THE ROMAN STAGE

W. BEARE, m A.
Professor of Latin in the University of Bristol

THE ROMAN STAGE

A SHORT HISTORY OF LATIN DRAMA
IN THE TIME OF THE REPUBLIC

HARVARD UNIVERSITY PRESS

CAMBRIDGE, MASSACHUSETTS
1951
PRINTED IN GREAT BRITAIN
 то
WILLIAM BLAIR ANDERSON
formerly Hulme Professor of Latin in the
University of Manchester
and
Kennedy Professor ofLatin in the University
of Cambridge
 PREFACE
Y A T T E N T I O N was first drawn to Plautine studies by
M Robert Seymour Conway, at that time (1924) Hulme
Professor of Latin in the University of Manchester. Much
interest had been aroused among English scholars by the
appearance in 1922 of Eduard Fraenkel's well-known work
Plautinisches im Plautus, dedicated to Gunther Jachmann and
acknowledging a special debt to Friedrich Leo. I found,
however, in the course of time that my interest lay not so much
in the relation of the Latin plays to their Greek originals, or
in those linguistic, metrical and textual studies, which in this
country are peculiarly associated with the names of W. M.
Lindsay and E. A. Sonnenschein, as in the light thrown by the
plays on the theatre for which they were intended. It seemed
to me that nearly all modern discussions of Latin drama
rested on certain assumptions, conscious or unconscious,
which might seem reasonable in themselves but were incapable
of proof and sooner or later led to serious difficulty. Ever
since the Renaissance the scholars of western Europe, even
while basing their standards on the supposed practice of the
Greeks and Romans, have in fact allowed contemporary
habits of thought to colour their conception of classical
antiquity. It might have been well to pay more attention
to the music, the art and drama of India and Java, of China
and Japan, if only to remind ourselves that our western
notions are not valid for all mankind, even at the present
day. My own limitations and misgivings have confined me
to a narrower field. I have tried to avoid using arguments
which depended either on subjective judgments, or on
evidence the validity of which was less than axiomatic. The
reader may be disappointed by my failure to reconstruct for
him the plot of some lost play, whether tragedy, togata or
Atellane ; he may even think it perverse in me to have
made so little use of the wealth of pictorial material which
Dr. Bieber has put at our disposal, or the many references to
drama which we find scattered throughout ancient literature.
On this last point perhaps I may quote what seems an
apposite remark of Mr. L. P. Wilkinson (Horace and his Lyric
Poetry, p. 19) : ' I t is difficult for us, who use words primarily
as a means of conveying what we believe to be the truth, to
penetrate the mind of Roman writers, who, like the Italians
rii
viii THE ROMAN STAGE

of to-day, often used them primarily " for effect." ' I have
placed little trust in the statements of such writers as the
elder Pliny, and have even reserved judgment with regard
to much of what is said by Vitruvius, however technical and
detailed. On the other hand, when Lucretius (iv. 75-83)
refers incidentally to the many-coloured awnings ' stretched
over great theatres' as they ' flap and flutter, spread every-
where on masts and b e a m s a n d in the subdued light of the
enclosed theatre ' tinge the assembly in the tiers beneath, and
all the bravery of the stage and the gay-clad company of the
e l d e r s w h o can fail to detect the note of truth ? But in the last
resort we come back to the plays as the primary source of
evidence which is certainly valid and relevant. It is also evidence
which is easily accessible to the reader, who may judge for
himself whether the arguments which I base on it are sound.
I owe a debt for help and encouragement to many friends,
including Professor P. J. Enk, of Groningen, Professor G. E.
Duckworth, of Princeton, Professor J. F. Mountford, of
Liverpool, and Professor J. A . K . Thomson, formerly of
King's College, London. Since it has been my misfortune
to have to differ from the views expressed on New Comedy
by Professor Gilbert Murray and Professor Τ . B. L. Webster,
I have special pleasure in acknowledging the courtesy with
which they have received my criticisms. My colleagues,
Professor H. D. F. Kitto and Dr. A. Momigliano, have read
my book in proof: I owe it to them that the mistakes are not
even more numerous. Mr. B. L. Joseph has allowed me to
see an advance copy öf his book on Elizabethan Acting, shortly
to be published by the Oxford University Press. If the
Elizabethan art of acting had closer affinity with Roman rhetoric
than with the naturalistic technique of our modern picture-
stage, how much more true must this be of the art of acting in
ancient Rome ! Among the American scholars whose works I
have consulted I would mention in particular Professor Tenney
Frank, Professor Roy Flickinger and Professor P. W. Harsh.
I have reserved to the last the names of Sir A . W. Pickard-
Cambridge, whose Theatre of Dionysus in Athens represents
perhaps the most important advance in our knowledge since
Dörpfeld's day, and Professor W. B. Anderson, whose W o r k has
lain mainly in other fields, but whose standards öf scholarship
are valid for us all. T o him I affectionately dedicate this book.
 CONTENTS
CHATTER ГАС·

I. INTRODUCTION ι
II. DANCE AND DRAMA : THE PRE-LITERARY PERIOD
IN ITALY 9
III. LIVIUS ANDRONICUS AND THE COMING OF
LITERARY DRAMA TO ROME 15
IV. NAEVIUS 23
V. PLAUTUS : LIFE AND LIST OF PLAYS 35
VI. GREEK NEW COMEDY 40
VII. THE FAMOUS PLAYS OF PLAUTUS 46
VIII. PLAUTUS : TREATMENT OF HIS ORIGINALS 53
IX. THE GENERAL CHARACTER OF ROMAN TRAGEDY 60
X. PACUVIUS 69
XI. COMEDY AFTER THE DEATH OF PLAUTUS 75

XII. TERENCE 81
XIII. THE OTHER COMPOSERS OF PALLIATAE 105
XIV. ACCIUS hi
XV. NATIVE COMEDY : THE FABULA TOGATA 120
XVI. POPULAR FARCE : THE FABULA ATELLANA 129
XVII. THE LITERARY ATELLANA 135
XVIII. THE MIME 141
XIX. THE LATIN PROLOGUES AND THEIR VALUE AS
EVIDENCE FOR THEATRICAL CONDITIONS 151
XX. THE ORGANIZATION OF THE ROMAN THEATRE 156
XXI. SEATS IN THE ROMAN THEATRE 163
XXII. THE SPECTATORS 165

XXIII. THE STAGE AND THE ACTORS* HOUSE 168

XXIV. COSTUMES AND MASKS 176

XXV. THE ROMAN ORIGIN OF THE LAW OF FIVE ACTS 188

XXVI. MUSIC AND METRE 211
XXVII. EPILOGUE : DRAMA UNDER THE EMPIRE 225
ix
χ THE ROMAN STAGE

APPENDICES
A. SEATS IN THE GREEK AND ROMAN THEATRES.
(С.Д. liii. 51-5) 233

B. SIDE-ENTRANCES AND ΠΕΡΙΑΚΤΟΙ IN THE HELLENISTIC

THEATRE. (C.Q.. xxxii. 205-10) 240

C. THE A N G I P O R T U M AND R O M A N DRAMA. (Hermathena

xxviii. 88-99) 348

D. CREPIDATA, PALLIATA, TABERNARIA, TOGATA.

(С.Д. liii. 166-8) 256

E. THE ROMAN STAGE CURTAIN. (Hermathma Iviii.

J 04->5) 359
F. CHANGE OF SCENE AND CHANGE OF SCENERY :
THE QUESTION OF ' S E T S ' 267

G. THE DOORS SHOWN ON THE STAGE 277

SHORT BIBLIOGRAPHY 287

INDEX 290
 ILLUSTRATIONS
Iphigeneia in Tauris, illustrating so-called prothyron.
Volute krater, Buckingham Collection . . . title page
(Mon. d. Inst. IV, pi. 51)
Reconstruction of Theatre at Oeniadae with thyromata facing 1
(Bieter,1 Jig. 332)
Spectators and Performers at the Etruscan Games „ 10
ι. Part of the Tomba della Scimmia at Chiusi
(Poulsen,1fig.so)
2. Part of small frieze in the Tomba delle Bighe
(Poulsen,1 fig. 19)
Scenes from the Etruscan Games . . . . „ 10
(Poulsen,1figs.4, 5 and 6)
' Chiron goes upstairs ' : Phlyax Vase-painting . „ 15
(Photo : British Museum)
Scene from Comedy : Relief from Pompeii in Naples
Museum . . . . . . . . . 38
(Photo: Alinari)
Jupiter and Mercury. Phlyax Vase-painting in the
Vatican
(PhotoMuseum
: Alinari) . . . . . » 5 0
Roman Theatre
1
Tickets. . . . . . page 167
(Bieber, fig. 455)
Poet and Tragic Muse: Choosing the Masks (?). Relief
in the Lateran Museum . . . . . facing 178
(Photo : Alinari)
Street Musicians : Mosaic by Dioskorides, from
Pompeii . . . . ' . . . . „ 222
(Photo : Alinari)
The Large Theatre, Pompeii . . . . . . 262
(Photo : Author)
The Roman Theatre, Vaison-la-Romaine, showing
curtain-slot . . . . . . . . 262
(Photo: Author)
Mechanism of drop-curtain, according to Mazois . page 266
(Fiechter* fig. 119)
The Roman Theatre, Orange . . . . facing 274
(Photo : F. Beau, Avignon-Monclar)
Relation of house-doors to the side-entrances . . page 27 7
Construction of Greek or Roman doorway . . „ 282
Woman unlocking front door . . . . . „ 285
(D. & S., s. v. ianua,*fig.4138)
Woman looking out of door . . . . ,, 285
(D. & S., s. v. ianua*fig. 4131)
1
From Bieber : History of the Greek and Roman Theatre by permission of
Princeton University Press.
1
From Pouben : Etruscan Tomb Paintings by permission of Oxford University
Press.
* From Fiechter : Baugeschichtliche Entwicklung des antiken Theaters by per-
mission
4
of Becksche Verlagsbuchhandlung, Munich
From the article on ianua in Daremberg et Saglio by permission of Messrs.
Hachette.
Reconstruction of theatre at Oeniadae with thyromata
 CHAPTER I

INTRODUCTION

I N THIS book I shall attempt to give a connected account of

the drama produced in the theatres of ancient Rome.
We still possess, under the names of Plautus and Terence,
twenty-six comedies translated from Greek into Latin and
intended for production on the stage. These works belong to
the late third and early second century before Christ. In
addition we have a number of fragments remaining from the
drama of Republican times, from tragedy and comedy,
translations of lost Greek plays and original compositions,
which help us to trace the history of the subject from the time
of Livius Andronicus to the end of the Republic. From early
Imperial times we possess, under the name of Seneca, ten
tragedies, which read as if they were intended not for produc-
tion but for declamation or private reading. Though framed
on Greek lines, these are not translations but independent
compositions ; indeed one of them, the Octauia, takes its
subject from contemporary Roman history, and has as its
heroine the ill-fated wife of the emperor Nero.
With the coming of the Empire the subject ceases to be a
unity; we have now an almost complete divorce between
literary drama, written for reading or recitation, and the stage,
almost monopolized by mime and pantomime. The contrast
between the Republic and the Empire is indeed one of the
most striking facts in the history of Roman drama ; yet as the
coming of the Empire only confirmed tendencies which had
already begun to display themselves under the Republic, it
will sometimes be helpful to refer, with due caution, to the
statements of late authors and to the monumental evidence
which Has come down to us from Imperial times.
The Latin plays no longer occupy the place which they
held at the time of the Renaissance, when they were the
acknowledged models of dramatic art. Even classical scholars
now prefer to read the extant Greek plays, original works by
I
2 THE ROMAN STAGE

the greatest of Greek writers, rather than Latin translations

and imitations. Yet the study of Latin drama is one of lasting
interest. The plays of Plautus are the first works of Latin
literature which have reached us complete, and his prologues
contain eye-witness descriptions of the generation of Romans
who defeated Hannibal. Apart from the translations made
by Plautus and Terence we do not possess a single complete
example of the social drama—the so-called New Comedy—
which was popular all over the Greek world of the third
century B.C. If our concern is with the theatre as a form of
popular entertainment, we may learn something from the
struggles of the Latin dramatists to adapt their Greek originals
to the taste of the boisterous crowds which flocked to the public
shows of Republican Rome. The influence which Latin plays
and Roman theories exercised on the dramatists of the
Renaissance makes it necessary for us to know something about
Plautus, Seneca and Donatus if we are to appreciate the
background of Ben Jonson and Shakespeare.
No connected acoount of the Roman theatre and drama
has reached us from Roman times. For our knowledge we
depend mainly on the surviving plays and their prologues,
supplemented by scattered references in ancient writers and
by the evidence of archax>logy and works of art. No one doubts
that of these three sources by far the most important is the text
of the plays. Archaeology can be of only subordinate value in
a matter which is primarily literary, dramatic and theatrical;
moreover the existing archaeological remains, reliefs, terra
cottas, etc., do not take us back earlier than the Empire. The
statements made by ancient writers belong for the most part
to the Empire, when theatrical conditions had changed, and
they were made by men who were often not primarily
concerned to tell the facts or in a position to discover them.
With the best will in the world, it is no easy matter to see
things as they are, and to describe them fairly and clearly.
The objective statement of fact can scarcely be said to have
been the main object of Roman writers, even of men as great
as Livy and Cicero. They had something higher in view :
To point a moral, or adorn a tale.
Much of our information is derived from third-rate gram-
marians, men whose outlook was narrowly bookish, whose
 INTRODUCTION 3
grasp of reality was weak, who strove to force the facts to fit
their theories. Their notions were to become canonical
principles for the scholars of the Renaissance, who regarded
Roman drama as the standard of perfection. Thus our
nascent English drama came to be subjected to rules, supposedly
Roman, which were never known to the ancient stage. This
at least we must concede to the scholars of the Renaissance :
they were influenced by Roman theories recorded in works
which had survived and by Latin plays still extant. A new
source of confusion was to be introduced by the shift of interest
from Roman to Greek literature. The modern classical scholar,
particularly the German scholar, is only too ready to form an
ideal picture of that unapproachable perfection which was
Greece, and by such a standard to judge—and to emend—the
material which has actually come down to us. And at all
times, from those of Cicero to our own, the plays of past ages
have been studied as literature, written to be read, rather than
as drama, designed for presentation on the stage.
The most obvious feature of Roman drama is its derivative
character. All our Latin comedies are based on Greek
originals, themselves now lost. It is plain, moreover, that the
Latin plays are not exact translations ; the translators omitted
what they thought dull, added what they thought would
interest their public and introduced other alterations of various
kinds. The scholars of Europe and America have made
ingenious attempts to distinguish between the Greek and
Roman elements in the extant Latin plays ; but the more
elaborate the arguments used, the more disputable the con-
clusions which are drawn. Much confusion would have been
avoided if writers had always made it plain whether they were
considering the Latin plays in themselves or using them as
evidence for the form of the Greek plays. My aim, at
any rate, is to consider the Latin plays in their present
form as evidence for literary and theatrical developments
in Rome.
This raises the question of how far our texts are sound.
It is clear that Plautus was not the author of those lines in the
prologue to the Casino, which refer expressly to a revival of the
play after his death. It is not likely that he was responsible
for both the alternative last scenes of the Poenulus, or that
Terence provided both the alternative last scenes to the
4 THE ROMAN STAGE

Andria—indeed in both cases it is evidently the second alterna-

tive which was added by a later hand. Occasional passages
in other prologues to Plautus' plays refer to him in a detached
manner hard to explain if he was indeed the author of the lines
in question. These passages, and the alternative endings, appear
to be interpolations introduced for revival performances—and
they are introduced at the beginning and the end of the plays,
where presumably there was room in the manuscript. But
when we consider the grounds on which other passages have
been suspected, we have to admit that they are inconclusive.
One reader may think a particular passage unworthy of the
supposed author; but what do we know about the style of
Plautus except what we infer from the text of his plays ? There
may be repetitions or even contradictions in the text ΐ but is
it not easier to suppose that these faults (if they are faults)
are due to the inadvertence of the original author than that
they were deliberately introduced by a later hand ? It may be
that for some time after the author's death his plays existed
merely in acting-editions, at the mercy of the producer who
happened to own them. It would be possible, no doubt, for
such a producer to make alterations—but we must suppose that
he would not do so without a reason. He might curtail a
scene which seemed dull, or throw in an occasional jest. But
as these were precisely the methods by which Plautus is sup-
posed to have altered his Greek originals, it is not likely that
we should be able to detect the work of the producer who carried
this process a step further. Ritschl used references in the
prologues to seats as a proof that these prologues were post-
Plautine. The truth is rather that these references (which
occur not merely in the prologues but in some of the plays)
are proof that some at least of the spectators in Plautus' day
were seated. There is an a priori probability that producers
would leave the text alone ; modesty, prudence and laziness,
the difficulty of writing in Plautine metre, the lack of room in
the manuscript, would suggest the advisability of not meddling
with the author's words. The prologues of Terence, written
in reply to contemporary critics, must have meant little or
nothing to the theatre-going public of later ages. Yet these
prologues have come down to us with the plays to which they
refer. Their preservation is itself proof of the respect with which
the author's manuscript was treated. It is not easy to suppose
 INTRODUCTION 5
that the archaic plays of Livius Andronicus were popular on the
stage after his death ; yet they survived until Cicero's day.
This suggests that the bulk of the dramatic production of the
Republican period was preserved. Notes of revival perfor-
mances were occasionally jotted down on the title page, but
we have no evidence of deletions, and the body of the play
probably remained in the form given to it by its author. There
was, it is true, some uncertainty about the authorship of
particular plays ; according to Varro, plays of other authors
came to be ascribed to Flautus, no doubt to enhance their
commercial value. But whatever changes were made, they
appear to have been made with an eye to performance ; in
other words, they belong to the period when drama was being
produced for the theatre.
Each Latin comedy that we possess, no matter what its
origin, is evidently a unity, the creation of some one man, who
knew the Roman theatre and public and adapted the play for
that public and that theatre. I do not share the usual view
that the Roman dramatists materially altered the structure of
their Greek originals. But even if we choose the other extreme
and narrow our conception of their creative activity to the
adaptation of their originals, scene by scene and line by line,
to the Latin language and the necessities of the Roman theatre,
the fact remains that these translated works have a life of their
own. They are worth studying for their own sake.
Much space is given i n ' o u r histories of literature to the
biographies of the dramatists. Where the early drama is
concerned such details are particularly open to doubt. The
Latin dramatists were not in their day regarded as of such
importance that anyone should trouble to record the facts
concerning them while those facts were available. They were
hard-working professional writers ; they gave their lives to
earning a living, and went to their graves leaving, it may be,
nothing to preserve their memory except the works which they
had written. Somehow these works survived, and later
generations of readers found themselves curious to know
something about the authors whose names they found on the
title-pages of the works they were reading. Attempts were
made to satisfy this curiosity by examining the text of the plays
or autobiographical details. Yet in the nature of things a
ramatist gets small opportunity for autobiography. Other
A 2
6 THE ROMAN STAGE

evidence was supplied by the manuscript, in which were

sometimes entered notes concerning the opening or revival
performances. Official records might mention the plays pro-
duced at the Games ; noble families had their archives.
Where there was no evidence, the biographers fell back
(perhaps without realizing it) on invention.
The Latin comedies can tell us little about the lives of the
men who wrote them ; but they are our best source of evidence
for the conventions and the practices of the Roman theatre.
It is agreed by all that we should try to read them as plays and
to picture them as performed on the Roman stage. But
modern readers are accustomed to a very different theatrical
convention ; we bring to our study of Latin drama assumptions
of which we ourselves may not be fully aware, but which are
only too likely to colour our picture and distort our judgment.
Consider so familiar a word as ' interlude' : how frequently
it is used by writers on Roman drama, and yet how modern are
its implications ! In trying to describe what actually took
place on the Roman stage, I have found myself forced to
sacrifice many of my earlier hypotheses and to try to get behind
some of my initial assumptions. The results of my enquiry
are largely negative. For example, when I find the dramatists
using various expedients, sometimes far-fetched, to overcome a
difficulty which would nowadays be met by the use of a
technical device, I infer that this particular device was unknown
to those dramatists. Simplicity was the keynote of the Roman
stage. The simpler the hypothesis that will explain the facts,
the more likely it is to be right.
The known facts are all too few. We must often be content
to remain ignorant. It is only too tempting to frame theories
to fill up the gaps in our knowledge ; but these theories raise
difficulties which necessitate the formation of other theories,
until we lose contact with reality. Attempts are made to press
into service material the relevance of which cannot be demon-
strated : now it is the illustrations in the manuscripts of
Terence, at another time the Italian vase-paintings, terra-
cottas and so forth. Fashions in research change ; scholars
are carried away by prevailing literary, artistic or even political
theories. No change of fashion can alter the fact that the
Latin plays are plays, and that, as plays, they may be studied
in their own light.
 INTRODUCTION 7

THE PLACE OF DRAMA IN ROMAN NATIONAL LIFE

Nearly all the early writers of Rome were dramatists.

Down to the death of Accius we hear of no verse writer who
did not devote at least part of his time to the making of plays
for the stage. In a period when there was as yet no reading
public whose support could maintain an author, the drama
offered almost the only means of earning a living with one's
pen. The Roman dramatists were not men of independent
means ; they depended partly on patronage, but such patronage
might soon have dried up had they not shown themselves
capable of securing the favour of the public for themselves and
their patrons. When men of the ruling class took up the com-
position of works in verse, the creative days of drama were
over. Up to then literature had been directed to utilitarian
ends. The value of drama as one means of entertaining the
public (who were also the electorate) was realized by the
authorities responsible for organizing the public games. From
the beginning the Roman theatre was supported and supervised
by the State ; its development seems to have been as much
the result of official policy as of native impulse. The Etruscan
dance and the drama of Greece seem alike to have been
imported by governmental action. To assimilate borrowed
material was the Roman way. To the end Roman drama was
derivative ; the fragments which we possess of original Latin
plays suggest that even native drama owed something to Greek
models. Unfortunately there is no extant example of a stage-
play with a plot invented by a Roman.
Nevertheless the Romans had a marked taste for the
theatrical, and Latin literature is rich in dramatic touches—
effective dialogue, vivid scenes, memorable character-sketches.
The Roman mentality had two rather incongruous elements—
a liking for the rhetorical and spectacular and a liking for the
ridiculous, for homely farce, satire and repartee. We can see
that the imported drama was coloured by these national
qualities : Roman tragedy was more rhetorical than its Greek
original, Roman comedy more inclined than Greek New
Comedy to farce. Our modern taste dislikes rhetoric and
enjoys satire ; hence we are apt to consider Roman comedy as
superior to Roman tragedy. This was not the view of Roman
critics, or of the theatre-going public. If anything, tragedy
8 THE ROMAN STAGE

outlived comedy on the stage ; down to the time of Cicero the

works of the tragic writers were performed before large and
excited audiences. Moreover, while we have no evidence
that tragedy was forced to admit an element of farce in order
to please the groundlings, we cannot fail to notice a solemn
note in comedy and even in the mime.
 CHAPTER II

DANGE AND D R A M A : THE P R E - L I T E R A R Y PERIOD

IN I T A L Y

T HE GREEKS found the origin of their drama in the dance.

For them ' dance' included every kind of significant rhyth-

mical movement. The feeling for rhythm lies deep in our
nature. Graceful movement gives pleasure both to the per-
former and to the beholder, especially if it is accompanied by
music or song. In the dance primitive man finds expression
for emotion of every kind ; in the dance, too, he gratifies his
instinct for imitation. Among primitive peoples dancing is a
communal activity. The tribe is faced at every turn with the
need for preservation from enemies, disease or hunger ; from
lack of game, of cattle, of crops, of children. Hence the war
dance, the dances in celebration of birth and marriage, the
magical dances intended to promote hunting and agriculture, to
exorcise the demons of sickness and to lay the ghosts of the dead.
For a farming community the most important and the
merriest event of the year is the harvest-home. Then, if ever,
the farmer can forget his cares and join with his friends in
unrestrained revelry, which may easily take a riotous or
licentious turn. The chief dance of the Hos of Bengal is in
January, when the granaries are full. During the dancing, as
in the Roman Saturnalia, masters and servants treat each
other as equals. There is much drinking of beer ; language
and conduct are freed from the customary restraints of deqency.1
Similar was the 4 Fescennine jesting ' of the Roman harvest-
home. Rival clowns, their features disguised in cork masks,
hurled jest and abuse at each other, to the delight of the
company. Such scenes were dear to the Italian heart. Horace
tells us how he and Maecena^were entertained by an exchange
of abuse between two Campanian clowns, and even Marcus
Aurelius Antoninus, heir to the Roman Empire, writes of how,
after a day's work with the labourers in the fields, finding
1 Enc. Brit., s.v.' Dance ' ; Frazer, Golden Bough, ix. 136.
9
 ΙΟ THE ROMAN STAGE

himself too tired to study, he sat in the kitchen and enjoyed

the rustic quips exchanged between his fellow workers. In
the Fescennine verses, the bantering dialogue of clowns at
the harvest-home, the Romans found the germ of their own
drama.
The origin of the name ' Fescennine ' was a puzzle to the
Romans, as it is to us ; some derived it from Fescennium, a
town on the confines of Etruria, others from fascinum, the male
organ of generation, widely regarded as a magic symbol. We
may feel that each of these suggestions is in itself plausible
enough to throw doubt on the other ; the Etruscans exercised
a strong influence on the development of scenic performances
in Rome, and the association of primitive drama with repro-
ductive magic is reflected in historic times by the wearing of
the phallus in Greek Old Comedy. 1
The distinction between solemn ritual and indecent revelry
seems clear enough to us, but it was not so clear to the Italians.
Indeed the two extremes seem to have gone naturally together.
The bride and bridegroom, about to enter into the high duties
of their new state amidst all the solemnity which religion
could afford, had to listen to the crudest jests, flung at them
in accordance with ancient custom. The victorious general,
riding through Rome in the superhuman pomp of the triumph,
could overhear the songs of his soldiers, which referred in the
language of the camp to his bodily infirmities and moral
peccadilloes. A t the head of the funeral procession, even at
the funeral of a dead and deified emperor, might on occasion
be seen an actor who wore the dead man's likeness and
mimicked his characteristic gestures. Indecency in drama
seems to owe its origin to ceremonies on which the very life of
the tribe was thought to depend, though its continuance in
more sophisticated times may be partly due to its power to
raise laughter.
Another matter in which we are perhaps inclined to draw
sharper distinctions than the Romans did is in the various
kinds of public spectacles which they included under the title
ludi, ' g a m e s T h e a t r i c a l performances, gladiatorial combats,
beast-fights and circus races were all l u d i ; their history shows
that they had much in common ; they often took place in the
1 If indeed it was worn, whether regularly or exceptionally ; but this is

unproved and much debated. See Cornford, Origin of Attic Comedy, p. 183.
 гзтгнгжт·

S P E C T A T O R S A N D PERFORMERS A T T H E E T R U S C A N G A M E S
ι Part of the Tomba della Scimmia at Chiusi
2 Part of small frieze in the Tomba delle Bighe

SCENES FROM THE ETRUSCAN GAMES

 DANCE AND DRAMA II

same buildings, and the general public went to them all with
much the same expectation of amusement and thrills.
The first circus performances of which Roman tradition
tells were those given by Romulus with the covert design of
securing wives for his followers from among the womenfolk of
their guests. However that may be, there is perhaps more
probability in the tradition that the Circus Maximus was built
by the elder Tarquin, first of Rome's Etruscan kings, and that
the performance consisted of chariot races and boxing matches,
the horses and the boxers being brought from Etruria. Grand-
stands were erected for the nobles ; the common people sät
on the ground. Every September the whole Roman population,
their harvest safely stowed away, were free to attend the Ludi
Romani at the Circus. There was no charge for admission ;
women and slaves were among the spectators ; the expenses
were borne by the State and the presiding magistrates.
Livy tells us that in the year 364-3 B.C. a superstitious
panic, due to pestilence, created a desire to celebrate the games
in some special way, in order to appease the wrath of heaven.
Dancers were imported, again from Etruria ; their graceful
movements to the sound of the flute 1 made a deep impression.
Amateur native performers tried to combine this decorative
Etruscan dance with the banter and mimicry of the immemo-
rial 1 Fescennine verses'; thus was developed a new type of
performance consisting of dance, repartee and song, accom-
panied by the flute. This ' mixed s h o w w h i c h Livy describes
as a satura or ' medley', though always somewhat formless,
soon ceased to be purely extempore ; the performers, now
professionals, were called histriones (the Etruscan word for
' dancers'), and continued for some time to entertain Rome
with their ' stage-shows ' (ludi scaenici).
In 264, the opening year of the First Punic War, came a
more sinister importation from Etruria, the gladiatorial
combat. That prisoners should be set to fight each other for
public entertainment was no great improvement on the
primitive custom of putting all male captives to death—a
custom which survived even to the time of Julius Caesar in the
triumphal procession (also derived from Etruria), when captives
were executed as the commander ascended the Capitoline hill.
1
' Flute ' is the conventional but inaccurate translation of tibia ; ' p i p e ' is
better. See ch. xx. On the meaning o f ' song ' see ch. xxvi.
12 THE ROMAN STAGE

Familiarity with such grim scenes may have affected the taste
of the theatre-going public ; Roman tragedy, as we shall see,
shows great interest in gruesome themes and details.
These developments are admirably illustrated in the
wonderful Etruscan tomb-paintings, which afford abundant
evidence of the popularity of the dance in Etruria at this early
period. In these paintings we see male dancers playing on the
flute and dancing-girls with castanets and bells ; a young
man and woman dance merrily together near a wine-bowl to
the music of the pipe and lyre. Everything is done to music ;
boxers box and wrestlers wrestle to the sound of the pipe ; the
piper keeps time for the cooks in the kitchen as well as for the
guests on the banqueting-couch. We are reminded of the games
of the Circus when we see processions of horsemen and chariots,
or a picture showing the lassoing of a horse ; more interesting
still is a picture of the aristocratic spectators, men and women,
sitting side by side in curtained ' boxes ' on grandstands, while
on the ground sprawl the common people. Especially striking
is a painting in the Tomba degli Auguri. On the left enters a
spectator, beckoning to a boy who brings his stool (just as, in
the Roman theatre of later days, chairs for the senators were
set in the orchestra). In the middle are a pair of wrestlers,
closely watched by the referee ; on the right is a grimmer
contest. A masked figure, called Phersu in the inscription,
holds a fierce dog by a long leash which he has wrapped round
the legs of his antagonist, who holds a club but is unable to use
it, because he has not yet succeeded in freeing his head from
a sack ; meanwhile the dog has torn great bleeding wounds
in his legs. In another picture we see the masked figure
(again named Phersu) running away. Some philologists
derive from the name of this masked gladiator the word persona,
the Latin term for a theatrical mask. Masks, grandstands,
curtains to protect the spectators, movable seats—these are
all features which will recur in the theatres of Rome.
The evidence of the tomb-paintings supports the Roman
tradition that their theatrical art, in its early stages, was largely
influenced from Etruria. Another early influence came from
Campania, where a rustic farce, performed by masked
characters (fabula Atellana) had long been popular. Greek
colonization had at an early date reached the shores of Cam-
pania, and, long before the introduction of written drama at
 DANCE AND DRAMA *3
Rome, contact with such towns as Naples may have been
familiarizing at least some Romans with Greek theatrical per-
formances, especially those of a popular nature such as the farces
which we see depicted on Greek vases from southern Italy. We
can well believe that before the beginnings of Latin literature
Rome had become familiar with stage-shows of some kind.
Livy describes them as saturae or ' medleys' because of the
formless mixture of dialogue and song. It appears from his
account that there were song-writers and composers of music
in third-century Rome.
Nothing has survived of these early theatrical compositions.
The word ' medley' is merely Livy's description, not a title ;
there was probably never a form of drama entitled Satura in
contemporary speech. But that the Romans had at an early
period formed a taste for dialogue mixed with song is strongly
suggested by their later treatment of Greek New Comedy.
Here they took over a form of drama which consisted almost
entirely of dialogue, and added to it so large an element of
music and song that the result seems more like an opera than
a play.

N O T E ON LIVY VII, 2, AND T H E «DRAMATIC S A T U R A '

Livy's brief review of the early history of dramatic performances

in Rome is the most detailed account which has come down to us.
Its value as history is qualified by Livy's obvious desire to prove a
thesis and the artificial nature of what seems an attempt to fit the
pieces of a jigsaw puzzle together. Livy's expressed object is to
show from what simple and comparatively innocent origins had
grown up what by his time had become a public nuisance. Like
most educated Romans, Livy despised the theatre. By his day the
stage had come to be occupied almost entirely by low farce and
seductive pantomime, performed by actors who possessed neither
legal rights nor moral reputation; the only performers, he says,
who in his day enjoyed citizen status were the actors of the traditional
Atellane farce. It may be that he or his authority had been struck
by the contrast between contemporary theatrical conditions and
the more respectable position of actors at an early period. Livy's
general view of Roman history is that it represents a moral decline;
the theatre is a particular case of this. He therefore stresses the
amateur character of the native performers (iuuenes, ' young
citizens ') in early Rome. What the relationship was between these
amateurs and the professional histriones is not clear. It was the
introduction by Livius Andronicus of Greek plays in translation
 14 THE ROMAN STAGE

which finally, according to Livy, forced the amateurs to confine

their theatrical activities to the performance of the Atellane farce.
This statement may be based on the position of Atellane actors in
Livy's own day, or in the time of his authority.
What evidence Livy had we cannot say. T h e Fescennine
dialogue, the Etruscan dance and the Atellane farce are facts,
confirmed from other sources; no doubt they influenced the
development of stage-shows at Rome.
The meaning of the term ' satura especially as applied to
drama, is one of the most debated problems in Latin literature.
I take satura as a purely Latin word, 1 connected with satis and
meaning ' full', ' stuffed As a description of a type of literary
composition or dramatic performance it means a ' m i x t u r e ' or
' medley ', whether of metres or of topics. The earliest saturae
of which we know anything, those of Ennius, were a hotch-potch
of remarks on philosophy, cookery and other topics, expressed
* sometimes in reflective soliloquy, sometimes in more dramatic
dialogue ', 3 in various metres and perhaps in prose as well. Clearly
these works were not plays. W e have no evidence as to the nature
of the saturae attributed more or less doubtfully to other dramatists.
Sometimes Satura seems to be the title of a comedy ; what it means
then we cannot tell.
The essence of the satura was freedom and lack of form.
Gradually the word came to take the sense of ' discussion of
current topics so moving quite away from the field of drama.
When Livy says that Andronicus introduced ' plays with a p l o t '
instead of the earlier saturae, he is saying little more than that
Roman dramatic shows were formless until Andronicus gave
them form.
1 The ' satyric' drama of the Greeks has in origin nothing whatever to do with

the Latin satura.

' Cf. " farce."
» Wight Duff.

situations. The supposed stage is shown on only some of the paintings. That the
costumes correspond to those of Old Comedy is unproved. The female figures are
shown as women, not as male actors masquerading as women.
 ' C H I R O N GOES U P S T A I R S '
РЫуах Vase-painting in the British Museum
 CHAPTER III

LIVIUS ANDRONICUS AND THE COMING OF

LITERARY DRAMA TO ROME

Y T H E middle of the third century B.C. we find theatrical

B activities established in various parts of Italy. In Etruria
public taste seems to have been satisfied with music and dance,
athletic, gladiatorial and equestrian performances. In Rome
the merry banter of the harvest-home had developed under
foreign influence into a professional but mainly improvised
performance consisting of song, dance and buffoonery. O n
Latium's southern frontier there had existed from early times
the Atellane farce, a kind of Punch and Judy show, which
displayed such stock types as the Clown, the Guzzler and the
Gaffer in ridiculous situations. In the Dorian settlements of
southern Italy and Sicily there were rude performances akin
to the mime, in which actors in a dress like that of Old Comedy
performed the so-called ' P h l y a x ' farces, burlesques of mytho-
logical themes, showing the gods in various undignified
situations—Zeus on a love-adventure, 6quipped with a ladder,
his path lit by Hermes, while his fair charmer gazes from a
window ; Apollo climbing the temple-roof at Delphi to escape
from Heracles, and the like ; such performances were not far
removed from mime, which was itself not easy to distinguish
from the activities of jugglers, acrobats and mountebanks of
all kinds. T h e special interest of the phlyax-farces is that, as
the vase-paintings show, the actors stood on a stage. Here we
have a contrast with classical Greek tragedy and comedy, in
which the centre of interest was the orchestra, where the chorus
danced and where, as it is now coming to be held, the actors
stood. 1
A t the beginning of the third century the mythological
burlesque (hilarotragoedia) had been given a literary form by
Rhinthon of Tarentum. W e may be sure that the Tarentines
1 But I cannot suppress a doubt as to how far the vase-paintings represent

actual stage performances. Some of the scenes depicted on them seem to violate
the conventions of ancient drama. They are essentially pictures of momentary
4
l6 THE ROMAN STAGE

were also interested in the higher forms of drama, both

tragedy and comedy. T h e Greek comic writer Alexis, himself
born at Thurii, had written a comedy entitled the Ταράντινοι.
T h e love of the Tarentines for the -theatre was proverbial,
and Pyrrhus, when called in to aid them in their struggle
against Rome, found himself (according to Plutarch) compelled
to close all places of public amusement in order to induce the
citizens to take the war seriously. This suppression of theatrical
activities (if we can believe the story) must have thrown a
certain number of actors out of employment. Plutarch tells
us that many Tarentines, unable to endure the new regime,
left their native town, and we may suppose that in this way
trained actors found their way to Rome, there to influence
such dramatic activities as the city possessed. Towards the
end of the war the Tarentines made honourable terms with the
Romans, and they contributed ships to the Roman navy in
the first Punic War.
T h e outbreak of war with Carthage brought Roman troops
back to Magna Graecia. For twenty years they campaigned
in Sicily. During that time they had an opportunity to become
familiar with Greek drama and the Greek theatre. T h e new
interest in the Greek way of life made itself felt at Rome in the
first year of peace. In the year 240 B.C. the Roman public
were entertained by the performance at the Ludi Romani of a
Greek play in Latin translation. T h e translation and the
production were the work of Livius Andronicus, a Romanized
Greek whom tradition connected with Tarentum.
O u r records concerning the founder of Latin literature and
drama are meagre and contradictory. W e cannot hope to
explain or reconcile them ; and when we have considered
what we know or surmise about the man and his work, there
may remain in our minds some sense of disproportion between
what he was and what he did.
T h e dramatist Accius, born in 170 B.C., asserted that
Andronicus had been taken prisoner by the Romans when they
captured Tarentum in 209 B.C., and that he did not produce his
first play until 197 B.C. In conformity with this chronology
is the statement of another early writer, Porcius Licinus
(c. 100 B.C.) that the Muse came to Rome in the second Punic
War. Horace tells us that it was not till after the Punic Wars
that the Romans began to apply their minds to Greek drama,
 LIVIUS ANDRONICUS 17
and Jerome gives us 187 B.C. as the date when Andronicus was
at the height of his fame. We can therefore understand the
interest aroused in Roman literary circles when an early
document was discovered which recorded the performance in
240 B.C. of a play by Andronicus. This document was seen by
Cicero, who refers to the matter no less than three times. We
can hardly doubt that Cicero was substantially in the right,
and that Accius was wrong. The archaism of Andronicus'
style would in itself prevent us from supposing that he was later
than Naevius and Plautus. Cassiodorus, too, tells us that he
produced a tragedy and a comedy at the Ludi Romani of 239.
The only other event in his career which can be dated is his
composition of the official hymn at the crisis of the Hannibalic
War in 207 B.C., when Hasdrubal was invading Italy. His
death seems to have occurred before 200 B.C., when the writing
of a State hymn was entrusted to another poet.
Apart from the evidently erroneous statement of Accius,
we have no evidence, other than general probability, to connect
Andronicus with Tarentum. It has been suggested that Accius
confused two captures of Tarentum—that of 209 and that which
presumably took place at the end of the Pyrrhic War, in
272 B.C. But we cannot be sure that Tarentum was captured
in the Pyrrhic War. Furthermore, if Andronicus was to write
the State hymn in 207, sixty-five years later, he can have been
little more than a child in 272. How then are we to suppose
that he acquired that knowledge of Greek and of Greek
literature which made him so significant a figure among the
' barbarous race of Romulus ' ? Jerome tells us (unfortunately
in connexion with an impossible date) that he was the slave
of Livius Salinator, whose children he taught, and that because
of his talents he was given his freedom. Suetonius says that
Andronicus, like that other ' half-Greek ' Ennius, gave lectures
on Greek literature and read out his own Latin compositions
to his class. There may be some truth in this ; Romans of the
upper ranks were no doubt beginning to realize the dis-
advantage of not knowing Greek, and were willing to pay
some one who would teach the language and literature to their
sons. Indeed the need for a school text may explain why
Andronicus made his translation of the Odyssey—a translation
which certainly seems to have been used in boys' schools in
the days of Horace, who retained a lively though not vindictive
ι8 THE ROMAN STAGE

memory of having been flogged through it by his teacher

Orbilius. T h a t Andronicus, a Greek in Rome, became
associated in some way with the powerful Livian family is
suggested by his name Livius, and goes well with the fact that
in the year 207, when he was chosen to write the hymn, a
Livius Salinator was consul. So far our evidence presents him
as a teacher and writer who could be called in by the authorities
to translate a Greek play for a special occasion, or to compose
a hymn. But there are three other references which show him
in a new light. Festus tells us that after the decisive victory
of the Metaurus a grateful government, wishing to do honour
to Andronicus, assigned the temple of Minerva on the Aven-
tine as a meeting-place and shrine for ' penmen and actors
scribae histrionesque, ' because Livius both wrote plays and
acted in them.' A n unknown scholiast (Glossae Salomonis)
says that he was the first both to perform and to write tragedies
and comedies. Best known of all the anecdotes, perhaps, is
the account given by his namesake, the historian Livy. Livy
tells us that Andronicus was the first to bring on the stage a play
with a plot (instead of the old saturae) ; that he rendered his
own musical compositions (suorum carminum actor) as did all
the writers of his day (id quod omnes tum яrant) ; that, as the
result of frequent encores, his voice gave w a y ; and that
accordingly he obtained the leave of his audience to employ a
boy (or a slave—Latin puer) to sing for him, while he himself
accompanied the song in dumbshow. Whatever we think
of these statements in themselves, we have no a priori right
to reject, as some of the Germans do, the implied connexion
between the writing of drama and its performance in early
Rome. As Livy himself declares, things had been different in
early times.
T h e introduction of drama at Rome raised a series of
practical problems. Some one had to make arrangements with
the magistrates, to secure or build a stage and dressing-rooni,
to maintain and train a troupe of actors, to supply them with
costumes, to hire a musician. It is not likely that a trained
actor-manager suddenly appeared to relieve Andronicus of
these burdens. More probably he had to arrange everything
himself. But whoever did it must have built on such founda-
tions as already existed. It appears from Livy's account of the
satura that there were already trained actors at Rome, who
 LIVIUS ANDRONICUS 19
were capable of giving a performance which included music,
song and clowning. The songs with a score ' written out. for
the flute-player ' must have been in some sort of metre. Here,
perhaps, we see the explanation of the great difference between
Latin plays and their Greek originals. All Greek plays,
whether tragedy or comedy, appear to have had a chorus.
The Latin comedies, at any rate, seem to have had no chorus.
What they did have were the cantica. A large part of
every Latin play was meant to be chanted or declaimed to
musical accompaniment. Those who declaimed the cantica
were the actors themselves. Thus the musical element,
which had been diminishing in New Comedy, was largely
developed by the Latin dramatists. It seems to me not quite
enough to look for the origins of the cantica in Greek tragedy,
or in some hypothetical development of opera among the
Greeks. They must have been introduced into Latin drama
because they were congenial to Roman taste. They occur
in purely native pieces such as the Atellanae, togatae and
mimes. Compared with the Greek originals, Roman plays in
general might be described in Livy's words as impletae modis
saturae, medleys in various metres. With the Greeks each
metre had traditional associations ; variety for its own sake
appealed to Roman taste.
We possess the titles of perhaps eight of Andronicus'
tragedies and of three of his comedies—all, beyond doubt,
adapted from Greek originals. We cannot tell how wide a
choice of originals was made available by the book-trade of
his day. What seems probable is that he chose tragedies from
the fifth century as well as later, comedy only from the New
Comedy which was so popular in the Greek theatres of his day.
The tragedies are Achilles, Aegisthus, Ajax the Whip-bearer,
Andromeda, Danae, The Trojan Horse, Hermione, Tereus and
perhaps Ino. The fragments of the Aegisthus show that it dealt
with the same subject as Aeschylus' Agamemnon—the return
of Agamemnon from Troy and his murder by his wife
Clytaemnestra, aided by her paramour Aegisthus ; Andronicus'
play is, however, a translation not of the Agamemnon but of
some lost original. Ajax the Whip-bearer dealt with the same
subject as the Ajax of Sophocles—the madness which came upon
Ajax when the arms of the dead Achilles were adjudged not
to him but to Odysseus. We shall find madness frequently
20 THE ROMAN STAGE

depicted in Roman tragedy and burlesqued in Roman comedy,

and the culminating horrors of Senecan melodrama were to
have their influence on the Elizabethan drama. The
Andromeda dealt with the thrilling and romantic story of the
maiden left as a prey to the sea-monster, and of her rescue
by Perseus. Melodrama and romance had gained ground
in later Greek tragedy and had a strong appeal for the
Romans. The Tereus told the ghastly tale of how Tereus,
king of Thrace, outraged Philomela, his wife's sister, and cut
out her tongue to prevent her telling of his crime, and of the
vengeance which the sisters wreaked on Tereus' son. It was
indeed a novel and exciting world to which the Roman public
were introduced by the new drama. Whatever Andronicus'
weakness in execution, we must admire the judgment which
led him to choose these tragedies as the drama and the Odyssey
as the epic which were to thrill and instruct the people of
Rome,
Of titles of comedies we have only three, the Gladiolus, the
Ludius and the Virgo or Verpus. The Gladiolus or ' Blade ' may
have portrayed one of those swashbuckling captains common
in the troubled period which followed the death of Alexander ;
the single extant fragment may be an ironic remark addressed
to the captain by some one to whom he was boasting of the
numbers he had slain : 'Were they fleas or bugs or lice ? Do
tell me.' The word ludius is used by Plautus of the professional
dancer familiar on the Roman stage ; whether the Ludius was
one of these, or a 4 Lydian', we cannot tell from the single
fragment : ' He dropped just as if he had been pole-axed.'
The title of the third comedy (' The Maid ' or ' The Circum-
cised M a n ' ) is too uncertain to afford a foundation even for
a guess as to the subject. If we may judge by the scanty
number of titles, Andronicus took little interest in comedy.
Terence, when referring to his own predecessors in comedy,
does not mention him, and his name does not occur in Volcacius
Sedigitus' list of the ten best comic dramatists. Perhaps he
was disqualified from writing comedy by his lack of colloquial
Latin ; Cicero, thinking of his style, compares his Odyssey to
some crude product of Daedalus, and says that his plays are
not worth a second reading.
Trained on Greek lines, Andronicus had to create a literary
style in a language which was not his own. No literary work,
 LIVIUS ANDRONIGUS 21
as far as we know, häd been composed in Latin before. The
language had its possibilities, but essentially it was heavier in
movement, less adaptable than Greek. Its great merit was its
gift of hitting the nail on the head. Alliteration emphasizes
this quality. We find in Andronicus' fragments some remark-
able examples :
confluges ubi conuentu campum totum inumigant,
' when across the flats in fury flow the watery floods amain
as Professor Wight Duff translates the line.
Metre was a great problem. Greek metres were based on
quantity, the time a syllable takes to utter. The typical and
primitive Latin metre, the Saturnian, seems to have been based
mainly on stress-accent. Crude as it must have appeared
when compared with the Homeric hexameter, this was the
metre which Andronicus was content to adopt for his version
of the Odyssey. In drama we find him using as the metre of
dialogue the iambic senarius, modelled on the Greek iambic
trimeter but characteristically Latin in rhythm. But we
also find in the fragments the trochaic septenarius, a
metre which, it is true, occurs in Greek tragedy and New
Comedy, but which appears to have been much more widely
used by the Latin dramatists. Indeed it had a long history in
popular Latin usage. It was the metre used by the soldiers
who marched behind Caesar in Rome :

urbani, seruate uxores ! moechum caluum adducimus.

And we find it in the Pervigilium Veneris :
eras amet qui nunquam amauit, quique amauit eras amet.
In the ' Trojan Horse ' we have a passage in cretics :
da mihi hasce opes
quas peto, quas precor ;
porrige, opitula !
All these metres, except the iambic senarius, were declaimed
to musical accompaniment; and all of them, including
the senarius, seem throughout the existence of stage-drama
to have been based on a compromise between stress-accent and
quantity.
3
22 THE ROMAN STAGE

The Romans looked back on Andronicus as a respectable

but somewhat colourless figure. His importance is as a
pioneer. He found Rome without literature or written drama.
He laid down the lines on which tragedy and comedy were to
develop for a hundred and fifty years.
 CHAPTER IV

NAEVIUS

N CN. N A E V I U S we have something more than an

I individual: we have a man. He is the earliest Italian whom
we feel we know as a human being. A greater contrast to his
predecessor, the respectable but colourless Andronicus, can
scarcely be imagined. In the whole of Latin literature we
shall hardly find a more original figure.
He must have been born not much later than 260 B.C., if
he was to take part in the First Punic War (264-241 B.C.).
Perhaps he was by birth a Campanian. Aulus Gellius records
the well-known epitaph, which he supposes to have been com-
posed by Naevius himself:
If deities immortal their tears for men might shed,
Our native Latin Muses would weep for Naevius dead ;
For since we laid our poet in Hades' treasure-store,
The true old Latin language is heard in Rome no more.
This epitaph Gellius criticizes as 1 full of Campanian' (or
' Gapuan ') ' pride Are we to understand from this remark
that Gellius thought Naevius to have been by birth a Cam-
panian, or is ' Campanian pride ' merely a general, proverbial
expression not intended to imply any connexion between
Naevius and Campania ? This second interpretation seems
rather unnatural. My feeling is that Gellius intends us to
understand that Naevius came not merely from Campania but
from the proud city of Capua itself. We have no further
reference to this matter, and we cannot tell from what source
Gellius derived his information, or whether it was correct. He
is probably mistaken in thinking that the epitaph was composed
by Naevius ; like that on Plautus which he quotes in the same
passage, it is probably the work of some admiring reader—
perhaps Varro himself, from whom Gellius derives so much of
his information about Latin literature.
One thing is clear : Naevius either spoke Latin from his
»3
ч THE ROMAN STAGE

infancy or acquired it sufficiently early to gain that mastery

of the language which is attested by the epitaph and by Cicero.
He must also have come to Rome early enough to regard
himself as a Roman and take a personal and active interest in
Rome's history and politics. We seem to be on sure ground in
affirming that he fought in the First Punic War. He said so
himself in the poem which he wrote on this w a r ; Varro
quoted his statement in his work on the Latin poets, and Gellius
in his turn (17.21) quotes Varro. In the course of his campaigns
he probably had opportunity to see something of Greek life
and the Greek theatre in the famous Greek cities of Sicily,
the main field of the tedious and bloody struggle. He may
also have brought back with him a bitter memory of harsh
military discipline and of the incompetence in high places
which sent so many splendid fleets and armies to the bottom
of the sea.
Peace brought with it the necessity to earn a living, and
the opportunity to do so. Andronicus had shown that even
in the ' barbarous' city of Rome there was a livelihood for
any one who could successfully adapt Greek plays for the
Roman stage. Naevius did not long delay. We read in
Gellius that in the year 235 B.C. two interesting events occurred
in Rome : the first Roman divorce and the production of a
play by Naevius at the public games. The action of Sp.
Carvilius Ruga in divorcing his wife simply because she had
failed to give him children was remembered as an important
event, and perhaps some priestly chronicle of events of religious
significance recorded under one year both this case and the
public games at which Naevius produced his play. In truth
the development of the theatre and the weakening of the
marriage tie were two aspects of the impact of Hellenism on
Roman life ; and in the plays of Plautus we shall presently
find marriage treated almost consistently as a subject for
mockery.
For the next thirty years or so (including the years of Rome's
death-grapple with Hannibal) Naevius seems to have given
his energies mainly to the theatre ; we have the titles of about
forty of his plays, showing an average of more than one play
each year. In the main, like nearly all Latin dramatists, he
based his work on Greek originals. But in the translations
of Naevius there was something new. Terence speaks of the
 NAEVIUS 25
' carelessness ' (neglegentia) of Naevius, Plautus and £nnius as
something more worthy of admiration than the dull pedantry
of their successors. Evidently Terence found that the plays of
these popular dramatists were somehow different from the
Greek which they professed and perhaps believed that they
were translating. Naevius, Plautus and Ennius may have
tried, in a general way, to render their originals faithfully ;
but their native genius and their instinctive knowledge of
popular taste compelled them to infuse a Roman flavour into
everything they wrote.
It is not always safe to judge a dramatist's outlook from
isolated passages in his plays, spoken we know not by whom or
in what circumstances. Nevertheless there is in the dramatic
fragments of Naevius a recurring note of independence, a love
of freedom and of free speech, which seems typical of the author.
' We will speak with the tongues of freedom at the feast of
Freedom's god.' ' I have always prized Freedom far above
wealth.' ' I hate folk who mutter : say plainly what you mean.'
The tragedy and comedy of the Roman republic were based
almost entirely on Greek originals ; yet for a man like Naevius
it would have been difficult to content himself with mere
translation. Somehow or other he would find a vent for his
native genius, his pride in Rome and his interest in contem-
porary life. Tragedy offered less opportunity than comedy;
and we know of only seven tragedies by him : Andromache (?),
Danae, Equos Troianus, Hector Proficiscens, Hesiona, Iphigenia and
Lycurgus. We notice the recurrence of two of Andronicus'
titles, the Danae and the Equos Troianus ; it would seem that
Naevius set himself the task of surpassing his predecessor on
his own ground. The Hector Proficiscens, translated from an
unknown Greek play, dealt with the subject so pathetically
treated by Homer, Hector's last leave-taking of his loved ones
before he goes to fight his fatal duel with Achilles. A line from
his farewell to his father Priam was particularly admired by
Cicero :

laetus sum laudari me abs te, pater, a laudato uiro.

How different is such a Hector from the Hector of Homer !
Homer gives us a human being, Naevius' line suggests a formal,
pedantic kind of hero, a Roman consul, very conscious of his
robes of office, addressing in set phrase his father, himself an
26 THE ROMAN STAGE

ex-consul. Listen again to Naevius' Lycurgus addressing his

bodyguard :
uos qui regalis corporis custodias
agitatis, ite actutum in frundiferos locos,
ingenio arbusta ubi nata sunt, non obsita.
In trying to achieve dignity, the Roman writers of tragedy
sacrifice simplicity. They write as if, like Aeschylus and
Euripides at the competition in Aristophanes' Frogs, they
meant to be judged by the weight of their lines.
Comedy offered more scope to Naevius' original talent.
Not that he composed original comedies ; indeed many of
the surviving titles are Greek, the others are in all probability
translated from the Greek, and the fragments bring before us
the characters and situations of Greek New Comedy. Where
Naevius' individuality appears is in his choice of plays to
translate, the spirit which he infused into his translations and
the unmistakable references to native Italian topics. The
purpose of comedy, at least for the Romans ofthat day, was to
arouse laughter. Ifjests, puns and topical allusions could serve
this purpose, then they were to be admitted regardless of any
incongruity with the context—in fact this very incongruity
might make them more amusing. The titles of Naevius'
comedies—we know more than thirty of them 1 —' The
Charcoal Burners ', ' The Potter ', ' The Soothsayer ', ' The
Races', 4 The Night-hawks' and so on, suggest themes from
common life, indeed from low life. It would not be difficult
to insert into such plays references to everyday matters in
Rome and its neighbour towns. There is an unmistakable
Italian flavour in a fragment of -conversation from ' The
Soothsayer ' : ' Who dined with you yesterday ? ' ' Guests
from Praeneste and Lanuvium.' ' I hope you gave each
party its favourite food—empty sow's paunch boiled for the
one, nuts in abundance for the other.' 2 In the Tunicularia
(the tunic was the wear of humble folk in Rome, Horace's
tunicatus popellus) some one says ' You would rob even Theo-
1
Acontizomenos, Agitatoria, Agrypnuntes, Apella, Ariolus, Astiologa, Carbonaria,
Clamidaria, Colax, Commotria, Corollaria, Dementes, Demetrius, Dolus, Figulus, Glaucoma,
Gymnasticus, Lampadio, Nagido, Nautae (?), Neruolaria, Paelex, Personata, Proiectus,
Quadrigemini, Stalagmus, Stigmatias, Tarentilla, Technicus, Testicularia, Tribacelus,
Triphallus, Tunicularia.
* Similarly Plautus inserts a description of Rome into the Curculio.
 NAEVIUS 87
dotus, who seated himself on an altar in a strong-room at the
Compitalia and wrapped himself round with mats in order to
paint with his cheap ox-hide brush a picture of the Lares at
play ! ' Here the language, at any rate, is very Roman.
We have about a dozen references in Roman writers to
a comedy of Naevius called Tarentilla, a title which probably
means ' The Girl of Tarentum.' Though Tarentum is in Italy,
the play was no doubt translated from the Greek ; Tarentum
was a Greek colony, and mention has already been made of a
play by the Greek writer Alexis entitled Ταράντινοι which
described the life of that gay city, much addicted to dining,
love-making and philosophy. But no Roman who saw
Naevius' play would have forgotten that for many years now
Roman soldiers had been campaigning among the Greek towns
of southern Italy and Sicily and had come to know there a
way of life more elegant and more dissipated than anything
that Rome had to show. Many of them must have beguiled
their off-duty hours in the company of girls like the heroine of
Naevius' p l a y 1 : ' Like a ball among a ring of players, she puts
herself at the disposal of all; to one she nods, to one she winks,
one she fondles, one she hugs, to one she gives her hand to
clasp, another's foot she presses with her own, she sings a duet
with one while signalling a message to yet another.' To judge
by the fragments, the play dealt with the adventures of two
young men from the country who squander their savings in
the luxurious town. They arrange a drinking-party ; one of
them gets drunk and has to be supported by his mistress.
Presently their angry fathers track them down. A stormy
scene follows ; finally the young men receive solemn advice
to leave the town and settle down to a sober married life on
their farms.
In other comedies we find the usual characters of New
Comedy. The Colax had not only the title-role, a ' flatterer'
or parasite, but also a braggart captain, miles gloriosus ; the
play was translated from Menander, and Terence gives us
the surprising and puzzling information that the translation
was the work of ' Plautus and Naevius' (or perhaps Plautus
or Naevius). Did the two dramatists collaborate, or did
Plautus re-write the play for a revival performance ? The
1
If these lines belong to this play, or indeed to Naevius at all. Isidore attributes
them to Ennius.
28 THE ROMAN STAGE

Agitatona (literally ' Charioteer Play ') told of a chariot race ;

some one (perhaps his father) says to Demea, ' Lest you should
say that I am thwarting you, I'll put them (the horses ?) at
your disposal for this one day ; afterwards I'll sell them while
they are actually running—unless you win.' We have another
scrap of conversation about the result: ' What ? have we
really won ? ' ' You have won.' ' I'm delighted ! * ' I'll tell
you how it happened.' We overhear a quarrel—perhaps
between father and son : ' You seem to object on purpose to
everything that I want, and to desire what I don't want.'
The Acontizomenos (' Struck by a javelin') had a prologue
which recommended it as a ' first-rate p l a y ' ; it seems to
have told of some one who was falsely accused of killing his
twin brother, and we also hear of a midnight murder. The
Agrypnuntes, ' Those who keep awake at night' (presumably
for mischief) dealt with a band of brigands who infested the
streets. One line seems to claim that the poet has shown up
these ' night-thieves ' on the stage. Was the Apella a portrait
of a Jew ? We remember Horace's Iudaeus Apella ; the name
is perhaps a pun on a and pellis, ' circumcised', and one of
Andronicus' comedies, according to Ribbeck, was entitled
Verptis, ' circumcised Ч1 All that survives of the Apella is two
remarks concerning onions. Elsewhere in the fragments we
have references to the stock themes of Roman comedy—the
power of love, the whipping of slaves, drunkenness, debauchery.
Young men run after courtesans ; they promise them all their
possessions ; one youth crudely wishes that the gods would
put his parents out of the way, no doubt that he may bestow
their property on his darling. 'References to banquets, para-
sites, exposed infants, twins, even quadruplets (no doubt all
alike) and titles like 1 The Madmen ', ' The Trick ', ' Dust in
his Eyes' (Glaucoma), suggest situations familiar to the reader
of Plautus. That there was a spice of indecency is suggested
by titles like Testicularia and Triphallus, and by some of the
fragments. The title Personata has been taken to mean ' The
Masked Play' (implying that in other plays masks were not
used), but it may only mean ' The Lady in the Mask
implying, perhaps, that the heroine wears a special mask as
a disguise for at least part of the play. It seems probable,
on the whole, that all actors on the Roman stage, except the
1
Another suggestion is that Apella means the ' Apulian Woman \
 NAEVIUS 39
mimes, normally wore masks ; there is nothing to warrant
our attributing to Naevius the first literary fabula Atellana.
Nor need we ascribe to him, because of his real or supposed
references to Italian topics, the invention of the comedy on
native Italian themes, fabula togata.
What he did, apparently, invent was the Roman historical
play, fabula praetexta or praetextata. The toga praetexta was the
purple-bordered toga worn by Roman magistrates; the
fabula praetexta dealt with the deeds of men who might have
worn such a toga, the heroes of Roman history or legend.
And so when Varro quotes from the Clastidium of Naevius an
iambic line referring to some one's triumphant return to his
native land, we may be pretty sure that this work was a play
dealing with the famous victory won at Clastidium in 222 B.c.
by the consul M. Claudius Marcellus. We also hear of a play
by Naevius entitled Romulus, perhaps the same as his Lupus,
' The Wolf', from which latter we have a fragment of a con-
versation between Vibe, king of Veii, and Amulius, the
usurping king of Alba Longa whom Romulus is destined to
slay. It is sometimes said that this or another play dealt with
the infancy of Romulus as well as with his triumphant achieve-
ments as a man. If that were the case, then indeed the
fabula praetexta was completely free from the ' Unity of Time '
which limited most classical plays to a day or little more
(two days in the case of the Heauton of Terence). The only
support for such a belief is a remark of Donatus (in reference
to the proverbial expression lupus in fabula, our ' Talk of the
Devil!') that there was no truth in the story that a wolf
appeared during the performance of a play by Naevius, in
the scene of the suckling of Romulus and Remus. I find it
equally difficult to suppose either that the Roman stage was
ever capable of showing human twins being suckled by a she-
wolf, or that Naevius would have thought it worth while to
stage the suckling of the twins without the presence of the
wolf. Apparently there was a play by Naevius dealing with
Romulus (and that we knew already) ; and some one had
the absurd notion of trying to explain lupus in fabula (which
really means ' the wolf in the story') by imagining that the
suckling of the twins by the wolf was actually shown in this
play.
Naevius' interest in Roman history and contemporary
 THE ROMAN STAGE

affairs led to the composition of another work which was not

a play, his Bellum Punicum, or narrative in Saturnian verse of
the first Punic War. Cicero makes Gato say that this work
gave the poet great pleasure in his old age—from which re-
mark it has been inferred that the poem was written during
Naevius' exile in Utica, when circumstances had cut him off
from his familiar literary medium, the stage. I find this very
difficult to believe. Possibly all that Cicero means is that
Naevius derived pleasure in his old age from re-reading his
poem ; and if this is his meaning, it is hard to see what
evidence he had for such a statement. We have other and
perhaps more reliable evidence that Naevius' later years were
clouded in misfortune.
According to a belief current in imperial times, Naevius'
jibes at the Roman aristocracy got him into trouble. A
commentator of the fifth century A.D. tells us that Naevius
lampooned the consular family of the Metelli in the line :
fato Metelli Romae fiunt consules,
4
The Metelli are made consuls at Rome by F a t e ' (and not
through any merit of their own). The ancient law of the Twelve
Tables prescribed capital punishment for slander (si quis
occentauisset siue carmen condidisset quod infamiam faceret fiagitiumue
alten) ; and an allusion to politics might be rewarded with
a flogging (cautum est ut fustibus feriretur qui publice inuehebatur).
So the Metelli (one of whom was consul in 206 B.C.) were
•athin their rights when they replied in the famous Saturnian :

dabunt malum Metelli Naeuio poetae,

' The Metelli will give a whipping to Naevius the poet.' This
story may be apocryphal; but Gellius, too, quotes lines of
Naevius which he thinks were in all probability a reference
to a scandalous episode in the life of the great Scipio : ' even
he who has gloriously achieved such mighty victories with his
strong arm, whose deeds thrill the world, who is the foremost
man in the eyes of all the nations—even he has known what
it is to be hauled away from his lady-love by his father, with
nothing on but his cloak.' (This passage may in fact have
referred not to Scipio but to some fictitious character, some
braggart warrior in a play.) A character of Plautus (M.G.
210-1) remarks ' I have heard that a foreign poet has his
 NAEVIUS 31
face fastened to a column, with two gaolers lying on him all
day ' (i.e. the two chains which bind him), and Festus tells us
that Plautus used the expression ' foreign p o e t ' {poeta barbarus)
of Naevius. Gellius states that Naevius was thrown into
prison for his incessant attacks on the nobles, and that in prison
he wrote two plays, the Hariolus and the Leo, in which he begged
pardon for his ρ cist offences, with the result that he was released
by the tribunes. Jerome, however, tells us that he was driven
from Rome by Metellus and other nobles, and died at Utica
in 201. These accounts are not easy to reconcile ; and if
Naevius' imprisonment was as rigorous as Plautus appears to
indicate, it is hard to see how he could have written plays in
gaol. Cicero mentions neither the imprisonment nor the
banishment, but tells us that an ' early commentary' gave
204 B.C. as the date of the poet's death—A date which Varro
thought too early. Evidently the date of Naevius' death was
a matter of dispute even so early as Cicero's day.
O n general grounds, the lines attributed in these anecdotes
to Naevius seem to me genuine ; as for the reply of the Metelli,
I have some suspicion of a Saturnian which is so easy to scan.
But even if we reject all the stories as mere gossip, it remains
true that this was the sort of gossip which attached itself to
Naevius. The best proof that his liberty of speech incurred
some signal punishment is the care with which subsequent
Latin dramatists resisted the temptation to allude to politics
or mention contemporaries by name. Naevius had taken the
first steps towards creating a new drama which would deal
with the vital interests of the Roman people ; but the heavy
hand of the State was to confine Latin playwrights to frivolous
or foreign themes.

THE FABULA PRAETEXTA AFTER NAEVIUS

In Naevius' two praetextae, the Romulus and the Clastidium,

we see already established the division of the praetextae into
(α) plays dealing with remote Roman history or legend, and
(b) plays celebrating victories won by Roman generals still
living or only recently deceased. Examples of (a) are Ennius'
' The Rape of the Sabine Woman ' (if that work was indeed
a play), Accius' Aeneadae or Decius, which dealt with the victory
of Sentinum in 295, achieved by the self-immolation of the
32 THE ROMAN STAGE

consul P. Decius Mus, and Aerius' Brutus, which told of how

the tyrant Tarquin had been expelled by Rome's first consul.
Examples of (b) are Ennius' Ambracia, which must have dealt
with the capture of Ambracia in 189 by Ennius' patron, the
consul M. Fulvius Nobilior (but it is not clear that this work
was a play), and the Paulus of Pacuvius, if we see in this work
a praetexta on the victory won at Pydna in 168 by the consul
L. Aemilius Paulus. This short list includes all the genuine
praetextae which we know to have been produced in Rome.
Contemporary history must have been a delicate matter
to handle on the stage. It seems to have been a rule of the
Roman theatre (at le?ot after the time of Naevius) that contem-
porary personages should not be mentioned by name. Cicero
{Rep. IV) says that in early Rome it was forbidden to blame
or praise any living Roman on the stage. We know, however,
that funeral games were held in honour of L. Aemilius Paulus
in 160, and that they included the performance of plays ; it
is conceivable that the Paulus was a praetexta specially written
for this occasion. The play dealt with a battle, and there was
a reference to a mountain track—presumably the track used
by the force under Scipio Nasica, which decided the issue at
Pydna. I find it more difficult to suppose that such a
praetexta could be performed at the triumphal games given by
the victorious general. It would seem, at any rate, that
praetextae dealing with contemporary history were written
for some special occasion, and presumably at short notice. To
make a play out of the events of a recent campaign must have
been a fairly difficult task. The meagreness of the extant
fragments suggests that such plays were short as well as few.
By way of contrast we must mention the praetexta written
by L. Cornelius Balbus on his own achievements in 49 B.C.,
and staged by him at the games which he gave at Gades in 43.
This must be regarded as merely one of the outrageous breaches
of decorum of which he was guilty (see Cie. ad Fam. x. 32).
Roman myth or remote history presented a more attractive
field for the dramatist. Some lines are preserved of Aerius'
Aeneadae or Decius which evidently dealt with the battle and
the ceremony of deuotio. From the Brutus Cicero has quoted
Tarquin's account of his dream, with the interpreter's reply,
which ends with an impressive reference to the future greatness
of Rome. Like Shakespeare's plays on English history, the
 NAEVIUS 33
praetexta found opportunities to appeal to patriotic sentiment,
that refuge for dramatists as well as for scoundrels. T w o other
lines of this play deal with the etymology of the word consul
and with the achievements of Tullius (i.e. King Servius
Tullius). This latter line, according to Cicero, was taken by
the audience on one occasion as a flattering reference to the
achievements of another Tullius (namely himself) in 63 B.C.
The Romans, including Cicero, had a craving for reading such
topical allusions into old plays ; this fact itself suggests that
direct reference to current events and living persons was dis-
couraged. The Brutus may have been intended to gratify
Accius' patron, the consul Junius Brutus, who celebrated his
Spanish victories with a splendid triumph in 136 B.C.
This seems to be all that can with probability be said about
the praetextae produced on the Roman stage. It seems
unlikely, indeed, that any other praetextae were ever composed
for performance. The late grammarian Diomedes, when
citing examples of this form of drama, mentions only the
Marcellus (i.e. the Clastidium), Decius and Brutus. The pro-
logue to Plautus' Amphitruo (lines 41-4) tells us that mention
had recently been made ' in tragedies' of the service conferred
on the State by Neptune, Virtue, Victory, Mars and Bellona ;
but the normal sense of tragoedia makes it fairly certain that
the reference here is to Roman adaptations of Greek tragedies.
We must conclude that the introduction of the historical
play by Naevius proved comparatively sterile. This form of
Latin drama has assumed undue importance in our eyes
because of the Roman grammarians' love of classification and
our own wish to find in Latin drama something corresponding
to Shakespeare's historical plays. Generally speaking, it seems
to have been beyond the power of the Roman dramatists to
construct a plot quite independently of Greek models. More-
over the mention of national history, even of remote history,
on the stage was rendered embarrassing by the sensitiveness of
both people and government on this subject. T h e comedy
of native manners (fabula togata) found its material in the
ordinary life of humble folk and of the country town, and its
plots may have owed something to New Comedy. In this way
it achieved sufficient success to supply full occupation for the
three writers whose names we know. The praetexta was never
more than a parergon for a few writers of tragedy, and its
34 THE ROMAN STAGE

remains do not extend beyond five or six certain titles and about
fifty lines. As a purely literary composition it maintained its
existence ; L. Cornelius Balbus circulated his manuscript
among his friends, who read it with malicious enjoyment, and
plays on remote or recent Roman history continued to be
written under the Empire—but not for the stage. Our only
extant example is the anonymous Octauia, written at some time
after Nero's death and having as its heroine his unfortunate
wife. This play seems to owe nothing to the genuine praetextae
of Republican times ; it is a mere imitation of Greek tragedy,
except that the subject is Roman. The author shows no more
regard than Seneca for the necessities of stage-production ;
he is clearly writing for private reading or the recitatio.
 C H A P T E R V

P L A U T U S : L I F E AND L I S T OF PLAYS

U R H I S T O R I E S of Latin literature present us with

O circumstantial details concerning the life of Plautus. We
are told that ' Titus Maccius Plautus' was born at Sarsina, in
Umbria, about the year 254 b.c., that at Rome he made some
money as a craftsman in the service of the theatre (which
perhaps means that he began as an actor), that he then
engaged in trade, lost his savings and was forced to hire himself
out as a worker in a mill, and that in such leisure moments as
were afforded by this occupation he wrote some plays, which
presumably were sufficiently successful to lead him to take up
the profession of playwright in earnest. His death took place
in 184. While there is nothing wildly improbable about this
story, critical examination shows that the Romans themselves
were by no means agreed as to the writer's name, his date and
the authenticity of his supposed works. Beyond question there
was a comic dramatist named Plautus, a highly successful
writer of the generation preceding that of Terence. At his
death he, like other dramatists, left his plays behind him in
manuscript form, some of them perhaps bearing on the title-
page the author's name in the genitive case, e.g. p l a v t i
c a s i n a , ' The Casina, by Plautus '. A few details of first per-
formances seem also to have survived, whether included in
the magisterial records of ludi publici or jotted down on the
manuscripts. In that unscholarly age these records soon fell
into confusion. The fame of Plautus as a popular entertainer
seems to have induced unscrupulous producers to pass off the
works of other dramatists as his. His own practice (if Gellius
is correct) of working over the plays of earlier writers may have
been partly responsible for the confusion. In the course of
time the total number of works attributed to him rose to one
hundred and thirty. Roman scholars strove to distinguish
the true from the false, but their methods were subjective and
their results were conflicting. Once they came to distrust the
35
зб THE ROMAN STAGE

evidence of authorship given on the title-page of the manu-

script, or by tradition, they had no criterion but their individual
sense of style for distinguishing the genuine plays of Plautus
(whether attributed by the manuscripts to him or to other
writers) from the plays of other writers which were falsely
attributed to him. Finally, in the time of Cicero, the scholar
Varro made a determined effort to clear up the confusion. He
appears to have drawn up three lists : first, plays which were
universally admitted to be Plautine ; second, other plays which
he himself, on grounds of style, believed to be Plautine ;
third, plays which, though attributed by some to Plautue, were
in his opinion the work of other writers. I n the first list, the
so-called Varronianae fabulae, were included twenty-one plays.
Varro's methods were indeed uncritical, but his authority
was great. Though the controversy continued for centuries
after his death, it cannot be a mere coincidence that our manu-
scripts of Plautus contain precisely twenty-one plays (of which
the last is a mere fragment). Evidently later editors saved
themselves trouble by accepting the Varronianae fabulae, and
only these. Our manuscripts therefore give us only those
plays which were, according to Varro, universally accepted in
his day. Varro himself did not claim that this list included
all the works of Plautus ; as we have seen, he put in his second
list those plays which, on grounds of style, he himself regarded
as Plautine, though they were questioned by at least some
other scholars. As for his third list, the plays attributed to
Plautus which, on grounds of style, he felt unable to accept,
he seems absurdly to have invented a dramatist named
' Plautius ' who had written at least some of them. Evidently,
Varro had in his mind such titles or references as Plauti Casina ;
as the genitive case o f ' Plautius ' would be identical with that
o f ' P l a u t u s t h e plays of Plautius would, he argued, in time
come to be attributed to Plautus.
It seems clear that, apart from the Varronianae fabulae,
there were genuine plays of Plautus which have not come down
to us. Terence can hardly have been mistaken in attributing to
Plautus the Commorientes ; yet this play does not appear in our
manuscripts. Terence also says, or appears to say, that there
was a play by ' Naevius and (? or) Plautus' called the Colax.
If we can believe that these two authors collaborated, or that
(as Gellius asserts) Plautus retouched the plays of earlier
 PLAUTUS : LIFE A N D LIST OF P L A Y S 37

writers, we can well understand that confusion as to authorship

had set in within a generation of Plautus' death. But if con-
fusion had begun thus early, can we even feel certain that the
twenty-one Varronianae fabulae are all from the hand of Plautus ?
T h e prologue to the Asinaria gives the author's name (in the
nominative) as ' M a c c u s ' ; the prologue to the Mercator
gives it (apparently) as Maccus Titus. (These two plays are
remarkably lacking in the metrical variety so characteristic of
Plautus.) The other prologues which mention the author's
name give it as ' Plautus'. The oldest manuscript (fifth
century A.D.) in one place gives the author's name (in the
genitive) as T . Macci Plauti. It would seem that the Romans
came to identify ' M a c c u s ' Maccus Titus ' and e Plautus '
by supposing that the author's full name was Titus Масс (i) us
Plautus ; but this identification was apparently denied (the
passage is very obscure) by the dramatist and literary historian
Accius, 1 who wrote more than half a century before the time
of Varro, More than one scholar of the present day (e.g.
Westaway and Norwood) has found the Mercator different in
style from the other plays. We must either suppose that
Plautus bore different names in his lifetime, or that two or
more individuals were confused by later generations. Whether
' Maccus ' and ' Plautus ' denoted the same person or not, it
is curious that both look like stage-names ; ' Maccus ' was the
clown in the Atellane plays, and ' Plautus' is said by Festus
to mean plants pedibus, ' Flat-foot', which reminds us of the
Latin name for the bare-footed mime, ' Flat-foot' (planipes).
O n the whole, it seems reasonable to believe that Varro's
list is substantially correct, though we must reserve judgment
as to the Mercator and the Asinaria. As for Plautus' date,
Jerome tells us that he died in 200, Cicero says that he died in
184, and Gellius speaks of him as a contemporary of Cato
(234-149). Fortunately there is good evidence to show that
Cicero's date is not far from the truth. There seems to be a
reference in the Miles Gloriosus to the imprisonment of Naevius.
The prologue to the Cistellaria speaks of the (second) Punic
war as still going on. For two plays the didascaliae, or records
of the first performance, have been preserved in our oldest

1 Sec Warmington, Fragments of Old Latin II, 587. Accius distinguishes

between Plautus and Titus Maccus, and says that the latter was not the author of
the Commorienies.
4
38 THE ROMAN STAGE

manuscript; these give the date of the Stichus as 200 and that of
the Pseudolus as 191. Cicero's date for Plautus' death may have
been inferred from the silence of the records as to new plays by
Plautus after that year. There would naturally be no corre-
sponding evidence as to the date of his birth. Cicero makes Cato
speak of Plautus in his old age as taking special pleasure in
two works, the Truculentus and the Pseudolus. This passage is
commonly understood to mean that these plays were written
when Plautus was an old man, and it may conceivably be based
on lost passages in Plautus' plays similar to the extant reference
in the Bacchides to the Epidicus as ' a play which I love as I do
my very self'. If the Pseudolus, produced (as we have seen)
in 191, was written when Plautus was an old man (i.e. sixty
or more), it follows that Plautus was born before 250. However
doubtful the argument, the conclusion is reasonable; it
allows us to regard Plautus as a junior contemporary of Naevius,
by whose original genius he seems to have been influenced and
from whose melancholy fate he probably took warning.
The biographical details given by late writers may be
nothing more than illegimate inferences from passages in
extant or lost plays. Legends are apt to gather about the early
careers of famous men, particularly if there is no reliable
information available. For example, the statement that
Plautus was born at Sarsina may be a mere inference from his
mention of the town in his Mostellaria. When we come to the
life of Terence we shall see how such biographies were con-
cocted. Gellius' statement that Plautus, when a mill-hand,
wrote certain plays is rendered rather doubtful by the fact
that the plays mentioned are not in our manuscripts, and were
therefore not among those universally accepted as Plautine in
Varro's day. Nevertheless it is not unlikely that Plautus
acquired a practical knowledge of the theatre at an early age,
that he knew poverty and hardship, and that he depended for
his livelihood on the success of his plays with the common folk
of Rome. If we hear of no patron, we may at least surmise
that he was shrewd enough to avoid offending the great. In
fact, he turns away from political affairs with the remark that
such matters should be left to the rulers of the state. If
Puritans murmured (as they well might do) at the tone of his
works or that of the new drama in general, he was ready to
pose as a champion of public morals—that is if we regard a*
 SCENE FROM COMEDY
Relief from Pompeii in Naples Museum
 PLAUTUS : LIFE AND LIST OF PLAYS 39

genuine the prologue and epilogue to the Captiui. With the

possible exception of the Mercator, the extant plays seem
sufficiently homogeneous in style (admittedly a fallible guide)
to confirm the general reliability of Varro's list and reveal
their author as essentially a man of the theatre, who troubled
himself little with politics, philosophy or literary art but made
it his business to win the crowds by plays which appealed to
their sense of fun, their interest in intrigue and their pleasure
in music, rhythm and rhetoric.
The twenty-one extant plays are the Amphitruo, Asinaria,
Aulularia, Bacchides, Captiui, Casina, Cistellaria, Cttrculio, Epidicus,
Menaechmi, Mercator, Miles Gloriosus, Mostellaria, Persa, Poenulus,
Pseudolus, Rudens, Stichus, Trimmmus, Truculentus, Vidularia.
Accidents to the manuscripts in the Middle Ages have reduced
the Vidularia to a mere fragment, and have seriously damaged
some of the other plays. Nevertheless we may reasonably
believe that we possess a larger body of drama from the hand
of Plautus than has been left to us by any other ancient
dramatist except Euripides.
All the plays of Plautus are palliatae, or adaptations from
the Greek, and presumably from Greek New Comedy. Our
only other complete examples of the palliata, the six plays of
Terence, are vastly different from the plays of Plautus. It seems
appropriate at this point to consider the character of New
Comedy itself, for which we have independent evidence in
the fragments of Menander's plays discovered in recent years,
as well as in a large collection of single lines or passages quoted
by later authors from Menander and his fellow-dramatists.
 CHAPTER VI

G R E E K NEW COMEDY

HE OLD Comedy of Aristophanes and his contemporaries

T dealt in outspoken terms with the political and social life
of Athens in the closing decades of the fifth century. That
war ended in a defeat for Athens which destroyed not only
her empire but much of the freedom of speech and the national
spirit which had made Old Comedy possible. Aristophanes
himself showed his recognition of the new order in his two
fourth-century plays, the Ecclesiazusae and the Plutus, which deal,
not with current politics, but with broad problems concerning
humanity at large. Why are public affairs so plainly mis-
managed ? They have always been in the hands of men ;
then give them over to the women, and see what happens.
Why is wealth so ill-distributed ? Because Plutus, the god of
wealth, is blind ; then give him Ms sight, and see what happens.
In addition to changes in theme and tone, the form of these
two fourth-century comedies (our only examples of what some
grammarians call Middle Comedy) is different from that of
their predecessors. The chorus, which had formed the original
element in drama and had decided the structure of Old
Comedy, is now, it seems, felt to be an anachronism. Its part
is shrinking, and occasionally its entrance is denoted by nothing
more than the stage-direction XOPOY, ' a performance by the
chorus \ Half a century later it seems to be entirely restricted
to occasional appearances which have little bearing on the
plot.
The final destruction of Greek independence by Macedon
in 338 B.C. was reflected in the tone of New Comedy. Finding
their city states reduced to political insignificance, men turned
their eyes to themselves and their own affairs. As private life
is more or less the same in every land, the New Comedy is
cosmopolitan in tone. The relations between husbands and
wives, between parents and children, between masters and
slaves; the adventures and misfortunes which affect private
40
 GREEK NEW COMEDY 41

life in a troubled epoch—piracy and kidnapping, the exposure

of unwanted children, the separation of relatives and their
subsequent re-union—these are among the themes of New
Comedy.
Every play must have an ending. The typical ending for
tragedy is the death of the hero. Comedy must end happily ;
so a fundamental question for the dramatists of New Comedy
was to find in ordinary life a happy ending. A captured
son may be restored to his father, a rascal may be outwitted,
the hero of the play may come into possession of wealth ;
but no other theme in private life can compare with love.
The difficulties of the lovers form the plot, their union marks
its conclusion. Old Comedy had often contained some sort of
marriage as part of the general revelry at the end ; more
relevant perhaps to New Comedy is the reunion of husband and
wife at the end of a romantic play such as Euripides' Helen.
But love 1 had not been a dominant element in fifth-century
drama ; it was now to exercise a decisive influence on comedy.
No play of Menander, according to Ovid, was without a love
interest. It was dramatic necessity which drove playwrights
to build their plays round the theme of love ending in union.
This theme seems so natural to us that we do not immediately
perceive that it was not so natural to the Greeks. With us,
marriage is founded on love ; with the Athenians things were
quite different. Athenian convention confined the daughters
of respectable families to their homes, where they met no men
except close relatives ; their marriages were arranged for them ;
they never saw their husbands until the wedding-day.2
Consequently, if a young gentleman is to fall in love at all, it
must usually be with a girl of inferior position—the daughter
of humble folk, a foreigner, a slave, a courtesan ; not being of
citizen class she cannot become his wife. The dramatists resort
to all sort of devices to get over this fundamental difficulty and
enable the love-plot to end in marriage. The heroine may
have been exposed at birth and picked up by strangers ; she
may have been kidnapped in infancy ; in either case she may
come into the hands of unscrupulous persons who bring her

1
By «pur' l o v e t h e Greeks usually meant sexual desire. It might be trivial or
serious, but there was nothi ig necessarilyadmirable in it.
1
This is the dramatic convention. In real life women had, or took, rather
more freedom.
42 THE ROMAN STAGE

up for the courtesan's profession, and thus she may make the
acquaintance of the hero and win his love. If such an affair is
to end in marriage, two points have to be established : we
must be assured that the heroine has had no dealings with any
man except the hero, and her true parentage and citizen
status must be revealed during the course of the play. Moral
standards are lax for men, strict for women ; it is generally
taken for granted that young men will amuse themselves with
courtesans, though such affairs should end when the young man
marries and settles down; but no woman of respectable
family is ever represented as unchaste. A modest girl may,
however, be the victim of misfortune ; if reduced to poverty,
she may accept the protection of a young man though legally
her lack of citizen status disqualifies her from becoming his
wife. Such a girl may display the most tender and single-
hearted affection towards her protector ; should a happy turn
of events prove that she is really his social equal by birth,
there will be no difficulty in the way of their marriage.
Maidens living respectably with their parents are allowed to
attend religious festivals, even at night; at such festivals casual
encounters may occur, perhaps with drunken youths, and
through no fault of her own a girl may find herself involved
in scandal. However, it always happens that the guilty
man turns up again ; his guilt and failure to acknowledge
it earlier are explained by the fact that he was intoxicated at
the time and that the darkness prevented both parties from
seeing each other clearly ; discovery of the truth leads to
repentance and marriage. By a further refinement of invention,
the girl who has been wronged in one of these encounters may
subsequently be married to her assailant without either of them
realizing the identity of the other; partial disclosure of the
facts may lead to separation, but knowledge of the whole truth
will bring reconciliation between husband and wife.
Modest girls and young wives were by Athenian convention
confined to their own homes ; all ancient drama has an out-of-
doors setting; consequently, we can hope to see but little of the
bride, still less to find in her a vivid personality. The virtues
of a wife are tenderness and fidelity ; wit, charm, talent in
music or dancing are the marks of a courtesan. The more
individuality a girl appears to possess, the less likely is she to
marry the hero. In these circumstances we must not hope to
 GREEK NEW COMEDY 43
find in New Comedy a love interest as we understand the term.
The hero's passion sets in train a series of events, and it is these
events which form the plot of the play. The heroine's part is
passive (especially if she is not a courtesan). The lovers never
meet without the presence of some third party on the stage.
With love itself, with the happiness of the lovers in each other's
society, the dramatists were perhaps little concerned ; how
little is illustrated by the Rudens, a play which is so contrived
that the parts of the hero and the heroine can both be taken
by the same actor. Indeed the fact that female parts were
taken by male actors must have deprived the Greek audience
of at least part of the interest which love scenes hold for us.
The Roman attitude to love was not the same as our own ;
nevertheless it was different from that of the Greeks. Love
between the sexes plays a comparatively small part in Greek
literature : it is the chief theme of the Roman elegiac poets,
and the love of Dido for Aeneas forms the most poignant
episode in Virgil's Aeneid. Not only did woman enjoy a
higher status in Rome than in Athens, but the Romans took
perhaps a more vigorous interest than the Greeks of Menander's
day in sex. Here we have the germ of possible alterations in
the Latin comedies, as compared with their Greek originals;
that such alterations were made I will try to show in the case
of Terence.
The hero of New Comedy—to judge by our Latin versions
—is usually either a good-natured rake, or sentimental and
lacking in initiative ; he is the son of a well-to-do father, and
depends for funds on his father's generosity or his slave's
ingenuity. Fathers are, however, usually niggardly, and with
good reason; with equally good reason wives, even the wives
of elderly husbands, are suspicious and shrewish. Children—
apart from an occasional slave-boy—do not appear. 1 The fathers
of Roman families must have looked askance on a picture of
family life so different from the Roman tradition. The heroes
of New Comedy, their relatives and their friends, belong to the
comparatively well-off classes ; they travel abroad on business,
they possess farms, they deal in land and cattle. Other
characters are marked off as belonging to particular callings,
each with its recognized mannerisms. We must be prepared
1 A baby in arms is brought on the stage in the EpitreponUs : cf. lines 85-88,

649-652.
44 THE ROMAN STAGE

to find the finer touches of characterization coarsened in the

Latin versions ; in Plautus, at any rate, the insistence is on
type rather than on individuality.
The form of New Comedy was extremely simple. The
typical setting was a section of a street, with the houses of
some of the leading characters in the background. The
audience are in the position of chance spectators of what is
happening on this section of the street. Characters go in and
out of the house-doors, or up and down the street on their way
to or from the country or harbour or the centre of the town.
The setting, whatever-it may be, remains unchanged through-
out the play ; the action of the play is confined to a single
day (exceptionally in the Heauton Timorumenos of Terence
a night intervenes). Modern editors divide the plays of
Menander into acts and scenes. This is rather misleading ;
the action of New Comedy was continuous ; characters came
and went, but there was no curtain, and no pause in the
performance. Occasionally in New Comedy a band of
revellers, guests, etc., may come along the street, give a song
and dance and depart or enter a house ; sometimes we find
actors withdrawing from the stage with the remark that they
wish to keep out of the way of the revellers. These dances—
we have evidence of several of them in the extant fragments of
Menander, including three in the Epitrepontes—are apparently
all that is left in New Comedy of the choral element. Whether
they occurred in all plays, or how often they occurred in any
one play, we cannot tell. They seem to be a survival, a
relic of tradition, and when they were omitted altogether (as
was apparently the practice of the Latin dramatists) their
omission seems to have had little effect on the structure of
the play. 1
In sharp contrast both to Old Comedy and to the Plautine
versions, New Comedy seems to have been almost without
musical effects. The bulk of the play was in plain speech; here
and there a change to trochaic metre indicates that a passage
of heightened emotion is delivered to musical accompaniment;
but our extant fragments show nothing like the wealth of
metrical and musical effects characteristic of Aristophanes and
Plautus. The prevailing tone, at least in Menander, seems to
have been sober and restrained. The zest in life has gone ;
1
See ch. xxv.
 GREEK NEW COMEDY 45
man cannot control his fate ; by resignation and restraint he
must seek for such happiness as can be attained.
No one would find such an attitude to life in the plays of
Plautus—even in those which, like the Stichus, seem to have been
translated from originals by Menander. No doubt Plautus
chose the liveliest originals he could find. Not all the writers
of New Comedy were as refined as Menander. But there can
be no doubt that we should ascribe to Plautus himself the
heartiness, the boisterous mirth, the delight in humour of
situation and vigour of dialogue, and some at least of the
indifference to artistic canons, which distinguish his plays
from what else we know of New Comedy.
We must take Plautus as we find him. We must give him
some of the credit for what is good, as well as the blame for
what is bad. Of his originals we know almost nothing except
what we can infer from his versions of them. If we think one
of his plots badly made, we cannot assume that the faults of
construction are introduced by the Latin writer. What he did
do was to choose his originals—we cannot tell how large a
number of Greek plays was available to him—and put them
into Latin. For the variety of his plots, for the metre and the
Latinity, he must be held responsible.
 CHAPTER VII

T H E F A M O U S P L A Y S OF PLAUTUS

T HE F O L L O W I N G summaries of some of Plautus'

plays may serve to illustrate the variety and limitations
of New Comedy.

(i) Amphitruo
The prologue is spoken by the god Mercury. He explains
that his father Jupiter has fallen in love with Alcumena, whose
husband, Amphitruo, is abroad in command of the army of
Thebes. Jupiter has assumed the likeness of Amphitruo to
deceive the virtuous Alcumena, whose company he is enjoying
at this moment (the time is early morning, before daylight).
Mercury has himself assumed the likeness of Amphitruo's
servant Sosia. The gods, he tells us, wear tokens to help the
audience to distinguish them from their human counterparts.
The real Sosia now enters from the harbour with tidings for
Alcumena : Amphitruo has won the war and will be home
almost immediately. Mercury accosts Sosia, claims to be
Sosia himself and drives the bewildered servant off the stage.
Sosia goes off to report this incident to Amphitruo. Jupiter
now comes out of the house, bids a tender farewell to Alcumena
and departs. A little later the real Amphitruo appears,
expecting a tender welcome. He is astonished to be received
with cold surprise by his wife, who thinks he is playing some
trick on her. His astonishment turns to fury when she tells
him that he has only just left her ; he hurls charges of in-
fidelity against her, which she receives with admirable dignity,
conscious of her innocence. Amphitruo goes off to find
witnesses ; Jupiter appears and cajoles the angry Alcumena
into a good temper; they go inside. When Amphitruo
returns, Mercury, still in Sosia's guise, bolts the door against
him, treats him as a stranger and pelts him from the roof. At
this point some scenes of the play are missing or in a fragmentary
condition ; evidently the real and the fake Amphitruo are
46
 THE FAMOUS PLAYS OF PLAUTUS 47

brought face to face, to the mystification of onlookers.

Finally Jupiter reveals himself in his divinity, explains the
trick which he has played, and soothes the unfortunate
Amphitruo, who accepts the extraordinary situation that
the twin brother of his newly-born son is Jupiter's child, the
infant Hercules.
The Amphitruo, described in the prologue as a tragicomedy,
is alone among the extant examples of New Comedy in bringing
gods on the stage as characters in the play. Their behaviour
is, indeed, unworthy of deity; but we must not take these
burlesques too seriously. Similar themes were popular in the
farces of Magna Graecia ; a phlyax-painting shows us Zeus
(Jupiter) on a love-adventure, escorted by Hermes (Mercury).
What is remarkable is the noble bearing of Alcumena, who
maintains in the most cruel circumstances her dignity, her
calm consciousness of innocence and her affection for her
husband. There is plenty of knock-about farce, but it is not
suffered to touch this queenly figure.1

(г) Aulularia.
An old Athenian citizen named Euclio, who lives in a very
poor way with his daughter Phaedria and his old servant
Staphyla, has found hidden treasure in his house, and now goes
in constant terror that some one else will discover his secret.
He does not know that Phaedria has been wronged by a young
gentleman of the neighbourhood ; this we learn from the
prologue, spoken by the household god, who adds that the
young gentleman's uncle (also unaware of Phaedria's situation)
is going to ask Euclio for her hand in marriage. We are now
given a vivid picture of Euclio's miserable state of suspicion.
When his next-door neighbour, the wealthy Megadorus (uncle
of the guilty Lyconides) comes to ask for Phaedria's hand,
Euclio at once suspects that he has got wind of the treasure ;
however, he gives his consent—provided that Megadorus does
not expect the bride to bring a dowry. During Euclios'
absence the cooks sent by Megadorus arrive to prepare the
wedding breakfast in his house. When he returns he drives
them out and hurries off with his treasure to find another
place of concealment. He is shadowed by Lyconides' slave, who
1
T h e spirit of the phlyax-painting is utterly different; and the situation
which it represents is quite unlike anything that happens in the Ampkitruo.
48 THE ROMAN STAGE

finally manages to unearth the treasure ; when Euclio discovers

his loss, his grief knows no bounds. Lyconides, who has come
to admit the wrong he has done to Phaedria and his desire to
make amends, imagines that Euclio's distress is due to know«
ledge of Phaedria's condition. Eventually the tangle is
straightened out, Lyconides recovers the treasure from his
slave, Euclio consents to the marriage of Phaedria and
Lyconides and apparently makes a wedding-present of the
treasure, much to his own ultimate relief and peace of mind.
Euclio is one of the outstanding characters of Latin drama.
Though not without some characteristics of the miser, he is
fundamentally a decent old fellow, crazed by the sudden
acquisition of wealth ; he is not a Shylock or a Harpagon,
though these famous characters owe much to him. As is
usual in New Comedy, the unfortunate heroine does not appear;
we only hear her moaning when in the pangs of childbirth.
Among the subsidiary characters Megadorus and his sister
Eunomia show a generosity of nature and a sense of social
responsibility which contrast favourably with the selfishness
of so many of Plautus' characters.

(3) Captiui.
An old Aetolian gentleman named Hegio had two sons ;
twenty years before the play opens one of them had been
carried off in infancy by a rascally slave to Elis and sold to a
gentleman who had given him the name Tyndarus and made
him personal slave to his own little son Philocrates. During
the war which is now being waged between Aetolia and Elis
Hegio's other son, Philopolemus, has been captured ; Hegio,
hoping to arrange an exchange of prisoners, has purchased
several Elean captives, his latest acquisitions being Philo-
crates and Tyndarus. They have secretly exchanged clothes
and names, expecting that Hegio will send the supposed slave
home to make arrangements for the ransom of his master. The
plan works admirably, and Philocrates departs. Unfortunately
Tyndarus is recognized by another of Hegio's captives, and
the poor old gentleman is so infuriated to learn how he has
been fooled that he orders Tyndarus off to the terrible
quarries, there to be worked to death. Philocrates returns
(with miraculous quickness, it must be confessed) from Elis,
bringing with him Philopolemus and the rascally slave who
 THE FAMOUS PLAYS OF PLAUTUS 49

had stolen Hegio's other son ; enquiries disclose that this son
is Tyndarus, who is thereupon hastily fetched back from the
quarries, and all ends happily.
As befits a comedy, the Captiui contains a lively plot and
some capital fun ; but it derives its unique quality from the
strong dramatic irony of the situation, the devotion of the
Captives to each other and the noble and generous atmosphere
which pervades the play.
(4) Menaechmi.
Two twin brothers had become separated in infancy ;
Menaechmus I, kidnapped and carried off to £pidamnus, has
had the good fortune to be made heir to a wealthy old gentle-
man, who has since died. Menaechmus II, having grown
to manhood in his native Syracuse, has started on a search for
his brother and is now newly arrived at Epidamnus with his
slave Messenio. The exact resemblance of the two brothers
leads to a ' comedy of errors ' which threatens to have serious
consequences; eventually the brothers meet, and all is explained.
The theme of mistaken identity was frequently exploited in
New Comedy, to judge by the number of plays entitled ' The
Twins ' or ' The Doubles '. The interest of the Menaechmi lies
chiefly in its well-handled if improbable plot and its amusing
scenes ; the two brothers alienate our feelings by their selfish-
ness, but some of the minor characters are well portrayed,
especially the doctor who is called in by the father-in-law of
Menaechmus I to treat the young man's supposed insanity.

(5) Miles Gloriosas.

Pyrgopolynices, a vain, amorous and timid ( braggart
c a p t a i n h a s secured possession of Philocomasium, mistress
of the young Athenian gentleman Pleusicles, and has also
purchased Pleusicles' slave Palaestrio, who had been kidnapped
by pirates. Palaestrio has managed to send word to Pleusicles,
who is now living in the house of his friend Periplectomenus,
next door to the captain. By breaking a passage through the
party wall Palaestrio has enabled the lovers to meet secretly.
Unfortunately another slave of the captain peers down through
the skylight and perceives Philocomasium in the wrong house ;
he has to be persuaded that the girl he has seen is really
Philocomasium's twin sister, a feat which is achieved by the
5° THE ROMAN STAGE

use of the secret passage and the girl's skilful performance

in the double role. Finally Pyrgopolynices is induced to lose
interest in Philocomasium by a report that his neighbour's
wife is in love with him ; he allows Philocomasium to depart
with a supposed ' ship's captain ' (really Pleusicles in nautical
guise) who claims to have been sent by her mother, while the
captain himself, in pursuance of his new intrigue, enters his
neighbour's house, there to be seized and threatened with the
barbarous punishment to which detected adulterers were
liable. When his cowardice has been thoroughly exposed he
receives a contemptuous pardon.
The brilliant portraits of the foolish captain, his cunning
parasite and the genial old bachelor Periplectomenus, as well
as the vigour of the play as a whole, compensate for certain
improbabilities and weaknesses, such as the extraordinary
ignorance on the captain's part as to his neighbour's domestic
circumstances and the fact that the two main themes—the
trick of the secret passage and the trick of the fictitious love-
message—seem to have little to do with each other.

(6) Mostellaria (' The Ghost Story').

Theopropides, an old gentleman of Athens, has been
abroad on business for three years ; during his absence his son
Philolaches lives a wild life under the direction of the slave
Tranio, and has even borrowed from a money-lender in order
to purchase the freedom of his mistress, the affectionate and
faithful Philematium. A merry drinking-party of the lovers
and their friends, Callidamates and his mistress Delphium, in
front of the house is cut short by Tranio's arrival with news
that he has seen Theopropides at the harbour. Tranio takes
command, hurries the revellers indoors, locks the house and,
when Theopropides arrives, scares him with a story that the
house is haunted and has been abandoned. Matters are
complicated by the arrival of the money-lender in search of
his dues ; Tranio, hastily building lie on lie, pretends that
Philolaches has borrowed money in order to purchase another
house. What house ? Why, the house next door. Theopro-
pides wishes to inspect the purchase ; the occupant appears ;
Tranio manages to keep the conversation at cross-purposes
during an amusing scene. Inevitably Theopropides finds out
that he has been deceived ; Tranio, threatened with disaster,
 JUPITER AND MERCURY
РЫуах Vase-painting in the Vatican Museum
 THE FAMOUS PLAYS OF PLAUTUS 51

takes refuge just in time on the altar and there manages by

native wit and impudence to hold his own until Callidamates
appears and gets Theopropides to grant a general pardon.
The irrespressible and endlessly resourceful Tranio is
perhaps the most delightful portrait in Latin comedy of the
' cunning slave \ The other portraits are all well drawn—the
spendthrift young men, the tender Philematium, her worldly-
wise old servant Scapha, the niggardly Theopropides, his
more genial neighbour Simo, and the money-lender. The plot
is one of the best-handled in Plautus.

(7) Rudens (' The Rope ').

An old Athenian gentleman named Daemones, whose
daughter had been kidnapped in infancy, and who has lost
his fortune through liberality, is living in humble exile in a
lonely cottage on the African shore near Cyrene. His daughter,
now called Palaestra, is in the power of the pimp Labrax, who
has brought her to Gyrene ; there a young gentleman from
Athens named Plesidippus has seen and fallen in love with her,
and has even paid part of her purchase-money to Labrax.
The pimp has tried to slip away with his slaves to Sicily, but
is wrecked ; Palaestra and her maid come to shore near the
cottage of Daemones and take refuge in the temple next door.
Labrax also swims to shore and tries to seize the girls ; Dae-
mones gives them protection, and Plesidippus arrives and drags
Labrax off to be tried in Gyrene. Gripus, a slave of Daemones,
fishes up a trunk containing (as he suspects) treasure ; the
trunk belongs to Labrax, and inside are the toys of Palaestra's
childhood, which proves that she is the daughter of Daemones,
and can therefore marry Plesidippus.
The rural setting, the story of the storm, shipwreck and
treasure-trove and the comparatively high moral tone of the
play as a whole, make the Rudens one of the most refreshing
examples of New Comedy. We may call it the most romantic
play of Plautus, if this term is understood not to include any
strong love-interest; the treatment of the love-theme is indeed
perfunctory.
Noteworthy features in the other plays of Plautus are the
cynical studies in debauchery afforded by the Bacchides and the
TruculentuSythe vivid portrait of the pimp Ballio in the Pseudolus
and. as a contrast, the moralizing tone of the Trinammus, which
52 THE ROMAN STAGE

play also contains the most amusing exposure of a trickster in

Plautus. The Epidicus (apparently one of its author's favourite
plays) has a plot of remarkable complexity. In the Asinaria
there is an effective scene in which a canny trader foils the plot
of two swindlers ; the Casino (a very popular play) is excep-
tional in the riotous indecency of its concluding scenes ; the
Cistellaria is unusually sentimental, a comidie larmoyante; the
Curculio has a striking opening scene, showing the lover calling
on his mistress in the darkness of early morning. In the
Mercator father and son are rivals in love ; this play is brilliant
at times, yet curiously lacking in the usual Plautine touches.
The Persa is remarkable in bringing a freeborn girl on the
stage ; though she is only the daughter of a parasite, she has a
dignity which contrasts sharply with the behaviour of the
usual courtesan. The Poenulus contains a not unsympathetic
portrait of a Carthaginian ; his remarks in Punic (or perhaps
pseudo-Punic) are ridiculously misinterpreted by a slave
in a scene which must owe much to the Latin translator.
The Stichus, one of the few plays known to be translated by
Plautus from Menander, ends like the Persa, in a slaves'
drinking-party. Finally the fragmentary Vidularia deals with
the fortunes of a gentlemanly young castaway who is forced to
hire himself out as a manual labourer.
 CHAPTER VIII

PLAUTUS : TREATMENT OF HIS ORIGINALS

KE MOST if not all of the other Latin dramatists,

Plautus was not so much an original playwright as one
who adapted Greek drama to Roman taste. We have no
convincing evidence that he invented a single plot or character,
or introduced into his originals any alterations which show
constructive power. Scarcely one scene among several
hundred can with certainty be ascribed to his independent
authorship. His originality shows itself firstly in the fact that
he limited himself to a single field, the translation of Greek
New Comedy, secondly in his choice of plays for adaptation,
thirdly in his intuitive perception both of what public taste
required and of the limitations under which he must work ;
but above all in his command of language and metre, of jest
and metaphor, of rhetoric and repartee.
In limiting himself to one field of drama he set an example
which was followed by nearly all subsequent dramatists. Both
in width of interests and in power of innovation he was inferior
to his contemporaries Naevius and Ennius ; he confined him-
self to comedy, not through lack of command of the tragic
style (witness the noble rhetoric of his Captiui) but, presumably,
because comedy was the field in which he thought he could
win greatest success. If his work was governed by practical
considerations (Horace accuses him of being entirely mercenary),
he seems at least to have taken some pride in his achievement;
he tells us that he prided himself in his Epidicus, and Cicero
makes Cato assert that Plautus admired his Pseudolus and his
Truculentus.1 These favourite plays have certain qualities in
common—a complicated plot, ρ lenty of trickery, some
remarkable turns of fortune, vigorous dialogue, metrical
variety, heartiness, cynicism—quali ties which are, indeed,
common to most of his plays, and which, when combined,
produce a result different from the extant work of any other
1
A remark which (at least) tells us something of Cicero's tastes.
5 5J
54 THE ROMAN STAGE

ancient dramatist. How wide a choice of Greek originals he

enjoyed is outside our knowledge. It is striking that he seems
to have taken so few plays from Menander—perhaps not more
than three out of the twenty-one, whereas with Terence the figure
is four out of six. Liveliness of plot and abundance of farcical
situations seem to have been what Plautus sought, not that
interest in theme and in personality which attracted Terence
to Menander. Scarcely a play of Plautus but contains at
least one scene of uproarious farce. Something may be
learned of his tastes by observing the cases where he changed,
or is thought to have changed, the title of his original. From
Philemon's somewhat moralistic ' T r e a s u r e ' (Thesaurus) he
translated his Trinummus ; the Latin title (if, indeed, due to
Plautus himself) shows that in his eyes the outstanding scene
was the encounter between the trickster (hired for ' three
n u m m i ' ) and the man whom he was impersonating. The
Rudens (translated from an unknown play by Diphilus) has
many virtues in our eyes :—its refreshing, unusual setting by
the lonely African shore, its romantic story of shipwreck and
treasure trove, its satisfying conclusion—honesty rewarded,
child and parent restored to each other, love triumphant—
but the Latin title, ' The Rope ', indicates that for the inventor
of that title the central scene was the tug-of-war between
Trachalio and Gripus for the possession of the treasure. The
Stichus was translated from one of two plays entitled ' The
Brothers ', by Menander. It has been criticized on the ground
that its three parts have little to do with each other : part one,
the faithful wives ; part two, the disappointed parasite ; part
three, the slaves' drinking-party ; but whereas part one and
two have in common at least the inimitable Antipho, the Roman
writer's own taste seems indicated by the change of title ;
Stichus, the merry slave, does not appear until part three.
For a dramatist whose only object is to amuse the ground-
lings, an obvious resource is obscenity ; indeed, no comic
dramatist of antiquity found it easy to resist this temptation.
Terence's most successful bid for popularity, the Eunuchus,
contains the most artfully sexual scene in Latin drama ; and
it is significant that the very play which we know to have been
revived soon after the death of Plautus, the Casina, is of all his
works the one which errs most grievously against decency.
There is, in fact, just enough obscenity in Plautus to т а к '
 PLAUTUS : TREATMENT OF HIS ORIGINALS 55

wonder why we do not find more. Apparently the police

imposed no such absolute ban here as they did in the matter
of personal slander and political satire. Of course views
change from age to age as to what is or is not objectionable.
The early editors of the Renaissance were shocked by such a
play as the Mercator,1 in which a father and son appear as
rivals for the affections of a slave-girl; to some modern
readers such a situation appears merely a piquant variation
of a stock theme of New Comedy, and some scenes of the play
possess a grace and brilliance which seem to come straight
from the Greek originals, untainted by coarseness of expression.
But why are there not more scenes in Plautus like the concluding
scenes of the Casina, so Italian in tone, so reminiscent of Ovid,
Petronius and Apuleius ? The truth appears to be that on the
topic of sex the Romans were highly sensitive, and public
taste imposed its own limits. We must remember that the
spectators included not only magistrates but women. That
some at any rate among the Roman audience were not in
favour of licence is suggested not only by the writer's conscious
pride in the high moral tone of the Captiui—' poets find few
plays like this, showing good men becoming better '—but also
in his care never to risk a jest which might sully the honour of
a free-born woman.
All theories which ascribe to him independent power of
plot-construction are based on hypothesis. One of the few
alterations made by him of which we can be fairly certain is
that, as Terence tells us, in the Commorientes he omitted (or
perhaps curtailed) one scene of his original, the Synapothnescontes
of Diphilus. This omission Terence ascribes to Plautus'.
' carelessness', and his explanation may be accepted, if we
interpret it as meaning that Plautus set no particular store on
fidelity of translation—he took what he wanted and left what
he did not want. The remark in the prologue of the Casina
that the young gentleman will not appear in the play, because
Plautus has broken down the bridge by which he was to return
home, may be merely a jesting way of saying that the plot
(even of the Greek original) did not allow this character to
appear. Of the expansion of certain scenes by the addition of
stock comedian's ' patter' there is, indeed, strong evidence.
The monologue of the Property-manager in the Curculio is
1
Sec the edition by Professor Enk, i. p. 22 ff.
56 THE ROMAN STAGE

certainly an addition by Plautus or a later hand. It is a

description of Rome ; it has no relation to the plot; its effect
is merely to entertain. Possibly we should also ascribe to
Plautus the long Punic passages in the Poenulus ; presumably
some of his audience had learned the language during the
war with Hannibal. That such enlargements might be
balanced by curtailment elsewhere is ä not unreasonable
supposition (though there seems to have been no fixed length
for a play, and some are twice as long as others) ; in this way,
perhaps, we can explain the omission (or curtailment) of the
lively abduction-scene in the Synapothnescontes. But of re-
modelling of the plot there is no convincing evidence. Perhaps
the play which has been most often suspected is the Stichtis,
and its three parts have often been felt to have little to do with
each other. But if Plautus drastically (and carelessly) re-
modelled the Stichus, how are we to explain the neatness with
which the roles are made to dovetail into each other, so that a
cast of only three actors could perform the whole play ?
To ascribe to Plautus the credit for his plots and characters
may be justified only in the sense that it was he who chose and
translated the plays which possessed these plots and characters.
The dignified Alcumena, the absurd yet pathetic Euclio, made
miserable by his wealth, the noble and moving farewell scene
of the Captiui, the deft intrigue, droll situations and amusing
characters of the Menaechmi, Miles Gloriosus and Mostellaria—
all these have indeed been transmitted to world literature by
Plautus, but he was not their creator. Nevertheless it would
be quite wrong to think of him as a mere translator. However
closely he follows his originals, he could scarcely have avoided
infusing his versions with his own spirit. Without necessarily
attributing to him any structural alteration, we can admit that
on a congenial theme he let his pen run away with him,
regardless of ' the necessary business of the p l a y T h e
dialogue between Sosia and his divine double, in the Amphitruo,
has surely been expanded by the Latin writer ; it is indeed the
longest scene in Plautus, and most of it is mere foolery—but
how Roman ! This quarrel between a street-bully and his
victim (si rixa est ubi tu pulsas, ego uapulo tanturn) is as racy of the
streets of Rome as the famous scene in Juvenal's third satire.
What the Romans themselves ascribe to Plautus was
command of metre, dialogue and jest. When he died ' all his
 PLAUTUS : TREATMENT OF HIS ORIGINALS 57

measures beyond measure wept for h i m a s his epitaph runs.

For dialogue he was ranked by V a r r o above all other Latin
comic dramatists, and Aelius Stilo said that, if the Muses were
to ?peak Latin, it would be the Latin of Plautus. Horace
grudgingly admits that the Romans have been only too ready
to praise Plautus' numeri and sales ; his wit is put by Cicero on
a level with that of Attic O l d Comedy ; Sidonius Apollinaris
regarded him as actually superior to the Greeks, and St.
Jerome used to console himself, after a night of weeping over
his own sins, by reading Plautus. It is hard to believe that
any one would turn to Menander for similar consolation. Even
supposing that many of Plautus' jokes came from his originals,
we must admit that it is not difficult to spoil a good jest in the
telling. T h e life and gaiety of his style could not have been
achieved by any obscura diligentia in translation. And when the
jokes depend (as they so often do) on a pun, possible only in
Latin, or on some topical allusion which would have no
meaning outside Italy, we must agree that here the full credit
is to be given to the Latin writer.
T h e fragments of the earlier dramatists show nothing
approaching Plautus' variety of metres (though we must always
regret that time has spared so little of Naevius), nor do later
writers appear to have rivalled him in this respect. In develop-
ing comedy on its musical side Plautus must have depended
to some extent on the development of instrumental talent
in R o m e ; without setting too high the skill of Pellio, who
perhaps undertook the title-role of the Epidicus (apparently
not to Plautus' own satisfaction) or that of Marcipor, slave
of Oppius, who played the music for the Stichus, we may at
least agree that the technique of such performers was not
created in a day. It is interesting to note that the metrical
effects are particularly rich in the plays which seem to have
been produced in Plautus' maturity or old age, while the
Mercator and Asinaria, which are unusually simple in metrical
structure, have for other reasons been thought to be early
works (if indeed they are by Plautus at all). T h e Miles,
which is also simple in metrical form, appears from the
reference to Naevius' imprisonment to be another of the early
works. O n the other hand the Cistellaria, which contains a
reference to the Hannibalic war as still going on (but apparently
drawing near its conclusion) shows a large range of metrical
 58 THE ROMAN STAGE

effects. With the Casina, of unknown date but certainly dis-

playing the author in characteristic vein, the variety has
become prodigious.
The speed and power of Plautus' dialogue, and the purity
of his style, were acknowledged by ancient critics ; nor do we
depend on ancient testimony alone, for we have some twenty
thousand lines of racy Latin, covering all the subjects within
the sphere of broad comedy—buffoonery, conviviality, endear-
ment, abuse, worldly wisdom and popular philosophy. What
could be more effective than to address an enemy a s ' Dug from
a dunghill ! ' (ex sterculino effosse) ? Where are the raptures of
love better portrayed than in the words of Phaedromus ?
sibi sua habeant regna reges, sibi diuitias diuites,
sibi honores, sibi uirtutes, sibi pugnas, sibi proelia :
dum mi abstineant inuidere, sibi quisque habeant quod suom est.
' Gie me a cannie hour at e'en,
My arms about my dearie O,
An' warldly cares, an' warldly men
May a' gang tapsalteerie О ! '
Such moments do not last long, and always in the background
there is the cynical spectator to point out love's follies—for this
is Latin comedy. Pseudolus, in the moment of his triumph,
grows for a moment grave at the thought of the vanity of
h u m a n wishes :
stulti haud scimus frustra ut simus, quom quod cupienter dari
petimus nobis, quasi quid in rem sit possimus noscere.
certa mittimus dum incerta petimus; atque hoc euenit
in labore atque in dolore, ut mors obrepat interim.
This is in the mood of the grave Roman satirist, with his eye
on the sun-dial:
dum serta, unguenta, puellas
quaerimus, obrepit non intellecta senectus.1
The solemn note must have struck a responsive chord even in
the hearts of that turbulent audience ; for mankind passes
easily from one extreme of emotion to the other. This mixture
of jest and gravity meets us elsewhere in popular Latin drama.
But Pseudolus does not linger on the heights :
sed iam satis est philosophatum ! nimi' diu et longum loquor.
1
Juv. ix. 128-9.
 PLAUTUS : TREATMENT OF HIS ORIGINALS 59

Every one can recognize in Plautus the uis comica which

Julius Caesar found wanting in Terence ; all the more striking,
then, is the occasional recurrence of the serious note. Nor
does this note always end in mockery. Something more than
farce and cyncism was needed to win from Lessing his tribute
to the Captiui as ' the finest play that has ever been put on the
stage \ The farewell of Tyndarus to the friend and master
for whose sake he is going into deadly peril is of a quality which
may move us, as it moves the honest Hegio, to admiring tears.
And when the deception practised on the old man has been
disclosed, and Tyndarus must pay the full penalty for his noble
lie, he faces the prospect of death with a spirit worthy of a
Roman legionary :
For if he break his word and not return,
And I be fated in this land to die,
Yet this my deed, when I am dead and gone,
Will be for ever bright and memorable :
That I from slavery, from foeman's might,
Rescued my master, made him free once more
To look upon his father and his home ;
And, when no other hope nor help was his,
To save my master's life, I gave—my own.
 CHAPTER IX

T H E GENERAL C H A R A C T E R O F ROMAN
TRAGEDY

T H E R E IS a widespread belief that tragedy was never

popular at Rome. For this belief no better evidence is
usually offered than a jest in the prologue of the Amphitruo :
' I will tell you the plot of this tragedy. What ? Are you
wrinkling your foreheads because I said it would be a tragedy ? '
The real reasons for the modern view are perhaps that only
fragments of the tragedy of the Republican period have
survived, and that the rhetorical style of these fragments makes
little appeal to modern taste, especially when we compare
them, as we sometimes can, with the originals by the great
Greek masters. It must be remembered, however, that Roman
tragedy continued to be performed for more than two hundred
years, that the Romans regarded Ennius, Pacuvius and Accius
as great tragic writers, that the tragedies of Ennius were
sufficiently well known to the common people to induce
Plautus to burlesque their style, that from the three leading
writers alone over seventy titles have come down to us, that
performances of tragedies in the time of Cicero were attended
by eager crowds, some of whom knew the classics of the stage
so well that at the first notes of the flute they could tell what
performance was to follow, and that on one occasion, when an
actor failed to take his cue, ' twelve hundred ' voices echoed
the words ' mater, te appello '. In fact Graeco-Roman tragedy
had at least as long a career on the Roman stage as any other
form of literary drama, and its effect on the popular mind
must have been far-reaching.
For the modern reader, however, Roman tragedy has little
appeal. To begin with, it was almost wholly derivative. We
have seen that the introduction of the native historical play
proved a failure, only five or six fairly certain examples being
recorded for the entire Republican period. In dealing with
Greek tragedy Latin translators seem to have shown even less
60
 GENERAL CHARACTER OF ROMAN TRAGEDY 6l

originality than in dealing with Greek comedy. Neither in

tragedy nor in comedy have w e any evidence that the Latin
writers (apart from Terence) introduced structural changes ;
they seem to have confined their alterations for the most part
to sentiment and style. 1 I n comedy they had the congenial
task of infusing crude vigour into the somewhat languid Greek,
and of scattering jests from a full sack. It is arguable that the
plays of Plautus would prove better reading than their Greek
originals, could w e compare them. But the R o m a n tragic
style was a poor equivalent for the style of Aeschylus or
Euripides. Compared with the Greek, the Latin usually seems
stiff, strained and unnatural. T h e R o m a n s appreciated the
very qualities which repel us. T h e y enjoyed melodramatic
effects, volleys of rhetoric, horrific plots and descriptions,
flamboyant personalities, superhuman virtue, incredible vice.
In another mood they were capable of laughing at these very
things ; ut paratragoedat carnufex ! ' how the knave rants ! '
as a character of Plautus remarks. O u r scraps of information
on the behaviour of the crowd at performances of tragedy
suggest that they were interested not so m u c h in the essential
dramatic qualities of the performance as in externals—im-
pressive staging, violent utterance and action, lines which
might be taken as topical, the arrival of distinguished spectators,
and of course any mishap which might befall either the actors
or any members of the audience. Y e t in fairness w e should
remember that it is usually the exceptional which is recorded.

W h i l e we possess the names of over a dozen translators

of Greek comedy, only five writers are certainly known to
have composed tragedy for the stage. Terence's prologues
suggest that the comic dramatist lived in a fiercely competitive
world ; the careers of the writers of tragedy overlapped
hardly at all. It seems that Livius was dead and Naevius was
in disgrace by the time Ennius came to R o m e , and Pacuvius
was eighty years old when Accius m a d e his first appearance.
Livius and Naevius were certainly contemporaries, but
Naevius' chief work was in comedy. T h o u g h Ennius and his
nephew Pacuvius were producing tragedies at the same time,
their relations m a y have been those of master and pupil rather
than of rivals. It is obvious that Naevius, Plautus, Gaecilius

ec 1 As for the modern theory that they combined different Greek originals,
s the discussion of contaminatio in the chapter on Terence.
62 THE ROMAN STAGE

and Terence had, each in his own way, a special inclination

for comedy ; but what induced the tragic writers to take up
tragedy, except the need to earn a living in the only way open
to a writer—the stage—and the knowledge that they had no
comic talent ? Ennius was a great poet and personality, but
it is hard to feel sure that he had any special dramatic gift ;
Pacuvius seems to have had a liking for complicated plots and
dramatic discussion of philosophical problems, and Accius had
a noted turn for repartee. But the fact that only fragments
have survived from the tragedies of these writers practically
confines our discussion of their powers to points of style.
Criticisms by ancient commentators are sometimes helpful;
sometimes the choice of subjects indicated by the titles is
suggestive. In a few cases we have the Greek originals, or
fragments of them, and elsewhere our knowledge of Greek
mythology helps us to reconstruct the plots. The lines of
development of Republican tragedy may to some extent be
inferred from what tragedy became under the Empire. But
the often-cited evidence of vase-paintings, wall-paintings, etc.,
can seldom be shown to be relevant; the mural decorators of
Pompeii probably derived their ideas of mythology from quite
other sources than Republican tragedy. If these illustrations
from imperial times are indeed taken from the stage, the
exaggerated stiffness of the figures and the dress may reflect
imperial rather than republican ideals. In imperial times
tragedy was something of a survival; such tragedies as were
produced may have been the by now archaic plays of Pacuvius
and Accius, and the whole performance may have been
consciously artificial. In earlier times, when Roman tragedy
was still vital, the style of acting and of dress may have been
considerably more natural.
If we ask what was the essential character of Roman
tragedy, perhaps the answer should be that it was a harsh,
exaggerated yet not wilfully unfaithful imitation of its Greek
originals. Differences in detail will be noticed when we deal
with the individual writers.

ENNIUS

As if by official arrangement, there always appears to have

been one recognized writer of tragedy in Rome as long as
 GENERAL CHARACTER OF ROMAN TRAGEDY 63

tragedy was still vital, and seldom more than one. The death
of Andronicus and the disgrace of Naevius left the way clear
for their successor, who, after completing his military service,
arrived in Rome in 204 B.C. as a member of the suite of M .
Porcius Cato. Quintus Ennius had a human weakness for
talking about himself. Our biographical details are probably
derived from his works, and for that reason fairly reliable.
They show him to us as poor yet independent, proud of his
native town and of his Roman citizenship, able to live on easy
terms with the great, yet himself belonging to the workaday
class of dramatists such as his fellow-lodger Caecilius and his
nephew Pacuvius. Born in the year 239 at Rudiae, in the
south of Italy, he prided himself on his knowledge of Greek,
Latin and Oscan (the three languages, we may add, of ancient
drama). From his arrival in Rome at the age of thirty-five he
devoted himself to his many-sided literary activity. A man of
versatile genius and broad human sympathies, an Italiot whose
imagination was fired by the literature of Greece and the
history of Rome, he brought into Latin literature and life a
fresh impulse of Hellenism. Like Livius Andronicus he earned
his living partly as a teacher and partly as a dramatist. Drama
was only one of his literary activities. Though Terence refers
to him together with Plautus and Naevius as one of those
careless but gifted writers whom he preferred to his own
laborious but dull contemporaries, it was as an epic poet rather
than as a dramatist that he was honoured by later generations.
He was apparently not regarded as the equal of either Pacuvius
or Accius in tragedy, while in comedy Volcacius Sedigitus puts
him tenth, adding that even this honour is granted only in
recognition of his early date ! We hear nothing elsewhere of
Ennius as a writer of comedy, and possess of his comedies only
two doubtful titles (Cupuncula, Pancratiastes) and three lines on
stock topics.

The tragedies of Ennius undoubtedly had great influence,

both as drama and as literature. We have twenty titles
{Achilles, Ajax, Alcumeo, Alexander, Andromacha, Andromeda,
Athamas, Cresphontes, Erechtheus, Eumenides, Hectoris Lytra, Hecuba,
Iphigenia, Medea Exul, Melanippa, Nemea, Phoenix, Telamo,
Telephus, Thyestes), with about four hundred lines ; we also
have what some claim to be the titles of two praetextae,
Ambracia and Sabinae, with a few lines.
64 THE ROMAN STAGE

The Hellenistic stage was dominated by the influence of

Euripides, and in Euripides Ennius found a writer who appealed
to his own questioning spirit and humanitarian sympathies.
An ancient scholiast tells us that Ennuis took ' very many '
tragedies from Euripides ; indeed more than half of his twenty
titles have with greater or less probability been considered
Euripidean. In the Hecuba, Iphigenia and Medea Exul we can
compare the Latin fragments with the complete originals.
From Aeschylus he appears to have derived his Eumenides;
from Aristarchus, another fifth-century tragedian, his Achilles
and some unspecified plays. A remark of Cicero 1 has some-
times been strained to mean that Ennius took some of his plays
direct from Homer. This is utterly unlikely ; all our evidence
goes to show that the original of each Latin tragedy or comedy
was a Greek tragedy or comedy.
As to Ennius' methods of translation, we have two appar-
ently contradictory statements by Cicero, who says in one place
that the Roman tragedians (including Ennius and Pacuvius)
translated ad uerbum, ' word for word ', in another that Ennius,
Pacuvius and Accius translated поп uerba sed uim,' not the words
but the sense \ 2 These statements may perhaps be reconciled :
the Roman tragedians followed their originals closely in
substance, deviating only in choice of words. 3 The only evidence
that Ennius allowed himself independence in more than style,
sentiment and metre is provided by Gellius, who quotes a song of
the chorus in his Iphigenia to which there is nothing correspond-
ing in the original, Euripides' Iphigenia at Aulis. In Euripides'
play the chorus are a band of young married women of Chalcis
who have come across the straits merely out of curiosity to
see the Greek fleet. Apart from singing occasional songs,
songs which are sometimes long and difficult and have only
indirect bearing on the story, they take practically no part
in the play. For some reason Ennius has, it seems, introduced
a party of soldiers. The canticum which he puts into their
mouths reminds us of many passages of Plautine patter ; it is

1
De Fin. I. iii. 7 : ut ab Hotnero Ennius, Afranius a Menandro solet. Cicero is
probably referring to Ennius' Annates ; in any case he is speaking of the insertion
of occasional passages from a Greek author into an original Latin work.
* De Fin. I. ii. 4 ; Acad. I. iii. 10.
3
Where Ennius translated Greek plays which have themselves been preserved
complete, practically all the Latin fragments can be compared with their source in
the Greek play.
 GENERAL CHARACTER OF ROMAN TRAGEDY 65

a short sermon on the theme of idleness, expressing the boredom

of the soldiers (neque domi nunc nos пес militiae sumus). Just so
a Plautine slave will weary us with a sermon on obedience or
some such topic in a style which suggests that the passage is
entirely the work of the Latin writer. If Ennius felt that the
Euripidean chorus had no real connexion with the plot and
could safely be omitted, we may concede that he showed some
dramatic insight. But he may merely have been repelled by
the difficulty of the Greek, or have thought that his R o m a n
audience would prefer a troop of soldiers to a party of young
women. Nor can we be certain that he did drop the Euripidean
chorus ; the song of the soldiers may have been added for its
own sake. In the canticum itself there is nothing to suggest
original dramatic power on his part. In his Medea Exul he
evidently kept the Euripidean chorus of Corinthian women.
T h e fact that Ennius translated Aeschylus' Eumenides (the
fragments can for the most part be referred to their origin
in the Greek play) makes us wonder whether he also trans-
lated the two other plays of the famous trilogy ; it is hard,
however, to believe that, if he did so, the Romans would
have endured hearing the three plays produced one after the
other. T h e Eumenides of Aeschylus forms a satisfactory story
in itself; the dramatic situation could of course have been
explained in a prologue.
I n metre he took complete freedom. Sometimes he keeps
the Greek metre, whether iambics (the opening lines of the
Medea, trochaics (the argument between Agamemnon and
Menelaus, Iph. iv.-vi.) or anapaests (the opening of the
Iphigenia) ; sometimes he turns iambics into trochaics (Medea's
altercation with Jason) or, in a still more emotional passage—
Medea's farewell to her children—he turns speech into what
is usually regarded as song :

. . . saluete, optima corpora !

cette manus uestras measque accipite !

O n the other hand we find that a lyrical utterance of Euri-

pides' chorus {Med, 1251 ff.) has been turned into trochaics.
I n general, where we can set Ennius' Latin side by side
with the Greek, we find that the version is reasonably close.
H e does not shrink from tränslating the boldest utterances of
Euripides, such as Medea's famous assertion that she would
 66 THE ROMAN STAGE

r a t h e r fight three battles t h a n b e a r o n e child. All t h e R o m a n

tragic writers seem to h a v e i n d u l g e d in sceptical outbursts ;
Euripides p u t s into Achilles' m o u t h (Ipk. A. 955) a c o n t e m p -
tuous reference to soothsayers ; E n n i u s e x p a n d s t h e passage
i n t o a m o r e detailed a t t a c k o n astrologers.
T h a t E n n i u s was truly a poet is revealed in t h e d r a m a t i c
f r a g m e n t s as clearly as in t h e Annates. T h e opening lines of
t h e Iphigenia—a lyrical dialogue between A g a m e m n o n a n d
a n old m a n — c o n v e y s a sense of t h e majesty a n d mystery of
t h e starry sky n o t inferior to t h a t of t h e Greek. Similar is
t h e feeling in a line preserved f r o m t h e Hecuba :
О magna templa caelitum, commixta stellis splendidis !
w h i c h p e r h a p s corresponds to
& στεροττά Αιος, & σκοτία ννξ.
T h e r e is p a t h o s in these lines of t h e Alexander w h i c h inspired
Virgil h i m s e l f :
О lux Troiae, germane Hector !
quid ita . . . cum tuo lacerato corpore,
miser, aut qui te sic tractauere nobis respectantibus ?
A n d Cicero testifies to t h e m o v i n g p o w e r of t h e l a m e n t in t h e
Andromacha :
О pater, о patria, о Priami domus,
saeptum altisono cardinc templum !

I n general, however, t h e most characteristic q u a l i t y of t h e

f r a g m e n t s of E n n i u s ' tragedies, especially w h e n c o m p a r e d
w i t h t h e original Greek, is t h e h e i g h t e n i n g of t h e rhetorical
effect. All o u r evidence goes to show t h a t R o m a n t r a g e d y
was f a r m o r e rhetorical t h a n its Greek models. T h e r e a r e
times w h e n rhetoric has its p l a c e : t h e altercation b e t w e e n
A g a m e m n o n a n d M e n e l a u s is a n e x a m p l e (Iph. i v - v i ) . W h e r e -
as in Euripides we h a v e

ΑΓΑ. τί Se σβ τάμα Sei φυΧάσσειν; ουκ αναίσχυντου τόδί ;

ΜΕ. οτι το βούλςσθαι μ εκνιζε' σο? Be δούλος ουκ %φυν.
ΑΓΑ. ουχί Seiva; τον ϊμον о1кеш οίκον ουκ еач 4μέ;
i n t h e L a t i n we h a v e
Ag. quis homo te exsuperauit usquam gentium impudentia?
Me. quis ted autem malitia ? . . . .
 GENERAL CHARACTER OF ROMAN TRAGEDY 67

or again

Ag. egone plectar, tu delinques : tu pecces, ego arguar ?

pro malefactis Helena redeat, uirgo pereat innocens ?
tua reconcilietur uxor, mea necetur filia?

Where the Greek simply shows us two angry men wrangling,

the Latin seems to echo the skilled thrust and parry of a
debate between two Roman advocates. This rhetorical
effect is achieved by the familiar devices of alliteration and
assonance, emphasis and repetition—often at the expense of
simplicity and truth to nature. A t times it can rise to grandeur
—as in the oath of Achilles (Ach. ii) :

per ego deum sublimas subices umidas,

unde oritur imber sonitu saeuo et spiritu;

or in Thyest. vii :
aspice hoc sublime candens quem inuocant omnes Iouem
(cf. Eur. inc. 935);

but the force of Jason's words to Medea :

ώ? "Ερως σ ήνά-γκασς
Tofot? αφυκτοις τονμον ΐκώσαι Βεμας,

is not improved by the rhetorical jingle in

tu me amoris magis quam honoris seruauisti gratia.
W e are irritated by a Cassandra who begins a passionate
outburst in this way :
mater, optumarum multo mulier melior mulierum.
Roman pedantry and excessive preoccupation with the form
of words shows itself in

amicus certus in re incerta cernitur,

a line thought well of by Cicero—who, when he champions

Ennius, is partly an advocate who feels that he holds a brief
for the earlier poets, partly a fellow-countryman who himself
suffers from the same defects.
Rhetoric is more than a trick of style : it is an attitude to
life. The writer who is excessively emphatic in words is apt
to lay too much stress on immediate effect in thought and
68 THE ROMAN STAGE

emotion. In Roman tragedy there is a tendency to over-

statement ; crude effects are substituted for simplicity and
naturalness. Ennius was a man of broad human sympathies,
possessed of a genuine poetic t a l e n t ; but even he is not free
from that straining after effect which was to become more and
more marked in Roman tragedy. It must be conceded to the
Roman tragedians that they were striving to create a worthy
diction ; if the attempt is momentarily abandoned, the tone
may become commonplace and sometimes, one might say,
class-conscious : compare
η δνσγeveia δ' ώ? e^et τι ·χρήσιμον'
και yap δακρΰσαι ραδίως αύτοίς εχεί
απαντά τ ειπείν τω Be yevvaup φύσιν
ävokßa ταύτα
with
plebes in hoc regi antistat loco : licet
lacrumare plebi, regi honeste non licet.
This class-consciousness may also account for the extraordinary
expansion of Euripides' phrase ' Corinthian women ' into
quae Corinthum arcem altam habetis, matronae opulentae,
optumates !
Perhaps we may sum up these criticisms by saying that Roman
tragedy did not succeed in developing a style which was at
once simple and dignified. That it did not do so is to be
explained quite as much by popular taste in Rome as by the
limitations of the writers of tragedy. The effects which to us
seem forced may have appealed to the listeners for whom they
were designed far more than would the artistic restraint of the
Greek originals.
Cicero admired the tragedies of Ennius almost to adoration,
was never tired of quoting from them and seems to assume
that his audience knows them too. He tells us that the Medea
was widely read in his day. Other Romans seem not to have
ranked Ennius as a dramatist quite as high as his great
successors in tragedy, Pacuvius and Accius. That his plays
were popular in his own day is shown by Plautus' burlesque
(Poen. ι) and Terence's reference ; how frequently they were
acted in the late Republic we do not know.
 C H A P T E R Χ

PAGUVIUS

A R C U S P A C U V I U S , son of Ennius' sister, was born in

220 B.C. at Brundisium. At Rome he occupied himself
with the writing of tragedy (some add satire) and with painting.
He lived on terms of friendship with Laelius. He produced a
play in his eightieth year (140 B.C.), one of the competitors in
that year being Accius, fifty years his junior and his successor
in tragedy. Later he retired to Tarentum, where he was
visited by Accius, then on his way to Asia. Accius stayed with
Pacuvius for a few days, and read his Atreus to the old dramat-
ist. Pacuvius died towards the year 130 B.C. A well-known
picture painted by him was preserved at the temple of
Hercules in the Forum Boarium.
Such is the account we piece together from various sources
of different authority. It was Accius himself, according to
Cicero, who vouched for the fact that Pacuvius and he pro-
duced plays on the same occasion. The connexion with
Brundisium is supported by the form of his name, which,
according to philologists, is Oscan. (We notice that he uses
the Oscan word ungulus for ' ring'.) The elder Pliny's
reference to the existence (in early times) of a picture by him
sounds reliable ; but Pliny does not seem to have seen this
picture himself, or to imply that it was extant in his own day.
If Pliny is right in calling Pacuvius Ennius' nephew, then
Jerome is wrong in saying that he was the son of Ennius'
daughter—which, according to Jerome's own dates, would
make Ennius a grandfather at the age of twenty. (Perhaps
the discrepancy arose from the ambiguity of nepos, which some-
times means ' nephew sometimes ' grandson \) Pacuvius'
friendship with Laelius may be a touch of fiction added by
Cicero to lend human interest to what is itself a fictitious
address ' On Friendship ' put into the mouth of Laelius.
Gellius' anecdote about Accius' visit to Pacuvius and their
discussion of the tragedy of Atreus seems suspiciously like that
6
70 THE ROMAN STAGE

other anecdote which shows Terence reading his Andria to the

elderly Caecilius. As for the well known and rather im-
pressive epitaph which Pacuvius is said to have composed and
caused to be inscribed on his tomb, some suspect that (like the
even better-known epitaphs of Naevius and Plautus) it is the
work of Varro, in whose book ' On Poets' Gellius found all
three. Certainly Gellius does not mention having ever seen
the tombs of any of these dramatists.
Pacuvius was the first Latin author to specialize in tragedy
—if we may venture to disbelieve the evidence of Diomedes
and Porphyrio that he wrote satire also. We possess the titles
of twelve tragedies—Antiopa (from Euripides), Armorum Iudicium
(? from Aeschylus), Atalanta, Chryses, Dulorestes (i.e. ' Orestes
as a Slave'), Hermiona, Iliona, Medus, Niptra (Sophocles),
Pentheus (somewhat different in plot from Euripides' Bacckae),
Periboea (Euripides' Oeneus ?), Teucer (Sophocles?), and a
praetexta, Paulus. The fragments amount to about four
hundred lines.
For so long a career twelve tragedies may seem a small
number. Perhaps he was a slow worker ; perhaps painting
took up much of his time and supplied part of his livelihood.
It is probable that he wrote with great care ; he was celebrated
for his learning, and his language shows elaborate attention
to sound. He was regarded by Cicero, Horace and Quintilian
as the leading Roman writer of tragedy, or second only to
Accius, who is sometimes said to have surpassed him in force.
Many of his plays enjoyed great popularity down to the end
of the Republic.
To what did he owe his success ? His plays are lost, as
well as their Greek originals, and we cannot now estimate the
dramatic power of a tragedy of Pacuvius. He seems to have
preferred plays with complicated plots (taken perhaps from
post-Euripidean originals) and to have specialized in pathetic
scenes. Possibly the most famous scene in Roman tragedy
was the opening of his Iliona. Polydorus, youngest son of Priam,
had been entrusted to the care of his sister Iliona, wife of
Polymestor, king of Thrace. She brought him up secretly as
her own son Deiphilus, pretending to her husband that
Deiphilus was Polydorus. On the fall of Troy Polymestor,
wishing to show his goodwill to the victors, murdered the
supposed Polydorus—really his own son. The play opened as
 PACUVIUS 71
the ghost of the murdered boy rose to implore his sleeping
mother for burial in tones which, as Cicero tells us, moved the
whole audience to tears. In the half-light of early morning the
sleeping form of Iliona was ' discovered ' on the stage ; pre-
sently, out of the recess in the stage (the curtain-slot, it seems,
according to the stage-practice of Cicero's day), the dismal
figure of the ghost appeared. Unfortunately on one notorious
occasion Fufius, the actor taking the part of Iliona, was really
asleep, and failed to hear the appeal of his fellow-actor
Catienus, who was taking the part of the ghost, until the whole
theatre joined in the cry ' Mother, I call to thee
Another favourite scene occurred in the Chryses. Orestes
and Pylades have been brought captive before Thoas ; he
wishes to discover which of them is Orestes, and to punish him
alone. T h e two captives vied with each other in generous
rivalry, each claiming to be Orestes—a contest which at the
first performance of the play brought the whole audience to
their feet shouting their applause. (This is, at any rate, what
Cicero makes Laelius say in the address ' O n Friendship'.)
In tragedy as in comedy the R o m a n crowd had a keen ear
for what seemed a topical allusion. A canticum from the
Armorum Iudicium was declaimed at the funeral of Julius Caesar.
O n e of its lines was particularly apposite : ' that I should have
spared the men who were to murder me ! '
T h e pathetic roles, such as that of Antiopa, gave R o m a n
actors a chance of starring ; thus Cicero tells us that he has
seen Rupilius, who specialized in the part of Antiopa. T h e
angry words of Telamo in the Teucer so inflamed the actor who
spoke them that his eyes seemed to burn out of the mask.
(Cie. de Orat. I I . 46. 193).
It would appear from these references that Pacuvius paid
attention to stage effects. His plays have those touches which
we miss in Seneca. W e can see the actors entering :

atque eccum in ipso tempore ostentum senem !

W e can see them eavesdropping on each other :

sermonem hie nostrum ex occulto clepsit, quantum intellego.

T h e stage-doors open audibly :

. . . quidnam autem hoc soniti est quod stridunt fores ?

72 THE ROMAN STAGE

So far we have been dealing with what Pacuvius derived

from his Greek originals. It is of special interest to notice
what modifications he introduced. As has already been said,
the testimony of Cicero is at first sight contradictory. In one
passage he speaks of Pacuvius' Antiopa (and other Latin
tragedies) as being translated ' word for word Elsewhere he
tells us that Pacuvius, Ennius and Accius rendered ' not the
words but the sense ' of their originals. It seems reasonable to
infer that the Roman writers of tragedy kept fairly close
(closer, perhaps, than the writers of comedy) to the general
sense of the Greek 1 ; but that on occasion they would modify
a passage for a special effect. Thus Cicero tells us that in the
final scene of the Niptra Pacuvius has actually improved on
Sophocles. Sophocles allowed the mortally wounded Ulysses
to express all his agony : Pacuvius makes him more restrained ;
even his mild reference to his sufferings is rebuked by his
stretcher-bearers, and he dies with the Stoic words :
conqueri fortunam aduersam, non lamentari decet :
id uiri est officium, fletus muliebri ingenio additus.
But in other passages, as might be expected, the Latin is more
rhetorical and violent than the Greek. We have already
mentioned the speech (which Cicero so much admired) in
which the aged Telamo bitterly accuses Teucer of having
deserted Ajax. The tone of Sophocles' play, to judge by the
fragment (516.N) is here comparatively restrained. Pacuvius
employs all the resources of his Latin to give a violent effect :
note the sigmatism and alliteration of the lines :
segregare abs te ausu's aut sine illo Salaminam ingredi,
neque paternum aspectum es ueritus, quem aetate exacta indigem
liberum lacerasti orbasti extinxti ?
This attention to sound-effect is equally marked in the
' foot-bath ' scene from which the Niptra took its title. The
aged nurse is about to wash the feet of the stranger (really
Ulysses) :
cedo tuum pedem mi, lymphis flauis flauum ut puluerem
manibus isdem quibus Vlixi saepe permulsi abluam
lassitudinemque minuam manuum mollitudine,
1 I sec no reason to suppose that Pacuvius combined material from two or more

originals. The fragment in his Teucer—patria est tibicumque est bene—is like
Aristoph. Plut. 1151 ; but perhaps Aristophanes himself borrowed it from
Pacuvius' original, or was quoting a popular saying.
 Ρ ACUVIUS 73
lines which at least one Roman reader (Gellius) thought
' delightful' (iucundissimi).
This pictorial quality is further illustrated in a description
of the sailing of the Greek fleet from Troy, and the coming on
of the storm :
flamma inter nubes coruscat, caelum tonitru contremit,
grando mixta imbri largifico subita praecipitans cadit.
Whatever the Greek original may have been, it is the
Latin writer who is responsible for the Latin of such passages
as these. Pacuvius' style is often natural and moving :
О flexanima atque omnium regina rerum oratio !
But he was criticized for weird compounds, as in the line :
Nerei repandirostrum incuruiceruicum pecus,
' The upturnsnouted and roundcrooknecked herd of Nereus'
(Warmington's translation). H e seems to have had a liking for
philosophical discussions : his Antiopa included a debate
between Antiopa's sons Amphion and Zethus on the respective
merits of the artistic and the practical life. Cicero quotes his
rendering of a passage in Euripides' Chrysippus on Heaven and
Earth as the universal parents of all life. The attack of sooth-
sayers which Cicero quotes from the Chryses seems to me Roman
in feeling :

For those who understand the speech of birds,

And learn more wisdom from another's liver
Than from their own, I vote one ought to hear
Rather than heed them.
In the Antiopa Amphion sets the ' townsmen' a riddle :
what is it that is
Fourfooted beast, slow-stepping, haunting fields,
Low set and rude and grim to gaze on. short
In head, snaky in neck, and disembowelled,
Without breath, yet with breathing sound endowed ? 1
When they give it up, he replies in one word—' Shellback'
(a pun on the two senses of testudo, ' tortoise' and ' harp '
—a pun which is also possible with the Greek chelus).
1
Warmington's translation.
74 THE ROMAN STAGE

Perhaps these passages helped to earn Pacuvius his reputa-

tion for ' l e a r n i n g B u t in a dramatist this quality may
degenerate into pedantry. Even in the well-known lines
describing Heaven and Earth as the universal parents—lines
translated from the Greek, and supposed to be uttered by a
Greek—he must needs turn aside to remark that what the
Greeks call aether is called caelum by ' our people the Romans.
Even Cicero displays uneasiness at lapses such as this. It is
Cicero, too, who tells us that Pacuvius' Latinity was poor.
A later generation of readers turned away from Pacuvius ;
they found his style harsh, his pathos wearisome, and grew
somewhat tired even of his ' l e a r n i n g B u t his popularity
on the stage for more than a century reflects credit both on
his dramatic power and on the ability of the Roman public
to appreciate a serious play.
 C H A P T E R X I

C O M E D Y AFTER THE DEATH OF PLAUTUS

T H E P R O L O G U E to Plautus' Casina refers expressly to

a revival performance of the play after the author's
death. T h e speaker remarks that wise judges will prefer
old wine and old plays—especially as the contemporary
comedies are worthless. In view of the general demand for
the plays of Plautus, his company is therefore reviving the
Casina, the premiere performance of which is still within the
memory of the older members of his audience. ' A t its first
appearance the play surpassed all other plays ; and yet at
that time there were living a garland of poets (flos poetarum)
who have now departed to the bourne where all must go.'
O f course the speaker of the prologue is bound to praise
the wares that he has to sell. He is claiming special merit
for the fact that he is offering not a new play but an old one.
Terence, we notice, takes care to emphasize that his plays are
new. In either case the object is to win the favour of the
public. W e have other evidence, however, that Plautus' plays
were immensely popular after his death. T h e ' garland of
p o e t s ' contemporary with Plautus must include Naevius ;
possibly also Ennius (though he seems not to have been very
successful in comedy) and if Ennius, then presumably Caecilius
too, as he is said to have died only a year after Ennius. But
to add Terence to the flos poetarum would put the date of the
Casina revival later than 160 ; this would presumably mean a
gap of thirty years or more between the first and second
performances of the p l a y — a n d in that case even the older
members of the audience could scarcely have seen the premiere.
T h e alternative is to include Terence, and perhaps Caecilius
also, among the writers of the worthless new comedies ; and
we know that both these dramatists had difficulty in winning
a hearing for their plays.
T h e success of Plautus with the crowd had evidently left
the public with an appetite for comedy, and at the same time
75
76 THE ROMAN STAGE

set a standard for later dramatists which they found it hard to

reach. W e now have a world of keen competition between rival
dramatists as well as rival troupes of actors. If we accept
Festus' statement that a guild for scribes and actors had been
founded in Andronicus' d a y — t h e collegium poetarum,—we
can imagine that the dramatists who met there discussed
their problems and the principles of their art. It was for them
either to rival Plautus on his own field, or to find new ways
to public favour. O n their solution of this problem their
livelihood would depend.

CAECILIUS

W e are told by Jerome that Caecilius was an Insubrian

Gaul (some gave his birthplace as Milan) and by Gellius that
he began his career in Rome as a slave. It is conceivable that
he had been made captive in one of the wars which the Romans
waged with the Insubrians about the years 222-219 B.C.
Statius, ' attendant', was a name often given to slaves;
Caecilius would no doubt be the name which he adopted from
his master on manumission. Looking again at Gellius, we
may suspect that he is trying to explain the poet's possession
of a slave-name ; brought to R o m e as a slave, the poet was
' therefore ' called Statius. Perhaps the slavery is merely an
inference from the name ; if so, it is not convincing ; not
every one called Statius was or had been a slave. If there is
anything in the story that Ennius, who prided himself on his
R o m a n citizenship, shared lodgings with Caecilius, we m a y
infer that Caecilius, like Ennius, was of citizen rank. It is
only too probable that nothing was known in Jerome's day
about his early career. M u c h more weight must be given to
Terence's assertion that Caecilius' early efforts as a playwright
had met with opposition and disappointment, but that the
actor-manager Ambivius Turpio had discerned his promise
and, after some failures, had succeeded in winning public
favour for his plays. A doubtful anecdote brings him into
direct contact with Terence, who is said to have been sent by
the magistrates to read his first play, the Andria, to Caecilius,
then the acknowledged master of comedy. T h e Andria, how-
ever, was not produced until 166, and Caecilius is said by
Jerome to have died in 168, the year after the death of Ennius,
and to have been buried near Janiculum.
 COMEDY AFTER THE DEATH OF PLAUTUS 77

We possess the titles of forty-two comedies : Aethrio, Andrea,

Androgynes, Asotus, Chalcia, Chrysion, Dardanus, Dauos, Demandati,
Ephesio, Epicleros, Epistathmos, Epistula, Ex Hautu Hestos, Exul,
Fallacia, Gamos, Harpazomene, Hymnis, Hypobolimaeus siue Sub-
ditiuos, Hypobolimaeus Chaerestratus, Hypobolimaeus Rastraria,
Hypobolimaeus Aeschinus, Imbrii, Karine, Meretrix, Nauclerus,
Nothus Nicasio, Obolostates siue Faenerator, Pausimachus, Philumena,
Plocium, Polumeni, Portitor, Progamos, Pugil, Symbolum, Synaristosae,
Synephebi, Syracusii, Titthe, Triumphus. The fragments amount
to 280 lines or part of lines.
With Caecilius we come to a crucial point in the history
of Latin comedy. Naevius and Plautus had treated their
originals with considerable freedom, had Latinized the titles
and had scattered allusions to Roman places and customs
through the plays. Caecilius gives us no topical references,
and even the titles of the plays are usually left in their Greek
form. After his time Greek titles are invariable. Varro's
assertion that Caecilius excelled other comic writers in plot
(<argumenta) presumably means that he chose well-constructed
originals. It appears that he was specially interested in
Menander : sixteen of his titles are also found in the list of
Menander's plays, Cicero speaks of Caecilius and Terence as
translators of Menander, and Gellius quotes a series of passages
from Caecilius' Plocium for comparison with the original by
Menander. Here again Caecilius stands half-way between
Plautus, with only three certainly Menandrian plays in
twenty-one, and Terence, with four Menandrian plays out
of six. Menander's plays, if we may judge by what remains of
them, were of a somewhat serious tone, and some evidence
would suggest that Caecilius shared this quality. Horace says
that Caecilius was considered to be distinguished by moral
earnestness (grauitas) ; Varro attributed to him emotional
power ; Cicero quotes him in court to illustrate his own
remarks on the relationship of fathers and sons. On the other
hand Velleius finds in him a rich vein of Latin wit (dulces Latini
leporis facetiae).
Fortunately we are enabled in his case, as in that of no
other Latin comic writer, to set extensive passages of the Latin
beside their Greek original. Gellius, wishing to illustrate his
general view that Latin comedies, though well enough in
themselves, were altogether inferior to their originals, takes
78 THE ROMAN STAGE

the Plocium as an example ; the Latin, he says, lags behind

the Greek of Menander in subtlety and truth to life, in delicacy
and style.
In the Plocium we have a familiar domestic situation.
Crobyle, a shrewish, suspicious, purse-proud wife, tyrannises
over her husband. The first passage quoted by Gellius from
Menander shows us the husband complaining to a friend.
What particularly galls him is that he has been forced to
dismiss a pretty servant-girl who has aroused Crobyle's
suspicions. Caecilius turns this passage from plain speech into
a canticum varying in metre from trochaic septenarius to
bacchiac and then to cretic. The main sense is unchanged ;
the expression is so different that we should not have recognized
the Latin as a translation of the Greek. The Latin writer has
thought out the whole situation for himself. Certainly his
version is coarse, but is it not lively and vigorous too ? Instead
of the rather too facile grace of the Greek we get a crude,
vivid energy, driving home its meaning with the hammer
blows of rhetoric. Roman literature excels in brief passages
of dramatic dialogue. Where Menander simply gives us the
husband's complaint, Caecilius' husband makes us hear the
nagging accents of Crobyle as well :
ita plorando orando instando atque obiurgando me optudit.
We have a vivid picture of her talking to her cronies, gloating
among them over her triumph :
quis uostrarum fuit integra aetatula
quae hoc idem a uiro
impetrarit suo quod ego anus modo
effeci, paelice ut meum priuarem uirum ?
A reference (reminiscent of Plautus) to Crobyle's unpleasant
breath is inserted by Caecilius without the slightest justification
in the Greek. Gellius observes that Caecilius preferred to
raise a laugh instead of giving his characters only such words
as they might naturally have been expected to utter. Elsewhere
he remarks that Caecilius' additions are akin to the style of
the mime. (Cf. Volcacius Sedigitus : Caecilio palmam Statio
do mimico).
Gellius is speaking of Caecilius as typical of Latin dramatists
in general : they are all well enough, considered in them-
selves, but their inferiority is at once apparent when they are
 COMEDY AFTER THE DEATH OF PLAUTUS 79

compared with their originals. We have other grounds for

assuming that Plautus and Naevius departed widely from
their originals ; besides altering speech to ' s o n g t h e y too
introduced every kind of rhetorical device, and they missed no
chance of raising a laugh, regardless of dramatic proprieties.
It was perhaps this coarse, lively quality in Caecilius' work
which seemed to Vellerns typical of Latin wit. As for his
reputed grauitas, Cicero's citation of an angry father's out-
burst may serve as an example :
egone quid dicam ? quid uelim ? quae tu omnia
tuis foedis factis facis ut nequiquam uelim.
istam in uicinitatem te meretriciam
cur contulisti ? cur inlecebris cognitis
non . . . refugisti ?
. . . cur alienam ullam mulierem
nosti ? . . .
dide ac dissice,
per me licebit . . .
. . . si egebis, tibi dolebit, mihi sat est
qui aetatis quod relicuom est oblectem meae.

Though the theme is a familiar one, the depth of feeling

revealed in this outburst seems to exceed that of similar
passages in Plautus and Terence.
We may wonder why Caecilius (the greatest Latin comic
writer, according to Volcacius Sedigitus and perhaps to
Cicero) met, like Terence, with opposition and public neglect
at the outset of his career. Was there something new in his
plays which at first repelled the public while attracting the
discerning actor-manager Ambivius Turpio ? Such a new
quality there certainly was in Terence's work : Terence,
as we shall see, was an original artist with principles of his own,
which he maintained although he knew they were not as yet
fully appreciated by the ' stupidus populus'. We have no such
evidence in the case of Caecilius. Though he does not seem
to have introduced the topical references to Roman things
so dear to Plautus—and so out-of-place in translated comedy
—he was not above adding crude jokes of which Terence
would have been ashamed. All we can say is that he left his
titles for the most part in their Greek form, that he seems to
have been something of a moralist, and that he was specially
attracted by Menander. Menander was the most refined
8o THE ROMAN STAGE

author of New Comedy, but in his own day he was not the
most successful on the stage. Whatever the new quality in
Caecilius' work, the public seems eventually to have come to
like it, even in his own lifetime (though we must never yield
too credulous an ear to Terence, whose object in making this
assertion is to persuade the public that in time they will like
his plays too). If there really was in Caecilius a tendency to
choose originals of a more serious type, in particular plays by
Menander, and to avoid any topical references which would
clash with the context, he had at least set foot on the road
which was to lead to the more fully Hellenized palliata of his
successors.
 C H A P T E R XII

TERENCE

ITH TERENCE we reach the climax of our story.

W Almost all the problems of Latin drama are raised by
the study of his prologues and plays.
O u r histories of Latin literature tell us that Publius Teren-
tius Afer was born in Carthage about the year 185 (some say
195) B.C. By birth he belonged to the native Libyan stock.
I n his boyhood he came to Rome as a slave of the senator
Terentius Lucanus. A t an early age he displayed not only
personal beauty but intellectual promise, and his master,
won by these qualities, gave the young ' A f r i c a n ' a good
education and set him free. As was usual for manumitted
slaves, the boy kept his slave-name Afer as his cognomen and
added it to the gentile name of his former master. H e was now
a youth of medium stature, slight build and dark complexion.
His abilities won him the friendship of some of the noblest
Romans of his day, especially Scipio Africanus Minor, who
was to achieve the final destruction of Carthage, and C .
Laelius, the chief figure in Cicero's dialogue ' O n Friendship
Thus the young Terence found an entry not only into R o m a n
society but into one of the most famous literary circles of the
ancient world, the group of writers and lovers of literature
who gathered round Scipio. T h e first of Terence's plays was
the Andria ; when he offered this to the aediles, he was told to
read it to the elderly dramatist Caecilius. Caecilius was at
dinner when the rather shabbily-dressed young stranger
arrived ; on hearing what he wanted, he told him to sit on
a stool and begin to read. T h e opening lines so impressed
Caecilius that he invited Terence to share his dinner, and
afterwards heard the play through with the greatest admira-
tion. T h e poet was now embarked on his dramatic career,
which was, however, destined to bring difficulty and dis-
appointment as well as brilliant and even unprecedented
success. T h o u g h encouraged by his noble friends, who were
81
82 THE ROMAN STAGE

even said to help him in the composition of his plays, he

found an enemy in a certain elderly playwright named
Luscius Lanuvinus, who spread slanderous rumours about
Terence's methods of composition and even caused a scene
at one of the rehearsals by accusing him of having stolen two
characters from an old Latin play. Terence was particularly
unlucky with his Hecyra, which twice failed owing to the
machinations of his enemies, who spread rumours among the
audience that more attractive entertainment was at that very
moment to be found elsewhere—rope-dancing, boxing and
so on. However, the steady support of the actor-manager
Ambivius Turpio achieved final success for even the Hecyra at
its third performance, in 160 B.c. Next year Terence sailed
for Greece to study Greek life and collect more Greek plays
for translation ; but some mischance befell him on his home-
ward voyage, and Rome never saw him again.
This is an interesting story. No other Latin dramatist
seems so well known to us as Terence. In particular his
personal and literary friendship with Scipio has caught the
imagination of later ages. But when we examine the sources
of our information, we soon find that the traditional account
is not always consistent with itself.
It is mainly derived from the Life of Terence written by
Suetonius in the second century A.D. and quoted at the
beginning of Donatus' commentary to Terence's plays.
Suetonius himself makes it clear that many of the details of the
Life were questioned by Roman scholars, whose conflicting
views he quotes. They had questioned the date of Terence's
birth. They had pointed out difficulties in the account of his
arrival in Rome from Africa. They had argued that the noble
friends whom he mentions (without naming) in his prologue
could not have been Scipio and Laelius, who were at the time
too young to be described in the terms which Terence uses,
and they had suggested other and perhaps more likely names.
They had fallen into complete contradiction about Terence's
financial circumstances, about the reason for his voyage, about
where he went and how he died. The most significant feature
of the controversy is that we find the controversialists citing
in support of their views no first-hand authority except Terence
himself. We begin to wonder whether any records of Terence
had survived to later ages except the text of his plays and
 T E R E N C E 83
prologues, with some notes of early performances. These
notes may have been contained in official records of public
games, but it is also conceivable that they had been jotted down
in the manuscripts of the plays. Indeed Suetonius tells us that
the unprecedented fee paid for the Eunuchus was entered on the
title-page of the manuscript. Here we have the origin of the
didascaliae which we still find prefixed to the plays of Terence
in our medieval manuscripts and modern printed editions.
T h e poet's name would appear on the title-page : we hear
nothing of disputes about authorship in regard to the plays of
Terence. But when we consider the utterly uncritical methods
of R o m a n scholarship, an uneasy suspicion may steal upon us
that all that survived of Terence was his manuscript, in which
notes of early performances had been inserted, and that from
study of what was contained in the manuscript, the prologues,
the didascaliae and perhaps the plays themselves, and finally
from pure invention, was constructed the ' L i f e ' of the
author.
It is always possible that the tradition contains some facts,
even if in origin these were derived from nothing but a shrewd
guess. But other guesses may not have been so shrewd. T h a t
Terence really was acquainted with the family of Scipio is
suggested by the statement in the didascalia that the Adelphi
was produced at the funeral games of Scipio's father, L .
Aemilius Paulus, in 160 B.C. T h a t he was born in Carthage
and came to R o m e as the slave of Terentius Lucanus may have
been an inference from his name, Terentius Afer ; if so, the
inference was unsound. T h e cognomen Afer is in itself no proof
of African blood, still less of slave origin, as is shown by the case
of Domitius Afer of Nimes, praetor in A.D. 25. (I have already
suggested that Gaecilius Statius was inferred to have been a
slave merely because the name Statius had servile associations.)
Fenestella pointed out chronological difficulties in the supposi-
tion that Terence had come from Africa ; and the very fact
that such difficulties could be pointed out suggests that there
was no incontrovertible evidence about Terence's birthplace.
I f he really was a slave, who would have troubled to record the
date of his birth ? T h e supper-party with Caecilius is difficult
to c r e d i t ; apart from the surprising detail that Terence,
friend of noble Romans, could not dress himself decently for
an important interview, the chronology is inconvenient:
84 THE ROMAN STAGE

Caecilius is said to have died in 168, and the Andria does not
seem to have been produced till 166. T h e conflict of evidence
or of views as to the identity of Terence's noble friends, as to
which of them helped him with his composition, as to why he
went abroad, where and how he died, and whether he died in
abject poverty or left an estate, suggests that nothing was known
about these matters. Evidently the records of his plays stopped
with the year 160. If a theory had to be invented to explain
the early cessation of so brilliant a career, what theory so
satisfactory as that of a voyage from which he was never to
return ? T h e very idea of a fatal voyage may have been
suggested by a passage in his last play, the Adelphi:

periit, abiit, nauem ascendit (' he has gone on board ship, he

has sailed away, he is no more.')

H o w such legends grow may perhaps be traced in the tale

of Terence's friendship with Scipio. Terence himself, personal
though his prologues are, takes care to name no living person—
not even himself. Accused of being aided in his work by noble
friends, he retorts that there would be no discredit in accepting
help from men to whom the whole state is indebted. Perhaps
the first extant reference to the subject after Terence's death is
contained in the bitter verses of Porcius Licinus, a violent
democrat of the first century B.C. Porcius is anxious to prove
that friendship with the great does a poor man no g o o d —
witness, he says, Terence's relations with Scipio, Laelius and
Furius,' who were the greatest nobles of his day and who left
him to die in abject poverty. Porcius is apparently supplying
these names from Terence's reference and his own view as to
who the most likely people would be ; no doubt in this attack
on the aristocracy the unpopularity of Scipio among the
popular party would make his name particularly appropriate.
Porcius gives the story a scandalous twist by accusing Terence
of having been the boy-favourite of Scipio and the others.
This we can hardly believe ; Scipio was born about 185, so
that Terence, who produced the Andria only nineteen years
later, cannot have been younger than Scipio and may quite
possibly have been older (a point made by even R o m a n
commentators). However, to the attempt to represent Terence,
the boy-favourite of Scipio, as at least no older than his patron
we may trace the commonly accepted dating of Terence's
 TERENCE 85
birth as 185. Other commentators argued that Scipio, Laelius,
and Furius were too young to suit Terence's description and
suggested what they thought were more likely names—again
with no authority except Terence's own words—but, as is the
way of the world, it was the most interesting names which
stuck in people's minds. Cicero takes the friendship with
Scipio and Laelius for granted, and further accepts the story
that they had assisted Terence in the composition of his plays.
Later writers are still more precise (though their accounts do
not agree with each other) ; they tell us which passages were
written by Terence's friends, they explain the circumstances
under which these passages were written, or even go so far as
to deny Terence any share in the authorship at all. How
could a person of such humble origins have written such
elegant Latin ? Clearly the real author must have been Scipio
(or Laelius). The sycophants of comparatively modern times
were attracted by a tale so flattering to the nobility, and thus
a legend which seems to have originated in class-conscious
hatred of the rich has been preserved by snobbery.
The only demonstrable connexion between Terence and
the family of Scipio is the fact that two of Terence's plays
were produced at the funeral games of Scipio's father, L.
Aemilius Paulus, in 160 B.C.
Having indicated certain suspicious features in a biography
which seems to have been largely constructed from Terence's
prologues, we have still to consider the prologues themselves.
Their authenticity is unquestionable ; the very fact that they
name no contemporaries is decisive against their having been
the work of a forger. In these prologues we have the personal
utterance of the author, defending himself against the attack
of the ' malicious old dramatist' whose identity he indicates
sufficiently by saying that he is the translator of Menander's
Phasma and Thesaurus. (As these translations survived, Roman
scholars had no difficulty in supplying the name of Terence's
enemy as Luscius Lanuvinus.) Can we ourselves hope to
reconstruct the dramatic career of Terence from the prologues
and the records of first performances ? Unfortunately first and
later performances are confused in our records, so that there is
hardly any agreement as to the order in which the plays were
written and produced, while Terence's own statements, made
in the course of bitter controversy, were intended to mislead
7
86 THE ROMAN STAGE

the public of his own day and have in fact misled readers of all
subsequent generations.
It is admitted by all that the order of Terence's plays is
uncertain. Dziatsko arranges the plays in this order of
performance :

Ι. Andria, at ludi Megalenses (April), 166 B.C.

2. Несутa, first performance, ludi Megalenses, 165 B.C.
(a failure).
3. Heautontimorumenos, ludi Megalenses, 163 B.C.
4. Eunuchus, ludi Megalenses, 161 B.C.
5. Phormio, ludi Romani (September), 161 B.C.
6. Adelphi, ludi funebres of Aemilius Paulus, 160 B.C.
7. Hecyra, second performance, ludi funebres of Aemilius
Paulus, 160 B.C. (again a failure).
8. Hecyra, third and successful performance, perhaps ludi
Romani, 160 B.C.

Lindsay's text arranges the plays thus : Andria, Heautonti-

morumenos, Eunuchus, Phormio, Hecyra, Adelphi. Other writers
suggest still other arrangements. It is agreed that the first
play was the Andria, produced in 166 B.C. (though the
prologue to the Andria might suggest that the writer has
already had a disappointment, whether with this or with some
other play), and that Terence's dramatic activity ended in
160 B.C., the year in which the Adelphi was produced. T h e
prologues to the Hecyra make it clear that this play failed twice
before its final success. But to try to form a satisfactory theory
as to the order of composition and production, a theory which
will reconcile or explain all the real or apparent contradictions
in our evidence, is to enter into a tangle of inconclusive
argument. Furthermore, to found any theory as to Terence's
literary development on the supposed order of his plays may
only too easily lead us to argue in a circle, using our pre-
conceived ideas of Terence's development to support our theory
of the order of his plays.
Let us approach the prologues of Terence, these baffling
but precious personal documents, with as open and yet as wary
a mind as possible. Let us remember that their order, like that
of the plays for which they were written, is to a large extent
unknown. Let us further remember that these prologues are
the ex parte utterances of a writer whose lack of candour is
 TERENCE 87
universally admitted, that they were written for delivery to an
uncritical crowd, whose suspicions they were designed to
allay for the moment so that an attentive hearing might be
secured for the play which was to follow, and that we have not
got the case of Terence's opponents. Critically examined, the
prologues reveal a literary and personal history of extraordinary
interest.
Terence's prologues are different from those of previous
dramatists. Up to now the main function of the prologue had
been to explain the dramatic situation to the audience ; Latin
writers had also used the prologue to put the audience in a good
humour and a readiness to give attentive hearing to the play.
Terence dropped the explanatory prologue altogether. Evi-
dently he thought that to make the situation reveal itself in the
course of the dialogue was preferable to giving the audience
a direct explanation. Here we have perhaps for the first time
a Latin dramatist consciously setting before himself an artistic
principle. 1 Terence was indeed a conscious dramatic artist,
perhaps the only one in the history of Latin drama. The other
use of the prologue, to put the audience into a favourable
humour, he would no doubt also have abandoned, had not
the attacks of his enemies forced him to use his prologues for
self-defence. The polemical prologue may have been his own
creation ; at any rate, all his prologues are polemical through-
out. They are all replies to charges brought by his enemies
with reference to his methods of composition ; indeed he
quotes his enemies as saying that but for their attacks he would
not have had any material out of which to make prologues at
all. He gives no explanation of the dramatic situation ;
he does not crack jokes with his audience ; his whole aim is to
prevent their minds being prejudiced against his play as a
result of the rumours which his enemies have spread.
What these rumours were we can only infer from Terence's
replies. The details of this literary controversy seem to have
been soon forgotten. Though Terence's prologues descended
with his plays to later generations, such references as we find
to them in later authors show no real knowledge of the points
at dispute. Donatus' commentary is composed merely of
1
There is perhaps some hint of a similar attitude in the prologues to the
Trinummus and Vidularia. But more probably Plautus is thinking of the reactions
of his audience ; he is anxious to assure them that the play will begin without delay.
88 THE ROMAN STAGE

superficial inference from the very lines which he professes to

explain. This point has not been sufficiently appreciated ;
modern writers have failed to distinguish between what is
valuable in Donatus and what is worthless, between his citations
of documents which are no longer extant and his uncritical
remarks on lines which we can read and interpret for ourselves.
He was an honest if unintelligent student; he had access to
the Greek originals from which Terence drew, and on frequent
occasions he quotes from or refers to these originals. Here his
evidence is trustworthy and precious ; but his attempts to reveal
the mind of Terence are infantile.
We gather from Terence's own words that he was accused
of weakness of style, of accepting literary help from others,
of stealing characters and passages from old Latin plays, of
' contaminating' his originals. Terence's style needs no
defence. That he accepted literary help from his noble
friends is unlikely in itself (though his style no doubt owes
something to the tone of the society in which he moved)
and has little relevance to our opinion of the plays as we
have them ; we can understand, however, that tact would
forbid him to give a direct rebuttal to this charge. The other
two charges, however spiteful, have in common one point
of undoubted fact—that Terence deliberately departed from
his Greek originals.
In translating the Eunuchus of Menander, Terence seems to
have added two characters—a braggart captain and a parasite.
H e sold this play to the aediles as a translation of Menander's
play, saying nothing about the addition which he had made.
It was arranged that a rehearsal should take place in the
presence of the magistrates. Luscius Lanuvinus managed to
gain admittance to this rehearsal. The performance began.
Suddenly Luscius jumped to his feet and cried out that Terence
was a thief: he had stolen the captain and the parasite from
an old Latin play, the Colax, translated from the Greek by
Naevius and (? or) Plautus. The charge of theft was an
ugly one ; we can understand that it was calculated to dis-
turb the minds of the aediles, whose interest in Terence's
play (for which they had paid) was of a practical rather than
of a literary character. In the prologue written for the public
performance of the play Terence recounts the incident and
replies to the charge. His defence is that the Colax was, in its
 TERENCE 89
original Greek, a play by Menander. From this Greek play
he had, he admits, borrowed the captain and the parasite
for his Eunuchus ; but he had not known that Menander's
Colax had already been translated into Latin. And if dramatists
are to be debarred from using stock characters a second time,
must they not also be debarred from using stock situations,
emotions, etc. ? and how then are they to compose a play
at all ? This reply is of particular interest. Terence admits
that his critics were right in tracing his captain and parasite
back to the Colax ; they erred only in supposing that he had
added these characters from the Latin Colax, whereas he had
in fact taken them from its Greek original, the Colax of
Menander. Opinion differs as to which side is here telling
the truth, Terence or his critics ; what nobody seems to have
noticed is the difficulty of supposing that the characters of
fiction can be transferred bodily from one work into another.
Every one has assumed that where Terence agrees with his
critics we may regard his statements as true.
Terence is evidently anxious to show that he had not taken
material from an old Latin play. His defence is that the
material in question was taken from a Greek play, and that
he had not known of the old Latin version. Similarly, in the
prologue to his Adelphi, translated from the Adelphi of Menander
(of course not that Adelphi from which Plautus derived his
Stichus) he tells us that he has added a scene fromi a play by
Diphilus, the Synapothnescontes. This play had already been
translated by Plautus under the title Commorientes, but Terence
asserts that Plautus had ' carelessly ' omitted the scene in
question. Again, in the Andria, translated from Menander's
Andria, Terence admits having added unspecified material
from Menander's Perinthia. Evidently, then, Terence was in
these cases willing to admit that he had departed from his
ostensible originals—but only to the extent of inserting material
taken direct from other Greek originals, material not (so far
as he knew) appropriated by previous Latin writers. This, if
true, would be an answer to the charge of theft; but is it an
answer to that other charge brought against Terence, that of
' contamination ' ? To consider what was meant by this charge,
let us turn again to what is agreed to be Terence's first play,
the Andria. In the prologue to this play we find him defending
himself against a taunt that he had ' contaminated ' his original.
9° THE ROMAN STAGE

His explanation is that there were two very similar plays by

Menander, the Andria and the Perinthia ; alike in plot, they
differed merely in language. In translating the Andria he had
felt free to add what he thought fit from the Perinthia. This,
he says, is what his enemies objected to when they said that
plays should not be ' contaminated'. Their accusation
showed their ignorance of the fact that Terence had merely
followed the example of Naevius, Plautus and Ennius, whose
' carelessness' Terence himself prefers to the ' laborious
dullness' of his enemies and rivals. In the. prologue to the
Heautontimorumenos Terence returns to this point ; his enemies,
he says, have accused him of ' contaminating' several Greek
plays in making a few Latin plays, but he claims ' good '
precedent for his methods, and does not propose to abandon
them.
What does this word contaminare mean ? Its ordinary sense
almost everywhere else in Latin, including Terence himself,
is ' sully ', ' stain ', and this is how Donatus, in his commentary,
explains i t ; but when he comes to interpret the words ' plays
should not be contaminated ' he remarks ' the reference is to
making one play out of several.' This second explanation
appears, however, to be merely an inference from Terence's
own words in the prologue to the Heautontimorumenos. It is
unlikely that a common Latin word meaning ' sully ', ' spoil',
should in two passages, and two only, bear the neutral
sense ' combine'. Taking the word in its ordinary sense,
we see that the sense of the Andria prologue is ' I have added
to my original passages from another Greek play. My enemies
object to this, arguing that (Greek) plays should not be
spoiled.' The accusation was that Terence had altered his
Greek originals and in doing so had spoiled them. His reply
contains two points : his alterations consist merely in the
addition of material from a second and very similar Greek
play, and (in making alterations at all) he can claim the
precedent of the great writers of the past, whose carelessness
he prefers to the laborious dullness of his rivals. These two
points are logically quite distinct; adding suitable material
from a second Greek source has nothing to do with ' careless-
ness '. Terence does not admit that either he or his pre-
decessors had ' contaminated ' their originals ; that was the
unpleasant word which his enemies had used of his methods
 TERENCE 91
(and which he quotes in Oratio Obliqua), but of course it would
have been absurd for him to say ' Yes, I, like the authors
whom I admire, have spoiled my originals What he admits,
in effect, is that he has not kept strictly to his originals ; when
he claims the precedent of earlier writers, he means that they,
too, did not keep strictly to their originals ; but whereas
their infidelities were, as he says, the result of ' carelessness ',
his alterations consisted merely in adding suitable material
from a second Greek source—a process which must obviously
have been deliberate.
Terence is here in a dilemma. The real reply to the charge
of altering his originals was that by altering he had tried to
improve them. But if he had made this reply his enemies would
have seized on it as an example of intolerable arrogance—
indeed, of sheer dishonesty. How dare a translator try to im-
prove on his original ? On the other hand Terence could
hardly agree that his alterations had been for the worse. He
therefore tries to make his infidelities seem as innocent as
possible—they consist, he implies, of mere verbal borrowings
from a second and very similar Greek play by the same author.
He adds that the great Latin dramatists of the past had also
treated their originals with freedom. Here he is skating on thin
ice ; the more he protests his fidelity to Greek sources in general,
the more unlike his methods will appear to those of the great,
' careless ' masters of the past. However, inconsistency between
his two lines of defence would not matter provided he could
leave his audience with the general impression that, after all,
this young dramatist had done nothing either very shocking or
very unusual. His critics were a different matter ; but he did
not fear them so long as they did not manage to prejudice the
public against him. Evidently they did not drop their
accusation. Terence's defence, they said, merely came to this :
in composing a Latin play he had spoiled not one Greek
original but two. To this Terence feels forced to reply in the
prologue to the Heautontimorumenos—but in the briefest and
vaguest of terms, as we have seen. Debarred from giving the
one honest reply, he is forced here, as elsewhere, to take refuge
in obscurity.
One point on which there was never any doubt was the
meaning of contaminare. What Terence cannot have foreseen
was that commentators for whom Latin was an acquired
92 THE ROMAN STAGE

language should have inferred from these two prologues that

contaminare could sometimes bear the special sense ' combine
Unfortunately modern scholars have understood Donatus also
to be offering this as a possible interpretation of the word, and
they have supported it by arguments based on etymology.
Hence it is that ' contamination ' is now a recognised term in
literary criticism, denoting the use of two or more originals in
the making of a single derivative work. From this error
scholars have been led into another for which they cannot
claim the support of either Donatus or any other ancient
writer ; they have supposed that it was this practice of com-
bining originals which Terence ascribed to his predecessors.
Consequently all discussions on the methods used by the Latin
dramatists give some space to considering how far each writer
fused different originals together. But from what has just been
said it should be clear (i) that there never was in Latin
literary history a recognized method of composition by fusion
of originals, (2) that the term contaminatio, ' spoiling', could
not possibly be used as a technical term for any recognized
literary method whatever. 1
O f course there was such a thing as plagiarism. Un-
acknowledged borrowing from various sources was a frequent
practice among ancient writers. Afranius tells us frankly that
his supposed ' native ' comedies contained material borrowed
from whatever source seemed suitable, whether Greek or
Latin. Few R o m a n dramatists would have felt qualms in
inserting in their plays, whether translations or original
compositions, an effective line or passage which they had come
across in their reading. But this haphazard procedure was
never recognized as a specific method of literary composition.
O f large-scale fusion of originals, of the combination of two
Greek plots into one Latin plot, we have no examples, even in
Terence. Terence says, it is true, that he inserted material
from the Perinthia into his Andria. Donatus gives us to under-
stand that he read through Menander's Andria and Perinthia,
presumably to find examples of these borrowings, and failed
to find them. Terence tells us that the two Greek plays were

1 T h e opposite of contaminatus is integer (cf. Cie. 7 o p . 69 Integra contaminatis,

iucunda minus iucundis, and perhaps the much-debated Heaut. 4, ex Integra Graeca
integrant comoediam hodie sum acturus, ' an unspoiled comedy taken from the unspoiled
Greek,' i.e. a straightforward translation).
 TERENCE 93
practically identical in plot, and different only in language.
Apart from the opening scene, Donatus could find only two
short passages in which the two Greek plays resembled each
other. Donatus was naturally puzzled ; why should Terence,
he asks, accuse himself of doing something which he has
apparently not done ? As it never occurs to Donatus to doubt
his author's good faith, he can only suggest that Terence means
that his alteration of the opening scene (a monologue in
Menander's Andria) into a dialogue was made in imitation of
the Perinthia. In Menander's Andria the old gentleman
appeared alone ; in the Perinthia the old gentleman was shown
conversing with his wife ; in Terence's Andria the old gentle-
man is shown conversing with his freedman. W e see from this
statement that Terence has introduced into his opening scene
a prosopon protatikon—a character who will never appear again
in the course of the p l a y — i n order to convert a monologue
into a dialogue. Once more we see the conscious dramatic
artist at work ; the procedure of Terence here is fully in keeping
with what we gather elsewhere of his methods. Donatus'
evidence with regard to what he found in Menander's play
is no doubt accurate. But it is difficult to accept his suggestion
that Terence's invention of the freedman in his Andria is based
on the appearance of the wife in Menander's Perinthia. T h e
dialogue between Terence's senex and his freedman must have
differed from the dialogue between Menander's senex and his
wife. T h e idea of using dialogue form instead of monologue
may, for all Donatus or anyone else can tell, have occurred to
Terence independently. T h e one example of borrowing from
the Perinthia which Donatus thinks he has discovered turns
out to be no example at all.
Plainly Donatus was not satisfied with his explanation. H e
read on, still looking for passages borrowed from the Perinthia,
and could find none ; but he did find a passage in Terence's
Andria (a three-line illustration of general application) which
reminded him of a passage in Menander's Eunuchtis1 (un-
fortunately he does not quote the Greek) and he exclaims with
obvious relief ' this is what was meant by that remark about
" spoiling plays " ', hoc est quod dicitur ' contaminari non
decere fabulas'.
It would seem that Terence's borrowings from the Perinthia
1 Men. 190 K . does not correspond with the passage in Terence.
94 THE ROMAN STAGE

cannot have been very extensive or obvious. Terence's own

words suggest that they were merely verbal—because in plot,
according to him, the two Greek plays were almost identical.
Why then does Terence lay such stress on the fact that he has
borrowed from a second Greek play ? Perhaps because he
wishes to divert attention from something more serious. He
has made other alterations, not by borrowing but by free
invention. These are the alterations which gave the real
grounds for the charge that he had ' spoiled ' his originals.
We owe to Donatus the information that the characters of
Charinus and Byrria, the second lover and his slave, have been
added by Terence, because they ' are not in Menander
The plot of Terence's Andria is that Pamphilus, in love with
the poor girl Glycerium, is being forced by his father to marry
Philumena, daughter of their next-door neighbour Chremes.
Pamphilus has never even seen Philumena ; we find him
wondering whether she is a monster of ugliness, whose parents
are trying to get rid of her. It is of course quite in keeping
with New Comedy and with Greek life that marriage should
be arranged between two young people of equal station who
have never met. Athenian convention confined the daughter
of a citizen to her parents' house (except when she took part
in religious festivals) until her marriage. Consequently, if a
young gentleman falls in love, it must in most cases be with a
girl who is not his equal in station. But Charinus, the young
gentleman added to the play by Terence, is passionately in
love with the young lady Philumena. We are not told how he
has made her acquaintance. Her father, though anxious to
find a husband for her, seems never to have heard of this very
eligible suitor. Apparently Terence has here introduced a
situation foreign to New Comedy—but quite possible in Roman
life. He must, then, have invented Charinus, and with him
his slave Byrria—and this is the natural interpretation of what
Donatus says. The two characters are not very skilfully added ;
once we know (thanks to Donatus) that they have been added,
we can easily see how it was done. They have no influence on
the main plot, and the scenes in which they appear could be
altered without difficulty so as to eliminate them altogether.
But the reason for the introduction of Charinus must have been
not, as Donatus suggests, to supply Philumena with a husband,
but to provide an interesting contrast in character and situation
 TERENCE 95
to Pamphilus. Here we have the motive of a creative dramatist,
even if the execution in this, Terence's first play, is faulty. The
presence of a second lover, as Donatus remarks, is not only
without authority in Menander's Andria, but is characteristic
of Terence, each of whose plays, except the Hecyra, contains a
double love-plot. It has been absurdly supposed that these
double love-plots are the result of fusing two originals together,
and that Charinus, for instance, must really have been taken
from the Perinthia. The essential fact about Charinus is that
he is a foil to the principal lover ; if taken from the Perinthia,
he must have been a foil to the principal lover in that play also.
This is directly contrary to what Donatus tells us ; indeed, it
would involve the conclusion that the principal lover and
therefore the main plot of Terence's Andria was taken from
Menander's Perinthia, not from his Andria. Charinus pre-
supposes Pamphilus, but Pamphilus could do without Charinus.
Therefore Terence must have invented Charinus. As for his
other double love-plots, Donatus says that he ' chose' plays
with two lovers in each, but unfortunately Donatus does not
add details ; if we are to press the word ' chose ', it will follow
that these double love-plots go back to the Greek originals.
This seems on the whole more probable than that Terence
invented the secondary lovers in these plays, though he may
have introduced modifications into the Greek plots and dialogue,
possibly to emphasize the love interest. 1
With this interest in love went an interest in sex. Of course
Aristophanes had not neglected the topic of sex, but his treat-
ment of it is frankly obscene ; sex is just one of the indecent
topics with which he provokes us to laughter. New Comedy
seems to have treated this topic with restraint, though we cannot
assume that all its writers were as refined or as subdued as
Menander. In Plautus we may discern a new outlook.
Plautus is full of boisterous mirth ; yet he is not like Aristo-
phanes. We do not find Plautus jesting in the Aristophanic
way on the bodily functions, but in the concluding scenes of the
Casina—scenes which have often been thought to owe much to
the Latin writer—we find a sexual topic developed for its own
sake. Terence's treatment of sex is not at all farcical : it is
1
The much-debated Heaut. 6, duplex quae ex argumento facta est simplici (duplici A)
I take to refer to the double love-theme. In view of line 4 (ex integra Graeca integram
comoediam) it is difficult to regard this play as other than a fairly close translation.
96 THE ROMAN STAGE

sentimental, suggestive, voluptuous ; in it we may see reflected

the outlook of those aristocratic friends for whose taste he
catered. In Greek New Comedy we find no reference to the
physical attractions of respectably bred women. The Charinus-
Byrria scenes of Terence's Andria show a different attitude.
There are references to the physical desirability of a young
lady, to the question whether she has been chaste or not, to the
possibility of her taking a lover after marriage. These touches
are probably the work of Terence. The Roman public liked
a spice of sex. The Casina, with its riotous concluding scenes,
received a revival performance shortly after Plautus' death.
Terence's own most successful bid for popularity, the Eunuchus,
contains the most salacious scene in ancient drama. The
youthful hero tells in detail how, disguised as a eunuch, he has
made his way into Thais' house and violated a closely guarded
virgin. Donatus tells us that in Menander this scene was a
monologue ; Terence, by adding a character to the cast, has
made it into a dialogue. The scene was famous in antiquity,
and is indeed of great power ; but its effect is, by general
admission, largely due to its dialogue form.
The mention of the Eunuchus brings us back to what is
perhaps the central problem in the study of Terence. We
have seen that Terence claims to have added to that play
two characters from the ΚόΧαξ of Menander. This would
seem to be a parallel to the addition (attested not by Terence
but by Donatus) of Charinus and Byrria to the Andria ; but
there is a difference. The parts of Charinus and Byrria could
be omitted without injuring the plot of the Andria. But the
parts of the captain and the parasite, though not of great
importance, could not be left out of the Eunuchus as we have it
without destroying the unity of the action. This unity of
action did not come about by chance, but was the conscious
creation of the dramatist. Therefore, if Terence really added
these characters to Menander's Ευνούχος, we must attribute
to him some share in the invention of the plot. He says,
moreover, that he took the two characters from the Κολάζ.
Whatever the truth may be which lies behind this statement,
one thing which we cannot believe is that Terence's play is
simply Menander's Εύνονγος with two characters added
from another play. If they were added, something else in the
Εύνοΰχοί must have been left out. Another difficulty is to
 TERENCE 97
decide what Terence means when he says that he transferred
characters from the Κ6\αξ. A character in a play exists
in the words he utters. But what he says is for the most part
related to the plot and the other characters of the play. We
do not know much about the Κολαξ ; but at any rate we know
that it was a different play from the Ευνούχος. It would be
absurd to suppose that the precise words uttered by the captain
and parasite in the Κ6\αξ could be put into the mouths of
characters set in the different context of the other play. We
have to conclude either that the Εύνοΰχος had its own
captain and parasite (in which case Terence presumably
took them over, possibly somewhat altered, along with the
rest of the play), or that Terence was responsible for putting
these two characters into the Eunuchus, which of course means
that he had to supply appropriate words for them, in so far
as what they say is related to the rest of the play. In neither
case can he have taken over from the Κόλαξ any words of
theirs which related to the plot of that play (and would there-
fore have been out of place in the Eunuchus). What he may well
have borrowed from the ΚόΧαξ would be the opening
soliloquy of the parasite and the opening dialogue between
the parasite and the captain, for these passages in our Eunuchus
have no functional value or reference to any particular context.
On hearing them Luscius Lanuvinus might have felt justified
in accusing Terence of stealing the two characters concerned
from the old Colax of Naevius or Plautus. (Indeed the dialogue
contains a joke which seems to have occurred in a play by
Andronicus.) Terence's object in his reply was not to state
objectively what he had done, but to achieve the practical
effect of assuring the public that he was giving them what
he had promised—his own translation from the Greek of
Menander. His reply had the merit of brevity; none of
those present was likely to detect its inherent fallacy (the
suggestion, that is, that dramatic characters can be trans-
ferred from one play to another). Evidently the crowd were
reassured ; the Eunuchus was Terence's most successful play.

We are now beginning to discern the secret of Terence's

methods. He was, in a limited sense, an original dramatist
forced by circumstances to pose as a mere translator. As a
dramatist he could not afford to ignore the general public,
but his opinion of their taste is revealed in his petulant reference
98 THE ROMAN STAGE

to the populus studio stupidus who left the Hecyra to see a rope-
dancer. His admitted object is to write plays of a high artistic
standard, sine uitiis. Though he professes to admire and
imitate the great ' careless' masters of the past, we can be
sure that his standards were very different from theirs. His
repeated reference to Plautus' ' carelessness ' in dealing with
his originals is itself a criticism ; yet his boast that his own
plays are ' faultless' suggests that, for him, mere translation
of the Greek original is not enough. 1 Donatus frequently
points out alterations introduced by Terence ; sometimes he
enlivens a scene by turning monologue into dialogue, some-
times he makes his characters behave more naturally than in
the Greek play (for example Micio, the comfortable old
bachelor of the Adelphi, is allowed by Terence at least to protest
before he is pushed into marriage), sometimes he adapts his
original to Roman customs and feelings. Donatus' comments
are supported by the evidence of the fragments preserved from
the Greek originals. While toning down difficult allusions
to peculiarly Greek institutions, Terence refuses to introduce
anything specifically Roman or Italian. He adds no topical
references, no allusions to contemporaries ; there is little
word-play, little buffoonery; the language is restrained ;
the metrical effects are subdued. Reading Terence we find
ourselves in a world which is neither characteristically Greek
nor aggressively Italian, but independent of place and time.
Terence's great interest is humanity. The keynote of his
dramatic technique is contrast of character. In the Heauton-
timorumenos, the Phormio and the Adelphi we have contrasted
pairs of old as well as of young men. In the Hecyra we have a
study of two elderly couples. In the Eunuchus there is a double
love-plot. The addition of the abduction-scene in the Adelphi
may have been intended to bring into bold relief one side of
Aeschinus' character, as conceived by Terence. Donatus
tells us that in Menander's Ευνούχος Chremes was a
' rustic' ; in Terence's play he is effectively portrayed as a
1
He criticizes his rival Luscius Lanuvinus for making the defendant speak
before the plaintiff in a trial-scene (Eun. 10-12). But Lanuvinus was presumably
following Menander (cf. the trial-scene in the Epitrepontes). Terence's criticism
implies that in Lanuvinus' place he would have altered Menander's arrangement
and made the plaintiff speak first. We must not take too seriously what is prob-
ably a mere debating-point, but at least we can say that for Terence translation
was not enough.
 TERENCE 99
timid creature, who has to be roused to action by the en-
couraging words of a courtesan ! Indeed one of the things
that struck ancient critics was the nobility of Terence's
courtesans. In character-drawing, according to Varro,
Terence was first among Latin writers of comedy.
We have seen that Terence treated his originals with some
freedom. Exactly what liberties he took we cannot always
divine. There may have been more divergencies than those
mentioned by Donatus ; for that matter, Donatus occasionally
points out where Terence's text agrees with the Greek originals,
and it is impossible to suppose that Donatus has given all the
examples of such agreement. Apart from the addition of
characters in the Andria and the Eunuchus and the insertion of
a whole scene in the Adelphi, the changes made seem to have
been slight. One of Terence's objects may have been to extend
the use of surprise as an element in drama. Ancient drama,
on the whole, does not seem to have aimed at surprise effects ;
in tragedy the myths were already known 1 : in comedy, where
the plot was invented by the dramatist, the danger of puzzling
the audience had to be taken into account. Plautus, adapting
Greek plays for a Roman audience, often wearies us with the
pains he takes to explain each turn of the plot beforehand.
Terence expects more of his audience. Perhaps his total
avoidance of preliminary explanations of the plot (such as
Menander had thought necessary, at least in his Perikeiromene)
is connected with a desire to startle his audience, and even at
times to mystify them a little. The revelation of Chremes'
bigamy (in the Phormio), Pamphilus' discovery (in the Hecyra)
of what is really the matter with Philumena, the extraordinary
trick played by Davus (in the Andria) in making the unwitting
Mysis act the desired part much more effectively than she could
have done had he previously disclosed his intentions, Demea's
sudden abandonment of his previous way of life (in the Adelphi)
and the success with which he turns the tables on Micio—these
scenes (which may indeed be purely Greek in origin, but are
unlike anything in Plautus) are not only ' d r a m a t i c ' in our
sense of the word ; well acted, they must have been excellent
' theatre'. Practical details of staging may sometimes be a
little obscure ; the movements of characters are not thrust on
our attention as they are by Plautus, who was above all things
1
Cf. the fragment from Antiphanes* ίΐοίητις (Koch 1 9 1 ) .
100 THE ROMAN STAGE

a man of the theatre. In command of rhythm and metre,

in flow of language, in farcical power and animal spirits,
Plautus is altogether the superior of Terence. Terence's
appeal is to the reflective ; but this does not mean that his
plays were a failure on the stage. That they were frequently
revived is suggested by the very fact that the notices of early
and later performances were confused in the records. To the
end of the Republic and even in imperial times they were
known to the theatre-going public ; Quintilian seems to
refer to contemporary performances of Terence's plays, and
the illustrations in the manuscripts have been thought to
suggest that their connexion with the stage reaches to even
later times.
We may wonder why of all the Republican dramatists only
Plautus and Terence have survived. The reason may be that
later ages found these two writers particularly good reading.
£ach in his own way was admitted to be a master of style. If
Plautus was more amusing, Terence was more polished and
more philosophic ; he was also easier to understand. The
excellence of the manuscript tradition is evidence of his
popularity in late antiquity and the Middle Ages. Few Latin
authors are in less need of a commentator. Free from difficult
expressions or topical allusions, written in the elegant Latin of
aristocratic Rome, his plays retain their attraction because they
deal effectively with topics interesting to all mankind.

NOTE ON ' C O N T A M I N A T I O '.

In m y account of Terence I have put forward what is, I believe,
a novel view on the subject of ' contamination '. As a narrative of
what Terence actually did I would not claim more for my hypothesis
than that it is consistent with itself and with the known facts. But
on the negative side I believe that there are substantial reasons for
questioning the modern view which has so powerfully affected the
whole study of Latin drama. I refer to the general belief that
(a) the word contaminare was sometimes used with the technical
sense of ' combine ' ; (b) the ' combination ' of plots was a wide-
spread practice among the R o m a n translators of Greek plays.
This belief is based ultimately on the occurrence of the word in
two prologues of Terence {And. 16, Heaut. 17), taken together with
the general evidence as to Terence's methods which he gives in
his prologues. Accepting ' contamination ' as a basic principle
 TERENCE ΙΟΙ

scholars have worked through the extant remains of L a t i n drama

in order to discover traces of such combination of plots, and though
their conclusions differ widely, m a n y of them will be found stated
as established truths in most works on early Latin literature or
drama.
M y reasons for opposing this formidable array of authorities
are : —
(1) T h e prevailing sense of contaminare in Latin is ' s u l l y ' ,
' stain ', ' s p o i l a n d it is unlikely that a word with so b a d a sense
could also have been used in the neutral and technical sense of
' combine
(2) ' Combine ' is not a satisfactory interpretation of the word
even in the two passages on which the theory is based.
(3) A p a r t from the prologues of Terence there is no statement in
the whole of ancient literature which suggests that the combination
of two Greek plots into one was ever attempted by any Latin
translator.
(4) T h e internal evidence of our extant Latin plays and
fragments affords no evidence of such combination.
(5) T h e combination of plays, in the sense in which the term is
widely used b y modern scholars, is a logical impossibility.
I will take these five points in more detail :
I. T h a t contaminare has in general a b a d sense is beyond
dispute. (Once or twice in late L a t i n it seems to be a mere synonym
for contingere, ' touch '.) Etymologically, no doubt, it means
' bring into c o n t a c t ' , but in practice this always implies ' with
something base This is h o w Terence uses i t ; (Eun. 552) :
ne hoc gaudium contaminet uita aegritudine aliqua.

Donatus, in his commentary on the Andria prologue, explains the

word as ' sully It is true that when he comes to the line
contaminari non decere fabulas
he remarks id est ex multis unam non decerefacere. This is not, however,
a new definition ; it is, I believe, an attempt to interpret the line in
the light of this very prologue and Heaut. 17. I f w e translate ' the
meaning (of the charge against Terence) is that it is not right to
make one play out of s e v e r a l ' , w e interpret Donatus correctly. T h e
Thesaurus does not recognize the neutral sense o f ' c o m b i n e b u t
treats all three occurrences of the word in Terence as examples of
the sense ' miscendo deprauare ' to spoil b y mixture
2. (a) I f contamino was a neutral, technical term denoting the
normal practice of Terence's predecessors, w h y was it used b y his
critics as the very ground of their complaint against him ?
(b) If the combination of plays was the practice which he
attributes to his predecessors and in which he declares that he
means to follow their example, w h y does he refer in the same breath
8
102 THE ROMAN STAGE

to their ' carelessness ' as more worthy of respect than the ' dull
laboriousness ' of his critics and rivals ? T h e combination of plots
cannot be achieved b y ' carelessness ', but needs conscious effort,
if indeed it is to be achieved at all.
(c) T h e process of combining pairs of Greek plays to form
single L a t i n plays necessarily means that the number of resultant
Latin plays will be less than the number of Greek originals. W h a t ,
then, is the point of the charge (Heaut. 17) that Terence had
' combined several Greek plays in making a few Latin plays ' ?
Is not this the very nature of combination ? T h e sting of the
accusation becomes apparent when w e give contaminate its usual
sense : that is, ' Terence has spoiled several Greek plays and made
only a few Latin plays.'
T h e conclusion is inescapable : contamino is not a neutral
word ; if applied to a literary process, it can only be applied by a
hostile critic. W h o first used it in this w a y ? Even if w e suppose
for the moment that Terence's predecessors had in f a c t ' combined '
plays, it seems unlikely that they had been criticized for so doing.
A t any rate Terence's words imply that the attack made on him
is something new ; for he says that, even though he himself has
merely been following precedent, his enemies ' understood nothing
that is, were unaware of the fact. It was Terence's enemies w h o
used the word about his methods ; and w h a t they meant was that
he was ' sullying ' his Greek originals. This would be an un-
reasonable term to use of combining one Greek play with another.
But it would make sense if they were simply accusing Terence of
somehow altering the original which he was ostensibly translating.
A n d if Terence in his reply alleges that all that he has done is to
insert material from a second Greek play, it does not follow that this
was w h a t he had been accused of doing. W h e n his enemies
accused him of ' spoiling ' his plays, they did not m e a n that he
' combined ' plays ; and w h e n he replied that all that he had done
was to ' combine ' plays, he did not mean that he had ' spoiled '
them. H e quotes the word contaminare in Oratio Obliqua, but
does not use it in his o w n person of his methods. H e admits that
there is some foundation of fact in what his critics say—-factum id
esse поп negat. T h e foundation of fact is that he has not been
content to keep strictly to his ostensible original ; and it is only in
this sense that he can claim to be following earlier example, that
of the ' careless ' Naevius, Plautus and Ennius.
3. T h e frequent references in L a t i n literature to the translation
of Greek plays into L a t i n always imply that the Latin translator was
dealing with one original at a time. Cicero goes so far as to say
that the Greek plays were translated ' word for word ', though
elsewhere he more reasonably observes that the Latin dramatists
rendered ' not the words but the sense '. Cicero occasionally
points out cases in which the translator departs from his original,
 TERENCE IO3

for example by changing the sense or expression in order to suit

R o m a n taste ; but nowhere have we any reference to alteration of
structure or to borrowing from another original.
4. Judged by its fruits, the theory of contamination has produced
nothing of positive value. Every play of Plautus has been declared
' contaminated ' by one scholar or another ; while Ribbeck, in
his standard work on R o m a n tragedy, attributes to the Latin
dramatists the most far-reaching powers of fusion and reconstruction.
None of these theories would ever have seen the light but for the
two passages in the prologues of Terence ; and even if the foundation
were sound, the superstructure is invariably crazy. O n this whole
subject I would refer the reader to Michaut's admirable chapter. 1
Where I differ from Michaut is in my view that ' contamination '
itself is unproved. Buteven if we are willing to retain the conventional
view that Greek plots were combined, I do not see how the evidence
at our disposal can ever afford illustration of such a proceeding.
T o see such combination at work we should need to be able to
compare a Latin play as a whole with its two Greek originals as
a whole. In fact all that we have to compare consists of passages,
usually short. Speaking generally, even where we have reason to
suspect that the Latin writer has departed from his ostensible
original because of some incongruity with the context, it does not
follow that the incongruous material is taken from a second Greek
source. Indeed it is usually of unmistakably R o m a n origin.
5. No doubt dramatists in ancient as in modern times borrowed
ideas from the works of their predecessors and contemporaries.
Similar situations, similar characters recur in play after play. This
is perhaps even more true of New Comedy than of modern drama,
where characters are more highly individualized. W e might
recognize that a character was modelled on Falstaff or Caliban ;
but how would one recognize that a stock character in New Comedy
was modelled on a similar character in a particular play ? Neverthe-
less, even in New Comedy, plays were not composed of inter-
changeable parts. Each play showed a particular group of
characters in a particular setting. W h a t each character said in that
play was addressed to one of the other characters in the p l a y — o r
even if a monologue, it was at any rate concerned with the situation
and plot of the play. Therefore, when Terence says that he
inserted two characters from the Colax into his version of the
Eunuchus, he must be saying either more or less than the truth.
For a character in a play, when removed from his setting, ceases to
exist; everything that he said in his original context, at least when
related to that context, would be meaningless in a different context.
T h e part would have to be written afresh ; the dramatist would
have to write the new words himself; and this is not borrowing, but
original composition. I am ready to suppose that Terence borrowed
1 Flaute ii. ch. 16.
 THE ROMAN STAGE

passages or even whole scenes ; but in so far as these borrowings

had to be fitted into the new play, the adaptation had to be done
by Terence. The theory of contamination is that the Roman
dramatists put together portions of Greek plays, adding a minimum
of original work in order to fit the Greek material together. I
think that the attempt to do this would inevitably involve the Latin
translator in original work of a complicated character—and this
is the negation of the contamination-theory, which denies to the
Latin dramatists all powers of dramatic construction, while in-
sisting that they made far-reaching changes in the structure of their
originals. That such changes were made I do not believe ; but if
they were made, then the Latin dramatist himself must be given
the credit for such unity as we are willing to concede to his play.
 C H A P T E R XIII

THE OTHER COMPOSERS OF PALLIATAE

T H E A C T I V E world of contemporary comedy to which

Terence's prologues introduce us has left little behind but
the names of a few authors, some comments on their work, a
few titles and a few fragments. In most cases we have nothing
to guide us as to their date ; perhaps they were contemporaries
of Caecilius or Terence, or came a little later in the second
century B.c. T R A B E A is said to have had emotional power ;
a fragment of an unnamed play shows us a lover in the full
tide of joyful expectation : ' I will soothe Madam's palm with
cash ! at a nod from me she will obey my wishes and desires ;
I will come to the door and push it with one finger, it will open,
all of a sudden Chrysis will see me. She will run to me, all
eagerness to throw herself into my arms ; she will be mine !
Fortune herself does not know such fortune as mine '. Perhaps
it is the same lover who later remarks in more reflective mood :
' I think that violent delight is a great source of folly
A T I L I U S wrote a play called ' T h e W o m a n h a t e r ' , Misogynos ;
he also seems to have gone in for tragedy, if it is the same
Atilius who made a bad translation of Sophocles' Electra ; his
style was harsh. A Q U I L I U S wrote a Boeotia; the passage
preserved to us is a parasite's lament on the invention of the
sundial, which has made his meals depend on the sun instead of
on his own appetite. V a r r o considered the style of this passage
proof that the play should be ascribed to Plautus. A n attempt
has been made to connect the passage with the introduction of
the sundial into Rome, but of course it is in all probability
translated from the Greek. L I C I N I U S I M B R E X wrote a
Neaera from which we have two lines reminiscent of a pass-
age in Plautus' Poenulus ; should we identify this writer with
P. Licinius Tegula, who wrote a hymn in 200 B.C., on the ground
that imbrex and tegula both mean ' tile ' ? J U V E N T I U S has
left two or three lines as well as a title—Anagnorizomene;
V A T R O N I U S seems to have written a comedy called Burra
105
 THE ROMAN STAGE

(—Greek Pyrrha). V A L E R I U S wrote a Phormio. W e know

a little more about L U S G I U S L A N U V I N U S , the ' m a l -
evolent old poet' of Terence's prologues. As he was ' old '
in 166 B.C., he must have been born about 2 1 0 B.C. or earlier.
He translated from Menander the Phasma (' Ghost'), which
told of a secret passage through a party wall used by a
maiden whose appearance in the other house (her mother's)
was at first taken to be that of a spectre by the young
gentleman who lived there ; on discovering his mistake he
fell in love with and finally married her. The Thesaurus
(' Treasure') told of a young spendthrift who, after running
through his property, sent his servant to make an offering
inside his father's tomb, as had been enjoined in his father's
will. The field in which the tomb stood had passed into
the hands of an avaricious old neighbour. In company
with the slave this neighbour visited the t o m b ; there they
found a treasure, which the avaricious neighbour promptly
seized. The young man brought an action to recover the
treasure ; the neighbour, though defendant, is made to speak
before the plaintiff—a point with which Terence taunts
Lanuvinus. Terence also finds fault with Lanuvinus for
introducing a ' running slave ' who orders the public to make
way for him, and for another scene in which a mad young man
thinks that he sees a hind fleeing from dogs and imploring
his protection. Lanuvinus' charges against Terence were that
his style was thin, that he ' spoiled ' (i.e. altered) his Greek
originals, and that he was a novice, dependent on his noble
friends, who had enabled him to enter on his literary career
without proper preparation. These charges suggest the
character and circumstances of Lanuvinus himself. Presumably
he had passed through a long apprenticeship—possibly as an
actor ; he was not admitted to the society of the great, and
disliked those who had this privilege ; in fact he had a class-
conscious, trade-unionist mentality. Some of Terence's
charges against him seem unfair ; the faults mentioned above,
if faults, were presumably taken over from his Greek originals.
Terence further accuses him of being ' a good translator and
a bad writer' who has turned good originals into bad Latin
plays because of his lack of style, and whose success on at
least one occasion was due to the actors rather than to his
own talents. Evidently Lanuvinus translated his originals
 THE OTHER COMPOSERS OF PALLIATAE I07

fairly closely. He himself criticized Terence's style as ' thin'

and ' s l i g h t w h i c h may perhaps suggest that his own was
pompous and turgid ; the two extant lines support this view :
Athenienses, bellum cum Rhodiensibus
quod fuerit, quid ego hie praedicem ?

The bitterness of the dispute suggests that Lanuvinus

regarded Terence as a dangerous rival. On the one side we
have the old writer laboriously turning his originals into dull
Latin, on the other his young rival remodelling his originals on
novel principles so as to produce new plays written in the
choicest style.
The final name is T U R P I L I U S (thirteen titles : Boethuntes,
Canephorus, Demetrius, Demiurgus, Epiclerus, Hetaera, Lemniae,
Leucadia, Lindia, Paedium, Paraterusa,1 Philopator, Thrasyleon ; the
fragments amount to about 200 lines). We are told that
Turpilius died at an advanced age in 103 B.c. ; he must there-
fore have been bom during the lifetime of Terence. It would
seem that the Hellenization of the palliata is now complete ; his
titles are all Greek, and his fragments contain nothing which
might not have been translated from the Greek.
We can follow the plot of the Leucadia to some extent from
the remarks of Servius and the fragments. The original was
by Menander. Phaon was a poor and ugly boatman of Lesbos,
who plied for hire between the island and the mainland.
Venus entered his boat in the guise of an old woman ; he gave
her a free passage, and in return she gave him a box of oint-
ment ; when he rubbed himself with it, he became irresistibly
attractive to all the women whom he met. We see one of
these women, who has spurned her previous lover ; the lover
expresses his bitter feelings about the airs which the boatman
gives himself and about the shameless way in which the girl
makes love to him. Phaon scorns her advances, and she
determines to commit suicide by throwing herself from the top
of the Leucadian cliff into the sea. We see her standing in
terror on the cliffs, looking at the crags, listening to the waves,
appalled at the loneliness ; then, resolved on her desperate
act, she commits herself to the gods (excluding Venus) and the
winds and casts herself headlong. Her struggles in the sea are
observed by some one who bids his oarsmen row up ; we hear
1
' The Female Spy or ' We're very wide awake
ιο8 THE ROMAN STAGE

her piteous request for a fire at which to dry herself; apparently

she has recovered her senses, and perhaps she is reconciled to
her former lover. Such a plot seems a refreshing change from
the usual urban setting of New Comedy ; how the play could
be staged without change of background is illustrated by the
Rudens. Presumably the women of Lesbos enjoyed more
freedom than those of Athens.
Another play translated by Turpilius from Menander is the
' Heiress ' (Epiclerus). A young man is being compelled by
his parents to marry a kinswoman who has been left an orphan
and heiress ; he bitterly resents his fate. The opening lines
of Menander's play show him coming on the stage from his
sleepless couch to tell the story of his life. Turpilius, after
the fashion of Terence, and possibly under the influence of the
opening . scene of the Curculio, has converted this opening
monologue into a dialogue between the young man and his
slave : ' Pray tell me sir, where you are hurrying before day-
break with only one slave ? ' ' I cannot sleep indoors,
Stephanio.' ' Why ? ' ' A s usual, my troubles drive away
sleep and force me out of doors in the dead of night.' The
Philopator dealt with a young man who ' loved his father '.
We gather that the heroine has been unfortunate, and hear her
(perhaps off-stage) lamenting her lot. Somebody drops a
letter which some one else picks up, observed by a third party ;
hence, we may presume, comes the disclosure of some secret,
bringing the play to a happy conclusion.
Five of the thirteen titles suggest a Menandrian origin.
The later palliata seems to have maintained the interest in
Menander which appears in Caecilius and still more in Terence.
The tone of the fragments seems Menandrian ; there is a lack
of farcical or violent scenes, of rough jests, of inflated language ;
among Latin dramatists Turpilius seems to have been in the
line of Terence. In metre we find plenty of trochaics, and
some cretics and bacchiacs.
Cicero speaks of a contemporary performance of Turpilius'
Demiurgus in which Roscius took part. This reminds us that the
palliata lived on the stage long after the death of its last
composer. Cicero refers to Ballio, the pimp of Plautus'
Pseudolus, as a figure still familiar to the theatre-going public
in his own day. Terence's didascaliae indicate revival per-
formances of several of his plays. We can thus see how
 THE OTHER COMPOSERS OF PALLIATAE IO9

questionable is the current view that the public got tired of the
foreign palliata and demanded something more native.
Indeed its exotic flavour and avoidance of dangerous, topical
references were among the chief attractions of the palliata ;
in plot, moreover, native comedy was probably altogether
inferior, even in the eyes of the Roman populace. Why the
supply of composers for the palliata failed before that of any
other type of drama is a difficult question ; but we may note
that within little over a generation after the death of Turpilius
the same fate seems to have overtaken all forms of literary
tragedy and comedy, whether translated or original. The old
classics were revived, but no new writers appeared. The
explanation may be that the Greek quarry was thought to be
exhausted, or that there had been a change in the social status
of dramatic composers. We now have on the one hand the
noble dilettante who writes plays for his own amusement and
that of his friends, on the other hand the professional composer
of subliterary theatrical pieces such as the mime. The feeling
is growing up that connexion with the theatre, including even
the writing of plays for public performance, is degrading. The
composition, even the translation, of tragedy and comedy
required a considerable standard of education, and those who
possessed such a standard would now have felt that they were
compromising their position by writing plays for the amuse-
ment of the general public.

NOTE ON VOLCACIUS SEDIGITUS AND HIS LIST OF

THE ' T E N BEST WRITERS OF COMEDY'.

About the end of the second century B.C. the writing of literary
history had begun at Rome. Among the topics discussed were the
dates of the dramatists and their respective merits ; one point in
particular was the order in which the writers of palliatae should be
placed with regard to quality. Volcacius Sedigitus drew up a list
of the ' ten best writers of comedy ' in these words :
multos incertos certare hanc rem uidimus,
palmam poetae comico cui deferant.
eum meo iudicio errorem dissoluam tibi,
ut, contra si quis sentiat, nil sentiat.
Caecilio palmam Statio do*mimico*.
Plautus secundus facile exsuperat ceteros.
110 THE ROMAN STAGE

dein Naevius qui*feruet*pretio in tertiost.

si erit quod quarto detur dabitur Licinio.
post insequi Licinium facio Atilium.
in sexto consequetur hos Terentius.
Turpilius septimum, Trabea octauum optinet.
nono loco esse facile facio Luscium.
decimum addo causa antiquitatis Ennium.
Too much importance should not be attached to what is manifestly
an individual opinion ; but Volcacius was not the only critic who
put Caecilius at the head of comic writers. We can understand
his putting Plautus and Naevius so high on the list (the description
of Naevius' fiery quality is uncertain : the Ms. read seruet) ; after
these three he does not seem to have cared very much how the other
writers should be arranged—' if there is to be a fourth place, it
will be given to Licinius '. It is surprising to find Terence only
sixth; his only outstanding stage-success, however, was the
Eunuchus, and the qualities of his style may not have been appreciated
until a later age. Ennius was not regarded as an important comic
writer—if we can discount the testimony of Terence {And. 18), and
Terence may have been thinking of Ennius as a great and popular
writer who, among other things, wrote comedies. But in putting
Ennius last Volcacius seems to be expressing some particularly
personal view. It is true that, in date, Ennius lived earlier than
most of the writers here mentioned ; and the implication of
Volcacius' words is that he was earlier than all the writers not
mentioned ; apparently, then, Livius as a writer of comedy had
already been forgotten, or was actually supposed to be later than
Ennius. Donatus, indeed, thought that Ennius was earlier than
Plautus.
 CHAPTER XIV

AGCIUS

A C C I U S is said to have been born in 170 B.C. Jerome's

L • date agrees well enough with Accius' own statement
that he was fifty years younger than Pacuvius. There are
independent records showing that the Accii were connected
with Pisaurum, in Umbria. Jerome tells us that the poet's
parents were ex-slaves ; whether this is true or not, there is
evidence that the poet enjoyed the friendship of some of the
highest men in Rome. During his long career he occupied
himself with grammatical, theatrical and literary studies
(Cicero, as a young man, discussed literature with him) and
above all with tragedy ; his career was brilliant, and he was
in old age the leading figure in the collegium poetarum. The
stories about Accius bring before us a man of great energy and
self-confidence ; in his character and industry, as well as in
his style, there was a daemonic force. He appears to have
been a hasty worker ; in his studies on literary history he
was capable of gross mistakes, and even those who most
admired his tragedies thought them inferior in learning and
care to those of Pacuvius. We have the titles of over forty
tragedies : Achilles, Aegisthus, Agamemnonidae, Alcestis, Alcimeo,
Alphesiboea, Amphitruo, Andromeda, Antenoridae, Antigone, Armorum
Iudicium, Astyanax, Athamas, Atreus, Bacchae, Chrysippus, Clute-
mestra, Deiphobus, Diomedes, Epigoni, Epinausimache, Erigona,
Enphyla, Eurysaces, Hecuba, Hellenes, Medea, Melanippus, Meleager,
Minos or Minotaurus, Myrmidones, Neoptolemus, Nyctegresia,
Oenomaus, Pelopidae, Persidae, Philocteta, Phinidae, Phoenissae,
Prometheus, Stasiastae or Tropaeum Libert, Telephus, Tereus,
Thebais, Troades) and two praetextae (Aeneadae or Decius,
Brutus), with about seven hundred lines. The range of titles
shows that he explored every field of tragedy—the Trojan,
Theban and Pelopid cycles ; originals of the fifth century
as well as later Greek tragedy, and independent compositions
III
112 THE ROMAN STAGE

on Roman historical themes. Unfortunately the fragments

are all short and scattered, and cast only fitful gleams on his
powers and methods as a dramatist. I have already mentioned
the well-known anecdote which tells how, on his way to Asia,
he visited Pacuvius, then living in retirement at Brundisium,
and read him his new play, the Atreus. Pacuvius found the
style impressive but harsh ; Accius replied that the best
fruits required most time to mellow. We need not believe in
the historical truth of this story (which reminds us of the
equally suspicious account of the interview between Gaecilius
and Terence), but at least it shows that some of Accius' work
was found crude by later generations; indeed his outstanding
quality was force, which might at times degenerate into
violence. Cicero calls him summus poeta, grauis et ingeniosus
poeta ; from Horace and Quintilian we gather that though
inferior to Pacuvius in learning he surpassed him in force and
elevation, while sharing his impressive dignity of style, senti-
ment and characterization ; Vellerns, who makes him the
central figure of Roman tragedy, says that though less careful
than Pacuvius he excelled him in energy (sanguis) ; Vitruvius
speaks of the vivid picture of the author which his works
brought to the reader ; Ovid bears tribute to his powerful
rhetoric (animosi oris).
Preference for plots and treatment of a violent, melo-
dramatic nature, majestic rhetoric, flamboyant character-
drawing—these qualities seem characteristic of Roman tragedy
in general, but are particularly marked in Accius. Another
very Roman quality was his command of effective repartee,
a command so evident in his dialogue that, according to the
story, he was asked why he had not made law his profession.
More surprising is the power of pictorial description and the
feeling for nature which his fragments occasionally display.
The Roman dramatists, even if incapable of creating plots or
characters on the large scale, seem at least to have thrown all
their energies into visualizing and intensifying each scene as
they came to it.
It is easy to show that Accius, like the other Roman tragic
writers, was overfond of emphasis. We can compare some
fragments of his Phoenissae with the original by Euripides.
Tyrrell translates the opening lines of the Greek play
thus :
 ACCIUS "3
О sun, that through the fires of the firmament
Cleavest thy way, and in thy golden car
Launchest the flames from thy swift coursers' feet,
Ill-starred the ray thou sheddest once on Thebes.
This is Tyrrell's rendering of Accius' version ;
О sun, that in thy glistering chariot borne,
With coursers swiftly galloping, dost unfold
A sheet of gleaming flame and burning heat,
Why with such baleful auguries and omens
Adverse giv'st thou to Thebes thy baleful light ?
Similarly, while Eteocles' command to Polynices to leave the
city is simply expressed in the Greek :
Then get thee from these walls, or thou shalt die,
Accius might make us fancy that we see Cicero driving Catiline
from Rome :
egredere, exi, ecfer te, elimina urbe !
No doubt rhetoric is at times appropriate to the dramatic
situation : what could be better than the famous retort of
the tyrant Atreus — oderint dum metuant ! (a phrase
apparently imitated from Ennius). But the continual straining
after rhetorical effect tends to eliminate all the delicate half-
tones necessary to credible character-drawing and to leave us
nothing but superhuman virtue and inhuman vice. Roman
tragedy instinctively turned to horrific, melodramatic themes,
such as the story of how Tereus violated Philomela and then
cut her tongue out so that her sister should not know of the
deed. There is force in Accius' picture of the lust-maddened
barbarian :
Tereus indomito more atque animo barbaro
conspexit earn ; hinc amore uecors flammeo
depositus facinus pessimum ex dementia
confingit.
Similarly Accius, like so many Roman dramatists, was attracted
by the theme of how Atreus served up to Thyestes a meal of
his son's flesh :
concoquit
partem uapore flammae, ueribus in foco
lacerta tribuit.
ii4 THE ROMAN STAGE

It is instructive to notice how Seneca, developing the innate

vices of Roman rhetoric to their limit, works up his description
of the same incident into a dozen ghoulish lines (Thyestes
760 ff.)
From the beginning Roman tragedy had tended to over-
statement ; with Accius this tendency has gone as far as it
well can ; with Seneca it goes further, and becomes absurd.
Splendid, flamboyant figures come before us as we read the
lines of Accius. Atreus, the tyrant, introduces himself thus :

en impero Argis, sceptra mihi liquit Pelops,

qua ponto ab Helles atque ab Ionio mari
urgetur Isthmos.1

The powerful retort of A j a x brings the speaker vividly before

our eyes :

uidi te, Vlixes, saxo sternentem Hectora ;

uidi tegentem clipeo classem Doricam ;
ego tunc pudendam trepidus hortabar fugam.

That stoic quality in Ajax's character, which was so much

appreciated by the Romans, is brought out by a famous line
taken ultimately from the Ajax of Sophocles :

uirtute sis par, dispar fortunis patris.

Telephus displays similar stoicism :

nam si a me regnum atque opes

eripere quiuit, at uirtutem non quiit.

We should, however, recognize that in Accius' rhetoric there

is much genuine feeling. These are perhaps the words of
Procne sorrowing over her mutilated sister :

О suauem linguae sonitum ! О dulcitas

conspirantum animae !

We find a certain grave humanity and sympathy with mis-

fortune (lines 187-8) :

abducite intro ; nam mihi miseritudine

commouit animum excelsa aspecti dignitas.
1 It is not certain that these often-quoted lines are by Accius.
 ACCIUS м
5
T o his Phoenissae, indeed, Accius seems to have added a com-
passionate touch not to be found in Euripides' p l a y — t h e care
for the wounded :
obit nunc uasta moenia, omnis saucios
conuisit, ut curentur diligentius.
In this compassion, this admiration for misfortune borne
with dignity, there is a Virgilian q u a l i t y — a n d we know that
Virgil was a student of Accius, and borrowed from his
Clutemestra some details of the storm in Aeneid i. There is a
feeling for nature in the description of early morning in the
Oenomaus (493-6) :
. . . before the dawn had lit
The earliest beams of day, when peasants rouse
Their sleeping horned oxen to the plough,
And split with the share the dewy steaming sod,
And lift the ridges from the yielding soil.
T h e longest fragment is a well-known description in his
Medea of the first ship, as seen by a peasant (391-402) :
In from the sea the mighty monster glides :
The billows hiss and roar ; in onward course
It dashes down their blows 'mid foam and spray;
The eddies boil and bubble, while it moves
Like some great storm-cloud torn from the sky
Or towering crag by wind and tempest driven
Or swollen water-spout, riding on high
Above the tumult of the tossing waves.
Or has the sea upturned its earthy bed,
O r deep in Ocean's caverns, trident-armed,
Has Neptune levered up some giant rock,
Now poised between the billows and the heaven;?
Accius would hardly have been a R o m a n had he not possessed
a strain of pedantry : Achilles quibbles as to whether his
stubbornness should be called pertinacia or peruicacia (a verbal
point which manifestly belongs to the Latin writer). Few
writers seem more imbued with grauitas ; there is a manly
note in
non genus uirum ornat, generi uir fortis loco
(' T h e man's the gowd for a' t h a t ' ) . Looking through the
fragments, we are struck with the frequency of the words
uir, uirtus, and by the fact that women seem less p r o m i n e n t
than in Ennius or Pacuvius.
 116 THE ROMAN STAGE

When we turn from style to substance and dramatic

structure, we are faced with the usual difficulty—the fragments
are too short and scattered to permit of certain conclusions.
Evidently Accius permitted himself some freedom in trans-
lation ; several of the fragments of his Bacchae and Phoenissae
have nothing corresponding to them in the extant Greek
originals by Euripides. We have no evidence of alteration in
structure. The explanatory prologue would have been at
least as necessary to the audience of Accius as to that of
Euripides. Certain fragments of the Bacchae and Phoenissae
come from the opening scenes, which themselves serve in the
Greek as explanations of the dramatic situation. Servius
tells us that the Atreus contained a genealogical chronicle ;
this would naturally come in the prologue. We have another
genealogical passage in inc. fab. iii.
Accius, like all the Latin dramatists, took complete
liberty with regard to metre ; fr. χ of the Phoenissae shows that
he has substituted recitative for the plain speech of Euripides,
presumably because of the emotional character of this scene,
in which Creon is told that he must sacrifice his son. That
Accius' Bacchae, like its original, had a chorus of Bacchants
is suggested by the line :
agite modico gradu ! iacite tirsos leuis !
No passage in Euripides' play quite corresponds to this line in
wording or in metre ; but the treatment of choral passages
was probably a matter in which the Latin translators allowed
themselves wide freedom. There are many passages in
Accius' fragments which purport to be spoken by groups of
people ; we have the plural titles {Bacchae, etc.) which in the
Greek plays certainly indicated the chorus ; but for all such
titles and passages the supposition of a spokesman with mute
attendants, appearing on the stage when required, is perhaps
an adequate explanation.
The staging of tragedy was presumably similar to that of
comedy, except that the back-scene would be deemed to repre-
sent not the doors of private houses but a palace, a cave or
whatever was appropriate. The palace door is mentioned
(line 29). The cave of Philoctetes is shown, surrounded by a
wilderness ; cf. line 557 :
contempla hanc sedem in qua ego nouem hiemes saxo stratus pertuli,
 ACCIUS 117
and line 554 :
quis tu es mortalis qui in deserta et tesqua te adportes loca ?
Even these lines were presumably spoken before the three-door
back-scene usual in comedy. Some would suppose that the
unwanted doors were screened from view, so that the central
door could represent the cave. 1 It is probable that, as in comedy,
the ' near distance ' lay on the spectators' right, the ' further
distance' on their left ; line 3 1 8 {ab classe ad urbem tendunt)
suggests that we shall soon see the warriors as they make their
way from the ' harbour ' side of the stage to the ' town ' side.
The action may begin at dawn ; the stars wane in the growing
light (line 693). The arrival of a new character in the early
morning is indicated by some one standing on the stage (line
123) :
sed quis hie est qui matutinum cursum hue celeranter rapit ?
A messenger arriving from a distance is bidden deliver his
report (line 499). The sound of the opening palace door
indicates that a new character is about to enter the stage
(line 29) :
sed ualuae resonunt regiae.
A speaker tells how he has overheard a conversation (line 281) ;
two characters come on the stage (i.e. into the ' open a i r ' )
in order to converse without the danger of being overheard
(line 292). Some one remarks that it is unseemly to walk
quickly (line 23) :
celeri gradu gressum adcelerasse dedecet.
These and similar touches are the hallmark of the dramatist
who intends his plays to be produced on the stage. No doubt
Accius derived them from his originals ; at any rate they mark
out his tragedies as different in purpose from the closet-
dramas of Seneca.

ROMAN TRAGEDY AFTER ACCIUS

A contemporary of Accius and one of the first Romans of

the governing classes to try his hand at dramatic composition
was G. J U L I U S C A E S A R S T R A B O (put to death in 87 B.C.).
1
See p. 259 and note.
9
ιι8 THE ROMAN STAGE

We have the titles of three plays : Adrastus, Teuthras, Tecmesa ;

a late grammarian tells us, with what authority we do not
know, that the author was particular to have the name
' Tecmesa' so pronounced (i.e. not Tecamesa) He was more
celebrated as an orator ; Cicero tells us that his tragedies lacked
power and that his style was too genial and witty to be strictly
tragic. The three lines which have survived are not enough
to enable us to judge his style for ourselves. He was a member
of the collegium poetarum ; we hear that Accius refused to
rise from his seat when Strabo entered, not out of disregard
for Strabo's high position, but because of conscious superiority
as a dramatist. We hear nothing further of Strabo as a tragic
writer ; we have no reference to the performance of any of
his plays, apart from the above-quoted remark of Marius
Victorinus about his insistence on having ' Tecmesa' pro-
nounced in the Greek way on the stage. Cicero seems to think
of him not as a practising dramatist but as a literary man,
one of whose interests was the composition of tragedies. On
the whole the impression we derive is that he was a dilettante
in drama, though sufficiently interested in it to give his
oratorical style a dramatic colouring. Similarly another
orator-dramatist of about this time, or perhaps of the previous
century, the Roman knight C. T I T I U S , is said by Cicero to
have sacrificed strictly tragic style to display his wit. Indeed
Afranius, the writer of togatae, made Titius one of his models.
We have nothing left of Titius' tragedies, but a fragment of
his oratory shows marked ability for social satire. Of
POMPILIUS, said to have been a pupil of Pacuvius, we have
a single line, and of SANTRA, perhaps to be identified with
the grammarian, we have the title Nuptiae Bacchi (?) and about
three lines. The strictly quantitative metre of these lines
suggests that Santra belongs to a somewhat later period.
It seems to be true that after the death of Accius few or
no more tragedies were written for the stage. The reason
cannot have been merely lack of popular interest; old
tragedies continued to be performed down to the end of the
Republic, and Aesopus, the tragic actor and friend of Cicero,
enjoyed a great reputation. Perhaps the reason was a change
in the attitude towards the writing of tragedy. On the one
hand noble dilettanti were beginning to amuse themselves
with the composition of tragedy ; on the other hand it was
 ACCIUS n
9
beginning to be felt that to write plays to be performed on the
popular stage was scarcely a worthy occupation, at any rate
for men of the class of these new noble writers. Perhaps there
is a first hint of this new attitude towards practising dramatists
in the liberty taken by a mime-performer in mentioning
Aerius' name on the stage ; though Accius was successful in
the action he brought against the offender, no Roman after
Accius' day seems to have taken up the writing of tragedy
as his profession. The Augustan age produced two famous
tragedies, written (to judge by the fragments) in strictly
quantitative metre, the Thyestes of Varius and the Medea of
Ovid ; we have the evidence of a scholiast that the Thyestes
was produced at the games in celebration of Actium (possibly
an exceptional occasion), but Ovid, later in life, protests that
he has never written for the stage, though some passages of
his Metamorphoses seem to have been used as libretti for panto-
mime. The only respectable professional dramatist of any
kind of whom we hear after the time of Accius is the mime-
writer Laberius ; and his career was to show in melancholy
fashion that it was now no longer possible to write for the stage
and to retain one's dignity.
 C H A P T E R XV

NATIVE COMEDY : THE FABULA T O G A T A

T HE R O M A N S had at all times an instinct for social and

political satire. It was natural that this instinct should
seek expression in their drama. Naevius, the first native-born
Italian dramatist, though like Andronicus basing his comedies
(so far as we know) on Greek originals, introduced a strong
native flavour by references to such topical themes as the habits
of guests from Praeneste and Lanuvium, or the picture of the
Lares painted for the ' festival of the cross-roads ' by the
imported artist Theodotus. Naevius' tendency to introduce
political themes and his attacks on certain nobles brought
him to disaster, and future Latin dramatists took care not to
offend in this way ; but in Plautus we still find a strong Roman
flavour of social satire and topical allusion. After Plautus'
day there is a change ; perhaps it was introduced by Caecilius,
but by the time of Terence at latest the palliata seems to have
renounced its old freedom to introduce topical allusions.
Aesthetic questions were now being vigorously discussed, as
the prologues of Terence indicate ; fidelity in translation
was felt to be important. It may be that the incongruity
between a Greek setting and Roman allusions came to be
regarded as objectionable. But in renouncing Roman topics
the palliata left the field open for the rise of a new, native
type of comedy, which would try to invent its own plots and
find its characters in Italian life.
The known writers of the fabula togata or ' comedy in
native dress' are Titinius, Afranius and Atta. Titinius is
usually supposed to have been the pioneer. His date is not
precisely known ; he may have been a contemporary of
Plautus, he may have been a successor of Terence. Afranius
certainly wrote after the time of Terence, for whose loss he
expresses regret; Vellerns, in different passages, speaks of
him as a contemporary of Pacuvius, Caecilius, Terence and
Accius. Atta is said to have died in 77 B.C. When we compare
T20
 NATIVE COMEDY : THE FABULA TOGATA 121

these dates with those of the palliata, the last author of which
was Turpilius, who died in 104 B.C., and when we further
reflect that the classics of the palliata were still familiar to the
theatre-going public in the days of Cicero, we may begin to
feel doubtful about the modern view that the togata owed its
origin to a public reaction from foreign comedy. The togata
arose while the palliata was still in its heyday 1 ; so far from
its driving the palliata off the stage, the two forms of comedy
long existed side by side. Greek comedy was the more popular
and important of the two, and much of its popularity was due
to the very fact that it was exotic (and did not incur any danger
of irritating native susceptibilities), but native comedy was at
least more successful than native tragedy. The praetexta was
never anything more than a parergon for writers of tragedy,
and the total number of certain titles can almost be counted
on the fingers of one hand. But the togata gave full-time
employment for three writers, and has left to us about seventy
titles.
The term togata could theoretically be applied to all
drama not derived from Greek sources, but usually it was
confined to comedy. The fate of Naevius warned its writers
not to make fun of the governing classes in Rome ; conse-
quently the togata finds its themes in the country towns
of Italy, 2 or among the ordinary folk of no specified town.
In this workaday world the ceremonial toga was not really the
characteristic wear, and the term togata was hardly a happy
description for the drama which dealt with the lives of such
folk. Accordingly we find as a synonym for togata (in the
sense of ' native comedy ') the term tabernaria, ' private-house
comedy'. Tabernae in early Latin meant ' private houses
in particular the houses of poor people, to which class most
(though not all) of the characters of the togata belong.
Of Titinius little is known. Varro praised the character-
drawing (or moral tone ?) of Titinius, Terence and Atta ; if
this order is to be taken as chronological (which is very
doubtful), he must have been earlier than Terence. He has
left us fifteen titles : Barbatus (' The Man with a Beard '),
1
But I do not feel convinced by the view of some scholars (e.g. Warmington,
Old Latin II. xv.) that togatae were written by Naevius—far less by Donatus' state-
ment (Com. iv. 5) that they were invented by Livius Andronicus !
2
Cicero seems to use the term togati of the Latins (as contrasted with the
Romans) : De Orat. iii. 1 1 . 43. See Appendix (d).
122 THE ROMAN STAGE

Caucus (' Blind '), Fullonia (' The Fullers'), Gemina (' The
Twin Sister'), Hortensius, Iurisperita (' The Lady Lawyer'),
Priuigna (' The Stepdaughter ') Procilia, Psaltria siue Ferentinatis
(' The Music Girl, or the Girl of Ferentinum '), Quintus,
Setina (' The Girl of Setia'), Tibicina (' The Flute-girl'),
Varus, Veliterna (' The Girl of Velitrae '), and Vlubrana (' The
Girl of Ulubrae '), with about one hundred and eighty lines.
C. Quinctius Atta is said by Jerome to have died at Rome in
77 B.C. and to have been buried at the second milestone on the
Via Praenestina. Varro, as we have seen, thought highly of
his character-drawing (or moral tone), and Fronto adds that
he excelled in feminine conversation. We have from him
eleven titles : Aedilicia (' The Aedile's Games'), Aquae Caldae
(' Taking the Waters '), Conciliatrix (' The Match-maker ' or
perhaps ' Procuress ') Gratulatio (' Well done ! '), Lucubratio
(' Up by lamplight'), Matertera (' Auntie '), Megalensia (' The
Games of the Great Mother'), Satura (' What you will'),
Socrus (' Mother-in-law '), Supplicatio (' The Entreaty '), Tiro
Proficiscens (' Off for his first campaign'), and about twenty
lines. L. Afranius seems to have been the most important
writer of togatae. We have forty-four titles : Abducta (' Gretna
Green '), Aequales (' Chums '), Audio (' Going—going—
gone ! '), Augur (' The Augur'), Brundisinae (' The Women of
Brundisium'), Bucco Adoptatus (' The Clown's adoption'),
Cinerarius (' The Hairdresser '), Compitalia (' The Crossroad
Festival'), Consobrini (' Cousins '), Crimen (' The Accusation '),
Deditio (' The Surrender'), Depositum (' Left on Trust'),
Diuortium (' The Divorce '), Emancipatus (' The slave set free '),
Epistula (' The Letter '), Exceptus (' Rescued from drowning '),
Fratriae (' Sisters-in-law'), Ida (' She who gets slapped ' ?),
Incendium (c Fire ! '), Inimici (' Enemies'), Libertus (' The
Freedman'), Mariti ('Married Men'), Materterae ('Aunts'),
Megalensia, Omen (' The Omen'), Patelia (' The Grain
Goddess ' ?), Pompa (' The Procession'), Priuignus (' The
Stepson '), Prodigus (' The prodigal Son '), Proditus (' Betrayed')
Promus ('The Butler'), Prosa (?), Purgamentum ('Trimmings'),
Repudiatus (' The rejected Lover ' ?), Sella (' The Chair '),
Simulans (' Not what he seemed to be '), Sorores (' The Sisters '),
Suspeda (' The suspected daughter ' ?), Τ alio (' Tit for tat'),
Temerarius (' Neck or Nothing'), Thais, Titulus (' The
Inscription ' ?) Virgo (' The Maid '), Vopiscus (' The Twin
 NATIVE COMEDY : THE FABULA TOGATA 123

that lived'), and over four hundred lines. Afranius, as he

himself admits, drew freely on the works of other dramatists,
whether Greek or Roman ; he thought highly of Menander
and Terence, and may have resembled them in some ways.
Quintilian regrets to admit that Afranius made use of boy-
love 1 as a dramatic theme, thereby (according to Quintilian)
revealing his own character. To suppose that a dramatist's
private life necessarily corresponds to that of any of his
characters seems uncritical, but Quintilian's testimony that
sodomy was employed as a dramatic motive by Afranius
is supported by Ausonius.2 Cicero speaks of Afranius' wit,
and says that he modelled his style on that of the orator and
tragedian C. Titius (whose date is itself, unfortunately,
uncertain). It would appear not only that Afranius was a
more gifted and fertile writer than Titinius or Atta (though not
equal to them in moral tone ; but perhaps Varro also was
thinking of his introduction of sodomy as a theme) but that
he cast his net more widely than they in search of subjects.
Our main evidence, that of the titles and fragments, does
not enable us to distinguish clearly between the work of these
different writers of togatae, or to reconstruct with probability
as much as a single plot. It would appear that a togata could
be of some complexity. The Vopiscus of Afranius introduces
the parents of the ' twin that lived ', a newly-married pair
(scarcely to be identified with these parents), a husband whose
wife has left him, a woman who feigns submission while really
playing on the violence of a man's temper, a trusted parasite,
a pampered slave, a door-keeper, a lady's maid. The Fullonia
of Titinius shows us a wrangle between fullers and weavers,
a wife who complains that her husband is squandering her
dowry, and a philanderer who, in fear of capture by an angry
husband, is about to commit himself (like brave Horatius) to
the Tiber. Afranius' Simulans shows some one advising a man
to ' feign ' anger with his son-in-law and pretend that he
wishes to take his daughter home ; we see a father and mother
quarrelling in the presence of their small child, some one
angrily abusing a prodigal, a would-be peacemaker and a
slave who is being turned out of doors. On the other hand we
have a statement by a scholiast that the cast of a togata,
like that of an Atellana, was smaller than that of a palliata ;
1 2
Antiphanes wrote a Paederastes and Diphilus a Paederastae. Ep. lxxix.
124 THE ROMAN STAGE

and on general grounds we should have suspected than an

average togata would be both shorter and simpler than the
borrowed type of comedy. No doubt in many cases the plot
of a togata was modelled on that of a palliata, though its
characters were necessarily taken from Italian life.
Sometimes at least there was a prologue ; in an unidentified
play of Afranius the god Priapus utters what appears to have
been an explanatory prologue, while Afranius' defence of
his literary methods, quoted from the Compitalia, presumably
occurred in the prologue, which must therefore have been
polemical in tone, after the fashion of Terence—whom
Afranius admired. The play sometimes opened in the early
morning (Epistula of Afranius, Lvcubratio of Atta). There was
probably no chorus ; Cicero's remarks (pro. Sest. 118) show
that the ' troupe', t h e ' actors' and t h e ' singers' were identical.
Though on the occasion mentioned by Cicero the actors
uttered in unison a denunciation of one of the characters,
directing their gaze significantly to where Clodius sat in one of
the front seats, this was evidently a ' put-up job '. The stage
represented as usual a street in front of some houses (Gratulatio,
Epistula) with the country lying on one side-entrance (Diuortium)
and no doubt, the centre of the town in the other direction
(Hortensius). The roof of the stage-building was perhaps
used in Titinius' Setina. The usual tricks of eaves-dropping
(Priuignus), of slaves swaggering along the street (Pompa), of
doors opening, characters appearing precisely as their names
happen to be mentioned, sounds of tumult ' w i t h i n e t c . ,
all occur.
The stock plots of the palliata turn on peculiar features of
Greek life, and would seem out of place in the togata. We are
told by Donatus that in the togata slaves were not allowed to
be shown as cleverer than their masters ; this typical theme of
the palliata had only been made tolerable by the exotic nature
of that type of drama. Very different was the discipline which
ruled in a Roman household. It would seem, then, that in the
togata we must have none of those swindling slaves who extract
money from their old masters in order to pay the expenses of
their young masters' amours. Yet in Afranius' Incendium the
old gentleman tells us, apparently in a prologue, that his son
and his slave have laid a plot, presumably in order to swindle
him. The wheels of the palliata were set in motion by love ;
 NATIVE COMEDY : THE FABULA TOGATA 125

yet we have seen how difficult it was for a respectable Athenian

girl to appear in public. In Roman life this convention did
not exist, at least to anything like the same extent ; young
women of respectable class can appear in public, can meet
eligible suitors, can accompany their fiances on a social call.
Hence all the devices whereby New Comedy succeeded in
enabling girls to meet their future husbands—kidnapping,
exposure of children, recognition-scenes, midnight festivals—
seem unnecessary ; mere mercenary gallantries too, need no
longer form the stuff of drama ; we find ourselves in an
atmosphere of family life not unlike our own, where marriage
is to some extent based on mutual attraction, and where
relatives, female as well as male, express their opinion about
a projected match. This does not mean that the togata could
provide a true love-plot, or even give love as important a place
as it occupies in New Comedy. Romance is impossible if the
course of love can run smooth ; the authors of the togata, if
they really desired to introduce a love-interest, would have to
invent their own ways of preventing lovers from winning
happiness too soon. There is no sign of a romantic interest
in the fragments of Titinius' Setina, where the reluctant suitor
doubts the wisdom of aspiring to a maiden of superior fortune,
or in the family councils of the Fratriae, the calculation that
a pretty girl will need less dowry and the hope for possible
advantages in kind if she marries the confectioner. Indeed
Afranius seems to be forced to resort to the themes of the palliata
in order to secure a sentimental interest ; the heroine of his
Exceptus is a Greek courtesan from Naples, her lover is for some
reason induced to attempt suicide, and her tender reproaches
reveal to him the depth of her love.
Another method of introducing a sentimental interest was
the enforced separation of a happily married couple. In the
Diuortium a father forces his daughters to leave their husbands,
in the Simulans he is advised to follow a similar course, in the
Vopiscus a husband seems to express his earnest but vain hope
of inducing his wife to return home. These are all plays of
Afranius. In general one would suspect that the togata was
not sentimental but lively, amusing and satirical (though prone
to occasional sententiousness). The opening scene of Afranius'
Epistula shows some eager gallant or belated reveller, bare-
headed and be-slippered, out of doors in the cold air of a winter's
126 THE ROMAN STAGE

dawn ; the other fragments of this play give us glimpses of a

drinking-party, of some one escaping from a street bully, of
some one dressed up as a girl, of a girl who can hardly restrain
her laughter while her mother storms in fury and of a general
state of domestic turmoil. In other plays we hear of husbands
trying to play the gallant—a stock theme in the palliata, it is
true, but in the togata there is this curious difference : the
erring husband chooses not the town but the country as the
setting of his amours. Here, perhaps, we have a concession to
Roman notions of propriety ; illicit gallantries must not be
flaunted in the Roman forum. Afranius' introduction of the
theme of sodomy is remarkable ; it shows that this type of
drama was not fettered by the sexual conventions which the
palliata had taken over from New Comedy. I n the palliata,
to borrow a phrase from the Curculio1, a man may love any one
except' maiden, wife or widow, youths or boys'; in other words
love affairs can only occur between the hero and a courtesan
or foreigner. In the togata not only do we have respectable
alliances between social equals, but we find the darker themes
of adultery, involving respectably born women, and of un-
natural vice. What we should like to know is how a coherent
plot was built up round such themes—but this is just what the
fragments do not tell us. It is sad to have to admit that in the
whole range of Latin drama we have no knowledge of the course
of as much as one plot invented by a Latin writer. 2 Perhaps the
togata did not contain a real plot (unless one borrowed from
the palliata), but was merely a series of loosely-connected
scenes taken from the everyday life of the streets, the shops,
the private houses of Rome or the country towns ; if so, it
may to some extent have resembled the old dramatic satura ;
and indeed Satura is the title of one of Atta's pieces.
What the titles and fragments do show is that the togata
depicted the life of ordinary Italian folk, especially of the lower
classes. Titinius' Fullonia dealt with the fullers, a favourite
butt for comedians—possibly because of the fact that one of
the cleansing agents they employed was urine. One of the
fullers complains' We fullers have no rest, day or n i g h t w h i l e
a weaver retorts ' But for our weaving you fullers would be
out of work '. In Titinius' Barbatus an embroiderer throws u p
his job, leaving his needle and thread to his master and mistress.
1 2
11. 36-7. I am speaking of drama intended for the stage.
 NATIVE COMEDY : THE FABULA TOGATA 127

In Afranius' ' B u t l e r ' we see two servant-maids wrangling ;

in one of Titinius' plays a drunken servant-girl is detected in
the act of stealing wool. People of higher rank occasionally
appear—the heiress and the lisping dandy in Titinius' ' Girl
of Setia ', the mistress of the house in his ' T w i n Sister The
personal names are usually typical R o m a n names, the local
names those of country towns : Titinius wrote a ' Girl of
Setia ', a ' Girl of Ferentinum ', a ' Girl of Velitrae ' ; Atta
wrote about a spa called ' A q u a e C a l d a e ' ; Afranius wrote
' T h e Ladies of Brundisium '. Y e t titles such as Megalensia
(Atta and Afranius), Compitalia (Afranius), which are the names
of R o m a n festivals, and references to the Tiber or to the custom
of throwing puppets into that river, would suggest that the
writers of togatae, for all their prudence, could not altogether
forgo the introduction of the capital. Afranius seems to have
introduced Greek names at times ; one piece of his deals with
a courtesan named Thais, another with a courtesan called
Moschis ; he quotes a Greek proverb about how Amyclae fell
through keeping silence ; and many of the familiar characters
of the palliata appear, especially in Afranius—parasites and
pimps, cooks and slaves and courtesans. Nevertheless the
prevailing atmosphere, even in Afranius, is Italian. Titinius
makes fun of the people of Ferentinum because of their craze
for things Greek ; Numerius, a character of Afranius, laughs
to scorn any Greek mannerism in conversation. Nothing can
be more R o m a n than the contemptuous reference in Titinius'
Quintus to people who speak Oscan or Volscian because of their
ignorance of Latin.

T h e togata must have enjoyed some popularity, at least

during the lifetime of its authors, for we have the titles of
seventy pieces. Revivals at a later period are not often
recorded ; the Simulans was performed in Cicero's day (was
its choice influenced by Clodius' enemies ? ), the Incendium was
staged in Nero's reign—but chiefly, we gather, in order to give
a spectacular display of the fire ; actors were allowed to keep
any property they could rescue from the burning buildings.
W e may suspect that the choice of old plays for revival was
often dictated by motives which had no connextion with the
dramatic merits of the play chosen. T h e purely literary
togata continued to be written down to Juvenal's day (but
perhaps Juvenal was one of those who included under this
128 THE ROMAN STAGE

title all plays on native themes). Melissus, a freedman of Mae-

cenas, invented the trabeata, a special type of togata dealing
with middle-class life (the trabea was the characteristic wear
of the equites.) One is reluctant to admit that Melissus can
have satirized the class to which his famous patron belonged;
perhaps the trabeata was akin to the praetexta, and dealt
kindly with the habits of Maecenas and his fellow-knights.
In any case, it was probably nothing more than a short-lived
literary curiosity. For the survival of the togata on the stage
we have no evidence ; Suetonius refers to an actor named
Stephanio, of the time of Augustus, who is called a togatarius—
but apparently what is meant by this term is that Stephanio
was the first to wear a toga when dancing in pantomime.
 CHAPTER XVI

POPULAR F A R C E : T H E FABULA ATELLANA

I N TRAGEDY the Romans of the Republican period were

content to adapt Greek originals ; for the Roman historical
play seems to have had little importance. But in the restless
search for new means of amusing the public, the Roman
dramatists gave literary form to certain popular forms of farce,
which had existed from early times as improvised performances
but were now, perhaps for the first time, to be set down on
paper and, in due course, published. Our information
concerning the Atellana and the mime is fragmentary ; their
importance in literature was probably small ; but we have
reason to believe that in the life of the common people, from
early times to the end of the Roman Empire, popular farce
played a greater part than all the literary forms of Roman
drama put together.
In antiquity, as in modern times, the Campanians had a
reputation for jest and merriment. Horace has left us an
account of how two Campanian buffoons indulged in a sparring-
match which was marked by high spirits, lively repartee and
merciless personalities. At an early period there seems to have
been developed in Campania a rustic farce which displayed
certain traditional characters in ridiculous situations. The
fabula Atellana is said by Diomedes to have derived its name
from a town of the Osci called Atella, where it had its origin.
Atella was about nine miles from Capua, on the road to Naples;
Oscan was an Italic dialect, akin to Latin, which was spoken
by the population of the southern Apennines who, in the
fifth and early fourth centuries, overran the greater part of
southern Italy, including Campania. It would seem, therefore,
that the Atellana was a form of entertainment popular among
the Oscan-speaking people of Campania, including the
inhabitants of Atella, and that the Romans, on becoming
acquainted with it, called it either ' Oscan farce' after the
district as a whole, or ' Atellane' after the town in that
i»9
I30 THE ROMAN STAGE

district—perhaps the town from which they had first derived

their knowledge of it. (We may remember that the ' Fescenn-
ine verses' derived their name, according to one Roman
theory, from the small Etruscan town Fescennium).
One would certainly imagine that the ' Oscan farce ' must
originally have been presented in the Oscan language. One of
the stock characters in ' some Atellane plays the ' Grand-dad '
Pappus, is said by Varro to have been called Casnar by the
Osci. This implies that the Atellanae survived in Oscan in
their original home after they had been introduced in Latin
to Rome. Strabo, indeed, tells us that though the Oscan race
had by his day died out, their dialect still survived among the
Romans, and was used for the ' staging and miming of poems '
in a ' certain native f e s t i v a l B u t we have other evidence that
the Oscan dialect was not intelligible to the ordinary Roman ;
and it is difficult to believe that the Atellanes were ever
performed at Rome—presumably for public entertainment—
in an unintelligible tongue. Strabo may have been misled by
the name ' Oscan farce' into imagining that the farce was
presented at Rome in Oscan ; or he may have misunderstood
some reference to the peculiar Latinity of the literary Atellanes
—which, judging by the fragments, was markedly rustic and
crude. Juvenal uses the term Oscan (in its Greek form Opici)
as ' b a r b a r i a n ' V a n d a l a n d Cicero's reference to the
transactions of some town council in Campania as ' Oscan
farces ' may also suggest that' boorish ' is what the term meant
to the Romans of his day. As for the name Atellana and the
connexion of the farce with the town of Atella, we cannot now
hope to know the facts. The explanations offered by Livy and
Diomedes have the air of plausible guesses, based on nothing
but the names ' Oscan farce' and ' Atellane play'. The modern
theories are purely theoretical; we have no evidence that
Atella played any particular part in the development of the
farce, or that the inhabitants of Atella were regarded as suitable
butts for jest. 1 Nor can we tell how far Greek influence affected
the development of the Atellana. Strabo's language shows that
he thought of it as a kind of mime ; but the Romans seem to
have drawn a sharp distinction between the Atellana and the
mime, and in its essential feature it differs from any Greek
performance known to us.
1
Mommsen, History of Rome. iv. 13.
 POPULAR FARCE: THE FABULA ATELLANA 131

This essential feature was the presence of certain traditional

characters. No ancient writer has left us a list of these char-
acters, but from the extant titles and fragments of the written
Atellanae we collect the names of four or five : Maccus,
Bucco, Pappus, and Dossennus, who is thought by some
modern authorities to be identical with Manducus. All these
characters seem to have had a certain family resemblance as
coarse, greedy clowns, whose animal characteristics were such
as might amuse a primitive and rustic audience, ever ready
to laugh at gluttony and drunkenness, at horse-play and obscene
jest. Attempts have been made to distinguish them more
clearly from each other by arguments based on etymology.
We know from Apuleius (Apol. 81) that Maccus and Bucco
were fools ; Plautus also (Bacc. 1088) uses ' B u c c o ' as a
synonym for ' fool'. Does Maccus mean ' stupid', or
' guzzler' ? Aristophanes (Eq. 396) uses μακκοΰ in the sense
'moon', 'be stupid'; and in the dialect of Sardinia the word
maccu means ' stupid'. But the word macco was once
used in Italian in the sense ' pap ' , 1 If Bucco comes from
bucca and means ' fat-cheeks ', does that indicate greed, or
stupidity, or boastfulness ? Pappus is the Greek πάππος,
' grandfather', a word used by Pollux in his list of comic
masks—' first grand-father ', ' second grand-father' and so on.
We have quoted Varro's statement that ' in some Atellanae
" old Pappus " is what the Oscans call " casnar " '. In the
literary fragments Pappus appears as the Old Fool. Are we
to derive the name ' Dossennus' from ' dorsum and to con-
clude that Dossennus was a hunch-back ? There is not a scrap
of evidence to support this view. That Dossennus was greedy
appears from a line of Horace concerning Plautus :
quantus sit Dossennus edacibus in parasitis,
' how monstrous a Dossennus Plautus appears among his
gluttonous parasites', which I take to mean ' what a super-
glutton Plautus shows himself' (because of his eagerness to
make money). In the belief that his name means ' hunch-
back ', and that hunchbacks are popularly regarded as
cunning, some scholars regard him as the Wise Fool, and press
into service certain of the fragments to support this view. As
1
See Walde, Lat. et. Wort (1940), and Meyer-Lübke, Rom. Et. Wort. s.v. The
difficulty is to find any historical connexion between the word meaning ' clown'
and the word meaning ' par '.
132 THE ROMAN STAGE

for Manducus, V a r r o tells us that his name comes from mando,

' c h e w ' : dictum mandier a mandendo, unde manducari, a
quo in Atellanis ad obsenum uocant Manducum. Here
Müller emends ad obsenum to Dossennum, identifying these two
characters, but there is no other evidence for such an identi-
fication. However that may be, Manducus is beyond doubt
the ' C h e w e r ' , an ogre with champing jaws. A passage in
Plautus (Rud. 535-6) transports us from this world of theoriz-
ing into the unmistakable atmosphere of a popular festival at
Rome. Labrax, his teeth chattering with cold, thinks o f ' hiring
himself out as a Manducus at the g a m e s ' . Juvenal gives us a
glimpse of a festival in some country town, with its grass-
grown theatre, and on the stage the ' traditional f a r c e ' with the
' hideous gaping m a s k ' which makes the terrified youngster
seek refuge in his mother's arms. Festus, explaining the word
' manducus tells us that in ancient festivals, along with other
alarming and ludicrous figures, an ogre appeared with huge
jaws, gaping mouth and chattering teeth. From the slang
manducare, ' champ the j a w s ' comes the Romance word for
' e a t ' (French manger). T h e Atellane farce has indeed left
a record behind it in the languages of modern Europe.
These are the characters of whom we hear. N o doubt
they did not always appear together. Pappus, according to
Varro's remark (quoted above), was seen in ' some Atellanae '
only. Indeed we have no evidence that any two of them took
part in the same performance except one of the literary
fragments, which shows Dossennus as present in a piece
entitled ' Maccus the Maiden
Such stock roles would, no doubt, require stock costumes
and masks. Indeed without a mask it would not be possible
to represent Manducus' huge champing jaws. Festus tells us
that the Atellane performers were specially denoted by the
title personati, as they were allowed always to wear their masks
on the stage, whereas other actors were compelled to lay theirs
aside. Perhaps Festus is here (in somewhat confused language)
contrasting the Atellani with the only other comic actors widely
known in the imperial age, the unmasked performers of mime.
T h e use of the mask, rendering play of feature impossible, would
make expressive gesture all the more i m p o r t a n t ; Juvenal
(vi. 72) refers to the ' gestures' of an Atellane actor, and
Tertullian {De Sped. 17) speaks of Atellani gesticulatores.
 POPULAR FARCE Г THE FABÜLA ATELLANA I33

Atellane pieces were apparently of no great length ; they are

referred to as Atellaniolae, ' little A t e l l a n e s t h e number of
characters in Atellanes, as in togatae, is said to have been
' rather s m a l l a n d their use as ' exodia ' or ' after-pieces '
would suggest that they corresponded in length to our
' curtain-raisers'. No doubt they were more or less impromptu
performances until they received literary form in the days of
Sulla. Nevertheless the Atellana had a plot containing
surprising complications, tricae Atellanae as they were called
(our word ' intrigue ' is derived from this technical expression).
But how could a short, impromptu performance have a
complicated plot ? Indeed a figure like Manducus seems to
have been more like an ogre in a dumb-show than a dramatic
character. Some light is thrown on this problem by the titles
and fragments, which show that a recurring motif was disguise
and masquerade. ' Maccus the Maiden ' might quite quickly
get into a complicated situation, reminiscent of the outrageous
adventures of Chalinus dressed as a bride in that very Plautine
comedy, the Casina. Even where there was no written
dialogue, we may suppose that the actors had been coached in
the outlines of the plot and were left to develop its humorous
possibilities as the inspiration of the moment might suggest
to them. They must often have fallen back on stock themes
which would allow them to show off their command o f ' patter'.
Indeed much Plautine dialogue is of this nature. We learn that
they were fond of introducing topical allusions, and of alter-
nately propounding riddles and trying to guess the answers.
What did consistency of plot matter, so long as the crowd
were amused ?
Livy's well-known summary of Roman dramatic origins,
however untrustworthy as history, can scarcely be ignored,
especially when confirmed by other evidence, or when it deals
with conditions prevailing in his own day. He tells that the
young citizens of Rome, when forced to abandon their amateur
dramatics by professional competition after the time of Andro-
nicus, revived the old Fescennine exchange of repartee ; this
later received the name ' afterpiece' and was blended with the
Atellane play, a form of drama which was never allowed to
get into professional hands. Hence in Livy's own day the
Atellani retained citizen status and served in the army ' as if
they had nothing to do with the s t a g e W e may well doubt
10
134 THE ROMAN STAGE

his ascription of amateur theatricals to third-century Rome ;

indeed the passage already quoted from Plautus shows that
the Manducus was paid for his services. In Livy's day the
theatre was largely given over to the mimes, who were all
infames. Possibly the Atellani was regarded as superior to the
mimes, in view of their long traditions. By Tacitus' day
this distinction seems to have disappeared. But from early
times all who performed for the amusement of the public,
whether ludii, histriones, Atellani or mimi, must surely
have done so for pay.
The Atellana was well known to Plautus, who refers to
Bucco and Manducus. Whether or not we identify with
Plautus t h e ' Maccus ' who wrote the Asinaria and the ' Maccus
Titus ' who wrote the Mercator, we cannot help wondering
whether this Maccus owed his name to having been an actor
in Atellanes. The use of the Atellana as an exodium illustrates
its popularity and its amusing character. Cicero, when dealing
in his De Oratore with the subject of jest, turns for examples
to the Atellanae of Novius. When Paetus, Cicero's elegant
friend, after mentioning a tragedy, the Oenomaus, turns to
jest, Cicero's reply is ' after the tragedy you have introduced
as exodium not, as was once the custom, an Atellana, but a
mime'. Paetus' polished jesting evidently savours of the town,
not the country, and is therefore compared not with the rustic
Atellana but with the sophisticated mime. Nevertheless the
Atellane survived in imperial times ; Juvenal shows that it
was still given as an exodium. The Roman grammarians
compare it in this respect with the Greek satyric play, and
like the satyric play it came to include not only rustic elements
but burlesque of mythology.
In order to form a closer acquaintance with the Atellane
of the late Republic we turn to the written Atellanes of
Pomponius and Novius. But the assumption of literary form
necessarily altered the character of this rustic farce ; Pomponius
and Novius could not help being influenced by other forms of
drama, and though they strove to retain the traditional
characters and the traditional crudity of jest and language,
the surviving titles and fragments remind us at times of the
palliata, the togata, the burlesque and the mime.
 CHAPTER XVII

THE LITERARY ATELLANA

L P O M P O N I U S of Bononia, described by Vellerns as the

• founder of the Atellana, is said by Jerome to have been

active in 89 B.C. Novius, who is mentioned with equal respect,
is quoted by Cicero in the De Oratore, the dramatic date of
which is 91 B.C. Evidently these two leading writers of Atellanae
were thought of as contemporaries. A t the beginning of the
first century farce seems to gain at the expense of higher forms
of drama ; from now on we hear of no writer who made a
living out of the writing of tragedy or comedy for the stage.
From Pomponius we have seventy titles, with fragments
amounting to nearly two hundred lines or parts of lines ;
from Novius we have forty-four titles with over a hundred
lines of fragments. T o distinguish between the styles of
these two writers would seem impossible, so meagre is our
information.
T h e large number of surviving titles supports our other
evidence that the pieces themselves were comparatively short;
a pair of titles like ' Hog(g) Sick ' and ' Hog(g) W e l l ' suggests,
indeed, that these two pieces were meant to be performed on
the same occasion. The most obvious Atellane feature of the
titles and fragments is the frequent reference to the stock
characters. Maccus appears to be the most popular figure ; we
have such titles as ' Maccus the S o l d i e r ' Maccus the Inn-
keeper ' Maccus the M a i d ' Maccus the E x i l e ' T h e T w i n
M a c c i ' ; apparently ' Maccus ' alone was a title which could
attract a crowd, as both Pomponius and Novius wrote plays
with this title. W e have also as titles ' Bucco the Gladiator
' Bucco adopted ' Pappus the F a r m e r ' T h e Bride of
P a p p u s ' Pappus Defeated at the P o l l ' (a title which
attracted both authors), and ' The T w o D o s s e n n i B u c c o ,
Pappus and Dossennus are also mentioned in the fragments,
and a fragment of Pomponius' Pictores appears to mention
Manducus (spelt Manduco) though perhaps only as an equiva-
13s
136 THE ROMAN STAGE

lent for ' glutton \ Other titles point to the maintenance of

the primitive rustic atmosphere : ' The P i g ' The Sow ',
' Hog(g) Sick ' Hog(g) W e l l ' The Farmer ' The Wood-
pile ' A-hoeing ' The V i n e - g a t h e r e r s ' The She-ass
' The S h e - g o a t O t h e r titles, again, suggest the life of the
town : ' All in W h i t e o r ' the C a n d i d a t e ' The Fullers
' The Inspector of Morals ' The Pimp Such characters
might easily be imagined in any small town of Campania, and
indeed some of the fragments remind us of the scribblings on
the walls of Pompeii. Here, then, the Atellana approached
the boundaries of the togata. Yet other titles remind us
of the palliata and New Comedy : Synephebi, Adelphi, Hetaera ;
a title like ' The Boy Favourite' may suggest kinship with
the Hellenized togata of Afranius or with the mime ; finally
there are titles which point to burlesque of tragedy : ' The
counterfeit A g a m e m n o n M a r s y a s , Phoenissae.
It would seem that the earliest writers to give the Atellana
a literary form, finding that the antics of Maccus and his
fellows were not sufficient material in themselves, were forced
to draw upon other sources. Nevertheless the traditional
element was preserved, at least in some of these pieces. Maccus,
in particular, appears in various situations, in which, we may
be sure, he conducts himself with due crudity and absurdity.
The fragments of ' Maccus the Soldier' refer to gluttony.
' Maccus the Exile' bids farewell to a door (presumably that
of a host) in terms which seem to echo Plautus. In Plautus'
Mercator the young Charinus, before departing for a far land,
takes tragic farewell of his father's door :

limen superum inferumque salue, simul autem uale !

So Maccus exclaims :
limen superum, quod mi misero saepe confregit caput,
inferum autem, ubi ego omnino omnis digitos defregi meos:

' Lintel, whereon I have often banged my unlucky head, and

threshold, where I have many a time broken all my toes
That Maccus might be dressed up as a Maid is shown by the
title Maccus Virgo ; and the feelings of some one who is deceived
by such a trick are expressed in a surviving fragment of the
Macci Gemini of Pomponius. In the ' Kalends of March ' of
 THE LITERARY ATELLANA 137
Pomponius we find some one being coached in the art of
adapting his voice to his woman's role. Elsewhere we find
characters disguising themselves by simply drawing their
cloaks over their heads ; we recall the similar use of the
ricinium or hood in the mime. W e can imagine that ' Bucco
the Gladiator ' was but a caricature of the Matador, so popular
with women as well as with men :
occidit taurum toruiter, me amore sauciauit.
' Bucco Adopted ' may deal with the theme ' If I were king ' ;
sudden changes in fortune were a favourite mime-theme
(modo egens, repente diues, as Cicero puts i t ; cf. Juv. x. 602-8).
Bucco's clownish wit appears when, on being asked to ' handle
the j o b cleanly he replies,' I have already washed my hands
Such over-literal interpretations are a stock form of jest in
comedy. T h e fragments do not allow us to determine the
character of Bucco more closely.
Pappus is beyond doubt the Old Fool. In him, as the
scarcely translatable title Hirnea Pappi shows, the disadvantages
of old age were satirized with true R o m a n vigour. His defeat
at an election was the theme of plays by both dramatists.
Pomponius shows him being consoled by the suggestion that
the crowd will alter their decisions later on ; in Novius' play
some young man informs him with brutal candour that his
electioneering will bring him not to the chair of office but to
the coffin. It is no doubt Pappus who, when asked ' W h y are
you complaining, D a d ? replies drily ' Would you have me
sing ? I ' m condemned '. Pappus is no doubt the c worthless
old f e l l o w ' of Pomponius' Praeco Posterior, whose young wife
is to be induced by her stepson to leave him ; in spite of his
early sacrifices at the temple of Venus (presumably for success
in love), a clown brings him news of disaster. Pappus is
naturally unfortunate as a husband ; his activities as a farmer
are interrupted by the vivid account of his wife's misconduct
(Pomponius' Pappus Agricola). T h e Romans seem to have
taken particular pleasure in the exhibition of dotard passion ;
Plautus makes full use of this theme, and perhaps our accounts
of the latter days of Tiberius have been coloured by the R o m a n
taste for this particular form of scandal. Manducus (in the
form Manduco) occurs in Pomponius 112. It is not clear that
any of our literary fragments refers to Manducus as a character,
ι38 THE ROMAN STAGE

but it can hardly be without significance that we find the

writers of Atellanae using manduco for ' guzzler ' and manducari
for ' champ the j a w s D o s s e n n u s was evidently greedy ;
in Pomponius' Campani some one proposes to distribute food to
' Dossennus and the fullers'.
In Novius' Duo Dossenni some one is called ' The terror of
the winejar'. In Pomponius' Philosophia Dossennus is asked
to say who stole the gold ; his reply is that he expects to be
rewarded for his information. In the Maccus Virgo he is seen
behaving in a very unseemly way to one of the pupils in his
school. Seneca mentions an inscription commemorating his
wisdom :
hospes, resiste et sophiam Dossenni lege.

On these references to him as a detective, a schoolmaster and

a wise man, coupled with the supposed meaning of his name—
' hunchback'—has been founded the theory that Dossennus
was the Learned Man, Scholasticus, whose cunning was indi-
cated by his hunch. But there is no necessary connexion,
even on the stage, between cunning and deformity, and
Dossennus' appearance as a schoolmaster may have been no
more characteristic of him than Maccus' appearance as a
Maid in the same play. It would appear, then, that we know
nothing about Dossennus except that he was probably greedy.
It is not possible to distinguish clearly between the different
characters of the Atellanae. Pappus was elderly and Manducus
was an ogre ; all were greedy and clownish.
The rustic origins of the Atellana are further indicated
in the homeliness of the language. A rustic defines wealth as
' a short-lived blessing, like Sardinian cheese '. The frequent
references to bodily functions may recall Aristophanes, but
contrast the Atellana with the plays of Plautus and Terence.
Rustics in Atellanae referred to a scortum (' wench ') as ' a bit
of skin ' (pellicula) ; indeed this type of drama had a vocabulary
of its own. The rustic character of the Atellana may have
helped to suggest its use as an after-piece, on the model of the
Greek satyric drama, which also was rustic in tone. But by
natural reaction the satyric drama seems to have communicated
its element of mythological burlesque to the Atellana, as
many of the titles indicate. The composers of this farce were
not unduly scrupulous about preserving its native quality,
 THE LITERARY ATELLANA 139
provided they could amuse their patrons. All forms of light
drama influenced each other. In the Atellana we meet familiar
characters of New Comedy such as the parasite and the pompous
quack-doctor ; elsewhere we find ourselves in the atmosphere
of the togata ; but perhaps most frequently we are reminded
of the mime, and indeed the Atellana must have tended to
become a form of masked mime. The title of one piece,
Exodium, reminds us of the use of Atellanae in general as ' after-
pieces ' ; another title, Satura (if it does not mean ' The Fat
Woman ') may suggest that this native farce was felt to be a
' M e d l e y ' or ' What you w i l l N o v i u s ' Mortis et Vitae
Iudicium is curiously reminiscent of the (non-dramatic) Satura
of Ennius which showed Mors and Vita engaged in debate.
Staging was probably modelled on the palliata. That the
usual three houses were shown in the background is suggested
by the fragment of the Pictores ' Here in the middle lives
Pappus, a worthless old fellow The metres used in the frag-
ments seem also to have been those of the palliata ; a gram-
marian tells us that the jaunty and rather vulgar iambic
septenarius was especially prominent in the Atellana. As
contrasted with the palliata, the Atellana was vulgar, trivial,
homely ; as contrasted with the mime, the Atellana was a
masked, rustic performance. Even when its themes were
mythological, we may suspect that the treatment was
thoroughly Italian. It was sufficiently different from any other
form of entertainment, and at the same time sufficiently
amusing, to maintain itself even on the imperial stage. Of its
history as a literary form subsequent to the days of Pomponius
and Novius we know little. (The dictator Sulla is said to have
written ' satyric comedies' in Latin ; these were scarcely
intended for the theatre.) But Mummius is said to have revived
the Atellana after a long period of neglect; we have a single
title (Riuinus) and three short fragments of his Atellanae.
Another writer, Aprissius (?), the form of whose very name is
doubtful, has left us a single fragment. Perhaps these writers
belong to the imperial period. The scantiness of the literary
remains suggests that after the time of Pomponius and Novius
the Atellane drama returned to a semi-improvised character.
We have a few fragments of Atellane cantica from the early
empire, one of them in Greek. The cantica would naturally
outlive the ordinary dialogue, which may have been little
140 THE ROMAN STAGE

more than comedians' patter, puns, riddles and so forth

{ilia obscura quae Atellani e more captant, as Quintilian puts it).
These fragments further illustrate the perennial interest of the
Atellana in politics, which led it even to indulge in veiled
allusions to the emperor's private life. The occasional use of
Greek may have been due to the desire to disguise, or give a
special flavour to, these dangerous allusions ; 1 it can hardly
have been frequently used, for Petronius' hero Trimalchio
speaks of the Atellana as something essentially Italian. In spite
of all the extraneous influences which acted on it, this native,
masked farce seems to have retained something of its primitive,
rustic, Italian character. As a form of literature it seems to
have been short-lived ; as a type of popular entertainment
it enjoyed great popularity during the early empire, and perhaps
formed part of the theatrical heritage which was handed on
by the ancient world to the Middle Ages.2
1
Of course Greek was widely understood in Rome—otherwise it would not
have been used on the comic stage.
2
Unfortunately the connexion with the Commedia dell' arte is incapable
of proof. Croce (Saggi sulla litteratura italiana del seicertto, pp. 219-220) refutes
Dieterich : £anni is probably not from Latin sanrtio but from Gianni, dim. of
Giovanni. See Lea, Ital. popular comedy, i, 54, 225-9.
 CHAPTER XVIII

THE MIME

F A L L the forms of entertainment which come within our

O survey, the mime was at once the most primitive and the
most permanent. In its earliest form it cannot be classed as
drama at all. A l l over the ancient world there were jugglers,
acrobats and public entertainers of all kinds, male and female,
who displayed their skill in the market-place, at festivals or
wherever and whenever they could secure patrons. A m o n g
these nameless mountebanks there were some with a special
gift for mimicry. T h e y could imitate with their voices the
neighing of horses, etc. (Plato Rep. 396 b) ; they had still
greater skill in gesture, an art which was carried to a high
pitch in the ancient world and which involved the use of every
l i m b — t h e mimi were akin to the acrobats—as well as of facial
expression. W e have been considering them so far as solo
performers, but assistants were required when the chief mime
wished to represent such everyday scenes as fruit-stealing or the
arrival of a quack doctor. These were stock subjects for the
deikelistai, little companies of players (probably masked) who
were popular among the Dorian Greeks. W e also hear of the
autokabdaloi (' improvisers '), whose name reminds us that the
performers of farce often depend a good deal on the inspiration
of the moment. T h e social status of such performers was low,
and their performance was of a simple kind. A rough platform
served to raise them above the heads of the crowd ; for scenery
a portable curtain was sufficient. T h e actors were concealed
behind the curtain till their turn came ; then, parting the
folds in the middle, they stepped into the public view. While
they performed a colleague might be collecting coins from the
spectators, as we see in a R o m a n wall-painting. 1 Xenophon
gives us a capital description of a performance by a boy and
girl, the property of a Syracusan dancing-master, who by
means of dance, gesture and words represented the love of
Dionysus and Ariadne. This performance was given in a
1 Reich, Mimtis, p. 540.
141
142 THE ROMAN STAGE

private house as an amusement for the guests ; one of those

present was Socrates. The girl was an acrobat as well as an
actress, and had previously impressed the company by her skill
in the sword-dance. Her dancing was, of course, accompanied
by music. Her social class is shown by the dancing-master's
admission that she was his concubine. Wide indeed was the
gulf between such performers and the actors who, in dignified
mask and costume, appeared in the theatre of Dionysus to
perform the tragedies of Aeschylus. Not that the mimes or
similar low-class performers were always maskless ; the
deikelistai seem to have owed their name to their use of the
deikelon, or mask ; the phlyakes, who burlesqued mythological
scenes for the entertainment of Greek towns in south Italy,
often wore masks of grotesque appearance, as well as tight-
fitting costumes, trousers and the phallus, a costume claimed to
resemble that of Aristophanic comedy. 1 It is not easy to dis-
tinguish sharply between the various types of popular enter-
tainers, but at least we may say that no respectable actor would
appear on the Greek stage without a mask. An element of
indecency clung to the mime from the beginning ; its aim was
mere amusement, the mimicus risus. The solo performer, who
could mould his features as he pleased and suit his voice to the
character he was representing, was a familiar figure at all
periods ; the company performances would naturally tend to
grow more elaborate with time. Hence we find the general
division of mimes into Paegnia, a generic term but on the whole
indicating performances of a slight, trivial character, and
Hypotheses, or ' plots perhaps taken over from drama proper
and rendered by the mimi in their own style. Another general
division was into the spoken and sung mime. 2 But to be too
precise about the classification of so primitive, popular and
widespread a form of entertainment would be vain labour.
Free from the shackles of social position, technical traditions
and a written text, the mimi were a law to themselves.
The popularity of farce among the Dorian Greeks, who
colonized much of Magna Graecia, meant that the mimi were
well known in south Italy from an early period. Athenaeus
tells us of a ' maskless ' actor named Cleon, who lived about
1
If we may assume that the vase-paintings do represent the performances
of the phlyakes. But see p. 15, note.
2
This distinction is perhaps unreal : see ch. xxvi.
 THE MIME 43
300 B.c. and was the best of the ' Italiot mimes Nor can
such a type of entertainment have long remained confined to
the Greek towns. Linguistic barriers were no hindrance to the
art of gesture and facial expression ; the simple requirements
of the troupes of strolling players—a rough platform on which
to set up their curtain and a few costumes—required no
elaborate outlay or preparations. T h e contacts of Rome with
Hellenism which arose out of the Pyrrhic W a r and the struggle
for Sicily meant that many Romans must have become
familiar with the mime at a time when the literary drama of
R o m e had not yet begun its career. T h e growth of a slave
population familiar with Greek provided the nucleus of an
audience for the ' ludus G r a e c u s T w o years after the intro-
duction of literary drama by Andronicus came the foundation
of the festival of Flora, a festival which either was from the
beginning or became an important occasion for the performance
of mimes ; the merry festival was riotously celebrated by the
common folk, and licence went so far as to sanction the appear-
ance of mime-actresses naked on the stage. W e are told that
in 212 B.C., at the first performance of the ludi Apollinares,
there was a dance by an elderly mime named Pompilius. Here,
then, in the third century, was an actor who had already
grown old in his profession. T h e influence of popular farce
on the development of literary Latin comedy may have been
considerable. Plautus' adaptations of Greek New Comedy
contain much jesting, buffoonery and horseplay of a kind
which he can scarcely have found in his literary originals, but
which would have been quite appropriate in the mime. His
very name, ' Flat-foot', may perhaps suggest that he had him-
self acted as a planipes or barefooted mime. His successor,
Caecilius, seems to have displayed in his style something of the
tone of mime. Volcacius perhaps refers to him as mimicus (the
reading is doubtful), and Gellius complains that he introduced
into his Greek originals nescio quae mimica—coarse jests, out
of keeping with the context, and added merely to raise a laugh.

By this time the strolling companies of mimes were becoming

familiar throughout the civilised world. A terracotta lamp
found in Athens shows us three mimologi who perform a
hypothesis called ' T h e M o t h e r - i n - l a w a title presumably
taken from literary comedy. A t its highest level, indeed, the
hypothesis approached drama proper, from which it differed
144 T H E
ROMAN STAGE

chiefly in its preoccupation with character-drawing and

situation rather than with plot. But there still remained the
solo performers, who prided themselves on their skill in
impersonation. The epitaph of one of them, Vitalis, boasts
of his skill in moulding his features and his voice to suit his
part whether male or female. Perhaps our earliest Latin
record of a mimus is the epitaph of Protogenes, slave of Clulius,
which was found near the town of Amiternum :
Protogenes Clouli suauei heicei situst mimus,
plourima que fecit populo soueis gaudia nuges.
Bücheler thinks the epitaph not much later than the time of
Ennius. Protogenes was evidently a slave of Greek origin who
amused the crowd with his ' merry trifles'; presumably he was
a solo performer of paegnia. From the age of Sulla, or perhaps
later, comes the epitaph found at Rome which commem-
orates Eucharis, slave and later freedwoman of Licinia.
Eucharis was a mime actress who had just been reaching
renown when she was cut down by death at the age of fourteen.
The epitaph tells us of the grief of the girl's father at the early
death of this little maiden, beloved by her parents and her
mistress, and so skilled in her profession that it seemed that the
Muses themselves had taught her. She had recently danced at
games given by the ' n o b l e s a n d she is described as the first
actress to have appeared on the ' Greek s t a g e I t seems
probable, however, that women took part in the mime from
the beginning, even in Italy, and from the second century
B.c. we have a farewell poem addressed by Antipater of Sidon
to the actress Antiodemis on her departure for Rome.
Genuine drama demanded a free atmosphere ; the mime
could and did maintain itself even in the courts of despots.
We find that the dictator Sulla was fond of the mime, and that
one of his friends was the lysiode 1 Metrobius. About this time
farce seems to have gained at the expense of higher drama ; the
Atellane performance now rises for a while above the literary
horizon, soon to be followed by the mime. On the whole,
however, the mime remained sub-literary. We can picture
these small companies of strolling players, men, women and
children, travelling from town to town like gypsies, setting up
their simple stage and curtain in some market-place and giving
1
T h e lysiode played female characters but wore male attire.
 THE MIME 145
their show. Chief among them was the leading actor or
actress (archimimus, archimima), to whom the rest were little more
than foils. Improvisation was probably the rule. The
arch-mime would perhaps begin by announcing the title, or
even summarizing the plot (habebant suum auctorem qui,
antequam mimum agerent, fabulam pronuntiaret, as Isidore
tells us) ; he was almost continuously on the stage, and he
kept the dialogue so much under his control that' the second
actor in the mime ' was a phrase denoting one who, as we should
say, ' played second fiddleThe actor secundarum partium
took such roles as the clown or fool; one of his methods of
raising laughter was probably to take the words of the archi-
mimus in too literal a sense—an old trick even in Plautus' day.
The distinctive costume was the ricinium, apparently a square
hood which could either be thrown back or drawn forward so as
to conceal the head, the centunculus, a patch-work jacket, tights
and the phallus ; the head was shaven, the feet were bare.
It is not to be supposed, however, that all the characters in the
performance were so attired. The elegant philanderer of
whom Ovid speaks must have been dressed as a man of fashion,
and a beautiful actress would appear in the most attractive
and expensive clothes. The size of companies was perhaps not
so large as it later became ; three mimologi are shown on the
Attic lamp (they presumably had a female colleague to act the
title-role of the mother-in-law), and Ovid speaks of a cast of
three—the peccant wife, the foolish old husband and the
foppish lover.
The plots were simple, the denouements often abrupt. A
stock theme was the appearance of some one in a novel situation
—a poor man suddenly become rich, for example (modo egens,
repente dims, as Cicero puts i t ) . Juvenal, when speaking of
foundlings adopted into great houses, is reminded of the mime.) 1
The riches might not last, and the ex-millionaire might have to
disguise himself in his cloak and hurry off the stage. Abrupt
endings were natural in these short, improvised pieces ; Cicero
pointedly contrasts the mime in this respect with the regular
1
We may remember how Augustus, on his death-bed, asked the spectators
whether he had played the ' mime of life ' in fitting fashion. Was he thinking
of his extraordinary elevation to supreme power, of the dominating yet lonely
part which he had played, of the fact that circumstances had forced him to adopt
a pose, or of the unsatisfactory, inartistic and even farcical form which real life often
takes—mimus, nonfabula, not a play but a mime ?
146 THE ROMAN STAGE

drama, telling us how, when the plot had reached a deadlock,

some character would make his escape, the scabillarii would
sound the signal for ' curtain', and the aulaeum would rise
from the floor to conceal the stage from view. Another form
of curtain which was perhaps more particularly associated
with the mime was the siparium, possibly a portable curtain
behind which the actors awaited their cue. Diomedes gives
as a possible (but surely unlikely) explanation of the term
planipes the fact (if it is a fact) that originally the actors
of mime stood not on the raised stage but on the level of the
orchestra, where they set up their outfit and performed their
piece. A corrupt passage of Festus seems to mean that the
mimes performed in the orchestra while the stage was being
set for a new play. Diomedes and Festus appear to be thinking
in terms of the Greek theatre, which had a large orchestra,
suitable for interludes. Apuleius describes how in the theatre
of Corinth, after a preliminary dance (presumably given in the
orchestra), the great aulaeum was raised, the siparia were folded
up, and a xjiost elaborate scene was disclosed. By Apuleius'
day it appears that the drop-curtain had come to be operated
as in our modern theatres (see Appendix (e)). In Cicero's
time the mimi had evidently performed on the stage, like all
other performers in a Roman theatre. Apparently the curtain,
whatever its form, was of more importance to the mime than
to other forms of drama.
Short, amusing, topical, utterly unrestrained by any
considerations of technique or decency, yet capable of adopting
on occasion the most sententious style, the mime came nearer
than any other form of drama to the real tastes of the Roman
populace. It would be difficult to describe in more detail so
formless or rather Protean a type of composition ; but some
further light may be derived from a study of the literary
fragments.

THE LITERARY MIME

The first Latin writer to give the mime written form was
Decimus Laberius (b. 106 B.c.), a Roman knight of high stand-
ing and character. Needless to say, he did not choose to act in
his own mimes ; his position in society was so established that
he was able to refuse Clodius' request for a mime with the taunt
that all the demagogue could do to punish him would be to
 THE MIME 147
send h i m for a short period of exile in D y r r h a c i u m (a reference
to Cicero's recent b a n i s h m e n t ) . Bitter i n d e e d was t h e blow to
his pride, w h e n , in his sixtieth year, h e was forced b y Caesar
to a p p e a r o n t h e stage as a competitor w i t h a y o u n g m i m e -
writer a n d actor, t h e ex-slave Publilius Syrus. W e still possess
t h e dignified a n d m a n l y p r o l o g u e w h i c h L a b e r i u s u t t e r e d on
this occasion, w i t h its significant w a r n i n g to t h e dictator
himself:
M a n y he needs must fear whom many fear.

C a e s a r a w a r d e d t h e prize to Syrus ; to L a b e r i u s h e gave half

a million sesterces a n d a gold ring, t h e r e b y signifying t h e res-
toration of equestrian r a n k w h i c h L a b e r i u s h a d forfeited by
a p p e a r i n g in a m i m e . Nevertheless his fellow-knights failed to
m a k e r o o m for h i m w h e n h e tried to resume his seat as a
spectator. Cicero called mockingly f r o m t h e seats of t h e
senators 4 1 would m a k e r o o m for you, were I n o t so short of
r o o m m y s e l f ' (an allusion to Caesar's p a c k i n g of t h e senate) ;
w h e r e u p o n L a b e r i u s replied w i t h considerable presence of
m i n d ' I t is surprising t h a t y o u a r e short of r o o m , for y o u usually
sit u p o n two seats at t h e s a m e time ! ' I n t h e prologue of his
next m i m e h e included a dignified allusion to t h e transience of
a d r a m a t i s t ' s p o p u l a r i t y . A y e a r or t w o later, in 43, h e died
at Puteoli.
W e h a v e forty-two titles a n d a b o u t one h u n d r e d a n d forty
lines or p a r t s of lines. T h e contrast between t h e long extract
f r o m a single prologue a n d t h e m e a g r e r e m n a n t s of t h e mimes
themselves has b e e n t h o u g h t to indicate t h a t only t h e prologue
of a m i m e was fully written out, while t h e rest of t h e piece was
to some extent improvised. T h e titles are p a r t l y Greek (Caco-
mnemon, Necyomantia, Ephebus) b u t mostly in L a t i n ; some a r e
reminiscent of t h e palliata (Aulularia), r a t h e r m o r e of t h e t o g a t a
{Aquae Caldae, Compitalia, Fullo, Saturnalia, Staminariae). Anna
Perenna m a y h a v e b e e n a mythological burlesque, akin to t h e
Rhintkonica. T h e m i m e , considered as a type of d r a m a , is
so formless t h a t it is h a r d to say w h a t a typical mime-title would
b e ; as we hear, however, of dogs w h i c h p e r f o r m e d in mimes, it
is interesting to find titles like Catularius a n d Scylax, while Cen-
tonarius r e m i n d s us of t h e mime-clown's traditional motley, t h e
centunculus. Paupertas m a y suggest some sort of d e b a t e between
t h e personified figures of Poverty a n d Riches ; w e r e m e m b e r
148 THE ROMAN STAGE

Novius' title Mortis et Vitae Iudicium. Gemelli, ' T h e Twins

suggests a favourite N e w Comedy subject, mistaken identity ;
but as the mimes are supposed to have been usually maskless, it
is hard to see how the necessary resemblance was obtained
between two actors ; perhaps ' T h e Twins ' are Castor and
Pollux, and the play a burlesque. T h e titles ' R a m ' Bull',
' Twins ' Crab ' Virgin ', suggest that Laberius wrote a
whole set of pieces named after the signs of the zodiac—one
would infer that such pieces must have been short.
There is evidence in the fragments that these mimes were
indecent in subject and language and that they dealt with the
traditional mime-theme, adultery (for which Laberius seems
to have invented a set of technical terms, moechimonium,
adulterio, adulteritas) and unnatural vice. A line in the ' Needle-
woman ' (' our mistress loves her stepson to distraction')
suggests some such plot as that of the Oxyrhynchus Mime,
including perhaps an attempt to get rid of the unwanted
husband by poison ; we have three references to drowsiness or
sleeping-draughts (3, 10, 86) which perhaps point to the
felonious administration of drugs. T h e guilty stepmother may
have been detected in time ; perhaps she is betrayed by the
stepson, who remarks ' I see your wife, m y stepmother, being
stoned by the people '. Another mime-theme, the utterance of
an outlandish tongue or gibberish (such as the pseudo-Indian
of the Oxyrhynchus Farce), may appear in the remark ' I
didn't realize that you spoke M o o r i s h ' . T h e typical mime-
theme of stupidity may have been prominent in ' T h e M a n with
a bad Memory with its f r a g m e n t ' This is that dolt who came
to me from Africa two months ago T h e traditional fondness
of the mime for topical allusions of the most biting kind is
illustrated by a reference to some plunderer of his province
(such as Verres), an allusion to Caesar's creation of Market-
overseers and his supposed plan to legalize bigamy, and a line
in the famous prologue ' ' R o m a n citizens ! we are losing our
liberties ! ' T h e homely style of some passages is illustrated in
the regrets expressed by some one for having come out of his
cottage or shop (taberna) in the rain. T h e fuller, treading the
clothes in the tank, is compared to a Balearic crane. Philo-
sophers are always fair game for the comedian ; we have
scoffing references to the Cynics and Pythagoreans. Did the
Ephebus contain a scene in heaven, where the gods discuss the
 THE MIME I49
destiny of R o m e ? Apparently Jupiter is asked to curb the
lust and licence of the gens togata ; perhaps it is Juno who
declares with bitterness that the gods have bestowed on the
gens togata their wide empire. A striking passage of the Restio,
perhaps from the prologue, sets forth the feelings of a rich and
miserly father who sees his riches being squandered by his son.
' T h e philosopher Democritus of Abdera set a shield to catch
the sun's rays in order to dazzle and blind his own eyes and
prevent himself from seeing the good fortune of wicked citizens.
So I too wish to make the gleam of money blind the end of my
days, in order not to see m y worthless son in luxury.'
W e still find the Latin comic genius pursuing the pun in
such lines as :
delenimenta
ad amorem deliramenta, ueneficia autem beneficia
sunt.
There is a Plautine ring in
non mammosa, non annosa, non bibosa, non procax.
W e are told, indeed, that Laberius took liberties with the
language, used forms the Latinity of which was doubtful and
made slang fashionable (e.g. manuarius for ' thief').
Horace admits Laberius' farcical power, but says that this
is not enough in itself to entitle his pieces to admiration as
' fine poems '. T h e fragments are all written in the usual
metres of stage declamation ; of ' l y r i c a l ' metres (so-called)
there is only the slightest trace.
T h e successful rival of Laberius, Publilius Syrus, was, as
his name implies, a slave from Syria, gifted with the wit for
which Syrians were noted. His qualities of mind as well as his
grace of body won for him freedom, education and patronage.
He made his fame as a writer and actor of mimes in the pro-
vincial towns of Italy ; then, coming to Rome for the ludi
Caesaris of 46 or 45, he challenged his rivals to extempore
performance, and was victorious over them all, including
Laberius. T h e high place in Caesar's favour of the ex-slave
mime-writer once more illustrates the liking of despots for the
entertainments of the common people. As we should expect
in the case of pieces which seem to have been largely extempore
very little is left of Syrus' mimes. There are two doubtful
titles, two not very enlightening fragments and a passage
11
150 THE ROMAN STAGE

eighteen lines long denouncing the extravagance of the

Romans, which is quoted by Petronius' hero Trimalchio
as the work of Syrus, but is usually thought now to be an imi-
tation of Syrus' style composed by Petronius himself; most
surprising of all, we have a vast collection of moral sayings and
epigrams attributed to Syrus ; many of them must be genuine.
Their high ethical standard surprised Seneca, who observes
that they are worthy of tragedy rather than mime. Evidently
the Romans liked a graver element to be included even in the
outrageous foolery of the mime ; such lines as
О uitam misero longam, felici breuem !
show that Syrus had more in him than mimicae ineptiae and
jests aimed at the gallery (nerba ad summam caueam spectantia),
and we are told that these very epigrams were capable of
bringing down the house (поп uides quemadmodum theatra consonent
quotiens aliqua dicta sunt quae publice adgnoscimus et consensu uera
esse testamur ?).
With Laberius and Syrus we have come down to the very
end of the Republic. All the forms of drama known to the
Republican stage have now been described. In accordance
with the general plan of this work, I will now try to give
some account of how these plays were produced and staged,
relying for information where possible on the text of the plays
themselves.
 CHAPTER XIX

THE LATIN PROLOGUES AND THEIR VALUE AS

EVIDENCE FOR THEATRICAL CONDITIONS

A L L T H E plays of Terence and most of the plays of

•IJLPlautus are introduced by prologues. The prologues of
Terence are the author's replies to the attacks of his enemies ;
clearly, therefore, they are written by Terence himself, and
they reflect the conditions of his own day. T h e origin of the
Plautine prologues is not so clear. T h e prologue to the Casina
specifically refers to a revival performance of the play some
time after the death of Plautus, but still within the lifetime of
some who had seen the play when it was first produced.
Evidently this prologue is, at least in part, post-Plautine, and
equally evidently the post-Plautine part at least is not translated
from a Greek original but is of Latin origin. T h e two-line
prologue to the Pseudolus (' Y o u had better get up and stretch
your legs ; a long Plautine play is coming on the stage ') also
seems to be post-Plautine. T h e speaker of the prologue to the
Menaechmi remarks ' I bring you Plautus—on my tongue,
not in my hand Would Plautus have referred to himself
quite in these terms ? Hardly any Plautine prologues can be
positively said to contain a contemporary reference which must
have been written by Plautus himself. (But see Cist. 197-202).
I think, however, that even where we may suspect the presence
of post-Plautine insertions, such insertions should still be
ascribed, like the Casina prologue, to a period not long after
the death of Plautus. They are presumably the work of
producers—for who but a producer would have had either the
opportunity or the motive for tampering with the text ? They
belong therefore to a period when the theatre was active and
before the text had been established by editors for the benefit
of a reading public. They are thus good evidence for the
theatre of the second century. O n the other hand some parts
of some prologues may conceivably be translated from the
Greek—presumably by Plautus himself. But references to
152 THE ROMAN STAGE

theatrical conditions can scarcely have been taken over from

the Greek original. Descriptions of the actual conditions in the
Greek theatre would not have been appropriate to the Roman
theatre; and no sane dramatist would have taken over
descriptive passages of such a kind (supposing he found
them in his original) without modifying and adapting them
to the conditions of his own day.
We have very little direct evidence concerning the pro-
logues of Greek New Comedy. We know that gods sometimes
appeared to give the audience necessary information about the
dramatic situation. The Perikeiromene of Menander contained
a prologue of this kind, spoken by the goddess Misapprehension,
of which part is preserved. The tone of this unique fragment
is matter of f a c t ; the goddess appears to take it for granted
that the spectators are interested in the play, and she makes no
attempt to enliven her narrative with jokes, appeals, promises
or topical allusions of any kind. This is a ' deferred ' prologue,
spoken, that is, not at the beginning of the play but after some
scenes have already been performed. It is quite as sober in
tone as the explanatory prologues of Euripides. We have also
some slight references in Middle and New Comedy to literary
topics—and one or two of these seem to occur in prologues ;
indeed, they deal with the use of the explanatory prologue and
kindred matters. The prologues of Plautus and Terence are
quite different from what we know of the Greek prologues.
The difference is that between information and propaganda.
The Latin writers are not able to assume that their audience is
interested in the play. Their prologues are designed above all
things to secure a hearing for the play.
The Plautine prologues have a genial air. The speaker
takes his audience as he finds them. His object is to give them
what they like. The prologues of Terence are earnest, dis-
trustful, plaintive. Terence wishes to give the audience what
they ought to like. It is hard to say which writer gives us the
more dismal impression of the Roman public.
The purely informative prologue would not have done, it
seems, for the Roman audience. The dramatist had to win
their attention at the very beginning. But how ? That was
the problem. An appeal for silence, a joke, a promise that the
play will be amusing, a topical allusion might be successful.
Alternatively it might be better to have no prologue but begin
 THE LATIN PROLOGUES T
53
the play right away with a lively or impressive scene. A
possible compromise is to give the needed information in a
' deferred' prologue, such as that spoken by the goddess
Auxilium in the Cistellaria, which is factual enough to be
considered a direct translation from the Greek. W h e n Plautus
does give information at the beginning of the play, we notice
that he sometimes promises to be b r i e f :

nunc argumentum accipite atque animum aduortite ;

quam potero in uerba conferam paucissuma (Men. 5, 6).

So too As. 8, Trin. 4-5. Evidently the dramatist was in a

dilemma. If he gave too much information the audience
might get bored. If he gave too little they might get confused.
He may have felt that it was only by being discursive, by
interjecting comedians' patter and by tricks of all kinds, that
he could hold the attention of his audience while imparting
the necessary information. H e had to consider tactics rather
than strategy.
For the Romans the theatre was essentially a place of
entertainment ; what they expected of comedy was that it
should amuse them. T h e theatre had to compete with other
forms of entertainment such as rope-dancing. Failure to make
an immediate impression might result in empty benches, as
the two prologues to Terence's Hecyra show. Faced by such
a possibility, the dramatist is compelled to cry his wares :

inest lepos ludusque in hac comoedia ;

ridicula res est !

(As. 13-4). Such promises were intended to secure the

necessary silence and attention :

ita huic facietis fabulae silentium,

(Amph. 15 ; so too Capt. 54, Poen. 3).

T h e Plautine prologues were meant to amuse, and should
not always be taken literally. For instance we can hardly
assume that the comic injunctions issued to married ladies,
nurses and courtesans were intended to be obeyed ; but they
do at least imply that woman were present among the audience.
T h e prologues of Terence were not meant to amuse, but
sometimes at least they were meant to mislead.
154 THE ROMAN STAGE

DEVELOPMENT OF THE DRAMATIC FESTIVALS IN ROME

All dramatic performances in Rome were given as part of

some festival. From the beginning the festivals were associated
with the State religion, and they always retained an official
air. Apart from festivals intended for a particular occasion
(euch as funeral games) all ludi were under the charge of
magistrates ; expenses were defrayed by the State, the pre-
siding magistrates, or other wealthy individuals, and entry was
free for the whole population. The ludi scenici, or public
festivals which included dramatic performances, were :

Made Magistrates: Duration

Title In honour of Instituted
scenic Date under Empire

ι Romani Jupiter ? 6th cent. 364 B.C. Cur. aediles 4-19 Sept.
15 Sept.
2 Florales Flora 238 (made ? Cur. aediles 28 April-
annual 173) 28 April 3 May
3 Plebei Jupiter ? 220 By я 00 Pleb. aediles 4-17 Nov.
15 Nov.
4 Apollinares Apollo 212 By 169 Praet. urb. 6-13 July
13 July
5 Megalensia Gt. Mother 204 By 194 Cur. aediles 4-10 April
4 April

The festivals included other amusements—circus races,

gladiatorial fights, beast-fights, wrestling, boxing, rope-dancing
etc ; sometimes these rival attractions were taking place
simultaneously with the dramatic performance. The place of
performance varied ; races were naturally given in the circus ;
in the days before Rome had a permanent theatre, plays would
be shown in an improvised theatre, erected, perhaps, near the
temple of the god in whose honour the festival was given. There
were also special games, such as the ludi uotiui given at the
inauguration of some public building, games in celebration of
a triumph or other happy event, and funeral games in honour
of some deceased nobleman. Like the regular ludi, these special
ludi continued to grow more lengthy and magnificent; by the
end of the Republic and still more under the Empire there
were frequent occasions during the summer months on which
the public could see dramatic performances. In the winter,
from late November to March, open air performances were
 THE LATIN PROLOGUES 155
necessarily suspended ; Juvenal tells us that then the devotees
of the theatre had to be content with private theatricals.
While all ludi aimed at winning public favour, and those
who gave the games taxed their resources more and more
extravagantly to secure this result, State supervision kept the
freedom of the performers within certain bounds. T h e fact
that so large a part of the electorate were assembled had a
political significance. T h e y were worth entertaining because
they had votes ; but any attempt to use the theatre for a
political demonstration was watched with a jealous eye. T h e
comparative decency of Latin comedy (as distinct from farce
and mime) may perhaps be attributed to the official character
of the games ; but at the merry Florales, as we have seen,
licence went so far as to allow the mime-actresses to appear
naked.
This is a general account of the festivals of the Republic.
In the time of Plautus things were on a simpler scale. Kurrel-
meyer calculates that in 240 B.C. there was probably only one
day in the year for plays ; by 214 this had increased to four
days of the Ludi Romani. From 200 we can add at least one
day of the Ludi Plebei, and from 191 at least one of the Ludi
Megalenses (didascaliae to Stichus and Pseudolus). ' Before
200 B.C. therefore there is no definite evidence that there were
more than four days on which the public games included ludi
scaenici, though by 190 B.C. the number may have been
increased to seven or eight.' (Kurrelmeyer, Economy of Actors
in Plautus, p. 7). More than one play may have been produced
in the same day {Pseud, prol.), but still the opportunities for
dramatists and actors were very limited—and it is not easy to
see how the actors earned their living during the 51 weeks
when they were not acting. Perhaps they took part in Atellane
farces, mimes and the like. Plautus himself may have done
so, if the explanation of his name as indicating that he was
plants pedibus is to be connected with planipes, ' flatfoot' or
' b a r e f o o t t h e mime. Maccus, the author of the Mercator
and Asinaria, according to the prologues, bears the name of
one of the stock Atellane roles. T h e actors—and playwrights—
of early Rome seem to have earned a meagre livelihood by
entertaining the public and serving the great as opportunity
offered—by dancing, singing, clowning, by writing lampoons
or panegyrics—as their talents and their training allowed.
 CHAPTER XX

THE ORGANIZATION OF THE ROMAN THEATRE

I N T H E interval between the writing of a play and its

appearance on the stage there is much to be done. O n the
organization of the early R o m a n theatre we have little in-
formation ; Plautus is almost silent, Terence is evasive, and the
later writers may not have known the facts. Between the
impecunious dramatist who wanted to sell his play, the general
public who wanted to be amused at some one else's expense,
and the ambitious magistrate who was willing to supplement
the State grant (lucar) out of his own resources, an essential
link was the producer and actor-manger. Such a man was
Ambivius Turpio, the producer of Terence's plays. H e bought
the plays at his own expense (pretio emptas meo, Нес. 57),
though hoping no doubt to recoup himself with the money
paid him by the magistrates (Eun. 20). In a prologue written
for him by Terence {Нес. 14 ff), he claims to have encouraged
Caecilius by bringing out his plays in spite of the unfavourable
attitude of the public. H e must have been the elderly actor
who spoke the prologue to the Η. T. He has a tone of dignity
and authority as a man who is conscious of having befriended
struggling dramatists, one who has not aimed at mere profit
but has even incurred financial risks in fostering talent for the
benefit of the public.
W e do not know what sort of bargains the producer struck
with the dramatist on the one hand and the magistrates on the
other. T h e common-sense view seems to be that the dramatist
sold his manuscript outright to the producer, who acquired
thereby the right to perform the play as often as he liked and
found practicable. If we are to take Sacch. 214-5 at its face
value, Plautus must have quarrelled with Pellio after selling the
Epidicus. ' T h o u g h I am as fond of the Epidicus as of my
own self, I cannot bear seeing it if Pellio is taking the (leading)
part.' This suggests that revival performances of Plautus'
plays were given in his own lifetime, but that he had no control
156
 THE ORGANIZATION OF THE ROMAN THEATRE I57

over them. Terence evidently kept on better terms with

Ambivius Turpio. After each of the two failures of the Hecyra
the manuscript was returned to the author, who added on
each occasion a prologue. T h e first of these prologues, after
mentioning how the opening performance was ruined by the
counter-attraction of the rope-dancer, adds ' Now the play is
as good as new, and the dramatist did not wish to produce it
again in order to sell it again '. Like all of Terence's personal
utterances, this cryptic remark is presumably a reply to
an accusation. I imagine that rumour had said ' He is
trying to revive his old play merely to pocket a second fee
Such a charge would indeed carry with it the absurd impli-
cation that failure might be more profitable than success.
Terence briefly answers ' T h a t is not my r e a s o n T h e true
reason was that neither Terence nor Ambivius Turpio was
willing to sit down under defeat ; moreover Terence was
proud of the Hecyra, and meant to persist until he secured it
a fair hearing.
T h e manuscript of a play presumably remained the property
of the producer until he sold or bequeathed it to some other
producer. It owed its preservation to its commercial value.
Some plays, it is true, were preserved which had little value to
a producer—for example the archaic plays of Andronicus.
T h e prologues of Terence had no commercial value, but
they were preserved as part of the precious manuscript.
Plautus does not trouble to mention that his plays are
new ; Terence takes care to point out that his plays have not
been seen before. It may be that by Terence's day producers
had grown over-fond of reviving the plays which they already
possessed, and for the use of which they had no need to fee an
author. T h e post-Plautine prologue to the Casina actually
claims that plays, like wine, are all the better for being o l d —
for the new plays are worthless. This is making a virtue of
necessity, or at least economy. A new play meant a fee to the
writer. Suetonius tells us that Terence's Eunuchus won for
its author the unprecedented sum of 6000 sesterces, and that
this payment was recorded on the title page of the manuscript.
T h e other fees received by Terence must therefore have been
smaller. In six years he produced six plays ; his income must
have been meagre compared with the half-million sesterces
earned annually by the actor Roscius in the next century.
i58 THE ROMAN STAGE

Terence says that failure would mean starvation for him.

T h o u g h drama offered almost the only means of earn-
ing a living b y one's pen, it can hardly have yielded a
fortune.
Ambivius T u r p i o was responsible not only for the purchase
of the play from the author b u t also for its performance ;
he acted with the other members of his company. These
functions m a y sometimes have been divided ; the prologue to
the Asinaria (line 3) speaks of the grex or troupe of actors, the
domini or ' masters' of the troupe, and the conductores or persons
w h o ' u n d e r t o o k ' the contract to produce the play. We
hear occasionally of the choragus or property-manager,
who hired the costumes to the actors ; in the Curculio Plautus
brings him on the stage to express his uneasiness about the
prospect of recovering his property.
T h e actors are usually called histriones ; another word is
cantores, as all actors had to be able to declaim. T h e y were
organized into a troupe (grex, caterua). It is commonly said
that all actors were slaves, or at best freedmen. This is true
of professional actors in imperial times, but w e have some
evidence that the acting profession had not always been so
despised. L i v y speaks of a time when the y o u n g citizens
of R o m e had thought it no shame to act in public, and of
how A n d r o n i c u s , ' like all (the dramatists) of his d a y ' had acted
in his o w n plays. W h a t truth there is in these remarks w e do
not know. It seems clear that Ambivius T u r p i o was a m a n of
some consequence, while in Cicero's d a y the comic actor
Roscius was the friend of Cicero ; Sulla m a d e him a knight,
and Cicero thought him worthy to be a senator. T h a t Roscius
began as a slave is a probably wrong inference from Pliny
N . H . vii. 128. H e made a large fortune b y his acting, he was a
teacher of his art and wrote a book about it. His contem-
porary, the tragic actor Aesopus, left a fortune of twenty
million sesterces. Evidently the profession of acting was not
in itself a b a r to social a d v a n c e m e n t ; a sharp contrast is
presented b y the mime. Laberius, as w e have said, felt
disgraced w h e n compelled by Julius Caesar to perform in one
of his own mimes (domum reuortor mimus) : by his appearance
on the stage he forfeited his equestrian status, which was
restored to him after the performance b y the Dictator.
Nevertheless the social status of the average actor in tragedy
 THE ORGANIZATION OF THE ROMAN THEATRE 159

or comedy was probably not high. T h e very fact that actors

were organized in a troupe under a dominus suggests something
not far removed from slave status. O f course they enjoyed a
far pleasanter lot than the degraded slaves of an Italian
plantation ; indeed when we consider on how few occasions
in the year plays could be performed, we may feel that actors
needed some master or patron to maintain them, or that they
made a living as public entertainers, dancers, singers, etc. 1
T h e epilogue to the Cistellaria tells the audience that the actors
are retiring to take off their costumes ; then those who have
acted well will get a drink, those who have acted badly
will be thrashed. This is perhaps a joke, but we cannot
easily imagine such a remark being made in the theatre
of Dionysus. T h e increasing vulgarity of the R o m a n
audience must have tended to degrade the status of those
who performed to amuse that audience. Polybius tells us
that in 167 B.C. the eminent Greek musicians assembled on
the stage found that the readiest way to please the crowd
was to indulge in a mimic battle.
It is often said that in Rome companies were larger than
in Greece. There seems to be no evidence for this except the
remarks of late grammarians, who may have been thinking
of the crowds brought on the stage in shows of the late
Republic and the Empire. T h e R o m a n stage-tradition was
derived from the Greek ; in Greece the number of trained
actors was kept small. T h e Greek use of masks and the limi-
tation of the number of characters who may speak in any
one scene made it possible for a company of three or four
to divide between them a considerable number of parts,
thereby securing economy and efficiency. How far the Roman
actors followed the Greek custom of each taking more than one
part we cannot t e l l ; 2 all we can say is that the Latin plays
are so constructed that a company of five trained actors
(aided on occasion by mutes) could perform almost any
scene in Latin comedy.
Competition between different companies was keen ;
every member in each team was therefore on his mettle. W e
have references to prizes awarded to companies or individual
1 Cf. Rudens 535.
2 There was certainly some doubling on the Roman stage ; cf. ibo, alius nunc
fieri uolo. (Poen. 126).
ι6ο THE ROMAN STAGE

actors, and indeed to unfair canvassing by actors for such

prizes. We hear of how they stationed their supporters
throughout the theatre with instructions to applaud at the
right moment, of solicitation in person or by letter, and even
of attempts to bribe the magistrates.
Grassus Diues, curule aedile in 2 1 1 B.C., is said to have
been the first to present gold and silver crowns. That good
acting might give a bad play undeserved success is suggested
by Terence (Phor. 9, 10) ; Plautus remarks that bad acting
may spoil a good play (Bacch. 214-5).
A musical element was present in all Greek and Roman
drama. Greek New Comedy had reduced this element,
but not removed it (it occurs in Menander's dialogue, as
well as in the choral dances), and on the Latin stage it was
increased to a degree which we may find hard to realize. The
greater part of most Latin plays was declaimed to musical
accompaniment. The musician had two pipes, each about
twenty inches long, bound to his mouth by a bandage round
his head, so as to leave his hands free to work the stops. The
pipes were called pares if they were of the same length and
impares if one was longer than the other. With one, it is
supposed, the musician played the air, with the other an
accompaniment. 1 Such pipes have been found at Pompeii,
and we have illustrations showing the musician standing
among the actors on the stage and accompanying their words.
Of all the wealth of metres which we find in the plays of Plautus
only one—the six-foot iambic line—is employed for plain
speech without musical accompaniment.2 And so in the
Stichus of Plautus, when the slaves are holding a drinking-
party and decide to give the musician a drink too, the metre
changes to six-foot iambics for seven lines (762-8) while the
tibicen is drinking. A prefatory note in the oldest manuscript
tells us that the music for the Stichus w a s ' made ' by Marcipor,
slave of Oppius. Perhaps he was both composer and per-
former. Flaccus, slave of Claudius, ' made' the music for
Terence's plays. The tibicen must have had considerable
powers of memory; each leading' part,' it would appear, had its
1
This view is rejected by the Oxford Classical Dictionary (s.v. Musk, 9, ii.) :
' if there was some device to allow the pipes to speak separately, the two instru-
ments may have combined to produce a more extended scale."
2
It should perhaps be added that the Romans did not distinguish as sharply
as we do between speech and song ; see chapter xxvi.
 THE ORGANIZATION OF THE ROMAN THEATRE l6l

traditional music, and as soon as the tibicen struck up the

overture, connoisseurs knew what character was to appear. 1
It is commonly said that on the Roman stage trained
vocalists were employed to sing the cantica or songs, while
the actor accompanied the words in dumb show. This
seems to have arisen from confusion in Livy's account between
the pantomime and drama proper. To try to divide the
spoken parts from the sung parts in this way would make
Latin comedy, as we have it, unstageable. We must suppose
either that the actors were unable to ' sing' (in which case
they must have remained silent throughout the greater part of
the play) or that they were able to ' sing ' (in which case they
had no need of the vocalists). To suppose that the vocalists
sang from behind the scenes is inconsistent with the account
in Livy, our sole authority on this matter.
We hear incidentally in the prologues of various servants
of the theatre. The crier (praeco) may sometimes have been
called on to ask for silence at the commencement of a per-
formance [Poen. n ) . The dissignator (Poen. 19-20) showed
people to their seats. The conquistores (Amph. 65) saw that
order was kept—but perhaps this was not a regular office.
There must have been stage carpenters, scene shifters and
the like, who could probably don mask and costume and take
a ' mute' part at a pinch. In the course of time men were
needed to work the curtain ; to these the signal was given
when time-keepers known as scabillarii2 tapped their specially
designed wooden shoes on the floor.
A more exalted official of the late Republic was the censor
of plays. In earlier times it may be that the magistrates who
sponsored the performance were left to satisfy themselves
that the play to be shown contained nothing offensive. The
prologue to the Eunuchus tells us of a preliminary performance
given in the presence of the magistrates and of certain other
interested persons such as Terence's rival, Luscius Lanuvinus.
While there is nothing in this passage to indicate that a
preliminary performance was in any way an unusual expedient,
we have no other reference to such an arrangement. Perhaps
the continual criticism of Terence's methods had induced the
1
Cie. Acad. II. ao, primo inflatu tibicinis Antiopam esse auint, aut Andro-
macham, cum id nos ne suspicemur quidem. But is the reference to drama, or to
some early form of pantomime ?
1
See the illustration in Daremberg et Saglio.
l62 THE ROMAN STAGE

magistrates on this occasion to demand an opportunity to see

the play before it was shown to the public.
It is not easy to say whether any notice of a forthcoming
performance was given to the public. In later times notices
were displayed giving an account of the various attractions
to be provided at the games. Some of Plautus' prologues
seem to take it for granted that the public do not know the
name of the play they are going to see or of its Greek author
or Latin adaptor. Terence, on the contrary, does not find it
necessary to give his name in his prologues. In the prologue
to the H. Т., after announcing the tide, he says that he would
add his own name and that of the Greek writer but for the
fact that they are already known to most of the audience.
Indeed it would appear that there had been only too much
public discussion before the appearance of each of Terence's
plays. Whatever previous notice had been given, it could not
be taken for granted that every member of the audience had
seen it. I have already referred to Cicero's remark that the
opening bars of the musical overture indicated to connoisseurs
the name of the character represented. This implies that neither
connoisseurs nor the general public had previously known what
character was to appear. What is certain is that in Imperial times
prospective theatregoers in Pompeii were informed by notices
painted in public places that if they went to the theatre on a
certain date they would find awnings 1 to shield them from the
sun and showers of perfumed water to temper the sultry air.
Similarly, while it seems plain from the prologues of Plautus
that there were no programmes to help the spectators to
follow the play, Ovid speaks of programmes as being available
at circus races and gladiatorial shows, and of how skilful
use of the programme might enable a young gentleman to
ingratiate himself with the girl sitting in the next seat. But
we have no evidence that programmes were ever used in the
theatre. In Imperial times there were theatre-tickets in-
dicating where the holder was to s i t ; 2 probably these were
held by certain privileged spectators. Reservation of seats
for senators and (later) equites had been a practice from
Republican times onwards, and seems to have caused much
ill-feeling among the under-privileged.
1
Lucretius (iv. 75-83) mentions the magnificent coloured itla spread over
the whole theatre, which, by excluding the glare of the sun, added to the effect
2
of the decorated stage. See page 167 and Bieber, Theater, figs. 455-7.
 C H A P T E R XXI

SEATS IN T H E ROMAN THEATRE

E T U S now imagine ourselves present at a performance

L л in the time of Plautus or Terence, and look about us in the
theatre. T h e theatre consisted of two main parts, the scaena and
the cauea, the scaena for the actors, the cauea for the spectators.
L e t us take the cauea first and ask ourselves w h a t sort of accom-
modation was provided for the spectators, and w h a t can be
known about the spectators themselves.
A difficulty meets us at the outset. W e r e seats provided for
the spectators ? Ritschl denied this on the authority of a
passage in Tacitus (Ann. xiv. 20), and consequently regarded
all references in the prologues to seats as proof that these
prologues could not have been written before the year 145 B.C.
Ritschl's view is now generally abandoned. (See A p p e n d i x (a).)
W e m a y consider at this point a number of references in
L i v y to seats in the first half of the second century B.C. In
194 special seats were, according to L i v y , assigned to the
senators. In 179 the consul Lepidus ordered the construction
of a ' theatre and stage ' near the temple of Apollo (theatrum et
proscenium ad Apollinis) ; here the word theatrum denotes,
as in Greek, the a u d i t o r i u m — w h a t the Romans called the
cauea. I n 174 the censors undertook the construction of a stage
w h i c h was to be of use to magistrates giving shows in the future
(scaenam aedilibus praetoribusque praebendam), but this
again seems to have been only a temporary structure. I n 155
a stone theatre was planned and its building begun near the
Palatine, but Puritanical opposition, led b y the consul Publius
Cornelius Nasica, not only secured the destruction of this
theatre b u t prevented for some time the customary erection of
tiers of seats for the spectators, w h o were therefore forced to
stand (Livy Epit. xlix). N o permanent stone theatre was
erected until the building of the theatre of P o m p e y in 55 B.C.
In the light of these records w e can perhaps understand w h y
Tacitus makes the diehards of Nero's day claim that in the good
163
164 THE ROMAN STAGE

old days, before the building of Pompey's theatre, both the stage
and the seating had been mere temporary structures, and that
at an earlier period still the spectators had been compelled to
remain standing ' lest, if they were allowed to sit, they should
spend whole days idling in the theatre
T h e R o m a n theatre had from an early period drawn upon
itself the suspicions of respectable citizens. W e can see an
attempt to placate this Puritanical opposition in the prologue
and epilogue to the Captiui. In the eyes of such opponents of
the theatre the occasional erection of a temporary structure,
to be demolished after the games were over, was one thing :
the building of a permanent stone theatre quite another. T h e
project of building the Theatre R o y a l in Bristol in the year 1764
was opposed by the Quakers and Methodists, who feared that
it would ' diffuse an habit of idleness, indolence and debauchery
throughout this once industrious c i t y B u t there can be no
doubt that seating, of however limited a character, was
provided for the public of Plautus and Terence.

NOTE ON CURIO'S REVOLVING THEATRES

T h e elder Pliny (N.H. 36, §§ 116-120) gives a circumstantial

account of the two wooden theatres built back to back by C . Curio
in 50 B.C., which at a given moment revolved on their.axes (while
the spectators still occupied their seats) so as to meet along the
diameter in the form of an amphitheatre. This account, rejected
by P. W . (s.v. amphitheatrun), seems to me to involve a geometrical
impossibility, as well as engineering problems which would baffle
the twentieth century. Unfortunately its rejection casts some doubt
on Pliny's previous description (§§ 114-5) of the theatre of Scaurus,
with its three-storied scene-buiiding and its 80,000 seats. As for
Pliny's reference (35, § 23) to the scena erected by Claudius Pulcher
(? the aedile of 99 B . C . ) , SO realistically painted in perspective that
ravens thought the painted tiles were real, its chief value is to confirm
our other evidence that by ' scene-painting ' was meant merely the
decoration of the scene-building. Cf. Val. Max. 2, 4, 6, and see
below, p. 267.
 CHAPTER XXII

THE SPECTATORS

A L L C L A S S E S went to the theatre. The show was given

j f j L b y the magistrates. Livy speaks of special places being
reserved after 194 B.C. for the senators. The prologue to the
Poenulus contains injunctions to married ladies (matronae),
nurses with their infant charges, prostitutes, slaves (who are
forbidden to occupy seats), lackeys (pedisequi), attendants on
the magistrate (lictores). That women were present is stated
by Terence also [Нес. 35). 1 Admission to the theatre was free ;
the theatre tickets of Imperial times (which indicated where the
holders were to sit) were probably as yet unknown. A tumul-
tuous crowd of every age and condition and of both sexes
poured into the theatre in search of fun and excitement,
shouting to each other, laughing, quarrelling, fighting for
seats. T h e State, which kept so jealous an ear for what was
said on the stage, seems to have made little or no attempt to
control the behaviour of the audience. The ushers (dissigna-
tores) may have done something to maintain order ; but the
playwright and the actors knew that it rested on their joint
efforts to secure a hearing for the play, for no one else would
come to their assistance.
In the modern theatre we pay for admission, and have a
natural inducement to get our money's worth. We enter a
building, and forget the outside world. Artificial lighting
concentrates our eyes and our thoughts on the stage. It was
quite otherwise in the early Roman theatre. The spectators
might become aware that there were rival attractions close at
hand. The first performance of the Hecyra failed because the
public were more interested in a rope-dancer and a pair of
boxers. The second, after a promising start, was broken up
by a rumour that a gladiatorial display was about to take
place.
1 Vitruvius (V. iii. i) tells us that ' at the play citizens with their wives and

children remain seated in their enjoyment."

Ifi
12 5
 THE ROMAN STAGE

Apart from the reservation of special places for the senators

after 194 B.C., the rüle seems to have been first come, first
served. T h e front seats 1 would naturally fill up soonest. Even
after the play had begun the ushers might still be escorting late
arrivals to their places (Poen. 19-20). People at the back
had difficulty in hearing (Capt. 11-14). In one prologue a
gentleman at the back is invited to walk up nearer to the stage.
In another prostitutes are cautioned not to seat themselves on
the stage itself {Poen. 17-18). Ladies are warned not to
chatter ; nurses must not bring squalling children to the
theatre. There must have been considerable bustle and
confusion, particularly at the beginning. T h e actors had to
make themselves heard over this din, and it is not surprising
to find the speaker of a prologue declaring that he will not
burst his lungs for any one {Capt. 14). Once the play was
fairly started the actors might hope that the spectators would
at any rate keep still. Indeed it m a y have been difficult to
leave one's seat till the end of the performance. Hunger and
thirst affected audiences then as now, but no provision for
refreshment seems to have existed in the theatre. T h e audience
are bluntly told {Poen. 6-10, 30-1) that there is nothing to
satisfy their appetite except the play itself. Outside the theatre
vendors of eatables plied their trade during the festival; the
lackeys are advised {Poen. 40-3) to make a dash to a neighbour-
ing cookshop while the cakes are still hot. In the Greek
theatres, we are told, vendors of refreshments made the rounds
of the auditorium during the performance ; apparently this
was not the custom in Republican Rome. Suetonius tells us
that Augustus was shocked to see a knight drinking in the
theatre, and sent a messenger to him to say ' W h e n I want a
drink, I go h o m e ' . ' Yes,' replied the k n i g h t , ' but the Emperor
is not afraid of losing his place.' Decorum was probably less
strict in the back seats. A play could probably be got through
in a couple of hours ; those who wished might then make their
escape before the next performance began.
Even after the days of Plautus and Terence, it seems that
the actors had no light task in confronting so motley an
audience. Public attention was fickle ; the spectators were
as much interested in each other as in the play. T h e arrival
1 Except those (if any) which were reserved for privileged persons such as

senators.
 THE SPECTATORS 167
of an important public personage in the front seats, a remark
on the stage which might seem to be a topical allusion, a
mishap of any kind, might take their thoughts off the play.
Nevertheless we have evidence that the Roman audiences
were capable of following a scene with close attention. The
generous contest between Orestes and Pylades, each trying
at the risk of his own life to save the other, brought the whole
theatre to their feet in enthusiasm. The very existence of
plays like the Captivi suggests that we should not too readily
despise the public for whose entertainment they were written.

Roman Theatre Tickets

 C H A P T E R X X I I I

THE STAGE AND THE ACTORS' HOUSE

U R N O T I O N S of the appearance of the Plautine

stage are derived to some extent from the still visible
remains of the imperial theatres, the extant illustrations
dating from imperial times, and the description written by
Vitruvius in the time of Augustus. Such information may
be misleading. There is nothing more certain than that the
imperial theatre was ornate, there is nothing more likely
than that the Plautine theatre was simple. From the evidence
of the plays I will endeavour to describe the simplest arrange-
ment which would permit these plays to be staged.
T h e essential feature of the Roman theatre from the earliest
times was the wooden stage ; Plautus calls it scaena or pro-
scaenium,1 and the Latin for ' dramatic festival' is ludi scaenici.
It was probably not more than five feet high, but may even
in the days of Plautus have been of considerable length and
some depth. 2 Between the stage and the foremost tier of seats
lay a flat space, corresponding roughly to the Greek orchestra
or ' dancing-place ' and called orchestra by later generations
of Romans, but not normally used by Roman performers ;
here some movable seats were sometimes set for distinguished
spectators. From the orchestra a short flight of steps led up
to the stage ; this flight of steps would have been convenient
for any member of the public who wished to appear on the
stage (like the prostitutes who are forbidden to do so in Poen.
17-8), but it does not seem to have been used by the actors in
the performance of a play.
Behind the stage stood the actors' house or dressing-room,
the front wall of which formed the permanent back-scene.
A t either end the stage was enclosed by the projecting wings
of the actors' house. T h e front wall of the house was pierced
1 Scaena can mean (1) scene-building, (2) stage, (3) a picture, whether on the

stage (e.g. one of the panels displayed on the periacti) or not.

8 This is the usual view, but in considering the evidence of the plays we should

remember that actors can make surprising use of even a small stage.
168
 THE STAGE AND THE ACTORS' HOUSE 169

by three doorways containing folding-doors (normally kept

closed). A t either end of the stage an open passage or
side-entrance led into the projecting wing (Vitruvius calls
it uersura) of the house. Thus actors had five means of
communication between the stage and the dressing-room: the
three doorways in the back-scene and the two side-entrances.
The doors must have been solidly constructed, for they had
to stand a good deal of hard knocking. Apart from the doors,
the back-scene seems to have been a blank wooden wall.
The house was of modest height, and had a practicable roof.
(There was as yet no roof over the stage). In Amph. 1008
Mercury announces his intention of climbing on to the roof
from inside the house in order to drive Amphitruo from the
door by emptying pots on his head.
The only object permanently to be seen on the stage was
the altar, which figures in so many plays. There was probably
as yet no scenery and no attempt to adapt the setting for any
particular play or scene of a play. There was no drop-curtain ;
the back-scene, such as it was, lay permanently open to view. 1
Our only evidence on these matters is the text of the plays
—that is, the words which the dramatists put into the
mouths of their characters. We have no direct evidence as to
how these plays were actually staged. The responsibility for
the staging and production belonged not to the dramatist but
to the actor-manager. The dramatist was acquainted with the
general conditions of the contemporary stage ; but we have
no evidence that he instructed the actor-manager as to how
particular plays, or scenes in plays, were to be shown. It
seems to me fallacious to argue from a particular reference in
a play that there was something unusual in the way in which
that play was actually staged—whether at the first production
or at a revival performance. The actor-manager no doubt had
his own ideas—which may not have been those of the dramatist,
or of other managers. All that we can hope to establish is the
general conditions necessary for the production of our Latin
plays—which do not contain (and perhaps never did contain)
any stage-directions other than what is implied by the words
put into the mouths of the characters.
1
See Appendix (e). Vitruvius (V. v. 7) remarks that musicians would turn
towards the stage-doors when they wished to obtain superior tonus. Evidently the
doors were still of wood in his day, though the rest of the scene-building was now
built of stone.
170 THE ROMAN STAGE

THE CONVENTIONS AND PRACTICES OF

ROMAN STAGECRAFT
Roman drama owed what success it achieved not to the
craft of the scene-painter but to the art of the dramatist and
actor. That art was exercised according to the conventions
of the ancient theatre, conventions accepted almost uncon-
sciously by the contemporary audience, yet puzzling and sur-
prising to the reader of another age and country. All drama
rests upon convention of some kind. It is absurd that people
should discuss their most intimate concerns within hearing of
the public ; yet without this convention drama would be
impossible. Convention enables the imagination to redeem
material deficiencies ; yet a certain amount of realism in dress,
scenery and properties has often been found helpful. When we
attempt to use the Latin plays as evidence for the way in which
they were staged, we have frequently to ask ourselves whether
the passage we are considering is an instance of convention
or of realism. The fact that an actor mentions some object
as present may sometimes be evidence that that object was
actually shown on the stage ; at other times we know that the
object was not and could not be shown to the eye, and therefore
had to be suggested to the imagination by words and gestures.
The entry of an actor carrying a lamp may serve to indicate
that it is early morning ; the lamp is real, the darkness is
indicated by the actor's actions and words.
The modern convention which enables our theatre-going
audience to see into the interior of a house would have startled
the Greeks and Romans. Their basic convention was quite
different. The stage represented for them an open street,
or some other open place; they were the general public
assembled on the other side of the street or in the open country,
and looking at the buildings which fronted on the street or
open space. Every scene, in order to be shown on the stage,
had to be thought of as taking place in the open air. In
Mediterranean countries much does take place in the open
which in our latitudes would occur indoors ; but the real
and sufficient reason for staging a banquet, a toilet-scene or
a confidential conversation on the street was that otherwise
such a scene could not be staged at all.
The expedients to which the dramatists are forced to
 THE STAGE AND THE ACTORS' HOUSE 171

resort by this convention are evidence of the validity of the

convention itself. If it is necessary to disclose what is supposed
to be taking place within the house, a character on the stage
may be asked to peep inside the door and report what he sees
(Bacch. 833 if.). The spectators can never look through those
doors with their own eyes. The test case is the banquet in
the Asinaria. The party begins indoors (11. 745, 809-10),
but at 828 the revellers appear ; are we to understand that,
as the Loeb editor would have it, the doors are opened and
they are shown just inside ? Admittedly this would make it
easier to understand why Artemona, who comes out of her
own house at line 851, does not see them till the Parasite
points them out to her at line 880. But this solution raises
more difficulties than it removes, and the evidence of the
text is against it. Lines 828-9 (omitted by the Loeb editor,
following Leo) show the party coming out and the slaves
setting tables for them. Artemona's failure to see them at
first is to be explained not by the presence of any physical
obstacle but by the convention that the actor does not see
what the dramatist does not want him to see ; in other words,
she takes care not to look in their direction. Her dramatic
irruption on the party would lose its effect if we supposed it
to take place within the actors' house ; the wrangle between
husband and wife must have occurred on the open stage. At
line 940-1 Artemona carries her husband off, and the two
remaining members of the party go into Philaenium's house.
Our modern stage convention enables us to look for a
while into the interior of a house, which at the end of the
scene will be hidden from us by the drop curtain. The ancient
stage, which represented the open street, lay, like the street,
permanently open to view. The plays of Plautus are so
constructed as to make it clear that no drop curtain was used
or known. Each play begins with an empty stage ; char-
acters come from their houses or from one of the side-entrances,
usually explaining in their opening words who they are, whence
they have come and whither they are going. At the end of
the play some pretext is made for getting the characters off
the stage, but before they go one of them informs the audience
that the play is over and asks for their applause. There is
no such thing as an opening or final tableau ; and at the end
of the play the stage is empty and set for the beginning of the
172 THE ROMAN STAGE

next play. If in the course of the play some object has to be

brought on the stage for a particular scene, we presently see
it being taken away again when it has served its purpose.
Of this we have a striking example in the Mercator : When the
cook has set down his dishes at the door and walked off in
high annoyance, the unhappy husband has to ask his wife
to have the now detested dishes taken inside, lamely remarking
that they will improve the family meal. (11. 800-2.)
The absence of a curtain implies the absence of special
scenery for any one play or scene in a play. The opening
scenes of the Rudens present us with a picture of a wild land-
scape, covered with rocks, full of caves, overgrown with rushes.
If we suppose that these features were actually depicted on
the stage, then we have to face the dilemma that either they
remained on the stage throughout the rest of the play (in
which they are entirely ignored) or that they were bodily
removed under the eyes of the spectators during the course
of the play. It is much more likely that the references to the
natural surroundings were addressed to the imagination, and
that the efforts of the two girls to find each other (220-243)
are frustrated for so long by nothing but the fact that they
take care not to look in the right direction. It would, no
doubt, as Dr. Pickard-Cambridge has suggested to me (cf. his
Theatre of Dionysus, p. 68), have been a simple matter for the
stage-carpenter to contrive a suitable setting for a particular
occasion ; but such a setting would have been a positive
embarrassment when the occasion for it was over. Yet how
could it be removed ? The same considerations seem to apply
to the setting of a play as a whole. That play succeeded play
with the same permanent background seems to be indicated
by Pseud, ι, 2, and Men. 72-6. Such as it was, therefore,
the scenery probably remained unchanged from play to
play.
Reduced to the barest terms, the permanent scenery
consisted of the plain wall at the back with its three doors,
the two projecting wings with the side-entrances, the flat roof
of the actors' house, and on the stage itself an altar. The
roof is actually used in the Amphitruo (1008 and frag, iv-vi)
and would seem to be mentioned in M.G. 156 ff. and Rud.
85 ff. ; the five entrances were in constant use. The three
doors at the back could represent one, two, or three separate
 THE STAGE AND THE ACTORS' HOUSE 173

houses 1 ; perhaps the side-entrance on the right of the spectators

was supposed to lead to the near distance, that on the left to the
more remote distance. Thus if the action were laid in town
the right-hand side-entrance might lead to the town centre, that
on the left to the country and harbour. In the Rudens the
setting is a lonely spot near the African sea-shore ; in the back-
ground are the cottage of Daemones and the temple of Venus,
represented by two of the three doors ; the side-entrance
on the right would lead to the beach near by, that on the left to
the town and harbour of Gyrene. In Amph. 333, set in the
city of Thebes, Sosia is speaking as he enters from the harbour,
i.e. from the spectators' l e f t ; Mercury, facing the audience,
says ' some one is speaking on my rightOther relevant
passages are Men. 555, Rud. 156, And. 734. T h e significance
of the stage-doors and the side-entrances in any particular
play is usually made clear in the prologue and in the remarks
of the characters. 2
In the normal way any one entering or leaving any of the
1 houses' or making his way between the town centre on one

side and the harbour or country on the other must appear

on the stage. When for some reason the dramatist wishes to
move a character from one of these places to another without
bringing him on to the stage he falls back on another con-
vention, the use of the angiportus (or angiportum).
The word angiportus means ' street*, and can be used even
of the street represented by the stage. Its special significance
in drama is to denote the back-street supposed to run behind
the houses which face on to the stage, connecting them by
means of their back-doors and gardens with each other and
also with the town, harbour and country. T h e angiportus is
not shown and could not be shown to the eye ; its use is
exceptional, and requires express mention ; it is a device which
enables the dramatist to escape at times from the general
rule that a character who leaves the stage by a particular door
or wing must return by the same door or wing. 8
In order to make the plot easily intelligible to the spectators
1 That unwanted doors were temporarily concealed by a curtain is con·

ceivable, but unproved. In the H.T. the third door is used once, and only once, in
the play. That the Greek dramatists sometimes used a two-door setting which the
Latin dramatists have altered to a three-door setting is altogether unlikely ; see
Appendix (g). s See Appendix (A).

* See Appendix (e). In Rud. 156, And. 734, I have to suppose that the speaker
has turned slightly away from the spectators.
174 THE ROMAN STAGE

(and incidentally to help the actors when rehearsing the play)

it is usual for each entrance on the stage to be announced before-
hand. These announcements, which are part of the dialogue,
serve as stage-directions, and were probably meant to give a
waiting actor his cue. The general custom that a character
already on the stage shall warn the audience that some one else
is about to appear leads to some interesting results. That a
character standing on the stage, which represents a street,
should be able to see farther up and down the street than the
spectators can see, and so should be able to perceive some one
approaching from the town-centre or harbour before they can
do so, is quite natural ; but it would not seem so natural for
him to announce that some one is coming out of one of the
house-doors, which faces the audience and is behind his back.
This difficulty is got over by making the opening door creak
and so draw the actor's attention. Like the whole of the actor's
technique, this device is of Greek origin. Sometimes the Greek
dramatist describes the person coming out from the house as
' striking ' the door (i.e. in the act of opening it). This phrase,
misunderstood in late antiquity, gave rise to the absurd notion
that the early Greeks knocked on doors not only when they
wanted to go in but also when they were coming out. 1
A grand principle of the stage was that a character saw and
heard only what the dramatist meant him to see and hear.
We are familiar with the situation in which one character spies
on another and overhears his words while himself remaining
(at least temporarily) unobserved. Both actors are on the
stage, in full view of the spectators. There is no physical
object on the stage behind which the eaves-dropper can take
shelter. He usually secures his temporary invisibility merely
by standing at the back of the stage, and he is always liable to
be detected as soon as the other actor allows himself to look in
the right direction. This is absurd, perhaps, but no more
absurd than that other convention of the ancient stage, the
' aside ', audible to thousands of spectators, yet inaudible, or
only partly audible, to a character standing only a few yards
from the speaker.
When we reflect on the fact that all the actors 2 wore masks
1
See Appendix (g).
1
Excluding, perhaps, the speaker of the prologue when appearing in his own
person.
 THE STAGE AND THE ACTORS' HOUSE 175

and that a very large part of their lines was not spoken but
declaimed to the accompaniment of the flute-player, we realize
that the Roman style of acting must have differed widely from
the naturalistic, conversational style of our own day. Our
actors talk to each other ; the Roman actors declaimed to the
audience. They stood, where possible, well to the front of
the stage ; they faced the audience, they kept their eyes on the
audience, they aimed above all things at making their words
carry even to the farthest seats.1 If an actor entered from a
house-door he usually walked to the front of the stage, looking
neither to his left nor to his right. This made it easy for others
already on the stage to step back and observe him. A character
coming in from a side-entrance would also address himself to
the audience and turn as best he could towards them while
making his way to the middle of the stage. It is thus not
surprising if for some time he fails to notice the presence of
other characters on the stage. To face the audience was very
necessary if one's words were to carry to the back of the large,
open-air theatre. The art of the actor lay not in naturalism or
in mimicry but in clear utterance conveying the appropriate
emotion and supported by appropriate gesture. It seems
probable that the actor made no attempt to alter his voice
according to whether he was taking the part of a gentleman or
a slave, a man or a woman. In the same way the dramatist
makes all his characters talk the same kind of Latin. The art
of gesture, on the other hand, was carried in ancient times to
a height which we can scarcely comprehend. When we try to
imagine a scene from a Roman play, we must picture to our-
selves the masked actors, their gestures, their carefully plotted
movements on the stage, their voices raised in rhythmical
declamation, while the flute-player stepped up now to one,
now to another,2 playing the accompaniment for each actor in
turn. When we consider how widely the style of Roman
acting must have differed from our own and how misleading
Roman writers often are on this subject, we realize how
dangerous it is to use our own notions of fitness as a guide to
the practice of the Roman stage.
1
Vitruvius (V. iii. 4) points out how important it is that the case-endings of
words should be audible even in the " gods " . Of course I do not mean that an
actor was never allowed to turn away from the spectators. See p. 173, note a.
1
Cie. Pro Mur. is, a6.
 CHAPTER XXIV

COSTUMES AND MASKS

I N T H I S chapter my object is (i) to describe the costumes

worn by actors in Republican times, (2) to show how these
costumes were turned to account by the dramatists. The
evidence used will be that of the extant plays and fragments.
It is usual to supplement this evidence by reference to the
statements of Pollux, Donatus and other late writers and to
Campanian reliefs, wall-paintings, terra-cottas and the illus-
trations in certain medieval manuscripts of Terence. Un-
fortunately all of this material is of doubtful value, interesting
when it is supported by the evidence of the plays, but otherwise
only too likely to give a wrong impression.
Costume on the Roman stage evidently varied according
to the type of drama, for it was used as the basis of classification
of the different types of drama. O u r complete plays belong
to what Donatus {De Com. vi §§ ι and 6) calls the palliata,
comedy in which the characteristic dress was the pallium or
Greek mantle of everyday wear. T h e Roman actors of the
palliata, like the Greek actors of New Comedy, wore the
ordinary Greek dress of contemporary life, with certain
modifications which will be described. I shall begin by summar-
ising the essential features of ordinary Greek dress, most of
which are mentioned in the plays.
T h e undergarment was the Ionic chiton (Latin tunica),
a linen or woollen shirt with holes for the neck and arms.
Sometimes it had sleeves. It was drawn on over the head,
and fastened with a girdle by which it could be tucked up
if necessary. Sometimes the chiton was the only garment.
But it was usual to wear outside the chiton the himation
(Latin pallium), a woollen wrap, rectangular in shape, which
was drawn round the body and adjusted in various ways. It
was usually fastened with a clasp over one shoulder. Sometimes
the pallium was the only garment. As the ordinary pallium
was somewhat cumbrous, we find that where freedom of
176
 COSTUMES AND MASKS 177
movement was required the pallium was not worn, or was
worn in a special way. Workmen might wear the tunic
alone. Others might wear outside the tunic the chlamys, a
kind of light-weight pallium, often gay in colour (like our
blazer).
Footwear for town (e/tßaSe?) was normally light: sandals
(υποδήματα, soleae; sometimes the Romans used the word
crepidae, from the Greek κρψτίς), and slippers (called by the
Romans socct). The sandal was fastened with thongs and
bands ; the soccus had no fastenings.
The head was normally uncovered. But Greek workmen
wore caps of felt (ττΐΚοΐ) or leather (κνναί). The broad-
brimmed hat (petasos, caused) came with the chlamys from
Thessaly, the two forming part of the characteristic costume
of Athenian youths when serving in the cavalry. The petasos
had a band which fastened it round the head, and a strap
under the chin which could be used by the wearer to let the
hat hang down his back. The Roman cap (pilleus) was
similar to the conical 7rt\o? ; it was in particular the head-
wear of newly emancipated slaves. The petasos was sometimes
worn by Roman invalids or valetudinarians, as for example
Augustus ; Caligula sanctioned its use in the theatre as a
protection against the sun.
Greek women, also, wore suitable sandals, chiton and
himation (the Latin word for a woman's cloak is palla). As
in real life, both in Athens and in Rome, there was no dis-
tinctive garb for slaves ; some town slaves might wear much
the same clothes as their masters, others went barefoot. 1
Rustics, whether slave or free, might wear goatskins.
No doubt in real life there was much greater variety than
would appear from this summary. Aristophanes mentions
many kinds of garments and of footwear of which we hear
nothing in New Comedy. No doubt, again, practical con-
siderations would enforce simplicity of wardrobe on the actors
of this cosmopolitan form of drama. We are often, indeed,
asked to believe that custom prescribed a particular costume
for every age and profession. Thus Donatus (i. 29-30)
dealing with stage practice, tells us ' Old men in comedy are
dressed in white, as that is said to be the oldest style ; young
men wear garments which contrast with each other in colour.
1 But not, I think, in the palliata ; see p. 180.
i78 THE ROMAN STAGE

Slaves in comedy wear a short garb, either because of the

poverty of early times or to allow of free movement. Parasites
wear their mantles w r a p p e d ' (presumably in some special
way). ' White is the colour for a cheerful character ; a man in
trouble wears shabby clothes. Purple (purpureas) is the colour
for the rich, red (puniceus) for the poor. A Captain wears
a purple chlamys ; girls are dressed in foreign style. A pimp
wears a costume of variegated hue ; a harlot is given a yellow
mantle to indicate her avarice. Trailing robes (syrmata)
are worn by characters in grief to show their neglect of per-
sonal appearance.' Pollux, on the other hand (iv. 119 if.)
says that young men wear a red or dark purple mantle, pimps
a ' d y e d ' (presumably brightly dyed) tunic and ' flowery'
(presumably gay) mantle (as did the courtesans at Athens,
according to Phylarchus), that young men wear white linen,
heiresses white (tunics) with a fringe, old women ' apple-
coloured ' (perhaps ' quince-yellow') or sky-blue clothes,
except for the priestesses, who wear white ; bawds a bright red
ribbon round the head. W e have also Pollux's detailed
description of forty-four masks used in comedy (nine for
old men, eleven for young men, seven for slaves, three for old
women, five for young women, seven for courtesans and two
for servant-maids). It is customary for modern writers
to say that as soon as a character appeared on the stage the
spectators knew from his appearance a good deal about
him : ' par elles ' (the colours) ' en meme temps que par les
masques, le public ötait souvent instruit sur le champ de
l'äge, de la condition, du caract£re meme des personnages
qui entraient en scene. Cette convention avait passe dans
la comoedia palliata des Romains.' (Navarre, Le Thidtre
grec, pp. 225-6).
I follow Navarre in holding that the Roman stage-
convention was based on the Greek—even with regard to the
use of masks. T h e actor's mask covered the whole head
(Gellius v. 7) and the hair belonging to a mask was coloured
to suit the part. Thus in general the elderly would appear
grey-haired, or in some cases partly bald ; the young would
usually have dark hair ; most slaves would have red hair.
Women's hair would look quite different from that of men.
But even these statements must be accepted with some reserve,
in view of the evidence of the plays.
POET AND TRAGIC MUSE
C H O O S I N G T H E MASKS (?)
Relief in the Lateran Museum
 COSTUMES AND MASKS I79
Let us consider the practical necessities of the stage. As
performance was continuous, actors had to be able to change
their costumes in a very short time (sometimes within six
lines).1 Costumes, especially elaborate costumes, were ex-
pensive. The natural tendency would be to reduce the
property manager's outfit to a limited number of costumes
of as few types and as simple style as possible. Masks on the
other hand were probably cheap. They had to be well
differentiated from each other, for it was the mask which
gave each character its identity. As in real life, so on the stage,
one character would be known from another not by the clothes
he wore but by his features. But if we were to enforce on the
property-manager the rules laid down by Pollux and Donatus,
we should greatly increase the extent and cost of the outfit.
Moreover these writers are not in agreement with each other,
or even with themselves. Old men, we are told by Donatus,
are dressed in white ; but white is also the colour for a cheerful
character. What is an old man to wear when he is not
cheerful ? Even if we dismiss these minor questions, the main
fact remains that Donatus and Pollux, as well as the ancient
artists, were thinking in visual terms. But in the plays we
get the impression that the appeal was not to the eye but to the
ear and the imagination of the public. When a new character
appears, we are told in plain language who he is. It is only
in special circumstances that stress is laid on his costume.
Men dressed somewhat differently from women ; slaves of the
rougher type, and some foreigners, were dressed to suit their
part; otherwise one would gather from the plays that all
characters wore much the same clothes, a tunic, a pallium
and (presumably) a pair of slippers or sandals. In fact
costume is comparatively unimportant in New Comedy.
But we are more aware of it in Plautus than in Terence or
Menander.
Plautus, as a native of Italy, is conscious that his characters
are dressed like Greeks, Graeci palliati {Cure. 288), in contrast
to the spectators, who wear the toga (Amph. 68). He also
realizes that the actor's costume costs money (Cure. 464-6;
Pers. 157-160; Trin. 8 5 7 - 9 ; Pseud. 1184-6; Amph. 85),
particularly when it is out of the ordinary. Little is said of
footwear. Slaves wear slippers (Epid. 725 ; Trin. 720) like
1 Kurrelmeyer, Economy of actors in Plautus, p. 19.
 ι8ο THE ROMAN STAGE

free men (Heaut. 124). Young men and women wear sandals,
which they order to be removed when they are about to recline
on a couch {Most. 384 ; True. 478, 631). On one occasion
special sandals are worn as part of a foreign-looking disguise
(Pers. 464). Perhaps we should conclude that ' slippers' and
' sandals' mean much the same thing, or that, if they were
different, the normal wear was slippers. Indeed the soccus
became a synonym for comedy of this kind (as contrasted with
the high boot or cothurnus of tragedy, or with the barefooted
mime). No character is said to be barefooted, or to have
any footwear other than slippers or sandals. Though the
tunic and the mantle are often mentioned, there is little
evidence that one tunic or one pallium differed from another.
A woman's palla differed from a man's pallium (Men. 1 9 1 - 2 ,
cf. Aristophanes Eccles.) and a man may be disguised in
woman's clothes (Cas. 769-70). A bridegroom wears white
clothes (Cas. 767-8). Old men are evidently recognizable
as such at sight—but this may be a matter of the mask and
gait and not of the costume. The chlamys is worn instead of
the pallium by soldiers, some travellers and some foreigners.
The tunic is sometimes allowed to trail by women (and
innkeepers). (Po. 1298-1303). There is a jesting reference to
a sleeved tunic (Ps. 738).
We must allow for the fact that this is comedy, and that
a dramatist may sometimes think fit to make fun of what is
really the established convention, whether in dress or in any-
thing else. The general impression I derive is that costume
was more or less the same for all characters, that the profession
of a character was sometimes indicated by such attributes as
the cook's knife, the fisherman's hooks and line or his net,
the soldier's sword and so on, and that it was the mask which
gave the character his individuality within the play as a
person of a particular age and temperament.1 Where costume
becomes important is when it is used to disguise a character.
The disguise is usually outlandish: A character puts on a
chlamys and a travelling hat and pretends to be a foreigner.
He cannot change his mask, for the mask is the character.
Therefore, if he wishes to deceive some one who knows him,
1
The courtesan can pose as a respectable woman by merely altering her head-
dress and her gait (M.G. 791-2, 872). (Apparently, then, it was possible to alter
the appearance of the hair which went with the mask). In the fabula togata
he would need a long robe as well (see p. 184).
 COSTUMES AND MASKS 181
he must conceal his mask by wearing a patch over one or both
eyes. The spectators will of course penetrate his disguise—for
they must be able to recognize him. His victim will be de-
ceived, because the dramatist so ordains it. (Pers. 155-6 ;
M.G. 1 1 7 7 - 8 1 ; Ps. 7 3 5 ; Trin. 771, 8 5 1 ; Cure. 392-5,
461-5, 505, 54З75)·
T h e dramatist uses costume for his purposes ; he is not
the slave of convention. Since the two Menaechmi must
appear alike, the traveller from Syracuse is dressed just
like his brother who appears from his own house. In the
Amphitruo both Amphitruo and his slave Sosia wear hats,
as having just arrived from abroad. But so do their divine
doubles, Jupiter and Mercury, who are pretending to be
leaving for abroad (142-6), but in fact do not leave. The
writer of this prologue has considered a problem of staging :
in the case of doubles, how are the spectators to know one from
the other ? The solution here is that each of the gods will wear
a distinguishing mark attached to, or under, his hat. Hence,
we may suppose, the need for hats—which are actually worn
by Sosia and Mercury in the opening scene, but are never
mentioned again. Indeed Mercury must have removed his
hat before he can think of putting on a garland (999). In
fact there is no need for the hats or the distinguishing marks :
Sosia and Mercury, when they meet in the opening scene,
speak in character, and nobody could mistake one for the other.
Besides, Sosia carries a lantern. T h e two Menaechmi are sup-
posed to be exactly alike ; yet it is impossible to mistake one
for the other. Each makes his identity clear by his words ;
each enters and leaves by his own side-entrance ; the stolen
palla which one or the other carries in turn helps further to
identify him.
In the Casina (769, 814 ff.) a man dresses up as a girl.
Apparently the' bride's' breast is padded realistically (848); she
does not speak, and we can imagine that her face was hidden
by the bridal veil, worn in Greek as well as in Roman weddings.
The plot of the Captiui turns on the exchange of identity
between master and slave. According to the prologue (37)
Philocrates and Tyndarus have changed clothes ; but this
is not in keeping with the text of the play ; for when Aris-
tophontes meets and recognizes Tyndarus (541) he expresses
no surprise at Tyndarus' clothes ; from which it would seem
13
 THE ROMAN STAGE

that neither Philocrates nor Tyndarus had worn a dress

which was indicative of their status. It would further seem
that neither wore a mask which was distinctively that of a
slave. Hegio takes Philocrates to be the slave ; yet we know
that he is really the gentleman ; besides, he will later appear
(922) in his true character, and the spectators must be able
to recognize his features. His appearance is described (647-8);
we at once notice that he has red hair—like all the slaves
whose hair is described in comedy. Evidently, then, red hair
was not confined to slaves. Tyndarus cannot wear a slave-
mask either ; for Hegio takes him for the master, and we
know that he is really Hegio's son. Menaechmus of Epi-
damnus, when rescued by his brother's slave, addresses
him simply as ' y o u n g m a n ' (1021, 5). Some slaves dress
like dandies (e.g. Tranio in the opening scene of the
Mostellaria). There is a difference between one slave and
another, no doubt ; Tranio is more elegant than Grumio,
Trachalio is superior socially to Sceparnio or Gripus. But
we have no evidence that on the stage, any more than in
real life, it was possible to tell a slave at sight from a freeman.
(Cf. Pseud. 610, where Harpax says to Pseudolus ' are you a
free man or a slave ? ' also Amfih. 343.)
T h e stress laid on such properties as the cook's knife,
or the soldier's sword, itself suggests that the clothes worn
were not in themselves indicative of profession. It is said
by Pollux that the parasite carried an oil flask and scraper as
a professional badge. This is perhaps merely an inference
from the plays ; the parasite is the poorest creature on earth ;
all he has are the indispensable articles of toilet—but perhaps
he keeps them at home. O n the stage all he has—all he can
pledge in a wager—is his p a l l i u m 1 (and presumably tunic
and slippers).
In the prologue to the Amphitruo (116-7) Mercury refers
to his ' slave's get-up ', seruile schema. W e cannot tell what
in his appearance was particularly slave-like ; he wears tunic
and cloak, travelling hat and beard. Perhaps all he means by
seruile schema is ' in human guise ' (instead of appearing in
divine form). So common is the wearing of the pallium with
town-slaves that they have a special way of wrapping it
round their necks when in a hurry.
1 Cure. 355 ; St. 230.
 COSTUMES AND MASKS 183

W e have thus reached the surprising conclusion that

costume may have played comparatively little part in helping
the spectators to follow a play. T h e costumes were not
particularly distinctive. T h e mask distinguished the individual
as such, but did not reveal his profession or status. It is taken
for granted ; Plautus seldom or never draws attention to its
use. O f course the references to facial expression in Latin as
in Greek plays must not be taken to mean that the actor's own
features were visible. Change of expression was impossible.
Quintilian indeed tells us of a mask with one cheerful and one
serious eye, which enables the wearer to display to the spectators
a different appearance according to the needs of the play. Such
devices seem difficult to reconcile with the needs of the real
stage. It is far better to suppose that language referring to
change of feature is addressed to the imagination of the
spectators, aided by the gestures of the actor.

STAGE COSTUME IN FORMS OF DRAMA OTHER

THAN THE PALLIATA

Here we no longer have complete plays to help us, and the

fragments throw little light on the subject. In tragedy trans-
lated from the Greek the actors wear Greek dress ; V a r r o used
the term palliata of derivative tragedy as well as comedy. W e
nowhere find derivative tragedy distinguished from derivative
comedy by the dress of the actors ; what we do find is that
the high boot worn by R o m a n tragic actors (the Romans called
it cothurnus) is used as a synonym for tragedy, just as soccus is
used to denote comedy. T h e Greeks did not use κόθορνος in
this sense ; with them the cothurnus was worn by women (Ar.
Frogs 47, 5 5 7 ; Lys. 657) and hunters. W e have no compelling
reason to believe that Greek tragic actors wore any exaggerated
footwear; the high tragic boot, as well as the use of the word
cothurnus to denote it, may have been of Italian origin. Horace
uses cothurnus of ' t r a g e d y ' . But it seems unlikely that the
Republican actors wore boots with the immensely thick soles
which we see on the ivory statuette from Rieti. Perhaps we
should picture the actors of Ennius' day as wearing a somewhat
rich form of contemporary Greek dress ; Plautus (Capt. 61-2)
indicates that the outfit for comedy was not adequate for the
presentation of tragedy.
 THE ROMAN STAGE

In native drama native dress was the rule ; on this point

the Romans were sensitive. The toga would therefore be
worn with appropriate Roman footwear both in native comedy
and in historical drama ; both were included, according to
Diomedes, under the title togata. But the characteristic
purple-striped toga worn by Roman magistrates was used
specifically to distinguish the historical play, fabula praetexta
or praetextata, from native comedy as well as other forms of
drama. The meagre fragments of the praetexta tell us nothing
about costume ; in the togata we have references to the toga
in Tit. 25, 44, to special patrician shoes (116), to the perfumed
tunic and toga of some dandy (138), to ' white togas and dirty
tunics' (167-8), to some one found out-of-doors in sandals
(which a Roman would have regarded as indoor wear, suitable
for banquets) ; see Afr. 105. It seems clear, therefore, that
the wear for actors in native comedy was toga, tunic and
shoes.1
The fabula Atellana and the mime had each its own
distinctive style. The Atellana was characterized by traditional
masks ; we meet a bald fisherman {Pomp. 119) and a bald
messenger (135). Masks are mentioned (Nov. 2). The mime
was characterized by the bare feet of the typical performer.
Laberius refers to the ' toga-wearing race' (43, 45). He
mentions a tunic (61).
Such is the meagre evidence which I have been able to
glean from the fragments. It may be that these varieties of
drama did not permanently adhere to the traditional costumes,
but that they all tended to influence each other, in costume
as well as in literary form.

THE USE OF MASKS ON THE ROMAN STAGE

Greek drama was always a masked performance. The

advantages of the use of masks are that it enables a small
company to represent a considerable number of characters :
that the less important parts can thus be as well rendered as
the leading roles ; that the exaggeration inherent in mask-
design enables spectators even at a distance to distinguish the
1
Afranius ( 1 3 3 - 4 ) indicates that courtesans normally wore a short dress,
although for safety's sake they might sometimes assume the stola of the R o m a n
matrons.
 COSTUMES AND MASKS

features of the character represented ; that women's parts can

be taken b y men ; that the appearance of characters can be
made to correspond with the requirements of the play ; above
all, perhaps, that the identity of the actor is concealed. T h e
R o m a n acting tradition was derived from the Greek. Never-
theless w e are told b y a late grammarian, Diomedes, that at
first wigs, not masks, were used on the R o m a n stage ; masks,
he says, were introduced b y Roscius to hide his squint. Donatus
names as the first ' masked actors ' Cincius Faliscus in comedy
and Minucius Prothymus in tragedy. Perhaps Donatus is
using the wordpersonati simply a s ' actors ' (of course in masks),
and his meaning is that these were the first R o m a n actors.
A t any rate it is evident that Diomedes and Donatus are not
in agreement. Cicero, speaking of the importance of an
orator's play of features, speaks approvingly of those elder
contemporaries w h o ' did not speak very highly even of Roscius
when he had his mask on '. A g a i n Cicero m a y be using
personatus as synonymous with ' a c t o r ' , his meaning being
that the orator necessarily has an advantage in the eyes of
good judges over even the best of actors, because his features
are visible. Possibly this very passage of Cicero was the source
of Diomedes' statement. It is the custom nowadays to accept
Diomedes' evidence, thus denying the use of masks to the
R o m a n stage in the time of Plautus and Terence. Further
evidence is sought in the references to facial expression w h i c h
occur in Latin comedy, e.g. Adelphi 643 ' he has b l u s h e d ; all
is well But we find similar references in Greek drama, which
was certainly performed by masked actors. Terence uses
persona in the applied sense of ' character in a p l a y ' (Eun.
26, 32, 35). I f the original sense of persona was ' m a s k '
(possibly connected with the masked figure Phersu w h o appears
in some Etruscan paintings) we cannot explain Terence's
applied use except on the assumption that Terence was familiar
with the use of the mask to indicate the actor's role. The
mask was traditional in the fabula Atellana, and must have
been familiar to the R o m a n s even before the introduction of
Greek drama in translation. It is hard to see w h y in adapting
Greek d r a m a they should have denied themselves so useful a
device. Naevius is said b y Festus to have written a comedy
1 O f course the actor could not blush at will. The simple explanation of this
and similar passages is to be found in the use of gesture.
l86 THE ROMAN STAGE

entitled ' Personata ' (? personata fabula), and Plautus' fond-

ness for the theme of identity of appearance itself suggests that
he took the mask for granted. He uses the word fiersolla,
diminutive of persona (Cure. 192), in the sense of ' ugly face
T h e frequent descriptions of the appearance of characters—
including their faces—are more likely to have been descriptions
of masks than of the actual features of members of the company.

COSTUME OF PROLOGUE

T o judge from Нес. 9, orator ad uos uenio ornatu prologi, it

might seem that there was a special costume for the speaker of
the prologue. Nevertheless this prologue is spoken by Ambivius
Turpio in his own character ; he reminds the audience of
how he had assisted Caecilius, and of his two failures to get a
hearing for the Hecyra. It would therefore seem that he is not
wearing a mask. W e must compare with this line Poen. 123,
126. In the second of these lines the Prologue tells the
spectators that he must now leave them in order to become
' a different character that is, to change his costume and take
a part in the play. In line 123 he has said ' I will go and put
on m y c o s t u m e w h i c h suggests that at the moment he is
speaking in his own character, without a mask. A special
prologue-mask would seem quite inappropriate for the speaker
of any of Terence's prologues, and also, probably, for the
Asinaria, Captiui, Casina, Menaechmi, Poenulus, PseudoIiis, Trucu-
lentus, and Vidularia. But when the prologue is spoken by one
of the characters in the play, he will naturally wear the costume
and mask of his part. In the Mercator, for example, the hero
himself introduces the play to us. In the Rudens, again, the
prologue is spoken by the god Arcturus, who apparently wears
a ' bright star T h e deferred prologue is always put into the
mouth of a character in the play or some other fictitious person,
such as ' Help ' o r ' Misapprehension T h e weight of evidence
seems to be against the view that there was a special costume
for the Prologue as such. In all probability he wore (if not
the toga) the usual tunic, pallium and slippers, but no mask ;
and was therefore recognizable at once as speaking in his own
person and on behalf of the dramatist. Apparently, then, in
Нес. 9., Ambivius Turpio is explaining that he comes on the
stage without a mask, in the usual style of the Prologue, to
 COSTUMES AND MASKS 187

make a special plea. T h e r e is thus no reason to suppose that

he was carrying an olive branch—a suggestion p u t forward on
a very doubtful interpretation of an illustration to the Adelfiki,
in which it is perhaps intended to show the speaker as carrying
a cypress branch, since the play had been produced (as the
artist could read in the didascalia) at the funeral games
for L. Aemilius Paulus.

N O T E ON POLLUX'S LIST OF MASKS

If we may judge by the evidence of works of art, as well as by
Pollux's list, mask-design was formal : it was not realistic. That
is why I think it dangerous to use our modern notions of realism
in order to connect a particular mask with a particular character
is extant drama. Either actual stage-usage was more flexible than
is usually allowed, or dramatists were indifferent to our anxiety
that the masks should harmonzie with the personality.
 CHAPTER XXV

THE R O M A N ORIGIN OF THE L A W OF FIVE ACTS

T H E I N T R O D U C T I O N of the five-act law, and of act-

division itself, into Renaissance d r a m a was due to the
desire to obey" classical precept and example. Horace h a d
laid it d o w n for all time that a play should have five acts if it
was to be a success. Moreover the plays of Terence h a d been
divided according to this principle by R o m a n editors, and so
came down in five-act form to the modern world. It would
seem natural to suppose that the l a w was based upon the
practice of classical dramatists. Y e t the most careful researches
of modern scholars have been unable to discover clear evidence
of five-act division in any Greek play which has reached us
complete, or in the plays of Plautus or Terence. Consequently
some have held that the l a w was a mere invention of theorists,
whether of Hellenistic or of R o m a n times. T h e discovery in
1905 and later of considerable portions of Menander's text
introduced a less sceptical view. T h e occasional occurrence
of the stage direction chorou, ' a performance b y the chorus
in these almost non-choral plays seemed clearly to divide them
into acts ; and though the evidence did not show how m a n y
such choral interludes occurred in any one play, it was found
that there were at least three of them in the Epitrepontes.
Further researches might well be expected to demonstrate the
existence of a fourth interlude in this particular play, and in
other plays as well ; and all agreed that a play with four
interludes must necessarily contain five acts. A n earlier event
of some importance was the appearance in 1884 of Prou's
paper ' Les theatres d'automates en Grece in which he drew
the attention of scholars to the account given b y Hero of
Alexandria of the Nauplios-show, with its five scenes, and
claimed it as a five-act play of Hellenistic times. T h e import-
ance of this new evidence was that it seemed to confirm in some
measure the statements of Donatus and Evanthius. F r o m them
w e gather (a) that each play of Terence should be divided into
188
 ROMAN ORIGIN OF THE LAW OF FIVE ACTS l8g

five acts ; (b) that this five-act arrangement was not introduced
by Terence, who tried in fact to eliminate act-division
altogether ; (c) that V a r r o found acts, not only in Terence but
also apud Graecos ipsos ; (d) that the Greek dramatists divided
acts by means of the chorus ; (e) that Menander, when he
suppressed the chorus, still ' left a place for it ', 1 A l l these
statements seemed now to be capable of explanation. Men-
ander had removed the chorus from taking part in the action
of the play ; he had, however, marked in his manuscript places
where it was to appear and perform an interlude. These
interludes divided each play of Menander (and perhaps of
New Comedy in general) into five sections—what the Romans
called actus. Terence, by omitting the interludes, had elimin-
ated the outward sign of act-division. T h e task of the R o m a n
scholars was therefore to restore this division into five acts.
And though they do not seem to have carried out this task very
successfully, at any rate we may assume that the Latin plays
were translated from five-act originals, and all that we have to
do is to find where those choral interludes occurred which
separated the acts from each other.
This reasoning was not universally accepted. T h e negative
view is expressed with admirable sense and humour by
Michaut. 2 But a positive theory has a tactical advantage over
scepticism. Research in Menander is now largely a matter of
arranging the scanty remains of each play according to the
five-act law. N o w it is one thing to divide into acts a play
which has survived complete ; such a procedure, whether
justified or not, at least leaves the text unchanged. But
Menander's text is largely a matter of conjecture ; and the
kind of conjectures put forward will be materially affected
according to whether their author accepts or rejects the
five-act law.
In 1940 there appeared Weissinger's ' Study of Act
Divisions in Classical Drama ' (Iowa Studies, no. ix). Weis-
singer believed, as most scholars do, that the law was of Greek
origin—that is to say that it had at any rate been formulated
by critics of the Alexandrian if not of an earlier age. H e had,
however, to confess that it cannot be shown to hold good
1 Compare with the present view the remark of K ö r t e in 1900 (Neue Jahrb.

fur das Klass. Alt. V . 88) : was sich der brave Euanthius mit den Worten ut choros
tollerent locum eis relinquentes gedacht hat, weiss ich nicht, vermutlich gar nichts.
2 ComMie romaine, i. 184-196.
 THE ROMAN STAGE

either for fifth-century tragedy, or for O l d C o m e d y , or for

Menander, or for Plautus and Terence. ' It is likely ' (page
60) ' that the five-act convention was becoming established in
Menander's day, but it is unsafe to conclude from the scanty
evidence that the rule had already become de rigueur.'
I venture to suggest that the trouble is not our lack of
evidence but our confusion of thought. O n e of the merits
of Weissinger's work is that he forces us to consider certain
fundamental questions which have too long been disregarded.
T h e first concerns the meaning of ' a c t O u r notions of
what it implies are based partly on the usage of the modern
theatre, partly on our reading of plays, particularly, it m a y be,
the plays of Shakespeare. In our theatres the curtain falls
at the end of each a c t ; the lights are switched on in the body
of the theatre, and the disciplined spectators are free to leave
the theatre for a stated time. For those w h o choose to remain
in their seats a light musical interlude is usually provided.
Meanwhile, behind the curtain, the scenery is prepared for
the next act. A t the end of the interval the lights are switched
off in the body of the theatre, the curtain rises, and the atten-
tion of all is- concentrated on the brilliantly lighted stage.
In other words, modern developments have made act-
division highly convenient both for the spectators and for the
management. . T h e dramatist, too, finds that he can turn
act-division to account b y employing it to denote lapse of
time. O u r modern convention is that the dramatic time
covered b y a scene does not greatly exceed the actual time it
takes to perform. But the action of the play as a whole m a y
be supposed to occupy several years. T h i s is made possible
by the existence of pauses between the acts, and the value w e
conventionally give to these pauses. T h e realistic nature
of our scene-painting and the use of programmes h a v e helped
to establish in our minds an association between the interval
and a change of place and time.
Conditions and conventions were very different when
act-division first appeared in our drama. T h e Elizabethan
theatre had neither drop-curtain nor realistic scenery nor
any of those modern devices which have made act-division so
convenient for us. T h e introduction of act-division was due
not to its intrinsic value but to respect for classical tradition,
as embodied in the precepts of Horace and Donatus and the
 ROMAN ORIGIN OF THE LAW OF FIVE ACTS igi

five-act division imposed by R o m a n editors on the plays of

Terence. T h e native English tradition was continuous
performance ; ' acts and s c e n e s t o quote Sir Edmund
Chambers (.Elizabethan Stage, iii. 199) ' which are the outward
form of a method of construction derived from the academic
analysis of Latin comedy and tragedy, make their appearance,
with other notes of neo-classical influence, in the court tragedies,
in translated plays, and in a few others belonging to the same
milieu of scholarship N o scheme of act-division other than
the five-act scheme was recognized (cf. Hunter, Review of
English Studies, vol. II, 1926, p. 295). In other words act-
division in Elizabethan drama was not a natural development ;
it was directly due to respect for R o m a n precept and example ;
and where observed in theatrical practice it necessitated
act-pauses. T h e word ' a c t ' is defined by Cotgrave (1611)
as a ' pause in a Comedy or Tragedy and we have the same
sense in a post-Shakespearian stage-direction in the Folio
of the ' Midsummer Night's Dream,' at the end of A c t I I I
—'' they sleep all the a c t e ' . These pauses had to be filled up
with interludes—dumb-shows, musical effects, or, as in Ben
Jonson's Catiline, reflective utterances of the Chorus.
It should be realized that the act-pauses were not at first
used to indicate change of scene or lapse of time. T h e native
tradition had left the dramatist free to suppose his scene
changed as often as he pleased—a freedom which Sidney
satirizes in his remarks on Gorboduc. T h e classical influence
brought into fashion not only the five-act rule but also the
unities of place and time, as we see them in Ben Jonson's
plays. Undoubtedly Shakespeare allows changes in place
and lapse of t i m e — b u t then it is not certain that Shakespeare
observed a c t — a n d scene—division. A l l that Professor Dover
Wilson will grant (Rev. of English Studies, vol. I l l , 1927, p. 395)
is ' that the company and the audience found a short break
convenient in the middle of a long play like Hamlet. But
such a break had no structural significance ; might occur at
any point in the play where the stage was left clear ; and was
a mere matter of theatrical convenience.' It would seem,
then, that Shakespeare did not use a pause in performance
to convey the sense of lapse of time—even of so long a lapse of
time as the sixteen years' interval in The Winter's Tale,
where he brings T i m e himself on the stage to give the informa-
 192 THE ROMAN STAGE

tion in plain English to the audience. O f course Shakespeare

k n e w o f the five-act law, and k n e w that some of his fellow-
dramatists observed it. I f he, then, k n o w i n g the act-pause
as a fact in the contemporary theatre, did not use it to c o n v e y
the sense of lapse of time, w e must a d m i t that there is n o
necessary connexion between the act-pause and the lapse of
d r a m a t i c time. L a t e r dramatists, while retaining the classical
act-pauses, a b a n d o n e d the classical unities ; consequently
they were able to e m p l o y the act-pauses in the w a y w i t h w h i c h
w e are familiar. B u t if Granville Barker and D o v e r Wilson
are right, it w o u l d b e a complete fallacy to use the changes
of place and time w h i c h Shakespeare allows as evidence that
he intended his plays to be divided into acts.
T h e question before us is w h e t h e r act-division was observed
b y a n y dramatist of classical antiquity ; this question cannot
be answered if w e are not clear in our o w n minds as to w h a t
w e m e a n b y ' a c t ' . T h o u g h there is some variation in m o d e r n
definitions of the term, it seems to be agreed that the act
must be followed b y a definite and perceptible pause in per-
formance. W h i l e this pause m a y be, and usually is, occupied
with an interlude of some kind, this interlude must not a d v a n c e
the action of the play : indeed the spectators must k n o w
b e y o n d d o u b t that the interlude is not an essential part of the
play. M o r e o v e r each act should be an artistic unity in itself,
forming a clearly defined part of the action of the p l a y as a
whole. I t is thus plain that ' a c t ' has a theatrical as well as a
d r a m a t i c significance. W h e n K u i p e r (Grieksche Origineelen,
p. 257) says ' w e find for the Greek Andria five distinct chapters,
corresponding to the five acts : (1) Exposition ; (2) A p p a r e n t
S u c c e s s ; (3) Frustration of Simo's scheme ; ( ^ R e c o g n i t i o n ;
(5) D e n o u e m e n t h e is evidently thinking of the ' a c t ' as a
d r a m a t i c entity, c a p a b l e of definition in literary terms. B u t
if there were no o u t w a r d and objective indication of w h e r e
each of these ' chapters ' ended and the next b e g a n , h o w could
the spectators b e expected to appreciate the artistic unity of
each act, or even to k n o w w h e n it was complete ? T h e r e f o r e ,
even on literary grounds, the act-pause is essential to the act.
T h i s is n o w h e r e denied ; indeed it is generally taken for
granted, even b y students of Greek d r a m a . T h u s M a i d m e n t
(' T h e later C o m i c C h o r u s C.Q. 1935, vol. X X I X , p. 15)
says that ' the Heauton definitely demands four pauses.' T h a t
 ROMAN ORIGIN OF THE LAW OF FIVE ACTS I93

is to say, if the Heauton were being produced at the present

day, we should feel that four pauses were appropriate. W e r e
such pauses in fact observed in the practice of the Greek or
R o m a n theatre ? I f w e cannot answer this question in the
affirmative, then w e must admit that the modern sense of the
term ' a c t 5 was u n k n o w n in the ancient world.
I doubt if any one seriously believes, or has ever believed,
that the performance of a Greek play was broken at regular
intervals by blank pauses. W e know t h a t — a t least in the
fifth c e n t u r y — i t was the custom to produce several plays in
one day. C o m p a r e d with modern standards, Greek plays
were short. A t the end of each play there was necessarily a
break, which would give to the spectators the breathing-space
which we find necessary in the modern theatre, and which is
afforded to us b y the break between acts. Within the bounds
of each individual play continuous performance was the rule.
Acted and choral portions alternated, but both were equally
part of the play. This is indeed admitted on all sides ; never-
theless an attempt has been m a d e to find the germ of act-
division present in the separation of the acted portions from
one another b y the choral odes, and the use of μέρος to denote
each separate ' p a r t ' which was performed by the actors.
Here we must guard against the common confusion between
genuine Greek and G r a e c o - R o m a n evidence. W h e n M a r c u s
Aurelius (xii. 36) speaks o f τά πέντε μέρη as necessary to
form a complete play, w e m a y presume that he is using
' p a r t ' as the nearest Greek he can find to translate the L a t i n
actus.1 T o go back to undoubtedly Greek sources, it is clear
that μέρος could m e a n for Aristotle or his interpolator (Poet.
1450a, 1452^) n o t ' a c t ' b u t ' constituent p a r t ' or ' m e m b e r ' ;
thus, quantitatively considered, the ' m e m b e r s ' were prologue,
episode, exode and choral song (the last-named including
parode and stasimon). O n the other hand there is evidence
that, w h e n it was necessary to refer the reader to a parti-
cular passage in a play, the play could be regarded as divided
into Part O n e , Part T w o and so on. Aristophanes (Frogs
1119-20) speaks of the prologue as the ' first p a r t ' of a tragedy,
and the scholiast ad loc. seems to be merely echoing his words
w h e n h e c o m m e n t s 6 yap προΧογος μέρος πρώτον της τραγωδίας.
Great importance is attached by L e o (PI. Forsch., 2nd ed.
1 Cf. Frete, La Structure des comMies de Plaute, p. g.
ig4 THE ROMAN STAGE

p. 230) to a passage in the hypothesis to the Andromache in

which the speech of Hermione is said to occur eV τω hevrepy
μέρει, ' i n the second p a r t ' . This speech actually occurs
immediately after the parodos, that is to say, at the beginning
of the first episode. As the Andromache has a prologue, it
seems clear that the prologue is here regarded as the ' first
p a r t t h e first episode as the ' second p a r t T h e ancient
Life of Aeschylus says that in the Niobe the heroine sits silent
' until the third p a r t T h e s e are, I believe, the only examples
of this use of μέρος which can be cited from the pre-Roman
period. Arrian (Teubner ed., 1. 24. 15) reports Epictetus
as saying that some king in tragedy, when things go wrong
with him, cries out ' about the third or fourth p a r t ' certain
words which actually occur in the exodos of the Oedipus Rex
(a play containing a prologue and four episodes, as well as the
exodos). Epictetus, like Marcus Aurelius, m a y have been
thinking in terms of Horace's five-act rule. W h a t the three
pre-Roman passages suggest is that, for the purposes of
reference, the choral odes were regarded as dividing the play
into parts ; thus the prologue would be Part One, the first
episode Part T w o , and so on. Weissinger (p. 16) argues that
the parodos forms part of the second μέρος, as against
Flickinger, who put it in the first μέρος, I venture to suggest
that the parodos and the stasima were not regarded as belonging
to the μέρη but as forming the boundaries between the μέρη.
(If it were desired to refer to a particular ode, a ready-made
nomenclature was available—the ' p a r o d o s t h e ' first
stasimon' and so on). Evidently Greek scholars made use
of the obvious fact that the choral odes divided the acted
portion of a play into sections, which they called ' parts But
μέρος, as used for this purpose, may not have conveyed any
of the theatrical implications associated with the word ' a c t ' .
T h e essential characteristic of act-division is the pause in
performance. T h e choral odes of fifth-century drama could
not be regarded as pauses in the performance : on the con-
trary they were a vital element in the performance. But in
time the importance of the chorus declined. In Menander's
plays it seems to have been confined to occasional appearances
in which it danced and perhaps sang ; but its songs have not
been recorded. T h e general view is that these appearances
were so unconnected with the action of the play that they had
 ROMAN ORIGIN OF THE LAW OF FIVE ACTS I95

the effect of breaking up that action into chapters, each of

which was a unity in itself. The appearance of the chorus,
in other words, was not part of the play : it was a signal
that for the time being the action was suspended. While the
chorus was dancing, the spectators were experiencing that
sense of pause which we derive from act-division. The drama-
tist composed his play with these pauses in mind : he used them
to give that sense of lapse of dramatic time which is given by the
act-pause in modern drama.
This theory may have derived support from the tradi-
tional belief in the high Hellenistic stage on which the actors
stood, twelve feet above the orchestra in which the chorus
performed ' interludes between the successive acts (Haigh,
Attic Theatre, 3rd. ed. by Pickard-Cambridge, p. 128). But
Dr. Pickard-Cambridge has now abandoned his belief in the
raised stage, even for the Hellenistic theatre (Theatre of
Dionysus, 1946, p. 165). The view that the Menandrian
chorus never took any part in the action seems to me incapable
of proof. It is generally supposed (cf. Maidment, I.e., p. 20)
that the guests who enter the house of Chaerestratus (Epitr.
195-201) are the chorus of the play, whose appearance at
this point is indicated by the direction χορού. A few lines
later Habrotonon hurries out of the house ' remonstrating
with one and another of the revellers who try to detain h e r '
(Allinson). The chorus have a relationship to one of the
characters of the play ; they enter his house as his guests ;
they accost one of the actors, and are addressed by her. How can
it be claimed that the appearance of such a chorus is not part
of the play, but is an interlude between two acts of the play ?
To suppose that the chorus could be regarded in this light
is to suppose that a revolutionary change had occurred in
the position of the chorus. Instead of being part of the play
(albeit a part of diminishing importance) it had become
extraneous to the play (and thereby assumed a new importance
in the sense that it now divided the play into separate episodes).
Yet it is generally agreed that in some Hellenistic plays, if
not in those of Menander, the chorus still took part in the
action. And the old tragedies were still produced, displaying
the chorus as an integral part of the play. How could the
contemporaries of Menander have held two radically different
views of the function of the chorus ?
196 THE ROMAN STAGE

In fact we know very little about the Menandrian chorus.

W e do not even know whether, once it had made its appearance,
it remained within view for the rest of the play, or withdrew
as soon as it had performed the dance or sung the song required
of it. Körte inclines to the second view (article on { Neue
K o m ö d i e ' in P.W. xi, 1266 ff.). This chorus of revellers is
commonly asserted to be a return to the κώμος or ' r e v e l '
out of which Old Comedy sprang. This assumption would
perhaps afford a basis for the other assumption that the
Menandrian chorus had become so external to the action that
it could be felt, and in fact was felt, to divide that action into
sections. But how does this square with the theory that N e w
Comedy is more closely akin to the later tragedy than it is to
O l d Comedy ? — a theory which is necessary if we are to link
the R o m a n references to actus-division in tragedy with the
structure of New Comedy.
T h e chorus was the original element in Greek drama, the
stuff out of which was developed the acted part (at least in
tragedy). As long as we can trace the history of Greek drama,
the chorus was retained. In Elizabethan drama, on the other
hand, the chorus was something external and alien to the
native tradition ; it was inserted by classicizing dramatists
to divide the play into acts. Is there not all the difference in
the world between the stage-direction ' Chorus : M u s i c ' in
Ben Jonson's Sejanus and the direction χοροΰ in Menander ?
Flickinger (Class. Phil., vii, 1912, page 33-4) suggests that ' in
Agathon's Mss. there were gaps indicated between the acts.
In actual performance suitable odes were thrown in (έμβόΧιμα)
T h e trouble is that we have no evidence for these ' gaps '
except the existence of the odes. Would there have been gaps
if there was no chorus ?
Nevertheless it is mainly on these supposed ' interludes '
that the case for five-act division in Menander is founded.
This should logically mean that there were four such interludes
in each play of Menander. T h a t is how Legrand understood
the matter ; the five-act law was for him a four-interlude law.
If, therefore, we had the complete text of Menander, we should
expect to find the stage-direction χορού occurring exactly four
times in every play. But the largest number of occurrences of
χοροΰ so far assigned to any one play is three, in the Epitrepontes
(Maidment, page 17). And, as Conrad points out (Technique
 ROMAN ORIGIN OF THE LAW OF FIVE ACTS I97

of continuous action in Roman Comedy, page 9) ' even if one or

several plays of Menander are discovered to be divided into
five acts, it would still remain uncertain whether all, or any
large number, of the writers of New Comedy were rigidly
bound by a law of five acts or not
The argument from the proved occurrences of χορον to
the five-act law is so obviously a non-sequitur that it is in
many cases tacitly abandoned. The editors of Menander
profess to discover act-division even where the word χορον is
not known to have occurred, or is even believed not to have
occurred. Thus Capps can say in his introduction to the
Perikeiromene (Four Plays of Menander, page 144) : ' according
to the divisions of the play adopted in this edition, the chorus
makes its appearance after the second act and does not furnish
an entertainment between the third and fourth acts.' But
if there was no interlude at the end of the third act, how would
Menander's audience have known that the third act had ended?
Allinson and Capps suppose Act I of the same play to finish
with Agnoia's speech. At this point the text is preserved
complete, and there is no trace of a χορον between this and the
following scene (cf. Maidment, page 17). On the Hero Capps
(page 6) remarks ' it is noteworthy that no chorus is mentioned
in the list' (of characters). ' Possibly the entertainment
provided between the acts was of too informal a character to be
dignified by the name " chorus In other words Capps is
prepared to make act-division quite independent of the
appearance of a chorus. Weissinger, on the other hand, bases
act-division on the use of χορού (page 54), and consequently
has to conclude (page 61) that ' not more than a possibility
exists that the plays of Menander uniformly had five acts
If we grant, as we must, that ' acts ' have no reality unless
there is some external mark of division between them, and if
we have no evidence that the chorus appeared four times in
any play of New Comedy, then we must either abandon our
belief in the five-act law or suggest some other mark of division
than a ' choral interlude Körte's article on Komödie in
P.W. suggests that on occasion the flute-player supplied an
interlude. For this there is no evidence at all in Greek drama, 1
and _л1у one example in Latin comedy, with which I shall deal
1
Aristophanes, Birds 280, gives the stage-direction (auXel), ' the note of the
flute is heard \ This is not an interlude but an integral part of the action.
14-
198 THE ROMAN STAGE

presently. 1 Are we to understand that the flute-player supplied

all the interludes in some plays, the chorus all the interludes
in other plays, or that there were plays in which the interludes
were supplied partly by the chorus, partly by the flute-player—
but that, however achieved, the interludes always amounted to
four ? Such suppositions may be justified if there is independ-
ent evidence of a five-act division in New Comedy ; but
belief in such a division cannot properly be based on the known
occurrences of χοροΰ in Menander.
This is where the Nauplios puppet-show, with its five
episodes, seems particularly relevant. Prou's learning and
enthusiasm convinced Legrand, Leo and others that here at
last we have a Greek drama with five acts and four act-pauses.
Hero's interest in the puppet-show is purely that of an engineer.
Having dealt with the ν-πάηοντα αυτόματα, ' movable puppet-
shows ' which were mounted on a movable platform, lay
permanently open to view, and consequently could present
only a single scene, he turns to the στατα αυτόματα, or
stationary puppet-show. This seems to have been a box
mounted on a pillar. The box had folding doors, which, when
opened, revealed movable figures seen against a painted
background. I give a somewhat abbreviated version of Hero's
description (Prou, page 207 ff.) :
The problem is to make the puppet-show (pinax)
open of its own accord and reveal the figures inside in
motion befitting the theme of the story ; then the pinax is to
close automatically and after a short interval to open again,
revealing other figures, some or all of which are if possible
to be in movement. This process is to be repeated several
times (πXeovaKKi). The arrangement employed by the early
designers was simple. When the pinax was opened, there
appeared in it a painted head. This moved its eyes, raising
and lowering them repeatedly. The pinax was shut and then
opened again ; the head had disappeared, but painted figures
were seen arranged in accordance with some story (et? τινα
μνθον Βιεσκβυασμβνα). The pinax would shut and open once
more, revealing another arrangement (Ьшвесьч) of figures to
complete the tale. There were thus only three different
movements : that of the doors, that of the eyes and that of the
curtains.
1 pp. 204-5
 ROMAN ORIGIN OF THE LAW OF FIVE ACTS I99

But in our time designers have introduced interesting

tales into their puppet-shows, and have made use of many and
varied movements. As I promised, I will describe one show
which seemed to me to be the best. The story set forth in it
was that of Nauplios. This is how it was divided up (τά Se
κατά, μέρος el%ev όντως) :
A t the beginning the pinax opened and displayed twelve
painted figures, arranged in three rows. These represented
some of the Greeks preparing their ships and getting ready
to launch them. These figures moved, some sawing, some
chopping. . . . They made a loud noise, as in real life.
After a sufficient time had elapsed, the doors shut and opened
again, and there was a new arrangement (Βιάθεσις). It
showed the ships being launched by the Greeks. T h e doors
again shut and opened, and nothing was visible in the pinax
except sky and sea. Presently the ships were seen sailing
by. . . . Again the pinax shut and opened. There were now
no ships to be seen, but Nauplios was there brandishing his
torch and Athene standing beside him. A flame burned above
the pinax. . . . Again it shut and opened, revealing the wreck
of the fleet and A j a x swimming in the sea. A mechanism in
the top of the pinax was raised, there was a peal of thunder,
a lightning flash fell on Ajax, and his figure vanished. T h e
pinax closed, and the story was ended.
Prou's treatment of this passage is coloured by his view
that the puppet-show was a form of drama. Weil (Journal
des Savants, July 1882, p. 418) pointed out that only the fourth
scene was suited for dramatic treatment. ' Les conditions du
spectacle dramatique et du spectacle mecanique sont trop
difftrentes.' I must add that the terms employed by Hero
are not reminiscent of the stage. He speaks of a ττίναξ, not a
θέατρον; he calls each separate scene a διάθεσις or ' arrange-
m e n t L e o (PI. Forsch, p. 230), claims that Hero uses
μέρος in the sense ' a c t ' — q u o t i n g the words 'ίνα μηδέν των
ητροΐίρημένων . . . μέρος той πίνακος φαίνηται. T h e passage
in question (149 β : Prou, p. 239) deals with the need for
care that the mechanism described shall not ' be visible in the
front part of the pinax '- 1
I can find in Hero's account only one reference to the
1 Was Leo thinking of the occurrence of μέρος in the passage quoted above
(rh. Ы . . .) ?
200 THE ROMAN STAGE

theatre : w h e n he comes to suggest h o w the effect of thunder

m a y be given (141 γ : Prou, p. 209) he describes h o w similar
effects are p r o d u c e d ev rot? θεάτρου.
T h a t there w e r e five scenes in this particular show seems to
be fortuitous. T h e earlier show referred to h a d three scenes,
and n o w h e r e in this connexion does H e r o stress or indeed
mention the n u m b e r five. T h e closing and opening of the doors
and the episodic nature of the show w e r e essential if a story
was to be told at all in this m e d i u m . I n fact, so far f r o m the
Nauplios-show supporting the case for the five-act l a w in
d r a m a , the only link w i t h d r a m a w h i c h W e i l could find in it
was precisely the fact that it has five parts. T h i s w o u l d b e
interesting if there was a n y independent evidence that the
n u m b e r five played some part in the construction of the d r a m a
of Hero's d a y . B u t no such evidence exists. A n d if w e c a n
base any opinion on the earlier development of d r a m a , the
tendency was to m o v e a w a y f r o m five as the n o r m a l n u m b e r
of μέρη. I quote Weissinger (without accepting all his figures) :
' I t is true that all of Aeschylus' extant plays h a v e five such
sections except the t w o w h i c h h a v e no prologues. B u t in
Sophocles only one (Philoctetes) has five, three h a v e six a n d three
h a v e seven. I n Euripides only one (Troades) has five, fifteen
h a v e six, and one (Medea) has seven '.
Nevertheless it is assumed b y most scholars that the five-
act l a w found expression in Hellenistic theory : that this was
p r o b a b l y one of the things w h i c h H o r a c e derived f r o m
Neoptolemus of P a r i u m and included in the Ars Poetica
(congessit praecepta Neoptolemi τοΰ Παριανού de arte poetica поп
quidem omnia sed eminentissima, as Porphyrio tells us). A s s u m i n g
that this is so, h o w do w e suppose Neoptolemus to h a v e f r a m e d
the rule ? N o t , surely, that all plays had five μέρη ; for the
n u m b e r of μέρη depended not on theory b u t on the a c t u a l
n u m b e r of choral odes a n d on the presence or absence of a
prologue. T h a t all plays ought to h a v e five μέρη ? I n that
case Neoptolemus was perhaps u p h o l d i n g the practice of
Aeschylus as against that of Euripides a n d Sophocles. This
is hypothetical, but it w o u l d be even more hypothetical to
suppose that Neoptolemus was referring to the practice of
dramatists of his d a y , a n d that they h a d returned in this respect
to the arrangement favoured b y Aeschylus. W h a t w e h a v e
failed to find is a n y Greek evidence that the Greeks recognized
 ROMAN ORIGIN OF THE LAW OF FIVE ACTS 201

the five-act law, or act-division of any kind, either in practice or

in theory. T h e case for the law rests on Latin evidence alone.
For the practice of the L a t i n dramatists we have the evidence
of the plays. W e have also some conflicting statements b y the
grammarians. Evanthius tells us (Donatus, T e u b n e r ed., i. 18)
that the Latin comic dramatists ' did not even leave a place
for the chorus ', thus obscuring the act-division, as he adds.
Donatus (i. 266) says that T e r e n c e ran all five acts into one
(uult poeta noster omnes quinque actus uelut unum fieri). Further-
more, Donatus' rule for restoring the act-divisions (i. 38) is
to look for empty-stage intervals at which it is possible to assume
a performance b y the chorus or the flute-player ; and as he
complains of the difficulty of this procedure, it seems that the
Latin text offered him no clear guidance. T h e implication is
that Donatus' text of Terence contained no mention of choral
odes or interludes. So too Diomedes (iii. 14) : Latinae igitur
comoediae chorum поп habent. (Kaibel, Com. Graec. fr. i. 61).
But in the Liber Glossarum (Kaibel, Com. Graec. fr. i. 72) w e
read aput Romanos quoque Plautus comoediae choros exemplo
Graecorum inseruit. {Corp. gloss, v. 181. 7, de com. Graec. 72 K . ) .
It is difficult to suppose that the writer of this remark had seen
a text of Plautus containing stage-directions corresponding to
χοροί). Consequently it is generally supposed that the reference
is to such passages as the ' song ' of the fishermen in the Rudens
(290-305), or the presence of the aduocati in the Poenulus
( 5 0 4 - 8 1 6 ) — b o t h presumably taken over from the Greek
original. Schanz (Rom. Lit.-Gesch. i. 131) further refers us
to Bacch. 107, Heaut. 170, where there is no mention of a chorus
in our texts, but traces are supposed to remain of a chorus in
the original. Others refer us to such purely R o m a n insertions as
Cure. 462-486 (description of R o m e b y the property-manager).
T h e O x f o r d editors of T e r e n c e actually insert the stage-
direction (Saltatio Conuiuarum) after Heaut. 170, referring us
in a footnote to Skutsch's article in Hermes, vol. xlvii (1912).
T h e suggestion put forward b y Skutsch (and independently
by Flickinger, Class. Phil. vii. 24-34, a l s o i n ! 9 1 2 ) was that in
Menander's play the chorus had entered at this point. This
suggestion seems to raise more difficulties than it solves
(cf. J a c h m a n n , Plautinisches und Attisches, pp. 245 ff., and
Drexler, Hermes lxxiii (1938) p. 65 ff.) 1 ; but in any case it
1 See Hermathena lxxiv (1949) pp. 26-38.
202 THE ROMAN STAGE

referred to Menander's play, not to Terence's adaptation,

and for the Oxford editors of Terence to insert a choral interlude
into the middle of a scene is a liberty which calls forth a protest
from Jachmann (p. 245, note 1) and Weissinger (p. 64).
There may conceivably be traces in the Latin plays of the
appearance of the chorus in the Greek originals. But if act-
division can ever be proved on internal arguments alone, then
any such arguments based on the structure of the Latin play
are presumably valid in the first place for the performance of
the Latin play and only indirectly for the performance of the
Greek play. T h e suggestion has been made that we should
postulate intervals and interludes to obviate awkward en-
counters between characters leaving and characters entering
the stage. Several examples are cited by Miss Johnston (Exits
and Enframes in Roman Comedy, 1933, pages 106-119), who
suggests that a brief interlude by the flute-player might allow
one actor to get away before another entered. It may be so,
though, as Harsh remarks (C.W. 1935, page 163), ' I t is not
an assumption which can definitely be proved c o r r e c t M i s s
Clifford, on the other hand (Dramatic Technique and the
Originality of Terence, C.J. 26, pages 605-618), regards these
awkward encounters in the wings as proof of the clumsiness
with which Terence adapted his originals. But surely we may
suppose that the stage was on occasion left vacant for a few
seconds, in order to allow one character to depart before another
entered, or to give a solitary actor time to enter a house and
return. (Harsh cites Aul. 627, As. 809, and from Greek drama
Eccles. 513 and Ale. 861).
Far more important is the view that intervals must be
assumed to allow for action off stage. In default of external
evidence the case for intervals can be founded only on the
text of the plays and our own sense of what is fitting. T h e
question arises whether we can assume that the theatrical
convention of the Greeks and Romans was in this respect like
our own. In modern practice, while there is no limit on the
dramatic time occupied by the play as a whole, the dramatic
time covered by each separate scene, as has already been said,
does not greatly exceed the time which the scene takes to
perform. T h e difference between the total time of action and
the two or three hours of performance is covered by the
intervals, which can represent any desired period. In ancient
 ROMAN ORIGIN OF THE LAW OF FIVE ACTS 203

drama the total dramatic time (if we may use so modern a

phrase) was as a rule limited to one day ; but the dramatist
was free within that limit to give any time-value he liked to
any scene or choral ode. There was, therefore, no dramatic
need for intervals of the kind familiar to us ; and indeed it is
generally held that in fifth-century drama at least there were
no such intervals. Some of the most striking lapses of time,
and all the changes of place, occur when the chorus is not
present. In Eumenides 234 the chorus of Furies is driven off the
stage ; in the very next line we find that we have been trans-
ported from Delphi to Athens. Even if we assume a pause in
the performance here, it cannot have been occupied by a
choral ode ; for the chorus are absent. T h e parabasis of the
Lysistrata ends at line 705 ; between 705 and 706 five days are
supposed to elapse (cf. line 881). Here Weissinger actually
suggests that the chorus was removed for some minutes ' to
make the break more decisive through the external device of a
completely empty stage ' (page 38). A t all events it must be
admitted that the gap in time here is not covered by a choral
interlude ; but few will acept Weissinger's suggestion that the
chorus left the theatre merely to give the sense that time was
elapsing.
Nevertheless it is assumed that in New Comedy pauses in
performances were employed to give the effect of a lapse of
time. Let us at least be clear whether we are discussing the
Latin plays or their Greek originals, and whether we are
postulating blank intervals or interludes. For blank intervals
in the performance of a play by Menander there is no evidence ;
for interludes there is the evidence of the χορού passages, which
is usually regarded as decisive. Even Conrad (page 71) after
making out a strong case for continuous performance in the
case of the Latin plays, can say ' there still remain a small
minority of vacant stages which everybody must admit furnish
reasonable grounds for the view that there were some essential
pauses in the action, at least in the Greek original, and
possibly even in the Latin adaptation. T h e most striking are a
few vacant stages which separate the withdrawal and re-
appearance of the same character or group of characters.
Now it is always proper to use this small number of cases as
evidence of χορού in the Greek original'.
T h e argument appears to be that the Greeks would have
204 THE ROMAN STAGE

found continuous performance intolerable in these cases,

although the Latin translator may not have felt the difficulty.
But if continuous performance can be assumed in Eum. 234-5
and similar cases, and if the R o m a n audience would tolerate
it in the cases Conrad has in mind, do we know enough about
Menander to be certain that he would have inserted χοροί5 in
those cases ? T h e argument assumes, too, that in these passages
the R o m a n translator made no other change than to omit
χορον. If he introduced other alterations, then we cannot use
his text as evidence for the structure of the Greek play.
W e are on firmer ground when we use the Latin plays as
evidence for the practice of the Roman theatre. Were there
blank intervals, deliberately employed for this purpose ? T h e
most striking instance of lapse of time is Heaut. 409-10. A t
409 the guests go indoors for supper. A t 410 Menedemus
appears from next door and remarks ' the dawn is breaking
It is difficult to assume that Menander had here brought on
a chorus of revellers going to the party ; for the party is
already complete. (Flickinger supposes a chorus of ancillae
here and a chorus of conuiuae at 171 (C.P. vii. 28) ). As for
the Latin play, does it help to assume a blank interval of
perhaps five minutes, or a solo by the flute-player ? Let us
consider the prologues of Plautus and Terence, so different in
style, and yet so alike in their anxiety to induce the spectators
to give the play a hearing. Is it likely that once the play was
under way the Latin dramatist would deliberately bring the
performance to a dead stop, merely to convey the sense of
lapse of time ? Let us consider, in particular, the brief
prologue to the Pseudolus : ' Y o u had better get up and stretch
your loins ; a long play by Plautus is coming on the stage ', 1
Surely this implies (a) that play followed play, (b) that once
a play had begun the spectators would have no break until it
was ended. T o assume that the flute-player was required to
give a solo at such passages as Heaut. 409-10 is to assume
(a) that the dramatist brought the action to a stop ; (b) that,
realizing the danger of such a course, he then tried to bridge
the gap by a flute-solo. Is it not simpler to suppose that there
was no pause ?
T h e only known example of a flute solo in Latin drama
(Ps. 573a) is the exception which proves the rule. Pseudolus,
1 So, too, the last line of the Epidicus.
 ROMAN ORIGIN OF THE LAW OF FIVE ACTS 205

left alone on the stage, suddenly announces that he will enter

the house to think out a plan ; he promises the spectators
that he will not be long absent, and that in the meantime the
flute-player will entertain them. H e then, evidently, goes
indoors, and presumably the flute-player does strike up an air,
for when Pseudolus reappears at the next line of the play
the metre has changed from the rhythm of speech to that of
' song
Here, beyond doubt, is a break in the action of the play.
It is occupied by an interlude. But that interlude does not
indicate the lapse of any dramatic time. Nothing happens
while Pseudolus is off the stage 1 ; he claims indeed to have
formed his plan, but we never hear what the plan is, and a
new turn of events makes it unnecessary. M y explanation
of this unique passage is founded on the quite exceptional
circumstances. Pseudolus' role is very heavy. T h e dramatist
wishes to give him a break in the longest continuous spell of
duty to be found in Latin comedy (lines 1-766). Therefore,
as soon as Pseudolus is alone on the stage, he makes a trans-
parent excuse and retires, assuring the audience that he will
soon return and that they will not be without entertainment
during his absence. Plainly the dramatist thought that this
expedient, however necessary here, involved some risk, and
that may be why we find no other instance of it in our extant
plays.
In default of other evidence, act-division is based on the
occurrence of empty-stage breaks. Unfortunately their
number varies from two (in the Mostellaria and Andria) to ten
(in the Rudens and Adelphi). If act-division is to be identified
with empty-stage break, then only two or three of Plautus'
plays and perhaps none of Terence's obey the five-act rule.
Attempts are often made to distinguish the breaks which con-
stitute act-division from those which do not, the usual criterion
being the lapse of dramatic time. Even if the critics were
agreed on this matter (and they are not), it is difficult to see
how they can introduce five-act division into plays which
have less than four breaks. In truth the evidence of the Latin
plays (in the form in which they have come down to us) is

1 Freti (Structure dramatique des Comedies de Piaute, p. 12) has noticed that the
play gains nothing from Pseudolus' absence (rien n'empechait Pseudolus de rester
en setae pour imaginer sa fourberie).
2 θ6 THE ROMAN STAGE

overwhelmingly against the validity of the five-act law so

far as the Roman theatre is concerned.
The fact remains that the Romans bequeathed to us the
five-act law. T o be more exact, Horace said that a play
should have five actus. The word actus means ' performance '
by the adores. When Terence says {Нес. 39) primo actu
placeo, he means ' the beginning of the performance was
successful'. (Similarly, in Ad. 9, in prima fabula means ' in
the beginning of the p l a y ' . ) But if the performance of the
actors should be broken up into sections by something not
performed by the actors, then each of these sections would
itself be an actus. Accordingly Horace, when instructing
Roman literary men on the composition of choral tragedy,
says that if a play is to be successful it must have five acts
and no more than five. In other words Horace is recom-
mending that a play should be neither too short nor too long ;
a prologue, three episodes and an exodos, separated by choral
odes relevant to the plot, will be about right.
Horace's principle of act-division was the choral ode :
the chorus, he says (A.P. 194-5) is n o t t o s i n g ' between the
acts ' except on themes relevant to the plot. It is plain that
Horace is not concerned with how to divide up a non-choral
play into acts ; still less is he to be taken as affirming that
every play ever written necessarily consists of five acts.
A passage of Cicero (Q.F. 1. χ л 6.46) has sometimes been
taken to imply that he regarded the third act as the climax
of a play. ' Let this third year of your command,' he says,
' like a third act, seem to have been the most finished and
splendid of all.' I think that no emphasis is to be placed
here on the word ' third ' where it is applied to ' a c t ' : if we
translate 1 let this third year—a third act as it were—be the
best. . .' we interpret his meaning correctly. 1 (Cicero speaks
elsewhere of the ' fourth a c t ' of Verres' wickedness, Verr. ii.
16. 18.)
One of Cicero's friends, the antiquarian, Varro, was
especially interested in act-division. According to Donatus
(ii. p. 192) Varro said that we must not be surprised to find
the acts in the Hecyra and other plays unequal in the number
of lines and scenes, since act-division depended on subject-
matter, not on equality of length, and that this was true not
1
Cf. Freti, op. cit., p. 9.
 ROMAN ORIGIN OF THE LAW OF FIVE ACTS 207

only of the Latin writers but of the Greeks themselves. This

passage comes immediately after Donatus' proposed division
of the Hecyra into five acts. Let us grant for argument's sake
that this five-act division of Terence's plays goes back to Varro.
It certainly does not go back to Greek sources ; what Greek
would have concerned himself with Terence ? Varro, w h o
died in 27 B.C., cannot have read the Ars Poetica ; but Horace
may have read V a r r o or Varro's R o m a n authority. Two
things are involved : the practical rule that five acts are the
right amount for a play, and the academic theory that any
play, even if non-choral, can be analysed into five acts. It is
just conceivable, as has been remarked above, that Neo-
ptolemus of Parium had said that the best Greek plays consisted
of five μερη, but it is inconceivable that he had said that all
plays consisted of five μέρη, if he meant by that a prologue,
three episodes and an exodos, for such a statement would be
contrary to known fact.
T h e attempt to analyse each play of Terence into five
actus was evidently the work not of Greek but of R o m a n
scholars. There is an obvious contradiction involved in the
attempt to divide a play which is admitted to be non-choral
into parts the separate existence of which depends on choral
interludes. Indeed Evanthius tells us that the omission of the
choral interludes by the Latin comic dramatists has made it
difficult to discern the five-act division in their plays (Donatus
i. 18 W . : postremo ne locum quidem reliquerunt, quod Latini fecerunt
cornici, unde apud illos dirimere actus quinquepartitos difficile est).
Elsewhere (i. 38) a suggestion is made for solving this problem :
we must look out for empty-stage intervals at which an inter-
lude by the chorus or the flute-player may be postulated. I n
another passage (i. 266) Donatus explains that Terence ran
all five acts of the Eunuchus into one for fear the bored spectators
might take the opportunity afforded by act-intervals to leave
the theatre. Nevertheless the Eunuchus, like other plays,
must have its five acts, which the Greek dramatists divided by
means of the chorus (hoc etiam ut cetera huiusmodi poemata quin-
que actus habeat necesse est choris diuisos a Graecis poetis).
It is on these passages that the case for the five-act law
depends. T h e argument may be put thus : the Greek drama-
tists observed five-act division, depending on choral interludes.
Terence, translating these five-act choral Greek plays, omitted
2o8 THE ROMAN STAGE

the chorus and so obscured the act-division—that is to say,

the act-division of the Greek originals, which in them was
clearly indicated by choral songs, or the stage direction χοροΰ.
The argument implies that, apart from omitting these choral
odes or stage-directions, Terence kept the structure of the Greek
original as he found i t ; for of course if he had altered the
structure of his Greek original, Donatus would have had no
right to introduce the act-division of the Greek original into
Terence's adaptation. It implies too, that the Roman scholars
had the Greek originals in front of them and were able to
compare them with Terence's adaptations. (We have other
evidence that this was the case—Donatus often refers to the
words of the Greek original). 1
But if the only difference introduced by Terence was to
omit the choral songs or directions for interludes which were
clearly indicated in the text of the Greek plays, why was the
task of restoring the five-act division of these Greek plays so
difficult for the Roman scholars of a later day ? All they had
to do, ex hypothesi, was to note where the indications of act-
division occurred in the Greek text and to insert act-division
at corresponding places in the Latin text. They did nothing of
the kind. The suggestions that Donatus gives for establishing
act-division and the difficulty which he and his predecessors
admittedly found in establishing it alike imply that comparison
of the Latin with the Greek text afforded no solution of the
problem. Indeed Donatus nowhere says that the Greek
originals of Terence were divided into five acts.2 All that is
clear from his remarks is that Roman scholars found ' actus ' in
Greek plays ; and what they meant by ' actus' is clear from
the words choris diuisos a Graecis poetis. ' Old Comedy',
according to Evanthius (i. 18) ' was at first purely choral ;
gradually actors came to be added and five-act division was
the result. Then the gradual disappearance of the chorus
resulted in New Comedy, in which not only is there no regular
chorus, but there is no place left for choral interludes. For
after spectators began to be difficult to please and formed the
habit of leaving the theatre when the actors withdrew and left
1 This seems to invalidate Professor Webster's suggestion to me that Donatus

relied on summaries of Menander such as the TTfpio^tu of Homerus Sellios, on

which see Körte, Berl. Phil. Woch. xxxviii. 787 f.
2 Apuleius' reference (Florida 16) to the third act of a play of Philemon seems

—like the whole anecdote—to be a Roman invention.

 ROMAN ORIGIN OF THE LAW OF FIVE ACTS 20g

the singers to continue the play, the poets were compelled first
to remove the chorus, merely leaving room for them (in the
text). This was the practice of M e n a n d e r , a practice which is
to be explained as has been indicated, and not on the hypo-
theses put forward b y other writers. Finally the dramatists
did not even leave room (in their text) for the chorus, and this
was the method followed b y the L a t i n comic dramatists, so
that it is difficult to recognize the five-act division in their
plays.'
Considered as an account of the development of Greek
drama, the remarks of Evanthius and Donatus involve
absurdities which, as M i c h a u t says (I. 191), would do honour
to a professional humorist. Another contradiction is presented
when w e come to the theory that comedy (excluding the
prologue) consists of protasis, epitasis and catastrophe, of
which protasis is defined as primus actus initiumque dramatis
(i. 22, cf. i. 27). T h e scholars of the Renaissance did their
best to reconcile the two theories, five-act and three-part, but
I agree with L e o (PI. Forsch, p. 232) that the two are mutually
exclusive. T o j u d g e b y the terminology, the division into
protasis, epitasis and catastrophe is Greek. T h e five-act theory
seems to be nothing but the product of R o m a n pedantry,
based ultimately on Horace's practical hint—possibly itself
derived from some remark of V a r r o . I f w e are to look for a
Greek origin for Horace's law, it can only have been some
statement such as ' the number of episodes in a tragedy should
be three ' (or ' the number of choral songs should be four ') ;
and if any Greek writer ever framed such a rule (a supposition
for which there is no j o t of evidence), he can only have justified
it by adding some such words as ' in accordance with the
practice of Aeschylus, not of Euripides
Continuous performance, w h i c h is the negation of act-
division, was the rule for ancient drama from Aeschylus to
Terence. It is usual to cite Seneca as a dramatist w h o obeyed
Horace's law. I f b y this is meant merely that the number of
choral songs in a play of Seneca is commonly four, the argument
is perhaps sound ; but it must be pointed out that in five cases
he has merely taken over the structure of his model. In the
Octavia there are five odes, and in the Phoenissae there are none.
T h e very most that can be said of Seneca is that he shows a
marked preference for four as the number of the choral odes in a
2IO THE ROMAN STAGE
-
play. He does not number off the divisions of the play as
' Act One,' ' Act T w o a n y more than did the Greek
dramatists. His odes are often relevant to the plot, and cannot
be regarded merely as entertainment designed to fill an interval.
But what puts Seneca out of court is that in all probability his
plays were designed not for performance but for reading by
a single reciter. No doubt the structure of his plays had much
to do with the introduction of the five-act law into modern
stage-usage ; but Seneca himself was not concerned with
stage-usage. There was no act-division, there were no acts,
in his plays, because those plays were not meant to be acted.
 CHAPTER XXVI

M U S I C AND M E T R E

T H E C O M E D I E S of Plautus have been compared to such

works as The Beggar's Opera or the light operas of Gilbert
and Sullivan. 1 Such comparisons, however stimulating, are
apt to mislead. Latin comedy was indeed written in various
metres, most of which were intended to be accompanied by the
music of the pipes (tibiae) ; the actors evidently illustrated
their words with gesture and sometimes with dance ; but the
essential element of all opera is song. T o us the difference
between song and speech is obvious ; but it is by no means
clear that this difference was recognized by the Romans or
the Greeks.
O u r ancient sources of evidence, however discordant in
other ways, agree that a Latin comedy (excluding the prologue)
consisted of two elements, diuerbium (or deuerbium) and
canticum ; and that these two taken together make u p the
whole of a play except the choral part—which, we are told
in the same connexion, 2 did not exist in Latin comedy.
Naturally we assume that canticum means ' song ', and con-
sequently that diuerbium means ' speech '. This is the obvious
sense to attach to the terms as used by Livy in his famous
account (VII. ii) of how, when Andronicus' voice gave way as
the result of taking too many encores, he employed a boy to
sing, standing in front of the piper, while he himself, freed from
the necessity of using his voice, was able to concentrate on
miming the canticum with appropriate gestures. This was,
according to Livy, the origin of the custom whereby the actors
merely mimed the sung part (ad m a n u m cantari) and confined
their vocal powers to the delivery of the diuerbia. Perhaps
Livy is here thinking of some form of theatrical performance
of his own day in which there was a clear distinction between
1
See Lindsay, Early Latin Verse, p. 263, and ' Plautus and the Beggar's Opera '
C.R. 37 (1923), P· 67·
' Diomedes, G . L . K , i. 491, 84.
211
212 THE ROMAN STAGE

the roles of the actor and the vocalist or chorus ; we know, for
example, that the performance of the pantomime was assisted
by vocalists (Lucian De Salt. 64). It may be, indeed, that the
attempt to find a historical explanation of this is the origin of
the anecdote concerning Andronicus. A t all events it seems
clear that L i v y regarded the canticum as something which
required a special vocal effort, greater than what was needed
for the diuerbium. Diomedes, however, tells u s 1 that the
diuerbia are those parts of comedy in which two or more
persons converse, whereas in a canticum only one person must
be present (or, if there is a second character, he must merely
overhear the other and must not converse with him). In other
words, for Diomedes canticum means ' monologue diuerbium
' dialogue ' ; which is a perfectly clear distinction, but quite
different from that between 4 song ' and ' speech '. Elsewhere 2
Diomedes recognizes the connexion of canticum with cano ;
contrasting Greek with Latin plays {drama, ' a c t i o n w i t h
fabula, from/απ, ' speak '), he a d d s ' for in Latin plays there are
more cantica which are s u n g ' (plura sunt cantica quae
canuntur). If he means by this ' more cantica than there are
in the Greek plays it is not easy to follow him ; the cantica,
so far as we can tell, were a R o m a n innovation. 3 But we cannot
reconcile Livy and Diomedes by supposing that all monologues
were sung, all dialogues spoken ; for there can be no doubt
that many monologues were spoken.
A t the head of certain scenes in some of the manuscripts of
Plautus occur the letters D V or C. W e are told by Donatus
(.Adelphi, praef. 1. 7) that D V , placed after the names of the
characters at the head of a scene, represents diuerbium, and
it may be inferred that С stands for canticum. In general, D V
appears over scenes written in iambic senarii (whether dia-
logues or monologues), С over scenes in other metres (whether
monologues or dialogues). 4 W e are told {de com. p. 30 W.)
that the actors spoke the diuerbia, and that the cantica were
set to music composed not by the poet but by a musician
(diuerbia histriones pronuntiabant, cantica uero temperabantur
1 G.L.K, i. 491, 24. 2 ib. 490.

s Whatever the origin of the metres—whether in Greek tragedy, or in New

Comedy, or in Alexandrian or Italiot mime—the distinctive feature of the Latin
cantica is the use of metrical variety to produce dramatic effects.
* Two scenes in senarii—Pers. IV. vi, Trin. IV. iv—seem to have the mark С ;
t W o—possibly three—scenes in trochaic septenarii—Capt. I l l i, Cos. IV. iii and
perhaps Epid· I. ii—are apparently marked D V .
 MUSIC AND METRE 213
modis non a poeta sed a peri to artis musicae factis). Donatus
(Adelph. praef. p. 4 W.) says that some cantica were marked
M M G (presumably mutatis modis canticum) because they were
sung with frequent changes of music (saepe mutatis per
scaenam modis cantata). In his prefaces to the other plays
he echoes these remarks, but without adding any very precise
information ; e.g. ad Phorm. p. 346 W.—diuerbiis quoque
facetissimis et gestum desiderantibus scaenicum et suauissimis
ornata canticis fuit. These verbal echoes, this lack of clarity,
may tempt us to doubt whether Donatus or Diomedes attached
any definite meaning to the words they were using.
W h e n we examine the metres used by Plautus, we find
great variety. Less than half of his work is in iambic senarii ;
a great part of the remainder is in trochaic septenarii and
iambic septenarii and octonarii, while the numerous other
metres (trochaic octonarii, anapaests, cretics, bacchiacs, etc.)
are grouped together in all sorts of ways, though no one of them
is kept up for long. In the whole of Plautus there are about
1300 iambic septenarii and 400 iambic octonarii ; trochaic
octonarii are few and scattered, while trochaic septenarii run
into thousands. It is generally agreed that the passages in
senarii were diuerbia, intended to be spoken, and that the
cretics, bacchiacs, etc., were cantica and were meant for song.
But what of the long scenes in septenarii and iambic octonarii ?
It has been suggested that they were delivered in something
between ordinary speech and song, a kind of recitative or chant
to musical accompaniment. It is even claimed that we know
the Greek name for this form of utterance—7ταρακαταλογη 1.
By means of somewhat tenuous arguments based partly on
Greek, partly on Latin evidence, we arrive at the conclusion
that there were three kinds of dramatic non-choral utterance
used alike on the Greek and on the R o m a n stage—speech, song
and recitative. But the R o m a n writers speak of only t w o —
diuerbium and canticum. Was recitative included by them
under diuerbium or under canticum ? Were there two kinds
of diuerbium—speech and recitative—or two kinds of canticum
—recitative and song ? Was M M C the special mark of the
true songs—and if so, must a true song have 'frequent changes of
music' and presumably of metre? O r was one of the three forms
of delivery left without a Latin n a m e — a n d , if so, which one ?
1 See Haigh, / l4< Theatre, pp. 266-чI.
15
214 THE ROMAN STAGE

It is generally agreed that the passages in iambic senarii

were meant for utterance without accompaniment, and that
all the other metres were meant to be accompanied b y the
pipes. T h e senarius is no doubt modelled (with a difference)
on the Greek iambic trimeter, the normal vehicle of dialogue
in Greek d r a m a . Aristotle (Poet, iv) tells us that of all metres
the iambic is the best adapted to represent speech (Χΐκτικόν),
and that it was therefore employed w h e n speech or dialogue
(Xefi?) was introduced into tragedy. H e also says (ib. vi)
that some effects in drama are produced b y metre alone,
others b y music (μέΧος) ; and w e infer that there was at least
one metre without musical accompaniment. L u c i a n (De Salt.
27) satirizes an extravagant tragic actor w h o at times ' goes
about singing the iambics ', w h i c h seems to imply that they
were not meant to be sung. Cicero (Orat. lv. 183-4) speaks
of the senarii of the Latin comic poets as scarcely distinguishable
from ordinary prose ; and as he refers in the same passage to
bacchiac verses w h i c h are accompanied b y the pipes, the
implication is that the iambic verses are not so accompanied.
W e must, of course, b e careful not to confuse Greek and R o m a n
practice, or early and late evidence. But with regard to the
practice of the R o m a n stage in Plautus' day we have a striking
piece of evidence in Stichus 758 ff. U p to this point the metre
of the s c e n e — a merry slaves' drinking-party—has been trochaic
septenarius. A t 758 the piper is invited to pluck his pipes from
his mouth and have a drink. T o allow for his drinking, the
metre changes for seven lines to iambic senarius. T h e n he is
told to p u f f out his cheeks and play a new tune in return for his
draught of old wine. As he begins to play, the metre changes
from senarii to an iambic octonarius followed b y iambic
septenarii, etc. E v i d e n t l y 1 the actsrs dance :

Sang, qui Ionicus aut cinaedicust, qui hoc tale facere possiet ?
(iamb, oct.)
St. si istoc me uorsu uiceris, alio me prouocato.
(iamb, sept.)
Sang, fac tu hoc modo.
St. at tu hoc modo.
Sang. babae !
St. tatae !
Sang. papae !
St. pax ! (iamb, sept.)
1 Cf. E.L.V. p. 262.
 MUSIG AND METRE 215
Sang, nunc pariter ambo. omnis uoco cinaedos contra.
(uers. Reizianus)
satis esse nobis non magis potis quam fungo imber.
(uers. Reiz.)
St. intro hinc abeamus nunciam : saltatum satis pro uinost.
(iamb, sept.)
uos, spectatores, plaudite atque ite ad uos comissatum.
(iamb, sept.)
W e can almost see the actors dancing as they utter these
lines. Evidently the septenarii, like all the other metres except
the senarii, were accompanied by the pipes. Moreover we
have the express statement of Cicero (Tusc. I. xliv. 107) that
the actor was accompanied by the pipe when delivering a
passage of trochaic septenarii (cum tarn bonos septenarios
fundat ad tibiam). T h e lines quoted are
neu relliquias sic meas siris, denudatis ossibus,
taetra sanie delibutas foede diuexarier.
T h e question still remains : was there a difference in the
method of utterance between the senarii and the other metres ?
Furthermore, was there a difference between the way in which
some metres were uttered to musical accompaniment and the
way in which other metres were uttered to musical accompani-
ment ? When Cicero (Tusc. I. xliv. 106), after quoting some
iambic octonarii, says haec pressis et flebilibus modis, qui totis
theatris maestitiam inferant, concinuntur, ' these lines are sung
to so subdued and mournful a melody as to draw tears from the
whole audience', does he mean that they were sung as we
understand song ? A n d , if so, do these octonarii differ in the
method of utterance from any other of the metres intended for
accompaniment ?
O f R o m a n music we know very little. W e have, however,
some information about the music of the Greeks, who no doubt
influenced the Romans in this as in the other arts. As Monro
points out (The Modes of Ancient Greek Music, 1894, p. 113),
' several indications combine to make it probable that singing
and speaking were not so widely separated from each other in
Greek as in the modern languages with which we are most
familiar.' As Greek was in the fifth century B.C. spoken with
a predominantly pitch or musical accent, Greek speech was
more like music than ours. Aristoxenus, a pupil of Aristotle,
points out (Monro, p. 115) that ' there are two movements of
2l6 THE ROMAN STAGE

the voice, not properly discriminated by any previous writer :

namely, the continuous, which is the movement characteristic of
speaking, and the discrete or that which proceeds by intervals,
the movement of singing. In the latter the voice remains for
a certain time in one note, and then passes by a definite
interval to another. In the former it is continually gliding by
imperceptible degrees from higher to lower or the reverse.'
Nicomachus (first century A.D. or later) says (Enchiridion,
p. 4) that 1 if the notes and intervals of the speaking voice are
allowed to be separate and distinct, the form of utterance
becomes singing.' So long as the accent remained predomin-
antly musical, it is difficult to see how it could have been
ignored in utterance ; in other words, it is difficult to suppose
that the natural pitch of the syllable was violated so that the
voice might agree with the instrument. But when the pitch
accent began to give w a y to a stress accent, the difference
between speech and song might become more obvious.
Aristides Quintilianus (between the second and fourth century
A.D.) ' recognizes a third or intermediate movement of the
voice, that which is employed in the recitation of p o e t r y '
(Monro, p. 116). It is by no means clear that such a distinction
could have been made in the fifth century B.C. T h e view
that ' recitative' was employed in Greek drama is really
founded not on any precise and positive evidence but rather
on the fact that the Greeks do not seem to differentiate clearly
between ' speech' and ' song ', 1 So far indeed as the evidence
goes, it may be said that there were not three distinct forms of
dramatic delivery but only one, a form of speech more musical
than our speech but not at all corresponding to our idea of song.
T h e Romans appear to have had from early times a
predominantly stress accent. T h e y would therefore, we may
imagine, have been more predisposed than the Greeks to
recognize a distinction between speech and song. Nevertheless,
like the Greeks, they seem frequently to confuse these terms.
So Horace (Odes I V , xii. 10) talks of e speaking a s o n g '
(dicere carmen), and O v i d claims that his elegiacs were ' sung
all over R o m e ' , totam cantata per urbem {Trist. I V , χ. 59).
W e scarcely notice this confusion of terms, so familiar has it
become. N o doubt there are other references in Latin literature
1 So Aristotle (Poet, xii) defines the napobot as η πρώτη λίξιι όλου χοροί,
' the first speech of the whole chorus
 MUSIC AND METRE 217
to ' speech ' and ' song' which show that a distinction was
sometimes made ; but if the terms were so often confused in the
Augustan age and later, we can hardly assume that the
modern distinction between ' speech ' and ' song' was actually
observed in theatrical practice in the time of Plautus.
We are dealing not with a verbal distinction but with a
real problem. A possible clue may be afforded by the dramatic
uses to which the different metres are put. We find that
senarii are used invariably in prologues, and that the verse,
if not already in senarii, tends to change to that metre when
documents are being read aloud (Asin. 751 ff., Bacch. 997 ff.,
Cure. 429 ff.) ; when an oath is being dictated (Rud. 1338) ;
when the speakers are recounting dreams (Cure. 246, Merc. 125,
Rud. 593) ; in a scene of quiet persuasion (Amph. 882) ;
after a display of insanity (Men. 872) ; where a serious piece
of business has to be transacted (Cist. 747, Cure. 635). Sep-
tenarii, usually trochaic, are normal in epilogues and concluding
scenes, and are very commonly used in lively, bustling scenes.
Iambic octonarii have a slower movement ; Lindsay calls them
the metre of soliloquy,1 and we can see how suitable they are
to represent the timorous gait of Sosia through the dark streets :

qui me alter est audacior homo aut qui confidentior,

iuuentutis mores qui sciam, qui hoc noctis solus ambulem ? 2
Cretics, bacchiacs and other ' lyric ' metres are often used to
depict terror or some other intense emotion. We may take as
an example of dramatic crescendo the first three scenes in the
' Third A c t ' of the Rudens. At line 593 Daemones appears
alone and relates his dream ; he speaks in senarii. At line 615
Trachalio hurries out of the temple, bawling for help in some-
what burlesque style ; the metre is trochaic septenarius. At
line 664 the terror-stricken girls appear ; their despair finds
its natural expression in cretics. But elsewhere the ' lyric '
metres (cretics, bacchiacs, etc.) are used with comic effect.
Moreover many spirited scenes are in senarii : cf. Rud.
780 ff., 839 ff. Indeed an effective heightening of tension is
created by the change from septenarii to senarii in Rud. 450,
where (after a scene of badinage) the music stops abruptly as
Ampelisca catches sight of the hated Labrax. There is no
more dramatic scene in Plautus than Capt. 659 ff., where the
1 a
E.L.V. 277. Amph. 153, 4.
2l8 THE ROMAN STAGE

change from hurrying trochaic septenarii to business-like

senarii effectively reflects the change in Hegio's mood from
bewilderment to resolve. In these and other cases the sudden
silence of the accompaniment creates an atmosphere in which,
as we might s a y , ' one could hear a pin drop Plautus' choice
of metre is in general decided not (as in Greek drama) by tradi-
tion but by the mood of the moment and the love of variety for
its own sake. Certain metres are almost inevitable ; we take
it for granted that a hurrying messenger will speak in septenarii
(cf. Capt. 768, Cure. 280, Most. 348, Trin. 1008), while a
sedate traveller arriving from abroad will use senarii (Men.
226, Most. 431, St. 402). But sometimes it is not so much the
choice of metre as the change itself which is dramatic. In
Rudens 413 Ampelisca, speaking in iambic septenarii, knocks
at a strange door :

heus, ecquis in uillast ? ecquis hoc recludit ? ecquis prodit ?

T h e door bursts open and the angry Sceparnio appears with
an explosive trochaic line :
quis est qui nostris tarn proterue foribus facit iniuriam ?
Here the dramatic effect depends on abruptness ; it is unlikely
that there was a pause while the piper changed to the new
rhythm. In other words the piper had to take his cue from
the actors' words ; his part was altogether subordinate to theirs.
A totally different effect is achieved in Cure. 158, where
again the metre changes to trochaic septenarius as the door
opens stealthily at the husky whisper of the old duenna :
placide egredere et sonitum prohibe forium et strepitum cardinum,
ne quae hie agimus eru' percipiat fieri, mea Planesium.
T h e bacchiac, which in Rud. 259 ff. represents the dignified
entry of the priestess, is used in Men. 753 to portray the futile
effort of the old man to mend his pace :
ut aetas mea est atque ut hoc usu' facto est,
gradum proferam, progrediri properabo.
It is evident that Plautus was keenly interested in metre and
rhythm, and that he used them for dramatic effect in a w a y
which can scarcely be paralleled in Greek literature. Compared
with him Terence is uninspired. But all the R o m a n dramatists,
to the best of their ability, freely employed metrical variety to
bring out the theatrical effects at which they aimed. A l l
 MUSIC AND METRE 219
drama was in metre ; the value of the accompaniment was to
emphasize the metre. But to argue from the presence of the
accompaniment that certain scenes were sung presents us with
a formidable dilemma. Song, in our sense of the word,
demands a special vocal and mental e f f o r t ; while the actor is
singing everything else—gesture, dance, etc.—must take second
place ; for the time being the action of the play is held up.
N o w it is practically certain that all the metres in a Latin play
except the senarii were meant to be accompanied. But these
metres, taken together, occupy the greater part of each play.
It is difficult to suppose that they were all meant for song.
Therefore most critics strive to distinguish the true songs from
the mass of recitative. But any attempt to do so must be
arbitrary ; Latin writers do not hint at such a distinction, and
we have to fall back on our own sense of what is fitting.
Perhaps the most plausible example of song in Plautus is
the eight-line serenade in cretics which Phaedromus addresses
to the bolts of the door which shuts him from his mistress
[Cure. 147-54). H e has asked his slave ' Shall I go up to the
door and sing a serenade ? ' and the slave replies, ' I won't say
no or yes, seeing that your character is so changed.' T h e n
Phaedromus begins :

Bolts, ah, bolts, I greet you gladly ;

Take my love and hear my plea,
Hear my prayer, my supplication,
Fairest bolts, ah, favour me.
Change to foreign dancers for me,
Spring, I pray you, spring on high,
Send a wretched man his dear love,
Love that drains his life-blood dry.
Look ! they sleep, those bolts most base
Will not budge to do me grace !
(Nixon's translation.)

If the Curculio were being presented on the modern stage,

this ' serenade ' would undoubtedly be sung. So ineradicable
is our habit of attributing to antiquity our own outlook that
we do not stop to enquire whether in fact the R o m a n actor
used for these lines a form of utterance different from what he
had used for the preceding dialogue.
T h a t Plautus knew the charm and power of instrumental
music is indicated in several passages. In the Casina, as the
220 THE ROMAN STAGE

supposed ' bride ' is b e i n g brought out o f doors, the w a i t i n g

b r i d e g r o o m calls on the piper to ' fill the w h o l e street w i t h the
sweet m a r r i a g e tune and together w i t h L y s i m a c h u s h e cries
' hymen hymenaee о hymen ! ' I n St. 7 6 1 - 2 the piper is
bidden ' h u r r y u p and p l a y us some c h a r m i n g sweet w a n t o n
air to m a k e us tingle to our finger-tips.' Pseudolus, leaving the
stage for a minute or two, promises the spectators that d u r i n g
his brief absence they will be entertained b y the piper (573 a).
Nevertheless it is v e r y doubtful w h e t h e r w e can regard the
serenade in the Curculio as a song in our sense of the w o r d .
T h e word occentare (which I h a v e translated ' sing a s e r e n a d e ' )
cannot b e pressed too far : Plautus himself uses the w o r d
elsewhere (Pers. 569) of the uproar m a d e b y a p a r t y of a n g r y
revellers before they burst into a house. 1 T h e serenade is
indeed m a r k e d out f r o m its i m m e d i a t e context b y the cretic
metre ; nevertheless that context is itself in metres usually
considered ' l y r i c a l ' ; the preceding dialogue between Phae-
dromus and his slave (138-46) is in anapaests, the following
f e w lines are in glyconics and dochmiacs, and one of t h e m
seems m e a n t not for song b u t for whisper :
st, tace, tace !—taceo hercle equidem.—sentio sonitum.
T h e content of the serenade is not particularly lyrical ;
the effect is comical rather than sentimental. I cannot discover
a n y m a r k e d l y lyrical quality in the so-called ' l y r i c a l ' passages
in Plautus. H e seems to m e to reach his highest note in the
trochaic septenarius. T h e r e is no passage more like a drinking-
song in tone than St. 729 :
haec facetiast, amare inter se riualis duos,
uno cantharo potare, unum scortum ducere.
hoc memorabilest : ego tu sum, tu es ego, unanimi sumus ;
unam amicam amamus ambo, mecum ubi est, tecum est tamen ;
tecum ubi autem est, mecum ibi autemst; neuter neutri inuidet.
O n the m o d e r n stage these lines w o u l d readily lend themselves
to song. Y e t so far as the metre is concerned they are in no
w a y different f r o m the rest of this long, lively scene of dialogue
in trochaic septenarii. N o w h e r e does Plautus c o m e nearer to
the spirit of a m o d e r n love-song than in Cure. 178-80 :
sibi sua habeant regna reges, sibi diuitias diuites,
sibi honores, sibi uirtutes, sibi pugnas, sibi proelia :
dum mi abstineant inuidere, sibi quisque habeant quod suom est.
1 Merc. 408 (occentent ostium) might be taken in either sense. See p. 30 and note.
 MUSIC AND METRE 221

These lines again form merely part of a long scene of dialogue

in the same metre. By way of contrast, what specially lyrical
quality can any one find in the bacchiacs with which Eunomia
admonishes her brother {Aul. 120 ff.) :
uelim te arbitrari med haec uerba, frater, . . . ?
The one point clearly established is that Plautus relied on
the piper to accompany all metres except the iambic senarii.
What music the piper would supply was probably left to him.
He was not a composer but an instrumentalist ; the didascaliae
give his name after that of the chief actor, and he seems to
have been a slave (' Marcipor, the slave of Oppius', ' Flaccus,
the slave of Claudius '). No doubt he had at his command
a number of simple phrases and chose whichever of them
seemed best suited to the metre of the scene. His art, such as it
was, was entirely ancillary to that of the actors. At the re-
hearsals he would presumably have become familiar with the
play. The letters С and D V (if we may suppose that they were
written in the author's manuscript, which was also no doubt
the actors' book) would make clear at the beginning of each
scene whether it was to be accompanied or not. Plautus was
an artist in rhythm, and he well knew how much the rhythm
of his lines depended on accompaniment if it was to have its
full effect. Cicero {Orat. lv. 183-4) tdls us that, but for the
presence of the piper, even some bacchiac verses in tragedy
would seem like prose :

quaenam te esse dicam ? qui tarda in senectute . . .

quae, nisi cum tibicen accessit, orationi sunt solutae simillima.
We distinguish between ' s o n g ' and instrumental accom-
paniment ; for the Greeks and Romans this was almost
impossible. The word cantare is used of the instrument :
cf. tibicinam cantantem {Most. 934). The word canticum
meant ' utterance to musical accompaniment' without any
implication that the utterance, considered vocally, was
anything more than rhythmical speech. Cicero {Sest. 1 1 8 )
uses the words cantores, histriones and actores of the same
persons. But in the same passage (122) he refers to some lines
in septenarii as having been spoken by the actor (ilia quem ad
modum dixit); and as he applies the verb ' agere ' to the actor's
rendering alike of the septenarii and of the anapaests, he
seems to imply that the style of utterance was not markedly
222 THE ROMAN STAGE

different. Plautus {Pseud. 366) makes Ballio address his

critics as cantores probos ! meaning not ' how sweetly you sing ! '
but ' what lungs you have ! ' So canto and cano are used by
the dramatists of any loud utterance—e.g. the crowing of a
cock (M.G. 690)—or droning repetition (Tr. 287, Ph. 495).
Whatever the metre, the words are said to be ' s p o k e n '
(dico or loquor) : so after a passage in cretics and other metres
uttered by Sosia the listening Mercury remarks (Amfih. 248) :
nunquam etiam quicquam adhuc uerborum est prolocutus
perperam.
In Cas. 166 Myrrhina, after beginning a speech which is
partly in anapaests, partly in cretics, asks ' ecquis haec quae
loquor audit ? ' It seems unlikely that Cas. 213 was meant to
be sung, though the metre is glyconic :
st! tace !—quid est ?—em !—quis est quem uides ?—uir eccum it.
W e have plenty of evidence that R o m a n as well as Greek
actors were expected to have loud voices ; the all-important
thing was to make themselves audible. Terence's actor-
manager, Ambivius Turpio, complains of the strain put on
the actor's strength by his violent efforts and loud shouting :
clamore summo cum labore maxumo (H.T. 40).
There is no reference in the plays to any one's powers of singing,
as distinct from speaking, and no mention of a special vocalist.
T h e lyrical metres are distributed among all the characters :
thus in the Casina Pardalisca, Olympio and Lysidamus are
all on the stage together, each speaking for the most part in
lyrical metres (814-54) ; then Myrrhina, Pardalisca and
Cleostrata (855-73) ; then Olympio, Cleostrata and Myrrhina
(874-936). If cretics, bacchiacs or any other metres denote
' s o n g ' , then Plautus must have taken for granted that the
actors would all be able to sing. A single vocalist could not
have undertaken the singing of all the cantica, no matter how
narrowly we try to define the term.
Beyond all doubt Plautine comedy is drama, not opera.
Nevertheless it had a musical element. In some measure
Plautus here resembles Shakespeare. Each wrote at a time
when the native stress accent was struggling to assert itself
against quantitative metres imported from abroad, or from
classical tradition. Each wrote verses intended for musical
 S T R E E T MUSICIANS
Mosaic by Dioskoridesj from Pompeii
 MUSIC AND METRE 223
accompaniment, which he expected others to supply. That
there was music and song in Elizabethan drama was in part
due to the classical tradition ; poets since the days of Homer
had been said to ' sing ' their verses, and no one stopped to ask
what the Greeks had meant by ' song.' 1 The Elizabethan age
itself was pre-eminently an age of song. This was made
possible by unique musical and cultural conditions; there was a
close relation between poets and musicians, and a general love
of music and poetry. The essential difference between the
Plautine cantica and Shakespeare's songs may be made more
plain when we consider that Shakespeare's songs (1) are short ;
(2) are easily distinguishable from their context; (3) are
composed in stanzas, intended to be sung to a repeated air ;
(4) are lyrical not only in form but in sentiment ; (5) though
often of great dramatic importance,2 nevertheless do not
advance the action in quite the same way as does the dialogue ;
(6) are assigned to a particular character, obviously because
his part was meant to be taken by an actor with a good voice ;
(7) are usually referred to in the context in terms which
indicate specifically that they were sung. ' A mellifluous
voice, as I am a true knight ! ' ' My voice is ragged ; I know
I caiinot please you.' ' Truly, young gentleman, though there
was no great matter in the ditty, yet the note was very
untunable.' The cantica of Plautus satisfy none of these
conditions. If the term includes all passages accompanied by
music, they occupy the greater part of each play. If we exclude
the septenarii, the other metres are so mingled that it is impos-
sible to isolate the so-called ' songs'. No one has succeeded in
discovering any trace of 4 strophic ' composition in Plautus.3
1
See Bruce Pattison, Music and Poetry of the English Renaissance, p. 2 0 - 1 .
Perhaps the music which Homer played on his lyre was not unlike the music
played by the Norman minstrels (ib. p. 24) : ' a little snatch of melody served for
a single line and was repeated throughout the composition. This sounds terribly
tedious to us, but the attention of the audience was no doubt concentrated on
the story.' There is much uncertainty as between declamation and song : the
O.E. expression ' singan and secgan '—' sing and say'—suggests that Epic
' came to be declaimed
* See Noble, Shakespeare's Use of Song, p. 12.
3
Even Crusius, who believes that he has proved the existence of Responsion
(in his sense of the term), has to admit that ' syllabic responsion scarcely exists
in Plautus ' (Die Responsion in den Plautinischen Cantica, 1929, p. 2). The true
explanation for the frequent examples of symmetry in Plautus has been pointed
out by Lindsay (Early Latin Verse, p. 1 1 3 ) : ' he likes to make a reply the exact
counterpart of a question . . . stroke is exactly matched by counter-stroke, cut
by parry . . . This is a feature of all lively comedy, English as well as Latin.'
224 THE ROMAN STAGE

T h e passages in cretics, etc., are no whit more lyrical than

certain passages in the septenarii. T h e normal action of the
play is carried on as freely in one metre as in another, although
when audibility is specially important the accompaniment
stops altogether and the metre becomes the senarius. Cretic
and bacchiac passages are distributed as freely as any others
among the characters. There is no specific reference to song,
the musical use of the human voice, in Plautus, and it is
probable that in his day the very idea o f ' song ' (in our sense
of the word) did not exist.
T h e much-debated question as to the origin of the cantica
would seem to be largely unreal. Whatever metrical hints the
Latin dramatists gathered from external sources, their use of
the metres is characteristically Roman. Professor Fraenkel
has rightly exorcized the phantom of ' Hellenistic o p e r a '
conjured up by the imagination of Leo and Wilamowitz
(Plautinisches im Plautus, p. 333). Unfortunately he himself
clings to the view that Plautine comedy is distinguished from
its Greek originals by its ' interchange of spoken and sung
scenes' (ib. p. 323). There is no song in Plautus. His achieve-
ment was to exploit the dramatic possibilities of rhythm as no
writer before or after him has done. T h e nameless poet
who composed his epitaph showed sound judgment when he
made Laughter and R h y t h m the chief mourners at the
dramatist's grave.
 CHAPTER XXVII

EPILOGUE : DRAMA UNDER THE EMPIRE

T H E T H E A T R E was a characteristic part of Roman life

and civilization. Consequently under the Empire we
find R o m a n theatres, great or small, springing up in every
province. W h e n we ask what kinds of performances took
place in these buildings, the answer is doubtful and dis-
appointing. Such information as we possess suggests that the
entertainment normally provided in the imperial theatres
consisted of trivial or degrading performances, whether mime,
recitation, pantomime or even gladiatorial combat.
T h e establishment of the Empire helped to confirm certain
tendencies which had already become visible in Republican
drama, and also to create others. L o n g before the end of the
Republic the supply of new plays for the stage had practically
ceased. T h e Augustan age gave birth to two famous tragedies,
Varius' Thyestes and Ovid's Medea (almost certainly these were
original works, not translations). According to one scholiast
the Thyestes was produced at the games in celebration of the
victory of Actium ; if this information is accurate, it appears
to be our last record of the performance of a new Latin play.
Otherwise such few performances of literary tragedy or comedy
as are recorded are revivals of old plays. For example Seneca
(ep. 80) seems to speak of a contemporary performance of
Accius' Atreus. Y e t the inevitable change in literary style
would make the works of the Republican tragedians seem more
and more archaic as time went on. For Quintilian Pacuvius
and Accius are the rude pioneers of R o m a n tragedy, which
culminates in writers like Ovid. But as O v i d himself, long
after the publication of the Medea, protests that he never was so
depraved as to write for the stage, it would appear that
Quintilian is here thinking of tragedy simply as a literary form,
not as something intended for performance. Something of
the spirit of tragedy may have survived in the dramatic
recitations; Nero, we are told, ' s a n g ' Orestes the Matricide,
22Ö THE ROMAN STAGE

Oedipus Blinded, Hercules Mad. These seem to have been

cantica depicting certain scenes of tragic character ; sometimes
the words were Greek. The vogue of these recitations, such as
it was, itself suggests· that tragedy proper was no longer
familiar on the stage. Far more important was the pantomime,
introduced in 22 B.C. by Pylades and Bathyllus. This typically
Roman performance (the Greeks called it the ' Italian dance ')
stood at the opposite extreme to the dramatic recitation ; the
central figure was the masked dancer, who performed scenes in
dumbshow, while appropriate words were sung by a chorus.
Lucian's lively account shows that the themes were taken from
mythology—for example the love of Ares and Aphrodite and
the snare set for the guilty pair by the injured husband
Hephaestus. The libretto sung by the chorus was specially
written for the performance. The writing of such libretti was
a lucrative if degrading occupation which attracted even such
poets as Lucan and Statius. The performance took place on
the stage ; some believe that elaborate scenery (illustrated,
perhaps, in the Pompeian wall-paintings) formed the back-
ground.1 Some passages in ancient authors would suggest, it
is true, that the pantomime was regarded as drama, whether
tragic, comic or satyric ; but in reality it seems to have been
something quite new and unconnected with drama proper.
The modern view that tragedy was decomposed, so to speak,
into the dramatic recitations and the pantomimic dances is
altogether artificial. The most obvious link of the pantomime
with drama was the stage setting ; its themes, taken from
mythology, might naturally be tragic ; at its highest level it
may have appealed to that section of the public who would in
other circumstances have been interested in tragedy ; but our
evidence is that the essential attraction of pantomime was the
supple, artistic, expressive, passionate, sometimes exquisitely
lascivious movements of the dancer—and what has such a
performance to do with true tragedy ?
Occasional attempts are made by modern writers to show
that the tragedies attributed to Seneca were intended for
performance on the stage. They happen to be the only
complete Latin tragedies which we possess, and their interest
as literature would be greatly increased if we could regard them
1
The usual view ; but there is no evidence to support it. See Purser's article
on Pantomimus (Smith's Dictionary of Antiquities, ii. 334-6).
 EPILOGUE Г DRAMA UNDER THE EMPIRE 227

as genuine drama. But if anything is clear about Roman

drama as a whole, it is that no one wrote for the stage except
to make money. This is true even of Republican times, when
the theatre was still fairly respectable ; under the Empire the
Romans of the literary class regarded the theatre with some-
thing like horror. It is incredible that Seneca, one of the
richest men in Rome and a man who openly admits his dis-
taste for close contact with the common people or their
amusements, should have composed plays intended to win
the favour of the general public. The dramatist who writes
for the stage must take into account not only the tastes of his
audience but the requirements of the stage ; and the internal
evidence of the Senecan plays shows that the author has not
visualized the actions of his characters. The usual technique
of bringing characters on or taking them off is ignored. We
often realize that a person is conceived as present only by the
fact that a speech is put into his mouth. We cannot tell when
he leaves the stage except by the fact that no more words are
attributed or addressed to him. A long speech is attributed to
Clytaemnestra (Agam. 108-124) > Y e t i* appears from the
remarks of the other person present (126) that Clytaemnestra
has been silent; the speech must therefore represent her
thoughts. Things happen which could not have taken place
on the classical (or indeed any) stage : Hercules shoots his
wife and children ; afterwards he tries to commit suicide by
shooting himself with his own bow and arrow ; the fragments
of corpses are pieced together, etc. If at times the author
seems vaguely conscious of a stage, the explanation is that
though his plays are original works, he is using as his models
Greek plays, which were of course intended for performance.
There is no evidence that Seneca (or whoever the author was)
was imitating the old Latin tragedies ; the Senecan tragedies
are simply artificial imitations of Greek tragedy, worked up in
the style of the Silver Age, and they are meant to be read or
declaimed, not to be acted.
The Octauia, our sole extant example of the Roman
historical play, is generally thought to have been composed
by some unknown author shortly after the death of Nero, to
which event there is a ' prophetic' reference circumstantial
enough to indicate knowledge of the facts. While in style less
v
rilliant and somewhat more natural than the tragedies of
228 THE ROMAN STAGE

Seneca, it is equally regardless of stage requirements. T h e plot

is based on the facts of history, but is worked up on the model
of Greek tragedy. T h a t Octavia, Nero's innocent and un-
fortunate wife, should have a nurse in whom to confide is
perhaps admissible ; but that his hardened mistress Poppaea
should also be assigned a nurse is more than we can stand.
It is doubtful whether there is any connexion between this
piece and the old Republican praetextae, which celebrated the
achievements of R o m a n consuls in battle. T h e Octauia is a
purely literary and artificial treatment of recent history on the
lines of Greek tragedy.
Even in Republican days the R o m a n government had been
very sensitive to political allusions, however veiled or even
unintentional, if uttered on the stage. Under the Empire such
allusions might bring appalling consequences to the author
and the actor concerned. T h e tragedies of Seneca as well as
the Octauia present absolute rulers in an odious l i g h t — a
consideration which makes it still less likely that they were
composed for production in the presence of the general public
(including the Emperor himself) during the reign of Claudius,
Nero or Vespasian. For although the occasional revival of an
old tragedy, such as Accius' Atreus, might be sanctioned (and
our evidence suggests that such revivals were infrequent), and
although a spice of political allusion, however dangerous, was
in the very nature of the mime, the composition and perfor-
mance of new tragedies which might rouse popular feeling
against the imperial system was another matter.
It is highly unlikely that any of the tragedies which have
come down to us from the Empire was ever performed until
the Renaissance. Then their influence was all-important, and
went far to determine the nature of modern drama. In
A.D. 1551 - 2 Seneca's Troades was performed at Trinity College,
Cambridge. T h e Elizabethan audiences, almost as inured to
scenes of blood as the Romans, liked the Senecan blend of
moralizing and melodrama. Juliet's Nurse, the Ghost in
Hamlet, the Tyrant as portrayed in Richard the Third, are figures
which owe something to Seneca. Fortunately for the world,
the Elizabethan drama was the product of many other factors
as well as the influence of Senecan classical tragedy.
T h e position of comedy under the Empire was no better than
that of tragedy. T h e very word comoedus is used of a slave who
 EPILOGUE : DRAMA UNDER THE EMPIRE 229

reads extracts from comedy as an entertainment for guests at

dinner. Purely literary comedies were composed : Pliny tells
us that he has heard Vergilius Romanus ' reading to a few
listeners ' his imitation of Old Greek Comedy (which included
a flattering reference to Pliny himself). Vergilius had also
written imitations of New Comedy, which Pliny compares in
excellence to the works of Plautus and Terence. 1 Quintilian,
in his reference to Latin comedy, mentions no composer later
than Terence ; nevertheless he refers to contemporary actors of
comedy, who wore masks, played the stock roles of the palliata,
and were so moved by their own acting as to be seen in tears
after they had removed their masks. Apparently, then, there
were still performances of genuine high comedy; but what were
the comedies so performed ? Immediately after his mention
of the contemporary comic actors Demetrius and Stratocles,
Q_uintilian quotes a passage of comedy to illustrate his meaning
— a n d the quotation is from Terence. This suggests that the
comedies performed in the Imperial theatres were the old
classics. Such revivals were probably exceptional; Pliny
speaks of comedy, as well as tragedy, as suitable for declamation;
the performances which he mentions are mimes and panto-
mines. Juvenal speaks of togatae as still being written in
his d a y — b u t apparently only for recitation. T h e only togata
which we know to have been performed under the Empire is
the Incendium of Afranius, and this was revived merely in order
to give a spectacular display of a stage fire. If the palliatae
and togatae written under the Empire were as oblivious of
stage needs as are the tragedies of Seneca, we can understand
that theatrical managers, if they wished to produce a comedy,
were forced to draw on the Republican classics.
If we can judge by the history of Greece and Rome, a
certain measure of public freedom is necessary to the production
of true drama. It was difficult for the Romans of the Empire
to feel any sense of corporate freedom or responsibility, or even
to take pride in the greatness of the universal Empire ; patriot-
ism is impossible in a State which fills the world. In such
circumstances men's interests were confined to themselves,
their own affairs and the trivial interests and follies of everyday

1 Cf. the epitaph ( C I L . ix. 1164) of Μ . Pomponius Bassulus (second century

A.D.) : ' lest I should live a life of bestial sloth, I translated some dainty plays of
Menander, and composed some new plays myself'.
16
230 THE ROMAN STAGE

life. Under foreign domination the Greeks had indeed pro-

duced New Comedy ; the Romans, overwhelmed under their
own Empire, gave themselves up to a merely sensual existence.
In their theatres pantomime took the place of tragedy, while
comedy gave way to farce. Since the sole aim was to tickle the
jaded palate of the public, producers not only lavished all the
resources of wealth and technique on their extravagant
productions, but also descended to the lowest depths of the
disgusting and the obscene. Even Livy regarded the theatre
of his day as a danger to public morals and the existence of
the State ; soon sexual displays were visibly presented on
the stage, and stage ' executions ' were carried out in reality
(by substituting for the actor a condemned criminal). While
we must always bear in mind that it is usually the exceptional
and the extreme which is recorded, and that much of our
information comes from sources hostile to the stage, the
cumulative weight of testimony makes it clear that the dramatic
standards of the Empire were altogether below those of the days
when such plays as the Captiui and the Adelphi were written
and produced for the amusement of the people.
The fabula Atellana maintained its place on the stages of
the first century A.D., both in Rome and in the provincial
theatres. It was still a masked performance, given as an
' after-piece ' ; it still had a rustic flavour, and it still brought
on the stage the familiar figures of Maccus, Bucco, Pappus,
Manducus and Dossennus ; its performers were still fond of
quips, riddles and topical allusions ; their status, whatever it
had been in Livy's day, seems now to be no better than that
of the other theatrical performers, whether slaves or freedmen.
Tacitus speaks with contempt of the enormous vogue and the
degraded morality of the Atellana in the time of Tiberius.
We hear occasionally of Greek cantica and of mythological
themes, but on the whole the Atellana seems to have preserved
its Italian, country-town flavour. The wealthy freedman
Trimalchio, so proud of his humble Italian origin that he
compelled his flute-player to play only Latin airs, also insisted
that the troupe o f ' comedians ' which he had bought should
perform only Atellane plays. We hear practically nothing of
literary Atellanae, or of the authors of such Atellanae as were
performed on the stage. Perhaps after the time of Pomponius
4nd Novius this type of farce had returned to its sub-literary,
 EPILOGUE : DRAMA UNDER THE EMPIRE 23I

semi-improvised form. No doubt in the course of time it grew

more and more like the mime, though the use of masks must
always, one would imagine, have been a characteristic feature.
After the end of the first century A.D. we hear no more of the
fabula Atellana.
The composition of purely literary mimes was one of the
amusements of dilettanti; belle componis mimos, as Martial
remarks to one of these writers. Perhaps these literary mimes
were mimiambi, such as were written by Pliny's versatile
friend, Vergilius Romanus, character studies intended for
recital, modelled on the mimiambi of Herondas. The mime
intended for performance on the stage remained sub-literary, un-
metrical and largely impromptu. The term, as in earlier times,
includes any ' imitative' piece from sheer mimicry to playlets
introducing several characters and containing several scenes.
No Latin mimes have survived ; we have, however, an
acting edition of a Greek mime preserved from the second
century after Christ, containing six or seven short scenes.
The action takes place in front of a house (the door of which
could conveniently be represented by the opening in the curtain
through which the performers make their appearance). The
beginning of the mime is lost. The opening words show us the
archimima or ' leading l a d y ' in the character of a faithless
wife, trying to seduce Aesopus, one of her slaves. He is in
love with a fellow-slave, Apollonia, and rejects his mistress's
advances. She then orders the lovers to be taken away and
left to die. A later scene shows the ' body ' of Aesopus being
brought in ; the other slaves pretend that he has thrown
himself from a height (actually they have drugged him for
his own safety) ; his mistress mourns his death, but soon
consoles herself with the company of another slave, Malacus,
with whom she conspires to poison her husband, whose 4 body '
is presently brought in. Now comes the turning-point. The
old man gets u p and denounces his guilty wife and Malacus;
they are led off to punishment, while Aesopus and Apollonia
are found to be alive and well, and all ends happily.
The dialogue is in prose ; probably the actors—or at least
the archimima, who has by far the most important role—felt
free to expand it at will. The sordid theme and the startlingly
indecent language seem to be characteristic of the mime in
general.
232 T H E ROMAN STAGE

In indecency the mime reached incredible heights. Not

only was adultery a stock theme ; the Emperor Heliogabalus
ordered its realistic performance on the stage. It was natural
that the Christian church should set itself against the mime,
and equally natural that the actors should retaliate by mocking
Christian sacraments, much to the delight of the crowd.
Gradually the church won the upper hand. In the fifth
century all performers of mime were excommunicated. In the
sixth century Justinian closed the theatres. Yet the mime
lived on. Its simple requirements—a stage and a curtain—
could be supplied in any public place or private house, and in
such settings it continued to entertain audiences who were now
nominally Christian. Though forced to drop its burlesque of
the sacraments, it still scandalized the Fathers by its indecency
and the immorality of its performers. Yet, as one of the last
strongholds of paganism, the mime did not lack defenders.
About A.D. 500 the sophist Choricius of Gaza wrote his
apology for actors (i.e. performers of mime), while the mime-
actress Theodora, whose ' strip-tease' acts had delighted the
public of Byzantium, succeeded in captivating the affections
of Justinian himself, and was elevated to the imperial throne.
How far the mime survived the fall of ancient civilization
is a doubtful point. So simple and elemental a type of perfor-
mance might arise independently in many ages and countries.
Yet it is hard to feel confident that the classical mime ever
wholly ceased to be. Certainly the Middle Ages had their
mimes. In its latest phase the classical mime was the last
representative of classical drama ; its strolling performers,
' birds of passage ' as a Greek poet calls them, had taken over
all that was left of a great tradition. Somehow or other they
may have handed on their craft to their successors of the Dark
and Middle Ages. When the darkness clears away we see a
new drama arising. Perhaps its performers were historically
linked with the strolling companies of classical times. However
that may be, the re-birth of classical literature was to ensure
that the new drama would grow to its full stature under the
shadow of the drama of Rome.
 APPENDIX A

SEATS IN T H E G R E E K AND R O M A N THEATRES

(Classical Revue, Vol. liii. pp. 51-5.)

T H E R E are several passages in ancient authors 1 which, taken

together, seem to imply that the spectators of Roman drama during
its productive period were forced to stand. This view was readily
accepted by nineteenth-century scholars, eager to find as many
differences as possible between Greek and Roman practice ; it is,
of course, incompatible with the frequent references to a seated
audience in the prologues of Plautus' plays ; 2 such references were
therefore regarded as proof that the prologues themselves were
post-Plautine. 3 We may, indeed, agree that a prologue is the part
of a play which is most liable to modification at the hands of later
producers; 4 and with the prologues we might be willing also to
sacrifice the concluding words of the Truculentus :
spectatores, bene ualete, plaudite atque exsurgite.

There are other references to seats, however, within the body of the
plays (Aul. 719, Cure. 644-7, P° en - 1224) ; and these passages set us
a pretty dilemma : are we to regard them as examples of somewhat
mechanical translation from the Greek, or as post-Plautine
insertions ?
The supposed evidence against the existence of seats seems to
imply further that the spectators were forced to stand not merely
from lack of accommodation but because of decrees designed to
check luxury and idleness.6 Such legislation must have been
singularly unpopular with the theatre-goers who had to remain on
their feet throughout long plays, the actors and playwrights who
were handicapped by the discomfort thus inflicted on their public,
and even the magistrates who gave the shows in order to win the
favour of the electorate. Every one present would, of course, be
aware that in Greek theatres no farther away than Pompeii seating
accommodation was provided for all. In Poen. 1224 a character
remarks ' in pauca confer : sitiunt qui s e d e n t ' Gut it short: the
1
Given in P.-W., s.v. theatrum.
1
Amph. 65 ; Capt. 12 ; Poen. 10 ; Pseud. 1.
3
This view is still upheld, e.g. by Wight Duff, Literary History of Rome, 1927,
p-157·
4
Cf. the prologue to the Casina, which refers expressly to a revival performance.
5
Cf. Tac. Ann. xiv. 20 : stantem populum spectauisse, ne, si consideret
theatro, dies totos ignauia continuaret.
233
234 THE ROMAN STAGE

stalls are thirsty Is this, perhaps, an example of mechanical

translation from the Greek ? But how irritated would a standing
audience be at such a reference to the (comparatively trifling)
discomforts of a seated audience ! How tactless the dramatist who
would allow such a reference to remain ! This line must, therefore,
be a post-Plautine insertion (in spite of its Plautine ring) ; we may
admit that jokes of this kind might well be added by an interpolator.
But it is not so easy to explain away Aul. 718-9 :

quid est ? quid ridetis ? noui omnes, scio fures esse hie complures,
qui uestitu et creta occultant sese atque sedent quasi sint frugi.

' Eh, what's that ? What are you grinning for ? I know you, the whole
lot of you ! I know there are thieves here, plenty of 'em, that cover them-
selves up in dapper clothes and sit still as if they were honest men ! ' 1

Here Euclio, robbed of his gold, is addressing the audience—

and he refers to them as seated. Jesting references to the knavishness
of the audience may, of course be paralleled in Greek (e.g. Frogs,
274-6), and it would be natural to assume that Plautus is here
translating his original ; but we cannot suppose that the lines
quoted were addressed to a standing audience ; they cannot, there-
fore, be an example of mechanical translation from the Greek, 8 but
must ex hypothesi be another post-Plautine insertion. But I now
wish to draw attention to a remarkable passage in the Curculio
(64З-7) :

THER. nutrix quae fuit ? PLAN. Archestrata.

ea me spectatum tulerat per Dionysia,
postquam illo uentum est, iam, ut me conlocauerat,
exoritur uentus turbo, spectacla ibi ruunt,
ego pertimesco. ibi me nescio quis arripit . . .

These lines occur in the vital αναγνώρισα or ' recognition ' scene.
T h e heroine is explaining how she was kidnapped in her infancy.
Her nurse had taken her to the show at the Dionysiac festival.
' We had scarcely arrived, and I been put in my place, when a
perfect hurricane arose ; the seats caved in—I was so terrified !
Then some one or other seized me. . . .' (Nixon's translation.)
So far as I know, the authenticity of these lines has not been ques-
tioned ; in fact, they form an appropriate and almost essential
part of an essential scene. There can be no reason, therefore, for
doubting that they are from the hand of Plautus himself. From
this it would seem to follow that even in Plautus' times the word
spectac{u)la had acquired the sense ' seats for spectators ' in which
1
Nixon's translation.
* Especially in view of the apparent reference to togas, togae cretatae,—which
seem to have been worn as early as the fifth century (Livy I V . xxv. 13).
 APPENDICES 235
we find it regularly used in Classical and Silver Latin. 1 Moreover,
as the scene described is Greek in detail, and as the passage would
have been as essential to the Greek original as to the Latin transla-
tion, it would seem to follow that Plautus is here translating from
the Greek.
T h e scene of the Curculio is in Epidaurus (line 341) ; the kid-
napping of the heroine, however, must have occurred somewhere
else. 2 O f the Greek original of the Curculio we know nothing ; but
as Athens was the birthplace and focus of New Comedy, 3 it is
natural to suppose that this play also was written and first produced
at Athens, and further, that the heroine should herself be regarded
as Athenian by birth (like Palaestra in the Rudern). T h e kidnapping
would, therefore, have occurred during the festival of Dionysus in
Athens. I know of no evidence that special seats were erected for
the spectators of the processions on such occasions ; the natural
assumption is that the kidnapping occurred in the theatre. 4
Apparently, then, we are to understand that the spectators' seats
in the Athenian theatre collapsed in a storm. T h a t this incident
had actually occurred, and within the memory of the spectators
who first saw the original of the Curculio, is suggested by Thera-
pontigonus' remark (line 651) : ' memini istanc turbam fieri
T o justify this conclusion, I must now endeavour to show
(1) that the word spectacla in the passage quoted can only mean
' spectators' seats ' ; (2) that, even in the time of New Comedy,
there were still, in the Athenian theatre, seats capable of being
blown down by a storm—in other words, wooden seats supported
on scaffolding.
T w o passages in Suetonius show clearly that spectacula can be
used of the part of an amphitheatre occupied by the spectators
(Cal. 35 : ' hunc spectaculis detractum repente et in harenam
deductum Thraeci comparauit' ; Dom. 10 : ' patrem familias
quod Thraecem murmilloni parem, munerario imparem dixerat,
detractum spectaculis in harenam canibus obiecit' 5 ). T h e references
here would presumably be to the tiers of stone seats : but that
spectacula can also mean ' grand stands ' is indicated by T a c . Ann.
X I V . 13 : ' exstructos, qua incederet, spectaculorum gradus, quo
modo triumphi uisuntur ' ; and that the ' gods ' in the amphi-
theatre occupied such wooden stands is suggested by Suet. Aug. 44,
1 Cie. Pro. Sest. 124 ; Livy I. xxxv. 8 ; Ovid Met. x. 668 ; Tac. Ann. xia. 13 ;

Suet. Calig. 35, Dom. 10.

2 Kidnapped children are always taken to another town : see Captivi,
Menaechmi, Poenulus, Rudens.
' Half of Plautus' plays are staged at Athens.
4 T h e vexed question whether women were admitted to the Athenian theatre

would seem to be settled, for the late fourth century at least, by this very passage.
1 The pater familias would presumably be sitting (within earshot of the emperor)

in one of the rows of stone seats which formed the lower two maeniana of the Flavian
Amphitheatre.
236 THE ROMAN STAGE

where we are told that, to allay a panic, the emperor left his place
and sat in that part of the building which seemed threatened with
collapse. 1 T h a t the occupants of the spectacula normally sat even
in the time of Cicero is shown by Pro Sest. 1 2 4 : ' m a x i m u m uero
populi R o m a n i iudicium uniuersi consessu gladiatorio declaratum
est . . . in hunc consessum P. Sestius . . . uenit . . . tantus
est ex omnibus spectaculis . . . plausus excitatus . . Fin-
ally we have a description in L i v y of the ' grand stands ' of the
primitive Circus (I. x x x v 8) : ' tum primum circo, qui nunc
maximus dicitur, designatus locus est. loca diuisa patribus equiti-
busque ubi spectacula sibi quisque facerent ; fori appellati ; spec-
tauere furcis duodenos a b terra spectacula alta sustinentibus pedes '.
Evidently the Fathers stood—or (perhaps more probably) s a t —
on platforms twelve feet high, which were supported by forked
poles.
Not only can spectacula mean ' seats at a show ' : it is the only
precise word for such seats (though we find subsellia, scamna, sedilia,
used for the ' seats ' benches ', or ' rows '). T h e primary meaning
of the word is ' show ' spectacle ' , its only other sense is ' seats '
(or ' stands ') for the spectators of a show ; nowhere does it mean
' stage scenery '. In the Curculio passage it cannot bear its primary
meaning : how could the ' show ' be blown down b y the wind ?
Evidently something collapsed and thereby spread panic among
the spectators ; w h a t could this have been unless the seats or
scaffolding on which they sat or stood ? W e should also consider
the words ' ut me c o n l o c a u e r a t ' , which surely describe how the
nurse set the little girl on a seat. T h e only reason for resisting this
argument is that w e seem to have proved too much ; in trying
to show that the R o m a n spectators sat on wooden seats, we seem to
h a v e shown that the Greek spectators sat on wooden seats too.
W e are so accustomed to picture the audience of Sophocles
sitting on rows of stone seats that w e find it hard to realize that the
evidence is against such a view. Even Haigh, after proving to his
o w n satisfaction {Attic Theatre, pp. 83-6) that the seats in the theatre
of Dionysus were not of stone but of wood until long after the end
of the fifth century, is apt to forget his conclusions and slip back to
the earlier view. I doubt the relevance of the archaeological
evidence ; while Puchstein tries to date the stone seats as far back
as the last years of the fifth century, Dörpfeld, supported by Haigh,
assigns them to a date not earlier than the middle of the fourth
century ; and Haigh suggests that they owe their origin to the
reconstruction of Lycurgus, w h o was minister of finance between
338 and 326. T h e inscriptions on the seats seem to belong to the
age of Hadrian. I turn to the literary evidence.
T h e only word I can find in fifth-century Greek for ' seats in the
1 cum consternatum ruinae metu populum retinere . . . nullo modo posset,
transiit e loco suo atque in ea parte consedit quae suspecta maxime erat.
 APPENDICES 237
theatre ' is Ικρία (Ικρια).1 T h e general sense of this word appears
to be ' planking resting on uprights ' ; thus in Homer it denotes
' decks ' ; in Herodotus v . 16 it is used of the platforms of lake-
d w e l l e r s (Ικρία ίττι σταυρών υψηλών εζευγιaeva iv μεστ) εστηκε rrj λίμντ)}.
Sometimes in mediaeval authors it seems to be used of upright
poles ; nowhere have I found it used to denote an object made
of any material other than wood. O f its use for * seats in the
theatre ' w e have two fifth-century examples :
A r . Thesrn. 395 :
ωστ' (νθΰε (ϊσιόντα από των Ικρίων
ΰποβλεπουσ' ημά$

(of the men coming home to their wives from the theatre) ;
Cratin. Incert. 53 (Meineke) :
χαΐρ', ω μΐγ' άχρ(ΐόγ(Χω! όμιλε rati επίβ&αΐ!,
rrjs ήμετεραί σοφίαs κριτήг Λριστι πάντων'
(ΰδαίμον' ετικτέ σε μητηρ Ικρίων ψ·όφησιν

(where the poet is addressing the Athenian audience, w h o delight

in the ' clatter of the benches spectaculorum strepitu, as Meineke
translates it).
I have found no other word in Greek drama for ' spectators'
seats ' (as opposed to the ' seats of honour ' of the πρόεδροι).
For further information about these theatrical Ικρία w e must
turn with due caution to the mediaeval lexicographers :
PhotlUS ( n i n t h c e n t . ) , S.V. Ικρία : τά «и ту άγορα. αφ' ων ίθεωντο
τους Διονυσιακούς άγώναϊ πριν η κατασκεαυσθηναι το iv Διονύσου θεάτρον.
S u i d a s (с. ΙΙΟΟ A.D. ? ) , S.V. 'ικρία : όρθα ζνλα . . . και τά των
θεάτρων . . . έπι ^ύλων -γαρ εκάθηντο. πριν γενηται το θεάτρον, £ύλα
εδίσμευον και ούτως ϊ θεωρούν. 'Αριστοφάνης ®εσ μοφοριαζοΰσαις κ.τ.λ.
(as already quoted.)
I d . s . v . Α ι σ χ ύ λ ο ? : . . . φυγών δε (Is Siκελίαν δια. το πεσεΐν τα Ικρία
επιδεικνυμενου αυτοί . . .
I d . S.V. Πρατίνας : . . . άντηγωνίζετο δε Α ϊιτχνλω . . . έπι της ό
όλυμπιάδος . . . επιδεικνυμενου Sc τούτον συνέβη τα Ικρία, εφ' ων εστηκεσαν
οΐ θεαταί, πεσειν, και «κ τούτου θεάτρου ΐνκοδομήθη Άθηναίοις.
G r a m m . Bekker. Anecd. p. 354, 25, s.v. αιγείρου θεα: Άθήνησιν
αίγειρος ην, ης πλησίον τα Ικρία εττήγννντο εις την θεαν προ του θεάτρον
γενεσθαι. οΰτω Κρατίνος.
S u i d a s , e t c . , S.V. άπ' αιγείρου θεα : η άττο των εσχάτων αίγειρος yap
επάνω ην του θεάτρον, άφ' ης οί μη εχοντες τόπον έθεώρουν.

See also Hesychius, s.v.

Haigh's explanation of these conflicting statements is, briefly,
that at the beginning of the fifth century the Athenian audience sat
on wooden benches rising in tiers one above the other, and resting
1 πρΰτον ξϋλον seems to be used only of the front bench at the Pnyx or in the
courthouse.
238 THE ROMAN STAGE

on wooden supports ; that these ικρία collapsed during a per-

formance in 499 B.C. ; that the Athenians then built, not stone seats,
but an earth embankment to support the wooden seats, which were
still used : this earth embankment took the place of the ' " ikria "
or wooden supports on which the seats had previously rested ' ;
that, finally, the stone seats were constructed towards the end of
the fourth century. This is a desperate attempt to reconcile the
statements of Suidas with Dörpfeld's discovery that ' the earth
foundations of the present auditorium . . . consist of two layers.
T h e upper one belongs to the fourth century, as is shown b y the
fragments of pottery embedded in i t ; the lower one is proved b y
similar evidence to be not later than the fifth But Haigh's view
that the Ικρία (which he wrongly takes to mean the supports of the
seats) were done a w a y with soon after 499 is contradicted b y the
passages I have quoted from Cratinus and Aristophanes, which
speak of Ικρία as still in use much later. So prominent a feature
of the theatre were they in 410 that ' to come from the Ικρία ' is
used for ' to come from the theatre '. T h a t they were still m a d e of
wood is indicated, I think, b y the phrase of Cratinus ' the rattle of
the benches ' ; stone seats would make but little ψόφησις. W e
m a y compare Pollux's remark s.v. ·πτ€ρνοκοπίΐν : τό μ,ίντοι та ίδώλια
τ α ΐ ϊ πτίρναις κατακρονίΐν τττ. iKeyov' Ιποίονν 8e τοντο οπore τίνα ΐκβάλοκν.
Haigh explains : ' T h e Athenians had also a peculiar custom of
marking their disapproval of a performance by kicking with the
heels of their sandals against the front of the stone benches on which
they were sitting '. N o t so peculiar a custom, if the benches were
of wood ! H a i g h has wantonly inserted the word ' stone against
his own theory and against c o m m o n sense as well.
W e notice that Suidas in one place says that the spectators
stood on the ίκρία and in another that they sat on them. Some of
of our ' grand stands ' at football grounds do not allow the occupants
to s i t ; they merely give elevation. These are, of course, intended
for less wealthy folk. Perhaps w e should picture the highest and
most remote Ικρία as being of this uncomfortable type. T h a t the
general public sat, at least by the time of Aristophanes, is indicated
b y Av. 793-6, where the birds suggest that if one of the spectators
should espy in the part of the theatre reserved for members of the
βονλη the husband of the lady with w h o m he is carrying on an
intrigue, a pair of wings would enable the lover to visit his mistress
and then resume his seat in the theatre (avtfis αν καθίζΐτο).
A b o u t the end of the fourth century w e find several references to a
seated audience : cf. Heges. Adelph. 29 :

ϊτολΧούί f'yi) σφάδρ' ο'δα των καθήμενων.

But the evidence suggests that the audience during the fifth a n d
early fourth century were accommodated on tiers of w o o d e n
scaffolding which itself rested on a n artificial earth bank ; most о
 APPENDICES 239
them sat, but the less fortunate people on the topmost tiers had to
stand ; others, again, were perched on trees. I f w e add to our
picture the itinerant vendors of wine and edibles (Philoch. ap.
A t h . p. 464E : παρά. St τον άγωνα πάντα οίνος αντοΐί ωνοχο(ΐτο και
τραγηματα π'αρΐφίρίτο) and the din of the wooden benches, w e
have a scene reminiscent of D e r b y D a y .
T h e evidence points to the reconstruction of the theatre as the
result of a collapse of the scaffolding. I suggest that this is a correct
view, but that the date of this particular accident (there m a y have
been m a n y like it) and of the consequent reconstruction has been
placed far too early. T h e only large-scale reconstruction of which
w e know was the building of stone seats, perhaps towards the end
of the fourth century. T h e original of the Curculio m a y well have
been produced about this time, and the reference to a collapse of
seats m a y have been a reference to a real event.
T h e r e is therefore nothing, it would appear, to prevent our
supposing that Cure. 644-7 w a s translated by Plautus from his Greek
original. I f w e grant, then, that the word for ' seats at a s h o w '
was already established in Latin at that time, it becomes more than
ever difficult to resist the conclusion that the plays of Plautus were
written for presentation before a seated audience.
It is now becoming recognized that the external evidence against
seats in the time of Plautus is weaker than was supposed, and the
whole argument based on such supposed evidence is demolished b y
the latest Pauly-Wissowa, s.v. iheatrum.1 Perhaps the internal
evidence cited in this article will serve to drive a nail in the coffin of
Ritschl's theory and help to re-establish confidence in the prologues
of Plautus as coming, in the main, from his own pen.
1 Pauly-Wissowa does not discuss how the curious belief arose that the Roman

authorities, while officially arranging for theatrical performances, should for

puritanical reasons have forbidden the spectators to sit. I suggest (1) that the
wooden scaffolding or benches of Plautus' day left no visible trace for future ages
to contemplate ; (2) that some race-conscious Roman antiquarian, rather hurt
at finding imposing tiers of stone seats in the theatres of Greek towns, tried to explain
the absence of similar remains in Rome in a way which would flatter his country-
men's pride in the mos maiorum.
 APPENDIX В

SIDE-ENTRANCES AND ΠΕΡΙΑΚΤΟΙ IN THE

HELLENISTIC THEATRE
(Classical Quarterly, Vol. xxxii. pp. 205-10.)

T H E greatest confusion prevails among modern writers as to the

use of the side-entrances in New Comedy and its Latin derivatives.
T h e statements on this subject made by editors and others, whether
confident or hesitating, differ widely from one another, and are
seldom supported by any real consideration of the ancient evidence.
In 1933 Professor M a r y Johnston published a careful treatise,
entitled Exits and Entrances in Roman Comedy (W. F. Humphrey Press,
New York), in which she discussed the internal evidence afforded
by the Latin plays, and came to the conclusion (page 151) t h a t ' on
the stage of the Roman theatre the side-entrance to the right of the
spectators was used for entrances and exits of characters from and
to the city and the forum, and that the side-entrance to the left
of the spectators was used for entrances and exits of characters
moving from and to the port and foreign parts, and, probably,
from and to the country as w e l l ' . With regard to Greek usage,
Professor Johnston was content to accept the orthodox view ' that
the side-entrance (parodos) at the spectators' right led to the harbour
or the market-place and that at their left into the country, since the
scene was regularly placed in Athens and since these were the actual
topographical relationships in the Athenian theatre ' (Flickinger,
page 208). Her conclusion, therefore, involved a discrepancy
between Greek and R o m a n usage as far as the harbour was
concerned.
T h e possibility of such a discrepancy has been admitted by
other writers (e.g. Flickinger, page 234) ; but it raises certain
difficulties. A t line 461 of the Captiui Ergasilus enters on the empty
stage. He has come from the forum (cf. lines 478 and 490) and is
on his way to the harbour (line 496), in which direction he departs
after line 497 (cf. lines 768 ff.). He had no intention of calling on
Hegio, whose offer of a cena aspera he regards merely as a last
resource. T h e only pretext for his appearance upon the stage must
therefore be that he has to cross it in order to get from the forum to
the harbour. This is perfectly natural if Professor Johnston is right
as to the Roman convention. But how are we to visualize Ergasilus'
movements in the Greek original of the play ? Had Ergasilus some
errand at Hegio's house, all reference to which has been suppressed
by Plautus ? O r is the whole scene an insertion by Plautus ?
240
 APPENDICES 241
If our solution of every difficulty in Plautus is to take the
traditional form of assuming that he has tampered with his original
in some way, we shall inevitably involve ourselves in a hopeless
tangle of subjective argument. Yet either of the alternatives
suggested seems preferable to assuming that, in the Greek play, the
parasite walked on, delivered his monologue, and then walked off
by the same side-entrance.
T h e Roman dramatists must have taken over their use of the
side-entrances from Greek sources ; not merely literary sources,
but the usage of the Hellenistic theatres of Magna Graecia. Why,
then, should they have heaped up difficulties for themselves by
modifying Greek convention in one vital point ? Few will accept
the view of Fensterbusch (Philol. LXXI, 1926) and Kelley Rees
{A.J.P. 32, 1 9 1 1 , page 400) that Roman topography had something
to do with the matter. Are we really so certain as to what Greek
usage was ?
T h e relevant passages in ancient authors appear to be :
(a) Vitruvius, V . vi. § 8 :—ipsae autem scenae suas habeant
rationes explicatas ita uti mediae ualuae ornatus habeant aulae
regiae ; dextra ac sinistra hospitalia ; secundum autem spatia ad
ornatus comparata, quae loca Graeci irtpiaKrovs dicunt ab eo
quod machinae sunt in eis locis uersatiles trigonae, habentes
singulae tres species ornationis, quae cum aut fabularum mutationes
sunt futurae, seu deorum aduentus cum tonitribus repentinis,
uersentur mutentque speciem ornationis in frontes ; secundum ea
loca uersurae sunt procurrentes, quae efficiunt una a foro altera a
peregre aditus in scenam.
(b) Pollux, I V . xix. §§ 125-7 :—παρ' ΐκάτΐρα. δί των δυο θυρών των
7 Γ £ ρ ί την μίσην ά λ λ α ι δ υ ο α ν c t e v , μία ίκατίρωθΐν, προς a s αί πΐρίακτοι
συμπΐπηγασιν, η μίν δ ε £ ι α τα ί £ ω π ό λ ί ω ς Βηλοΰσα, η δ ' αριστΐρα τα ίκ
πόλ.(ως, μάλιστα τα ΐκ λ,ιμίνος. και θίονς те θαλαττίονί (πάγα, και πάνθ'
οσα Ιπαχθίστΐρα όντα η μηχανη φίρίΐν ά δ υ ν α τ ί ΐ . (I δ* ίπιστραφίΐίν α [
πτρίακτοι, ή 8ΐζια μίν άμείβΐΐ τόπον, άμφότιραι δί χωράν υπαλλ,άττουσιν.
των μέντοι παρόδων η μίν δΐζια άγρόθfv η ΐκ λιμΐνος η ίκ ι τ ό λ ί ω ϊ a y t i ,
οι δ ί a X X a ^ o ö i i ' πΐζοί άφικνόνμ(νοι κ α τ ά την Ιτίραν ΐΐσίασιν.
Pollux further tells us that the keraunoskopeion was a ' high
periactus ' (§ 130) ; he couples it with the bronteion or, ' thunder-
maker ' ; immediately afterwards he describes the theologeion (on
which the gods manifest themselves) and states that the κατα,βλήματα
were painted curtains or boards which were ' dropped ' (κατφάλΚπο)
on the periacti, and showed such views as a mountain, a river or
the sea, or whatever else might be suitable to the play (§ 1 3 1 ) .
There is also a passage in the Vita Aristophanis (quoted by
Haigh, А. Т., page 194, note) to the effect that, if the chorus entered
4
as from the city ', it used the ' l e f t ' : if from the country,
the right.
I can find no reference to any other authoritative texts on the
242 THE ROMAN STAGE

subject of side entrances, or for that matter on the nature and use
of the periacti. T h e passage in the Vita deals, of course, with the
entry of the chorus into the orchestra by means of the orchestral
πάροδοι. I am assuming 1 that there was a stage in the Hellenistic
as in the Roman theatre, and that it is to this stage that
Pollux and Vitruvius are referring ; but the author of the Vita
may fairly be quoted as evidence for the view that on the stage, as
in the orchestra, the town and country entrances were opposite to
each other. But any fair-minded reader of Haigh's note, referred
to above, will agree that the expressions ' r i g h t ' and ' l e f t ' in both
Pollux and the Vita cannot be understood without further evidence.
Such evidence does not exist, so far as the Hellenistic theatre is
concerned. T h e confident statements made by modern writers
about the influence of the topography of Athens on the growth of
the convention which we are discussing are entirely in the air. 2
T h e origins of a dramatic convention should be sought in
dramatic, not in topographical, conditions. T h e scene of most
New Comedy plays was a street—usually in Athens ; characters
very frequently come on from the forum or go off towards the forum ;
the harbour is mentioned rather less frequently ; still less frequently
is reference made to the country. Professor Johnston calculates
(pp. 38 ff.) that, of the twenty-six Latin plays, twenty-four are set
in town ; of these five require forum, harbour and country ; thir-
teen require forum and harbour only ; three require forum and
country only ; in three plays of Plautus only the forum (town)
is actually required (but these three plays are (a) the Asinaria, in
which ' the entrances are confused, and the movements of the
characters cannot be followed satisfactorily' ; (b) the Casina, in
which frequent reference is made to the country, though no one
actually goes there ; (c) the Persa, in which some pretence is made
of using the harbour entrance) ; only two plays, the Rudens and the
Heauton, are set in the country. If we assume that the harbour lies
in the same direction as the forum, we leave the opposite entrance
unused in some two-thirds of the plays. If, on the contrary, we
follow Professor Johnston in opposing the forum to both country
and harbour, we shall find that both side-entrances are used,
actually or in appearance, in all the plays with a city setting, with
the exception of the Asinaria. But in the plays with a country
setting circumstances may be quite different. It is clear that in the

1 But Dr. Pickard-Cambridgc has now abandoned his early belief in the raised

stage, even for Menander's theatre (Theatre of Dionysus, 1946, p. 165). I accept
his view, which in no way affects my argument.
* They seem to me founded on fallacy. I do not believe a theatrical audience
is so conscious of topography as the theory implies. And if the Greeks were, how
can we explain the fact that New Comedy was produced in various theatres all
over the Greek world, with quite different topographical settings ? Or that, as
this article shows, the convention varied according to where the play was supposed
to take place ?
 APPENDICES 243
Rudens both the town and harbour of Gyrene are thought of as
lying in the same direction ; on the opposite side is the beach.
In the Heauton the town as a whole lies on one side ; there is no
specific mention of the open country, which we may presume lies
on the opposite side, but it can well be imagined that when Clitipho
is sent off for a walk (line 590) in order to get him out of the way, it
is towards the open country that he directs his footsteps.
If we now turn to the remains of Menander's plays we find a
higher percentage of country settings. T h e scene of the Hero is a
country district near Athens ; we hear that a party of hunters will
arrive from the town (Fr. Sabb., Capps, page 21) ; we are also led
to expect that Laches will arrive from Lemnos (line 65), presumably
via the Piraeus and the town. ' T o w n ' and ' harbour ' would, in
such circumstances, naturally be thought of as lying in the same
direction ; on the opposite side would be the open country, the
farms of neighbours, etc. T h e Epitrepontes is also set in the country
near Athens ; we get the general impression that ' t o w n ' and
' country ' lie on opposite sides ; there is no use made of the
4 harbour'. T h e scene of the Perikeiromene is a street in Corinth ;
opposition of market-place and country gives a natural setting ;
there is no mention of the ' harbour '. T h e scene of the Samia is
given by Capps as a street in Athens ; the use of the forum entrance
is probably indicated in line 69—IK rrjs [άγορά?] ; there is no
mention of either harbour or country. T h e Gorgias seems to be set
in the town (cf. line 79) ; Davus comes in from the country
(Allinson, line 32).
So far the results of our inquiry are meagre. But in the
Citharistes we have evidence of a more interesting nature.
Moschion, while on a visit to Ephesus, had seen the procession
of freeborn maidens bringing offerings to ' Diana of the Ephesians ',
and had fallen in love with one of them, daughter of Phanias, a
harp-player, who lived next door to his father Laches in Athens.
It seems that Moschion has now married this lady ; when the play
opens he has just returned to Athens, and is expecting his wife by
another ship. This ship has not yet arrived. T h e opening scene,
as given by Allinson, shows Moschion (?) on his way from the har-
bour to the market-place, talking to a friend, and followed by slaves
carrying luggage. He has sent a messenger to summon his father
from the country. He is very worried that his bride's ship has not
yet arrived, and is pouring his troubles into the sympathetic ear of
his friend, and proposes to finish the story as they go together to
the market-place. As he passes his father's house he issues a hasty
order to his slaves : ' Let someone take these things into the house
out of sight as quickly as possible '. Moschion and the friend
depart for the market-place ; Laches enters from the country. He
is mystified at the summons he has received from his son. He is
going to look for Moschion indoors ; if the lad is not there, he will
244 THE ROMAN STAGE

go on to seek him at the market-place. Laches goes into his house ;

Moschion returns from the market-place, wondering whether his
father has yet arrived. Laches appears : Moschion greets him
with warmth, and embarks on the story of his love affair.
In this summary I have followed Allinson's reconstruction
closely (cf. also Sudhaus) ; if he is right, then we have a setting
which can only be visualized on the assumption that the harbour
entrance was opposite to the market-place entrance. Otherwise
how can we account for Moschion's appearance on the stage, in
front of his father's house, when hurrying from the harbour to the
agora ? It is true that the harbour is not mentioned in the extant
fragments of the scene ; but the young man's reference to his
anxiety at his wife's having failed to arrive and his fear lest some
misfortune has befallen her at sea, coupled with his order to the
slaves, seems to make it clear that he himself has only just arrived
from abroad, and has not yet entered his father's house.
I have failed to find any evidence in the remaining fragments of
Menander. I now turn to the statements of Vitruvius and Pollux,
quoted above. The concise words of Vitruvius ' una a foro altera
a peregre ' agree with the view I have expressed, so far as they go ;
the trouble is that Vitruvius has not given us enough detail. Pollux
tells us that the ' right-hand parodos' leads ' from the country or
the harbour or the city This cannot be right; emendation,
however, is impossible without assuming the answer to the very
question which we are discussing. In each of the passages quoted
we find the side-entrances closely connected with the periacti.
Combining the statements of Vitruvius and Pollux, we find that
these devices were revolving three-sided stands, set one at either
end of the permanent back-scene, between the side-doors and the
side-entrances. Each of the three sides of a periactus displayed a
different scene. The right-hand periactus, according to Pollux,
showed ' the region outside the city ', the left-hand ' the things
from the city, especially the things from the harbour ' : it also
introduced sea-gods, and whatever was too heavy for the μηχανή.
To revolve the right-hand periactus alone indicated a change of
τόπος, to revolve both periacti a change of χώρα. Vitruvius, on the
other hand, tells us that the revolution of the periacti denoted
either a change of play or the arrival of a deity, accompanied by
sudden peals of thunder.
None of the descriptions which I have read in modern works
(including the 1934 and 1937 P.W.) makes any real attempt to
deal with this curious and apparently contradictory evidence ;
even the sober Haigh ( A . T p a g e s 197-9) is both inadequate and
fanciful. All writers seem to agree in regarding the periacti as
appliances for changing scenery, ' the only appliances for changing
scenery that are mentioned by the ancient Greek writers ' (Haigh).
But New Comedy does not appear to have allowed for air· "bange
 APPENDICES 245
of setting within the course of any one play. There is no evidence
for such a change either in Plautus, or in Terence, or in the
fragments of Menander. From the beginning to the end of any one
play the audience were confronted with the usual three-door
back-scene. And what would such a back-scene have to do with
the Καταβληματα which Pollux says were placed on the periacti,
showing a mountain, a river, or the sea ? How artificial, too,
would be the ' curious conventional custom ' (Haigh) whereby one
periactus was turned to denote a ' slight' change of scene—' merely
from one part of the same district to another whereas ' when the
action was transferred to an entirely new district, then both the
periacti were turned round, and the scenery was changed at each
end '. Such an account can have been penned only by a writer
who was thinking chiefly in terms of fifth-century drama. We can
understand Haigh's conclusion that ' it must have been chiefly in
the intervals between successive plays that the periacti were em-
ployed ' ; in fact Vitruvius has told us that they were so used (cum
fabularum mutationes sunt futurae). But the periacti seem by their
nature to have been designed for rapid alteration. And what of
Vitruvius' other remark that they denoted the arrival of gods, and
of Pollux's reference to the introduction of sea-gods ' and objects too
heavy for the mechane ' ? ' It is possible that, of the two sides of the
periaktos which were out of sight of the audience, one contained a
small ledge or balcony, on which the sea-god took his stand. As
the machine rolled round, he would come suddenly into v i e w '
(Haigh).
Let us look again at Vitruvius and Pollux. We observe :—
(a) that both writers refer to the periacti in close connexion
with their accounts of the side-entrances ;
(b) that both writers speak of ' arrivals '—deorum aduentus,
6(οι'ς . . . ΰαλαττιονς indyti—as in «оше way connected with the
use of the periacti; with which evidence we may couple the ίκ
of Pollux—' the things from the city ', ' the things from the harbour ;
(c) that Pollux's description of the scenes shown on the periacti—
" the region outside the city ' the things from the city ', ' the
things from the harbour '—is strongly reminiscent of the use of the
side-entrances to denote arrivals from or departures to ' country
' town ' and ' harbour ' ;
and when we further remember that, while there were only
two side-entrances, there were three conventional significations to be
shared between them, we shall, I think, be forced to the conclusion
that the function of the periacti was to indicate to the audience the
conventional significance to be attached to the side-entrances at
any given moment in the play.
Let us suppose that the scene is the usual street in Athens. At
the back are the usual three doors, denoting three houses, or two
houses and a temple or the like. T h e significance of these doorways
17 . 1
246 THE ROMAN STAGE

will be made clear to the audience (a) in the prologue, if there is

one ; (b) b y frequent references in the course of the play. More-
over these doorways in the back-scene will retain their significance
unaltered throughout the course of the play. O n either hand are
the side-entrances. T h e i r function differs from play to play ; it m a y
be altered more than once within a play. T o avoid the necessity of
issuing frequent reminders to the audience, use is made of the
periacti, set close to the side-entrances ; one m a y present a view of
the market, the opposite one a rural scene (a river, a mountain,
etc.). A n arrival from abroad is anxiously expected ; suddenly
one of the periacti swings round, presenting to the audience a
picture of the sea, the harbour, a ship, a dolphin or the like. This
is a change of τόπος. T h e play has ended ; another, with an
entirely fresh setting, perhaps in a foreign town, is to be brought on.
It will still require the three doorways ; but a revolution of both
the periacti will give these doorways the appropriate framework ;
this is a mutatio fabulae, a change of χώρα. A deity is to manifest
himself, with appropriate thunder and lightning. A b o v e the stage
is set the ' high periactus also called the ' flash 5 (πίρίακτος υψηλή,
κίραυνοσκοπΰον) ; on one, two or all of its sides (we m a y
suppose), a fiery streak is painted on a dark background. The
' high p e r i a c t u s ' turns—or perhaps whirls ; the βροντάον, or
' thunderer rattles, and the deity appears b y the celestial side-
entrance, the μηχανή or θίολογΐΐον. Such an entry w e m a y
suppose was made b y Jupiter at the end of the Amphitruo :—
strepitus, crepitus, sonitus, tonitrus : ut subito, ut prope, ut ualide tonuit !
(line 1062);
ardere censui aedis, ita tum confulgebant (line 1067) ;
and finally : —

sed quid hoc ? quam ualide tonuit ! di, obsecro uestram fidem ! (1130)

and the god appears in majesty, delivers his decree, and departs
whence he came (ego in caelum migro, line 1143). A sea-god,
however, cannot well appear from the s k y ; to herald his arrival
one of the side periacti will revolve to show a view of the sea, and
the god will walk in b y the neighbouring side-entrance. Groups
of deities, collectively too heavy for the μηχανή, will enter in some
such w a y ; ghosts and infernal deities will appear as it were from
the depths of Hades ; the varieties of the detachable καταβλήματα
are unlimited.
I have made no attempt to interpret the terms ' r i g h t ' and
' l e f t ' as used by Pollux. O n any view his account is confused, and
he is drawing on different and imperfectly understood authorities.
Professor Johnston's independent study of the L a t i n plays has led
her to put the ' f o r u m ' entrance to the spf^t®^«' right u«.
 APPENDICES 247

' h a r b o u r ' and ' c o u n t r y ' entrances to the spectators' left. 1 T h e

only qualification I would add is that this arrangement should be
limited to plays with a ' t o w n ' setting ; a setting in the country
m a y well have necessitated placing the harbour in the same
direction as the town, as is, in fact, the case in the Rudens, and
perhaps in the Hero ; our arrangement of the Rudens will depend
on our interpretation of ad dexteram in line 156 and again in line 254.
W h a t I a m most concerned with is to show that, so far from there
being any discrepancy between Greek and R o m a n practice, there
is every reason to suppose that the Romans adopted unchanged the
use of the side-entrances which they found prevailing in the
theatres of M a g n a Graecia.
1 This goes well with Pollux's statement that it was the ' right-hand'
periactus (i.e. that to the spectators' left) which indicated ' the things outside the
city,' and which also was revolved to show a change of ' place '. The other
periactus, displaying a view of the city-centre, would not need to be turned so
frequently.
 APPENDIX С

T H E A N G I P O R T U M AND R O M A N D R A M A
(Hermathena, Vol. xxviii. pp. 88-99.)

T H E word angiportum is commonly taken to mean ' alley or even

1
blind alley ' cul-de-sac '. As to its use on the Roman stage, the
conventional view is given by Professor Mary Johnston {Exits and
Entrances in Roman Comedy, 1933, page 15) :—' An alley or passage
(1angiportum) was supposed to lead back from the street between
two houses '. But Mr. P. W. Harsh (Classical Philology, vol. XXXII,
No. 1, pages 45 if.) argues that' alley ' is a misleading translation ;
that the word simply means ' s t r e e t a n d may be used in comedy
even of the street upon which the houses front {Pseud. 961), though
elsewhere it is ' sometimes a more secretive place than the stage
i t s e l f a n d is ' apparently thought of as running behind the houses
portrayed on stage and that even in Terence, Adelphi 578, the
sense ' cul-de-sac ' is entirely dependent on the additional words
поп peruium. The discrepancy between these views seems to justify
some attempt to review the evidence so far as stage usage is
concerned.
The usual word in comedy for the street on which the houses
front is platea, of which Harsh lists ten examples ; uia is also used
' in a general way for the thoroughfare on which the characters
stand (cf. С as. 856 ; St. 606, etc.) '. In Ps. 1234-5 w e find uia
apparently opposed to angiporta ; Ballio, leaving for the forum,
remarks to the spectators :
nunc ne exspectetis dum hac domum redeam uia ;
ita res gestast: angiporta haec certum est consectarier.
Ballio does not, in fact, appear again, and we must suppose that he
returned to his house by the back entrance. Here, then, angiportum
would be the back street upon which the back entrance opened ;
the plural may be used merely as an equivalent of the singular,
or may include the other streets through which Ballio would pass
on his way from the forum to the back of his house. Perhaps,
however, uia depends on the addition of hac for its meaning here,
and would not by itself bear the same meaning as platea. To quote
Harsh (page 49, note 10) : ' uia is also used in a general way for
the thoroughfare on which the characters stand (cf. Cas. 856 ;
Stich. 606, etc.) Turning to the passages here cited, we find :
acceptae bene et commode eximus intus
ludos uisere hue in uiam nuptialis {Cos. 855-6)
248
 APPENDICES 249
' After our nice, enjoyable entertainment inside, here we are out on
the street to watch the wedding games ' (Nixon's translation).
Here uitt means ' the open s t r e e t a s opposed to the privacy of
indoors.
GE. non tu scis quam—ecflictentur homines noctu hie in uia ?
Рам. tanto pluris qui defendant ire aduorsum iussero (St. 606-7).

Here, again, there is a demonstrative to make the meaning of uia

clear, and the general sense is, as before, ' in public ' in the open
street', as opposed to the safety of indoors.
The passages cited by Harsh for the use of platea (apart from
Вас. 632, 1 where it is merely a proposed emendation) are :
ne quis in hac platea negoti conferat quicquam sui (Capt. 795).
suaui cantu concelebra omnem hanc plateam hymenaeo mi (Cas. 799).
parasitum tuom
uideo currentem ellum usque in platea ultima (Cur. 277-8).
sterilis hinc prospectus usque ad ultumamst plateam probe (Mil. 690).
sed quis hie est qui in plateam ingreditur . . . ? (Trin. 840).
sed quis hie est qui hue in plateam cursuram incipit ? (Trin. 1006).
in hac habitasse platea dictumst Chrysidem (An. 796).
ilia se(se) interea commodum hue aduorterat
in hanc nostram plateam (Eun. 3 4 3 - 4 ) .
si te in platea offendero hoc post unquam . . . (Eun. 1064).
sed hie quis est senex quem uideo in ultima platea ? . . . (Phor. 2 1 5 ) .

In all these passages the sense is ' this s t r e e t a n d there is no

suggestion of a contrast with ' indoors'. Nor is platea here used in
the sense o f ' way ' or ' means of getting to any particular place '.
All the examples of uia cited in Lodge's Lexicon Plautinum would
seem to come under one of these three heads—(x) ' in public ',
(2) ' way ', (3) ' highroad' : cf. Cas. 675 : de uia in semitam
degredere. It would appear, then, that uia and platea are not
interchangeable terms so far as comedy is concerned.
The conventional view of the typical stage setting is that the
two side doors (hospitalia) represent the entrances of two houses,
while the central door (ualuae regiae) is the entrance to an ' alley'
separating the two houses, and would, as such, presumably stand
open during the whole of the play. There are obvious objections
to such an arrangement. Why should the principal door be
assigned so secondary a function ? If it remains open, would not
the spectators be able to see inside, and, if so, what would they see ?
The interior of the greenroom ? A blank wall ? If the door is
usually shut, we have the absurdity of an actor being forced to open
it in order to enter an alley. Nowhere is there any reference to the
opening of a door on such an occasion.
We are forced to suppose that the door stood permanently open
1
Here I differ entirely from Dalman.
250 THE ROMAN STAGE

during the course of a play requiring such a setting as w e are

discussing, and that, a few feet inside, a screen stood displaying a
view of an alley. Actors w h o were supposed to leave the platea b y
means of the angiportum would enter the doorway, step behind the
screen, and so disappear from the view of the audience. If this was
the arrangement adopted, w e still have to ask ourselves :
(a) W h a t happened w h e n (as in the Pseudolus) the setting
included three houses as well as an angiportum (cf. line 1235) ?
(b) W h y is the angiportum, in fact, so seldom used, and for so
limited a range of services ? T o quote Professor M a r y Johnston
{Exits and Entrances, page 37) : ' N o entrances are made through
an angiportum at any time, so far as our material shows. I n most
passages where the angiportum is referred to at all it is mentioned in
accounting for movements of the characters when they are unseen
and off the stage, if it is not convenient to bring them on the stage '.
(ιс) W h y is there no reference to this use of the central doorway
in Vitruvius or Pollux ?
(d) W h e r e is the ' alley ' supposed to lead ? W h a t is its relation
to the garden which is occasionally mentioned in connexion with
i t ? (Cf. As. 741-2 ; Most. 1044-5).
(e) W h y should angiportum be translated ' a l l e y ' in certain
passages, when elsewhere it clearly means ' s t r e e t ' — o f t e n , it is
true, with a suggestion of secrecy, but still' street' and n o t ' alley ' ?
For the relation of the angiportum to the hortus and the posticum,
or ' back-door ', certain passages in the Mostellaria are of particular
importance. In line 928 Theopropides sends T r a n i o to the country.
Tranio, as he departs, remarks to the audience :

nunc ego me iliac per posticum ad congerrones conferam.

A s Theopropides remains on the stage after Tranio's departure,

it is difficult to suppose that Tranio, instead of leaving the stage b y
the ' country ' side-entrance, should slip back to the central door
and so into the angiportum. T o execute such a manoeuvre under
the eyes of his master would have been disastrous. W e must think
of him as actually going off the stage by t h e ' c o u n t r y ' side-entrance.
His next appearance is at line 1041. H e tells us (lines 1043-5) :

nam erus me postquam rus misit, filium ut suom arcesserem,

abii iliac per angiportum ad hortum nostrum clanculum,
ostium quod in angiportu est horti, patefeci fores,
eaque eduxi omnem legionem, et maris et feminas.

This is how the present writer would arrange the stage setting :

angiportum ostium horti angiportum

hortus
posticum
L. (rus) House of Simo Ноше of Theopropides R. (forum)
 APPENDICES 251
Tranio goes off the stage at line 932 by the left-hand side-entrance,
but by voice and gesture informs the audience that he intends to
make his way round to the ' back door '. He sets forth his move-
ments in more detail in lines 1043-6. He made his way from the
side-entrance to the street which ran at the back of the two houses,
entered from this back street the garden behind Theopropides'
house by means of the garden gate, got into the house of Theo-
propides by the back-door, and was then able to remove the
occupants through the back-door, the garden, the garden gate, and
so into the back street, and off to the forum, if that is where they
ultimately go. At line 1041 Tranio enters by one of the side-
entrances—perhaps R, as it seems more natural to imagine that
Philolaches and his friends would go to the town than to the
country or harbour. Line 1076 would seem to imply that by this
time Tranio was standing near the ' country ' side-entrance ; in
reply to Theopropides' question he remarks that the country-folk
are on their way to town, and that Philolaches will shortly arrive.
This would mean that during his monologue (1041-1062) he has
crossed the stage. Where does he hide when Theopropides appears
(line 1064) ? In the left-hand side-entrance ? This would seem
to be the only retreat available, if we abolish the ' alleyway and
would make his entry at line 1075 quite natural. 1
These conclusions may be tested with reference to the other
plays which make use of the back-entrance.
(Ι) As. 7 4 0 - 3 : ARG. L e o n i d a , curre, opsecro, p a t r e m hue orato u t ueniat.
L E . i a m d u d u m est intus. ARG. h a c q u i d e m n o n uenit.
LE. angiporto
iliac per h o r t u m c i r c u m iit c l a m , ne quis se uideret
h u e ire f a m i l i a r i u m : ne uxor resciscat metuit.

The scene of the play is a street in front of the houses of Demaenetus

and Cleaereta. At line 126 Demaenetus left for the forum. At
line 545 Leonidas and Libanus enter from the forum, where they
have met Demaenetus. The slaves talk to each other until Argy-
rippus and Philaenium come out of Cleaereta's house (line 585).
All four characters have since been on the stage. It would, there-
fore, have been impossible for Demaenetus to make his way from
the forum and across the stage to the front-door of Cleaereta's
house without being seen by Argyrippus. When informed by
Leonidas that Demaenetus has, in fact, found his way into Cleaereta's
house, Argyrippus naturally replies : ' He certainly didn't come
this w a y ' . Leonidas explains that the old gentleman, afraid of
being seen by any of the servants, who might tell his wife, has
' stolen round that way by the back street and through the garden
1 I now hold that there were no hiding-places on the stage. Tranio simply
steps back, and Theopropides is obliging enough not to see him. But see pp. 254-5
on the Phormto, where there is express reference to the side-entrance.
252 THE ROMAN STAGE

A t line 830 w e are shown Demaenetus in front or coming out of

Cleaereta's house. T h e stage setting will, therefore, be :
angiportum angiportum
hortus
L. House of Demaenetus House of CI. R . (forum)

T h e r e is nothing to show the relative position of the two houses,

but clearly Demaenetus did not enter the ' alley ' from the stage.
(2) Cas. 6 1 3 - 4 : a b i et aliud cura, ego i a m per h o r t u m iussero
m e a m istuc transire u x o r e m a d u x o r e m t u a m . —

W e may arrange the setting in this w a y :

(angiportum) (ostium) (ostium) (angiportum)
hortus Lysidami hortus Alcesimi
(posticum) (posticum)
L. (rus) House of Lys. House of Ale. R . (forum)

T h e two gardens would, of course, be separated by a roughly

built wall (maceria). T h e r e m a y be a gate in this wall, but as the
evidence of other passages rather suggests that there was usually пЪ
such gate, it seems safer to suppose that Alcesimus would send
M y r r h i n a through their own garden and into the angiportum and
so round to the garden of Lysidamus. In fact the mention of the
garden implies the existence of an angiportum, and though there is
no express reference to an angiportum in the play, its existence would
make the movements of the bridal procession much more intelligible.
A t 815-6 the procession comes out of Lysidamus' house, ostensibly
on its w a y to the country, though Lysidamus' real object is, of
course, to escort the ' bride ' to Alcesimus' house, which has been
cleared of occupants. T o enter b y the front-door in full view of
Lysidamus' own house would be a risky proceeding, and there is
no reference to any such entry. Clearly, the procession leaves b y
the ' country ' side-entrance, and w e are to think of them as making
their w a y round by the angiportum and garden into the back of the
house. A t 875 Olympio, the ' b r i d e g r o o m r u s h e s out on the
stage from the front-door of Alcesimus' house.

(3) Ep. 660-1 : EP. T h e s p r i o , exi istac per hortum, adfer domum
a u x i l i u m mihi,
magnast res.

Epidicus, shouting at the door of Chaeribulus' house, bids his fellow-

slave Thesprio go ' out through the garden there ' and so ' home '
to the house of their master, Periphanes.
(4) Merc. 1007 : iliac per h o r t u m nos d o m u m transibimus.

Eutychus has just invited Demipho indoors, and adds thatDemipho's

son is inside. Demipho accepts the invitation, and remarks that
he and his son will go home ' that w a y , through the garden
 APPENDICES 253
(In other words, the spectators must not expect to see them re-
appear ; the play is at an end).
(5) Pers. 444-6 : Tox. abi istac trauorsis angiportis ad forum ;
eadem istaec facito mulier ad me transeat
per hörtum. Do. iam hie faxo aderit. Tox. at ne
propalam.

Dordalus is to enter his house (istac) and thence to make his w a y b y

' cross-streets ' to the forum, where he will have the money tested
which Toxilus has handed him as payment for Lemniselenis ; at
the same time (eadem) he is to send Lemniselenis through the garden
into the house of Toxilus' master, thus avoiding publicity. A t line
448 Dordalus departs ; he returns at line 470, apparently by the
forum side-entrance ; in other words, he must have sent Lemniselenis
into the house of Toxilus' master before he went to the banker's.
T h e dramatist thus avoids having to bring Lemniselenis on the
stage at this point.
(5a) Pers. 678-9 : Tox. per angiportum rursum te ad me recipito
iliac per hortum.

Sagaristio is to pretend that he is a foreigner ; after entrapping

the unfortunate Dordalus he is to complete the deception b y
departing through the left-hand side-entrance, as though on his
w a y back to the harbour. Cf. lines 676-7 :

ubi argentum ab hoc acceperis,

simulato quasi eas prosum in nauem.

O n c e off the stage, he is to slip round b y the angiportum into the

garden and so rejoin Toxilus indoors. This plan is carried o u t ;
Sagaristio departs for the harbour at line 710, and reappears
(line 763) from inside the house of Toxilus' m a s t e r ; he is
accompanying Lemniselenis.

(6) Stichus 431-2 : amicam ego habeo Stephanium hinc ex proxumo,

tui fratris ancillam.
„ 437 : iam hercle ego per hortum ad amicam transibo
meam . . .
„ 449-52 : . . . est etiam hie ostium
aliud posticum nostrarum harunc aedium :
(posticam partem magis utuntur aedium)
ea ibo opsonatum, eadem referam opsonium :
per hortum utroque commeatus continet.

T h e stage-setting is perhaps as follows :

(1angiportum) (angiportum)
hortus /tortus
ostium posticum
L. House of Epignomus House of Pamphilippus House of Antipho R,
 254 THE ROMAN STAGE

T h e r e is no specific reference to the angiportum, but its presence is

necessary to enable Stichus to make his w a y from the garden of
Epignomus' house to market and back to the house of Pam-
philippus, from the front-door of which he appears at line 641.
T h e r e is another reference to the garden entrance at line 614,
where Pamphilippus declares that he will use it in order to make his
w a y to Epignomus' house for supper. Stephanium must also make
frequent use of it, as she is cooking in both houses. W e are left
in some doubt as to whether there is direct communication between
the two gardens.

(7) Pseud. 960-2 : Simia, pretending to be entering from the harbour,

remarks :
habui numerum sedulo ; hoc est sextum a porta proxumum
angiportum, in id angiportum me deuorti iusserat;
quotumas aedis dixerit, id ego admodum incertum scio.

A n d at line 971, addressing Ballio, he asks :

ecquem in angiporto hoc hominem tu nouisti ?

In these passages angiportum, is used of the street on which the houses

front. Ballio's house is situated, not in an ' a l l e y b u t beside the
houses of Simo and Callipho. Pseudolus, entering from the forum
(line 905) with Simia, indicates Ballio's house with the words
(line 952) tertium hoc est.

'Я) True. 303-4 : quid maceria ilia ait in horto quae est, quae in noctes
singulas
latere fit minor, qua isto ad uos damni permensust
uiam ?
,, 248-9 : sed is clam patrem etiam hac nocte ilia per hortum
transiluit ad nos.

Strabax has been paying nightly visits to Phronesium, climbing

over the wall which separates the gardens of the two houses.
Either, therefore, there was no gate in this wall, or if there was a
gate it was kept locked at night, and Strabax was not trusted with
a key.

(9) Phor. 891-2 : sed hinc concedam in angiportum hoc proxumum :

inde hisce ostendam me, ubi erunt egressi foras.

Here w e are probably to understand that Phormio steps into the

right-hand side-entrance. 1 A t 898 he shows himself, declaring
loudly his intention of looking for Demipho. T h u s this passage,
which might at first sight be thought to support the ' conventional '
theory that the angiportum was an ' alley ' between two houses, and
1 Not simply as a hiding-place, but in order that Demipho and Chremes may

presently see him coming on to the stage from this side-entrance.

 APPENDICES 255
could be entered from the stage, really disproves i t ; Phormio's
entry from such an alley might arouse the suspicions of Demipho
and Chremes, whereas they would find it quite natural that he
should enter from the ' forum 5 side-entrance, presuming that he
is coming from his lodging, where Demipho would expect him to
have been since he last saw him (line 440).
W e m a y reconstruct the setting of the Phormio as follows (v.
Johnston, Exits and Entrances, page 32) :

(.angiportum) (angiportum)
(hortus) {hortus) (hortus)
L. Demipho's House Chremes' House Dorio's House R.

A t line 314 Demipho, having ordered Geta to fetch Phormio and

Phaedria to fetch Antipho, enters his house, announcing that he
will p a y his devotions to the household gods and thence (inde) make
his w a y to the forum and bring some friends back with him for the
interview with Phormio. Phaedria starts obediently in the right
direction, but slips into Dorio's house (line 310) as soon as Demipho's
back is turned. G e t a goes off to fetch Phormio (line 310) and
reappears with him at line 315. A t line 346 they see Demipho
approaching from the forum. Clearly he must have left his house
b y the hortus and angiportum, as he has not reappeared on the stage.
A gesture at the word inde would make his intention clear to the
spectators. N o doubt the explanation of lines 829-30 is similar ;
here Phormio refers to a visit to Dorio's house which he has just
made, though in fact w e have not seen him enter i t ; he must have
m a d e his entry and departure b y the back.
It would appear from our study of the Phormio that the side-
entrances can be used as temporary hiding-places, but also that
the term angiportum can be used of the streets into which they lead.
N o r is this surprising, for in Pseud. 960, 961, 971, as w e have seen,
angiportum is used of the platea itself.

(10) Adelphi 908-9 : atque hanc in horto maceriam iube dirui

quantum potest : hac transfer : unam fac domum.

This passage seems to prove that the gardens of two adjoining

houses should be thought of as separated by a single wall, through
which there is no gateway ; else w h y should Aeschinus have to
pull the wall down in order to convey his bride from one house to
the other ? Still less is there any room for an angiportum between
the two gardens.
It appears abundantly clear that the ' alley ' of which editors
so often speak could have had neither place nor function on the
R o m a n stage. T h e only means of entry and exit were the house-
doors and the side-entrances.
 APPENDIX D

CREPIDATA, PALLIATA, TABERNARIA, TOGATA

(Classical Review, Vol. liii. pp. 166-8.)

T H E Roman classification of the different kinds of drama according

to the characteristic dress or footwear used by the actors seems to
have grown up haphazard and never to have achieved a satisfactory
or agreed form. The loci classici are Diomedes (Keil, Gram. Lat. i),
pages 489-91 ; Evanthius, De Fabula, ch. iv ; Donatus, De Comoedia,
ch. vi, §§ ι and 5 ; Donatus on Ad. 7 (for Evanthius and Donatus
see Wessner's Teubner edition) ; Lydus, De Mag. i. 40. In these
passages attempts are made to classify drama ; some of the technical
terms employed are used, somewhat more casually, by other
writers such as Horace {A.P. 288) ; a comparison of all such
passages shows that the Romans themselves differed as to the
meaning of certain terms which our literary histories are apt to
employ with perhaps unjustified assurance.
Diomedes tells us that at first togata was a general term, including
(apparently) all forms of drama not translated from the Greek.
The corresponding term for all forms of drama derived from Greek
sources he gives as palliata, quoting Varro : Graecas fabulas ab
habitu aeque palliatas Varro ait nominari. Diomedes admits that
a communis error has grown up of limiting togata to one form of native
comedy, namely tabernaria, so that people speak of the togatae of
Afranius, while even Horace, he regrets to say, contrasts togatae
with praetextae. Diomedes himself uses palliata to include tragedy,
comedy, satyric drama and mime, while under togata he includes
praetextata, tabernaria, Atellana, planipedia. The word palliata occurs
in two passages, viz. Donatus, De Com. vi, §§ ι and 6 ; in both these
passages it is used in its modern sense of Latin adaptations of Greek
comedy. Evanthius, who is concerned with native Latin drama,
and Lydus, who is classifying tragedy, have no occasion to use the
word ; but it is surprising to find that Donatus on Ad. 7, in his
attempt at a complete classification of drama, does not mention
palliata : his list here is tragedy, comedy, togata, tabernaria, praetex·
lata, crepidata, Atellana, μΙμο%, Rhinthonica. As it is inconceivable
that Donatus, in his commentary on Terence, should have left out
of consideration the very form of drama which Terence composed,
we must suppose that one of the nine terms in this list can bear the
meaning ' Latin adaptation of Greek comedy '. So far, then,
against two uses of palliata in the modern sense, we have one where
156
 APPENDICES 257
it means something different and one where some other word takes
its place. What is this other word ? If we suppose that Donatus
includes under ' c o m e d y ' not only Greek originals but Latin
adaptations from them, how are we to explain his distinction
between tragedy and crepidata ? Lydus, loc. cit., tells us τίμνιται
( ^ τ ρ α γ ω δ ί α ) CIS κρηπι&αταν και πμιιιτ(£τάταν' ώΐ' η μίν κρηπιΆάτα ίΑΚηνίκας
ίχ(ΐ νποθίσας, η δί πραιτίζτατιι Crepidata Would then mean
ρωμαϊκά·;.
Latin adaptations of Greek tragedy—and this is the sense in which
most modern writers use the word ; but it cannot be the sense in
which Donatus uses it, if the text and logic of his note on Ad. 7
are sound. Apparently, then, the only two ancient writers who
use the word crepidata use it in different senses.
In view of this conflict among our authorities, perhaps we should
consider the words palliata and crepidata themselves. Drama was
classified according to characteristic dress and footwear, and the
pallium was recognized as the characteristic Greek dress even in
Plautus' day {Cure. 288). The crepida (κμηπίς) was a kind of Greek
open shoe or sandal ; Gellius (xiii. 2 1 . 5 ) gives crepidula as a Greek
equivalent of solea, a half-shoe (quibus plantarum calces tantum
infimae teguntur). The numerous illustrations in Daremberg and
Saglio show the great difference between this sandal and the high
boot which became for the Romans the symbol of tragedy under
the title cothurnus. There is plenty of evidence that the Romans
associated the pallium and crepida as typical Greek wear (Livy,
xxix. 19. 12, cum pallio crepidisque inambulare in gymnasio ;
Suet. Tib. 13, redegit se, deposito patrio habitu, ad pallium et
crepidas ; Cicero, Pro Rab. Post. 25-7, speaks of the pallium,
chlamys, crepida, and soccus as typically Greek—crepida and
soccus seem to mean much the same thing in this passage). The
crepida was an everyday type of shoe, such as might be worn by
slaves (Cie. In Pis. 38, ' crepidatus, ueste seruili'). Finally, as we
see from Plautus, Pers. 464, ' hanc hospitam autem crepidula ut
graphice decet! ', the crepida was worn in comedy.
The essential point about the pallium and crepida is their associa-
tion as typical Greek wear. ' Fabula palliata' and ' fabula
crepidata ' should therefore be synonyms denoting Latin drama of
Greek origin. Varro, followed by Diomedes, used palliata as an
inclusive term for all such derivative drama ; more generally,
however (as Diomedes admits), the term was confined to derivative
comedy. There may have been a similar variation with regard to
the meaning of crepidata. When Lydus uses it of tragedy derived
from Greek sources the emphasis is on Greek rather than on tragedy ;
he might just as well have used palliata (in its Varronian sense).
Normally, however, the associations of the crepida and the pallium
are with everyday life, and so with comedy ; and, though Sophocles
is said (Vita, ch. 4) to have introduced white κρηπΐ&ίς into tragedy,
no Roman could have used crepida as synonymous with cothurnus
258 THE ROMAN STAGE

to denote tragedy as opposed to comedy. T h e normal term for the

footwear of the comedian is socctis ; but where it was desired to
emphasize the Greek origin of a form of Latin comedy, the word
crepidata would have been exactly in place. T h e word missing from
Donatus' list is therefore some word specifically denoting Latin
tragedy derived from Greek sources ; and he may have left it out
for the very good reason that no such word existed.
As for togata and tabernaria, we have seen that Diomedes uses
togata to include native drama of all kinds, while protesting against
the communis error which identified it with tabernaria, the correct
term for the work of Afranius and Atta. We may agree with
Diomedes to this extent, that togata was not a very happy term for
these native comedies depicting humble life in country towns
(where, as Juvenal tells us, nobody wore the toga except as a
shroud). A word denoting the dress of humble folk might have been
more in place—perhaps tunicata (cf. Horace's ' tunicatus popellus'
and Naevius' early effort in this field, the ' Tunicularia'.) However,
the word actually chosen was not descriptive of dress at all, b u t of
the house of the characters ; ' tabernae ', says Diomedes, had once
been a general term for private houses ' quod tabulis tegerentur ',
and we remember that Horace uses it specifically for poor houses,
' pauper um tabernae '. Modern writers usually regard tabernaria
as a lower-class form of togata, and even suggest that it was Titinius
who wrote tabernariae, plays about people who lived in ' booths '
(e.g. his play about the Fullers), whereas the more polite Atta and
Afranius wrote togatae. I t is surprising, in that case, that Diomedes
refers to Atta and Afranius as writers of (togatae) tabernariae, while
Titinius is nowhere spoken of except as a composer of togatae. True,
Donatus twice mentions togata and tabernaria side by side, as though
they differed in some way, and Evanthius, defining the terms,
derives togata ' ab scaenicis atque argumentis Latinis', tabernaria
' ab humilitate argumenti ac s t i l i ' ; but this is a distinction without
a difference ; the fragments show that all native comedy dealt
(perhaps for reasons of prudence) with humble life. T h e grammar-
ians are sometimes apt to be over-scientific ; having the two terms,
they tried half-heartedly to distinguish between them. Diomedes
gives himself away when, contrary to his own theory, he refers to
Atta as a togatarum scriptor, just as he weakens his argument about
palliata by using palliati in the sense of comic actors. Finally, Donatus
lets the cat out of the bag when he tells us that some people refer
to togatae as tabernariae.
O u r conclusion is that, in spite of some variation and confusion,
the general usage of antiquity identified palliata and crepidata as
denoting ' adaptations of Greek comedy ', togata and tabernaria as
denoting ' native Latin c o m e d y ' , and that the modern attempts
to explain crepidata as ' derivative tragedy and tabernaria as ' a
lower-class form of togata ' are unsound.
 APPENDIX Ε

THE ROMAN STAGE CURTAIN

(Hermathena, Vol. lviii. pp. 104-15).

T H E drop-curtain was perhaps the most notable Roman contri-

bution to stage technique. When was it introduced, and why ?
W h a t was the relation of the aulaeum to the siparium, and how
were they operated ?
With the curtain as a form of decoration ΟΓ Scenery (таратгстасг/ха)
the Greeks of classical times may have been acquainted. Further-
more, the occasional use of a small portable curtain to conceal part
of the stage, an unwanted door in the background, etc., is so
obvious a device that it seems hazardous to deny it to the Greeks
altogether. 1 For the use by the Greeks of a large drop-curtain,
concealing the whole of the stage, there is no evidence ; indeed
the development of Greek stage technique, especially in its later
forms, as illustrated by our extant plays, seems to postulate a stage
permanently open to view. T h e Latin adaptations by Plautus and
Terence also read as if designed for a curtainless stage. T h e char-
acters are brought on at the beginning and taken off at the end in a
w a y which implies that opening or closing tableaux were impossible.
A n isolated passage such as Capt. 1,
hos quos uidetis stare hie captiuos duos,
cannot outweigh the evidence of the plays as a whole. Still less
was a curtain available within the course of a play ; hence objects
no longer wanted have to be removed under the eyes of the
spectators, whether without or with apology. Just as from Hamlet's
remark :
I'll lug the guts into the neighbour room,
we infer the absence of a drop-curtain in Elizabethan times, so
when Lysimachus suggests that the dishes left by the cook in front
of the door should be taken inside, adding the lame remark that
they will supplement the family meal (Merc. 800-2), we clearly see
the embarrassment to which the dramatist was put by the absence
of a curtain on the Greek and Roman stages.
T h e aulaeum is said by Donatus (de com. 12. 3) to have been
introduced in 133 B.C. (aulaea quoque in scaena intexta sternuntur,
quod pictus ornatus ex Attalica regia R o m a m usque perlatus est;
1 Nevertheless I do not now think that doorways were concealed in this way :

see Appendix ( / ) .
259
2бо THE ROMAN STAGE

pro quibus siparia aetas posterior acccpit) ; cf. Varro de uit. pop·
Rom., Serv. ad Aen. i . 697. These passages seem to mean that
embroidered curtains were among the luxuries introduced from
Attalus' palace, along with chlamides, pallae, plagae, uasa aurea ;
the use of the aulaeum on the stage, at least in the form of a drop-
curtain, was probably a Roman invention, which must have been
subsequent to 133 B.C. It is first mentioned by Cicero, pro Cael.
65 (56 B.C.) ; he speaks of it as rising at the end of a mime. O u r
most detailed picture of the rising of the aulaeum is supplied
by Ovid (Met. x. 111-4), who shows that the curtain rose
smoothly and that only when it was fully raised did the
figures embroidered on it become completely visible ; they then
seemed to be standing on the edge of the stage. Similarly Virgil
(1Georg. iii. 25) refers to the ' embroidered Britons lifting the
aulaea Clearly the curtain was raised to conceal the stage ; and
Phaedrus, who lived about this time, describes the lowering of the
curtain (aulaeo misso) at the beginning of a pantomime. From
Horace (Ep. 11. i. 189) we learn that the aulaeum was ' kept d o w n '
throughout an elaborate stage display, lasting for four hours or
more, and that to wait for the aulaeum is to wait for the end of
the play {A.P. 154).
We have seen that the drop-curtain had been introduced to the
Roman stage by the year 56 B.C. at latest. W h y it was introduced
is further indicated by Cicero in the passage referred to. He is
contrasting a regular, artistically constructed play with a mime,
which comes to an abrupt halt at a given signal : mimi ergo iam
axitus, non fabulae : in quo cum clausula non inuenitur, fugit
eliquis e manibus, dein scabilla concrepant, aulaeum tollitur. In a
regular play the spectators receive ample warning that the play is
coming to an end, even before the final ' plaudite ' ; but it was
precisely because of the formless nature of the mime that its con-
clusion had to be marked by some external device—and the device
used, according to Cicero, was the drop-curtain. T h e vogue of the
mime and of similar performances may, therefore, explain the
introduction of the drop-curtain itself.
Early in the first century B.C. we notice certain new features
in Roman dramatic composition and theatrical production.
Theatrical performances tend to become at once more splendid
in display and more trivial in content ; the supply of new
literary plays for the stage dries up, the writing of literary plays
now becomes a private amusement for noble dilettanti, and the
stage is given over to farce and display, with the occasional
revival of an old play. W e hear that Sulla was fond of the company
of μϊμ.01 and γίλωτοποίοι, that he wrote farces in Latin, and that
Pomponius and Novius, writers of Atellane farces, were active about
this time (Plut,. Sulla 2 ; N Athen, , ν ^ 78);. T/he introduction of
painted scenery in 99 B.C. (Val. Max. 11 4 ) 1 may also h^ve helped
 APPENDICES 261
to make the use of a front-curtain desirable. O f course, once
introduced, the drop-curtain would have been used not only for
mimes but also for regular drama ; indeed, as we shall see, Cicero
gives some evidence on this point.
In our modern theatres the curtain is drawn up towards the
roof to leave the stage visible, and lowered so as to touch the stage
at the end of the performance. Though the earliest drop-curtains
in the post-Renaissance theatres appear to have worked in the
opposite way, the convenience of the modern arrangement is
obvious The Roman method of working the aulaeum clearly
raised two problems—how to prevent the curtain, when lowered,
from encumbering the stage, and how to raise and lower it at all
when, as is generally supposed, there was nothing overhead to which
ropes could be attached. When we find in the remains of Roman
theatres traces of a recess under the stage floor, near the front of the
stage and running its full length, we seem to have the answer to the
problem of where the curtain was housed during the performance.
Indeed we have some evidence that this ' curtain-slot' lent itself
to other purposes, such as the emergence o f ' ghosts ' from the under-
world ; it was from here that Catienus appeared to address his
famous ' mater, te appello ' to the deaf ears of his drunken fellow-
actor Fufius in a revival of the Шопа of Pacuvius ; cf. Cie. Sest.
126 : emergebat subito, cum sub tabulas subrepserat, ut ' mater,
te appello ' dicturus uideretur, and the scholiast : in eo est argu-
mentum ita dispositum ut Polydori umbra secundum consuetudinem
scaenicorum ab inferiore aulae(i) parte procedat. The operation
of the curtain is thought to be connected with the row of holes
sometimes found in the floor of the curtain-slot. Mazois suggested
that a hollow post reached from the bottom of each hole to the
level of the stage, while a narrower post could be made to slide
down into the larger post or (by means of a rope attached to the
bottom of the inner post and the top ©f the outer post) to project
above stage level, raising the curtain with it. By this means it
would be possible for some one in one of the wings to raise or lower
the curtain by pulling a rope or releasing it, even though there was
nothing overhead to which pulleys could be attached. For
simplicity's sake I am giving Mazois' theory (so far as I understand
it) in the form accepted by Fiechter (Baugesck. Entwich, des ant.
Theat.), namely, that there were only two sections in each post.
Fiechter reproduces Mazois* sketch (fig. 119). I do not feel
competent to discuss the archaeological evidence, and am willing
to accept the view that the holes were designed for narrower posts
inside larger ones. Nor do I feel able to say whether the engineering
problem involved in raising the huge curtain by this means could
1 The words of Valerius Maximus and Pliny the Elder (35, 23) may mean

only that the scene-building was painted decoratively for architectural effect; see
Appendix ( / ) .
18
 2Ö2 THE ROMAN STAGE

be solved by the Romans. Since the posts must have risen above
the stage when the curtain which they supported was raised, and
since it is hard to suppose that when the curtain was lowered the
posts were still left standing above the stage, thereby interfering
with the spectators' view, we seem driven to accept the theory of
telescoping sections. I still have two difficulties. First, is it not
hard to believe that the Romans passed straight from the curtainless
stage to this elaborate method of working the drop-curtain ?
Would it not be easier to imagine that the aulaeum was at first
operated in some simpler and more obvious w a y ? Second, if we
find evidence that the working of the drop-curtain in the later
empire cannot be explained on Mazois' theory, are we not bound
to consider the whole matter afresh ?
Leaving the aulaeum for the moment, I turn to the other form
of stage-curtain, the siparium. T h e word is a varient of supparum,
a topsail, or woman's upper garment.
T h e siparium is nowhere described as being raised or lowered ;
Apuleius twice refers to its being folded up. It must have already
been in common use in Cicero's time, as is shown by his phrase
post siparium, ' behind the scenes ', in contrast to in exostra, ' in full
view ' (de prou. cons. 14). Donatus (I.e.), after observing that the
aulaeum had been superseded on the stage by the siparium, defines
this as a mimicum (v.l. minutum) uelum quod populo obsistit
dum fabularum actus commutantur. Festus remarks : siparium,
quo in scenis mimi utuntur, dictum ait Verrius a uelamento, quod
uocetur alias aulaeum. Seneca (dial. ix. xi. 8) uses siparium =
' mime ' in contrast to cothurnus = ' tragedy '. Juvenal (viii. 185)
speaks of a bankrupt noble hiring his voice out to the siparium as
an actor in the Phasma of Catullus (a mime-writer), and the
scholiast adds siparium uelum est sub quo latent paradoxi cum in
scaenam prodeunt, aut ostium mimi. This last remark is perhaps the
most illuminating. T h e siparium appears to have been a portable
screen behind which jesters could hide when waiting their turn to
appear (perhaps shown in Dieterich's Pulcinella, pi. 2). It served the
mime-actors both as front-curtain and as back-scene. They could
set it up where they chose ; they stood behind it until each actor's
turn came to appear, when he made his way through a parting in
the middle of the curtain and so displayed himself to the audience.
In the performance this parting would serve as the house-door, the
need for which is seen in the Oxyrhynchus mime. When the mimes
gained access to the theatre they would still bring their siparium
with them. Diomedes gives as an alternative explanation of the
term planipes the fact that the mimes had at one time stood not on
the high stage but down in the orchestra, where they used to set
up their instrumenta and give their performance. W h a t can the
instrumenta have been except the siparium ? Festus says of the
mimes : solebant (enim saltare) in orchestra dum (in scaena actus
T H E ROMAN THEATRE, VAISON-LA-ROMAINE
SHOWING C U R T A I N - S L O T
 APPENDICES

fa)bulae componerentur. In the R o m a n theatre, since the

orchestra was occupied b y the chairs of the senators, the mimi, like
all other performers, must have appeared on the stage, as is shown
b y Cicero's statement that at the end of their performance the
aulaeum was raised. E v e n on the R o m a n stage, however, the
traditional siparium might still have been useful. W e m a y picture
the mime, when performed in the R o m a n theatre, as an interlude
between oth^r performances (embolium) or an after-piece (exo-
dium) ; if the performers had appeared from the same house-doors
and acted against the same background as the players in the drama
which had just been staged, the dramatic illusion might have been
disturbed. 1 W e m a y perhaps conjecture that, when the aulaeum
was raised at the end of the play, behind it the mimi set u p their
siparium and waited for the aulaeum to be lowered again. How-
ever we conceive the matter, it is clear from the evidence of Cicero
and Juvenal that the aulaeum and siparium were both in use in
their times.
For the use of the aulaeum and siparium together w e have
evidence in Apuleius, Met. x . 29. Lucius, the Ass, is waiting his
turn to appear on the stage of the theatre at Corinth ; meanwhile
he seizes the chance to crop the grass as he peers in at the open door.
Presumably he is standing in the parodos, and thus commands a
view of the stage and the orchestra. A s a curtain-raiser boys and
girls give a Pyrrhic dance ; then the trumpet sounds, and aulaeo
subducto et complicitis sipariis scaena disponitur. Butler translates:
' the great curtain fell away, the lesser curtains were drawn back,
and the stage was arrayed before our eyes.' This translation gives
a clear picture ; the preliminary dance must have taken place in
the orchestra ; then the curtains are withdrawn, and the spectators
see a most elaborate scene on the stage (' there was a mountain of
wood,' etc.). Bulle (Untersuchungen an griech. Τ heat., page 285,
ι) takes a similar view, translating subducto as ' nach unten gezogen
T h e trouble is that subducere nowhere means ' lower ' : it can,
and frequently does, mean ' raise ' (for which meaning Georges
quotes this very passage). Fiechter {op. cit., page 120) understands
Apuleius to m e a n that the aulaeum was raised to conceal the stage,
the siparia being folded together for the same reason ; then,
behind the curtains, the new scene is built up ; unfortunately
Apuleius has forgotten to tell us that, when the new scene is ready,
the aulaeum is lowered again to reveal it. This view at least does
justice to subducto. But would the audience have sat patiently,
gazing at the aulaeum, while so elaborate a scene was being
prepared behind it ? In Met. i. 8, aulaeum tragicum dimoueto et
siparium scaenicum complicato, the effect of folding the siparia is
evidently to reveal something, not to conceal i t ; therefore, in the
present passage, complicititis sipariis must m e a n that the siparia
1 This argument now seems to me to lack force ; see Appendix ( f ) .
264 THE ROMAN STAGE

were folded together so as to reveal the stage, and aulaeo subducto

is part of the same process. A more plausible translation would be
the aulaeum was removed this being a possible sense of subducere
but if the process referred to is really the lowering of the curtain,
subducere, which frequently means ' l i f t w o u l d be an unfortunate
word to choose ; and if Apuleius' meaning is that the aulaeum is
carried away altogether, we have to suppose that the method of
operating it as a drop-curtain had been abandoned in the second
century. Fiechter, indeed, though taking the aulaeum here to be
a drop-curtain, worked in the classical method, believes for quite
other reasons that the classical method of working it was altered
in the second century. In three theatres of this period (at Dugga,
Timgad and Athens) there are postholes, but there is no trace of a
curtain-slot ; he therefore thinks that from this time on the curtain
was attached on the morning of the performance to the movable
posts and taken away again at the conclusion of the performance.
If we try to apply such a suggestion on the present instance, we
have to suppose either that the aulaeum rose and fell between the
various items of the day's programme ; and when it fell, as there
was no longer any curtain-slot, it must have remained heaped on
the stage while the posts disappeared underground (a view which
does not solve the difficulty presented by subducto, and adds the
complication that the curtain is left lying on the stage during the
main performance), or that the aulaeum is no longer used as a
drop-curtain at all. But we have three references from the fourth
century to the rising or falling of the aulaeum. T w o come from
Ammianus Marcellinus : (a) xvi. 6. 3, Dorus euanuit et Verissimus
ilico tacuit uelut aulaeo deposito (where the lowering of the curtain
marks the end of the performance) ; (b) xxviii. 6. 29 : ut ne quid
cothurni terribilis fabulae relinquerent intentatum, hoc quoque
post depositum accessit aulaeum (where, again, the lowering of the
curtain marks the end of the tragedy, after which comes the
exodium). T h e third passage refers to the raising of the curtain.
Donatus (Eun. praef. 1. 5) tells us that Terence avoids pauses
between the acts, ne ante aulaea sublata fastidiosus spectator exsurgeret,
' lest the spectator should grow weary and leave his seat before the
curtain rose ' (at the beginning of the next act, presumably).
Incidentally these passages suggest that the statement already
quoted from Donatus to the effect that the aulaeum had been
superseded by the siparium can scarcely have been universally true.

A drop-curtain was, then, known to Ammianus Marcellinus

and to Donatus, and it worked just like our modern curtain, being
raised at the opening of the performance and lowered at the end.
T h e Thesaurus takes this view of the passages in Marcellinus. T h e
passages in Donatus and Apuleius can be explained, I think, only
in this way. A l l our literary references to the drop-curtain sub-
sequent to the age of Tiberius point, therefore, to its having worked
 APPENDICES 265

in the reverse way to that which was known to Cicero, Virgil,

O v i d and Phaedrus. It must now have been kept up during the
performance in such a w a y as not to obstruct the view. But,
according to Mazois' theory, the curtain could not have been raised
higher than the top of the inmost poles, and to lift both curtain and
poles out of sight would be impossible. So far as I know, the
archaeologists are not aware of this new problem, and consequently
have no solution to offer.
T o keep a drop-curtain raised so that it does not obstruct the
view of the stage necessitates the existence of some fixed object above
the stage to which ropes can be attached. T h a t ropes were used
for drawing up other objects above the heads of the spectators is
shown by Plutarch, Sulla xi, where we hear of a figure of Victory
being made to soar aloft in the theatre at Pergamum. Juvenal
(iv. 122) speaks of boys being hoisted to the awnings spread high
above the spectators' heads :
et pegma et pueros inde ad uelaria raptos.

T h e drop-curtain of the later period, at least, must have been

drawn up to something. One might conjecture that permanent
posts were erected, one at each end of the stage, perhaps concealed
in the wings. Another possibility is that the curtain was drawn up
towards the stage roof. But if we grant this for the later period, may
we not grant it for the earlier period as well ? Fiechter thinks that
the stage roof of the uncovered theatres is a vestige of the roof which
covered the entire building in the theatrum tectum ; he also
believes that the theatrum tectum in Pompeii dates from 80 B.C.
T h e need to protect the stage with a permanent roof goes well with
the other developments (such as the painting of the scene-building)
which took place early in the first century. Once given a stage
roof, whether high or low, the idea of a drop-curtain seems obvious.
No doubt experiments were made, as in modern times ; the
first plan, apparently, was to raise the curtain to conceal the
stage and to lower it into a recess to reveal the stage. I suggest
that the telescoping posts 1 were a development of this first plan ;
they were designed not to supply the power which raised the curtain,
but to ensure that it rose evenly, and perhaps to take some of the
weight. There may have been some simple locking device which
would hold the posts rigid when they were fully extended and which
could be released when it was desired to lower the curtain again.
In other words, the system of telescoping posts was merely supple-
mentary to the essential and common-sense principle that the
curtain was raised and lowered by ropes attached to some fixed
object, which was at least as high as the top of the fully raised
curtain itself.
W e have seen reason to believe that at some time between the
1 If they ever existed.
266 THE ROMAN STAGE

life of Phaedrus and that of Apuleius the new system came in of

operating the curtain as in the modern theatre. The top of the
curtain must now have been permanently raised ; the bottom was
drawn up to it to reveal the stage and let fall to conceal the stage.
The telescoping posts and the curtain-slot were thus made useless ;
and in fact Fiechter finds (page 123) that in the second century the
curtain-slot disappears. We need not, however, suppose that the
change of system occurred everywhere at the same time ; much
must have depended on local circumstances, on the size of the theatre,
the nature of the performances for which it was mainly designed, etc.
At Verulamium the post-holes appear to have been filled up at an
early period, while the curtain-slot remained in use. May we not
suspect that each theatre had, at any given period, its own method
of working its curtain ?
Apuleius twice speaks of the aulaeum and the siparium being
used together for the purpose of concealing the stage ; when the
aulaeum rises to reveal the stage, the siparia are folded together
for the same purpose. The siparia in these references can scarcely
have been the portable screens used by the mimes. Butler is
perhaps right in translating complicitis sipariis as ' the lesser curtains
were drawn back ' ; the picture appears to be that whereas the
aulaeum concealed the stage as a whole, the siparia were hung in
the wings. When we roll up a window-blind and fold back the
side-curtains we are performing a somewhat similar operation.
Such siparia would be particularly useful if the aulaeum did not
reach quite from end to end of the stage. It may be worth adding
that at Dugga, where no trace of the curtain-slot has been dis-
covered, an inscription records the presence of siparia : scaenam
cum siparis et omamentis omnibus et . . . The
mention of σάφαροι in an inscription at Ephesus
also suggests that the term siparia may some-
times have been used of a more permanent
form of stage equipment than the portable screens
of the mime-actors. We must not forget Festus'
remark that Verrius identified the siparium with
the aulaeum. When Donatus tells us that the
aulaeum has in his day been superseded by the
siparium, he appears to be referring to the perform- j jj Γ
ance of Terence's comedies. We have reason to "
believe that any such performances in the late У
empire were strongly influenced by the mime ;
thus we are told by Donatus that actresses were
now employed for female parts. So the influence of
the mime may have helped to bring back the Mechanism of
siparium into favour, and it is well to recall the drop-curtain
frequent indication of small curtains in the illus- according to
trations to the manuscripts of Terence. Mazois
 APPENDIX F

CHANGE OF SCENE AND CHANGE OF S C E N E R Y : T H E

Q U E S T I O N OF ' SETS '

L I K E all Greek drama, the plays of Plautus and Terence were

written for an open-air, curtainless theaire. So long as the stage
lay permanently open to view, there was, in my opinion, no attempt
to alter the visible background, though the imaginary scene might
be changed at will. So in the medieval drama the stage might
represent any desired place—a freedom satirized by Sidney but
still retained by Shakespeare. Change of scene in Greek drama
seems to be confined to the earlier period ; in the Eumenides the
scene shifts from Delphi to Athens, in the Peace from Earth to
Heaven, in the Frogs from Earth to Hades.
The innumerable theories as to Greek stage scenery which
have been put forward may be arranged in two main groups. The
common principle of the first group is that the actors' house,
otherwise called the skend or permanent scene-building, was itself
specially adapted to the needs of each play. The principle of the
second group is that the permanent scene-building was concealed
by ' sets '. In default of any real evidence in favour of either of these
views, their supporters appeal to the text of the plays and to the
evidence of vase-paintings and wall-paintings supposed to be in
some measure inspired by theatrical performances. Unfortunately
a major premiss is always lacking from such arguments ; we have
no proof that verbal descriptions of the imaginary surroundings
were pictorially represented in the physical setting, or that ancient
illustrations, even when they contain details which are agreed to
be theatrical in origin, are in other respects faithful copies, or
copies at all, of theatrical performances.
Aristotle (Poet, iv) tells us that σκηνογραφία and the third
actor were introduced by Sophocles. Vitruvius, however, asserts
(vii. praef. §11) that it was Agatharchus who first made a scaena,
and that he did so when Aeschylus was active as a dramatist (or
perhaps ' under the instruction of Aeschylus '). In the view of
Pickard-Cambridge (Theat. of Dion., page 124) Agatharchus
' painted an architectural design in perspective on the flat back-
ground '. Apparently what he did was to paint the front of the
scene-building so as to give it a decorative, architectural effect,
sufficiently dignified to serve as the permanent background for
dramatic performances of all kinds. The early plays of Aeschylus
(Suppliants, Persae, Seven Against Thebes, Prometheus Bound) perhaps
268 THE ROMAN STAGE

belong to a time when there was no scene - building. But

from the Oresteia onwards the action of every extant play is
specifically related to some sort of building or dwelling in the back-
ground, whether it be temple, palace, private house or houses,
tent or cave. From the Oresteia onwards we also find the three-actor
rule established. Apparently the purpose of the skend was to
house the three actors, and the task of Agatharchus was to decorate
the skeni front. From now on all Greek drama was acted before
a permanent background, the front of the scene-building, with its
three doors. I f this view is correct, it would be misleading to
translate σκηνογραφία as ' scene-painting \
O u r modern convention of realistic scenery is bound up with a
whole complex of circumstances utterly remote from the Greek
theatre. O u r theatre is an enclosed building, artificially lighted.
By darkening the auditorium and lighting up the stage we concen-
trate the attention of the spectators on the stage, for that is all that
they can see. T h e Greeks sat in theatres open to the sky ; if they
turned their eyes away from the actors, they could look at real
sunlit scenery of the most spectacular kind. Whatever Agatharchus
and his successors tried to do, they can hardly have attempted to
compete with Nature. If the spectators fixed their attention on the
drama, it was because they were interested in the drama itself—
the words of the actors and their actions.
There is no real evidence that the Greeks made any attempt to
suit the background to the play. T h e assumption that they did so
is based on nothing but our own convention of realistic scenery ; it
always defeats itself by leading to most unrealistic results.
It is agreed that in the theatre of Dionysus several plays were
performed in one day. In 431 B.C. Euripides presented the Medea,
the Philoctetes, the Dictys and the satyric drama Theristai, after
which presumably came a comedy. Thus the imaginary back-
ground changed from palace to sea-shore, then to country, and
finally to two or three private houses. If we assume that the
visible background was somehow changed between one play and
another, we must allow that any such change was made in a very
short time and in full view of the audience. If realism is what we
desire, how do we suppose that a realistic-looking palace was
changed (for example) into a realistic-looking cave ? If, on the
other hand, we are content with slight and purely symbolical
alterations ; if, with Professor Percy Gardner (' T h e Scenery of
the Greek Stage J.H.S., xix, 1899, pages 252-68) ' we may
suppose that a few rocks strewn on the stage . . . would well
suffice to satisfy the audience ' in the case of a satyric drama, then
we have evidently abandoned our demand for realism, and may as
well accept the convention in all its stark simplicity. A few rocks
strewn on the stage would be of small aid to the imagination ;
instead of helping the spectator to forget the inadequacy of the
 APPENDICES 269
background, they would tend to underline i t ; and they would
presently (ex hypothesi) have to be cleared away before the next
play could begin.
If we abandon the view that structural alterations, sufficiently
convincing to be worth while, could be made in the brief interval
between play and play, and if we still crave for a visible background
corresponding to the imaginary setting, we may be tempted by the
alternative hypothesis o f ' sets '. This hypothesis has now received
a notable adherent in Dr. Pickard-Cambridge. He holds (Theatre
of Dionysus, pages 122-33) that changes of scene were indicated
' by the use of painted canvas or screens or panels, which could be
easily moved into position ' ; in other words, that there were ' sets
for each type of scene, easily t r a n s f e r a b l e w h i c h the theatre
servants could put in position ' openly in the presence of the
spectators without offending their susceptibilities He does not
claim that this is more than a theory; in fact he discounts certain
obscure remarks in ancient authors (page 122, note 2) which have
sometimes been regarded as evidence for these screens. 1 (The
καταβλήματα mentioned by Pollux, iv. 131, belong to the ΊΓίρίακτοι,
on which see Appendix (b) .) T h e reason for putting forward the
hypothesis at all is presumably that it seems to offer an easy and
practical way of suiting the background to the needs of each play in
turn. Nevertheless it raises serious difficulties, some of which were
pointed out by Professor Percy Gardner in the article already
cited : cf. page 258 : ' There is no justification for the notion that
the scene-painter of the fourth century would stretch across the
upper part of the зкепё a great canvas screen representing the
sky and across the lower part of that front another canvas screen
representing some particular place, and pierced with doors
corresponding to the doors from the skeni on to the stage '.
I have already pointed out that the doors in R o m a n Comedy
have to endure a good deal of knocking and kicking. This raises no
difficulty if we are dealing with the sturdy wooden doors in the
scene-building ; would it be possible if the doors were made of
canvas or light wooden material ? I do not know how far in front
of the skene these screens are supposed to have stood. Mercury,
in the Amphitruo, standing on the house roof, empties pots on the
head of Amphitruo below ; this is readily understandable if the
house-front is also the front of the scene-building, not so convincing
if the house-front is a canvas screen, separated by some space from
the front of the scene-building. I do not see where these great
screens were housed, nor how they were moved. As Navarre
remarks (Le Theatre grec, pages 80-1) ' Par quels moyens materiels
1 He thinks, however, that the nickname proskenion applied to Nannion should

be taken in the sense of ' screen' (p. 157). I would give the term its ordinary
tense, ' front of scene-building because this, like Nannion, owed its decorative
appearance to paint.
 270 THE ROMAN STAGE
ces divers decors 6taient-ils obtenus ? Vraisemblablement ä l'aide
de chassis peints, appliqu6s contre la fa$ade de la зкёпё et glissant
sur une coulisse. Toutefois les dimensions considörables de ces
chassis (plus de 20 n i t r e s de long au thdatre d'Athänes, par
exemple) les eussent rendus difficilement maniables.' He has
accordingly to suppose that the Greeks were already acquainted
with the scaena ductilis or some other Roman device ; in other words
he admits that the use of sets is a pure hypothesis, and one not free
from difficulties.
But the chief logical argument against the theory of ' sets '
is not that they would have been difficult to manipulate but rather
that they would (ex hypothesi) have afforded only too quick and
convenient a method of changing the scene. For if the Greeks really
felt the need of changing the visible background to suit the imaginary
setting, and if they solved the problem so successfully, we should
have expected that as Greek drama developed change of imaginary
setting would become more common. The opposite is the case.
T h e freedom from the unity of place which we notice in fifth-
century drama seems to have disappeared in the time of Menander.
If the fragments of his plays and the versions by Plautus and Terence
are any guide, the normal setting of New Comedy is a section of
street in front of one, two or three houses. I have already pointed
out that the wild scenery of the early part of the Rudens is completely
ignored for the greater part of the play, and was in all probability
left to the imagination, whereas Daemones' cottage remains
visible from first to last. Imaginary scenery can be forgotten when
no longer required, but physical objects would have to be physically
removed. T h e scene-building and its doors are the permanent
reality ; indeed the prologue to the Menaechmi (70-6) shows that
the same setting serves for play after play :
' This city is Epidamnus during the presentation of this play ;
when another play is presented it will become another town '
(Loeb. tr.).

To change the play will change the town as well,

And other folk within these doors will dwell;
For now a pimp you'll see, and now a lover,
And now a cross old father you'll discover,
Poor man or beggar, gipsy, toff or tout;
It all depends on what the play's about.

So far from any attempt being made to adapt the visible back-
ground to the imaginary setting, it was the setting which had to be
adapted to the permanent background. Where the background
helped, it was turned to good account; where it did not help, it
had to be ignored. But the difficulty of ignoring it was felt more
keenly as the public grew more sophisticated ; and the result is
seen in New Comedy, with its stereotyped setting.
 APPENDICES 27I

THE STAGING OF 'INTERIOR' SCENES: THYROMATA

OR PROTHYRA?
T h e action of every extant specimen of N e w C o m e d y is specific-
ally related to one, two or three buildings, the doors of which are
used b y the characters for their exits and entrances. T h e r e can
therefore be no question of change of the imaginary setting.
Nevertheless it is supposed b y some writers that special arrange-
ments were m a d e for the display of certain scenes which in real
life would probably take place indoors. T h e two possible alterna-
tives are :
(1) that such scenes were staged in the doorway or a little
distance inside the doorway. This implies that the doors in the
back-scene were w i d e — t h e so-called thyromata. (For illustrations
of attempted reconstructions see Bieber, figs. 296, 297, 303, 306).
(The thyromata, according to Bieber (page 222), were ' wide open-
ings into which backdrops and other decorations were set, or which,
w h e n left open, served to exhibit interior scenes '.)
(2) that they were staged in front of the house-doors, but inside
the so-called prothyra. These would be projecting open porches,
the roofs of which were supported by pillars. (For illustrations see
Pickard-Cambridge, figs. 9-29).
It will be seen at once that these two alternatives are opposed
to each other. Nothing could well be more different from the
opening of a door wide in order to display an interior scene than
the partial concealment of that door behind a projecting porch.
Moreover the view just quoted concerning the use of the thyromata
contradicts itself; if the doorways were filled with scenery, they
could not at the same time be used as doors ; and practicable
doors in the background are a n absolute necessity, if N e w C o m e d y
is to be staged at all.
T h e thyromata and prothyron theories have this in common :
they confine the actor in an enclosed space, where his movements
must necessarily be less free, his gestures less visible and his words
less audible than if he stood on the open stage. T o balance these
disadvantages, two claims are m a d e :
(1) that intimate scenes would be more convincing if framed
within a building ;
(2) that instances where one character is not seen b y another
are easier to understand if w e suppose that the unseen character
was concealed within a doorway or a projecting porch. Thus, for
example, Professor Nixon, in his L o e b fcdition of Plautus, assumes
that the banqueting-party in the Asinaria (line 830 ff.) takes place
inside Cleareta's house. T h e weakness of this supposition has
already been pointed out ( p - 1 7 1 ) .
A party takes place indoors in the Bacchtdes, but the spectators
do not see i t ; A r t a m o opens the door an inch or two (832) and
272 THE ROMAN STAGE

Nicobulus and Chrysippus peer in and discuss what they see.

Lines 720-3 show that the interior of the house is visible only to
those standing at the open door. The spectators have to rely on
the report given by the actor peering inside the door, just as they
are accustomed to be told by the actor on the stage of an impending
arrival from one of the wings.
There is no clear evidence that any character is thought of as
visible to the spectators until he actually makes his entrance on the
stage. The doors seem normally to be shut, except while a character
is entering or coming out. If Theopropides is surprised to find his
door shut in broad daylight {Most. 444), Antipho seems somewhat
surprised too when he finds his daughter's door open (Stich. 87).
The Loeb editor occasionally suggests that a character uses his
own doorway as a place of concealment, while observing something
taking place on the stage. This may seem plausible enough, but
there is nothing in the text to support such a view. If Cleostrata
is really standing in her doorway (Car. 562) to spy on her husband,
it is surprising that at line 573 he exclaims ' but there's my wife in
front of the house ! I'm afraid she's not deaf and that she's heard
all this.' In truth the conclusive argument against the theory that
intimate scenes were staged inside the house-door is that the text
usually speaks of the characters in such scenes as being in front of
the house (Pers. 756 ; True. 490, 583, 631 ; St. 147, 683).
The alternative theory is that interior scenes were shown in a
projecting prothyron or porch in front of the house-door. This is
the view of Lundstrom (' Aussen oder Inner ', Eranos, i, page 95,
1896), Kelley Rees (' The Prothyron in Greek plays ', C P . x, 1915,
page 1 1 7 if.) and Dalman ('De aedibus scaenicis comoediae nouae',
Leipzig, 1929) ; it has been searchingly criticized by Legrand
(Daos, 1910, pages 434-444). While Pickard-Cambridge is in
general opposed to the view that such porches were shown in the
permanent back-scene (pages 75-100), he is willing to allow
(page 174) that ' in the few plays in which scenes were certainly
acted in a portico in front of the main house, such a portico could
obviously be erected ad hoc in front of any of the doors of the
skene', referring in a foot-note to the Mostellaria and the Stichus.
The crucial passage on which the whole theory rests is admitted
by both Kelley Rees and Dalman to be Most. 817. Two houses are
mentioned in this play, the houses of Theopropides and Simo. 1
In line 817 Tranio, looking at Simo's house, says to Theopropides :
uiden uestibulum ante aedis hoc et ambulacrum quoiusmodi ?

' Do you see this porch in front of the house here, and what a fine
promenade ? ' Theopropides replies ' Yes, very handsome, upon
my word *. Rees comments : ' Thus the vestibulum formed a part
of the stage building and was visible to actors and spectators \
1
See plan on p. 250.
 APPENDICES 273
Identifying it with the ambulacrum, he describes it as a ' promenade
porch Having thus established the existence of a visible porch
of some size, he proceeds to find a dramatic use for it. It would
serve admirably, he thinks, for drinking-parties, toilet-scenes and
the like, which the text expressly states to take place in front of the
door ; it would also serve on occasion to explain why one character
does not see another. In the Mostellaria he finds three such
scenes. At line 157 we see Philematium, assisted by Scapha,
preparing her toilet in front of Theopropides' house (cf. line 295),
while Philolaches, who had arrived at line 84 from the forum and
is therefore supposed to be standing in the street, looks on un-
observed. Here, as Rees points out, ' if the women were busy with
the toilet inside the house, Philolaches would have to stand upon
the stage between the women and the audience in order to see them.
In this case the women would have seen him '. Moreover, when at
length he addresses them (11. 295-6) he bids Scapha ' get inside
and take this trumpery in with you '.
At line 308 slaves bring tables and couches, and the lovers, as
they take their places at table, see their friends Callidamates and
Delphium coming along the street (in uia, 326) to join them. Soon
all four are carousing together. But at 348 Tranio arrives from
the harbour with the bad news that Theopropides has returned ;
he packs them all indoors (391) and locks the door behind them
(425-6). These, then, are the two ' interior ' scenes which Rees
supposes to have been acted inside an open porch in front of the
door of Theopropides' house. Later in the play Callidamates'
slaves, who have come to fetch their master, begin to knock at
Theopropides' door (898) ; Theopropides comes out of Simo's
door at 904, yet does not see them until 935. Rees infers t h a t ' they
were partially concealed ' from Theopropides owing to the fact
that they were standing in the porch. Similar scenes occur, as we
have seen, in other plays, though nowhere else do we have any
indication as to the appearance of the porch. 1 For this Rees relies
on vase-paintings supposed to be reminiscent of theatrical scenes
and showing figures standing in or near a roofed and columned
building like a small temple. If the argument from vase-paintings
is valid at all, we seem compelled to suppose that such porches were
a normal part of the visible background, not only in tragedy but
also in comedy. There is great difficulty in supposing that such an
imposing structure was run up merely to serve the purposes of a
particular play, and removed afterwards. And as nothing in our
comedies would suggest that one house front was more imposing
than another, it is difficult to refuse all three doors their porches. In-
deed Rees refers to a Pompeian wall-painting in the Casa dei
Gladiatori which shows three doors, each with its portico. But
if three porches formed part of the background, it is not easy to see
1
But see As. 425 ; Plant, frag, 146 (Lindsay).
 274 THE ROMAN STAGE

h o w plays were staged in which only one or two houses are

mentioned. I t is one thing to ignore a plain door set unobtrusively
in a flat w a l l ; it is another matter to pretend that a visible pillared
and roofed porch does not exist.
Everything turns on how we interpret Most. 817. According to
K e l l e y R e e s , ' ambulacrum is a descriptive term for the uestibulum.'
A t line 756 T r a n i o tells Simo that Theopropides is planning to
improve his house in view of his son's approaching marriage. H e
wishes to add women's quarters, baths, an ambulacrum and a
porticus ; he therefore proposes to visit Simo's house and take it as
a model, having heard that it is well provided in these respects—
especially with provision for shade in hot weather. (Simo retorts
that, on the contrary, the sun stands at his door all day long, like
a debt-collector.) N o w the ambulacrum which Theopropides
wishes to add to his house must be like the ambulacrum in Simo's
house which he wishes to take as a model. I f Simo's ambulacrum
is a ' promenade porch ' in front of the door, then no such porch
can be supposed to exist outside Theopropides' door ; for clearly,
if the visible back-scene showed such porches in front of both doors,
it would be absurd to say that Theopropides intends to build w h a t
is manifestly there already. But if Theopropides' door has no
porch, then the ' interior scenes ' staged in front of his door cannot
be supposed to take place inside a porch. T h u s the prothyron-
theory breaks down at the outset. I n fact a ' porch ' would be
irrelevant to Theopropides' building scheme ; the association of
the ambulacrum with women's quarters, porticus and baths shows
that w h a t is meant is a covered walk in the interior of the house.
It follows that his model, the ambulacrum in Simo's house, must
also belong to the interior. Looking at line 817 again w e notice that
the phrase ante aedis goes with uestibulum and does not necessarily
refer to ambulacrum. I imagine that at the words, ' D o you see
this uestibulum in front of the house here ? ' T r a n i o opens the door
slightly and then exclaims, ' A n d that promenade ? H o w fine ! '
pointing at the imaginary peristyle in the interior. T h e spectators
cannot see it ; but there is much obvious make-believe in Tranio's
description of w h a t he sees. H e asserts that he can see a picture of
a crow making fun of two vultures, while Theopropides stubbornly
denies that he can see anything of the kind. A little later T r a n i o
scares the unhappy old gentleman b y a sudden reference to the
house-dog. W h o can believe that there was a visible picture or a
real house-dog ? T h e uestibulum is visible enough ; but all it
need consist of is the door with its two doorposts and its joints
(818-29), and the space just in front of it. As w e have seen, Simo
complains that the sun stands at his door all day long, like a debt-
collector. Presumably the reference is to his front door, where the
collector would naturally stand (so in this very play the money-
lender Misargyrides makes his w a y to Theopropides' front door),
 APPENDICES 275
I f so, then there was apparently no roofed porch to shade Simo's
front door.
T h e r e is another indication that Simo's door leads straight from
the house into the street, without any intervening porch. A t line
1063 Theopropides, at last aware of the trick played upon him,
sets a trap for Tranio. H e tells Simo's slaves to stand inside the
threshold of Simo's house, while he himself takes u p his position
' in front of the house '. So far from wanting to conceal himself in
a porch, his object is by a show of affability to lure T r a n i o to
approach the door, inside which the invisible slaves are waiting.
(In fact T r a n i o has already seen Theopropides coming out of Simo's
door). A porch would have been very much in the w a y .
T h e r e remains the opening scene of the Stichus. In the back-
ground are the houses of Pamphila, Panegyris and their father
Antipho. 1 T h e play opens with the appearance of the sisters. T h e y
seat themselves on a couch. Presently A n t i p h o comes out of his
o w n door on his w a y to Panegyris' house. H e does not see his
daughters, nor they him, for some thirty lines. A t line 87 he
approaches Panegyris' house, remarking ' I'll go in. But the door's
open '. T h e two daughters then hear his voice, come u p to greet
him, and get him to sit down with them.
Is this opening scene supposed to occur inside Panegyris' door-
w a y ? or inside a porch in front of her door ? T h e porch is ruled
out b y Antipho's words ' T h e door's open '. For if the daughters
were sitting in the porch, they would be in front of the door, and
he would see them before he noticed the door. I f they are actually
in the interior of the house, w e have to infer that A n t i p h o goes
inside to sit d o w n with them. It is clear from the text that they
hear Antipho's voice before they see him. This would be in-
explicable if at that moment he is entering the doorway inside
which they are sitting. But it is readily to be understood if they are
on the open stage, perhaps a little to one side, while A n t i p h o stands
behind them, looking with surprise at the door which Panegyris
has left open. T h e open door makes him realize that she must
have come out. W h e n they hear his voice, they turn round and
see him. T h e fact that for thirty lines they have failed to do so is
nothing unusual in N e w C o m e d y ; the actors took care not to look
in the right direction. T h e only unusual feature is the presence of
a couch, stool and cushions on the stage without any mention of
their being brought out or taken in again. W e m a y suppose that
the two women bring these articles out with them at line 1, and
perhaps that Panegyris carries them indoors at the end of the scene.
However that m a y be, line 147 (nunc, soror, abeamus intro) shows
that up till then the sisters have been on the open stage. Similarly
the slaves' drinking-party with which the play concludes takes place
on the open stage ; as it ends Stichus remarks ' L e t us go inside ' ;
then, turning to the audience, he asks for their applause.
1 See plan on p. 253.
276 THE ROMAN STAGE

THE OPENING SCENE OF THE HEAUTON T1M0RUMEN0S

T h e theory of ancient stagecraft which I have put forward will,
I believe, be found applicable to every scene in New Comedy.
Most of the difficulties which have been thought to exist are due
either to baseless assumptions or to failure to read the text with care.
T h e opening scene of the Heauton Timorumenos is a case in
point. Menedemus and Chremes are shown in conversation,
Menedemus still carrying the heavy rake with which he has been
working all day on his farm. T h e view that he is actually ' dis-
covered ' at work on the farm (represented by the stage) is, I
believe, quite foreign to ancient stagecraft; it is, moreover,
demonstrably out of keeping with the text. There was no curtain
in the days of Terence, and therefore no opening ' tableau ' ; the
stage represented merely the open space in front of the houses of
the characters. As is normal in ancient drama, the play opens
with an entry on the bare stage—in this case the entry of Chremes
and Menedemus from the ' c o u n t r y ' side-entrance. (So the
Epidicus opens with the entry of two characters from one of the
wings.) T h e opening words of Chremes explicitly refer to Mene-
demus' farm as being ' in the neighbourhood As the two are
evidently walking from the country towards their own houses, the
spectators will realize that they are going home. This obvious
fact, taken in conjunction with Chremes' words (11. 15-18 of this
scene) ' I never go out so early or come home so late but I see you
on your farm, digging or ploughing or else carrying something '
will make the situation clear. Chremes has evidently walked back
with Menedemus from his farm, and therefore the time is evening.
When Chremes urges Menedemus to put down his rake, his words
ne labora mean not ' Don't go on w o r k i n g ' but ' don't distress
yourself by holding so heavy an i m p l e m e n t ' ; compare the meaning
of labor in a few lines earlier (si quid laborist nollem) and of
laborare in Cicero Phil. v. 6 §18 ('lest his friends should suffer
fatigue if they had to carry the shields themselves ').
NOTE ON SCAENA VERSILIS A N D SC A EN A DUCTILIS
These terms occur in Servius' note on Georg, iii. 24-5. Virgil's
meaning is (I think) clear : the play is ended, the periacti revolve
(versis frontibus), the curtain rises and the stage disappears (scaena
discedat). Servius seems by scaena versilis to mean the periactus, by
scaena ductilis perhaps an alternative method of drawing the panels
off the stands instead of revolving the periactus. When Servius
says that this was done to show another picture behind the first,
he can scarcely be referring to the back-scene, which, as Virgil
expressly says, is concealed by the rising curtain. See p. 263, note 2.
But Servius or his authority may have taken Virgil's discedat as
' parts asunder '.
 APPENDIX G

THE DOORS SHOWN ON THE STAGE

THE view taken in this book is that, in addition to the entrances

in the wings (open throughout the course of the performance) there
were three practicable doors, perhaps all of the same size, in the
permanent back-scene, set unobtrusively but visibly in the wall of
the scene-building, and that these were the doors which the actors
used for the purposes of the play ; it being understood that ?iny door
which was not required in a particular play was for the time being
simply disregarded. This I imagine to have been the rule ever since
the construction of the scene-building in the fifth century ; but it
is particularly relevant to Roman Comedy. The theory that the
projecting paraskenia, or wings, were themselves sometimes used
to represent the houses of characters 1 seems to me to violate a
general principle of staging in comedy—that the stage represented
a section of street in front of the houses of the characters.8
The accompanying sketch will make clear the relation of the
house-doors to the side-entrances, as I understand it.

i lt J I
3 paces(see
Bacch .832)

That there were three doors at the back of the stage is stated by
both Pollux and Vitruvius.3 The statement of Pollux that the
central door was reserved for the principal actor, the door to the
right for the second actor and the door to the left for the least
important character, absurd as it is, seems inspired by an attempt
to connect the trinity of doors with the trinity of actors, and thus
confirms the statement that there were three doors. It will be
noticed that Vitruvius and Pollux speak of three doors in connexion
with tragedy as well as with comedy, and offer explanations of the
use to which each of the three doors is put in tragedy. These
1 See e.g. Pickard-Cambridge, Theatre of Dionysus, p. 59.
1 See Gomme, Essays in Greek History and Literature, p. 353, note 1.
* Cf. Haigh, Attic Theatre, p. 189, and p. 241 above.
19 *77
278 THE ROMAN STAGE

explanations are difficult to reconcile with the needs of our extant

tragedies, nearly all of which require only one entrance (sometimes,
perhaps, a side-door is used as well). In comedy the usual setting
requires two doors, but a few plays need three and a few only one.
No comedy requires more than three doors. The problem is solved
when we recognize that the number three was imposed not by the
needs of the play but by the permanent background, and that Pollux
and Vitruvius are merely endeavouring to suggest how this perma-
nent background could be reconciled with the varying requirements
of tragedy and comedy. When Haigh 1 tells us t h a t ' it is necessary
to distinguish carefully between the permanent doors in the wall
surrounding the stage, and the temporary doors or entrances which
were left when the scenery had been put up he is introducing a
distinction which was entirely unknown to the Greeks.
T o this extent I agree with the general view of Frickenhaus,2
who by identifying the permanent structure of the scene-building
with the visible background of all performances sweeps away the
whole basis of the fanciful theories as to the stage setting of particular
plays which have been advanced by Wilamowitz, Murray and many
others. Unfortunately Frickenhaus puts forward an equally fanciful
theory as to the number of permanent doors available in Athenian
New Comedy. He admits (p. 8) that, in the fifth century, three
doors were sometimes required—as in the Acharnians, the Peace and
the Lysistrata ; he quotes a fragment of Eupolis (24 Kock) : ' they
live here in three huts, each in his own home '. He thinks that the
middle door was large and opened inwards (i.e. into the scene-
building) in order to admit of the use of the ekkyklema, while the
two side-doors were smaller, opened on to the stage, and had no
ekkyklema. He supposes that the Athenian writers of New Comedy
gave up the use of the ekkyklema and also of the central door,
confining themselves to the two side-doors, each of which was used
at least once in each a c t ; meanwhile plays written elsewhere than
at Athens retained the use of the middle door. Therefore, when in
Latin comedies translated from Attic originals we find that three
doors are required, we are to take it that the third door has been
introduced by the Latin writer, bent on ' contaminating ' Greek
plays and influenced by such (supposedly) non-Attic originals as the
Curculio, the action of which takes place at Epidaurus and requires
three doors. A theory which depends on such uncertain assumptions
as contamination, act-division, the ekkyklema and a fundamental
difference between stage-convention in Athens and in other Greek
towns, scarcely merits serious criticism, though Pickard-Cambridge 3
thinks that Frickenhaus ' makes a strong case for supposing the
third house in Plautus' Stichus and Terence's Phormio and Hecyra to
1
A.T., p. 188.
2
Die altgriechische Bühne, 1917.
• Thfatre of Dionysus, p. 173, note 6.
 APPENDICES 279
be due to contamination It is always possible to argue that some
detail in a Latin play is unsatisfactory, and one may, if one wishes,
take the further step of supposing that the weakness is due to the
Latin translator. I can only point to the obvious fact that Latin
comedy requires sometimes one door, sometimes three, but usually
two ; and as this is true of fifth-century Greek comedy also, the
natural inference is that in this respect the Latin writers were
following their originals. If Plautus and Terence sometimes added
a third door for the fun of the thing, it seems surprising that they
did not make more use of it. According to Frickenhaus himself,
the added third door in the Phormio and the Hecyra is mentioned only
once ; in other plays (e.g. the Aulularia) to remove it would neces-
sitate a surgical operation. Perhaps it is enough to say that
Frickenhaus has not convinced Dalman, 1 who quotes Jachmann's
view that in the Aulularia the temple is essential to the plot, and
cannot be supposed to have been added by Plautus.
I come now to a famous crux. Did the three house-doors open
inwards or outwards ? Plutarch (Life of Publicola, ch. 20) tells us
t h a t ' whereas the doors of other houses at that time opened inwards,
the street door of Publicola's house was made to open outwards,
to show by such an honourable distinction that he was always ready
to receive any proposals for the public service. All the doors in.
Greece, they tell us, were formerly made to open so, which they
prove from those passages in the comedies where it is mentioned
that those that went out knocked loud on the inside of the house
first, to give warning to such as passed by or stood before them, lest
the doors in opening should dash against them.' A similar state-
ment with regard to early Greek practice is made by Helladius, a
grammarian of the fourth century A.D. (who is probably echoing
Plutarch, or Plutarch's authority). T h e remarks of Plutarch and
Helladius have been taken as reliable evidence with regard both
to stage-practice and to real life, and are stated as fact by many
editors. They were vigorously attacked more than a hundred
years ago by Becker, 2 who rejected them as untrue both of Greek
behaviour and of the construction of the house-door. T h e argument
was taken up by W . W . Mooney, 3 who agreed with Becker that
there was no such thing as ' knocking from within ', and that house-
doors in real life usually opened inwards. Mooney was nevertheless,
led by other evidence, not considered by Becker, to conclude that
the house-doors on the stage opened outwards. A different view
of this evidence was taken by Dalman (op. cit.). But Plutarch's,
statement seems still to be accepted by most classical scholars, and.
we find Professor Murray 4 repeating it as recently as 1942.
1 De aedibus scaenicis comoediae nouae, 1929.
* Charicles, Eng. ed., p. 54, note 32, and page 269.
' The House-Door of the Ancient Stage, 1914.
4 Rape of the Locks, p. 113.
 THE ROMAN STAGE

I shall begin by examining Plutarch's words. As he says ex-

pressly that, according to his authorities, Greek doors ' formerly'
opened outwards, he clearly implies that this was no longer true in
his own day. It is also apparent that the only evidence advanced
by his authorities for their statement as to former practice was the
(alleged) fact that in comedy persons going out of their houses gave
a preliminary knock to warn passers-by, who might otherwise have
been struck by the opening doors. This again implies that house-
doors which opened outwards opened into the public street. They
cannot, therefore, have been pictured as placed within a deep recess
in the house-front, or alternatively as protected by a projecting
porch ; for in either of these cases a door opening outwards would
have opened into the recess or the porch, and would not have
endangered any one in the public street. On this point we have
evidence from Aristode. He states 1 that ' Hippias of Athens
offered for sale upper stories that projected over the public streets,
together with flights of steps, railings and doors that opened out-
wards. The owners of the buildings bought them, and in this way
a large sum of money was collected.' Evidently Hippias was
exacting fines from persons whose houses encroached in one way or
another on the street; therefore doors that opened outwards opened
on to the street.
Plutarch's authority is wrong in asserting that persons coming
out of doors in comedy ' knock ' ОП their doors (κόπτουσι και
ψοφοΰσι τάς αύτων θίίρας). The word κόπτω, like its Latin equiva-
lent pulto or pulso, is never used in extant comedy of persons going
out of doors. Moreover we have express statements in ancient
authorities that κόπτω was used of knocking from outside, while the
noise made by persons going out was denoted by the word φοφω.
It has been made clear by Becker and Mooney that ψοφώ (Latin
crepo) denotes not a deliberate knock but any kind of noise—in this
case the noise made by the mere act of opening the door. The
evidence of archaeology is accepted by Mooney as showing that
in real life house-doors normally opened inwards ; from which it
follows that persons coming out of doors had no need to warn
passers-by. Nevertheless he concluded that stage-doors opened
outwards because of the use of words for ' strike ' or ' push '
(πλήττω, pello) with reference to characters coming out of doors,
especially when in a violent temper ; thus Menander, Sam. (Loeb
ed.) 88-9, 154-5, 353> has the phrase την θΰραν τ-ίπληχ*, ' he
has struck the d o o r a n d Terence, Ad. 788, writes quisnam a
me pepulit tam grauiter foris, ' who has banged so violently on
my door ? ' (again of some one coming out). These phrases have
apparently confirmed Murray in his view that persons coming out
of doors gave a knock to warn passers-by ; though it is difficult on
this view to understand the mental condition of Niceratus (Sam. 353),
1
Econ. ii. 4. I quote the Loeb translation.
 APPENDICES 281
who, in the act of bursting out of doors with threats of slaughter,
pauses to give a polite warning to the general public. Mooney, on
the other hand, understands the phrase to refer to the action of
' pushing ' the door open from within ; which of course implies
that the door opens outwards. There is further the situation in the
Wasps of Aristophanes. Here Philocleon is trying to force his way
out of the house, while the other characters, standing outside the
door, are trying to keep him in. If the door opened inwards, one
would expect that Philocleon would try to pull it open, while his
opponents would pull the other way, but Aristophanes speaks of
pushing (line 152).
Dalman (op. cit.) disagrees with Mooney's interpretation of
τΓΐ·π·ληχ€ν, pepulit, and quotes A. von Gerkan's view that these
verbs merely denote the shaking which necessarily accompanies the
opening of a door. But I think it will be helpful if at this point I
turn from the literary evidence to consider the structure and opera-
tion of an ordinary Greek or Roman house-door. Unfortunately
this subject presents us with technical problems of the most delicate
nature ; nevertheless certain fundamental points are clear, and seem
to me decisive with regard to theatrical practice.
There are three distinct problems : (1) the construction of the
house-door in real life ; (2) the construction of the stage house-door ;
(3) the use of the stage house-door for dramatic purposes.
The main parts of a Greek or Roman doorway were the sill
below, the lintel above and the doorposts or jambs at each side.
In the space thus framed were hung the two wings (fores, ualuae)
of the double door. Instead of being attached with hinges to the
doorpost, each wing swung on metal-covered pivots set at top and
bottom of the axis which turned in sockets excavated in the sill and
lintel in an angle cut out of the inner side of the doorpost.
It is obvious that the position of the sockets in relation to the
jambs decides the question whether the door opens inwards or
outwards. The archaeological evidence shows that the sockets
were on the inner side of the jambs. Moreover the sill was cut away
on the inner side so that the door, when closed, should strike against
it. It is evident that doors working on this principle would be
clumsy and noisy in operation.
The house-doors were fastened by means of bars and bolts.
The bar stretched from one post to the other, and fitted into sockets
in each post. At the bottom of each wing, and near the central
division, was a bolt which could be pushed down into a hole in
the sill. These fastenings were on the inner side of the doors.
O n general grounds we should expect the stage-doors to corre-
spond to those of real life. The evidence of the plays suggests that
they were similar in construction. The bars and bolts mentioned in
the plays appear to be where one would expect to find them—on the
inside of the doors. But the decisive point is the position of the
282 THE ROMAN STAGE

hinges, if w e m a y use this term of the pivots and sockets. It has just
been said that the doors were noisy in operation. In real life,
persons w h o wished to enter the house without being heard had no
remedy but to open the door as gently as possible, lifting it at the
same time. 1 But persons wishing to leave the house quietly had
another resource : they could pour water in one of the sockets set
in the sill. So the wife in Aristophanes' Thesmophoriazusae, stealing
out of doors to meet her lover, pours water on the hinge (lines 487-8).

Slot (b) Slot(b)

I
(Wing) (Wing)

Sill

Street

Here of course the speaker is referring to a real house-door. But

in the Curculio (lines 158-61) the old duenna ' pours a little water
underneath ' (suffundam aquolam) before opening the door from
within. Evidently the hinges of this stage-door were on the inside ;
therefore the door must have opened inwards, just like a real door.
(The object of pouring wine on the sill outside—line 8 0 — WEIS not to
silence the door but to tempt the old duenna to come out.)
I turn to the references to the stage-door in the Wasps. In the
opening scenes of this play, the inmate of the house is being kept
1 Lucian, Dial. Meret. xii. 3.
 APPENDICES 283
indoors against his will. Therefore the house-door has the unusual
function of keeping the occupant inside. To understand the passage
where both sides are said to push against the door, we should note
that, before old Philocleon had got to the door from the inside, Sosia
had already, as a precautionary measure, been ordered to throw hi*
weight against the door from outside (line 142). The effect of this,
with a door which opened inwards, would be to force it against its
bolts and thus make it difficult for any one inside to withdraw them.
Now it is clear from the text that the bar and bolts are inside, and,
also that Philocleon cannot undo them. Bdelycleon, addressing Sosia,
who is pushing against the door from outside, expresses anxiety (line
155) lest his father will ' nibble off the door-pin' which fastened the
bar in position.
Of course we must not fall into the error of supposing that every-
thing mentioned as happening inside the house, and therefore out
of sight of the spectators, does in fact happen. At line 177 Bdelycleon
announces that he will go inside, disregarding the fact that the doors
are supposed to be bolted. It is much easier for the spectator to
forget what is unseen and imagined than what is visible. We have
noticed that Euclio's instruction to his submissive housekeeper
(Aul. 103-4) to ' shut the door with both bolts ' cannot have been
carried out ; for in line 242 he enters his house without difficulty.
In general I imagine that stage-doors were not really bolted, at
least during the performance of a play. To bolt them could have
had no effect on the spectators, who were unable to see the inside
of the doors ; it might, however, have had awkward consequences
the next time some one on the stage was required by the plot to
enter the door which had been bolted.
I turn to Wasps 199-201. Philocleon has again been driven
indoors ; then Bdelycleon, anxious to prevent him from escaping
again, tells some one to
" pile a heap of stones against the door,
A n d shoot the door-pin home into the bar,
And heave the beam athwart it, and roll up,
Quick ! the great mortar-block."

(Rogers' translation). These words would at first sight be taken to

refer to visible action carried out on the stage—in other words,
outside the closed door. If so, then the door could be barred on the
outside, which seems to imply that it opened outwards ; for other-
wise what would be the use of piling stones against it ? Those who
take this view must, if they are consistent, suppose that all the elabo-
rate measures referred to were actually carried out. If so, it is sur-
prising that the chorus of Philocleon's fellow-judges, who presently
appear, do not themselves offer to unbar the door and remove the
stones, etc. But in fact they do not mention either the bar or the
stones ; and as their advice to Philocleon is to let himself down from
284 THE ROMAN STAGE

the window, 1 it would seem that they cannot get at the fastenings of
the door. I would therefore take the alternative course of assuming
that Bdelycleon's words are addressed to some one within the house,
who is asked to bolt and bar the door from within.2 In fact it is
plain that the door is not barred either on the outside or on the
inside ; for at line 529 Bdelycleon bids some one ' bring out the
memorandum-book' ; and nothing is said about the need for
removing the stones, mortar-block, etc. This inconsistency does
not matter if the barring of the door at lines 199-201 is supposed to
take place from within ; for as the spectators have never seen the
stones, mortar-block or bar, they are ready to forget about them.
It seems to have been possible to lock a house-door, or sometimes
one of the inner doors, from the outside by means of a key, in such
a way that with©ut the use of the key it could not be opened again,
even by those inside. The husband mentioned in Lysias' speech
De Caede Eratosthenis complains that his wife had locked him into
his bedroom, and had then taken the key away with her. In the
Mostellaria, Tranio, after sending the revellers indoors, has the key
brought out to him. He bids the messenger fasten the door from
inside, declaring that he himself will secure it from the outside.
We must suppose that he does then lock the door, or at least go
through the motions of locking it. The dramatic effect of his action
is, I think, to assure the spectators that even if those inside the house
forget or disregard Tranio's instructions, they will nevertheless not
be able to open the door ; and in fact the door remains unused for
the rest of the play.
In everyday usage the front door was normally closed, but not
barred or bolted, except by night. Plutarch {De Curios. 3) remarks
that it is not the ' done thing' to enter another man's house without
knocking ,· and this implies that it would be physically possible to
open the door. If people did not normally force their way into the
houses of others, the reason was not that the door was barred :
it was partly good manners, partly fear of the dog and the door-
keeper ; indeed, given the abundance of domestic servants, the
clumsy doors could scarcely have been opened without attracting
the attention of some one. Evidently it was possible to close the door,
so that it did not accidentally fly open, without having to use the
bolts or bar. We hear of the use of door-handles and knockers, but
we cannot be certain that there was anything corresponding to our
modern door-handles, which operate a catch shutting the door
securely yet leaving it possible for any one to open it from either side.
Similarly in the theatre we may take it that the house-doors were
1
' So now to the window lash the cord ' (1. 379). But Philocleon seems to be
on the roof, not at a window (350, 355). There is nothing to show that he comes
out through a window.
2
That some domestics have been left indoors is indicated in line 166, where
Philocleon bids them give him a sword.
 APPENDICES 285

normally shut. Sometimes characters in plays express surprise to

find a house-door shut. I do not agree with Mooney that what
really surprises them is to find the house-door locked or bolted.
I do not see how any one could have known from outside whether
the door was locked, unless he tried to open it. But it has just been
pointed out that people did not normally try to open the doors of
others. Moreover we have a case of a character expressing surprise
at finding the door open (St. 87) ; and we can scarcely suppose that
by ' open ' Antipho meant ' unlocked as he has just declared his
intention of going in, which would have been impossible if the door
was locked. I take it that Panegyris' door is wide open, the reason
being that she has failed to pull it to when coming on the stage at
the beginning of the play. Taking the word ' s h u t ' in its plain

1ШСТ
—4WW "' *m ""' » H W

Θ Φ Φ mi //"τπττνττϊπττ·
Φ® 9 β И| r ft _i>.
-Ш-Ш-Г A •\
Woman unlocking front door Woman looking out of door
sense, one has to admit that the closed but unbolted door of Roman
comedy displays surprising resistance to the violent thumping and
kicking to which it is so often subjected. Perhaps the stiffness of
the hinges kept the door in position. On one occasion it is de-
lfl&rately left ajar (Men. 3 5 1 , 362).
All our evidence supports the reasonable view that the doors
shown on the stage corresponded both in appearance and in con-
struction with the doors of real life, except that, in the days when the
theatre was itself a temporary structure, the house-doors were
presumably slight affairs, made completely of wood. I imagine
that the scene-building was used as a store-room for theatrical
properties ; if so, it was presumably possible to lock the doors,
or at least one of them, from outside.
Everything on the inner side of the doors was invisible to the
spectators, and had to be left to their imagination. In Bacch. 8 3 4 - 5
the door is supposed to open into the dining-room, which is visible
to the characters on the stage when they open one wing of the door
an inch or two. In Most. 8 1 7 the (slightly opened) door gives a
286 THE ROMAN STAGE

view of the colonnade at the back of the house. All this and much
more the spectators were willing to take on trust. Of course they
knew that what really lay behind the doors was the interior of the
dressing-room, with actors waiting for their cues, perhaps changing
their masks or costumes under the watchful eye of the manager.
The art of the theatre consists partly in making imaginative
use of material means and limitations. The very clumsiness of
ancient house-doors was turned to good account by the dramatists,
aided by the goodwill of the spectators, who were familiar with the
convention and readily joined in the game of make-believe without
which the theatre could never have come into existence. If we try
to see Roman drama with Roman eyes, we shall enjoy it more
keenly; and the true end of all drama is enjoyment.
 SHORT BIBLIOGRAPHY

TEXTS

I n citing passages from Latin drama I have used the following

texts :
Plautus. Oxford edition (W. M . Lindsay).
Terence. Oxford edition (Lindsay and Kauer).
Tragicorum Romanorum Fragmenta (O. Ribbeck).
Comicorum Latinorum Reliquiae (O. Ribbeck).
Seneca. L o e b edition (F. G . Miller).

I have found Professor P. Nixon's L o e b edition of Plautus most

useful, as he takes a consistent and realistic view on matters of
staging. T h e text follows that of Leo, w h o omits many lines on
purely arbitrary grounds. Professor A . Ernout's Bude edition is
admirably sound and scholarly. Professor Ε. H . Warmington's
L o e b edition of the Fragments of Old Latin is accurate and refreshingly
independent.
T h e English reader m a y consult Professor G . E. Duckworth's
The complete Roman Drama for translations of all extant plays,
including the post-classical Qyerolus, with useful notes.

ILLUSTRATIONS

There is a wealth of material in Dr. Margarete Bieber's History

of the Greek and Roman Theater (Princeton, 1939). O n the interpre-
tation of works of art I find myself in general agreement with
Dr. A . W . Pickard-Cambridge, Theatre of Dionysus in Athens
(1946), reviewed by J . T . Allen, C.P. 42 (1947), pp. 257-9, and
M . Bieber, A.J.P. 69 (1948), pp. 97-100.

SUMMARIES OF E A R L I E R WORK

I have constantly consulted Schanz-Hosius, Geschichte der

Römischen Literatur, V o l . I. For a critical treatment of the
whole subject no other work has been so useful as that of
G . Michaut, Histoire de la comedie romaine, Vols. I and II. Pauly-
Wissowa's Real-Encyclopaedia is, of course, indispensable as a work
of reference, though coloured by the habits of mind characteristic
of much German scholarship. Daremberg et Saglio's illustrated
dictionary is admirably lucid and helpful on technical points.
Other general works of great value are :
287
 THE ROMAN STAGE

Allen, J. Т., Stage Antiquities of the Greeks and Romans and their
influence (1927).
Flickinger, R., The Greek Theater, ed. iv (1936).
Legrand, P., Doos, Tableau de la comedie grecque pendant la periode
dite nouvelle (1910). Still perhaps the most complete account of
New Comedy. (English translation, Greek New Comedy, by James
Loeb.)
Lejay, P., Plaute (1925).
I mention below some books and articles to which I owe a
particular debt, or which illustrate interesting trends of study :
Buck, С. H., A chronology of the plays of Plautus (Baltimore, 1940).
Duckworth, G. E., ' The structure of the Miles Gloriosus,'
C.P. 30 (1935), pp. 228-48. (Dealing with the hypothesis of con-
tamination.)
' P l a u t u s : the other nineteen plays,' C.W. (Dec., 1947),
pp. 82-91. This is a very fair review of recent work.
Duckworth, G. E., Review of Arnaldi, ' Da Plauto a Terenzio,'
A.J.P. 70 (1949), pp. 221-4.
Dunkin, P. S., Post-Aristophanic Comedy (Illinois, 1946). Reviewed
by Duckworth, A.J.P. 68 (1947), pp. 419-26.
Enk, P. J., ' Quelques observations sur la maniere dont Plaute
s'est comport6 envers ses originales,' Rev. de Phil. 64 (1938),
pp. 289-54.
Frank, Tenney, ' Naevius and free speech,' A. J.P. 48 (1927),
pp. 105-10. Life and Literature in the Roman Republic (California Press,
1930). ' Status of actors at Rome,' C.P. 26 (193 r), pp. 11-20. ' The
actor Pellio,' A. J.P. 53 (1932), pp. 248-51. Review of Jachmann,
Plautinisches und Attisches, A.J.P. 53 (1932).
Gow, A. S. F., ' O n the use of masks in Roman Comedy,'
J.R.S. II (1912), p. 65.
Harsh, P. W., Studies in Dramatic ' Preparation ' in Roman Comedy
(Chicago, 1935).
Iambic Words and Regardfor Accent in Plautus (Stanford University
Press, 1949).
Hough, J. N., ' The Development of Plautus' Art,' C.P. 30
(1935). PP· 43-57·
' The understanding of Intrigue : a study in Plautine Chrono-
logy,' A. J.P. 60 (1939), pp. 422-35.
Keller, Ruth Mildred, ' Iste deiktikon in the early Roman
Dramatists ' ( Т А Р А 77 (1946), pp. 261-316). (Disproves Lindsay's
theory that the use of demonstratives is a guide to the movements of
actors on the stage. Reviewed by Whatmough, C.P. 44(1949), ρ. 214.)
Kuiper, W. E. J., Grieksche Origineelen an Latijnsche Navolgingen,
Amsterdam (1936).
' Two Comedies by Apollodorus of Garystus : Terence's
Hecyra and Phormio ' (Mnemosyne, Supp. 1, 1938). Reviewed by
Tredennick, C.P. 53 (1939), 66.
 BIBLIOGRAPHY 289

Knapp, С., ' References in Plautus and Terence to plays,

players and playwrights,' C.P. 14 (1919), pp. 35-55· ' References to
painting in Plautus and Terence,' C.P. 12 (1917), pp. 143-7.
Little, А. M . G., ' Plautus and popular d r a m a ' {Harvard
Studies in Class. Phil., 49 (1935), pp. 205-28).
Momigliano, Α., Review of Robinson, Freedom of Speech in
the Roman Republic, J.R.S. (1942), pp. 120-4.
Noble, R., Shakespeare's use of song (1923).
Norwood, G., The art of Terence (1923). Plautus and Terence
(1932). (Readable and independent, if somewhat extreme.)
Post, L. Α., 'Dramatic Infants in Greek,' C.P. 34 (1939),
pp. 198-208.
Prescott, H. W . , ' Link Monologues in Roman Comedy,' C.P. 34
(1939), pp. 1-23, 116-126. ' T h e antecedents of Hellenistic
Comedy,' C.P. 12-14 (1917-9).
Poulsen, F., Etruscan Tomb Paintings (Oxford, 1922).
Sedgwick, W. В., ' Plautine Chronology,' A. J.P. 70 (1949).
' The origin and development of Roman Comic metres,' Classica
et Mediaevalia (1949), pp. 171-81.
Saunders, C., Costume in Roman Comedy (Columbia, 1909).
Shisler, F. L., ' The use of stage business to portray emotion
in Greek Tragedy,' A.J.P. 66 (1945), pp. 277-97.
Sihler, E. G., ' The Collegium poetarum at Rome,' A.J.P. 26
(1905) p. I.
Webster, Т . B. L., ' South Italian Vases and Attic Drama,'
C.Q. 42 (1948) pp. 15-27.
' The Masks of Greek Comedy,' Bulletin of the John Rylands
Library, 22 (1949)·
Westaway, K . W., The original element in Plautus (Cambridge
»917)·
 INDEX

Accius, 16, 3 1 , 32, 60-2, 64, 1 1 1 - 1 9 , Cleon (mime actor), 142

225 Clifford, H. R., 202
act, 44, 188-210 collegium poetarum, 18, 76, 1 1 8
acting, actor, 1 1 , 13, 14, 18, 71, 76, commedia dell'arte, 140
158-60, 175, 221, 229-32 conquistor, 161
actress, 141, 143, 144, 266 contaminatio, 88-94, 100-4
Aesopus (actor), 158 Cornford, F. M., 10
Afranius, 92, 120-7 costumes, 176-87
Allinson, F. G., 197, 244 cothurnus, 180, 183, 257
Ambivius Turpio, 79, 156-8 curtain, 44, 71, 141, 146, 1 7 1 - 2 , 259-
amphitheatre, 164 66, 276
Andronicus, Livius, 5, 16-22, 23, 61, Dalman, С. O., 249, 279, 281
63. 76, 157 dance, 9-12, 44, 214-5, 226
angiportus, -um, 173, 248-55 Demetrius (actor), 229
Antiodemis (mime-actress), 144 Dieterich, Α., 140, 262
Aprissius, 139 Diomedes (grammarian), 185, 201,
Aquilius, 105 212-3, 256, 262
archimimus, -a, 145, 231 dissignator, 161
Atellana, 12, 13, 15, 29, 37, 129-40, diuerbium (deuerbium), 2 1 1 - 2 4
184, 230-1 Donatus, on act-division, 188-210
Atilius, 105 on costumes, 177-9, 185
Atta, 120-7 on the curtain, 266
aulaeum. See curtain on forms of drama, 256-8
awnings, praef., 162 on music, 212 ff.
on Terence, 82, 87, 88, 90, 92-5, 98-9
doors, 1 1 7 , 168-9, 174, 249-55. 277-85
Balbus, L . Cornelius, 32, 34
Dörpfeld, W., 236-8
Bassulus, M. Pomponius, 229
Dossennus, 131 ff.
Bathyllus, 226
doubling of parts, 43, 159
Becker, W. Α., 279-80
Duff, J . Wight, 21, 233
Bieber, Μ . , praef, 271
Dziatsko, K., 86
Bucco, 1 3 1 ff.
Bulle, Η., 263
Butler, Η. Ε., 263 Enk, P. J . , 55
Ennius, 17, 25, 31, 60-8
Eucharis, 144
Caecilius, 61, 75-80, 81, 83-4 Evanthius. See Donatus
cantica, 19, 64, 65, 71, 161, 2 1 1 - :
226
cantor, 221. See actor fabula, types of, 256-8
Capps, Ε., 197 Fensterbusch, С., 241
Catienus, 71, 261 Fescennine verses, 9 - u , 14
Catullus (mime), 262 Fiechter, E. R., 261, 264-6
cauea, 163 Flaccus (tibicen), 160, 221
centunculus, 145, 147 Flickinger, R., 196, 201, 204, 240
Chambers, Ed., 191 flute, flute-player. See tibia, tibicen.
choragus, 158. See also producer Fraenkel, E., 224
Choricius of Gaza, 232 Freti, Α., 193, 2об
chorus, 19, 44, 65, 188-210 Frickenhaus, Α., 278-9
Cincius Faliscus, 185 Fufius, 71, 261
 INDEX 29I
Gardner, Percy, a68-g Monro, D. В., 215-6
Gerkan, Α . von, 281 Mooney, W . W., 279-84
gesture, 132, 141, 183 Mountford, J. F., praef.
ghosts, on stage, 71, 246, 261 Mummius (dramatist), 139
Gomme, A . W., 277 Murray, Gilbert praef., 278-80.
music, 9-12, 19, 44, 57, 159-61, 210-24
Haigh, A. E., 195, 236-θ, 241, 244, 277
Harsh, P. W., 202, 248-9
Naevius, 23-33, 37. 38> 57. 61. 75. 9».
hilarotragoedia, 15 n o , 185
histrio, 11, 18, 221. See actor
Nauplios puppet-show, 188, 198-200
Hunter, Mark, 191
Navarre, O., 178, 269, 270
Neoptolemus of Parium, 200 ; cf. 209
ikria, 237-8
Nero, 225, 227-8
illustrations in manuscripts, 6, 100,
New Comedy, 40-5, 240-7 and passim
176, 266
Nixon, P., 219, 271 ; cf. 171
indoor scenes, 171
Noble, R., 223
interlude, 6
Norwood, G., 37
Jachmann, G., 201, 202, 279
Johnston, Mary, 202, 240-6 oeeentare, 30, 220
Juventius, 105 orchestra, 12, 168
Joseph, B. L., praef. Oscan farce, 129, 130
Oxyrhynchus farce, 148
knocking. See doors Oxyrhynchus mime, 148, 231
Körte, Α., 189, 196, 197, 2θ8
Kuiper, W. E. J., 192 Pacuvius, 69-74
Kurrelmeyer, Carrie, 155 palliata, 176, 256-8
pantomime, 1, 226, 230
Laberius, Decimus, 146-50 Pappus, 131-4, 137
' left and ' right' on stage, 173, 240-7 Pattison, Bruce, 223
Legrand, P., 196, 198 Pellio (actor), 57, 156
Leo, F., 171, 193, 198, 199, 209, 224 performance, notice of, 162
Licinius Imbrex, 105 performance, preliminary, 88, 161-2
Lindsay, W. M., 211, 217, 223 periacti, 240-7, 269, 276
Livius. See Andronicus persona, 12, 185. See mask
Livy (VII, 2), 11, 13-14, 18, 133, 211 phallus, 10, 145
Lodge, Lexicon Plautinum, 249 Phersu, 12, 185
Lucan (as writer of libretti), 226 phy takes, 13, 15, 142
lucar, 156 Pickard-Gambridge, A . W., praef., 195,
Ittdi, 10, 11, 154-5 242, 269, 271, 277
Lundström, V . , 372 planipes, 37, 143, 146
Luscius Lanuvinus, 82, 88, 97, 106-7 Plautius, 36
Plautus, 35-59 and passim
Maccus, 37, 131 ff. Pollux, 178, 187, 241 ff., 277-8
Maidment, K . J., 192, 196-7 Pompilius (dramatist), 118
Manducus, 131 ff. Pompilius (mime-actor), 148
manuscript, author's, 4, 156-7 Pomponius, 134-40, 260
Marcipor, 57, 160, 221 Porcius Licinus, 16, 84
masks, 12, 71, 142, 175, 178-87 praeco, 161
Mazois' theory of curtain, 261 ff. praetexta(ta), 1, 29, 31-4, 184, 228
Melissus (dramatist), 128 producer, 156-8
Menander, 41-5, 77-80, 188-208 programmes, 162
metre, 21, 44, 65, 116, 211-24 prologue, 151-5, 233-9
Metrobius (mime-actor) 144 property-manager. See choragus
Michaut, G., 189, 209 proscenium, 163, 168
mime, 1, 8, 15, 37, 78, 119, 130, 132, protasis, epitasis, catastrophe, 209
134, 141-50, 184, 231-2, 260-6 prothyron, 271-5
Minucius Prothymus, 185 Protogenes (mime-actor), 144
Mommsen, Т., 130 Prou, V . , 188, 198-200
292 THE ROMAN STAGE

Publilius Syrus, 147-50 Strabo (dramatist), 117

puppet-show. See Nauplios puppet- Strabo (geographer), 130
show Stratocles (actor), 229
Purser, L. C., 226 Syrus. See PubliSus Syrus
Pylades (actor), 226
tabernaria, 121, 256-8
Quintilian, 183, 225, 229 Terence, 81-104 passim
Theatre Royal, Bristol, 164
recitationes, 225-9 Theodora (empress), 232
Rees, Kelley, 241, 272-4 thyromata, 271, 272
Reich, H., 141 tibia, 11, 160, 211
revivals, 4, 5, 75, 127, 156, 157, 229 tibicen, 18, 19, 160, 198, 204,205, 211-24
Rhinthon, 15 ; cf. 147 tickets, 162
Ribbeck, O., 103 Titinius, 120 if.
ricinium, 145 Titius (dramatist), 118
Ritsehl, F., 4, 163, 239 togata, 29, 120-8, 184, 229, 256-8
roof (of scene-building), 46, 169, 172 Trabea (dramatist), 105
roof ( of stage), 169, 265 trabeata, 128
Roscius, 158, 185 Turpilius, 107-g
Rupilius (actor), 71 Tyrrell, R. Y., 112-3

samio, 140 uersura, 169

Santra (dramatist), 118 unities, 29, 44, 192
satura, 13, 14
scabillarii, 146, 161 ; cf. 260 vacant stage, 203-5
scaena, scaenicus, 11, 163, 168 Valerius (dramatist), 106
scaena ductilis, scaena uersilis, 270, 276 Varius, 225
scene, scenery, 172, 261, 267-76 Varro, 5, 24, 36, 37, 57, 77, 206-7
seats, 4, 11, 12, 162-4, 273-9 Vatronius, 105-6
Seneca, 1, 2, 71, 117, 150, 209, 210, Vergilius Romanus, 229
225-8, 262 Virgil (Georgics iii, lines 24-5) 260, 276
Shakespeare, 2, 190-2, 222-3, 267 Vitruvius, 169, 175, 241 ff., 277-8
side-entrances, 169, 172, 173, 240-7 Volcacius Sedigitus, 63, 109-10
siparium, 146, 259-66
Skutsch, F., 201 Warmington, Ε. H., 73, 121
soccus, 179, 180, 258 Webster, Τ. B. L., prtuf., 208
spectacula, 234-9 Weil, H-, 199, 200
spectators, 162, 165-7 Weissinger, R. Т., 189, 190, 197, 200,
stage, 15, 168-9 202, 203
stage-conventions, 170-5 Westaway, K . W., 37
Statius, Caecilius. See Caecilius Wilamowitz, U. von, 224, 278
Statius, Papinius (as writer of libretti), Wilson, Dover, 191
226 window, 172, 284

PRINTED BY MORRISON A N D GIBB LTD., LONDON AND EDINBURGH