Art and Marriage in Early Byzantium

Art and Marriage in Early Byzantium
Author(s): Gary Vikan

Source: Dumbarton Oaks Papers, Vol. 44 (1990), pp. 145-163
Published by: Dumbarton Oaks, Trustees for Harvard University
Stable URL: http://www.jstor.org/stable/1291624
Accessed: 06-06-2016 16:00 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.
Dumbarton Oaks, Trustees for Harvard University is collaborating with JSTOR to digitize, preserve
and extend access to Dumbarton Oaks Papers
This content downloaded from 193.170.112.180 on Mon, 06 Jun 2016 16:00:50 UTC
All use subject to http://about.jstor.org/terms
Art and Marriage in Early Byzantium

GARY VIKAN
This riage
articlehelped
addresses
theByzantine
question of
how marto shape
material
culture, or, put another way, how the material culture
of Byzantium reflects the rituals and values of

marriage. The question properly encompasses all
facets of betrothal and marriage, the ceremonial,
legal, moral, and sexual. And it should include as
well all potential categories of marriage-related artifact, from illustrated historical chronicles portraying real marriages and the coins issued to com-
ture on folio 125r (Fig. 1), for example, represents

the union of Emperor Constantine VII and Helen,
the daughter of Romanos I, Lekapenos.2 This
scene, when coupled with middle Byzantine texts

of the marriage ceremony and passages in the Book
of Ceremonies distinguishing the wedding stephanos
from the imperial stemma, and surviving marriage
crowns of the period, like the pair in the Byzantine
Museum of Athens (Fig. 2), graphically evokes the
experiential reality of a middle Byzantine wed-
memorate them, to the rings, crowns, and belts

used to punctuate and confirm various stages of
ding.3 Similarly straightforward in its approach is
the ritualized union, to the gifts exchanged by the

couple or offered by friends, to the amulets worn
by one or the other to protect their union and ensure successful procreation.
the Stathatos Collection, Athens (Fig. 3), whose bezel bears simply the words "I, Goudeles, give the
betrothal ring to Maria."4
neous neither in character nor in quantity across

the millennium of Byzantine history. That from
late Antiquity is at once richer in its variety and
more subtle in its iconography than is the relatively
sparser evidence from the middle and late Byzantine periods, which tends to reflect a more literal
approach. Characteristic of the latter are the mar-
125r. See Estopafian, Skyllitzes matritensis, no. 305.
The evidence, although abundant, is homoge-
riage scenes in the twelfth-century copy of the

John Skylitzes chronicle in Madrid.' The miniaSpecial thanks are due to Leslie S. B. MacCoull for help with
papyrological sources used in this article.
After fifty years the most useful general treatment of early
Byzantine marriage remains that of K. Ritzer's 1940 Wirzburg
dissertation, here cited in its French translation, Le mariage dans
les iglises chrtiennes du Ier au XIe si'cle (Paris, 1970). As for
marriage-related art of the period, the standard publication is

still the insightful though partial treatment by E. H. Kantoro-
wicz, "On the Gold Marriage Belt and the Marriage Rings of
the Dumbarton Oaks Collection," DOP 14 (1960), 3-16.
'See S. Cirac Estopafian, Skyllitzes matritensis, I. Reproducciones
y miniaturas (Barcelona-Madrid, 1965), nos. 133, 221, 305, 325,

481, 491, 511, 549; and, for the dating of the manuscript, N. G.
Wilson, "The Madrid Scylitzes," Scrittura e civilta 2 (1978), 209-
19. See also A. Grabar and M. Manoussacas, L'illustration du
manuscrit de Skylitzes de la Bibliothique Nationale de Madrid, Bib-
liotheque de l'Institut Hellenique d'Etudes Byzantines et Post-
Byzantines de Venise 10 (Venice, 1979), 159, for a brief, general

discussion of the manuscript's marriage iconography.
an inscribed middle Byzantine betrothal ring in
2 Figure 1 = Madrid, Biblioteca Nacional, cod. 5-3, no. 2, fol.
For the Byzantine betrothal and marriage ceremonies, see

P. N. Trempela, "He akolouthia ton mnestron kai tou gamou,"
Theologia 18 (1940), 101-96 (based on manuscripts of the 9th
through 17th centuries). For the distinction between the two
crowns, see Constantine Porphyrogenitus, De ceremoniis aulae byzantinae, II, 6-9.
Figure 2 = Athens, Byzantine Museum, nos. 7663a, b. See

P. A. Drossoyianni, "A Pair of Byzantine Crowns," XVI. Internationaler Byzantinistenkongress, Akten 11/3 = JOB 32.3 (1982), 529-
38. Drossoyianni is overly cautious in identifying these
crowns-which are a match for those in the Madrid Skylitzes,

clearly form a pair, are large enough to be worn, and bear a
passage from the Psalms (20:3) used in the marriage ceremony
(Trempela, "He akolouthia," 153 f)-as specifically marriage
crowns. Because their main inscription is an invocation, on behalf of "Romanos the spatharokandidatos together with his wife
and children," she sees them as "votive offerings to a church"
(even though made of tin-plated copper). Yet Byzantine marriage rings do occasionally bear invocations (see M. Hadzidakis
[Chatzidakis], "Un anneau byzantin du Mus&e Benaki," BNJ 17
[1939-43], nos. 80, 89; and note 97 below), and it was not un-
usual for late Roman marriage rings to invoke long life in a
general and anticipatory way on behalf of the groom's family
(see F. Henkel, Die r6mischen Fingerringe der Rheinlande [Berlin,
1913], 322). Noteworthy also are late Roman sarcophagi where-
on the dextrarum iunctio, as generally symbolizing the matrimon-
ial bond, takes place in the presence of the couple's children;

see L. Reekmans, "La 'dextrarum iunctio' dans l'icono-
graphie romaine et palkochretienne," Bulletin de l'Institut Historique Belge de Rome 21 (1958), 68 f.
4MNHCTPON AIAOMH FOYAEAHC MAPHA. Figure 3 =
Athens, Stathatos Collection. See Collection Heline Stathatos, 2:
146
GARY
VIKAN
By contrast, the evidence of the early Byzantine

gold marriage belt at Dumbarton Oaks (Fig. 31) is
much more complex.5 Christ and Dionysos are interlocked in a common thematic conceit, and the
seemingly unrelated notions of concord, grace,

and health are inscriptionally linked as "from
God." The paired clasps are dominated by an apparently realistic representation of the dextrarum
iunctio, which actually had no precise visual counterpart in the ritual it evokes.6 Yet the belt itself
played an instrumental role at quite a different
juncture in the same ceremony. All of this makes
the Dumbarton Oaks belt much less a document of
an experiental reality than one of a deeper, conceptual reality.

Further distinguishing the earlier from the later
evidence is that the latter is so sparse and hetero-
geneous that it cannot properly be said to constitute a historically distinct unit within later
Byzantine art history generally. Early Byzantine
marriage-related art, by contrast, has its own identity and tradition within the history of late antique
material culture as, from the fourth to seventh
century, it gradually emerges out of later Roman

art and is progressively adapted to the needs of a
Christian, east Mediterranean clientele. This marriage art will, therefore, constitute the core of this
wedding ring, with the former, the annulus pronubus, in part functioning as, or at least symbolizing,
the prenuptial arrabon or earnest money.7 Both

rings would have been worn on the third finger of
the left hand, the anularius, since it was then still
commonly believed that a nerve or sinew ran directly from that finger to the heart.8 As Aulus Gellius had explained it:
When the human body is cut open as the Egyptians
do and when dissections, or anatomai as the Greeks
phrase it, are practiced on it, a very delicate nerve is
found which starts from the annular finger and travels to the heart. It is, therefore, thought seemly to give
to this finger in preference to all others the honour of
the ring, on account of the close connection which
links it with the principal organ.9
In late Antiquity marriage rings were worn by

both men and women, although only the bride-tobe would customarily receive the annulus pronubus.10 Thus, in theory at least, three rings would
typically be associated with each marriage." But in
fact, the number of rings extant from this period
which are identifiable by their inscription or imagery as specifically related to betrothal or marriage is relatively small, and of those that do survive, a disproportionately high number are made
of gold.'2 In part this may reflect a perpetuation in
paper.
During the early Byzantine period, as today, the

single object most intimately associated with marriage was the ring. And then, as now, there was a
distinction between the betrothal ring and the
SIn addition to the customary commitment of money and a

ring, other less common betrothal gifts included crosses, jewelry, and clothing. See Ritzer, Le mariage, 127-29; and for the
distinction between the betrothal and the wedding ring, Henkel, Fingerringe, 337-39.
8 Macrobius, Saturnalia, VII.13; and Isidore of Seville, De ec-
clesiasticis officiis, 11.20.8. See Henkel, Fingerringe, 337, 341, and
11th century. See Nog 'EXX. 13 (1916), 212-21. For a pair of
ring no. 1033, showing clasped hands holding a heart. The Orthodox betrothal service, as it had taken shape in the middle
Byzantine period (Ritzer, Le mariage, 129 f), and as it survives
today, stipulates that the betrothal rings-ta daktylidia tou arrabonos-be placed on the third finger of the right hand, the hand
(1228-34) and his wife Anna Komnene, daughter of Emperor
(London, 1941), 81.
Hadzidakis, "Un anneau," no. 90; and S. de Ricci, Catalogue of a
Collection of Ancient Rings Formed by the Late E. Guilhou (Paris,
l Of course this does not include those rings made for husband or wife to mark significant anniversary dates, birthdays,
or the new year. Certainly such rings did exist (Henkel, Finger-
(no. 47.2.2294; along with no. 47.2.2293, which appears to be
ringe, 338), and there is no reason to think that their inscriptions
its copy). The appropriateness of an identically inscribed set of

gold and silver betrothal rings is confirmed by the text of the
Byzantine betrothal ceremony, which assigns a gold ring to the
rings.
Les objets byzantins et post-byzantins (n.1., n.d.), no. 33. "Goudeles"
is a well-known Byzantine family name, attested as early as the

similar betrothal rings in gold and silver formerly in the Guilhou Collection, which bear the word mnestron and what have
been taken to be the names of King Stephan Radoslav of Serbia
Theodore Angelos Komnenos Doukas of Thessalonike, see

1912), nos. 853, 855. The silver ring is now at Dumbarton Oaks
man and a silver ring to the woman; see Trempela, "He ako-
louthia," 117.
5See M. C. Ross, Catalogue of the Byzantine and Early Medieval
Antiquities in the Dumbarton Oaks Collection, Volume II: Jewelry,

Enamels, and Art of the Migration Period (Washington, D.C., 1965),
no. 38.
6See C. Reinsberg, "Concordia," Spditantike undfriihes Christen-
tum (Frankfurt am Main, 1983), 312; Reekmans, "La 'dextrarum

iunctio'," 25; and note 114 below.
with which divine authority and power was customarily expressed (Trempela, "He akolouthia," 134 f).
9Noctes atticae, X. 10, in J. Carcopino, Daily Life in Ancient Rome
10 Henkel, Fingerringe, 339; and Ritzer, Le mariage, 127 f.
or imagery would differ fundamentally from true marriage
12There is a striking absence of identifiable marriage rings

from such rich late antique archaeological sites as Corinth and
Sardis. See G. R. Davidson, Corinth, Volume XII: The Minor Objects (Princeton, 1952), 227-48; and J. C. Waldbaum, Metalwork
from Sardis: The Finds through 1974 (Cambridge, Mass.-London,
1983), 128-33. For the disproportionate number of late antique
gold marriage rings (identifiably Christian, and of eastern Mediterranean manufacture), see notes 22, 45, and 97 below, listing
extant examples in all media of the three main iconographic

types.
ART AND MARRIAGE IN EARLY BYZANTIUM 147
Byzantium of the Roman custom, attested by Pliny,
finger of the left hand, or if it were found to match
and in part it simply may have been an unintended

byproduct of the church fathers' preaching against
the wearing of many rings; for if, as Clement of
Alexandria stipulated, a housewife could wear just
a ring portrayed on that finger in a contemporary

mural or on a sculpture.17 Such opportunities,
however, are rare, and in any event one cannot be
certain that the anularius was reserved solely for
betrothal and marriage rings. One is left, therefore, with the supposition that a significant portion
of giving an unadorned iron hoop in betrothal,
one ring-a signet to seal doors, cabinets, and

food storage vessels around the home-she and
her husband may have chosen to make that ring
function as a marriage ring as well, even if its device were not specific to that role.13 In either case
the marriage (or betrothal) ring would be unidentifiable from our vantage point.
Viewed more broadly, this disproportion of gold
to bronze suggests that it was primarily among the
wealthy and powerful of early Byzantium that
marriage rings were taken to be a significant genre

of personal adornment in their own right.14 Had
the situation been otherwise, it would have been a
simple matter for Byzantine craftsmen to replicate
gold marriage rings in silver, bronze, or even glass,
just as they replicated gold invocational rings and
belt fittings, among other jewelry items, in various
less expensive media.'15 Furthermore, the close

iconographic bond between those relatively few
surviving gold marriage rings and contemporary
gold coinage, and the frequency with which the

bridal couple on ring bezels appears to be "impe-
rial" in portrait type, headgear, or dress, corroborate the notion that the clientele for early Byzantine marriage rings was substantially confined to
the topmost level of society.' 16
How can an early Byzantine marriage ring be

identified? Theoretically, even a simple iron or
bronze band, or an anonymous gold band set with
an unincised stone, could be linked to marriage if
it were discovered through excavation on the ring
"3Pliny, Natural History, XXX.12; Henkel, Fingerringe, 337;
and J. Dl61ger, "Anulus pronubus," Antike und Christentum 5
(1936), 188-200. Clement of Alexandria, Paidagogos, III.11.

"4Although evidence from the Roman period (Henkel, Fingerringe, 339) suggests that then, too, many couples of lesser
means chose simple metal bands for betrothal and marriage,
nevertheless there survives a much higher proportion of inscriptional and iconographic bronze marriage rings from that
period than later, in Byzantium.
15See, e.g., G. Vikan and J. Nesbitt, Security in Byzantium: Lock-
ing, Sealing and Weighing, Dumbarton Oaks Byzantine Collection
Publications 2 (Washington, D.C., 1980), 5, 16.
16 For the differential impact of early Byzantine marriage leg-
islation on various strata of society, see E. Patlagean, Pauvrete
iconomique et pauvret6 sociale a Byzance, 4e-7e siecles, Civilisations
et societes 48 (Paris, 1977), 117. Marvin Ross (Catalogue, no. 66)

was so persuaded by the imperial character of some early Byz-
antine marriage rings as to suggest, with no other evidence, that
they were issued by the court in multiples, like medallions, to

commemorate imperial marriages.
of the hundreds of simple metal bands extant

from the period once functioned as marriage
rings, and with the knowledge that only a relatively
few, mostly precious metal, rings were explicitly
destined for marriage through their inscriptions
or iconography.
Many Roman marriage rings are identifiable by
the words they bear. Paired names are common, as
are a variety of phrases evoking such generic ma-
trimonial good wishes as long life, fidelity, and

happiness (e.g., Prudentia Rodani vivas, concordia
nostra perpetua sit)-phrases which, by the later

fourth century, were often given a specifically
Christian slant (e.g., Tecla vivat deo cum marito seo).18
By contrast, Byzantine inscriptional marriage

rings of any period are uncommon, with the mnestron didome ring cited above a rare exception.19 Beyond this, there are just a few invocational rings
with both male and female names (e.g., "Lord,
help Basil and Anna"), and even fewer rings with
personalized twin bezel devices (e.g., a bisected

oval bezel at Dumbarton Oaks, with "Basil-
Theodora").20 At no time in Byzantine history can
such rings be said to constitute a genre in their

own right.
In effect, a Byzantine marriage ring can be identified almost solely by virtue of the image, often
with accompanying inscription, that it bears on its
bezel. Whether it might be a betrothal ring, a marriage ring, or perhaps an anniversary ring is not
explicitly clear, although it is reasonable to assume
that those rings showing the dextrarum iunctio or in-
scribed with homonoia ("concord") would not be be"7See Henkel, Fingerringe, 342 f. Clement of Alexandria, in
the passage from the Paidagogos cited above (note 13), stipulated
that a man should wear his seal ring on his little finger, below
the lowermost knuckle. For a gold marriage ring found in a
3rd-century sarcophagus, see Henkel, Fingerringe, 337 f. For a

match between the monumental arts and an extant ring, compare the ring on the left hand of the second lady-in-waiting
behind Empress Theodora in the San Vitale procession mosaic
with, for example, Ross, Catalogue, no. 6E.
"See Henkel, Fingerringe, 322.
'9See also B. Segall, Museum Benaki, Athens: Katalog der Gold-
schmiede-Arbeiten (Athens, 1938), no. 256a.
20Ross, Catalogue, nos. 119, 126. Compare Henkel, Finger-
ringe, nos. 819 if, for Roman friendship rings with similar be-
zels.
148
GARY
VIKAN
trothal rings, since that ceremonial act and the

moral ideal evoked by that word would by then not
yet have assumed their significance for the
couple.21
was in common use in the West (Fig. 6: Speratu[s]Beneriae).24 The meaning of its composition is no
more complex than it appears to be; much as these
two figures formally complement one another
within the compositional field they share, so they
The earliest type of Byzantine marriage ring,

exemplified by a superb example in the Dum-
barton Oaks Collection (Fig. 4: Aristophanes-
Ouigil[a]ntia), is that whereon husband and wife, in

bust portraits, face one another in full profile.22
There are two basic bezel variants within the type,
an inverted pyramid with a deeply cut retrograde
device (Fig. 4), and a thin disc with a superficially
cut direct device (Fig. 5 [detached bezel]);23 the
former-whereon the groom appears at the right,

with his fibula on the "left" shoulder-was clearly
intended for sealing, whereas the latter apparently
was not. Both variants usually show a cross be-
tween the couple, but on the sealing version of this

ring type the cross is always small, whereas on the
non-sealing version it is usually large. Further-
more, the latter rings are characteristically lightweight and simply executed, whereas the former
are usually heavy, very finely crafted, and occasionally personalized.
By the fourth century this type of marriage ring,
sometimes with a cross or a Christian inscription,

21tSee Henkel, Fingerringe, 337-39. Although a dextrarum iunc-
tio may have taken place in confirmation of the betrothal contract (Ritzer, Le mariage, 128), it most likely would have been
between the parents.
22Figure 4 = Washington, D. C., Dumbarton Oaks Collection,
no. 47.18. See Ross, Catalogue, no. 50. For others, which may be
taken to be Byzantine by virtue of their findspots or inscrip-
tions, see F. Cabrol and H. Leclercq, "Anneaux," DACL 1.2

(1924), fig. 678 (gold, with cross); O. M. Dalton, Catalogue of
Early Christian Antiquities and Objects from the Christian East in the
Department of British and Medieval Antiquities and Ethnography of
the British Museum (London, 1901), no. 207 (gold, with cross [=
our Fig. 10]); E. Dauterman Maguire, H. P. Maguire, and M. J.
Duncan-Flowers, Art and Holy Powers in the Early Christian House,
Illinois Byzantine Studies 2 (Urbana-Champaign, 1989), no. 83
(bronze, with cross); W. de Grtineisen, Art chretien primitifdu haut
et du bas moyen-age (Paris, n.d.), no. 460A (silver, with cross); de
Ricci, Catalogue, no. 406A (silver, with cross: Theodotis); F. H.
should be understood as complementing one another spiritually in the life they share. Not surprisingly, this formula, which was then also commonly
used in the West for co-portraits of Sts. Peter and
Paul, had a long history in Roman coinage.25 For

example, a medallion of ca. 257 (Fig. 7), inscribed
concordia augustorum, shows Gallienus and Salonina
face-to-face in a composition designed at once to

evoke familial solidarity and imperial harmony.26
When, as is often the case, an object or symbol appears between the two heads (e.g., a scepter, victory crown, or cross), the intention was simply to
augment the solidarity concept by identifying that
authority, reward, or belief that unites the pair.
The double-profile marriage ring was taken

over by Byzantium from the West fully developed
in design and meaning, and already adapted for a
Christian clientele. Although the first extant eastern examples may date as early as the mid-fourth
century,27 the core of the sealing version of this
ring type probably falls in the first decades of the
fifth century. This is suggested by archaeological
evidence associated with closely comparable rings
discovered in the West, and by parallels between
specific eastern rings (Fig. 4) and eastern coinage
(Fig. 8: Eudocia, 423; Fig. 9: Theodosius II, 420).28
The non-sealing version should probably be

24Figure 6 = London, British Museum. See Marshall, Cata-
logue, no. 208. For the appearance of this ring type in the West,
see Henkel, Fingerringe, 337-39; M. Henig, A Corpus of Roman
Engraved Gemstones from British Sites, BAR British Series 8 (Ox-
ford, 1978), no. 790; and Ross, Catalogue, no. 50. Many ex-
amples in a variety of media and bezel designs are known; most

are signets, and inscriptions are common. Occasional examples
(Henkel, Fingerringe, nos. 98, 99) show just a single profile bust,
suggesting that such rings may have been made in pairs.
25 See J. P. C. Kent, Roman Coins (New York, 1978), nos. 156,
188, 219, 258, 276, 351, 383, 389, 457, 480 (= our Fig. 7), 523,
578, 643, 707. For such double portraits of Peter and Paul, see
Marshall, Catalogue of Finger Rings, Greek, Etruscan and Roman in

the Department of Antiquities, British Museum (London, 1907), no.
F. Cabrol and H. Leclercq, "Pierre (Saint)," DACL 14.1 (1939),

figs. 10220-23, 10240, 10245.
couple]); Ross, Catalogue, nos. 51, 52 (gold, with cross [no. 52 =
480.
our Fig. 5]); G. Vikan, Byzantine Objects of Daily Life in the Menil
Collection, Volume I (forthcoming), no. R25 (bronze).
To this list may be added a one-sided, 5th-century Byzantine
shows a male portrait strikingly like the amethyst intaglio por-
273 (gold, with cross [repousse, with a child between the
lead sealing at Dumbarton Oaks (Shaw no. 47.2.1971; unpub-
lished), apparently made with such a marriage ring bezel.

23 Figure 5 = Washington, D.C., Dumbarton Oaks Collection,
no. 53.12.61. See Ross, Catalogue, no. 52; and for such a bezel
still attached to its hoop, Cabrol and Leclercq, "Anneaux," fig.
678.
26Figure 7 = London, British Museum. See Kent, Coins, no.
27The Menil Collection bronze ring (see note 22, above),
trait in Berlin usually taken to be Constantine I. See E.
Zwierlein-Diehl, Antike Gemmen in deutschen Sammlungen: II,

Staatliche Museen Preussischer Kulturbesitz, Antikenabteilung, Berlin
(Munich, 1969), no. 545.

28Figures 8, 9 = London, British Museum. See, respectively,
Kent, Coins, nos. 751, 744. For a summary of the western archaeological evidence, see Ross, Catalogue, no. 50.
placed more centrally in the fifth century, because
and for specific figure types, from which such sa-
model, and the cross has come to dominate its
were usually deleted.34
it is clearly further removed from the Roman
field.29 Following the evidence of imperial portrai-
ture on the coinage, however, one would not expect the profile ring type to have continued much
beyond the later fifth century.30
There is no question that the figures portrayed

on the cruder, non-sealing version of this ring type
are generic portraits only, evocative in a general

way, by the jewelry that they wear, of people of
high rank.31 Representatives of the more luxurious, more expertly crafted signet version, however,
seem to have a portrait-like specificity, and in the
case of the finest surviving example, at Dumbarton
Oaks, the figures are identified by name. Were
these paired profiles intended to be realistic portrayals of Aristophanes and Vigilantia? Almost certainly not, since both are repeated virtually identically on an uninscribed gold marriage ring in the
British Museum (Fig. 10),32 and although Marvin

Ross had suggested that these two rings might
have been made for husband and wife, it is much
more likely that they were simply products of

the same Constantinopolitan(?) workshop or gold-
smith-expensive stock items to be personalized
on demand. In fact, very few marriage rings of this

type-or, for that matter, of any other type from
the period-bear identifying inscriptions, and yet
many among them, including simple bronze examples, show striking similarities to well-known
imperial portrait types.33 The commonality of the
rings in our Figures 4 and 10, therefore, would not
be one of shared individual portraiture, but rather
lient imperial elements as diadems and pendilia

Before moving on to the next common early
Byzantine iconographic scheme for evoking marriage, mention should be made of the appearance
of the double-profile marriage formula in two con-
texts other than rings. There are a number of

stamped glass pendants of eastern Mediterranean
manufacture bearing juxtaposed profile portraits
of a man and a woman, in some cases separated by
a small cross and in others, recalling the vivas inscription common on Roman marriage rings, by
the word zoe.35 Although this genre of object has
pre-Christian roots, it was still in common use during the fifth century, when these specimens should
be dated.36 Although too few marriage pendants

and necklaces survive to allow for the possibility
that they once constituted a significant category of
late antique jewelry in their own right, it is noteworthy that they are, in different forms, attested
both earlier in the West (e.g., by a Piazza della Consolazione treasure necklace [Fig. 19]), and later in
the East (e.g., by a Mersin treasure necklace [Fig.
18]).37 But unlike rings, crowns, and belts, neck-
laces seem not to have had a designated role to
play in the marriage ceremony.

Brief mention should also be made of two closely
interrelated tiny silver boxes-one found in

Thrace (Fig. 11: homonoia) and the other apparently in Cilicia-which bear juxtaposed male-
4 The deletion of imperial regalia is especially obvious in the

case of the Dumbarton Oaks and British Museum rings (Figs.
one of shared workshop technique, and of common dependency on familiar coin types. Indeed,
4, 8-10).
this is the message of early Byzantine marriage art

generally; namely, that coinage provided the ultimate models, both for basic iconographic concepts
and one with neither cross nor inscription, see Objects with Semitic Inscriptions, 1100 B.C.-A.D. 700 (and) Jewish, Early Christian
and Byzantine Antiquities, Zurich, L. Alexander Wolfe, and Frank
5An unpublished example with a small cross is preserved in

the Corning Glass Museum (no. 59.1.205). For a zoe pendant,
Sternberg, Auktion XXIII (November 20, 1989, Zurich), nos.
288, 294.
29The increasing dominance of the cross, from the early 5th

into the 6th century, is a phenomenon traceable (for example)
in Byzantine coinage and among Byzantine flat weights. For the
former, see P. Grierson, Byzantine Coins (London, 1982), 4, 34-
36Although hundreds of such stamped glass pendants sur-
vive, mostly from the eastern Mediterranean region and North

Africa, with pagan, early Christian, and Jewish iconography, the
genre has received little scholarly attention. See J. Philippe, Le
36; and for the latter, Vikan, Menil Collection, chap. 6.

3?Grierson, Coins, 4. There are, however, Byzantine lead seal-
monde byzantin dans l'histoire de la verrerie (Ve-XVIe sivcle) Bologna,
Sts. Peter and Paul separated by a small cross. See, e.g., G.
function; their continuity into the 5th century is attested by occasional examples with stylite saints or with the Chrismon.
37 See Age of Spirituality: Late Antique and Early Christian Art,
ings datable to the 6th century bearing juxtaposed profiles of
Schlumberger, Melanges d'archeologie byzantine (Paris, 1895), 299,
no. 4.
" Note especially Ross, Catalogue, no. 51, where the bride's
earrings and necklace are clearly discernible, as is the groom's
elaborate fibula.
32Figure 10 = London, British Museum. See Dalton, Cata-
logue, no. 207; and Ross, Catalogue, no. 50.

33See note 27 above.
1970), 37 f. Many specimens appear to have been amuletic in
Third to Seventh Century, ed. K. Weitzmann (New York, 1979),
no. 281; and A. Banck, Byzantine Art in the Collections of the USSR
(Leningrad-Moscow, n.d.), no. 102 (and below). See also Ross,

Catalogue, no. 66 (the Anastasius and Ariadne solidus at Dumbarton Oaks, mounted as a necklace clasp [= our Fig. 23]). For
the late Roman counterpart, in cameo-carved jet, to these eastern glass marriage pendants, see Henig, Corpus, no. 759.
150
GARY
VIKAN
female profile portraits separated by a monumen-
tal jeweled cross.38 Although the significance of
their other decoration, and even their function,

have never been satisfactorily explained, it should
be clear from the dominant role of the double portraits and from the "concord" inscription on the
Thracian box (cf. Fig. 12) that they were in some
way connected with marriage, or at least with the
joint action of a married couple.39 In this respect
they recall the much grander mid-fourth-century
Roman casket of Proiecta from the Esquiline treasure, which, even if it does not include in its iconographic program the domum deductio of the Ro-
man wedding as some had thought, nevertheless

may justifiably be linked to the marriage ceremony,
thanks to the paired frontal portraits of husband
and wife on its lid and to the vivatis in Christo in-
scription just beneath it.40 Kathleen Shelton has

suggested that the casket, as a container for bath
articles, was a gift to the bride, and that its imagery
was chosen to reflect the traditional bride's toilet,
taken on the evening before the wedding.41
What was the function of the Thracian and Cil-
ician boxes? Although traditionally, if sometimes

unenthusiastically, identified as reliquaries, they
small and early in date.42 Moreover, their iconography would be unusual for that role, since in ad-
dition to paired marriage portraits, the Cilician

box bears two votive(?) images of St. Konon accompanied by the word hygia ("health") and an animal
frieze incorporating patently apotropaic elements.
Grabar found these features anomalous on a con-
tainer intended for relics, but, not recognizing a

possible marriage connection, he could suggest no
alternative interpretation.43 Perhaps these boxescontaining some relic-like apotropaica, or maybe
rings-were specifically associated with the marriage ceremony, an event in which both the notion
of "health" and the warding off of evil had significant roles to play (see below).44
Just as the profile soon substantially disappeared
from Byzantine coinage, so also it disappeared

from the bezels of Byzantine marriage rings. By
the sixth century, paired frontal bust portraits (Fig.
12: homonoia, Theou charis) and variations on the
dextrarum iunctio (see below) had become the stan-
dard iconographic formulae.45 Nearly all surviving

42Those few metal boxes of comparable size and date in
would be, among extant reliquaries, unusually
which relics have been found seem to have been made significantly earlier than their outer protective containers, and quite
38Figure 11 = Sofia, Narodni Muzej, no. 2519. It measures

2.9 x 4.5 x 3.0 cm. See H. Buschhausen, Die spitriimischen Me-
nos. B19, B20, C12; and Age of Spirituality, nos. 568, 569.
43Grabar, "Isaurie," 57 f. For closely comparable votive images in repousse silver, see M. Mundell Mango, Silver from Early
Byzantium: The Kaper Koraon and Related Treasures (Baltimore,
1986), 244.
44 For a small, undecorated bronze box of the period discov-
tallscrinia undfriihchristlichen Reliquiare, 1. Teil: Katalog, WByzSt
11 (Vienna, 1971), no. B3; and for the Cilician box, now in
Adana, no. B4. See also A. Grabar, "Un reliquaire provenant
d'Isaurie," CahArch 13 (1962), 49-59; and idem, "Un reliquaire
provenant de Thrace," CahArch 14 (1964), 59-65. These two
boxes are so closely related to one another-and so unlike other

silver containers of the period-as to indicate a common origin;
because Thekla and Konon, both Isaurian saints, appear on the
box now in Adana, which was apparently found in Cilicia (or
Isauria), the two may be localized to that region of southeastern
Asia Minor. See Grabar, "Isaurie," 55 f; and idem, "Thrace," 65.
39Buschhausen, Metaliscrinia, 182, 185 f, 205 f, was closer to
the mark in labeling these as "private donor portraits," derived
from imperial portrait types; he had difficulty, however, in rec-
onciling his inappropriately early dating of the Thracian boxto the second quarter of the 4th century-with the dominant
cross, and with his conviction that both boxes were made to be
reliquaries (which then hardly existed). Grabar, on the other
hand, correctly dated the boxes later (the Thracian box ca. 400,
the Cilician box to the mid-5th century), but mistakenly identified the paired profiles as Constantine and Helen, overlooking
the fact that the imperial diadems had been deleted from the
coin models, and choosing to downplay the fact that no such
iconographic formula for that pair of saints is known to have
existed at such an early date.
40These two elements-paired frontal portraits and a vivas in
deo inscription-are united on a contemporary marriage ring
bezel preserved in Spalato. See O. Pelka, Altchristliche Ehedenkmiller (Strasbourg, 1901), 131, no. 50. For the iconography and
function of the Proiecta casket, see K. J. Shelton, The Esquiline
Treasure (London, 1981), 27 f.
41Shelton, Treasure, 28.
possibly for other purposes. See Buschhausen, Metallscrinia,
ered with a phos-zoe ring in it, see Objects with Semitic Inscriptions,
no. 339; and for various textually and archaeologically attested

uses for small boxes during this period, including as containers
for jewelry, see J. Duffy and G. Vikan, "A Small Box in John
Moschus," GRBS 24.1 (1983), 93-99.
Several iconographic and stylistic elements of the Thracian
and Cilician boxes (e.g., the profile marriage portraits, the

Lamb of God, Christ Enthroned with Apostles) are paralleled
on the bezels of a closely interrelated group of four luxurious

gold rings, two of which are the near-twin marriage rings at
Dumbarton Oaks and in the British Museum discussed above
(Figs. 4, 10). For that group, which should be dated with the
boxes to the first half of the 5th century, see G. Vikan, "Early
Christian and Byzantine Rings in the Zucker Family Collection,"
JWalt 45 (1987), 33, figs. 5-8. The repetition of juxtaposed profile portraits of Sts. Peter and Paul on the top and back of the
Cilician box may have been intended to complement and reinforce the "concord" notion implicit in the double marriage portrait appearing on the box's two ends.
45Figure 12 = Washington, D.C., Dumbarton Oaks Collec-
tion, no. 59.60. See Ross, Catalogue, no. 4E. For others, see Byzantium: The Light in the Age of Darkness, New York, Ariadne Gal-
leries (November 2, 1988-January 31, 1989, New York), no. 39

(gold, with crowns: homonoia) (now in the Royal Ontario Mu-
seum, Toronto); Dalton, Catalogue, no. 133 (gold and niello,
with Christ: Theou [?] homonoia); Gruneisen, Art chretien, no. 456
(gold "bague reliquaire" [i.e., thick detached bezel]); P. Orsi, Sicilia bizantina, I (Rome, 1942), 158, fig. 73 (gold and niello, with
examples of the frontal portrait ring type are gold,

although, unlike profile portrait rings, these exist
primarily in just one version, with a series of inter-
changeable subordinate elements.

The typical frontal portrait ring is characterized
by a thick, usually partially nielloed bezel soldered
to a sturdy wire hoop. The round field of its nonsealing device is dominated by a flared-arm cross,
flanked on the left by the groom and on the right
by the bride; frequently a bust portrait of Christ
appears just above the cross (Fig. 13: homonoia).46
The exergue is reserved for an inscription, which
will usually be the word homonoia; the upper and
lower edges of the bezel may be inscribed as well,
and this secondary inscription will usually be the
phrase Theou charis ("grace of God"). Stylized marriage crowns are sometimes added, in the form of
simple arcs or semicircles suspended over the

heads of the bride and groom, even though in
some cases (Fig. 13) the bride appears to be wear-
ing a crown already. Two virtually identical bronze

rings of this type (Fig. 14: hygia) stand somewhat
apart from the core group by virtue of their medium, bezel shape, and relative cross size, and because their exergue bears the word "health."47 The
more characteristic gold examples, although sub-
ject to iconographic variations, are nevertheless
closely interrelated through small details of style

and technique; this suggests a common origin and
a restricted chronology.48 Their precious material,
substantial weight, and generally fine craftsmancrowns and Christ: homonoia); Paris, Cabinet des Medailles, Seyrig Collection, unpublished (bronze, with crowns: hygia); Ross,
Catalogue, no. 67 (gold and niello, with crowns and Christ: hom-
onoia [= our Fig. 13]); no. 68 (gold and niello, with Christ:
Theou charis); J.-M. Spieser, "Collection Paul Canellopoulos (II),"

BCH 96 (1972), no. 9 (gold, with dove: homonoia, Theou charis [=
our Fig. 17]); no. 11 (gold [raised device]: charis); Toronto, the
Royal Ontario Museum, no. 986.181.3, unpublished (gold, with
Christ: Theou homonoia); Vikan, Menil Collection, no. R26
(bronze, with cross: hygia [= our Fig. 14]); idem, "Zucker," fig.
9 (gold [raised device]).
46Fig. 13 = Washington, D.C., Dumbarton Oaks Collection,
no. 53.12.4. See Ross, Catalogue, no. 67.
47Figure 14 = Houston, Menil Collection, no. R26. See Vi-
kan, Menil Collection, no. R26. Its unpublished twin, in the Cabinet des Medailles, is much better preserved; on it the crowns
are clearly visible, as are the first three letters (upsilon, gamma,
iota) of the word hygia, which is only faintly discernible in the
exergue of the Menil ring.
48Orsi, Sicilia bizantina, I, 158, fig. 73 and Dalton, Catalogue,
no. 133 are very much alike, as are Ross, Catalogue, no. 4E (=
our Fig. 12), Byzantium, no. 39, and Spieser, "Canellopoulos,"
no. 9 (= our Fig. 17). Reputed findspots include Constantinople (Ross, Catalogue, no. 67), Beirut (Dalton, Catalogue, no.
133), and Sicily (Orsi, Sicilia bizantina, I, 158, fig. 73); this would
suggest central dissemination of the ring type, if not the actual
rings, from Constantinople.
ship indicate wealthy clients, though surprisingly,

not one example of this ring type has been found
with a personalized inscription.49 The word homonoia suggests a role in marriage, to the exclusion
of betrothal, and significant variations in hoop size
presuppose use by both men and women.50
Although frontal bust portraits of the sort decorating the lid of the Proiecta casket were common
among most categories of late Roman marriagerelated art (e.g., sarcophagi, gold glass), they were
not characteristic of Roman wedding rings.5' In

this case Byzantium seems to have arrived at the
iconographic formula on its own, perhaps developing it directly from the western-derived profile
type in its later, non-sealing variant, with the larger
cross (Fig. 5).52 But whatever its immediate origins,

the core of the frontal portrait group clearly belongs in the later sixth to early seventh century.
This is indicated by the ring design itself, whose
thick disk bezel more typically bears a cross monogram, datable after ca. 540, by archaeological evi-
dence associated with related rings and treasures

(e.g., the Mersin treasure, discussed below), by the
distinctive three-figure configuration of those bezel devices with Christ en buste (Fig. 13), which parallels Byzantine glass weights datable to the later
sixth century (Fig. 15: Euthaliou), and by general
similarities between those bezels without Christ
(Figs. 12, 14) and coins and bronze weights issued

during the reign of Justin II and Sophia (Fig. 16).53
490On the finer examples (Fig. 12) it is clear that the groom
wears a chlamys with fibula and the bride a necklace and earrings.
5"Compare, for example, Ross, Catalogue, nos. 4E, 68.
"For sarcophagi and gold glass, see F. Cabrol and H. Leclercq, "Mariage," DACL 10.2 (1932), figs. 7649, 7664, 7665,
7667, 7670. For a rare Roman ring bezel with frontal portraits,
see Pelka, Ehedenkmiiler, 131, no. 50.
52The Grineisen ring bezel (Art chretien, no. 456), with frontal
busts flanking a large cross, is remarkably like the non-sealing

profile version. Frontal imperial bust portraits (both male) separated by a small cross appeared in the East in the 5th century
on bronze flat weights (see Dalton, Catalogue, no. 437), and a

small cross appears between seated emperors on the coinage
issued during the brief joint reign of Justin I and Justinian I
(Grierson, Coins, pp. 2, 19); neither, however, appears directly
to have inspired the ring bezels, whose compositions are quite
different.
53 Figure 15 = Houston, Menil Collection, no. GW12. See Vikan, Menil Collection, nos. GW9, GW12, and chap. 7. The rings
closest in composition to the glass weights are those published
by Dalton (Catalogue, no. 133) and Orsi (Sicilia bizantina, I, 158,
fig. 73).
Figure 16 = Geneva, Musde d'Art et d'Histoire. See N. Dtirr,

"Catalogue de la collection Lucien Naville au cabinet de numismatique du Musde d'Art et d'Histoire de Geneve," Genava 12
(1964), no. 311; and Vikan, Menil Collection, chap. 6. For the
coinage, see A. R. Bellinger, Catalogue of the Byzantine Coins in the

Dumbarton Oaks Collection. Volume One, Anastasius I to Maurice,
152
GARY
VIKAN
Although close interrelationships among the gold

members of this group argue against a long tradition, a firm terminus ante quem for the type is difficult to fix, since striking compositional parallels
exist between the bronze representatives and
eighth-century coins and lead sealings.54
Concord is iconographically conveyed on these

rings with the same directness and simplicity that
it had been conveyed on the earlier profile rings.
Moreover, this type too has compositional and thematic parallels in imperial iconography, as we have
just seen. Justin II and Sophia are united on coins
and bronze weights, evoking their imperial and familial unity, and although the figures represented
on the glass weights are different from those on
the rings-the emperor is above and the eparch of
Constantinople and another high official belowtheir iconographic message is basically the same,
for much as husband and wife complement one
another in the domestic sphere under the uniting
authority of Christ, so the two officials complement
one another in the economic sphere under the

uniting authority of the emperor. Iconographi-
cally, what distinguishes these rings from the earlier type, aside from the portraits, is their much
greater emphasis on Christian imagery; specifically, the more prominent cross and the bust por-
trait of Christ.
Also iconographically distinctive to this ring

type, but not specifically Christian, are the crowns
that appear on about half the examples in the
group (Fig. 13).55 During the early Byzantine period crowns were essential ingredients in the mar-
riage ceremony, and among the Orthodox they
have remained so ever since.56 Like most other as491-602 (Washington, D.C., 1966), pl. LVIII. For the dating of
this ring design (via the cross monogram), see Ross, Catalogue,
no. 69; and Vikan, "Zucker," 39; and for the related archaeological evidence, see Ross, Catalogue, no. 4E; and Dalton, Catalogue,
no. 189. The motif of Christ en buste above a cross is paralleled
on the ca. 600 pewter Holy Land flasks preserved at Monza and
Bobbio. See A. Grabar, Les ampoules de Terre Sainte (Paris, 1958),
passim.
54See Grierson, Coins, pl. 17, 302; and G. Zacos and A. Veglery, Byzantine Lead Seals (Basel, 1972), no. 260.
55 Marriage crowns, as distinct from whatever secular headgear bride or groom might wear, do not appear on rings of the
profile portrait type already discussed, and they are very rare
on the dextrarum iunctio rings which will be considered next. Al-
though in a few instances among rings of the latter category,

Christ (or Christ with the Virgin) appears to be touching the
heads of the couple (Figs. 25, 26), in no case can it be seen that
a crown is being set in place.
56See Ritzer, Le mariage, 95 f, 136 f; K. Baus, Der Kranz in
Antike und Christentum, Theophaneia: Beitrige zur Religionsund Kirchengeschichte des Altertums (Bonn, 1940), 93-109;
and note 3 above.
pects of the ceremony, the custom of crowning the

bride and groom was taken over by the Byzantines
from the Romans. The Roman marriage crown,
made of flowers and sacred plants, was initially rejected by the church fathers-most notably Tertul-
lian (De corona) and Clement of Alexandria-be-
cause of its superstitious associations.57 But by the

fourth century it was apparently in common use
among Christians at both ends of the Mediterra-
nean, signifying in a general way the spiritual

triumph of a couple united in Christ-a notion
that John Chrysostom developed further, making

of the crowns the symbolon tes nikes ("symbol of vic-
tory") of the bride and groom over sensual pleasure.58 And although no Church requirement yet
existed, it was becoming increasingly common for
a priest to perform the crowning; indeed, when

Emperor Maurice married the daughter of his
predecessor Tiberius, it was the patriarch of Constantinople who put the wedding crowns in place.59
Crowns of various sorts, sometimes held by

Christ or simply suspended in mid-air, appear
prominently in the marriage iconography of
fourth-century Rome, most notably on sarcophagi

and in gold glass vessels.60 Although they may ap-
proximate in their leafy appearance the crowns
then used in the marriage ceremony, their iconographic role, like that of the figure holding them,
was essentially symbolic.61 In this respect they parallel the victory crown suspended in the hand of
God over the head of the empress on contempo-
rary coinage (e.g., Fig. 8),62 and thus differ fundamentally from the real marriage crowns being set
in place in the marriage miniatures of the twelfthcentury Skylitzes chronicle (Fig. 1). The same applies to the schematic crowns on the bezels of sixthcentury Byzantine marriage rings, which, like their
counterparts in the Christian art of fourth-century
Rome, may have symbolized a general sort of spiritual triumph or perhaps the more specific moral
victory evoked by John Chrysostom. But in either
case, they probably still were believed to carry a
measure of that amuletic power ascribed to their
pre-Christian counterparts. This is suggested by

57See Baus, Kranz, 99, 100; and Ritzer, Le mariage, 95.
58See Baus, Kranz, 101; Reekmans, "La 'dextrarum iunctio',"
69-77; Kantorowicz, "Marriage Belt," 8; and Ritzer, Le mariage,

95, 136 (PG 62, col. 546).
59See Ritzer, Le mariage, 136 f (Theophylaktos Simokattes).
60See Cabrol and Leclercq, "Mariage," figs. 7647, 7662, 7664,
7665; Reekmans, "La 'dextrarum iunctio'," 69-77; and, for a ring
with suspended crowns, Henkel, Fingerringe, no. 404.
61See Reekmans, "La 'dextrarum iunctio'," 69-77; and Reinsberg, "Concordia," 315 f.
62See Kent, Coins, no. 727 (etc.).
a passage from the Psalms (5:12), "Thou hast
between a Christian couple of the sixth century
crowned us with a shield of favor," inscribed on two
and a pagan couple of the second, for as defined
seventh-century Byzantine dextrarum iunctio mar-
in the Digest of Justinian, "Marriage is a union of a
whose bezels show Christ between the bride and
lifetime...."65 But among the Romans a harmoni-
groom, raising his hands to the tops of their heads

as if touching, placing, or blessing crowns.
achievement, whereas in Byzantium matrimonial
riage rings which will be discussed below-rings
We have already seen that inscriptions-names

and a variety of good wishes-were common on
both pagan and Christian marriage rings of the
late Roman period in the West; they were rare,
however, on the early profile type of Byzantine
marriage ring. It is only with the later, frontal portrait type that words appear as a major ingredient
on Byzantine marriage rings. Specifically, there are
three different inscriptions characteristic of the
group, one of which is traditional and two innovative. The traditional inscription, in the exergue of
most of the rings, is the word homonoia or "concord," which was likely derived directly from the
familiar Roman evocation of (and personification

for) unity of mind and purpose, concordia.63 Homonoia might be applied to a political or dynastic al-
liance, or to a simple private marriage, but how-
ever simple the alliance, its meaning for the
married couple was at once specific and profound,
since for them "concord" was a succinct statement
of the prevailing late antique moral code governing their relationship.64 Most clearly articulated by
the Stoics, the belief was that one's wife was no
longer a piece of expendable property, she was a

lifelong companion and friend. Thus marriage,
man and woman and a consortium for an entire
ous lifelong union was viewed as a personal

homonoia could only be achieved through divine
grace. Novel 26 of Leo VI, summarizing the spirit

of Justinian's Novel 22, makes this explicit by characterizing marriage as "a most important and valuable gift, granted by God the creator, to man."66
Byzantium's additions to the traditional Roman
formula of frontal bust portraits convey this shift
in emphasis in the clearest possible terms. The
large, ubiquitous cross is one telling addition, the

coupling on a few rings of the genitive Theou with
homonoia ("concord of God") another, and the ap-
pearance of the bust of Christ over the cross on

many rings yet another. But the most direct statement of Byzantium's modification of the old Stoic
ideal comes in the innovative inscription shared by
nearly half these rings: Theou charis. The state of
marital concord evoked in traditional terms by the
paired portraits and the word "concord" can now
be understood only as something available to mankind solely through the "grace of God."67
An iconographically unique ring (Fig. 17: homonoia, Theou charis), though one still clearly belonging to the frontal portait type, at once corroborates
and refines this interpretation.68 Basically it re-
now for the first time based on the idea of the
couple, was meant to be a durable state of affection

between two honorable, civic-minded individuals.
To this extent there is no distinction to be drawn
63See H. Zwicker, "Homonoia," RE 8 (1913), 2256-69; and
Reinsberg, "Concordia," 312-17. Although the Greek word
homonoia, in the sense of the Latin concordia, had a long tradition
on eastern Mediterranean coinage, and occurs occasionally as
well on pre-Christian Greek marriage rings (with the motif of

joined hands), its appearance here, in 6th-century Byzantium,
likely owes its immediate inspiration to the pervasive western
concordia marriage tradition of the later Roman period (which
was in turn dependent on the earlier Stoic homonoia tradition).
For the coinage, see Zwicker, "Homonoia," 2268; and for the
rings, Henkel, Fingerringe, no. 222; and Henig, Corpus, app. 30.
The homonoia inscription with bride, groom, and cross is documentable for the first time in the East in the 5th century, with
the Thracian silver box discussed above (Fig. 11). The close relationship of this box's iconography to that of contemporary
profile marriage rings suggests that the word homonoia may, at

that period, have entered the eastern marriage art tradition via
(now lost) rings of that type.
64See A History of Private Life, I: From Pagan Rome to Byzantium,
ed. P. Veyne (Cambridge, Mass.-London, 1987), 33-47 (P.
Veyne).
65Dig, 23.2.1; see The Digest ofJustinian, trans. A. Watson (Phil-
adelphia, 1985), II, 657. For the continuity of this definition

into the late Byzantine period, see P. D. Viscuso, A Byzantine
Theology of Marriage: The "Syntagma kata stoicheion" of Matthew
Blastares, Ph.D. dissertation (Catholic University, 1988), 71 f.

For a more general overview of the continuity from pagan to
Christian marriage, see D. Herlihy, Medieval Households (Cambridge, Mass., 1985), chap. 1.
66See Les Novelles de Leon VI, ed. and trans. P. Noailles and A.
Dain (Paris, 1944), Novel 26.
67See Kantorowicz, "Marriage Belt," 8-11. A comparable
transformation from pagan to Christian values occurred on

Byzantine bronze flat weights during the same period. Around
400, paired imperial portraits flank a tyche to evoke the idea that
the empire's prosperity depends on harmonious co-rulership le-
gitimized by the polis; by the later 5th century a cross has replaced the tyche, implying that divine sanction is now preemi-
nent; and in the 6th century imperial imagery is usually
dispensed with altogether, its place taken by a large cross and,

quite often, by Theou charis. By then it was only through the
"grace of God" that honest commerce could be guaranteed, for
in the face of chronic dishonesty in the marketplace, the state
had been forced to resort to depositing a standard set of weights
in the largest church of the city. See Vikan, Menil Collection,

chap. 6.
68Figure 17 = Athens, Canellopoulos Museum. See Spieser,
"Canellopoulos," no. 9.
154
GARY
VIKAN
peats the iconographic configuration of the Dum-
barton Oaks ring illustrated in Figure 12, but
Theou charis has been shifted to the exergue with

the homonoia inscription, and a bird with a branch
in its beak has been placed above the cross. This

seemingly incongruous addition (or substitution,
for Christ) is in fact a subtle visual quotation from

a popular early Christian salvation scene in which
a dove returns to the Ark with an olive branch,
confirming to Noah and his family that they have
been spared from the flood (Gen. 8:10, 11).69 In

this respect the bird symbolizes hope, salvation,
and more important here, divine grace.

The other innovative inscriptional element, supplanting homonoia on the two bronze rings in the
group (Fig. 14 [much worn]), is hygia. Is it simply a
banal good wish of the sort evoked at all weddings,
even now-the Byzantine equivalent of the Ro-
man wedding acclamation feliciter-or did it carry

a more specific meaning, comparable to homonoia
and Theou charis? The word "health," in various

forms (hygia, hygieia, hygienousa phori, hygienon chro),
sometimes coupled with other words (hygia-zoe,

hygia-chara), or occasionally personalized (hygia
Ioannou), appears on a variety of early Byzantine
artifacts. On commercial stamps and on water
buckets for use in the baths, the intention seems
simply to have been to convey a good wish to the

consumer or user, whereas belts and rings bearing
the word "health" might reasonably be taken to be
so inscribed for amuletic purposes-which was undoubtedly the intention when hygia appears, as it
occasionally does, on earthen pilgrim tokens from
holy shrines and on medico-magical armbands

used by women to treat gynecological ailments.70
The fact that the word appears here on an object
worn on the body suggests, prima facie, a high level
of intentionality, but even if this were so, hygia's po69For a ring with both dove and Ark on its bezel, see R. Garrucci, Storia della arte cristiana (Prado, 1880), pl. 478, 10, and for
other parallels, pls. 477, 478; Dalton, Catalogue, pls. I, nii; T.

Klauser, "Studien zur Entstehungsgeschichte der christlichen
Kunst, IV," JbAC 4 (1961), 141, 142; and Vikan, "Zucker," 33,
fig. 5.
70For the stamps and buckets, see Vikan, Menil Collection,

S120 and chap. 5; and M. Mundell Mango, C. Mango, A. Care
Evans, and M. Hughes, "A 6th-century Mediterranean Bucket
from Bromeswell Parish, Suffolk," Antiquity 63, 239 (1989),
295-311. For a relevant belt fitting and some hygia rings, see
The Malcove Collection, ed. S. D. Campbell (Toronto, 1985), no.
107 (hygienon chro); and Dalton, Catalogue, nos. 149-52. For the
pilgrim tokens and the armbands, see G. Vikan, "Art, Medicine,
and Magic in Early Byzantium," DOP 38 (1984), 69 f, 75 f (and
for a doctor's instrument case and a doctor's pill stamp so inscribed, pp. 66, 69 f).
tential amuletic role in relation to marriage is not

immediately apparent.
Dioscorus of Aphrodito, in his "Epithalamium
for Paul and Patricia" (ca. 566), seems to be using

the notion of health on its most banal, least intentional level when he wishes the couple "a life without illness."71 But in his "Epithalamium for Isakios," his poetry turns to the more serious topic of
demons and magic:
Soon you shall see children on your lap, sweeter than

nectar, loved for their beauty-a blessing that your
troubles have merited, as you do honor to the ability
of your forebears. Go away, evil eye; this marriage is
graced by God .... 72
Isakios' marriage is "graced by God" toward the

proper goal of every marriage: procreation.73 And
this, too, is a concern at the heart of Justinian's
Novel 22, whose summary by Leo VI was already
quoted in part above:
Marriage is a most important and valuable gift,
granted by God the creator, to man. For not only does

it give aid to nature exhausted by death, not allowing
the human species, devoured by death, to disappear
entirely, but also it gives men great joy in another way,
thanks to the children born to it.74
The paramount importance of the reproductive

aspect of Byzantine marriage, evoked in Dioscorus'
epithalamium and in Novel 26, reappears in even
clearer terms in the opening statement of Novel 2

of Justin II, which restored divorce by mutual con-
sent: "There is nothing more honored by men

than marriage, from which result children, the
succession of future generations, the peopling of

villages and towns, and the creation of good polity."75 The operative principle of this legislation is
that an unhappy marriage-one corrupted by "irrational hatred"(alogon misos) -necessarily renders

impossible the central goal of marriage, childbearing, and therefore can be dissolved by mutual consent, divorce being an effective treatment for the
7See L. S. B. MacCoull, Dioscorus of Aphrodito: His Work and
His World (Berkeley, 1988), 81 f.
72 MacCoull, Dioscorus, 111.
73The following discussion of the impact of demographic exigencies on Byzantine marriage poetry, morality, and law-and
ultimately, on Byzantine marriage art-owes much to a paper

entitled "Byzantine Models of Marriage" delivered.by Angeliki
Laiou at Dumbarton Oaks in 1985, and to Judith Herrin's paper
in the 1989 Dumbarton Oaks Spring Symposium, entitled "Divorce." Special thanks are due to both scholars for kindly allowing me to read their papers in typescript. See also Patlagean,
Pauvrete, chap. 4.
74 See note 66 above.
75Zepos,Jus, I, 3-4, 75 (A.D. 566).
disruptive force of that "mischievous demon"

(skaios daimon) afflicting the couple.76 Similarly,
marriages rendered childless as a consequence of,
among other things, homosexuality, impotence, or
disease (leprosy) could be terminated by divorce.77
This means, in effect, that a successful, fruitful
marriage would be one blessed with both harmony
and good health, the one being no less essential
than the other. It is in this spirit that Dioscorus of
Aphrodito ends his "Epithalamium for Paul and
Patricia":
Give Paul a pleasant and happy marriage with Patri-
cia, a life quite without illness .... Paul and beloved

Patricia.... Give them the . . . of indissoluble har-
mony, as they hold children on their laps, and a bright
and peaceful life, worthy of poetic praise... .78
In a society suffering from high infant mortality,

an alarming morbidity rate among young women,
and a chronic labor shortage-in a society to which
God had given marriage to insure its very survival-the words hygia and homonoia, as invocations
of amuletic (i.e., birth-facilitating) power from
God, could be viewed as both complementary and

inscriptionally interchangeable.79 And as for the
linking of magic with marriage, that bond is as old
and intimate as the bond linking magic with childbirth.80s The date of the Roman wedding was chosen to avoid certain unlucky days, the ring finger
was chosen for its supposed connection with the
heart, the entrails of an animal were examined to
discover the marriage omens, crowns were worn to

ward off danger, ribald songs were sung to avert
the evil eye, and the bride was carried over the
threshold, according to some, to avoid an unlucky
first step. The passage from Dioscorus adjuring

the evil eye, the hygia inscription, and the excerpt
from Psalm 5 invoking a "shield of favor," suggest
that at least some of the superstitious flavor of the
76 Ibid., 3-5.
SCauses for divorce are enumerated in Justinian's Novels 117
and 134, and in the Ecloga of A.D. 741 (references from Herrin,
"Divorce").
78 MacCoull, Dioscorus, 81 f.
79For an overview of the mortality issue, see A. Laiou, "Family, Byzantium," Dictionary of the Middle Ages, ed. J. R. Strayer, 4
(1984), 595 f. E. Kitzinger ("Reflections on the Feast Cycle in

Byzantine Art," CahArch 36 [1988], 72 note 71), apparently unaware of Byzantine marriage legislation and hygia rings, failed
to recognize the inherently complementary nature of concord
and health in his analysis of Byzantine marriage ring iconogra-
phy.
8"For superstitious elements in the Roman marriage, see M.

Johnston, Roman Life (Glenview, Illinois, 1957), 132-37; and for
magic and childbirth in late Antiquity, see Vikan, "Art, Medi-
cine, and Magic," 76-84.
pagan marriage ceremony survived into Byzantine

times. More such evidence will emerge below.
Before proceeding to the last major category of
early Byzantine marriage iconography, brief mention should be made of the appearance of the
frontal portrait formula outside the context of
rings. In the late nineteenth century a treasure of

gold jewelry was discovered near Mersin, in Cilicia.
Now in Leningrad, it includes, among various earrings, rings, bracelets, belt fittings, and necklaces
characteristic of the later sixth to early seventh
century, an unusual pendant necklace with a large
repouss6 medallion of an emperor being crowned,
and twenty repouss6 links bearing identical frontal
bust portraits of a man and woman separated by a
cross, with hygia in the exergue (Fig. 18).81 N. P.
Kondakov, in his 1896 publication of the treasure,
identified the pair as "emperor and empress?,"

whereas Andre Grabar, in his extensive study of
this necklace in the 1951 Dumbarton Oaks Papers, labeled them as Sts. Constantine and Helen, while
acknowledging that no contemporary visual evidence could be cited as corroboration, and apparently overlooking the fact that neither wears the
appropriate imperial headgear.82 But clearly, the
links' "concord" male-female composition and hygia inscription are much more appropriately associated with contemporary marriage iconography
(Figs. 12, 14).3 Highly plausible on its face, this
interpretation is rendered more attractive by the

presence in the Mersin treasure of a marriage ring
and, more important, by the fact that Grabar had
already concluded, primarily on the basis of the

large medallion, that this necklace fulfilled some
magical function.84 In this respect it recalls the Pi-
azza della Consolazione marriage necklace in the

81 See N. Kondakov, Russkie klady. Izsldovanie drevnostej veliko-
knjazeskago perioda (St. Petersburg, 1896), 187-91; A. Grabar,

"Un medaillon en or provenant de Mersine en Cilicie," DOP 6
(1951); Banck, Byzantine Art, nos. 102-5, 107; Ross, Catalogue,
no. 4; and B. Deppert, "Early Byzantine Jewelry, Fourth to Seventh Century A.D.," Byzantine Jewelry, ed. D. Content (forthcom-
ing), section xii. This treasure is datable by virtue of the cross

monogram borne by one of its belt tabs, and because of its general stylistic and typological similarities with other treasures datable through coin finds.
Figure 18 = Leningrad, State Hermitage Museum, no.
omega 107-8. See Banck, Byzantine Art, no. 102; and for a line
drawing, Kondakov, Russkie klady, fig. 101.
82Kondakov, Russkie klady, 187 f; and Grabar, "Un medaillon,"
29 f.
83Spieser ("Canellopoulos," no. 9) and Deppert ("Jewelry,"
section xII) both noted the similarity between these link medallions and contemporary marriage rings.
84See Grabar, "Un medaillon," 46.
156
GARY
VIKAN
Metropolitan Museum of Art (Fig. 19) which, although western and earlier, combines in its two
pendants traditional marriage iconography and
magic.85 In this case the larger, gold pendant bears
the marriage imagery-profile portraits of husband and wife about to be crowned-and the
magic comes in the form of a smaller, hematite

pendant, a Greco-Egyptian gem amulet bearing
the so-called Abrasax, a snake-legged solar deity
with the head of a rooster.86 Although this creature
was believed to wield multivalent powers, the material into which its image is cut, hematite, was val-
ued in late Antiquity above all for its supposed

styptic quality, verified by the red powder pro-
duced when this lustrous black mineral of iron ox-
ide is crushed.87 Hematite and blood control, in
identification with Constantine and Helen. Addi-
tionally, it is important to be clear as to what these

objects are, for as stamped earthen tokens they be-
long to a familiar category of late antique artifact-the earthen eulogia ("blessing")-which has
been shown to be characteristically amuletic in its
intended function and often specifically medicinal
in its desired effect.90 Moreover, while many such
tokens, like those showing St. Symeon Stylites the
Younger, may be linked through their iconography
to specific pilgrimage shrines, others, like those
showing Christ en buste, the Virgin and Child, or
the word "Solomon" with a coiled serpent (Fig. 21),
were apparently topographically anonymous, and
therefore, in effect, simply portable bits of sacred
magic.91 The same is likely true of the paired bust
the sense of uterine bleeding, were key ingredients
tokens, which are functionally as well as icono-
in contemporary medical magic, and here, under
graphically equivalent to the Mersin treasure necklace links (Fig. 18) and the bronze hygia marriage
rings (Fig. 14), insofar as the amuletic invocation
the umbrella of marriage, this amuletic power may

have been invoked as a protective aid in childbear-
ing-specifically, as an antidote to the uterine

bleeding symptomatic of miscarriage.88
"Sts. Constantine and Helen" or "Sts. Peter and
Paul" are the labels conventionally applied by antiquities dealers to the stamped earthen discs with
paired frontal portraits (Fig. 20) which formed the
major part of a hoard of several hundred "pilgrim
tokens" recently dispersed on the international art
market.89 According to the dealers' explanation,

the figures are shown flanking a cross because
these objects were issued by Emperor Heraclius in
celebration of his returning the True Cross to Jerusalem. However, that the bust to the left is
slightly larger than the other argues against the

possibility that they might be Peter and Paul, and
the absence of contemporary visual parallels or
distinguishing imperial regalia militates against an

5 Figure 19 = New York, the Metropolitan Museum of Art,
Rogers Fund, 1958, no. 58.12. See Age of Spirituality, no. 281.
860On the Abrasax, see C. Bonner, Studies in Magical Amulets,
Chiefly Graeco-Egyptian (Ann Arbor, 1950), chap. 9.
"See Vikan, "Art, Medicine, and Magic," 81. Comparable
powers were attributed to heliotrope (red-flecked green calce-
dony), which may in fact be the material of this gem amulet. See
Age of Spirituality, no. 281 (wherein this gem is labeled "hematite"), and no. 398 (an anti-hemorrhaging amulet with Christian
iconography also labeled "hematite" but likely heliotrope.
"8See ibid., 83.
89Figure 20 = Baltimore, the Walters Art Gallery, nos.

TL90.9.1, 2, 3, unpublished. For published tokens from this
hoard, which in addition to the present iconographic type includes a type comprising a Greek cross with what appear to be
the letters for nike ("victory") between its arms, see Objects with
Semitic Inscriptions, no. 172. For the early Byzantine appearance
of such a formula (specifically, on bread stamps), see G. Galavaris, Bread and the Liturgy (Madison, 1970), 72-76.
of "health" conveyed with an inscribed word on
the necklace and rings is conveyed on the tokens

with their very substance.92 One form of magic was
to be worn and the other, presumably, to be con-
sumed, as we know Symeon tokens were, either

powdered and rubbed on the body or, more rarely,
taken internally with water.93
How, finally, might a hygia marriage ring or

necklace, or an earthen eulogia marriage token
with health-giving properties, have exercised its
powers to enhance the couple's chances for successful procreation? Hematite, as we have seen,
was used to control bleeding (here, presumably,
miscarriage), and other contemporary Byzantine
amulets for women were used to induce "calm" in
the uterus; indeed, one category bears an incantation addressed directly to the hystera ("womb"), accusing that "dark and black one" of "coiling like a
serpent, hissing like a dragon, and roaring like a
lion," and then admonishing it to "lie down like
a lamb."94 Why? One first assumes that this calmness would enhance the organ's fertility (or at least,
discourage dysmenorrhea), and that was probably,
on some level, the intention. But it is also true that
the "wandering womb" was believed to be the

90See Vikan, "Art, Medicine, and Magic," 67-73, 81-83.
91 Figure 21 = Paris, Robert-Henri Bautier Collection. See Vikan, "Art, Medicine, and Magic,";' 81 f (notes 109-15).
92 Some Symeon tokens are inscribed with the word hygia, and
Symeon's Vita makes it abundantly clear that this is how they

were used. See Vikan, "Art, Medicine, and Magic," fig. 2, 68-
73.
9 See Vikan, "Art, Medicine, and Magic," 69 f.
94See ibid., 77 f.
source of both physical and emotional malcontent

in a woman.95 A magical papyrus text of sixth- to
seventh-century date recommended to treat "the
ascent of the uterus" (pros metras anadromen) makes
this explicit:
I conjure you, O Womb, [by the] one established over
the Abyss . . . that you return again to your seat, and
that you do not turn [to one side] into the right part
of the ribs, or into the left part of the ribs, and that
you do not gnaw into the heart like a dog, but remain
indeed in your own intended and proper place, not
chewing [as long as] I conjure by the one who, in the

beginning, made the heaven and earth and all that is
therein. Hallelujah! Amen!
Write this on a tin tablet and 'clothe' it in 7 colors.96
Like the anularius, via its special nerve, the uterus,

by virtue of its ability to "wander," was believed to
have direct access to the heart, though not for

good but for evil. A uterus in place was one in a
state of "health," and accordingly more effective
for procreation, but a uterus in place was also one
that could not "gnaw at the heart," and thus could
not, one may assume, create in the heart that irrational hatred (alogon misos) which, as effectively as
any illness, could inhibit procreation. Thus again,
the words homonoia and hygia, under the umbrella
of marriage, could at once be complementary and
amuletic.
More or less contemporaneously with the frontal
bust portrait type appeared yet another formula

for evoking marriage, featuring three (rarely four)
standing figures, with the groom on the left, the
bride on the right, and between and slightly behind them, Christ (Fig. 22).7 Unlike the other two
95See A.A. Barb, "Diva metrix," JWarb 16 (1953), 193-238
(esp. 214 note 23); and Bonner, Amulets, 90.

96The Greek Magical Papyri in Translation, ed. H. D. Betz (Chicago-London, 1986), 123 f (Papyri graecae magicae, VII.260-71).
97Figure 22 = Washington, D.C., Dumbarton Oaks Collec-
tion, no. 61. 3. See Ross, Catalogue, no. 66. For other such rings,
see A. Banck, "Dva vizantiiskikh zolotykh perstnia iz sobranija
Ermitazha," Trudy Gosudarstvennogo Ermitazha. Kul'tura i iskusstvo
narodov vostoka 6 (1961), 31-39 (see also, Banck, Byzantine Art,

nos. 103b, 106c) (no. 103b = gold, Christ draws the couple together: homonoia[?]; no. 106c = gold, niello, and set stones, oc-
tagonal hoop, Christ presents the couple: homonoia, Psalm
5:12); J. Boardman and D. Scarisbrick, The Ralph Harari Collec-
tion of Finger Rings (London, 1977), no. 115 (gold, octagonal
hoop, Christ presents the couple, crowns: homonoia); Byzantium,
no. 41 (gold [detached bezel], Christ presents the couple,

crowns: homonoia); Cabrol and Leclercq, "Mariage," fig. 7685
(gold, Christ draws the couple together: homonoia); C. Cecchelli,

"LUanello bizantino del Museo di Palermo," Miscellanea Guillaume
de Jerphanion = OCP 13 (1947), 40-57 (gold, niello, and set
stones, octagonal hoop with locus sanctus scenes, Christ presents
the couple: Psalm 5:12); Dalton, Catalogue, nos. 129-32 (no.

129 = gold, niello, and set stones, octagonal hoop with locus
ring types, this one evokes an action, but although

the dextrarum iunctio is usually taken to be that action, only about a third of these rings actually show
the bride and groom clasping hands. On the others, they either approach one another under the
guidance of Christ (Fig. 24: euchi), or stand fron-
tally, slightly apart, as if being presented or

crowned by Christ (Fig. 25: homonoia)-or, more
rarely, by Christ and the Virgin (Fig. 26: homonoia).98 The groom appears to be wearing a tunic
and chlamys with fibula, and the bride a long robe,
sometimes with a veil-presumably the flammeum
of the Roman ceremony.99 The spiked headgear
that she wears on a few rings and, with the veil, on
the Dumbarton Oaks marriage belt (Fig. 31), is

probably just a residual feature carried over from
the coin models upon which this iconographic type
depends.100 As evidence of this, one may cite two
sanctus scenes, Christ and the Virgin bless the couple: homonoia;
no. 130 = gold and niello, octagonal hoop, Christ draws the
couple together: homonoia, John 14:27; no. 131 = gold, niello,
and set stones, octagonal hoop, Christ blesses the couple: homonoia, Theotoke boethei amen [= our Fig. 25]; no. 132 = gold,
niello, and set stones, octagonal hoop, Christ blesses the couple:
homonoia, John 14:27); O. M. Dalton, Byzantine Art and Archae-
ology (Oxford, 1911), 545 (gold and niello: John 14:27); de
Ricci, Catalogue, nos. 822, 845, 848, 861 (no. 822 = gold, dextrarum iunctio?; no. 845 = gold, dextrarum iunctio?; no. 848 = gold,
dextrarum iunctio; no. 861 = gold, dextrarum iunctio); C. Kondoleon and A. Gonosov~i, The Art of Late Rome and Byzantium: A
Catalogue of the Collection of the Virginia Museum of Fine Arts
(forthcoming), acc. no. 66.37.7 (gold, octagonal hoop, Christ
presents the couple: homonoia); Ross, Catalogue, nos. 64-66, 69

(no. 64 = gold, dextrarum iunctio; no. 65 = gold, dextrarum iunctio?; no. 66 = gold, dextrarum iunctio [= our Fig. 22]; no. 69 =
gold, niello, and set stones, octagonal hoop with locus sanctus
scenes, Christ and the Virgin bless the couple: homonoia, Kyrie
boethi tous doulous sou Petrous Theodotis, John 14:27 [= our Fig.
26]); Vikan, "Zucker," 34 f, figs. 10, 11 (fig. 10 = gold, Christ
draws the couple together: euchi [= our Fig. 24]; fig. 11 = gold,
octagonal hoop, Christ blesses the couple: homonoia, Theotoke
boethe Georgi[ou] Plakelas); W. F. Volbach, Mittelalterliche Bildwerke
aus Italien und Byzanz. Staatliche Museen zu Berlin: Bildwerke des
Kaiser Friedrich-Museums (Berlin-Leipzig, 1930), no. 6810 (gold

and niello, octagonal hoop, Christ blesses the couple?: Theotoke
boethe Meaele? [Melane?] amen).
98 Figure 24 = Baltimore, The Walters Art Gallery, Zucker

Family Collection, no. TL10.1985.048. See Vikan, "Zucker," 34
f. Figure 25 = London, British Museum. See Dalton, Catalogue,
no. 131. Figure 26 = Washington, D.C., Dumbarton Oaks Collection, no. 47.15. See Ross, Catalogue, no. 69.
99See Carcopino, Rome, 81; and compare Reinsberg, "Concordia," fig. 130 (the so-called pronuba sarcophagus, whereon the
bride wears the wedding veil).
100In his discussion of the Dumbarton Oaks marriage belt
(Catalogue, no. 38), Ross rightly observed that the trace of an

imperial diadem on the bride "indicates an imperial prototype"
(e.g., a commemorative coin or medallion from an imperial
wedding). But contrast the over-interpretation of this vestigial

headgear by L. M. Fuchs, in "The Gold Marriage Belt at Dumbarton Oaks," BSCAbstr 13 (1987), 22 f. See also Kitzinger, "Reflections," 72 note 72, for remarks on the seeming imbalance
158
GARY
VIKAN
items from an early Byzantine treasure from Tre-
bizond at Dumbarton Oaks: a marriage solidus of
Anastasius and Ariadne set as a necklace clasp

(Fig. 23) and a marriage ring whose bezel bears
what is apparently its copy (Fig. 22).101 While the

groom on the ring has been stripped of Anastasius'
imperial diadem, the bride, quite inappropriately,
has retained from Ariadne's portrait both diadem
and pendilia. Just two rings in this group show mar-
riage crowns, which, hovering over the couple's

heads, match those more frequently encountered
on frontal portrait rings. And finally, three closely
interrelated representatives of this type, including
that at Dumbarton Oaks illustrated in Figure 26,
bear tiny locus sanctus Christological scenes on the
faceted surfaces of their octagonal hoops.
There are as many known rings of this type

(about two dozen examples) as there are of the
other two types put together. All are gold, none
seems to have been designed for sealing, about a
third bear some inlaid niello, and on most of those,

the garments of the bride and groom are (or were)
inset with semiprecious stones (Figs. 25, 26).102
Well over half are inscribed on the exergue of the
bezel (with homonoia or, in one case, euchi), and

many bear secondary inscriptions as well (an invo-
cation, John 14:27 or Psalm 5:12), either on the

faceted surface of the hoop or, more rarely, on the
rim of the hoop or the bezel. About half have oc-
tagonal hoops, and about as many either a round
or an oval disk bezel; others have conical bezels or
square bezels with projecting lobes. There is much

greater variety in design, iconography, and style
among these marriage rings than among the oth-,
ers, even though there are closely interrelated

subgroups in the series, most notably that asso-
Although the terminus post quem for this iconographic type may be fixed at ca. 450, with the first
documented appearance in the East of the dextrarum iunctio showing Christ as a symbolic pronubus
on a coin type issued to commemorate the marriage of Marcian and Pulcheria (Fig. 27),104 most of
the surviving representatives of the type should be
assigned to the seventh century. This is suggested
by the fact that one of the three octagonal locus
sanctus rings, which by its weight alone (23.1
grams) presupposes an important owner, was reli-
ably reported to have been part of a treasure

found near Syracuse, where Constans II resided
with his court from 663 until his assassination five
years later.105 These three rings, for which the

iconographically related pilgrim flasks in Monza
and Bobbio (Fig. 29) provide a terminus post
quem of ca. 600, are in turn linked in design, iconography, and style to the marriage rings in Lon-
don and Leningrad cited above (note 103), which
are themselves related to another marriage ring in
London (Fig. 25), and one in Berlin.106 Furthermore, this core subgroup of seven rings includes
enough peculiarities (e.g., substantial niello decoration, inset stones, square bezels with lobes, the
John 14:27 inscription) to draw yet more rings into
the seventh century.107
Other chronological points of reference for the

type include the Dumbarton Oaks ring illustrated
in Figure 22, which may date as early as ca. 500,
and the marriage ring from Mersin, datable with
its treasure to the later sixth to early seventh century. That the former ring is only distantly related
to most of the others and that the Mersin ring has
just one close relative-the ring illustrated in Fig-
ciated with the three octagonal rings bearing locus
ure 24-together reinforce the notion that the

large subgroup around the Syracuse octagonal
sanctus scenes.103
ring constitutes, in the seventh century, the core of
the surviving evidence. Indeed, a chronological
between the bride's elaborate headgear on some of these rings

and the simpler, usually indistinct headgear of the groom. He
also implies, quite rightly, that most of these seemingly "imperial" rings (e.g., our Fig. 22) are by their light weight and lack
of technical sophistication unlikely candidates for imperial patronage.
spread of roughly 150 years would help to explain

the rich variety among these rings-as do their re-
104 Figure 27 = Glasgow, Hunterian Museum. See G. Zacos
101 Figure 23 = Washington, D.C., Dumbarton Oaks Collection, no. 59.47. See Ross, Catalogue, no. 66 (for both objects).
'102See Banck, "Zolotykh," 35-39 (a ring inset with an emerald
and a garnet).
and A. Veglery, "Marriage Solidi of the Fifth Century," NCirc

68.4 (1960), 73 f.
105On date and patronage see, most recently, Kitzinger, "Reflections," 62, and note 72. Of all Byzantine marriage rings this
one, by virtue of the headgear of its bride and groom, its Syra-
(Cecchelli, "L'anello," 40-57; Dalton, Catalogue, no. 129; and

Ross, Catalogue, no. 69), as well as Dalton, Catalogue, no. 132,
and Banck, Byzantine Art, no. 106c. Also closely related is an
cuse findspot, and its substantial weight, has the strongest claim
to courtly (if not actual imperial) patronage.
103 Included in that subgroup are the three locus sanctus rings
octagonal locus sanctus ring without explicit marriage iconography in the Walters Art Gallery. See P. Verdier, "An Early Christian Ring with a Cycle of the Life of Christ," The Bulletin of The
Walters Art Gallery 11.3 (1958).
o06For the Berlin ring, see Volbach, Bildwerke, no. 6810. For
the dating of the flasks, see J. Engemann, "Palistinische Pilger-
ampullen im F. J. Dolger Institut in Bonn," JbAC 16 (1973),
25 f.
107For the rings with these characteristics, see note 97 above.

puted findspots, ranging from Trebizond, Mersin,
and Constantinople, to Syracuse.
The immediate inspiration for this ring type,
which has no antecedent in the West,s08 was most

likely the marriage solidus issued in 450 to mark
the union of Marcian and Pulcheria (Fig. 27: feliciter nubtiis), or perhaps that of 491, commemorating the marriage of Anastasius and Ariadne (Fig.
23: feliciter nubtiis).109 These coins, in turn, drew on
a long Roman numismatic tradition analyzed by
Ernst H. Kantorowicz, Louis Reekmans, and, most
recently, by Carola Reinsberg."0 The standard

three-part configuration, involving two contract-
ing parties (usually husband and wife) clasping

right hands, and a third, facilitating party (characteristically Concordia, concord personified) between them, with hands resting on their shoulders,
was in more or less continuous use on Roman coin-
age of the later second and third centuries. Much
like the simpler "concord" compositional conceit

of juxtaposed profile portraits (Fig. 7), this formula was used to evoke, celebrate, and propagan-
cluding that by marriage in 206 of Caracalla and

Plautilla (Fig. 28: concordiafelix).111'
It was with the marriage solidus of Marcian and
Pulcheria in the mid-fifth century that this old
secular numismatic formula was Christianized,
through the substitution of Christ for Concor-
dia."2 On the level of implied ritual, this meant
that the couple was now coming together under
the guidance of Christ-pronubus instead of

Corcordia-pronuba-recognizing that in the traditional Roman wedding it was a once-married ma-
tron, the pronuba, who brought the couple together, whereas with increasing frequency in early
Byzantium it was a priest who performed that
function.113 But even if on some level of implied
ceremonial verisimilitude Christ could be imagined as standing in for his symbolic equivalent, the
priest, who in turn could be imagined as standing
in for his functional equivalent, the matron pronuba, the iconography as presented on the coins and
rings should be understood as essentially sym-
bolic.114 Thus, as the union of Caracalla and Plau-
dize all sorts of imperial and familial alliances, in-
'08 The western dextrarum iunctio marriage ring shows instead
just the joined hands of the couple. See Henkel, Fingerringe,

no. 87. Interestingly, this ring type was not taken up in Byzan-
tium.
"09There is an earlier Byzantine marriage issue, of 437 (or

439), whereon Theodosius II takes the place of Concordia in
overseeing the union of Licinia Eudoxia and Valentinian III.
See, for the three solidi, Zacos and Veglery, "Marriage Solidi,"
73 f; and Kent, Coins, no. 752. For the dynastic implications of
these issues, see Reekmans, "La 'dextrarum iunctio'," 78 f; and
K. G. Holum, Theodosian Empresses: Women and Imperial Dominion
in Late Antiquity (Berkeley, 1982), 209.
To determine that one solidus was the source of the ring type
to the exclusion of the other, it would be necessary to date securely one of the rings (or a belt) before the year 491-which
has not been done-or to differentiate between the two coin

issues iconographically, and then to isolate those distinctive
characteristics among the rings. The bride and groom are
nimbed, and Christ is apparently beardless on the earlier of the

two issues, whereas on the later solidus he is clearly bearded, and
bride and groom are without nimbi. But while Christ appears
to be beardless on some of the rings and the couple perhaps
nimbed (Fig. 24), on others he is clearly bearded, and it is ap-
parent that the bride and groom generally do not have nimbi.
This could mean that both coins exercised an influence on the
rings, although it is also possible that the rings' basic iconographic type, whichever its archetype, was successively adapted
to suit prevailing tastes. That many of these rings clearly date
well beyond the 5th century, and that at least one (Fig. 22) appears to draw directly on the 491 issue, together suggest that
the later coin may have been the proximate source of inspira-
tion for the ring type.
"0Reekmans, "La 'dextrarum iunctio'," passim; Kantorowicz,

"Marriage Belt," passim; and Reinsberg, "Concordia," passim.
Also directed toward this same basic subject, but less useful, is
an article by C. Frugoni, "L'iconografia del matrimonio e della
coppia nel medioevo," Il matrimonio nella societ& altomedievale, Set-
timane 24 (Spoleto, 1977), II, 901-63.
"'Figure 28 = London, British Museum. See H. Mattingly

and E. A. Sydenham, The Roman Imperial Coinage, Vol. IV, Part 1.
Pertinax to Geta (London, 1936), pl. xii, 7. For the chronology of
the type on Roman coinage, see Reekmans, "La 'dextrarum iunc-
tio'," 34; and Reinsberg, "Concordia," 312.

"12This transformation on the coinage was conceptually anticipated in an epithalamium by Paulinus of Nola (Carmen XXV;
Kantorowicz, "Marriage Belt," 9): "By those of his who marry
in this (Christian) law, Jesus stands as pronubus, and he changes
water into the nectar of wine." For other such evidence of the
Christianization of the traditional Roman marriage during this
period, see Ritzer, Le mariage, 130-41; and Kantorowicz, "Mar-
riage Belt," 8 f.
Outside the realm of coinage (notably, on sarcophagi and in

gold glass) the dextrarum iunctio compositional scheme had already been adapted for a Christian clientele in the West during
the 4th century (Reekmans, "La 'dextrarum iunctio'," 66-77; and
Kantorowicz, "Marriage Belt," 8, fig. 7); however, it was done in
such a way (e.g., with a diminutive Christ holding leafy crowns
over the couple) that no direct iconographic link can be established either with the 5th-century Byzantine marriage solidi or
with the 6th- to 7th-century Byzantine marriage rings.
"3See Ritzer, Le mariage, 76, 136-38. Gregory of Nazianzus
wrote to Proclus regarding the wedding of the latter's daughter

(PG 37, col. 316D; Ritzer, Le mariage, 137 f): "It is quite voluntarily that I would have assisted at this celebration, in order to
put one into the other the hands of the young couple and their
two hands in the hands of God."
"14While the dextrarum iunctio, as a graphic, age-old evocation
of the contractual and spiritual bond of marriage, was at this

period an essential ingredient in the wedding ceremony (B.
K6tting, "Dextrarum iunctio," RAC 3 [1957], 883), this threefigure iconographic configuration should not be interpreted as
a "snapshot" of some moment in the ritual (Reekmans, "La 'dextrarum iunctio'," 25), no more than the iconography of the Riha
and Stuma patens should be taken as literally accurate to a moment in the contemporary communion liturgy. Textual evidence that the pronuba(us), whether matron or priest, actually
brought the couple's hands together is slight (Reinsberg, "Con-
160
GARY
VIKAN
tilla is one overseen and characterized by Concordia, the union of Marcian and Pulcheria is one
overseen and characterized by Christ. But there is
a crucial difference, for while the ideal concord of
a pagan marriage could be achieved through the

combined virtue of the partners and could be expressed symbolically by their clasped hands (even
in the absence of Concordia), the ideal Christian

union required Christ's presence. That he was
there, united with and uniting the couple, is what

was essential, not whatever the bride and groom
might be doing in relationship to one another.
Thus, while among the Romans a ring showing
just clasped hands was sufficient to symbolize marriage and was commonly used ceremonially in con-
junction with it, among the Byzantines it would

have been insufficient, and in fact, it seems never
to have been used. Moreover, while the Romans
emphasized the couple's symbolic hand gesture in

the presence of Concordia, the Byzantines, as we
have already seen, more often than not dispensed
with it in the presence of Christ-presumably because it was no longer considered crucial.
How are the various moments or actions represented on the Byzantine rings of this type (Figs. 22,
24-26) to be interpreted? Clearly, the intention
was not to show sequential stages of an unfolding

marriage ceremony, but rather to convey symbolically the single, simple message that this marriage
was one sanctioned by Christ. Of overriding importance is the presence of Christ as the dominant
visual element; full length and at the center, he is
iconographically equivalent to the dominant cross
on rings of the frontal portrait type. But on these
rings a more graphic, ceremonially evocative message of Christian benediction is conveyed through
Christ's supervisory role in bringing the couple

physically together in the dextrarum iunctio and,
more importantly, through the imposition of his
hands upon them in blessing."115 Moreover, both ac-
tions are inscriptionally mirrored in the word euchi
("prayer") in the exergue of one ring (Fig. 24),

which undoubtedly refers to the blessing then customarily pronounced on the couple by an invited
priest.116 In those few instances where both Christ
and the Virgin appear on the ring bezel, the former blessing the groom and the latter the bride,
the intention seems simply to have been to invoke
the Virgin's blessing in addition to that of Christ,
much as the help of the Virgin is explicitly invoked
on the inscribed hoops of several of these rings

(Fig. 25: Theotoke boethei amen).117
The appearance of locus sanctus iconography on

the faceted hoops of three rings in this group (e.g.,
Fig. 26) adds further support to the idea that the
early Byzantine marriage ring-indeed, early Byz-
antine marriage art generally (Figs. 14, 18, 20,

31)-was significantly amuletic.1s Familiar from
and apparently invented for Palestinian eulogia oil

flasks of the sort preserved at Monza and Bobbio
(Fig. 29),"'9 these Holy Land scenes were trans-
ferred from those miracle-working pilgrimage artifacts to various items of jewelry which, by virtue
of their inscriptions and complementary imagery,
were clearly intended to function as amulets.120
Closest to these rings is a distinctive category of
medico-magical armband produced in the eastern
Mediterranean region during the sixth and seventh centuries (Fig. 30: heis Theos ["one God"],
"6 For early evidence of the blessing of the bride and groom
by a priest through the imposition of his hands, and for John
Chrysostom's specific references to the euchai and eulogiai that
the priest pronounces as part of the marriage benediction, see
Ritzer, Le mariage, 104-10, 134 f (PG 54, col. 443).
"'7To support Kantorowicz's ("Marriage Belt," 13) ingenious
interpretation of those rings with both Christ and the Virgin as
portraying, in response to Eph. 5:25, a typological bond between the terrestrial couple and "the exemplary concord of
King and Queen of Heaven," one would need stronger evidence

for the textual and iconographic impact of that Ephesians marriage epistle (5:22-33) in early Byzantium. And even then, one
would still have to question how mimesis could be involved, since
unlike the examples cited by Kantorowicz from Roman coinage
(his figs. 31a-c), bride and groom are here not imitating but
cordia," 312), and it would seem unlikely that the early Christian community would have so readily taken over a pagan iconographic scheme (e.g., onto sarcophagi and gold glass) if it were
understood to have had some sacramental significance. One
should also take as more symbolic than literal the passage from
Gregory of Nazianzus cited in note 113 above.
"5On all but one ring in this series (Fig. 24), Christ appears
to be touching the couple, either on the shoulders, the hands,
or the head. Since in no instance can it be seen that he (or the
Virgin) holds marriage crowns (which in two instances are
shown separately, hovering over the couple), this gesture, even

when directed toward the head, should be taken as one of blessing and not coronation. A similar situation obtains on coins of
the middle Byzantine period whereon the Virgin is seen to

touch the crowned head of the emperor in benediction. See P.
Grierson, Byzantine Coinage, Dumbarton Oaks Byzantine Collection Publications 4 (Washington, D.C., 1982), 26.
rather receiving the actions of Christ and the Virgin.

"'For this interpretation of the locus sanctus rings' iconography, see Vikan, "Art, Medicine, and Magic," 83; and Kitzinger,
"Reflections," 62.
19Figure 29 = Monza, Cathedral of St. John the Baptist, ampulla 2. See Grabar, Les ampoules, pl. v. The scenes on the ring
and flask are: Annunciation, Nativity, Visitation, Presentation in
the Temple (ring), Baptism, Crucifixion, Women at the Tomb
(flask), Christ Appearing to the Women (ring), Ascension
(flask). For an iconographic analysis of these rings, plus that at

the Walters Art Gallery sharing these scenes but lacking explicit
marriage imagery (Verdier, "An Early Christian Ring"), see En-
gemann, "Pilgerampullen," 20-22.
2"0For the issue of how, in contemporary belief and piety, this

transference of locus sanctus image, and with it, sacred power,
took place, see G. Vikan, "Sacred Image, Sacred Power," Icon

(Washington, D.C., 1988), 14-18.
r4a
/r .u
-hz4
(~4T~T A
7?7W4*
4Ia~w*#I
4O74ALu ~7N6
Aeow~opcpo&.?\ou pOLT~~ OY4~ ~P'Xd4
1 Madrid, Biblioteca Nacional, cod. 5-3, no. 2, fol. 125r, marriage miniature (photo: Madrid, Biblioteca Nacional)
4k
Yk
2 Athens, Byzantine Museum, nos. 7663a, b, marriage crowns (after Drossoyianni,

"Byzantine Crowns," pl. i)
AIAO~i~
3 Athens, Stathatos Collection, betrothal ring

(after Collection HelMne Stathatos, II, no. 33)
4 Washington, D.C.,
Dumbarton Oaks Collection,
no. 47.18, marriage ring
(photo: Dumbarton Oaks)
6 London, British Museum,
5 Washington, D.C.,
Dumbarton Oaks Collection,
marriage ring (photo: the
no. 53.12.61, marriage ring
Trustees of the British
bezel (photo: Dumbarton
Museum)
Oaks)
7London, British Museum,
medallion (photo: the Trustees

of the British Museum)
*i
8 London, British Museum, solidus

(photo: the Trustees of the British
Museum)
9 London, British Museum, solidus

Museum)
;ni~
10 London, British Museum, marriage

ring (photo: the Trustees of the
British Museum)
;Owl
41
If
3 14
. ,I, ; . m
7Ie
11 Sofia, Narodni Muzeji, no. 2519,

marriage(?) box (after Buschhausen,
Metallscrinia, no. B3)
4X:
12 Washington, D.C., Dumbarton

Oaks, no. 59.60, marriage ring
14 Houston, Menil Collection,
no. R26, marriage ring

(photo: Menil Collection)
13 Washington, D.C., Dumbarton

Oaks, no. 53.12.4, marriage ring
15 Houston, Menil Collection,
no. GW12, glass weight

(photo: Menil Collection)
17 Athens, Canellopoulos
16 Geneva, Mus&e d'Art et
d'Histoire, flat weight
(photo: Mus6e d'Art et
d'Histoire)
Museum, marriage ring

(after Spieser,
"Canellopoulos," no. 9)
,kiAl
ZW 1?
AA-?,Nm m-
18a detail
14,
-ivj
KIM,
18 Leningrad, State Hermitage Museum, no. omega 1078, marriage necklace (photo: State Hermitage
Museum)
19 New York, the Metropolitan Museum of

Art, Rogers Fund, 1958, no. 58.12,
marriage necklace (photo: Metropolitan
Museum of Art)
20 Baltimore, the Walters Art Gallery, no.

TL90.9.1, 2, 3, amuletic marriage tokens
(photo: same)
21 Paris, Robert-Henri
Bautier Collection,
amuletic token (photo:

Robert-Henri Bautier)
22 Washington, D.C.,
Dumbarton Oaks
Collection, no. 61.3,
marriage ring (photo:

Dumbarton Oaks)
23 Washington, D.C.,
Dumbarton Oaks
Collection, no. 59.47,

solidus mounted as necklace
clasp (photo: Dumbarton
Oaks)
24 Baltimore, the Walters Art
Gallery, Zucker Family

Collection, no.
TL 10.1985.048, marriage
ring (photo: the Walters
Art Gallery)
25 London, British Museum,
marriage ring (photo: the
Trustees of the British
Museum)
I'',
26 Washington, D.C., Dumbarton Oaks Collection, no. 47.15, marriage ring (photo: Dumbarton Oaks)
QIt
4b
27 Glasgow, Hunterian Museum, solidus

(photo: same)
28 London, British Museum, aureus

Museum)
kV
29 Monza, Cathedral of St. John the Baptist, ampulla 2,

pilgrim flask (after Grabar, Les ampoules, pl. v)
dpI
30 Columbia, University of Missouri Museum of Art and Archaeology (olim, Cairo, Fouquet
Collection), amuletic armband (after Maspero, "Bracelets-amulettes," fig. 1)
31 Washington, D.C., Dumbarton Oaks Collection, no. 37.33, marriage belt (photo: Dumbarton Oaks)

Psalm 90).121 Such armbands are distinguished by
a few recurrent apotropaic inscriptions-notably

Psalm 90: "He that dwells in the help of the High-
est .. ."-and by a ribbon-like design highlighted
is pursuing you." . . . I have used this method many

times, and I thought it inappropriate not to draw your
attention to it, since it has a power against the ill-
ness.125
with medallions bearing one or more of the flasks'

locus sanctus scenes plus such patently magical images as the Holy Rider, the Chnoubis, the pentalpha, and various "ring signs."122 Made of silver or,
more rarely, bronze, they were designed specifi-
In addition, there is the apparently amuletic inscription from Psalm 5 ("Thou hast crowned us
with a shield of favor"), already discussed, which
appears on one of the octagonal locus sanctus rings,
and on a second octagonal ring, without scenes.126
Chnoubis, "the Master of the Womb," specifically
Beyond the words homonoia and euchi, Psalm

5:12, and occasional invocations calling upon the
cally for women and, through the magic of the
for control of the uterus. 123 While there is nothing
Lord or the Theotokos to help husband and wife,
about the locus sanctus marriage rings to suggest

the same amuletic specificity, the parallel between
these two object types is suggestive, especially in
one additional text, the opening passage from

John 14:27, is inscribed on the hoops of several
rings in this series (Fig. 26): "Peace I leave with
"wandering womb" of the earthen marriage tokens
you, my peace I give unto you." 127 The intention

was probably to invoke a generic sort of matrimon-
light of the possible proactive role against the
(Fig. 20) discussed above.
Also amuletic is the ring's very shape, the octagon, which is shared by more than half the rings
belonging to the dextrarum iunctio iconographic
type. This is indicated by the high rate of coincidence between octagonal hoops and rings bearing
magical words (e.g., heis Theos, Psalm 90), magical
symbols (e.g., the Evil Eye, the pentalpha, "ring

signs"), and magical figures (e.g., the Holy Rider,
the Chnoubis),124 and also by the fact that the oc-
tagon is the shape prescribed by Alexander of

Tralles for medico-magical rings designed to treat
colic:
Take an iron ring and make its hoop eight-side, and

write thus on the octagon: "Flee, flee, O bile, the lark
121 Figure 30 = Columbia, University of Missouri Museum
of Art and Archaeology. See J. Maspero, "Bracelets-amulettes
d'epoque byzantine," Annales du service des antiquitis de l'Egypte 9
(1908), 246-58, fig. 1 (line drawing); and Vikan, "Art, Medicine, and Magic," 75. The scenes on the armband are: Annunciation, Nativity, Chnoubis and "ring signs," Baptism, Crucifixion, Women at the Tomb, Holy Rider, Ascension (far left).
122See Vikan, "Art, Medicine, and Magic," 75-77, and, for an
armband bearing the word hygia, fig. 10.
123 For the issue of uterine magic, see the article cited in the
preceding note. I took up the gender-specific question as it relates to these armbands in an unpublished paper entitled "The
Magic of Silver in Early Byzantium," presented as part of the
1986 NEH sponsored symposium, "Ecclesiastical Silver Plate in
Sixth-Century Byzantium," jointly hosted by the Walters Art
Gallery and Dumbarton Oaks.
124 For examples of such rings, see Vikan, "Art, Medicine, and
Magic," figs. 13, 15; and Objects with Semitic Inscriptions, nos. 25,
340, 347 (and nos. 329, 336, and 338, for faceted hoops that
are too crudely executed to be identified specifically as eight
sided). Conversely, bezels bearing invocations, monograms, or
iconic images are rarely associated with octagonal (or even faceted) hoops. Clearly, this evidence does not square with Kantorowicz's notion ("Marriage Belt," 13 f; repeated by Kitzinger,
"Reflections," 72 note 71) that the correlation was with "concord," by way of a perceived identity of shape between octagonal ring hoops and the octagonal Church of Holy Savior in
Antioch, originally devoted to Homonoia-Concordia.
ial peace, like that implicit in the word homonoia,

and that referred to repeatedly, in one form or another, in the text of the later Byzantine marriage
ceremony: "protect these your servants in peace

and harmony."128 No doubt the same marital peace
was on Dioscorus of Aphrodito's mind when, in his
"Epithalamium for Paul and Patricia," he wished
the couple:
... indissoluble harmony, as they hold children and
grandchildren on their laps, and a bright and peace-
ful life, worthy of poetic praise . . . . 129
Final mention should be made of the appearance of this dextrarum iunctio iconography outside
the realm of rings, on a pair of luxurious gold

marriage belts, one from the de Clercq collection
now preserved in the Louvre, and the other at
Dumbarton Oaks (Fig. 31: ex Theou homonoia;
charis, hygia).130 Both are datable with the marriage

'125 Alexander of Tralles, VIII.2. See Alexander von Tralles, ed.
and trans. T. Puschmann (Vienna, 1878-79), II, 377.

126 See the list of rings in note 97 above.
127 This same passage, which is otherwise unusual among the

minor arts of early Byzantium, is found on one of the amuletic
silver armbands described above. See W. Froehner, Collection de
la Comtesse R. de Bearn (Paris, 1905), 10-12. Interestingly, Theou
charis does not appear on these rings, perhaps because Christ's
active participation in the wedding benediction obviates it.
121See Trempela, "He akolouthia," 133; Kantorowicz, "Marriage Belt," 11; and Kitzinger, "Reflections," 72 note 71.
'129See MacCoull, Dioscorus, 81 f. But it is also possible that a
more specific, amuletic sort of "kardia peace" was being invoked
through the addition of John 14:17, comparable to that of the
magical uterine papyrus quoted above ("do not gnaw into the
heart"), since verse 17 concludes with the phrase "Let not your
heart be troubled, neither let it be afraid." One is reminded
again of that alogon misos which, springing from the heart, can
inhibit a happy marriage and thus procreation.
13"Figure 31 = Washington, D.C., Dumbarton Oaks Collection, no. 37.33. See Ross, Catalogue, no. 38. For the de Clercq
belt, see A. de Ridder, Collection de Clercq. Catalogue, tome VII: Les

bijoux et les pierres gravies, premiere partie, les bijoux (Paris, 1911),
162
GARY
VIKAN
rings to the sixth to seventh century, both are said
then to have functioned as well on a more practi-
gion, and both show basically the same design: two

large central repouss6 medallions linking twenty
(or 21) smaller medallions. The smaller medallions
show dionysiac figure types of various ages, pagan
couple's childbearing goal. Epithalamia like that in

honor of Athanasius suggest it, and magical papyri
substantially confirm it. A seventh-century "prayer
for pregnancy" papyrus in the Pierpont Morgan
to have come from the eastern Mediterranean re-
gods (Hermes, Apollo), and tyches, whereas the

larger clasp medallions bear the familiar dextrarum
iunctio with Christ as symbolic pronubus, much as it
appears on the solidus of Arcadius and Ariadne

(Fig. 23). Disregarding differences in style and
cal, therapeutic level in helping to fulfill the
Library contains a Coptic charm to be used by a
husband in order to get his wife pregnant.'33 First,

as if anticipating the later wedding Akolouthia, a series of biblical parallels is invoked: God made man
in his own image, God promised "our mother
quality (the Dumbarton Oaks belt is far superior),

the two belts are remarkably alike, and disregarding technique and size, their central clasps are remarkably like marriage ring bezels. Even their inscriptions are similar to those of the rings, with the
man."
God concord; grace, health," and the de Clercq ex-
priate ingredient in a late antique wedding,
Dumbarton Oaks belt bearing the words "from
ample showing "wear in good health; grace of
God" (hygienousa phori; Theou charis).
What is new is not the word hygia, but rather its

greater prominence, and, more important, the interlinking of the Christianized dextrarum iunctio
with various pagan figures, notably Dionysos. But
in fact, during late Antiquity Christ and Dionysos
could easily be accommodated under the encompassing umbrella of marriage. Again, Dioscorus of
Aphrodito provides an illuminating textual counterpart, in his "Epithalamium for Athanasius":
strong Athanasius ... I have beheld [you] as another
new Dionysos; for truly those who look upon the
wine, Love's adornment, passing it closely in goblets
one to another, have prayed to Poseidon the nurturer
for you, O bridegroom .... 11
Wine, "love's adornment," was an essential part of

late antique wedding festivities, as it remains today
and as it was at Cana.132 Moreover, wine seems
no. 1212; and E. Coche de La Fert6, L'antiquite chretienne au Musde du Louvre (Paris, 1958), no. 47. An unpublished early Byzantine belt(?) in the Hermitage, consisting of copper repousse
links, shows in the large medallion at its center the bust portrait
of a woman enframed by the words charis and hygia. This, too,
may be a marriage belt.
Among the 7th-century David silver plates there is one (Age
of Spirituality, no. 432) that portrays the marriage of David and
Michal (1 Sam. 18:27) under the authority of King Saul (-pronubus). Iconographically it is conceived as a courtly wedding,
much like that portrayed on the marriage solidus of Licinia Eu-
doxia and Valentianian III (ca. 437), where Theodosius II takes

the place of Concordia-pronuba (see note 109 above). The plate
includes as well a pair of flute players and, in its exergue, two

money bags and a modius. These were probably added to evoke
the musical merriment of a traditional Roman wedding (Ritzer,
Le mariage, 130-32), and, perhaps, the dowry gifts.
131See MacCoull, Dioscorus, 86 f. See also Macrobius, Saturn-
alia, I.18.9-10.
132 For the perpetuation of pagan festivities at early Christian
weddings, see Ritzer, Le mariage, 130-32.
Sarah" a child, et cetera. Then, in order to put the

charm to use, the husband is instructed to read its
text over a cup of wine which, when offered to his
wife, will ensure that she be "graced by the seed of
Not only was Dionysos with his wine an appro-
whether Christian or pagan, so were any number

of mythological models for love and its fulfillment.
Dioscorus' "Epithalamium for Matthew" makes
this graphically clear:

Bridegroom, bend your mind to love; Zeus himself in
heaven, because of Europa's beauty, is known to have
become a bull; for love of Leda he was esteemed a
swan. Carry your Europa over the threshold, not over
the sea; go to bed with your Leda, but don't worry
about wings.34
But what, finally, of the belt itself; where did this

seemingly utilitarian article of jewelry fit into the
marriage ritual? According to the Greek Vita of St.
Alexis, it was in the nuptial chamber that the
groom gave his bride her (ring and) wedding belt

(zone).135 And where more appropriately could this
have taken place, for following the protocol of the
Roman wedding, it was there on the nuptial couch
that the groom-after the bride had been delivered by the matron-pronuba-loosed the nodus herculeus which, hours earlier, she had fastened on the
girdle with which her wedding gown (and virgin's
modesty) had been secured.'36 Belt, wine, Christ,
and Dionysos were in that moment symbolically

and functionally intertwined in the literal fulfillment of the mandate of marriage.
The history of marriage-related art in Byzan-
tium, insofar as it constitutes an identifiable tradition in its own right, is one substantially confined
"' See F. D. Friedman, Beyond the Pharaohs: Egypt and the Copts
in the 2nd to 7th Centuries A.D. (Providence, R.I., 1989), no. 108.
134 See MacCoull, Dioscorus, 108.
"I See A. Amiaud, La lgende syriaque de Saint Alexis, l'homme de
Dieu (Paris, 1889), 12 f; and Ross, Catalogue, no. 38.
'36See Carcopino, Rome, 82.
to late Antiquity, to rings, and to the wealthy. Indeed, it is at once characterized and circumscribed
by two superb rings at Dumbarton Oaks: the

double-profile signet of Aristophanes and Vigilantia from the early fifth century (Fig. 4) and the oc-
tagonal locus sanctus amulet ring of Peter and
Theodote from the mid-seventh century (Fig. 26).

Moreover, much as the Byzantine marriage rite developed directly from the Roman ceremony, so the
Byzantine marriage ring developed directly from
its Roman antecedent. Sequentially, over half a
millennium, Rome and Byzantium engendered an

unbroken tradition of marriage art thematically
dominated by the Stoic ideal of the harmonious
couple-by paired portraits, by the dextrarum iunctio, and by the inscribed words concordia and homo-
noia. Throughout, coinage remained the crucible

for iconographic invention and the reservoir for
figure types, and throughout, iconography remained symbolic and "portraiture" anonymous.
Superficially, the bezel of a Byzantine marriage
ring of the seventh century (Fig. 26) looks much
like a Roman commemorative marriage coin of the
third (Fig. 28), and in fact, there is a direct genealogical link between the two. But over the centuries
much had changed, beyond the mere substitution
of the Greek homonoia for concordia. Not only has
Christ-pronubus taken the place of Concordia-
pronuba, the Virgin Mary has been inserted beside

him, dispensing her blessing touch to the head of
the bride just as Christ does to the groom, while
the couple stands passively apart, facing forward.
What is paramount now is no longer the traditional contractual gesture of clasped hands, but
rather Christian benediction, since among the Byz-
antines a harmonious marriage was no longer
viewed as an achievement of husband and wife,

but as a gift from God. The old Stoic ideal has here
been given a Christian reinterpretation, and reflecting that, marriage iconography and marriage
inscriptions have acquired their own distinctly
Christian character.
There is another critical distinction to be drawn
between a Byzantine marriage ring of the seventh
century and its Roman antecedent-or, for that

matter, its fifth-century Byzantine antecedent.
Over those two hundred years or so the very func-
tion of the marriage ring-beyond its immediate

ceremonial role-had been fundamentally altered, insofar as what almost invariably used to be
a signet was now more often than not an amulet
(and never a signet). Hygia inscriptions, octagonal
hoops, locus sanctus scenes, and the "shield of favor" excerpt from the Psalms all contributed to
the marriage ring's new amuletic empowerment,
which is most appropriately interpreted in light of
the demographic concerns pervading Byzantine

marriage law and poetry as relating specifically to
childbearing.
But even as late as the sixth and seventh centu-
ries, the pre-Christian roots of Byzantium's marriage art (and marriage) are still unmistakably pre-
sent, most notably on the Dumbarton Oaks
marriage belt (Fig. 31). In much the same way that
Dioscorus of Aphrodito invokes God's protection
on the wedding couple almost in the same breath
that he evokes for them the model of Leda and the
swan (while adjuring the evil eye), the goldsmith
of the Dumbarton Oaks belt interlinks Christ with
Dionysos and his thiasos, while simultaneously invoking an amuletic sort of "health," ex Theou. Such
was the breadth and richness of marriage art in
early Byzantium.
The Walters Art Gallery
Baltimore

Art and Marriage in Early Byzantium

Uploaded by

Copyright:

Available Formats

Art and Marriage in Early Byzantium

Uploaded by

Document Information

Original Description:

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

Art and Marriage in Early Byzantium

Uploaded by

Copyright:

Available Formats

Art and Marriage in Early Byzantium

Author(s): Gary Vikan

Art and Marriage in Early Byzantium

of Byzantium reflects the rituals and values of

ture on folio 125r (Fig. 1), for example, represents

the daughter of Romanos I, Lekapenos.2 This

scene, when coupled with middle Byzantine texts

experiential reality of a middle Byzantine wed-

memorate them, to the rings, crowns, and belts

ding.3 Similarly straightforward in its approach is

the ritualized union, to the gifts exchanged by the

neous neither in character nor in quantity across

125r. See Estopafian, Skyllitzes matritensis, no. 305.

The evidence, although abundant, is homoge-

riage scenes in the twelfth-century copy of the

dissertation, here cited in its French translation, Le mariage dans

les iglises chrtiennes du Ier au XIe si'cle (Paris, 1970). As for

marriage-related art of the period, the standard publication is

y miniaturas (Barcelona-Madrid, 1965), nos. 133, 221, 305, 325,

19. See also A. Grabar and M. Manoussacas, L'illustration du

manuscrit de Skylitzes de la Bibliothique Nationale de Madrid, Bib-

liotheque de l'Institut Hellenique d'Etudes Byzantines et Post-

Byzantines de Venise 10 (Venice, 1979), 159, for a brief, general

an inscribed middle Byzantine betrothal ring in

2 Figure 1 = Madrid, Biblioteca Nacional, cod. 5-3, no. 2, fol.

For the Byzantine betrothal and marriage ceremonies, see

Figure 2 = Athens, Byzantine Museum, nos. 7663a, b. See

38. Drossoyianni is overly cautious in identifying these

crowns-which are a match for those in the Madrid Skylitzes,

(Trempela, "He akolouthia," 153 f)-as specifically marriage

usual for late Roman marriage rings to invoke long life in a

general and anticipatory way on behalf of the groom's family

(see F. Henkel, Die r6mischen Fingerringe der Rheinlande [Berlin,

1913], 322). Noteworthy also are late Roman sarcophagi where-

on the dextrarum iunctio, as generally symbolizing the matrimon-

ial bond, takes place in the presence of the couple's children;

graphie romaine et palkochretienne," Bulletin de l'Institut Historique Belge de Rome 21 (1958), 68 f.

4MNHCTPON AIAOMH FOYAEAHC MAPHA. Figure 3 =

Athens, Stathatos Collection. See Collection Heline Stathatos, 2:

By contrast, the evidence of the early Byzantine

seemingly unrelated notions of concord, grace,

played an instrumental role at quite a different

juncture in the same ceremony. All of this makes

the Dumbarton Oaks belt much less a document of

an experiental reality than one of a deeper, conceptual reality.

century, it gradually emerges out of later Roman

the prenuptial arrabon or earnest money.7 Both

the ring, on account of the close connection which

links it with the principal organ.9

In late Antiquity marriage rings were worn by

During the early Byzantine period, as today, the

distinction between the betrothal ring and the

SIn addition to the customary commitment of money and a

clesiasticis officiis, 11.20.8. See Henkel, Fingerringe, 337, 341, and

11th century. See Nog 'EXX. 13 (1916), 212-21. For a pair of

(1228-34) and his wife Anna Komnene, daughter of Emperor

(London, 1941), 81.

Hadzidakis, "Un anneau," no. 90; and S. de Ricci, Catalogue of a

Collection of Ancient Rings Formed by the Late E. Guilhou (Paris,