
SS.Ecclesiae Rituum p.301.
5
Rendered by R.F.Littledale into English as the hymn Come down, 0 Love Divine.
Unfortunately Zazzera cites Luke 12:49 ('I came to cast fire upon the earth'). In this
verse fire almost certainly signifies the eschatological fire of judgement.
6
'(Le) feu nouveau, image de la lumiere et de la charite que nous re~evons par
Notre-Seigneur Jesus-Christ.' 1844 Langres Directory p.53.
7
Rao p.161.
8
Rationale VI.80 p.350. This phrase occurs in the B-group of the Benediction-fonnulas
(Appendix 5). Used in the vocative case, it is both a title and an attribute of God.
187
PART ill
The New Fire Procession

Chapter One
THE NEW FIRE AND THE PROCESSION
(This section describes the new ft.re procession on Holy Saturday. In places where
the new ft.re was kindled on Maundy Thursday and Good Friday the processions on
those two days were identical except that they did not cuhninate in the lighting and
blessing of the Easter candle.)
There can be little doubt that the new ft.re procession was a feature of pre-Christian
ritual which was adopted into the liturgy of the Christian Church. The necessity of
transferring the new ft.re from its place of reservation on Maundy Thursday, or place of
kindling on Holy Saturday, also perpetuated a primitive element of the Lucernarium
viz. the bringing in of the lamp, and within its revitalised Christian milieu resulted in the
emergence of an elaborate ceremonial which invested the procession with its own dis-
tinctive character. From the evidence of the early ordines we fmd that two main tradi-
tions existed within the Romano-Gallican Church : (i) that in which a small candle, lit
with the new fire, was borne in procession for the lighting of the Easter candle in
church; and (ii) that in which the Easter candle itself was carried in procession, having
previously been kindled with the new fire.
(i) The Gelasian sacramentaries do not state how the ft.re was conveyed to the
Easter candle. It is likely, however, that a small lighted candle was carried from the
place of reservation. 1 The practice of lighting the small candle from the new or reserved
frre is first attested in the eighth-century Ordo 26.2 It is also found in the ninth-century
Ordo 293 and Pontifical of Poitiers,4 and in the Regularis Concordia of the tenth cen-
tury .5 Thereafter this became the practice of the vast majority of churches in both the
cathedral and monastic traditions. Information about the size of the small candle is
meagre. However, it is recorded that at Canterbury the candle should weigh half a
pound, and that it should not previously have been used. 6 In some places the upper
candle of the Tenebrae hearse was used. 7
1
1f the ft.re was reserved beneath the altar, a small taper would still have been required to
transfer the flame.
2
0R 26.14, by inference.
3
0R 29.59.
4
Poitiers p.138.
5
PL 137.349B.
6
HBS 23 p.380.
7
Feasey p.l89.
189
After the candle had been lit from the new frre, it was common practice to light
another small taper and to place it in a lantern. 1 Gavanti: tells us that this was done in
case a high wind should extinguish the processional candle;2 and this is confrrmed by
Lanfranc and the Gilbertine Ordinal, and by evidence from St Lo and Nidaros. At St
Paul's, Rome, at Besan~on, and at Nidaros the lantern was carried by a boy; but the
Decrees of Lanfranc, implemented at St Vedast 's Abbey and at Durham, specify the
magister puerorum. 3
The processional candle was carried into church raised aloft. 4 A symbolic inter-
pretation was subsequently attached to the raising of the light; but our sources give no
indication as to why the candle was originally held on high. There are a number of
possible reasons. (i) It was a feature of a corresponding pre-Christian ritual. (ii) It was
done so that the fire would be visible to all those participating in the ceremony. (iii) It
was raised as a gesture of thankfulness that the minor miracle of producing fire had
been performed. (iv) The bringing of light into the gloom of the church symbolised
Christ's leaving the darkness of His tomb, as Van Doren observed.5 The candle raised
on high would have visually expressed the notion of His triumph and symbolised His
victory over the forces of darkness.
A number of different devices were used for supporting the candle.
1. The Pole. The use of the pole6 is enjoined in Lanfranc 's Decrees and in most of the
early surviving monastic customaries. In more recent times a pole continued to be used
at Amiens and Le Mans, and most likely in other churches which clung to their Gallican
traditions. At Cambrai the pole was painted red, in sympathy perhaps with the colour of
the fire. 7 (Table 39).
2. The Reed. In respect of the surviving documentary evidence the use of a read ante-
dates that of a pole. It figures in a number of early ordinei and is found in some other
early sources.9 Some of the later medieval Roman service-books 10 offer a choice
1
0r sconsa, 'a screened light', as at Fleury and Besan~on.
2
Vol. I p.234.
3
For the references in this paragraph, see Tables 39-41.
4
It was customary for this candle to be borne by the officiating deacon. Maigne-d 'Amis
ascribes the duty to a subdeacon (Lexicon VII p.438).
5
La ceremonie p.78.
6
Baculus,pertica, virga: all three terms are found.
7
Compare the phrase rutilans ignis from the Romano-Gallican Preface.
8
0R 26.4; OR 28.26; OR 29.15; OR 31.29. The latter ordo records its use only on
Maundy Thursday; but it was presumably used on the other two days of the Triduum.
9
Poitiers, PRG, andAlcuin.
10
0rdo XIV, Ordo XV, CA 1706, and Bindo F.
190
Church/Document Date Comment Source
Fleury 10C d Albers V p.143
Lanfranc c.1070 abc Decrees PL 150.467B
Besan~on llC a DAMR 3.13 p.127 (M 56)
Cluny llC d PL 149.659A
Farfa llC acd Albers I p.48
Fruttuaria llC acd Albers IV p.54
Gembloux llC acd Albers n p.93
St-Benigne, Dijon llC DAMR 3.13.34p.126(M 1150)
St Paul's, Rome llC c DAER 3.22 p.124 (M 1184)
Nidaros c.1210 ONE p.232
Haymo (OM) 1243 b HBS 85 p.209
StDenys c.1273 DAMR 3.13.34 p.126 (M 1158)
PGD c.1296 m xxxii.7 p.239
Reims 14C a DAER 22.5.2 p.97 (M 261)
Missale Romanum 1474 b HBS 17p.175
Wurzburg 1477 Ordinary np
Toulouse 1490 b DAER 4.23.6 p.127 (M 311)
Austin Friars 1491 Missal np
Melk: 1495 b Missal fol.lxviii
Rouen 1640 b Ritual p.305
Cambrai 1699 Gav./Mer. IV p.161
Amiens 1752 Missal p.182
LeMans 1789 Ceremonial p.127
Bayeux 1790 b Missal p.168
a Reference to the use of a small candle.

b Offers choice of device.
c Spare candle in lantern.
d Evidence for Maundy Thursday only. '
Table 39. The use of a pole for the new frre.
between a reed and a pole; others, 1 which relate more closely to the ceremonial of the
papal court, specify 'reed' alone. Similarly the influential Pian Missal of 1570 specifies
the sole use of a reed.
The length of the ree<f was traditionally three and a half cubits or ten palms,3
both measurements being approximately the height of a man. More recent manuals
have recommended a measurement of about five feet. 4 Some liturgical comrnentators5
insisted that the reed should not be a pole or a rod; nor should it be an imitation reed. 6
For symbolically it was important. According to Rupert of Deutz7 it represented the
1
Haymo's Ordo Missalis, PGD, and MR 1474.
2
0r baculus ( 1491 Missal of Augustinian Friars np ).
3
Desideri p.150; Bisso I p.79. The former measurement is first recorded in Ordo XIV
10rdo Caietani), PL 78.1218B.
For instance, Fortescue and O,Connell (6th edition) p.
5
Gavantus/Merati IV p.155; Desideri p.149.
6
Loan p.283. He admitted that in practice a thin pole 5' long was often used.
1
De Div.Off. V, PL 170.169C
191
reed which the soldiers gave to Jesus after his trial (Mt 27:29). Bisso, followed by
Desideri, also links the reed with the Passion; 1 but the symbolism is forced. Possibly
alluding to the use of the serpent-rod or the serpent-candle, which had all but disap-
peared from the ceremonial of Holy Saturday by his day, Bisso claimed that the reed
signified the Passion of Christ, and that, just as a reed is used to kill serpents, so the
Passion of Christ destroys the Devil.
It became customary in some churches to decorate the reed with flowers, thus par-
alleling the ornamentation of the Easter candle. The earliest references to the practice
come from two writers in the middle of the seventeenth century .2 Both state that some
of the actual reed should be visible, the latter adding that this is done 'because it (the
reed) is not devoid of symbolism'.3 Van Gennep also wrote that flowers were attached
to the upper portion of the reed (1.3 p.1257). The importance ofleaving part of the reed
free of flowers is mentioned in two eighteenth-century ceremonials,4 and by other com-
mentators such as Merati and Desideri.5 The latter also mentions its being decorated
with other ornaments. At Constance the reed was in fact a pole, at the end of which was
the effigy of a deacon bent backwards; and above its head were the words 'Here (is
placed) the twisted candle'.6
It might be argued that the reed was a development of the pole, suggested by the
scriptural precedent of Matthew 27:29. Caution, however, is recommended here for two
reasons. The evidence for the reed antedates that for the pole by some 250 years; and
there is only a tenuous link between the reed of the New Testament and that used for
conveying the new fire. In all likelihood the reed and the pole derive from different
liturgical milieux. Moreover it is tempting to see in the use of the reed an echo of the
myth of Prometheus' theft of fire from heaven; 7 or at least a Gallican version of the
myth whose re-enactment in a pre-Christian religious context the Church incorporated
into her own ceremonial. Corroborative evidence, however, to commn direct continuity
is lacking; but the pagan ancestry of the ritual seems almost certain.
3. The Spear. The use of a spear for the transportation of the candle is trrst attested in
the tenth-century Regularis Concordia; and the fact that it also appears in Lanfranc's
Decrees as an alternative to a pole may suggest that the spear was a development or
1
Hierurgia I p.78 and Praxis p.l49, respectively.
2
Corsetti (1656) p.316 and De Bralion (1657) p.247. Colti (1772) adds that it may also
be adorned with gold (ll p.156).
3
ld enim mysterio non caret (ibidem).
4
Augustinian Friars (1714) p.307 and Capuchins (1775) p.128.
5
Gavantus/Merati IV p.155 and Praxis p.150, respectively.
6
Hic Cereus torquatus. Ceremonial of Constance cited in Gav./Mer. IV p.161.
7
lnterestingly, the plant fennel, in which the fire was stolen, grows to a height of 5 feet.
192
Document Date Comment Source
Ordo 26 750-75 a OR26.6
Ordo 28 c.800 OR 28.26
Ordo 29 870-90 a OR 29.15
Poitiers c.900 a Pontifical p.215
PRG c.950 a Vol. 2 §220 p.58
Alcuin c.lOOO a Lib.de Div Off PL 101.1205C
Beneventum llC a Odermatt p.273
V allombrosa llC Albers N p.249
CMG llC d Albers V p.32
Rupert c.llll De Div.OffV, PL 170.169C/D
OrdoXII c.1190 PL 78.1076C
OrdoAlbini c.1190 Liber Censuum II p.130
M. Cassino 12C Monte Cassino A (PR XII I p.292)
PRXII 12C l.xxxii.7 p.239
Haymo(OM) 1243 b HBS 85p.209
Marseilles 13C b /LEMp.84
OrdoXIV c.1310 PL 78.1218B
CA 1706 c.1350 ZRKMp.213
BindoF. 1377 ZRKMp.274
Aries 14C d DAER 4.22 p.117 (M 30)
OrdoXV 14C PL 78.1321C
MR 1474 1474 b HBS 17 p.175
Melk 1495 b Missal fol.lxvili
Salzburg 1507 ad Missal fol.lxxxvi
Aquileia 1519 Missal p.91
Cosenza 1549 Missal fol.115
Camaldolese 1634 a Ceremonial p.82 and p.84
StLo c.1700 ac De Moleon p.403
Austin Friars 1714 a Missal p.307 and p.309
Bayeux 1790 Missal p.168
Nantes 1837 Missal p.199
a Reference to the use of a small candle.

d Evidence for Maundy Thursday only.
Table 40. The use of a reed for the new ftre.
refmement of the latter - unlike the reed which we maintain was inherited from
pre-Christian ritual. Its use may have been suggested by the biblical precedent of the
soldier's lance of John 19:34 in the same way that the use of the reed recalled Matthew
27:48. Alternatively, the spear with its sharp point provided a suitable instrument for
affixing the candle, and, having become a feature of the new ftre ceremonial, was
subsequently endowed with a symbolic interpretation. John of Avranches commented
1
that the spear recalled the Crucifixion.
'Lib.de Off.Eccl.,PL 147.49A: Christus in cruce suspensus.
193
The surviving evidence would suggest that in addition to the houses of the Benedictine
order the spear was used mainly in England and Northern France. In the romanised cer-
emonial at Palencia the two small candles, which were lit from the new fire, were
placed on small spears (hastuli). 1 Somewhat surprisingly, the twelfth-century Roman
Ordo Albini prescribes the use of a spear; other Roman ordines stipulate a reed.
Church/Document Date Comment Source

Regularis Concordia c.970 a PL 137.491B
John of Avranches c.1070 d PL 147.49A
Lanfranc c.1070 abc Decrees, PL 150.467B
OrdoAlbini c.1190 Liber Censuum IT p.l30.
Gilbertines 12C c HBS 59p.39
Norwich c.1265 acd HBS 82p.81
Canterbury 13 c HBS 28p.274
Hereford 13 c Missalp.97
St Vedast, Arras c.1300 ac HBS 86p.160
Westminster 1370 c HBS 5 col574
Durham 14C ac Missal p.l85
Tongres 15C Ordinary p.164
York 14C Missal p.109
St Mary's, York c.1400 HBS 75p.275
Barking 1404 HBS 65 p.101
Salisbury 1502 a Processional p.84
Palencia 1568 e Missal fol.c
Rouen 1640 Ritual p.305
Chalons 1748 Missal p.178
Bayeux 1790 Missal p.168
a Reference to small candle.

d Evidence for Maundy Thursday only.
e Use of two small spears.
Table 41 The use of a spear for the new fire.
4. The arundo serpentina. This device is discussed in Chapter 2.
1
1568 Missale Pallantinum fol.c.
194
Chapter Two
THEARUNDO SERPENTINA
(i) Its description and use
The arundo serpentina was a reed or, more likely, a pole, the upper part of which
tenninated in the effigy or representation of a serpent. 1 Three variations of the device
are known. (i) Either the end of the wooden pole was carved in the likeness of a
serpent, or a graven image of this creature was attached to the end of the pole.2 At
Braga the arundo was a bronze winged dragon with three candles issuing from its
mouth. 3 Elsewhere the candle, which was inserted into the serpent's mouth, tenninated
in a triple ramification. 4 The use of the serpent-reed is attested in the seventeenth
centlli)? and also in the eighteenth century. De Moleon refers to the arundo used at
Rouen Cathedral at the beginning of the latter century; but records that the carving of
the serpent had disappeared.6 Its use was still permitted at Bayeux at the end of the
same century. 7 (ii) The arundo serpentina was also a reed or pole to the end of which
was affixed a candle twisted to resemble a snake. Examples of this type appear on some
of the Exultet rolls of Southern Italy, and show the candle either protruding from a spike
or socket at the end of the pole, or entwined around the upper section of the shaft. 8 (iii)
The tenth-century Regularis Concordia attests the combination of the two variants
described above in the same device:
ferentes hastam cum imagine serpentis ...et...candela, quae more serpentis inftxa
est.9
The combination of serpent-reed and serpent-candle was also familiar to Durandus:
1
In addition to serpens it was also known as coluber (Poitiers p.215), and draco (1790
Missal of Bayeux).
~e latter type is well exemplified in the woodcut illustration in the 1502 Sarum
Processional (p.75). Also in Wordsworth, Ceremonies and Processions, p.84.
3
King, LPS p.224. At Worcester also the serpent held three candles (Antiphonary p.69).
4
For instance,in the Roman rite. See also Table 43.
5
Feasey (The Paschal Preconiurn p.259) refers to an illustration of a boy, dressed as an
angel with wings, lighting the Easter candle with a wax serpent twined about a rod, in
Le Tableau de Ia Croix represente dans les ceremonies de Ia Sainte Messe, printed by
F.Mazot in 1653.
6
Voyages Liturgiques p.304.
7
1790 Missal ofBayeux p.168.
8
A very, Plates LXXII, CXX, and CXXXIII.
9
'Bearing a spear with an effigy of a serpent. .. and ... a candle, which is inserted to
resemble a snake.' PL 137.491B.
195
In some churches also during these seven days (Easter Week), when they go to the
font for Baptism, the effigy of a serpent placed on a pole leads the procession. A
twisted candle, lit with the new fire, is fixed on the head of the serpent. With this
the Easter candle and all the other lights of the church are.lit. 1
and is attested at Toulouse in the late fifteenth century.2 A triple twisted candle emerg-
ing from the serpent's mouth is found at Braga in the mid-sixteenth century. At Bayeux
in the thirteenth century the serpent held an unspecified number of twisted candles in its
mouth : habeat draco in ore candelas plures retortas.
The use of the serpent was not confmed to the lighting of the Easter candle. We
have already noted in the above-quoted excerpt from the Rationale that Durandus men-
tions the lighting of the lamps in some churches by means of the arundo - a practice
attested by evidence from Vallombrosa. 3 Moreover the serpent was used at other
ceremonies and on other days. Its use at the blessing of the font and at Baptism is not
only attested by Durandus in the same above-quoted passage from the Rationale, but is
also found in the rite of Braga and in the revised Mozarabic rite. 4 It featured in the new
1ue processions on Maundy Thursday and Good Friday at Auch;5 and at Bayonne the
serpent-candle was lit for the reading of StJohn's Passion on Good Friday, perhaps as a
liturgical allusion to Jesus' reference to the raising of the serpent in the same gospel (Jn
3:14).6 At Rouen a winged dragon was borne in procession on Ascension Day by a
verger in a purple robe, and placed at the feet of the Blessed Virgin Mary. 7
(ii) The origin of the arundo serpentina
The origin of the serpent-reed or serpent-candle should not be sought in the

emergence and development of the triple candle; rather, it was the triple candle which
developed from the serpent-candle, as we shall show. In fact the earliest reference to
the serpent-candle is to found in the ninth-century Pontifical ofPoitiers (p.215); the
earliest ordines are silent concerning the shape of the candle. The presence of liturgical
1
ln quibusdam etiam ecclesiis, in his septem diebus, quando descenditur ad fontes, ante-
fertur quidam serpens imaginarius, super virgam; et candela novo lurnine accensa,
super caput serpentis retorta affigitur, ex qua cereus paschalis et omnes aliae ecclesiae
accenduntur. Rationale VI.89 p.377.
2
Deferentes virgam sculptam in figuram serpentis... Nam ex igne novo accendebatur cer-
eus in modum serpentis efformatus. Martene, DAER 4.23.6 p.l27 (M 311).
3
1503 Missal fol.xcv.
4
King, LPS p.224 and Missale Mixtum, PL 85.470A, respectively.
5
1491 Missal, cited by Feasey, The Paschal preconium p.259.
6
1543 Missal p.42. It was put to one side after the reading and taken to the sacristy at
the end of the liturgy.
1
Guide de Ia France Mysterieuse p.814.
196
Document Date Comment Source
Poitiers c.900 c Pontifical p.215
Reg. Cone. c.970 CR PL 137.491B
St Vito, Verdun lOC R Albers Vp.l22
John/Avranches c.1070 c Lib.de Off.Eccl. PL 147.49A
CMG llC c Albers V p.32
Fleury llC R Albers V p.143
Rupert c.llll R De Div.Off.V PL 170.169C/D
Corbie 12C DAMR 3.13.34 p.126 (M 1146)
PGD c.1296 R ID.iv.8 p.588
Bayeux l3C Ordinary p.135
Hereford l3C R · Missal p.97
Marseille 13C /LEM.p.84
Worcester 13C Antiphonary p.69
Strasbourg 1364 c DAER 4.24 p.162 (M 35)
Westminster c.1370 R Missal HBS 5 col.574 ·
Aries 14C c DAER 4.22 p.117 (M 30)
StMary's, York c.l400 R HBS 75p.275
Toulouse 1490 CR DAER 4.23.6 p.127 (M 331)
Auch 1491 Missal, cited by Feasey, The
Paschal Preconium p.259
Uzes 1495 R Missal fol.lxiii
Salisbury 1502 R Processional p.75
Spires 1512 c Agenda fol.xciii
Coutances 1557 R Missal fol.lxvii
Braga 1558 Missal fol.xcvi
Austin Canons 1579 Ordinary fol.137
Rouen c.1700 R De Moleon p.304
Bayeux 1790 R Missal p.168
c Use of serpent-candle.
R Use of serpent-reed.
CR Use of serpent-candle and serpent reed in combination.
Table 42. Evidence for the use of the arundo serpentina.
features and ritual elements of probable pagan provenance in this pontifical 1 and in
ceremonies described in other documents2suggest a pre-Christian antecedent for the
setpent-candle also. For there can be little doubt that the ceremony of the new f.tre pre-
dates the arrival of Christianity.
If the use of the setpent-candle derives from a pre-Christian religious milieu -the
importance of the snake in northern European worship is widely recognised - it is diff.t-
cult to see how it can represent the malign aspects of that creature within the context of
the new frre. The fire-breathing setpents and dragons, inherited from pagan folklore,
1
For instance, the three apotropaic weather-candles (§406), and the carved wooden
model of a turret city (§407), both on p.216. .. ·
~e circumambulation of the new frre at Breslau and Wurzburg (p.114 ), and the noise
at the conclusion of Tenebrae.
197
appear within the context of Christian theology as creatures symbolising vice or evil or
paganism itself. Numerous are the instances of hermits, bishops, and saints who in
times past had done battle with and vanquished such monsters. 1 One should not try to
see a link between this type of setpent and that which bore the new fire. Nor should one
try to fmd in the new-fire setpent an echo of an otherwise-irrecoverable myth relating to
the conquest of the powers of darkness by the superior strength of the deities of heaven
represented by fire and light.
On the other hand the fire-setpent may have been a tangible relic of the belief that
this being symbolised the force of power and life, which was visually represented
amongst the ancient Greeks by the caduceus. Again, it has been pointed out that in a
caduceus-like device the setpents may stand for the past and the future, while the wand
represents eternity.2 The weakness of both these theories is that no account is taken of
the presence and use of fire; and since the ceremony under study revolves around the
production of ftre and its transportation, it is safer to look for the origin of the fire-setp-
ent elsewhere. H its origin is to be sought in mythology, a more plausible explanation
would be to see in the frre-setpent a visual representation and re-enactment of the myth
found in many parts of the world, according to which the thief of the frre stolen from the
gods was a bird or a beast. 3 Interestingly, a woodcut illustration in the Sarum Pro-
cessional of 1508 depicts the head of an animal, almost certainly a boar, with a candle
protruding from its mouth, as the termination of the arundo serpentina (p.75). However
the lateness of the drawing, together with allowance made for artistic licence, the
uniqueness of the creature, and the otherwise-universal use of a snake for the transporta-
tion of the new fire seem to rule out any direct connection between the Sarum device
and the above-mentioned myth.
A propos of this myth it is appropriate to mention at this point those instances in

which the Easter candle was lit by means of a dove. In an Exultet roll of the early
twelfth century from Monte Cassino there is an illustration of a dove lighting the Easter
candle.4 This may be the artist's own graphic means of expressing the symbolism sug-
gested by the account of Jesus' baptism in the Jordan. On the other hand the dove may
represent the actual device used for the kindling - a sculptured bird atop a pole, or even
a contrivance which enabled a metal or ceramic dove to be lowered from the ceiling or
1
For instance, StRomain (Rouen), St Vigor (Bayeux), St Nicaise (Vaux), StJulien and
St Leon (Le Mans), St Bienheure (Vendome), St Clement (Metz), St Martial (Bor-
deaux), St Martha (Tarascon), St Florent (Seaumur).
2
Varley p.126.
3
Frazer, Apollodorus IT Appendix m p.327.
4
Avery Plate LXII.
198
to be swung in a lateral movement to the wick of the candle. 1 The former device was
used to light the Easter candle at Tongres in the fifteenth century; a candle was placed
in the dove's mouth.2 A similar appliance may have been in use at York in the four-
teenth century. For we read in an inventory :
inveniet cereum paschalem et omnia ad eum pertinencia tam in coloribus, floribus

et cordis quam in aliis pertinentibus ad columbam. 3
It is difficult to see what else columbam may refer to.
We have previously referred to the likely scripturally-inspired origin of the reed

and of the spear (p.193). In attempting to discover a biblical antecedent for the
fire-serpent we may with confidence dismiss the suggestion that it was inspired either
by the Cherubim of Genesis 3:24 or by the Seraphim of Isaiah 6:2, since serpentine
characteristics were attributed to neither beings; although frre was associated attribut-
ively with the former and circumstantially with the latter. Moreover allusions to these
creatures seem incongruous within the context of the new frre. Both John of Avranches
and Rupert of Deutz state that the fire-serpent recalls the fiery serpent which Moses set
upon a pole (Numbers 21:8-9). The former alludes4 to John 3:14 and, just as Jesus com-
pared himself to the bronze serpent which Moses set up in the desert, he symbolically
identifies Christ with the frre-serpent used in church. However, he does not observe
that, just as the uplifting of Moses' serpent and of Christ brought salvation, so fire in a
sense brings life to men. The identification of Christ with the fire-serpent represents an
instance of secondary or expository symbolism which characterised medieval interpreta-
tion of the liturgy, and which to the mind of modem man often seems inapposite and
forced.
Rupert of Deutz offers5 a mystical and more explicitly allegorical interpretation of

the fire-serpent, and at the same time presents a symbolic interpretation of both the reed
and the rod, and links both in a secondary comparison. The reed, which represents that
which the soldiers gave Jesus after his trial, was foreshadowed by the rod which Moses
1
The artificial descent of fire at Easter was not unknown in the East. At the beginning
of the eleventh century the Christian writer Abelfaragius records the allegation that at
Jerusalem the iron chain, which held the lamp above the Holy Sepulchre, was probably
greased with oil of balsam and ignited from the roof. Masudi, a Moslem historian, had
previously alluded to the production of fire by a clever device. Goodrich-Freer
f.P·107-8.
Ordinary pp.164-5.
3
'He will find the Easter candle and all its accessories,viz. paints, flowers, and ribbons,
as well as the accoutrements for the dove.' Bradshaw and Wordsworth IT p.98.
4
Lib.de Off.Eccl., PL 147.49A.
5
De Div.Off. V, PL 170.169C/D.
199
turned into a snake; so that, in the same way that the rod was transformed into a snake
and then became a rod again, so Jesus (prefigured by the rod) lives, will die, and then
will transform the deaths of sinners by rising to life again.
It is possible that Rupert, though writing some decades later than John of
Avranches, preserves an older tradition which relates, not to the origin of the serpent-
candle, but to the reason for its adoption within the Paschal liturgy. As we have noted
elsewhere, 1 the theme of the entire Paschal liturgy is rooted in the Old Testament types
and prefigurements of Exodus and Numbers : from the allusions and references to
Moses and the Burning Bush in the formulas for the blessing of the fire at the start of
the liturgy, to the ceremony of Baptism which recalls the entry of the Israelites, God's
chosen people, into the promised land. The writer is therefore of the opinion that, in
view of the Mosaic motif running through the Paschal liturgy, the rod on which the new
fire was borne into church was seen to be foreshadowed in the rod which Moses used at
the court of Pharaoh; and that the image of the serpent, borrowed possibly from a corre-
sponding pre-Christian new frre ritual, became attached to the end of a pole to com-
memorate and to portray vividly the transformation of the rod as described in Exodus
4:3. It is significant that just as the rod was turned into a serpent shortly after the
beginning of Moses' mission, so the serpent-rod was used in the early stages of the new
frre ceremony. The writer also believes that the development of the serpentine candle
emerging from the serpent's mouth was a visual portrayal and liturgical re-enactment of
the swallowing of the snakes by Aaron's rod in Exodus 7: 12.2
'See Part II Chapter 3 (ii).

2
A copper-gilt and enamel cross from the Meuse region, now in the British Museum,
shows Moses and Aaron flanking a brazen serpent on top of a column. It dates from the
third quarter of the twelfth century.
200
Chapter Tirree
THE TRIPLE CANDLE
(i) Description and construction
The use of the triple candle is :ft.rst attested in the Pontificale Romanum of the twelfth
century .1 It refers to triplicem candelam coniunctam, 'a candle twisted into three
branches', a device found in nearly all subsequent Roman documents up to and includ-
ing the Pian Missal of 1570, as well as in the rites of some religious orders and of some
churches outside Italy? Corsetti refers3 to the triple candle as 'in calce unum'; England,
describing the papal ceremony of the last century, mentions4 that the three candles 'part
from a common stock'; and Van Gennep, writing in the earlier part of the present cen-
tury, refers to the candle with three branches.5 With the eventual adoption of the
Roman rite by most of the churches in the West, the use of the triple candle became
almost universal. However, up to the liturgical changes of 1955 a single candle was still
used for bearing the new fire in a number of French dioceses6 whose rites still preserved
features of their traditional ceremonial.
It would appear that elsewhere three separate candles were arranged at the end of
the reed or pole in a triangular formation. This is the arrangement prescribed in the
Roman Missal of 1474. The Ritual of Evesham refers to three cereoli; and the mention
of a three-branched candlestick, as opposed to single candle, at Aquileia, Lyon, Vallom-
brosa, and StMary's, York indicates that three individual candles were used. At Tours
the candleholder was known as the rastrum, 'the three-pronged hoe'. In some instances
it is not clear whether it was the candle or the candlestick which was triple. The 1836
Missal of Auch refers to the triple arundo; whilst in a manual of the Augustinian Friars
the descriptive phrase triangulo distinctis is equally ambiguous.
Twentieth-century manuals permitted the use of either three separate tapers oi a

triple candle with a single stock. 7 The disposition of the candles in a trident-like forma-
1
For this and other documentary references relating to the use of the triple candle, see
Table43.
2
Its use was unknown in the Milanese and Mozarabic rites, and in those of the Cister-
cian and Dominican orders.
3
Praxis p.316.
4
Ceremonies of Holy Weekp.l19.
5
Manuell.3 p.1257. The Capuchin candle had one foot and three branches.
6
For instance, Agen, Autun, Bayonne, Carcassonne, Digne, St Brieuc, St Die, and
Vannes. Survey of 1984.
7
0'Loan p.283; Fortescue and O'Connell (4th ed.) p.337.
201
tion was prohibited, 1 even though the Caeremoniale Episcoporum of 1600 contains an
illustration of the three candles arranged in this very way (p.298). The combination of
setpent-reed and triple candle has been noted above on page 1 CJ S.
Document Date Source

PRXII 12C I xxxii.7 p.238
Corbie 12C DAMR 3.l3.34p.126(M 1146)
Bee c.1200 DAMR 3.13.34 p.127 (M 1153)
Apamea 1214 DAER 4.24 p.160 (M 25)
Haymo(OM) 1243 HBS 85p.209
Evesham c.1250 HBS 6 col.80*
PGD c.1296 illiv.8p.588
Worcester 13C Antiphonary p.69
Bayeux 13C Ordinary p.135
OrdoXIV c.1310 PL 78.1218B
CA 1706 c.1350 ZRKM p.213
Durham 14C Missal p.185
StMary's, York c.1400 HBS 75p.275
Missale Romanum 1474 HBS 17p.175
Austin Friars 1491 Missal np
Tongres 15 c Ordinary p.164
V allombrosa 1503 Missal fol.xcv
Monte Cassino 1507 Missal fol. 91
Aquileia 1519 Missal fol. 91
Braga 1558 Missal fol.xcvi
Missale Romanum 1570 1950 t.e. p.186
Camaldolese 1634 Ceremonial p.84
Carmelites 1664 Missal p.157
Cistercians 1689 Ritual p.245
Evreux 1740 Missal p.186
Cahors 1760 Missal p.173
Poitiers 1767 Missal p.245
Lyon 1771 Missal p.189
Poland 1819 Manual IT p.473
Vatican 19C England p.119
* Evidence for Maundy Thursday.
Table 43. Evidence for the use of the triple candle.
1
Fortescue and O'Connell ibidem.
202
(ii) The origin of the triple candle
No single explanation can adequately account for the emergence of the triple
candle and its use within the new fire ceremony; surviving documentary evidence
would suggest that its origins are to be sought in a number of different liturgical
milieux. It is maintained 1 that the small candle which bore the new fire into church
became a triple taper to match the threefold cry of Lumen Christi; and indeed this
theory is difficult to discredit in view of the fact that the triple candle and the triple
acclamation do occur together in PR XII, our earliest evidence for the former. Also in
support of this view there is the illustrated evidence of the Exultet rolls of Southern
Italy: some2 of the poles have a single twisted termination, perhaps representing a ser-
pent, and one3 has a double twisted end to the shaft. It is not difficult to believe, in view
of the evidence of the latter, that the number of candles may have been increased from
one to three to match the number of cries, or to symbolise the Trinity, or to accommo-
date both. The objection that such an explanation does not satisfactorily account for the
springing of the candles from a central stock is partially removed when one considers
the relative difficulty of entwining three candles around a central pole compared with
the ease of af:ftxing a candela triplex.
The evidence of the slightly later Ordo Ecclesiae Lateranensis, however, does not
support the theory that the number of candles was increased to three in order to achieve
· the above-mentioned liturgical symmetry. For according to that document the deacon,
prior to chanting the triple Lumen Christi, carried into church on the reed
plures candelas in unum glomeratas, ne a vento leviter extinguantur.4
The functional purpose of the several candles could not be more clearly stated. The
mention of plures suggests that a definite number had not been fixed; but the fact that
an indefinite number was used clearly shows that as far as the Lateran Church was con-
cerned their number had not been increased to correspond to the cries of Lumen Christi.
The subsequent use of three candles in the Roman rite could just as easily have been
determined by that number's symbolic representation of the Trinity as by a deliberate
design to achieve numerical correspondence with the acclamations of Lume~ Christi.
Subsequently the trinitarian significance of the device in question was highlighted.
1
Bugnini and Braga p.189.
2
Avery Plates CXX and CXXXIII.
3
Avery Plate CXXXVIII.
4
'Several candles bound together, so that they might not easily be extinguished by the
wind.' OEL p.61.
203
Louis Thomassin commented that 'we light the tripartite candle in honour of the Trinity,
believing that, bathed in the light of Jesus Christ, we have knowledge of the inner mys-
teries of the Trinity'. 1
Indeed the evidence of OEL and the service-books of other churches2 in which the
three candles were lit at the same time strongly suggests that the close rapport which
existed between each of the three candles and the corresponding cry of Lumen Christi,
as exemplified in the Pian Missal - of which more presently - was unknown during most
of the Middle Ages. In support of this position we can make two important observa-
tions. (i) The acclamation of Lumen Christi did not feature in the rites of a number of
churches in which the triple candle was used? (ii) The triple candle was borne in
procession to the singing of the Inventor rutili at Evesham, Salisbury, York, and
Tongres.4
The signifcance of the phrase in unum5 is not immediately clear. It may indicate
simply that the candles were bound together around the reed and not arranged at fixed
intervals from each other; or it may signify that the candles sprang from a central
stock - a method of arrangement more easy to accomplish, as we suggested above. The
serpentine theme involving the reed and the candles, which is attested in other docu-
ments, was unknown in the Roman rite.
Further support for the functional origin of the triple candle comes from England
and Ireland. At Barking and Durham an unspecified number of candles were affixed to
the top of the spear for the reception of the new frre. 6 Similarly in the Old Irish Missal
the number of candles is not stated (p.126). However, in the Gilbertine rite five candles
were used. 7 In view of the reason given in OEL for the use of several candles, it is not
difficult to see in the multiple use of candles attested by these four documents a precau-
8
tion against the sudden quenching of the processional fire by the elements.
1
In Trinitatis honorem Cereum in tres divisum accendimus, rati nos Jesu Christi lumine
fusos Trinitatis penitiora Mysteria nosse. De Dierum Festorum ll.l4 p.72.
2
For instance, Monte Cassino and Coutances, and also the Carmelite Missal. See Table
43 for all documentary reference to the triple candle.
3
For instance, Bee, Braga, Lyon (all missals including that of 1904), and Rome, accord-
ing to the testimony of Haymo's Ordo Missalis, and most pre-Tridentine missals. In
twenty-two of the latter service-books (from 1474 to 1561) which were examined by the
writer, only two (1558 and 1560 Missals) contain the triple cry of Lumen Christi.
4
See Chapter 4, Table 45.
5
From the above-quoted excerpt from OEL.
6
HBS 65 p.lOl and Missal p.186, respectively.
7
HBS 59 p.39. It is unlikely that this number was chosen to match the number of grains
of incense inserted into the Easter candle.
8
At Durham a candle in a lantern was also used in case of an emergency (Missal p.186).
204
There remain to be considered three further possible factors which may have con-
tributed to the emergence and use of the triple candle. (i) The first has already been
touched upon. We saw that the eighteenth-century liturgist Thomassin suggested that
the processional candle became tripartite in honour of the Trinity. 1 Jean Grancolas also
put forward this explanation. 2 However, we cannot be sure whether the number of
candles was fixed at three as a gesture to honour the Trinity, or whether the trinitarian
association was subsequently added to this device. Corsetti pointed oue that the three
candles springing from a central stock signified both the trinity and the unity of God.
Other writers4 have commented upon the trinitarian symbolism of the candles, Dom
Gaspar Le Febvre noting in the twentieth century that the device anticipated the Bap-
tism in the Trinity which the catechumens in former times had undergone.5 Van
Doren's claim that the use of the triple candle is purely allegorical is unwarranted.
Moreover, his statement that this candle represents Christ in His divinity and in His
humanity seems curious, as does his comparison of the reed bending in the wind with
the humiliation of Christ during His Passion. (ii) It might be argued that the increase in
the number of candles from one to three constituted an elaboration of the serpent-
candle, and symbolised more realistically the swallowing of the snakes, as narrated in
Exodus 7:12. However, while such a theory is attractive, it must be advanced
tentatively in view of the absence of corroborative evidence. It is true that at Braga a
triple candle emerged from the mouth of the serpent-reed; however the late appearance
of the device and the known influence of Cluny suggest that the use of the triple candle
in this rite was a later development. (iii) hi an eleventh-century Exultet roll from Bari
there is an illustration of the triple candle, which, if part of the original picture, would
provide the earliest evidence for the device. 6 Significantly, the pole is tilted forwards,
the position in which it should be held according to a number of early documents. 7
Moreover, it should be borne in mind that Bari was a Byzantine dependency until 1071;
and though the local Italian rite was used in that city, the influence of Byzantine cer-
emonial, which included the episcopal triple hand-candle,8 should not be completely
ruled out.
1
De Dierum Festorum 11.14 p. 72.
2
Commentarius p.316.
3
Praxis p.316.
4
Desideri, Praxis p.150; Thurston, UIW p.415.
5
1928 Daily Missal p.828.
6
Avery Plate XIX. A very is of the opinion that the pole is a later addition.
7
0R 26.9; OR 29.17; Alcuin, PL 101.1205D.
'7he 'tPtKTtptov is mentioned in Byzantine liturgical texts of the tenth century.
205
(ill) The lighting of the triple candle
The functional purpose of the triple candle necessitated the kindling of all three
wicks at the same time. 1 There is no evidence before the sixteenth century to show that
the three candles were lit one by one in close conjunction with each cry of Lumen
Christi. 2 Even after the practice had become established of uttering the three cries inter-
mittently during the procession into church, the lighting of all three candles at the same
time continued at Vallombrosa, at Cosenza, in the Carmelite rite, and possibly at
Aquileia. 3
The custom of lighting one candle in sympathy with each acclamation of Lumen
Christi is fust attested, perhaps smprisingly, as late as 1570 in the Missale Romanum of
Pope Pius V, and seems to have had its origin within the liturgy of that church. The
practice was subsequently adopted by a number of diocesan churches, mostly French,
and by some religious orders.4
The establishment of the close rapport between the candle and the cry of Lumen
Christi resulted in the lighting of the fust candle inside the church,5 and necessitated the
transportation of the flame from the source of the new fire to the door of the church in
readiness for the lighting of the first candle. A number of different devices are known
to have been used to fetch the new fire. They include a small candle, 6 which was some-
times placed in a lantern if a strong wind was blowing;7 a busia, which consisted of two
wax-wicks twisted together for the better preservation of the flame; 8 and a gossypium
ceratum, which was a length or roll of cotton covered with wax. 9 A manual of cere-
monies for Poland mentions a wax-coated spill for lighting the triple candle or that in
the lantern if necessary. 10 The use of ligna sulphurata fust appears in the
1
1507 Missal of Monte Cassino fol.92; Ordinary ofTongres p.164.
~practice has to be inferred in some manuals e.g. the 1634 Camaldolese Ceremo-
nial (p.84) and the 1775 Cappuchin Ritual (p.128).
3
1503 Missal fol.xcii; 1549 Missal fol.115; 1664 Missal p.157; and 1519 Missal
fol.91, respectively.
4
Missals ofEvreux (1740) p.187; ofCahors (1760) p.173; ofPoitiers (1767) p.245; of
Auch (1836) p.192; and of Nantes (1837) p.199. Religious orders include Augustinian
Canons (1579 Ordinary fol.137); Camaldolese (1634 Ceremonial p.84); and Capuchins
p775 Ritual p.128).
See also Chapter 4 Section (ill).
6
Caeremoniale Episcoporum p.298; Maison du Roy p.400; Ceremonial of Lyon p.474.
At Lyon, where there was no new fire procession, the triple candle was lit only at the
announcement of the Exultet.
7
Gavanti: . I p.234.
8
Merati p. 78.
~artinucci n p.241.
10
1819 Manual IT p.468.
206
Camaldolensian Ceremonial of 1634. 1 They were sulphur-tipped or sulphur-coated
splints of wood whose original purpose was to transfer the new fire from the burning
woodpile or chafmg-dish to the small candle, 2 or to the triple candle in instances where
all three lights were kindled together. 3 They were subsequently used, it would appear,
to bring the new fire into church and to light one of the three candles prior to the singing
of the first Lumen Christi, as at Auch and Nantes,4 and also to light the Easter candle
itself.5
(iv) The disposition and disposal of the processional candle(s)
Several commentators mention the stand for holding the reed once the Easter
candle had been lit.6 This could be made of marble or wood, materials also specified in
a number of ceremonials. 7 The Augustinian Ceremonial enjoined that it should be
placed on the Epistle side of the altar. At Uzes the setpent-reed was placed next to the
archdeacon's seat. 8 According to the Pontifical of Poitiers the deacon handed the reed to
the sacristan after he had lit the Easter candle (p.215); and among the Cistercians the
small candle taper used for lighting the Easter candle was blown out after the singing of
the Exultet and Preface.9 At the Cathedral of StJohn Lateran the reed was taken into
the sacristy after the two standard candles and the seven lamps had been lit. 10 A rubric
in the Sarum Missal states simply that the reed should be moved after the conclusion of
the Preface, a direction enjoined by Corsetti. 11 Merati prescribes that a drip-pan should
be placed below the reed-candle to catch the wax as it falls (p. 77). Elsewhere the extin-
guishing of the reed-candle or of the triple candles probably did not take place until the
conclusion of the ceremony, as at Lyon. 12 The Augustinian Friars allowed their three
candles to bum until after the end of Vespers on Holy Saturday. 13 At Vallombrosa the
church lamps were lit by using the triple candle. 14
1
Page 82. The mention of them by Gavantus in 1652 (Thesaurus I p.233) would suggest
that their use in the sixteenth century was not confmed to the rite of this order.
2
1634 Camaldolese Ceremonial p.82 and p.84; 1662 Ceremonial of Paris p.375.
3
For instance, in the Cistercian rite (1689 Ritual p.247).
4
1836 Missal p.192 and 1837 Missal p.199, respectively.
5
See Part N Chapter 11.
6
Gavantus/Merati N p.155; Desideri p.144; Gattinari p.143.
7
For instance, those of the Augustinian Friars (1714) p.311; of the Capuchins (1775)
r,.125; and of the Camaldolese (1634) p.82.
1495 Missal fol.lxiii. The rubric adds 'or in a convenient place'.
9
Nom.Cist. p.105. A reed was not used in this rite.
1
<This took place during the singing of the Preface. OEL p.61.
11
Dickinson, Missale p.343 and Praxis p.320, respectively.
12
1838 Ceremonial p.418.
13
14
1503 Missal fol.xcv.
207
Chapter Four
THE PSALMS, THE INVENTOR RUTIU, AND THE LUMEN CHRISTI
The earliest docrnents record that the procession of clergy and people moved into
church with the new ftre in silence. 1 From the middle of the tenth century the practice
arose of singing one or more psalms, or Prudentius' hymn, Inventur rutili. Alternatively
the cries of Lumen Christi punctuated the silence of the procession in some churches.
(i) The Psalms
The chanting of psalms during the return of the procession with the newly-kindled
frre was prescribed in Lanfranc's Decrees. It was also enjoined in a number of earlier
Benedictine custornaries as well as in the Cluniac and Carmelite rites. 2 The tradition
survived for several centuries in a number of cathedrals, mainly it would seem, as a
result of monastic influence. It is not difficult to believe that the return-psalms were
introduced into the new frre ceremonial as a counterpart to those sung on the way to the
new fire in an attempt to achieve a sort ofliturgical symmetry. Psalms 26, 66, and 79
were almost certainly chosen because of their mention of the light of the Lord. Psalms
23 and 147 tell of the triumph of the Lord and of his glorification, the latter actually
mentioning God's control offrre. The choice of Psalms 69 and 119, and to some extent
Psalm 56, seems somewhat obscure. Their penitential aspect makes them more fitting
to have been sung before the kindling of the new fire, as indeed the Penitential Psalms
were sung in a number ofplaces. 3
At Bayeux the choir sang the antiphon Clamaverunt ad Dominum cum tribularen-
tur as they returned into church; whilst at Rouen, before the deacon began the Exultet,
they sang the antiphon Cum rex gloriae Christus infernum debellaturus intraret et
chorus angelicus portas principum tolli praeciperet.4 This antiphon was also sung at
Hereford at the Sepulchre on Holy Saturday night (Breviary p.324).
(ii) The Inventor rutili5
1
0R 26.9; OR 29.17; OR 31.63; OR 32.17; PRG II p.97; CMG (Albers V p.32); the
ordo ofCorbie and Ordinal ofM.Cassino (DAMR 3.13.34 p.126, M 1145 and M 1139
respectively); and Alcuin, PL 101.12050.
2
For these and other relevant references, see Table 44.
3
The Consuetudines Cluniacenses (Albers II p.47, Antiquiores C) record the older tradi-
tion of using the Penitential Psalms on the return.
4
1780 Semaine Sainte p.495 and 1497 Missal np, respectively.
5
For the implications of the internal evidence of the hymn, see Appendix 12. For a list
of churches where this hymn was sung after the blessing of the new fire, see Table 45.
208
Church/Order Psalms Date Source
Farfa 53 56 66 69 c.lOOO PL 150.1199
Fruttuaria 23 26 66147 c.lOOO Albers IV p.54
V allombrosa 66 c.l040 Albers IV p.249
Dijon 53 56 66 69 147 c.1050 DAMR 3.13.34 p.126
Cluny 53 56 79 c.1060 PL 149.659A
Lanfranc 53 56 66 69 c.1070 PL 150.467B
Sigibert 26 27 53 c.1070 Albers n p.93
Bee 53 56 66 69 c.1200 DAMR 3.13.34p.126
Norwich 26 c.1265 HBS 82p.81
Evesham 2627 c.1250 Ritual col.80
St Vedast, Arras 53 56 c.1300 HBS 86p.160
Cannelites 66 c.1312 Ordinary p.l71
Lyre 53 56 66 69 c.1400 DAMR 3.13.34 p.126
Burgos 66 1546 Missal fol.cili
Palencia 66 1568 Missal fol.c
Rouen 26 1640 Ritual p.307
(The numeration of the psalms is taken from the Septuagiant.)
Table 44. The processional psalms after the new fire.
The singing of this hymn of Prudentius after the kindling of the new fire is frrst
attested in the Romano-Germanic Pontifical; but according to that document its use was
geographically restricted. 1 Although Lanfranc refers to the Inventor rutili in his
Decrees,2 there is no evidence for its use in any monastery of Italy, Spain or Switzer-
land. Indeed, with the exception of Aquileia, the hymn was sung only in churches to the
east and north of the Alps. 3 It was particularly popular in Germany as is clear from a
glance at Table 45. According to Lanfranc the Inventor rutili was sung by two choir-
boys who were standing close to the bishop's throne as the procession made its way
from the place where the new frre had been kindled and blessed; but the practice
developed in which the singers themselves joined the procession, and a chorus, formed
of those participating, repeated the frrst verse of the hymn as a refrain between the sing-
ing of subsequent verses. At Durham and Westminster two brothers led the singing, at
Exeter two boys, and at Barking the duty-priest for the week and a priest representing
the Chapter; whilst among the Gilbertines two candle-bearers (ceroferarii), or two
1
PRG n §345 p.97. Manuscripts c and K both state that a procession in silence was the
norm. It is only C which adds: Aliqui tamen hie cantant hymnum Prudentii, 'However
some sing Prudentius' hymn at this point.'
2
PL 150.467B. Since Lanfranc also mentions the singing of psalms after the kindling of
the new frre, presumably the Inventor rutili was alternative to them; although Lanfranc
does not refer to a choice.
3
Since the singing ofthe hymn was not universal according to PRG, Dendy's statement
(p.140) that the hymn was used in the Roman rite for a time must be challenged: ' ... the
Inventor rutili. .. at Rome only enjoyed a brief period of use during the ascendancy of the
Ordo Romanus Antiquus [=PRG]'. Not all features ofthe new fire ceremony found in
P RG passed into the Roman rite, e.g. the kindling of fire on Maundy Thursday.
209
others performed the duty. Instances of its use after the Council of Trent are few. It
survived at Sens and at Perigueux until the eighteenth cenury; and is even found in a
ceremonial for Le Puy as late as 1836.
Church/Document Date Source

PRG (MS C) c.950 II §345 p.97
Lanfranc c.1070 PL 150.467B
Rupert of Deutz c.llll PL 170.149B
MCP 12C HBS 39 pp.169-70
Ireland c.l200 Old Irish Missal p.126
Evesham c.1250 HBS 6 col.90
Norwich c.1265 HBS 82 p.91
LeMans c.1295 DAER 4.24.3 p.146 (M 89)
Salisbury 13C HBS91
Worcester 13 c Antiphonary p.69
Exeter 1337 HBS 37
Strasbourg 1364 DAER 4.24.3 p.146 (M 35)
Anderlecht 14C Ordinary p.88
Westminster c.1370 HBS 5 col.578
York 14C Missal p.llO
StMary's, York c.1400 HBS 75p.292
Barking 1404 HBS 65 p.lOl
Wiirzburg 1477 Ordinary np
Verdun 1481 Missal fol.lxiiii
Breslau 1483 Missal np
Freising 1487 Missal fol.cii
Basel 1488 Missal fol.xcii
Trier 1488 Missal fol.cii
Cologne 1494 Missal fol.cxxii
Prague 1498 Missal fol.xci
Hildersheirn 1499 Missal fol.xcvi
Tongres 15C Ordinary p.164
Esztergom 1501 Missal fol.lxxiiii
Pass au 1503 Missal fol.lxxxiv
Halberstadt c.1505 Missal fol.lxx
Mainz 1507 Missal fol.xcii
Saltzburg 1507 Missal fol.xciii
Hamburg 1509 Missal fol.xc
Bremen 1511 Missal follxxxv
Spires 1512 Agenda fol.xcvi
Minden 1513 Missal fol.ciii
Lund 1514 Missal fol.xc
Breslau 1519 Missal fol.lxxix
Meissen 1520 Breviary np
Abo c.1522 Manual p.240
Auxerre 1537 DAER 4.24.3 p.146 (M 39)
Liege 1540 Missal fol.lxxvii
Ratisbon 1570 Ritual np
Sens 1715 Missal p.238
Perigueux 1782 Missal p.159
LePuy 1836 Ceremonial p.376
Table 45. Evidence for the Inventor rutili.
210
(iii) The cry of Lumen Christi
(a) Origin.
Dom Bernard Capelle,1 followed by others/ would find the origin of the triple cry
of Lumen Christi of the Roman rite, together with the complementary refrain of Deo
gratias, in the corresponding Mozarabic ceremonial of Holy Saturday. According to the
latter rite the bishop emerged from the sacristy with the lighted Easter candle and pro-
claimed 'Deo gratias'; to which the congregation responded three times in like
manner. 3 As the procession moved into the choir, they sang the antiphon Lumen verum
inluminans omnem hominem in hunc mundum venientem. Capelle argues that this ritual
found its way to Rome by way of Gaul, Milan, and Central Italy (p.ll7), citing as evi-
dence for this route a debatable instance of Mozarabic influence in the Old Gallican
Missal,4 and the testimony of what he believed to be an eleventh-century Milanese ordo
(MS Vat.lat 10673), but which has been shown to be a Beneventan gradual.5
It is true that similarities do exist between the Spanish rite and that of Central
6
Italy : the lighting of the bishop's candle, the procession into church with that candle,
and the threefold acclamation of the congregation; but the greater number of differ-
ences which exist between the two rites should make us very cautious in trying to detect
the influence or dependency of the one rite on the other. Moreover, it has yet to be
shown what liturgical contact or interchange existed between Spain and Southern Italy
in the period AD 700 to AD 900 when the Beneventan rite is most likely to have been
susceptible to the liturgical influences of other churches, and when any importations of
Spanish provenance are most likely to have occurred. Political conditions in Spain,
1
La Procession pp.116-7.
2
Bugnini and Brage pp.189-90; Dendy p.138.
3
See also Appendix 13.
4
lt is true that the prayer in the Vetus Missale Gallicum (PL 72.363) entitled PRAEFA-
TIONE CERAE 'does seem to indicate a procession where all carried lights' and may
· have originated in a liturgical milieu, such as the Mozarabic, where this ritual did take
place. On the other hand the references in the prayer to vinculis ..disruptis, illumina-
tionem, and candoris suggest a baptismal setting in which accensa luminaria will be
the candles of the neophytes. Evidence for the existence of these lights at the time
comes from Amalarius (Liber de Ord.Ant. 44.8).
5
Hesbert p.189.
6
As attested in the Beneventan Gradual and Missale Antiquum. For the text of these two
documents, see Hesbert p.188.
211
however, during this crucial time would suggest that exchanges of liturgical forms and
practices between Spain and Southern Italy were most unlikely, especially any issuing
from Spain.
The new frre ceremonies of the Beneventan and Mozarabic rites differed from
each other in the following respects:
Mozarabic Benevefltan
1. Easter candle brought into church. Easter candle already in position.
2. Bishop cries 'Deo gratias' at the sac- Deacon cries threefold Lumen Christi at
risty door. ambo.
3. Singing of antiphon lumen verum dur- Silent procession.

ing procession.
4. Fire kindled and Easter candle lit in a Fire not necessarily kindled inside the
darkened room. building. 1
5. Lighting and blessing of a lamp. No use of lamp.
6. No blessing of new frre. Blessing of new fire.
If the Beneventan rite had been subjected to Mozarabic influence, we might have
expected a greater correspondence of ceremonial detail and fewer divergencies. The
phrase ex occulto does not necessarily refer to a darkened sacristy - see Note 1 below;
and even the congregational cry of Deo gratias is not completely parallel in the two
rites. Moreover, although certain features of the Spanish rite are of a venerable
antiquity, it is by no means certain that all the elements recorded in the tenth-century
1
The rubric of the Gradual states : De quo igne accendetur cereus; et, quasi ex occulto,
proferatur in publicum. 'From this frre the Easter candle will be lit; and, just as the frre
has been kept in a place of concealment, so let it be brought forth for all to see.' Den-
dy's claim (p.132) that this 'suggests the theatrical procession with lights from a dar-
kened room' is unconvincing. Not only is there no evidence for the use of more than
one light in the Beneventan procession; rather, the phrase suggests the locus secretior,
familiar from the Roman rite, which was a place well hidden from view. Moreover, a
rubric earlier in the Gradual prescribed the kindling of the new frre by means of a fire-
stone 'or in some other way' (alio livet <= quolibet> modo). If the latter included the
use of a lens, the ritual could hardly have been performed 'in a darkened room'.
212
Antiphonary date from the time ofElipandus (c.AD 718-802). In Appendix 12 we have
argued in favour of the importation of a number of Gallicanisrns into the Mozarabic rite,
including the cry of Deo gratias.
All the early evidence for the Lumen Christi is to be found in documents of Cen-
tral Italian provenance. 1 Indeed, as we observed above, the threefold cry is only found
outside Italy amongst the religious orders whose own rites closely followed that of
Rome2 - it is even absent from the ceremonial of many northern Benedictine houses, not
being prescribed in Lanfranc' s Decrees - and in churches which were influenced by
Roman ritual, such as Marseill£'or which used romanised Gallican missals, such as ~ --~
Chalons and Poitiers. Havin{discounted a Mozarabic provenance for this liturgical fea- ......
ture, it is not difficult to fmd its origin within the Romano-Gallican tradition. In Section
22 of the eighth-century Ordo 19 we read :
If night comes while they are eating and it is necessary to kindle a light, as soon as
the brother who carries the light enters, he says, so that all may hear, 'The Light of
Christ'. All reply 'Thanks be to God'. After a blessing from the senior he puts
the light in its place.
As Andrieu observed, Benedict's Rule provided for the evening meal to end
before night-fall. However the 'Strasbourg liturgical historian', who dated this ordo to
the years AD 781-90, suggested somewhat carelessly that, in view of the climatic con-
sideration relating to Section 22, the ordo may well have come from north of the Alps. 3
Bad winter evenings do occur in Italy; and presumably the evening meal was always
taken at the same hour. This ordo, therefore, could equally have originated in Italy.
Indeed, it seems difficult to escape the conclusion that this greeting with its response
came to be used in the Holy Saturday liturgy once the Vigil began to be held in the late_
afternoon or early evening.
(b) The development of the ritual.
The ritual of the Lumen Christi passed through a number of stages in its develop-
ment before it reached the form familiar from the Pian Missal of 1570.
'For a list of rites in which the Lumen Christi was a feature and for the documentary
references, see Table 46.
2
In the Cistercian rite there was only a single cry at the altar (1669 Missal p.155);
whilst the Lumen Christi did not feature in that of the Premonstatensians (King, LRO
p.190).
Les Ordines Romani ill p.212.
213
C hurchJDocument Date Source
Beneventum• c.1000 Hesbert p.188
V allombrosa llC Albers N p.249
Lateran c.l140 OEL p.61
Monte Cassino A • 12C PR XII I p.292
PRXII 12C I xxxii.7 p.239
Apamea 1214 DAER 4.24. p.160 (M 25)
Marseille 13C ILEM p.84
OrdoXIV c.1310 PL 78.1218C
CA 1706 c.1350 ZRKM p.214
Uzes· 1495 Missal fol.lxili
Camald91ese 1503 Missal fo1.89
Missale Romanum 1570 1950 t.e. p.188
Frejus c.1600 De Rubeis p.327
Cistercianst 1689 Ritual p.247
Cahors• 1760 Missal p.173
Capuchins 1775 Ritual p.128
* There was only one processional candle.

t Lumen Christi was acclaimed only once.
Table 46. Evidence for the acclamation of Lumen Christi.
1. The oldest form of the ritual, and that from which subsequent variations of the
ceremonial developed is found in the Beneventan rite of the late tenth century, as
attested in the Gradual and M issale Antiquum. According to these two documents the
threefold Lumen Christi together with the responsorial Deo gratias was acclaimed at the
ambo after the lighting of the Easter candle by the deacon and just prior to his chanting
of the Exultet. 1 . Monte Cassino A may attest the same practice. 2
1
As the bearer of the Lumen Christi it was the duty of the deacon to proclaim its pres-
ence.
~e text of Monte Cassino A and Monte Cassino B, both of the twelfth century, is
printed in Andrieu's edition of PR XII (I pp.292-3). The problem of the former docu-
ment and in particular the interpretation of the clause : Acolytus vero portat cereum ad
ammonem, is discussed in Appendix 11. If cereum refers to the Easter candle, this
document attests Beneventan practice. If, however, the reed-candle is to be understood,
the document relates to the second stage of the development, as does Monte Cassino B.
In either interpretation it is the acolyte's voice that is heard.
214
2. In Monte Cassino B the cry of Lumen Christi is directed by the deacon at the
reed-candle which has been brought into church, lit with the new fire, by an acolyte.
Only after the last response of Deo gratias is the Easter candle lit. The so-called sub-
urbicarian practice, described in PR XII, suggests that the triple cry was uttered by the
deacon standing close to the Easter candle near to the altar, again before that candle was
lit.'
3. The next stage of development reveals an elaboration of ceremonial. For,

whereas the deacon or acolyte had previously stood in the same position to utter the
cries,2 the proclamations of Lumen Christi became incorporated into the new ftre pro-
cession in such a way that the first Lumen Christi was heard at the door of the church,
the second in the nave, and the final cry at the altar. 'This arrangement subsequently
obtained unti11970, surviving the liturgical reforms of 1955, though with the Easter
candle replacing the triple candle. In some rites the stational norm of door, nave, and
altar was not observed. The third acclamation occurred at the ambo if it was customary
to locate the Easter candle in that position. Ordo XIV prescribed the choir for the sec-
ond station. This also took place at Frejus and Cahors. At Marseille the procession
emerged from the sacristy door and stopped at the altar of the Blessed Virgin Mary, in
the choir, and at the reading desk next to the Easter candle. In the Roman rite since
1970 the nave stopping has generally been omitted, the acclamations occurring at the
new fire, in the doorway, and at the altar.
The processional chanting of the Lumen Christi is first attested at V allombrosa in

the eleventh century; and it may be that the practice originated within that very monas-
tery. According to earlier practice the procession had moved into the church in silence,
as we noted at the beginning of this chapter. At Vallombrosa, however, the unique
practice obtained of chanting a psalm and uttering the triple Lumen Christi during the
procession. A glance at Table 44 shows that contemporary monastic practice was to
chant several psalms on the return into church; at Vallombrosa only one psalm was
sung. It is the writer's belief that at this monastery in former times more than one psalm
was chanted during this part of the ceremony; and that the processional Lumen Christi,
borrowed from a Central Italian monastic milieu, was deliberately included in the cer-
emonial either as an musical feature additional to the psalms, or as a replacement for
those psalms which were previously chanted once the procession had entered the
1
I xxxii.10 p.241. So also PGD (Ill iv.S-9 p.588). For the so-called suburbicarian prac-
tice, see Appendix 15.
2
Sicardus records that this took place at the doors of the church. Mitrale, PL 213.3238.
215
church. The triple acclamation of the Lumen Christi during the procession subsequently
found its way into the Roman rite in the twelfth century. At Rome the duty was per-
formed by the junior cardinal deacon. 1
4. The tmal stage in the development saw the utilisation of the triple candle to
reinforce dramatically the significance of the threefold cry. It had formerly been the
practice to light all three candJes from the new fire at the same time. This is implied in
Ordo XIV : with each successive cry of Lumen Christi the deacon raised the candle
higher. The practice is also attested in the two Roman missals of 1558 and 1560.
According to the Pian Missal of 1570 the triple candle was carried into church unlit.
The deacon who bore the candle then lowered the reed and one of the candles was lit
with the new fire. Thereupon the deacon chanted the flrst Lumen Christi. The ritual
was then repeated in the centre of the church; and after the procession had reached the
altar, the last candle was lit, and Lumen Christi was announced for the third and tmal
time. In prescribing the west end of the chancel, the centre of the chancel, and the altar
steps as the three stations for the Lumen Christi, the compilers of the Caeremoniale
Episcoporum were perhaps taking a realistic view of the small number of laity who
were likely to attend the Vigil, then held on Saturday morning (ll.27 p.297). The
Roman Missal of 1574 and the Vallombrosan Missal of 1503 both record that the triple
candle was handed to an acolyte and subdeacon, respectively, after the chanting of the
third Lumen Christi. They held it until it was time to light the Easter candle.2 With
each cry of Lumen Christi all genuflected except the cross-bearer. 3
1
PR XII I xxxii p.239 and OEL p.61. The history of the Lumen Christi in the Roman rite
is not at all clear. In addition to the two above-mentioned documents, it is also found in
Ordo XII (1192) PL 78.1076C; PGD (c.l296) ill p.588; Ordo XIV (1311) PL
78.1121; CA 1706 (c.1350) ZRKM p.214; and Bindo F. (1377) ZRKM p.276. It is not
found in Haymo's Ordo Missalis, or in the Dominican rite, both of which were
modelled closely on papal ceremonial (SMRL I p.44; King, LRO p.338), or in the 1474
Missale Romanum.and subsequent missals except two printed in Venice in 1558 and
1560. It appears in the mandatory Pian Missal of 1570. Van Dijk (I p.82) attributed
the omission of the Lumen Christi in Haymo's ordo to the fact that it was not known
outside Rome; but this, as we have seen, is clearly incorrect, as is Dendy's claim that it
was 'probably kept out by the popularity of the Inventor rutili (p.140).' Both the 1558
and the 1560 Missals refer to the custom being observed 'in certain places' (HBS 33
p.84), one of which was presumably the papal court, and state that it is a priest, as
opposed to a deacon in MR 1474 and other Roman missals who performs the duty.
2
HBS 33 p.85 and 1503 Missal fol.xcii.
~e genuflection of the deacon who held the reed is frrst attested in the Roman rite.
(Ordo XV, PL 78.1321C). The practice is also found at Cahors (1760 Missal p.172).
According to the 1600 Caeremoniale Episcoporum the deacon should both genuflect
and raise high the triple candle simultaneously (p.298).
216
Thus a close rapport was established between the lighting of the three candles and
the threefold acclamation. It inevitably resulted in directing the attention of the
congregation away from the Easter candle, and in the close identification of the 'light of
Christ' with the flames ofthe three candles. Nor did the trinitarian significance of the
number of candles go unnoticed. 1 Indeed it might be said that a slight shift in emphasis
occurred in the status and role of the triple candle vis-a-vis the aspect of God which the
light represented. For the triple light and the threefold cry to some extent blurred the
distinction between God's light and Christ's light as expressed in the Nicene Creed, and
detracted from the centrality of the latter within the Paschal vigil. Gueranger went so
far as to assert that the threefold cry of Lumen Christi expressed the revelation of the
divinity of the Three Persons of the Trinity .Z The greeting of the triple candle also had
the effect of detracting from the significance of the Easter candle, especially since it was
not kindled in the majority of churches until the singing of the Preface was
half-completed. This had the result of marring the close relationship between the
Candle and the flame, and of assigning to the column of wax almost the function of a
totem. 3
The practice of chanting the first Lumen Christi in a deep voice, and the
subsequent cries at a successively higher pitch, is first found in the above-mentioned
eleventh-century Customary of V allombrosa. Apropos of our contention referred to
above, this may suggest a Central Italian provenance for the custom, although there is
no hint of this practice in the two Beneventan documents. The ritual is also attested in
P R XII and PGD, and was subsequently adopted in the majority of churches in which
the Lumen Christi was acclaimed.
The emergence of this liturgical feature may be explained in two ways. (i) If we
are correct in our belief that the custom originated in a Central or Southern Italian
liturgical milieu, the rise in pitch in the deacon's voice may have developed in
correspondence with his ascent of the ambo in three stages. The weakness of this
theory is that there is no evidence that his ascent of the ambo was a gradual one. (ii)
More likely, perhaps, is the writer's own suggestion that the rise in pitch occurred so as
to enable those standing at some distance from the deacon to hear his acclamation. A desire to
make oneself more audible is usually achieved by an utterance or shout at a higher level
1
The prominence of the triple candle and the superstitious awe in which it came to be
held is referred to by two writers. Sir James Frazer recorded that at the end of the last
century in the Abruzzi fragments of the three candles were used as charms against
lightning (Golden Bough 7.1 p.122); and well into the present century Estella Canziani
wrote that at Isemia, 'If three drops of wax from the three candles lit by the priest from
the new trre drop on anyone's hat, that person is safe against lightning, provided he
keeps his hat on.' (Through the Apennines p.328.)
2
Liturgical Year pp.SSS-9.
3
Harbert p.236.
217
of sound, especially if the ftrst attempt at attracting attention was considered ineffective
and the strength of the voice insufficient. Moreover, the twice-repeated cry of Lumen
Christi would serve to emphasise the importance and significance of the flame which
the deacon held in his hands.
References for Table 36 on p.157. (NB M. = Missal.)

(1) OR 26.3. (2) OR 28.25. (3) OR 29.14. (4) II p.57 8215.
(5) PL 101.1205C. (6) np. (7) M. fol.89. (8)Cerem. p.125. (9) Ordinary
p.171. (10) Ritual p.245. (11) M. fol.xlix. (12) M. fol.lxxxv. (13)
Van Dijk II p.245. (14) M.fol.lxvii. (15) DAMR 3.15.10 p.142. (16)
PL 137.491B. (17) M.fol.xci. (18) M.p.182 {19) Ordinary p.87. (20) M.
fo1.91. (21) M.p.239· (22) M.fol.xci. (23) M.p.41. (24) Hesbert p.185.
(25) Cerem.p.315. (26) M.p.225. (27) M.fol.xcvi. (28) M.np. (29)M.p.172.
(30) M.p.194. (31) M.p.171. (32) M.fol.115. (33) Cerem.p.329. (34) M.
fol.lxxiii. (35) M.p.186. (36) Frazer, GB 10 p.126. (37) De Rubeis p.327.
(38) M.fol.lxx. (39) M.fol.xcvi. (4o) M;p.126. (41) Bellotte p.813. (42)
M.p.186. (43) Antiphonary p.280. (44) M.p.159· (45) Ordinary np. (46) M.
p.219. (47) M.p.189. (48) M.p.214. (49) M.fol.lxvii. (50) M.fol.xcii.
(51) M.p.169. (52) M.p.200. (53) M.p.158. (54) M.fol.109. (55) M.p.198.
(56) M.fol.c. (57) Cerem.p.375. (58) M.fol.lxxxiv. (59) M.p.158. (60)
M.fol.lxix. (61) M.p.204. (62) Manual p.305. (63) M.p.209. (64) Manual
p.20. (65) M.fol.lxxxv. (66) M.fol.lxxii. (67) M.p.186. (68) M.fol.lxxv.
(69) M.p.163. (70) M.p.206. (71) M.p.191. (72) M.fol.ci. (73) M.p.227.
(74) M.fol.liiii. (75) M.p.215. (76) Ordinary np.
218
Chapter Five
THE EASTER CANDLE AND THE PROCESSION
The bringing in of the fire and the blessing of the light that it provided are the two
principal elements which are derived from the Lucernarium and present in all the rites
of the western tradition. The Milanese rite preserved the primitive practice of using a
lamp for the bearing of the light. In other traditions in which the new fire ceremony
was combined with the blessing of the Easter candle, three different practices developed
of conveying the new f'rre to the place where the Easter candle was to be blessd. The
most common method of employing a small candle or the triple candle to carry the frre
has been dealt with in the above Chapters 1-3. The second and third ways of bringing
the new f'rre into church involved the same fundamental procedure, but admit the use of
the Easter candle as a characteristic feature of the procession.
(i) The beari~g of the Easter candle: 1 Lit
The tradition in which the Easter candle was borne in procession already kindled
is first attested in the eighth-century Ordo 28 :
Et, accenso cereo, procedunt simul omnes de sacrario cum ipso cereo in ecclesia
cum silentio, nihil cantantes, et ponitur in candelabro ante altare. 1
There seems little doubt that the Easter candle has been substituted for the lamp of the
Lucernarium in order to bear the new frre into church. Interestingly, in the Mozarabic
rite, in which the new frre was also borne into church by means of the Easter candle, the
use of the lamp was also retained? In Appendix 11 we have shown that in the twelfth
century in Central Italy almost the same ceremonial involving the procession with the
lighted Easter candle was still in use. In that latter rite the Easter candle was subse-
quently taken to the ambo for the singing of the Exultet and for the Prophecies. At
Naples in the fourteenth century the cimiliarcha carried the Easter candle into the
church lit;3 whilst the same practice at Bourges as late as the eighteenth century is
almost certainly a survival of the same tradition attested in Ordo 28 and Ordo 31 ;4 for
1
'And after the Easter candle has been lit, they all accompany it in procession into the
church in silence. There is no singing. The Candle is placed in a candelabrum in front
of the altar.' O.R. 28.59. The practice is also attested in Ordo 31 (§63).
2
For the Mozarabic rite, see Appendix 13. The custom of using a lamp, probably the
result of Mozarabic influence, was also maintained at Ripoll (Sacramentary p.92).
3
Mallardo p.33.
4
1741 Missalpp.225-6.
219
that Gallican rite contains other primitive elements, such as the litany before the Exultet,
and the blessing of the Easter candle with the original form of the Veniat quaesumus. 1
The revival of this tradition following the liturgical reforms of 1955 was part of the
attempt to emphasise the importance of the Easter candle and to restore it to its former
position of centrality within the Vigil liturgy. The claim that the Easter candle ceased to
be carried in procession because in some churches it had become too large and too
heavy cannot be sustained.2 (a) In some rites, such as the Ambrosian, it seems unlikely
that the Candle was ever carried in procession. (b) We shall see in a later chapter that
very large candles were carried during the course of the old Roman liturgy. (c) The evi-
dence would suggest that the Easter candles began to assume massive proportions long
after the custom of bearing them in procession had generally fallen into desuetude.
(ii) The bearing of the Easter candle: 2.Unlit
The custom of bearing the Easter candle in procession unlit appears to be a syn-
thesis of traditions : for the small candle, lit with the new ftre, was also carried in the
same procession. The earliest mention of the bearing of the unlit Easter candle is found
in the Sacramentary of Corbie and in PRG, both of the tenth century .4 It is not clear
from the former document how the fire was taken to the altar where it was blessed. The
evidence of PRG is discussed below. At Aquileia the unlit Easter candle was borne
along with the triple candle ablaze; 5 and at Palencia two small candles on spears
accompanied the unlit column of wax from the cloister.6
(iii) The evidence ofPRG
The composite character of the Pontifical is apparent from the variant rubrics
recorded by its Manuscripts C and K; and the task of identifying the separate strands
and elements which make up its new fire ritual is not made easy by the fact that the new
frre ceremony took place on each day of the Triduum.
Sections 342 and 346 of PRG are of principal concern for our study. The former
records that after the procession had assembled outside the church
cereus ponitur in loco mundo, 'the candle is set in a clean place'.
1
See Part N Chapter 10, especially p.290.
2
Bugnini and Braga p.189.
3
Part N Chapter 16.
4
PL 78.336B and PRG ll §§342 and 346 pp.94 and 98.
5
1519 Missal fol.91.
6
1568 Missal fol.c.
220
Since we are informed later that the cereus was lit with the new fire and put on a reed
(§345), it would seem to follow that this was the candle which was placed in loco
mundo. There are, however, two difficulties over this interpretation. The sign of the
cross is made over the candle and the benediction-formula, Deus mundi Conditor, is
pronounced over the new fire. The ascription of undue importance to the small candle
with this formula, is not only unique to PRG, but recalls the blessing of the Easter
candle according to the Gelasian sacrarnentaries. 1 This fact and the setting of the candle
'in a clean place' strongly suggest that in §342 cereus is to be identified with the Easter
candle, and that the rubrics of this section relate to its consecration outside the the
church and presumably close to the source of the new fire. The structure of the ritual
differs from that found in the Gelasian sacrarnentaries in that, whereas the latter
comprises the bringing of fire, the sign of the cross, the lighting of the candle, and the
blessing, PRG omits the kindling at this stage. It is not difficult to believe that at a
former time the Easter candle was lit outside the church, and borne thus in procession
into the building; and that, with the merging into a combined ritual this practice and the
tradition in which the new fire was carried on a reed-candle, the bearing of the Easter
candle, also ablaze, was seen as a superfluous duplication. Hence it carne to be borne
unlit.
Now it is true that the rubrics of PRG do not actually state that the Easter candle
was borne unlit into church. However, after the procession had entered the church, we
are informed :
Et illurninantur ex eo VTI larnpades ante altare quae tarnen prius sine lurnine erunt
ita compositae, ut absque ullo irnpedirnento possint accendi. Cereus vero magnus
qui benedicendus est, ponitur in candelabro ante altare in medio ecclesiae...2
At fust sight it might appear as though ponitur relates to the position of the Easter
candle as a result of its being set in position prior to the start of the ceremony; and such
a descriptive sentence would not be out of place within the context. However, ponitur
bears a passive verbal force which relates to an action and not to a state, and should be
interpreted 'they place'. Confurnation of this translation is provided by the use of
another verb in the passive voice, also from the same rubric of §342 : illuminantur.
This can only mean 'are lit', or expressed actively, 'they light'. Moreover, the use of
the periphrastic verb erunt compositae to describe the lamps prior to their being lit,
1
See Part IV Chapter 10 p.289. ·
2
'The seven lamps in front ofthe altar are kindled from it [the reed-candle]. They had
previously been placed unlit in such a way that they could be lit without any hindrance.
The great candle which is to be blessed is placed in the centre of the church and in front
of the altar.,' (The writer's italics.)
221
makes it very likely that erit positus 1 or some equivalent expression would have been
used instead of ponitur, if the Easter candle had already been in position before the strut
of the service.
We conclude that PRG contains sufficient information for the attestation of Stage
2 in the development of the use of the Easter candle, which is sununarised below.
Stage 1 : Easter candle lit outside church with new fire.

Easter candle carried into church in procession.
Stage 2 : Reed-candle lit outside church with new fire.

Easter candle carried into church unlit.
Stage 3 : Reed-candle carried into church lit with new fire.

Easter candle already in position inside church.
1
Literally 'will have been placed'.
222
PARTN
The Easter Candle and the Paschal Vigil

Chapter One
THE ORIGIN OF THE EASTER CANDLE
It is generally believed that the lighting and blessing of the Easter candle was a liturgi-
cal development of the Lucernarium of Holy Saturday. It not only ensured the survival
of that service, admittedly in an altered form; but without destroying the traditional
structure of the Lucernarium the incorporation of the new element transformed the old
ceremony, and, combined with the ritual of the new fire, became integrated into the Pas-
chal vigil to produce the liturgy of Holy Saturday .1 Elsewhere2 we have referred to the
elements in the Paschal vigil which survived from the Lucernarium. These were :
- the bringing in of the lamp

- the officiating deacon
- the offering of light to God3
The carrying of the lamp survived unchanged in the Milanese and Mozarabic rites,
and in other western rites as the bearing of either a candle, lit with the new fire and
placed on a pole, spear, or reed, or the Easter candle itself. It is the deacon who still
officiates at the service; and the offering of light is a feature of the formula for the
blessing of the Candle in all the western rites.
It is difficult to disagree with Gregory Dix's description of the Lucernarium as an

'originally utilitarian ritual' in which the purpose of the lamp was 'to give light to the
lector' .4 The functional use of the lamp or the candle, which later replaced the lamp in
most rites, was first pointed out by De Vert.5 However, a number of scholars have
denied the utilitarian origin of the Easter candle. Berliere, either unaware of the ante-
cedents of the Easter candle or choosing to ignore them, claimed that its purpose was
never functional because it could not provide enough light for illuminating the rest of
the church building; and that its origin was symbolic.6 Likewise Thurston, in an
1
The basic structure of new f'rre, blessing of Easter candle, Vigil-readings, and Baptism,
which still obtains today, had been achieved in some parts of Gaul by AD 800.
2
See p. 219 and p.287.
3
For the Dialogue and the Blessing of the Light, see Part N Chapter 10.
4
Shape p.23.
5
Paschal co1.328.
6
Le derge pascal p.107.
224
uncharacteristic lapse of scholarship, asserted that the ceremony involving the Easter
candle was designed from the beginning with a strictly mystical and symbolical mean-
ing.•
It hardly needs to be mentioned that it was unnecessary to illuminate the rest of the
church du.ri_ng the proclamation of the Exultet and the Preface. Indeed in the Roman rite
it was intended that the Exultet and part of the Preface should be chanted in semi-dark-
ness?
To argue that the beeswax candle replaced the lamp of the Lucernarium as the
source of light for the reader is correct to a point. It is true that the small taper or the
Easter candle itself replaced the lamp as the means by which the light was introduced
into church. In the Mozarabic and Milanese traditions the lamp continued to feature in
the ceremony. In the latter rite it was present at the blessing of the font. 3 In the Spanish
ceremony, however, it was carried into church along with the Easter candle, and con-
tinued to bum throughout the remainder of the Paschal liturgy, maintaining almost par-
ity with the Easter candle.
In the next chapter we will show that the Easter candle emerged as the principal
feature of the Lucernarium of Holy Saturday in the region of Northern Italy as early as
the fourth century. Before we survey the early evidence for the Candle, two questions
remain to be considered. (i) Why was the lamp of the Lucernarium replaced by a
candle of beeswax as the means of illumination for the reader? (ii) Why did this devel-
opment ftrst take place in Northern Italy?
Since the origin of the Easter candle is shrouded in obscurity, it is perhaps under-
standable why liturgical commentators in the past without exception have avoided
addressing themselves directly to answering the first question. It seems unlikely that
the Easter candle was borrowed from a pagan religious milieu with a comparable ritual,
or that it was suggested by such features of pre-Christian worship such as sky-pillars4 or
sacred trees. 5 The view that a large candle was used in order to provide a great light for
1
Lent and Holy Week p.408.
2
After the Vigil carne to be held in the earlier part of Holy Saturday, a sufficient amount
of natural daylight, especially on a sunny day, would render the flame of the Easter
candle unnecessary for the provision of light by which to read. (For the provision of
light at the Paschal vigil, see Chapter 15.)
3
1768 Missal p.125.
4
Cook, Zeus Vo1.2 pp.36ff.
5
Duval, Les Dieux de La Gaule pp.116-7.
225
the reader of the lessons merely side-steps the issue. For apart from the fact that there is
no evidence to suggest that in the fourth century the Easter candle was especially large,
a lamp of considerable dimensions would have been equally suitable for this purpose.
A clue to the solution of this problem may be provided by the internal evidence of
the songs composed in honour of the Easter candle, the Prefaces. 1 Now it is true that
there are prominent references and allusions to the Passover, to Baptism, and to the Res-
urrection in all the Prefaces; but the pre-eminent theme is the praise of the Candle itself
and the significance of this source of light. When we turn to study the two surviving
benedictiones cerei of Ennodius, we discover that in addition to the allusion to the bees
and the generation of beeswax, which is common to all the Prefaces, the composition of
the Candle is for him of profound significance. In both laudes he identifies the three
constituent elements of the Candle viz. the wax, the wick, and the light.
(a) species trino conpaginatae consortio societatis propemodum mysticae glutino

coniunguntur, quarum ceram paravit nectareis partubus feta virginitatis, papyrum
ad alimenta ignium lympha transmisit, lumen adhibetur e caelo.
(b) venerandis compactam elementis facem tibi, Domine, mancipamus in qua

trium copula munerum ... unum, quod de fetibus fluminum adcedunt nutrimenta
flammarum : aliud quod apum tribuit interemerata fecunditas ... ignis etiam caelo
infusus adhibetur. 2
For Ennodius it is not the physical elements in themselves which are important; it
it the hallowing of them by the direct intervention of God in the historical process of
redemption recorded in Scripture. For the wax, produced parthenogenetically by bees -
so it was believed - symbolised Jesus' birth from a pure virgin; the papyrus, which
served as a wick in the Candle, grew in river-water, the element hallowed by Jesus
through his own baptism; and the silent flame recalled the Burning Bush with its
foliage still intact. The Easter candle, therefore, represents the God-hallowed material
world, and by extension the whole of creation, which as a resUlt of the Incarnation and
'These are discussed more fully in Chapter 10.

2
(a) 'Elements, joined together in triple partnership, are united by the bond of an almost
mystical fellowship. The virgin-born bee has prepared the wax for her nectareous par-
turition; the water has produced the papyrus for the sustenance of the ft.re; and light is
admitted from the sky.' (b) 'We offer to you a torch composed of elements to be
revered, a union of three gifts : one, which from the plants of the rivers enables the fuel
for the flames to burn, and another, which the unsullied fertility of the bees provides.
Fire, sent from the sky, is also added.' The excerpts are from Preface I and Preface II,
respectively. Pinell, La Benediccio p.93 and p.95. (The writer's italics.) The text is also
in Hartel pp.415-22. In the latter passagefetibusjluminum, 'the produce of rivers',
refers to the papyrus, which was obtained from the River Nile and the marshes of Egypt.
226
the advent of end-time following the Resurrection, has now become potentially
redeemable. Therefore, the offering of the Candle, this microcosm of the world, would
symbolise and once more re-enact the union of heaven and earth in anticipation of the
fmal redemption of all creation at the close of the Age; and reinforce the eschatological
theme of the Paschal liturgy. This union of human and divine is twice referred to
explicitly in the Romano-Gallican Preface of the Tridentine Missal :
Nox in qua terrenis coelestia, humana divina coniuguntur.
Ut Cereus iste ... supernis luminaribus misceatur. 1
The latter reference, of which there is also an echo in the Beneventan Preface,2
underlines and reinforces the importance of the Candle; and the eschatological
significance of this particular source of light largely explains why a candle composed of
beeswax was used in preference to an oil-lamp, as the medium for the offertory of light
in the Vigil liturgy.
It is very likely that its shape was not an inconsiderable factor in the adoption of a
candle for the provision of light at the liturgy of Holy Saturday. For in contrast with a
lamp the column of wax provided a much more vivid and realistic symbol of the pillar
of fire that featured prominently at the Paschal vigil.
Any attempt to account for the Northern Italian provenance of the Easter candle
must remain speculative in our present state of knowledge. This part of Italy was not
especially noted for apiculture or for the production of a superior quality of beeswax.
1
'A night in which heavenly things are united to those of earth.' and 'That this Candle
may be mingled with the heavenly lights.'
2
Pinell, La Benediccio p.96.
227
Chapter Two
THE EASTER CANDLE : EARUEST REFERENCES
It would seem that the lighting and blessing of the Easter candle is a purely western
liturgical development; and the earliest references to the ritual strongly suggest a geo-
graphical provenance in Northern Italy. The earliest reference to the Easter candle
occurs in a letter, 1 written by Jerome in AD 384, to a deacon named Praesidius in the
church at Piacenza. Praesidius had previously asked Jerome to compose a carmen cerei
for him; but Jerome is unwilling to comply for two main reasons. His first objection is
to the style of language which he is expected to use in composing this song of praise.
Previous writers had used a florid form of language reminiscent of the fourth book of
Virgil's Georgics. 2 This is all very soothing on the ear, says Jerome, but it is not in
keeping with the office of a deacon, especially as on these occasions ecclesiastical
superiors are listening in silence to a minister who does not possess sacerdotal authority,
nor with the sacraments of the Church, nor with the season of Easter. His other and
principal objection is to the whole notion of the ceremony involving the Easter candle.
Anticipating in a way some of the reformers of the sixteenth century, he claims that the
ceremony and some of its features are unscriptural. 'Read the Old Testament', he
fulminates. 'There is no instance of the use of wax. 3 And where can you fmd a refer-
ence in the New Testament to a wax taper.' Elsewhere in the letter he refers to the
whole proceedings as a 'rather vulgar ritual'.
We must bear in mind, when assessing the evidence of Jerome, that in many ways
he was not typical of his age. Therefore, we must disregard his undisguised antipathy to
this ceremony and try to evaluate the evidence in an unbiassed way. Four conclusions
emerge. (i) At Eastertide in Piacenza there existed a ritual involving the use of a wax
candle, at which a carmen cerei was sung by way of a benediction. (ii) It had become
traditional to compose the song in a recognised poetic style. (iii) The allusions in the
carmen to bees and honey and beeswax, to which Jerome took so much exception,
strongly suggest that the content and theme of these fourth-century laudes cerei were
similar to those of the later extant examples. (iv) It was the officiating deacon who was
responsible for the singing, though not necessarily the composition, of the carmen, and
who occupied a position of prominence during the ceremony, a feature of the rite that
has survived to this day.
1
Epistola ad Praesidium: de Cereo paschali. PL 30.188-9.
2
Interestingly, Virgil was born near Mantua, only 55 miles from Piacenza.
3
The use of candles in Palestine was not widespread in pre-Christian times.
228
The existence of a recognised style of composition shows that the use of the wax-
candle at the Easter vigil was well-established at Piacenza, and almost certainly at
Milan into whose liturgical orbit Piacenza and other cities in Northern Italy came.
Indeed Jerome, who was himself born at Strido near Aquileia, is clearly familiar with
the ceremony, even if he found it distasteful. His sneer suggests that the rite was popu-
lar; and this may well indicate that it had been in existence for a number of years. Lack
of evidence, however, prevents us from assigning even a rough date to the first
appearance of the ceremony. Moreover, the situation is complicated by the fact that the
use of the Easter candle emerged within the context of the well-established Lucerna-
rium; so that even a date in the earlier part of the fourth century must remain conjec-
tural.
The next firm evidence for the ceremony comes from St Augustine. His own
words that he had written a short carmen in praise of the Easter candle 1 clearly shows
that he was familiar with this Paschal ritual and may well have actively participated in it
on one occasion. The composition of his taus cerei must have been undertaken before
AD 391 when he was a deacon, since in that year he was consecrated bishop. We are
not told where the carmen was sung; but it may have been on the occasion of his visit
to Milan, where we have suggested this Paschal ceremony was well-known. However,
it may have taken place in North Africa. It is unfortunate that we possess no evidence
for the Easter candle in that region, unless Augustine's words are applicable to North
Africa.
Indicative of the popularity of the ceremony and also of its early appearance in
Central Italy is the decree of Pope Zosimus (AD 417-18):
Per parrocia (sic) concessa licentia cereum benedici.2
It was generally3 believed in subsequent centuries, solely on the strength of this state-
ment, that the ceremony surrounding the Easter candle had been instituted throughout
the churches of the West, except Rome, by this pope. However, we have already
observed that the rite was well-established in Northern Italy in the fourth century. It
was assumed that paroccia referred to the dioceses outside Rome, and that the Church
1
Quod in laude quadam Cerei breviter versibus dixi. De Civitate Dei 15.22.
2
'Permission was granted to bless the Candle throughout the parishes.' (The writer's
italics.) Liber Pontifical is I p.225.
3
Amalarius, Lib.Off. 1.18.1; Sicardus, Mitrale PL 213.3238; PR XII I xxxii.8 p.240;
Durandus, Rationale VI.80 p.350.
229
of Rome at this time was powerful enough to influence and even sanction the liturgies
of other churches in the West. 1 The earliest evidence for the blessing of the Easter
candle within the Cathedral of StJohn Lateran dates only from the tenth century. 2
As additional evidence for the continuity, if not the provenance, of the ceremony
in Northern Italy, there survive the two laudes cerei,3 written at the beginning of the
sixth century by Ennodius who became Bishop of Pavia in AD 517. Pope Gregory also,
writing4 c. AD 595 to Marinianus, Bishop of Ravenna, has occasion to refer to the bless-
ing of the Easter candle as performed in that northern Italian city.
Evidence for the use of the Easter candle in the East is wanting. Feasey5 produces
no corroborative evidence to support his extravagant claim : 'The rite [of the Easter
candle] undoubtedly came from the East, either from Jerusalem or Antioch'. Dendy6
would see a possible origin for the Easter candle in the East. He adduces in support of
his suggestion the statement of F.C.Burkitt7 : 'The Saturday of Annunciation has been
the name of Easter Eve or Holy Saturday among the Jacobites ever since the middle of
the sixth century: no doubt the name is derived from some 'announcement' of Easter
tidings corresponding to the Western Exultet.' All this is clearly conjecture and guess-
work. Neither can attempts to see any connection between the Easter candle and Con-
stantine's lavish display of street illumination in Constantinople8 be taken seriously. It
is true that Egeria' s evidence is ambiguous; for she wrote c .380 of the liturgy of the
Church in Jerusalem that 'they keep the Vigil like us'. 9 If it could be shown that the
Easter candle was known in Spain, Egeria's probable homeland, in the last quarter of
the fourth century, it could be argued, as Thurston does 10 that it was a feature of the
Jerusalem liturgy at that time. However, there is no mention of the Easter candle by the
Christian Spanish poet Prudentius, who wrote some twenty years after Egeria's visit to
Jerusalem, in any of his poems; and in view of the fact that the Easter candle is so sig-
nificant a liturgical feature, whose symbolism readily lends itself to poetic composition,
1
For the attitude of the Church of Rome towards local customs elsewhere during the
pontificate of St Gregory, see p.l75. For other intetpretations of paroccia, see Appen-
dix 15.
2
PRG IT p.97. Interestingly, the institution of the Easter candle according to the 1507
Missal of Salzburg (fol.xciiii), which derives directly from PRG, is attributed to Pope
Gelasius (AD 492-6).
3
Pinell, La benediccio pp.92-95 (Hartel's text). Also in PL 63.258-262.
4
Epistle XI.33. PL 77.1146.
5
The Paschal Candle p.353.
6
The Use of Lights p.131.
7
In ITS (1923) p.425.
8
Eusebius, Vita Constantini 4.22. PG-20.1169.
~ilkinson, Egeria' s Travels p.138.
10
The Exultet p.514.
230
his silence would seem to be conclusive for its absence from the rites of Spain at that
time. The earliest finn evidence for the Easter candle in Spain comes from the year
AD633. 1
1
Fourth Council of Toledo, Canon 9 PL 84.369B.
231
Chapter Three
THE PREPARATION, COMPOSmON, AND COST OF THE EASTER CANDLE
Our sources tell us little about the preparation of the Easter candle. At Gembloux it was
set up after Sexton Holy Saturday. 1 However, at Barking Abbey the Candle was pre-
pared on Good Friday, and at Fleury on Monday of Holy Week. 2 At the two last-men-
tioned monasteries it was necessary to inscribe the insignia on the Candle in readiness
for the Vigil.
Traditionally the Easter candle has been made of beeswax, as the Preface follow-
ing the Exultet proclaims. However, the advantages of beeswax, a more pleasant odour
and a slower rate of combustion than tallow, are matched to some extent by its scarcity
and inevitably its cost. In the following chapter we shall show how the use of the Judas
to some extent curtailed excessive expense. At Lyon the use of a wooden or metal
Easter candle with a bougie inserted in a socket at its upper end was forbidden? How-
ever, in some of the large churches of Paris in the last century, two Easter candles were
used : one made of wax which was taken to the font, the other being a tin or wooden
candle-like fixture of great height placed in the choir, which was supported by a large
candlestick.4
The problem of obtaining candles of pure beeswax became acute during the last
century. For instance, in 1857 the Bishop of Charleston in South Carolina asked Pope
Pius IX for permission to use tallow candles because of the scarcity and price of bees-
wax. 'Let the recent malpractice of making candles from tallow be stopped, ' 5 came the
reply. However, a decree of the S.C.R., dated 14 December 1904, permitted the
addition of other substances to the beeswax provided that the proportion of beeswax
was in maxima parte.6 Even then, some authorities still insisted on a beeswax content
of 75%7 In the remote missions of Oceania the use of whale oil or blubber was
allowed.
1
Albers II p.99.
2
HBS 65 p.101 and Novarinus p.17, respectively.
3
4
Paschal col.333.
5
Van der Stappen p.92: Inductus abusus adhibendi candelas ex sevo eliminetur.
6
DACSR. Maxima pars was interpreted 65%.
7
Van der Stappen p.89.
232
In England during the later Middle Ages, mention of the charges for the prepara-
tion of the Easter candle is 'to be found in every book of church accounts' . 1
1
Gasquet, Parish Life p.181. A useful list of churchwarden's accounts is to be found on
page xi of that work.
233
Chapter Four
THE SIZE OF THE EASTER CANDLE
It is not unreasonable to suppose that from its earliest appearance within the Paschal
liturgy the Easter candle was always distinguishable from other candles by its size,
however modest the differential may have been. The flrst indication not simply of its
size but of its unusual size appears in the account of the fire that occurred in the Church
of St Stephen in Naples in the eighth century:
Ecclesia Salvatoris, quae de nomine sui auctoris, Stephania vocitatur divino - quod
flens dico - iudicio igne cremata est. Moris enim fuit, ut cereus sanctus, inormi
mensura porrectus, propter dominicae resurrectionis honorem a benedictionis
exordio usque ad alterius diei missarum expleta sollemnia non extingueretur.
Nocte igitur quadam ipsius festivitatis, cum solito dimitteretur accensus, cunctis
quiescentibus, ignis per aranearum forte congeriem in laquearia ipsius ecclesiae
pervenit, et sic demum aestuavit in omne aedificium. 1
Mallardo refers to the both the 'extraordinary length of the candle' and the height of the
surviving column which had supported the Candle, and which he presumably had seen.
However, even if the column does date back to the eighth century, there is no
corroborative evidence to substantiate the claim that the Candle which caused the fire
was of similar dimensions. The Easter candle at the not-too-distant Monastery of
Monte Cassino was quite small; yet being atop a very tall stone column,2 it could be
said to be 'inormi mensura porrectus' and was certainly impressive enough to honour
the Lord's resurrection. In later Neapolitan practice it was small enough to be carried in
procession by the cimiliarcha, the cathedral treasurer. 3 Moreover, it is unlikely that the
1
'The Saviour's church, which is named after its founder, Stephanus, was destroyed by
the fire of divine judgement. I weep as I write this. The Holy Candle stood extremely
high as a mark of honour for the Lord's resurrection. It was customary for it to bum
from the beginning of the blessing until the solemnities of the Masses on Easter Day
were completed. During the night of one Holy Saturday, when the Candle was
unattended as usual and all were resting, some drapery in the church caught fire; and
after the fire had reached the roof, the whole building was eventually engulfed in
flames.' Gesta episcoporum Neapolitanorum (Waitz's edition p.426). Text in
Mallardo, La Pasqua p.22.
2
Zamecki p.17.
3
The ceremonial of Holy Saturday according to the fourteenth-century Constitutions of
John Orsini, in Mallardo, La Pasqua p.33.
234
size of the candle envisaged by Mallardo had shrunk to quite the proportions of the
portable wax colwnn of Orsini's Constitutions, even allowing for a reduction in its size
as a result of the conflagration. 1
An interval of more than 200 years separates the incident at Naples from the
palaeographic evidence of the Exultet rolls of Southern Italy. Even making allowances
for some artistic licence in the execution of the detailed liturgical scenes. these ele-
venth-century and twelfth-century drawings record a considerable range in the size of
the various Easter candles depicted. The above-mentioned Candle at Monte Cassino
appears to measure between 18" and 24" in height,2 and those at Bari and Mirabella
about three feet; 3 whilst the Easter candle at Gaeta stands at about five feet from the
ground. 4 All the Candles depicted in the rolls taper towards their apexes. Candles
shaped in this way give the columns which support them a more aesthetically-satisfying
termination; their lower centre of gravity gives them greater stability; and their almost
pointed ends render the wicks more accessible for kindling and less prone to inoppor-
tune failure and embarassing extinction.
The practice of fashioning Easter candles of very large and, to the modem way of
thinking, excessive proportions is well documented throughout the later Middle Ages
and beyond, in both the cathedral and monastic (mainly Benedictine) liturgical tradi-
tions. Not all Benedictine houses, however, were extravagant in this respect. The
Roman Church encouraged the use of a Candle of considerable proportions.5 Feasey 's
statement6 that the Easter candle had of necessity to be of great size so as to last
throughout the night vigil, is manifestly untrue. Some candles of very small proportions
are able to bum for twenty-four hours.
At Lincoln the Easter candle weighed 42lbs,7 and those at Bury St Edmunds and
in the Lateran Basilica both contained 80 lbs of wax;8 whilst the use of desuper in the
rubric of the Ordinal of Barking in connection with the fiXing of the great candle in its
1
For us the cause of the fire is a matter for speculation. The drapery or curtains (ara-
nearum congeries - the meaning is uncertain) had caught fire and set the ceiling alight.
The cloth could just as easily have been set on fire closer to the floor as at a point near
the roof.
2
Zamecki p.17.
3
Avery Plates XII and LVI, respectively.
4
Avery Plate XXX.
5
The Caeremoniale Episcoporum prescribes: Praeparetur cereus Paschalis praegrandis
~ll.27.1). The writer's italics.
'Ancient English Holy Week Ceremonial p.192.
7
Bradshaw and Wordsworth ll p.291.
8
HBS 99 p.53 and Gavantus/Merati N p.l54, respectively. A solid beeswax candle of
this weight could have the dimensions of9'6" x 5".
235
candelabrum, seems to imply the hoisting and lowering of a great weight. 1 We are
informed that at Westminster in 1558 the Paschal column consisted of three hundred-
weight of wax? The massive Easter candle which Emperor Maximilian presented to
the church at Echtemach in Luxembourg in 1512 weighed 354lbs.3 A
seventeenth-century manual of Rouen, which recommended4 that the Easter candle
should be 'grandioris formae', was given a generous interpretation in that city. For at
the begi_nning of the eighteenth century it is recorded that not only was the Candle in
Rouen Cathedral twenty-five feet in height, but that those in the Churches of Saint-
Ouen, Notre-Dame de la Ronde, and Saint-Sauveur were also of a similar dimension. 5
It is recorded that the Easter candle at Salisbury stood thirty-six feet high;6 but
the claims made for the colossal height of those at Durham and Norwich cannot be sub-
stantiated. The Candle at the former cathedral was probably not as tall as is generally
believed; 7 whilst the assertion by Feasey that the Easter candle at Norwich was as tall
as that at Durham, that is, it almost reached to the roof of the cathedral, was based on
the mistaken belief that the circular aperture at the junction of the vault-ribs in one of
the bays was used for lighting the wick of the Easter candle. 8
Great care must be exercised when interpreting the statistical information relating
to the height and weight of Easter candles. For instance, we learn that the Easter candle
at Rouen stood twenty-five feet high and weighed forty pounds;9 yet that at Seville,
which stood only two feet higher, weighed 1500 lbs. 10 It is clear from a comparison of
these two sets of statistics that a number of factors must be taken into consideration in
assessing the authenticity and value of the recorded data, before any firm conclusions
can be reached. (i) There may be errors in the transmission of the text; or descriptions
may contain oft-repeated mistakes which have not or cannot be verified. (ii) Heights,
and sometimes weights, may be based on a visual assessment. (iii) Human nature is
often prone to exaggeration, especially in situations in which large numbers are
involved. (iv) Errors may occur when converting from one system of measurements to
another. (v) The constitution of the Easter candle must be determined. (vi) The
method of its manufacture and its composition must be known.
1
HBS 65 p.97.
2
Machyn p.169.
3
Passmore p.216.
4
1640 Ritual p.305.
5
De Moleon p.318 and p.32l.
6
1508 Processional p.73.
7
See also Appendix 8.
8
Ancient English Holy Week Ceremonial p.193.
9
Feasey, ibidem p.l93.
1
00oblado (p.299) records 80 arrobas. (l arroba = 18.75lbs avoirdupois).
236
Of the six above-mentioned points for consideration the last two probably account
for most of the seeming discrepancies and exaggerations. Strictly speaking, the Easter
candle consists only of a column of wax. However, a false candle-stock or Judas 1 may
have been added below the Candle so as to increase the height of the Candle, and in
some instances, to double its overall vertical dimension, thereby giving an impression of
considerable loftiness. On the other hand the candlestick may have been included in the
estimation of the height of the Easter candle, especially if the former was columnar in
design and of similar diameter to the Candle.
Two aspects of the manufacture of the Easter candle deserve our attention. (i) The
ratio of a candle's diameter to its height is disproportionate in that the halving of the
former does not result in the doubling of the latter. For instance, the height of a candle
measuring 9' 11" x 5" becomes 20' 0" when the diameter is reduced by only 11/ 2" to 3 1/ 2".
(ii) The manufacture of very tall candles, especially those with a diameter insufficient to
maintain the rigidity of the wax column, necessitated the insertion of a wooden2 or
metallic core for a large section of its length in order to ensure its continuing vertical
position. 3 It follows that the weight of the Candle would be substantially increased if
its core were a metal rod of considerable diameter.
The great height of some Easter candles is also conf"mned by a description of the
manner in which they were lit. At Seville a chorister climbed a gilt-iron mast which
stood close to the Easter candle. At its summit was a railed-in platform, similar to a
ship's crow's-nest, on a level with the top of the Candle. From this platform the Candle
was both lit and trimmed; and the melted wax was also drawn off with a large iron
ladle.4 At the Lateran Basilica a portable pulpit was wheeled into the church so that the
deacon could light the Candle; 5 whilst at Durham, where we are informed that the
Easter candle was square, a long pole was kept in the triforium above the choir for the
purpose ofkindling.6 At St Leonard's, Leau also, where the candleholder was 5.68m
high (18'7"), the deacon had to climb to the triforium for the same pwpose. 7
Since the Easter candle came to be fashioned in large dimensions either to repre-
sent the importance of the Resurrection or to symbolise the pre-eminence of the light of
Christ, it is not surprising that the size and sturdiness of the holder were increased to
1
See also below, pp.239-40.
2
As at Bourges (Paschal col.333).
3
1838 Ceremonial of Lyon p.479. The insertion of a core also had the advantage of
reducing the expense incurred by the use of beeswax, a costly raw material.
4
Doblado p.299.
5
Feasey, The Paschal Preconium p.254.
~e writer's interpretation of Fowler, Notes on the Rites of Durham p.9.
7
Callewaert p.l41.
237
counteract the increased weight superimposed thereon and to provide additional sup-
port. It also ensured that from an aesthetic point of view the Candle and the candlestick
maintained a satisfactory relationship with each other; and at the same time underlined
even more forcefully the importance of the person and the event it commemorated.
The Easter candle was either supported by a metal candelabrum, frequently of

exquisite craftsmanship, or it rested on the top of a stone column, often embellished
with suitably-appropriate sculptures and other decorations. 1 The use of the column to
elevate the Candle is first reliably attested at Monte Cassino in the eleventh century;2
but, as we have suggested at the beginning of this chapter, a column may have been in
use at Naples in the eighth century. There can be little doubt that its use in the Paschal
liturgy of the Romano-Gallican tradition was suggested by the reference in the Preface
following the Exultet to the pillar of frre 3 in the Book of Exodus, and by the identifica-
tion of the Easter candle with that pillar. 4 The significance of the column in this
respect is noted by Macri.5
The great height of the column6 or candelabrum, rendering the Easter candle
beyond of the deacon's reach, may in some instances account for the absence of the five
grains of incense in the ritual - at Durham possibly, for example - or explain why the
grains were inserted into the Candle during the preparations for the Vigil, as, for
example, at Barlcing. 7 However, the problems created by the use of a lofty support for
the Candle were not insuperable. The Gilbertine Ordinal and the Sarurn Missal both
permitted the choice of inserting the grains of incense either into the Easter candle or
into the candleholder; 8 whilst a Parisian handbook of ceremonial and a Polish manual
both enjoin that ladders should be provided for this ritual act. 9
Without specifying a maximum size for the Easter candle the Roman Church rec-
ommended moderation in smaller churches, chiefly, one suspects, for the avoidance of
unnecessary expense. Commentators on the Roman liturgy stated that the Easter candle
should appear larger than ordinary candles, and suggested a weight of between eight
1
Perhaps the fmest are those at St Paul's-outside-the-Walls in Rome, and in the Baptist-
ery in Florence.
2
Zamecki p.17
3
'Columnae illuminatione'.
4
' lam columnae huius praeconia novimus.' In the Ambrosian Preface we read : 'ecce
iam ignis columna resplendet.'
5
Hierolexicon p.142.
6
At St-Maurice, Angers it stood above 12' high. De Moleon p.60.
7
HBS 65 p.lOl.
8
HBS 59 p.40 and 1515 Missal fol.cxi, respectively.
9
1662 Ceremonial p.374 and 1819 Manual ll p.478.
238
and ten pounds. 1 Somewhat paradoxically the Easter candle in the Mother of Cathe-
drals weighed 80 lbs, as we have already noted. At Paris 12 lbs of wax was used;2
whilst in some churches it weighed 33lbs to commemorate the traditional age of
Jesus. 3 The above-mentioned Polish manual of Roman ceremonies also mentions the
use of a 'rather long rod' for lighting the Easter candle in the event of its being very tall.
However, since the candlestick is presumably included in the height, we can gain no
indication of the vertical dimension of the Candle itself. In the Cistercian rite, whose
ceremonial was characterised by austerity, the recommended weight of the Easter
candle was 3lbs.4 It need hardly be added that in many instances the size of the Easter
candle would have reflected the wealth of a church or monastery; and at times would
have been determined by the availability of beeswax.
Today the Easter candle is distinguishable from other liturgical candles by its size;
but both in height and in diameter it is generally of very modest proportions. In Great
Britain and Ireland the Easter candle rarely exceeds 36" x 2". In parts of France, how-
ever, the tradition of tall candles is still perpetuated to some extent, although the giant
columns of old are no longer to be seen. Of the thirty-six cathedrals for which the
writer has information, a Candle of at least one metre (39") in height is used in thirty,
and in eleven of these the Candle stands at l.Sm (58").5 The reduction in the size of the
Easter candle in modem times has rendered some of the tall marble columns obsolete.
Moreover, since the Candle is now borne into church in the procession from the new
frre, the placing of it on the top of the tall shaft during the ceremony is now impracti-
cable, as at Lyon. In that church a portable iron candlestick is also used for the ease of
carrying the Candle to the font for Baptism.
The Judas
In some churches a desire to increase the height of the Easter candle either to
enhance its significance, or to offset to some extent the need to use a small beeswax
candle, or simply to render it more conspicuous resulted in the utilisation of a device
known as the Judas. This was a shaft of wood,6 shaped and painted to resemble a
candle, and attached to the base of the Easter candle, thus forming a lower false stock
and so increasing the height of the Candle. The latter was attached to the Judas by
1
Desideri p.150; Gavantus/Merati IV p.154; DHCR I p.470 (about 10 lbs).
2
3
Grancolas p.318; Gavantus/Merati IV p.154.
4
1689 Ritual p.244.
5
Survey of 1984.
~e use of metal for this purpose is unknown.
239
means of a spike protruding from the end of the Judas, or by a spring of wire, 1 or, one
might reasonably suppose, by inserting the end of the Easter candle, neatly chamfered,
into a socket at the top of the Judas. Sometimes the lower false candle of wood was
coated with wax rather than painted. The Judas at St Mary-at-Hill in London in 1511
weighed 7 lbs?
There is much uncertainty about the origin of the term Judas. W. Cooke3 traced
the origin of the word to Hebrews 7:14 : Christ, typified by the Easter candle, sprang
out of (the tribe of) Judah. Alternatively, since it is said that the stocks of other candles
were also called Judases, 4 the connection, if it is biblical, is more likely to have been
with Judas Iscariot: his false nature and his sham relationship with Jesus.
1
As at East Cheap in the fifteenth century. Feasey, Paschal Candle p.364.
2
Peacock p.163. He refers to the destruction of these Judases in the early years of the
reign of Queen Elizabeth I (pp.106,163,164).
3
Cited by C. Wordsworth, Medieval Services in England p.l68.
"There is a reference in the accounts of St Christopher-le-Stocks in 1488 to the wooden
stocks for processional candles : 'vi judas staves for torches painted' (Freshfteld p.ll9).
The decorated Judases at Epworth in 1566, mentioned by Peacock (p.77), were prob-
ably stocks of this type, and not intended for use in supporting the Easter candle.
240
Chapter Five
THE POSmON OF THE EASTER CANDLE
Apropos of the position in which the Easter candle stood prior to the chanting of the
Exultet, two main traditions can be identified during the later Middle Ages. Both are
discussed regardless of whether or not the Candle was borne in procession. For if the
Candle had previously been carried in procession, it continued to be placed in the same
position it had formerly been set in, once it was no longer borne into church.
1. The Gallican tradition.
The placing of the Easter candle in front of the altar is implicitly attested·in three
Gelasian sacramentaries and in two other documents. 1 The relevant rubric, quoted at
the beginning of Chapter 6, which describes the arrival of the archdeacon ante a/tare,
leaves us in little doubt that the Easter candle was also in this position. Since it is also
known from five ordines 2 that there were seven lamps in front of the altar, it is not clear
whether the Easter candle was placed between them and the altar, or whether it stood to
the west of those lamps. However, the slightly later PRG, which also mentions the
seven lamps, expands the rubric by adding 'in medio ecclesiae' and clarifies the
description of the Candle's position by adding that the clergy and the people gather
around it.3 Presumably the seven lamps remained outside the circle formed by the con-
gregation.4 A number of documents state that the Easter candle is placed in front of the
altar without indicating whether the location was in the sanctuary or in the choir. At
Soissons it was placed in the sanctuary ,5 and according to the Cistercian rite it stood on
the sanctuary step.6 It is possible that in some instances the rubric was left deliberately
1
GeV p.68; GeAng p.52; GePr p.55; OR 30A.15; Gradual of St Gregory, cited in
Schelstrate IT p.142.
2
0R 26.9; OR 28.30; OR 29.17; OR 31.13; OR 32.5.
3
The suburbicarian variant of PR XII has a similar phrase qualifying ante a/tare
(Ip.240).
"The twelfth-century Ritual of Soissons would appear to be exceptional in this respect
(DAER 4.24 p.161, M 305). According to this service-book the Easter candle stood in
the middle of the sanctuary, and the seven-branched candlestick was placed on the sanc-
tuary step (ad ascensum presbyterii). It is unlikely that the middle candle of this meno-
rah and the Easter candle were one and the same, since we are informed that a lectern
was placed next to the Candle; but its position between the altar and the menorah poses
the interesting question of how visible the Easter candle and the officiating deacon were
to the congregation, assuming that the seven-branched candelabrum was of no mean
f.roportions. The rite of Soissons was unusual in other respects- see p.l29 and p.l65.
See Note 4 above.
6
1689 Ritual p.244. In fact the middle step from which the Abbot gave his blessing.
241
vague so as to allow some flexibility of position. Martene, writing at the beginning of
the eighteenth century, mentions the sanctuary steps as one of the possible locations for
the Easter candle. 1
The practice of placing the Easter candle in the centre of the choir survived in
France well into the nineteenth century, as Table 47a shows; and was re-established in
the Roman rite following the liturgical changes of 1955. It is usual today to leave the
Candle in the choir only for the duration of the Vigil; and subsequently to remove it to
the sanctuary or to the ambo if one exists. However practice is flexible. Sometimes the
Candle remains where it was blessed. At times it is placed conveniently near the font to
be lit at the ministration of Baptism.
2. The Italian tradition.
Although the practice was not confined exclusively to the churches of Italy - see
Table 47b - the placing of the Easter candle next to the ambo almost certainly had its
origin in that country. The earliest evidence is palaeographic : the Exultet rolls of
Southern Italy provide graphic testimony for the practice from the tenth to the twelfth
centuries. 2 Indeed, the medieval ambo, together with its column or holder for the Easter
candle, survives in a number of Italian churches, especially in the South. The choice of
this position was perhaps inevitable, since the prophecies, an important element in the
Paschal vigil, were read from this raised platform; so that the juxtaposition of the
Candle and the Bible emphasised the link, and visibly expressed the close rapport that
existed between the Light of Christ and the Word of God.
Documentary confirmation for the ambo-position of the Easter candle is found in

an eleventh-century Beneventan ritual,3 and in the twelfth-century Monte Cassino A
and Monte Cassino B. Its position by the ambo is also enjoined by PR XII and other
Roman service-books, including the Roman Missal of 1570, and by Gueranger in the
nineteenth century, even though in some churches custom was at variance with official
directives. For the Pian Missal of 1570, though in theory mandatory for most churches
which acknowledged the primacy of the see of Rome, did not make recognition of or
allowance for the fact that not all churches possessed an ambo. Ordo XIV, Ordo XV,
and the 1474 Missale Romanum make mention of the ornatum pulpitum on which the
missal was placed and next to which stood the Easter candle. The term almost certainly
refers to the ambo, which in many Italian churches was elegantly constructed and
1
DAER 4.24.8 p.147.
2
Avery Plates XII, XXX, XLIV, LXXVIII.
3
0dermatt p.273.
242
Church/ Source Church/ Source
Document Document
PRG II §346p.98 Mende 1766 Missal p.199
Corbie PL 78.336B St Bertrand 1773 Missal p.209
Alcuin PL 101.1216B Tours 1784 Missal p.191
Paris 1662 Cerem. p.374 LaRochelle 1835 Missal p.186
Maison du Roy Sem.Sainte p.401 Autun 1845 Missal p.238
Sees 1742 Missal p.185
Table 47a. The Easter candle in the centre of the choir.
Beneventum Odermatt p.273 OrdoXIV PL 78.1218C

Milan Beroldus p.11 0 OrdoXV PL 78.1322A
Rome (12C) PR XIII p.239 MR 1474 HBS 17p.175
Monte Cassino A PR XII I p.292 Strasbourg DAER 4.24 p.162
Monte Cassino B PR XII I p.293 Chalons 1748 Missal p.178
Apamea DAER 4.24 p.160
Table 47b. The Easter candle at the ambo
Salisbury HBS 91 p.22 Soissons 1745 Missal p.163

Dominicans 1482 Missal fol.89 Carcassonne 1749 Missal p.194
V allombrosa 1503 Missal fol.xcii Seville Doblado p.299
Carnaldolese 1634 Cerem. p.82 Coutances 1825 Cerem. p.329
Amiens 1745 Missal p.182 Verdun 1832 Cerem. p.313
Table 47c. The Easter candle at the Gospel side of the altar.
Perigueux 1782 Missal p.158 Metz 1829 Missal p.158

LePuy 1783 Missal p.159 Toulouse 1832 Missal p.205
Table 47d. The Candle in the choir or at the Gospel side.
Reg.Conc. PL 137.494C Mainz 1507 Missal fol.xcii

Gembloux Albers II p.99 Aquileia 1519 Missal fol.91
V allombrosa Albers IV p.249 Cistercians 1689 Ritual p.244
Rome (Suburbic.) PR XII I p.240 Angers De Moleon p.80
Soissons DAER 4.24 p.161 Cahors· 1760 Missal p.172
Naples Mallardo p.33 Viennat 1782 Missal p.214
• or outside the sanctuary
t outside the sanctuary
Table 47e. The Easter candle in front of the altar.
Tables 47a-e
exquisitely adorned, and not to the legile or reading-desk, which was used of necessity
for the Exultet and prophecies in churches which possessed no ambo. At Milan the
ambo was known as the tribunum (Beroldus p.110).
243
3. The later development.
In the same way that the Book of the Gospels is regarded in the Orthodox tradition
as a verbal icon of Christ, the column of wax and the flame of the Easter candle that
symbolised the Word of God in the Trinity came to be identified closely with the
reading of that Word in the Gospellection. Hence, at the Gospel of the ftrst Mass of
Easter, the customary honoriftc lights accompanying the written Word were, and indeed
still are, dispensed with, since the Light of the Gospel was held to be visibly present in
the flame of the Easter candle. In churches which possessed an ambo from which the
Gospel was read, this had always been the practice. Elsewhere, the Gospel was read to
the north of the altar - to its left as one faces it. The earliest evidence for the Easter
candle being set in this position is found in the Sarum rite of the thirteenth century. 1
Attestation of the practice in the Dominican rite, though from a later period, would
suggest that the ceremonial of the Black Friars had influenced the English rite in this
respect. 2
The Gospel position is also found in Spain at Seville, and was also a feature of the
Mozarabic rite; 3 and Baldeschi and De Vert both mention that this was the usual
position for the Easter candle.4 The position is also found in a number of French
diocesan rites, some permitting the alternative position of in medio choro. Paschal,
commenting on this very aspect of the ritual, states that the position on the Gospel side
was found in country districts.5 It is likely that in many small country churches a
central position in the choir was not always suitable. This consideration may have been
in the minds of the compilers of the Ordinal of Nidaros, which prescribes that the Easter
candle should be set 'in a convenient place'.6
The Easter candle was also set at the Gospel side of the altar at the conclusion of
the Preface at Limoges,7 at Milan,8 and at Biasca and other Swiss towns which followed
the Ambrosian rite.9 In some churches, of both the Roman and Milanese traditions, in
'See Table 47c for reference and other instances of this position.
~e Sarum rite contained other Dominican influences, for example, the reckoning of
Sundays after Trinity, rather than after Pentecost as in the Roman rite.
3
Martene, DAER 4.42 p.l64.
4
Ceremonial p.271 and Explication IV p.l33.
5
La Liturgie col.329.
6
0NE p.232. This directive may partly have been determined by the existence ofsmall
churches in the more remote parts of this far-flung diocese, which in addition to the
whole of Norway included Iceland, Greenland, and the Isle of Man.
7
1830 Missal p.227.
8
Beroldus pp.lll-2; King, LPS p.359.
~g. Holy Weekp.98.
244
which the use of the ambo fell into desuetude, the Easter candle was set in this position
before the start of the service, 1 and the Exultet was proclaimed and the prophecies read
from a sanctuary lectern.
Paschal also mentions that the Candle was sometimes placed at the Epistle side of
the altar. 2 Examples of this position occur at St Mary-in-Trastevere in Rome, and in
the Florentine baptistery. Other authors record only slight variations of position. De
Vert refers to the Easter candle at the altar rail in some churches; and De Moleon
informs us that at Rouen Cathedral the Candle formerly stood between the Tomb of
Charles V and the three sanctuary lamps?
1
Baldeschi p.271 and King, LPS p.359.
2
3
Explication IV p.133 and Voyages p.318, respectively.
245
Chapter Six
THE EASTER CANDLE AND THE SIGN OF THE CROSS
(i) The sign of the cross made with a gesture of the hand
The consecration of the Easter candle is first attested in the eighth-century Gelasian
Sacramentary :
Veniens archidiaconus ante altare, accipiens de lumine quod vi feria absconsum

fuit, faciens crucem super cereum, et illuminans eum Et completur ab ipso
benedictio cerei. 1
This rubric is found with only slight variations of wording in two other Gelasian
sacramentaries and the Sacramentary ofReirns,2 and in two other documents. 3That the
act of consecration was achieved by making the sign of the cross is not in doubt; what
is not clear is the manner in which this was done. Rabotin, followed by Capelle,
suggested that it was achieved either by the archdeacon's hand raised in the act of
benediction, or with his hand holding the small candle which had been lit from the
reserved fire and with which the Easter candle was kindled. 4 Both cite the practice at
Poitiers5 c .AD 900 and the custom observed by a number of bishops of the Greek
Orthodox Church of performing this act of benediction with a lighted candle in their
hand. 6 However, the absence of any reference in any of the above-mentioned
documents to a small candle and to the sign of the cross being made in this way make
this argument difficult to sustain; although the silence of the rubrics is not conclusive in
view of the fact that there are other details of the ritual which are unrecorded.
There is no good reason to doubt that the primitive practice was to make the sign
of the cross with a simple gesture of the hand. An alternative reading in Ordo 30A adds
manu sua to the phrasefaciens crucem super cereum by way of clarification. 7 How-
1
'The archdeacon comes in front of the altar, takes a light from the fire that was hidden
on Good Friday, and, after making the sign of the cross over the candle, lights it. Then
he completes the blessing of the Candle.' GeV, Mohlberg p.68 §425. The document is
a Frankish recension of a Roman book, and the ceremonial described here relates to the
Gallican Church.
2
Angouleme p.52; Prague p.55; PL 78.336B, respectively.
3
0R 30A.l5; and the Gradual of St Gregory, cited by Schelstrate II p.142.
4
Les Grains d' Encens p.224 and Le rite de cinq grains p.5, respectively.
5
Poitiers p.215.
~abotin, however, had reservations: 'J'hesite cependant a interpreter Je geste Gelasien
f.ar la rubrique poitevine' (p.225).
OR 30A.l5, Manuscript R.
246
ever, it is possible that the latter phrase without the additional manu sua in some
instances came to be interpreted as an injunction to incise a cross into the wax of the
Easter Candle. For (a) the incision of a cross did eventually replace the tracing of one
manually in most churches; and (b) the very fact that the adjectival phrase in the
above-mentioned reading of Manuscript R is qualified by manu sua may suggest that
super cereum was capable of two interpretations. In classical Latin super bore the
meaning of 'above', 'over', or 'on top of' regardless of whether or not physical contact
was involved. 1 The preposition in on the other hand was normally employed to convey
the notion of one object resting upon another. In later Latin super was regularly used to
express the higher position of one object in relation to another with physical contact, as
in super virgam, super caput, and super candelabrum. Even in modem French sur may
mean 'on' and 'above' or 'over'. The incision of a cross on the Easter candle, therefore,
may have been partly the result of this somewhat ambiguous rubric. The writer has sug-
gested below an instance of where the Candle may have been marked with a visible
cross.
In the rubric quoted at the beginning of this chapter there is no indication of any
formula of benediction before the Easter candle was lit and prior to the chanting of the
Exultet. Now in addition to this initial consecratory gesture the Sacramentary of Prague
also records that the sign of the cross was made with the hand once during the Exultet
and once during the Preface.2 The performimg of the two additional gestures raises the
possibility that the initial cross was actually cut into the surface of the beeswax, the two
subsequent gestures serving to reinforce or confirm that which had been visibly incised.
PRG with its synthesised ceremonial records a hand-gesture over the candle at the new
fire (IT p.95 §342) and an incision prior to the lighting of the Candle (p.97 §346) fol-
lowed by a hand-gesture at incensi sacrificium vespertinum during the Preface (p.98
§347).
Although the incision of the cross in the Easter candle was adopted by most
churches, the tracing of a cross with the hand in the air survived at Vallombrosa,3 and at
Basel4 where a choice of performing or omitting this act of benediction was given dur-
ing the Preface. On the other hand at Nidaros in the early thirteenth century the deacon
was forbidden to raise his hand during the Preface,5 either because this action was
1
The use of super in the latter sense is found in Gregory's letter to Marinianus : preces
~uae super cereum ... dici solent. PL 77.1146.
At sancti + huius and cereus+ iste, respectively. GePr p.55 §95. The Sacramentary
of Gellone also attests the latter gesture. GeGe p.95 §678C.
3
1503 Missal fol.xcv : at in honore nominis tui ofthe Preface.
4
1488 Missal fol.xcii: at cereus iste sit benedictus of the Preface.
5
0NE p.233. At the words cereus iste sit benedictus.
247
considered to be a superfluous duplication of the act of benediction, or perhaps the dea-
con was held to be acting ultra vires in appropriatin g to himself the sacerdotal authority
of which the making of the sign of the cross was expressive.
(ii) The sign of the cross made by incision
In the same way that there is some uncertainty over the interpretation of the phrase
super cereum, so there exists some doubt about the precise meaning of the preposition
in as it occurs in the phrases in cereo and in eo [=cereo]. For in classical as well as in
medieval Latin this preposition may be translated into English either as 'in' in the sense
of 'existing within' or 'inserted in', or as 'on' in the sense of 'lying on' or 'on the sur-
face of'. The sign of the cross made in cereo in the latter sense implies either a tracing
with the fmger or thumb which leaves no visible mark upon the wax, or a cruciform
anointing with oil or chrism which leaves a visible cross on the candle, however feint,
against the pale background of the beeswax. A cross made in cereo in the former sense
ensures that a permanently visible mark will be incised into the candle.
In the tenth-century Antiphonary of Leon we read : (After he has been given the
candle) 1
Faciens episcopus in ipso cereo A .e.t Q
Now whist the possibility should not be ruled out completely that this cross was traced
with chrism, the addition of the two Greek letters would suggest that both the cross and
the letters were designed to be visible to the faithful, once the Candle had been mounted
in a prominent position, and that all three markings were incised into the ~ax of the
Candle. This view is reinforced by the omission of any reference to the use of chrism in
the rubrics, which are quite detailed.
According to the contemporary Pontificale Romano-Germanicum a cross was the

only marking on the Candle? Again there is no mention of chrism, and its use for the
purpose of making the cross seems unlikely. It is tempting to see the rubric as evidence
for the incision of a cross as opposed to one effected by chrismation, in view of the con-
temporary practice at Corbie, described below, and in view of the subsequent practice of
the Roman Church, as recorded in later service-books. 3 It must be borne in mind,
1
Page 280. 'The bishop, making this cross on the Candle .... ' For the use of this cross
elsewhere, see Chapter 9 p.276. ·
2
Archidiaconus .. facit crucem in eo, 'The archdeacon ... makes a cross on it.' PRG IT
~346p.97.
PR XII I xxxii.lO pp.240-1 (Suburbicarian variant); OEL p.60.
248
however, that liturgical practice in the tenth century varied markedly even within the
commes of a small area; and the later Roman evidence of the twelfth century should
not necessarily be regarded as corroborative for the practice of two centuries earlier.
The tenth-century Sacramentary of Corbie also records :
Faciant crucem de incenso in cereo et scribatur annus Domini.

atque A et 0. 1 .
In the next chapter we will argue that a cross had been incised prior to and in readiness
for the reception of the five grains of incense. At Gembloux in the eleventh century it is
difficult to believe that the pre-traced cross on the Candle was not incised? Similarly,
the Customary of the German Monasteries and Alcuin both record the presence of a
cross on the Candle, which is almost certain to have been incised.3 Apropos of the
above-mentioned Mozarabic evidence from l.£on there is no mention or suggestion of
the use of chrism for this purpose; and the incision of other marks on the Candle makes
it almost certain that the cross had also been engraved.
The earliest indisputable references to the incision of the cross are to be found in
the two Roman documents referred to above. That contained in PR XII relates to the
ceremony as observed in the suburbicarian churches of Rome. They state :
cum stylo facit crucem in ipso cereo (PR XII) and

4
(erigat cereum) scribatque in eo crucem (OEL).
In both instances it is the duty of the officiating deacon to carve the cross during the
ceremony and in the presence of the congregation, in the same way that the sign of the
cross was made by the archdeacon according to the Gelasian sacramentaries, or by the
bishop at Leon. According to an alternative tradition the cross was incised before the
ceremony began. This is attested in three of the documents we have already met,5 and it
can also be inferred with confidence at Rouen in the eleventh century, and in those
Benedictine monasteries which followed strictly the regulations prescribed by Lan-
1
' Let them make a cross on the Candle out of incense, and let the year of the Lord be
inscribed and the letters A and 0.' PL 78.336B
2
Albers n p.99.
3
Albers V p.38 and PL 101.1216B, respectively. Both date from c.AD 1000.
4
'He incises a cross in the Candle with a stiletto' and '(let him set the Candle in posi-
tion) and let him inscribe a cross on it.'
5
Viz. Sacramentary of Corbie, the Customary of Sigibert, and the Customary of the Ger-
man Monasteries. (PL 78.336B, Albers IT p.99 and V p.38.
249
franc. 1 At Barkin!f at the beginning of the fifteenth century the cross was incised on
Good Friday along with the chronological information. This would suggest that it
existed as much to provide guidelines for the insertion of the grains of incense as to
convey to an onlooker the fact that the Easter candle had been consecrated. For the
cruciform arrangement of the five grains now provided this indication of the Candle's
consecration, and so rendered the incision of a cross a preliminary act prior to the inser-
tion of the grains. Moreover, the practice of gouging the five holes in the Candle for the
reception of the grains made the cutting of a cross completely dispensable even for the
purpose of providing guidelines.
In the Milanese rite the presence of a cross upon the Easter candle is first attested
in the Ambrosian Missal of 1560 (fol.l13). There is no reference to it in earlier missals
of that rite. Since it was inscribed before the start of the ceremony and mentioned only
in conjunction with the insertion of the grains of incense - which in that rite then
occurred after the conclusion of the Preface - its primary purpose may have been
functional : to indicate the spatial disposition of the grains. 3
The surviving evidence would therefore suggest that the incision of a cross on the
Easter candle was a development of the earlier practice of tracing a cross with the hand
in the air, a concretisation of the manual gesture, analogous perhaps to the materialisa-
tion of the incensation of the Easter candle, which resulted in the implantation in the
wax of the five grains of incense in a cruciform arrangement.4
(iii) The sign of the cross made with fire
In the Pontifical ofPoitiers of c.900 we read (p.215):
Qui durn propter cereurn venerit, inclinans se ad altare et ter signum crucis de ipsa
harundine cum candela accensa faciens contra cereurn, inluminat eum.5
1
Acta Vetera, PL 147.176C and Decrees, PL 150.4460, respectively. According to both
documents the insertion of the grains of incense formed part of the preparation of the
Easter candle.
2
HBS 65 p.lOl.
3
At Milan the marking of the cross was probably borrowed from the Roman rite, as
were the five grains of incense.
4
See Chapter 7 especially pp.255-60.
5
'When he [the deacon] has approached the E~ster candle, he bows towards the altar,
and making the sign of the cross three times with the candle on the reed in front of the
Easter candle, he lights it.'
250
Dom B .Capelle had no hesitation in fmding the antecedents of this practice in the rub-
rics of the Gelasian Sacramentary. His claim that 'll est manifeste que le Pontifical
s 'inspire du Sacramentaire' is based on doubtful linguistic similarities between the two
sets of rubrics, and on the fact that in both rites there was no physical contact between
the consecrator of the Candle and the Candle itself. Inevitably there will be elements
common to both rites, such as the tracing of the cross in the air and the lighting of the
Candle; but liturgical development that may have taken place over the period of about
150 years which separates the two documents, and the incorporation of elements, such
as the serpent-shaped candle, drawn from different religious and cultural milieu, should
make us very wary of speaking of the dependence of the Pontifical upon the Sacramen-
tary in this respect.
Although the practice of placing a lighted candle on a reed is also found in three
earlier ordines, 1 in the contemporary PRG, and in many later documents,2 Poitiers alone
records the use of this candle to trace a cross in the air in front of the Easter candle as a
gesture of consecration. It is true that the rubrics of the Pontifical are much more
detailed than those of the ordines; but in view of the silence of the latter documents on
this point, we must conclude that the use of the lighted candle for forming the sign of
the cross was unique to Poitiers.
(iv) The use of chrism for the sign of the cross
The marking of the Easter candle with a cross of chrism is found in two manu-
scripts relating to the Beneventan rite : the late tenth-century Missale Antiquum and a
mid-eleventh-century gradual. 3 They both record that a cross was traced on the wax
after the Candle had been lit, but before the cry of Lumen Christi and the singing of the
Exultet. The practice is also found in the Diocese ofValence4 where the Easter candle
was anointed shortly before the insertion of the five grains of incense. It is just possible
that a survival of this ritual action is to be found in the rite of the Church of Lyon.
1
Viz.OR 26.9; OR 28.26; OR 29.15.
2
See Table 40 on p.193.
3
Beneventum MS Vi.33 and Rome MS Vat.lat.10673, respectively. The text describing
the kindling of the new fire and the lighting of the Easter candle is given in R.-J.Hes-
bert, Antiphonale Missarum p.188. H.M.Bannister, who claimed (Miscellanea Ceriani
p.135) that the Gradual was a fragment of an ordo ambrosianus, which preserved the
ancient usage of the Milanese Church, has been shown to be in error (Hesbert, ibidem
p.189). Hesbert's conclusion was apparently unknown to Dendy (The Use of Lights
p.l32).
1504 Missal fol.lvi.
251
There the grains of incense were first anointed with chrism before being inserted into
the Candle. However, in the following chapter we will suggest that this may have been
done for another reason (p.265).
(v) The origin and significance of the chrismal cross
It is not difficult to see in this act of chrismation a situation analogous to the

consecration of a church or altar in the Romano-Gallican rite, which is attested as early
as the tenth century, in which the anointing of inanimate objects formed part of the cer-
emonial; 1 and just as the anointing of the altar and the walls of a church was commem-
orative of the blessing which they had received, so the chrismation of the Easter candle
recalled its consecration by the deacon, and corresponded to the cross either incised into
the wax or delineated by the grains of incense in other western traditions. At the same
time the anointing of the Candle must also have been interpreted in a powerfully sym-
bolic way in that it reinforced the intimate connection between the Easter candle and
Baptism, recalling the chrismation of the baptizands.2
Capelle would find a reference to the origin of this practice in Prudentius' poem
Inventor rutili, dux bone,luminis,3 which dates from the late fourth or early fifth cen-
tury, and which was sung at the daily, or possibly weekly, office of the Lucernarium.4
He cites lines 155 and 156 :
Lumen quod famulans offero suscipe (1.155)

Tinctum pacifici chrismatis unguine (1.156)
and comments 'The fine candle, which is going to illuminate the evening prayer, now
that the sun's rays have disappeared, has therefore first been anointed with "peaceful
chrism". This blessing forms part of the ritual, for it is quite difficult to understand this
text in a purely symbolic and spiritual seiJSe.
1
PRG I: §52 p.144 for the anointing of the altar; §57 p.145 for the anointing of the
walls of the church.
2
1nterestingly, the Ambrosian Preface contains a reference to baptismal chrismation at
the point where the grains of incense were formerly inserted: Christi vero populus
insignitur fronte ... chrismate (1934 Missal t.e. p.40 of Repertorium). This invites specu-
lation that the word chrismate may once have functioned as an internal rubric, and the
anointing of the Candle may have taken place at this juncture.
3
Hymn V of the Cathemerinon. Cunningham's edition pp.23-28.
4
For a discussion of the Paschal elements in this hymn, see Appendix 12.
252
'Let us mark the first stage: anointing- obviously in the form of a cross- with
holy chrism on the wax itself. It concerns the ordinary candle. All the more does the
same apply to the Easter candle. ' 1
That the couplet refers to the office of the Lucernarium and in particular to the
deacon's offering of light, as Capelle says, there can be little doubt; but Capelle's inter-
pretation of the second line is open to question, and in particular the precise nature of
the deacon's offering. Moreover, he assumes that the Lucernarium, of which the
Spanish-born Prudentius treats, is that form of the service as found in Spain; whereas
there is no firm evidence to support a claim that the form of this service which Pruden-
tins had in mind was exclusively Spanish? There are a number of reasons why Capel-
le's conclusions should be challenged.
(i) It is generally believed that the office of the Lucernarium derived its name from
the fact that the lighting of a lamp (lucerna) and the offering of the light of that lamp
formed the central features of that service. If this office was celebrated daily and if the
lamp was anointed at each celebration, the use of a new lamp each day is implied, since
the act of chrismation in all liturgical situations is normally performed only once.3 This
difficulty and the belief that chrismation took place forced Capelle to assume that a
fresh candle was used each day for this service. Although the use of a candle at the
Lucernarium should not be ruled out, his description of the daily candle as 'beau' is
somewhat extravagant.
(ii) It is unnecessary to translate lumen as 'candle' in line 156. It is true that the
word may refer to the object from which the light emanates, and in later transalpine
Latin it is synonymous with cereus. 4 However, the translation of lumen in the couplet
as 'lamp' or as 'candle' focuses undue attention on the participial phrase which com-
prises the whole of line 156. This phrase will then refer to the consecratory cross of
1
Le beau cierge qui va illuminer Ia priere du soir, maintenant que se sont ereintes les
clartes du jour, a done ere prealablement oint du "chreme pacifique". Benediction
rituelle, car il est bien dijficile d' entendre ce texte au sens purement symbolique et
spirituel.
Marquons ce premier jalons: onction- en forme de croix evidement- avec le saint
chreme, sur Ia eire elle-meme. II s' agit du cierge ordinaire. A plus forte raison en
est-il ainsi du cierge pascal. (Le rite de cinq grains p.4).
1bus Capelle. His discussion of the blessing of the Easter candle according to the
Mozarabic rite in the very next paragraph begins 'En Espagne encore ... ' Admittedly,
Prudentius was a Spaniard by birth and wrote poems about Spanish martyrs; but
Capelle overlooks the possibility that what was true of the Lucernarium in Spain may
well have been applicable to the form that the office took in Italy and in Gaul.
3
In Baptism, Conf"mnation, Ordination, and Extreme Unction; in the consecration of
churches, altars, and bells; and in the blessing of the Easter candle.
4
Andrieu, Les Ordines Romani II p.266.
253
chrism and will bear the meaning of 'anointed with the unction of peace-bringing
chrism'. To the present writer this seems to place the emphasis of the couplet on the
fact that there is a chrismal cross on the lamp (or candle) rather than on the fact that the
deacon is offering a lamp (or candle). If, on the other hand, lumen bears its primary
meaning of 'light' in the sense of the emanation of illumination, tinctum may then be
rendered 'diffused' and unguine will then refer to the impregnation of the air with bal-
sam (chrismatis). This is the language of poetry; and in this couplet we surely have a
reference to the use of scented lights in divine worship, attested elsewhere by
Prudentius and by other writers. 1 It would seem, therefore, that Capelle's attempt to
establish a fourth-century precedent for the anointing of the Easter candle is unwar-
ranted.
1
Peter of Paris mentions a porphyry candelabrum which was brought up from the font
and carried a vase of gold, from which a wick of amianthus, set in balsam, 'diffused a
great light'. (Cited by Gaillard, Holy Week p.llO.)
The use of scented candles is mentioned by Prudentius :
nectar... guttatim /acrimis stillat o/entibus, *
and by Paulinus :
Lumina ceratis adolentur odora papyris, **
and is almost certainly referred to by the same author in :
Sed quis odor nares al/abitur aethere manans
Unde meos stringit lux inopina oculos. ***
Later evidence for the use of scented candles is to be found in Gregory of Tours'
description of the baptism of Clovis :
Balsama diffunduntur, micant fragrantes odore cerei. ****
* Cathemerinon V, PL 59.820A.
** Poem XIV, PL 61.467B.
*** Poem XXV, PL 61.6370.
**** Historia Francorum ll.31. PL 71.226B.
254
Chapter Seven
THE FNE GRAINS OF INCENSE
(i) The development of the ceremony
The practice of inserting five grains of incense into the Easter candle in the shape of a
cross is first reliably atte.sted in the tenth-century Sacramentary of Corbie :
Faciant crucem de incenso in cereo (et scribatur annus Domini, atque A et 0). 1
Doubts about the interpretation of the phrase de incenso have been expressed by
Capelle? He argued that de incenso bore the meaning of 'with the fire'. He claimed
that incensum here retained its primary meaning of 'fire', and instanced the use of the
word with this meaning in two phrases, one in the Preface following the Exultet and the
other in the prayer, Veniat quaesumus. 3 In both instances it had undergone a shift in
meaning from 'f"rre' to 'incense'. He also cited, as corroborative evidence, the practice
at Poitiers in the ninth century of tracing a cross with a lighted candle in front of the
Easter candle.
The interpretation of incensum as used in its original meaning in the Preface and in
the Veniat quaesumus is generally accepted by scholars and is not in dispute here.
However, it is difficult to accept Capelle's interpretation of de incenso within the con-
text under discussion and to believe that the sign of the cross was made with a lighted
candle, as at Poitiers. For this adjectival phrase will then have to bear the meaning of
'with the fire', a translation of de which is, to say the least, forced. Similarly the phrase
in cereo will have to be rendered 'in front of the Candle' instead of 'on the Candle'.4
Apart from the fact that the clause appears to have the literal meaning of 'let them make
a cross of incense on the Candle', the close association of this rubric with the directive,
quoted in parenthesis, in the second half of the sentence, strongly suggests that the cross
also was physically imposed upon the wax of the Candle. Moreover, the use of the
impersonal faciant makes it difficult to envisage the tracing of a cross in the air by one
1
' Let them make a cross on the Candle out of incense, (and let the year of the Lord and
the Greek letters A and Q be inscribed).' PL 78.336B (= M 106). The evidence of the
Pontifical of Egbert is discussed in the next chapter, p.272 n.2.
2
Le rite de cinq grains p.8.
3
Suscipe incensi huius sacrificium vespertinum and super hunc incensum, respectively.
The latter phrase is found in the original version of the prayer. See also Chapter 8 pas-
sim.
4
Compare the corresponding phrase in Poitiers (p.215): contra cereum.
255
individual using a lighted candle. Again, the sign of the cross in Poitiers is made with
the fire on a reed. In the absence of evidence to the contrary and in the light of later
practice, it must be assumed that the cross was formed by the insertion of five grains of
incense at the salient points familiar from later practice, as opposed to the embedding of
two continuous lines of incense grains into the wax at right angles to each other. 1
The insertion of the grains of incense as a preliminary act prior to the blessing of
the Easter candle is also enjoined in Lanfranc's Decrees, in the contemporary Acta Vet-
era of Rouen, and in the Customary of Sigibert.2 It was performed at Essen in the four-
teenth century; and at Barking, also in that century, the grains were inserted into a
pre-incised cross on Good Friday.3 In all these instances there is no suggestion of any
benediction-formula prior to their insertion, and, as decorations, they were commemor-
ative in a way which ~e shall discuss presently.
Two traditions existed relating to the point in the ceremony at which the grains
were inserted : (i) before the blessing of the Easter candle - already noted in the previ-
ous paragraph - and (ii) during the Preface following the Exultet. 4 On account of the
close rapport that existed between the insertion of the grains and the censing of the
Easter candle, and because the Candle came to be censed during the Preface for the rea-
sons we will advance in the next chapter, the latter practice became by far the more
common and eventually the norm throughout the Western Church.
The censing of the Easter candle vis-a-vis grains of incense is found in three tradi-
tions, referred to, for convenience, as modes, and summarised below in Table 48.
At frrst sight the insertion of the five grains of incense into the Easter candle, immedi-
ately followed by the censing of the same (Mode C), appears puzzling; for the frrst
action was in a sense an honouring of the Candle with incense, so that the second action
wouls seem to be a superfluous reduplication of the first. A similar claim could be
made for the ritual according to Mode B where the two actions were separated by an
unspecified interval of time. Bearing in mind that liturgical practice did vary from rite
to rite, it is not difficult to see in the modes three stages in the development of the incen-
sation of the Easter candle; and to recognise that in the course of liturgical development
anomolies and duplications do arise : either as a result of
1
That this method may have been used, however, still remains a possibility.
2
PL 150.4660; PL 147.176C; Albers ll p.99, respectively.
3
0rdinal p.66 and HBS 65 p.lOl, respectively.
4
Hereafter referred to simply as the 'Preface'.
256
Gembloux llC Albers n p.99
Essen 14C Ordinal p.66
Carcassonne 1749 Missal p.196
Perigueux 1782 Missal p.159
LePuy 1783 Missal p.159
Meaux 1845 Missal p.169
Mode A. Grains of incense inserted before the Exultet. No
incensation of the Easter candle .
Evesham• c.1250 HBS 6 col.92

Norwicht c.1265 HBS 82 p.91
Marseille* 13C /LEMp.84
Durham* 14C Missal p.l88
Westminster* c.1370 HBS 5 col.582
St Germain des Pres• 1394 DAER 3.15.7 p.142
Jumieges (Ordinal) De Vert IT p.35
Pontivy (Ritual) " "
St Vincent, Metz " "
* at in huius igitur noctis
t at sacrificium vespertinum
+at suscipe, sancte pater
Mode B. No grains of incense inserted into the Easter candle. 'This
was censed during the Preface. (See also p.272 n.2.)
Corbie (Sacramentary) 10C PL 78.336B

Rouen (Acta Vet.) llC PL 147.176C
Lanfranc c.1070 PL 150. 467C
Mode C. Insertion of grains of incense before the start of the
ceremony. Incensation during Preface.
V allombrosa llC Albers IV p.249

St Benigne, Dijon llC DAMR 3.15.7 p.142
Lateran Church c.l143 OELp.61
Narbonne 1528 Missal fol.lxxxv
Vendome De Vert IT p.35
Sees (Ordinal) " "
Mode D. Insertion of grains of incense, and incensation of the
Easter candle during the Preface.
Table 48. The grains of incense vis-a-vis the censing of the Candle.
misinterpretations or reinterpretations of language, or as the result of attempts to

harmonise biblical symbolism with liturgical practice, or as the result of synthesising or
attempting to synthesise into one ritual ceremonial actions drawn from varying liturgical
milieux.
257
The most likely explanation for this double ritual act is perhaps to be found in the
parallel development of another consecratory action involving the use of incense, which
provides a credible analogy to the ritual of the grains. The Gelasian Sacramentary
prescribes 'an offering of incense upon the altar' at the suggestion of and to underline
the spiritual incense in the prayer for the consecration of an altar. 1 In practice this
offering entailed the tracing with the thurible of a cross above the altar according to
ninth-century and tenth-century sources. By the twelfth century the ritual had
undergone a transformation : a cross made of incense grains was burned upon the altar?
Rabotin called this 'a kind of materialisation of a primitive rite'. 3 The writer believes
that the censing of the Easter candle underwent a similar materialisation at an earlier
period, but resulted in a transformed outcome. For, whereas the dedicatory cross of
incense upon the altar was made to burn, the incense inserted into the Easter candle was,
for obvious reasons, diverted from its natural use, and transformed into a permanent and
visible symbol; and it is not difficult to discern the two main contributory factors which
helped to bring about this change. (i) From a distance the visibility of a cross incised
into a candle of bleached wax would be greatly enhanced if the arms of the cross were
highlighted either by two continuous lines of incense grains arranged at right angles to
each other in the incision, or by the insertion of single grains of incense at the five
salient points of the cross. (ii) The identification of the Easter candle with the body4 of
Christ, and the association of the cross, already incised into the Candle, with his
suffering and death may have invited the insertion of the five grains of incense, as
symbols of his wounds, at the points of the cross just referred to.
We therefore believe that the insertion of the grains of incense does comprise a
duplication of the first act of incensation with the thurible in that the second ritual action
both reinforced and expressed the latter in a more permanent and material way; and that
this duplication of censing, attested by the six documents of Mode D, represents an
intermediate stage in the development of the rite. We now tentatively suggest an outline
history of the ritual, showing the stages by which the development may have occurred.
Stage 1. The censing of the Easter candle before the Exultet. There is no documentary
evidence for the censing of the Easter candle at this stage of the ceremony. However, a
number of observations may be made in support of our view that at an earlier period an
initial act of incensation took place before the Candle was blessed. (i) We know that the
1
Mohlberg, GeV §692 p.l08.
2
Small heaps of incense were placed in the four comers and in the middle of the altar
during the consecration, and ignited. PR XII I p.200.
3
Les Grains d' Encens p.225.
4
Medieval writers compared the wax of the Candle with the human substance of Christ.
See Chapter 14 pp.307-8.
258
censing of other inanimate objects took place prior to their consecration, for instance, at
the above-mentioned dedication of an altar in the Gelasian Sacramentary. (ii)
According to the three Gelasian sacramentaries, Ordo 30A, and the Reims Sacramentary
the cross is traced super cereum and the Candle is lit before the blessing.' It is difficult
to believe that the Candle was not censed also. (iii) The relevant descriptions in the
above-mentioned documents are short and contain only a minimum of detail. Mention
of a routine action, such as incensation, is almost likely to have been omitted. (iv) It is
difficult to account otherwise for the insertion of the grains of incense in the rites of
those churches listed in Mode A, and to believe that they were originally inserted merely
as decorations
Stage 2. We suggested above that the insertion of the grains of incense represents a
materialisation of the act of incensation. The presence of the grains of incense on the
Candle, therefore, prior to its being blessed (Mode A )presupposes that an act of
incensation had already occurred. The grains thus embedded as a permanent and visible
indication that the Easter candle had been honoured with incense, the action of censing
with the thurible was either seen to be superfluous, since in effect the Candle had been
censed twice, with the result that it disappeared from the ceremonial; or it was
transferred to another point in the ceremony- a situation which we fmd in Stage 3.
Stage 3. The censing of the Easter candle during the Preface and the reasons which led
to the occurrence of that ritual act midway through the taus cerei are dealt with at length
in Chapter 8. In some churches the grains of incense continued to be inserted prior to
the blessing of the Candle; but, having lost their close rapport with the former initial act
of incensation, now assume an almost-decorative function. Tills stage in the
development of the ritual of the grains vis-a-vis the censing of the Easter candle is to be
identified with Mode C (p.257). (In the churches included in Mode B the cruciform
materialisation of the act of censing the Easter candle had not taken place.)
Stage 4 ... Here the censing of the Easter candle and the materialisation of that action in
the form of the insertion of the five grains of incense into the Candle both occur during
the Preface. Tills corresponds to Mode D.
Stage 5. We now reach the fmal phase in the development of the ritual. Here we must
envisage a situation in which the incensation of the Easter candle during the Preface
was regarded as an unnecessary duplication, as it were, of the insertion of the five grains
of incense. Since the presence of the grains in the wax provided visible and permanent
'For references, see Chapter 6 p.246.
259
evidence that the Candle had been honoured with incense, the censing of the Candle
with the thurible ceased to be carried out. We believe that this development had
occurred in the earlier Romano-Gallican rite from which the later medieval Roman rite
derived.
(ii) The significance of the five grains of incense
It is possible to identify three main traditions relating to the use of the five grains
of incense.
1. The negative use of the grains of incense.
Since liturgical development was never uniform and variety of practice could and
did occur even within one small geographical area, it is perhaps not swprising to fmd
that the ritual of the grains did not feature in the ceremonial of a considerable number of
churches during the Middle Ages. Table 491ists some of the churches in whose
service-books there is no mention of the grains of incense.
The insertion of the grains of incense was unknown in the ancient rites of Beneventum,
Braga, Milan, and Toledo; and it is clear from the table below that the absence of this
feature was not confmed to any particular geographical area. The Missal of Breslau
contains a prayer for the blessing of the incense (fol.lxxix ). Interestingly, this prayer,
Veniat quaesumus, is the second of the four prayers prescribed in that missal for the
blessing of the frre, and occupies the same position as the corresponding prayer in the
Durham Missal. In the latter church this was a prayer for the blessing of the incense in
the thurible. It should be borne in mind that the absence of rubrics relating to the grains
of incense does not necessarily imply that the ritual was unknown in that rite. 1 The
presence of eight Benedictine houses in the above list shows that liturgical practice was
not uniform in the monasteries of that order, a fact not unnoticed by Lanfranc. 2
2. The cross of incense grains as a decoration.
We noted above on page 259 that in churches in which the grains of incense had
been inserted into the Easter candle before the ceremony began, they had come to be
regarded almost as markers highlighting the five salient points of the cross, rather than
'It is just possible that the reason for the silence of some of the documents in the above
list lies in the fact that the grains of incense had been inserted before the start of the
ceremony.
2
Decrees. PL 150.467B
260
Church Date Document/Source
Leon· lOC Antiphonary
Mainz c.950 PRG
Ripoll 1038 Sacramentary
Beneventum llC M issale Antiquum
Evesham c.l250 Ritual
Norwich c.1265 Customary
Marseille 13C ILEM
Westminster c.1370 Missal
StGermain des Pres 1394 DAMR 3.i5 p.142 (M 1165)
StMary's, York c.1400 Ordinal
Durham 14C Missal
Tongres 14C Ordinal
Milan* 1475 Missal
Braga· 15 c King, LPS p.223
Saintes c.1500 Missal
Pass au 1503 Missal
Cambrai 1507 Missal
Hamburg 1511 Missal
Breslau 1519 Missal
Jumieges t Ordinal (De Vert II p.35)
Pontivy t Ritual ( " " )
St Vincent, Metz t Ordinal ( " " )
t De Vert does not mention the date.

* The ritual of the grains was subsequently introduced as a result of
Roman or monastic influence into the rite ofToledo in 1500, of Braga
in 1512, and of Milan in 1560, according to the respective missals for
those years.
Table 49.
as visible symbols recalling the incensation of the Candle. This is almost certainly
borne out by the fact that there is no instance from those places where the grains were
inserted prior to the blessing of the Easter candle that they themselves received a
blessing. It is also significant that in five of the six 1 medieval instances in which the
insertion of the grains featured as a preliminary act, other insignia were also placed on
the Candle as part of its preparation. Only the Acta V etera of Rouen is silent on this
point. A number of churches, listed in Table 50a, record the insertion of the incense
grains during the Preface as a decorative and commemorative act. There is no
benediction-formula for these grains, which seem to serve chiefly as the means of
delineating the cross on the Easter candle, and only secondarily as reminders that in
former times the Candle had been censed during the Preface.
1
The documents are listed in Mode A and Mode C on p.257.
261
Church Date Document/Source
Fruttuaria llC Customary (Albers IV p.65)
V allombrosa llC Customary (Albers IV p.249)
Cluny llC Customary (PL 149.663B)
Wurzburg 1477 Ordinal np
Trier c.1487 Missal fol.ciii
Cologne 1494 Missal fol.cxxv
Bremen 1511 Missal fol.lxxxvi
Spires 1512 Agenda fol.c
Meissen 1520 Breviary np
Tournai 1540 Missal fol.lxx
Sens 1715 Missal p.242
Table50a
Cistercians 1119 Nom.Cist. p.105
Soissons c.1185 DAER 4.24 p.161 (M 305)
Nidaros 13C ONEp.232
Exeter 1337 Ordinal (HBS 37 p.322)
Table50b
Table 50. Evidence for the decorative function of the grains.
The documents in Table 50b prescribe the insertion of the grains during the Preface and
do not include a prayer for their blessing. However, since they also omit any prayers
for the blessing of the fire, they have been listed separately. It should be noted that the
ftrst, second, and fourth documents in Table 50b also enjoin the inscribing of other
insignia on the Candle as part of its preparation.
3. The representational function of the incense grains.
In Chapter 14 we shall discuss the symbolism of the Easter candle, and the shift in
emphasis regarding the nature and significance of the Candle. In the same way that the
whole Candle - flame, wax, and wick - came to represent different aspects of the human
nature of Christ, in addition to symbolising his divine light, so the grains of incense,
decorative attachments recalling the incensation of the Candle, assumed a more realistic
function and became vivid representations of the wounds of Christ in the wax that was
his 'flesh'. This developed from a modification or change in two aspects of the ritual
(i) A blessing came to be pronounced over the grains prior to their insertion into the
Candle. (ii) It became the practice in places to bear the incense grains in procession
along with the new ftre. As a result they acquired greater prominence within the
ceremonial, achieving an importance almost on a par with the new fire. At Lyon the
anointing of the grains almost certainly recalled the treatment of Christ's wounds
The ritual involving the blessing and insertion into the Candle of the five grains of
incense developed along three different lines.
262
Mode I. The grains are :
(a) blessed shortly after the commissioning ofthe new fue;
(b) carried in procession (in churches where a procession is held);
(c) inserted into the Candle at the words suscipe, sancte pater ... incensi of the
Preface.
This is the mode that was ultimately adopted by the great majority of churches in the
West mainly through Roman or monastic influence. It is fust attested in the
twelfth-century Pontificate Romanum, 1 though there is no procession according to that
document, and in a contemporary Premonstratensian sacramentary.2 The aspersion and
censing of the grains, though absent from the 1474 Missale Romanum, featured in the
Roman Missal of 1570 and entered the Ambrosian rite in the revised Missal of 1902.
Mode II. The grains are blessed and inserted during the Preface. This mode is found in
the following rites :3
Evesham (c.1250)
St Vedast, Arras (c.l300)
Seville (1507)
Braga (1512)
Rennes (1523)
Troyes (1736)
At Braga the fust of the two prayers used for the blessing of the grains began during the
Preface at the words in huius igitur noctis; the second presumably finished in time for
the grains to be inserted at the usual place (suscipe, sancte pater). At Rennes the
incense was also aspersed at this point. We are not told the precise point at which the
incense grains were blessed in any of the other churches.
Mode Ill. The grains are blessed and inserted before the Exultet. This third tradition
occurred in the rites of the group of churches in France, listed in Table 51.
This variant recalls the practice of inserting the five grains of incense prior to the
blessing of the Easter candle, which was attested in the six medieval instances in Table
48 (Modes A and C); but derives from an older tradition in which the Easter candle was
1
PR XII I pp.238-40.
2
Weyns p.69 (Manuscript B).
3
Respective references: HBS 6 col.91; HBS 86 p.160; Missal fol.lxxviii; Missal np;
Missal fol.lxxiii; Missal p.27.
263
Church 1 Date B enedictionformuld- Source
Bourges 1741 D.D.N.qui suscepisti Missal pp.225-27
Carcassonne 1749 Veniat quaesumus Missal p.l96
LePuy 1783 D.D.N.qui suscepisti Missal pp.l59-60
Perigueux 1784 D.D.N.qui suscepisti Missal pp.l59-60
Meaux 1845 Veniat quaesumus Missal p.l69
Table 51
was kindled before the singing of the Exultet. 3 All five churches, listed in Table 51,
preserve this feature; and in all, except Meaux, the Easter candle was lit prior to the
insertion of the grains
The reformed Mozarabic rite, as authorised by the Missale Mixtum of 1500,

strictly falls into this category, Mode III. However, since the ceremonial it prescribes
was the result of a attempt to harmonise two different traditions, the insertion of the
grains before the lighting of the Easter candle was deliberately contrived and not the
result of normal liturgical development.
(iii) The insertion of the grains of incense
The implantation of the grains of incense into the Easter candle prior to the com-
mencement of the Vigil, or even prior to the chanting of the Exultet, enabled those
involved in the preparation of the Candle to ensure that the grains were embedded
securely in the wax. With the insertion of the grains during the taus cerei the possibility
always existed that one or more of the grains might fall from the wax through the failure
of the resin to adhere to the Candle for one reason or another. We therefore fmd that the
directions in a number of service-books4 state that the five small holes should previ-
ously have been gouged out of the side of the Candle in readiness for the reception of
the grains; and until 1955 this practice was officially sanctioned by the Church of
Rome. Desideri adds the sensible advice that the holes should be lower than the part of
1
Bourges, Le Puy, and Perigueux belong to a subgroup whose ceremonial is character-
ised not only by the use of the prayer D .D .N.qui suscepisti for the blessing of the grains
of incense, but by the survival of the Veniat quaesumus as a preliminary blessing of the
Easter candle. Amiens too might be classed with these three churches, except that in the
former church the grains were inserted into the Candle more Romano during the Pref-
ace.
2
For the text of these formulas (V) and (a), see Appendix 5.
~e use of the V eniat quaesumus for the blessing of the Easter candle belongs to the
same ancient tradition.
4
0rdo ofNidaros p.232; Strasbourg Ordinal (DAER 4.24 p.l62); Besan~on Ceremonial
p.314; Paris Ceremonial p.374; 1488 Missal of Basel fol.xcii; 1498 Missal of Prague
fol.xci.
264
the Candle which will bum. 1 A rubric from Magdeburg states that the incense should be
inserted sub signo crucis? This may be intetpreted 'in the form of a cross' or 'under the
sign of the cross'. The former interpretation is to be preferred in view of the
almost-universal observance of the cruciform arrangement.
At Lyon the grains were ftrst warmed so as to soften the wax, and then anointed
with chrism before being affixed. 3 Rome also allowed the application of heat. 4lt is
recorded that at Valence the grains were embedded into the salient points of an incised
cross which had frrst been anointed with chrism. 5 This oil in the above-mentioned
instances may also have acted as an adhesive; although we have suggested an alterna-
tive origin for the use of the chrism at Lyon. 6 It was also recommended that the grains
should be large, or that several grains should be fused together so as to be conspicuous
at a distance. In more recent times the grains have sometimes been held in wooden or
metal cases to which a spike has been attached for their insertion into the wax. As a
result the devices have been referred to as nails. The term is also used of the small
spike of red wax with a grain of incense in its head, which is inftxed into a special-
ly-prepared cavity. Both the word nail and the colour of the wax which highlights the
grains against the pale background of the Easter candle, appropriately reinforce the
symbolism of the incense. (See Section (v).)
As a result of the liturgical reforms of 1955 the primitive practice was restored :
the grains were thenceforth to be inserted immediately before the Easter candle was lit,
and subsequently blessed. 7 In the Roman Missal of 1970 there is no reference to the
insertion and blessing of the grains. All markings on the Easter candle have now been
made optional.8
Traditionally the grains of incense have been inserted by the officiating deacon or
the priest who acts for the deacon. According to the Customary of St Be'nigne, Dijon9
the armarius performed this task while the deacon was censing the Easter candle. At
1
Praxis p.150.
2
1503 Missal fol.xciiii.
3
4
Fortescue and O'Connell (6th ed.) p.343.
5
1504 Missal fol.liiii.
6
See p.252.
7
Diekmann p.129.
8
1970 Missale Romanum p.267; Harbert p.236. The Milanese rite follows Roman
practice in this respect (1981 Missale Ambrosianum p.243), as does the recently-re-
vised Anglican service (Lent, Holy Week, Easter p.229).
~artene, DAMR 3.15.7 p.142 (M 1150).
265
St Vedast, Arras and at Cordoba the sacristan gave the deacon one grain to insert, and
himself inf'txed the other four. 1 At the Cathedral of StJohn Lateran the grains had to
be of the purest incense?
According to the rubrics of the vast majority of provincial and diocesan rites the
grains began to be inserted following the words curvat imperia in the Preface. In the
next chapter we shall observe why this was thought to be an apposite juncture for carry-
ing out the insertions. It also provided a suitable point in the Preface at which to pause.
4
In some sources3 the break is prescribed at the words in huius igitur noctis, in others
at sacrificium vespertinum; whilst in a number of service-books5 it is enjoined at sus-
cipe, sancte pater. These slight variations did not substantially affect the performance
of this ritual action, and different interpretations based on these variations are not
justified. Since there is no indication to the contrary in the rubrics of any service-book,
we must assume that the deacon interrupted his singing at the point indicated by the
words, in order to insert the grains into the Candle; and that on the completion of this
duty he resumed his c_hanting of the laus. At Cologne, however, the rubrics in the Mis-
sal of 1495 specify a fixed point in the Preface for the insertion of each grain (fol.cxxv ).
Within the sentence beginning in huius igitur noctis the following arrangement is
found:
1st grain inserted at suscipe

2nd grain inserted at vespertinum
3rd grain inserted at oblationem
4th grain inserted at solemni
5th grain inserted at apum
Subsequent missals of Cologne present a difficulty; for whilst they specify the moment
for each insertion, they refer to only four grains.
1
HBS 86 p.l60 and 1561 Missal fol.cix, respectively.
2
0ELp.61.
3
For instance, Haymo (OM), HBS 85 p.205 and Customary of St-Germain-des-Pres,
DAMR 3.15.7 p.142 (M 1165).
4
For instance, PR XII I p.240 and 1561 Missal of Cordoba fol.cix.
5
0rdinal of St Vedast, Arras (HBS 86 p.160); 1481 Missal of Verdun; 1519 Missal of
Aquileia fol.91.
266
1514 Missal 1525 Missal
1st grain at suscipe 1st grain at suscipe

2nd grain at vespertinum 2nd grain at vespertinum
3rd grain at solemni 3rd grain at per ministrorum
4th grain at manus 4th grain at de operibus
The reduction in the number of grains from five to four is inexplicable from the evi-
dence of the missals alone. The only other known instance where only four grains of
incense were used occurs in a fourteenth-century ordinary of the Collegiate Church at
Essen. Here four grains of incense occupied the outer points of the cross. However,
although some connection could be argued on the grounds of the geographical proxim-
ity of the two cities, additional evidence is required before a liturgical link between the
two churches can be established. Moreover, the central grain at Essen consisted of
myrrh, not incense.'
Now the 1518 Missal of Ratisbon records that the grains of incense were inserted
at the words in odore(m) suavitatis (fol.xcvii). In the next chapter we shall argue that
the lighting of the Easter candle at suscipe, sancte pater indicated that incensi (the fol-
lowing word) had retained its original meaning of 'fire' within this context. It is, there-
fore, just possible that the Ratisbon Missal preserves the original point in the Preface at
which the Easter candle was censed, since these words invited the incensation of the
Candle almost as an internal rubric.
We have shown that the five salient points of the incised cross provided the obvi-
ous places at which to impress the grains into the wax of the Candle. Contemporary
practice is to insert them in the form of a Greek cross; and illustrations from manuals2
of the past, which prescribe the order for the insertion of the grains, would suggest that
this shape has long been traditional. However, at Le Puy, Nantes, and La Rochelle a
Latin cross was used. 3 The grains were normally inserted in the order shown below,
regardless of the shape of the cross. This corresponds to the personal sign of the cross
in the Western Church in which the cross-stroke is made from left to right.
1
4 2 5
3
'Ordinary p.66.
2
For example, Martinucci IT p.242 and Merati p.81.
3
1783 Missal p.160; 1837 Missal p.203; 1835 Missal p.191, respectively.
26T
At Auch 1 and at Toulouse2 the grains were arranged in the form a five-pointed star.
This shape was, according to De Vert (II p.37), suggested by the occurrence of vesperti-
num in the above-mentioned phrase from the Preface, the word being the adjectival
form of vesper, 'the evening star'.
At Salisbury, Exeter, and in the Gilbertine houses there existed a choice of affix-
ing the grains either to the Easter candle or to the candelabrum which supported it. 3 The
phrase si attingi potest in the Gilbertine Ordinal suggests that the Easter candle might be
beyond the reach of the deacon's hands if the candlestick was very tall. The practice of
affixing the grains to the false stock of the Candle was forbidden according to a nine-
teenth-century ceremonial of Lyon. 4 It maintained that the union of the grains with the
wax was a prerequisite of a valid benediction of the Candle; and that the drops of wax
that fell into the baptismal waters were inefficacious , since the Easter candle had not
received an authentic blessing. At Salisbury, Angers, and in the churches of the Dio-
cese of Lyon, grains were also inserted into the cereus minor, the smaller Easter candle
that was carried to the font. 5
(iv) The grains of incense at Milan
The ceremony of the grains of incense in the Ambrosian rite has an interesting his-
tory in that we plot its gradual adoption from the Roman rite in three stages. (i) In the
Missal of 1475 the Easter candle is censed at the conclusion of the Preface (fol.lxxxi).
There is no mention of the grains of incense. (ii) According to the Missal of 1560 five
grains of incense are inserted into the salient points of a pre-traced cross after the incen-
sation of the Candle which follows the deacon's song (fol.113 ). (iii) In the Missal of
1594 we fmd that the grains are now inserted during the Preface at the words chrismate,
non cruore (fo1.95). This is almost certainly the result of Roman influence; for the
1
2
De Vert II p.37. He refers to some other churches without specifying them.
3
Sarum Missal (Warren) p.270; HBS 37 p.322; HBS 59 p.40, respectively.
4
5
Sarum Missal (Warren) p.270; 1731 Ceremonial p.261; 1838 Ceremonial p.479,
respectively.
268
Candle, formerly lit before the singing of the Exultet, now receives its flame also during
the taus. The Preface in the Roman rite had been characterised by these ritual interrup-
tions since the twelfth century. 1
(v) The symbolism of the grains of incense
Medieval commentators were not slow to fmd biblical echoes in the use of the five
grains of incense, and to attach symbolic interpretations to them. We therefore
encounter symbolism at two levels : that recalling and suggested by two incidents in the
Gospel narrative, and that which had wider theological implications.
(a) Writing c.llll Rupert of Deutz,2 followed by Sicardus and later Durandus, 3
saw represented in the grains of incense the spices and ointment which the women
brought to the grave of Jesus. In general detractors of medieval symbolism, such as
Capelle,4 have not been slow to point out that the body of Jesus was anointed with
myrrh and aloes, and not with incense (John 19:39). Their criticisms, however, have
been largely unjustified on two counts. (i) A symbol is but one object representing
another object, and by the very nature of a symbol cannot be that which it represents.
The incense was understood to represent the women's myrrh because of the redolent
properties characteristic of both substances; and its use was all the more apposite in
those churches in which the grains of incense were carried in procession from the new
frre to the Easter candle. (ii) Incense and myrrh are not dissimilar substances. In fact,
in some types of incense myrrh is an ingredient. The correspondence between the
incense, and the myrrh and aloes brought by the women, was further suggested by the
relative points during the Roman Preface at which the grains were inserted and the
Easter candle was lit. For the Resurrection, symbolised by the lighting of the Candle,
did not take place until after the embalming.
'Unlike the interruptions in the Roman Preface, which follow each other at fair~, 'I ..Jrt
intervals and which are largely determined by words acting almost as internal rt~brics
(p.XXX and p. 316'), those in the Ambrosian Preface occur at the end of well-defmed
sections, and divide the taus into four more or less equal parts. The wording of those
phrases in the Preface at which the three actions occur (viz. the lighting of the Candle,
the afftxing of the grains, and the illumination of the church lights) is not unrelated to
the nature of those actions. Of particular relevance here is the action involving the
grains and the sentence from the Preface which invites that action: Christi vero populus
insignitur fronte, non inguine; lavacro, non vulnere; chrismate, non cruore. Here this
Roman-derived ritual action of inserting the grains bears little, if any, relationship to the
post-prefatorial act of incensation that occurred in the Milanese rite. The cross formed
by the grains of incense is symbolic and anticipatory of the chrismal cross of Baptism.
2
De Div.Off. V, PL 170.173C.
3
Mitrale, PL 213.324A and Rationale VI.80 p.351, respectively.
4
Le rite des cinq grains p.ll. Van Doren on the other hand was equally dogmatic : 'Ils
ne doivent pas representer les clous de Ia croix. Ils signifient, d'apres les liturgistes, les
onguents precieux dont on entoura le corps du Seigneur.' Le cierge paschal p.75.
269
The infixing of the five grains of incense into the Candle also recalled the five
wounds of Jesus. The symbolism, which is still associated with the grains, is first men-
tioned by Durandus c .1280; 1 but it is difficult to believe that the identification of the
grains with the wounds was not made much earlier. For the increase in the ceremonial
importance of the grains, noted earlier in this chapter, is almost certainly linked with the
symbolic identification of the wax of the Easter candle with the human flesh of Christ.
Again, Capelle's objection to the symbolism on the grounds of the lack of positional
correspondence is somewhat unjustified. A symbol does not necessarily mirror an
object in every small detail.
(b) At a theological level the grains of incense had a more profound and also a
double significance. For, since they were inserted at the words sacrificium vespertinum,
they not only recalled the evening sacrifice in the Temple under the dispensation of the
Old Law, which itself prefigured the sacrifice of Christ;2 They also vividly suggested
the sacrifice of Christ himself for three reasons. The grains were inserted into his body;
they were disposed in a cruciform arrangement and recalled the Cross; in the Middle
Ages the ceremony had come to take place at the time of day corresponding to the hours
of the Crucifixion. From a less objective point of view but still within a sacrificial con-
text, the grains of incense together with the flame of the Candle symbolised the sacrifice
or offering of the people to God by the people and on behalf of the people. 3
1
Rationale VI.80 p.351.
2
Beleth, Rationale, PL 202.110; Zazzera p.299.
3
Parvio, Manuale, p.213.
270
Chapter Eight
THE INC ENS ATION OF TilE EASTER CANDLE
The use of incense in the ceremony of the Easter candle belongs strictly to a study of the
history of that element; but the prominence of the five grains of incense in the ceremo-
nial necessitaies some siudy of the use of incense in order lo account for presence of
those grains within the ritual. The attempt to trace the history and development of the
use of incense within the ceremonial from the surviving documentary evidence is
fraught with difficulties. Practices involving its use, inherited from differing liturgical
traditions, could vary markedly even throughout one small geographical area. At times
a document has few rubrics and much has to be understood. At times the rubrics,
because of their brevity, are either vague or ambiguous. With regard to the latter type
of directives perhaps the greatest difficulty has been in deciding how to interpret the
stark instruction 'A prayer for the blessing of the incense', where it is not clear whether
incense relates to the substance in the thurible with which the new fire or the Easter
candle is to be censed, or whether the word refers to the five grains later to be inserted
into the Easter candle.
In Chapter 7 we looked at the close and intimate connection which the writer
believes existed between the censing of the Easter candle and the affixing of the five
grains of incense, and traced the development that occurred in the use of the grains of
incense vis-a-vis the incensation of the Candle. In this chapter we shall consider the
incensation of the Easter candle, noting the three points during the ceremony when this
ritual act is known or is thought to have occurred, viz. prior to, during, and following
the blessing of the Easter candle.
(i) Incensation before the Exultet. Direct evidence for an initial act of incensation
is wanting; but we have tried to show in the previous chapter that the insertion of the
five grains of incense prior to the blessing of the Easter candle presupposes that this rit-
ual act at one time occurred at this stage of the ceremony. Although evidence is again.
lacking, it is possible that an initial act of incensation of the Candle was transferred to
the newly-kindled fire, if a distinction is made between the censing of the fire and the
censing of the Easter candle. •For, if a situation had formerly existed in which the cens-
ing of the Candle had been followed by the insertion of the five grains of incense, and
if, as we argued in the previous chapter, this was seen as a duplication of the act of
• Such a transfer could explain why the new fire was censed ~ it
had been aspersed. In all other instances of blessings incensation
precedes aspersion.
271
incensation, it is not unreasonable to believe that the action involving the thurible was
transferred to the new fire; whilst the presence of the grains on the Candle were held to
be tantamount to its being censed.
(ii) Evidence for the incensation of the Easter candle during the Preface is plenti-
1
ful, and dates from the eleventh century, if not the tenth. 2 It must be assumed that the
incense was frrst blessed before it was used to honour the Candle; but this is not always
stated in the sources. For instance, it is not mentioned by Lanfranc. However, accord-
ing to the eleventh-century Acta Vetera3 of Rouen the incense was blessed with the
Veniat quaesumus after the procession had returned into church; and at Durham and St
Germain des Pres,4 and almost certainly at Norwich and Westminster,5 the incense was
1
For the 18 instances of this practice, see Modes B, C, and in Table 48 on p.257.
1nere are two documents from this century which may attest the practice. (i) Accord-
ing to the Sacramentary of Corbie (PL 78.336B) the bishop blesses both the new frre
and the incense while the deacon is consecrating the Easter candle. Unfortunately there
is no indication as to whether the incense was used to cense the new frre or the Candle,
or both. (It is true that a cross of incense had been placed on the Candle prior to the
start of the ceremony; but it is most unlikely that it is this incense which the bishop
blesses.) (ii) The evidence of the Pontifical of Egbert is more problematical. It contains
the following rubric (p.130) :
Benedictio incensi in Sabbato Sancto antequam benediceris cereum, et
ipsum debes rnitti in cereum in ipso loco ubi dicitur suscipe incensi.
and is followed by Formula (b) for the blessing of the incense (Appendix 5). The rubric
is generally cited as an early instance- Thurston claimed it was the earliest (Holy Satur-
day p.14) - of the practice of inserting the five grains of incense into the Candle (Rabo-
tin p.222; Capelle, Le rite des cinq grains p.9); but, if so, it antedates the next oldest
rubric for the blessing of the grains by about 200 years. Now it is not impossible that an
isolated piece of evidence such as this should have survived from a minor liturgical
milieu; but the interval of time which separates it and the evidence from Corbie raises
some doubt about its authenticity. Since, however, the terminus post quem for this
document is AD 1000, even allowing for the possibility that this rubric is an interpola-
tion, a closer examination of the language of the rubric would not be out of place. For
the use of the second person singular strongly suggests that the directive is the addition
of a later hand
The second part of the rubric is generally rendered : ' ... and you ought to put it (ip-
sum) into the Candle at the very point where the words suscipe sancte occur' on the
assumption that ipsumhlefers to the five grains of incense. The writer, however, would
suggest that ipsum, w 'chis singular, refers, not to the grains, but to the thurible of
incense, since he is of the opinion that mitti here has its basic meaning of 'send' or 'di-
rect', and that in cereum should be. rendered 'towards the Candle' or 'against the
Candle'. Moreover, the rubricist has distinguished carefully between the use of the
preposition in meaning 'in' or 'on' and suggesting position, and in governing the accus-
ative case and conveying the notion of forward movement.
The clause ipsum debes mitti in cereum will then be interpreted 'you ought to
cense the Candle'. If the suggestion is correct, the Pontifical should now be cited as
evidence for the censing of the Easter candle in the tenth century, but not for the inser-
tion of the grains.
3
PL147.176C.
4
Missal p.l86 and DAER 4.24 p.l58 (M 250), re~;pectively.
5
HBS 82 p.91 and HBS 5 col.576, respectively.
272
hallowed at the same time as the fue. At Evesham the blessing took place during the
Preface. 1 At St-Germain-des-Pres and at Evesham the deacon circumambulated the
Easter candle as he swung the censer.2
Apropos of the censing of the Easter candle during the taus cerei it was Claude de
Vert who fust drew attention3 to the fact that in the following invocation in the Preface
suscipe, sancte Pater, incensi huius sacrificium vespertinum
incensi had come to be intetpreted 'incense' as a result of a shift in meaning which the
word had undergone. There seems little doubt that, at the time that the Preface was
composed, incensum signified 'fue '; 4 and that this is how the word was originally inter-
preted within the above-quoted context. Confumation of this view is to be found in the
1518 Missal ofRatisbon (fol.cxvii) in which we encounter what appears to be an
intermediate stage in the development of the ceremonial. For the rubric states that the
Candle should be lit at suscipe, sancte Pater, and that the grains of incense should be
inserted at in odore(m) suavitatis. We have suggested on page 267 that the incensation
of the Candle may originally have occurred at the latter point in the Preface, and may
have subsequently been transferred under the influence of incensi to the earlier position.
Only at Ratisbon, it would appear, did the earlier practice survive. In the Ordinary of
Tongres the Easter candle is lit at suscipe, sancte Pater; but it contains no rubrics relat-
ing to the grains of incense.
In the laus cerei the dominant theme of Christ's resurrection is vividly supplem-
ented by the contrast between the darkness of night, which symbolises the evil and
wickedness of the old order, and the re-emergence and triumph of the light of Christ, of
which the Easter candle with its bright flame, is the symbol. The change in the interpre-
tation of incensi5 not only detracted from the centrality of the Candle's light, but intro-
duced into the blessing of the Candle the offering of incense in addition to the offering
of the Candle and its light - and this occurred at the beginning of a new section of the
taus. The offerillg of incense now having been introduced into the text of the Preface,
1
HBS 6 col.90, at the words Haec nox est in qua.
2
DAMR 3.15.7 p.142 (M 1165) and HBS 6 col.92, respectively.
3
Explication IT p.35.
4
Literally 'that which has been kindled'.
5
The English translation of OHS 'this evening sacrifice of burning light' restored the
original and correct interpretation of incensi (Diekmann p.135). Sadly, ICEL were con-
tent with 'the sacrifice of praise' for the Roman Missal of 1970 (PJSM p.322).
273
the way was now open for an act of incensation to take place at the words which made
reference to that element; and this in tum resulted in the insertion of the five grains of
incense into the Candle at this precise point.
It was perhaps inevitable that incensi should undergo this shift in meaning from
'fire' to 'incense'; for according to Amalarius 1 the altar at Vespers was censed at the
second verse of Psalm 140 :
Dirigatur oratio mea sicut incensum in conspectu tuo : elevatio manuum mearum
sacrificium vespertinum.
The verse was particularly apposite vis-a-vis the invocation in the Preface since (i) the
deacon's prayer was being offered in conspectu tuo; (ii) the words of the psalm incen-
sum and sacrificium vespertinum were directly paralleled in the Preface; and (iii), as we
noted in Chapter 6, in some churches2 the sign of the cross was traced in the air at the
word incensi - an action possibly echoing the elevatio manuum of the psalm.
(iii) The censing of the Easter candle after the Preface occurs in the Ambrosian
rite at Milan. Though not mentioned by Beroldus in the twelfth century, it is prescribed
in the Ambrosian Missal of 1475 (fol.lxxxi) and in the revised Missal of 1902.3 In the
revision of the Ambrosian rite following the Second Vatican Council, the incensation of
the Easter candle still takes place at the conclusion of the Preface.4 This contrasts with
the revised Roman rite in which incensation takes place prior to the preconium. The
Milanese practice is also found at Essen in the fourteenth century; 5 and may also have
featured at Ripoll and at Saintes. For at the Spanish monastery a prayer for the blessing
of the incense follows the Preface;6 whilst at Saintes a rubric indicates that the benedic-
tion of the element occurred at the same juncture.7
1
De Eccl.Off. IV.7, PL l05.1181C.
2
PRG IT §347 p.98.
3
Repertorium p.47, typical edition.
4
1986 Messale Ambrosiano p.59.
5
0rdinary p.66.
6
Sacramentary p.92.
7
Missal of c.1500 fol.lxxxi.
274
Chapter Nine
THE INSIGNIA ON THE EASTER CANDLE
(i) Alpha and Omega
The delineation of the two Greek letters A and non the Easter candle is attested in sur-
prisingly few docwnents. It is found in the Sacramentary of Corbie and in the contem-
porary tenth-century Mozarabic Antiphonary of Leon, 1 in the twelfth-century
Pontificate Romanum and in two other closely-related Roman docwnents,2 and in the
fourteenth-century Ordinal of Essen (p.58). Their occurrence in the so-called suburbi-
carian tradition recorded in PR XII and their presence in the Sacramentary of Corbie
point to a Gallican provenance for the practice. In Appendix 11 we argue that the
Easter candle together with the markings on the Candle entered the Mozarabic rite as a
result of Gallican influence.
It is generally believed that the A and .Q were depicted on the Easter candle to
indicate that Christ was Lord of the Ages and 'potentate of time'; and to a point this is
true. However, the inclusion of the date, that is, the nwnber of years from the Incarna-
tion, also on the surface of the Candle, rendered the two Greek emblems superfluous, in
that the date also came to represent the same concept and to signify Christ's sovereignty
over time. It is almost certain that for this reason the depicting of A and .Q on the
Candle generally fell into desuetude, surviving only in isolated instances in the later
Middle Ages, such as at the Collegiate Church in Essen. 3
Although it is unknown when the two Greek letters were first inscribed on the
Easter candle, it is not difficult to discern the reason for their delineation. In earlier
times the Paschal vigil was observed in an expectation of the second coming of Christ,
and in commemoration of the Resurrection, the event which had ushered in the New
Age. Nor is it difficult to realise why, within the other-worldly context of the Paschal
liturgy, the eschatological name or attribute of Christ, found in St John's Revelation
(20:6), should be chosen and used both to identify the Candle with Christ within this
eschatologically liturgical context, and to indicate to the faithful the identity of the
Candle. 4 The depicting on the Candle of this title from Scripture is reminiscent of the
1
PL 78.336B (M 106) and Antiphonary p.280, respectively.
2
PR XII I xxxii.lO p.241 (suburbicarian variant); OEL p.60; Pontifical of Apamea
(DAER 4.24 p.160).
3
0rdinary p.58.
4
Especially if the Candle was of very modest proportions.
275
practice within the Orthodox tradition of including the name or an abbreviation of the
name on the icon of a saint. The inscribing of the earthly name of Jesus may have
seemed inapposite within the futuristic context of the liturgy, being evocative of the
inscription above the Cross, and in some way diverting attention towards the Passion
rather than anticipating the Resurrection and the Second Coming.
In two manuscripts' a pictorial representation is preserved of the two letters in a

spatial relationship to the cross. Presumably they bear some likenes to the devices
actually delineated on the Easter candles. The later and more elaborately-designed
device from Essen also incorporates the Chi-Rho monogram and a trio of crosses, sym-
bolising the Trinity rather than echoing Golgotha.
Figure 1 (Leon) Figure 2 (Essen)
The similarity of the position of A and Q in both devices suggests that the two crosses
derive from a common ancestor in spite of the differences in time and location which
separate their respective delineations. The position of the letters at the extremities of
the arms of the cross suggests that they represent the notion that time belongs to and is
held in the hands of Christ, even when he is on the cross.
(ii) The Year
Earlier liturgical commentators2 cited the following passage from Bede's De Tern-
porum Ratione, written in AD 725, as early evidence for the inscribing of the year on
the Easter candle :
1
Antiphonary of Leon p.280 and Ordinal of Essen p.58.
2
For instance, Novarinus p.l7 (c.l635); Martene, DAER 4.24.7 (c.1700).
276
Sancta quidem Romana et apostolica Ecclesia bane se fidem tenere et ipsis testatur
indiculis, quae suis in cereis annuatim scribere solet, ubi tempus dominicae passio-
nis in memoriam populis revocans, numerum annorum triginta semper et tribus
annis minorem quam ab eius incamatione Dionysius ponat, adnotat. Denique
anno ab eius incamatione iuxta Dionysium septingensimo primo, indictione quarta
decima, fratres nostri qui tunc fuere Romae, hoc modo se in natali Domini in cer-
eis sanctae Mariae scriptum vidisse, et inde descripsisse referebant : 'A passione
Domini nostri Jesu Christi anni sunt DCLXVITI' .1
Martene's comment was: Patet insciptionem olim in Paschate adhibitam, per totum
annum in cereo conservatam fuisse. 2
The French liturgist was correct in finding the origin of the custom of inscribing
the year in the practice of the Roman Church; but in the mistaken belief of his time that
the Easter candle was known at Rome in AD 701, he interpreted Bede 's in cere is as
though the noun were in the singular form in cereo, and thus avoided a difficulty, as
Dendy pointed out (p.138). Since the candles inS. Maria Maggiore, which Bede's fel-
low monks visited on Christmas Day, were not Easter candles - he clearly refers to the
existence of more than one candle in this basilica - we must try to establish which were
the candles in that church that bore the inscription. A number of preliminary observa-
tions must be made. First, we may at once eliminate this as a reference to all the
candles of the basilica, since it is unlikely that all of them were made of beeswax (cera).
Secondly, the candles in question were either only used on rare occasions, since the date
was still visible at the end of December, or were exceedingly large, so that even with
constant use they had not burned down to that part of the stock which bore the inscrip-
tion. Moreover, the wording ofBede's reported inscription, if authentic, would have
required the use of candles of no mean diameter. Thirdly, Bede informs us that the
candles were renewed every year and, if so, lasted for the whole year, as Martene con-
jectured. Fourthly, the candles were closely associated with the events of Holy Week,
and in particular of the Triduum.
1
'The holy and apostolic Roman Church bears witness that she keeps this faith even by
the very marks which it is her custom to inscribe on her candles each year. For, in order
to remind the people of the season of the. Lord's Passion, she registers (on the candles)
the number of years less thirty-three which Dionysius calculated from the Incarnation.
Accordingly, in the seven hundred and first year from the Incarnation by Dionysius'
reckoning, our brothers who were in Rome at the time, said that on Christmas Day they
saw written on the candles of StMary's Church [S. Maria Maggiore] and made a note
of the following : 'From the Passion of our Lord Jesus Christ 668 years'. Chapter 47.
PL 95.494B-495A.
2
'1t is clear that the inscription was formerly applied at Easter and kept on the Candle
throughout the whole year.' DAER 4.24.7.
277
In view of these observations it is difficult to escape the conclusion that the wax
candles, to which Bede is referring, are none other than the two man-sized candles
which figured prominently in the Roman Paschal vigil. 1 It is significant that in the Pon-
tifical of Poitiers, which contains elements from both the Roman and the Gallican Vigil
traditions, both man-sized candles bear an inscription (p.215). The writer believes that
the imprinting of information on these candles was a practice of some antiquity, even
when the Pontifical of Poitiers was compiled. The objection that the Paschal vigil at
Rome was held only in the Cathedral of St John Lateran is not insurmountable. The
early ordines Romani relate primarily to papal ceremonial in the Mother of Cathedrals~
and there is no evidence to suggest that a Vigil was not held in the other basilicas of
Rome, apart possibly from S. Croce in Gerusalemme.
With the fusion of ceremonial elements from both the Roman and the Gallican
Vigil traditions in the region of northern Gaul and the resulting diminished importance
of the two Vigil-candles, it is not difficult to envisage the transference of the inscription
from the two large candles of the Roman tradition to the single Easter candle~ and to
understand why the date, previously reckoned from the Passion, subsequently expressed
the number of years from the Incarnation. For the former were the visible symbols of
the Passion that characterised the Roman Vigil; the single Candle represented the Light
of the world and the moment that the Light entered the world. 2
The inscribing on the Easter candle of the year reckoned from the Incarnation is
first attested c .AD 900 in the Pontifical of Poitiers. In addition to the Sacramentary of
Corbie3 it is mentioned in the Customary of the German Monasteries ( c .1 000) and in
Lanfranc's Decrees;4 whilst its appearance in three Roman documents5 may be the
result of the Benedictine influence of Monte Cassino, unless we have at Rome a sur-
vival of the above-mentioned custom of dating the candles, which was described by
Bede. Apart from the above-mentioned documents, but excluding Poitiers, there are
very few instances in which the annus Domini alone was inscribed on the Easter candle.
In the vast majority of instances it formed the point de depart or nucleus for the devel-
opment of the charta, which provided additional chronological and liturgical informa-
tion. For this reason we have deferred a discussion of its subsequent history and
survival until the next section.
'See Chapter 16 for their use and significance.

2
1t may be significant that in the earlier laudes cerei we encounter a close connection
between the supposed parthenogenesis of the bee and the production of wax on the one
hand, and the birth of Jesus, and the Virgin Mary on the other.
3
PL 78.336B.
4
Albers V p.38 and PL 150.4460, respectively.
5
Viz. PR XII I p.241; OEL p.60; Pontifical of Apamea (DAER p.l60 (M25)).
278
The inscribing of the year disappeared from the Roman rite sometime after 1250,
probably as a result of the liturgical influence of the Franciscans, in whose service-
books this ritual act did not appear; and perhaps also as a result of a wish on the part of
the Roman Church that this symbol of the risen Chriast should be devoid of decorative
excess. It is not found in the Roman Missal of 1474. Likewise it must have disap-
peared from the ceremonial of churches which were influenced by or had adopted the
Rom<:m rite. The instance of Salzburg at the beginning of the sixteenth century is a rare
and late survival of this practice. 1
The Church has always stressed that Christianity is an historical religion and that
its founder was born at a specific point in time. The dating of the candles, which Bede
referred to, belongs to that tradition, first found in the gospels, of rooting the message of
salvation fmnly in history; and serves to remind2 the faithful of this fact within the con-
text of the liturgy. With the development of the charta this functional purpose was
underlined by means of the inclusion of additional information, as we shall observe
presently. The medieval mind, however, was not slow to endow the practice of dating
the Easter candle with a symbolic interpretation. John Beleth explained that the practice
conveyed the notion that time belongs to Christ, and at a different level of interpretation
observed, in a comparison that relates to saintliness or intensity of faith rather than to
actual numbers, that in the same way that the year symbolises Christ, the divisions of
the year stand for the twelve apostles, and the days represent the Christian faithful. 3
Honorius of Autun designates Christ as the acceptable year of the Lord; and extends
Beleth's symbolism to include baptised children, who are equated with the hours. 4 The
symbolism is echoed by Sicardus, who refers to Christ as both the summation and the
consummation of time, 'the ancient and the fulness of days' .5 Sicardus in the same
passage also refers to 'fertilitas fructuum in anno', perhaps echoing the words of Psalm
64 : 'You crown the year with your bounty'.
(iii) The Charta
It is said that the De Temporibus of Bede 'did much to establish the practice of
dating events from the Incarnation' (DCC). It is therefore significant that the dual
method of dating events used by Bede, that is, the year from the Incarnation and the
indictio, was also found on the Easter candle at Poitiers in the ninth century, and at a
1
1507 Missal fol.xciiii.
2
'/n memoriam populis revocans.' PL 95.6670.
3
Rationale. PL 202.112A.
4
Gemma Animae. PL 172.6670.
5
Maximus annus antiquus et plenus dierum. Mitrale, PL 213.3230.
279
number of other churches. 1 The inclusion of this second item of information which, it
may be argued, is both superfluous and of relevance only within a fifteen year period,
may be attributable to Bede. It is significant that in AD 701 the inscription on the
candles at Rome, as recorded by Bede, did not mention the indictio.
The inclusion of the epace from the eleventh century onwards not only provided
an additional piece of chronological infonnation,3 but underlined the prominence and
importance of the Easter candle at Eastertide, almost as a subtle reminder. Although in
this respect it recalls the purpose of the inscription on the eighth-century Roman Vigil
candles, there is no direct evidence to link the recording of the epact with the earlier
Roman practice. As with the inscribing of the year, medieval commentators attached a
symbolic interpretation to the indictio4 and to the epact. Sicardus wrote that the indictio
epitomised the actions of men, and the epact the succession of ages and the passing of
5
time. Both, he declared, were ordained and disposed by Christ.6 Later instances of
churches which conf'med the information to the year, the indictio, and the epact are rare.
In the influential Roman rite the insignia were absent. Elsewhere, and especially in
France, the inclusion of additional information resulted in the emergence of the charta.
However, at Barking at the start of the fifteenth century the custom was maintained of
recording only the year, the indictio, and the epact; 7 whilst at the end ofthat century it
was still observed in the Cathedral of Prague.8
The numerals and letters of the insignia may have been incised into the wax like
the cross. If so, the incisions would have to have been filled in with some form of colour-
ing, in order to make them conspicuous and visible from a distance. It is equally likely,
however, that the information was painted onto the wax of the Candle. For not only
could the information be then read with ease; should the same Candle be required for
use the following year, the alteration of the numbers could much more easily be accom-
plished. At the beginning of the twelfth century Cistercian practice was to write the
year, the indictio, and the epact on a cartula or small piece ofparchment.9 This was
done in all probability because the Cistercians used only a small Easter candle, since the
1
Poitiers p.215. Also at Gembloux (Albers IT p.99); Essen (Ordinal p.58); and Nidaros
(ONE p.232).
1be age in days of the moon on 1 January of a given year.
3
According to the Customary of Farfa the indictio was not included. Albers I p.54.
4
Beleth also refers to it as the 'aera'. PL 202.112A.
5
So also Durandus, Rationale VI.80 p.351.
6
Mitrale, PL 213.3230.
7
HBS 65 p.101.
8
1498 Missal fol.xci.
9
Nom.Cist. p.104. The number of entries was subsequently increased to eleven. It was
customary for the armarius (treasurer) to produce the charta (Martene, DAMR 3.15.8
p.142), as at Farfa (PL 150.1203C) and Fleury (Martene, ibidem.).
280
surface available for inscribing moderately-large letters was inadequate for the repre-
sentation of the necessary information. Tills is the earliest recorded instance of the use
of a charta.
The precedent having been set, the charta was not only adopted into the rites of
other religious orders and churches; but within a relatively short space of time it was
realised that it was possible to include more liturgical and religious information on the
parchment by increasing its size. Merati informs' us that the cycle of liturgical feasts
for the whole year was originally incised into the wax of the Candle and only subse-
quently transferred to the parchment. Tills is borne out by the Customary of Fleury,
according to which fourteen items of information were inscribed on the surface of the
Easter candle,3 and by the 1512 Missal of Spires.4 Merati also tells us that the charta
was also known as Breve anni, 'the summary of the year'.
Chartae varied enormously in length and correspondingly in the range of informa-

tion they conveyed. All retained the original nucleus of the year, the indictio, and the
epact. The Dominican chart had two additional entries relating to the order and the
dominicalletter; 5 whilst the Premonstratensians displayed the names of the Pope, the
abbot, the bishop, and the king6 as well as the movable feasts. In post-Tridentine times
some of the charts became even longer; and in addition to a large amount of biblical,
liturgical, and astronomical information included events of both medieval and recent
history. For instance, at the Church of St Gudila in Brussels the victory of John, Duke
of Brabant on 5 June 1288 at the Battle ofWoeringen was recorded. 7 As mentioned
above, the chart at the Monastery of Fleury had fourteen entries; whilst there were
eighteen at Reirns Cathedral, thirty-five at Chalons, and forty-eight at Rouen. 8 At the
last-mentioned cathedral the deacon read aloud the contents of the chart at the con-
clusion of the Preface.9 In many churches the names of the treasurer or sacristan and
1
Gavantus/Merati N p.155.
2
'0fficii ordo per totum annum.'
3
Novarinus p.17.
4
Agenda Spirensia fol.xcili.
5
Ceremonial of 1520 in King, LRO p.357. The indictio, epact, and dominicalletter have
disappeared from the present-day chart. See Appendix 14.
~ssal of 1578 cited by King, LRO p.190.
7
Zazzera p.300.
8
For the years 1585, 1708, and 1678, respectively. Martene, DAER 4.24 pp.146-7. For
the full text of all three, see Appendix 14.
9
De Moleon p.318.
281
precentor were also written on the vellum plaques. 1 It was also common practice to
place at the head of the chana a formula of benediction or consecration , introduced by
benedictus est cereus ... or consecratus est cereus .. ?
The custom of affiXing the tabula paschalis to the Easter candle survived within
the cathedral tradition until recent times. At Amiens, where a charta was in use until
1969, it was the responsibility of the Secrhaire General to make a new one each year.
The custom survives in some houses of the Dominican Order; but disappeared from the
Cistercian and Premonstratensian rites after the Second Vatican Council.
There were a number of reasons for the demise of the charta, foremost of which
was perhaps its absence from the Roman rite and the growing influence of that rite over
the years in France, where the tradition of affixing a chart to the Easter candle seems to
have been very strong, and lasted longest. The holding of the Paschal vigil on the
morning of Holy Saturday detracted from the ritual and ceremonial in general, and must
also be regarded as a contributory factor in its disappearance. Beaudin's assertion3 that
the display of information on the parchment destroyed the 'rapport intime' between the
message and significance of the information, and the Easter candle, is justified in the
same way that the insertion of the grains of incense into the candlestick rather than the
Candle detracted from their significance and reduced their purpose to the level of mere
ornamentation. On the positive side it could be argued that this loss of rapport resulted
in the chart acquiring an identity independent of the Easter candle, and an informative
role comparable to the depiction of biblical scenes in stained glass windows. It must be
remembered that the Cistercians used a chart for more than 800 years, and many of the
French churches employed this device for centuries.
The chart was normally affixed to the lower part of the Easter candle, the central
portion of the stock being reserved for the grains of incense. Paschal's statement4 that
the charts were not always fastened to the Candle is both vague and frustrating since he
does not elaborate. Presumably some were fixed to the candlestick; and perhaps it was
not unknown for some to be displayed in a prominent place in close juxtaposition to the
Easter candle, for instance on the ambo. The parallel which Durandus draws5 between
the chart and the superscription above the Cross should not be adduced in support of the
view that the chart was aff'"txed to the upper part of the Candle. Such a position would
1
Paschal col.329.
2
See Appendix 14.
3
Le ci'erge pascal p.21.
4
5
Rationale Vl.80 p.351.
282
clearly not be practicable. The large chart at Rouen was fastened at a man's height 1 to
the huge column of wax that was the Easter candle, so that it could be read with ease.
At Essen the chart completely encircled the Candle; 2 whilst the custom at Spires
recalled earlier practice. For in that latter church the four entries, viz. the year, the
golden number, the day of the cathedral's consecration, and the name of the bishop or
burgomaster,3 were written on the wax rather than on the parchrnent. 4
Evidence for the physical appearance of the chart is meagre. The rubric in the
Ordinal of Essen states that the information should be written 'in large and beautiful
letters' (p.58); and De Moleon referred to the chart at Rouen as a fine parchment.5 It
would appear that the chart resembled the page of an illuminated manuscript. Those
executed until recently at Arniens belonged to this tradition. From the lower edge hung
two seals. One was made of green wax, the episcopal colour, and impressed with the
arms of the diocese; the other, of purple wax, indicating that the status of Ami ens
Cathedral was that of a minor basilica, was stamped with the arms of the cathedral
chapter. The seals were attached to the chart by ribbons of corresponding colours.
These devices are still preserved in the Treasury of Arniens Cathedral.6 For their chart
the Dominicans use the black and white shield of their order, and inscribe it with letter-
ing of contrasting colours. 7 That in use at Blackfriars in Oxford measures about eight
by six inches.
(iv) Portraits and decorations
Bemado Bisso records8 that it was the custom of some churches to paint the like-
ness of the bishop, or of the patron saint, or of any saint on the Easter candle; but he
does not give any indication of the extent of this practice. However, a French source in
the early part of the last century informs us that the Easter candle was usually decorated
with different portraits of saints.9 The same writer adds that edifying 'objects' made of
gold were also used as decorations.
1
De Moleon p.318.
2
0rdinal p.58.
3
It is not clear to whom presule refers.
4
1512 Missal fol.xciii.
5
'Beau velin'. Voyages liturgiques p.318.
~e writer is indebted for this information to the Secretaire General of that diocese,
L'Abbe P.Grey.
7
See Appendix 14.
8
Hierurgia I p.l80.
9
DHCR I p.470.
283
The practice of painting candles is of long standing. 1 It is mentioned by St Pauli-
nus in the fifth century. 2 Mention of colours (coloribus) in an inventory of accoutre-
ments for the Easter candle at York in the thirteenth century probably relates to the
paints that were used to decorate the Candle. Bellotte records that painted candles were
frequently used in the former ceremonies of the Church of Laon; but he does not men-
tion the Easter candle. 3 However, at Seville we are informed that paint was applied to
the newly-cast column of wax. 4 illustrations in some late nineteenth-century and early
twentieth-century missals depict zig-zags and other geometric patterns on the central
portion of the Candle's stock. It is probably these that Van Doren has in mind when he
criticises the candlemakers of his day for hiding the fragrant symbol of Christ's 'pure
and glorious flesh' behind strips of coloured paper.5
(v)Flowers
The adornment of the Easter candle with flowers is first attested in the Mozarabic
rite of the tenth century.6 The Candle was festooned with flowers or garlands during
the reading of the first prophecy. In Central and Southern Italy the surviving Exultet
rolls provide pictorial evidence for the practice in that part of the world during the elev-
enth and twelfth centuries; 7 and at York in the thirteenth century, flowers appear in an
inventory of accessories for the Easter candle. 8 Rock cites a reference in Pamelius'
Liturgicon to the twining of flowers around the Easter candle in an old Ambrosian mis-
sal.9 Their use as adornments of the Easter candle has survived to modem times - so
great is the force of tradition that, although not prescribed by any of the manuals of
Roman liturgy for this occasion, the Memoriale Rituum 'suggests their use "if custom-
ary" on certain days'. 10 In former times at Dixmude, Nieuport, Veyme, and other places
in West Flanders, branches were fastened to the candlestick in addition to the flowers
and leaves. Known as the 'Paschal Tree', it was a visual expression of the arbor de cora
1
Candle-painting was formerly a trade in its own right. Pierin del Vaga was one such
craftsman.
2
De S.Felice Natalatium, Carmen VI (PL 61.491B): Ast alii pictis accendant lumina
ceris.
3
RELR p.812.
4
Doblado p.299.
5
Le Cierge Paschal p.15. In a communication dated 1 August 1988 with a leading finn
of candlemakers in England, the present writer was informed that the handmade-candle
foreman could not recall such designs in the forty-two years of experience in his craft.
6
Antiphonary of Leon p.284.
7
Avery, Plates XLVI, LXV, and LXIX.
8
Bradshaw and Wordsworth IT p.98.
9
Church of our Fathers I p.167. The present writer has failed to locate this reference in
either the cited or the original work. It is just possible that it existed as a marginal anno-
tation in one of the above-mentioned books.
1
~ortescue and O'Connell (11th edition) p.29.
284
et fulgida. 1 Today, at Annecy not only is the candelabrum festooned with garlands; a
vase of flowers is placed in front of it to show the importance of the Easter candle? In
most cathedrals and churches it is usual to leave the Candle adorned with flowers for the
whole season of Easter. At Vannes Cathedral, however, flowers are used only at the
Vigil itself; whilst at Lyon they remain in position only till the end of Easter week. 3
Their use in pre-Christian religious rites may have been a contributory factor in
the adornment of the Easter candle with flowers. However, it is much more likely that
the practice of decorating the Candle with flowers was an internal liturgical develop-
ment, suggested by references in the laudes cerei to the source of the beeswax. For not
only did the presence of flowers fulfil a decorative function in furnishing a floral setting
or foil for the Candle; it provided a physical and tangible complement to the floral allu-
sions of the Song, a visual representation of one of the themes of the Preface, and
served as a forceful reminder of the origin and source of the beeswax. In later centuries,
when references to the bees and the flowers had all but been excised from the taus cerei
in some churches, although the rapport between the wording of the Preface and the
flowers was subsequently lost, the floral decorations survived.4
In former times in Hertfordshire Holy Saturday was one of the great flow-
er-gathering days of the year. Particularly sought after was the rare Pasque flower
(anemone pulsatilla) with its purple petals.5
1
Callewaert p.l40.
2
Survey of 1984.
3
Survey of 1984.
4
References to flowers are found in four different Prefaces in the western rites. The
clause aliae vertunt flares in ceram of the Gallican Preface (Vich Sacramentary p.3;
Miss.Gall.Vet. p.36; Bobbio Missal p.77; S.Gall348 p.83; Jumieges Missal p.92) is par-
ticularly apposite in view of the juxtaposition of the flowers and the beeswax. In the
Milanese Preface (Manuale Ambrosianum p.201 and subsequent Ambrosian missals)
not only is the importance of the flowers mentioned; but" Christ is identified with both
the wax of the Candle and with the flowers: Quid enim magis accommodum magisque
festivum quam iesseico jlori floreis excubemus et tedis? praesertim cum et sapientia de
semetipsa cecinerit: Ego sumflos agri, et /ilium convallium. ('What is more fitting and
more festive than that we keep watch for the Flower of Jesse with floral torches?
especially when even Wisdom has sung of herself: I am the flower of the field, and the
lily of the valleys.') Similarly in the Beneventan Preface (Pinell, La Benedicci6 p. 96)
the importance of the flower to the bee is stressed: flore utuntur coniuge,flore fungun-
tur genere,flore domos instruunt,flore divitias convehunt,flore ceram conficiunt.
('They use the flower as a spouse; they gain their offspring from the flower; they
construct thir homes from the flower; they gather their riches from the flower; they
make their wax from the flower.') There is also a reference to flowers in the Mozarabic
benedictio cerei (Pinell, ibidem p.117). In this latter Preface it is unlikely that the
flowers, which were placed around the Easter candle during the reading of Genesis 1,
represented nature or creation.
5
Jones-Baker p.133
285
(vi) Crucifix
C.Callewaert records that in some Belgian churches an image of the Crucified was
attached to the Easter candle. 1 Like the grains of incense, this emblem of the Passion,
fastened to the waxen symbol of the Resurrection, betokened the mystery of the Cross
and bore witness that the Crucified had risen from the dead; in the same way that the
crown·of thorns, which encircles some Easter candlesticks, signifies the triumphal king-
ship of Christ.
(vii) Branches
At St Maartenskerk in St Ghislain there is a fifteenth-century candlestick near to

the crown of which three small branches spring from the main stem so as to form
candleholders vertical with and parallel to it. There is a copy in a church in Bruges. It
is said that the three candles represent the three Marys who went to embalm the body of
Jesus and who were the first witnesses of the risen Lord.2 At Capua the three candles on
a similar device were said to represent the Trinity .3 In addition to three, Callewaert also
mentions candlesticks with two, four, and six branches, but does not elaborate. It is not
difficult to see in the latter type the survival or development of the menorah.
1
De Paaschkandelaarp.141.
2
Callewaert, ibidem.
3
286
Chapter Ten
THE BLESSING OF THE EASTER CANDLE: THE FORMULA
(i) The Romano-Gallican A and Milanese traditions
The structure of the formula found in both these traditions is tripartite :
(i) The invitatory proclamation or preconium

(ii) The Dialogue
(iii) The Preface or taus cerei
In view of the survival of other elements 1 from the Lucernarium in the ceremonial sur-
rounding the lighting and blessing of the Easter candle, it is not difficult to discern the
origin of the Dialogue and Preface in the latter ritual, and to recognise that the dialogue
and prayer of thanksgiving in the Apostolic Tradition 2 are liturgical features expanded
and transformed into the second and third sections of the later formula.
It is not known when the structure of the formula achieved its present form. The
scheme of Exu/tet/Dialogue/Preface, which still obtains today, had already been fixed
by the eighth century; but the inclusion of this non-Roman ceremony into both the
Gregorian and Gelasian Sacramentaries must have been of recent occurrence in that
century. The first element, the preconium proper, serves as an introduction to the Pref-
ace; but the date of its incorporation into the formula as a whole cannot be determined
before the eighth century by surviving documentary evidence. An earlier date,
however, can be inferred with confidence in view of the age of Ennodius' laudes cerei. 3
We have seen that the Dialogue was a feature of the Lucernarium. With regard to the
third element, the Preface, there can be little doubt, to judge from the surviving Bene-
ventan, Gallican, and Milanese Prefaces, with their expansive references to flowers and
bees, that poetic language was a characteristic of the laudes cerei to which Jerome took
1
See page.224.
~e Dialogue, followed by the prayer of thanksgiving for light, divine, natural, and
man-made, lacks '"Up with your hearts" because that is said only at the offering' (Cum-
ing, Hippolytus p.23 §25). With the development of the Lucernarium into the blessing
of the Easter candle, which included the offering of light, the omitted couplet was
incorporated.
3
The close similarity between the Romano-Gallican and Milanese preconia, and their
omission from the beginning of both of the Prefaces of Ennodius (to be dated to
c .AD 51 0) would suggest that the wording of the Exultet had become fixed in a formula
common to a number of western rites at an early date.
-287
exception. If so, the Prefaces must have achieved their fmal form and contained the
same themes, familiar from later compositions, well before the end of the fourth cen-
tury.
There is evidence in the fourth century that the taus cerei could be composed
either in prose or in verse. Capelle has shown that the Preface of the Romano-Gallican
tradition was in all likelihood written by St Ambrose. 1 On the other hand we have St
Augustine's own clear testimony that his laus cerei had been written in verse. 2 The
three extant lines of his Preface show that he had composed it in dactylic hexameters.
Archdale King, amongst others, is of the opinion that the hymn Ignis Creator Igneus,
found only in the Antiphonary of Bangor/ is the sole remaining example of a taus cerei
in verse, and may have been written by St Ambrose, himself a no mean composer of
hymns. 4 There is also in existence the Escoriat Preface, a taus cerei composed in
verse,5 and the poem of Drepanius Florus, De Cereo Paschali, with its references to the
composition of the wick and the apian origin of the wax.6
A study of the composition and content of the Preface is strictly beyond the scope
of this work; but it has to be observed that the Prefaces of the Romano-Gallican , Mila-
nese, and Beneventan liturgies, and also the taudes cerei of Ennodius contain the same
themes or motifs; and that the offering of light and the eulogy of the bees are elements
peculiar to the Paschal Preface, and represent a development and expansion of the
simple prayer of thanksgiving found in the Apostolic Tradition.
Romano-Gallican Milanese Beneventan Ennodius I Ennodius II

Passover/ Lamb/ Resurrection Mystery of Mystery of
Resurrection Sacrifice Creation/ Creation/
Resurrection Renewal
Offering of Light Offering of Light Offering of light Offering of light
Light of candle of candle
Eulogy of bees Bees Bees Bees (Bees)
Eschatology Eschatology Divine Pro- Divine Protec- Renewal
tection tion
1
L'"Exultet" Pascal pp.219-46. The 1488 Missal of Basel attributes the work to the
bishop (fol.xcii); but both M.Huglo and Dom B.Fischer (whom Huglo cites on p.88)
contest its Ambrosian authorship (p.87).
2
De Civit.Dei 15.22. On the strength of this remark the Church for centuries regarded
Augustine as the author of the Romano-Gallican Preface, as is evidenced, amongst other
testimony, by the Sacramentary of Fulda (Schmidt I p.425), Grancolas (p.319), and even
the 1830 Missal of Limoges (p.220). Huglo (p.81) casts doubt on the authenticity of the
three verses.
3
HBS 10 p.ll.
4
LRCp.417.
5
Reconstructed text in Pinell, La Benediccio pp.97-100.
6pL 61.1087-88.
288
Similarly the Ignis Creator Igneus incorporates the themes of Passover/Resurrection,
the Candle, Light, and Bees. The three ritual Prefaces and the two laudes cerei of
Ennodius differ from one another in that each seems to place a greater emphasis on one
particular theme, or aspect of a theme, but without detriment to the overall structure and
recognised pattern of the composition. Prominent in the Romano-Gallican Preface are
the references and allusions to the Passover and the Crossing of the Red Sea. A strong
eschatological tone characterises the Milanese laus cerei; whilst in the Beneventan
Preface the theme of flowers and bees is particularly conspicuous. We have already
observed that Ennodius dwells on the physical characteristics and composition of the
Easter candle vis-a-vis the instances of God's intervention in history (pp.226-7).
In what could be interpreted as a posthumous vindication of Jerome's strictures,

the Roman Church excised from the Roman Preface the offending and incongruous
Virgilian language, leaving only two short references to the bee : de operibus apum and
apis mater eduxit. It was left to the revisers of the post-Vatican ll Missal to remove all
references to that creature, in view of the modem composition of Easter candles 1 which,
it was thought, rendered allusion to the bee meaningless.
(ii) The Romano-Gallican B tradition
l.The Deus mundi Conditor. Archdale King referred correctly, in the opinion of
the writer, to two types of Preface within the Romano-Gallican tradition, which he des-
ignated Type A and Type B.2 The latter has been discussed in the previous section.
Type A comprises the single prayer Deus mundi Conditor. It is found as the laus cerei
in the Gel asian Sacramentary, the Sacramentary of Autun, and in Ordo 30A;3 but as the
sole formula for the consecration of the Easter candle it is found nowhere else. In spite
of its clear reference to the Easter candle, Pinell called the formula an oracio Romana;4
and Deshusses excised it from the supplement5 to the Gregorian Sacramentary as one of
those blessings in codex R 'which are scarcely re<;.ognisable '.6 A brief examination of
the contents of this prayer, however, "reveals that its structure closely resembles that of
the Romano-Ga1fican Preface. It is true that the offering of light is placed frrst; but the
formula is not strictly a Preface in that the opening words do not suggest a continuity of
the Dialogue, a characteristic of the standard Preface. Rather, it resembles the initial
1
See Harbert p.240.
2
LRCp.417 .
3
. GeV (Mohlberg) p.68; GePh p.63; OR 30A.l5.
4
La Benediccio p.85.
5
Considered by that author to have been compiled by St Benedict of Aniane
c. AD810-l5.
6
Le Sacramentaire Gregorien p.42.
289
invocations of later prayers for the blessing of the new ftre. After the prologue, there
are three sections, introduced respectively by igitur, ergo, and igitur, whose themes are
the Resurrection, light, and bees. It closes with a further invocation recalling the escha-
tological themes and language of both the Romano-Gallican Preface and Ennodius'
laudes cerei.
1bis last pericope, here linked to the preceding section by ergo, became detached
from the rest of the prayer, and came to constitute a formula of blessing in its own right,
the Veniat quaesumus! The Deus mundi Conditor (without the Veniat quaesumus)
subsequently became a prayer for the blessing of the new f'"rre on Holy Saturday within
the Germanic tradition. 2
2. The Deus mundi Conditor and the Romano-Germanic Exultet/Preface. The

presence of both these benediction-formulas for the Easter candle in the Sacramentaries
of Angouleme and Gellone3 would at fust sight appear to constitute an unnecessary
duplication of consecration. As with the so-called double blessing in the Mozarabic
rite, the difficulty largely disappears when it is realised that the Deus mundi Conditor is
an invocatory blessing; whilst the Exultet/Preface formula has much in common with
the notion of berakah. In the Spanish rite an interval of time and space separated the
two blessings. In the Gallican tradition, since both formulas were pronounced concur-
rently, it was perhaps inevitable that one of them should disappear as a prayer of
consecration- as indeed did happen- or be diverted to serve some other pwpose.4 In
fact, within the Germanic tradition the Deus mundi Conditor became a formula for the
blessing of the new fire on Holy Saturday, as we have already noted; though in that
development it shed its fmal pericope.
3. The Veniat quaesumus. The fmal pericope of the Deus mundi Creator, once
linked to that blessing by ergo, subsequently became either a prayer for the blessing of
the new fire, or in the majority of churches including that of Rome, the prayer for the
blessing of the five grains of incense. However, in a small number of churches within
the Gallican tradition it survived as the formula for the blessing of the Easter candle,
once the major portion of the Deus mundi Conditor ceased to fulfil this function. These
French churches thus represented the survival of the tradition attested above in the
1
Also found as a blessing for the new ftre (pp.l24-5), as a blessing for the grains of
incense (p.263), and as a blessing of the cereus minor (p.291).
2
As such, it is found in PRG (IT p.95), and was used at Abo (Manual of c.l522 p.238), at
Mainz (1507 Missal fol.xcii), at Salzburg (1507 Missal fol.xciii), and at Ratisbon (1570
Ritual np).
3
GeAng p.52 and GeG pp.92-3, respectively.
4
According to the Pontifical of Poitiers the prayer was used to commission the three
apotropaic candles lit at the 'altar of the fonts' (p.216).
290
Sacramentaries of Angouleme and Gellone. At Bourges, Perigueux, and Le Puy 1 the
Easter candle was blessed by means of invocation by the celebrant with the Veniat
quaesumus, after it had been kindled with the new ftre, but prior to the chanting of the
Exultet by the deacon. This double benediction was comparable to Mozarabic practice.
At Amiens, where the Easter candle was consecrated according to the Roman rite, the
Veniat quaesumus was used to bless the cereus minor.2
(iii) The Mozarabic tradition
Mention has already been made of the double blessing of the Easter candle in the
Mozarabic rite, reminiscent of the practice attested in the Sacramentaries of Angouleme
and Gellone. In the former rite the lamp was also blessed with the use of both an invo-
catory formula and a berakah-type benediction. Neither the blessing of the lamp nor the
blessing of the Candle are preceded by a preconium corresponding to the Exultet, and
the Dialogue appears as the detached triple acclamation of Deo gratias. The themes of
the Candle-Preface are light, Baptism, and the composition of the Easter candle. The
latter recalls the laudes cerei of Ennodius.
(iv) The Exultet!Dialogue/Preface as a blessing
Within the Roman rite the chanting of the Exultet and the Preface by the deacon is
analogous to the reading of the Gospel. During the performance of this blessing, which
has thus acquired the status of a Gospel-passage, the congregation remains standing,3
those in the choir turning to face the deacon. At Milan, prior to the liturgical revisions
that followed the Second Vatican Council, a lengthy interruption occurred at the end of
the first pericope of the Preface. During this a subdeacon and the sacristan withdrew to
the sacristy to fetch the lamp containing the new flre. On their return the Easter candle
and the two Vigil-candles were lit. 4 This in effect divided the Preface into two distinct
blessings, as is clear from the rubric which follows the words in veritate proveniunt at
the end of the first pericope : hac benedictione finita, 5 and from the fact that at that junc-
6
ture the congregation sat down and remained seated until the conclusion of the Preface.
1
1741 Missal p.225; 1782 Missal p.159; 1783 Missal p.159, respectively. Apropos of
this prayer used in the same context Atchley (p.139) cited the 1845 Missal of Parniers.
(The copy which he consulted was destroyed during Hitler's war.) He was in error,
however, to include the Pontifical of St-Germain-des-Pres. For this document, see
above, Pt ll Chapt.3 p.125.
2
1752 Missal p.182.
3
Fortescue and O'Connell (11th ed.) p.306. It is also prescribed in the 1669 Cistercian
Missal p.155.
4
1560 Missal fol.llO.
5
'When this blessing is fmished.' 1475 Missal fol.lxxx.
6
1560 Missal fol.llO.
291
At Lyon the Exultet was chanted after the reading of the prophecies. 1 1bis prac-
tice recalls the arrangement attested in Ordo 29? Innnediately prior to the Exultet the
first litany was sung. 3 1bis also occurred at Paris, Besan~on, and Lu~on, 4 and at
Bourges.5 At Coutances, Bayeux, Rouen, and Norwich an antiphon preceded the
Exultet. 6
There may be some link between the Milanese ritual and the practice attested in
PRG. According to the rubrics of the latter (II p.97), the Exultet was termed the first
blessing of the Candle and was read by the archdeacon quasi in modum legentis. 7 He
then raised his voice for the Dialogue and presently began to chant the Preface. 8
1
1771 Missal p.226.
2
0R 29.48.
3
1510 Missal of Lyon fol.lxix.
"These three churches are cited by Jounel, La Semaine Sainte p.147.
5
1741 Missal p.226.
6
1557 Missal fol.lxviii; Semaine Sainte p.495; 1497 Missal np; and HBS 82 p.91,
respectively.
7
Literally, 'as if in the manner of one reading'.
8
lnde vero accedit in consecrationem cerei, decantando quasi canonem : 'Then he began
the blessing of the Candle in the same way that he would chant the Canon (of the
Mass)'.
292
Chapter Eleven
THE UGHTING OF THE EASTER CANDLE
(i) Before the Exultet
Two traditions are found within the primitive practice of lighting the Easter candle
before it was blessed. (i) The procession into church with the Candle previously
kindled at the new fire has been discussed in Part ill. 1 (ii) Within the second tradition
the Easter candle was lit just prior to the commencement of the Exultet, and to this
tradition belong all those churches in which the Easter candle was either borne into
church unlit, or was already in position by the altar or the ambo, prior to the start of the
ceremony. The tradition is attested in a number of early sacramentaries and pontiftcals,2
in Alcuin and the Regularis Concordia,3 and is mentioned by Micrologus and Honorius
of Autun. 4 The latter adds that the kindling occurred at this point because 'Christ has
the light from the beginning'. The practice was also observed at Monte Cassino in the
twelfth century. 5 At the beginning ofthe eighteenth century De Vert found that the
Easter candle was lit prior to the start of the blessing in a large number of churches;6
and this is borne out by the rubrics of a number of French diocesan missals.7 Thurston
drew attention8 to the fact that the clause in the Roman Preface
(columnae) ...quam in honorem Dei rutilans ignis accendir
is a form of internal rubric, and implied, at the time that the Ia us was composed, that the
Easter candle was already lit. The lighting of the Candle prior to the blessing meant that
the Exultet and Preface were sung without interruption. For in the older documents,
cited below in Note 2, the ritual involving the grains of incense was unknown; whilst in
the rites of the first four French cathedrals mentioned below in Note 7, the grains of
1
Chapter 5 §1.
2
GeV p.68 §425; GePr p.55; GeAng p.52; OR 29.48; PRG IT p.97 §346; Poitiers
r,.215. It can be inferred in Ordo 26 with confidence.
PL 101.1216B and PL 137.494C, respectively.
4
PL 151.1016B and Gemma Animae, PL 172.668C, respectively.
5
PRX//Ip.293 (=M 1139).
Taschal col.330.
7
Carcassonne (1749) p.196; Perigueux (1782) p.159; Le Puy (1783) p.159; Meaux
(1849) p.169. Additional evidence for the practice comes from Albi (DAER 4.24.8
p.147 and Lyon (1487 Missal cited by King, LPS p.61), and from Ordo XIV (PL
78.1218).
8
The Exultet and the Paschal Candle p.517.
9
'Which the ruddy f""tre has kindled to the honour of God.' Another intetpretation of the
word italicised by the writer is discussed below in the next section.
293
incense were inserted before the deacon began the Exultet. Reference to Table 52
shows that in these four French cathedrals the lighting of the Easter candle prior to the
Exultet vis-a-vis the insertion of the grains of incense, also at a point before the blessing
of Candle, represents the first stage in the development of these two ritual actions as
features which subsequently interrupt the Preface in most western rites.
According to two twelfth-century Roman pontificals' the Easter candle was also
lit before it was blessed. These documents are representative of Stage 2 in Table 52,
since they both attest the insertion of the grains of incense during the Preface. The evi-
dence of later Roman service-books, however, is confusing. The twelfth-century Ordo
of the Lateran Church, Durand's pontifical, and the slightly later Bindo F esulani place
the kindling and insertions during the Preface,2 as does the Missale Romanum and all
subsequent Roman Missals up to and including the Tridentine Missal of 1570. Ordo
XIV and CA 1706, on the other hand, concur with PR Xl/. 3 As we noted when we dis-
cussed the triple Lumen Christi, a certain fluidity of ceremonial existed within Roman
ceremonial up to 1570.
One of the most noticeable alterations to the Paschal ceremonies, resulting from
the liturgical revisions of 1955, was the restoration of the primitive practice, which had
survived in the four above-mentioned cathedrals as late as the eighteenth century, of
singing the Exultet and Preface without interruption, the lighting of the Candle being
transferred to the beginning of the ceremony, its position of old.
Stage 1. Easter candle lit before Exultet.

Incense grains inserted before Exultet.
Stage 2. Easter candle lit before Exultet.
Incense grains inserted during Preface.
Stage 3 Easter candle lit during Preface.
Incense grains inserted during Preface.
Table 52
(ii) During the Preface following the Exultet
The practice of lighting the Easter candle at ignis accendit in the Preface is first
attested in the tenth-century Sacramentary of Corbie. 4 In almost all other rites which
1
PR XII I xxxii.8 p.240 and the Pontifical of Apamea (DAER 4.24 p.l60, M25).
2
0EL p.61; PGD ill p.588; and ZRKM p.214, respectively.
3
PL 78.1218C and ZRKM p.214, respectively.
4
PL 78.336B.
294
perpetuated this tradition, it is at this same point that the Candle is lit. 1 Mention of the
practice at Rouen in the following century ,2 and its inclusion by Lanfranc in his Decrees
and by Ulric in the Customary of Cluny/ together with the evidence from Corbie,
strongly suggest a monastic provenance in Northern France. Its adoption and use within
the monastic tradition4 ultimately led to its appearance in most of the cathedral rites in
the Romano-Gallican tradition, including that of Rome, and in the rite of Milan.5
It is not difficult to see why the practice developed of kindling the Easter candle
during the Preface. Elsewhere6 we observed that the censing of the Easter candle and
the insertion of the five grains of incense during the Preface were the result of the
ambiguous interpretation of incensi huius sacrificium, a phrase which invited the oppor-
tunity to match word with action. The intrusion of the ritual action into the Preface was
further facilitated by the fact that the words which had come to be regarded as an
internal rubric occurred at the start of a new section of the Preface, introduced by igitur.
The break: having occurred in the Preface for the incense, the precedent was now set for
a further interruption, suggested, so the writer believes, by another ambiguity of lan-
guage. For at the conclusion of the very next sentence occurs accendit, 'has kindled',
which, as we have already seen, indicated that the Easter candle was alight before the
start of the Exultet. Since, however, accendit could equally be construed as being in the
present tense of the verb, it is the belief of the writer that the alternative translation
'kindles' invited the lighting of the Easter candle at this point, and created a materialisa-
tion, as it were, of the primary concept inherent in this verb. It thus became an internal
rubric or cue for the deacon or official who applied the fire to the wick of the Candle
(iii) The Agent
As a rule the Easter candle was lit by the officiant, usually a deacon, who chanted
the Exultet and Preface; and it is likely that the deacon performed these two duties in the
fourth century, when the ceremony is first recorded. According to some early sacra-
1
In Haymo's Ordo Missalis (Van Dijk IT p.246) the Candle is lit at divisus in partes; in
the 1543 Missal of Bayonne (p.46) at sed iam columnae; in the 1568 Missal of Palencia
(fol.cvi) at reddit ecclesia; in the 1543 Missal of Paris (fol.lxxx) at praeconia novimus;
and in the Ordinary ofTongres (p.165) and the 1518 Missal ofRatisbon (fol.cxvii) at
suscipe, sancte pater.
2
Acta Vetera, PL 147.176C.
3
PL 150.467C and PL 149.663B, respectively.
4
By the eleventh century it featured at Fruttuaria and Vallombrosa in Northern Italy
~Albers IV p.65 and p.249, respectively).
Beroldus p.110. In the Milanese rite the opening words of the second pericope ecce
iam ignis columnae resplendet were acclaimed three times by the deacon, the congrega-
tion responding Deo gratias after each cry. Martene, DAER 4.24.11 p.148. For four
Gallican exceptions, see Section 1 above.
6
Chapter 8 pages 273-4.
295
mentaries and other service-books 1 the officiant was the archdeacon. This dignitary
also functioned in this capacity at Vienne and Soissons;2 whilst at Troyes it was the
senior archdeacon. 3 At Naples4 and in the Mozarabic rite5 the bishop lit the Easter
candle; whilst Beneventan practice allowed either the bishop or a priest to perform the
task.6 In the churches listed below in Table 53 the Candle was lit by a variety of other
officials and clerics.
Acolyte Monte Cassino (DAMR 3.15.10 p.143, M 1139

Lisieux (possibly)
(1752 Missal p.193)
Serjent LeMans (1789 Ceremonial p127)
Subdeacon Rome (HBS 17 p.175)
Albi (DAER 3.24.8 p.147, M 1)
Sacristan Lanfranc (PL 150.467C)
Chartres (13 C. Ordinal p.l11)
Provost St-Germain-des-Pres (DAMR 3.15.7 p.142, M 1165)
Precentor Premonstratensians (King, LRO p.l90)
Table 53
In none of the above-listed rites and documents is there any indication as to why the
officiating deacon did not light the Easter candle. In some instances it may have been
thought that the deacon would be distracted from his chant if he performed this duty. In
other churches the height or location of the Easter candle would have rendered this task
difficult for the deacon. 7 However, according to the service-books of the vast majority
of the churches within the Roman, Gallican, and Germanic traditions, as well as in the
Milanese rite, the lighting of the Easter candle was performed by the officiating deacon.
8
In the papal ceremonial the task fell to the junior cardinal deacon. Although the
Roman Missal of 1474 mentions a subdeacon,9 the Tridentine Missal specifies the
officiating deacon.
'GeV §425 p.68; Gefr p.55; GeAng p.55; OR 30A.l5; PRG II p.97.
2
De Moleon p.23 and Martene, DAER 4.24 p.161, M 305, respectively.
3
1736 Missal p.228.
4
Mallardo p.33.
5
Leon Antiphonary p.280.
6
Hesbert p.188.
7
As at Durham and Leau (p.237). Both Gavanti (p.166) and Bauldry (p.191) state that
the Easter candle could be removed from its holder to allow its being kindled. The Pol-
ish Manual also permitted this concession to facilitate the insertion of the grains of
incense (p.477).1n the Hereford Missal accendit is sustained for several notes, thus
allowing time for the Candle to be lit at that very word (p.l 02).
8
PR XII I.xxxii.7 p.239; C.A.1706 (ZRKM p.213).
9
HBS 17 p.175.
296
(iv) The Means
In the Milanese rite the bringing in of the lamp lit with the new ftre was almost
certainly a survival from the Lucernarium. Within the Romano-Gallican tradition the
Easter candle was lit with either the serpent-candle, the reed-candle, or the triple candle.
In France the practice emerged of transferring the new ftre from the single processional
candle or one of the triple candles to the wick of the Easter candle by means of a
sulphur-coated splint (sulphuratum). None of the missals which attest this procedure'
states the precise point at which the trre was transferred. The mention of the triple
candle in two of the missals (Auch and Nantes) again demonstrates how this functional
means of bringing in the new frre had developed into a vivid and symbolic presentation
of the Trinity with an almost separate existence and purpose of its own.
1
Besan~on (1707) p.317; Toulouse (1832) p.211; La Rochelle (1835) p.191; Auch
(1836) p.192; Nantes (1837) p.203; Meaux (1845) p.169.
297
Chapter Twelve
THE EXTINGUISIDNG OF THE EASTER CANDLE
In the history of the rite before 1100 two traditions existed relating to the length of time
the Easter candle continued to bum following the Paschal vigil : (i) that in which the
Candle burned either contit1uously or intermittently until it was disposed of at the end of
Easter week, and (ii) that in which the Candle was lit at certain services throughout the
whole of the Easter season.
We shall have occasion to refer to the evidence of Micrologus and Honorius in the
next chapter when we consider the fragmentation of the Easter candle at the end of
Easter Week (pp.303-4 ). Suffice it here to state that neither writer gives any indication
as to whether the Easter candle burned continuously or intennittently during that period.
At Rouen in the eleventh century the Candle was lit at every mass in Easter Week. 1 It is
very difficult to know whether or not the practice at Vienne2 c .1700, where the Easter
candle was kept alight day and night until Easter Saturday, had survived over the cen-
turies, or whether the custom was a recent neo-Gallican revival? The alternative tradi-
tion in which the Easter candle was lit at every major feast until Ascension Day is first
encountered c.1150 in the Gilbertine rite; 4 but may have been known in the Cistercian
rite some fifty years earlier. 5
The disastrous results of leaving the Easter candle at Naples unattended during the
night of Holy Saturday were described above at the beginning of Chapter 4. Evidence
elsewhere for the continuous burning of the Easter candle during the remainder of Holy
Saturday and throughout all of Easter day is plentiful; and though the majority of it is
late, it is likely that in nearly all instances it attests a centuries-old tradition. The cus-
tom became established of extinguishing the Candle after Compline on Easter Day.
Some of the churches where this practice is attested are listed in Table 54 below.
1
2
De Vert D p.38.
3
At Tours Cathedral and the Collegiate Church of St Martin, also in Tours, the Easter
candle burned continuously until Low Sunday. (Guyet p.294.)
4
HBS 59 p.40.
5
According to Guignard (p.l17) the Easter candle should remain where it was blessed
until Ascension Day. It is difficult to believe that it remained unlit during the whole of
this period.
298
Church Date Source
Rouen 11 C. PL 147.176C
Cistercians 1119 Nom.Cist. p.105
Gilbertines c.1150 HBS 59p.40
St Augustine's, Canterbury 13C. HBS 28p.274
Salisburyt 13C. HBS 91 p.24
Exeter 1337 HBS 37 p.322
Bursfeld c.1500 DAER 3.15.7 p.142, M 1179
Paris 1662 CeremonjaJ p.179
Amiens c.1700 De Vert II p.38
Angers 1731 Ceremonial p.261
Bourges 1741 Missal p.233
Lisieux 1752 Missal p.194
Reims* 1770 Missal p.213
Lyon 1771 Missal p.203
Tours 1784 Missal p.200
Coutances 1825 Ceremonial p.329
LaRochelle 1835 Missal p.193
Autun 1845 Missal p.247
t According to the Sarum Missal (Warren I p.270) it was extinguished at Vespers

on Easter Day, the office also prescribed by Lanfranc (PL 150.476C) and
enjoined in the Premonstratensian rite (Missal of 1578 cited by King, LRO
p.l90).
:j: The rubric of thls missal banishes any doubt that the Easter candle burned
throughout the night of Holy Saturday : 'The Easter candle should bum during
the whole of Saturday, the night whlch follows, as well as on Easter Day continu-
ously until Compline.' Direct evidence for the burning of the Candle during the
night of Holy Saturday also comes from Laon (Bellotte p.814) and the 1597
Missal of Metz in whlch the rubric was a quotation from the Preface : flamma.ts
eius Lucifer matutinus inveniat, 'let the morning star fmd her flames'.
Table 54
At V allombrosa it remained lit usque mane; 1 and the Ordinal of St Mary's, York refers
to the Candle being extinguished the following day without specifying at whlch service
thls took place.Z At Naples in the eighth century we learn that it was put out after Mass
on Easter Day. 3 The 1845 Missal.ofMeaux enjoins that the Easter candle should bum
at each service on Easter Day (p.173). This could be interpreted to mean that the
Candle was extinguished after every service on that day; but the significance of the
Easter candle vis-a-vis the importance of Easter Day makes thls unlikely. Moreover
there is no other recorded instance of such a practice.
1
Albers IV p.220.
2
HBS 75 p.292.
3
Mallardo p.22.
299
In some churches the Easter candle remained alight beyond the evening of Easter
Sunday. At Sens it was extinguished after Lauds on Easter Monday; 1 whilst it burned
continuously until Tuesday, presumably until after Compline on that day had been sung,
at Verdun 2 and Oermont-Ferrand? The Sarum rite, followed by that of Exeter,4 pre-
scribes that the Candle be lit for Mass, Matins, Vespers, and Compline on Easter Day,
Monday, and Tuesday.5 Roman practice, defmed in the decree of the S.C.R. dated 19
May 1607, stipulated that the Easter candle should be lit on the three days of Ea~ter only
at Mass and at Vespers. 6
In the period between Easter and Ascension Day universal practice, which
included the Roman as permitted by the above-mentioned decree, was to light the Easter
candle on all intervening Sundays. The decree of 1607 added that other customs, if
occurring during Eastertide, should be kept. A considerable number of churches7 did in
fact light the Candle on all major feast days between Easter and Ascension Day. After
Compline on Easter Day in the Cistercian rite the Easter candle was not lit again until
Vespers of Ascension Day. 8
It was perhaps inevitable that the time which the Easter candle remained in church
should be increased from seven to forty days, seeing that the Candle had come to repre-
sent the visible presence of Christ on earth after his resurrection. 9 Its removal from
church symbolised his disappearance from human sight. In the majority of churches,
including those of Rome, Lyon, Braga, and Milan, the Easter candle was extinguished at
the end of the Gospel on Ascension Day, and removed from church at the end of the
service. In some churches variant traditions had grown up over the years, so that we
fmd that the Easter candle was extinguished fmally at other times :
1. End of Mass on Ascension Day- Old Carmelite rite (King, LRO p.268).
2. Friday after Ascension Day- Salisbury (Warren I p.270) and Exeter (HBS 37
p.322).
'Missal of 1715 p.244.

2
Albers V p.123. The Customary of St Vitus states that the Easter candle is not extin-
guished 'until the third day'. De Vert mistakenly interpreted this 'Wednesday' (Expl-
ication n p.38).
3
De Vert IT p.38. See also previous note.
4
HBS 37 p.322.
5
Warren I p.270.
6
DACSR and Philippeau p.146.
7
For instance, Exeter (HBS 37 p.322); Gilbertines (HBS 59 p.40); Bursfeld (DAMR
3.15.7 p.l42, M 1179); St Augustine's, Canterbury (HBS 28 p.284); Poitiers (1767
Missal p.253).
8
King,LRO p.l04.
9
Desideri p.l51.
300
3. Compline on Friday after Ascension Day- Premonstratensians (King, LRO
p.190).
4. Compline of Ascension Day- Cistercians (King, ibidem).
5. After None on Ascension Day- Portuguese custom mentioned in decree of
S.C.R. dated 20 December 1783.
6. Trinity Sunday- Worcester (Feasey, The Paschal Candle p.357).
7. Vigil of Pentecost- Albi, Paris, Rouen (Feasey, ibidem).
8. 'At Pentecost'- Bursfeld (Martene, DAMR 3.15.7 p.142).
9. Compline of Pentecost- Nantes (1837 Missal p.205).
10. At assumptus est in coelum in the Gospel for Ascension Day - Soissons (1745
Missal p.169) and Calmrs (1760 Missal p.173).
11. At the 11th hour on Ascension Day- Tulle (Martene, DAMR 3.15.7 p.142).
12. Wednesday after Ascension Day- Durham (Raine p.9).
301
Chapter Thirteen
THE DISPOSAL OF THE EASTER CANDLE
Five different ways are known in which the Easter candle was disposed after it had
served its purpose either at the conclusion of Easter Week or at the end of the Season of
Easter.
(i) In places, such as Spires, 1 where the Easter candle was large, or where only a
small portion of the wax had been consumed, the use of the same Candle was permitted
for the following and even for subsequent years. At Lyon it was prescribed that if the
same Candle were to be used the following year, it should not be blessed a second time. 2
Presumably this entailed only the omission of the Exultet and Preface from the Paschal
vigil. A sentence in Sicardus' Mitrale should not be cited as contemporary evidence for
the reuse of the Easter candle.
cereus renovatus et illuminatus Christum significat. 3

'
At first sight renovatus appears to mean 'renewed' and to suggest that the Easter candle
from the previous year is being used. However, the use of renovatus in this sense
would imply that the same Candle was used year after year- possible, but unlikely; and
his reference to the disposal of the Candle in a later passage4 shows that he does not
intend us to understand that the wax of the previous year's Candle was reworked to pro-
vide the Candle for the following year. It is much more likely that renovatus here refers
to the changed state of the Candle, from a mass of lifeless beeswax to the consecrated
and spirit-charged column in the focus of the Paschal vigil.
(ii)ln some places, such as at Westminster, the old Easter candle was reworked
with the addition of new wax;5 whilst at Seville the huge column of wax was broken up
and recast. 6 This corresponds to the contemporary practice of a number of churches
where the unused portion of the Candle is returned to the manufacturer for recasting.
1
Agenda (1512) fol.xcili.
2
3
'The Candle, renewed and lit, signifies Christ.' PL 213.323C.
4
PL 213.325A.
5
Feasey, The Paschal Candle p.361.
6
Doblado p.299.
302
(ill) According to the Constitutions of Walter Cantilupe the remainder of the
Easter candle was used for the manufacture of small altar candles and of candles for the
use of the poor, and for providing tapers at the funerals of paupers.•
(iv) The making and distribution of Agnus Dei's from the wax of the previous
year's Easter candle is a physical counterpart or an extension in a material dimension of
the practice, ftrst encountered in the West in the Mozarabic rite, of sharing the light of
the Candle with the assembled faithful. The receiving of the light on the candle of each
man and woman was thus paralleled by the distribution of the wax of the Easter candle
via the Agnus Dei, which imparted to each recipient whatever inherent virtue the Candle
was held to possess.
The making of these medallions is attested in Rome and the diocesesoutside Rome
as early as the eighth century.2 However fum evidence for their production from the
remains of the Easter candle is relatively late? It is true that Sicardus draws an analogy
between an Agnus Dei and a fragment of the Easter candle; but he does not actually
state that the former was composed of wax from the Candle.4
(v) An earlier form of the custom from which the above-mentioned practice
almost certainly developed demonstrates the awe in which the Easter candle came to be
held and the almost magical properties with which it was supposed to be endowed. It
also shows clearly that the superstitious beliefs of pre-Christian Europe, far from being
extinguished by the advent of Christianity, lived on vigorously as part of the subculture
of medieval Christian society. In the early part of the eleventh century Micrologus
records that during Easter Week fragments of the Easter candle were distributed to the
people 'for the fumigation of their possessions'. 5 Sicardus also states the purpose of the
fragments to be' ad fumigandos' .6 Small pieces of wax were presumably broken off the
Easter candle and were burnt in the homes of the faithful to render, it was believed,
through the permeation of the smoke both the house and its contents immune from the
assaults of the Devil. The purificatory theory of J.G.Frazer/ that f"tres were supposed to
avert hail, thunder, and lightning caused by witches, was not new to that anthropologist.
Grancolas wrote8 in the eighteenth centul."y that the lighting and blessing of the Easter
1
Feasey, Ancient Holy Week p.204. The funeral tapers are also mentioned by Wilkins I
r,.57l.
OR 26.7-8.
3
Piccolomini I p.137 and Grancolas p.319.
4
Mitrale, PL 213.325A.
5
Ad subfumigandum rebus eorum. PL 151.1016B.
6
Mitrale, PL 213.325A.
1
Golden Bough 10 p.342.
8
Commentarius p.319. ,
303
candle was held to be sovereign against lightnings, tempests, and the many dangers in
life. In fact the apotropaic virtues of the Candle were recognised as early as the time of
Ennodius. In his longer surviving Preface the request is made that a fragment of the
wax candle may be sovereign 'against blasts of wind and buffetings of storms' and 'a
wall for the faithful should an enemy attack'. 1
It is claimed2 that an allusion to the practice of distributing fragments of the

Easter candle to the faithful as talismans is contained in the prayer Veniat quaesumus,
which, as we already noted, was originally the concluding pericope of the Deus mundi
Conditor:
in quocumque loco ex huius sanctificationis mysterio aliquid fuerit deportatum,

expulsa diabolicae fraudis nequitia, virtus tuae maiestatis assistat.3
However, the evidence is ambiguous, since the medium of sanctification is not stated
and could be either the wax or the fire. For the practice of distributing the new ftre is
well-attested. 4
Honorius of Autun, who also mentions the practice, added that the possession of a
fragment of the Easter candle, which represents Christ, symbolised a share in Christ for
the faithful at the general resurrection. 5 This teaching also gives us a small insight into
how the Church achieved a sort of modus vivendi with some pagan beliefs and practices,
and at the same time reveals how the Church in her tum was to some extent influenced
by superstitious beliefs.
The tradition of fashioning talismans from the wax of the Easter candle survived
into the twentieth century. Bisso describes the use of charms in the house or in the
fields against illusiones diabolicae (amonst other evils) as a practice of the past;6 but at
Bourges in the eighteenth century globuli of wax, stamped with a cross, were distributed
after Mass on Ascension Day, and placed above the thresholds of houses as a protection
against storms. 7 More recently, wax crosses, made from the Easter candle, were fas-
tened to the doorposts of the churches of Capua, a practice still observed at the Monas-
'Pinell, Benediccio p.93.

2
Van Doren; La ceremonie dufeu nouveau p.77 note 3.
3
' Into whatever place a portion of this sanctifying mystery shall be carried, may the evil
of Satan's guile be driven thence and may the power of your majesty be present.'
4
See Part ll Chapter 8 p.185.
5
Quia Christus in resurrectione ultima fidelibus in prernio tribueretur. Gemma animae
PL 172.6670.
6
Hierurgia I p.l80.
7
1741 Missal p.233.
304
tery of Monte Cassino at the beginning of this century .1 In his History of Reims
Flodoardus (AD 894-966) records a story which further illustrates the supposed
miraculous potency of the Easter candle. When the bodies of Rufinus and Valerius
were being transferred to the cathedral, the Easter candle caught fire - a form of divine
recognition of their saintliness.2
'Latis p.127.
2
PL 135.326A/B.
305
Chapter Fourteen
THE SYMBOLISM OF THE EASTER CANDLE
In Chapter one we suggested that a beeswax candle was used at the Vigil of Holy Satur-
day in preference to the traditional lamp of the Lucernarium both because of its cosmic
significance and because its shape and size vividly symbolised the column of fire in
Exodus. 1 We noted that in Ennodius' laudes cerei the lighted beeswax candle not
merely symbolised, but actually existed as a tangible microcosm of creation. Although
the Romano-Gallican Preface, contains two eschatologi cal allusions, it has as its
pre-eminent theme the Passover and the passing of the Israelites through the Red Sea.
These events were understood to be prefigurations of the Christian Vigil and Baptism.
It was perhaps inevitable, therefore, in view of the dominant references in the

Romano-Gallican Preface to the events in Exodus that the Easter candle, which pro-
vided the visual symbolic link between the Old Testament narrative and the liturgical
re-enactment of that narrative at the Paschal vigil, should undergo an elaboration of
symbolism which emphasised the significance of the Easter candle both in its Old Tes-
tament setting and especially within the immediate context of the Vigil liturgy.
As early as Amalarius2 we fmd an intensification of that typology, characteristic of

medieval biblical interpretation, which sought to find the face of Christ present through-
out the whole of the Old Testament. Apropos of the Paschal vigil the Easter candle
readily lent itself to its being interpreted as a symbol of Christ, since the pillar of fue in
the Book of Exodus was seen to foreshadow the coming of Christ during the Christian
Passover. The identification of Christ with the Easter candle was made by a number of
medieval writers3 who elaborated the symbolism and significance of the large burning
column of wax.
Just as the Easter candle symbolises the presence of the Lord in the fiery column,4
which led the Israelites through the Red Sea from bondage into a new life, so within a
Christian liturgical context, in which it was carried before the catechumens to Baptism,
the Candle was seen to represent Christ leading the Christian faithful to a new life. The
spread of light from the Easter candle imparts a share in the merits effected by Baptism,
1
See pages 226-7.
2
Liber Officialis 1.1.18: Columna ista [of Exodus] Christum praefigurabat.
3
Hugh of St Victor, De Off.Eccl. PL 177.451-2; Rupert of Deutz, De Div.Off.
PL 170.171B; Honorius of Autun, Gemma Animae PL 172.668C; Robert Paululus
PL 177.451; Sicardus, Mitrale PL 213.324B.
4
Macri p.l42; Thurston, The Exultet p.509.
306
not only to those who receive the light, but to the inanimate lamps and candles of both
the church building and the hearths and homes of the faithful; 1 and foreshadows and
anticipates the ultimate redemption of all creation, the doctrine familiar from the laudes
cerei of Ennodius.
Beaudoin drew attention to the fact that the Easter candle symbolises both the per-
son and the work of Christ;2 and this is well borne out by the above-mentioned medi-
eval commentators and in particular by Durandus. 3 For him the Easter candle has a
threefold significance. At one level it represents both the new teaching of Christ and
the new life in Christ, available to all and symbolised in the sharing of the light of the
Candle. Like Rupert of Deutz and Honorius of Autun before him, Durandus identifies
the light of the Easter candle with the Holy Spirit. In the same way that the disciples
received the Holy Spirit from Christ, so all the candles should be lit from the Easter
candle.
Durandus also echoes Rupert when he writes that the light of the Easter candle
symbolises the Resurrection. Since the unlit Candle conveys the notion of Christ in
death and repose,4 so the actual kindling of the wax column's wick represents the very
instant that Christ arose from the dead. We noted above5 that at Naples the Easter
candle was fashioned to a great height in honour of the Resurrection, the size reflecting
the magnitude of the one who rose. Thurston, commenting on the largeness of the
Candle, added that a great light should typify the True Light.6
The third symbolic aspect of the Easter candle was suggested by the three physical
components of the lighted candle, not according to Ennodius' conception of the Candle
as a microcosm of creation, but using the analogy which Augustine drew between a
candle and a human being. 7 According to the African Doctor of the Church the bees-
wax, the wick, and the flame of a candle corresponds to the flesh, the soul, and the
intelligence of a man. With the identification of the Easter candle with Christ, it was
perhaps inevitable that the Ennodian conception of the Candle at a higher and cosmic
level should yield to the more readily-grasped personal Augustinian view of the Candle.
Since the analogy already existed between the supposed parthenogenetically-produced
1
For the taking home of the new fire, see Pt IT Chapter 8 p.185.
2
Le cierge pascal p.24.
3
Rationale VI.80 pp.350-1.
4
Rock, Hierurgia p.407.
5
Chapter 4 p.234.
6
Lent and Holy Week p.408.
1
Sermones lnediti I (PL 46.819). There is some uncertainty over the authorship of this
work.
307
wax of the bee and the human flesh of Christ, who was born of a virgin, a tripartite
identification of the wax, the wick, and the flame of the Easter candle with the person of
Christ was readily made.
In a wider sense the beeswax, which symbolised Christ's flesh, was also held to
represent his humanity, an analogy suggested perhaps by Clement of Alexandria, who
stated that wax was symbolic of human frailty.' Durandus also follows Augustine in
likening the wick of the Candle to the soul of Christ; but identifies the flame of the
Candle with Christ's divinity rather than his intelligence or intellect. Augustine had
viewed the candle in light of human existence. In more recent times it has been pointed
out that for the Christian the Augustinian view of the candle is a salutary reminder of
his own position and standing vis-a-vis the Easter candle as the symbol of Christ. For
he should find in the Candle an image of himself, since the wax, the wick, and the flame
symbolise respectively his body, his soul, and his faith. 2
Elsewhere3 we have observed how the custom of extinguishing the Easter candle
on or near Ascension Day considerably narrowed its significance, and how the directing
of the cries of Lumen Christi at the triple candle detracted from the importance of the
Paschal column of wax. It was generally regarded that the Candle represented the vis-
ible resurrected presence of Christ,4 and for that reason was extinguished at Ascension-
tide. One of the aims of the liturgical reforms of 1955 was to restore the Easter candle
to its former status as a symbol of the timeless and wriversal presence of Christ.
* * * * *
Arnalarius is also our earliest authority for attaching a symbolic interpretation to
the smaller candle which was associated with the Easter candle. 5 This single candle
stood for the twelve apostles, who accompanied Christ during his ministry and were
responsible for the spread of Christ's light. For Christ had said to them 'You are the
light of the world'; and the second candle is a liturgical reminder of Christ's words.
'Stromateis 4, cited by Novarinus p.l9.

2
Berliere, Le cierge pascal p.114.
3
Pt IV Chp.l2 p.300 and Pt 3 ChapA p.217.
4
DHCR I p.471
5
Followed by Sicardus,!E•.!!!:•and Durandus, Rationale VI.80 p.352. See also below,
Chapter 17 pp.338-9.
308
Chapter Fifteen
THE PROVISION OF LIGHT AT THE PASCHAL VIGIL
Little is known about the provision of light at the primitive Paschal vigil in the West.
The evidence of the eighth-century and ninth-century ordines, 1 however, shows that
prior to the commencement of the Vigil, the church was in total darkness.
Subsequently, it has been the universal practice !hroughout the western Church, even
when the Vigil came to be anticipated and held in the morning light of Holy Saturday,
to extinguish all the lights of the building prior to the start of the service. This is stiU a
feature of the 1970 Roman and 1986 Church of England rites. In the Middle Ages it
was sometimes the practice to extinguish the lights after the kindling of the new fire. 2
The darkened church provided a congruous ambience and created an appropriate

atmosphere for the reading of the lections which formed the principal feature of the
Vigil. At one level the reading of the Word in the gloom of night could be viewed as a
liturgical re-enactment of John 1:5;3 and at another the darkness of the church
symbolised the sin of the world soon to be dispelled by the light of the Resurrection,
once the lamps and candles had been rekindled with new or rehallowed fue. With the
anticipation of the Vigil in the afternoon and subsequently in the morning of Holy
Saturday, these dramatic and atmospheric results were lost.
There are no grounds for believing that the Paschal vigil was conducted in total
darkness. It is true that the Vigil-lections could have been memorised and recited
without the aid of the written word, so allowing the service to be conducted in a
complete absence of light; for we saw that in sixth-century Gaul Matins of Holy
Saturday was held in such circumstances.4 However, a comparison between the two
services is invalid, since at that office of Matins well-memorised psalms were chanted
by a relatively small group of monks. During the Vigil, however, lengthy portions of
Scripture were read to a congregation which included children. The difficulties
attendant upon holding this service in total darkness seem obvious.
1
0R 26.9 and OR 29.17; OR 31.67, respectively.
2
Missal of Lesnes (HBS 95 p.47); MR 1474 (HBS 17 p.175).
3
With the emergence and use of the Easter candle, the symbolism became more
appropriate.
4
See Part I Chapter 1 pp.3-6.
309
Nevertheless the service was not held in an abundance of light. In the next chapter
we will show that at the old Roman Vigil illumination for the purpose of reading was
provided by two man-sized candles, lit from frre either reserved on Maundy Thursday'
or kindled on Good Friday and reserved until the following day. We learn from the
Pontifical of Regensburg that the two large candles were blessed prior to the reading of
the lections (p.l25). Their benediction at this point can also be inferred from Zachary's
letter and from Ordo 29? The church thus remained in semi-darkness throughout the
readings and the blessing of the font, until the cry of Accendite, following the Agnus Dei
after Baptism, instructed the neophytes to light their candles and the sacristans to kindle
the lamps of the church. The sudden appearance of light at the conclusion of the Vigil,
heralding the Resurrection and symbolising the triumph of the Light of the World over
sin and death, provided a vivid contrast with the sepulchral atmosphere3 which pervaded
the ceremonial that had just ended. The faithful were now able to participate in the
milieu of both physical and eschatological light in which the first Mass of Easter was
celebrated.
The incmporation of the Easter candle and its ritual into the Roman Vigil resulted
in a diversification of the ceremonial relating to the provision of light vis-a-vis the read-
ing of the Vigil-lections and the blessing of the font. Whilst in some rites the tradition .
persisted of reading the Vigil-lections in semi-darkness, in some churches the provision
of an abundance of light4 occurred either at the beginning or close to the beginning of
the Paschal vigil.
(i) The Vigil in semi-darkness
We have already observed that with the increasing anticipation of the Vigil in
some churches in the West the dependence on liturgical light for both reading and dra-
matic effect had largely disappeared.5 The churches, listed in Tables 55, 56, and 57,
which perpetuated the tradition of holding the Vigil in darkness, had the following
scheme for the provision of light :
'Letter of Zachary, PL 89.951B.

2
0R 29.45.
3
The large candles recalled the two angels of Luke 24:4 (p.328-9). The darkened church
obviously suggested the Tomb. The aspect of death is further reinforced by the remark
of St Ambrose that 'the font has the shape and appearance of a sort of tomb' (Sermons
on the Sacraments ill. I).
~at is, with the illumination of the whole church.
5
Nevertheless there are a number of cathedrals whose unlit interiors on a dull late-
March morning would have provided a fitting atmosphere of gloom.
310
(a) The new fire procession.
(b) The lighting of the Easter candle at ignis accendit.
(c) The general illumination of the church after the blessing of the
font, cued (in some churches) by the cry of Accendite.
The replacement or displacement of the two Vigil-candles by the Easter candle, as the
principal source of liturgical light, ensured the continued supply of sufficient functional
light necessary for the reading of the prophecies. However, the substitution was also a
highly significant development; for,. whereas the two Vigil-candles suggested sep-
u1chrallight and the mourning of the Church for the dead Christ, the light of the Easter
candle was both Paschal and resurrectional. For, given the prominent Exodus-typology
especially in the Romano-Gallican Preface, the ftre of the Candle symbolised histori-
cally the presence of God in the Burning Bush and in the Fiery Column, and within the
immediate eschatological context of the Paschal vigil anticipated the inrush of light at
the dawn of the New Age, which wou1d banish the darkness of this world's oppression
and sin. At the same time, while the tlfSt chapter of Genesis was being read, the burn-
ing of the Easter candle, 'consecrated in honour of your name', demonstrated that the .
Word of God was in existence even at the Creation. The Easter candle 'burning over
the pages of the Old Testament is a sign of the presence of Christ from the beginning'. 1
Although only six of the documents listed in Tables 55-57 speciftcally mention the
kindling of the altar lights, it is safe to assume that most of the other documents include
them in their mention of 'all the lights of the church'. Lanfranc's injunction includesthe
lights both ante and circa the altar, that is, both the functional and cu1tic lights of the
choir and sanctuary. Further support for the implied inclusion of the altar lights comes
from the alternative tradition, outlined in the next section, in which these liturgical
lights were kindled prior to the start of Mass.
'Harbert p.241. The symbolism is valid regardless of when the Easter candle is lit, pro-
viding it precedes the reading of the prophecies. The weakness of the recent form of
service, produced by the Joint Liturgical Group (Gray pp.76 ff.) and incmporated into
Lent, Holy Week, Easter produced by the Liturgical Commission of the Church of Eng-
land (pp.223 ff.), in which the Vigil-lections precede the ceremony oflight,is the nec-
cessity of providing 'essential' light for the reading of these lessons.
311
Lanfranc's Decrees· c.1070 PL 150.468A
Carcassonne 14C DAER 4.24 p.150 ,M 56
Rosslyn Missal c.1300 HBS 15 p.35
Aries 14C DAER 4.24 p.150, M 31
Mende 14C DAER 4.24 p.150, M
187
Bazas 1503 Missal fol.lxi
Narbonne 1528 DAER 4.24p.l50, M
203
Auxerre 1537 DAER 4.24 p.150, M 39
Tournai 1540 Missal fol.lxxili
Bayonne 1543 Missal p.49
Osma 1561 Missal fol.c
Besan~on 1766 Missal p.234
* Indicates that the altar lights were also lit.
Table 55 Evidence for the illumination of the church at the conclusion of

the Vigil following the triple cry of Accendite.

Ordo 27 750-800 OR 27.62
Ordo 28 c.800 OR 28.81
Poitiers c.900 Poitiers p.220
PRG c.950 1I p.llO
Regularis Concordia c.970 PL 137.4940
Regensburg c.980 Pontifical p.l30
Alcuin c.1000 PL l01.1221C
Cluny· llC PL 149.663C
Cluny 1510 Missal fol.lilii
Farfa llC Albers I p.56
Fruttuaria* llC Albers IV p.68
Fleury llC Albers V p.l46
Vallombrosa· llC Albers IV p.250
Gembloux llC Albers 1I p.l01
CMG llC Albers V p.38
St Vito, Verdun llC Albers V p.146
Besan~on llC DAER 4.24 p.150, M 56
Reims c.l200 Ordinary p.130
St-Martin, Tours l3C Ritual p.58
St Vedast, Arras c.1300 HBS 86p.160
Strasbourg 1364 DAER 4.24 p.l63, M 35
Westminster c.1370 HBS 5 col.589
Hereford 1502 Missal p.112
Camaldolese• 1503 Missal fol. 94
Vallombrosa 1503 Missal fol.xcv
Lyon 1510 Missal fol.ci

the Vigil following a single cry of Accendite.
312
Church Date Source
Salzburg 1507 Missal fol.ciii
Liege 1540 Missal fol.Ixxxv
Paris 1666 Missal p.260
Angers• 1731 Ceremonial p.267
Lisieux 1752 Missal p.203
St Bertrand 1773 Missal p.226
Perigueux 1782 Missal p.l71
Metz 1829 Missal p.175
La Rochelle 1835 Missal p.203
Autun 1845 Missal p.258

the Vigil without the cry of Accendite.
The cry of Accendite can be traced back to the eighth-century Stational Mass of
Easter Day at the Cathedral of StJohn Lateran, where the subdeacon issued the order to
kindle the lights of the basilica before the start of the service, once the ceremonies of the
Vigil had ended. 1 It was heard at Lyon, Regensburg, and in other churches where a
Roman-type Vigil was held; 2 and survived in numerous French dioceses long after it
had disappeared from the Roman rite. 3
The earliest sources attest a single cry of Accendite. There is no evidence of any
link between the twice-repeated order enjoined by Lanfranc and the triple acclamation
of Lumen Christi when the new flre was brought into church. In the two instances4
where both the Lumen Christi and the Accendite featured within the same rite, the latter
was proclaimed only once. In places the threefold Accendite was announced by the
choir alta voce.5 At Bayonne the deacon held the serpent-candle; whilst at Osma the
deacon ascended an altar step with each subsequent cry. In both these two churches the
choir made the response of Deo gratias after each shout. At Aries and Narbonne the
response of Lumen Christi followed the first two acclamations and Deo gratias the
1
0rdo I, PL 78.9400.
2
0ne of the features of the old Roman Vigil which survived at both Lyon, where the
Roman rite was introduced by Leidrad c.AD 800, and Beneventurn, was the position in
the liturgy of the Vigil-lections. At Lyon the prophecies were read by natural light
because of anticipation and without the use of liturgical light before the Easter candle
was blessed. In the latter rite the kindling of the new fire and the blessing of the Easter
candle took place between the eleventh and fmal readings (Hesbert p.188).
3
lt is not found in PR XII or in subsequent Roman service-books.
"The Customary ofVallombrosa and the Camaldolese Missal. For these and subsequent
references, see the Tables.
5
ln Lanfrimc 's Decrees and in the Missal of Bazas.
313
third. In Auxerre Cathedral a white-robed choir boy raised his voice with each subse-
quent cry of Accendite. In the rites of the churches listed in Table 57 there is no indica-
tion that the cry featured in the ceremonial.
(ii) The Vigil in the light
1. illumination before the blessing of the Easter candle~
(a) The Mozarabic tradition. It is generally agreed that a number of elements in

the Mozarabic new fire ceremony had their origin in the fourth-century liturgy of
Jerusalem; 1 Certain features, such as the striking of the new fire and the blessing of the
Easter candle together with its Preface, are importations from Gaul. However, there
seems to be little doubt that the following have their provenance in the rite of
Jerusalem:
- the sacristy in total darkness

- the lighting of the lamp by the bishop
- the lighting of the clergy candles
- the entry of the bishop into church
- the sharing and spread of light from candle to candle
The initial ceremony completed, the Easter candle is consecrated and the Vigil-lections
are read in a blaze of light from the candles of the faithful. This Jerusalem-derived
Vigil, in which the congregation participates throughout in the newly-blessed light con-
trasts markedly with the type of Vigil held in semi-darkness, which was described
above in Section (i).
(b) Other traditions. The custom which obtained in the churches of Cordoba,
Bourges, and Carcassonne resembled Mozarabic practice in that the Vigil-lections were
read with the church lights ablaze. With these three churches should be included Uzes
and Passau, where the church lights were kindled at the conclusion of the Preface (Table
58). It is possible that at Cordoba the influence of the Mozarabic rite may be detected,
although somewhat strangely the Easter candle was lit only during the Preface at ignis
accendit. The antecedents of this feature at Bourges and Carcassonne are more difficult
to explain. Spanish influence seems very unlikely. One possible explanation is that we
1
See Appendix 13.
314
have in these two churches a survival and development of the ceremonial of PRG. For
according to that pontifical both the seven lamps in front of the altar and the Easter
candle were kindled before the archdeacon began the Exultet. 1
(i) Church illumination before the Exultet

Mozarabic rite 7C Leon Antiphonary pp.280-1
Cordoba 1561 Missal fol.ciiii
Bourges 1741 Missal p.226
Carcassonne 1749 Missal p.196
(ii) Church illumination after the Exultet

Uzes 1495 alllights Missal fol.lxiii
Passau 1503 other candles Missal fol.lxxxiv
Table 58 Evidence for the illumination of the church either before or after the belssing
of the Easter candle.
2. illumination during the blessing of the Easter candle.
From about the middle of the eleventh century there emerged the practice of kind-
ling other lights during the Preface in addition to the Easter candle. It is first attested in
the Acta Vetera of Rouen2 : the two small Vigil-candles were lit during the Preface at
the words divisus in partes. The lighting of these candles at this point is also found in
the 1511 Missal ofNimes and in an Ordinary of the Regular Canons of St Rufinus,3 and
in the Dominican rite. 4 The practice suggests a liturgical representation and visual inter-
pretation of the subsequent clause mutuati (tamen) luminis detrimenta non novit.5 It is
not difficult to realise why this practice was first extended to include the seven altar
lamps,6 and subsequently all the lights of the church, as at Salisbury. 7
Later development, centring around the point at which the lights, both functional
and liturgical, were kindled, resulted in a diversity of practice. One suspects that alter-
native points were adopted because the cue of divisus in partes was separated by only
three words from ignis accendit, the point at which the Easter candle was lit. Three
separate traditions developed involving this point.
1
11 p.98 §346. The practice is also attested in Alcuin (PL 101.1216B).
2
PL 147.176C.
3
Both documents are cited by De Vert, Explication II p.37.
4
1504 Missal fol.lxxxix and 1908 Missal p.l69. According to Dominican practice the
two acolytes' candles were lit at divisus in partes, and a further dissemination of light
~for general illumination) occurred at apis mater eduxit.
'Suffers no detriment from its light being borrowed.'
6
0EL p.61 and PR XII I.xxxii.8 p.240.
7
For references, see Table 59.
315
(i) At suscipe, sancte pater. In most other rites the five grains of incense were
inserted at this point. That the following phrase incensi huius sacrificium vespertinum
originally related to the light of the Candle, and not to the incense, strongly suggests
that the tradition of lighting other candles at this point was very old. Regrettably, the
only known instance of its occurrence is at Tongres in the fifteenth century. Here the
other candles mentioned presumably included those of the acolytes, and the church
lights.
(ii) In a number of churches the kindling of the other lights also occurred at the
same point (ignis accendit) at which the Easter candle was lit. The kindling of these
lights may have occurred here because it was felt that, once the new fire had been used
to kindle the Easter candle, there was no obvious reason to delay further acts of illumi-
nation; or because it demonstrated visibly the truth of the assertion, soon to be heard by
all : luminis detrimenta non novit; or possibly because it avoided a further interruption
in the chanting of the Preface.
(iii) A pause for the secondary act of illumination at apis mater eduxit was
favoured by an even larger number of churches, including Rome, and by the majority of
the monastic orders. These words, which closed the short pericope eulogising the bee,
provided a suitable break in the Preface in the same way that the insertion of the grains
and the lighting of the Easter candle also occurred at the end of a section. In churches
which retained the Preface containing the lengthy eulogy of the bees, the interruption
for this secondary act of illumination occurred at virgo permansit. Other points during
the Preface where secondary illumination is known to have occurred are listed
in Table 59.
In a large number of service-books the rubrics are silent regarding the point at
which additional illumination occurred. In the rubrics of a number of church rites it is
not clear which lights were kindled during the Preface. The word lampades without a
qualifying phrase or adjective may signify either the altar lights or the church lights or
both. For instance, the rubric at mater eduxit in the 1762 Missal of Paris relates to the
acolytes' candles and the lampades (p.239). The Missal of 1666, however, specifies
'the lamps hanging in the choir' (p.244).
That development and flexibility of practice could and did occur within the same
rite is perhaps best exemplified in the various Roman documents which attest the vari-
ous points at which the church , the acolytes', and the altar lights were kindled in the
period AD 950 to 1574. As we observed at the beginning of this chapter, in the
primitive Holy Saturday rites of the Roman Church the two large Vigil-candles had
316
(i) At suscipe, sancte pater
Tongres 15 c 'other candles' Ordinary p.l65
(ii) At ignis accendit
StMary's York c. 1400 other lights HBS 75p.292
Carmelites c. 1312 acolyte and church Ordinary p.31
Exeter 1337 other church lights HBS 37 p.322
Langres 1492 other church lights Missal fol.lxv
Seville 1507 other lights Missal fol.lxxviii
Wiirzburg 1509 acolyte only Missal fol. 244
Cologne 1514 other church lights Missal fol.lxxv
Coutances 1557 other church lights Missal fol.lxx
Palencia 1568 church and altar Missal fol.cvi
Chalons 1748 acolyte and lamps* Missal p.183
Lyon 1771 acolyte and church Missal p.195
Bayeux 1790 other candles Missal p.170
Meaux 1845 acolyte and lamps• Missal p.169
(iii) At mater eduxit
Haymo(OM)t c. 1243 lamps* Van Dijk II p.246
Rome 1477 lamps* Miss. Rom. np
Wiirzburg 1497 other lights Ordinary np
Coimbra (AC) 1597 lamps· Ordinary fol.137
Camaldolese 1634 lamp of high altar and Ceremonial p.85
church lights
Rouen 1640 church lights Ritual p.306
Carmelites 1664 church lamps Missal p.162
Paris 1666 acolytes and hang- Missal p.244
lamps in choir
Cistercians 1669 all the lamps* Missal p.159
Besan~on 1682 all lights Ceremonial p.333
Evreux 1740 lights Missal p.192
Amiens 1752 lamps* Missal p.79 (Supp1)
Maison du Roy 1741 lights Sem.Sainte p.401
Mende 1766 other lights Missal p.205
Poitiers 1767 all lights Missal p.251
Capuchins* 1775 nearby lamps Ceremonial p.129
Bayeux 1790 altar lights Missal p.170
Poland 1819 (As Camaldolese) Manual p.479
Coutances 1825 acolyte and church Ceremonial p.333
Lu~on 1828 acolyte and lamps* Missal p.220
Limoges 1830 church lights Missal p.255
Toulouse 1832 candles and lights Missal p.211
(iv) At divisus in partes
Salisbury c. 1300 church lights Leggp.l18
Salisbury c. 1486 church lights Missal fol.lxxxiv
(v) At virgo permansit
Burgos 1546 church lights Missal fol.cvii
Cosenza 1549 all lamps Missal fol.118
Braga 1558 other lights Missal fol.xcvi
(vi) At non novit
Braga 1558 2 other candles Missal fol.xcvi
* lampades.
t Actually at 0 vere beata nox.
t Other lights at 0 vere beata nox.
Table 59 Evidence for illumination during the Preface
317
been lit before the reading of the lections, and the church lights, together with the
candles of the neophytes, were kindled after the cry of Accendite. Subsequent changes
came about as follows.
(a) The Church Lights. PRG is silent on this score. The twelfth-century Ordo of
the Lateran Church, however, states that all the lights of the church should be lit by the
sacristan at the Kyries which introduce the Mass of Easter.' This procedure had also
been enjoined in Lan:franc 's Decrees : after the cry of Accendite all the church lights
should be kindled, including those in front of and around the altar. 2 Now according to
Haymo's Ordo Missalis of c.l243 'lamps' were lit during the Preface at mater eduxit. 3
In view of the later reference in the same document to altar lights (p.248), the lampades
referred to during the Preface are almost certainly the lamps which hung in the choir.
Late fifteenth-century and early sixteenth-century Roman rnissals4 retain the rubric :
(accendunt) lampades ante altare; 5 but the majority of later books, including the Tri-
dentine Missal of 1574 omit the phrase ante a/tare, and state simply: Hie accenduntur
lampades. The vaguenes of the rubric, which may relate to the church lights, the choir
lights,6 or even the altar lights, thus allows some flexibility of practice. 7 In more recent
times all the lights and lamps in the church, except the altar candles, were kindled at
mater eduxit. The latter were lit after the Litany. 8
1
0EL p.73: omnia luminaria et lampades ecclesiae.
2
PL 150.468A :ante et circa a/tare. The evidence of later Roman missals strongly
suggests that the phrase (in the missals) lampades ante a/tare, which formerly referred
to the liturgical lights that subsequently became altar candles, indicates the functional
lights of the choir. These, viewed from the nave, did indeed hang in front of the altar.
Those disposed circa a/tare were the cultic lights. The evidence, such as it is, would
suggest that in Lan:franc's time lights were not placed upon the altar, rather around it.
3
Van Dijk II p.248.
4
For instance, MR 1500 MR 1501; MR 1506; MR 1520.
5
The corresponding rnid-prefatorial rubric in the Roman Missals of 1474, 1477, 1484,
and 1491 (HBS 33 p.85, np, np, fol.93, respectively) seem to present a difficulty~ for
they state : Hie accenditur lampas ante a/tare, 'Here a lamp is lit before the altar'. The
use of the singular lampas may be explained in three ways. (Its appearance in success-
ive missals seems to rule out an error.)
(i) It is used generically and is here to be interpreted 'light'. (ii) It refers to the
ftrst of the chancel lamps to be lit. (iii) The sanctuary lamp is intended.
The interpretation of lampasas 'light' is poetic and would be very unusual within
a rubrical context. The second explanation implies that one of the chancel lights was in
some way special. Support for the third possibility is to be found in the 1634 Camaldo-
lese Ceremonial, in which we read that, at the words mater eduxit, the lamp of the high
altar was lit by an acolyte p.85). The church lights were also lit at this point.
Somewhat surprisingly the rubrics of MR 1477, MR 1484, and MR 1491, which
form the preamble to the Exultet, state that at the words mater eduxit: Hie accendunt
lampades.
6
As in the 1509 Missal ofWiirzburg fol.244.
7
This is further illustrated by the fact that the Caeremoniale Episcoporum of 1600
enjoins that the church lights should be lit during the Litany (p.303) and the Ceremonial
of Benedict XIII states that this should occur just before the start of Mass.
8
318
(b) The Altar Lights. Reference has been made in the above section to the practice
of kindling the altar lights prior to the start of Mass. The Lateran Missal and Durand's
Pontifical also provide evidence for the practice in the thirteenth century. 1 Earlier prac-
tice had been to light the seven lamps before the Exultet. 2 Subsequently they were lit
during the Preface. According to PR XII and the Lateran Ordo3 the kindling of these
lights took place at luminis detrimenta non novit. This tradition survived at Palencia. 4
(c) The two candles. Custom varied regarding the moment at which the two
candles, which had survived from the old Roman-type Vigil,5 were lit. The displace-
ment of the two man-sized candles from the centrality of the liturgy either resulted in
their assuming a very minor role in the ceremonial or in the merging of their function
with the honorific episcopal candles in some churches or the acolytes' candles in others.
Largely, one suspects, because they had acquired different functions within the liturgy,
it is perhaps not surprising to fmd a variation in the points at which they were kindled
during the ceremonial. At Chartres the two bishop's candles were lit probably during
the Preface; 6 whilst at St-Germain-des-Pres the two Vigil-candles, their status reduced,
were lit immediately after the conclusion of the Preface. 7 This is the point at which
they were lit at Salzburg, where even in the fifteenth century the two candles were of
considerable dimensions. 8 At Vallombrosa, however, the two candles were lit after the
cry of Accendite .9 The lighting of the two acolytes' or torch-bearers' candles is unre-
corded in many documents. At Perigueux and Meaux, for instance, they were lit before
the Exultet, as was the Easter candle10 • Elsewhere, the kindling of the church lights
presupposes that these two candles were already ablaze. The Sarum ~isal of c .1486
states that the illumination of the church was the responsibility of the torch-bearers :
ceroferarii accendunt candelas per ecclesiam. 11
1
Schmidt IT p.61 0 § 112 and PGD ill p.588, respectively. Bauldry in the eighteenth cen-
!UIY also prescribes this point (p.193).
111.us PRG ll p.98 §346.
3
I.xxxii.8 p.240 and OEL p.61, respectively.
4
1568 Missal fol.cvi.
5
For these candles, see Chapter 16 pp.326 ff.
6
0rdinary p.111.
7
Martene, DAER 4.24 p.l59, M 230.
8
1507 Missal of Salzburg fol.xcvii.
9
Albers IV p.250.
10
1782 Missal p.159 and 1845 Missal p.l69, respectively.
11
Fol.lxxxvi. 'The torch-bearers kindle the lamps throughout the church.'
319
smaller Easter candle, was lit is commed to the Regularis Concordia. According to this
document the lighting of this candle took place at the conclusion of the Preface. 1 At
Salisbury it seems to have been lit as soon as the larger Candle was aflame. 2
(e) The Sanctissimum light. Few service-books or manuals refer to the light before
the reserved sacrament. The rubric of the fourteenth-century Customary of Laon3
seems to imply that the lights before the Corpus Christi were relit immediately after
they had been extinguished. This momentary loss of light is also attested by De Bralion
and Gavantus. 4 The nineteenth-century Polish Manual states that the sanctuary lamp
was lit during the Preface at mater eduxit, but does not indicate at what point it was
extinguished (p.479). The 1775 Capuchin Memoriale concurs with the foregoing
manual, but adds that the lamp was extinguished before None (p.l30). On the other
hand Desideri5 and more recent liturgical handbooks6 insist that this light ought never
to be extinguished.
3. illumination after the blessing of the Easter candle.
At Passau and Uzes7 the church lights were lit at the conclusion of the Preface.
We noted above that according to the Regularis Concordia the cereus minor and the
two Vigil-candles were lit at this point during the service; but the fact that that
document enjoins the kindling of the church lights after the cry of Accendite makes the
influence of Benedictine practice unlikely.
(iii) The rites of Lyon and Milan
The ceremonial of the primatial church of France has undergone a number of

changes in respect of the provision of light at the Paschal vigil. From c .AD 800, when
the Roman rite was first introduced at Lyon, until 1771, the cry of Accendite at the
conclusion of the Vigil signalled that the lights of the cathedral should be lit. fu 1771
the contemporary Roman practice was adopted: the acolytes' candles and the church
lights were kindled during the Preface, and the altar candles after the blessing of the font. 8
1
PL 137.494C.
2
Implied from Dickinson p.341.
3
Martene, DAMR 3.15.7 p.142, M 1164.
4
Caeremoniale p.255 and Merati-Gavantus p.166.
5
Praxis p.l44.
6
For instance, Fortescue and O'Connell (11th ed.) p.301.
7
1503 Missal fol.lxxxiv and 1495 Missal fol.lxiii, respectively.
8
1771 Missal p.195.
320
The traditional cry of Accendite, which had been excluded at that time, was restored to
the liturgy in 1904, inunediately before the Agnus Dei. 1 Present practice is modelled on
the Roman reforms of 1955 and 1970.
The earliest evidence for the Ambrosian rite dates from the twelfth century. The
Easter candle and the diaconal candles were lit at the words Ecce iam ignis which intro-
duce the second pericope of the A_mbrosian Preface? According to the Missal of 1594
the lampades were kindled at the closing words of the third pericope ut coruscus
adveniet (fol.97). To that rubric the Missals of 1669 and 1901 add 'and the lights of the
church'. 3 Contemporary practice is to kindle all the church lights and candles as the
procession with the Easter candle moves into church. 4
1
1904 Missal p.221.
2
Beroldus p.llO and all missals up to that of 1901.
3
Martene, DAER 4.24 p.169 and 1934 Missal t.e. p.40 (Repert.),respectively.
4
1986 Missal p.54.
321
Chapter Sixteen
THE ROMAN VIGIL AND THE GALLICAN RITE
(i) Outline description and the Vigil-candles
The evidence for the older Roman Vigil is contained in five of the ordines Rmr.ani,
which range in date from c.AD 600 to c.AD 800. 1 None of these five docwnents
describes the service in great detail - the information supplied by Ordo 11 , for instance,
is extremely meagre; nevertheless collectively they provide sufficient details to enable
us to reconstruct with confidence the structure of the ceremony which took place during
this period on the night of Holy Saturday in the Cathedral of StJohn Lateran.
After the clergy had donned their vestments in the sacristy, 3 they moved thence in
silent procession4 into church escorting the two man-sized Vigil-candles,5 and making
their way to the altar to take up their appointed positions. The two candles were borne
in procession by two junior officials or clerics in minor orders,6 and held by their
bearers who stood one either side of the altar. 7 From the evidence of Ordo 23 it is not
clear whether the two candles were lit in the sacristy. However, Ordo 24leaves us in
little doubt that the kindling took place in the main body of the church and in alllikeli-
1
0R 11, OR 16, OR 23, OR 24, and OR 30B. All five docuents attest papal practice.
Ordo 11 contains what appears to be an anachronistic rubric relating to the blessing of
the Easter candle: postea impletur <cerei> benedictio (Andrieu's parenthesis), 'after-
wards the blessing (of the candle) is completed'. It is tempting, in view of the uncer-
tainty over cerei, to emend the text by reading cereorum for cerei. This would accord
well with the mention of the two candles in the next section and possibly account for the
substitution of the singular form of the noun by a scribe who, perhaps familiar with the
blessing of a single candle, had superscribed cerei above cereorum by way of a query.
2
From the evidence of OR 23.24 and OR 30B.37 it would appear that only deacons (in-
cluding the archdeacon and subdeacons) were present at this ceremony. OR 30B.30
adds that the archdeacon presided. According to the same authority (ibidem) the Pope
only made his appearance at the Paschal ceremonies after the Agnus Dei.
3
0R 16.36.
4
0R 16.36.
5
1t is true that only OR 11.90 mentions the size of the candles (staturam hominis
habentes); but this phrase describing the two candles at the Paschal vigil is also found
in PRG (IT p.99 §348), Poitiers (p.215), Alcuin (PL 101.1216C), and the 1507 Missal of
Salzburg (fol.xcvii). It can be reasonably inferred that these lights were present at the
Vigil described by the other ordines, even though they are not specifically mentioned.
6
0R 23.24 mentions regionarii, OR 24.41 notarii, and OR 30B.37 subdeacons. The dif-
ference in title or status is here not significant. Both regionarii and notarii may well
have been subdeacons during this period; and under the canons of the Roman Church,
it may have been possible to hold the offices of notarius and regionarius concurrently.
Alternatively, the discrepancy may be attributed to the different era which each ordo
attests. Moreover, the mention of subdeacons in Ordo 30B may well be a Gallican sub-
stitution for a church where papal notarii and regionarii were unknown.
7
0R 30B.37 adds that they stood to the rear of the altar. ·
322
hood in the sanctuary. 1 Elsewhere,2 we have seen that the ftre used to light the candles
had been reserved for that purpose either on Maundy Thursday or in the ninth century
after its kindling on Good Friday.
We shall refer later in this chapter to the size and to the signiftcance of the size of
the two baptismal candles or Vigil-candles. Regarding their physical appearance there
is no good reason why the descriptive phrase staturam hominis habentes, 'man-sized',
should not be interpreted literally. The argument that a ftve or six foot high candle of
solid wax could not have been carried in procession because of its weight cannot be sus-
tained. A candle, the size of a human being, can be manufactured without an excess-
ively wide diameter,3 making quite feasible its transportation by one person over the
short4 distance covered by the procession during the course of the Roman Vigil.
Moreover, the very use of the phrase staturam hominis habentes to describe these
candles indicates that they were unusually large; 5 and there is other evidence for the
existence and use in church of very large candles. At Assisi there is a thirteenth-century
fresco depicting a server at Mass holding a ftve or six foot high candle; and in a docu-
ment from Haughrnond Abbey, dated 1341, there is mention of twelve candles, each
weighing 6 lbs, which burned around the tomb of Richard, Earl of Arundel.6 However,
candles half as tall as those indicated by this phrase and set in candle-holders standing
two or three feet from the ground, would give the impression from a distance of attain-
ing the height of a man. The use of the verb tenere in two of the ordines7 to inform us
that these candles were being held during the course of the Vigil is inconclusive since
these large candles would rest on the ground whilst those who held them were station-
ary, regardless of whether they were mounted in candlesticks or not. It is true that the
Romano-Germanic Pontiftcal states that the man-sized candles stood in candleholders; 8
but in view of the centrality of the Easter candle in the rite described by this document,
it was perhaps inevitable that the size of these two candles should be reduced as their
1
Sabbato sancto veniunt ornnes in ecclesiam et tunc illuminantur duo cerei, tenentibus
duobus notariis, unus in dextro cornu altaris et alter in sinistro. OR 24.41. The evidence
is also found in the Sacramentary of St Eligius §88. Ecclesiam is almost certainly
referring to the main body of the church, although the inclusion of the sacristy in the
term should not be ruled out.
2
See Part IT Chapter 1 pp.99-100.
3
Modern candles measuring 24" and 36" can be manufactured with diameters of as little
as 0.75" and 1.25" respectively. A tallow candle measuring 5' x 2" would weigh
approximately 6.44 lbs. A beeswax candle would be slightly heavier.
4
Even allowing for the fact that the baptistery of St John Lateran was a building
detached, as now, from the cathedral itself. ·
~phrase is not found outside the context of the Paschal vigil.
6
Document 1245, p.227 in U.Rees, !h2_ Cartulary.
7
0R 24.41 : cerei tenentibus duobus notariis; OR 308.37: tenentesfaculas.
8
PRG IT p.99 §348.
323
role and status in the Vigil diminished. The use of holders to increase the overall height
of the two Vigil-candles would continue to justify their description of 'man-sized' and
furnish a reminder of their former size.
(ii) The Lucemarium at Jerusalem
Egeria' s mention 1 of the importance of the Lucernarium in the cycle of daily

offices at Jerusalem should not be taken as evidence that her readers would be unfamil-
iar with the service or that the office was unknown in Egeria's native Galicia or Aqui-
taine.2 For the service which Egeria witnessed in Jerusalem had undergone a unique
development under the direction ofthat city's innovative and dynamic bishop, Cyril,
who utilised its topography and historic sites for liturgical experimentation and change.
The Jerusalem Lucernarium must have contrasted strongly with the lamp-lighting ser-
vice familiar to Egeria in her own native land. 3 That a close relationship existed in the
Jerusalem rite between the Lucernarium of Holy Saturday and the Paschal vigil is clear
from the Peregrinatio; 4 but the entry of the bishop into the cave5 of the Anastasis, the
lighting of the candle he held from the lamp that burned perpetually at the Tomb, and
the sharing of the frre with the faithful were features of a Lucernarium which were
peculiar to Jerusalem and which together formed a preliminary ceremony in themselves,
but were not an integral part of that office. This introductory service provided an
opportunity for each of the participants to have a share in the blaze of light in which the
readings from Scripture were proclaimed.
(iii) The development of the Roman Vigil from the Lucemariurn
When we examine the structure of the Roman Paschal vigil of the seventh and
eighth centuries, we find that, whilst there are features inherited from the local Roman
tradition, some elements which are common to both the Roman and Jerusalem cere-
monies strongly suggest the influence of the latter church's liturgy on the former. The
most striking similarity is the sepulchral nature of the place in which the fire is reserved
for the lighting of the candles at the Vigil. At Jerusalem a flame burned continuously at
1
Wilkinson pp.66-69 and 123.
2
For Egeria's country of origin, see Wilkinson p.3.
3
Her statement : Vigiliae autem paschales sic fiunt quemadmodum ad nos, 'They keep
their Paschal vigil like us' (Duchesne, Christian Worship p.512), relates only to the
Vigil itself and not to any ritual that preceded it. There is no evidence that the Lucerna-
rium in the western Church was other than a simple daily ceremony involving the light-
ing and blessing of a lamp.
4
Wilkinson pp.66-69.
5
Egeria uses spelunca to denote the Sepulchre (Duchesne, ibidem p.493 ).
324
the Tomb, symbolising the undying and unquenchable Spirit of Jesus. 1 At Rome the
unique topographical conditions of Jerusalem could obviously not be reproduced;
nevertheless, it is our contention that the pattern of services at Rome during the Tri-
duurn reflected the topographical and liturgical conditions that obtained in Jerusalem.
For the services at Rome during the latter part of Holy Week were not confmed to one
ecclesiastical building, as they were in most other churches. The main services of Good
Friday including the night office of Matins/Lauds were held in the Church of S.Croce in
Gerusalemme, with a result that no lights burned in StJohn Lateran from the conclusion
of the Pedilavium on the evening of Maundy Thursday until the start of the Paschal vigil
during the evening of Holy Saturday.2 The absence of illumination in the cathedral for
the duration of nearly two days perhaps inevitably suggested the gloom and silence of a
tomb, especially at this time of the liturgical year. Moreover, we have already 3
observed that the fire for the lighting of the two baptismal candles (or Vigil-candles)
was reserved in secretiore loco; and in this remote chamber, where the fire was so care-
fully tended, it is possible to see a parallel derived from the cave or chapel in the Anas-
tasis in Jerusalem. 4 Further evidence in support of the view that during Good Friday
and Holy Saturday the Cathedral of StJohn Lateran liturgically represented the Holy
Sepulchre in Jerusalem will be presented later in this chapter.
As in the Jerusalem rite the liturgy of Rome on the evening of Holy Saturday
began with a ceremony of light , which in the latter city appears to be a survival of the
primitive Lucernarium,5 the short description of which we fmd in Hippolytus. 6 Accord-
ing to this account, after the onset of evening a deacon brings in a lamp, 'and standing
in the midst of all the faithful who are present', he exchanges an initial greeting with
them and utters the prayer of thanksgiving for the light. 7 Allowing for liturgical devel-
opments over a period of about four centuries, the preliminary ceremony to the Roman
Vigil contains elements in common with and recognisably derived from the
Lucernarium. (i) The ceremony takes place late in the day. The start of the service at
1
For the symbolism of lights at tombs, see Appendix 6. ·
2
During this period there was no illumination at the night office of Holy Saturday in St
John Lateran. See Part I Chapter 2 pp.9-10.
3
Part IT Chapter 7 p.172.
4
Whether the number of lamps, i.e. three, can be attributed to the influence of Jerusalem
is open to question. For, whilst Egeria in the fourth century and Antonius of Placentia
in the sixth mention that only one lamp burned at the Sepulchre (Duchesne, Christian
Worship p.493 and Geyer p.171, respectively), Arculf, writing c.AD 700, refers to
twelve (Wright p.2).
5
As a daily office this service had disappeared from the Roman liturgy as a result of the
replacement of the old cathedral tradition by the monastic one. Bradshaw p.l23.
6
Cuming's text, p.23 §25.
7
The writer disagrees with Curning who has argued that the prayer was said by the
bishop. Both the sense and certainly the grammar demand that the understood subject
of 'shall give thanks' should be the deacon.
325
the ninth hour 1 or at the eighth ho~ seems to confirm the generally-held view that by
the eighth century the Vigil in the West had come to be anticipated. 3 (ii) Light is car-
ried into church from without. The increase in the number of lights from one in the
Apostolic Tradition to the two of the Roman ordines will be discussed presently. The
difference does not alter the fact that light is carried. (iii) The prominence of the dea-
cons. From Ordo 23 and Ordo 30B 4 it would appear that only deacons and subdeacons
were present at this ceremony. We suggested at the beginning of this chapter that the
regionarii of Ordo 23 and the notarii of Ordo 24 were members of the diaconate.
Ordo 30B .30 adds that the archdeacon presided. Even allowing for development within
the liturgy the association of the diaconate with the lighting of the evening lamp seems
to have been perpetuated throughout the centuries.
The similarity between the two services referred to above leads us to draw a
number of conclusions. (a) The procession and the blessing of the candles prior to the
start of the Roman Vigil was a development of the primitive Lucernarium in which the
evening lamp was carried and blessed by a deacon. At Jerusalem the unique conditions
had resulted in the bishop's appropriation of the deacon's function. (b) The anticipated
hour for the Vigil was a Roman development. In Jerusalem the Vigil continued to take
place in the evening. (c) The bringing in of the light, which had been in the primitive
Roman Lucernarium both a functional and a symbolic act, had become primarily a sym-
bolic ritual.
In the Paschal vigil according to the Ambrosian and Mozarabic rites the prelimi-
nary ceremony also preserved the central feature of the primitive Lucernarium viz. the
bringing in of light.5 The Roman ceremony differed from two of these rites6 in one
significant respect : in the ceremonies of the latter one light was carried in procession;
whereas at Rome, and possibly at Milan, there were two. 7 The divergence of Roman
practice (and possibly the Milanese) in this respect is not easy to account for in view of
the likely influence of Jerusalem on all three western rites. One might have expected
the use of only one lamp or candle at Rome, as at Milan and in Spain. The Roman
exception may be explained in a number of ways.
1
0R 16.38.
2
0R 30B.37.
3
For the anticipation of services during the Triduum, see Appendix 10.
4
0R 23.24 and OR 30B.37, respectively.
5
The use of a lamp survived in both rites. See Appendix 13.
~at is, the Mozarabic and the Jerusalem rites.
7
1t is very significant that at Vespers in the (former) Ambrosian rite two cantari or
candlesticks were placed beneath the table next to the altar, and removed at the start of
the evening hymn. (Borella p.251.)
326
(i) Since the two man-sized candles were carried in procession to the baptistery
later the same evening for the blessing of the font, and subsequently brought back to
their former position behind the altar in readiness for the start of the Easter Mass, 1 it
could be argued that the number of lights was deliberately increased from one to two so
as to provide the escort of two lights2 for the Pope, who officiated both at the blessing of
the font and at the Mass which followed. 1his theory presupposes a change from the
use of one baptismalla...rnp at 3.n earlier period to the use of two candles by the seventh
century?
(ii) The evidence ofTertullian and the Apostolic Tradition 5 shows that in some
4
churches in the West more than one lamp featured at the Lucernarium. Two lamps,
therefore, may originally have been used at Rome; and with the fusion of the
Lucernarium and the Vigil proper into a single ceremony, they may have been replaced
by the two candles which escorted the Pope.
(iii) A third suggestion would also explain why they were so large - neither of the
previously-mentioned theories can account for their size. The phrase describing the two
candles, staturam hominis habentes, is significant. The language of liturgical rubrics
and directives, unlike the language of prayer, has never been characterised by florid or
poetic turns of phrase. We should perhaps have expected a term such as magni cerei to
indicate in a rubric candles of great size. For this reason the writer believes that not
only should the phrase in question be understood in a literal sense to indicate that the
candles were as tall as a human being, but that the candles were intended to represent
human beings, or, as we shall see presently, beings in the form of men; and that for that
reason their height was increased deliberately in order to achieve that intention.
Support for the theory comes from the ninth-century Pontifical of Poitiers :
After this (the blessing of the Easter candle) two man-sized candles are immedi-
ately lit from the Easter candle and held by two no~arii on either side of the altar.
1
0R 30B.44 and .61.
2
According to the solecistic Ordo 4, the Pope enjoyed two honorific lights at every
liturgical occasion : Deinde oblationarius inluminet duos cereos ante secretario pro
luminaria pontificis, quod est consuetudo omni tempori (§7). See also Partl Chapter 2
pp.9-10.
A baptismal lamp was formerly used at Milan ( 1768 Missal p.125).
4
Apology 39.
5
Cuming, Hippolytus p.23 §25. This evidence, which may well relate to the Roman
Church, is admittedly ambiguous. The plural/amps appears in the title of the chapter;
but the direction in the text states that the deacon should bring in a lamp.
327
The candle on the right has been marked by the notarius who holds it 'The angel
sitting at the head'. The left one is marked 'The angel sitting at the feet'. The
candles symbolise the two angels in the sepulchre ... (Luke 24:4). 1
If the use of the two man-sized candles and the symbolism associated with those
candles are derived from Roman practice, then the two large candles lit at the Roman
Vigil also represent the men 'in dazzling apparel'. To this claim there are two immedi-
ate objections; (i) Poitiers, a Gallican pontifical of c.AD 900 should not be used as evi-
dence for the practice of the Roman Church nearly three hundred years earlier. (ii) The
symbolism was suggested by the size of the candles and became attached subsequently.
Although both objections must be examined carefully, neither is fatal to our

theory. Two observations should be made regarding the first. It is true that a period of
about 300 years separates the earliest mention of the two candles in Ordo 11 and the
description of them in Poitiers. However, the evidence of Ordo 30B, which is to be
dated to c.AD 800, shows that the two candles still figure prominently at the Vigil; and
although Poitiers itself was compiled c.AD 900, the description of much of the ceremo-
nial would have been valid for an earlier period. This pontifical is mainly Gallican in
respect of the new fire ceremony and the use of the Easter candle. Two features,
however, viz. the reservation of fire from Good Friday and the use of the two man-sized
candles derive from Roman practice. There is, therefore, a strong possibility that this
symbolic interpretation was not only known at Rome, but that it had originated from a
Roman milieu; although it must be admitted that there is no documentary evidence
from the Roman ordines to support this contention. 2
The case for the Roman origin of the symbolism of the two angels gains support
from a closer look at the circumstances surrounding the emergence of the symbolism.
As a rule a symbol is attached subsequently to the object it represents; and there are
many instances where the Church has found in the minutiae of Christian ceremonial
allusions and prefigurations in the Scriptures. The representation of the two angels by
the two man-sized candles seems to be a typical example; but the circumstances of the
liturgy in which the candles were used also invites the possibility that it was not the size
of the candles which suggested the symbolism. Rather, it was the mention of the two
1
Qua expleta, statim illuminantur duo cerei staturam hominis habentes de cereo bene-
dicto, et tenentur altrinsecus a duobus notariis in dextra parte et leva. Et dexter quidem
habet sibi impressum a notario Angelus ad caput sedens. Sinister vero a suo Angelus ad
pedes sedens. Quique in typo duorum angelorum in sepulchro domini ... (p.215).
1bis is perhaps not surprising in view of the laconic style of the five ordines cited at
the beginning of this chapter, and the brevity of their rubrics, compared with, for
example, PRG and Poitiers.
328
angels in Luke 24:4 which resulted in their representation by two candles staturam
hominis habentes and presence in the Roman ceremonies of the Triduum. Then again,
it is possible that this symbolism, already attached to two smaller lights, brought about
an increase in their height in order to achieve a more accurate and meaningful relation-
ship between object and symbol.
On the other hand, if it was the size of the candles which suggested the symbolism
in the first instance, it is necessary for us to ascertain not only why the candles were so
large, but why they numbered two. With regard to their number we have already
referred to the Pope's two honorific lights, and suggested the possibility of two cande-
lae lucernales. 1 However, it is difficult to believe that either of these two sets of lights
would have attained the height of a human being, had it not been for the existence of
some potentially-influential aspect of the Paschal liturgy. Later writers associated the
two candles with the apostles;2 but the application of that symbolism in this instance is
not valid, since in later times there is a clear contrast between the two candles and the
single Easter candle representing Christ. The possibility that the candles stood for the
two natures of Christ is both an unlikely and an unsatisfactory explanation. There is no
hint of this in any of our sources; and such mystical symbolism is more characteristic
of Orthodox theology. Moreover, the fact that both candles were lit with the same fire
makes this interpretation awkward. In the five papal ordines, cited at the beginning of
the chapter, there is no hint of what the candles represented; nor can any assistance be
derived from the Milanese and Spanish rites in which a single light was used.
Since we cannot readily account for either the number or the size of the candles
which represented the angels, we must examine the possibility that the desire to repre-
sent visually the angels at the Vigil resulted in the introduction of the two man-sized
candles into the liturgy of Holy Saturday, from which illuminations had hitherto been
absent. If so, this would assign the ceremony with its additional luminous feature to
that category of services envisaged by Herbert Thurston, when he wrote (of the Easter
candle): 'In this case we have, I believe, a ceremony that was really designed from the
beginning with a strictly mystical and symbolical meaning. ' 3 However, it is doubtful if
any religious ceremonies have such origins - even Thurston admitted that there were not
many - and this is true of many elements and features of Christian ritual, which have a
functional or utilitarian origin and subsequently become endowed with a symbolical
interpretation. Such, we believe, is the origin of the presence, as opposed to the height,
of the two man-sized candles. Once, however, a ritual or an element within a ritual has
1
'Lucernarium-lamps'.
2
For instance, Honorius of Autun, Sacramentary, (PL 172.748B).
3
Holy Saturday p.4.
329
been endowed with a symbolical significance, that ritual or element may undergo a
development which will result in a closer approximation between it and the object that it
represents.
Such a development appears to have occurred within the ritual involving the two
candles at the Roman Vigil. It would satisfactorily account for both their size and their
number. Two small lights were carried into church at the start of the Vigil, a surviving
feature of the primitive Lucernarium which at Rome had disappeared on every other
day of the year. These lamps or candles were placed, or more likely held, one each side
of the altar while the Vigil-lessons were read. We have already noted that the only ser-
vice to have taken place since the evening of Maundy Thursday in the Cathedral of
StJohn Lateran, wherein the Vigil was held, was the night office of Holy Saturday. The
latter service was conducted in darkness. We suggested above that, because of the
absence of illumination during this period, 1 the entire church was held to symbolise the
Tomb of Jesus. The sepulchral atmosphere is further emphasised by the use of the dim
light in which the Vigil-lections were read. This would recall the time of day when the
women visited the Tomb on the first Easter morning. Furthermore, the announcement
of the two angels was of a prophetic nature, as was the content of the Vigil-lections. It
is our contention, therefore, that, suggested by the content of Luke 24:4, the two lights
which had survived from the primitive Lucernarium were increased in size to convey
the impression of human height. Moreover, if the two large candles, which were held
either side of the altar, had the same significance as those in Poitiers, the altar coming
between the two candles would have represented the Tomb; and although documentary
evidence is lacking for the altar at StJohn Lateran, at St Remigius' Abbey the hiding of
the fire behind or under the altar possibly shows that the notion of the altar as a tomb
was not unknown. 2 The increase in the size of the candles would have resulted in the
provision of more light for those who read. The candles were subsequently carried in
procession to the font. Since the Pope was honoured with two smaller candles in the
normal course of events, it does not seem unreasonable to suppose that at the blessing of
the font the two man-sized candles served a dual purpose; and that, when they had been
returned to the altar for the start of the Easter Mass, they then symbolised the presence
and authority of the Pope.
* * * * *
1
Apart from the reserved fue which was hidden from view.
2
It would satisfactorily explain the practice at this monastery. See below, Section (iv)
p.332.
330
In the earlier chapters of Part N we saw that our knowledge of the Easter candle
in this early period' was derived almost exclusively from references in letters and from
surviving laudes cerei. In the period between the earliest reference to the Candle and
the frrst mention of it within the context of a service,2 the ceremony in which the Easter
candle figured prominently had been adopted by churches throughout a considerable
part of Europe north of the Alps, almost certainly as a result of Milanese or Aquileian
influence. 3 However, during this same period Roman influence was also at work; and
the results of this we shall note in the following sections. However, sufficient purely
Gallican documentary evidence is extant for us to observe the main features of the
Easter candle ceremony in Gaul before the importation of Roman elements.
(iv) The eighth-century Gallican rite: Mode A
All of our sources4 refer to the Easter candle as cereus without any adjectival
qualification. None of them mentions its being brought into church as part of a pro-
cession, though Ordo 17 records the entry into church of the deacons 'without lighted
candles' (§102). The possibility, therefore, should not be ruled out completely that it
was carried into church, in view of the evidence of later centuries and in view of the fact
the sources in question concern themselves primarily with the lighting and blessing of
the Candle. The officiant at the ceremony was a deacon or the archdeacon,5 who, after
making the sign of the cross, lit the Candle with the frre which had been reserved for
that purpose the previous day, and began his benediction. The frre was obtained from
the candle, or more likely the lamp, which had been concealed from view at the con-
clusion of the night office in the early hours of Good Friday morning, or from
1
That is, c .AD 400-700.
~at is, from the late fourth to the mid-eighth century.
3
By the fourth century Milanese jurisdiction extended as far as Rhaetia; and by the fifth
century Chur, Augsburg, and Ratisbon were subject to Aquileia. As late as the eleventh
century these three cities were still using Ambrosian liturgical books. (Borella p.105).
4
OR 17.103; OR 30A.15; GeV p.68; GeAng p.52; GePr p.55; Sacramentary of St
Remigius' Abbey p.328; and the Gradual of St Gregory. A fragment of this last-men-
tioned source is to be found in Macri, Hierolexicon p.142. These references also relate
to the ceremony as a whole.
5
Five of the documents cited in the previous note mention the archdeacon. Ordo 17
refers to ille qui cereum benedici (sic) debet, 'he who has to bless the Candle'- pre-
sumably, a deacon. The Sacramentary of St Remigius' Abbey states that the duty was
performed by the sacerdos, who may have been a priest, the abbot, or even the bishop.
This is the only recorded instance of the ceremony being performed by someone other
than a deacon or an archdeacon. The preceding rubric rules out the possibility that sac-
erdos is here used to indicate any member of the clergy.
331
f'rre reserved on Maundy Thursday. 1 It is the writer's belief that at St Remigius' Abbey
the fire was reserved behind or under the altar. 1bis would explain why the officiant
moved to the rear of the altar :
Deinde veniens sacerdos ab oratione vadit retro altare, accipiens de lumine quod
sexta feria absconsum fuit. 2
This location would also be most convenient for the concealing of a lamp at the
conclusion of the night office of Good Friday. It was observed above on page 88 that in
the Roman rite the last candle at Tenebrae was hidden in this very place. As to the
position of the Easter candle, all our sources except the Sacramentary of St Remigius
state that the officiant stands ante a/tare. In view of this it is almost certain that the
Candle was placed also in front of the altar, presumably in a central position.
(v) The later Gallican rite: Mode B
An alternative Gallican tradition, also free from Roman influence, is contained in

the tenth-century Sacramentary of Corbie. 3 Although the document relates to a period
some two hundred years after that which the documents in the previous section attest,
elements present in this rite are also found in the ninth-century Ordo 31 and in the
eighth-century Ordo 26 and Ordo 28, and this would suggest that the tradition recorded
by the sacramentary was much older than the tenth century.4
The ceremony described in this document differs from the eighth-century Gallican
rite in the above Section (iv) in five important respects. (1) The service is presided over
by the bishop with all orders of clergy in attendance. (2) The Easter candle is borne in
procession into church5 before being set down in medio choro. (3) The Easter candle is
then consecrated by a deacon who sings the Exultet after he has received a blessing
from the bishop. The blessing of the Candle by the deacon6 preserves the primitive
practice of the Church, first found in the fourth century,7 and suggests that its
performance by the archdeacon was a later development. (4) The frre for lighting the Easter
1
Quintaferia. Thus the Gradual of St Gregory (Macri p.142). For the reservation of the
frre, see Part I Chapter 2 p.15.
2
'Then, the prayer ended, the priest goes behind the altar and receives a light from the
fire which had been concealed on Good Friday ... ' Sacramentary p.328.
3
PL 78.336B/C = M 106.
~e blessing of the Easter candle is not contained in Ordo 26, only alluded to. Ordo 28
and and Ordo 31 should be used as corroborative evidence with some caution because
of their Roman accretions.
5
Also attested in OR 28.59 and OR 31.63.
6
Also attested in OR 28.60.
7
See Chapter 2 p.228.
332
candle is that newly kindled on Maundy Thursday and reserved until Holy Saturday,
unlike the fire of Mode A which was reserved on Good Friday. The reservation of fire,
newly kindled with a flint is first attested in Ordo 26 (§3). The sacramentary also con-
tains the fuse recorded instance of the blessing of the new frre (and the incense) by the
bishop (or a priest), but does not include the prayer of consecration. The frre and the
incense are blessed while the deacon is chanting the Exultet. (5) The lighting of the
Ea-;ter candle during the Preface at ignis accendit contrasts markedly with the practice
in Mode A where the Candle is already alight at the start of the benediction, and marks
the beginning, as far as documentary evidence is concerned, of that process of liturgical
development which caused the Ia us cerei to undergo a series of interruptions.
Although most of the documentary evidence for the ceremonial surrounding the
Easter candle in tenth-century Gaul attests the presence of the two Roman Vigil-candles
and the lighting of the Easter candle prior to the chanting of the Exultet, the significance
of the practice at Corbie should not be underestimated; for the twin candles ultimately
either declined in liturgical importance or disappeared completely from the Paschal
rites, and it became universal practice to light the Easter candle at ignis accendit. We
can attribute with some confidence to Lanfranc the triumph of these two features of the
Corbeian rite : the absence of the two Vigil-candles and the lighting of the Easter candle
during the Preface. For in spite of the liturgical development which took place else-
where in the century which intervenes between the Sacramentary of Corbie and Lan-
franc's work, both these features are to be found in the elaboration of ceremonial set out
in his Decrees; and Corbeian parentage seems very likely. For on the one hand Corbie
was a very influential Benedictine centre in Northern France c.lOOO; and on the other
Lanfranc was for a time Abbot of St Stephen's Monastery at Caen, and the local or
regional liturgical variations of the area may well have been assimilated by him during
his sojourn in that part of France. Certainly, the directions for the ritual surrounding the
blessing of the Easter candle according to the Benedictine Regularis Concordia of the
previous century are quite different from those enjoined by Lanfranc. 2
(vi) The Roman Vigil in Gaul
In Section 1 of this chapter we discussed the evidence of the five Roman ordines
in an attempt to make a partial reconstruction of the form that the Paschal vigil took at
the papal court in Rome. There seems little doubt that this Roman form of the Vigil
was in use in a number of churches in Gaul, since all five ordines have their provenance
1
0n the assumption that the sacramentary antedates PRG.
2
A certain amount of ceremonial variation existed from one Benedictine house to
another, a fact Lanfranc himself commented upon in his Decrees (PL 150.4678).
333
in that country, though they attest Roman practice, and since the Roman form of ser-
vices was encouraged by Charlemagne in his desire to impose liturgical uniformity
throughout the length and breadth of his vast domains. In addition to these five ordines
the Sacramentary of Autun also contains the Roman form of the Paschal vigil. Like
some of the other Gelasian sacramentaries, it includes the Deus mundi Conditor, the
benediction-formula for the Vigil-candles. 1 We do not know how widely the Roman
form of the Vigil was adopted in the regions to the north of the Alps. Charlemagne's
onJy-partial success is reflected in the emergence of the Romano-Gallican synthesis
from the liturgical disorder of the eighth and ninth centuries.
1
GePh p.63. The Sacramentary was formerly referred to as the PhillippsGelasian Sacra-
mentary.
334
Chapter Seventeen
THE ROMANO-GALLICAN SYNTHESIS
(i) The diversity of practice
It was perhaps inevitable that with the importation of the Roman form of the Vigil into
Gaul its fusion with the local Gallican rites not only took place but was achieved with
differing results. The diversity of practice was known to Amalarius; and the recom-
mendation in his Liber de Ordine Antiphonarii, written c.AD 830, should be seen as an
attempt by him to promote the synthesis of Roman and Gallican ceremonial practices.
Hoc est quod dico : reservetur ignis de sexta feria, ut inluminetur cereus qui poni-
tur in vice columnae ignis benedicendum, qui ab initio benedictionis inluminatus
est, et cum benedictus est, ab eo inluminetur secundus cereus. 1
However, a compromise along these lines was bound to be forced- in fact there no evi-
dence that his suggestion was adopted by any church at the time, since the significance
of the two Vigil-candles would have been lost, if one had been converted into the Easter
candle and the other had been made subordinate to it. It is true that at Aquileia in the
sixteenth century ,2 the status of the two candles corresponds to that prescribed by Ama-
larius; but we cannot be certain that the presence of these two candles was the result of
the recommendations of Amalarius. However, it is very likely that the candle lit at the
conclusion of the Ia us cerei in the Benedictine rite, which we shall discuss in Section
(ii), corresponded to the secundus cereus, enjoined by Amalarius.
That the two candles envisaged by Amalarius were of unequal status is clear both
from the symbolism which he attaches to each and from their use at the blessing of the
font. For him the larger candle is Christ; whilst the smaller one, which takes its light
from the larger, represents the company of apostles to whom Christ said 'You are the
light of the world'. 3 After the Vigil-lections both candles are to be borne in front of the
catechumens to Baptism, but only the larger one is immersed in the water. 4 The Bene-
1
'This is my recommendation: let fire be reserved from Good Friday to light the Candle
which represents the column of flre and which is set in position to be blessed. It is lit at
the beginning of the blessing, and after it has been consecrated, a second candle is lit
from it.' WOA XLIV.8.
2
1519 Missal fol.98.
3
Liber officialis 1.20.2.
4
Liber officialis 1.20.2 and 1.26.1-4.
335
dictine development, which we shall discuss towards the end of this chapter, took place
probably as a result of the influence of Amalarius' writings; but there were practical
considerations also.
The diversity of practice, which had prompted Amalarius' solution for a synthesis
of ceremonial, manifested itself in a number of different ways. Again for convenience
we will refer to them as modes.
Mode 1. The description of the lighting and blessing of the candles in the late
ninth-century Ordo 29 shows, not a fusion of two traditions, but a juxtaposition of two
quite separate services. 1 The two Vigil-candles of the Roman rite are lit with fire newly
kindled on Holy Saturday, and are placed one either side of the altar. After the reading
of the Vigil-lections the clergy depart together with the two candles. On their return,
the Easter candle is lit with ftre kindled on and reserved from Maundy Thursday. The
separate performance of both the ceremonies on Easter Eve would appear to be a transi-
tional stage in the fusion of both rituals. Ordo 29 is also unique in that it is the only
early Roman ordo to record a separate kindling of ftre on all three days of the Triduum.
Mode 2. This variation has features in common with the eighth-century Gallican
2
rite but belongs more to the Roman than the Gallican tradition It is found in the
tenth-century Pontifical of Wolfgang Bishop of Regensburg :
Diaconus ... accipiat de igne qui sexta feria fuerat excussus de lapide. et incendat
duos magnos cereos. et faciens crucem benedicat eos. 3
It would appear from this pontifical that the Easter candle was unknown at Regensburg
in the tenth century. Alternatively, the section relating to the Easter candle may be
missing from this document. The two large candles, which later escort the bishop,4 and
the new ftre kindled from a stone on Good Friday are clearly Roman elements; but the
blessing performed by the deacon and the sign of the cross would appear to come from
the Gallican tradition.
'OR 29.45-48.
2
See Chapter 16 Section (iv).
3
'Let the deacon receive some of the ftre which had been kindled from a stone on Good
Friday; and let himlight the two large candles; and after making the sign of the cross
let him bless them.' Pontifical p.125.
4
Recalling the Pope's honorific lights- see pp.9-10. Pontifical p.130.
336
Mode 3. Mode 1 and Mode 2 may represent isolated instances of attempts to har-
monise the different liturgical traditions. In Mode 3, however, we have what appears to
be a fusion of the Gallican and Roman uses of light, in which the principal features of
both would appear to retain their original importance and significance. Thus in the
eighth-century Gelasian Sacramentary of Gellone 1 it is the Easter candle which is
kindled and blessed; whilst at the blessing of the font only the two Vigil-candles are
mentioned. 2 Although there is no indication in the rubrics of this sacramentary, it is rea-
sonable to assume that these two candles were also alight during the reading of the lec-
tions. Support for this view comes from the late ninth-century Pontifical of Poitiers,
which also records the very point in that rite at which these candles were lit. 3 We learn
that the two man-sized candles are held by notarii who stand one either side of the
Easter candle or the altar; and that they are kindled with fire taken from the flame of
the Easter candle at the conclusion of the laus cerei. Poitiers concurs with the Sacra-
mentary of Gellone in attesting the presence of only the two Vigil-candles at the bless-
ing of the font (ibidem p.216).
Evidence for the widespread use of the Romano-Gallican synthesis throughout

Western Europe is plentiful. At Mainz in the tenth century the two man-sized candles
in candlesticks stood one either side of the altar and were lit at the conclusion of the
laus cerei.4 This arrangement is also found in the slightly-later Alcuin; 5 but it is poss-
ible to detect in that document a decline in the importance of the two Vigil-candles,
since at the blessing of the font the choice is permitted of immersing either the two large
candles or the Easter candle. The placing of the two candles in candlesticks also indi-
cates a reduction in their physical size.6 The use of two smaller candles placed either
side of the Easter candle is documented as early as the eleventh century at Rouen; 7 and
by Beleth,8 Honorius of Autun,9 and Durandus. 10 The latter informs us that this practice
obtained 'in most churches'.
1
GeGe pp.92-3.
2
GeGep.99.
3
Poitiers p.215.
4
PRG IT p.99 §348.
5
PL 101.1216C.
6
Evidence for this is to be found in PRG and Alcuin, ibidem; PR XII l.xxxii.8 p.240;
Pontifical of St-Germain-des-Pres (DAER 4.24 p.159, M 230); and at Marseille in the
thirteenth century (ILEM p.84 ).
7
8
Rationale, PL 202.110C.
9
Sacramentary, PL 172.748B.
10
337
At Milan in the twelfth century two small candles were held by deacons one either
side of the Easter candle, and two large candles stood at the font. 1 By the seventeenth
century the use of the latter had been discontinued. 2 At Braga the Missal of 1512 refers
to the two candles lit during the Preface (np), but they are not mentioned in the Missal
of 1558. We referred above to the practice at Aquileia and noted that the use of two
candles in that church corresponded to the prescription of Amalarius. In the primitive
Mozarabic rite no additional candles were lit to accompany the Easter candle. How-
ever, according to the Missale Mixtum of 1500 two candles are to be lit from the flame
of the Easter candle. 3 Of these Martene informs us that one stood to the left ofthe
altar, whilst the other was placed behind. 4 No reason is given for this arrangement. At
the Benedictine monastery of Monte Cassino in the twelfth century the two additional
candles were placed ad altare. 5 Since the Easter candle stood next to the ambo, the
phrase should perhaps be interpreted 'on either side of the altar'. This is the position
mentioned in the contemporary Pontifical of St-Germain-des-Pres.6 Here the Easter
candle may also have stood by the ambo.
We mentioned above the statement of Durandus that in most churches two small
candles stood either side of the Easter candle. Their absence from the minority of
churches may be explained in three ways. (i) As (originally) baptismal candles, they
would have lost their raison d' etre, if the ceremony of the blessing of the font no longer
took place, as happened in some monasteries. (ii) There may have been some churches
in which the use of the two candles was unknown, especially if they preserved a Galli-
can rite free of Roman influence in this respect. (iii) We have already observed that
their function and the function of the two acolytes' or torch-bearers' candles may well
have been merged; or they may have disappeared if the bishop was accorded two
honorific candles.
(ii) The Benedictine development
There can be little doubt that at the Monasteries of Monte Cassino and
St-Germain-des-Pres the ceremonial surrounding the kindling and blessing of lights at
the Easter vigil followed the injunctions of the tenth-century Benedictine Regularis
Concordia. This enjoined upon the houses of that order the principal features of the
1
Beroldus p.11 0.
~ey appear in Ambrosian missals of the sixteenth century, but not in the Missal of
1669.
3
PL 85.442C.
4
DAER 4.24 p.164.
5
PR XIII p.293 and Martene, DAMR 3.15.10 p.143. See also Appendix 11.
~artene, DAER 4.24 p.159, M 230.
338
Romano-Gallican synthesis viz. the lighting and blessing of the Easter candle, and the
kindling of the two candles, which were held by acolytes one each side of the altar. In
addition to these lights the Regula also directs that another smaller candle should be
kindled at the conclusion of the Ia us cerei, 1 but gives no indication as to its purpose.
Since its function was not to provide a source of fire in an emergency ,2 it either consti-
tuted the second candle, which we saw at the beginning of this chapter had been
enjoined by Amalarius, albeit reduced in size, or it acted as a surrogate for the Easter
candle, which for one reason or another remained in front of the altar or next to the
ambo, while the font was being blessed. In later times it had the latter function, as is
evidenced at Worcester where it was known as the cereusfontium,3 and at Barking. 4
The Gilbertine Ordinal suggests that this small candle was only used where the blessing
of the font took place.5 The use of the cereus minor was not confmed to the monastic
tradition. Sicardus refers to it, again in a baptismal context. 6 For him the small candle
signifies the Order of the Apostles, since Christ~ who is symbolised by the larger Easter
candle, addressed his disciples 'You are the light of the world', and since the apostles,
like Christ, illuminate the Church. Durandus mentions its existence, but does not com-
ment upon its purpose. 7 It featured in the rites of Salisbury8 and Aquileia;9 and at the
Church of St-Agnan in Orleans it accompanied the Easter candle in the procession to
the font as late as the eighteenth century. 10 There is also a reference to this candle at
Reims. 11 In the older rite at Braga this candle was lit with the serpent-candle before
being dipped into the font.
1
PL 137.494C.
2
In the event of the failure of the reed-candle or setpent-candle, an additional source of
ftre would have been available from the candle in the lantern or from one of the candles
of the acolytes.
3
Antiphonary p.69.
4
HBS 65 p.lO 1.
5
HBS 59 p.40.
6
Mitrale, PL 213.324D-325A.
7
Rationale IV.80 p.352.
8
13 C Missal (Warren) p.270.
9
1519 Missal fol.98.
'CUe Moleon p.209.
11
1770 Missal p.224. See also Appendix 9 p.358.
339
Appendixes
APPENDIX 1 - The Omissions .......................................................................... 341
APPENDIX 2 - The Roman and Galli can traditions .. .... .. .. .. .. .. .... .. .. ...... .. .. .. .. .. . 342
APPENDIX 3 - The Origin of the Seven Lamps .. .. .... .. .. .. .. ...... .. .. .. .. .. ...... .. .. .. .. .. 344
APPENDIX 4 - The non-observation of the new fire and Easter candle

ceremonies .. ............................ ........ .............................. .............................. ...... .... 347
APPENDIX 5 - The Formulas for the benediction of the New Fire .................. 348
APPENDIX 6 - The Sepulchre .......................................................................... 353
APPENDIX 7 - Illumination on Good Friday at the Passion and at the

Adoration of the Cross ......................................................................................... 355
APPENDIX 8 - The Paschal Column and Candelabrum .................................. 356
APPENDIX 9 - The Blessing of the Font and Baptism ...................................... 358
APPENDIX 10 - Anticipation ............................................................................ 361
APPENDIX 11 - Monte Cassino A .................................................................... 362
APPENDIX 12 - The Inventor rutili of Prudentius ............................................ 365
APPENDIX 13 - The Mozarabic and Milanese Rites ........................................ 366
APPENDIX 14 - The Chartae. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..... .. .... 368
APPENDIX 15 - The Suburbicarian Dioceses .................................................. 371
340
APPENDIX 1 - The Omissions
Since the eighth century in the Gallican Church and the tenth in the Roman rite, the
night offices of Matins/Lauds on Maundy Thursday, Good Friday, and Holy Saturday
were distinguished from the night offices of the rest of the days of the year by the
omission of the customary versicles and responses. The Deus in adiutorium and the
lnvitatory were not said at the start of the service; no blessing was requested at the start
of the lessons; and the Kyrie eleison was omitted at the conclusion of Lauds. Above
all, the Gloria Patri doxology was omitted at the close of each psalm. 1 These same
omissions, which are also found in the Office for the Dead and represent the ftmereal
aspect of Tenebrae, were 'made designedly to mark the deep mourning in which the
Church is plunged' ,2 and seem to point to the great antiquity of the office. 3
In the Middle Ages these omissions of the beginnings and endings were given an
allegorical interpretation. According to Rupert of Deutz4 Christ, who is the A and Q,
the beginning and the end, is dead during this period. Hence they are omitted. This
explanation is also to be found in the Ordinal of StMary's, York. 5 Durandus also men-
6
tions that the Gloria's etc. are not said because Christ is lying in the tomb.
1
Sicardus, Mitrale, PL 213.297A. In the now-superseded night office of Holy Saturday
according to the Ambrosian rite, the Gloria Patri was omitted from the end of the
twenty-three psalms and one canticle. Breviarium Ambrosianum, Pars Prima.
~urston, UIW p.243.
3
Batiffol p.93.
4
De Div.Off. V, PL 170.148A.
5
HBS 75 p.271.
6
341
APPEND IX 2 - The Roman and Gallican traditions
There has been a tendency in the past amongst some writers, when dealing with the
development of the Roman and Gallican liturgies in the period AD 700 to AD 950 to
assume the existence of a uniform Roman rite and similarly a uniform Gallican rite.
The picture of two contrasting and at times mutually interacting liturgies could not be
further from the reality of the situation. It is probably true that, as the temporal power
of the papacy increased, in the areas which were subject to the Roman Church, the pres-
sures towards conformity were considerable; but in the areas of so-called Gallican
influence, particularly in the decades before the reign of Charlemagne, the existence of
several liturgically-independent churches and the absence of a central unifying force
inevitably resulted in a variety of differing local rites.
It is beyond the scope of this work to investigate the ancestries and origins of the
Gallican liturgies. It seems quite likely, however, that the various Gallican rites, ulti-
mately deriving from the same parent ancestor as the Roman rite and developing inde-
pendently, at times under the pressure of local pagan customs, were influenced to a
greater or lesser degree by the liturgies of Aquileia and Milan, depending on their
geographical proximity to those centres, probably more than is generally recognised;
and even in the period of increased Roman influence, from AD 789 onwards, the Mila-
nese Church in all likelihood continued to be influential, especially in those areas where
the boundaries of the Gallican and Milanese churches were coterminous.
Prior to the Decree of Conformity in AD 789 it can be safely said that there was
considerable variety in the forms of service found throughout the Gallican Church.
When St Chrodegang, after a visit to Rome in AD 751, introduced the Roman rite at
Metz, that church became 'a Roman lighthouse in a sea of liturgical disorder'
(A.A.King). In view of the state of political unrest in Gaul and Western Germany in the
preceding centuries, liturgical conformity had not been possible. It was Charlemagne
who attempted to impose throughout his vast domains the Roman rite, which he had
introduced into the royal chapel of his capital at Aachen, as part of his attempt to unify
his extensive empire.
The attempt can only be described as partially successful and slow, if his intention
was to replace completely the Gallican rites with the Roman. For it was not until
c.AD 810 that the Roman rite was introduced at Lyon by Leidrad; and in AD 832 the
night offices of the Triduum with which Amalarius was familiar were decidedly non-
Roman.
342
After the death of Charlemagne the Treaty of Verdun in AD 843 destroyed the
political unity which had facilitated the partial imposition of liturgical uniformity. The
next hundred years or so witnessed the gradual gallicanisation of the Roman rite,
especially in the region at the centre of which lay Mainz, whence originated the Roma-
no-Germanic Pontifical, the ancestor of the future Tridentine Missal and Breviary.
343
APPENDIX 3 - The Origin of the Seven Lamps
The presence of seven lamps or candles which stood or were suspended from the ceiling
in front of the high altar is more widely attested in the Gallican than the Roman tradi-
tion. 1They are mentioned in five ordines of Gallic an origin2 as featuring either at Lauds,
where they were extinguished gradually, or at the illumination of the church on the
evenings of Maundy Thursday or Good Friday. They are also attested in three other
early documents. 3 Ordo 30B, which purports to describe papal ceremonial, mentions
them within the context of the extinction of light at Tenebrae (§28); but Tenebrae as
such was unknown at Rome at the end of the eighth century, the date assigned to that
ordo; 4 so that the description of that service in the ordo relates to a non-Roman church.
However, there is no reason to believe that seven lamps did not burn before the high
altar of St John Lateran in Rome.
The seven lamps, however, must not be confused with the seven candles, borne in
procession by acolytes, which were always present at any Stational Mass which the
Pope celebrated.5 These lights were carried by six acolytes and one subdeacon or head
acolyte, and were placed before the altar immediately after the Peace. They are to be
distinguished from the two honorific (civic) candles which preceded the Pope on every
occasion,6 and which at Mass were placed behind the altar. 7
The function of these seven candles and what they represented or symbolised are
matters for some debate. The city of Rome had been divided by Pope Fabian
(AD 236-50) into seven ecclesiastical districts, using in all likelihood as a basis the
fourteen civil regions which the emperor Augustus had constituted. Ordo I records that
whenever the Pope celebrated Mass, six acolytes and one subdeacon for the district 'on
duty' for the day would act as candle-bearers.8 The candle of each of these seven min-
isters may well have represented one of the seven districts of Rome. That all seven
ministers on any one occasion should come from the same district, and that six of them
were severally representing the other six districts does seem somewhat strange. It
1
It has yet to be shown that the seven-branched menorah, which stood in the sanctuary
of a number of larger churches, was ever substituted for the seven altar lamps. At Dur-
ham, for instance, the central holder of the menorah held the Easter candle. (Raine p.9).
2
0R 26.9; OR 28.30; OR 29.17; OR 31.13; OR 32.5.
3
Poitiers p.138; PRG IT p.57 §213; Alcuin, Lib.de Div.Off., PL 101.1205D.
4
See especially pp.11-13, above.
5
0rdo I records the ceremonial for the Mass of Easter Day. This description of the lit-
urgy is applicable to any of the eighty-nine Stational Masses.
6
See above, pp.9-10.
7
0R 4.7 and OR 30B.37.
8
Atchley, Ordo Romanus p.39. For the concept of the division of duties among the Jew-
ish priesthood, see inter alia Luke 1:8.
344
seems much more likely that, if we are dealing with a situation which involved dis-
trict-representation, there would have been one acolyte from each of the districts. But
since we are told that the seven acolytes were all from the same region on any one
occasion, it casts some doubt on whether each candle stood for one district.
The writer believes that a much more profound symbolism was attached to the
seven pap£11 candles, The celebration of a Stational Mass by the Pope was a time of
great importance and solemnity. He was accompanied by all the dignitaries of the
Church, since the occasion represented both the unity and the totality of the Roman
Church at its most solemn function. To a point the presence of all the Roman clergy
symbolised the fact that the Mass was being celebrated on behalf of the whole city of
Rome. In a more visible and dramatic way the presence of the seven processional
candles also symbolised the major constituent elements of the church, on whose behalf
the whole congregation had gathered, in the following way.
We read in the Book of Revelation (1:12 and 20) that the seven golden lampstands
symbolised the Seven Churches of Asia, and by extension the whole of the Church in
the Roman province of Asia. The writer believes that the symbolism was borrowed by
the Roman Church in view of the existence of the seven so-called' Constantinian basil-
icas within the city of Rome; and that each of the seven processional candles in question
represented one of those seven major basilicas. At what period this feature of seven
lights first appeared in Rome we can but hazard a guess. We have already observed that
the concept of seven patriarchal basilicas arose after the time of Sixtus ill. It is possible
that the use of the seven candles is much older and that they previously represented
some other ecclesiastical heptad, either in the world-wide Church or found locally in
Rome. We might consider, for instance, the seven seasons of the Christian year,2 the
seven suburbicarian dioceses, the seven Stational Masses at the Cathedral of St John
Lateran,3 or the seven martyrs, venerated at Rome, who are mentioned in the communi-
cantes prayer in the canon of the Tridentine Mass.
When we turn to consider the origin of the seven stationary lamps which stood or
hung before the high altar and which were extinguished at Tenebrae, it is not difficult to
believe that these lights were scripturally-inspired, like the seven processional candles
'It is now known that the Basilica of S. Maria Maggiore, unlike the other six, is not of
Constantinian foundation, but dates from the pontificate of Sixtus ill (AD 432-40). It
seems unlikely, therefore, that the notion of the seven major churches of Rome arose
before the second third of the fifth century.
2
Advent, Christmas, Epiphany, Lent, Easter, Pentecost, Common Time.
3
lt is interesting to note that of the 89 Stational Masses at Rome 49 (ie 7 x 7) were cel-
ebrated in the major basilicas: StPeter's 13, S.Maria Maggiore 12, StJohn Lateran 7,
St Paul's 6, S.Lorenzo 4, S.Sebastiano 4, S.Croce in Gerusalemme 3. ·
345
of the Roman rite, also from the last book of the New Testament. According to Revel-
ation 4:5 'before the throne bum seven torches of fire, which are the seven spirits of
God'. Within the context of the liturgy the throne of God is represented by the altar,
and the seven 'torches of fire', which bum perpetually and have an essential relation-
ship with God are quite clearly portrayed by the seven liturgical lamps which bum in
front of the altar. 1 For they are permanent lights whose use is not restricted to one
service. On the other hand the seven candles, which formed part of the Pope's pro-
cession, were placed temporarily in front of the altar and their use was confmed to papal
High Mass. The objection cannot be sustained that two sets of seven lights would be
unlikely; for there is still a simultaneous use of seven acolytes' torches and seven altar
candles at High Mass in the papal liturgy today.
In the same way that the flame of each lamp in Revelation 4:5 was perpetually
burning, it is possible that the seven lamps of the earthly liturgy were never allowed to
be extinguished, except once a year. The evidence of Ordo 308 would suggest that this
is likely to have occurred at the conclusion of Matins/Lauds of Good Friday, the last of
the lamps being removed rather than extinguished at the end of the service.2 Possible
support for this view is to be found in the rubrics of a number of documents which
relate to the entry into church on Maundy Thursday of the procession bearing the new
fire. All attest the concern that the seven lamps should be rekindled speedily and effi-
ciently.3
'Another, though less likely, explanation is given by John the Deacon. He explains that
seven is a sacred number, since it is the sum of four, the number of Gospels which attest
the Trinity, and three, the Trinity itself. (PL 59.4038.) On the face of the tribunal above
the apse in the Church of S.Prassede in Rome is a mosaic portraying the Lamb who is
flanked on either side by three and four candlesticks. These are said to be 'allegorical
of the seven mysteries' (Forbes p.257).
2
0R 308.28. See The Development ofTenebrae, Stage 2, pp.11-13.
30R 26.9; OR 29.17; PRG ll p.58 §220; Alcuin, PL 101.1205D.
346
APPENDIX 4 - The non-observation of the new fire and Easter candle
ceremonies
In view of the simplicity and austerity of its rite the ceremonies of the new fire and the
Easter candle have never been observed in the houses of the Carthusian order. Follow-
ing the Second Vatican Council some members of the order felt that the Easter candle
ought to be adopted; but the idea was not accepted. 1
Formerly the Easter candle was not blessed in the Church of St Stephen and the
Church of the Holy Cross in Lyon, the congregations of those churches repairing to the
cathedral for the ceremony.2 There was a similar custom for the faithful of the parochial
churches of Evreux and its suburbs to attend the mother church, and subsequently to
take some of the new fire home. 3
In spite of the restoration of the Paschal rites to their former times following the
liturgical reforms of the Roman Church in 1955, the ceremonies are not observed in all
parish churches; and only a small number of Anglican churches observe the ceremonies
of Holy Saturday, which closely follow those of the Roman rite. At Amiens Cathedral
they fell into desuetude in 1969, having been revived in their present form in 1955.
This was mainly the result of poorly-attended services during those fourteen years. 4
1
Letter to the writer from Fr Bernard O'Donovan O.C. of St Hugh's Charterhouse, Hor-
sham, England, dated 22 June 1984.
2
1771 Missalp.194.
3
1740 Missal p.187.
4
Letter to the writer from L 'abbe P.Grey, Secretaire General of that diocese.
347
APPENDIX 5 - The Formulas for the benediction of the New Fire
A. Deus, qui per Filium tuum, angularem scilicet lapidem, claritatis· tuae ignem fidelibus contulisti :
productum e silice, nostris profuturum usibus, novum hunc ignem sanctifica: et concede nobis, ita per
haec festa paschalia coelestibus desideriis inflammari; ut ad perpetuae claritatis, purls mentibus,
valeamus festa pertingere. Per.
* PRG has caritatis.
A l. Deus, qui per Filium tuum claritatis tuae ignem fidelibus contulisti, novum bunc ignem + sanctifica,
et concede nobis, ita per haec festa paschalia caelestibus desideriis inflammari, ut ad perpetuae claritatis
puris mentibus valeamus festa pertingere. Per.
B. Exaudi nos, lumen indeficiens, Domine Deus noster, unici luminis lumen; fons luminis, lumen auctor
luminum, quae creasti et inluminasti; lumen angelorum tuorum, sedium, dominationum, principatuum,
potestatum et omnium intelligibilium, quae creasti; lumen sanctorum tuorum. Sint lucemae tuae animae
nostrae; accedant ad te et inluminentur abs te; luceant veritate, ardeant caritate; luceant et non
tenebrescant, ardeant et non cinerescant. Benedic hoc lumen, o lumen, quia et hoc, quod portamus in
manibus, tu creasti, tu donasti. Per haec lumina, quae accendimus, de hoc loco expellimus noctem; sic et
tu expelle tenebras de cordibus nostris. Simus domus tua lucens de te, lucens in te; sine defectu
luceamus et te semper colamus; in te accendamur, et non extinguamur.
B 1. Dominus Deus, Pater omnipotens, lumen indeficiens, qui es conditor omnium luminum : benedic
hoc lumen, quod a te sanctificatum atque benedictum est, qui illuminasti omnem mundum : ut ab eo
lumine accendamur, atque illuminemur igne claritatis tuae : et sicut illuminasti Moysen exeuntem de
Aegypto, ita illumines corda, et sensus nostros; ut ad vitam et lucem aetemam pervenire mereamus. Per.
B 2(a). Domine Deus noster, Pater omnipotens, exaudi nos, lumen indeficiens : tu es sancte conditor
omnium luminum, benedic, Domine, hoc lumen quod a te sanctificatum atque benedictum est. Tu
inluminasti omnem mundum, ut ab eo lumine accendamur et illuminemur igne claritatis tuae, sicut ignem
inluminasti Moyse, ita illuminabis cordibus et sensibus nostris, ut ad vitam aetemam pervenire
mereamur. Per.
B 2(b). Domine Deus noster, Pater omnipotens, exaudi nos lumen indeficiens: tu es sancte conditor
omnium luminum, benedic, Domine, hoc lumen quod a te sanctificatum atque benedictum est. Tu
inluminasti omnem mundum, ut ab eo lumine accendamur et inluminemur igne claritatis tuae, sicut igne
inluminasti Moysen; ita inluminatis sensibus et cordibus nostris ut ad vitam aetemam petvenire
mereamur. Per.
B 2(c). Exaudi nos, lumen indeficiens : tu es sancte conditor omnium luminum, benedic, Domine, hoc
lumen quod a te sanctificatum atque benedictum est. Tu inluminasti omnem mundum, ut ab eo lumine
accendamur et inluminemur igne claritatis tuae, sicut ignem inluminasti Moyse, ita iluminabis cordibus et
sensibus nostris, ut ad vitam aetemam petvenire mereamur. Per.
B 2(d). Dominus Deus noster, Pater omnipotens, exaudi nos lumen indeficiens: Tu es sanctus conditor
omnium luminum; bene + die, Domine, hoc lumen quod a te sanctificatum et benedictum est, ut ab eo
lumine accendamur, et illuminemur igne claritatis tuae. Sicut ignem illuminasti in rubo tempore Moysis,
ita illuminare digneris corda nostra et sensus nostros, ut ad vitam aetemam pervenire mereamur. Per.
B (2e). Deus pater omnipotens, exaudi nos, lumen indeficiens; tu es sancte conditor omnium luminum,
benedic, domine, hoc lumen, quod ate sanctificatum atque benedictum est. Tu illuminasti omnem
mundum, ab eo lumine accendimur et illuminamur igni claritatis tuae; igni ergo quo illuminasti Moysen
illumina quaesumus corda et sensus nostros ut ad vitam aetemam pervenire mereamur. Per.
B 3. Domine Iesu Christe lumen indeficiens exaudi nos tu qui es sancte conditor omnium luminum.
benedic domine hoc lumen quod a te sanctificatum atque benedictum est : tu qui iiJuminasti omnem
mundum : ut accendamur et illuminemur igne claritatis tuae, sicut illuminasti domine Moysen legiferum
tuum. ita illumina corda nostra et sensus. ut ad vitam etemam pervenire mereamur. Per.
348
B 4. Domine Deus noster Pater omnipotens, lumen indeficiens comlitor omnium luminum, exaudi nos
famulos tuos et benedic hunc ignem :qui tua sanctificatione atque benedictiooe consecretur. Tu illunri-
nans omoem hominem venientem in hunc mundum illumina conscientias conlis oostri igne tuae caritatis
ut tuo igne igniti tuo lumine illuminati : expulsis a conlibus nostris peccatorum tenebris : ad vitam te
illustrante pervenire mereamur aetemam. Per.
B S(a). The first part of this prayer is identical with B 4 above. After aetemant it continues: Et sicut
illuminasti ignem Moysi famulo tuo per columnam ignis ambulanti in mari rubro : ita illustra nostrum
lumen : et candela quae de eo fuerit accensa in honore maiestatis tuae semper perserveret benedicta : ut
quicumque ex eo lumine portaverit sit illuminatus lumine gratiae spiritualis. Per.
B S(b). Domine Deus Pater omnipotens lumen indeficiens, exaudi nos famulos tuos et benedic hunc
ignem qui tua sanctificatione atque benedictione consecrator : tu domine qui illuminans omoem hominem
venientem in hunc mundum illumina conscientias conlis nostri igne tuae caritatis. ut tuo igne igniti et
illUOlinati expulsis a conlibus nostris peccatorum tenebris ad vitam te illustrante pervenire mereamur
aetemam. Per.
B S(c). Domine Deus Noster Omnipotens, lumen indeficiens, conditor hominum, exaudi nos famulos
tuos et bene + die hunc ignem, qui tua sancti + ficatione et bene + dictione consecratus est. Tu qui illumi-
nas omoem hominem venientem in hunc mundum, illumina tenebras conlis nostri, et conscientias nostras
igne tuae caritatis, ut tuo lumine illuminati, expulsis a conlibus nostris peccatorum tenebris, ad vitam, te
illustrante, pervenire mereamur. Per.
B S(d). Domine Deus Noster Omnipotens, Lumen indeficiens, conditor omnium luminum, exaudi nos
famulos tuos et benedic hunc novum ignem qui tua sanctificatione consecretur. Tu illuminas omnem
hominem venientem in hunc mundum, illlumina quaesumus conscientias cordis nostri igne tuae caritatis
ut tuo igne igniti tuo lumine illuminati expulsis a conlibus nostris peccatorum tenebris ad vitam te illus-
trante pervenire mereamur aetemam.
B S(e). Domine Deus noster audi nos pater omnipotens lumen indeficiens qui es sanctorum splendor
luminum : et bene + die domine hoc lumen quod a te conditum est : qui illuminas omnem hominem
veoientem in hunc mundum : ut a te vero lumine accendamur : et illuminemur igne claritatis tuae : illu-
mina etiam corda et sensus nostros ut ad vitam aetemam pervenire mereamur. Per.
B 6. Domine Deus noster Pater omnipotens qui es lumen indefiens et conditor omnium luminum, benedic
et sanctifica hunc ignem : ut per te sancificatum et benedictum, qui illuminasti omnem mundum, ab eo
accendamur atque illuminemur igne claritatis tuae, quo illuminasti Moysen famulum tuum : ut, illumina-
tis cordibus et sensibus nostris, ad vitam aetemam pervenire mereamur. Per.
B 7(a). Dominus Deus omnipotens, lumen indeficiens et conditor omnis luminis, exaudi nos indignos
famulos tuos, et benedicere huius novi ignis lumen, quod a te vero lumine nobis donatum est, ut et tui
am oris accendamur illustratione, et verae caritatis illuminemur igne : et sicut Moysen famulum tuum
mirabili illuminasti splendore, ita corda et sensus nostros illuminare digneris; ut peccatorum labe expiati,
ad te, qui aetema vita es, pervenire mereamur. Qui cum Deo Patre et Spirito Sancto vivis.
B 7(b) Dominus Deus omnipotens, lumen indeficiens et conditor omnium luminum, exaudi nos famulos
tuos, et benedicere digneris huius novi ignis lumen, quod a te vero lumine nobis donatum est, ut et tui
am oris accendamur splendore, ita corda et sensus nostros illuminare digneris, ut expiati peccatorum labe,
ad te qui vita aetema es pervenire mereamur. Per.
B 8. Dominus sancte pater, omnipotens aetemus deus, exaudi nos lumen indeficiens. Tu es enim,
domine deus noster, conditor omnium luminum. Benedic, domine, et hoc lumen, et bane caeram, quod a
te incensum, sanctificatumque ac benedictum est. To qui inluminasti omoem mundum, ab eo lumine
accendamur, et illuminemur igne claritatis tuae. To es ignis qui famulo tuo moysi in rubo apparuisti. Tu
es columna ignis qui populum israhel in nocte defendebas et inluminabas. Tu enim tres pueros de fomaci
ignis liberasti, domine, cum filio tuo ihesu christo et sancto spiritu tuo, qui in igne super apostolos singu-
los die pentacosten, et post tempus super Comelium, cum omni domo suo, tibi primum ex gentibus cre-
dentem de celo descendisti, ut sicut eos omnes conseruasti et inluminasti, ita sensus nostros cordaque et
anintas nostras in hac paschali sollempnitate et omni vitae nostrae tempore inluminare igne spiritus sancti
digneris, ut ad vitam aetemam pervenire mereamur in caelis. Per.
349
B 9. Domine Deus, Pater Omnipotens, lumen indeficiens, qui es conditor omnium luminum, qui illumi-
nasti omnem mundum, benedic lumen hoc, quod a te sanctificatwn est atque benedictum, ut ab eo accen-
damur et illuminemur lumine claritatis tuae et sicut illuminasti cor Moysi ad rubum ardentem et non
comburentem, quando eum misisti in Egyptum ad Pharaonem et cwn ad te ascendisset in montem Synai,
ita illuminare cordibus nostris et sensibus, per te qui via, veritas et vita es, ad vitam permanentem et
lucem aetemam pervenire mereamur. Per.
B 10. Dominus Deus noster pater omnipotens lumen indeficiens :exaudi nos qui es sancte conditor
omnium luminum et benedicere digneris hoc lumen quod a te sanctificatum est atque benedictum et sicut
illwninasti omnem mundum tuo lumine ut ab eo accendamur et illuminemur lumine claritatis tuae : ita
illnmina corda nostTa et sensus nostros et ideo deprecamur te domine deus noster : ut sicut illuminasti
ignem Moysi famulo tuo per columnam spiritualem ambulantem im mari rubro ita intelligentias nostras
illustra : ut ad vitam aetemam pervenire mereamur et candela quae de hoc lumine fuerit accensa in honore
maiestatis tuae : semper perseveret benedicta et quicumque acceperit de eodem lumine sit illuminatus
lumine spiritualis gratiae. Per.
B 11. Domine Deus noster, Pater omnipotens, qui lumen indeficiens omniumque creator es luminum,
hone ignem bene + dicere et sanctificare digneris, ut per Ftlium tuum, qui lux vera est mundum illumi-
nans universum, ab isto, sicut Moyses, accendamur tuaeque claritatis igne sic iugiter illustremur, ut tam-
quam lucis filii ad vitam perveniamus aetemam. Per.
C. Domine sancte, Pater omnipotens, aeteme Deus : benedicentibus nobis hone ignem in nomine tuo, et
unigeniti Filii tui Dei ac Domini nostri Jesu Christi, et Spiritus Sancti, co-operari digneris; et adiuva nos
contra igniti tela inimici, et illustra gratia coelesti : Qui vivis et regnas cum eodem Unigenito tuo, et
Spiritu sancto, Deus : per omnia saecula saeculorwn.
C 1. Domine sancte, Pater omnipotens, aeteme Deus : nobis qui hunc ignem in nomine tuo, et Filii tui,
Dei et Domini nostri Jesu Christi, et Spiritus Sancti, bene + dicimus, et sanctifi + camus, co-operari dig-
neris, et nos adiuvare, Qui vivis, et regnas cwn eodem Unigenito tuo, et Spiritu Sancto Deus.
D. Domine sancte, Pater omnipotens, aeteme Deus, benedicere et sanctificare digneris ignem istwn,
quem nos indigni per invocationem Unigeniti Filii tui Domini nostri Jesu Christi benedicere
praesumimus : tu clementissime, eum tua benedictione sanctifica, et ad profectwn humani generis prove-
nire.concede. Per eundem Dominum nostrum Jesum Christum Ftlium tuum, qui tecwn vivit et regnat in
unitate Spiritus Sancti Deus per omnia saecula saeculorum.
E. Domine sancte, Pater omnipotens, aeteme Deus, quia in nomine tuo et Filii tui Dei ac Domini nostri
Iesu Christi et Spiritus Sancti benedicimus et sanctificamus hone ignem, adiuva nos per eundem Domi-
num qui tecum vivit in unitate eiusdem Spiritus Sancti.
E 1. Domine sancte Pater omnipotens aeteme Deus in nomine tuo et Filii tui ac Domini nostri Iesi Christi
et Spiritus Sancti, benedicimus et sanctificamus lumen hoc. Adiuva nos. Qui vivit
F. Domine Deus, Pater omnipotens, conditor omnium rerum, te invocamus ut benedicas et consecres
hunc ignem sicut benedixisti rubum in quo apparuisti Moysi. et sicut illuminasti cor eius per visibile
lumen maiestate tua invisibili ita et corda nostra potencia divinitatis tuae invisibiliter per hunc visibilem
ignem illuminare digneris. Per.
G. Coelesti lumine quaesumus, domine, semper et ubique perveni, ut misterium cuius nos participes esse
voluisti, et puro cemamus intuitu et digno percipiamus affectu. Per.
H. Domine sancte, Pater omnipotens, aeterne Deus, in nomine tuo et filii tui domini nostri Iesu Christi et
spiritus sancti, benedicimus hone ignem et eum cwn cera, et omnibus eius alimoniis sanctificamus et
signo crucis Christi Iesu filii tui altissimi signamus, ut intus vel foris, non quod nocet incendat, sed omnia
ad usus hominum necessaria calefaciat sive illuminet et quae ex hoc igne fuerint conflata vel calefacta,
sint benedicta et omni humanae saluti utilia, ut non, cum Nadab et Abiu ignem tibi offerentibus alienum,
incendamur, sed cum Aaron pontifice et filiis eius Eleazaro et lthamaro, hostias tibi pacificas, sancti spiri-
tus igne assatas immolare valeamus et semper eiusdem spiritus sancti igne vitia nostra ure, confaque luce
scientiae tuae illurnina et animas nostras fidei calore clarifica Per.
350
I. Benedictio Dei + Patris omnipotens et Fi+lii et Spiritus + Sancti descendat et maneat super hoc lumen
et incensum istud.
J. Omnipotens sempiteme, aeteme Deus Creator omnium rerum, te bumiliter deprecamur, ut bunc ignem
novum, caelo terrae largitum, sanctificare et benedicere digneris, et sicut in adventu too elementa
contagione peccatorum polluta purgare polliceris; ita et hie ignis novus tabemaculis fidelium tuorum
babendus purgationem perfectam obtineat, et obstaculis invictus contra omnem nequitiam maligni spiritus
perficiatur.
K. Rogamus te domine deus omnipotens ut digneris benedicere bunc ignem ne impediat domum bane in
qua accenditur. Per.
L. Domine sancte, Pater omnipotens, etemus Deus, beoedicere et sanctificare digneris ignem istum qui
nos indigni suscipimus per invocationem bunigeniti filii tni domini nostri Iesu Christi quem bodie in
templo presentatum, iustum Simeonem diu expectantem in ulnas suscepisse novimus et salutare tuum
ante faciem omnium populorum, esse lumen scilicet gentibus et gloriam plebi tue Israel pro propbetico
spiritu docuit, te quaesumus domioe ut benedicere digoeris lumen istud et omnibus illud manibus
gestantibus verum lumen tue majestatis concede, ut ad te cognoscentes per viam virtutum ad te valeamus
pervenire, qui in trinitate perfecta unus et gloriaris deus per omnia saec. saeculorum.
M. Adesto, quaesumus Domine, supplicantibus, tuaque praeferentia cousecrare ac benedicere buius

incrementa ignis dignare, ut omnibus in te sperantibus fugato daemonico phantasmate tribuas lumen et
colorem spiritus tni saluberrimum.
N. Domine Deus noster, lumen indeficiens, benedicere, + sanctificare, + et cousecrare digneris bane ignis
creaturam: ut eo in tni honorem utentes, expulsis a cordibus suis peccatorum tenebris, ad vitam, te
illustrante, pervenire mereantur aetemam. Per.
0. Domine sancte Pater omnipotens aetemus Deus, lux et splendor universarum creaturarum, qui ex
nihilo cuncta producere dignatus es, te humiliter deprecamur : ut bane creaturam ignis benedicere et
sanctificare digneris in nomine dilectissimi Filii tni, et in virtute Spiritus sancti: ut nobis ad obsequium et
ad lumen esse facias tam animae quam corporis, et ut nibil in eo nobis adversarius damni irrogari possit;
sed sit nobis in adiutorium per virtutem et potestatem tuam, salvator mundi. Qui vivis et regnas.
P. Deus qui Moisi famulo tuo in specie ignis, rubo ardente, apparuisti: quique etiam Sanctum Spiritum
tuum Dominum nostrum de caelo promissum super apostolos igneis linguis descendere fecisti : nobis
quoque famulis tuis caritatis ignem tribuere dignare; et hunc novum ignem de lapide prosilitum nobis in
usum profuturum sanctifica : et concede nobis ita per haec festa paschalia coelestibus desideriis
inflammari, ut ad perpetua festa purgatis mentibus pertingere valeamus. Per.
Q. Oramus te, domioe deus noster, ut caereus iste in honorem nominis tni cousecratus ad noctis huius
caligioem destruendam indeficiens perseveret. In odorem suavitatis acceptus supernis luminaribus
misceatur. Flammas eius lucifer matutinus inveniat, me, inquam, lucifer qui nescit occasum, llle qui
regressus ab inferis humano generi serenus ioluxit
R. Domine Deus, Pater Omnipotens, lux vera et fons omnium; benedic + et sanctifica Cereum istum, ut
ab eius lumine accendamur atque illuminemur igne claritatis tuae : et sicut illuminasti Moisen exeuntem
de Aegypto, ita illumines corda et sensus nostros, ut ad vitam et lucem aetemam pervenire
mereamur. Per.
S. Domine Omnipotens, Eteme Deus, beoedicentibus nobis hunc ignem novum productum e silice :
nostris perfecturum usibus quem nos indigni invocatione unigeniti filii sui tni domini nostri Iesu Christi
benedicere persumimus; tu clementissime deus cum tua benedictione sanctifica ad perfectum bumani
generis pervenire concede ut nos celestibus desideriis inflammati ad festa ventura purgatis mentibus
pertingere valeamus. Per.
T. Domine Jesu Christe qui filios Israel egyptiaca servitute depressos: moyse famulo tuo ductore per
colurnnam nobis in die ad terram repromissionis perduxisti benedic hone ignem benedictiooe tua
coelesti : ut per tni amoris ignem spiritualiter mentibus nostris infusum conburat in nobis omne viciorum
incendium : ut tua ineffabili luce illuminati : ad illius esum perducamur agni qui passionis suae vulnere
potens est totius mundi peccata delere. Qui tecum vivit...
351
U. Omnipotens sempiteme deus mundi conditor luminis siderumque fabricator : per cui us ineffabilem
potentiam omnis claritas sumpsit exordium : te in tuis opibus invocamus : aperi nobis quaesumus labia
nostra ad confitendum nomine tuo et ad laudem g]oriae tuae ut dignere celebrare mereamus sacrum
officium : qui in hac sacratissima nocte vigilia de donis tuis cereum tuae suppliciter offerimus maiestati.
V. Veniat. quaesumus, omnipotens Deus, super hunc /hoc/ incensum larga tuae benedictionis infusio: et
hunc noctumum splendorem invisibilis regenerator accende; ut non solum sacrificium, quod hac nocte
litatum est, arcana luminis tui adrnixtione refulgeat; sed in quocumque loco ex huius sanctificationis
mysterio aliquid fuerit deportatum, expulsa diabolicae fraudis nequitia. virtus tuae maiestatis
assistat Per.
W. Domine Sancte Pater Omnipotens Aeteme Deus, lumen quod in nomine tuo et filii tui dei ac domini
nostri iesu christi et spiritus sancti benedicimus et sanctificamus, quaesumus ut a te beoedictum sit et
sanctificatum, eoque utentes exterius, interius spiritualiter calefieri mereamur. Per.
* * * * *
Formulas for the blessing of the incense
a. Domine Deus noster, qui suscepisti munera Abel, Noe et Abraham, sacrificium Aaron, Samuelis, et
Zachariae, et omnium sanctorum tuorum, incensum istud bene+dicere, et de manibus nostris in odorem
suavitatis recipere digneris; ut ornnes gestantes, tangentes, et adorantes illud, virtutem et auxilium Spiri-
tus sancti percipere mereantur.
b. Deus omnipotens, Deus Abraham, Deus Israel, Deus Jacob, immitte in bane creaturam incensi vim
odoris tui vel virtutem, ut sit servulis tuis vel ancillis munimentum tutelaque defensionis, ne intret hostem
in viscera eorum, aditumque et sedem habere non possit.
352
APPENDIX 6 - The Sepulchre'
The placing of a light by the Sepulchre is first attested in the eleventh century by John
of Avranches, at Rouen, in Lanfranc's Decrees, and in the Customaries of Sigibert and
Fruttuaria? The practice may be considerably older,3 although the reposition of the
Sanctissimum after the Mass of Maundy Thursday is attested no earlier than the tenth
century.4 Two traditions relating to the presence of lights at the Sepulchre may be
identified : one, in which the light or lights were extinguished on Good Friday, and the
other, in which the Sepulchre remained illuminated until the conclusion of the Vigil on
Holy Saturday or until Easter Day.
( 1) According to the tradition recorded by the ftve above-mentioned documents, a

light burned continuousll from the reposition of the Sanctissimum until the conclusion
of Lauds on Good Friday. This practice survived at Notre-Dame in Rouen until at least
the beginning of the eighteenth century .6 Subsequently it became customary to relight
the candle with the new frre on Good Friday;7 and eventually to allow the candle to
burn continuously from the reposition on Maundy Thursday to the consuming of the
Sacrament on Good Friday. 8
A number of churches attest the use of two or more candles during the presence of
the Sanctissimum in the Sepulchre.9 With the elaboration of ceremonial following what
Jounel called the 'eucharistic triumphs' ofthe Counter-Reformation the adornment of
the Sepulchre with both lights and other decorations became what might today be con-
1
Usually known today as the 'altar of repose' or 'place ofreposition'. Here the Host,
consecrated on Maundy Thursday, is reserved for use at Communion (the Mass of the
Pre-sanctified) on Good Friday. ,
2
Lib.de Off.Eccl., PL 147.50B; Acta Vetera, PL 147.173C; PL 150.460A; Albers ll
f.·93; Albers IV p.58.
The practice of honouring the dead with lights has a venerable antiquity. For pagan
practice see, for example, Dennis ll p.388, Marwick p.93, Briggs p.72, and Hare, Days
p.248. For its use in Christian worship see, for example, Canziani p.105, Lees p.168,
and especially Dendy pp.99-107.
4
Regularis Concordia, PL 137.495.
5
Acta Vetera and Lanfranc (ibidem): lumen continue ardeat.
6
De Moleon pp.300-301.
7
As at Laon (DAMR 3.15.7 p.142, M 1164) and St-Martin d'Ainay, Lyon (DAER 4.22
p.125, M 175).
"This occurred at Nidaros (ONE p.226), Lund (1514 Missal fol.lxxxiiii), Magdeburg
p503 Missal foi.lxxxvii), and Cambrai (1507 Missal fol.lxvi).
2 candles: Canterbury (l-IBS 23 p.380), Coutances (1825 Ceremonial p.315), Liege
(1492 Ordinary np), and Rome (Ordo XV, PL 78.1306). 4 candles :Barking (HBS 65
p.93). At Lyon, Mende, Salzburg, and Vienna four candles were held temporarily dur-
ing the reposition of the Corpus Christi. An unspecified number burned at Burgos
(1546 Missal fol.xciii), Cosenza (1549 Missal fo1.104), Hilde1sheim (1499 Missal
fol.xcvi), Poitiers ( 1767 Missal p.240), and in the Cistercian lite ( 1669 Missal p.139).
353
sidered excessive. 1 This emphasis on magnificence was positively encouraged by the
Caeremoniale Episcoporum in its recommendations for the preparation of the chapel of
repose. 2 Surrounding the Pauline Chapel of StPeter's in Rome in the last century
burned nearly 600 candles. The diarist Samuel Rogers commented that the arrangement
was 'elegant in the highest degree' .3 Following the liturgical changes of 1955 the
'traditional simplicity of the Roman liturgy' was restored. It was enjoined that the Sep-
ulchre should be moderately adorned, 4 and that a single light should burn at the con-
clusion of the Adoration of the Sanctissimum at midnight on Maundy Thursday.5
(2) According to the alternative tradition the Sepulchre was honoured with light
until the conclusion of the Vigil on Holy Saturday. It is possible that in some churches
the Sepulchre may also have served as a place of reservation for the older unconsumed
Hosts. However, it had become customary in places by the thirteenth century to place a
cross in the same Sepulchre as the Host.6 Its presence in the Sepulchre invited and
resulted in its being honoured by light also. At Hereford one candle7 and at York two
candles8 burned from Maundy Thursday to Easter Day. Whereas at Salisbury one of
the two candles was extinguished after the removal of the Sanctissimum; the other
remained alight until the procession on Easter Day before Matins.9 This procession is
also attested in Hungary. 10
1
At Palencia in the sixteenth century there were 'very many candles'. 1568 Missal
fol.lxxxviiii.
2
Quo pulchrius magnificentiusque poterit, muftis luminibus ornatum (11.23.2).
3
Hale p.274.
4
Pastoral Instruction 11.8; Jounel, Le nouvel ordo p.29.
5
6
At York (Missal p.107) and Durham (Raine pp.l0-11).
7
HBS 26 p.324.
8
Misssal p.107.
~ssal (Dickinson) p.337.
10
1815 Ritual pp.440-41.
354
APPENDIX 7 - Illumination on Good Friday at the Passion and at the
Adoration of the Cross
Two traditions relating to the use of light at the Good Friday liturgy existed within the
rites of the Western Church. The use of liturgical light when the Passion was read and
the Cross was venerated in the Gallican Church is attested by documentary evidence, 1
and can be inferred with confidence for the Mozarabic rite. 2 Moreover, Amalarius'
statement concerning the practice of the Roman Church implies that he was familiar
with the use of light at the Good Friday liturgy. Both at Salzburg3 and in the Camaldo-
lese rite4 two altar candles burned during the reading of StJohn's Passion.
The absence of illumination during the Good Friday liturgy of the Roman rite is
attested in the ninth century by Amalarius, who wrote that in the Roman Church on
Good Friday all fire was extinguished from the sixth to the ninth hour in commemor-
ation of the Cruciftxion.5 In its liturgical application this entailed extinguishing the altar
lights before the reading of StJohn's Passion and rekindling them after the Adoration of
the Cross. 6 Roman practice was generally followed throughout the western rites includ-
ing those of Braga and Lyon. 7 In the Ambrosian rite, however, the altar lights and the
candles of the acolytes were extinguished at emisit spiritum during St Matthew's
Passion, and were not rekindled until after the Easter candle had been lit the following
day. 8
1
PRG II p.86 §304.
2
Leon Antiphonary pp.275 ff.
3
1507 Missal fol.lxxxvii.
4
1503 Missal fo1.82.
5
Liber Officialis 4.22.2.
6
Caeremoniale Episcoporum p.278 and p.284.
7
1558 Missal fol.xciii and 1771 Missal p.178, respectively.
8
King, Holy Week p.98.
355
APPENDIX 8 - The Paschal Column and Candelabrum
Many of the surviving Paschal candlesticks in the churches of Rome are twisted shafts
of verd-antique with bases of gilt bronze or white marble. Amongst the most impres-
sive are the voluted columnar candlesticks of S.Clemente and S.Lorenzo, both inlaid
with mosaics and bearing a spike on their summits to receive the Easter candle, that of
S.Maria Cosmedin with its cosmatesque craftmanship, and the twelfth-century column
in St Paul's, embellished with a series of bassi-relievi depicting the Passion and the
Resurrection. Other examples of Paschal pillars with well-executed sculpture are to be
found in the monastic church at Farfa, in the Capella Palatina of Palermo Cathedral, in
the Church of SS.Nereus and Achilleus, and in the Baptistery at Florence.
The practice of using a pillar was not confined to Italy. At Spires a choice existed
between a column and a candelabrum; 1 whilst at Angers in the eighteenth century a tall
column of marble stood in front of the altar.2 Likewise at Lyon a tall spiral column of
marble stands by the ambo. Indeed the rubrics of a number of French diocesan missals
recommend that the Paschal candelabrum should resemble a column. 3
The Paschal candlestick has been characterised by two attributes, size and beauty.
That at Barking in the fourteenth century was described as large,4 as was that at
Besan~on in the seventeenth century; whilst the Easter candle at Autun stood in a
5
candelabrum that was both grandius and colurnnar.6 It need hardly be stated that the
candlestick should be as tall, if not taller, than the Easter candle. Likewise, it is only
fitting that the importance and significance of the Candle should be matched by a holder
of noble craftmanship. It was even recommended that the candelabrum of a country
church should be pulchre elaboratum.1 We are also informed that the candleholders
were artistically worked often in the form of an angel,8 even in smaller churches. 9 Most
of these appear to have been replaced by candlesticks of a much simpler design. 10 At
Ushaw College in England there are four angels at the base ofPugin's elaborately-exe-
1
Agenda fol.xciii.
2
De Moleon p.80.
3
Sees (1742) p.185; Mende (1766) p.l99; Perigueux (1782) p.158; Tours (1784)
p.191; Metz (1829) p.158; La Rochelle (1835) p.l86.
HBS65p.97.
5
6
1845 Missal p.238.
7
Desideri p.150.
8
A survival perhaps of the old Roman Vigil? See p.327 and p.330.
9
DHCR I p.470; Gattinari p.142.
10
Survey of 1984.
356
cuted eight-foot high candelabrum, the fmgers of the raised right hands pointing
upwards towards the light, a vertical movement which also characterises the postures of
the ceramic figures of pilgrims at St John's Cathedral in Portsmouth.
There is little evidence from previous centuries to make us believe that Paschal
candlesticks were generally fashioned from wood rather than metal; although it is likely
that the use of the latter often reflected the fmancial resources of a church. Of those in
use in thirty-five French cathedrals, eighteen were made of copper or bronze, four were
of iron, and the rest were made of wood. 1
At Durham the central branch of the enormous seven-branched candlestick was

used to hold the Easter candle. 2 According to a description of this candelabrum
recorded in 1593, many years after its destruction, its feet were flying dragons; there
were representations of the four evangelists; and it was studded with precious stones
and embellished with intricately-wrought metalwork. 'The Paschall in latitude did con-
taine almost the breadth of the Quire, in longitude that did extend to the height of the
lower vault, wherein did stand a long piece of wood reaching within a man's height to
the uppermost vault roofe of the church. ' 3
1
Survey of 1984.
~ese candelabra, made in imitation of the menorah in Solomon's Temple, became
common in Western Europe as a result of the Crusades. Although there were a number
of them in use in the cathedrals of England, the only surviving examples are to be found
in Central Europe- for instance, at Essen, Brunswick, ~sterneuburg, and Prague (Zar-
necki pp.134-5). They are still in use today at Arhus, Lund, and Ribe.
3
Raine, Rites of Durham p.9. It is generally believed that the 'long piece of wood'
refers to a Judas which was inserted into the central holder to increase the size of the
Easter candle. It is more likely, however, that this length of wood was a pole with a
small candle at one end, used for lighting the Easter candle. If there was a hole in the
roof of the cathedral - the description was based on the memory of an old man - it is
unlikely to have been used for lighting the Candle. The writer believes that the pres-
ence of an aperture in the roof may have been linked to the use of a block and tackle, or
similar device, which may well have been necessary for hoisting the large Easter candle
into its holder.
357
APPENDIX 9 - The Blessing of the Font and Baptism
The use of light at the blessing of the font is ftrst attested in the Roman rite c .AD 600. 1
Ordo 11 informs us that the two man-sized Vigil-candles were taken in procession to
the font in front of the Pope? Two later ordine? state that the two candles were low-
ered into the baptismal water when the Pope or presiding priest uttered the prayer Des-
cendat in hanc plenitudinem. The practice is also attested in two eighth-century
Gelasian sacramentaries,4 and in an eleventh-century baptismal ordo from Northern
Italy.5 It is also alluded to in Zachary's letter to Boniface.6 The presence of the two
large candles in the Ambrosian rite of the twelfth century is mentioned by Beroldus
(p.111); but there is no evidence that they were lowered into the water.
The immersion into the font of the Easter candle, which replaced the two Vigil-
candles of the Roman rite, in the churches of Gaul and Germany is frrst found in the
tenth-century Pontificale Romano-Germanicum; 7 and the practice became widespread
throughout the western rites. Alcuin records a transitional stage in which the choice is
allowed of immersing either the two Vigil-candles or the Easter candle. 8 By the end of
the fifteenth century it had also become customary for the priest to dip the Easter candle
three times into the font, withdrawing it twice and sinking it to a lower level each time,
and repeating in an ever-higher tone Descendat in hanc plenitudinem.9 In many
churches some of the molten wax of the Easter candle was allowed to drop onto the sur-
face of the baptismal water in the form of a cross. At Reims the choice was allowed of
using either the Easter candle or another candle. Presumably the latter was the cereus
fontium, and was used if the Easter candle was too large to carry in procession. 10
1
It is true that at the baptism of the Jews of Auvergne c.AD 500, Gregory of Tours
refers to the flickering of the candles and the burning of the lamps (PL 71.326). The
description, however, is too imprecise for us to comment with confidence upon the use
of light from a liturgical point of view.
2
0R 11.90.
3
0R 23.24ff and OR 30B.46.
4
~ p.99 and~ p.69.
5
Lambot p.xxxv.
TL 89.951B. See p.170 for the text ofthe letter.
7
PRG IT p.l04. Although the two Vigil-candles were present at the ceremony described
by PRG, there is some uncertainty as to whether they also were dipped into the water.
8
De Div.Ojf, PL 101.1219A/B.
~us a very large number of missals. The silence of a few missals (e.g. those of
Valence (1504) fol.lxiiii and Narbonne (1528) fol.xcv) on this score would suggest that
a single immersion obtained in a number of churches.
10
1770 Missal p.224.
358
The Sacramentary of Ripoll enjoins that the candles be placed in the water (p.97).
It is not clear whether cer[e]os refers to the two Vigil-candles or to the candles of the
baptizands. Support for the latter type comes from the Pontifical of Poitiers, which
quite clearly states that the unlit candles of the children about to receive Baptism are
placed in the water. 1
At Salisbury the Easter candle, which was present at the font, was not lowered into
the water. Instead drops of wax dripped onto the surface of the fontal water.Z The
silence of Beroldus would suggest that at Milan in the twelfth century the Easter candle
was not carried to the font. Its absence on this occasion is also attested at Biasca
c.AD 900,3 and at Nidaros in the early thirteenth century. 4
The earliest reference in the West to the baptismal candles held by the neophytes
occurs in a work formerly attributed to St Ambrose.5 Martene cites the letter of Marcus
of Gaza to Arcadius on the occasion of the baptism of Theodosius the Y ounger;6 and
we had occasion above to refer to the conversion of the Jews of Auvergne.
In the period from c .AD 500 to 1000 the custom of handing an unlit candle to each
baptizand is recorded only three times. In the ninth century Amalarius states that the
neophytes' candles were lit after the last litany;7 and Alcuin records that they were lit
after the precentor had proclaimed 'Accendite' .8 The plunging of these candles into the
font at Poitiers has been referred to above. At what point during the service or during
their period of preparation the catechumens received their candles is unknown. Accord-
ing to the Northern Italian baptismal ordo, they brought their candles to the scrutinies.9
Attestation of the use of a formula, which subsequently became widespread

throughout the western rites, to accompany the presentation of a candle first occurs in
the Missal of Robert of Jumieges. 10 It reads : 'Receive the irreproachable candle; guard
your baptism; so that when the Lord comes to the wedding-feast, you may meet him in
the court of heaven for ever and ever.' The twelfth-century Ritual of Soissons and the
thirteenth-century Sa1um Missal contain slight variations. The former
1
Deponunt in fontem cereos baptizandorum infantum non illuminatos (p.216).
2
Missal of c.1300 (Legg) p.l29.
3
Sacramentary p.70.
4
0NE p.234.
5
De Lapsu Virginis, PL 16.372.
6
De Antiquis Ecclesiae Ritibus Vol.1 p.54.
1
Liber de Ord.Ant. XLIV.8.
8
De Div.Off., PL 101.1221C.
9
Lambot 12, 14, 29.
'OWilson, The Missal pp.99-100.
359
begins: 'Receive the irreproachable, burning candle ... '; 1 whilst the latter
starts : 'Receive the irreproachable candle; guard your baptism; keep the command-
ments ... '.2
1
DAER 4.24 p.161, M 305.
2
Missal of c.1300 (Legg) p.131.
360
APPENDIX 10 - Anticipation
Reference has already been made to the anticipation of the new fire ceremony and the
Easter vigil in the afternoon of Holy Saturday as early as the eighth century. 1 The sing-
ing of Tenebrae at Rome in the fourteenth century in the late afternoon of Wednesday,
Thursday, and Friday of Holy Week would suggest that the blessing of the Easter candle
took place much earlier in the day. 2 Evidence from the late-fifteenth and si.'{teenth cen-
turies for the kindling of the new fire at noon has been presented in Table 27, and for
the performance of the same ceremony at about 9.00 a.m. in Tables 24 and 26.
However, we also observed that in the seventeenth century some Benedictine

houses continued to sing Tenebrae in the early hours of the morning, a practice that the
Cistercians maintained until the time ofthe Second Vatican Council. 3 As late as the six-
teenth century at Braga the liturgy of Holy Saturday continued to be performed at
night. 4 Jounel suggests that the anticipation of the services of the Triduum was not
universal in France until after 1600; but he does not instance any church where antici-
pation did not occur.5
In the cathedrals of the Catholic areas of Eastern Europe the liturgy of Holy Satur-
day has always been celebrated during the late evening of that day. Patrick Leigh Fer-
mor graphically describes the ceremony at Esztergom in 1933.6 The writer is given to
believe by eye-witnesses that a similar situation obtained in the larger churches of
Poland.
•seep
2
0rdo XIV, PL 78.1204B.
3
The writer is grateful to the Rt Rev. John Moakler, Abbot of Mount St Bernard Abbey,
Coalville for this information. Letter dated 21 June 1984.
4
King, LPS p.223.
5
La Semaine Sainte p.146.
6
Between The Woods pp.15-16.
361
APPENDIX 11 - Monte Cassino A
We hope to show that the first of the two twelfth-century rites of Monte Cassino,
referred to here as Monte Cassino A, 1belongs to the same tradition as that recorded in
the late eighth-century Ordo 28 in respect of the ceremonies of Holy Saturday. For both
attest a procession which accompanies the lighted Easter candle. The evidence of
Monte Cassino A is, however, ambiguous; and it is not immediately clear from the rub-
rics how the Easter candle featured in the initial stages of the ceremony. Much of the
difficulty revolves around the interpretation of cereus in the following passage from that
document :
Et accenso cereo, procedunt omnes de secretario cum ipso cereo in aeclesiam cum
silentio et, posito in candelabro cereo, portante acolito, procedunt ad altare.
AcolJ!us vero portat cereum ad arnmonem et dicit tribus vicibus : Lumen Christi,
plane?
From PRG and Monte Cassino B we know that cereus may either refer (a) to the
small candle used for bringing the new fire into church, or (b) to the Easter Candle, or
(c) to one of the two man-sized candles of the Roman Vigil. In view of the mention of
only one candle, we may at once discount the third possibility. However, in view of the
absence of any article in Latin, definite or indefmite,the first occurrence of cereus in the
above passage could equally be (a) or (b). Likewise, the second occurence of the word
presents a choice of interpretation. The identification of cereo, which is the substantival
element in the Ablative Absolute, depends on the interpretation of posito. This may be
translated either (i) 'the acolyte who bears the candle places it in a candlestick', or (ii)
'an acolyte bears the candle (ie the Easter candle) which has (previously) been placed in
a candlestick', since it is not clear who is carrying the lighted candle (ie the ftrst cereo).
Then again, the candle which the acolyte takes to the ambo may be a small lighted
candle, if the Easter candle, previously unmentioned, was already in position by the
ambo. The situation becomes even more complex, if a distinction is then made between
the fust and the second occurrence of cereo. For the former would refer to the small
candle used to bring in the new fue, and the second to the Easter candle, especially as
ipso seems to give some emphasis to the noun which it qualifies.
1
Printed as §1 on p.292 of PR XII in M.Andrieu, Le Pontifical Romain au Moyen Age.
The contemporary Monte Cassino B (p.293 §2) is to be identified with the twelfth-cen-
tury Ordinary of Monte Cassino, known to Martene (= M 1139). The exact milieu of
Monte Cassino A is unknown; but its comparison with Monte Cassino B by the
compiler of the latter, and the presence of the Lumen Christi, would suggest a central
Italian location not far from Monte Cassino.
2
PR XII I p.292 § 1. The writer's italics.
362
Most of the difficulties disappear, however, when we compare the rubrics of Ordo
28, supplemented by those of Ordo 31, with those contained in Monte Cassino A.
(Table 60).
Ordo 28 (§§58-63) Ordo 31 (§63-67) Monte Cassino A

1. Robing at 9th hour Robing at 9th hour
2. Candle is lit Candle to be blessed is Candle is lit
lit in the
3. Procession from sac- sacristy. Procession from sac-
risty risty
4. Silent procession Silent procession
5. Candle in candle- Candle to be blessed in Candle in candlestick
stick candlestick
6. at altar at altar at altar
7. (Lessons later read at Candle is taken to ambo
ambo) (Cry of Lumen Christi)
8. Deacons asks for (Archdeacon sings) Deacon asks for bless-
blessing ing
9. Exultet Exultet Exultet
10. Two candles lit Two man-sized candles Two candles lit
lit
11. New fire taken to New trre taken to every
every home home
Table60
The similarities between Ordo 28 and Monte Cassino A leave us in little doubt that
the latter perpetuates the ceremonial of Ordo 28 in its entirety and has added to the rit-
ual only the cry of Lumen Christi. It is remarkable that so little change had occurred in
the liturgy of this tradition over nearly 400 years. Ordo 31 clearly belongs to the same
tradition in spite of the smaller number of rubrics. Its importance, however, for our
present argument lies in the fact that, in enlarging on the identity of the candle in Rub-
rics 2 and 5, it enables us to interpret cereus correctly in the above-quoted passage from
Monte Cassino A, and to show conclusively that the Easter candle was lit in the
sacristy and borne in procession into church, trrst to the altar and then to the ambo. The
passage may now be translated :
363
After the Easter candle has been lit, they all proceed from the sacristy with the
Candle into church in silence. The acolyte who bears the Candle places it in a
candlestick, and they move to the altar. Then the acolyte takes the Candle to the
ambo and proclaims three times: Lumen Christi.
364
APPENDIX 12 - The Inventor rutili of Prudentius
In view of the relatively late attestation 1 for its earliest use on Holy Saturday, it would
seem that this hymn, which was originally composed for use at the Lucemarium, was
recommissioned for use at the new fire ceremony, when that ritual was incorporated
into the Paschal liturgy of the Gallican Church, probably in the eighth century;2 and it
is not difficult to see why the hymn came to be sung on Holy Saturday, asstLming that
its use had generally fallen into desuetude with the disappearance of the daily office of
the Lucernarium. 3
The principal themes of the hymn are God's provision of light which brings salva-
tion, and deliverance from the darkness of this world's evil. The tone of the hymn is
both eschatological and strongly expectational with references to the Deliverance in
Exodus, being reminiscent of the themes of the Romano-Gallican Paschal Preface.
After an initial invocation to God as the source of light (1.1 ), the petition for the provi-
sion of light for the faithful, which looks forward to the return of that light after its dis-
appearance, suggests an approaching conclusion to a commemoration of the Crucifndon
(1.4 ). References to the Passover (11.37 -8), Christ's descent to the underworld (1.127),
the keeping of a watch (1.137), and the Resurrection (1.132) strongly suggest the Paschal
vigil of Holy Saturday. This is reinforced by a reference to the production of fire from a
flint (1.7). It is not difficult to understand why the hymn was once thought to have been
composed for the Vigil of Holy Saturday.4
We have referred elsewhere to the highly poetic language ofPrudentius.5 This

being taken into account, it would seem that the content of the Inventor rutili related to
the primitive vigil of Saturday which was held in preparation for the weekly commem-
oration of the Resurrection every Sunday. It is possible that the Lucernarium, being
originally a daily service, had developed into a weekly office held every Saturday
evening; and that Prudentius' hymn was written for this weekly celebration.6
1
Tenth century. See Chapter 4 (ii) and Table 45 on p.211.
2
0rdo 26 is our earliest evidence for that ceremony.
3
Although the hymn was written by Prudentius, who was himself a Spaniard, there is no
evidence for its use on Holy Saturday in the Mozarabic rite. In Chapter 4 (ii) we have
shown that its use was conf"med almost exclusively to the regions east and north of the
Alps.
4
Mabillon IT p.141.
5
Part IV Chapter 6 p.254.
6
A weeki y performance of the Lucernarium is favoured by Capelle, Le rite des c inq
grains p.3.
365
APPENDIX 13 - The Mozarabic and Milanese Rites
1. The Mozarabic Rite. 1
In a comparative study of the liturgies of Holy Saturday the development of the

ceremony of light from the Lucernarium is more apparent in the Mozarabic rite than in
any other western rite. For in addition to the lighting and bearing in procession of the
Easter candle, the lamp, which had formerly been the principal source of light at the
Lucernarium, continued to be accorded similar veneration. The influence of the
Jerusalem liturgy of Holy Saturday was also obvious in the Spanish ritual. There is
general agreement amongst scholars that the production of ftre by the bishop in a sac-
risty from which all light had been excluded, and the procession of light into the church,
to be followed by the Vigil, was derived directly from Jerusalem.
Not all the precursors of the Mozarabic rite, however, are to be found in the East,
as Capelle was forced to admit. 2 In Chapter 10 of Part N we have attempted to account
for what Dendy saw as a superfluous second blessing of both the lamp and the Easter
candle. The presence, however, of both the lamp and the Easter candle, and the blessing
of both, are clear indications of the composite character of the Spanish rite. The writer
believes that the three principal features of the ritual which do not have their prov-
enance in Jerusalem, viz. the Easter candle, the striking of the ftre, and the triple Deo
gratias were elements imported into Spain from Gau1, possibly as early as the sixth
century. Elsewhere we have shown that the Easter candle had its origin in Northern
Italy (pp.228-30), the markings on the Candle in Rome(pp.277-8), and the kindling of
ftre in Northern Europe (pp.183-6). The threefold cry of Deo gratias is almost cer-
tainly derived from the triple Lumen Christi, which we have argued had its origin in
Central or Southern Italy.
The influence of Jerusalem occurred in the three centuries or so which separate the
visit of Egeria to Jerusalem from the Moorish conquest of Spain at the beginning of the
eighth century. A date nearer the former event seems more likely. The Gallican influ-
ence may be assigned to a later date. We have suggested the sixth century, in view of
the evidence of Canon 93 of the Fourth Council of Toledo, which was held in AD 633 :
1
Uon Antiphonary pp.280-3 and Ferotin pp.210-15.
2
La Procession p.109.
3
PL 84.369B.
366
The Lamp and the Candle are not blessed in some churches on Easter Eve ... .It is
fitting for unity and peace that the [same] rite be observed in the churches of Gali-
cia.1
The fact that the lamp and the Easter candle did not feature in the same ceremony in all
parts of Spain by AD 633 would suggest that the combination of both these liturgical
light -sources within the same ceremony was not of great antiquity in that country.
The changes resulting from the introduction in 1500 of the M issale M ixtum as the
official Mass-book of the Mozarabic rite hardly affected the ceremonies involving the
new fire and the blessing of the Paschal light. The insertion into the Easter candle of
the five grains of incense and the introduction of the two additional candles, which were
lit from the Easter candle, hardly affected the character of the ceremonial.2
2. The Milanese Rite.
Like the Holy Saturday liturgy of the Mozarabic rite, the corresponding Ambro-
sian ceremonial of Milan had ritual elements inherited both from the Lucernarium and
from the liturgy of Jerusalem. Prominent amongst the former was the lighting of a lamp
from the new f"rre and the carrying of that lamp for the lighting of the Easter candle.
This lamp was also used to light the two large candles which burned during the blessing
of the font. 3
It is very likely that the use of two churches in the twelfth century for the new frre
ceremony derived from the practice at Jerusalem. For it is significant that the new fire
was kindled and blessed in the Church of the Holy Sel'ulchre in Milan, and subse-
quently taken to the 'summer' church, wherein the Paschal vigil took place.4 These
buildings corresponded to the Anastasis and the Martyrium in Jerusalem.5 In later
centuries the new frre was kindled in the sacristy of the cathedral, as in the Mozarabic
rite, the former arrangement being preserved to a point. The use of the sacristy is per-
petuated in the rite revised after Vatican II, although a location outside the church is
now permitted.6
1Dendy wrongly translates 'the churches of Gaul' (p.130).

2
PL 85.442C.
3
Beroldus pp.ll0-111; 1768 Missal p.125.
4
Beroldus pp.l 09-110.
5
For the Paschal vigil at Jerusalem, see Bertoniere pp.121ff.
6
1981 Missal p.242.
367
APPENDIX 14 - The Chartae.
The Charta at Reims Cathedral in 1585
Benedictus est hie Cereus in honorem & laudem Domini nostri JESU-CHRISTI, qui
cum Patre & Spiritu-sancto vivit & regnat Deus in saecula saeculorum Amen
Annus ab origine mundi juxta V.M.V.C.LIX Epacta lunae Paschalis

Hebraeorum supputa- semper quaerenda est inter
tionem est Octavum diem Martii &
Annus ab Incamatione M.V.C.LXXXV. quintum Aprilis
Domini Dies Paschae hoc anno inci-
Annus a passione eiusdem M.V.C.LII debat in 14. Aprilis, sed ne
Annus a Nativitatis B. Virgi- M.V.C.LXXXXIX Judaeos sequi videamur,
nis Mariae differtur in 2l.diem. Anno
Annus ab Assumptione ejus- M.V.C.XXXVI sequenti erit 6.Aprilis.
dem Annus Cycli solaris XXVI
Annus Calendarii Gregoriani IV Annus lndictionis XIII
refonnati Annus pontificatus Sanc- est XIII
Littera Dominicalis usque ad tissimi D. D. N.Papae
Circumcisionem est F. & Gregorii XIII
deinceps toto sequente Annus a conversione & bap- M.LXXXVI
annoerit E. tismo Clodovei primi ex
Aureus numerus est 9. anni Francorum regibus
vero sequentis est 10. Christianam fidem com-
Epacta eo ipso die cui plexi
respondet in Calendario, Annus aetatis Christianissimi XXXIV
novam lunam in·_singLLli.S nostri regis Henrici tertii
mensibus indicat Annus regni ejusdem XI
Annus Archiepiscopatus XI
Reverendissimi D. D.
Ludovici a Guisia
368
The Charta at Chalons Cathedral in 1708
Benedictus est hie Cereus in honorem Agni immaculati Domini nostri Jesu-Christi. BB.
Virginis Mariae, BB. Protomartiris Stephani, omniumque Sanctorum & Sanctarum.
Anno Periodi Julianae VI.M.CD.XXI. Anno a Constantinopoli per CC.UV.

Anno Aerae Cbristianae M.VD.VIII. Turcos expugnata
Anno reparatae salutis M.VI.LXXV. Anno ab Henrici magni C. :XV.
Anno post missos in GaUiam Camuti inauguratione ter-
viros Apostolicos & praes- tio Calendas Martii
ertim B. Memmium hujus Anno Pontificatus SS.Domini VIII.
Ecclesiae Cathalaunensis Clementis Papae Unde-
conditorem cirni
Anno post Attilam Hunno- M.II.LVI. Anno Regni Cbristianissimi LXV.
rum Regem a Meroveo Regis Nostri Ludovici
Francorum Rege & Aetio Magni
Romanorum duce prolli- Anno Ordinationis Dlus- XII.
gatum in Campis Catha- trissimi Domini D. Gasto-
launicis. nis Joannis Baptistae
Anno a Francorum Regno M.CC.LXXXVIII. Ludovici de Noailles
condito Episcopi & Comitis
Anno a translatione Regni v.cc.:xx. Cathalaunensis Franciae
Francorum a Caroli Magni Paris
stirpe ad Capetanos Anno ab extructo hujus Tern- LXXIV.
Anno a baptizato Clodoveo M.CC.XIV. pli propilaeo
Franciae Rege die Natali Anno a secundo bujus Templi XLI.
Domini496 incendio
Anno a vastatis prima vice V.LXX. Anno Cycli Solaris currente Vllll.
fortuito incendio hisce Anno Cycli decennovennalis XVIII.
sacris aedibus seu numeri aurei
Anno Dedicationis bujus V.LXI. Anno Cycli Indictionis I.
Templi ab Eugenio Ponti- Anno Cycli Epactarum vn.
fice Maximo 26 Octobris Littera Dominicalis G.
Anno Expugnationis Jero- V.:XX. Littera Martyrologii G.
solymorum a Saladino Concurrens vn.
Anno ab extinctione Ordinis CCC.LXXXXVD. Pasch a VIII.Aprilis.
Templariorum in Concilio Rogationes XIIII.Maji.
generali Viennensi 6. Pentecostes XXVD.Maji.
nonas Maji Dies Adventus II.Decembris.
Anno ab Elepbantiacorum N.XCVD. Dominica Septuagesimae XXVD.J anuarii.
crimine & nece Feria Cinerum XIII.Februarii.
Anno ab Urbe per Armenia- CC.LXXIX. Dominicae Post Pentecosten XXVI.
cos liberata
369
The Charta at Rouen Cathedral in 1678
Annus ab origine mundi 5678 Dominica I. quadragesimae 19 Februarii

Annus ab universali diluvio 4033 anni sequentis
Annus ab Incamatione Domini 1678 Dies Paschae anni sequentis 2 Aprilis
Annus a passione ejusdem 1645 Annus ab institutione Sancti 1419
Annus a Nativitate B. Mariae 1692 Melloni
Annus ab assumptione ejus- 1628 Annus a transitu ejusdem 1368
dem Annus ab institutione S.Ro- 1032
Annus indictionis 1 mani
Annus cycli solaris 7 Annus a transitu ejusdem 989
Annus cycli lunaris 7 Annus ab institutione S.Au- 1046
Annus praesens a Pascha 7 doeni
praecedente usque ad pas- Annus a transitu ejusdem 1033
cha sequens est communis Annus a dedicatione hujus 614
adund. Epacta. Ecclesiae Metropolitanae
Aureus numerus 7 Annus ab institutione Rollonis 766
Littera Dominicalis B primi duds Nonnanniae
Littera Martyrologii g Annus a transitu ejusdem 760
Tenninus Paschae 17 Aprilis Annus a coronatione Guillelmi 604
Dies Paschae 10 Aprilis primi duds Nonnanniae in
Lunaipsius 5 Aprilis regno Angliae
Annotinum Paschale 18 Aprilis Annus ab obitu ejusdem 590
Dies rogationum 16 Maji Annus a reductione ducatus 474
Dies Ascensionis 19 Maji Nonnanniae ad Philippum
Dies Pentecostes 29Maji II. Franciae regem
Dies Eucharistiae 9 Junii Annus ab alia reductione duca- 228
Dominicae a Pentecoste usque 25 tus Nonnanniae ad Caro-
ad adventum lum VII. Franciae regem
Dominica prima adventus 27 November Annus Pontificatus SS. Patris 2
Littera dominicalis anni A & DD. Innocentii Papae XI.
sequentis Annus ab institutione R. Patris 7
Annus sequens est 1679 Comm.Ord. & D.D. Francisci IV.
Littera Martyrologii anni T Archiepiscopi Rotomagen-
sequentis sis & Nonnanniae primatis
Dominicae a Nativitate 4 Annus a nativitate Christi a- 40
Domini usque ad septuage- nissimi principis Ludovici
simam anni sequentis XIV. Frandae & Navarrae
Tenninus septuagesimae anni 11 Februarii regis
sequentis Annus regni ipsius 35
Dominica septuagesimae anni 29 Januarii
sequentis
Consecratus est iste cereus in honore Agni immaculati, & in honore gloriosae virginis
ejus genetricis Mariae.
.... .
-:;·
..
Anno ab Incarlia ticfne Domini 1989
·.
Anno a confi.rmatione"·ordinis 773
Anno a
/
tr~itu beati ~~inici
~
768
Benedictus est cereus ~t~
ad honorem
Domini nostri
Jesu Christi
The Dominican Chart
370
APPENDIX 15 - The Suburbicarian Dioceses
Much of the uncertainty over the interpretation of the term 'suburbicarian' vis-a-vis the
blessing of the Easter candle arises from Duchesne's statement that 'the ceremony [of
the Easter candle] was so popular that the Popes, although they did not adopt it in their
own church, were obliged to permit its use in the "suburbicarian" diocese'. 1 He was
conunenting on the authorisation granted by Pope Zosimus (AD 417-18) to churches
subject to papal jurisdiction pennitting them to introduce the ceremony of the Easter
candle:
per parrocia [paroccias] concessa licentia cereum benedici.2
According to Duchesne parocciae referred to the suburban parish churches of the Dio-
cese of Rome. Since, however, there is no documentary evidence before the tenth cen-
tury for the existence of the Easter candle in the papal rite,3 it seems most unlikely that
for over 500 years the rite as observed in the Cathedral of St John Lateran resisted the
introduction of the Easter candle which was blessed in other churches within the City
and Diocese of Rome. It would appear that Duchesne overlooked the fact that paroccia
at this period in the Church's history indicated 'diocese' rather than 'parish'. It is poss-
ible that the seven ancient Suburbicarian Dioceses within the inunediate vicinity of
Rome are here intended; but our observation above regarding the parish churches of the
Diocese of Rome is almost equally applicable in this instance. It is the writer's belief
that parocciae refers either to those dioceses beyond the inunediate vicinity of Rome,
which were subject to papal influence, or to those regions which had been converted to
Christianity by missionaries who owed their allegiance to Rome, or, most likely, to
both.
This view is based partly on the evidence of Ordo 25, which states that the bless-
4
ing of the Easter candle took place in suburbanis civitatibus. Moreover, Amalarius had
observed:
5
Romanis ita agentibus, nobis praeceptum est a papa Zosimo benedicere cereum.
'Christian Worship p.252.

2
'Permission was granted throughout the parishes for the Easter candle to be blessed.'
Duchesne, Liber Pontificalis I p.225.
3
It is f'rrst attested in the Roman rite in PRG c.950.
4
0R 25.2, dated c.800. The phrase relates to cities and states outside Rome as far as
Gaul.
5
'This is the practice at Rome [the making of wax Agnus Dei's]; we were permitted to
bless the Easter candle. Liber Officialis 1.18.1.
371
It is clear from Amalarius' statement that Zosimus' decree had been binding in regions
beyond the commes of Italy.
The phrase 'ordo suburbicaire' was applied somewhat loosely by Capelle (and
accepted by Andrieu) to Ordo 26, a Roman ordo with Gallican influences. 1 Chavasse
argues in favour of an intermediate zone between the Lateran Church and the neigh-
bouring dioceses of Italy, corresponding to the seven Suburbicarian Dioceses, for the
provenance of Ordo 26.2 His theory, however, is based on a questionable understanding
of the relationship of the compiler of the ordo vis-a-vis the grammatical subject of
faciunt in §8.
1
Andrieu, Les Ordines Romani ill p.322.
2
Le Sacramentaire pp.103-4.
372
Key to bibliography
The pwpose of the key is to provide a rapid reference to the primary sources listed in Sections (b) and (c)
of the bibliography. The vast majority of the documents which the writer has consulted relate to or are
associated with specific locations. Reference should therefore be made in the first instance to the left-
hand column. The figure in the right-hand column relates to the corresponding entry in either Section (b)
or Section (c) in the bibliography. A number of primary sources, mainly those not associated with a
specific place, are listed in Section (a) of the bibliography.
Figures in italics and preceded by the letter M refer to the documents cited by Edmond Martene in
De Antiquis Ecclesiae Ritibus, and identified and classified with corresponding numeration by A.-G.Mar-
timort in La documentation liturgique de Dom Edmond Martene.
Abo Manual c.1522 261 Braga Missai i558 99

Ainay, St- Missal 1531 M 175 Bremen Missal 1511 59
Martin Breslau Missal 1483 10
Ales Breviary 1758 142 Breslau Missal 1519 73
Amiens Missal 1555 94 Bressanone Missal 1511 60
Amiens Missal 1752 139 Burgos Missal 1546 90
Amiens Pontifical llC 209 Bursfeld Missal 1498 29
Anderlecht Ordinary 14C 245a Bury Customary c.1234 229
Angers Missal 1489 18 StEdmunds
Angers Ceremonial 1731 126 Cahors Missal 1760 143
Angers Ceremonial 1734 127 Cahors Sacramentary 12 C M 97
Angers, St- Pontifical 14C M 348 Camaldolese Missal 1503 38
Aubin Camaldolese Ceremonial 1634 113
Angouleme Sacramentary 8C 201 Cambrai Missal 1507 49
Apamea Pontifical 1214 M 25 Canterbury Customary 13C 272
Aquileia Missal 1519 72 Canterbury, Missal 13C 265
Arbuthnott Missal c.1480 220 St Augustine
Aries Pontifical 14C M 30 Capuchins Ceremonial 1775 152
Arras Missal 1508 54 Carcassonne Missal 1749 138
Arras, St Ordinal c.l300 199 Carcassonne Sacramentary 13 C M 98
Vedast Carmelites Missal 1504 43
Auch Missal 1836 175 Carmelites Missal 1664 ll6
Auch Ritual 1838 177 Carmelites Ordinary c.1312 289
Auch Sacramentary c.1000 215 Cassino Missal 1507 50
Augsburg Missal 1555 97 Cassino ordo 12 c 248
Austin Friars Missal 1491 20 Cassino Ordinary c.llOO M 1139
Austin Friars Ceremonial 1714 123 Chalonss.M. Missal 1748 137
Autun Sacramentary 8C 235 Chalons s.M. Missal 1543 M84
Autun Missal 1555 95 Chilons s.S. Ordinary 13C M 67
Autun Supplement 1556 96 Chartres Missal 1782 156
Autun Missal 1845 180 Chartres Ordinal 13C 210
Auxerre Breviary 1545 128 Chartres, Sacramentary 10 C M 75
Auxerre Missal 1537 M 39 St-Pierre
Avranches Pontifical 12C M 339 Chartres, Ordinary 12C M 76
Bamberg Missal 1499 31 St-Jean
Bangor Antiphonary 7C 279 Chester Customary llC 193
Barking Ordinal 1404 275 Chezal-Ben. Ceremonial 1531 M 1180
Basel Missal 1488 17 Chezal-Ben. Breviary 1586 M 1181
Bayeux Ceremonial 1677 119 Cistercians Missal 1487 14
Bayeux Sem.Ste 1730 125 Cistercians Missal 1669 118
Bayeux Sem.Ste 1780 154 Cistercians Ritual 1689 121
Bayeux Missal 1790 162 Cluny Missal 1510 57
Bayeux Ordinary 13 c 203 Cluny Customary llC 193
Bayonne Missal 1543 214 Coimbra Ordinary 1579 106
Bazas Missal 1503 42 Coire Missal 1589 108
Beauvais Ritual 1783 159 Cologne Missal 1494 24
Beauvais Pontifical 15 c M 212 Cologne Missal 1514 66
Bee Missal c.1270 239 Cologne Missal 1525 81
Benedictines Missal 1481 6 Cologne Missal 1626 112
Benedictines Missal 1518 71 Cologne Ritual 12C MSGg
Beneventum Missal llC 227 15*
Beneventum Ritual llC 257 Corbie ordo 10C M 1145
Bergamo Sacramentary c.900 260 Corbie Ordinary 12C M 1146
Besan~n Ceremonial 1682 120 Cornie Customary 13C M 1147
Besan~on Ceremonial 1707 122 Cordoba Missal 1561 102
Besan~on Missal 1766 145 Cosenza Missal 1549 91
Biasca Sacramentary c.lOOO 234 Cosenza Ordinary c.1220 M 107
Boulogne Ritual 1780 155 Coutances Missal 1557 98
Bourges Missal 1741 131 Coutances Ceremonial 1825 166
Braga Missal 1512 61
373
Dijon, Customary llC M 1150 Magdeburg Missal 1503 40
St-Benigne Mainz Missal 1507 52
Dominicans Missal 1482 9 Mainz Pontifical 14C M 190
Dominicans Missal 1504 44 Marseille Manual 13C 202
Dominicans Missal 1908 188a Meaux Missal 1845 181
Dominicans H.W.Offices 1927 188b Meissen Breviary 1520 76
Durham Missal 14C 221 Melk: Missal 1495 26
Durham Ritual 269 Melun Missal 1489 M 185
Eichstadt Missal 1517 68 Mende Missal 1776 146
Essen Ordinary 14C 197 Mende Pontifical 14C M 187
Esztergom Missal 1501 36 Metz Missal 1829 168
H<;ztergom Missal 1512 62 Milan Manual llC 252
Evesham Ritual c.1250 287 Milan Missal 1475 1
Evreux Missal 1740 130 Milan Missal 1560 100
Exeter Ordinal 1337 207 Milan Missal 1594 109
Farfa Customary llC 193 Milan Missal 1768 148
Fleury Customary llC 193 Milan Missal 1902 186
Fontevrault Missal 1534 84 Milan Missal 1981 192a
Freising Missal 1487 15 Milan Missal 1986 192b
Freising Missal 1579 107 Minden Missal 1513 64
Fruttuaria Customary llC 193 Nantes Missal 1837 176
Fruttuaria Customary llC 193 Narbonne Missal 1528 82
Fulda Sacramentary 10 C 263 Narbonne Ordinary 14C M 202
Gastine Ordinary 12C M 126 Nevers Sacramentary c.l050 266
Gellone Sacramentary 8C 216 Nidaros Breviary 1519 74
Gilbertines Ordinal 12C 288 Nidaros Manual 13C 217
Glandeves Missal c.1420 M 144 Nidaros Ordinary 13C 228
Grandmont Customary llC M 1184 Norbertine Sacramentary 12 C 283
Halberstadt Missal c.1505 47 Norbertine Breviary 1930 189b
Hamburg Missal 1509 55 Norway Missal 1519 75
Hereford Breviary 13C 225 Norwich Customary c.l265 274
Hereford Missal 1502 236 Noyon Missal 1541 88
Hilde sheim Missal 1499 32 No yon Sacramentary 9C M 202
Hungary Ritual 1815 163 Osma Missal 1561 103
Ireland Missal c.1200 281 Palencia Missal 1568 104
La Rochelle Missal 1835 173 Palermo Missal 1130 270
Langres Missal 1492 22 Paris Missal 1543 89
Langres Missal 1517 69 Paris Ceremonial 1662 115
Langres Directory 1844 179 Paris Missal 1666 117
Langres Ordinary 13C M 168 Paris Missal 1762 144
Laon Rite (1662) 198 Paris Breviary 1778 153
Laon Customary 14C M 1164 Paris Pontifical 15 c M 212
Laon Ordinary 13C M 156 Paris, Royal Sem.Ste 1741 132
LeMans Ceremonial 1789 161 Chapel
Le Mans Ordinary 13C M 89 Passau Missal 1503 41
LePuy Missal 1783 158 Perigueux Missal 1782 157
Le,..Puy Ceremonial 1836 174 Poi tiers Missal 1524 80
Leon Antiphonary 10C 200 Poi tiers Missal 1767 147
Lesnes Missal 13 c 241 Poitiers Missal 1498 M 222
Liege Directory 1492 23 Poland Manual 1819 164
Liege Missal 1540 85 Prague Missal 1498 30
Liege, Ordinary 13C 278 Prague Sacramentary 8C 213
StJames Ratisbon Missal 1518 70
Limoges Missal 1830 169 Ratisbon Ritual 1570 105
Limoges Ordinary 1630 M 158 Regensburg Pontifical 10C 226
Limoges Missal 1483 M 159 Reims Missal 1770 149
Limoges Breviary 1495 M 160 Reims Ordinary c.1200 204
Linkoping Ritual c.1525 223 Reims, Sacramentary c.800 204
Lisieux Ritual c.1530 83 Abbey
Lisieux Ritual 1744 135 ofStRemi-
Lisieux Missal 1752 140 gius
La beck Missal 1505 46 Reims Ordinary c.1300 M 251
Lucon Missal 1828 167 Reims Ritual 1585 M 252
Lund Missal 1514 67 Reims Missal 1491 M 254
Lyon Missal 1510 58 Reims Ritual 14C M 261
Lyon Missal 1771 150 Rennes Missal 1523 79
Lyon Ceremonial 1838 178 Rennes, Missal 12C M 239
Lyon Missal 1846 182 St-Melan
Lyon Missal 1904 187 Rhenau Sacramentary 8C 231
Lyon Ordinary c.l200 M 173 Rhein au Ritual 1114 240
Lyre Customary c.1400 M 1154-5 Ripoll Sacramentary 1038 259
Magdeburg Missal 1480 5 Rome Missal 1477 2
374
Rome Missal 1484 11 St Florian Ritual 12 c 222
Rome Missal 1491 21 StMalo Missal 1503 39
Rome Missal 1500 33 Strasbourg Ritual 1364 134
Rome Missal 1501 35 Strasbourg Ordinal 1364 M 35
Rome Missal 1506 48 Tongres Ordinary 15 c 245b
Rome Missal 1520 77 Tool Missal 1551 92
Rome Ceremonial 1600 110 Tool Ordinary 14 c M 1160
Rome Ritual 1848 183 Toulon Missal 14 c M 310
Rome H.W.Offices 1897 185 Toulouse Missal 1832 170
Rome Missal 1950 190 Toulouse Missal 1490 M 311
Rome Missal 1970 191 Tournai Missal 1540 86
Rome Missal 1754 208 Tours Missal 1784 160
Rome Missal 1474 249 Tours Sacramentary 9C M 320
Rome Missal 1574 249 Tours Missal 13C M 324
Rome Breviary 1879 253 Tours, Ritual 13C 264
Rome Missal 1558 lOla St-Martin
Rome Missal 1560 101b Trier Missal c.l487 16
Rome Missal 1928 189a Trier Ordinary c.l300 242
Rouen Breviary 1480 4 Troyes Missal 1736 129
Rouen Missal 1497 27 Upsala Missal 1513 65
Rouen Ritual 1640 114 Utrecht Missal 1497 28
Saintes Missal c.1500 34 Utrecht Missal 1540 87
Salisbury Missal C.1486 13 Uzes Missal 1495 25
Salisbury Ritual 13C 205 Uzes Ordinary 14C M 346
Salisbury Missal 13 c 212 Valence Missal 1504 45
Salisbury Ordinal c.l210 224 Vallombrosa Missal 1503 37
Salisbury Customary 13C 224 Verden Missal 1486 12
Salisbury Processional 1517 238 Verdun Missal 1481 7
Salisbury Missal c.l300 247 Verdun, St Customary llC 193
Salisbury Breviary 1531 262 Vito
Salisbury Missal 13C 282 Verdun Ceremonial 1832 171
Salzburg Missal 1507 53 Verona Ritual 1609 111
Saragossa Missal 1552 93 Vich Sacramentary 1038 258
sees Missal 1742 133 Vienna Missal 1782 165
sees Ritual 1834 172 Westminster Missal c.l370 246
Senlis Ordinary 14C M 301 Westminster, Customary 13C 273
Sens Missal 1520 78 StPeter
Sens Missal 1715 124 Worcester Antiphonary 13C 255
Sens Ritual 1555 M 299 Worms Missal 1490 19
Seville Missal 1507 51 Wiirzburg Directory 1477 3
Soissons Missal 1745 136 Wi.irzburg Missal 1481 8
Soissons Ritual 1753 141 WUizburg Missal 1509 56
Soissons Ritual 1856 184 York Missal 14C 237
Soissons Ritual c.ll85 M 305 York Breviary c.1050 243
Spires Missal 1512 63 York, Ordinal c.1400 268
St-Bertrand Missal 1773 151 StMary
St-Denys Customary c.1273 M 1158
St-Germain- Customary 1394 M 1165 * For this document cited by Martene (DAER
des-Pres 4.24.3 p.145), see Martimort.
St-Germain- Pontifical 12C M 230
des-Pres
375
BIBLIOGRAPHY
(a) Medieval writers and documents (see also Key to bibliography)
Peter Abelard : in PL 178.

Acta Vetera of Rouen Cathedral (11 C): in PL 147
Alcuin : Liber de divinis officiis in PL 101.
Amalarius (c.830): Liber Officialis in Vol.2, and Liber de Ordine Antiphonarii in
Vol.3 ofNo.232.
Bede: in PL 90-95, except A History of the English Church and People, [ET L.
Sherley-Pricej (Penguin Books, Harrnondsworth, 1955).
Beroldus: No.251.
Bindo Fesulani (AD 1377): in No.267.
The Bobbio Missal: in No.250.
Codex Avignon 1706 (c.1350) : in No.267
Drepanius Florus: in PL 61.
Durandus: (see Bibliography).
The Pontifical of Egbert : No.230.
Flodoardus: in PL 135.
The Gelasian Sacramentary: No.256.
The Gradual of Beneventum : in Hesbert.
The Gregorian Sacramentary : No.284.
Gregory of Tours: Historia Francorum in PL 71.
Haymo's Ordines : Nos.233 and 276.
Honorius of Autun : Gemma Animae, Sacramentarium, Speculum in PL 172.
Hugh of St Victor : in PL 177
John Beleth : Rationale in PL 202.
John of Avranches: Liber de officiis ecclesiasticis in PL 147.
Lanfranc : Decreta Pro Ordine Sancti Benedicti in PL 150.
The Lateran Missal : in Schmidt Vol.ll.
The Lateran Ordo: No.219.
The Leofric Collectar : No.211.
The Leofric Missal : No.280.
Liber Ordinum (Mozarabic): No.218.
Magdelen College Pontifical: No.281.
Micrologus : in PL 151.
The Missal of Robert of Jumieges : No.286.
Missale Antiquum (Beneventum): in Hesbert.
Missa/e Mixtum : in PL 85.
Monte Cassino A andB : inNo.194(i).
Nomasticon Cisterciense : No.271.
Ordines Romani 11-30B: in No.195.
Ordines Romani X-XV : in PL 78.
Ordo Albini :in Fabre Vol.2
Ordo Romanus I: in No.196.
The Constitutions of John Orsini: in Mallardo.
The Pontifical of Poitiers, : No.254.
Pontificate Gulietmi Durandi (13 C): No.194 (iii).
Pontificate Romanae Curiae (13 C): No.194 (ii).
Pontificate Romano-Germanicum : No.277.
Pontificate Romanum (12 C): No.194 (i).
Prudentius: Carmina in No.206.
Rabanus Maurus : in PL 107.
Regutaris Concordia : in PL 137
Robert Paululus : in PL 177.
The Rosslyn Missal : No.244.
Rupert ofDeutz (c.1110): De Divinis Officiis inPL 170.
Sacramentarium Vetus (11 C): in PL 151.
Sicardus: Mitrale in PL 213.
St Eligius Sacramentary : in Menard.
The Wulfstan Collectar : in No.211.
376
(b) Original primary sources
1. Missale Ambrosianwn (tpm), Milan, 1475.

2. Missale Romanwn (tpm), Venice, 1477
3. Ordinarius Herbipolensis (tpm), Spires, 1477
4. Breviariwn Rothomagense (tpm), Rouen, 1480.
5. Missale ad usum ecclesiae Magdeburgensis, Magdeburg, 1480
6. Missale Benedictinwn (tpm), Bamberg, 1481.
7. Missale ... .Virdunensis Ecclesiae, Paris, 1481.
8. Missa1e Herbipolense (tpm), Wurzburg, 1481.
9. Missale ad uswn Ordinis Praedicatorum (tpm), Venice, 1482.
10. Missale Vratislavense (tpm), Mainz, 1483.
11. Missale Romanwn (tpm), Ntimberg, 1484.
12. Missale Virdunense (tpm), Magdeburg, 1486.
13. Missale ad uswn ecclesiae Sarum (tpm), Basel, c.1486.
14. Missale Cisterciense (tpm), Strasbourg, 1487.
15. Missale Frisingense, Bamberg, 1487.
16. Missale Treverense (tpm), Basel, c.l487
17. Missale Basiliense (tpm), Basel, 1488
18. Missale Andegavense, Paris, 1489.
19. Missale Wormatiense (tpm), Basel, 1490.
20. Missal of the Augustinian Friars (tpm), Ntimberg, 1491.
21. Missale Romanwn (tpm), Papie, 1491.
22. Missale ad usum ecclesiae Lingonensis, Paris, 1492.
23. Ordinarius ecclesiae maioris Leodiensis, Cologne, 1492.
24. Missale Coloniense (tpm), Cologne, 1494.
25. Missale Uceciense (tpm), Mainz and Lyon, 1495.
26. Missale Benedictinae religionis Monachorum cenobii Me/licensis, Ntimberg, 1495.
27. Missale secundum usum insignis ecclesiae Rothomagensis, Rouen, 1497.
28. Missale insignis ecclesiae Traiectensis, Paris, 1497.
29. Missale ordinis Sancti Benedicti burszfeldensis ordinis, Spires, 1498.
30. Missale emendatum iuxta rubricas Pragensis ecclesiae, Leipzig, 1498.
31. Liber missalis secundum ordinem ecclesiae Bambergensis, Bamberg, 1499.
32. Missale Hildersemense, Ntimberg, 1499.
33. Missale secundum usum romanae ecclesiae, Lyon, 1500.
34. Missale Santonense (tpm), Paris, c.1500
35. Missale Romanum, Venice, 1501.
36. Missale Strigoniense (tpm), Lyon, 1501.
37. Missale monasticum secundum consuetudinem ordin.is Vallisumbrosae, (Venice,
1503), reprinted in London, nd.
38. Missale monasticum secundum ordinem Camaldulensem, Venice, 1503.
39. Missale ad usum insignis ecclesiae maclonensis, Rouen, 1503.
40. Missale Magdeburgense (tpm), Ntimberg, 1503.
41. Missale Pataviense, Vienna, 1503.
42. Missale ....ad usum ecclesiae vasatensis, Roelle, 1503.
43. Missale secundum usum Carmelitarum, 1504, reprinted in London, 1936.
44. Missale Praedicatorum, Venice, 1504.
45. Missale ad usum ecclesiae valentinensis, Valence, 1504.
46. Missale Lubicense (tpm), Speyer, 1505.
47. Missale Halberstatense (tpm), Mainz, c.l505.
48. Missale Romanum, Venice, 1506.
49. Missale secundum usum venerabilis ecclesiae Cameracensis, Paris, 1507.
50. Missale .... secundum morem et ritum Casin(msis congregationis, Venice, 1507.
51. Missale secundum usum almae ecclesiae hyspalensis, Seville, 1507.
52. Missale Maguntinum, Mainz, 1507.
53. Missale Saltzburgense, Venice, 1507.
54. Missale ad usum insignis ecclesiae Atrebatensis, Arras, 1508.
55. Ordo missalis secundum ritum laudabilis ecclesiae Hamburgensis, Strasbourg,
1509.
56. Missale secundum usum ecclesiae Herbipolensis, Lyon, 1509.
377
57. Missale Ordinis Cluniacensis, Paris, 1510.
58. Missale ad usum lugdunensis ecclesiae, Lyon, 1510.
59. Missale secundum ritum ecclesiae Bremensis, Strasbourg, 1511.
60. Missale secundum ritum ecclesiae Brixinensis, Basel, 1511.
61. M issale secundum ritum et consuetudinem almae Braclulrensis ecclesiae, Salman-
tice, 1512.
62. Missale secundum chorum ecclesiae Strigoniensis, Venice, 1512.
63. Agenda Spirensia, Spires, 1512.
64. Missale Mindense (tpm), Niirnberg, 1513.
65. Missale Upsalense (tpm), Basel, 1513
66. Missale Diocesis Coloniensis, Paris, 1514.
67. Missale Lundense av ar 1514, Faksimiledition, Malmo, 1946.
68. Missale secundum chorum et ritum Eystetensis Ecclesiae, Niirnberg, 1517.
69. Missale diocesis Lingonensis, Paris, 1517.
70. Missale secundum usum ecclesiae ratisponensis, Bamberg, 1518.
71. Missale .... Ordinis Sancti Benedicti, Hagen, 1518.
72. Missale Aquileyensis Ecclesiae, (Venice, 1519), reprinted in Brussels, 1963.
73. Missale secundum Rubricas Vratislavensis diocesis, Basel, 1519.
74. Breviaria ad usum ritumque sacrosanctae Nidrosiensis Ecclesiae, Oslo, 1519.
75. Missale pro usu totius regni Norvegiae, 1519, reprinted in Oslo, 1959.
76. Breviarius ... Misnensis Ecclesiae, Meissen, 1520.
77. Missale ad Romanae ecclesiae usum, Paris, 1520.
78. Missale Senonense, Paris, 1520.
79. Missale Redonense (tpm), Paris, 1523.
80. Missale Pictaviense (tpm), Limoges, 1524.
81. Missale Diocesis Coloniensis, Paris, 1525.
82. Missale ad usum sanctae Narbonensis ecclesiae, Lyon, 1528.
83. Manuale sacerdotum continens ecclesiae sacramenta et modum administrandi, Pes-
taudemer, c.1530.
84. Missal of Fontevrault, Paris, 1534.
85. Missale ad usum insignis ecclesiae Leodiensis, Paris, 1540.
86. Missale ad usum insignis ecclesiae Tornacensis, Antwerp, 1540.
87. Missale ad verum cathedra/is ecclesiae Traiedensis ritum, Antwerp, 1540.
88. Missale ad usum insignis ecclesiae Noviomensis, Paris, 1541.
89. Missale ad usum ecclesiae Parisiensis, Paris, 1543.
90. Missale secundum consuetudinem Burgensis ecclesiae, Burgos, 1546.
91. Missale secundum consuetudinem Ecclesiae Cosentinae, Venice, 1549.
92. Missale Tullense (tpm), Toul, 1551.
93. Missale Cesaraugustanum, Saragossa, 1552.
94. Missale Ambianense, Paris and Amiens, 1555.
95. Sact·orum Codex iuxta ritum Ecclesiae Heduensis, Lyon, 1555.
96. Supplementum Missalis Aeduensis c.1700
97. Missale secundum ritum Augustensis ecclesiae, Dillingen, 1555
98. Missale .... Constanciensis Ecclesae, Rouen, 1557.
99. Missale iuxta usum et ordinem Almae Bracarensis Ecclesiae, Lyon, 1558
100. Missale Ambrosianum, Milan, 1560.
101. (a) Missale Romanum, Venice, 1558. (b) Missale Romanum, Venice, 1560
102. Missale Cordubensis Ecclesiae, Cordova, 1561.
103. Missale secundum usum ....sanctae ecclesiae Oxomensis, Cordova, 1561.
104. Missale Pallantinum, (C.Femandez aValtodano), Palencia, 1568.
105. Obsequiale ...secundum antiquum usum ... Ecclesiae Ratisbonensis, lngolstadt,
1570.
106. Ordinaria dos Canonicos Regulares de Ordem do ...S.Augustinho de congregat;ao
de sancta Cruz de Coimbra, Lisbon, 1579.
107. Missale Frisingense, Munich, 1579.
108. Missale secundum ritum Curiensis ecclesiae, Constance, 1589.
110. Caeremoniale Episcoporum, 1st edition, Rome, 1600.
111. Rituale Ecclesiae Veronensis, (Alberto Valerio), Verona, 1609.
112. Missale Coloniense, Cologne, 1626.
378
113. Ceremoniale Eremitarum Camaldulensium S.Romua/di, Florence, 1634.
114. Sacerdotale seu Maunuale Ecclesiae Rothomagensis, Rouen, 1640.
115. Caeremoniale Parisiense ad usum Urbis et Diocesis Parisii, Paris, 1662.
116. Missale Fratrum Beatae Dei Genetricis Virginis Mariae de Monte Carmelo, Paris,
1664.
117. Missale Parisiense, (Harduin de Pereft.xe), Paris, 1666.
118. Missale Cisterciense, Paris, 1669.
119. Ceremonial pour I' Eglise et le Diocese de Bayeux, Caen, 1677.
120. Ceremonial du diocese de Besanfon, (A.P. de Grammont), Besans;on, 1682.
121. Rituale Cisterciense, Paris, 1689.
122. Ceremonial du Diocese de Besanfon, Besans;on, 1707.
123. Sacrae Caeremoniae ... iuxta ritum ... Fratrum Eremitarum SAugustini, Rome, 1714
124. Missale Senonense, (H.F. de Ia Hoguette), Sens, 1715.
125. Semaine Sainte a /'usage du Diocese de Bayeux, (J.-D. de Cheylus), Bayeux,
1730.
126. Ceremonial de I' Eglise d' Angers, Chateuagontier, 1731.
127. Diurnale Andegavense, Angers, 1734.
128. Breviarium Autissiodorense, Paris, 1736.
129. Missale sanctae ecclesiae Trecensis, (J.B.Bossuet), Troyes, 1736.
130. Missale Ebroicense, (P.-J.-C. de Rochechouart), Paris, 1740.
131. Missale Bituricense, (F.H. de Roye de Ia Rochefoucauld), Bourges, 1741.
132. L'Office de Ia Semaine-Sainte a /'usage de Ia Maison du Roy, (L'Abbe de Belle-
garde), Paris, 1741
133. Missale Sagiense, (L.-F. Neel de Christot), Paris, 1742.
134. Rituale Argentinense, (Armand, Cardinal de Rohan), Strasbourg, 1742.
135. Rituale Lexoviense, (H.-I. de Brancas), Paris, 1744.
136. Missale Suessionense, (F. Due de Fitz-James), Paris, 1745.
137. Missale ...Ecclesiae Cathalaunensis, (Choiseul-Beaupre), Chalons-sur-Marne,
1748.
138. Missale Carcassonense, (A.Bazin de Besons), Paris, 1749.
139. Missale sanctae Ambianensis ecclesiae, (D'Orleans de 1a Motte), Amiens, 1752.
140. Missale Lexovicense, (H.-I. de Brancas), Lisieux, 1752.
141. Rituel du Diocese de Soissons, (F. Due de Fitz-James), Paris, 1753.
142. Breviarium Alesiense, Paris, 1758.
143. Missale Cadurcense, (B.-D.-R. du Guesclin), Paris, 1760.
144. Missale Parisiense, (C.-G.-G. de Vintimille), Paris, 1762.
145. Missale Bisuntinum, (A.-C. de Choiseul-Beaupre), Besans;on, 1766.
146. Missale Mimatense, (G.-F. de Choiseul-Beaupre), Paris and Mende, 1766.
147. Missale Pictaviense, (M.-L. de Beaupoil de St-Aulaire), Poitiers, 1767.
149. Missale sanctae ecclesiae metropolitanae Remensis, (C.-A. de Ia Roche-Aymon),
Paris, 1770.
150. Missale Sanctae Lugdunensis Ecclesiae, Lyons, 1771.
151. Missale Convenarum, (C.-A.-G. Osmond), Toulouse, 1773.
152. Memoriale Rituum sive Caeremoniale in Ordine FF Minorum S.Francisci Cappu-
cinorum, Rome, 1775.
153. Breviarium Parisiense, Pars Vema, Paris, 1778.
154. Semaine Sainte a /'usage du Diocese de Bayeux, Bayeux, 1780.
155. Rituel de diocese de Boulogne, (F.-1. de Portz de Pressy), Boulogne, 1780.
156. Missale Carnotense, (1.-B.-J. de Lubersac), Chartres, 1782.
157. Missale Petrocorense, (E.-L. de Grossolles de Flamarens), Paris, 1782.
158. Missale ecclesiae Aniciensis, (M.-1. de Galard de Terraube), Paris, 1783.
159. Rituale Bellovacense, (F.-1. dele Rochefoucauld), Beauvais, 1783.
160. Missale Turonense, (1.-M.-F. de Conzie), Paris, 1784
161. Ceremoniale de I' Eglise Cathedrale de St-Julian du Mans, Le Mans, 1789.
162. Missale Bajocense, (J.D. de Cheylus), Lyons, 1790.
163. Rituale Agriense, (S.Fischer de Nagy-Szalatnya), Budapest, 1815.
164. Manuale Caeremoniarum Romanarum ...ad usum Ecclesiarum Poloniae, Part II,
Warsaw, 1819.
165. Missale ad usum provinciae Viennensis, (1782 Missal), Graz, 1822.
379
166. Ceremonial du Diocese de Coutances, (P.Dupont-Poursat), Coutances, 1825.
167. Missale Lucionense, (R.F.Soyer), Lyon, 1828.
168. Missale Metense, (1.-F. Besson), Metz, 1829.
169. Missale Lemovicense, Limoges, 1830.
170. Missale Tolosanum, Toulouse, 1832.
171. Ceremonial a I' Usage de Ia Cathedrale et du Diocese de Verdun, (Mgr de Ville-
neuve), Verdun, 1832.
172. Rituale Sagiense, (L.-F. Neel de Cristot), Sees, 1834.
173. Missale Rupellense, (D.J.Bemet), Paris, 1835.
174. Ceremoniale aI' usage de I' Eglise du Puy, (L.-J.-M. de Bonald), Le Puy, 1836.
175. Missale Auscitanum, (J.-F. de Montillet), Auch, 1836.
176. Missale Nannetense, (J.M. Micolon de Guerines), Nantes, 1837.
177. Rituel a I' usage du Diocese d' Auch, (Cardinal D'Isoard), Auch, 1838.
178. Le Ceremonial de Ia sainte Eglise de Lyon, Lyon, 1838.
179. Pastorale Lingonense, Langres, 1844.
180. Missale Aeduense, (B.-U.-J.-M. du Trousset D'Hericourt), Autun, 1845.
181. Missale Me/dense, (R.-F. Gallard), Meaux, 1845.
182. Missale Sanctae Lugdunensis Ecclesiae, Lyon, 1846.
183. Rituale Romanum, (as revised and enlarged by Benedict XIV), Rome, 1848.
184. Rituale seu Mandatum insignis ecclesiae Suessionensis, Soissons, 1856.
185. Officia Hebdomadae Sanctae (According to the Breviary) (Tournai, 1897).
186. Missale Ambrosianum, t.e. (authorised 21 December 1901), Milan, 1902.
187. Missale Romanum in quo antiqui ritus Lugdunenses servantur, Lyon, 1904.
188. (a) Missale iuxta ritum Sacri Ordinis Praedicatorum, Rome, Tournai, 1908. (b)
Office of Holy Week according to the Dominican rite, (Burns, Oates, Wahbourne,
London, 1927).
189. (a) Daily Missal, edited by Dom G.Lefebvre, Bruges and London, 1928. (b) Bre-
viarium Praemonstratense (Burns, Oates, Washbourne, 1930).
190. Missale Romanum (ex decreto sacrosancti Concilii Tridentini), Rome, 1950.
191. (a) Missale Romanum typical edition (Rome, 1970). (b)Pope John Sunday Missal
(Kevin Mayhew, Leigh-on-Sea, 1978).
192. (a) Missale Ambrosianum typical edition, (Casa Archivescovile, Milan, 1981). (b)
Messale Ambrosiano (Casa Archivescovile, Milan, 1986).
(c) Modern editions of primary sources
193. Albers, B., Consuetudines Monasticae, 5 vols (Stuttgart, Vienna, Monte Cassino,
1900-11).
194. Andrieu, M., Le Pontifical Romain au Moyen-Age, Vol.I: Le Pontifical Romain
du XII siecle, Vol.ll : Le Pontifical de la Curie Romaine au XIII siecle, Vol.ill : Le
Pontifical de Guillaume Durand, Studi e Testi Nos.86, 87, 88 (Vatican, 1938, 1940,
1940).
195. Andrieu, M., Les Ordines Romani du Haut Moyen-Age, 5 vols, Spicilegium
Sacrum Lovaniense (Louvain, 1931-61).
196. Atchley, E.G.C.F., OrdoRomanus Primus, Library ofLiturgiology and Ecclesio-
logy for English Readers, Vol.6 (London, 1905).
197. Arens, F., Liber Ordinarius der Essener Stiftskirche (Paderbom, 1908).
198. Bellotte, A., Observationes ad ritus Ecclesiae Laudunensis Redivivos, (Paris,
1662).
199. Brou, L., The Ordinal of St Vedast' s Abbey, Arras, Henry Bradshaw Society
Publications, Vol.86 (London, 1957).
200. Brou, L. and Vives, J., Antifonario Visigotico Mozarabe de Ia Catedral de Leon,
Monumenta Hispania Sacra [Series Latina, Vol.V.l] (Barcelona, Madrid, 1959).
201. Cagin, P., Le Sacramenta ire Gelasien d' Angouleme (Macon, 1919).
202. Chevalier, U., Institutions Liturgiques de I' Eglise de Marseille, Bibliotheque
Liturgique, XIV (Picard et Fils, Paris 1910).
380
203. Chevalier, U., Ordinaire et Coutumier del' Eglise Cathedrale de Bayeux,
Bibliotheque Liturgique, Vlli (Picard et Fils, Paris, 1902).
204. Chevalier, U., Sacramentarium Abbatiae S.Remigii Remensis, Bibliotheque Litur-
gique, Vll (A.Picard, Paris, 1900).
205. Collins, A.J., Manuale Sarum, Henry Bradshaw Society Publications, Vol.91
(London, 1960).
206. Cunningham, M.P., Aurelii Prudentii Clementis Carmina, Corpus Christianorum
[SeriesLatina, CXXVl] (T~olt, 1966)e
207. Dalton, J.N., Ordinale Exonense, Henry Bradshaw Society Publications, Vol.37
(London, 1909).
208. De Azevedo, E., Vetus Missale Romanum Monasticum (Rome, 1754).
209. De Beauville, V. and Josse, H., Le Pontifical d' Amiens (du xr siecle) (T.Jeunet,
Amiens, 1885).
210. Delaporte, Y., L'Ordinaire chartrain du XIIr siecle (Societe Archaeologique
d'Eure-et-Loire, Chartres, 1953).
211. Dewick, E.S. and Frere, W.H., The Leofric Collectar, Vol.2, Henry Bradshaw
Society Publications, Vol.56 (London, 1921 ).
212. Dickinson, F.H., Missale ad usum insignis et praeclarae ecclesiae Sarum, (Bum-
tisland, 1861-83).
213. Dold, A. and Eizenhofer, L., Das Prager Sakramentar (Beuroner Kunstverlag,
Beuron in Hohenzollem, 1949).
214. Dubarat, V., Le Missel de Bayonne de 1543 (Pau, Paris, Toulouse, 1901).
215. Duffour, J., 'Fragments d'un Ancien Sacramentaire d' Auch' in Archives Histo-
riques de Ia Gascogne (Uonce Cochareaux, Auch, 1912).
216. Dumas, A., Liber Sacramentorum Gellonensis, Corpus Christianorum [Series
Latina, CLIX] (Turnholt, 1981).
217. Faehn, H., Manuale Norvegicum (Presta Handbok), Libri Liturgici Provinciae
Nidrosiensis Medii Aevii, Vol.1 (Oslo, 1962).
218. Ferotin, D.M., Le Liber Ordinum (en usage dans I' eglise wisigothique et mozarabe
d' Espagne), originally published in Paris, 1909 (Gregg International Publishers,
Famborough, 1969).
219. Fischer, L., Ordo Officiorum Ecclesiae Lateranensis (Munich and Freising, 1916).
220. Forbes, A.P., Liber Ecclesiae Beati Terrenani de Arbuthnott (Bumtisland, 1864).
221. Fowler, J.T., 'The Durham Missal', in The Rites of Durham, Surtees Society
Publications, Vol.107 (Durham, 1903), pp.172-191.
222. Franz, A., Das Rituale von St Florian aus dem zwolften Jahrhundert (Freiburg irn
Br., 1904).
223. Freisen, J., Manuale Lincopense, Breviarium Scarense, Manuale Aboense (Pader-
bom, 1904).
224. Frere, W.H., The Sarum Customary (Cambridge U.P., 1898).
225. Frere, W.H. and Brown, L.E.G., The Hereford Breviary, Henry Bradshaw Society
Publications, Vol.26 (London,1904).
226. Gamber, K. and Rehle, S., Das Sakramentar-PontifikLll des Bischops Wolfgang von
Regensburg (F.Pustet, Regensburg, 1985).
227. Gamber, K., Missale Beneventum von Canosa (F.Pustet, Regensburg, 1972).
228. Gjerl~w, L., Ordo Nidrosiensis Ecclesiae, Libri Liturgici Provinciae Nidrosiensis
Medii Aevi, Vol.2 (Oslo, 1968).
229. Gransden, A., Consuetudines Burienses, Henry Bradshaw Society Publications,
Vol.99 (London, 1973).
230. Greenwell, W., The Pontifical of Egbert, Surtees Society Publications, Vol.27
(Newcastle, 1853).
231. Hanggi, A. and Schonherr, A., Sacramentarium Rhenaugiense (Freiburg (Sw)
1970).
232. Hanssens, J.M., Amalarii Episcopi Opera Liturgica Omnia, 3 vols, Studi e Testi,
Nos 138,139,140 (Vatican, 1948-50).
233. (Haymo of Faversham), The Ordines of Haymo of Faversham, Henry Bradshaw
Society Publications, Vol.85 (London, 1953).
234. Heirning, 0., Das Ambrosianische Sakramentar von Biasca, Corpus Ambrosiano
Liturgicum II (Munich (W.F.), 1969).
381
235. Heiming, 0., Liber Sacramentorum Augustodunensis Corpus Christianorum,
~eries latina CUXB] (fumholt, 1984).
236. Henderson, W.G., The Hereford Missal (Leeds, 1874).
237. Henderson, W.G., Missale ad usum insignis Ecclesiae Eboracensis, Surtees
Society Publications, Vol.59 (Newcastle, 1872).
238. Henderson, W.G., Processionale ad usum insignis ac praeclarae Ecclesiae Sarum
(Leeds, 1882).
239. Hughes, A., The Bee Missal, Henry Bradshaw Society Publications, Vol.94 (Lon-
don, 1963).
240. Hiirlimann, G., Das Rheinauer Rituale, Spicilegium Friburgense, No.5, (Univers-
itatsverlag, Freiburg (Sw.), 1959).
241. Jebb, P., Missale de Lesnes, Henry Bradshaw Society Publications, Vol.95
(Worcester, 1964).
242. Kurzeja, A., Der A/teste Liber Ordinarius der Trierer Domkirche, Liturgiewis-
senschaftliche Quellen und Forschungen, Vo1.52 (Munich (W.F.), 1970).
243. Lawley, S.W ., The York Breviary, Surtees Society Publications, Vol. 71 (Newcas-
tle, 1879).
244. Lawlor, H.J., The Rosslyn Missal, Henry Bradshaw Society Publications, Vol.15
(London, 1899).
245. Lefevre, P.F., (a) Les Ordinaires des collegiales Saint-Pierre aLouvain et
Saints-Pierre-et-Paul aAnderlecht (Publications Universitaires de Louvain, Lou-
vain, 1960). (b) L'Ordinaire de Tongres: I. Le Temporal, Spicilegium Sacrum
Lovaniense, Vol.34 (Louvain, 1967).
246. Legg, J.W., Missale ad usum Ecclesiae Westmonasteriensis, Vol.2, Henry Brad-
shaw Society Publications, Vo1.5 (London, 1893).
247. Legg, J.W., The Sarum Missal (Oxford U.P., 1916).
248. Leutermann, T., Ordo Cassinensis hebdomadae maioris, Miscellanea Cassinese,
20 (Monte Cassino, 1941 ).
249. Lippe, R., Missale Romanum Mediolani, 1474, 2 vols, Henry Bradshaw Society
Publications, Vols 17 and 33 (London, 1899 and 1907).
250. Lowe, E.A., The BobbiiJ Missal, Henry Bradshaw Society Publications, Vol.58
(London, 1920).
251. Magistretti, M., Beroldus: Ecclesiae Ambrosianae Mediolanensis Kalendarium et
Ordines Saec. XII (J.Giovanola, Milan, 1894).
252. Magistretti, M., Manuale Ambrosianum, Monumenta Veteris Liturgiae Ambrosia-
nae, Part IT (Milan, 1904).
253. Marquess ofBute, The Roman Breviary, Vol. I (Winter) (W.Blackwood & Sons,
London and Edinburgh, 1879).
254. Martini, A., II Cosidetto Pontificale Di Poitiers (Herder, Rome, 1979).
255. Mocquereau, A., L' Antiphonaire Monastique de Worcester, Paleographie Musicale
Xli (Berne, 1971).
256. Mohlberg, L.C. with L.Eizenhofer and P.Siffrin, Liber Sacramentorum Romanae
Aeclesiae Ordinis Anni Circuli, Rerum Ecclesisticarum Documenta, Fontes IV
(Herder, Rome, 1960).
257. Odermatt, A., Ein Rituale in Beneventanischer Schrift, Spicilegium Friburgense
26 (Universitatsverlag, Freiburg (Sw ), 1980).
258. Olivar, A., El Sacramentario de Vich (Madrid and Barcelona, 1953) .
. 259. Olivar, A., Sacramentarium Rivipullense (Madrid and Barcelona, 1954).
260. Paredi, A., Sacramentarium Bergamense (Bergamo, 1962).
261. Parvio, M., Manuale seu Exsequiale Aboense (c.1522) (Societas Historiae Eccle-
siasticae Fennica, Helsinki, 1980).
262. Procter, F. and Wordsworth, C., Breviarium ad usum insignis ecclesiae Sarum
(Cambridge U.P., 1882).
263. Richter, G. and Schonfelder, A., Sacramentarium Fuldense, 1912 edition reprinted
as Henry Bradshaw Society Publication, Vol.101 (London, 1977).
264. Rituel de Saint-Martin de Tours - Xlll siecle, Documents et Manuscrits (Paris,
1899).
265. Rule, M., The Missal ofSt Augustine's Abbey, Canterbury (Cambridge U.P.,
1896).
382
266. Sacramentarium ad usum Aecclesiae Nivernensis, (Gregg International Publishers,
Famborough, 1969).
267. Schinnnelpfennig, B., Die Zeremonienbucher der Romischen Kurie in Mittelalter
(Tubingen, 1973).
268. Stanbrook, Abbess of and Tolhurst, J .B .L., The Ordinary and Customary of St
Mary's, York, Vol.2, Henry Bradshaw.Society Publications, Vol.75 (London,
1937).
269. Stevenson, J., Rituale Ecclesiae Dunelmensis, Surtees Society Publications,
Vol.lO (Newcastle, 1841).
270. Terrizzi, F., Missale Antiquum S. Panormitanae Ecclesiae, Rerum Christianarum
Docurnenta, Fontes XII (Herder, Rome, 1970).
271. Sejalon, H., Nomasticon Cisterciense (Solesmes, 1892).
272. Thompson, E.M., The Customary of the Benedictine Monasteries ofSt Augustine,
·Canterbury and StPeter, Westminster, Henry Bradshaw Society Publications,
273. Thompson, E.M., The Customary of StPeter, Westminster, Vol.2, Henry Bradshaw
Society Publications, Vol.28 (London, 1904).
274. Tolhurst, J.B.L., The Customary of Norwich, Henry Bradshaw Society Publica-
tions, Vol.82 (London, 1948).
275. Tolhurst, J.B.L., The Ordinate and Customary of Barking Abbey, Vol.1, Henry
Bradshaw Society Publications, Vol.65 (London, 1927).
276. Van Dijk, S.J.P., The Ordinals of Haymo of Haversham and related documents,
1243-1307, Sources of the Modem Roman Liturgy, Vol.ll (Leiden, 1963).
277. Vogel, C. and Elze, R., Le Pontifical Romano-Germanique du dixieme siecle, 2
vols, Studi e Testi, 226 and 227 (Biblioteca Apostolica Vaticana, Vatican, 1963).
278. Volk, P., Der Liber Ordinarius Des Lutticher St Jakobs-Klosters (Aschendorff
Verlag, Munich (W.F.), 1923).
279. Warren, F.E., The Antiphonary of Bangor. 2 vols, Henry Bradshaw Society Publi-
cations, Vo1.4 and Vol.lO (London, 1894-95).
280. Warren, F.E., The Leofric Missal (Oxford U.P., 1883).
281. Warren, F.E., Missale Vetus Hibernicum (London, 1979).
282. Warren, F.E., The Sarum Missal Alcuin Club Collections, No.ll (London, 1911).
283. Weyns, N.I., Sacramentarium Praemonstratense, Biblioteca Analectorum Prae-
monstratensium (Averbode, Premontre, 1968).
284. Wilson, H.A., The Gregorian Sacramentary, Henry Bradshaw Society Publica-
tions, Vol.49 (london, 1915). See also Menard in general bibliograpy.
285. Wilson, H.A., The Magdelen College Pontifical, Henry Bradshaw Society Publi-
cations, Vol.39 (London, 1910).
286. Wilson, H.A., The Missal of Robert of Jumieges, Henry Bradshaw Society Publi-
cations, Vol.ll (London, 1896).
287. Wilson, H.A., Officium Ecclesiasticum abbatum secundum usum Eveshamensis
monasterii, Henry Bradshaw Society Publications, Vol.6 (London, 1891).
288. Woolley, R.M., Ordinale Gilbertinum, Henry Bradshaw Society Publications,
289. Zimmermann, B., Ordinaire de l'Ordre de Notre Dame duMont Carmel par Sibert
de Beka (vers 1312), (A.Picard et Fils, Paris, 1910).
(d) General bibliography (modern writers)
Aheame, P. and Lane, M.,. Pontifical Ceremonies (Burnes, Oates, Dublin, 1942).
Albers, D.B., 'Die Altesten Consuetudines von Vallumbros', Revue Benedictine, 23
(1911), pp.432-36.
Alleau, R., Guide de Ia France Mysterieuse (Paris, 1964 ).
Andrieu, M., 'Le Pontifical d'Apamee', Revue Benedictine, 48 (1936), pp.321-48.
Armellini, M., Lezioni di Archeologia Cristiano (F.Cuggiani, Rome, 1898).
383
Atchley, E.G.C.F., A History of the Use of Incense in Divine Worship, Alcuin Club Col-
lections, No.13 (London, 1913).
Avery, M., The ExultetRolls of Southern Italy, (Princeton, 1936).
Avery, M., 'Beneventan Lections', Studi Gregoriani 1 (1947), pp.433-58.
Babylonian Talmud, [ET !.Epstein], Shabbath 1 (Socino Press, London, 1938).
Baggs, C.M., The Ceremonies of Holy Week at the Vatican and StJohn's Lateran
(Rome, 1839).
Baldeschi, J., The Ceremonial according to the Roman Rite [ET J.D H Dale] (Charles
Dolman, London, 1873).
Ba..ring-Gould, S., The Lives of the Saints 3rd edition, (J.Hodges, London, 1878-82).
Baronius, C., Annales Ecclesiastici, [Ed. J.D.Mansi and D.Gregorius], 3~ vols (Lucca,
1738-59).
Bannister, H.M., 'Ordine "Amb'Osiano" per la settimana santa', Miscellanea Ceriani
(Ulrico Hoepli, Milan, 1910).
Barbier de Montault, X., 'Le Chandelier Pascal a Rome', Messager des Fideles, I
(1884-5), pp.73-75.
Batiffol, P., A History of the Roman Breviary, [ET E.M.Y.Bayley] (Longmans, Green,
London, 1912).
Bauldry, M., Manuale Sacrorum Caeremoniarum (Venice, 1762).
Baumstark, A., Comparative Liturgy, [ET F.L.Cross] (A.R.Mowbray & Co., London,
1958).
Beauduin, L., 'Le cierge pascal', La Maison-Dieu, 26 (1951) pp.23-27.
Beauduin, L., 'Liminaire', La Maison-Dieu, 45 (1956) pp.S-8.
Benoit-Castelli, G., 'Le Praeconium Paschale', Ephemerides Liturgicae, LXVII (1953),
pp.309-394.
Berliere, U., 'Le cierge pascal', Revue Benedictine, 5 (1888), pp.106-116.
Bernal, J., 'Vicisjtudes literarias e hist6ricas de la oraci6n hispana "Exaudi nos, Lumen
indeficiens" ',Miscellanea Liturgica in onore di Sua Eminenza if Cardinale Gia-
como Lercaro II (Rome,1967).
Bertoniere, G., The Historical Development of the Easter Vigil and Related Services in
the GrrekChurch, Orientalia Christiana Analecta 193 (Rome, 1972).
Bishop, E., Liturgica Historica (Oxford U.P., 1918).
Bisso, B., Hierurgia sive Rei Divini Peractio (Genoa, 1686).
Bolton, C.A., 'The New Fire', Irish Ecclesiastical Record, 59 (1947), pp.215-20.
Bonnet, H.N., 'Les Realisations', La Maison-Dieu, 26 (1956) pp.55-72.
Borella, P., II rito ambrosiano (Morcelliana, Brescia, 1964).
Bouyer, L., 'The Mystery of Easter' Downside Review, 66 (1947-48), pp.117-126 ..
Bouyer, L., The Paschal Mystery [ET Mary Benoit] (Allen and Unwin, London, 1951).
Bouyer, L., 'Saintes Vigiles', La Maison-Dieu, 26 (1951), pp.ll-22.
Bradshaw, H. and Wordsworth C., Statutes of Lincoln Cathedral (Cambridge U.P.,
1897).
Bradshaw, P.F., Daily Prayer in the Early Church, Alcuin Club Collections, No.63
(London, 1981).
Briggs, K.M., The Folklore of the Cotswolds (Batsford, London, 1974).
Bugnini, A., 'De Reformatione Liturgica Generali', Ephemerides Liturgicae, LXX
(1956), pp.414-29.
Bugnini, 0. and Braga C., 'Ordo Hebdomadae Sanctae Instauratus', Ephemerides
Liturgicae, LXX (1956), pp.81-228.
Bumpus, T.F., The Cathedrals and Churches of Belgium, (Werner Laurie, London,
1928).
Bumpus, T.F., The Cathedrals and Churches of Northern Italy, (Werner Laurie, Lon-
don, 1907).
Butler C., Benedictine Monachism, 2nd edition, (Longmans & Co., London, 1924),
Cabrol, F., Les Origines Liturgiques (Paris, 1906).
Cabrol, F. and Leclercq, H., Monumenta Ecclesiae Liturgica (Paris, 1900-02).
Cabrol, F. and Leclercq, H., Dictionnaire d' Archeologie Chretienne etde Liturgie (Pa-
ris, 1903-53)
Callewaert, C., 'De Paaschkandelaar', Liturgisch Parochieblad S.Pietersabdiy, 4
(1922), pp.l40-2.
384
Caimcross, H., Lamburn, E.C.R., Whatton, G.A.C., Ritual Notes, 8th edition (W.Knott
& Son, London, 1935).
Cameron, M.L., Old Etruria and Modern Tuscany (Methuen, London, 1909).
Canivez, J .M., 'Le Rite Cistercien', Ephemerides Liturgicae, LXIII ( 1949), pp.276-
311.
Canziani, E., Through the Apennines and the Lands of the Abruzzi (Heffer, Cambridge,
1928).
Capelle, B., 'L'"Exultet" Pascal Oeuvre de Saint Ambroise', in Miscellanea Giovanni
Marcati, Studi e Testi, 121 (Biblioteca Apostolica Vatican, Vatican), pp.219-46.
Capelle, B., 'La procession de Lumen Christi au Samedi-Saint', Revue Benedictine, 44
(1932), pp.105-19.
Capelle, B., 'Le rite des cinq grains d'encens', Les Questions Liturgiques et Parois-
siales, 17 (1932), pp.3-11.
Casel, 0., 'Zur Feuerweihe',Jahrbuchfur Liturgiewissenschaft, 2 (1922), pp.90-l.
Catholic Encyclopedia: see under C.G.Herbermann.
Chavasse, A., Le Sacramentaire Gelasien (Tournai,1958).
Colti, J.B., Dictionarium sacrorum rituum (Pistoria, 1772).
Cook, A.B., Zeus, 3 vols (Cambridge U.P., 1914-40).
Corsetti, B., Praxis Sacrorum Rituum ac Caeremoniarum (Brussels, 1656).
Cox, J.C., 'The Lights of a Medieval Church' in Curious Church Gleanings, edited by
W.Andrews (W.Andrews,& Co., Hull, 1896), pp.57-59.
Crawford, F.M., Ave, Roma Immortalis (Macmillan, London, 1903).
Crossley, F.H., The English Abbey, revised by B.Little (Batsford, London, 1962).
Corning, G.J., Hippolytus: A Text for Students, Grove Liturgical Studies, 8 (Grove
Books, Bramcote, Notts., 1979).
Dalmais, I.H., Introduction to the Liturgy, [ET R.Capel] (Chapman, London, 1961).
Davies, J.G., Holy Week, A Short History (Lutterworth Press, London, 1963).
De Bralion, N., Caeremoniale Canonicorum (Paris, 1657).
De Choin, Bishop Joly, Instructions sur le rituel de Toulon (Lyon, 1790).
De Conny, M., Ceremonial Romain, 3rd edition (Maison Mequiquoi, Paris, 1858).
De Grassis, P ., De Ceremoniis Cardinalium et Episcoporum in eorum diocesibus
(Rome, 1564).
De la Croix, J.F., Dictionnaire historique des cultes religieuses, 4 vols (Versailles,
1820).
De Moleon, - see Des Marettes.
Dendy, D.R., The Use of Lights in Christian Worship, Alcuin Club Collections, No. 41
(London, 1959). ·
Dennis, G., The Cities and Cemeteries of Etruria, 2 vols (Dent, London/ Dutton, New
York, 1907).
DeRubeis, B.M., De Vetustis Liturgicis Aliique Sacris Ritibus.. .in aliquibus Forojulen-
sis Provinciae Ecclesiis (Venice, 1754).
Deshusses, J., Le Sacramentaire Gregorien (Freiburg (Sw.), 1971).
Desideri, F., Praxis Sacrorum Rituum (Florence, 1739).
Des Marettes, J.B.Le B. (Lesieur de Moleon), Voyages Liturgiques de France, ou
Recherches faites en diverses villes du royaume (Paris, 1718).
DeVaux, R., Ancient Israel: Its Life and Institutions, [ET J.McHugh], 2nd edition
(Darton, Longman, Todd, London, 1980).
De Vert, C., Explication simple, litterale, et historique des Ceremonies de I' Eglise, 4
vols (Paris, 1709-13).
Diekmannp.L., The Masses of Holy Week and the Easter Vigil (Longmans, Green &
Co., London, New York, Toronto, 1957).
Dix, G., The Shape of the Liturgy, 2nd edition (Dacre Press, Westminster, 1947).
Doblado, L., Letter from Spain (Henry Colburn, London, 1822).
Dolger, F.J., 'Sol Salutis. Gebetund Gesang im christ lichen Alterturn', in Liturgieges-
chichtliche Forschungen, Vol. 4-5 (Miinster (W.F.), 1925), pp.364-5.
Dolger, F.J., 'Das Karsarnstag Feuer aus der Kristal Linse', Antike und Christentum, 6
(1940), pp.286-96,
Du Cange, C. du F., Glossiarium Mediae et lnfimae Latinitatis, New edition (Le Favre,
Niort, 1886).
385
Duchesne, L., Christian Worship: Its Origin arul Evolution, [ET M.L.McClure], 3rd
English edition (S.P.C.K., London, 1910).
Duchesne, L., Le Liber Pomificalis (Paris, 1886).
Dudden, F.H., The Life and Times of St Ambrose (Oxford U.P., 1935).
Durandus, G., Rationale Divinorum Officiorum (Lyon, 1562).
Duval, P.-M., Les Dieux de Ia Gaule, Mythes et Religions, 33 (Paris, 1957).
Dykmans, M., L'(Euvre de Patrizi Piccolomini, Vol.2, Studi e Testi, 294 (Biblioteca
Apostolica Vaticana, Vatican, 1982).
Dymond, R., 'The Parish of St Petrock', Report and Transactions of the Devonshire
Association, XIV (1882), pp.410-13.
Eisenhofer, L., Handbuch der Katholischen Liturgik, Vol.l (Freiburg im Br., 1932).
Eisenhofer, L. and Lechner, J., The Liturgy of the Roman Rite, [ET A.J. and J.F.Peeler]
(Herder, Freiburg/ Nelson, Edinburgh, London, 1961 ).
England, J., The Ceremonies of Holy Week in the Chapels of the Vatican (Rome, 1833).
Fabre, M.P., (editor), Liber Censuum, 2 vols (A.Fontemoign, Paris, 1905).
Farbridge, M.H., Studies in Biblical and Semitic Symbolism (New York, 1970).
Feasey, H.P., Ancient Holy Week Ceremonial (T.Baker, London, 1897).
Feasey, H.P., 'The Easter Sepulchre', American Ecclesiastical Review, 32 ( 1905),
pp.337-55.
Feasey, H.P., 'The Paschal Candle',American Ecclesiastical Review, 34 (1906),
pp.353-71.
Feasey, H.P., 'The Paschal Preconium', American Ecclesiastical Review, 36 ( 1907),
pp.249-61.
Fermor, P.L., Between the Woods arul the Water (Penguin Books, London, 1987).
Fletcher, B., A History of Architecture, 17th edition (The Athlone Press, London,
1961).
Forbes, S.R., Rambles in Rome (Thomas Nelson, London, 1892).
Forcadell, A., 'Ritus Carmelitarum Antiquae Observantiae', Ephemerides Liturgicae,
LXIV (1950), pp.5-52.
Fortescue, A., The Ceremonies of the Roman Rite Described, 2nd edition (Burns and
Oates, London, 1919).
Fortescue, A., The Mass, a Study of the Roman Liturgy (London, 1912).
Fortescue, A. and O'Connell, J.B., The Ceremonies of the Roman Rite Described, 4th
edition (Burns, Oates, and Washbourne, London, 1932).
edition (Burns, Oates, and Washbourne, London, 1937).
edition (Burns, Oates, and W ashbourne, London, 1960).
Fowler, J.T., Memorials of the Church of SS Peter and Wilfred, Ripon, Surtees Society
Publications, Vol.81 (Newcastle, 1882).
Fowler, J.T., Notes on the Rites of Durham, Surtees Society Publications, Vol.107
(Durham, London, Edinburgh, 1903).
Franklin, J.W., The Cathedrals of Italy (Batsford, London, 1958).
Franz, A., Die kirchlichen Benediktionen in Mittalalter, Vol.1 (Freiburg (im Br.), 1909),
reprinted by Akademische Druck-U. Verlagsanstalt, Graz, 1960.
Frazer, J.G., (editor), Apollodorus, Vol.2, Loeb Classical Library (Heinemann, 1921).
Frazer, J.G., The Fasti of Ovid, 5 vols (Macmillan, London, 1929).
Frazer, J.G., The Golden Bough, 13 vols (Macmillan, London, 1923).
Frazer, J.G., Myths of the Origin of Fire (Macmillan, London, 1930).
Freshfield, E., 'On the Parish Books of St Margaret-Lothbury, St Christopher-le-Stocks,
and St Bartholemew-by-the-Exchange', Archaeologia, 45 (1877), pp.57-126.
Fry, T., (editor), The Rule ofSt Benedict (Minnesota, 1981).
Gaillard, J., 'Les antiennes de Matines et de Laudes de Ia Semaine Sainte', Revue
Gregorienne, 28 (1949), pp.41-57.
Gaillard, J., 'Feu', Catholicisme, 4 (1956), col.1231-32.
Gaillard, J., 'The Great Night', [ET W.Busch], Worship, 28 (1953-54), pp.l68-175.
Gaillard, J., Holy Week and Easter, [ET W.Busch] (Collegeville, 1954).
Gasquet, F.A., Parish Life in Medieval England (Methuen, London, 1906).
Gattinari, T., Sacrorum Caeremoniarum Enchiridion (Parma, 1771).
Gavantus, B., Thesaurus Sacrorum Rituum, Vol.1 (Paris, 1652).
386
Gavantus, B., Thesaurus Sacrorum Rituum, revised by C.M.Merati (Venice, 1823).
Gerbert, M., Vetus Liturgica Alemannica (Saint Blaise, 1776).
Geyer, P., (editor), Itinera Hierosolymitana saeculi /Ill-VIII, Corpus Scriptorum Eccle-
siasticorum Latinorum, Vol. 39, (Vienna, 1898).
Goodrich-Freer, A., Inner Jerusalem (Constable & Co., London, 1904).
Gougard, L., 'De la veilleuse', Ephemerides Liturgicae, XLVI (1932), pp.435-8.
Gougard, L., Christianity in Celtic Lands [ET M.Joynt] (Sheed and Ward, London,
1932).
Grancolas, J., Commentarius Historicus in Breviarium Romanum (Venice, 1734).
Gray, D.C., (editor), Holy Week Services, 2nd edition (SPCK, London, 1983).
Grisar, H., Das Missale im Lichte romischer stadtgeschichte (Freiburg (im Br.), 1925).
Gueranger, P., The Liturgical Year: Passion Time and Holy Week, [ET L.Shepherd]
(James Duffy, Dublin,1870).
Guignard, P., Monuments primitives de Ia Regie Cistercienne (Dijon, 1864).
Guyet, C., Heortologia sive de festis propriis locorum et ecclesiarum (Venice, 1728).
Gy, P.-M., 'Semaine Sainte et triduum paschal', La Maison-Dieu, 41 (1955), pp.8-10.
Hale, J.R., (editor), The Italian Journal of Samuel Rogers (Faber, London, 1956).
Harbert, B., 'A Song at Twilight', Liturgy, Vol.7 No.6 (1983), pp.232-43.
Hare, A.J.C., Days near Rome (Kegan, Paul, London, 1907).
Hare, A.J.C., Walks in Rome, 2 vols, 5th edition (Daldy, Isbiter, London, 1876).
Hartel, G., Magni Felicis Ennodi opera omnia, C.S.E.L., 6, (Vienna, 1882).
Heintz, J., 'La celebration de la vigile pascale en France', La Maison-Dieu, 37 (1954),
pp.120-22.
Herrera, M.L., 'Some Typical Spanish Traditions', in Folklore Studies in the Twentieth
Century, edited by V.J.Newall (Woodbridge and Totowa, 1980).
Hesbert, R.-J., 'L'Antiphonale Missarum de l'ancien rit beneventan, Ephemerides
Liturgicae, LXI (1947), pp.l53-210.
Heuser, H.J., 'The Tenebrae and the New Fire', American Ecclesiastical Review, 36
(1907), pp.225-31.
Hillgarth, J.N., 'Visigothic Spain and Early Christian Ireland', Prodeedings of the Royal
Irish Academy, 62 (1962), No.6.
Houssaye, M., 'Les Ceremonies de la Semaine Sainte', Revue des Questions Histo-
riques, 28 (1878), pp.447-87.
Huglo, M., 'L'auteur de l'Exultet pascal', Vigiliae Christianae, 7 (1953), pp.79-88.
Hutton, E., Assisi and Umbria Revisited (Methuen, London, 1953).
Hutton, E., A Wayfarer in Unknown Tuscany (Methuen, London, 1909).
Jelmini, A., 'Die Feier der Osternacht in Der Schweiz', Liturgisches Jahrbuch, 3
(1953), pp.225-29.
Jones, C.P.M., A Manual for Holy Week (London, 1967).
Jones-Baker, D., The Folk/pre of Hertfordshire (Batsford, London, 1977).
Jounel, P., 'Les Missels Diocesains Fran~ais do 18e siecle', La Maison-Dieu, 141
(1980), pp.91-96.
Jounel, P., 'Le nouvel ordo de 1a semaine sainte', La Maison-Dieu, 45 (1956),
pp.16-35.
Jounel, P., 'La Semaine Sainte en France aux XVIr et XVIIT' siecles', La Maison-Dieu,
41 (1955), pp.132-52.
Juglar, J ., 'A propos de la vigile pascale', Ephemerides Liturgicae, LXV (1951 ),
pp.182-6.
Jungmann, J.A., The Early Liturgy, [ET F.A.Brunner] (Darton,Longman,Todd, Lon-
don, 1960).
Keating, G., The History of Ireland, [ET J.O'Mahoney] (New York, 1857).
Kellner, K.A.H., Heortology, [ET anon.], International Catholic Library (London,
1908).
King, A.A., Eucharistic Reservation in the Western Church (Mowbray, London, 1962).
King, A.A., 'Holy Week in Ambrosian Switzerland', Westminster Cathedral Chronicle,
1951, pp.96-99.
King, A.A., Liturgies of the Past (Longmans, Green, London, 1959).
King, A.A., Liturgies of the Primatial Sees (Longmans, Green, London, 1957).
King, A.A., Liturgies of the Religious Orders (Longmans, Green, London, 1956).
King, A.A., The Liturgy of the Roman Church (Longmans, Green, London, 1975
387
Kirk, G.S., The Nature ofGreekMyths (London, 1974).
Kirk, R.E.G., The Accounts of the Obedientiars of Abingdon Abbey, Camden Society
Publications, Vol. 51 (new series), 1892.
Kirschbaum, E., The Tombs ofSt Peter and St Paul, [ET J.Murray] (Seeker and War-
burg, London, 1959).
Kitzinger, E., Early Medieval Art, 3rd edition (British Museum Publications, London,
1983).
Klauser, T., A Short History of the Western Liturgy, [ET }.Halliburton] (Oxford U.P.,
London, 1969).
Krautheimer, R., Corbett, S., Frank, W., Corpus Basilicarum Christianarum Romae
(Vatican, 1967).
Latis, A., 'De Praeconio Paschali', Ephemerides Liturgicae, XVI (1902), pp.l23-32.
Le Braz, A., The Land of Pardons, 5th edition (Methuen, London, 1924).
Lees, D.N., Tuscan Feasts and Tuscan Friends (Chatto and Windus, London, 1907).
LeFebvre, G., (editor), Daily Missal (Liturgical Apostolate, Bruges & London, 1928).
Lerevre, P.F., 'La Liturgie de Premontre', Ephemerides Liturgicae, LXII (1948),
pp.195-229.
Lesage, R., Vestments and Church Furniture, [ET F.Mwphy] (Burns, Oates, London,
1960).
Le Vavasseur, L., Ceremonial selonle rit romain (d' apres J .Baldesclzi) (Paris, 1859).
Levi-Strauss, C., From Honey to Ashes, [ET J. & D.Weightman] (London, 1973).
The Liturgical Commission of the Church of England, Lent, Holy Week, Easter
(Church House Publishing, Cambridge U.P., S.P.C.K., 1986).
D.G.M., 'La 1iturgie de Naples au temps de St.Gregoire', Revue Benedictine, 8 (1891),
pp.481-93 and 529-37.
Mabillon, J., De Liturgia Gallicana (Paris, 1685).
McArthur, J.A., The Evolution of the Christian Year (London, 1953).
Macdonald, A.B., Christian Worship in the Primitive Church (T. & T.Clark,
Edinburgh, 1934).
Mackinnon, A.G., Rome (Seeley, London, nd).
Macri, D., Hierolexicon sive Sacrum Dictionarium (Rome, 1677).
Maigne D'Arnis, W.H., Lexicon Manuale ad scriptores mediae et inftmae latinitatis
(J.P.Migne, Paris, 1866).
Male, E., The Early Churches of Rome, [ET D.Buxton] (E.Benn, London, 1960).
Mallardo, D., 'La Pasqua e Ia settimana maggiore a Napoli', Ephemerides Liturgicae,
LXVI (1952), pp.3-36.
Mannix, D.P., Those About To Die (Hamilton & Co., London, 1958).
Marcora, C., La Vigilia Nella Liturgia, Archivio Ambrosiano, VI, (Milan, 1954).
Martene, E., De Antiquis Ecclesiae Ritibus, 4 vols (Venice, 1764).
Martimort, A.G., La documentationliturgique de Dom Edmond Martene, Studi e Tesi,
279 (Vatican, 1978).
Marwick, E.W., The Folklore of Orkney and Shetland (Batsford, London, 1975).
Masson, G., The Companion Guide to Rome (Collins, London, 1965).
Menard, H., Liber sacramentorum ... editus ex missali Ms Sancti Eligii, (Paris, 1642).
Merati, C.M., Ceremonies oftfte Church (Selections) (London, 1837).
Mercenier, E., La Pr£ere des Eglises de Rite Byzantin 2 vols, 2nd edition edited by
G.Bainbridge (Editions de Chevetogne, 1953).
Mohrmann, C., 'Exultent divina mysteria ',Ephemerides Liturgicae, LXVI ( 1952),
pp.274-81.
Moreton, B., The Eighth-Century Gelasian Sacramentary (Oxford U.P., 1976).
Morlot, F., 'La Nuit Pascale: Enquete',La Maison-Dieu, 67 (1961), pp.ll0-158.
Nicholls, J.G., The Diary of Henry Machin, Camden Society Publications, Vol. 42 (old
series), 1848.
Novarinus, A., Sacro-prophana (Lyon, 1635).
O'Connell, J.B., 'The New Order of Holy Week ll', Clergy Review, 41 (1956), p.131.
O'Dea, W.T., 'Artificial lighting prior to 1800', Folk-Lore, LX1I (June 1951) pp.314-7.
Oeconomia domus domini seu Liber de Sacramentorum administratione .. .in urbe Tho-
losa (Paris, 1553).
O'Kane, J., Notes on the Rubrics of the Roman Ritual, 4th edition (J.Duffy & Sons,
Dublin, 1868).
388
O'Loan, D., The Ceremonies of some Ecclesiastical Functions (Dublin, nd).
O'Shea, W.J., The Meaning of Holy Week (Collegeville, Minnesota, 1958).
Pamelius, J., Liturgica Latinorum (Cologne, 1571).
Paschal, J.-B.-E., La Liturgie Catholique (J.-P.Migne, Paris, 1844).
Pascher, J., 'Die Osterkerze', Liturgisches Jahrbuch, 2 (1952), pp.132-34.
Passmore, T.H., In Further Ardenne (J.M.Dent, London, 1905).
Peacock, E., English Church Furniture (John C.Hotten, London, 1866).
Peacock, E., 'Hearse: how a word has changed its meaning' in Curious Church Glean-
ings edited by W.Andrews (W.Andrews & Co., Hull, 1896), pp.214-17.
Pellicia. A.A., Polity o.fthe Christian Church [ET J.C.Bellett] (Masters & Co., London,
1883.
Philippeau, H.-R., 'Note sur !'utilisation liturgique de cierge pascal' La Maison-Dieu
31 (1952), pp.145-49.
Pinell, J., 'La Benedicci6 Del Ciri Pasqua! i Els Seus Textos', Liturgia, II (Montserrat,
1958), pp.85-100.
Pinell, J., 'Vestigis Del Lucemari a Occident', Liturgia, I (Montserrat, 1956), pp.91-
149.
Podhradsky, G., A New Dictionary of the Liturgy, edited by L.Sheppard (Geoffrey
Chapman, London, 1967).
Powicke, F.M. and Cheney, C.R., Councils and Synods relating to the English Church,
2 vols (Oxford U.P., 1964).
Pugin, A.W., A Glossary of Ecclesiastical Ornament and Costume (C. Tilt, London,
1844).
Rabotin, H., 'Les Grains d'Encens du Cierge Pascal',La Vie et Les Arts Liturgiques, 7
(1920-21), pp.221-225.
Rahner, H., Greek Myths and Christian Ritual, [ET M.Benoit] (Bums, Oates, London,
1963.
Raine, J., The Rites of Durham, Surtees Society Publications, No.15 (London and
Edinburgh, 1842.
Rees, A.D. and Rees, B.R., Celtic Heritage (Thames and Hudson, London, 1961).
Rees, U., The Cartulary ofHaughmond Abbey (Shropshire Archaeological Society and
University of Wales Press, Cardiff, 1985).
Reinhold, H., 'Lugano and Holy Week', Worship, 28 (1953-54), pp.426-432.
Righetti, M., Manuale di Storia Liturgia, 2 vols, 2nd ed. (Ancora/Milan, 1955).
Rock, D., The Church of our Fathers, 4 vols (John Murray, London, 1905).
Rock, D., Hierurgia, 2nd edition (London, 1851).
Rosenwein, B.H., Rhinoceros Bound: Cluny in the Tenth Century (Philadelphia, 1982).
Ross, A., The Folklore of the Scottish Islands (Batsford, London, 1974).
Sacrarum caeremoniarum sive rituum ecclesiasticorum sanctae Romanae ecclesiae
libri tres (Venice, 1582).
Schelstrate, E., Antiquitas Ecclesiae (Rome, 1697).
Schmidt, H.A.P., Hebdomada Sancta 2 vols (Herder, Rome, Freiburg im Br., Barce-
lona, 1956-57).
Schmitz, P., L' Histoire de I' Ordre de Saint-Benoit, 7 vols, Les Editions de Maredsous
(Maredsous, 1942-56).
Schoenbechler, R., 'The New Insignia on the Easter Candle', American Ecclesiastical
Review, 128 (1953), pp.273-80.
Scullard, H.H., Festivals and Ceremonies of the Roman Republic (London, 1981).
Seasoltz, R.K., New Liturgy, New Laws (Minnesota, 1980).
Seeck, 0., (editor), Notitia Dignitatum (Berlin, 1876).
Shepherd, M.H., The Paschal Liturgy and the Apocalypse (Lutterworth Press, London,
1960).
Shutt, F.J., 'Triduum Sacrum', Clergy Review 36, (1951), pp.20-27.
Sladden, D. and Lorimer, N., Queer Things about Sicily (Treheme, London, 1905).
Smith, L.M., Cluny in the Eleventh and Twelfth Centuries (P.Allan, London, 1930).
Stevenson, K., 'The Ceremonies of Light: Their Shape and Function in the Paschal
Vigil Liturgy', Ephemerides Liturgicae, XCIX (Mar/Apr. 1985), pp.170-185.
Stokes, W., Lives of the Saints from the Book of Lismore (Oxford, 1890).
Sullivan, J., 'Setting the Theme: Subliminal Influence of Liturgy "A School of
Prayer"', Spiritual Life, Vol.27 No.I (1981), pp.14-23.
389
Thibaut, J.B., Ordres des offices de La semaine sainte aJerusalem du ive au xe siecle
(Paris, 1929).
Thomassin, L., De Dierum Festorum Celebratione (Venice, 1729).
Thompson, S., A Motif-Index of Folk-Literature, 6 vols (Copenhagen, 1955-57).
Thomson, P.D., Parish and Parish Church (T.Nelson, London, 1948).
Thurston, H., 'The Early Cultus of the Blessed Sacrament', The Month, Apri11907,
pp.382-88.
Thurston, H., 'The Easter Sepulchre', The Month, Apri11903, pp.404-414.
Thurston, H., 'The Exultet and the Paschal Candle', The Month, April1896,
pp.502-18.
Thurston, H., Holy Saturday (Catholic Truth Society, London, 1937).
Thurston, H., Lent and Holy Week (Longmans, London, 1904).
Thurston, H., 'The Sign of the Cross', The Month December, 1911, pp.586-602.
Thurston, H., Tenebrae (Catholic Truth Society, London, 1931).
Thurston, H., 'Votive Candles', The Month, August 1932, pp.141-151.
Tyrer, J.W., A Historical Survey of Holy Week, Alcuin Club Collections, No.29 (SPCK,
London, 1932).
Ughelli, F., Italia Sacra, 10 vols (Venice, 1717-22).
Van Der Stappen, J .F., Sacra Liturgia (Mechlin, 1906).
Van Dijk, S.J.P., Sources of the Modern Roman Liturgy, 2 vols (Leiden, 1963).
Van Doren, R., 'La ceremonie do feu nouveau au Samedi-Saint', Les Questions Litur-
giques et Paroissiales, 13 (1928), pp.74-78.
Van Doren, R., 'Le cierge pascal', Les Questions Liturgiques et Paroissiales, 14
(1929), pp.65-77.
Van Gennep, A., Manuel de Folklorefranf:ais contemporain (Paris, 1946-58).
Varley, D., Seven- The Number of Creation (London, 1976).
Viale, E., 'La benediction du feu nouveau', La Maison-Dieu, 26 (1951), pp.41-52.
Vinitor, G., Compendium Sacrorum Rituum et Caeremoniarum (Cologne, 1685).
Vogel, C., Introduction aux sources del' histoire du culte chretien au moyen age (Cen-
tro Italiano di Studi sull' Alto Medioevo, Spoleto, 1966).
Wade-Evans, A.W., Life ofSt.David, (Rhygyfarch), Translations of Christian Litera-
ture, Series 5, (London, 1923).
Walsh, M.J., 'Notes on Fire-Lighting Ceremonies', Folk-Lore, LVIII (June 1947),
pp.282-4.
Warren, F.E., The Liturgy and Ritual of the Celtic Church (Oxford, 1881).
Warren, F.E., The Sarum Missal, Part 1, Library of Liturgiology and Ecclesiology,
Vol.8, (London, 1911).
Weiser, F.X., A Handbook of Christian Feasts and Customs (New York, 1958).
Wilkins, D., Concilia Magna Britanniae et Hiberniae, 4 vols (London, 1737).
Wilkinson, J., Egeria' s Travels (SPCK, London, 1971).
Wiseman, N., Four Lectures on the Offices and Ceremonies of Holy Week (London,
1839).
Wordsworth, C., Ceremonies and Processions of the Cathedral Church of Salisbury
(Cambridge U.P., 1901).
Wordsworth, C., Medieval Services in England (T.Baker, London, 1898).
Wright, T., Early Travels in Palestine, (Bohn's Antiquarian Library, London, 1848).
Wuest, J., Matters Liturgical, [ET of Collectio Rerum Liturgicarum by T.W.Mullaney,
lOth edition (New York and Cincinnati, 1959).
Yamold, E., The Awe-Inspiring Rites of Initiation (St Paul Publications, Slough, 1972).
Zamecki, D., The Monastic Achievement (London, 1972).
Zazzera, P., SS. Ecclesiae Rituum Divinorumque Officiorum Explicatio (Rome, 1784).
390
(i) Index of medieval writers and documents
Peter Abelard, 91 Ordo 17, 4, 9, 15, 100-1,331

Alcuin, 24, 27, 34, 35, 38, 40,65-6,74-5,80-1, Ordo 19, 213
82, 102, 108, 111,131, 133, 140, 143, 157, Ordo 23, 9, 10, 13, 14, 15, 101, 322, 326
190,193,205,208,243,249,293,312,315, Ordo 24, 322-3
322,337,346,358,359 Ordo 25, 371
Amalarius, (i) Liber Offidalis, 34, 35, 39-40, 52, Ordo 26, 14, 18-19, 20, 22, 23, 36, 37, 64, 65, 72,
64,72,85,100, 109,229,274,306,308,355, 80-1,85,93, 101, 108, 111, 131-2, 140-1, 153,
371. (ii) Liber de Ordine Antiplwnarii, 11-13, 155,157,189,190,193,205,208,241,251,
99,109,177-8,186,210,335,353,359. 293,309,322,344,346,365,372
Ambrose, 1 Ordo 27, 22, 23, 65, 72, 108, 312
Angouleme Sacramentary (GeAng),100-1, 241, Ordo 28, 17, 22, 37, 65,80-1,82, 85, 91, 92, 108,
246,259,290,293,331 123, 132, 135, 140-1, 153, 157, 185, 190, 193,
Augustine, 1 219,241,251,312,332,344,363-4
Aurelian, 4 Ordo 29, 18, 20, 22, 24, 34, 37, 64,68-73,74,
Autun Sacramentary, (GePh), 289,358 80-1, 85, 104-5, 109, 111, 132, 135, 140. 153,
Jobn of Avrancbes, 30, 36, 38, 40, 59, 85, 104, 157,189,190,193,208,241,251,292,293,
115, 134, 135, 147, 150, 160, 185-6, 193-4, 309,310,336,344,346
197, 199-200 Ordo 30A, 15, 16, 92, 100-1, 241, 246, 259, 289,
Bangor, Antipbonary of, 288-9 331
G.Barozzi, Pontifical of, 181 Ordo 30B, 6, 13, 16, 37, 87, 92, 100-1, 322-3,
Bede,162, 166, 175,276-7,279 326, 344, 345-6
Jobn Be1etb, 35, 40, 41, 43, 44, 45, 57, 82, 86, 91, Ordo 31, 22, 23, 37, 64, 74,80-1, 85, 92, 108,
112, 153, 159, 163, 165, 167, 168, 169, 186, 132,135,140,153,190,208,219,241,309,
269,279,337 332,363-4
Beroldus, 106,243,244,274,295,320,367 Ordo 32, 24, 34, 38, 70, 72, 80, 85, 185, 208, 241,
Bindo Fesulani, 139, 181, 190, 193, 216, 294 244
Bruno of Segni, 44 Ordo I, 313,344
C.A. 1706,139,181,190,193,202,212,216, Ordo XI, 112, 139
294,296 Ordo XII, 193, 216
Jobn Cassian, 2 Ordo XIV, 20, 181, 190, 193, 202, 214, 216,
Corbie Sacramentary, 115, 130, 162, 165, 168, 242-3, 294, 361
220,243,249,255,257,272,275,278,294, Ordo XV, 20, 181, 190, 193, 216, 242-3
332,333, Ordo Albini, 129, 139, 145, 193, 194
Drepanius Horus, 288 Ordo of Benedict, 181
Durandus,32,40,41,43,44,45,57,64,83,87, Ordo of Cencius, 181
89-90,109,159,169,187,195-6,229,269-70, ordo of Corbie (M 1145), 108, 114, 132, 140, 208
282,307,308,337,338,339 PaulinusofMilan,3
Egbert Pontifical, 255, 272 Poitiers, 20, 22, 23, 24, 34, 37, 64, 68, 70, 74,
Hodoardus, 305 80-1, 82, 85, 105, 111, 131-2, 135, 164-5, 185,
Ge1asian Sacramentary (GeV), 100-1, 124, 125, 189,190,193,195,196-7,207,250,251,256,
241,246,258,259,289,293,331 290,293,312,322,327-8,330,337
Gellone Sacramentary, (GeGe), 100-1, 247,290, Pontifical of Durand (PGD), 117, 130, 139, 145,
334,337,358 181,191,197,216,294
Gregorian Sacramentary, 289 Pontifical of the Roman Curia (PRC), 117, 122,
Gregocy of Tours, 4, 5, 142, 254, 358 139
Haymo ofFaversbam, (i) OB, 43, 85, 87. (ii) OM, Pontificale Romano-Germanicum (PRG), 20, 23,
112,139,148,191,193,202,204,216,266, 24,35,36,37,39-40,52,65, 70, 72, 73, 74,
295,317,318 80-1,82,85, 95, 100, 102-4, 106, 108, 111,
Honorius of Autun, (i) Gemma Animae, 40, 41, 117,123,124,125,126,129,130,131-2,133,
77,92, 165,169,186-7,279,293,298,304, 134, 135, 138, 140, 143, 157, 165, 175,
306-7,377. (ii)Speculum, 45, 161, 186,329 179-80,185,190,193,208,209,211,220,
Hugh of St Victor, 89, 161, 165, 169, 306 230,241,243,247,248,251-2,290,293,312,
~c.40, 77,104,108,110,112,115,127, 315,318,322,323,328,337,343,346,355,
132,136,148,190,191,194,208-9,211,213, 358,371
250,256-7,260,295,296,311-12,318,333, Pontificale Romanum (PR XI[), 104, 106, 112,
353 117,121,124,127,129,138-9,146,179,181,
Lateran Misal, 181-2, 319 193,201-2,203,214-15,216,217,229,241,
Lateran Ordo (OEL), 42, 67, 112, 139, 181, 203-4, 242-3,248,249,258,263,266,275,278,294,
216,248,249,275,278,294,312,318,319 296,312,319,337,338,362-4
Leofric Missal, 121, 126 Prague Sacramentary, 100-1, 117, 122, 241, 246,
Liber Ordinum, 130, 366 247,259,293,331
Magdalen College Pontifical, 115,211 Regu/aris Concordia, 85, 91, 93, 110, 111, 130,
Micrologus, 293, 298, 303 131-2,147,149,189,192-3,195,197,243,
Missale Mixtum, 128, 264, 338, 367 293,312,320,333,339,353
Nomasticon Cisterciense, 112, 114, 130, 155,262, Ritual of Rbeinau, 118
280 Robert of Jumieges, Missal of, 285, 360
Ordo4, 344 Robert Paululus, 187, 306
Ordo 11, 322, 328, 358 Rosslyn Missal, 312
Ordo 16, 9, 322
391
Rupert ofDeutz, 16, 45, 77, 108, 118, 134, 146, Sicardus,35,40,44,57, 112,146,155,159,165,
159,160,165,167,191,193,197,199-200, 169,186,215,229,269,279,280,302,303,
211,269,306-7,341 306,308,341
Sacramentarium Vetus, 165 Wulfstan Collectar, 112
St Remigius' Abbey, Sacramentary of, 330-2
(ii) General index - mainly personal and geographical
Abo, 104,113,116,122,124,125,211,290 Basel, 113,121,136,157,211,247,264,288

Abruzzi, 217 Bayeux, 120, 129, 138, 147, 152, 191, 193-4,
Accendite, 310-14,318-20,359 195-7,202,208,292,299,317
Agen, 129,152,201 Bayonne, 110,131,137,152,196,201,295
Agnus Dei, 105,303 Bazas, 312-13
Aix, (Aachen), 152, 158, 342 beams,25,27,28,29
Albi, 108,165,293,296,301 Beauvais, 118, 126, 130
Ales, 43,61 Bec,40,53,118, 147,204,209
Altar lights, 31-3, 67, 86-87, 311, 319 Bede, 276-8
Amalarius, 11-13, 99 beeswax, 308
ambo, 242-4 Bel, 184
Ambrose, 1,3,288,310 Belgium, 184
Ambrosian rite, 6, 7, 8, 10, 136, 138, 220, 238, Betley, 129, 152
252,263,274,326,355,358 Beltane fires, 184
~ens,39,113, 146,152,155,157,190,243, Benedict, St, 2, 20, 93, 213
264,282,291,299,317,347 Benedict of Aniane, 289
Anastasis,1, 183, 324-5 Benedict XIV, 89
Anderlecht, 157 Benedictines,20,21, 112,118,194,213,235,
Angers, 44, 80-1, 89, 113, 118, 151, 152, 243, 249,260,278,320,333,335,361
268,299,313,356 Benedictus, 85-88, 103, 67
Annecy, 158,285 Beneventum, 105,107,112,117-8,154,157,193,
Anticipation,20,21, 131,136,151,156,326 210-12,214-5,217,227,242,251,260-1,285,
Antioch, 230 288,296,313
Antonius of Placentia, 325 berakah, 126,291
Apamea, 112,139,146,202,214,243,275,278, Bergamo, 83
294 beryl, 128, 161-3, 165
Apostolic Tradition (Hippolytus), 287-8,325-7 Besan~on, 43,. 58, 76, 80-1, 86, 118, 128, 136,
Aquileia, 113, 124, 157, 193, 201-2, 206, 209, 151,157,190-1,264,292,297,312,317,356
211,214,220,229,243,266,331,339,342 Bianco da Siena, 187
Arbuthnott, 113 Biasca, 244, 359
Arculf, 325 Birmingham Oratory, 19, 61·
Aries, 130, 193, 197, 312-13 Boniface, 9, 99, 170, 175, 184
Arras,120, 137,152,158 Boulogne, 118
Arras, St Vedast, 44, 58, 62, 112, 113, 115, 130, Bourges, 125,157,219,237,264,291,292,299,
146,190,194,209,263,266,312 304,314
arundo serpentina, 194-200 Braga, 43, 110, 118, 121, 122, 132, 136, 150, 157,
Ascension Day, 300-01, 304, 308 195-7,202,204,205,260-1,263,300,317,338,
aspersion, 127-8, 263 339,355,361
Auch, 108,118,121,128,136,146,185,196-7, Bremen, 113, 121,211,262
201-2,206-7,214,268,297 Breslau, 113, 114-6, 122, 125, 157, 197, 211,
Augsburg, 331 260-1
Augustine, 1,2,229,288,307 Bressanone, 123
Augustus, 344 Brittany, 184
Austin Canons, 197,202,206,214 Bruges, 286
Austin Friars, 121, 146, 147, 157, 191, 193,201-2, Brussels, St Gudila, 281
207 Burgos, 113, 115, 118, 120, 165,209,317,354
Autun, 115,121,147,151,152,157,201,243, Bursfeld, 113, 118, 150, 299, 301
299, 313, 356 Bury St Edmunds, 235
Auxenre,21,61,80-1, 151,183,211,312,314 Cabaret, J., 167
AveUana, 120, 125, 132, 165 caduceus, 198
Avignon, 32 Caen, St Stephen's Monastery, 333
Avranches, 115 Cahors, 115,118,120,137,146,157,202,214-5,
banners, 114, Bayonne, 118, 132, 152, 157 216,243,301
Baptism, 306, 335 Camaldolese, 43, 84, 86, 115, 120, 136, 157, 193,
baptismal lamp, 327 202,206-7,214,243,312-13,317-18,355
Bari, 205, 235 Cambrai, 190-1,261,353
Baddng,23,40, 77,113,118,125,136,162,164, candles, size of, 234-9, 323, 327-30
165,194,202,204,210-11,232,235,238,249, Canterbury, Christ Church, 63
256-7,280,339,354,356
392
Canterbury, St Augustine, 23, 40, 62, 83, 85, 11o, Egeria, 230, 324
119,121,132,189,194,299,354 Eichstadt, 123
Capua, 286, 304 Elipandus, 213
Capu~,61, 147,148,157,201,206,214,317, England,90, 106,117,184,233
320 Ennodius, 226, 230, 287-8, 306-7
Carcassonne, 115, 121, 147, 151, 152, 157, 201, epact, 279-80
243,257,264,293,312,314-5 Epworth, 240
Carmelites, 45, 81, 113, 115, 120, 136, ISS, 157, Essen,256-7,267,275-6,280,283
202,204,206,208-9,300,317 Esztergom, 113, 122, 137, 157,211,361
Carthusians, 347 Eveshrun,l10, 115,121,127,130,131,132,
Cassian, John, 2 201-2,204,209,211,257,261,263,273
Centndltruy,210,215-7,219,229,284 Evreux,121 136,147,155,157,186,202,206,
cereus minor, 268, 291, 320, 335, 339, 359 214,299,317,347
Chalons-sur-Mame, 194,213,281,317,370 Exeter, 23, 39, 40, 41, 77, 85, 112, 149-50,
Chalons-sur-Saone, 42, 43 210-11,262,268,299-300,317
Charlemagne, 14 Exeter, Synod of, 62
Charles V, 244 Exodus, see Moses.
Charleston, SC, 232 Exultet, 226, 285, 287-92
charta, 279-83,369-71 Exultet rolls, 195, 198, 203, 205, 235, 242, 284
Chester, 85, 112 Fabian, 344
Chezal-Benoit, 42, 113 Fana,22,30,42,43,46,48,52,54,56,60, 75,
chrism, 248,251-4,265 80-1,85,88,110,111,114-5,132,146,191,
Chrodegang,342 209,280,312,356
Chur, 331 Fire, reservation of, 9, 11
church lights, 318 fire-drill, 153, 181
Church of England, 265, 309 fire festivals, 123
Circus Maximus, 54 fire symbolism, 186-7
Cistercians, 21, 45, 88, 112, 114, 118, 121, 127, Fleury,40,43,52,53, 110,111,148,162,165,
128,130,147,150,155,157,201-2,207, 168, 190-1 197, 232, 280-1
213-4, 239, 241, 243, 262, 280-2, 298-9, 301, flint, 153-7, 159-60
317,354,361 Florence, 157,238,356
Clermont-Feraod, 300 flowers, 148, 157, 192, 284-5
cloister, 147-8 font, blessing of, 15
Clotaire I, 4 Fontanelles, 112
Clovis, 254 Fontevrault, 118
Cluny,21,42,43,52,53,56, 75, 76,80-1,85, formulas, 116-26, 264, 348-52
109, 110, 111, 114-5, 118,125, 130, 132,145, Franciscans, 21, 32, 43, 112, 157
157,161,162,163,191,205,208-9,295,312 Freising, 113, 116, 121, 136, 155, 165, 211
Coimbra, 317 Frejus, 147,157,214-5
Cologne, 113,116,120,155,211,262,266,317 Fruttuaria, 22, 40, 46, 47, 59, 85, 110, 115, 121,
colours, 284 126,191,209,262,295,312,353
Constance, 192 Fulda, 120, 288
Constantine, 230 Gaeta, 235
Constantinople, 230 Gallican rites, passim, and esp.331-3.
Corbie, 27, 42, 43, 53, 111,114, 116, 130,165, Gelasius, 2
197,202,243,248-9,257,295,333 Gembloux, 40, 46, 47, 59, 80, 85, 87, 115, 132,
Cordoba, 113, 120, 138,266,314-5 146,191,209,232,243,249,256-7,280,312,
Cosenza, 113, 121, 150, 157, 193, 202, 206, 214, 353
317,354 German monasteries, 110, 132, 197, 249, 278, 312
Cotswolds, 184 Germany, 156
Coutances,31,36,43, 76,80-1,86,118,147,152, Getbsemaoe, 97
157,197,204,243,292,299,.317,354 Gilbertines,40,51,60, 77,85,89, 112,147,165,
crucifix, 286 194,204,210-11,238,268,298-9,339
Cyril of Jerusalem, 324 gradual psrums, 52-57, 76
Deus mundi Creator, 103, 104, 119, 124, 126, Graodmont, 42, 43
221, 289-90,334 Great Budworth, 51
Digne, 129,158,201,209 Gregory I, 175
Dijon, 152 Gregory of Tours, 3, 4, 5, 7, 142, 254
Dijon, St-B~nigne, 42, 43, 48, 49, 52, 53, 60, 110, Gregory VII, 42, 182
111, 115, 118,127, 128, 132,161, 162, 165, hearse,28-32,39,46-63, 189
191,257,265 hekal, 58
Dixmude, 284 honorific lights, 9, 15
Dominicans,21,32,33, 76,87, 113,150,155-6, honorific lights, 9, 15, 327, 329,336, 344
157,201,216,243,244,281-3,315,371 Halberstadt, 116, 157, 211
Downside Abbey, 61 Hamburg, 113, 122, 211, 261
Drepanius Florus, 288 Haughmond Abbey, 323
Durhrun,112, 115,121,127,136,190,194,202, Hephaestus, 184
204,205,210-11,236,237,238,257,261, Hereford, 40, 41, 112, 120, 194, 197, 208, 296,
272,301,312,354,357 312,354
East Cheap, 240 Hertfotdshire, 285
Easter candle, 219-22, Pt IV passim Hildersheim, 113, 118, 125, 138, 157,211,354
Echtemach, 213 Holy Sepulchre, 112, 183, 199, 324-5, 330
393
Hugh (Abbot), 42 Lyre, 112,118,147,209
Hungary, 354 Magdeburg,265,353
Ignis Creator lgneus, 288 Mrunz,9,38,104, 113,116,122,123,124,133,
incensation, 263, 271-4 137,157,147,171,211,243,261,290,337,
incense, 124-5, 127-8, 258-60, 269 343
indictio, 279-80 Mantua, 228
lnoocent m, 43 Marinianus, 230
Inventor rutili, 204, 209-11, 252-4, 365-6 Marseille, 136,193,197,213-14,215,257,261
heland, 112,120,146,157,184,204,211 Martin V, 32
Isemia, 217 Martyrium, 183
Jacobites, 230 Maxima redemptionis, 20,91
Jerome, 228-9, 289 Maximilian,236
Jerusalem, l, 96, 97, 112,122, 151, 172,183,199, Nleaux, 118,138,151,157,257,264,293,297,
230,314,324-6,366-7 299,317,319
John, Duke of Brabant, 281 Meissen, 116,211,262
Judas (false candle), 232,237,239-40 Melk, 113,121,138,148,157,191,193
Judas candle, 90 Mende, 115,121,147,151,152,157,243,312,
Judas hand, 82 317,354,356
Junrieges,257,261 menorah, 51, 58,286, 344,~57
LaRochelle, 115,121,147,151,157,243,267, Meu, 121,136,147,151,157,243,299,313,342
297,299,313 Meu, St Vincent, 257,261
Lanfranc,53 Milan, 1, 3, 10, 105, 107, 112, 118-9, 120, 121-2,
Langres, 21, 113, 121, 136 127,128,130,150-1,155,157,177,179,183,
Laon,30,42,43,53,112, 130,146,157,284,299, 201,210,219,25,243,244,249,260-1,265,
320 268-9,284-5,287-9,291-2,295,296-7,300,
Lateran, StJohn, 9, 11, 42, 44,101, 107, 145, 148, 321,331,338,342,367
150,171-3,176-8,207,214,230,235,237, Minden, 113,120,211
239,257,266,278,313,325,344,371 Minor Rogation Days, 13
Latin Patriarchate, 112 Mirabella, 235
Laval, 158 Monte Cassino, 27, 31, 36, 37, 42, 67, 108, 113,
Le Mans, 44, 152, 190-1, 211, 296 121,138,148,157,193,198,202,204,206,
Le Puy, 120, 128, 147, 155, 157, 210-11, 243, 234-5,238,278,293,296,305,338
257,264,267,291,293 Monte Cassino A and B, 362-4
Leau,237 Montpellier, 152
Leidrad, 178 Moses, 122-3,159,187,199-200,226,289,306,
Leidrad, 343 311
lens, symbolism of, 169 Mozarabic rite, 8, 103, 106, 112, 117-8, 120, 123,
lens, use of, 161-9 126,128,155,179,183,201,210-11,219,
Leo I, 2 225,244,264,284-5,291,296,303,314-5,
Leon, 13 326, 355, 366-7
Leo IV, 178-80 myrrh, 269
Leon, (see also Mozarabic rite), 112, 128, 150, Nancy, 152
151,157,248-9,261,275-6 Nantes, 120, 121, 147, 157, 193,202,206-7,214,
Lesnes,112, 128,130,155 267,297,299,301
liege,45, 113,125,128,157,211,313,354 Nap1es,147,219,238,243,296,299,307
liege,StJames,44 Naples, St Stephen, 234-5
light, diSposition of, 24-33 Narbonne, 113,118,125,149,150,165,257,
light, provision of, 22-3 312-3
lights, extinction of, 64-84 Ner Tamid, 172
lights, numbers at Tenebrae,29, 30, 34-45 Nevers, 44, 152
lights at tombs, 142, new fire, symbolism, 145-6, 186-7
lirnoges,20, 121,152,157,244,288,317 Nidaros,40, 77,80-1,85,110,112,120,128,130,
Limoges, St-Martial, 167 136,190-1,244,247,262,264,280,353,359
lincoln,40,83,235 Nieuport, 284
lincoln Cathedral Statutes, 62 Nile, 226
limeux, 120,125,147,157,296,299,313 Nimes, 152, 315
London, St-Mary-at-Hill, 240 North Africa, 2, 229
London, St Christopher, 240 North Italy, 225,227,228-9
London Oratory, 19 Norwich, 30, 40, 73, 77, 78, 80-1, 85, 88, 110,
LUbeck, 137 115,127,132,194,209,211,236,257,261,
Lucernarium, 116, 119, 120, 122, 174, 189,219, 272,292,
224-5,229,253,287,297,306,324-7,330, Noyon, 113
363-6, 366-7 Omissions, 9, 341
Lu~on, 155, 121, 157,292,317 oratorium, 140-45
Lumen Christi, 203-4,206-7,210-18,251,308, Orleans, 152
313,366 Orleans, St-Agnan, 40, 87,339
Lund,113,211,353 Orthodox Church, 205, 244, 246, 276
Lyon, 110,114,120,121,149,150,152,157, Osma, 113, 121, 137, 165,312-3
178,201-2,204,206,207,232,237,239,251, Palencia, 113, 115, 122, 123, 128, 148, 157, 194,
262,265,268,285,292,293,299-300,302, 209,220,295,317,319,354
312-3,317,320-1,347,354,355,356 Palermo, 118, 356
Lyon, StMartin d'Ainay, 118, 155, 161, 165,353 Palestine, 28
394
Pamiers, 152, 291 Rome, StPeter's, 354
Paris,39,42,43,44,58,86, 115,121,128,152, Rouen,21,30,38,40,60,85,87, 109,113,118,
157,232,238,239,243,264,292,295,301, 121, 130, 136, 147, 151, 157, 183, 191, 194,
313,316,317 195-7,208-9,236,245,248,256-7,281,283,
Paris, Church of the Virgins, 21 292,295,298-9,301,315,317,353,370
Paris, Royal Chapel, 59, 121, 147,243,317 Rufinus, 305
Paris, St Victor, 21 sacred trees, 275
parrocia, 229 Saint Arnaud, 13
Paschal II, 182 Saint John Passion, 15, 196
Pasque flower, 285 Saint Maur, Monastery of, 21
Passau, 113, 116, 120, 123, 157, 211, 261, 314-5, Saint Peter's tomb, 25
320 Saint Vanoe, Monastery of, 21
Pau1inus, (of Milan), 2, 3, 254 Saintes, 26i, 274
Paulinus, (of Nola), 254, 284 Salisbury,28,30,40,41, 77,80-1,85,112,115,
Pavia, 230 121, 127, 136, 149-50, 157, 183, 194, 195,
Pedilavium, 325 197-8,204,207,211,236,238,243,244,268,
Perigueux, 120,147,155,157,210-11,243,257, 299,300,317,319,339,359-60
264,291,293,313,319,356 Salzburg, 108, 111, 116, 124, 126, 132, 157, 185,
Piacenza, 228-9 193,211,230,279,290,313,319,322,354,
Pius IX, 232 355
Poitiers, 1,3,4,5,6,10,23,25,51,56,64,65, Saru:tissimum light, 83-4, 142, 320
113,118,120,136,146,157,202,206,213-4, Saragossa, 113, 120, 150, 157
246,255,299,312,317,354,359 Saturday of Annunciation, 230
Poland, 147,202,206,238,239,296,317,320, Scandinavia, 184
361 Scotland, 184
Pontivy,257,261 Sees, 115,121,147,155,157,243,257,299
portraits, 283 Senlis,40
Portsmouth, 357 Seos, 113,118,137,148,165,210-11,262,300
Portugal, 301 Serbia, 184
Praesidius, 228 seven lamps, 10, 15-19, 22-4, 67, 241, 344-6
Prague, 116,155,165,211,264,280 Seville,4,80-1,82,86,92, 113,118,136,157,
Preface (laus cerei), 287-92,315-7 236-7,243,263,302,317
Premonstratensians, 40, 86, 87, 120, 150, 213, Sistine Chapel, 28, 83, 87, 92
281,296,301 sky-pillars, 275
processional candle, 189ff, 207 Society of St Pius X, 19
processional lamp, 225, 291, 306, 314, 326 Soissons, 31, 40, 60, 128, 130, 137, 150, 157, 161,
processional pole, 190 165,167-8,241,243,262,296,301,360
Prometheus, 190 Solemn Prayers, 11, 186
Prudentius, 1,174,209,230,252-3,365-6 Solesmes, 19
psalms, other, 114-16, 208-9 Southltaly,210-12,242,284
psalms, penitential, 114-18, 208 Spain, 184,210,230-1,253
Pythagoreaoism, 184 spear, 192-4
Radegonde,4 Spires, 113, 116, 121, 128, 165, 197,211,262,
Ratisbon, 104, 113, 114, 116, 124, 125, 138, 155, 281,283,302,356
165,211,267,273,290,295,331 St-Germain-des-Pres, 53, 121, 125, 132, 134-5,
Ravenna, 230 257,261,272-3,296
reed, 190 St-Pierre-sur-Dive, 112
Regensburg,185,310,312-13,336 StBertrand, 115,121,147,151,155,157,243,
Reims, 39, 44, 109, 110, 111, 115, 118-9, 121, 313
127,129,151,152,157,158,191,246,281, St Brieuc, 152,201
299 St Denys, 53, 118, 148, 191
Reims, St Remigius' Abbey, 330-2 St Die, 152,201
Renoes,113,118,263 St~orian,118, 125
reservation of fire, 9, 11, 140-5, 333 St ~our, 152
Rhaetia, 331 St Ghislain, 286
Richard, Earl of Arundel, 323 St Lo, 190, 193
Rig Veda, 184 St Malo, 113, 118
Ripoll,106,112, 118,165,219,261,274,359 StPierre d'Orval, 167
Rodez, 151, 152 Stanford-in-the-Vale, 61
Roman Breviary, 58 Strasbourg, 112,116,118,126,136,146,152,
Roman rite, passim, but especially 49, 88, 129, 158,197,211,243,264,312
138-9,145,170-82,204,216,238,242,263, Strido,229
265,279,300,316,318-19,322-30 Subiaco, 42
Rome, other churches, 356 suburbicarian, 215,230,248,275,371-2
Rome, S.Croce in Gerusalemme, 10, 107, 177, Swabia, 153
278 Tenebrae,name, 19,91
Rome, S.Maria (Aventine), 42 Tertullian, 327
Rome, S.Maria in Trastevere, 245 Theodore, 10, 11-13,99-100, 105, 177
Rome, S.Maria Maggiore, 277 Theodosius, 359
Rome, S.Maria Maggiore, 277 Tibaes, 110
Rome, St Paul's, 42, 43, 46, 48, 50, 85, 88, 110, Toledo, 4th Council of, 231
190-1, 238,356 Toledo- see Mozarabic rite also, 130, 150, 260
395
Tong~es,31,44,62,67, 73, 75,79, 113,118,126, Vatican, 202
147,194,199,202,204,206,211,261,273, Vatican ll, 8, 19, 93, 129, 160, 274, 289, 291, 347,
295,316-7 361
Tool, St Epvre, 112, 118, 147 Vendome, 257
Toulouse, 108,118,121,136,147,157,191, Veneratio of the Cross, 6, 33, 186
1~7.243,268,297,317 Veniat quaesumus, 118-22, 124-5, 255, 264,
Tournai, 113, 120,262 290-1,304
Tours, 118, 121, 138, 157, 163, 165, 201, 243, Veniun,43, 76,80-1,86,113,123,125,137,151,
298-9,356 243,266,300
Tours,St~artin,298,312 Verdun, St Vito, 197, 312
Trent. Council of, 31, 42, 210 Verdun, Treaty of, 343
Triduum, decay of, 151, 156 Versailles, 152
Trier, 40, 61, 77, 79,80-1,85, 113, 120, 134, 157, Vesta, 184
211,262 Veyme, 284
triple candle, 201-7 Vich,106, 112, 118, 126
Trowbridge, 60 Vienna, 121, 147, 151, 157, 243, 354
Troyes,118,129,152,157,263,296 Vienne, 130, 296, 298
True Cross, 4, 5 Virgil's Georgics, 228
Tulle, 151,158,301 WMes, 184
two candles, 319-20, 322-30, 335-8, 358 Westminster, 112, 121, 127, 136, 194, 197,
Ulric,42,295 210-11,236,257,261,272,302,312
UpsMa, 164, 165, 169 Worcester, 41, 62, 85, 87, 88, 110, 147, 195, 197,
Urban VITI, 156 202, 211, 301, 339
Ushaw College, 357 Wurzburg, 113,114,116,118,126,157,191,197,
Uzes,42,43, 113,118,120,147,197,207,214, 211, 262, 317-18
314-15, 320 York, 38, 39, 40, 41, 43, 46, 75, 77, 86, 87, 89,
VMence, 113, 157,251,265 112,120,149-50,194,204,211,284,354
VMerius, 305 York,St~ary.53,58,59,60,62,82,85,87,88,
Valladolid, 110 89, 109, 11, 131, 132, 135, 142-3, 151, 162,
Vallombrosa, 115, 118, 121, 136, 155, 157, 193, 164,168,194,197,201-2,211,261,299,317,
196-7,201-2,206,207,209,214,215,216, 341
217,243,247,257,262,295,299,312-3,319 Zachary,9,99, 101,170-6,180,310
Vandals,2 Zoroastrianism, 184
Vannes, 129,152,201,285 Zoshnus,229,371-2
396

6756 4060

Uploaded by

Copyright:

Available Formats

6756 4060

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

6756 4060

Uploaded by

Copyright:

Available Formats

Durham E-Theses

Fire and light in the western triduum: their use at

tenebrae and at the paschal vigil

The stage-by-stage development of Tenebrae is described showing the

The copyright of this thesis rests with the author.

Thesis submitted to the University of

1989 ·_,·_,'_ ·~: . I,:.·.-

There is no monograph in existence devoted specifically to a study

The present work is written primarily from an historical point of

Tho officG of }latins/Lauds, known as Tenebrae~ emerged from a mon=

The holding of the Vigil in anticipation was already established by

By the end of the fifteenth century it had become the practice in

In 1955 the ~fforta of the Liturgical Movement over a period of

liturgically and commemor~tively ~ore meaningful ~d tlhich appro~imatod

In 1984 the writer undertook a survey of the Paschal vigil as held

It had been intended to examine a number of other documents, mainly

'Blackfriars Correspondence' relates to communications received from

The writer wishes to express his gratitude to the following persons

A reference in the text to a published work appears in a footnote

AC Augustinian Canons ~ ~ Breviarii

Alcuin Liber de divinis officiis, formerly attributed to Alcuin.

~-~· Ephemerides Liturgicae.

MVLA Monumenta Veteris Liturgiae Ambro~ianae.

The Numeration of the Psalms

Greek Septuagint Hebrew

PART III : THE NEW FIRE PROCESSION

1. THE NEW FIRE AND THE PROCESSION 189

PARr IV : THE EASTER CANDLE AND VIGIL

1. THE ORIGIN OF THE EASTER CANDLE 224

Matins, known in the West as vigiliae, had developed within Christian

1 Daily Pr~er PP• 112-13. 2 Wilkinson PP• 66-69.

Pierre Batiffol, treating of Matins in the West in the early Middle

The earliest mention of the night office of Matins in the West

The first indication we have for the liturgical use of light, or

had brought back from Jerusalem in the above-mentioned year.

Gregory supplies us with three items of information which are of eap=

It is clear that the light represents Christ; and its emanation

There are three additional considerations which support the view

A clue to the content of the office of Matins/Lauds at Poitiers in

(ii) We have already suggested that, although the Gallican Ordo 17

(iii) Reference has already been made to the all-night monastic

THE DEVELOPMENT OF TENEBRAE

\'le suggested in the previous chapter that the tenth-century Matins/Laude

1 The practice of extinguishing gradually the lights at Matins/Lauds

The funereal aspect of the entire Triduum ia apparent from the

Of greater importance, however, for our study are the statements in

1 OR 23.1 and OR 17.93 respectively. The former, though written in

adopted by the papal court in the fourth century in imitation of those

There is no evidence to suggest that the provision of light at the

Summary of evidence - Stage 1

1 OR 23.10. The writer accepts Andrieu's suggestion that the phrase

corded by Amalariua of Metz on the occasion of a visit by him to that

At first sight Theodore appears to have ignored Amalarius' question con-

1 Officium must be translated 'service', and not 'office' which would

Two facts concerning the custom of Amalarius' church emerge as being

It is clear from the statement of Amalarius and from the reply of

Summary of evidence - Stage 2

In the evidence of Ordo 30B, which constitutes Stage 3 in the devel-

1 Martini's dating. Vogel p.160.