Fairweather-Seneca The Elder PDF

Seneca the Elder
JANET FAIRWEATHER
CAMBRIDGE UNIVERSITY PRESS

Cambridge
London New York New Rochelle
Melbourne Sydney
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521231015
© Faculty of Classics, University of Cambridge 1981
This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 1981

This digitally printed version 2007
A catalogue record for this publication is available from the British Library
ISBN 978-0-521-23101-5 hardback

ISBN 978-0-521-04435-6 paperback
To my mother
CONTENTS
Preface page ix
Abbreviations xi
PART I. THE PLACE OF SENECA THE ELDER IN LITERARY HISTORY

1. Seneca the Elder: a man of his time 1
2. The declamatory anthology 27
3. The criticism 50
PART II. SENECA THE ELDER ON THE HISTORY OF ELOQUENCE

1. Oratory and rhetorical theory up to his own time 77
2. The history of declamation 104
3. The decline of rhetoric in the early Empire 132
PART III. FIVE ASPECTS OF DECLAMATION: THE ELDER SENECA'S EVIDENCE

1. Inventio 151
Divisio 152
Colores 166
2. Dispositio 179
3. Elocutio 190
Phrasis electa 190
Genus dicendi ardens et concitatum 200
Vitiosa: some general terms denoting bad taste 214
4. Memoria 228
5. Actio 235
PART IV. THE PLACE OF EARLY IMPERIAL DECLAMATION IN LITERARY

HISTORY: THE ELDER SENECA'S EVIDENCE
1. Asianism, Atticism, and the styles of the declaimers 243
2. Declamation and literary modernism in the early Empire 304
Appendix: Clausula usage of Seneca, Latro and Fuscus 326

Contents viii
Notes 328
Bibliography 377
Indexes
1. General index 385
2. Index of passages cited 409
PREFACE
The English-speaking world was without a monograph on Seneca the

Elder until the end of 1978. Now it has two, but I trust that no
reader will consider this, the second to appear, wholly super-
fluous. Certainly the recently published book by Professor L.A.
Sussman, The elder Seneca (Leiden, 1978), is in no way dependent
on any work of mine. For my part, I have for a considerable time
had access to Sussman's doctoral dissertation of 1969, but it only
became available to me three years after I had begun my Senecan
researches, by which time I had passed the stage when it might
have had a formative effect on my views. Certain points of organ-
ization in Section III of this book were suggested to me by
Sussman1s dissertation, but I am not otherwise substantially in-
debted to it. As a consequence of the publication of Sussman's
new book I have added many references to it in footnotes, but have
made only three minor alterations to my main text. Inevitably our
fields of investigation overlap, but we have to a surprising extent
chosen to concentrate our attention on different aspects of the
elder Seneca. Unlike Sussman, I have not attempted to say much
about the Senecan historical fragments of the Nachleben of the
declamatory anthology, but I have ranged over a number of areas
outside the scope of his book. Some, it may fairly be alleged, are
rather remote from the person of Seneca the Elder. Students of
classical Greek and Hellenistic education may, I hope, find some-
thing of interest in these pages, as may those interested in the
rhetoric of Cicero's day, in the Greco-Roman biographical tradi-
tion, or in Roman elegy. Thanks to the generosity of the editors
of this series I have been allowed to include in my book complex
discussion of some highly technical matters. Consequently I am
able to present an analysis of the processes involved in the
composing and declamation of controversiae and suasoriae much more
detailed than any available till now, to demonstrate how far
Preface
research into the authenticity and style of the declamatory frag-

ments may be carried without the aid of systematic lexicography,
and to give serious consideration to the ways in which the elder
Seneca as a literary critic differed from Cicero and Quintilian in
aims and presuppositions.
This book is a revised version of a Ph.D. dissertation sub-
mitted to the University of London in 1977. To Dr J.B. Hall, my
former supervisor, I owe a particular debt of gratitude for his
seemingly inexhaustible good cheer, his detailed and patient
criticism of my work, and for various helpful suggestions for the
restoration of the elder Seneca's text. The authorship of these
conjectures is duly acknowledged wherever they are considered, as
is my indebtedness to Dr Miriam Griffin and Dr M. Winterbottom for
discussion of various points of interpretation. I am grateful to
Dr Winterbottom and Professors E.J. Kenney and C O . Brink for
reading and commenting on my work and for recommending its accept-
ance by the editors of Cambridge Classical Studies. I must also
express gratitude to the Department of Education and Science for
the award of a Major State Studentship (1967-70) and to the
Mistress and Fellows of Girton College, Cambridge who, by electing
me to the Jex-Blake Research Fellowship for 1973-6, enabled me,
after a period spent chiefly in Canada engaged in Greek studies,
to resume work on a project which could hardly otherwise have been
completed. Finally, the kindness and generosity shown to me in
the darker days of my career by my mother, to whom this book is
dedicated, should not go without recognition.
Cambridge J.A.F.
June 1979
Postscript.
I wish to thank Mrs Janet Chapman and the officers of the Cambridge
University Press for their part in the production of this book and
also Dr Marie Lovatt and Dr J.C. McKeown for their assistance with
proof-reading.
ABBREVIATIONS
Brugnoli Suetonius, De grammaticis et rhetoribus,

ed. G. Brugnoli (Leipzig, 1963)
Diels/Kranz, Vorsokr. Die Fragmente der Vorsokratiker, ed. H.
Diels, W. Kranz, 3 vols. (Berlin-Grunewald,
1951-2)
FGrHist Die Fragmente der griechischen Historiker,
ed. F. Jaco.by (Berlin/Leiden, 1923-58)
GL Grammatici Latini, ed. H. Keil, 8 vols.
(Leipzig, 1855-80)
GRF Grammaticae Romanae Fragmenta, ed. H.
Funaoli (Leipzig 1907)
HRR Historicorum Romanorum Reliquiae, ed. H.
Peter (Leipzig, 1870-1906; vol.1 ed.2
1914-16)
9
LSJ A Greek-English Lexicon, ed. H.G. Liddell,
R. Scott, H.S. Jones, with supplement
(Oxford, 1968)
OLD Oxford Latin Dictionary (Oxford, 1968 - )
ORF Oratorum Romanorum Fragmenta liberae rei
publicae, ed.3 H. Malcovati (Turin, 1967)
PIR2 Prosopographia Imperii Romani saec.I, II,
III, ed.2 E. Groag, A. Stein (Berlin/
Leipzig, 1933)
RE Paulys Real-Encylop'ddie der classischen
Altertumswissenschaft (Stuttgart, 1894 - )
RLM Rhetores Latini Minores, ed. K. Halm
(Leipzig, 1863)
Rh.Gr. Rabe Hermogenis Opera (Rhetores Graeci VI), ed.
H. Rabe (Leipzig, 1913, repr. Stuttgart,
1969)
Rh.Gr. Spengel Rhetores Graeci, ed. L. Spengel, 3 vols.
(Leipzig, 1853-6, repr. Frankfurt, 1966)
Rh.Gr. Spengel-Hammer Rhetores Graeci ex recognitione Leonardi
Spengel (Leipzig, 1894)
Rh.Gr. Walz Rhetores Graeci, ed. C. Walz, 9 vols.
(London and elsewhere, 1832-6, repr.
Osnabriick, 1968)
ThLL Thesaurus Linguae Latinae (Leipzig, 1900 - )
Abbreviations xii
Note: For an explanation of symbols used in references to the text

of Seneca the Elder and other ancient authors see p.377.
PART I
THE PLACE OF SENECA THE ELDER IN LITERARY HISTORY
1 SENECA THE ELDER: A MAN OF HIS TIME
The elder Seneca had every reason to consider himself singularly

well-equipped for the task he was undertaking when he compiled the
work which has come down to us under the title Oratorum et
rhetorum sententiae divisiones colores. This task was to pro-
vide a detailed record of the rise of the type of rhetorical dec-
lamation which had become a remarkably important feature of the
literary life of Rome in the early years of the Empire, that is,
the declamation of controversiae (speeches for the prosecution or
defence in imaginary court cases) and suasoriae (speeches offering
advice to mythical or historical figures as they face crucial de-
cisions) , in public as well as in private, by adult amateurs -
often eminent men - as well as by schoolboys and professional
teachers of rhetoric.
Born most probably in the fifties B.C.,2 at any rate back in
the days when Cicero had still been alive, by the time he came to
write his work on the declaimers the elder Seneca was able to draw
on an unusually intimate knowledge of his subject acquired over
seventy years or more. He could remember the days before public
declamation by non-professionals had become accepted practice
(Contr. X pr. 4: ...nondum haec consuetudo erat inducta), and he
knew that it was in the last years of the Republic, when he was a
child, that the word declamare had first been used as a technical
term to describe the habit, newly popular among Roman orators, of
delivering practice speeches in the privacy of their homes (Contr.
I pr. 12). He therefore felt able to refer to the declamation
with which his readers would be familiar as something which had
been born within his own life-time: ...facile est mihi ah incun-
abulis nosse rem post me natam (ibid.).
Actually, most of the features of modern declamation had been
emerging long before its birth in its final form.3 Many of the
declamation themes recorded in Seneca's collection show signs of
The place of Seneca the Elder in literary history 4
having been invented for use in Greek schools of rhetoric, prob-

ably several centuries earlier,4 and there is evidence that ju-
dicial themes of the controversia type were already being set as
exercises for Roman schoolboys in the earlier half of the first
century B.C.5 We also have the testimony of Suetonius that the
grammaticus Antonius Gnipho was giving regular public declamations
in Latin and Greek to distinguished audiences in the year of
Cicero's praetorship, 66 B.C.6
Seneca was aware that some kind of declamation had been known
before his time, but he seems to have underestimated the degree of
continuity between old and new. In Contr. I pr. 12 he presents an
account of Roman declamation according to which it had gone
through three distinct phases: in the first the thesis had been
the dominant type of exercise; then came the exercises which
Cicero called causae; finally the controversia had come to the
fore. This account over-simplifies the facts, and is in some ways
positively misleading.7 It does not tell us, for instance, that
Cicero in his maturity came to regard the thesis as preferable to
less abstract types of exercise as a means of educating the ora-
tor, or that themes of the controversia type had been invented
long before that name had been applied to them. In ignorance of
this last fact, Seneca could set about his task with a light heart,
not worrying that declaimers active earlier than his life-time
might have anticipated his contemporaries in their treatment of
the standard themes.^
The one great speaker he regretted not having heard was
Cicero. Age alone had not prevented him but, because of the ex-
treme difficulty of travel between Spain and Italy during the
Civil Wars, he had been kept at home as a child in Corduba, his
birth-place,10 instead of being sent to Rome for his entire edu-
cation, as was to be, and may already have been, the custom with
children in his family:11
omnes autem magni in eloquentia nominis excepto Cicerone videor
audisse; ne Ciceronem quidem aetas mihi eripuerat, sed bellorum
civilium furor, qui tune orbem totum pervagabatur, intra coloniam
meam me continuit: alioqui in illo atriolo in quo duos grandes
praetextatos ait secum declamasse, potui adesse illudque ingenium,
quod solum populus Romanus par imperio suo habuit, cognoscere et,
quod vulgo aliquando dici solet, sed in illo proprie debet, potui
Seneca the Elder: a man of his time 5
vivam vocem audire. (Contr. I pr. 11)

Even supposing Seneca had been in Rome in the late Republic,
it is improbable that a mere provincial boy would really have been
admitted to the atriolum where Cicero and the consuls-designate
declaimed, about as probable as the ancient tradition, with which
Seneca's wishful thinking about overlapping chronology may be
compared, that Demosthenes (384-322 B.C.) had been known by his
great predecessor Lysias (c.459-c.38O B.C.). 1 2 As it was, by
staying in Spain he had the chance to meet at a very early age a
boy who was to become the leading declaimer in Rome in the years
that followed Cicero's death. Seneca's close friendship with
Porcius Latro, which lasted a prima pueritia usque ad ultimum eius
diem (Contr. I pr. 13), must have been one of the main reasons for
the enthusiastic interest in declamation which he retained as an
adult.
Seneca and Latro went as condiscipuli to the school of a me-
diocre rhetorician called Marullus (Contr. I pr. 22). Whether his
school was in Spain or Rome is unclear, but the former hypothesis
is by no means improbable.1^ According to Suetonius, grammatici
had long been active in the western provinces (Be gramm. et rhet.
3.6 Brugnoli), and in the Republican period their curriculum had
often included rhetoric in addition to the usual grammatical
studies (4.6). Even supposing that rhetoric had never been taught
in Corduba before the elder Seneca's boyhood, the educational
emergency brought about by the Civil Wars, when the sons of even
the wealthiest families could not be sent to Rome, was just the
kind of situation likely to prompt an enterprising rhetorician to
set up a school there. The outstanding gifts of the two boys were
apparent from early days. Seneca displayed a prodigious capacity
for memorizing vast quantities of unrelated data:
nam et duo milia nominum recitata quo erant ordine dicta reddebam,
et ab his qui ad audiendum praeceptorem mecum convenerant singulos
versus a singulis datos, cum plures quam ducenti efficerentur, ab
ultimo incipiens usque ad primum recitabam. (Contr. I pr. 2)
At declamation itself he was out-shone by Latro, who worked his
way to the top of Marullus' class (Contr. I pr. 24), but he had
just the talents to fit him to play Boswell to Latro's Johnson.
However, when Latro set himself up in Rome as a professional
The place of Seneca the Elder in literary history
rhetorician, Seneca did not confine his attention to his friend's

school alone. Rome in the last three decades B.C., the time when
Latro was active there, was full of men more or less eager to dis-
play their declamatory skills to the public, and Seneca, to judge
from the number of declaimers cited in his work, seems to have had
plenty of leisure to go and hear as many of them as he wished.
We do not know exactly how often the leading professional
rhetores would declaim to the public, but that five or six times a
year was thought infrequent can be deduced from what Seneca says
about Albucius:
Instatis mihi cotidie de Albucio: non ultra vos differam, quamvis
non audierim frequenter, cum per totum annum quinquiens sexiensve
populo diceret <et> ad secretas exercitationes non multi in-
rumperent; quos tamen gratiae suae paenitebat. (Contr. VII pr. 1)
Antonius Gnipho, according to Suetonius, had declaimed non nisi
nundinis.11* Some of the professionals liked to demonstrate their
virtuosity by declaiming for very long stretches of time: when
Albucius did declaim to the public, he might go on for hours on
end: saepe declamante illo ter bucinavit (Contr. VII pr. I ) , 1 5
and we hear of one occasion when Latro, incensed by hostile criti-
cism of his eloquence, declaimed a suasoria for three days (Contr.
II.4.8). Some of these professional recitals were the occasion of
distinguished social gatherings. For instance, one of Latrofs
declamations was attended by Augustus in the company of Agrippa
and Maecenas (Contr. II.4.12f.). They rivalled poetry recitations
(a parallel development in the social life of Rome)* 6 for the
attention of the city's literary connoisseurs.
It was often possible to listen to declamations by up-and-
coming students in the schools, as well as the masters, but dif-
ferent rhetores differed in the extent to which they let their
pupils perform. Latro refused even to listen to his pupils'
efforts, let alone to correct their work or allow them to declaim
publicly in his school - declamabat ipse tantum et aiebat se non
esse magistrum sed exemplum - but he and the Greek declaimer
Nicetes were alone in being able to take this egotistical line
(Contr. IX.2.23). Their pupils were obliged to display their tal-
ents elsewhere. Thus Seneca heard Latro's pupil Florus declaiming,
but non apud Latronem (ibid.). Other rhetoricians let their
Seneca the Elder: a man of his time
students perform in their own schools, and ran the risk of being
up-staged by them. Among these was Cestius Pius, one of whose
young pupils, Alfius Flavus, attracted such large audiences that
the master hardly dared declaim immediately after him (Contr.
1.1.22).
Sometimes adult amateurs, including men well established in
public life, were invited to declaim in a school to an audience of
students and other interested people. Aietius Pastor, according
to Seneca, declaimed a certain controversia in Cestius* school
when he was already a senator (Contr. 1.3.11). Probably decla-
mations were also given by amateurs to invited guests in private
houses: if M. Lepidus, before whom the senator Scaurus declaimed
(Contr. X pr. 3 W ) , was the consul for A.D.ll he can hardly have
kept a regular school, even if he is identical to the teacher of
Nero (son of Germanicus) mentioned in Contr. II.3.23. Other
public figures were more self-effacing about their declamatory
pursuits, viewing declamations as mere exercises for the real
business of forensic speaking (Contr. Ill pr. 1: domesticas
exercitationes), rather than as an excuse for self-advertisement
(Contr. X pr. 4: frivolae iactationis). Their motives for adopt-
ing this view-point were, according to Seneca, usually mixed: good
orators could be disappointing as declaimers, sometimes they
realized this and acted accordingly (Contr. Ill pr. 1; IV pr. 2);
they might also reject public declamation as part of an effort to
present themselves as defenders of old-fashioned austerity (Contr.
X pr. 4 ) .
Seneca somehow managed to hear even the most reluctant de-
claimers at their private exercises. He was probably among those
who barged in (inruperent) to hear the private declamations of
Albucius; at least, he gives quite a detailed account of why it
was a waste of effort to do so (Contr. VII pr. 1). He was able to
criticize the declamatory manner of Cassius Severus (Contr. Ill
pr. Iff.), who declaimed only raro...et non nisi ah amicis coactus
(Contr. Ill pr. 7 ) , and of T. Labienus, who never admitted the
public (Contr. X pr. 4 ) . He even had the chance to see Asinius
Pollio as an old man giving a lesson in declamation to his grand-
son Marcellus Aeserninus (Contr. IV pr. 3) - a remarkable
privilege, perhaps to be explained by a friendship between Pollio

and Senecafs family originating in the Civil Wars when, in 43 B.C.
Pollio had his headquarters in Corduba.17
What Seneca was doing in these years apart from listening to
declamations is unclear. There is no evidence that he was ever a
rhetorician himself. Any plans he may have had for a political
career were abortive, and how serious his aspirations in that di-
rection may have been cannot be determined now. In a digression
addressed to Mela, his youngest son, in Contr. II pr. 3f., he
refers to his elder sons1 entry into politics in quibus ipsa quae
sperantur timenda sunt, and then has this to say about himself:
ego quoque eius alioqui processus avidus et hortator laudatorque
vel periculosae dum honestae modo industriae duobus filiis
navigantibus te in portu retineo.^Q There is a certain ambiguity
in the word alioqui. Does it have a purely temporal sense here,
in which case does it mean 'at one time1, so that we have a con-
fession here on the elder Seneca's part of early political am-
bitions, or merely 'at other times'? Or does it mean 'otherwise',
that is, either 'if circumstances were different' (i.e. if poli-
tics were not so dangerous), or 'in other cases' (pointing a con-
trast between his ambitiousness with regard to other people's
careers and his contentment with Mela's lack of ambition)?
In her recent work on the Senecas*9 Mrs M. T. Griffin assumes
that it means 'at one time' and that Seneca the Elder is recalling
that he once had eager political ambitions himself. The strongest
argument in favour of this view is that avidus + genitive seems
not to be used of mere warm approval for other people's actions,
but rather has to mean 'greedy on one's own behalf';20 an octogen-
arian, such as the elder Seneca must have been by the time of
writing, could hardly be 'otherwise' or 'at other (recent) times'
eager for his own advancement. Against this view it must be ob-
jected that in none of the other instances of the use of alioqui
by Seneca the Elder or the declaimers he quotes which are cited in
the Thesaurus Linguae Latinae,21 does it appear to have unambigu-
ously a temporal sense equivalent to olim.22 Most often in their
usage it seems to mean 'otherwise', 'if this were not so'.2^ So
perhaps we ought to argue that greed on behalf of one's family was
not so far removed from personal greed that avidus could not be
used with reference to it. Also worth considering is the recent
Oh.*
suggestion of J. B. Hall ZH that et should be deleted after avidus,

so that we should read: ego quoque eius alioqui processus avidus
hortator laudatorque vel periculosae dum honestae modo industriae
... The difficulty of avidus is removed if it qualifies hortator.
It is clear that Seneca the Elder remained an egues,25 for
whatever reason, all his life, but there is nothing to show that
he was involved in any of the standard occupations of his class,
private business or public financial administration. For all we
know, the running of his family's estates may have been his only
responsibility. His wide knowledge of the Roman schools shows
that he must have spent a good deal of time in the capital, but it
is possible that he travelled quite frequently between there and
Corduba, witness in particular his interest in Spanish orators,
quibus, quo minus ad famam pervenirent, non ingenium defuit sed
locus (Contr. X pr. 13) , 2 6 The journey need not have been unduly
arduous in peace-time. When the elder Pliny (NH XIX.1.4) declares
that the voyage from Gades to Ostia could be made under sail in
less than seven days, he is assuming perfect weather conditions,
but the winds would have had to be consistently adverse for it to
take as long as a month. 27 In view of this we need not assume
that a man, having arrived in Rome from Spain, would necessarily
stay there for years on end. It is therefore even more futile
than is generally recognized to attempt to trace the elder
Seneca's movements over the years. We do know that at least one
of his three sons by Helvia, the lady, very much his junior,28 to
whom we find him married in the last decade B.C., was born in
Corduba, but that this child, the future philosopher, was in due
course taken to Rome by an aunt, 'illius manibus in urbem perlatus
sum'.29
Seneca's interest in declamation lasted well past the time of
Latro, who tragically took his own life to escape from a quartan
fever, according to St Jerome, two years before the birth of
Christ.30 The occasion on which Seneca heard Asinius Pollio,
already an old man (iam senem), giving a lesson to his grandson
(Contr. IV pr. 3) may have been within Latro's life-time,31 given
that Pollio had been born in 76 B.C., but several of his refer-
ences to declaimers belong to a rather later period. The incident
in which Cestius Pius unkindly criticized the young Quinctilius
Varus, son of the general, at the time when he was Germanici gener
ut praetextatus (Contr. 1.3.10) can hardly have happened before
A.D.16.32 The education of his sons gave him an excuse, if one
were needed, for keeping in touch with the world of rhetoric in
his old age. He refers with disgust to the style of Musa, a
rhetorician whom his sons occasionally went to hear (Contr. X pr.
9); he tells us that he had actually been with them, evidently on
more than one occasion, to hear Scaurus declaim (Contr. X pr. 2 ) .
In the latter years of his life he must have found some compen-
sation for the loss of Latro in friendship with another distin-
guished declaimer, Iunius Gallio. Repeatedly in addressing his
sons he refers to him as Gallio noster or vester.^3 He rates him
among the four greatest declaimers of his day, indeed as Latro's
only serious rival:
primum tetradeum quod faciam, quaeritis? Latronis, Fusci, Albuci,
Gallionis. hi quotiens conflixissent, penes Latronem gloria
fuisset, penes Gallionem palma. (Contr. X pr. 13)
Novatus, SenecaTs eldest son, later assumes the name of Iunius
Gallio,34 presumably adopted by his father's friend. One of the
lost prefaces must surely have been devoted to praise of Gallio as
a declaimer.
About the political inclinations of his family and friends in
these years he is uncommunicative.35 His two references to
Sejanus, key figure in the politics of Tiberius' reign, are not
illuminating: in Contr. IX.4.21 it is with the wit of Asilius
Sabinus, who happened to encounter in prison certain Seianianos
locupletes, that he is primarily concerned, and in Suas. 2.12
his main intention is to praise the eloquence of Attalus the
Stoic; the fact that this philosopher was exiled by Sejanus is
only mentioned in an aside. We do know that Iunius Gallio suf-
fered exile on the grounds (unsubstantiated) of association with
Sejanus, after the disastrous misfiring of something which Tacitus
classed as meditata adulatio (Ann. VI.3). Events of this kind
must have left a mark on the elder Seneca's attitudes and have to
be considered as part of the background to his memorable words on

politics fin which the very objectives which are sought after are
to be feared1 in Contr. II pr. 4.
The digression in the second preface in which these senti-
ments are expressed is a most interesting passage, full of clues
(unfortunately open to varying shades of interpretation) about the
elder Seneca1s outlook at the time when, surprisingly late in
life, Novatus and Seneca were setting out on the cursus honorum.
haec eo libentius, Mela, fili carissime, refero, quia video
animum tuum a civilibus officiis abhorrentem et ab omni amjgitu
aversum hoc unum concupiscentem, nihil concupiscere. t u ^
eloquentiae tamen studeas: facilis ab hac in omnes artes discursus
est; instruit etiam quos non sibi exercet. nee est quod insidias
tibi putes fieri, quasi id agam ut te bene cedentis studii favor
teneat. ego vero non sum bonae mentis impedimentum: perge quo
inclinat animus, et paterno contentus ordine subdue fortunae mag-
nam tui partem. erat quidem tibi maius ingenium quam fratribus
tuis, omnium bonarum artium capacissimum: est et hoc ipsum
melioris ingenii pignus, non corrumpi bonitate eius ut illo male
utaris. sed quoniam fratribus tuis ambitiosa curae sunt foroque
se et honoribus parant, in quibus ipsa quae sperantur timenda
sunt, ego quoque eius alioqui processus avidus et hortator
laudatorque vel periculosae dum honestae modo industriae duobus
filiis navigantibus te in portu retineo. sed proderit tibi in
ilia quae tota mente agitas declamandi exercitatio, sicut Fabiano .
profuit: qui aliquando cum Sextium audiret nihilominus declamita-
bat, et tarn diligenter ut putares ilium illi studio parari, non
per illud alteri praeparari. (Contr. II pr. 3f. W)
Seneca the Elder here breaks off from his pen-portrait of Fabianus,
declaimer turned philosopher, and suggests that what he has been
saying (haec) ought to be of special interest to his youngest son
Mela. Why this should be so does not emerge till the last sen-
tence of the passage quoted. It appears that Mela had decided to
live a life of philosophical quietude, far from the world of poli-
tics, and therefore could see no point in continuing his rhetori-
cal studies. His father puts the case of Fabianus before him as
an object lesson: here was a distinguished philosopher who had
continued declaiming after his philosophical conversion, and with
no diminution of effort.
The contrast between the unambitious Mela and his brothers
with their political aspirations calls forth some remarkably
interesting observations from their father, the exact implications
of which, however, are often far from clear. Mrs Griffin in her
recent work on the Senecas has taken Contr. II pr. 3-4 as evidence
that the elder Seneca, frustrated in his own early ambitions for a
political career, looked upon public life as honesta industria
and, having perhaps gone as far as to resort to plots (insidias)
in order to ensure that Novatus and Seneca should be well trained
as orators, gave them whole-hearted backing when they decided to
embark on their 'noble voyage'.37 A. Vassileiou38 has also re-
cently asserted that Seneca the Elder was very ambitious on his
sons1 behalf, though he does not view these ambitions as the out-
come of an early disappointment in his own career. Neither of
these two interpretations of the passage pays sufficient attention
to the fact that it is Mela, the philosophical quietist, whom
Seneca the Elder is explicitly commending in the passage, and that
in doing so he says things which are extraordinarily ungenerous to
the two elder brothers. What had they done to deserve the as-
persions cast on their intellects and the use they were making of
their talents in the words est et hoc ipsum melioris ingenii
pignus, non corrumpi bonitate eius ut illo male utarist Neverthe-
less, elsewhere in the same passage the elder Seneca does appear to
suggest that he gave strong backing to his elder sons' ambitions.
The suggestion that Seneca's denial of plots indicates that
he had gone to all lengths to ensure that Novatus and Seneca en-
tered politics may readily be discounted: all that he denies in
Contr. II pr. 3 is that he wants Mela to keep up his study of
rhetoric merely because it is something which he is good at; in
the next sentence he makes it clear that he regards rhetoric as a
useful training for other occupations besides public life. It is
in the sentence where he refers to public life and honesta
industria that we find him expressing himself in such a way that
it seems difficult to argue, as one might have done on the
strength of his previous remarks, that he approved of Mela, but
not of his two elder sons.
Three problems of interpretation are raised by this long sen-
tence. What is the precise, significance of the opening words, sed
quoniami What is the meaning of alioquil And why does the sen-
tence end without any expression of pleasure by Seneca the Elder
at the prospect of keeping Mela in harbour?39 To start with the
second question, we have noted the difficulty of finding examples

in the usage of Seneca and the declaimers quoted by him of the use
of alioqui as equivalent to olim, so that unless one is convinced,
as I am not completely, by the argument that eius...processus
avidus cannot possibly refer to the elder SenecaTs avidity for any
but his own advancement, it seems hazardous to interpret alioqui
in this narrowly restricted sense to the exclusion of any other,
and view the words eius...avidus as an unambiguous confession of
early political ambitions on his part. The most natural inter-
pretation does seem to be that alioqui means 'otherwise', 'in
other cases1, i.e. in particular, Tin the case of my other
sons',^ even though, as we have seen, he casts doubts on the wis-
dom of their ambitions in previous sentences. The interpretation
1
of alioqui as meaning 'otherwise in the sense 'if public life
were not so dangerous' (referring back to the words, in quibus
ipsa quae sperantur timenda sunt) has a certain attraction, as it
would make the elder Seneca's avidity for political advancement
conditional on circumstances not found in the Rome of his day, and
hence not inconsistent with his praise of Mela's quietism. But it
is difficult to make sense of ego quoque if one adopts this
interpretation.
The problems raised by the beginning and the end of the sen-
tence are best considered together. The only solution to them, I
suggest, is to postulate that the text at least at the end of the
sentence is corrupt. Some expression there of warm indulgence for
Mela seems demanded by the context: perhaps between navigantibus
te and retineo the word libenter has dropped out, owing to its
slight resemblance to ...tibus te, or, to adopt an alternative
suggested by J. B. Hall, retineo is a corruption of retin<ere
gaud>eo or the like.41 If we accept that some such supplement is
required at the end of the sentence, quoniam at the beginning
becomes much more intelligible. Sed before quoniam still seems
puzzling. A connective meaning 'and' or 'therefore' might have
been expected. Perhaps it is a corruption of et 4 2 or ita^^
(the intrusive s being the product of dittography of the last
letter of the preceding word utaris), in which case quidem in the
previous sentence is picked up by sed not here but later on, in
the emphatic words sed proderit tibi...declamandi exercitatio...

Translate then, perhaps:
'But' (or 'And1 or 'Therefore1) 'since your brothers are busying
themselves with ambitious aims and preparing themselves for the
forum and public careers, in which the very objectives which are
sought after are to be feared, even I, though in other cases eager
for this advancement and an encourager and praiser of hard work,
however dangerous' (or, deleting et, 44 'though in other cases an
eager encourager of this advancement and a praiser of hard work,
however dangerous') '<am glad to> keep you in harbour.'
The difficulty remains that the elder Seneca does not seem
consistent in the attitude he expresses towards public office in
Contr. II pr. 3-4, one moment implying that it is a bad use of the
intellect, and the next representing himself as an enthusiast for
it despite its dangers. Prolonged consideration of the passage
leaves me with an uneasy feeling that Seneca the Elder, thoroughly
familiar with the arguments for and against the thesis, 'an
capessenda sit sapienti respublica1,45 was all too ready to be
deflected from a clear expression of his views into phrase-making
appropriate to whichever side of the debate he happened to be
touching on.116
But, whatever his true attitude was towards his elder sons'
ambitions, two interesting facts do come to light in Contr. II pr.
3-4 which we could not have learnt from any other source, first,
that it was Mela and not Seneca who was regarded when young as the
philosopher of the family, and secondly that when the elder sons
entered politics their father was to some extent conscious of the
dangers into which their aspirations might lead them. He was not
to know that Mela's decision to remain paterno contentus ordine
would not save him from the calamity which was to befall the whole
family.47
The Oratorum et rhetorum sententiae divisiones colores was
one of two literary projects to which the elder Seneca devoted
himself in the last years of his life, the other being a study of
recent Roman history, starting from the beginning of the Civil
Wars, which he kept up to date practically until the day he
died.48 He probably regarded this other work as much the more
important, being well aware that historiography was a weightier
matter than scholastic rhetoric (see Suas. 5.8; 6.16).
That the declamatory anthology was composed at least in part

in the last years of his life, is shown by various references
which could not have been made before the 30s A.D.: an allusion to
the fall of Sejanus in 31 (Contr. IX.4.21); another to the death
of Scaurus in 34 (Suas. 2.22). The fact that he refers to the
writings of Cassius Severus (Contr. Ill pr. 3) and Cremutius
Cordus (Suas. 6.19, 23), whose books were publicly burnt in A.D.
12 and 25 respectively, not to be republished until early in
Caligula's reign, has been adduced as evidence that the passages
in question were not written before 37. 4 9 This is open to doubt,
for it is unlikely that any attempts in antiquity to suppress
books were rigorous enough to stop their circulation in private
hands, and Seneca, who was over eighty in the 30s A.D., might have
thought it an old man's privilege to write about anything he
chose, and to leave it to his literary executors to decide the
right time for publication. But as one historical fragment which
may well be by the elder Seneca describes the death of Tiberius,50
a date in the reign of Caligula for the compilation of the de-
clamatory anthology is by no means to be ruled out.
It seems that after his death, which we know has to be dated
before 41, 5 1 the younger Seneca decided to suppress at least some
of his father's works, though in the biography which he wrote De
vita patris, now lost except for one fragment, he was full of
praise for them:
si quaecumque composuit pater meus et edi voluit, iam in manus
populi emisissem, ad claritatem nominis sui satis sibi ipse pro-
spexerat: nam nisi me decepit pietas, cuius honestus etiam error
est, inter eos haberetur, qui ingenio meruerunt ut puris titulis
nobiles essent.52*
The elder Seneca's decision to work in his old age on two
very different projects - the political history and the work on
the declaimers - is one of several pieces of evidence that he was
a much more complex character than is commonly supposed. Even
within the declamatory anthology we find indications that he was
one of those people who have, in private life, an irrepressible
charm and sense of humour which it takes only friendships, or
even the memory of friendships, to keep alive, but who are at the
same time profoundly pessimistic and austere in outlook when they
consider the state of the world as a whole. The younger Seneca

has this to say about his father's Histories:
quisquis legisset eius historias ab initio bellorum civilium, unde
primum veritas retro abiit, paene usque ad mortis suae diem, magni
aestimaret scire, quibus natus esset parentibus ille, qui res
Roma<nas...53*
The parenthesis, unde...abiit, is interesting. It has to be taken
as the younger Seneca's own comment, but he would surely not have
made it if it did not reflect a dominant theme in his father's
work. The phraseology should be compared with Contr. I pr. 6:
...ut possitis aestimare, in quantum cotidie ingenia decrescant
et nescio qua iniquitate naturae eloquentia se retro tulerit. It
is of course hazardous to speculate on the character of Seneca's
historical work,54 now so hopelessly lost, but it seems very un-
likely that it was wholly cheerful in tone. True, in the decla-
matory anthology he has some kind things to say about Augustus as
a 'most clement man' (Contr. IV pr. 5) and as a ruler who allowed
himself to display human feelings and friendliness in his letters
(ibid.), while allowing others to insult his right-hand men with-
out fear of retribution: tanta autem sub divo Augusto libertas
fuit, ut praepotenti tune M. Agrippae non defuerint qui ignobili-
tatem exprobrarent (Contr. II.4.13). It also seems that he had
come to terms in some ways with the fact that the Republic was no
more: mihi videtur admiratione dignus divus Augustus, sub quo
tantum licuit, sed horum non possum misereri, qui tanti putant
caput potius quam dictum perdere (ibid.); he finds it remarkable
that Labienus retained his enthusiasm for the Pompeian cause in
tanta pace (Contr. X pr. 5 ) . On the other hand, against book-
burning, the punishment unknown in the Republic which Labienus was
the first to suffer, he vents great fury (Contr. X pr. 6f.), only
equalled by that with which he inveighs against the young men of
his day who have neglected eloquence and chosen instead to aim for
turpia multo honore quaestuque vigentia (Contr. I pr. 7 ) . We have
seen in Contr. II pr. 4 his realization of the dangers which faced
men entering public life in post-Augustan Rome.
In a historical fragment ascribed to 'Seneca' by Lactantius,

the rise and fall of Rome is described in terms of the stages in a
human life; the Empire with its return to monarchy is reckoned as
the second childhood of the state:

haec fuit prima eius senectus, cum bellis lacerata civilibus atque
intestino malo pressa rursus ad regimen singularis imperii recidit
quasi ad alteram infantiam revoluta. amissa enim libertate, quam
Bruto duce et auctore defenderat, ita consenuit, tamquam sustent-
are se ipsa non valeret, nisi adminiculo regentium uteretur. (Fr.
hist. 1 = Lact. Inst. div. VII.15.14)
It is open to dispute whether this gloomy assessment of Roman
history belongs to Seneca the Elder or his son the philosopher.55
In my view there is a limit to the lengths to which one should go
in order to deny the elder Senecafs authorship of it. True, it
can be argued of this, as of any other piece of historical analy-
sis or narrative, that it could have originated as an exemplum in
a philosophical work, but seeing that it is known that the elder
Seneca wrote historias ab initio bellorum civilium..., and we have
famous parallels in, for example, the first book of Thucydides,
and the beginning of Tacitus1 Annals for the inclusion of a sum-
mary of the history of a state from its origins in works otherwise
devoted to recent history, it seems unreasonable to argue that the
fragment is more likely to be by Seneca the philosopher than his
father. Again, it is not at all self-evident what moral point
could have been illustrated by the second of the Senecan histori-
cal fragments printed by Muller, a graphic account of the death of
Tiberius:
Seneca eum scribit intellecta defectione exemptum anulum quasi
alicui traditurum parumper tenuisse, dein rursus aptasse digito
et compressa sinistra manu iacuisse diu immobilem subitoque
vocatis ministris ac nemine respondente consurrexisse nee procul a
lectulo deficientibus viribus concidisse. (Suet. Tib. 73.2)
The historical sentiments in the fragment preserved by Lactantius
accord well enough with what we know about the elder Seneca's out-
look from his extant work;56 note that, even if there is no place
in its bleak scheme for kind words about Augustus, the author's
view is that there has been a deterioration since Rome's prima
senectus. This fragment could well have formed the preface of his
lost work. We simply do not have enough information on which to
base a denial that the elder Seneca's Histories were ever pub-
lished. Remember that it would only need one scholar to read
through a manuscript, excerpting as he went along, for fragments
of it to enter the literary tradition.
To be on the safe side, however, we must restrict our atten-

tion to the declamatory anthology when looking for evidence of the
elder Seneca's character. In this work the prevailing tone is one
of cheerful bonhomie and wit, as Seneca reminisces about people
who had been dear friends to him, and others who at least had im-
pressed or amused him, in years gone by. The work was a pleasure
to him, as he declares in the first preface:
est, fateor, iucundum mini redire in antiqua studia melioresque ad
annos respicere et vobis querentibus, quod tantae opinionis viros
audire non potueritis, detrahere temporum iniuriam. (Contr. I pr.
1)
It was particularly pleasant to have a chance to recall his
friendship with Latro:
in aliis autem an beneficium vobis daturus sim nescio, in uno ac-
cipio: Latronis enim Porcii, carissimi mihi sodalis, memoriam
saepius cogar retractare et a prima pueritia usque ad ultimum eius
diem perductam familiarem amicitiam cum voluptate maxima repetam.
(Contr. I pr. 13)
Having been on extremely close terms with Latro and other leading
figures responsible for the promotion of public declamation, he
was not at all disposed to adopt a censorious attitude towards
rhetoricians and their activities, though he knew perfectly well
that there were some people who took the view that it was mis-
guided to regard declamation as more than useful private practice
for true public oratory, and others who even doubted the educative
value of controversiae. Indeed, he quotes the arguments of two
such critics at length (Contr. Ill pr. 8ff.; IX pr. Iff). But he
was himself convinced of the usefulness of a rhetorical education
even for people who had no intention of becoming public speakers
of any kind in later life: facilis ah hac in omnes artes discursus
est; instruit etiam quos non sihi exercet (Contr. II pr. 3 ) . He
voices no objection to the practice of public declamation - to
have done so would have been disloyal to the memory of Latro.
That he was far from outraged by the scholastic outlook is evident
when he expresses a preference for a bona fide scholasticus over a
declaimer who aped the manner of forensic speakers (Contr. X pr.
12). The tolerance which is characteristic of his criticisms of
the declaimers is summed up by some words in the tenth preface:
nee sum ex iudicibus severissimis, qui omnia ad exactam regulam
R7*
derigant:0 multa donanda ingeniis puto; sed donanda vitia, non
portenta sunt (Contr. X pr. 10). Some of the more stupid de-
claimers1 efforts call forth caustic rebukes. For instance, just
before his declaration of tolerance in Contr. X pr. 10 he has been
saying of Musa, non ergo, etiamsi iam manu missus erat, debuit de
corio eius nobis satis fieri? But on other occasions he suffers
fools remarkably gladly, making such whimsical comments as non
minus stulte Aemilianus quidam Graecus rhetor, quod genus stult-
orum amabilissimum est, ex arido fatuus, dixit... (Contr. X.5.25)
and nihil est autem amabilius quam diligens stultitia (Contr,
VII.5.11).
Very occasionally, though, a powerful outburst of pessimism
breaks through for a while, and disrupts the easy-going geniality
which is the usual mark of his literary criticism. On the two
main occasions when this happens the style becomes far more agi-
tated and much nearer to the manner of the rhetoricians than
usual. A long section of the first preface is given over to a
heated discourse on the decline of eloquence since Cicero's time.
The following is a typical sample:
torpent ecce ingenia desidiosae iuventutis nee in unius honestae
rei labore vigilatur: somnus languorque ac somno et languore
turpior malarum rerum industria invasit animos, cantandi saltandi-
que obscena studia effeminatos tenent; [et] capillum frangere et
ad muliebres blanditias extenuare vocem, mollitia corporis certare
cum feminis et inmundissimis se excolere munditiis nostrorum
adulescentium specimen est. quis aequalium vestrorum quid dicam
satis ingeniosus, satis studiosus, immo quis satis vir est?
(Contr. I pr. 8-9)
The only other passage in the work at all comparable with this
comes in the tenth and last preface (Contr. X pr. 6-7), a tirade
perhaps deliberately placed there as a counterbalance to the one
on the decline of eloquence at the beginning of the work. Several
other motifs which first appear in the first preface are certainly
taken up in the tenth, with the evident intention of unifying the
work. 58 This time the subject of Seneca's anger is book-burning:
bono hercules publico ista in poenas ingenio<rum ver>sa crudelitas
post Ciceronem inventa est; quid enim futurum fuit, si triumviris
libuisset et ingenium Ciceronis proscribere? sunt di immortales
lenti quidem, sed certi vindices generis humani et magna exempla
in caput invenientium regerunt, ac iustissima patiendi vice quod
quisque alieno excogitavit supplicio saepe expiat suo. quae vos,
dementissimi homines, tanta vecordia agitat? (Contr. X pr. 6)

Modern writers on the elder Seneca have had a tendency, when
describing his character, to stress the sterner side of it,5^ his
'old fashioned austerity',6^ to the exclusion of all else. There
is no denying that he had such a side. Admittedly, it needs to be
taken into consideration that the first of his major pessimistic
tirades is on a theme, known as convicium saeculi (Contr. II pr.
2) or insectatio temporum (Suas. 6.9), which was a well-known
rhetorical commonplace, and that attacks on book-burning were
likewise part of the declaimers1 repertoire, required notably when
one was declaiming Suasoria 7: deliberat Cicero, an scripta sua
conburat, promittente Antonio incolumitatem si fecisset. Seneca
will therefore not have needed to think as hard as usual when
composing these passages, and his memories of other people's
treatments of the themes may have given him more than his usual
fluency. But that does not mean that he was insincere when he
wrote them.
However, bitter reflections on Roman decadence and the loss
of free speech do not bulk very large in Seneca's books on decla-
mation, and it is a mistake to blind ourselves to the evidence
provided by these books that he was a man of considerable toler-
ance and geniality, and, at least in his literary tastes, a mod-
ernist. We must be wary of sweeping statements by modern writers
to the effect that he was the personification of all the ideals
which made Rome great, a latter-day Cato, weighed down with
gravitas and stiff with antiquus rigor, statements such as are the
key-note of the somewhat rhapsodic accounts of his character given
by "Rornecque and Edward: 'il est profondement penetre de 1'esprit
de l'ancienne Rome: aussi bien son ide~al est-il le type parfait du
vieux Romain, Caton l'Ancien1;61 'Seneca's character has undoubt-
edly the gravitas, dignitas and constantia of the old Roman.'62
In fact, especially in his literary preferences, he appears
deeply affected by the spirit of the new era in which he lived.
If he had been a straightforward strictus Cato he would hardly
have had much patience with a man so devoted to art for art's sake
as Porcius Latro, or made remarks so attuned to the Ovidian age 63
as his facetious comments on dry stylists in Contr. II.1.24:
aridi declamatores fidelius quos proposuerunt colores tuentur:

nihil enim illos sollicitat, nullum schema, nulla sententia. sic
quae malam faciem habent saepius pudicae sunt: non animus illis
deest sed corruptor.
We have in the elder Seneca the curious phenomenon of a laudator
temporis acti se puero whose youth had been spent in a milieu far
from austere.
The idea that the elder Seneca was a man of antique severity,
devoted to old Roman traditions and contemptuous of new-fangled
philosophy, derives from no less an authority than his second son,
but there is some reason to doubt whether the younger Seneca's
references to his father can be trusted to provide more than a
conventional portrait of what an aged paterfamilias, born back in
the good old days of the Republic, ought to have been like. It
seems to have been the rule, rather than the exception, among the
Annaei Senecae, to present stylized portraits of any members of
their family mentioned in their writings. The elder Seneca leads
the way by representing his sons throughout his work on the
declaimers (with the notable exception of Contr. II pr. 3-4, where
he refers to their choice of careers) as thoroughly immature young
men, interested only in scholastic sententiae, unwilling to apply
their minds to the serious study of history (Suas. 6.16, 27) or
even to the drier aspects of declamation (Contr. I pr. 22). He
never mentions that his second son was interested in philosophy.
Again, the younger Seneca, when writing about his mother, uses
language suspiciously reminiscent of the standard declamatory de-
scriptions of female types. Compare, for example, Cons. Helv.
16.4, non faciem coloribus ac lenociniis polluisti: numquam tibi
placuit vestis, quae nihil amplius nudaret, cum poneretur, with
some words from a declamation by Latro (Contr. II.7.4), prodite
mihi fronte in omne lenocinium composita, paulo obscurius quam
posita veste nudae.^ He also represents his father as contemptu-
ous of philosophy (Ep. mor. 108.22: philosophiam oderat). Here he
is totally misleading: there is inescapable evidence that the
elder Seneca shared his son's admiration for the philosopher
Fabianus and the sanctis fortibusque praeceptis to which he dedi-
cated himself, and for the Stoic Attalus, for his subtilitas as
well as for his eloquence (Suas. 2.12). 65 On another occasion the
younger Seneca asserts that his father had not wished Helvia to
immerse herself too deeply in philosophical learning:
itaque illo te duco, quo omnibus, qui fortunam fugiunt, confugi-
endum est, ad liberalia studia: ilia sanabunt vulnus tuum, ilia
omnem tristitiam tibi evellent. his etiam si numquam adsuesses,
nunc utendum erat; sed quantum tibi patris mei antiquus rigor per-
misit, omnes bonas artes non quidem comprendisti, attigisti tamen.
utinam quidem virorum optimus, pater meus, minus maiorum consuetu-
dini deditus voluisset te praeceptis sapientiae erudiri potius quam
imbui! non parandum tibi nunc esset auxilium contra fortunam sed
proferendum. propter istas quae litteris non ad sapientiam
utuntur sed ad luxuriam instruuntur, minus te indulgere studiis
passus est. beneficio tamen rapacis ingenii plus quam pro tempore
hausisti. (Cons. Helv. 17.3f.) 66
These remarks are best taken as telling us more about the elder
Seneca1s attitude towards women than towards philosophy, that is,
supposing we do not take this passage as just one more case of
Senecan persiflage, intended to excuse a son for preaching to his
mother, rather than as certain evidence that Helvia*s desire for
philosophic enlightenment was discouraged by her husband. Cer-
tainly the younger Seneca felt no more obliged to tell the whole
truth when referring to his fatherfs character in his philosophi-
cal works than the elder Seneca had done when referring to the
tastes of his sons in his declamatory anthology. Seneca the
Younger never mentions his fatherfs interest in scholastic rhet-
oric, which would not have been consistent with the picture he
wished to present of his father as a man maiorum consuetudini
deditus, distinguished for his antiquus rigor (Cons. Helv. 17.
3-4).
Modern scholars have added two main embellishments to the
younger Seneca1s picture of the old man, both of which deserve re-
examination. First, we are to believe that his devotion to the
consuetudo maiorum included an attachment to traditional re-
ligion.67 For all we know this may have been the case, but surely
the one sentence adduced as evidence (Contr. X pr. 6): sunt di
immortales lenti quidem, sed certi vindices generis humani etc.,
coming as it does from one of his untypically declamatory passages
(the tirade against book-burning), and expressing, albeit elo-
quently, a fairly commonplace sentiment, is not necessarily suf-
ficient basis for an assumption that he paid more than lip-service
to the immortal gods.

The other main characteristic modern writers are fond of
attributing to him is intense patriotism. According to W. A.
Edward, 'He is a provincial; but for Rome and things Roman he has
a burning enthusiasm that surpasses that of the true sons of the
eternal city. 168 However, Mrs Griffin in a recent article has
taken a new look at the evidence, and, finding the elder Seneca
less thoroughly Romanized than his philosopher son, has laid more
emphasis than is usual on signs of his attachment to his native
province.6^ One assertion commonly made about him that needs to
be questioned more than it has been, is that he was prejudiced
against the Greeks.7^ Now, it is true that he once pits the
achievements of Roman eloquence in the Ciceronian age against
'insolent' Greece: quidquid Romana facundia habet, quod insolenti
Graeciae aut opponat aut praeferat, circa Ciceronem effloruit
{Contr. I pr. 6 ) . But before we lay too much weight on this
remark, we might note that it had a close parallel in a sententia
from a declamation by Cestius Pius, a Greek declaimer notorious
for his contempt for Cicero:71 iniuriam ilium facturum populo
Romano, cuius linguam <in locum pr>incipem extulisset, ut
insolentis Graeciae studia tanto antecederet eloquentia, quanto
fortuna (Suas. 7.10 M ) . 7 2 It looks as if insolens Graecia was a
standard phrase, a famous quotation, maybe, whose source is no
longer known, which one might use with no particular depth of
feeling. Another of the elder Seneca's expressions of patriotism,
taken seriously by Edward, though not so much so by Bornecque,73
sed nolo Romanos in ulla re vinci (Contr. X.5.28), would be less
ambivalent if the contest in which he is pitting Romans against
Greeks at this point were not a competition for the most insane
sententia. It is necessary to notice these facts about these
supposed key passages, because detailed study of Seneca's criti-
cisms of individual Greek sententiae does not reveal undue harsh-
ness of judgement, as A. F. Sochatoff74 was the first to point
out. J. Buschmann, who wrote two papers about Seneca the Elder,
one dealing with the Greek declaimers mentioned by him and another
with what he calls his 'enfants terribles', ought to have made the
point that these two classes of declaimers overlap very little.75
When Seneca criticizes samples of bad taste by the Greek declaim-

ers he is no harder on them, as a rule, than on their Roman
counterparts.76 Such a departure from mildness as the use of the
superlatives corruptissime and furiosissime against two Greek
declaimers in Contr. X.5.21 is very exceptional. Elsewhere we
find him giving as high praise as he ever awards to a declamatory
sententia to one by a Greek: Diocles Carystius dixit sententiam,
quae non in declamatione tantum posset placere sed etiam in
solidiore aliquo scripti genere... (Contr. I.8.16). 77 Nor does
such cautious weighing up of merits as we find, for example, in
the same section, when he considers a sententia by Dorion, betoken
blind prejudice: Dorion dixit rem paulo quidem elatiorem, quam
pressa et civil is oratio recipitf sed qua egregie attonitos patris
adfectus exprimeret... In the several passages where he considers
the relative merits of Greek and Latin sententiae on the same
theme,7^ Roman declaimers are just as likely to win the prize for
corruptum or obscenity as for excellence. In the international
competition of Contr. 1.4.10-12 which opens with the words, ex
altera parte multa sunt pulcherrime dicta; sed nescio an Graecis
nostri cessuri sint, it is Albucius who finally emerges as victor,
but the opposition, in Seneca's opinion, was strong:
Vibius Rufus dixit: adulter meus exit et commodo suo. Hybreas
hunc sensum optime dixit:... Dionysius, filius eius Dionysii, qui
Ciceronis filium docuit, elegans magis declamator quam vehemens,
hunc sensum et vehementer dixit et eleganter: ...Vibius Rufus
dixit: quam otiosi, quam securi adulteri transierunt praeter
oculos meos, praeter filii manus! Latro cum exeuntis adulteros
descripsisset adiecit: adulescens, parentes tuos sequere. Nicetes
illam sententiam pulcherrimam, qua nescio an nostros antecesserit:
... sed illud Albuci utique Graecos praeminet: cum pugnantem se
<in> acie descripsisset, dixit: me miserum quas manus adulter
effugit! et illud Albuci: 'non potuiT inquit 'matrem occidere.'
quo excusatior sis, adice: Tet patrem1. (Contr. I.4.11f. M)
And in Contr. 1.4.12 the contest ends with a reference to a most
stupid imitation of a Greek sententia by one of Seneca's Roman
'enfants terribles1:
dixerat Nicetes:... Murredius dum hanc sententiam imitari vult,
stultissimam dixit: reliqui in acie pugnantes manus.
In Contr. 1.2.22 Seneca takes pains to ensure that we do not gain
the impression that obscenity was a fault peculiar to Greek de-
claimers. After citing a sweeping statement by Scaurus on the
subject, hoc autem vitium aiebat Scaurus a Graecis declamatoribus

tractum, qui nihil non et permiserint sibi et inpetraverint, and
two Greek examples, he goes on to remark: in hac controversia de
sacerdote non minus obscene dixit Murredius: fortasse dum repellit
libidinem, manibus excepit (Contr. 1.2.23). A protracted contest
between Greeks and Romans in Contr, X.4.18-23 opens with the
words:
Celebris haec apud Graecos controversia est; multa ab illis
pulchre dicta sunt, a quibus non abstinuerunt nostri manus, multa
corrupte, quibus non cesserunt nee ipsi.
The last stages of this competition deserve close consideration,
as they lead up to one of Senecafs rare generalizations, which is
often quoted in isolation and in context is rather puzzling. In
§22-3 Seneca shows how two corrupt sententiae by Glycon met their
match in three equally bad Latin ones:
9
corruptam dixit sententiam: xpouaaxw T L S xnv £upav rav
X <tva> TtpoaayaYfl TUS. et illam: dye, ov 6e xAaue, au 6e
dpi^vet. a) xaxajv auucpwvLwv! sed nostri quoque bene insanierunt.
Murredius dixit: producitur miserorum longus ordo, maior pars se
sine se trahit. et Licinius Nepos: ut solvendo sis, in poenas
quotiens tibi renascendum est? illud Sparsus dixit, quod non
corruptum tantum sed contrarium dicebat esse Montanus: ' solus
plura habes membra quam tot hominibus reliquisti.1 ita enim hie
potest videri laesisse rem publicam, si multi sunt debilitati;
apparet autem non esse multos, si plura habet membra quam debili-
tatis reliquit. et illud aeque aiebat ab illo corrupte dictum:
f
prodierunt plures mendici quam membra.f
It is at this juncture that we read:
Graecas sententias in hoc refero, ut possitis aestimare, primum
quam facilis e Graeca eloquentia in Latinam transitus sit et quam
omne, quod bene dici potest, commune omnibus gentibus sit, deinde
ut ingenia ingeniis conferatis et cogitetis Latinam linguam
facultatis non minus habere, licentiae minus.
This conclusion seems in its latter part quite out of joint with
what has come immediately before. One might have expected Seneca
to say rather: et cogitetis Latinam linguam facultatis minus
habere, licentiae non minus.®^ Perhaps he did, and our text is
at fault.
To sum up, Seneca the Elder was very much a man of his time,
not the embodiment of all the old Roman virtues and prejudices
which he is sometimes made out to have been. While he occasion-
ally expressed pride in Roman achievements, patriotism did not
seem to him so serious a matter that he could not make jokes about
it, 81 and he did not let his judgement of individual Greek senten-
tiae be distorted by a general bias against Graeculi - a term
which, significantly, he uses only once, and then with good humour
and in a passage where he may well be quoting Cestius (Suas,
1.6) , 8 2 He shows no sign anywhere in his extant writings of the
disdain for philosophy attributed to him by his second son.83
With regard to literature he was a modernist. This is not only
apparent from his enthusiastic interest in the activities of the
rhetoricians and in the intricacies of the art taught in what a
really old-fashioned Roman would have called their schools of
impudence,8lf but also from the range of his references to the
higher forms of literature: with the single exception of his quo-
tation from Cato in Contr. I pr. 9 he makes no reference to any
pre-Ciceronian Roman orator;85 nor does he mention any Latin
historians or poets earlier than Sallust and the Neoterics.86
Where politics were concerned, he appears to have had a certain
amount of admiration for Augustus, though, if the Senecan fragment
preserved by Lactantius (fr. hist. 1) comes from his Histories, he
viewed the return of Rome to the rule of one man as in itself a
sign of an advanced stage of decline. Certainly in his opinion
that Roman rhetoric had degenerated seriously during his life-time
(Contr. I pr. 6ff.), in his awareness of the dangers which faced
men of his sons1 generation entering public life (Contr. II pr.
3), and in his fulminations against the suppression of literature
(Contr. X pr. 6f.), he reflects some of the disillusionment
typical of the era in which he was writing.
2 THE DECLAMATORY ANTHOLOGY
The elder Seneca mentions several reasons why he came to write

about declamations and declaimers. First, he will have us believe
that the theme of his work was suggested by his sons:
Seneca Novato, Senecae, Melae filiis salutem.
Exigitis rem magis iucundam mihi quam facilem: iubetis enim quid
de his declamatoribus sentiam, qui in aetatem meam inciderunt,
indicare et si qua memoriae meae nondum elapsa sunt ab illis dicta
colligere, ut, quamvis notitiae vestrae subducti sint, tamen non
credatis tantum de illis sed et iudicetis. (Contr. I pr. 1)
This request from his sons is perhaps fictional or semi-fictional.
To claim that one was writing at the request of some person was,
like the epistolary greeting, a standard convention among ancient
writers of prefaces to works whose utility needed to be empha-
sized.1 To make out that one*s purpose in writing was to give
information to some member of one's family had been a cliche in
classical didactic writing ever since Hesiod had addressed the
Works and days to his brother. To take two Roman instances, Cato
in, for example, the dictum quoted by Seneca in Contr. I pr. 9:
orator est, Marce fili, vir bonus dicendi peritus, and Cicero in
his Partitiones oratoriae, had addressed rhetorical precepts to
their sons, the latter in a baby-simple catechistic fashion which
his rebellious offspring can hardly have appreciated.2 Admittedly
Cicerofs son was only thirteen when it was written. On the other
hand Novatus, Seneca and Mela were adults by the time they are
supposed to have made their request (see Contr. II pr. 4 ) , and
their father had no reason besides literary convention to talk
down to them in the way he sometimes does (e.g. Contr. I pr. 22;
Suas. 6.16), as if they were foolish youths, uninterested in the
intricacies of argument or the serious study of history, unin-
terested in anything, in fact, except sententiae. The way he
refers to his sons1 request as a command (exigitis...iubetis...)
is in line with old-established conventions of preface-writing.
Precedents are to be found in Hellenistic technical writings:

Janson^ cites Archimedes, De sphaera et cylindro II pr. init.,
eTeaxeuAds you ypacâu T&V TtpoBAnyonrwv fas a7io6eCêLS, as a typi-
cal example. Among Roman writers the 'command1 was absolutely
conventional, as one of the younger Pliny's anecdotes, about an
elegist called Passenus Paulus, shows:4
is cum recitaret, ita coepit dicere: 'Prisce, iubes...' ad hoc
Iavolenus Priscus (aderat enim ut Paulo amicissimus): 'ego vero
non iubeo.' cogita qui risus hominum, qui ioci. est omnino
Priscus dubiae sanitatis. (Ep. VI.15.Iff.)
Seneca represents his work on the declaimers as a collection
of rhetorical specimens for his sons to examine with a view to
imitation. Anticipating Quintilian in this,5 he argues that it is
not a good idea to take a single writer, however excellent, as
one's only model:
Facitis autem, iuvenes mei, rem necessariam et utilem, quod non
contenti exemplis saeculi vestri prioris quoque vultis cognoscere:
primum quia, quo plura exempla inspecta sunt, plus in eloquentiam
proficitur. non est unus, quamvis praecipuus sit, imitandus, quia
numquam par fit imitator auctori. haec rei natura est: semper
citra veritatem est similitudo. (Contr. I pr. 6)
He does not, however, exclude examples of bad taste from his col-
lection, arguing that there was something to be learnt from exam-
ining even these:
omnia autem genera corruptarum quoque sententiarum de industria
pono, quia facilius et quid imitandum et quid vitandum sit docemur
exemplo. (Contr. IX.2.27)
As it turned out, at least one of his sons did learn something
from his collection. Several studies have revealed close paral-
lels between turns of phrase in the younger Seneca's works and
sententiae recorded in his father's anthology.6 It is as well to
remember that the younger Seneca was not dependent solely on his
father's work for his knowledge of declamation, so that one can
never be sure that a declamatory-sounding phrase in his works is
not based on something remembered from his school-days, rather
than from his father's books. That he did take a considerable
interest in his father's work on declamation is proved, however,
by some obvious imitations of his father's literary criticism he
makes in the description of Fabianus' style in Ep. mor. 100.
Although he dedicates the anthology to his sons, the elder
The declamatory anthology 29
Seneca makes it clear early on that he has reasons for writing it

other than their alleged demands for sententiae. It seems that he
was depressed by the lack of any comprehensive or accurate written
record of the work of the declaimers famous in his youth. The
situation had arisen where new declaimers could plagiarize the
sententiae of their predecessors with impunity, as few people
could remember who had originally composed them. This was a situ-
ation he aimed to rectify by the publication of his work:
sententias a disertissimis viris iactas facile in tanta hominum
desidia pro suis dicunt et sic sacerrimam eloquentiam, quam prae-
stare non possunt violare non desinunt. eo libentius quod exig-
itis faciam et quaecumque a celeberrimis viris facunde dicta
teneo, ne ad quemquam privatim pertineant, populo dedicabo. ipsis
quoque multum praestaturus videor, quibus oblivio inminet, nisi
aliquid, quo memoria eorum producatur, posteris tradetur. fere
enim aut nulli commentarii maximorum declamatorum extant aut, quod
peius est, falsi. itaque ne aut ignoti sint aut aliter quam debent
noti, summa cum fide suum cuique reddam. (Contr. I pr. 10-11 W)
It is evident that all along he had a wider readership than his
immediate family in view.
The Oratorum et rhetorum sententiae divisiones colores is a
work of a curious design, unparalleled in extant ancient litera-
ture. From certain things Seneca says, it appears that its form
was suggested by the arrangement of a gladiatorial show, and it is
probably best to take him at his word and look no further for
possible sources of its inspiration.7 The task Seneca had set
himself was two-fold: on the one hand, he wanted to give general
descriptions of the style and personal traits of the leading de-
claimers of his acquaintance, and on the other, he wanted to pro-
vide a detailed record of the treatment by as many declaimers as
possible of particular controversiae and suasoriae. How to com-
bine these two aspects of his work in a unified scheme was a prob-
lem. The solution he adopted was this: he would introduce the
leading declaimers by means of pen-portraits, not all together at
the beginning of the work, but in a series of prefaces to separate
books, which would provide at intervals welcome diversions from
the series of detailed surveys of declamations which would make up
the bulk of the work. The first preface would be rather more di-
verse in content, and longer, than the others, and would serve as
an introduction to the whole, like the opening procession at the
Circus, to which he whimsically alludes at the end of the preface:

sed iam non sustineo diutius vos morari: scio, quam odiosa res
mihi sit Circensibus pompa (Contr. I pr. 24). This first preface
would have as a prominent feature a portrait of his great friend
Latro; later ones would serve to introduce one declaimer or sev-
eral. He would keep his readers waiting right to the end for the
formal introduction of some of the celebrities quoted in his sur-
veys. In doing this he was adopting a technique for holding an
audience's attention familiar to anyone in show-business:
quod munerarii solent facere, qui ad expectationem populi detin-
endam nova paria per omnes dies dispensant, ut sit, quod populum
et delectet et revocet, hoc ego facio: non semel omnes produco;
aliquid novi semper habeat libellus, ut non tantum sententiarum
vos sed etiam auctorum novitate sollicitet. acrior est cupiditas
ignota cognoscendi quam nota repetendi. hoc in histrionibus, in
gladiatoribus, in oratoribus, de quibus modo aliquid fama pro-
misit, in omnibus denique rebus videmus accidere: ad nova homines
concurrunt, ad nota non veniunt. (Contr. IV pr. 1)
After each of the ten prefaces came surveys illustrating the
way in which about eight controversia themes were treated by vari-
ous declaimers, some of whom would have been formally introduced
in previous prefaces, others not. Particular prominence tends to
be given in these surveys to any declaimer introduced in the pre-
face immediately before: thus Latro, introduced in the first pre-
face, dominates book I, but is less frequently quoted as the work
progresses; Fabianus is repeatedly quoted in book II. At the end
of the preface Seneca usually makes an effort to point forward to
some feature in the surveys to follow: thus he tells us at the end
of the first preface that the first controversia in the collection
will be one which he remembers having heard Latro declaim as a
schoolboy (Contr, I pr. 24), and at the end of the second preface,
in which he describes Fabianus, he makes the promise, fin hunc
ergo libellum quaecumque ah illo dicta teneo conferam1 (Contr, II
pr. 5 ) . 8
The surveys of controversiae all have the same basic plan,
the threefold structure of which, sententiae, divisiones9 colores,
is alluded to in the title of the work. After quoting in full the
theme of the controversia and the laws under which it was to be
tried, Seneca first assembles, normally without comment, samples,
very diverse in kind, extracted from a number of declaimers*

treatments of the theme. Presumably these samples comprise the
sententiae mentioned first in the title, as divisiones and colores
are dealt with in the second and third sections of the surveys
respectively. It cannot be, however, that any of these substan-
tial extracts taken as a whole could have been called a sententia,
for this word is not used in a technical sense by Latin writers on
rhetoric to mean anything much longer than a single sentence,10
except when they are referring to the delivery of verdicts - which
is not the function of by any means all the extracts in the first
section of the Senecan surveys11 - or to a type of progymnasma
akin to the chria (Quint. 1.9.3). Rather, Seneca^ extracts give
us series of sententiae, and very little but sententiae, if we
give that name to all the types of sentence so called by Quinti-
lian in VIII.5.Iff.12 together with the descriptive sentences
which, it would appear from Contr. IX.2.24, Seneca was prepared to
describe loosely as sententiae as well.
The logical sequence between the sentences in the samples is
sometimes extremely tortuous, and the question arises whether
Seneca is presenting continuous prose exactly as first delivered
by the declaimer, or whether he is stringing together particularly
notable sentences which did not follow one another so closely in
the original declamation. No certain answer can be given in every
case, but careful and detailed consideration has been given to the
matter by both Bornecque and Winterbottom, who in their editions
place dashes (Bornecque in both text and translation, Winterbottom
in the translation only) wherever there is discontinuity of
thought between adjacent sententiae. It does not seem as if
SenecaTs usual practice was to string together single sententiae
which had borne no relation to one another in the declamation as
originally delivered. Usually at least two or three sentences in
succession do seem to be linked closely together, and perhaps some
of the jerky transitions marked by Bornecque and Winterbottom with
dashes ought to be attributed to the declaimers rather than to
Senecan excerpting. Occasionally, for example in the long extract
from Latro s declamation of Contr. H . 7 , 1 ^ it seems clear that
f
Seneca was quoting continuous prose at considerable length.11*

The sententiae recorded in the opening section of Seneca's

surveys are of many different kinds. They were taken, one sus-
pects, from all the different parts of a declamation, some from
the prooemium, some from the narratio, and so on. Just occasion-
ally it is indicated that he is quoting from the narrative of the
case, thus: CESTI PII narratio (Contr. 1.2.7), but usually only
the name of the rhetorician is given (in the genitive). Normally
most of the extracts treat the thema from whichever point of view,
whether that of the prosecution or the defence, was most favoured
by the declaimers, but often, though not invariably, the section
devoted to sententiae ends with a few extracts from treatments of
the less popular side of the case under the heading, pars altera.
After the sententiae comes a section devoted to divisio.
Here we are given analyses of the scheme of argument used by a few
of the leading rhetores, notably Latro, with whose perceptiveness
in the detecting of basic issues Seneca wished to impress us
(Contr. I pr. 20f.). He outlines the main questions at issue, and
the subsidiary ones arising from them with admirable clarity, but
very concisely, using a very few standard terms and phrases to
guide the reader, too few for the liking of H. Bardon,15 a cen-
sorious critic of the elder Seneca's style, vocabulary and criti-
cal techniques. Seneca makes it clear from the outset that the
scope of the sections on divisio in his surveys is to be strictly
limited: he will state the main questions raised by each de-
claimer, but it will be no part of his plan to give any detailed
account of the way in which separate lines of argument were devel-
oped, 'nee his argumenta subtexam, ne et modum excedam et pro-
position1 , since, so he alleges, his sons would not be interested
(Contr, I pr. 22). He nevertheless does sometimes quote a few
sentences of argumentation to show how some of the basic questions
were treated.
It is in the third and final section of the surveys that we
find the only extensive passages of literary criticism outside the
prefaces. In the surveys of controversiae this section begins
with consideration of the colores used by each declaimer. Color
was a technical term meaning the complexion which one gave to the
past actions of the litigants in a controversial^ For, as long
as he did not actually contradict the given facts of the case, the
declaimer was free to give whatever damning, or mitigating, inter-
pretation he chose, to actions described in the thema. Seneca
does not, however, restrict himself to the discussion of colores
in the closing section of his surveys. He usually digresses into
other, often more interesting, topics. There is much comparative
criticism. Seneca shows a great interest in the question who imi-
tated whose sententia, and whether the imitation was an improve-
ment on, or a frigid over-working of, the original idea. Usually
he is concerned with imitations by one rhetorician of another, but
sometimes he also enlightens us about the debt owed by certain
declaimers to the poets, and by certain poets to the rhetoricians
(e.g. Contr. II.2.8; Suas. 3.5). Occasionally he turns aside from
the subject of declamation altogether and recalls discussions of
modern poetry in which some of the leading critics and patrons of
Augustan Rome took part (e.g. Contr. VII.1.27; Suas. 1.12), or
momentarily lets his interest in historiography come to the fore,
as in Suas. 6.14-25, where he compares recent historians' accounts
of the death of Cicero. Sometimes in this section too, Seneca
feels prompted, on mentioning some declaimerfs name, to sketch his
character or to tell some illuminating anecdote about him.
Usually at the very end of the section, but sometimes earlier
in it too (e.g. Contr. IX.1.12ff.; X.5.19ff.), we find a few sen-
tentiae by Greek declaimers quoted, sometimes set out for compari-
son with Latin sententiae on similar themes. The comparison of
Greek with Latin sententiae may follow fairly naturally from what
has come before, but quite often the Greek examples seem almost to
be thrown in as an afterthought. (The nearest parallel for this
rather strange arrangement comes in Valerius Maximus' presentation
of history for rhetoricians, the Facta et dicta memorabilia, where
in each section a few foreign exempla are added after the more
numerous Roman ones.) The medieval scribes made valiant attempts
to cope with the unfamiliar Greek script where Seneca required it,
but sometimes gave up the unequal struggle and left the names of
Greek declaimers without their sententiae. Seneca, then, allotted
more space to Greek declaimers than is now apparent. But it is
improbable that he originally devoted as much attention to them as
to their Latin counterparts. The survey of Contr. X.5, which

seems to be complete, and is unusually rich in Greek sententiae,
contains no Greek quotation more than thirteen words long. We
ought perhaps to question whether Seneca relegated as many Greek
sententiae right to the end of the surveys as we find there now in
our texts. The surveys often end with what seems like a rag-bag
of sententiae, sometimes probably consisting of displaced addenda,
miscellaneous sentences which some scribe, correcting his manu-
script, noted as omitted earlier in the survey. However, it is
usually rather difficult to see where those Greek sententiae left
till last could have been placed earlier, and in Contr. 1.5.9, if
the text is to make good sense, a transposition is required which
actually restores some Greek sententiae to the end of the sur-
17*
vey.L'
As well as dealing with controversiae in ten books, Seneca
compiled surveys of suasoriae. Seven of these are extant, and
more, it seems, were part of Seneca's plan, for in Contr. II.4.8
he promises an account of a suasoria not among those we have. It
is uncertain, though, whether any more surveys of suasoriae than
are now extant were actually written or published. The colophon
which follows the suasoriae in some manuscripts, lannei senecae
oratorum et reto(retho V)rum sententiae divisiones colores suas-
oriarum liber primus (I B) explicit incipit liber secundus (II
<feliciter> B), to give the version of BV, cannot be regarded as
evidence for a second book of suasoriae.*® The order and number-
ing of the books in the manuscripts which preserve the fuller ver-
sion of the text of Seneca's collection, are as follows:
Suasoriae = liber I suasoriarum
Contr. I = liber II
Contr. II = liber III
Contr. VII= liber IV
Contr. IX = liber V
Contr. X = liber VI.
The liber secundus promised in the colophon to the suasoriae in
these manuscripts is none other, then, than the first book of
controversiae, and suasoriarum is to be taken as a genitive of
definition, not as a partitive genitive.
One question which seems never to be considered seriously,
but ought to be, though it is probably unanswerable, is where in

the plan of the whole work Seneca intended the surveys of suas-
oriae to fit. The manuscripts which give them, place them, as we
have seen, before the controversiae; modern editors print them
after Contr. X. Neither position seems very satisfactory. The
problem is that the preface to Contr. I is unquestionably intended
to act as an introduction to the whole collection, and the preface
of book X seems just as obviously intended to draw the work to-
wards its close. In the first preface Seneca has expressed de-
light at the opportunity which his sons' request has given him, to
relive happier years, redire in antiqua studia melioresque ad
annos respicere (Contr. I pr. 1 ) , and has accepted the task with
the words: fiat quod vultis: mittatur senex in scholas (Contr. I
pr. 4 ) . In the tenth preface he harks back to his initial en-
thusiasm, but now professes to be weary of his task:
Quod ultra mihi molesti sitis, non est: interrogate, si qua vul-
tis, et sinite me ab istis iuvenilibus studiis ad senectutem meam
reverti. fatebor vobis, iam res taedio est. primo libenter ad-
silui velut optimam vitae meae partem mihi reducturus: deinde iam
me pudet, tamquam diu non seriam rem agam. hoc habent scholastic-
orum studia: leviter tacta delectant, contrectata et propius ad-
mota fastidio sunt. sinite ergo me semel exhaurire memoriam meam
et dimittite vel adactum iureiurando, quo adfirmem dixisse me quae
scivi quaeque audivi quaeque ad hanc rem pertinere iudicavi.
(Contr. X pr. 1)
Though one may be sure that this passage is packed with literary
convention, and it seems quite possible that Seneca, far from
being weary of declamation, went indefatigably on, after writing
book X of the controversiae, to compile his surveys of suasoriae91^
it seems inconceivable that book X, prefaced as it was by all
these protestations of weariness and reminiscences of Contr. I pr.
Iff., was not intended to be the final book of the collection. On
the other hand, the character of the first preface, and Seneca's
promise in Contr. II.4.8 of a survey of the suasoria about Theod-
otus 'cum ad suasorias venero1, make it clear that the manuscripts
are wrong in placing the liber suasoriarum before the controver-
siae.
Some radically different speculation is required, and one
line to consider is this. Perhaps Seneca's original plan was that
each preface should serve as an introduction to surveys of both
controversiae and suasoriae. (The controversiae would have to

come first, for, as we have seen, Seneca leads into them at the
end of the prefaces.) However, his editors failed to recognize
his intention, perhaps because he had left writing the surveys of
suasoriae till last, and had died without completing them or mak-
ing it clear where in the scheme of the whole work they were to
go. I suggest this because our extant 'liber suasoriarum1 looks
suspiciously like two fragments wrongly linked together.
A notable feature of Suas. 2, 3 and 4 is that they contain a
series of curiously laboured reminiscences of Seneca*s description
of Arellius Fuscus in the second preface (Suas. 2.10, 23; 3.7; 4.5
cf. Contr. II pr. 1 ) . Suas. 1 also contains the only extant quo-
tation in the collection outside the second book of Controversiae
from a declamation by Fabianus, of whom Seneca has said at the end
of the second preface, in hunc ergo libellum quaecumque ah illo
dicta teneo conferam (Contr. II pr. 5 ) . Can it be, then, that the
first five suasoriae - five, because the reference to Fuscus in
Suas. 4.5 points forward to the fifth suasoria - were originally
intended to follow the controversiae of book II and share the same
preface? The main objections to such a theory are that the suas-
oria on Theodotus, promised in Contr. II.4.8, is not to be found,
and that the surveys we have contain no quotations from suasoriae
declaimed by Ovid in his youth, which one might have expected in
view of Seneca's words in Contr. II.2.12: declamabat autem Naso
raro controversias et non nisi ethicas; libentius dicebat suas-
orias. The only plausible ways to counter both these objections
are to surmise either that some other suasoria-surveys preceded
our Suas. 1 in the hypothetical continuation of book II, or that
Seneca turned to the suasoriae he intended to add to book II (our
Suas. 1-5) considerably later than the controversiae in it; that
he only re-read the preface of the book before setting to work on
the suasoriae, and had forgotten the promise given in Contr.
II.4.8 and the expectations which might be aroused by his remarks
in Contr. II.2.12.
Suasoriae 6 and 7 stand rather apart from the rest: Fuscus
does not feature nearly so prominently in them. These two long
surveys would follow happily after the controversiae of book IV:
Haterius, introduced in the fourth preface, is given the prominent

first position in both surveys; Pollio, who is contrasted with
Haterius in Contr. IV pr. 2ff. also figures conspicuously in the
historiographical digression of Suas. 6 (14f., 24f.). Both suas-
oriae concern the death of Cicero, and would follow easily after
the last controversia of book IV, which was also about a victim
of the proscriptions:
Bello civili patronus victus et proscriptus ad libertum confugit.
receptus est ab eo et rogatus, ut operas remitteret. remisit con-
signatione facta. restitutus indicit operas, contradicit.
(Contr. IV.8 thema)
The surveys of suasoriae follow much the same plan as those
of controversiae: sententiae are followed by divisiones. The only
difference is that, as colores are a feature peculiar to judicial
rhetoric and are therefore not found in suasoriae, the surveys end
with a section which is even more of a mixed bag than usual, and
includes unusually long and intriguing digressions.
One of the most vexed questions about the elder Seneca is how
far he relied on his memory when compiling his anthology. The
question arises because, in the section of the first preface
where, in accordance with a well-known convention of prefatory
writing, he expresses doubts about his ability to undertake the
projected work,2^ he discourses at length about his memory, once
miraculously tenacious, but now weakened by old age. The impli-
cation is that it is on this faculty that he is expecting to have
to rely:
memoria est res ex omnibus animi partibus maxime delicata et
fragilis, in quam primam senectus incurrit. hanc aliquando <adeo>
in me floruisse, ut non tantum ad usum sufficeret sed in miraculum
usque procederet, non nego... [Remarkable examples follow]
...nunc et aetate quassata et longa desidia, quae iuvenilem quoque
animum dissolvit, eo perducta est ut, etiamsi potest aliquid prae-
stare, non possit promittere. diu ab ilia nihil repetivi: nunc
quia iubetis quid possit experiar et illam omni cura scrutabor.
ex parte enim bene spero: nam quaecumque apud illam aut puer aut
iuvenis deposui, quasi recentia aut modo audita sine cunctatione
profert; at si qua illi intra proximos annos commisi, sic perdidit
et amisit ut, etiamsi saepius ingerantur, totiens tamen tamquam
nova audiam... controversiarum sententias fortasse pluribus locis
ponam in una declamatione dictas, non enim <semper> dum quaero
aliquid invenio, sed saepe quod quaerenti non comparuit aliud
agenti praesto est... (Contr. I pr. 2-5 W)
Many scholars have been quite willing to countenance the idea that
Seneca1s only source, or at least his main source, was his memory,
and they have a number of good reasons for doing so. The examples
Seneca gives of the miraculous memory of his youth - his ability
to recall two thousand separate names in the correct sequence, or
over two hundred isolated lines of poetry in reverse order - are
impressive, and we must not be over-sceptical about his claims, for
a glance at any encyclopaedia1s entry on memory or mnemonics will
convince one that the human memory is capable of almost anything,
especially when trained in a discipline in which memorizing is
regarded as a necessity or a virtue. And doubtless many a
specialist in geriatrics would attest that Senecafs account of his
ability to recall things heard in youth better than those of more
recent years, is psychologically convincing. Boissier in his
f 1
article, Les ecoles de declamation a Rome , and Bonner in Roman
declamation, show themselves firm believers in Seneca's memory,2-1
and Winterbottom has recently reaffirmed their view, though he
refers us in a footnote to Bornecque for evidence that Seneca
'certainly had some written sources1.22
Bornecque expresses scepticism about Seneca's memory in no
uncertain terms: 'enfin sa memoire, si extraordinaire fut-elle, ne
pouvait suffire a. un pareil effort1 ,2^ and proceeds to list a num-
ber of written sources which Seneca mentions and could have used.
Other doubters have included G. L. Hendrickson24 and 0. Immisch,25
who seem to have arrived independently at the view that it is
naive to regard Seneca's picture of himself racking his brains for
memories of declamations long ago as anything but a preface-
writer's convention taken over from the tradition followed by
writers of dialogues. Their view has been elaborated in a recent
American thesis, C. W. Lockyer's 'The fiction of memory and the
use of written sources: convention and practice in Seneca the
Elder and other authors',2^ a careful piece of work in which,
however, the evidence as to Seneca's possible use of written
sources is pressed much further than it will go.
Here is a dispute that will never be settled conclusively.
We have seen plenty of evidence that Seneca was fully conscious of
prefatory conventions. In the vexed passage about his memory he
is adopting at least one such. To express misgivings about one's
capacity to undertake the task at hand had been standard practice

among the Attic orators, and had become enshrined in the Greco-
Roman theory about prooemia: benevolentia...comparatur...si quae
incommoda acciderint aut quae instent difficultates, proferemus...,
as Cicero puts it (De inv. 1.16.22). So it is not unreasonable to
be somewhat sceptical about the extent of Seneca's powers of recol-
lection, when one considers that far-fetched claims about reliance
on memory were commonplace among writers of dialogues and related
types of literature. It is true that Seneca writes about his
memory at surprising length, which makes one less disposed to re-
gard the passage as a mere prefatory cliche than if he had dis-
missed the matter briefly, but we should bear in mind that he was
capable of alluding to the fiction of his sons' frivolous tastes
more than once in his work. Note too that even some of the obser-
vations in Contr. I pr. 2-5 which look most like the fruit of
Seneca's own experiences may in fact have been traditional in
mnemonic theory. We should compare Quintilian's remarks in
XI.2.6f. on the vagaries of memory:
quid? non haec varietas mira est, excidere proxima, vetera in-
haerere? hesternorum inmemores acta pueritiae recordari? quid
quod quaedam requisita se occultant et eadem forte succurrunt?
nee manet semper memoria sed aliquando etiam redit?
Two questions arise, if we adopt a sceptical view of Seneca's
claims about his memory, and it is important to distinguish be-
tween them. First, did he make use of written records of the
declaimers' work published by his contemporaries? Secondly, did
he draw on private written records which he had collected himself?
Seneca did not pretend that there was no information about
the declaimers active in his life-time available to him except
that which was stored in his head: in Contr, I pr. 11 he openly
recognizes the existence of various published works in his field
of interest. However, he is scathingly critical of them. The
value of his work, as he sees it, will be that, by drawing on his
private store of information, he will be able to supplement and
correct the inadequate, and often misleading, data contained in
these books, where they relate to the declaimers who had been
known to him personally.
It ought therefore to come as no surprise to find Seneca
occasionally making passing references to written records, and to

find that these references are sometimes accompanied by hostile
criticism. In Contr. X pr. 12, for instance, he assigns a decla-
mation, falsely ascribed to Latro, to its true author: Capitonem,
cuius declamatio est de Popillio, quae misero Latroni subicitur,
and in Contr. 1.3.11 he finds a stupid color used by Iunius Otho
all the more intolerable because the man had written books about
colores.
The extent to which Seneca actually used such sources is
tantalizingly unclear. Take, for example, one of the most
interesting instances discussed by Lockyer:27
Montanus Votienus Marcellum Marcium, amicum suum, cuius frequenter
mentionem in scriptis suis facit tamquam hominis diserti, aiebat
hanc dixisse sententiam... (Contr. IX.6.18)
Lockyer jumps to the conclusion that 'Here Seneca is drawing on a
written work which records the sayings of someone else1, and notes
also, on aiebat, 'The word itself suggests reporting by word of
mouth, although here written sources are being used.' Now, there
is absolutely nothing here to prove that Seneca took the sententia
in question from a book by Montanus, and we are free to take him
at his word and believe that he had heard Montanus orally ascrib-
ing it to his friend. On the other hand, Lockyer's instinct could
very well be right.
Seneca's explicit references to contemporary works on decla-
mation are very brief and scattered. They are not of a kind which
would lead one to assume that he kept the works in question by his
side as he wrote. Indeed, even if one is sceptical about Seneca's
memory, it is unnatural to suppose that he went through the cum-
brous process of looking up references in a wide selection of
papyrus rolls every time he wished to refer to a modern authority.
This brings us to the other question: whether Seneca's work
could have been based on some private written record he had kept
of declamations heard throughout his long life. I do not find
this hypothesis improbable. Quintilian refers to a practice,
current among schoolboys of his time, of collecting noteworthy
extracts from other people's declamations: commentariis puerorum
in quos ea quae aliis declamantibus laudata sunt regerunt
(II.11.7). It is not inconceivable that an adult enthusiast might

continue to keep his collection up to date. We may have a tell-
tale sign that Seneca's anthology is an edited version of such
commentarii, in his occasional remark, in lieu of criticism, that
such and such sententia 'was praised': hoc loco dixit Latro rem
valde laudatam (Contr. VII.1.18); hoc loco dixit Gallio illam
sententiam, quae valde excepta est (Contr. X.2.10). For the keep-
ing of collections of literary extracts on a very large scale by
adults there is clear evidence in the younger Pliny's description
of his uncle's private notes:
electorumque commentaries centum sexaginta mihi reliquit, opistho-
graphos quidem et minutissimis scriptos; qua ratione multiplicatur
hie numerus. (Ep. III.5.17)
The elder Pliny in the preface to his Naturalis historia takes
great pride in the number of obscure sources he has perused, but
it does not occur to him to write about the menial processes of
excerpting and sorting his material which he must have gone
through before producing the final version of his work, except in
so far as these processes are implied in the word lectione (NH
pr. 17). It may well have been the case that considerations of
literary dignity, the idea that it would be unthinkable to confess
in an artistic preface to the use of a private cache of decla-
matory extracts - res omnium sordidissimae, lacking even the
distinction of having originated in recondite writings - may have
been sufficient to persuade Seneca the Elder to throw a smoke-
screen around his methods in the shape of an elaborate disqui-
sition on his memory. Unwillingness to admit to the usefulness of
writing as an aid to the memory was a time-honoured theme in
ancient literature; it is well summed up by the Egyptian myth of
Theuth, discoverer of writing, related by Plato in Phaedrus 274c-
275b.28 As Lockyer's dissertation has made clear, claims to have
remembered vast quantities of material are commonplace in an-
tiquity, and, though it is fair to believe that people then took
more pains to train their memories and use them than we do, no one
will be credulous enough to believe, for instance, Athenaeus when
he claims to be reporting in the Deipnosophists the conversations
at a banquet at which he was personally present29 (Deip. I.2a).
Athenaeus1 epitomator recognizes the fact of the matter when he

says of Athenaeus1 claim to have been present: 6payaT0upy£L 6e T6V
6baXoyov 6 'Adnvauos CnA.^ IIAaTU)VLH(J3 (Deip. I.If.). Seneca's
disquisition in Contr. I pr. 2-5, though surely containing some
genuine personal reminiscences, has to be viewed against the back-
ground of the old Platonic tradition of hostility towards the
written word.
More important than the question of what written sources, if
any, the elder Seneca used, is the question whether the samples of
eloquence he attributes to the various declaimers are in fact
accurate records of the styles of these individuals. Bornecque,
citing the work of M. Sander and H. T. Karsten, and adducing some
additional arguments of his own, assures us that they are. 30
Unfortunately, though his conclusion is unlikely to be wrong, we
find that he arrived at it on the basis of untrustworthy and in-
adequate data. The reader will scarcely need to have it pointed
out that the parallel he notes between the versions of a sententia
by Gallio given by Seneca in Contr. II.3.6 and Quintilian in
IX.2.91 is no evidence at all for Seneca's reliability: modern
editors of Seneca have drawn on Quintilian's version, and editors
of Quintilian on Seneca's, to reconstruct the text, and even then
they cannot make the ancient authorities agree over which past
tense Gallio used, eras or fuisti. It is less obvious that
Bornecque's discussion of the clausulae used by Seneca and the
declaimers is riddled with errors, and that very little of the
work of Sander and Karsten, to which he refers, is of any rel-
evance to the question of the authenticity of the declamatory
fragments.
Bornecque claims to be able to show that different declaimers
varied in the extent to which they kept to the canonical rhythms
for sentence-endings, and that they differed again from Seneca,
who was more negligent in this respect than any of them. But if
we look closely, for example, at Bornecque's selection of alleged
irregular clausulae from the fragments of Votienus Montanus' dec-
lamations, we find that they include some examples from the analy-
sis of his divisio in Contr. IX.2 (not X.2, as Bornecque's
printers will have us believe), for which Seneca is responsible,
not Montanus; in one of the cases cited Seneca is actually quoting

Montanus in indirect speech (Contr. IX.2.18: responsurum).
Bornecque also assumes that the rhythm ^±L is objectionable,
whereas in fact it was one much favoured by Cicero. De Groot
gives the figures: Cicerofs speeches - 9.7%; 'normal' - 5.4%.31
Here is a subject which requires complete re-examination - a
difficult undertaking, for when an analyst of prose-rhythm tackles
the Latin of Seneca's declaimers he is faced, not only with the
usual hazards of his occupation,32 but with the difficulty of
deciding just how many of the paratactic cola and commata in which
their extracts abound end with sufficiently strong breaks in the
sense to warrant strong clausula rhythms. An element of subjec-
tivity inevitably intrudes. It is with some hesitation therefore
that I offer tentatively a few conclusions derived from study of
samples, selected and analysed as honestly as lay within my capa-
bilities, of 100 clausulae each from Seneca (prefaces), Porcius
Latro and Arellius Fuscus.33
Most of the clausula rhythms found in each of these samples
either belong to five types strongly favoured by Cicero (however
un-Ciceronian in effect the Silver Latin versions may be owing to
the preponderance of short words at the cadence), or may be re-
garded as resolutions of these types, viz.:
The figures are as follows: Seneca - 72% (inclusive of 23% resol-

utions); Latro - 72% (17% resolutions); Fuscus - 80% (15% resol-
utions) .
However, of the eight clausula rhythms which occur five or
more times in any one of the samples, two are of types generally
avoided by Cicero,31* viz.:
Occurrences (%)
Seneca Latro Fuscus Cicero 'Normal'

(speeches)
16 12 13 16.2 7.4
16 13 6 6.4 23.5
12 11 18 25.3 17.2
10 13 15 8.3 2.9
10 3 4 2.8 2.2
7 4 2 1.8 6.2
6 13 12 4.9 4.4
5 6 7 9.7 5.4
Several points emerge from these provisional findings which
suggest that clausula study on a larger scale might prove valu-
able. 35 It-would be interesting if Fuscus proved to be signifi-
cantly fonder throughout of the unresolved standard clausulae
than Seneca in his prefaces; also if, unlike Seneca and Latro, he
consistently followed Cicero in abjuring the double spondee and
the rhythm — u u y . One might have predicted that the friends
Seneca and Latro would agree to a large extent in their rhythmic
preferences: the apparent discrepancy in their attitude towards
the rhythms -w-^v^, -^-wwv^ and — ^ ^ ^ would seem all the more
significant if it were sustained consistently throughout a larger
sample. It does look very much as if we are dealing with three
distinct stylists.
Bornecque36 was over-generous in his estimate of the work of
M. Sander and H. T. Karsten on the language of Seneca the Elder
and the declaimers. Sanderfs Sprachgebrauch des Rhetors Annaeus
Seneca^1 though obviously the product of enormous labours, gives
us little help in deciding whether or not the declamatory extracts
are authentic, as no clear distinction is made in his lists of
usages between the Latin of Seneca and that of the men quoted by
him. The one observation by Sander on the question of authen-
ticity which Bornecque refers to, namely that Latro alone uses the
term idcirco whereas the others use ob hoc, ob illud etc., appears
similarly in isolation in the introduction of Sprachgebrauch I. 38
We are referred back for further examples to Sanderfs earlier
dissertation, Quaestiones in Senecam Rhetorem syntacticae
(Greifswald, 1872), a work which Bornecque does not appear to have

used. As copies of this dissertation are rare, and the question
of the authenticity of the fragments is treated less cursorily in
it than elsewhere, though still not at all adequately, it seems
worth quoting SanderTs observations in full, only bringing the
system of reference up to date:3^
...non pauca vocabula sunt, quae Seneca singulos rhetores pro-
nuntiantes faciat, a quibus tamen ipse abhorreat, ipse rursus
saepius talibus vocibus utitur quas rhetores aliis eiusdem
notionis verbis circumscribunt. utriusque rei ex permultis
exemplis nonnulla eligam. Latro enim locutionibus naturale est ut
(Contr. VII.2.8), facile est ut (Contr. IX.2.3), idcirco (Contr.
II.7.8) utitur, quae apud Senecam ipsum non inveniuntur.t+0 item
Asinius Pollio duobus locis namque (Suas. 6.24), quando vi causali
(ibid.), item Albucius Silus expecto ut (Contr. VII.4.1), mereo ut
(Contr. VII.8.1) scribit; item Papirius Fabianus enimvero (Contr.
II.1.11); Arellius Fuscus mos est ut (Contr. 1.5.8), aequum est ut
(Contr. IX.3.I); 41 Cestius Pius (?) licet ut (Contr. IX.4.10); 42
Pompeius Silo iubeo ut (Contr. 1.2.15);43 Cremutius Cordus aliter
ac (Suas. 6.19) ;**** Titus Livius ut cum participio futuri con-
iunctum (ibid. 17), aliaque multa quae Seneca ipse evitavit. e
contrario Seneca (Contr. I pr. 15f.) ideoque locutione utitur,
ceteri rhetores constanter et ideo; lex est ut (Contr. I pr. 7); 1 * 5
spero ut (Contr. IX.2.18); iam vero (Contr. I pr. 17; III pr. 4 ) .
Unfortunately the value of these observations, promising though
they may seem, is limited. Presumably this summary lists the
variations of usage which Sander found most significant 'ex per-
multis exemplis1, but it is marred by inaccuracies: the words in
Contr. 1.2.15, iubemur ut, are surely to be attributed to Seneca,
not Pompeius Silo; the attribution of the phrase speramus ut in
Contr. IX.2.18 to Seneca is probable but uncertain; the attribu-
tion of licere...ut in Contr. IX.4.10 to Cestius, which he marks
as doubtful, is indeed pure guess-work, and Cestius is found else-
where (Suas. 2.6) using licere + infin.; it also needs pointing
out that if Seneca is quoting Latro verbatim in Contr. 1.1.13,
II.6.5 and VII.6.13, Latro used oh hoc and ob id in his divisiones.
Moreover, the expressions cited by Sander as peculiar to particu-
lar stylists in Seneca's anthology are all very rare ones; where
one example of a usage is cited it is probably the only example
Sander could find; at any rate, when attempting to check his
claims, I found that one could not expect to find in a single
declamation-survey more than one of the abnormal usages he refers
to in Quaestiones 5f. It does not seem that we have a large

enough sample of the writing of Seneca the Elder to be sure that
he would always have avoided all the expressions which Sander
reckons to have been peculiar to the declaimers quoted by him.
H. T. Karsten makes some attempt in his Elocutio rhetorica
qualis invenitur in Annaei Senecae Suasoriis et Controversiis1*^
to distinguish between the usage of Seneca and the various de-
claimers when listing abnormal expressions, but only in the case
of very rare ones; in the case of commoner abnormalities, the
distribution of which might be of much interest to us, he gives us
just book and section references and expects us to be content with
the information (given after a list of abbreviations of rhetor-
icians1 names in capital letters) that Werba ac nomina, quibus
litteras maiusculas non adscripsi, pleraque ab ipso Seneca non-
nulla a compluribus rhetoribus usurpata sunt.'1*7 Even in the case
of great rarities he sometimes lapses from his good intentions,
omitting to tell us, for instance, that the phrase supplex
accadens genibus in Suas. 6.3 for which he found no parallel, was
Latrofs. His introductory observations on linguistic differences
between the declaimers are quite untrustworthy - note in particu-
lar that he is quite wrong when he suggests that Latro used the
word alioquin in a sense peculiar to himself48 - and his method of
listing rarities, like SanderTs, seems calculated to exhaust the
patience of the best-intentioned reader. The fundamental trouble
with the approach of these two scholars is that they never make it
clear to what extent their presentation of material is selective.
What is needed before any serious work on the language of the de-
claimers can be done is nothing short of a complete index verborum
to Seneca the Elder, in which usages of Seneca himself are clearly
distinguished from those attributed to the various declaimers and
again from those found in passages where Seneca appears to be
reporting the words of critics or declaimers either obliquely or
verbatim. Until such a work exists there seems little hope of
pin-pointing subtle differences between the usages of the various
declaimers in a way which will convince the sceptical reader that
Seneca recorded their words faithfully.49
There is certainly some general correlation between the

styles of the declaimers as they are described by Seneca in his

prefaces and the styles found in the extracts attributed to them.
Particularly striking in the extracts are certain peculiarities of
the rhetorician Arellius Fuscus and his independent-minded pupil
Fabianus to which Seneca refers in the second preface.50 However,
that there is such a correlation is not necessarily proof of the
complete authenticity of the fragments: a fraudulent or imprecise
anthologist would have been certain to make the extracts he pre-
sented display the stylistic qualities he had mentioned when
criticizing the writers to whom they were attributed. In view of
this, it seems worth drawing attention to two correspondences of a
type which it is hard to account for if Seneca has not given us
authentic fragments. In Contr. VII pr. 4f. Seneca mentions that
Albucius used sometimes to imitate Fabianus and Hermagoras. It is
interesting to find that parallels between Albucius1 and Fabianus1
sententiae occur, not in the controversiae which one imagines
Seneca treated while the seventh preface was still fresh in his
mind, but in the first Suasoria (Suas. 1.3, 9-10), and that we
have to wait until Contr, X.I for the one clear parallel between
Albucius and Hermagoras (Contr. X.I.I, 15). 5 1
To the general rule that the declamatory extracts confirm
well enough what Seneca the Elder says in his prefaces about the,
men to whom they are attributed, there appears, however, to be one
glaring exception. The case of Votienus Montanus obliges us to
question our most fundamental assumptions about the soundness of
the elder Senecafs methods. In Contr, IX pr. 1 we read: Montanus
Votienus adeo numquam ostentationis declamavit causa ut ne
exercitationis quidem declamaverit. Seneca then reports how, when
he asked him why he took this line (rationem quaerenti mihi), he
answered with an uncompromising tirade against the pretensions of
the schools of rhetoric (§1-5). Yet in Contr. VII.5.12 we are
told that Votienus Montanus was toto animo scholasticus, and
furthermore, books IX and X of the Controversiae contain twenty-
two references to his treatment of declamations.52
No attempt yet made to account for this contradiction is at
all convincing. 0. Gruppe,53 in an attempt to eliminate the
phrase toto animo scholasticus from Contr. VII.5.12, had recourse
to the wildest of conjecture, and even if one accepted his theory

one would still have to explain away the presence of declamatory
extracts attributed to Montanus in Contr. IX and X. R. Hess51* and
55
Bornecque looked for a third reason, apart from the two he is
said in Contr. IX pr. 1 to have rejected, why Montanus might have
treated declamatory themes. The explanation of Hess that he did
not deliver full declamations but frequented the schools ut
disputaret5^ has surely to be discounted in view of Contr. IX.4.14:
Montanus partem accusatoris declamavit. Bornecque1s suggestion,
f
il declama non par ostentation ou pour s'exercer, mais vrai-
semblablement pour faire comme tout le monde', fails to take into
account the fact that ostentatio and exercitatio mark opposite
poles in the range of possible reasons for declaiming, and that
the effect of Seneca's use of polar expression in the crucial
sentence is to lead the unbiased reader to conclude that Montanus
did not declaim at all. The hypothesis advanced by W. Hoffa 57
that illud 'Montanus Votienus adeo numquam ostentationis declamavit
causa, ut ne exercitationis quidem declamaverit' magis ad verba
Votieni quam ad facta spectat, ut ipsius Votieni verba hoc loco
tamquam suo ore reddere Senecam censeamus, and that the origin of
1
Seneca's statement in Contr. IX pr. 1 lay in a claim, ego non
declamo' made by Montanus himself, seems totally implausible,
though it points the way towards an explanation of the mystery
which is perhaps not so incredible. This is that in the ninth
preface Seneca derived his material not, as he suggests, from an
actual conversation he had once had with Montanus, but from one of
the scripta to which he refers in Contr. IX.6.18,58 perhaps a
dialogue in which Montanus, though himself toto animo scholasticus,
presented an interlocutor who declared that he never declaimed,
even for the most austere of educational motives, and proceeded to
inveigh against the schools in the words reported by Seneca in
Contr. IX pr. 1-5. Such an explanation does not do much credit to
poor Seneca, and one may hope it is not the right one. An
alternative perhaps worth considering is that the attribution to
Votienus Montanus of a hard line against the schools is the result
of textual corruption in Contr. IX pr. 1, most likely arising from
an ambiguous abbreviation in Seneca's original manuscript or an
early copy of it. One would expect the man introduced in the
ninth preface to have been quoted perhaps elsewhere by the elder
Seneca as a critic, but never as a declaimer. Very few people
mentioned by him come into this category, but interestingly enough
they include two whose names, if abbreviated, might well have been
confused with that of Votienus Montanus, namely, Valerius Messala,
who shared his initials, and the poet Iulius Montanus.Dy
The question whether Seneca^ quotations give us authentic
records of the declaimer^ words must remain unanswered at least
until tihere exists a full index verborum to the work, and the
prose rhythm of Seneca and the men quoted by him has been studied
more thoroughly. Nevertheless, it is surely justifiable to adopt
the working hypothesis that, before the text became corrupt in the
course of transmission, the extracts were accurate. One fails to
see why Seneca the Elder, if he did not possess a reliable record,
memorized or written, of the actual words of the multitudinous
declaimers he quotes, would have undertaken the labour of com-
piling the type of anthology which he has given us. It seems
about as plausible that he should have set out to forge thousands
of declamatory fragments as that Athenaeus, for instance, should
have forged all his quotations from Greek comedy. If the elder
Seneca had been content with partially accurate remembrances, or
had a taste for forgery, why did he not, for example, take on the
much easier task of passing off as Latro's some complete declama-
tory compositions of his own, with or without a Latronian core?
It will be assumed in my discussion of declamatory styles that
Seneca recorded the words of the declaimers faithfully, though
textual corruption has inevitably diminished the reliability of
his record.
3 THE CRITICISM
The chief glories of Senecafs criticism are the pen-portraits,

mainly to be found in the prefaces. These remarkable psychologi-
cal and critical descriptions have been more often imitated than
discussed. Fine imitations of his manner were produced in this
country particularly in the seventeenth century, when English
prose had an incisiveness and luminosity strangely akin to
Seneca1s, and hard to recapture now. Following the example of
Montaigne who, for instance in his essay Du parler prompt ou
tardif, had looked to the elder Seneca for reflections from which
morals could be drawn -
On recite de Severus Cassius, qufil disoit mieux sans y avoir
pense; qu'il devoit plus a la fortune qu'a sa diligence; qu'il luy
venoit a profit dfestre trouble en parlant, et que ses adversaires
craignoyent de le picquer, de peur que la colere ne luy fit re-
doubler son eloquence. (Cf. Contr. Ill pr. 6 ) 1 -
English essayists took to adapting Senecafs anecdotes. Here is
Abraham Cowley imitating Suas. 2.17 in his essay On greatness:
...he would eat nothing but what was great, nor touch any Fruit
but Horse-plums and Pound-pears. He kept a concubine that was a
very Gyantess, and made her walk, too, alwaies in chiopins, till
at last, he got the surname of Senecio Grandio, which, as Messala
said, was not his Cognomen, but his Cognomentum.^
But earlier than Cowley, and probably more out of admiration for
Seneca the Elder as a sensitive observer of literary qualities
than as a purveyor of anecdotes suitable for moralization, Ben
Jonson in his commonplace book, the Discoveries, had paraphrased
several passages from Senecafs prefaces, for example:
One, though hee be excellent, and the chiefe, is not to bee imi-
tated alone. For never no Imitator, ever grew up to his Author;
likeness is alwayes on this side Truth: Yet there hapnTd, in my
time, one noble Speaker, who was full of gravity in his speaking.
His language (where hee could spare, or passe by a jest) was nobly
censorious. No man ever spake more neatly, more presly, more
weightily, or sufferfd lesse emptinesse, lesse idleness, in what
hee utterfd. No member of his speech, but consisted of the owne
graces: His hearers could not cough, or looke aside from him,
The criticism 51
without losse. Hee commanded where hee spoke; and had his Judges
angry, and pleased at his devotion. No man had their affections
more in his power. The feare of every man that heard him, was,
lest hee should make an end.^
Though Jonson is paraphrasing, after the initial comments on imi-
tation based on Contr. I pr. 6, Seneca1s description of Cassius
Severus (Contr. Ill pr. 4, If.), a marginal note, fDominus
Verulanus' (i.e. Francis Bacon, Lord Verulam), assures us that he
did indeed have a particular contemporary in mind. Later he actu-
ally adapts Senecan criticisms (Contr. Ill pr. 6; I pr. 6) to
English statesmen named in the body of his text:
Lo: Egerton, the Chancellor, a grave, and great Orator; and best,
when hee was provok'd. But his learned and able (though unfortu-
nate) Successor is he, who hath fill'd up all numbers; and per-
form1 d that in our tongue, which may be compar'd, or preferr'd,
either to insolent Greece, or haughty Rome.
Seneca1s phrase insolenti Graeciae (Contr. I pr. 6) appealed to
Jonson so much that elsewhere in some famous verses he compares
Shakespeare1s comedies favourably with
all that insolent Greece, or haughtie Rome
Sent forth, or since did from their ashes come.5
Thus Seneca's prefaces served as an important model for Jonson's
attempts at a type of writing uncommon then in English, the de-
scription of the literary qualities of contemporaries.
But, inspiring though they once proved to be to eminent
moralists and critics, Seneca's character sketches do not lend
themselves to systematic analysis or classification.6 They vary
enormously in length; they are also very diverse in their con-
stituents. Some give us mainly biographical information; others
contain quite extensive literary description. For example, the
fourth preface consists of the introduction of a contrasting pair
of men, Asinius Pollio and Q. Haterius. The pen-portrait of
Pollio consists almost entirely of reflections on his character
and pleasing anecdotes about his private life; true literary
criticism is restricted to one sentence:
floridior erat aliquanto in declamando quam in agendo: illud
strictum eius et asperum et nimis iratum ingenio suo iudicium adeo
cessabat ut in multis illi venia opus esset quae ab ipso vix in-
petrabatur. (Contr. IV pr. 3)
On the other hand, Haterius' introduction, though it opens with an
anecdote illustrative of his character, consists mainly of a long

and detailed analysis of his declamatory manner: we are told about
his excessively rapid delivery, his inability to refrain from
developing each point in an argument at inordinate length, the
lack of order in his declamations which resulted from this fail-
ing, and the eccentricity of his diction in relation to the canons
of taste usually accepted in the schools (Contr. IV pr. 7-11).
The techniques by which Seneca brings to our attention the
salient characteristics of the declaimers range from the simplest
to the most oblique. Most often, of course, he describes charac-
ter and style in simple direct statements: the sentence quoted
above on Pollio, floridior...impetrabatur (Contr. IV pr. 3), will
serve as one example. His talent for the analysis of style by
direct means is perhaps best illustrated by his description of the
declamatory manner of Arellius Fuscus (as compared with that of
his pupil, Fabianus) in Contr. II pr. 1:
erat explicatio Fusci Arelli splendida quidem, sed operosa et im-
plicata, cultus nimis adquisitus, conpositio verborum mollior,
quam ut illam tarn sanctis fortibusque praeceptis praeparans se
animus pati posset; summa inaequalitas orationis, quae modo exilis
erat, modo nimia licentia vaga et effusa: principia, argumenta,
narrationes aride dicebantur, in descriptionibus extra legem
omnibus verbis, dummodo niterent, permissa libertas; nihil acre,
nihil solidum, nihil horridum; splendida oratio et magis lasciva
quam laeta.
Though this passage shows that Seneca was quite capable of vivid
evocation of style by means of a densely packed collection of
specifically critical nouns and adjectives, he did not usually go
about criticism in this way. He does not blind us with the ex-
cessive use of technical terminology which can make some ancient
critics heavy reading. His critiques normally come mixed with
illustrative examples and anecdotes. For instance, he could have
dismissed the declaimer Musa with a few insulting adjectives,
tumidus, corruptus, insanus, or the like. Instead, he gives us
specific examples of his bad taste:
omnia usque ad ultimum tumorem perducta, ut non extra sanitatem,
sed extra naturam essent. quis enim ferat hominem de siphonibus
dicentem fcaelo repluunt1 et de sparsionibus Todoratos imbres1...?
(Contr. X pr. 9)
Anecdotes serve to add precision to critical statements otherwise
The criticism 53
open to more than one interpretation, and may incidentally shed

light on aspects of the subjectfs character which are nowhere
described in plain terms. Thus an anecdote intended to illustrate
Latro's fondness for sententiae tells us much also about his
relationship with the master who taught him rhetoric:
hoc quoque Latro meus faciebat, ut sententias amaret. cum con-
discipuli essemus apud Marullum rhetorem, hominem satis aridum,
paucissima belle, sed non vulgato genere dicentem, cum ille
exilitatem orationis suae imputaret controversiae et diceret:
f
necesse me est per spinosum locum ambulantem suspensos pedes
ponere', aiebat Latro: 'non mehercules tui pedes spinas calcant,
sed habentf; et statim ipse dicebat sententias, quae interponi
argumentis cummaxime declamantis Marulli possent. (Contr. I pr. 22)
As an alternative to his practice of illustrating direct
statements with examples or anecdotes, Seneca sometimes prefers to
allude to criticisms made by the declaimers and members of their
audiences, without always making it clear whether he agrees with
the critic quoted or not. For example in Contr, II.3.22 Cestius
finds fault with a color used by Fuscus in a controversia about a
raptor who has received pardon from the father of the raped girl
within the prescribed time-limit, but not from his own:
Fuscus Arellius dixit: magnam partem legis consumpsi nee de mora
queror: raptae pater rogabatur. Cestius non probabat, et hac
sententia usus est cum hunc colorem argueret: dum vult videri
rogatum diu raptae patrem, efficit ut videatur suum diu non
rogasse; malo autem videri huius patrem tarde exorari quam tarde
rogari.
Seneca gives us no indication what he personally thought of the
color, and, seeing that neither rhetorician cited is one for whom
he expresses whole-hearted approval, it is impossible to be sure
whether or not he approved of anything in Cestius1 criticism
beyond the neat manner in which it was expressed. That he was by
no means always in agreement with criticisms which he quoted with-
out comment is evident from the extreme case of the third preface,
eleven sections of which (8-18) are devoted to an attack by Cassius
Severus on the pretensions of the rhetoricians, to which he makes
no answer, although it seems clear from other evidence, notably
from Contr. II pr. 4, that he himself firmly believed in the value
of an extended rhetorical education. Fortunately he does some-
times indicate his disapproval of other peoplefs criticisms, as
for example in Contr. 1.3.10 where he condemns Cestius1 cruel jibe
at Quinctilius Varus the younger, and similarly he does sometimes

note when a criticism he is quoting seems to him to contain the
mot juste, as, for example, when he quotes Pollio on Albucius'
style in Contr. VII pr. 2: sententiae, quas optime Pollio Asinius
albas vocabat, simplices, apertae, nihil occultum, nihil in-
speratum adferentes, sed vocales et splendidae.
Uncertainty whether Seneca shared the views of the critics he
quotes is one of the problems which we have to face when discuss-
ing his criticism. The only safe policy to adopt is not to regard
as Senecan any quotation from another critic which he does not (as
he does in the case of Pollio's adjective, albas) specifically
praise. Not that it is always a simple matter to tell whether
Seneca is quoting someone else, or criticizing on his own account.
A grammatical idiosyncrasy worth watching for in his writing is
his apparent tendency to mix direct with indirect speech. It
often seems unlikely that a change to direct speech signifies that
he has stopped reporting. A good example comes in his report of a
lecture by Cestius on the different varieties of suasoriae (Suas.
1.5):
aiebat Cestius hoc genus suasoriarum <alibi> aliter declamandum
esse [quam suadendum]. non eodem modo in libera civitate dicendam
sententiam, quo apud reges, quibus etiam quae prosunt ita tamen,
ut delectent, suadenda sunt. et inter reges ipsos esse discrimen:
quosdam minus, alios magis osos veritatem; facile Alexandrum ex
iis esse, quos superbissimos et supra mortalis animi modum in-
flatos accepimus. denique, ut alia dimittantur argumenta, ipsa
suasoria insolentiam eius coarguit; orbis ilium suus non capit.
itaque nihil dicendum aiebat nisi cum summa veneratione regis, ne
accideret idem, quod praeceptori eius, amitino Aristotelis,
accidit...
It would be unnatural to take the sentence, denique...capit, as
other than part of Cestius1 disquisition, even though it is wholly
cast in direct speech, especially as its logical consequence,
introduced by itaque, is definitely ascribed to him.
Seneca's willingness to stand back, as it were, and let other
critics air their views, is a sign that he has given us a remark-
ably objective survey of the literary circles in which he moved.
However, to someone attempting to pin-point the place of the elder
Seneca's literary criticism in relation to that of more self-
assertive critics like Cicero and Quintilian it poses something of
The criticism 55
a problem to find that he does not invariably cap the opinions of

others by a statement making his own opinion on the issues quite
clear.
Also, though it is possible, as we shall see, to form a pic-
ture of the types of teaching which he accepted on many issues of
rhetorical theory which fall within the five topics into which the
orandi ratio was customarily divided, inventio, disposition
elocutio, memoria, actio (cf. Quint. III.3.1), this can only be
done by collecting very scattered references from all parts of the
work, most of which are descriptive of a particular declaimer's
approach to a particular thema, rather than being cast in the form
of didactic generalizations. Hence, before we can compare his
views with those of, for example, Cicero and Quintilian, we have
to impose a framework on his criticism which bears no relation to
the way in which it was originally set out. For Seneca the Elder
is unusual among ancient critics in feeling no desire to arrange
his criticism according to the rigid divisions of oratory pre-
scribed in the rhetorical handbooks.
When Dionysius of Halicarnassus, who was living and working
in Rome in SenecaTs younger days, wanted to discuss an oratorfs
style, he did so with a certain thorough-going professionalism and
according to f a system so rigorously defined that it may be set
out in tabular form1.8 S. F. Bonner, in his monograph on this
critic, actually sets out a table showing how Dionysius aimed, by
considering in turn word-choice, composition, and figures, and by
weighing up the writer's virtues and failings., to assign his style
to one of three categories: plain, middle, or grand.9 Seneca, by
contrast, shows no desire to be systematic. He alludes, of
course, in passing, to diction, composition figures, and genera
dicendi, but he evidently felt no need to discuss each aspect of
style in turn when writing about a declaimer, and certainly had no
desire to categorize anyone's style as plain, middle, or grand.
Of course, unlike Dionysius, he was writing about men he had known
personally, not about classical orators of three or four hundred
years before. That made a difference. Seneca could start his
descriptions with a tabula rasa9 uninhibited by any accumulation
of critical tradition. Still, a less independent spirit would
probably have fallen back on the three-style system of literary

classification, even when describing contemporaries.
Seneca1s independence from the Hellenistic and Ciceronian
tradition of didactic writing on rhetoric is evident again in the
absence from his criticism of generalizations about certain large
literary polemics which other critics found of absorbing interest,
and in his lack of concern to tell us plainly where he stood in
relation to the opposing parties in any battle of books. The
controversy between Atticists and Asianists which had seemed so
important to Cicero is mentioned by Seneca surprisingly rarely.
He applies the epithet Asianus to a mere handful of declaimers and
only on one occasion refers to any Attici.10 Similarly he makes
only a few references to the followers of Apollodorus of Pergamum
and Theodorus of Gadara, whose disagreements were certainly a live
issue in the schools of the early Empire, and does not swear al-
legiance to the ideals of either party.^ Probably it was chiefly
because he assumed adequate knowledge of these polemics on the
part of his readers that he felt no need to pontificate on the
questions being disputed, for he was not to know that future gen-
erations would be almost wholly dependent on his books for a
knowledge of the Latin rhetoric of his time.
Perhaps the elder Seneca, as a critic, may be likened, if not
to the proverbial man who cannot see the wood for the trees, then
to one who assumes that the topography of the wood is common
knowledge (the wood in question - the whole Greco-Roman rhetorical
tradition - having been mapped out adequately by other men before
him) and prefers to examine a few single trees in detail; this he
does with an artist's eye, relying on instinct to show him the
important peculiarities which make each an individual, rather than
approaching them with systematic flora in hand, as a man like
Dionysius of Halicarnassus might, with a view to determining genus
and species.
There are certain recurring themes to be found in Seneca's
pen-portraits, but they have nothing to do with classifications
set out in rhetorical handbooks.
We find he is particularly intrigued, for example, by
people's inconsistencies. He devotes a long section of the first
The criticism 57
preface (Contr. I pr. 13-15) to a description of Latro's unalter-

able habit of working too hard and playing too hard: in utramque
partem vehementi viro modus deerat: nee intermittere studium
sciebat nee repetere etc. In the second preface Arellius Fuscus*
extreme unevenness of style occasions comment: summa inaequalitas
orationis, quae modo exilis erat, modo nimia licentia vaga et
effusa.,. (Contr. II pr. 1). The main theme of the third preface
is the phenomenon, exemplified by Cassius Severus, of men who made
good orators, but mediocre declaimers: quosdam disertissimos cog-
novi viros non respondentes famae suae cum declamarent, in foro
maxima omnium admiratione dicentes, simul ad has domesticas
exercitationes secesserant desertos ab ingenio suo (Contr. Ill
pr. 1 ) . He remarks also on a certain discrepancy between Severus1
character and his oratorical manner: nee enim quicquam magis in
illo mirareris, quam quod gravitas, quae deerat vitae, actioni
supererat (Contr. Ill pr. 4 ) . In the fourth preface, Pollio's
unwillingness to declaim in public is contrasted with his import-
ant contribution to the rise of public literary performances of
another kind; his declamatory style is contrasted with his ora-
torical manner (Contr. IV pr. 2f.). Haterius' theoretical atti-
tude towards the arrangement of argument is contrasted with his
practice: dividere controversiam putabat ad rem pertinere, si
ilium interrogares, non putabat, si audires (Contr. IV pr. 9 ) .
Albucius, introduced in the seventh preface, changed his manner
according to the size of his audience, so Seneca tells us (Contr.
VII pr. 1 ) , and his style was marred by his mixing of incompatible
types of vocabulary: splendidissimus erat: idem res dicebat omnium
sordidissimas (Contr. VII pr. 3). In the tenth preface, Scaurus1
potential as an orator is contrasted with his actual achievements
(Contr. X pr. 2-3); the censorious image Labienus wished to pro-
ject is contrasted with his true character (ibid. 4 ) ; Musa is rep-
resented as talented enough, but deficient in other respects
(ibid. 9 ) .
Seneca is also interested in the extent to which external

factors enhanced or damaged a man's literary reputation. We are
told of Alfius Flavus: semper autem commendabat eloquentiam eius
aliqua res extra eloquentiam: in puero lenocinium erat ingenii
aetas, in iuvene desidia (Contr, 1.1.22). Cassius Severus, we

learn, owed a good deal of his reputation to his physical pres-
ence: primum tantundem erat in homine quantum in ingenio: corporis
magnitudo conspicua, suavitas valentissimae vocis... (Contr. Ill
pr. 3 ) . The minor declaimer L. Magius would hardly have attracted
an audience if he had not been the son-in-law of Livy (Contr. X
pr. 2 ) ; on the other hand, distinguished family connections could
prove disastrous for one's career, as happened in the case of
Asinius Gallus: ...magnum oratorem, nisi ilium, quod semper
evenit, magnitudo patris non produceret, sed obrueret (Contr. IV
pr. 4 ) . Nevertheless a really exceptional man might rise to emi-
nence from the most unpromising beginnings. Seneca marvels at the
way Labienus overcame his many disadvantages:
magnus orator, qui multa impedimenta eluctatus ad famam ingeni
confitentibus magis hominibus pervenerat quam volentibus. summa
egestas erat, summa infamia, summum odium, magna autem debet esse
eloquentia, quae invitis placeat, et cum ingenia favor hominum
ostendat, favor alat, quantam vim esse oportet, quae inter ob-
stantia erumpat! (Contr. X pr. 4)
Another theme which Seneca harps on is the way in which de-
claimer s abused, or improved on, their natural gifts. Neglect of
vocal exercises, together with inadequate sleep, ruined Latrofs
originally sound voice: vox robusta, sed surda, lucubrationibus et
neglegentia, non natura infuscata (Contr. I pr. 16); in general,
his irregular habits of life undermined his health:
iam vero quin rem inimicissimam corpori faceret, vetari nullo modo
poterat: post cenam fere lucubrabat nee patiebatur alimenta per
somnum quietemque aequaliter digeri, sed perturbata ac dissipata
in caput agebat; itaque et oculorum aciem contuderat et colorem
mutaverat. (Contr. I pr. 17)
Alfius Flavus was another man careless of his natural vigour, but
in his case it was simple indolence, together with a taste for
poetry, that brought about his downfall (Contr. 1.1.22). The
choice of a life of philosophic calm, involving as it did the sup-
pression of excessive emotions, was in itself an admirable thing,
but could ruin one as an effective declaimer. This was the case
with Fabianus: cum veros conpressisset adfectus et iram doloremque
procul expulisset, parum bene imitari poterat quae effugerat
(Contr. II pr. 2 ) . On the other hand people sometimes made an
effort to improve their natural rhetorical talents. Latro
The criticism 59
improved his memory by learning mnemonic techniques: memoria ei

natura quidem felix, plurimum tamen arte adiuta (Contr. I pr. 17) .
In general, the way to overcome mediocrity was by hard work:
diligentiam, maximum etiam mediocris ingenii subsidium (Contr.
Ill pr. 7 ) .
It is clear that the elder Senecafs criticism is of a kind
set apart from the main stream of ancient writing on rhetoric to
which Cicerofs Rhetorica and Quintilianfs Institutio oratoria
belong. Seneca1s most characteristic type of criticism is the
pen-portrait in which the subjectTs human foibles are usually
given as much attention as his style. In addition there is tech-
nical analysis of argumentation in the sections about divisio, and
comparative criticism of varying complexity in the closing stages
of his surveys of declamations. One important respect in which he
differs fundamentally from Cicero and Quintilian is that he does
not set out to give straightforward instruction to the student of
rhetoric. His approach is only didactic to the extent that his
examples are intended to serve as models for imitation or warnings
of faults to be avoided. He rarely lays down any general rules
about the way the various parts of a declamation should be
treated. Still less Ms his aim to theorize about the qualities of
the hypothetical ideal orator. To adopt a convenient modern div-
ision of literary criticism into three categories,-12 Seneca1 s
criticism is primarily fdescriptive1, rather than legislative*
or 'theoretical1.
Most of the classics of ancient criticism belong to the lat-
ter two categories. Indeed, scant recognition has been given in
histories of the critical tradition to the fact that descriptive
criticism existed in antiquity at all. It did exist, but its
practitioners tend not to be given the name of 'critics1: we know
them rather as philosophers, grammatici, scholiasts, epistolo-
graphers, historians or biographers. It is with the biographical
tradition that the elder Seneca's work has most affinity. This
becomes clear if we look at passages in Suetonius' Lives of the
Caesars in which the literary preferences of the various emperors
are described.
For example, in the Life of Augustus straightforward
description of Augustus1 style of speaking is enlivened by al-

lusions to his ipsissima verba on the types of rhetoric he tried
to avoid:
genus eloquendi secutus est elegans et temperatum, vitatis
sententiarum ineptiis atque concinnitate et 'reconditorum
verborum1, ut ipse dicit, ffetoribus', praecipuamque curam duxit
sensum animi quam apertissime exprimere. (Aug. 86.1)
Suetonius also illustrates the emperor's tastes by quoting dis-
paraging remarks he made about named individuals:
cacozelos et antiquarios, ut diverso genere vitiosos, pari
fastidio sprevit, exagitabatque nonnumquam; in primis Maecenatem
suum, cuius 'myrobrechis', ut ait, Tcincinnos' usque quaque per-
sequitur et imitando per iocum irridet. sed nee Tiberio parcit et
exoletas interdum et reconditas voces aucupanti. M. quidem
Antonium ut insanum increpat, quasi ea scribentem, quae mirentur
potius homines quam intellegant. (Ibid. 2)
There is no closer parallel in Roman literature than this for the
way Seneca so often enlivens his criticism by recording illuminat-
ing comments made by a declaimer about himself, or vituperation
flung by one rhetor at another. Suetonius, like Seneca, is not
content with stereotyped generalizations about style: indeed, in
his efforts to pin down the essential qualities of Augustus1
writing, he goes into even more detail than Seneca ever does, not
only listing colloquialisms he was prone to use, but even noting
peculiarities of his orthography (Aug. 87f.). Like Seneca,
Suetonius thought it as important to discuss a man's delivery and
memory as his diction and style, and tried to find a psychological
explanation for the manner he adopted in public. Seneca had dis-
cussed with interest Albucius' unwarranted fear of improvisation
(Contr. VII pr. 2) and the lack of self-confidence which made him
unable to give up imitating other people's styles and settle on
one of his own (Contr. VII pr. 4f.). Suetonius tells us (Aug. 84)
that Augustus also avoided impromptu speaking, and was so afraid
of memory lapses that he took to preparing anything important he
had to say in writing, even his more serious conversations with
his wife.
Suetonius, of course, wrote later than the elder Seneca, and
seems to have known his work, at least in an epitomized form,13 so
it would be more satisfactory if one could find an equally close
parallel for the mixture of stylistic analysis and psychological
The criticism 61
description characteristic of Seneca1s pen-portraits in earlier

biographical writing. Unfortunately the remains of such pre-
Suetonian biography as would be of interest to us are extremely
fragmentary.14
Parallels may be adduced between the elder Seneca and certain
Greek biographical miscellanists who, like him, chose to write
about their contemporaries. A certain similarity between Seneca1s
manner and that of Antigonus of Carystus, who wrote in the third
century B.C. about recent philosophers, was noted by Wilamowitz.*5
Georg Misch 16 has looked even further back and seen Ion of Chios
as a representative of an allied tradition.
Ionfs Epidemiai, an anecdotal record of personalities to be
encountered in the fifth century B.C., included verbatim accounts
of literary conversations, such as the one in which Sophocles took
part about the correct language in which to describe rosy cheeks.
Ion concludes this anecdote (FGrHist 392 F 6) with a reflection on
an inconsistency in Sophocles' character: he did not, so we are
told, show in public life the same cleverness he displayed in
poetry and witty conversation. A remark of Plutarch (Pericles
5.3) confirms that Ion, like Seneca, delighted in presenting per-
sonalities as two-sided, oocntep Tpayuxriv 6u6aaxaAuav d^touvxa xnv
TL
dpexnv sfxeuv itdvxws xal aaxupuxov yepos.
How far the parallel noted by Wilamowitz between Antigonus
and Seneca is valid can be illustrated by two fragments in par-
ticular. A fine example of the theme of contradictions in charac-
ter was to be found in his pen-portrait of Menedemus. This philo-
sopher's violence in debate was contrasted with his mildness in
private life, and the generosity which he extended even to his
academic enemies:
ev 6e TOILS CnTnaeau, cpncrCv, &6e ydx^yos ?iv wad' UTtajTuua cpepwv
aTtfieL. oyws 6' o$v xououxos ev xous Aoyous uitapx^v ev TOLS '
y p
oycos auxov eft eTtotncre xnv yuvaCxa iiapaireycljas ex AeAcpwv
XaAxl6os, euAa$ouyevriv xds KAwireuas xe xal xds xa§' 66ov
C . (Diog. Laert. 11.136)
Another fragment, on Polemon (Diog. Laert. IV.17), may be compared
with Seneca's marvelling at the way the young Fabianus suppressed
his emotions (Contr. II pr. 2) : xoaouxov 6e eTtoxetvau x6 ?i§os
dp£auevov cpuXoaocpeuv, wax' euu TCXUTOU ax^yotxos TTJS yopcpns

yeveuv.
Whether there existed any comparable descriptions of rhetor-
icians by Hellenistic writers is something of a mystery. The
first book of Philostratus1 Lives of the sophists, which one might
have expected to draw on any such works as there were, proves to
be the reverse of informative on this point. In the text as we
have it, the gap in the succession of sophists between Aeschines,
who was in Philostratus' view the founder of the 'second soph-
istic', and Nicetes of Smyrna, a rhetorician of the first century
A.D., is bridged only by a meagre praeteritio in which three minor
rhetoricians, Ariobarzanes of Cilicia, Xenophron of Sicily, and
Peithagoras of Cyrene, are dismissed as unimportant (Vit. soph,
1.511). Whether this extraordinary gap is due to a lacuna in the
text, or the unavailability to Philostratus of any sources relat-
ing to the last three centuries B.C. is an open question.
Certainly the Hellenistic writers who wrote compendious works
on famous men included in their works biographical sketches of
orators and rhetoricians. For instance, Hermippus 'the Calli-
machean'17 wrote books about Aristotle and his pupils, who in-
cluded the orator Demetrius of Phalerum, and about Isocrates and
his school, which included Isaeus, Demosthenes, Hyperides, and
many minor orators and rhetoricians. However, to reconstruct the
life story of a great man already dead, using secondary sources and
his own writings, which is what Hermippus and most Hellenistic
biographers were usually doing, requires quite different methods
from those employed by an anecdotist who is describing his contem-
poraries, and may result in a very different kind of end-product.
A certain amount of literary criticism will have been mixed with
the biographical narrative in such Lives, but it may have been of
low quality and very negligible in scope, very much less important
in these Lives than the establishing of chronology and master-
pupil successions, matters which rarely interested Seneca the
Elder. The pseudo-Plutarchan Lives of the ten orators, which may
be regarded as reflecting the more serious side of Hellenistic
biography, serve to show how it was possible for an ancient bio-
grapher to write about an orator's life and works without bearing
The criticism 63
any great resemblance to Seneca the Elder or Suetonius.

The elder Seneca nowhere leads us to suppose that he was
widely read in Greek prose literature, so that any resemblance
between his Barockstil (to use Wilamowitz1 expression)18 and that
of Antigonus of Carystus and others is, if not completely coinci-
dental, most likely the result of his knowledge of Latin imi-
tations of Hellenistic biographical writings. He could very well
have read some works of this type in the course of his study of
Roman history. The problem is to know which Latin authors could
possibly have influenced him.
We happen to know that Seneca was interested in the way that
the pen-portrait had become an increasingly important motif in
historiography:
quotiens magni alicuius <viri> mors ab historicis narrata est,
totiens fere consummatio totius vitae et quasi funebris laudatio
redditur. hoc, semel aut iterum a Thucydide factum, item in
paucissimis personis usurpatum a Sallustio, T. Livius benignus
omnibus magnis viris praestitit; sequentes historici multo id
effusius fecerunt. Ciceroni hoc, ut Graeco verbo utar,
19
Livius reddit. (Suas. 6.21)
The character sketches of Sallust and Livy may well have influ-
enced him to some extent: antithetical presentation of conflicting
traits is certainly to be found in them. In Livyfs 'epitaph' for
Cicero, quoted with admiration by Seneca in Suas, 6.22, we read:
ingenium et operibus et praemiis operum felix, ipse fortunae diu
prosperae; sed in longo tenore felicitatis magnis interim ictus
vulneribus...omnium adversorum nihil ut viro dignum erat tulit
praeter mortem...
SallustTs pithy description of Catiline comes rather nearer to
Seneca's manner than the rotund period from which the Livian
antithesis above has been excerpted:
L. Catilina, nobili genere natus, fuit magna vi et animi et
corporis, sed ingenio malo pravoque... animus audax subdolus
varius, cuius rei lubet simulator ac dissimulator: alieni ad-
petens, sui profusus, ardens in cupiditatibus: satis eloquentiae,
sapientiae parum: vastus animus inmoderata, incredibilia, nimis
alta semper cupiebat. (Cat. 5)
Note here, in addition to the antitheses, the way the portrait is
built up in places by means of nouns and adjectives in the nomina-
tive, devoid of verbs. We have seen something similar in Seneca's
description of Fuscus (Contr. II pr. 1: summa inaequalitas
orationis etc.). There is no frivolity, though, in the pen-

portraits of Sallust and Livy, no jotting down of revealing anec-
dotes and clever witticisms remembered from the time when the men
described were alive. The models, if any, for these features of
Seneca1s technique, must have lain elsewhere, most likely in anec-
dotal records of earlier writers1 lives.
Four Latin authors are listed by St Jerome (De viris ill.
praef.) as eminent predecessors of Suetonius in the writing of
works De viris illustribus, namely Varro, Santra, Nepos and
Hyginus. Varro, who wrote a book De poetis (GRF 2O9ff.) which was
used by Suetonius and may have influenced his biographical tech-
nique, also composed a work called the Imagines (GRF 214f.),2^
comprising descriptions of seven hundred famous men, some of whom
must surely have been orators. Santra, cited by Suetonius as an
authority on poets, certainly wrote something about the history of
rhetoric, for Quintilian (XII.10.16 = Santra fr. 13, GRF 388)
cites an account by him of the rise of Asianism. Two of Nepos1
works on famous Romans, the brief Life of Cato and the longer
Atticus are extant; he also wrote a Life of Cicero (GRF 4O5ff.)
and other works containing biographical material (HRR II 26ff.).
Hyginus wrote a work called De vita rebusque illustrium and
another entitled Exempla (HRR II 72f.).
It is unfortunate that so little survives of Hyginus* bio-
graphical works, for he was a Spaniard, and a great friend of Ovid
(Suet. De gramm. et rhet. 20.2 Brugnoli), and on both these counts
Seneca the Elder might well have taken an interest in his writings.
It appears that Hyginus did occasionally touch on literary topics
(fr. 1, HRR II 72), but the main interests he displays in his
meagre extant fragments - in the property of great men (frs. 2f.),
in the origines of cities (frs. 5ff.) and in the rationalizing of
myths (frs. 15f.) - were not ones shared by the elder Seneca as
far as we know. It must be doubted whether Varro greatly influ-
enced Seneca. The fragments of his works on literary history re-
veal him as a scrupulous user of the more cautious (if sometimes
misguided) methods evolved by Hellenistic scholars for recon-
structing the past. Nothing suggests that he was capable of vivid
metaphorical description of style, or exuberant anecdotal writing,
The criticism 65
and, however fluent a writer he may have been, it seems unlikely

that he could have given nearly as much individual attention to
each of the seven hundred men in his Imagines as Seneca gives to
his rhetoricians. As for Santra, we lack evidence that he took an
interest in the details, as opposed to the broad outlines, of the
history of rhetoric. But it is unsafe to argue from silence.
Nepos, in the Lives of Atticus and Cato which have survived com-
plete, emerges as one of those biographers who keep more or less
closely to the conventional form of classical eulogy: a survey in
chronological order of a man's achievements is followed by a
string of paragraphs singling out his particular virtues. Nothing
could be further from Senecan informality. Literary description
does find some place in these Lives, as it is permissible to eu-
logize a man for being cupidissimus litterarum (Cato 3) or
antiquitatis amator (Atticus 18), but though Nepos displays in
these Lives a certain gift for outlining the contents of books
succinctly, he makes only the most perfunctory and colourless
remarks about style. On the other hand a fragment of another of
Nepos1 works (De inl. viris fr. 17, HRR II 40) gives us praise of
Cicero expressed in terms strikingly close to Seneca's talk of
Cicero's viva vox (Contr. I pr. 11) and the challenge of Roman
oratory to insolent Greece (ibid. 6):
Cornelius Nepos in libro de historicis Latinis de laude Ciceronis:
non ignorare debes, unum hoc genus Latinarum litterarum adhuc non
modo non respondere Graeciae sed omnino rude atque incohatum morte
Ciceronis relictum. ille enim fuit unus, qui potuerit et etiam
debuerit historiam digna voce pronuntiare, quippe qui oratoriam
eloquentiam rudem a maioribus acceptam perpoliverit, philosophiam
ante eum incomptam Latinam sua conformarit oratione.
We are not actually obliged to look to Nepos in order to find an
explanation for the fulsomeness of Seneca's expression when he
praises Cicero: there are sufficient parallels for it in the de-
claimers' treatments of Suas. 6 and 7. But this, as we shall
21
see, is not the only case where we find in a fragment of one of
Nepos' lesser-known works a close analogue for the type of data,
and manner of presenting it, found in those rare passages in
Seneca's anthology where he refers to the rhetoric of the late
Republic. Such works as Nepos' libellus quo distinguit litteratum
ab erudito - - and perhaps his Exempla2-^ were evidently richer in
2 2
detailed linguistic investigation and literary allusion than his

extant Lives of Atticus and Cato. However, there is nothing in
his fragments that seems to bear a striking resemblance to the
pen-portraiture of contemporaries in Seneca's prefaces.
Of course, Varro, Nepos, Santra and Hyginus were not the only
pre-Suetonian authors to write about the life and works of famous
Romans. For instance, Atticus wrote a work called Imagines along
the same lines as VarroTs work of the same name (Nep. Att. 18).
How much biographical speculation and criticism had accumulated
around Republican poets by the time of Seneca the Elder is par-
ticularly well illustrated by the Suetonian Life of Terence in
which we find cited variously as authorities on Terence's life and
as critics of his comedies, not only Nepos, Varro, and Santra, but
Fenestella, Porcius Licinus, Volcacius, C. Memmius, Q. Cosconius,
Afranius, Cicero and Caesar. Prose writers seem to have been less
thoroughly discussed by grammarians than poets. However, Cicero
in his Brutus had ensured that the achievements of his predecessors
in Roman oratory would not wholly be forgotten, and provided the
elder Seneca with precedents for his use of the auyxpLats in Contr.
IV pr. 2ff. and Suas. 2.15. Tiro (GRF 39Off.), as well as Nepos,
wrote a Life of Cicero; he was also the author of Epistulae de
rebus grammaticis rhetoricisque, which included (fr. 6) analysis
of a speech by Cato. Certainly known to Seneca the Elder was a
work by Votienus Montanus in which he made frequent reference to
the eloquence of his friend Marcius Marcellus (Contr. IX.6.18).
This isolated reference to books by Montanus, the existence of
which is nowhere else attested, should warn us of our hopeless
ignorance of the true amount of written literary criticism which
may have been available to Seneca. There may have been many works
written in the early Empire containing a mixture of biographical
and literary information which, however, enjoyed too limited a
circulation for any record of them to be preserved. It should be
noted, for instance, that Suetonius must have had a second source,
in addition to an epitomized version of the elder Seneca's sev-
enth preface, for his Life of Albucius Silus (De gramm. et rhet.
30 Brugnoli).24
There is no evidence, though, that any author wrote character-

The criticism 67
sketches of the rhetoricians of the late Republic resembling those

in Seneca's prefaces. The index of FunaioliTs Grammaticae Romanae
fragmenta, which lists Roman grammatical works, including bio-
graphical writings, from the period up to, and including, the
reign of Augustus,25 records no work entitled De rhetoribus. It
is noteworthy that the Republican rhetoricians, only four in num-
ber, whose lives are sketched by Suetonius (De gramm. et rhet.
26-9 Brugnoli) are not treated in nearly as much detail as
Albucius. So, in the absence of any obvious precursor to Seneca
the Elder in his mode of pen-portraiture, we have to treat him as
an individualist and innovator, which maybe is indeed what in some
respects he was.
Individualists are, of course, the most difficult of all lit-
erary figures to discuss. In the elder Seneca's case the chief
difficulty, as we have seen, lies in the fact that, as he is pri-
marily a writer of 'descriptive' criticism, he does not lend him-
self to comparison with either his main predecessor or his main
successor in the tradition of Roman rhetorical criticism, Cicero
or Quintilian, whose work belongs chiefly to the categories of
'legislative' and 'theoretical' criticism. Even the Brutus, with
its multitude of brief descriptions of Republican orators, is a
vehicle for Cicero's theorizing, notably his polemic against the
Atticists and his search for the ideal orator. No whimsical anec-
dotes or references to the peculiar quirks of individuals, such as
were acceptable in the Senecan or Suetonian pen-portrait, were
allowed, even in this work, to disrupt the smooth classicism which
Cicero thought appropriate for literary theory. The result is
that some modern readers have been disappointed by the lack, in
Cicero's descriptions of orators in the Brutus, of the liveliness
prevalent in Seneca's prefaces.26
Before we can make any useful comparisons or point out con-
trasts between the elder Seneca and the literary theorists, we
therefore have to process his criticism, to reconstruct, as far as
possible, the type of Art of Rhetoric he would have written, had
he chosen to write in that form. Two attempts to isolate theories
about the proper treatment of a declamation implicit in Seneca's
criticism have indeed already been made by the American scholars,
A. F. Sochatoff and L. A. Sussman, but as neither Sochatofffs ob-

servations in his all too brief article, fBasic rhetorical the-
ories of the elder Seneca 1 , 27 nor the relevant chapters in
Sussman1s doctoral dissertation and recent book,28 provide a
wholly satisfactory treatment of the subject, a new and indepen-
dent study along similar lines will not come amiss.2^
There is a danger, though, that if we sort SenecaTs criticism
into a rigid scheme far removed from his own informality, we shall
lose sight of the sort of critic he was. To counteract this
danger some attempt must also be made to use the pen-portraits as
Seneca the Elder intended them to be used, as an aid to the ap-
preciation of the fragments of the major declaimers.3^
I shall not be following, in my study of SenecaTs criticism,
the lines mapped out by H. Bardon in his book, Le vocabulaire de
la critique litteraire chez Seneque le Rheteur,31 and his article,
'Mecanisme et stereotypie dans le style de Seneque le Rheteur',32
because, as no critical reader can fail to notice, his approach is
flawed by a fundamental misconception. Bardon regrets, basically,
that Seneca the Elder was not Cicero. The most important part of
his book consists of a 'Lexique1, in which he lists critical terms
used by the elder Seneca, and notes where these terms were used by
Cicero and Quintilian, and occasionally other critics, and where,
significantly, they were not used by them. This 'Lexique' has
some usefulness as a source of cross-references, but it is marred,
unfortunately, by many serious omissions and errors.33 it i s also
unfortunate that in the chapters which follow, Bardon takes the
line that it is totally deplorable that, after all the hard work
expended by Cicero in building up a Latin critical vocabulary of
considerable richness, Seneca does not use all the terminology put
at his disposal by the great master. BardonTs demonstration that
Seneca's critical vocabulary is very different from Cicero's is
valuable, but his complaints about Seneca's sins of omission are
in many ways unfair.34 First, he does not pause to consider that
Cicero's coinages of critical terms are not all uniformly success-
ful, and that not all were thought worthy of re-use even by his
keenest admirers. G. M. A. Grube has noted Cicero's use in De or.
III.52.199 of three alternative terms to designate a 'type' of
The criticism 69
style, as an example of Cicero's 'impatience with precise tech-

nical terminology1.35 It is easy to see why later critics were
likely to reject two of the three terms, habitus, color, figura,
when writing about style in general terms. Figura was increas-
ingly to take on the very specific technical sense of figure of
speech, figure of thought; color, among the rhetoricians of the
early Empire, has a meaning which is unrelated to stylistics.
Rejection of some Ciceronian terms might mean a certain lessening
of the richness of critical language, but it did not mean any
diminution of clarity. It should be noted that Quintilian some-
times rejects as imprecise pure Latin critical terms, such as
Cicero always tried to use, and prefers to revert to the Greek
word his predecessors had been trying to translate: ...enthymema
(quod nos commentum sane aut commentationem interpretemur, quia
aliter non possumus, Graeco melius usuri)... (Quint. V.10.1). So
Seneca is in good company when he uses latinized forms of Greek
terms.
In his strictures on the elder Seneca, Bardon does not make
due allowance for the fact that Cicero's rhetorical criticism is
far greater in bulk than the purely critical sections of the dec-
lamatory anthology. It is quite unreasonable of him, on this
count alone, to complain that, 't/ne autre faiblesse du vocabulaire
critique chez Seneque consiste dans le petit nombre des termes qui
le constituent par rapport a celui de Cic&ron',36 and to illus-
trate this statement by totting up the number of adjectives Seneca
applies to the word vox, and contrasting the total arrived at
unfavourably with the number of epithets applied by Cicero to the
same word. But even if Seneca's criticism were equal in bulk to
Cicero's, it would still not be fair to expect him to use an
identical and equally large selection of critical terms. It is
not just that the language of criticism was bound to evolve with
the passing of time, and that it would not be appropriate to write
about school rhetoric in exactly the same terms as were applied to
true oratory. The fact is simply that descriptive criticism
requires a different vocabulary from - in particular far fewer
strictly technical terms than - legislative or theoretical
criticism.
The last four hundred years of English criticism, it has been

observed, have been marked by the gradual ousting of the latter
two types of criticism, for example, books of rhetoric offering
instruction to the writer, and essays on literary aesthetics, by
descriptive criticism, 'the youngest1 in English 'of the three
forms, by far the most voluminous, and the only one which today
possesses any life and vigour of its own 1 . 3 7 Parallel with this
trend, it is noticeable that 'the terminology of criticism dimin-
ishes steeply1.38 This decline is connected with the increasing
horror with which formal rhetoric has come to be regarded by
English writers, a phenomenon which was not paralleled in the
Roman world, where the scholastics saw to it that rhetoric sur-
vived, whatever else might decline, and the situation was never
reached where a man sufficiently well educated to write descrip-
tive criticism would do so completely without reference to the
standard body of rhetorical terminology. But there is something
inherent in the nature of literary description which makes it un-
likely that a descriptive critic would use the whole range of
terms found in school text-books and theoretical essays, and this
applies to the ancient world as much as to the modern. The de-
scriptive critic may want to allude to someone's fondness for a
particular rhetorical figure: it does not follow that he will feel
called upon to mention all the figurae listed in the standard
handbooks, however well-acquainted with their definitions he may
have been. He may want to find an adjective to describe the voice
of a particular man whose delivery contributed significantly to
his success or failure: it does not follow that he will want to
survey exhaustively all the possible types of vociferation. It is
unreasonable of Bardon, then, to complain of Seneca's vocabulary
that, fil existe des mots du vocabulaire critique qui ne sont
absolument pas representes'.39
Nor would the ancient descriptive critic, if he had any

imagination at all, restrict himself to the vocabulary provided
for him by writers of Arts of Rhetoric. It would not, for
example, be a sign of the best critic if, when assessing the
essential qualities of an individual's style, he kept rigidly to
the standard epithets attached by theorists to the genera dicendi.
The criticism 71
The more gifted critic will draw much of his vocabulary from non-
scholastic sources, applying to the written word images of light
and shade, or density and fluidity, and transferring to criticism
nouns and adjectives originally applied to extra-literary pursuits,
the martial arts, for example, or crafts involving skilful con-
structive processes, building, carpentry, weaving and so on. This
is what the elder Seneca does, not, of course, without Ciceronian
precedent,40 but choosing a new selection of words in which the
love of the striking and colourful, characteristic of Silver
Latin, is apparent. Bardon, who seems to have felt he must ex-
clude from his Tlexique' any words which were not strictly tech-
nical terms or metaphors well established in criticism by Cicero,
does the richness of Senecafs language much less than justice.
Take, for example, a remarkable sentence from his description of
Fabianus: deerat illi oratorium robur et ille pugnatorius mucro,
splendor vero velut voluntarius non elaboratae orationi aderat
(Contr. II pr. 2 ) . Bardon includes in his list oratorius,
splendor, elaboratus, and of course, oratio, but not robur,
pugnatorius, mucro, or voluntarius,
Bardon1s selection of words is unnecessarily perverse, but
there is a real difficulty in distinguishing, in SenecaTs writing,
between truly critical vocabulary and words of biographical de-
scription. Such a distinction is of doubtful validity anyway, in
view of the ancient saying, talis hominibus...oratio, qualis vita
(Sen. Ep. mor. 114.2). For this reason, no select list of his
critical terminology is likely to be wholly satisfactory. Once
more one feels the need for a complete index verborum to the elder
Seneca^ work.
The elder Seneca was not a great stylist. Schott41 was being
over-indulgent towards him when he wrote, de cuius scriptoris stylo
ita iudicare non dubitem, nihil esse in lingua Latina cum a
Cicerone Fabioque discesseris scriptum purius aut elegantius. He
belongs to a period of stylistic transition: he seems to have been
feeling his way towards that finesse of pointed expression for
which his son was to be famous, but he is not uniformly successful
in achieving it. One has to agree with Bardon that sometimes his
striving after effect seems na'ive, as for example, when he says of
the child prodigy Alfius Flavus, cum praetextatus esset, tantae

opinionis fuit, ut populo Romano puer eloquentia notus esset
(Contr. 1.1.22), and when he records how a certain Dionysius,
elegans magis declamator quam vehemens hunc sensum et vehementer
dixit et eleganter (Contr. I.4.II). 42 Sometimes also, as we have
seen, he obscures his exact standpoint on moral issues by writing
too much declamatory verbiage about them.
Yet one could fairly apply to his prose-writing the assess-
ment with which he concludes a pen-portrait of one of the de-
claimers: redimebat tamen vitia virtutibus et plus habebat quod
laudares quam cui ignosceres... (Contr. IV pr. 11). Mixing of
metaphors, a feature of style carried to excess by some of his
contemporaries, for example Valerius Maximus, occurs in his
writing too: his references in quick succession to oratorical oak-
wood and a combative dagger in Contr. II pr. 2 provide a good
example. But whereas in the hackneyed moralizing of Valerius
Maximus in such a passage as VI.9 ext. 7,
caduca nimium et fragilia, puerilibusque consentanea crepundiis
sunt ista quae vires atque opes humanae vocantur. affluunt
subito, repente dilabuntur: nullo in loco, nulla in persona
stabilibus nixa radicibus consistunt; sed incertissimo flatu
fortunae hue atque illuc acta, quos sublime extulerunt, improviso
recussu destitutos profundo cladium miserabiliter immergunt,
the effect of muddling together allusions to childrenfs toys,
rising tides, firm roots, blasts of fortune and so on, is decid-
edly tasteless, in the elder Seneca^ underivative type of liter-
ary criticism a certain unconventionality in the deployment of
imagery seems a positive virtue. The younger Seneca in Ep. mor.
100, where he echoes a number of his fatherfs criticisms of
Fabianus,^^ may show greater refinement of taste than his father
when he paraphrases the words, deerat...mucro (Contr. II pr. 2 ) ,
thus: deest illis oratorius vigor stimulique quos quaeris et
subiti ictus sententiarum (§8), but this version is surely much
less memorable than the original.
Nor is the elder Seneca1s colourfulness in the language of
his literary descriptions the only feature of his writing which
one would not necessarily want to sacrifice in favour of the
greater smoothness and refinement cultivated by some other classi-
The criticism 73
cal writers of criticism and belles lettres. One also senses a

great charm and obvious warmth pervading his writing as he remi-
nisces about his friends among the declaimers. In particular one
notices that he has an admirable capacity to sympathize with them
in the misfortunes which they brought upon themselves by an
approach to life which he recognized to be dangerously unbalanced.
His heart goes out to Latro, for instance, whose unremitting in-
tellectual exertions caused him, in addition to physical suffering
(Contr. I pr. 17), another type of exhaustion, ingenii lassitud-
inem, quae non minor est quam corporis sed occultior (ibid, 15),
and to Albucius, who after the miscalculated use of a declamatory
device in a law-suit had lost him his case, felt obliged to retire
permanently from the courts, being, as Seneca explains, homo
summae probitatis, qui nee facere iniuriam nee pati sciret (Contr.
VII pr. 7). Besides, by such anecdotal digressions as the one
about the megalomaniac Seneca Grandio (Suas. 2.17), later to be
paraphrased so enjoyably in Abraham Cowley's essay On greatness,
and by such wry comments as his quasi-Ovidian strictures on dry
stylists in Contr, II.1.24, he earns a high rank among the wits of
the ancient world.
PART II
SENECA THE ELDER ON THE HISTORY OF ELOQUENCE
1 ORATORY AND RHETORICAL THEORY UP TO HIS OWN TIME
The extent of the elder Seneca's familiarity with the classics of

Greek and Roman oratory and rhetorical theory cannot be estimated
with any hope of accuracy but if, as seems probable, he had not
nearly the same depth of knowledge as Cicero of the rhetorical
traditions of Greece and the Roman Republic, this is nothing to be
wondered at in view of the contrast between the kinds of education
which the two men received in their youth. Seneca the Elder, con-
fined in his formative years to a remote and war-torn western
province, and unfortunate in having to learn the elements of rhet-
oric from the uninspired Marullus, had none of the educational
experiences which were to enable Cicero to write about oratory
with his air of incontrovertible authority. Not for him, unless
our information is grossly misleading, anything comparable to the
old-fashioned tirocinium fori, the prolonged study of Greek philo-
sophy, the travels around the schools of distinguished rhetor-
icians in Greece and Asia Minor, which Cicero describes in the
autobiographical section of the Brutus (89.3O4ff.).
Nevertheless Seneca the Elder was able to overcome to a cer-
tain extent the limitations of his formal education. As an adult
enthusiast in the auditoria of the declaimers he had the chance to
learn much, just by listening to the discussions and quips of the
rhetoricians and distinguished members of their audience, which
would sharpen his critical sensitivity. His study of history must
have further broadened his outlook. On the other hand it is
questionable whether he ever made any considerable study of the
works which we consider the classics of rhetorical theory, or of
the speeches of Greek, or pre-Ciceronian Roman orators. Of
course, we must beware of relying too much on an argument from
silence: the fact that Seneca does not mention a book does not
mean that he had never read it; his subject-matter does not call
for constant reference to the more venerable orators and
Seneca the Elder on the history of eloquence 78
theorists. But references to rhetoric before his own time are

strikingly rare in his criticism, and it seems indicative of a
decided modernity in his outlook that when he wants to give extra
authority to a critical pronouncement he tends to refer to what
Latro said on the matter, or Pollio, or Messala - men he had actu-
ally heard airing their views - never, for example, to what Cicero
said in the De oratore. The frame of reference of Seneca's criti-
cism is for the most part his own life-time. On the few occasions
when he looks back beyond his personal experience one needs always
to ask whether he could not have been relying on hear-say and
secondary sources for his information.
Consider first his references to Greek rhetoric outside the
schools of declamation. Seneca the Elder was by no means ignorant
of Greek, for he was clearly able to listen to declamations by
contemporary Greeks with discrimination. It is true that his quo-
tations from their work are much briefer than those from Latin
declamations, but it cannot be argued that his familiarity with
their declamations could have been confined to an acquaintance
with sententiae which had been singled out for praise or condem-
nation by their Latin-speaking contemporaries, as in Suas. 3.6 he
explicitly refers to a visit which he and Gallio had made to
Nicetes' school: memini una nos ab auditione Nicetis ad Messalam
venisse. It is all the more remarkable that his allusions to the
great days of Attic oratory are so few, and so little suggestive
of first-hand knowledge of the authors concerned.
In Contr. Ill pr. 8 he quotes an unappreciative criticism of
Plato's Apology by Cassius Severus: eloquentissimi viri Platonis
oratio, quae pro Socrate scripta est, nee patrono nee reo digna
est. Severus is arguing that Plato was among those who could only
be successful in one genre; he was not as good an orator as a
philosopher. Of the Attic orators proper only Demosthenes and
Aeschines receive mention. Demosthenes was famous enough as a
historical figure to have his life-story pillaged for exempla even
by Latin rhetoricians. Musa, for instance, in Contr, VII.3.4 re-
1
calls Demosthenes suicide: 'habuit' inquit 'Demosthenes venenum
et bibit.' idem ego tibi pater quod Demostheni Philippus?
Seneca, then, may be displaying no more than the general knowledge
Oratory and rhetorical theory up to his own time 79
any Roman might have when he objects to Nicetes' anachronistic use

of an imitation of Demosthenes1 Marathon oath in a suasoria about
Xerxes (Suas. 2.14). Another reference, however, the startling
implications of which will be discussed later,* concerns Demos-
thenes' style. We are told of Calvus: compositio quoque eius in
actionibus ad exemplum Demosthenis viget:^ nihil in ilia placidum,
nihil lene est, omnia excitata et fluctuantia (Contr. VII.4.8).
This is a striking critical remark, but whether it implies first-
hand familiarity with Demosthenes' speeches is questionable.
Demosthenes' vehemence must have been well known, and, as we shall
see,3 there is some reason to believe that in Contr. VII.4.8
Seneca is recalling remarks he had heard Latro make. Seneca's one
reference to Aeschines is an insignificant one. At pains to dis-
tinguish the famous orator from a minor rhetorician called
Aeschines active in his own day, Seneca says in passing that dec-
lamation was not known in the great man's time: Aeschines, non
ille orator - tune enim non declamandi studium erat -, sed hie ex
declamatoribus novis dixit... (Contr. 1.8.16). Evidently he did
not know of the tradition that the origins of modern declamation
belonged to the time of Aeschines (Quint. II.4.41: circa Demetrium
Phalerea) , or even in his school (Philostr. Vit. soph. I.481).1*
These are the elder Seneca's only references to Attic ora-
tory, unless one counts the sentence from ps.-Demosthenes, In ep.
Phil. 13 (based on 01. 11.20), which he quotes with only slight
deviation from the received text, but, misled by Arellius Fuscus,
wrongly attributes to Thucydides (Contr. IX.1.13). There is ab-
solutely no mention of Antiphon, Andocides, Lysias, Isocrates,
Isaeus, Lycurgus, Hyperides, or Dinarchus. Greek rhetorical
theorists before his own time are similarly neglected: we find no
allusion to the fifth-century sophists, nothing, if we except
Cassius Severus' opinion of the Apology (Contr. Ill pr. 8) and a
garbled5 reference by Cestius to Aristotle's cousin Callisthenes
in Suas. 1.5, about Plato or Aristotle, no mention by name of
Theophrastus, or Hermagoras, the second-century theorist whose
views on the scope of the orator's work, and methods of analysing
forensic argument, exercised so much influence on Cicero and
Quintilian.6 Even granted that as a writer of descriptive
criticism, Seneca had no compelling need to make constant refer-

ence to works of rhetorical theory, the number of famous names
conspicuous by their absence is remarkable.
His neglect of contemporary Greek rhetorical theorists is not
so complete, as he does make just a few references to the dogmatic
disagreements between the followers of Apollodorus of Pergamum and
Theodorus of Gadara (neither of whom he cites as a declaimer).7
We learn from him about one of the major issues over which Apollo-
dorus and Theodorus differed: whether it was invariably necessary
to narrate the facts of a case.8 The orator Vallius Syriacus,
criticized during a trial by his opponent Bruttedius Niger for
omitting to account for the movements of the accused, is reported
to have explained: primum non apud eundem praeceptorem studuimus:
tu Apollodorum habuisti, cui semper narrari placet, ego Theodorum,
cui non semper (Contr. II.1.36). Seneca notes the debilitating
effect that a rigid adherence to Apollodorean dogma could have on
an orator: sed Turrinus pater multum viribus dempserat, dum Apollo-
dorum sequitur ac summam legem dicendi sectam putat (Contr. X pr.
15). He also reports minutiae of the Apollodoreans' theory: their
views on ambiguous controversia themes, of which Contr, 1.2 is an
example: aiebat [sc. Latro"] Apollodoreis quidem fixa esse themata
et tuta, sed hie non repugnare controversiam huic suspicioni; non
enim ponitur adhuc virginem <esse> et multa sunt propter quae
credibile sit non esse (Contr. 1.2.14);^ and on the treatment of
themes where the question arose whether a beneficium had been
received: Dionysius Atticus, a pupil of Apollodorus, agreed with
Gallio that if it were at all possible to get an adversary to con-
cede that no benefit had been received, we should try to achieve
this, but that otherwise we should pass the matter over quasi
donemus et possimus quidem facere controversiam, sed nolimus
(Contr. II.5.11). Dionysius also pointed out that one could argue
in some cases that, though a benefit had been received, it was an
insignificant one (ibid.), and another Apollodorean, Moschus, pro-
duced a further quibble as to whether one could speak of beneficia
received by a husband from his wife and vice versa: non est bene-
ficium, sed officium facere quod debeas (Contr. II.5.13).
On the Theodoreans Seneca has less to say, merely noting, in

addition to their views on the dispensability of narratio, the

fact that Tiberius was among their number: Tiberius ipse Theo-
doreus offendebatur Nicetis ingenio... (Suas. 3.7). He does not
explain what connection there was, if any, between Theodorean
doctrine and Tiberius1 prejudice against the ecstatic rhetoric of
Nicetes.10
Seneca does not commit himself to allegiance to either
school. Though he shows a certain lack of sympathy for the Apollo-
doreans in Contr. X pr. 15, it does not follow that he was a Theo-
dorean. He seems to have been tolerant, like the Theodorei, of
the omission of narratio (Contr. 1.2.19, cf. IX.2.18) in certain
circumstances, but does not, for example, adopt the idiosyncratic
terminology they used for the analysis of argumentation: Seneca
uses the term quaestiones, following in this the usage of
Hermagoras et Apollodorus et alii plurimi scriptores (Quint. III.
11.3), where a Theodorean would have spoken of capita generalia
and capita specialia (ibid.).
Among SenecaTs references to the two schools there is nothing
that need imply that he had read any critical work by Apollodorus
or Theodorus. He owes his information about the Apollodoreans1
insistence on water-tight controversia themes to Latro. The
reference to Tiberius1 connection with the Theodoreans comes from
a long report of gossip about Nicetes. It is never to Apollodorus
and Theodorus themselves that he refers, but to their pupils, men
whose opinions it was quite possible to learn about without con-
sulting books, just by listening to lectures, the gossip of the
declamation halls, and repartee in the law-courts.
Similarly, the sprinkling of Greek critical terms to be found
in his books could have been gleaned here and there from the lec-
tures and gossip of the schools. There is no reason to suppose
that Seneca owes a particular debt to any single Greek theorist
for his terminology, which includes, as is only to be expected,
words from every stratum of the language, from old-established
terms like tragoedia and historia to others which were much more
recent coinages (though, given the gaps in our evidence, we cannot
know quite how recent) like idiotismus and metaphrasis.^^ Some
are loan-words well entrenched in Latin, e.g. comicus,*2 others,
like ethicos and problemata philosophumena appear monstrous in the

Latin garb in which they appear in our texts. (But perhaps it was
a copyist, rather than Seneca himself, who was responsible for the
transliteration in such cases.) ^ No consistent criterion**4 can
be seen to determine the preference for Greek terms over Latin by
Seneca and the critics quoted by him: there seems no particular
reason why they should so often use the terms prooemium and
epilogus when there existed well-established pure Latin terms for
all the standard sections of a speech; again, Seneca seems just as
happy using figrura15 as its Greek equivalent schema, and it seems
legitimate to regard insania and corrupte as synonymous in his
criticism with cacozelia and cacozelos. ^ So, although some of
his Greek terms are words for which no adequate, all-embracing and
unambiguous Latin equivalent had been coined, e.g. tragoedia, no
rational process of selection called for the use of all of them.
Now, irrational selection is what is to be expected if Seneca's
grecisms are derived from random reminiscences of a jargon orig-
inally used in impromptu, unwritten criticism. That this is their
origin becomes quite clear if we observe that in his prefaces, the
most formal and carefully composed parts of his work, Greek terms
are relatively less frequent than in the surveys of declamations,1'7
where he is less concerned with smooth artistic finish, and if we
note how in a number of passages Seneca quotes other people using
Greek terms which he elsewhere adopts for his own use, for example:
Contr, VII pr. 8: clamabat Albucius: non detuli condicionem;
schema dixi9 and Contr. X.4.25: P. Vinicius...hunc aiebat sensum
disertissime apud Nasonem Ovidium esse positum...occiso Achille
hoc epiphonema poni... It is not surprising to find a free use of
Greek terms in the conversation of the Latin critics of Seneca's
time, for we may deduce from the frequent appearance of Greek ex-
pressions in the letters of Cicero and his friends where they are
discussing literature, that conversations about rhetoric and
poetry in the late Republic would also have been full of them.
It seems quite possible, then, that the whole of the elder
Seneca's knowledge of Greek oratory and rhetorical theory could
have been picked up at second hand from his friends and acquaint-
ances in and around the declamation halls. Significantly, he
said of Latro, whose formal education had been identical with his
own: Graecos...et contemnebat et ignorabat (Contr. X.4.21) - this
with reference to contemporary Greek rhetoricians. It would not
be fair to say that the same was true of Seneca, who was at least
a frequenter of Greek declamations, but this statement is indica-
tive of the extent to which these two men, who were to exercise,
the one through his offspring, the other through his pupils, a
considerable influence on the development of Silver Latin, were
cut off from the main stream of Greco-Roman literary tradition by
an uninspired and perhaps wholly provincial*® education.
There is a remarkable scarcity in SenecaTs criticism of al-
lusions to Republican Roman orators other than Cicero. The only
orator before Cicero's generation to receive mention is the elder
Cato, who is invoked in grand style in Contr. I pr. 9:
erratis, optimi iuvenes, nisi illam vocem non M. Catonis, sed
oraculi creditis. quid enim est oraculum? nempe voluntas divina
hominis ore enuntiata; et quem tandem antistitem sanctiorem sibi
invenire divinitas potuit quam M. Catonem, per quem humano generi
non praeciperet, sed convicium faceret? ille ergo vir quid ait?
f
orator est, Marce fili, vir bonus dicendi peritus.1
The quotation is no proof that Seneca had read CatoTs works. Not
only was Cato occasionally the subject of declamatory exempla^^
but the particular dictum cited is one to be found quoted in hand-
books of rhetoric after SenecaTs time, and may well have been
excerpted for scholastic use earlier. It is given great promi-
nence by Quintilian, who in his final book bases his whole dis-
cussion of the qualities of the orator on Catofs definition (XII.
l.lff•) and by Fortunatianus, who quotes it right at the beginning
of his rhetorical catechism: Quid est rhetorical bene dicendi
scientia. quid est orator? vir bonus dicendi peritus (RLM
81.4f.).
Seneca never alludes at all in his criticism to any of the
other oratorical worthies of the second century B.C., praised in
Cicero's Brutus. It may be suspected that his view of early Roman
oratory was similar to that of Velleius Paterculus (1.17.3):
at oratio ac vis forensis perfectumque prosae eloquentiae decus,
ut idem separetur Cato (pace P. Crassi Scipionisque et Laelii et
Gracchorum et Fanii et Servii Galbae dixerim) ita universa sub
principe operis sui erupit Tullio, ut delectari ante eum
paucissimis, mirari vero neminem possis nisi aut ab illo visum aut
qui ilium viderit.
Certainly his conception of literary history resembles that
of Velleius in other respects: remember his desire to be numbered
amongst those who could have seen Cicero (Contr. I pr. 11), and
his contention that the only Roman oratory to rival the Greeks
circa Ciceronem effloruit, and that omnia ingenia, quae lucem
studiis nostris attulerunt, tune nata sunt (Contr. I pr. 6-7). It
seems that Seneca wished to glorify the first generation of liter-
ary men he had known - misguidedly, one may think, in the case of
the rhetoricians whom he must have had primarily in mind - by
associating them with the Ciceronian age. About the orators who
might truly be described as circa Ciceronem he has little to say,
and, though his outrageous failure to qualify the statement in
Contr. I pr. 11, omnes autem magni in eloquentia nominis excepto
Cicerone videor audisse, may be blamed on a momentary lapse of
concentration combined with a taste for pithy expression, it will
not come as a surprise if we find no evidence that Seneca had any
very profound or extensive familiarity with late Republican
oratory.
Seneca professes to be a great admirer of Cicero, but then,
it was normal in the schools of rhetoric to revere Cicero uncriti-
cally. No matter how violent a reaction the declaimers* own rhet-
oric represented against the Ciceronian ideal, they were prepared
to declaim with enormous fervour on themes based on (more or less
apocryphal) stories about the great orator's life and death:
Contr. VII.2: De moribus sit actio.
Popillium parricidii reum Cicero defendit; absolutus est. pro-
scriptum Ciceronem ab Antonio missus occidit Popillius et caput
eius ad Antonium rettulit. accusatur de moribus.
Suas. 6: Deliberat Cicero, an Antonium deprecetur.
Suas. 7: Deliberat Cicero, an scripta sua conburat promittente
Antonio incolumitatem, si fecisset.
The death of Cicero may seem a surprisingly dangerous topic for
the declaimers of the early Empire to deal with. Perhaps part of
its charm lay in the danger. The rhetoricians were aware which
were the most sensitive issues, and generally avoided mention of
them. The extracts given by Seneca from Contr. VII.2 include
none where the declaimer explicitly mentions that Cicero had once
been hailed Pater Patriae, despite the fact that the theme of this
controversia almost cries out for sententiae complaining that
Popillius, having killed his own father, had gone on to murder the
Father of his Country.2^ The declaimers were normally careful,
when speaking of the proscription in which Cicero was doomed, to
lay the blame entirely on Antony. Seneca remarks on Albucius1
treatment of Suas, 6, ...solus ex declamatoribus temptavit dicere
non unum illi esse Antonium infestum (Suas. 6.9).
Cicero was undoubtedly read, if not always with any degree of
concentration, by the average student in the schools of the early
Empire. His diction already seemed old-fashioned: Seneca finds it
remarkable that the declaimer Haterius ...quaedam antiqua et a
Cicerone dicta, a ceteris deinde deserta dicebat... (Contr. IV pr.
9). Yet his reputation remained unassailable. The egotistical
Cestius might rate his own eloquence above Cicero1s, and win his
pupils over to his point of view, but public opinion was against
him. Cassius Severus complains:
pueri fere aut iuvenes scholas frequentant; hi non tantum
disertissimis viris, quos paulo ante rettuli, Cestium suum prae-
ferunt sed etiam Ciceroni praeferrent, nisi lapides timerent.
quo tamen uno modo possunt praeferunt; huius enim declamationes
ediscunt; illius orationes non legunt nisi eas quibus Cestius
rescripsit. (Contr. Ill pr. 15)
But it was not just the fear of a stoning that kept the scholas-
tics from expressing open contempt for Cicero. The need to praise
him uncritically in declamations perpetuated an unnaturally high
regard for him through a period when in reality all his ideals in
oratory and politics were being rejected. The declaimers inter-
larded their sententiae in Contr. VII.2, Suas. 6 and Suas. 7 with
allusions to Cicero's speeches, sometimes direct quotations.21
Their favourite speeches seem to have been the last of the
Verrines,22 the Catilinarians,23 the Pro Milone, to which Cestius
had done the honour of providing an answer,21* and above all the
second Philippic, the speech which Juvenal was later to single out
as the prime example of Ciceronian eloquence: te, conspicuae divina
Philippica famae \ volveris a prima quae proxima (10.125f.).25
Thus the declaimers of the elder Seneca's time anticipated in
their choice of Ciceronian allusions the view of one of the inter-

locutors in Tacitus1 Dialogus:
nee Ciceronem magnum oratorem P. Quinctius defensus aut Licinius
Archias faciunt: Catilina et Milo et Verres et Antonius hanc illi
famam circumdederunt... (Dial. 37.6)
To the rhetoricians these speeches were a quarry of resonant ex-
pressions of righteous indignation: tuis verbis, Cicero, utendum
est: 'o tempora, o mores'.' , 2 6 and of dicta about the proper atti-
tude towards death: mors nee immatura consulari nee misera
sapienti;^7 mori enim naturae finis est non poena,2-® which one
could imagine Cicero recalling as he faced the prospect of ex-
ecution. They also provided lurid invectives against tyrannical
opponents which the declaimers might re-use when condemning
Popillius1 evil deed: ut uno ictu pereat tantum dabo: pro Cicerone
sic liceat pacisci? (Sepullius Bassus in Contr. VII.2.1 cf. Verr.
II.5.118), or recall when dissuading Cicero from submission to
Antony:
videbis ardentes crudelitate simul ac superbia oculos; videbis
ilium non hominis, sed belli civilis vultum; videbis illas fauces,
per quas bona Cn. Pompei transierunt, ilia latera, illam totius
corporis gladiatoriam firmitatem; videbis ilium pro tribunali
locum, quern modo magister equitum, cui ructare turpe erat, vomitu
foedaverat: supplex accadens genibus deprecaberis? eo ore, cui se
debet salus publica, humilia in adulationem verba summittes?
(Latro in Suas. 6.3; cf. Phil. II.63-4; Verr. II.5.161)
Sometimes, but not often, less famous speeches of Cicero are al-
luded to. Cestius quotes from the Pro Roscio Amerino: quid tarn
commune quam spiritus vivis, terra mortuis, mare fluctuantibus,
litus eiectis? (Contr. VII.2.3 = Rose. Amer. 72 - a somewhat no-
torious piece of Cicero's early oratory, whose faults he recog-
nizes himself in Or. 30.107). Another sententia of Cestius1 could
have been suggested by the first Philippic, or the Pro Marcello
(Suas. 6.4). 2 9 Varius Geminus in an unusual treatment of Suas. 6
gave Cicero what Cassius Severus approvingly called vivum
consilium, pointing out that innumerable people in the provinces
were indebted to Cicero and would shelter him if he fled from
Italy:
Siciliam dixit vindicatam esse ab illo, Ciliciam a proconsule
egregie administratam, familiares studiis eius et Achaiam et
Asiam, Deiotari regnum obligatum beneficiis, Aegyptum et habere
beneficii memoriam et agere perfidiae paenitentiam. sed maxime
ilium in Asiam et in Macedonian! hortatus est in Cassi et in Bruti

castra. (Suas. 6.11)
Here is an allusion to the Pro rege Deiotaro, as well as one to
his prosecution of Verres. Varius Geminus also declaimed the
altera pars of this suasoria, recommending that Cicero should
yield to Antony. He pointed out, among other things, how enemies
often are persuaded to give up their hostility, instancing how
Cicero himself had been prevailed upon to defend Vatinius, whom he
had attacked in an earlier speech (Suas. 6.13 W ) . 3 ^ Varius
Geminus1 imaginative grasp of the details of Cicero's career seems
to have been exceptional: ..,Cassius Severus aiebat alios
declamasse, Varium Geminum vivum consilium dedisse (Suas. 6.11),
but presumably Latro and others would at least have been able to
reel off Cicero's achievements cursu (see Contr. I pr. 18).
Were the declaimers familiar with any writings of Cicero
other than speeches? Some at least of Cicero's correspondence was
certainly accessible in the early Empire: in Suas. 1.5, perhaps
quoting from a lecture by Cestius, Seneca gives an explicit refer-
ence to a letter from Cassius to Cicero (Ad fam. XV.19.4). The
quotation, though not exact, is close to the original.3* But none
of the other possible verbal echoes of the letters found in the
declaimers' utterances is so close as to be unquestionably the
product of actual reading of the originals. History books and
school tradition could have transmitted all that the average de-
claimer knew of them. For example, though it is interesting to
note the parallel between Haterius' sententia in Suas. 6.1, in hac
causa etiam Ciceronem verba deficient, and expressions in Ad fam.
II.11.1, putaresne umquam accidere posse, ut mihi verba dessent...?
and XIII.63.1, non putavi fieri posse, ut mihi verba dessent, the
verbal similarity is not close enough to prove that Haterius had
read the Ad familiares, though he may have done.32 Nor can his
use of the words iam nostra peracta sunt (Suas. 6.1) be regarded
as firm evidence that he knew the letter to Brutus in which Cicero
quotes from Plautus' Trinummus: mihi quidem aetas acta ferme est:
tua istuc refert maxime (Trin. 319 = Cic. Ad Brut. 8.2 Watt).
Similarly, where the declaimers refer to topics to be found dis-
cussed in Cicero's philosophical and rhetorical treatises, the
inspiration for their words need not have come directly, or even
indirectly, from Cicero. Fuscus1 resounding words in Suas. 6.5,
non te ignobilis tumulus abscondet etc., could have been inspired
by any number of sources (given that Fuscus, if not a Greek him-
self, was much inclined to imitate Greek originals)33 besides
Cicerofs De senectute 82, with which they have been compared.34
Declamatory references to the oratory of the Republic may or may
not imply knowledge of the Brutus, Varius Geminus could have
learnt of Cicerofs studies in Achaia and Asia from this source,
but his reference to them (Suas. 6.11) is of too general a kind
for us to be sure. Capito may have derived the knowledge he dis-
plays in Contr. VII.2.6 about Pompey's relations with Hortensius
from the reference to the speech Pro Cn. Pompei bonis in Brut.
64.230, but it would be unnatural to suppose that Cicero was the
only authority to mention this presumably quite important speech.
How familiar Seneca himself was with Cicero's works can
hardly be determined, seeing that his subject-matter was not such
as to require constant reference to them, but it is important to
notice that the majority of his Ciceronian allusions could very
well have been picked up from secondary sources, oral and written.
Some of what he says about Cicero has a highly rhetorical
flavour untypical of his criticism. Probably he was influenced by
the declaimers1 eulogistic treatment of Cicero35 when he wrote in
Contr. I pr. 11:
illudque ingenium quod solum populus Romanus par imperio suo
habuit, cognoscere et, quod vulgo aliquando dici solet, sed in
illo proprie debet, potui vivam vocem audire.
As we have seen,36 in the passage where he pits the Roman elo-
quence of Cicero's time against insolent Greece (Contr. I pr. 6)
his expression is closely paralleled by a sententia by Cestius in
Suas. 7.10.
Other references are more or less certainly reports of what
other critics had said about Cicero. In Contr. VII.3.8-9 he ac-
knowledges his debt to Cassius Severus for an account of the his-
tory of Roman humour, in which Cicero figures prominently.
Another important Ciceronian allusion comes in the course of a
passage throughout which it may be suspected that Seneca is
quoting Cestius (Suas. 1.5-8). 37 This is the passage discussed

earlier as an example of Senecafs eccentric mixing of direct and
indirect speech. If we set this allusion, an explicit reference
to a letter from Cassius to Cicero, in its context, it will be
apparent that we cannot use it as firm evidence that the elder
Seneca had read any of Cicerofs correspondence. Cestius is talk-
ing in this passage about the way one ought to address a monarch
like Alexander:
...nihil dicendum aiebat nisi cum summa veneratione regis, ne
accideret idem, quod praeceptori eius, amitino Aristotelis,
accidit, quern occidit propter intempestive liberos sales; nam cum
se deum vellet videri et vulneratus esset, viso sanguine philo-
sophus mirari se dixerat, quod non esset L-X^P. O£6S itep Te peet
yaN&peaau deouaov. ille se ab hac urbanitate lancea vindicavit.
eleganter in C. Cassi epistula quadam ad M. Ciceronem missa
positum: multum iocatur de stultitia Cn. Pompei adulescentis, qui
in Hispania contraxit exercitum et ad Mundam acie victus est;
deinde ait: Tnos quidem ilium deridemus, sed timeo, ne ille nos
gladio dvTLyuxTripLari. ' in omnibus regibus haec urbanitas
extimescenda est. aiebat itaque apud Alexandrum esse <sic>
dicendam sententiam, ut multa adulatione animus eius per-
mulceretur... (Suas. 1.5f.)
After eleganter...positum we should perhaps understand esse aiebat
rather than est.38
Worth considering as a group are three passages concerned
with rhetorical schools and amateur declamation in the late
Republic: Contr. I pr. 11; ibid. 12 and II pr. 5. These passages
have two things in common besides similarity of subject-matter:
they contain detailed information which could only have been
obtained from the writings of Cicero or sources contemporary with
him, and they are closely paralleled by passages in Suetonius* De
grammaticis et rhetoribus. Contr. I pr. 12, the main Senecan
passage to deal with the history of Roman declamation before his
own time, will be discussed at length in relation to other ancient
evidence, including that of Suetonius, in the next chapter.^ The
Suetonian parallel passage is De gramm. et rhet. 25 .8f. Brugnoli.
The reference to Cicero as a declaimer in Contr. 1 pr. 11, aliogui
in illo atriolo in guo duos grandes praetextatos ait secum
declamasse potui adesse... has a counterpart in De gramm. et rhet.
25.3:
Cicero ad praeturam usque etiam Graece declamitavit, Latine vero
senior et quidem cum consulibus Hirtio et Pansa quos discipulos et

grandes praetextatos vocabat.
The information recorded here seems of a type likely to have been
derived from Cicerofs correspondence. Such was certainly the
ultimate source of the information given in Contr. II pr. 5,
primus omnium Latinus rhetor Romae fuit puero Cicerone Plotius, as
we see from the parallel passage in De gramm. et rhet. 26.1:
L. Plotius Gallus. de hoc Cicero in epistula ad M. Titinium sic
refert: equidem memoria teneo pueris nobis primum Latine docere
coepisse Plotium quendam. ad quem cum fieret concursus quod
studiosissimus quisque apud eum exerceretur, dolebam mihi idem non
licere. continebar autem doctissimorum hominum auctoritate qui
existimabat Graecis exercitationibus ali melius ingenia posse.
Sometimes in this group of related passages Seneca gives us
details which are not in Suetonius, for example the reference to
Cicero's atriolum in Contr, I pr. 11, and in the history of decla-
mation in §12, among other things, a quotation illustrative of
Calvus1 use of the term declamare. On other points, however,
Suetonius provides us with fuller information, which he could not
have derived from Seneca. It looks very much as though the two
authors were using data derived from a common source, presumably a
grammatical work dealing with the history of Roman declamation.40
If this is the case, Contr. I pr. 11 and II pr. 5 provide no more
certain proof that Seneca had read any of Cicero's correspondence
than does the reference to a letter of Cassius to Cicero in Suas.
I.5.1*1
But Seneca the Elder was not incapable of original research.
At least his discussion of the events surrounding the death of
Cicero, a subject which must certainly have come within the scope
of his Histories, looks like the fruit of his own investigations.42
In Contr. VII.2.8 he criticizes the declaimers for assuming that
Popillius was the assassin of Cicero, having previously been de-
fended by him on a charge of parricide:
Popillium pauci ex historicis tradiderunt interfectorem Ciceronis
et hi quoque non parricidi reum a Cicerone defensum, sed in
privato iudicio: declamatoribus placuit parricidi reum fuisse.
sic autem eum accusant tamquam defendi non possit, cum adeo possit
absolvi, ut ne accusari quidem potuerit.
This passage provides us with an important insight into the re-
lationship between declamation themes and historiography. Note
that Seneca relies here on secondary sources, historici, for all

his information, but maybe this was inevitable, for there may
never have existed primary written documentation for the identity
of Cicero's assassin, and any there was is likely to have been
suppressed; Cicero's speech for Popillius may have been so unim-
portant that it was never published. We must not underrate
Seneca's carefulness as a historical investigator when he did have
access to primary evidence. In Suas. 6.14ff. he makes a study of
narratives of Cicero's death and estimates of his career by a var-
iety of historians and poets. One extract he examines is a pass-
age, damaging to Cicero's reputation, from Pollio's speech, Pro
Lamia. At least, this passage was to be found in the published
version of this speech, but Seneca tells us that he had learnt
from people actually present at the trial that it had not been in
the speech as Pollio delivered it; he also notes that Pollio had
not had the audacity to include his libellous allegations in his
Histories, and deduces from this that they were quite baseless:
...adieceratque his alia sordidiora multo, ut [tibi] facile
liqueret hoc totum adeo falsum esse ut ne ipse quidem Pollio in
historiis suis ponere ausus sit. huic certe actioni eius pro
Lamia qui interfuerunt, negant eum haec dixisse - nee enim mentiri
sub triumvirorum conscientia sustinebat - sed postea conposuisse.
(Suas. 6.15 W)
Here is a warning to us that, while we must make due allow-
ance for the elder Seneca's sponge-like capacity for absorbing
other people's criticisms, and for the likelihood that in Contr. I
pr. 12 and allied passages he was dependent on a grammatical
source, it cannot be assumed that he was totally without indepen-
dence as a researcher into matters outside his own experience. It
might seem quite legitimate to suggest, on the strength of the
available evidence, that Seneca could have written such literary
criticism as he gives us without having any more knowledge of
Cicero's orations than the average declaimer had, if as much. It
would be easy to argue that his quotation in Suas. 6.27 from the
Pro Archia, one of the less famous speeches, Sextilius Ena fuit...
talis, quales esse Cicero Cordubenses poetas ait, <pingue> quiddam
sonantis atque peregrinum (= Arch. 26) must have been one which
every Corduban schoolboy knew. But it would be unsafe to go so
far as to suggest that, for example, his passing observations re-

lating to Ciceronian diction in Contr. IV pr. 9 need not imply any
careful reading of Cicero, being perhaps a reminiscence of anti-
Hater ian gossip in the schools. The man who was interested enough
in the last years of the Republic to compare versions of Polliofs
Pro Lamia was quite probably much more widely read in Cicero's
oratory than we can judge on the evidence of his criticism.4^
Nobody will expect the elder Seneca to draw heavily on
Cicero's philosophical works. There is, as it happens, a passage
in Contr. I pr. 19 which it is relevant to compare with Tusc.
disp. 1.24.59: Seneca denies that the feats of memory ascribed to
Hortensius and Cineas are anything to be unduly amazed by, whereas
Cicero marvels at these two men as examples of brilliant memor-
izers, along with Simonides, Theodectes and Metrodorus Scepsius.
It need not be, however, that Seneca in Contr. I pr. 19 is con-
sciously challenging Cicero's opinion. The fact seems to be that
the same examples of mnemonists were cited automatically by any
teacher of rhetoric when dealing with the topic of memoria. Quin-
tilian was later to cite Simonides and Hortensius in this connec-
tion (XI.2.11ff.); the elder Pliny (NH VII.88) cites Cineas.
How familiar Seneca was with Cicero's rhetorical treatises is
another, and a more important, matter. Possible verbal echoes of
these works are extremely scarce. One is found in Contr. II pr.
5, just before one of the references paralleled in Suetonius:
...Blandum rhetorem...qui <primus> eques Romanus Romae docuit;
ante ilium intra libertinos praeceptores pulcherrimae disciplinae
continebantur, et minime probabili more turpe erat docere quod
honestum erat discere. nam primus omnium Latinus rhetor Romae
fuit puero Cicerone Plotius.
The echo is of Orator 42.145:
num igitur...est periculum ne quis putet in magna arte et gloriosa
turpe esse docere alios id quod ipsi fuerit honestissimum discere?
It is possible that this verbal reminiscence was transmitted by
the grammatical source which told Seneca about Plotius, even
though it is not paralleled in the extant part of the De gramm. et
rhet., which is also lacking in any reference to Blandus. For
though the scrappiness of Suetonius' text poses insoluble prob-
lems, and it is impossible now to discover for certain the
identity of the grammatical work from which he and Seneca seem to

have derived information about Republican schools and declaimers,
one observation at least will be appropriate here. A serious can-
didate for the role of author of the lost source - he would not
have liked to be called a grammaticus - is Cornelius Nepos, whose
minor work quo distinguit litteratum ab erudito was known to
Suetonius, and included pernickety differentiation between liter-
ary terms which may be compared with the drawing of the distinc-
tion between dicere and declamare and other terminological
niceties in Contr. I pr. 12i 44
Cornelius quoque Nepos libello quo distinguit litteratum ab
erudito, litteratos vulgo quidem appellari ait eos qui aliquid
diligenter et acute scienterque possint aut dicere aut scribere,
ceterum proprie sic appellandos poetarum interpretes, qui a
Graecis grammatici nominentur. (Suet. De gramm. et rhet. 4.2 =
Nepos, De inl. vir. fr. 18*, HRR II 40)
Now, in another fragment quoted by Suetonius, Nepos displays an
interest, shared by Seneca in the remarks on Blandus where the
Ciceronian phrase comes, in a highly unusual subject, the breaking
of class barriers by men of letters:
L. Voltacilius Pilutus...primus omnium libertinorum, ut Cornelius
Nepos opinatur, scribere historiam orsus non nisi ab honestissimo
quoque scribi solitam ad id tempus. (Suet. De gramm. et rhet.
27.If. = Nepos, De inl. vir. fr. 16*, HRR II 40)
Nepos, of course, must have had a very wide familiarity with
Cicero's writings, being intimate with his circle, and the author
of a Life of Cicero as well as one of Atticus. And it is not out
of the question that Nepos could have referred to the unusual
career of Blandus, as he himself lived on into the twenties B.C.
and the knightly rhetor was one of the older generation of
Seneca's declaimers.45
Another briefer echo of Cicero's rhetorical theory is found
in Contr. I pr. 12, one of the passages which certainly have
Suetonian parallels. Cicero had written in De or. 1.34.157:
educenda deinde dictio est ex hac domestica exercitatione. In
Contr. I pr. 12 we are told that Calvus explicitly distinguished
between declamare and dicere, and an illustrative quotation is
elucidated by the gloss, alterum putat domesticae exercitationis
esse, alterum verae actionis. This elucidation may in this case
be the work of Seneca's missing source, but the phrase reappears

in pure Senecan criticism in Contr. Ill pr. 1: simul ad has
domesticas exercitationes secesserant, desertos ah ingenio suo.
However, it is uncertain whether Seneca remembered Cicero's use of
this phrase, which may well have been a watch-word in his day
among people who adopted an old-fashioned attitude towards decla-
mation.
Another, very different, possible case of Ciceronian influ-
ence is to be found at the very end of Seneca's tenth preface. At
the end of the De oratore we find commendation of the young
Hortensius, whose great promise was already evident at the dra-
matic date of the dialogue, first by Catulus, his father-in-law,
and then by his friend Crassus (III.61.228f.). Similarly, Seneca
the Elder closes his last preface with praise of a promising young
man, the son of Clodius Turrinus, an intimate friend of the
family:
inde filius quoque eius, id est meus - numquam enim ilium a vobis
distinxi -, habet in dicendo controversiam paternam diligentiam,
qua vires ingenii sui ex industria retundit. hoc et in ipso
genere vitae sequitur, ad summa evasurus iuvenis nisi modicis
contentus esset, et ideo dignus est cuius tarn modestis
cupiditatibus Fortuna praestet fidem. (Contr. X pr. 16)
Both Cicero and Seneca the Elder seem to be aligning themselves
with a tradition which went back to Plato, who introduces praise
of the young Isocrates at the end of the Phaedrus (278e-279b). Of
course we do not know how widely this Platonic device was used
among Roman writers of the late Republic and early Empire, but De
or. III.61.228f. must surely have been the most famous example of
its use. Even if not directly inspired by this passage, Contr. X
pr. 16 is a tantalizing reminder that Seneca the Elder must have
read much in the course of his long life to which he never refers
in his extant books. The fact remains that any study he may have
made of Cicero's rhetorical treatises left surprisingly little
mark on either his critical ideals or his vocabulary.46
Another debatable question is whether he had a first-hand
knowledge of the speeches and criticism of Calvus. He once quotes
a remark by Calvus about declamation, but it comes in one of the
passages (Contr. I pr. 12) where he may be suspected of depending
on a grammatical source. That the quotation is to be ascribed to

Calvus rather than to Cicero seems clear, despite a certain ambi-
guity in the text of the sentence which introduces it.
...ipsa Meclamatio1 apud nullum antiquum auctorem ante Ciceronem
et Calvum inveniri potest, qui declamationem <a dictione>
distinguit; ait enim declamare iam se non mediocriter, dicere
bene; alterum putat domesticae exercitationis esse, alterum verae
actionis.
<a dictione> add, Gertz; <ab oratione> Bursian. declamare
iam se Bonnet; declamare iam ne M (est add. in marg. M^);
declamare est P; declamare iam NSa; declamare est iam 3v.
Let us accept GertzT supplement and BonnetTs conjecture: there
still remains a certain clumsiness of construction in this pass-
age. After the plural antecedent, Ciceronem et Calvum, one ex-
pects a plural verb in the relative clause. Instead we find the
singular distinguit, with a quotation following it which illus-
trates one manfs usage of the term declamare. The singular verbs
in the remainder of the sentence: ait...putat... forbid us simply
to emend distinguit to distinguunt. It would be unjust to suggest
that the clumsiness is attributable to lack of care on Senecafs
part in epitomizing his source, for it is clumsiness of a kind
unparalleled in his work, and particularly surprising found in an
otherwise artistically written preface. Perhaps we ought to
transpose ante Ciceronem et Calvum, placing them after inveniri
potest. Alternatively, as J. B. Hall has recently argued,47
either Ciceronem et or et Calvum may be regarded as an interp-
olation, most probably the former, as it is much more likely that
the name of Cicero who fhas held the stage since section 11 of
this preface1 should have crept in, than that of Calvus; the pres-
ence of the words ante Ciceronem in a sentence a few lines above
could have precipitated the interpolation of a gratuitous refer-
ence to Cicero.48*
The other reference to Calvus in Seneca1s criticism may seem
at first sight totally to confute any notion that his reading in
Republican oratory could have been very limited. The remarkable
passage Contr. VII.4.6-8 presents an estimate of Calvus quite
opposed to that put about by Cicero in his rhetorical works. 49
Cicero had represented his rival Calvus as a speaker who,
despite many admirable gifts, was less effective than he might
have been owing to his excessive self-criticism and his desire to

be reckoned an orator in the Attic tradition:
sed ad Calvum - is enim nobis erat propositus - revertamur; qui
orator fuit cum litteris eruditior quam Curio turn etiam accuratius
quoddam dicendi et exquisitius adferebat genus; quod quamquam
scienter eleganterque tractabat, nimium tamen inquirens in se
atque ipse sese observans metuensque ne vitiosum colligeret, etiam
verum sanguinem deperdebat. itaque eius oratio nimia religione
attenuata doctis et attente audientibus erat inlustris, <a>
multitudine autem et a foro, cui nata eloquentia est, devorabatur.
Turn Brutus: Atticum se, inquit, Calvus noster dici oratorem
volebat: inde erat ista exilitas quam ille de industria con-
sequebatur. dicebat, inquam, ita; sed et ipse errabat et alios
etiam errare cogebat. (Brut. 82.283f.)
Cicero proceeds to point out to Brutus that there were many var-
ieties of Attic oratory: the style of Lysias so admired by the
modern aspirants to the name of Attici was certainly one type, but
Demosthenes was no less Attic, nor was Hyperides, nor, for that
matter, Demetrius of Phalerum. The fundamental mistake of the
Roman Atticists was, according to Cicero, that they assumed that
one only needed ieiunitatem et siccitatem et inopiam (82.285) to
emulate the Attic style, and that they failed to recognize that
Demosthenes was the greatest Athenian orator of them all. Therein
lay the reason why, so he says, they failed to hold anyoneTs at-
tention in court:
ne illud quidem intellegunt, non modo ita memoriae proditum esse
sed ita necesse fuisse, cum Demosthenes dicturus esset, ut con-
cursus audiendi causa ex tota Graecia fierent. at cum isti Attici
dicunt, non modo a corona, quod est ipsum miserabile, sed etiam ab
advocatis relinquuntur. (84.289)
In Contr. VII.4.6-8, on the other hand, Calvus is not only
represented as an orator so totally uninhibited where actio was
concerned that he would rush around the court while delivering his
speeches, he is actually said to have imitated Demosthenes in the
rhythmic arrangement of his words:
Calvus, qui diu cum Cicerone iniquissimam litem de principatu
eloquentiae habuit, usque eo violentus actor et concitatus fuit,
ut in media eius actione surgeret Vatinius reus et exclamaret:
rogo vos, iudices: num, si iste disertus est, ideo me damnari
oportet? idem postea, cum videret a clientibus Catonis, rei sui,
Pollionem Asinium circumventum in foro caedi, inponi se supra
cippum iussit - erat enim parvolus statura, propter quod etiam
Catullus in hendecasyllabis vocat ilium fsalaputium disertum1 - et
iuravit, si quam iniuriam Cato Pollioni Asinio accusatori suo
feeisset, se in eum iuraturum calumniam; nee umquam postea Pollio

a Catone advocatisque eius aut re aut verbo violatus est. solebat
praeterea excedere subsellia sua et inpetu latus usque in
adversariorum partem transcurrere. et carmina quoque eius, quam-
vis iocosa sint, plena sunt ingentis animi. dicit de Pompeio:
digito caput uno
scalpit. quid credas hunc sibi velle? virum.
conpositio quoque eius in actionibus ad exemplum Demosthenis
viget: 50 * nihil in ilia placidum, nihil lene est, omnia excitata
et fluctuantia. hie tamen in epilogo, quern pro Messio tune tertio
causam dicente habuit, non tantum leniter componit sed
> 51
< > cum dicit: fcredite mihi, non est turpe misereri1,
et omnia in illo epilogo fere non tantum emollitae compositionis
sunt sed infractae. (Contr. VII.4.6-8)
This critique is not the only ancient example of a view of
Calvus radically different from Cicero's. The Catullan hendeca-
syllables cited as an illustration of Calvus1 stature are also
evidence of the popular appeal of his oratory:
risi nescioquem modo e corona
qui cum mirifice Vatiniana
meus crimina Calvus explicasset
admirans ait haec manusque tollens
f
Di magni, salaputium disertum.T (Carm. LIII)
Quintilian (X.I.115) tells us that critics in his day were divided
between those who trusted Cicero's judgement on his rival and
those who preferred Calvus to all other writers; his own verdict
lies between the two extremes:
inveni qui Calvum praeferrent omnibus, inveni qui Ciceroni
crederent eum nimia contra se calumnia verum sanguinem perdidisse;
sed est et sancta et gravis oratio et castigata et frequenter
vehemens quoque. imitator autem est Atticorum, fecitque illi
properata mors iniuriam si quid adiecturus sibi, non si quid
detracturus, fuit.52
The real reasons for the difference of opinion cannot be un-
covered now that all but the most meagre fragments of Calvus1 ora-
tory has been lost. Cicero was to some extent at least sincere in
the opinion of Calvus he expresses in Brut. 82.284: it is true
that in a letter to Calvus which embarrassingly got into the wrong
hands he appears to have praised aspects of his writing very ef-
fusively, but even here he seems to have exhorted Calvus to aim at
greater forcefulness. He explains to Trebonius:
ego illas Calvo litteras misi non plus quam has, quas nunc legis,
existimans exituras; aliter enim scribimus quod eos solos quibus
mittimus, aliter quod multos lecturos putamus; deinde ingenium
eius maioribus extuli laudibus quam tu id vere potuisse fieri
putas, primum quod ita iudicabam: acute movebatur, genus quoddam

sequebatur, in quo iudicio lapsus quo valebat, tamen adsequebatur
quod probarat; multae erant et reconditae litterae, vis non erat;
ad earn igitur adhortabar; in excitando autem et in acuendo
plurimum valet, si laudes eum quern cohortere. habes de Calvo
iudicium et consilium meum; consilium, quod hortandi causa
laudavi, iudicium, quod de ingenio eius valde existimavi bene.
(Ad fam. XV.21.4) 53
Cicero was obviously being extremely unfair to Calvus in suggest-
ing in Brut. 84.289 that he exercised no power over bystanders in
the corona, but perhaps the Vatinian orations, to which Catullus1
poem and other favourable estimates of his talent refer, were of a
very much more brilliant style than the rest of his speeches: note
how in Tacitus' Dialogus Aper, the arch-disparager of Republican
oratory, has this to say about his contemporaries' reading habits:
quotus enim quisque Calvi in Asicium aut in Drusum legit? at
hercule in omnium studiosorum manibus versantur accusationes quae
in Vatinium inscribuntur, ac praecipue secunda ex his oratio; est
enim verbis ornata et sententiis, auribus iudicum accommodata, ut
scias ipsum quoque Calvum intellexisse quid melius esset, nee
voluntatem ei, quo <minus> sublimius et cultius diceret, sed
ingenium ac vires defuisse. (Dial. 21.2)
Most of Calvus1 extant prose fragments come from the Vatinians.^
Still, Cicero's grudging estimate of his rival's style seems hard
to accept as wholly justified. Perhaps A. D. Leeman has got to
the heart of the matter when he suggests that the character of
Calvus' oratory 'could make Cicero's elaborate pathos look a
little cheap in the eyes of his young contemporaries',55 and we
should regard Brut. 82.283ff. as the product of his discomfort at
realizing this.
The extant fragments of Calvus' orations56 certainly provide
evidence of greater rhetorical artistry and forcefulness than one
would expect of an Atticist if one takes seriously Cicero's pic-
ture of the would-be Attici as men misguidedly aping nothing but
the Lysian plain style. They serve better to support the state-
ments about Calvus' vigorous and agitated compositio in Contr.
VII.4.8. Study of their clausulae yields particularly interesting
results.
De Groot's clausula De Groot's clausula

statistics: statistics:
Cicero's speeches: 'Normal1 Demosthenes: Thucydides
in percentages57 in percentages58
Fr. 22 (In Vat.) = Quint. VI.1.13
9.7 : 5.4 factum esse ambitum scitis omnes et hoc 4.8 : 3.4
vos scir(e) omnes sciunt.
Fr. 23 (In Vat.) = Quint. IX.2.25

25.3:17.2 perfrica frontem et die te digniorem qui 18.9:14.2
praetor fieres quam Catonem.
Fr. 25 (In Vat.) = Aquila Rom. RLM 35.4

cf. Quint. IX.3.56
Not non ergo pecuniarum magis repetundarum quam Not
available maiestatis, neque maiestatis magis quam available
Plautiae legis, neque Plautiae legis magis
quam ambitus, neque ambitus magis quam
omnium legum omnia iudicia perierunt.
Fr. 26 (In Vat.) = Iul. Sever. RLM 366.5

2.9 : 1.9 hominem nostrae civitatis audacissimum, de 0.8 : 3.1
factione divitem, sordidum, maledic(um)
accuse
Fr. 28 (In Vat.) = Charisius RLM 296.21

—v—v ad ita mihi Iovem deosque inmortales — ^—^
25.3:17.2 velim bene fecisse, iudices, ut ego pro 18.9:14.2
_w—^_w certo_habeo, si parvuli pueri de ambitu -^—w-w
4.0 : 2.4 iudicarent. 3.0 : 2.3
Fr. 32 (Pro Mess.) = Sen. Cpntr^ VII.4.8

4.7 : 2.4 credite mihi, non est turpe misereri. 3.5 : 1.2
Fr. 36 (incertae sedis) = Charisius RLM 102.20

4.9 : 4.4 quorum praedulcem cibum stomachus ferre 4.8 : 3.7
non potest.
These are the only fragments of any length extant from Calvus1
orations. Fr. 36 may not give us the end of a sentence.
Obviously one might wish for very much fuller evidence. How-
ever, fr. 25 in particular shows how unjustifiable it is to trust
implicitly Cicero's picture of the Atticists as deliberate seekers
of exilitas, forever draining the life-blood of their eloquence by
excessive purging away of what they regarded as vitiosum (Brut.
82.283f. cf. Or. 9.28ff.), and how unreasonable, despite Cicero,
Or. 23.75ff., it is to regard as the only Atticist style the
plainest of the three types of prose distinguished by Cicero, with
its avoidance of rhythmic artifice (ibid. 23.77) 59 and brilliant
ornatus (ibid. 23.78). 60 There is no doubt that Calvus has to be

classed among the Atticists: indeed he is the one Roman Republican
orator of whom we are explicitly told that he aspired to the epi-
thet Atticus (Brut. 82.284). 61 Yet the style of fr. 25 is such
that it would have been condemned as an example of Asianic bombast
long ago, had it been ascribed to anyone but Calvus. More to the
point, it can serve as an illustration of the Demosthenic com-
positio alluded to in Contr. VII.4.8: compare the famous example
of HALua£Igradatio from the De corona (179), cited by Demetrius
(Eloc. 270), and in translation by Quintilian (IX.3.55) just
before the Calvan fragment:
OUM e^nov uev xatjxa, oux eypaâ 6e, ou6' eypacjâ yev, oux
ercpeaBeuaa 6e, ou6' enpeaBeuaa yev, oux eiteoaa 6e OnBaCoug...62
For the clausula of fr. 25 we have no statistics to show its rela-
tive frequency in rhythmic and virtually non-rhythmic prose, but
as it is a resolution of the much favoured rhythm, cretic +
spondee, there is no reason to suppose that Calvus would have been
thought inartistic in choosing this cadence. Examination of the
clausulae of the other fragments suggests that Calvus1 preferences,
at least in the Vatinians, were similar to Cicero's, and not too
far from those of Demosthenes. We must be careful, then, not to
attach too much weight to Cicero's allegations (Or. 9.28ff.) that
the Atticists invariably avoided rhythm in their prose.63 Note
also that two sentences out of the seven end with a dichoreus ,sl*
a rhythm which we should therefore stop regarding as an index of
Asianist tendencies on the strength of Cic. Or. 63.212: insistit
autem ambitus modis pluribus, e quibus unum est secuta Asia maxime,
qui dichoreus vocatur, cum duo extremi chorei sunt, id est e
singulis longis et brevibus. Bear in mind Shewring's statement
about Demosthenes1 rhythms, 'His only well-defined clausula is
-v^-^',65 and the fact which emerges from Shewring's researches,
that various of the earlier Attic orators, including Lysias, had
shown a decided preference for this rhythm. As for the general
tone of the fragments of Calvus' oratory, all but one of them dis-
play the vehemence which Quintilian found in his speeches
frequenter (X.I.115). The exception is fr. 32 which, as we have
seen, was cited expressly to illustrate how Calvus knew how to
employ a 'soft1 approach untypical of his style, when this was

appropriate.
The passage describing Calvus in Contr. VII.4.6-8 is by any
standards a remarkable piece of criticism. But does its existence
mean that Seneca the Elder was familiar with the works of Atticist
critics now lost? Once more we have to consider whether he may
not be presenting data received at second or third hand. In
Contr. VII.4.6 he has been reporting some observations made by
Latro about the need for a change of style when one wishes to
arouse pity, for example in a peroration:
aiebat...ipsam orationem ad habitum eius quern movere volumus ad-
fectus molliendam. in epilogis nos de industria vocem quoque
infringere et dare operam ne dissimilis orationi sit orator; com-
positionem quoque illis mitiorem convenire.
The function of the passage about Calvus, which follows immedi-
ately after this, is to illustrate the point which Latro was
making: Calvus was normally a violentus actor et concitatus, and
normally aimed at vigorous and agitated composition but if he had
to deliver a pathetic epilogus, such as that of the Pro Messio, he
would change his style and use far gentler rhythms.66 Now, given
the fact that Seneca elsewhere has been found to have a tendency
to move from indirect to direct speech while all the time report-
ing another critic,67 it is quite likely that the passage about
Calvus originally formed part of Latrofs lecture, even though
Seneca gives it in direct speech. And, given the fact that Latro
'despised and was ignorant of the Greeks' (Contr. X.4.21), it must
be doubted whether the comparison of Calvus' rhythms with Demos-
thenes', if this lecture was Seneca's source for it, could have
been the fruit of Latro's own reading. What we have in Contr.
VII.4.6-8, perhaps, is Latro displaying his historiarum omnium
summa notitia by reeling off a selection of the acta of Calvus
(cf. Contr. I pr. 18). Lest anyone should suggest that the ma-
terial in this passage is not the sort of historical data which a
rhetorician was likely to learn as potentially useful for decla-
mations, consider Contr. X.I.8 where Latro mentions, in a list of
exempla of the violence of accusations flung around in time of
civil war, the fact that:
in Cn. Pompeium terra marique victorem fuit qui carmen conponeret,
uno, ut ait, digito caput scalpentem.

The allusion is to the very poem of Calvus quoted in Contr. VII.
4.7. 68
There is precious little else, either in the critical parts
of Seneca's books or in the declamatory extracts, about Republican
oratory. Seneca tells an anecdote in Contr, I pr. 19 about
Hortensius1 brilliant memory, but gives us no criticism of his
oratorical manner. Capito, as we have seen, alludes to Hortensius1
late speech Pro Cn. Pompei bonis, but tells us nothing about this
speech which could not have been deduced from its title (Contr.
VII.5.6). The allusion to Calvus by Latro in Contr. X.I.8 is only
one item in quite a long series of illustrations of the prevalence
of invective in the late Republic, which take us into lesser-known
areas of the history of rhetoric:
Macerio qua violentia in absentiam Metelli strepuit! M. Cato
Pulchro obiciente furtorum crimina audivit. quae maior indignitas
illius saeculi esse potuit quam aut Pulcher accusator aut reus
CatoI in Cn. Pompeium terra marique victorem fuit qui carmen con-
poneret, uno, ut ait, digito caput scalpentem; fuit aliquis qui
licentia carminis tres auratos currus contemneret. M. Bruti
t sacratissimi t 6 ^ eum eloquentia lacerat, cum quidem eius civili
sanguine non inquinatas solum manus sed infectas ait; atque ille
tamen, cum tres consulatus ac tres triumphos scinderet, adeo non
timuit ne esset reus ut etiam disertus esse curaverit.
But such a display of erudition is extremely unusual. The de-
claimers have nothing to say about Pompey and Caesar as orators
beyond this reference to Pompeyfs eloquence in defending himself,
and a quotation by the minor declaimer Tuscus of Caesarfs famous
veni, vidi, vici (Suas. 2.22). Occasional mention is made in the
suasoriae addressed to Cicero of Antony's inadequacy as a speaker:
Haterius, for example, recalls in Suas. 7.3 the eloquence of
Antony's famous grandfather, M. Antonium ilium indignum hoc
successore generis. Cassius Severus mentions Sallust as an
example of the inability of any one man to excel in more than one
genre: orationes Sallustii in honorem historiarum leguntur (Contr.
Ill pr. 8 ) .
There is no way of being sure, then, how much Greek or Repub-
lican Roman oratory and literary criticism Seneca the Elder had
read. Nearly all his allusions to such literature could be ex-
plained as the outcome of receptiveness to the views of better-
informed critics in and around the schools of declamation or as

derived at second hand from an erudite, but perhaps quite brief,
account of the rise of Roman declamation. On the other hand, the
trouble he takes to assess the value of sources relating to the
last days of Cicero, and his use of the Platonic and Ciceronian
motif of praise for a rising young talent at the end of his series
of ten critical prefaces, serve as reminders that an extreme
scepticism which would deny him any reading beyond that of the
narrowest scholasticus cannot be wholly justified.
2 THE HISTORY OF DECLAMATION
declamabat autem Cicero non quales nunc controversias dicimus, ne

tales quidem, quales ante Ciceronem dicebantur, quas thesis
vocabant. hoc genus materiae, quo nos exercemur, adeo noyum est,
ut nomen quoque eius novum sit. controversias nos dicimus: Cicero
causas vocabat. hoc vero alterum nomen Graecum quidem, sed in
Latinum ita translatum, ut pro Latino sit, scholastica, contro-
versia multo recentius est, sicut ipsa 'declamatio' apud nullum
antiquum auctorem ante Ciceronem et Calvum inveniri potest, qui
declamationem <a dictione> distinguit;1 ait enim declamare iam se
non mediocriter, dicere bene; alterum putat domesticae exercita-
tionis esse, alterum verae actionis. modo nomen hoc prodiit; nam
et studium ipsum nuper celebrari coepit: ideo facile est mihi ab
incunabulis nosse rem post me natam. (Contr. I pr. 12)
The account of the rise of declamation in Contr. I pr. 12 is
an excursus prompted by the reference Seneca has just made to
Cicero's late declamations in the company of Hirtius and Pansa.
It is not intended, therefore, to give more than an account of
Roman declamation from around the time of Cicero to his own day.
The indebtedness of the Roman rhetoricians to their Greek counter-
parts, hinted at only in the remarks on the term scholastica, is
not given the attention which would be its due in a full account
of ancient rhetorical education. Nevertheless it is essential for
a complete understanding of Contr. I pr. 12 to consider what is
known about rhetorical exercises used in Greece before Seneca's
time - the Roman thesis, suasoria and controversia all had Greek
antecedents - and how ancient authorities viewed the history of
Greek education.2
The %eob£ was an abstract philosophical proposition on which
one could discourse at length from two opposed standpoints. The
question for discussion might be a matter of pure theory
(cognitionis), for example: verine sint sensus; sitne ius id quod
maiori parti sit utile; ecquisnam perfecte sapiens esse possit; or
it might be of more practical application (actionis), for example:
quibus officiis amicitia colenda sit; quern ad modum sit res
publica administranda; quern ad modum in paupertate vivendum.^ The
The history of declamation 105
factor which distinguished the thesis from all other exercises was
its total abstraction, the absence from its theme of any reference
to particular people or particular circumstances (see Cic. Or.
14.46).
The practice of speaking at length for and against decrees
went back to the fifth-century sophists. According to Diogenes
Laertius, Protagoras iipaJTOS MaTe6eu£e T & S itpos Tag §eaeus eno-
Xebpnaeus (Diog. Laert. IX.53). The discourse on the theme, ws
xaptaxeov un ep&vxu yaAAov n epwvxt attributed to Lysias in
Platofs Phaedrus (227c; 23Oe-234b) is an early example. According
to Quintilian, Aristotle and Theophrastus were responsible for
initiating the use of the -deaus as an educational exercise (Quint.
XII.2.25). Cicero tells us that Aristotle encouraged his pupils
to treat the themes with rhetorical elaboration:
haec igitur quaestio a propriis personis et temporibus ad universi
generis rationem traducta appellatur §eaus. in hac Aristoteles
adulescentis non ad philosophorum morem tenuiter disserendi, sed
ad copiam rhetorum in utramque partem, ut ornatius et uberius dici
posset, exercuit; idemque locos - sic enim appellat - quasi
argumentorum notas tradidit unde omnis in utramque partem
traheretur oratio. (Cic. Or, 14.46)
The catalogue of Aristotle's works includes references to Geaets
eTiLxetpnyaxuxal xe/ , Geaeus epcoTUMal 6' - the tradition represen-
ted by the show-speech of Lysias in the Phaedrus evidently
continued - Geaeus cpuALwat 3' » Geaeus iiepL (JÛXTIS a' (Diog. Laert.
V.24); Theophrastus is credited with Geaeus x6' (ibid. 44) and
Geaeus <aXXo> y' (ibid. 49),h
The Seaus appears to have remained in use amongst the later
Peripatetics, and to have gained acceptance by the Academics too,
despite the strong hostility towards rhetoric which persisted long
among members of that school.5 At any rate, Cicero represents the
use of the deaus as peculiarly characteristic of both Academics
and Peripatetics: quae exercitatio nunc propria duarum philo-
sophiarum de quibus ante dixi putatur (De or. III.27.107).6 These
two philosophical schools did not exercise a complete monopoly
over the êaus, but were near enough to doing so as to call forth
a lament from one of the speakers in the De oratore at the way he
and others whose main care was for rhetoric, had unfairly been
deprived of their patrimony, de nostra possessione depulsi in parvo
et eo litigioso praediolo relicti sumus (ibid. 108), whereas apud

antiquos - he does not identify them - the discussion of wide ab-
stract issues had been considered the province of people like him-
self, a quibus omnis de rebus forensibus dicendi ratio et copia
petebatur (ibid. 107). It is true that Hermagoras, an influential
rhetorician of the second century B.C.,7 had reasserted the view
that the treatment of abstract issues (Seaets) was as much a part
of oratory as judicial, deliberative and epideictic rhetoric con-
cerned with particular events (uftodeaeus) .® But it seems that the
Hellenistic rhetoricians devoted very little space in their treat-
ises to the consideration of the ways in which abstract themes
might be treated, and consequently their claim to be concerned
with more than mere pettifoggery did not constitute a serious
challenge to the philosophers. According to a speaker in Cicero's
De oratore the rhetoricians of his time - the dramatic date of the
dialogue is 91 B.C. - exercised an almost complete monopoly over
the teaching of the kind of themes which Hermagoras had classed as
UTtofteaetc;, illud alterum genus quod est temporibus, locis, reis^
definitum, despite the interest recently taken by Philo the Aca-
demic in this type of subject-matter (see Tusc. disp. II.3.9), but
on the other hand they assigned the $eaus to a sadly subordinate
position in their educational scheme: alterum vero tantum modo in
prima arte tradenda nominant et oratoris esse dicunt; sed neque
vim neque naturam eius nee partis nee genera proponunt, ut praeter-
iri omnino fuerit satius quam attactum deseri... (De or. III.28.
110).
The history of rhetorical exercises of the uioêats type is
harder to trace. Regrettably little relevant information has sur-
vived from the Hellenistic schools,10 so that it is easy to be
misled by the elder SenecaTs talk of rem post me natam (Contr. I
pr. 12) into thinking that suasoriae and controversiae were indeed
phenomena which sprang up from nowhere in Augustan Rome. It is a
curious fact that, though it is clear that Greek rhetoricians were
declaiming on just the same themes as their Latin-speaking col-
leagues in the time of Seneca the Elder, the first extant Greek
writings which are generally reckoned to give us treatments of
rhetorical themes equivalent to suasoriae and controversiae date
from the later Empire.11 Also, when looking for evidence of the
teaching methods of Hellenistic rhetoricians, one is continually
faced with the difficulty that, while it is unquestionable that
the chief concern of these men was with the types of rhetoric
which Hermagoras classed as uitodeaeLS, and that it is unlikely
therefore that they set their pupils no exercises more advanced
than those known as progymnasmata, it is impossible usually to
prove which, if any, of the numerous examples of fictitious delib-
erations and law-suits found in works indebted to Hellenistic
theory like Cicero's De inventione and the Rhetorica ad Herennium,
were actually used as subjects for declamation in their schools.
To be on the safe side we must look for other sources of evidence
for the antecedents of suasoriae and controversiae,^^
Philostratus in Vit. soph. 1.481 divides the history of Greek
school rhetoric into two periods or 'sophistics1:
n yev 6rj dpxaCa
x aocpuaxtxri
p xal xa pcpLAoaocpouyeva UTtoxuêyevri
&&
auxa &TtoT&6nv l eg ynxos, 6i,eAeyexo yev ydp
xal d "Kepii dv6pelas,
d l
6baAeyexo 6e Ttepl 6uxat6xrixoc;, npwwv xe itcpu xau §ewv xal oitn
otTteaxnydxLaxat n i,6ea xou xoayou. n 6e yex' eKeuvnv, nv ouxl
veav, dpxaua yap, 6euxepav 6e yaAAov Tipoaprixeov, xous itevnxas
UTiexuiiwaaxo xal xous iiAouaCous xal xous dpuaxeas xal xous
xupdvvous xal xots es ovoya UTiodeaets, ecp' as T\ uaxopua ayeu. ?ipê
6e xris yev dpxatoxepas Fopyuas 6 Aeovxuvos ev 'OexxaAous, xns 6e
6eux£pas Auax^vriS 6 'Axpoynxou xwv yev 'A^nvriau TIOAOXLXWV exueawv,
Kapua 6e evoyuAnaas xal 'Po6(j)9 xal yexex^bplcovxo xas UTtodeaets ou
yev xaxd xexvriv, ou 6e duo Fopylou xaxd xo
According to Philostratus, then, the first sophistic, beginning
with Gorgias, was noted for the extended rhetorical treatment of
philosophical themes, for example, questions about courage, jus-
tice, cosmology. These, despite Philostratus1 loose usage of the
words UTtoxodeyevn and uiroêaetg in connection with them, may be
assumed to have included the exercises called deaeus whose history
we have been tracing. Philostratus also mentions a rather differ-
ent type of theme characteristic of the first sophistic 'about
heroes and gods'. Gorgias' Helen and Palamedes, Alcidamas'
Odysseus, and Antisthenes' Aias and Odysseus may be taken as
examples of this type. Aeschines is represented as having founded
the second sophistic after his expulsion from Athens (in 330 B.C.).
Characteristic of this new educational movement was the use of
themes peopled by 'poor men, rich men, men renowned for bravery,
and tyrants1, and themes derived from history, containing refer-

ence to named individuals. Now, the conventional figures pauper,
dives, vir fortis and tyrannus are all familiar from Roman decla-
mation themes, the last two types proclaiming their non-Roman
origins for all to hear, and of course historical subjects find a
place in the Roman collections too, more often among suasoriae
than controversiae. It seems highly likely, then, that the themes
which Philostratus is describing are Greek forerunners of contro-
versiae and suasoriae, especially in view of the fact that Quinti-
lian (II.4.41) dates the origins of judicial and deliberative
exercises to approximately the time of Demetrius of Phalerum: nam
fictas ad imitationem fori consiliorumque materias apud Graecos
dicere circa Demetrium Phalerea institutum fere constat. Demet-
rius governed Athens for the ten years before 307, and became li-
brarian at Alexandria in 297, so that Quintilianfs dating agrees
with Philostratus1. Quintilian does not seem to have heard the
innovation ascribed to Aeschines; his sources were more inclined
to attribute it to Demetrius, but he questioned whether there was
any external evidence to corroborate this tradition.
The clearest evidence^ for the use of rhetorical exercises
on deliberative themes in the Hellenistic world is provided by
Polybius in a passage where he criticizes the historical methods
of Timaeus. He attacks, among other things, that historianTs man-
ner of presenting xas 6nunyopCas xal xas TtapaxAriaeLs, exu 6e xous
Ttpea$euTLMOUs Aoyous (XII.25a.3). Polybius1 complaint is that
Timaeus, instead of reporting the speakers1 true words in full or
giving the gist of them, puts completely imaginary words into
their mouths, as one might do when holding forth in school on a
set theme:
ou yap T<* priêvTct yeypacpev, ou6' ws eppn^ri xax' dArideuav, aAAa
itpoêuevos ws 6eu pnSnvau, navxas eâpL^yeuxau xous pnêvxas
Aoyous Mat xa Tiapenoueva xous Ttpayyaauv ouxws J) s a v e u
x u s e v 6 u a x p u $ i J L i t p o s u u o ^ e a u v
E it i» x e ^ P o i n . . .caaTiep dn66eL^LV xris eauxou 6uvaueci)S
Ttououuevos, dtAA' OUH e^nyncruv xwv wax' dAn-êtav etpnyevcav.
(XII.25a.5)
Later (25 .Iff.) he gives explicit examples of Timaeus1 failing:
words which he placed in the mouths of some of Sicily's most sen-
sible rulers were inferior even to the normal products of school
k
rhetoric (25 .8). He criticizes Timaeus for attributing to Hermo-
crates, one such ruler, a speech at a peace conference, which,
consisting as it did largely of a prolonged lecture, full of inept
mythological allusions, on the differences between war and peace,
was an insult to the intelligence of his listeners, the leading
citizens of Gela, who had been in favour of peace from the outset.
He concludes his critique with these words:
6n TLOL HOT' av aAAous expnaaxo Aoyous n fipocpopaus
LOV apTU yevoyevov Ttepl 6taTpb3as nat <T&S> in xtov uito-
wv TioAuTipayyoauvas xal gouAoyevov TtapayyeAyaTLH&s in TWV
Ttapeitoyevwv TOLS TipoawTtOLS TIOLSZOSOLL TTIV eiruxeupncTLV* doneZ yap
<oux eT>epots, aAAa TOUTOLS O^S Tuyatos 'EpyoMpaxnv nexpr\o%o,{,
cpnau. (26.9)
The expression . . .itapayyeAyaTUxws*l+ in TCOV iiapeiioyevwv TOUS
TipoawiroLS 7ioteuadat xnv eTiLxeLpncn/V, taken together with the fact
that HermocratesT speech comes in the context of negotiations
about peace, and fwar and peace1 had been classed by Aristotle
(Rhet. 1.4.7) as one of the five standard topics of deliberative
rhetoric, must lead us to conclude that the rhetorical exercise
which Polybius has in mind is the deliberative prosopopoeia, a
variety of declamation in which the suasor actually adopts the
r61e of a historical personage, rather than being merely an
unnamed adviser of some great man, which was the case in some
other suasoriae (cf. Quint. III.8.52). The speech in Timaeus1
Histories to which Polybius next turns his attention (26 .1) is
even more clearly to be compared with this sub-species of suasoria:
Timoleon exhorts the Greeks to do battle with the Carthaginians.
Rhetorical exercises in imitation of deliberations were evi-
dently in use, then, in the second century B.C., and nothing makes
it improbable that they were known in the time of Aeschines and
Demetrius of Phalerum. We must call into question, indeed,
whether they originated as late as Aeschines1 time. With the
deliberative orations attributed to historical characters by
Timaeus we may compare certain historically dubious speeches in
Herodotus. For example in 1.71 Sardanis the Lydian makes a speech
intended to dissuade Croesus from invading Cappadocia. Like Timo-
leon in Timaeus ap. Polyb. XII.26 .4 Sardanis lays considerable
emphasis on the fact that the enemy were wearers of trousers.
Sen. Suas. 4 provides us with an example of a suasoria used in

the Roman schools about the advisability of an invasion: deliberat
Alexander Magnus an Babylona intret, cum denuntiatum esset illi
responso auguris periculum. Again, in Hdt. III.8Off. the Persians
deliberate in set speeches whether to aim for a democracy, an oli-
garchy or a monarchy. These Aoyot ctTitaTOt. . . evCotab 'EAArivoov, in
which Herodotus nevertheless wants his hearers to believe, may be
compared with the triplices...suasoriae known to Quintilian (III.
8.33). 15
In the case of exercises of the controversia type as well,
the question is not whether the dating of their origins given by
Philostratus and Quintilian is too early, but rather whether it
may not be far too late.
Independent evidence for their use in Hellenistic schools
seems to be lacking.16 A distinction has to be drawn between the
controversia type of judicial declamation (characterized by the
arrangement of conventional figures, pauper, dives, vir fortis,
tyrannus etc. or, more rarely, historical personages, in complex
and purely imaginary legal tangles) 17 and another type of display
speech ad imitationem fori in which the declaimer makes a reply to
a speech once delivered in court by a famous orator, or alterna-
tively tackles the case from the same standpoint as the original
orator, but attempts to improve on it. It is an example of this
variety of rhetorical exercise which is preserved in the third-
century B.C. papyrus Berl. Pap. P. 9781:18 the declaimer has
chosen the role of Leptines and defends himself against the
charges levelled against him by Demosthenes in Or. XX. The
1
Apologies of Plato and Xenophon, and Polycrates Prosecution of
Socrates (Diog. Laert. 11.38) may be regarded as forerunners of
this type of writing. Similar rhetorical tours de force were
occasionally attempted in the elder Senecafs time: such were
Cestius* speech in Milonem (Contr. Ill pr. 16), Gallio's re-
scriptum Labieno pro Bathyllo Maecenatis (Contr. X pr. 8 ) , and
Latro's most eloquent oration pro Pythodoro Messalae (Contr. II.
4.8 W ) . 1 9 *
The best place to look for the antecedents of controversiae
is in the set of Tetralogies ascribed to Antiphon, fine examples
of sophistic rhetoric which are generally accepted as dating from

the fifth century B.C.,20 even if maybe not to be attributed to
the orator Antiphon. General affinities between these specimen
speeches and later judicial declamations have of course long been
recognized,21 but it has not been appreciated just how closely the
first and third Tetralogies in particular are related, in themes
though not in treatment, to the Roman controversiae.
In the first Tetralogy the given facts of the case appear to
have been these. A man has been found murdered one night in an
isolated spot, the slave with him having been mortally wounded.
Neither has been robbed of his cloak. The man who is charged with
the murder is a rich man, who was a known enemy of the murdered
man, and had recently been accused by him of embezzling sacred
monies. The case was still waiting to be heard at the time of the
Diipoleia when the murder took place. The wounded slave, when
asked to identify the murderer by the people now bringing the
charge, presumably the murdered man's relatives, named the rich
man, but since then has died.
Nothing more than these sparse facts is taken for granted by
the author of the Tetralogy2-2- and they contain nothing,2^ apart
from the specification that the dead men were not robbed of their
cloaks, rather than simply not robbed, the assumption that a slave
might be worth robbing, and the reference to an explicit date,
which a declaimer of the early Roman Empire would not have found
very familiar. How similar in kind the theme of the first Tetra-
logy is to those of later controversiae will be most simply demon-
strated if we translate all the given facts, except these three
minor details, into the language of the Roman scholastics, using a
pastiche of phrases from themes known to Seneca the Elder and
ps.-Quintilian:
qui divitem inimicum peculatus reum detulerat, inter moras iudicii
cum servo nocte processit. ille occisus inspoliatus in solitudine
inventus est, servus iuxta cadaver vulneratus. placuit propinquis
quaeri a servo quern percussorem cognosceret. ille divitem
nominavit et decessit. dives caedis reus est. 24
The affinities of this subject with Contr. VII.5 and X.I are par-
ticularly close. The most difficult questions at issue in Tetra-
logy I are these. First, a rich man known to be an enemy of the
murdered man is liable to be suspected of the crime, given that

the murder was not accompanied by robbery, but is probability a
sufficient ground for conviction?25 Secondly, how far can the
evidence of the slave be relied on, given that slaves are in gen-
eral an unreliable class, and that this one, being dead, cannot
be subjected to further interrogation? Contr. X.I presents us
with a case in which quidam, cum haberet filium et divitem
inimicum, occisus inspoliatus inventus est. The son in this case
does not reckon that the argument from probability alone will
stand up in court, and is looking for further evidence: ...ait:
'accusabo, cum potero' (Contr. X.I thema). He uses it however
when the dives brings a counter-charge, e.g. Contr. X.I.3: quisquis
fuit [sc. percussor], quasi dives spolia contempsit (cf. Tetr. I.
1.4). In Contr VII.5 the main question at issue is how seriously
one should take the testimony of a witness whose reliability must
be doubted, in this case a five-year-old child:
rumor erat de adulterio procuratoris et matris familiae. quodam
tempore pater familiae in cubiculo occisus inventus est, uxor
volnerata, communis paries perfossus; placuit propinquis quaeri a
filio quinquenni, qui una dormierat, quern percussorem cognosceret;
ille procuratorem digito denotavit.
The prosecutor here, as in Tetr. I, has only probabilities from
which to argue, apart from this one dubious piece of testimony.
The second Tetralogy concerns involuntary homicide, a crime
which is frequently mentioned in controversia themes (e.g. Contr.
IV.3 which features a man imprudentis caedis damnatus; ps.-Quint.
CCXLVIII, CCCV), though only as an issue subsidiary to the theme
as a whole. The accidental killing in Tetr. II takes place in the
gymnasium (Tetr. II.1.1): a boy is struck down by a javelin thrown
by an athlete. Athletes do not appear very frequently in Roman
declamation themes, but, for example, in Contr. V.3 two brothers
fight each other to the death as pancratiastae at Olympia. The
theme on which Tetr. II is based is very similar to one which
Pericles is supposed to have debated at great length with Prot-
agoras. According to Plutarch (Per. 36.3), Pericles' son Xanth-
ippus maliciously publicized his father's interest in sophistry:
yap TUVOS axovTb^ uaTaâvTOS 'ETILTUUOV TOV $apaaAuov
al xaTaxTeCvavTOS, nyepav cfAnv avaXwaat yexa npcoTayopou
6uanopouvTa 7i6xepov TO OLKOVTUOV r\ TOV BaAovTa yaAAov fl TOUS

xaxa TOV OPÔTOITOV Aoyov atTUous XPH T°u
Note that the reference to aywvodeTag suggests that the accident

in this case, like the fratricide in Contr. V.3, is supposed to
have happened during organized Games, rather than while the jav-
elin thrower was practising as in Tetr. II; also that the theme,
like those of the veteres controversiae in Suetonius' history of
Roman declamation,26 is supposed to have been based on a real
occurrence.
The third Tetralogy concerns a fight between two drinking
companions (cf. the luxuriosi of controversia themes e.g. Contr.
II.6; IV.1), an old man and a young man. The old man was worsted
in this fight (Tetr. III.1.6; 3.2), and, although he received
medical attention, died a few days later (2.4; 3.5; 4.8). The
young man is charged with murder, but the questions arise whether
he was fighting in self-defence when he struck his victim (2.2ff.;
3.2ff.; 4.2ff.), and whether indeed the physician by his incom-
petence may not have been responsible for the old man's death
(2.4; 3.5; 4.8). Once more there are interesting comparisons to
be made with Roman controversia themes. In Tetr. I we have un-
covered the archetypal 'rich man'; here we have two type figures,
veos and TipeaBus, paired, so that the sophist can differentiate
between the two parties with clarity, and incidentally, when it
suits his case, can arouse prejudice against the accused, as for
example in 1.6: u&peu 6e xal axoAaola Tiapotvwv eus av6pa itpeaBuTnv,
TUTtTGov Te xal Tivuywv ews Tns (JJUXHS aneaTepnaev auTOV. For an
assault by an inebriated young man on an old man, in this case
pitiable for other reasons, see Contr. IV.1:
amissis quidam tribus liberis cum adsideret sepulchro, a luxurioso
adulescente in vicinos hortos abductus est et detonsus coactus
convivio veste mutata interesse. dimissus iniuriarum agit.
The question of the responsibility of a physician in a case where
death is not immediate arises again in Calp. Flacc. Decl. XIII,
where the dispute is between two medici, both of whom claim to
have poisoned the tyrant:
tyrannus suspicatus sibi venenum datum ab eo medico quern in arce
habebat, torsit eum. ille pernegavit. misit ad medicum civitatis.
dixit datum illi ab illo venenum, sed se remedium daturum. dedit

poculum, quo exhausto statim periit tyrannus. contendunt de
praemio.
The dating given by Quintilian and Philostratus, then, for
the introduction of judicial exercises seems to be far too late.
The sophists of the fifth century B.C. were already composing
speeches on fictitious subjects which, if they were stated in bare
outline, as they surely must have been, would have closely re-
sembled controversia themes. Their manner of treating the themes
was different: the Tetralogies are sets of four speeches, two for
the prosecution and two for the defence, presented in the sequence
in which they would be delivered in a real trial, whereas it does
not seem to have been normal to declaim controversiae alternately
for and against, in imitation of the law-courts.27 A declaimer of
the Roman Imperial period would not have given a specific date on
which the crime occurred, or offered his slaves to give evidence
under torture as the sophist does in Tetr, 1.4.8. But the Tetra-
logies have an unquestionable claim to be regarded as the direct
ancestors of controversiae, and it seems as if their lineage
stretched some way further back.
Scholars have been puzzled by the fact that the Tetra-logies
presuppose a legal system akin to, but much more fluid than, that
which obtained in late fifth-century Athens, a system in which,
for example, the participants in Tetr. II have to start from the
assumption that any homicide is punishable, whereas under the
Draconian Code, in which it was accepted that cpovog dxouauos was
not a crime of the same order as cpovos exouauos, it was deemed in
certain circumstances, such as when, as in this case, the accident
took place ev a$Aous, to be no crime at all. 28 Another notable
feature of the Tetralogies is the presence of Ionicisms in their
language. K. J. Dover has suggested as the most probable expla-
nation of them, Tthat the Tetralogies represent the imitation of
an Ionic genre by an Athenian1, pointing out that 'most of the
purely Ionic words - as distinct from those which may as well be
labelled tragic as Ionic - are legal words*.2^ Perhaps, then,
the prototypes of controversiae were first contrived by Ionian
sophists for use in cities where laws such as we find associated
with TexpaAoyuau and controversiae, akin to those of Athens, but

in some ways more backward than them, were in force, all this
in a period when the problems of cities afflicted by insolent
tyrants in their citadels and unreasonably demanding tyrannicides,
so striking an anachronistic feature of the Roman exercises,31 had
recently become matters of consuming interest.32
About the Greek background to the declamatory practices of
his own time Seneca the Elder seems to have had only the vaguest
knowledge. He tells us that declamation was unknown in the time
of Aeschines (Contr. 1.8.16), misinformation most probably derived
from tirades by critics of the schools, such as Petronius puts
into the mouth of one of his characters in Sat. 2, to the effect
that the great figures of classical Greek literature had managed
perfectly well without declamation:
nondum iuvenes declamationibus continebantur, cum Sophocles aut
Euripides invenerunt verba quibus deberent loqui. nondum
umbraticus doctor ingenia deleverat, cum Pindarus novemque lyrici
Homericis versibus canere timuerunt. et ne poetas [quidem] ad
testimonium citem, certe neque Platona neque Demosthenen ad hoc
genus exercitationis accessisse video. 33
Seneca was acquainted with Greek technical names, thesis^1* and
scholastica, for different types of declamation themes (Contr. I
pr. 12); he knew that there were a few standard quaestiones used
by Greek declaimers in controversiae, which the Romans had
T
removedf (submoverunt - Contr. 1.7.12 cf. 1.8.7); he also dates
(in Contr. II pr. 5) the origins of Latin rhetorical education to
the boyhood of Cicero, and yet in Contr. I pr. 12 refers to a
stage in scholastic history before Cicerofs time. But he nowhere
refers explicitly to the debt owed by Rome to Greek educational
methods or shows any sign of having made a clear assessment of it.
As a history of Roman declamation Contr. I pr. 12 can be
criticized on a number of counts. It should first be considered
in relation to Suetonius, De gramm. et rhet. 25.8-9 Brugnoli, an
account which, like Seneca1s, distinguishes three stages in the
rise of Roman declamation, but describes the first two with a
greater wealth of detail.
1st stage
Seneca
quales ante Ciceronem dicebantur quas thesis vocabant.
Suetonius (25.8)
nam et dicta praeclare per omnes figuras, per casus et apologos
aliter atque aliter exponere et narrationes cum breviter ac presse
turn latius et uberius explicare consueverant, interdum Graecorum
scripta convertere ac viros inlustres laudare vel vituperare,
quaedam etiam ad usum conmunis vitae instituta turn utilia et
necessaria turn perniciosa et supervacanea ostendere, saepe fabulis
fidem firmare aut demere quod genus thesis et anasceuas et cata-
sceuas Graeci vocant: donee sensim haec exoluerunt et ad contro-
versiam ventum est. 35
(Cf. Quint. II.1.9: an ignoramus antiquis hoc fuisse ad augendam
eloquentiam genus exercitationis, ut thesis dicerent et communes
locos et cetera citra complexum rerum personarumque quibus verae
fictaeque controversiae continentur?)
2nd stage
Seneca
declamabat autem Cicero non quales nunc controversias dicimus...
controversias nos dicimus; Cicero causas vocabat.
Suetonius (25.9)
veteres controversiae aut ex historiis trahebantur sicut sane non-
nullae usque adhuc aut ex veritate ac re, si qua forte recens
accidisset: itaque locorum etiam appellationibus additis proponi
solebant. sic certe conlectae editaeque se habent ex quibus non
alienum fuerit unam et alteram exempli causa ad verbum referre:
aestivo tempore adulescentes urbani cum Ostiam venissent litus
ingressi, piscatores trahentes rete adierunt et pepigerunt bolum
quanti emerent. nummos solverunt. diu expectaverunt dum retia
extraherentur. aliquando extractis piscis nullus infuit sed
sporta auri insuta. turn emptores bolum suum aiunt, piscatores
suum. venalicius cum Brundisi gregem venalium e navi educeret,
formoso et pretioso puero quod portitores verebatur bullam et
praetextam togam inposuit. facile fallaciam celavit. Romam
venitur, res cognita est, petitur puer quod domini voluntate
fuerit liber in libertatem. olim eas tappellationes Graeci syn-
tasist vocabant... ^6*
3rd stage
Seneca
...quales nunc controversias dicimus...hoc enim genus materiae,
quo nos exercemur, adeo novum est, ut nomen quoque novum sit.
controversias nos dicimus...hoc vero alterum nomen Graecum quidem,
sed in Latinum ita translatum, ut pro Latino sit, scholastica,
controversia multo recentius est, sicut ipsa Meclamatio1 etc.
Suetonius (25.9)
..,mox controversias quidem [sc. vocabant] sed aut fictas aut
iudiciales.
The exact relationship between the Senecan and Suetonian
accounts is unclear. The main doubtful points are these. First,

did Seneca1s source give in full the long list of progymnasmata
alleged by Suetonius to have been dominant in the first stage, or
did it just mention the thesis (plus perhaps a reference to
communes locos and cetera citra complexum rerum personarumque in
general, as in Quint. II.1.9), so that what Suetonius is giving us
is post-Senecan elaboration? Secondly, were the verbatim quo-
tations from veteres controversiae to be found in Seneca's source?
If so, they could explain in what way Seneca might have imagined
the first Roman judicial exercises to have differed from the later
controversiae, but we cannot be sure of this. What the two ac-
counts clearly do have in common is a belief that at first Roman
rhetorical education did not deal with themes of the unodeats
type; that when judicial exercises were first used in Rome they
were of a type different from the controversial it was not until a
later stage, we are to suppose, that the controversia was intro-
duced.
For the pre-Ciceronian period, it is true, there is precious
little documentation, and it may seem conceivable that the pro-
gymnasmata of Greek primary education were introduced to Rome
before the more advanced exercises. But it must be doubted
whether, once Hellenizing had begun, Roman practicality would have
wasted much patience on the thesis, the chria and the fable, be-
fore demanding instruction in techniques of persuasion useful for
senatorial debates and the law-courts. Besides, we have seen the
evidence of the De oratore as to how little attention was normally
paid to the thesis by rhetoricians in the period immediately be-
fore Cicero's time.37 One might be less sceptical about the
Senecan-Suetonian belief in the early dominance of the thesis etc.
in Roman education, if the later parts of the accounts corresponded
closely to the truth, but, as we shall see, they seem wrong in
several respects.
First, Seneca's statement that Cicero did not declaim theses
is erroneous. Secondly, themes of the kind later called contro-
versiae were known to Cicero, and one such is explicitly said by
him to have been in use at the dramatic date of the De oratore.
Thirdly, public declamation by adult schoolmasters had been known
in Rome at a period earlier than Seneca's generalization about the

novelty of declamatio suggests.
Cicero's attitude towards the thesis altered somewhat in the
course of his life, but was always favourable. In his early De
inventione he criticized Hermagoras for his suggestion that the
treatment of wide abstract issues was part of the orator's prov-
ince, but he took this line not out of any disrespect for philo-
sophy, rather for the opposite reason, because he believed that
such matters were too weighty to be treated as a subordinate part
of an orator's work: nam quibus in rebus summa ingenia philo-
sophorum plurimo cum labore consumpta intellegimus, eas sicut
aliquas parvas res oratori attribuere magna amentia videtur (De
inv. 1.6.8). Later he came to regard his De inventione as an
immature work (De or. 1.2.5), having meanwhile arrived at a much
more glowing estimate of the orator's competence and r61e in so-
ciety. In his mature rhetorical works he encourages the orator to
derive all the benefit he can from the schools of philosophy, and
near the beginning of the Orator he makes the significant claim:
fateor me oratorem, si modo sim aut etiam quicumgue simf non ex
rhetorum officinis sed ex Academiae spatiis extitisse... (Or.
3.12). In particular he sees that it is highly beneficial to the
orator if he can penetrate to the basic abstract question (infinita
quaestio = %eobs) underlying the complex facts of the case about
which he has to speak:
...orator, non ille vulgaris sed hie excellens, a propriis
personis et temporibus semper, si potest, avocet controversiam;
latius enim de genere quam de parte disceptare licet, ut quod in
universo sit probatum id in parte sit probari necesse; haec igitur
quaestio a propriis personis et temporibus ad universi generis
rationem traducta appellatur §e:aos. (Or. 14.45f.)
Besides requiring his ideal orator to be something of a philo-
sopher, Cicero had come to demand eloquence of the philosopher
too. He describes at the beginning of the Tusculan disputations
(written in 45 B.C.) his own efforts towards the creation of a
Latin philosophical language; in particular he tells us that he
had taken to discoursing in the manner of the Greeks on philo-
sophical subjects, in the company of friends on his Tusculan
estate. Jokingly he refers to the copious and ornate philosophical
discourses produced on these occasions as the declamation of his

old age:
hanc enim perfectam philosophiam semper iudicavi, quae de maximis
quaestionibus copiose posset ornateque dicere, in quam exerci-
tationem ita nos studiose dedimus, ut iam etiam scholas Graecorum
more habere auderemus: ut nuper tuum post discessum in Tusculano,
cum essent complures mecum familiares, temptavi quid in eo genere
possem. ut enim antea declamitabam causas, quod nemo me diutius
fecit, sic haec mihi nunc senilis est declamatio. (Tusc. disp. I.
4.7)
One might be tempted to dismiss this account of Cicero's scholae
at Tusculum as a fiction, an artificial setting of the scene in
preparation for the ensuing discourses, but it is clear from other
sources that in later life Cicero really had come to believe in
the declamation of fteaeus as an educational method preferable to
those used by the rhetoricians of his time.^8 In a letter to his
brother (Ad Q. fr. III.3.4) he explicitly contrasts his own prefer-
ence for the $eaus with his nephew's liking for the less abstract
types of rhetorical exercise:
Cicero tuus nosterque summo studio est Paeoni sui rhetoris,
hominis, opinor, valde exercitati et boni; sed nostrum instituendi
genus esse paulo eruditius et dexuKWTepov non ignoras.
This letter dates from 54; in another written in March 49 he actu-
ally lists a number of %eoeLg (in Greek) which he was using to
exercise himself at that time. He explains:
...ne me totum aegritudini dedam sumpsi mihi quasdam tamquam
§eaeus quae et TtoALTOxau sunt et temporum horum, ut et abducam
animum ab querelis et in ipso de quo agitur exercear. (Ad Att. IX.
4.1)
The questions he was setting himself all concerned the problem of
how to act under a tyranny.
Seneca's flat denial in Contr, I pr. 12 that Cicero did not
declaim the kind of exercises called theses is therefore one of
his most unfortunate errors. If, on the other hand, he had said
that theses were not in common use in the average rhetorical
school of Cicero's time, that would probably have been a true
statement of the facts.
Seneca was wrong in suggesting that the controversia was a
type of exercise as novel as its newly coined name.^ We have
looked at the evidence for the use of similar themes remarkably
early in the history of Greek education: the introduction to Rome
of judicial themes of the type popular under the Empire has to be

dated at least as early as Cicero's boyhood. In the same letter
to his brother in which he expresses his preference for the deaus
(Ad Q. fr. III.3.4) Cicero makes some interesting remarks about
the type of education which his nephew was receiving at the school
of Paeonius:
...neque ego impediri Ciceronis iter atque illam disciplinam volo
et ipse puer magis illo declamatorio genere duci et delectari
videtur. in quo quoniam ipsi quoque fuimus, patiamur ilium ire
nostris itineribus; eodem enim perventurum esse confidimus. (ibid.
5)
Here we learn that CiceroTs own education had been along much the
same lines as the training in declamation enjoyed by his nephew,
which dealt with themes less abstract than the deats. However
much terminology had changed, the declamations of Cicero's youth,
to which he refers in Tusc. disp. 1.4.7: ut antea deelamitabam
causaSf quod nemo me diutius fecit... and Brut. 90.310: com-
mentabar declamitans - sic enim nunc loquuntur - saepe cum M.
Pisone et cum Q. Pompeio aut cum aliquo cottidie, idque faciebam
multum etiam Latine, sed Graece saepius... must have been funda-
mentally similar to those which Q. Cicero's son, along with most
students and rhetoricians of the very late Republic, preferred to
§£OZL£, though probably they were more varied in type and more
rigorously treated than the themes declaimed in the school of
Paeonius.40 Striking evidence that there was, in fact, a remark-
able degree of continuity in educational practice from the early
first century B.C. through to the time of Quintilian, comes in the
De oratore, where Cicero represents M. Antonius, grandfather of
the triumvir, as recommending a training in the techniques of
judicial oratory more rigorous than that provided by the easy
exercises which were generally all the practice which students of
rhetoric were given:
hoc in ludo non praecipitur; faciles enim causae ad pueros
deferuntur; lex peregrinum vetat in murum ascendere; ascendit;
hostis reppulit: accusatur. nihil est negoti eius modi causam
cognoscere; recte igitur nihil de causa discenda praecipiunt;
[haec est enim in ludo causarum formula fere.] at vero in foro
tabulae testimonia. pacta conventa stipulationes, cognationes
adfinitates, decreta responsa, vita denique eorum, qui in causa
versantur, tota cognoscenda est. (11.24.100)
Antonius1 complaints foreshadow the criticism so often levelled

against the rhetoricians of the Empire,41 that they were not pro-
viding adequate training for the forum; the particular exercise
quoted is also given by Quintilian as an example of a type of
controversia in which one cannot quibble over the interpretation
of the law, but has to consider solely the character and motives
of the individual who has infringed it:
...tale genus controversiarum in quo nullum argumentum est quod ex
lege ipsa peti possit, sed de eo tantum de quo lis est quaerendum
est. fperegrinus si murum escenderit capite puniatur. cum hostes
murum escendissent, peregrinus eos depulit: petitur ad supplicium.T
(Quint. VII.6.6)
This theme, it should be noted, contains, like the typical
controversiae of the Empire, nothing but a bare statement of the
facts of the case; it includes none of the inessential scene-
setting that distinguishes the veteres controversiae which
Suetonius assigns to the second stage of Roman declamation. When
these Italian realist themes, set in Ostia and Brundisium and pur-
porting to be based on real events, were invented, and how
seriously they rivalled the starker type of theme in popularity at
any stage, are intriguing but unanswerable questions. The problem
of the ownership of treasure fished out of the sea, which is at
issue in the first theme, was a legal conundrum known to Plautus,
who exploits its possibilities in the long dispute over the rope
(with trunk attached) from which the Rudens takes its name
(938ff.). Note particularly the legalistic quibbling in 975ff.:
GR. mare quidem commune certost omnibus. TR. adsentio:
qui minus hunc communem quaeso mihi esse oportet vidulum?
in mari inventust communi. GR. esne inpudenter inpudens?
nam si istuc ius sit quod memoras, piscatores perierint.
Plautus must have expected his audience to appreciate these soph-
istries to some extent, and this fact might seem to suggest that
the topic disputed was a familiar feature of Roman education even
in Plautus1 time. However, as the Rudens was an adaptation of a
Greek original by Diphilus (see Rud. 31f.), perhaps we should de-
duce no more than that it was discussed in Hellenistic schools:
Plautus1 audience may have been expected to laugh at the scene as
illustrating a typical ineptum.. .negotium et Graeculum.1*2- The
1
origins of Suetonius veteres controversiae must remain a mystery.
They may well be genuine early exercises: the central theme of the
first must have been debated in the Hellenistic world; in the
first Tetralogy of Antiphon we have noted the mention of a
specific date (ALUTtoAeuous 4.8) not essential to the main issue of
the case, like aestivo tempore in the first vetus controversia.
On the other hand, one also has to consider the possibility that
they were antiquarian fabrications of some quite late rhetor, who,
like Quintilian (II.10.4), would have liked to see a greater el-
ement of realism in declamation themes.
Whatever the truth of this matter, Seneca is clearly mis-
leading when, referring to the modern controversia he says, hoc
enim genus materiae, quo nos exercemur, adeo novum est, ut nomen
quoque novum sit (Contr. I pr. 12). His denial that Cicero de-
claimed the sort of exercises now called controversiae also seems
to conflict with some other evidence, though it may well be that
Cicero, when declaiming, used a wider variety of UTtodeaeus than
found favour in the schools of the elder Seneca's time. De or.
11.24.100 shows that Cicero knew of at least one school exercise
which Quintilian could class as a controversia; it is fairly safe
to infer that he had been set such exercises as a boy, even if
later he came to prefer the tougher training methods which he
makes M. Antonius advocate in the De oratore. In addition, Seneca
the Elder himself knew a tradition, conflicting with that set out
in Contr. I pr. 12, that Cicero had once declaimed a controversia
very like Contr. 1.4:
color pro adulescente unus ab omnibus, qui declamaverunt, intro-
ductus est: fnon potui occidere', ex ilia Ciceronis sententia
tractus, quam in simili controversia dixit, cum abdicaretur is,
qui adulteram matrem occidendam acceperat et dimiserat: ter non
<... (Contr. 1.4.7)
Perhaps we should not lay too much weight on this piece of school
tradition, originating as it may in the realms of Ciceronian
pseudepigrapha, or simply in some rhetorician's fanciful wish to
ascribe a time-honoured color to the fount of eloquence himself.
But it is not totally implausible that some memory of the nature
of Cicero's declamations, at least his late ones, should have been
preserved; nor is it inconceivable that, in order to indulge the
tastes of the younger generation, Cicero declaimed controversiae
(though not necessarily only controversiae) in the company of

Dolabella, Hirtius, and Pansa.
Certainly controversiae were not the only sort of judicial
declamations to find favour in the late Republic. Sometimes poli-
ticians would practise all the possible variations of invective
for use against their opponents. Cicero jeers at the way that
Antony 'ipse interea XVII dies de me in Tiburtino Scipionis declami-
tavit1... (Phil. V.19). Alternatively they might declaim in order
to prepare themselves for rebutting their opponents in an important
speech: Cn. Pompeium quidam historici tradiderunt sub ipsum civile
bellum quo facilius C. Curioni promptissimo iuveni causam Caesaris
defendenti contradiceret repetiisse declamandi consuetudinem
(Suet. De gramm. et rhet. 25.4 Brugnoli).^ Cicero's declamations,
in line with M. Antonius' requirements in De or. II.24.99f., may
likewise often have been concerned very specifically with contem-
porary legal issues and politics on which he might be required to
speak. As we shall see, the word commentari in Republican usage
can mean 'to prepare one's material (orally) for a forthcoming
speech', as well as 'to declaim' in general, and to Cicero, per-
haps, there was not always a clear distinction between these two
activities. Or sometimes he may have harked back to past events
and contrived the ideal speech which he might have delivered in
the circumstances: bear in mind that some of his most famous
speeches were never actually delivered and could therefore in a
sense be described as rhetorical exercises. There may be a hint
that his late uno-deaeus included treatments of contemporary prob-
lems in Ad Att. XIV.22.1 where he sets out the attitudes of
Hirtius and his friends towards Caesar in terms of what he calls
a
autem hanc habent eamque prae se ferunt, clarissimum

<virum> interfectum, totam rem publicam illius interitu per-
turbatam, inrita fore quae ille egisset simul ac desisteremus
timere, clementiam illi malo fuisse, qua si usus non esset, nihil
ei tale accidere potuisse.
Just as the deaeus which he set himself in later life included
some which helped him to thrash out the ethical problems at issue
in the politics of his time (cf. Ad Att. IX.4.1), so his unodeaeus
may sometimes have been of a kind which would help him to organize
rigorously his private deliberations about people and events at

times of crisis, enabling him, for example, to examine all facets
of the question where to place one's allegiance after the death of
Caesar.
However, even if the declamations of Cicero and other eminent
men of the late Republic were sometimes of a different, and more
serious, order than those of the Augustan scholastics, Seneca's
history of declamation has to be faulted for obscuring a consider-
able degree of continuity which must have existed in rhetorical
education between Republic and Empire. For there can be no doubt
that, whatever Cicero might say or do, throughout this period the
pupils of humble rhetoricians continued to debate the legal rights
of the foreigner who had scaled the city walls in order to repel
the invader.
Similarly, Seneca is misleading when he claims that the prac-
tice called declamatio was a very recent innovation: modo nomen
hoc prodiit; nam et studium ipsum nuper celebrari coepit: ideo
facile est mihi ah incunabulis nosse rem post me natam (Contr. I
pr. 12). Leaving aside the question of the Greek background, it
seems that the practice of regular declamation by adults in Rome
to considerable audiences antedates the use of the word declamatio
to describe it. Suetonius writes about the rhetorical teaching of
M. Antonius Gnipho, a grammaticus who taught primum in Divi Iuli
domo pueri adhuc, deinde in sua privata in the following terms:
docuit autem et rhetoricam ita ut cotidie praecepta eloquentiae
traderet, declamaret vero non nisi nundinis. scholam eius claros
quoque viros frequentasse aiunt, in his M. Ciceronem, etiam cum
praetura fungeretur. (De gramm. et rhet. 7.3 Brugnoli)
Cicero was praetor in 66 B.C. It may be that orators, as opposed
to rhetoricians, never declaimed in public as early as this, and
that rhetoricians did not admit the public to their schools as
often as they were later to do, but Seneca is over-simplifying the
facts of the matter when he calls declamation 'rem post me natam1.
Contr. I pr. 12, then, contains several major distortions of
the history of Roman declamation. On the other hand the lexico-
graphical data in it about the development of declamatory termin-
ology from Cicero's to Seneca's own time, seem to be sound, as far
as can be judged from independent investigation of Cicero's usage.
Causa is the term used by Cicero when referring to the decla-

mations of his early years in Tusc. disp. 1.4.7: ut antea
declamitabam causas, quod nemo me diutius fecit, sic haec mihi
nunc senilis est declamatio. These causae9 presumably to be
identified with the declamations in both Greek and Latin to which
he refers in Brut. 90.310, will often, perhaps most often, have
been judicial exercises, but not always. Causa was used by Cicero
and others as a translation of Hermagoras1 term UTIOSEQLS: Quinti-
lian makes the equation succinctly in III.5.7: finitae autem sunt
[sc. quaestiones"] ex complexu rerum personarum temporum ceterorum-
que: hae uuoêaeus a Graecis dicuntur, causae a nostris. Now,
according to Hermagorean theory, uitodeae us /causae could be of
three different types: judicial, deliberative, or epideictic:
quales sunt, quae in litibus, quae in deliberationibus versantur,
addat, si quis volet, etiam laudationes (De or. 11.16.65). It is
hinted here that in practice the terms were rarely used with ref-
erence to epideictic, but later in the De oratore Cicero gives an
example of a deliberative causa: placeatne a Karthaginiensibus
captivos nostros redditis suis recuperari (De or. III.28.109). So
it seems that the exercises which Cicero called causae could have
included themes of the suasoria, as well as of the controversia,
type. The invectives which Antony and others practised could
also, theoretically at least, have been so described.
The word controversia is used by Cicero in a variety of tech-
nical and non-technical senses,4^ but not, as far as one can make
out, in the precisely limited sense: 'judicial theme for decla-
mation1. The primary meaning of controversia is 'quarrel1: nulla
controversia mihi tecum erit9 says Megadorus in Plautus' Aulularia
(261); a quarrel of the type customary among shepherds is called a
controversia by Cicero in Pro Cluentio 59.161. In technical con-
texts controversia is sometimes used by Cicero as a rendering of
the Greek dycpLaBnTnaus, that is, the fundamental cause for dispute
underlying a deaus or uitodeaus. Compare Cic. De inv. 1.6.8 =
Hermagoras fr. 6a Matthes: Hermagoras...qui causam esse dicat rem,
quae habeat in se controversiam in dicendo posi tarn cum personarum
certarum interpositione, with Theon, Prog. Rh. Gr. 11.120.13
Spengel = fr. 6d Matthes: deaus eaxlv
-upoaojioov wptayevwv waL laans nepuaTaaecos. . .

Controversia in such contexts is something underlying, or con-
tained within, a causa, but elsewhere Cicero uses it, like causa,
as equivalent to UTtoSeaus. In De or. III.28.109 he outlines the
theory of the Academics and Peripatetics about §eaeus and uito-
as follows:
omnem civilem orationem in horum alterutro genere versari: aut de
finita controversia certis temporibus ac reis; 45 * hoc modo:
placeatne a Karthaginiensibus captivos nostros redditis suis
recuperari? aut infinite de universo genere quaerentis: quid
omnino de captivo statuendum ac sentiendum sit? atque horum
superius illud genus causam aut controversiam appellant eamque
tribus, lite aut deliberatione aut laudatione, definiunt; haec
autem altera quaestio infinita et quasi proposita consultatio
nominatur.
Causa and controversia are explicitly equated in this analysis of
the types of rhetoric. What Cicero does not do is to restrict the
use of controversia to the judicial type of unoêaus, still less
to the type of fictitious U7i6$eaLS set for declamation. So it
seems that Seneca was right to regard the usage of controversia
accepted in his time as non-Ciceronian. It also seems plausible
that, as Suetonius appears to have suggested (De gramm. et rhet.
25.9 Brugnoli), when controversia was first used with reference to
judicial declamations, it had to be qualified by the adjectives
ficta or iudicialis, if confusion was to be avoided.^ The word
still retained some of its ambiguity in Quintilian's time: verae
fictaeque controversiae are distinguished in II.1.9.
The term scholastica, according to Seneca controversia multo
recentius, is indeed not to be found in any work earlier than
Senecafs own. It is unmistakably used as a feminine noun in
Contr. II.3.13, III pr. 12 and X.5.12. The dative and ablative
plurals scholasticis are found in a number of other contexts,1*7
and as Bonner has pointed out 48 there is no reason to assume, as
lexicographers have generally done, that these are neuter, rather
than feminine, plurals. Bonner surmises that the term was a
Latinization of a Greek word r\ axoAaaxtKn, formed from axoAaaxLHog
as pnTOpUMn developed from pnxopLKOS, but evidence is lacking for
this development. It seems relevant to compare the term suasoria,
cognate with the adjective suasorius.
To turn now to Seneca's statement that declamatio itself was

not found used by authors earlier than Cicero ( ? ) ^ and Calvus, it
should be noted that the quotation he gives to illustrate Calvus'
usage, ait enim declamare iam se non mediocriter, dicere bene, is
evidence for the use of the verb declamare, not of the noun
declamatio. Seneca has sacrificed lexicographical precision for
the sake of euphony when he states that ipsa declamatio, rather
than declamare, was not to be found used by early authors.50 Now,
whatever else the quotation may illustrate, it does not, assuming
it is by Calvus, represent the first ever use of the verb de-
clamare. Calvus was two years old when in 80 B.C. Cicero taunted
his opponent in the Pro Roscio Amerino (29.82) for making alle-
gations, quae mihi visus est ex alia oratione declamare quam in
alium reum commentaretur. Here, as in other early contexts,
declamare probably means simply 'to rant', and commentari is best
taken to mean 'prepare a speech (for a forthcoming trial)', so
that the implication of the whole would be: 'he seemed to me to be
ranting on about charges extracted from some other speech, which
he was preparing for use against some other defendant'. It is
true that commentari has sometimes, in Cicero's works, much the
same meaning as declamare was to have in the Empire, but it was a
word with quite a wide range of meanings. It is used with refer-
ence to other arts besides rhetoric, even the gladiatorial art, to
mean 'remind oneself of one's skills' i.e. 'to practise', e.g. De
or. III.23.86: magister hie Samnitium summa iam senectute est et
cotidie commentatur; it can even be used of the self-training by
which one prepares for death (Tusc. disp. 1.31.75). Where rhet-
oric is concerned it can have two different, though related, mean-
ings. Sometimes it serves as a translation of yeAexav, the sur-
prisingly imprecise word used by the Greeks throughout antiquity
to mean 'declaim rhetorical exercises': this is how Cicero uses it
in Brut. 89.305: ...et scribens et legens et commentans oratoriis
tantum exercitationibus contentus non eram; and ibid. 90.310:
commentabar declamitans - sic enim nunc loquuntur saepe cum M.
Pisone et cum Q. Pompeio. Elsewhere, however, he uses it to refer
to the preparation of real speeches, as for example in Brut.
22.87:
unum quasi comperendinatus medium diem fuisse, quem totum Galbam

in consideranda causa componendaque posuisse; et cum cognitionis
dies esset et ipse Rutilius rogatu sociorum domum ad Galbam mane
venisset, ut eum admoneret et ad dicendi tempus adduceret, usque
ilium, quoad ei nuntiatum esset consules descendisse, omnibus ex-
clusis commentatum in quadam testudine cum servis litteratis
fuisse, quorum <alii> aliud dictare eodem tempore solitus esset.
interim cum esset ei nuntiatum tempus esse, exisse in aedis eo
colore et eis oculis, ut egisse causam, non commentatum putares.
In Rose. Amer. 29.82 commentari is used in association with oratio
and reus, so it seems best to take it there too as referring to
preparation for a real speech, and not to assume that declamare,
used in the same sentence, need have had strong scholastic over-
tones as early as 80 B.C. 51
Declamare is used by Cicero of shouted invective in general,
e.g. Verr. II.4.149: qui pro isto vehementissime contra me
declamasset (70 B.C.); Ad fam. III.11.2: ne in quemvis inpune
declamari liceret (50 B.C.); and also of exercises intended to
improve the speaker's vocal powers rather than his skill in argu-
mentation: De fin. V.2.5: quo in loco ad fluctum aiunt declamare
solitum Demosthenem (45 B.C.). Similarly declamatio is sometimes
used by Cicero to mean just 'abusive ranting': Mur. 21.44: non
declamatio potius quam persalutatio (63 B.C.); 5 2 it is used to
mean 'elocution practice' by the Auctor ad Herennium (III.12.20).
Declamare seems to have taken on the absolute sense illus-
trated by Seneca's quotation from Calvus, 'to declaim rhetorical
exercises (as opposed to true speeches)', only in the last years
of Cicero's life. The advent of the new usage is heralded by the
first appearance of declamator in 54 B.C.: non vobis videtur cum
aliquo declamatore...disputare? (Plane. 34.83), 53 and disparaging
remarks about illo declamatorio genere (Ad Q. fr. III.3.4 - also
54 B.C.), clearly with reference to exercises less abstract than
§eaets. When, however, in the Brutus, composed in 46, Cicero uses
the iterative declamitare in conjunction with commentari, he still
apologizes for his neologistic expression: commentabar declamitans
- sic enim nunc loquuntur (90.310). From these examples it is
evident that Cicero was not responsible for giving declamare and
related terms their new absolute sense. (It is interesting that
Seneca should cite Calvus in this connection.) In his latest
letters and treatises Cicero uses declamare and declamatio in

their new sense, despite his distaste for the new fashion for de-
claiming when carried to extremes: haud amo vel hos designates,
qui etiam declamare me coegerunt, ut ne apud aquas quidem
acquiescere liceret (Ad Att. XIV.12.2 - 44 B.C.). He evidently
recognized that declamatio had come to stay, though he would have
been happier if the instituendi genus...deTuxwiepov which he pre-
ferred (Ad Q. fr. III.3.4) and which he adopted as his senilis
declamatio (Tusc. disp. 1.4.7), had appealed more to the younger
generation. The statement in Contr. I pr. 12 that the usage of
declamatio and declamare current in Seneca's day was of recent
origin, seems therefore to have been quite correct, as does the
account of the order in which new Latin names - causa, contro-
versia, scholastica - were given to non-abstract declamatory
exercises.
Contr. I pr. 12 presents sound lexicographical material, but
this is set within a distorted picture of the whole development of
Roman declamation. In particular we need some explanation why, in
Seneca's account, the thesis is represented as exclusively pre-
Ciceronian, in flagrant contradiction to the facts, and the contro-
versia as a very recent innovation and as something quite distinct
from the causa as Cicero knew it.
The idea that the dominant pre-Ciceronian exercise had been
the thesis could have arisen out of a false inference from De or,
III.27.107, where Cicero says of the abstract types of exercise:
quae exercitatio nunc propria duarum philosophiarumf de quibus
ante dixi, putatur, apud antiquos erat eorum, a quibus omnis de
54
rebus forensibus dicendi ratio et copia petebatur Maybe
Seneca's source affirmed the precedence of deaets positively, like
Quintilian in II.1.9: ...antiquis hoc fuisse...genus exercita-
tionis ut thesis dicerent et communes locos et cetera citra com-
plexum rerum personarumque..., rather than denying, as Seneca
does, that Cicero declaimed them.55
The information (correct) that Cicero called his rhetorical
exercises causae could have been gained directly from Tusc. disp.
1.4.7,56 one of Cicero's most explicit references to his prolonged
self-education in early manhood. The denial that Cicero declaimed
controversiae, though possibly based on genuine recollection that

some of Cicerofs UTtoSeaeus were different from those of later
schoolmen, may however simply be another inference from Cicerofs
use of the term causa in Tusc. disp. 1.4.7, false this time. Note
how, in the problematic opening sentences of Contr. I pr. 12,
no clear distinction is drawn between innovation of genre and
innovation of terminology:
declamabat autem non quales nunc controversias dicimus, ne tales
quidem, quales ante Ciceronem dicebantur, quas thesis vocabant.
hoc enim genus materiae, quo nos exercemur, adeo novum est, ut
nomen quoque eius novum sit. controversias nos dicimus: Cicero
causas vocabat.
Perhaps one should detect here a simple inability on Seneca's part
to conceive that Cicero might have given the name causa to the
exercise which later generations called controversia.
But the lack of any reference to Cicero's late philosophical
declamations cannot be explained as due to false inference. Nor
can it be convincingly argued that Seneca's source cannot have
known the passages in which Cicero refers to his decrees. Tusc.
disp. 1.4.7, which tells us about Cicero's causae, also tells us
of his senilis declamatio on philosophical themes. The letters
Ad Atticum, although maybe not available when Seneca was writing,57
were sufficiently accessible in the 30s B.C. for Nepos to read
them (Nepos, Att. 16.2ff.).
To help explain why no reference to Cicero's taste for veaeus
found a place in any of the ancient accounts of the history of
Roman declamation, it may be relevant to compare Philostratus'
sketch of the rise of the two sophistics. According to Philo-
stratus, the first sophistic dealt exclusively with xa cptAo-
crocpouueva; themes pertaining to particular men and situations were
the innovation of Aeschines at the beginning of the second. No
mention is made of the Tetralogies ascribed to Antiphon, or of the
tradition known to Plutarch (Per. 36.3) that Pericles had indulged
in dispute with Protagoras about the accidental death of a named
individual; nor is there any reference to the continued inclusion
of the themes here assigned to the first sophistic, among the pro-
gymnasmata of later Greek education. Philostratus' account seems
governed by a presupposition that the simple and abstract must
precede the complex and concrete. There is no place in it for

recognition that the two main species of sophistical themes might
exist side by side. So it is with Contr. I pr. 12: the abstract
thesis - regarded by most rhetoricians of the time as a simple
preparatory exercise - is assigned to the first stage of Roman
declamation, and we are left to suppose that the more complex
causa and its successors superseded it completely.58 Thus Roman
education, like Greek, is made to go through a first sophistic
before it can reach its second phase. The uncomfortable evidence
of such passages as Tusc. disp. 1.4.7, on CiceroTs taste for
philosophical declamation, and De or. 11.24.100, on the early use
of a controversia theme, had simply to be ignored, if there was to
be no disruption of a satisfying notion of steady progress along
lines familiar from Greek literary history.
3 THE DECLINE OF RHETORIC IN THE EARLY EMPIRE
To Seneca the Elder, the time of Cicero seemed the golden age of
Roman eloquence; its glory had lingered on for perhaps one more
generation - he was grateful for the illumination shed on his
studies by the brilliant minds 'born at that time1 (Contr. I pr.
7) l - but since then rhetoric, in his view, had shown a rapid de-
cline. The causes of this decline were to be the topic of much
discussion throughout the first century A.D.,2 just as the decay
of Greek eloquence after the downfall of the democracies appears
to have been the subject of earlier scholarly debate.3 Seneca's
is one of the earliest of the Roman discussions, and it is
interesting to consider which of the explanations for the decline
that were to become more or less standard, he does, or does not,
give.4
In Contr. I pr. 7 he offers us three alternative explanations:
in deterius deinde cotidie data res est, sive luxu temporum -
nihil enim tarn mortiferum ingeniis quam luxuria est - sive, cum
praemium5* pulcherrimae rei cecidisset, translatum est omne
certamen ad turpia multo honore quaestuque vigentia, sive fato
quodam, cuius maligna perpetuaque in rebus omnibus lex est, ut ad
summum perducta rursus ad infimum, velocius quidem quam ascend-
erant, relabantur.
He then proceeds to give a highly coloured description of contem-
porary decadence, which owes much to the cliches of declamatory
convicium saeculi (Contr. I.pr. 8-10). ^
Luxus temporum, if we are to judge from this description,
consisted in the dedication of modern youth to bone idleness
(somnus languorque) or else, what was worse, to malarum rerum
industria and the obscena studia of singing and dancing; the
fashion was for effeminate behaviour nisi in libidine, and Seneca
finds it unsurprising that no orators were emerging among the
younger generation: in hos ne dii tantum mali ut cadat eloquentia:
quam non mirarer, nisi animos in quos se conferret eligeret.
The concept that good oratory depended on good morals was an
The decline of rhetoric in the early Empire 133
old one, pin-pointed in the requirement of the elder Cato, which

Seneca quotes in Contr. I pr. 9, that an orator should be vir
bonus as well as dicendi peritus.7 According to the younger
Seneca (Ep. mor. 114.1) the saying talis hominibus...oratio qualis
vita originated as a Greek proverb, and was often to be heard on
the lips of his contemporaries; in his view it was unquestionable
that the prevalent style of oratory at any time was a reflection
of the morals of the community at large (ibid. 2) so that:
argumentum est luxuriae publicae orationis lascivia, si modo non
in uno aut in altero fuit, sed adprobata est et recepta.
Indeed, contemporary luxury was again and again blamed in the
ancient world for the decline of all the arts, not just oratory.
Ps .-Longinus, referring to yeyaXas (puaets in general, regards their
non-appearance as the result of a combination of (puAoxpnuaTua and
the cpLXr)6ovLa which enslaved and submerged the lives of its devo-
tees (Subl. 44.6). The debasement of poetry was associated by
satirists with luxury and effeminacy (e.g. Pers. Sat. 1.32ff.;
98ff.). A character in Petronius' Satyr icon attributes the de-
cline of the sculptor's art and all the branches of philosophy, as
well as the decline of oratory, to the low moral standards of his
contemporaries vino scortisque demersi (§88). In addition, it was
a commonplace of historians, stated for example in Livyfs preface
(9ff.), that political decadence was the consequence of increasing
luxury and avarice.
Luxus temporum is Seneca's first explanation for the decline;
his second: sive, cum praemium pulcherrimae rei cecidisset, trans-
latum est omne certamen ad turpia multo honore quaestuque
vigentia,Q is expressed in somewhat veiled language, but it seems
certain that the allusion is to changed political conditions.
That the pulcherrima res is eloquentia is clear enough, but what
did Seneca consider to have been its 'reward', and what were the
turpia multo honore quaestuque vigentia to which men were now
devoting their energies? The rewards of eloquence as Cicero saw
them are conveniently listed in Pro Caelio 19.46. Arguing that an
orator's training is very hard work, Cicero asks: an vos aliam
causam esse ullam putatis cur in tantis praemiis eloquentiae,
tanta voluptate dicendi, tanta laude, tanta gratia, tanto honore,
tarn sint pauci semperque fuerint qui in hoc labore versentur?^

The rewards mentioned here surely sum up what the writers have in
mind in discussions of the decline of eloquence where the loss of
praemia/e%a%\a is referred to. Besides the elder Seneca, Tacitus
and ps.-Longinus touch on this topic.
According to ps.-Longinus, there was a popular view sometimes
regarded as hackneyed (lucrxeuxeov exeuvtp xtp dpuXouyev^; we are
asked in Subl. 44.2), that great oratory could not exist outside
free, republican states, indeed that democracy was required for
the fostering of all greatness: f) 6riyoxpaxla x&v yeyaXwv dyadn
xudrivos. The idea was an old one: Aristotle had associated the
development of formal rhetoric with the rise of democracies in
Sicily (Cic. Brut. 12.46). Cicero, after reporting this in his
Brutus, notes how conspicuously lacking in orators the non-
democratic Greek cities had been, and in De or. 1.8.30 he again
remarks on the way that eloquence flourished in omni libero populo.
Similar observations must have been made in a lost part of Tacitus1
Dialogus, for after a lacuna we find in Dial. 36ff. criticism of
the assumption that the conditions which produced great oratory
were .necessarily a good thing. It is in summarizing the common-
place political explanation for the decline of rhetoric that ps.-
Longinus refers to frewards1:
4 xe yap,
y , cpncrLv,
p , oxavfi xa cppovnyaxa
p xwv yeyaXocppovwv
yp n eXeu-
depua xall emeXTtlaau,
l l d'ya
xal ' 6teyeCpeLV
C xo TtpoSuyov xns Ttpos dXXn-
Xous epo6os xal xns itepl xd itpwxeua cptXoxuylas. exu ye ynv 6ua xd
itpoxelyeva ev xats TioXtxelaus e it a % X a exdaxoxe xd (Jjux^d ftpo-
xepnyaxa xwv pnxopwv yeXexwyeva dxovaxaL xau oEov exxpuêxau xal
XOLS Trpdyyaat xaxd xo etxos eXeudepa auvexXdyTieu. (Subl. 44.2-3)
(noXuxelaus should be taken here in the sense Republican govern-
ments1 for which meaning see LSJ s.v. itoXtxeLa III.2.) Since the
loss of the 'rewards' available to orators a new era has seen the
rise of servility sanctioned as right behaviour and a genius for
nothing but flattery:
u 6e vuv eouxayev' ecpp ri y
6 t x al
l a s , xots auxous eêau xal eitLxnSeuyaauv eC duaXcav ext
cppovnydxwv yovov oux eveaiiapyavajyevob xal d'yeuaxot xaXXuaxou xal
yovuywxdxou Xoywv vdyaxos, xnv eXeuêpLav' ecpn 'Xeyw 6td7[ep ou6ev
oxt yn x o X a x e s ex$alvoyev yeyaXocpuets. ' (ibid. 3)
Some sentiments of a r e l a t e d kind are expressed by the elder Pliny
in NH XIV.1.5f. Pliny i s bewailing the decline of the a r t s in
general, rather than just oratory, and he holds responsible for

the decline not the loss of republican government, but the sheer
massiveness and material wealth of the Roman Empire, which has en-
gulfed a multitude of small states where the arts, so we are to be-
lieve, had flourished thanks to the lack of riches of any other kind,
and to the patronage of rulers desirous of immortality (NH XIV.
1.4). But he shares with the philosopher quoted by ps.-Longinus
the conviction that the decline of the arts and the fall of vitae
pretia have been accompanied by the rise of an all-pervasive ser-
vility. He associates this with the increasing importance which
the possession of wealth had assumed in the Rome of his day:
postquam senator censu legi coeptus, iudex fieri censu, magistra-
tum ducemque nihil exornare quam census, postquam coepere orbitas
in auctoritate summa et potentia esse, captatio in quaestu
fertilissimo, ac sola gaudia in possidendo, pessum iere vitae
pretia, omnesque a maximo bono liberales dictae artes in con-
trarium cecidere, ac servitute sola profici coeptum. hanc alius
alio modo et in aliis adorare, eodem tamen habendique ad spes
omnium tendente voto; passim vero etiam egregii aliena vitia quam
bona sua colere malle. ergo, Hercules, voluptas vivere coepit,
vita ipsa desiit. (ibid. 5-6)
It must be concluded that the praemium pulcherrimae rei, the
loss of which Seneca deplores, was certainly no mere financial
reward, but rather the laus, gratia and honor which Republican
orators had been able to earn, and that, if pressed to explain
what he meant by turpia multo honore quaestuque vigentia, he would
probably have spoken of the varieties of servility and sycophantic
crawling (in particular, perhaps, legacy-hunting) by which influ-
ential men of his time had attained their wealth and respected
positions.
Seneca's third explanation: sive fato quodam, cuius maligna
perpetuaque in rebus omnibus lex est, ut ad summum perducta rursus
ad infimum, velocius quidem quam ascenderant, relabantur, belongs
to an old-established tradition of theorizing about the rise and
fall of phenomena: fthe famous parabola shape - ascent, highest
point of axyn and inevitable decline - so dear to antiquity1.1^
As Polybius puts it (VI.51.4): ...TOVTOS xal awyaxos xal noXtTeuas
xau iipdêtos toil xus auCnats xaxd (puatv, yexd 6e xauxnv dxyn,
xdrceuxa cpdCaus...11 Albucius, a declaimer fond of philosophizing,
brought a generalization on these lines somewhat inappositely into
his rendering of Suas. 1 (§3): guidquid ad summum pervenit, in-

cremento non relinquit locum, and the younger Seneca uses similar
language in his Consolatio ad Marciam (23.2f.): quidquid ad summum
pervenit, ad exitum prope est...ubi incremento locus non est,
vicinus occasus est.
It is a remarkable fact that Cicero had already, in Tusc.
disp. II.1.5, prophesied the forthcoming demise of rhetoric from
natural causes: atque oratorum quidem laus ita ducta ab humili
venit ad summum, ut iam, quod natura fert in omnibus fere rebus,
senescat brevique tempore ad nihilum ventura videatur. Velleius
Paterculus, with whom Seneca may be seen to share other views on
literary history,12 treats the question of the rise and fall of
genres, including oratory, in some detail, and his account is the
best commentary on Seneca's third explanation for the decline.
Velleius is impressed by the way that literary genres, once they
have reached perfection, only flourish for a short while, and he
subjects literary history to a somewhat procrustean treatment in
order to prove his point, making out for example that, just as
Greek drama only flourished at its peak for a short time, there
was similarly no philosophy of any importance before Socrates or
after the death of Plato and Aristotle, no rhetoric of note before
Isocrates or after eius auditores eorumque discipulos (I.16.3ff.).
He takes the view that the classical period of Roman historio-
graphy and poetry had been confined to a span of eighty years, now
ended; Roman oratory, of course, had reached its peak suJb principe
operis sui Tullio, and there were few orators before him whose
work could give the reader any pleasure, no orator whom one could
really admire who had not either seen Cicero, or been seen by him
(I.17.2f.).13 Though unable to find a reason which satisfies him
completely for the brief efflorescence of genres, he regards the
following as the most plausible explanation:
alit aemulatio ingenia, et nunc invidia, nunc admiratio imitatio-

nem accendit, naturaque quod summo studio petitum est, ascendit in
summum difficilisque in perfecto mora est, naturaliterque quod
procedere non potest, recedit. et ut primo ad consequendos quos
priores ducimus accendimur, ita ubi aut praeteriri aut aequari eos
posse desperavimus, studium cum spe senescit, et quod adsequi non
potest, sequi desinit et velut occupatam relinquens materiam
quaerit novam, praeteritoque eo, in quo eminere non possumus,
aliquid, in quo nitamur, conquirimus, sequiturque ut frequens ac

mobilis transitus maximum perfecti operis impedimentum sit.
Such was the type of theorizing current in Seneca's time about a

natural law affecting literature whereby decline was inevitable
after a brief period of perfection. But Seneca has shown himself
in Contr. I pr. 6 slightly less gloomy than Velleius about the
apparent impossibility of emulating the achievements of one's pre-
decessors when the genre in which one is working seems to have
reached its dxun already. He sees a possible way out of the
impasse in a less limiting sort of imitatio than that envisaged by
Velleius: ...quo plura exempla inspecta sunt, plus in eloquentiam
proficitur. non est unus, quamvis praecipuus sit, imitandus, quia
numquam par fit imitator auctori, and the provision of plura
exempla is one of the purposes of his collection of declamatory
extracts (ibid.).
It seems misguided to assume that only one of Seneca's
explanations for the decline of oratory can contain any truth.
Modern critics who discuss them tend to find the political expla-
nation by far the most attractive.14 Bonner, for instance, dis-
misses luxus temporum on the one hand as 'a rhetorical theme' and
the theory about the natural law of rise and fall as an 'easy ex-
planation. . .hardly satisfactory, and...in any case, probably
little more than a commonplace'. But such a process of elimin-
ation only over-simplifies, and perhaps distorts, our picture of
what seemed in antiquity a complex and puzzling phenomenon.
The first and third explanations should not be dismissed so
lightly. If we look at Seneca's descriptions of individual de-
claimers, we find in the case of Alfius Flavus a young man whose
considerable natural talents were threatened with extinction by
just the sort of indolence which has been attacked in the tirade
against the morals of the younger generation (somnus languorque -
Contr. I pr. 8 ) : ipse omnia mala faciebat ingenio suo; naturalis
tamen ilia vis eminebat, quae per multos annos, iam et desidia
obruta et carminibus enervata, vigorem tamen suum tenuit (Contr.
1.1.22), and in Scaurus an orator who notoriously failed to re-
alize the potential of his gifts owing to what Seneca calls his
perpetua desidia (Contr. X pr. 3) and who was, if we are to be-

lieve other ancient references to him, a man much given to blatant
impuritas (Tertullian, De Pallio 5) and luxurious living (Petr.
Sat. 77). Thus the rhetorical effusions in Contr, I pr. 8ff. were
not totally divorced from the realities of the times, though of
course Seneca would have been wrong if he imagined that orators of
earlier generations had all been paragons of virtue. In the case
of Cassius Severus we have a man whose state of mind may be com-
pared to that which Velleius Paterculus attributes to the gener-
ation after one which has produced inimitable masterpieces.
Cassius has abandoned Cicerofs ideal of an orator perfect in all
departments, arguing that no one can excel in more than 6ne kind
of writing, and that in ipsa oratione quamvis una materia sit,
tamen ille, qui optime argumentatur, neglegentius narrat, ille non
tarn bene implet quam praeparat etc. (Contr. Ill pr. 10). By
Tacitus1 time Cassius Severus had come to be regarded as the orig-
inator of the new phase in Roman rhetoric, and according to Aper
in the Dialogus (19.1) he turned the course of rhetoric away from
ilia vetere atque directa dicendi via as the result of a conscious
intellectual decision: non infirmitate ingenii nee inscitia
litterarum transtulisse se ad aliud dicendi genus contendo, sed
iudicio et intellectu. Though Cassius does not reject the whole
genre of oratory simply because it has reached its dxyn, he re-
sembles the typical post-classical generation of Velleius1 account,
in deciding that it is hopeless to aspire to the grand ideals of
the past, and (at least according to Aper's interpretation) in
consciously setting out in a completely new direction.
SenecaTs theory about the downfall of the praemium pulcher-

rimae rei ought likewise to be considered against the background
of his many casual remarks about early Imperial orators.15 From
these there emerges quite a varied picture of the state of oratory
in the time of Augustus and Tiberius. The art appears to be far
from dead, or suffering from any total collapse in the esteem ac-
corded to it, even if free speech has been somewhat curtailed; an
optimist might, on the basis of what Seneca says about oratory
outside the first preface, agree with Aper in Tacitus' Dialogus
that eloquence in the Empire was still praesidium simul ac telum,
quo propugnare pariter et incessere sive in iudicio sive in senatu

sive apud principem possis (Dial. 5.6), and might ask as he does:
fama et laus cuius artis cum oratorum gloria comparanda est? (7.3).
The leading orators of the period are named by Cassius
Severus as Pollio, Messala and Passienus (Contr. Ill pr. 14), 1 6
and later tradition, as we have seen, rated Cassius himself as a
highly important figure. Seneca gives a long description of his
oratory in Contr. Ill pr. 2ff. Plenty of other lesser orators re-
ceive discriminating attention from Seneca: Mamercus Aemilius
Scaurus (non novi quemquam, cuius ingenio populus Romanus pertinac-
ius ignoverit. Contr. X pr. 2-3; cf. Suas. 2.22); L. Arruntius
(Contr. VII pr. 7); Asilius Sabinus (Contr. IX.4.17ff. - illud non
probavi, quod multa in re severa temptavit salse dicere); Asinius
Gallus (Contr. IV pr. 4: magnum oratorem, nisi ilium, quod semper
evenit, magnitudo patris non produceret sed obrueret); Fabius
Maximus (Contr. II.4.11: ...nobilissimus vir fuit, qui primus foro
Romano hunc novicium morbum, quo nunc laborat, intulit); Furius
Saturninus (Contr. VII.6.22: maius nomen in foro quam in declamati-
onibus habuit); Iunius Otho (Contr. II.1.34); T. Labienus (Contr.
X pr. 4f.: magnus orator, qui multa impedimenta eluctatus ad famam
ingeni confitentibus magis hominib^.s pervenerat quam volentibus) ;
Tuscus (Suas. 2.22: homo quam inprobi animi, tarn infelicis
ingenii); Vallius Syriacus (homo disertus - Contr. IX.4.18; cf.
Contr. II.1.34ff.); Varius Geminus (Contr. VI.8 extra; cf. Suas.
6.11f.); L. Vinicius (quo nemo civis Romanus in agendis causis
praesentius habuit ingenium - Contr. II.5.19f.); P. Vinicius
(Contr. VII.5.11: exactissimi vir ingeni); M. Vipsanius Agrippa
(Contr. II.4.13); Votienus Montanus (Contr. IX.5.15: homo rarissumi
etiamsi non emendatissimi ingeni; cf. Contr. VII.5.11; IX pr. 1;
6.10, 18).
Orators seem, from Seneca1s passing remarks, to have been by

no means short of work.*7 A statement intended to illustrate
Cassius Severus1 exceptional carefulness as an orator sheds un-
expected light on the frequency with which one might be expected
to deliver speeches:
numquam tamen haec felicitas illi persuasit neglegentiam. uno die
privatas plures <quam duas> non agebat et ita ut alteram ante
meridiem ageret, alteram post meridiem, publicam vero numquam

amplius quam unam uno die. (Contr. Ill pr. 5)
How often, then, would a careless orator speak? Some indication
is found in Senecafs description of the hasty manner in which
Scaurus prepared his speeches: saepe causam in ipsis subselliis,
saepe dum amicitur discebat (Contr. X pr. 2 ) .
Some limitation of free speech there certainly was. The
libertas which in Augustus1 time had allowed one to joke about M.
Vipsanius Agrippafs lowly middle name is referred to as a thing of
the past:
tanta autem sub divo Augusto libertas fuit, ut praepotenti tune M.
Agrippae non defuerint qui ignobilitatem exprobrarent. Vipsanius
Agrippa fuerat, <at> Vipsani nomen quasi argumentum paternae hum-
ilitatis sustulerat et M. Agrippa dicebatur. cum defenderet reum,
fuit accusator qui diceret: 'Agrippa Marce et quod in medio est1 -
volebat Vipsanium intellegi -; fuit qui diceret: Tconcurrite!
Agrippa malum habebit. responde sis, Marce uterque.T mihi
videtur admiratione dignus divus Augustus, sub quo tantum licuit..
(Contr. II.4.13 W)
Even the reign of Augustus had seen the burning of Labienus1
books, 18 deplored in Contr. X pr. 5ff., though Seneca does not
blame the Princeps for it. Now the situation had arisen where one
might have to decide whether it was worthwhile caput potius quam
dictum perdere; Seneca advises discretion (Contr. II.4.13).
An orator was not necessarily so inhibited by having to con-
duct a case before the Emperor that he could not produce attract-
ive oratory in the new style, full of pleasing sententiae. Of P.
Vinicius1 accusation of Votienus Montanus we read in Contr. VII.
5.12:
accusaverat ilium apud Caesarem, a colonia Narbonensi rogatus. at
Montanus adeo toto animo scholasticus erat, ut eodem die, quo
accusatus est a Vinicio, dic<eret: fdel>ectavit me Vinici actio1;
et sententias eius referebat. eleganter illi dixit Surdinus:
rogo: numquid putas ilium alteram partem declamasse?
Moreover, to judge from what Seneca says about Sabinus1 plea for
an allowance (Contr. IX.4.2Of.), the Roman Senate still sometimes
heard even late in Tiberius' reign speeches in which both pathos
and humour were to be found:
multa ilium diserte dixisse memini, cum introductus esset ex
carcere in senatum postulaturus, ut diaria acciperet. tune dixit
de fame questus: nihil onerosum a vobis peto, sed ut me aut mori
velitis aut vivere. et illud dixit: nolite, inquam, superbe
audire hominem calamitosum:

saepe qui misereri potuit misericordiam rogat.
et cum dixisset Seianianos locupletes in carcere esse: homo,
inquit, adhuc indemnatus, ut possim vivere, parricidas panem rogo.
cum movisset homines et flebili oratione et diserta, redit tamen
ad sales: rogavit, ut in lautumias transferretur: non est, inquit,
quod quemquam vestrum decipiat nomen ipsum lautumiae; ilia enim
minime lauta res est.
Of course, grovelling flattery towards the Emperor marred
some of the public oratory of this period; Contr. VI.8 extra pre-
serves an example: Varius Geminus apud Caesarem dixit: Caesar, qui
apud te audent dicere magnitudinem tuam ignorant, qui non audent,
humanitatem. (We do not know in what context Seneca quoted this.)
The prosecutors in trials for maiestas must have brought eloquence
into some disrepute: note that Tuscus, who prosecuted Scaurus on
this charge, is the one orator in whom Seneca fails to find a re-
deeming feature (Suas. 2.22). And the orator speaking before
Tiberius ran the risk, according to Tacitus, of having his most
carefully contrived flatteries misinterpreted: the calamity of
Iunius Gallio, who in 31 framed a proposal designed to please
Tiberius, only to incur the suspicion of being a two-faced
satelles Seiani (Tac. Ann. VI.3), 1 9 must have saddened his friend,
the elder Seneca, and served to remind him that in a political
career ipsa quae sperantur timenda sunt (Contr. II pr. 4 ) .
Outside the political sphere, lurid trials still sometimes
provided opportunities for sensational, if not great, oratory.
One such is described in Contr. II.1.34ff.:
erat genus iudici tale: speciosum iuvenem dominus suus deprehend-
isse cum uxore in cubiculo testatus est et ob hoc uxorem suam
dimisit. hoc nomine servum adulteri postulatum dominus non
defendebat, mulier, in quam petebatur praeiudicium, tuebatur.
The advocates displayed much ingenuity in dealing with this case,
in particular in finding a color to exculpate the woman, cum in
cubiculo visa esset cum servo et marito, and the onlookers at the
trial were vastly entertained by witticisms bandied around about
the 'Theodorean1 avoidance of narration indulged in by one of the
counsels.20 Other notable cases recalled by Seneca include the
trial of Moschus the rhetorician for poisoning (Contr. II.5.13),
in which Pollio spoke unsuccessfully for the defence, and the
fiasco in the Centumviral Courts which forced Albucius to restrict
his activities from then on to the schools of rhetoric (Contr. VII

pr. 6ff.).
Outside Rome eloquence appears to have been in quite a
healthy state. Of course, to an ambitious urban Roman this fact
would seem of little importance: note the dictum of Censorinus,
quoted by Cassius Severus in Contr. Ill pr. 12, that candidates
for public office in the municipia were exerting themselves in a
dream world (in somniis laborare).2* Seneca nevertheless had a
high enough regard for his native province to think the leading
orators of Spain worth praising. At the end of the tenth preface
we are told about Gavius Silo, a man of whom Augustus said,
1
numquam audivi patrem familiae disertiorem1 (Contr. X pr. 14),
and Clodius Turrinus, the account of whose career shows that at
least in Spain eloquence could still bring enormous prestige, not
to mention financial gain:
causas nemo diligentius proposuit, nemo respondit paratius; et
pecuniam itaque et dignitatem, quam primam in provincia Hispania
habuit, eloquentiae debuit. natus quidem erat patre splendid-
issimo, avo divi Iuli hospite, sed civili bello attenuatas domus
nobilis vires excitavit, et ita ad summam perduxit dignitatem ut,
si quid illi defuerit, scias locum defuisse. (Contr. X pr. 16)
There is plenty of evidence, then, even within Seneca^ own
work, that the opportunities for eloquence in the early Empire
were abundant. To what extent the prestige - the laus, gratia and
honor - which could be earned by an orator, was less than it had
been under the Republic is something which it is impossible to
gauge properly now, but the dominant position in Roman education
of the schools which ostensibly trained the orators of the fu-
ture 22 is some index of the esteem in which oratorical skill con-
tinued to be held under the Empire. Thus Seneca's second expla-
nation for the decline of eloquence is not such a resoundingly
complete truth that all other theories should be dismissed as
superficial by comparison. Bear in mind that, like the others, it
was probably taken over from earlier theorizing about the Hellen-
istic world.
Senecafs threefold theory de causis corruptae eloquentiae is
as interesting for what it leaves out as for what it includes.
First, it includes no suggestion that the loss of the praemium of
eloquence and the resultant change in people's activities was any-

thing but deplorable, whereas in the discussions of the subject by
Tacitus and ps.-Longinus we find extensive debunking of the view
that the loss of the conditions afforded to the orator by a free
Republic was altogether a bad thing.
In the long summing-up speech which draws Tacitus' Dialogus
to its close Maternus makes it clear that he has no regrets about
the demise of the perturbatio and licentia of the Republic, with
its popularist legislation, its all-night speeches, its pros-
ecutions of the powerful, family feuds, aristocratic factions, and
never-ending enmity between Senate and plebs:
quae singula etsi distrahebant rem publicam, exercebant tamen
illorum temporum eloquentiam et magnis cumulare praemiis vid-
ebantur, quia quanto quisque plus dicendo poterat, tanto facilius
honores adsequebatur, tanto magis in ipsis honoribus collegas suos
anteibat, tanto plus apud principes gratiae, plus auctoritatis
apud patres, plus notitiae ac nominis apud plebem parabat. (Dial.
36.4)
If civil strife and the activities of people like Catiline, Milo,
Verres and Antony were the price one had to pay for great oratory,
he for one was no*", prepared to pay it (37.6ff.). In ps.-Longinus1
treatise doubt is cast on the whole idea that the loss of Republi-
can government was the cause of the disappearance of great litera-
ture. The view that an undemocratic constitution enslaves the
intellect is put by ps.-Longinus into the mouth of 'one of the
philosophers' (Subl. 44.1); speaking in the first person (44.6) he
proposes that it is 6 waxex^ nywv Tag eirtûyLas aiteptopLaTOS
ouToau TtoAeyog in particular moral enslavement caused by love of
money and pleasure, rather than r\ xns oilxouuevris eupnvri, which has
stifled literary greatness. It is noteworthy that Seneca the
Elder, whatever reservations he had as to whether free speech
should be left completely untrammelled,2^ did not, in his main
statement on the decline of eloquence, look on an intellectual
desert and call it peace.
But the respect in which Seneca's theorizing in Contr. I pr.
7 is most conspicuously different from other discussions of the
decline of eloquence is the lack of any hint in it that the edu-
cation provided by the schools of rhetoric might in any way be to
blame.
The remoteness from real life of the themes of controversiae

and suasoriae is again and again jeered at by critics in the first
century A.D. 24 People were for ever pointing out that all the
years spent assiduously in declamation on these fanciful themes
turned out to be of no use at all when the student of rhetoric
finally entered the forum. Encolpius in Petronius* Satyricon puts
the matter forcibly when addressing Agamemnon the rhetorician
(Sat. If.):
nunc et rerum tumore et sententiarum vanissimo strepitu hoc tantum
proficiunt, ut cum in forum venerint, putent se in alium orbem
terrarum delatos. et ideo ego adulescentulos existimo in scholis
stultissimos fieri, quia nihil ex his, quae in usu habemus, aut
audiunt aut vident, sed piratas cum catenis in litore stantes, sed
tyrannos edicta scribentes, quibus imperent filiis ut patrum
suorum capita praecidant, sed responsa in pestilentiam data, ut
virgines tres aut plures immolentur, sed mellitos verborum
globulos et omnia dicta factaque quasi papavere et sesamo sparsa.
qui inter haec nutriuntur, non magis sapere possunt, quam bene
olere, qui in culina habitant, pace vestra liceat dixisse, primi
omnium eloquentiam perdidistis.
It was not hard to find illustrations of the irrelevance of many
controversia themes to modern life: in Tacitus1 Dialogus we find
singled out for ridicule tyrannicidarum praemia aut vitiatarum
electiones aut pestilentiae remedia aut incesta matrum, all sub-
jects common in the schools but rarely or never encountered in the
forum (Dial. 35.5). 2 5 Quintilian saw the justice of such com-
plaints: nam magos et pestilentiam et responsa et saeviores
tragicis novercas aliaque magis adhuc fabulosa frustra inter
sponsiones et interdicta quaeremus (II.10.5), and recommended cer-
tain reforms: judicial themes of a more realistic type might be
invented, and some of the more pernicious tendencies which had
developed in the schools - the perpetual use of high-flown
language and the total exclusion of humour - could be done away
with:
utinamque adici ad consuetudinem posset ut nominibus uteremur et
perplexae magis et longioris aliquando actus controversiae
fingerentur et verba in usu cotidiano posita minus timeremus et
iocos inserere moris esset: quae nos, quamlibet per alia in
scholis exercitati simus, tirones in foro inveniunt. (ibid. 9)
A type of bombast favoured in declamation had been well parodied
by Petronius:
num alio genere furiarum declamatores inquietantur, qui clamant:
'haec vulnera pro libertate publica excepi; hunc oculum pro vobis
impendi: date mihi ducem, qui me ducat ad liberos meos, nam
succisi poplites membra non sustinent1? (Sat. 1)
Evidently drawing on a Greek source,26 he blames the taste for
such levibus...atque inanibus sonis on Asianic influences newly-
come to Athens: nuper ventosa istaec et enormis loquacitas Athenas
ex Asia commigravit... (§2).
In general the schools were criticized for providing excess-
ively sheltered conditions which gave the students a false sense
of security, later to be rudely shattered as soon as they went out
to speak in the forum. In the Dialogus Messala complains:
at nunc adulescentuli nostri deducuntur in scholas istorum, qui
rhetores vocantur...<in> quibus non facile dixerim utrumne locus
ipse an condiscipuli an genus studiorum plus mali ingeniis ad-
ferant. nam in loco nihil reverentiae est, in quem nemo nisi
aeque imperitus intrat; in condiscipulis nihil profectus, cum
pueri inter pueros et adulescentuli inter adulescentulos pari
securitate et dicant et audiantur... (35.Iff.)
Quintilian singles out for criticism the frequent applause with
which the students greeted each other's efforts (II.2.9ff.), and
the dangers inherent in the scholastic practice of declaiming only
one side of a controversia at a time (XII.6.5, cf. VII.3.20),
which meant that the declaimer could assume that any facts not
made clear in the thema were in his own favour (VII.2.54). Such
observations were, as we shall see, already commonplace among hos-
tile critics of the schools in the elder Seneca's time.
Blame for the deep entrenchment of the pernicious practices
customary in the schools of rhetoric was directed variously
towards the rhetores, the students, and their parents. Agamemnon,
the rhetorician in the Satyricon, concedes the truth of the criti-
cisms levelled against the schools, but points out the difficult-
ies of the teacher's position; he has to give the pupils what they
want:
nihil nimirum in his exercitationibus doctores peccant, qui
necesse habent cum insanientibus furere. nam nisi dixerint quae
adulescentuli probent, ut ait Cicero, 'soli in scholis
relinquentur'. (Sat. 3)
He finds that there is no demand from their parents either, for a
strenuous type of education:
parentes obiurgatione digni sunt, qui nolunt liberos suos severa
lege proficere. primum enim sic ut omnia spes quoque suas

ambitioni donant. deinde cum ad vota properant, cruda adhuc
studia in forum pellunt et eloquentiam, qua nihil esse maius con-
fitentur, pueris induunt adhuc nascentibus. (§4)
Quintilian, speaking as a representative of the profession, has
fault td find with the rhetoricians as well: for various reasons
they have a tendency to spend more time than they need in covering
the standard curriculum:
sed culpa est in praeceptoribus prima, qui libenter detinent
<quos> occupaverunt, partim cupiditate diutius exigendi mercedulas,
partim ambitione, quo difficilius <videatur esse> quod pollicentur,
partim etiam inscientia tradendi vel neglegentia: proxima in nobis,
qui morari in eo quod novimus <quam> discere quae nondum scimus
melius putamus. nam ut de nostris potissimum studiis dicam, quid
attinet tarn multis annis quam in more est plurimorum, ut'de iis a
quibus magna in hoc pars aetatis absumitur taceam, declamitare in
schola et tantum laboris in rebus falsis consumere, cum satis sit
modico tempore imaginem veri discriminis et dicendi leges
comperisse? (XII.ll.14f.)
This indictment has the ring of truth.
That there should be no trace of criticism of rhetorical edu-
cation in Contr. I pr. 7 is all the more remarkable seeing that
later in his work the elder Seneca reports two attacks by acquaint-
ances of his on the whole practice of declamation which are even
more devastating and comprehensive than Petroniusf anti-scholastic
invective. Cassius Severus is reported in the third preface as
having criticized the schools severely for encouraging in those
who frequented them an exaggerated self-esteem which proved to be
quite unwarranted once they entered public life:
agedum istos declamatores produc in senatum, in forum: cum loco
mutabuntur; velut adsueta clauso et delicatae umbrae corpora sub
divo stare non possunt, non imbrem ferre, non solem sciunt, vix se
inveniunt; adsuerunt enim suo arbitrio diserti esse. non est quod
oratorem in hac puerili exercitatione spectes. quid, si velis
gubernatorem in piscina aestimare? (Contr. Ill pr. 13f.)
In particular he pours scorn on the way that ignorant and taste-
less students preferred Cestius and Latro to the leading orators
of their time, Pollio, Messala and Passienus, and only refrained
from expressing a preference for their favourite rhetorician over
Cicero for fear of a stoning (§14f.). In the ninth preface Seneca
reports an even more vehement attack on the schools of declamation.
It opens with an allegation that school practices were having a
damaging effect on the oratory of the forum:
qui declamationem parat, scribit non ut vincat, sed ut placeat.

omnia itaque lenocinia [itaj conquirit; argumentationes, quia
molestae sunt et minimum habent floris, relinquit; sententiis,
explicationibus audientis delinire contentus est. cupit enim se
approbare, non causam. sequitur autem hoc usque in forum declama-
tores vitium, ut necessaria deserant, dum speciosa sectantur.
(Contr. IX pr. If.)
The conditions which made for the scholastics1 unwarranted over-
confidence are attacked one by one:
accedit etiam illud, quod adversarios quamvis fatuos fingunt:
respondent illis et quae volunt et cum volunt. praeterea nihil
est, quod errorem aliquo damno castiget; stultitia eorum gratuita
est. vix itaque in foro futurus periculosus stupor discuti
potest, qui crevit, dum tutus est. quid, quod laudationibus
crebris sustinentur et memoria illorum adsuevit certis intervallis
quiescere? cum ventum est in forum et desiit illos ad omnem
gestum plausus excipere, aut deficiunt aut labant. adice nunc,
quod <animus> illis nullius interventu excutitur: nemo ridet, nemo
ex industria obloquitur, familiares sunt omnium vultus. in foro,
ut nihil aliud, ipsum illos forum turbat. (§2f. M)
As an illustration we are given an anecdote about Latro, who was
once called upon to speak in a court-case, but was so unaccustomed
to outdoor speaking that he began his speech with a solecism, and
had to get the case moved indoors before he would complete it (§3).
The fundamental failing of declamation as an educational method,
our critic goes on to argue, is that it is so much easier than the
work for which it purports to be a training: gladiators and ath-
letes are trained far more sensibly, required as they are to prac-
tise tests of skill and endurance more arduous than those they
will face in the arena (§4). The schools of rhetoric provide no
such preparation for the adverse conditions to be encountered by
orators in the forum:
in foro partem accipiunt, in schola eligunt; illic iudici blandi-
untur, hie inperant; illic inter fremitum consonantis turbae
intendendus animus est, vox ad aures iudicis perferenda, hie ex
vultu dicentis pendent omnium vultus. (§5)
Trained in such an environment, novices in the forum only mature
into orators after long subjection to ridicule and much hard work
(ibid.).
The elder Senecafs own attitude towards declamation seems to
have been much less severe. He was aware that the schools served
sometimes as an escape route for failed orators (Contr. VII pr. 6;
IX.4.18), but he was not persuaded by any of his contemporaries1
attacks on the schools to abandon his conviction that rhetorical

education was a useful training even for people who were not
planning to enter public life (Contr. II pr. 3). His friendship
with Latro, one must suspect, was a prime cause of his tolerance
towards the declaimers. We must make no mistake about it: this
tolerance was considerable. Modern critics have often suggested
that Seneca thought that declamation ought to return to its proper
place as a mere educational exercise,27 but there is no evidence
that this was so. Bornecque asserts: !J1 n'ignore pas que la
declamation n'est qu'un moyen de former les jeunes gens; la preuve
en est que, de tous les professeurs, celui qu'il admire le plus
est Latron, qui, j'aurai 1'occasion de le montrer, maintient la
declamation dans ses justes limites.' But this will not do as
proof. Certainly Latro did not suffer from gross lapses of taste,
and organized his declamations in a straightforward manner, but it
is important to remember that he was foremost among Latin rhetores
in avoiding the chore of actual teaching (Contr. IX.2.23) and that
he was alleged to have been a total failure as an orator (Contr.
IX pr. 3 ) . By expressing admiration for Latro, Seneca set himself
apart from those who condemned, with censorium supercilium raised,
the practice of declaiming for the sake of frivolous ostentation
(Contr. X pr. 4 ) . He did not regard declamation as a form of lit-
erary activity of the same seriousness as true oratory or the
writing of history: note, in addition to his description of the
declamatory anthology as non seriam rem in the somewhat conven-
tional leave-taking of Contr. X pr. 1, remarks he makes in Contr.
1.8.16: Diocles Carystius dixit sententiam, quae non in declama-
tione tantum posset placere sed etiam in solidiore aliquo scripti
genere, and in Suas. 5.8 (referring to Gallio): hoc loco
disertissimam sententiam dixit <dignam> quae vel in oratione vel
in historia ponatur... Nevertheless, school rhetoric interested
the elder Seneca sufficiently for him to devote a long work to the
memory of the declaimers he had known, and to present posterity
with minute and sympathetic documentation of the various facets of
their art.
PART III
FIVE ASPECTS OF DECLAMATION: THE ELDER SENECA'S EVIDENCE
Quint. III.3.1: omnis autem orandi ratio, ut plurimi maximique

auctores tradiderunt, quinque partibus constat: inventione,
dispositione, elocutione, memoria, pronuntiatione sive actione
(utroque enim modo dicitur).*•
1 INVENTIO
The declaimer1s first task, as he set about composing a contro-

versial was to decide whose part he would take in the imaginary
law-suit: in foro partem accipiunt, in schola eligunt (Contr. IX
pr. 5 ) . Some controversia themes were so devised that it was
possible to make out a fair case both for and against the imagined
defendant, but usually one side's case was very much easier to
justify than the other1s, and few declaimers chose to defend the
more obviously villainous characters in the themes, see e.g.
Contr. X.4.15: pro illo qui debilitabat expositos pauci admodum
dixerunt... Also perhaps at this stage one might decide whether
to speak in the person of the prosecutor or defendant himself, or
whether to take the part of an advocate.2 There were various
standard reasons for adopting the latter procedure, never men-
tioned by Seneca the Elder, doubtless because he assumed they were
common knowledge. Ps.-Quintilian explains the rules as follows:
in plerisque controversiis plerumque hoc quaerere solemus, utrum
ipsorum persona utamur ad dicendum, an advocati: vel propter
sexum, sicut <in> feminis, vel propter aliquam alioqui vitae vel
ipsius, de quo quaeritur, facti deformitatem. (Decl. Min. CCLX,
61.12ff. Ritter)
Seneca's declaimers impersonate advocates when speaking on behalf
of women (e.g. Contr. 1.2, IV.6) and slaves (e.g. Contr. III.9),
and sometimes on account of facti deformitatem (Contr. X.5.17ff.),
but Seneca finds it worthy of note only when a declaimer decides
to act as an advocate in order to facilitate a more hard-hitting
prosecution, e.g. in a case where a son is prosecuting his father
and would be inhibited by pietas if he had to speak in person
(Contr. 1.7.13),^ or in order to allow for a more comprehensive
defence involving the use of a variety of colores (Contr. 1.7.17;
VII.2.13).
When setting out to declaim a suasoria one could choose to
give advice either for or against a course of action. Here again,
Five aspects of declamation 152
one side tended to be more popular than the other: faced with the
subject, deliberat Cicero, an scripta sua conburat promittente
Antonio incolumitatem, si fecisset (Suas. 7 ) , none of the de-
claimers would advise Cicero to save his skin, and Seneca comments:
huius suasoriae alteram partem neminem scio declamasse; omnes pro
libris Ciceronis solliciti fuerunt, nemo pro ipso, cum adeo ilia
pars non sit mala, ut Cicero, si haec condicio lata ei fuisset,
deliberaturus non fuerit. (Suas. 7.10)
The choice whether to take the part of the deliberator himself or
of someone advising him seems to have been determined by school
tradition. In Suas, 2 and 5 the subject-matter demanded that one
should speak as a participant in a gathering assembled to make a
decision, but only convention can have determined the fact that in
Suas. 3 one took the part of the deliberator, Agamemnon, but in
Suas. 1, 4, 6 and 7 that of an adviser to the man deliberating.4
The declaimer next proceeded to that part of his work gener-
ally described in Roman rhetorical theory as inventio. Seneca the
Elder never uses this term, though he can hardly not have known
what it meant. In his view (and presumably this was the view gen-
erally accepted in the schools of his time), the task of inven-
tion1 as it faced someone preparing a controversial consisted of
two parts so distinct that they required separate consideration:5
first, divisio, the isolating of the basic issues to be disputed,
and secondly, the choice of colores, ways of interpreting the
given facts of the case. A suasoria involved only divisio, no
colores.
Seneca clearly felt that divisiones and colores were worthy
of as much close attention on the part of the critic as matters of
form and style. Indeed, the bulk of his criticism outside the
prefaces derives more or less directly from a consideration of
these topics.
DIVISIO
Divisio in the elder Seneca's view was the foundation (Contr. I
pr. 21: fundamentum) on which a declamation was built.^ In the
completed declamation it would be overlaid with such an elaborate
superstructure (Contr. I pr. 21: superstructis tot et tantis
molibus) that its existence might hardly be perceptible to the
Inventio 153
audience. Indeed, the outpouring of brilliant language would

often serve to obscure flaws of divisio in a declamation, ipsa
enim actio multas latebras habet, nee facile potest, si quo loco
subtilitas defuit, apparere, cum orationis cursus audientis
iudicium impediat, dicentis abscondat (Contr. I pr. 21), and it
was quite possible for a declaimer to win acclaim without paying
much attention to the careful organization of arguments. For
instance, Seneca says of the young Ovid, tune autem cum studeret,
habebatur bonus declamator. hanc certe controversiam ante
Arellium Fuscum declamavit, ut mihi videbatur, <illo>'7 longe
ingeniosius, excepto eo, quod sine certo ordine per locos dis-
currebat (Contr. II.2.9), and again, molesta illi erat omnis
argumentatio (ibid, 12). One hostile critic quoted by Seneca went
as far as to advance the generalization that the typical declaimer
omitted argument altogether, argumentationes, quia molestae sunt
et minimum habent floris, relinquit (Contr. IX pr. 1 ) , but this
was probably an overstatement. In Seneca's view, at any rate, a
good declaimer was one who paid due attention to the 'virtue of
oratory1 subtilitas9 a fine precision in the analysis of the ques-
tions at issue. His friend Porcius Latro had laid particular
emphasis on the need for a well-ordered divisio: it had been his
practice, before delivering any declamation in full, to present an
unadorned statement of the lines of questioning which he was in-
tending to pursue, a practice which, to Seneca's regret, later
rhetoricians had not imitated (Contr. I pr. 21).
The divisiones of suasoriae were usually extremely simple.
When treating Suas. 4 (Deliberat Alexander Magnus an Babylona
intret, cum denuntiatum esset illi responso auguris periculum),
Arellius Fuscus raised the single issue8 of the validity of proph-
ecy: in hac suasoria nihil aliud tractare Fuscum scio quam easdem
quas supra rettuli quaestiones ad scientiam futuri pertinentis
(Suas. 4.4). Occasionally a declaimer would dispense with divisio
altogether and would simply put on a display of enthusiasm:
Triarius omni dimissa divisione tantum exultavit, quod Xersen
audiret venire: adesse ipsis novam victoriam, nova trophaea (Suas.
5.7). Usually, though, one's persuasion would be divided into two
or three main sections, called simply partes (Suas. 2.11; 6.8-9)
or quaestiones (Suas. 1.8-9; 4.4). Thus Cestius1 treatment of

Suas. 1 was divided into two parts:
...quam sic divisit, ut primum diceret, etiamsi navigari posset
Oceanus, navigandum non esse...deinde illam quaestionem subiecit,
ne navigari quidem Oceanum posse. (Suas. 1.8)
In the suasoria about the three hundred Spartans, Arellius Fuscus
made use of a tripartite plan described by Seneca as divisione...
ilia volgari:
...non esse honestum fugere, etiamsi tutum esset; deinde: aeque
periculosum esse fugere et pugnare; novissime: periculosius esse
fugere... (Suas. 2.11)
Note that the declaimers did not feel at all tied to the classic
plan for persuasion of which Quintilian says in III.8.22, partes
suadendi quidam putaverunt honestum utile necessarium. ego non
invenio huic tertiae locum. It was sometimes used, as for example
by Cestius in Suas. 6 (§10): ...mori tibi utile est, honestum est,
necesse est, ut liber et inlibatae dignitatis consummes vitam.
But it was just one of many alternatives.
Occasionally Seneca presents an analysis of the subsidiary
issues raised by one of the main quaestiones in a declaimer's
treatment of a suasoria. For example we are told that Fabianus,
having adopted the same two-fold division as Cestius for Suas. 1,
dealt with the second part, on the impossibility of crossing
Oceanus, as follows:
ut primum negaret ullas in Oceano aut trans Oceanum esse terras
habitabiles. deinde: si essent, perveniri tamen ad illas non
posse; ...novissime: ut posset perveniri, tanti tamen non esse.
(Suas. 1.10)
Seneca also describes in detail how Cestius subdivided a section
in his treatment of Suas. 3, where he denied the validity of
prophecy:
...deos rebus humanis non interponere arbitrium suum; ut inter-
ponant, voluntatem eorum ab homine non intellegi; ut intellegatur,
non posse fata revocari. si non sint fata, nesciri futura; si
sint, non posse mutari. (Suas. 3.3)
Here is evidence of quite complex argument, encroaching on the
territory of the philosophers. But, on the whole, suasoriae
afforded little scope for subtilitas in argumentation. Seneca
specifically apologizes for the lack of that quality in the
suasoria on the trecenti Lacones (Suas. 2.10).
Inventio 155
The issues raised by a controversia theme and its accompany-

ing laws were usually far more numerous, and it required more
acumen on the part of the declaimer to arrange the questions for
dispute in an appropriate order, duly subordinating the minor
issues to the greater ones.
Seneca distinguishes three types of question which one might
treat in a controversia: the quaestio iuris,^ in which one asked
whether the defendantTs action was legal, 'an liceat...?', 'an
possit...?'; the quaestio aequitatis,10 the treatment of which is
sometimes given the technical name tractatio,^^ in which one con-
sidered whether his action was morally right, 'an debeat...?', 'an
oporteat...?'; and the quaestio coniecturalis^^ in which one con-
jectured whether something had, or had not, happened.
Most of the questions found in Seneca's analyses are of the
first two types. Seneca appears perfectly clear in his mind as to
the difference between law and equity. We read, for example, in
Contr. 1.1.13, Latro illas quaestiones fecit: divisit in ius et
aequitatem, an abdicari possit,- an debeat. The distinction was a
very old one, and not peculiar to Roman legal thought.13 Aristotle
notes in his Rhetoric (1.13.2) that the concept of an unwritten
law which can overrule the written one was already recognized by
Sophocles, Empedocles, and Alcidamas. The Latin terminology used
by Seneca is found already in Cicero's works, e.g. De or. 1.56.240:
pro aequitate contra ius dicere; Part. or. 28.100: in earn formam
causarum, in qua quale sit ambigitur quae in aequitate et in
iure maxime consistit.
Generally questions of law were treated first, before any
discussion of equity, e.g. Contr. 1.8.7: prima quaestio ilia ab
omnibus facta est vulgaris: an filius ob id, quod sui iuris sitf
abdicari possit; deinde: an debeat. haec tota tractationis est;
Contr. II.5.14: Latro ex suo more has non quaestiones putabat, sed
membra illius ultimae partis ex aequitatis quaestione pendentis.
It was obviously best from a psychological point of view to leave
the section which played most on the emotions till last.
Quaestiones iuris
The section of the divisio which dealt with points of law was
often quite elaborately constructed. Characteristically there
would be two or three main legal questions at issue, from each of
which a number of lesser questions might spring. Senecafs analy-
ses make the relationship between quaestiones quite clear:14 his
practice is simply to set down the main questions first, and then
to list the subsidiary questions, explaining in a few words how
they were related to the main ones; after that, subdivisions of
the subsidiary questions, if any, might be listed, as for example
in Contr. I.2.13f.:
Latro in has quaestiones divisit: an per legem fieri sacerdos non
possit; etiamsi lex illi non obstat, an sacerdotio idonea sit. 15
an lege prohibeatur, in haec duo divisit: an casta sit, an pura
sit. an casta sit, in haec divisit: utrum castitas tantum ad
virginitatem referatur an ad omnium turpium et obscenarum rerum
abstinentiam...etiamsi ad virginitatem tantum refertur castitas,
an haec virgo sit...an pura sit, in haec divisit: an, etiamsi
merito occiderit hominem, pura tamen non sit homicidio coinquinata;
deinde: an merito occiderit hominem innocentem uti corpore
prostituto volentem.
The complexities set out diagrammatically below (p. 161) are here
described with the minimum of fuss: Seneca found no need for an
elaborate technical terminology in his analyses, no need even to
distinguish by means of adjectives between greater and lesser
questions, as Bardon would like him to have done. 16 Generally he
calls all questions, great and small, simply quaestiones. (Once
he uses the diminutive, quaestiuncula, but curiously enough with
reference to a quaestio which seems to have been of some import-
ance in the divisio of at least one declaimer who included it -
Contr. IX.6.15, cf. ibid. 10.) As a change from quaestiones, he
sometimes substitutes the word pars, using it either to mean a
major part of the whole divisio (Contr. II.5.15; IX.6.15), or a
part of a quaestio (Contr. IX.4.10). Such lines of speculation as
are not strong enough to stand on their own as quaestiones are
variously described as particulae (Contr. 1.3.8: ...particulas
incurrentes in quaestionem); membra (Contr. II.3.15: haec omnia
quasi membra in aliquam quaestionem incurrentia tractabat, non ut
quaestiones); or argumenta (Contr. II.3.16: Fuscus parum hoc
putabat valens esse tamquam quaestionem, satis valens tamquam
Inventio 157
argumentum).
Quaestiones/'tractationes aequitatis
Questions of equity called for more expansive treatment than was
appropriate for legal argument. It was presumably for this reason
that the term tractatio was sometimes preferred to quaestio as a
word to describe the discussion of aequitas:
Contr. 1.1.14: adiecit quaestionem [Gallio] alteram: an si abdicari
possit etiam adoptatus, <possit> ob id vitium, quod, antequam ad-
optaretur, noturn fuit adoptanti. haec autem ex aequitatis parte
pendet et tractatio magis est quam quaestio.
Contr. 1.2.14: an idonea sit, in tractationes quas quisque vult
dividit...
Contr. 1.4.6: an oportuerit, tractationis quidem est, quam ut quis-
que vult variat.
Contr. II.2.5: reliqua, cum ad aequitatem pertineant, tractationis
sunt.
Contr. II.5.16: et hoc contra Latronem dicebat [sc. Blandus] : quo-
modo istam quaestionem putas in aequitatis tractationem cadere,
cum quid liceat quaeratur, non quid oporteat?
Contr. IX.1.9: Latro in has quaestiones divisit...novissume: an si
adfectu et indignatione ablatus non fuit in sua potestate,
ignoscendum illi sit. hoc non tamquam quaestionem, sed, ut illi
mos erat, pro tractatione aut loco.
However, quaestio was so much the obvious word to use to denote an
enquiry of any sort that it is not surprising that Seneca does not
invariably suppress it as the name for a question of equity (as
opposed to the treatment of such a question):
Contr. II.5.14: Latro ex suo more has non quaestiones putabat, sed
membra illius ultimae partis ex aequitatis quaestione pendentis.
Contr. VII.4.4: Albucius non iuris illam fecit quaestionem sed
aequitatis, ita tamen, ut et iuris adiungeret...
Contr. VII.8.8: novissimam quaestionem fecit aequitatis...
On numerous other occasions he applies the term quaestio to ques-
tions of the type, 'an debeat...?f, or 'an oporteat...?' (e.g.
Contr. 1.1.13; II.3.12).
With one exception all types of question about moral obli-
gation came under the heading of aequitas. The exception was the
question, often raised in controversiae, whether an action consti-
tuted a beneficium requiring gratitude. This had to be classed as
a quaestio iuris as it was dealt with under the declamatory lex
ingrati (see e.g. Contr. II.5.10ff.).
Quaestiones coniecturales
The adjective coniecturalis is taken from the terminology used by
Latin theorists when translating the axaaus theory of Hermagoras
of Temnos.17 Status (or constitutio) coniecturalis was the name
given to one of the four basic types of issue distinguished by
Hermagoras, the type in which one raised simple questions of fact,
such as occideritne Aiacem Ulixes...bonone animo sint erga populum
Romanum Fregellani...si Karthaginem reliquerimus incolumem, num
quid sit incommodi ad rem publicam perventurum? (Cic. De inv.
1.8.11).
Seneca gives the name controversia coniecturalis to two of
the declamations in his collection, meaning that they hinge on
questions of fact. These are Contr. VII.3, where we have to de-
cide whether a son who has been disowned and then forgiven three
times was preparing poison for his father or himself, in uno
homine coniectura duplex est (Contr. VII.3.6), and Contr. VII.7
where the issue is whether the dying general was referring to his
father when he gave the warning 1cavete proditionem1. Seneca re-
gards the divisio of such controversiae as a simple matter: ...con-
iecturalis est et habet quasi certum tritumque iter (Contr. VII.7.
10); non puto vos exigere divisionem, cum coniecturalis sit contro-
versia (Contr. VII.3.6).
Quaestiones coniecturales were also raised occasionally in
more complex controversiae which also involved questions of law
and equity. In Contr. II.2.5 we find Latro making a very reason-
able conjecture: had the husband acted with ill intent against his
wife? Seneca approvingly calls it optimam quaestionem con-
iecturalem, and indeed it is one of the most important questions
raised by the theme. On the other hand in Contr. 1.5.8-9 we
hear of Cestius trying a conjectural question which is no more
than an innuendo, and is rightly condemned as such by Latro:
Cestius et coniecturalem quaestionem temptavit: an haec cum
raptore conluserit et in hoc rapta sit, ut huic opponeretur.
Latro aiebat non quidquid spargi posset suspiciose, id etiam
indicandum. colorem hunc esse, non quaestionem; earn quaestionem
esse, quae impleri argumentis possit. Cestius aiebat et hanc
Inventio 159
posse impleri argumentis. (Contr. 1.5.8-9 W)

From this we may deduce why guaestiones coniecturales are found so
rarely in the divisiones. Questions of fact had to be more than
mere insinuations in order to be classed as quaestiones rather
than as colores. Also, we are told in Contr. 1.2.14 that the
Apollodoreans objected in principle to themes which left a major
question of fact in doubt, such as that of Contr. 1.2, which does
not specify whether the would-be priestess is actually a virgin.
The Apollodoreans1 objections must have helped to make conjectural
questions unpopular in the schools of Seneca's time.
In his divisio of Contr. II.2 Latro included quaestiones of
all three types, but to do so was rare:
Contr. II.2 themai Vir et uxor iuraverunt, ut, si quid alteri
obtigisset, alter moreretur. vir peregre profectus misit nuntium
ad uxorem, qui diceret decessisse virum. uxor se praecipitavit.
recreata iubetur a patre relinquere virum; non vult. abdicatur.
Latro's divisio (Contr. II.2.5 M ) :
ius: an pater abdicare possit propter matrimonium.
an, etiamsi non malo adversus uxorem animo

- aequitas: [fuit] maritus fecit, tamen tam temerarius
et inconsultus relinquendus sit...
quaestio - .
. 7 . : an etiam malo adversus uxorem a m m o fecerit.
com ecturali s
Latrofs plans for Contr. I.I and 1.2 exemplify a more normal type
of divisio, in which coniectura is excluded, or allotted only a
very minor place in the scheme.
Contr. I.I themai
Liberi parentes alant aut vinciantur.
Duo fratres inter se dissidebant; alteri filius erat. patruus in
egestatem incidit; patre vetante adulescens ilium aluit; ob hoc
abdicatus tacuit. adoptatus a patruo est. patruus accepta
hereditate locuples factus est. egere coepit pater: vetante
patruo alit ilium, abdicatur.
Latro f s divisio (Contr. 1.1.13):
(1) an abdicatus non

desinat filius esse
(2) an is desinat qui non

tantum abdicatus sed
etiam ab alio adoptatus
est.
r-ius
an abdicari possit...an (3) an quisquis patrem non
necesse fuerit ilium patrem aluit puniatur
alere et ob id abdicari non
possit, quod fecit lege (4) an aliquam filii lex
cogente. excusationem accipiat
(5) an in hoc accipere

potuerit.
(1) an, etiamsi ille

indignus fuit qui
aleretur, hie tamen
-aequitas recte fecerit, qui
an abdicari debeat aluit
(2) an dignus fuerit, qui

aleretur.
Contr. 1.2 themai

Sacerdos casta e castis, pura e puris sit.
Quaedam virgo a piratis capta venit; empta a lenone et prostituta
est. venientes ad se exorabat stipem. militem, qui ad se
venerat, cum exorare non posset, conluctantem et vim inferentem
occidit. accusata et absoluta et remissa ad suos est: petit
sacerdotium.
Inventio 161
Latrofs divisio (Contr. I.2.13f.):
—(1) utrum castitas tantum

ad virginitatem re-
feratur an ad omnium
— I. an casta sit - turpium et obscenarum
rerum abstinentiam.
—(2) etiamsi ad virginitatem

tantum refertur
castitas, an haec
virgo sit.18
• ius
an per legem
fieri sacerdos
non possit
r—(1) an, etiamsi merito

occiderit hominem, pura
tamen non sit homicidio
I— II. an pura sit -
coinquinata
—(2) an merito occiderit

hominem innocentem uti
corpore prostituto
volentem.
an idonea sit tam infelix:
— (1) ut caperetur
— (2) ut veniret
— (3) ut lenoni potissimum

- aequitas
— (4) ut prostitueretur
etiamsi lex illi
non obstat, an
— (5) ut occidere hominem
sacerdotio idonea
sit. cogeretur
— (6) ut causam diceret.
In Contr. 1.1.13 Seneca tells us that the old type of divisio,

by which he means primarily the type which had been promoted by
Latro, was simpler than that now prevalent: divisio contro-
versiarum antiqua simplex fuit; recens utrum subtilior an tantum
operosior <sit> ipsi aestimabitis. It will be evident from the
samples of his divisiones given above that Latro had an eye for
the essential issues in a case. It appears from some remarks
Seneca makes that he made conscious efforts to make his divisiones

simpler than was customary in the schools, and that his intention
was to bring scholastic practice closer to that of the courts:
Latro semper contrahebat et quidquid poterat tuto relinquere
praeteriebat. itaque et quaestionum numerum minuebat et locos
numquam attrahebat; illos quoque, quos occupaverat, non diu
dicebat, sed valenter. hoc erat utique praeceptum eius, quaedam
declamatorem tamquam praetorem facere debere minuendae litis
causa, quod in hac controversia fecit. (Contr. VII.7.10)
In Contr, II.3.12 we find him elaborating on this teaching and
naming some issues which he reckons can safely be regarded as
settled:
Latro eleganter dicebat quasdam esse quaestiones, quae deberent
inter res iudicatas referri, tamquam, an quidquid optaverit vir
fortis aut tyrannicida accipere debeat: quasi iam pronuntiatum sit
non debere, nemo iam hanc quaestionem tractat, sicut ne illam
quidem, an quidquid pater imperat faciendum sit.19 inter has
putabat et hanc esse, an pater ob dementiam, quae morbo fieret,
tantum accusari a filio debeat; aiebat enim manifestum esse e lege
et de officio patris quaeri et fingi quasdam controversias, in
quibus pater furiosus probari non possit <nec> absolvi tamen
propter impietatem nimiam, libidinem foedam. quid ergo? aiebat;
numquam utar hac quaestione? utar, cum aliis deficiar.2^
However, Asinius Pollio was contemptuous of Latro*s attempts to
imitate forensic methods:
Pollio Asinius aiebat hoc Latronem videri tamquam forensem facere,
ut ineptas quaestiones circumcideret, <sed> in nulla magis ilium
re scholasticum deprehendi. remittit, inquit, earn quaestionem,
quae semper pro patribus valentissima est. ego [semper] scio
nulli a praetore curatorem dari, quia inicus pater sit aut impius,
sed quia furiosus; hoc autem in foro esse curatorem petere, quod
in scholastica dementiae agere.21 (ibid, 13)
Of course, no declaimer worth his salt would simply take over
and re-use the divisio of Latro or any of the other leading mas-
ters of rhetoric. He had to exercise his ingenuity to think of at
least one innovation. He might decide to revive one of the
quaestiones which Latro and others had rejected as res iudicatas,
or else he might ransack the theme and accompanying laws in search
of some previously unnoticed legal or ethical problem. For
instance in Contr, II.3 (Raptor raptae patrem exoravit, suum non
exorat. accusat dementiae) Gallio raised the question which Latro
passed over: whether a man can be accused dementiae for anything
other than simple insanity, and then went on to add the ingenious
quibble: an agi cum patre dementiae possit ob id, quod fecerit,
Inventio 163
non ob id quod facturus sit (Contr. II.3.14). The declaimer was

encouraged to frame new questions by a scholastic rule that one
should take into consideration every word in the laws accompanying
the controversia theme: praeceptum...quo iubemur ut, quotiens
possumus, de omnibus legis verbis controversi am faciamus (Contr.
1.2.15). A departure by the so-called novi declamatores22 from
f
Latro s plan, which is mentioned in Contr, 1.4.6, is based on just
such close reading of a declamatory law:
novi declamatores illam quaestionem temptaverunt ex verbo legis
natam 'adulterum cum adultera qui deprehenderit, dum utrumque
corpus interficiat, sine fraude sit.1: an nemo possit occidere
nisi qui deprehenderit...
Another source of new quaestiones which the novi declamatores
might look to was in the Greek tradition of declamation; see, for
example, Contr. 1.1.14: novi declamatores Graecis auctoribus ad-
iecerunt primam illam quaestionem: an adoptatus abdicari possit.
However, Seneca tells us of certain quaestiones which were
popular with Greek declaimers but flatly rejected by their Roman
counterparts. The basic disagreement lay in the fact that the
Greeks assumed that the tyrannicida and the vir fortis had certain
statutory rights whereas the Romans refused to accept that they
had more than a strong moral claim for privileged treatment:
Contr. 1.7.12: Graecorum improbam quaestionem satis erit in eius-
modi controversiis semel aut iterum adnotasse: an in tyrannicidam
uti pater hac lege possit. quasi sacras et publicas manus esse,
in quas ne piratae quidem licere quicquam putent. nostri hoc
genus quaestionis submoverunt.
Contr. 1.8.7: Graeci illam quaestionem primam solent temptare,
quam Romanae aures non ferunt: an vir fortis abdicari possit. non
video autem, quid adlaturi sint, quare non possit. nam quod et
vir fortis est et totiens fortiter fecit, non plus iuris illi ad-
fert, sed plus commendationis.
But the declaimers stopped a long way short of really trying to
bring controversiae into line with modern Roman legal practice.
So far we have been considering divisiones of an orthodox
type, in which quaestiones iuris and the issues arising from them
are treated before the tractatio aequitatis, and the quaestio
coniecturalis if there should be one. Sometimes declaimers broke
away from this type of format and preferred to present the ques-
tions at issue in a more artificial, but more striking, way. This
procedure was called figura dividere. Here are some examples:

Contr. 1.2.16: Albucius figura divisit controversiam; dixit enim:
putemus tres sacerdotium petere: unam quae capta est, alteram quae
prostitit, tertiam quae hominem occidit: omnibus nego; et sic
causam contra singulos egit.
Contr, II.1.23 W: Argentarius omnes priores transit partes, statim
ad hoc venit: debueritne patri parere an non debuerit; et in
figuram contulit declamationem. 'volof inquit faliquis filium
abdicet qui petit a patre paupere ut in adoptionem diviti daretur;
quam bonam1 inquit 'causam pater habebit! dicet hie...1, deinde
sic transit, cum declamasset earn controversiam, quae posita non
erat: fsi ille filius malam causam habet, ego bonam habeo.' con-
tulit suam causam cum illo.
Contr. II.4.9: Cestius bella figura egit: 'dementia1 inquit 'res
est sanitati contraria. non quaeram extra exemplar sani hominis,
ad quod patris mei <mentem exi>gam: ipsum sibi comparabo. fuit
aliquando sanus: tune quid faciebat? oderat luxuriam, vitia
castigabat. hunc tarn severum senem putabitis sanum, si vobis in
lupanari ostendero?' sic declamavit, ut patri accusatorem patrem
daret et ilium argueret sibi ipsum comparando.
Contr. IX.6.14: Silo a parte patris comparationem fecit inter se
matris et filiae <et> totam hac figura declamavit...
The divisio of a controversia was similar, but not identical,
to the organization of arguments for a law-court speech. We have
noted the tendency of some declaimers to prune away questions as
res iudicatas, some of which it would be vitally important to
raise in a real court case (Contr. II.3.12f.), but the principal
difference is that outlined in Contr. X.5.12. It appears that one
was expected to compress into a single declamation what an orator
would take two actiones to discuss:
hanc controversiam magna pars declamatorum sic dixit, <vel>ut
<non> controversiam divideret, sed accusationem, quomodo solent
ordinare actionem suam in foro qui primo loco accusant; in
scholastica, quia non duobus dicitur locis, semper non dicendum
tantum sed respondendum est. obiciunt, quod hominem torserit,
quod Olynthium, quod deorum supplicia imitatus sit, quod tabulam
in templo Minervae posuerit. si Parrhasius responsurus non est,
satis bene dividunt. nihil est autem turpius quam aut earn contro-
versiam declamare, in qua nihil ab altera parte responderi possit,
aut non refellere, si responderi potest.
Seneca seems quite content here with the scholastic custom; his
rejection of the tactics appropriate to a real-life actio prima
may be compared with Latro's views on the treatment of the contro-
versia about Popillius:
Latroni non placebat ilium sic accusari, quomodo quidam
Inventio 165
accusaverunt: obicio tibi, quod occidisti hominem, quod civem,

quod senatorem, quod Ciceronem, quod patronum tuum. hac enim
ratione non adgravari indignationem sed fatigari. statim illo
veniendum est ad quod properat auditor; nam in reliquis adeo bonam
causam habet Popillius ut, detracto eo quod patronum occidit,
nihil negoti habiturus sit; patrocinium eius est civilis belli
necessitas. itaque nolo per illos reum gradus ducere <per> quos
potest tutus evadere. licuit enim in bello et hominem et civem et
senatorem et consularem occidere, ne in hoc quidem crimen est,
quod Ciceronem, sed quod patronum. naturale est autem ut, quod in
nullo patrono fieri oportuit, indignius sit factum in Cicerone
patrono. (Contr. VII.2.8 M)
That at some stage it had become customary - as it had not
been when Antiphon's Tetralogies were composed - for the functions
of two actiones to be covered in one scholastic declamation is
curious, just as it seems extremely strange that it was very much
the exception rather than the rule among Roman declaimers to
arrange any sort of mock trial in which prosecution was pitted
against defence. (The only place where Seneca seems to allude to
any such arrangement comes in Contr, IV pr. 3, where he says of
Asinius Pollio,
audivi autem ilium et viridem et postea iam senem, cum Marcello
Aesernino nepoti suo quasi praeciperet. audiebat ilium dicentem
et primum disputabat de ilia parte, quam Marcellus dixerat:
praetermissa ostendebat, tacta leviter implebat, vitiosa co-
arguebat. deinde dicebat partem contrariam.)
It is evident from what Seneca and Latro say in Contr, X.5.12
and VII.2.8 that in some ways it may well have been easier to
organize the actio prima of a simple prosecution than a well-
balanced controversia, but, as critics of the schools were quick
to point out, the declaimer's task was much easier than the ora-
tor's in one important respect. In the actio secunda an orator
would have to refute the view of the case presented by his op-
ponent in his first speech. This might be an enormously difficult
task. The declaimer met with no such opposition. It was left to
him to imagine what objections his opponent might make to his own
interpretation of the case, and there was a strong temptation not
to credit one's imagined adversary with much intelligence:
adversarios quamvis fatuos fingunt: respondent ill is et quae
volunt et cum volunt (Contr. IX pr. 2).
COLORES
Seneca used the term color in a way which had no precedent in
Cicerofs works. To Cicero color had meant the distinctive charac-
ter of a style,23 or else colourfulness of expression.24 Xpwya is
used similarly by most Greek writers on rhetoric (LSJ s.v. IV 1,2);
it is only in some late examples of theory relating to axaaeus and
modes of defence that we find it used to mean what color meant to
SenecaTs rhetorician friends, that is, the complexion which a
speaker gave to the actions of the accused in a lawsuit or judicial
declamation (LSJ s.v. IV 4). How it was that the Roman declaimers
came to use color in this sense would have to remain a complete
mystery, were it not that D. Matthes, in his Teubner collection of
the fragments of Hermagoras of Temnos, has extracted from obscurity
two significant scholia on the ITepu axdaewv of Hermogenes of
Tarsus. These show us that the Senecan use of color was in line
with Hermagorean usage:
ri xd dn' dpxns oifxPi> xeAous or)\ieZcL Ttououyevos 6 xaxriyopos
6oxeu $udceadaL T O V 6txaaxnv xal TieCSetv
C UJS xoO eyxaAouyevou
evexev xaDxa leitoCrixev 6 cpeuywv, 6eu itpos xo\3xo dywvCceaâL xov
(peuyovxa xal yn xou eibcpepoyevou d6uxnyaxos evexev cpaaxeuv
n etpnxevau rj xo uddos auyBeprixevat • xoOxo yap eaxLV n
xns auxuas, o xP^P a TtpoaayopeuouaLV ot 'Epyayopeuou*
x 6e xwv arc' dpxHs « X P L xeAous 9 Xuaus 6e yexd d
wau eaxat r\ dvxuaxaxLxri ff yexaaxaxuxn r\ dvxeyxAnyotxtxri rf
auyyvwyovLxn- dvxuaxaxtxri yev, edv ocpeAos xt TipoXaPajyeâ
f
dvaAay$dva) xous duoxripuxxous, uva yn diropouvxes eul xAoinv n
GTutBouAriv xpaircavxau1 . xdxeuva 6e oyouws dvxuaxaxbxd, 6'x' av 6'nAa
e x w v xptvnxau xupavvC6og enuêaews* epet ydp 6'xu fcpuAdxx(Ai x^j TioAet
ets dvayxauov eitudoauv1 . .. xal duo eAeou eaxl yexddeaus, ws ercl xoO
ddixovxos x6 veoa(payes awya, O I L TeAeaJv edauxov'. (Porphyrius in
Hermog. Stat., Rh. Gr. IV 397.8 Walz = Hermagoras fr. 14a Matthes)
The scholion given by Matthes as fr. 14b (incert. auct. schol.
min. in Hermog. Stat., Rh. Gr. VII 308 adn. Walz) repeats the
statement that the Hermagoreans gave the name xP&V101 to tne
yeTot-
Seats xfjs auxuas in the course of an explanation of the way in
which a prosecutor may rebut a plea of the type called auyyvwyn.
It may be deduced that the term XP&VK* w a s used with reference
to pleas in that part of Hermagorean theory, alluded to by Quin-
tilian (VII.4.4ff.), in which types of defence were first dis-
tinguished as either xax' dvxtAny^tv (qua ipsum factum, quod
obicitur, dicimus honestum esse), or xax' dvxtdeatv (in quo factum
Inventio 167
per se improbabile adsumptis extrinsecus auxiliis tuemur). At any

rate, the adjectives which Porphyrius applies to the yexadeaus xns
auxCag, which the Hermagoreans called xP&ya, are derived from the
names which the Greek theorists gave to the four standard var-
ieties of defence xai' dvxudeauv, namely: dvxCaxaaus (a plea that
an action, though criminal, was justified by some benefit which
ensued); yexaaxaaus (a plea by which one transferred the blame for
the crime on to some other person or thing) ; avxeyxXriyct (a plea
that the criminal act had been committed in response to some prior
injury); and auyyvwyn (a plea that the crime was of a pardonable
nature). 25
It is curious that Seneca the Elder should use the term color
in the sense which the Hermagorei gave to xP&Va> while Cicero and
Quintilian never do so in all their elaborate accounts of the rel-
evant parts of Hermagorean theory.26 Here, as in a number of
other cases, the schools of rhetoric from which Seneca derived his
terminology appear to have taken over their technical vocabulary
from sources quite outside the Ciceronian tradition.
It is unclear whether Seneca was familiar with a classifi-
cation of colores under the headings: dvxuaxaaus (comparatio, com-
parativum, compensatio); yexaaxaaus (remotio criminis, trans-
latio); avxeyxAnya (relatio/translatio criminis); auyyvwyn (con-
cessio, excusatio, venia),27 He certainly shows no inclination to
use any such terms when discussing colores, but that could be be-
cause he presupposes familiarity with them. He did know of a work
on colores in four books - was it laid out according to the four-
fold classification of types of defence? - by Iunius Otho, a rhet-
orician of his own day, and in the course of describing this work
he refers to a fault in it as having been derived ab antiquis qui
artem dicendi tradebant (Contr. II.1.33). Otho included so many
dreams among his colores that Gallio wittily called his books
Antiphontis libros (the allusion being to the work IlepL xpuaeoos
oveCpcov by the fifth-century sophist Antiphon);28 OthoTs willing-
ness to approve of colores which could not be supported or refuted
by argument2^ was, we are told, a pernicious tendency inherited
from his predecessors in the field. Seneca expresses disapproval
on the ground that a speaker using such colores will normally be
suspected of not telling the truth.30 The reference to antiquis

qui artem dicendi tradebant is intriguing. It has to be counted
among the handful of indications that Seneca the Elder, though
more influenced by oral school criticism than by Cicero, cannot
have been entirely unread in rhetorical theory - that is, unless
we argue that he may be quoting criticisms by Gallio throughout
the discussion of Otho's books, or that the antiqui in question
may have been no earlier than the exponents of the antiqua divisio
mentioned in Contr. 1.1.13.
Seneca nowhere defines the term color, but he mentions dis-
tinctions drawn by Latro between it and the terms quaestio and
defensio. According to Latro (Contr. 1.5.9), a mere innuendo
should not be classed as a quaestio, but rather as a color, for a
quaestio has to be capable of being backed up by argumentation:
earn quaestionem esse, quae impleri argumentis possit. In Contr.
VII.6.17 we find Latro distinguishing between color and defensio:
a parte patris magis defensione opus esse dicebat Latro quam
colore; evidently he thought of a color as being something flims-
ier than a full-scale defence. Note that Latrofs view that
colores might include insubstantial suggestions, not necessarily
capable of proof, is the view criticized by Seneca - or was it
Gallio? - in Contr. II.1.33, if my interpretation of that passage
is correct, as a vitium derived ab antiquis, qui artem dicendi
tradebant.
According to Seneca one might introduce a color right at the
beginning of a declamation and use it throughout, e.g. Contr. I.I.
24: hoc colore usus est, quern statim a principio induxit ;
Contr. 1.4.8: Albucius non narravit, sed hoc colore egit ab initio
usque ad finem. Alternatively one might bring colores into play
primarily in the narratio: Contr. VII.1.20: Latro ilium introduxit
colorem rectum in narratione, quo per to tarn actionem usus est;
Contr. 1.4.8: Silo hoc colore narravit (cf. Contr. X.2.13, 15, 17).
Sometimes people used up all the resources of their colores in the
narratio, but this, at least in the view of Pollio, was not a good
idea:
Asinius Pollio dicebat colorem in narratione ostendendum, in
argumentis exsequendum. non prudenter facere eos, qui in
Inventio 169
narratione omnia instrumenta coloris consumerent; nam et plus

illos ponere quam narratio desiderasset et minus quam probatio.
(Contr. IV. 3 extra)
In Contr. VII.1.21 Albucius includes a number of colores in his
argumentation: Albucius in argumentis plura posuit et omnes fere
colores contrectavit.
Some colores are described by Seneca as straightforward,
simple, or openly expressed. Colores by Latro are frequently so
described:
Contr. 1.1.16 M (on the young man who, though disinherited by
father and uncle in turn, supported them in their hour of need):
Latro colorem simplicem pro adulescente <habuit>: habere non quo
excuset, sed quo glorietur.
Contr. II.6.6 (on the father who followed his son's example in
riotous living):
Cestius <a> parte patris aiebat simulationem luxuriae significandam
magis quam profitendam...Latro aperte putabat simulationem con-
fitendam.
Contr. VII.1.20 (on the young man who would not punish his brother
for parricide in the traditional manner):
Latro ilium introduxit colorem rectum in narratione, quo per totam
actionem usus est: non potui occidere.
Contr. VII.2.10 (on Cicero's assassin):
colorem pro Popillio Latro simplicem habuit: necessitate coactum
fecisse.
Simple colores are also ascribed to Pollio: ipse autem hoc colore
usus est, quern aiebat simplicissimum (Contr. IV.6 extra); and to
Romanius Hispo: ...simpliciter putavit agendum (Contr. VII.7.14).
Both criticize the over-subtle efforts of other declaimers.
Scaurus was another person critical of unnecessary subtlety: he
described Iunius Otho as a man who would whisper public announce-
ments in one's ear (acta in aurem legere, Contr. II.1.39).
In the case of some controversiae it was immediately obvious
what sort of color one should adopt. It might be that one simply
had to inveigh against one of the characters in the theme, e.g.
Contr. 1.2.21: alterius partis color nihil habet difficultatis...
dicendum est in puellam vehementer, non sordide nee obscene; Contr.
II.3.17: omnes infamaverunt raptae patrem quasi cum raptore con-
ludentem; Contr. IV.8 extra: omnes invecti sunt in libertum.
Other controversiae posed more problems, and the need for a
more subtle approach in such cases was recognized by Seneca. He

commends Otho for his success in dealing with difficiles contro-
versias...in quibus inter silentium et detectionem medio
temperamento opus erat (Contr. II.1.33), even though he appreci-
ates the wittiness of ScaurusT remark about his whispered announce-
ments. To use Seneca's terminology, difficult themes needed to be
treated suspiciose (Contr. II.1.34), per suspiciones et figuras
(Contr. VII.1.20), schemate (Contr. II.4.10). Otho's treatment of
Contr. II.1 provides one example of an approach of this kind. The
theme was this:
Dives tres filios abdicavit. petit a paupere unicum filium in
adoptionem. pauper dare vult; nolentem ire abdicat.
Otho took the part of the poor father and gave veiled indications
of his motives in a series of sententiae:
itaque et hanc controversiam hoc colore dixit, tamquam in
emendationem abdicatorum et reconciliationis causa faceret. hoc
non detegebat, sed omnibus sententiis utebatur ad hoc tendentibus..
(Contr. II.1.37)
There were often disagreements as to whether the subtle
approach was better than the straightforward. For example, when
Cestius and Latro considered how to treat Contr. II.6 (Quidam
luxuriante filio luxuriari coepit. filius accusat patrem
dementiae), they disagreed as to which color was best:
Cestius <a> parte patris aiebat simulationem luxuriae significandam
magis quam profitendam. ita, inquit, apparebit ilium simulasse si
etiamnunc simulat; si desinit simulare, ostendit iam sibi nihil
opus esse eo consilio, quasi filius emendatus sit; emendatum autem
esse non concessit, et adsidue dixit nihil magis se quam inter-
vallum hoc luxuriae timere; intermissa vitia vehementius surgere.
Latro aperte putabat simulationem confitendam. incipio, inquit,
non tantum honestum senem sed prudentem defendere si, quod vitium
videri poterat, efficio consilium. quare potius significet quam
dicat frugi <se> esse? (Contr. II.6.6 W)
Latro, as so often, champions the cause of directness.
With the remarks of Seneca and the declaimers as to where a
controversia called for subtle colores and where simpler tactics
were preferable, it is relevant to compare Quintilian's obser-
vations on the so-called controversiae figuratae (IX.1.14; 2.65ff.)
Seneca, as we have seen, refers to occasions when declaimers pro-
ceeded per suspiciones et figuras or schemate. Quintilian tells
us that one curious usage of the term schema which was current in
Inventio 171
the common parlance of his day derived from the definition of that
term given by Zoilus (the fourth-century critic better known for
his attacks on Homer). Zoilus defined axnya as a device by which
one gave the impression of saying something other than what one
was actually saying. Quintilian's contemporaries used the term
controversia figurata in a way which presupposed that one accepted
this definition of axnya/figura (IX.1.14). According to Quinti-
lian, declaimers gave the name figuratae to controversiae of three
sorts:
eius triplex usus est: unus si dicere palam parum tutum est, alter
si non decet, tertius qui venustatis modo gratia adhibetur et ipsa
novitate ac varietate magis quam si relatio sit recta delectat.
(IX.2.66)
Quintilian discusses at great length {ibid. 65-80) the excessive
subtlety which was often encouraged by the scholastic teaching
about this sort of figuration; he would have agreed with Latrofs
statement about figures in general in Contr, I pr. 24: schema
negabat decoris causa invention, sed subsidii...
From time to time Seneca describes a color as durus, strictus,
or asper. A color will be so described if it involves the attri-
bution of hatred or unusual severity to a character in the theme,
sometimes even to the person whom one is defending. Romanius
Hispo and Latro were noted for their readiness to adopt this kind
of color:
Contr, IX.3.11: Hispo Romanius erat natura qui asperiorem dicendi
viam sequeretur; itaque hoc colore egit, ut inveheretur tamquam in
malum patrem et diceret crudeliter exponentem, perfide recipientem.
Contr. X pr. 15: Latro...dicebat quosdam esse colores prima facie
duros et asperos: eos non posse nisi actione probari...multa se
non persuadere iudici, sed auferre.
Further examples may be found in:
Contr. 1.8.8 (on the war hero who is disinherited by his father
for wishing to enter battle for the fourth time)
colorem a parte patris quidam duriorem fecerunt; voluerunt enim
videri invisum filio patrem: itaque ilium malle cum hostibus
vivere quam cum patre;
and Contr. VII.1.24 (on the man who casts his brother adrift in an
open boat instead of punishing him for parricide in the traditional
manner)
Hispanus duro colore usus est: hoc, inquit, supplicium tamquam
gravius elegi.
The opposite of durus and asper as used in such contexts is tutus

(Contr. X pr. 16), and the verb used to denote the toning down of
harshness in colores is temperare (Contr, 1.1.24).
Sometimes the declaimer would explain the defendant's actions
as the result of some more or less irrational impulse.31 His
defence might be that he was overcome by momentary anger or mad-
ness, or that he had been motivated by portents, a vision, or, as
we have noted, a dream. Declaimers who adopted such colores laid
themselves open to criticism and ridicule from some quarters, but
they used them nonetheless, as they had the advantage over more
subtle defences of providing opportunities for emotional out-
pourings in the grand style.
For example, the defence of the father in Contr. 1.7, who
wrote to the pirates offering them double the ransom demanded if
they cut off the hands of his son (who admittedly had twice com-
mitted fratricide, but with just cause), was one of the hardest
tasks the declaimers ever set themselves. Fuscus, Pompeius Silo,
and Sparsus, taking the part of the cruel father, all chose to
hold forth about the anger they had felt at the son's various
actions:
colorem pro patre alius alium introduxit. Fuscus iratum se illi
confessus est fuisse, quod fratrem in conspectu patris occidisset,
et huic loco vehementer institit, quom nemo hoc tyrannus, nemo
pirata fecisset... Silo Pompeius et ipse iram fassus est: aiebat
enim non habiturum fidem si negasset iratum fuisse; sed irae
causam non dixit quam Fuscus; transeundas aiebat eas offensas
quibus ille gloriaretur; hanc causam posuit, quod relictus esset
ab unico filio, quod invito se navigasset, cum videret <se> senem,
orbum, iam paene egentem: iam turn ilium fugisse ne aleret; et ad
preces patrem deduxit et rogavit in epilogo filium. et Sparsum
hoc colore declamasse memini, hominem inter scholasticos sanum,
inter sanos scholasticum. (Contr. I.7.14f.)
Latro went one step further: the father's action, according to
him, was not the result of mere anger - nor, as Cestius (Contr.
1.7.16) had suggested, was it all part of an ingenious plot to
avoid paying the ransom - the balance of his mind had been totally
deranged:
Latro totum se ab istis removit coloribus, et advocavit vires suas
tanto totius actionis impetu ut attonitos homines tenuerit; hoc
enim colore usus est: nescio quid scripserim. olim iam mihi ex-
cussa mens est. ex quo vidi filium unum in arce, alterum in
Inventio 173
adulterio, tertium in parricidio, ex quo respersus sum fili

morientis sanguine, ex quo relictus sum solus, orbus, senex, odi
meos. hie color illius viribus adprobandus est; quanta enim vi
opus est ut aliquis accusando se miserabilem faciat! (Contr.
1.7.16)
The most notable specialist in portents and visions was
Arellius Fuscus:
Contr. 1.1.16: Fuscus ilium colorem introduxit, quo frequenter uti
solebat, religionis: movit, inquit, me natura, movit pietas, movit
humanorum casuum tam manifesto approbata exemplo varietas. stare
ante oculos Fortuna videbatur et dicere talia: esuriunt qui suos
non alunt. (Cf. Contr. 1.8.15; II.1.27)
Iunius Otho was famous for his dreams:
Contr. VII.7.15: erat autem ex somniatoribus Otho: ubicumque ilium
defecerat color, somnium narrabat. (Cf. Contr. II.1.33)
Sometimes declaimers used more than one color in a contro-
versia. Arellius Fuscus did not restrict himself to describing
his anger at the fratricidal son in Contr. 1.7; he went on to ex-
plain his cruel letter to the pirates as follows:
illud ad excusandam epistulae crudelitatem adiciebat: scripsi
piratis non eo animo ut manus tibi praeciderentur, sed ut ex-
probrarem tibi cruentatas in conspectu patris fraterno sanguine
manus. tuto autem scribebam; sciebam enim piratas non facturos
nisi pecuniam accepissent, quam non mittebam: denique nee praecid-
erunt. et si sperassent, utique praecidissent: sed apparuit illas
epistulas irascentis esse, non promittentis. (Contr. 1.7.14 W)
One might have thought this was a rather risky scheme! When a de-
claimer uses more than one color Seneca calls the procedure
miscere colores (Contr. 1.7.17; IV.6 extra), and the complex
colouring itself, mixtus color (Contr. IV.6 extra). As it was
generally best for a defendant speaking on his own behalf to use
one plea consistently, if he were to sound convincing, it was
thought wise for the declaimer to assume the role of an advocate
if he wished to mix many colores. For example, Albucius (Contr.
1.7.17) impersonated an advocate when attempting to defend by
various means the father who had offered the pirates double the
ransom money if they would cut his son's hands off: Albucius omnes
colores miscuit, et, ut hoc liberum sit, patronum patri dedit nee
voluit narrare. When mixing colores one had to be very careful
not to appear to 'protest too much1. When explaining in Contr.
IV.6 one's refusal to say which of two long-absent sons was the
offspring of one's second wife, it was better, as Pollio pointed
out (Contr. 4.6 extra), to use the simple color, fscio, sed non
indico, quia pueris hoc utile est1, than to confuse the issue by
mixing two colores, fnescio, sed etiam si scirem, non indicarem.'
When deciding on colores the declaimer had to avoid, above
all, choosing one which ran contrary to his own interests. For
instance, if he were arguing that the condemned priestess who had
survived her fall from the cliff-top ought to be thrown down again,
it would be a mistake to claim that any type of divine inter-
vention had taken place. Cestius points this out in Contr. 1.3.9:
contra sacerdotem qui<dam> dixerunt: videri deos infestos illi in
hoc earn servasse, ut diutius torqueretur. aiebat Cestius malle se
casu videri factum quam deorum voluntate; nam si semel illos
intervenire huic rei fatemur, manifestius erit <contra> poenam
servatam esse sacerdotem quam in poenam.
Similarly, a declaimer taking the part of the poor father in
Contr. II.1 was liable to lose the sympathy of his hearers if he
expressed a desire for wealth:
color a parte patris aliquid curae desiderat. quidam induxerunt
patrem cupidum divitiarum, quod invidiosum est in hoc visum, 2
quia ita divitias filio dare vult ut filiis eripiat. itaque Latro
optimo colore usus est: in hoc, inquit, te in adoptionem volo
dare, ut facilius per te abdicati reconcilientur. (Contr. II.1.30)
Severe critics were hard on declaimers who, instead of work-
ing out a good color consistently, yielded to the temptation of
bringing in sententiae which were more elegant than useful to
their case. Thus Messala criticized the sentimental picture con-
jured up by Albucius of an illegitimate child who followed his
grandfather out of the room as if of right (and was consequently
adopted by him):
Albucius ethicos, ut multi putant, dixit - certe laudatum est cum
diceret -: exeuntem <me> puer secutus est. non probabat hanc
Messala sententiam: non habet, inquit, fiduciam si mavult videri
recepisse puerum quam adduxisse; et sine ratione est adoptatum
esse non quia debuerit sed quia secutus est. (Contr. II.4.8)
This seems an unnecessarily grudging criticism. Seneca appears to
have been more tolerant. At any rate aridi declamatores who keep
too faithfully to the colores they have chosen, and refuse to be
led astray by any opportunity for a pleasing schema or sententia,
are likened by him to plain women, chaste for lack of a seducer
(Contr. II.1.24).
It was against the rules of controversia-writing to use
Inventio 175
colores that introduced elements which were not mentioned in the

thema, or which contradicted in any respect the facts of the case
as set out in it. 33 Take, for example, Contr. IX.5, the theme of
which ran as follows:
Quidam duos filios sub noverca amisit: dubia cruditatis et veneni
signa insecuta sunt. tertium filium eius maternus avus rapuit,
qui ad visendos aegros non fuerat admissus. quaerenti patri per
praeconem dixit apud se esse. accusatur de vi.
A declaimer speaking for the prosecution had to think up a very
good excuse for the fact that the grandfather had not been ad-
mitted to see the dying boys. Even good declaimers found it hard
not to introduce factors which could not legitimately be deduced
from the theme:
Silonis Pompei color fuit, ut Latroni videbatur, qui controversiae
repugnaret; dixit enim venisse avum ad inbecillum puerum. ad
aegros non semper admitti, utique ad eos qui graviter aegrotent;
saepe et patrem non admissum; sic avo quoque intempestive venienti
dictum: 'nunc non potesT; statim cum convicio abisse. in altero
idem fecisse. Latro aiebat hunc colorem optumum esse si res ita
esset, sed recipi non posse, quia ponatur: 'non est admissus1; sub
hoc themate intellegere nos non hoc illi dictum: 'nunc non potes1,
sed 'ex toto non potes'. (Contr. IX.5.10)
It was also not thought allowable to make any claim about the
facts of the case which could not be substantiated without a wit-
ness. Arellius Fuscus is criticized in Contr. VII.2.12 W for
breaking this rule:
Fuscus Arellius hoc colore usus est: Antoni se partem secutum ut,
si quid posset, Ciceroni prodesset; facta proscriptione ad genua
se Antoni procidisse, deprecatum esse pro Cicerone; offensum
Antonium dixisse: 'eo magis occide quern mori non vis.' hie color
displicebat Passieno, quia tad testem ducitt; nam, si hoc fecit
Popillius, non tantum quod defendat non habet sed habet quod
glorietur.
Whatever Seneca wrote where the manuscripts give us the doubtful
reading ad testem ducit, which is obelized by Winterbottom,3*4 it
seems certain that Passienus' objection must have been to the lack
of any witness's evidence to prove that Popillius had acted as
honourably as Fuscus was making out. A color contrived on behalf
of Parrhasius, in the controversia in which he is charged laesae
rei publicae for torturing an aged Olynthian whom he had bought to
serve as his model for a painting of Prometheus, was open to a
similar objection:
tGallionist color intolerabilis est; dixit enim <se> senem ex

noxiis Olynthiis emisse; quod si illi licet fingere, non video
quare non eadem opera dicat et conscium proditionis Lastheni
fuisse et se poenae causa torsisse. (Contr. X.5.18 W)
The inventor of colores had, above all, to aim at verisi-
militude. The more distinguished declaimers regularly attained
this, but the lunatic fringe provided a wide range of far-fetched
and totally unconvincing colores to amuse Seneca and his friends:
Contr. 1.3.11 (on the priestess thrown from the cliff)
Pastor Aietius hanc controversiam apud Cestium dixit iam senator,
et hunc colorem optimum putavit: sic veneficiis corpus induruit ut
saxa reverberet inultum. Cestius hunc corripuit et dixit: fhoc
est quare ego auditores meos invitem ad alios audiendos ire?f
Contr. VII.3.8 (on the ter abdicatus who was discovered mixing
poison)
Murredius pro cetero suo stupore dixit medicamentum se parasse ad
somnum, quia adsiduae sollicitudines vigiliarum sibi consuetudinem
<fecerint>.
Contr. IX.4.22 (on the young man who beat his father at the com-
mand of the tyrant, after his brother had thrown himself from the
citadel to avoid doing so)
Murredius non degeneravit in hac controversia; nam colorem
stultissimum induxit: voluit, inquit, et hie sequi fratris
exemplum: dum retineo, dum luctor, visus est patrem cecidisse.
Convincing characterization was important, especially in certain
controversiae which were described as ethicae (Contr. II.2.12).
These were the only controversiae which Ovid cared to declaim, and
presumably Contr. II.2, from which we have an Ovidian extract, is
an example of one. Fabianus found difficulty after his philo-
sophical conversion with the imitating of other people's emotions
that was required of the declaimer: cum veros conpressisset ad-
fectus et iram doloremque procul expulisset, parum bene imitari
poterat guae effugerat (Contr. II pr. 2 ) . The convincing
portrayal of character could be a delicate matter: the declaimer
had to be careful, when bringing out the merits of the person
whose part he had taken, not to be suspected of some related
fault. Thus Aeschines the declaimer was criticized for the way in
which he portrayed the father who prevented his valiant son from
going into battle for a fourth time:
hoc loco Aeschines ex novis declamatoribus, cum diceret: non me
gloria cupidiorem tui fecit, non omnibus admiranda virtus:
Inventio 177
f
confiteborf inquit fadfectus patris, quos ut quisque volet inter-
pretetur: OUTWS OIV waL 6euAov ecptXouv. f videbatur hie, dum
indulgentiam exprimit, non servasse dignitatem patris. (Contr.
1.8.11)
The question of the appropriateness of a declaimerfs style to the
character he was trying to portray is touched upon by Seneca in
Contr. 1.8.16:
Dorion dixit rem paulo quidem elatiorem quam pressa et civilis
oratio recepit, sed qua egregie attonitos patris adfectus
exprimeret.
Latro discusses in Contr. VII.4.6 the style to be adopted when
pleading on behalf of someone who seeks pity rather than vengeance:
Latro dixit pro matre summisse et leniter agendum, non enim,
inquit, vindictam sed misericordiam quaerit, et cum eo adulescente
consistit in quo ita exigit pietatem ut impediat. aiebat itaque
verbis quoque horridioribus abstinendum quotiens talis materia
incidisset; ipsam orationem ad habitum eius quern movere volumus
adfectus molliendam.
The declaimer also had to bear in mind which characters
needed to be treated with particular respect. A son must obvi-
ously not speak harshly about his father, and, in view of this,
Pompeius Silo put the prosecution of the cruel father in Contr.
1.7 into the hands of an advocate:
...descripsit mores hominis impii, cruenti, quia per liberos non
posset per piratas tyrannidem exercentis: quae ut liberius
diceret, patronum filio dedit. (Contr. 1.7.13)
Similarly, the poor man's son in Contr. II.1 could not simply
inveigh against the rich man who wished to adopt him:
de colore magis quaesitum est: an adulescens debeat in divitem
aliquid dicere. quid enim faciet? dicet in eum qui tantum
honoris illi habet, et in amicum paternum, non dicet in eum quern
fugit? (Contr. II.1.24)
Fabianus found a way out of this difficulty by attacking riches in
general, rather than the rich man in particular (Contr. II.1.25).
There are a few indications that a certain verecundia was rec-
ommended if one had to inveigh against a woman:
Contr. 1.2.21: dicendum est in puellam vehementer, non sordide nee
obscene.
Contr. VII.1.20: fuerunt et qui in novercam inveherentur; fuerunt
et alii, qui non quidem palam dicerent, sed per suspiciones et
figuras, quam rem non probabat Passienus et aiebat minus verecundum
esse aut tolerabile infamare noveram quam accusare.
Slaves, on the other hand, were held to be of no account at
all. One might have thought that the theme of Contr. VII.6 would
be seen as a golden opportunity for meditations to the effect that
misfortune might reduce any man to slavery:
Tyrannus permisit servis dominis interemptis dominas suas rapere.
profugerunt principes civitatis; inter eos qui filium et filiam
habebat profectus est peregre. cum omnes servi dominas suas
vitiassent, servos eius virginem servavit. occiso tyranno reversi
sunt principes; in crucem servos sustulerunt; ille manu misit et
filiam conlocavit. accusatur a filio dementiae.
Perhaps it is slightly unsafe to draw conclusions from the number
of quotations taken by Seneca from each of the two sides of a
controversial but it seems striking that he quotes far more de-
claimers as having prosecuted the father of the bride than as
having defended him. Of those who spoke for the defence, only
Albucius seems to have adopted the philosophical approach:
Albucius et philosophatus est: dixit neminem natura liberum esse,
neminem servum; haec postea nomina singulis inposuisse Fortunam.
denique, inquit, scis et nos nuper servos fuisse. rettulit
Servium regem. (Contr. VII.6.18)
The standard attitude of the declaimers seems to have been that
the father's action was exceptionally hard to excuse; as Latro put
it: a parte patris magis defensione opus esse...guam colore
(Contr. VII.6.17).
2 DISPOSITIO
A finished controversia normally consisted of four main sections:

the principium or prooemium (e.g. Contr. I.1.24f.); the narratio
(e.g. Contr. 1.1.21); the argumenta or argumentatio (e.g. Contr.
1.6.9; II.2.12) and the epilogus (e.g. Contr. IV pr. 8 ) . l A suas-
oria did not require a narratio, and so seems to have consisted
simply of a principium (e.g. Suas. 7.14) followed by various
quaestiones, and presumably some kind of conclusion (see e.g.
Contr. VII.7.19, quoted below), though Seneca does not in fact
mention any epilogus in his extant surveys of suasoriae.^ Seneca
assumes a knowledge of the nature and functions of these sections
of a declamation, and restricts his attention to details of treat-
ment and departures from the norm.
About the prooemium he has little to say. There is an
interesting reference, though, in Contr. VII.7.19, to the way
Cestius criticized his pupils when they used an unsubtle trick
which he called echo, that is, when they opened a declamation with
a quotation which they would bring in once more at the end:
illud et in hac controversia et in omni vitandum aiebat Cestius,
quotiens aliqua vox poneretur, ne ad illam quasi ad sententiam
decurreremus. sicut in hac apud Cestium quidam auditor eius hoc
modo coepit: fut verbis ducis vestri, iudices, incipiam, cavete
proditionem1; sic finivit declamationem ut diceret: ffinio quibus
vitam finit imperator: cavete proditionem.' hoc sententiae genus
Cestius echo vocabat et <sic> dicenti discipulo statim exclamabat:
uuepxriv nx^: u t i-n ilia suasoria in qua deliberat Alexander an
Oceanum naviget cum exaudita vox esset: Tquousque invicte?f ab
<hac> ipsa voce quidam coepit declamare et in hac desit; ait illi
Cestius desinenti: ev aou uev Xri^co, oeo 6' apôyau. et alteri,
cum descriptis Alexandri victoriis, gentibus perdomitis, novissime
poneret: fquousque invicte?1, exclamavit Cestius: tu autem quo-
usque? (Contr. VII.7.19 W)
Arellius Fuscus is said to have composed his principia, like his
narrationes and argumenta, in a very arid style quite different
from the florid elaboration in which he indulged to excess in
descriptive passages (Contr. II pr. 1 ) . Fuscus stood in marked
contrast to Passienus, who attracted a large audience for his

principia and epilogi, according to Cassius Severus (Contr. Ill
pr. 10), but lost it while he delivered the body of his speech:
Passienus noster cum coepit dicere, secundum principium statim
fuga fit, ad epilogum omnes revertimur, media tan turn quibus
necesse est audiunt.
Seneca singles out for special approval an ingenious device
for making the transition from prooemium to narratio:
Hermagoras in hac controversia transit a prooemio in narrationem
eleganter, rarissimo quidem genere, ut <in> eadem re transitus
esset, sententia esset, schema esset, sed, ut Latroni placebat,
schema quod vulnerat, non quod titillat: ...ex altera parte
transit a prooemio in narrationem Gallio et ipse per sententiam
sic: quidni filium mihi nolim cum isto communem esse, cum quo
utinam communem nee patrem habuissem? (Contr. 1.1.25)
Quintilian was to take a much less favourable view of this
practice:
ilia vero frigida et puerilis est in scholis adfectatio, ut ipse
transitus efficiat aliquam utique sententiam et huius velut prae-
stigiae plausum petat, ut Ovidius lascivire in Metamorphosesin
solet; quem tamen excusare necessitas potest, res diversissimas in
speciem unius corporis colligentem: oratori vero quid est necesse
surripere hanc transgressionem, et iudicem fallere qui ut ordini
rerum animum intendat etiam commonendus est? (IV.1.77f.)
Seneca did not need to tell his readers what a normal
narratio was like, but quite often he touched on the question of
whether it was ever appropriate to omit it, a question over which,
as we have seen,3 the Theodoreans disagreed with the Apollodoreans.
From time to time declaimers did suppress the narratio. For
instance, Cestius decided that it would be best not to dwell too
long on the curriculum vitae of the would-be priestess in Contr,
1.2, when speaking on her behalf:
Cestius timuit se in narrationem demittere; sic illam trans-
cucurrit: haec dixit in sacerdote futura maxime debere aestimari:
pudicitiam, innocentiam, felicitatem. quam pudica sit, miles
ostendit; quam innocens, iudex; quam felix, reditus. etiam
habemus quandam praerogationem sacerdoti ab ipso numine datam,
licet isti obiciant fuisse illam captivam, lenoni postea servisse,
causam novissime dixisse: inter tot pericula non servassent illam
dii nisi sibi. (Contr. 1.2.19 W)
Votienus Montanus, realizing that the conduct of the proconsul
Flamininus as described in Contr. IX.2 thema was morally inexcus-
able, decided that a narratio would not help his case:
Dispositio 181
quaedam controversiae sunt in quibus factum defendi potest,

excusari non potest; ex quibus est et haec. non possumus efficere
ut <reus> propter hoc non sit reprehendendus; non speramus ut
ilium iudex probet sed ut dimittat; itaque sic agere debemus tam-
quam pro facto non emendato, non scelerato tamen. itaque negabat
se pro Flaminino narraturum Montanus, sed iis quae obiciuntur
responsurum. (Contr. IX.2.18)
On the other hand, Albucius did without a narratio in Contr. 1.4
because he felt he had too good a case: Albucius non narravit, sed
hoc colore egit ab initio usque ad finem: ego me defendere debeo?
(Contr. 1.4.8). As a substitute for a full narratio one could sum
up in what was called a propositio the basic question at issue:
Albucius...patronum patri dedit nee voluit narrare. a propositione
coepit: alimenta pater a filio petit (Contr. 1.7.17). Seneca
says nothing against the occasional suppression of the narratio,
and we may suspect that he considered the rigidity expected by
people like the Apollodoreans unreasonable,1* anticipating in this
the view of Quintilian:
plerique semper narrandum putaverunt: quod falsum esse pluribus
coarguitur. sunt enim ante omnia quaedam tam breves causae ut
propositionem potius habeant quam narrationem. (Quint. IV.2.4)
However, some of the instances of omitted narratio which Seneca
mentions without adverse comment would have aroused Quintilian's
anger. To shirk the difficulties of a case in the way that
Cestius did in Contr. 1.2, and Montanus in Contr. IX.2, was to ask
for defeat:
sed quatenus etiam forte quadam pervenimus ad difficilius
narrationum genus, iam de iis loquamur in quibus res contra nos
erit: quo loco nonnulli praetereundam narrationem putaverunt. et
sane nihil est facilius nisi prorsus totam causam omnino non
agere. sed si aliqua iusta ratione huiusmodi susceperis litem,
cuius artis est malam esse causam silentio confiteri? nisi forte
tam hebes futurus est iudex ut secundum id pronuntiet quod sciet
narrare te noluisse. (Quint. IV.2.66)
Another declamatory practice which annoyed Quintilian was
that of automatically inserting an ornate digression at the end of
the narratio, regardless of relevance:
plerisque moris est, prolato rerum ordine, protinus utique in
aliquem laetum ac plausibilem locum quam maxime possint favorabil-
iter excurrere. quod quidem natum ab ostentatione declamatoria
iam in forum venit, postquam agere causas non ad utilitatem
litigatorum sed ad patronorum iactationem repertum est, ne, si
pressae illi qualis saepius desideratur narrationis gracilitati
coniuncta argumentorum pugnacitas fuerit, dilatis diutius dicendi
voluptatibus oratio refrigescat. (IV.3.If.)

Descriptive digressions on a wide variety of subjects are quite
common in the extracts from the declamations,5 and, as Bonner6 has
observed, Seneca does sometimes write about descriptiones in a way
which might suggest that they were already as much an essential,
fixed, part of the declamation as were prooemium, narratio, argu-
menta, and epilogus: 'The descriptio...became a digression, and
that it was regarded as an integral part of the speech is clear
from the fact that Seneca says (Contr. 1.4.8-9) "et in descriptione
dixit" and "et illud post descriptionem adiecit" exactly as he
uses in narratione and in argumentis,' Seneca certainly antici-
pates Quintilian in objecting when a descriptio is inserted with-
out concern for relevance:
in ea descriptione, <quam> primam in hac suasoria posui, Fuscus
Arellius Vergilii versus voluit imitari; valde autem longe petit
et paene repugnante materie, certe non desiderante, inseruit.
(Suas. 3.4)
But it does not seem to have been the case that in Seneca's time
descriptiones were invariably confined to the position between
narratio and argumenta where Quintilian's contemporaries normally
placed them. On the rare occasions when Seneca mentions the con-
text of particular descriptiones, they come in the course of the
narratio (Contr. VII.1.26, 27 - Artemo, Cestius), or in the argu-
menta (Suas. 6.8 - Latro). Arellius Fuscus' long digression
describing weather-signs alluded to in Suas. 3.4 must have formed
part of his treatment of the last item in his divisio, the argu-
ment that Agamemnon would eventually be able to sail without
having sacrificed Iphigenia, as the present calm was due to the
blind forces of nature, and the will of the gods could not be
known by mortals (Suas. 3.3 cf. ibid. 1 init.).
The argumentatio was the section in which the declaimer
developed his chosen quaestiones and tractationes. It is evident
from Seneca's discussions of divisio that at least some of the
leading declaimers took more trouble over this part of their work
than critics of the schools would admit (see e.g. Contr. IX pr. 1).
Seneca takes pains to show that Latro's method of argumentation
sadly underrated by certain contemporary critics, was in fact
greatly to be admired. He is critical of declaimers who set out
Dispositio 183
their arguments in too arid a manner (Contr. II pr. 1 ) , and those

whose sole aspiration was to split hairs with clarity; he is all
in favour of logical precision of the finest sort (subtilitas),
but only as a foundation on which greater things could be built.
Defending Latro he writes:
nihil est iniquius his, qui nusquam putant esse subtilitatem, nisi
ubi nihil est praeter subtilitatem; et in illo cum omnes oratoriae
virtutes essent, hoc fundamentum superstructis tot et tantis
molibus obruebatur, nee deerat in illo, sed non eminebat; et
nescio an maximum vitium subtilitatis sit nimis se ostendere.
(Contr. I pr. 21)
Each quaestio or tractatio would be developed in the completed
declamation by means of argumenta, which might be supported by
moralizing loci or historical exempla and perhaps ornamented by
descriptiones. Seneca makes it clear at the beginning of his work
that it is no part of his plan to discuss argumenta in detail
(Contr, I pr. 22), but he does not keep rigidly to his programme,
and quite frequently gives a sketch of the way in which a de-
claimer elaborated certain of his quaestiones, sometimes even
quoting his actual words, as for example in Contr. 1.5.4 W:
Latro primam fecit quaestionem: non posse raptorem qui ab rapta
mori iussus esset servari. si legatus, inquit, exire debet, per-
ibit; si militare debet, peribit; si ius dicere debet, peribit; si
raptam ducere debet, a'eque peribit. si is te ante rapuisset et
nuptias optasses, interposito deinde tempore antequam nuberes hanc
vitiasset, negares ilium debere mori rapta iubente? etc.
Loci are distinguished from quaestiones in Contr. VII.7.10
where Seneca describes with approval LatroTs technique of argumen-
tation:
Latro semper contrahebat et quidquid poterat tuto relinquere
praeteriebat. itaque et quaestionum numerum minuebat et locos
numquair attrahebat; illos quoque quos occupaverat non diu dicebat
sed valenter.
But there are one or two instances where a locus seems hardly dis-
tinguishable from the treatment of a minor quaestio, e.g. in Suas.
5.6:
Argentarius his duobus contentus fuit: aut non venturum Xersen aut
non esse metuendum si venerit. his solis institit... locum movit
non inutiliter: iudicare quidem se neque Xersen neque iam quemquam
Persarum ausurum in Graeciam effundi; sed eo magis trophaea ipsis
tuenda, si quis umquam illinc venturus hostis esset, ut conspectu
trophaeorum animi militum accenderentur, hostium frangerentur.
The difference we are supposed to detect perhaps is that a locus
introduces only in passing ideas outside the quaestiones of the

divisio proper. Loci is the term Seneca uses to describe totally
disorganized arguments in Contr. II.2.9.
The word locus, as we see, is not restricted in Seneca1s
usage to what we would call a commonplace, but that does not pre-
vent it from being used many times with reference to stock moral-
izing: philosophumenon locum...quomodo animi magnis calamitatibus
everterentur (Contr. 1.7.17); de fortunae varietate locum (Contr.
1.8.16); adoptionis locum (Contr. II.4.13); locum de indulgentia
liberorum in patres (Contr. VII.5.13); ilium locum: quam multa
populus Romanus in suis imperatoribus tulerit (Contr. IX.2.19);
ilium locum: quamvis sceleratos parentes velle tamen innocentes
liberos suos esse (Contr. IX.6.19); locum contra somnia et deorum
providentiam (Suas. A.4). Many other obvious commonplaces may be
seen in use in the declamatory extracts.7 The function of such
loci was to set the issues being disputed in a wider context.
Cicero (see Or. 14.45) would have approved of the principle be-
hind their use, but not of the mechanical triteness with which
declaimers usually treated them.
Historical examples had long been recognized as one of the
most important resources available to a speaker: Aristotle saw
their function as equivalent to that of induction in dialectic
(Rhet. 1.2.8), just as the function of enthymemes in rhetoric was
parallel to that of syllogisms in philosophy. So it was that some
knowledge of Greek and Roman history was essential to the de-
claimer and that a noteworthy part of Latrofs stock-in-trade was
historiarum omnium summa notitia (Contr. I pr. 18). In the de-
clamations of Seneca's day we find historical exempla introduced
either singly, to provide a precedent for a particular action, or
several at a time. In Contr. VII.1.15 the young man who has set
his parricidal brother adrift in an open boat is made by Musa to
point out that:
centurio Luculli Mithridaten non potuit occidere - dextra simul ac
mens elanguit - pro bone Iuppiter, Mithridaten quam non dubium
parricidam!
In Contr. IX.2.19 Votienus Montanus builds up a full-scale locus
by assembling instances of the way that the Romans had generally
Dispositio 185
turned a blind eye to the faults of their generals:

ipse Montanus ilium locum pulcherrime tractavit, quam multa
populus Romanus in suis imperatoribus tulerit: in Gurgite
luxuriam, in Manlio inpotentiam, cui non nocuit et filium et
victorem occidere, in Sulla crudelitatem, in Lucullo luxuriam,
in multis avaritiam.8
The declaimer did not need to be a very profound student of his-
tory in order to be able to contrive exempla. Ready-made collec-
tions of exempla existed, for instance the work by Nepos mentioned
earlier.9 Valerius Maximus1 Facta et dicta memorabilia, published
towards the end of the elder Seneca's life-time, is a good example
of the sort of compilation on which students of rhetoric could
draw;10 it provides lists of exempla, Roman and foreign, suitable
for the treatment of any commonplace from de religione observata
(I.I) to de his, qui per mendacium se in alienas familias in-
seruerunt (IX.15). There is an important reference in Contr. VII.
5.12f. to the mechanical use of memorized exempla by scholastics
in places where they were not necessarily appropriate:
gravis scholasticos morbus invasit: exempla cum didicerunt, volunt
ilia ad aliquod controversiae thema redigere. hoc quomodo ali-
quando faciendum est, cum res patitur, ita ineptissimum est
luctari cum materia et longe arcessere, sic quomodo fecit in hac
controversia Musa, qui, cum diceret pro filio locum de indulgentia
liberorum in patres, venit ad filium Croesi et ait: mutus in
periculo patris naturalia vocis impedimenta perrupit, qui plus
quam quinquennio tacuerat. quia quinquennis puer ponitur, putavit
ubicumque nominatum esset quinquennium, sententiam fieri...
The argumentatio could be expansive or minimal according to
the declaimerTs wishes. The critic who suggested that declaimers
customarily neglected argumentation because it was too troublesome
and undecorative (Contr. IX pr. 1: argumentationes, quia molestae
sunt et minimum habent floris, reliquit) was overstating the case,
but we do have, in the young Ovid, an example of a declaimer who
avoided argument when possible: molesta illi erat omnis argument-
atio (Contr. II.2.12). According to Seneca, when Ovid declaimed
Contr. II.2, sine certo ordine per locos discurrebat (Contr. II.
2.9). Haterius was another declaimer incapable of presenting his
arguments in a controlled manner; he suffered from excessive
fluency:
nee verborum illi tantum copia sed etiam rerum erat; quotiens
velles eandem rem et quamdiu velles diceret, aliis totiens
figuris, aliis tractationibus. (Contr. IV pr. 7)

If we are to believe Seneca (Contr. IV pr. 8 ) , he had to be in-
structed by a freedman when to move on to the next locus or to the
epilogus:
iubebat eum ille transire, cum aliquem locum diu dixerat: trans-
ibat; insistere iubebat eidem loco: permanebat; iubebat epilogum
dicere: dicebat.
However, the abandonment of the sequence, narratio ->
argumentation did not necessarily mean total disorganization.
There seems to have been a standard alternative arrangement,
capable of many variations, but sufficiently distinctive for the
practised listener to be able to follow what the declaimer was
doing. Seneca alludes to this arrangement when criticizing a
habit which Albucius had of developing each guaestio as if it were
a controversial
erat et illud in argumentatione vitium, quod quaestionem non tam-
quam partem controversiae sed tamquam controversiam implebat.
omnis quaestio suam propositionem habebat, suam exsecutionem, suos
excessus, suas indignationes, epilogum quoque suum. (Contr. VII
pr. 2)
Something like this format: propositio - exsecutio - excessus -
indignationes, seems to have been used by Albucius in Contr. 1.7.
(17):
...nee voluit narrare. a propositione coepit: alimenta pater a
filio petit...ad defendendum venit [= exsecutio~] .. .philosophumenon
locum introduxit, quomodo animi magnis calamitatibus everterentur
|_= excessusj . . .deinde anthypophoran sumpsit: mentiris; ille vero
iratus fuit. cogis, inquit, me dicere iratum tibi merito fuisse
etc. [= indignatio] .ll
Seneca refers to the treatment of the epilogus in two pass-
ages in book VII. The main question at issue in Contr. VII.4 was
whether the young man in the theme was right in wanting to go and
ransom his father from the pirates, rather than stay at home to
look after his mother who had gone blind from weeping. Seneca
records how Latro, unlike a certain Greek declaimer, Apollonius,
who was tactlessly vehement in his perorations, recommended that,
even if one were speaking in favour of the young man's decision to
go to his father, one should treat the mother gently in one's con-
clud ing word s:
in epilogis vehemens fuit Apollonius Graecus: <at periculosum est.>
nihil non; et domi manere et flere.12* Latro dixit pro matre
Dispositio 187
summisse et leniter agendum, non enim, inquit, vindictam sed

misericordiain quaerit, et cum eo adulescente consistit in quo ita
exigit pietatem ut impediat. aiebat itaque verbis quoque
horridioribus abstinendum quotiens talis materia incidisset; ipsam
orationem ad habitum eius quern movere volumus adfectus molliendam.
in epilogis nos de industria vocem quoque infringere et vultum
deicere et dare operam ne dissimilis orationi sit orator; con-
positionem quoque illis mitiorem convenire. (Contr. VII.4.5f. W)
Probably still quoting Latro (despite the change to direct
speech), Seneca goes on to cite a precedent for the quelling of
vehemence in the epilogus from Calvus' speech Pro Messio (ibid.
8).13
The other passage concerns the best treatment of the epilogus
in controversiae in which the defendant has both to defend himself
and to inveigh against someone else whom he alleges to be the
guilty party. Three possible courses were open to the declaimer:
to defend himself first and make his accusations second; to mix
defence with accusation; or to make his accusations first and then
defend himself. The discussion, as printed by MUller and Winter-
bottom, runs as follows, but the text seems unsatisfactory at
several points:
has controversias, quae et accusationem <habent et defensionem>,
non eodem ordine omnes declamaverunt. quidam fuerunt qui ante
defenderent quam accusarent, ex quibus Latro fuit. Fuscus
Arellius: debet, inquit, reus in epilogo desinere. optime autem
epilogum defensioni contexit; et homines magis defendenti quam
accusanti favent. ultima sit pars quae iudicem faventem possit
dimittere. quidam permiscuerunt accusationem ac defensionem, ut
comparationem duorum reorum inirent, et crimen simul reppulissent
statim transferrent; ex quibus fuit Cestius. hoc non semper ex-
pedit. utique ei qui inbecilliorem partem habet non est utile
comminus congredi; facilius latent quae non comparantur. in hac
controversia non sunt duo sed tres rei; noverca enim procuratori
coniungitur. itaque a fili parte utique aiebat prius accusandum,
quia unum deberet crimen defendere, duo obicere, et adulteri et
caedis. (Contr. VII.5.7f.)
The subject of the last sentence, as Winterbottom notes ad loc,
can hardly be Arellius Fuscus or Cestius, and a name seems to be
missing: Cestius is ruled out because he mixed accusation with
defence, Fuscus because he asserted that 'debet...reus in epilogo
desinere1, a dictum which I take to mean that the accused should
leave off defence in the epilogus (and should turn into a pros-
ecutor). 14 The missing name. I believe, is that of Latro, whose
name also seems out of place where it now stands in the second
sentence of this passage. Two considerations prompt this sugges-

tion. First, Latro is recorded in Contr. VII.4.5f. to have been
in favour of gentle pathos, rather than vindictiveness, in epi-
logis, so that it seems improbable that in this controversia,
given that alternatives were open to him, he was among those who
chose to defend themselves first and to devote the latter part of
their declamation to accusations; rather one would expect him to
leave defence till last. Secondly, the text of the sentences,
quidam fuerunt...desinere, must be viewed as suspect on internal
grounds: it is contrary to Senecafs practice, when alluding to
something which a declaimer has said, to state his name baldly and
then to follow it immediately with a quotation + inquit; Seneca
always prefaces a quotation + inquit with some introductory sen-
tence in which a different verb of saying is used, or in which the
declaimerfs approach is described in general terms, e.g.:
Contr. VII.2.10: Marcellus Aeserninus eundem colorem aliter in-
duxit. cogitabat, inquit, secum Antonius: quod Ciceroni ex-
cogitabo supplicium?
Contr. VII.2.13: solus ex declamatoribus in Ciceronem invectus
est. quid? ille, inquit, cum Antonium hostem iudicaret et omnis
Antoni milites, non intellegebat se et Popillium proscripsisse?
Contr. VII.4.6: Latro dixit pro matre summisse et leniter agendum,
non enim, inquit, vindictam, sed misericordiam quaerit...
Contr. VII.5.11 W: Vinicius...solebat hanc sententiam Saeniani
deridere... Saenianus in hac eadem controversia dixerat: nihil
puero est teste certius, utique quinquenni... haec finitio,
inquit, ridicula est...
Contr. VII.5.15: Murredius mimico genere fatuam sententiam dixit'.
facit, inquit, quod solet: pro amatore sanguini suo non parcit.
Contr. VII.6.18: Albucius et philosophatus est: dixit neminem
liberum esse, neminem servum...denique, inquit, scis et nos nuper
servos fuisse.
Read therefore, in Contr. VII.5.7:
quidam fuerunt, qui ante defenderent quam accusarent, ex quibus
[Latro] fuit Fuscus Arellius. debet, inquit, reus in epilogo
desinere.
and in § 8:
itaque a fili parte utique <Latro> aiebat prius accusandum, quia
unum deberet crimen defendere, duo obicere, et adulteri et caedis.
It has to be supposed that Latro's name was at some stage omitted,
noted in the margin as missing, and wrongly reinserted by a later
Dispositio 189
copyist.
Difficulties remain. In the passage, quidam fuerunt possit
dimittere, we are presented not with one recommendation as to the
treatment of these controversiae, but two: first we are told of
declaimers who chose to put defence before accusation, and then we
are given arguments in favour of the reverse procedure, in which
defence comes last. We have to consider where it is that Seneca
turns from considering one view and goes on to the other. The
dictum, debet...reus in epilogo desinere, belongs, in my opinion,
as in that of Seneca1s translators, to the exposition of the first
view, and means that the accused should leave off defence in the
epilogus.i:> But the opening clause of the next sentence, optime
autem epilogum defensioni contexit, expressing as it does approval
for the practice of weaving together the epilogus with the defens-
ive part of a speech, most likely favours the placing of defence
after accusation. it3 Now, the subject of this sentence can hardly
be Fuscus, if we accept that he preferred the sequence: defence ->•
accusation. More likely reus 'the defendant' (in general) is to
be regarded as its subject, in which case Bursian's suggestion
(which Kiessling printed in his edition) of the prospective future
contexet for contexit has much to recommend it: 'But best of all
he will weave his epilogus on to the defensive part of his speech.
Whatever the true interpretation is of these words, it needs to be
made clear that at least in the following two statements (et
homines...possit dimittere) Seneca does not support those who put
defence first and prosecution second. He presents arguments
against those qui ante defenderent quam accusarent, just as, later
in his discussion, he points out the disadvantages of mixing accu-
sation with defence in certain cases. It is only to the third
possible procedure, accusation followed by defence, that he gives
his full approval.
3 ELOCUTIO
Omnia ergo habebat [sc. Cassius Severus], quae ilium, ut bene

declamaret, instruerent: phrasin non vulgarem nee sordidam, sed
electam, genus dicendi non remissum aut languidum, sed ardens et
concitatum, non lentas nee vacuas explicationes, sed plus sensuum
quam verborum habentes... (Contr. Ill pr. 7)
PHRASIS ELECTA
Tastefulness in the choice of diction, in the elder Seneca1s
opinion, was one of the factors which ought to make for a good
declaimer.1 It emerges from his reports of critical discussions
about declamation that the scholastics of his time hedged them-
selves around with all manner of prohibitions about diction. It
was not only the use of barbarisms and obscenities that was for-
bidden: any word with mundane associations, any archaism or neo-
logism, was liable to be scrutinized with varying degrees of ped-
antry. Seneca tells us much of interest about these prohibitions,
while himself adopting a fairly tolerant attitude towards all but
the most monstrous breaches of the rules.
Even over the matter of barbarisms we have to distinguish
between Seneca's attitudes and those of the stricter critics of
his time. Seneca did not, of course, positively advocate barbar-
ism, but he had nothing disparaging to say about the diction of
Latro, of whom Messala, Latini...sermonis observator diligentissi-
mus, once said on hearing him declaim, sua lingua disertus est
(Contr. II.4.8). Yet he was evidently not completely deaf to the
strangeness, by metropolitan standards, of Spanish Latin: at any
rate he was reminded by some passages in Sextilius Enafs poetry of
Cicero's reference in the Pro Archia to the rich and foreign res-
onances of Corduban poetry (Suas. 6.27). As for the vexed ques-
tion of the admissibility of Grecisms to Latin, we find that in
writing literary criticism he was willing to sacrifice Latinitas
in the interests of clarity, showing as he does markedly less
Elocutio 191
concern than Cicero to find pure Latin equivalents for the techni-
cal terminology of the Greek theorists.2
Once more, in the case of words with mundane associations
(verba sordida) and colloquial language in general, we must be
careful not to confuse Senecafs own opinions with those of the
more rigid critics, The schools of his time, so he tells us,
could not bear the use of Tsordid1 words any more than they could
tolerate obscenities: quaedam enim scholae iam quasi obscena re-
fugiunt nee, si qua sordidiora sunt aut ex cotidiano usu repetita,
possunt pati (Contr. IV pr. 9 ) . The idea, apparently, was that it
was not appropriate to mention humble utensils and undignified
necessities in the highly cultivated realm of artistic prose. In
Contr, VII pr. 3 Seneca gives examples of items which he classes
as res...omnium sordidissimas; unfortunately the text of this list
is corrupt, but it seems to have included vinegar, fleabane, lan-
terns and sponges. It is in the schools of rhetoric then (though
one must also bear in mind the possible effects of prolonged study
of epic poetry with the grammatici) that we may find one source of
the tendency whereby a Silver Latin orator might choose to refer
to Hibericas herbas rather than spartum (Quint. VIII.2.2), and
which was to make Tacitus refuse to call a spade a spade and
write, amissa magna ex parte per quae egeritur humus aut exciditur
caespes (Ann. 1.65).
To judge from Seneca's wording in Contr. IV pr. 9, quaedam...
scholae iam quasi obscena refugiunt... the prohibition on mundane
words was a fairly recent development. Some declaimers, in fact,
for various reasons still rejected the restriction. They might be
just old-fashioned, or deliberate archaizers, or might not want to
be classed with the scholastics. There were even people in the
schools who admired the use of 'sordid1 diction above all else:
consectari autem solebat [sc. Bassus Iulius~J res sordidas et
inveniebat qui illas unice suspicerent (Contr. X.I.13). Bassus
spurned convention to the extent of referring to a dog guarding a
front door: non mehercules te ferrem, si canem ad ostium
alligasses (ibid).
An important point to notice is that to the declaimers of the
early empire the use of undignified words seemed old-fashioned.
On one occasion Seneca mentions archaizers qui verba antiqua et

sordida consectantur (Contr. IX.2.26). In fact it was not to a
very distant past that they were looking back. One result of the
scholastic ban on verba sordida was that it imposed restrictions
on an important means of invective which had been much used by
Cicero himself. For example, Seneca remarks, Rufus Vibius erat,
qui antiquo genere diceret; belle cess it illi sententia sordidi-
oris notae: praetor ad occidendum hominem soleas poposcit (Contr.
IX.2.25). Now, this sententia reminds one of the Verrines.
Cicero had thought it quite permissible to sneer at praetors for
cavorting round in slippers, purple cloaks and ankle-length
tunics, supported by their mulierculae (Verr. II.5.86), but by
SenecaTs time that sort of thing evidently seemed dated. The use
?
of verba sordida like soleae, so important in Cicero s techniques
of ridicule, was already being rejected by the modernists at the
very beginning of the Silver 1 age of Latin prose.
However, the taboo was not yet universally accepted, and was
still regarded as peculiar to the schools. Thus a man like
Albucius, who did not want to appear a purely scholastic speaker,
sprinkled his declamations with the unmentionable words:
splendidissimus erat; idem res dicebat omnium sordidissimas -
acetum et puleium et tdammam et philerotemt laternas et spongias:
nihil putabat esse quod dici in declamatione non posset, erat
autem ilia causa: timebat ne scholasticus videretur. dum alterum
vitium devitat, incidebat in alterum, nee videbat nimium ilium
orationis suae splendorem his admixtis sordibus non defendi sed
inquinari; et hoc aequale omnium est, ut vitia sua excusare malint
quam effugere. (Contr. VII pr. 3f. W)'
Albucius met with a mixed reception. Seneca notes how one of his
expressions, panem quern cani das, patri non das?, was admired by
some people, but derided by others (Contr. 1.7.18). This senten-
tia, or something very like it, evidently remained well known for
a long time. Quintilian tells us that it was admired in his youth
but would be quite unacceptable to the scholastics of the time
when he was writing:
id tamen in declamatoribus est notabilius, laudarique me puero
solebat fda patri panem1, et in eodem fetiam canem pascis1. res
quidem praecipue in scholis anceps et frequenter causa risus, nunc
utique cum haec exercitatio procul a veritate seiuncta laboret
incredibili verborum fastidio ac sibi magnam partem sermonis
absciderit. (VIII.3,22f.)
Elocutio 193
Quintilian himself could not accept this restrictive canon of dic-

tion. He emphasized the value of vulgar words for the heightening
of contempt and for humorous effect, and appreciated Cicero's art-
istry in this field:
an cum dicit in Pisonem Cicero Tcum tibi tota cognatio serraco ad-
vehatur', incidisse videtur <in> sordidum nomen, non eo contemptum
hominis quem destructum volebat auxisse?
he asks in VIII.3.21. He proposes that the only occasions when
words should be excluded should be when they were beneath the dig-
nity of the subject-matter, or when they were blatantly obscene:
sed ne inornata [quae] sunt quidem, nisi cum sunt infra rei de qua
loquendum est dignitatem, excepto si obscena nudis nominibus
enuntientur. (ibid. 38)
Seneca the Elder does not by any means always accept un-
critically the scholastic teaching on verba sordida. True, with
reference to the would-be priestess of Contr. 1.2 he writes,
dicendum est in puellam vehementer, non sordide nee obscene, and
he cites as 'sordid' Iulius Bassus' expressions, extra portam hanc
virginem and ostende istam aeruginosam manum, and Vibius Rufus'
observation, redolet adhuc fuliginem fornicis (Contr. 1.2.21).
Often, though, he takes a line rather different from that of the
modernist declaimers. He regards the use of colloquialisms as a
positive virtue, but one not far removed from a vice, and hence a
dangerous procedure:
idiotismos est inter oratorias virtutes res quae raro procedit;
magno enim temperamento opus est et occasione quadam. hac virtute
varie usus est [sc. Albucius] : saepe illi bene cessit, saepe
decidit. nee tamen mirum est si difficulter adprehenditur vitio
tarn vicina virtus. (Contr. VII pr. 5)
He goes on to remark how successful Gallio had been in using this
device, even as a youth, quod eo magis mirabar, he adds, guia
tenera aetas refugit omne, non tantum quod sordidum sed quod
sordido simile est (ibid. 6 ) . This remark on the tastes of the
young hardly seems a universal truth: what it does prove is how
strong a hold the scholastic demands for refined diction exercised
over the younger generation of Seneca's time.
In view of his pronouncement that idiotismos was a virtue,
however dangerously close to being a vice, it is not surprising
that we find Seneca elsewhere in his criticism showing tolerance
and sometimes approval for conversational language in declamation.

For example, he considers the effect of a vulgarism (the ethic
dative mihi) in a sententia by Vibius Rufus fnot bad1: Vibius
Rufus videbatur cotidianis verbis usus non male dixisse: ista
sacerdos quantum mihi abstulit (Contr. 1.2.23). It is with defi-
nite approval that he notes a colloquial phrase used by Romanius
Hispo: Hispo Romanius bello idiotismo usus est: dixerunt, inquit,
amici: eamus ad raptae patrem: hoc curemus, illud domi est (Contr.
II.3.21 W) . The reason why Seneca approves is probably that the
slang expression, illud domi est9 is just the sort of language one
might expect from the friends of the defendant in this particular
controversia. We may compare his remarks on Cornelius Severus'
lines,
stratique per herbam
'hie meus est' dixere 'dies'.
A pernickety grammarian called Porcellus had objected that the
Roman soldiers in Severus1 poem were guilty of a solecism when
they said 'hie meus est dies1 rather than 'hie noster'. Seneca
answers his criticism very effectively:
...in sententia optima id accusabat quod erat optimum, muta enim
ut 'noster' sit: peribit omnis versus elegantia, in quo hoc est
decentissimum, quod ex communi sermone trahitur: nam quasi
proverbii loco est: fhic dies meus est1; et, cum ad sensum
rettuleris, ne grammaticorum quidem calumnia ab omnibus magnis
ingeniis summovenda habebit locum; dixerunt enim non omnes simul
tamquam in choro manum ducente grammatico, sed singuli ex iis:
f
hic meus est dies.1 (Suas. 2.13)
Once again Seneca is appreciative of the poignant effect of an
every-day phrase. Another example of his approval of idiotismos
comes in Contr. VII.5.9 where a sententia containing the col-
loquial word rivalis is included in a list of res bellae: Brutus
Bruttedius cotidiano verbo significanter usus est: rivalem, inquit,
occidit, amicam sauciavit. He actually uses the word rivalis him-
self in a passing remark in Contr. II.6.12: Agroitas Massiliensis
longe vividiorem sententiam dixit quam ceteri Graeci declama tores,
qui in hac controversia tamquam rivales rixati sunt. It is not
always the case, then, that Seneca shared the attitudes of the
modernist scholastics towards the use of everyday language, even
though he considered that it would help one to be a good declaimer
Elocutio 195
if one used a vocabulary non vulgarem nee sordidam (Contr. Ill

pr. 7 ) .
In Contr. 1.2.23 Seneca gives his assent to the principle
that obscenity should be avoided: longe recedendum est ab omni
obscenitate et verborum et sensuum: quaedam satius est causae
detrimento tacere quam verecundiae dicere. This veto sounds un-
compromising enough, but what it really means is not immediately
clear when one considers the unseemly themes of some of the
controversiae, and the lurid descriptions which were called for
when one treated them. Certainly the declaimers do not talk the
language of the Petronian gutter, but Contr. 1.2 is, as Bornecque
puts it, fde tous les ouvrages de l'antiquite celui qui nous
fournit le plus de documents sur le recrutement et 1'organisation
interieure de certaines maisons que les Romains, sans fausse
pudeur, designent par leur nom1.^
Where did Seneca and the declaimers draw the line? It seems
that lesbianism (Contr. 1.2.23: controversiam de illo qui tribadas
deprehendit et occidit...), rape (e.g. Contr. 1.5), and the
interior organization of brothels (Contr. 1.2), were considered
subjects perfectly acceptable for treatment in the schools, so
long, at any rate, as mention of these subjects was demanded by
the theme of the controversia. But indecent insinuations un-
warranted by the theme, whether relating to perversions or
straightforward adultery, were liable to be frowned on when they
affected some character - a free-born child, or a woman claiming
to be a virgin - whose verecundia ought to be respected. Thus
Seneca designates as obscene an innuendo by Murredius in Contr.
1.2.21 about the alleged virginity of the ex-prostitute, unde
scimus an cum venientibus pro virginitate alio libidinis genere
deciderit? and Scaurus silenced, by quoting 'inepta loci1 from a
Priapeion by Ovid,1* a certain declaimer who had said in a contro-
versia in which a woman sues her husband for failing to consummate
the marriage: novimus istam maritorum abstinentiam, qui etiamsi
primam virginibus timidis remisere noctem, vicinis tamen locis
ludunt (Contr. 1.2.22). Seneca also disapproved of Vibius Gallus'
insinuation in Contr VII.5.14 that the bailiff who is rumoured to
be the lover of the materfamilias might also be the father of her
little son, and Cestius made the point that the little boy ought
to be spared such allegations:
Gallus Vibiiis inprobam dixit sententiam cum caedem describeret:
occidit, inquit, maritum, novercam laesit, puero pepercit: etiam-
nunc putabat suum. valde enim puero Cestius aiebat parcendum...
Apparently Cestius, and the Greek Hermagoras, treated the boy with
greater delicacy, but in what ways exactly is not clear, given the
brevity of the sententiae ascribed to them and the doubtful state
of the text.
Seneca does not make any comments in person on the origins of
the taste for obscenity in the schools. He quotes Scaurus as
having alleged that it originated among the Greek declaimers, qui
nihil non et permiserint sibi et impetraverint (Contr. 1.2.22),
but it is interesting, as an indication of Seneca1s attitude
towards the Greeks, that he caps the examples Scaurus gives from
the work of the Greek declaimers Hybreas and Grandaus, with a
Roman example (by Murredius) which he considers no less obscene
(ibid. 23).
During the first century A.D. all manner of attitudes towards
propriety of language were current: some people advocated total
avoidance of verba sordida, while others maintained that no word
is indecent in itself and that if the subject-matter to be dealt
with is disgusting, no paraphrasing will conceal the fact (Quint.
VIII.3.39). Quintilian ruled that words have no inherent ugli-
ness, giving fnakedf obscenities as the only exception. He ap-
peals to the criterion of fRoman modesty1 (ibid). Seneca's view
of the matter seems not unlike Quintilian1s, though he had no
illusion that verecundia was invariably characteristic of the
Romans.
The next question to consider is how far the use of archaic
expressions on the one hand and neologisms on the other was
thought acceptable in the 'choice diction' of the declaimers. We
have seen how the use of verba sordida for denigration had been
rejected by some of the scholastics as old-fashioned. This should
prepare us for the striking fact that, as early as Seneca's time,
some words used by Cicero were already regarded as archaic.
Seneca contrasts Haterius' compliance with the modernists'
Elocutio 197
prohibition on verbis calcatis et obsoletis with his taste for

archaisms: quaedam antiqua et a Cicerone dicta, a ceteris deinde
deserta dicebat (Contr. IV pr. 9 ) . Haterius1 rapid delivery could
not hide the oddity of his expression: adeo quidquid insolitum
est, etiam in turba notabile est (ibid). The clearest evidence
for Seneca's attitude towards archaisms comes in his survey of
Contr. IX.2, a declamation where the subject-matter, the execution
of a prisoner at a banquet on the orders of the praetor Flamininus,
tempted the declaimer to imitate Ciceronian techniques. Seneca
regarded Vibius Rufusf sententia about the praetor's slippers,
praetor ad occidendum hominem soleas poposcit, as successful (Jbene
cessit), but he could not share Asinius Polliofs approval for
another sententia: at nunc a praetore lege actum est ad lucernam
(Contr. IX.2.25). It was evidently too much to combine in one
sentence the ancient formula, lege actum with the sordid word,
lucerna.5 Seneca goes on to quote a witticism about archaizers
which Livy attributed to a rhetorician called Miltiades:
Livius de oratoribus qui verba antiqua et sordida consectantur et
orationis obscuritatem severitatem putant aiebat Miltiaden
rhetorem eleganter dixisse: o u TO 6e£u6v ualvovTau. (Contr. IX.
2.26)
In his comments on this dictum Seneca disapproves, as he does
elsewhere (e.g. Contr. II pr. 2 ) , of the kind of austere writing
which tends to obscurity, but he does not make any general state-
ment in agreement with Miltiades1 distaste for archaisms. The
fact seems to be that, while he would not go so far as Pollio in
expressing approval for archaic expressions, he considered that
the genus dicendi antiquum did have some positive merits. At any
rate his opinion of Scaurus, who was an exponent of such a style,
was that nihil erat illo venustius, nihil paratius (Contr. X pr.
2 ) , and he considered Labienus, who combined color orationis
antiquae with vigor novae, a great orator (ibid. 5 ) . It was,
indeed, usual among ancient critics to express distaste for what
they considered to be extreme archaism, while appreciating the
aura of gravity which the tasteful deployment of time-honoured
words might give to one's style. Thus Cicero likens the imitation
of Thucydides to a taste for wines dating from the consulships of
Opimius and Anicius, less drinkable than other mature, but rather
more recent vintages (Brut. 83.287f.)> and Quintilian commends the

artistic use of archaisms as seen in VirgilTs poetry, while dis-
approving of extremes: adspergunt illam, quae etiam in picturis
est gravissima, vetustatis inimitabilem arti auctoritatem. sed
utendum modo nee ex ultimis tenebris repetenda (VIII.3.25).
Seneca nowhere in his extant books gives his personal opinion
whether the declaimer should feel free to use newly coined words.
In Contr. VII.6.21 he reports the disapproval with which some of
his contemporaries greeted one neologism:
Hispo Romanius dixerat: maritum autem ego istum vocem raptorem
serotinum? verbum hoc quasi apud antiquos non usurpatum quibusdam
displicebat. eiusdem verbi significatione, ut extra reprehensionem
esset, usus est Gavius Sabinus cum diceret nondum esse consummatam
adversus servos publicam vindictam: etiamnunc in domo nostra
residuus raptor est.
But he does not commit himself to this disapproving attitude. The
reason why he considered Marullus to be using language licenter in
a sententia quoted in Contr. II.2.7 may have been that the word
reludere which he used was a new coinage; it is certainly a very
rare word. If so, it is indicative of a certain tolerance towards
neologisms on Seneca*s part that, though he thinks it bold to use
such a word, he considers that it expresses the meaning intended
well enough: licenter verbo usus est satis sensum exprimente...
But the best evidence for Senecafs attitude towards innovation in
prose diction lies in the way he writes himself. The vocabulary
which he uses for criticism is, as Bardon has shown, not that of
Cicero.6 It seems hardly likely, then, that he would have agreed
entirely with any scholastic ban on new formations. Once more
Quintilian (VIII.3.31ff.) illustrates a sensible attitude which
one might take if one rejected the rigidity of the schools. Quin-
tilian was familiar with a pedantic prohibition on neologisms: his
own teachers had persisted in forbidding the use of piratica,
musica and fabrica, and there had been a certain Celsus who banned
the invention of new words by the orator. Quintilian, while re-
garding Latin as less well adapted to the formation of new words
than Greek, realized that nothing would stop the evolution of
language. He gives examples of several words, now well estab-
lished, which had once been considered novelties, and argues that,
Elocutio 199
although men of his day might have lost primitive man's right to
invent new root-words, they were still free to form new compounds.
Seneca trusts in this right to reuse old roots when he employs
words like pugnacitas (Contr. 1.2.16), efficaciter (Contr. VII pr.
3) and detractus (Suas. 7.11).
However, he does seem to have conceived of the existence of a
law of diction which writers ought not to infringe. Arellius
Fuscus was one declaimer guilty of breaking it:
in descriptionibus extra legem omnibus verbis, dummodo niterent,
permissa libertas; nihil acre, nihil solidum, nihil horridum;
splendida oratio et magis lasciva quam laeta. (Contr. II pr. 1)
Fuscus' offence, as we shall see,7 lay in his introduction to
prose of whole clusters of markedly poetic words. But poets too
were liable to be upbraided for literary licentia. Seneca says of
Ovid: verbis minime licenter usus est nisi in carminibus, in
quibus non ignoravit vitia sua sed amavit (Contr. II.2.12). The
examples of this 'licence' given by Seneca were the lines semi-
bovemque virum semivirumque bovem (Ars am. 11.24) and et gelidum
Borean egelidumque Notum (Am. 11.11.10), together, almost cer-
tainly, with a third example now lost. Here, though, it is Ovid's
way of pairing contrasting words, more than the diction itself,
which seems to be being faulted: semibos was probably a neologism,
but none of the other words seems inherently objectionable. Cer-
tain declaimers had gone beyond the limits permissible even in
poetry in the use of metaphors. Seneca criticizes one such de-
claimer, Musa, in Contr. X pr. 9:
quis enim ferat hominem de siphonibus dicentem 'caelo repluunt' et
de sparsionibus 'odoratos imbres' et in cultum viridarium
'caelatas silvas' et in tpicturat8 'nemora surgentia'?
Elsewhere he recalls how Maecenas^ had suggested that a declaimer
could learn how this fault might be avoided by observing Virgil's
discretion in the use of figurative language:
corruptissimam rem omnium, quae umquam dictae sunt, ex quo homines
diserti insanire coeperunt, putabant Dorionis esse in metaphrasi
dictam Homeri, cum excaecatus Cyclops saxum in mare reiecit:
< >.10 haec quo modo ex corruptis eo perveniant, <ut> et
magna et tamen sana sint, aiebat Maecenas apud Vergilium intellegi
posse. tumidum est: opous opos ditocntaTau. Vergilius quid ait?
rapit
haud partem exiguam montis. (Aen. X.128)
ita magnitudini studet, <ut> non inprudenter discedat a fide, est
inflatum: xal x £ L P^ a 3aAXexau vnaog. Vergilius quid ait [qui] de

navibus?
credas innare revolsas
Cycladas. (Aen. VIII.691f.)
non dicit hoc fieri, sed videri. propitiis auribus accipitur,
quamvis incredibile sit, quod excusatur, antequam dicitur. (Suas.
1.12 M)
Seneca did not think that declaimers should permit themselves
extra legem omnibus verbis...libertas (Contr. II pr. 1), and prob-
ably his ideal in declamatory diction was that of Latro, whose
iI *
admirer Ovid (Contr. II.2.8), 11 is said to have used words minime
licenter in prose (ibid. 12). Nevertheless he is tolerant towards
all but the worst offences of the declaimers. At the conclusion
of his onslaught on Musa's diction he dissociates himself from
those critics who expect everything to conform ad exactam regulam,
and declares, multa donanda ingeniis puto; sed donanda vitia, non
portenta sunt (Contr. X pr. 10).
GENUS DICENDI ARDENS ET CONCITATUM

The hypothetical good declaimer of Contr. Ill pr. 7 possesses,
besides a choice vocabulary, command of a style characterized as
non remissum aut languidum sed ardens et concitatum, and as con-
taining non lentas nee vacuas explicationes,12 sed plus sensuum
quam verborum habentes. The pursuit of the fiery and agitated in
style was clearly one of the main ideals of the early Imperial
rhetoricians and writers under their influence: Quintilian was
later to describe Lucan as ardens et concitatus (X.I.90). Another
word used to distinguish the style of the new rhetoric from that
of a previous age was vigor: in Labienus1 style, we are told in
Contr. X pr. 5, both the color antiquae orationis and the vigor
novae were in evidence.
The vigor of the new style stemmed ultimately from what the
theorists called compositio,1^ that is, the order of words, the
collocation of long and short clauses, and prose-rhythm - ordo,
iunctura9 numerus (Quint. IX.4.22). Arellius Fuscus emerges as
the most notable offender against the elder Seneca's taste in
compositio. We learn that Fabianus, after his philosophical con-
version, had found Fuscus1 compositio no longer acceptable,
mollior quam ut illam tarn sanctis fortibusgue praeceptis praeparans
Elocutio 201
se animus pati posset (Contr. II pr. 1). Seneca makes it clear

that in old age he too had come to find Fuscus' nimius cultus et
fracta compositio distasteful (Suas. 2.23); he suspects nonetheless
that these very faults would appeal to the young (ibid.), and re-
calls how in his own youth, when Fuscus was the acknowledged mas-
ter of the polished style (Suas. 4.5),
...nihil fuisse...tam notum quam has explicationes Fusci, quas
nemo nostrum non alius alia inclinatione vocis velut sua quisque
modulatione cantabat. (Suas. 2.10)
Faults of compositio, castigated by ancient theorists, which may
frequently be detected in samples of Fuscus1 more flowery prose11*
include the unnatural use of hyperbaton (see Quint. IX.4.28); the
stringing together of too many short words in succession (see
Quint. IX.4.22); and the contrived repetition of the same rhythms
(see Quint. IX.4.55f.). At the other extreme from the compositio
of Fuscusr florid descriptions, so we are led to suppose, was that
of Cassius Severus: compositio aspera et quae vitaret conclusionem
(Contr. Ill pr. 1 8 ) . 1 5 Latro, in so many respects Senecafs ideal
declaimer, appears to have been in general an avoider of mollis
compositio, though in Contr. VII.4.6 he is said to have recommended
a change of tone and the use of gentler rhythms in pathetic
perorations: compositionem...illis mitiorem convenire.1^ A particu-
lar instance of poor compositio is discussed in Contr. IX.2.23f.:
illud, quod tamquam Latronis circumfertur, non esse Latronis pro
testimonio dico et Latronem a sententia inepte tumultuosa vindico;
ipse enim audivi Florum quendam, auditorem Latronis, dicentem non
apud Latronem...: refulsit inter privata pocula publicae securis
acies; inter temulentas ebriorum reliquias humanum everritur
caput. numquam Latro sic composuisset, ut, quia publicam securem
dicturus erat, diceret privata pocula, nee in tarn mollem com-
positionem sententia eius evanuisset; nee tarn incredibilis umquam
figuras concipiebat ut in ipso triclinio inter lectos et tlocot et
mensas percussum describeret. (Contr. IX.2.23f. W)
The exact implications of Seneca's assertion, nee...evanuisset,
are unclear. The reference can hardly be just to what we would
regard as the clausula of the sentence in question, everritur
caput, for in my sample of Latro!s clausulae the rhythm -^-v^
occurs in 13% of the examples considered;17 nor is the type of
word-division exemplified by Florus1 clausula unparalleled in
—^ — ^—
Latro*s usage - compare trecesimus dies in Contr. II.3.1; nor is
the use at the end of such a clausula of a word naturally con-

sisting of two short syllables - compare Contr. 1.1.1: alimenta
non daret - despite the view of theorists (see e.g. Quint. IX.4.
106) that pyrrhics at the end of sentences produced instability.
One cannot suggest that what Seneca objected to was the fact that
the whole second part of the sententia has a markedly trochaic
rhythm: inter temulentas ebriorum reliquias human(urn) everritur
caput, for it is possible to find Latronian parallels even for
extended trochaic sequences.18 So most likely his objection was
not so much to the rhythm used as to the means by which it was
attained, notably the unnatural ordering of the words humanum
everritur caput.
Characteristic of the genus dicendi ardens et concitatum was
the brilliance and frequency of its sententiae.1^ The long ex-
tract from Latro's version of Contr. II.7 (§lff.), which is gener-
ally agreed to consist of continuous prose rather than a collec-
tion of excerpts, and which will be analysed in a later chapter,20
gives a fair idea of the style of a lover of sententiae (Contr.
I pr. 22), who abhorred empty passage-work (Contr. VII.7.10).
Senecafs description in the first preface of Latrofs training
methods and memorized stock-in-trade provides the best guide he
gives us to the main component parts of declamatory prose:
solebat autem et hoc genere exercitationis uti, ut aliquo die
nihil praeter epiphonemata scriberet, aliquo die nihil praeter
enthymemata, aliquo die nihil praeter has translaticias quas
proprie sententias dicimus, quae nihil habent cum ipsa contro-
versia inplicitum, sed satis apte et alio transferuntur, tamquam
quae de fortuna, de crudelitate, de saeculo, de divitiis dicuntur;
hoc genus sententiarum supellectilem vocabat. solebat schemata
quoque per se, quaecumque controversia reciperet, scribere.
(Contr. I pr. 23)
It is interesting that Seneca and Latro recognize the grouping,
epiphonemata, enthymemata, sententiae proper, for in doing so they
anticipate the classification of the main types of sententiae
which Quintilian gives us (VIII.5.Iff.). Quintilian's discussion
of these types provides the best available commentary on the
Senecan passage.
Sententiae proper are quasi-proverbial utterances: antiquiss-
imae sunt quae proprie, quamvis omnibus idem nomen sit, sententiae
Elocutio 203
vocantur, quas Graeci gnomas appellant, declares Quintilian (VIII.

5.3). Though of general import, they can have special relevance
to the particular case at issue - ad rem - or to some particular
person - ad personam (ibid.). Some theorists attempted to define
the role of the sententia within rhetorical syllogisms, either as
part of an enthymema, or as the conclusion of an epichirema,^ but
Quintilian feels that these definitions are not always valid. He
prefers to classify gnomic sententiae as either single - simplicem
- or double - duplicem (ibid. 4 ) . He has given examples of the
1
singleT type when discussing sententiae with relevance ad rem and
ad personam (ibid. 3): nihil est tarn populare quam bonitas and
princeps qui vult omnia scire necesse habet multa ignoscere. He
groups with this type sententiae which include an explanation for
the truth they are expressing - ratione subiecta - for example:
nam in omni certamine qui opulentior est, etiam si accipit
iniuriam, tamen quia plus potest facere videtur (ibid. 4 ) . His
example of a double sententia is obsequium amicos, veritas odium
parit (ibid.). Gnomic sententiae did not have to be cast in the
form of plain statements, but could be expressed in all manner of
indirect and striking ways, per omnes...figuras (ibid. 5). Some
theorists laid down that there were ten possible ways of express-
ing them, but Quintilian could not see why there could not be
still more; he mentions as possible types sententiae expressed per
interrogationem, per comparationem, infitiationem, similitudinem,
admirationem... Such figured sententiae could be more forceful
than the straightforward sort; he cites the example of Virgil's
usque adeone mori miserum est? (Aen. XII.646), so much more in-
cisive than mors misera non est (VIII.5.6).
After a few more remarks on the sententia proper, in VIII.5.9
Quintilian proceeds to discuss the enthymema. Following Cicero
(Top. 13.55) closely, he observes that though this term could mean
'thought1 (omne quod mente concepimus) in general, it had acquired
the specialized meaning of sententia ex contrariis because, so he
suggests, this was the greatest of all the varieties of 'thought1,
ut Homerus 'poeta', 'urbs' Roma. Though the enthymema was a func-
tional part of rhetorical argument and had been treated as such by
Quintilian in an earlier book (V.14.Iff.), it is included by him
under items of style as well because it could be used merely ad

ornatum, long after one's hearers needed to be convinced of the
point being made (VIII.5.10). It could be used, like the epi-
phonema which he is about to discuss, at the end of a paragraph as
a sort of parting shot. The only example he gives of an enthymema
here is a Ciceronian one: quorum igitur inpunitas, Caesar, tuae
clementiae laus est, eorum te ipsorum ad crudelitatem acuet
oratio? (Lig. 10). Closer to the style of declamatory sententiae
than this or any of the enthymemes ex pugnantibus cited by Quinti-
lian in V.14.2ff. are the examples from drama which Cicero gives
in Topica 13.55:
hoc metuere, alterum in metu non ponere!
earn quam nihil accusas damnas, bene quam meritam esse autumas
male merere? id quod scis prodest nihil; id quod nescis obest?
There is a fine Latronian example of a declamatory enthymeme in
Contr. 1.1.2: imitationem alienae culpae innocentiam vocas? Simi-
lar in type and perhaps also to be regarded as enthymemata are
such sententiae as eo iam perductus erat, ut omnem spem ultimorum
alimentorum in ea domo poneret in qua habebat <et> abdicatum et
inimicum (ibid. 1 ) , and ergo fame morientem videbo per cuius
cineres iuraturus sum? (ibid. 3 ) .
In giving enthymema2-^ the restricted technical sense of
sententia ex contrariis Quintilian in VIII.5.9 is following a
tradition related to Aristotle's theory of yvwyoAoyua rather than
to his theory of argumentation. In his discussion of rhetorical
argument Aristotle had used evûynyot as a term of very wide appli-
cation: an enthymeme was nothing less than the rhetorical equiv-
alent of the dialectician's syllogism:
TO 6e TLV&V OVT03V CTepov T I 6tot xauTa auygauveuv itapa xauxa T(J)
e£vaL, n xadoAou rj ws tub TO TCOAU, exet yev auAAoyuayos
6e evûynya
(Rhet. 1.2.9: 'When certain things being posited, something dif-
ferent results by reason of them, from their being true, either
universally or in most cases, this is called in dialectic a syllo-
gism, in rhetoric an enthymeme.') Usually in an evduynya the
orator, out of consideration for his hearers, leaves out some of
the steps of reasoning which a dialectician would give in full in
a auAAoytayos; usually, though not always, because the orator's
Elocutio 205
subject-matter is human behaviour with all its vagaries, the prem-

isses dealt with in an evûynyot tend to be probabilities rather
than universal truths (ibid. 12ff.). But, in theory at least,
there could be rhetorical enthymemes corresponding to all the
syllogisms of dialectic, and they could include proofs from con-
sequents as well as proofs from contraries. Quintilian comes
closer to this wide conception of an enthymeme in his fifth book,
where he is dealing with argumentation (V.10.Iff.), than in his
passage about sententiae in the eighth.
In his discussion of yvwyoXoyta in Rhet. II.21.Iff. Arist-

otle considers a variety of types of sententious utterance. Some
he looks upon as constituting parts of enthymemes, but yvojyat, of
one type, which he declares to be the most highly regarded, are
described as ev-duynyaxuxal:
au 6' evûyriyaxuxal yev, oux evduynyaxos 6e yepos* aunep xau
yaXuax' eu6oKuyo\3auv.2^ eoal 6' a^xau ev oaaus eycpalvexau xou
Xeyoyevou x6 auxtov,21* olov ev x$
ddavaxov opynv yn (puXaaae ^VTITOS iSv
xo yev yap (pdvau yn 6etv del cpuXaxxeuv xnv opynv yvwyn, T O 6e
Tipoaxelyevov f§vr|T6v ovxa 1 xo 6ud T L Xeyeu. 6'yooov 6e xau xo
T0V
dvaxd XPH ^vaxov, oux d^dvaxa x6v ^vaxov cppovetv. (§6)
Aristotle does not actually state that enthymematic yv&yau have to
be the equivalent of syllogisms from contraries, but both the
examples he gives could be regarded as such.
The author of the Rhetorica ad Alexandrum takes us a step
nearer the Roman equation of enthymemata with sententiae ex con-
trariis when he stipulates that evduynyaxa (the fourth of his
types of direct proof) have to do with contraries and are best
expressed in as few words as possible:
evduynyotxa 6e eaxtv ou yovov xd x$ Aoycp xal xri Ttpdêu evavxuouyeva,
OLXXOL xau X O L S d'AAous dâau. Aifyn 6e iioXXa yexucov, ws ev xw
eêxaaxuKtj) e*6eu eupriTau, xal axoTtwv, eu nr\ 6 X6yos feauxco]
evavxuouxau n xd neftpayyeva X O L S 6uwauoLS r| TCO voycp n xaj auy-
cpepovxu n xw xaX^ rj xaj 6uvaxlp n xaj pa6C(p n T $ C U K O X L n xaj ndeu xou
Xeyovxos n T W eêu xwv Tipayydxwv. xd yev o5v xouauxa xwv evûyn-
ydxwv naxd xaiv evavxuwv exXrj^Teov. xd 6' evavxua xouxous uiiep
nywv auxwv 6eu Xeyetv duocpauvovxas Tag updêus xdg nyexepas xotu
xous Xoyous evavxuouyevoug X O L S d6uxoLS xal xous dvoyous xal xous
dauycpopotg xal X O L S X W V dvdpwuwv xwv novripwv ndeau, xal auXXn$6nv
xous yox^npots voyucoyevous eSvau. 6eu 6e xouxwv exaaxa auvdyeuv
J)S eus Bpax^Taxa xal cppdêuv oxu yaXuaxa ev oXuyots xots ovoyaaL.
xd yev o^v evduynyaxa xouxov xov xpoirov TioXXa nounaoyev, xal ouxws
auxous d'puaxa xP^croyeSa. (Rhet. Alex. 1430a)
It is to this sort of theory, not to the elaborate precepts about

argument in Aristotlefs Rhetoric and logical treatises, that
Cicero is referring when in Topica 13.55 he notes that what the
rhetoricians call an evûyriya is equivalent to the dialecticians1
third mode of conclusio:
cum autem aliqua coniuncta negaris et ex eis unum aut plura
sumpseris, ut quod relinquitur tollendum sit, is tertius appellatur
conclusionis modus, ex hoc ilia rhetorum ex contrariis conclusa,
quae ipsi evduynyaxa appellant...25
The author of the Rhetorica ad Herennium uses the Latin term con-
trarium in the same sense (IV.18.25).26
It is almost certain that when Seneca the Elder and Latro
referred to enthymemata they meant sententiae ex contrariis,
though just conceivably they were among those, with whom Quinti-
lian disagrees (VIII.5.4), who classed all sententiae cum ratione
as enthymemes (see V.10.2). At any rate, it is totally unlikely
that they would have used the term to refer to rhetorical syllo-
gisms in general, as Aristotle does in Rhet. 1.2.9. For though
Quintilian knew of definitions of enthymema much wider than
sententia ex contrariis, he only discussed them in the part of his
work concerned with argumentatio (V.10.Iff.). When he turned to
consider enthymemata as items of elocutio it was the definition
sententiae ex contrariis that alone sprang to his mind (VIII.5.9).
And as Seneca, when writing about LatroTs practice methods in
Contr, I pr. 23, mentions enthymemata in conjunction with epi-
phonemata and sententiae proper, it is clear that he is thinking
of them as stylistic rather than argumentative devices.
Quintilian (VIII.5.11) defines epiphonema as rei narratae vel
probatae summa adclamatio, and as examples cites Virgil?s tantae
molis erat Romanam condere gentem (Aen. 1.33), and Cicerofs facere
enim probus adulescens periculose quam perpeti turpiter maluit
(Mil. 9 ) . 2 7 This is a type of sententia of which Silver Latin
authors were extremely fond. Quintilian considered that it had
been used to excess:
sed nunc aliud volunt, ut omnis locus, omnis sensus in fine
sermonis feriat aurem. turpe autem ac prope nefas ducunt respirare
ullo loco qui adclamationem non petierit. inde minuti corruptique
sensiculi et extra rem petiti: neque enim possunt tarn multae bonae
sententiae esse quam necesse est multae sint clausulae. (VIII.5.
Elocutio 207
13f.)
Sententiae proper, enthymemata and epiphonemata, the types of
sententiae practised regularly by Latro, are the types which
Quintilian discusses first and in most detail, but he observes
that certain additional new varieties had come to be favoured by
orators and rhetoricians in recent years. We must therefore not
necessarily expect to be able to classify all the sententiae used
by Latro and his colleagues under the three main headings. Some
of the new varieties, including sententiae ex inopinato, alio
relatae, aliunde petitae, sententiae playing on simple geminatio;
and sententiae...a verbo (these last usually in the worst of
taste), are listed and illustrated by Quintilian in VIII.5.15-25.
Quintilianfs sententiae...a verbo are of the type which the elder
Seneca and his contemporaries sometimes called sententiae Publi-
lianae, after Publilius Syrus, the writer of mimes - which was
unfair to Publilius, according to Cassius Severus:
Cassius Severus, summus Publili amator, aiebat non illius hoc
vitium esse, sed eorum, qui ilium ex parte qua transire deberent
imitarentur, <non imitarentur> quae apud eum melius essent dicta
quam apud quemquam comicutn tragicumque aut Romanum aut Graecum...
(Contr. VII.3.8)
We find examples of the so-called Publilian sententiae in Contr.
VII.3.8:
Murredius...Publilianam sententiam dedit: abdicationes, inquit,
suas veneno diluit; et iterum: mortem, inquit, meam effudit;
again in VII.4.8:
in hac controversia Publilianam sententiam dedit Festus quidam
rhetor...fuit autem Festi sentential 'captus est, inquit, pater.1
si te capti movent, et haec capta est. et quasi non intellexisse-
mus, ait: nescitis dici 'captos luminibus1?
(the blind mother of the theme refused to let her son go to ransom
her husband from the pirates); and also in VII.2.14, where the
text unfortunately is corrupt.
Besides devoting days to the practice of each of the main
types of sententiae, Latro, we are told, would write out in full
any schemata required in his declamations (Contr. I pr. 23). We
may look to Quintilian1s ninth book (IX.1.1-3.102) for a detailed
- though confessedly not exhaustive28 - survey of the figurae or
schemata recognized in his day. Quintilian's definition of a
figura is: conformatio quaedam orationis remota a communi et

primum se offerente ratione (IX.1.4). LatroTs opinion of the
proper use of figures is given in Contr. I pr. 24:
schema negabat decoris causa inventum, sed subsidii, ut quod
[palainj aures offensurum esset si palam diceretur, id oblique et
furtim subreperet. summam quidem esse dementiam detorquere
orationem cui esse rectam liceret.
With this view of the matter Quintilian was in full agreement:
figures were an aid to persuasiveness, not merely decorative,
nam etsi minime videtur pertinere ad probationem qua figura quid-
que dicatur, facit tamen credibilia quae dicimus, et in animos
iudicum qua non observatur inrepit. (IX.1.19)
Like Latro, he considered that one should not use figures all the
time:
et oratio habet rectam quandam velut faciem, quae ut stupere in-
mobili rigore non debebit, ita saepius in ea quam natura dedit
specie continenda est. (IX.3.101)
Seneca mentions surprisingly few figurae by name, considering
the large number of types known to Quintilian.29 It is usually
only the grander types of figure which he finds worthy of note,
and more often those which theorists would have classed as figures
of thought than those classed as figures of speech, on which dis-
tinction see Quint. IX.1.17:
inter plurimos enim, quod sciam, consensum est duas eius esse
partes, 6uavoLas, id est mentis vel sensus vel sententiarum (nam
his omnibus modis dictum est), et Aeêws, id est verborum vel
dictionis vel elocutionis vel sermonis vel orationis: nam et
variatur et nihil refert.
The only figures of the latter class which he mentions by name are
the tricolum30 (Contr. II.4.12, IX.2.27) and the tetracolon (Contr.
IX.2.27). These come to his attention because they were being
seriously misused by inferior speakers:
hanc controversiam cum declamaret, Maximus dixit [quasi] tricolum
tale qualia sunt quae basilicani sectantur. dicebat autem a parte
patris: omnes aliquid ad vos inbecilli, alter alterius onera, de-
tulimus: accusatur pater in ultimis annis, nepos in primis tabdica-
tur millust, haec autem subinde refero quod aeque vitandarum
rerum exempla ponenda sunt quam sequendarum. (Contr. II.4.12 W)
(Muller reconstructs the end of the tricolon as follows: nepos in
primis <adoptatur, in mediis> abdicatur filius.)
sed ne hoc genus furoris protegere videar, in Flaminino tumidissime
dixit Murredius: praetorem nostrum in ilia ferali cena saginatum
meretricis sinu excitavit ictus securis. et illud tetracolon:
Elocutio 209
serviebat forum cubiculo, praetor meretrici, career convivio, dies

nocti. novissima pars sine sensu dicta est, ut impleretur numerus;
quern enim sensum habet: 1serviebat dies noctif? hanc ideo sen-
tentiam rettuli quia et in tricolis et in omnibus huius generis
sententiis curamus ut numerus constet, non curamus an sensus.
omnia autem genera corruptarum quoque sententiarum de industria
pono, quia facilius et quid imitandum et quid vitandum sit docemur
exemplo. (Contr. IX.2.27)
Figures of thought to which Seneca refers are the testamenti
figura:
Contr. IX.3.14: Gallio autem elegantissime dixit a parte patris
cum ultima per testamenti figuram tractaret: quandoque ego mortuus
ero, tune mihi heres sit: vis interrogem, uter?
(cf. Pollio's parody of a will, Quint. IX.2.34f.);
the iusiurandum:
Contr. VII pr. 6f. (on Albucius): haec ilium sollicitudo fugavit a
foro et tantum unius figurae crudelis eventus. nam in quodam
iudicio centumvirali, cum diceretur iurisiurandi condicio ali-
quando delata ab adversario, induxit eiusmodi figuram, qua illi
omnia crimina regereret. placet, inquit, tibi rem iure iurando
transigi? iura, sed ego iusiurandum mandabo: iura per patris
cineres, qui inconditi sunt, iura per patris memoriam; et executus
est locum, quo perfecto surrexit L. Arruntius ex diverso et ait:
accipimus condicionem; iurabit. clamabat Albucius: non detuli
condicionem; schema dixi. Arruntius instabat. centumviri rebus
iam ultimis properabant. Albucius clamabat: ista ratione schemata
de rerum natura tolluntur. Arruntius aiebat: tollantur; poterimus
sine illis vivere.
(cf. Suas. 2.14; 7.14; Quint. IX.2.62f., 95);
ironia:
Contr. 1.7.13: et cum diu pressisset [sc. Silo Pompeius] ilium
tyranni patrem esse, adiecit: aude postulare ut illud tibi prosit,
quod tyrannicidae pater es. Blandus hunc sensum, cum postero die
declamaret, in ironiam vertit, et, cum obiecisset quod tyranni
pater esset, adiecit: nolite ilium aversari; habet quod adponat:
et adulteri pater est.
(cf. Quint. IX.2.44);
anthypophora or contradictio:
Contr. 1.7.17: deinde anthypophoran sumpsit [sc. Albucius]:
mentiris; ille vero iratus fuit. cogis, inquit, me dicere iratum
tibi merito fuisse;
Suas. 2.18: [sc. Statorius Victor] sumpsit contradictionem: 'at1
inquit 'trecenti sumus'; et ita respondit: trecenti, sed viri, sed
armati, sed Lacones, sed ad Thermopylas; numquam vidi plures
trecentos.
(cf. Quint. IX.2.106; 3.87).
Finally, though Seneca never uses the term praesumptio (on which
see Quint. IX.2.16ff.), he mentions two sub-species of this im-

portant figure, praeparatio and confessio:
Contr. VII pr. 3: itaque moderatio est adhibenda, ut sit ilia
praeparatio, non confessio.
As we have seen,31 Seneca also mentions a class of figurae dis-
tinct from any recognized by Quintilian, figures of divisio,
whereby the whole of a declaimerTs argumentation was arranged in
an abnormal fashion.
It has been observed earlier32 that the term locus is not
restricted in the elder Seneca's criticism to the meaning 'mor-
alizing commonplace1, and some types of loci may suitably be con-
sidered under the heading of elocutio. The development of an
elaborate figura, Albucius1 ill-fated iusiurandum, is classed as
a locus in Contr. VII pr. 7. In Contr. X.5.23 we read how all the
declaimers taunted Parrhasius for torturing his model for a paint-
ing of Prometheus, and once more Seneca uses the term locus: ilium
locum omnes temptaverunt: quid, si volueris bellum pingere? quid
si incendium? quid, si parricidium? The word is also used with
reference to passages of description33 in Contr. II.1.26:
solebat autem [sc. Vibius Gallus] sic ad locos pervenire, ut
amorem descripturus paene cantantis modo diceret: 'amorem de-
scribere volo' sic tamquam 'bacchari volo.' deinde describebat et
<ut> totiens coepturus repetebat: 'amorem describere volo.'
As to the proper length and style for loci, Seneca probably shared
the views of Latro who, on the rare occasions when he introduced
such passages, spoke non diu...sed valenter (Contr. VII.7.10).
Numerous descriptive passages are to be found in Seneca's
declamatory extracts, for example: Contr. Il.l.lOff. - battle-field,
extravagant building and resultant evils, beauties of simple life
(Fabianus); II.5.6 - torture (Fabianus); VII.1.4, 26 - storm at
sea (Haterius, Artemo); IX.2.4 - drunken orgy (Iulius Bassus);
X.4.2 - mangled beggars (Cassius Severus); Suas. 1.1, 2, 4, 11-13 -
Oceanus; 2.1 - treacherous straits (Arellius Fuscus); 3.1 -
variation of weather, phases of the moon (Fuscus); 5.1 - the
broken spirit of Xerxes (Fuscus). Descriptiones could be extended
set-pieces, like most of these, but it must not be imagined that
Seneca always has such long digressions in mind when he uses the
terms, descriptio, describere: in Contr. VII.5.10 there is a
Elocutio 211
reference to a description of the step-mother's wound, a subject

on which even Musa could hardly have expatiated for very long; a
descriptio mentioned in Contr. 1.4.8 appears merely to have been
an extra-vivid piece of narrative:
et in descriptione dixit: cum me vocavit pater, fhocf inquam
'putavit supplicium futurum morte gravius, si adulteram filio
ostenderit.' et illud dixit: exierunt adulteri inter patrem
debilem et filium stupentem.
Other critics would have classed such descriptions as this and the
one mentioned in the analysis of Latro's divisio for Suas. 6 (§8)
in which Cicero was invited to envisage acerbitatem servitutis
futurae, as examples of the figure sub oculos suJbiectio/uiroT^TCcoaus
(see e.g. Quint. IX.2.4Of.).
The term explicatio, which Seneca uses in the stylistic re-
commendations of Contr. Ill pr. 7, seems to serve in his usage as a
synonym for descriptio. At least, the one thing which the pass-
ages with reference to which he uses this term have in common is
an element of description. That an explicatio does not have to be
long is proved by a passage in Contr. VII.1.27:
soleo dicere vobis Cestium Latinorum verborum inopia <ut> hominem
Graecum laborasse, sensibus abundasse; itaque, quotiens latius
aliquid describere ausus est, totiens substitit, utique cum se ad
imitationem magni alicuius ingeni derexerat, sicut in hac contro-
versia fecit, nam in narratione, cum fratrem traditum sibi de-
scriberet, placuit sibi in hac explicatione una et infelici: nox
erat concubia, et omnia, iudices, canentia <sub> sideribus muta
erant.
Consequently no translation of explicatio suggestive of prolonged
unfolding seems wholly appropriate. Now, particular explicationes
of Arellius Fuscus are referred to in Suas. 2.10, 2.23 and 4.5:
huius suasoriae feci mentionem, non quia in ea subtilitatis erat
aliquid, quod vos excitare posset, sed ut sciretis, quam nitide
Fuscus dixisset vel quam licenter; ipse sententiam <non> feram;
vestri arbitrii erit, utrum explicationes eius luxuriosas putetis
an vegetas. Pollio Asinius aiebat hoc non esse suadere, sed
ludere. recolo nihil fuisse me iuvene tarn noturn quam has explica-
tiones Fusci, quas nemo nostrum non alius alia inclinatione vocis
velut sua quisque modulatione cantabat. at quia semel in mentionem
incidi Fusci, ex omnibus suasoriis celebres descriptiunculas sub-
texam, etiamsi nihil occurrerit, quod quisquam alius nisi suasor
dilexerit. (Suas. 2.10)
sed ne vos diutius infatuem, quia dixeram me Fusci Arelli explica-
tiones subiecturum, hie finem suasoriae faciam. quarum nimius cul-
tus et fracta compositio poterit vos offendere, cum ad meam aetatem
veneritis; interim <non> dubito, quin nunc vos ipsa, quae offen-
sura sunt, vitia delectent. (Suas. 2.23)
et quia soletis mihi molesti esse de Fusco, quid fuerit, quare
nemo videretur dixisse cultius, ingeram vobis Fuscinas explica-
tiones. (Suas. 4.5)
The passages which are referred to here as presenting explica-
tiones are Suas. 2.If., 3.1 and 5.Iff.; all of which, though not
wholly descriptive, contain descriptions.^ That it is quite
possible that when the elder Seneca uses explicatio he is referring
to the descriptive elements within the extracts, rather than to
the extracts themselves, is evident from the way in which, to vary
his terminology, he uses descriptio in a parallel passage, when
referring forward to another, not wholly descriptive, Fuscine
extract (Suas. 4.Iff.):
iam, <si> vultis, ad Fuscum revertar et descriptionibus eius vos
statim satiabo, ac potissimum eis quas in simili huius tractatione
posuit, cum diceret omnino non concessam futurorum scientiam.
(Suas. 3.7 W)
If this reasoning is correct, the slow-moving, empty explicationes
which the imagined good declaimer of Contr. Ill pr. 7 avoids, are
verbose descriptive passages. It was not only in descriptions,
however, that Seneca disliked verbosity.
Seneca1s preference for fiery agitation did not prevent him
completely from appreciating richness and fluency of language.
Rather he deplored the inability of those declaimers whose style
was notable for copia and facultas to exercise control over their
material. Thus he does not object to Albucius1 copious use of
language, which left one with no complaint about the deficiencies
of Latin: non posses de inopia sermonis Latini gueri, cum ilium
audires^^ (Contr. VII pr. 3 ) ; but he does complain about the over-
discursive philosophizing to which he was prone in his private
declamations (ibid. 1) and his habit of developing each guaestio
as if it were a controversia in itself (ibid. 2 ) . Similarly,
after acknowledging Haterius1 remarkable services to the Latin
language: solus omnium Romanorum, guos modo ipse cognovi, in
Latinam linguam transtulit Graecam facultatem (Contr. IV pr. 7 ) ,
he finds much to censure in this declaimer, who warranted compari-
son with a muddy torrent (ibid. II), 3 6 being both excessively
rapid in his delivery (ibid. 7) and totally unable to gauge when
Elocutio 213
he had said enough about any topic (ibid. 8 ) . A type of repeti-

tiveness particularly associated with the declaimer Votienus
Montanus37 is castigated as a vitium in Contr. IX.5.17.
Seneca disapproved of extreme sparseness and brevity as well
as excesses of loquaciousness. Latro as a schoolboy had so taken
exception to the exilitas of his teacher Marullus* style that he
went as far as to interpolate sententiae into his argumentation
while he was declaiming to the class (Contr. I pr. 22). Seneca,
likewise a pupil of Marullus, calls this master hominem satis
aridum (ibid.), and he can be scornful of aridi declamatores in
general (Contr. II.1.24). We have seen how he observed that the
inopia of Cestius1 Latin prevented him from imitating poetic
models successfully (Contr. VII.1.27). He is critical too of the
obscureness which results from extreme brevity, a fault which
marred the declamations of Fabianus, and was still occasionally
to be found in his philosophical works, in which quaedam tarn
subito desinunt, ut non brevia sint, sed abrupta (Contr. II pr.
2).
Inaequalitas is mentioned as a fault of style in Contr. II
pr. 1. It emerges from Seneca's criticism elsewhere that he did
realize that the declaimer needed to vary his tone to some extent;
what he was averse to was the juxtaposition of totally opposed
styles within a declamation. When considering Seneca's views on
colores we have seen how he occasionally displays a concern that
the declaimer should not adopt a tone inappropriate to the charac-
ters involved in the case at issue;38 he ponders whether some
words quoted in Contr. 1.8.16 were not too high-flown for pressa
et civilis oratio, and, like Latro (Contr. VII.4.6), is in favour
of a certain restraint in the use of language when women are con-
cerned (Contr. 1.2.21). We have noted also 39 how Latro recommended
a change of compositio when one's epilogus was given over to a
plea for mercy (Contr. VII.4.6). But Seneca seems to have be-
lieved quite strongly that a man should basically have one style
and keep to it. He is critical of the way in which Albucius, in-
consistent through lack of self-confidence, ranged through all the
genera dicendi known to him, and ended up a worse declaimer than
he had originally been:
nulla erat fiducia ingenii sui et ideo adsidua mutatio; itaque dum
genera dicendi transfert et modo exilis esse volt nudisque rebus
haerere, modo horridus et valens potius quam cultus, modo brevis
et concinnus, modo nimis se attollit, modo nimis se deprimit,*
ingenio suo inlusit et longe deterius senex dixit quam iuvenis
dixerat. (Contr. VII pr. 5)
In particular he finds fault with the uncomfortable combination in
Albucius' declamations of the supervacuus strepitus characteristic
of the schools, with the verba sordida of old-fashioned forensic
oratory. It is not clear, though, that Albucius actually juxta-
posed passages in completely different styles within the same dec-
lamation. This was a practice particularly characteristic of
Fuscus, whose habit of delivering his principia, argumenta and
narrationes in a dry style, but of indulging in all manner of syn-
tactical complication and audacity of diction in his descriptiones,
is so vividly recalled in Contr. II pr. l.4^
The Fuscine manner was perhaps the most important rival to
the genus dicendi ardens et concitatum. It seems to have sup-
planted the Latronian ideal in the favour of the scholastics in
the later first century A.D., and even to have come to influence
oratory outside the schools. At least, when Quintilian refers in
IV.3.Iff.141 to the practice of separating the plain style of the
narratio from the pugnacity of the argumentation by means of a
flowery digression, he declares that this custom was followed by
most people (plerisque), and, though it was in origin a product of
declamatory ostentation, it had now entered the forum. Seneca
realized the importance of Fuscus as a declaimer, awarding him,
together with Albucius, a place in his primum tetradeum. But
predictably, it is to Latro, along with Gallio, another master of
consistent sententiousness42 - of whom, unfortunately, we have no
full critical appraisal in Seneca's extant books - that he assigns
the highest honours: hi quotiens conflixissent, penes Latronem
gloria fuisset, penes Gallionem palma (Contr. X pr. 13).
VITIOSA: SOME GENERAL TERMS DENOTING BAD TASTE

Frequently in his surveys of the declamations Seneca pauses to
give his verdict whether or not a particular sententia is in good
taste. In criticizing the bad he often uses the words corruptus,
Elocutio 215
corrupte, corrumpere, and occasionally also cacozelia, cacozelos.

It seems worth demonstrating to what extent his judgements fall
into line with the criteria for distinguishing corrupti generis
oratio set out by his son in Ep. mor. 114.1, and the account of
cacozelon given by Quintilian in VIII.3.56-8.
In Ep. mor. 114.1 the younger Seneca undertakes to answer a
query of Lucilius as to the cause of the popularity at certain
times of various types of corrupt style:
Quare quibusdam temporibus provenerit corrupti generis oratio
quaeris et quomodo in quaedam vitia inclinatio ingeniorum facta
sit, ut aliquando inflata explicatio vigeret, aliquando infracta
et in morem cantici ducta; quare alias sensus audaces et fidem
egressi placuerint, alias abruptae sententiae et suspiciosae, in
quibus plus intellegendum esset quam audiendum; quare aliqua aetas
fuerit quae translationis iure uteretur inverecunde.
Describing the vice cacozelon in VIII.3.56-8 Quintilian appears to
equate it with corrupta oratioi both comprise a wide range of
stylistic faults, but do not include rerum vitia.
cacozelon, <id> est mala adfectatio, per omne dicendi genus
peccat; nam et tumida et pusilla et praedulcia et abundantia et
arcessita et exultantia sub idem nomen cadunt. denique cacozelon
vocatur quidquid est ultra virtutem, quotiens ingenium iudicio
caret et specie boni fallitur, omnium in eloquentia vitiorum
pessimum: nam cetera parum vitantur, hoc petitur. est autem totum
in elocutione. nam rerum vitia sunt stultum commune contrarium
supervacuum: corrupta oratio in verbis maxime inpropriis, redund-
antibus, compressione obscura, compositione fracta, vocum similium
aut ambiguarum puerili captatione consistit. est autem omne caco-
zelon utique falsum, etiam si non omne falsum cacozelon: tett
dicitur aliter quam se natura habet et quam oportet et quam sat
est. totidem autem generibus corrumpitur oratio quot ornatur.
But the faults of content, stultum, commune, contrarium and super-
vacuum,1*3 a r e not the only vitia Quintilian distinguishes from
cacozelon, which is only one item in a long list he gives of
possible departures from appropriate ornatus, the others being:
cacemphaton/deformitas (VIII.3.44ff.); TaneLvuois/humilitas (48f.);
eAAeu<|>us - cum sermoni deest aliguid, guo minus plenus sit (50f.);
6yoeu6eua - guae nulla varietatis gratia levat taedium (52);
macrologia and pleonasmos (53f.); periergia (55); male dispositum/
avouxov6unTOV (59); male figuratum/aoxTwiOLTLOTOv (ibid.); male
conlocatum/-HCLHoov\)%eTO\) (ibid.); Zap6uayos (ibid.) and the related
vice in Latin of mixing different types of diction, sublimia
humilibus, vetera novis, poetica vulgaribus (60). Just why
Quintilian should have distinguished some of these faults from

cacozelon is not self-evident: it seems inconsistent to distinguish
macrologia, pleonasmos and periergia from cacozelon and then to
list abundantia among the species of this last-named fault; to
mention compositio fracta as a form of corrupta oratio, and then
to include xaxoa6v§eTOV as a distinct item in the continuation of
his list of inornata ,1+If It looks very much as though Quintilian
is presenting in this section on departures from ornatus a con-
flation of two traditions, in one of which cacozelon was used to
cover all aspects of stylistic bad taste, quidquid est ultra
virtutem, and another in which it was given some more restricted
meaning, and so could be named as one item in a long list of
faults of style.45
Seneca and Quintilian agree in classing as species of
corrupta oratio or cacozelon the following faults: tumidity
(Seneca: inflata; Quintilian: tumida, abundantia, verbis...redund-
antibus); the abuse of metaphors (Seneca: translationis iure
uteretur inverecunde; Quintilian: verbis...inpropriis); effeminacy
in the rhythmic arrangement of words (Seneca: explicatio...in-
fracta et in morem cantici ducta; Quintilian: compositione fracta);
the expression of far-fetched ideas (Seneca: sensus audaces et
fidem egressi; Quintilian: arcessita); obscurity arising from ex-
cessive concision (Seneca: abruptae sententiae et suspiciosae, in
quibus plus intellegendum esset quam audiendum; Quintilian: com-
pressione obscura). Quintilian mentions some other species in
addition.
Similarly in the elder Senecafs criticism corruptus and its
cognates are used with reference to a wide range of faults.
Corrupte is used as the opposite of bene in Contr. IX.2.21, ex
altera parte multa bene dicta sunt, multa corrupte..., and of
pulchre in Contr. X.4.18, multa ab illis pulchre dicta sunt...
multa corrupte. Cassius Severus uses corruptius as virtually
equivalent to peius in Contr. Ill pr. 15: non illi peius dicunt,
sed hi corruptius iudicant. In Contr. IX.2.27 Seneca affirms his
intention to show quid vitandum sit by recording omnia...genera
corruptarum...sententiarum. The use of the verb corrumpere may
be illustrated by Contr. IX.5.17: habet hoc Montanus vitium:
Elocutio 217
sententias suas repetendo corrumpit; dum non est contentus unam

rem semel bene dicere, efficit ne bene dixerit.
The term sanus is sometimes used as the opposite of corruptus,
and insanire to denote the activity of writing in a fcorrupt'
manner: et volebam vos experiri non adiciendo iudicium meum nee
separando a corruptis sana; potuisset [et] enim fieri ut vos magis
ilia laudaretis quae insaniunt (Suas, 1.16); corruptissimam rem
omnium quae umquam dictae sunt ex quo homines diserti insanire
coeperunt (ibid. 12). In Contr. IX.2.27f. the synonymous terms
insania and furor are used with reference to corrupt style; this
passage also serves to illustrate the close connection in the
elder Seneca's mind between the concepts of corruptum and caco-
zelia:
sed ne hoc genus furoris protegere videar, in Flaminino tumid-
issime dixit Murredius1*6. . .omnia autem genera corruptarum quoque
sententiarum de industria pono, quia facilius et quid imitandum et
quid vitandum sit docemur exemplo. ecce et illud genus cacozeliae
est quod amaritudinem verborum quasi adgravaturam res petit...et
illud quod Saenianus dixit habet sui generis insaniam... (Contr.
IX.2.27f. W)
Criticisms in which the elder Seneca uses the terms corruptus,
corrupte, corrumpere; insania, insanus, insanire; furor, furiose;
and cacozelia, cacozelos,^ can be used to illustrate all the main
points in the analyses of corrupt style given by Seneca the
Younger and Quintilian.
To consider first tumidity, the use of the synonyms tumidus
and inflatus in connection with corruptus is well exemplified by
Maecenas' remarks on Dorion's paraphrase of Homer in Suas. 1.12.1*8
It seems that two types of faulty expression were considered tumid
by Seneca the Elder: not just redundance, but also the bombastic
use of tropes and verba impropia.^^ In Contr. IX.2.27 one of the
examples of tumidity, the genus furoris represented by Murredius
in his treatment of this controversia (tumidissime dixit) is a
tetracolon with a last member which Seneca considered redundant:^
serviebat forum cubiculo, praetor meretrici, career convivio, dies
nocti. One of several things wrong with the last51 of the utter-
ances by Musa which are described in Contr. X pr. 9 as ad
ultimum tumorem perducta, ut non extra sanitatem sed extra naturam
essent, is the redundant use of circumlocutions: quidquid avium
volitat, quidquid piscium natat, quidquid ferarum discurrit,

nostris sepelitur ventribus. quaere nunc cur subito moriamur:
mortibus vivimus. But other quotations which Seneca considers
tumid seem objectionable not so much because of redundance - they
may even be quite concise - as on account of the perverse avoid-
ance of plain language and the abuse of various tropes. MaecenasT
reason for describing as tumidum and inflatum DorionTs Homeric
paraphrase, opous opos ontocnraTaL. . .xal x £ ^P^ a BaAXeiau vnaos, was
that his use of hyperbolic language lacked the restraint which was
so admirable in Virgilfs poetry:52 haec quo modo ex corruptis eo
perveniant ut et magna et tamen sana sint aiebat Maecenas apud
Vergilium intellegi posse; he quoted to illustrate his point
Virgilfs haud partem exiguam montis (Aen. X.128) and credas innare
revolsas\cycladas (Aen. VIII.691f.).53 In Contr. X pr. 9 Seneca
cites as illustrations of Musafs tumor a series of far-fetched
metaphors: caelo repluunt, odoratos imbres, caelatas silvas,
nemora surgentia,^ as well as the invective against meat-eating
quoted above, which contains, besides a certain amount of redun-
dance, a far-fetched funerary metaphor, nostris sepelitur
ventribus, and a grotesque use of metonymy,55 mortibus (rather
than mortuis) vivimus. Another sentence by Murredius, praetorem
nostrum in ilia ferali cena saginatum meretricis sinu excitavit
ictus securis, deemed very tumid by Seneca in Contr. IX.2.27,
includes another bold use of funerary language in an unexpected
context (ferali cena, with reference to a banquet at which an
execution took place); the application of the term saginatum
(fattened up) to a human being offends against classical standards
of restraint in metaphor; Murredius1 notion that the praetor was
only roused by the fall of the executionerTs axe made heavy de-
mands on his listeners1 credulity.56 In regarding both redundance
and misuse of tropes as species of tumidity Seneca the Elder ap-
pears to differ from his son who distinguishes the improper use of
metaphor (his fifth species of corrupti generis oratio) from
inflata explicatio which is listed first in Ep. mor. 114.I.57
Next in the younger Seneca's list comes explicatio infracta
et in morem cantici ducta (cf. Quint. VIII.3.57 compositione
fracta). As we have seen, the elder Seneca in his youth had
Elocutio 219
enjoyed chanting the explicationes of Arellius Fuscus to various

melodies (Suas. 2.10), but he makes it clear in Suas. 2.23 that he
no longer admired the fracta conpositio of the passages which in-
spired such chanting:
quarum nimius cultus et fracta conpositio poterit vos offendere
cum ad meam aetatem veneritis; interim <non> dubito quin nunc vos
ipsa quae offensura sunt vitia delectent.
Vibius Gallus, who opened his descriptions paene cantantis modo by
proclaiming 'amorem', or whatever, fdescribere volo1, earns criti-
cism in Contr. II.1.25 for deliberately imitating insane stylists
(insanos), as if insania and furor were desirable. Perhaps it was
the affected rhythm which Seneca deemed corrupt when Glycon began
an evocation of the discordant music of beggars1 cries with the
y u y y — \j v^v_/ —f —
words, aye, au 6e xAaue, au 6e 3pnveu. (Contr. X.4.22)

There is no shortage of sensus audaces et fidem egressi, to
use the younger Senecafs terms, among the sententiae which his
father and the critics quoted by him consider corrupt or insane.
Maecenas finds it admirable that, unlike Dorion, the author of the
corrupt and tumid paraphrase of Homer quoted in Suas, 1.12,
Virgil, when treating similar subject-matter, ita magnitudini
studet <ut> non inprudenter discedat a fide. In Contr. X.5 58
Seneca designates two sententiae as corrupt or insane expressly
because they are based on incredible or impossible presuppositions:
Sparsi sententia in descriptione picturae habet aliquid corrupti:
f
et, ubicumque sanguine opus est, humano utitur1; dixit enim quod
fieri non potest. {Contr. X.5.23)
Spyridion honeste <dixisse> Romanos fecit; multo enim vehementius
insanit quam nostri phrenetici. voluit videri volturios ad
tabulam Parrhasi advolare...aeque familiariter in templum
volturios subire putavit quam passeres aut columbas; dixit enim:
aapxocpaya aou y' r\ ypacpn nTtaxa Ccjia • (ibid. 27-8)
The theme of Suas. 2 5 ^ inspired both audaces sensus about writing
in blood and impossible claims about the victory of dead Spartans;
insanire and cacozelos are among the terms used to criticize them:
insanierunt in hac suasoria multi circa Othryadem: Murredius, qui
dixit: fugerunt Athenienses; non enim Othryadis nostri litteras
didicerant. Gargonius dixit, Othryades, qui perit ut falleret,
revixit ut vinceret. Licinius Nepos: cum exemplo vobis etiam
mortuis vincendum fuit. Antonius Atticus inter has puerilis
sententias videtur palmam meruisse; dixit enim: Othryades paene a
sepulchro victor digitis vulnera pressit ut trophaeo tLaconemt
inscriberet. o dignum Spartano atramentuml o virum, cuius ne
litterae quidem fuere sine sanguine! Catius Crispus, municipalis

<rhetor>, cacozelos dixit post relatum exemplum Othryadis: aliud
ceteros, aliud Laconas decet; nos sine deliciis educamur, sine
muris vivimus, sine vita vincimus. (Suas. 2.16 W)
The term arcessita, used by Quintilian in his analysis of caco-
zelon to denote far-fetched expressions, is not paralleled in the
elder Seneca1s criticisms of the style and content of sententiae,
but he uses the expression longe arcessito colore in Contr. 1.6.9,
and longe arcessere with reference to exempla in Contr. VII.5.13.
The elder Seneca's view of allusive brevity seems to have
been that explicationes...plus sensuum quam verborum habentes
(Contr. Ill pr. 7) were commendable in declamation (in this re-
spect his critical position differs from that of his son in Ep.
mor. 114.1), but that obscurity was a fault which well deserved
the name of furor. In Contr. II pr. 2 he observes how Fabianus1
style long after his philosophical conversion was still marred by
the obscurity and abruptness he had learnt to cultivate in the
school of Fuscus; in his view these qualities were vitia:
...luxuriam quidem cum voluit abiecit, obscuritatem non potuit
evadere; haec ilium usque in philosophiam prosecuta est. saepe
minus quam audienti satis est eloquitur, et in summa eius ac
simplicissima facultate dicendi antiquorum tamen vitiorum remanent
vestigia, quaedam tam subito desinunt, ut non brevia sint, sed
abrupta. (Contr. II pr. 2)
In Contr. IX.2.26f. he does not hesitate to follow Livy and the
rhetorician Miltiades in regarding the obscurity sought after by
archaizers as just as much a variety of furor as the opposite vice
of redundance:
Livius de oratoribus qui verba antiqua et sordida consectantur et
orationis obscuritatem severitatem putant aiebat Miltiaden
rhetorem eleganter dixisse: eul TO 6e£u6v uatvovTat. tamen in his
etiamsi minus est insaniae minus spei est; illi qui tument, illi
qui abundantia laborant, plus habent furoris, sed plus et corporis;
semper autem ad sanitatem proclivius est quod potest detractione
curari; illi succurri non potest qui simul et insanit et deficit,
sed ne hoc genus furoris protegere videar, in Flaminino tumid-
issime dixit Murredius...
The views expressed by Livy and Miltiades should be compared with
the reference to the Attic fever of the archaizer Annius Cimber in
ps.-Verg. Catal. 2. 6 0 It may be appropriate to note at this point
that another feature of style which the elder Seneca rather sur-
prisingly described as a kind of cacozelia was that mainstay of
Elocutio 221
the old-fashioned prosecutor, the use of harsh diction for vilifi-

cation:61
ecce et illud genus cacozeliae est, quod amaritudinem verborum
quasi adgravaturam res petit; ut in hac controversia Licinius
Nepos dixit: reus damnatus est legi, perit fornici. (Contr. IX.
2.28)
Puerile word-play was regarded as a form of corrupt style by
the elder Seneca as well as Quintilian, who complains of vocum
similium aut ambiguarum puerili captatione in VIII.3.57 along with
the other vices which we have been considering. In Suas. 7.11 we
are told that a certain declaimer, whose name has accidentally
dropped out of the text,
dixit.. .sententi.am cacozeliae genere humillimo et sordidissimo,
quod detractu aut adiectione syllabae facit sensum: Tpro facinus
indignum! peribit ergo quod Cicero scripsit, manebit quod
Antonius proscripsit?'
The craze for fPublilian' sententiae is described as an insania by
Moschus in Contr. VII.3.8.
The theory of corrupti generis oratio was closely bound up,
as we see from Sen. Ep. mor. 114, with the discussion of the
causes of the decline of Roman eloquence, which seems to have been
an echo of earlier Greek theorizing about the corruption of ora-
tory after Athens, along with the other city-states, lost her in-
dependence. For this reason, if for no other, it might seem safe
enough to conjecture that the theory of corrupt eloquence in all
its details had been imported into Latin criticism by the Atti-
cists of Cicero's time, concerned as we know them to have been
with the imitation of Greek orators belonging to the time before
this decline set in. Besides, the interesting point emerges from
a consideration of the elder Seneca's use of the terms corruptus,
corrumpere in connection with cacozelia and insania and their
cognates, that in his time corruption was regarded as the opposite
of sanum and sanitas. One remembers that the words sanitas and
insania figure in Cicero's critique of the Atticists in the Brutus
(82.284) and are commonly believed to have been favourite slogans
of these anti-Ciceronians.62
Actually the truth of the matter may be much more complicated.
If the reports of Tacitus and Quintilian on Cicero's obtrectatores
axe. based on genuine correspondence,6^ Cicero certainly was
accused by contemporary critics, including Calvus and Brutus, of a

number of the faults which later critics considered species of
cacozelon or corruption:
satis constat ne Ciceroni quidem obtrectatores defuisse, quibus
inflatus et tumens nee satis pressus sed supra modum exultans et
superfluens et parum Atticus videretur. legistis utique et Calvi
et Bruti ad Ciceronem missas epistulas, ex quibus facile est
deprehendere Calvum quidem Ciceroni visum exsanguem et attritum,
Brutum autem otiosum atque diiunctum, rursusque Ciceronem a Calvo
quidem male audisse tamquam solutum et enervem, a Bruto autem, ut
ipsius verbis utar, tamquam fractum atque elumbem. (Tac. Dial.
18.4ff.)
at M. Tullium...habemus...in omnibus quae in quoque laudantur
eminent is simum. quern tamen et suorum homines temporum incessere
audebant ut tumidiorem et Asianum et redundantem et in repetitio-
nibus nimium et in salibus aliquando frigidum et in compositione
fractum, exultantem ac paene, quod procul absit, viro molliorem...
(Quint. XII.10.12)
But nowhere is it made clear that these critics who considered
Cicero parum Atticus grouped the faults of which they accused him
together in a comprehensive theory of cacozelon or corrupta oratio.
And, probable though it is that sanitas and insania were used in
Latin Atticist criticism of the late Republic, the only thing
proved about the use of these words in Cic. Brut, 82.284 is that
they were terms acceptable to Cicero. For in this passage, far
from rejecting sanitas as an ideal, Cicero uses the term in a
description of a hypothetical type of Atticism to which he himself
would not be averse:
nam si quis eos qui nee inepte dicunt nee odiose nee putide Attice
putat dicere, is recte nisi Atticum probat neminem. insulsitatem
enim et insolentiam tamquam insaniam quandam orationis odit,
sanitatem autem et integritatem quasi religionem et verecundiam
oratoris probat.64
Cicero also recognizes in Brut. 55.202 that inflatum et corruptum
were faults to be avoided by any orator who favoured a copious
style. The fact may be that the notions of sanitas and corruptum
had already been established in Latin literary criticism so long
that, no matter what ideals of style a critic might believe in, he
would find it difficult to avoid using these terms completely, but
was free to reinterpret them according to his taste, and might
place as much or as little emphasis on them as he pleased.65
Cicero uses such terms very sparingly,66 usually without
Elocutio 223
strong technical overtones,67 and never presents us with a fully

worked out theory of corrupti generis oratio or cacozelia. This
may be because the critics who attacked him were in his view over-
fond of these concepts, though we cannot be sure. He actually
seems to have succeeded in never using in all his Rhetorica one
type of abusive term they used against him, tumens, tumidus, tumor;
he also seems to have avoided using inflatus in a derogatory sense,
and only to have lapsed from his resolve in Brut, 55.202, where it
is used in conjuction with corruptus.
Cicerofs position seems to have been that, though sanitas was
an admirable ideal and corruptum was to be avoided in diction and
style, his tastes in sanitas and incorruptum were less narrowly
restricted than those of some of his contemporaries. It is well
known that after Brut. 82.284 Cicero moves on from expressing
general assent to the principle that Attic sanitas was admirable,
into a tirade in which he points out the diversity of styles
within Attic oratory, and in particular argues the case for pre-
ferring Demosthenes to Lysias. His usage of the term incorruptum
provides interesting further testimony of the catholicity of
Cicero*s tastes: he finds this quality in the language, unaffected
by modern developments, sometimes preserved by women (De or. III.
12.45); in the critical taste of Athenians (Or. 8.25); in philo-
sophical writing (ibid. 19.64); in the style of Attic orators as
late as Hyperides, Aeschines, Lycurgus, Dinarchus and Demades
(Brut. 9.36); in the diction of Q. Catulus (ibid. 35.132).
Quintilian maintains in VIII.3.57 that cacozelon and corrupta

oratio comprise only faults of elocutio, not the rerum vitia:
stultum, commune, contrarium, supervacuum.^Q This distinction, if
known to Seneca the Elder, was not observed by him with total
rigidity. Indeed, it is a somewhat unreal distinction, for
arcessita and sensus audaces et fidem egressi are objectionable
for their content more than for their style and yet are regarded
as species of cacozelon and corrupti generis oratio. And stultus
was a term with such wide connotations in non-literary contexts
that it would be unrealistic to expect its usage to be confined
within narrow bounds in literary criticism. Together with such
terms as fatuus and ineptus and the nouns stupor and ineptiae,^
it was obviously handy for the criticism of such varieties of
cacozelia as the trivial word-play of ûblilian' sententiae.
Nevertheless there seem to be certain contexts in which the elder
Seneca always uses such terms as stultus, fatuus, ineptus in pref-
erence to corruptus, insanus, cacozelia, namely with reference to
incredible colores, futile legal quibbling, contrarium, and the
related fault of anachronism, all faults of inventio rather than
elocutio, or, to use Quintilian's terms, rerum vitia.7®
That the terms stultus and ineptus were applied to much the
same range of failings is best illustrated by Contr. VII.5.8-11 W:
circa vulnus novercae71 quidam bellas res dixerunt, quidam ineptas.
prius ilia quae belle dicta sunt referam...(10) ex illis qui res
ineptas dixerant primus ibi ante omnis Musa voster, qui cum vulnus
novercae descripsisset adiecit: at, hercules, pater meus tamquam
paries perfossus est. Murredius: patrocinium putat esse causae
suae quod sanguinem misit. Nepos Licinius ait: non est istud
vulnus, sed ludentis adulteri morsus. Saenianus ex ilia stultorum
nota sententiam protulit: non vulneravit, inquit, novercam sed
viri sui sanguine aspersit; cum ilia vulnerata ponatur. Vinicius,
exactissimi vir ingeni, qui nee dicere res ineptas nee ferre
poterat, solebat hanc sententiam Saeniani deridere...: nihil puero
est teste certius, utique quinquenni; nam et ad eos pervenit annos
ut intellegat, et nondum ad eos quibus fingat. haec finitio,
inquit, ridicula est: fnihil est puero teste certius, utique quin-
quenni1; puta nee si quadrimus puer testis est nee si sex annorum.
illud venustissime adiciebat: putes, inquit, aliquid agi: omnia in
hac sententia circumspecti hominis sunt, finitio, exceptio; nihil
est autem amabilius quam diligens stultitia.
An equation between fatuus and ineptus is to be found in Contr.
VII.4.3:
Buteo fatuam quaestionem moverat primam: an lex quae de alendis
parentibus lata esset ad patres tantum pertineret. illis omnia
privilegia data et ipsam poenam non alentium signum esse non
muliebris potestatis. res est ineptior quam ut coarguenda sit...
The overlapping of his usage of stultus, ineptus, fatuus and
their cognates, with that of such words as corrumpere and insania
can be seen in Contr. IX.6.11:
Triarius multo rem magis ineptam, quia non invenit illam sed
conrupit...
and Contr. X.5.24f.:
sed si vultis audire supra quod non possit procedere insania,
Licinius Nepos ait: si vultis digne punire Parrhasium, ipse se
pingat. non minus stulte Aemilianus quidam Graecus rhetor, quod
Elocutio 225
genus stultorum amabilissimum est, ex arido fatuus dixit: diio-

C Ilappaatov, \ir\ deAnaas ypacpetv e^ uuwv dpxexuiov eupri.
Cases where the terms stupor, fatuus, and stultus are applied
to sententiae of the 'Publilian1 sort are to be found in:
Contr. VII.2.14: et Murredius non est passus hanc controversiam
transire sine aliqua stuporis sui nota. descripsit enim ferentem
caput et manum Ciceronis Popillium et Publilianum dedit: Popilli,
quanto aliter reus Ciceronis <tangebas caput> et tenebas manum
eius!
Contr, VII.5.15: Murredius mimico genere fatuam sententiam dixit,
cum dixisset novercam disputare contra filii sui testimonium:
facit, inquit, quod solet: pro amatore sanguini suo non parcit.
79*
Suas. 7.14: Gargonius, <fatuorum>/z amabilissimus, in hac suas-
oria dixit duas res quibus stultiores ne ipse quidem umquam
dixerat; unam in principio: nam, cum coepisset scholasticorum
frequentissimo iam more a iureiurando et dixisset multa tita quam
primum tantum timeatt73 quantum potest, Tita aut totus vivat
Cicero aut totus moriatur ut ego quae hodie pro Ciceronis ingenio
dixero nulla pactione delebo.1 alteram rem dixit, cum exempla
referret eorum qui fortiter perierant: 'Iuba et Petreius mutuis
vulneribus concucurrerunt et mortes faeneraverunt.f
The second of the quotations in this last passage may be compared
with the fPublilian? sententiae of Murredius quoted in Contr. VII.
3.8: abdicationes, inquit, suas veneno diluit; et iterum: mortem,
inquit, meam effudit. The other one is not so easily classified;
it comes closest to the type of writing criticized elsewhere for
tumidity. Hence Shackleton Bailey's ingenious conjecture, ait, ut
7U*
quam primum tantum tumeat, quantum potest,' seems to point in
the right direction; note that the sententia in question came near
the beginning of Gargonius1 declamation.
Unreasonable questioning of the established interpretation of
a law is criticized in a passage (Contrt VII.4.3) which has already
been quoted to illustrate the use of fatuus and ineptus in close
conjunction.75 It seems relevant to compare Quintilian's dictum
in V.12.8: in rebus vero apertis argumentari tarn sit stultum quam
in clarissimum solem mortale lumen inferre.
Unconvincing ingenuity in the invention of colores is casti-
gated with words denoting stupidity on a number of occasions. Two
passages concerning Murredius1 habitual foolishness in the choice
of colores (Contr, VII.3.8: Murredius pro cetero suo stupore...;
Contr. IX.4.22: Murredius non degeneravit in hac controversia; nam
colorem stultissimum induxit) have already been quoted in another
connection.76 Other good examples of stupid and inept colores are

found in:
Contr. 1.3.11 (on the priestess thrown from the cliff)
Othonem Iunium patrem memini colorem stultum inducere, quod eo
minus ferendum est quod libros colorum edidit. ffortassef inquit
'poenae se praeparavit, et ex quo peccare coepit cadere condidi-
cit.'
Contr. IX.2.20 (on the execution at Flamininus' banquet)
Triarius ineptum introduxit colorem: sermo erat, inquit, in con-
vivio contemni nimiam praetoris lenitatem; alios fuisse pro-
consules, qui cotidie animadverterent, huius anno nullum esse
occisum. dixit aliquis ex convivis: 'ego numquam [iratus] <vidi
hominem occidi>.' dixit et mulier: 'et ego numquam.' iratus quod
dementia sua contemptui esset, 'curabo' inquit 'sciant non deesse
mihi <severitatem.' adducitur> sceleratus, quern videre lucem
ultra non oportet.
(Cf. also Contr. 1.4.7; VII.7.14.)
Seneca never seems to use corruptus, insania, cacozelia and their
cognates with reference to colores.
The fault of contrarium, the making of statements contrary to
the given facts of the case, or to one's own interests,77 could be
committed in a sententia which was also 'corrupt' (tumid and far-
fetched) :
illud Sparsus dixit quod non corruptum tantum sed contrarium
dicebat esse Montanus: 'solus plura habes membra quam tot hominibus
reliquisti.' ita enim hie potest videri laesisse rem publicam78
si multi sunt debilitati; apparet autem non esse multos si plura
habet membra quam debilitatis reliquit. et illud aeque aiebat ab
illo corrupte dictum: 'prodierunt plures mendici quam membra.'
(Contr. X.4.23)
But inepte and stultus are the words used to criticize errors of
this type in Contr. 1.6.11 and VII.5.10:
Contr. I.6.10f. W (on the young man who swore to marry the pirate
chief's daughter, but is now ordered by his father to divorce her
and marry a rich orphan)
<Iulius Bassus>...adiecit iurasse se per patrem. Triarius dum
sententiam puerilem7^* captat, inepte dixit iurasse se et per
orbam. aiebat enim Cestius male deseri hanc orbam <si per eam>
etiam iurasset. Latro aiebat <alterum> quoque iusiurandum ineptum
esse; nihil enim minus convenire quam aliquem per patrem iurare
patrem relicturae.
Contr. VII.5.10 W 8 0
Saenianus ex ilia stultorum nota sententiam protulit: non vulner-
avit, inquit, novercam sed viri sui sanguine aspersit; cum ilia
vulnerata ponatur.
Elocutio 227
Triarius also laid himself open to a charge of contrarium when, in

his ineptus color about the banquet of Flamininus (Contr. IX.2.20),
he referred to the meretrix of the theme merely as a mulier, and
made her out to have been just a seconder of proposals for greater
proconsular severity.
The clearest case of the use of terms denoting stupidity in
the criticism of notions which conflict with the true sequence of
historical events comes in Suas. 2.22:
sed, si vultis, historicum quoque vobis fatuum dabo. Tuscus...cum
hanc suasoriam declamaret, dixit: Texpectemus, si nihil aliud hoc
effecturi, ne insolens barbarus dicat: veni, vidi, vici 1 , cum hoc
post multos annos divus Iulius victo Pharnace dixerit.
The suasoria in question was the one about the Spartans at Thermo-
pylae. Another declaimer tried to make a rhetorical point out of
the glorious name of Thermopylae - as if it had been possible to
do so before the Spartans1 stand against Xerxes - and, probably
for this reason among others, was criticized for stupidity:
decentissimi81* generis stultam sententiam referam Victoris
Statori, municipis mei...sumpsit contradictionem: fatf inquit
'trecenti sumus'; et ita respondit: trecenti, sed viri, sed armati,
sed Lacones, sed ad Thermopylas; numquam vidi plures trecentos.
(Suas. 2.18 M)
The elder Senecafs usage of the word ineptus in connection
with stupidity provides interesting corroboration of the statement
in Cic. De or. II.4.17f. that it was a word with no Greek equiv-
alent:
quern enim nos ineptum vocamus, is mihi videtur ab hoc nomen habere
ductum, quod non sit aptus, idque in sermonis nostri consuetudine
perlate patet; nam qui aut tempus quid postulet non videt aut
plura loquitur aut se ostentat aut eorum, quibuscum est, vel
dignitatis vel commodi rationem non habet aut denique in aliquo
genere aut inconcinnus aut multus est, is ineptus esse dicitur.
hoc vitio cumulata est eruditissima ilia Graecorum natio; itaque
quod vim huius mali Graeci non vident, ne nomen quidem ei vitio
imposuerunt; ut enim quaeras omnia, quomodo Graeci ineptum
appellent, non reperies. omnium autem ineptiarum, quae sunt in-
numerabiles, haud sciam an nulla sit maior quam, ut illi solent,
quocumque in loco, quoscumque inter homines visum est, de rebus
aut difficillimis aut non necessariis argutissime disputare.
Seneca the Elder makes it clear that he did not consider ineptiae
a monopoly of the Greeks, but he confirms the fact, implied in
these remarks in the De oratore, that ineptus, despite appearances,
was not equivalent to the Greek d
4 MEMORIA
According to Quintilian (III.3.4), Albucius had advanced the her-

etical view that the art of oratory consisted of only three, not
five, parts: memoria and actio were to be excluded as being
natural gifts. Quintilian disagreed with Albucius, though he had
to concede that the opinion that actio was the product of natural
endowments was nothing new, but even had the support of the soph-
ist Thrasymachus. Seneca the Elder says nothing about this theory
of Albucius, but his own position with regard to one side of the
issue is clear enough. He was convinced that memory could be
assisted by art: Latrofs memory, for example, was natura guidem
felix, plurimum tamen arte adiuta (Contr. I pr. 17).
As we have seen,1 Seneca claims in Contr. I pr. 2 to have
been the proud possessor in youth of a memory which verged on the
miraculous: he had been capable of remembering two thousand random
names and reproducing them in the correct sequence, and of memor-
izing more than two hundred assorted lines of poetry and then
reeling them back to his amazed school-fellows in reverse order.
About Latrofs memory too he makes claims calculated to astound his
readers:
in illo non tantum naturalis memoriae felicitas erat, sed ars
summa et ad conprehendenda quae tenere debebat et ad custodienda,
adeo ut omnes declamationes suas, quascumque dixerat, teneret
etiam. itaque supervacuos sibi fecerat codices; aiebat se in
animo scribere. cogitata dicebat ita ut in nullo umquam verbo eum
memoria deceperit. historiarum omnium summa notitia: iubebat ali-
quem nominari ducem et statim eius acta cursu reddebat; adeo,
quaecumque semel in animum eius descenderant, in promptu erant.
(Contr. I pr. 18)
He goes on to assure his sons that there is nothing to prevent
them from achieving similar feats:
video vos, iuvenes mei, plus iusto ad hanc eius virtutem obstup-
escere; alia vos mirari in illo volo: hoc quod tantum vobis
videtur, non operosa arte tradi potest. intra exiguum paucissim-
orum dierum tempus poterit quilibet facere illud, quod Cineas
fecit, qui missus a Pyrrho legatus ad Romanos postero die novus
Memoria 229
homo et senatum et omnem urbanam circumfusam senatui plebem

nominibus suis persalutavit; aut quod ille fecit, qui recitatum a
poeta novum carmen dixit suum esse et protinus <ex> memoria
recitavit, cum hoc ille cuius carmen erat facere non posset; aut
quod fecit Hortensius, qui a Sisenna provocatus in auctione per-
sedit per diem totum et omnes res et pretia et emptores ordine suo
argentariis recognoscentibus ita ut in nulla re falleretur,
recensuit. cupitis statim discere? suspendam cupiditatem vestram
et faciam alteri beneficio locum; interim hoc vobis in quo iam
obligatus sum persolvam. (Contr, I pr. 19 W)
Thus he leaves the topic of memory without giving us any infor-
mation about the mnemonic technique in the efficacy of which he so
firmly believed.
It is to Cicero (De or. II.86.35Off.), the Rhetorica ad
Herennium (III.16.28ff.) and Quintilian (XI.2.Iff.) that we must
look for descriptions of the chief classical system of mnemonics.2
The assumption fundamental to this system was that visual images
were more easily remembered than the written word: acerrimum...ex
omnibus nostris sensibus esse sensum videndi (Cic. De or. 11.87.
357) . It was also noticed that the sight of places revisited had
the power to bring back to one's mind all manner of details about
the previous occasions when one had been there:
nam cum in loca aliqua post tempus reversi sumus, non ipsa
agnoscimus tantum sed etiam quae in iis fecerimus reminiscimur,
personaeque subeunt, nonnumquam tacitae quoque cogitationes in
mentem revertuntur. (Quint. XI.2.17)
The memorizer was therefore recommended to fix in his mind an
image of some complex piece of architecture:
loca discunt quam maxime spatiosa, multa varietate signata, domum
forte magnam et in multos diductam recessus. in ea quidquid
notabile est animo diligenter adfigunt, ut sine cunctatione ac
mora partis eius omnis cogitatio posset percurrere. (Quint. XI.2.
18; cf. Cic. De or. 11.86.354; Rhet. Her. III.16.29ff.)
Against the background of the various parts of the imagined
building - its rooms, for example, or the spaces between its
columns - he was told to set mental pictures designed to remind
him of the various items to be memorized:
imagines sunt formae quaedam et notae et simulacra eius rei quam
meminisse volumus; quod genus equi, leonis, aquilae memoriam si
volemus habere, imagines eorum locis certis conlocare oportebit.
(Rhet. Her. III.16.30; cf. Cic. De or. 11.86.354; Quint. XI.2.
19ff.)
Their arrangement in relation to the architectural background was
intended, of course, to remind him of the correct order in which

they were to be recalled (Cic. De or. 11.86.354: ut ordinem rerum
locorum ordo conservaret). The items to be memorized might be
words (single words or whole lines of poetry), or events, for
example the circumstances of a crime alleged to have been commit-
ted:
duplices igitur similitudines esse debent, unae rerum, alterae
verborum. rerum similitudines exprimuntur cum summatim ipsorum
negotiorum imagines conparamus; verborum similitudines constitu-
untur cum unius cuiusque nominis et vocabuli memoria imagine
notatur. (Rhet. Her. III.20.33; cf. Cic. De or. 11.86.354; Quint.
XI.2.19)
If the items to be memorized were names of concrete objects
there was no difficulty about conceiving imagines of them. Quint-
ilian, though sceptical about some of the claims of the mnemonists
(XI.2.22, 24ff.), recognizes the usefulness of the visual mnemonic
system for such feats as the one which Hortensius was said to have
performed at an auction:
equidem haec ad quaedam prodesse non negaverim, ut si rerum nomina
multa per ordinem audita reddenda sint. namque in iis quae didic-
erunt locis ponunt res illas: mensam, ut hoc utar, in vestibulo et
pulpitum in atrio et sic cetera, deinde relegentes inveniunt ubi
posuerunt. et forsitan hoc sunt adiuti qui auctione dimissa quid
cuique vendidissent testibus argentariorum tabulis reddiderunt,
quod praestitisse Q. Hortensium dicunt. (ibid. 23f.)
Rather heavier demands were made on one's imagination when abstrac-
tions had to be remembered, though for the commoner ones standard
imagines, whether invented by the memorizer himself or derived from
some theorist, could be used over and over again.^ Quintilian
suggests two simple images to denote abstract concepts: sit autem
signum navigationis ut ancora, militiae ut aliquid ex armis (XI.
2.19). There were two main ways in which the memorizer could re-
mind himself of names. The first was to conceive a picture of
some namesake of the man he wished to remember, either a famous
historical figure or one of his own friends who happened to share
the same name: si Fabius forte sit tenendus, referamus ad ilium
Cunctatorem, qui excidere non potest, aut ad aliquem amicum qui
idem vocetur (Quint. XI.2.30). The other method was to derive an
image from the etymology of the man's name: quod est facilius in
Apris et in Ursis et Nasone aut Crispo, ut id memoriae adfigatur
Memoria 231
unde sunt nomina. origo quoque aliquando declinatorum tenendi

magis causa est, ut in Cicerone, Verrio, Aurelio (ibid. 31). To
memorize a line of poetry one was required to go through some very
odd mental contortions:
cum verborum similitudines imaginibus exprimere volemus, plus
negotii suscipiemus et magis ingenium nostrum exercebimus. id nos
hoc modo facere oportebit:
iam domum itionem reges Atridae parant.
<hunc versum meminisse si volemus, conveniet primo> in loco con-
stituere manus ad caelum tollentem Domitium cum a Regibus Marciis
loris caedatur - hoc erit fiam domum itionem reges'; in altero
loco Aesopum et Cimbrum subornari ut ad Iphigeniam in Agamemnonem
et Menelaum - hoc erit 'Atridae parant1. hoc modo omnia verba
erunt expressa. (Rhet. Her. III.21.34)4
Some of the visual punning recommended as an aid to the memorizing
of facts in a lawsuit is equally bizarre:
rei totius memoriam saepe una nota et imagine simplici conprehend-
imus; hoc modo, ut si accusator dixerit ab reo hominem veneno
necatum et hereditatis causa factum arguerit et eius rei multos
dixerit testes et conscios esse. si hoc primum, ut ad defendendum
nobis expeditum sit, meminisse volemus, in primo loco rei totius
imaginem conformabimus; aegrotum in lecto cubantem faciemus ipsum
ilium de quo agetur, si formam eius detinebimus; si eum non
agnoverimus, at aliquem aegrotum non de minimo loco sumemus, ut
cito in mentem venire possit. et reum ad lectum eius adstituemus,
dextera poculum, sinistra tabulas, medico testiculos arietinos
tendentem. hoc modo et testium et hereditatis et veneno necati
memoriam habere poterimus. (ibid. 20.33)
The memorizer was supposed to form images in the same way to ex-
press the other counts against the accused, and to set each in
sequence against its architectural background (ibid. 16.29).
Rival mnemonic systems existed, notably the astrological
method of Metrodorus, with its provision of three hundred and
sixty distinct loci, to which Quintilian refers contemptuously in
XI.2.22. But the principle that images of the items to be remem-
bered should be set against a background was the same even here.
The detailed discussions of memory in the Rhetorica ad Herennium
and Quintilianfs eleventh book can be relied upon as providing a
rough guide at least to the principles on which the elder Seneca's
mnemonic system must have worked, and probably more than that,
seeing that the feat of Hortensius which is mentioned in Contr. I
pr. 10 is associated by Quintilian with the architectural method
(XI.2.23f.). The secret behind the elder Seneca's ability in
youth to reproduce countless names or lines of poetry in sequence

is laid bare. What is not made clear is how such a mnemonic sys-
tem could have been of much assistance to him in the word-for-word
memorizing of thousands of declamatory excerpts, many dealing with
the same faceless characters and stock situations. He would have
needed to go through life envisaging hieroglyphs for every brick
in Corduba. That is not to deny, however, that reports of seem-
ingly miraculous feats of memory, however achieved, are widespread
throughout human history.5
There is no doubt that the memorizing of declamatory models
for imitation was encouraged in the schools of rhetoric, along
with the kind of rote-learning which provided Latro with a ready
supply of exempla.^ Pupils of a traditionally-minded rhetorician
might perhaps be recommended to look to CiceroTs speeches for
their models, though there was a growing tendency, deplored by
Cassius Severus, for students to learn by heart the declamations
of their teacher to the exclusion of classical exemplars; this was
the practice in the school of Cestius, where the students read no
Cicero except the speeches to which their master had composed
answers, and did not memorize even these (Contr. Ill pr. 15).
Poetic extracts were also sometimes recommended as rhetorical
models worth remembering:
P. Vinicius, summus amator Ovidi, hunc aiebat sensum disertissime
apud Nasonem Ovidium esse positum, quem ad fingendas similes
sententias aiebat memoria tenendum. occiso Achille hoc epiphonema
poni:
quod Priamus gaudere senex post Hectora posset,
hoc fuit.
(Contr. X.4.25; Ov. Met. XII.6O7f.)
Twice, however, Seneca complains that there had been during
his life-time a marked decline in the amount of memorizing custom-
ary among students of rhetoric; one result of this had been an
increase in undetected plagiarism:
quis est, qui memoriae studeat? quis est, qui non dico magnis
virtutibus, sed suis placeat? sententias a disertissimis viris
iactas facile in tanta hominum desidia pro suis dicunt et sic
sacerrimam eloquentiam, quam praestare non possunt, violare non
desinunt. (Contr. I pr. 10)
...memini auditorem Latronis Abronium Silonem...recitare carmen
in quo agnovimus sensum Latronis...tam diligentes tune auditores
erant, ne dicam tam maligni, ut unum verbum surripi non posset;
Memoria 233
at nunc cuilibet orationes in Verrem7 tuto licet pro suis <dicere>.

(Suas. 2.19 W)
Orators and declaimers of Senecafs time did not always write
out their speeches and declamations in full before committing them
to memory. Extemporization was known in both fields, though total
reliance on it was not normal, and even some of the most competent
speakers suffered from a curious lack of confidence which pre-
vented them as a rule from improvising. Haterius declaimed ad-
misso populo ex tempore (Contr. IV pr. 7 ) , but Albucius, that
other master of declamatory fluency, did not trust his powers of
improvisation sufficiently to use them in public: dicebat enim
citato et effuso cursu, sed praeparatus. extemporalis illi
facultas, ut adfirmabant qui propius norant, non deerat, sed
putabat ipse sibi deesse (Contr. VII pr. 2 ) . Cassius Severus
always wrote very full notes before speaking in the courts, though
it was a remarkable fact that when forced to make an impromptu
reply to anyone he would excel himself:
sine commentario numquam dixit nee hoc commentario contentus erat,
in quo nudae res ponuntur, sed ex maxima parte perscribebatur
actio; ilia quoque, quae salse dici poterant, adnotabantur; sed
cum procedere nollet nisi instructus, libenter ab instrumentis
recedebat. ex tempore coactus dicere infinito se antecedebat.
numquam non utilius erat illi deprehendi quam praeparari; sed
magis ilium suspiceres, quod diligentiam non relinquebat, cum illi
tarn bene temeritas cederet. (Contr. Ill pr. 6)
Some declaimers, especially beginners, must similarly have felt
the need to write out very full commentarii before venturing to
declaim, but there was an alternative practice which came part way
between declamation completely prepared in writing and total im-
provisation. Latro devoted days on end to writing (Contr. I pr.
14: cum se ad scribendum concitaverat, iungebantur noctibus dies
et sine intervallo gravius sibi instabat nee desinebat, nisi de-
fecerat), but he does not seem to have written out full versions
of his declamations. He wrote out in full the schemata required
in a declamation (Contr. I pr. 23); he may also have written down
the sententiae to which he devoted whole days of practice (ibid.),8
and in the case of enthymemata and epiphonemata these must gener-
ally have been contrived with particular declamations in mind.
Latro's carefully worked-out divisiones also look like the product
of written preparation,9 though we are told nothing about the way

in which they were composed. Nevertheless to some extent he evi-
dently relied on mental, unwritten, preparation: cogitata dicebat
ita, ut in nullo umquam verbo eum memoria deceperit (Contr. I pr.
18). Quintilian, when weighing up the arguments for and against
the written word as an aid to memory, was to reflect on the way
that quae per plures dies scribimus ediscenda sint, cogitatio se
ipsa contineat (XI.2.10). What Seneca says about the tenacious-
f
ness of Latro s memory for cogitata bears out the truth, at least
where a gifted man was concerned, of the latter half of Quinti-
lian^ observation; Latro was also fortunate in not needing, un-
like most people, to spend days memorizing what he had written,
despite the fact that he was not a slow or anxious writer:
numquam ille quae dicturus erat ediscendi causa relegebat: edidic-
erat ilia, cum scripserat. id eo magis in illo mirabile videri
potest, quod non lente et anxie, sed eodem paene quod dicebat
impetu scribebat. illi, qui scripta sua torquent, qui de singulis
verbis in consilium eunt, necesse est quae totiens animo suo ad-
movent novissime adfigant; at quorumcumque stilus velox est,
tardior memoria est. (Contr. I pr. 17f.)
His memory was not only good in the short term: he could remember
all his past declamations (ibid. 18).
It would probably be fair to surmise that declaimers varied
in the extent to which they prepared their work in writing as
widely as orators, whose ranks included in this period the negli-
gent Scaurus, who used to learn the essentials of a forthcoming
lawsuit while getting dressed or actually after arrival in court
(Contr. X pr. 2 ) , as well as the painstaking Cassius Severus,
whose preparations are described in Contr. Ill pr. 6.
5 ACTIO
Pronuntiatio a plerisque actio dicitur, sed prius nomen a voce,

sequens a gestu videtur accipere. namque actionem Cicero alias
'quasi sermonem', alias 'eloquentiam quandam corporis1 dicit.
idem tamen duas eius partis facit, quae sunt eaedem pronuntiati-
onis, vocem atque motum: quapropter utraque appellatione in-
differenter uti licet. (Quint. XI.3.1)
The standard view of rhetorical theorists was that actio,*
like memoria, was not wholly the product of the speakerTs natural
gifts (e.g. Quint. III.3.1), though in the elder Seneca's day
Albucius, not without some support from earlier tradition, re-
jected the normal inclusion of actio among the five parts of
orandi ratio (ibid. 4 ) . The attitudes of Seneca and Latro towards
this question seem closely comparable with those of Cicero and
Quintilian. Seneca does not deny that prior exercitatio of a kind
could be beneficial to the speaker's actio, but, in common with
the theorists, he does not wish orators to sound like actors, and
Porcius Latro is said to have rejected certain forms of vocal and
physical exercise which we know from Cicero and Quintilian to have
formed part of the training of actors and singers.
It was recognized to be an advantage for the declaimer to
have, as Latro had, a naturally strong voice (robusta - Contr. I
pr. 16), best of all one which combined strength with sweetness,
such as Cassius Severus possessed, suavitas valentissimae vccis -
quamvis haec inter se raro coeant, ut eadem vox et dulcis sit et
solida (Contr. Ill pr. 3 ) , but it was possible to be a successful
declaimer without a perfect voice. As one critic pointed out, it
was easier to keep the attention of an audience in the schools
than to project one's voice and personality in the forum: illic
inter fremitum consonantis turbae intendendus animus est, vox ad
aures iudicis perferenda, hie ex vultu dicentis pendent omnium
vultus... (Contr. IX pr. 5 ) . 2 Latro's success was not impeded by
the fact that his voice was not as clear as it should have been,
as a result of his contempt for training in actio and a tendency

to overwork, vox robusta sed surda, lucubrationibus et neglegentia,
non natura infuscata (Contr. I pr. 16). One might have thought
that Gargonius1 vocal qualities would have ruled out any prospects
for him of a career in rhetoric, and yet he succeeded in becoming
master of a school: Gargonius fuit Buteonis auditor, postea scholae
quoque successor, vocis obtusae sed pugnacissimae, cui Barrus
scurra rem venustissimam dixit: centum raucorum vocem habes (Contr.
1.7.18).
A strong rib-cage was reckoned important for effective voice-
production: Latro coped with the heavy demands made on his voice
beneficio...laterum {Contr. I pr. 16). One way of exercising the
latera was to go for healthy walks, latus ambulatione reparare
(ibid.). This Latro neglected to do, but fortunately at various
times he had led a sufficiently active life for Seneca to be able
to say corpus illi erat et natura solidum et multa exercitatione
duratum (Contr. I pr. 16). 3 Another exercise recommended for
the voice and similarly neglected by Latro because, according to
Seneca, he could not unlearn ilium fortem et agrestem et Hispanae
consuetudinis morem ; utcumque res tulerat, ita vivere (ibid.),
was intended to give the speaker command over all the gradations
of tone which he might require: per gradus paulatim ab imo ad
summum perducere...a summa contentione paribus intervallis
descendere (ibid.).
In order to deliver a declamation in the fiery and agitated
style one needed to muster all one's strength - Contr. VII pr. 1:
omnes vires advocabat [sc. Albucius] - so as to gather sufficient
impetus to carry one through. Latro found no difficulty in sum-
moning up the necessary energy, numquam impetus ardentis animi
deseruit (Contr. I pr. 16), and so, in spite of his lack of care
for his voice and physique, he was able to amaze his audiences
with sustained forcefulness - Contr. 1.7.16: advocavit vires suas
tanto totius actionis impetu, ut attonitos homines tenuerit - and
to deliver the high points of his declamations summis clamoribus
(Contr. 1.1.21; VII.2.9).
For violent delivery, as for agitated compositio, we are told
in Contr. VII.4.7 that Calvus provided a precedent: we are informed
Actio 237
that he was a violentus actor et concitatus, who had been known to

leave his place in court and, impetu latus, to run right over to
the benches of his opponents. It is also implied (Contr. VII.4.8)
that the epilogus of his Pro Messio could serve as an illustration
of Latro's teaching that one should desist from agitated delivery
for special effects of pathos (Contr. VII.4.6: in epilogis nos de
industria vocem quoque infringere et vultum deicere et dare
operam, ne dissimilis orationi sit orator). Cicero would have
been surprised to find his rival credited with that controlled
variability of actio which he had recommended for his ideal orator:
volet igitur ille qui eloquentiae principatum petet et contenta
voce atrociter dicere et summissa leniter et inclinata videri
gravis et inflexa miserabilis... (Or. 17.56).
Seneca does not tell us much about the gestures which ac-
companied declamation beyond his reference to Latro's teaching
that the speaker's facial expression, like his voice, should ac-
cord with his subject-matter (Contr. VII.4.6). He mentions how
the declaimer Seneca Grandio at one juncture stood on tip-toe and
raised his hands in the air before uttering the portentous words,
gaudeo, gaudeo: totus Xerses meus eriti (Suas. 2.17), but as this
attempt to increase his height was but one manifestation of the
megalomania for which this individual was noted,5 it cannot be
supposed typical of declamatory gestus. Much use must however
have been made in the schools of what Cicero had described as
quasi sermo corporis (De or. III.59.222); at any rate the gestures
of declaimers were evidently often applauded: cum ventum est in
forum et desiit illos ad omnem gestum plausus excipere, aut de-
ficiunt aut labant (Contr. IX pr. 2 ) . Senecafs remarks on Cassius
Severus in Contr. Ill pr. 3 show that he was well aware that im-
pressive physical appearance (corporis magnitudo conspicua) could
contribute greatly to any speaker's success.
It appears that it was normal to declaim standing up (see
e.g. Suas. 2.17), but Albucius in his private declamations began
sitting, and only stood up when he started to feel inspired (Contr.
VII pr. 1). Latro's custom was to sit while giving his preliminary
plain statement of the divisio, before he began to declaim (Contr.
I pr. 21). We learn from Contr. IX.3.13 that some declaimers at
least changed their dress to suit the language in which they were
going to declaim: non...contenti unius linguae eloquentia, cum
Latine declamaverant, toga posita sumpto pallio quasi persona
mutata rediebant et Graece declamabant.
Faults of actio mentioned by Seneca the Elder include, be-
sides the raucousness of Gargonius and the exaggerated gestures of
Seneca Grandio, excessive rapidity of speaking, inability to pro-
ject violent emotions, and sing-song delivery. The velocitas
orationis of Haterius provoked Augustus to quip, Haterius noster
sufflaminandus est (Contr. IV pr. 7). Fabianus, having endeavoured
to overcome his own emotions in accordance with his philosophical
beliefs, became too tranquil to enact effectively the highly-
wrought emotions of the characters whose part he had to take in
declamations (Contr. II pr. 2 ) . When Arellius Fuscus abandoned
the dry style of his principia, narrationes and argumenta in
favour of his more luxuriant manner (ibid, 1 ) , the compositio of
his elaborate sentences was such that they almost demanded to be
sung (Suas. 2.10). Vibius Gallus is reported to have introduced
his descriptions paene cantantis modo (Contr. II.1.25).6 One re-
calls how Cicero (Or. 17.57) refers to the epilogus paene canticum
characteristic of Phrygian and Carian rhetoricians. Quintilian
was to complain that in the years since Cicero had written the
habit of chanting had become more and more widespread among ora-
tors until in his day he might well ask: quisquamne, non dico de
homicidio sacrilegio parricidio, sed de calculis certe atque
rationibus, quisquam denique, ut semel finiam, in lite cantat?
(XI.3.58). It is not, however, warrantable to assume that this
habit was as generally prevalent in the elder Seneca's time as in
Quintilian's.
In the third preface Seneca distinguishes between the pro-
nuntiatio of Cassius Severus and the type which one would expect
from an actor: pronuntiatio quae histrionem posset producere,
<nec> tamen quae histrionis posset videri (Contr. Ill pr. 3 W ) .
This distinction was one which it was customary for critics to
draw when discussing actio: Cicero, before Porcius Latro, had
questioned the need for importing into oratorical training such
practice methods, evolved by Greek tragic actors, as the exercise
Actio 239
in which vocem cubantes sensim excitant eandemque, cum egerunt,

sedentes ab acutissimo sono usque ad gravissimum sonum recipiunt
et quasi quodam modo conligunt (De or. 1.59.251). Quintilian (XI.
3.19ff.) contrasts the moderate amount of attention which the ora-
tor should pay to actio with the training methods of phonasci.
Interestingly, he considers it unreasonable to expect the orator
to find time even for those healthy walks spurned by Latro, let
alone for elaborate vocal exercises, nam neque certa tempora ad
spatiandum dari possunt tot civilibus officiis occupato, nee prae-
parare ab imis sonis vocem ad summos nee. semper a contentione
condere licet, cum pluribus iudiciis saepe dicendum sit (ibid.
22).
Much less in line with the teaching of the classical theor-
ists is a remark made by Seneca in Contr. X pr. 12. Referring to
Iulius Bassus he comments: ...demptam velles quam consectabatur
amaritudinem et simulationem actionis oratoriae. nihil est in-
decentius quam ubi scholasticus forum quod non novit imitatur.
That Seneca the Elder should have affirmed, even if half in jest,
the belief that it was not desirable for scholastics to imitate
forensic actio, sets him apart both from the uncompromising enem-
ies of the declamatory schools, and from Quintilian who, though
fairly tolerant of the use of suasoriae and controversiae in rhe-
torical training, would never have lost sight of the idea that
their function was to train future orators.
PART IV
THE PLACE OF EARLY IMPERIAL DECLAMATION IN LITERARY HISTORY:
THE ELDER SENECA'S EVIDENCE
1 ASIANISM, ATTICISM, AND THE STYLES OF THE DECLAIMERS
Eduard Norden, who in his Antike Kunstprosa viewed the whole his-
tory of Hellenistic and Latin prose style as a centuries-long
battle* between the rival tendencies of Atticism and Asianism, was
confident that what he called the new style (der neue Stil) charac-
teristic of early Imperial declaimers, could in its entirety be
regarded as a manifestation of Asianism.2 Various points emerge,
however, from the elder Seneca's criticism of the declaimers which
suggest that Norden1s diagnosis was an over-simplification. First,
it is evident that, however uniform in style declamatory sententiae
as excerpted by the elder Seneca may seem to readers today, the
sensitive ears of the anthologist himself detected many distinct
genera dicendi in use among the declaimers of his time, not just
one all-embracing type of modernity.3 Secondly, though the elder
Seneca does use the term Asianus in his criticism, it is note-
worthy that he does so very sparingly, applying this epithet only
to four Greek declaimers and probably, though the text of the
passage in question is doubtful, to one Latin declaimer, namely
Fuscus, whose style, as we have seen,4 had certain characteristics
which set it apart from that of Latro and the genus dicendi ardens
et concitatum which was Seneca's ideal.
Norden based his contention that the new Latin style of the
early Empire was a species of Asianism on four passages from
ancient authorities.5 He cites first Suetonius, Aug. 86, a de-
scription of the efforts of Augustus to steer a middle course in
style. Augustus, we are told here, was contemptuous of two dis-
tinct kinds of extremists, cacozelos et antiquarios, ut diverso
genere vitiosos, laughing on the one hand at Maecenas for his
r
myrobrechis.. .cincinnos\ and on the other at Tiberius for his
predilection for exoletas...et reconditas voces; he also found
fault with Mark Antony for the eccentric wavering of his style
between archaism fit to rival that of Annius Cimber, Veranius
The place of early Imperial declamation in literary history 244
Flaccus and Sallust, and imitation of the Asiaticorum oratorum

inanis sententiis verborum volubilitas. From this passage Norden
infers that the dominant trends in the prose style of the Augustan
age must have been towards archaism on the one hand and, on the
other, towards a modern style, whose exponents were criticized for
cacozelia and considered imitators of Asiatic orators. He ac-
knowledges that some stylists avoided both extremes, but names
only Augustus in this connection among prose writers of the elder
Seneca?s period. The second passage which Norden cites is Plut-
arch, Ant, 2.5, where it is said of Antony: exP^xo 6e T $ HaAouyev^
yev 'Aauxvqj CnA(p TWV Xoywv dvSouvTt yaXuaxa HOIT' CMeCvov TOV
Xpovov... He fails to point out that the time of Antony's ac-
tivity as an orator was rather too early for evidence about his
style and that of his contemporaries HOLT* exeovov TOV XP° VOV >
necessarily to have any relevance to a consideration of early
Silver Latin. He also omits to observe that Suetonius1 analysis
of Antony's rhetorical manner is more complex than Plutarch's.
For further evidence that the modernism of early Imperial Latin
rhetoric should be considered a form of Asianism, Norden then
turns to Petronius' Satyricon, the extant remains of which open
with an attack on the style of declamation represented by the
rhetor Agamemnon, whose school Encolpius has been visiting. After
a fine parody of a melodramatic type of declamation, 'haec vulnera
pro libertate publica excepi, hunc oculum pro vobis impendi: date
mihi ducem qui me ducat ad liberos meos, nam succisi poplites
membra non sustinent1, and various strictures on the empty racket
of sententiae9 on unrealistic declamation themes, and on mellitos
verborum globulos et omnia dicta factaque quasi papavere et sesamo
sparsa, Encolpius alleges that:
nuper ventosa istaec et enormis loquacitas Athenas ex Asia com-
migravit animosque iuvenum ad magna surgentis veluti pestilenti
quodam sidere afflavit, semelque corrupta regula eloquentia stetit
et obmutuit. (Sat. 1-2)
Norden's last piece of evidence is Quintilian's remark on actio in
XI.3.58: Cicero illos ex Lycia et Caria rhetoras paene cantare in
epilogis dixit: nos etiam cantandi severiorem paulo modum ex-
cess imus . The modern development referred to,^ as we have noted,
need not be assumed to have become general as early as the elder
Asianism, Atticism, and the styles of the declaimers 245
Seneca's time.
Full consideration of Norden's equation of the early Imperial
'New Style1 with Asianism must be postponed until all the main
varieties of declamatory prose recognized by the elder Seneca have
been considered in turn.7 Meanwhile let us look at Seneca's ex-
plicit references to Asiani. They are very few in number.
Contr. 1.2.23 W (Seneca is quoting Scaurus on the subject of ob-
scenity)
Hybreas, inquit, cum diceret controversiam de illo qui tribadas
deprehendit et occidit, describere coepit mariti adfectum, in quo
non deberet exigi inhonesta inquisitio: eyw 6* eaxonncja rcpoxepov
T6V av6pa, <et> eyYeYEvnxat xws n itpoaeppontxat. Grandaus, Asianus
aeque declamator, cum diceret in eadem controversial 'non ideo
occidi adulteros [non] paterentur', dixit: eu 6e cpnXappeva uouxov
eXa$ov.
Contr. IX.1.12f.: Latro dixit: filiam tuam dimittam? quid
adultero faciam? pro una rogas, duos eripis. hanc Hybreas aliter
dixit sententiam: aol 6e, youxe, TI notnaw; yn xal aoO KaXXtas
Tiaxnp eaxuv; haec tota diversa sententia est a priore, etiamsi ex
eadem est petita materia. ilia non est similis sed eadem quam
dixit prior Adaeus, rhetor ex Asianis non proiecti nominis, deinde
Arellius Fuscus: dxotpLaxos aou 6oxw, KaXXua; oux o£6as nov3 you xnv
XOtpLV e6wxas; hanc sic mutavit Arellius Fuscus: non dices me,
Callia, ingratum: unde redemeris cogita. memini deinde Fuscum,
cum haec Adaei sententia obiceretur, non infitiari transtulisse se
earn in Latinum; et aiebat non commendationis id se aut furti, sed
exercitationis causa facere.
(After this, Fuscus defends himself by citing an imitation by
Sallust of a sententia allegedly by Thucydides.)
Contr. IX.6.16 (Seneca is quoting Pompeius Silo)
dixit, inquit, Hybreas: xt o$v; eêuaaxo xaxa xns L,6uas Suyaxpos;
oux* dXXa xaxot xns eyns. hanc sententiam Fuscus Arellius, cum
esset ex Asia<nis>, non casu dixit, sed transtulit ad verbum
quidem: quid ergo? inquit, mentita est de filia sua? immo de mea.
Asianis Schultingh; asia MSS8*
Contr. X.5.21: hie est Craton, venustissimus homo et professus
Asianus, qui bellum cum omnibus Atticis gerebat. cum donaret illi
Caesar talentum, in quo viginti quattuor sestertia sunt Athenien-
sium more: n npoades, cpnatv, n acpeX' , tva yn 'AXXLXOV ?.
This last passage contains the only reference in the whole of the
elder Seneca's writing to the Attic side of the polemic.^ It is a
frivolous reference, but the words professus Asianus^® used in the
description of Craton's animosity are significant in showing that
Asianus cannot simply be regarded as a geographical designation
when it occurs in the elder Seneca's criticism.
As we see, all Seneca's explicit references to the rhetoric
of Asiani have to do with minutiae, that is, imitations of par-

ticular sententiae in particular controversiae. However, it is
notable that the style of Arellius Fuscus differs strikingly from
that of most of his contemporaries in the Latin schools, and it
may well be that it is only accidental that no passage is extant
in which Seneca designates his genus dicendi, or one of his genera
dicendi, as Asianic.
The chief peculiarity of Fuscus' style was his habit of di-
versifying an otherwise dry declamatory manner with extraordi-
narily florid descriptive passages, unmanly in compositio and out-
rageously bold in diction. In excerpts from his drier passages
his style does not seem fundamentally different from that of
Latro. Compare for example Latro in Contr. 1.2.1:
o egregium pudicitiae patrocinium: 'militem occidi1! at hercule
lenonem non occidisti. deducta es in lupanar, accepisti locum,
pretium constitutum est, titulus inscriptus est: hactenus in te
inquiri potest; cetera nescio,
with Fuscus in Contr. 1.2.5:
ne metue puella: pudica es; sed sic te viro lauda, non templo.
meretrix vocata es, in communi loco stetisti, superpositus est
cellae tuae titulus, venientem recepisti: cetera, etiamsi in
communi loco essem, tamen potius silerem.
But there is nothing in the fragments of Latrofs declamations to
compare, for example, with the explicatio quoted from Fuscus1
treatment of the theme, Agamemnon deliberat an Iphigeniam immolet
negante Calchante aliter navigari fas esse:
non in aliam condicionem deus fudit aequora, quam ne omnis ex voto
iret dies; nee ea sors mari tantum est: caelum specta, non sub
eadem condicione sidera sunt? alias negatis imbribus exurunt
solum, et miseri cremata agricolae lugent11 semina, et haec
interdum anno lex est; alias serena clauduntur, et omnis dies
caelum nubilo gravat: subsidit solum, et creditum sibi terra non
retinet; alias incertus sideribus cursus est, et variantur tempora,
neque soles nimis urguent neque ultra debitum imbres cadunt: quid-
quid asperatum aestu est, quidquid nimio diffluxit imbre, invicem
temperatur altero; sive ista natura disposuit, sive, ut ferunt,
luna cursu gerit - quae, sive plena lucis suae est splendensque
pariter adsurgit in cornua, imbres prohibet, sive occurrente
nubilo sordidiorem ostendit orbem suum, non ante finit quam lucem
reddit -, sive ne lunae quidem ista potentia est, sed flatus, qui
occupavere, annum tenent: quidquid horum est, extra iussum dei
tutum fuit adultero mare. (Suas. 3.1 M)
There could be no better illustration than this sentence of
Seneca's criticism, erat explicatio Fusci Arelli splendida guidem
sed operosa et implicata (Contr. II pr. I ) . 1 2 Sentence construc-

tion of this exceptional complexity must have seemed most abnormal
to an audience accustomed to the tersely paratactic genus dicendi
ardens et concitatum. Balancing words serve to mark the limits of
the various sections of the elaborate, not to say top-heavy,
structure:
alias ... alias ... alias ...
quidquid ... quidquid ...
sive ... sive ...
sive ... sive ...
sive ...
quidquid ...
Within this framework Fuscus works out such small-scale effects of
rhetorical symmetry as the isocolon,
KJKJ— \J — — — U U — — — \J —
alias negatis imbribus exurunt sol(um) et 14 syllables
miseri cremat(a) agricolae lugent semma 14 syllables
and may well have made conscious efforts to ensure that the open-
ing long parallel sections,
alias...lex est 35 syllables
alias...retinet 37 syllables
alias...cadunt 39 syllables
were of comparable, but progressively increasing, length. He
certainly seems to have aimed at conformity of rhythm at the open-
ings of certain successive clauses, for example,
\j KJ — <u — —
alias negatis imbribus...
miseri cremat(a) agricolae...
alias serena clauduntur...
and again,
siv(e) ista natura disposuit...
siv(e) ut ferunt luna cursu gerit...
a rhythm on which he then produces a slight variation at the open-
ing of the next clause,
quae sive plena lucis suae (e)st...
which is resolved in the next into
splendensque pariter adsurgit...
Yet the effect of the whole is not one of discipline, but of that
wandering effusiveness of which Seneca complains in Contr, II pr.
1 (nimia licentia vaga et effusa).
All through the explicatio of Suas. 3.1 over-poetic snatches
of rhythm catch the attention, minor Ionics here, neque soles

nimis urguent..., anapaests there, alYas lhcelctus sideribus. ..
We have, in fact, in this passage, a fine object lesson in the
reasons why we find classical theorists recommending that rhetori-
cal prose should be rhythmical, but not strictly metrical,13 and
we need no further explanation why Seneca and his friends when
young had been tempted to set Fuscus' explicationes to music
(Suas. 2.10). But Fuscus' style in Suas. 3.1 somehow has none of
the grace of poetry; in particular the flow is broken by the
occasional intrusion of jerky successions of short and prosaic
words: et haec interdum anno lex est^1* ...; non ante finit quam
1ucem reddit.. .
It is interesting that the clausula with which the sentence
ends, adultero mare, is comparable, both in rhythm and in the con-
trived placement of a pyrrhic in the final position, with the
type, everritur caput, criticized in Contr. IX.2.24 as mollis and
not in Latro's manner.1^
Suas, 3.1 will also serve to illustrate Seneca's criticism of
Fuscus' diction: in descriptionibus extra legem omnibus verbis,
dummodo niterent, permissa libertas {Contr, II pr. 1). One cannot
fail to notice the highly poetic colour of the expression. Again
and again Fuscus chooses words and usages which in the Augustan
age - and it has to be remembered that he was already famous in
the elder Seneca's youth (Suas, 2.10) - one would not have ex-
pected to encounter in prose: aequora and serena; gravare transi-
tive, meaning 'to weigh down' (ThLL s.v. gravo IA lb); subsidere
with reference to an inanimate substance (cf. Lucr. V.493);
credere terrae (cf. Virg. Geor. 1.224); variare (cf. Ov. Fast.
III.449) and urgere (cf. Hor. Carm. 1.22.20) with reference to the
weather; asperare used of the effects of heat (cf. Val. Flacc. V.
368); splendere used literally to mean 'to shine'; sordidus used
metaphorically of the dark side of the moon; flatus in the sense
'winds' (ThLL s.v. flatus IB); 1 6 occupare referring to winds which
'invade' a year.
It comes as no surprise to learn that Fuscus was known to
make conscious efforts to imitate Virgil in his prose - efforts of
which Seneca is suitably scornful (Suas. 3.4-5 W ) :
in ea descriptione <quam> primam in hac suasoria posui Fuscus

Arellius Vergilii versus voluit imitari; valde autem longe petit
et paene repugnante materia, certe non desiderante, inseruit. ait
enim de luna: 'quae, sive plena lucis suae est splendensque
pariter assurgit in cornua, imbres prohibet, sive occupata nubilo
sordidiorem ostendit orbem suum, non ante finit quam lucem reddit.
at Vergilius haec quanto et simplicius et beatius dixit:
luna, revertentes cum primum colligit ignes,
si nigrum obscuro comprenderit aera cornu,
maximus agricolis pelagoque parabitur imber. (Georg. 1.427-9)
et rursus:
sin...
pura nee obtunsis per caelum cornibus ibit. (Ibid. 432-3)
We are told furthermore that one of his reasons for aspiring to
imitate Virgil was a desire to please Maecenas:
solebat autem Fuscus ex Vergilio multa trahere, ut Maecenati
imputaret; totiens enim pro beneficio narrabat in aliqua se Vergi-
liana descriptione placuisse; sicut in hac ipsa suasoria dixit:
'cur iste <in> interpretis ministerium placuit? cur hoc os deus
elegit? cur hoc sortitur potissimum pectus quod tanto numine
impleat?! aiebat se imitatum esse Vergilianum 'plena deo'. (Ibid.
5)
This link with Maecenas is most interesting, for it is in the
fragments of Maecenas1 prose, which Suetonius mentions as exemp-
lifying cacozelia in the time of Augustus, that we find writing
most closely comparable with the Fuscine manner. The younger
Seneca, who in Ep. mor. 114.4 describes Maecenas' style as eloqu-
entiam ebrii hominis involutam et errantem et licentiae plenam,
quotes several examples to illustrate his point:
quid turpius 'amne silvisque ripa comantibus'? vide ut 'alveum
lyntribus arent versoque vado remittant hortos'. quid? si quis
'feminae cinno crispat et labris columbatur incipitque suspirans,
ut cervice lassa fanantur nemoris tyranni.1 'inremediabilis
factio rimantur epulis lagonaque temptant domos et spe mortem
exigunt.' 'genium festo vix suo testem.' 'tenuisve cerei fila et
crepacem molam.' 'focum mater aut uxor investiunt.' (Ibid. 5 =
Maecenas fr. 11 Lunderstedt)17
Another example, ipsa enim altitudo attonat summa (fr. 10 Lunder-
stedt), is recorded in Ep. mor. 19.9. Seneca wonders why Maecenas
did not simply say of the mountain in question, 'attonita habet
summa1. Quintilian complains in particular about Maecenas' use of
unnatural hyperbaton and deliberate seeking out of 'lascivious'
rhythms:
quaedam vero transgressiones et longae sunt nimis...et interim
etiam compositione vitiosae, quae in hoc ipsum petuntur, ut
exultent atque lasciviant, quales illae Maecenatis: 'sole et

aurora rubent plurima1; Writer sacra movit aqua fraxinos'; 'ne
exequias quidem unus inter miserrimos viderem meas' (quod inter
haec pessimum est, quia in re tristi ludit compositio).
(IX.4.28 = fr. 16 Lunderstedt)
Arellius FuscusT explicatio in Suas. 3.1 is open to criti-
cisms similar to these levelled against Maecenas1 style: there
were simpler ways of expressing in Latin the concept, fit is
cloudy1, than Fuscus1 serena clauduntur, et omnis dies caelum
nubilo grravat;18 hyperbaton of the type represented by miseri
cremata agricolae lugent semina is abnormal in prose; Fuscus
shared with Maecenas a fondness for affectation in rhythm and un-
usual vocabulary. It is not surprising then that the wording of
parts of the elder Seneca's critique of Fuscus in Contr. II pr. 1,
explicatio...implicata... [sc. oratioj nimia licentia vaga et
effusa, should resemble the terms in which his son condemns
Maecenas1 style in Ep. mor. 114.4.
The eccentricities of Fuscus1 descriptive manner also have
close parallels in the fragments (FGrHist 142) of Hegesias, one of
the chief founding fathers of Asianism.19 Norden's criticisms of
this stylist in Die antike Kunstprosa2-® could be taken over almost
point by point and applied to the Fuscine extract in Suas. 3.1.
Hegesias1 compositio, samples of which are analysed by Norden,
features efforts at isocolon:
L uavCaL irpoanTataaaa TIOALS 13 syllables
Tpaya)6Cas eXeeuvoxepa yeyove (F 10) 13 syllables
There is an abundance of audacious metaphors, for example a funer-
ary image in F 16: at 6e yuvatxes yexrixîncrav ets Maxe6ovuav, xfiv
iioAuv da([»aaaL xuva Tponov. Hegesias had a fondness for making ab-
stract nouns the subjects of sentences, for example: n yev oftv
eAitls auxri auve6payev eus TO TOAUCXV (F 5.10f.); TOUS 6' aAAoug
opyn Tipoacpaxos euCynpaTO (F 5.18), a tendency which Fuscus was
to share, as we see not only in Suas. 3.1 where we read, et omnis
dies caelum nubilo gravat, but elsewhere, for example in Suas. 5.1:
suis ira ardet ignibus et in pacta non solvitur. The use of peri-
phrasis21 was recognized in antiquity as particularly character-
istic of the Asiatic style (Quint. XII.10.16: circumitu...
enuntiare). Thus, when Hegesias means ex TOU oupavou he writes
en TT\£ HOT' oupavov yepu6os (F 12). Similarly Fuscus, wishing to

describe sunny intervals and scattered showers, relates that quid-
quid asperatum aestu est, quidquid nimio diffluxit imbre, invicem
temperatur altero.
The justifications for giving the name Asianist to the de-
scriptive manner of Arellius Fuscus are many. The question we now
have to ask is whether there are any such grounds for attaching
the same label to the genus dicendi ardens et concitatum, the type
of style favoured pre-eminently by Latro. This style differed from
the Fuscine, as we have seen, in that explicationes were kept
brief (Contr. Ill pr. 7 ) , and a fiery intensity was consistently
aimed at; Latro, for instance, would never dwell on loci at any
great length (Contr. VII.7.10). Yet the fault of aridity was
avoided by the best exponents of this style. Latro even as a
schoolboy had seen scope for sententiae where his more cautious
master had not (Contr. I pr. 22), and the effect of a Latronian
declamation was one of considerable richness while, at least in
the elder Seneca's view, a certain subtilitas of structure always
provided a sound foundation for the elaborate whole (ibid. 20f.).
The most extended example we have of Latro's style is the
extract from his treatment of Contr. II.7, the theme of which was
as follows:
quidam, cum haberet formonsam uxorem, peregre profectus est. in
viciniam mulieris peregrinus mercator commigravit; ter illam
appellavit de stupro adiectis pretiis; negavit ilia, decessit
mercator, testamento heredem omnibus bonis reliquit formonsam et
adiecit elogium: 'pudicam repperi.' adit hereditatem. redit
maritus, accusat adulteri ex suspicione.
It seems that, contrary to his usual practice, the elder Seneca
may actually have recorded the whole of this declamation of
Latro's, but the last part of it (though perhaps not very much)
is now lost. The extant part of the declamation seems to consist
of the prooemium (§1); a vestigial narratio leading up to the
propositio,22 1
poteram ego salvo pudore meo nihil de hereditate
suspicari, in qua etiam nomen auctoris ab uxore doctus sum?1 (§2
W ) ; the rest (§3ff.) is given over to argumentation. The contro-
versia was of the type which Seneca called coniecturalis and, un-
less we have lost a great deal of Latro's argumentation, it would
appear that his divisio was of the uncomplicated sort customary in

the treatment of such themes.23 Throughout the extant argumen-
tation Latro seems to be treating just one quaestio coniecturalis
about the wife's character, utrum adultera sit an pudica. He
diversifies his arguments by means of excessus and indignationes.
At the opening, and just before the end of §1 we find care-
fully constructed periods involving more subordination of clauses
than was usual in the fiery and agitated style.
Quamquam eo prolapsi iam mores civitatis sunt ut nemo ad suspic-
anda adulteria nimium credulus possit videri, tamen ego adeo longe
ab eo vitio afui ut magis timeam ne quis in me aut nimiam patien-
tiam aut nimium stuporem arguat quod tarn seram querellam detuli.
illud, iudices, mihi tormentum est, quod notata iudicio vestro, ut
multiplicatam dotem perdat, plus tamen ex quaestu habitura est
quam quantum damnatae perdendum est...
But these are the only concessions Latro makes to the tradition of
prefatorial rotundity. For the rest, one sententia follows another
throughout §1 in brilliant succession: non accuso adulteram nisi
divitem factam; ex ea domo ream protraho in qua iam nihil meum
est, and so on. The prooemium ends with the indignant exclamation:
tantum in istam dives amator effudit ut post poenam quoque ex-
pediat fuisse adulteram.
Latro abstains from providing a full-length narratio for the
good reason that the husband, whose part he is taking, was absent
abroad at the time of his wife's alleged adultery. He contents
himself with drawing the jurors' attention to the rumours which
were supposedly circulating about his wife and her benefactor, be-
fore putting it to them, in the brief propositio quoted above,
that it was inevitable that he should be suspicious (§2).
He next (§3) begins the argumentation with an anthypophora
designed to anticipate the argument that a beautiful woman such as
his wife could not be blamed for not remaining chaste, and that it
was inevitable, at any rate, that she should attract lovers.
Latro's counter-argument is that any married woman, if she so
wishes, can thwart the hopes of potential admirers by dressing
sombrely and behaving with sufficient modesty:
matrona, quae <tuta> esse adversus sollicitatoris lasciviam volet,
prodeat in tantum ornata quantum ne inmunda sit; habeat comites
eius aetatis quae inpudicum, si nihil aliud, in verecundiam
annorum movere possit; ferat iacentis in terrain oculos; adversus

officiosum salutatorem inhutnana potius quatn inverectmda sit; etiam
in necessaria resalutandi vice multo rubore confusa <sit>. sic se
in verecundiam pigneret <ut> longe ante inpudicitiam suam ore quam
verbo neget. in has servandae integritatis custodias nulla libido
inrumpet.
Latro next (§4) launches into an indignatio in two parts, first an
imaginative description of women in seductive clothes behaving
shamelessly, and then a further imago21* in which the would-be
seducer's go-between is imagined (likely story!) to have been
flogged and tortured:
prodite mihi fronte in omne lenocinium composita, paulo obscurius
quam posita veste nudae, exquisito in omnes facetias sermone,
tantum non ultro blandientes ut quisquis viderit non metuat ac-
cedere: deinde miramini si, cum tot argumentis inpudicitiam prae-
scripseritis, cultu, incessu, sermone, facie, aliquis repertus est
qui incurrenti adulterae se non subduceret* internuntium, puto,
ilia sollicitatoris arripi et denudari iussit, flagella et verbera
et omne genus cruciatus poposcit, in plagas deterrimi mancipi vix
inbecillitatem muliebris manus continuit. nemo sic negantem iterum
rogat.
Note that both cases of sub oculos subiectio are kept brief. The
z:>
sententia which closes this secticm is an obvious applause-
winner .
In his next argument Latro makes all the capital he can out
of the words in the theme, ter illam appellavit de stupro adiectis
pretiis. From the fact that three approaches were made he deduces
(at the risk of being accused of contrarium)2^ that her refusals
took the form of silence: quod proximum est a promittente, rogata
stuprum tacet (§6). He assumes that the gifts were accepted and
magnifies them to vast proportions, skirting the difficulty of
describing them by means of the useful device of praeteritio:
inspicite adulterae censum; ex eo inpune sit quod adulter isti
dedit si est aliquid quod non dederit. quid singillatim omnia
percenseo? quid ego non emi in mundo tibi? <me> miserum: maritus
cum omni censu meo inter munera adulteri lateo. (Ibid.)
There follows in the next section (§7) another indignatio
opening with the heavily ironic exclamation, o nos nimium felici
et aureo, quod aiunt, saeculo natosi Latro sums up in a resound-
ing enthymema the fundamental contradiction in the wife's pleas as
viewed by the suspicious husband: sic etiam qui inpudicas quaerunt
pudicas honorant?
Most of the rest of the extract is given over to detailed

consideration of the will. First Latro reads out a passage from
the imagined will and reflects how much it sounds like the will of
a husband bequeathing his property to his wife; suppose he were to
believe that his wife was chaste and wished to add a final codicil
to his own will, he would have to look to the will of this adulter
for suitable wording (§7). Then (§8), after reading again from
the will he muses on the implausibility of the notion that the
family of his wifefs admirer included no chaste women and that no-
where on his travels had he met any women but prostitutes, that
everywhere he went he had been looking for a chaste woman to be
his heiress. After a plain statement of his allegation of adultery
and a few more pointed remarks on the decadence of the age, Latro
concludes his examination of the will with a figura testamenti , 2 7
inviting the jurors to consider, along with the will of his wife's
admirer, a proposal for his own will, the better to decide which
contains the more accurate description of his wife's character
(§9 W ) :
atque ego, si hunc morem scribendi recipitis, in conspectu vestro
ita scribam: 'uxor mea heres <ne> esto, quod peregrinante me
adamata est, quod heres ab adulescente alieno ac libidinoso
relicta est, quod tarn infamem hereditatem adit.' a duobus vos
testamentis in consilium mitto: utrum secuturi estis? quo ab
adultero absolvitur, an quo damnatur a viro?
The extract of Latro's declamation as preserved in the 'full
text' manuscripts ends with sententiae on the theme that true
chastity provokes nobody to comment upon it. It is quite possible
that remarks on this theme marked the end of Latro's argumentation,
but it does seem that we lack at least his epilogus, and it is
doubtful whether any sententiae from it have been preserved in the
Excerpta.28
Contr. II.7.1-9 illustrates well Latro's virtuoso command
over the various types of sententiae and figurae and his terseness
in the treatment of loci. Illustrations of his compendious grasp
of the acta of historical personages (Contr. I pr. 18) are lacking
in this controversia but are available elsewhere, for example in
Contr. X.I.8. More significant is the absence from Contr. II.7.1-9
of any protracted descriptiones; one searches in vain for these in
Latrofs other fragments too. Also lacking, surely, are instances

of the kinds of bad taste which Seneca castigates in the work of
less distinguished declaimers with such words as corruption and
ineptum.2^
That LatroTs declamatory style was not identical with that of
Fuscus would be apparent even if Seneca had not alluded specifi-
cally to the difference as he does in the words diversion — dicendi
genus in Contr. II.2.8.30 It does not seem an unreasonable hypo-
thesis that the elder Seneca regarded Latrofs style as distinct
from that of the Asiani. Note how in Contr. IX.1.1231 he takes
pains to point out that a sententia of Latrofs was quite different
from one, concerned with similar subject-matter, by the Asianus
Hybreas, whereas Fuscus in the same controversia had copied Adaeus,
another Asianus, exactly. On another occasion Seneca goes as far
as to say that it was out of the question that Latro could have
imitated any Greek:
Artemon dixit: TOL yev TWV d'AAwv eupâxa* TiAeu, yewpyet. Ta 6'
nyexepa dvontnpcr xpecpeb d'pa TOV oAoxAnpov. hanc sen tent iam Latro
Porcius virilius dixit, qui non potest <de> furto suspectus esse;
Graecos enim et contemnebat et ignorabat. (Contr. X.4.2Of.)
No other statement made by Seneca the Elder proclaims as
emphatically as this one, about Latrofs lack of indebtedness to
Greek models, the fact that a new age of Latin prose-writing had
dawned, very different from the one which had preceded it. Latro,
educated perhaps entirely in far-off Corduba,32 seems to have been
cut off in his formative years even more completely than his
school friend Seneca from the Greek-based literary culture which
Cicero and his familiares had taken for granted. Was it that
Latrofs family - much less well off, perhaps, than the Annaei
Senecae, for it was exceptional for an eques to become a rhetor-
ician in their time (Contr. II pr. 5) - did not possess any Greek-
speaking slaves, so that he was deprived of the chance of becoming
bilingual at home in the way that was customary (see Quint. I.I.
12-14) among the Roman upper classes?33 In the circumstances it
is not self-evident that there need be any justification for de-
scribing the style of Latro in terms taken over from the late
Republican controversy between Atticists and Asianists, implying
as they do imitation of Greek models, whether Attic or Asiatic.
That is not to deny, however, that Greek influences of either sort

could have reached Latro in his boyhood by an indirect route and
without his recognizing them for what they were.
Let us consider some possible formative influences on Latro
as a stylist. First there was the idiom of his native Spain:
Messala's jibe, sua lingua disertus est (Contr. II.4.8), serves to
show that this was a factor to be reckoned with. Secondly there
was the poetry which he studied with the grammaticus which presum-
ably, in those days before the poems of Virgil and Horace had
become the standard school texts, comprised the Latin drama and
epics of the Republic. Insofar as these poems were modelled on
Greek originals, a certain amount of Hellenic literary technique
would have been absorbed by Latro through the study of them. The
terse rhetoric of the speeches contained in them should not be
ignored as a possible source of inspiration for Latro's senten-
tiousness. The acquaintance with drama of a schoolboy from
Corduba need not, by the way, have been confined to his grammati-
cal studies: a performance of a praetexta at Gades during the
Civil Wars is attested by Pollio in Ad fam. X.32.3.
After his years with the grammaticus Latro moved on to study
rhetoric in the school of Marullus. The question we must now ask
is whether the rhetorical manner he learnt to cultivate there was
(even if he and his friends did not realize it) in any real sense
Asianic. In other words, was the sententious manner customary in
the treatment of controversiae and suasoriae in the time of Seneca
the Elder a direct descendant of a style originating in the Greek
schools of Asia Minor some time in the previous three centuries?
Several factors make this hypothesis seem quite attractive.
First, the popularity of sententiousness seems to have been uni-
versal among the declaimers of the time: Cestius Pius from Smyrna^
declaimed in a style not fundamentally different from that of
Latro the Corduban, and it is clear from Seneca's quotations that
brief sententiae were sought after by Greek-speaking, as well as
Latin, rhetoricians. Secondly, it is a well known fact that the
ponderous conservatism of the Hellenistic rhetorical school tra-
dition was such that even arguments by a critic of Cicero's
stature in favour of change made no lasting impact upon it. For
instance, Cicero's desire for the thesis to be given an important

place in the rhetorical curriculum35 does not seem to have had the
slightest effect on the subsequent development of Greco-Roman
education. Also, as we have seen, the Hegesian brand of Asianism,
despite everything which its detractors said against it, survived
the centuries, presumably fostered by a continuous tradition in
certain Hellenistic schools, to re-emerge in the Latin explica-
tiones of Arellius Fuscus.36 Is it not highly plausible, then,
that the more arid style of Fuscus - not so different from that of
Latro, and presumably quite similar to that of Marullus, whom
Seneca describes as hominem satis aridum (Contr. I pr. 22) - had
its origins in the schools of Asia Minor? We have seen that the
Hellenistic age saw a movement towards brevity in the treatment of
judicial declamations as a whole, in that it became customary to
declaim only one speech on a given theme rather than four;37
furthermore, brevity in the composition of all enthymemata was
already encouraged by the author of the Rhetorica ad Alexandrum.^^
Yet there are objections which may fairly be levelled against
the supposition that a prototype of the genus dicendi ardens et
concitatum had been developed in the schools of Hellenistic Asia
several centuries earlier. A passage in Cicero's Brutus might be
cited in support of the notion that the 'pointed' style of the
early Roman Empire had a precedent in an early Asiatic type of
rhetoric:
genera autem Asiaticae dictionis duo sunt: unum sententiosum et
argutum, sententiis non tam gravibus et severis quam concinnis et
venustis, qualis in historia Timaeus, in dicendo autem pueris
nobis Hierocles Alabandeus, magis etiam Menecles frater eius fuit,
quorum utriusque orationes sunt in primis ut Asiatico in genere
laudabiles. aliud autem genus est non tam sententiis frequentatum
quam verbis volucre atque incitatum, quale est nunc Asia tota, nee
flumine solum orationis, sed etiam exornato et faceto genere
verborum, in quo fuit Aeschylus Cnidius et meus aequalis Milesius
Aeschines. in his erat admirabilis orationis cursus, ornata
sententiarum concinnitas non erat. (Brut. 95.325)
It is to the first of Cicero's two types of Asianism that we must
direct our attention. Fortunately, though the speeches of Hiero-
cles and Menecles of Alabanda are hopelessly lost, there do sur-
vive a number of fragments of the Histories of Timaeus, which
according to Cicero were representative of this style.39 But one
searches in vain in these fragments (FGrHist 566) for antici-

pations of the style we find in the declamations of Porcius Latro.
Polybius, who criticizes Timaeus for imitating the techniques of
school rhetoric when putting speeches into the mouths of histori-
cal characters, actually quotes from one such speech, in which
Timoleon exhorts the Greeks to enter battle with a numerically
superior Carthaginian army (XII.26a = Timaeus F 31). Here if any-
where, seeing that the speech was of the suasoria type, one would
expect to find any anticipations there were in Timaeus1 style of
Latronian sententiousness. Timoleon is made to encourage the
Greeks not to consider the numerical strength but the unmanliness
of the enemy:
Mai, yap TTJS At3uns diraans auvex&S
x otxouuevris xal
p , l
6'uws ev xaug Ttapotuuabs, oxav Ttepl epnytag p
(3ouA(A)ue§a Ttounaab, Aeyeuv nyas "epnyoxepa xfjs AL3uriSrt9 oux e m TT\\>
ep TIP lav cpepovxas T 6 V Aoyov, dAA' enl xnv dvav6pLav xwv xaxoux-
^0 xa§oAou 6e, cpnû, T I S av (po3ri§eln xous d'v6pas, OL'TLVES
(puaews T O U T O T O L S dv^pcouoLS 6e6a)KuLag L 6 L O V Tiapd TO. AotTid xaiv
Aeyw 6e xds x e ^ P a S , xauxas nap' 6'Aov x6v 3ôv evxos xuiv
X e x ° V T £ S dupaxxous Ttepucpepouat; x6 6e yeyuaxov o n xaL uno
xous x^TcovtaKotSs cpnaC, uepbCwyaxa cpopouatv, uva yn6' oxav dno-
•ddvwauv ev xaus ydxats cpavepoL yevcovxau xous uitevavxtots. . .
Nothing in this speech, surely, or in any other of Timaeus' frag-
ments, bears any noticeable resemblance to the genus dicendi
ardens et concitatum. Rather the naive techniques of persuasion
in this fragment seem to hark back to the speeches in Herodotus,
notably the speech of Sardanis in 1.71. Only the very feeble
argument about the proverbial expression, epnyoxepa xfis Auftuns,
marks it out as the product of a more pedantic age than that of
Herodotus; compare the point made by Hegesias in FGrHist 142 F 14:
6ebv6v xriv x^Pa^ d'cntopov eu'vau xfiv TOUS Ziiapxous xexouaav.
From Cicero's remarks in Brut. 95.325 it appears that the
Timaean variety of Asiatic style, characterized as sententiosum et
argutum, had been popular with certain orators when he was a boy,
but that since then it had been superseded throughout Asia by a
more voluble style, non tarn sententiis frequentatum guam verbis
volucre atque incitatum, quale est nunc Asia tota... It may well
be that this second style had already largely ousted the first by
the time when Cicero came to study with Asian and other Greek
rhetoricians. This supposition would account for the fact that it
is volubility rather than sententiousness that we think of as the

dominant characteristic of his early oratory. It would accord
well also with the fact that, as far as we can judge from the
fragmentary evidence available, a flowing periodic style seems to
have been the one generally favoured by Roman orators throughout
the first half of the first century B.C.41
To sum up, the more sententious of the two Asiatic styles
distinguished by Cicero went out of fashion at the beginning of
the first century B.C. Traces of it remained in the later
speeches of Hortensius, though these were not much admired (Brut.
95.327), and perhaps it was never ousted from the Asian schools
completely, but continued to be favoured in a few circles, together
with the Hegesian elements which may have been inseparably linked
to this genus dicendi. But this possibility is of no relevance to
our present enquiry, as there is no evidence that the Timaean type
of Asianism in any way resembled the kind of style which Latro
learnt to cultivate in the school of Marullus.
Why there should have been a general move towards greater
terseness in rhetorical style in the latter half of the first cen-
tury B.C. is mysterious. M. Aper in Tacitus' Dialogus attributes
the change to the need for speakers to appeal to an increasingly
sophisticated audience and, with the advent of the Empire, to take
account of certain changes in the nature of jurisdiction:
at hercule pervulgatis iam omnibus, cum vix in cortina adsistat
quin elementis studiorum, etsi non instructus, at certe inbutus
sit, novis et exquisitis eloquentiae itineribus opus est, per quae
orator fastidium aurium effugiat, utique apud eos iudices qui vi
et potestate, non iure et legibus cognoscunt, nee accipiunt
tempora sed constituunt, nee expectandum habent oratorem dum illi
libeat de ipso negotio dicere, sed saepe ultro admonent atque alio
transgredientem revocant et festinare se testantur. (19.5)
There may well be much truth in this analysis, though it does not
explain everything.42
It might be suggested that the newly introduced practice of
declaiming controversiae and suasoriae in the Latin schools could
have been a major factor in the rise of the fashion for senten-
tiousness: the stock characters in controversia themes in particu-
lar are placed in the sorts of situation which might seem ex-
pressly designed to prompt the composition of pointed ironic
sententiae; besides, the subject-matter to be treated by declaimers

of these themes was much sparser than the circumstantial data
available to orators in even the simplest of court cases, and so
they could not indulge in otiose repetition without inviting ad-
verse criticism from their hearers (Contr. IX.5.15).
However, it would be wrong to assume that controversia themes
were bound to be treated in a pithy style: the arguments in Anti-
phonfs Tetralogies are expressed in language more diffuse than
Latronian Latin; the Declamationes maiores of ps.-Quintilian,
though far from deficient in brief and striking sententiae, con-
tain a rather larger proportion of complex periods in their compo-
sition than does, for example, Latrofs version of Contr. II.7. We
may take as an illustration a passage from Decl. mai. XVII which
is a treatment of the theme which appears in Seneca's collection
as Contr. VII.3. It concerns a young man whose father has tried
unsuccessfully to disown him three times and who now, having been
discovered mixing poison, is accused of intended parricide.
o pertinacissimum accusatorum genus victi parentes! dum auctori-
tatem nominis vestri fortius imperiosis adseritis adfectibus et,
ne pudorem paenitentiam fateamini, contumacia vindicatis errorem,
calamitatibus meis accessit, ut ter absolverer. namque ut erat in
supervacuo odio mei senex prima lite1*^ deprensus, ferre non
potuit quod reddebar invito, et quia a iudicibus non inpetraverat
ut abdicaret, apud se tenuit, ne desineret hoc velle. credidit
tandem*41* aliquid profecturum querelarum errore repetito et
speravit iuxta contentionem suam, ut lassesceret pro me iusta
miseratio. quid facerem igitur? quo verterem iam fatigatam
innocentiam? nee exire me decebat ex domo, ne viderer quidquid
vos non credideratis, agnoscere, nee expectare poteram, cum mihi
rursus aliam seriem malorum minaretur, quod me iam coeperat pater
contentione qua vobis irascebatur, odisse. (Decl. mai. XVII.3)
There is no reason to suppose, then, that controversia themes
could not have been treated in the orotund style prevalent towards
the end of the Republic. Still less is it self-evident that the
style used in suasoriae had to be pithy: there is no reason why a
declaimer, if he cared to do sufficient research into the mytho-
logical or historical background of his chosen subject, should not
have treated deliberative themes with great elaboration, and in a
correspondingly copious style.
More to the point, perhaps, we have noted in passing that the
speeches in epic and drama which Latro must have studied with the
grammaticus could have been a formative influence on his style.

It is worth considering whether the style of rhetoric to be found
in poetry might not have appealed very widely to the educated
youth of Latrofs generation throughout the Greek and Roman world,
impatient, for whatever reason, with the long-windedness which had
been acceptable when Cicero had started out on his oratorical
career. The conventions of epic and drama had always demanded a
terser style of rhetoric than had been customary in the law-courts
or assemblies of either Athens or Rome. We have seen how the
examples of enthymemata taken from drama by Cicero in his Topica
seem stylistically akin to Silver Latin rhetoric. The same may be
said of Cicero's own verse translations of tragedy included in the
Tusculan disputations. Take, for example, part of his version
of a speech of Hercules in torment from the Greek of Sophocles'
Trachiniae:
non Graia vis, non barbara ulla immanitas,
non saeva terris gens relegata ultimis,
quas peragrans undique omnem ecferitatem expuli:
sed feminae vir, feminea interemor manu.
o nate, vere hoc nomen usurpa patri
neve occidentem matris superet caritas.
hue adripe ad me manibus abstractam piis.
iam cernam mene an illam potiorem putes.
perge, aude, nate, illacrima patris pestibus,
miserere! gentes nostras flebunt miserias.
heu.' virginalem me ore ploratum edere,
quern vidit nemo ulli ingemescentem malo!
accede, nate, adsiste, miserandum aspice
eviscerati corpus laceratum patris!^^
Here surely - and observe that this is a rendering of an impeccably
Attic original - is rhetoric not so very far removed from the
melodramatic declamatory style later to be parodied by Petronius.
But it is not only in the poetry of the Republic and its
classical Greek models that we find precedents for the senten-
tiousness favoured by Latro and his contemporaries. Cicero's late
speeches contain passages which were surely a major source of
inspiration to the early Imperial declaimers. Consider the per-
oration of the second Philippic, in particular the following
resounding words:
defendi rem publicam adulescens, non deseram senex: contempsi
Catilinae gladios, non pertimescam tuos. quin etiam corpus
libenter obtulerim, si repraesentari morte mea libertas civitatis

potest, ut aliquando dolor populi Romani pariat quod iam diu
parturit! etenim si abhinc annos prope viginti hoc ipso in templo
negavi posse mortem immaturam esse consulari, quanto verius nunc
negabo seni? (Phil. II.118f.)
What better example could there be of the genus dicendi ardens et
concitatum at its most elevated? For once Cicero confutes the
famous criticism of 'Longinus' that whereas his was a fire which
burnt widely and steadily, that of Demosthenes was the fire of
thunderbolts (Subl. 12.4).
Cicero, of course, recommended Demosthenes above all other
Attic orators as the model for his contemporaries to imitate (Brut.
84.288ff.). They perhaps hardly needed the advice. There is a
statement in Opt. gen. or.1*^ 2.6 to the effect that nemo est
orator qui Demostheni <se> similem nolit esse, and some people at
least seem to have considered that Calvus had succeeded in attain-
ing to the Demosthenic ideal, for Senecafs excursus on Calvus in
Contr. VII.4.6-8 clearly derived at least in part from the criti-
cism of people who held this view. Now, in this passage Seneca
is quite probably quoting Latro.*47 And, significantly, the
stylistic virtues supposedly shared by Calvus and Demosthenes
which are singled out for praise, vigour and agitation (compositio
quoque eius ad exemplum Demosthenis viget: nihil in ilia placidum,
nihil lene est, omnia excitata et fluctuantia), are very much the
same as those which he reckons in Contr. Ill pr. 7 to be the mark
of a good declaimer.
It looks, then, as though the rhetoricians of the early
Empire who favoured consistent fiery agitation saw themselves as
the successors of the late Republican admirers of Demosthenes'
style and preserved at least a few memories of Atticist, non-
Ciceronian, theory on the subject. But, as we have seen,48 the
information about Calvus in Contr. VII.4.6-8 may well have been
taken from a collection of exempla, and among the men who exer-
cised most influence in promoting the agitated style were some,
including Latro, who were ignorant of Greek. It is not surprising
that the attempts of such narrowly educated westerners to emulate
the Demosthenic ideal produced results which do not strike the
modern reader forcibly with their resemblance to the thunderbolts
of the great Athenian.

Certain tendencies in Republican rhetoric, the rhetoric of
the stage as well as that of the forum, can be seen, then, to
foreshadow the extreme sententiousness universally favoured in the
early Imperial schools. But let us admit that the difference be-
tween late Cicero and the pointed style evidently sought after to
some extent by every schoolman from Latro of Corduba to Cestius of
Smyrna is too great to be accounted for as the outcome of gradual
literary evolution. (The declaimers from the eastern provinces
had no such excuse as Latro not to have approached closer than
they did to the style of Demosthenes.) Let us also admit that the
difference is not wholly accounted for in the sociological specu-
lations of Aper in the Dialogus.^ Possibly one of the main
reasons for it lies in certain emphases which had become standard
in the curricula of rhetorical schools throughout the Greco-Roman
world, notably the restriction of the philosophical and historical
education of future orators to the rote-learning of ready-made
loci and exempla, and the limiting of their training in inventio
to practice in the contrivance of divisiones and colores, and
their training in elocutio to practice in the composition of
sententiae, this last element in the curriculum being, if any safe
deduction may be drawn from the elder Seneca's low estimate of his
sons1 tastes expressed in Contr. I pr. 22, the only aspect of
declamation which interested the non-analytical mind of the aver-
age student of rhetoric.
Where and when this drastic narrowing of the rhetorical
curriculum originated is unknown. The encouragement of wide
learning in history and philosophy had probably never been con-
sidered by any but the most exceptional of Hellenistic rhetoricians
as part of their duties.50 However, the restriction of training
in inventio and elocutio to instruction in the contriving of sen-
tentiae, divisiones and colores may well have been quite a recent
development, owing its universal acceptance throughout the Empire
of the elder Seneca's day to the widespread use of one or two
influential text-books, now lost. Note that the term color, as
used by Imperial rhetoricians, does not seem to have been of any
great antiquity.51 It is indeed by no means out of the question
that the inventor of the system, and hence one of the main pro-
moters of sententiousness, was a Latinus rhetor of the late
Republic, or a Greek teacher of the same period, prompted to
simplify the rhetorical curriculum by Roman parents who wanted
rapid results if they were to supply him with pupils.
It may still be, though, that the first impulse towards agi-
tated sententiousness was purely Greek, and had a root cause quite
different from curricular reorganization - disgust at the redun-
dant style fostered in certain earlier Hellenistic schools, maybe,
or even a hankering after Laconic brevity.52 But any Greeks re-
sponsible for innovation in this direction are unlikely to have
been Asianists: volubility seems still to have been the main
characteristic of Asian orators and their Roman imitators in the
40s B.C.5^ It may be relevant to note in this connection that a
tantalizingly fragmentary passage of Philodemus (Rhet. II.218f.
Sudhaus) seems to discuss Athenian rejection of periodic structure
in relation to school rhetoric. Seneca the Elder certainly does
not seem to have considered the style he most admired to be
Asiatic: he finds nothing inconsistent about including in Contr.
VII.4.6-8 expressions of approval for Demosthenes, Calvus and
Porcius Latro, and, when he refers to Asiani, he gives us no
warrant for supposing that he considered himself, or the declaimers
he most admired, to be among their ranks.
Of the two declamatory genera dicendi so far discussed, the
Fuscine and the Latronian, it was probably Latrofs which had the
greater influence on the development of early Silver Latin. At any
1
rate, neither of Fuscus two most distinguished pupils, Ovid and
Fabianus, were whole-hearted in their admiration for their master.
In Contr. II.2.8f. we are given an intriguing account of
Ovid's relations with Fuscus and Latro. Unfortunately, at an im-
portant juncture in this passage the received text seems suspect:
hanc controversial!! memini ab Ovidio Nasone declamari apud rhetorem
Arellium Fuscum, cuius auditor fuit; nam Latronis admirator erat,
cum diversum sequeretur dicendi genus, habebat ille comptum et
decens et amabile ingenium. oratio eius iam turn nihil aliud
poterat videri quam solutum carmen. adeo autem studiose Latronem
audit ut multas illius sententias in versus suos transtulerit. in
armorum iudicio dixerat Latro: mittamus arma in hostis et petamus.
Naso dixit:
arma viri fortis medios mittantur in hostis;

inde iubete peti. (Met. XIII.121f.)
et alium ex ilia suasoria sensum aeque a Latrone mutuatus est.
memini Latronem in praefatione quadam dicere quod scholastici
quasi carmen didicerunt: non vides ut immota fax torpeat, ut ex-
agitata reddat ignes? mollit viros otium, ferrum situ carpitur et
rubiginem ducit, desidia dedocet. Naso dixit:
vidi ego iactatas mota face crescere flammas
et rursus nullo concutiente mori. (Am, I.2.11f.)
tune autem cum studeret habebatur bonus declamator. hanc certe
controversiam ante Arellium Fuscum declamavit, ut mihi videbatur,
<illo> longe ingeniosius, excepto eo quod sine certo ordine per
locos discurrebat.
H
The problem lies in the meaning of nam in the opening sentence.
A clear contrast needs to be pointed between the fact that Ovid
was a pupil of Fuscus and the fact that he was an admirer of
Latro, for, given that the two rhetoricians favoured different
genera dicendi, the latter fact does not explain the former. Such
crtr*
was the reasoning which must have led Gertz^ to propose the con-
jecture, tamen Latronis admirator erat. But an alternative way of
emending the passage might be by transposition, as follows:
hanc controversiam memini ab Ovidio Nasone declamari apud rhetorem
Arellium Fuscum, cuius auditor fuit, cum diversum sequeretur
dicendi genus; nam Latronis admirator erat. 56 *
f
I remember this controversia being declaimed by Ovidius Naso at
the school of the rhetor Arellius Fuscus, whose pupil he was (at
that time - therefore perf.), although he favoured a different
style of speaking (from his master's); for he was (always -
therefore imperf.) an admirer of Latro.1
This transposition, if correct, would be of some importance, for
we would have to abandon any notion that the dicendi genus which
Ovid preferred as a student of rhetoric was that of his master
Fuscus, and take it that Latro was the more important influence
on him.
Various arguments might be used against this suggestion and
in favour of preferring an interpretation which made Ovid an ad-
herent of the Fuscine dicendi genus. First, it is said of Ovid's
prose later in Contr. II.2.8, oratio eius iam turn nihil aliud
poterat videri quam solutum carmen. Might the same not truly be
said of Fuscus' style in his explicationesi It might indeed be
suggested that the words, habebat ille...transtulerit, look like
a commentary on an assertion that Ovid was an admirer of Latro in
spite of favouring Fuscus' style. Again, Ovid is said not to have
organized his argumentation systematically in Contr. II.2; in this

respect he was no true follower of Latro. He preferred suasoriae
to controversiae (Contr, II.2.12); Fuscus is said to have declaimed
suasoriae 'libentissime' (Suas. 4.5). 5 7
Yet there is much to be said in favour of the supposition
that Latro had more influence on Ovid's style than Fuscus. Wander-
ing and over-elaborate descriptive digressions are distinctly not a
typical feature of Ovid's poetry. Would one not expect to find
numerous examples of them, especially in the Heroides where Ovid
comes closest in his poetry to indulging his taste for suasoriae,
if he had been a true devotee of the Fuscine manner? Given that
he was a poet, it would be ludicrous to view his habit of imitating
Virgilian phrases as evidence of a predilection for the prose-
poetry of contemporary Asianists. He had in any case better taste
than they, as may readily be seen from a comparison between two
phrases concerning a river and its banks, one by Maecenas and one
from Ovid's Amores. Seneca the Younger in Ep. mor. 114.5 attrib-
utes to Maecenas the words, amne silvisque ripa comantibus (fr. 11
Lunderstedt), and Ovid opens Am. III.6 with the line, amnis
harundinibus limosas obsite ripas (cf. Virg. Eel. 7.12f., hie
virides tenera praetexit harundine ripas\Mincius). Note how
Ovid's invocatory description of the river contains none of the
ambiguity of case endings which renders Maecenas' words obscure;
note too that it is confined to one line of poetry. Ovid's view
of the best length for descriptions is summed up in his criticism,
reported in Contr. VII.1.27, of some lines by Varro of Atax (fr. 8
Morel):
desierant latrare cane's urbesque silebant;
omnia noctis erant placida composta quiete.
solebat Ovidius de his versibus dicere potuisse fieri longe
meliores si secundi versus ultima pars abscideretur et sic
desineret:
omnia noctis erant.
And it is a view which we find reflected in his poetic practice;
as T. F. Higham has observed, 'Nature's operations on the grandest
scale are more than once depicted by Ovid with equal concise-
ness.' 58 It seems relevant to recall that Latro preferred loci of
all kinds to be kept short (Contr. VII.7.10) and how the fiery and
agitated style as described by Seneca contained non lentas nee

vacuas explicationes, sed plus sensuum quam verborum habentes
(Contr. Ill pr. 7). In the important matter of description, then,
Ovid in his poetry is more Latronian than Fuscine, and there is no
reason to suppose it was otherwise in his prose. That there is no
extended descriptio in the sample of his declamation in Contr, II.
2.9ff. would prove nothing if considered in isolation, but we hap-
pen to be informed in Contr. II.2.12 that Ovid verbis minime
licenter usus est nisi in carminibus... In his prose, therefore,
he can hardly have been a very faithful follower of his master
Fuscus, whose licentia of style in parts of his declamations is
described by Seneca in Contr. II pr. 1 and elsewhere.
Having dismissed the hypothesis that poetic diction in the
Fuscine manner was a feature of OvidTs prose, one next has to con-
sider whether there are rhythmical qualities in it which mark it
out as nihil aliud...quam solutum carmen, and whether in this re-
spect Ovid has to be regarded as a follower of Fuscus rather than
Latro. Here we move on to dangerous ground. For one thing, it is
not only resemblances to dactylic verse in Ovidfs prose which
could be of significance in this enquiry for, although we only
know Ovid as a composer of hexameters and pentameters, he was
celebrated in antiquity also for his tragedy, the Medea, which was
presumably written in a diversity of metres but principally in
iambics. Further, in the analysis of prose-rhythm the simple
truth has always to be remembered that all Latin sentences consist
of collections of long and short syllables:59 hence it is imposs-
ible to write prose in which sequences of iambs, trochees, cretics,
dactyls, spondees and so on, will never occur.60 So the mere fact
that it is possible to analyse snatches of the Ovidian declamatory
extract (Contr. II.2.9-11) as quasi-metrical, for example: quid
est quod ill(am) ab indulgentia su(a) avocet (§9) as iambic; fer
socer felicitatem and exciderat lurantibus (§11) as trochaic; sic
\J — _ y_, _ _ v^» — — \J ^ . . . — — —
amant. pauca nosti, pater, crimma (§10) as cretic; tu nobis

religiosum (§9) as dactylic; ecc(e) obiurgator nostn qu(am)
— — v^v^— w — ^ — ^
effrenato (ibid.) as heavily spondaic; queritur pencul(um) eius
— — — <<J <U — ^J — \J
(§10) as anacreontic; and si solus moriturus essem (§11) as hippo-

nactean, need not by itself account for the elder Seneca's
statement that Ovid's youthful prose style resembled solutum

carmen, unless it proves to be the case that there is a much more
pervasive metrical element in his prose than in that of other de-
claimers, in particular that of Latro.
This does not seem to be so, as can be demonstrated (to take
a passage at random) from Contr, II.7.3, where Latro admits such
quasi-metrical sequences as: tempus est ludices (cretic); sollici-
taretur potuit and quam faciem quamvis (dactylic); habeat comites
(anapaestic); nequ(e) est quod dicat (iambic); -sum salutator(em)
mhumana potius qu(am) mverecunda <sit> (trochaic). It is true
that similar rhythmic sequences can be extracted from samples of
the 'dry' style of Arellius Fuscus as well as from his descrip-
tions, for example from Contr. II.1.4: luvenum sine patre (dac-
tylic); navigabo, militabo and pauperes sumus qu(i) habemus
(trochaic). The fact remains that the study of sentence rhythm
gives us no grounds for supposing that a declaimer whose prose
could be described as resembling solutum carmen could not have
been a follower of Latro, rather than Fuscus, in his dicendi
genus.6l
Even more dangerous territory is entered if it is suggested
that it was not so much the actual rhythms of Ovid's declamatory
prose that justified the description solutum carmen, as rhythms
implied in his various sententiae, that the disiecti membra poetae
could be rearranged and with a certain amount of patching up could
be turned into acceptable verses in the Ovidian manner. It can be
pointed out, certainly, that it takes little remoulding to convert
quid ad patrem pertinet quod amantes iurant? (Contr. II.2.10) into
pertinet ad patrem quod vos iuratis, amantes?^2 But clearly such
observations are unlikely to help us to decide which dicendi genus
Ovid preferred, for surely a sufficiently determined versifier
could achieve similar results by tampering with many a declamatory
sententia, Fuscine or Latronian.
Perhaps in any case, despite the terms in which Ovid himself
refers to his early attempts at writing (Trist. IV.10.25f.),
sponte sua carmen numeros veniebat ad aptos,
et quod temptabam dicere versus erat,
it is wrong to suppose that Seneca had rhythmical qualities
primarily in mind when he wrote, oratio eius iam turn nihil aliud
poterat videri quam solutum carmen. Maybe it was rather in Ovid's
taste for amatoriis...sententiis (cf. Contr. III.7 extra), cer-
tainly apparent in Contr. II.2.9-11, that Seneca chiefly saw the
characteristics of the poet foreshadowed in his declamatory prose.
Affinities may certainly be found between the extract from Ovid's
declamation and the elegiac tradition. It is not just that the
theme of Contr. II.2 posed the question of how much weight should
be attached to lovers1 vows, a topic also treated in elegy, for
example .in Tib. El. I.4.21ff. Most strikingly a sentence in §9,
in which the word pater is used during an address to a father-in-
law, parce pater: non peieravimus, is surely an imitation of
Tibullus1 protestation of his innocence before Jupiter in El. I.
3.51f.:
parce, Pater, timidum non me peiuria terrent,
non dicta in sanctos impia verba deos.
Tibullus1 first book appeared in 27 or 26 B.C., when Ovid was
about sixteen years old. Here is evidence that the younger man,
presumably before briefly entering public life at about the age of
nineteen (Trist. IV.10.29ff.), was given to imitating at least one
poet in his declamations, thus producing something which could
fairly be called solutum carmen.
To return to the hypothesis that Ovid preferred the dicendi
genus of Latro to that of Fuscus, there is nothing in Seneca's
criticism that makes it impossible to accept. Obviously in some
respects Ovid differed in his tastes from both his official and
his unofficial mentor. The repetitiousness which was to mar some
of his verse had a counterpart in the speeches and declamations of
Votienus Montanus (Contr. IX.5.15ff.), but not, as far as one can
tell, in the style of either Fuscus or Latro. His manner of
arranging arguments was probably in general less rigorous than
that of Latro, but this was a respect in which he differed from
Fuscus too, as we learn from Contr. II.2.9. In other ways he re-
sembled both Fuscus and Latro. His liking for suasoriae was cer-
tainly a taste which he shared with Fuscus, but we must not be
misled by the fact that Fuscus happens to figure more prominently
than Latro in the extant surveys of suasoriae into thinking that
Latro despised this type of declamation: remember that in Contr.

II.4.8 he is reported to have declaimed a certain suasoria about
Theodotus for three days on end. Latro was just as inclined as
Fuscus to declaim controversiae ethicae, the other type of decla-
mation favoured by Ovid (see Contr. II.2.12): like Fuscus and Ovid
he declaimed Contr. II.2 (see §1); as we have seen,63 Seneca gives
us a very long extract from Latro's treatment of Contr. II.7,
which is certainly another example of this type of controversia.
Incidentally, there are no grounds for being surprised that Ovid,
if he was an admirer of Latro's genus dicendi, should have remained
a pupil of Fuscus, seeing that Latro refused to listen to his
pupils1 declamations or correct their work (Contr. IX.2.23).
One more point of interpretation needs to be settled before
we leave the question of Ovid's preference in rhetorical style.
If we accept the transposition of the words nam Latronis admirator
erat to the end of the first sentence of Contr. II.2.8, we have to
abandon any notion that the words in the third sentence likening
Ovid's oratio to solutum carmen allude to a debt which he owed to
the Fuscine style. Consequently autem in the fourth sentence can
no longer be taken as pointing a contrast between adherence to
Fuscus' dicendi genus and admiration for Latro. If it does point
an antithesis it is between the fact that Ovid's poetic gifts were
apparent in his declamations and the fact that his declamatory
tastes left their mark on his poetry. But perhaps this is to read
a more rigorous logic into Seneca's reminiscences than was inten-
ded. Autem in his usage sometimes serves as a very vague connec-
tive with no particular antithetical force.64
That Fabianus, Fuscus' other most distinguished pupil, re-
acted against his master's style is clearly stated in Contr. II
pr. Iff.
Cum repeterem, quos umquam bene declamantes audissem, occurrit
mihi inter alios Fabianus philosophus qui adulescens admodum
tantae opinionis in declamando, quantae postea in disputando fuit.
exercebatur apud Arellium Fuscum, cuius genus dicendi imitatus
plus deinde laboris impendit, ut similitudinem eius effugeret,
quam impenderat, ut exprimeret.
(There follows the description of Fuscus' style quoted above,
pp. 77f.)
ab hac cito se Fabianus separavit et luxuriam quidem cum voluit

abiecit, obscuritatem non potuit evadere; haec ilium usque in
philosophiam prosecuta est. saepe minus quam audienti satis est
eloquitur, et in summa eius ac simplicissima facultate dicendi
antiquorum tamen vitiorum remanent vestigia, quaedam tarn subito
desinunt, ut non brevia sint, sed abrupta. dicebat autem Fabianus
fere dulces sententias et, quotiens inciderat aliqua materia, quae
convicium saeculi reciperet, inspirabat magno magis quam acri
animo. deerat illi oratorium robur et ille pugnatorius mucro,
splendor vero velut voluntarius non elaboratae orationi aderat.
vultus dicentis lenis et pro tranquillitate morum remissus; vocis
nulla contentio, nulla corporis adseveratio, cum verba velut in-
iussa fluerent. iam videlicet conpositus et pacatus animus; cum
veros compressisset adfectus et iram doloremque procul expulisset,
parum bene imitari poterat quae effugerat. suasoriis aptior erat;
locorum habitus fluminumque decursus nemo descripsit abundantius.
numquam inopia verbi substitit, sed velocissimo ac facillimo cursu
omnes res beata circumfluebat oratio.
Not surprisingly we can find among the fragments of Fabianus1
declamations some which typify the standard sententious style of
the early Imperial schools, an arid version of which was to be
found in Fuscus' prooemia, narrationes and argumenta, others in
the Fuscine sing-song descriptive manner, and still others in a
rather different style, which we may suspect to belong to the
period after his conversion to philosophy.
Contr. II.2.4 will serve to show that Fabianus was fully
capable of hammering out clever sententiae and contriving the
sorts of figures of speech and thought which most delighted the
scholastic audiences:65
'non possum1 inquit 'relinquere virum.' quicquam non potes, quae
mori potes? paene qui falsae mortis nuntium miserat verae recepit.
vir, dum nimis amat uxorem, paene causa periculi fuit; uxor, dum
nimis amat virum, paene causa luctus fuit; pater, dum nimis amat
filiam, abdicat. servate, <di>, totam domum amore mutuo labor-
antem. fmoriarT inquit: hoc patri minaris, viro promittis. potes
sine viro pati; peregrinationem eius tulisti. facilius potest
carere eo, cui spiritum debet, quam eo, cui inpendit.66
His description of Ocean in Suas. 1.4 shows him to have been a
competent imitator of the more eccentric side of Fuscus1 art:
quid? ista toto pelago infusa caligo navigantem tibi videtur ad-
mittere, quae prospicientem quoque excludit? non haec India est
nee ferarum terribilis ille conventus. inmanes propone beluas,
aspice quibus procellis fluctibusque saeviat, quas ad litora undas
agat. tantus ventorum concursus, tanta convulsi funditus maris
insania est; nulla praesens navigantibus statio est, nihil
salutare, nihil notum; rudis et inperfecta natura penitus recessit.
ista maria ne illi quidem petierunt qui fugiebant Alexandrum.
sacrum quiddam terris natura circumfudit Oceanutn. illi qui iam

siderum collegerunt meatus et annuas hiemis atque aestatis vices
ad certam legem redegerunt, quibus nulla pars ignota mundi est, de
Oceano tamen dubitant, utrumne terras velut vinculum cludat an in
suum colligatur orbem et in hos per quos navigatur sinus quasi
spiramenta quaedam magnitudinis <suae> exaestuet; ignem post se,
cuius augmentum ipse sit, habeat an spiritum. quid agitis, con-
militones? domitoremne generis humani, magnum Alexandrum, eo
dimittitis quod adhuc quid sit disputatur? memento, Alexander:
matrem in orbe victo adhuc magis quam pacato relinquis.
Notice in particular the highly poetic quality of, for example,
the words in the opening sentence, ista toto pelago infusa caligo,
and his usages of statio9 meatus and spiramenta; also the elabor-
ate construction of the sentence (illi qui,.,an spiritum) in which
he describes the bafflement of learned men as to the nature and
extent of Oceanus; note too the occasional repetition of rhythmic
patterns at the beginning of phrases:
ista toto pelago...
navigahtem tibi...
— w — — —
tanta convulsi...
nulla praesens navigantibus...67
One can also appreciate that if he continued to use such language
as we find in the passage, an in suum colligatur orbem...habeat an
spiritum, when he came to deal with questions of natural science
in his philosophical works, he might justly have been accused of
obscurity.
For a passage which may well belong to the period after
Fabianus1 break with Fuscus we may look to the long extract, Contr.
II.1.10-13. The subject-matter, convicium saeculi , is of a type
which, so we have been told in Contr. II pr. 2, the future philo-
sopher found particularly inspiring. The quest of riches is seen
as the cause of civil wars:
ecce instructi exercitus saepe civium cognatorumque conserturi
[praelium] manus constiterunt et colles equis utrimque complentur
et subinde omnis regio trucidatorum corporibus consternitur; ilia
turn in multitudine cadaverum vel spoliantium sic quaesierit ali-
quis: quae causa hominem adversus hominem in facinus coegit? - nam
neque feris inter se bella sunt nee, si forent, eadem hominem
deceant, placidum proximumque divino genus; quae tanta vos pestis,
cum una stirps idemque sanguis sitis, quaeve furiae in mutuum
sanguinem egere? quod tantum malum <huic> uni generi vel fato vel
forte iniunctum? an, ut convivia populis instruantur et tecta
auro fulgeant, parricidium tanti fuit? magna enim vero et lauta
sunt propter quae mensam et lacunaria sua <nocentes> potius quam
lucem innocentes intueri maluerint. an, ne quid ventri negetur

libidinique, orbis servitium expetendum est? in quid tandem sic
pestiferae istae divitiae expetuntur si ne in hoc quidem, ut
liberis relinquantur? (Contr. Il.l.lOf. W)
As an instance of the general corruption brought about by riches,
modern extravagance in building is berated:
quid tandem est quod non divitiae corruperint? primum, si inde
incipere velis, aedes ipsas, quas in tantum extruxere ut, cum
domus ad usum ac munimentum paratae sint, nunc periculo, non
praesidio <sint>: tanta altitudo aedificiorum est tantaeque viarum
angustiae ut neque adversus ignem praesidium nee ex ruinis ullam
[yillam] in partem effugium sit. (§11)
After enlarging on the topics of the use of expensive building
materials and the consequent anxiety of the rich about possible
accidental fires, arson and looting (all themes, like so much else
in Fabianus1 declamation, paralleled in Senecan philosophy and
Imperial satire), 68 he goes on to reflect that the rich act as if
they had never seen the true beauties of nature:
o paupertas, quam ignotum bonum es! quin etiam montes silvasque
in domibus marcidis et in umbra fumoque viridia aut maria amnesque
imitantur. vix possum credere quemquam eorum vidisse silvas
virentisque gramine campos, quos rapidus amnis ex praecipitio vel,
cum per plana infusus est, placidus interfluit; non maria umquam
ex colle vidisse lenta, aut hiberna cum ventis penitus agitata
sunt: quis enim tarn pravis oblectare animum imitamentis possit si
vera cognoverit? videlicet <haec illis placent> ut infantibus
quae tangi comprehendique manibus aut sinu possunt; nam magna non
capit exigua mens. ex hoc litoribus quoque moles iniungunt con-
gestisque in alto terris exaggerant sinus; alii fossis inducunt
mare: adeo nullis gaudere veris sciunt, sed adversum naturam
alieno loco aut terra aut mare mentita aegris oblectamenta sunt.
et miraris <si> fastidio rerum naturae laborantibus iam ne liberi
quidem nisi alieni placent? (§13)
Fabianus1 fluency, and his particular aptitude for topographical
description are well illustrated in this last extract.
It is easy to imagine these sentiments being expressed by a
young man imbued with a newly found moral earnestness, but none-
theless one must be careful not to overstate any claim that
Fabianus in this passage departs from the style of Fuscine rhet-
oric. If faced, for instance, with the first sentence of Contr.
II.1.10 out of context, who would dare to state categorically that
it could not have been by Fuscus? But examination of Fabianus1
clausulae in the extract Contr. II.1.10-13 (25 in all, discounting
those where textual restoration has been necessary) perhaps
confirms one's subjective impression that it must have been com-

posed after the break with Fuscus.6^ Of the 16 clausula types
found only 5 occur more than once. Three of these certainly are
types favoured by Fuscus: —v->—« (Fuscus: 18%) and - ^ — ^ (Fuscus:
12%) occur 3 times each; — ^-^^ (Fuscus: 13%) occurs twice. But
Fabianus departs from Fuscine practice in tolerating the hexameter
ending -^^-^ twice70 in this small sample (Fuscus 4%), and, most
strikingly, in using the rarely favoured rhythm ^ ^^ 4 times71
1
(Fuscus 1%). One of Fuscus favourite clausulae —v-»—^>±> (15%) is
used only once in this passage by Fabianus. How far he is depart-
ing from the usage of Fuscus and other rhetoricians in sentence
rhythm as a whole and in diction might be an interesting subject
for further investigation, but until much more research has been
done on the styles of the declaimers it would be dangerous to
assert that there is anything remarkable about, for example, the
frequency of elision (or hiatus) in the sentence, tantaâltitudo
aedificiorum est, tantaeque viarumângustiae ut neque_adversus
ignem praesidium nee ex ruinis ullam in partem effugium sit, or
the presence of such long and unsubtle alliterative sequences as,
civium cognatorumque conserturi manus constiterunt et colles equis
virisque complentur et subinde omnis regio trucidatorum corporibus
consternitur. The presence of such series of long syllables as
aedes ipsas quas xn tant(um) extruxere, might similarly strike an
unwary analyst new to the field as an indication that Fabianus was
reacting against conventional rhetorical technique, but, as it
turns out, parallels for them can be found in the usage of both
Fuscus (e.g. Suas. 3.1: haec mterd(um) anno lex est) and Latro
(e.g. Contr. II.7.3: irritamentum spem corrumpendi; si tant(um) m
formonsa sperari posset, quantum...). Such are the hazards of
close stylistic analysis when applied to the elder Seneca's decla-
matory collection.
Fabianus was to be one of the philosophers most admired by
the younger Seneca. In De brev. vit. 10.1 he is described as non
ex his cathedrariis philosophis sed ex veris et antiquis and in
Ep. mor. 40.12 as vir egregius et vita et scientia et, quod post
ista est, eloquentia. Ep. mor. 100 consists of a long defence of
Fabianus against various strictures which Lucilius, Seneca's
correspondent, is said to have made on his style. In a number of

respects the younger Seneca shares the opinions expressed by his
father in Contr. II pr. Iff. 72 All through the letter he is in-
sistent that it is inappropriate to require a*philosopher to aim
at the rhetorical standards which one expects from other types of
prose writer. Lucilius had criticized Fabianus' compositio ad-
versely (Ep. mor. 100.1), but Seneca has nothing but praise for
the untroubled flow of his language, so suitable for the conveying
of moral precepts,
Fabianus mihi non effundere videtur orationem sed fundere; adeo
larga est et sine perturbatione, non sine cursu tamen veniens.
illud plane fatetur et praefert, non esse tractatam nee diu tortam.
sed ita, ut vis, esse credamus: mores ille, non verba composuit et
animis scripsit ista, non auribus. (§2)
his disregard for modern fashions in the choice and arrangement of
words,
Fabianus non erat neglegens in oratione sed securus. itaque nihil
invenies sordidum: electa verba sunt, non captata, nee huius
saeculi more contra naturam suam posita et inversa, splendida
tamen quamvis sumantur e medio. (§5)
and his rejection of sententious brevity in favour of greater
breadth of expression,
sensus honestos et magnificos habes, non coactos in sententiam sed
latius dictos. (ibid.)
To read these criticisms reinforces one's suspicions that the
declamatory excerpt, Contr. II.1.10-13, is not far removed styli-
stically from Fabianus1 mature philosophical writings: certainly
there is a scarcity of ingeniously contrived sententiae in it, and
the diction and style, though not without parallels in the poetry
of his day,7^ do not offend the ear by any outlandish artifici-
ality; one can imagine Fabianus1 audience already sensing that, as
Seneca was to put it in Ep. mor. 100.11, ad profectum omnia tendunt,
ad bonam mentem: non quaeritur plausus.
The younger Seneca's appreciation for Fabianus' unosten-
tatious mode of expression did not, of course, deter him from
using in his own philosophical works a style rather more in accord-
ance with the taste of his rhetorically educated contemporaries.
In particular his fondness for sententiousness is well known.
However, like Fabianus, he seems to have rejected the excesses of
the Fuscine explication Contrast, for example, with Fuscus1 di-

gression on weather signs in Suas, 3.1, Seneca's expression when
he describes the human mind contemplating the heavenly bodies
(Quaest. nat. I pr. 12),
secure spectat occasus siderum atque ortus et tarn diversas con-
cordantium vias; observat ubi quaeque Stella primum terris lumen
ostendat, ubi columen eius summumque cursus sit, quousque
descendat; curiosus spectator excutit singula et quaerit. quidni
quaerat? scit ilia ad se pertinere,
or windless cloudy weather (ibid, V.3.1),
hoc falsum esse vel ex eo colligas licet quod tune minime ventus
est cum aer nubilo gravis est; atqui tune plurima corpora se in
angustum contulerunt et inde est spissarum nubium gravitas.
In both passages the sentence structure is clear and simple, ob-
scurities such as he criticizes elsewhere in Maecenas' descriptions
are absent, and there is none of the licentia of vocabulary for
which Fuscus, in common with Maecenas, was notorious. Seneca's
usages of concordare9 columen and in angustum all have precedents
in extant Republican prose,7*4 and it may well be accidental (due,
perhaps, merely to the fact that Cicero preferred ethics to phys-
ics) that we appear not to have a non-poetic precedent for the use
of spissus with reference to the density of matter. Seneca's
philosophical style might perhaps be described as a novel blend
between the discursiveness and verba non captata of Fabianus and
the genus dicendi ardens et concitatum favoured in the schools of
rhetoric.
Once we have recognized that Seneca the Younger, though a
modernist, shows himself, both by his admiration for Fabianus and
by his rejection of the style of descriptive writing character-
istic of Fuscus, an opponent of the phenomenon which his age re-
garded as Asianism, it becomes easier to understand how it was
that he not only felt capable of preaching to his readers in Ep.
mor. 114 on the diverse types of corrupt style, but could also
join in a debate on the causes of the corruption which was in all
probability75 an echo of earlier speculations of Greek theorists
who deplored the decline of oratory in the Hellenistic world.
To return to the early Imperial declaimers, none of them
seemed to the elder Seneca's sensitive ears to have sounded
exactly like Fuscus or exactly like Latro. Nearest to Fuscus,
perhaps, of the declaimers criticized in his prefaces,76 was Musa,

a rhetorician whose floruit must have been much later than that of
Fuscus, seeing that the elder Seneca's sons were able to attend
his declamations (Contr. X pr. 9 ) . That Musa is recorded to have
used such far-fetched metaphors as caelo repluunt, odoratos imbres,
caelatas silvas and nemora surgentia77 (ibid.) seems clear enough
evidence as to the type of description which he favoured. Some
other egregious examples of his descriptive technique are quoted
in Suas. 1.13 where, after quoting a sententia (now lost) by one
Menestratus, Seneca remarks:
efficit haec sententia ut ignoscam Musae, qui dixit ipsis Charybdi
et Scylla maius portentum: 'Charybdis ipsius maris naufragium1 et,
ne in una re semel insaniret: fquid ibi potest esse salvi ubi
ipsum mare perit?'
Elsewhere we are told of another declaimer, Vibius Gallus, who was
unique in making positive efforts to attain insania and in par-
ticular was notorious for his manner of introducing descriptions:
Gallus Vibius fuit tam magnae olim eloquentiae quam postea
insaniae, cui hoc accidisse uni scio, ut in insaniam non casu in-
cideret, sed iudicio perveniret; nam dum insanos imitatur, dum
lenocinium ingeni furorem putat, quod cotidie simulabat ad verum
redegit...solebat autem sic ad locos pervenire, ut amorem de-
scripturus paene cantantis modo diceret: 'amorem describere volo1
sic tamquam fbacchari volo.1 deinde describebat et <ut> totiens
coepturus repetebat: 'amorem describere volo.1 (Contr. II.1.25f.)
Vibius Gallus evidently had some redeeming features, for Seneca
says of his description of riches in Contr. II.1, multa facunde
explicuit, corruptius quam Fabianus sed dulcius (§26), but for
sheer bombast it would be difficult to beat a portentous utterance,
ad mortis auctorem per vulnera sua ruunt, in his description of
concitatissuma...in morte rabies in Contr. IX.6.2. Clearly he and
Musa were kindred spirits, and there may have been affinities be-
tween his manner of delivery and that of Fuscus.78
Latro's closest rival, in the elder Seneca's view (Contr. X
pr. 13), and probably the declaimer who most resembled him, was
Iunius Gallic There must surely have been some sort of tribute
to him in one of the prefaces now lost, even though Gallio was
still alive when Seneca was writing and was evidently a great
friend of his three sons, 79 who therefore did not need to be told
about his gifts as an orator and declaimer. It is most unfortunate
that only a few of Seneca's passing remarks about his declamations

are extant.
Gallio's style was to be adversely criticized, though dis-
tinguished from that of Maecenas, by Vipstanus Messala in Tacitus'
Dialogus,
ceterum si omisso optimo illo et perfectissimo genere eloquentiae
eligenda sit forma dicendi, malim hercule C. Gracchi impetum aut
L. Crassi maturitatem quam calamistros Maecenatis aut tinnitus
Gallionis... (Dial. 26.1)
and almost any extract from Gallio's declamations will serve to
show why a critic surveying the whole development of Roman elo-
quence should have dismissed his style as tinnitus. Take, for
example, Contr. II.3.6:80
'rogo' inquit. nunc? hie? sic? si volebas rogare, admovisses
propinquos, amicos, maiorum imagines, lacrimas, repetitos alte
gemitus. testor deos, sic rogaturus fui puellae patrem. 'quando'
inquit <'misereberis?' cum> vultum in supplicis habitum summiseris,
cum dixeris: 'paenitet quod rapui, quod te priorem non rogavi',
cum dixeris te dementem fuisse, deliberabo cum amicis, deliberabo
cum propinquis, deliberabo cum tua matre. me miserum, quam paene
promisi! <dura>, anime, dura; here fortior eras, et multum habeo
quod deliberem: diversi me adfectus distringunt, inter reum et
patrem distrahor; hinc iniuria est, hinc natura.
Fortunately, we are not left completely in the dark as to why
Seneca the Elder, by contrast, had a high opinion of Gallio. We
are told that he displayed remarkably good taste in the use of
colloquial language (idiotismos): iam adulescentulus cum declamaret
apte et convenienter et decenter hoc genere utebatur; quod eo magis
mirabar quia tenera aetas refugit omne non tantum quod sordidum
sed quod sordido simile est (Contr. VII pr. 6 ) . In this respect
Seneca evidently regarded him as superior to Albucius, another
member of the primum tetradeum (Contr. VII pr. 3; cf. X pr. 13).
That Gallio was appreciative of the power of others to sway the
emotions of their hearers is evident from his praise of Cassius
Severus in Contr. Ill pr. 2, and presumably he tried to emulate
what he admired. Certainly we see from his testamenti figura in
Contr. IX.3.14 that he was a virtuoso in the treatment of figures,
than which, in Quintilian's view (IX.1.21), adfectus nihil magis
ducit. On certain occasions we find him contriving sententiae of
the types most admired by his friend: in Contr. 1.1.25 he effects
a transition from prooemium to narratio by means of a sententia
and, just before, Seneca has praised Hermagoras1 use of this very
device; in Suas. 5.8 Seneca singles out one of GallioTs sententiae
for some of the highest praise he ever gives, hoc loco disertissi-
mam sententiam dixit, <dignam> quae in oratione vel in his tori a
ponatur: diutius illi perire possunt quam nos vincere. To us, as
to Gerth, who in his Pauly-Wissowa article comments fsehr
sophistisch',81 it may not seem worthy of such high commendation,
but it should be borne in mind that this sententia was uttered in
the course of a very unusual treatment of the suasoria in question.
Gallio was the only declaimer whom Seneca remembered taking the
part of a representative of the peace party at Athens when faced
with renewed threats of invasion from Xerxes (Suas. 5.8), and
Seneca is elsewhere found to be appreciative when a declaimer pre-
sents vivum consilium rather than merely reeling off stock argu-
ments (Suas. 6.11). When choosing lines of argument Gallio was
unafraid of quaestiones durae (Contr. IX.1.10), a characteristic
he shared with Latro (Contr. X pr. 15). In the matter of colores
Gallio agreed with Seneca that excuses about dreams were a poor
substitute for more rational pleas (Contr. II.1.33). To some
limited extent, then, we are able to form an impression of GallioTs
talent, and to appreciate how near it came to the elder Senecafs
declamatory ideal.
All the other declaimers fell short of it, but for a wide
variety of reasons. Cassius Severus, though he possessed all the
qualities which ought to have made him a good declaimer (Contr.
Ill pr. 7 ) , was one of those men who, though highly successful in
the forum, were deserted by their talent when they turned to
declamation: non tantum infra se, cum declamaret, sed infra multos
erat (Contr. Ill pr. 7 cf. ibid. 1). His declamations struck
Seneca as uneven: declamationes eius inaequales erant, sed ea,
quae eminebant, in quacumque declamatione posuisses, inaequalem
earn fecissent (ibid. 18); other notable characteristics of his
style included compositio aspera et quae vitaret conclusionem,
sententiae vivae (ibid.). Seneca apologizes for not being able to
give an adequately representative sample of Severus1 declamatory
manner: iniquom tamen erit ex his eum aestimari, quae statim sub-
texam; non enim haec ille optime dixit, sed haec ego optime teneo
(ibid.). Now that the full text of Contr. III-VI and VIII is lost
the sampling available to us is all the more inadequate: only two
extracts of any length and a few isolated sententiae are extant.
The report of his harangue against the pretensions of the schools
(Contr. Ill pr. 8-18) may perhaps be looked to for evidence of the
gift for forceful impromptu speaking to which Seneca refers in
Contr. Ill pr. 6, ex tempore coactus dicere infinito se ante-
cedebat. numquam non utilius erat illi deprehendi quam prae-
parari. But it is impossible for us now to assess the full range
of Severus1 style.
The longest of the declamatory extracts reveals an unexpected
gift on Cassius Severus1 part for gruesome description:
hinc caeci innitentes baculis vagantur, hinc trunca bracchia
circumferunt, huic convulsi pedum articuli sunt et <ex>torti tali,
huic elisa crura, illius inviolatis pedibus cruribusque femina
contudit: aliter in quemque saeviens ossifragus iste alterius
bracchia amputat, alterius enervat, alium distorquet, alium de-
lumbat, alterius diminutas scapulas in deforme tuber extundit et
risum <e> crudelitate captat. produc, agedum, familiam semivivam,
tremulam, debilem, caecam, mancam, famelicam; ostende nobis
captivos tuos. volo mehercules nosse ilium specum tuum, illam
humanarum calamitatium officinam, illud infantium spoliarium. sua
quoique calamitas tamquam ars adsignatur: huic recta membra sunt,
et, si nemo moratur, proceritas emicabit: ita frangantur, ut humo
se adlevare non possit, sed pedum crurumque resolutis vertebris
reptet. huic <eximii oculi sunt>: extirpentur radicitus. huic
[non] speciosa facies est: potest formonsus mendicus esse; reliqua
membra inutilia sint, ut Fortunae iniquitas in beneficia sua
saevientis magis hominum animos percellat. <sic> sine satellitibus
tyrannus calamitates humanas dispensat. (Contr. X.4.2 M)
A number of the words used by Severus in this passage seem to be
either neologisms or new to prose: extundere, spoliarium, reptare,
extirpare, speciosus. Other expressions, however, contribute
something of an archaic flavour to the passage: famelicus, a
favourite word of the ante-classical comic poets, appears not to
have been used at all by Cicero; the long string of adjectives in
the sentence, produc, agedum, familiam semivivam, tremulam,
debilem, caecam, mancam, famelicam, may be compared with Plaut.
Merc. 630, ad mandata claudus, caecus, mutus, mancus, debilis.
Interestingly, it emerges from Contr. VII.3.8f. that Severus had
an interest in Republican drama - he refers to Pomponius and
Laberius as early Latin punsters - and that he did not eschew even
the mimes of Publilius as a source of expressions worth imitating.

In the other main extant fragment from his declamations (Contr.
IX.2.12) we find him echoing another sort of archaic expression,
the language of Roman law:
ne de servo quidem aut captivo omni loco aut omni genere aut per
quos libebit aut cum libebit supplicium sumi fas est, adhibeturque
ad ea magistratus ob custodiam, non ob laetitiam.
In both these declamatory extracts we detect something of the
censoriousness which, according to Seneca, characterized his ora-
tory so long as he refrained from joking: quamdiu citra iocos se
continebat, censoria oratio erat (Contr. Ill pr. 4 ) . Less clear
is what Seneca meant when he described his compositio as aspera et
quae vitaret conclusionem (ibid. 18). This last word, Thomas'
emendation of the MSS reading compositionem, probably refers to
artistic clausula-rhythm82 rather than to 'periodic structure1, ^
the avoidance of which was so general in his time that Seneca would
hardly have thought it worthy of comment. Too few of Severus1
declamatory sententiae are extant to allow systematic investi-
gation of his rhythmical practice. One which may be singled out
as jerkily constructed is the anthypophora in Contr. VII.3.10,
quare ergo nunc non morens? dicet aliquis.8l+ For almost in-
variably declaimers, when imagining possible retorts of an adver-
sary, would say dicet aliquis or the like first, before voicing
the hypothetical objection.
Severus1 harangue against the rhetoricians in Contr. Ill pr.
8ff. is illustrative of his wit, for which however there can have
been little scope in his declamations; funniest of all is his
story about Cestius Pius in §16:
memini me intrare scholam eius cum recitaturus esset in Milonem;
Cestius ex consuetudine sua miratus dicebat: si Thraex essem,
Fusius essem; si pantomimus essem, Bathyllus essem, si equus,
Melissio. non continui bilem et exclamavi: si cloaca esses,
maxima esses, risus omnium ingens; scholastici intueri me, quis
essem qui tarn crassas cervices haberem. Cestius Ciceroni re-
sponsurus mini quod responderet non invenit, sed negavit se
executurum nisi exissem de domo. ego negavi me de balneo publico
exiturum nisi lotus essem.
Other passages in the harangue serve to show how unsubtle he might
be when speaking impromptu - hence perhaps his effectiveness on
such occasions.85 We have to take it on trust from the elder
Seneca that Severus normally composed speeches with the utmost

care, always expunging anything redundant or irrelevant to his
case: nemo minus passus est aliquid in actione sua otiosi esse;
nulla pars erat, quae non sua virtute staret, nihil in quo auditor
sine damno aliud ageret (Contr. Ill pr. 2). No one would have
guessed this from the style of the opening of the explanation he
gives why he was not so successful in declamation as in oratory,
where he hammers home the point that specialization was a common
phenomenon with quite unnecessary verbosity, showing by means of
example after example that poets, historians, philosophers, strong
men, dogs, horses, actors and gladiators as well as orators tended
to be successful in only one, sometimes very limited, sphere of
activity (Contr. Ill pr. 8-11).
Cassius Severus was later to be considered a very important
figure in the transition from the old to the new style of Latin
oratory at the beginning of the Empire - the first of the moderns
in the view of Aper in Tacitus1 Dialogus,
...quern primum adfirmant flexisse ab ilia vetere atque derecta
dicendi via, non infirmitate ingenii nee inscitia litterarum
transtulisse se ad illud dicendi genus contendo, sed iudicio et
intellectu. vidit namque, ut paulo ante dicebam, cum condicione
temporum et diversitate aurium formam quoque ac speciem orationis
esse mutandam. (19.If.)
Vipstanus Messala later in the dialogue does not dispute Severus1
primacy in the new style, though he takes a less glowing view of
his achievement:
equidem non negaverim Cassium Severum, quem solum Aper noster
nominare ausus est, si iis comparetur qui postea fuerunt, posse
oratorem vocari, quamquam in magna parte librorum suorum plus
bilis habeat quam sanguinis. primus enim contempto ordine rerum,
omissa modestia ac pudore verborum, ipsis etiam quibus utitur
amis incompositus et studio feriendi plerumque deiectus, non
pugnat sed rixatur. ceterum, ut dixi, sequentibus comparatus et
varietate eruditionis et lepore urbanitatis et ipsarum virium
robore multum ceteros superat. (26.4)
The elder Seneca nowhere in his extant books singles out
Severus as the originator of the new oratory, though it is appar-
ent from his descriptions of Scaurus and Labienus in the tenth
preface86 that he recognized a distinction between an old and a
new style of oratory. Seneca only gives us Severus1 own opinion
as to who were the leading orators of his day, and he had the
modesty to exclude himself, naming instead Pollionem Asinium et

Messalam Corvinum et Passienum, qui nunc primo loco stat (Contr.
Ill pr. 14). Something of the bile and brawling criticized by
Vipstanus Messala in the Dialogus can be detected in Severus1 on-
slaught on the rhetoricians, and there is perhaps a good example
of contempt for any predictable ordo rerum in the way he moves
from human to animal examples of specialization and back again in
Contr. Ill pr. 8-11. Seneca does anticipate the critics of the
Dialogus in thinking highly of Severus as an orator, considering
him to be underrated by people who had not heard and seen him in
action:
non est quod ilium ex his quae edidit aestimetis; sunt quidem et
hie quibus eloquentia eius <agnoscatur; tamen auditus> longe maior
erat quam lectus. non hoc ea portione illi accidit qua omnibus
fere, quibus maiori commendationi est audiri quam legi, sed in
illo longe maius discrimen est. (Contr. Ill pr. 3 W)
It is a great pity that we have lost so many of the fragments of
his declamations which Seneca must originally have provided, even
though they would have given us no adequate notion of the strengths
of his oratorical manner.
Cestius Pius, singled out by Cassius Severus as typifying the
lunacy of the rhetorical schools, was an interesting character.
A. D. Leeman has described him as Venomous 1 , 87 but this seems
rather a hard verdict on a man who, though he compromised his own
talent by his willingness to play to the gallery, and may have dis-
couraged certain of the younger generation from the pursuit of any
high literary ideals they might otherwise have had, can hardly be
reckoned more noxious than the average poseur.
Cestius was a Greek (from Smyrna according to St Jerome), 88
but overturned the normal order of things by choosing to work as
a Latin rhetorician (see Contr. IX.3.13). The most notable fact
about Cestius as a stylist would seem to be that his command of
Latin was somewhat limited and that for this reason, if for no
other, his explicationes were not of the same order as those of
Arellius Fuscus:
soleo dicere vobis Cestium Latinorum verborum inopia <ut> hominem
Graecum laborasse, sensibus abundasse; itaque, quotiens latius
aliquid describere ausus est, totiens substitit, utique cum se ad
imitationem magni alicuius ingeni derexerat, sicut in hac
controversia fecit, nam in narratione, cum fratrem traditum sibi

describeret, placuit sibi in hac explicatione una et infelici; nox
erat concubia, et omnia, iudices, canentia <sub> sideribus muta
erant. (Contr. VII.1.27)
Despite his inadequacies, however, his self-esteem knew no bounds,
and one of his chief poses was to belittle Cicero, encouraging his
pupils to learn by heart his own declamations, but only to read
those speeches of Cicero to which he, Cestius, had composed
answers (Contr. Ill pr. 15). He was once pointed out to Cicero's
drunken son at a banquet as Cestius, qui patrem tuum negabat
litteras scisse, and duly received a horse-whipping (Suas. 7.14).
For all this, Cestius was an influential teacher, preferred,
along with Latro, by the younger generation, to the leading ora-
tors of his time (Contr. Ill pr. 14f.). He numbered among his
pupils, besides Argentarius, so thorough-going a plagiarist of his
master's sententiae that Cestius called him Cesti simius or 6
TttSnxos you (Contr. IX.3.12), Alfius Flavus, a remarkable child
prodigy (Contr. 1.1.22), and Surdinus, a gifted youth, a quo
Graecae fabulae eleganter in sermonem Latinum conversae sunt (Suas.
7.12). Mature men also sometimes declaimed in his school, for
example the senator Aietius Pastor (Contr. 1.3.11).
It appears from the quite numerous criticisms by Cestius
quoted by Seneca the Elder that he was by no means lacking in
natural good taste and discrimination. His remarks in Suas. 1.
89
5ff., on the need to vary one's tactics in suasoriae to suit the
character of the person one is addressing, are perfectly sensible.
Elsewhere we often find Cestius intolerant of lapses of taste or
inconsistencies, whether committed by his pupils or by rival rhet-
oricians: unfortunate doubles entendres (Contr. 1.3.10; Suas. 7.
12); breaches of propriety (Contr. 1.5.3; 6.11); statements damag-
ing to the declaimer's own case (Contr. 1.3.9; II.3.22; 6.8);
incredibility (Contr. 1.3.11); the discussion of scarcely relevant
philosophical quibbles (Contr. VII pr. 8f.); naive use of the de-
vice of echo (Contr. VII.7.19); fanciful metaphorical expression
(Contr. VII pr. 9 ) ; even incorrect Latin usage (Contr. 1.3.9) -
all these faults call forth scorn from Cestius. Yet it was a
curious fact that Cestius did not eliminate from his own work the
faults which he criticized in others1:
tantus autem error est in omnibus quidem studiis, <sed> maxime in

eloquentia, cuius regula incerta est, lit vitia quidam sua et
intellegant et ament. Cestius pueriliter se dixisse intellegebat:
'mater, quid est venenum?'; deridebat enim Murredium qui hanc
sententiam imitatus in epilogo, cum adloqui coepisset puellam et
diceret: fcompone te in periclitantium habitum, profunde lacrimas,
manus ad genua dimitte, rea es f , fecerat respondentem puellam:
pater, quid est rea? et aiebat Cestius: quod si ad deridendum me
dixit, homo venustus fuit, et ego nunc scio me ineptam sententiam
dicere; multa autem dico non quia mihi placent sed quia audientibus
placitura sunt. (Contr. IX.6.11f.)
Asinius Pollio, introduced by Seneca in the fourth preface,
was obviously a much greater man than Cestius, and differed from
him enormously in his attitude towards declamation, regarding it,
so Seneca believed (Contr, IV pr. 2 ) , as an exercise too far
beneath his intellectual dignity for him to perform in public.
Yet he resembled Cestius in not demanding of himself, when de-
claiming, the high standards of which he was capable, and which
he demanded of others. In his declamations he adopted a style
different from that of his speeches, and was less rigorously self-
critical than usual:
floridior90 erat aliquanto in declamando quam in agendo: illud
strictum eius et asperum et nimis iratum ingenio suo iudicium adeo
cessabat, ut in multis illi venia opus esset, quae ab ipso vix in-
petrabatur. (Contr. IV pr. 3)
Insufficient fragments are extant from which to form an ad-
equate impression of Pollio's oratory, but there is certainly a
marked contrast between the style of the quotation from his Pro
Lamia in Suas. 6.15 (= Pollio fr. 19, ORF3 519):
itaque numquam per Ciceronem mora fuit quin eiuraret suas [esse]
quas cupidissime effuderat orationes in Antonium; multiplicesque
numero et accuratius scriptas illis contrarias edere ac vel ipse
palam pro contione recitare pollicebatur
and the declamatory extract in Contr. VII.1.4 W:
aequas mihi praebete aures: dabo vobis etiam damnatum absolvendum.
f
vivit* inquit 'frater1; non credo. 'servavit1 inquit 'me';
fecisti ut crederem. haec est summa rerum gestarum: in ea domo in
qua facile parricidium creditum est, ego fratrem occidere non potui,
frater patrem. fquid mihi cum ista tabula? semel mori volo.1
Note particularly the extraordinary number of disyllabic words in
the latter example and the resulting jerkiness of effect.^*
Another declamatory extract begins with a sentence cast in a form
which seems to have been a favourite among the least austere of
the declaimers, inter nuptiales fescenninas in crucem generi

nostri iocabantur (Contr, VII.6.12), with which we may compare,
for example, Contr, 1.1.6: inter funestas acies armatae manus in
foedus porriguntur (Fuscus); Contr, IX.2.24: inter temulentas
ebriorum reliquias humanum everritur caput (Florus). Without
Seneca's testimony one would never have believed it possible that
Asinius Pollio, with his reputation for rugged compositio (Sen.
Ep. mor. 100.7) and archaizing tendencies (Quint. X.I.113), could
have declaimed in this manner. It is not as if Pollio's strictum
...et asperum...iudicium (Contr. IV pr. 3) was not provoked by the
efforts of other declaimers. He was hard on Latro, who prided
himself on the quasi-forensic simplicity of his divisiones, for
leaving out, when he declaimed Contr, II.3, a quaestio on which
the whole case would have hinged if it were being heard in the
courts (Contr, II.3.13). He condemned the famous explicationes in
Fuscus1 suasoriae as not being aimed at persuasion (Suas, 2.10).
His archaizing tastes are reflected in his approval of Vibius
Rufus? sententia in Contr. IX.2.25, at nunc a praetore lege actum
est ad lucernam,^*- and also perhaps in his revival of the ancient
practice of alternating declamations for the prosecution and the
defence when teaching his grandson Marcellus Aeserninus (Contr.
IV pr. 3 ) .
Q. Haterius, introduced along with Pollio in the fourth pref-
ace, fell short of Seneca's ideal in declamation mainly because of
his excessive verbosity (Contr, IV pr. 7ff.), a trait which was
evidently also a feature of his oratory (Sen. Ep. mor. 40.10f.;
Tac. Ann. 4.61). He was noted for declaiming impromptu and at
great speed:
declamabat autem Haterius admisso populo ex tempore: solus omnium
Romanorum, quos modo ipse cognovi, in Latinam linguam transtulit
Graecam facultatem. tanta erat illi velocitas orationis ut vitium
fieret. itaque divus Augustus optime dixit: 'Haterius noster
sufflaminandus est1; adeo non currere sed decurrere videbatur.
nee verborum illi tantum copia sed etiam rerum erat: quotiens
velles eandem rem et quamdiu velles diceret, aliis totiens figuris,
aliis tractationibus, ita ut regi posset nee consumi. (Contr. IV
pr. 7)
He was evidently a highly-strung individual, having been moved to
tears by the pathos of his own declamation on one memorable
occasion, and it was at times like this, when his emotions were
aroused, that he was at his most powerful (Contr, IV pr. 6,11).
His worst fault seems to have been the lack of clear structure in
his declamations. We are told that he had to employ a freedman to
tell him when he had treated any given topic sufficiently and
needed to move on to the next (ibid, 8 ) . His argumentation did
not give the impression of being at all well organized, dividere
controversiam putabat ad rem pertinere, si ilium interrogares, non
putabat, si audires (ibid, 9 ) ; Seneca never refers to him in his
analyses of divisiones,
Some idea of Haterius' technique of amplification may be
gained from an examination of the extract from his treatment of
Suas. 6, deliberat Cicero an Antonium deprecetur:
sciant posteri potuisse Antonio servire rem publicam, non potuisse
Ciceronem. laudandus erit tibi Antonius; in hac causa etiam
Ciceronem verba deficient, crede mihi, cum diligenter te custodi-
eris, faciet tamen Antonius quod Cicero tacere non possit. si
intellegis, Cicero, non dicit froga ut vivas1, sed Troga ut
servias.1 quemadmodum autem hunc senatum intrare poteris, ex-
haustum crudeliter, repletum turpiter? intrare autem tu senatum
voles in quo non Cn. Pompeium visurus <es>, non M. Catonem, non
Lucullos, non Hortensium, non Lentulum atque Marcellum, non <tuos>,
tuos, inquam, consules Hirtium ac Pansam? Cicero, quid in alieno
saeculo tibi? iam nostra peracta sunt. M. Cato, solus maximum
vivendi moriendique exemplum, mori maluit quam rogare - nee erat
Antonium rogaturus - et illas usque ad ultimum diem puras a civili
sanguine manus in pectus sacerrimum armavit. Scipio, cum gladium
<in> pectus abdidisset, quaerentibus qui in navem transierant
militibus imperatorem fimperatorf inquit fbene se habet.T victus
vocem victoris emisit. 'vetat1 inquis '<me> Milo rogare iudices1;
i nunc et Antonium roga. (Suas, 6.If. W)
As early Imperial rhetoric goes, this passage has quite a flowing
quality, canorum illud et profluens, to use Tacitus' expression
(Ann. IV.61). Haterius allows himself plenty of time to dwell on
each point. The first four sentences, for example, form a group,
in that the first and the fourth relate to the proposition that
Cicero cannot be a slave, and the middle two to the related notion
that he will not be able to remain silent. Haterius is indulging
in ring-composition, one of the oldest devices in the impromptu
speaker's technique. The list of missing members of the Senate
(quemadmodum Pansam) does not lack amplitude; two examples of
noble suicides are given where one might have sufficed.
C. Albucius Silus was another declaimer lacking in a due

sense of proportion and given to over-developing the various sec-
tions of his declamations (Contr. VII pr. If.), 9 3 * but the elder
Seneca evidently preferred his efforts to those of Haterius. On
Haterius his verdict had been that despite the muddy flow of his
rhetoric (cum turbidus flueret) there was more to praise in it
than to pardon (Contr. IV pr. 11); he is even more generous to
Albucius, going as far as to include him in his primum tetradeum
(Contr. X pr. 13).
The chief features which recommended his work were the
splendor which generally characterized his style (Contr. VII pr.
2); his straightforward, unambiguous sententiae (ibid.); his skill
at arousing the right reactions in his audience (ibid. 3: adfectus
efficaciter movit, figurabat egregie, praeparabat suspiciose) ,9l+
and his apparently effortless command of inexhaustibly rich
supplies of Latin expression (ibid.).
The long extract from Albucius' treatment of Contr. VII.1
which follows immediately after his introduction in the seventh
preface may be taken to illustrate some of the best features of
his style. It certainly contains some noteworthy figures of
speech and thought. Take, for example, a passage from near the
beginning of §2, where the young man who cast his brother, con-
victed of parricidium, adrift in an open boat, rather than punish-
ing him in the traditional manner, defends his action:
a me frater ut viveret non impetravit, ut fugeret non impetravit:
nihil aliud impetravit quam ut aliter quam in culleo moreretur.
malam causam habeo, ut inter fratres. ubi spes? in gubernaculo?
nulla est. in remigio? ne in hoc quidem est. in comite? nemo
repertus est naufragi comes, in velo? in antemna? omnia [paene]
instrumenta circumcisa sunt, adminiculum spei nullum est.
The first sentence provides examples of uaoxwAov (Quint. IX.3.80),
repetitio - repetition of like openings - and conversio -
repetition of like endings (Rhet. Her. IV.13.19):
ut viveret non impetravit
ut fugeret non impetravit
nihil aliud impetravit...
The section beginning with in gubernaculo, by way of a change,
takes the form of a tetracolon with members of progressively
increasing length:
in gubernaculo? nulla est.

in remigio? ne in hoc quidem est.
in comite? nemo repertus est naufragi comes,
in velo? in antemna? omnia instrumenta circumcisa sunt,
adminiculum spei nullum est.
Each member begins with the same word, as had the first two mem-
bers of the preceding sentence, but Albucius avoids the jingling
effect which would have resulted if each had ended similarly, by
varying the close of the third one. There are other clever
touches. Notice the calculated decrease in the number of syl-
lables in the second word of each member: gubernaculo 5, remigio
4, comite 3, velo 2. Notice also how, to avoid excessive styl-
ization, Albucius asks in the last member not one but two rhetori-
cal questions. The sentence, malam causam habeo, ut inter fratres,
may be taken as an instance of one of those simple sententiae,
quas optime Pollio Asinius albas vocabat (Contr. VII pr. 2); it
can also be regarded as an example of the figure confessio nihil
nocitura (Quint. IX.2.51). This was classed as a figure of
thought, whereas isocolon, tetracolon, repetitio and conversio
count as figures of speech. But Albucius1 distinguished perform-
ance in the use of these two main classes of figurae was probably
not all that Seneca had in mind when he wrote of him in Contr. VII
pr. 3, figurabat egregie, for, as we have seen,95 he occasionally
refers to a procedure figura dividere, and one instance of it is
credited to Albucius in Contr. 1.2.16.
For all its merits, however, Albucius1 declamatory manner
lacked the aequalitas which Seneca could have wished for (Contr.
VII pr. 3 ) . In particular he had the habit of flouting declama-
tory convention by using verba sordida which sounded to Seneca1s
ears most incongruous amidst his otherwise resplendent diction
(ibid. 4 ) . 9 6 Examples of his taste for sordid words may be found
in Contr. 1.5.9; 7.18; X.I.14. Furthermore Albucius1 style de-
teriorated as he grew older, owing to his inability to settle down
and use any one style consistently. He was for ever imitating the
last stylist to whom he happened to have been listening. Seneca
could remember him listening to Fabianus the philosopher with
codices at the ready, oblivious to all else, and similarly listen-
ing, stupified, to the eloquence of Hermagoras, filled with an
ardent desire to imitate him (Contr. VII pr. 4f.). Subject as he

was to such overwhelming enthusiasms, he ranged from time to time
through all the available genera dicendi, and never fulfilled the
great promise of his youth by developing a distinctive style of
his own: sometimes he would be exilis and expound only bare facts,
sometimes he would be horridus et valens potius quam cultus, some-
times brevis et concinnus, sometimes too elevated, at other times
not elevated enough (ibid. 5 ) . 9 7
The nature of Senecafs declamatory extracts is such that we
are prevented from illustrating all the variations of Albucius1
style. Terseness in the extracts may, for example, on many oc-
casions perhaps be attributable to Senecan excerpting rather than
to Albucius1 aping of an arid declaimer. It is however possible
to point out a few parallels between his sententiae and others by
Fabianus and Hermagoras.
ALBUCI SILI. Terrae quoque suum finem habent, et ipsius mundi
aliquis occasus est; nihil infinitum est; modum <tu> magnitudini
facere debes, quoniam Fortuna non facit. magni pectoris est inter
secunda moderatio. eundem Fortuna victoriae tuae quern naturae
finem facit: imperium tuum cludit Oceanus. o quantum magnitudo
tua rerum quoque naturam supergressa estl Alexander orbi magnus
est, Alexandro orbis angustus est...non procedit ultra spatia sua
caelum, maria intra terminos suos agitantur. quidquid ad summum
pervenit, incremento non relinquit locum, non magis quicquam
ultra Alexandrum novimus quam ultra Oceanum. (Suas. 1.3)
Fabianus philosophus primam fecit quaestionem eandem: etiamsi
navigari posset Oceanus, navigandum non esse. at rationem aliam
primam fecit: modum inponendum esse rebus secundis. hie dixit
sententiam: ilia demum est magna felicitas quae arbitrio suo con-
stitit. dixit deinde locum de varietate fortunae et, cum de-
scripsisset nihil esse stabile, omnia fluitare et incertis motibus
modo attolli, modo deprimi, absorberi terras et maria siccari,
montes subsidere, deinde exempla regum ex fastigio suo devolutorum,
adiecit: 'sine potius rerum naturam quam fortunam tuam deficere.1
secundam quoque quaestionem aliter tractavit; divisit enim illam
sic ut primum negaret ullas in Oceano aut trans Oceanum esse
terras habitabiles. deinde...novissime: ut posset perveniri,
tanti tamen non esse. hie dixit incerta peti, certa deseri;
descituras gentes si Alexandrum rerum naturae terminos super-
gressum enotuisset... (Ibid, 9f.)
The sentiments italicized, commonplace though they may appear, do
not seem to have any particularly close parallels in the extracts
from other declaimers1 treatments of this suasoria quoted by
Seneca. Very few sententiae by the declaimer Hermagoras are
extant, but there is a clear parallel between one of them in

Contr. X.I.15, e£x ev £X$P°us (puaet TiappnauaaTris xaxriY0PG^v
6uvayevos, and another by Albucius, dum libere loquitur multos
offendit (ibid. 1 ) . Albucius1 tendency to use on occasion too
high-flown a style (Contr. VII pr. 5: modo nimis se attollit) had
evidently already begun before he devoted himself entirely to
school rhetoric: the fateful figura iurisiurandi which precipi-
tated his retirement from the law-courts (ibid. 6ff.) was a fine
instance of it. His liking for verba sordida is sufficient expla-
nation for SenecaTs criticism (ibid. 5 ) , modo nimis se deprimit.
Votienus Montanus is described in Contr. IX.5.15 as homo
rarissumi etiamsi non emendatissimi ingeni. As to the rareness of
his intellect, we have to take Seneca's word for it, but we are
given a fair indication of the nature of its flaws in the several
references to him outside the problematic ninth preface.98* Like
Albucius he had a regrettable tendency to forget when he was en-
gaged in a real lawsuit and to act as if he were back in the
schoolroom (Contr. VII.5.12). Like Cestius he is reported to have
jeered at the ineptiae of others (Contr. IX.6.10), but did not ex-
clude them entirely from his own declamations (Contr. VII.5.11f.).
However, his most distinctive and interesting failing was his
habit of composing strings of sententiae all making the same point,
a failing which, it was recognized, he shared with Ovid:99
habet hoc Montanus vitium: sententias suas repetendo corrumpit;
dum non est contentus unam rem semel bene dicere, efficit ne bene
dixerit. et propter hoc et propter alia quibus orator potest
poetae similis videri solebat Scaurus Montanum inter oratores
Ovidium vocare; nam et Ovidius nescit quod bene cessit relinquere.
(Contr. IX.5.17)
Seneca records two examples of Montanus1 repetitiveness, the first
from his speech for one Galla Numisia, accused of poisoning her
father, one twelfth of whose property she stood to inherit.
uncia nee filiae debetur nee veneficae.
in paternis tabulis filiae locus aut suus debetur aut nullus.
relinquis nocenti nimium, innocenti parum.
non potest filia tarn anguste paternis tabulis adhaerere, quas aut
totas possidere debet aut totas perdere. (Ibid. 15f.)
Apparently, though Seneca could not recall it all, he said much
more to the same effect. Seneca comments: nihil non ex eis bellum
est, si solum sit; nihil non rursus ex eis alteri obstat. His
other example comes from Montanus1 treatment of Contr. IX.5, which

concerns a boy who, after the death of his brothers in suspicious
circumstances, was seized by his grandfather from the care of his
stepmother.
erras, pater, et vehementer erras: quos perdidisti non quaeris,
quem quaeris non perdidisti.
puer iste <si> invenitur perit.
quisquis puero favet ne inveniatur optet.
puer, nisi avum sequitur, fratres secuturus est; desine quaerere
quem si inveneris sic perdes ut invenire non possis.
rapuit isturn avos ne raperet noverca.
unum tantum pater ex liberis suis quaerit qui salvus est. (ibid,
16)
Probably the sententiae quoted followed immediately upon one
another, as do the three which he cites for the sake of comparison
from Ovid, Met. XIII.503-5:
ne multa referam quae Montaniana Scaurus vocabat, uno hoc con-
tentus ero: cum Polyxene esset abducta ut ad tumulum Achillis
immolaretur, Hecuba dicit:
cinis ipse sepulti
in genus hoc pugnat.
poterat hoc contentus esse; adiecit:
tumulo quoque sensimus hostem.
nee hoc contentus est: adiecit:
Aeacidae fecunda fui.
aiebat autem Scaurus rem veram: non minus magnam virtutem esse
scire desinere quam scire dicere. (Contr. IX.5.17) 100
Among the declaimers introduced in Seneca's tenth preface,
none of them consistently first-rate, several appear to have had
styles distinct from any we have so far encountered.
By Seneca's standards Scaurus and Labienus were more or less
old-fashioned stylists. In Contr. X pr. 2 he classes Scaurus1
oratorical style unequivocally as antiquum, praising his verborum
...non vulgarium gravitas and the authority lent to his utterances
by his distinguished appearance. His only regret is that Scaurus
had ruined his chances of becoming a great orator through negli-
gence (ibid. 3 ) . Labienus affected the contempt for public decla-
mation still evidently customary among orators of his generation
(ibid. 4 ) , but he appears to have made more concessions than
Scaurus to modern tastes in style: color orationis antiquae, vigor
novae, cultus inter nostrum ac prius saeculum medius ut ilium
posset utraque pars sibi vindicare (ibid. 5 ) . We are more ad-
equately provided with samples of Labienus' style than of Scaurus1
A celebrated attack of his on the unpunished vices of the upper

classes - a subject doubtless congenial to this writer, whose
books were burnt as a consequence of his belief in free speech
(ibid.) - is quoted in Contr. X.4.17f. W:
Labienus tarn diserte declamavit partem eius qui debilitabat ex-
positos quam nemo alteram partem, cum illam omnes disertissimi
viri velut ad experimentum suarum virium dixerint. ilium autem
locum vehementissime dixit: vacare homines huic cogitationi, ut
curent quid homo mendicus inter mendicos faciat! principes,
inquit, viri contra naturam divitias suas exercent: castratorum
greges habent, exoletos suos ut ad longiorem patientiam inpudic-
itiae idonei sint amputant, et, quia ipsos pudet viros esse, id
agunt ut quam paucissimi sint. his nemo succurrit delicatis et
formosis debilibus. curare vobis in mentem venit quis ex soli-
tudine infantes auferat perituros nisi auferantur; non curatis
quod solitudines suas isti beati ingenuorum ergastulis excolunt,
non curatis quod iuvenum miserorum simplicitatem circumeunt et
speciosissimum quemque ac maxime idoneum castris in ludum con-
iciunt. in mentem vobis venit misereri horum quod membra non
habeant. quidni illorum quod habent? et hoc genere insectatus
saeculi vitia egregia figura inquinatum et infamem reum maiorum
criminum inpunitate defendit.
Nobody would be tempted to date Labienus1 invective earlier than
the Augustan age: such turns of phrase as contra naturam divitias
suas exercent and quia ipsos pudet viros esse, id agunt, ut quam
paucissimi sint bear the unmistakable stamp of their period. Yet
the rhythms favoured in an earlier age reverberate in LabienusT
sequences of polysyllabic words, for example: longiorem patientiam
inpudicitiae and iuvenum miserorum simplicitatem circumeunt, even
though the use of iuvenum simplicitatem to denote iuvenes be-
longs to the new era. Such circumlocutions as curare vobis in
mentem venit instead of curatis, and in mentem vobis venit misereri
instead of miseremini, also distinguish him from the modernist
devotees of fiery agitation.
We have already considered the style of Musa, described in
Contr. X pr. 9.101 The chief distinguishing feature of Moschus,
who is introduced next (ibid. lOf.) seems to have been excess in
the employment of figurae:
Moschus non incommode dixit, sed ipse sibi nocuit; nam dum nihil
non schemate dicere cupit, oratio eius non figurata erat sed prava.
itaque non inurbane Pacatus rhetor, cum illi Massiliae mane oc-
currisset, schemate ilium salutavit: fpoteramf inquit 'dicere: ave
Mosche.r
Such cautious qualifying of terms was a variety of the figure
praesumptio to which Quintilian refers in IX.2.18: verborum quoque

vis ac proprietas confirmatur...reprehensione: 'cives, inquam, si
hoc eos nomine appellari fas est', and it is indeed to be found in
MoschusT fragments in Contr. X.1.3: mortuo patre meo - timeo enim
ne quis sibi iniuriam fieri putet si dixero 'occiso'. and in
Contr. X.3.1: inquinasti filiae sanguine penates. quamquam quid
ego dico penates, tamquam in domo perierit?
Study of the fragments suggests that there was more that could
have been said about Moschus* style than Seneca cares to mention
in the tenth preface. It is clear that he was a declaimer, much
like Fuscus, with two styles: a dry style for narrative and
argumentation (e.g. Contr. II.3.4; X.1.3; 2.17; 6.1) and the quasi-
poetic descriptive manner which we find in Suas. 1.2:
inmensum et humanae intemptatum experientiae pelagus, totius orbis
vinculum terrarumque custodia, inagitata remigio vastitas, litora
modo saeviente fluctu inquieta, modo fugiente deserta; taetra
caligo fluctus premit, et nescio qui, quod humanis natura subduxit
oculis, aeterna nox obruit. 102
According to Porphyrion,103 Moschus hailed from Pergamum; it was
only after conviction on a charge of poisoning that he went west
to teach at Massilia (Contr. II.5.13).
Fulvius Sparsus and Iulius Bassus, introduced in Contr. X pr.
llf., are both criticized for harshness. Sparsus attempted to
imitate Latro, but lacked the originality to seem more than a
mimic:
Sparsus autem dicebat valenter1^4 sed dure. ad imitationem se
Latronis derexerat nee tamen umquam similis illi erat, nisi cum
eadem diceret. utebatur suis verbis, Latronis sententiis.
Parallels between Sparsus and Latro do not actually seem to be
very common in the fragments, though quite large samples exist of
both men's writing. In Contr. VII.6, however, two sententiae by
Sparsus: eligitur maritus quern sanus pater dotalem dedisset, and
gener tuus ipsis nuptiis crucem meruit (§3), may be compared with
two by Latro: cum infelici face ad dotalem suum nova nupta de-
duceretur...exhorrui... and vocat servum et, quia crucem non
meruerat, mereri iubet (§9).
Iulius Bassus was the declaimer whose amaritudo and imitation
of forensic actio provoked Seneca to make the remark, nihil est
indecentius quam ubi scholasticus forum, quod non novit, imitatur
(Contr. X pr. 12).* 0 5 One means by which Bassus attempted to

dissociate himself from scholastic convention and to acquire
amaritudo was evidently by the use of verba sordida (e.g. Contr.
1.2.21; 3.11; X.1.13); 106 in the first two of these passages he
also defies the declaimers1 ruling that women were to be respect-
fully treated.107 We have to take Seneca at his word that Bassus
aped the practice of the forum in his delivery, but one can well
imagine one particular passage from his treatment of Contr. I.7 108
being accompanied by a wide variety of gestures:
nolo me tarn bene alas quam ego te alui; nolo ignoscas mihi: quid-
quid passus es, quidquid timuisti, patiar: posce flagella, scinde
rugas. ustus es? subice ignes, semimortuam hanc faciem, quae
tantum in contumeliam suam spirat, quia extingui non potest,
exure. si parum est, fac quod ais ne piratas quidem fecisse,
manus praecide. exhibeo tibi. hae sunt illae quae quidlibet
scribunt. ubi est gladius tuus? stringe. (Contr. 1.7.9)
The very fact that Capito, introduced next (in Contr. X pr.
12), was, by contrast with Bassus, a bona fide scholasticus made
Seneca look upon him with a certain whimsical indulgence. He even
goes as far as to say that Capito was in his declamationibus, quae
bene illi cesserunt, nulli non post primum tetradeum praeferendus.
Nevertheless he still considered it regrettable that Capitofs
declamation about Popillius was commonly attributed to fpoor
Latro1 (ibid.). Reading the extract (Contr. VII.2.5ff.) which
Seneca gives from this declamation, we can discover one of the
reasons why it might have been damaging to Latrofs reputation:
Capito appears to have been suffering, when he wrote it, from that
disease by which, according to Seneca in Contr. VII.5.12, schol-
astici in general were afflicted, a lack of moderation in the use
of exempla:
Ciceronem quisquam potuit occidere qui audiit? Minturnensis palus
exulem Marium non hausit; Cimber etiam in capto vidit imperantem;
praetor iter a conspectu exulis flexit; qui in crepidine viderat
Marium in sella figuravit. non possumus de Popillio queri: eodem
loco patronum habuit quo patrem. Cn. Pompeius terrarum marisque
domitor Hortensi se clientem libenter professus est; et Hortensius
bona Pompei, non Pompeium defenderat. Romulus, horum moenium
conditor et sacratus caelo parens, non tantam urbem fecit quantam
Cicero servavit. Metellus Vestae extinxit incendium, Cicero Romae.
glorietur devicto Hannibale Scipio, Pyrrho Fabricius, Antiocho
alter Scipio, Perse Paulus, Spartaco Crassus, Sertorio et Mithri-
date Pompeius; nemo hostis Catilina propius accessit. (Contr. VII.
2.6f. W)
None of the three Spanish declaimers, Gavius Silo and the

Clodii Turrini, father and son, praised at the end of the tenth
preface (§14ff.)> appears, either from Seneca's critical remarks or
from the extracts attributed to them, to have been a very distinc-
tive stylist. True, we may find sententias...excitatas, insidi-
osas, aliquid petentis (Contr. X pr. 15) in the elder Turrinus'
fragments, but the same could be said of almost any extract in
Seneca1s anthology. It remains an important fact that such men
lived, cultivating in such remote cities as Tarraco (ibid. 14) and
T
eloquent' Corduba (Martial 1.61.7) standards, both in declamation
and oratory proper, which seemed to Seneca the Elder comparable
with those of the metropolis (Contr. X pr. 13), thus helping to
ensure that the economic prosperity of early Imperial Spain was
matched by a good measure of cultural prosperity.
Among the declaimers introduced in Seneca's prefaces there
are, as we see, exponents of all the genera dicendi mentioned in
the account of influences on Albucius in Contr. VII pr. 5 ^ 9
(except perhaps anyone who could be described as consistently
brevis et concinnus), and representatives of a few more styles
besides. And nowhere does Seneca give us any warrant for suppos-
ing that all these diverse stylists were considered either by
themselves or by their contemporaries as belonging to a single
united Asianist movement. He uses the term Asianus just enough
to show that the battle of the books between Atticists and Asian-
is ts was not a dead issue, but he uses it in a way that makes it
implausible that he regarded himself or Latro, his favourite de-
claimer, as belonging to the Asianist camp, and so sparingly that
it might not seem unreasonable to deduce that in his time it was
very much the exception, rather than the rule, for a rhetorician
to be classifiable as ex Asianist10
However, Asianist rhetoricians were probably rather more
numerous and influential than examination of Seneca's extant
criticism alone might lead us to believe. True, surprisingly few
declaimers are mentioned by Seneca as having imitated or inspired
sententiae by the five explicitly named Asiani. Among Greek de-
claimers, Dionysius the Younger seems to have composed a sententia
along the same lines as one by Hybreas which Seneca quoted in
Contr. 1.4.11, though now that both fragments are lost we cannot
know if he imitated him closely, and one Glaucippus Cappadox
possibly inspired a sententia by Adaeus in Contr. IX.2.29. In
Contr. X.4.20 M certain Latin declaimers are reported to have
copied another of Adaeus1 sententiae, fsed sic1, Seneca is at pains
to assure us, Tut putem illos non mutuatos esse aperte hanc senten-
tiam, sed imitatos.* Rather surprisingly, one of them was Blandus,
Fabianus1 second teacher (cf. Contr. II pr. 5 ) ; the others were
Moschus and Fuscus. We cannot now tell how close Vibius RufusT
sententia in Contr. 1.4.11 came to Hybreas' treatment of the same
idea. It is perhaps remarkable, in view of the obvious importance
of Arellius Fuscus in the scholastic circles of his day, how sel-
dom Seneca tells us of straightforward imitations of his senten-
tiae, divisiones and colores, or of precedents for them, in the
work of other declaimers. There is a certain affinity between a
sententia of his in Contr. 1.4.10 and one of Latro's (ibid.), but
Fuscus1 version is condemned as illius sententiae frigidius...
contrarium. In Contr. VII.3.7 Fuscus uses the same color as
Albucius, sed aliter. In Contr. 1.1.15 Cestius is found force-
fully elaborating the concluding quaestio of Fuscus1 divisio, and
correspondingly in Contr. X.4.21 Fuscus is found rendering a
sentiment of Cestius1, taken from the Greek of one Damas Scombros,
aliter, that is, as it happens, with greater elaboration and force.
But, for a reason already explained,111 Cestius was incapable of
being a successful imitator of Fuscus in all respects, and in any
case, as we see from Contr. II.3.22, he was not an uncritical ad-
mirer of him. An interesting possible candidate for inclusion
among the Asiani is Haterius, who in Contr. IX.6.16 is named as
the author of a sententia very similar to one which Fuscus trans-
lated word for word from Adaeus: 112 modestius hanc sententiam
vertit Haterius: quid ergo? mentita est? quidni ilia mentiretur
de accusatoris sui filia? Perhaps his verbosity should be re-
garded as an imitation of what Augustus, criticizing Antony,
called Asiaticorum oratorum inanis sententiis verborum volubilitas
(Suet. Aug. 86.3), even though Seneca says nothing more precise
about the sources of inspiration for his fluency than that they
were Greek (Contr. IV pr. 7).
Probably some, if not all, of the enfants terribles casti-

gated frequently by Seneca for corruptum9 insania and cacozelia,
should also be grouped with the Asiani. That is not to accept the
presumption**3 that the use of these terms in antiquity was necess-
arily confined to criticism directed against professed Asianists,
any more than their use was confined exclusively to the remarks of
critics of the extreme Atticist persuasion, which, as we have
seen, 114 was not the case in CiceroTs time, and certainly was not
in the early Empire, when we find even Maecenas extolling the vir-
tues of 'sane1 expression and condemning tumidity (Suas. 1.12).
It should be noted that Seneca the Elder did not consider that
insania was a stylistic tendency which anyone, apart from Vibius
Gallus, the great exception, cultivated deliberately; rather
people were liable to lapse into it casu, by accident (Contr. II.
1.25). But, unquestionably, many of the faults which he criti-
cizes as insane or corrupt115 are the same which marred the style
of Hegesias, Maecenas and Arellius Fuscus. And it has to be re-
membered that even the declaimers whom Seneca liked least, Murre-
dius, Gargonius and the rest, each had their regular gatherings of
pupils, and doubtless exercised more influence on the impression-
able young than he would have wished. It seems quite likely too
that in the period immediately following the elder Seneca's death,
men of their stamp came to dominate the scholastic world as they
had not in the days when Latro flourished, hence QuintilianTs
strictures in XI.3.58 on the universal prevalence of sing-song
delivery.
However, even if we assume that Asiani were more numerous in
the schools of the elder Senecafs time than the scarcity of ex-
plicit references to them might suggest, it still seems unjusti-
fiable to follow Norden in his supposition that all the varieties
of the fnew style1 of the early Empire were manifestations of
Asianism.116 Of the passages cited by him in this connection,
Suet. Aug. 86.3 and Plut. Ant. 2.5, concerned as they are with
Mark Antony's style, relate to a period too early for them to de-
mand discussion here, and Quint. XI.3.58 refers to a period too
late. So, indeed, does Petr. Sat. Iff., but this passage, being
closer in date to Seneca the Elder, and containing as it does a
reference to Asianism in the course of an attack on modern de-

claimers in general, cannot be so lightly dismissed as without
relevance to our present investigations.
In Sat. If. Petronius' versatile hero Encolpius is declaiming,
on his own admission (§3) , against declamation, and in particular
that of the rhetorician Agamemnon to whom he has just been listen-
ing. It is in the course of this tirade that he makes the alle-
gation that this windy and monstrous loquaciousness of his was a
recent immigrant to Athens from Asia: nuper ventosa istaec et
enormis loquacitas Athenas ex Asia commigravit... (§2). He argues
that style of this kind had been a pernicious influence on young
talent, and points out that the great lyric poets and tragedians
had not needed to declaim and nor, moreover, had Plato and Demos-
thenes. It should be noted that, with the exception of the opening
parody of declamatory style which, like all the dialogue in the
Satyricon, is in Latin, the whole of Encolpius1 harangue has to do
with Greek literary history. It sounds suspiciously like the
parroting of stock arguments which had been in circulation ever
since scholars had started to object to the way that pernicious
oriental influences had 'recently' begun to pollute Athenian ora-
tory. Encolpius' use of nuper may be compared with the use of
itpwriv by Dionysius of Halicarnassus in De rhet. vet. 1:
n yev 'Axxtxri youaa xal dpxaâ xau auxox^wv axuyov euXncpeu
y y px x y p
x&v eauxrjs exiieaouaa dyadwv, n 6e ex TLVWV Bapd§pwv xfis '
ex$es xal rcpwnv dcptxoyevn, Muan n ^pvyla T L S rj Kapuxov TL xax6v,
'EXXnvudas n£uou 6touxetv uoXets dueXdaaaa TWV XOLVWV xriv exepav,
n dyadfis xriv cpuXoaocpov Mai n yauvoyevn xnv awcppova.
The fact remains that the dispute between Encolpius and Agamemnon
is supposed to be taken as referring to the Roman schools of Petro-
nius1 time: Encolpius has been attacking declamatores in general
(§1) just before he makes his reference to Asiatic loquacitas, and
it emerges from §5, where Agamemnon describes his educational
ideals, that the declamatores in question are rhetoricians en-
trusted with the education of Roman youths in the post-Ciceronian
era.
But how seriously ought we to take this passage in the Satyri-
con as an analysis of Imperial declamatory style? How are we to
1
weigh Encolpius generalization that the bombast of the declaimers
was of Asiatic origin against the view discernible in the elder

Seneca's criticism that, while some declaimers were Asiani, others
were not?
Perhaps one might argue that we need see no conflict between
the two views: Asiani could have been so much more in the ascend-
ant in Petronius1 day than in the elder Seneca's that to say that
by then they had come to dominate the schools totally might not
have been much of an over-statement. One could also argue that
Seneca, no less than Encolpius, would have found fault with the
type of style parodied in Sat. I.**7
But maybe this would be to pay more respect to Encolpius1
criticism than is its due. It is surely to be doubted whether the
type of critic portrayed by Petronius in the person of Encolpius
is likely to have been a very reliable authority on the question
of the origins of early Imperial sententiarum strepitus, seeing
how he is prepared to regurgitate anti-Asianist criticisms which
had obviously long been out of date, without making even so much
of an attempt to adapt them to a modern Roman context as to sub-
stitute Romam ex Asia for Athenas ex Asia. One can fairly expect
such a critic to have known that suasoriae and controversiae had
Hellenistic precedents, for this fact had been pointed out by
scholars prior to Quintilian (see II.4.41); he would also be
likely to know that Asiatic orators had been known for their ver-
bosity before the rise of the Roman Imperial schools of rhetoric,
for Cicero (Brut. 95.325) was among the authorities for this in-
formation. But it does not seem at all likely that such a man
would have thought very deeply about the possible origins of the
particular variety of loquacitas represented by rhetoricians like
Agamemnon, or that any Roman scholar of the early Empire would
have been disposed to do the necessary research in the archives of
the Hellenistic schools to discover whether the genus dicendi
ardens et concitatum came into favour first in Asia, or elsewhere.
If we have to choose between the evidence of Seneca the Elder
and Petronius1 Encolpius on this issue, it is surely more sensible
to take seriously the objective reporting of the older authority
than to accept without question the partisan and hackneyed gener-
alizations in Petr. Sat. 2. And the elder Seneca gives us no
warrant for supposing that the whole of early Imperial school

rhetoric ought to be labelled as Asiatic. His criticism rather
serves to undermine that supposition. Asianism there was in some
of the schools of his day, but in the van of the Augustan declama-
tory movement we find Latro, a man ignorant of Greek rhetoric
(Contr. X.4.21), who, if not himself responsible for expressing
the highest praise for Calvus the Atticist which has survived from
antiquity (Contr. VII.4.6ff.), was the closest friend of our
authority for it,11** and whose novelties of expression were open
to criticism from purists (Contr. II.4.8) probably as occidental
rather than oriental.
A fact which makes it all the more implausible that all the
declaimers of the elder Seneca's time, Greek and Latin, could have
been regarded by any contemporary critic as all representatives of
a united Asianist movement is that Dionysius of Halicarnassus, who
lived in Rome from 30 B.C. until his death some time after 8 B.C.,
and thus all through the period when Latro was the leading Latin
rhetorician,11^ took the view that Asianism was no longer an im-
portant force in the schools of his day. When Dionysius wrote
disparagingly about the meretricious style of expression recently
arrived from out of the hell-holes of Asia, and referred to it in
f
De rhet. vet. 2 as the new rhetoric1, TT\ 6e vea...[sc. pnxopLXfl]
- thereby suggesting to Norden the term 'der neue Stil1 which he
uses as a synonym for 'Asianismus1^^® - he was referring not to
the rhetoric of his own time, but to a type of style favoured in
the recent past but now almost universally rejected, thanks to the
good taste of the Romans (§3), in favour of a return to the so-
briety of the old (i.e. classical Attic) style (§2). He has this
to say about the Greek rhetoric of his time:
e£u) yap oAuywv TLVGOV 'AauavaJv noXewv, aCs 61/ dyaduxv Bpa6eud
eaxuv n TWV xaX&v ya^naus, <*L XouTtal TtenauvTaL TOUS cpopxuxous xal
4>uxpouc; xal dvata^nTous dyomujaau Xoyous, xwv yev itpoTepov yeya en'
auTOts cppovouvTwv ail6ouyevu)V rf6n xau waxa ytwpov duauToyoAouvTOJV
itpos TOUS exepous, ei yn Tuves TtavTaitaatv dvtdtTws exouau, TWV 6e
vecaaxl TOU yaSnyaxos dnToyevwv eus xaxatppovnauv dyovTcov TOUS
Xoyous xal yeXwxa uououyevwv xnv en' auTous cntou6riv. (ibid.)
Here we have a reference to a phase in the history of Greek rhet-
oric of which there is no mention in Encolpius1 tirade in Petro-
nius, Sat. If. It is interesting to compare Dionysius1 optimism
about the tastes of Greek students of rhetoric at the time of his

sojourn in Rome,*2* with the elder Senecafs pleasure at recalling
the standards of declamation customary in those 'better years1
(Contr. I pr. 1), when men born within Cicero's life-time had
still been the shining lights of the literary world (ibid, 7).
Suetonius in Aug. 86.2 refers to cacozelos et antiquarios as
standing at the two opposite poles of this literary world.
Maecenas, who, as we have agreed, deserved the epithet Asianus as
much as Arellius Fuscus deserved it, is named by Suetonius as a
representative of the cacozeli, while Tiberius and Annius Cimber
are classed as antiquarii. What Suetonius does not say is that
all prose writers of the Principate were deemed by Augustus, or
anyone else, to have adhered without exception to one or other of
the alternative types of extremism. Augustus himself, we are
given to understand (ibid.), tried to find a golden mean somewhere
between them. And this, surely, is what most of the declaimers
quoted by the elder Seneca would have claimed to have been doing,
whatever we may think of their efforts now. Within the range of
styles to be heard in the auditoria of the declaimers contemporary
critics detected good taste as well as bad. Amateurs visiting the
schools and professional rhetoricians alike considered they had
every right to criticize inferior declaimers for lapses of taste,
without issuing any general condemnation of the sententiousness so
widely - we may think excessively - cultivated in the rhetorical
prose of their day. The modern critic of Silver Latin literature
therefore needs to exercise great caution if he wishes to use such
terms as corruptum9 cacozelia, tumor and cultus in contexts where
no ancient critic has anticipated him. Still more, he should use
the term 'Asianist1 only with the utmost circumspection, bearing
in mind that this epithet may certainly be applied to the type of
style favoured by Fuscus in his explicationes, but not, it seems,
despite the fact that Asiani were undoubtedly among those who in
the elder Seneca's day participated in the general fashion for
sententiousness, to the consistently fiery agitation of Latro, or,
to put the matter in Tacitean terms (see Dial. 26.1), it may
safely be used with reference to the calamistros Maecenatis, but
not, however much distaste we may feel for this kind of writing,
as a term with which to issue general condemnation of the tinnitus

Gallionis.
2 DECLAMATION AND LITERARY MODERNISM IN THE EARLY EMPIRE
In the early years of the elder SenecaTs life-time it appears that

literary Latin had undergone a profound and rapid change. A
startling indication of the rapidity with which it had come about
is that he found it remarkable (Contr. IV pr. 9) that Haterius
used words which were antiqua et a Cicerone dicta, a ceteris
deinde deserta. Consequently, though he does occasionally dis-
tinguish between an old style and a new, for example in his de-
scriptions of Scaurus and Labienus in the tenth preface,1 it is
not to be imagined that men described as antiqui in his criticism,
for example the declaimer named Crispus whom he calls an antiquum
rhetorem in Contr. VII.4.9, were the kind of stylists whom Augustus
would have referred to as antiquarii (Suet. Aug. 86.2). It is
evident from his remarks on the old and new types of divisio in
Contr. 1.1.13 that to the elder Seneca veteres meant the gener-
ation of Porcius Latro. For he promises, 'ego exponam quae aut
veteres invenerunt aut sequentes2- adstruxerunt1, and yet he never
gives examples of the simple divisio controversi arum antiqua
earlier than those of Latro and the rhetoricians active at the
time when he rose to fame. Extreme archaism does not appear, from
anything the elder Seneca says, to have been a strong force in
early Imperial declamation. Even such a stylist as Pollio, of
whom Quintilian was to write, a nitore et iucunditate Ciceronis
ita longe abest ut videri possit saeculo prior (X.I.113), became
extraordinarily self-indulgent when he declaimed.^
Seneca the Elder, as we have seen, pays lip-service in Contr.
I pr. 9 to the elder Cato and in Contr. VII.4.8, quoting Latro
maybe, to Demosthenes, but otherwise shows remarkably little
interest in pre-Ciceronian oratory.** And it was not only in the
case of rhetoric that he seems to have turned his back on the past
and directed practically his whole attention towards modern litera-
ture: in historiography, philosophy and poetry too his interest
Declamation and literary modernism in the early Empire 305
seems to have been confined almost exclusively to the writings of

the last generation of the Republic and his own contemporaries.
That his neglect of the older Latin literature was the result
of distaste for it, rather than mere ignorance, is evident from
his disagreement with Pollio over the acceptability of a sententia
in Contr. IX.2.25, and most strikingly in his agreement with
Livy's remarks to the effect that the pursuit of verba antigua et
sordida and a severitas which resulted in obscurity was as much a
form of insania as the opposite vice of tumidity (ibid. 26f.).
By agreeing with Livy over this matter Seneca the Elder falls into
line with the attitude expressed by Cicero towards the Thucydideans
(Or. 9.3Off.).
In the field of historiography it is clear that the elder
Seneca's chief interest was in the writings of modern Romans from
Sallust onwards, though he does mention both the greatest Greek
historians (Quint. X.I.73: longe duos ceteris praeferendos),
Herodotus and Thucydides. He was mistaken in believing (Suas. 2.
11) that a famous dictum of Leonidas, in which he urged his men to
eat their morning meal, destined as they were to dine in Hades,
was to be found in Herodotus,6 but one at least of his references
to Thucydides seems the fruit of better scholarship. In one of
his rare displays of erudition (Suas. 6.21), he notes that Thucy-
dides had once or twice employed the device of eituTatpuov, in which
the historian, having narrated the death of a great man, assesses
his achievements in a consummatio totius vitae et quasi funebris
laudatio; this device, he goes on to observe, was used occasion-
ally by Sallust, more frequently by Livy, and even more profusely
by Livy's successors. As D. R. Stuart has pointed out, such
euuTacpua do indeed occur very occasionally in Thucydides.7
Whether the research which sought out this honourable precedent
for the practice of Sallust and his successors was the elder
Seneca's own may be doubted. Sallustfs debt to Thucydides was not
a subject in which only dedicated historians were interested.8 It
was even discussed by Arellius Fuscus, of all people, in an argu-
ment aimed to confute someone (an Atticist?) who disapproved of
his close imitation of the Asian rhetorician Adaeus:
do, inquit, operam ut cum optimis sententiis certem, nee illas
corrumpere conor sed vincere. multa oratores, historici, poetae

Romani a Graecis dicta non subripuerunt sed provocaverunt. tune
deinde rettulit aliquam Thucydidis sententiam: 6euvau yap at
euipa^Cau auyxpuâu nal auaxuaaaL xa exaaxwv ayapTnyaTa. deinde
Sallustianam: res secundae mire sunt vitiis obtentui. cum sit
praecipua in Thucydide virtus brevitas, hac eum Sallustius vicit
et in suis ilium castris cecidit; nam in sententia Graeca tarn
brevi habes quae salvo sensu detrahas: deme vel auynpuâi, vel
auo"Huaaau, deme enaaxoov: constabit sensus, etiamsi non aeque
comptus, aeque tamen integer, at ex Sallusti sententia nihil demi
sine detrimento sensus potest. (Contr. IX.1.13 W)
The Greek sententia quoted actually came from ps.-Demosthenes (in
ep. Phil. 13), not Thucydides, but the fact that Seneca did not
notice the error in such a point of detail tells us nothing about
the extent of his Greek reading, for Livy, too, was deceived:
T. autem Livius tarn iniquus Sallustio fuit ut hanc ipsam senten-
tiam et tamquam translatam et tamquam corruptam dum transfertur
obiceret Sallustio. nee hoc amore Thucydidis facit, ut ilium
praeferat, sed laudat quem non timet et facilius putat posse a se
Sallustium vinci si ante a Thucydide vincatur. (Contr. IX.1.14)
Was Livy, one wonders, present in Fuscus1 audience, or was this
example of SallustTs art a famous one which he discussed on
another occasion?
None of the Latin historians prior to Sallust receives any
mention in Seneca's criticism. As we have seen, there is a par-
allel for the elder Seneca's apparent neglect of pre-Ciceronian
oratory in the views expressly stated by Velleius Paterculus in
a digression (1.16-17) where he reflects on the brief efflorescence
of the various literary genres. There is a similar parallel for
Seneca's apparent lack of interest in the pre-Sallustian Roman
historians in Velleius' remarks on historiography:
historicos etiam, ut Livium quoque priorum aetati adstruas,
praeter Catonem et quosdam veteres et obscuros minus octoginta
annis circumdatum aevum tulit... (1.17.2)
Seneca's interest in historiography extended to writers who sur-
vived Livy, whose death in A.D. 17 presumably marked the end of
Velleius' 'less than eighty years' of great history writing, for
Cremutius Cordus, quoted in Suas. 6.19, 23, did not die until A.D.
25, Bruttedius Niger, quoted in Suas. 6.20, survived to be impli-
cated in the fall of Sejanus in 31, and he himself continued
writing history almost right up to his death well on in the 30s.9
That is not to say that he would necessarily have disputed
Velleius1 contention that Livy was the last of the great his-
torians. In the remarkable passage, Suas. 6.14-27, where he com-
pares and contrasts various writers' narratives of the death, and
estimates of the achievements, of Cicero, it is Livy who, among
the historians, receives the highest praise. Seneca approved of
the way that Livy was generous (benignus) in the distribution of
Gituxacpua to the great men of history (§21), and admired in par-
ticular his tribute to Cicero: ut est natura candidissimus omnium
magnorum ingeniorum aestimator T. Livius, plenissimum Ciceroni
testimonium reddidit (§22). He also found much to admire in the
eituxacpuov on Cicero by Asinius Pollio, another historian whose
literary activity was confined to Velleius1 fless than eighty
years1 (§25: adfirmare vobis possum nihil esse in historiis eius
hoc quern rettuli loco disertius, ut mihi tune non laudasse
Ciceronem sed certasse cum Cicerone videatur), even though he dis-
approved of the anti-Ciceronianism in Pollio!s writings, as exemp-
lified by his outrageous allegation in the speech Pro Lamia (so
baseless that he dared not include it in his Histories) that
Cicero had promised to recant and write as many speeches for
Antony as he had previously written against him (§14f.), and by
the narrative in which he made out that Verres, when proscribed,
had faced his death with great courage (§24). By contrast, Seneca
does not single out any of the other historians whom he quotes in
Suas. 6.14ff. for particular praise, and he is highly critical of
Cremutius Cordus* tribute to Cicero: Cordi Cremuti non est operae
pretium referre redditam Ciceroni laudationem; nihil enim in ea
Cicerone dignum est, ac ne hoc quidem, quod \_paene~] maxime tolera-
bile est (§23). However, he must have had the eloquence of these
three historians in mind as well as that of Livy and Pollio when
he referred in §25 to tot disertissimis viris. Quintilian, re-
viewing Roman historiography in X.l.lOlff., was to take the line
that though Sallust and Livy, equal in his view to Thucydides and
Herodotus respectively, were Rome's greatest historians, Aufidius
Bassus and Cremutius Cordus had much to recommend them.
The only philosophers to whom Seneca the Elder, as opposed to

the critics quoted by him, refers are all modern ones. The refer-
ence to Plato in Contr. Ill pr. 8 occurs in the course of Cassius
Severus' remarks on specialization; the passing allusion to Arist-

otle in Suas. 1.5 comes within a quotation from Cestius. Among
modern Roman philosophers, Seneca the Elder mentions Fabianus
(Contr. II pr. passim), Sextius, Fabianus1 mentor (Contr. II pr.
4 ) , and Attalus, a Stoic who suffered banishment at the instigation
of Sejanus and was, so he tells his sons, magnae vir eloquentiae,
ex his philosophis, quos vestra aetas vidit, longe et subtilissimus
et facundissumus (Suas. 2.12).
All these three were philosophers admired by the younger
Seneca. In Quaest. nat. VII.32.2, for example, there is an ap-
proving reference to the Sextiorum nova et Romani roboris secta,10
and extended praise for the elder Sextius in Ep. mor. 64.2ff.
Fabianus, whose lectures the younger Seneca had attended (Ep. mor.
52.11), is praised for the sincerity of his teaching in Brev. vit.
10.1, for the excellence of his way of life, his learning and his
eloquence in Ep. mor. 40.12, and is defended in Ep. mor. 100 from
criticisms which had been levelled against his style. As for
Attalus, we have in Ep. mor. 108.3 a lively account of the younger
Seneca's enthusiasm for his teaching: haec nobis praecipere Attalum
memini, cum scholam eius obsideremus et primi veniremus et nov-
issimi exiremus, ambulantem quoque ilium ad aliquas disputationes
evocaremus, non tantum paratum discentibus sed obvium.
There is not a trace of disapproval in the elder Seneca's
references to these men, in spite of his son's statement in Ep.
mor. 108.22 that the reason why his father had made him give up
vegetarianism in his youth was that he abhorred philosophy:
in primum Tiberii Caesaris principatum iuventae tempus inciderat:
alienigena turn sacra movebantur et inter argumenta superstitionis
ponebatur quorundam animalium abstinentia. patre itaque meo
rogante, qui non calumniam timebat sed philosophiam oderat, ad
pristinam consuetudinem redii; nee difficulter mihi ut inciperem
melius cenare persuasit.
Why he interpreted his father's motives in this way is mysterious.
Perhaps it was because the elder Seneca's view that discretion was
sometimes the better part of valour (expressed, for example, in
Contr. II.4.13) seemed out of keeping with the antiquus rigor he
is credited with in Cons. Helv. 17.3, 11 that the younger Seneca is
so insistent that his father did not act out of fear. What we can
be sure of is that it was none of the three philosophers praised

by father and son alike who persuaded Seneca the Younger to adopt
his vegetarian diet, but one Sotion, a philosopher never mentioned
by his father in the extant books, who differed from Sextius in
using the Pythagorean doctrine of the transmigration of souls as
an argument against meat-eating, in addition to the usual medical
and moral arguments (Ep. mor. 108.17ff.).
Insofar as the elder Seneca can be said to have had a philo-
sophical position, it appears to have anticipated in certain re-
spects the views later to be expounded by his second son, which is
hardly surprising in view of their shared admiration for the
Sextians and Attalus.
In Contr. IV pr. 4ff. and Suas. 2.15 he tells two pairs of
anecdotes describing the contrasting reactions of fathers to the
death of their sons; the first pair concerns two Romans, Pollio
and Haterius, the second, two Greeks, Lesbocles and Potamon.
These anecdotes are accompanied by reflections which may be com-
pared with similar sentiments in the younger Senecafs Ep. mor. 63.
In Contr. IV pr. 6, recalling Polliofs fortitude in not withdraw-
ing from his normal social life even on the day of his son's
death, he marvels: o magnos viros, qui fortunae succumbere ne-
sciunt et adversas res suae virtutis experimenta faciunti With a
similar ideal in mind the younger Seneca remarks in Ep. mor. 63.1:
illud, ut non doleas, vix audebo exigere; et esse melius scio.
sed cui ista firmitas animi continget nisi iam multum supra
fortunam elato? We see from the elder Seneca's recollections
about Fabianus (Contr. II pr. 2) that it was assumed by the
Sextians, to whose sanctis fortibusque praeceptis (ibid. 1) he had
dedicated himself, that emotions ought to be suppressed; they also
taught that one should fight adversity with virtues (Sen. Ep. mor.
59.7). Both the Senecas, however, had their doubts whether such
an ideal was attainable by ordinary people, or indeed wholly ad-
mirable: on Lesbocles and Potamon the elder Seneca remarks,
utriusque tamen adfectum temperandum puto: hie durius tulit
fortunam quam patrem decebat, ille mollius <quam> virum (Suas. 2.
15), and his son, consoling Lucilius on his bereavement in Ep.
mor. 63.1, similarly feels it allowable for a man to succumb to
his emotions in moderation: nobis autem ignosci potest prolapsis

ad lacrimas, si non nimiae decucurrerunt, si ipsi illas repressi-
mus. nee sicci sint oculi amisso amico nee fluant; lacrimandum
est, non plorandum. These are commonplace sentiments no doubt,
for it was hardly possible for a Roman to write about bereavement
without reference to the conventional topics of consolatio - the
poor younger Seneca cannot recall his deep grief at the death of
Annaeus Serenus (Ep. mor. 63.14) without envisaging himself inter
exempla...eorum quos dolor vicit - but it is interesting nonethe-
less to see how closely the attitudes of father and son correspond
over this matter.
Distaste for contemporary luxury was so commonly expressed by
thinkers, serious and not so serious,12 in the early Empire, that
nothing precise can be determined about the elder Seneca's philo-
sophical position on the basis of his tirade on the subject in
Contr. I pr. 7ff., but certainly both Sextius (Sen. Ep. mor. 108.
18) and Attalus (ibid. 23) held strong views on the subject. As
for the respect in which the elder Seneca most differed from his
second son, his non-participation in public life (whether the
result of conscious choice or force of circumstances) and his
approval of Mela's decision to live free from desires and thereby
safe, to a large extent, from the whims of Fortune (Contr. II pr.
3: perge quo inclinat animus, et paterno contentus ordine subdue
fortunae magnam tui partem), it is interesting that the career of
the elder Sextius provided a classic precedent for the choice of
such a life: honores reppulit pater Sextius, qui ita natus ut rem
publicam deberet capessere, latum clavum divo Iulio dante non
recepit; intellegebat enim quod dari posset et eripi posse (Sen.
Ep. mor. 98.13).
Velleius Paterculus makes no mention of any great period of
Roman philosophical writing in his remarks on the brief flowering
periods of the literary genres. Probably he thought Rome's philo-
sophical achievements totally unworthy of comparison with those of
Greece; such a view would be in line with his opinion (1.16.4)
that no Greek philosophers worth mentioning had flourished much
after the time of Plato and Aristotle. Velleius does, however,
express an opinion as to the limits of the great age of Roman
poetry: with the exception of tragedy and comedy which flowered

early (1.17.1), Rome's only poetry of note, he considers, had been
composed in those same seventy odd years, up to the death of Livy,
which had produced the best Roman historiography (ibid. 2: nee
poetarum in antiquius citeriusve processit ubertas). And in the
case of poetry as well as historiography none of the elder Seneca's
Latin quotations antedates that golden age whose limits Velleius
was so early able to discern.
The only Greek poet to receive mention in the elder Seneca's
extant books is Homer (set-book of the Greek grammarians).13
Homeric quotations and paraphrases were sometimes included in
Greek declamations (Contr. 1.8.15; X.2.18; Suas. 1.12), and oc-
casionally, it would seem, in Latin ones too (Contr. 1.7.14).
Wits, Greek and Roman, made effective use of Homeric tags (Contr.
VII.7.19; IX.3.14; Suas. 7.14). But we find no real Homeric
criticism in the elder Seneca's books, unless we count Maecenas'
discussion of literary magnitudo in Suas. 1.12, which was initially
prompted by the failure of the declaimer Dorion to paraphrase
Homer without tumidity.11*
In striking contrast to Cicero, the elder Seneca never quotes
from, or even (if we except Cassius Severus' reference to Pomponium
Atellanarum scriptorem in Contr. VII.3.9) alludes to, any early
Roman dramatists or epic poets. This can scarcely be because he
had not read them: his school-days belonged to a time when their
works, and not yet those of Virgil and Horace, were the staple fare
of the pupils of Latin grammatici, and from them, presumably, came
the verses which he memorized with such facility as a boy (Contr.
I pr. 2 ) .
Several late Republican poets active within Velleius' 'less
than eighty years' receive mention, but all in passages where
Seneca is more or less certainly quoting other critics. Cicero's
poetry is contemptuously dismissed in Contr. Ill pr. 8 by Cassius
Severus: Ciceronem eloquentia sua in carminibus destituit. The
mimicus Publilius Syrus has his reputation vindicated in Contr.
VII.3.8f. by the same critic, who also mentions in passing Lab-
erius, another mime-composer active in the late Republic (ibid.
9). In Contr. VII.1.27 Julius Montanus suggests that some fine
lines by Varro of Atax were the inspiration for some even finer
ones by Virgil, which Cestius had failed miserably to emulate in
prose:
Montanus Iulius, qui comes fuit <Tiberii>,15 egregius poeta,
aiebat ilium imitari voluisse Vergilii descriptionem:
nox erat et terras animalia fessa per omnis
alituum pecudumque genus sopor altus habebat. (Aen. VIII.26f.)
at Vergilio imitationem bene cessisse, qui illos optimos versus
Varronis expressisset in melius:
desierant latrare canes urbesque silebant;
omnia noctis erant placida composta quiete. (Fr. 8 Morel)
(Note that Montanus does not, as a modern commentator would, point
to Apollonius Rhodius, Arg, III.749f.,
ou6e xuv&v uAaxn E T ' dva TITOXLV, OU $poos fiev
nx^^S* cruyri 6e ueAauvouevnv ex £V 6'pcpvnv,
as the original which lay behind Varrofs description of night.)
Words from poems by Catullus and Calvus, as we have seen,16 are
quoted in the appreciation of the latterfs oratory in Contr. VII.
4.7.
But the Latin poets most often quoted by the elder Seneca are
Virgil and Ovid. Of Virgil we are told by Cassius Severus that
his eloquence deserted him in prose, just as Cicerofs deserted him
in poetry (Contr. Ill pr. 8 ) . We are possibly indebted to the
elder Seneca rather than to his son for some remarks by Julius
Montanus on Virgilfs delivery recorded in Donat. Vit. Verg. 29.
We certainly owe him a debt of gratitude for the evidence recorded
in Suas, 1.12 of how Maecenas appreciated VirgilTs avoidance of
bombast. Seneca himself shows an admiration for Virgilian
17
beauties of style when he expresses a preference for Aen, XI.
28Off. over lines on a related theme by one Abronius Silo, even
though these contained an imitation of a sententia by Latro, si
nihil aliud, erimus certe belli mora (Suas. 2.19):
postea memini auditorem Latronis Abronium Silonem...recitare
carmen in quo agnovimus sensum Latronis in his versibus:
ite agite, <o> Danai, magnum paeana canentes,
ite triumphantes: belli mora concidit Hector.
...sed ut sciatis sensum bene dictum dici tamen posse melius,
notate prae ceteris quanto decentius Vergilius dixerit hoc quod
valde erat celebre, 'belli mora concidit Hector1:
quidquid ad adversae cessatum est moenia Troiae,
Hectoris Aeneaeque manu victoria Graium
haesit.
Messala aiebat hie Vergilium debuisse desinere: quod sequitur:

et in decimum vestigia rettulit annum
explementum esse; Maecenas hoc etiam priori conparabat. (Ibid.
19-20 W)
Once again it seems that Seneca had heard a discussion of Virgil
in which Maecenas had participated. He praises Virgilian lines
again in Suas. 3.5, this time verses from the Georgics (I.427ff.),
which he considers to have been expressed simplicius et beatius
than the prose paraphrase of them which Fuscus had contrived in
the hope of pleasing Maecenas.18
Virgil, like Homer, was not only imitated by declaimers, as
for instance by Cestius (Contr. VII.1.27) and Fuscus (Suas. 3.1,
4f.), but was sometimes quoted by them verbatim, as by Fuscus
and one of his pupils about whom we are told an amusing anecdote
in Suas. 4.4f.:
declamitarat Fuscus Arellius controversiam de ilia quae, postquam
ter mortuos pepererat, somniasse se dixit ut in luco pareret...
<cum> declamaret et a parte avi non agnoscentis puerum tractaret
locum contra somnia et deorum providentiam et male de magnitudine
eorum dixisset mereri eum qui illos circa puerperas mitteret,
summis clamoribus ilium dixit Vergili versum:
scilicet is superis labor est, ea cura quietos
sollicitat. (Aen. IV.379f.)
auditor Fusci quidam, cuius pudori parco, cum hanc suasoriam de
Alexandro ante Fuscum diceret, putavit aeque belle poni eundem
versum et dixit:
scilicet is superis labor est, ea cura quietos
sollicitat.
Fuscus illi ait: si hoc dixisses audiente Alexandro, scisses apud
Vergilium et ilium versum esse:
capulo tenus abdidit ensem. (Aen. 11.553)
From this story and from the jokes in Suas. 3.6f. based on the
i q*
allegedly Virgilian phrase, plena deo,1^ it emerges that Virgil

was already sharing the fate of Homer in providing useful quo-
tations for the repertoire of wits.
Ovid, himself once a student in the schools of declamation
(Contr. II.2.8ff.), also in due course shared Virgil's privilege
of having his poetry imitated in declamatory prose. P. Vinicius,
a great admirer of Ovid, recommended that one should store in
one's memory ad fingendas similes sententias an epiphonema from
the Metamorphoses (XII.6O7f.):
quod Priamus gaudere senex post Hectora posset,
hoc fuit. (Contr. X.4.25)
In Contr. III.7 extra, Cestius disapproves of a sententia by

Alfius Flavus which he alleges to have been based on lines by
Ovid:
Alfius Flavus hanc sententiam dixit: ipse sui et alimentum erat et
damnum. hunc Cestius quasi corrupte dixisset obiurgans: apparet,
inquit, te poetas studiose legere: iste sensus eius est qui hoc
saeculum amatoriis non artibus tantum sed sententiis implevit.
Ovidius enim in libris metamorphoseon dicit:
ipse suos artus lacero divellere morsu
coepit et infelix minuendo corpus alebat. (Met. VIII.877f.)
Ovid's poems, like Virgil's, provided wits with apt quotations.
Thus Scaurus silences Murredius by quoting from one of his
Priapeia (Contr. I.2.22). 20
As we have seen,21 there is much of interest to be learnt
about Ovid both as poet and as declaimer from Contr. II.2.8ff. and
other passages in Seneca's anthology. It comes as no surprise to
be told in Contr. II.2.12 that he preferred suasoriae to contro-
versiae , only declaiming the type of controversiae known as
ethicae of which Contr. II.2 (about the suicide pact of a married
couple) is presumably an example. We also learn that his argumen-
tation in this controversia had no clear structure (ibid. 9: sine
certo ordine per locos discurrebat) and that in general molesta
illi erat omnis argumentatio (ibid. 12) - facts which are of
interest to the reader puzzled by the strange order in which
topics are sometimes presented in, for instance, the Heroides,
though they should not necessarily always be relied on as a de-
fence of the received text where hopeless disorganization seems to
reign. An amusing anecdote is told in Contr. II.2.12 about Ovid's
fondness for lapses of taste in his poetry:
verbis minime licenter usus est nisi in carminibus, in quibus non
ignoravit vitia sua sed amavit. manifestum potest esse, quod
rogatus aliquando ab amicis suis ut tolleret tres versus, invicem
petit ut ipse tres exciperet in quos nihil illis liceret. aequa
lex visa est; scripserunt illi quos tolli vellent secreto, hie
quos tutos esse vellet: in utrisque codicillis idem versus erant,
ex quibus primum fuisse narrabat Albinovanus Pedo, qui inter
arbitros fuit:
semibovemque virum semivirumque bovem (Ars am. 11.24);
secundum:
et gelidum Borean egelidumque Notum (Am. 11.11.10);
<tertium: ...> 22 *
ex quo adparet summi ingenii viro non iudicium defuisse ad com-
pescendam licentiam carminum suorum sed animum. aiebat interim
decentiorem faciem esse in qua aliquis naevos esset.

In addition, Ovid's taste for brevity in description is evidenced
by his suggestion, quoted in Contr. VII.1.27, that certain lines
of Varro's would benefit from pruning. His habit of composing
several consecutive sententiae on the same theme is seen from
Contr. IX.5.15ff. to have been criticized by his contemporaries
and to have had parallels in the speeches and declamations of
Votienus Montanus. Unfortunately Seneca does not make it clear if
the poet influenced the orator or vice versa, or whether their
resemblance was merely coincidental, for Scaurus, whose remarks on
the subject are quoted by Seneca, was as ready to refer to Montanus
as inter oratores Ovidium as he was to give OvidTs repetitions the
name of Montaniana (ibid. 17). That Ovid frequently included con-
scious imitations of Virgil in his poetry, non subripiendi causa,
sed palam mutuandi, hoc animo ut vellet agnosci (Suas. 3.7 W) was
attested by his friend Gallio.
Curiously lacking in Seneca's range of references to Roman
poetry is any direct quotation from Lucretius, the earlier el-
egists, or Horace. However, Seneca may well have had in mind the
lines of Lucretius,
sed veluti pueris absinthia taetra medentes
cum dare conantur, prius oras pocula circum
contingunt mellis dulci flavoque liquore (I.936ff.= IV.llff.)
when on one occasion he referred to the way in which children have
to be tricked into drinking medicine:
et quia hoc <propo>situm recta via consequi non potero, decipere
vos cogar, velut salutarem daturus pueris potionem. sumite pocula.
(Suas. 6.16 M ) 2 3 *
At any rate Quintilian, wishing in III pr. 4 to use the same image,
actually quoted the Lucretian lines. Again, Seneca includes in
his criticisms of Haterius some words, cum torrentis modo magnus
quidem sed turbidus flueret (Contr. IV pr. 11), which remind one
of Horace's reference to the muddy flow of Lucilius, cum flueret
lutulentus (Serm. I.4.11).21+
The elder Seneca's taste in modern poetry extended to an
admiration for several of the lesser epic poets of his day, in-
cluding some who certainly or probably survived into the reign of
Tiberius. Julius Montanus, who is described as an egregius poeta
in Contr. VII.1.27 was actually a friend of Tiberius: the younger

Seneca, less enthusiastic about him than his father, calls him
tolerabilis poeta et amicitia Tiberi notus (Ep. mor. 122.11).
As for the types of modern epic poetry alluded to by the
elder Seneca, it is interesting that whereas there is one quo-
tation (the one by Abronius Silo, Latrofs pupil, deemed in Suas.
2.19 bene dictum but inferior to similar lines in Virgil) 25 which
comes from an epic concerned wholly or in part with the Trojan
Wars, his other four extracts from minor hexameter poets come from
epics about Roman history of the post-heroic era. Was the prefer-
ence of Lucan foreshadowed in his grandfather's taste in modern
poetry?
Cornelius Severus is quoted on two occasions, once for lines
describing the Roman army on the eve of a crucial battle (Suas.
2.12) and again for a passage lamenting the death of Cicero (Suas.
6.26). Seneca has doubts whether the lines in Suas. 2.12 do jus-
tice to the greatness of the Roman people, but he admires none-
theless the poet's telling use of colloquial language.26 He has
no hesitation in preferring this poet's lament for Cicero to any
treatment of the same theme by a prose historian, and, given his
high regard for historiography, this is praise indeed: nemo tamen
ex tot disertissimis viris melius Ciceronis mortem deploravit quam
Severus Cornelius (Suas. 6.25). A propos of one of the lines in
1
Severus lament, conticuit Latiae tristis facundia linguae, Seneca
notes that it was inspired by another by one Sextilius Ena, a
Corduban poet whom he describes as homo ingeniosus magis quam
eruditus, inaequalis poeta et plane quibusdam locis talis quales
esse Cicero Cordubenses poetas ait, <pingue> quiddam sonantis
atque peregrinum (Suas. 6.27). Reciting this line, deflendus
Cicero est Latiaeque silentia linguae, in the house of Messala on
one occasion Ena had caused annoyance to Pollio, though he evi-
dently pleased Cornelius Severus, who was also a member of the
audience:
Pollio Asinius non aequo animo tulit et ait: 'Messala, tu quid
tibi liberum sit in domo tua videris; ego istum auditurus non sum,
cui mutus videor1, atque ita consurrexit. Enae interfuisse
recitationi Severum quoque Cornelium scio, cui non aeque displi-
cuisse hunc versum quam Pollioni apparet, quod meliorem quidem sed
non dissimilem illi et ipse cofliposuit. (Ibid.)

A quotation from a poem by Albinovanus Pedo in Suas. 1.15 has as
its theme an episode of even more recent Roman history, the North
Sea voyage of Germanicus.27 Seneca regarded the lines quoted from
Pedo's poem as superior to any of the declaimers1 attempts to
describe Oceanus:
Latini declamatores in descriptione Oceani non nimis viguerunt;
nam aut minus descripserunt aut <nimis> curiose. nemo illorum
potuit tanto spiritu dicere quanto Pedo... (Ibid.)
Both of the elder Seneca's extended quotations from epic, Cornelius
Severus1 lament for Cicero and Pedofs description of the North Sea,
clearly belong stylistically to the beginnings of Silver Latin.28
That he praises both these quotations highly makes it questionable
whether he would have agreed entirely with Velleius1 opinion that
no Roman poetry of note had been composed since the time, towards
the end of the second decade A.D., which was marked by the death
of Ovid as well as that of Livy, in Velleius1 view Rome's last
great historian.
The almost total absence of references to early Latin litera-
ture in the elder Seneca's books probably reflects a decided mod-
ernity in his literary outlook. It is not safe to infer, however,
that such modernist disregard for the past in a man of his time
was the outcome of a preoccupation with contemporary school rhet-
oric. Seneca the Elder belonged to a generation which had re-
ceived in the schools of grammatici all too thorough a grounding
in the Latin poetry of the third and second centuries B.C. Posi-
tive aversion to such literature was expressed most famously by
Horace, born in 65 B.C., that is, perhaps no more than ten years
before Seneca the Elder, and brought up on Livius Andronicus in
the school of the fearsome Orbilius (Epist. II.1.69ff.). There is
no record that Horace attended any school of rhetoric, and in
Epist. 1.2.Iff. he seems anything but envious of Lollius, his
addressee, whom he imagines busy declaiming in Rome while he him-
self is re-reading Homer at Praeneste. Horace's preference for
modern over early Roman poetry was a conscious choice, which he
justified in Epist. II.1.18ff. in the face of opposition from
enthusiasts for the old poets, who included, it seems, not just
academic archaizers, but a large theatre-going public (ibid. 60-3).

Horace's view of the poems he had learnt at school was that,
though they did not deserve to be consigned to the bonfire, they
were far from being the flawless masterpieces people sometimes
made them out to be:
non equidem insector delendave carmina Livi
esse reor, memini quae plagosum mihi parvo
Orbilium dictare; sed emendata videri
pulchraque et exactis minimum distantia miror;
inter quae verbum emicuit si forte decorum, et
si versus paulo concinnior unus et alter,
iniuste totum ducit venditque poema. (ibid. 69ff.)
Thus, the elder Seneca was in good company if, in adult life, he
chose to turn his attention towards poetry more polished than any
he had learnt at school. Similarly, if he had no taste for early
Roman historiography, there were very respectable precedents for
such an attitude in the criticism of early annalists in Cicero's
De oratore (II.12.51ff.), and in Livy's disdain for archaizing
stylists (Contr. IX.2.26). In the case of Roman philosophy it
was hardly possible to argue by analogy with Greek literary his-
tory (cf. Hor. Ep. II.1.28ff.) that old was best, seeing that
philosophical Latin was so largely the recent creation of Cicero
and Lucretius. As for oratory, Cicero's Brutus provides ample
evidence that already among the literary men of the late Republic
interest in archaic Latin speeches was negligible. The only Latin
speeches known by Cicero to be extant which were earlier than the
elder Catofs were apparently Appi Caeci oratio de Pyrrho et non
nullae mortuorum laudationes (Brut. 16.61), none of which he
imagines would appeal to his contemporaries. Even the speeches of
Cato, who died, as Cicero points out, 'annis LXXXVI ipsis ante me
consulem* , and whom it was therefore wrong to think of as belonging
to remote antiquity (ibid. 15.61), were very seldom read any
longer: Catonem vero quis nostrorum oratorum, gui quidem nunc sunt,
legit? aut quis novit omnino (ibid. 17.65). It is evident that
Cicero's researches into the oratory of the second century B.C.
did not represent a kind of activity normal among men of his time.
Yet nobody could mistake the Brutus for an archaizer's manifesto.
Cicero was perhaps the most firmly convinced of all late Republican
critics that modern Latin oratory, and especially his own, was
superior to all that had come before it. As we have seen,29 he

declares in Tusc. Disp. II.1.5 that the glory of oratory had risen
ab humili till it had arrived in his own day ad summum.
The distaste for the archaic felt by modernists in the age of
Augustus, as in the late Republic, was clearly the product of a
positive and eminently justifiable pride in the excellence of
recent Roman literary achievements. It did not owe its origins to
narrowness in the teaching of Latini rhetores. However, consider-
ably before the end of what Velleius considered to be the great
age of Roman poetry and historiography, we can detect certain new
tendencies in the writings of authors educated during or after the
Civil Wars, which may surely be suspected of having originated in
the schools of declamation. The elder Senecafs passing criticisms
are occasionally helpful in confirming that it was indeed from the
rhetorical schools that these tendencies are likely to have arisen.
It is a commonplace in the criticism of early Imperial litera-
ture that 'the most outstanding influence discernible in the 1st
century, largely determining its character, is...the rhetorical
element1.30 Of course, no Roman writing with any pretensions to
artistry had ever been entirely unrhetorical, but never before had
techniques of deliberative and judicial rhetoric, employed with
such an unremitting search after effect, been so generally preva-
lent in all the literary genres. Particularly characteristic of
this new literature are tight argumentative sequences each capped
by an epiphonema. In addition, moralizing loci, exempla, and
descriptiones are repeatedly drawn from what one senses is a
fairly limited stock-in-trade.
Probably these tendencies are to be explained largely by a
change in educational custom.31 Horace attended the school of a
grammaticus, but there is no evidence that he went on to study
with a rhetor. The latter stages of Virgil's education, according
to Donatus* Life (§15), were devoted chiefly to the study of medi-
cine and, more especially, mathematics. Rhetoric may have been
among the cetera studia which he touched on, but Donatus tells us,
immediately after his remarks on Virgil's education, that his one
appearance as an orator was a failure. When Virgil and Horace
were growing up it was probably still not thought normal to study
rhetoric unless one was destined for public life; the old Roman
attitude described by Antonius in the De oratore (11.13.55) may
well have still prevailed: nemo...studet eloquentiae nostrorum
hominum, nisi ut in causis atque in foro eluceat.
It seems, however, that in the generations which succeeded
that of Horace it became increasingly abnormal for a Roman with
any pretensions to culture to have been without a period of train-
ing in one of the schools of rhetoric. It was admittedly because
his father wished him to enter public life that Ovid, the leading
poet of the first generation to be educated under the Principate,
was made to study rhetoric (Trist. IV.10.21ff.). More remarkable
is the case of Fabianus who, some time after he had begun studying
rhetoric, decided that his mission in life was to be a philosopher.
Amazingly, even though the praecepta which he embraced were of a
kind which militated against the emotional ostentation required of
declaimers, he continued to declaim conscientiously for some
time after his philosophical conversion (Contr. II pr. 4 ) . In
doing this, Fabianus indicated his acceptance of the educational
principle, in which Seneca the Elder (and doubtless many of his
contemporaries) firmly believed, that a rhetorical training - that
is, given the limited curriculum of early Imperial schools, prac-
tice in the declamation of controversiae and suasoriae - was of
value not only to the future orator, but also to students whose
intention was to distinguish themselves in artes quite remote from
the oratory of the forum: facilis ah hac in omnes artes discursus
est; instruit etiam quos non sibi exercet (Contr. II pr. 3). The
assumption, which had originated in Greece and particularly in the
school of Isocrates (see Cic. De or. 11.13.55ff.), that rhetoric
could provide useful techniques for other types of writers besides
orators, was at last becoming accepted in Rome.
That did not mean that everyone agreed that declamation pro-
vided a good training even for prospective orators. As we have
seen, Seneca quotes two most virulent attacks on this assumption
in the third and ninth prefaces.32 He also tells several anec-
dotes in which orators are recorded to have behaved as if they
were in a school-room rather than in a law-court, with results
ranging from the amusing to the disastrous. The bystanders at the
trial reported in Contr. II.1.34ff. were treated to a comic alter-

cation between counsel for the prosecution and the defence about
the rival theories of Apollodorus of Pergamum and Theodorus of
Gadara relating to narrationes; they expressed their appreciation
magnis...clamoribus.33 No doubt among them there were many
habitues of the schools of rhetoric, interested, as Seneca the
Elder was, in the ingenious colores put forward by the defence in
this difficult and sensational case. Albucius felt obliged to
retire from his forensic activities after a figura iurisiurandi he
had incautiously used was taken literally by his opponent, who
thereupon secured his clients acquittal (Contr. VII pr. 7), 3l+ and
how out of touch with reality a scholasticus might become is
further revealed by the story, related in Contr. VII.5.12, that
when Votienus Montanus had a charge brought against him by the
citizens of Narbo, he greeted his prosecutor's speech with
pleasure, and singled out sententiae from it for special praise,
thereby prompting the wit Surdinus to ask, 'numquid putas ilium
alteram partem declamasse?1 Such anecdotes, together with the
invectives of Cassius Severus (Contr. Ill pr. 12ff.) and the
critic quoted in the ninth preface, serve to prove that declama-
tory methods and attitudes were starting to infiltrate into the
law-courts.
It was the elder Seneca's view that a rhetorical education
could benefit a future philosopher (Contr. II pr. 3f.). Yet the
declamations which Fabianus so assiduously delivered in the school
of Blandus after his philosophical conversion were not the theses
which Cicero had considered the appropriate sort of themes for a
philosopher to practice upon in order to develop fluency:35 they
were controversiae and suasoriae. There was certainly scope for a
certain amount of philosophizing in these types of themes: large
ethical questions were often raised in the tractationes aequitatis
in controversiae, and sometimes declaimers of suasoriae devoted
the whole of their argument to the consideration of such questions
as whether the gods intervened in human affairs (Suas. 3.3) or
whether it was possible to predict the future (Suas. 4.4). It
would not be fair then to suggest that the teaching of the rhetor-
icians in Seneca's day constituted a complete rejection of Cicero's
notion that the ideal orator should always seek to isolate the
abstract issue at stake in any dispute in which he had to partici-
pate (Or, 14.45: a propriis personis et temporibus semper, si
potest, avocet controversiam), even though Cicerofs works on rhet-
orical theory seem to have had precious little influence on the
curriculum of the schools of that time.
However, the reason why Seneca the Elder believed declamation
could be of benefit to philosophers cannot have been that he con-
sidered there was scope in it for extensive practice in philo-
sophizing. For, doubtless sharing with Latro (Contr. VII.7.10) a
general preference for tautness in the exposition of loci, he con-
demns in Albucius' private declamations ilia intempestiva in
declamationibus eius philosophia which sine modo tune et sine fine
evagabatur (Contr. VII pr. 1). From the anecdote which he tells
about Albucius and Cestius in Contr, VII pr. 8 we can see that
Seneca's view that philosophizing in declamation ought to be kept
within proper limits was not totally unreasonable:
nee in scholasticis tamen effugere contumelias poterat Cestii,
mordacissimi hominis. cum in quadam controversia dixisset
Albucius: quare calix si cecidit frangitur, spongia si cecidit non
frangitur? aiebat Cestius: ite ad ilium eras; declamabit vobis
quare turdi volent, cucurbitae non volent.
Cicero himself, surely, would not have considered the subject of
Albucius1 speculation a suitable topic for rhetorical treatment.
Nevertheless, intempestiva seems quite a strong word with which to
criticize the large philosophical element in this rhetorician's
declamations. It therefore seems appropriate to consider the
comment quoted from Contr. VII pr. 1 in relation to that anti-
Ciceronian tradition in Imperial rhetorical theory supported by
Aper in Tac. Dial. 19.Iff. and condemned by Quintilian in III.5.
12ff.
Aper argues that, now that philosophical knowledge had become
widely disseminated among the Roman public, the treatment of ab-
stract issues in speeches would not have the same effect on
listeners as it had in the old days when, si quis odoratus philo-
sophiam videretur atque ex ea locum aliquem orationi suae in-
sereret, in caelum laudibus ferebatur (Dial. 19.3). The represen-
tatives of this tradition met by Quintilian do not seem to have
presented nearly so intelligent a rationale as that of Aper for

their view that universales quaestiones were useless in oratory
(Quint. III.5.12). It was on a bogus authority that they placed
their reliance, namely some books falsely ascribed to Hermagoras
of Temnos (ibid. 14). They would have done better to have cited
the passage in the De inventione (1.6.8) where Cicero as a young
man had argued against the true Hermagoras1 inclusion of deaeus
within the province of rhetoric, even though he was later to
change his mind completely over this matter.36
We see from De inv. 1.6.8 that one did not need to be a hater
of philosophy to think it unsuitable to discuss purely abstract
issues in oratory and, though the followers of ps.-Hermagoras en-
countered by Quintilian were doubtless very narrow men, it is not
justifiable to assume that Seneca the Elder was like them in
every respect. After all, his sons received a philosophical as
well as a rhetorical education. It just seems that he thought
expansive philosophizing did not mix well with declamation.
Eloquence, on the other hand, seemed to him an excellent quality
in a philosopher: Attalus is praised in Suas, 2.12 as magnae vir
eloquentiae. Presumably the reason why he considered that rhet-
orical training had benefited Fabianus was that it had enabled him
to expound his precepts persuasively to a public accustomed to
ornate rhetoric, whose reaction to a crabbed style of dialectic
would have been that expressed by Antonius in Cicerofs De oratore:
in philosophos vestros si quando incidi, deceptus indicibus
librorum, qui sunt fere inscripti de rebus notis et inlustribus,
de virtute, de iustitia, de honestate, de voluptate, verbum
prorsus nullum intellego; ita sunt angustis et concisis disputatio-
nibus inligati. (11.14.61)
About the possible value of rhetorical training for a his-
torian, such as he was himself, the elder Seneca says nothing
whatsoever, beyond praising one of Gallio's declamatory sententiae
in Suas. 5.8 as worthy of inclusion in a historia9 though presum-
ably history was one of the disciplines he had in mind when he
claimed in Contr. II pr. 3 that there was an easy path from rhet-
oric in omnes artes. He does, however, tell us several interesting
facts about certain historical fabrications which circulated in
the schools of rhetoric. He traces the origin of the theme of
Suas. 7, in which Cicero is faced with the choice of death or the

burning of his books, to a malicious passage in Pollio's Pro
Lamia:
nam quin Cicero nee tam timidus fuerit ut rogaret Antonium nee tam
stultus ut exorari posse eum speraret nemo dubitat, excepto Asinio
Pollione, qui infestissimus famae Ciceronis permansit. et is etiam
occasionem scholasticis alterius suasoriae dedit; solent enim
scholastici declamitare: deliberat Cicero an salutem promittente
Antonio orationes suas comburat. haec inepte ficta cuilibet
videri potest. Pollio vult illam veram videri; ita enim dixit in
ea oratione quam pro Lamia edidit. (Suas, 6.14f.)
On the other hand it is the declaimers who are blamed for the
invention of a story that Popillius, the man named by a few, but
not all, historians, as the assassin of Cicero, had once been de-
fended by the great orator when faced with a charge of parricide
(Contr. VII.2.8). 37 However, as none of the historians known to
Seneca took over the story of the parricide trial, we cannot use
this intriguing piece of source criticism as evidence that fictions
were being imported from declamatory themes into historical narra-
tive at this date, as they certainly were to be later, in the
Gesta Romanorum.^^ Seneca unfortunately leaves it unclear whether
the story, which was accepted by some historians, that Popillius
had been defended by Cicero on some lesser or unspecified39
charge, was of declamatory or historical origin.
That poets were coming under the influence of rhetoricians is
evident from his recognition of imitations of Latro in the verses
of Ovid (Contr. II.2.8) and Abronius Silo (Suas. 2.19). It is not
the case that the only qualities which Seneca the Elder admired in
poetry were the declamatory virtues: he detects in Pedo's lines on
the North Sea a certain spiritus which he did not find in any
declamatory descriptions of Ocean (Suas. 1.15); he is sensitive,
more sensitive than the grammatici, to the poignancy of a colloqui-
alism in a poem by Cornelius Severus (Suas. 2,12); he prefers the
simple felicities of lines in Virgil's Georgics to a convoluted
paraphrase of them by Arellius Fuscus (Suas. 3.5). We are also
reminded by Seneca that it was not only declamation that served in
his time to encourage ostentatiousness among writers: the custom
of recitatio to which he refers in Contr. IV pr. 2 and Suas. 6.27
was another important outlet for literary ambitio. The fact
remains that Seneca does not make an exception of poetry when he

asserts that rhetoric provides a training for 'all the arts1, and
that he seems not in the least disapproving when he recalls in
Contr. II.2.8 Ovid's indebtedness to the sententiae of Latro.
It seems to be implied in the assertion, facilis ah hac in
omnes artes discursus est (Contr. II pr. 3 ) , that rhetoric is the
central literary discipline, seeing that from it the other artes
are approached by travelling in diverse directions. In a very
real sense declamatio did occupy a central position in the liter-
ary culture of the elder Seneca's day. A rhetorical training
seems to have come to be regarded as an essential tertiary stage
in education, as it had not been in Rome at the dramatic date of
the De oratore, and probably still was not when Virgil and Horace
were boys. Some of the distinguished members of the older gener-
ation met by Seneca in the auditoria of the declaimers were little
disposed, at least when engaged on serious writing, to change the
stylistic habits which they had learnt in an earlier age, but
writers of the elder Seneca's own and subsequent generations had
been brought up from boyhood among the clangor of sententiae and
the rant of standard exempla and descriptiones, with results which
are widely apparent in the literature of the early Empire.
APPENDIX
Clausula usage of Seneca, Latro and Fuscus: a trial sample*
Seneca Latro Fuscus

I 16 12 13
II 12 11 18
III 10 13 15
IV 6 13 12
V 5 6 7
49 55 65
Resolutions
I Rl -w-v^w 10 3 4
R2 -^w-w 2 2 2
R3 *U\J<U\J\J — ^ 1 — —
R4 v^v^ w - 1 -
R5 ^^w-www 2 -
I I I Rl -w-^^ww 1 3 1
R2 uuu KJ±L 2 — 2
R3 -www-^^ - - 3
IV R «^^^-^w - 3 -
V Rl ^/^v^ 2 2 -
R2 ^^—wwv^^ ]_ — —
R3 -ww—^^ 1 1 2
R4 -uu-uuuy ^ ]_ —
R5 —v^^-ww - 1 1
23 17 15
Clausula usage of Seneca, Latro and Fuscus 327
Seneca Latro Fuscus

Also (cf. V)
" v-,w 2 3 1
Rl w-^-^ 1 2 -
R2 ^ wwuy — — 1
Spondaic, dactylic and anapaestic rhythms

16 13 6
7 4 2
1 3 4
- 1 1
- 1 -
1 - 2
- 1 3
25 23 18
100 100 100

NOTES

1
This is the least garbled version of the title, preserved
e.g. by BV in the colophon to the liber suasoriarum; we have
no way of being sure that it is Senecan, but it is difficult
to imagine anyone much later than Seneca1s time bothering to
draw attention to the fact that his work includes extracts by
oratores (i.e. amateur declaimers) as well as rhetores.
2
This is as precise a dating as can safely be deduced from
Contr. I pr. 11, cf. Miriam Griffin, fThe elder Seneca and
Spain', JRS LXII (1972), 5, and pace L.A. Sussman, The elder
Seneca (Leiden, 1978), 20.
3
See pp.lO4ff.
4
See pp.lO7ff.
5
The evidence is collected by S.F. Bonner, Roman declamation
(Liverpool, 1949), 16ff.; note particularly Cic. De or. II.
24.100, bearing in mind the dramatic date of the dialogue.
6
Suet. De gramm. et rhet. 7.3ff. Brugnoli.
7
For further discussion see pp.lO4ff.
8
Ad Q. fr. III.3.4, cf. E.G. Sihler, 'eexuxcoTepovT, AJPh XXIII
(1902), 283ff.
9
Though he had heard an older contemporary, Arellius Fuscus,
talking about his master's treatment of Contr. 1.7 (see §14),
and appears to accept that Cicero declaimed something very
like Contr. 1.4 (see §7).
10
On Corduba see M.T. Griffin, JRS LXII (1972), Iff.; in par-
ticular on the difficulties of travel from Corduba to Rome in
43 B.C. see Asinius Pollio's.letter to Cicero, Ad fam. X.31.
11
See Sen. Cons. Helv. 19.2 on the younger Seneca's apparently
very early removal to Rome; PIR2 A 611 on Lucan's childhood;
M.T. Griffin, JRS LXII (1972), 6 on the education of children
from leading provincial families in the late Republic.
12
Ps.-Plut. Vit. X orat. 836a, 848c.
13
Mrs Griffin, JRS LXII (1972), 6, and Sussman, The elder
Notes to pages 5-8 329
Seneca 20f. argue otherwise. It is not possible to date the

arrival of Seneca and Latro in Rome at all precisely. The
earliest datable anecdote from Seneca's Roman days is the
reference in Contr. II.2.9 to a declamation by Ovid, which
presumably was composed before 24 B.C. when his formal edu-
cation ended (Trist. IV.10.30), but, as I surmise (p.269),
after the publication of Tibullus' first book in 27 or 26.
However, it seems from Contr. II.2.8 that Latro was already
well established as a rhetorician by this time. On the other
hand, Sussman's inference (p.20) from Contr. I pr. 3 that
Seneca reached Rome while still a boy (puer) is invalid.
14
Suet. De gramm. et rhet. 7.3 Brugnoli.
15
Most Roman references to the blowing of a trumpet to mark the
time refer to the night watches, as M. Winterbottom notes ad
loc., but maybe the practice of sounding the trumpet three
times in the daylight hours mentioned, admittedly in a Greek
context, in Sen. Thyest. 798f. was not unknown in Rome.
16
Contr. IV pr. 2, cf. A. Dalzell, 'C. Asinius Pollio and the
early history of public recitation at Rome', Hermathena
LXXXVI (Nov. 1955), 20ff.
17
Cf. M.T. Griffin, JRS LXII (1972), 5. N.b. It may conceivably
have been in Corduba that Seneca first heard Pollio declaim
(Contr. IV pr. 3 ) , not in Rome as Sussman (The elder Seneca
22) and others have assumed.
* 8 The text of this sentence is discussed further later, pp.l2ff.

19
JRS LXII (1972), 9; 'Imago vitae suae', Seneca, studies in
Latin literature and its influence, ed. C.D.N. Costa (London,
1974), 5; Seneca (Oxford, 1975), 33f.
20 As was pointed out to me by Mrs Griffin per litt. 24.2.75.
21
To which we are obliged to resort for lack of any specialized
lexicographical work on Seneca the Elder which can be trusted.
22
ThLL s.v. I 3. See also n.48, p.337, on the failure of H.T.
Karsten to prove his contention that Latro used alioqui in
the sense alio tempore.
23
ThLL s.v. II 1,2. There appear to be only three occasions
besides Contr. II pr. 4 where the elder Seneca uses the word.
In Contr. I pr. 11, alioqui in illo atriolo...potui adesse,
it clearly means 'otherwise1. In Contr. IV pr. 8, regi autem
ab ipso non poterat; alioqui libertum habebat, cui pareret...,
Winterbottom obelizes alioqui, and clearly it does not serve
to refer us to a time or circumstance different from that tc
which the state of affairs described in the preceding clause
pertained; maybe, however, the reading alioqui is correct and
has the meaning 'at any event' (see OLD s.v. 2b; ThLL I lb).
The exact meaning of alioqui in Contr. IX.5.11 is uncertain,
but cannot be fat one time1: Gallio utrumque miscuit et hoc

colore, qui videri potest alioqui thema evertere, paratius
usus est.
2h
Per litt. 25.10.76. For the linking of weighty phrases of
this type with -que cf. especially Cic. Div. Caec. 20.65:
quern actorem causae suae socii defensoremque fortunarum
suarum potissimum esse voluerunt.
25
Tac. Ann. XVI.53, XVI.17, cf. Contr. II pr. 3.
26
For a detailed account of the evidence see M.T. Griffin, JRS
LXII (1972), 6-8. One of Mrs Griffin's general conclusions
(12ff.) about the elder Seneca is that he seems less deraci-
nated from his provincial background than his son the philo-
sopher.
27
See L. Casson, Ships and seamanship in the ancient world
(Princeton, 1971), 27Off.
28
See Cons. Helv. 18.9 for evidence that her father was still
alive in the 40s A.D.
29
Cons. Helv. 19.2. Probably literally as a babe in arms, but
on the ambiguity of ferre with reference to children going
for long journeys see Contr. IX.6.13.
30
Chronici canones ed. J.K. Fotheringham (London, 1923), 251.
31
Not necessarily just before PollioTs death in A.D. 5 as Mrs
Griffin (op. cit. 8) and Sussman (The elder Seneca 23)
assume.
32 See M.T. Griffin, loc. cit.
33
Contr. II.1.33; 5.11,13; III pr. 2; VII pr. 5; X pr. 8; Suas.
3.6.
34
E.g. Tac. Ann. XV.73; XVI.17.
35
But see M.T. Griffin, Seneca (Oxford, 1975), 43ff. for a dis-
cussion of the evidence for the activities in public life of
known connections of the Annaei Senecae in this period, and
an attempt to account for the late debut of Novatus and Seneca
in politics.
36
The punctuation and the substitution of tu for the MSS
reading ut were both suggested by E. Thomas, 'Schedae
criticae novae in Senecam Rhetorem1, Philologus Supptbd VIII
(1900), 232f. Modern editors prior to Winterbottom sadly
obscured the sense of the opening of the digression &nd the
connection between praise of Mela and the pen-portrait of
Fabianus, by misguided attempts to emend the words, hoc unum
studeas, to suit their own preconceptions. See especially
Millerfs apparatus, and his own version: [hoc unum con-
cupiscentem\ nihil concupiscere <nisi> ut eloquentiae tantum

studeas. Mtiller and the others must have been troubled by
the fact that Seneca here praises someone who has chosen
philosophic calm, whereas he is supposed, on the strength of
Ep. mor. 108.22, Cons. Helv. 17.4, to have hated philosophy.
They, and later writers on the elder Seneca, e.g. Bornecque,
Decl. 14, Edward, Suasoriae xxvi, Sussman, The elder Seneca
48, have wanted him to be praising Mela for single-minded
devotion to rhetoric. But this makes nonsense of the di-
gression in its context.
37
Seneca (Oxford, 1975), 33f.; cf. JRS LXII (1972), 9.
38 T
A propos d'un passage de Seneque le Pere (Contr, II pr. 4 ) :
la psychologie d'un pere ambitieux pour ses enfants au ler
siecle ap. J . C , Latomus XXXII (1973), 162ff.
39
Vassileiou, Latomus XXXII (1973), 162ff., sees in the bleak-
ness of the words, te in portu retineo, evidence that the
elder Seneca was less than enthusiastic about Mela's inten-
tions.
1+0
So Winterbottom in his translation and per litt. 19.11.75.
1+1
Per litt. 25.10.76.
1+2
Cf., e.g., Contr. I pr. 21, et in illo...; et nescio an...
for Seneca's use of et to link what amount to whole sentences
with different subjects.
3
Cf. Contr. I pr. 4, ita ex memoria... for the opening of a
sentence with ita in a sense equivalent to itague.
hh
See p.9.
45
Cf. Sen. Ep. mor. 14.13, and, for a detailed treatment of the
theme, Seneca's De otio.
1+6
Unlike the aridi declamatores of whom he remarks in Contr.
II.1.24 that no figure or sententia would distract them from
keeping faithfully to their chosen colores.
1+7
Tac. Ann. XVI.17.
1+8
Sen. fr. 98 Haase (from the younger Seneca's De vita patris) ,
quoted pp,15f.
1+9
So Bornecque, Decl. 24; Edward, Suasoriae xxvi; M.T. Griffin,
JRS LXII (1972), 4, 11; Sussman, The elder Seneca 92.
50
Fr. Hist. 2 Miiller, on which see p.17.
51
I.e. before the younger Seneca's exile (see Cons. Helv. 2.5).
Evidence for the place of his death discussed by Mrs Griffin
(op. cit. 8) is surely inconclusive.
52
Sen. fr. 98 init. Haase inserts scriptorum between puris and
titulis, unconvincingly. Could puris titulis perhaps mean
f
whose memorials boast no cursus honorum1?
53
The second part of fr. 98, continuing straight on from the
extract quoted above. After diem <perductas> or the like
seems to be required, cf. Contr. I pr. 13: ad ultimum eius
diem perductam familiarem amicitiam. As it is, the phrase
introduced by usgue ad dangles too loosely.
^ Much speculation there has been nonetheless. Sussman, The

elder Seneca 137ff. provides bibliography and an introduction
to the main arguments about its possible scope, time of com-
position and publication, content and sources.
55
Mrs Griffin, JRS LXII (1972), 10, argued against the attri-
bution of the historical fragments to the elder Seneca.
Sussman, The elder Seneca 137 is more ready to accept them as
remnants of the lost Histories.
56
The elder Seneca was not wholly funspeculative' as Mrs
Griffin (loc. cit.) suggests. What about his speculations on
the decline of eloquence (Contr. I pr. 7)?
57
Here I adopt Bornecque's conjecture derigant, RPh XXVI (1902),
373, as against Kiesslingfs derigam, preferred by MUller and
Winterbottom. The MSS divide between those reading redicam,
redigam ('full text1 MSS) and reducant, redigant or the like
(excerpta MSS). However, for arguments suggesting that the
'full text1 MSS are in general more to be trusted than the
excerpta see H. Hagendahl, 'Rhetorica', Apophoreta Gotoburg-
ensia V. Lundstr'6m oblata (GBteborg, 1936), 299ff.; M. Winter-
bottom, 'Problems in the elder Seneca1, BICS XXI (1974), 24ff.
58
See p.35.
59
E.g. Bornecque, Decl. 16ff.; Edward, Suasoriae xxvii ff.
6
^ Edward's translation (xxvii) of a phrase in Sen. Cons. Helv.
17.3.
61
Bornecque, Decl. 16. The assertion that Cato was his ideal
rests on Contr. I pr. 9 (part of the tirade on the decline of
eloquence) in which a famous dictum of Cato's, orator est,
Marce fili, vir bonus dicendi peritus is quoted with hearty
approval. But one does not deduce that Fortunatianus was
steeped in early Republican ideals just because at the begin-
ning of his Ars rhetorica (81.4f.) he adopts Cato's defi-
nition of orator; see p.83.
62
Edward, Suasoriae xxviii; cf. also now Sussman, The elder
Seneca 26ff.
^ 3 Cf. Ovid. Am. 1.8.43, casta est quam nemo rogavit.

64
Also compare Cons. Helv. 16.3 with Fabianus in Contr. II.5.7.
65
See p.308. On Seneca's misrepresentation of his father in
Ep. mor. 108 see now Sussman, The elder Seneca 21 f.
66
For the attitude cf. Juv. 6.434ff.
67
E.g. Bornecque, Decl. 16; Edward, Suasoriae xxviii; Sussman,
The elder Seneca 26f.
68
Edward, op. cit. xxix; cf. Bornecque, Decl. 17; Bonner, Rom.
decl. 147; Sussman, The elder Seneca 26.
69 JRS LXII (1972), 12ff.
70
Edward, Suasoriae xxix; Bornecque, Decl. 17; Bonner, Rom.
decl. 147.
71
See Contr. Ill pr. 17; IX.3.13.
72
Cf. Quint. X.I.105: oratores vero vel praecipue Latinam
eloquentiam parem facere Graecae possunt: nam Ciceronem
cuicumque eorum fortiter opposuerim.
73
Edward, Suasoriae xxix; Bornecque, Decl. 17.
7h
'Basic rhetorical theories of the elder Seneca1, CJ XXXIV
(1939), 350.
75
J. Buschmann, Charakteristik der griechischen Rhetoren beim
Rhetor Seneca (Parchim, 1878), Die 'enfants terribles' unter
den Rhetoren des Seneca (Parchim, 1883). In fact, in the
opening pages of the first paper he helped to promote the
error.
76
Adverse criticisms of Greek declaimers, but by no means
always severe ones, are included or implied in Contr. 1.1.25;
2.23; 4.7, 10; 6.12 (Winterbottom deletes et greca); 7.12;
8.7, 11; II.3.23; 6.13; VII.1.25; 4.10 (where, however,
Seneca is quoting other people's views); 5.15; IX.5.17
(reading sed genere corrupto); X.4.18, 22f.; 5.21-5, 27f.;
Suas. 1.12f., 16; 2.14.
77
For praise and other relatively favourable criticism of Greek
declaimers see Contr. 1.1.25; 4.7, llf.; 7.18; 8.16; II.1.39;
6.12 (ars inculta uncharacteristic of those under Greek
influence); IV pr. 7 (see n.80, p.334); VII.1.26; 5.14; IX.
2.29; X.4.18; 5.21f.; Suas. 2.14; 7.14.
78
Contr. 1.1.25; 2.22f.; 4.10-12; II.3.23; VII.1.25f.; 4.10;
5.14; IX.1.12f.; 5.16f.; X.4.18-23; 5.19-28; Suas. 1.13; 2.14.
79
So Mliller, following AB, which give the declaimer's name in
Greek l e t t e r s .
80
Cf. Contr. X pr. 3: hi caloris minus habent, neglegentiae non
minus. On facultas as a characteristic of the Greeks, rarely
reproduced by Latin declaimers, see Contr. IV pr. 7; for
licentia in Latin see Contr. II pr. 1; Suas. 2.10 (Arellius
Fuscus); Contr. II.2.7 (Marullus); ibid. 12 (Ovid's poetry).
81
See, e.g., Contr. X.5.28, quoted p.23.
82
See p.54.
83
See pp.308-10.
84
See Cic. De or. III.24.94; cf. Suet. De gramm. et rhet. 25.1
Brugnoli.
85
See pp.77ff.
86
See pp.305-7; 310-19.
The declamatory anthology

1
Cf. T. Janson, Latin prose prefaces (Stockholm, 1964), 7ff.
and, for discussion of Seneca's use of prefatory conventions,
Sussman, The elder Seneca 51ff.
2
C.W. Lockyer, ?The fiction of memory and the use of written
sources: convention and practice in Seneca the Elder and
other authors1 (Diss. Princeton, 1970), 195ff., gives further
examples of this device in Roman literature.
3
Latin prose prefaces 21.
4
Cited by Janson, op. cit. 61.
5
Quint. X.2.11, 24ff.
6
E. Rolland, De 1'influence de Seneque le p$re et les rheteurs
sur Seneque le philosophe (Ghent, 1906); C. Preisendanz, fDe
Senecae rhetoris apud philosophum filium auctoritate1,
Philologus LXVII (1908), 68ff.; C.S. Rayment, 'Echoes of the
declamations in the dialogues of the younger Seneca1, CB XLV
(1969), 51f., 63.
7
It is conceivable, however, that the format of his work had
been partially anticipated in some anthology of extracts from
the ten canonical Attic orators in ten books with a preface
to introduce each orator. Cf. Cic. Opt. gen. 6.17.
8
See also Contr. Ill pr. 18; IV pr. 11; VII pr. 9; X pr. 16.
9
There are a few exceptions, e.g. in Contr. 1.5.1; 1.7.10.
10
Some enthymemata might consist of two contrasting rhetorical
questions or exclamations (see p.204); see also Rhet. Her.

IV. 17.24 for sententiae quae dupliciter efferuntur-.
11
Bornecque's generalization (Decl. 33): fles Sententiae, ou
l f on cherche dans quelle mesure le cas pose tombe sous le
coup d f une loi donn§e', does not cover all we find in the
opening extracts of Seneca's surveys.
12
See pp.2O2ff. and Sussman, The elder Seneca 36.
13
Analysed on pp.251ff.
14
See also for example Contr. II.1.10-13; Suas. 3.1.
15
H. Bardon, Le vocabulaire de la critique litteraire chez
SGneque le Rheteur (Paris, 1940), 68ff. On Bardon1s attitudes
see my pp.68ff.
16
Seneca rarely uses color in the Ciceronian sense of Stylistic
colour1; see Bardon, op. cit. 19. On this different technical
usage, see my pp,166ff.
17
As I argue in CR N.S. XX (1970), llf., hunc sensum Vibius
Rufus...nisi nox defecisset should be transposed to fit
between impleri argumentis and hunc sensum a Latinis iactatum.
A similar case of displacement is perhaps to be found in
Contr. X.4.25. The words P. Asprenas...fortassis pater est
could very well have been placed originally after the
sententia by Blandus in Contr. X.4.20 which ends hie fortasse
meus est, but later have been omitted owing to a confusion of
like endings, and consigned to the end of the survey by a
corrector.
18 As Edward (Suasoriae 156) assumes.
19
They contain his latest historical allusion in the work
(Suas. 2.22), but there is no certainty that they were com-
posed last; cf. M.T. Griffin, JRS LXII (1972), 11. On the
liber suasoriarum see now also Sussman, The elder Seneca 69ff.
20 Cf. Janson, Latin prose prefaces 16, 25.
21
G. Boissier, Revue des deux mondes XI (1902), 480ff. trans.
W.G. Hutchison in Tacitus and other Roman studies (London,
1906), 163ff.; Bonner, Rom. decl. vii.
22
Vol. I x.
23
Bornecque, Decl. 28f.
2l+
'Literary sources in Cicero's Brutus and the technique of
citation in the Dialogue', AJPh XXVII (1906), 197 n.l.
25
'Wirklichkeit und Literaturform', RhM LXXVIII (1929), 114f.
26
Princeton, 1970. Sussman, The elder Seneca 75ff., whole-
heartedly accepts LockyerTs conclusions.
27
Lockyer, Diss. 170f.
28
Cf. Lockyer, Diss. 42ff.
29
Ibid. 142ff.
30
Bornecque, Decl. 25ff. citing M. Sander, Der Sprachgebrauch
des Rhetors Annaeus Seneca, 2 vols. (Berlin, 1877, 1880);
H.T. Karsten, De elocutione rhetorica qualis invenitur in
Annaei Senecae Suasoriis et Controversiis (progr. Rotterdam,
1881).
31
A.W. de Groot, Der antike Prosarhythmus (Groningen, 1921),
106f.
32
For example, unlike the analyst of verse metre, he has no
certain way of telling whether a naturally short syllable
before a mute + liquid is to be regarded as long or short,
and" he can never be sure whether the possibility of synizesis,
hiatus etc. is to be discounted in particular cases.
33
The samples were taken from the following passages: Seneca -
Contr. I pr. 1-16; II pr. 1-5; III pr. 1-7, 18; Latro - Contr.
1.1.1-3; 2.1; 3.1; 4.1; 5.1; 6.1; 7.1-2; 7.10; 8.1; II.1.1;
2.1; 3.1; Fuscus - Contr. 1.1.6; 2.5; 3.3; 4.5; 4.8; 5.2; 6.7;
7.5; 8.2; II.1.4-8; 2.1; 3.3; 3.9; 4.4; 4.5; 5.4. In making
this sample I aimed to select only sentence endings which
corresponded to major breaks in the sense. In doing so I did
not abide rigidly by the sentence division of any one editor,
though I was probably influenced to some extent by the punctu-
ation of Winterbottom, whose edition I used. Detailed stat-
istics, which however are not to be regarded as providing
more than preliminary soundings, will be found in an appendix
on pp.326f.
31+
See de Groot, Der antike Prosarhythmus 106, from which all
the figures given below for the frequency of clausula rhythms
in CiceroTs speeches and 'normal1 Latin usage are taken. On
de Grootfs methods see L.P. Wilkinson, Golden Latin artistry
(Cambridge, 1963), 139ff.
35
In a study of a larger sample one would be able to take
account of discrepancies in the frequency with which the less
common clausulae occur, in addition to the distribution of
the eight types considered here.
36
Decl. 25f.
37
See n.30 above.
38
Sprachgebrauch I 1.
39
Sander referred to Bursianfs edition by page and line in his
notes. I have omitted several apparently false references
from among the instances of normal usages which he cites for
the sake of comparison. I am indebted to Dr J. Mejer of
Copenhagen for the gift of a photocopy of this dissertation.
1+0
Sander, Quaestiones 5 n.4, adds: fpro idcirco ceteri rhetores
oh id, ob hoc' and refers us to another note, 38 (actually
36) n.l, in which he lists the following examples: ob id -
Contr. 1.1.13 (Latrofs divisio), 14 (Gallio's divisio); ob
hoc - Contr. I.I thema; II.6.4 (Fabianus), 5 (Latrofs
divisio); VII.6.13 (Latro's divisio); Suas. 2.21 (Seneca
paraphrasing a thema); propter hoc - Contr. 1.5.3 (Pompeius
Silo), 8 (Seneca reporting Fuscus); 7.16 (Cestius); X.I.10;
propter id - Contr. 1.8.8 (where id is conjectural).
14
* = Sander, Quaestiones 5 n.ll: cf. Suas. 6.13 (Varius GeminusT
divisio: iniquum + infin.); Contr. 1.6.5 (Iulius Bassus:
aequum + infin.).
1+2
= Sander, Quaestiones 5 n.12: (for licere + infin.) cf. Suas.
5.4 (Seneca reporting Cestius); Contr. 1.1.14 (Gallio's
divisio); IX.2.17 (Pompeius Silo^ divisio).
1+3
Sander, Quaestiones 6 n.l, gives examples of iubere + infin.
which it would be superfluous to reproduce.
1+4
= Sander, Quaestiones 3 n.l: cf. Contr. VII. 1.2 (Albucius:
aliter quam); 2.13 (Romanius Hispo: aliter non potuisse pacari
rem publicam quam...) .
1+5
= Sander, Quaestiones 6 n.4: (for lex + infin.) cf. Contr.
1.1.5 (Asprenas); II.3.10 (Marullus).
1+6
See n.30, p. 336 above.
hl
Eloc. rhet. 11.
1+8
At least in Contr. I pr. 11 Seneca uses alioqui not in the
sense ceterum as Karsten suggests in his entry s.v. (Eloc.
rhet. 16), but in the same sense, alio casu, in which Latro
customarily uses it; in none of the examples cited by Karsten
does it appear to bear the meaning alio tempore, as he claims
it sometimes does in Latrofs usage; there is a contradiction
between Karstenfs observation on p.10 'de peculiari signifi-
catione particulae alioquin apud Latronem* and the documen-
tation on p.16 s.v., from which it emerges that Gallio and
Hispo used alioqui to mean alio casu.
1+9
I have unfortunately not had access to the work of M. Cerrati,
La grammatica di A. Seneca il rhetore (Turin, 1908), but C.J.
Fordyce's description of it in his review of H. Bardon, Le
vocabulaire de la critique litteraire chez Seneque le Rheteur
in CR LX (1946), 129 as fuseful, though not always accurate1
does not encourage one to suppose that it has all the merits
lacking in the work of Sander and Karsten.

50
As Sander (Quaestiones 6) and Karsten (Eloc. rhet. 10)
observe. See my pp.245-51; 270-4.
51
See pp.29Of.
52
Contr. IX.1.3,10,12; 2.11,13,18,19,22; 3.5,10; 4.5,11,14,15,
16; 5.3,6,16; 6.3,19; X.2.12; 3.16.
53
0. Gruppe, Quaestiones Annaeanae (Diss. Berlin, 1873), 28.
54
R. Hess, Quaestiones Annaeanae (Diss. Kiel, 1898), 4ff.
55
Bornecque, Decl. 200.
56
I.e. to discuss points of declamatory technique, rather than
to deliver full declamations. For this use of disputare
(distinct from that relating to philosophical discussion,
examples of which are collected by G. Francois in his article,
'Declamatio et disputatio\ AC XXXII (1963), 513ff.) see
Contr. VII.3.10, X pr. 15.
57
W. Hoffa, De Seneca Patre quaestiones selectae (Diss.
G'dttingen, 1909), 33ff.
58
See p.40 on the use of aiebat perhaps with reference to these
writings.
59
References to M. Valerius Messala Corvinus as an orator,
critic and wit occur in Contr. 11.4.8,10; III pr. 14; Suas.
1.7; 2.17,20; 3.6f.; 6.27. Iulius Montanus is mentioned as a
critic in Contr. VII.1.27; also in fr. 3 Muller, which however
may be by the younger Seneca (cf. Ep. mor. 122.11). It is un-
likely that the name of Iulius Montanus unabbreviated could
have been corrupted into Votienus Montanus in Contr. IX pr. 1:
the name Votienus is so much the lectio difficilior as com-
pared with Iulius, and it occurs twice in the one section.
If we accept the hypothesis that someone other than Votienus
Montanus was introduced at the beginning of the ninth preface,
we have to account for the fact that almost all of the extant
fragments of Votienus come in the ninth book. Perhaps they
were equally numerous in the eighth book and, if the remarks
in Contr. VII.5.11f. were not sufficient introduction, he had
been mentioned in the preface to Contr. VIII. If this were
so, the name of Votienus Montanus would have been familiar to
the hypothetical early scribe who made the mistake, when he
came to copy Contr. IX pr. 1.
The criticism
1
Montaigne, Essais ed. M. Rat (Paris, 1962), I 36f.
2
Cowley, Essays and other prose writings ed. A.B. Gough
(Oxford, 1915), 179f.
3
Jonson, Works Vol. VIII ed. C.H. Herford, P. and E. Simpson
(Oxford, 1947), 59Of.
k
Op. cit. 591.
5
Op. cit. 391. See also Jonson1s use of Contr. IV pr. 7-11 in
his prose criticism of Shakespeare, quoted and discussed by
Sussman, The elder Seneca 170f.
6
Sussman, op. cit. 95ff., now provides much fuller discussion
of Senecafs pen-portraiture than any previously available.
7
Or just conceivably a book; cf. the use of aiebat in Contr.
IX.6.18.
8
S.F. Bonner, Dionysius of Halicarnassus (Cambridge, 1939), 23.
9
Op. cit. 24.
10
See pp.245f.
11
See pp.8Of.
12
Cf. G. Watson, The literary critics (Harmondsworth, 1962),
llff.
13
See W.D. Lebek, fZur Vita des Albucius Silus bei Sueton1,
Hermes XCIV (1966), 36Off.
14
See F. Leo, Die griechisch-r'6mische Biographie nach ihrer
literarischen Form (Leipzig, 1901 repr. Hildesheim, 1965);
W. Steidle, Sueton und die antike Biographie (Munich, 1951);
A.D. Momigliano, The development of Greek biography
(Cambridge, Mass., 1971).
15
Antigonos von Karystos (Berlin, 1881), 82.
16
History of autobiography in antiquity (London, 1950), I 299 =
Geschichte der Autobiographie3 (Frankfurt, 1947), I 312.
17
Now edited by F. Wehrli, Schule des Aristoteles, Supptbd I
(Basel, 1974).
18
Antigonos von Karystos 82.
19
For the neuter form see ThLL s.v. epitaphius.
20
On the nature of this work see Momigliano, Development of
Greek biography 96f.
21
See p.93; n.68, p.346.
Notes to pages 65-70 34O
22
HRR II 40.
23
See n.68, p.346, on fr. 24*, but it is not certain exactly
which of Nepos* works this fragment comes from.
2h
See W.D. Lebek, Hermes XCIV (1966), 36Off. Only sections 3
and 5 of the Suetonian Life are certainly derived from Contr.
VII pr.
25
But n.b. this work is not as comprehensive as it at first
sight appears; for example it does not include by any means
all the fragments of Nepos concerned with literary matters.
26
Notably J.F. D'Alton, Roman literary theory and criticism
(London, 1931) , 546: 'His prefaces alone would win for him a
high rank among Roman critics. It is probable that he would
not have achieved so much, if Cicero had not smoothed the
path for him, but his judgements are generally characterised
by a freshness and spontaneity that we often miss in Cicero.1
27
CJ XXXIV (1939), 347ff.
28
'The elder Seneca as a critic of rhetoric1 (Diss. N. Carolina,
1969), 70ff.; The elder Seneca (Lei,den, 1978), 115ff.
29
See pp.151-239.
30
See pp.246-96.
31
Paris, 1940.
32
AC XII (1943), 5ff.
33
Notoriously, Bardon makes no use of Mlillerfs edition; many
other serious failings in his lexicographical method and con-
clusions were soon pointed out by reviewers of Vocabulaire9
e.g. J. Marouzeau, REL XVIII (1940), 2O3ff.; A. Cordier, RPh
XVII (1943), 22Off.; W. Stegemann, PhW LXIV (1944), 172ff.;
C.J. Fordyce, CR LX (1946), 129. See also Sussman, The elder
Seneca 103f.
34
His chapter headings: II Faiblesses: obscurites; III
Faiblesses: Appauvrissement de certains emplois; IV
Faiblesses: pauvrete du vocabulaire, will give some indi-
cation of the tone of his work.
35
The Greek and Roman critics (London, 1965), 179.
36
Bardon, Vocabulaire 85.
37
Watson, The literary critics 14.
38
Op. cit. 28.
39
Bardon, Vocabulaire 85.
1+0
See E. Fantham, Comparative studies in Republican Latin
imagery, Phoenix Suppl. Vol. X (Toronto, 1972), 137ff., on
the De ora tore.
1+1
Editio Commeliniana (Heidelberg, 1603/4) praef. He was
dissenting fa viro doc to, qui memoria nostra Senecam hunc
Rhetorem leviter doctum, ac quasi proletarian appellat'.
Schottfs judgement is cited with whole-hearted approval by
Edward (Suasoriae x ) .
1+2
Bardon lists a number of more or less inelegant expressions
used by Seneca the Elder in his article, AC XII (1943), 5ff.
1+3
For a comparison between the two Senecas1 accounts of
Fabianus1 style see A.D, Leeman, Orationis ratio (Amsterdam,
1963), I 264ff.
Oratory and rhetorical theory up to his own time

1
See pp.95ff.
2
On the text see n.50, p.345.
3
See pp.lOlf.
** Discussed further on pp.lO4ff.

5
Cf. Edward and Winterbottom ad loc.
6
Fragments collected by D. Matthes (Teubner, Leipzig, 1962).
7
G.M.A. Grube, fTheodorus of Gadara1, AJPh LXXX (1959), 337ff.
gives a well-balanced account of Senecafs references to their
teaching, in relation to the more extensive data in Quintilian
and the 'Anonymus Seguerianus* Rh. Gr. 1.2 352ff. Spengel-
Hammer.
8
Cf. Anon. Seguer. U3ff. (Rh. Gr. 1.2 372).
' The theme does not state explicitly whether the ex-prostitute
who wishes to become a priestess has actually, as she claims,
kept her virginity.
10
Grube, AJPh LXXX (1959), 335ff. argues against the view that
Theodorus favoured impassioned rhetoric free from rules, but
does not cite Suas. 3.7 in this connection, as one might well
do.
11
*I6uA)TLau6s first appears in Philodemus, Poet. 11.71, but is
credited to earlier Stoic theory by Diog. Laert. (VII.59);
U£T&(ppaaus is not attested in Greek sources before Plut. Dem.
8.2. See n.14, p.342, for Senecan references.
12
Used in several senses by Plautus and in the sense Twriter of
comedy1 by Cicero (Or. 55.184).
13
Note how in Suas. 6.21 a transliteration is provided by a
second hand in B, where AB 1 have Greek characters. Sometimes
(e.g. in Contr. X.4.21f. M) even Greek declaimers1 names have
been transmitted in Greek letters.
14
Except that the Greek words never seem to be metaphors for
describing the criticfs subjective impressions of the nuances
of style. Most of them are technical terms referring to
literary occupations and genres, themes for rhetorical or
philosophical treatment, the major sections of a literary
work, figures of speech or thought and elements of diction:
anthypophora (Contr. 1.7.17); cacozelia (Contr. IX.1.15;
2.28; Suas. 7.11); cacozelos (Suas. 2.16); comicus (Contr.
VII.3.8); echo/r)x&* (Contr. VII. 7.19); enthymema (Contr. I
pr. 23); epilogus (Contr. Ill pr. 10; IV pr. 8; VII.4.5*,6,8;
5.7; IX.6.12*,13); epiphonema (Contr.'1 pr. 23; X.4.25);
eituxdcpuov (Suas. 6.21, see n.19, p.339); ethicos* (Contr. II.
2.12); ethicos (Contr. II.3.23; 4.8; Suas. 1.13); grammaticus*
(Suas. 2.13); hendecasyllabi* (Contr. VII.4.7); hermeneumata
(Contr. IX.3.14); (hexis - Bursian's conjecture in Contr. VII
pr. 2 ) ; historia (Suas. 5.8; 6.15; 6.25); historicus (Suas.
2.22; 6.14,16); idiotismos (Contr. II.3.21; VII pr. 5 ) ;
ironia (Contr. 1.7.13); metaphrasis (Suas. 1.12); mimicus*
(Contr. VII.5.15); mimus (Contr. VII.3.9); pantomimus (Contr.
Ill pr. 16*; Suas. 2.19); phantasia (Suas. 2.14); philo-
sophumenos (Contr. 1.3.8; 7.17); phrasis (Contr. Ill pr. 7;
VII pr. 2*); problema (Contr. 1.3.8); prooemium (Contr. I.I.
25; VII.1.26; X.I.13); rhetor (Contr. VII.4.8,9; IX.2.26; X
pr. 10,11; Suas. 2.12); schema (Contr. I pr. 23f.; II.1.24;
3.22; VII pr. 7*,8; IX.2.22; X pr. 10); schola (Contr. I pr.
24*; 7.18; III pr. 13*, 15f.*; IX pr. 5*; X pr. 11; Suas. 2.
15,21); scholastica (Contr. I pr. 12; II.3.13*; III pr. 12*;
VII pr. 8; IX.5.15; X.5.12, see S.F. Bonner, fRhetoricaf, CR
LXI (1947), 86); scholasticus - adj. (Contr. IX pr. 4*, 5 ) ;
scholasticus - noun (Contr. 1.6.10; 7.15; II.2.8; 3.13*,19;
III pr. 16*; VII pr. 9; 5.12; Suas. 3.6); soloecismos (Contr.
IX pr. 3; Suas. 2.13); syllaba (Contr. 1.7.18*; X pr. 11;
Suas. 7.11); tetracolon (Contr. IX.2.27); tetradeum* (Contr.
X pr. 13); thema (Contr. 1.2.14; IX.5.11); thesis (Contr. I
pr. 12; VII.4.3); tragicus (Contr. VII.3.8); tragoedia (Suas.
3.7); tricolum (Contr. II.4.12).
Systematic lexicography will probably reveal that this list
is not exhaustive. It contains merely the examples cited by
Bardon in Vocabulaire, plus a few (marked with an asterisk)
which I happen to have noticed to be missing from his Lexique.
15
Contr. 1.1.15; 2.16; II.1.23; 3.18; 4.9; 6.6; VII pr. 6; IX.
2.24; X.4.18.
16
See especially Contr. IX.2.27f. for the interchangeability of
these words; Seneca1s use of them is discussed in detail on
pp.214ff.
17
See n.14, p.342.
18
See p.5.
19
E.g. Contr. VII.6.17.
2{
^ By contrast, the Seianiani were described as parricides for
their offences against the Emperor (Contr. IX.4.21), even
though Tiberius was never officially dubbed Pater Patr'iae.
21
See Winterbottom1s notes on these declamations.
22
Contr. VII.2.1 (Sepullius Bassus) cf. Verr. II.5.118; Suas.
6.3 (Latro) cf. Verr. II.5.161; also Suas. 6.11 (Varius
Geminus on Cicero and Sicily in general).
23
Contr. VII.2.10 (Marcellus Aeserninus) cf. Cat. IV.3 etc.;
Suas. 6.3 (Latro) cf. Cat. 1.2 etc.; Suas. 6.12 (Varius
Geminus) cf. Cat. IV.3 etc.
2h
Suas. 6.2 (Haterius) cf. Mil. 92,105; Suas. 7.3 (Cestius) cf.
Mil. 101; Contr. Ill pr. 15 (Cestius1 declamation in Milonem).
25
Contr. VII.2.5 (Haterius) cf. Phil. 11.64; Contr. VII.2.10
(Marcellus Aeserninus) cf. Phil. 11.119 (where Cicero recalls
Cat. IV.3); Suas. 6.3 (Latro) cf. Phil. II.63-4; Suas. 6.4
(Pompeius Silo) cf. Phil. 11.64; Suas. 6.5 (Triarius) cf.
Phil. 11.67; Suas. 6.7 (Argentarius) cf. Phil. 11.77 etc.;
Suas. 6.12 (Varius Geminus) cf. Phil. 11.119 (with Cat. IV.3);
Suas. 7.2 (Cestius) cf. Phil. 11.24; Suas. 7.5 (Pompeius Silo)
cf. Phil. 11.20; 5 and 60.
26 Latro in Suas. 6.3 quoting Cat. 1.2.
27
Words extracted from a sentence in Cat. IV.3 (cf. Phil. II.
119) quoted in Contr. VII.2.10 and Suas. 6.12.
28
Suas. 7.3 = Mil. 101 paraphrased.
29
Phil. 1.38; Marc. 25 or possibly Ad fam. X.I.I.
30
Cf. Winterbottom, BICS XXI (1974), 26.
3
* Suas. 1.5: nos quidem ilium deridemus, sed timeo ne ille nos
gladio dvTLyuxTnpLap; Ad fam. XV.19.4: scis, quam se semper a
nobis derisum putet; vereor, ne nos rustice gladio velit
32
Parallel noted by Winterbottom ad loc; cf. his notes on
Suas. 6.4 (Cestius) and 6.8 (Latro) where sententiae are
compared with Ad fam. X.I.I; but the similarity is only a
general one.
33
Contr. IX.6.16.
34
Winterbottom ad loc. also compares Veil. Pat. II.66.5; ps.-
Quint. Decl. CCLIII (37.11 Ritter).
35
Though perhaps also by Nepos, see p.65.
36
See p.23.
37
Though all or part of the digression on Antonyfs visit to
Athens is perhaps Senecan.
38
Cf. Suas. 6.4, where Winterbottom compares a sententia by
Cestius with, among other passages, Ad fam. X.I.I.
39
Pp.lO4ff.
1+0
On certain vaguenesses in the exposition of the data in
Contr. I pr. 12, which make it seem unlikely that Seneca was
himself the scholar responsible for the underlying research,
see pp.l29f.
14
* On the question of how much, if any, of Cicero f s correspon-
dence was available for public perusal in Seneca1s day see
p.130.
hl
Cf. M.T. Griffin, JRS LXII (1972), 10.
1+3
On the unreliability of arguments ex silentio see the dis-
cussion of Seneca1s references to poets, pp.311ff.
1+4
Quoted p. 104.
1+5
He taught Fabianus (Contr. II pr. 5 ) ; more than that cannot
be said with certainty - the dating of Latro's departure from
Rome to 15 B.C., and the assumption that he did not return,
used by Bornecque (Decl. 194) in dating Blandus1 birth, are
baseless.
46
Not that his vocabulary was entirely distinct from Cicerofs;
Bardon's lexique in Vocabulaire llff. lists many examples of
usages shared by the two critics. But Sussman never
adequately justifies his assertion (The elder Seneca 95)
that f Seneca 1 s critical outlook substantially reflects the
influence of Cicero...1
hl
PACA XII (1973), 11.
1+8
But for another case where Cicero's name has surely been
interpolated without any such provocation see Sen. De clem.
I.10.1 where modern editors have let the following statement
about Augustus slip past unemended: iam Domitios, Messalas,
Asinios, Cicerones, quidquid floris erat in civitate,
clementiae suae debebat.
1+9
Cf. A.D. Leeman, Orationis ratio I 138ff.; W.D. Lebek, Verba
prisca (GBttingen, 1970), 83ff.
50
reget A riget BVD: viget Jahn. Rigere is not found in
CiceroTs rhetorical works; nor does rigidus seem to be used
of literary qualities by the rhetorical theorists; vigere, on
the other hand, is used with reference to prose rhythm in
Cic. Or. 64.215. For another comparison of Calvus with
Demosthenes see Plin. Ep. 1.2.2, quoted n.62, p.346.
51
Here Muller prints his supplement summisse, which is also
adopted by Winterbottom, but molliter9 suggested to me by Mr
Duncan Kennedy, seems more convincing.
52
Cf. ORF3 492ff. for the testimonia relating to Calvus' ora-
tory; Lebek, Verba prisca 88f. for a summary of modern
attempts to account for the disagreement over its merits.
53
On the dating of this letter see D.R. Shackleton Bailey ed.
Ad familiares (Cambridge, 1977), ad loc.
54
See p.99.
55
Orationis ratio I 139. Cf. Lebek, Verba prisca 95f. on the
fact, recognized in later antiquity, that Cicero's manner was
quite far removed from that of Demosthenes.
56 0RF3
57
A.W. de Groot, Der antike Prosarhythmus (Groningen, 1921),
107.
58
Handbook of antique prose rhythm (Groningen, 1919) , 196.
Unfortunately de Groot gives no figures for Lysias. He
regards Thucydides, with some reservations, as virtually
reproducing the 'normal1 frequency of rhythms in Greek.
59
primum igitur eum tamquam e vinculis numerorum eximamus.
sunt enim quidam, ut scis, oratorii numeri, de quibus mox
agemus, observandi ratione quadam, sed alio in genere
orationis, in hoc omnino relinquendi.
60 turn removebitur omnis insignis ornatus quasi margaritarurn etc.
61
On the identity of the Roman Attici see A.E. Douglas, 'M.
Calidius and the Atticists', CQ N.S. V (1955), 241ff. es-
pecially 245f. where he excludes from the ranks of the
Atticists everyone but Calvus and, probably, Brutus; also
his Cicero - Greece and Rome: New Surveys in the Classics II
(Oxford, 1968), 38f. where he excludes Brutus as well,
following K. Barwick, intr. ed. Brutus (Heidelberg, 1949), 15,
who argues that Cicero would not have dedicated the Atticists
to Brutus if he had belonged to their movement (an invalid
argument, seeing that Caesar dedicated his De analogia to
Cicero, with whose views he was not in agreement), and F.
Portalupi, Bruto e i neo-atticisti (Turin, 1955), who sees
Brutus' style as the product of his philosophical training
and the natural austerity of his character. But the time has
come for a reaction against the type of reasoning which seeks

to eliminate from consideration as an Atticist any orator who
does not conform strictly to Cicerofs prescription for the
plain style (or who, though conforming to it, may have had
extra-literary reasons for doing so). Future discussions of
Roman Atticism ought to take the fragments of Calvus, the one
Latin orator known to have aspired to the epithet Atticus
(Cic. Brut. 82.284), as their starting point. Lebek, Verba
prisca 97, is right to stress that there were diverse tend-
encies within the Atticist movement.
62
Cf. PI in. Ep. 1.2.2: temptavi enim imitari Demosthenen semper
tuum, Calvum nuper meum, dumtaxat figuris orationis; nam vim
tantorum virorum, 'pauci quos aequus...' adsequi possunt.
63
And consequently not to conclude automatically that any
orator whose style is known to have been rhythmical must be
excluded from their number. The case of M. Calidius (cf.
A.E. Douglas, CQ N.S. V, 1955) needs to be reconsidered.
64
Discussed by Lebek, Verba prisca 86 n.16. I accept the modern
convention of regarding -^-w and -\J— as the same clausula
type. Note also Calvus fr. 27: vehement issime probare.
65
W.H. Shewring, fProse rhythm and the comparative method II f ,
CO XXV (1931), 12ff., esp. 13f.
66
Note the successions of short syllables: credite mihi, non
— — ^j \j \^>
est turpe miseren.

67
Cf. p.54.
68
N.b. A fragment of Nepos (Exempla, fr. 24*, HRR II 33) con-
tains a Catullan allusion amongst miscellaneous information
about Mamurra.
69
So Winterbottom; MUller prints his conjecture, sceleratissimi
ca1umnia toris.
The history of declamation

1
On the textual problem here see p.95.
2
For detailed discussions see: R.P. Robinson ed. Suetonius, De
gramm. et rhet. (Paris, 1925), 35ff.; H. Throm, Die Thesis
(Paderborn, 1932); W. Hofrichter, Studien zur Entwieke lungs-
geschichte der Deklamation (Diss. Breslau Univ. publ. Ohlau,
1935); W. Kroll s.v. 'Rhetorik% , RE Supplbd VII (1940),
1039ff.; S.F. Bonner, 'Rhetorica1, CR LXI (1947), 84ff.; Rom.
decl. Iff.; M.L. Clarke, fThe thesis in the Roman rhetorical
schools of the Republic*, CQ XLV (1951), 159ff.; E.M.
Jenkinson, TFurther studies in the curriculum of the Roman
schools of rhetoric in the Republican period1, Symbolae
Osloenses XXXI (1955), 122ff.; S.F. Bonner, Education in

ancient Rome from the elder Cato to the younger Pliny (London,
1977), 65ff.; Sussman, The elder Seneca 2ff.
3
This is a selection of the examples in Cic. Part. orat.
18.61ff., a passage which illustrates the appallingly complex
way in which the different varieties offteaeuswere classi-
fied by ancient theorists. For a useful list of $eaug topics
see Bonner, Rom. decl. 3-5.
^ On the part played by Aristotle and Theophrastus in promoting

the %£OL£ see also Quint. XII.2.25; Theon, Prog. 69.1
(Rh. Gr. II Spengel).
5
See Cic. De or. I.11.45ff.
6
Cf. De or. III.18.67; Brut. 31.119f.; Acad. 1.19; De fin.
IV.5; Quint. XII.2.25.
7
See n.6, p.341.
8
Cic. De inv. 1.6.8; Quint. III.5.5; 7ff.; 12ff. etc. =
Hermagoras fr. 6a-e Matthes. HermagorasT claim that abstract
questions were included in the province of rhetoric was
challenged by Poseidonius (Plut. Pomp. 42.5).
9
Though perhaps we should read reJbus; see n.45, p.351.
10
Few papyri giving fragments of school exercises of any sort
survive from before the Roman Imperial period. See R.A. Pack,
The Greek and Latin literary texts from Greco-Roman Egypt2-
(Ann Arbor, 1967), nos. 2495-2559: oratory (including rhetori-
cal exercises); nos. 2642-2751: elementary school exercises.
** Lucianfs Phalaris II is a suasoria (the Delphians deliberate

whether to accept the bronze bull of Phalaris as a gift), his
Tyrannoktonos and Apokeryttomenos are controversiae, the
latter on the same theme as Contr. IV.5; the speeches of
Aelius Aristides include show-pieces of the suasoria type
e.g. LII. On sophistic declamations later than these see W.
Morel, W. Kroll, s.v. 'Melete', RE XV.1 (1931), 496ff.; G.
Kennedy, The art of rhetoric in the Roman world (Princeton,
1972), 553ff.
12
I accept that, as Bonner (Rom. decl. 15) points out: fthe
illustrations of the various types of axaaus given in the
rhetorical handbooks are remarkably like simple and straight-
forward declamatory exercises1.
13
One cannot regard the interest in deliberative oratory shown
in Aristotle's Rhetoric or the Rhetorica ad Alexandrum as
certain proof that school exercises of the suasoria type were
in use when these works were written. Spengelfs ueAexftv
TtoAAdxus in Rhet. Alex. 1422a fin., cited by Bonner, Rom.
decl. 11, is conjectural.
ll+
TtapayYeXyaTuxfts i n Sext. Emp. P.H. 1.204 has the meaning
f
hortativelyf; no other example is cited in LSJ.
15
Cf. Rhet. Her. III.2.2 for other examples of multiple themes
for deliberation.
16
Though Philodemus (Rhet. 1.134 Sudhaus) alludes to yuynyaxa
.. . TWV 6 u x a v [ u M Si v] xa [u] auylBouXeuTtx [wv x]a[l
71
[p?a] $£UTIKGJV Xoy[a)]v.
17
As we shall see, there is no need to share Bonner's doubts
(Rom. decl. 12) as to whether judicial exercises more complex
than mere type declamations such as 'against a rich man 1 or
1
against a tyrant1 could have existed as early as Aeschines1
time.
18
Rhetorische Papyri ed. K. Kunst (Berlin, 1923), 4ff.
19
I accept that this, the MSS reading, is probably right (cf.
Contr. X pr. 8) rather than MUllerfs pro Pythodoro <reo>
Messalae (cf. Suas. 6.24), but I disagree with Winterbottom's
inference that it was a very eloquent speech by Messala pro
Pythodoro that Latro recited. Latrofs recitation was pro-
voked by an unkind remark which Messala had made about his
style. The verb recitare is used of all three of these mock
law-court speeches; this seems to suggest that they were more
carefully prepared than the average declamatio, hence perhaps
Seneca's curious use of the term oratio with refetence to
Latro f s display speech.
20
See K.J. Dover, 'The chronology of Antiphon's speeches', CQ
XLIV (1950), 44ff.
21
Bonner, Rom. decl. 12; cf. Clarke, CQ N.S. I (1951), 161.
22
Murder: Tetr. 1.1.2 xov cp6vov; time and place: 1.4 awpl TWV
VUMTWV...€:V epnulqt; slave still alive when found: 1.9 l
yap £ T L apdels; no robbery: 1.4 ixovres. . .to. uyaxua
the accused a rich man: 2.12, 3.8; enmity between him and the
murdered man: 1*5 ex TtaXoiLoO. . .ex^POS wv; prosecution for
embezzlement not yet heard: 1.6-8; date of crime: 4.8;
identification by slave: 1.9 dvaxpuvoyevos ucp' ny&v, TOOTOV
yovov ecpn YVWVCIL T & v TICXL6VTU)V auxous; his subsequent death is
assumed in, e.g., 1.4 ou y&P c*v a ^ v T V <ixoXo6§(p 6ue(p§eupev
auTOV and his unavailability for interrogation under torture
is a crucial issue discussed in 2.7, 3.4, 4.7.
23
Contrast the comparative wealth of circumstantial detail even
in such a simple speech as Antiphon's Prosecution of the
stepmother. None of the allegations made about the rich
enemy in addition to the charges outlined above, or state-
ments made by him in self-defence, goes beyond the approved
varieties of mud-slinging and white-washing described by
ps.-Quintilian in Decl. min. CCCLI and CCCLII.
24
gui .. .reiun detulerat cf. ps.-Quint. CCCXIX; divitem inimicum
cf. Contr. X.I; peculatus cf. Contr. IX.1 (for peculatus as
equivalent to LepCJv xAoTtfi see Bonner, i?om. decl. 106); inter
moras iudicii cf. ps.-Quint. CCCXIX; cum servo: for slaves in
controversiae see, e.g., ps.-Quint. CCCLXIV, Contr. III.9;
VII.6; nocte cf. ps.-Quint. CCLXXII; processit cf. ps.-Quint.
CCCLXIV; occisus inspoliatus...inventus est cf. Contr. X.I;
in solitudine cf. ps.-Quint. CCLXXXI; servus...nominavit cf.
Contr. VII.5: ...pater familiae in cubiculo occisus inventus
est, uxor volnerata, communis paries perfossus (hence my
nominative); placuit propinquis quaeri a filio quinquenni,
qui una dormierat, quern percussorem cognosceret; ille
procuratorem digito denotavit; iuxta cadaver cf. ps.-Quint.
Decl. mai. II; et decessit cf. Contr. II.4; dives caedis
reus est cf. ps.-Quint. CCCLXIV.
25
Actually this argument is not pressed as unsubtly as one
might have expected in the first Tetralogy; the emphasis is
on the improbability that ordinary robbers committed the
crime, rather than on the likelihood that the rich man was
responsible.
26
De gramm. et rhet. 25.9 Brugnoli, see p.116.
27
See pp.l64f.
28
Dem. Or. XXIII 54, cf. 51 for ascription to Draco.
29
CQ XLIV (1950), 58.
30
On similarities to Attic law see especially U.E. Paoli,
'Droit attique et droit romain dans les rheteurs latins1,
Revue historique de droit frangais et etranger XXXI (1953),
175ff.; for an extremely archaic law used in connection with
controversiae see Contr. IX.4: gui patrem pulsaverit manus ei
praecidantur which, as Bonner notes (Rom. decl. 96), has an
exact parallel in the Code of Hammurabi of 2100 B.C. (§195) -
text and translation in G.R. Driver, J.C. Miles, The
Babylonian laws (Oxford, 1955), II. (This appears to be the
only parallel between declamatory law and this code.)
Whereabouts in the Greek world mock law-suits might have
originated remains to be traced; we may rule out any of the
Sicilian colonies where the Code of Charondas was used, for
this well-documented code (Arist. Pol. passim; Diod. Sic.
Xll.llff.) shows no particular affinities with known declama-
tory law. Sicily remains a possibility (Cic. Brut. 12.46),
but remember the sophists' diverse origins and wide travels.
31 E.g. Contr. 1.7; II.5; IV.7; V.8; IX.4.
32
Note that, if the use of 6ua<pde(,peuv in Tetr. II, e.g. in
4.9, eav 6uacpdapfi, is taken to refer to the death penalty,
the legal situation in that Tetralogy is not identical with
Roman declamatory law, in which five years1 exile is the
standard penalty for involuntary homicide (see Bonner, Rom.
decl. 98ff.). But need it be interpreted in this way?

33
See Philod. Rhet. 11.97 Sudhaus, fr. VIII, for evidence of a
dispute about the extent of Aeschines1 education.
3i+
He was well aware of what constituted a thesis, see e.g.
Contr. VII.4.3.
35
These exercises, with the exception of translation, all
belong to the class which Greek rhetoricians called pro-
gymnasmata. In the Roman education of Suetonius' day they
were dealt with by the grammatici, the Roman rhetores con-
sidering such simple exercises beneath them; the Greeks
however seem always to have taken them more seriously (Quint.
1.9.6). For an ancient account of the history of progym-
nasmata see Theon, Prog, 65ff. (Rh. Gr. II Spengel); for the
individual exercises see this work passim; also for dicta...
exponere cf. Quint. 1.9.2; narrationes...explicare cf. Quint.
1.9.3; II.1.8; 4.15ff.; viros...vituperare Quint. II.1.8;
4.2Of.; quaedam...ostendere (theses) Quint. II.1.9; 4.24f.
(cf. also Quint. II.1.11; 4.22f. on communes loci); fabulis
...demere cf. Quint. II.4.18f.
36
Brugnoli's text, except for the last sentence, where the
obelizing is mine.
appellationes Graeci OaKHUCS (synthesis graeci eas
appellationes E); appellationes Graece W$MQ Roth, Hofrichter,
Bione, qui crucem anteposuit; appellatione Graeca Schott,
Reiff.; tappellationest < > Graeci Robinson; appellationes
del. Brugnoli.
syntasis OB; syntaxis WNGDV Reiff, Hofrichter, Bione; syn-
thesis L (corr. syntaxis) HE Roth; sintaxis IM; sintesis K;
sinthasis UC; sinthesis A; synthasis Q; auvdeaeus S; auvxaêus
Stegemann; auaxaaets Bonner; Reacts Gronovius, Bentley; uito-
%kocL£ Wolf, Volkmann; auvTaaeus Brugnoli.
Despite Bonnerfs arguments in 'Rhetorical , CR LXI (1947),
84ff., I find it hard to believe that the exercises were ever
called appellationes, and am inclined to think the conjecture
appellatione Graeca is right (cf. appellatio...Graeca in De
gramm. et rhet. 4.1). As for syntasis etc., we should bear
in mind that the earliest extant MSS of the De gramm. et
rhet. belong to the fifteenth century and derive from a
single archetype brought to Italy from Hersfeld by Enoch of
Ascoli around the middle of the century. All but perhaps one
of the manuscript readings recorded are therefore Renaissance
conjectures or miscopyings. Clearly the word began with syn-
or sin- in the Hersfeld codex, but we need not be so sure
that the word Suetonius wrote began with this prefix as we
would have to be if a wide range of medieval manuscripts were
extant, all attesting such a reading. In addition, the Greek
word seems to have been transmitted in Latin characters; un-
comprehending western scribes could have distorted it con-
siderably in the transmission of the text before the
Renaissance. These facts need pointing out because the only
conjecture which corresponds well with the known usage of
Cicero's time is UTtoSeaeug, rather remote from the MSS

readings. For uitoSeaus = causa, controversia, see pp,125f.
and especially Quint. III.5.7: hae UTtodeaets a Graecis
dicuntur, causae a nostris, and observe that Seneca in the
section of Contr. I pr. 12 parallel with the Suetonian
passage refers to the use of the term causae.
37
See pp.lO5f.
38
Cf. E.G. Sihler, 'GeTUxdrrepov' , AJPh XXIII (1902), 283ff.
39
This must be the conclusion even if, on the grounds that
theoretical works do not necessarily reflect accurately the
curriculum of contemporary schools, we prefer not to assume
that all the numerous imaginary legal situations cited as
illustrations of technical points in handbooks like the
De inv. and Rhet. Her. need have been actually used as
themes for declamation in the schools of the Roman Republic.
1+0
Including perhaps more of the type represented by the decla-
mation about Leptines in Berl. Pap. P. 9781, on which see
p.110.
41
Cf. pp.l44ff.
1+2
The expression is Cicero's: Tusc. disp. 1.35.86.
1+3
Cf. the invectives against Cicero and against Sallust in the
Sallustian appendix, but R. Syme, Sallust (Cambridge, 1964),
314ff. for arguments against a late Republican dating for them.
kk
Cf. Bonner, Rom. decl. 22, 29.
1+5
Rebus, suggested here in a marginal note in Cod. O (see
apparatus in K. Kumaniecki's Teubner, 1969, ad loc.) may be
preferable both in this phrase and the parallel one in §110
(quoted p.106 above) to the MSS reading reis printed by the
editors. Rei only figure in judicial rhetoric, whereas the
example of a finita controversia given here belongs to the
genus deliberativum, and controversiae, according to the
theorists being reported, could also be epideictic. Cf.,
however, Cic. Or. 14.46: a propriis personis et temporibus...
1+6
mox controversias quidem [sc. vocabant] , sed aut fictas aut
iudiciales. I am unconvinced by the interpretation of these
words assumed by Bonner, Rom. decl. 18f.
h7
Contr. VII pr. 8; IX.5.15; Quint. IV.2.30; VII.1.14.
1+8
'Rhetorica', CR LXI (1947), 86.
1+9
Cf. p.95, on the text.
^ His source may have made the matter clearer.

^ Contrast Bonnerfs arguments, Rom. decl. 28. But the fact

that the Auctor ad Her. calls declamatio an exercise is
irrelevant, for he is referring to voice-training, and in any
case the traditional dating of the Rhet. Her. to the Sullan
period has been called into question since Bonner wrote by
A.E. Douglas, fClausulae in the Rhetorica ad Herennium as
evidence of its date 1 , CQ N.S. X (1960), 65ff.
52
Though Bake f s conjecture denuntiatio is printed in the OCT.
53
Cf. the use of declamatio in Plane. 19.47.
5lf
As Clarke, CQ N.S. I (1951), 166, has pointed out.
55
That the thesis alone is assigned to the pre-Ciceronian
period in Contr. I pr. 12 is more easily explained if we
imagine that Seneca's source gave some generalizations,
similar to those in Quint. II.1.9 about the early use of
abstract exercises, rather than a detailed list of pro-
gymnasmata such as we find in the Suetonian account.
56
Quoted p.119. Causa is not used with reference to rhetorical
exercises in the more famous autobiographical passage, Brut.
89.3O4ff.
57
Cf. D.R. Shackleton Bailey ed. Cicero's Letters to Atticus
Vol. I (Cambridge, 1965), 59ff.
58
Clarke, CQ N.S. I (1951), 165f., notes how in Ciceronian
theory general questions are regarded as prior logically to
particular questions, and suggests that fthis supposed logical
priority may have given rise to the belief in their historical
priority1.
The decline of rhetoric in the early Empire

1
See p.84.
2
Cf. H. Caplan, fThe decay of eloquence at Rome in the 1st
century1, Studies in speech and drama in honor of Alexander
M. Drummond (Ithaca, N.Y., 1944), 295ff.; L.A. Sussman, 'The
elder Seneca's discussion of the decline of Roman eloquence1,
CSCA V (1972), 195ff.
3
See Tac. Dial. 36.1, where we read after a lacuna: eadem
ratio in nostra quoque civitate antiquorum eloquentiam pro-
vexit, and ps.-Longin. Subl. 44, which contains statements
about the conditions provided for great literature by
6nuoHpaTux in general, and ev TCXUS TioXtTetaus in the plural
(44.2-3). N.b. with J.W.H. Atkins (Literary criticism in
antiquity (Cambridge, 1934), II 210ff.) and D.A. Russell (ed.
Longin. SuJbl. (Oxford, 1964), xxii ff.), I think the treatise
On the sublime belongs most likely to the first century A.D.,
pace G.M.A. Grube (The Greek and Roman critics, 34Off.) who
argues in favour of a later date.
4
Cf. Bonner, Rom. decl. 42ff.; Sussman, Diss. 191ff.
5
MUller and Winterbottom print Thomas1 conjecture, pretium
but see n.8 below.
6
Cf. Contr. II pr. 2; Suas. 6.9; Contr. II.1 passim.
7
Cf. Quint. XII.1.1.
8
Here, pace Thomas and the editors who adopt his conjecture
pretium, I am content with the MSS reading praemium. If
gloria (Mil. 13.34) and laus (Off. 11.45) can 'fall1 (cadere)
in Cicero, as can almost any other abstraction in classical
prose, there seems no reason why praemium...cecidisset is
unacceptable Latin; see ThLL s.v. cado II 2.
9
Cf. Aperfs unorthodox arguments early in Tacitus1 Dialogus
(5.3ff.) to the effect that orators in his day were, unlike
poets, well rewarded for their labours. Note especially his
use of the term voluptas (§6 passim), and his rhetorical
question in 7.3 (quoted p.139) about the glory won by
orators. More in line with Senecafs view is Maternus'
suggestion in Dial. 41.5 that his colleagues would have
enjoyed laus et gloria if they had lived in an earlier age.
10
H.I. Marrou, A history of education in antiquity, trans,
(from ed.3) G. Lamb (New York, 1964), xiv = Histoire de
1'education dans l'antiquit§s (Paris, 1965), 20.
11
Cf. Sail. lug. 2: omniaque orta occidunt et aucta senescunt.
12
See pp.83f., and pp.3O6ff.
13
See pp.83f.
14
Caplan, 'Decay of eloquence1 319; Bonner, Rom. decl. 42f.;
Sussman, The elder Seneca 12ff. Contrast, however, the views
of E.P. Parks, The Roman rhetorical schools as a preparation
for the courts under the early Empire (Baltimore, 1945).
15
Of course, ideally it should also be considered in relation
to all the information given by historians and others about
the senatorial debates and major trials of the period, but to
make an adequate attempt at this task would be outside the
scope of the present work. Parks, Rom. rhet. schools,
provides much useful material.
16
On the difficulty of the words, Passienus, qui nunc primo
loco stat in this passage see D.R. Shackleton Bailey, CQ N.S.
XIX (1969), 326, but the text has to stand: cf. Contr. II.5.7':
Passienus temporis sui primus orator.
17
Cf. Parks, Rom. rhet. schools 19ff.
18
Dio LVI.27.1, Suet. Calig. 16.1, cf. M.T. Griffin, JRS LXII
(1972), 14.
19
Cf. M.T. Griffin, Seneca (Oxford, 1975), 48; but n.b. the
association with Sejanus is only Tiberius' conjecture and too
much weight cannot be placed on it.
20
Though Seneca uses the scholastic term color in his analysis,
it is clear that this was a real trial and not a controversial
as two actiones (see pp.l64f.) and the non-scholastic pro-
cedure calumniam iurare are mentioned.
21
The quotation is not necessarily indicative of Seneca's own
attitudes.
22
Cf. Parks, Rom. rhet. schools 19ff.
23
See Contr. X pr. 5 (on Labienus): libertas tanta, ut
libertatis nomen excederet, et, quia passim ordines hominesque
laniabat, Rabienus vocaretur. animus inter vitia ingens et
ad similitudinem ingeni sui violentus et qui Pompeianos
spiritus nondum in tanta pace posuisset; cf. Contr. II.4.13.
24
Cf. Bonner, Rom. decl. 7Iff.
25
Cf. the sarcastic allusions to declamation themes in
Juvenal's Satires, especially 7.15Off., though the subject
here is the unpleasantness of the rhetorician's job, rather
than the decline of eloquence.
26
See pp.299ff.
27
Bornecque, Decl. 20; Edward, Suasoriae xii; Bonner, Rom.
decl. 71f. But Sussman, Diss. 158ff., presents the evidence
more accurately; see his remarks in The elder Seneca 17.
Inventio
1
The orator was supposed to tackle the five parts of his work
in the order given here, as Quintilian explains in III.3.10.
2
See Winterbottom's edition, Vol. II, 635, Index II, s.v.
'advocates'.
3
But cf. Contr. II.6.10.
** See Bonner, Rom. decl. 53.

5
Cf. Sussman, Diss. 83.
6
On divisio see Bornecque, Decl. 102-5; Bonner, Rom.
decl. 56ff.; Bardon, Vocabulaire 68ff.; Sussman, The elder

Seneca 38ff., lllff. Seneca does not suggest that a bare
statement of disputed issues 'regularly occurred after the
narration1, as Sussman (p.Ill) asserts.
7
Supplied by J.B. Hall.
8
An issue, however, which raised a number of subsidiary lines
of enquiry, see p.154, on Cestius1 treatment of Suas. 3.
9
Contr. 1.7.12; II.2.5; VII.4.3, 4.
10
Contr. II.5.14; VII.4.4; 8.8.
11
Contr. 1.1.14; 2.14; 4.6; II.2.5; IX.1.9.
12
Contr. 1.5.8; II.2.5.
13
The Roman schools of rhetoric were less important as promoters
of aequitas than is suggested by e.g. Bornecque, D&cl. 132,
'...le sentiment de l'equite, qui s'introduit vers cette epoque
dans le droit1, Parks, Rom. rhet. schools 76, who makes the
suggestion that the declamations of Seneca's time fmay have
contained the germs of one of the most important phases in
Roman law, that is, the development of equity1. The concept of
equity was in fact as well entrenched in the minds of Greek
and Roman orators by Senecafs time as it would ever be; it
hardly needed the schools to promote its 'development1. For
equity in Cicero's writings see Bonner, Rom. decl. 46.
llf
H. Bardon, in a chapter gloomily entitled 'Faiblesses:
obscurites% {Vocabulaire 68ff.) gives a very hostile account
of Seneca's expression in his sections on divisio; I have
aimed to provide something of a corrective to his strictures.
etiamsi lex...idonea sit is a question of aequitas.

16
Vocabulaire 69ff.
17
For the terminology of axdats theory, Greek and Latin, see
Hermagoras frs. 9-13 Matthes; Bonner, Rom. decl. 12ff. Note
that Seneca never uses the adjectives which describe the
other three types of OT6LOL£: definitiva, generalis,
translativa (Cic. De inv. 1.8.10), nor indeed the terms
aT&oxs, status or constitutio. See, however, also on color
(pp.l66ff.) for another apparently Hermagorean element in
Seneca's critical vocabulary.
18
Here a question of fact is raised which was left ambiguous in
the thema.
19
Cf. Contr. II.1.20: Silo Pompeius sic divisit: coepit a vetere
et explosa quaestione, an in omnia patri parendum sit.
20
However, in Contr. II.6.5 we find Latro using this volgarem
quaestionem, quam solebat fastidire

21
I.e. Pollio considers that Latro, if he wished to imitate the
practice of the law-courts, should not have omitted to argue
that the accused, however cruel, was not mad. I cannot
accept, pace Winterbottom ad loc., that the last sentence is
Seneca's rather than Pollio's.
22
On whom see n.2, p.373.
23
E.g. De or. III.25.96; 52.199; Seneca himself uses the word
similarly in Contr. X pr. 5: color orationis antiquae.
2k
E.g. De or. III.25.100; Brut. 87.298.
25
See e.g. Hermog. Stat. 38.16ff. (Rh. Gr. VI Rabe); and for
Latin expositions of the same theory, Cic. De inv. 1.11.15;
Quint. VII.4.7ff.
2
^ Possibly the explanation is that the term was taken from
works misattributed to Hermagoras of Temnos, such as those
deemed spurious by Quintilian in III.5.14.
27
See Hermagoras ed. Matthes, 73ff. (index rerum).
28
Diels/Kranz, Vorsokr.* II 367f.
29
colores...qui non possunt coargui: surely not just 'irrefut-
able' colores (so Winterbottom), for the adoption of these
could hardly have been considered a vitium from any point of
30
The text of Contr. II.1.33 seems corrupt in several respects.
The muddle in the MSS after coargui: non ut omniasse ut non
esset (-se B) AB; non ut omniS (-iaa D T ) sseverent ut non
esse VD gives scope for such varied reconstruction that it
seems a case for the obelus - Shackleton Bailey (CQ N.S. XIX
(1969), 324) does not even accept the usual assumption that
somnia or somniasse was in the original, and suggests that
non ut omniasse is f a dittography of ut non esse(t) with non
added for good measure*. Falsum in the sentence, sed ridiculum
possit9 should surely be deleted. Read then: illi enim
colores probabant qui non possunt coargui fnon ut omniasse ut
non esset aliquo nomine offensus.f sed ridiculum est ad-
fectari quod [falsum] probari non possit. non multum interest
in causa sua falsum aliquis testem det an se: alteri enim
credi non debet, alteri non solet.
31
Cf. Bonner, Rom. decl. 60, on the 'swoon1.
32
Should the words in hoc visum be regarded as a mangled ditto-
graphy of invidiosum9 and deleted?
33
Cf. Quint. IV.2.28.
testem adducit A. Augustinus. As alternatives consider ad

testem ducit <controversiam>, or (a suggestion of J.B. Hall)
ad test<imonium r>em ducit.
Dispositio
1
Cf. Bornecque, Decl. 53f.; Bonner, Rom. decl. 54; Sussman,
The elder Seneca 115ff.; for examples of complete contro-
versiae see ps.-Quint. Decl. maiores.
2
References to some complete works of the suasoria type are
given in n.ll, p.347.
3
See p.80.
4
Cf. Contr. X pr. 15.
5
See pp.210ff.
^ Rom. decl. 58; cf. Sussman, The elder Seneca 116.

7
See the index of commonplaces, colours and rhetorical terms
in Winterbottom1s edition, Vol. II 635ff.
8
The words luxuriam in multis are attested only by the
Excerpta MSS and, seeing that Gurges has already been men-
tioned as an example of luxuria, one would happily delete
them as an inept interpolation, if it were not that Lucullus1
name is more associated with luxury than avarice. At all
events the text of the end of this sentence seems suspect.
9
See n.68, p.346.
10
On the dating of this work see J. Wight Duff, A.M. Duff, A
literary history of Rome in the Silver Age^ (London, 1964) ,
54f.
11
Cf. my analysis of Latro's treatment of Contr. II.7, pp.
251ff.
* 2 I accept that Kiessling and subsequent editors are right to

transpose the words in epilogis...Graecus from Contr. VII.4.3
(after passurum) to a position before Latro's remarks on
epilogi. It seems less certain that the declamatory quotation
here assigned to Apollonius belongs to him, rather than some
other declaimer whose name has dropped out. Perhaps in
epilogis Graecus came immediately before Latro dixit...
13
Cf. pp.96-7.
14
Cf. the translations of Bornecque: ! Le coupable doit dis-
paraitre dans la p£roraison', and Winterbottom: fIn the per-
oration the accused should be over and done with.'
15
I recognize that, if my deletion of Latro were rejected, and
it could be argued that Fuscus was making a new point, the
question would have to be asked whether debet...reus in
epilogo desinere could not mean just the reverse, i.e.: 'It
ought to be the defendant who concludes the speech in the
epilogus.1 But I suspect the declamatory Latin for that
would be: debet...reus cum epilogo desinere cf. Suas. 1.2:
tempus est Alexandrum cum orbe et cum sole desinere. For
desinere in cf. Sen. De prov. 4.14: omnes considera gentes in
quibus Romana pax desinit, Germanos dico et quiquid circa
Histrum vagarum gentium occursat.
16
It might conceivably be regarded as part of FuscusT critique
and be taken to mean (reading Bursian's contexet): 'but he
will weave the epilogus (i.e. his concluding accusations)
on to the defensive part of his speech as well as possible1 -
but would there be any particular point in a smooth tran-
sition when the defendant is changing his tactics radically
and is eschewing the psychological advantages to be gained by
putting defence last?
Elocutio
1
The fullest existing discussions of Seneca on diction are
those of A.F. Sochatoff, CJ XXXIV (1939), 347ff., and Sussman,
Diss. 113ff., 121ff.; The elder Seneca 118ff.; but none of
these is detailed enough to illustrate all the respects in
which Seneca differs from the stricter critics.
2
See n.14, p.342.
3
Decl. 121.
Priap. 3.7f.: quod virgo prima cupido dat nocte marito \ dum
timet alterius vulnus inepta loci.
Cf. Contr. VII pr. 3 for inclusion of lamps among res...

sordidissimas.
6
Vocabulaire llff. That is not to suggest, of course, that
there are no elements common to the vocabulary of the two
critics.
7
P.248.
° Mlillerfs conjecture, plantaria9 has much to recommend it.

The context demands reference to something which could be
fancifully described as 'rising groves1: plantaria, 'saplings1
or 'tree-nurseries1, fulfil the requirement admirably.
9
Though, to judge from Sen. Ep. mor. 114.4ff., Maecenas1 own
prose was notoriously full of just this type of licentia.
10 No lacuna is marked in Winterbottom1s text, but surely a

Greek quotation is lacking at this point.

11
On the text of this passage see pp.265ff.
12
On the meaning of this word see pp.211f.
13
See Dion. Hal. De comp. for a full treatment of this subject
in relation to Greek style by a near-contemporary of Seneca
the Elder. See also pp.43ff. on the clausulae preferred by
Seneca, Latro and Fuscus, and my review of Winterbottom1s
Loeb, JRS LXVI (1976), 27Off. for a critique of Hagendahlfs
statements about clausulae in the declamatory extracts.
Sussman's account of compositio (The elder Seneca 126f.) is
in several respects misleading: (in-)fractus does not mean
'choppy1 in Seneca's usage (see n.44, p.360); I do not
believe that either scholastici or Atticists 'virtually
neglected1 rhythmic considerations.
14
See the analysis of Suas. 3.1, pp.246ff.
15
See p.281.
16
See pp.101-2.
•^ See appendix on p.326; the rhythm of the preceding clausula,

publicae securis acies (III Rl in my appendix) was not dis-
dained by Latro either.
18
See p.268.
19
Cf. Sussman, The elder Seneca 127ff.
20
See pp.251ff.
21
The name given by some theorists to the proof ex con-
sequentibus; see Quint. V.10.2.
22
Much helpful information about the use of this term is given
by H. Schepers s.v. Enthymem, Historisches Worterbuch der
Philosophie II (Darmstadt, 1972), 527ff.
23
Cf. Cic. Top. 13.55; Quint. VIII.5.9; the recurrence of this
kind of remark about the high esteem in which such y^&U0^
were held proves that there must be a degree of continuity in
the tradition between the theory of Aristotle's time and
Cicero's.
24
Cf. the definition of enthymema as sententia cum ratione
mentioned by Quintilian in V.10.2.
25
This observation may be Cicero's own. It is not justifiable
to infer (with Schepers, op. cit. 531) that evdiiuriya was so
used in any Stoic dialectical source used by Cicero, seeing
that he explicitly attributes the usage to rhetoricians.
26
Cf. Quint. V.10.2 on the usage of Cornificius.
27
For another example see Contr. X.4.25, quoted on p.313.
28
Around IX.3.87 Quintilian starts to lose patience with the
proliferation of types of expression classed as figurae by
some critics of his day.
29
Cf. Sussman, The elder Seneca 12Off.
30
So the MSS in the only passage (Contr. II.4.12) where the
singular occurs, but that is not to say it is certain, or
even probable, that Seneca the Elder preferred that form to
tricolon; he may even have written the word in Greek letters;
see n.13, p.342.
31
Pp.l63f.
32
See p.184.
33
Actually, though, one cannot make a rigid demarcation between
descriptiones and moralizing loci: a declamatory description
of amor was bound to be moralistic in tone; the description
of Xerxes in Suas. 5.1 could fairly be classed as a treatment
of the locus, 'quomodo animi magnis calamitatibus
everterentur' (cf. Contr. 1.7.17).
34
See list of descriptions, pp.21Of.
35
Cf. Lucr. 1.832: patrii sermonis egestas with which Winter-
bottom (note on Contr. VII pr. 3) contrasts Cicero's belief,
expressed, e.g., in Nat. deor. 1.4.8, that the inferiority of
Latin to Greek in verborum...copia was a thing of the past.
36
See p.315.
37
See pp.291f.
38
See pP.176ff.
39
See p.101.
k0
Quoted p.52.
1+1
Quoted pp,181f.
h2
See pp.278-9.
1+3
Cf. Quint. V.13.16 for the grouping: contraria et supervacua
et stulta.
**** Given that compositio is the standard Latin translation of

o\)v§eoL£ (OVOU&TGOV) . For the meaning of fracta see Quint.
1.10.31 (on modern theatre music) effeminata et impudicis
modis fracta; Demetr. Eloc. 189, auvdeats 6e dvontauaTLxn Mat
eouxuux TOLS x e x X a a y e v o u s *al daeyvous

yexpoug, oEa yaAuata ra EwTctdeta 6ua TO yaXaxwTepov and
ps.-Longin. Subl. 41.1, ytxpoTiotov 6' ou6ev OUTCOS ev xous
u<J>nXous d)£ pu^yos x e x X a a y e v o s Xoywv xal aeaoBrr
yevos, o£ov 6r) nupptxôu xal Tpoxotuou xal 6tx6peôu, xeXeov
eus opxno'Ttxdv
1+5
There is no unanimity among Greek critics as to the scope of
xaxoCnXov. in Demetr. Eloc. 186 xaxo£r)Xov is the fault into
which the writer falls if he fails in an attempt at elegance
(T$ yXacpuptji); it is distinguished from 6 c|>uxpos xaP a*Tnp, the
fault which comes of failing to attain to grandeur (xtp
yeYa*01tPeTieO. In ps.-Longin. Subl, 3.4 xaxoCnXov is
associated with puerility (yetpaxuai6es) , the fault which
results from striving after prettiness and sweetness, rather
than with tumidity (TO O L 6 O D V ) , the vice of unsuccessful
grand stylists. On the other hand in Hermog. Inv. IV.12 (Rh.
Gr. VI.202 Rabe) xaxoCnXov is made to embrace even more
faults than it does in Quintilianfs account: TO 6e xaxo^TiXov
yuveTau n xaTa TO dt6uvaTov rf XCXT& TO dvaxoXouôv, o xal
evavTuwyd eaTuv, n xaTa TO auaxpov n xaTa TO daeges n xaTa TO
a6uxov r[ xaTa TO Trj cpuaeu TioXeyLov, xa$' oils Tp6nous xal
dvaaxeudcoyev yaXuaTa Ta 6unYnyotTa ex3dXXovTes ws d'nuaTa.
Nor do the Latin grammarians who use the term agree as to its
definition. According to Diomedes (GL I 451.8ff.), cacozelia
est per affectationem decoris corrupta sententia, cum eo ipso
dedecoretur oratio, quo illam voluit auctor ornare. haec fit
aut nimio cultu aut nimio tumore. According to Sacerdos (GL
VI 455.12f.), cacozelia est, quae fit duobus modis, aut
magnarum rerum humilis dictio, aut minimarum oratio tumens.
1+6
There follow the examples of Murrediusf tumidity quoted on
pp.2O8f.
1+7
The following instances of the use of these words have been
taken into consideration; those marked with an asterisk are
not included in Bardonfs 'Lexique1 in Vocabulaire llff.:
corruptus: Contr. III.7 extra*; VII.1.25; 4.10; IX.2.27*;
5.17; X.4.22,23; X.5.23*; Suas. 1.12,16*; corrupte: Contr.
II.1.26; III pr. 15*; IX.2.21; X.4.18,23; 5.28; corrumpere:
Contr. II.6.8 (conjectural)*; IX.5.17*; 6.11; insania: Contr.
II.1.25*; VII.3.8*; IX.2.26,28; insanus: Contr. II.1.25*;
insanire: Contr. IX.2.26*; X.4.22*; 5.27; Suas. 1.12,13;
2.16; furor: Contr. II.1.25*; IX.2.27*; furiose: Contr. X.5.
21,23; cacozelia: Contr. IX.1.15; 2.28; Suas. 7.11; caco-
zelos: Suas. 2.16. It goes almost without saying that Bardon
is wrong in translating cacozelia as 'emphase, pathos1 and
cacozelos as femphatiquement1.
h8
Quoted in full on pp.l99f.
1+9
On verba impropria see Quint. VIII.2.2ff., where, after
citing the expressions Hibericas herbas and duratos muria
pisces contrived by orators in attempts to avoid undignified
expression, he avers that such departures from plain language
constitute a fault: id apud nos inproprium, ctMUpov apud

Graecos vocatur, quale est 'tantum sperare dolorem', aut,
quod in oratione Dolabellae emendatum a Cicerone adnotavi,
'mortem ferre', aut qualia nunc laudantur a quibusdam,
quorum est 'de cruce verba ceciderunt'.
50
Senecafs criticism is quoted in full on pp.208-9.
51
Others are quoted on p.199.
52
On hyperbole see Quint. VIII.6.67ff. With Maecenas1 criti-
cisms compare Hermog. Inv. IV.2 (Rh. Gr. VI.2O2f. Rabe) on
Homer's avoidance of HCLH6£,T)\OV in the passage which Dor ion
paraphrased.
53
Quoted in context on p.200.
5k
See p.199.
55
Cf. Quint. VIII.6.23ff.
56
Cf. SenecaTs remark on a similar sententia by Florus: numquam
Latro tarn incredibilis umquam figuras concipiebat ut in
ipso triclinio inter lectos et tlocot et mensas percussum
describeret (Contr. IX.2.24 W ) .
57
Quoted on p.215.
58
Contr. X.5 thema: Laesae rei publicae sit actio. Parrhasius,
pictor Atheniensis, cum Philippus captivos Olynthios venderet,
emit unum ex Us senem; perduxit Athenas; tor sit et ad
exemplar eius pinxit Promethea. Olynthius in tormentis perit.
ille tabulam in templo Minervae posuit. accusatur rei
publicae laesae.
59
Suas. 2 thema: Trecenti Lacones contra Xersen missi, cum
treceni ex omni Graecia missi fugissent, deliberant an et
ipsi fugiant.
60
Cited by Quintilian, VIII.3.28. On Cimber and the notion of
Attica febris see Leeman, Orationis ratio I 163; Lebek, Verba
Prisca 160ff.
61
See pp.l91ff.
62
E.g. Leeman, Orationis ratio I 187: fthe Atticist ideal and
slogan of sanitas1.
63
For references to scholars sceptical of their authenticity
see G.L. Hendrickson, fCicero's correspondence with Brutus
and Calvus on oratorical style 1 , AJPh XLVII (1926), 235ff.
Hendrickson rejects their arguments, as does Leeman, op. cit.
140f. Testimonia relating to this correspondence are col-
lected by W.S. Watt, Ciceronis Epistulae III (OCT, 1958),
168ff.
Cf. Brut. 13.51: nam ut semel e Piraeo eloquentia evecta est,

omnis peragravit insulas atque ita peregrinata tota Asia est,
ut se externis oblineret moribus omnemque illam salubritatem
Atticae dictionis et quasi sanitatem perderet ac loqui paene
dedisceret. hinc Asiatici oratores non contemnendi quidem
nee celeritate nee copia, sed parum pressi et nimis
redundantes; Rhodii saniores et Atticorum similiores.
65
N.b. the term xaxoCriAov was not peculiar to extreme Atticists
in Greek criticism. Demetr. Eloc. in which it occurs (186)
is written in koine Greek with only occasional Atticisms; see
G.M.A. Grube, A Greek critic: Demetrius on style (Toronto,
1961), 46ff., 133ff.
66
See K.M. Abbott, W.A. Oldfather, H.V. Canter, Index verborum
in Ciceronis rhetorica (Urbana 111., 1964).
67
E.g. De or. 11.75.305: ex quo etiam illud adsequor, ut si
quis mihi male dicat, petulans aut plane insanus esse
videatur; Brut. 66.233: C. Fimbria...qui omnia magna voce
dicens verborum sane bonorum cursu quodam incitato ita
furebat tamen, ut mirarere tarn alias res agere populum, ut
esset insano inter disertos locus; Or. 71.236: composite et
apte sine sententiis dicere insania est; ibid. 70.232:
quantum autem sit apte dicere, experiri licet, si aut com-
positi oratoris bene structam conlocationem dissolvas per-
mutatione verborum; - corrumpatur enim tota res ut [et] haec
nostra in Corneliana et deinceps omnia
68
This grouping of stultum with contrarium among the rerum
vitia has some precedent in Cicerofs remarks in Part. orat. 38.
133 on the unfortunate consequences which are inevitable when
a defence counsel's interpretation of an ambiguous written
statement is rejected: fore uti multa vitia, stulta, iniqua,
contraria consequantur.
69
The following passages have been taken into consideration;
those marked with an asterisk are not included in Bardon's
'Lexique' in Vocabulaire llff.: stultus: Contr. 1.3.11; 4.12;
V.2 extra; VII.5.10*; IX.4.22; X.5.25; Suas. 2.18*; 7.14;
stulte: Contr. X.5.25; stupor: Contr. VII.2.14; 3.8; IX pr.
2; Suas. 2.21*; fatuus: Contr. VII.4.3; 5.15; IX pr. 2;
X.5.25; Suas. 2.22; ineptus: Contr. 1.4.7; 6.11*; II.3.13*;
VII.3.10; 4.3; 5.10*, 11, 13; 7.14; IX.2.20; 6.11,12*;
inepte: Contr. 1.6.11*; IX.2.24*; X.I.12*; ineptiae: Contr.
IX.6.10.
70 See p.215.
71
The part of the theme of Contr. VII.5 relating to the step-
mother's wound is quoted on p.112.
72
Supplied by H.J. MUller.
7
^ Miller and Winterbottom here print conjectures.
7h
CQ N.S. XIX (1969), 321. See, however, the objections of L.
Hakanson, 'Some critical notes on Seneca the Elder 1 , AJPh
XCVII (1976), 121ff., who proposes: nam cum coepisset schol-
asticorum frequentissimo iam more, ut quam primum tantum
tumeant quantum potest, a iureiurando, et dixisset multa,
<ait> 'ita aut...
75
See p.224; compare also Polliofs remarks on Latro's elimin-
ation of ineptae quaestiones in Contr. II.3.13, quoted on
p.162.
76
See p.176.
77
See Contr. II.4.12; X.4.23 for the expression; also pp.
174f. for colores running contrary to themes and to the
declaimer's interests.
78
In this controversia much hinged on the question whether the
crime committed by the mutilator of beggars, however des-
picable, constituted res publica laesa.
79
Here Mliller prints Gertz' conjecture, parilem.
80
On the stepmother's wound, see p.112.
8
* Here Winterbottom prints Shackleton Bailey's conjecture
licentissimi (see CQ N.S. XIX (1969), 320), but the MSS
reading decentissimi seems defensible. Seneca the Elder was
capable of making such statements as nihil est autem
amabilius quam diligens stultitia (Contr. VII.5.12).
Decentissimi generis stultam sententiam can be viewed as one
of a series of whimsical allusions to fatuity in Suas. 2:
compare sed si vultis historicum quoque fatuum dabo (§22);
sed ne vos diutius infatuem (§23).
Memoria
1
Pp.37ff.
On mnemonic techniques in antiquity, the Middle Ages and the

Renaissance, see F.A. Yates, The art of memory (London, 1966).
Sussman, The elder Seneca 131f. includes a brief discussion
of Seneca's remarks on memory.
^ Cicero (De or. 11.88.359) mentions a system in which standard

symbols are contrived to represent even conjunctions, but
Quintilian (XI.2.25) considers such a system impracticable.
** So printed by H. Caplan (Loeb ed. 1954), on whose divergences

from the editions of F. Marx see his introduction, xxxvii ff.
5
Cf. p.37.
6
See Contr. I pr. 18 and pp.l84f.
7
The standard example of great length in literature, cf. Tac.
Dial. 20.1; Quint. XI.2.25.
8
Note, however, Contr. I pr. 22, for evidence of his ability
to improvise sententiae orally.
9
Quintilian (XI.2.36f.) was appreciative of the usefulness of
a well-constructed divisio as an aid to memory: nam qui recte
diviserit, numguam poterit in rerum ordine errare: certa sunt
enim non solum in digerendis quaestionibus sed etiam in
exequendis, si modo recte dicimus, prima ac secunda et dein-
ceps, cohaeretque omnis rerum copulatio, ut ei nihil neque
suhtrahi sine manifesto intellectu neque inseri possit.
Actio
* Sussman, The elder Seneca 132ff. includes a discussion of

Seneca's remarks on delivery.
2
Cf. Contr. VII.4.10 on uncritical applause in the schools.
^ Cf. Contr. I pr. 14: cum vero se silvis montibusque

tradiderat, in silvis ac montibus natos, homines illos
agrestis, laboris patientia et venandi sollertia provocabat
et in tantam perveniebat sic vivendi cupiditatem, ut vix
posset ad priorem consuetudinem retrahi.
4
Cf. Contr. VII.4.6: Calvus, qui diu cum Cicerone
iniquissimam litem de principatu eloquentiae habuit...
5
Suas. 2.17: nam et servos nolebat habere nisi grandes et
argentea vasa non nisi grandia. credatis mihi velim non
iocanti, eo pervenit insania eius, ut calceos quoque maiores
sumeret, ficus non esset nisi mariscas, concubinam ingentis
staturae haberet.
6
Quoted in full on p.277.
Asianism, Atticism, and the styles of the declaimers

1
E. Norden, Die antike Kunstprosa (Leipzig, 1898 repr.
Stuttgart, 1958) 1.152: %Wir werden nun spaterhin den Jahr-
hunderte lang dauernden Kampf dieser beiden Richtungen zu
verfolgen haben, und zwar wollen wir dabei den Asianismus
als den "neuen Stil", den Atticismus als den "alten Stil"
bezeichnen.1 Norden's use of these terms is criticized in
Wilamowitz1 important article 'Asianismus und Atticismus1,
Hermes XXXV (1900), Iff.
2
Antike Kunstprosa I 263-73.
^ See especially Contr. VII pr. 5, quoted on p.214.

4
See pp.2OOf.; 214; 238.
5
Antike Kunstprosa I 263-5.
6
Cf. p.238.
7
See pp.296ff.
8
N.b. such an expression as cum esset ex Asia would be con-
ceivable at least in colloquial Latin (cf. Plaut. Bacch. 472;
Rud. 737) but for the expression ex Asianis see Contr.
IX.1.12.
9
It is improbable that the minor declaimers Dionysius Atticus,
trained by Apollodorus of Pergamum (Contr. II.5.11), and
Antonius Atticus, criticized for extreme puerility in Suas.
2.16, were distinguished for their Attic preferences in
rhetorical style.
10 Asianus is Bursian's certain conjecture; the MSS have asinius.
11
Lugent is Haase's conjecture; Winterbottom reverts to the MSS
reading legunt, which for me, however, strikes a false note.
For funerary imagery in the prose of cacozeli and Asiani see
pp.218, 250.
12
Quoted in context on p.52.
*^ E.g. Cic. De or. III.48.184, where the recommendation,

orationem...non adstricte, sed remissius numerosam esse
oportere, is attributed to Theophrastus.
*** Edward ad loc. actually suggests that this clause freads like
a marginal note inserted in the text1.
15
See pp.2Olf.
16
Normally confined to poetry at this date, though there was
precedent for its use in prose in Cic. De or. 11.44.187.
17
P. Lunderstedt, De C. Maecenatis fragmentis {Commentationes
Philologicae Jenenses XI fasc. 1) (Leipzig, 1911). For
further bibliography relating to the text of these very
obscure fragments see the apparatus to L.D. Reynold's OCT of
Seneca, Epistulae morales, ad loc.
* 8 Contrast, e.g., Sen. Nat. quaest. V.3.1: aer nubilo gravis

est.
19
Strabo XIV. 1.41: 6"s ?ip£e uaAtaTot xou 'Aauxvou Aeyouevou
CnAou; cf. Cic. Or. 69.230.
20
I 134-9.
21
However, as Norden (Antike Kunstprosa I 138) notes, this
phenomenon was already present in the style of Alcidamas, who
is criticized for it by Aristotle in Rhet. III.3.3.
22
Cf. p.186.
23
Cf. p.158.
lh
Cf. Contr. 1.6.12.
25
Assuming that Wachsmuth was correct in transposing it to this
point; see Mlillerfs apparatus.
26
Cf. p.226; p.364 n.77.
27
Cf. p.209.
28
The excerpta following those taken from the extant Latronian
passage (seram querellam...defendatur uxor) seem to represent
variant treatments of topics treated earlier in Latrofs
arguments, and are probably to be ascribed to other declaimers.
Note that the Excerpta show that Seneca recorded at least one
extract from the treatment of the altera pars of this decla-
mation, and therefore did not confine his attention to
LatroTs treatment of the husband's case.
29
Unless we accuse him of contrarium in §6; cf. pp.226f.
30
I cannot accept the view of K. BUchner, TEin Stilwechsel
Ovids 1 , Mus. Helv. XIII (1956), 183, that the diversion
dicendi genus referred to was poetry as opposed to rhetoric.
When referring to literary genres other than rhetoric the
elder Seneca uses the terms artes (Contr. II pr. 3) and
studia (ibid, 4 ) ; Cassius Severus uses opus in a similar
sense in Contr. Ill pr. 8. The term genus dicendi seems
always to refer to a type of rhetorical style; see Contr.
II pr. 1; III pr. 7; VII pr. 5; IX.5.17; X pr. 2.
31
Quoted on p.245.
32
Cf. p.5.
33
Bonner, Roman education 20ff. discusses the various levels of
tuition provided by Greeks in Roman households and, p.178,
the amount of Greek required before a boy could profit from
the Greek exercises and study of texts set by grammatici.
34
See Jerome on Olympiad CXCI,4 (p.249 Fotheringham).
35
See pp.H7ff.
36
See pp.245ff.
37
See pp.llOff.; 165.
38
See p.205.
39
On Timaeus1 style see also A.E. Douglas, ed. Brutus (Oxford,
1966) ad loc.
**" Although Polybius quotes Timaeus in indirect speech up to

this point, the lack of any marked change of style in the
subsequent direct quotation shows that he is keeping close to
the original from the outset.
*** See ORF^ passim. Admittedly sententiousness is not entirely

lacking in early Ciceronian oratory, but it is not to be
supposed that it need be the product of Asiatic influence.
Compare, e.g., Rose. Am. 89, tempus hercule te citius quam
oratio deficeret, with Isocr. Archid. 81; Dem. De cor. 296.
For similar sententiae in declamation see Contr. 1.5.9.
1+2
Why, for instance, if iudices so disliked having their time
wasted and their intelligence insulted, did a plain style not
become more generally favoured, and why did the treatment of
philosophical and historical topics in rhetoric not become
less, rather than more, trite?
43
lite Obrecht: luce MSS; Lehnert.
1+4
tandem scripsi: tamen MSS.
1+5
Tusc. disp. II.8.2Of. trans. Soph. Trach. 1060ff. Sussman,
The elder Seneca 37, also recognizes that drama provided
precedent for declamatory sententiousness.
1+6
A. Dihle, fEin Spurium unter den rhetorischen Werken Ciceros'
Hermes LXXXIII (1955), 3O3ff., argues that this work is not
by Cicero.
hl
See pp.lOlf.
48
P.101.
49
Quoted on p.259.
50
See p.106, on their attitude towards philosophy.
51
See pp.l66ff.
52
See Philod. Rhet. II.216ff. Sudhaus.
53
See Cic. Brut. 95.325; Suet. Aug. 86.3.
54
I am unconvinced by the interpretation of this sentence pro-
posed by K. BUchner, Mus. Helv. XIII (1956), 180: 'Naturlich
ist sein zweiter Teil keine BegrUndung, sondern korrigiert
wie so h&ufig eine Erwartung Oder Vorstellung. Offenbar war
bekannt die Verbindung Ovids mit Latro. Seneca sagt betont,

Ovid sei der auditor, der Horer des Arellius Fuscus gewesen:
jawohl; denn die allgemein bekannte Verbindung mit Latro war
erst sp'dter wirksam.' For objections to this last assertion
see n.30, p.367. Nor does it seem quite natural to interpret
it as Winterbottom does in his translation, 'I remember this
controversia being declaimed by Ovidius Naso at the school of
the rhetor Arellius Fuscus - Ovid being his pupil. He was an
admirer of Latro, though his style of speech was different.1
To take nam Latronis admirator erat to mean 'for his position
in relation to Latro was merely that of an admirer1 places
too much strain on the Latin, given that the elder Seneca was
not, merely chatting elliptically to his sons and other people
of his time who were perhaps already aware of Ovid's debt to
Latro, but was writing for posterity (Contr. I pr. 11).
Posteri could not be expected to take the words to mean any-
thing other than ffor he was an admirer of Latro f .
^ 5 See Mlillerfs apparatus ad loc.

56
We have to suppose that a scribe, after copying cuius auditor
fuit, felt a certain compulsion to bring like closer together
with like and, skipping cum...genus, went straight on to copy
nam Latronis admirator erat. (It was perhaps at this stage
that the excerpta were compiled. They give the reading:
Ovidius Naso apud Arellium Fuscum magistrum suum hanc contro-
versiam declamavit; nam Latronis admirator erat,) Later a cor-
rector noted in the margin the words omitted and the next copy-
ist reinserted them incorrectly in the main body of the text.
For cases of transposition where a similar sequence of events
has to be postulated see, e.g., Contr. II.7.4/5 (Wachsmuth),
Contr. VII.4.3/5 (MUller), and n.17, p.335.
57
On Ovid's affinities with Fuscus see T.F. Higham, 'Ovid and
rhetoric', Ovidiana ed. N.I. Herescu (Paris, 1958), 35f.
58
Op. cit. 39; he cites as illustrations Met. 1.292; II.263f.;
Pont. III.1.19f. Even where Ovid includes poetic descrip-
tions of greater length, e.g. Met. XI.474ff. (shipwreck),
XII.39ff. (Fama), his style loses none of its characteristic
tautness.
59
Cf. Quint. IX.4.61: neque enim loqui possum nisi e syllabis
brevibus ac longis
60
For examples of metrical snatches from all the best classical
prose stylists see Quint. IX.4.72-8.
61
N.b. later in Contr. II.2.8 it is reported that the
scholastici learnt a sentence from a praefatio by Latro
quasi carmen.
62
I am indebted to Dr J.B. Hall for this rendering. Experiment
along these lines was initiated by R. Ehwald in his Ad
historiam carminum Ovidianorum recensionemque symbolae
(Progr. Gotha, 1892).

63
Pp.251ff.
6I+
See, e.g., Contr. II.1.25f.: hie controversiam postero die
quam erat a Fabiano dicta declamavit; solebat autem sic ad
locos pervenire, ut amorem descripturus paene cantantis modo
diceret...; VII pr. 3: adfectus efficaciter movit, figurabat
egregie, praeparabat suspiciose. nihil est autem tarn
inimicum quam manifesta praeparatio.
65
It includes two examples of anthypophora (anticipation of
objection) and one tricolon, two of the very few kinds of
figurae which the elder Seneca mentions by name (see pp.
2O8f.).
66
For the theme see p.159.
67
But n.b. such repetitions were not confined to the Fuscine
descriptive manner. A Latronian example of repeated opening
rhythm in conjunction with anaphora may be found in Contr.
II.7.5: quotiens absentis viri nomen inploravit? quotiens,
quod rion una peregrinaretur, questa est?
68
See WinterbottomTs notes ad loc.
69
For Fuscus1 preferences see pp.326f.
70
...effugium sit; ...amnesqu(e) imitantur.
7
* .. .corporibus consternitur; .. .proximumque divmo genus;
.. .parricidium tanti fuft; .. «divitiae corrupennt.
72
Cf. Leeman, Orationis ratio I 264ff. on the two passages.
Leeman's supposition, however, that there was a difference of
opinion between the two Senecas as to the necessary constitu-
ents of a sententia (266f.) can hardly be right: the younger
Seneca's criticism, deest illis oratorius vigor stimulique
quos quaeris et subiti ictus sententiarum (§8) is a para-
phrase of his father's deerat illi oratorium robur et ille
pugnatorius mucro (Contr. II pr. 2 ) ; the fact more likely is
that whereas Fabianusf declamations often enough included
fere dulces sententias, they were less common in his philo-
sophical works.
73
E.g. in the Satires and Odes of Horace, where he treats
similar themes.
7Lf
See ThLL s.w.
75
See p.352 n.3.
76
See also, however, p.294, on Moschus.
77
See p.199.
78
See p.238.
79
See p.10.
80
Perhaps a famous passage; at any rate Quintilian alludes to
one sentence in it in IX.2.91. The theme of Contr. II.3 is
as follows: raptor raptae patrem exoravit, suum non exorat.
accusat dementiae; the legal position presupposed is that
raptor, nisi et suum et raptae patrem intra dies triginta
exoraverit, pereat.
81
S.v. Iunius 77, RE X (1917), 1035ff.
82
See Cic. Brut. 8.33 cf. Or. 50-51.169.
83
So Winterbottom1s translation after e.g. Quint. IX.4.22.
81+
Though n.b. the rhythmic sequence was not without parallel
in the usage of Seneca and Latro; see appendix, p.326 (V R4) .
85
See Contr. Ill pr. 6.
86
See pp.292f.
87
Orationis ratio I 231.
88
See n.34, p.367.
89
See pp.54, 89.
90
Bardon, Vocabulaire 33, is totally unconvincing in his defi-
nition of floridus as ffleuri (c'est a dire: releve d'images) T;
such a description certainly does not fit PollioTs declamatory
manner. Probably Seneca meant by floridior something like
sententiis dulcior. Cf. Quint. XII.10.60 on the middle
(floridum/dvdripov) style: medius hie modus et tralationibus
crebrior et figuris erit iucundior, egressionibus amoenus,
compositione aptus, sententiis dulcis, lenior tamquam amnis
et lucidus quidem sed virentibus utrimque ripis inumbratus,
though n.b. SenecaTs criticism is singularly deficient in
references to the classical theory of three styles which is
being described by Quintilian in XII.10.58ff.
91
N.b. theorists on compositio advised against the use of too
many short words in succession e.g. Dion. Hal. De comp. 12.
92 See p.197.
93
Given Albucius* fluency, BursianTs conjecture in §2, non
hexis magna, can hardly be right.
9
The remarks on Albucius1 technique for manipulating emotions
call for some elucidation. We have observed (p.208) that
Quintilian regarded figurae as devices aimed at persuasion,
not as mere ornaments, and that this view was shared by Latro;
hence the inclusion of figurabat egregie between two criti-

cisms relating to persuasiveness. Suspiciose cannot mean fde
maniere a eveiller les soupcons' (so Bardon, Vocabulaire 56),
seeing that the declaimer wants above all to inspire con-
fidence in his hearers. Winterbottom's interpretation in his
Loeb translation, 'He was...skilled at allusiveness in his
preparation1, seems the right one. Praeparatio, on which
Seneca (loc. cit.) proceeds to comment, nihil est autem tarn
inimicum quant manifesta praeparatio; apparet enim subesse
nescio quid mail, was the most common of the several var-
ieties of the figure praesumptio/Tipb\r\\i<\>L£ recognized by the
theorists; it is used cum id quod obici potest occupamus
(Quint. IX.2.16) and consists of hints made usually in the
prooemium as to the arguments which one is intending to use.
Seneca (loc. cit.) also advises moderation to prevent one's
praeparatio from turning into a confessio. A confessio was
another variety of praesumptio in which one conceded a point
to one's adversary before proceeding to argue against him on
other counts.
95
Pp.l63f.
96
See p.192. In §2 the words magna...phrasis seem suspect.
97
The whole passage is quoted on p.214.
98
The evidence of this preface has to be considered separately
for reasons explained on pp.47ff.
99
Cf. pp.269f.
100
Here I have adopted Wachsmuth's transposition, desinere...
dicere, whereas Mliller and Winterbottom keep the MSS reading,
scire dicere quam scire desinere.
101
See pp.276f.
102
Cf. Fabianus' description of Ocean quoted on p.271f.
103
Ad Hor. Epist. 1.5.9.
104
Valenter is the reading of the excerpta MSS; the editors
prefer the lectio difficilior of the 'full text' MSS,
violenter9 but it seems to me too difficult to accept that
Seneca linked two expressions of opprobrium with sed. For
the use of valenter with reference to Sparsus' model, Latro,
see Contr. VII.7.10.
105
See p.239.
106
See pp,191ff.
107
See p.177.
108
On the theme of this controversia see p.172.
109
Quoted on p.214.
110
See p.245.
111
See pp.283f.
112
See p.245.
113
See Norden, Antike Kunstprosa I 267.
llh
See pp.221ff.
115
See pp.216ff.
116
Antike Kunstprosa I 263ff.
117
The parody is of a type of declamation which calls for much
gesticulation; compare the passage of Iulius Bassus quoted on
p.295, and SenecaTs disapproval of his actio (Contr, X pr.
12) . Such a rhythm as date iriiha ducem.. . (assuming ducem,
deleted by K. MUller after Jacobs, to be in fact Petronian),
might well have been considered by Seneca as a case of
emollita compositio. Cf. Contr. VII.4.8, discussed on p.101,
a passage which proves, incidentally, that even 'effeminate1
rhythms were not a monopoly of the Asiani. See also p.261,
for a comparison between the Petronian parody and the
emotional rhetoric of Attic drama, as translated by Cicero.
118 See pp.lOlf.

119
See pp.5f.
120
Antike Kunstprosa I 152.
Norden (op. cit. 265f.) dismisses Dionysius as fein falscher

Prophet1, and his account of an Atticist revival as the wish-
ful thinking of a Stubengelehrter.
Declamation and literary modernism in the early Empire

1
See pp.292f.
2
These sequentes are apparently given the alternative name of
novi declamatores in Contr. 1.1.14, but why Seneca should
have classed some of the declaimers he does as novi (e.g.
Cestius, loc. cit.) is mysterious since, as Hoffa (Quaest.
sel. 7ff.) has observed, there is not always a chronological
justification for describing them as new. It is on the basis
of Contr. I.1.13f. that I make a very tentative suggestion in
my forthcoming article, fThe elder Seneca and declamation1,
in ANRW, n.163, that the term might have originated in
attempts by Latro to pose as a preserver of antique standards,
in which he denigrated anyone who departed from his boasted
classicism as novi.
3
See pp.285f. But n.b. paradoxically we have noted certain
seemingly archaic features in the declamatory fragments of
Cassius Severus who was to be considered later (Tac. Dial,
19.1; 26.4) as the first of the moderns. Also bear in mind
that Dionysius of Halicarnassus (Rhet. vet. 2) thought he
detected an almost universal movement back to classicism
among students of Greek rhetoric in his day.
4
See pp.77ff.
5
See pp.22Of.
6
See Edward and Winterbottom ad loc.
Epochs of Greek and Roman biography, Sather Classical Lec-

tures IV (Berkeley, Calif., 1928), 36. The most famous
example of ThucydidesT use of this device is the assessment
of Pericles after the mention of his death in II.65.6; the
brief appraisal of Nicias in VII.86.5 is certainly another
example; the more extensive estimate of Themistocles1 gifts
in 1.138.3 precedes, rather than follows, the narrative of
his death. On the term eiLTacptov see n.19, p. 339.
8
See R. Syme, Sallust (Berkeley and Cambridge, 1964), 274ff.
9
See pp,14ff.
^ Cf. Ep. mor. 59.7: Sextium ecce cum maxime lego, virum acrem,
Graecis verbis, Romanis moribus philosophantem.
11
Quoted on p.22.
12
See Winterbottom1s notes on Contr. II.1 passim, and for
declamatory examples his Index III (Vol.11 638ff.) s.w.
luxury, rich men, poor men.
13
See H.I. Marrou, Histoire de l'education dans l'antiquit^
246ff.; Bonner, Roman education 212f.
14
Quoted on pp.l99f.
15
Cf. Sen. Ep. mor. 122.11, quoted on p.316.
16
See p.97.
17
Cf. Ovid's remark, aiebat interim decentiorem faciem esse in
qua aliquis naevos esset (Contr. II.2.12), for the meaning of
decentius in Seneca's appreciation of Virgil.
18
See pp.249.
* 9 The phrase does not occur in our texts of Virgil. See Edward
and Winterbottom ad loc; E. Fraenkel, Horace (Oxford, 1957),
199 n.l; F. Delia Corte, 'Plena deo T , Maia XXIII (1971),

102ff.
20
See p.195.
21
See pp.264ff.
22
Muller is more certain than Winterbottom that there is a
lacuna here. Certainly it is possible, as Winterbottom
observes ad loc., that the third quotation was unknown to
Seneca, but it seems a remote possibility, seeing how fond
story-tellers always are of three-part narration. Seneca
would surely have felt the need to apologize supposing he
had forgotten the third versus.
23
Winterbottom prefers to del-ete the MSS 1 si tarn before recta
via. For a different reconstruction of this corrupt piece of
text see D.R. Shackleton Bailey, CQ N.S. XIX (1969), 321.
24
C O . Brink, Horace on poetry: prolegomena to the literary
Epistles (Cambridge, 1963), 159, compares Callim. Hymn to
Apollo 108f.:
'Aaaupuou TtOTayouo yeyas poos, aXXa xa uoXXa
Auuaxa yns xal noXXov ecp' u6axb aupcpexov eXxeL.
25
See p.312.
26
See p.194.
27
Probably that of the younger Germanicus in A.D. 16. See
Winterbottom ad loc.
28
See H. Homeyer, 'Klage urn Cicero. Zu dem epischen Fragment
des Cornelius Severus 1 , Annales Univ. Saraviensis (phil.
Fak.) X (1961), 327ff.; V. Bongi, 'Nuova esegesi del
frammento di Albinovano Pedone 1 , Istituto Lombardo di scienze
e lett. Rendiconti (Classe di Lettere) LXXXII (1949), 28ff.;
V. Tandoi, 'Albinovano Pedone e la retorica guilio-claudia',
Studi Italiani di Filologia Classica XXXVI (1964), 129ff.
29
P.136.
30
So C.W. Mendell, Latin poetry, the age of rhetoric and satire
(Hamden, Conn., 1967), 10.
31
Though n.b. some Republican grammatici taught a considerable
amount of rhetoric. See p.5.
32
See pp.l46f.
33
See p.80.
34
See p.209.
35
Cf. pp.H8ff.
36
See p.118.
37
See pp.90f.
38
See L. Friedl&nder, De Senecae controversiis in Gestis
Romanorum adhibitis (Kb'nigsberg, 1891); MUller's ed., introd.
Vllf. n.l.
39
So Dio, XLVII.11.1.
BIBLIOGRAPHY
EDITIONS
The text in my quotations from Seneca the Elder normally represents
the consensus of the editions of H.J. MUller (Vienna, 1887, repr.
Hildesheim, 1963) and M. Winterbottom (Loeb ed. with English trans-
lation, Cambridge, Mass, and London, 1974). Where these editions
differ I indicate which of them I am following by means of the
letter M or W appended to the book and section reference, usually
only giving reasons for my preference or including any discussion
of textual matters in cases where the interpretation of Senecafs
criticism is at issue. Asterisks are used to indicate any notes
which contain discussion of problems in the text of Seneca the
Elder or other ancient authors.
Occasional mention is also to be found in this book of all
the other nineteenth- and twentieth-century editions of Seneca the
Elder, besides those of MUller and Winterbottom, namely those of:
C. Bursian, Leipzig, 1857
A. Kiessling, Leipzig, 1872, repr. Stuttgart, 1967
H. Bornecque, Paris, ed.1 1902 and ed.2 1932 (with French
translation)
W.A. Edward, The Suasoriae of Seneca, Cambridge 1928 (with
English translation and commentary).
Among earlier editions I cite only the Editio Commeliniana of
the two Senecas by A. Schott and M.A. Muretus, Heidelberg, 1603/4.
The preface by Schott from which I quote is also printed amongst
other introductory material in the important Elzevir edition,
Amsterdam, 1672, with commentaries by J. Schultingh and others.
Classification and description of the early editions is provided
in an article by H.D.L. Vervliet, 'De gedrukte Overlevering van
Seneca Pater1, De Gulden Passer XXXV (1957), 179ff.
SECONDARY WORKS
No attempt is made here to do more than set down a list of works
(excluding those on non-literary matters) cited in this book. A
useful, though not exhaustive, bibliographical study is provided
in J.E.G. Whitehornefs 'The Elder Seneca: a review of past work1,
Prudentia I (1969), 14ff. This will be supplemented in due course
by a detailed bibliography by L.A. Sussman, and my own article,
'The elder Seneca and declamation', both forthcoming in Aufstieg
und Niedergang der romischen Welt.
Bibliography 378
Abbott, K.M., Oldfather, W.A., Canter, H.V. Index verborum in

Ciceronis rhetorica (Urbana, 111., 1964).
Atkins, J.W.H. Literary criticism in antiquity (Cambridge, 1934).
Bardon, H. Le vocabulaire de la critique litteraire chez Seneque

le Rheteur (Paris, 1940)
'Mecanisme et stereotypie dans le style de Seneque le Rheteur1,

AC XII (1943), 5ff.
Barwick, K. ed. Cicero, Brutus (Heidelberg, 1949).
Boissier, G. 'Les ecoles de declamation a Rome 1 , i?evue des Deux

Mondes XI (1902), 48Off.; trans, by W.G. Hutchison, 'The
schools of declamation at Rome1 in Tacitus and other Roman
studies (London, 1906), 163ff.
Bongi, V. fNuova esegesi del frammento di Albinovano Pedone',

Istituto Lombardo di scienze e lett.: Rendiconti (Classe di
lettere) LXXXII (1949), 28ff.
Bonner, S.F. Dionysius of Halicarnassus (Cambridge, 1939).
'Rhetorica', CR LXI (1947), 84ff.
Roman declamation in the late Republic and early Empire

(Liverpool, 1949).
Education in ancient Rome from the elder Cato to the younger

Pliny (London, 1977).
Bornecque, H. Les declamations et les declamateurs d'apres

SSneque le Pere (Lille, 1902, repr. Hildesheim, 1967).
f
Le texte de Seneque le Pere 1 , RPh N.S. XXVI (1902), 36Off. and
XXVII (1903), 53ff.
Brink, C O . Horace on poetry: prolegomena to the literary

Epistles (Cambridge, 1965).
f
BUchner, K. Ein Stilwechsel Ovids?', Mus. Helv. XIII (1956),
180ff.
Buschmann, J. Charakteristik der griechischen Rhetoren beim

Rhetor Seneca (Progr. Parchim, 1878).
Die 'enfants terribles' unter den Rhetoren des Seneca (Progr.

Parchim, 1883).
Caplan, H. fThe decay of eloquence at Rome in the first century',

Studies in speech and drama in honor of Alexander M. Drummond
(Ithaca, N.Y., 1944), 295ff.
Cerrati, M. La grammatica di A. Seneca, il retore (Turin, 1908).

Bibliography 379
Clarke, M.L. 'The thesis in the Roman rhetorical schools of the

Republic1, CQ XLV (1951), 159ff.
Cordier, A. Review: Bardon, Vocabulaire, RPh 3me ser. XVII

(1943), 22Off.
D'Alton, J.F. Roman literary theory and criticism (London, 1931).
Dalzell, A. f C. Asinius Pollio and the early history of public

recitation at Rome 1 , Hermathena LXXXVI (Nov. 1955), 2Off.
T
Delia Corte, F. Plena deo f , Maia XXIII (1971), 102ff.
Dihle, A. 'Ein Spurium unter den rhetorischen Werken Ciceros*,

Hermes LXXXIII (1955), 3O3ff.
T
Douglas, A.E. M. Calidius and the Atticists1, CQ N.S. V (1955),
241ff.
'Clausulae in the Rhetorica ad Herennium as evidence of its

date 1 , CQ N.S. X (1960), 65ff.
ed. Cicero, Brutus (Oxford, 1966).
Cicero: Greece and Rome, New Surveys of the Classics II (Oxford,

1968).
Dover, K.J. fThe chronology of Antiphon's speeches1, CQ XLIV

(1950), 44ff.
Duff, J. Wight, Duff, A.M. A literary history of Rome in the

silver age 3 (London, 1964).
Ehwald, R. Ad historiam carminum Ovidianorum recensionemque

symbolae (Progr. Gotha, 1892).
Fairweather, J.A. f A transposition in the text of Seneca the

Elder, Contr. I.5.91, CR. N.S. XX (1970), llf.
Review: M. Winterbottom ed. and trans. Seneca the Elder, JRS

LXVI (1976), 27Off.
f
The Elder Seneca and declamation1, forthcoming in Aufstieg und
Niedergang der r&mischen Welt.
Fantham, E. Comparative studies in Republican Latin imagery,

Phoenix, Suppl. Vol. X (Toronto, 1972).
Fordyce, C.J. Review: Bardon, Vocabulaire, CR LX (1946), 129.
Fraenkel, E. Horace (Oxford, 1957).
Francois, G. 'Declamatio et disputatio\ AC XXXII (1963), 513ff.

Bibliography 380
Friediander, L. De Senecae controversiis in Gestis Romanorum

adhibitis (Progr. KBnigsberg, 1891).
Gerth, K. s.v. 'lunius' 77 (Iunius Gallio), RE Xi (1917), 1035ff.
Griffin, M.T. 'The elder Seneca and Spain1, JRS LXII (1972), Iff.
T
Imago vitae suae?, Seneca: Studies in Latin literature and its
influence, ed. C.D.N. Costa (London, 1974), Iff.
Seneca (Oxford, 1975).
de Groot, A.W. Handbook of antique prose rhythm (Groningen, 1919).
Der antike Prosarhythmus (Groningen, 1921).

f
Grube, G.M.A. Theodorus of Gadara1, AJPh LXXX (1959), 337ff.
A Greek critic: Demetrius on style (Toronto, 1961).
The Greek and-Roman critics (London, 1965).
Gruppe, 0. Quaestiones Annaeanae (Diss. Sedini, 1873).
Hagendahl, H. 'Rhetorica', Apophoreta Gotoborgensia V. Lundstr'dm

oblata (Gbteborg, 1936), 299ff.
Hakanson, L. fSome critical notes on Seneca the Elder1, AJPh

XCVII (1976), 121ff.
Hall, J.B. 'Seneca, Controversiae I praef. 12 f , PACA XII (1973),

11.
Hendrickson, G.L. 'Literary sources in Cicero's Brutus and the

technique of citation in the dialogue', AJPh XXVII (1906),
185ff.
'Cicero's correspondence with Brutus and Calvus on oratorical

style', AJPh XLVII (1926), 235ff.
Hess, R. Quaestiones Annaeanae (Diss. Kiel, 1898).
Higham, T.F. 'Ovid and rhetoric', Ovidiana, ed. N.I. Herescu

(Paris 1958), 32ff.
Hoffa, W. De Seneca Patre quaestiones selectae (Go'ttingen, 1909) .
Hofrichter, W. Studien zur Entwickelungsgeschichte der Deklamation

(Diss. Breslau Univ., publ. Ohlau, 1935).
Homeyer, H. 'Klage um Cicero. Zu dem epischen Fragment des

Cornelius Severus', Annales Univ. Saraviensis (phil. Fak.) X
(1961), 327ff.
Bibliography 381
Immisch, 0. 'Wirklichkeit und Literaturform1, RhM N.F. LXXVIII

(1929), 113ff.
Janson, T. Latin prose prefaces (Stockholm, 1964) .
Jenkinson, E.M. 'Further studies in the curriculum of the Roman

schools of rhetoric in the Republican period1, Symbolae
Osloenses XXXI (1955), 122ff.
Karsten, H.T. Elocutio Rhetorica qualis invenitur in Annaei

Senecae Suasoriis et Controversiis (Progr. Rotterdam, 1881).
Kennedy, G. The art of rhetoric in the Roman world (Princeton,

1972).
n
Kroll, W. s.v. Rhetorik\ RE Supptbd VII (1940), 1039ff.
Lebek, W.D. Zur Vita des Albucius Silus bei Sueton1, Hermes XCIV
f
(1966), 36Off.
Verba Prisca: Die Anfange des Archaisierens in der lateinischen

Beredsamkeit und Geschichtschreibung (Gb'ttingen, 1970).
Leeman, A.D. Orationis ratio (Amsterdam, 1963).
Leo, F. Die griechisch-r'dmische Biographie nach ihrer

literarischen Form (Leipzig, 1901, repr. Hildesheim, 1965).
Lockyer, C.W. The fiction of memory and the use of written

sources: convention and practice in Seneca the Elder and
other authors (Diss. Princeton, 1970).
Lunderstedt, P. De C. Maecenatis fragmentis, Commentationes

Philologicae Jenenses XI fasc. 1 (Leipzig, 1911).
Marouzeau, J. Review: Bardon, Vocabulaire, REL XVIII (1940),

2O3ff.
Marrou, H.I. Histoire de 1'Education dans I1antiquity (Paris,

1965); trans from ed.^ by G. Lamb, A history of education in
antiquity (New York, 1964).
Misch, G. Geschichte der Autobiographie Vol. I (Frankfurt, 1947);

trans. Misch and E.W. Dickes, A history of autobiography in
antiquity (London, 1950).
Momigliano, A.D. The development of Greek biography (Cambridge,

Mass., 1971).
Morel, W., Kroll, W. s.v. *Melete\ RE XVi (1931), 496ff.
Norden, E. Die antike Kunstprosa (Leipzig, 1898, repr. Stuttgart,

1958).
Bibliography 382
Pack, R.A. The Greek and Latin literary texts from Greco-Roman
Egypt2 (Ann Arbor, 1967).
Paoli, U.E. 'Droit attique et droit romain dans les rheteurs

latins', JRevxze Historique de Droit francais et Stranger XXXI
(1953), 175ff.
Parks, E.P. The Roman rhetorical schools as a preparation for the

courts under the early Empire, Johns Hopkins University
Studies in Historical and Political Science, Series LXIII
no. 2 (Baltimore, 1945).
Portalupi, F. Bruto e i neo-atticisti (Turin, 1955).
Preisendanz, C. 'De Senecae rhetoris apud philosophum filium

auctoritate', Philologus LXVII (1908), 68ff.
Rayment, C.S. 'Echoes of the declamations in the dialogues of the

younger Seneca1, CB XLV (1969), 51f., 63.
Robinson, R.P. ed. Suetonius, De Grammaticis et Rhetoribus (Paris,

1925).
Rolland, E. De 1'influence de SgnSque le p£re et des rheteurs sur

Seneque le philosophe (Ghent, 1906).
Russell, D.A. ed. 'Longinus', On the sublime (Oxford, 1964).
Sander, M. Quaestiones in Senecam rhetorem syntacticae (Diss.

Greifswald, 1872).
Der Sprachgebrauch des Rhetors Annaeus Seneca, 2 vols. (Berlin,

1877, 1880).
Schepers, H. s.v. %Enthymem% , Historisches W6rterbuch der

Philosophie II (Darmstadt, 1972), 527ff.
Shackleton Bailey, D.R. ed. Cicero's letters to Atticus I

(Cambridge, 1965).
'Emendations of Seneca "Rhetor1", CQ N.S. XIX (1969), 320ff.
ed. Cicero, Ad Familiares (Cambridge, 1977).
Shewring, W.H. 'Prose rhythm and the comparative method' , CQ XXIV

(1930), 164ff. and XXV (1931), 12ff.
Sihler, E.G. '8eTLxu>Tepov', AJPh XXIII (1902), 283ff.
Sochatoff, A.F. 'The basic rhetorical theories of the elder

Seneca', CJ XXXIV (1939), 347ff.
Stegemann, W. Review: Bardon, Vocabulaire, PhW LXIV (1944), 172ff.
Steidle, W. Sueton und die antike Biographie (Munich, 1951).

Bibliography 383
Stuart, D.R. Epochs of Greek and Roman biography, Sather

Classical Lectures IV (Berkeley, California, 1928).
Sussman, L.A. fThe elder Seneca as a critic of rhetoric1 (Diss.

Chapel Hill, N. Carolina, 1969).
'The elder Seneca's discussion of the decline of Roman

eloquence', Cal. Stud. Class. Antig. V (1972), 195ff.
The elder Seneca (Leiden, 1978).
Syme, R. Sallust (Berkeley and Cambridge, 1964).
Tandoi, V. 'Albinovano Pedone e la retorica guilio-claudia',

Studi Italiani di Filologia Classica XXXVI (1964), 129ff.
Thomas, E. 'Schedae criticae novae in Senecam rhetorem',

Philologus Supptbd VIII (1900), 159ff.
Throm, H. Die Thesis - Ein Beitrag zu ihrer Entstellung und

Geschichte, Rhetorische Studien XVII (Paderborn, 1932).
Vassileiou, A. 'A propos d'un passage de S^neque le Pdre (Contr.

II praef. 4 ) : la psychologie d'un pere ambitieux pour ses
enfants au I e r sidcle ap. J . C , Latomus XXXII (1973), 162ff.
Watson, G. The literary critics (Harmondsworth, 1962).
von Wilamowitz-Mbllendorff, U. Antigonos von Karystos (Berlin,

1881).
'Asianismus und Atticismus', Hermes XXXV (1900), Iff.
Wilkinson, L.P. Golden Latin artistry (Cambridge, 1963).
Winterbottom, M. 'Problems in the elder Seneca', BICS XXI (1974),

20ff.
Yates, F.A. The art of memory (London, 1966).

INDEXES
GENERAL INDEX
Abronius Silo,232,312,316,324 Aemilianus,19,224

abruptus,215-16,220,271 Aemilius Lepidus, M. (consul
A.D.11),7
abstract issues in rhetoric,
104-7,118-19,130-1,176,184, Aemilius Lepidus, M. (prae-
321-3,352 n.58; see also ceptor of Germanicus1 son),
thesis 7
abstract noun as subject of Aemilius Scaurus, Mamercus,7,
sentence,250 15,24-5,57,137,139-41,169-
70,195,234,245,282,291-2,
abundantia,215-16,220
304,315
Academics,105-6,118,126,
aequitas distinguished from
347 n.5
ius,155-61,355 n.13
accusatio within a decla-
Aeschines (Attic orator),
mation, 187-9
78-9,107-9,115,130,223,
actio: fifth part of the art 348 n.17
of rhetoric,55,60,96,101,
Aeschines (declaimer),79,176
149,153,212,228,235-9,244,
271,294-5,354 n.1,373 n.117; Aeschines of Miletus,257
actio in classification of
Aeschylus of Cnidos,257
theses,104; speech in a
court case,164-5,354 n.20, Afranius, L.,66
365 n.l
agitated style,96-7,190,206-
actors, techniques of,238-9 14,243-303,319-25
Adaeus,245,255,297,305 Agroitas of Massilia,194
adfectus,101,187,278,288 Aietius Pastor,7,176,284
admirator distinguished from Albinovanus Pedo,314,317,324,
auditor,264-70,368 n.54, 375 nn.27,28
369 n.56
Albius Tibullus,269,329 n.13
adversaries, answers to
Albucius Silus, C.,6,7,10,24,
imaginary,147,165,209,281
45,47,57,60,66-7,73,82,85,
advocates impersonated by 135,141,157,164,168-9,173-4,
declaimers,151,173,177 178,181,186,188,192,209-10,
212-14,228,233,235-7,278,
Aelius Aristides,347 n.ll
288-91,296-7,321-2,337 n.44,
Aelius Donatus,312,319 339 n.13,340 n.24,371 n.94
Aelius Seianus, L.,10,15,141, Alcidamas,107,155,367 n.21
306,343 n.20,354 n.19
Indexes 386
Alexandria,108 critic,18-19,23-5,28,50-73,
132-325; his influence on
Alfius Flavus,7,57,58,72,137,
later literature,28,50-1,60,
284
334 n.6,338 n.1,339 nn.2-5,
alioqui,11-14,46,329 nn.22-3, 340 n.24; knowledge of pre-
337 n.48 Augustan rhetorical theory
and practice,77,79-131,304,
alliteration,274
of pre-Augustan epistolo-
allusiveness,17O,215,22O graphy,87,13O,343 n.31,344
nn.38,41,352 n.57, of
amabilis,19,224,264,364 n.81
historiography,90-2,304-7,
amaritudo in prosecution, of philosophy,14,20-2,78,88,
220-1,294-5 304-5,307-10, of poetry,26,
304-5,311-18,324-5, see also
poetry
anachronism,5,224,227
Annaeus Seneca, L. (the
analogy and anomaly,198-9 Younger): his birth in Cor-
duba,9; his early move to
anasceuae and catasceuae,
Rome,328 n.11,330 n.29;
116,350 n.35
addressed by his father upon
Anaximenes, see Rhetorica entry into public life,11-14;
ad Alexandrum considered by his father
less intelligent than Mela,
Andocides,79
11; death of father before
anecdotes serving to enliven his exile,331 n.51; wrote
criticism,52-3,64,67,281, biography of his father,15-
285,314-17 16,331 n.48,332 nn.52,53;
acted as his fatherfs liter-
anger adduced as color,172
ary executor,15; had high
Annaeus Lucanus, M.,200,316 regard for his father's
Historiae9l6; historical
Annaeus Mela,8,11-14,21,27-9,
fragments sometimes attri-
228,277,310,330 n.36,331 n.
buted to him,16-17; por-
39
trayal of his father in
Annaeus Novatus (Gallio) , philosophical works,21-2,
10-14,21,27-9,228,277,330 n. 332 n.60,333 n.65; his
35 father's view of his philo-
sophical teachers,308-10;
Annaeus Seneca (the Elder):
dedication to him and his
evidence for his life and
brothers of the declamatory
personality,3-26,307-10,328
anthology,27-9,228; in-
nn.2,10-13,329 n.17,330 nn.
debtedness to his father's
26,29,31,35,36,331 nn.39,51,
work,28,71-2,334 n.6; his
332 nn.55,56,61,333 nn.76,
view of the decline of elo-
77,334 n.80; historical
quence, 133,276; his accept-
writings,15-17,332 nn.52-6;
ance of natural decline
form and character of his
after perfection,136; his
declamatory anthology,27-49,
views on corrupt style in
transmission of its text,
relation to his father's
33-7,48-9,254,333 n.79,335
criticisms,215-21,276; his
n.17,338 n.59,360 n.30,367
judgement on Maecenas'
n.28; his evidence for the
style,249-50,266,358 n.9;
way in which declamations
treated same philosophical
were composed and presented,
commonplaces as Fabianus,
151-239; as a literary
Indexes 387
273; his philosophical style Antonius Gnipho, M.,4,6,124

in relation to Fabianus and
Aper, M. (in Tac. Dial.),98,
the declaimers,275-6,333 n.
138,259,263,282,322-3,353
64,334 n.6,366 n.18; at-
n.9,368 n.42
tended declamations by Musa,
277; his judgement on Pol- aperte,169,297
lio f s style,286; his opinion
of Iulius Montanus,316 Apollodorei,56,80-1,159,180-1
Apollodorus of Pergamum,56,
Annius Cimber,22O,243,3O2,362
80-1,159,180-1,321,341 nn.
n.60
7,8,366 n.9
Annius Milo, T.,85-6,110,143,
Apollonius (declaimer),186,
287
357 n.12
avouxov6unTOV,215
Apollonius Rhodius,312
Anonymus Seguerianus,341 nn.
appellatio,116,350 n.36
7,8
applause,147,237,275,321,365
avTeyK^nua (type of defence),
n.2
166-7
ditpeii:ris,227
anthypophora,186,209,252,342
n.14,370 n.65 Aquila Romanus,99
Antigonus of Carystus,61-3, arcessitus,215-16,220,223
339 nn.15,18
archaism,190,196-8,220,243-4,
Antiphon (Attic orator),79, 280,286,304,318,374 n.3
348 nn.20,23
Archimedes,28
Antiphon (author of Tetra-
ardens,190,200-14,247,252-64,
logies) ,109-15,122,130,
300,302
165,260,348 nn.20-3,349 nn.
24-5,32 Arellius Fuscus,10,36,43-7,52-
3,57,153,156,173,175,179,
Antiphon (author of work on
182,187-9,199-201,210-14,
the interpretation of
219-20,238,243,245-51,255,
dreams),167,356 n.28
257,264-77,283,286,294,297-
antiquarian tastes in litera- 8,302,305,313,324,326-7,328
ture, 191-2, 196-8,220, 243, n.9,334 n.80,336 n.33,337 n.
286,302,317-18 40,358 nn.15,16,359 n.13,369
nn.54,56,57,370 nn.67,69
antiquarius,24 3,302,304
Argentarius,164,284,343 n.25
antiguus,168,22O,3O4; see
also oratio antiqua argumenta, argumentatio,32,81,
147,153,156-7,168,179,182-6,
dvTLCTTaaLS (type of defence) ,
204-5,206,214,238,251,271,
166-7
314
Antisthenes,107
argutus,257-9
Antonius, M. (consul 99 B.C.),
aridus,53,174,213-14,257,294
120-3,320,323
Ariobarzanes of Cilicia,62
Antonius, M. (triumvir),20,60,
84-7,102,120,123,125,143, Aristotle,54,62,79,105,109,
152,175,188,221,243-4,261-2, 134,136,155,184,204-6,310,
287,297-8,307,324,344 n.37 347 nn.4,13,349 n.30,359 n.
23,367 n.21
Antonius Atticus,219,366 n.9
Indexes 388
Arruntius, L.,139,209 222,245,345 n.61,346 n.63,

363 n.64
ars = activity requiring edu-
cation,18,168,320-5,367 n.30 Atticus, see Pomponius;
Dionysius
ars distinguished from natura,
59,228 audacity in invention and
style,215-16,219-20,223,246
Artemo(n),182,210,255
auditor distinguished from
a r t s of r h e t o r i c , 5 5 - 6 , 5 9 , 6 7 - 8 , admirator,264-70,368 n.54,
77-8,149,354 n . l
369 n.56
aaXTiyctTUQTOV , 2 1 5
Augustus, Caesar,6,16-17,26,
Asianism,56,100,222,243-264, 59,67,138,140,142,238,243-5,
296-303,365 n.l 286,293,297,302,304,319,344
n.48
Asian orators,244,255,257,264,
297,299,363 n.64 autem,189,270,370 n.64
Asianus, * Aai,av6s,56,222,
243-6,255,264,296,298,300,
302,366 nn.8,10,11,19
Bacon, Francis,51
Asilius Sabinus,10,139-41
bad taste,23-5,214-27
Asinius Gallus, C,58,139
barbarism,190
Asinius Pollio, C.,7-8,37,45,
Barrus,236
51-2,54,57,58,78,91-2,96,
139,141,146,162,165,168-9, bellus, belle,194,224
173-4,197,209,211,256,283,
285-6,289,304-5,307,309,
beneficiurn,80,157-8
316,324,328 n.10,329 nn.16, bilingual declaimers,120
17,330 n.31,344 n.48,356 n.
biographers as literary
21,364 n.75,371 n.90
critics,59-62,64-6
asper,171-2,201,279,281,285-6
biographical tradition, re-
Atellan farce,311 lation of the elder Seneca's
work to it,51,56-67,339 nn.
Athenaeus,41-2,49
13,16,18
Athens and Athenians,145,223,
biographies of orators and
264,299-300
rhetoricians,62-7
athletes in rhetorical exer-
biography of the elder Seneca
cises,112
by his son,15-16,322 nn.52,
Attalus (Stoic philosopher), 53
10,21,308-10,323
Blandus, see Rubellius
Attica feJbris,22O,362 n.60
bombast,144-5,214-20,277,299-
Atticism,56,67,95-102,221-3, 300,312
243-64,296-303,345 n.61,346
book-burning,15-16,19-20,22,
n.63,363 n.65,365 n.1,366
140,293
n.9,373 n.121
brevity,190,200,210,213-16,
Attic orators,78-80,96-101,
218,220,254,259,266-7,271,
223,255,262,299,301,334
275,290,296,312-13,315
n.7,345 n.55,368 n.41
Bruttedius Brutus,194
Atticus, 'ATTUKOS,56,96-101,
Indexes 389
Bruttedius Niger,80,306 263,281,283-5,291,297,313,

322,337 nn.40,42,343 nn.24,
Buteo,224,236
25,32,344 n.38,355 n.8,373
n.2
characterization,151,176-8,
cacemphaton,215 194,213
cacozelia, cacozelos,82,215- chauvinism ascribed to the
24,243-4,249,302,342 nn.14, elder Seneca,23-6
16,361 nn.45,47,362 n.52,
children in rhetorical themes,
363 n.65,366 n.ll
174,185,195,285
Caesar, see Iulius
chria,31,116-17,350 n.35
calamistri,278,302
Xpwua and associated Greek
Calidius, M.,345 n.61,346 n.63 terminology,166-7,356 n.26
Caligula (C. Iulius Caesar Cicero, see Tullius
Germanicus),15
Cineas,92,228
Callimachus,375 n.24
circumlocution,217-18,249-51,
Callisthenes,79 293,367 n.21
Calpurnius Flaccus,113 circumstantial detail excluded
from rhetorical themes,111,
calumniam iuro,354 n.20
114,121,348 n.23
canticum,215-16,218,238
Civil Wars,5,8,77,165
causa,104,116,125,130-1,350 n.
civilis applied to language,
36,352 n.56; see also [hypo-
213
thesis] /uTiodeaus
class barriers broken by men
Capito,40,88,102,295
of letters,93
Cassius Dio,354 n.18,376 n.39
Claudius Caecus, Appius,318
Cassius Longinus, C,87,89-90
Claudius Marcellus Aeserninus,
Cassius Severus,7,15,5O,53,57, 7,165,188,286,343 nn.23,25
58,78-9,85-8,102,138-40,142,
clausulae,42-4,98-101,201-2,
146,180,190,201,207,210,216,
273-4,281,326-7,345 n.58,
232-4,237-8,278-83,307-8,
346 n.64,359 nn.13,17,370
311,321,367 n.30,374 n.3
nn.69-71,371 n.84
Catiline, see L. Sergius
Clodius Turrinus (the elder),
Catilina
94,142,296
Catius Crispus,22O
Clodius Turrinus (the younger),
Cato, see Porcius 94,296
Catullus, see Valerius codices used by declaimers,
228,289
Catulus, see Lutatius
Celsus, see Cornelius cogitatio in relation to
memory,234
Censorinus,142
cognitio (in classification of
Cestius Pius, L.,7,10,23,26, theses),104
32,45,53,54,85-9,110,146,
colloquialism,60,144,193-5,
154,158,164,170,174,179-82,
316,324
187,196,211,213,226,232,
Indexes 390
color in criticism of style, coniectura, coniecturalis,

166,197,200,356 n.23 155,158-9,251-2
color = complexion put on consolatio,310
facts of case,30,32-3; term
constitution see aiaats
derived from 'Hermagorean*
theory of defence,166-7,356 contests, literary, imagined
n.26; distinguished from by the elder Seneca,23-5
quaestio,159; distinguished
contradictio,209
from defensio,178; specimen
credited to Ciceronian de- contrarium: fault in the
clamation, 122; term used in treatment of themes,215,
Senecan criticism of ora- 223-4,226-7,253,284,360 n.
tory,141,321,354 n.20; 43,364 n.77; with reference
colores in third section of to enthymemata,203-6,360 n.
controversialsurveys,30,32- 26,363 n.68
3; constituting half of de-
controversial meaning,3,117,
clamatory inventio,152,263;
122,125-6,130; place in the
book by Iunius Otho about,
history of school rhetoric,
40; introduced at various
104-31,300,347 n.ll; treat-
points in declamation,168-9;
ment, 114, 145, 149-2 39, 260,
straightforward,169; subtle,
286,357 n.l; importance in
170-1; harsh,171-2; con-
the history of Roman litera-
cerned with irrational
ture, 143-8, 319-25
behaviour and supernatural
events,172-3,279; incredible, controversiae ethicae, see
174-7,224,356 nn.29,30; ethicos
damaged by irrelevant sen-
controversiae veteres9116-17,
tentiae,174; contrary to
121-2
theme,175; requiring support
of witness,175; linked with Cornelius Celsus,198
characterization,177-8;
Cornelius Dolabella, P.,123,
mixed,151,173-4
362 n.49
comicus,81,207,342 nn.12,14
Cornelius Hispanus,171
commentarii,40-1,233
Cornelius Nepos,64-6,93,130,
commentor,123,127-8 339 n.21,340 nn.22,23,25,
344 n.35,346 n.68
communis,215,223
Cornelius Scipio Aemilianus,
compositio,43-4,55,96-101,
P.,83
200-2,213,215-16,222,236',
246-50,267-8,273-5,279,281, Cornelius Severus,194,316,324
286,326-7,345 nn.57-59,359
Cornelius Tacitus,10,17,86,98,
nn.13,17,360 n.44,366 n.13,
134,138,141,143-5,191,221-2,
369 nn.59-62,370 nn.67,69-
259,263,278,282,286-7,302,
71,371 n.90
322-3,330 nn.25,34,331 n.47,
concinnus,214,290,296 352 n.3,353 n.9,365 n.7,
368 n.42
concitatus,96,101,190,200-14,
237,247,252-64,300,302 Cornificius, see Rhetorica
ad Herennium9360 n.26
conclusio: in rhetoric,201,
279,281; in dialectic,206 conversion 288-9
confessio,210,289,372 n.94 convicium saeculi,19-20,271-3,
293
Indexes 391
convoluted style,247-51 124,243-325,328 n.l; termi-

nology relating to,3,104-9,
copia,212,286-8
115-31; ancient views on the
Corduba,4,8,9,91,190,232,255- value of,26,121,129,143-8,
6,263,296,328 n.10,329 n.17 319-25; relationship with
higher forms of literature,
corruption, political,133-5
14,24,28,108-10,143-8,162,
corruptus and cognates,24-5,28, 181,214,220,258,260-70,
82,199,214-25,244,255,302, 291-2,299,304-25
342 n.16,361 n.47,363 n.67 decline of eloquence, theories
Cosconius, Q.,66 about,19-20,132-48,215,221,
352 nn.2,3,353 nn.8,9,11,14,
costume of declaimers,237-8
15
Cowley, Abraham,50,73,339 n.2
dedications of literary works,
Crassus, see Licinius 27-9,345 n.61
Craton,245 defence HOT' dvrlAnycKv dis-
tinguished from defence Mar'
Cremutius Cordus, A.,15,306-7 avrCêauv,166-7
Crispus,3O4 defensio within a declamation,
168,187-9
criticism: Tdescriptive1,59,
64-73; 'legislative1,59,67- deformitas,215
70; 'theoretical1,59,67-70
deliberativus, see rhetoric,
critics reported by the elder deliberative
Seneca,33,47-9,52-5,60,66,
80-3,91,101-2,146-7,167,181, Demades,223
187-200,207,218-20,226,228, Demetrius (author of Be elo-
235,281-2,284-6,307-8,311- cutione),100,360 n.44,361
19,358 n.l n.45,363 n.65
cu!tus,2Ol,214,29O,3O2,361 n. Demetrius of Phalerum,62,79,
45 96,108-9
Curiatius Maternus (in Tac. democracy,132-4,143
Dial.) ,143,353 n.9
Demosthenes,5,62,78-9,96-7,
cursus honorurn, see public 100-1,110,115,128,223,262-4,
life 304,345 nn.50,55,349 n.28;
see also ps.-Demosthenes
descriptio,181-3,210-12,214,
219,246-51,254-5,265-7,271,
Damas Scombrus,297
277,280,283-4,319,325,360
decens, decenter,227,264,278, n.33,369 n.58
364 n.81,374 n.17
desino,187,189,357 n.14,358
declamatio and cognates,94-5, n.15
104,117-20,123-4,127-30,352
dialogues, conventions of,
nn.51-53
41-2,335 n.24
declamation: its origins and
6taxpt8ri meaning 'school1,108
history,3-4,79,104-31,328
n.5,346 n.2; as a school dichoreus,99-100,346 n.64
exercise,3,5-7,107-8,117,
diction,55,60,85,190-200,246,
143-8,286; as practised by
248,267,275-8,280,284,289
adults,3,6-8,93-4,105-7,117,
Indexes 392
didactic literature, con- divisio antiqua distin-

ventions in,27-8 guished from that of novi
declamatores,304; divisio
digression,181-2,214,249,371
of law-court speeches,162,
n.90
164-5,286; divisio as an
Dinarchus,79,223 aid to memorization,365 n.9
Diodes of Carystus,24,148 Dolabella, see Cornelius
Diodorus Siculus,349 n.30 Donatus, see Aelius
Diogenes Laertius,61,105,110, Dorion,24,177,199,217-19,311,
341 n.ll 362 n.52
Diomedes (grammarian),361 n.45 doubles entendres,284
Dionysius (father and son),24, dream adduced as color,167,
72,296 172-3
Dionysius Atticus,8O,366 n.9 durus, dure,171-2,279,294,372
n.104
Dionysius of Halicarnassus,
55-6,299,301-2,339 n.23,359
n.13,371 n.12,373 n.121,374
n.3
echo/nxw,179,284,342 n.14
Diphilus,121
editions of the elder Seneca,
directness,161-2,169-70,208 377
disorganization,153,182,184-6, effeminacy,19,216,222,246,360
266-7,269,287 n.44
dispositio, second part of the effundo, effusus used pejor-
art of rhetoric,55,149,179- atively, 24 7 ,275
89,354 n.l
Egerton,51
disputatio: in schools of
electus applied to diction,
rhetoric,48,338 n.56; philo-
190-200,275
sophical,270,308,323,338 n.
56 elegance,72,140,162,180,361
n.45
dives, stock figure in rhe-
torical themes,107-8,110-13, elevation,214,290-1
170,174,177,348 nn.17,22,
elision,274
349 nn.24,25
ellipse (e'AXeucKs) ,215
divisiones: in second section
of declamation-surveys,30-2; elocutio, third part of the
constituting half of de- art of rhetoric,55,149,177,
clamatory inve/itio, 152,263; 190-227,263,354 n.l
their function,152-3; dif-
embezzlement in rhetorical
fering amounts of trouble
themes,111,348 n.22,349 n.
spent on them,32,182-3,233;
24
Senecafs technique of analy-
sing them,152-61,355 n.14; emotions, appeals to,101,140,
divisio not a separate sec- 174,176-7,186-9,237-8,271,
tion of declamation,152-3, 278,286-7,371 n.94
355 n.6; of suasoriae,153-4;
of controversiae,155-65, Empedocles,155
251-2; differing degrees of Empire, see Roman Empire
complexity,162-3,304;
Indexes 393
empty verbiage,190,200,212, 183-5,232,254,263,282,295,

214-15,217,244 310,319,325,357 n.8,368 n.
42
emulation of famous speeches,
291 exilitas and associated terms,
96,99,213,290
'enfeints terribles' among the
declaimers,23,176,298,333 explication 147,190,200,211-12,
n.75 215-16,218-20,246-51,257,
265-7,283-4,286,302
English literary criticism,
50-1,70,339 n.5 exsecutio,186
enthymema,69,202-7,233,257, extemporization, see impromptu
261,334 n.10,342 n.14,359 speaking
nn.22-5
exultans9215,222
enuxeiPnua/epichirema,203,
359 n.21
Fabianus, see Papirius

108-9
Fabius Maximus, Paullus (con-
epideictic, see rhetoric,
sul 11 B.C.),139,208
epideictic
Fabius Quintilianus, M.
epilogus,82,101,179-80,182,
(Quintilian): arrangement of
186-9,213,237,244,254,285,
his theory of oratory,55,
342 n.14,257 nn.12,14,358
149,354 n.l; his approach to
nn.15,16
criticism different from the
epiphonema,82,202,204,206-7, elder Seneca1s,54-5,59,67;
232-3,313,319,342 n.14,360 likewise his critical termi-
n.27 nology, 68; used Latinized
Greek critical terms,69; his
epistolographers as critics,
style in relation to the
59,97-8,275,346 n.62,370 n.
elder Senecafs,71; data
72
shared with the elder Seneca,
enuTa(puov,63,305,307,339 n.19, 42,83,315,353 n.7,371 n.80;
342 n.14,374 n.7 on learning Greek,255; on
schoolboys' notes,40; on
epitome of the elder Seneca
models for imitation,28,334
available to Suetonius?,60,
n.5; on progymnasmata,31,
339 n.13
116-17,129,350 n.35,352 n.
equity, see aequitas 55; on origin of Seats,105,
116-17,129,347 n.4,352 n.55;
ethicos,174,176,270,314,342
on origin of declamation,79,
n.14
108; use of the term uito-
eulogy,65,116,125,307,318,350 $eaus,351 n.36; on delibera-
n.35 tive prosopopoeia,109; on tri-
plices suasoriae,110; cited
Euripides,115 judicial theme known to
Excerpta,254,332 n.57,367 n.28, Cicero,120-2; his use of the
369 n.56 term controversia,121,126,
of the term scholastica,351
excerpting by the elder Seneca,
n.47; on divisio,365 n.9; on
31
Hermagorean theory of de-
excessus,186,252 fence, 166-7, 356 nn.25,26; on
works falsely ascribed to
exemplum,33,64-5,83,101-2,
Indexes 394
Hermagoras of Temnos,323; Festus,2O7

on techniques of argument,
fides = credibility,199-200,
205,359 n.21; on rerum
215-16,219,223
vitia,223-5,360 n.43; on
contradiction of themes, figura and associated terms,
356 n.33; on quibbling,225; 55,69,82,170-1,180,207-10,
on controversiae figuratae, 254,293-4,342 n.15,346 n.62,
170; on figurae,171,207-10, 360 n.28,362 n.56,371 n.94
278,288-9,360 n.28,371 n.94;
on transitions,180; on figura, divisio by means of,
avoidance of narratio,181; 163-4,289
on descriptiones,181-2; on figuratae controversiae,171
three-style theory,371 n.90;
on verba sordida,191-3; on figures of speech,208-9,271,
obscenity,196; on neo- 289
logisms, 198; on archaisms, figures of thought,208-10,271,
304,362 n.60; on peri- 289
phrasis, 250; on compositio,
200-1,249-50,360 n.44,369 figuro,171,288,371 n.94
nn.59,60; on sententiae, floridus,285,371 n.90
202-4,206-7,359 nn.23,24,
360 n.26; on cacozelon,215- Florus,6,201,286,362 n.56
18,220-1,361 n.45; on verba fluency,212-13,257
impropria,361 n.49; on
hyperbole,362 n.52; on forcefulness,72,96-8,186,297
metonymy,362 n.55; on con- Fortunatianus,83,332 n.61
clusio,371 n.83; on sing-
song delivery,238,244-98; fractus and associated terms,
on vocal exercises,239; 97,201,215-16,218-19,222,
cited Santra on Asianism, 359 n.13,360 n.40
64; cited Cato on oratory, free speech in Rome,16,19-20,
83,353 n.7; his references 26,84-5,133-5,138-43,293,
to Calvus,97,99-100; his 354 n.23
reference to Cicero's ob-
trectatores,221-2; remark on frigidity,222,297,361 n.45
Cicero's Verrines,365 n.7; Fulvius Sparsus,25,172,219,
judgement on Pollio,304; 226,294,392 n.104
judgement on Lucan,200; his
criticisms of schools of Furius Saturninus,139
rhetoric,144-6; his views on
furor and cognates,24,217,219-
relation between philosophy
20,361 n.47,363 n.67
and rhetoric,322-3,347 n.8;
view of relative merits of Fuscus, see Arellius
Greek and Roman oratory,333
n.72
Gades,9,256
fable (type of progymnasma),
Galba, see Sulpicius
117,350 n.35
Gallio, see Annaeus, Iunius
facultas,25,212,334 n.80
Gargonius,219,225,236,238,298
Fannius, C.,83
Gavius Sabinus,198
fatuus,224-5,227,363 n.69,364
n.81 Gavius Silo,142,296
Fenestella,66 genus dicendi,55-6,70,190,200,
Indexes 395
213-15,246,255,259,264-70, Helvia, wife of the elder

282,290,300,367 n.30 Seneca,9,21-2
Germanicus (Iulius Caesar),7, hendecasyllabi,96-7,342 n.14
10,317
Hermagoras (declaimer),47,180,
Germanicus (Iulius Caesar) the 196,279,289-91
younger,375 n.27
Hermagoras of Temnos,79,106-7,
Gesta Romanorum,324,376 n.38 118,125-6,158,166-7,323,341
n.6,347 n.8,355 n.17,356 nn.
gestures,235,237,271,295,373
26,27; see also ps.-Herm-
n.117
agoras
gladiatorial shows and the
Hermagorei,166-7
form of the elder Seneca1s
anthology,29-30 hermeneumata,342 n.14
Glaucippus Cappadox,297 Hermippus,62,339 n.17
Glycon Spyridion,25,219 Hermogenes of Tarsus,166,356
n.25,361 n.45,362 n.52
ta, Aristotelian
theory of,204-5 Herodotus,109-10,258,305,307
Gorgias of Leontini,107 Hesiod,27
Gracchus, see Sempronius hexis,342 n.14,371 n.93
gradatio9100 hiatus,274,326 n.32
grammatici,5,59,64-7,90,92-3, Hierocles of Alabanda,257
191,194,260-1,311,317,319,
Hirtius, A.,4-5,90,104,123,
342 n.14,350 n.35,367 n.33
129,287
grand style, failed attempts
Hispanus, see Cornelius
at,199-200,290,311,361 n.45
historia,81,101,116,184-5,
Grandaus,196,245
228,342 n.14
Greek loan-words in Latin
historians as literary
criticism,69,81-2,190-1,341
critics,59
n.11,342 nn.12-14,16
his toricus,91,342 n.14
Greeks contemporary with the
elder Seneca,23-6,78,82-3, historiography,14-17,21,26,
196,211,227,245,255,283-4, 63-4,101,109-10,136,184-5,
296-7,301-2,333 nn.75-9,334 228,254,257-8,263,279,304-7,
n.80,367 n.33 318,323-4
Homer,115,199,203,217-19,311,
317,362 n.52
habitus,69,101 Horatius Flaccus, Q.,248,256,

315,317-20,370 n.73
Haterius, Q.,37,51-2,57,85,87,
horridus,214,290
102,185-6,196-7,210,212,233,
238,286,288,297,304,309,315, Hortensius Hortalus, Q.,92,94,
343 nn.24,25 1O2,229-J1,287,295
health, concern for,73,236,239 humilitas,215
heckling,147,195,281 humour,73,88,140,144,193,222,
281,311,313
Hegesias of Magnesia,250-1,
257-9,298,366 n.19
Indexes 396
Hybreas,24,196,245,255,296-7 inessential detail excluded

from rhetorical themes,111,
Hyginus, see Iulius
114,121-2,260
hyperbaton,201-2,249-51
inflatus,200,215-18,222-3
hyperbole,218,362 n.52
inopia sermonis Latini ,212,
Hyperides,62,79,96,223 283,360 n.35
[hypothesis] /urcodeaus,106-7, inquit,188
117,122-6,130-1,347 nn.8,9,
insania and cognates,82,199,
350 n.36
217,219,221-2,224-5,277,305,
[hypotyposis] /u ,211 342 n.16,361 n.47,363 n.67
insanity: charge of,162-3;
plea of,172-3
Iavolenus Priscus,28 inventio, first part of the
art of rhetoric,55,149,151-
idiotismus/Ld LOOT uayos, 81,
78,224,263,354 n.l
193-5,278,341 n.11,342 n.14
involuntary homicide in rhet-
illustrations in Senecan
orical themes,112-13
criticism,52,60
Ion of Chios,61
imago = vivid evocation of
scene,253,367 n.24; in Ionian origin postulated for
mnemonic theory,229-32 tetralogiaet114-15
imitation,28,136-7,211,245-6, ironia,209,342 n.14
262,264-5,269,289-91,297,
Isaeus,62,79
305-6,311-16
isocol on /UCJOXOJAOV, 24 7 , 2 5 0 ,
implicatus,246-51
288-9
impromptu speaking,60,233-4,
Isocrates,62,79,94,136,320
280,286-7,365 n.8
iudicialis, see rhetoric,
improprius,215-16,361 n.49
judicial
inaequalitas,213,279,289
Iulius Bassus,191,193,21O,226,
inconsistencies (theme in 239,294-5,337 n.41,373 n.117
Senecan criticism),56-7,61,
Iulius Caesar, C.,66,102,
279,285,289-90
123-4,204,227,310,345 n.61
incorruptus,223
Iulius Hyginus, C.,64,66
incredibility,199-200,215,219,
Iulius Montanus,49,311-12,315,
223-6,284,361 n.45
338 n.59
indignatio,186,252
Iunius Brutus, M.,17,87,96,
indirect speech in the elder 102,222,345 n.61,362 n.63
Seneca's criticism,54,89,
Iunius Gallio,10,41,42,78,110,
101,344 n.37
141,148,157,162,167-8,176,
inelegant expressions in the 180,193,209,277-9,315,323,
elder Seneca's criticism, 337 nn.40,42,48,354 n.19,
71-2,340 n.34,341 nn.41,42 371 n.81
ineptus and cognates,222, Iunius Iuvenalis, D.,85,333
224-7,255,285,291,363 n.69, n.66,354 n.25
364 n.75
Indexes 397
Iunius Otho,40,139,167-70,173, 234-5,239,259,291,319-21,

226 353 n.15
ius distinguished from legal language,197,281
aequitas,155-61
Lesbocles,309
iusiurandum,209,291,321
lex - written law,156
lex dicendi,80,199-200,248
lex ingrati,157-8
Jonson, Ben,50-1,339 nn.3-5
lexicographical work on the
Juvenal, see Iunius
elder Seneca,32,42-6,68-71,
126,337 n.40,42,48,49,340
n.33
, see cacozelia liber suasoriarum, place in
Seneca's collection,34-7,
,215-16
335 n.19
361 n.44
licentia and cognates,25,198-
200,247,249,267,276,314,
358 n.9
Licinius Calvus, C.,79,93-102,
104,127,187,222,236-7,262,
Laberius, D.,280,311
264,301,312,345 nn.50-6,61,
Labienus, T.,7,16,57,58,139- 346 nn.62,64,66,362 n.63,
40,197,282,292-3,304,354 n. 365 n.4
23
Licinius Crassus, L.,94,278
laconic brevity,264,368 n.52
Licinius Crassus, P.,83
Lactantius,16-17,26
Licinius Lucullus, L.,184-5,
Laelius, C.,83 287,357 n.8
Latin usage, faults in,147, Licinius Nepos,25,219,224
190,194,256,284
literary genres, brief flower-
Latini rhetores, Republican, ing of,83-4,306-7,310-11,317
90,264,319
Livius, T.,45,58,63,133,197,
Latro, see Porcius 220,305-7,311,317-18
laudatio, see eulogy and Livius Andronicus, L.,317-18
rhetoric, epideictic
Livy, see T. Livius
law: Athenian,114-15,349 n.30;
locus = passage, commonplace,
Roman,162-3,281; unwritten,
116,153,162,181,183-6,210,
155; natural, concerning
251,254,263,265,277,314,319,
rise and fall of phenomena,
322,360 n.33; in mnemonic
132,135-8; see also ius
theory,229-32
laws connected with contro-
locus communis (elementary
versiae,30,114-15,157-8,
exercise),116-17,350 n.35
162-3,349 nn.30,32; of
Charondas,349 n.30; of Lollius, M.,317
Draco,114,349 n.30; of
Hammurabi,349 n.30 Lucan, see Annaeus
Lucian of Samosata,347 n.ll
law-courts,139-42,149,164,
Indexes 398
Lucilius, C.,315 Massilia,194,293-4

Lucilius Iunior, C,275,309 Maternus, see Curiatius
Lucretius Carus, T.,248,315, Mela, see Annaeus
360 n.35
Lucullus, see Licinius
membrum (in divisiones),156
Lutatius Catulus, Q.,94,223
Memmius, C.,66
luxuriosi in rhetorical
memoria, fourth part of the
themes,113,169-70
art of rhetoric,38-9,60,102,
luxury, as cause of decline, 149,185,228-35,284,354 n.l,
132-3,137-8,143,293,310, 364 nn.2,3,365 n.9
374 n.12
memory, the elder Senecafs
Lycurgus,79,223 supposed reliance on,37-42,
228,279,334 n.2
Lysias,5,79,96,98,lOO,lO5,223,
345 n.58 Menecles of Alabanda,257
Menedemus,61
Menestratus,277
Maccius Plautus, T.,87,121, Messala, see Valerius,
125,280,342 n.12,366 n.8 Vipstanus
macrologia,215-16 metaphors,215,250; as used in
literary criticism,71-2,341
Maecenas, C.,6,60,110,199,217-
n.40; in declamation,199-
19,243,249-50,266,278,298,
200,218,248,250,277,284; in
302,311-13,358 n.9,362 n.52,
poetry,199-200
366 n.17
metaphrasis/yeTacppaai, s,81,199-
Magius, L.,58 200,341 n.11,342 n.14
magnitude?, 199-200,219,311 yeTaaxaaus (type of defence),
maiestas, charges of,141 166-7
yauvoyotL,197,220 metonymy,218
mala adfectatio,215 Metrodorus Scepsius,92,231
male conlocatum,215 Milo, see Annius
male disposition,215 Miltiades (rhetorician),197,
220
male figuration,215
mimicus, mimus,225,281,311,
manuscripts of the elder
342 n.14
Seneca, see Annaeus
Seneca (the elder) misattributed: classical quo-
tations, 79, 305-6, 313, 374 n.
manuscripts of Suetonius, De
19; declamations,29,40,49,
gramm. et rhet.,350 n.36
295
Marcellus, see Claudius,
mixing of different types of
Marcius
diction,192,215,289-91
Marcius Marcellus,40,66
mixtus color,173-4
Marullus,5,53,77,198,213,
mnemonics,38-9,228-34,364 nn.
256-7,259,334 n.80,337 n.45
2,3
Indexes 399
mock trial,165 Nicetes of Smyrna,62

modernist tastes in litera- Nonius Asprenas, P.,335 n.17,
ture ,20-1,26,78,191-2,196-9, 337 n.45
276,282,293,304-25,374 n.3
Novatus, see Annaeus
mollis and associated terms,
novi declamatores,163,373 n.2
97,100-1,200-2,248,345 n.51,
373 n.117
monarchy,16-17,26
obscenity,24-5,156,169,190-1,
monotony (6uoeL6eua),215
193,195-6,284
Montaigne,50,338 n.l
obscurity,197,213,215-16,220,
Montanus, see Iulius, Votienus 266,271-2,305
moralizing commonplaces,72,184 openness,169,208,225,297
Moschus, see Volcacius operosus,246-51
muddy torrent,288,315,375 n.24 opus,367 n.30
mud-siinging,348 n.23; see oratio antiqua,197,200,282,
also color 292-3,304
murder in rhetorical themes, oratio nova,197,200,282,292,
111-12,348 n.22,349 nn.24,25 304
Murredius,24-5,176,188,195-6, oratory: Greek,77,82,115,350
207-9,217,219,224-5,298,361 n.33, see also Attic ora-
n.46 tors, Asian orators; Roman
(pre-Ciceronian),20,26,77,
Musa,10,19,52,57,78,185,199-
83-4,304,306,317-20,332 n.
200,211,217-18,224,277
61, (late Republican),84-7,
94-103,136,261-3,304, (con-
temporary with the elder
Seneca),18-19,80,136,138-42,
Narbo,321
146-8,278-83,291-2,296,315,
narratio,32,80-1,141,168-9, 319-21; declamation as a
173,179-82,186,214,238, training for,18,123-4,143-8,
251-2,271,278,284,321 214,263-4,319-21,325; as an
art distinct from declama-
narratio (type of progym-
tion ,18,69,93,104,164-5,
nasma),116
220-1,239,279,285,320-1
natural gifts, use and abuse
ornatus, faults concerning,
of (theme in Senecan criti-
215-16,361 n.45
cism) ,58-9
orthography, Suetonius*
neologism, 190,198-9,280
interest in,60
Neoterics,26,97,312
Orbilius Pupillus, L.,317-18
Nepos, see Cornelius
Ostia,9,116,121
Nero (eldest son of Germani-
Otho, see Iunius
cus),7
new style,197,200,243,245,282, Ovidius Naso, P.,20-1,36,73,
298,301 82,153,176,180,185,195,199-
200,232,248,264-70,291,313-
Nicetes (declaimer),6,24,78,81 15,317,320,324-5,329 n.13,
Indexes 400
332 n.63,334 n.80,358 n.4, Peripatetics,105,126,347 nn.

337 n.30,368 n.54,369 nn.55- 4,6
58,62,370 n.66,374 n.17
periphrasis, see circum-
locution
Persius Flaccus, A.,133
Pacatus,293
Petronius Arbiter,115,133,138,
Paeonius,119-20 144-6,195,244,298-303,373 n.
117
Pansa, see Vibius
phantasia,342 n.14
pantomimus,342 n.14
Philo (Academic philosopher),
Papirius Fabianus,ll,21,28,3O,
106
36,45,47,52,58,61,71-2,154,
176-7,200,210,213,220,238, Philodemus,264,341 n.11,348
264,270-6,289-90,297,308-9, n.16,350 n.33,368 n.52
320-1,323,333 n.64,337 n.40,
341 n.43,344 n.45,370 nn.64, philosophers as literary
70-2,372 n.102 critics,59,274-5,370 n.72
philosophumenos I'(puAoaocpouyevo s,
parental attitudes towards
82,107,130,184,342 n.14
rhetorical education,18,
145-6,319-25 philosophy: in relation to
rhetoric,78,104-7,115,118-
parricide,84-5,343 n.20
19,130-1,176-8,184,212,220,
pars (in divisiones),153,156 223,238,259,263,270-6,284,
299,308,320-3,368 n.42;
pars alters,32,140,151-2,321,
Roman attitudes towards,
367 n.28
21-2,26,78,118-19,136,259,
particula,156 308-10,320,374 n.10; see
also Annaeus Seneca (the
Passenus Paulus,28
elder)
Passienus,139,146,175,177,18O,
Philostratus,62,79,107-8,110-
283,353 n.16
14,130-1
Pater Patriae,84-5,343 n.20
phrasis,190-200,342 n.14,372
patriotism, ascribed to the n.96
elder Seneca,23-6
physicians in rhetorical
patrons, literary,33,135,199- themes,113
200,249-50,313
pietas,15,149,177,187
pauper9 stock figure in rhet-
Pindar,115
orical themes,107-8,110,170,
174,177 Piso, see Pupius
Peithagoras of Cyrene,62 plagiarism,29,232-3,245,284,
297,305-6
pen-portraits,29,50-3,63-4,
67-8,340 n.26 Plato,41-2,78-9,94,103,105,
110,115,136,307,310
Pergamum,294,366 n.9
Plautus, see Maccius
Pericles,112,130,374 n.7
pieonasmos,215-16
periergia,215-16
Plinius Caecilius Secundus,
periodic sentence structure,
C. (Pliny the Younger),28,
246-8,252,260,264,272,275,
41,345 n.50,346 n.62
285
Indexes 4O1
Plinius Secundus, C. (Pliny 90,200-2,204,206-8,210-11,

the Elder),9,41,92,134-5 213-4,226,228,232-9,245-6,
248,251-70,274,276-7,279,
Plotius Gallus, L.,90,92
286,297,304,312,316,322,
Plutarch,61,130,244,298,341 n. 324-7,329 nn.13,22,336 n.33,
11; see also ps.-Plutarch 337 nn.40,48,343 nn.22,23,
25,26,32,334 n.45,348 n.19,
poetry: Greek,115,136,261,299,
355 n.20,256 n.21,357 nn.ll,
311-12,317,373 n.117; Roman
12,358 n.15,359 nn.13,17,
Republican,26,66,136,260-1,
362 n.56,364 nn.75,77,365
263,280,304-5,310-12,317;
nn.8,3,367 n.28,29,369 n.54,
Augustan,6,136,199-200,232,
56,61,371 nn.80,94,372 n.
249,264-70,312-20,329 n.16,
104,373 n.2
370 n.73,374 nn.17,19,375
nn.24,28; post-Augustan,311- Porcius Licinus,66
12,315-17,375 nn.28,30;
Porphyrion,294,372 n.103
influencing declamation,33,
199-200,213,232,248-50, Porphyrius,166-7
260-1,272,294,311-14,368 n.
45; influenced by declama- Potamon,309
tion, 33, 232-3, 264-70, 27 3, praemium and associated terms,
319,324-5 132-5,138-9,142-3,353 nn.8,9
'pointed1 style, origins of, praeparatiOy praeparo,210,288,
257-64,296-303 372 n.94
Polemon,61 praesumptio,209-10,294,372
n.94
Pollio, see Asinius
praeteritio,253
Polybius,108-9,135,258,368
n.40 prefaces, function of the
elder Seneca's,27-30
Polycrates,110
prefatorial conventions,27-8,
polysyllabic words,293
38-9,334 n.l
Pompeius Magnus, Cn.,16,86,
pressus,213,222,363 n.64
88-9,101-2,123,287,295
Principate, see Roman Empire
Pompeius, Q.,120,127
principium,168,179-80,214,238
Pompeius Silo,45,164,168,172,
175,177,209,245,337 nn.40, probability, arguments from,
42,343 n.25,355 n.19 111-12
Pomponius, L.,280,311 proJblema,82,342 n.14
Pomponius Atticus, T.,65-6,93, progymnasmata,31,107,116-17,
130,352 n.57 130,350 n.35
Porcellus,194 , see praesumptio
Porcius Cato, M. (the censor), pronuntiatio,149,235-9
20,26,27,65-6,83,133,304,
prooemium,32,82,179-80,251,
318,332 n.61
271,278,342 n.14
Porcius Latro, M.,5-6,9-10,18,
20-1,24,30,32,40-1,43-6,49,
proposition 181,186,251-2
53,57-9,73,78-80,83,101, prose rhythm,43-4,95-101,
146-8,153,155-65,168-72, 200-2,247-50,267-8,273-4,
174-5,177-8,180,183-4,186- 285-6,326-7,336 nn,31-5,
Indexes 402
345 nn.57-9,346 nn.61,63-6, iecturalis,155,158-9,252,

359 nn.13,17,366 n.8,369 nn. 355 n.12; infinita,118,
59-62,370 nn.67,70,71 130-1,352 n.58, see also
thesis; (de-) finita,106,
prosopopoeia,10 9,15 2
125,130-1,352 n.58, see also
Protagoras,104,112,130 hypothesis
provinces: education in,5,77, quaestio in Hermagorean
255,328 n.13; oratory in, classification of rhetoric,
142,296,354 n.21 106-31
pseudepigrapha,29,40,49,323, quaestiuncula,156
351 n.43,356 n.26,368 n.46
quibbling, futile,224-5
ps.-Cicero, see Tullius
Quinctilius Varus (the
Cicero, M.
younger),10,54
ps.-Demosthenes,79,110,306,
Quintilian, see Fabius
348 n.18,351 n.40
ps.-Hermagoras,323,356 n.26 quotations from classics in
declamations,85-6,102,311,
ps.-Longinus,133-4,143,262, 313,343 nn.22-32
352 n.3,361 nn.44,45
ps.-Plutarch,62,328 n.12
ps.-Quintilian,111,151,260, rapid speaking,212,238,257,
344 n.34,348 n.23,349 n.24, 286,363 n.64
357 n.l
recitatio, red to, 6,281,324,
ps.-Sallust,351 n.43 329 n.16,348 n.19
ps.-Virgil,220 rectus,169,208
psychology: in declamations, redundance,215-17,222,282,363
101,186-9; the elder n.64
Senecafs interest in,56-61,
reliability of Seneca's decla-
67,73; see also character-
matory anthology,42-9
ization, ethicos
public life: the elder
religionis color,172-4
Senecafs view of,8-14,16-17, repetition 288-9
26; declamation as a train-
repetition: of similar rhythms,
ing for, see oratory
201; of similar clause open-
Publilius Syrus,207,221,224-5, ings, 288-9; of similar clause
281,311 endings,288
puerility,215,221,226,285,361 repetitiveness, general,213,
n.45,364 n.79 222,260,269,291-2
Pupius Piso Frugi, M.,120,127, replies to famous speeches,
193 110,348 nn.18,19,351 n.40
Republican government,16-17,
134-5,143
quaestio: in theory of div- rerum vitia,215,223-7,360 n.43
isio,153-65,168,179,182-3,
res iudicatae,162,164,355 nn.
279,286,297; iuris,155-7,
19,20,356 n.21
355 n.9; aequitatis,155-61,
321-2,355 nn.10,15; con- rewards of eloquence,132-5,
Indexes 403
138-42,353 nn.8,9 Sallustius Crispus, C.,26,63,

102,244-5,305-7,351 n.43,
Rhetorica ad Alexandrum,205-6,
353 n.11,374 n.8; see also
257,347 n.13
ps.-Sallust
rhetor,145-8,255,319,328 n.l,
sanitas, sanus,199,217,220-3,
342 n.14
363 n.64
rhetoric: judicial,106,108,
Santra,64-6
110-15,117,120,122-6,130-1,
139-42,144,347 nn.11,12, Eap6tay6s,215
348 nn.16,19,22,23,351 nn.
39,40,357 n.l, see also con- schema,82,170-1,180,202,207-
troversia; deliberative,106, 10,233,342 n.14; see also
108-10,117,125,140-1,347 nn. figura, figures of speech
11,13,348 nn.14,15,351 n.45, and of thought
see also suasoria; epideic- scholay342 n.14
tic,106,116,125,351 nn.43,45
scholastics,104,116,126,342
Rhetorica ad Herennium,128, n.14
206,229-31,288,335 n.10,348
n.15,351 n.39,352 n.51 scholasticus9l8,^7,103,162,
265,295,321,342 n.14
Rhodians,363 n.64
scholiasts,59,166
rigeo and cognates,22,345 n.50
schools of rhetoric: Hellen-
ring-composition,287 istic, 106, 108-10, 12 1,256-7,
role in declamations, choice 264,300; Roman,5-7,26,104,
of,151-2 110,143-8,263-4,317,319-25;
in Roman provinces,5,328 n.
Roman Empire: its massiveness, 13
135; political life,16-17,
Scipio, see Cornelius
26; Senate,140-1; law-
courts, 138-42,259; peaceful Sempronius Gracchus, C,83,278
conditions,16,143; moral
decline,134-5; see also Sempronius Gracchus, Ti.,83
book-burning, free speech Senate, oratory in,140-1,353
n.15
Roman literature in relation
to Greek,23-6,33,65,84,88, Seneca, see Annaeus
163,196,198,212,227,255-64,
Seneca fGrandiof,50,73,237-8,
283-4,286,290-1,296-303,333
n.72 365 n.5
sententia (type of pro-
Romanius Hispo,169,171,194,
gymnasma),31,116,350 n.35
198,337 nn.44,48
sententiae: in opening sec-
Rubellius Blandus,92-3,157,
tions of declamation-surveys,
209,297,321,335 n.17,344 n.
30-2,334 n.10,335 n.ll; fond-
45
ness of declaimers for,147,
325; practice in composition
of,202; three main types
(enthymemata, epiphonemata,
Sabinus, see Asilius sententiae proper) distin-
Sacerdos (grammarian),361 n.45 guished, 202-7; other var-
ieties,207; 'Publilianae1,
Saenianus,217,224,226 207,221,225; varying quality
St Jerome,9,64,330 n.30,367 n. of,214-15,243; detailed
34 criticism of,170,214-15,
Indexes 4O4
245-6,297; Greek and Roman Socrates,78,110,136

specimens compared,23-5,245;
soloecismus,147,194,342 n.14
used to counteract exilitas,
174,213; feature of 'middle1 solution carmen,264-70
style,371 n.90; damaging sophistic, first,107,130-1
colores by irrelevance,174;
used to effect transitions, sophistic, second,62,107,131
178; repetitiveness in their
Sophocles,61,115,155,261,368
use,291-2; specimens by
Latro,213,251-4,365 n.8, by n.45
Ovid,268-9,292,315, by sordidus, sordide,169,190-7,
Fabianus,271,275,37O n.72, 214,220-1,275,278,289,295,
by Gallio,278-9, by Albucius 358 n.5
imitating others,290-1, by
Votienus Montanus, 291-2, by Sotion,309
Spanish declaimers,296; Spain,4-5,8,23,77,89,190,236,
possible origins of taste 256,296,328 nn.10,11,13,
for them,257,259-64,300-3; 329 n.17,330 n.26
specimens from oratory,140,
321; from Senecan literary Sparsus, see Fulvius
criticism,71-2; declamatory specialization,78,102,138,282,
specimens fit for higher 311
forms of literature,148;
imitated by poets,312-13 spiritus,324
sententiosus,257-9 splendor and cognates,52,71,

192,246-51,271,288
Sepullius Bassus,86,343 n.22
aTaats,158,347 n.12,355 n.17
Sergius Catilina, L.,63,85-6,
Statorius Victor,209,227
143,261,295
status, see axaaus
Severus, see Cassius, Corne-
lius Stoicism,20-2,307-10,345 n.61
Sextilius Ena,91,19O,316 Strabo,366 n.19
Sextius, Q.,308-10,374 n.10 strictus,171
Sextus Empiricus,348 n.14 students of rhetoric, their
Shakespeare,51,339 n.5 attitudes and behaviour,5,
21,28,85,145-7
short words, successions of, studium,367 n.30
201,248,278,285,366 n.14,
371 n.91 stultus and cognates,19,215,
223-7,363 nn.68,69,364 n.81
Sicily,86,108-9,134,349 n.30
style,52,55-6,60,68-9,98,200-
Simonides,92 27,243-303,319-20,325,371
simplex,169,313 n.90
sing-song delivery,211,215-16, suasoria: definition,3; treat-
218,238,244,271 ment, 151-239, 2 60, 2 79, 284,
sitting and standing,237 286,347 n.ll; importance in
the history of Roman litera-
slaves in rhetorical themes, ture, 143-8, 319-25; place in
111-12,114,149,177-8,348 the history of school rhet-
n.22 oric, 104-10, 125, 266, 269-71,
300,347 nn.11,13
Smyrna,256,263
Indexes 405
suasorius,126 Terentius Afer, P.,66

suib oculos sujbiectxo,211,253 Terentius Varro, M.,64-6,
339 n.20
subtilitas,153-4,161,183,251
Terentius Varro, P. (Ata-
subtlety in use of colores,
cinus),266,312,315
169-70
Tertullian,138
success dependent on externals
(theme in Senecan criti- testamenti figura,209,254,278
cism),57,237
tetracolon /TCTPOIHWAOV , 208-9,
Suetonius Tranquillus, C.,4-6, 288-9
17,59-61,63-4,66-7,89-90,
tetradeum,214,278,288,295,342
92-3,113,115-17,121-6,243-4,
n.14
298,302,304,328 n.6,334 n.84,
349 n.26,350 nn.35,36,351 tetralogia,110-15,130,165,257
n.46,354 n.18,368 n.53
thema,30,32,111,145,174-5,342
Sulpicius Galba, Servius,83, n.14
128
themes of controversiae and
summisse,187,345 n.51 suasoriae,30,32,80,84,90,
107-16,119-22,125,130-1,
supernatural occurrences
144-5,151-61,170-5,178,260,
mentioned in declamations,
324,341 n.9,351 n.39,354
144,167,172-3
n.25
supervacuus,214-15,223
themes, rule against contra-
Surdinus,284,321 diction of, 174-5; rule
against assuming facts
surveys of declamations, the
extraneous to,174-5
elder Seneca1s,29-34,251,
334 n.7 Theodectes,92
suspicio and cognates,170, Theodorei,56,80-1,141,180
215-16,288,372 n.94
Theodorus of Gadara,56,8O-l,
sycophancy,16,135,141 141,321,341 nn.7,8,10
sylla2>a,221,342 n.14,369 n.59 Theon,125,347 n.4,350 n.35
syllogism,204-5 Theophrastus,79,105,347 n.4,
366 n.13
(type of defence),
166-7 thesis/%£OLS,104-6,115-19,123,
125,129-31,257,321,328 n.8,
syncrisis/avynpuaus,66,309
342 n.14,350 nn.35,36,351
n.38,352 n.55
Theuth, myth of,41
Tacitus, see Cornelius
Thrasymachus,228
three-style theory,55-6,99,
Tarraco,296 371 n.90
teaching methods criticized, Thucydidei,197-8,3O5
120-1,143-8,162,285
Thucydides,17,63,79,99,197,
tempero,172 245,305-7,345 n.58,374 n.7
Terence, see P. Terentius Afer Tiberius Caesar,10,17,60,81,
Indexes 406
138,140-1,243,302,308,312, 105,349 n.30; rhetorical

315,343 n.20 exercises practised by him,
4-5,104,116-24,328 nn.5,9;
Tibullus, see Albius
remarks on the % eats,105-6,
Timaeus,108-9,257-9,368 nn.39, 117-19,130-1,328 n.8,347 nn.
40 3-6,351 n.38; his use of the
term ui6deats,106,125,129-30,
tinnitus,278,303 347 n.9,351 nn.36,45,352 n.56;
Tiro, see Tullius his use of the term causa,4,
119,125,129-30,352 n.56; his
tirocinium fori,ll use of the term controversia,
title of Seneca's declamatory 125-6; evidence for his fam-
anthology,3,328 n.l iliarity with controversia
themes,121-3,131; his use of
tractatio aequitatis,155,157, declamo and cognates,96,120,
182-3,321-2,355 n.ll 124,127-30,344 n.48,352 nn.
tragicus,207,342 n.14 52,53; remarks on the rhet-
orical training usual in his
tragoedia,81-2,342 n.14 time,120-1,123,145,246,319-
transfer of training,18,319- 20,325,328 n.8,334 n.84;
25 his approach to criticism
different from the elder
transition,180,186 Seneca's,54-6,59,67,69,168;
translatio,215-16 similarly his critical
terminology,68-9,198,344 n.
translation,116,261,284,350 46; his imagery in critical
n.35 works,71,341 n.40; his pref-
travel between Spain and erence for pure Latin terms
Italy,9 in criticism,190-1; charac-
teristics of the Brutus,66,
Triarius,153,226-7,343 n.25 83,318,340 n.28,345 n.61;
tricolum/Tplxu)Aov,208,342 n. his anticipations of Silver
14,360 n.30,370 n.65 Latin style,261-3,368 nn.41,
45,373 n.117; his style in
tropes,199-200,217-18 relation to the elder
Tullius Cicero, M.: his early Seneca's,71; some of his
travels,77; his extensive expressions soon found old-
fashioned ,192-3,196-7,304;
education,77,225; his time
points of detail concerning
regarded as the classical
his diction,330 n.24,342 n.
period of Roman oratory,132,
12,353 n.8,362 n.49,366 n.
136,318-19; his oratory seen
16; his view of Demosthenes,
as challenging that of
96,262,345 n.55; of lesser
Greece,23,84; free speech in
Greeks,121,351 n.42; of
his time,19; techniques of
Asian rhetoric,238,244,257-9,
ridicule in his speeches,
300,363 n.64,368 n.41; of
192-3; lengthy speeches by Calvus,95-100,345 nn.53,59,
him,233,365 n.7; his view of 60,365 n.4; his obtrecta-
the rewards of eloquence, tores,221-2,362 n.63; delib-
133-5; the elder Seneca born erative and judicial themes
within his life-time,3-5; in De inventione,107; dis-
his view of the relationship tinction drawn by him be-
between rhetoric and philo- tween aequitas and ius,155,
sophy, 105-6, 184, 321-3, 352 355 n.13; his debt to Herm-
n.58; view of the early his- agoras of Temnos,79,167,355
tory of rhetorical education,
Indexes 407
n.17,356 n.25; on prooemia, Tullius Cicero, Q. (the

39; his view of the scope of elder),119-20
the Latin language,360 n.35;
Tullius Cicero, Q. (the
his use of the term color
younger),119-20
with reference to style,166,
356 nn.23,24; his theory of Tullius Tiro, M.,66
three styles,68-70,345 nn.
tumidus and cognates,199,215-
59,60; his view of archaism,
18,220,222-3,302,305,311,
197-8,280,305,318-19; his
361 nn.45,46
remarks on the enthymema,
203-4,206,359 nn.23,25; his turbidus 9212
view of obtrusive rhythm in
Tuscus,102,139,141,227
prose,366 n.13; clausulae
favoured and avoided by him, tutus,172
43-4; use of the term con-
tyrannicide,163
clusio,371 n.82; his use of
sanitas and cognates,221-3; tyrannus, stock figure in
his use of words later found rhetorical themes,108,110,
in theory of corrupta oratio, 115,163,348 n.17
222-3,298; of those used to
criticize rerum vitia,363 n.
68; his remarks on the term
ineptus9227; on memory,229- unification of declamatory
30,364 n.3; on actio,235, anthology,efforts towards,
237-9; his philosophical 19-20,29-30,34-7
works,276,318; his poetry,
261,311-12,368 n.45; his
opinion of Corduban poets,
91,190; rhetorical themes vagus,247
concerned with his life,37,
valens, valenter,162,210,214,
84-5,102,152,164-5,175,188,
211,225,287,324,351 n.43;
290,294,374 n.104
fulsome praise of him by Valerius Catullus, C.,96-7,
declaimers,65; contempt of 312,346 n.68
certain students of rhetoric
Valerius Flaccus,248
for him,146; the elder
Seneca1s collection of trib- Valerius Maximus,33,72,185
utes to him by historians
Valerius Messala Corvinus, M.
and poets,33,63,307,316-17;
49,78,110,139,146,174,190,
praise of him by Nepos,65;
256,283,313,316,338 n.59,
lost biography of him by
344 n.48,.348 n.19
Nepos,64; another by Tiro,
66; knowledge of his works Vallius Syriacus,8O,139
in the early Imperial
Varius Geminus,86-8,139,141,
schools of rhetoric,85-8,
337 n.41,343 nn.22,23,25
146,232-3,284,343 nn.22-32,
344 n.38; the elder Seneca's Varro, see Terentius
knowledge of his works,78,
vehemens,72,97,186
88-94,102-3,130,344 n.41,
352 n.57; works misattributed Velleius Paterculus,83-4,
to him,122,351 n.43,368 136-8,306-7,310-11,317,319,
n.46 344 n.34
Veranius Flaccus,243-4
Tullius Cicero, M. (the
younger),24,27,284 verbosity,212-13,215-16,222,
Indexes 408
282,286-8,297,299-300 47-9,66,139-40,180-1,184-5,
213,216,226,269,291-2,315,
verecundia,177,195-6
321,338 n.59
Vergilius Maro, P.,182,198-
200,203,206,218-19,248-9,
256,266,312-16,319,324,374
nn.17,19; see also ps.- white-washing of defendants,
Virgil 348 n.23; see also color
Verres, C.,85-6,128,143,192, wit,73,141,281,311,313-14,321
233,307
witnesses: not appealed to by
Vibius Gallus,195-6,210,219, Roman declaimers,114,175-6,
238,277 357 n.34; unreliable, in
rhetorical themes,111-12
Vibius Pansa Caetronianus, C.,
4-5,90,104,123,129,287 women: education of,21-2,223,
333 n.66; characters in
Vibius Rufus,24,192-4,197,286,
declamation,21,144,149,177,
297,335 n.17
186-7,195
vigor and cognates,97,197,262,
word-play,221
345 n.50
written preparation of de-
violentus, violenter,96,101,
clamations ,202,228,233-4,280
237,372 n.104
written sources available to
Vinicius, L.,139
the elder Seneca,39-42,47-8
Vinicius, P.,82,139-40,188,
written word, tradition of con-
224,232
tempt for,41-2
Vipsanius Agrippa, M.,6,16,
139-40
Vipstanus Messala (in Tac.
Xanthippus,112
Dial.),145,278,282-3
Xenophon,110
vir fortis, stock figure in
rhetorical themes,107-8, Xenophron of Sicily,62
110,163,171
Virgil, see Vergilius
virtus,72,193,215 young versus old in rhetorical
themes,114
vitium, vitiosus,19,72,96,99,
213,214-27,285,363 n.68
vituperation,116,350 n.35
Zoilus,171
viva vox,65
vivuzn consilium,86,279
voice,65,69-70,97,101,147,
235-9,271
Volcacius Moschus,80,141,221,
293-4,297,370 n.76
Volcacius Sedigitus,66
volubility,212-13,257,260
Votienus Montanus,25,40,42-3,
INDEX OF PASSAGES CITED
ABRONIUS SILO ARCHIMEDES 12.46 134,349 n.30

13.51 363 n.64
Morel, Frag. 232,312,316, De sphaera et 28 15.61 318
Poet.Lat.120 324 cylindro II 16.61 318
pr. init. 17.65 318
22.87 127-8
AELIUS ARISTIDES 31.119 347 n.6
ARISTOTLE 31.120 347 n.6
Or.HI 347 n.ll 35.132 223
Rhet. 55.202 222-3
66.233 363 n.67
ALBINOVANUS PEDO 1.2.8-9 184,204,206 82.283 96,98-9
1.2.12-14 205 82.284 96-100,221-3,
Morel, Frag. 317,324,375 1.4.7 109 345 n.61
Poet.Lat. nn.27,28 1.13.2 155 82.285 96,98
115-16 II.21.1-6, 205,359 n.23 83.287-8 197-8
III.3.3 367 n.21 84.289 96,98
87.298 356 n.24
89.304-92.317 77,352 n.56
ATHENAEUS 89.305 127
ap. Arist. 90.310 77,120,125,
Rhet.I.13.2 Deip.I.la-2a 41-2 127-8
95.325 257-9,300,368
ap. Arist. n.53
Rhet.III.3.3 95.327 259
BERL.PAP.P 9781 110,348 n.18,

351 n.40
.6.8 118 ,125,323
ANONYMUS SEGUERIANUS (Rh.Gr. 347 n.8
1.2.352-98 Spengel-Haramer) .8.10 355 n.17
.8.11 158
341 nn.7,8 CALLIMACHUS .11.15 356 n.25
.16.22 39
Hymn.Apoll. 375 n.24
ANTIGONUS OF CARYSTOS 180-1
ap. Diog. 61 .2.5 118

Laert.II.136 CALPURNIUS FLACCUS I .8.30 134
.34.157 93
ap. Diog. 61-2 Decl.XIII 113 : .56.240 155
Laert.IV.17 thema .59.251 238-9
1.3.9 106
1.4.17-18 227
ANTIPHON CALVUS 1.12.51-13.54 318
1.13.55-8 320
Tetralogia I 110-15,122, (ORF3 492-500) 1.14.61 323
165.348 nn.22, 1.16.65 125
23,349 nn.24, fr.22 99-100 11.24.99 123
25 fr.23 99-100 11.24.100 4,120,122-3,
fr.25 99-100 131,328 n.5
Tetralogia II 110,112-15, fr.26 99-100 11.44.187 366 n.16
165.349 n.32 fr.28 99-100 11.75.305 363 n.67
fr.32 97,99-101,187, 11.86.350- 229
Tetralogia III 110,113-15,165 237 88.360
fr.36 99-100 11.86.354 229-30
11.87.357 229
ANTIPHON (sophist) Morel, Frag.Poet.Lat.84-7 11.88.359 364 n.3
III.12.45 223
Ilepl xploews 167 fr.18 97,101-2,312 III.18.67 347 n.6
ovelpuiv (Diels/ III.23.86 127
Kranz Vorsokr.^ III.24.94 26,334 n.84
II 367-8) CATULLUS III.25.96 356 n.23
III.25.100 356 n.24
Carm.LIII 97,312 III.27.107 105-6,129
ANTISTHENES III.27.108 106
III.28.109 125-6,351 n.45
Aias 107 CHARISIUS (ed. K.Barwick) III.28.110 106,347 n.9,
351 n.45
Odysseus 107 102.20-2 99 III.48.184 366 n.13
296.21-4 99 III.52.199 68-9,356 n.23
III.59.222 237
APOLLONIUS RHODIUS III.61.228-9 94,103
CICERO
Arg.III.749-50 312 Opt.gen.
(a) Rhetorica
2 .6 262,368 n.46
AQUILA ROMANUS (RLM 22-37) Brut. 6 .17 334 n.7
35,4-8 99 8.33 371 n.82

9.36 223
Indexes 410
Or. 47 352 n.53 Eloc.

83 128
3.12 118 186 361 n.45,363
8.25 223 F osc.Amer. n.65
9.28 99-100 189 360 n.44
9.29 99-100 2 86 270 100
9.30 99-100,305 32 127-8
9.31-2 305 39 368 n.41
14.45 118,184,322 DEMOSTHENES
14.46 105,118,351 \'err.
n.45 Or.XVIII (De Ccr.)
17.56 237 [1.4.149 128
17.57 238 [1.5.86 192 179 100
19.64 223 [1.5.118 86,343 n.22 296 368 n.41
23.75 99 [1.5.161 86,343 n.22
23.76 99 Or.XXIII
23.77 99,345 n.59
23.78 99-100,345 (c) Letters 51 349 n.28
n.60 54 349 n.28
42.145 92 Ad Att.
50-51.169 371 n.82
55.184 342 n.12 [X.4.1 119,123 (DEMOSTHENES)
63.212 100 {IV.12.2 129
64.215 345 n.50 {IV.22.1 123 In ep.Phil.13 79,306
69.230 366 n.19
70.232 363 n.67 Ad Brut. (ed. W.S.Watt) DIO CASSIUS
71.236 363 n.67 8.2 87
XLVII.1.1 376 n.39
Part.Orat. Ad fam.
LVI.27.1 354 n.18
18.61-19.67 104,347 n.3 [1.11.1 87
28.100 155 [II.11.2 128
38.133 363 n.68 {.1.1 343 nn.29,32, DIODORUS SICULUS
344 n.38
Top.13.55 203-4,206,261, {.31 328 n.10 XII.11-19 349 n.30
359 n.23 {.32.3 256
<III.63.1 87
(b) Speeches (V.19.4
-i 87,89,343 n.31 DIOGENES LAERTIUS
{V.21.4 97-8,345 n.53
Arch.26 91,190 11.38 110
Ad Q.f. 11.136 61
Cael.46 133-4 IV. 17 61-2
[II.3.4 4,119,128-9, V.24 104
Cat. 328 n.8 V.44 105
[II.3.5 120 V.49 105
1.2 86,343 nn.23, VII.59 341 n.ll
26 IX. 53 105
IV.3 86,343 nn.23, d) Philosophica
25,27
Acad.I.19 347 n.6 DIOMEDES (GL I 295-529)
CIuent.161 125
1Oe fin. 451.7-10 361 n.45
Div.Caec.20.65 330 n.24
[V.2.5 347 n.6
Lig.10 204 if.2.5 128 DIONYSIUS OF HALICARNASSUS
Marc.25 86,343 n.29 9e off.II.45 353 n.8 De rhet.vet.
Mil. Atat.deor.I.4.8 360 n.35 1 299,373 n.121

2-3 301,373 n.121
9 206 ?en.82 88,344 n.34 374 n.3
34 353 n.8
92 343 n.24 Vusc.disp.
101 86,343 nn.24, DONATUS
28 [.4.7 119-20,125,
105 343 n.24 129-31 Vit.Verg.
[.24.59 92
Mur.44 128,352 n.52 [.31.75 127 15 319
[.35.86 351 n.42 29 312
Phil. [1.1.5 136,319
[1.3.9 106
1.38 86,343 n.29 [1.8.20-1 261,368 n.45
II.5 343 n.25 FORTUNATIANUS [RLM 81-134)
11.20 343 n.25
11.24 343 n.25 CICERO) 81.4-5 83,332 n.61
11.60 343 n.25
11.63 86,343 n.25 J rn Sallustium 351 n.43
11.64 86,343 n.25 invectiva
11.67 343 n.25 GORGIAS
11.77 343 n.25 CORNELIUS SEVERUS (Morel,
11.118 261-2 Frag.Poet.Lat .118-19) Helena 107
11.119 86,261-2,343
nn.25,27 "r.ll 194,324 Palamedes 107
V.19 123 i r.13 316-17,375 n.28
Indexes 411
HEGESIAS (FGrHist 142) NEPOS
F5 250 Chronici canones (ed. J.K. Atticus

F10 250 Fotheringham)
F12 250-1 16 130
F14 258 01ymp.CXCI 367 n.34 18 65-6
(p.249)
Olymp.CXCIV 9 Cato 3 65
HERMAGORAS OF TEMNOS (p.251)
(ed. D. Matthes) Cicero (GRF 64
De viris ill. 64 405-6)
fr. 6 125-6, 347 n praef.
fr.9 355 n. 17 De inl.viris . (HRR 11.35-40)
fr.10 355 n. 17
fr. 11 355 n. 17 fr.16* 93
fr.12 355 n. 17 fr.17 65,344 n.35
fr.13 355 n. 17 fr.18* 65-6,93
fr.14 166-7
6.434-56 333 n.66 Exempla (HRR 11.26-34)
7.150-77 354 n.25
HERMOGENES 10.125-6 85 fr.24* 65-6,340 n.23
346 n.68
Inv. (Rh.Gr.VI.93-212 Rabe)
202.4-8 361 n.45 LACTANTIUS

202.16-203.18 362 n.52
Inst.Div. 16-17,26,332
Stat. (Rh.Gr.VI.28-92 Rabe) VII.15.14 nn.55,56
38.16-39.19 356 n.25

1.2 .11-12 265
LIVY
1.8 .43 20-1 ,332 iu63
II. 11.10 199, 314
Praef.9-12 133 Ill .6.1 266
1.71 109,258 ap. Sen. 63-4 Ars Am. II.24 199, 314
III.80-2 110 Suas.6.21
Fast.III.449 248
Medea 267
HESIOD (LONGINUS)
Op. 10 Subl.
1.292 266 ,369 n.58
3.4 361 n.45 II.263-4 266 ,369 n.58
12.4 262 VIII.877-8 314
41.1 361 n.44 XI.474-572 369 n.58
Od.IX. 481 199-200,21 •4.1 143 XII.39-63 369 n.58
362 n.52 44.2 134 XII.607-8 232 ,313
44.3 134 XIII.121-2 265
44.6 133.143 XIII.503-5 292
HORACE
Pont.III. 369 n.58
Serin.1.4.11 315 1.19-20
Epist. Apokerytto- 347 n. 11 Priap.3.7-8 195 ,314,358

menos n.4
1.2.1-4 317
II.1.18-68 317-18 Phalaris II 347 n. 11
II. 1.69-75 317-18
Tyrannoktonos 347 n. 11 IV.10.25-6 268
Carm.1.22.20 248 IV. 10.29-36 269,320,329
n.13
LUCRETIUS
HYGINUS (HRR II.72-7)
1.832 360 n. 35
fr.l 64
1.936-8 315 PERSIUS
frs.2-3 64
IV.11-13 315
frs.5-13 64
V.493 248 Sat.1.32-5
frs.15-16 64
PETRONIUS
ION OF CHIOS (FGrHist 392) MAECENAS (ed. P. Lunderstedt)
Satyr.
F6 61 fr.10 249
F15 61 fr.ll 249,266 1 144-5,244,
fr.16 249-50 298-303
2 115,144-5,244,
ISOCRATES 298-303
MARTIAL 3 145,298-303
Archid.81 368 n.41 4 145-6,298-303
1.61.7 5 299
77 138
IULIUS SEVERIANUS (RLM 355-370) 133
366.5-7 99
Indexes 412
PHILODEMUS (PLUTARCH) V.14.1-4 203

VI.1.13 99
Poet.II.71 341 n.ll Vit.X orat. VII.1.14 351 n.47
VII.2.54 145
Rhet. 836a 5,328 n.12 VII.3.20 145
848c 5,328 n.13 VII.4.7-12 356 n.25
1.134 348 n.16 VII.6.6 121
11.97 350 n.33 VIII.2.2 191,361 n.49
11.216-17 368 n.52 POLLIO VIII.2.3-4 362 n.49
11.218-19 264,368 n.52 VIII.3.21 193
(Cic.) ad Fam. 328 n.10 VIII.3.22-3 192
X.31 VIII.3.25 198
PHILOSTRATUS (Cic.) ad Fam. 256 VIII.3.28 362 n.60
X.32 VIII.3.31-7 198
V it.Soph. VIII.3.38 193
ORF 516-26, 91-2,285,307 VIII.3.39 196
1.481 79,107-15, fr.19 VIII.3.44-55 215-16
130-1 VIII.3.56-8 215-27
1.511 62 VIII.3.59-60 215-16
POLYBIUS VIII.5.1-25 31,202-7,
359 n.23
PLATO VI.51.4 135 VIII.5.3-6 202-3
XII.25 a 108 VIII.5.9-10 203-6,359
Apologia 110 XII.25k 108-9 nn.21-5
Socratis XII.26.9 109 VIII.5.11-14 206-7
XII.26a 109,258-9 VIII.5.15-25 207
Phaedrus VIII.6.23-7 362 n.55
VIII.6.67-76 362 n.52
227c 105 PORPHYRION IX.1.1-3.102 207
23Oa-234b 105 IX.1.4 208
274c-275b 41 Schol.Hor. 372 n.103 IX.1.14 170-1
278e-279b 94 Epist.I.5.9 IX.1.17 208
IX.1.19 208
IX.1.21 278
PLAUTUS IX.2.16-18 209-10,294,
372 n.94
AuJ.261 125 QUINTILIAN IX.2.25 99
IX.2.34-5 209
Bacch.472 366 n.8 Inst.Or. IX.2.40-1 211
IX.2.44 209
Merc.630 280 1.1.12-14 255 IX.2.51 289
1.9.2 350 n.35 IX.2.62-3 209
Rud. 1.9.3 31,350 n.35 IX.2.65-80 170-1
1.9.6 350 n.35 IX.2.91 42,371 n.80
31-2 121 1.10.31 360 n.44 IX.2.95 209
737 366 n.8 II. 1.8 350 n.35 IX.2.106 209
975-8 121 II.1.9 116-17,126,129, IX.3.55 100
350 n.35,352 IX.3.56 99
Trin.319 87 n.55 IX.3.80 288
II.1.11 350 n.35 IX.3.87 209
II.2.9-13 145 IX.3.101 208
PLINY THE ELDER II.4.15-17 350 n.35 IX.4.22 200-1,371 n.83
II.4.18-19 350 n.35 IX.4.28 201,249-50
N.H. II.4.20-1 350 n.35 IX.4.55-6 201
II.4.22-3 350 n.35 IX.4.61 369 n.59
Praefatio 17 41 II.4.24-5 350 n.35 IX.4.72-8 369 n.60
VII.88 92 II.4.41 79,108-15,300 IX.4.106 202
XIV.1.4 135 II.10.4 122 X.I.73 305
XIV.1.5 134-5 II.10.5 144 X.1.90 200
XIV.1.6 134-5 II.10.9 144 X.I.101-4 307
XIX.1.4 9,330 n.27 II.11.7 40-1 X.I.105 333 n.72
III pr.4 315 X.I.113 286,304
PLINY THE YOUNGER III.3.1 55,149,235, X.I.115 97,100
354 n.l XI.2.1-51 229
Ep. III.3.4 228,235 XI.2.6-7 39
III.3.10 354 n.l XI.2.10 234
1.2.2 345 n.50,346 III.5.5 347 n.8 XI.2.11-16 92,93
n.62 III.5.7-9 125,347 n.8, XI.2.17 229
III.5.17 41 351 n.36 XI.2.18 229
VI.15.1-3 28 III.5.12-15 322-3,347 n.8, XI.2.19-22 229-31
356 n.26 XI.2.23 92,230-1
III.8.22 154 XI.2.24-6 230-1,364
PLUTARCH III.8.33 110 n.3,365 n.7
III.8.52 109 XI.2.30 230
Ant.2.5 244,298 III.11.3 81 XI.2.31 230-1
IV.1.77-8 180 XI.2.36-7 365 n.9
Dem.8.2 341 n.ll IV.2.4 181 XI.3.1 235
IV.2.28 356 n.33 XI.3.19-24 239
Pericles IV.2.30 351 n.47 XI.3.58 238,244,298
IV.2.66 181 XII.1.1 83,353 n.7
5.3 61 IV.3.1-2 181-2,214 XII.2.25 105,347 nn.4,6
36.3 112-13,150 V.10.1 69,206 XII.6.5 145
V.10.2 206,359 nn.21, XII.10.12 221-2
Pomp.42.5 347 n.8 24 XII.10.16 64,250
V.12.8 225 XII.10.58-65 371 n.90
V.13.16 360 n.43 XII.11.14-15 146
Indexes 413
(QUINTILIAN) SENECA THE ELDER I .1.2 204,336 n.33

I .1.3 204,336 n.33
Decl. Mai. (ed.G. Lehnert) Historiae I .1.5 337 n.45
I .1.6 286,336 n.33
II thema 349 n.24 Fr.l MUller 16-17,26,332 I .1.13 45,155,157,
XVII thema 260 nn.54-6 160-1,168,304,
XVII.3 260,368 nn.43, Fr.2 MUller 15,17,332 nn. 337 n.40,373
44 54,55 n.2
I .1.14 155,157,163,
Decl. min. (ed.C. Ritter) Oratorum et rhet orum senten- 337 nn.40,42,
355 n.10,373
CCLX (61.12- 151
16) Title 328 n.l 1 .1.15 82,297,342 n.15
CCXLVIII thema 112 T .1.16 169,173
CCLIII (37.11) 344 n.34 Controversiae ] .1.21 179,236,295
CCLXXII thema 349 n.24 1 .1.22 7,58,72,137,284
CCLXXXI thema 349 n.24 I pr.l 18,27,35,302, 1 .1.24 168,172,179,208
CCCV thema 112 336 n.33 1 .1.25 179-80,278-9,
CCCXIX thema 349 n.24 I pr.2 5,37-42,228-32, 333 nn.76-8,
CCCLI sermo 348 n.23 311,336 n.33 342 n.14
CCCLII sermo 348 n.23 I pr.3 37-42,228-232, 1 .2 thema 151,159-60,195
CCCLXIV thema 349 n. 5f 329 n.13,336 1 .2.1 246,336 n.33
n.33 1 .2.5 246,336 n.33
I pr.4 35,37-42,228- ] .2.7 32
32,336 n.33 1.2.13 156,161,168,
I pr.5 37-42,228-32, 355 n.15
RHETORICA AD ALEXANDRUM 336 n.33 ] .2.14 80,155-7,159,
I pr.6 16,23,26,28,51, 161,342 n.14,
1422a 347 n.13 65,84,88,137, 355 n.ll
1430a 205-6,257 336 n.33 1 .2.15 45,163
I pr.7 16,26,45,84, 1.2.16 82,164,199,
132-48,302, 289,342 n.15
RHETORICA AD HERENNIUM 310,336 n.33, 1 .2.19 81,180
353 nn.5,8 ] .2.21 169,177,193,
III.2.2 348 n.15 I pr.8 19-20,26,132, 195,213,295
III.12.20 128 137-8,310,336 ] .2.22 24-5,195-6,
III.16.28- 229 n.33 314,333 n.78
24.40 I pr.9 19-20,26,27, 1 .2.23 25,194-6,245,
III.16.29- 229,231 83,132-3,138, 333 nn.76,73
19.32 304,310,336 ] .3.1 336 n.33
III.20.33 230-1 n.33 ] .3.3 336 n.33
III.21.34 231,364 n.4 I pr.10 19,26,29,132, 1 .3.8 82,156,342 n.14
IV.13.19 288 138,232,310, ] .3.9 174,284
IV.17.24 335 n.10 336 n.33 1 .3.10 10,53-4,284,
IV.18.25 206,360 n.26 I pr.ll 3-5,29,39,65, 330 n.32
77,84,88-90, ] .3.11 7,40,176,226,
328 nn.2,11, 284,295,363
329 n.23,336 n.69
n.33,337 n.48, ] .4.1 336 n.33
SACERDOS (GL VI 427-546) 369 n.54 ] .4.5 336 q.33
I pr.12 3,4,89-91,93-5, ][.4.6 155,157,163,
455.12-13 361 n.45 104-31,336 n. 355 n.ll
33,342 n.14, 1[.4.7 122,226,328
346 n.2,350 n. n.9,333 nn.76,
SALLUST 34,351 n.36, 77,363 n.69
352 n.55 [.4.8 168,181-2,211,
Cat.5 63-4 I pr.13 5,16,18,57,336 336 n.33
n.33 [.4.9 182
lug.2 353 n.ll I pr.14 57,233,336 n. [.4.10 24,297,333
33,365 n.3 n.76,78
I pr.15 45,57,73,336 [.4.11 24,72,297,333
(SALLUST) n.33 nn.77,78
I pr.16 45,58,169, [.4.12 24,333 nn.77,
In Ciceronem 351 n.43 235-6,336 n.33 78,363 n.69
invectiva I pr.17 45,58-9,73, t.5 thema 195
228,234 [.5.1 334 n.9,336
I pr.18 87,101,184, n.33
SANTRA (GRF 384-9) 228,234,254, [.5.2 336 n.33
365 n.6 [.5.3 284,337 n.40
fr.13 64-5 I pr.19 92,228-32 [.5.4 183
I pr.2O 32,251 [.5.8 45,155,158-9,
I pr.21 32,152-3,183, 337 n.40,355
SCHOLIA IN HERMOG.STAT. 237,7.51 n.12
I pr.22 5,21,27,32,53, [.5.9 34,158-9,168,
Rh.Gr.IV.397. 166-7,356 n.26 77,183,202,213, 289,335 n.17
8-30 Walz 251,257,263, [.6.1 336 n.33
(Porphyrius) 328 n.13,365 [.6.5 337 n.41
n.8 [.6.7 336 n.33
Rh.Gr.VII.308 166-7,356 n.26 I pr.23 82,202-10,233, [.6.9 179,220
adn. Ualz 342 n.14,359 [.6.10 226,342 n.14
nn.21-5 [.6.11 226,284,363
I pr.24 5,30,82,171-2, n.69,364 n.79
208,342 n.14 [.6.12 333 n.76
I.I thema 159,337 n.40 [.7 thema 172,349 n.31
1.1.1 202,204,336 [.7.1 336 n.33
n.33
Indexes 414
Controversiae (cont.) II.1.13 210,272-5,335 II.4.9 82,164,342

n.14,356 n.21, n.15
1.7.2 336 n.33 370 nn.70,73 II.4.10 170,338 n.59
1.7.5 336 n.33 II.1.20 355 n.19 II.4.11 139
1.7.9 295 II.1.23 82,164,342 II.4.12 6,208,342 n.14
1.7.10 334 n.9,336 n.15 360 n.30,364
n.33 II.1.24 20-1,73,82, n.77
1.7.12 115,155,163, 174,177,213, II.4.13 6,16,139-40,
333 n.76,355 331 n.46,342 184,308,354 n.
n.9 n.14 23,364 n.77
1.7.13 151,177,209, II.1.25 177,219,238, II.5 thema 349 n.31
342 n.14 277,298,361 II.5.4 336 n.33
1.7.14 172-3,311,328 n.47,370 n.64 II.5.6 210
n.9 II.1.26 277,361 n.47 II.5.7 333 n.64
1.7.15 172,342 n.14 II.1.27 173 II.5.10 158
1.7.16 172-3,236,337 II.1.30 174,356 n.32 II.5.11 80,158,330
n.40 II.1.33 167-8,170,173, n.33,366 n.9
1.7.17 151,173,181, 279,330 n.33, II.5.12 158
184,186,209, 356 nn.29,30 II.5.13 80,141,158,
342 n.14,360 II.1.34 139,141,170, 294,330 n.33
n.33 320-1,354 n.20 II.5.14 155,157,355
1.7.18 192,236,289, II.1.35 139,141,320-1, n.10
333 n.77,342 354 n.20 II.5.15 156
n.14 II.1.36 80,139,141, II.5.16 157
1.8.1 336 n.33 320-1,354 n.20 II.5.19 139
1.8.2 336 n.33 II.1.37 170 II.5.20 139
1.8.7 115,155,163, II.1.39 169,333 n.77 II.6 thema 113,170
333 n.76 II.2 thema 159,269,314 II.6.4 337 n.40
1.8.8 171,337 n.40 II.2.1 270,336 n.33 II.6.5 45,337 n.40,
1.8.11 177,333 n.76 II.2.4 271 355 n.20
1.8.15 173,311 II.2.5 155,157-9,355 II.6.6 82,169-70,
1.8.16 24,79,115,148, nn.9,11,12 342 n.15
177,184,213, II.2.7 198,334 n.80 II.6.8 284,361 n.47
333 n.77,350 II.2.8 33,200,255, II.6.10 354 n.3
n.33 264-70,313-14, II.6.12 194
324-5,329 n.13, II.6.13 333 nn.76,77
II pr.l 36,52,57,63, 342 n.14,367 II.7 thema 251,270
179,183,199- n.30,368 n.54, II.7.1 31,202,251-2,
201,213-14,238, 369 nn.55-8, 260,270,357
246-51,267, 61,370 n.64 n.ll
270-5,308,309, II.2.9 153,184-5,264- II.7.2 31,202,251-2,
334 n.80,336 70,313-14,329 260,270,35)
n.33,367 n.30, n.13 n.ll
370 n.72 II.2.10 267-9,313-14, II.7.3 31,202,251-5,
II pr.2 20,58,61,71-2, 369 n.62 260,268,270,
176,197,213, II.2.11 267-9,313-14 274,357 n.ll
220,238,270-5, II.2.12 36,82,153,176, II.7.4 21,31,202,251,
308-10,336 179,185,199- 253-5,260,270,
n.33,341 n.43, 200,266-7,270, 357 n.11,367
352 n.6,370 313-14,334 n. n.25,369 n.56
n.72 80,342 n.14,374 II.7.5 31,202,251,
II pr.3 8,11-14,18,21, n.17,375 n.22 253-5,260,270,
26,148,270-5, II.3 thema 162,371 n.80 357 n.ll,369
308,310,320-5, II.3.1 201,336 n.33 n.56,370 n.67
330 n.36,331 II.3.3 336 n.33 II.7.6 31,202,251,
nn.45-7,336 II.3.4 294 253-5,260,270,
n.33,341 n.43, II.3.6 42,278,371 357 n.ll,367
367 n.30,370 n.80 n.29
n.72 II.3.9 336 n.33 II.7.7 31,202,251,
II pr.4 8-9,11-14,16, II.3.10 337 n.45 253-5,260,270,
21,27,53,141, II.3.12 157,162,164 357 n.ll
308,320-3,329 II.3.13 126,162,164, II.7.8 31,45,202,251,
nn.18-23,330 286,342 n.14, 253-5,260,270,
n.24,331 nn.37- 363 n.69,364 357 n.ll
47,336 n.33,367 n.75 II.7.9 31,202,251,
n.30 II.3.14 162-3 253-5,260,270,
II pr.5 30,36,89-90, II.3.15 156 357 n.ll
92-3,115,255, II.3.16 156 II.7 Excerpta 254,367 n.28
270-5,297,308, II.3.17 169
336 n.33,344 II.3.18 82,342 n.15 III pr.l 7,51,57,94,
n.45 II.3.19 342 n.14 279,336 n.33
II.1 thema 170 II.3.21 81,194,341 III pr.2 51,139,278,
II.1.1 336 n.33 n.11,342 n.14 282,330 n.33,
II.1.4 268,336 n.33 II.3.22 53,82,284,297, 336 n.33
II.1.5 336 n.33 342 n.14 III pr.3 15,58,139,235,
II.1.6 336 n.33 II.3.23 7,82,333 n.76, 237-8,283,336
II.1.7 336 n.33 78,342 n.14 n.33
II.1.8 336 n.33 II.4 thema 349 n.24 III pr.4 45,51,57,139,
II.1.10 210,272-5,335 II.4.4 336 n.33 281,336 n.33
n.14,370 nn. II.4.5 336 n.33 III pr.5 139-40,336
71,73 II.4.8 6,34-6,82,110, n.33
II.1.11 45,210,272-5, 174,190,256, III pr.6 50-1,139,233-4
335 n.14,370 301,338 n.59, 280,336 n.33,
nn. 70,71,73 342 n.14,348 371 n.85
II.1.12 210,272-5,335 n.19
n.14,370 n.73
Indexes 415
III pr. 7 7,59,139,190- V.2 extra 363 n.69 VII.4.1 45

214,220,243,251, V.3 thema 112-13 VII.4.3 155,224,225,
262,267,279,336 V.8 thema 349 n.31 311,342 n.14,
n.33,342 n.14, 350 n.34,355
367 n.30 VI.8 extra 139,141 n.9,357 n.12,
III pr.8 18,53,78-9, 363 n.69,369
102,280-3,307, VII pr.l 6,7,57,212, n.56
311-12,320, 236-7,288,322, VII.4.4 155,157,311,
367 n.30 329 n.15 355 nn.9,10
III pr.9 18,53,280-3, 54,60,186,212, VII.4.5 186-8,342 n.14,
VII pr.2 233,288-9,342 357 n.12,369
320
III pr.10 18,53,138,180, n.14,371 n.93, n.56
280-3,320,342 372 n.96 95-102,177,
n.14 57,191-2,199, 186-8,201,213,
III pr.ll VII pr.3 210,212,278, 237,262,264,
18,53,280-3,
320 288-9,358 n.5, 301,342 n.14,
III pr.12 18,53,126,142, 360 n.35,370 365 n.4
280-3,320-1, n.64,371 n.94, 95-102,236-7,
342 n.14 372 n.96 262,264,301,
III pr.13 18,53,146, VII pr.4 47,60,192,289- 312,342 n.14
230-3,320-1, 91 79,95-102,187,
342 n.14 VII pr.5 47,60,81,193, 207,237,262,
III pr.14 18,53,139,146, 213-14,289-91, 264,301,304,
280-4,320-1, 296,330 n.33, 342 n.14
338 n.59,353 341 n.11,342 VII.4.9 304,342 n.14
n.16 n.14,366 n.3, VII.4.10 333 n.76,78,
III pr.15 18,53,85,146, 367 n.30 361 n.47,365
216,232,280-4, VII pr.6 82,141-2,147, n.2
320-1,342 n.14, 193,209,278, VII.5 thi 111-12
343 n.24,361 291.342 n.6 VII.5.6 102
n.47 VII pr.7 73,82,139, VII.5.7 187-9,342 n.14,
III pr.16 18,53,110, 141-2,209-10, 357 n.14,358
280-3,320-1, 291,321,342 nn.15,16
342 n.14 n.14 VII.5.8 187-9,224
III pr.17 18,53,280-3, VII pr.8 82,141-2,284, VII.5.9 194,224
320-1,333 n.71 291,322,342 VII.5.10 210-11,224,
III pr.18 18,53,201,279- n.14,351 n.47 226,363 n.69,
83,320-1,334 VII pr.9 284,334 n.8, 364 n.80
n.8,336 n.33, 342 n.14 19,139,188,
371 nn.82,83 VII.1.1 288 224,291,338
III. 7 extr, 269,314,361 VII.1.2 288-9,337 43 n.59,363 n.69
n.47 VII.1.3 288 47-9,140,185,
III.9 them 151 VII.1.4 210,285 291,295,321,
VII.1.15 184 338 n.59,342
IV pr.l 30 VII.1.18 41 n.14
IV pr.2 6,7,37,57.66, VII.1.20 168-70,177 VII.5.13 184-5,220,363
285,324,329 VII.1.21 169,177 n.69
n.16 VII.1.24 171 VII.5.14 195-6,333 nn.
IV pr.3 7,9-10,37, VII.1.25 333 n.76,78, 77,78
51-2,57,66, 361 n.47 VII.5.15 188,225,333
285-6,304,329 VII.1.26 182,210,333 n. n.76,342 n.14,
n.17,371 n.90 76,78,342 n.14 363 n.69
IV pr.4 37,58,66,139, VII.1.27 33,182,211, VII.6 thema 178
165,309 213,266,283-4, VII.6.3 294
IV pr.5 16,37,66,309 311-13,315-16, VII.6.9 294
IV pr.6 37,66,287,309 338 n.59 VII.6.12 286
IV pr.7 37,52,66,185-6, VII.2 thema 84 VII.6.13 45
212,233,238, VII.2.1 86,343 n.22 VII.6.17 168,178
286,297,333 VII.2.3 86 VII.6.18 178,188
n.77,334 n.80, VII.2.5 295.343 n.25 VII.6.21 198
339 n.5 VII.2.6 88,295 VII.6.22 139
IV pr.8 37,52,66,179, VII.2.7 295 VII.7 thema 158
186,213,286-7, VII.2.8 45,90-1,164-5, VII.7.10 158,162,183,
329 n.23,339 324 202,210,251,
n.5,342 n.14 VII.2.9 236 266,322
IV pr.9 37,52,57,66, VII.2.10 86,169,188, 169,226,363
85,92,191,197, 343 nn.23,25, n.69
286-7,304,339 27 VII.7.15 173
VII.2.12 175,357 n.34 VII.7.19 179,284,311,
IV pr.10 37,52,66,286, VII.2.13 151,188,337 342 n.14
339 n.5 n.44 VII.8.1 45
IV pr.ll 37,52,66,72, VII.2.14 207,225,363 n. VII.8.8 155,157,355
212,286-8,315, 69 n.10
334 n.8,339 VII.3 thema 158,260
n.5 VII.3.4 IX pr.l 18,47-9,139,
IV.1 thema 113 VII.3.6 158 146-7,153,182,
IV.3 thema 112 VII.3.7 297 185,291,320-1,
IV.3 extra 168-9 VII.3.8 81,88,176,207, 338 n.59,372
IV.6 thema 151 221,225,280, n.98
IV.6 extra 169,173-4 311,342 nn.12, IX pr.2 18,47-9,146-7,
IV.7 thema 349 n.31 14,361 n.47, 165,237,291,
IV.8 thema 37 363 n.69 320-1,363
IV.8 extra 169 VII.3.9 88,280,311, n.69,372 n.98
342 n.14 IX pr.3 18,47-9,146-8,
VII.3.10 281,338 n.56, 291,320-1,342
363 n.69,371 n.14,372 n.98
n.84
Indexes 416
Controversiae (cont.) 269,291-2,315, X.4.19 25,333 n.78

333 n.78,338 X.4.20 25,255,297,333
IX pr.4 18,47-9,146-7, n.52 n.78,335 n.17
291,320-1,342 IX.5.17 213,216,269, X.4.21 25,83,101,255,
n.14,372 n.98 291-2,315,333 297,301,333
IX pr.5 18,47-9,146-7, nn.76,78,361 n.78,342 n.13
151,235,291, n.47,367 n.30, X.4.22 25,219,333 nn.
320-1,342 n.14, 372 n.100 76,78,79,342
372 n.98 IX.6.2 277 n.13,361 n.47
IX.1 thema 349 n.29 IX.6.3 338 n.52 X.4.23 25,226,333 nn.
IX.1.3 338 n.52 IX.6.10 139,156,291, 76,78,334 n.80,
IX.1.9 155,157,355 363 n.69 361 n.47,364
n.ll IX.6.11 224,284-5,361 n.77
IX.1.10 279,338 n.52 n.47,363 n.69 X.4.25 82,232,313,335
IX.1.12 33,245,255,333 IX.6.12 284-5,342 n.14, n.17,342 n.14
n. 78,338 n.52, 363 n.69 X.5 thema 362 n.58
366 n.8 IX.6.13 342 n.14 X.5.12 126,164-5,342
IX.1.13 33,79,245, IX.6.14 164 n.14
305-6,333 n.78 IX.6.15 156 X.5.17 151
IX.1.14 33,306 IX.6.16 245,297,343 X.5.18 151,176
IX.1.15 33,82,342 n.15, n.33,366 n.10 X.5.19 33-4,151,333
361 n.47 IX.6.18 40,48,66,139, n.78
IX.2 thema 180,197 339 n.7 X.5.20 33-4,333 n.78
IX.2.3 45 IX.6.19 184,338 n.52 X.5.21 24,33-4,245,
IX.2.4 210 333 nn.76-8,
IX.2.11 338 n.52 X pr.l 35,148 361 n.47
IX.2.12 281 X pr.2 10,57-8,139- X.5.22 33-4,333 nn.
IX.2.13 338 n.52 40,197,234,292, 76-8
IX.2.17 337 n.42 304,367 n.3O X.5.23 33-4,210,219,
IX.2.18 43,45,81,181, X pr.3 7,57,137-9,292, 333 nn.76,78,
338 n.52 334 n.8O 361 n.47
IX.2.19 184-5,338 n.52, X pr.4 3,7,57-8,139, X.5.24 33-4,224-5,333
357 n.8 148,292,304 nn.76,78
IX.2.20 226,363 n.69 X pr.5 16,139-40,197, X.5.25 19,33-4,
IX.2.21 216,361 n.47 200,292-3,354 224-5,333 nn.
IX.2.22 82,342 n.14 n.23,356 n.23 76,78,363 n.69
IX.2.23 6,148,201,270 X pr.6 16,19-20,22, X.5.26 33-4,210,333
IX.2.24 31,82,201-2, 26,140 n.78
248,286,342 n. X pr.7 16,19,26,140 X.5.27 33-4,219,333
15,359 n.17,362 X pr.8 110,330 n.33, nn.76,78,361
n.56,363 n.69 348 n.19 n.47
IX.2.25 192,197,286, X pr.9 10,52,57,199, X.5.28 23,33-4,219,
305 217-18,277, 333 nn.76,78,
IX.2.26 192,197,220, 293,358 n.8 334 n.81,361
305,318,342 X pr.10 18-19,82,200, n.47
n.14,361 n.47 293,332 n.57, X.5.19 33-4,333 n.78
IX.2.27 28,208-9,216- 342 n.14 X.6.1 294
18,220,305,342 X pr.ll 293-4,342 n.14,
n.14,361 n.47 372 n.104 Suasoriae
IX.2.28 82,217,221,342 X pr.12 18,40,239,
n.14,361 n.47 294-5,373 n.117 1 thema 152
IX.2.29 297,333 n.77 X pr.13 9,10,214,277-8, 1.1 210
IX.3.1 45 288,296,342 n. 1.2 210,294
IX.3.5 338 n.52 14 1.3 47,136,290
IX.3.10 338 n.52 X pr.14 142,296 1.4 210,271-2,372
IX.3.11 171 X pr.15 80-1,171,279, n.102
IX.3.12 284 296,338 n.56, 1.5 54,79,87,89-
IX.3.13 237-8,283,333 357 n.4 90,284,308,
n.71 X pr.16 94,142,172, 343 n.31
IX.3.14 209,278,311, 296,334 n.8 1.6 26,89,284
342 n.14 X.I thema 111-12,349 n.24 1.7 89,338 n.59
IX.4 thema 349 n.31 X.I.I 47.291 1.8 89,154,284
IX.4.5 338 n.52 X.I.3 112,294 1.9 47,154,290
IX.4.10 45,156 X.1.8 101-2,254,346 1.10 47,154,290
IX.4.11 338 n.52 n.69 1.11 210
IX.4.14 48,338 n.52 X.I.10 337 n.4O 1.12 33,81,199-200,
IX.4.15 338 n.52 X.I.12 363 n.69 210,217-19,298,
IX.4.16 338 n.52 X.I.13 191,295,342 311-12,333 n.
IX.4.17 139 n.14 76,341 n.ll,
IX.4.18 139,147 X.I.14 289 342 n.14,358
IX.4.19 139 X.I.15 47,291 n.10,361 n.47
IX.4.20 139-41 X.2.10 41 1.13 82,210,277,333
IX.4.21 10,15,139-41, X.2.12 338 n.52 nn.76,78,342
343 n.20 X.2.13 168 n. 14,361 n.47
IX.4.22 176,225,363 X.2.15 168 1.15 317,324,375
n.69 X.2.17 168,294 n.27
IX.5 thema 175 X.2.18 311 1.16 217,333 n.76,
IX.5.3 338 n.52 X.3.1 294 361 n.47
IX.5.6 338 n.52 X.3.16 338 n.52
IX.5.10 175 X.4.2 210,280-1 2 thema 152,219,362
IX.5.11 329 n.23,342 X.4.15 151 n.59
n.14 X.4.17 293 2.1 210,212
IX.5.15 139,260,269, X.4.18 25,82,216,293, 2.2 212
291,315,342 333 nn.76-8, 2.6 45
n.14,351 n.47 342 n.15,361
n.47
Indexes 417
2.10 36,154,201, nn.23,25,27 40.10-11 286

211,219,238, 6.13 87,139,337 40.12 274,308
248,286,334 n.41 52.11 308
n.80 6.14 33,37,91,307, 59.7 309,374 n.10
2.11 153-4,305 323-4,342 n. 63.1 309-10
2.12 10,21,308,316, 14 63.14 310
323-4,342 n.14 6.15 33,37,81,91, 64.2-5 308
2.13 194,342 n.14 285,307,323-4, 98.13 310
2.14 79,209,333 nn. 342 n.14 100.1-12 28,72,274-5,
76-8,342 n.14 6.16 14,21,27,33, 286,308,341
2.15 66,309-10,342 91,307,315, n.43,370 n.72
n.15 342 n.14,375 108.3 308
2.16 82,219-20,342 n.23 108.17-21 309-10
n.14,361 n.47, 6.17 33,45,91,307 108.22-3 21,26,308-9,
366 n.9 6.18 33,91,307 331 n.36,333
2.17 50,73,237,338 6.19 15,33,45,91, n.65
n.59,365 n.5 306-7 114.1 133,215-23
2.18 209,227,363 6.20 33,91,306-7 114.2 71
n.69,364 n.81 6.21 33,63,91,305, 114.4 249-50,358 n.9
2.19 232-3,312,316, 307,342 nn.13, 114.5 249,266,358
324,342 n.14 14,374 n.7 n.9
2.20 313,338 n.59 6.22 33,63,91,307 114.6-8 358 n.9
2.21 337 n.40,342 6.23 15,33,91,306-7 122.11 316,374 n.15
n.14 6.24 33,37,45,91,
2.22 15,102,139, 307,348 n.19 Ouaest.nat.
141,227,335 n. 6.25 33,37,81,91,
19,342 n.14, 307,316,342 I pr.12 276
363 n.69,364 n.14 V.3.1 276,366 n.18
n.81 6.26 91,307,317, VII.32.2 308
2.23 36,201,211-12, 375 n.23
219,364 n.81 6.27 21,91,190,307, Thyestes 798-9 329 n.15
316-17 324
3 thema 152 338 n.59 Fr.98 Haase 14-16,331 n.48
3.1 182,210,212, (De vita 332 nn.52-4
246-51,274,276, 7 thema 20,84,152, patris)
313,335 n.14, 323-4
359 n.14,366 7.2 343 n.25
nn.11,14,16 7.3 86,102,343 SEXTUS EMPIRICUS
3.3 154,182,321 nn.24,28
3.4 182,248-9,313 7.5 343 n.25 P.H. 1.204 348 n.14
3.5 33,248-50,313, 7.10 23,88,152
324,374 n.19 7.11 82,199,221,
3.6 78,313,330 n. 342 n.14,361 SOPHOCLES
33,338 n.59,342 n.47
n.14,374 n.19 7.12 284 Trach.1060-80 261,368 n.45
3.7 36,81-2,212, 7.14 179,209,225,
313,315,338 n. 284,311,333
59,342 n.14, n.77,363 nn. STRABO
374 n.19 69,72,73,364
n.74 XIV.1.41 366 n.19
4 thema 110,152-3
4.1 212 Liber suasori- 34,328 n.l
4.2 212 arum: colophor,
i SUETONIUS
4.3 212
4.4 153,184,313,321 Fr.3 MUller 312,338 n.59 De gramm. et rhet. (ed.
4.5 36,201,211-12, G. Brugnoli)
266,313
SENECA THE YOUNGER 3.6 5
5 thema 152 4.1 350 n.36
5.1 210,212,250, Cons.Helv. 4.2 93
360 n.33 4.6 5
5.2 212 2.5 331 n.51 7.3 6,124,329 n.14
5.3 212 16.3 333 n.64 7.4 4,328 n.6
5.4 337 n.42 16.4 21 20.2 64
5.6 183 17.3 20,22,26,308, 25.1 334 n.84
5.7 153 332 n.60 25.3 89-90
5.8 14,81,148,279, 17.4 22,26,331 n. 25.4 123
323,342 n.14 36,333 n.66 25.8-9 89,113,115-22,
18.9 330 n.28 126,349 n.26,
6 thema 84,152 19.2 9,328 n.ll, 350 nn.35,36,
6.1 87,287 330 n.29 351 n.46
6.2 86,287,343 nn. 26 67,89-90
24,27 Cons.Marc.23. 136 27 67,93
6.3 46,86,343 nn. 2-3 28 67
22,23,25,26 29 67
6.4 86,343 n.25, De brev.vit. 274,308 30 60,66,339 n.13
344 n.38 10.1
6.5 88,343 n.25 De vit.Caes.
6.7 343 n.25 De clem.I.10.1 344 n.48
6.8 153,182,211 Aug.
6.9 20,85,153,353 De prov.4.14 358 n.15
n.6 84 59-60
6.10 154 Ep.mor. 86 59-60,243-4,
6.11 86-8,139,279, 297-8,302,304,
343 n.22 14.13 331 n.45 368 n.53
6.12 86,139,343 19.9 249 87-8 59-60
Indexes 418
De vit.Caes. (cont.) VALERIUS MAXIMUS
Tib.73.2 17,332 n.55 I. 1 185

VI .9 ext .7 72
Cal.16.1 354 n.18 IX .15 185
De poetis (GRF 64-6

209-12)
Imagines (GRF 64-5

I.I 17 214-15)
1.65 191
IV. 61 286-7
VI.3 10,141 VARRO OF ATAX (Morel, Frag.
XV.73 330 n.34 Poet.Lat.93-9)
XVI.17 14,330 n.34
fr.8 266,312,315
Dial.
5.3-10.8 138-9,353 n.9 VELLEIUS PATERCULUS

18.4-6 221-2,362
n.63 1.16.3 136
19.1-23.6 322-3 1.16.4 136,310
19.1 138,282,374 1.16.5 136
n.3 1.17.1 311
19.2 282 1.17.2 136,306-7,311
19.3 322 317
19.5 259,263,368 1.17.3 83-4,136
n.42 1.17.6 136-7
20.1 365 n.7 1.17.7 136-7
21.2 98 II.66.5 344 n.34
26.1 278,302-3
26.4 282
35.5 144 VIRGIL
36.1-37.5 134,143,145
37.6 86,134,145 Eel.7.12-13 266
37.7-8 134,145
41.5 353 n.9 Georg.
1.224 248
TERTULLIAN 1.427-9 249,313,324
1.432-3 249,324
De Pallio 5 138
Aen.
THEON Progymn. (Rh.Gr.11.60- 1.33 206

130 Spengel) 11.553 313
IV.379-80 313
65-72 347 n.4,350 VIII.26-7 312
n.35 VIII.691-2 200,218
120.13-15 125-6 XI.288-90 312-13
X.128 199,218
XII.646 203
THUCYDIDES
1.1-21 17 (VIRGIL)
1.138.3 374 n.7
II.65.6-13 374 n.7 Catal.2 220,362 n.60
1.3.51-2 269
1.4.21-6 269 Apologia
Socratis
TIMAEUS (FGrHist 566)
109,258,368
n.40
TIRO (GRF 390-403)
fr.6 66
frs.8-11 66
VALERIUS FLACCUS
V.368 248

Fairweather-Seneca The Elder PDF

Uploaded by

Copyright:

Available Formats

Fairweather-Seneca The Elder PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fairweather-Seneca The Elder PDF

Uploaded by

Copyright:

Available Formats

What is the main subject and scope of the document?

What is the main subject and scope of the document?

What are some of the major sections or parts discussed in the document?

What are some of the major sections or parts discussed in the document?

Seneca the Elder

CAMBRIDGE UNIVERSITY PRESS

Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK

This publication is in copyright. Subject to statutory exception

First published 1981

ISBN 978-0-521-23101-5 hardback

PART I. THE PLACE OF SENECA THE ELDER IN LITERARY HISTORY

PART II. SENECA THE ELDER ON THE HISTORY OF ELOQUENCE

PART III. FIVE ASPECTS OF DECLAMATION: THE ELDER SENECA'S EVIDENCE

PART IV. THE PLACE OF EARLY IMPERIAL DECLAMATION IN LITERARY

Appendix: Clausula usage of Seneca, Latro and Fuscus 326

The English-speaking world was without a monograph on Seneca the

research into the authenticity and style of the declamatory frag-

Brugnoli Suetonius, De grammaticis et rhetoribus,

Note: For an explanation of symbols used in references to the text

The elder Seneca had every reason to consider himself singularly

having been invented for use in Greek schools of rhetoric, prob-

vivam vocem audire. (Contr. I pr. 11)

rhetorician, Seneca did not confine his attention to his friend's

privilege, perhaps to be explained by a friendship between Pollio

suggestion of J. B. Hall ZH that et should be deleted after avidus,

be considered as part of the background to his memorable words on

second question, we have noted the difficulty of finding examples

the emphatic words sed proderit tibi...declamandi exercitatio...

That the declamatory anthology was composed at least in part

consider the state of the world as a whole. The younger Seneca

In a historical fragment ascribed to 'Seneca' by Lactantius,

the second childhood of the state:

To be on the safe side, however, we must restrict our atten-

dementissimi homines, tanta vecordia agitat? (Contr. X pr. 6)

aridi declamatores fidelius quos proposuerunt colores tuentur:

to the immortal gods.

When Seneca criticizes samples of bad taste by the Greek declaim-

subject, hoc autem vitium aiebat Scaurus a Graecis declamatoribus

The elder Seneca mentions several reasons why he came to write

Precedents are to be found in Hellenistic technical writings:

Seneca makes it clear early on that he has reasons for writing it

Circus, to which he whimsically alludes at the end of the preface:

very diverse in kind, extracted from a number of declaimers*

Seneca was quoting continuous prose at considerable length.11*

The sententiae recorded in the opening section of Seneca's

to their Latin counterparts. The survey of Contr. X.5, which

but ought to be, though it is probably unanswerable, is where in

controversiae and suasoriae. (The controversiae would have to

Haterius, introduced in the fourth preface, is given the prominent

capacity to undertake the task at hand had been standard practice

occasionally making passing references to written records, and to

(II.11.7). It is not inconceivable that an adult enthusiast might

Athenaeus1 epitomator recognizes the fact of the matter when he

not Montanus; in one of the cases cited Seneca is actually quoting

The figures are as follows: Seneca - 72% (inclusive of 23% resol-

Seneca Latro Fuscus Cicero 'Normal'

(Greifswald, 1872), a work which Bornecque does not appear to have

to in Quaestiones 5f. It does not seem that we have a large