CAMBRIDGE CLASSICAL TEXTS AND

COMMENTARIES
EDITOllS

C. 0. BRINK D. W. LUCAS F. H. SANDBACH

10
THE TRAGEDIES OF
ENNIUS
THE TRAGEDIES OF
ENNIUS
THE FRAGMENTS
EDITED WITH AN INTRODUCTION
AND COMMENTARY

BY
H.D.JOCELYN
Reader in Latin in the
University of Sydney

CAMBRIDGE
AT THE UNIVERSITY PRESS
1967
 Published by the Syndics of the Cambridge University Press
Bentley House, 200 Euston Road. London, N.W. 1
American Branch: 32 East 57th Street, New York, N.Y. 10022

C Cambridge University Press 1967

Library of Congress Catalogue Card Number: 67-11525

Printed in Great Britain

at the University Printing House, Cambridge
(Brooke Crutchley, University Printer)
 CONTENTS
PREFACE pagevu

INTRODUCTION 1

I Athenian Drama and the Roman Festivals 3

Il The Hellenising of the Roman Stage 12

m Athenian Drama and the Roman Poets 23

IV The Form of Roman Tragedy 29
V Ennius 43
VI The History of the Text of Ennius' Tragedies 47
VII The Titles of Ennius' Tragedies 58

THE FRAGMENTS 65

COMMENTARY 159

ADDENDA 427
BIBLIOGRAPHY 428
CONCORDANCE I 433
II 438
III 441
INDICES

I Ennius' Tragic Vocabulary 445

D Ennius' Metres 465
m Matters Discussedin Introduction and Commentary 468
V
 PREFACE

This book contains material from a dissertation submitted in

the year 1961to the University of Cambridge for the degree of
Doctor of Philosophy. No alterations of substance were made
to the typescript after June 1966.I have tried, using printed
editions of the authors who quote Ennius, to present the evidence
which exists concerning each tragedy whose title is known and
to discuss the questions to which this evidence is relevant. I have
put together under the rubric INCERTA those pieces of verse
which are quoted by ancient authors in company with Ennius'
name and which have been attributed by one modem scholar
or another to the tragedies.
My debts are many. The Faculty of Archaeology, History
and Letters of the British School at Rome, the Faculty Board
of Classics in the University of Cambridge, the Council of
St John's College, Cambridge, and the Senate of the University
of Sydney gave me financial help at various times. Dr W.
Ehlers allowed me in the northern summers of 1957 and 1958
to use the archive and library of the Institute for the Thesaurus
Linguae Latinae. Professor 0. Skutsch read and criticised a
section of my typescript, showed me unpublished work of his
own and discussed several problems with me at great length.
Mr D. W. Lucas and Dr D. R. Shackleton Bailey read sections
of my typescript and made criticisms. Professor C. 0. Brink
read many drafts of the whole work, devoted much time to
writing extensive criticisms of these drafts, tolerated occasior.al
unwillingness to accept good advice and encouraged me in
moments of weariness to persevere. The officers of the
Cambridge University Press indulged my vagaries and lavished
their skill on the somewhat forbidding material with which
vu
 PREFACE

I supplied them. Professor S. Mariotti and Mr F. H. Sandbach

corrected and improved the proof pages in many places. I
thank these whom I have named and the many others whose
friendship, knowledge and wisdom I exploited.
H. D.JOCBLYN
New Haven, Connecticut
March1967

Vlll
INTRODUCTION
 I. ATHENIAN DRAMA AND THE
ROMAN FESTIVALS
The twenty-two dramatic treatments of Greek heroic legend
whose remains are discussedin this volume were adapted from
Athenian tragedies between the years 203 B.c. and 169 for per-
formance at ludi scaenid.These ludi scaeniciformed an impor-
tant element both of the regular yearly festivals managed by
the civil magistrates in honour of luppiter, Apollo, Ceres,
Magna Mater and Flora and of those held for some special
purpose, such as to thank a deity for a magistrate's military
success or to honour the spirit of a deceased member of the
ruling class.1 They were believed to have been introduced to
Rome from Etruria in 364 as a means of placating the divine
senders of a plague. i The old agricultural festivals of the so-
called Calendar of N uma, which continued to be managed by
the priests, had no place for them.3 The earliest adaptation of an
Athenian play which scholars of the first century B.c. could find
recorded was performed at Judiscaenidin 240, the year follow-
ing the first capitulation of Carthage to Roman arms.4
1
Sec Habel, RE Suppl. v (1931), s.v. ludi publid, 6o8 ff., L. R. Taylor,
T APhA LXVm (1937), 284 ff. For the continuing religious associations of the
ludi scaenid see J. A. Hanson, Roman Theater-Temples (Princeton, 1959),
pp. 3 ff.
a Cf. Festus, p. 436.23 tf., Llvy 7.2, Valerius Maximus 2.4.4.
3 Naeviw (Com. IIJ) identified the old Liberaliawith the Attic f110M1a.
Plautw however (Cist. 89, 156, Cure. 644, Pseud.S9) seems to have thought the
differences too great to justify the identification. Tertullian's statement (De
spect.10. 19) that scenic games were properly calledLiberaliais based on second-
hand theorising rather than firm knowledge.
4 H. Mattingly Jr., CQ N.S. vn (1957), 159-63, produces no good reason
for doubting the veracity of the antiqui commentariiconsulted by Varro
(Gellius 17.21.42), Atticw and Cicero (Brut. 72, Tusc. 1.3, Cato so). Accius,
whose Didascalica set the first production of a play by Livius Andronicus in 197,

3 I-2
 INTRODUCTION

Scholars and literary amateurs of the first century B.c. regarded

those plays of the previous two centuries whose scenes were set
in Greece as being the work of the men who wrote the Latin
acting scripts and yet were conscious ~t particular Greek
tragedies and comedies underlay them all.1 The Greek philo-
sophical dialogues which Cicero adapted were full of quota-
tions from Attic drama. Cicero sometimes replaced these with
quotations from the scripts of Latin stage adaptations and some-
times with his own translations of the verses quoted. In the first
case he either left the quotation anonymous or named the Latin
adapter while in the second he always named the Greek

may have been quite well aware that Naevius and Plautus had produced plays
before this date (sec W. Hupperth, Horaz iiberdie scamicaeoriginesder Romer
[Diss. Koln, 1961], pp. s ff., 10 ff., H. Dahlmann, 'Studien zu Varro, "De
poetis'", Abh. Ale.d. Wiss. u. d. Lit. Mainz, Geistes-u. Sozialw. Kl. 1962, Nr.
10, 29 ff., C. 0. Brink, 'Horace and Varro', EntretiensHardtIX (1963], 192).
E. Fraenkel seems to have based his view that Livius produced only one play,
and that a tragedy, in 240 (RE SuppL v [1931], s.v. Liuius, 598 f.) on Cicero,
Brut. 72, Tusc. 1. 3, Cato s. These passages should not be pressed nor, it
must be admitted, should Gcllius' words Liuius poetafabulas dome . .. coepit.
However according to Cassiodorus (Chronica)Livius produced both a tragedy
and a comedy in 239. This date is a clear error for 240 but the rest of Cassio-
dorus' statement may be reliable. For Livius as the founder of both comedy and
tragedy cf. Donatus, De com. s .4, Gloss. Lat. 1128, s.v. comoedia,1 568, s.v.
tragoedia.
' There is nothing in ancient discussionsof republican tragedy and comedy to
support the statement frequently made in modem times that the Latin poets
occasionally wrote quite independently of particular Greek models (cf. J. J.
Scaliger, Coniectaneain M. TerentiumVarronemde linguaLatina [Paris, 1565],
p. 6, H. Columna, Q. Ennii PoetaeVetustissimiquaesupersuntFragmenta[Naples,
1590], p. 408, F. Nieberding, flias HomeriabL. Attio Poetain DramataConuersa
[Conitz, 1838], p. 3, T. Ladewig, Analeda Scenica[Neustrelitz, 1848], pp. 8, 29,
38, G. Boissier, Le poete Attius [Paris, Nimes, 1857], p. 6o, W. S. Teuffel,
CaeciliusStatius,Pacuuius,Attius, Aftanius [Tiibingen, 1858], pp. 23, 29, U. von
W ., in T. von Wilamowitz-Moellendorff, Die dramatische TechnikdesSophokles
[Berlin, 1917], p. 315 n. 1,J. Vahlen, EnnianaePoesisReliquiael[Leipzig, 1903],
p. ccvn [very tentatively], E. M. Steuart, A]Ph XLVU [1926], 276,J. Heurgon,
Enniusn [Paris, 1958], p. 143). Cicero, Fin. 1.7 refen to the relations between
Ennius' Annalesand Homer's epics; Gloss. Lat. I 568, s.v. tragoedia(tragoedias
autemEnnius FEBB omnesex Graecistranstulit),to plays on Roman historical
themes such as the Sabinae(cf. Horace, Ars 285-8).

4
 ATHENIAN DRAMA AND THE ROMAN FESTIVALS

dramatist. 1 Scholars normally at this time gave the name of the

Latin poet and his Latin tide when quoting from an acting
script but very occasion.allygave the title of the Greek original;
on one occasion a play is quoted with the name of the author of
the original as well as that of the adapter. 2 The writers of the
third and second century acting scripts were by now thought of
with varying degrees of admiration as the fathers of an indi-
genous Roman literature and it is clear that performances of
these scripts were advertised with their names to the fore.
. When the early adaptations were being made a different
attitude to them probably prevailed among the managers of the
ludi scaeniciand the citizens who attended the theatre. The
literary and artistic culture that had spread out from Athens
over the whole Greek-speaking world was then making the
same appeal to certain of the Roman governing classas it had to
the rulers of Etruria and other barbarians in earlier centuries.
The works that had been produced for performance at the
festivalsof Dionysuswere among the brightestjewels of that
culture. They could have been produced in the original Greek
at the Roman ludi, as in later times they perhaps sometimes
were,3 but one of the purposes of the ludi was to impress the
peers of the presiding magistrate and their clients. To many of
these Greek culture in its less adulterated forms was an alien and
suspect thing. The pride of the majority in Roman race, lan-
guage and tradition was satisfied by the form in which the
Greek works were presented; the enthusiasm of the minority
1
Cf. G. Przychocki, FAsXXXII(1929). 215 ff.• Fraenkel. GnomonVI (1930).
663. Seneca followed Ciccro's practice; contrast Epist. n5 .14-15 with 95. 53
and 102. 16.
3
Sec below• p. 59.
3 The evidence collectcd by F. G. Welckcr. Du grkchischenT,agodienmit
Rucksichtauf den tpischenCyclus (RhM Suppl. n. Bonn. 1839-41). pp. 1323 ff..
is of an ambiguous kind. Polybius 30.22. Tacitus. Ann. 14.21. Plutarch. Brut.
21 do not ncccssarily refc:£to drama at all. Suetonius (Jui. 39. Aug. 43) talks of
performances by 'omniwn linguarum bistrioncs •.

s
 INTRODUCTION

for classical Greek culture by their advertisement as the works

of the classical Athenian tragedians and comedians. Slaves and
foreign immigrants did the work of adaptation and it is un-
likely that their names carried much weight while they lived.
The uncertainty among Roman scholars about the authorship
of certain comic scripts popularly ascribed to Plautus 1 may
have been partly due to failure by the magistrates of earlier
times always to record the Latin poet's name in their commentarii.
A Sophocles might be made a magistrate at Athens but at Rome
a magistrate could be pilloried for having an Ennius in his
retinue. 1
The opening speech of the Rudensdescribes a storm as follows
(83-8): pro di immortales,tempestatem quoiusnwdiI Neptunusnobis
nocte hac misit proxuma. I detexit uentus uillam- quid uerbis
opust? I non uentusfuit uerum AlcumenaEuripidi, I ita omnis de
tectodeturbauittegulas; I inlustriores
Jedt Jenestrasqueindidit.It is
unclear whether the speaker has in mind a stormy reaction by
the heroine of Euripides' play to her husband's accusations3 or a
real storm within the action of that play ;4 likewise whether the
identification stood in the Diphilian originals or was added by
Plautus himsel£ 6 In any case one cannot imagine such a state-
1
Cf. Terence, Eun. 25 (?), Varro, Ling. 6.89, Gellius 3.3.
i For the traditional Roman suspicion of the maker of verses cf. Cato, Mor. 2
poeticaeartishonosnon erat; si quis in ea re studebataut sesead conuiuiaadplicabat
g,assatoruocabatur. 3 C£ Plautus, Amph. 812 ff.
4 So R. Engelmann, Ann. 1st. Corrisp.Arch. xuv (1872), 16, Beitriigezu

Euripides I: Alkmene (Berlin, 1882), p. 11, ArchiiologischeStudien zu den

Tragikem (Berlin, 1900), pp. S9-00,
5 So S. Vissering, ~estiones Plautinae I (Amsterdam, 1842), p. 42, T.
Bergk. Ind. lectt.Marburg 1844, XI ( = Kl. phil. Sehr. 1 22s), Ladewig, Anal.
seen.p. 5.
6
So Fraenkel, Plautinischesim Plautus (Berlin, 1922), p. 68 ( = Elementi
Plautiniin Plauto[Florence, 1900], p. 64); cf. the Addendato the Italian transla-
tion, p. 403. The addition of the Greek poet's name suggests Plautine author-
ship; for the late fourth century method of referring to famous tragedies cf.
Menander, Epitr. 767 Tpaytk'l)Vfpa, 0-01~i')cnvf~ A(iyris6~1'1"(contrast Aristo-
phanes, Thesm. 134 ff.).
6
 ATHENIAN DRAMA AND THE ROMAN FESTIVALS

ment being made on the Roman stage unless a play about

Alcmena had already appeared there and been advertised as
being in some sense Euripides' •AA1q111V1l- Admittedly there is no
undisputed evidence elsewhere for the existence of an adapta-
tion of this play, 1 but considering the fragmentary record of
third and second century dramatic productioni we should not
be surprised to find such a reference in the Rudens.The pro-
logues of the Poenulus(v. I) and the Eunuchus(vv. 9, 19-20)
refer to Latin versions of Greek plays in this way although it
was theatrically possible for the speakers to use the first century's
customary mode of reference if it had then been in general use.
The only Greek plays which are said by knowledgeable
ancient authorities to have been adapted for performance at
Roman ludi in the late third and early second centuries were
originally composed for the festivals of Dionysus at Athens in
the fifth, fourth and early third centuries.3 The only dramatists
among the famous Greeks mentioned in Plautus' comedies are
the Athenians, Euripides,4 Diphilus and Philemon.5 The forms
of comedy written by Epicharmus and others at Syracuse in the
fifth century and by Rhinthon at Tarentum in the third and
the imitations of Athenian drama made by Machon and the

• O.Ribbcck (sec Corollariumin TragicorumRomanorumFragmenta'[Leipzig,

1871], p. LXDI) took Marius Victorinus, Gramm. VI 8.6 ff. to refer to tragic
personages rather than to titles; hence Alcumenadocs not appear in his 'index
fabubrum •. Plautus, Amph. 91 ff. can be plausibly interpreted as referring to the
play postulated. Sec also below, p. 63.
a Of the 130 comic titles once attributed to Plautus we now know only 53.
Several tragic titles occur only once or twice in our sources, sometimes even
without the Latin adapter's name (Laomeckmat Schol. Vcron. Verg. Am. 2. 81 ;
Penthesileaat Fcstus, p. 206.3).
3 Bcrgk. Commmtationumde Reliquiis ComoediaeAtticae Anliquae Libri II
(Leipzig, 1838), p. 148, explained the fragment of Varro's De pottis quoted by
Priscian, Gramm.u 469. ')-dei,uk ad Siculosseadplicauit-asreferring to Plautus
(c£ Horace, Epist. 2.1.58) but thought that Varro was speaking generally of
the reputed np&Tas eupnfis of the comic genre.
4 Rud. 86.

s Most. 1149. Philcmon was not, of course, an Athenian by birth.

7
 INTRODUCTION

tragedians of the 'Pleiad' for the festival of Dionysus estab-

lished at Alexandria by Ptolemy Philadelphus (285-247) 1 ob-
tained for themselves a certain literary notoriety and it is not
utterly impossible that plays from Alexandria1 and Magna
Graecia3 were adapted for the Roman stage. Ennius adapted
non-dramatic works by Epichannus, Archestratus, Sotades and
Euhemerus for the private delectation of his aristocratic patrons
and perhaps imitated Callimachus'AITiain the proem of his
epic Annales4but the public presentation of dramatic works
outside the classicalAthenian repertoire would have been quite
another matter. As late as 45, when a large number of poetical
works from Alexandria and other Hellenistic centres had
achieved a sort of second-grade classicalstatus and were being
enthusiastically studied and imitated by poets writing in Latin,
Cicero chose adaptations of the non-dramatic poems of Eu-
phorion to set against what he believed to be a version by
Ennius of a Euripidean tragedy.5 New plays continued to be
presented at the Dionysiac competitions of Athens until well
into the first century A.D. 6 but no playwright productive
1
Cf. Wilamowitz, Hellenistische Dichtung1 (Berlin, 1924), pp. 166 ff.
3
So A. Rostagni, RFIC xuv (1916), 379-97 ( = ScrittiMin. n ii 3-22); cf.
M. Lcnchantin De Gubernatis, MAT Lxm (1913), 389 ff., Ennio (Torino,
1915), pp. 62 ff. For detailed criticism ofRostagni's argument see N. Terzaghi,
AATLx (1925), 66off. (=Stud. Graec.et !.At. [Torino, 1963], 686ff.).
3 So T. B. L. Webster, 'Alexandrian Epigrams and the Theatre', in Miscel-
laneadi studiAlessandriniin memoriadi AugustoRostagni(Torino, 1963), S31-43,
HellenisticPoetryand Art (London, 1964), pp. 269 ff., 282 ff., 290 f.
4 So K. Dilthey, De Callimachi Cydippa (Leipzig, 1863), pp. IS f., F.
Skutsch, Aus VergilsFrUhzeit(Leipzig, 1901), pp. 34 ff., S. Mariotti, Lezioni su
Ennio (Pcsaro, 1951), p. 60, 0. Skutsch, The AnnalsofQ!intus Ennius (London,
1953), p. 10; contraE. Rcitzenstcin, in Festschrift
R. Reitzenstein(Leipzig, 1931),
63 ff., R. Pfeiffer, CallimachuSI(Oxford, 1949), p. II, G. Marconi, RCCMm
(1961), 224 ff.
s Tusc. 3 .45. For Cicero's belief (apparently false) that Ennius' Andromacha
was an adaptation of a tragedy by Euripides see Opt. gen. 18.
6
Cf. Dio Chrysostom, Or. 19. The competition for new plays was over by
the late second century {Lucian, Enc. Dem. 27). Ot mpl TOV tu6wo-ov-rexvtTCXt
often included tragic and comic poets as well as actors (sec below, p. 16).
8
 ATHENIAN DRAMA AND THE ROMAN FESTIVALS

between 240 B.C. and the fall of the Roman senatorial regime
seems to have gained more than an ephemeral repute. Ziegler's
view 1 that the Latin playwrights adapted contemporary Greek
work thus lacks all probability.
Besides the six classical Athenian dramatic poets Roman
audiences are reported to have seen in Latin linguistic dress
Aristarchus, a contemporary of Euripides whose name figured
in accounts of the development of tragedy and from whose
hand seventy titles were known to ancient scholars,i Alexis,
Posidippus, Apollodorus and Demophilus.3 Only the last
mentioned is absent from the Greek record but that may be an
accident and there is no good reason to suppose that the origi-
nal of the Asinaria was composed either outside Athens or by a
contemporary of Plautus. Of the comedians Menander was
plainly the one most often performed and of the tragedians
Euripides.
The classical six had certainly acquired most of their later
pre-eminencein Greeceby 240 B.C. A competitionfor actors
held at Athens in 254 employed three old satyr plays, three old
tragedies and three old comedies, one each by Diphilus,
Menander and Philemon. 4 Aristophanes' comedy Bcrrpax01
shows Aeschylus, Sophocles and Euripides standing out in
public esteem as early as the end of the fifth century. From
1
RE 2 VIii (1937), s.v. Tragoedia,1984.
i Souda A 3893 Adler. A third century B.c. papyrus (FlindersPetrie
Papyri Part 11, ed. J.P. Mahaffy [Dublin, 1893), pp. 158-9 [ = nr. 1594
Pack1 ]) contains scraps of what looks like a collection of epigrams ad-
dressed to famous poets; in them and their titles are legible the names
Aristarchus, Astydamas and Cratinus. See R. Rcitzenstein, BPhW XIV (1894),
155-9.
3 At Epist. 2. 1 . 163 ff. Horace is talking generally. He names Thespis
because of his fame as the ,rpC,Tosevpnfisof tragedy; cf. above, p. 7 n. 3, on
the reference to Epicharmus at v. 58.
4 See B. D. Meritt, HesperiaVII (1938), 116 ff., A. Korte, HermesLXXIII
(1938), 123 ff., A. W. Pickard-Cambridge, The DramaticFestivalsof Athens
(Oxford, 1953), pp. 123 f.

9
 INTRODUCTION

the record of theatrical revivals in Athens and elsewhere1 and

the character of allusions to tragedy in fourth and third
century comedyi it is clear that Euripides' popularity far
surpassed that of the other two quite early and continued to
do so.
Menander gained a similar position among comedians, al-
though exactly when is hard to say.3Nevertheless for some time
other tragedians and comedians continued to have their works
revived. There are recorded performances of tragedies by the
fourth-century poet Chaeremon and the otherwise unknown
Archestratus4 and performances of comedies by Anaxandrides,
Posidippus and Philippides.5 The scholars of the Alexandrian
Museum thought Ion and Achaeus worthy to stand beside
Aeschylus, Sophocles and Euripides,6 doubtless in obedience
to a common opinion. The Ptolemaic rubbish tips of Oxy-
rhynchus and other Greek settlements of the Egyptian country-
side provide evidence of a taste in tragic poetry more extensive
1
For the literary and inscriptional evidence see Welcker, Die griech. Trag.
pp. 127s ff., A. Miiller, Lehrbuchder griechi.schen Buhnenalterthumer (Freiburg,
1886), pp. 390 £, N]bb xx:m (1909), 36 ff., Pickard-Cambridge, Dramatic
Festivals,pp. 100 ff., M. Kokolakis, ID.crroovXII (1900), 67 ff. The evidence
from figurative monuments collected by Webster (CQ XLII [1948], IS f.,
HermesL:xxxn [19s4], 29s ff., MonumentsIllustratingTragedyand Satyr Play
[BICSL Suppl. XIV (1962)]) is ambiguous; it has to be interpreted with the aid
of knowledge provided by literature and lapidary inscriptions; it does not pr<>-
vide new knowledge.
a See A. Pertusi, Dioni.soXVI (19s3), 27 ff., XIX (19s6), III ff., I9S ff.
3 On Menander's posthumous f.ame see A. Dain, Maia xv (1963), 278 ff. The
material concerning comedy in late antiquity collected by Webster (AJA
LXVI [1962], 333 ff.) refers almost exclusively to Menander.
4 I.G. v 2. n8 (saec. n a Chr. = S.I.G. 1 1080). Given the context,
Webster's view (HellenisticPoetry and Art, p. 16) that Archestratus was a
contemporary poet seems most unlikely. L. Moretti's doubts (Athenaeum
xxxvm [1900], 272) as to whether I.G. XII 1. 12s (saec. n a Chr.) refen
to the classicalSophocles are likewise unjustified.
s I.G. n• 2323a (Anaxandrides 311), 2323 col. iii (Posidippus 181), col. v
(Philippides 1ss).
6
Anecd. Grace. Par. IV 196.20, Tzctzes, Proleg.Lye. p. 3.8 Scheer.

10
 ATHENIAN DRAMA AND THE ROMAN FESTIVALS

and catholic than that possessed by the schoolmasters of the

urban centres of later antiquity. 1
Where playwrights were concerned the tragic and comic
repertory of the third and second century Roman theatre prob-
ably reflected that of contemporary Greek theatres. But in
actual plays the Romans seem to have had their own taste.
Many scholars have noted the extraordinary preponderance
among surviving tragic titles of those connected with the Tro-
jan cycle of heroic legends.i These legends had long been of
particular interest to the ruling families of Greek states of
recent origin; in these there was a keen desire to find the
same linkswith the heroic past as Argos, Thebes and Athens pos-
sessed.3The wanderings of the Greek and Trojan heroes after
the destructionofTroywere easilyembroidered to suit anystate
in the Mediterranean area. The families of the citiesofEtruria and
Latium, as of other non-Greek speaking communities, early
took an interest in the legends that fascinated their Greek neigh-
bours. Long before the earliestadaptationof an Attic tragedy
speakersof Latin imitated as best they could the sounds of the
Greek heroes' names4 and were accustomed to see representa-
tions of incidents from the legends on works of art.5 There is
1
On the comparatively large number of Euripidean plays absent from the
selection of the later schools which are represented in Ptolemaic papyri see
C. H. Roberts, MusHx (1953), 270. The proportion of tragic papyri which can-
not be attributed to the classicaltrio seems to be much greater among the
Ptolemaic than among the Roman; cf. nos. 169 ff. and 1707 ff. in the second
edition of R. A. Pack's catalogue (Ann Arbor, 1965).
a E.g. Welcker, Die griech. Trag. pp. 1344, 1350, Ribbeclc, Die riimische
Tragiidieim Zdu,Jter tierRepublilt (Leipzig, 1875), p. 632.
3 Cf. T. J. Dunbabin. PBSR XVI (1948), II ff., for the claimsof South Italian
cities to be founded by Greek heroes after the fall of Troy.
4 On the inscriptions on the Praeneste mirron and caskets (C.I.L. 11 547-70)
see R. S. Conway, Tht lt4lic Dialects(Cambridge, 1897), pp. 315 ff. On the
forms of names usedby the early poets see J. Wackemagel, PhilologusLXXXVI
(1931), 143 (,.. Kl. Sehr.I 7SS).
s Sec I. S. Ryberg, An ArchdtologicalRecordof Romefrom the Sevtnlh to the
&cond CenturyB.c. (London, 1940 [Studies and Documents xm]), passim.
II
 INTRODUCTION

good evidence that by the end of the third century many Roman
aristocrats liked to think of themselves as the descendants of the
Trojans led to Latium by Aeneas.1 At least one family, the
Mamilii of Tusculum, was advertising itself by the end of the
second century as sprung from Ulysses.i During the first
century the Iulii claimed descent from Aeneas himself and the
Memmii from Mnestheus.3 These genealogical obsessions must
have been encouraged by the constant performance of the old
adaptations of tragedy at the ludibut it would not be fanciful to
see in them one of the forces directing the choice of the magi-
strates who first had the adaptations made in the third and second
centuries.

II. THE HELLENISING OF THE

ROMAN STAGE
The history of the ludi scaenicidown to 240 is wrapped in
obscurity. The accounts which serious scholars wrote have
come down to us in somewhat garbled form and it is clear that
they were based much more on Greek theories about the
origins of Attic drama4 than on documentary evidence about

• The time of this story's birth and the speed of its growth arc disputed. For
the extreme views on either sidecf. J. Perret, Lts originesde la ligendetroyennede
Rome (Paris, 1~). pp. 451 ff., A. Alfoldi, Early Rome and the Latins (Ann
Arbor, 1963), pp. 278 ff.
a Sec Fcstus, pp. 116. 7 ff., Livy 1.49.9, Dionysius Hal. 4.45. 1, F. Munzer,
RomischeAdelsparteien undAdelsfamilien(Stuttgart, 1920), p. 65 (on numismatic
evidence).
3 Sec Alfoldi, 'The Main Aspects of Political Propaganda on the Coinage of
the Roman Republic', in R. A. G. Canon and C. H. V. Sutherland (edd.),
Essaysin Roman Coinagepresentedto H. Mattingly (Oxford, 1956), pp. 79 f.
4 Sec F. Leo, HermesXXIV (1889), 67 ff. (= Ausg. kl. Sehr. 1283 ff.), XXXIX
(1904), 63 ff., G. L. Hendrickson, A]Ph xv (1894), 1 ff., XIX (1898), 285 ff., F.
Solmscn, T APhA LXXVIII (1947), 252 ff., J. H. Waszink, VigiliaeChristianaen
(1948), 224 ff., K. Meuli, MusH·xn (1955), 206 ff., Brink, EntretiensHardt IX
(1963), 175 ff.
12
 THB HELLENISING OF THE ROMAN STAGE

what went on in Rome before 240. Nevertheless some elements

of these accounts seem to come from intelligent observation of
post-240 stage practices and their general evolutionary ap-
proach is preferable to that of certain modem idealist accounts.
The survival of words of probable Etruscan origin, like
histrio,persona1 and scaena, ~ in the vocabulary of first-century
theatrical practice suggests that some of the early performers
did come from Etruria and may have brought Etruscan theatri-
cal ways with them. The word saturaitself can be plausibly
interpreted as Etruscan.3 The word must have denoted at one
time some sort of stage performance. It can hardly be a mere
invention on the model of Aristotle's TOacnvp1K6v.4All, how-
ever, that Livy's story at 7 .2 .4 f[ implies is that the histriones
presented on a scaenaat public festivals arrangements of words
in a variety of metrical patterns accompanied by pipe music and
called saturae.There is no suggestion that these saturaeinvolved
consistent acts of impersonation.5 Indeed the use of the word as
a book title by Ennius, Pacuvius6 and Varro7 and what we know
of the form of their books 8 carries the very opposite suggestion.
1
Sec W. Dcccke, EtruskischeForschungmund Studim VI {Stuttgart, 1884),
p. 47, F. Skutsch, ALL xv (1908), 145, P. Friedlander, Ciotta n (1910), 164,
v. Blumenthal, RE X1X i (1937), s.v. Persona:1. Dit Thtatermaske,1036 ff.
a Sec W. Schulz.e, Kuhns Zeitschr.u (1923), 242 ( = Kl. Sehr. 638 f.); A.
Ernout discusses objections raised to Schulz.e's view in BSL xxx (1929),
122 n. 2 ( = Philologita1 [Paris, 19,46),p. son. 2).
3 Sec P. Meriggi, Studi EtruschiXI {1937), 157, ICJ7, B. Snell, SIFC N.S. XVD
(1940), 21s f. For criticism see F. Althcim. Geschichteder l4teinischtnSprache
(Frankfurt am M., 1951), pp. 363 ff.
4 The title Saturaattributed to the scenic poets Naevius {Festus,p. 3o6. 29 ),

Pomponius (Gramm.n 200.7, 282. 16) and Atta {Isidore, Orig. 6.9) looks like
a usefulpiece of evidence but does not materially help the argument .
.5 Talk of 'dramatic' saturais confused and misleading.
6
Diomedes, Gramm.1,485.32ff. 7 Q..uintilian,Inst. 10.1.93.
8 The 'JU'#tuor libri saturarummentioned by Porphyrio (Hor. Sat. 1. 10.46)
and quoted by Nonius, Macrobius and the Danieline Servius must be a late
arrangement. There is no getting round the plain words of Diomedcs (Sue-
tonius): olim carmenquad ex uariispoematibusamstabatsatira uocabaturquale
saipsenmt Pacuuiuset Ennius.
13
 INTRODUCTION

The aristocracies of Etruria imported from Greece and had

made by local artists objects decorated with pictorial representa-
tions of the heroic legends as early as the sixth century. 1
Etruscan or Etruscanised names were given to the gods and
heroes. a A writer of the late second or early first century, a
certain Volnius, is said3 to have composed tragedies in the
Etruscan language. Modem scholars have often suggested that
Volnius worked in a tradition that went back beyond 240. The
idea is not an improbable one but unambiguous evidence is
lacking. Excavation of tombs has produced frescos portraying
many aspects of pre-240 scenic performances, but so far of
nothing like an Attic tragedy. The nearest approach to drama is
in pictures of entertainments involving a masked performer. 4
Vase-paintings and reliefs on cinerary urns which have heroic
themes can all be easily interpreted as going back to Greek
originals.5 One can, however, leave open the possibility that
forerunners and contemporaries of the 240 B.c. Latin-speaking
histrionesperformed Etruscan adaptations of the Athenian
classics.
Campanian farce was performed at ludiscaenidin Rome dur-
ing the first century and later both in Latin and in Oscan. 6
Plautus makes an obvious reference to the stock character Bucco
1
Sec J. D. Beazley, EtruscanVase-Painting(Oxford, 1947), p. 8, L. Banti,
Die Welt der Etrusker (Stuttgart, 196<>),pp. n6 f., R. Hampe and E. Simon,
GriechischeSagen in der.frUhenetruskischenKunst (Mainz, 1964).
:a See E. Fiesel. Namen des griechischen Mythos im Etruskischen(Gottingen,
1928). 3 Varro, Ling. S. SS·
4 SeeS. De Marinis, Bnciclopedia dell'ArteAntica VI (1965), s.v. Phersu,n9.
s A series of third-century cinerary urns (E. Brunn and G. Korte, I rilievidelle
urneetrusche1-m[Rome and Berlin, 1870-1916)) are often taken to be based on
a Greek artist's venions of scenes from Euripidean tragedies. A. Piganiol,
Recherchessur lesjeux romains(Strasbourg, 1923), pp. 32 ff., argued that they
represent Etruscan adaptations of plays by Euripides made for performance at
funeral games.
6
See Varro, Ling. 7 .29, Strabo S .233, Cicero, Fam. 7. I. 3, Diomedes,
Gramm.I 489. 32 ff.

14
 THE HELLENISING OF THE ROMAN STAGE

at Baah. 1088 and another to Manducus at Rud. 535. The name

Maccus 1 suggests that Plautus may at one time have been him-
self an actor in this kind of drama .. Livy leaves it unclear when
precisely he thought Campanian farce was first performed at the
ludi scaenidbut it seems likely that the aediles would have pre-
sented this kind of drama before they tried the more sophisti-
cated Atheniankind. There were plainly no links between the
histrionesand the performers of farce in Livy' s own day and it is
therefore hazardous to supppse either that they had more than
accidental relations in the thirqcentury or that the form of
Campanian drama had any considerable influence on the form
which Athenian drama assumed on the Roman stage.
The years between 240 and the middle of the next century
saw considerable reorganisation of the old magisterial festivals
as well as the establishment of new ones. One of the principal
innovations was the regular performance ofLatin versions of the
classics of Attic drama. Many Greek states had reformed their
ancient festivalsin similar fashion or introduced new ones on
the model of the Athenian festivals of Dionysus. i As early as
the fifth century Archelaus of Macedon established a festival in
honour of Zeus and the Muses of which scenic performances
were an important element.3 Philip,4 Alexanders and the suc-
cessor kings 6 often had tragedies and comedies performed at the
monster public festivals they delighted in arranging. Even the
old musical festivals of Pythian Apollo were forced to admit
the Athenian newcomers, tragedy and comedy. 7 To cater for the
1
Asin. n, Mere.10, Gellius, 3. 3; cf. Varro, Ling. 7. 104 (?: Maccius F), Leo,
1 (Berlin, 1912), p. 85.
PlautinischeForschungen
2
Sec W. W. Tam and G. T. Griffith, HellenisticCivilisation' (London,
1952), pp. n3 ff. on Greek. festivals between Alexander's death and 189 B.c.
3 Diodorus 17.16.3. 4 Demosthenes 19.192ff.
S Plutarch, Alex. 4, 29, Athcnaeus 12. 538 F.
6 Diodorus 20.108.1.

7 Plutarch, Mor. 674D, Philostratus, Soph. pp. 238.20 ff., 269. 1 ff., Apoll.
6. 10 (pp. 109.35 ff.).
IS
 INTRODUCTION

demand guilds of itinerant theatrical workers and performers,

ol mpl Tov A16waov nxviTa1, 1 established themselves. Stray
references in literature and a.great nwnber of lapidary inscrip-
tions record their activities from early in the third century in
most parts of the Greek-speaking world. On at least two occa-
sions in the first half of the second century 2 they performed at
Rome.
The Greek cities with which members of the Roman ruling
class had their first direct contacts after the long period of
cultural isolation following the expulsion of the second
Tarquin3 were those of southern Italy and Sicily. Three of the
four known dramatists of the third century came to Rome from
thisarea.Hiero II of Syracuse visited Rome in 237, according to
Eutropius4 ad ludos spectandos.Permanent theatres had been
erected in the wealthier cities as early as the fifth century .5
Athenian drama was well known. Aeschylus visited the court of
Hiero I and produced the Titpaa1 and the Al-rvaia1in Syracuse.6
Those Athenian prisoners taken in 413 who could recite
Euripides were, according to legend, 7 released by their admiring
captors. The tyrant Dionysius not only invited Athenian poets
1
See Welder, Die griech. Trag. pp. 1303 ff., 0. Liiders, Die Dionysischen
Kunstler(Berlin, 1873), P. Foucart, De collegiisscenicorum
artificumapudGraecos
(Paris, 1873), Poland, RE 2 vii (1934), s.v. Technitai (Nachtrage), 2473 ff.,
Pickard-Cambridge, DramaticFestivals,pp. 286 ff.
:a At the ludi celebrated by M. Fulvius Nobilior in 186 (Livy 39.22.2) and
those celebrated by L. Scipio in the same year (Livy 39.22. 10). Cf. Polybiw
30.22 for the ludi of L. Aniciw in 167. There is no need to suppose that they
performed in tragedy or comedy.
3 See Altheim, Epochender romischenGeschichte 1 (Frankfurt am M., 1934),
pp. 128 ff., A. Blakeway, JRS xxv (1935), 136, Ryberg, An Archaeological
Recordof Rome, pp. 3, 79, 107, 204 f.
4 3. 1-2.
5 On the theatre of Syracuse, which seems to have been dedicated about
•6o B.c., sec G. E. Rizzo, n TeatroGrecodi Siracusa(Milan, Rome, 1923). On
the other theatres of Magna Graecia sec B. Pace, Dionisox (1947), 266 ff., A.
von Gcrkan, in FestschriftA. Rumpf (Cologne, 1965), pp. 82 ff.
6
Vita, 9, 18. 7 Plutarch, Nie. 29.

16
 THE HELLENISING OF THE ROMAN STAGE

to his court but composed tragedies himself for performance at

the Athenian festivals.1 An early South Italian kalyx krater
(400-390) depicting the punishment of a thief in a stage comedy
has an inscription in what is possibly Attic dialect and iambic
verse. :i It is probably an accident that activity by ot mpl TOV
~16waov TE)(ViTCX1 is not recorded until the mid first century.3
Nevertheless the Dorian cities had theatrical traditions of their
own, independent at least in origin of the Athenian theatre, 4
and it is possible that the texts of Athenian plays were used by
actors costumed differently from Athenian actors and on stages
different in character from the Athenian.5 In any case one may
reasonably seek the inspiration of at least some of the changes
made to the Roman ludiscaenicibetween 240 and the turn of
the century in the stage practices of Magna Graecia.
However one imagines the amalgam of Etruscan, Italiote
Greek and perhaps Oscan elements in the practice of the late
third century Roman histrionesit is plain that the adapters of
Athenian tragic and comic scripts had to deal with a theatrical
situation very different in the one case from that which faced

' Sec Niese, RE v (1905), s.v. Dionysios [1), 900 f., for references.
3
See Beazley, AJA LVI (1952), 193, Webster, CQ xm (1948), 25, in
Fe.stschriftB. Schweitzer (Stuttgart, 1954), pp. 26o f.
3 Plutarch, Brut. 21 (Naples). J.C. XIV 12, 13 {Syracuse), 615 {Rhegium) are
widated.
• On the so-called Cf>).VCO<ES see Pollux 9. 149, Athenaeus 14. 621 P ( = Sosi-
bius)'.
5 It is only a possibility. The evidence of painted pottery will not bear the
weight of the interpretations often placed upon it. Late fifth and fourth century
Apulian, Lucanian and Campanian wares have representations of stage per-
formances of both comic (see A. D. Trendall, Phlyax Vases [BICSL Suppl.
VIII (1959)), Webster, MonumentsfllustratingOld and Middle Comedy [BICSL
SuppL IX (1900))) and tragic (see Webster, MonumentsIllustratingTragedyand
Satyr Play)type. The costume of the comic acton is clearly local. In the absence
of fully comparable Athenian material it is difficult to pass judgment on the
elaborate costume of the tragic acton. The low stage raised on columns with
steps leading up to it seems quite different from what we know of the contem-
porary Athenian stage.

2 17 J TO
 INTRODUCTION

Euripides in the fifth century and in the other from that which
faced Menander in the late fourth; different again from that of
contemporary Athens and other Greek-speaking states.
Roman stages in stone at the time of Vitruvius differed con-
siderably in dimensions from contemporary Greek. 1 The stages
in wood of the third and second centuries no doubt differed
similarly from contemporary Greek stone structures. Archaeo-
logical investigation and close study of the surviving dramatic
texts have made it clear that the classicalAthenian stage differed
yet again in dimensions, and quite radically. 2
The flat area directly in front of the stage where the Athenian
choruses sang and danced was occupied by seats for part of the
audience.3 Adaptations of comedy clearly dispensed with the
dancing chorus. Ancient students tell us so4 and surviving comic
scripts either do not have the actor's introduction of the first
choral ode conventional in late fourth century Attic comedy5 or
replace it with something else. Plautus writes at Bacch. 107:
simul huic nesdoquiturbarequi hue it decedamus, 6 while Terence

leaves an awkward break at Haut. 170.7 The performance of the

tibicenindicated at Pseud.573a and the lecture of the choragusat
Cure.462-86 probably replaced choral odes in Plautus' originals.
The difficulty which ancient students, who must have possessed
the Greek scripts, found in imposing a five-act division onthe

I 5 •6 •2, 5 •1•
2
See Pickard-Cambridge, The Theatre of Dionysus in Athens (Oxford,
1946),pp. 5 ff.,Webster, Greek TheatreProduction{London, 1956), pp. 22, 170.
3 Vitruvius should not be taken to imply at 5 . 6. 2 that before 194 this area
was left free of spectators; see Livy 34. 44. 5.
4 Cf. Donatus, Ter. Eun. praef. 1. 5, Euanthius, De Jab. 3. 1, Diomedes,

Gramm.1491 . 29 f. The statement at Gloss.lAt. 1 128. 6-apud Roma,wsquoque

Plautuscomoediae chorosexemploGraecorum inseruit-may refer to the fishermen
of Rud. 290-324 and the aduocatiof Poen. 504-816.
s Cf. Antiphanes, fr. 91, Alexis, fr. 107, Menander, Epitr. 33-5, Perik. 71-2,
Dysk. 230-2.
6
Cf. Leo, HermesXLVI (19n), 292 ff.
7 Cf. F. Skutsch, HermesXLVII (1912), 141 ff.

18
 THE HELLENISING OF THE ROMAN STAGE

Latin versions suggests that the Latin poetsoften covered up the

breaks by rewriting the actors' dialogue. Classical tragedy dif-
fered from comedy in that the chorus sometimes had a role
integral to the action and it was almost impossible for Latin
adapters to remove it in these cases. Plautus appears to refer to a
Roman tragic chorus at Amph. 91-2. Ancient students inter-
preted parts of Latin tragic scripts as utterances of a chorus 1 and
the remains of Ennius' adaptation of Euripides' M,'}6e1acontain
unmistakeable signs~ that the original chorus of Corinthian
women was retained. Horace' s discussion of the chorus at Ars
193 ff. may be based on a Greek theoretical discussion and may
at times adduce arguments quite inapplicable to Roman
practice,3 but it is hard to believe that the despised republican
playwrights abandoned the chorus and that Horace passed over
in silence such a divagation from the Attic practice he so much
admired. 4 The Roman choruses could not have danced in the
classical Greek manner.5 The extent to which they were present
1
Cf. Varro, Ling. 6.94, Gellius 19. 10. 12 (but see commentary on fr. xcrx),
Terentianus Maurw 1934.
a See Commentaryon frs. cv and ex.
3 Cf. Latte, HermesLX (1925), 5 ff., Brink, Horaceon Poetry: Prolegomena
(Cambridge, 1963), pp. n4 f.
4 H. Plank, Q. Ennii Medeacommentario perpetuoillustrata(Diss. Gottingen,
1807), pp. 56 ff., and A. La Penna, Maiav (1952), 95, seem to banish the chorus
from the early republican stage. Most students allow it in some form; cf. A. G.
Lange, VindiciaeTragoediae Romanae(Leipzig, 1822), p. 22 n. 31, Welcker, Die
griech.Trag.p. 1368, Grysar, SB Vienna 1855, 365 ff., 384 ff., 0.Jahn, Hmnesu
(1867), 227 ff., Ribbeck, Die rom. Trag.pp. 632 ff., L. Mueller, Q. Ennius. Eine
Einleitungin das Studiumder romischenPoesie(St Petersburg, 1884), p. 79, Leo,
De TragoediaRomana ObseruationesCriticae (Gottingen, 1910), pp. 14 ff.
(= Ausg. kl. Sehr. I 203 ff.), Pl. Forsch.•,p. 96, GeschichtederromischenLiteratur
1 (Berlin, 1913), pp. 193 ff., E. S. Duckett, Studiesin Ennius (Diss. Bryn Mawr,
1915), pp. 53 ff., Fraenkel, Pl. im Pl. p. 336 (= Elementi,p. 320), Wl. Strzelecki,
in TragicaI (Wroclaw, 1952), pp. 54 ff.
5 Post-fifth-century Attic tragedians reduced considerably the role of the
chorus. There is considerable evidence that later productions of classical plays
sometimes dispensedwith a full chorus; cf. E. Capps, AJA x (1895), 287, Leo,
RhM LII (1897), 509 ff., A. Korte, N]bb v (1900), 81 ff., P. Venini, DionisoXVI

19
 INTRODUCTION

during the actions of tragedies and their mode of giving utter-

ance must remain obscure. 1
The left-hand end (from the point of view of the spectators)
of the Roman stage pretty clearly was the conventional point of
entry for travellers from abroad. 2 There is some evidence that
this reversed the Attic convention.3 .
At Athens tragic acting and comic acting were considered
quite different skills.4 At Rome on the other hand it was not
unusual for an actor to perform in both genres.5The scholars of
late antiquity thought6 that the adapters of comedy had more
than the classical Athenian trio of actors to employ.7 If masks
were not in use before the end of the second century 8 there had
to be as many actors as characters. The early adapters of comedy
seem to have taken advantage of a greater number of available
actors in order to alter whole scenes of their originals. The
Stichus is the only one of Plautus' pieces that could be easily

(1953), 3, Webster, HermesLXXXII (1954), 294 ff. A third century s.c. papyrus
of Euripides' 1 lmr6AvroSomitting the choral lyrics basnow turned up (Pap.
Sorb. 2252). W. S. Barrett, Euripides:Hippolytos(Oxford, 1964), p. 438 n. 2
compares Pap. Hib. 1 4, long thought to be of Euripides' Olvevs.
1
Leo's theory that Ennius always made the coryphaeus utter trochaic tetra-
meters and kept the other choreutae mute goes far beyond the evidence. At
least one Ennian choral utterance (fr. ex) is more plausibly scanned as in lyric
verses and Leo himself admitted lyric scansion of choral utterances in the other
tragedians.
:a Cf. Plautus, Amph. 333, Men. SSS, Rud. 156, Terence, Andr. 734, M.
Johnston, Exits and Entrancesin Roman Comedy(New York, 1933), pp. 64 tf.
3 Pollux, 4. 126. See W. Beare, CQ xxxn (1938), 205 ff. ( = Roman Stage',
pp. 248 ff.), Pickard-Cambridge, Theatreof Dionysus,pp. 234 tf.
4 C£ Plato, Politeia3. 39S A.
s Cf. Plautus, Poen. 1 ff., Cicero, Orat. 109.
6
C£ Diomedes, Gramm.1490.27 ff.
7 The existence of the convention of three actors at the time when classical
tragedy and classicalcomedy were actually written basbeen denied; for the
evidence and discussion see Pickard-Cambridge, DramaticFestivals,pp. 137 ff.,
K. Schneider, RE Suppl. vm (1956), s.v. 'YnoKptfllS, 190 ff. Menander's
~VO'KOAoS can be arranged for three actors; c£ G. P. Goold,Phoenix :xm (1959),
144 £, J. G. Griffith, CQ N.S. x (1900), 113 ff.
8 See below, p. 22.

20
 THB HELLENISING OF THE ROMAN STAGE

performed with three principals. 1 Terence's pieces, on the other

hand, are perhaps all just actable with three. Nevertheless the
didascaliae of the Adelphiand the Hecyramention each two prin-
cipals, something unparalleled in Greek didascaliae.i Terence
was much more faithful than hispredecessors to the texts of his
originals but the producers of hisplays may have still employed
the traditional Roman number of actors.3 Horace' s polemic at
Ars 192 suggests that contemporary producers of tragedy at
Rome still employed more than three actors.
The amount of singing required in the actors' parts in Attic
drama was very much less than in the Roman versions4 but
there is no evidence that the Athenian actors did not do their
own singing. According to Livy 7. 2. 9-10 the Roman histriones
merely mimed the canticawhile a singer accompanied the
tibicen.It is difficult to believe that tragedy and comedy were
performed this way. The trochaic and anapaestic verses of
Ennius' Andromachaquoted by Cicero at Sest. 120-2 were cer-
tainly uttered by the actor Aesopus. At De orat.1 .254 and Leg.
1 . 11 Cicero talks of the actor Roscius singing in old age. Livy
may have been thinking of the contemporary pantomime. 5
There is very little evidence as to how the histrioneswere
costumed in the third century and early second. One piece6
suggests that tragic actors wore the dress that normally distin-
guished the priestly order of jlamines. The accounts of comic
costume by the Greek Pollux (4. 118-20) and the Roman
Donatus (De com. 8 . 6) do not tally and may reflect real dif-

I C£ Leo, NGG, Phil.-hist. Kl. 1902, 391.

a C£ thatof the b.WKOMJS.
3 K. Dziatzko, R.hMxx (1865), 591, and F. Schoell,N]bb CXIX (1879), 41 ff.,
discussthe Terentian didascaliaewithout reaching any satisfactory conclwion.
4 See below, p. 29. Euripides, Or. 1369 ff. would have needed a highly

skilled singer.
s C£ Ribbeck, Die ,om.T,ag. p. 634.
6 Gloss. Lat. I 128, 568, Scrviw auct. Aen. 4.262; see I. Hilberg, WSt xm
(1891), 170 £
21
 INTRODUCTION

ferences between Greek and Roman stage practice. Romans of

the first century believed that the wearing of masks by the
histrionesin performances of tragedy and comedy was a recent
innovation. 1
At Athens the writers of tragedy and comedy were usually
native Athenians, sometimes of aristocratic birth. During the
fourth century a number of foreigners came to the city to
practise the two arts. There was a tradition of poetry held in
honour by all.The prime object of the poet was to win the prize
in the competition to which he submitted hiswork. Performers
likewise were usually citizens of respectable status and com-
peted for prizes from the year 449 onwards. 2 At Rome writing
verse became in the course of time a respectable activity for
gentlemen, acting never. In the third and early second centuries
both the making of acting scripts and the performance of them
were lowly activities.3 A magistrate would use either his own
slaves, freedmen and clients for the spectacles he was providing
at the ludi or hire professional poets and actors. 4 There is no
clear evidence for poetic competitions during republican times.5
1
Sec Cicero, De oral.3. 221, Fcstus, s.v. persona/a,p. 238. 12 ff., Donatus, De
com. 6.3, Diomedes, Gramm. 1489.11 ff., C. Saunders, A]Ph XXXD(1911),
58 ff. The sceptical attack on the tradition by A. S. F. Gow, JRS D (1912),
65 ff., and Beare, CQ XXXDI (1939), 139 ff. (c£ Roman Stage', pp. 192 ff.,
303 ff.) is unconvincing. Donatus' belief (Ad. praef. 1 .6, Eun. praef. 1 .6) that
Terence's actors wore masks was probably based on the illwtrations of the text
current in hisday; see Leo, RhMxxxvm (1883), 342 f. The prologue ofPlautus'
Poenuluswas spoken by one of the actors who changed his costume in some
way (123, 126) in order to play one of the ordinary roles. This shows nothing
about the distribution of roles within the action.
a On Athenian actors see Pickard-Cambridge, DramaticFestivals,pp. 127 ff.
3 See B. Warnecke, N]bb XXXDI (1914), 95 ff., RE vm ii (1913), s.v. Histrio,
2117, 2125 ff. T. Frank's challenge to the orthodox view (CPh XXVI [1931],
1 1 ff.) is unconvincing.
4 For the giving of a pretium to poets cf. Terence, Eun. 20, Horace, Epist.

2. 1 . 175, Gelliw 3. 3, Suetoniw, Vit. Ter. 3. 14.

s Plautus, Cas. 17 and Terence, Phorm. 16-17 are clearly metaphorical; cf.
Volcaciw Sedigitus ap. Gell. 15 .24, Horace, Epist. 2. I. 181.

22
 THE HELLENISING OF THE ROMAN STAGE

Comic poets often had to plead for a quiet hearing. They

ended their scripts with an appeal for applause, as in their
originals, 1 but made no request for victory. 2 Tragic scripts seem
to have ended similarly.3 There were however acting competi-
tions. 4 The language of Plautus, Trin. 7o6 suggests that they
went back beyond the time of composition of this script.5
As time passed the differences between the external condi-
tions of the Roman stage and those of the contemporary Athe-
nian lessened but during the period when the first adaptations
of classicalAthenian dramawere made they remained consider-
able and inevitably affected the poetic form of the adaptations.

III. A THE NIAN DRAMA AND THE

ROMAN POETS
The only evidence for what the Latin poets themselves thought
about the processof adaptingthe Athenianclassicsis foundin
five comic prologues from the years 166-161, those of Terence's
Andria, Hauton timoroumenus,Eunuchus,Phormioand Adelphi.
Terence can be seen struggling against two contradictory but
rdated movements of taste, one favouring the work of the
older poets against the new, the other demanding faithful
versions of the Athenian classics.He appeals to the commonly
1
C£ Posidippus, Pap. Heid. 183; Mcnandcr, Dysk. 965 ff., Epitr. fr. 11
Korte, MisoumenosD ➔ col. ii. 36-8 Turner, Sikyoniosin Pap. Sorb. 2272 e col.
=
B. II ff. ( -4,20ff. Kassel).
3
There was sometimes a prayer for the success of Roman arms in the
prologue; c£ Plautus, A.sin. 15.
3 Ct. Horace, Ars I SS, Q..uintilian,Inst. 6. 1. 52; contrast Euripides, I. T.
1497 ff., Or. 1691 ff., Phoin. 1764 ff.
4 C£ Cicero, Alt. 4. 15. 6. It is noteworthy that the ivy of Dionysµs {sec

Olck, RE v {1905), s.v. Epheu, 2838) was replaced by Apollo's pahn (sec
Steier, RE xx i {1941), s.v. Phoinix, 401).
5 Plautus, Amph. 69 and Poen. 31 may be acton' interpolations from a later
time.
23
 INTRODUCTION

recognised authority of Naevius, Plautus and Ennius in defence

of the liberties he has taken with his originals.1 The argument
proceeds within the context of comedy but we need not suppose
that Naevius and Ennius regarded the adaptation of tragedy as
fundamentally different. The Athenians always kept the trage-
dian' s craft rigidly distinct from the comedian' s2 and required
much less originality from the tragedian in the matter of plot
construction.3 At Rome on the other hand Livius and his im-
mediate successors cultivated the two genres together and
brought their respective forms much closer than they had been
at Athens.4
Terence's thinkingmade the conventionally sharp ancient
distinction between the subject-matter and plot of a play (argu-
mentum) and the verbal manner in which the poet presented
those things (oratio,stilus).5The distinction between one argu-
mentumand another was at the same time much less sharp for
him than it was for certain contemporary critics and is for the
modem scholar. Speaking of Menander' s TJeptv&fa and •Av6pfa
he declared: qui utramuisrectenorit ambasnouerit.6 The scanty
remains of the TJeptv&fa make it appear to us to have had a very
different argumentumfrom the •Av6pfa but a poet busy provid-
ing scripts for set occasions would naturally have been more
struck by their similarities. Terence's predecessors appear to
have been much busier than he was and we can well believe
that they regarded many plays as very like others and did not
1
Amir. 18-19; cf. Haut. 2er1.
3
Cf. Plato, Politeia3. 395A. Ion 534 c and Symp. 223 D deal with an ideal
situation.
3 Cf. Antiphanes, fr. 191. On the views of Aristotle (Poet.9.1451b21--6,
14.1453b25) see Brink, Horace,pp. 103 ff.
◄ Sec below, pp. 31ff, 38ff.
S Amir. 9-12, Haut. 6, 46. Cf. Aristotle, Poet. 6. 1449 b24 ff., Horace, Ars
40 ff., 119 ff., Plutarch, Mor. 347B-F (anecdote about Menander's mode of
composition).
6 Amir. 10. Cf. also Eun. 41 nullumstiam dictumquodnon dictumsit prius.

24
 ATHENIAN DRAMA AND THE ROMAN POETS

think close fidelity to the particular argumentum

of a play much
of a virtue.
In discussingthe school exercise of ,rapacppaa1s1 and his own
adaptations of Greek philosophical dialoguesi Cicero, who was
well acquainted with the texts of many Greek and Latin plays
and knew at first hand the workings of the Roman theatre,
thought it proper and persuasive to use the analogy of the
behaviour of the old republican poets towards the texts of their
originals.
The schools of third and second century Greece employed
two exercises akin to the translation of poetical works from one
language into another. In the one poetic texts were interpreted
word for word in the everyday language.3 In the other a rheto-
rical equivalent of the substance of a poem or prose work was
sought; the student had to present this substance in such a way
as to please the sort of audiences he was being trained to per-
suade; it was important that he should use the vocabulary and
mode of expression appropriate to public oratory.4 Rom.an
schools took over both exercises, at first using Greek texts and
then, as the indigenous literature grew in bulk, Latin ones. The
continued use of Greek texts naturally came under attack and
defenders of the practice tossed up the notion that the para-
phrast should attempt to improve on his original.5 This notion
ceases to be surprising when one remembers the sharp ancient
distinction between a work of art's substance and its verbal or
material form; the one was constant and to an extent beyond
1
Opt. gen. 18. :a Ac. 1. 10, Fin. 1.4-7.
3 Cf. the paraphrases of the Iliad fowtd on papyrus (nos. 1157 ff. in Pack's
catalogue) and the literary references to the exercise collected by H.-1.
Marrou, Histoirt de l'lducationdans l'antiquiti' (Paris, 1958), pp. 231 ff., G.
Giangrande, EranosLX (1962), 152 ff.
• Cf. Plutarch, Dem. 8.2, Dio Chrys. Or. 18.19, Theon, Progymn. 4,
Hermogenes, Meth. 24.
s Cf. Q!!intilian, Inst. 10.5.5, Pliny, Epist. 7.9.3.

25
 INTRODUCTION

time and space, the other was very much fixed in its creator's
particular time and could not be exactly reproduced in another's;
adaptation alone was possible. The old dramatic adaptations of
Sophocles, Euripides and Menander were much admired in
some quarters in the mid first century and some were thought to
excel their originals. 1 Accordingly Cicero argues in his treatise
De optimogenereoratorumthat the paraphrases he and others
made of the Attic orators were of the same character as the
dramatic adaptations and deserved the approbation accorded
these.
At Ac. I. 1-10 Cicero is concerned to rebut the view that
readers who knew Greek would not bother with philosophical
dialogues written in Latin. Such men, he argues, read with
pleasure the Latin adaptations of Aeschylus, Sophocles and
Euripides made by the poets of the previous century, qui non
uerbaseduim Graecorumexpresserunt poetarum;they will get even
more pleasure out of the philosophical dialogues of Roman
writers si, ut illi AeschylumSophoclemEuripidem,sic hi Platonem
imitenturAristotelemTheophrastum.His distinction between uis
and uerbais the one informing Opt. gen. 14-23.i The poets, in
his view, did not offer a word for word representation of the
substance of their originals but achieved in their own medium
effects similar to those which the Athenian poets had achieved
in theirs. Both sets of poets had their own excellences.
At Fin. 1 .4-7Cicero makeshis case against those who scorned
Latin philosophical dialogues more detailed and sophisticated.
Such men, he argues, read fabellasLatinas ad uerbume Graeds
expressasand must therefore be foolish to ignore works of a
more serious kind which do not merely reproduce the words of
someone like Aristotle in the manner of a grammatical exercise
1
Cf. Cicero, Tusc. 2.49, Varro ap. Suet. Vit. Ter. 3.
:a Cf. also the hexameters of Caesar and Cicero on Terence (Suetonius, Vit.
Ter. 7).

26
 ATHENIAN DRAMA AND THE ROMAN POETS

but treat Aristotelian themes afresh in the manner of a Theo-

phrastus. Where his own dialogues depend closely on a classical
Greek philosopher they depend, he suggests, in the way that
Ennius' epic depends on Homer and Afranius' fabulaetogatae on
Menander, not in the way Ennius' Medeadepends on Euripides.
1

The character of his argument forces him to attribute less

independence to the dramatic poets and more to himself than
he does at Ac. 1. 1-10. 3 Nevertheless he does not assimilate the
republican plays completely with the grammatical kind of
translation. He writes of them as possessing a literary value of
their own independent of their degree of fidelity to the original
Greek. He dismissesAtilius as a bad writer, not as an inaccurate
translator.
Ancient scholars regarded the tragedies and comedies of the
republican period as different in kind from their originals. The
argumentabelonged to Greece, only the words to the Latin
poets.3 It is easy to misunderstand their view. It was not one
denigratory of the Latin poets. There would have been a strong
temptation for Romans to consider the construction of a plot a
smaller matter than the provision of adequate words. Such
detailed comparisons of the texts of the Latin plays and their
Greek originals as were made4 resembled those made between
Latin and Greek poems united only in genre, for example
1
Cf. Afranius, Com. tog. 2s-8, Macrobius, Sat. 6. 1. 1-7.
2
It was not only in his public orations that Cicero tailored his statements of
fact according to circumstance and audience; one might compare with Ac.
1.1-10 and Fin. 1.4-7 his description of the libri academiciat Att. 12.s2.3:
apograplu,sunt, minore laborefiunt (on the context of this remark sec D. R.
ShackletonBailey, Towardsa Text of Cicero'Ad Atticum' [Cambridge, 1900),
pp. 61 f.).
3 Cf. Donatus, Tcr. Andr. 9 scribilenim Terentiusqui uerbaadhibettantum;
facit Menanderqui etiamargumentumcomponit.Leo, Pl. Forsch.•,p. 87 n. 1, com-
pares Marcus Pomponius Bassulus, C.L.E. 97, Menandri paucas uorti scitas
Jabuw... ipsus etiam sedulofinxi nouasand Pliny, Epist. 6.21.
4 Cf. Gcllius 2.23, 11.4, the scholia to Terence's comedies passim.
 INTRODUCTION

between the Aeneid and Homer's epics.1 Argument about the

text of a Latin play and its correct interpretation based itself on
such comparisons 1 with a rarity that is significant. Most discus-
sion, including that of the Latin poet's alleged misunderstand-
ings of his original, was carried on in terms deriving from the
discussion of rhetorical irapacppams.3Students were interested in
how effectively the Latin poet used his medium in dealing with
the Greek substance. We thus often find the same language used
to describe his degree of successas is used of Virgil's treatment
of a Homeric moti£
Modem views of the relationship between the comedies and
tragedies of the early republic and the Attic plays that lie behind
them have often been subconsciously coloured by the attitude
to the translation of the Hebrew and Greek scriptures imposed
until recently by the Christian sects upon their members; the
words of the original as well as its substance were in orthodox
theory sacred and should either be left alone or respected by the
translator; the idioms and rhetorical modes of the translator's
own culture ought not to replace those of the original. 4 The
practice of successful translators of these writings has of course
always diverged to a greater or lesser extent from the orthodox
theory but the word 'translation' has come to have associations
that can only be misleading in the discussion of pagan Latin
literature. Accordingly I make little use of it.
1
Cf. Sidonius' comparison of the Htcyra of Terence with Mcnander's
'Em-rphrOV'T'ES (Epist.4. 12).
a Cf. Donatus, Amir. 204, 483, 592, 801, Eun. 46, Htc. 58.
3 The ovy,cp101s of two authors writing within the same genre wasa regular
exercise of the Greek schools; sec F. Fock, Hmnts tvm (1923), 327 ff.
4 Cf. Hieronymus, Epist. 51.
 INTRODUCTION

IV. THE FORM OF ROMAN TRAGEDY

Thirty-two tragic scripts have been transmitted to us from the
Athenian theatre more or less in their entirety. Of these we
possess small sections of the Latin adaptations of possibly
twelve. All twelve seem to have contained a much lower pro-
portion of that type of verse which was uttered by the actor
without any accompaniment from the piper 1 than did their
originals.i The remains of third and second century Latin
tragedy taken together show a tendency for poets to move
1
My argument does not require an answer to the question how Greek
and Roman acton actually delivered verses while the piper was playing. At
Athens stichic tetrameters were recited in a mode (napcxiccrr~oyri) different
from that used for trimeters on the one hand and from that used for verses of
lyric type on the other (for the sparse evidence see Pickard-Cambridge,
DramaticFestivals,pp. 153 ff.). Ancient editors of Latin comic scripts deployed a
set of symbols-DV (deuerbium),C (canticum)and MMC (mutatismodiscanti-
cum:see Donatus, Ad. praef. 7) which suggests that Roman actors used a similar
triple mode of delivery. I assume that both at Athens and at Rome iambic tri-
meters were normally spoken without accompaniment (this is implied in such
comparisons with ordinary speech as at Aristotle, Poet.4. 1449a21-8, 22. 1459a
10-14. Rhet. 3.8.1408 b 32-6, Cicero, Orat. 189-91, Horace, Ars 80-.2) while all
other types of verse received an accompaniment of some kind (cf. Aristo-
phanes, Om. 222, Menander, Dysk. 879, Xenophon, Symp.6. 3, Plautus, Pseud.
573 a, Stich. 762,769, Cicero, De orat.1.254, Orat. 183-4, Ac. 2.20, Diu. 2.113,
Tusc. 1 . 1o6). There is evidence for occasional Greek (Plutarch. Mor. 1140 f.,
Athenaeus,14.636B, Lucian. Dt Salt. 27, Pap. Osl. 1413 [see SO XXXI (1955),
1 ff.]) and Rom.an (see Commmtaryon fr. crv; the irregularities in the distribu-
tion of the symbols C and DV in the Palatine manuscripts of Plautus are rela-
tively very few and should be treated as erron of transmission [so Ritschl.
RhM XXVI (1871), 616 (= Opusc. m 22 ff.), Bergk, PhilologusXXXI (1872),
229 ff. (= Kl. phil. Sehr. 1192 ff.)] rather than as evidence for republican the-
atrical practice [so R. Klotz, Gnmdzuge altromischerMetrik {Leipzig, 1890),
pp. 379 ff., A. Klotz, Wurzb.Jbb. n (1947), 305 f.]) departures from the norm
but these are too few to affect my argument significantly. I have not found it
useful always to distinguish between musically accompanied stichic verses and
'lyrics'.
, For the replacement of Greek trimeters by other types of verse in par-
ticular casessee Commentaryon frs. LXXXIV, cv, cvn, CVDI. CIX.

29
 INTRODUCTION

closer to Attic practice as time passed.1 Only one script of the

kind of Attic comedy adapted at Rome survives in its entirety 2
but the lengthy pieces of Menander's 1Th6K1ovand Caecilius'
adaptation quoted by Aulus Gellius (2. 23) and the pieces of
Terence's originals quoted in the scholia confirm the accuracy
of Diomedes' statement at Gramm. I 490. 22-3: in Latinis. ..
fabulis plura sunt canticaquaecanuntur.3The difference however
grew less as the second century advanced. 4
If, as seems likely,5 the Latin poets not only dispensed with the
choral songs of Attic comedy but modified the actors' parts at
those points where, in their originals, the choral songs had been
performed, the resultant continuous action would have been
something very different from what Athenian audiences of the
time of Menander were accustomed to see. The action of the
average Attic tragedy at the time of Euripides could be said to
have possessed a continuous character inasmuch as the interests
of the members of the chorus were involved to a greater or
1
Trimeten form about 40 per cent ofLivius' measurable venes, 35 per cent
ofNaevius', 30 per cent of Ennius', 45 per cent of Pacuvius', SS per cent of
Accius'. Frequent uncertainty about the scansion of small fragments reduces
the value of these figures. The difficulty of isolating choral fragments and the
general uncertainty about the extent of the Roman tragic chorus' role make
comparisons with Greek percentages thoroughly ambiguous. I note for what it
is worth the fact that trimeten form about 65 per cent of the verses of the
'Opia-rris.
a On the relationship between the vene quoted by Festus, p. 174.18 as by
Plautusin Dyscoloand Menander's Avc:nco).os see Strzelecki, GIF :m (1959),
305 ff., C. Q!!csta, RCCM1 (1959), 307 ff., T. Mantero, in Menandrea(Univ.
di Genova, 1st. di Fil. Class., 1900), 125 ff.
3 Cf. Qyintilian, Inst. 10. 1.99 lied Terenti scriptaad ScipionemAfriamum
rejerantur(quaetamensunt in hoegenereelegantissima et plus adhuchabituragratiaesi
intrauersustrimetrossteti.ssent),
Marius Victorinus, Gramm.VI78 . 20 sdoplurimos
affimuzreTerentianasuel maximefabulasmetrumac disciplinamGraecarumcomoedi-
arumnon custodi.sse, id est quasMenanderPhilemonDiphiluset ceteriediderunt.
4 In c. 18,400 Plautine verses there are c. 8,200 trimeters (38 per cent), in c.
6,000 Terentian 3,100 (52 per cent). Menander's AVCTKo).os has 813 trimeters in
969actors' venes (8-4per cent).
s See above, pp. I 8 f.

30
 THE FORM OF ROMAN TRAGEDY

lesser extent in it. Nevertheless the songs of the chorus very often
had little relevance to the action and were always performed
away from the area used by the actors. The Latin poets did not
abolish the tragic chorus but what they could do with it was
limited by the fact that their choral performers had to use the
same comparatively restricted area as their actors; the elaborate
dancingof fifth-century Athens was impossible and without this
the songs of the Attic poets would have had little theatrical
value. It is possible, therefore, that they dispensed with some of
the songs of their originals and pruned others severely. In any
case the action of a Roman tragedy as it revealed itself in the
theatre would have had much more continuity than that of any
fifth-century Athenian play.
The dramatic material which the Latin poets put into
musically accompanied verse differs not only in quantity but
also in kind from what can be found in the scripts of classical
Attic tragedy and comedy. The Attic comedians rarely went
beyond the simpler stichically arranged types of verse1 and did
not allow the action to progress in scenes accompanied by
music. The tragedians often gave their actors metrically com-
plex utterances and made them take part in lyric dialogues both
with the chorus and with one another. It would be difficult
however to point to a long musically accompanied scene in
which vital information was given and the action made serious
progress.i
It is plain that the first Latin poets set the scripts of comedy
and tragedy they chose to adapt in one dramatic mould and that
during the time of the republic this mould was never broken
1
F. Marx. PlautusRudens: Text und Kommentar(Abh. Siichs.Ak., Phil.-hist.
Kl. xxxvm (1918), v), pp. 254, ff., collects the little that can be found oflyric
verse. There is somewhat more in the remains of the so-called' Middle Comedy'
(see Leo, R.hM XL (1885], 164,).
i Cf. Wilamowitz, Mtnandtr. Das Schiedsgericht (Berlin, 1925), p. 169 n. 1,
G. Pasquali, EndclopediaItalianaXXVII(1935), s.v. Plauto,528.

31
 INTRODUCTION

despite an ever increasing desire on the part of critics for the

main features of the two Attic genres to be faithfully repro-
duced. Two historically credible kinds of explanation have
been offered of their behaviour. According to one1 they
adapted the Attic scripts to a contemporary type of musical
stage performance. According to the other 2 they simply
merged the theatrical forms of Attic tragedy and Attic
comedy. The latter explanation does not cover all the facts,
the former has to postulate an entity for whose existence
there is little solid evidence. The issue is an important one and
cannot be brushed aside with idle talk about Roman creative-
ness. The paucity of our knowledge of what went on during
the third century in the theatres of Etruria and the Greek and
Oscan speaking cities of the South makes it incapable of
settlement.
The scholars who first edited and discussed the early Latin
theatrical scripts found it possible to use the same general type
of metrical description as had been applied at Alexandria and
Pergamum to the scripts of Attic drama and other Greek
poems.3 Some areas of the Latin scripts gave more trouble than
1
Cf. Wilamowitz, NGG, Phil.-hist. Kl. 1896, 228, GriechischeVersleunst
(Berlin, 1921), p. 125, 0. Crusius, PhilologusLV (1896), 384, Leo, Die plauti-
nischenCantuautuldiehellmistischeL yrik (Abh. Gott.GesellschaftN .F. 17 [ 1897]),
pp. 78 tf., Gesch. pp. 121 tf., R. Reia:enstein, NGG, Phil.-hist. Kl. 1918,
233 tf., 0. lmmisch, SB Heidelberg,Phil.-hist. Kl. XIV (1923), 7, 1 tf., H.
Drexler, PlautinischeAkzentstudienn (Breslau, 1932), pp. 358 tf. These discus-
sions concentrate their attention on Plautine comedy and more or less ignore
tragedy.
a Cf. Bergk, PhilologusXXXI(1872), 246 n. 23 ( = Kl. Phil. Schr.1207 n. 23),
Wilamowitz, Hermes xvm (1883), 248 f., Fraenkel, Pl. im Pl. pp. 340 tf.,
366 tf. ( = Elementi,pp. 324 tf., 346 tf.; see also Addenda,p. 439 ), RE Suppl. VI
(1935), s.v. Naevius, 632 tf.
3 Two closely related systems of description can be observed in the remains
of Latin metrical discussion and traced back to Greek theorists (see Leo,
HermesXXIV[1889], 280 tf.). From one system come the terms senarius,sep-
tenariusand octonarius,which first appear in Varro (cf. Diomedes, Gramm. 1
SIS. 3 tf., Rufinus, Gramm.VI SSS. S tf.) and Cicero (0rat. 184, 189, Tusc. 1. 1o6)

32
 THE FORM OF ROMAN TRAGEDY

others. Stichic arrangements of iambic, trochaic, cretic,

bacchiac, anapaestic and dactylic verses were readily dis-
cernible, although the number of departures from the classical
patterns and the comparatively high degree of metrical and
prosodical licence caused disquiet.1 The two manuscript tradi-
tions of Plautine comedy and stray remarks in late grammatical
writing provide evidence on how the more complex areas were
divided into metrical cola.1 Very little is known about how the
resulting cola were analysed. Here the absence of exact Attic
analogues was particularly bothersome to scholars.3
The scholars' mode of description may have encouraged the
strange notion that the Latin poets got their metrical patterns
from careless and incompetent examination of the scripts of
Attic drama. 4 Modem students of republican drama have found
this notion unpalatable and many put forward the equally im-
plausible one that Livius Andronicus-an immigrant Greek-
deliberately altered the patterns in the light of his intuitive
perceptionof the nature of the Latin language.5Attemptshave

and are still commonly used in modem discussion. Those who coined these
terms do not seem to have thought that their objects of description differed in
nature from Greek verses. Many ancient (e.g. Anon. Gramm. VI 286. 14: hie
Latine senariusquodpedessex simplicesluibeat,Graecetrimderquodtresav3vy{cxs
habeatappellatur)and modem students (e.g. those who allow word accent as an
clement in the structure of early Latin vcnc) have made the two sets of terms
cover differences of metrical nature. I use the terms trimcter, tetrameter and
catalcxisiD talking of both Greek and Latin verse simply for descriptive con-
venience and leave the theoretical issues open.
1
Cf. Marius Victorinus, Gramm.VI 81. 1 ff., Euanthius, De Jab. 3. 3.
3
Sec below, p. 51.
3 Cf. Charisius, p. 375.13 ff. (De satumio), Rufinus, Gramm. VI 561.8 ff.

(quoting a bewildered comment by Siscnna on a passage of apparent anapaests

in Plautus).
4 Cf. Horace, Ars 251 ff., Epist. 2. 1 .69 ff. Marius Victorinus (Gramm. VI
78. 19 ff.) records a notion based on similar premises, to the effect that those
who adapted the comedies of Menandcr took their metrical patterns from
Aristophanes.
s Cf. R. Klotz, Grundzuge,pp. 29 ff., Leo, Gesch.pp. 62 ff.

3 33 JTO
 INTRODUCTION

been made to argue that the catalectic trochaic tetrameter

(septenarius)1
and the iambic trimeter (senarius):i used by Livius
and other adapters of Attic drama owed something to verse
forms already established in Rome in 240. These attempts can
be criticised in detail3 but their approach to the problem is basi-
cally sound. The Latin dramatic poets had Latin-speaking
actors and a Latin-speaking audience to deal with as well as
scripts in Attic Greek. It is likely that the earliest of them would
have arranged the words of the new type of stage performance
in patterns like those which actors were used to mouthing and
audiences to hearing.
The most striking features of the Latin scripts are the presence
of large blocks of verses which Attic scripts either do not show
at all or do so only sporadically in lyric contexts, 4 the presence
of certain types of short verse quite absent from Attic scripts,5
the polymetry and absence of strophic corresponsion in extended
1
C£ Immisch, SB Heidelberg,Phil.-hist.Kl. XIV (1923), 7, 27 ff., Fraenk.el,
Hermes LXII (1927), 357 ff. ( = Kl. Beitr. D II ff.), Altheim, Geschichtede,
lateinischenSprache,pp. 366 ff.
:a C£ Altheim, Ciotta XIX (1930), 24 ff., S. Mariotti, Livio Andronicoe la
traduzioneartistica(Milan. 1952), p. 33 n. 3.
3 C£ G. Pasquali, Prtistoria dtlla poesia RomatUJ(Florence, 1936), pp.
46 ff.
4 I note in the meagre remains of Latin tragic scripts blocks of acatalectic

trochaic (e.g. Ennius 185--6,296-7) and iambic tetrameten (e.g. Ennius 322-
31; c£ Sophocles' satyr play 'lx\lEV'Taf,238 ff.), of cretic (e.g. Ennius 81-3; cf.
Aeschylus, Choe. 783, 794, Hik. 418-27, Euripides, Or. 1419-24; the 'cretics'
of Eubulus [fr. n2] and Aristophanes arc of a quite different kind [see Leo,
Die plaut. Cant. p. 74)) and bacchiac tetrameters (e.g. Ennius 290, 293-5; c£
Aeschylus, Ag. no3, 1no, Choe. 349-50, 367-8, Prom. n5, fr. 23, fr. 341,
Sophocles, Phil. 396-7, 5n-12, Euripides, Ba. 994, 1016, Hel. 64,2-3, Herakles
906,Ion 1446-7, Or. 1437-40, Phoin. 295, 1536, Rhes. 706-8, 724--6, Aristo-
phanes, Thesm. 1143-4).
s E.g. Livius, Trag. 20-2 da mihi hasu opts, quaspeto, quasprecor:porrige,
opitula(-v- v-1-v- -v-1-1....1'),8,Lv -; c£ Plautus, Most. 6<Jo-'746,Pseud.
1285-334; but for a different metrical interpretation see Leo, RhM XL (1885),
166, De Trag.Rom. p. 13 [ = Ausg. kl. Sehr.1202)). On Roman 'clausulae' see
Marius Victorious, Gramm.VI 79. 1 ff.

34
 THE FORM OF ROMAN TRAGEDY

lyric structures, 1 the high degree of metrical 2 and prosodical3

variation in some types of verse, the extremely frequent co-
incidence of metrical and rhetorical units,4 the regularity of
certain caesurae and diaereses,5the avoidance of certain shapes
of words in particular metrical positions6 and the blurring of the
sharp Attic distinctions between comic and tragic verses. There
is a little evidence suggesting that certain types of verse used
only in comedy at Athens appeared in both the tragic and the
comic scripts of the Latin poets. An ancient writer on metric7
seems to have read a catalectic iambic tetrameter in a tragic
script8 and at least two others can now be read among the
extant fragments.9 The same writer also quotes 10 a catalectic
anapaestic tetrameter whose language is thoroughly tragic and
several extant fragments of indubitable tragic origin are open
1
The fragmentary transmission makes definite statements about the tragic
scripts difficult but I note Ennius 43--9+69---73, 80-94. Similar structures are rare
in Attic scripts (but cf. Euripides, Or. 1369 ff., Aristophanes, Batr. 1309 ff.).
The corrcsponsion which F. Crusius (Die R.esponsionin den plautinischen
Cantica (PhilologusSuppl XXI 1, 1929]) and others have claimed to find in
various Roman lyrics (for more recent discussion of the issue see G. Maurach,
Untasuchungenzum Aufbau plautinischerLieder [Gottingen, 1964]) is quite
different from the Attic type.
, The use of long syllables in the 'pure' elements of iambic and trochaic
metra was one of the main causes of the criticism mentioned above, p. 33. For
metrical hiatus in tragedy see Commentaryon v. 17, fr. xv, fr. xxvnh, v. 126, fr.
LXVI. v. 154; for 'split anapaests', etc., on fr. xn.
3 For iambic shortening in tragedy see Commmtary on fr. rx; for ille (two
morae), etc., on v. 33; for prosodical hiatus, on v. 173, fr. XCVI.
• See Commentaryon fr. ex.
S See Commentaryon fr. IV.
6
See Commentaryon v. 126, fr. CIV, fr. cue.
7 Gramm. VI 613. 12 (for the writer's identity see G. Schultz, HermesXXII
(1887], 265, Leo, HermesXXIV [1889], 282 n., R. Heinze, SB Leipzig, Phil.-hist.
Kl. 1918, 4, p. 21 n. 1 [ = Vom Geist desRomertums',p. 241 n. 32)).
8
J. Kraus (RhMVIII (1853], 531 ff.), L. Mueller (N]bb XCVII (1868), 432 f.)
and othen have dismissed the verse quoted-haec bellicosus(Carrio: bellicosis
cod.) cui pater mater (Carrio: mater pater cod.) cluet Minerua-as a gram-
matical concoction. It could also, of course, be a paratragic verse from comedy.
9 Pacuviw 131-2, Accius 64-5. See Strzelecki, in TragicaI, pp. 43 1f.
10
Gramm. VI 614. 3.
35
 INTRODUCTION
to this metrical interpretation. 1 The so-called 'uersus Reizianus 'i
has been detected by modem scholars in Naevius, Trag. 133
and Ennius 246. 4
The number of perceptible differences in the metrical
structure of Latin comic and tragic scripts is small.
The comic scripts contain short groups of words which could
be interpreted as dactylicS but nothing like Ennius, Trag.43-6
(four dactylictetrameters)or 250 (dactylichexameter).6 Early
Latin tragic trimeters and tetrameters show, when examined in
large groups, somewhat less resolution of long elements and
replacement of short elements by longs and double shorts than
do contemporary comic verses, less synaloephe and synizesis of
adjacent vowels, less tolerance of hiatus7 and 'split anapaests'
and less treatment of normally iambic sequences of syllables as
pyrrhic. No shape of trimeter or tetrameter, however, seems to
occur in comic scripts and not in tragic. 8 Whereas comic ana-
1
E.g. Ennius S, 365-6. Sec Strzelecki, in TragicaD (Wroclaw, 1954),
pp. 93 ff.
a This vcne does occur in Attic tragedy(e.g. Sophocles, Ai. 4~) but is
extremely rare.
3 W. M. Lindsay, Early Latin Verse (Oxford, 1922), p. 279 n. 2.
4 Sec Commmta,y on fr. cax.

S Sec W. Meyer, Oberdie Beobachtungdes Wortaccmtesin de, altlateinischm

Poesie(Abh. Bayer.Ak. xvu, 1886), pp. 93 ff. Leo (Die plaut. Cant. pp. so, 52)
however argued that Plautus' -vv-vv -vv- is a form of the glyconic.
6
More tragic hexamctcn may lurk among the verses quoted in our sources
with only Ennius' name and assigned by Vahlcn to the Annales.
7 Sec B. Maurenbrecher, Hiatus und Vnschleifungim altm Lattin (Leipzig,
1899), pp. 225 f.
8
Strzelecki (De SenecaeTrimetroIambicoQ_uaestiones Selectae[Krak6w, 1938),
pp. 94 ff.) follows F. A. Lange (~stiones Mdricae [Diss. Bonn, I 8s1]) in
arguing that tragic trimctcrs and tctramctcn of the republican period as well as
of the imperial period never have a short syllable occupying the clement pre-
ceding a final crctic-shapcd word o,r word-group. Q!µte savage emendation is
required to make some transmitted verses conform to 'Lange's law'. However,
even if such a law did bind republican tragic verses, the strong tendency of
comic verses to have a long syllable at the point in question would have made
its rhythmical effect scarcely perceptible to a listening audience. The operation
 THE FORM OF ROMAN TRAGEDY

paests contain with great frequency 1 sequences of syllables

normally iambic scanned as pyrrhic, there is little prosodic
variation of any kind in tragic anapaests. The latter have diaere-
sis between metra and divide disyllabic elements between words 1
rather more often than do comic anapaests. The cretic and
hacchiac verses of comedy stand apart from other types of
comic verse in the small amount of metrical and prosodical
variation they contain3 but not, apparently, from their tragic
counterparts.
This blurring of the Attic metrical distinctions springs not
only from the fact that there was one and not two dramatic
traditions at Rome but alsofrom the different way in which the
Latin-speaking poets and their audience viewed the personages
and the subject-matter of the two Attic genres. For the average
Athenian the themes of tragedy were ancient history, the heroes
remote and extraordinary beings. It was fitting that they should
speak in tight rhythmical patterns. Comedy on the other hand
usuallydealt with the contemporarylife of Athens,by the late
fourth century almost exclusively so. Its chief personages were
city-dwelling property holders and the rhythms of their speech
needed to be only sufficiently distinct from those of every day
to mark their slightly elevated position in Athenian society.
For Romans of the third century B.C. the personages of both
genres were almost equally remote in space and time. They
differed in little more than social class and poets would have

of'Ponon's law', on the other hand, helped to make a group of Attic trimeten
or tetrameters rhythmically very different from a group of their comic
counterparts.
1
With much greater frequency, in fact, than iambic and trochaic verses; see
A. Spcngcl, Refonnvorschliigezur Metrik der lyrischen Versartenbei Plautus
(Berlin, 1882), pp. 309 ff.
2
SeeJ. Perret, REL xxxm (1955), 352 ff.
3 See 0. Scyffert, De &cchiacorumVersuum Vsu P1"utino(Diss. Berlin,
1864), A. Spcngel, Reformvorschl.pp. n3 ff., 193 ff., G. Jachmann, Glotta VII
(1916), 39 ff., RhM LXXI (1916), 527 ff.

37
 INTRODUCTION

found the sharpness of the metrical differences between the

plays they were adapting meaningless from a theatrical point
of view. They did not, however, totally obliterate the
differences. Scholastic tradition must have had a strong hold
even then. 1
The general linguistic form of the two Roman genres no
more reproduced the Attic situation than did the metrical. At
Athens even in the sixth century it was understood and ex-
pected that poets should use for serious themes a vocabulary
remote from that of everyday life. The tragedians had audi-
ences acquainted with several different types of poetry whose
forms had been shaped in communities speaking distinct
dialects of the Greek language and which enjoyed universal
cultural prestige. They found a ready acceptance of the conven-
tion according to which one set of non-Attic words informed
the utterances of the actors and another those of the chorus. The
comedians first gained public recognition some time after the
conventions of tragedy had been firmly established and always
thought of their genre as the polar opposite of tragedy. By the
time of Menander their normal vocabulary scarcely differed
from that used in contemporary polite society.2 In third-century
Latium there seem to have been no commonly recognised
traditions of public poetry-extempore compositions are an-
other matter-and speakers of the Roman dialect probably
already looked down on others. The Latin poets could not
have reproduced the Attic situation even if they had·so desired.
In place of the three very distinct vocabularies of the Attic stage
they offered one, based largely on that regularly used in the
1
The formulations of some scholan (e.g. R. Klotz, Grundzuge,p. 22) on
this issue are misleading. See Lindsay, E.L. V. p. 274.
a For the language of tragedy see Aristotle, Poet. 22 .1458 a 18 ff., Rhet.
3.7.1408 a 10 ff. The observations concerning the two genres at Cicero, Opt.
gen. 1, Qyintilian, Inst. 1. 8. 8, 10. :2. 22 probably derive to some extent from
Greek sources.
 THE FORM OF ROMAN TRAGEDY

houses of the great Roman families1 but drawing also on the

special languages of religion,i law,3 and public administration4
and apt to resurrect obsolete morphology,5 to produce new
words 6 and to vary the usage of common words and phrases7
for the purpose of amplifying the tone of discourse. At the same
time they indulged freely in hyperbole 8 and metaphor9 and
created artificial patterns of words and phrases to a much greater
extent than any of the classical Attic poets. 10 Some Roman
critics 11 deplored the absence of the sharp Attic distinctions of
language between republican comedy and tragedy. Neverthe-
less the mode of speech normal to the personages of tragedy
had from the beginning a certain character of its own, being
much more artificial and remote from that of every day, and
was frequently aped for comic effect by the slaves of early
comedy. n The modern student cannot often isolate in the re-
mains of the Latin tragic scripts words and turns of phrase and

• For the aristocratic connections of most of the Latin poets there is good
evidence. The silence of our sources in regard to the rest should not be treated
as evidence of popular connections. The management of the ludi was firmly in
the hands of the aristocracy.
1
See Commentaryon fr. IV, v. 41, v. 4-2, fr. xvm, fr. xxvnh, fr. XXXIV,
v. 103, fr. LJCCCVI, fr. ex, v. 247, fr. CXXI, v. 280, v. 287.
3 See Commentaryon fr. XIV, vv. 48-9, fr. xxvnh, fr. XLV, fr. LXIV, fr. LXXXVII,
v. 272.
4 See Commentaryon fr. I, vv. S, 6, 74-s, 123, 137, fr. LXXV, vv. 173, 200,
212, 214, fr. CV.
S See Commentaryon vv. 37, S9, 79, n2, ISI, 183, 248.
6 See Commentaryon vv. S, 17, 20, 25, 26, 4S, 68, 91, 93, fr. xxxvm, vv. no,

III, II3, IIS, 136, 138, ISO, 169, 246, 279.

7 See Commentaryon vv. 2, 33, 3S, 39, 43, 44, 46, 61, 71, 72, 74-s, 76-,JJ,88,
9S, IOS, 127, 134, 156, 166, 168, 232, 267, 299.
8
See Commentaryon vv. 24, 41, 73, 272, 281.
9 See Commentaryon v. 3, fr. xn, vv. 35, 43-4, S3, S7, 96, fr. XLII, vv. 143,
144, 165, 171, 180, 187, 188-9, 216, 229-30, 243, 245, 304.
10
See Commentaryon fr. I, vv. 4, 6-,JJ,8, 9, frs. IX, XIV, vv. 17, 19, 21, 24, 34,
39, S4, s1, 62, 63, 74-s, 1f>..JJ. 83, 87, 90, 94, 100, 10s, 192.
11
E.g. Horace, Ars 86 ff., Gellius 2.23 .21. Cf. Schol. Soph. Ai. n27, Schol.
Eur. Atulr. 32, Plutarch, Mor. 853c-D.
i:a Cf. Plautus, Pseud.70~ ... ut paratragoed41 camufex.
39
 INTRODUCTION

label them with certainty as peculiarly tragic. Nevertheless

some words and phrases can be shown from their comparative
frequencies in the tragic and comic scripts and the contexts
in which they occur to have possessed a more elevated tone
than others. 1 The Latin distinctions were ones of degree, not
of kind.
As the second century advanced and the two Roman genres
came more and more to be cultivated by specialistpoets, the
language of comedy moved away from that of tragedy and
approached the common language. i That of tragedy on the
other hand seems to have become more elaborate and artificial.3
Both the Attic and the Latin poets varied their metrical
patterns according to the substance and tone of the actor's
speech. At Aeschylus, Pers.176, for example, the Q!!een inter-
rupts a highly emotional dialogue with the chorus in trochaic
tetrameters to recount a dream in trimeters. 4 The dialogue in
various musically accompanied metres at Plautus, Persa482-
548 is interrupted twice while a letter is read out in trimeters.5
At Plautus, Rud. 1338 iambic tetrameters cease as one of the
actors begins to recite an oath. Such changes of metre within a
scene are on the whole uncommon. A scene is usually set either
in trimeters or in musically accompanied verses. In Attic scripts
an increase in uncommon vocabulary and rhetorical embellish-
ment of the phrase becomes noticeable when trimeters are
1
Sec Commtntaryon vv. 3, 7, 9, 13, 17, 18, 21, 22, 29, 66, 89, 100, 116, 162,
178, 213, 215, 218, 219, 222, 225, 237, 238, 242 et al.
i Cf. Euanthius, Defab. 3. 5 tam illud est admirandum(scil. in Tertntio) quodet
moremretinuitut comoediamscriberetet temperauitaffectumne in tragoediamtransili-
ret, quod cum aliis rebus minime obtmtum et a Plautoet ah Afranio et Appio et
multisJere magniscomidsinutnimus.
3 Cf. Cicero, Orat. 36 on the styles of Ennius and the other writen of
tragedy.
4 Cf. the two speechesfrom Accius' Brutusquoted by Cicero, Diu. 1 . 44; the
king describesa dream in trimeters, the coniectores
give their interpretation in
trochaic tetrametcn.
s Cf. Plautus, &ah. 99"1tf., P.seud.998 tf.
40
 THE FORM OF ROMAN TRAGEDY

abandoned even if only stichic tetrameters replace them. 1 In the

comic scripts of Plautus the difference in stylistic level between
the trimeters of the action and the musically accompanied
verses is very pronounced. In those of Terence this difference is
much smaller but still perceptible. In the remains of the late
second century tragic scripts of Accius, trimeters seem not to
differ from the musically accompanied verses. The small
amount of material available from early tragedy and the un-
certain scansion of many small fragments make it hard for one
to be positive, but the proposition that early writers of tragedy
did not make a distinction between the two types of verse
parallel with that made in contemporary comedy is on general
grounds unlikely. There are some signs that Ennius kept the
trimeters of the tragic action much less elaborate than other
verses. i Prologue trimeters, as in comedy, would have occupied
a special position.3
1
C£ Wilamowitz, SB Btrlin 1916, 73 f. ( = Kl. Sehr. I 423 f.).
3
See Commentaryon v. 9, fr. IX, v. 17, v. 19.
J The fact that the early Latin poets imitated in their own way the genre dis-
tinctions of Greek poetry was pointed out by Ritschl (Partrgazu Plautusund
Ttrtnz 1 [Leipzig, 1845], p. 112) and Bergk (N]bb LXXXm [1861], 631 [ = Kl.
phil. Sehr.1 302 ]) • The latter alsoremarked on the way in which the sung vencs
of drama {both tragedy and comedy) differed stylistically from the spoken.
Important illwtrations of the way epic style differed from tragic have been
given by Fracnk.el. RE Suppl. v (1931), s.v. Livius, 6o3 ff. and of the distinc-
tions between the two types of venc within comedy by Leo (NGG, Phil.-hist.
Kl. 1895, 415 ff. [ = Ausg. kl. Sehr. 149 ff.]), Fracnk.el (Pl. im Pl. pp. 209 ff.
[= Elemtnti, pp. 199 ff.], Iktus und Akzent im lateinischenSprechvtrs[Berlin,
1928], p. 93 n. 1), E. Lindholm (StilistischeStudien zur Erweiterungder Satz-
glitdtr im Lateinischtn[Lund, 1931], pp. 94 ff.) and H. Haffter (Untersuchungen
zur oltlateinischtnDichtersproche [Berlin, 1934], passim).Haffter asserts (p. 124)
that 'die tragischen Senare im Gebrauche aller bier bcsprochenen Stilmittel in
keincr W eisc von den Langversen der gleichen Gattung sich unterscheiden •.
The general approach of these scholars has beenattacked by M. Lenchantin De
Gubcmatis (Athtnaeumxm [1935], 278 ff.) from the point of view of idealist
aesthetics and by M. Leumann (MusH IV [1947], 116 ff. [ = Kl. Sehr. pp.
131 ff.]) on empirical grounds. Lenchantin contributes nothing of substance to
the argument. Leumann merely reiterates that the Latin distinctions are not
exactly parallel with the Greek distinctions.
41
 INTRODUCTION

Three sources have been suggested for the origin of the types
of phrasal elaboration characteristic of certain parts of early
Roman drama: the modes of public speaking taught in Greek
schools and practised by Roman politicians, 1 the Attic Tpay1K11
M~•sand the formulae of Roman law and religion. 2 A parallel
for every type of Latin elaboration can be found in the scripts of
Attic tragedy but some, like alliteration, are relatively uncom-
mon and the rate of occurrence of the others rarely approaches
even that of the spoken verses of Terentian comedy. The
paucity of what survives of Roman legal and religious formulae
makes discussion difficult. Many were taken over as they stood
by the dramatists or used as models for new creations but they
never seem to have provided more than a light antique colour-
ing to the style of an actor's speech. Our actual knowledge of
third and second century Roman oratory is slight but we can
guess that men prepared to tolerate the hellenisation of their
community's religion and public art would not have neglected
the powerful aids to politicalsuccessofferedby the techniques
of speaking taught in contemporary Greek schools. Those Latin
poets who had received their education in Greek-speaking com-
munities could not have escaped the influence of the pflTOpES.
Those who had close personal relations with their Roman
patrons could not fail to be interested in the debates of the
senate, the law courts and the assemblies.3 Here, in turns of
phrase and sentence with which orators raised their discourse
above the commonplace and worked on the emotions of their
hearers, lay ready material for dramatists trying to construct a
poetic style for a people poor in distinctively poetic traditions.
It seems unlikely that Cato would have delivered himself
1
Cf. E. Norden, Die antikeKunstprosan 1 (Leipzig, 1898), pp. 839, 889, Leo,
Analecta Plautina II (Gottingen, r898) ( = Ausg. kl. Sehr. 1123 ff.), Gesch.
pp. 34 ff., F. Eckstein, PhilologusLXXVD (1921), 173.
:a Cf. Fraenkel, Pl. im Pl. pp. 356 ff. ( = Elementi,pp. 338 ff.).
3 For Ennius' interest in public speaking see Ann. 303 ff.

42
 THE FORM OF ROMAN TRAGEDY

publicly in language more elaborate than that of his philhellene

peers and elders, and yet those long passages of his orations
which are preserved usually show much more adornment than
is observable even in the scripts of Attic tragedy. 1 The trimeters
of comic prologues have a high degree of elaboration, Terence's
at times even more than his musically accompanied verses, and
it is perhaps significant that the prologists of the Amphitruo"and
the Hautontimorumenus3 cast themselves explicitly as oratores.It
would be foolish to assert that the Attic Tpay11Ct') M~1shad no
influence on poets constantly adapting Attic plays but the forms
of elevated speech already familiar to third and second century
Roman audiences should be considered the dominating
influences.

V. ENNIUS
Like all the early poets of whom we know anything Ennius
came from outside Latium and must have acquired his Latinas a
second or third language. He was a Messapian of high birth4
who in adolescence had received a Greek literary educations
and in manhood had served as a soldier of fortune in one of the
South Italian units of the Roman army, 6 high enough in the
1
Suppl. xxvm 2, 1935]) perceived this
R. Till (Die SpracheCatos[Philologus
but drew the unlikely conclusion that Cato's oratorical style was inBuenced by
that of Ennius' poetry. :a 20, 33 £, 50. 3 II £
4 Sec Silius 12. 393, Scrvius, Aen. 7. 691, Souda E 1348 (Aelian). It is
possible, of course, that Ennius had claimed only to come from the land. of
King Messapus.
s This is all that Festus, p. 374. 8 £ and Suetonius, De gramm.1 mean by the
terms Graecusand semigraecus. Whether he had received a philosophical or
rhetorical training as well we cannot say for sure but the general features of his
poetry and certain particular fragments (c£ among those of tragedy IV, ex,
CLXXXVDI)suggest he had both. Sec H. Frankel, HermesLXVIl (1932), 308 ff.,
LXX (1935), 62 ff.
6 Sec Cornelius Nepos, Cat. 1. 4, Cicero, Cato 10, Ps. Aurel. Victor, De uir.
illustr. 47 .1. On Iapygian and Messapian units fighting in 225 see Polybius
2.24.11.

43
 INTRODUCTION

ranks 1 to be able to make the acquaintance of the quaestor M.

Porcius Cato. He lived with one servant in his own house in the
artisans' quarter on the Aventine and for some time pursued the
profession of grammaticus. i He is known to have been on inti-

mate terms with various politically active members of the

aristocracy, at first with Cato, who brought him to Rome in his
entourage in 203, and later with men of factions hostile to
Cato's. The political debates of 203-169 sometimes seem3 to
make themselves heard in the scripts of the tragedies he adapted
for the festivals managed by his aristocratic patrons. Like
Livius and Naevius he adapted comedies as well as tragedies and
like Naevius he composed plays of the tragic type on themes
from Roman history.
In a medieval glossary there appears in a discussion of the
history of tragedy the interesting note: tragoediasautemJere
omnesex Graedstranstulit,plurimasEuripidis(Lindsay: Euripides
codd.), nonnullasAristarchi{Lindsay: Aristarchuscodd.).4 The
source of the note is obscures but there is no reason to thinkit an
invention. It may come from one of Varro' s treatises by way of
Suetonius. One of Ennius' titles, Achilles, is attributed else-
where6 to Aristarchus and four, Alexander,Andromacha,Hecuba
and Medea,7 to Euripides. Of the seventy tragic scripts of
1
Silius 12. 394-5 LatiaequesuperbumIuitisadonuibatdextramdecussuggests that
he was a centurion, but such statements need to be treated with caution.
i I do not share Fraenkel's doubts about Suetonius, De gramm. 1 (RE Suppl.
V [1931), 6o1).
3 See Comtnffltaryon frs. LXXXIV, cv.
4 Gloss. Lat. 1 568. Cf. Donatus' addition to Suetonius' Life of Terence, 10,
duaeab Apollodorotranslal4eesse dicuntur comico,Phormio et Hecyra; quattuor
reliquaea Menandro.
s Cf. H. Usener, RhM xxvm (1873), 417 ff. ( = Kl. Sehr. m 36 ff.). On the
'Glossary of Ansileubus' or 'Liber Glossarum' in general see G. Goetz, Abh.
siichs.Ges., Phil.-hist. Kl. xm (1893), 256 ff., Lindsay, CQ XI (1917), 119 ff.
GlossariaLatina 1 {Paris, 1926), p. 8, S. Timpanaro, StudUrb, Serie B, XXXI
6
(1957), 178. Plautus, Poen. 1, Festus, p. 282. 10.
7 Varro, Ling. 7. 82, Cicero, Opt. gen. 18, Gellius u. 4, Cicero, Fin. 1. 4.

44
 ENNIUS
Aristarchus p<messed by the Alexandrian library we know
1
from Greek sources the title of only one, T<XVTCXAC>S, but of the
seventy catalogued under the name of Euripides we seem to
know all the titles and can guess with a fair amount of certainty
at their general themes. i If the story of Ptolemy and the
Lycurgean text of the classical tragedians3 has any basis in fact
these could have been the only ones surviving in Athens in
c. 330 B.C. and it is most unlikely that scripts of the other plays
would have survived for long anywhere else. Andromacha,
despite the statements ofV arro, Ling. 7. 82 and Cicero, Opt.gen.
I 8, cannot be regarded as an adaptation of any of the Euripi-
dean seventy.4 Neither can Aiax, Eumenides,Hedoris lytra,
Nemea or Telamo.There is nothing on the other hand in the
remains of Alexander,Andromeda,Eredheus,Hecuba,lphigenia,
Melanippa,Phoenix, Telephusor Thyestesto make one wish to
deny Euripidean provenance. About Athamas, Alcmeo and
Cresphontesthere is doubt. The title Medeacovers two plays, one
quite certainly a version of Euripides' MT)6eta,the other pos-
sibly a version of his Alyevs.V ahlen is wrong to dismiss5the
plain statement of the glossary article that Ennius adapted more
than one play by Aristarchus. There are several possibilities
among the extant titles and in any case we cannot be sure that
these titles exhaust Ennius' tragic production. 6
The texts which Ennius worked from very likely resembled
1
F. Bws, LittrarischesCentralbl.1893, 1434, restored hrl TOO'Ax1M]fo>sTOV
•AptO"Tapxovin the Flinden Petrie papyrus mentioned above, p. 9 n. 2.
a Cf. Wilamowitz, AMlecta Euripidea (Berlin, 187s), pp. 144 ff., W.
Schmid, GtschichledtrgmchischenLittratur1 iii (Munich, 1940), pp. 329£. There
arc a number of titles too many; see E. G. Turner, in Acta of the 9th Inttr-
Mtional Congressof Papyrology, Oslo 1958 {Oslo, 1962), 1 ff., for further
evidence that two scripts bore the title <l>p(~oS-
3 Plutarch, Mor. 841 P, Galen, Hipp. Epid. 2. 4.
4 See Commentary,pp. 236 f[ S E.P.R. 1 , p. ccr.
6 See Varro, Ung. s. 14 ( = fr. CLXXXVI)and below, p. 62, 7. 13 ( = fr. cxc)
and below, p. 62.

45
 INTRODUCTION

those found in certain third century B.C. papyri from which the
Alexandrian colometry of the lyrical passages is absent.1 Philo-
logical examination of the tragic texts had already progressed
a long way by 203 i and Ennius doubtless possessed and used
commentaries upon them. The suggestion, however, that what
they contained sometimes radically affected Ennius' own
version3 seems an unlikely one. We must distinguish between
versions of poems made for dramatic presentation to audiences
largely ignorant of literary Greek and versions of such poems as
Aratus' <J>oov6µwa made for circulation in books. Only the
latter were aimed at people who knew the Greek originals (and
perhaps, in some cases, the scholarly apparatus as well) and who
would set the versions against the originals.4 Ennius had no
motive that we can see for worrying about the exact interpreta-
tion of the Greek tragic texts.
Unlike the classical Greek dramatists Ennius probably did
not compose his own musical scores.5 Some of the classical
scores still existed in his day 6 but even then Greek musicians
1
For an accowit of these papyri sec G. Zwitz, An Inquiryinto the Transmis-
sion of the Plays of Euripides(Cambridge, 1965), pp. 250 £
:a C£ Wilamowitz, Einleitung in die attische Tragodie (Berlin, 1889),
pp. 134 tf.
3 C£ Leo, Pl. Forsch.•p. 98, Gesch.p. 192.
4 On Cicero's version of Aratw' <l>cnv6j.lE\l(X and the scholia see Leo,
HermesXLIX(1914), 192 £ (= Ausg. kl. Sehr.1279 f.); on Varro Atacinus and
the scholia to Apollonius' epic see E. Hofmann, WSt XLVI (1928), 161. Accep-
tance of H. Frankel's view (HermesLXVn [1932], 306; c£ Mariotti, Livio
Andronico,p. 28) of the relationship between the scholia and Livius' version of
the Odyssey would entail regarding this poem as quite a sophisticated piece of
work.
s C£ the dicbsca1iae to Terence's comedies and Donatus, De com. 8.9
deuerbiahistrionespronuntiabant,canticauerotemperabantur modisnon a poetased a
peritoartismusicaefactis.Cicero however talks at Leg. 2.39, perhaps loosely, of
'modi Liuiani et Naeuiani'.
6
The discussion of Dionysius Hal. at De comp.uerb. 11. 63-4 implies that
students of his day had access to at least some of the score to Euripides'
•opm11s. A iii-ii B.C. papyrus of the score to vv. 338-43 of the tragedy (no. 4n
in Pack's catalogue) survives; see Tumer,JHS LXXVI(1956), 95 tf.

46
 ENNIUS

were composing new scores for parts of the old scripts1 and it
seems likely that the music composed for the Roman tibiae,and
that for the cxvA6sof classicalAthens differed as widely as did the
metrical patterns imposed on the Latin and Greek words.

VI. THE HISTORY OF THE TEXT OF

ENNIUS' TRAGEDIES
Ennius' plays were well known to theatre goers in 166 B.c.3and
no doubt had been for some time. 4 These and other plays of the
early period5 were constantly produced at first century B.C.
festivals. The last public performance of a play by Ennius
which is clearly recorded took place at the ludi Apollinaresof
S4 B.C. 6 Accius' Tereuswas performed in its entirety in 447 and
extracts from Pacuvius' Armorum iudiciumand Atilius' Electra
1
Cf. Latte, Eranosm (1954), 125 ff., LIII (1955), 7S f., S. Eitrem and L.
Amundsen, SO XXXI (19SS), .26ff.
2
An instrument that probably came to Rome from Etruria (cf. Livy
9. 30. s-10); it is usually assumed to have been of the same character as the
fifth-<:entury Athenian a(JM,.
3 SecTerence, Amir. 18.
4 Plautus, Batch.214 is the earliest evidence for the performance of old plays
at the Roman festivals. Many imitations of Ennian passages, of which the most
convincing is Batch.933 o Troiao patria o Pergamumo Priameperiistisenex ~
Trag. 87 o paur o patri4 o Priami domus,have been found in the comic scripts
attributed to Plautus. There is a clear reference to Ennius' Achillesat Poen. 1
(sec Commentaryon fr. I) but no certainty that Plautus (d. c. 184) made it.
s Varro expected the readers of his treatise De linguaLatinato be acquainted
with the plot of a play by Livius about Teucer (7. 3) but Cicero's disdainful
remarks at Brut. 71 suggest that only the learned knew Livius' plays at fint
hand. The EquosTroianusproduced by Pompey in SS (see Cicero, Fam.7. 1. 2)
was probably Naevius' tragedy.
6
SecCicero, Att. 4. 1 s.6. It is uncertain what kind of performance Ovid is
referring to at Rem. 383, or even whether the script of Ennius' Amlromachawas
used.
7 Sec Cicero, Att. 16. 2. 3, 16. s. 1, Phil. I. 36. Cf. Ribbcck. Q!y,estionum
saniurum Mantissa,in TragicorumLatinorumReliquiae(Leipzig, 1852), p. 326.

47
 INTRODUCTION
were sung at Julius Caesar's funeral games. 1 At the time when
Horace was composing Epist. 2. I . 23 ff. the old poets still
enjoyed great popularity in the theatre and the emperor
Augustus notoriously favoured their work. 1 Nero (A.D.54-68)
had the Incendiumof Afranius staged with a realism that gained
notoriety.3 Q_uintilian'sdiscussion at Inst. 11. 3. 178-82 suggests
that at least some of the comedies of Terence were regularly
performed in his time. It is difficult to tell when fresh informa-
tion based on knowledge of the contemporary theatre ceased to
be inserted in commentaries upon these plays but the reference
to use of female actors at Donatus, Ter. Amir.716 looks quite
late.
Only three names of post-republican comic poets are certainly
known, those ofFundanius,4 Marcus Pomponius Bassulus,5and
Vergilius Romanus. 6 Q!!intilian could find none of sufficient
commonly recognised standing to set beside the classical
Greeks.7 Many on the other hand tried their hand at tragedy.
The Thyestesof Varius was performed at the festival held to
celebrate Octavian' s victory at Actium 8 and reliable witnesses
attest the performance of tragedies by Pomponi us in the theatre. 9
Some wrote tragedies explicitly for private recitation. 10
That other tragedians whose names are recorded intended
their work for performance in the public theatres is not stated
in the ancient sources and has often been denied by modem
students. Ovid's Medea was much admired as late as Q!!in-
tilian' s time. 11 The plays of V arius, Gracchus, Pomponius and
1
Sec Suetonius, Iul. 84. 3
Sec Suetonius, Aug. 89. 1.
3 Suetonius, Ner. II. 4 Horace, Sat. 1. 10.42.
5 C.L.E. 97. 6
Pliny, Epist. 6.21. 1-4.
7 Inst. 10. I .99.
8
Cod. Paris. 7530, cod. Casin. 1086.
9 Pliny, Epist. 7. 17. II, Tacitus, Ann. I 1. 13. For relations between Pom-
ponius and Seneca see Q_uintilian.Inst. 8. 3 . 31.
10
Sec Tacitus, Dial. 2 ff.
11
Sec Q_uintilian,Inst. 10. 1.98, Tacitus, Dial. 12.
 HISTORY OF THE TEXT OF ENNIUS' TRAGEDIES

Seneca were drawn upon by the elder Pliny in his treatise on

Latin morphology and by Caesius Bassius in his treatise on Latin
metric. These treatises are probably the ultimate sources of the
quotations scattered through the handbooks of late antiquity.
Seneca's alone of the mythological plays written in Latin sur-
vived in their entirety until the age of printing. 1
The almost total silence of our literary sources concerning the
production of complete tragedies at the festivals of Latin-
speaking communities after the first century A.D. 2 may have no
significance at all. Most extant late writers, both pagan and
Christian, were hostile to the theatre3 and, if plays enjoying the
educational prestige attaching to the names tragedy and
comedy had been performed there, these writers would have
been unwilling to advertise the fact. The notion that the mass of
the population of Rome became too degenerate, both morally
and culturally, to tolerate high comedy and tragedy has no-
thing to commend it. Gladiatorial contests and knock-about
farce were without a doubt extremely popular long before the
advent of the Attic type of drama and the performers of the
latter always had to contend with audience behaviour which
would have been unthinkable in Athens. 4 There is no reason to
think that mass tastes ever altered in one direction or the other,
but some that, except during the period of economic collapse in
the third century A.D., a superficial knowledge of and respect
for the high poetic genres increased among the upper classes.
1
Varius' Thyestes may have reached the age of Charlemagne (sec A. E.
Housman, CQ XI [1917], 4-2, and, contra,Lindsay, CQ XVI [1922], 180).
:a The evidence is collected by Welcker, Im griech.Trag.pp. 1319, 1477 ff.,
A. Miillcr,'Das Biihncnwcscn in dcr Zcit von Constantin dcm Grossen bis
Justinian ', N]bb xxm (1909), 36 ff. (esp. 40 f.), L. Friedlander, Darstellungen
ous tin Sittengeschichte Roms8 (Leipzig, 1921), vol. 11, pp. 1 ff., 112 ff., 118 ff.
3 This hostility was no new thing among pagan aristocrats (cf. Cicero, Sest.
119, S. Rose.,46,I.ivy 7. 2) no mere reflection of Plato's ancient prejudices.
4 Sec Plautus, Amph. 51 ff., Terence, Hee. 1 ff., 25 ff. Cf. Polybius' descrip-

tion (30. 22) of the ludi celebrated by Anicius.

49 JTO
 INTRODUCTION

Terence's work formed part of the regular syllabus of Latin

literature 1 while that of the Attic dramatists was necessary
reading for every student of Greek. 2 Roman houses of the
wealthier sort were frequently adorned with figurative repre-
sentations of various aspects of comic and tragic theatrical per-
formances.3 It seems to me likely that those who managed the
Judiwould have at least occasionally paid obeisance to cultural
tradition and allowed the performance of plays commonly read
in the schools. One might have expected a comedy of Terence,
possibly a comedy of Plautus4 or an Attic play according to the
original script,5 but hardly any work of the other Roman
dramatic poets. 6
Somewhat more is known and can be conjectured about the
fate of those Ennian scripts which were recorded in books and
circulated among the Roman reading public.
In the early part of the second century B.C. the scripts of plays
performed at Judiscaenicimay have been seen by few people
outside the actor's companies. As late as 161 Terence could
claim, apparently in good faith, that he did not know that
Menander' s K6ACX~ had akeady been adapted for performance on
1
With Cicero, Sallust and Virgil Terence formed the 'quadriga • of
Arusianus Messius (see Cassiodorus, Inst. diu. 1 s.7).
a On knowledge.of Greek in the West sec P. Courcelle, Lts lettresgreCIJUtS en
occidtnt,deMacrobeaCassiodorel(Paris, 1948), H.-I. Marrou, Saint Augustinet "2
Jin de la cultureantique' (Paris, 1958), pp. 27 ff., 631 ff.
3 For the archaeological evidence sec M. Bieber, The Historyoftht Greekand
Roman Theater (Princeton, 1961), pp. 227 ff., Webster, AJA LXVI (1962),
333 ff. Bieber and Webster grossly overestimate the value of this evidence.
Changes in the form of artistic representations need not rcficct the fashions of
the contemporary theatre.
4 Amobius 7. 33, however, is no evidence for a performance of the Amphitruo.
s See above, p. s n. 3. It is perhaps significant that Juvenal speaks of the per-
formance of tragedy at Roman festivals (6.67 ff., 396 f.) but names Sophocles
alone of actual tragedians (6.634 ff.). Even Persius refen to Menander's
Ewovxcs(s. 161 ff.) rather than to Terence's adaptation (contrast Horace, Sat.
2.3.259ff.).
6
For their neglect in the schools sec below, pp. SSff.

so
 HISTORY OF THE TEXT OF ENNIUS' TRAGEDIES

the Roman stage.1 The sources of Suetonius' treatise De gram-

matids et rhetoribusknew of no serious philological activity in
Rome until the year after Ennius' death and the state of the text
of the extant comedies of Plautus makes it plain that the first
scientific editors had to deal with scripts which had passed
through the hands of professional actors and were rarely, if ever,
able to obtain a Plautine autograph. 2 Ennius' tragic scripts were
frequently reused at the early festivals and they may have suf-
fered the same sort of damage that can be observed in the
extant text of Plautine comedy.
Editions of the Annales employing the conventions estab-
lished at Alexandria for the editing of classicalGreek poetry are
recorded with the names of C. Octauius Lampadio and Q.
Vargunteius3 but no ancient author mentions an edition of
Ennius' tragedies. Nevertheless we may guess on the analogy of
what the extant texts suggest for the comedies of Plautus and
Terence4 that they were also edited in the Alexandrian manner,
each one prefixed with 616a01«XA{cn containing, amid other
information, the title and author of the Greek original and
with Ennius' name and the title commonly given to his adapta-
tion suffixed. The discussionsof the metricians5and such quota-
tions as that of v. 44 by Cicero at Orat. 155 provide solid
evidence that the ancient editions presented the lyric passages
divided into cola.
1
Eun. 30 ff.
:a For the theatrical transmiuion of the texts of comedy sec F. Osann,
Atuilectdcrltiui (Berlin. 1816), pp. 147 ff., Ritschl. Parerga,pp. 88 ff.
3 Cf. Gcllius 18. s. II, Suctonius, De g,amm. 2, Fronto, p. 15. 13 ff. van den
Hout, Frag. Paris. Gramm.VII 534.4, Timpanaro, SIFC N.S. XXI (1946), 49 ff.
4 Cf. W. Studemund, in Festgruss derphilologischen Gesellschaftzu Wurzburg
an die XXVI. VersammlungdeutscherPhilologen(Wiirzburg, 1868), p. 48, Leo,
RhMn {1885), 161 ff., Die plaut. Cant. pp. 5 ff., Pl. Forsch. 1 , pp. 29 ff., Gesch.

pp. 356 ff., Pasquali, Suma delta tradiziDMe criticadel iestol (Florence, 1952),
pp. 350 ff.
5 One is echoedby Cicero at De orat.3. 183. Cf. Varro ap.Rufin. Gramm.VI
556.7 (on Accius).
51
 INTRODUCTION

Between 169and the period from which the first extant prose
works containing literal quotations come considerable changes
in Latin orthographical convention took place. There are com-
paratively few traces in these quotations of the orthographical
peculiarities which mark early second century public inscrip-
tions and the text of Plautus' comedies offered by the Ambrosian
codex. To what extent modernisations come from the first
philological edition of the tragedies rather than from the heed-
lessness of quoters or the scribes who copied the works of the
quoters cannot now be fully ascertained. 1 One would suppose
that titles were peculiarly liable to modernisation, especially
those of tragedies still performed in the late republican theatre.
I have tried to avoid medieval and renaissance forms in printing
titles and quotations but not to impose any further rationality
on the chaos of the ,rapa6oaas.
During the late second and most, if not all, of the first
century B.C. the tragedies of Ennius and other Latin poets were
read and studied intensively in schools. 2 They were usedas a
quarry by rhetoricians seeking examples of certain types of
argument and figurated speech as well as by grammarians seek-
ing unusual words and anomalous forms of accidence and syn-
tax. Orators quoted famous passages in the Senate and the law
courts.3 Literary men bandied brief, allusive quotations with
each other in private letters4 and introduced quite lengthy
1
Cf. the way quotations of Herodotus in Greek rhetorical works (e.g. the
treatise TTeplUlflOVS)tend to lose their ionicisms.
i Cf. Rhetor inc. Her. 4. 7 on the copying out of Ennian sententiaeand

Pacuvian messenger-speeches; Cicero, De orat. I . 246 on the learning of

Pacuvius' Teucn by heart.
3 C. Titius, 'uir aetatis Lucilianae •, quite certainly alluded in a speech to
Ennius 72-3 (Macrobius, Sat. 3 .13. 13). L. Sempronius Atratinus may, like
Crassus and Cicero (Cael. 18), have alluded to the Medea at the trial of
M. Caelius Rufus in 56 B.c. {Chirius Fortwlatianus 3. 7).
4 Cf., apart from Cicero, Varro, Epist. Iuli Caesarisap. Non. p. 263. 3

(~Cicero, Att. 13 .47. 1). Fam.9.16.4 shows that Paetus quoted from Accius'
Omomausin a letter to Cicero.
52
 ,
HISTORY OF THE TEXT OF ENNIUS TRAGEDIES

quotations into philosophical dialogues.1 Poets made elaborate

imitations not only in the genre of tragedy 1 but also in others,
particularlythat of epic.3
A very considerable amount of what we possess of Ennius'
tragic writing is embedded in works· of this period preserved
either more or less whole, or in epitomes such as that by
Pompei us Festus of the lexicon of Verrius Flaccus, or in quota-
tions by the authors of late antiquity.• Not all thiscomes directly
from whole texts of the tragedies or even from the memory of
the quoters. The practice of borrowing examples from previous
writers rather than collecting them afresh had already begun in
the sphere of teclmical writing.5
1
C£, apart from Cicero, Varro, fepcwro616CXC11<CXAOS ap. Non. p. 261.7.
i On Asinius Pollio see Tacitus, Dial. 21.
3 For Virgil's Aeneid and tragedy sec Maaobius, Sat. 6. 1-3, Norden, P.
VergiliusM.aro:Ameis Buch VI• (Leipzig, 1916), pp. 241, 263 f., 304,370£
4 C£ the quotation of the Alcmeoin the piece of Cicero's Hortensiusquoted

by Priscian (Gramm.u250. 12); of the Me~a in the piece ofVarro's fepoVTo61-

6aalc<XAoS quoted by Nonius (p. 261. 7).
s On the quotations of poetry in the anonymous rhetorical treatise addressed
to Herennius and Cicero's De inuentionesee D. Matthes, Lu.strumm (1958),
81 ft:; in the extant books of Varro's De lingua Latina R. Reitzcnstein, M.
Termtius Varround]ohannaMauropus(Leipzig, 1901), H. Dahlmann, Varround
die hellmistischeSprachtheorie(Berlin, 1932), R. Schroter, 'Studien zur varro-
nischen Etymologie', Abh. Ale.d. Wiss. u. d. Lit. Mainz, Geistes-u. Sozialw. Kl.
1959, Nr. 12; in the lexicon of Vcrrius Flaccus, R. Reitzenstein, Vmianische
Forschungen(Breslau, 1887), Strzelecki, Q_uaestiones Vmianae (Warsaw, 1932).
W. Zillinger, a scholar who made a very detailed study of Cicero's poetic
quotations (Cicero unddie altromischmDichter[Diss.Erlangen, 1911]), writes as
if Cicero always quoted either directly from a whole text or from memory of a
whole text. He notes (p. 84) without offering any explanation that the mode of
quotation employed in the Oratorvaries strikingly. At 164 a famous speech
(Trag. inc. 80-2: probably from Pacuvius' Iliona) is quoted as if it were well
known to the reader: nisiJo,u sic loquipamitet: 'qua tempestateHelenamParis' et
qu« sequuntur;Cicero's writings are full of similar quotations (e.g. Diu. 2.112,
Tusc. 3. 53, 3. 58, etc.). At 155, on the other hand, in a discussion of anomalous
accidence, two vcnes of tragedy (Ennius 44 and Pacuvius 82) metrically com-
plete and yet defective in sense are quoted without apology; this is a mode of
quotation frequently employed by the lexicographer Vcrrius Flaccus (cf.
Vahlen, E.P.R.•, p. LXVII, L. Rychlewska, F.osXL111 (1948/9), fasc. 1, 186 ff.). It

53
 INTRODUCTION

Ennius' Telephusseems to have been read in the school at-

tended by the fable writer Phaedrus(born in 17 B.c.). 1 However
it is doubtful whether, at least in the schools of the capital, this
and other tragedies long survived the attack of such partisans of
modem poetry as Q. Caecilius Epirota and Remmius Palae-
mon. i Early imperial prose and poetry show no sign of the sort
of acquaintance with the scripts of republican tragedy that
grammar-school study might have provided. The fact that
Seneca's quotations are few and mostly, if not all, at second
hand is much more significant than the frequency of his
denunciations.3 The links that can be found between his tragic
scripts and those of the republican poets4 are nowhere near as
extensive as those, for example, between Virgil's Aeneid and
Ennius' Annales and may be due to imitation of Ovid and
Varius, who, whatever formal critical views they acquired,
would have remained deeply affected by what they read at
school.5 The argument of Tacitus' Dialogusshows what conven-

is more likely that Cicero took the two verses from a grammarian'streatise
than that he himself employed deliberately in a literary dialogue a technical
mode of quotation. The quotation of two tragic verses (Trag. inc. 19-4--5)as if
they came from a scene of Terence's Phormioat Orat. 157 indicates even more
plainly Cicero's dependence on a grammatical source; this is a type of error
found repeatedly in authon who take blocks of quotations from their predeces-
son (cf. Str7.Clecki, ~- Vm. pp. 3 ff., on Paulw' epitome of Fcstw'
epitome ofVerriw Flaccw).
1
Cf. 3, cpil. 33-4. It is possible that Phaedrw read the trimctcr he quotes in a
collection of sententiae.On the use of such collections in schools see K. Homa.
RE Suppl. VI (1935), s.vv. Gnome, Gnomendichtung,Gnomologim, 74 ff., J.
Barns, CQ xuv (1950), 126 ff.
a Cf. Suetoniw, De gramm. 16, 23. These grammaticifollowed rather than
created a fashion; c£ Cicero, Tusc. 3 .45, Horace, Epist. 1. 19. 7-8, 2. 1. 50-3,
Ars 259--62, Ovid, Am. 1.15. 19, Trist. 2.259-00.
3 Cf. De ira 3.37.5, Epist. 58.5, ap. Gell. 12.2.2.
4 Cf. G. Carlsson, 'Die Obcrlieferung dcr Scneca-Tragodien', Lumls Univ.
Arsskrift,N.F. Avd. 1, Bd. 21, No . .s (1926), 58 ff. F. Strauss had denied that
there were any (De rationeinterSmecam et antiquasfabulasRDmanasintercedente
[Diss. Rostock, 1887)).
s Asiniw Pollio wrote tragedies in consciow imitation of the republican
poets (sec Tacitw, Dial. 21).
54
 HISTORY OF THE TEXT OF ENNIUS' TRAGEDIES

tional attitudes were like during the reign of the Flavian

emperors. For Curiatus Matemus, defending poetry against the
meers of the practical orator, Ovid's Medea and Varius'
Thyestesare the classics of tragedy; Marcus Aper, defending
modem oratory and attacking that of the ancients (i.e. of those
after Cassius), can bring against the latter as an apparently ir-
refutable charge the fact that they frequently imitated the style
of the republican tragedians. It would be wrong however to
accept Leo' s theory 1 that no one in Rome read the older repub-
lican poetry during the Julio-Claudian and early Flavian periods
and that texts disappeared from libraries both public and private.
The violence with which the fashionable view was expressed
and remarks like that of Seneca at Epist.114. 13-multi ex alieno
saeculopetuntuerba-suggest that some did not follow fashion.2
The fashion of admiring the old republican poets which M.
Valerius Probus was supposed to have introduced3 and which
made the emperor Hadrian express a preference for Ennius'
epic poetry over Virgil' s4 caused the scripts of at least some of
Ennius' tragedies to be sought out and consulted. Fronto's
pupil, Marcus Aurelius, the future emperor, Aulus Gellius (b.
123, educated in Rome)S and Apuleius (b. about 123, educated
in Carthage) seem occasionally to quote from them at first
hand. But while the books of the Annaleswere often read in
second-century schools6 the tragedies probably were not. In
1
See Pl. Forsch.1 pp. 26 ff.
a Tacitus (Ann. 13. 15) describesBritannicus uttering a poem that could have
been the famous canticum from Ennius' Andromaclu,(fr. xxvu).
3 Suetonius, De g,amm. 23. Signs of change can be seen in the tone of Martial
at 11.90.6 and Q_uintilianat Inst. 1.8.8 (conttast 10.1.97). On Probus see N.
Scivoletto, GIF xn (1959), 97 tf., K. Buchner, in H. Hunger et al., Geschichte
tier Textuberliejerungde, anti/em und mittelalterlichcnLiteratur1 (Zurich, 1961),
pp. 335 ff. 4 Spartianus, Hadr. 16.6.

s On Gellius' sources sec L. Merdclin, N]bb Suppl. m (1857-60), 633 ff., C.

Hosius, A. Gellii NoctiumAtticarumLibri XX, vol. 1 (Leipzig, 1903), pp. XVIff.
6
Cf. Gellius 16.10, 18.5, 20.10.2.

SS
 INTRODUCTION.

general discussions of archaic poetry preserved from this

century Ennius always figures as a writer of epic, never of
tragedy; Pacuvius and Accius are usually named as the great
tragedians of the Republic. I
The only person after Apuleius who can be shown with prob-
ability to have handled a roll or codex containing an Ennian
tragedy is the early fourth century grammarian Nonius
Marcellus. The Telephusand the Hectorislytraappear to have
been among those republican poems which he excerpted him-
sel£i At some time in the fifth or the sixth century a reader of
Orosius' Historiaeconsulted a copy of the seventh book of the
Annalesin order to gloss the historian's text.3 However, to judge
by Macrobius' remarks about the literary tastes of his contem-
poraries,4 not even this work could have found a great many
readers. The quotations of the tragedies which appear in works
composed after Nonius' lexicon are probably all second or third
hand. This cannot be indubitably demonstrated in every case
but the frequency with which the authors of late antiquity
admit to using older material5 and the number of coincidences
both between blocks of quotations in early and late technical
1
Sec Fronto, p. 131. 13, Gellius 13 . 2. 1, Diomcdes, Gramm. 1 490. 12 ff.
(Suetonius ?). This had long been the orthodox view (cf. Ovid, Am. 1. 1s.19,
Trist. 2.359, Velleius 1.17.1, 2.9.3, Columella I praef. 30, Penius 1.76-8,
Martial 11 . 90. 6, Tacitus, Dial. 20, Qyintilian, Inst. 10. 1 . 97) and was perhaps
as old as Horace, Epist. 2. 1. ss-6. Horace nevertheless thought Ennius'
dramatic verse worthy of assault. Cicero joined Ennius with Pacuvius and
Accius to form a classical tragic trio (De oral. 3 .27, Oral. 36, Opt. gen. 18,
Ac. 1. 10, Fin. 1 .4).
a Cf. Lindsay, Nonius Marcellus'Dictionaryof RepublicanLatin (Oxford,
1901); for further analysis of the structure of Nonius' work along the same
lines see Lindsay, PhilologusLXIV( 1905), 438 ff., Strzelecki, F.osXXXIV(1932-
3), n3 ff., De Flauio Capro Nonii Auctore (Krak6w, 1936), Rychlewska, in
TragicaII, pp. 117 ff.
3 See Norden, EnniusundV ergilius:Kriegsbilder ausRomsgrosserZeit (Leipzig,
1915), pp. 78 ff. 4 See Sat. 1.4.17, 6.1.2, 6.1.5, 6.3.9, 6.9.9.

S Cf., in particular, Priscian, Gramm.m 418. 10 {at the beginning of a work

which contains some of the longest quotations of republican tragedy extant).
 HISTORY OF THE TEXT OF ENNIUS' TRAGEDIES

writing and between arguments illustrated by particular quota-

tions of republican poetry in early and late literary prose1
should compel scholars to be at least hesitant in postulating a
direct connection between any tragic fragment preserved in a
work of late antiquity and a whole tragic text.
From this point the history of the text of Ennius' tragedies is
coincident with those of the many works which carry quota-
tions from them. Our evidence for some quotations is com-
paratively old (e.g. the fourth-century palimpsest of Cicero's
De republica),for others recent (e.g. the book printed at Basle in
1521 containing Rufinian's De.figurissententiarum et elocutionis),
for some copious and good, for others (particularly lexico-
graphical quotations) bad. I have found it necessary to express
doubt concerning, or disbeliefin, the irapa6oats at many points.
Sometimes the ancient quoter had a faulty text 2 or misread3 or
misunderstood an accurate text, 4 but most errors must come
from the inattention or incompetence of medieval scribes.
Lack of knowledge of the original context often prevented
them from reading obscure groups of letters correctly and con-
tinues to prevent modem scholars from correcting their errors
in a manner likely to convince others.
1
On Rnfinian see Marx, BPhW x (1890), 1008, A. Gantz, De Aquilae
Romani et Iulii Rufinumi Exemplis (Diss. Konigsberg, 1909), W. Schaefer,
Q!aestiDMsRhetoricae(Diss. Bonn. 1913); on Charisius and Diomedes, K.
Barwick, RemmiusPalaemonund dieriimislheArs Grammatica(PhilologusSuppl.
xv 2, 1922); on St Jerome, C. Kunst, De S.HieronymiStudiisCiceronianis(Diss.
Vienna, 1918), p. 142; on the Ciccronian scholiasts, P. Hildebrandt, De Scholiis
CiceronisBobiensibus(Diss. Gottingen, 1894), pp. 34 ff.; on Macrobius and the
Virgilian scholiasts, H. D. Jocelyn, CQ N.S. XIV (1964), 280 ff., xv (1965),
126 ff.; on Priscian. L. Jeep, PhilologusLXVn (1908), 12 ff., LXVm (1909), 1 ff.
3
Cf. the reading of v. 209 caesaeatddissentabiegnae. .. trabes (sec Com-
mentdry).
3 Cf. Nonius' placing of fr. LXXDI under the lemma VAGAS.
4 Cf. Vcrrius' placing of fr. xcvm under the lemma PEDVM, baculum.Similar
crron may have been made by Cicero or his source concerning the exitium of
v. 44 (Orat. ISS) and by Nonius' source, Lindsay's list 27 'Alph. Verb', con-
cerning the reg,edereof v. 7 (p. 166.21).
57
 INTRODUCTION

VII. THE TITLES OF ENNIUS' TRAGEDIES

Most of the titles which head the quotations made by the
ancient grammarians consist of proper names-Achilles, Aiax,
Alcmeo,Alexander,Andromacha,Andromeda,Athamas,Cresphon-
tes, Erectheus,Hecuba, Iphigenia, Medea, Melanippa, Nemea,
Phoenix, Telamo,Telephus,Thyestes.One title contains a Latin
adjective-Mede a exul; one the name of the author of the Attic
original-Achilles Aristarchi;three contain Greek words not
domiciled in second-century Latin-Andromache aechmalotis,
Eumenides,Hectorislytra (Lytra). Some of the spellings used by
the grammarians could not possibly come from Ennius himself
and there is reason to suppose that some of the very titles are
equally late. Certain modem scholars have inferred from the
titles AchillesAristarchi,Andromacheaechmalotis and Medeaexul1
that the grammarians quote from two Ennian plays about
Achilles, from two about Andromache and from two about
Medea. The substance of the grammarians' quotations supports
strongly the idea of two Medea plays but leaves the issue open
in the case of Achillesand Andromacha. i

Andromacheaechmalotis should be considered along with the

title Aiax mastigophorus which Nonius appears to attribute to
Livius Andronicus at p. 207. 32, the title PhiloctetesLemnius
which Varro attributes to Accius at Ling. 7. I I and the titles
Phasmaand Synaristosaeunder which pieces of the comedies of
Plautus otherwise known as Mostellariaand Cistellariarespec-
tively are quoted in Verrius Flaccus' lexicon.3 What has hap-
1
Stephanus' notion that Hectorislytra and Lytra denote distinct plays has
never fowid favour (FragmentaPoetarumVtterumLAtinorum[Geneva, 1564)).
a Sec Commentary,pp. 161 ff., 234 ff.
3 Cf. Fcstus, pp. 158.33, 394.18 for Phasttur (contrast p. 166. 19); pp. 390.8,
480. 23 (for new evidence on the text sec Fracnkel, PhilologusLXXXVn [1932],
II7 ff. [ = Kl. Beitr. u 33 ff.)) for Synaristosae(contrast p. 512. 10).

58
 THE TITLES OF ENNius' TRAGEDIES

pened in each case is that some grammarian, not necessarily

Nonius or Verrius, headed his quotation with the title of the
Latin poet's Attic original taken from the 616aO"KCXA(cn of the
edition he consulted. Sheer caprice would have been his motive,
not a desire to distinguish one play from another or to indicate
titles actually usedin the theatre for revival productions. 1
The mode of quotation used at Festus p. 282. ~Ennius in
Achille Aristarchi-has no exact parallel in the remains of Latin
lexicography. Nevertheless its similarity with the mode of
reference used in the prologue of Plautus' Poenulus 2
should be
treated as accidental. P. Scriverius3 thought that the quoter was
trying to distinguish the play he was quoting from a second
play about Achilles by Ennius; Bergk4 from a play about
Achilles by another Latin poet. I suggest a more humdrum
explanation: the quoter, not necessarily Verrius Flaccus in the
first instance,Swas simply looking at or recalling the 616aO'l<CXAfa1
at the beginning of the roll instead of the Latin title written at
the end and on the aiAAvf3os.
The titles Eumenidesand Hectorislytrawould have been in-
comprehensible to those members of Ennius' first audiences
who were ignorant of literary Greek. They do not appear out-
side the works of the grammarians Verrius, Nonius and
1
Plautus, Cas. 32 may simply give a translation ofDiphilw' title I0.11pov-
there is no need to suppose that the revival performance for which some
J.IEV01;
learned penon composed the extant prologue was not advertised with the title
Casuui. Osann, Anal. crit. p. 164, lumped Sortientesand Phasmatogether as
revival titles. Ritschl, Parerga,pp. 159, 165, 206, wr.ote more carefully; he
treated Sortiffltesas a revival title and Phasmaas a title used either for learned
purposes or for stage revivals. Many later scholan (e.g. A. 0. F. Lorenz,
Plautus: Mostellaria[Berlin, 1866], pp. 2 f., Lindsay, The Ancient Editionsof
Plautus [Oxford, 1904], p. 1 n.) have adopted Osann's view of the Festus
quotations alleging it to be Ritschl's.
3
Sec above, p. 7.
3 ColledaneaVeterumTragicorum Fragmmta(Leiden, 1620), p. 8.
4 Ind. lectt. Marburg18-44, XI ( = Kl. phil. Sehr.I 225).
S Sec Commentary,p. 161 n. 1.

59
 INTRODUCTION

Diomedes. It is possible that in the early second century the

plays were advertised with other titles and that these titles were
either unknown to 1 or ignored by the grammarians who used
the plays as a quarry for wiusual Latinity. The twenty-one
comedies which ancient scholars agreed to be genuinely
Plautine all had titles attaching to them which could be under-
stood by speakers ofLatin, some proper names (Amphitruo,etc.),
some straight translations of the Attic title (Mercator~ •Eµiro-
pos,etc.), some of a native Latin type (Cistellaria~ Iwap1a-roo-
aa1, etc.). 2 Terence retitled his version of Apollodorus'
'E,r161K036µevos as Phormiobut allowed his other five comedies
to be advertised with transliterations of their Attic titles
although these must have been incomprehensible to many
spectators. It is reasonable to suppose that Terence was follow-
ing a new fashion in tune with the growing Hellenisation of the
Roman stage and to suspect that all titles attaching to comedies
and tragedies of earlier times which require a knowledge of
Greek for their comprehension may have been bestowed by
students of the Latin scripts rather than their authors.3
1
Noniw seems to have possessed a roll containing the Hectorislytra; see
Commentary,p. 290.
a The fabula -aruz type of title continued to be used in the next century for
freshly composed Atellane farces.
3 The small number of purely Greek titles from the third and early second
century and of purely Latin titles from the late second was discussed by Osann,
Atuil. ait. pp. 161 ff., and Ritschl, Parerga,pp. 139 ff. Ritschl argued that
theatrical practice fluctuated until Plautw established the fashion of wing Latin
titles. This is historically implawible and the evidence can be interpreted more
economically in the way I have suggested. The trimeter Acontizomenus f abula
estprimeprobaquoted by Charisius, p. 273 . 1 I ff. may come from a prologue
written for a revival performance of Naeviw' version of an• AKoVTtlOIJEVOS (c£
Plautw, Cas. 31). On the other hand the practice of Caeciliw and Pacuviw, like
that of Terence, may well have fluctuated. The title HypobolimaeusR.astraria
which Noniw attributes to Caecilius on a number of occasions (pp. 16. 17, 40. 3,
89. 14,147.6, 176.6, sos.29) looks like a conflation of an old-style title, which
must have been Caeciliw' own, and a transliteration of the title of the Attic
original. Since Cicero quotes the titles Niptra (Tusc. 2.48) and Synephebi(Fin.
1.4, Opt. gen. 18, Nat. de"1.3. 72, Cato 24, Tusc. 1. 31) and since their use gave
6o
 THE TITLES OF ENNIUS' TRAGEDIES

The title Medeaexul, like the title Hedorprofidscensattributed

to Naevius by Priscian at Gramm.n 400. 1, could in principle
be interpreted as a mere translation of a Greek. title (e.g.
M1'}&1acpevyovaa)made either by Ennius himself or by a gram-
marian.1 Greek grammarians commonly invented such titles
when they wished to distinguish two scripts by the one poet
about the one hero or heroine. 2 However the tragic verses
quoted along with Medeaexul clearly belong to the version of
the extant Euripidean M{)6e1awhich is elsewhere quoted with
the title Medea.3Among the seventy tragedies of Euripides
known to the scholars of Alexandria4 there was one apart from
the extant M{)&1a which had the Colchianwoman as a per-
sonage. But this is regularly quoted in our sources as the
Alyevs. We should therefore suppose that Medea exul is a title
of the type of Parasituspiger and Parasitusmedicus(Plautus),
HerculesJurensand HerculesOetaeus(Seneca), and was applied
by some grammarian to Ennius' adaptation of Euripides'
M{)&1a in order to distinguish it from another tragedy by
Ennius about Medea.
A large number of fragments of tragic verse are quoted by
Cicero and Varro without any m•ention of the title of the par-

him stylistic pain (cf. Tusc. 3 .65 ille Terentianusipse se poeniens,id est lcxvrov
,iµc..,povµevoS)we may suppose that no othen were known for Pacuvius'
version of Sophocles' Nfn,-pa and Caecilius' venion of Menander's Iwicp11~01
in the middle of the first century. Whether these plays ever possessed Latin
titles is now anybody's guess.
1
The titles Famaator and Subditiuuswhich Nonius (pp. 150.2, 543 .23,
20.4.33, 514.31) attributes to Caecilius could be grammarians' translations of
Greek titles. Nonius quotes a Hypobolimaeusat p. 178. 16 (for Hypobolinuieus
Rast,aru, sec above, p. 6o n. 3) and an Obolostatesat pp. 98 .6, 154. 10,277.33,
279. 40, 508 • 12.
a The Soudagives an Afas µoov6'1WoS to Astydamas, an •Ax1"AMvs 8epo-tTO-
KTOVOS to Chaeremon and a 'H~fis mp1Ka16µev05to Spintharos. The
absence of companions for these three plays in our sources is due to the rarity
with which the three tragedians are quoted.
3 See Commentary,p. 342. 4 See above, p. 45.

61
 INTRODUCTION

ticular tragedy from which they come or their author. Some-

times, however, the hero or heroine who spoke them is named.
The first editor of the Ennian fragments, Stephanus, printed
under the extant titles only those fragments whose quoters
made a specific assignation. Columna and subsequent editors
have with varying degrees of confidence attempted to assign
for themselves the fragments which are quoted without assigna-
tion. They have assumed that Cicero and Varro knew at first
hand all the plays which are named by Verrius, Nonius, Pris-
cian and other grammarians and that they knew no play outside
those named in our grammatical sources. These assumptions arc
not necessarily valid. Where Ennius is concerned there is no
sign at all that Cicero knew the Andromeda,Nemeaor Phoenix,
few that he knew the Athamas,Erectheus,or Hecuba.On the
other hand he quotes no tragic verse as by Ennius which cannot
be easily assigned to one of the known titles. Varro quotes two
fragments as by Ennius which require textual emendation of a
very unconvincing kind to make them fit any of the twenty-
two plots to which we can attach titles.
At Ling. s.14 (fr. CLXXXVI)Varro quotes a fragment-o terra
Thraeca(Fleckeisen: trecaF} ubi Liberifanum inclutumI Maro
locaui-which would appear to the unprejudiced observer to
come from the prologue of a play set in Thrace which was
spoken by Maro the son of Dionysus (Liber ). Leo 1 compared
the opening verses of Euripides' 'HMK-rpa: & yiis 1r<XAatov
.,Apy<>s,'lvaxov ~{, 66ev... (,rAEUae,and• AAKflcrTtS: & 600'1CXT 1

•A6'11'\TEt'wols h-AflVlyoo••.• Ribbeck 2 altered locauito locauit

and interpreted the words as part of an apostrophe to the birth-
place of Eumolpus in the Erectheus.V ahlen3 gave them to the
Polymestor of the Hecuba.
1
Gesch.p. 187 n. 2. Cf. Bergk, Philologus:xxxm (1874), 291 ( = Kl. phi!.
Sehr. I 358).
a Qy_ust.san. p. 262. 3 E.P.R. 1 p. 194.

62
 THE TITLES OF ENNIUS ' TRAGEDIES

The fragment quoted by Varro at Ling. 7 .13 (fr. cxc)-

extemploacceptummenecato(Scaliger: negatoF) etfilium-can be
got into the Andromachawith Vossius' change of acceptumto
acceptam 1
or into the Hecubawith F. H. Bothe's further change
of .filium to filiam.1
In assigning the fragments I have followed the method of
Columna rather than that of Stephanus; more for the con-
venience of discussion thanout of any confidence in my own
ability to determine with certainty the plays which Cicero and
Varro quoted on particular occasions. Nevertheless some
assignations are more probable than others and I have tried to
render explicit the reasoning behind those which I make.
Titles are sometimes omitted from the quotations of Festus,
Nonius and Diomedes. In these cases it is unmethodical to
assign the quotations to plays which are never formally quoted
by these grammarians. The extant commentaries on Terence,
Virgil, Horace and Cicero rarely give a title when they refer to
Ennius. Where they do it is always a title we know from other
sources. I am accordingly sceptical about O. Skutsch's sugges-
tion3 that Terence quotes the first verse of an Alcumenaby
Ennius at Eun. 590-qui templacaelisummasonituconcutit4-and
that Donatus (see Ennius, Trag.fr. cuab) recognised the con-
nection. In any case, while it seems reasonably certain that
at Rud.
Plautus refers to a Latin version of Euripides' •AAKµf1V11
86,5 there are no good grounds for thinking Ennius the only
person to have written tragedy at Rome in the early decades of
the second century and thus necessarilythe author of the version.
1
Castigatwneset Notae, p. 17 (in Scriverius, Collectanea).
i F. H. Bothe, PoetaeStenid Latinorum,vol. v (Leipzig, 1834), p. 48, com-
pared Euripides, Hek. 391 vv,eis6i µ' a>J..a
8vya-rpl avµcpovevacrre.Columna's
discussion at Q. .EtuaiiFrag.pp. 362-3 foreshadowed both suggestions.
3 HSCPh LXXI (1967), 128.
4 Cf. Plautus, Rud. 1 quigentesomnesmariaqueet tmas mouet.
s See above, p. 6.
 THE FRAGMENTS

5 JTO
Testes cuiusque tragoediae ratione fere temporum
ordinaui. testium codices siglis notaui usitatis.
 ACHILLES
I
Plautus, Poen. 1 :
Achillcm AristarchimihicoQll]lcntari lul>et;
inde ·mihi principium capiam, ex ea tragoedia.
sileteque et tacete atque animum aduortite.
audire iubet uos imperator histricus ·(histrycus codJ.),
s bonoque ut animo sedeate in subselliis
et qui esurientes (D': esuplentes BCD1) et qui saturi
uenerint.
qui edistis multo fecistis sapientius;
qui non edistis saturi fite fabulis.
nam cui paratumst quod edit, nostra gratia
10 nimia est stultitia sessum inpransum incedere.
exsurge praeco, fac populo audientiam.
iam dudum exspecto si tuom officium scias;
exerce uocem quam per uiuisque et colis.
nam nisi -clamabis tacitum ( Turnebus: ta titum B: statim
CD) te obrepet fames.
1 s age nunc reside duplicem ut mercedem feras.
bonum factum, tesset edicta ut seruetis mea.
scortum exoletum ne quis in proscaenio
sedeat, neu (Camerarius:niue codJ.)lictor uerbum aut
uirgae muttiant,
neu dissignator praeter os obambulet,
20 neu sessum ducat dum histrio in scaena (scena codJ.)siet.
diu qui domi otiosi dormierunt, decet
animo aequo nunc stent uel dormire temperent.
serui ne obsideant, liberis ut sit locus,
uel aes pro capite dent; si id facere non queunt

67 5-2
 THE FRAGMENTS

25 domum abeant, uitent aneipiti infortunio,

ne et hie uarientur uirgis et loris domi
si minus eurassint (cura sint codd.)tquom eri ueniant
domumt.
nutriees pueros infantisminutulos
domi ut procurent, neu quae spectatum adferat
30 ne et ipsae sitiant et pueri pereant (CD: pertant B:
-peritent T) fame
neue esurientes hie quasi haedi obuagiant.
-·matronae tacitac spectcnt, tacitae rideant (redcant C:
rediant B: reddeant D),
canora hie uoce sua tinnire temperent,
domum sermones fabulandi eonferant
35 ne et hie uiris sint et domi molestiae.
quodque ad ludorum curatores attinet,
ne palma dctur quoiquam artifiei iniuria,
neue ambitionis eaussa extrudantur foras
quo deteriores anteponantur bonis.
40 et hoe quoque etiam quod paene oblitus fui:
dum ludi fiunt in popinam pedisequi
inruptionem facite; nune dum oeeasio est,
nune dum scribilitae aestuant oeeurrite.
haec quae imperata sunt pro imperio histrieo,
45 bonum hercle faetum, prose quisque ut meminerit.
ad argumentum none uieissatim uolo
remigrare ut aeque meeum sitis gnarures (T: siti signa
rures BCD);
eius nune regiones limites eonfinia
determinabo; ei (codd.Plauti:eius codd.Non. p. II .25)
rei ego sum factus finitor.

68
 ACHILLES

II
(a} Cicero, Verr. 2. 1.46: Delum uenit. ibi ex fano Apollinis
religiosissimo noctu clam sustulit signa pulcherrima atque
antiquissima, eaque in onerariam nauem· suam conicienda cura-
uit. postridie cum fanum spoliatum uiderent ii qui Delum
incolebant, grauiter ferebant; est enim tanta apud eos eius fani
religio atque antiquitas ut in eo loco ipsum Apollinem natum
esse arbitrentur. uerbum tamen facere non audebant, ne forte ea
res ad Dolabellam ipsum pertineret. tum subito tempestates
coortae sunt maximae, iudices, ut non modo proficisci cum
cuperet Dolabella non posset sed uix in oppido consisteret.
ita magnifluctus eidebantur. I

hie nauis illa praedonis istius, onusta signis religiosis, expulsa

atque eiecta fluctu frangitur; in litore signa illa Apollinis
reperiuntur; iussu Dolabellae reponuntur. tempestas sedatur,
Dolabella Delo proficiscitur.
(b) Schol. Gronouianus: ITA MAGNI FLVCTVS EICIBBANTVR.
Enniano hemistichio usus est ex ea tragoedia quae Achilles
inscribitur.
III
Festus, p. 282. 9:
prolatoaereastitit, 2

Ennius in Achille Aristarchi cum ait, significat clipeo ante se

protento.
IV
(a} Festus, p. 394. 33 : SVBICBS Ennius in Achille pro subiectis
posuit cum dixit nubes:
per egodeumsublimassubices 3
umidasundeoriturimbersonitusaeuoet spiritu.
m in Achille Sailiger:achillacin F
IV (a) 3 deum subiccs F 4 indc oritur imbcr sonitus acuo spiritu F
69
 THE FRAGMENTS

(b) Gellius 4.17.13: congruens igitur est ut subices etiam,

quod proinde ut obices compositum est, u littera breui dici
oporteat.
14: Ennius in tragoedia quae Achilles inscribitur subices pro
aere alto ponit, qui caelo subiectus est, in his uersibus: 'per ego
deum sublimas subices humidas unde oritur imber sonitu saeuo
et strcpitu'. plerosque omnes tamen legere audias u littera
producta.
(c) Nonius, p. 169. 1: SVBICES noue positum; non a subiciendo
sed altitudine. Ennius Achille (acille codd.):'per ego deum sub-
limas subicis umidas unde oritur imber'.

V
Nonius, p. 147. 18: OBVARARE, peruertere, deprauare; dictum a
uaris. Ennius Achille:
tnam consiliust obuarantquibustam concedithie ordo. s
VI
Nonius, p. 166. 20: REGRBDBRB, reuocare. Ennius Achille:
quonuncincertare atqueinoratagradum 6
regredere
conare?
VII
Nonius, p. 277.23: DEFBNDBRE, ttuerit depellere (debellare
ueri AA). Ennius Achille:
seruadues, dejendehostescumpotesdejendere. 8

VIII
Nonius, p. 472.26: PROWANT. Ennius Achille:
tmtatmortalesintersesepugnantproeliant. 9

V acille codd.:s nam consiliis ius Timpanaro

VD in achille Gen.: in chille (achillco lJI) que te B
VIII 9 inta mortalcs J..BA04 : ita mortalcs A.d: mortales interea Klussnum,i:
interca mortalcs Laclunann
 AIAX

IX
Isidorus, Di.ff 1.218: inter famam et gloriam: gloria quippe
uirtutum est, famauero uitiorum. Ennius in Achille:
summamtu tibipro malauitafamam extolles 10

et pro bonaparatamgloriam.
maleuolentes[enim] famam tollunt,beneuolentesgloriam.

AIAX

X
Cicero, Off. 1 . 114: suum quisque 1g1tur noscat ingenium
acremque se et bonorum et uitiorum suorum iudicem prae-
beat, ne scaenici plus quam nos uideantur habere prudcntiae.
illi enim non optumas sed sibi accommodatissimas fabulas
eligunt: qui uoce freti sunt Epigonos Medumque (Medeamque
X), qui gestu Melanippam (melenippam B: menalippam bX),
Clytemestram (clitemestram Lp: clitimestram c), semper Rupi-
lius quem ego memini Antiopam (Anthiopam Z), non saepe
Aesopus Aiacem.
XI
(a) Varro, Ling. 6.6: cum stella prima exorta ... id tempus
dictum a Graecis Acrn-ipa,Latine uesper; ut ante solem ortum
quod eadem stella uocatur iubar, quod iubata. Pacui dicit
pastor ... Enni Aiax:
lumeniubarnein caelocerno? 13

(b) Varro, Ling. 6. 81: cemo idem ualet. itaque pro uideo ait
Ennius: 'lumen iubarne in caelo cemo?'

IX 12 enim tkl. R.ibbtdt

XI (a) Ennii l..Mtus: cnnius F

71
 THE FRAGMENTS

(c) Varro, Ling. 7. 75: possunt triones dicti, vn quod ita sitae
stellae ut temae trigona faciant. taliquod t
76: 'lumen iubame in caelo cemo ?' iubar dicitur stella lucifer
tquaet in summo quod habet lumen d.iffusumut leo in capite
iubam. huius ortus significat circiter esse extremam noctem.
itaque ait Pacuius ...
XII
Festus,p. 48.2.3: TVWOS al)ti dixerunt essesilanos, alii (ali F)
riuos, alii (ali F) uehementes proiectiones sanguinis arcuatim
fluentis quales sunt Tiburi in Aniene. Ennius in Aiace:
ta iaxt missosanguinetepidotullii e.Jllantes
uolant. 14

XIII
Nonius, p. 393. 7: STATIM producta prima syllaba, a stando,
perseueranter et aequaliter significat ... Ennius Aiace:
qui rem cumAchiuisgesseruntstatim. IS

ALCMEO
XIV
(a) Cicero, De orat.3. 154: nouantur autem uerba quae ah eo
qui dicit ipso gignuntur ac fiunt uel coniungendis uerbis ut
haec: 'tum pauor sapientiam omnem (mihi add.M: sapientiam
mihi omnem 'pars integrorum'[Vahlen];cod. G Non. p. 16.7:
omnem sapientiam mihi cod.L Non.) exanimato (exanimo B)
expectorat (expectarat A) (v. 17)'; 'num non uis huius me
uersutiloquas malitias'. uidetis enim et 'uersutiloquas' et
'expectorat' ex coniunctione facta esse uerba, non nata. sed
saepe uel sine coniunctione ...

XII cnnius inaiacea. iax F 14 tulii F: tullii codd.Pauli

XDI Is achibidis AA

72
 ALCMEO

(b) Cicero, De orat.3 .217: aliud enim uocis genus iracundia

sibi sumat, acutum ... aliud miseratio ac maeror, Bexibile ...
218: aliud metus, demissum et haesitans et abiectum:
multissum modisdrcumuentus,morboexilioatqueinopia. 16
tum pauorsapientiamomnemexanimatoexpectorat.
taitert terribilemminaturuitaecruciatum
et necem;
quaenemoest tam.fimw ingenioet tantacon.fidentia
quin refugiattimidosanguenatqueexalbescatmetu. 20

219: aliud uis, contentum ...

(c) Cicero, Tusc.4.19: quae autem subiecta sunt sub metum,
ea sic definiunt: pigritiam ... terrorem ... timorem ... pauorem
metum mentem loco moucntem, ex quo illud Emu (vncM:
Ennius X) 'tum pauor sapientiam omnem exanimato (omnem
mihi ex anima codd.)expectorat (K2: expectaret X: expectoret
B: expelleret vr•c) (v.17)'; exanimationem ... conturbationem
... formidinem ...
(J) Cicero, Fin. 4.62: hoe uero te ferre non potuisse, quod
antiqui illi quasi barbati, ut nos de nostris solemus dicere,
crediderint (crediderunt RNV), eius, qui honeste uiueret, si
idem etiam bene ualeret, bene audiret, copiosus esset, opta-
biliorem fore uitam melioremque et magis expetendam quam
illius, qui aeque uir bonus multis modis esset, ut Emu Alcmeo,
'circumuentus morbo exilio atque inopia (v. 16)'.
(e) Cicero, Fin. 5.31: quis est enim aut quotus quisque (est
add.codd.Non. p. 224.18), cui, mors cum adpropinquet, non
'refugiat timido sanguen {refugiat timidos anguis BERNI:
refugiat timido sanguis N2V: fugiat timido sanguen codd.Non.)
atque exalbescat metu (v. 20)'.
(f) Cicero, Hortens.frag. ap. Prise. Gramm.II 250. 12: ut ait
Eruuus, 'refugiat timido sanguen atque exalbescat metu (v. 20)'.
XIV (b) 16 modis sum L 17 omnem mi P: omnem mihi VO 18 alter
L: om. M minitatur L 20 sanguineL

73
 THE FRAGMENTS

xv
(a) Cicero, Ac. 2. 52: illud enim dicimus non eandem esse uim
neque integritatem dormientium et uigilantium nee mente nee
sensu ... quod idem contingit insanis, ut et incipientes furere
sentiant et dicant aliquid quod non sit id uideri sibi et cum
relaxentur sentiant atque illa dicant Alcmeonis:
sedmihi neutiquamcorconsentitcumoculorumaspectu. 21

(b) Cicero, Ac. 2. 88: dormientium et uinulentorum et furio-

sorum uisa inbeeilliora essedicebas quam uigilantium siccorum
sanorum. quo modo? quia cum cxpcrrectus esset Ennius non
diceret se uidisse Homerum sed uisum esse,Alcmeo autem: 'sed
mihi neutiquam cor consentit- (v. 2 I)'. similia de uinulentis ...
89: quid loquar de insanis?... quid ipse Alcmeo tuus, qui negat
cor sibi cum oculis consentire, nonne ibidem incitato furore:
untiehaecjlammaoritur? 22

et illa deinceps:
tincede incedet adsunt;me expetunt.
quid cum uirginis £idem implorat:
fer mi auxilium,pestemabigea me,
jlammiferamhancuim quaeme excruciat. 25

caeruleae
incinctaeigni incedunt,
drcumstantcumardentibus taedis.
num dubitas quin sibi haec uidere uideatur? itemque cetera:
intenditcrinitusApollo
arcumauratumluna innixus;
Dianaf acemiadt a laeua. 30

qui magis haec crederet si essent quam credebat quia uide-

bantur ? apparet enim iam cor cum oculis consentirc.

XV (b) 23 incacde incacde V 26 ceruleae (cerulaeae .B)incincte igni codd.:

caeruleo incinctae angui Colutnn4
74
 ALEXANDER

(c) Festus, p. 162. 14:

NBVTIQV AM pro
cum ait 'sed
lorum aspect (v. 21)
neutiquam
XVI
Nonius, p. 127.13: IAMDIV pro~~·-· .Ennius Alcmeone:
f actumest iam.diu. 31

ALEXANDER
XVII
(a) Cicero, Att. 8 . 11 . 3 : uoluisti enim me quid de his malis
sentirem ostendere. npo8ecnr(300igitur, noster Attice, non
hariolans ut illa (Pius:utilia codd.)cui ncmo credidit sed coniec-
tura prospiciens: 'iamque marl (maria codd.)magno-(v. 43)'.
non multo, inquam, secus possum uaticinari. tanta malorum
impendet •1A1QS.
(b) Cicero, Orat. 155: atque etiam a quibusdam sero iam emen-
datur antiquitas, qui haec reprehendunt. nam pro deum atque
hominum £idem deorum aiunt. ita credo hoe illi nesciebant. an
dabat hanc licentiam consuetudo? itaque idem poeta qui
inusitatius contraxerat 'patris mei meum factum pudet (v. 37)'
pro meorum factorum, et 'texitur exitium examen rapit
(v. 44)' pro exitiorum, non dicit liberum ut plerique loquimur
... at ille alter in Chryse ...
156: ... atqui (uulgo:et quid Heerdegen:quid A: et qui L)
dixit Accius ...
(c) Cicero, Diu. I .66: inest igitur in animispraesagitio (uulgo:
praesagatio codd.)extrinsecus iniecta atque inclusa diuinitus. ea

XV (c) NEVTIQVAM pro nullo modo - Paulus

75
 THE FRAGMENTS

si exarsit acrius furor appellatur cum a corpore animus abstrac-

tus diuino instinctu concitatur.

sedquidoculisrapereuisa est derepenteardentibus? 32

ubi illapauloantesapienstuirginalit modestia?

mater,optumatummultomuliermeliormulierum,
missasum superstitiosis
hariolationibus; 3s
tnequet me Apollofatisfandis detnenttminuitamciet.
uirginesuereoraequalis,patrismei meumf actumpudet,
optumiuiri. meamater,tui me miseret,meipiget.
optumamprogeniemPriamopeperistiextrame. hoedolet:
men obesse,illosprodesse,me obstare,illosobsequi. 40

o poema tenerum et moratum atque molle. sed hoe minus ad

rem.
67: illud quod uolumus expressum est ut uaticinari furor uera
soleat.
adestadestJax obuolutasanguineatqueincendio. 41
muhosannoslatuit.duesferte opemet restinguite.
deus (restinguit deus AV) inclusus corpore humano iam, non
Cassandra loquitur.
iamquemarimagnoclassiscita 43
texitur.exitiumexamenrapit.
adueniet.fora ueliuolantibus
nauibuscompleuitmanus.litora.
tragoedias loqui uideor et fabulas.

XVD (c) 32 rabcrc lAmbimu 1573 33 (aut} ubi L«hmann uirgina1cB1

34 optumatum uulgo: optumarum Porson:optuma tum AVB 36 mcque
Grotius:namquc me Hottingn: namquc Ribbede 37 ucrcor Ribbecle:ucrOIIC*
(o tx c corr.)B: ucro AV 39 pcpcristi Marsus:rcppcristi AVB 40 mCllc
B 42 rcfcrtc AV 4S aducnit et fcra VS 46 complebit P
 ALEXANDER

{d) Cicero, Diu. multos nemora siluaeque, multos

1.114:
amnes aut maria commouent, quorum furibunda mens uidet
ante multo quae sint futura. quo de genere ilia sunt:
eheuuidete: 47
iudicauitinclitumiudidum inter Jeas tris aliquis,
quo iudido LacedaemoniamulierFuriarumuna adueniet.
eodem enim modo multa a uaticinantibus saepe praedicta sunt.
(e) Cicero, Diu. 2.112: at multi saepe uera uaticinati, ut Cas-
sandra: 'iamque mari magno-( v. 43 )' eademque paulo {populo
V) post: 'eheu uidete (v. 47)'.
113 : num igitur me cogis etiam fabulis credere? quae delec-
tationis (ed. Veneta1471: delectationes AVB) habeant quantum
uoles, uerbis sententiis numeris cantibus adiuuentur; auctori-
tatem quidem nullam debemus nee £idem commenticiis rebus
adiungere.
XVIII
Cicero, Diu, 1 . 42: haec etiamsi ficta sunt a poeta non absunt
tamen a consuetudine somniorum. sit sane etiam illud com-
menticium quo Priamus est conturbatus quia
matergrauidaparerese ardentemf acem so
uisa est in somnisHecuba.quofacto pater
rex ipse Priamussomniomentismetu
perculsuscurlssumptussuspirantibus

XVIl "'Trag. inc. ap. Q!!intil. Inst. 9. 3 . 77 Hecuba hoe dolet pudet
piget, Varro, Men. frag. ap. Non. pp. 112.21, 328 .28 adest fax
inuoluta incendio (incendii coJJ. Non. p. 328.29), Vergilius, Aen.
2. 569-7 4 Tyndarida aspicio ... illa ... Troiae et patriae communis
Erinys Iabdiderat sese; ex Vergilio Lucanus 10. 59
XVIlI *Ouidius, Epist. 17.237-8 fax quoque me terret quam se
peperisse cruentam I ante diem partus est tua uisa parens.
XVII (4) .f,8 inter add.W: intus VI>
XVIII so se om. H

77
 THE FRAGMENTS

exsacrificabat
hostiisbalantibus.
tum coniecturam postulatpacempetens, SS
ut se edoceretobsecrans
Apollinem
quoseseuertanttantaesortessomnium.
ibi ex oraclouocediuinaedidit
ApollopuerumprimusPriamoquifo_ret
postillanatustemperarettollere; 6o
eum esseexitium Troiae,pestemPergamo.
sint haec ut dixi somnia fahularum ...

XIX
Varro, Ling. 6. 83: ah auribus uerha uidentur dicta audio et
ausculto; aures (A. Spengel:auris 0. Mueller:audio F) ah aueo
(Laetus:ahaucto F), quod his auemus discere (uu{go:dicere F)
semper, quod Ennius uidetur hvµov ostendere uelle in Alexan-
dro cum ait:
iam dudumab ludisanimusatqueauresauent 62
auideexspectantes
nuntium.
propter hanc aurium auiditatem theatra replentur.

XX
Varro, Ling. 7.82: apud Ennium: 'Andromachaenomen qui
indidit recte ei indidit (v. 99)'. item:
quapropter
ParimpastoresnuncAlexandrumuocant. 64

imitari dum uoluit (Aldus:uolunt F) Euripidem (euripeden F)

et ponere hvµov, est lapsus; nam Euripides quod Graeca posuit,
hvµa sunt aperta. ille ait ideo nomen additum Andromachae,
quod &v6plµaxe-ra1(Aldus: andromache · quod andromachete
F); hoe Enni (ennii F) quis potest intellegere in uersu (Turnebus:
54 et sacrificabat H 56 doceret V 58 ubi B 6o temptaret tollere
H: tempora extollere vc
 ALEXANDER

inuersum F) significare 'Andromachae nomen qui indidit recte

indidit' aut Alexandrum ah eo appellatum in Graecia qui Paris
fuisset,a quo Herculem quoque cognominatum Alexicacon, ah
eo quod defensor esset hominum?

XXI
(a) Festus, p. 240. 10:
antiquos \1
dicisum est i
nifici ait neq\l~
in Alexandr9
tamidiot purusP\lt(us)
sycophanta est
quo certior sc
putatum dici sol
ta id est pura fact

(b) Gellius, 7. s.10: scriptum est autem 'purum putum' non in

Carthaginiensi solum foedere sed cum in multis aliis ueterum
libris tum in Q. quoque Ennii tragoedia, quae inscribitur
Alexander, et in satira M. Varronis, quae inscripta est Sls
irai&s ol Y~VTES-
XXII
Festus, p. 416.35: STOLIDVS, stultus. Ennius lib. 1. •. et m
Alexandro:
hominemappellat.'quid tiasciuit stolide?'non intellegit. 66

et Caecilius ...

XXI (a) PVTVS antiqui diccbant pro puro, unde putatac uitcs et arborcs, quod
decisis inpcdimcntis rcmancrcnt purac. aurum quoquc putatum dici solct, id
est expurgatum, et ratio putata, id est pura facta - Paulus
XXD 66 lasciuisScaliger

79
 THE FRAGMENTS

XXIII
Festus, p. 494. 33: TABNIAS Graecam uocem sic interpretatur
Verrius, ut dicat ornamentum esse laneum capitis honorati, ut
sit apud Caecilium ... Ennius in Alexandro:
uolansde caelocum coronaet taeniis.
Accius ...
XXIV
Macrobius, Sat. 6. 1.61 (7: dicam itaque primum quos ah aliis
traxit uel ex dimidio sui uersus uel paene solidos) : 'multi
praeterea quos fama obscura recondit (Verg. Aen. s .302)'.
Ennius in Alexandro:
multi alii aduentant,paupertasquorumobscuratnomina. 68

XXV
Macrobius, Sat. 6. 2. 18 ( 1 : locos locis componere sedet animo
ut unde formati sint quasi de speculo cognoscas): 'o lux
Dardaniae spes o fidissima Teucrum (Verg. Aen. 2.281)' et
reliqua. Ennius in Alexandro:
o lux Troiae,germaneHector, 69
quid ita cum tuo laceratocorporemiser?
aut qui te sic respectantibus
tractauerenobis?

XXIV *ex Vergilio Statius, Theb.6. 56o.

XXV *Vergilius, Aen. 2.281-6 o lux Dardaniae, spes o fidissima
Teucrum, I quae tantae tenuere morae? quibus Hector ab oris I
exspectate uenis? ut te post multa tuorum I funera, post uarios
hominumque urbisque labores I defessi aspicimus. quae causa
indigna serenos I foedauit uoltus? aut cur haec uolnera cemo ?,
6. 500-2 Deiphobe armipotens, genus alto a sanguine Teucri, I quis
tarn crudelis optauit sumere poenas? I quoi tantum de te licuit? ·
XXV 70 ita cumque tuo T miser (ades) Mariotti:miser (es) Vahlm

80
 ANDROMACHA

XXVI
Macrobius, Sat. 6. 2. 25: 'cum fatalis equus saltu super ardua
uenit I Pergama et armatum peditem grauis attulit aluo {Verg.
Aen. 6.515-16)'. Ennius in Alexandra:
nam maximosaltusuperauitgrauidusarmatisequus 72
qui suopartu arduaperdatPergama.

ANDROMACHA
XXVII
(a) Cicero, Sest. 120: quid fuit illud quod, recenti nuncio de illo
senatus consulto quod factum est in templo Virtutis ad ludos
scaenamque perlato, consessu maximo summus artifex et me-
hercule semper partium in re publica (Naugerius:in TR.PL. codd.)
tam quam in scaena optimarum, flens et recenti laetitia et
mixto dolore ac desiderio mei, egit apud populum Romanum
multo grauioribus uerbis meam causam quam egomet de me
agere potuissem? summi enim poetae ingenium non solum

XXVI *Lucretius 1.476-7 nee clam durateus Troianis Pergama

partu I inB.ammasset equos noctumo Graiugenarum; Vergilius,
Georg.3. 139-41 exactis grauidae cum mensibus errant, I non illas
grauibus quisquam iuga ducere plaustris, I non saltu superare uiam
sit passus, Aen. 2.237-8 scandit fatalis machina muros I feta armis,
I
328-9 arduus armatos mediis in moenibus adstans fundit equos,
6. 515-16 cum fatalis ecus saltu super ardua uenit I Pergama et
armatum peditem grauis attulit aluo; Ouidius, Ars 1 . 364 militibus
grauidum laeta recepit equum; Macro bi us, Sat. 3 . 13 . 13 nam Titius
in suasione legis Fanniae obicit saeculo suo quod porcuin Troianum
mensis inferant, quern il1iideo sic uocabant, quasi aliis inclusis ani-
malibus grauidum, ut ille Troianus equus grauidus armatis fuit.
XXVI 72 nwic maxima P

6 81 JTO
 THE FRAGMENTS

arte sua sed etiam dolore exprimehat. tqua enimt (qua enim
ui Koechly)
qui rempublicamcertoanimoadiuuerit 74
statuerit,steteritcumAchiuis-
uohiscum me stetisse dicehat, uestros ordines demonstrahat.
reuocahatur ah uniuersis.
re dubia
haut dubitarituitam o.fferreneecapitipepercerit.
121: haec quantis ah illo clarnorihus agehantur. cum iam
omisso gestu uerhis poetae et studio actoris (Heruagius:auctoris
codd.)et exspectationi nostrae plauderetur:
SVMMVM AMICVM SVMMO IN BBi.LO

- narn illud ipse actor adiungehat amico anirno et fortasse

homines propter aliquod desiderium adprohahant -
SVMMO INGENIO PRAEDITVM.

tum (G: tam P: iarn ed. Ascens.1531)il1aquanto cum gemitu

populi Romani _aheodem paulo post in eadern fahula sunt acta.
'o pater- (v. 87).'
me, me ille ahsentem ut patrem deplorandum putahat (ed.
Ascens.1531:putarat codd.), quern Q. Catulus, quern multi alii
saepe in senatu patrern patriae nominarant. quanto cum fletu de
illis nostris incendiis ac ruinis, cum pattern pulsum, patriam
adflictam deploraret, dornum incensan1euersamque. sic egit ut,
demonstrata pristina fortuna, cum se conuertisset,
'haec omnia uidi inflammari (v. 92)'
fletum etiam inimicis atque inuidis excitaret.
122: pro di immortales. quid? ilia quern ad modum dixit
idem. quae mihi quidern ita acta et scripta uidentur esse ut uel a
Q. Catulo, si reuixisset, praeclare posse dici uiderentur; is enim
XXVD (a) 77 haut Madvig: *ut (fuit aut) P: ut G dubitari G uitam
Naugerius:uiam codd. pepercerit uulgo: pcper P1 : pcpcrcit pee

82
 AND ROMACHA
libere reprehendere et accusare populi non numquam temeri-
tatem solebat aut errorem senatus.
'o (P: uero G) ingratifici Argiui (argui P 1), immunes (schol.
Bob.: inanes codd.)Grai, inmemores bencfici (bencfitii
codd.)'.
non erat illud quidem uerum; non enim ingrati, sed miseri,
quibus reddere salutem a quo (P: quod G) acceperant non
liceret, nee unus in quemquam umquam gratior quam in me
uniuersi; sed tamen illud scripsit disertissimus poeta pro tmet
(Telamone lAmbinus),egit fortissimus actor, non solum opti-
mus, de me, cum omnis ordines demonstraret, senatum equites
Romanos uniuersum populum Romanum accusarct.
'exulare sinitis (ed.Ascens.1531:sinite codd.), sistis (si istis codd.)
pelli (P: belli G), pulsum patimini '.
quae tum significatio fuerit omnium, quae declaratio uoluntatis
ah uniuerso populo Romano in causa hominis non popularis,
equidem audiebam (ed. Ascens. 1531: audiebamus codd.);
existimare facilius possunt qui adfuerunt.
123 : et quoniam hue me prouexit oratio, histrio casum meum
totiens conlacrimauit, cum ita dolenter ageret causam meam, ut
uox eius ilia praeclara lacrimis impediretur; neque poetae,
quorum ego semper ingenia dilexi, tempori meo defuerunt;
eaque populus Romanus non solum plausu sed etiam gemitu
suo comprobauit. utrum igitur haec Aesopum potius pro me
aut Accium dicere oportuit, si populus Romanus liber esset, an
principes ciuitatis? nominatim sum appellatus in Bruto.
'Tullius qui libertatem ciuibus stabiliuerat '.
miliens reuocatum est. parumne uidebatur populus Romanus
iudicare id a me et a senatu esse constitutum quod perditi ciues
sublatum per nos criminabantur?
(b) Cicero, De orat.3 . 102: numquam agit hunc uersum Ros-
cius eo gestu quo potest ... quid ille alter? 'quid petam praesidi
83 ~2
 THE FRAGMENTS

(praesidiicodd.)? (v.81)' quam leniter, quam remisse, quamnon

actuose. ins tat enim: 'o pater o patria o Priami do mus (v. 87) '.
in quo tanta commoueri actio non posset, si esset consumpta
superiore motu et exhausta.
(c) Cicero, De orat. 3. 183: est autem paean hie posterior non
syllabarum numero sed aurium mensura, quod est acrius
iudicium et certius, par fere cretico, qui est ex longa et breui et
longa, ut 'quid petam praesidi (praesidii codd.)aut exsequar?
quoue nunc (v. 81) '.

(d) Cicero, De orat. 3 .217: aliud enim uocis genus iracundia

sibi sumat, acutum, incitatum, crebro incidens ... aliud miseratio
ac maeror, Bexibile plenum interruptum fiebili uoce: ... et illa:
'o pater o patria o Priami domus (v. 87)'; et quae sequuntur
'haec omnia uidi (uidet M) infiammari, Priamo ui uitam euitari
(ui uitam uitari L: uitam euitaret M) (vv. 92-3)'. aliud metus,
demissum et haesitans et abiectum ...

(e) Cicero, Orat. 92: translata dico, ut saepe iam, quae per
similitudinem ah alia re aut suauitatis aut inopiae causa trans-
feruntur; immutata (Schutz:mutata AL), in quibus pro uerbo
proprio subicitur aliud quod idem significat sumptum ex re
aliqua consequenti.
93 : quod quamquam transferendo fit, tamen alio modo
transtulit cum (L: quod A) dixit Ennius tarcem et urbem
orbas (L: arcent urbem orbam A) (v. 83 ?) alio modo si pro
patria arcem dixisset et horridam Africam terribili tremere
tumultu cum dicit pro Afris immutate Africam t
(f) Cicero, Tusc. 1 . 8s: sit igitur aliquis qui nihil mali ha beat,
nullum a fortuna uolnus acceperit. Metellus ille honoratis (V2 :
honoratus X) quattuor filiis aut (K: at GRV) quinquaginta
Priamus, e (V2: om. X) quibus septemdecim iusta uxore natis.
in utroque eandem habuit fortuna potestatem, sed usa in altero
84
 AND ROMACHA
est. Metellum enim multi filii filiae nepotes neptes in rogum
inposuerunt, Priamum tanta progenie orbatum, cum in aram
confugisset, hostilis manus intcremit. hie si uiuis filiis incolumi
regno occidisset 'astante ope barbarica, tectis caelatis laqueatis
(vv. 89-90)' utrum tandem a bonis an a malis discessisset?tum
profecto uideretur a bonis. at certe ei melius euenisset, nee tam
fiebiliter illa canerentur: 'haec omnia uidi infiammari, Priamo
ui uitam euitari, Iouis aram sanguine turpari (vv. 92-4)'. quasi
uero ista ui (PetrusCrassus:uel codd.)quicquam tum potuerit ei
melius accidere. quodsi ante occidisset, talem (Dauisius:tamen
codd.)euentum omnino amisisset; hoe autem tempo re sensum
antisit malorum.

(g) Cicero, Tusc.I . 105: sed plena errorum sunt omnia. trahit
Hectorem ad currum religatum Achilles; lacerari eum et
sentire, credo, putat. ergo hie ulciscitur, ut quidem sibi
uidetur; at illa sicut acerbissimam rem maeret:
uidi, uiderequodmepassaaegerrume,
Hedoremcurruquadriiugo raptarier.
quern Hectorem, aut quam diu ille erit Hector? melius Accius
et aliquando sapiens Achilles: 'immo enimuero corpus Priamo
reddidi, Hectora (Nieberding:Hectorem codd.)abstuli'. non
igitur Hectora traxisti, sed corpus, quod fuerat Hectoris.

(h) Cicero, Tusc. 3 . 44: quaerendum igitur quern ad modum

aegritudine priuemus cum qui ita dicat:
'pol mihi fortuna magis nunc defit quam genus.
namque regnum suppetebat mi, ut scias quanto e loco,
quantis opibus, quibus de rebus lapsa fortuna accidat
(vv. 338-40)'.

XXVD (g) 79 curro GI(lR etfart. V1 (-u i,a ras.)

85
 THE FRAGMENTS

quid? huic calix mulsi impingendus est, ut plorare desinat, aut

aliquid eius modi? ecce tibi ex altera parte ab eodem poeta:
ex opibussummisopisegensHectortuae. So
huic subuenire debemus; quaerit enim auxilium.
quidpetampraesidiaut exequar?quouenunc 81
auxilioexili autfugaefreta sim?
arceet urbeorbasum. quo accedam? quoapplicem?
cui neearaepatriaedomislant,fractaeet disiectaeiacent,
Janafiammadefiagrata, tosti talii t stantparietes, ss
deformatiatqueabietecrispa.
scitis quae sequantur et ilia in primis (Tregder:illum primis X:
illud in primis yes'):
o pater,o patria,o Priamidomus,
saeptumaltisonocardinetemplum.
uidi egote adstanteope barbarica,
tectiscaelatislaqueatis, 90
auroeboreinstructamregi.fice.
45 : o poetam egregium. quamquam ab his cantoribus Eu-
phorionis contemnitur. sentit omnia repentina et necopinata
esse grauiora. exaggeratis igitur regiis (s':regis X) opibus, quae
uidebantur sempiternae fore, quid adiungit?
haecomniauidi injlammari, 92
Priamoui uitameuitari,
Iouisaramsanguineturpari.
praeclarum carmen. est enim et rebus et uerbis et modis lugubre.
eripiamus huic aegritudinem. quo modo? conlocemus in
culcita plumea ...

XXVD (h) 81 pracsidii X 82 exilii X fugae ~ Bentley: fuga X Ss alti

M'< 89 adstantem (m eras.in V) X 90 laqueatis codd.:lacuatis Seruius
aud. Atn. 1 • 726 91 regificem (m exp. K1B) X 94 sanguine KR' todd.
Non. p. 181.1: sanguinem GR 1 V

86
 ANDROMACHA
(i) Cicero, Tusc. 3. 53: Karthaginienses multi Romae seruie-
runt, Macedones rege Perse capto; uidi etiam in Pdoponneso,
cum essem adulescens, quosdam Corinthios. hi poterant omnes
eadem ilia de Andromacha (~: antromacha X) deplorare:
'haec omnia uidi (v. 92)'. sed iam (etiam KR) decantauerant
fortasse. eo enim erant uoltu, oratione, omni reliquo motu et
statu, ut eos Argiuos aut Sicyonios (sicionios K1R) diceres
(dicere X: co". vc), magisque me mouerant Corinthi subito
aspectae (aspecta X: co". V2) parietinae quam ipsos Corinthios,
quorum animis diuturna cogitatio callum uetustatis obduxerat.

XXVIII
(a) Cicero, De orat.2 . 1 ss: ' ... miror cur philosophiae sicut
Zethus ille Pacuuianus prope bellum indixeris '.
156: 'minime ', inquit Antonius; 'ac sic decreui philosophari

XXVII *Plautw, Baah. 933-4 o Troia o patria o Pergamum o

Priame periisti senex I qui misere male mulcabere quadringentis
Philippis aureis; Sallustius, lug. 14. 17 nunc uero exul patria domo,
solus atque omnium honestarum rerum egens, quo accedam aut quos
appellem?; Vergilius, Atn. 1 . 483 ter circum Iliacos raptauerat
Hectora muros, 2.241-2 o patria (o PATRIA. uersus Ennianus-
Seruius)o diuom domus Ilium et incluta hello Imoenia Dardanidum,
27r5 raptatw bigis ut quondam aterque crucnto I puluere perquc
pedes traiectw lora tumentis. I ei mihi (m MIHI. Ennii uersus -
Seruius)qualis erat, quantum mutatus ah illo I Hectore, 499-505 uidi
ipse furentem Icaede Ncoptolemum geminosque in limine Atridas, I
uidi Hecubam centumque nurus Priamumque per aras I sanguine
foedantem quos ipse sacrauerat ignis. I quinquaginta illi thalami, spes
tanta nepotum, I barbarico postes auro spoliisque superbi I pro-
cubuere; Porcius Latro ap. Sen. Contr.2. 1. I uidi ego magni exer-
citus ducem sine comite fugientem, uidi ... limina deserta... nam
quid ex summis opibus ad egestatem deuolutos loquar?; Tacitus,
Ann. 13. 15 ille constanter exorsus est carmcn quo euolutum cum
sede patria rebusque summis significabatur.

87
 THE FRAGMENTS

potius, ut Neoptolemus apud Ennium, "paucis; nam omnino

(quam omnino HE 2) haud placet'".
(b) Cicero, Rep. 1 . 30: Aelius Sextus ... Zethum ilium Pacuui
nimis inimicum doctrinae esse dicebat; magis eum delectabat
Neoptolemus Enni qui se ait philosophari (filosofari cod.)uelle
set paucis; nam omnino haud placere.
(,) Cicero, Tusc. 2. 1: Neoptolemus quidem apud Ennium
philosophari sibi ait necesse esse (est H) sed paucis; nam omnino
haud placere. ego autem, Brute, necesse mihi quidem esse
arbitror philosophari - nam quid possum praesertim nihil
agens agere melius? - sed non paucis ut ille ...
2: sed tamen in uita occupata atque, ut Neoptolemi tum erat,
militari pauca ipsa multum saepe prosunt et ferunt fructus ...
(d)Gellius s.1 s.9: hos alios talis argutae delectabilisque desidiae
aculeos cum audiremus uel lectitaremus neque in his scrupulis
aut emolumentum aliquod solidum ad rationem uitae pertinens
aut finem ullum quaerendi uideremus, Ennianum (ennianum
autem RV) Neoptolemum probabamus, qui profecto ita ait:
estpauds; nam omninohaudplacet.
philosophandum 95

(e) Gellius s. 16. s: sed hie aeque (Petschenig:

eaque codd.)non
diutius muginandum eiusdemque illius Enniani Neoptolemi, de
quo supra scripsimus, consilio utendum est, qui degustandum
ex philosophia censet, non in eam ingurgitandum.
(f) Apuleius, Apol. 13: da igitur ueniam Platoni philosopho
uersuum eius de amore ne ego necesse habeam contra senten-
tiam Neoptolemi Enniani pluribus philosophari.

XXIX
Cicero, Att. 4. 1 s.6: redii Romam Fontei causa a.d. vn. Id.
Q_uint. veni spectatum (Graeuius:spectaculum codd.)primum
magno et aequabili plausu (sed hoe ne curaris; ego ineptus qui
88
 AND ROMACHA
scripserim); deinde Antiphonti operam. is erat (Victorius:
miserat codd.)ante manu missus quam productus. ne diutius
pendeas, palroam tulit; sed nihil tarn pusillum nihil tarn sine
uoce nihi1tam - uerum haec tu tecum habeto. in Andro-
macha tamen maior fuit quam Astyanax ( Victorius:Astyanax
nam Cratander:astya [uel astia] nam /l: astra nam TT<I>), in
ceteris parem habuit neminem.

XXX
Cicero, Opt. gen. 18: huic labori nostro duo genera reprehen-
sionum opponuntur. unum hoe: 'uerum (uu~o: uerbum G)
melius Graeci '. a quo quaeratur ecquid (uu~o: et quid G)
possint ipsi (uu~o: illi G) melius Larine? alterum: 'quid istas
potius legam quam Graecas?' idem (uu~o:id est G) Andriam et
Synephebos nee minus Terentium et Caecilium quam Menan-
drum legunt, nee Andromacham aut.Antiopam aut Epigonos
Latinos trecipiuntt, sed tamen Ennium et Pacuuium et
Accium potius quam Euripidem et Sophoclem legunt. quod
igitur est eorum in orationibus e Graeco (Lambinus:a greco G)
conuersis fastidium, nullum cum sit in uersibus?

XXXI
Cicero, Ac. 2. 20: quam multa quae nos fugiunt in cantu exau-
diunt in eo genere exercitati qui primo inflatu tibicinis Antio-
pam esse aiunt aut Andromacham, cum id nos ne suspicemur
quidem.
XXXII
Cicero, Diu. 1 . 23 : sus rostro si humi A litteram inpresserit,
num propterea suspicari poteris Andromacham Enni (Ennii
Acvc) ah ea posse describi?

89
 THE FRAGMENTS

XXXIII
Varro, Ling. s. 19: omnino feo magis puto a chao chount et
hinc caelum, quoniam, ut clixi, 'hoe circum. supraque quod
complexu continet terram ', cauum caelum. itaque dicit
Andromaca Nocti:
quaecauacaeli 96
signitenentibus
con.fids
bigis.
et Agamemno: 'in altisono caeli clipeo (vv. I 88-9) '; cauumenim
clipeum. et Ennius item ad cauationem: 'caeli ingentes fornices'.

XXXIV
Varro, Ling. 7. 6: templum tribus modis dicitur: ah natura, ah
auspicando, a similitudine; natura in caelo, ah auspiciis in terra, a
similitudine sub terra. in caelo templum dicitur, ut in Hecuba ...
in terra, ut in Periboea ... sub terra, ut in Andromacha:
Acherusiatemplaalta Orci salueteinfera. 98

XXXV
Varro, Ling. 7. 82: apud Ennium:
Andromachaenomenqui indiditrede indidit. 99

item ... imitari dum uoluit (Aldus: uolunt F) Euripidem

(euripiden F) et ponere m,µov, est lapsus. nam Euripides
(euripedes F) quod Graeca posuit, hvµa sunt aperta. ille ait ideo

XXXIV *Trag. inc. ap. Cic. Tusc. I . 48 quae est anus tam delira
quae timeat ista quae uos uidelicet si physica non didicissetistimeretis
'Acherunsia templa alta Orci pallida leti nubila Qetio nubila GK1 (b
post o add.K']R: let/lo nubila V [leto nubila B]}tenebris loca'?
XXXV *Cicero, Att. 2. I . s ego illam odi male consularem. ea est
enim seditiosa, ea cum uiro bellum gerit.
xxxm Andromacha Nocti 'quae Lattus: Andromeda Nocti 'quae Scaligtr:
androma noctiquc F
XXXV 99 recte ei indidit F
90
 ANDROMACHA
nomen additum Andromachae quod av6pl l-lCXXETa1 (Aldus:
andromache· quod andromachete F).hoe Enni (EnniiF) quis
potest intellegere in uersu (Turnebus:inuersum F) significare
'Andromachae no men qui indidit recte indidit' aut ...

XXXVI
Varro, Ling. 10.70: tde generet multi utuntur non modo
poetae sed etiam plerique thaec primot omnes qui soluta
oratione loquuntur ( • • •) dicebant ut quaestorem praetorem
sic Hectorem Nestorem. itaque Ennius ait:
Hectorisnatumde Troianomuroiactari. 100

Accius haec in tragoediis largius a prisca consuetudine mouere

coepit et ad formas Graecas uerborum magis reuocare.

XXXVII
Festus, p. 384.16: ysSIdicebantur
Naeuius: 'odi' inquit
inde aperte dice
times? Ennius in sexto
ntus in occulto mussa
s in Andromacha: di 101
9n est: nam mussare si
s in Agnorizomene:
'quod potes sile cela oc tege
tace mussa mane'. SVM pro eum

XXXVIII
Nonius, p. 76. I: AVGIFICAT, auget. Ennius Andromaca:
tquid fit seditio tabesne an numerust augificatfsuost. 102

XXXVI 100 iactari uulgo: lactari F

XXXVD SVMMVSSI,murmuratorcs. Naeviw: •odi' inquit •summussos, proinde
aperte dice quid sit'. Terentiw (sic) mwsare pro taccre posuit cum ait: •sile ccla
occulta tege tacc mussa' - Paulus
XXXVIII andromaga L 1 102 tabetne Lipsius nwneros ed.print.
91
 THE FRAGMENTS

XXXIX
Nonius, p. 292.7: BXANCLAitB etiam significat perpeti. Ennius
Andromache aechmaloto:
quantiscumaerumnisiliumexanclauidiem.

XL
(a) Nonius, p. 401. 37: SVMMVM, gloriosum, laudabile ...
Ennius Andromache aechmaloto (andromaca haec malo codd.):
'annos multos longinque Qongique codd.)ah domo bellum
gerentes summum summa (summam AA) industria '.
(b) Nonius, p. 515. 12: LONGINQVB et LONGITBR pro longe.
Ennius Andromache aechmaloto:
annosmultoslonginquea domo 104
heliumgerentessummumsummaindustria.

XLI
Nonius, p. 504.18: LAVERBNT (uulgo:lauere codd.)etiam inde
manauit. Ennius Andromaca:
nam ubi introducta
estpuerumqueut lauerentlocant 1o6
in clipeo.
XLII
Nonius, p. 505 . 12 SONVNT etiam inde manauit. Ennius
Andromache aechmalotide:
nam nequeiratinequeblandiquicquamsinceresonunt. 108

XXXIX Andromache aedunaloto 'quantis Gerlach:Andromache aechmalo-

tidi 'quantis Roth: andromache malo torquantis (torquentis LBA) codd. 103
exanclaui cum diem AAL•
XL (b) accius andromache ei nwo (mala CA) codd.
XLI 1o6 puerorumque G 107 clypeo Aldus: cypeo LBamb.: cipeo HG
XLU andromace ethemapotide codd. 108 quiquam codd.:co". L 1

92
 ANDROMEDA

XLIII
Nonius, p. 515.24: RARBNTBR.•. Ennius Andromacha:
sedquasiautferrum aut lapis 109
duratrarentergemitumtconatur trabem t

XLIV
{a) Servius, Aen. 1 . 224: ueliuolum duas res significat, et quod
uelis uolatur, ut hoe loco, et quod uelis uolat, ut Ennius: 'naues
(nauius C) ueliuolas'. qui et proprie dixit.

(b) Macrobius, Sat. 6. 5. 10: rnulta quoque epitheta apud

( 1:
Vergilium sunt quae ah ipso ficta creduntur sed et haec a ueteri-
bus tracta monstrabo) despiciens mare ueliuolum (Aen. 1 .224)
... Ennius in quarto decimo ... idem in Andromache:
rapitex altonauesueliuolas. Ill

ANDROMEDA
XLV
Festus, p. 312. 7: QVABSO, ut significat idem quod rogo, ita
quaesere ponitur ah antiquis pro quaerere, ut est apud Ennium
lib. n ... et in Cresphonte ... et in Andromeda:
liberumquaesendum
causafamiliae matremtuae. 112

XLDI 109 quasi ferrum B'0 4 no conatu trahens Lipsius

XLIV (b) andromache rapit A: dromachera alpit N: dromachera capit P:
dromacera apice T: dromache rapit RF uelicolas F
XLV andromedoa liberum quae sedm (i.e. secundum) F

93
 THE FRAGMENTS

XLVI
(a) Festus, p. 448.19:
saxa et difficili
ri insuetae, aut
lere. Ennius jn An
tita saxo atque host
unde scrupulosam
in se asperi. Come!
Im: his tum iniectus
et quaedam dubitatio

(b) Nonius, p. 169. 25: pro scabra es (QH.icherat:

SCABRES
scapres pro scabres codd.). Ennius Andromeda:
scrupeoinuestitasaxoatqueostreistquam excrabrentt. 113

XLVII
Festus, p. 514.22: VRVAT. Ennius in Andromeda significat
circumdat, ah eo sulco, qui fit in urbe condenda uruo aratri,
quae fit forma simillima uncini curuatione buris et dentis, cui
praefigitur uomer. ait autem:
circumseseuruatadpedesa te"a quadringentos
tcaputt. n4

XLVIII
Nonius, p. 20. 18: CORPORARB est interficere et quasi corpus
solum sine anima relinquere. Ennius Andromeda:
corpuscontemplatur
undecorporaret
uulnere. us
XL VI (a) SCRVPI dicuntur aspera saxa et difficiliaattrcctatu; uncle saupulosam
rem dicimw, quae aliquid in se habct asperi - Paulus
XL VI (b) I I 3 scrcpeo L squamae scabrent Mncerus

94
 ANDllOMEDA

XLIX
(a) Nonius, p. 165.8: i"reciproca animum in quam odioset
Ennius Andromeda: 'rursus (riscus B.4) prorsus rcciprocat
fluctus (fructus codd.)feram'.

(b) Nonius, p. 384. 32: RVRSVS, retro ... Ennius Andromeda:

rursusprorsusredprocat
fluctus tferamt. 116

L
Nonius, p. 183. 18: VISCBRATIM, per uiscera (peruiscera CJ 04:
om. L ). Ennius Andromeda:
aliaJluctusdilfert dissupat 117
uisceratim
membra;mariasalsaspumantsanguine.

LI
Priscianus, Gramm. II 293 . s: inueniwitur tamen quaedam
pauca feminini generis, quae ex masculinis transfigurantur non
habentibus neutra, quae et animalium swit demonstratiua,
naturaliter diuisum genus habentia, quae differentiae causa
ablatiuo singulari 'bus' assumentia faciwit datiuum et ablati-
uum pluralem, quod nulla alia habet declinatio in 'bus' termi-
nans supra dictos casus,ut 'a' longam in eispaenultimam habeat,
ut 'his natabus ', 'filiabus ', 'deabus ', 'equabus ', 'mulabus ',
'libertabus ', 'asinabus' . . . et 'filiis' tamen in eodem genere
dictum est. Ennius in Andromeda:
filiis propterte obiectasum innocensNerei. 119

[id est natis pro natabus] id est Nerei filiabus.

U andromcadaRr: andromedia GK

9S
 THE FRAGMENTS

ATHAMAS
LII
Charisius, p. 314. 9 BVHOB ••• Ennius in Athamante:
his erat in oreBromius,his Bacchuspater, 120

illis Lyaeusuitis inuentorsacrae.

tum pariterteuhan euhiumt
ignotusiuuenumcoetusalternauice
inibatalacrisBacchicoinsultansmodo.

CRESPHONTES
LIII
Rhetor ineertus, Her. 2. 38: utuntur igitur studiosi {studioseM)
in eonfirmanda ratione duplici eonclusione hoe modo:
iniuriaabste adfidorindignapater. 12s

nam si inprobumesseCresphontemexistimas,
cur me huic locabasnuptiis?sin estprobus,
cur taleminuitaminuitumcogislinquere?
quae hoe modo eoncludentur, aut ex eontrario eonuertentur aut
ex simpliei parte reprehendentur {reprehendetur M). ex
eontrario (eontraria M) hoe modo:
nulla te indignanataadfido iniuria. 129

si probusest, collocaui.sin est inprobus,

diuortiote liberaboincommodis.
LII 120 hiserat Fabricius:is crat N 121 illisLyaeus uitis n1 : illis* uitis n: illis
lisaew uitis N 122 tum pariter euhan (euhoe euhoe) euhium Fabricius: tum
pariter euchoe neucheum Cauchiiexc. 123 iuuenem Cauchii exc. 124
insultans ed. princ.: insultas N
LIii 126 chresponthe H: chrespontem BC: chresponthem PTT:threspontem E
existimabas bd 127 sin est EC: sine si M 128 linqucre TTBC:liquere
M: relinquerc E 129 nuta M 130 est locaui sin (sin autem bi) E: es
collocabisin M: est te locaui Oudendorp 131 te liberabo Omnibonus:libero
te M: liberabo te E
 CRESPHONTES

ex simplici parte reprehendetur si (sed M) ex duplici conclu-

sione alterutra pars diluitur, hoe modo:
'nam si inprobum esse Cresphontem (esse chrespontem M:
threspontem esse E) existimas (existimabas bd),
cur me huic (huius Cd: his bi) locabas nuptiis? : : duxi
probum.
erraui. post cognoui et fugio (fugio nunc E) cognitum'.
ergo reprehensio huiusmodi conclusionis duplex est; auctior
(acutior CBE) ilia superior, facilior haec posterior ad cogi-
tandum.
LIV
Festus, p. 312. 7: QVABSO, ut significat idem quod rogo, ita
quaesere ponitur ah antiquis pro quaerere, ut est apud Ennium
lib. n ... et in Cresphonte:
dudt me uxoremliberorumsibi quaesendum
gratia. 13~

LV
Festus, p. 334. 8 : RBDHOSTIRB, referre gratiam ... nam et hos-
tire (hostiae F) pro aequare posuerunt. Ennius in Cresphonte:
audiatqueauditishostimentumadiungito. 133

LVI
Gellius 7. 16. 8: sed neque solus Catullus ita isto uerbo (i.e.
'deprecor';92 .3) usus est. pleni sunt adeo libri similis in hoe
uerbo significationis, ex quibus unum et alterum quae sub-
petierant apposui.
9: Q. Ennius in Erectheo non longe secus dixit quam
Catullus ... signat abigo et amolior uel prece adhibita uel quo
alio modo.
10: item Ennius in Cresphonte:
ego meaecum uitaeparcam,letum inimicodeprecer. 134

LIV chrcsponte F
LV chrcsponte F 133 audi Scaligtr:audis F
LVI 134 cum meae codd.:transp.Bothe
7 97 JTO
 THE FRAGMENTS

LVII
Nonius, p. 144.12: NITIDANT, abluunt (F3mg.: albunt F3:
aluunt L1C.4J)A.); dictum a nitore. Ennius Cresphonte:
topiet 13s
eamsecumaduocant,eunt adfontem, nitidantcorpora.

LVIII
Nonius, p. 471.2: SORTIRBNT pro sortirentur ... MODBRANT pro
moderantur ... SORTIVNT. Ennius Cresphonte:
an interse sortiunturbematqueagros. 137

LIX
Macrobius, Sat. 6.2 .21 (c£fr. xxv): 'nee te tua funera mater I
produxi pressiue oculos aut uulnera laui {Vergilius, Aen. 9. 486-
7)'. Ennius in Cresphonte:
nequeterraminicerenequecruentaconuestirecorpora 138
mihi licuitnequemiseraelauerelacrimaesalsumsanguinem.

ERECTHEVS
LX
Festus, p. 158. 10:
aerumnas. NBMINIS
et quis diceret cum sit
uitio creatis neminisque
us Erectheo : lapideosunt 1 40
cordemulti quosnon miseretneminis.NBMO
LVII crcspontc codd. 136 ad fortcm codd.:co". P
LVID crcsfontc codd. 137 inter scsc Vossius
LIX crcsiphontc P: crcssiphontc NRFA 138 conucrtirc T: conucrtcrc A
139 mihi corpora codd.:transp.Bothe
LX NBMINIS gcnitiuo casu Cato (sic)usus est, cum dixit: 'sunt multi cordc quos
non miscret ncminis' - Paulus
98
 EV MENIDES

LXI
Gellius 7. 16.9: Q. Ennius in Erectheo non longe secus dixit
quam Catullus (c£ fr. LVI}:
tqwt nunc
inquit
aerumnamea libertatemparo,
quibusseruitutemmea miseriadeprecor.

LXII
Macro bi us, Sat. 6. 4. 6 ( I : ego conabor ostcnderc hunc studio-
sissimum uatem et de singulis uerbis ueterum aptissime iudicasse
et inseruisse electa operi suo uerba quae nobis noua uideri facit
incuria uetustatis) : 'tum ferreus hastis I horret ager (Vergilius,
Aen. 11. 6o1-2) '. HORRET mire se habet. sed et Ennius in quarto
decimo ... et in Erectheo:
armat arriguntt horrescunttela. 1-43

EVMENIDES
LXIII
Nonius, p. 292. 18: EXANCLARE, effundere. Ennius Eumeni-
dibus:
nisi patremmaternosanguineexanclandouldscerem. 1#

LXI cricthco V: cripitco (cripite o B') codJ.Ncm.p. 290.i8 1-41 cui codd.
Non. crumnam ea (crumnam et a L) codd.Ncm.: crumpna ea V libcrtatc
para codd.Ncm. 14-2mcam miscriam (mcam miseria A' : mea miscriar E)
codd.Ncm.
LXII crccthco RFA: crcctco NP: cricthco T 143 horrigunt N:
arguntA

99
 THE FRAGMENTS

LXIV
Nonius, p. 3o6. 32: PACBSSBRB significat recedere. Ennius
Eumenidibus:
dicouicisseOrestem.uos ab hoefacessite. 145

LXV
Nonius, p. 474.35: OPINO pro opinor ... Ennius Eumeni-
dibus:
tacereopinoesseoptumumet pro uiribus
sapere,atquefabularitute noueris.

LXVI
Nonius, p. 505. 16: BXPBDIBO pro expediam ... Ennius
Eumenidibus:
id ego aecumac iustumfecisseexpediboatqueeloquar. 148

HECTORIS L YTRA

LXVII
Festus, p. 334. 8: RBDHOSTIRE, referre gratiam ... nam et hostire
pro aequare posuerunt. Ennius in Cresphonte ... et in Hectoris
lytris:
quaemeacomminusmachaeraatquehastathospius manut. 149

LXIV 145 edico L. Mueller: dico ego Scaliger Orestcm ed. princ.: orestcn
codd. nuossab (nuosab B) A.4 faccssit (faccssiL1) codd.:co". Vrbin.307
LXV 147 atquca AA tatc C.4
LXVI 148 ac iwtum fccisseJocelyn: ac iw fuisse Gulielmius: accius fecisse
codd.
LXVII inncctoris lyrisque mea F: co". Vrsinus 149 hostibis eminus
Timpanaro: hostiuit c manu Scalig:er

100
 HECT ORIS L YTRA

LXVIII
Nonius, p. 111.7: FVAM, sim uel fiam ... Ennius Hectoris
lytris:
at egoomnipotens ISO
ted exposcout hoeconsiliumAchiuisauxiliofuat.

LXIX
Nonius, p. 222.25: SPBCVSgenere masculino ... Ennius Lytris:
inferumuastosspecus. 1s2
LXX
Nonius, p. 3SS. 3: OCCVPARB est proprie praeuenire ... Ennius
Hectoris lytris:
Hedor tei summat armatoseducitforas 153
castrisque
castraultroiamf erreoccupat.
LXXI
Nonius, p. 399. 8: SPBRNBRBrursum segregare. Ennius Hectoris
lytris:
meliusest uirtuteius: nam saepeuirtutemmali 1ss
nanciscuntur;ius atqueaecumse a malisspernitprocul.

LXXII
Nonius, p. 407. 24: TBNACIA est perseuerantia et duritia. Ennius
Hectoris lytris :
tducet quadrupedum iugo inuitamt 157
Jomainf,enaet iungeualidaquorumtenaciainf,enariminis.
LXVIII haectoris lytris (listris BA) codd. I soa ego L Is I ted Bothe:
tc codd. auxilio Vossius:auxilii ,odd.
LXX hacctoris listris codd. 153 ui sumnu Mercerus foras Iunius: in foras
,odd. 154 confcrrc Vossius
LXXI hacctoris lystris (listris G: lytris L 1) codd. ISS ius Bentinus:ciw codd.
156 nanciscu tur L: nascuntur G atque cum L
LXXII hacctoris listris (lystris Bamb.)codd. 158 iungc Lipsius: iugc ,odd.
IOI
 THE FRAGMENTS

LXXIII
Nonius, p. 467.23: VAGAS pro uagaris ... Ennius Hectoris
lytris:
constititcredoScamander,arboresuentouacant. 159

LXXIV
Nonius, p. 469.25: CVNCTANT pro CWlCtantur ••• Ennius ...
idem Hectoris lytris:
qui cupiantdarearmaAehilli tut ipset cundent. 16o

LXXV
Nonius, p. 472. 21: CONMISERBSCIMVS .•. Ennius Hectoris
lytris:
tser uos et uostrum t 161

imperiumetfidem Myrmidonum,uigiles,conmiserescite.

LXXVI
Nonius, p. 489. 29: TVMVLTI. Ennius Hectoris lytris:
quidhoehieclamoris,quidtumultiest?nomenquiusurpatmeum?163

LXXVII
N onius, p. 490. 6: STRBPm pro strepitus. Ennius Hectoris lytris:
quidin eastrisstrepitiest?

LXXDI hcctoris (haectoris H&mb.) lystris (litris &mb.) codd. 159 arboris
codd. uacant Columna:uagant codd.
LXXIV hcctoris (hacctoris H&mb.) lytris (lystris H1) codd. 16o ut ipsi
Junius cunctet AA
LXXV cctoris lystris (hacctori listris &mb.) codd. 161 ser uos et uostrum
(useruos urorQ &mb.) codd.:per uos et nostrum Palmerius
LXXVI haectoris lystris (litris H1: listris H1) codd. 163 qui tumulti CA
LXXVD haectoris lystris (lytris &mb.) codd.164 strepitus G

102
 HECTORIS L YTRA

LXXVIII
Nonius, p. 504. 30 SONIT pro sonat ... Ennius Hectoris lytris:
aessonit,frangunturhastae,terrasudatsanguine. 16s

LXXIX
Nonius, p. 510.32: SABVITBR pro saeue ... Ennius Hectoris
lytris:
saeuiter
fortunaferro cernuntde uictoria. 166

LXXX
Nonius, p. 518. 3: DBRBPBNTB••• Ennius Hectoris lytris:
ecceautemcaligoobortaest, omnemprospectumabstulit. 167
derepentecontulitsesein pedes.

LXXXI
Diomedes, Gramm.I 345. 3: similiter halare et halitare (alare et
alitare A: alere et alitare B: halere et halitare M). Ennius in
Lytris:
sublimeiterut quadrupedantes
jlammamhalitantes.

LXXXII
Diomedes, Gramm.I 387. 21 : est tertium his (i.e.odiet memini)
simile, ut quidam putant - nee enim defuerunt qui hoe uerbum
praesentis temporis esse dicerent -, noui nouisti nouit; et id
simile est instanti et perfecto, ut memini ... apud ueteres

LXXVID hacctoris lystris codd. 165 acs sonit Nie. Fabn: et sonit codd.
LXXIX haectoris lystris codd.
LXXX hectoris (haectoris H&mb.) lystris codd. 167 aborta LA' B' 168
se sese G in pede C'
LXXXI lustris codd. 169 sublime iter ut B: lublime iter A: lublime item M
quadrupedaorisB alitantcs B

103
 THE FRAGMENTS

pluraliter huius uerbi instans colligitur, cum nomus dicunt pro

eo quod est nouimus, ita ut Ennius in Lytris:
nos quiescere
aequumest; nomusamboVlixem.

HECVBA
LXXXIII
Varro, Ling. 7. 6: tempi um tribus modis dicitur: ah natura, ah
auspicando, a similitudine; natura in caelo, ah auspiciisin terra,a
similitudine sub terra. in caelo templum dicitur, ut in Hecuba:
o magnatemplacaelitumcommixtastellissplendidis. 171

in terra, ut in Periboea ... sub terra, ut in Andromacha ...

LXXXIV
Gellius 11 . 4. 1 : Euripidis uersus sunt in Hecuba uerbis,
sententia, breuitate insignes inlustresque; Hecuba est ad Vlixen
dicens (293-5):
2:
TO s· a~{ooµa,Kav KaKOOS
{KaKoScodd.Gell.) Afy,J, TO aov
VlK~(m{ae1,m{6e1codd.Bur.)· A6y05yap~ T" a~OWTOOV
lwv
KaKTwv 6oKoVVTc..:>v (Porson: CXVToS
CXVToS codd.)ov TCXVTOV
o%1e1.
3: hos uersus Q. Ennius, cum earn tragoediam uerteret, non
sane incommode aemulatus est. uersus totidem Enniani hi sunt:
haectu _etsiperuersedices,facile Achiuosjlexeris: 172
nam cum opulentilocunturpariteratqueignobiles,
eademdictaeademqueoratioaequanon aequeualet.
LXXXII lustris codd.
LXXXID inecuba F 171 caelitum Scioppius:caeli tum F
LXXXIV 172 haec tametsi X 173 namque opulenti cum Scaliger:nam
opulenti quum Porson pariter et TT 174 dicta atque eadem TT
104
 HECVBA

4: bene, sicuti dixi, Ennius; sed ignobiles tamen et opulenti

avrl a~OWTCA>VKa\ 6oKOVV"TCA>V satisfacere sententiae non
uidentur; nam ncque omnes ignobilcs a~oOcn, neque omnes
opulenti ~oOcnv.
LXXXV
Nonius, p. 115 • 28: GVTIATIM ••• Ennius Hecuba:
1"tndehinc meae inquamt lacrumae
guttatimcadunt. 17s

LXXXVI
Nonius, p. 116.31: GRATVLAIU, gratias agere. Ennius Hecuba:
luppitertibi summetandemmaleregestagratulor. 176

LXXXVII
Nonius, p. 153 .22: PBllBITBRE, perire ... Ennius Hecuba:
set numquamscripstisquisparentemaut hospitem 177
necassettquos quist cruciatuperbiteret.

LXXXVIII
Nonius, p. 223 .24: SALVM neutri generis est uulgari consue-
tudine. masculini. Ennius Hecuba:
undantemsalum. 179

LXXXIX
(a) Nonius, p. 224. 6: SANGVIS masculino genere in consue-
tudine habetur ... neutro. Ennius Hecuba:
heu me miseram.interii.perguntlaueresanguensanguine. 180

LXXXV haccuba codd. 175 uide hWlc meac in quern MerctrUs:uidc hanc
mcae in quam Vossius lacrumac a..t: lacrimac P: meac L
LXXXVI 176 iupitcr codd.:corr. L 1
LXXXVD 177 scripstis Vossius:scripsistis codd. qui Junius 178 nccassct
quo quis cruciatu Junius: nccassat quos quis cruciatur codd.
LXXXIX (a) hccuba hcu IJ.4:hccubac hcu L 1 : hccuba chcu P 180 labcrc
codd.

105
 THE FRAGMENTS
(b) Nonius, p. 466.18: cum sit eluere et emaculare et
LAVARE,
aquis sordida quaeque purgare, uetustatis auctoritas posuit
etiam polluere ... Ennius Hecuba (heucuba LBamb.:ecuba G):
'heu me miseram (miserum C,f). interii. pergunt lauere sanguen
sanguine (sanguen om. L 1 )'.
(c) Nonius, p. 503. 38: LAVIT pro lauat ... inde tractum
LAVERE
est ... Ennius ... idem Hecuba: 'heu me miscrum. interii.
pergunt lauere Qabere codd.) sanguincm (LA,f: sanguen
sanguinem _&4C.d)'.
XC
Nonius, p. 342. 23: MODICVM in consuetudine pausillum. uolu-
mus significare; modicum ueteres moderatum et cum modo
(commodo AA: conunodum LJ>f) dici uolunt ... Ennius
Hecuba:
quaetibi in concubiouerecunde
et modicemoremgerit. 181

XCI
Nonius, p. 474.32: MISBRBTB. Ennius Hecuba:
miserete tmanust
dateferrum qui me animapriuem. 182

XCII
(a) Nonius, p. 494.3: PAVPBIUES pro paupertate. Ennius
Hecuba: 'senex sum: utinam moriar mortem oppetam prius-
quam eueniat quod in pauperie mea (pauperie J>f : paupericm
C..f)senex grauiter gemam'.
(b) Nonius, p. 507. 19: BVBNAT pro eueniat. Ennius Hecuba:
senexsum: utinammortemobpetampriusquameuenat 183
quodin pauperiemeasenexgrauitergemam.
XC haecuba L: heccuba Gm.H 181 in cubio AA.D-4
XCI ecuba codd.

1o6
 THE FRAGMENTS

IPHIGENIA
XCIII
Rhetor incertus, Her. 3 . 34: cum uerborum similitudines
imaginibus exprimere uolemus, plus negoti suscipiemus et
magis ingenium nostrum exercebimus. id nos hoe modo facere
oportebit. 'iam domum ultionem (domu ultionem l: domui
ultionem C: domi ultionem HPTTBC 212bdp: domum itionem
p m. 2, librariuscod. Bernensis469 in marg., Vidorius)reges
Atridae parant'. in loco tconstitueret (PTTB:construere H:
oportet constituere bi:constituere oportet CJ) manus ad caelum
tollentem Domitium cum a Regibus Marciis loris caedatur: hoe
erit 'iam domum ultionem (domii ultionem l: domii ultiones
M: domi ultionem bd) reges'. in altero loco Aesopum et
Cimbrum subornari, ut agant ( W. Kroll: ut ad M: ut uel
uagantemE) Iphigeniam (ephigeniam codd.), in Agamemnonem
et (PTTB:in agamen non emit H: in agamen nomen et C: om.E)
Menelaum (om.E): hoe erit 'Atridae parant'. hoe modo omnia
uerba erunt (CE: erant M) expressa.

XCIV
Cicero, Tusc. I. 116: clarae uero mortes pro patria oppeti-
tae non solum gloriosae rhetoribus sed etiam beatae uideri
solent ... Menoeceus non praetermittitur qui item oraculo edito
largitus est patriae suum sanguinem. Iphigenia ((nam) Iphi-
genia Vahlen)Aulide duci se immolandam iubet ut hostium
(sanguis superscriptum habetV) eliciatur suo.

XCV
(a) Cicero, Rep. I . 30: in ipsius patemo genere fuit noster ille
amicus dignus huic ad imitandum, 'egregie cordatus homo,
catus Aelius Sextus', qui egregie cordatus et catus fuit et ah
107
 THB FRAGMENTS

Ennio dictus est, non quod ea quaerebat quae numquam inueni-

ret, sed quod ea respondebat quae eos qui quaesissent et cura et
negotio soluerent, cuique contra Galli studia disputanti in ore
semper erat ille (utrumA in e an e in A co"ectasit, diiudicarenon
potuitZiegler)de Iphigenia (ifigenia cod.)Achilles:
astrologorumsignain caeloquidsit obseruationis? 185
cum Capraaut Nepa aut exoriturnomenaliquodbeluarum,
quodest antepedesnemospectat,caeliscrutanturplagas.
(b) Cicero, Diu. 2. 30: Democritus tamen non inscite nugatur,
ut physicus, quo genere nihil est adrogantius. 'quod est ante
pedes nemo spectat, caeli scrutantur plagas'. uerum is tamen
habitu extorum et colore declarari censet haec dumtaxat:
pabuli genus et earum rerum quas terra procreet uel ubertatem
uel tenuitatem; salubritatem etiam aut pestilentiam extis
significari putat.
(c) Seneca, Apocol.8 .2: si mehercules a Saturno petisset hoe
beneficium, cuius mensem toto anno celebrauit, Saturnalicius
(Biicheler:saturnaliaeius codd.) princeps, non tulisset illud,
nedum ah Ioue, quem (Gronouius:illum deum abiouem
[abioue VL] qui codd.),quantum quidem in illo fuit, damnauit
incesti .. .
3 : ... hie no bis curua corriget (Sonntag:corrigit codd.)?quid
in cubiculo suo faciat nescit (Bucheler:nescio codd.),et iam
caeli scrutatur plagas.
(d) Nonius, p. 145. 12: NEPAMquidam cancrum putant ad illud
Plauti (Cas. 443) 'retrouorsum cedam: imitabor nepam (nepa
C..4I)d)' et illud aliud 'aut cum nepa tessett'. dubium in
utroque. nam uere nepa scorpius clicitur.
(e) Donatus, Ter. Ad. 386 (386-8 istuc est sapere, non quod
ante pedes modost I uidere sed etiam ilia quae futura sunt I
prospicere): NON QVOD ANTB PBDBS MODO BST VIDBRB. hoe
108
 IP HI GENIA

sumpsit poeta de illo in (in V: om. C) tsyrumt (Syrium

Schoell:physicum Lindenbrog)peruulgato ancillae dicto: 'quod
ante pcdcsest non uident, caeli scrutantur (uidet caeli scrutatur
V) plagas'.
XCVI
(a) Varro, Ling. s.19: omnino teo magisputo a chao choullt
et hinc caelum, quoniam, ut dixi, 'hoe circum supraque quod
complexa continet terram ', cauum caelum. itaque dicit
Andromacha nocti (uide fr. XXXIII) •• •et Agamemno: 'in
altisono caeli clipeo '. cauum enim clipeum. et Ennius item ad
cauationem: 'caeli ingentes fornices (v. 319)'.
(b) Varro, Ling. 7. 73:
quidnoctisuidetur?in altisono 188
caeliclipeotemosuperat
(plaustri) stellassublimumagens
etiamatqueetiamnoctisiter.
hie multam noctem ostendere uol t a temonis motu; sed temo
unde et cur dicatur latet. arbitror antiquos rusticos primum
notasse quaedam in caelo signa, quae praeter alia erant insignia
atque ad aliquem usum tcultlll'aet tempus designandum
conuenire animaduertebantur.
XCV *Cicero, Tusc. 5 . 114 et cum alii saepe quod ante pedes esset
non uiderent, ille (FBY,-•c•1 b: illa X) in (s': om. XF) infinitatem
omnem peregrinabatur, ut nulla in extremitate consisteret, Nat.
Jeor.3 .40 singulas enim stellas numeras deos eosque (easque P) aut
beluarum nomine appellas ut Capram, ut Nepam (cod.Vrsini:lupam
AVB), ut Taurum, ut Leonem, aut rerum inanim.arum, ut Argo, ut
Aram, ut Coronam; Poeta incertus, Aetna 254-6 nam quae mortali
spes est, quae amentia maior, I in Iouis errantem regno perquirere
diuos, Itantum opus ante pedes transire ac perdere segnem (&hrader:
segne est G: segnes x); ex Cicerone Minucius Felix 12. 7, Augustinus,
Conf.10. 16. 25, Ambrosius, Noe 7. 17, Paulinus Nolanus, Epist.12. 5.
XCVI (b) 190 (plaustri) stcllas Joalyn sublimum agens Joalyn: sublimis
agcns Tumtbus: sublime agens F
109
 THE FRAGMENTS

74: eius signa sunt quod has septem stellasGraeci ut Homerus

uocant &µa~av et propinquum eius signum ~v (bootem
signum F), nostri eas septem stellas thoues et temonem et
prope eas axem t triones enim et boues appellantur a bubulcis
etiam nunc cum arant terram ...
75: temo dictus a tenendo: is enim continet iugum. et
plaustrum appellatum a parte totum, ut multa. possunt triones
dicti, vn quod ita sitaestellae, ut ternae trigona faciant taiiquod t
(c) Festus, p. 454.37:
septem stellae appell
bus innctis quos trio
appellent quod iun~
quasi terrionei:µ
quod id astrum Graec
partem quandam
Ennius: 'superat
et physici eum sumi:µ
conten ... temp
ainnt, quod ita snnt
ut ternae proxi~~
trigona
(d) Apuleius, Socr.2: suspicientes in hoe perfectissimo mnndi,
ut ait Ennius, clipeo miris fulgoribus uariata caelamina.

XCVII
Festus, p. 218.21: OB praepositione antiquos usos esse pro ad
testis est Ennius, cum ait lib. XIV ••• et in Iphigenia:
Acherontemobiboubi Mortis thesauriobiacent. 192

eiusdem autem generis esse ait obferre, obtulit, obcurrit,

oblatus, obiectus. mihi non satis persuadet.

XCVD ephigcnia W: hiphigenia X 192 obibo ed. print.: adibo W: adhibo

X obiacent X: adiacent W
110
 IPHIGBNIA

XCVIII
(a) Festus, p. 292.7: PBDVM est quidem baculum incuruum
quo pastores utuntur ad conprehendendas oues aut capras; a
pedibus. cuius mcminit etiam Vergilius in Bucolicis, cum ait
(s. 88): 'at tu sume pedum '. sed in co uersu, qui est in Iphigenia
Enni:
procede:gradumproferrepedum, 193
nitere,cessasojide?
id ipsum baculum (Vrsinus: iaculum F) significari cum ait
Verrius, mirari sarisnon possum, cum sit ordo talis, et per cum
significatio aperta: gradum proferre pedum cessasnitere.

(b) Schol. Veronensis, Verg. Eel. s.88:

pedum autem est
baculum recuruum quo pastores utuntur • I aut adminiculum
pedum sit, ut ait Ennius in Iphigenia (ifigenia): 'gradum pro-
ferre pedum nitere cessas o fide' • • I pastores pedes ouium
(Keil: bouium) retrahere soleant.

XCIX
Gellius 19. 10.4: ••. cumque architectus dixisset necessaria
uideri esse sestertia ferme trecenta, unus ex amicis Frontonis:
'et praeter propter' inquit 'alia quinquaginta' ...
6: 'non meum' inquit 'hoe uerbum est, sed multorum
hominum, quos loquentis id audias;
7: quid autem id uerbum significet, non ex me, sed ex
grammatico quaerundum est' ...
11 : atque ibi Iulius Celsinus admonuit in tragoedia quoque
Enni, quae Iphigenia inscripta est, id ipsum, de quo quaere-
batur, scriptum esse et a grammaticis contaminari magis
solitum quam enarrari.

XCVlD (a) iphigeniae F 194 o fide om. F etfortasseFestus

III
 THE FRAGMENTS

12: quocirca statim proferri Iphigeniam Q. Enni iubet. in

eius tragoediae choro inscriptos esse hos uersus legimus:
otio qui nescituti 195
plus negotihabetquamcum est negotiumin negotio.
nam cui quodagatinstitutumest fin.illist negotium,
id agit,( id) studet,ibi mentematqueanimumdeleetatsuum;
totioso initiot animusnescitquiduelit.
hoe idemest: em nequedominuncnos neemilitiaesumus. 200

imus hue, hinc illuc; cum illueuentumest, ire illinelubet.

ineertee"at animus,praeterpropteruitamuiuitur.
13 : ... 'audistine ', inquit, 'magister optime, Ennium tuum
dixissepraeter propter et cum sententia quidem tali quali seueris-
simae philosophorum esseobiurgationes solent? petimus igitur,
dicas, quoniam de Enniano iam uerbo quaeritur, qui (uulgo:
quid eodd.) sit remotus (Hosius: motus eodd.)huiusce uersus
sensus: U• mcerte errat ammus
•
praeter propter wtam
•
wwtur
• • "'
.

C
Julius Rufinianus, Rhet. 11, p. 41 . 28: ayavCXKTfl015,
indignatio,
quae fit maxime pronuntiatione. Ennius in Iphigenia:
Menelausme obiurgat;id meisrebusregimenrestitat. 203

Cl
Iulius Rufinianus, Rhet. 37, p. 47. 16: cruyKp1a15
siue &\ni8ea1s,
comparatio rerum atque personarum inter se contrariarum, ut
egoproiectorquodtu peccas.tu tdelinquast egoarguor. 204

pro malejactisHelenaredeat,uirgopereatinnoeens?
tua reeoncilietur
uxor, meanecetur}ilia?
XCIX 196 negotii codd. 197 in illis 6: in illo y negotio TTX 198 (id)
studet Ribbeclt 199 otioso in otio Lipsius 200 nee uulgo:de wdd. 201
hunc 6 hinc illinc Z: illuc hinc Q ire illuc Q lubct .Btroaldus:iubet
codd. 202 uita Salmasius
C 203 rcstitat Bentley: rcstat ed. Basil.
Cl 204 delinquis R. Stephanus 206 nccctur filia Columna:negetur £ilia mca
ed. Basil.-
112
 MEDEA EXVL; MEDEA

CII
Servius auctus, Aen. 1 . 52: sane uasto pro desolato ueteres
ponebant. Ennius Iphigenia:
quaenuncabste uiduaeet uastaeuirginessunt. 207
ponebant et pro magno. Clodiw Commentariorum: 'uasta
. . ,
mama magna.

MEDEA EXVL; MEDEA

CIII
(a) Rhetor incertus, Her. 2.34: item wttosa expos1t10 quae
nimium longe repetitur ... hie id, quod extremum dictum est,
saris fuit exponere, ne Ennium et ceteros poetas imitemur,
quibus hoe modo loqui concessum est:
utinamne in nemorePeliosecuribus 208

caesaacddissetabiegnaad terramtrabes,
neue indenauisinchoandiexordium 210

cepisset,quaenuncnominaturnomine
Argo, quiaArgiui in ea delectiuiri
uectipetebantpellem inauratamarietis
Colchis,imperioregisPeliae,per dolum.
nam numquameraerransmeadomoejferretpedem 21 s
Medeaanimoaegroamoresaeuosauda.
cm (a) 208 utinam CE: uti iam M 209 cacsa accidisset abiegna scripsit
F.nnius:forwse mauit rhdor: cacsae codd. accidissent BTT://// / cidissent P 1 :
accedissentH: cecidissent J>ICE abiegn.aeEBITCP':ad ign~ H: abi~ P1
210 nauis E: naucs M inchoandi b: inchoandas H: inchoansas P 1 : inchoans
TT:inchoanda B: inchoandae CP'd: inchoandum I 211 cepisset M: coepis-
set bdH'TTC: caepisset BI 212 argu M cjui {qua PB) argiui in eadem lecti
uiri M: qua argiui delecti uiri E 214- pclire M 215 nam (urioam C)
numquam era (hera C) errans mea domo efferret pedem BC: nam numquam
era mea errans medea domo efferret pedem d: nam numquam hera errans mea
me<lea efferret pedem (pellem b) domo bi: nam numquam errans mea efferret
pedem M 216 non habentcodices
rhetoris:medea egro amore saucia in cod.P
marginem. a additum
8 113 JTO
 THE FRAGMENTS

nam hie satis erat dicere, si id modo quod satis esset curarent
(curassent E) poetae: 'utinam ne era errans (hera errans CBE:
erra ueras H: erra erra errans P: trans TI) mea domo efferret
pedem (M: mea efferret pedem domo E) Medea (Medeas
HP 1) animo (b: animii d: homo H: homodo P: om. TIBCI)
aegro (egro b)amore saeuo (amores reuo M: amore BE: om.TI)
. '.
sauc1a
(b) Cicero, Inu. 1 . 91 : remotum est quod ultra quam satis est
petitur ... huiusmodi est illa quoque conquestio: 'utinam ne in
nemore Pelio securibus caesae accidissent (H3: accedissent M:
concidissent S2i: cecidissentJ) abiegnae (P3 in marg.:om. C) ad
terram trabes (vv. 208-9) '. longius enim repetita est quam res
postulabat.
(c) Cicero, Cael. 18: ... conduxit in Palatio non magno do-
mum. quo loco possum dicere id quod uir clarissimus, M.
Crassus, cum de aduentu regis Ptolemaei quereretur, paulo
ante dixit: 'utinam ne in nemore Pelio (peleo P) (v. 208) -'
ac longius mihi quidem (V: quidem mihi Tirr) contexere hoe
carmen liceret: 'nam numquam era (eram GE) errans (v. 215)'
hanc molestiam nobis exhiberet 'Medea (mede P) animo aegro
(aegra H) amore saeuo saucia (v. 216)'. sic enim, iudices,
reperietis quod, cum ad id loci uenero, ostendam, hanc
Palatinam Medeam eamque migrationem huic (Kayser:
migrationemque huic codd.)adulescenti causam siue malorum
omnium siue potius sermonum fuisse.
(d) Cicero, Fin. 1 . 4: in quibus hoe primum est in quo admirer,
cur in grauissimis rebus non delectet eos sermo patrius, cum
idem fabellas Latinas ad uerbum e Graecis expressas non inuiti
legant. quis enim tam inimicus paene nomini (nomini pene BE)
Romano est, qui Enni (ennii codd.)Medeam aut Antiopam
Pacuui (pacuuii codd.) spernat aut reiciat, quod se isdem
Euripidis fabulis delectari dicat, Latinas litteras oderit? ...
114
 MEDEA EXVL; MBDBA

5 : ... an 'utinam (Muretus:at utinam ABERN: aut umnam

V) ne in nemore (v. 208)' nihilo minus legimus quam hoe
idem Graecum, quae autem de bene beateque uiuendo a Platonc
disputata sunt, haec explicari non placcbit Latinc?
(e) Cicero, Tusc. 1.45: ... qui ostium Ponti uiderunt et cas
angustias per quas penetrauit ea quae est nominata 'Argo, quia
Argiui in ea {in ea add.Kc) delecti (KVSB: dilecti GRV 1) uiri
uecti petebant pellem inauratam arietis (vv. 212-13)'.
(f) Cicero, Nat. deor.3. 75: utinam igitur ut ilia anus optat 'nc
in nemore Pelio securibus caesae accidissent (cecidissent .BI)
abiegnae (abiegnc Ifl: abigne APVB 1) ad terram trabes (vv.
208-9)', sic istam calliditatem hominibus di ne dedissent ...
(g) Cicero, Fat. 34: itaque non sic causa intellegi debet ut quod
cuique antecedat id ei causa sit sed quod cuique {antecedat id ei
(co". ex id et] ... cuique in mg. antiquissimusco"ectoradd.A: III
antecedat id, lituradel. efficientes B: id et causa V) efficienter
(efficientur V1B1) antecedat {antecedant APVP), nee quod in
campum descenderim (descenderint VP) id fuissecausae (causa
B) cur pila luderem (lauderem AP).. .ex hoe gcnere illud est
vc
Enni (ennii Be) 'utinam ne in nemore Pelio securibus accidis-
sent {B1 : accedissent V: cecidissent A2B2)abiegnae (abi•egnac
A: abiaegne V) ad terram trabes {trabas V1) (vv. 208-9)'.
licuit uel altius: utinam ne in Pelio nata u11aumquam esset
(esseBP) arbor. etiam supra: utinam ne esset (esse VP) mons
Pelius. similiterque superiora repetentem regredi infinite
licet. 'neue inde nauis inchoandi cxordium cepisset (.BP:cocpis-
set AV} (vv. 210-11)'. quorsum haec praeterita? quia sequitur
illud: 'nam numquam era errans mea domo ecferret (haec
ferret APVP: ha paene erasaeB) pedem Medca animo aegro
am.ore saeuo (am.ores aeuo BP: amores VP: more AP) saucia
(vv.215-16)'. fnon ut eae res (ut heres V 2) causam adfcrrcnt
amorist
115 8-2
 THE FRAGMENTS

(h) Cicero, Top. 61: hoe igitur sine quo non fit ah eo in quo
certe fit diligenter est separandum. illud enim est tamquam
'utinam ne in nemore Pelio (v. 208) '. nisi enim accidissent
(AaV: cecidissent OjbcdLf3) ahiegnae ad terram trahes, Argo ilia
facta non esset,nee tamen fuit in his trahihus efficiendi uis neces-
saria. at cum in Aiacis nauim crispisulcans igneum fulmen
iniectum est, inflammatur nauis necessario.
(i) Varro, Ling.7. 33: sic dictum a quihusdam ut una canes, una
trahes 'remis rostrata per altum '. Ennius: 'utinam ne in nemore
Pelio (polio F) securihus caesa accidisset (cesa saccidissent F)
ahiegna ad terram trahes (vv. 208-9)'. cuius uerhi singularis
casus rectus (Sdoppius:recte F) correptus (Laetus:correctus F)
ac facta trahs.
(k) ~intilianus, Inst. s .10. 84: recte autem monemur causas
non utique ah ultimo esse repetendas, ut Medea: 'utinam ne in
riemore Pelio (v. 208)'. quasi uero id eam fecerit miseram aut
nocentem, quod illic ceciderint ahiegnae (AB: ceciderit abiegna
P) ad terram trahes (v. 209).
(l) Iulius Victor, Rhet. 12, p. 41 s .24: uitiosum genus argumen-
tationis ... ut remotum, quod ulterius quam saris est petitur,
ut: si in nemore Pelio non cecidissent tr-ahes,hoe scelus factum
non esset.
(m) Donatus, Ter. Phorm. 157: VTINAM uetus
NE PHORMIO.
elocutio utinam (ne) (add. Stephanus),ut Ennius in Medea:
'utinam ne in nemore Pelio umquam sectae cecidissent (cecidis-
sent caesae V) ad terram trahes (vv. 208-9) '.
(n) Hieronymus, Epist. 127. s.2:
hanc multos post annos imi-
tata est Sophronia et aliae, quihus rectissime illud Ennianum
aptari po test 'utinam ne in nemore Pelio (v. 208) '.
(o) Priscianus, Gramm. n 320 . 1s: uetustissimi tamen etiam
trahes pro trahs proferehant. Ennius in Medea: 'utinam ne
116
 MBDEA BXVL; MBDEA

in nemorc Pelio securibus caesa (cessa GL) accedisset (K:

accidissct r: cecidisset h Coloniensis)abiegna in terram (in
terram abiegna codd.)trabes (vv. 208-9) '.
(p) Priscianus, Gramm.m 423 . 3s: nee solum comici huiusce-
modi sunt usi iambis sed etiam tragici uetustissimi, ut Ennius in
Medea:'utinam ne in nemore Pelio securibus caesae cecidissent
abiegnae ad terram trabes, neue inde (RVBa: neuenide A}
nauis inchoanda exordium cepisset ( V: coepisset A) quae nunc

cm *Pomponius, Atell. ap. Non. p. 516. 12 occidit taurum

toruiter, me amore sauciauit (Lipsius: me amores amauit coJJ.);
Catullus 64. 1-7 Peliaco quondam prognatae uertice pinus I dicuntur
liquidas Neptuni nasse per undas I Phasidos (fascidicos X: al.
phasidos add.nng: fasidicos 0) ad fluctus et fines Aeetaeos (uulgo:
ceticos 0: oeticos X), I cum lecti iuuenes, Argiuae robora pubis
(uulgo:pupis 0: puppis X), I auratam optantes Colchisauertere
pellem I ausi sunt uada ~ cita decurrere puppi, I caerula uerrentes
abiegnis aequora palmis, 171-2 utinam ne tempore primo I Gnosia
Cecropiae tetigissent litora puppes, 250 multiplices animo uoluebat
saucia curas; Lucretius 4. 1048 idque petit corpus mens unde est
saucia amore; Horatius, Epod. 16. 57-8 non hue Argoo contendit
remige pinus, I neque impudica Colchisintulit pedem; Vergilius,
Eel. 4. 34-5 et altera quae uehat Argo I delectos heroas, 8 .47-8
saeuus amor docuit natorum sanguine matrem I commaculare
manus; Phaedrus 4.7.4-C) parua (Rigault: par P) libellum sustine
patientia, I seueritatem frontis dum placo tuae I et in cothurnis
prodit Aesopus nouis (Pithoeus:nobis P). I utinam J!CC (Bongars:ne
PR) umquam Pelii (Salmasius:pelei P) nemoris iugo I pinus
bipenni concidisset Thessala I nee ad professae mortis audacem uiam
I fabricasset Argus opere Palladio ratem; Poeta incertus ap. Gell.
19.11.4 anima aegra (coJJ.:aegra amore Camerarius)et saucia I
cucurrit ad labeas mihi; Apuleius, Met. 4. 32 sed Psyche uirgo uidua
domi residens deflet desertam suam solitudinem aegra corporis,
animi saucia (uulgo:animis audacia coJJ.); 0rosius 1. 12. 10 nolo
m,.miniue. Medeae amore _saeuosauciae et pignorum paruulorum
caede gaudentis.
117
 THE FRAGMENTS

nominatur nomine Argo, qua uecti Argiui dilecti uiri petebant

illam pellem inauratam arietis Colchis imperio regis Peliae per
dolum. nam numquam era errans mea domo efferret pedem
(vv. 208-15)'.
CIV
Cicero, De orat. 3 .217: aliud enim uocis genus iracundia sibi
sumat, acutum, incitatum, crebro incidens ... aliud miseratio ac
maeror, flexibile, plenum, interruptum, fiebili uoce:
quonuncme uortam?quoditer indpiamingredi? 217

domumpaternamne?anneadPeliae.filias?

CV
(a) Cicero, Fam.7. 6 . 1 : tu modo ineptias istas et desideria urbis
et urbanitatis depone et, quo consilio profectus es, id adsiduitate
et uirtute consequere. hoe tibi tam ignoscemus nos amici quam
ignouerunt Medeae quae Corinthum arcem altam habebant
matronae opulentae optumates, quibus illa manibus gypsatis-
simis persuasit ne sibi uitio illae uerterent quod abesset a
patria. nam
multisuamrembenegessereet publicampatriaprocul; 219

multiqui domiaetatemagerentproptereasunt improbati.

CIV *C. Gracchus ap. Cic. De orat.3 .214 et Q!!intil. Inst. II. 3. IIS
quo me miser conferam? quo uortam? in Capitoliumne? at fratris
sanguine redundat (L: madet M). an domum? matremne ut
miseram lamentantem uideam et abiectam?; Catullus 64. 177-81
nam quo me referam? quali spe perdita nitor? I Idomneosne (E.
Fraenkel:idoneos (idmoneos X] ne V) petam montes? at (Muretus:
a V) gurgite lato I discemens ponti (0: pontum X) truculentum
ubi diuidit aequor? I an patris (R: impatris 0: in patris G) auxilium
sperem? quemne ipsa reliqui, I respersum iuuenem fraterna caede
secuta?
CIV 218 paternam L ad Peliae uulgo: ad Paeliae P: appellarc M

118
 MEDEA EXVL; MEDEA

2: quo in numero tu certe fuisses, nisi te extrusissemus. sed

plura scribemus alias. tu, qui ceteris cauere didicisti, in Britannia
ne ah essedariis decipiaris caueto et (quoniam Medeam coepi
agere) illud semper memento:
qui ipsesibisapiensprodessenon quit nequiquamsapit. 2:n

(b) Cicero, Off. 3 .62: nemo est qui hoe uiri boni fuisse neget;
sapientis negant, ut si minoris quam potuisset uendidisset. haec
igitur est illa pernicies, quod alios bonos, alios sapientes existi-
mant. ex quo Ennius nequiquam sapere sapientem qui ipse sibi
prodesse non quiret. uere id quidem, si quid esset prodesse
mihi cum Ennio conueniret.

CVI
Cicero, Tusc. 3 . 63 : sunt autem alii quos in luctu cum ipsa
solitudine loqui saepe delectat, ut ilia apud Ennium nutrix:
cupidocepitmiseramnuncmeproloqui 222
caeloatqueterraeMedeaimiserias.

CVII
Cicero, Tusc.4. 69: quid ait ex tragoedia princeps ille Argo-
nautarum?
tu me amorismagisquamhonorisseruauisti
gratia. 22.4

quid ergo? hie amor Medeae quanta miseriarum excitauit

incendia. atque ea tamen apud alium poetam patri dicere audet
se (s'V3: sed X) coniugem habuisse illum 'Amor quem dederat
qui plus pollet potiorque est patre '.

CV (a) 221 ncquicquam R

CVI 223 Mcdcai Tumebus: medcac codd.
CVD 22.4 tum amoris K: tum ea moris R scruauisti td. Cratandritul:scruasti
codd.

119
 THE FRAGMENTS

CVIII
Cicero, Nat. deor.3 . 65:
nequaquamistucistacibit; magnainestcertatio. 22s

nam ut egoilli supplicaremtantablandiloquentia

ni ob rem-
66: parumne ratiocinari uidetur et sibi ipsa nefariam pestem
machinari? illud uero quam callida ratione:
qui uolt quoduolt ita dat(semper)se res ut operamdabit. 228

qui est uersus omnium seminator malorum.

ille trauersamentemi hodietradiditrepagula 229

quibusegoiramomnemrecludamatqueilli pernidemdabo
mihi maerores,illi luctum,exitium illi, exilium mihi.
hanc uidelicet rationem, quam uos diuino beneficio homini
solum tributam dicitis, bestiae non habent.
67: uidesne igitur quanto munere deorum simus adfecti?
atque eadem Medea (Media A1 VB1 ) pattern patriamque
fugiens
'postquam (posquam A) pater
adpropinquat iamque paene ut conprehendatur parat,
puerum interea obtruncat membraque articulatim diuidit
perque agros passim dispergit corpus; id ea gratia
ut, dum nati dissipatos artris captaret parens,
ipsa interea effugeret, illum ut maeror tardaret sequi,
sibi salutem ut familiari pareret parricidio.'
huic ut scelus sic ne ratio quidem defuit.

CVIII S istac ibit B1 : isthacc ibit V1:istaibit A V1B'

22 226 illi Ribbtde:illis
codd. subplicarcm A blandiloqucnti codd. 227 ni ob rem Vahlm : aniobc
B: niobcm AV 2~8 qui qult A 1 V1B: qui uult esse paucidtttriores,Dauisius
1
quod uult A V1B ut adat V1 (semper) sc Jocelyn 229 transucna V1
1
230 pcmiciem V1: pemitiem A .B': permiicm A V1B1 1 231 cxitium V1.B':
cxitum AV1B 1
120
 MEDEA EXVL; MEDEA

CIX
(a) Varro, Men. fr. ap. Non. p . .261.7: non uides apud
(Ennium) (add. Aldus) esse scriptum 'ter sub armis malim
uitam cernere quam semel modo parere'?
(b) Varro, Ling. 6.81 : cemo idem ualet. itaque pro uideo ait
Ennius ... ah eodem est quod ait Medea:
namter subarmismalimuitamcernere 232

quamsemelmodoparere.
quod, ut decemunt de uita eo tempore, multorum uidetur uitae
finis.
(c) Nonius, p . .261. 18: CBRNBRB rursum dimicare uel conten-
dere ... Ennius in Medea exule: 'nam ter (te codd.)sub armis
malim (maculim L 1) uitam (E1 : uita codd.)cemere'.

ex
Probus, V erg. Eel. 6.31-3 (namque canebat uti magnum per
inane coacta I semina terrarumque animaeque marisque fuissent
I et liquidi simul ignis) : ... Aemilius Asper cum hunc locum
(Aen.6. 7.24-6) adnotaret sic ait: 'haec membra naturae sic solet
iungere ut in tria diuidat. nam et alibi: "maria ac terras caelum-
que profundum (Aen. 1. 58)" et Homerus similiter: w ~
yatavrnvf, w6' ovpav6v, w & 66Aaaaav (II. 18.483)'. sed et
Homerum ipso hoe loco possumus probare quattuor elemen-
torum mentionem fecisse. nam w ~ yaiav rnv~ev Achilles
significatur, ut homo terrenus, cui arma fiebant, w 6" ovpavov
aerem scilicet in quo ista fiebant: ignem extrinsecus adhibebi-
mus in ipso Volcano. similiter et Ennius in Medea exule in his
uersibus:

CDC (b) 232 nam ,um lu,btt F: jortdsserumattulit Va"o multa ccmcre F

1.21
 THE FRAGMENTS

IuppitertuqueadeosummeSol qui resomnisinspicis 234

quiquetuo luminemareterramcaelumcontines
inspicehoef acinusprius quamfit. prohibessis
scelus.
nam (Keil:iam codd.)et hie Iuppiter et Sol pro igni, qui mare et
terram et caelum continet, (ut) (add.KeiQnon dubie (dubium
MP) caelum pro aere dixerit.

CXI
(a) Nonius, p. 38.29: BLIMINARB,extra limen eicere ... Ennius
Medea exule:
antiquaerilisfida custoscorporis, 237
quidsic te extraaedisexanimatameliminat?
(b)Nonius, p. 292.20: BLIMINARBestexire. EnniusMedeaexule:
'antiqua erilis (edilis codd.)fidacustos corporis quid tsitt extra
aedis exanimata (examinata [examinate L1 ] codd.)eliminat'.
Accius Phoenissis...
CXII
(a) Varro, Ling. 7.9: in hoe templo faciundo arbores constitui
fines apparet et intra eas regiones qua oculi conspiciant, id est
tueamur, a quo templum dictum, ut contemplare, ut apud
Ennium in Medea: 'contempla et templum Cereris ad laeuam
. '
asp1ce.
(b) Nonius, p. 469. 34: CONTBMPLA ••• Ennius Medea:
astaatqueAthenasanticumopulentumoppidum 239
contemplaet templumCererisad laeuamaspice.
CX 236 facimw V fit VE: sit MP prohibcssis Bothe:prohibcsse V: prohibe
ccP: prohibe prohibe esse M: prohibe E
CXI (a) cnnium et ea cxulc L 1 : cnniw metca cxulc L 1 : cnniw mcde cxu1eG
238 sic tc Mercerus:sit codd. cnnimatam climinat]octlyn: cxanimata climinas
Mercerus:cxaminata climina codd.
CXII (b) 239 anticum Roth: antiquum uulgo: anti cum codd. opulentcum
L0 4 2-40 et tcmplum Ccrcris ad lacuam aspicc non lu,bmt codius:fortasse
omisitNonius
122
 MBLANIPPA

CXIII
Nonius, p. 84.31: CBITB significat dicite ucl date; ab co quod
cedo ... Ennius in Medca:
salueteoptimacorpora.
cettemanusuestrasmeasqueaaipite.

CXIV
Nonius, p. 170.8: SVBLIMARB, cxtollere. Ennius Medea:
sol qui candentem
in caelosublimatfacem. 2,u

CXV
Nonius, p. 297. 16: BFFBllB significat profcrrc ... Ennius
·Medea:
utinamneumquampnedet cordiscupidocordepedemextulisses2.4-4

CXVI
Nonius, p. 467. 7: AVCVPAVI, actiuum positum pro passiuo...
Ennius Medca:
ftuctus uerborumauresaucupant. 24S

MELANIPPA
CXVII
Cicero, Off. I. 114: uideJr.x.

CXVIII
Gellius 5 • 11 . 11-14: ' ... inter enim pulcherrimam fcmioam et
deformissimam media forma quaedam est, quae et a nimiae
pulcritudinis periculo et a summae deformitatis odio uacat;

CXV 2.4-4McdeaColchisLipsius
123
 THE FRAGMENTS

qualis a ~nto Ennio in Melanippa perquam eleganti uocabulo

stata (PR: tecta V) dicitur, quae neque KOt'"I futura sit neque
,roivft '. quam formam (P2 : quam in formam codd.)modicam et
modestam Fauorinus non mi hercule inscite appellabat
'uxoriam'. Ennius autem in ista quam dixit (dixi R) tragoedia
eas fere feminas ait incolumi pudicitia esse quae stata forma
forent.
CXIX
(a) Nonius, p. 170. 10: SVPBRSTITBNT(superstent DA), saluent.
Ennius Melanippa:
regnumque
nostrumut sospitentsuperstitentque. 246

(b) Nonius, p. 176.2: sosPITBNT,saluent. Ennius Melanippa

(menalippa G): 'regnumque nostrum sospitent superstitentque '.

CXX
Nonius, p. 246. 9: AVSCVLTAREest obsequi ... Ennius Melanippa:
mi auscultanate:pueroscremariiube. 247

CXXI
Nonius, p. 469. 3 : AVGVRO••• Ennius Melanippa:
tcerto hie est nullat quin monstrumsiet.
hoeegotibi dicoet coniectura
auguro.

CXXII
Macrobius, Sat. 6. 4. 7 (ef. fr. LXII): 'splendet tremulo sub
lumine pontus (Vergilius, Aen. 7.9)'. TREMVLVM LVMEN de
imagine rei ipsius expressum est. sed prior Ennius in Melanippe:
luminesic tremuloterraet cauacaerulacandent.
CXIX {a) 246 supcntitcnt Addalius
CXX mcnalippa a,dd. 247 mi Vossius:mihi a,dd. crcmitari Bothe
CXXI mcnalippo {melanippo L1) · codd. 248 certo hie est nullum B~ :
certatio hie est nulla Passeratius
124
 PHOENIX

CXXIII
(a) Gellius 6. 9 . 1 s: praeterea inueni a uerbo scindo simili
ratione non sciderat, sed sciciderat dictum esse...
17: Ennius quoque ( • • •)
(b) Priscianus, Gramm. 11 516.14: scindo scidi. uetustissimi
tamen etiam scicidi (sciscidi R) proferebant ... Ennius m
Melanippa:
cum saxumsciciderit 251

NEMEA
CXXIV
Nonius, p. 183. 14: VENOR, circumuenior. Ennius Nemea:
teneorconsaepta,undiqueuenor. 252

CXXV
Priscianus,Gramm.11 171 . 4: hie et haecet hoe pecus. Enniusin
Nemea:
pecudidareuiuammarito. 253

potest tamen figurate hoe esse prolatum, ut si dicam 'aquila

maritus' uel 'rex auium '.

PHOENIX
CXXVI
Gellius 6. 17. 1: percontabar Romae quempiam grammaticum
... discendi magis studio et cupidine, quid significaret ob-
noxius quaeque eius uocabuli origo ac ratio esset....

CXXID (b) menalippa BD 251 scisciderit R

CXXIV 252 conscpta (conccpta G) codd.
CXXV nemia L: neinea G 253 pecudi t;c pecu1ico". K: pecodi G wua
Drtsd. 1 .z Erl. 1 .z Lips. z Krthlii: uerba Palmer
I2S
 THE FRAGMENTS
10: iam uero illud etiam Q. Enni quo pacto congruere tecum
(congrueret equum codd.)po test, quod scribit in Phoenice in his
uersibus?
seduirumuerauirtuteuiueretanimatum adiecitt 254
fortiterquetinnoxium uocaret aduersumaduersarios.
ea libertasest quipectuspurum etfirmumgestitat;
t aliaet resobnoxiosae noctein obscuralatent.

CXXVII
(a) Helenius Aero, Ter. Ad. 45-6 (semper parce ac duriter I se
habere) ap. Charisium, p. 257 .6: secundum antiquorum con-
suetudinem. nam et Ennius in Phoenice:
quamtibi ex oreorationemduriterdictisdedit. 258

(b) Nonius, p. pro dure ... Ennius Phoenice

512. 1: DVRITBR
(foenice G: fenice LH): 'quam tibi ex ore orationem duriter
dictis dedit '.
CXXVIII
Nonius, p. 91. 4: CVPIBNTBR, cupidissime ... Ennius Phoenice:
tstultus est qui cupidat cupienscupientercupit. 2 59

CXXIX
Nonius, p. 245. 30: ARGVTARI dicitur loquacius (H3: loquacium
codd.)proloqui. Ennius Phoenice:
ttum tu isti crederet atque exercelinguamut argutarier
possis. 260

CXXVI phonice codd. 254 animatum addecet Carrio 2ss innoxium

uacare P: innoxium stare Bentlty 257 obnoxiosae Addalius: obnoxiose V:
obnoxio se P: obnoxie se R
CXXVII (a) poinice N
CXXVDI foenice codd.
CXXIX Ennius Phoenice 'tum MtrctrUS:ennius quo enicetum codd. 26o
tum tu isti credc te Haupt: tum tu isti crcde L1 T~bus lingua LAA.&'
 TBLAMO

CXXX
Nonius, p. 507. 22: FAXIM,fecerim. Ennius Phoenice:
plus misersim si scelestum
faxim quoddicamfore. 261

CXXXI
Nonius, p. 510.32: SABVITBRpro saeue ... Ennius Phoenice:
saeuitersuspidonem
ferrefalsamfuttilum est. .262

CXXXII
Nonius, p. 514.12: FVTTILB, futtiliter. Ennius Phocnice:
ut quodfactumestfattile amid uosferatisfortiter. .263

CXXXIII
Nonius, p. 518. 3 : DBRBPBNTB••• Ennius Phoenice:
ibi tum derepenteex altoin altumdespexitmare.

TELAMO
CXXXIV
(a) Cicero, Nat. deor.3.79: Telamo autem uno uersu locum
totum conficit curdi homines neglegant:
namsi curent,benebonissit,·malemalis;quodnuncabest. .26s
(b) Cicero, Diu. 132: nunc illa testabor non me sortilegos
I.
neque eos qui quaestus causa hariolentur, ne (~: nee AVB)

CXXX fcnicc (foinicc &ml,.) codd. 261 sim Dtlrius: sum codd.scelcstum
d. prim:.: scclcstim (scclcstcmAA) coda.
CXXXI fonicc LG: focnicc H&mb. .262futtulum ~
CXXXB focnicc codd. .263amic:icodd.:a me id 0. Sltutsch:a me Butchtler
fcratis d. print. : fucratis codd.
CXXXlll focnicac LGH: phcnicc Bamb.

127
 THE FRAGMENTS

psychomantia quidem quibus Appius, anticus tuus, uti solebat,

agnoscere; non habeo denique nauci (Marsus:non ab eodem
sanci AV: non ab eodem sancti BP: non ab eoclem sanxi Be)
. Marsum augurem, non uican~s haruspices, non de circo astro-
logos, non lsiacos (isiagos BP) coniectores, non interpretes
somniorum; non enim sunt hi aut scientia aut arte diuini sed
superstitiosi uates inpudentesque harioli,
aut inertesaut insaniaut quibusegestasimperat, 2.66
qui sibi semitamnon sapiuntalterimonstrantuiam;
quibusdiuitiaspollicentur,ab iis drachumamipsipetunt.
de his diuitiissibi deducantdrachumam,reddantcetera.
atque haec quidem Ennius, qui paucis ante uersibus esse deos
censet, sed eos non curare opinatur 'quid agat humanum genus.'
ego autem, qui et curare· arbitror et monere etiam ac multa
praedicere, leuitate, uanitate, malitia exclusa (~: exclusam AVB)
diuinationem probo.

(c) Cicero, Diu. 2. 104: primum enim hoe sumitis: 'si sunt di,
benefici in homines sunt '. quis hoe uobis dabit? Epicurusne, qui
negat quicquam (••quicquam AB: atquicquam VP) deos nee
alieni curare nee sui? an noster Ennius (nostert AVP: noster-
renius BP),qui magno plausu loquitur adsentiente populo:
egodeumgenusessesemperdixi et dicamcaelitum, 270

sedeosnon
.
curareopinorquid
.
agathumanumgenus.
et quidem cur sic opinetur rationem subicit, sed nihil est
necesse dicere quae secuntur; tantum sat est intellegi, id sumere
istos pro certo quocl dubium controuersumque sit.

CXXXIV (b) 268 ab iis LAmbinus:ah his AVB dracmam A: dragmam VW:
drachman BP 269drachumam AVPBP:dracmam V': dragmam B'
CXXXIV (c) 271 curarare B

128
 TELAMO

CXXXV
Festus, p. 218.2: OBSIDIONBM potius dicendwn esse quam ob-
sidiwn adiuuat nos testimonio suo Ennius (pacuuius X) m
Telamone quom (cwnX) ait:
scibasnatumingenuumAiacemcui tu obsidionem
paras. 272

CXXXVI
Nonius, p. 85 .20: CLARET,
clara est uel clareat (uel dare CAJJA)
.
. . . Ennius Telamone:
nam ita mihi TelamonispatrisatqueAead et proauiIouis 2 73
tgratia ea estt atquehoelumencandidumclaretmihi.

CXXXVII
Nonius, p. 159. 38: PORCBT significat prohibet ... Ennius
Telamone:
deumme sentitf acerepietas,ciuiumporcetpudor. 275

CXXXVIII
(a) Nonius, p. 172.19: SQVALAM pro squalidam. Ennius
Telamone:
strataterraelauerelacrimisuestemsqualamet sordidam. 2 76

(b) Nonius, p. 503 .38: pro lauat ... LAVERB inde trac-
LAVIT
tum est ... Ennius Telamone: 'strata terra (terrae EP) lauere
Qabere codd.) lacrimis uestem squalidam (squalem P 1 ) et
sordidam'.

CXXXV 272 ingcnuum ed. princ.: ingcnium codd.

CXXXVI 273 atquc codd.:aui Bergk Acaci et Bergk: facict codd. touis L
CXXXVIII talamonc codd.

9 129 JTO
 THE FRAGMENTS

CXXXIX
Nonius, p. 475 .20: PARTIRBT pro partiretur ... Ennius Tela-
mone:
eandemme in suspidonemscelerispartiuitpater. 277

CXL
Nonius, p. 505. 35: AVDIBO pro audiam. Ennius Telamone:
moreantiquoaudiboatqueauristibi contrautendasdabo. 278

CXLI
Diomedes, Gramm. I 382. 10: cui enim in dubium cadit quin
abnuo abnuis dicamus? uerum apud uetcres et (Putschius:est
codd.) abnueo dictum annotamus, ut Ennius ... idem (BM:
item A) in Telamone (M: telamonem AB) ex eo futurum:
abnuebunt 279

TELEPHVS
CXLII
(a) Festus, p. 128.24: MVTTIRE (mutire codd.),loqui. Ennius in
Telepho:
palammuttireplebeiopiaculumest. 280

(b) Phaedrus 3, epil. 33:

ego quondam legi quam puer (Pithoeus:pueri PR)
sententiam
'palam muttire plebeio periculum est,
dum sanitas constabit pulchre meminero (Rittershausen:
memini P: memin. R).
CXXXIX 277 eamdcm codd. in om. AA: in me Dtlrius
CXL telemone codd. 278 atque CA Monttptss. Oxon.: nequc Lfd&f Paris.
7665 utendos CADA
CXLI 279 abnuebant B: abnueb M
CXLD (a) 280 muttirc (mutirc IR) codd.Pauli: mutirc codd.Ftsti
130
 TELEPHVS

CXLIII
(a) Festus, p. 440. 35:
cultum et sord
tum quod proximae
scium accedit; iQ
diti paludum squ
in Telepho: 'quam ue
Stoia'. SQVARROSOS

(b) Nonius, p. 537.23: STOLAM ueteres non honestam uestem

solum, sed etiam omnem quae corpus tegeret. Ennius Telepho:
tcedo et caueo cum uestitust squalidasaeptusstola. 281

idem in eadem:
regnumreliquisaeptusmendidstola. 282

CXLIV
N onius, p. 15 . 3 : BNODA significat explana; et (ex CJD1) quae
sit proprietas, manifestum est, hoe est, nodis exsolue ...
Ennius Telepho:
uerumquorumliberiletodati 283
sunt in bellonon lubenterhaecenodariaudiunt.

CXLV
Nonius, p. 232. 17: ADVORSVM rursum apud significat ...
Ennius Telepho:
teipsumhoeoportetprofiteriet proloqui 285
aduorsumtillam mihi t
CXLID (a) SQVAUDVM, incultum et sordidum; quod proximc similitudincm
habcat squama piscium, sic appcllatum - Paulus.
CXLm (b) tclcfo codd. 281 ccdo et cauco conucstitus Colunuu, 282
scptus codd. mcndici Iunius: mcdici codd.
CXLIV tclcfo codd. 28 3 laeto L
CXL V tclefo codd.
131
 THB FRAGMENTS

CXLVI
Nonius, p. 342. 6: MACTARB malo adficere significat ... Ennius
Telepho:
qui iliumdi deaequemagnomactassint
ma(o.

CXLVII
Nonius, p. 429. 1 : inter urbem et ciuitatem hoe interest.
urbs est aedificia (aedificatio H1), ciuitas incolae ... Ennius
Telepho:
sedduitatemuideoArgiuumincendere. 288

CXLVIII
Nonius, p. 490. 10: ITINER pro iter. Ennius Telepho:
deumquede consiliohoeitinercredoconatummodo.

THYESTES
CXLIX
(a) Cicero, De orat.3. 164: nolo esse uerbum angustius id quod
translatum sit quam fuissct (quam fuisset om.M) illud (ilium M)
propnum ac suum:
quidnamest obsecroquodte adiriabnutas?
melius esset uetas, prohibes, absterres; quoniam ille dixerat
'ilico istic. ne contagio mea bonis umbraue obsit (vv. 293-4)'.
(b) Cicero, Tusc. 3. 25: nunc aegritudinem si possumus depel-
lamus ... taetra enim res est, misera, detestabilis...

CXL VI tennius telefo ,odd. 287 dii codd.

CXL VIl telefus et LAA: telefo et BA 288 incendier Aldus
CXLVIIl telefo (telepo L 1) codd.
CXLIX (a) 290 quod M: quid L adiri L: abire M: adirier Vahlen 1854

132
 -THYESTES
26: qualis cnim tibi ille uidetur?
TantaloprognatusPelopenatusqui quondama socru 291
OenomaoregeHippodameam raptisnanctusnuptiis.
Iouis iste quidem pronepos. tamne ergo abiectus tamque
fractus?
nolite
inquit
hospitesad me adire.ilicoistic. 293
ne contagiomeabonisumbraueobsit.
tantauis scelerisin corporehaeret.
tu te (tune .R1) Thyesta damnabis or ha bisque luce propter uim
sceleris alieni?
CL
(a) Cicero, Pis. 43: neque ucro ego, si umquam uobis mala
precarer, quod saepe feci, in quo di immortales meas preces
audierunt, morbum aut mortem aut cruciatum precarer.
Thyestea est ista exsecratio, poetae uolgi animos, non sapientium,
mouentis, ut tu naufragio expulsus uspiam 'saxis furus asperis,
euisceratus' latere penderes, ut ait ille, 'saxa spargens tabo sanie
et (et om. W)sanguine atro '.

(b) Cicero, Tusc.I . 1o6: exsecratur luculentis sane uersibus apud

Ennium Thyestes, primum ut naufragio pereat Atreus; durum
hoe sane; talis enim interitus non est sine graui sensu; illa inania
ipsesummissaxisfixus asperis,euisceratus, 296
laterependens,saxa spargenstabosanieet sanguineatro.

CXLIX *Cicero, Verr.2. 1 • 8I nisi tanta acerbitas iniuriae, tanta uis

sceleris fuisset.
CL *Lucilius ap. Non. p. 405 . 3 latere pendens, saxa spargens tabo
sanie et sanguine atro.
CXLIX (b) 291 socru Bentley: soccro codd. 292 hippodamiam R.1 nactus
KR.': nactu'st Bentley 293 illic o G1 KV1 (1 exp. GIVI) 295 (mco)
tanta Bentley
133
 THE FRAGMENTS

non ipsa saxa magis sensu omni (omnia X: a exp. V 1 ) uacabwit

quam ille latere pendens cui se hie cruciatum censet optare. quae
( V2: quam X) essent dura si sentiret (s': sentirent X), nulla
(swit) (add.s') sine sensu. illud uero perquam inane
nequesepulcrumquo redpiathabeat,portumcorporis, 298
ubi remissahumanauita corpusrequiescat
malis.
uides quanto haec in errore uersentur. portum esse corporis et
requiescere in sepulcro putat mortuum. magna culpa Pelopis qui
non erudierit £ilium nee docuerit quatenus esset quidque
curandum.
CLI
Cicero, Orat. 183: in uersibus res est apertior, quamquam
etiam a modis quibusdam cantu remoto soluta esse uidetur
(uulgo:uideatur AL) oratio, maximeque id in optimo quoque
eorum poetarum qui Avp1Kola Graecis nominantur, quos (L:
eos A) cum cantu spoliaueris nuda paene remanet oratio.
184: quorum similia swit quaedam etiam apud nostros {L:
illos A), uelut illa {A:ille L) in Thyeste:
quemnamte essedicamqui tardain senectute- 300

et quae sequuntur; quae, nisi cum tibicen accessit, orationis

sunt solutae simillima.
CLII
Cicero, Brut. 78: C. Sulpicius Gallus ... hoe praetore ludos
Apollini faciente, cum Thyesten fabulam docuisset Q. Marcio
Cn. Seruilio consulibus, mortem obiit Ennius.

CLIII
(a) Cicero, Nat. deor.2.4: quid enim potest esse tarn apertum
tamque perspicuum, cum caelum suspeximus caelestiaque con-
templati sumus, quam esse aliquod numen {nomen B1 ) prae-

CLI 300 quin L senecta Bothe

134
 THYESTES

stantissimae mentis quo haec regantur? quod ni ita (quot nuta

B1) esset, qui potuisset adsensu omnium dicere Ennius 'aspice
hoe sublime candens quem inuocant (inuocat B1 ) omnes Iouem'
- ilium uero et Iouem et dominatorem rerum et omnia motu
regentem. et, ut idem (em• A) Ennius, 'patrem diuumque
h.ominumque' et praesentem ac praepotentem deum?
(b) Cicero, Nat. deor. 2. 64: sed ipse luppiter, id .est iuuans
pater, quem {V: partem quae B1 : partemque A} conuersis
casibus appellamus a iuuando (adiuuando B1) Iouem, a poctis
'pater diuomque hominumque' dicitur ...
65: hunc igitur Ennius, ut supra dixi, nuncupat ita dicens
'aspice hoe sublime (suplimim B1 ) candens, quem inuocant
omnes Iouem' planius (planus A 1) quam (PB: quem AHV 1 :
que Vi) alio loco idem 'tcwt quod in me est exsecrabor hoe
quod lucet (licet V} quicquid est (v. 342)'.
(,) Cicero, Nat. deor.3. 10: primum fuit, cum caelum suspexis-
semus, statim nos intellegere esse aliquod numen quo haec
regantur. ex hoe illud etiam 'aspice hoe sublime candens
(cadens A1 V} quem inuocant omnes Iouem'.

(d) Cicero, Nat. deor.3 .40: sit sane deus ipse mundus (mundus
deus ipse P). hoe credo illud esse 'sublime (sublime esse P}
candens quem inuoeant (inuoeans A1 ) omnes Iouem '.
(e) Festus, p. 400. 17: SVBLIMBM est in altitudinem elatum, ut
( • * *) Ennius in Thyeste:
aspicehoesublimecandensquemuocantomneslouem. 301

(f) Apuleius, Mund. 33: namque habitus orantium sic est, ut

manibus extensis (in) (uu~o: extensis codd.)caelum precemur.
Romanus etiam poeta sic sensit: 'aspice hoe sublime (BF: sub

CLm (t) SVBUMBM ( MPEGR: sublime TI: sublimcn L) est in altitudinem

clatum, id autem dicitur a limine superiorequia supranos est - Paulus

135
 THE FRAGMENTS

lumine PL} candens {B: cadens 6) quern inuocant (inuocans B}

omnes Iouem'. unde illa quae uidentur suntque omnibus prae-
stantiora, easdem sublimitates regionum tenent ...
(g) Probus, Verg. Eel. 6. 31-3: principem habuerunt Empedo-
clem Agrigentinum qui de his ita scribit: maapa 611 nav-rCA>v
~13~µaTCX npi:>Tovfaa1v Zrosaf)YTIS.•• ut accipiamus Zrosaf)YTIS
ignem, qui sit 3K\>vet candens, quod ignis est proprium, de quo
Euripides ... et Ennius: 'aspice hoe sublime candens quern
uocant omnes louem '.
CLIV
Nonius, p. 90. 13 : CONGLOMBRARB, inuoluere, superaddere.
Ennius Thyeste:
eheumeafortunaut omniain me conglomeras
mala. 302

CLV
Nonius, p. 97. 29: DBLBCTARB, inlicere, adtrahere. Ennius
Thyeste:
set me Apollo ipsedelectatductatDelphicus. 303

CLVI
Nonius, p. 110.11: FLACCBT, languet, deficit ... Ennius
Thyeste:
sinjlaccebuntcondidonesrepudiatoet reddito.

CLVII
Nonius, p. 255 .25: CRBPARB,ferire. Ennius in Thyeste:
sedsonitusaurismeaspedumpulsu increpat. 305

CLIV Ennius in Thiestc td. a. 1476: cnnius tithe codd. 302 chcu mca
Lachnumn:cumca codd.
CLV Thycstc set Mtrarus; thcestcs et codd. 303 dclficus codd.
CL VI 304 sin ftacccbunt Gulitlmius:inftacccbunt codd.

136
 INCERTA

CLVIII
Nonius, p . .261.13 : CBRNBRB,iudicare ... Ennius Thyeste:
impetremfacileab animout cernattwtalem babium t 306

CLIX
Nonius, p. 268. 9: CONTINGBRB, euenire ... Ennius Thyeste:
quammihi maximehie hodiecontigeritmalum. 307

CLX
Nonius, p. 369. 29: PVTARB, animo disputare ... Ennius Thyeste:
ibi quidagatsecumcogitatparatputat. 308

INCERTA
CLXI
(a) Terentius, Eun. 586-91:
egomet quoque id spectare coepi et, quia consimilem luserat
iam olim ille ludum, inpendio magis animus gaudebat mihi.
deum sese in hominem conuortisse atque in alienas tegulas
uenisse clanculum ! per inpluuium fucum factum mulieri !
at quern deum ! qui templa caeli summa sonitu concutit.
ego homuncio hoe non facerem?
(b) Donatus: QVI TBMPLA CABLI SVMMA (summo T) SONITV
CONCVTIT. ah auctoritate personae (persona TC) ut fit (sit C) in
exemplis. SONITV CONCVTIT. parodia de Ennio. TBMPLA CABLI
SVMMA. tragice (silma tragice V: sumairagica C: suma ragica T),
sed de industria, non errore.

CLVIII thyste codd. 306 cemet AA

CLIX tyeste codd.
CLX thieste codd. 308 cogitet LBA paritat Buechele,

137
 THE FRAGMENTS

CLXII
(a) Rhetor incertus, Her. 2.39: item uitiosum est cum id pro
eerto (E: quo id proeo P 1 : quid [u in ras.]praeco H: quae id
pro quo BC 1 ) sumitur quod inter omnes teonstat"t, quod etiam
nunc in eontrouersia (eontrouersia est blC: controuersia sit d),
hoe modo:
ehotu: di quibusestpotestasmotussuperumatqueinferum 309
paceminterseseconciliant,conferuntconcordiam.
nam ita pro suo iure hoe exemplo utentem (Kayser:usum E:
utuntur M) Thesprotum (M: threspontem E) Ennius induxit
quasi iam (E: quaesitam M) satis eertis rationibus ita (id ita HI)
esse demonstrasset.
(b) Cicero, lnu. 1.91: eontrouersum est in quo ad dubium
demonstrandum dubia causa adfertur hoe modo: 'eho tu: di
(P1 : dii C) quibus est potestas motus superum atque inferum
pacem inter seseconciliant,conferunt concordiam'.

CLXIII
Rhetor ineertus, Her.4. 18: eonpositio est uerborum eonstructio
... ea conseruabitur si ... et si uitabimus eiusdem litterae
nimiam adsiduitatem; eui uitio uersus hie erit exemplo - nam
hie nihil prohibet in uitiis alienis exemplis uti - : 'o Tite tute
Tati tibi tanta tyranne tulisti (ef.
Prise.Gramm.II 591 .12 et al.)'
et hie eiusdem poetae:
quicquamquisquamtquemquam t quemquequisqueconueniat
neget. 311
CLXII (a) 309 cho tu dii quibw E: studiis M est potcstas E: est potcst at H:
est potcst aut PB: cssc potcst aut TT:est potcstas ut C motw E: motum M
310 pace cnim M: paccm E scscM: scE
CLXIII 311 quicquam quisquam quemq; E: qui oquam quicquam qucmq
(qucmquam P) qucmq; P 1B: quicquam quisquam quemquam qucmquc P'C:
quicquam quicq quemq H: quicquam quicquam qucmq; TT
138
 IN CERT A

CLXIV
(a) Cicero, S. Rose. 89: uerum ego forsitan propter multi-
tudinem patronorum (~: patemorum I~) in grcge (~: gregem
I~) aclnumerer, te pugna Cannensis accusatorem sat 1,,onum
fecit. multos caesos non ad Trasumennum (trahasymennum I:
trasimennum ~) lacum sed ad Seruilium uidimus.
90:
quis ibi non est uulneratus
ferro Phrygio? 312
non necesse est omnis commemorarc Curtios Marios deniquc
Memmios ( Vrsinus:mammeos codd.)quos iam aetas a proeliis
auocabat, postremo Priamum ipsum senem, Antistium qucm
non modo aetas sed etiam leges pugnare prohibebant.
(b) Schol. Gronouianus: FBRRO FRVGIO. in Ennio haec fabula
inducitur, Achilles (Eberhard:achillis C) quo tempore propter
Briseidem cum Graecis pugnare noluit; quo etiam tempore
Hector classemeorum incendit. in hac pugna Vlixes uulneratus
inducitur et fugiens (ad) (add.Graeuius)Achillen uenit. cum
interrogaretur ah Aiace cur fugisset, ille ut celaret dcdccus
twtiumt: 'quis enim uulneratus ferro Frugio?'
CLXV
(a) Cicero, Balb. 36: in quo erat accusatons mterpretatio
indigna (uulgo: digna in codd.)responsione, qui ita dicebat
'comiter' esse 'communiter' quasi uero priscum aliquod aut
insolitum uerbum interpretaretur. comes (comites E) benigni
faciles suaues homines esse dicuntur; 'qui erranti (eranti P1 )
comiter (P: com [erasalineolasupra-m] G: comiti E) monstrat
{monstrant GE) uiam' benigne, non grauate; 'communiter'
quidem certe non conuenit.

CLXIV (a) 312 Brugio Osann

139
 THE FRAGMENTS

(b) Cicero, Off. 51: omnium (Zumpt: omnia codd.)autem

1.
communia hominum {ZLp: omnium c) uidentur ea quae sunt
generis eius quod ah Ennio positum in una re transferri in
permultas {ZL: multas ~) potest:
homoqui erranticomitermonstratuiam 313
quasilumendesuo lumineaccendat f acit.
nihilominusipsi lucetcum illi accenderit.
52: ... sed (et p) quoniam copiae paruae singulorum sunt,
eorum autem qui his egeant infinita est multitudo, uulgaris
liberalitas referenda est ad illum Enni (ennii codd.)finem 'nihilo-
minus ipsi lucet' ut facultas {ni[c]hilominus scilicet ut facultas
L 1 [c]: nihilominus ipsi luceat ut facultas L 2~) sit qua in nostros
simus liberales.
CLXVI
Cicero, De orat.1 . 199: quid est enim praeclarius quam honori-
bus et rei publicae muneribus perfunctum senem posse suo iure
dicere idem quod apud Ennium dicat ille Pythius Apollo, se
esse eum unde sibi, si non populi et reges, at omnes sui clues
consilium expetant suarum (suarum summarum codd. integri
pkrique: summarum codd.Lagomarsiniani) rerum incerti;
quosegomea opeex 316
incertiscertoscompotesque
consili
dimittout ne restemeretractentturbidas.
200: est enim sine dubio domus iuris consulti totius oraculum
ciuitatis.
CLXVII
Cicero, De orat.2. 221 : quocl est hominibus facetis et dicacibus
difficillimum, habere hominum rationem et temporum et ea
quae occurrunt {A1 : occurrant codd.), cum salsissime dici pos-

CLXV (b) 313 comiter ZL: comiti p: communiter c 31 s q nichilo c ipsi

luccat ZX: ipsi ut luccat ~
CLXVI 316 ope mca Reisig 317 compotes integripame omnes(Vahlen)
140
 INCERTA

sunt, tenere. itaque non nulli ridiculi homines hoe ipsum non
insulse interpretantur. dicere enim aiunt Ennium (L: enim
aiunt H: enim aut AE) flammam a sapiente (flammam a sapi-
ente L: pienti H [postxxxvi litt. sp.]AE) facilius ore in ardente
opprimi quam bona dicta tencat; haec scilicct bona dicta quac
salsa sint; nam ea dicta appellantur proprio iam nomine.

CLXVIII
(a) Cicero, De orat.3. 162: primum est fugienda dissimilitudo:
caeliingentesfornices. 319

quamuis sphaeram in scaenam, ut dicitur, attulerit Ennius,

tamen in sphaera fornicis similitudo inesse non potest.
(b) Varro, Ling. s.19: a chao cauum (Laetus:chouit F) et hinc
caelum ... et Ennius item ad cauationem: 'caeli ingentes
fornices'.
CLXIX
(a) Cicero, Rep. 1 . 49: concordi populo et omnia referente ad
incolumitatem et ad libertatem suam nihi1esse inmutabilius,
nihi1firmius ... itaque cum patres rerum potirentur numquam
constitisse ciuitatis statum; multo iam id in regnis minus
quorum, ut ait Ennius, 'nulla (Mai: nulla regni cod.) sancta
societas nee £ides est'.
(b) Cicero, Off. 1 .26: maxime autem adducuntur (ad hoe ad-
ducuntur p) plerique ut eos (uti eos p) iustitiae capiat obliuio
cum in (in om. p~) imperiorum honorum gloriae cupiditatem
inciderunt (Zp: inciderint cs').quod enim est apud Ennium,
nullasanctasodetasnee.fidesregniest, 320

id latius patet. nam quicquid eiusmodi est in quo non possint

(Z: possunt XK~) plures excellere, in eo fit plerumque (plerum-
quc fit p) tanta contentio ut difficillimum sit seruare sanctam
societatem.
 THE FRAGMENTS

CLXX
Cicero, Orat. Iss: idem poeta qui inusitatius contraxerat
'patris mei meum factum pudet' pro meorum factorum (if.fr.
xv I I b)... non dicit liberum ut plerique loquimur cum cupidos
liberum aut in liberum loco dicimus sed ut isti uolunt:
nequetuum umquamin gremiumextol/asliberorumex tegenus. 321
CLXXI
(a) Cicero, Drat. ISS: non dicit liberum ut plerique loquimur
... sed ut isti uolunt (ef.fr. CLXX) ••• et idem: 'namque
Aesculapi {L: excola••• A) liberorum (v. 326)'.
(b) Cicero, Tusc.2. 38: quin etiam uidemus ex acie efferri saepe
saucios, et quidem rudem illum et inexercitatum quamuis leui
ictu ploratus turpissimos edere. at uero ille exercitatus (exercitus
K 1) et uetus oh eamque rem fortior medicum modo requirens a
quo obligetur
o Patricoles
inquit ad uos adueniensauxiliumet uestrasmanus 322

petopriusquamoppetomalampestemmandatamhostilimanu-
nequesanguisullopotis estpactoprojluensconsistere
-
si qui sapientiamagisuestramorsdeuitaripotest; 32S

namqueAesculapiliberorumsauciiopplentporticus.
nonpotestaccedi.
certe Eurypylus hie quidem est ! hominem exercitum ! ubi
tantum luctus continuatur (s': luctum continuatus GKR) uide
quam non flebiliter respondeat, rationem etiam adferat cur
aequo animo sibi ferendum sit:
qui alteriexitiumparat,
eum sdre oportetsibiparatampestemut partidpetparem. 328

abducet (abducit V 2) Patricoles credo ut conlocet in cubili

CLXX 321 nequc tuum A: ncquc tu L
CLXXI (b) 326 esculapi K: aesculapii GR.V 328 exitium G'VS: cxitum
G1 V1 329 paratum Bentley
142
 INCERTA

( V2s':cubiculi X), ut uulnus obliget (non obliget K). si quidem

homo esset- sed nihil uidi minus (uidiminus vc: uidimus X).
quaerit enim quid actum sit: 330
eloque,eeloque,eresArgiuumproeliout se sustinet.

nonpotestecfaritantumdictisquantumf actissuppetit
laboris.
quiesce igitur et uolnus alliga. etiam si Eurypylus posset non
posset Aesopus. 333
ubiJortunaHectorisnostramacremacieminclinatam-
et cetera explicat in do lore; sic est enim intemperans militaris in
forti uiro gloria. ergo haec ueteranus miles facere poterit,
doctus uir sapiensque non poterit?

CLXXII
Cicero, Drat.16o: Burrum semper Ennius, numquam Pyrrhum;
u,. pate;,ecerunt
~r. Bruges, 334

non Phryges. ipsius antiqui declarant libri.

CLXXIII
Cicero, Fin. 2. 41 : nee uero audiendus Hieronymus, .cui sum-
mum bonum est idem quod uos interdum uel potius nimium
saepe (minimum sepe V: sepe minimum BE) dicitis, nihil
dolere. non enim si malum est dolor carere eo malo saris est ad
bene uiuendum. hoe dixerit potius Ennius:
nimiumboniest cui nihil est mali. 33S
nos beatam uitam non depulsio~e mali sed adeptione boni
iudicemus.
CLXXDI *Publilius Syrus, Sent. 430 nimium boni est in morte cui
nil sit mall.
331 ccfari V: hccfari K: haccfari GR 333 haectoris X inclinatam (dedit)
Ribbecle
CLXXD 334 Bruges Vidorius: frugcs FPM: phrugcs 0: phrygcs A
143
 THE FRAGMENTS

CLXXIV
Cicero, Tusc. 3 . s: at et morbi pemiciosiores pluresque sunt
animi quam corporis. hi enim ipsi odiosi sunt quod ad animum
pertinent eumque sollicitant,
animusqueaeger
ut ait Ennius
sempererrat; 336
nequepati nequeperpetipotest,cuperenumquamdesinit.

CLXXV
Cicero, Tusc. 3 . 39: quid ergo? huiusne uitae propositio et
cogitatio aut Thyestem leuare poterit aut Aeetam (aetam X:
oetam 1(2Rc),de quo paulo ante clixi,aut Telamonem pulsum
patria exulantem atque egentem ...
43 : ad hancine igitur uitam Telamonem illum reuocabis ut
. dinem ?....
1eues aegntu
44: ... quaerendum igitur quem ad modum aegritudine
priuemus (.K2R 2 V3:priuemur X) eum qui ita dicat:

pol mihifortunamagisnuncdefitquamgenus. 338

namqueregnumsuppetebatmi, ut sciasquantoe loco,
quantisopibus,quibusde rebuslapsafortunaaccidat.
quid? huic calix mulsi impingendus est, ut plorare desinat ...
ecce tibi ex altera parte ah eodem poeta: 'ex opibus summis opis
egens Hector tuae (if.ft. xxv11 h)'. huicsubueniredebemus ...
46: ... eripiamus huic aegritudinem ...

CLXXVI
Cicero, Tusc. 4. 70: quis est enim iste amor amicitiae? cur
neque deformem adulescentem quisquam amat neque for-
mosum senem? mihi quidem haec in Graecorum gymnasiis

CLXXV 339 neque K mi Grotius: mihi X

144
 INCERTA

nata consuetudo uidetur, in quibus isti liberi et concessi swit

am.ores. bene ergo Ennius:
jlagitiprincipiumest nudareinterciuiscorpora. 341
qui ut sint, quod fieri posse uideo, pudici, solliciti tamen et
anxii sunt, eoque magisquod se ipsi continent et coercent.

CLXXVII
Cicero, Nat. deor.2.64: sed ipse luppiter, id est iuuans pater ...
65 : hunc igitur Ennius, ut supra dixi, nuncupat ita dicens
'asp ice hoe sublime candens quem inuocant omnes Iouem'
planius quam. (PB: quem AHV1: que G V2) alio loco idem
tcwt quodin me est exsecraborhoequodlucetquicquidest. 342

CLXXVIII
Cicero, Diu. 1 . 88: Amphilochus et Mopsus Argiuorum reges
fuerunt sed iidem augures, iique urbis in ora marituma Ciliciae
Graecas condiderunt.atque etiam ante hos Amphiaraus etTiresias,
non humiles et obscuri neque eorum similes, ut apud Ennium est,
qui sui quaestuscausafictassuscitantsententias, 343
sed clari et praestantes uiri qui auibus et signis admoniti futura
dicebant.
CLXXIX
Cicero, Diu. 2. S7: Democritus quidem optumis uerbis causam
explicat cur ante lucem galli canant (B: cantant AV); depulso
enim de pectore et in omne corpus diuiso et mitificato (modifi-
cato V2) cibo cantus edere quiete satiatos; qui quidem silentio
noctis (noctis ex co". B), ut ait Ennius,
f auentf aucibusrussis 344
cantuplausuquepremuntalas.
CLXXVI 341 Bagitii X clues G(J)R"'
CLXXVII 342 qui Gul~lmius licct V
CLXXVID 343 ficta BI: factas V
CLXXIX 34S cantu veBI: cantus AV'W lausuque AvP
IO 145 JTO
 THE FRAGMENTS

CLXXX
Cicero, Diu, 2. 127: iam uero quis dicere audeat uera omnia
(AcB: ueras omnia AP: uera somnia V) esse somnia?
aliquotsomniauera
inquit Ennius sed omnianon necesseest. 346

CLXXXI
(a) Cicero, Off.1 .61 :itaquein probris (inprobisLpb)maximein
promptu est si quid tale dici po test: 'uos enim (etenim p<)iuuenes
animum geritis muliebrem, ilia uirgo uiri'; et si quid eiusmodi:
Salmacidaspoliasine sudoreet sanguine. 347

(b) Festus, p. 439. 10: SALMACIS nomine nympha, Caeli et Ter-

rae £ilia,fertur causa fontis Halicarnasi aquae appellandae fuisse
Salmacidis (uulgo: salamcidis F); quam qui bibisset uitio
inpudicitiae mollesceret. oh eam rem tque idt eius aditus,
angustatus parietibus, occasionem largitur iuuenibus petulanti-
bus antecedentium puerorum puellarumque uiolandarum (Augu-
stinus: uitolandarum F) quia non patet effugium (Timpanaro:
patefugium F). Ennius: 'Salmacida spolia (salmacidas polla F)
sine sanguine et sudore '.
CLXXXII
Cicero, Off. 2. 23 : omnium autem rerum nee aptius est quic-
quam ad opes tuendas ac tenendas (tuendas Lp: tuendas tenen-
tes P: tenendas ac tuendas H<)quam diligi nee alienius quam
timeri. praeclare enim Ennius:
quem metuuntoderunt;quemquisqueoditperiisseexpetit. 34 8

CLXXXII *Hieronymus, Epist. 82. 3 . 2 quern metuit quis odit,

quern odit perisse cupit.
CLXXX 346 non ncccssc est F: nonnunc haec esset AV: non nunc necesse est,
std nunc lineatransuersadel. B
CLXXXII 348 perissc c.odd. expetit ZLp: cupit c
146
 INCERTA

CLXXXIII
Cicero, Off.2. 62: propensior benignitas essedebebit in calami-
tosos nisi forte erunt digni calamitate. in iis tamen qui se
adiuuari uolent (uolunt b), non ne adfligantur sed ut altiorem
gradum ascendant, restricti omnino esse nullo modo debemus
sed in deligendis (diligendis Lp) idoneis iudicium et diligentiam
adhibere. nam praeclare Ennius:
benefactamalelocatamalefactaarbitror. 349

CLXXXIV
(a) Cicero, Off. 3. 104: sed in iure iurando non qui metus sed
quae uis sit debet intellegi (intellegi debet Xs'). est enim ius
iurandum affirmatio religiosa; quod autem affirm.ate et (Zps':
affirmat et c) quasi deo teste promiseris, id tenendum est. iam
enim non ad iram deorum, quae nulla est, sed ad iustitiam et ad
£idem (ad iustitiam et £idem Le: ad £idem et iustitiam p)
pertinet. nam praeclare Ennius:
o Fidesalmaaptapinniset ius iurandumIouis. 350

qui ius igitur (igitur ius Xs') iurandum uiolat, is £idem uiolat
quam in Capitolio uicinam Iouis optimi maximi (ioui c), ut in
Catonis oratione est, maiores nostri esse uoluerunt.

(b) Apuleius, Socr. s: neue per ista iuretur, cum sit summi
deorum hie honor proprius. nam et ius iurandum Iouis (Brant:
iurandum iouis iurandum codd.)dicitur, ut ait Ennius.

CLXXXV
Cicero, Lael.64: quid? haec ut (ut om.P) omittam, quam graues,
quam difficiles plerisque uidentur calamitatum societates ! ad

CLXXXIII 349 bcnefacta Z: bcne facta ~: beneficia X male facta codd.

147 I~
 THE FRAGMENTS

quas non est facile inuentu qui descendant (discendant P:

distendant M: discedant K). quamquam Ennius recte:
amicuscertusin re incertacernitur. 351

tamen haec duo leuitatis et infirmitatis plerosque conuincunt,

aut si in bonis rebus contemnunt aut in malis deserunt.

CLXXXVI
Varro, Ling. s.14: locus est ubi locatum quid esse potest, ut
nunc dicunt collocatum. ueteres id dicere solitos apparet ...
apud Ennium:
o terra Thraecaubi Liberifanum inclutum 352
Maro locaui.
CLXXXVII
Varro, Ling. s .23: terra, ut putant, eadem et humus; ideo
Ennium in terram cadentis dicere:
cubitispinsibanthumum. 354

CLXXXVIII
Varro, Ling. s .64: terra Ops, quod hie omne opus et hac opus
ad uiuendum, et ideo dicitur Ops mater quod terra mater. haec
emm
terrisgentis omnispeperitet resumitdenuo. 355

quae dat cibaria, ut ait Ennius, quae quod gerit fruges Ceres.
antiquis enim quod nunc G C (Lachmann:& F).
65: idem hi dei Caelum et Terra, luppiter {lupiter F) et Iuno
quod, ut ait Ennius,
isticest is luppiterquemdico,quem Graed uocant 356
aerem,qui uentusest et nubes,imberpostea,
atqueex imbrefrigus, uentuspostfit, aer denuo.
CLXXXVI 352 Thraeca Fleckeistn:treca F inclutum Gulielmius:inciuiO F
3S3 miro ut uid. ras.ex maro F locauit Ribbeck
CLXXXVUI 356 iupiter F 358 uentus Laetus: uentis F
148
 INCERTA

thaec proptert luppitersunt ista quaedicotibi.

tquat mortalisatqueurbesbeluasqueomnisiuuat.
quod hinc (L. Spengel: hie F) omnes et sub hoe, eundem
appellans dicit 'diuumque hominumque paterrex'.

CLXXXIX
Varro, Ling. 7. 12: tueri duo significat, unum ah aspectu, ut dixi,
unde est Enni (C. 0. Mueller:enI F) illud:
tueorte senex?pro Iuppiter! 361

... alterum a curando ac tutela ...

CXC
Varro, Ling. 7. 12: quare a tuendo et templa et tesca dicta cum
discrimine eo quod dixi.
13 : etiam indidem illud Enni (Scaliger:enI F) :
extemploacceptumme necatoet}ilium. J62

extemplo enim est continuo quod omne templum esse debet

continuo saeptum (septum F) nee plus unum introitum habere.

CXCI
Varro, Ling. 7. 16: Ennius:
ut tibi Titanis Triuiadederitstirpemliberum. 36 3
Titanis Triuia Diana est, ah eo dicta Triuia quod in triuio
ponitur fere in oppidis Graecis uel quod luna dicitur esse, quae
in caelo tribus uiis mouetur, in altitudinem et latitudinem et
longitudinem. Titanis dicta quod cam genuit, ut (L. Spengel:
ut ni F) Plautus (cf.Bacch.893),Lato.

3 S9 iupiter F
CLXXXIX 361 iupiter F
CXC 362 acccptii, cc ex u co". F: acceptam. Vossius necato Scaliger:negato F

149
 THE FRAGMENTS

CXCII
Varro, Ling. 7. 19: Enni (ennii F):
Areopagitaequiddederetquam pudam t 364
Areopagitae ah Areopago; is locus (Laetus:his locis F) Athenis.

CXCIII
Varro, Ling. 7. 48: apud Ennium:
quaequein corporecauacaeruleo 365
caelicortinareceptat.
caua cortina dicta quod est inter terram et caelum ad simili-
tudinem cortinae Apollinis; ea a corcle quod inde sortes primae
existimatae.
CXCIV
Varro, Ling. 7 .49: apud Ennium:
quin indeinuitissumpserintperduellibus.
perduelles dicuntur hostes.

CXCV
Varro, Ling. 7. 87: lymphata dicta a lympha; (lympha) (add.L.
Spenge~a Nympha, ut quod apud Graecos 9rns, apud Ennium:
Thelis illi mater. 368
CXCVI
Varro, Ling. 7. 89: apud Ennium:
si uolesaduortere
animumcomitermonstrabitur 36 9
comiter hilare ac lubenter cuius origo graeca Kooµos(Aldus:
comos F).
cxcm 365 quaeque in corpore caua caeruleo Bergk: caua quaeque in corpore
caeruleo Tumebus, Aduers. XXIII .17: quaeque in corpore causa ceruleo F
366 caeli cortina receptat Tumebus: c~lo orta nare ceptat F
CXCIV 367 uindictam uictis Scaliger
CXCV 368 Thelis Tumebus: thetis F
ISO
 INCERTA

CXCVII
(a) Varro, Ling. 7.93: euax uerbum nihil significat sed effu-
titum naturaliter est- ut apud Ennium:
hehae! ipseclipeusceddit. 370

(b) Grammaticus incertus, Gramm. v 574.24: clipeus (L:

clypeus M) generis masculini, ut Ennius: 'cecidit clipeus (L:
clypeus M) '.
CXCVIII
Varro, Ling. 7. 93 : euax uerbum nihil significat sed effutitum
naturaliter est- ut ... apud Ennium:
heu meapuellatspet quidemid succenset tibi. 37 1

CXCIX
Varro, Ling. 7. 101: apud Ennium:
uodbusconddetfacimus ett obrutus. 37.2

mussare dictum quod muti non amplius quam MV dicunt.

cc
Rutilius Lupus 1. 12: ~1acpopa. hoe schema ( est) (add.Ruhnken)
cum uerbum iteratum aliam sententiam significat ac signifi-
cauit primo dictum (Stephanus:dictum est codd.)... item in
Enni uersu (Halm: Enni uersus Meineke:uniuersum codd.):
mulierem:quidpotiusdicamaut ueriusquammulierem? 373

CCI
Paulus, p. 110. 16: MBTVS feminine (femine R) dicebant.
Ennius:
uiuaman moriarnullain me est metus. 374

cxcvm 371 ipse quidem Lachmann

CXCIX 37.2 facito musset L. Mutlla
CCI 374 mcto R
151
 THE FRAGMENTS

CCII
Festus, p. 166.11: NAVCVM ait Ateius Philologus poni pro nugis.
Cincius quod oleae nucisque intus sit. Aelius Stilo omnium
rerum putamen. glosematorum autem scriptores fabae grani
quod haereat in fabulo. quidam ex Graeco, quod sit val Kaiovxf,
leuem hominem significari. quidam nucis iugulandis, quam
Verrius iugulandam uocat, medium uelut dissepimentum.
Plautus ... et Naeuius ... sed (et) (uu{go:sed F) Ennius:
illicest nugatornihili non naucihomo. 37S

CCIII
Festus, p. 218.2: OBSIDIONBM potius dicendum esse quam
obsidium adiuuat nos testimonio suo Ennius (pacuuius X) in
Telamone quom (cum X) ait .. .item alio loco:
Hectorqui haudcessatobsidionem
obducere. 376

CCIV
Festus, p. 388. 25:
traxit. sosr
omnes ferre auc
Afranius in Ep
seruent tuis
'maxime Teu(crorum ductor quo sospite)
numquam'. Elli].
parentemet pa
sospitem.Ace 377
rite ad patri
set. Ennius uid

CCU 37S illic Vrsinus:illuc F . nihili Vrsinus:nihil F

CCIV SOSPES, saluus. Ennius tamen sospitem pro seruatore posuit. soSPIT ARE est
bona spe adficere aut bonam spem non fallere - Paulus

152
 INCERTA

ficare cum dix

liber. ceterum s
ait esse bone spe
spem non &lie
CCV
Festus, p. 482 . 8 : TOPPER significare ait Artorius cito fortasse
celeriter temere ... ira'; Sinnius (C. 0. Mueller:iras ennius F}
uero sic: 'topper fortasse ualet in Enni et Pauci scriptis; apud
Ennium est
topperquamnemomeliusscit.

CCVI
Festus, p. 488. 22:
~ro aspera difficilia aditu
{a asperasaxa tescatuor379
CCVII
Festus, p. 494. 12: TAM ••• at antiqui tam etiam pro tamen usi
sunt, ut Naeuius ... Ennius:
ille meaetampotispads potiri. 380

CCVIII
Paulus, p. 507. 12: VITVLANS, laetans gaudio ut partu (pastu
uitulus Augustinus).Ennius:
is habetcoronamuitulansuidoria. 381

CCIX
Aurelius, Ad Frontonemepist. p. 27. 25: m ea fortuna con-
stitutum in qua, ut Q. Ennius ait,
omnesdantconsiliumuanumatquead uoluptatemomnia. 382

CCVI TESCA, loca augurio designata.Cicero aspera ait cssc et difficilia-

Paulus: tcmpla antiqua tcsca cssc ait truerotaspcra difficiliaaditu. Ennius
ardua aspera saxa ttuost- lib. Vat. n. 2731 ( Mommsm, SB Berlin 1864, 64
[= Gu. Sehr. YII 276])
CCVII 380 patis V
153
 THE FRAGMENTS

CCX
Aurelius, Ad Frontonemepist.p. SS. 18: de Herode quod dicis,
perge, oro te, (et) (add.Haupt), ut Q...uintus
noster ait,
peruincepertinaciperuicacia.

CCXI
Helenius Aero, Hor. Carm. 3. II. 17-18 (Cerberus,quamuis
furiale centum I muniant angues eaput eius): ut ait Ennius:
angueuillosicanis.

CCXII
Nonius, p. 196.29: CAEMENTA neutri, ut est usu (usus B-4)•..
feminini. Ennius:
labat,labuntursaxa, caementae
cadunt.

CCXIII
(a) Nonius, p. 197. 28: QVISet generi feminino adtribui posse
ueterum auctoritas uoluit ... Ennius :
et quis illaecest quaelugubrisuccincta
est stola? 386

(b) Pompeius, Gramm.v 2o6. apud maiores nostros indif-

20:
ferenter inuenimus hoe pronomen, et 'quis uir' et 'quis
mulier' ... 'quis mulier' habemus et in Ennio et in Pacuuio et
in ipso Terentio.
CCXIV
Nonius, p. 205. 23 : PRBTVMneutri tantum generis esse uolumus
... masculini ... Ennius:
crassapuluis oritur,omnemperuolatcaelifretum.

CCX 383 peruicacia m1 : peruicaciam m1

CCXII 385 labwitur uulgo: lababwitur coda. caementae Iunius: caementa
codd.

154
 IN CERT A

CCXV
Hieronymus, Epist.6o. 14. 4: Naeuius poeta 'pati' inquit 'ncccsse
est multa mortalem mala '. unde ... prudenterque Ennius
plebes
ait
in hoeregiantestat:locolicet 388
lacrimareplebi,regihonestenon licet.
ut regi, sic episcopo. immo minus (plenius G) regi quam
episcopo (quam episcopo regi G: episcopo quam regi 't'). ille
enim nolentibus praeest, hie uolentibus.

CCXVI
Diomedes, Gramm.I 345 . I : item adeo adis: hoe iteramus adito
aditas dictitantes, ut Ennius:
ad eum aditauere. 390
CCXVII
Diomedes,Gramm.1 400. 1 s: plura enim uerba quae uulgo
passiuo more declinamus apud ueteres diuersa reperiuntur
enuntiata declinatione .... moro ... Ennius {Ennius om. B):
tan aliquid quod dono illi morare sed accipitet 391

CCXVIII
(a) Diomedes, Gramm. 1447. 5: homoeoteleuton est oratio
similibus clausulis terminata, id est pari uerborum exitu finita,
ut apud Ennium:
eosreduciquamrelinqui,deuehiquamdeseri 392
malui.
CCXV 388 plebis <I>:plebs mDB p.c.m• rcgio GAK<l>'f'B antistat GKB:
antistct m a.c.m• {antcsto in mg. m1 ): antistant 'f' 389 lacrimari <l>B
CCXVI ut Ennius 'ad cum aditaucrc' hue transposuitSdoppius
CCXVD 391 morarcs scd ~
CCXVm (a) 392 costc duci M: cos dcduci AB quam rclinqui om. ABM
deuihi M: dcuei A dcscrit M 393 maluit B
155
 THE FRAGMENTS

(b) Donatus, Gramm. IV 398. 24: homoeoteleuton est cum

simili modo (est cum similiter P: est oracio cum simili
modo S) dictiones plurimae finiuntur ut 'eos reduci (eos
reuei S) quam relinqui deuehi (relinquere deuei S) quam deseri
.,
malw.

(c) Charisius, p. 370.27: homoeoteleuton est oratio pari

uerhorum exitu finita ut 'hos reduci (seduci N) quam relinqui,
deuehi (deui N) quam deseri malui (ma!lu N) '.

CCXIX
Seruius auct., Verg. Georg.I. 12-13 (cui prim.a frementem I
fudit equom): nonnulli uero oh hoe' cui prima frementem fudit
aquam' legunt quod ueteres murmura aquae fremitum dicehant.
Ennius:
ageroppletusimbriumJremitu. 394

CCXX
(a) Seruius auct., Verg. Georg.4.17~1 (fulmina massis I cum
properant) : fulmina properant uetuste ait, ut ... et Ennius :
'festinat diem'.
(b) Servius auct., Verg. Aen. 9.401 (properet per uolnera
mortem) : aut deest adire (Commelinus:
audire F) aut deest ad ...
aut certe antique properet mortem, ut ... Ennius:
festiuumfestinantdiem. 39S

CCXXI
Seruius, Verg. Aen. 1.4 (saeuae memorem Iunonis oh iram):
cum a iuuando dicta sit Iuno quaerunt multi cur eam dixerit
saeuam et putant temporale epitheton quasi saeua area

CCXX (b) 39S festiuum Danielis: fcstinum F dies F

 IN CERT A

Troianos, nescientes quod saeuam dicebant ueteres magnam,

ut Ennius:
indutafuit saeuastola.
id est magna.
CCXXII
Seruius auct., Verg. Aen. 2.62 (certae occumbere morti):
OCCVMBBRE MORTI. nouae locutionis figura et penitus remota.
Ennius:
ut uos nostriliberi 397
defendant,pro uostrauita mortioccumbantobuiam.

CCXXIII
Seruius auct., Verg. Aen. 3 .241 (ferro foedare uolucres):
FOBDARB,cruentare. Ennius:
Jerrofoedati iacent. 399

CCXXIV
Seruius, Verg. Aen. 6. 686 (effusaeque genis lacrimae) : GENIS,
palpebris. Ennius de dormiente:
inprimitquegenaegenam. 400

CCXXV
Seruius, Verg. Aen. 7.319-20 (nee face tantum I Cisseis
praegnas ignis enixa iugalis) : ossms, regina Hecuba, £ilia
secundum Euripidem Cissei {ciseiASL), quern Ennius Pacuuius
et Vergilius sequuntur; nam Homerus Dymantis (dimantis
codd.)dicit. haec se facem parere uidit et Parin creauit qui causa
fuit incendii.

CCXXI 396 inducta (corr.) N

CCXXIJ 397 nft C: uft C 6 : uestri/ 398 uestra C'/ obuiam om.f
CCXXIV 400 inprimit M rcne H: genacque M

157
 THE FRAGMENTS

CCXXVI
I
Seruius, Verg. Aen. 9.255-6 (integer aeui Ascanius): INTEGBR
ABVI, adulescens cui aetas integra superest, unde Ennius

deosaeui integros 401

dicit quibus multum aeui superest.

CCXXVII
Seruius auct., Verg. Aen. 9. 747 (hoe telum mea quod ui dextera
uersat): VBRSAT, librat, iactat. et est Ennianum
uersatmucronem. 402

158
COMMENTARY
 ACHILLES

The tide Achillesis given to Ennius by Verrius Flaccus/Gcllius, 3 Nonius,3

Isidore• and a late commentator on Ciccro,s orations,5 Nonius gives the
same tide to Livius 6 and to Accius.7 One ofVerrius, quotations 8 allows us to
see that the author of Ennius, original was the mid nfth century poet
Aristarchus ofTegea. The reference to an Achilles.Aristarchi in the prologue
of Plautus, Poenulusmust be to Ennius, adaptation.9
The themes of the three tragedies are not readily discernible from the
fragments quoted. Livius may have dealt with incidents of Achilles, life on

1
Festus, p. 282.9 depends on Strzelccki's 's lists from archaic poetry' (cf.
~st. Vm. pp. 81 ff.); p. 394.33 on the '17 lists from archaic authon'.
s 4. 17. 14 depends on a grammatical source, probably one of the writings of
Probus or Sulpicius Apollinaris.
3 Pp. 147.19 and 166.20 depend on list 27 of Ijndsay's 41, 'Alph. Verb';

p. 169.2 perhaps on Gellius4. 17. 14 (cf. Strzelccki, in Aus tin olurtumswissm-

schaftlichtnArbrit Vollupolen[Sehr.tin &kl.far Altmumswiss. tin Dt. Aledd.tin
Wiss. zu Balin xm, Berlin, 1959), pp. 81 ff.); p. 277.26 perhaps on list 1
'Gloss. i'; p. 472.26 on list 26 'Gloss. iii' or list 27.
4 Nothing precise can be said about the source of Di.ff.1 .218. Isidore's poly-
mathy depends notoriously on the work of writers who were not even them-
selves given to original research (on Isidore in general seeJ. Fontaine, Isidoretk
Slvillt et la cultureclassiquedansl'Espagnewisigothiqau[Paris, 1959]). The effort
to distinguish carefully between words of related meaning had long been
encouraged by rhetoricians and lexicographen (see M. L. Uhlfelder, Dt
Proprid4te Strmonum utl R.trum: A Study and Critical F.dition [American
Academy in Rome, 1954), pp. 1 ff.).
S The source of Schol. Gron., Cic. Vm. 2. 1.46 is obscure. The identifica-
tion of noacknowledged quotations in the text of Ciccro's speeches was under-
taken by commentaton as early as the fint century A.D. and probably earlier
(cf. Asconius, Pis. 82).
6
The source of pp. 365. 37 and 473. 18 is obscure.
7 The source of pp. 98 .4, 110. 13, 277. 37 and 503. 34 is obscure. It is
certainly a different one from that which provided the Myrmidonesquotations
at pp. 109.31, 120.31, 137.33, 234.8, 261.24, 262.15, 396.31, 433.1 and
so2.1, namely Lindsay's list 17 'Accius iii' (cf. also P. Schmidt, Dt Nonii
Mdrcelliauctoribus g,ammaticis[Leipzig, 1868), pp. 52 ff., in particular p. 59 n.
26). H. Diintzer's theory (L. Liuii AndroniciFragmenta[Diss. Berlin, 1835),
p. 20) that the titles Achillesand Myrmidontsrefer to the one script thus remains
a possible one.
8
Festus, p. 282. 9. See Introduction,p. 59.
9 See Introduction, p. 7.
II 161 JTO
 COMMENTARY

Scyros, 1 Accius with the siege of Troy. 2 Ennius' tragedy was certainly set
near a field of battle and W elcker interpreted most of the fragments with
great plausibility as coming from a scene in which Agamemnon's ambassa-
dors attempt to persuade the sulking Achilles to rejoin the fight with the
Trojans.3 Later students have made only slight modifications to Wdck.er's
reconstruction.
Achilles appeared in three other tragedies by Ennius, Hectoris lytra,
Iphigeniaand Telephus. The Hectorislytra was set in the Greek camp before
Troy; so too theAiax, Pacuvius' Armorumiudidum,Accius' Armorumiudicium,
Myrmidones,Epinausimacheand Nyctegresia.Several of these plays are quoted
outside grammatical literature in loose and allusive ways,4 and it is rarely
possible for the modern student to assign them accurately. Even when
Ennius' name is mentioned we are hindered by our ignorance of how far
forward from theembassy the action of theAchilleswent and how far back
from Priam's visit to Achilles' tent that of the Hectorislytra went.
At S. Rose. 89 Cicero is speaking about the great number of professional
prosecutors slain during the Sullan proscriptions. There spring to his mind
famous slaughters in Roman history and in heroic legend, in particular that
of the Achaeans by their shipsat thehandsof Hector. One of his sentcnces-
quisibi non estuulneratusferro Phrygio?--consistsof three trochaic metra and is
said by a scholiast to come from an address by Ulyssesto Ajax outside the tent
of Achilles in a tragedy by Ennius (fr. CLXIV). Osann,S Weick.er and Bergk 6

1
Cf. E. Bickel, RhM LXXXVI (1937), 1 ff.
3
This would be certain if the title Myrmidonesreferred to the same script.
On the fragments of the Myrmidonessec Wclcker, Die AeschylischeTrilogie
(Darmstadt, 1824), pp. 420 ff., G. Hermann, De AeschyliMyrmidonibusNereidi-
bus PhrygibusDissertatio(Leipzig, 1833), pp. 6 ff. (= Opusc. v 1.40ff.), W.
Schadcwaldt, Hermes LXXI (1936), 49 ff. ( = Hellas und Hesperien [Ziirich,
1960], 191 ff.), C. B. Earp, A Study of the Fragmentsof Three RelatedPlays of
Accius (Diss. Columbia, 1939), pp. 24 ff., G. Barabino, in 'AVTi6oopovH. E.
Paoli oblatum(University of Genoa, 1956), pp. 57 ff.
3 Die griech. Trag. p. 933. Cf. Dtintzcr, L. Liuii Andron. Frag. p. 20. A.
Scholl's reconstruction (Beitriigezur Kenntnisde, tragischen PoesiederGriechen,1
[Berlin, 1839], pp. 485 ff.) is not worth discussing.
4 Horace doubtlessly has a Latin tragic Achilles in mind at Ars 120-2-
honoratumsifortereponisAchillemIinpiger,iracundus, inexorabilis,acerIiuraneget sibi
nata,nihil non adrogetarmis-but it would be foolish to guess which one. Cicero
distinguishes 'Achilles Homericus' (De orat.3 . 57, Tusc.3 . 18, 4. 52, Diu. 1 . 65,
Att. 9. 5. 3), 'Acciw et aliquando sapiensAchilles' (Tusc.1. 105) and 'ille delphi-
gcnia Achilles' (Rep. 1 . 30) but frequently alludes to tragic versions of incidents
of the Trojan siege without naming the author. S Anal. crit. p. 13 n. 6.
6
Ind. lectt.Marburg1844, x ff. ( = Kl. phil. Sehr. I 222 ff.), altering Vlixes
uulneratus... cum interrogaretur ab Aiace to Eurypylus... Achille.
162
 ACHILLES
assignedit to the Achilles.Ribbcck1 accepted Scholl's reconstruction of the
Heck>ris lytra,"according to which the action began even before Patroclus'
departure to do battle with the Trojans, pointed out the large number of
extant fragmentsreferring to battle and amgned quisibinonestuulneratus ferro
Phrygio?to that play. Scholl's reconstruction is by no means certain and
while there is positive evidencethat Cicero knew the saipt of the Achilles
there arc very few signsthat he knew the Hectorislytra.At an earlierpoint in
his speechfor Roscius,where he dcsaibcs the embassy of the dccurions of
Amcria to Sulla {26),there is a verbal paralldism with fr. VI of the Achilles.
On the other hand, when later in the speechhe dcsaibcs the murder of the
elder Roscius,he askshis hearers {98): non uersaturanteoculosuobisin CMde
Glaucia?non adestiste T. Rosdus?non suis manibusin curruconlocatAuto-
medontem ilium,sui salerisaarbissiminefariaequeuidoriaenuntium?and the
scholiast comments (p. 312. 17): AutomedonAchillisaurigafuit. posttaquam
AchillesHectoremuidt, posuitaurigamsuumin cum, ut iretet nuntiaretoccisum
Hectorem.This departs from Homer's version of the story and it is possible
that Cicero had a tragedy in mind. However the tragedy could hardly have
been Ennius' Achilles.The whole question of fr. CLXIV must accordinglybe
left open.
At Tusc.2. 38 Cicero quotes a dialogue in iambic tetrameters between
Eurypylus, who had beenwounded in battle, and Patrodus (fr. CLXXI). It is
clearthat we have here a dramatisationof the incident dcsaibcd by Homer
at II. II . 8o6 ff. G. Hcrmann3assigned it to Accius'Myrmidones but a phrase
is alsoquoted at Orat. 155 and Bcrgk4showed from an analysisof Cicero's
method of quotation here that Ennius must be the author. Bergk assigned
the dialogue to the Achilles;Ribbcck and Vahlcnon the other hand assigned
it to the Hectorislytra.The problem is similarto that of fr. CLXIV. Two points
arc worth making, although they do not help much to resolve the problem.
First,fr. CLXIV and fr. CLXXI appear to come from similarkindsof sceneand it
is a little difficultto imagine them both in the one play. Secondly, the
Eurypylus of fr. CLXXI appearsto have given afterhis dialoguewith Patroclus
a lengthy dcsaiption of the battle. The iambic tetrameters may replace
trimctcrs of the original but the argumentative structure of the scene must
belong to the Greek poct.S Such accounts of events offstagc were usually
given by unnamed messengerseven in late fifth century Attic tragedy.6 One

' Q!y,est.scm. pp. 276 f.; cf. Die rom. Trag. pp. 118 ff.
" Beilr. pp. -472ff. See below, p. 290.
3 De Atschyli Myrmidonibus, p. 8 ( =- Opusc. v 1-42).
4 Ind. ltctt. Marburg1844, vm ff. ( = Kl. phil. Sehr.1 220 ff.). Cf. below,
p. 205. s Andria28 ff., however, could be Terence's invention.
6
Fraenkcl. however (De Mediaet NouaComoedia ~aestionesStledae[Diss.
Gottingcn, 1912], pp. -43ff.), compares Pap. Hib. 1.-4 (p. 22: Euripides?).
163 II·2
 COMMENTARY

should hesitate to attribute such an advanced dramatic technique to

Aristarchus.
The words quoted by Cicero at Orat. 16o (fr. CLXXII)-uipatefecerunt
Bruges-and by Verrius at Festus,p. 218. 2 (fr. ccm)-Hector qui hauJcessat
obsidionem obducere-asby Ennius clearly belong to an account of Hector's
attack on the Greek encampment. The tragedy must be either the Achillesor
the Hectorislytra.
The words quoted by Varro at Ling. 7 .93 (fr. acvn): hehaetipseclipeus
ctddit,by Nonius at p. 205. 30 (fr. cCJCIV):aassapuluisoritur,omnemperuolat
caeliftetum, and by the commentators on Virgil's Aeneid at 3 .241 (fr.
ccxxm): ferrofoedatiiacent,and 9. 747 (fr. CCXXVII): uersatmuaonem,as by
1
Ennius belong to one or more battle descriptions. The Erectheus is as likdy
in each case to be the source as either the Achillesor the Hectorislytra.
The Ennian reference to Achilles' mother quoted by Varro at Ling. 7. 87
(fr. acv) could as easily come from the Achillesor the Htctorislytraas from
the Iphigenia.i
Ribbcck: compared the words quoted by Gcllius at 19. 8. 6, eo egoingenio
natussum: amidtiamatqueinimidtiaminftontempromptamgero,with Homer,
n.9.312-13 ~8pbsyap µ01 KElVOS6µC)s 'A{6ao1NA1J01V, I6sx' hepov
µev1<EV81J wi q,pea{v,6:Jv,,.o
& eifflJand assigned them to the Achilles.They
could have been spoken by the Achilles of the Hectorislytraor the Telephus
equally wdl,3 or indeed by several other Ennian heroes.Gcllius' mode of
quotation, Q. Ennius in illomemoratissimo libro,implies, however, that he is
not quoting a dramatic poem.• LadewigS suggested that at Mur.6o, nonmulta
sed(si) (siadd. ed. Ascens.I SI I)
peccas,inquitillefortissimouiroseniormagister,
peccaste regerepossum,Cicero quotes from a scene of Accius' Myrmidonesin
which Achilles is admonished by his one-time tutor Phoenix about his refusal
to help the hard-pressed Greeks. Ennius' Achillesis just as likely to be the
source.
I
The prologue of the Poenulusposes two questions of importance to students
of the Achilles.The first one is whether we have a parody of a scene set at a
place of public assembly; the second is whether we have extensive verbal
quotations of Ennius' script.

I Cf. fr. LJCII.

a Welcker, Die griech. Trag. p. 934 n. I, suggested the Achilles. Vahlen,
E.P.R. 1, p. 159, the lphigtnia.
3 Weick.er, Die griech. Trag. p. 489 n. II, assigned them to the Ttltphus.
They have no parallel in Euripides' I.A. and it is difficult to imagine a context
in which Ennius could have introduced them himself in his version of that play.
4 Cf. L. Havet, RPh XIV (1890), 45, Timpanaro, SIFC N.S. xxm (1948),
6 ff. 5 Anal. seen.p. 3 n. I.
164
 ACHILLES

Scholl, who seems to have believed that thecomic prologue was a mere
translation of that of the Kapx1166v1<>s without any necessary reference to
Ennius' version of Aristarchus' •AxlAAe:Js, argued that the tragic scene
showed theassembly describedin lluulI, which was summoned to consider
what to do about theplague sent by Apollo upon theGreek army. 0. Jahn 1

imaginedan assembly in which Achilles slewThcrsitcs. Vahlen3 drew atten-

tion to the one describedin lliad IX at which Agamemnon announced his
decision to placate Achilles. This kind of view is besetwith serious diffi-
culties. The staging of a scene set at a place of assembly is hard to visualise:
hard in the theatre of Dionysus at Athens, even harder in a Roman theatre
where the Bat space in front of the stage was occupied by part of the
audience. 3 In extant Attic tragedies the proceedin~ of assemblies arc
regularly reported by messengers ;4 the sceneof Aeschylus' Evµa,i&sis a
special case. Furthermore, if Scriverius' hypothesis of two plays by Ennius
about Achilles hasto be abandoned, we are faced with a shift of scene within
the one play: from theassembly ground to Achilles' tent. The scene shifts in
Aristophanes' Barpaxo1 ( 180) but nowhere in comedy of theclassical period
or its Roman adaptations;S in Aeschylus' AITvalcn (fr. 26 Mette) and
Evµevi&s(234, 488) and in Sophocles' early tragedy Alas (815) but no-
where in the tragedies of Euripides or other late tragedies so far as we can
determine. It is likdy that even if a Roman poet chose an early Attic tragedy
to adapt he would have obliterated any shiftsof scene. Two opinions are
admittedly possible on this point but Roman tragedies with shiftsof scene
should not be postulated unnecessarily. One could avoid the difficulties
I have sketchedby interpreting verses 3-49 of the prologue as a parodl of
commands given by Ennius' Achilles to the soldiers guardinghis tent.
E. Klussmann7 argued that v. 3 of the prologue, siletequeet t«ete atque
animumaduortiteand v. 11, exsurgepraeco Jaepopuloaudientiamwere verbatim
quotations of thescript of the Achilles.In his first edition of the fragments of
tragedy Ribbcck printed theseverses and part of v. 4, audireiubetuosimpera-
tor, as coming from the hand of Ennius. All subsequent editors of the
fragments of Ennius' tragedies have accepted this idea. Of editors of the
I
PoenulusLeo accepted vv. 3-4, siletequeet t«ete atqueanimumaduortite.audire
1
Hennesm (1869), 191. 2
E.P.R. 1 , p. n8.
3 See Introduction,
p. 18.
4 Cf. Euripides, Hek. 521 ff., I.A. 1543 ff., Or. 884 ff.

s H. W. Prescott, CPh vm (1913), 14 ff., tried to explain the muddle at

Amphitruo 629-32 by postulating a shift of scene from the harbour to the
palace.
6 For doubts about the usual view taken of the Poenulusprologue cf. F.
Stoessl, RE xxm ii (1959), Nachtrii'ge,s.v. Prologos,
2370.
7 N]bb Suppl XI (1845), 325.

165
 COMMENTARY

iubet uos imperator,as a quotation and rejected v. 11. Lindsay on the other
hand added v. 16, bonumf actum tcsset tdictaut seruetismta, to the quotations
allegedby Klussmann and Ribbcck.
The verses exsurgtpraeco,Jae populo audientiam 1
and bonumf actum tessct
edictaut seruttismea obviously spring straight from the discourse the comic
poet gives hisactor. There is no reason at all to treat them as quotations of
another play. Sileteque tt 14Ceteatque animum aduortite.I audirt iubet uos
imperatormay be of similarcharacter.
The word commtntariis wrongly interpreted at T.L.L. m 1864.19 as
'afferre •. There is no good paralld for this IJ1eaning.What we have is prob-
ably a piece ofactors' jargon: 'apply themind to,study, con (with a view to
acting a role)'; cf. Plautus, Cist. 509, True.73 5 ff., Cicero, S. Rose.82, Brut.
87, 301. The sentence indemihiprincipiumcapiamex ea tragotdianeed mean no
more than'I, who have often played the title role in the Achilles,will open
this comedy like the Achilles,i.e. with a general's speech'.
The verb sileti' and thecopula -quett3 had a pompous tragic ring but the
commonplace tacetemust have come ,rcxpcx ,rpoa6o1dav with similar effect
to histricusin v. 4 andfabulisin v. 8. Where one word of a tautological pair is
rarer and more devated thanthe other it normally comes second.4
Audire iubet uos imperatorlooks superficially like part of a quotation of a
tragic herald's address but one wonders why the prologising actor, who
elsewhere speaks with the voice of a magistrate, should begin his edict so
incffectivdy.
I should take the phrase as a pompous variant of uos audire iubeo and
compare the Euripidean Orestes' address to his dead father at Or. 1225-30
and the words of the Plautine Jupiter to Alcmcnareported at Amph. 1004-5 :
Alcumenaadestauxilium: nt time. I et tibi et tuis propitiuscaelicultoraduenit.S
Roman magistrates used a phrase lilce uos audire iubeo at contionts6 and
Caesar was not the only one to talk of himself in the third person; Cicero
addressed Catiline with exire ex urbe iubet consulhostem(Cati1. 1 • 13).

1
Cf. Plautus, A.sin.4faa nunciam{Linge: iam nunccodd.) tu praecoomntm
auritumpoplum.
3
Only four times in comedy.
' Rare in comedy and then only in phrases of some dignity concluding a
metrical unit; see T.L.L. vii 887. 36 tf.
4 For such tautologies in Roman tragedy see below on Ennius, v. 9.

s Cf. the way those personages of comedy who are given to paratragic
pomposity sometimes issue greetings in the third person: Plautus, Amph. 676,
Baah. 243, Epid. 126-7, Mere.713, Poen.685-6, Trin. 435-6, 1151-2, Terence,
Eun. 270-1.
6
C£ Livy 2. 7. 8 ibi audireiussisconsullaudare
fortunamcollegat,Horace, Sat.
2. 3 . 77 audireatquetogamiubeocomponere (parody).
166
 ACHILLES
Words and phrases from Ennius' Achillesdoubtlessly do occur in the
Poenulusprologue, perhaps here and there right down to v. 49, but the
modem scholar is in no position to isolatethem. The similarity between
siletequeet t«ete atqiuanimumoduortiteand Aristophanes, Thmn. 381 a{ya
cnc.:mairp6cnxeTOV vowpointed out by Fraenkd1 is not so very great if I
am right about the tone of t«ete and in any casecould be explainedby the fact
that both Athenian mclRoman public assemblieswere rather noisy. There is
no need to invoke Aristarchus as the link.betweenthe two verses.

II
The words which the scholiast alleges to come from Ennius' Achillesstand
out stylisticallyfrom Cicero's discourse.The use of itaas a causativeparticle,
frequent in drama,seemsnot to occur dsewhere in Cicero's orations. Words
of the length of eidebantur,much affectedby tragedy, rarely end a Ciceronian
sentence. From the preceding sentence0. Plasberg3extracted the tetrameter
ibi tum $Ubito $UntC0<1t't«tempestates
maxim«, remarking the absencedsewhere
in Cicero's orations of the verb cooriri.
The absenceof thisverb and the com-
monness of its simple form has no significance, for Cicero nowhere has
tempestasorta est or anything like it; the purist Caesar obviously found no-
thing poeticalin tempestascoortaest (c£ Gall.4. 28. 2, s.10. 2, 7. 61. 1, Ciu.
1 . 48 . 1 ). However the plural tempestates
would have raised the tone of the
sentence above the ordinary (seebdow on v. 46).
The alleged Ennian words can be scanned as either ita magniJluctus
eiciebantur v - or - v - ita magniJluctus
eidebanturv -. In neither caseis the
scholiast's term 'hemistichium' an accurate description. Similarphrases in
the scholia on Virgil's poemsfrequently cover what are not verbal quota-
tions of earlier hexametric poems but mere loose imitations.3
It is accordingly possible that Cicero was using antique phraseology to
embroider his near-miraculous story and had no thought ofEnnius' Achilles.
I think.it certain enough that a storm causedby an angryApollo formed part
of either the background or the action of thisplay. It was Apollo who caused
Agamemnon to insult Achillcs4 and ultimately brought about his death.
A commentator saw a similarity, either real or imaginary, between Cicero's
storm and Ennius' and pointed it out.5 The Leiden scholia look like the end
of a long processof epitomising. What we now read on Vm. 2.1.46 may be
the work of someone rewriting an earlier and perhaps more accurate note.
1
zu Aristophanes(Rome, 1962), pp. 119 ff.
Beobachtungen
:a In Vahlen, E.P.R. 1, addenda,p. 304.
3 Sec CQ N.S. xv (1965), 139 ff. 4 II. 1. 8 f.
s For detailedknowledge of republican drama in the extant commentaries
sec the Bobbio scholia to StSt. 102, 117, 118, uo, 122, 123, 126.
 COMMENTARY
lta magnifluctuseiciebanturneed be no more a verbatim quotation of tragedy
than is nisitantaacerbitas ~
iniuriae,tanta uissalerisfaissetat 2. 1. 81 { Ennius,
Trag. 295) or sempernobisuigilandum,semperlaborandum uidemusat 2. 5. 181
(~ Accius, Trag.214).

I ftuctlUeiciebantur: cf. Ovid, Fast. 3. 521 eiectasi forte tenebiturunda,

T.L.L. vii 309.17 ff.
III
The three words come from a clcscription ofbattle and probably refer to the
stance normally adopted by Ajax (cf. Homer, n.7 .219 ff., Anon. Culex
315£).

2 prolato aere: Festus' explanation could be misleading. Scutoproiecto

was probably the set phrase in the contemporary military language; cf.
~adrigarius ap. Gell.9.13. 16 constiterunt: Gallussuadisciplinascutoproiecto
cantabundus;Manlius. •.• Sisenna ap. Macr. Sat. 6. 4. 15 scutisproiectistecti. .•
conidunt,Livy 7 .10.9 proiectolaeuascuto... deiedt,32.25. IO cumproiectoprae
seclipeostaret.Ennius would have usedproferreto fit the fabled hugeness of
Ajax's shield. Here as elsewhere he gave his hero archaic weaponry; con-
temporary Roman soldiers carried the oblong scutummade of wood, linen,
leather and metal insets (Polybius 6. 23 ).
In Roman poetry relating to the heroic age aesis common for protective
armour in general, 1 but outside the phrase under discussion there is no clear
example of its standing metonymically for a particular shield. Ferrum
however frequently denotes a particular offensive weapon.

IV
This is the beginning of either an oath or a supplication, the normal forms of
which on the republican stage were per Iouemiuroet sim. and per deosim-
mortalesobsecroet sim. Ennius' personage invokes the divinised atmosphere.
This was a notion familiarto Greeks of the Hellenistic period but perhaps as
strange to early second century Romans as it was to fifth-century Athenians.
Such oaths as those at Euripides, Ion 870,&XA'ov TOA1bsir0Avacrrpov fSos.
and fr. 487,6µvvµ16' lepovcxtetp' ot1<11a1v A16s,,Aristophanes ridiculed as
both irreligious and nonsensical (Batr. 100, Thesm. 272; cf. Neph. 627,
814 ff.). Philemon however could make the new god speak a comic pro-
logue (fr. 91 ). Whether Ennius took this invocation from Aristarchus' play
or from his own general acquaintance with Euripidean tragedy and Greek

1
See T.L.L. I 1073. 73 ff. Cf. the same use of xw6sin Greekpoetry.
a Contrast Sophocles, Ant. 758 ov-r6v6' ·o).u'11fOV, 0. T. 1087.

168
 ACHILLES
philosophical speculation must remain uncertain. Certainly the divinised
atmosphere appears frequently in his remains: e.g. Trag.233-4, 342, 355-00.
Here it is invoked in the style usualwith thetraditional deities.
The philosophical tone suggests that Ennius' speaker might be Phoenix,
Acbi11es'on~e tutor.
The old editon scanned the words transmitted as two iambic trimeten.
This involves treating the fint syllable of subiasas short, contrary to the
testimony of Gelliusand thenormal behaviour of such words in republican
drama.1 Recent editon follow Lacbrnann2 in taking per egodeumsublinuu
subiasas the end of a trochaic tetrameter lacking thenormal diaeresis.3C£,
however, Vahlen, E.P.R.•, pp. LXX £

3-4 per ego deam mblimu mbices I amida1: the disjunction of per
from its object probably had an antique ring even in supplicationsduring the
second century. It doesnot occur in extant comic oaths (but c£ Ovid, Met.
3 .658, Fast. 2. 841) and is rare, comparativdy speaking, in comic supplica-
tions (c£ Plautus, Bach.905, Mtn. 990,Pom, 1387, R.JUJ. 627, TerenceyAndr.
538, 834). Egois usually accompaniedby an oblique caseof the second person
pronoun in these disjunctions but c£ Plautus R.JUJ. 627.
For the two adjectives in asyndeton split by their noun c£ Ennius, Ann.
395 indignohelloconfecitac.erbo, Trag.296 summissaxisfixus asperis,Plautus,
Persa707-8 longanominaI contorplicata habemus,R.JUJ.
907 quisalsislocisincolit
pisculmtis, Trin. 297-8 nil ego istosmororf aeceosmoresturbulosquibusboni
dedecorantse, Porcius Licinus, fr. 1 .2 bellicosam
in Romuligentemferam.
Sublimis(-4U) was a word of some dignity; it occurs nine times in republi-
can tragedy, not at all in comedy except in the set phrases sublimemrapere
(arripere)and sublimemferre(auferre).
*Subtx, 'something laid underneath' or 'something raised up from
underneath', 4 seems not to occur elsewhere in Latin. The adjective umidas
and therelative clauseuntieoriturimbersonitusaeuoet spirituindicate that the
lower atmosphere is referred to.S The rareness of similar formations in
tragedy makes it unlikdy to be a poetic coinage. Ennius may be using a
word of the architectural vocabulary metaphorically: subiceswould be the
1
M. W. Mather, HSCPh VI (1895), 83 ff., 99, 130 ff., and C. Exon, Herma-
thma xm (1905), 129 ff., find five certain casesin which the first syllableof a
iacio compound is treated as short.
1
On Lucretius3 .227 (Leipzig, 1850). 3 Cf. Ennius, Trag.74,151,196,231.
4 For subicere,
'raise up', cf. Virgil, Eel. 10.73-4, Georg.2. 19, 4.385, Aen.
12.287-8. Sub- in origin indicated a movement from below; see P. Lejay,
RPh XL (1916), 181 ff.
s Cf. Euripides, fr. 941. 1-2 6p~ TOVV\VOV
Tov6' &m,pov al6'pa IKal yi'iv
mp,~,xove·vypatsIv &yKa>.ats ;
 COMMENTARY

blocks out of which the vault of heaven was construct.cd; for similar meta-
I
phors c£ Euripides,Hek. 1100-1 ovpav1ov V\fll"ITfflS 4s~aepov,Ennius,
Trag. 319caeliingentesfomices(c£ Septuag. Is. 40.22 6 CTTr)aclS c;.,s
Kcxµapav
Tovovpav6v), Ann. 49 caelie,aerulatempla, 6o cae,u,cula . .. caeli,615 porta• •.
caeli,Sat. 4 pilatas. .. aetherisoras,Accius, Trag. 531 alto ab limine caeli. The
absence of closeparallels for some of the Latin metaphors in extant Attic
tragedy no doubt reflects the fact that arches and vaults were unknown in
the public buildings of fifth-century Athens. 1

4 unde oritur imber: where a traditional deity is invoked, in both Greek

and Roman literature, a relative clausefrequently indicates his permanent
habitat or function :1 cf., in Roman drama, probably reflecting thelanguage
of Roman cult as much as that of the Attic originals, Ennius, Trag. 234 sol
quires omnisinspicis,Accius, Trag. 581-2 sol qui.•. Jlammam... explicas,Trag.
inc. 19-20 o sancteApollo qui umbilicumcertumterrarumobtinesI undt super-
stitiosaprimumsaeuaeuasituoxforas, 216 Liber qui augustahaeclocaCithaeronis
colis,Plautus, Bacch.172-3 Apollo qui aedibusI propinquosnostrisaccolis,Poen.
1187-8 luppiterquigenuscolisalisquehominum,per quemuiuimusuitalemaeuom,
I quempenes spes uitaesunt hominumomnium.
sonitu saeuo et spiritu: c£ Accius, Trag. 392 ingentisonituet spiritu,Virgil,
Aen. 6. SS1 saeuason4te.
The play with thesounds u and s may be intended to represent the stormy
weather being referred to; c£ Plautus, Amph. 1o62 strepitusaepitus sonitus
tonitrus, Pacuvius, Trag. 336 strepitusfremitus clamor tonitruum, Horace,
Epod.10.19 lonius udocumremugienssinus, Virgil, Georg.4.200 f[ tum sonus
auditurgrauiortractimquesusu"ant, etc., Aen. 1.85 f[ utui EurusqueNotusque
ruont,etc. Yet alliteration and assonance are so common in the remains of
tragedy, in every kind of context3 and particularly at the ends of metrical

1
Sec H. Plommer, Andmt and ClassicalArchitecture(London, 19s6),
pp. 244 f£
a Sec Norden, AgnostosTheos (Leipzig, 1913), pp. 168 ff.
3 For alliteration of noun and determinant cf. Livius 33 maremagnum,37
struicessaxeas,Naevius 8 praesenspretium,21 corporiscustodias,Ennius 3 sublimas
subices,4 sonitu saeuo,29 arcumauratum,34 muliermelior,43 mari magno,44
exitiumexamen,S1 sortessomnium,96 cauacaeli,139 salsumsanguinem,171 stellis
splendidis,189 caeliclipeo,207 uiduaeet uastaeuirgines,215 eraerrans,216 animo
aegro, 223 Medeai miserias,237 custoscorporis,239 opulentumoppidum,2.-44,
cupidocorde,250 cauacaerula,254 uerauirtute,2s6 pectuspurum,296 summissaxis,
305 pedumpulsu, 333 acremaciem,Pacuviw 70 propinquitateparti, 80 antiqui
amid, 82 portentumpauos, 177 reginarerum,232 caelitumcamilla,246 manuum
mollitudine,251 sparteisserilibus,2s2 scruposamspecum,309 dictioDelphis, 386

170
 ACHILLES
units1 that it is always hazardous to link thesound of a phrase with its sense.
In the present caseit is worth pointing out that thetriple alliteration of B is
not as common as that of c and P, soundswhich beginjust as many words
in the tragic lexicon, and that u (long) assonancesare quite uncommon.
Ennius took this dement of his tragic style from his Latin predccesson
rather than from Attic tragedy, which, although not averse to phonetic
effects, sought them very sparingly.3 Their presence in contemporary
hymns and prayers3enabled them to lend weight to the utterances of the
tragic heroes.
V
Timpanaro's restoration namconsiliisiusobuarant..• 4 is the most economical
and convincing so far offered. The resulting anapaestswould come from a
denunciation by Achillesof his Greek enemies.

5 obaarant: not elsewherein Latin; the simpleverb uarareseemsto occur

only in the technical vocabularyof the land surveyors (c£ Grom. pp. 28 5~.
28~1 ). In the case of verbs of the common language tragedy often forms
compounds with ob-where comedy uses the simple form or another com-
pound: e.g. Ennius 41 obuoluere,192 obire and obiocere,3.26 opplere, 376
obducere,Pacuvius 213 obnuntiare,Accius285 obtui (319), 344 oblectare,Trag.
inc. 254 obrigescere.

quibas tam concedit hie ordo: the more influential among the Greek
~OV\T)ES,

indcoremiracundiam, 387 consiliumcicur,418 saeuitiamSalaciu, Accius71 animo

ug,oto, 150 salutisspessupremas,176 satiassanguinis,297 abundantemantiquam
amnem, 331 luciftralampade,421 liberorumleto, 4S3 ffl4UISUts misericordia,
467
uis uerit4tis,468 animumatrocit4tem,501 horridahonutitudo,517 serpentissqwmae,
s6s uis uolnms, 566 stellasseptem,S73 crepituclangente,581 candidocurru,583
aduersoaugurio,647 moremultarum,659 satiassupplici.
1
For alliteration of the three concluding words cf. Nacvius 13, 28, 3S, 4S,
Ennius -4, 34, SS, 62, 118, 138, 203, 211, 217, 219, 250, 258, 259, 262, 281, 287,
303, 310, 317,318,329, 3-47,350, 365, Pacuvius93, 118, 2,46, 292, Accius 16, 22,
158, 198, 200, 201, 223, 229, 288, 43-4, 44S, 449, -4S3, S09, 520, S73, 630.
2
For a passage roughly parallel with Ennius, Trag. 3-4 cf. Aeschylus, Hik.
33-6 Ma 6t A<XiAcmt I XElµCA>VO'TV'TT(t>
f:}poVTijCTTEpo,rij T' I l>µf:}pot6pota(vT'
~µ01s ayp(asI ~oS cwrf\acnms 6AotVTo.For alliteration of noun and deter-
minant cf. Sophocles, Ai. SS, 44S; of the last three words in a metrical unit cf.
Aeschylus, Ag. 1-430, Sophocles, Ai. 44S, 687, El. 210, Euripides, Ba. 1298,
Rhes. 383.
3 Cf. those recorded by Cato, Ag,. 132 ff.
4 SIFC N.S. xxu (19-47), 69 ff. Cf. Havet, RPh XXXI (1907), 133.

171
 COMMENTARY
For tam with verb cf. Plautus, .Bacch.778 ( = adeo). Ennius, Trag. 380
(- tamen).
Hie ordowas a phrase regularly usedby Roman SCllaton referring to their
assembled peen (e.g. Cicero, Pis. 6, 8, 40, 4S et al.). Cf. Plautus' parody at
Cist. 22-4 decetpol, mea Seknium, I hunt esseordinembeniuolentisinter se I
benequeamidtiautier.
VI
For a parallel situation cf. Seneca.Thy. 429-30 quaecausacogit,genitor,a patria
g,tulumI refme uisa? Ennius' speaker may be one of the ambassadon sent to
placate Achilles.
Cicero perhaps hasthispassage in mind when at S. Rose.26 he says of the
ambassadon sent to Sulla: Ameriam re inoratareuerterunt.The phrase re
inorataoccun nowhere else in Latin. Cicero hasremagereand remdicerequite
commonly in his orations, remorarenot at all; he hascausamagereand causam
dicerealso commonly. causamorareonly once (in the early Pro Q!inctio43; cf.
Brut. 47).
6 incerta re atqae inorata: for the word-order cf. Plautus, Amph. 33
iustamremetfacilem,Terence, .Atulr.132 benedissimulatumamoremet celatum;
see Haffter, Untersuchungen, p. 73.
Orare,•speak, say• (cf. Varro, Ling. 7. 41 ). seems to have been obsolete
even in Ennius' time in the vemacular except for the legalistic phrases
aequomorareand ius orare.
In Ennius' time the absolute phrase, especially where it had a preponder-
antly temporal significance,characterised the officiallanguage rather than the
ordinary vernacular. It is a noticeable feature of the style of speeches in
comedy parodying the reports which generals made of their achievements; 1
wherever else it appears the context has a markedly formal tone. It is not
very common even in tragedy; Ennius has SI quofacto. 71 respectantibus .•.
nobis,176 maleregesta,299 remissa.humanauita, Pacuvius 411 o«identesole.

~ gndum Iregredereconare: Nonius' source may have madethe sort

of error that Caesellius made over the gender of cor in Ann. 382 (Gcllius
6. 2. 3 ff.). &g,edereis not elsewhere recorded in Latin and reg,ediis common
only in the historians. The grammarian could have mistaken reg,edere,a
second penon singular present indicative, 1 for an infinitive and amare, an
imperative, for an indicative.
1
E.g. Plautus, Amph. 188 f., 654 ff., &«h. 1070 f., Persa1S3 ff. Sec Fraenkel.
Pl. im Pl. pp. 236 f.• Addend4to Italian trans. p. 429, Haffter, Untersuchungm,
P· 49.
:a The form normally ended in -re in republican drama; for statistics on
Plautus' usage sec C. M. S. Miiller, Ciotta xvn (1928), 137 ff.
172
 ACHILLES

Gradus,'act of taking a step', occurs only once in comedy outside Plautus

(Terence, Phorm.867), is avoided by Caesar, and is rardy used by Cicero.
Sallust and Livy, on the other hand. have it often in accounts of military
activities. Republican (12instances)and Scnccantragedy (6s) have it in all
kinds of contexts. Plautus' 22 instancescan all be interpreted as parody of
either military or tragic language; he has neithergradumregredtre nor gradum
refme nor gradumreptdare(so Pacuvius. Trag.400); redire(142 instances;
Terence 89), reuerti(27; Terence 10), reuenire(17; Terence o), readere(10;
Terence o) expressthe idea for him.
For the etymological figure cf. Euripides, Alie. 869n66a ffll£VOOV,
Plautus,Poen.s14 istocgrassarig,adu.Suchfiguresfrequentlyoccur in the plays
of Plautus in all kinds of contexts from theludicrousto thehighly pathetic.
They are much lesscommon in the rest of comedy.1 Ennius has them often
in both tragedy (36 f atisf andis,48 iudicauit. .. iudidum,93 uitameuitari,11s
corpus... corporaret,211 nominaturnomine,236 fadnus... fit) and epic (Ann.
souoceuoaibam,77 curantes..• cumcura,244/aceretfadnus,314/actumquefadt,
452 longiscere longe,458 riserunt.. .risu); in the rest of tragedy I find only
Naevius 39 injlexujlectitur,Pacuvius 239 nuncupastinomine,Trag. inc. 82
partu... parit, 185 dominaredomino.They occur fairly frequently in Attic
tragedy but it was their use by Roman public men in solemn contexts3 that
made them attractive to a poet seeking !l theatrical equivalent of the Attic
TpaylJcr\ ~lS.
If conareis to beunderstood in Nonius' sense,it means 'are you about to?,
are you preparing to?'; cf. Pacuvius, Trag.227, Terence, Phorm.52, Haut.
240, Cicero, Fam. 5.12.1, Propertius 1.6.19.

VII
These words would have beenaddressedto the ~nlking Achilles.For their
n.
sentiment c£ Homer, 9.251 cpp<llEV6noos~avaolow aAE~O'ElSKCll<OV
i'jµap, Aeschylus.My,m. fr. 132 Tl TI'OT.
av6po6cmcrovCXKOVOOV lfiK6irov
ov mAa8e1strr· &pooyav;
The Latin wording suggeststhe conditions under which contemporary
Romans fought rather than those of theGreekheroic age. Homer's Achilles
was in no sensea fdlow citizen of Agamemnon, Ulysses,Ajax and therest.
Neverthelessthe Attic tragediansrepresentedtheexpedition againstTroy as
promoting a panbd)enic interest and employed arguments proper to city-
state patriotism (e.g. Euripides, I.A. 1368-401 ). To be compared with
Ennius' tetrameter are two passagesof Plautus, theslaveToxilus' parody of
1
Sec Haffier, Vnursuchungen,pp. 10 ff.
a Cf. Lex XII tab. 12.2 si seruosju,tumfaxit noxiamut no(x)it.

173
 COMMENTARY

the Roman general's prayer of thanksgiving at Persa753-62 (753 lwstibus

uictisciuibussaluis)and Pleusides' monologue on the dishonourable actions
caused by love at Mil. 1284~ ( 1289 mittoiamut occidi
Achilksciuispassusest).

8 serua clues, defende hostes: such asyndctic antithesis is extremdy

common in tragedy: c£Ennius 24fer mi auxilium,pestemabigea me, 38 tui me
tniseret,meipigetet al.
For the rhyming metra c£Ennius, Trag.205 Helenaredeatuirgopereat,224
tu me amorismagisquamhonoris,Naevius, Com. 76 alii adnutatalii adniaat,
Plautus, Amph.68 3 sicsalutasatqueappellas,1013 in medicinis
intonstrinis,
Bacch.
64, Cure.179, 285, 286, 297, Epid. 198, 230, 231, Men. 403, 1015, Mere.493,
833, Mil. 331, Pseud.695, True. 824- The laws governing the structure of
trochaic tetrameters all but prevent such rhymes in Attic tragedy. 1 Comedy,
despite its looser metrical structure, has very few. To the Aristophanic
examples collected by Fraenkel, HermesLXII (1927), 365 ff. ( ==Kl. Beitr. n
19 ff.), may be added Menandcr, Dysk. 717.
Ennius' use of defendereprobably had an archaic tone even at the time of
writing. The only parallel in comedy is Plautus, Rud.774 ut illasseruesuim
defendas(contrast Bacch.846 et al.).

cum potes defendere: verbs are frequently repeated in adjacent clausesin

both tragedy and early comedy.2

VIII
W elcker took this fragment to refer to a particularbattle and to come from
a speech made by one of the ambassadors to the ~nlking Achilles. Kluss-
mann' s palaeographically attractive restoration of the corrupt beginning,
int(ere}a, would make such an interpretation almost inevitable. Mortales,
however, unaccompanied by any determinant, is taken most naturally as
'men in general'. The fragment could come from a moralising speech com-
paring the bellicosity of men with the peaceful ways of the gods.3

9 mortales: substantive mortaliswas probably little used in the vernacular.

But tragedy hasit 5 times as against uir20 and homo10. The Plautine ratio is

1
The nearest I can find is Euripides, I.A. 869 X~Tt 1-1•Iv Tats o-cxtcn
cpepvaTs.
a Outside the prologues (And,. 10, 17, 18-19, Haut.13-14, 20-1, Eun. 4-S,
27-8, 41, Phorm.9-10, II, 22-3, 31, Hee. 10-11, Ad. 18-19) Terence tends to
avoid such repetition.
3 The gods depicted in early Greek epic and extant Attic tragedy are very far
from peacefulbut certain philosophers substituted a more edifying picture (cf.
Cicero, Nat. deor.2. 78), and Ennius seems to have made one of his tragic
heroes give voice to their doctrine at fr. CLXII.
174
 ACHILLES

45:24,6:780, the Terentian 0:65:177, that of the rest of comedy 2:5:23.

Caesaravoids the word and Cicero uses it sparingly (in the set phrases: multi
""""2ks, omnu m., amcti m.). The historians have it often.
pagnant proeliant: for the near tautology c£ Homer, 11.2. 121 ,ro).eµ{-
3Etvfi6tµax_eoeoo, Plautus, Cure.179 sibipugnassibiproelia,Lucretius 2.118
proeliapugnas edere,4.967 pugnareac proelia obire, Virgil, Aen. 11.912
ineantpugnaset proeliatemptent.Such tautologies are much more common
in Roman tragedy (c£ Ennius 65, 117, 148, 148, 156, 181, 198, 207, 246,
276, 285, 303, Pacuvius 32, 36, 44, 98, 175, 223, 303, 340, 369, 384, Accius
66, 90, 114, 136, 554, 587, Trag. inc. 73, 145) than in Attic. The Latin
poets were probably using the associations they had in the minds of their
hearerswith the abundant style of the official language. 1 Only one of the
Ennian examples occurs in a trimeter. In comedy Plautus restricted them by
and large to the trimeters of the prologue and the musically accompanied
verses of the action.:i Later drama employed them much more sparingly,
tragedy blurring the old distinction between trimetcrs and musically accom-
panied verses.
The 'asyndcton bimembre' has many parallels in the remains of Ennius'
tragedies (36, 65, 74-5, 90, 91, 116, 117, 166, 303), fewer in those of Pacuvius
(175, 263, 310, 384) and Accius (66, II4, 394), still fewer, comparatively
speakjng, in Attic tragedy.3 The official language again probably provided
the impetus.4 Comic asyndcta either consist of old-fashioned proverbs or
formulae of the official language or occur in the company of other stylistic
devicesaping the style of tragedy.
The deponent form proelianturmay have been regular in the common
language in Ennius' time (there is, of course, no significance in its absence
from comedy) as it was later and replaced with proeliantin thisverse for the
sake of homocoteleuton.
IX
The text which the codicesof Isidore present is full of difficulties.
The verbal repetition extolles. .. tolluntis of a type common in republican
drama5 but the two sententiaedo not seem quite appropriate to the one con-
1
Cf. C.I.L. 11 614 (189 B.c.) L. Aimilius L. f inpeiratordecrtiuit... agrum
oppidumqu.quodea tempestateposedisent,itempossiderehaberequeiousit.
:a Cf. Haffter, Untersuchungen,pp. 63 ff.
3 Of casesaccompanied by alliteration and homoeoteleuton I note Aeschylus,
Ag. 1410 arriSn<ESarrha~, 1553, Pers. 426, Euripides, Alk. 173, Or. 1302.
4 Cf. Lex XII tab. 6. 3 ususauctoritas
fundi bienniumest, 8. 10 uinctusuerberatus
igni necariiubetur.
S Cf. Pacuvius, Trag. 38-9 non tam ilium adpetit. .. quam ilium eumpse
(Vossius: tum ipsumcodd.) lapidem.•. petit, 410 intuenturneetuendisatietascapie,
175
 COMMENTARY
text. Whereasthe first dealswith evil liven and good liven, the second deals
with theevilly disposed and thewell disposed. Editors of Isidore have treated
the second as written by the bishop 1 but Ribbcck' s removal of enimproduced
a perfectly good trochaic tetrameter. L. Mueller inserted an idem alibi in
front of the tetrameter.
The Latinity of summamtu tibi pro malauitafamam extolleset pro bona
paratamgloriamis dubious at only one point. Comedy, Caesar, Cicero, the
historians and the classicalpoets have the phrase gloriam(laudem)parere
regularly. But gloriamparareis possible in republican tragedy. The metre of
Ennius, Trag.141 protects the hemistich libertatem paroagainst any suspicions
that might be raised against it by the regularity of libertatemparereelsewhere
in Latin. Ribbeck produced, with two alterations, quite regular trochaic
I
verse: summamtu tibi pro malauitafamamextolles,pro bonapartamgloriam.
The treatment of malaand bona,words normally iambic in shape,as pyrrhic
has good parallels in the iambic and trochaic verses of tragedy: e.g. Livius,
Trag.n, Ennius, Trag. 16, 41, 84, 163, 164, 181, 187, 258, 323, 327, 331,
Pacuvius, Trag.s~. 236, Accius, Trag.133, 147. Vahlcnproduced trochaic
rhythm without alteration to the tradition by dividing the epithet malaand
its noun uita between verses. I can find no good parallel for this in early
dramatic vcnc. 1
The distant disjunction of the antcposcd epithet summamcounts somewhat
against trochaic or iambic measurement. There are no parallels in any kind of
tragic versebeforeAccius(Trag. .134 -s,
1 658-9).3
I have no solution to theseproblems.
For the sentiments c£ Cato, Or. 17 si quiduosperlaborem rectefaceritislabor
illea nobisdto recedet,bentf actuma uobis,dumuiuitis,nonabsadet;sedsi quaper
uoluptaltmnequiter faceritis,uoluptasdto abibit,nequiter
f actumilludapuduos
sempermanebit,Cicero, Tusc.3 . 3 est enimgloriasolidaquaedam reset expressa,
nonadumbrata: eaestconsentiens lausbonorum,incorruptauox bentiudicantiumde
excelltntiuirtute,ea uirtuti resonat°tamquamimago.quaequia rectefactorum

potest,Plautus, Aul. 202-3, &lcch.980, Capt.65-6, Men. 122-3, Pstud. 549-50,

Terence, Hee. 8, Phoms.265, Fraenkcl, in Studienzur Textgeschichttund Text-
leritile(Cologne, 1959), pp. 21 £ If one excludes phrases like 1<a-rf6ET't&-n
(Euripides, Med. 1252; see Elmsley ad loc.) the type is very uncommon in Attic
drama.
• So too Scrivcrius; see Collectanea,p. 182.
1
But cf. Ennius, Trag.283-4 dati I sunt, 316-17 ex I inctrtis.
3 Ennius and Pacuvius disjoin epithet and noun with scarcely any greater
boldness or frequency than do the comedians; like the latter they disjoin much
more often in musically accompanied verses than in spoken trimcten. Accius
appean to have made no distinction between sung and spoken verse and to
have been about three times as free as his prcdcccsson in placing epithets.
 AIAX
plerumquecomesest,nonest bonisuirisrepudiatula. ilLiautem,quaese eiusimita-
trictmesseuolt, temerariaatqueinconsiderata
etplerumquepeccatorum uitiorumque
lauJmrix, JamapopuLiris,simulationehonmatisf ormameiuspukhritudinemque
corrumpit.
For the distingnisbing of words related in meaning c£ Euripides, Andr.
~a ~a µvp{o1ai S..,pPar&'>vI ovSw yey&'>cn~{OTOv~YKCA>-
31S>-22 cl'>
acxsµfyav, I EOKM1a 6" ots ~ lo-r' aAri&{cxs wo I ev6ooµov{300, Med.
1229-30 6AJ:kn,
1
s· hnppvwTOSarrvxio-reposI&XA.ovyw<>lT &, &Mes,
1

ev6a{µoov6 &, oO, Accius, Trag. 4-9, 296,Trag. inc. 30. Grammatical and
rhetorical theorisingprovided the impetus as much in seconckcntury Rome
as in filth-century Athens.

10 tu tibi: c£ Plautus, Ann. S24, Cure. 9, Pseud.

936, et al.

&rnam extolles: c£ Nacvius, Trag. 7 desubitofamamtolluntsi quamsolam

uiderein uia,Plautus, Persa3sI inimicif amamnonita ut natastf erunt,Trln. 186
ha.seine
propterresmaledicasf anuuf erunt,689 nemi honef amamdi.fferant.There is
a tendencyin tragedy to replace simple verbs regular in the common lan-
guage by compounds with ex-; sec below on v. 20.

AIAX

The title Aiax is given to Ennius by Vcrrius 1 and Nonius. 2 Nonius gives the
same title to Livius.3 Livius' one fragment, mirumuideturquoJsitf actum iam
Jiu,is of indeterminate subject-matter. 4 The fragment ofEnnius' play quoted
by V crrius suggests that the death of Ajax was the main theme. V ahlen
interpreted the fragment quoted by Nonius in such a way as to carry the
action back to the debate over whether Achilles' armour should be given to
Ajax or to Ulysses.
Greek sources record the title Alcxsagainst the names of Sophocles,
Carcinus, Theodectesand the younger Astydamas. Scaliger's view5 that
Ennius adapted Sophocles' extant Alcxswould be upset if Vahlen were right
about fr. xm or if it could be shown that Livius had already adapted this
play.
1
Fcstus,p. -482.S, a P. 393 • IS,
3 P. 127. 20. The source is obscure at all three places.
4 Ribbcck (Die rom. Trag. p. 26) and Leo (De Trag. Rom. p. 6 [ = Ausg. ltl.
Sehr. I 195)) refer it to Greek ingratitude towards Ajax, Ribbcck supplying
mirumuideturquodsitf actum iamdiu( oblitosesse)?,Leo ( mortalibus)mirumuidetur
quodsitfactum iam diu; cf. Sophocles, Ai. 646 tf., 1266 tf.
s Coniea. Varr. Ling., on 6.6.
12 177 JTO
 COMMENTARY
At least two other republican tragedies, Pacuvius' Amwrum iudidumand
Accius' tragedy of the same name, had Ajax the son of Tclamon as a central
figure. The action of Accius' tragedy almost certainly included the death of
Ajax.1 Ajax appeared as a minor figure in a scconclEooiao tragedy, either the
Achillesor the Hectorislytra?
Cicero distinguishesbetween 'Homericus Aiax' (Diu. 2.82, Tusc.4.49)
and •Afax fabulacque' (Scaur.3). He refen to tragic presentations of the
debate between Ajax and Ulysses before the judges (Off. 3 .98), Ajax's
demented attack on Uly~ (De orat. 3. 162, Ac. 2. 89) and Ajax's death
( Tusc.4. 52, Off. 1 . 113, Scaur.3). Since he nowhere quotes Livius verbally
and on one occasion refcn to his plays with some dicdaio,3we may suppose
that the Aiax to which he refers at Off. 1. 114 is Ennius' tragedy.
Ribbcck4 attributed the trimetcr quoted by Cicero at Off. 1 . 61 without
name of author and by Festus at p. 439. 17 with that of Ennius (fr. CLXXXI)
-Salmadda spoliasinesudoreetsanguine-to a sceneof the Aiax in which Ajax
molested a herd of cattle, imagining the cattle to be the Greek princes who
had deprived him of Achilles'armour.SThe attribution is plausible.It would
not conffictwith the view that Ennius adapted Sophocles' Af cxs. In thisplay
Ajax's treatment of the cattle was describedby Athena but a Latin adaptation
might well have substituted an actual scene. In adapting the •A&Aq,o{of
Menander, where, it scems,6 the young man's seizure of the girl from the
procurer was merely reported, Terence added after the introductory dia-
logue between the two old men a sceneshowing the actual seizure.This must
be the scenewhich, in the prologue,7 Terence claims to be a careful vcnion
of a sceneofDiphilus' Iwcnro8vtj0'1<0VTES omitted from Plautus' adaptation
of that play. Ennius of course was one of the playwrights to whose authority
Terence appealed in justification of his free treatment of Mcnandcr' s text.
However I leave the fragment among the incerta.
The Laurentian codex ofVarro's treatiseDe linguaLatinahasat 6.6 a piece
of vcne introduced by the words 'cnnius aiax '. At three points in books s and
6 Varro heads quotations of tragedy with the name of the speak.er(s. 18-20,
6.2-7, 6.81-2), twice clearly distinguishing this name with that of the
author of the tragedy: 6.6 Pacuidicitpastor,6. 7 in BrutoCassii (sic) quoddicit
Lucretia(c£ 7. 3 etiamTeucerLiuii post zy annosah suis qui sit ignoretur).
1
Welcker, RhMm (1829), S4 n. 39, and G. Hermann, De Aeschyli Tragoediis
Fata Aiaciset TeucriComplexisDissertatio(Leipzig, 1838), pp. s ff. ( = Opusc.
vn 365 ff.), thought that Pacuvius'tragedy also extended as far; but cf. Leo, De
Trag.Rom. p. 12 (= Ausg. kl. Sehr. 1201).
s Ennius, Trag. fr. cuavb. See above, p. 162. 3 Brut.71.
• ~st. seen.p. 272; cf. Die rom. Trag. pp. 131f.
s Cf. Trag. inc. 47-8.
6 Cf. Ad. 88-92. 7 Ad. 6-u.
178
 AIAX
Cicero frequently employs thesame method of citing drama.1 His purpose is
sometimes to mark off a Latin poet's handlingof the pcnonage from a
funous Greek handling but usually to separate two well-known Latin
baudlin~ We may thercforc2CCCpt as very probable Laetus' reading Enni(i)
A.iax at Varro, Ling.6.6. 3

Scaligerexcluded thewords which Varro quotes-lumen iubarnein caelo

cerno--fromEnnius' Aiax on thegrounds that there is nothing correspond-
ing in thetext of Sophocles'Alas. So much the worse, one mightreply, for
theview that .Enniusadapted Sophocles'play. In factthewords clearlycome
from a scene set just beforedawn3andone needpostulatevery littlechangeto
theopening of theAlas in order to accommodate them in a typical second
century B.c. Latin version. It is impossible to imagine them in either the
Achilltsor the Hectorislytra.l therefore follow Columna in printing them as
a fragment of theAiax.
Some words quoted by Nonius at p. 207. 32-praest4tUr uirtutilausgeluset
multoociustuenio LBA/ucntoPt tabtsdt-in order to illustrategelu neuter
appear to contain a sententiarelating to thetransitorinessof a man's repute
among his fellows.Ribbcck4 saw thesimilarity with Sophocles.Ai. 1266--71
cpeO TOV8av6V'ToSws T<JXEla TlS~parols I xap1s 61appelKOO ,rpo6ova"
&Afamm, I el aov y• 66' <XV'l'IP ovs· hrl aµncpC:,v '>..oyoov, I Alas, h'
IO')(ElµvfiaTtV,OUav1rOAAaKISI Tf\V0'1)V irpoTEfvc.,,v,rpoOK~ 'fNXf\V
Sop{· IciA.A'oiXETcxt 611 iraV"TaT<XVT' App1µµh,aand. with thehelp of cod.
Vrbinas 308, corrected the introductory titus liuiuspisi adt«ematico foro to
Liuius in Aiace mastigophoro. Sophocles' extant Alas was often called by
ancientscholars Alas µaanyoq,oposs so as to distinguish it from his tragedy
concerning the Loaian Ajax. There can be little doubt therefore that proe-
st4tur • •. tabesdt comes from a Latin version of Sophocles' Alas µaaTtyo-

1
E.g. Tusc. 1. 105 Acdus et aliquandosapiensAchilles, 1. 1o6 apud Ennium
Thytstes, 2. 1 Neoptolemus. •• apud Ennium, 2 • 48 in Niptris ille sapientissimus
Graeciae. .. Pacuuius,3 . 62 ille AgamemnoHomericuset idemAccianus,3 . 63 ilia
apudEnnium nutrix, 3 .65 ille Terentianus'ipse sepoeniens',4.67 ille apud Tra-
beam,4.67 Naeuianusille... Hedor, Fin. 1.3 TerentianusChremes,4.62 Enni
Alcmeo, Rep. 1. 30 Zethum ilium Pacuui, Nat. deor. 2. 89 ille apud Accium
pastor,Diu. 1.44 Superbi Tarquinisomniumde quo in Bruto Aai loquituripse,
1 .131ille Pacuuianusqui in Chryst physicusinducitur,2. 133 PacuuianusAmphio,
De oral. 1 . 199 apud Ennium. .. ille Pythius Apollo, Fam. 7. 33 Philoctetesapud
Acdum.
a Vahlen, E.P.R. 1, p. 95, RhM XVI {1861), 578 (= Ges. phil. Sehr. 1417),
suggested independently Enni(i) Aiax and EnnianusAiax.
3 Cf. A. La Penna, Maia v (1952), 94.
4 Monatsb.Preuss.Ak. d. Wiss. z. Berlin1 {1854), 45-6.
s Cf. the hypothesis of this play and Athenaew 7. 277 c.

179 12·:2
 COMMENTARY
cp6pos. 1
It is, however, by no means certainthat Livius was the author of the
version. I should like to dagger pisi and mark after it a lacuna; this lacuna
would have contained gelu in the nominative or accusative case and another
name, perhaps that of Ennius rather than that of Livius, as the author of Aiax
mastigoplwrus. I should regard the Aiax mastigoplwrus piece as an intruder in
Nonius' gelu/gelusarticle.
The type of ablative of comparison presented by the transmitted text is
unparallded in archaic Latina but gelu can hardly be taken otherwise and
such ablatives as that in Plautus, Poen. 812 leuiorplunuiestgratia arc not very
differcnt.3It is thus plain that the quotation doesnot illustrateNonius'
lemma, gelu neuter (or, for that matter, gelus masculine).
The third book of Nonius' dictionary (' de indiscretis generibus ') appears
to have useda work of Flavius Caper as its base and to have supplemented
the quotations it drew from Caper with quotations from the 41 listsdrawn
on in the other books.4 A quotation of an Aiax mastigophorus of Livius could
not come at this point from any of the 41 lists. There is no evidence that
Caper quoted any Livian tragedy by name, very little that he quoted any at
all.S The historian Titus Livius, on the other hand,who is quite absent from
the 41 lists,6 seems to have been quoted often by Caper.? Two quite certain
quotations of the historian appear in Nonius' third book8 and one rather
dubious one.9 I should therefore postulate that Nonius took from Caper a
piece of the historian illustrating gelu neuter and added from memory the

1
The arguments which K. Ziegler, RE 2 VI ii (1937), s.v. Tragoedia,1985,
brings against i<!entifying the substance of prMStatur . .. tabescitwith that of
Sophocles, Ai. 1266 ff. arc quite inconclusive. For the introduction of imagery
foreign to the original cf. Livius, Carm. fr. 16 igitur demum Vlixi co, Jrixit prae
~
pauore Homer, Od. 5 .297; Ennius, Trag. 229-30 ille traunsamenlt mi hodie
~
tradiditrtpagulaI quibusegoiramomnemrecludam Euripides, Med. 365 ff.
a See E. Lofstedt, Syntactica1• (Lund, 1942), pp. 304 ff.
3 See Drexler, GymnasiumLXm ,(1956), 165.
4 Sec Strzelecki, Bos XXXIV (1932/3), 113 ff., De FlauioCapro,pp. 1 ff.
S Ribbcclc includes in his collection the pieces attributed simply to 'Livius'
at Nonius, p. 197. 31 and Priscian, Gramm.II 231. n.
6 TitusLiuius at p. 368 .29 is an obvious error. The words uestispullapurpurea

ampla could be a version of Homer, Od. 19. 225. Nonius quotes Liuius in
Odyssiaatp. 475. 16 and 493. 16 and introduces apparent venions ofOd. 1. 169
and 1. 136 with Liuius at pp. 509. 28 and 544. 20. Livian tragic titles come into
the dictionary in plenty from the alphabetical lists of adverbs and verbs and
from glossaries dependent on Verrius Flaccus.
7 Cf. Priscian, Gramm.u 141.1- 171.15, 194.1-282.18.
1
P.196.16TitusLiuius(= 6.40.18),p.197.20Liuiuslib •.x11(= 22.14.8).
9 P. 194.20 Liuius lib.1.x: aurataeuaginaeauratabalteaillis erant;cf. Homer,
Od. II .610.
180
 AIAX
Aiax mastigophorus piece, fajljng t.o see that it did not illustrate hislemma. He
might have seenit in his alphabeticallist of verbs illustrating theuncommon
1
tabesare. I should furtherpostulatethat the archetypeof our manuscripts of
Nonius' dictionary bad lost all but the fust word of theTitus Liuius quota-
tion as well as thename of theauthor of the Aiax mastigophorus. 2
In any case
praestatur ••. tabescitshould be excluded ttom thetragic fragments of Livius.
There is at least some possibilitythat it belongs to Ennius.

X
See above, p. 178.
XI
Columna's inclusion of thealiquodof Ling. 7. 75, which makesno senseat all
in thecontext ofVarro's preceding discussion,in theEnnian 6.-agment3 is un-
convincing. The fragment occurs at 6.6 and 6.81 with no trace of an
aliquodand one canoffer no explanation as to why theword should have been
omitted at thesetwo placesand included at 7. 75. The in altisonocaeliclipeoof
5.19 is deprived of its introductory Agamemnoat 7. 73 becausethat is not the
style of introduction normal in book r,4but extended to quidnoctisuideturin
altisonocaeliclipeotemosuperatstellassublimeagensetiamatqueetiamnoctisiter
becausea different point is at issue.

13 lumen iubame: iubaris sometimes taken as a neuter adjective with

lumen.The word appearsasa masculinenoun in thehighly artificiallanguage
ofEnnius' epicSbut is normally a neuter. The terms of thediscussionat Ling.
6. 6 and 7. 76 suggest that Varro took it as a noun here. Vahlen punctuated
his text: aliquodlumen, iubame?in caelo,cerno,interpreting it as 'aliquod
lumen {q,ws T1)in caclo cemo: iubame est?"
For iubaras themorning star c£ Ennius, Ann. 557, Lucretius 5 .697,Anon.
Aetna 333; contrast Plautus' diumastella(Men.175) and Pomponius' lucifer
(Atell. 74).
1
Strzelecki, De FlauioCapro,p. 30 n. 1, sees the difficulty of tracing it to
either Caper or the 41 lists and assumes it to be from a marginal scholium.
a Similar lacunae have to be postulated at pp. 75.8, 90.10, u6.8, 170.12
{liuiusantiopa!),176. 12, 209.25, 223 .2, 479. 13, SIS. 12, et al.
3 Cf. Vahlen, Ind. lectt.Berlin 1880, 14 ( = Op. ac.I us).
4 At 7. 72-6 Varro quotes many of the same passages as he does at s.18-20,
and 6.81-2, in all probability from the same source; he alters the mode
6.1,,-J'J,
of introduction of each passage to one of those usual in book 7.
s 557 intaeafagitalbusiubarHyperioniscursum(~ 6 q>(l)O'f6pos). Cf. Anon.
Aetna 333.
6
E.P.R. 1, p. 122. Cf. Op. ac. I us.

181
 COMMENTARY

Lu~n was a word of poetic tone: it occurs 12 times in republican tragedy,

its synonym lux only twice; in comedy on the other hand lux occun 29,
lumtn 4 times (at Plautus, Cist. 643, True.518 as a pocticism for uita;at Cure.
95, 117 in paratragic dialogue). Most fonnations in-mm had a similartone;
H. Ploen1 counted 18 in 1,940 verses ofrepublican tragedy, 8 of which do not
occur at all in comedy, and only 24 in 30,000 verses of comedy.

cemo: in the meaning 'perceive' this verb occurs 8 times in republican

~ ~
9) andonly 7 in comedy ( uidere
tragedy( uidere45, audire11, intellegere
1,270, audire540, intellegere
125).
XII
The corpse of Ajax is beingdescribed: c£ Sophocles,Ai. 917-19 ov&ls &,,
6cms KaitlAcs, TAOOfl ~Atrre1vIq,vaooVT'&,oo irpos ~tvas 1KTEfOtvfas I
1rAT}~ µe).aveb, atµ• arr·olKE(QS O'~S. 1411-13 rn yap8EpµalI
avp1yyes &,oo ,vo-0001l,10.avI µwas.
For theimagery uolantc£Homer,OJ.22. 18-19CXVT(Ka s·
of tullii e.fllantes
CXVAoS &va ~tvas iraxvs i\1'8EvI aTµarcs av6poµio10, Sophocles, Ant.
1238-9 Kal ,vmoov 6~tav ~• f,oilv I AMCij irape1~ fOtvfov O'Ta-
1'ayµarcs, Euripides, Hek. 567-8 -riµve1m6fip~ ,rvevµaros 61appoas·
I KpoVVOl s· ~~pow, Rl,es. 790-1 8Epµas& KpOWoS &airarov ,rapa
a~ats I ~• µe sva6v1!)aKov-ros atµarcs vrov, Lucretius 2.354 san-
guinisexspiranscalidum depem>re jlumen, Virgil,Aen. 9.414-15 uc,luitur ille
uomenscalidumdepectore}lumenIfrigidus,700-1 redaitspecusatriuulnerisutulamI
spumantem,et.fixoferrumin pulmonetepesdt,11 . 668 sanguinisilleuomensriuos.
Vahlen understood the transmitted text as (ferroperCUJSUS iacet}I Aiax:
uolant.The resulting enjambement iacetI
missosanguinetepidotullii e.J!lantes
Aiax, the asyndeton Aiax: missoand the dactylic word sanguineoccupying
the third foot of the trochaic tetrameter are all highly unusual Compara-
tivdy speaking, even fewer so-called 'split anapaests' and the like are trans-
mitted in the iambic and trochaic verses of early tragedy than in those of
comedy3 and most that are transmitted occur in thecompany of other kinds
of anomaly.3The only credible restoration is Hermann's animammissosangui
1
De Copiae VerborumDifferentiisinJerVaria PoesisRomanaeAntiquioris
GeneraInteradmtibus(Diss. Strasbourg, 1882), p. 36.
3
The situation in comedy was observed and described by Hermann. Elt-
mentaDoctrinuMetrlcae{Leipzig,1816), p. 78, Ritschl, Prolegomena to Plautus'
Trinummus(Bonn, 1848), pp. CCLXXff., Lachmann on Lucretius 2.719 and
C. F. W. Mueller, Nachtriige zurplautinischen
Prosodie(Berlin, 1871), pp. 12 ff.
Opinions have differedwidely about the caseswhich defy the rigid rules laid
down by these early critics.
3 Cf. Ennius, Trag. 33, 86, 204, 263. For a permissive discussionof the
Enoiao examples sec Vahlen, HermesXVD (1882), 604 ff.
182
 AIAX
tepidotullii tj/lanttsuoldltt.1 However the ablative sanguiseems to be un-
exampJedin extant Latin and present participlesrarely govern objectsin the
remainsof republican drama.2

~ miao IIIDg1linetepido: theparticiple seems to have a present force;

cf. Nacvius, Trag. IS l«tus sum lauJmime abste patera lmulatouiro,Livy
1.37.1, 2.36.1, Scrvius, Vcrg. Georg.1.206-7.
For mittere'cmittcre • cf. Luaetius 2. 194 quoJgenuse nostrocummissuscor-
poresanguisIemicatex.sultans altespargitqueauorem.It is possiblethat sanguinem
mitterewas theregular phrasein Ennius' day but more likely that we have an
archaisingremoval of the customary prdix; cf. in Attic tragedy alvdv,
ija8cn, &v1!lcn<E1v,1CTE(ve1v,6Mvva1, in Ennius' adaptationsmissasum(3S =
emissasum),uereor(37 = reuereor), sumptus(S3 = consumptus), sl4tuerit(7S =
constituent),locant (106 - collocant),linquere {128 = relinquere),umunt
{166 = dtumunt), ibit (22s = abibit),p•at {3o8 = comp•at),puldt (308 -
reputat).
tallii: not elsewhere,it seems, in recorded Latin; perhaps a word of rustic
tone even in Ennius' day. Tragedy admitted few words which obviously
originated outside thedialectof Latin spokenby the Roman upper classes.
Thr-classicalgrammariansbranded theungulusof Pacuvius, Trag.64 and 21s
and thcfamulusofTrag. inc. 138 as Oscan. For words of Greek origin sec
below on v. 67.

eftl•atel: for the intransitive use cf. Lucretius 6. 682, 699.Intransitive

fo,,e is very common.
XIII
Ribbcck set thesewords in a speechlike that which Ulyssesdeliversat the
end of Sophocles'Al~ (c£ 1338-41 &XA'CXVTOV (µ,r~ wr· fy~ TOlOVS'
tµol I OVKavrcrnµaacnµ' CXV COO'T£µ~ Mye1vI Iv' 6v6p' lSetv&ptaTOV
6aot I Tpo(av aq,1K6µea8a,
'Apye(c.,.,v, ,r~~v 'Ax~). Vahlen on the
other hand3 suggested that they come from a speechby Nestor in which he
advisesthat the Trojans should be consulted asjudges in the dispute between
Ulyssesand Ajax. Neither view is cntirdy satisfactory.
n.
The words refer to crraS(fl vaµ(VT)(cf.Homer, 13. 313 f., 713, Strabo
10.-449), a form of fighting in which Telamonian Ajax was pro-eminent.

1
On Sophocles,Ai. 918, ed. 3 (accordingto Ribbcck). Cf. Euripides, Ba.
620 8v"'6v~. Or. 1163 1T<XVTC,,)S
IKTivtwvlf'VX-iiVlµfiv, Plautw, Persa638
animamecflauit,Virgil, Aen. 2. 532 conciditac multouitamcum sanguinefadit.
3
C£, however, Ennius, Trag.55, 63, 10s, 19<r1, Terence, And,. 75, Eun.
584, Hee. 163, 365-6. 3 E.P.R. 1, pp. ea, 121.

183
 COMMENTARY
I should interpret them as part of a sentence to the effectthat Ajaxwasone of
those quiremcumAchiuisgessn-untstdtim,i.e. as belonging to a freeversion of
Sophocles, Ai. 1266 ff. (esp. 1269--70 oo (Agamemnon) av (Ajax),ro).-
AaKlSI 'lT\Vcniv irpcrrEiv<A>v irpo(iKaµes'f'VX~V6opf).
The phrase remgererebelonged properly to the official language (cf. Varro,
Ling. 6. 77 imperatorquoddiciturresgererein eonequef acitnequeagit,sedgerit,id
est sustinet).Where warlike operations were concerned cum could be used
not only of hostility to an enemy (Cicero, 'Prou.IS, Off. 3. 108, Livy
7.26.13, 9.16.2, 28.12.1) but also of co-operation with a colleague
{Anon.Bell.Aft.10.1, Livy 10.21.14, 21.40.3).

15 Achiuis: 6 times in tragedy and once in comedy (Plautus, Batch.

936) of
the heroes who besieged Troy. Graed,the name by which second-century
Romans knew the Greeks, does not occur in tragedy. There are three in-
stances however of Grotcia.Achiuiwould have been the old Latin name for
the Greeks settled in Campania (' Axoorof); how it was preserved until the
time of the tragic poetsis hard to say.

atatim: not in republican drama outside the five passages quoted by

Nonius to illustrate the meaning 'perseueranter et aequaliter'. Terence,
Phorm.790, Plautus, Amph. 239 and 276 neither confirm nor disprove
Nonius' strange doctrine about the quantity of the a. The text of the
Afranius piece is unsound while the Ennius piece under discussion could be
scanned either x - quiremcumAchi:uis gesseruntstatimor, no less plausibly, qui
remcumAchiuisgesseruntstatimv - ; in the latter casethe law of Bentley and
Luchs (seebelow on v. 308) would require a long a.

ALCMEO

Nonius gives the title Alcmeoto Ennius once 1 and to Accius nine times. 2
The stories about the hero AlCineorelated by the mythographers are many

1
The source ofp. 127.15 is Lindsay's list 28 'Alph. Adverb'.
3
Most of the quotations come from Lindsay's list 5 'Accius i'. About
pp. 181. 16 and 18-4.1, where the codices give Alcmena/-eas the title, there is
some doubt. Ribbeck argued (N]bb LXXVU(1858), 192 ff. [in Ritschl, Opusc.n
512 ff.]; cf. Ritschl, RhMvm [1853), -477, xn (1857), 102 f. [ = Opusc.n-475 f.,
-485 f.]) from the discussion of Marius Victorinus at Gramm. VI 8. 6 ff. and
Priscian at Gramm.n 29.7 ff. and from the readings of the codices ofNonius'
lexicon at pp. 16. 19 (alcimeone),116. 1-4 (aldm«hone) and 393 .28 (ale~)
that Ennius spelled the hero's name 'Alcumco • and Accius 'Alcimco '. Two dis-
tinct questions have been muddled here: tint, what form did the dramatists use
184
 ALCME0
and difticult to sort out. 1 He killed his mother Eriphyle because of her
betrayal of his father Amphiaraus. Like Orestes, an Argive of later times, he
was pursued by the Furies from one foreign city to another and protected
against these demons by Apollo. He produced children by at least three
women, Manto, the daughter of the Theban seer Teiresias, Alphcsiboea
(Arsinoe,according to Apollodorus), the daughter of Phegeus, king of
Psophis, and Callirrhoe, the daughter of the river god Achelous. The neck-
lacewith which Eriphyle had beenbribed to betray Amphiaraus was alsothe
instrument of Alanco's destruction. When Callirrhoe learned that it had
becs1given to her predecessorAlphtsiboea she demandedit for herself.
Alanco went to Psophis to recover it and was slain by the brothers of
Alphcsiboea.
The one fragment of Ennius' tragedy quoted by Nonius-f actumest iam
diu--appean to refer to the killing of Eriphyle and to set the scene away
from Argos at a time long aftcrwards.2 Of the eight fragments of Accius'
tragedy two are potentially informative. That quoted at p. 487. 29, qui
duut cum te uideritsocerumgeneribustanuunesse inpietatem?,was very
plausibly interpreted by W dcker3 as coming from an address to Phegeus
by one of his sons after the discovery of Alcmco' s plot to recover the
neddacc. No satisfactory interpretation of the quotation at p. 393 . 26,
suosdeseruitliberos. sunt, has been found. G. Kr6kowski4 referred
superstites
it to the childrenAlcmco had of Manto and set Accius' tragedy in Corinth
around the theme of Alcmco's reunion with the children. According to
Apollodorus (3 . 7. 7. 2) Alanco had entrusted them to the care of the
king of Corinth. It is thus difficult to explain the word deseruitin the
Accian fragment. Psophis remains a more likely scene for the tragedy
than Corinth.
Alcmco was quite certainly a personage of Accius' Epigoniand possibly of

in the text of their plays?; second. what form appeared at the colophon and
upon the cri~vj3os of the rolls consulted by Nonius and the lexicographen
upon whom Nonius drew? The texts of republican drama present the name
only twice: Plautus at Capt. 562 almost certainly wrote Alcumeus;at Accius,
Trag.78 one could restore the genitive of this form or the classicalAlcmeonis(so
L. Mueller, Dt A«ii FabulisDisputatio (Berlin, 1890 ], pp. 8 f., referring to
Varro, Ling. 10.70 and Accius' introduction of Greek forms of proper names)
equally as well as Alcimeonis(Ribbeck: almeoniscodd.).
1
SeeC. Robert, Diegriechische HeldtnS1Jge mi (Berlin, 1921), pp. 956 ff. The
principal sources are Apollodorus 3 . 7. 5 ff. and Pausaoias a.24. 7 ff.
:a Cf. Cicero's reference to the exsultati-o
... seniummatriddarumat Har. resp.
39.
3 Die griech.Trag. p. 283.
4 In TragiC411, 57 ff. (68).

185
 COMMENTARY
the Eriphyla1 and the Alphesiboea.iThe Epigoni was set in Argos and dealt
with the killing of Eriphyle and the second Argive expedition against
Thebes. The one fragment of the Eriphyla-PalLu bicorporanguiumspiras
trahit-is enigmatic. Those of the Alphesiboeahave been interpreted to refer
to Alcmeo• s second visit to Psophis and his death at the hands of the brothers
of Alphesiboea3 but if thiswas the theme of Accius• Alcmeoone ought not to
postulate without very good reason a second handling of it. L. Muellcr4
referred the fragments quite plausibly to a play about the revenge Alphesi-
boea took on her brothers,5 H.J.Mette 6 to a play about Alcmco's first com-
ing to Psophis and marriage with Alphesiboea.
Cicero frequently mentions Alcmco ( Tusc.3 . 11. Fin. 4. 62, Ac. 2. 52, 88,
89); he describeshim as a uir bonus (Fin. 4. 62 )7 and quotes two tragic
speeches from his mouth. one in trochaic tetrameters (Deorat. 3. 154, 218,
Tusc.4.19, Fin. 4.62, 5 .31, Hortensiusap.Prise. Gramm.11250. 12-fr. XIV),
the other in a mixture of anapaests and no longer discernible metra (Ac. 2. 52,
88-9-fr. xv). Ennius'name is three times connected with theformer butnot
at all with the latter. Columna gave it to Enniw 8 and subsequent editors
have all followed him.
The speech in lyrics is addressed to a uirgoand describesan attackby the
Furies upon Alcmeo and the arrival of Apollo and Diana to rescue him.The con-
text of Cicero• s discwsion and Alcmco• swords sedmihi neutiquam corconsentit
cumoculorumaspectu make it clearthat the demons are figments of the hero's
imagination and could hardly have beenrepresented on the stage by actors. 9

1
Q!!oted only by Priscian at Gramm.u 236.6.
a Q!!oted by Nonius at pp. 15.6, 73 .25, 136. 16, 279.35, 280.4, ,469.2s.
485. 2s, 497. 2, s12 .14 from Lindsay's lists 'Accius i '. the same source as pro-
vided most of the Alcmeoquotations. P. Schmidt (De Nonii Marulli audoribus
grammatids,pp. 52-6s) demonstrated this fact and thus destroyed Bothe's
theory that Alcmeoand Alphesiboeawere alternative titles of the one play.
3 Cf. Scholl, Beitr. pp. 132 ff., Welcker, Die griech. Trag. pp. 278 ff.,
Hartung, EuripidesrestitutusI (Hamburg, 1843), pp. 187 ff., H. Grotemeyer,
De L. Accii tragoediis(Diss. Munster. 1851), p. ss. Ribbeck, Die rom. Trag.
pp. s01 f., Schadewaldt, HermesLXXX (1952). 46 ff. ( = Hellas und Hesperien,
316 ff.).
4 De Acciifabulis, pp. 10 f.; cf. Robert, Die griech.Heldensage mi, p. 963.
6
s Cf. Propertius 1 .1s. 15-16. Lustrum IX (1964). 142 f.
7 Cf. Ovid, Met. 9. 408 factopius et sceleratus eodem.
8 Stephanus (p. 109) had expressly excluded it.

9 This was fint seen by Wilamowitz, Ind.schol.Gottingm1893, 12-17 ( = Kl.

Sehr. I 185-91). cf. G. Perrotta, SIFC N.S. VI (1928). 127-32. Cicero thrice (S.
Rose.67, Pis. 46, Leg. I .40; cf. Suetonius, Ner. 34) refers to stage Furies using a
poetic phrase-taedis ardentibus-which occurs in the tragic anapaests. Ennius'
Eumenidescontained real stage Furies and it is possible that the phrase occurred
186
 ALCMEO
The audiencewould have seen only Alcmco's reactions,as in the parallel
situation of Euripides'•~s. 1

Ciccro's phraseology at Ac. 2.89, quid? cum uirginisfidem implorat .• .•

containsnothing foreign to his usagc:3but it loob as if he is echoing the
tragic Alcmco's actualaddressto a person unknown. Otherwisehe would
have given the person's name or status (i.e. andll« or filiae). In any case,
whether the phraseologybelongsto Ciceroor to Ennius,it suggestsstrongly
that the personaddressedappearedto be Alcmco'ssocialsuperior (i.e. some
citizen of the town to which he had come). In such passagesof republican
comedy as Plautus,Amph. 376profolemThebanidues,Amph. 1130 di obseao
uostramfa/em, Men. 999-1000 ptrii, opseao uestramfa/em I Epidamnienses
subuenitedues,Rud.615-24 proCym,ensespopulatesuostramegoimploro fa/em
.. .ferteopeminopiae. .. uostramiterumimploro folem I quipropehie adestis . ••
f erte suppetiasthe wordfoleshad its old senseof' protectivepower• and called
to mind the client-patronrclationship.3If in comedy a character entrusted
birnsc:lfto thefidesof a socialinferioras at Terence,Eun. 885-7, nuncegote in
hacre mi orout adiutrixsits; I egome tuaecommendo et committofole; I te mihi
patronamcapioThais,te obseao,a mildly humorouseffectwas being sought
from the incongruityof the language.The word uirgonormally denoted in
tragedy and comedy a young woman of free status not yet married or a
woman in the handsof a slavedcalcrwho had not yet been sold on the retail
market. Hartung's view4that the uirgowas Tisiphonc,the daughter whom
Alcmeohad had by Manto and boughtby mistakein a slavemarket, and
that the scene was parallelwith that of Euripides,Ion 725 ff. is therefore
hardly a acdiblc onc.SWclckcr's vicw6 that the uirgowas the daughter of

there. On the other hand Ciccro's language tends to take on a poetic colouring
whenever he speaks of the heroic world, whether or not he has a particular
Roman tragedy in mind; the only other occurrenceof taedain his works is at
Vm. 2 ••• 106, where he rccowits the wanderings of Ceres. Zillingcr, Cimo
und die altromischenDithter, p. 29, is wrong in any case to treat the three
passages as ttstimoniato the Alcmeo.
1
Or. 253 ff. C£ Aeschylus, Chot. 10•8 ff., Euripides, El. 13~. I. T. 281 ff.
Scrvius appears to say in a note on Aen. •••73 that in a play of Pacuvius real
Furies attacked Orestes as he came out of Apollo's temple but Virgil's com-
parison of the figures of Dido's dream with the demons that beset Penthcus
and Orestes only has point if these were imaginary demons.
a Sec T LL VI i 66s. 73 ff. onfakm alicuiusinplorore.
3 Sec Fracnkcl, RMI uca (1916), 193 ff. ( = Kl. Beitr.I 21 ff.).
4 Euripidesrestitutusn (Hamburg, 184-4), p. 536; c£ Ladewig, .Anal. sun.

p. 29.
s Hartung himself was troubled by the word uirgoand suggested that it was
used to prevent anybody thinking that Alcmcowas guilty of incest as well as
6
matricide. Dit gritch.Trag. pp. S7S ff.
187
 COMMENTARY
Phcgeus is tenable but one finds it a little odd that the king's unmarried
daughter should bestanding in the street in such a way as to beaccostedby a
derelict madman.1 I therefore suggestthat the uirgowas a priestessof Apollo/
the god of purification and healing who played the same role in the story of
Alcmeo as in that of Orestes.If so the sceneof action would bein front of a
temple and Psophiscould perhaps beruled out as the city to which Alcmeo
had come; any action there would have taken place outside Phegeus' palace.3
The trochaic tetrameters which Cicero quotes at De Orat. 3 .218 are
corrupt at a vital point but I thinkit can beargued4 that they were spoken by
an Alcmeo aware that the Furies belonged to his own imagination rather
than the external world and afraid of the outcome of a legal trial that he was
to face.
Many Greek tragedians are credited with an •AA~v: Sophocles,
Euripides, Timotheus (1.G.n 3091. s f.), Agathon, Astydamas,Theodectes,
1

Euaretus {I.G. n1 2320 . .26f.) and Nicomachus of AlexandreiaTroas (Souda

N 396).Euripideswrote two plays which the more carefulof ancient scholars
distinguishedwith the phrases 610:Kopfv8ovand 610:'t'oocptoos. Without
knowing about the 610:Kop(v6ovquotations BentleySpostulatedthe exis-
tence of two plays, one set in Psophis and dealing with Alcmeo's first visit,
his purification by Phcgeus and his marriage with Phcgeus' daughter.
Wdcker 6 and Wilamowitz7 reconstructed the •AA~ 610:'t'oocpl6os
according to Bentley's view, Wilamowitz pointing out that the distinguish-
ing phrasein thetitleseemedto excludeas the theme Alcmco's secondvisit to
Psophisand murder there at the handsof Phegeus' sons. Scholl8 and Hartung9
however reconstructed the play around thistheme and fairly recently there

1
In Attic drama, as in fifth and fourth century Attic society, neither un-
married girls nor matronsnormally went outsideunaccompaniedor conversed
with strangen; cf. Sophocles,Ant. S78 f., El. s16 ff., Euripides,Amir. 876 ff.,
El. 343 f., 1072 ff., Hele. 974 f., Heraleltidai474 ff., I.A. 73S, 1028 ff., Phoin.
88 ff., Aristophanes,Lys. 16, Menander, Dysle. 20s ff., 218 ff., Headlam and
Knox on Herodas 1.37 {Cambridge,1922).
a Cf. Euripides, Tr. 2s2-3 ~ Tav TOVCl>o(~ ,rcxp8wov,~ ytpas 6 I XPVC7~
1<6µasl&.>1<'~pov 36av.
3 A fourth-centurytragedy might have had a palaceand a temple. No more
than one stage building need be postulatedfor any extant fifth-centurypiece
except the' AvSpoµaxfl, which was not written for performancein Athens;sec
Pickard-Cambridge,Theatreof Dionysus,pp. 30, s2 ff.
• See below on fr. XIV.
s EpistolaadJo. Millium (Oxford, 1691), pp. 17 ff.
6
Die grlech.Trag. pp. S7S ff.
7 Ind. schol. Gottingen 1893, 12 ff. {= Kl. Sehr. 1 18s ff.). Cf. Euripides:
Heraleles'{Leipzig,189s), p. 123 n. 22.
8 Beitr. pp. 132 ff. 9 Euripidesrestitutus1, pp. 187 ff.

188
 ALCMEO

has come to light a scrap of papyrus which seems to many to settle the issue
in their favour. G. Vitelli saw that Pap. S.I. 13. 1302 and the quotation of
Euripides' "AA~ at Stobaeus 4.19.25 coincided and W. Schadewaldt1
interpreted the enlarged fragment as coming from the•~ 61a't'ooq,t-
6os: from a scene in which Phegcus, having learnt of Alcmco's return to
Psophis, tellsa slave and the chorus not to reveal the news to his daughter.
Schadcwaldt's arguments however arc far from conclusiveand he makes no
attempt to explainthe title phrase 61a 't'c:.)ft6os.:a
As Ennius' original threeGreek plays have beenproposed: the •AAIQlioov
61a 't'c:.)fl6os,3the •AAlq.lE(A)v
61a Kop{v6ov4and Thcodcctcs' •AAK~v.s
Sincewe know Ennius to have adapted many tragediesby Euripides there
is a good chance that Euripides provided the original of the Alcmeo.The
threefragments usually assigned to this play do not exclude the •~
6ux Kop{v8ovor an• AAK~ 61a't'ooq,t6osdealing with the first visit. It
would be difficulthowever to make them dealwith the second visit. I inter-
pret fr. XIV as spokenby an Alcmcoafraid of the outcome of a legal trial.
Some verses quoted by Stobacus, apparently as from Euripides' •~v
(AKµ'8. 12; fr. 67), refer very plainly to such an episodeand were argued by
T. Zidi6ski6 to have a metrical technique typical of the later Euripides. The
•~v 61a Kop{v6ovwas produced after Euripides' death while the
•MlqJiCA>V 61a't'ooq,t6osappeared in 43 8 B.C. Both Zielinski's argument and
mine arc shaky ones, mine much more than Zielinski's, but if they have
arrived at the truth they give grounds for thinking that Hartung was right
about Ennius' original. The identification of words? and common motifs8
as Euripidcan docs not hdp the argument. Such thingscould come from any
play about Alcmcoor any play about a matricide. One might as well argue

1
Hmnu LXXX (1952), 46 ff.{= Hellas und He.sptriffl,316 ff.).
a Sec the criticism of H. van Looy, Zts VerlorenTragediesvan Euripides
(Brussels, 1964), pp. 78-103 and 310-12 (summary in French).
3 So A. Matthiae, EuripidesIX (Leipzig, 1829), p. 20, Wdcker, Die grieeh.
Trag.pp. S1S ff., Wilamowitz, Ind. sehol.Gottingen1893, 14 ff. ( = Kl. Sehr.1
187 ff.), Vahlen,E.P.R.1 , p. ccu,Lco, HermesXJ.D (1907), 153 (= Ausg.kl.Schr.
II 409), Geseh.p. 190 n. 1.
4 So Hartung, EuripidesrestitutusII, p. 534; cf. Ladewig, Anal. seen.p. 29
(arguing that Ciccro's phrase uir bonusexcludes the two visits to Psophis).
5 So Ribbeck (tcntativdy), ~st. seen.268 f., Die rom. Trag.pp. 197 ff., G.
Perrotta, SIFC N.S. VI (1928), 127 ff.
6 Mnfflwsyne L (1922), 319 ff.
7 Welcker compared exalbescat of v. 20 with Hcsychius 1, p. 237 Latte, s.v.
&pycxfmv.
8
Wilamowitz compared the fit of madness with Tatian 24; Leo Alcmeo's
inopui (v. 16) with Photius Berol. 39.8.
 COMMENTARY
ttom the similarity which Columna observed between inade inade odsunt
yap
adsuntme me expenmt (histext of v. 23) and Euripides, Or. 257 CXV'Tat
<XV"Tat
irA11afov8p<t)01(0Va{ J.lOV that Ennius adapted the "Opm-t1ls.

XIV
Corruption at the head of v. I 8 stands in the way of an uncontrovenial
interpretation of this ttagment. I shallargue that the speakerallegoriscd the
Furies as what he suffered in consequence of mental and physical illness,exile
and lack of ttiends and the spirit of Eriphyle as his fear of the result of a trial
he was soon to face.
The alterof v. 18 hasalways given trouble. Columna stopped the quotation
of Ennius' Alaneo at expectorot, 1
apparently thinkingthat Cicero began the
quotation of another play at alter.Plank? and Bothe correctly extended the
quotation to metu but offered no explanation of alter. After nfty yean of
honourable struggle with the problem Vahlen despaired with the words
'Alter qui explicem non habeo, nisi forte interceptum est, errore aut consulto
de more Ciceronis, quod excipiebat Alter'.3 This does not, of coune, solve
anything.
Ribbeck: proposed in his first edition and stuck in his third to the emenda-
tion mater,imagining that Alcmeo,like his fellow matricide Orestes, was
punued by the spirit of his mother as well as by the Furies. He could have
adduced in suppon Aeschylus, Eum. 94-139 and Euripides, Or. 255-6. 4 The
emendation is much superior to Vahlen's similarlyinspiredultor.S Such
emendations assume that Ennius' tetrameters describea vision of demons, if
not actual demons. But this is what the anapaests quoted by Cicero at At.
2. 89 do (tt. xv) and it is not likdy that Ennius' play had two scenes of this
kind. Furthermore the verse form and the wording of the ttagment under
discussion suggest that Alcmeo is speaking at a moment of comparative
rationality 6 and trying to explain why 'neutiquam cor consentit cum
oculorum aspectu'.
In the phrase multis sum moJis drcumumtus there is the same military
metaphor as in v. 23 me expetunt, 24/er mi auxilium, 2.6indnctoeigni inceJunt,

1
Stephanus gave circumuentusmorboexilio otque inopia to the Alaneo but
seems to have lost Cicero's quotation at De orot.3 .218.
2
Q. Ennii Medta, p. 105.
3 E.P.R. 1 , p. 123.
4 Cf. also Virgil. Aen. 4.471-3, Ovid, Met. 9.409-10 ottonitusque
malis,exul
mentisf/uedomusqueI uultibusEummidummatrisf/ueagitabiturumbris(of Alcmco
himself).
S Ind. lectt.Berlin 1888-9, 3 (= Op. ac. 1 -401).
6
Cf. vv. 34-42 and 43-9 (Cassandra's speech of apology in trochaic tetra-
meters followed by her prophecy in dactyls and other lyric verses).
190
 ALCMEO

27 drcumstant.But instead of the three Furies1 three less demonic entities

surround Alcmeo: morbus,exiliumand inopia.Morbusis his sick state of body
and mind,i exiliumhis homdessness,3 inopiahis lack of funds4and powerful
friends.S The three allegorical Furies are accompanied not by the spirit of
Eriphyle but by pauor,a state of neurotic fear. The mode of attack employed
by pauoris described in vv. 17-18. Some scholars have tried to-make pauorthe
subject of the verb minaturas well as of expectorat,emending alter accord-
ingly. 6 But the shrewdest suggestion has come from Bergk who proposed7
that alterterribilembe altered to atrabilismihi.This emendation is falsebut it
did not deserve thescorn that Ribbeck poured upon it. 8 Greek philosophers
devoted much speculation to thephenomena of clairvoyance and prophecy9
and many explained them as mental aberrations arisingout of an excess of
black bile. It is unlikdy that Ennius, prone as he was to indulge in tragic
philosophising, 10 and ready to represent the emotion of fear in physical terms
in v. 20, would have introduced black bile stark naked without a word of
explanation. But Bergk' s feeling that prophecy is involved in Ennius' dis-
course was just. Ennius' discourse makes senseif for alteris substituted some-
thing like animusor mensenim.
1
The late fifth century Attic tragedians make the Furies three in number; cf.
Euripides, Or. 408, 434, 1650, Tr. 457, E. Wiist, RE Suppl vm (1956), s.v.
Erinys, 122.
a Cf. Euripides, Or. 34-s hrrni6ev &yp(c;tO"VVrCXICElS VOO<t> VOO'elI TAfiµwv
'O~s 6&, 227 ff., 395,407, 480, 792, Soo, 881 ff., 1016, Timocles ap. Athen.
6.223s 6 voaoov 6~ µav1KG'.)s'AAK~wv' ~o.
3 The normal lot of the parricide in the heroic world of tragedy (cf.
Aeschylus, Choe. 1034 ff., Euripides, Hera/des1322, Hile. 148, I. T. 80, 929, Or.
898-9oo) and historical Athens (cf. Demosthenes 23 . 42 ). The sentenceof death
passed on Orestes in Euripides' •o~s (902-56)is represented as a barbarous
one.
4 For the exile's poverty cf. Euripides, Herakles1325 ff., Med. 551 ff., Seneca,

Thy. 303 £, 923 ff.

S C£ Plautus, Rud. 617 ferte opeminopuieand see below on vv. 89 (ops)and
173 (opulentus).The parricide of the heroic world of tragedy was spurned as
one dangerously polluted (c£ Euripides, I. T. 947 ff.; contrast Homer, II.
2.661 ff., Od. IS .280-1); drought struck the cities where Alcmeo was allowed
to live (Apollodorus 3. 7. 5. 3).
6
Adeo was ,uggested by Vahlen (Ind. lectt.Berlin 1887-8, 6 n. [ = Op. ae.1
382 n.]), tetrum by L. Mueller (Q. Enni CarminumReliquuie [St Petersburg,
1884], p. 113). .
7 Philologusxxxm (1874), 283 ( = Kl. phil. Sehr. 1 362).
8
Ad TragkosAddendain ComkorumRomanorumFragmental(Leipzig, 1873),
pp. CXIII-CXIV, RhM XXIX (1874), 218.
9 C£ Plato, Phaidr. 244D, Tim. 71s, Cicero, Diu. 1.63, 81, Lucretius
10
I. 132 ff., 4.757 ff. See above on fr. IV, below on fr. xax.
191
 COMMENTARY
The prophetic powers of the mind are frequently mentioned in Attic1
and Roman drama.i It is evil that normally appcan to themind when it is in
a prophetic state.
For minari•prophesy evil at the handsof a third party' cf. Cicero, Cann.
fr. 10. 28-9 oracla. .. tristisminitantiawus, Manilius 1. 892-4 t4liasignificant
lucmtesuq,e cometdt:Ifunna cumf atibusumiunt, terrisque minanturI ardentes
sinefine rogos,Seneca,Nat. 2. 39. 2. ubirebusquietisneeagtntibusneecogitantibus
quicquamfalmen quideminterumit et aut minaturaut promittit aut monet,Otd.
20-1 thalamosparmtis Phoebuset dirostorosI gnato minaturinpia incestos face,
Octauia236-7 gmtibus cladesnouasI minanturastra,Statius, Theb.2. 348 ext4
mirumtiadiuos,Ps. Q_uintil. Dtcl.4 .12 minatusestmihimanusmeasmeusanimus.
T.LL. IV 1219. 6-, seemsto take uit4t as a genitive depending on cruciatum
and compares Porphyrio, Hor. S4t. 1 .1. 78 pukher tt g,auis sensusrecusantis
diuitias,quat cruciatumuit4t dominoadferant.But Alcmeois alreadysuffering
mental torture; it is not merdy threatened or prophesied. I should take uit4t
as uit4t (meae)(for the omission cf. v. 295 tdnta uis saltris in corpore[meo]
haatt, vv. 3,.0-1 ut sciasquantoe loco,I quantisopibus,quibusde rebuslapSti
fortuna [mea) aaulat), i.e. as (mihi) uiutnti. For this personal use of the
abstract uit4 cf. Plautus, Asin. 16-17 sicut tuom uis unicumgnatum tuae I
superwe uit4t sospitemet superstitem,Trin. S1 deosqutorout uit4t tuaesuperstes
supptt4t,Cicero, Mil. 86 nequtulloin locopotiusmortem lacerariqumnin quouit4
essetdamnata, Sest.83 eiusigituruitamquisquamspoliandamomamentisessediat,
cuius mortemonumdam monumentosempiternoput4retis?,Propertius 2. 1 • 73-4
Maeanas... et uit4t et mortigloria iust4 meae,4.7.69 sic mortislacrimisuit4t
StltUltnUS amores,Valerius Maximus 3 . 4. s cuius uita triumphauit,mors Papia
legedamnata est, 9. 2 ext. 1 o Etnuci. •. amari uitdt pariter« mortistortores.
The object of fear is auciatus et nex. The real demons of Aeschylus'
Eummidesand theimaginaryones of Euripides' plays3 threaten Orestes with
thesevery things.But if my interpretation of thefragment is correct Ennius'
1
E.g. Aeschylus, Pers. 10-11 tcaK6J.IQV'rtS
••• 8uµos, Euripides, Amlr. 1072
,rp6µCXV'TIS8vµos ~ Tt ,rpoa6oK~ Trag. inc. 1761TT)6ci.>v 6' 6 8vµas lv6o&v
µ(XVTE\Jn'cn,Menandcr, Misoumenosc,l,16-17 Turner µ(XV"TEVE8' 'fl'fNXT\
Tt µou
fha KCXK6v· I 6i6ou<a.
1
E.g. Plautus, Aul. 178praesagibat
mi animus.frustra
me irequomexibamdomo,
&ah. 679 aimus uunistoc dictoplus praesagiturmali, Terence, Haut. 236
ntscioquidprofectomi aimus praesagit"'4li, Pacuvius, Trag. 78 propemodum
animusconiectura de more eius augurat,Seneca, Thy. 957-8 mittit luctussigna
faturi I mms, antt suipraesagamali;cf. in dactylic verse, Calvus, fr. 10 mms mea
dira sibi praedicmsomnia uaecors,Virgil, Aen. 10. 843 ddgnouitlongtgemitum
praesagamalimens.
3 Aeschylus, Eum. 267,305 f., 422 f., Euripides, Or. 26o f. (cf. 438 ff.), I. T.
285 f.

192
 ALCMEO

Alaneo must fear them from somewhere else,most likely thecivil authori-
ties. Euripides allegorised Orestes' imaginary assailant$ as thenaggings of a
guilty conscience. 1 The fourth~tury orator Acschines allegorised the
Furies of the contemporary stage as the insatiable lusts of a bad man's heart
driving him on to furthercrime. 2 An allegory very similarto theone I am
attributing to Ennius appears in Lucretius' venification ofEpicurean teachings
at 3 .978-1023: atqueea nimirumquaecumque Acherunte I
profundo proditasunt
essein uitasuntomnianobis. .. Cerberus et Furiaeiamueroet lucisegestas.• . I qui
nequesunt usquamneepossuntesseprofeao:Isedmetusin uitapoenarumpro male
f actisI est insignibusinsignis,scelerisque
luella,I careeret lwrribilisde saxo iactus
deorsum,I uerberacamifices roburpix Lunminataedae;I quaetamen etsi absuntat
menssibiconscia f aais Ipraemetuens adhibetstimulostorretque jlagellis,I neeuidet
intereaquiterminusessemalorumIpossitneequaesitpoenarumdeniquefinis I atque
eademmetuitmagishaecnt in morte grauescant. IhieAcherusia fit stultorumdenique
uita.The demons with which the poets populate the underworld are here
reinterpreted as the punishments which await the guilty man in this world.
A difficulty remains: the Alcmeo of an Attic play might have had to fear
death from the civil authorities ifhe faceda trialin Argos but any trial in a
foreign city, if we can go by the analogy of that of Orestes in Athens as
depicted by Aeschylus, could have resulted, at the wont for him, only in
expulsion. I therefore suggest that, whereas thetext of theplay Ennius was
adapting made Alcmeo express fear about theoutcome of a trial to decide
whether he should be admitted to the community he had approached,
Ennius gave thehero the fears that a contemporary Roman parricide might
have. There are signsthat he rewrote the trial scene of Aeschylus' Eumenides
with Roman conditions likewise in mind.3
L. Hostius was executed just after theHannibalic war for murderinghis
father,4 in a manner that greatly impressed the theatre audience for whom
Ennius was writing. In adaptations of Attic comedy Plautus alludes at least
twice quite unmistakeably to the culleusin which Roman parricides were
put.S The phrase cruciatum et necemperhaps came straight from the language
6
of the lawyers and in any case could not hdp drawing theattention of the

1
Or. 395 ff.: cf. Cicero, S. Rose.67,Pis. 46, Har. resp.39, Leg. 1 .40, Parad.
18, Juvenal 13. 193-5, Q!!intilian, Deel. 31.~. p. 236.9.
:a Tim. 190. 3 See below on fr. LXIV. 4 Plutarch, Rom. 22.5.

S Epid. 349 ff., Pseud.212 ff. For the origin of this practice, apparently wi-
known to Greek communities, see Latte, T APhA LXVlI {1936), 24 ff., RE
Suppl. vn {1940), s.v. Todesstrafe, 1614, ZSavSt LXVlI{1950), 51 ff.
6
Cicero uses the nominal phrase crudatuset morsregularly in referring to
judicial executions (Verr. 2. 5. 72, 134, 138, 153 et al.; cf. Ulpian, Dig. 4.6. 3 et
al.). The word nex was probably already uncommon in ordinary parlance in
Ennius' own day; it occurs only 8 times in republican drama {~ mors69).
13 193 JTO
 COMMENTARY
audience to contemporary Roman judicial practice. The Romans regularly
tortured aiminals before executing them. 1 The Athenians on the other
hand were notoriously gentle. Torture docs not appear in the executions
their tragic poets describediand the only case recorded from historical
Athens is of a slave condemned for poisoning.3 A proposal to torture the
condemned Phocion was greetedwith horror. 4

16 maim sum modil: thearchaic dramatists normally joined multisand

modisso closely as to measure the phrase - u u - or even to write it as
multimodis.But there is no need to entertain Bothe' s multismodissum c. For
thedisjunction cf. Plautus, Bacch.
507 a: egoistancmultisulciscar
modis.Five of
thesix verbs in thepassage are hoisted towards thehead of their respective
clauses. This type of word order is a marked feature of the archaic tragic
style. And in all archaic writing, and some classical, whenever a part of esse
for some reason precedes its complement, it tends to go to the second position
of the clause; see M. Seyffen,Cicero:Latlius,ed. 2 (Leipzig, 1876), pp. 441-
2, J. Wackernagel, IF 1 (1892), 428-9 ( .. Kl. Sehr.196-8).

circumuentus: cf. Horace, Ars 169 multasentmcircumueniunt

incommoda.
morbo uilio atque inopia: for the syntactical arrangement of thetriad
(a,b atque c) cf.Plautus, Bocch.1113, 1181, Men. 174, Terence, Ad. 846, Haut.
777-8, 893, Ennius, Ann. 108.

17 tum pauor sapientiam nmoem •voimato expectorat: I follow the

reading of the 'codices mutili' at De oral. 3. 218; for the omission of the
I
anaphoric pronoun cf. Plautus, Cas. 621-2 occidi, cor metu mortuumst,
membra(mihi)miseraetremunt,Terence, Eun. 665-6 audieram . .. uerum(mi.bi)
miseraenon in mentemuenerat.Tragedy was prone to omit such pronouns
when the context was unambiguous; cf. Ennius 19-20, 38, 304, Pacuvius
257, 259, 263, 26-4, 266, 328, 350, 351, Accius 13 et al. All the penonal
pronouns and pronominal adjectives are, perhapsbecause of theirundignified
lack of size, much less common in tragedy than in comedy.S Hiatus at the
1
Cf. Cicero, Vm. 2. 5. 14, 118. I.ivy's statement, in aliisgloriarilied, nulli
gentium mitiortsplacuisstpoenas(1 .28. 11), is pious nonsense.
1
Cf. the discussion of Orestes' punishment at Euripides, Or. 857 ff.
3 Antiphon 1 .20. See Latte, RE Suppl. vn (1940), 1009-10.
4 Plutarch, Phoc. 35.

s Ennius has egoetc. about once in 49 words, Pacuvius 1 in 53, Accius 1 in 57,
Terence 1 in 27. Tu etc. occurs about once in 75 words ofEnnius, 1 in 62 ofPacu-
vius, 1 in 77 of Accius, 1 in 37 of Terence. ls etc. occurs about once in 188 words
of Ennius, 1 in 145 of Pacuvius, 1 in 182 of Accius, 1 in 6o of Terence. Signi-
ficantly isetc.isextremelyrareintheelevatedlanguageofEnnius' epic,extreme-
~
ly common in the remains of his prose treatise EuhtmnUS(1 in 700 1 in 25).
194
 ALCMEO
diaeresis of the trochaic tetrameter is not elsewhere transmitted in the re-
mains of Ennius' tragedies1 but is so frequently in the comedies of Plautus.
Terence seems to have avoided it2 but Accius allowed it a few times3and it
.would be hypcraiticalto deny it to Ennius.
Fear is one of theincorporalia most commonly personified in republican
drama: c£ Naevius, T,ag.40 nosduplicat aduenientistimos,Pacuvius, T,ag. 292
ecfarequ« co, tuum timidiw tarite~ Accius. Trag.122 uoshienonmertetmetus,
Plautus, Amph. 1079, Cas. 6s3, 704, Cist. 688a, Mil. 1233, RMJ.686, 703.
Incorporalia and abstracts govern transitive verbs much more often in Latin
tragedy thanin Attic. In the latter the construction scarcely occurs outside
gnomic utterances and passageswhere the linebetween personification and
demonology is hard to draw (c£ Aeschylus, Ag. 1306, Choe.288, Eum. 88,
~
Hilt.736, Pers.703, Prom. 181 Homer, n.11.37, 13.299,Aeschylus, '1kb.
4s). In Latin tragedy I note it at Livius 13, Naevius 40, Ennius 17, 2s, 68, 102.
1s6, 199, 222, 271, 275 (bis),30s, 333, Pacuvius 6, 58, 6o, 67, 109, 119, 137,
1so,161,170,179,211,222,224,241,2s7,276(bis),277,292,299,303,306
(bis),309, 368, 374, 388, 396, 412; Accius 13, 16, 88, 98, 99, 102, 118, 122,
138, lj2, ISS, 176,188,272,286,303 (bis),, 316,344. 3S2, 387,419,453,456,
479, 492, so3, s16, SSS, s64, 566, s68, S8o, 587, 621; Trag. inc. 58, 93, 148,
247. Only two of the Enoian examples occur in spoken trimetcrs but
Pacuvius and Accius appear to have constructed their trimetcrs in this way
more often thanthey did their musically accompanied verses. Occasionally,
as in theverse under discussion, the verb governs an abstract as well: e.g.
Pacuvius 60,276,292, 306; Accius 99, 176, 188, 3S2, 4S6, 621; Trag. inc. 58,
148 (c£ Aeschylus, Hilt. 498; Sophocles, Ant. 389, 1028; Euripides, Phoin.
944: about theonly clear examples in thesethree plays). The construction is
comparativdy rare in comedy and where it does occur it tends to be accom-
panied by other devicesdesigned to elevate thestyle; sung versesshow it very
much more often thando spoken; see Hafner, Untersuchungen, pp. 86 ff.
For tum 'praeterea' c£ Ennius, Ann. 77 curantesmagnacumcuratumcupientes
regni,Plautw, Men. 258 ff., Terence, Phorm.327 £
Pauorwas a word of some elevation, occurring 5 times in tragedy and not
at all in comedy. It is absent from Caesar and from Cicero, except for three
passagesin theTusculans (4. 16, 4. 19, s.52) where he is making fine psycho-
logical distinctions. Abstract formations in -or tended generally to have a
lofty tone; Ploen counted26in 1,940verses of tragedy and only 3S in 30,000
of comedy.

1
But cf. vv. 148, 149.
a For a defence of the few casestransmitted sec A. Klotz, Hmnes LX (1925),
325 f[
3 Trag.13, 149, 231, 301, 451, 476.

195 13-2
 COMMENTARY
Sapienliaoccun twice elsewhere in tragedy. It is more common (21
instances)in comedy than other abstracts in -entiaand -antiabut rare com-
pared with its synonym consilium.
For sapientiam . .. expeaorat cf. Plautus, True. 77-8 PhrcmesiumI suom
nomenomneex peaoreexmouitmeo,Batch.653, Pseud.144, True.6o3, Catullus
76. 22, Lucretius 3 . 908, 4. 908. The verb expectorareoccurs dsewhcrc in
archaicand classicalliterature only in Accius (Trag. 301, 595). The type of
formation was probably as bizarre to archaic ears as to classical.1 The only
parallds I can find are, in comedy, egurgitare(Plautus, Epid. 582) and, in
tragedy, eliminare(Ennius238, Pacuvius 134, Accius448, 592).
For the redundancy of pauor. .. exanimatocontinued by terribilem . •.
timido. .. metu c£ Plautus, Amph. 1079-81 etukm nosformido timidastmore
inpulit. .. ita mihi animusetiam nunc abest,Cist. 688-8 a intus paueo et Joris
f ormido,ita nuncutrubiquemetusme agitat.
18 terribilem: elsewhere in republican drama only at Trag. inc. 96.
Formationswith-bilis (-biliter)are numerousin tragedy, comparativdy rare
in comedy.

minatur: scarcdy preferable to minitatur,the reading of the 'codices

integri'; both forms occur frequently in drama and in Cicero's writings, the
latter having the simple predominating (39:22 in the orations, 10:4 in the
philosophicaldialogues,8: 5 in the letters, neither in the rhetorical writings
at all), the former the intensive (23:8 in comedy, 1:1 (?] in tragedy).

uitae cruciatumet necem: see above, p. 193.

IS)-20 quae nemo est . .. quin refugiat timido sanguen atque ex-
albescatmeta: Vahlen's supplement (cum aduentareuideat)removes the
anacoluthonat the priceof even greater syntacticalcomplication.It would be
better to understandrefugiattimidosanguenatqueexalbescat as a pictorial sub-
stitute for a verb like timeatand quae(i.e. crudatumet necem)as its object; c£
Aeschylus, Theb.28!}-90 µtpiµvcn 3CA>'TTVp0V0'1 T<Xpj:x>sI TOV aµcprreixi\
AEoov,Hik. 566-7 6dµcrn 8vµov I ir@.AoVT' ~iv &1'l8ii, Euripides, Or.
86o, Ba. 1288.
For the physicaleffect of fear c£ Aeschylus,Ag. 1121-2 hrl m Kap6{av
(6pcxµeKpoKof3acp11s I O'Tayoov,Aristotle,fr. 243 Rose, Lucretius 3. 152 f[

19 tam mmo ingenio et tanta conJidentia: the second phrase repeats

the substanceof the first at slightly greater length (5: 8 syllables)and with
choicer vocabulary. Plautus has (Rud. 645) quis homosttantaconfidentiabut
1
Cf. Q..uintilian,Inst. 8 . 3 . 3 1.
196
 ALCMEO

comedy on thewhole has the phrase-type 'est+ablative of abstract noun'

much less often than the type 'est ingenio+adjective•. The anaphora tam • ••
tantaelevates the tone further.
The long verses of Ennian tragedy and Plautine comedy share a marked
tendency to make thesecond phrase of a pair longer than thefirst. Where the
substance of the second repeats that of the first in comedy, the speaker is
usually a personage of low degree taking off the ways of speaking usualin
tragedy. Such tautologies are rare in Attic tragedy and restricted to highly
emotional passages; 1 they are common in the Latin adaptations: Livius 21,
Ennius 19, 40, 61, 83, 163, 174, 193-4, 204, 205--0, 217, 252, 278, 310,
Pacuvius so, 155, 164, 256, Accius 15, 6o, 87, 154, 207, 364, 365, 510, 583,
6o8. Only two of the Ennian examples are in trimeten, one of thePacuvian
and three of the Accian. This may, however, be partly an accidental effect of
the length of the vene. i

20 refugiat: once elsewhere in republican tragedy (Trag. inc. 189), and

once in comedy (Caecilius 236) with little perceptible difference in meaning
from the extremely common simple form of the verb. Ennius' respectare (71),
Pacuvius' retinere(263) and Accius' remanere(447), reticere(95) and reuisere
(336) similarly replace the simple forms usedby the common language.

sanguen: not elsewhere in republican drama; for sanguisc£ Ennius, Trag.

324,Trag. inc. 209, Plautus,Mere.550;for sanguinemPlautus,Bacch. 372 et al
Sanguenappean in epic at Ennius, Ann. 113, in oratory at Cato, fr. 211.

enlbescat: only here, Cicero, De orat. 1.121, Ac. 2.48 and Gellius
in v. 17 and amplifies
12. 1. 12. The prefix picks up exanimatoand expectorat
the tone rather than alten or extends the senseof what is being said.Other
compounds with ex- likewise replacing the simple form in tragedy are
edocere(Ennius 56, Pacuvius 374), eloqui(Ennius 330, Accius 301), enitere
(Accius 235 1 ), exaudire(Accius 281), exposcere(Ennius 151), exsacri.ficare
(Ennius 54), exsuperare(Trag. inc. ap. Cic. Tusc. 4. 77, Pacuvius 404),
exsusdtare(Accius 199 ). Plautus' exaugere,exdissertare,
exobsecrare,exputare
erogitare,extumereappear to be of a paratragic character. The prefix AK-
performed a similar function in Attic tragedy.3

1
Cf. Euripides, Hel. 483 Ti ,oo;T( M~e..:>;,Alk. 21 8avelv•.. Kai 1,1ETaaTfjvat
~(ov, 108 l&tyes q,vxas,!&l'yeSSt ,pevoov. ·
a On the whole subject see Lindholm, StilistischeStudien,pp. 94 ff., Hafficr,
Untersuchungen, pp. 53 ff.
3 See Wilamowitz, Heralele.s 1 , on v. 155, A.Meillet, AperfUd'unehistoirede la

languegrecque'(Paris, 1930), p. 209, Fracnkel, Aeschylus:Agamemnon(Oxford,

1950), on v. 1033.
197
 COMMENTARY
The inchoativeform of the verb probably had a poetic tone. Suchforms
are comparativdy rare in the fragmentsof Ennius' tragediesbut pullulatein
those of his Annalesand the rest of republicantragedy.This situationmay be
partly due to the nature of the sourcesof Nonius' lexicon,which included
only two actualscriptsof Ennius' tragedies.Ploen1 counted 85 inchoative
formationsin comedybut it would be wrong to thinkthat many camefrom
ordinary Roman speech.64 of them occur only in Plautus' playsand many
of these nowhere else in Latin. Vcry often a paratragic tone is plainly
detectablein the context of occurrence.

xv
For the context seeabove, p. 187. I shouldadd only that Apollo and Diana
were of a similarnature to the three Furies:figmentsof Alcmeo'simagina-
tion rather thanreal personagesappearing on stage. They perhaps sym-
bolisedAlcmeo•s hope of release&om his sufferings.
The words of v. 21, sedmihi neutiquamcorconsentitcumoculorum aspectu,
suggestthat Alcmeo is speakingat a moment of relative sanity.:iIt would
thereforebe preferableto scanthem as part of a trochaictetrameter(lacking
the final two dementsv - ) thanas anapaests.However a convincingsupple-
ment ishard to find.Ribbeck'saspectu (truci)isquite unsatisfactory;Pacuvius,
Trag. 2-3 quadrupes . .. aspectutrud bdongs to a quite different kind of
context.
Cicero quotes neitherv. 22 nor 11. 23 in its entirety and scribeshave cor-
rupted what he docsquote of the latter.3Ribbeckprinted v. 23 as a very un-
convincingcatalectictrochaictetrameter-inadunt inaduntadsuntadsuntme
med expetunt-and vv. 24-7 as two acatalectictetrameters,thus making the
dactyl-mapedend of ardentibus coincide with the end of a foot.4 The old
editors took vv. 24-30 as a seriesof anapaesticdimeterswith hiatusbetween
v. 28 and v. 29. In his Epistleto Mill (p. 26) Bentley pointed out that the
anapaesticsystemsof classicalGreekand Roman poetry avoidedthis kind of
hiatus.Biichderargued that v. 28 was defectiveat the beginning;RibbeckS
1
De Cop. Verb.Diff. p. 79.
3
Cf. Euripides, Hel. S1S ou ,rov fPO~ ~ ro,To 6' 6µµa µov VOO'El;
=
3 Vahlcn {cf. Ind. lectt.Berlin 1887-8, 7 [ Op. at. 1 383)) takes incedeincede
as addressed by Aletnco to himself. Of his parallels only Euripides, El. 112-13
could be penuasive and there the interpretation of the Greek is disputed.
In any case it is hard to see what point such a self-address on Alcmco's part
would have.
4 Cf. above on fr. XII. A. Spengcl. Reformvorschl. p. 192, drew attention to
this and the lack of pure theses in Ribbeck's alleged trochaics .
.5 Coroll.pp. XVII f.
 ALCMEO
supplied(tccum)intendit,Vahlcn1 hac,Plasbcrg2dextra.None of thesesugges-
tions is convincing and the words transmitted make reasonablesenseas they
stand. The only paralld hiatus in tragic anapaests,that betweenEnnius,Trag.
91 and 92, coincideswith a strong rhetorical pause but the small number of
archaic anapaestssurviving and the large number of anomalies transmitted
in other kinds of archaic verse make regularisation a dubious procedure.

21 cam oculorum upectu: the phrasesoculisaspicere et sim. arc common

in comedy (Plautus, Cas.939-40, Men. 1001, Mil. 1217 et al.) but the abstract
noun aspectus occurs only twice (Plautus,Epid. 572; Turpilius 75).Tragedy,
on the other hand, has aspectusfive times. The formation of abstract sub-
stantives in -tus (-sus)was much affected by the more dcvated genresof
archaic poetry; Plocn counted 63 such formations in I ,940 versesof tragedy
and only 125 in 30,000 of comedy.
Cum oculorumaspectustands for little more than cum oculis;c£ Euripides,
I.A. 233-4 Tav ywoot<Etov ~1v 6µµ6:rc,.w I ~ 1TAl)aooµ1,I. T. 1167 ~1v
6' 6µµ6:rCA>v~pµcxmi. This type of periphrasis, in which the adjectiveor
defining genitive carriesthe main idea,is common in both Attic and Roman
tragedy: where nouns in-tus (-sus)arc concerned I note Ennius 123 iuuenum
coetus,245.fructusuerborum,305 pedumpulsu,309 motussuperumatqueinferum,
394 imbriumfremitu, Pacuvius 68 triplicem . .. partum,303 beluarumacf erarum
aJuentus,328 paternumaspectum,AcciusSIS ex tuo... .uuu, 6o8 bellijluctus,618
mediocrisatu.

22undehaecflarnrnaoritar: for fire as a signof the onset of madness cf.

Aeschylus,Ag. 1256 {Cassandra) ,rarrai olov TO irOp • trripxEToo 6i µ01,
Sophocles,El. 887-8 elsTf µ01 I ~~aaa 8a;\m1 T<i>6'av11tda-r(t')1TVpf;,
Virgil, Aen. 7. 354-6 acdumprima luesudosublapsauenenoI pertemptatsensus
atqueossibusimplicatignemI nealumanimustotopercepitpectortjlammam.
Flammaappears to have beena word of dcvatcd tone. It occurs 10 times in
tragedy, only 4 in comedy(~ ignis 1:26),.

23 fincedeincedet adsunt: Alcmeo docs not name the fiends whom he

imagines to be attacking him; probably ominiscausa(cf. Sophocles, O.K.
128, Euripides, Or. 409).
Alesseis commonly used in prayers invoking the aid of a deity (c£ the
auguralprayer quoted by Serviusat Aen. 8. 72 adestoTiberinecumtuis undis,
Catullus 62. 25 et al.). For use of normally auspiciousvocabulary from the
sacrallanguage in inauspiciouscontexts see below on v. 144,
1
Ind. lectt.Berlin 1887-8, 4 n. ( = Op. ac.1 380 n.).
a Using an idea put forward by Zillingcr, Ciceround die altromisthtnDichter,
p. 109 n. 3.
199
 COMMENTARY
me expetant: the compound verb picks up exilio... exanimatoexpectorat
... exalbtscatin fragmentXIV and is itselfpicked up by excrudat.For txpttere
'petcre, attack' I can instance only Plautus, Ba«h. 51 duaeunum expetitis
palumbem.

24 fer mi auxilium, pestam abige a me: chiasmw is a common adorn-

ment of the more emotional speechesof both Attic tragedy (e.g. Euripides,
Alk. 215-17, Ba. 74-S, 902-3) and the Latin adaptations (e.g. Enniw 143,
159, 165, 310, Pacuviw 77, 92, 113-14, 155, 253.......,
276, 335, 336, Acciw, IS,
156, 303, 365).

I
24-5 pestem abige a me, flarnrnif.-am baneaim quae me excra-
ciat: vision and reality here begin to fuse,so that it is not clear whether
Alcmeo is referring to the Bames of the Furies' torches or to the fever
(irvpn-6s) from which he is suJfering.Pestismay be a tragic hyperbole for
febrisor morbus;it is nonnally used of a plague which attacks a whole pop-
ulation (c£ Enniw, Ann. 559). The phrasesmorbumabigereandfebremabigere
are common in Latin medical writing (c£ T.L.L. I 97. 41 ff.).
Flammiferam baneuimis a periphrasisfor luutantasjlammas;c£ Trag. Grace.
inc. 90 irvpos ... µwos,Lucreti:w 2.215 uisjlammea, Anon. Aetna 567
inandiuis.
The compound adjectivejlammifer does not occur again before Ovid.
Such adjectives are rare in comedy {Plautw has dulcifer,furcifer,trifarcifer,
lucrifera,jlabellifera),
comparativdy common in tragedy (f,ondiferat Naeviw
22,frugiferat Trag. inc. 164, lwrriferat Pacuviw 82 and Acciw 566, luciferat
Acciw 331, mortiferat Trag. inc. 87 ). Classicaldactylic poetry employs them
sparingly, Seneca'smythological tragedy freely.1

26 caeraleae incinctae igni incedunt: Columna rewrote this as caeruleo

incinctatanguiincedunt,thinking, no doubt, that there was enough talkof fire
elsewherein the canticum and explicitly quoting in support of hisalterations
Virgil, Georg.4.482-3 coeruleosque implexaecrinibusanguisI Eumenidesand
Ovid, Met. 4.482-3 cruorerubentemI induiturpallam,tortoqueincingiturangue.
To these might be added Hesiod, Asp. 233-4 hrl Se lWV1J01 6p6:KovreI
6o100&-mJoopewr'hrtKVp-rwov-n K6:p11va(the Gorgons), Acschylw,
Choe. 1049-50 cpcnoxfTooves Kai 1ml'AEKTaYflµwatj 1TVKVOIS 6p6:Kov01v,
Euripides, Ba. 697-8KCXTcnniK"TOVS 6opcxsI 6cpecn l<CrTEl(.OO'aV"rO (Mac-
nads), Catullw 64.258 stse tortisserpentibusincingebant.
Against himselfin favour of caeruleae(sc.Furiae)Columna quoted Orph.
H. 70.6-7 KVav6xpOOTOl6vaO'O'at, arracrrpa,n-ovaa1 arr' 6aaoov I

1
Cf. K. Miinscher,Bursians
Jahresb.cxcn (1922), 200 ff.

200
 ALCMEO

&nn'lv avTavyii cpmos aapKOCp86pov alyAf\V. To this should be added

Hesiod,Asp. 249 KfipesKVaveoo, Virgil,Am. 7. 346 huicdea"1ffllkisunumde
crinibusanguem,Statius, Theb.1.110 et caerukiredeuntin pectoranodi.
lncinctaeigni can stand, interpreted either as 'wearing belts of fue' (no
more bizarre than Euripides, I. T. 288 fi 6' ii< XlTOOVCA>V nvp irviovaa)
or as' armed with fiery weapons• (cf.Virgil, Am. 9. 74facibuspubesaccingitur
atris, 12. 811 jlammis cincta [Iuno], Seneca, Tro. s6o-1 ). Ennius' love of
redundant expressionneedsno further illustration.
The form caeruleus occurs once elsewherein republican drama,likewisein
anapaests(Plautus, Trin. 834).
lncinctaeis piaed up by incedunt. •• interulit. •. innixus. The compound
verb incingereis absent from comedy and classicalprose. Other in--com-
pounds replacing simple verbs of the common language in tragedy arc
inaudire(Pacuvius 35), inaurare(Ennius 213), inesse(Ennius 22s). inftnu,re
(Ennius 158 ), inlwrrescere
(Pacuvius411 ), inniti (Ennius 29 ), insultare(Ennius
124), inuestire(Ennius 113), inuisere(Accius 237).

27 circamstant cum udentiba, taedis: for thetorches of the Erinycs cf.

Aristophanes,Plout.423-S la(A)S "Epavvslanv lK TpayCf)S{as · I ~Arnetyi
TOI µav11<0vTl Kai Tpay't)6tK6v.1-&XA' OVK fxet yapsc;sas. Acscbines,
Tim. 19() Kcx6arrepa, Tats Tpay't)S{oosTlotvas °'<XWEIV Kai KoA03e1v
~aiv 1'µµwoos.There is no clear evidence in the text of Evµev{&s that
Aeschylus' Erinyes carried torches.
~
. Taedaoccurs twice in tragedy, not at all in comedy ( Jax 1 : s); it was
probably a word of the sacral language.

28 crinitaaApollo: <l>or~ &lc.epao1<6µT}s n.

(Homer, 20. 39); cf.Virgil,
Am. 9.638. Crinesusually denoted the hair of the Roman matronain its
characteristicarrangement; c£ Plautus, Mil. 792, Most.226 (in both passages
theAttic original is handled freely).

29 arcum aaratam: for Apollo's golden bow cf. Aeschylus,Bum. 181-2

I
'IMT}VOVapYfl<Tn\V 6q,tv, )(p\/Ol)Aarov e&lµtyyoS ~~µEVOV.
Ennius' use of the adjective auratus,properly 'gilded', rather than aureus
may be a piece of rationalism.

lam: 'the curve of the bow'; cf. the use of the denominative lunareat
Propertiw 4. 6. 2 s adem geminos. . .lunarat in arcus,Ovid, Am. 1. 1. 23
lunauitque
genu sinuosum fortiterarcum.

innixus: not elsewhere in republican drama; niti occurs 6 times in

comedy, 4 in tragedy.
201
 ""C~•----~----- ---.-
...#.-............... -·----------··-·

COMMENTARY

Diana &cem iacit: Apollo's sister normally carries a bow (Euripides.

Hipp. 167, 1422, 1451, Naevius ap. Macr. Sat. 6. 5. 8, Accius, Trag.52, 167)
but for the torch c£ Sophocles, 0. T. 204 ff., Tr. 205 ff., Euripides, I. T. 21.

XVI
See above, p. 185.

ALEXANDER

The title Alexanderis given to Ennius by Varro, Verrius, Gcllius and

Macrobius. Varro, Verrius and Gcllius quote what they do to illustrate
points of lexicography. Macrobius' quotations come from an account of
those verses and passagesof Virgil's poems which were thought to have been
based on work by the older Latin poets.
Alexander was another name of the Trojan hero Paris. One story, as old as
the Kwpta and Homer, n.24.28 ff.,told of how the three goddesses, Hera,
Athena and Aphrodite, came to him when he was mindingcattle on Mount
Ida and got him to settle a dispute as to who among them was the most
beautiful. Another story, as old as Pindar, Paian8 e. s ff., told of how he was
exposed at birth on the advice of seers, rescued and raised by herdsmen of
Mount Ida; at athletic games celebrated in hismemory by hisfather Priam
he competed with and defeated hisbrothers, Hector and Deiphobus; angry
at beingdefeated by an apparent slave, Deiphobus plotted to kill Alexander
but at the critical moment the factsof Alexander's birth came out and he was
received back into the royal house. 1
Tragedies by Euripides and Sophocles dealt with Alexander's recognition. i
The pieces attributed to Ennius' Alexandercan all be interpreted with some
plausibility as coming from a play with the same theme.

1
For the second story see Apollodorus 3. 12. 5. 2 tf. and Hyginw, Fab. 91.
Its relations with the fint story are obscure; cf. C. Robert, Bild u,ul Lied
(Berlin, 1881), pp. 233 tf., Hermes11. (191-4),315 f., Die griech.Heldensagem ii
(Berlin, 1923), pp. 977 tf., 1071 tf.
, Sophocles, fr. 90 N. 1 (Steph. Byz. p. 139.20) is quite unambiguous.
Direct quotations of Euripides' •A)J~avSposin Greek authors are uninformative
and early students were led badly astray by their belief that its title was
•AAe~avSpa. J.Barnes (Euripides:quaeextantomnia[Cambridge, 1694]) seems to
have been the first to make the recognition of Alexander Euripides' theme.
Osann, in F. A. Wolf's AnaledaLitterarian (Berlin, 1818), pp. 529 tf., got rid of
and argued that Hyginw, Fab.91 summarised the plot. The
the title• AAe~avSpc:x
matter was put beyond all doubt by the papyrus fragments published in 1922
(by W. Cronert, NGG, Phil.-hist. Kl., 1-17).

202
 ALEXANDER
C. 0. Mueller referred Varro's quotation at Ling. 6.83 (fr. XIX), iam
dudumab ludis animusatqueauresauent I auideexspectantesnuntium,to the
games at which Alexander defeated his brothen.
Welcker 1 interpreted Macrobius' quotation at Sat. 6. 1.61 (fr. XXIV),
multi alii aduentantpaupertasquorumobscu,atnomina,as coming from the
messenger'sspeechwhich recounted thegames and asdescribing Alexander's
companions. This interpretation is confirmed to some extent by thefu:t that
Virgil's imitation occun in hislist of those who came to compete in thefoot-
race at the funeral games for Anchises(Am. s .293 ff.).In Homer's foot-race,
which was Virgil's main object of imitation at this point, there competed
only threeprominent heroes, Ajax, Odysseus and Antilochus (n. 23. 740 ff.).
Virgil's imitations frequently reflectthecontext as well as thewording of the
older Latin poet's verses.,
Macrobius' quotations at Sat. 6.2. 18 and 25 (frs.xxv and xxvi) refer to
events which occurredin time afterthedeath of Alexander. They were for a
long time partly responsible for thenotion that Ennius wrote two plays, an
Alexanderand an Alexandra,thelatter dealing with theprophetess Cassandra
and set after the fall of Troy. Taking up a hint of Columna's, Vossius inter-
preted the quotations as from a speech by Cassandra describing a vision of
thefuture. This interpretation is supported by the factthat Virgil imitates the
fint in thespeech made by the ghost of Hector to the sleeping Aeneas on the
night of Troy's fall {Aen.2.281 ff.) and both in the conversation between
Aeneas and the shade of Deiphobus in the underworld (Aen.6. sooff.).
At Ling. 7. 82 (fr. xx) Varro quotes a verse of Enni~opttr Parim
pastoresnuncAlexanarumuocant-which can only come from a play about
Alexander's early life. Varro asserts that Ennius took his etymology straight
from Euripides and it would seem fair to suppose that he believedtheLatin
Alexanderto be an adaptation of Euripides'• Mi~av6pas.3This play was the
only one of theAlexandrian seventy"in which such an etymology could have
been found.
Commenting on Virgil, Aen. 7, 31Sr20 neeface tantum I Cisseispraegnas
ignisenixa iugalis,where Juno isprophesying the same troubles for Aeneasand
his bride Lavinia as struck Parisand Helen, Scrvius comments: ossms regina
Hecubafilia secutulumEuripidemCissei quem Ennius Pacuuiuset Vergilius
sequuntur;nam HomerusDymantisdidt. haecsef acemparereuiditet Parinaeauit
quicausafuitincendii(fr. ccxxv).5Columna and sucettding .-.ditonof Ennius
1
Die griech.Trag.p. 467.
a Cf. Norden, Bnniusund Vergilius, passim,AeneisVJ1,pp. 365 f.
3 F. Yater, Untersuchungen uber die dramatischePoesieder Griechen1 (Berlin,
1843), p. 23 (c£ Leo, Gesch.p. 189 n. 2), tried to deny that this was implied by
Varro's discourse. 4 See Introduction,p. 45.
s Cf. Am. 10. 705 ff. and Servius' note.
203
 COMMENTARY
thought that Serviusrefers to Ennius' version of Euripides' •EKaf:nl. B. Snell
1
argucd that Servius' comment shows him to have believed that Ennius
adapted the •AM~av6p<>s. It shows nothing of the sort. The bulk of what
Servius says probably comes from the same mythological source as lies
beneath Hyginus, Fab.91. 1 •.• Hecubae,Cissti siueDynumtisfiliae. uxor eius
praegnasin quiett uidit sef acemardenttmparereand Apollodorus 3 • 12. S •2
'EK~v T11V AvµavTOS fi ~ TtWScpaat Kia~ ... !~ •EK~ Ka9'
vnvovs OOAOV TEKElV 61cnrvpov, TOVTOV & ,raaav hnwµeo6cn T11V
1r6A1vKcxiKOOEIV. Th~ similarity between Hyginus and Ennius, Trag. so-1
mater grauida parerese ardenttmf aam Iuisaestin somnisHecubais an accidental
one.3 The information about Euripides' view ofHecuba's parentage might
refer to the •EK<Xfnl(v. 3) or to several of his other plays including the
•Me~av6p<>s.The information about Ennius' view would have beenadded
by Servius or some pr~·Uog scholiast and might just as readily refer to
Ennius' version of the 'EKal:n,as to hisAlexander.Nothing is implied about
the relationship between Greek and Latin plays.
The pieces attributed by ancient authors to the Latin Alexanderare not
such as could prove wrong Varro' s belief that it had a Euripidean original.
A number of pieces of tragic verse quoted by Cicero without either the
name of the author or the title of the play are commonly given by modern
scholarsto the Alexander.
At severalpoints of hisdialogue on divination Cicero illustratesthe pheno-
menon of ecstaticprophecy with the utterances of the Cassandraof a Latin
tragedy as shebeholdsa seriesof visionsof the events which culminatedin the
destructionofTroy. Diu. 2. 112 (fr. xvne) givestheproof,if proofberequired,
that 1 .67 (fr. xvnc) and 1. 114 (fr. xvnd) quote the same spcech.3There is,
however, one difficulty.The quotation at 1.67 strongly suggests that the
tragedy is set within the walls of Troy (ciuesfertt opem},while the context of
the quotation at 1 . 114 suggeststhat Cicero believedit to be set on Mount Ida
or perhaps by the seashore (multosnemorasiluaequemultosamnesautnuiriacom-
mouent).Cicero's dialogue shows many signsof hasty composition and it
could be that he decorated hisadaptation of someGreek account of the causes
of ecstaticprophecy with more from the same speechwhich he had quoted at
1 • 67 and did not stop to think whether it was completely appropriate.4

1
Euripides'Alexandros(HermesEin2Clschr.v [1937)), p. 59.
2
There is no common linguistic peculiarity such as the se. .. uisaest which
links Ovid, Epist. 17 .237-8 with Ennius.
3 T. C. W. Stinton, Euripidesand theJudgementof Paris (London, 1965),
pp. 68 f., toys with the idea that separate speeches arc quoted.
4 C£ Philippson, BPhW XLII (1922), 102. At Tusc. 3 .44 £ Cicero quotes a

speech by the Ennian Andromache (fr. xxvnh) although only a speech by a

male personage would fully suit the context.
204
 ALEXANDER
It has been argued• that while the verses quoted in 1 .66 and 1 .67 come
from thesame play the first group comes from near theend and thesecond
from near the begiooiog, Now it is true, as Cicero bimsrlf admits, that the
first group does not illustrate thephenomenon of ecstasy (opoem4tenerumet
moratumatquemolle.sedhoeminusad rem). Furthermore some scholarr have
been tempted to reverse Cicero's order, thinking that thesecond group were
accompanied in utterance by the unmaidenly behaviour remarked on and
apologised for in the first group.3 But the rationale of Cicero's mode of
quotation, his &ilure to make clearin his discoursethat the fust two verses
quoted were not uttered by Cassandra and theenormous extent of hisquota-
tion, suggest that he consulted a text of the tragedy rather than hismemory
and copied out the verses which struck his sensibilities, both those relevant
to hisargument and those not. A papyrus roll would have carried the text in
Cicero' s day and rolls were far more easily consulted in one place thanin two
and more likely to be read frontwards than backwards. As to thesecond point
it is an error to think that a seer's description of hisor her vision was neces-
sarily accompanied by signsof violent physical disturbance ;4at Euripides,
T,. 308-461 physicaldisturbance seems to have come fust, then rational
discourse,then prophecy.
Cicero's discussion of thecorrect form of the second declension genitive
plural at Orat. 155-6 takes examples from the work of three writers of
tragedy. The third is mentioned by name, Accius; the second, 'ille alter in
Chryse', must be Pacuvius; the fust can only be Ennius. These were in
Cicero's day thethree classic Latin tragedians.5Two of thefour examples of
Ennius' usage come from the speech of Cassandra quoted at Diu. 1 . 66-JJ.
Study ofCicero's mode of quotation throughout the whole discussion of the
aw&ots of individual words at Orat. 149-64 confirms our impression that
Diu. 1 . 66-JJ quotes from the one speech. The quotation of texitur exitium
examenrapit,a metrical unit makingno sense, at 155 and theattribution of
two tragic verses to Terence's Phormioat 157 suggests that here Cicero's
examples arc drawn from a grammarian's treatise rather thanfrom hisown
direct reading. Errors of attribution are common in theblocks of illustrative
passages which went from one grammarian to another. Verrius Flaccus is
known to have been prone to quoting units of metre rather than units of

1
Tentatively by G. Murray, Greek Studies (Oxford, 1946), p. 130, and
strongly by F. Scheidweilcr, PhilologusXCVII(1948), 324 n. 1.
' E.g. Diint1.er,Ztitschr.f. d. Alt. 1838, 61-2, Hartung, Euripidesrestitutusn,
p. 244.
3 Cf. Euripides, Or. 281 ff.
4 For this cf. Plautus, True. 6oo ff., Virgil, Aen. 6.46 ff., Seneca, Ag. 710 ff.,

Lucan 5. 161 ff.

s Cf. Cicero, Opt. gen. 18, Ac. 1. 10, De orat. 3 .27.
205
 ________
---......... --·----~ ·~-- ---

COMMENTARY
sense;1 he must have beenfollowing a practice already well establishedat the
time of composition of the Orator.Grammarians alsotended to quote from
successivepartsof the one speechor scene;the lexicon of Nonius Marcellus
provides many examples. At Orat.1s1 a quotationofTer. Plwrm.390 follows
one of 384, while the three quotations of Pacuvius' Chrysesat Is s appear to
be successiveverses. One would expect the two Ennian pieces, palrismei
meumfactumpudetand texiturexitiumexamenrapit,to be of similarcharacter.
Columna suggested in his commentary that the verses in question ~
longed to the same scene as provided those attributed by Macrobius to the
Alexanderat Sat. 6.2. 18 and2s. Ribbeck was thefust editor to include them
in a collection of the Alexanderfragments. There is no place for them in any
other known tragedy of Ennius or, indeed, of his republican fellows.a
At Diu. 1 • 42, in an account of prophetic dreams, Cicero quotes twelve
trimeters from the prologue of a Latin tragedy describing the dream which
led to the exposure of Alexander on Ida (fr. xvm). B. Heath3 seemsto have
beenthe fust to assign them to the Alexander.It is possible that they come
from one of the other known republican tragedies which dealt with the fate
of members of the Trojan royal house, EquosTroianus,Hectorproficiscens,
Andromacha,Hectorislytra, Hecuba, 4 niona, Astyanax or Troades.The pro-
logue of Euripides' <l>o{v1aaa1,a tragedy set at the time of the Argive siegeof
Thebes, went back to events before the birth of Oedipus. I have nevertheless
thought it hypercritical to separate a description of Alexander'sbirth from
the only Latin tragedy known to deal with him.
A number of brief pieces quoted by Cicero, Verrius, Q...uintilian and
Charisiushave been attributed at various times to the Alexander.In no caseis
the substance sufficientlyspecific to justify the attribution.
Barnes5set in Cassandra's canticum the iambics and trochaics quoted by
Cicero at Diu. 2. II S, o sancteApollo qui umbilicumcertumterrarumobsides,I
undesuperstitiosa primumsaeuaeuasituoxf era.
Snell6set the trimeter quoted without author's name in Festus' epitome of
Verrius at p. 306. 10, omnisaequalisuincebatquinquertio,in the messenger's
speech describing the games. Ribbeck7had compared Sophocles,'HMKTpa
69<>-3 and assigned the verse to Atilius' version of this play. There is no
evidencehowever that Verrius ever quoted Atilius.
1
Sec Introduction,p. 53.
2
Bothe made the strange suggestion (P.S.L. vol. v, p. 275) that they be
placed in Accius' Hecubaor Troades.
3 Notae (Oxford, 1762), p. 163.
4 Aldus Manutius and Welcker (Die griech. Trag. p. 463 n. 3) proposed

Ennius' Hecuba:Bothe Accius' Hecuba.

s Euripides,p. s17.
6 Euripides'Alexandros,p. 38. 7 Die rom. Trag. p. 609.

206
 ALEXANDER
Hartung1 set thetrimcter given to Ennius in Paulus' epitome of Festusat
p. 507. 12 (&. ccvm), is luJ,et coronamuituLmsuiaoria,in a speech by Dei-
phobus. C. Lcfk.e1 compared part of the Strasbourg papyrus of Euripides'
•~av6pos: ,roOw( V a)v et,,K<XAAfV1Ka Cxc,.w~,, j I - ,rciv 6:0'TV
,rA11pot Tpwucovyavpovµevos(- &. 43 .41-2 Snell); thisappearsto come
from a conversation betweenHecuba and Dciphobus after Priam' s decision
to let Alexmder's victory stand.
Lange3set theexample of 6µ01cmAEVTOvquoted by Q_uintilianat Inst.
9. 3. 77, Hecuhahoedoletpudetpiget,in Cassandra'seanticum,after thewords
Vahlcn4 suggested that the phrase might come
superstitionslutriolationibus.
after illos ob~qui. Its substance would be quite otiose anywhere in the
cantitum.N. Terzagbi5interpreted it as either a misquotation by Q_uintilian
or an imitation of Ennius by another tragedian.
Ribbeck6 set the example of emotional speech quoted by Charisius at
p. 315 .19, heusheuspaterheusHector,in thesame canticwn. The phrase seems
as otiose here as does Hecuhahoedoletpudetpiget.

XVII
(e) Cicero quotes remnants of a scene in which Cassandrabehaves at first
with virginalpropriety, then goes into an ecstatic trance, returns momen-
tarily to mental normality, realises the cause of her ecstasy, and is finally
confronted by a series of visions of the fate of Troy and its rulingfamily.
Verses 32-3 are spoken by the chorus leader? ancl vv. 41-2 provide in-
directly theanswer to his question. It was thepresence of Alexander either
inside thepalaceor on stage and a dim perception of hisidentity that caused
Cassandra'seyes to burn and her general demeanour to alter. Her plea to the
citizensof Troy (other bystanders as wellas members of thechorus) in v. 42
only makes dramatic sense if the spectatorshave seen or can see on stage the
physicalAlexander.8 Verse 39, optumamprogeniemPriamo peperistiextrame,on

I EuripidesRatitutUSD, P• 238.
a De EuripidisAlexandto(Diss. Munster, 1936), p. 63.
3 VindiciaeTragoediaeRomanae,p. 48.
4 E.P.R. 1, p. 128. S BFC XXXIV (1927/28), 43.
6
Q!!atst.seen.p. 26o.
7 &liton of Cicero and Enniw seem to me quite wrong in making Hecuba
the speaker without at the same time changing uisaest to uisaes and ilia to ista.
8
The appealto the body of citizens at a moment of danger (see W. Schulze,
•Beitr. z. Wort- und Sittengeschichte11 ', SB Berlin 1918, 481 ff. [ = Kl. Sehr.
160 ff.)) is a common motif of tragedy (c£ Sophocles,O.K. 884, Euripides,
Heralcleidai 69--10, I. T. 1422-30, Or. 1621-4) and comedy (c£ Aristophanes,
Neph. 1322-3, Plautw, Aul. 406, Mm. 1000, Most. 1031, Rud.615-26, Terence,
Ad. 155-6).
207
 COMMENTARY
the other hand sounds very oddly if Alexander has already been recognised
by his family. There is tragic irony here. Only the spcctaton arc as yet fully
aware that the bearer of fire and destruction has arrived. We cannot now
decide whether Alexander is visible to the spcctaton as Cassandra speaksor
whether he is off stage, i.e. inside the palace. There are good parallelsfor both
possibilities. In Euripides' BaKxoo Pentheus sees the handsome stranger
standing on stage as a bull (920 ff.)and Agave describesthe head of Pentheus
which she carries in her hands as that of a mountain lion (n68 ff.). In
Aeschylus' •AyaµeµvoovCassandracries out at the culminatingpoint of her
series of visions of past and future events in the Atreid palace 6-rrtxe Tfis
f3oosI TOV.Tavpov(n25-6), meaning Clytaemnestra and Agamemnon
whom the spectators have just seen enter the palace.
W elcker made Cassandrautter her prophecies as Dciphobus was attempt-
ing to kill Alexander (Hyginus, Fab.91.6); Hartunia£icrAlexander had
been recognised as Priam's son. Vater,3 on the other hand, argued that the
scene comes from very early in the play, long before Alexander had arrived
from the arena where the games were held. Such a view would demand a
very forced interpretation of Ennius' words but some sort of parallel is
offered by Ovid's account at Epist. s .115-20 of a vision which Cassandra
saw before Helen's arrival in Troy.
Ps. Longinus refers at Is.4 to a graphic utterance by Cassandrain a play by
Euripides, quoting the beginning of a trimeter: aXA' G>cp0.1mro1T~
(fr. 935). Wilamowitz4 pointed out that of the Alexandrian seventy the only
one to which Ps. Longinus could be referring is the •AfJ~avSpos. Hartung' s
guess5that Ps. Longinus quotes the original ofEnnius' duesferte opemis thus
very likely correct.
Lefke6 argued that vv. 32-3 (reading rabererather than ,apere) were a
version of those Euripidean trimeters preserved in Strasbourg as
X - u - X - u ]flSf\1<ova•rn'oS
X - u - X - u ~ ]CXJ<XEVEIcppwa.
Snell7accepted thisargument and argued further from the appearance of the
papyrus that the verses belonged in the first scene near verses restored by
Wilamowitz as
,ral6a K[(?:a®Spav~
Kai µi>tv& )Sop1<a
c'r>[Se
fl1<ovaa]va&'.n-oov irapos.
<l>o1f3E{oov

1 3
Die griech.Trag.pp. 472 ff. EuripidesRestitutusn, pp. 244 £
3 Untersuchungen, p. 23; c£ Snell, Euripides'Alexandros,pp. 24 ff.
4 Anal. Eurip. p. 148. S EuripidesRestitutusn, p. 244.
6 De Eur. Alex. p. 97. 7 Euripides'Alexandros,p. 25.

208
 ALEXANDER
To Snellthewords preservedimplied that Cassandraprophesiedthedestruc-
tion of Troy in thisfirst scene.One of two embarrassingconclusionswould
have to follow from this: either that Vater's interpretation of the Enoiao
scene was correct, for two prophecies could hardly be cxpcctedin theone
play, or that Ennius did not adapt the Euripidean play at all It is likely
enough that Caswidra did appear in thefint sceneof the•~av6pos and
that someone did say of her: ~Ml cppiva.But no one could argue that
this phrase comes neccmrily from a description of her behaviour at the
moment of speaking. Several pombilities are open.
lamqueat v. 43 indicates that Cicero omitted1 the beginningof the'IA1as
mtilorum, perhaps an account of theGreek chieftainsassemblingtheirforces at
Aulis. Ennius' Cassandra actually sees the Greek shipssweeping across the
water (texitur. . .rapit),thesoldiersdisembarked on theshore (compleuit), the
three goddessesjudged on Mount Ida (iudicauit).She perceives the evil to
come intdlectually rather than visually; hencethefuture tenses odutniet. ..
aduenkt.i
While convening with her mother the Eooiao Cassandra speaks in
stichic trochaic tetrameters and continues with theseverseseven aftershehas
realised who the stranger is. From Ps. Longinus' quotation it is clear that
iambic trimeters stood at thispoint in Euripides' play. The fint scenesof the
Latin prophecy are described in dactylic verses; the more sophisticated
spectatorswould perhaps have associatedthe rhythm with that of the Greek
oraclesrecordedin the Sibyllinebooks.Whetherany Latin oracleswere
composed in hexameters before thepublication of Ennius' Annalawe do not
know.3 Other scenesof the prophecy are describedin metres now difficultto
discern.The change of metre at v. 43 is parallded at Euripides, Tr. 444,
where Caswidra passesfrom trimeters to trochaic tetrameters for the climax
of her prophecy, and Seneca, Oed.233, where thecontents of an oracle are
given in dactylic hexameters after a description of themedium's behaviour
in trochaic tetrameters. It was rash of Strzdecki4 to argue from an alleged
imitation by Senecaat Ag. 693-774 that Euripides cast Cassandra'sprophecy
in iambic trimeters from beginningto end.

1
For Cicero's habit of making omissionsfrom the middle of a quotation
without warning cf. De orat.2. 327and Terence,Amlr. 117-29,Att. 1. 3. 10 and
Terence, Eun. u4-1s.
a For prophecies (future tenses)alternatingwith visions (presentand perfect
tenses)cf. Aeschylus,Ag. 1090-172,Pindar, Pyth. 4.-49,8 .4s ff., Horace, Cams.
l . 1S, Virgil, Am. 6. 86 ff., Tibullus 2. S . 39 ff.
3 0. Skutsch,CR N.S. vm (19s8),47,dcniesthepossibilitywithunnccessary
firmness.
4 Dt SentcaeAgtllrffflftlOM Euripidis~ Altxandro (Wroclaw, 1949),p. 21.

209 JTO
 COMMENTARY
32 sed quid oculisrapere uisa est: for the use of std when something
new is seen or heard to happen on stage c£ Plautus, Amph. 270, 1130 et al.
Lambinw' rabere1 is quite unnec.essary.In any caseraberewould be used
metaphorically, 'rage like a mad dog', and hardly need to be accompanied
by uisaest.
For oculisrapere,'use the eyes as if they were handsto grab with• c£Trag.
inc. 48 oculispostremumlumenradiatumrape,Plautus, Poen. 277 haectanta
oculisbona concipio,Virgil, Georg.2.230 locumcopiesoculis.The normal
phrase in comedy is oculisaspicere(uidere,intueri,amere). Attic tragedy has
6q,6CXAµots (6µµ(X(71, 6aao1s) 6pav (&pKE06a1)but nothing like oculis
rapere.Plautus frequently transfersthe qualities and functions of one part of
the body to another for comic effccf with no analoguesin theNm and few
in the •Apxata.

oculi, • •• derepente ardentibm: c£ Plautus, Capt.S94 ardentoculi(of an

apparent madman), Virgil, Aen. 2.404-s Cassandra ..• ad caelumtendens
ardentialuminaJrustra.
Derepenteoccurs 4 times in tragedy, 6 in comedy (Plautus, Men. 874,
Most. 488, Terence, Hee. 518, SS4, Turpiliw 3, Com. inc. 22). The simple
form repenteoccurs only once in tragedy but 13 times in comedy. The con-
texts of occurrence of derepente in comedy confirm the statisticalindication
that it had a high-falutin tone.

33 ubi ilia paulo ante sapiem tuirginalitmodestia: Lachmann (on

Lucretiw 1 . 186) removed the hiatuspauloIanteby supplying( aut) ubi. .. and
accepted the old interpretation of uirginalias uirginali(s),3
despitethe unusual
division of the fifth foot of the tetrameter4 and the necessity of scanning ilia
as two morae.S This neverthelessremains the most economicalrestoration so
far suggested. 0. Skutsch argues tcntativdy for sapiensilia uirginalis faulo
anteubi modestiaor ubi iliapauloante (apta) sapiensuirgin[al]inwdestia.
For ilia paulo ante. .. modestiacf. Plautus, PerStJ38s non tu nunc lwminum
moresuides. . ?, Poen.72s rem aJuorsuspopuli saepeleges.
Sapere,sapienswually have a personal reference in both tragedy and
comedy but c£ Pacuviw, Trag. 6 quodconiectura(Roth: consectura codd.)
sapiensaegrecontuit(Vossiw: contuUtcodd. ). Tragedy is prone to apply to the
1
This was suggestedin the margin and the notes of his 1573 edition. The
notion that it occun in manuscripts seems to come from a misreading of
Muretus, VariarumLectionumlibriXV (Antwerp, 1580), IX 19.
1
See Fraeokel. Pl. im Pl. p. 107 n. I ( = Elementi,p. 101 n. 1).
3 Cf. Leo, Pl. Forsch.',p. 290 n. 1. 4 Cf. above on fr. :xn.
s See F. Skutsch. Plautinischesund Romanisches(Leipzig, 189.2),pp. n6 f.
6
HSCPh LJOO (1967), 133.
210
 ALEXANDER
inanimate and the abstract adjectivesand participles usedproperly of living
things: e.g. Ennius 35 superstitiosishariolationibus, 53 curis. .. suspirantibus,
Pacuvius 36, 47, 109, 147, 386, Accius 33, 3 8, 46, 133, 315,398,412, 4S3, 501,
Trag. inc. 20, 58, S9, 114-

U mater, optamatum multo mulier melior mulierum: there is no

needto tamper with optumatum(optumatum codd.) or any of the other words
transmitted. Asconius and Nonius cite passages of Cicero cont.iining the
genitive plural optunuitum1
while considerable contamination of the vowd
and consonantal stems of the third declension is recorded dsewhere in early
Latin texts. Optumarumwould be more normal (c£ C.I.L. 11 9. I £ honeoino
ploirumecosentiontRomaneduonorooptumofuise uiro, Plautus, Amph. 676-7
AmphitruouxoremsalutatlaetussperatamsuamI quamomniumThebisuir unam
esseoptumamdiiudicat,Capt. 836 quantumest hominumoptumorumoptume)but
.Enni.usappean to have madehis Medeacallthe women of Corinth matrotuJe
opulentaeoptumates(seebelow on fr. cv) and Plautus has the similar forma-
tion summasthree times of members of the .Athenian land-owning class
(Gist.25, Pseud.227, Stich. 492; c£ the humorous infumasat Stich. 493). In
the political pamphlets of the first century the senatorial supporters of the
statusquoreferred to themselvesregularly as optumates.iThe word suggested
socialand moral excellenceas well as wealth and power (ops).
For the pleonasm optumatum. .. meliorc£ Aeschylus,Hik. 524-6 µCXK<Xf)Gl)VI
µCXKapT<X'TE Kai rueoov I rue16-rcrrov, Sophocles, 0. T. 334, KCXKOOV
KaKlCTTE,Naevius, Com. I 18 pessimorum pessime,Plautus,Capt. 836 optumorum
optume, Men. 817 miserorummiserrumus,Horace, Sat. 1.3.136 magnorum
maxime.
For mulier melior mulierumcf. Aeschylus, Hile. S24-S &vex~&vcocroov
µCXK<Xf)Gl)VI µCXKapT<X'TE, Sophocles, 0. T. 66o TOVTI'av'T(a)V 8e<;;)vOeov
,rpoµov, Plautus, Capt. 333 optumusquehominumes homo, 540, 825, Trin.
1115, Terence, Phorm.853, Hee. 861, Ad. 218.
For thecomparative meliorrepresenting the superlative c£ Plautus, Capt.
825 regumrex regalior,Varro, Rust. 2. s. 10 Epirotid meliorestotius Graedae.
Ennius aims at a certain kind of parech~s comparatively rare in Attic
tragedy (c£ however Sophocles, Tr. 947 TTOTEPa TTpOTEPov hncmvoo,
Euripides,I. T. 1339 HMEN HMEN01,Phoin.1174 EXWvEXWpu et al.)
but very common in the Latin adapt.itions (e.g. Naevius 3S ACReM ACRi-
Moniam,39 inFLExV FLECtitVr, Pacuvius 37 praeGRAnDi GRADu, S3
STVdlo obSTVplda, 203 uMoREM riMaREM, 246 MiNVaM MaNVuM
et al.).

1
Nonius, p. 409. 32, Asconius, p. 71 . 14 Stangl.
3
Cf. Rhct. inc. Her. 4.45 et al.

211
 COMMENTARY

It is unlikely that Euripides made his Cassandraaddressher mother in such

a bloated style (c£ Tr. 3S3 µiynp 11'VKCX3E ... ). The early Latin adapters of
Greek poetry tended to add descriptive matter to vocatives (e.g. Livius,
~
Cann. fr. 14 sanctapuer Saturni}ilia regina Od. 4.513 T16-rvta•HpTl,
~
Ennius, Trag. 237 antiquaerilisfida custoscorporis Euripides, Med. 49
11'aAcnov oi1<oov t<'riiµa&a-rrofvrisiµf\s, Trag.234 summeSol qui resomnis
~
inspids Euripides. Med. 1251 ,raµcp<rlis aK"rls •AM(ov)and to other cases
in solemn narrative (e.g. Livius, Cann. fr. IS apud nymphamAt1'mtisfiliam
~
Calypsonem Od. 4.557 wµcp11siv µeyapo1ai K~\l\f/0\IS. Ennius, Trag.
~
214 imperioregisPeliae Euripides, Med. 6 TlU\(q:).We may have here a
rcfiection of the modes of speech employed in upper-class Roman society
on formal occasions.

35milla sum aupentitiosis bariolationibm: the expressionis as bizarre

and resistant to rational analysisas that of the Aeschylean Cassandra at Ag.
1209-ri)(vcnaiv w6eo1stP1l"1Ml·
I should guess that Ennius thought of his Cassandra as a horse ridden by
the power of clairvoyance; c£ Virgil, Aen. 6.77-101 .. . Phoebi nondum
patiens. .. si• •. possitexcussissedeum. . .fatigatI os rabidum,fera cordadomans,
fingitquepremendo. .. eafrenafarenti Iconcutitet stimulossubpectoreuertitApollo.
This field of imagery is commonly usedin connexion with non-prophetic
madness (e.g. Aeschylus.Prom.S97, 883, Euripides, Hipp. 237, I. T. 81, Or.
36, El. 1253). Here there are obvious sexualassociations.1
For missasum, 'emiua sum, immissa sum•, c£ Cicero, Fin. I • 2 d!fficikm
quandam temperantiam postu1'mtin eo quoJsemeliam missumcoerceri reprimique
non potest,Tibullus I. 4. 32 qui prior Eleo est carceremissusequus,Propertius
3 . II. 62, Seneca, Med. 874. For the replacement of compound verbs with
their simple forms in tragedy see above on v. 14.
',rapacrrar11<6s,prophetic', is normally usedof personsin
Superstitiosus,
republican drama1 but cf. Trag. inc. 18 superstitiosa .•. uox.
The noun hariol4tio occurs elsewhere in Latin only at Gcllius IS. 18. 3. The
great majority of forms in -tio in republican drama {Ploen counts 18 in
tragedy, 123 in comedy) retain a strong verbal force.
The pluralsof abstractnouns and incorporalia occur frequently in tragedy:
e.g. Ennius 36 fatis, 44 exitium, S3 curls, 103 aerumnis,223 miserias,231
1
Female ecstaticseen were sometimes regarded as Apollo's concubines; sec
Herodotus 1. 182, Ps. Longinus 13 . 2, Pausaoiu 10. 12. 2, Apollodorus, Epil.
6. 3, Norden, Aen. VI•, pp. 145 f., Latte, HThR xxxm (1940), 9, P. Amandry,
IA MantiqueapollinienneaDtlphes (Paris, 1950), pp. 117 f.
:a Plautus, Amph. 323, Cure. 397, Rud. 1139, Pacuvius, Trag. 216. See E.
Linkomies,Arctos D (1931), 73 ff., E. Benveniste,REL XVI (1938), 3S, Latte,
RomischeReligionsgeschichte (Munich, 196<>),p. 268 n. 1.
212
 ALEXANDER

moerores,
Pacuvius 81, 155, 175, 319, 356, Accius IS, 36, 42, 73, 94. 109, 154,
156, 160, 171, 175, 195, 219, 224, 293, 352, ,468, 481, sso,SS3, S77, 587, 621,
694-They are not excluded from comedy, even by Terence, and do not seem
to have beenfelt as such outright poeticisms as their equivalents were on the
Attic stage.

36 tnequefme Apollo &tu fandisdementem inuitam ciet: it would

benatural to understand transmittednequeas negativing the whole sentence1
if it were not for the contc:xt.Vahlen takesthe word closelywith inuitamand
compares Aeschylus, Ag. ~202 ff. and Euripides, Tr. 451 ff. Neither of these
passagesimplies that Cassandrawelcomed or enjoyed the physical presence
of Apollo. In any case Ennius' Cassandra goes on to express shame at her
periodic fits. The relevant parallds are Aeschylus, Ag. 1174 ff., 1215 f.,
Euripides, I.A. 760, Tr. 408 ff.,where Cassandrais assumed to behave under
an cxtemal, unwanted compulsion.1 Hottingcr's ru,mque me Apollois palaeo-
graphically attractive but the division of thefirstfoot dactyl is odd (seeabove
on fr. xn); likewise the use of """'i"' before a consonant (absent from
comedy; but c£ Livius, Cann. fr. 20, Nacvius, Trag.41 (?), Ennios, Trag.
339). Ribbeck's namqueApollo is a little more convincing than Grotius'
mequeApollo (for the omission of the anaphoric pronoun see above on
v. 17).
Farioccurs five times in tragedy, of the utterances of ordinary men as well
as of seers. Comedy has it only in the idiomatic phrases f ando aaipere
(Plautus, Amph. 588) andfando mulire(Epid. 496) and the comic creation
ingeniumfansatqueinfans(Persa174). The substantivefatum, 'statement about
the future', sometimes 'event (usually evil) prophesied', perhaps never
'fate',3 is fairly common in both genres. For the etymological figure c£
Euripides, Phoin.409fxp11a••ASpcxo-r<t) A~las XP1laµovTiva.
Ciet probably comes from the same field of imagery as missasum.

37 uirgines uereor aeqaalis: for uereor,'alaxvvoµm ', I can find no

exact parallel. It seemsto be a poetical replacement of the compound reuereor
(c£ Plautus, Epid. 173.filiumreuereor).
Cassandrahas ceasedto be herself properly a uirgo;sheis now the paelex
of Apollo. Aeschylus had represented his Cassandra as a socialoutcast: Ag.
1273-4 K<XAOVµM')
6~ cpo1TQS ~ ayvp-Tp1aI 1TTooxos TaA<nvaA1µo8vfls
Women of the rulingclassdid not go in for ecstatic prophecy
ftveox01,111v.
either in fifth-century Athens or in second-century Rome.

1
But cf. Terence, Phorm. 114.
a Cf. Timpanaro, SIFC XXI (1946), 58 f.
3 See below, p. 305.

213
 COMMENTARY
patris mei meam factmn pudet: 'I am ashamedat what my father
thinksof thethingsI do.' Pudetdocs not seemto be accompanied elsewhere
by two genitives in this way.
Genitive plural meumis transmitted three times elsewhere in republican
tragedy (Pacuvius So, 401, Accius 424), not at all in comedy except in the
company of maiorumand parentum.

39 progeniem . .. peperiati: the stockphrase in republican drama was

puerumetc. parere.Here we have a poetic substitute preserving the allitera-
tion. Progeniesoccurs three times elsewhere in republican drama (Terence,
Plumn. 39S, Pacuvius, Praet. 1, Accius, Trag. so; cf. Lucilius Soo}in the
meaning 'ancestry'; for the meaning 'of&pring' cf. C.1.L. 11 1s progeniem
genui, Cicero, Tusc. 1 . 8s Priamumtanta progenieorbatum.

hoe dolet: for the rhyme det • .. pudtt. .. piget•.. dolet extendingover
four trochaic tetrameters I can find no parallel in Greek or Latin drama. Over
two it is quite common: e.g. Aeschylus, Pers. 171 f., 24,7f., Euripides, &.
618 f., 622 f., 628 f., I.A. 338 f., 361 f., 392 f., Plautus, Amph. 308 £, 3S8f.,
366 f., 368 f., 413 f., 41S f., 43S f., 547 f., 806 f., 1127 f.

.fOmen obeue, illos prodeae, me obstare, illos obleqai: for theex-

clamatory accusative and infinitive involving a pronoun or pronominal
adjective and the particle -nt cf.Plautus, Pseud.202 a: hundnt hiehomintmpati
colereiuuentutemAtticam,371, Terence, AIUlr.689, Eun. 644, Cicero, S. Rose.
9S,Horacc, Sat. 1.9.72-3,2.4.83,2.8.67, Virgil.Aen. 1.37. Asimi]arr.ypc
of expression is common in Attic drama and oratory; cf. Aeschylus, Ag.
1662 ff. @.Aa Tovas· iµol µara(av y°AC>aaav &s· arrav6(aai. .. , Bum.
837~. Sophocles, Ai.410-11, Demosthenes 21 .209, 2s .91. Elaborate gestur-
ing would have probably accompanied an excited tone of voice.
For the tautology of the two phrases sec above on v. 19.
Similar anaphora of the personalpronoun occurs in highly emotional
passages of comedy; cf. Plautus, Asin. 148 te egout ulciscar, te egout dignaes
perdam,Men. 101s, Mil. 331, 1386, True.441, Terence, Eun. 193 ff.

41 adett adett &x obuolata angaine atque incendio: this vision

looks like a doublet of the dream which Hccuba had of giving birth to a
fiery torch when pregnant with Alexander.Among the many accounts of
thisdream only Ovid's (Epist.17 .237 ff.)mentions fire as well as blood. For
the combination cf. Euripides, I. T. 288 mip 1MO\Jaa Kal '6vov,Virgil,
Aen. 2. 210 oculossuffectisanguintet igni.Blood was a common ingredient of
Roman portents (e.g. Cicero, Diu. 1.46, 1.98, 1.99, 2.ss, Livy 22.1.10,
21.63.13, 23.31.1s et al.).

214
 ALEXANDER

For the gcmination of the present indicative cf.Euripides, Alie.259 &yel µ'
&ye1 !,If TIS (vision of death), Virgil. Aen. 6.46 deuseca deus,Ovid, Met.
15 . 677en deus est deusest. Comedy doubles vocatives and imperatives fairly
frequently.
Obuoluerewas certainly archaic in tone in Ciccro's day. Of the three
examples in his writings two arc in legal contexts and one in a reference to
heroic legend (Vm. 5.72, Inu. 2.149, Orat. 74); he has inuoluereoften. We
cannot be certain about Ennius' day: the only other occurrence of obuoluere
in republican drama is at Plautus, Most.424 in a speechin high-falutin style
by a slave; inuolueredocs not occur at all.
Incendiumis used elsewhere in republican drama only metaphorically
(Plautus, Asin. 919, Mere. 590); in classicalLatin it is used of fues rather
larger than those one would associatewith a torch.

42 ferte opem: a high-falutin variant off erte auxilium;perhaps from the

sacrallanguage (c£ Trag. inc. 241, Terence, Amir.473, Ad. 487, Turpilius,
Com. 118); the tone is elevated at Plautus, Bacch.637, Mil. 1387, Rud.617.

43-4 iamque mari magno classis cita I texitur: c£ Virgil,Aen. 6. 4-5 et

litoracuruaeI praetexuntpuppes.Cicero uses a related field of imagerywhen
he turns Homer, n. 2.303-4 x8t3a TI: 1<aiirpootf 6-r' ~ AvA{6a vf\ES
'Axoooov I Tl~VTO l(<Xl((X TTp1aµ'¼) t<alTpooai~VO'OO into Argolicis
primum ut uestitaest classibusAulis, I qr44e Priamoclademet Troiaepestemque
{erebant(Diu. 2. 63 ).
For the alliterative phrase maremagnum,'open sea•, c£ Livius, Trag.33,
Ennius, Ann. 445, Lucretius 2.1, Cicero, Fam. 16.9.4, Rep. 6.21.
For the omission of in c£ Lucretius 2 . 1 mari magnoturbantibusaequora
uentis,Virgil, Aen. 3 . 204 mamus pelago,6. 697 stant sale Tyrrhenoclasses,
Cicero, Fin. 5 •9 naturasicab iis inuestigata est, ut nullaparscaelomari terra,ut
poeticeloquar,praetermissa sit. There is a closely parallel phrase at Sophocles,
Tr. 114-15 t<Vµcrr'avevpei 1T6VT'¼) ~· rn16VTaT' t6o1. Ennius fre-
quently dispenseswith prepositions in the manner of Attic tragedy: e.g.
at Trag.45, 76, 214, 218, 299, 330.
Citusaccompaniesa verb in four obviously paratragic passagesof comedy
(Plautus,Amph. 244, 1111, 1115, Stich. 39.1); c£ Sophocles,O.K. 307 &Op'
6:«p{~eroo TCX)(VS et al. The adverbial use of adjectivesis common in Ennius'
epic poetry (e.g. Ann. 21, 35, 40, 42, 80, 85, 203, 519) but rare in early
tragedy; cf. however Accius 396, 403, 570, 576, Trag. inc. 63, 75.
For the present tense in a vision of the future c£ Homer, Od. 20, 351 ff.,
Aeschylus,Ag. 1072-177, Euripides, Alk. 252-5, Virgil, Aen. 7 .98 (Servius'
reading of the text), Ovid, Epist. 5. 115-.20. The so-called 'prophetic future•
is somewhat different; see below on v. 61. Cassandraactually seesthe events
as they happen.
215
 ___
- - ....---.......__, ======-:;,-=-=----·
------

COMMENTARY
44 entium eDmen rapit: c£ Virgil, Atn. 12. 4so illeuolatcampoque atrum
rapitagmenaperto,Livy 3. 23. 3 dtatumagmenTusculumrapit,23. 36. 3 dtatum
agmenpraeterCapU4mrapit,2s . 3s.1 adHasdrubalem Hamilcarisdtatumagmen
rapiunt.
Examenis usedinstead of agmenfor the sakeof alliteration with exitium;
but cf. Plautus, True. 314 nequeistucinsegtstitergoeogetexamenmali, Pro-
pertius 2. 32. 41 in tantostuprorumexamine.
.&,pert, 'macto move quickly', is a usage foreign to comedy but com-
mon in tragedy: e.g. Ennius 111 rapit ex alto nauesueliuolas,Accius 396,
Trag. inc. 196, 237.

45-6 ... fera ueliaolantibus I naaibus compleuit manm litora: c£

Virgil, Atn. 6. s-6 iuuenummanusemieatardensI litusin Hesptrium.Many take
nauibusas instrumental. It is perhaps a pregnant ablative of separation (i.e.
nauibuseg,essaeompleuit).
The tense of compleuitis confirmed by thetextually quite secure tractauere
of v. 71.
The disjunction puts a heavy emphasis on fera; c£ the way in which
Virgil's Aeneas speaks of the Greeks and their gods at Atn. 2. 29, hie saeuus
ttndebatAchilles,and 2. 326-7 f erusomniaIuppittr Argos I transtulit.
Ships frequently fly in ancient poetry {Aeschylus, Hik. 734, Pers. SS9,
Euripides, Hel. 147, Med. 1, T,. 1086, Virgil, Atn. 3. 124), sometimes by
means of their oars {Homer, OJ. 11. 12s, Euripides, I. T. 1346, Propcrtius
4.6.47), sometimes, as here, by means of their sails (Hcsiod, Erg. 628,
Aeschylus, Prom. 468, Euripides, Hipp. 752, Ennius, Trag. 111, Lucretius
4. 390, 5 . 1442, Virgil, Atn. 3 . 520, Ovid, Pont. 4. 5 . 42 ).
The ueliuolanstype compound is common in archaic Latin epic (Naevius,
Cann. fr. 30 arquittnens,Ennius, Ann. 81 altiuolans,181 bellipottns,181 sapi-
entipottns,195 belligerans, 303 suauiloquens, 458 omnipottns,541 altitonans,ap.
schol. luuenalis 7.134 [see Wessner, BPhWxun (1923), 572 ff.] multisonans)
and in tragedy {Ennius 97 signittnens, 150 omnipotens,Accius 52, 167
arquittnens,127 armipottns,Trag. inc. 36 crispisuleans, 150 praepotens).
Plautus
has a number of -potens compounds in prayers and paratragic speeches
(omnipotensat Poen.275; multi- at Bacch.652, Cas. 841, Trin. 820; prae- at
Poen.1182; ante-at Trin. 1116; salsi-at Trin. 820; uiri-at Persa252; eaeli-at
Persa755 ). They have no counterpart in Greek and very few appear in classi-
cal Latin poetry. It is tempting to say that therepublican poets got them by
adoption or extension from the contemporary sacral language but extant
prayers have nothing like them.
The plural litorastands for the singular in republican drama only here and
at Accius, Trag. 526 (anapaests); c£ ma{ in Attic tragedy. Other plural
concretes standing for singulars in Latin tragedy are aequora{Accius

216
 ALEXANDER

224), caua{Ennius96), delubra{Accius 527),freta (Trag. inc. 183), lumina

{Naevius28),pectora (Trag. inc. 96), Pergama(Livius 2, Ennius 74), sceptra
{Accius3, Trag. inc. 104), su,gna{Accius335), teaa (Trag. inc. 242), temp/a
(Ennius 98, 171, Pacuvius 311, Accius 529), terga(Livius, Ino 3), thesauri
{Ennius 192).

(d) Haupt observed trochaic rhythm in the words of this quotation. 1

.EMUuidetemight just as readily be regarded as a colon rcizianum as theend
of an acatalcctic tetrameter.:'
The tense of iudicauitshould be preserved along with that of compkuitin
v. 46, troctauereinv. 71 andsuperauitinv. 72. The future would beabsurdafier
the imperative uidete.Cassandra'smode of speechis the same as that of the
Acschylcan Cassandra at Ag. 1214-22, Euripides' Alccstisat Alie.259-63
and Virgil's Venus at Aen. 2.004-18 (aspia.•• Neptunus.•. fandamenta
quatit•••JunoScaeassaeuissima port.asIprimatenet. .• iamsummasarcesTritonia
- respia- Pallas I insedit. •. ipst paler. .. ) ; all arc pointing out things
invisible to ordinary mortals.
Cassandra could scarcely be referring to common knowledge about the
historical past;3 uidetemakesit plain that she is describinga vision. The
vision on the other hand could be of a past event. Ancient diviners were
capable of seeingthe past as well as the future.4 Cicero, however, obviously
understood iudiC11Uit as referring to the future and although he was capable of
mishandlingpiecesof dramatic poetry for thesakeofhis argument we should
not abandon his interpretations of them without good reason.
Whether the vision is of the past or the future it occurs out of chrono-
logical order.S This is odd. Aeschylus' Cassandra (Ag. 1072 ff.) has her
visions in strict chronological order; so too Seneca's (Ag. 726 ff.).
If the vision is of the past the Latin Alexanderwould have been aware
from thebeginning of the action that he was something other than an ordi-
nary slave. Yet in Euripides' tragedy therecognition was as much a surprise
to him as it was to his parents. The debate with Dciphobus assumesthat he
was preciselywhat he seemed to be. If, on theother hand, thevision is of the
future, it is odd that after being rccogniscd as a prince of theroyal house he
1
Philologusm (1848), 547 ( = Opusc.I 209).
a Sec Mariotti, Lezioni su Ennio,p. 134. But for the rarity of this colon in the
company of trochaic versessee Maurach, PhilologusCVII (1963), 252.
3 So Blicheler,RhM xxvu (1872), 477.
4 Cf. Homer, 11.1. 70, Hesiod, Th. 38, Aeschylus,Ag. 1072-330, Euripides,

I.T. 1259--68, Virgil, Gtorg. 4.392-3, Ovid, Met. 1.517-18.

s Cicero's words at Diu. 2. n.z--Cassandra 'iamque marl magno' eadtmque
paulopost 'eheu uidete'-establishes the order of the fragments.Plank, Q. Ennii
Medea,p. 118, was wrong to reversethem.
217
 COMMENTARY

should go back to herding cattle on Ida. Certainly in epic poetry the Trojan
princes arc often describedasherdsmen1 and in tragic allusionsto thestory of
the judgment Alexander is usually described as a herdsman.:.i However the
•AAE~av6p<>s, like many of Euripides' plays, brought theheroes into a world
similarto the contemporary one, where a member of a wealthy and power-
ful family was the last person likely to spend time tending cattle.
Thephrase inclitumiudiciumis odd in thedramatic context. One could take
the epithet as belonging in substancewith interdeastris.3But this is a bold
cnallage for republican tragedy. The tragedian may bewritingwith thefame
ofthejudgment in hisown day in mind. At Trag.211-12, with no prompt-
ing from his original, Ennius makes his actor speak like a contemporary
scholiastrather than a person of the heroic age.
The words Lacedaemonia mulierFuriarumunaaduenietare usually taken as a
close translation of a Greek sentence containing the word 'Ep1ws.4 The
Roman adaptcn of tragedy certainly called the "Epl\MS who punued
matricidesFuriae.SCicero6 translated Sophocles,T,. IOSI-2 Kaefi'f'EV ooµo1s
Totstµots 'Ep1woovIvcpcnrrovaµq,{~Af\O'TpoV as haeemeinretiuituestefuri-
ali. At Euripides, Tr. 4S6--7there is a sentence parallel in structure with
Ennius': oVJ<tr'CXV q,8avo1sCXV cxvpavIO'Tio1sKapa6oKOOV, I cl>sµ{av
TplOOV 'Ep1wv ,;;a~µ• t~a~oov)(8ov6s.In substance,however, it is not.
Euripides' emphasisis on the wickednessof Agamemnon; Cassandrawill be
the immediate causeof hisjust destruction and is fittingly compared with an
'Ep1ws.Ennius'emphasisis on thewickednessof Helenherself;sheis com-
pared with a Furiabecause Cassandrathinksof her as a bringer of fire and
destruction to Troy; the causeof her coming, Alexander'sdecisionin favour
of Aphrodite, could hardly be thought a punishable act of wickedness.
The semantic area of 'Ep1ws in Attic tragedy and that of Furiain re-
corded Latin are not exactly co-extensive.The 'Ep1ws was a spirit of ven-

1
C£ Homer, II. s.313, 6.25, 4,21 ff., II. 101 ff., 14.445, 15. 547 £, 20. 89 ff.,
188 ff.; for storiesrecordedoutsideHomer see Robert, Bild undLied,p. 234, Die
Grieeh.Heldensagem ii, p. 978 n. 3.
:a C( Sophocles, fr. 469, Euripides, Atulr. 274 ff., Hee. 629 ff., Hel. 23 ff.,
357 ff., 678, 708, I.A. 76, 180 ff., 573 ff., 1283 ff., Trag. Grace. inc. 286, Accius,
Trag.610, Seneca, Ag. 730 ff. (contrast the vagueness of Euripides, Tr. 919 ff.).
3 I am grateful for the suggestion to E. W. Handley. for enallage in
tragedy see below on v. 57.
4 So they were understood by Virgil at Aen. 2. S69-'73: Tyndaridaaspido. ..
Troiae et patriae communisErinys; by fraenkel, on Aesch. Ag. 749, p. 347
(' Euripides has adopted this conception in his Alexandrosif, as is probable, the
words ofEnnius go back to the original'); and by Wiist, RE Suppl. VIII (19s6),
118.
6
s Cicero, S. Rose. 67, Pis. fr. 3, 46, Leg. 1.40, Parad.18. Tuse. 2.20.

218
 ALEXANDER

gcancc and justice, never one of random destruction.1 Furiaon the other
hand was frequently wed, like the abstractspestis,pernicies,labes,exitium,to
indicate mad, wicked and/or destructive persons whose presence made
other personsequally mad, wicked and/or destructive: e.g. Cicero, Dom. 99,
102, Har. ,esp. 4, II, Sest. 33,-39, 109, 112, ·Vat. 31, 33, 40, Pis. 8, 26, 91, Fam.
1.9.15, Q!.int.fr. 3.1.11, Horace, Sat. 1.8.45, Livy 21.10.11, 30.13.12.
Not one_of the personsso indicatedhad any kind ofjustice on his or her side,
at leastnot in theview of theindicator. What Furiadenoted before the Latin
poetstook it up itis hard to say. Certainly it was the horrific externalaspectof
theAttic 'Ep1vvs,their fierytorches and serpentine adornmen~ rather than
their role as the keepers of a primitive system of justice that impressed the
imagination of all poets after thefifth century, both Greek and Latin.
What Cicero quotes at Diu. 1 . 114 may therefore be not a closeadaptation
of verses of Euripides' •AM~av6poSbut a cardess injection by Ennius into
thespeechhe wrote for hisCassandraof matter quite foreign to the•~av-
Sp<>s.The words eheuuidete. .. oduenietcouldhave led into a description,based
on one in the• ~av6p<>s, of a vision of the night of Troy's fall with Hden
waving, like a Furiafrom a tragedy, a torch to signaltheGreek attack.ers.2
Snell3argued from the shape of thedebate between Hecuba and Hden in
Euripides' T~Ses that there had been no reference to the contest of the
goddessesin the 'Me~av6p<>s,the first member of the 415 B.c. trilogy:
whereas Helen merdy alludes to the dream of Hecuba (919 ££), something
which was describedat length in the prologue of the •~av6pos, shegives
a most detailed account of how thethree goddessescame to Alexanderon
Ida; the words with which she introduces her account, Mev& TaniA011t'
6Kovaovoos fxa, imply that Euripides is setting something new before his
audience; again the scorn which Hecuba in her reply pours upon Hden' s
story4and the rationalismwith which sheexplains Alexander'sdeparture for
Sparta, treating Aphrodite as a symbol of the foolishnessengendered by
sexuallust, would have appeared rather strange to the audienceof 415, if, in
the earlier play, the divindy inspired Cassandrahad describeda vision of the
beauty contest. ScheidwcilerSand Strzelecki6 have argued that Ennius'
1
Wiist, RE Suppl vm (1956), n8, takes too narrowly modem a view of
justice. The exact point at Aeschylus, Ag. 749 and Euripides, Or. 1389 is
obscure and disputed but in neither case could Helen be regardedasa criminal;
in the one the crime was Alexander's insult to Mcnelaus' hospitality, in the
other Laomedon's perjury.
2
C£ Virgil. Aen. 6.517 ff., Tryphiodorus 512 ff.
3 Euripides'Alexandros,pp. 53 £
4 Stinton, Euripidesandthe]udgtment,pp. 38 n. 1, 64 ff., arguesthat Hecuba
docs not express disbelief at 969ff.
6
s Philologusxcvn (1948), 334. De SenecaeAgamemnone,p. II n. 19.
219
 COMMENTARY

words eheu uidete.iudicauitinclitumiudiciumetc. destroy Snell's position.

Things are the other way around. Properly interpreted the Latin words
confirm it. 1

48-9iudicauit inclitum iudiciam . .. quo iadicio: the spelling inclutus

seems to be the earliestrecorded in inscriptionsand there are traces of it in the
manuscript tradition of severalrepublican authors. The word was probably
not in common usein Ennius'day. It occursonly three times in comedy, all
threein paratragic contexts (Plautus,Mil. 1227, Persa251, Pseud.174), three
times in tragedy. ClassicalLatin confines it to poetry and history.
The phrase iudidumiudicaredocs not occur elsewherein republican drama
or in Cicero. Rem iudicareand iudidum dareare Cicero's regubr phrases.
Ennius however may have employed a contemporary legalism;,.for etymo-
logical figure in the language of law and poetry see above on v. 6.
Relative clausesrepeat the antecedent noun fairly frequently in the de-
vated parts of comedy (e.g. Plautus, Aul. 561, Cas. 3~11, Epid. 41,
1015-16, Rud. 997, Terence, Haut. 20, Hee. 10-11). The phenomenon docs
not seem to occurin Attic drama but is common in Latin legal texts.3Many
of the Latin dramatic examples have as here a juridical Bavour.

48-9 ... inter deas tris aliquis •••Lacedaemoma mulier Furiarum

una adueniet: there is probably a deliberate antithesis here between the
three Olympiangoddessesand the three underworldFuriat.
For the appellationof Hden cf.Euripides, Amir.486 fi Aa1<atva,Horace,
Carm. 3.3.25, Virgil, .Aen.6.511.
For Furiarumuna adueniet,'she will arrive like one of the three Furies', cf.
the construction of Horace, Epist. 2. 2. 99-100 disadoAlcaeuspunctoillius;
ille meoquis? IquisnisiCallimachus?, Propertius 2. 2. 6-7 incedituel lout digna
soror,I aut. .. Pallas,Martial 1. 62. 6 Penelopeuenit,abit Helene.

XVIII
There is some dispute as to where the quotation begins.~ia is given by some
editors to Cicero's discourse,by others to the tragedian's trimeters.
The words sit sane illud commentidumquo Priamusest conturbatus form a
logical unity which the addition of quiamatergrauidaparerese ardentem f acem
1
Snell himself saw the difficulty upon which Scheidweiler and Strzelecki
pounce; see Euripides'Alexandros,p. 54.
i C£ C.I.L. 11 583 (122 B.c.) .4 ioudiciumioudicatio
leitiJqueaestumatio,quei
quomqueioudiciumex h.l. erunt,eorumh.l. esto.
3 See W. Kroll, Glottam (1910-12), 8 f.

220
 ALEXANDER
uisdest in somnisHtcubaembarrasses if it does not destroy. It is therefore
tempting to make quia begin a syntactically independent quotation. But
such quotations usually have a logical unity; they do not beginwith adver-
bial clauseshanging in the air. In any caseit is difficult to imaginewhat could
have formed the substanceof the principal clauseomitted from the quotation
of the tragic narrative.
I do not think it has beennoticed that Cicero's discourse paraphrases to
some extent what he actually quotes: quo Priamusest conturbatus quo ~
facto. .. Priamus. .. mentis rMtu perculsus.It is more usual for Cicero to
introdua: a direct quotation with a paraphrase of what comes before (e.g.
Tusc. I . 4S, I . 69). The dialogue on divination was composed in obvious
hasteand we must regard the quiain question as Cicero' s own inelegant link
betweenhis allusion to the dream of Hecuba and his quotation of Ennius'
poetic account. Vahlcn's suggestion that Cicero omitted a word like mea
from the head of what he quotes is attractive.
There have beenmany guessesat the identity of the prologue speaker:
Diintzcr suggested Cassandra, Hartung V cnus, Cronert Hecuba, W.
Schmid1 some subordinate person. An interpretation by 0. Skutsch of fr.
:xxm would seem to me to give some support to Hartung's view.
In adapting Euripides' story of Hecuba's dream Ennius was forced
by the exigencies of his medium to treat the dream as a Roman prodi-
gium in need of procuratio.The phrases exsacrificabathostiis. .. paam
petms . .. tdidit. .. temperarettollert seem to have been borrowed from
the contemporary sacral language. There are signsthat the substance of the
Euripidean narrative has been altered out of respect for Roman religious
feeling.
Most other extant accounts of Hecuba's dream 3 make Hecuba take reli-
gious action.3 Certainly no Roman paterfamiliaswould have allowed a
woman to interfere in so important a matter4 and it is possible that Ennius
departed from the Euripidean account at this point.5
The fear which the Latin narrative attributes to Priam may also be an
Enniao addition. The Romans regarded all prodigiaas signsof divine dis-

1
GeschidttedergriechischmLiteraturI iii, p. 475 n. 11.
a Hygi.nus, Fah.91, Dictys 3 .26, Schol Eur. Andr. 293, Schol A, Hom. II.
3. 325, Tzetzes, Anteh. 43; aliterLycophron 224, Apollodorus 3. 12. s.2, Ovid,
Epist. 16.47 ff.
3 Cf. Aeschylus, Pers. 176, Euripides, Htk. 87, I. T. 42.
4 See Snell, Euripidu' Alexandros,p. S9 n. 3.
s Cf. Terence's alteration to Apollodorus's narrative at Phorm.91; according
to Donatus: Apollodorustonsorem ipsum nuntiumfacit, qui dia# se nuper~llae
comamob luctum ahstulisse,quod scilicetmutASSeTermtium, ne extemis moribus
sptd4roremRomanumofftnderet.
221
 COMMENTARY

pleasure and fear was their automatic response.1 For the persons of Ionian
epic and Attic tragedy a npascould presage good as readily as evil. The
words of Clytemnestra at Sophocles, El. 644 ff. (esp. 646-7 el l,liv 1TEftlVEV
toeAa ... el 6' !x8pa) arc typical. Cicero adapted II. 2.320-1 f\µets s•
maares8avµ<X30µEV olov rnix&ri-j ~ ow6E1va fflAWf)<X 8e(;.')y
elafiA8 1

lt<CXT6µ(3as as nosautemTIMIDI stantesmirabilemonstrumI uidimusin mtdiis

diuomuersarieraris (Diu. 2.64), thus importing a quite un-Homcric but
thoroughly Roman attitude to the prodigy.
To judge by Andr.293 ff. and Tr. 919 ff., Euripides very likely made
Apollo's order one to kill the child.1 Ennius may have usedthe legal
circumlocution tolleretemperarerather than a word like necare(c£ Plautus,
True. 399) out of euphemism or with contemporary Roman practice in
regard to unwanted children in mind.3
Hyginus, Fab.91, Dictys 3 .26, Schol Eur. Andr.293, Schol. A, Hom. II.
3. 325 and Tzctzcs,Anteh.43 makeHccuba herselfconsult µavnis/coniectores
about the dream, Apollodorus 3 . 12. 5. 2 makes Priam consult Acsacus,
while Mythogr. Vatic. 2. 197 makes Priam interpret the dream himself;
according to the Latin trimetcrs, on the other hand,Priam goes direct to
Apollo. Robcrt4 and SnellSargue that the mythographers, for all their
divergencesamong themselves,reflect Euripides' account of events more
truly. In the view of Robert and Snellthe Greeksalwaysconsultedµavrus in
such circumstances;they never approached the gods directly.
I think it can be demonstrated with some probability that Ennius here
followed Euripidesquite faitbfully.Apollo was himselfa µavns, the µavns
ofZcus(c£ Acschylus,Eum.17-19nxY1'}s 6ev1v2EvslveeovKT{a~cppwaj
t3E1map-rov T6v6eµav-nvw8p6vo1s· I A1c,s1T()Offl'TT'IS s·laTl /\~{~
1TCXTp6s) and at the moment of prophecy no clear distinctionwas made be-
tween the God and his human medium (c£ Cicero, Diu. 1 .67 deusinclusus
corporehumanoiam, non Cassandraloquitur).Where the Delphic oracle was
concernedthe Greekpoets spoke regularlyof the god as the giver of answers
(e.g.Homer, Od. 8. 79-81, Pindar,01. 7. 32-4, Aeschylus,Prom.655~. Cho.
269-90.Sophocles,0. T. 69-150, 787-93, 0.K. 84-105, Euripides,Hel. 148-
50, I. T. 82-94, 972-8, Ion 66-75, Med. 667--81, Phoin.13-20). Prose writers

1
Sec R. Bloch, Lts prodigesdansl'antiquitlclassique
(Paris, 1963), pp. 13 ff.,
84 ff.
a However parallel stories in Attic drama sometimes vary for no apparent
reason; contrast Sophocles, 0. T. 1174 and Euripides, Phoin.25 on the baby
Oedipus.
3 Cf. Terence's departure from Menander's text at Ad. 275; according to
Donatus: Menandermori ilium uoluisse.fingit, Tertntiusprofagert.
• Diegriech.Htldmsagem ii, p. 980.
s Euripides•Alexandros,p. S9 n. 3.
222
 ALEXANDER

on the other hand represented the answer as coming from the Pythian
priestess (e.g. Herodotus passim;cf. Cicero, Diu. 1. 81). The only references
to the medium in Attic tragedy are at Aeschylus, Bum. 1 ff., Sophocles, 0. T.
7n ff. and Euripides, Ion 91-2. We may believe that Euripides made Priam
(or, rather, Hecuba; see above, p. 221) consult an Asiatic oracle of Apollo
and that Ennius followed him in speakingof the god as the µavns; the
mythographersmade various guessesas to who the human medium was.1

So-I parere1e ardentem &cem I aia est in somnis: Buecheler' s parere

(ex) se as.siroilatesEnnius' construction to the form normal in archaic and
classicalLatin.2 However Ovid's imitation at Epist. 17 .237-8,fax quoqueme
tmet quamsepeptrisstcruentamI antediempartusesttuauisaparens,justifies the
transmitted text; cf. also Plautus, Cure. .200-1 hacnode in somnisuisussum
uiderierIproculsedertlongea meAesculapium,Gellius Is.22 . 8 uisi,msibiesseait
in quieteceruam. .. ad se reuerti.Ennius cowiated the two logical construc-
tions partrese. .. faam uiditand parert. .. facemuisaest.3

51 quo &cto: picked up at v. SS by tum, at v. 58 by ibi. Despite the ab-

sence of this phrase in particular from the rest of republican drama and the
rareness in general of the ablative absolute and the conjunctive relative
pronoun one should not tinker with the transmitted text.

52-3 ... somnio mentiametu Iperculms: Lambinus regularised mentis

to mentembut such genitives are common in republican drama (cf. Plautus,
Epid. 138 desipiebam
mentis,Trin. 454 satintu's sanusmentisaut animitui, Aul.
105 disaucioranimi,Terence, Hee. 121 animiut incertusforet).

53 caris sumptus nupirantibus: curausually occurs in the singular in

both tragedy (s times) and comedy (52 ). There is no other certain example
of the plural in tragedy, but in elevated parts of comedy there are five
(Plautus, RJul.221, Terence, And,. 26o, Hee. 230, 817, Turpilius 52).
For curissumptuscf. Terence, Plwrm.340 illeet curaet sumptuabsumitur.But
sumptushere probably stands for consumptus. For the metaphor cf. Homer, n.
6.202 ov 6vµov Kari&>v (Cicero, Tusc.3 .63 ipsesuumco, edens),(Theo-
critus] 30.21 6 1r68os Kai TOV faoo µVEAovta6(e1, Ennius, Ann. 335-6
curamue. .. quaenuncte coquit,Plautus, Epid. 320 exspectandoexedormiseratque

• For gods speaking directly to inquirers in Latin poetry cf. Plautus, Men.
84<>-1,Virgil, Aen. 3. 84--99, 7. 81-101, Seneca, Thy. 679 ff. For the human
medium cf. Seneca,Oed.223 ff.
2
E.g. Plautus, Mil. 389, Cicero, Diu. 1.49.56, 57 et al.
3 Cf. Trag. 329 (tum scirt oportetsibiparatampestemut partidpetpareni=
paratampestem+paratumpt.stemut participet).
223
 COMMENTARY

exenteror,True.593 qui ipsusse comest,Luactiw 3 .993-4 exest anxius angor

I aut. .. sdndunt.. . curae,Catullw 66 . 2 3 exeditcura medullas,Livy 40. 54. 1
maeroreconsumptus . .. deassit.
Suspirantibusis perhaps better taken as a bold heighteningof thepersonifi-
cation (c£ Sophocles, Ai.9571,1cnvoµa,o1s &xeow) rather thanas an instance
of the 'tropw quo id quod efticit ex co quod efficitur ostendimw '. 1 The effect
is much more bizarre than that of any of the adjectives and participles
defining abstracts listedabove on v. 33. Closely parallel is the humanist
reading of Luactiw 5. 747 aepitansdtntibusalgor.

5-4exsacrificabat hostiil balantibus:for the offering of sacrificeafter

dreams, visions, omens, portents, etc. c£ Aeschylus, Pers. 201 ff:, Choe.
523 f£, Sophocles, El. 405 ff:,Menander, Dysk.417 ff:,Plautw, Amph. 1126--
7 iube uasapura actutumadornari mihi, I ut Iouis supremimultis hostiispaam
expetam,Cure.270-3 paam ab AesculapioI petas,nefortt tibi eueniatmagnum
malum,I quodin quietetibiportentumst . .. quoeres ,,u,lt uort4t tibi, Virgil, Aen.
3. 176 ff:, 4. 56 f£, 5. 743 f£, 8. 544 ff:, Tibullw 1. 5. 13.
The compound exsacrificare occurs nowhere else in recorded Latin.
Vahlen interprets: '!Kevea8cn uelut ayosHerodotw {6.91) aliique'. But
there is no question of pollution here. We may regard the form as simply
a poetic coinage; exsaaificabat forms a trio with the unwual, if not unique,
compounds edoceret and edidit;c£ above on v. 20.
The simple saaificareoccurs 21 times in comedy. It is absent from Caesar
and Cicero (except for Nat. deor.2.67) but common in Livy. The word
probably belonged to the sacral language in the first century, and was
avoided by the purists bccawe of its similarityto certain poetical formations
with -ficus,-.ficareand -.ficabilis.
For the narrative imperfect standing within a succession of aorist and
historic present tenses c£ Naeviw, Carm.fr. 3 postquamauemaspexitin templo
Anchisa,Isacrain mensaPenatiumordineponuntur;I immolabataureamuictimam
pulchram,fr. 24 isquesusumadcaelumsustulitsuasfrcsammullwt gratulabatur
diuis, Virgil, Aen. 3. 32 f£ insequor ... et... sequitur... Nymphas uenerabar
agrestisI Gradiuumque patrem,Geticisquipraesidetaruis,I ritesecundarent uisus
omenqueleuarent.I ...auditur. ..• The contexts are allliturgical but what signi-
ficance thishas,if any, I cannot say. See Mariotti, Btllum Poenicum, pp. 75 f.
Hostiawas a word of the sacral language, 'victim offered to appease the
wrath of the gods•.
Balanswas probably a poeticism (c£ Enniw, Ann. 186, Lucretiw 2.369,
6. 1132, Virgil, Georg.1.272, Acn. 7. 538) modelled partly on such sacral

1
8. 6. 27; cf. Rhct. inc. Her.4. 32. For a discussion of this pheno-
Q...uintilian
menon in classical poetry see Norden, Aen. Vl 1, pp. 138 f.

224
 ALEXANDER

terms as (hostia)bidens,lactens,etc., and partly on the GreekµT)KQS(of sheep

Euripides, Kykl. 189; of goats Homer, n. II. 383, Sophocles, &. 468).
Forthethree-syllablerhymesuspirANTIBVB .. . bauNTIBVSiniambictri-
meten c£ Trag. inc. 133-5 caelum nitescere,arboresftondEBCERE, uites I
I
l«tificaepampinispubESCERE, ramibacarum ubertmeincuruESCERE,Plautus,
Pseud.67-8, 805-6. It occun sporadically in the trimeters of Attic tragedy:
e.g. Sophocles, Ai. 807-8, 1085-6, 0. T. 110-11, Euripides, Hipp. 727-8,
Herakleidai541-2, Med. 408-9, Agathon, &. 11.

55 tum coniectaram poatulat: for coniectura of the interpretation of

dreams c£ Plautus, Cure. 246, Rud.612, 771; for somniumconicertPlautus,
Cure.253 (~ Euripides, I. T. 55TOwap6' w6eavµ~ooT6&). The usage
seems to have arisen from the use of lots in dream interpretation (sortes
).1
conicert
Abstract formations with -ura are not common in republican drama.
Ploen counts only 4 in tragedy, 20 in comedy.

paceinpetens: for pax, 'grace, favour', c£ Ennius, Trag. 38o, Plautus,

Amph. 32, 1127, Cure.270, Mere.678; thisIl'\eaoiog was probably still alive
only in liturgicalphrases like pacempetoet al. No exact equivalent needhave
stood in the prologue of the •AAt~cxvSpos; such dreams were not an auto-
matic signof divine displeasure for the Greek mind. C£ Cicero's version of
Homer, n. 2. 305-6 f}µels6' aµq,i mpi KPfi"'lVlepc>VSK<XT<X j3ooµovsj
ru11ro-aas
Fp6oµev 6:ecxvaToaoij bc<XToµ~: nos drcum laticesgelidos
fumantibusarisI aurigerisPiuomplacantesnumi;i3tauris(Diu. 2. 63).

56 ut se edoceret obsec:rans: the compound edocereoccurs twice in

tragedy, the simple docereonce; in comedy the ratio is 6:26.

57 quo sese uertant tantae sortes somnium: for the sequence of tenses
-postulat (historic present) .. . edoceret
... uertant-cf. Plautus, Pseud.795-6
quin ob eamrem Orcusredpereadse huncnoluitI ut essethie qui mortuiscenam
coquat.
With quo seseuertantsortesEnnius is revivifying the faded metaphor of
coniectura.
The lots are shaken in the turning urn. C£ Horace, Carm.2. 3 . 26-8
omniumI uersatururnaseriusociusI sorsexitura,Virgil, Aen. 3. 375-6 sicfata
deumrex I sortituruoluitqueuices,Seneca, Tro. 974 uersatadominosurnacaptiuis
dedit.
For the enallage in tantaesortessomnium,c£ Ennius, Trag. 123 ignotus
iuuenumcoetus,401 deosaeuiintegros,Pacuvius, Trag.20 meapropagessanguinis,
1
C£ G. P. Shipp, CR u (1937), 209 ff.
IS 225 JTO
 COMMENTARY

Accius, Trag. s66-7horriferAquilonisstridor,Trag. inc. ap. Cic. Tusc. 1.48

pallida leti. .. fuca.Republican tragedy normally keeps epithets with the
nouns to which they belong in sense.
For tantus, 'so extraordinary', cf. Plautus, Amph. 1036 nequeegounupuun
tantamira~ uidisseanseo, 1057 ita tanuimira in oedibussunt (aaa,Accius,
'Praet.31 di rem tantamhaut temereinprouisoofferunt.

58--9ibi ex orado uoce diuinaedidit I Apollo: cf. Pindar, 01. 7. 32-4

Tc';> 6 xpvaoK6µcxsevloo&os
IJEV l~ a6V'TOV vaoov1TAOOV Ielm, Euripides,
I. T. 976-7 wreOOevcxv611v Tp{iroSoslie xpvaoO ACXl(OOV, I <J>otj36s µ•
hreµ'f'E&Opo... , Plautus, Men. 840 ecceApollomihi ex oracloimperat.
For the phrase ex oracloederecf. Cicero, Fin. s.79, Ad Brut. 1 • 2 a. 3.
Edere,'announce publicly', belonged properly to the officiallanguage.
For edereof the pronouncementsof the keepersof the Sibylline oraclescf.
Livy 21. 62. 8, 22. 10. 10 et al.
For the pleonasmof uocediuinaediditApollocf. Homer, Od. 8. 79 Xf>E{CA>v
µu6t')aaro <J>ot~ 'Air6Uoov, Pindar, bth. 8.45 ~ qxnoKpov(SoosI
wmo1aa8ea, Aeschylus,Ag. 20selm cpCA>VOOV, Choe.279mcpcxvaKCA>V elm,
Sophocles,Ai. 7S7 ~ 1¥1 Afyoov,Ant. 227 'fNXT\ yap1106cxiroUcx µ01
µv6ovµwl), Catullus, 64. 321 taliadiuinofuderuntcarmine f ata, Cicero, Carm.
fr. 22.22 tum Calchashate estfidenti uocelocutus(Diu. 2.63 = Homer, n.
2.322).

59-60 primm Priamo qui foret I postilla natus: Joremas the perfect
participle'sauxiliaryoccurson only nine other occasionsin republicandrama
(Pacuvius,Trag.300, Plautus,Amph.21, Cure.449, Mil. 1083, Most.494,800,
Poen.262, Rud.218, Caecilius,Com.4S); thricethe participleis natus(contrast
Plautus, Cas. 46, Persa634, Terence, Hee.279). The contexts of occurrence
in comedy have a tone either of legal formality or of paratragedy.

puerum . .. temperaret tollere: for temperareand infinitive the only

parallelsin drama are Plautus, Poen.22, 33, 1036; in all three passagesthe
speak.erappears to be aping the language of the magistrates.
For the phrasepuerum. .. tollerecf. Trag. inc. 8s egocumgenuitum morituros
sciuiet huicrei sustuli,Plautus, Amph. 501 quoderit natumtollito, True. 399,
Terence, Andr.219, 401,464, Haut. 626-7, Cicero, Alt. II .9.3, Horace, Sat.
2. s.46 et al. The phrase reflectsa Roman family ritual wherein the pater-
familiaspicked up from the hearth the child he intended to rear as his own.1
It is often saidithat Attic drama usedthe verb avoopeta6a1of a similarritual.
1
Cf. Augustine,Ciu. 4. 11.
:aCf. E. Samter, Familienfesteder Griechen und Romer (Berlin, 1901),
pp. 59ff., 0. Schrader,Spraclwergleichungund Qena,1907), p. 345.
Urgeschichte8
226
 ALEXANDER

This appears to be untrue. In contexts parallel with the Roman ones the verb
T~IV is used{e.g. Menander, Perik.370, 382); avaapeio6<Xl refers regularly
to the taking up to rear of another person's child (e.g. Aristophanes, Neph.
531, Menander, Sam. 140, 159, 196, Epit. 74, 87, 154, Perik. 14). 1 Plutarch
uses avcnpeio6cnto tum the tollereof his Latin source when speaking of the
children Antony had by Cleopatra (Ant. 36) but this tellsus nothing about
fifth and fourth century B.c. Attic usage. There is no direct evidence that
Athenian families employed a ritual like the Roman. On the fifth day after
birth the Athenian child whom the father had decidedto have reared was
carried ceremonially around the hearth.i One can only guessat the signifi-
cance attached to thetwo ritualsin Athenian and Roman society respectively
but it must have been roughly similar. Athenian society is no more likely
than Roman to have possessedboth.

61 eam eue exitium Troiae, pestem Pergamo: while Attic drama has
regularly Texnipycxµcx(except in two lyric passages,Euripides, I.A. 773, Tr.
1065, where the city as a whole is clearly meant), Roman has Pergamum
(except at Livius, Trag. 2, Ennius, Trag. 73, Seneca, Ag. 206, Troad.472,
889). Pergamum indicated the arx, Pergamaperhaps the walls surrounding the
arx, Troiathe urbsagrique.For Troia(Tpofcx)and Pergamum(Tlipycxµcx) in
conjunction c£ Sophocles, Ph. 353, Euripides, Amlr.292-3, Plautus, Bacch.
1053, Virgil, Aen. 2.554-6, 3.349-51.
"Essestands for fore. The present tense is frequently usedin Attic drama in
prophecies (e.g. Aeschylus, Ag. 126, Prom. 848, Euripides, Hipp. 47,
Aristophanes,Hipp. 127) as well as in descriptions of visions. For theindirect
reporting of such prophecies c£ Aeschylus, Choe.1030-2 TOVm.,66µcnrr1v
I
A~{cxv,XPiiacnrr·~µol irpa~cnrrcxµevTCXVT' I
mbs cxhfcxsKCXICTlS elven,
Apollodorus 3 . 12. s.3 ovras(sc.AiaCXKQS) el1m>v Ti)s,rcrrpf6osyevioeoo
TOV ircxt6cx arrooAEacxv a<eetvcnTo~ tdAfve.3
For eumesseexitium Troiaec£ Plautus, Bacch.1054 sdui tgo iamduaumJore
meexitiumPergamo, Virgil,Eel.3. 101, Aen. 2. I9G-I, Horace, Carm.I. 15 .21.
Thepredicative dative is normal (e.g. Plautus, Bacch.947,953, Cicero, Mur.
56 et al.). Ennius was probably seekingconcinnity with thefollowing phrase.
Exitium is not elsewhere used of single individuals in drama or in Cicero;
pestis,on the other hand, is commonly so used (e.g. Plautus, Pseud.20.t,
Terence, Ad. 189).

1
Tollerewas used in this sensealso by the Roman poets: e.g. Plautus, Cist.
124, 167, 184 et al.
3
C£ Plato, Theait. 16oE. Plautus appears to have generalised a detailed
reference to this ritual at True.424.
3 For the phenomenon in classicalLatin seeE. WJStrand,Horace'
s NinthEpode
(Goteborg, 1958}, pp. 29 ff., 49 ff.
227
 COMMENTARY

XIX
For the context of this fragment sec above, p. 203.
Welcker thought that Priarn was the speaker. The Strasbourg papyrus
revealed that the Euripidean Priarn was present at the games. Cronert
accordingly made Hecuba the speaker.

62-3 ab ludis . .. nuntium: c£ Plautus, Bacch.197 quodab illoc attigisset

nuntius,528, Cicero, Alt. 13 .47. 1 (alluding to a passage of tragedy).

62 animus atque aures auent: a periphrasis for audireaueo.C£ Sopho-

1
cles,Ai. 686Tovµov c!>vt~ dap, Euripides, Hipp. 173 T{1fOTfoTl µa8etv
lpaT<Xl 'fNXT),912 "" yap,ro8ovaa irma Kap6{a KAVElV,Hyps. I . 3 •I 5
Ta& µ01Ta& 8vµbs 16etvITTCXl, Plautus, Cist. SS4animusaudireexpetit,
Catullus 46. 7 iam mensprattrepidansau.etuagari,Cicero, Phil. 5 . 13 auet
animusapudconsiliumilludpro reo dicere(a passage of excited rhetoric).
Auere does not occur elsewhere in republican drama. Cicero alone of
classical writers uses it frequently.

63 auide expectantes: the adverb is probably to be construed with

expectantesrather than auent.Rhetorical pause usually coincides with metrical
pause. Where a verb is accompanied by an etymologically related adverb the
adverb usually precedes: e.g. Ennius, Trag.259 cupientercupit,Plautus, Capt.
250 memoriter meminisse,Cas.267 cupidecupis,Cure.535propereproperas,688
properepropera, Most.985 meeius• .. miseremiseret, Persa427 ualideualet,Poen.
6o6 sapientersapit,921 iterumiterem,Pseud.358c."Ursimcu"am,Rud. 1265 iterum
•.. itera,True.61 tempestiuotemperent,354 nitidenitet,Pomponius, Atell. 109
memoremeministi.However the adverb does occasionally follow: e.g. Plautus,
Amph. 417 memoratmemoriter,Men. 146 ecquidadsimulosimiliter,151 caueo
cautius,Poen.562 meministismemoriter,Afranius, Com. tog. 294 cu"e cursim,
365 meminimemoriter.
XX
For the Greek etymology sec Apollodorus 3 . 12. 5 . 4 hpeq,a, ovoµaaas
TJap,v• yev6µevos & veav{aKoS KooiroAAoov61acptpc..:,v
K<XAAE1
TE KOO
~00µ13 •AM~av6posirpoaoovoµaaet,, A1JaTQS
cxv815 &µw6µevosKai TOts
iro1µv{o1s&Ae~aas.
For the giving of a name according to the character or circumstances of
the person to bear it c£ Aeschylus, Prom. 848 ££,Euripides, Hel. 8 ££,Ion
661 ££,I. T. 32 £, Plautus, Capt. 285 ££,Pseud.653 ££,Stich. 174 f£ Plautus
was obviously not troubled by the sort of objection which Varro has to
Ennius' method of adaptation; the three passages I have quoted could all be
hisown additions to the substance of hisoriginals. The Attic tragedians were
228
 ALEXANDER
in the same theoretical difficulty with the portrayal of barbarians; they
ignored it except when it was dramatically convenient to draw attention
to linguisticdifferences.1
XXI
65 purus putus: c£ Plautus,Pseud.988-9 oh I Polymachaeroplagides Ipurus
putus est ipsus, 1200 purusputus hie sucophantast,Varro, Men. 432 (Nonius,
p. 27 .23) Chrysosandalos locatsibiamiculamde lacteet ceraTarentinaquamapes
Milesiaecoegerintex omnibusjloribuslibantes,sine osseet neruis,sinepelle,sine
pilispuramputamproceramcandidamtener:am f ormosam.

XXII
W dcker referred this fragment to an account of Deiphobus addressingthe
victorious Alexander; Ribbeck to one of how Alexander's favourite steer
was taken away from him.

66 hominem appellat: 'cum appellat'; the usage is common in repub-

lican drama; c£ Terence, Amir. 744 and Donatus ad loc.;J. Kohm,
Altlateinische
Forschungen(Leipzig, 190s), p. 89 n. 1, counts 120 instancesin
Plautus, 34 in Terence.

stolide: 14 times in republican drama as against stultus102, absent from

Caesar and Cicero (except Top. S9), fairly common in the historians; the
word probably had a high-falutin tone even in Ennius' day. Plocn counts
40 formations with -idusin comedy and 27 in tragedy, 10 of whichdo not
appear at all in comedy.
XXIII
Wdcker, Hartung and Ribbeck referred thisfragment to•Epoos/Amor,C. 0.
Mueller to NiK11/Victoria. The latter was commonly represented in ancient
art and literature as winged and bearinga crown and fillets, c£ Euripides,
I. T. 1497--9 er,µfycx aeµVfl NiK11 Tov !µov I ~{OTOvKcx-rixo1s I Kcxlµ'I'\
:A.,'tyo1s<TTE~v<1cx, Sallust ap. Macrob. Sat. 3. 13. 8 praetereatum
sedentiin transennademissumVictoriaesimulacrumcum machinatostrepitutoni-
truum coronamei imponebat,tum uenienti ture quasideo supplicabatur, Livy
7. 12. 13 unicumducemquinihilagentisibile caelodeuolaturam
in sinumuictoriam
censeat.
W dcker suggestedtwo possiblecontexts :2 a descriptionof thegamesby a
messengerand Cassandra'saccount of her vision of the contest of the god-
desses.W dcker' s first idea seems to me to have been the better one. We
1
E.g. Aeschylus, Ag. 1061, Hik. us. C£ Pacuvius, Trag.364.
:a Die griech.Trag.pp. 467, 1s82.
 COMMENTARY

now know (sec fr. 18 Snell) that in Euripides' play a messenger described
Paris being crowned by human hands as victor in the contest. It is unlikdy
that any ancient tragedian would have had a personagedescribea visible
intervention by a supernatural beingin such events. I would suggest that the
Latin trimeter comes from the messenger's report of a prayer uttered by
Alexander to his patroness Venus 1 asking for victory in the contest; c£
Odysseus' prayer to Athena during Homer's foot-race at n. 23. 768 ff.,
Cloanthus' prayer to the •di quibus impcrium est pclagi' during Virgil's
boat-race at Aen. s .235 ff.
In the view of 0. Skutsch 2 the verse is best interpreted as the opening of a
prologue spoken by Victoria)This seems to me possible on the further sup-
position that Ennius replaced a prologue by Aphrodite in his original with
one by the abstract Victoria.It is difficult to imagine NfKT1, who was associ-
ated in the Athenian tnind with Athena, opening a Euripidcan play about
the periodof Troy's prosperity. At Rome on the other hand Victoriawas
associated with Venus rather than with Minerva.

67cum corona et taeniil: c£ Thucydides 4. 121 ••• S,,µoaf(itµa,)(pva4">

<n"E~'t) . .. l6f(it& hcnvfow -reKai irpoo,'ipxoVTO c:xrmpa6A1')Tij,
Aristophanes, &tr. 392-5 Kai I Tfisaf\stopTfisa~foos I irafacxvra Kai
O'K~av1j1CT}aavTaTa1v1ova6cn, schol.adloc.Tcnv1ova6cn • mrrlTOV
a-rE<p<XVOVa6cn • Tcnv{ayap To {xxµµaTOv<n"Eq,av<)V, Plato, Sympos. 212B
~~Cl)~ cnrrov KlTrOV"l'iTlVl ~'tl 6aaet Kal toov,Kai Tcnvfas
~OVTa ml TfisKEq>CXAT}S ,rawTI'OAAQS, Virgil, Aen. s .269 (describing the
victon in the boat race) puniceisibanteuinctitemporataenis, Scrvius ad loc.
PVNICBIS TAENIS uittisroseis;et significatlemniscatas quaesuntdeftondibus
coronas
f asciis,et sicut Varrodicit,magnilumorissunt.•taenis'autemmodo
et discoloribus
uittis; aliasuittarumextremitatesdicitut (7. 3s2) 'fit longaetaeniauittae'.
Corona,•rncpavos ', is common in republican drama; taenia,'Tcnv{a' rare.
Coronamust have been a borrowing by the sacral language from South
Italian Greek; K~VT) docsnot appear in extant Attic literature in the mean-
ing •corona,rncpavos '. Taeniaalsoprobably came from this source rather
than the Attic plays adapted by the poets. The common language contained a
large number of obvious borrowings from Greek. The poets restricted these
in comedy largely to the speech of slaves and pcnons of low degree and
usedthem even lessin tragedy. Ennius hasonly barbaricus,corona,drachuma,
exanclare,machaera,
ostreum,philosophari,
salus,stola,taenia,thesaurus.
1
Pompey's temple to Venus uictrix was called by Tiro aedes Vidoriae
(Gclliw 10. 1.7).
a Expressed in private conversation, 11 October 1965. Sec now HSCPh LXXI
(1967), 126 f.
3 Plautw refers to such a tragic prologue at Amph. 41-2.

230
 ALEXANDER

XXIV
For the context of thisfragment see above, p. 203.

68 aduentant: only here in tragedy and five times in comedy; aduenire on

the other hand is extrcmdy common in both genres. Tragedy frequently
uses an intensive form with no perceptible difference of meaning from the
simple form normally wed in comedy: e.g. abnutare(Ennius 290 ), aJitare
(Ennius 390), concursare {Accius 398), incrtpit4re{Trag. inc. 234), initare
(Pacuvius 1), pmogiwe (Pacuvius 315), proiectare{Ennius 204), rapt4re
{Ennius 79), subiectare (Pacuvius 334), succussare (Accius 568), uolitare(Trag.
inc. 183). Comedy haslarge nwnbcrs of similar verbs banned from the Latin
of theclassical purists; it is erroneous to label themall vulgarisms; many must
have been poetic formations.

paupertu quorum obscarat nomina: a periphrasis for pauperesatque

~
c£ v. 309 diquibusestpotestasmotussuperumatqueinferum di superi
ignobile.s;
atqueinferi.
For the sentiment c£ Euripides, El. 37-8, Phoin.442-3, fr. 326.3-4, fr.
362. II ff.
XXV
For the context of thisfragment see above, p. 203. For the idea of a mutilated
spirit c£ Homer, OJ. 11 .40-1, Seneca, Oed. 617 ff. For theincident described
cf. Homer, II. 22.463-4, Sophocles, Ai. 1030-1, Euripides, Amir. 107-8,
Ennius, Trag.78-9, Virgil, Aen. 1.483, 2.272-3.
It is difficult to fit the words transmitted in Macrobius' codicesinto any 0.1.
the metrical patterns found in republican drama and yet they are not
indubitably corrupt. The question quid ita cum tuo laceratocorporemiserhas
seemed to many to lacka verb but an actor could well have interpreted ita
with a gesture of the hand; c£ Plautus, RJUJ. 809 adsistiteambo.sic. audite
nunciam,Ennius, Trag.293 nolitehospites.adme aJire.ilicoistic.
Trochaic and bacchiac scansion have been tried. Vahlen 1 produced a
trochaic system: o lux Troiae,germane I
Hector, quiditacumtuo lacerato I
corpore
miser(es) aut quite sic respectantibus I tractauerenobis?;Leo a slightly more
2

convincing system: o lux Troiae,germaneHector,I quidita cum tuo laceratoI

corpore I I
(abiectus's)miserautqui te sic respectantibustradauerenobis?Mariotti3
took o lux Troiae,germaneHectoras an :anapaestic dimetcr and the rest as
bacchiac verse: quid ita cum tuo laaratocorporeI miser( atles)aut qui te sic

1
Hmrres :m (1877), 400, xv (1880), 262.
2
Dt Trag.Rom.p. 20 ( = Ausg. kl. Sehr.1 210). Cf. Dieplaut. Cant.pp. 29 ff.
3 Uzioni SU Ennio, p. 133.
231
 COMMENTARY

I tractauerenobis?1 However of theten alleged bacchiac metra

resptctantibus
only one is purdy so and wherever a run of indisputable bacchiacsoccurs
in contemporary comedy pure metra predominate.i I leave the question
open.

'9 o lax Troiae: 'o hdp of Troy'; cf. Homer, Od. 16.23 T11M1,1ax,e
yAVKEpOV cpaos, Sophocles,El. 1354-5 oocpiATarovq,C>s, ooµ6vosO"CJTT\P
661,100vI •Ay~µvovos, Euripides, Hek. 841 oo&airOT', ooµfytaTOv
·ru11a1v cpaos. Heraklts531 00cpiATar'avSpC>V, 00cpaos IJOAOOV1Tarp{.
Trag. inc. ap. Page, G.LP. 162 f£ •EKTOp ... 1TaTPGtKai fµol µfya q,C>s,
Accius, Trag.163 a: (He)ctorlux Dardaniae,Plautus, Mil. 1344 quaeres? quid
uideo?lux salue,Stich.618 o lux oppidi.

70 quid ita cum too lacerato corpore miser: for theadnominal use of
cum cf. Ennius, Trag. 235 quiquetuo cum luminemareterramcaelumcontines,
Ann. 54 Tiberinetuo cumjlumine sancto,Plautus, Cure. 191 tun etiam cum
noctuinisoculis'odium' me uocas?,Pseud.')67 heustu qui cum hirquinabarba
astas.

71 aut qai te sic respectantibus tractauere nobis: for autintroducing a

question which is not an alternative to the previous one cf. Plautus, Amph.
409 quid igituregodubitoaut qur non intro eo in nostramdomum?, Poen. 994
quoiattsestisaut quoex oppido?
With the phrase respectantibus . .. nobisis to be contrasted Plautus, Amph.
998 hiedeludetur . .. uobisinspectantibusand Cicero, Manil. 33 cumpropeinspec-
tantibusnobisclassisea. .. a praedonibuscaptaatquedepress a est. The absolute
phrase with inspectanteor inspectantibus is normal in Caesar and Cicero.
For tractart,'drag up and down, round and about', cf. Ennius, Ann. 137
per aequoracampi,Lucretius 3 . 88~ malismorsuque
tractatus f erarumI tractari.
The stylisticlevd of the termination in-ereis obscure. D. W. Pye3counted
12 instancesin tragedy against 4 of-irunt, 52 in Plautine comedy against 100,
28 in Terentian comedy against 25. Ennius' tractauere differs from the great
majority of comic forms in -erein occurring before a word beginning with a
consonant. Havet4 thought that on the few occasionswhere this happens in
comedy the poet had some special rhythmical or stylistic motive.
1
Strzelecki, in Tragica
1, 61 f., approves Mariotti's colometry but supplies ess
instead of ades.
a Cf. A. Spengel, Rtformvorschl. pp. 26o ff.
3 Trans.Phil. Soc. 1963, 1 ff. 4 RPh XXXI (1907), 230 ff.

232
 ALEXANDER

XXVI
Cassandradescribesa vision with nammaximosaltusuperauitgrauidusarmatis
equusand expresses a foreboding about the future with qui suopartu ardua
perdatPergama.Modern editors follow Vossiusin writing superabit.There is
no good syntacticalreason for abandoning the transmitted text. Perdatis a
prospectivesubjunctive.1 For the sequencec£ Plautus, Epid. 285 repperilulec
te qui abscedat suspido, Cicero, S. Rose.32 etiamneadsubselliacumferro atque
telisuenistisut hie aut iuguletisaut condemnetis
?
The image of the pregnant mare leaping appean first in extant Greek
literature at Aeschylus,Ag. 824-6: 1TOA1V 6111µa6wev"ApyelovSaKos,I
l,r,rov veoaac,s,
&am611cpopos ~. I11Tl611µ• 6povacxs&µ,t maa&>v
6vmv; c£ Euripides, T,. 9-12 6 yap Tlapvaa1osI <l><A>J<EVS
"EfflloS,µ11xa-
valm TT<XUaSos I fylruµov• l,r,rov -mJX~V ~apµoacxs, I ,rvpyoov
hreµ'f'EYMoS 6Mep1ov13Ph"cxs, Lycophron 342, Antiphilus, A.P. 9.156,
Tryphiodorus 389. Something of its clarity is allowed to fade by Ennius'
masculineadjectivegrauidus.
The words which Macrobius quotes show no sign of corruption and yet
they are difficult to reduce to metrical order. Nam maxumosaltu superauit
grauidusarmatisequuscan be treated as an iambic tetrameter acatalecticwith-
out diaeresis.Such verses are common in comedy both in company with
other types of verse (breaking into anapaestsat Plautus, Amph. 1o61, into
trochees at Amph. 10"/I) and in stichic runs. They are also common in
tragedy (c£ vv. 138-9, 163-4, 171, 322-32). Vahlen treats qui suo partu
arduaperdatPergamaas trochaic despite the resulting dactylic word arduain
the third foot (seeabove on fr. xu).
72-3 nam maxumo salta 111perauit ... Pergama: Ptrgamahas to be
taken &iro Ko1voO with superauit;c£ Plautus, Persa524acsuopericlois emat
qui eam mercabitur.
Saltus, 'leap', occurs only here in republican drama; it is absent from
Caesar and Cicero (except for Cato 19) but frequent in dactylic verse, both
archaic(Ennius,Ann. 129, Lucretius5. 559, 1318) and classical{Virgil,Georg.
3 .141, Aen. 2. 565 et al.).
Superareis usedof physicalmovement only here, v.189 and Plautus, Stich.
365 (c£ 279) in republican drama; it is absent from Caesar and Cicero but
common in Virgil (Georg.3 .141, 3 .270 et al.) and Livy (6.2. II et al.).
73 qui 1uo partu ardaa perdatPergama: the plural Pergamaoccurs
elsewherein republicandramaonly at Livius, Trag.2. Here Ennius may have
in mind thewalls of the citadel; hencetheplural (aftermoenia).
1
Cf. Plautus, Asin. 29 die. •. quodte rogem,Catullus 64. 330,Virgil, Aen.1 . 287.
233
 COMMENTARY

Arduus is normally used of mountains, mountain paths, etc.; rardy of

walls, ramparts, etc. in either classicalprose (but c£ Varro, Rust. I. 14. 3,
Cicero, Rep. 2. 11, Tacitus, Hist. 3. 30) or classicalverse (but c£ Virgil, Aen.
12. 745, Ovid, Met. 3. 61 ).

ANDROMACHA

The title Andromacha is given to Ennius by Cicero, Nonius, 1 Macrobius and

Servius.2 The Danidine Servius gives the same title to Novius, a writer of
abulaeAtellanaewho sometimes parodied tragic themes (Georg.1 . .266).
Behind three ofNonius' citations (p. 76.1 andromaca, p. 504. 18 andromaca,
p. sIs.28 andromacha) appears to lie the archaic form Andromaca; behind four
(p. 292. 7 andromache malo tor, p. 402. 3 andromacahaec malo, p. 505. I2
andromace ethemapotide, p. 515. 13 andromache eimalo)a classicaltransliteration
of' Av6poµcxx11 cxtxµCXACA)'Tfs.3In the one case Ennius' own title would have
been given, in the other the title of hisGreek original. 4 Nothing in the pieces
quoted alongside these titles suggests that two distinct scripts were in the
possession of the grammarians who first quoted them.
The title Andromacha appears without an author's name at Cicero, Tusc.
3 .53, Ac. 2.20, Alt. 4. 15 .6, Varro, Ling. 7 .6, and Festus, p. 384.21. Com-
bination with Tusc. 3 . 44 and Orat. 93 makesit certain that at Tusc. 3. 53
Cicero means a play by Ennius. Since no Latin tragedy AnJromachaapart
from that by Ennius is known editors have interpreted the other passages
likewise.5 Ovid may refer to the play at Rem. 383.
1
All seven ofNonius' citations appear to come from the sources defined by
Lindsay as list 27 'Alph. Verb' and list 28 'Alph. Adverb'. At the end of the
article on summus (pp. 401.23 -402.5) comes the same passage which at
p. 515. 13 illustrates the adverb longinqueand quite certainly comes from list 28.
It must have been added to the summusarticle when the book De uariasignifica-
tione sermonumwas being organised alphabetically. The presence of Accius'
name at p. 515. 12 must be an error of the transmission.
:a Servius, Aen. 1 . 224 and Macrobius, Sat.6. s.10 appear to draw ultimately
on the same source; see CQ N.S. xv (1965), 135 ff.
3 The fint to glimpse the truth was C. L. Roth, who, in the edition of
Nonius which he published with F. D. Gerlach at Basic in 1842, suggested in
the apparatus at p. 292. 8 Andromachaaechmalotide. C£ Propcrtius 2. 20. 1-2
quidflesI anxiacaptiuatristiusAndromacha?Ribbeck's Andromaca aedmalotide(ed.
2, p. 23) and Wolffiin's Andromacaaecmaloto(ALL vm [1893], 234) confuse
things. 4 See Introduction, pp. 58 f.
5 Very little is preserved of the quotation at Fcstus, p. 384.21. It is odd that
(Enniu)s in Andromachashould have been written rather than idem in
Andromacha.Perhaps one should restore (Nouiu)s in Andromacha.
234
----·-.

ANDROMACHA
The personage who gave hername to the tragedy was the wife of Hector
and the action took place after the fall of Troy. Opinions have differed about
the scene. Columna set it in the camp of the Greek chieftains preparing to
return to their homes and made the ki11ing of Andromache' s son Astyanax
the central theme. Merccrus 1 interpreted Nonius as quoting from two
scripts, one an adaptation of Euripides' extant •Av6poµax11 with the title
Andromache-Molottus, the other an adaptation of the Tp~Ses. Diin~
rejected the idea of two scripts and by .adducing some dubious parallels
between the Ennian fragments and Seneca's Troadesmade the kj)]ing of
Polyxena as well as that of Astyanaxa theme of the play. W dcker3 imagined
one play with a title like Andromache in Molottisand set its scene at the coun
of Helenus, the brother of Hector, in Epirus. 4 Roth's elucidation of the text
ofNonius, p. 292.7 as EnniusAndromache «chmalotidiconfirmed Columna's
view. One might compare Euripides, Tr. 677-8 vav~Aovµm s· I
fy6> 11'p0S
·ruas· atxµCXA(A)TOS ~ 6oOAOV l'JYOV.The heroine of Euripides' •Av6po-
'10X11is frequently referred to as atxµ<XAOOTis or atxµ<XA(A)ToS (583, 871,
908, 932, 962, 1059, 1243) but thisis to distinguishherfrom Neoptolemus'
regular spouse Hermione; one could hardly imagine the word in a title of
the play. LadewigS suggested that Cicero' s argument at Tusc. 3. 53 only
has point if Andromache' s lament referred to recent events. Ribbeck 6
settled the issue by diroctingattention to Cicero, Att. 4. 15 .6. The central
event of Ennius' play must have beenthe kiJliog of As7anax.7 Proper inter-
pretation of Cicero, Orat. 93, Tusc. 3 .44 and Sest. 121 confirms Diintzer's
view9 that the killing of Polyxena also took place during the action of
the play.
Speeches by and about Andromache (Varro, Ling 5 . 19,10 7.82), Astyanax
(Varro, Ling. 10.70) and Ncoptolemus (Cicero, De orat.2. 156, Rep. 1.30,

1
In his second edition of Nonius (Paris, 1614). He grantedthe possibility of
a title 'Av6poµax1l1'lv Mo>.OTTots.
i Zeitschr.f d. Alt. 1838, 47 ff. 3 Dk g,iech. Trag. p. 1200.
4 Cf. Euripides, Amir. 1243 ff., Virgil, Aen. 3 .294 ff.

s Anal. seen.p. 36.

6
Qiy,est.seen.pp. 257 ff.
7 Tcrzaghi, BFC xxxu (1925), 16 f., nevertheless wanted to put the verses
about the women of Sparta quoted by Cicero at Tusc. 2. 36 ( = Trag. inc.
205-:-8) into Ermius' Andromaclia because of their similarity with Euripides,
Andr. 595 ff.
8
Sec below on fr. xxvna.
9 Accepted by Ribbcck and tacitly controverted by Vahlcn in E.P.R. 1 ; cf.
Vahlen's interpretation of Varro, Ling. 7.6.
10
I supply Amlroma(ca)Noctiwith Pomponius Lacfusrather thanAndrom(ed)a
Noai with Scaliger.

235
 COMMENTARY

Tusc.2.1, Gcllius 5.15.9, 5.16.5, Apuleius, Apol. 13) quoted along with
Ennius' name have no place in any other play of his which is quoted in our
1
sources and should be placed in the Andromacha.
One other republican tragedy about the kiHjng of Astyanax is known,
Accius' A.stya,uix.iThe monument set up for P. Numitorius Hilaruson the
via Salaria late in the first century B.c. represents against a Roman stage
background Ulyssesleading away Astyanaxand Andromachc standing by.3
Speculationabout theauthor of this play is vain.
I have followed Columna in givingthetwo trimetcrs quoted by Cicero at
Tusc. 1. 10s to the Andromacha(fr. XXVII g) for four, perhaps not sufiicient,
reasons.First, theform of HectoremexcludesAcciusas theauthor ;4secondly,
Ciccro's ilia suggeststhat he has already before referred to thespeechfrom
which he quotes and thequite certain referenceto theAndromacha at 1 • 85 is
theonly referenceone could connect. Thirdly, thelong quotation from the
Andromachaat 3 .44-6 (fr. XXVII h) shows that Ennius' heroine did indeed
speakof Hector in trimetcrs; fourthly theuidiof v. 78 seems to be picked up
at vv. 89 and 92.
Cicero and Varro seem to have believedthat Euripideswas theauthor of
Ennius' original. On theimplicationsof Ling. 7. 82 secabove, p. 203. At Opt.
gen. 18 Jahnmade largo-scaledeletions in order to cure an obviously ailing
text: idem.Andriamet Synephebosnee minus [Tcrcntium et Caccilium quam
Mcnandrum lcgunt nee] Andromachamaut Antiopam aut EpigonosLatinos
reripiunt,[sedt:amenEnnium et Pacuuium et Acciumpotius quam Euripidem
et Sophoclcm legunt]. This seems rather drastic surgery but even if we per-
mit it we must still take seriouslythe statements that arc made. The alleged
intcrpolator could have got his information about .Andrid,Synephebiand
Antiopafrom Fin. 1.4 but that about Andromachaand Epigoniappears only
here and it is difficult to sec how it could have been invented. An intcr-
polator with no genuine knowleclgcwould surdy have insertedthename of
Aeschylus.5
The extant •Av6poµ<X)(Tl cannot be a candidate. The play known to our-
selvesand to the ancientsas T ~Ses has Andromachc as a personage and
deals with the kj)ling of Astyanax. Many of the fragments of Ennius'
Andromachahave verbal analogues in this play as well as in Euripides'

1
On Neoptolemus see Ribbeck, Q]y,est.seen.p. 258. Welcker, Die griech.
Trag. pp. 1374 f., assumed the existence of a Philoctetesof Ennius in order to
accommodate him.
i On this play see Zielinski, Bos XXXI (1928), 2 ff.

3 See Pickard-Cambridge, DramaticFtsti11als, p. 186 and fig. 57.

4 Cf. Varro, Ling. 10.70.
5 Cf. Cicero, Ac. 1 . 10.

236
 ANDROM.ACHA
•Av~ and 'Emf:nl. 1
Diintzcr and~ suggested that Ennius
used the Tp(tXX6es as the basis of his play and added speeches and scenes
from •Av6poµax11,'EK~11and other plays.
This theory implies a degreeof freedom in adaptation far beyond that
attributed by Terence to the early Latin poets.3 In any casewhatever play
Varro had in mind it was not the Tpctx,e~. The etymology of Andromache's
name in question does not occur in our text of this tragedy and could not be
imaginedin any other.4 Furthermore there is no evidence that the Tp<t)a6ES
was ever called•Av6po'10XT) alxµCXAOOTis. The only purpose of the epithet
alxµCXACA>Tf swould be to distinguisha script with the title•Av6poµax11 from
that of the extant •Av6poµax11and yet an epithet more unsuited to such a
purposeit is hard to imagine.
Columna suggested that there existed another Euripidean •Av6poµax11
adapted by Ennius and known to Varro and Cicero but now quite lost.5This
looks like a theory which could not be proved falsebut there is much to say
against it. The title• Av6poµax11 alxµCXA(A)T{swould have been a bad means
of distinguishingthe hypothetical play from the extant. Av6poµax11. In any
casethe scholium on Euripides, Amir.445 implies that the scholarsof Alex-
andria could find no play by Euripides with the title •Av6poµax11in the
record of plays produced at Athenian festivals.We know the names and can
guess at the themes of the seventy Euripidean tragedies possessedby these
scholars6 and there is no place among them for a second play about the killing
of Astyanax or even for one containing the etymology of Andromache's
name referred to by Varro. It is possiblethat such a play, written by Euripides
for performance outside Athens, survived in the possessionof actors and was
adapted by Ennius. Possible,but unlikely. It is even more unlikely that Varro
and Cicero alone of our witnessesshould have known about it.
The conclusion seems inescapable that Cicero and Varro were in error,
perhaps having been misled by a corrupt didascalicnotice like that now pre-

1
These were noticed long ago by Scaliger (Coniect.Varr.Ling., on 10. 70),
who wrongly attributed the Latin fragments in questionto the Hecuba,and by
Columna. 0. Schonberger (HermesLXXXIV (1956], 255 £) ought not to have
produced as a novelty Columna's identification of 88 ff. and Euripides, Tr.
479 ff.
a EuripidesRestitutusu. p. 284. C£ Wilamowitz, HermesLX (1925), 286 n. 2
(= Kl. Sehr.IV 374 n. 2).
p. 23.
3 See Introduction, 4 See below on fr. xxxv.
s C£ Welcker,DieGriech. Trag.p. 1203, Vahlen,E.P.R.1, p. ccm. The pieces
attributed to an •Av6poµCl)(T\ by Schol. Arist. Batr. 105 (fr. 144) and Stobaeus
1.3.23 (fr. 151) probably belong to the •Av6poµ!6a. Artemidorus, p. 284.13
Pack, whatever the truth about the verse quoted (Syleus, fr. 687. 1), refers to a
6
scene of the extant •Av6poµCl)(fl. See Introduction,p. 45.

237
 COMMENTARY
fixed to Tcrencls Hecy,a.where the name of the famous Mcnandcr has
replaced that of the obscure Apollodorus, or elsehaving applied theirgcncral
knowledge of Ennius' predilection for Euripides falsely to the particular case.
Such similarities as have been pointed out between the Latin Andromacha
and the threeEuripidean plays arc eitheraccidental (the plays have a similar
background of events) or may be due to the fact that Ennius translated a
Greek imitator of Euripides. In the fourth ccntury 1 and even in the fifth?
Attic tragedians were not averse to copying a predecessor's phrases and
themes when handlinga similar subject. Fourth-century tragedies seem to
have beenfilled, like contemporary comedies, with echoes of the universally
popular Euripides. A play about Andromache and Astyanax is recorded
against the name of Antiphon.3 There is insufficient evidence to discuss
purposefully Mord' s thcory4that the papyrus published by E. Lobel in Greek
Poetryand Life (Oxford, 1936), 295 ff., comes from a text of Antiphon's
play and that this play was Ennius' original.

XXVII
(a) From section 118 to section 123 Cicero describesscenic performances at
ludiof some extraordinary character held on the day when the Senate decreed
his recall from exile. He alleges that the actors and the audience understood
various speeches in the old second-century plays performed as referring to
himself and his enemy Clodius.S He names the fabula togatahe quotes at
section n8 as the Simulansand thefabula praetexta at 1.23as theBrutuJ.In the
intermediate sections he quotes three pieces Gfmusically accompanied tragic
verse, the middle one of which (section 121) plainly comes from the canticum
quoted at length at Tusc.3 . 44-5 and attributed explicitly to Ennius at Orat.
93 and to the Andromacha at Tusc.3. 53. He introduces thismiddle quotation
with the words ab eodempaulopost in eademfabula sunt acta.It would seem
therefore that the fust of the anonymous quotations-qui rempublicamcerto
animo adiuueritstatueritsteteritcum Achiuis. .. re dubia haud dubitarituitam
offerreneecapitipepercerit-belongsto the Andromacha and that the third could

1
Athenaeus 10.454B-B alleges that Theodectes borrowed from Agathon
and Agathon from Euripides the device of describing the letters of the name
8HIEYI.
s The hypothesis of Aeschylus' TTEpaoo reports an opinion of Glaucus that
Aeschylus copied the <l>o(v1aaa1 of Phrynichus. Servius (Am. 4.694) alleges
that the Thanatos of Euripides' •MKT1o-r1s was borrowed from Phrynichus.
3 See Aristotle, Eth. Bud. 7.4.1239a37.
4 BPh W LVII (1937), 558 ff.
s Cf. Cicero's account of the behaviour of the actor Diphilus at the ludi
Apollinaresof 59 at Att. 2. 19. 3 and Suetonius' accounts of certain actors of
imperial times at Aug. 53, 68, Tib. 45, Galb. 13.
238
 ANDROMACHA
(but need not necessarily) belong there too. In section 123 Cicero names the
poet of the Brutusas Accius and the actor Aesopus. There is no necessary
implication that Accius and Aesopus were responsible for the play or plays
quoted in sections 120-2. On theother hand the words at thebeginningof
section 123, et quoniamhue me prouexitoratio,histriocasummeum totienscon-
laaimauit,cumita dolenter
ageretcausammeam,ut uox eiusiliapraeclaralacrimis
impediretur;neque poetae, quorum ego semper ingenia Jilexi, tempori meo
defuerunt,
strongly suggest that the quotations of the previous sections came
from more than one play but that only one actor was involved. 1
A grave problem immediatelypresents itself.The canticum from the
Andromachawas spoken by theheroine herself compJaining of the destruc-
tion wreaked by the Greeks both upon her father's city and upon her
husband's. The words quoted before the canticum----quirem publicamcerto
animoadiuueritstatueritsteteritcumAchiuis• .. reJubiahaudJubitarituitamofferre
neecapitipepercerit-on the other hand laud someone who has co-operated
with the Greeks. The words quoted after, o ing,atifidArgiui inmoenesGrai
inmemortsbenefid,I exularesinitis,sistispelli, pulsumpatimini,upbraid certain
races of Greeks for ingratitude to someone who has helpedthem.
In 1565 Lambinus substituted Telamofor Simulansin 118 and pro Telamone
for pro me in 122. In 1573 he let Simulansbe. His notes show that he was
bothered by the presence of verses from Andromache' s canticum. Paulus
Manutius1 argued that everything in 120-2 came from Accius' (sic)AnJro-
macha3and that both theexiled Telamo and thebereftAndromache appeared
in this play.
In 1828 there turned up certain scholia4 which, though deriving in their
present form from a fourth-century commentary on Cicero's orations, go
back ultimately to quite early and genuinely learned work. They appear to
assert that everything in 120-2 was uttered at a performance of theEurysaces
of Accius by the actor Aesopus:
SVMMVS ARTIFBX actorillis temporibus
notissimustragicarumf abularumAesopus
egisseAcdi fabulam,quaescribiturEurysaces{Gaumitz: eurysaa C), ita ut per
omnemactioniscursumtemporareipublicaesigni.ficarentur et quodammodo (Orelli:
quodamC) Ciceronis fortuna deploraretur.ex quo illudprobartc.ontendit,omnes
prorsushominesetiaminfimaeplebisrestitutionisuaepromptissime suffragatos.
o IMMVNBS GRAI et haecuerbasunt de tragoedia:in qua uerbumistudimmunes
ingratossignificat,quemadmodum ********~******* munificos dicebanteosqui
1
For one actor appearing in several plays cf. Cicero, Att. 4. 1 s.6. This seems
to have been regular practice for a long period at the Athenian festivals; see
Pickard-Cambridge, DramaticFestivals,pp. 94 ff.
a Commentarius viii (Venice, 1579), p. 62.
3 Stephanusincludcd the piece quoted by Nonius at p. SIS. 12 in such a play.
4 Ed. A. Mai (Rome, 1828).

239
 COMMENTARY
gratitt liberatesexstitissent.ergouersusomnestragiciad ipsumCiaronemir).ay(CA>S
(Leo: ab AesopoLuterbacher: ******* C) eonuertuntur,ut aliudquitkmin opere
poetieoJuerit, aliudueroin ipsiusactorissignificationibus.
Madvig 1 declared that the scholiast was in error and that only the verses
o ingrati.fidArgiui, immunesGrai, immemoresbenefid,exularesinitis,sistispelli,
pulsumpatiminibelong to the Eurysaces.He made no suggestion as to how the
two pieces quirempublicamcertoanimoadiuueritstatuerit,steteritcumAthiuis and
re dubiahauddubitauituitam ojferreneeeapitipeperc.erit
might be placed in the
Andromacha.Bergk' made Andromache appeal to the Greeks on her own
behalf,reminding them of Priam's former services to them. Ladewig3 made
her ask the Greeks for mercy for Ajax, whom they were about to stone for
the rape of Cassandra. J. Bake 4 deleted in eademfabula and suggested that
three tragedies are quoted in 120-2, one anonymous, a second the Andro-
machaand the third Eurysaces.
The way out now universally accepted was suggested first by F. Nieber-
ding5and taken up by Ribbcck.6Nieberding settheEurysacesin Salamis around
the events narrated by Justinus at 44. 3 and interpreted qui rempublieametc. as
spoken by someone about Teuccr after he had been refused admittance to
Salamis by Eurysaces, Telamo's heir.7 Ribbcck combined the views of
Lambinus with the new information offered by the scholia and set the
Eurysacesin Aegina whither he imagined Tclamo had Bed after a revolution
on Salamis. He interpreted qui rempublieametc. as spoken about Tclamo by
Eurysaccs who had come to Acgina with Teucer looking for hisgrandfather.
Both Nieberding and Ribbeck treated the anapaests as inserted into Accius'
canticum from Ennius' Andromachaby the actor Acsopus.
There is good evidence that ancient actors played fast and loose with
authors' scripts. The scholia to the tragedies of Euripides arc full of allega-
tions of scenic interpolation. 8 In the very passage under discussion Cicero
says that the actor added the tetrameter summum amicumsummo in bello

1
'De emendandis Ciceronis orationibus pro P. Sestio et in P. Vatinium
disputatio', in Opusculaacademica (Copenhagen, 1834), 49S £
i Ind. leett.Marburg1844, vm. Bergk later withdrew; the text is altered at Kl.
phil. Sehr. 1 219 £
3 Anal. seen.p. 36.
4 Scholicahypomnemata 1 (Leiden, 1837), p. 119. C£ Bergk, PhilologusXXXI
{1872), 238 n. 9 (= Kl. phil. Sehr. 1200 n. 9).
s llias Homeriab L. Attio Poetain DramataConuersa(Conitz, 1838), p. 2s.
6 In Tragicos RomanorumPoetasConiedanea:SpecimenI {Diss.Berlin, 1849),
p. 34, ~aest. seen.pp. 328 £, Die rom. Trag.pp. 419 ff.
7 C£ Welcker, Die griech. Trag. pp. 197 ff., Robert, Die grieeh.Heldensage
m ii 2, p. 1483.
8
C£, for example, Med. 84, 148, 169, 228, 3S6, 380, 910.
 ANDROMACHA
summoingeniopraeditumto the Latin tragedian's script. Nevertheless it is hard
enough to believe with Ladewig that the one personage uttered the verses
quoted in 12<r2 in different scenes. To have even the most patriotic of acton
put them into the mouth of the one personage in the one scene is ludicrous.
If, however, we assume that Acsopw played two roles in Andronu,c/ui
Cicero' s quotations can be satisfactorily interpreted. With qui rempublicam
certo animoadiuuerit statuerit,steteritcum Athiuis and re Jubialuiuddubitarit
uitamoffme neecapitipepercerit he was speaking the part of a Greek. Later in
the play he assumed or rcassumed the role of Andromache; henceCiccro's
carefully chosen words abeodempaulopost in eadem f abula.Maskswere cer-
tainly employed by Roman acton in the mid first century, 1 whatever hap-
penedin Ennius' own day? The Greek practice of performing tragedies with
no more thanthree speaking ~cton probably came in along with the masks
that made it possible.
~i rempublicameertoanimoadiuueritstatuerit,steteritcum Athiuis and re
JubiaJu,uJdubitarituitamoffereneeeapitipepercerit could come from a speech
by Ulyssesin a debate over the demand of Achilles' ghost for the blood of
Polyxcnc, pointing out the debt owed to the dead hero (c£ Euripides, Helt.
136 ££,309 ££).
My argument in brief is that Cicero describedperformances by Acsopus in
Ennius' Andronu,cha, .Accius''Eurysaces and Brutusand that in the Andromacha
.Acsopusplayed two roles, that of Ulysses and that of Andromache henclf; a
scholar recognised the tetrameten o ingratifici ArgiuiinmoenesGraiinmemores
benefici,I exulare(sinitis sistispelli, pulsumpatiminias coming from the
'Eurysaas;someone cpitnmisiog this scholar's commentary extended the
Eurysacesquotations to include those which in reality came from the
Andromaclui.
Ennius' verses conjure up the city-state patriotism of second-century Rome
rather than the panhellcnism which the .Attic tragedians attributed to the
heroeswho besieged Troy. The Euripidcan Ulysses (Helt. 30!}-10) says of
.Achilles -fiµtv s· 'AxvJ-..Evs a~loS TIµi)s, ywoo, I 8avcl>v\l1rip yf\s
·ruaSoS KaAAlCTT avTlP·
1
The Latin tragedian makes his hero USCone of
the most hackneyed metaphon of Roman public life.
The tetrameter which Acsopus is reported to have added in the 57 B.C.
performance of Andronu,c/ui looks as if it borrowed its play with the adjective
summusfrom v. 105. Praeditusis very rare in republican drama (Plautus,
Amph. 218, Terence. Andr. 98).
79'-5 .. . rem publicarn certo animo adiauerit I statuerit: for certo
animoc£ Cicero, ~inct. 71, Anon. Bell.Aft.41.
For rempublicamadiuuarec£ Cicero, Phil. 8.30, 10.26, Off. 1.123.
1
Cf. Cicero, De orat.2. 193, 3. 221. i See Introduction,
p. 22.

t6 241 JTO
 COMMENTARY
I can find no other instance of rempublicamsl4lutrt. It must be a tragic
variant of rempublicamconstituert(Cicero, De orat.I. 37, Phil. 2. 92, 10. 22
et al.). For the metaphor, difficult to parallel closely in Attic tragedy 1 or the
language of Athenian politics, cf. Ennius, Ann. soomoribusantiquisres stat
Romanauirisque,Cicero, Vm. 3 .223 quirempublicamsistere negatposstnisi.• . ,
Virgil, Aen. 6. 857-8 hie remR.onumam I
magnoturbantetumultu sistettquts.

1teterit cam Achiais: c£ Caesar, Ciu. 1.61 quaesuptriorehello cum

Strtoriosteterantduitatts.
For the triple homoeoteleuton c£ Ennius, Trag.90 teais caelatislaqueatis,
308 cogitatparatputat, Accius, Trag.437 constititcognouitsensit,ssotiulatu
questugemitu,Trag. inc. 21 dokt pudttpigtt. The phenomenon unobscured by
synaloephe is hard to find in Attic tragedy; but cf. Aeschylus, Prom.691
m'n1crraAvµcrra 6efµcrra, Sophocles, El. 1235 ~T)vpn• i\A&-r"Et&e•.
76-7re dubia I bauddubitarituitam oft"ere:for the phrase redubiac£
Accius, Trag. 38-9 ut quaetum absentemrebusdubiisconiugemI tetineritnunc
prodatultoremand contrast Plautus, Epid. 113 is est amicusqui in re dubiart
iuuat, Capt. 4o6, Most. 1041, Sallust, lug. 14.5, Livy 7.30.3, 29.25.3. In rt
(rebus)+epithet is the normal type of phrase in republican drama; the only
other exceptions are at Trag. inc. 1.51and two paratragic passages of Plautus,
Peru 7S3 and True.7.5.
For the play with verb and etymologically related adjective cf. Naeviw,
Trag. IS laetussum laudarime abste pater a laudatouiro,Ennius, Trag. 351
amicuscertusin re incertacernitur,Accius, Trag.314 probisprobatumpotiusquam
multisfort.
Caputoffme seems to have beenthe more common high-falutin variant of
st offmt (Plautus, Capt.230-1, Lucretius 3. 1041, Cicero, Sull. 84 et al.). For
uitamoffme cf. Cicero, Sest. 61.

nee capiti pepercerit: c£ Sophocles, El. 979-80 ~ Tolow ~8pots ev

~l:nlK6olv 11'0TEI 'f'VXTlS&cpetST}aCXV"TE 11'povcrn'rn1v cp6vov,Euripides,
Herakles1146Tf Sf\Ta cpe(Soµoo'f'VXT)S lµf\s; ,Ennius, Trag.134 egocummeae
uitaeparcamletuminimicodeprecer, Plautus, Rud.222 uitaehauparco,Cicero,
Nat. deor.3 . 1 spatriaeconsulerent,
uitaenonparcerent.Vitaealicuiusparcere
seems
to have been the common variant of alicuiparcerebut cf. Accius, Trag. 294,.
This type of pleonasm whereby a negative phrase repeats the substance of
the preceding positive phrase is fairly common in Attic tragedy: cf. Soph~
clcs,Tr. 474 waaoo TaA1')8ES ov~ Kp\J\f/01,100,Euripides,Hek. 668 6AOOAQS
.oVKEr'Elet al. It is rare in republican drama.
1
But cf. Euripides, Tr. 116~1 ~\1
Tpo(av ,ro,-1 j maowav 6p8~cnnw, Hik.
c.2.29-30aupa I is 6peov tCM"fl.
 ANDROMACHA

(g) For thecontext of this fragment sec above, p. 236.

,S uidi, uidere quod me pauaaegemune: contrast Plautus, Bacch.1016

persuasumstf acerequoiusme nuncf actipudet,Cist. 504 foci saepequodf actum
queror,Men. 9SS ut parenturquibusparalisopusest et al.
For the omission of sum c£ Plautus, Men. 119, Most. 84 et al.

79 ctlffll quadriiugo: c£ Virgil, Georg.3.18, Am. 12.162, Ovid, Am.

is the normal term in comedy (Plautus, Amph.422 et al.) as
3 . 2. 66; quadrigae
in classical prose and verse.

raptarier: twice elsewhere in republican drama (Plautus, Aul. 632, Cist.

216) as against raperess times.The infinitive form in -ier seems to be no more
frequent in tragedy thanin comedy; the common language usedthe form in
-i.

(h) Andromache seems to have described Hector's death in trimeters and

then broken into cretic tetrameters to lament her hdplessness. For the use of
cretics in similarcontexts c£ Plautus, Gas.621 ff., Rud. 664 ff. For the con-
trast between thetrimeters carrying thenarrative and the musically accom-
panied verses carrying the expression of personalfedings c£ Euripides, Amlr.
1-116, Terence; Amir.206-27, Haut. 213-380, Eun. 292-390. Andromache
goeson to mourn the ruinedstate of the palace of her father Eetion in
trochaic tetrameters and theunseemly end of the once prosperous Priam in
anapaestic dimeters.
A number of the details of Ennius' metrical patterns are odd. Only one
other cretic tetrameter with an initialchoriamb is transmitted in the remains
of republican verse with a fair degreeof certainty (Plautus, True.624) and
not so very many with any kind of choriamb. 1 The choriamb can be re-
moved from v. 82 by treating the second i of auxilioas consonantal3 but this
would be to replace one anomaly with another. Hiatus and syllabaancepsare
common enough after the second metron in cretic verse but not after
others.3 The two casesof hiatus in v. 82 make it a very odd verse. The third
foot of v. 86 is constructed in an unusual way; 4 the i of abietecould be treated
1
A. Spengel, Reformvorschl.pp . .21 ff., banned all choriambs from crctic
verse.
a For this phenomenon in dramatic verse see Marx, Rudens,pp. 194 f., 0.
Skutsch, Prosodischeund metrischeGesetze der Iambenkurzung (Gottingen,
1934), pp. 43 f. In epic hexameters Ewtlus has aujum (Ann. 94), &ruilius (Ann .
.2s1: conpeluitSeruiliussic) and insidj,antes
(Ann. 436).
3 Cf. Ritschl, Prol. Trin. p. CXCVI, C. F. W. Mueller, Plaut. Pros.p. 6.27.
4 See above on fr. XII.

243 16-2
 COMMENTARY

as prosodically null I but this again would be to replace one anomaly with
another. Lindsay scanned the words deformatiatqueabietecrispaas an ana-
paestic dimetcr 1 but offered no explanation of why Ennius should have
changed the rhythm in the middle of the account of the burning of Eetion' s
palace. The dactyl in the final foot of v. 87 has few relatively certain ana-
logues among the anapaestic verses of republican drama (but cf. Terence,
And,. 625, Accius, Trag. 571,Trag. inc. 96).
Despite these metrical anomalies the sense and rhetoric of all the words
transmitted except faga in v. 82 and alii in v. 85 appear to be excellent.
The emendations offered by analogising metricians are singularly un-
pcrsuasive.3
Ennius makes his Andromache speak of herself as a stateless person in
terms of Roman law and socialpractice (opis egens. .. quid. .. praesidi? . •.
quouenuncauxilioexili autfugae?... arceet urbeorba... quoapplicem?).One
might compare the way in which Euripides' Medea draws upon the special
language of Athenian law (Med. 386-8 Tis µe St~e-rooir6A1s; I Tfs yfiv
6:0VAOV Kal 66µovs ~eyyvovs I ~M>S-rrapaax~v ~VC1£T<XlTovµov
Stµcxs;)and contrast the quite general language of his Hecuba (Hek. 158--04
~µ01 µ01 I Tis aµwe1 µ01. ,ro{a ywva; I ,ro{a Se 1T6A1s;I q,po06oS
,rpw~vs. q,pov601-rratSes.I ,ro{av ii TavTav ii KE{vavI a-n{xoo; ,rot s·
,iaoo; -rrovTIS 6eoovI ii Saiµoovvc;>vhrapooy6s; ). It looks as if Andro-
mache hasbeen offered her freedom by her Greek captors 4 and is bewailing
her inability to profit from such freedom.
Ennius lays much emphasis on the religious aspects of the destruction of
the palaces of Eetion and Priam, using, to some extent unavoidably, the
Roman sacral language (araepatriaedomi. . .Jana. .. domus,saeptumaltisono
cardinetemplum... Iouis aram),and perhaps importing with Andromache's
horror at the pollution by blood of Jupiter's altar feelings foreign to the
world portrayed by Attic tragedy. Andromache speaks as if the destruction
and pollution of the altars affectsquite significantly her own wellbeing. One
is reminded of the tone of Cicero' s speech De Jomosua5and the strange con-
cern of Virgil's Aeneas for the images of the Troiaepenates.The Homeric
Andromache says nothing about altars in her account of the destruction of
her father's kingdom (II. 6.414-28). Euripides' Trojan women are similarly

1
Cf. in epic verse Ennius, Ann. 104 Ner(i)enem,Lucretius 2.991 or(i)undi.
Marx, Rudens,p. 195,allowed the phenomenon in dramatic verse.
i ALL VD (1892), 596.
3 Cf. A. Spengel's transposition of Priami and domus (Reformvorschl.
p. 326).
4 Cf. Agamemnon's qualified offer to Hecuba at Euripides, Hek. 754f.

5 Cf. 109 quidestS4tldius

. .. quamdomusuniuscuiusque
duium?hiearaesunt,hie
foci, hiedi Ptnates.
244
 ANDROMACHA
unconcernedfor the altarsof Troy. 1 It is the gods of Attic tragedy who worry
about altars.2 The human disposses.,cd have other worries.3

8o ex opibua 1ammi1 opis egem Hector taae: Vahlen compared

Plautus, Mere.111 ex summisopibusuiribusque usqueexperirenitere4andsupplied
uocote, :wimilating the substance of the Latin words to that of Euripides,
Andr.523-5 w1T001S 1T001Set8e aav
j xetpa Kai 66pvavµµcxxovj lCTT}aaf-
µav TTp1aµov mxt and Tr. 587-90 µ6Ao1scI>1r001sµ01.. . aas
Saµ<XpTOS
@.Kap.It is better to take the phrase ex opibussummisas of thesame type-
as those in Euripides, Hek. 55--0 cI>µf\np i\TIS '1<TVpcxw11<00v 66IJ(.l)vI
SovAaOVi\µap eI&s.Ennius, Ann. 312--13 mortdlansummumFortunarepenteI
reddidite summoregnoutf amulinfimusesset,Livy 2. 6. 2 ne se ortum,eiusdem
sanguinis,extorrem,egentemex tanto modoregnocum liberisadulesantibusante
oculossuosperiresinerent.For opts' socialposition, power, influence'c£Plautus,
Cist.•94 nequeopesnostraetom suntualiJM~ tuae,Ennius, Trag.339-40 ut
sdas quantoe loco,I quantisopibus,quibusde rebuslapsaf ortunaoaidit.For ops
'protection afforded by such power to socialequals•c£ Plautus, Persa256 ei
egentiopemadferam,Sallust, Hist. 2.47.4 cum egensalienaeopisplura mal4
exspectarem,Cicero, De orat.1 • 184 praesidiumclientibus,opemamicis. .. porri-
gentema.tquetendentem.
For the polyptoton c£ Accius, Trag. 124 ut meaope opesTroiaeinteg,em.

81 quid petam praesidi aat exequar?: exequa,is usually interpreted as

'quaeram'; P. Schmid, T.L.L. vii.1850.26, refers to Cicero's remark
qiu,eritenim auxiliumand compares Plautus, Epid. 572 quaeexanimataex-
sequituraspeaumtuom.The resulting pleonasm is possiblefor Ennius but can
be removed by interpreting thesecondverb as' consequar'; c£ Plautus, RMJ.
261-2 bonama.tque obsequentemdeamatquehaudg,auatamIpatronamexsequontu,
benignamque multum.
Praesidium denoted in the legallanguage the protection which a patronus
gave his elientes;c£ Cicero, De orat.1 .184-

81-2 quoue nunc Iawdlio exili aut fugae&etasun?: Bentley'sfugaeis

an absolutely necessarycorrection. Fugamakes no sense.For the coupling of
exiliumandfagac£ Plautus, Mere.652 quismodustibiexiliotandemeuenietlqui
1
Cf. Amir. 103 ff., 394 ff., Hek. 47S ff., 90S ff., Tr. 98 ff., 474 ff., S77 ff.,
1277 ff. At Tr. 1317 the &~ l,,IO.a8pa simplytake their placein the catalogueof
destruction (cf. Andr. 1024 ff.).
:a Cf. Aeschylus,Ag. s2s ff., Euripides. Tr. IS ff., 69 ff.
3 It is not impossible that Ennius follows a Greek text closely; cf. Sophocles,
0. T. 1378-9, Euripides, Herakleidai876 ff.
• Cf. Plautus, Mil. 620 ea te txpelert tx opibussummis•
.245
 COMMENTARY
finisfagoe?,Cicero, Caul.1 . 22 et al.The lawyers made a fine distinction:c£
Marcianw, Dig. 48. 22. 5 txilium triplextst: out art«um locorum interdictioaut
latafaga. •. aut. .. rtlegauoin insulam.If Enniw was adaptinghisAttic original
closelyat thispoint he would probably have had theono-wordphrasecpvy,'i
in front of him.
For -ue connecting questionsc£ Plautw, Asin. 636, Rud.503. The com-
parative rarity of theusage suggeststhat it had a formal, perhaps legalistic,
tone.
For autlinking near➔ynonyms c£ Plautw, Amph. 90,4 si sissanu.s autsapias
sans et al.
Auxilium denoted theactive assistancegiven to both climtn and amid in
times of trouble.

83 arceet mbe orbe.mm: cmoAfsel1,11. A formulaof theofficiallanguage

is employed. However there could have beena very similarphrasein Ennius'
original; c£ Euripides,Mtd. 771 µoA6\rres6o-tV Kai TI'OAtaµaTT<XAAa6os.
For the Roman formula c£ the Fctial oath quoted at Fcstus,p. 102. 11 si
scimsfallo tum mt Ditspitersaluaurbt arcequeboniseiciatut tgo hunclapidtm
(N.B. Polybiw' translation, 3. 25. 8 ,rmoov TOOV l!iAA.oovact)3<>1,1WOOV
W
Tats 16foos,rcrrpfaw, w TOTSISfoasv61,101s, hrl ToovISfoov~foovlepG>v
), Caeciliw,Com.146quiquasiadhomscaptuslibtrseruiosaluaurbtatqut
Tacp<A>V
arce(an addition to the Menandrian original), Cicero, Diu. 2.69 ad arcem
urbtmquerttinendam(concerning the effect of an oracle), Livy 4.61.9,
24.37.6, 31.45.6, 37.37.2.

quoaccedam?quoapplicem?: c£ Sallwt,lug. 14. 17 quoaccedam autquos

appellem? The cretic rhythm suggeststhat we have aaldam(aaid4m)rather
than acddam.The phrase adgmua is to be understood. For theuse of quoin
reference to persons c£ Plautus, Mere.803, Mil. 119, Stich. 142. This and
similarusages(e.g. of undt)are not common in republicandrama; they may
have had a legalistictone. Applicartcertainly recallsvery forcibly the client-
patron relationship; c£ Terence, Andr.924-5 ille. .. adplicat . .. ad Chrysidis
patremse,Haut.393 hi se aduosadplicant, Cicero, Dt orat.1. 177 quiRomamin
exiliumumisset,cui Romoeexulart ius esset,si se ad aliquemquasipatronum
applicauissetintestatoquetsstt mortuus- nonnt in ta causaius applicationis . .•
patefactum esta patrono?For theintransitiveuse of
in iudicioatqut inlu.stratum
applicarec£ that of ejjlareat v. 14 and of recipereat 298. Supplicarewas
regularly intransitive.
For the anaphoraof the sameform of the interrogativeand the isocolonc£
Euripides,Alk.Tf ovirpos 1,1Ma6po1s, Tf ovTij& iroMts;, 863irot ~. irot
aToo;Tf "}J;yoo, Tf SEµfi;, Plautus, Mm. 114-16 rogitas ... quidpetam,quid
feram.Such phrases,wbcrcthere are only two members, are normally
.246
 ANDROMACHA
linkedby Si in Attic drama and aut in Roman. The asyndeton increased
the emotional intensity of the utterance.

S. cui nee arae patriaedomi stant: the legalflavour of the preceding

verses suggests that neeis an archaism (c£ Lex XII tab. s.4 cui suus hms nee
tscit, 8. 16 quodneemanifostum erit)rather thana Graecism ('ne ... quidcm',
ovSi).

85 6ma fLammadeJlagrata:fanum appean to be here usedin its original

sense of 'consecrated place'; c£ Livy 10.37.1sfanum tantum,id est locus
temploeffatus,fuerat.
Neither dejlagrare nor jlagrareoccurselsewherein republican drama. In
classicalprose and verse the simple verb is much commoner than the
compound.

I
8~ totti faliit stant puietes, defonnati atque abiete crispa: all
editon seemto have accepted the old correction altiand take deformatiatque
abiettcrispato refer to parietts.But the latter, though roasted, still stand
upright (tosti.. . stant).Other thingsmust have lost their original shape.Alii
could be answered by another alii in the part of the quotation omitted by
Cicero. If altiis read something like posttsmust be assumed to have stood in
the second member of the antithesis. Cicero all but says that the phrase
deformatiatqueabiettcrispais incomplete (scitisquaesequantur).For similar
incomplete quotations c£ Tuse. 1. 34, 1. 106, 2. 39, 3. S3, 3. s8.
Deformare doesnot occur elsewherein republican drama. In classicalLatin
the word is confined to poetry and history.

8'] o pater o patria o Priami domas: c£ Sophocles, 0. T. 1394-s c:'>

T16A~ Kal K6pn,8eKal TCX ,r&-rp1aI A6y~ 1TCXAatCX &.:>µae•,Euripides,
Med. 166 c:'>,ra-reper,1r6A1s.
For the parechesis pattr ~patriac£ Plautus, Capt. 43, Men. 1083, 1090,
Mere.660.
For the anaphora and ascendingtricolon c£ Sophocles, Ant. 891-2 G>
TV~, c:'> wµq>Etov, KcrraaKacp1'\s
c:'> I ol1<1101s
&e(cppovpos, Terence, Ad.
790 o eaelum ottrraomariaNeptuni.This type of phrasal arrangement, in which
the third of three dements balances or outweighs the previous two, is much
more common in Roman dramathanin Attic and seems, relativdy speaking,
both more common in tragedy than comedy and more common in the
musically accompanied parts of comedy thanin the trimeters; see Lindholm,
StilistischeStudien,pp. 26 ff.
For the vocative use of the nominative form c£ Trag. inc. 184-s o ®mus
antilfU",heuquamdispariI ®minare®mino, Novius, Attll. 4,0 o ®musparata

247
 COMMENTARY
pulchratf amiliatfestiuiter,Euripides, Plwin.1 soo
er,661,10S
er,Soµos. This use is
quite distinct from Plautus' appositional nominatives (Asin. 691mi Lihane
ocellusaureuset al.) and may justly be termed a Graecism.1

88 saeptumaltisono cardinetemplum: a bizarre phrase. However, that

the seat of Troy's government should bea templumrdiects the ideas of con-
temporary Rome; c£ Varro, elaayooy1K4> adPomp.,ap.Gell.14.7. 7 docuitque
••. nisiin locoperauguremconstituto, quodtemplumappellaretur,senatusconsultum
f actumesstt, iustumid nonfaisse;proptereaet in curiaHostiliaet in Pompeia. ••
cumprofana ea locafuissent,templa esseperauguresconstituta,Servius auct. Aen.
1 . 446 ... eranttamen templa in quibusauspicatoet publiceresadministrarentur
et
senatushaberi posset,Cicero, Ltg. 2. 26 quorumhiemundusomnistemplumessetet
domus,Virgil, Aen. 7. 174 (the palace of the kingofLanuvium), Seneca, Thy.
901-2 (the palace of Atreus). For s«pire in connection with the enclosure of
the templumc£ Varro, Ling. 7. 13 omnetemplumessedtbtt continuoseptumnee
plus unumintroitumhabere,Festus,p. 146. 12 MINORA TBMPLAfiunt ohauguribus
cum locaaliquatabulisaut linteissepiuntur,ne uno ampliusostiop41eant,certis
uerbisdtfinita. itaquetemplumest locusita tjfatus aut ita septusut ex unaparte
pateatangulosque adfixoshabeatad terram.
Door pivots are not mentioned in the remains of Attic tragedy although
the appearance of penons from insidethe stage house is frequently heralded
by a noise from the opening door.a Gracchus mentions them twice.3Neither
Attic nor Roman comedy refers much to them.4
Altisonus, 'V'f'Tlxils',
occurs elsewhere only at Ennius, Trag. 188, Ann.
S7S, Cicero, Carm. fr. 19. 1, Seneca, Ag. 582, Here.0. 530, Phaedr.1134,
Juvenal n. 181. The only other possible formation with-sonus in republican
tragedy is suauisonus(Naevius 20, Accius 572). There are none in comedy.

89 uidiego te adstante ope barbarica:it is more likdy that a scribe

smoothed adstanteto adstantem at Tusc.3 . 45 thanthat Cicero twistedEnnius'
phrase to suit himv.lf at 1. 85. Ennius is referring to the military support
given to Troy by her Asian allies.C( Virgil, Aen. 8. 68 s-8hincopt barbarica

1
SeeJ. Svennung, Anredeformen (Uppsala, 1958), p. 267.
:a Euripides, Hel. 859-00, Ion 515-16, Or. 1366.
3 Trag. 1 o gratacardo regiumeg,essumindicans,2 sonatinpulsaregiacardo;cf.
Seneca,Mtd. 177 sedcuiusidu regiuscardostrepit?Contrast Pacuvius, Trag.214
ualuaesonunt,Accius, Trag.30 sedualuaeresonuntregiM,470 atqueadto ualuas
soneresensiregias.
4 The OT~, 'socket', appears at Aristophanes, Them,. 487, fr. 255,
Hermippus, fr. 47.9. The context at Plautus, Cure. 94 and 158 is paratragic.
The normal advertisement for the newcomer from the stage house in Attic
comedy is ~'llCE Tl'iv8upavet sim., in Roman Joris conaepuitet sim.
248
 ANDROMACHA
uariisqueAntoniusarmisI uictorohAutoroepopuliset litorerubro,I Aegyptum
uiresqueOrientiset ultimasecumI Bactrauehit.
Bap~tK6s docs not occur in Attic tragedy and so i• nn1ikdy to be the
immediate source of Ennius' barbaricus. The latter occurs elsewhere in
republicandrama only at Pacuvius,Trag.270 and three high-falutinpassages
of Plautus, Capt. 492, 884, Cos. 748. In dauical Latin its use is rcstrictcd,
except for I.ivy 25.33.2, to poetry. &rbarusis frequent in comedy and
classicalprose. Similarforms in tragedy are Delphicus{Ennius303), modicus
{Ennius 181), musicus{Pacuvius 114), tffltbricus{Pacuvius 158), trabicus
(Pacuvius400).Plautushasa largenumber at the endsof iambictrimeters and
trochaic tetrameters. Their frequency in the Roman official language
(poplicus,ciuicus,hosticus,col.onicus
et al.) gave them a formal tone.
For fl&p~/barbarus in the mouth of a non-Greek without any
pejorative connotation cf. Aeschylus,Pers. 254-5 6'1(a)Ss• avayK1l-rrav
ava,nv~cn,ra8os, I TT'Pacn.crrparosyap was&,.~ pap~
(a Persian messengerspeakingto his queen), Plautus, A.sin.11 Demophilus
scripsit,Maccusuortit barbare(i.e. Latine), Trin. 19 Philemoscripsit,P"1utus
Cicero, Orat. 16o absurdumerat.. . in barbaris
uortitbarbare, auihusGroecam
littnam adhibere.

90 tectia caelatialaqaeatia: tectisis a poetic plural; cf. Trag. inc. 213,

242 and contrast Plautus, Amph. 1008 et al.
Coffered ceilingsin private dwellings always indicatedgreat luxury;
cf. Lucretius 2.28, Cicero, Tusc. 5.62, Horace, Corm. 2.18.1, Anon.
Culex 63-4, Seneca. Nat. I prol 8, .Ep. 90.42 (non impendebant caelata
laqutaria ), 115 . 9, Plutarch,Lycurgus13 {storyofleotychidcs the ddcr ). That
they are not mentioned in the remainsof either Greek or Roman comedy is
perhaps an accident. Aristophanes,Sph. 1215 and Diphilus, fr. 61 seem to
refer to them. The lackof mention of them in extant Attic tragedy could be
due to a ddiberate avoidanceby the poets of anachronism.If so Ennius has
here acldcd an architectural detail foreign to his original. Nothing cor-
responding with the impluuiummentioned at Amph. no8, Mil. 159 et al,
Eun. 589 could have stood in the comediesadapted by Plautus and Terence.

91 aaro ebore instructam regifice: for instruere'furnish richly' cf.

Plautus, Bacch.373, Cicero, Nat. deor.2 .95, Virgil, A.en. 1.637-8.
For ivory in domesticfurniture cf. Homer, OJ.4.72-3 X<XAKOV TE crre-
po,niv Ka6&;,µara t'lxftwra, IxpvaoO T" t'}1JK-rpov TE Kerlapyvpov f\6'
Wq,cxv-ros, 23 .1~200 Mxos ... SatMAAoovXP\JO'Ct> -n Kerlapyv~ 1'}6'
rucpavn. Euripides,1.A. 582-3 ~oov ,rapo11ew Soµc,.w ~ crras
(of Paris). There are surprisinglyfew mentions of ivory in extant Attic tra-
u.
gedy {Sophocles,fr. 1025.7, Euripides. I.A. 582, Achaeus, 25.2). The

249
 COMMENTARY

notorious wealth of Troy is usually symbolised by gold (cf.Euripides, Andr.

168-9, Hek. 492, Hel. 928, Tr. 18, 994-5).
R.egificus(-e) occun only here in republican drama; it is rare in classical
Latin (the lexica quote Virgil. Aen. 6.6o5, Val. Flacc. 2.652, Silius 11 .271).
Drama has the synonymous regius(-e) eight times. Other formations with
-ficus in tragedy are ingratificus(Accius 364), hostificus{Accius So, 83),
laetificus (Pacuvius 414); comedy has bent.ficus,
(Trag. inc. 134), largificus dam-
nificus, tklenificus,falsificus,fumificus,fartificus, lucrificus{?), magnificus,
malejicus, mirificus,
munificus, spurcificus,
ueneficus. The common occurrence of
magnificus, male.ficus
and ueneficus suggests that these belonged to the common
language but the rest may be poetic creations. With the exception of the
tragedies of Seneca1 classical poetry has very few such formations.

92 iaftarnrnari:elsewhere in republican drama only at Accius, Trag.14

(for the rarity of jlamma sec above on v. 22). Inandereoccurs 13 times in
comedy, once in tragedy.

93 ui uitam euitari: for the sound play cf. Turpilius, Com. 202 quibus
rebusuita anumtuminuitarisolet,Trag. inc. 131 uirginemme quondmn inuitam
per uim uiolatluppiter.
Euitare(uita) occurs elsewhere in Latin only at Accius, Trag. 348 and
Apuleius, Met. 3 . 8. For the formation cf. Plautus' exanimare(probably from
the common language), exossare,elinguare,exdorsuare, exoculare,euiscerare.
A different type of formation is represented by expectorare at v. 17.
Besidesperire,mori,necareEnnius' tragic fragments contain a wide variety
of periphrases for ki1ling and dying: uitam euitari,144 rtUJternosanguine
exanclando,18o perguntlaueresanguensanguine,182 me anirru,priunn, 183
mortemobpetam,192 Acherontemobibo,230 perniciemdabo,283 letodatisunt,
299 remissahumanauita, 323 oppetomalampestem,328 exitiumparat, 329
pestemut participet. Those of his successon have much less:Pacuvius 148 leto
dabit,286 mortem. .. offeras,289 mactoinferis,Accius 117 letoojferes,43 3 morte
inbuturummanus,491 mittisleto.Plautus is probably parodying contemporary
tragedy at Asin. 606-11 ... a uitaabiudicabo . .. morti tktkre. .. uitamlargiar . ..
uitam esse amissurum,Bacch.849 exheredemfecero uitae suae, 869aniniam
amborumexsorbebo.Few of the Eunian expressions have exact parallels in
Attic tragedy.

94 louis anm sanguine tmpari: no altar of luppiterstood in a Roman

private house or even in the R.egiaor any of the curiae.Ennius is representing
the ZevslpKEloS(KaT)O'loS) of hisoriginal as best he can.With the exception

1
Sec F. Skutsch, Glotta D (1909/10), 16o f. (= Kl. Sehr. 386).

250
 ANDROMACHA
of the paratragic Amphitruo (c£ 1127) Roman comedies make theLares the
object of domestic cult (c£ Plautus, Aul. 2, 386, Mere. 830 ff. (contrast
Euripides, Phoin. 631 ff.), Mil. 1339, Rud.1207, Trin. 39).
For aram sanguineturparic£ Lucretius 1 • 84--0Aulide quopacto Triuiai
uirginisaram I Iphianassaiturparuntsanguinef oedeI ductortsDanmundeltai,
primt1uirorum,Sallust, Hist. 1.47, Virgil, kn. 2.499-502. The Euripidcan
accounts of Priam's death (Hele.21-4 hn:l & Tpofa e••EKToposT arr6A- 1

1'vrcn I 'fNX,;,mrrJXt><X e•mia K<X'TEC1Kaq,Tl,I <XVToS & lx,.>"'4>irpos

8eo6µ,;T<t> mTVEt I a~ls •AxtX>J.<.i:,s
ircn6QShe µ1cn'6vov, Tr. 16-17
,rpos5! t<pT)m&wj:xiepo1sI 'lffflTCt.>KE TTp(aµosZT)vosA()J<E(ov 8avoov,
1
481-3Kal Tov cpvrovpyovTTpfaµov... KaT~ lcp'A{)1<Efct>iru~)
arc unconcerned with the dirtyingof the altar and it is possiblethat Ennius
is here adding :i horrific detailfrom the sphereof Roman tabu.
The Olympian deities of Attic tragedy certainly objected to suppliantsat
their altars being slainand only Artemis, to the horror of the poets who
relate the stories,ever demandedto be propitiated with human blood.2 But a
regular and often mentioned part of theact of sacrificewas to pour the blood
of the sacrificedanimal over the altar.3 This was done either indirectly by
meansof a a~Tov or directly by cutting theanimal's throat over the altar
itsel£4 The mode of sacrificepractised at Rome differed in various details
from theheroic and historical Athenian modes, for example in theboiling of
the entrails of the slainanimal in vesselsseparately. There is a strange reti-
cencein our sourcesaboutwhat happenedto the animal'sbloodin sacrifices
ritu Rommu,to Iuppiterand the other di superi.5Livy's horrified description of
Samniteritual (10.41) and Virgil'saccountofDido's terror as the wine sheis
1
Cf. Pindar, Paian6. IIJ ff.
1
Cf. the story of the substitutionof the deer for Iphigcnia and the language
of Euripides' messenger,I.A. 1595 ~ µ1'!µ1CM)i3CA>µOv E\iyEvelcp6v<t).
3 Cf. Aeschylus,Theb. 275 µ1'iA01a1v alµ6:aaoVTCXS amcxs Euripides, Ion
8EG'>V,
I 126--7~ acpayatC11 Atowaov -rrtrpasI SEvcntE61aaas 1Tat6os &VT'btrnip(CA>\I.
4 Cf. Aeschylus,Ag. 232, Euripides, El. 813 ff., Hel. 1561 ff., I.A. 1578 ff.,
I.T. 26 ff.
s This was noticed by W. Warde Fowler, The ReligiousExptrima of tht
Roman Peoplt (London, 1911), pp. 180, 196. Later writers (e.g. G. Wissowa,
ReligionundKultusde, Romer,ed. 2 [Munich, 1912], pp. 416 ff., Latte, RE IX i
(1914], s.v. immolatio,1129, Rom. Rei. p. 388) have seen no significancein it.
Nothing about the ritusRomanusis to be learnt from Ciris525 sanguinetaurorum
supplex rtsperserataras, Catalepton14.8 uictimasaaatos spargethonortfocos,
Seneca, Phatdr. 498-9 non cruor'4rguspias I inundataras.The silence of the
A~id, stuffed as it is with Roman antiquarian lore, is eloquent. Lucretius
4.1236-7 sanguine... conspergunt aras, s.1201-2 Mas sanguine... spargert,and
Amobius 3 . 24 nisi ~corum sanguinedtlibutas suas consptxerintarulas,suos
deserunt. .. pratsidiatusarc concernedwith religious ritual in general.
251
 COMMENTARY

pouring upon the altar turns to blood (Am. 4.453-5) suggest that the
Romans fearedto let blood of any sort staincertain altars.If thisis so we have
part of an explanation why the Roman antiquarians believed that animal
sacrificewas unknown m early Rome.1 Ennius' phrase Iouisaramsanguine
turparimay therefore be not a literal translationof some Attic poet's phrase
but an original creation aimed at superstitionspeculiarto second-a:ntury
Romans.
Turparedocsnot occur elsewherein republicandramaand is very rare m
classicalprose.
The two-syllable rhyme nmning over three anapaestic dimcten...
injlammari . .. euitmi. .. turpariappearsnot to be paralleledm ancient drama.
Euripideshasone-syllablerhyme over two dimetersoften enough (e.g. Helt.
70-1, 111-12, 116-17, 127-8, 130-1, 142-3), Plautus two-syllable rhyme
over two (e.g. &ah. 1094. Rud.955~a).

XXVIII
Aulus Gdlius quotesphilosophmulum estpaucisas the very words of Ennius.
Efforts to extract something different from Cicero's various paraphrascs-
decreuiphilosophari . .. paucis;se aitphilosophari
utilesedpaucis;philosopharisibi
ait necesse essesedpauds-should be abandoned.
In his first edition of the tragic fragments Ribbcck got from Gdlius' dis-
courseat s.16. s the trochaictetrameter degustandum exphilosophianonin eam
1
ingurgitandum. L. Mercklin's arguments against the particular form of
Ribbcck'sverse have some weight but the mode of expressionseems a little
drasticeven for Gelliusand we may m fact have a paraphrasediverging only
slightlyfrom Ennius' own words. For the phrase degustare ex philosophiac£
Labcrius, Mim. 36 sequereme in latrinumut aliquidgustts ex Cynicahaeresi,
Tacitus, Dial. 31. 7 sed eum qui quasdamartts haurire,omneslibaredebtt,
Q!!intilian,Inst. 12. 2. 4 qui litterasuelprimisut aiuntlabrisdegustmit,Schol
Pers. 3 . 52 certe degustasti philosophuun.For in philosophiamingurgitare c£
Cicero, Fin. 3 . 7 quasihelluarilibris.
The words q>tA6a~ and q,~oooq,etvdo not occur m extant Attic tra-
gedy) The activity of q>tAoaoq,{a however is frequently representedunder
the traditional heroic masb by the innovating Euripides: TOlOVTOS arnv
&ef, TIX'l')(XA>tKairpoac.ma eloay(A)vq>tAOO'Oq>OWTa. 4 Neoptolemus'

1
C£ DionysiusHal. Ant. Rom. 2.74.4, Ovid, Fast. 1.337 ff., Pliny, Nat.
18. 7, Plutarch,Rom. 12, Num.8, 16. Mere pythagoreanisingwill not sufficeas
an explanation.
a N]bb Suppl.m (1860), 668.
3 Trag. Grace.me. 522 looks like comedy.
4 Schol.Hipp. 953; cf. Athenae,us13. 561 A.
252
 ANDROMACHA
modified anti-intellectual sentiments were probably common among men of
affiun both at Athens 1 and at Rome. 2 Ennius may have found them in his
original or insertedthem himscl£3In any casehe would have taken the word
pmlosopharifrom the common language.

9Sphilosorhaodurnest paucis: one must understand uerbis;cf. Plautus,

Aul. 1 nequismireturqui sim pauciseloquar,Terence, Haut. 10 nuncquamob
paucisdabo.
rem haspartisdidicerim

baudplacet: this and similar phrases occur at Plautus, Mere.349, Stich.

297, Pseud.653; nonplacetet sim. are much more common.

XXIX
Andromache was notoriously tall (cf. Ovid, Ars 2. 645, 3 . 777, Juvenal
6.503, Dares 12); henceAntiphon's unsuitability for the role.

XXX
See above, p. 236.
XXXI
To explainthis remark of Cicero's schobrs often adduce Donatus, De com.
8 . 11 : huiusmodicarminaad tibiasfiebant. ut his auditismulti ex populo ante
Jicerent,quamfabulam acturi scaeniciessent, quam omninospectatoribus ipsius
antecedens tituluspronuntiaretur.But Cicero is patently referring to the per-
sonage from whom the aria comes, not the play as a whole. It is worth
quoting hisfurther remark at 2 • 86: simul.injlauittibicena peritocannenagnosci-
tur. Donatus' statement may come from scholastic misinterpretation of Ac.
2. 20 itself rather than observation of stage practice.
Cicero could have in mind the opening aria of a play .4 If so it is hard
to believe that connoisseurs came to the theatre not knowing what plays
were to be performed. On the other hand, if he is referring to an aria
within a play it is hard to believe that the connoisseurs recognised what
personage was about to give utterance from knowledge of the music rather
than from knowledge of the action of the play.
1
Cf. Plato, Gorg. 484c.
:a Cf. Cicero, Off. I. 19, Fin. I. I, Tacit:us, Agric. 4.
3 For Ennius' freedom in the matter of sententiaesec below on fr. LXXXIV,
fr. CV, fr. CVIII.
4 Cf. Euripides' 'lqnywEta 1'lv AvAf6t, CPfiaoS,Plautus' Cistellaria,Epidicus,
Persaand Stichus,which all departed from the normal in having, according to
the surviving scripts, musically accompanied openings.

2.53
 COMMENTARY
These difficulties can be avoided by the assumption that Cicero had in
mind a theatrical show like that given at the funeralgames of the murdered
Julius Caesar, where certain arias from Pacuvius' Armorum iudiciumand
Atilius' adaptation of Sophocles' 'HAEKTpa particularly befitting the occa-
sion were performed (Suetonius, Jui. 84, Appian, B.C. 2. 146). On such an
occasion it might well be that those with trained can would first recognise
the source of each aria.
XXXIII
Pomponius Laetus' Arulroma(ca) Nocti ispalaeographically at least as likdy as
Scaliger's Androm(ed)aNocti and evidence exists that Varro and/or his
sources knew the Andromacha, none that they knew theAndromedaof either
Ennius or Accius.
The alleged similarity between the Latin anapaests and Aristophanes,
Thesm.1065-8 wvv~tepee I OOSµaKpOVhnrevµa 61001<EIS I acrrepoe,~a
vGn'aSacpproova'Ial&eposlepas ITOOcnµVOT<XTov 61' 'OAvµ,rov,saidby
a scholiast to be TOV1rpoA6yov-nis •Av6poi,&t6cxs elafx>Afl,is not so very
close. That Ennius gives the stars to Night's chariot rather than to theheaven
matters little. More importantly Ennius' confidssuggests that darkness is dis-
appearing; the Greek verses present the scene as still quite dark. In any case
the similarities that do exist may be accidental. The personages of Attic tra-
gedy frequently address the night: e.g. Sophocles, El. 201 ff.,Euripides, El. 54,
Hele.68, Or. 174. Comedy sometimes parodies the device, sometimes uses it
seriously: e.g. Aristophanes, Batr.1331, Pap. Ant. 1s recto 9, Menander, fr.
789, Plautus, Mere.3-5. Night's chariot might turn up anywhere; cf. Aeschy-
lus, Choe.660-1 Ta,(WE s·, 00SKai WKToS &pµ"hrefYeTat I C1KOTelv6V,
I
fr. 69.5~ µeAav{,r,rov,rpocpvycl>vlepas WKToSaµoAy6v, Euripides, Ion
1150-1 µeAaµ'Tml'AoS SEvv~acn{pc.:rrov3VY0lSI 6x1w'rnCXAAEV, Theo-
critus 2. 166arnpes evKaAOlO Kar' &nvya WKToS61fa6o{, Cicero, Atat.
189-90 hietamenaeternoinuisenslocacurriculo ~
nox Isig,uideditnautis( Aratus
408 apxat11 w~).Virgil, Aen. 5. 721 et nox atrapolum bigissubuectatenebat,
Tibullus 2. I • 87-8 iamnoxiungitequos,currumque sequunturImatrislasduosidera
'ulua choro,Anon. Culex 202 iam q:uatitet biiugesoriensErebeisequosnox.
96 caua caeli: cf. Ennius, Trag. 2s0 cauacaerula,Varro, Men. 270 caeli
cauernas, Cicero, Atat. 2s2 et latecaelilustrarecauernas(translating nothing in
Aratus ), Carm.fr. 3. s, Lucretius 4. 171, 391, 6. 252, Cicero, Atat. 313-14 et
quantosradiosiadmusde luminenostroI quis huncconuexumC4tlicontingimus
~
orbem( Aratus 541 ooaov 6' 6cp8<XAµoto ~AfiS arro-refVETatCXV)'Tl),
Virgil, Aen. 4. 4S 1 taedetcaeliconuexatueri,6. 241 superaadconuexa ferebat.The
visible sky is frequently represented in republican tragedy as a hemispherical
envelope or vault: cf. Ennius, Trag. 188-9 in altisonocaeliclipeo,319 caeli
ingentesfornias, 365~ caua. . . caelicortina,Naevius ap. Varr. Ling. 7. 7 hemi-
254
 AND RO MACHA
sph«rium ubi a,nclu,aierulaseptumstat. Such metaphon appear in late Greek
poetry: e.g. Syncsius,Hymn. 9. 87 MoS ovpcxv&'>v,Oppian, Cyn. 1.281
al6ep{o10"1yvaAOlO'l.Euripides' heroes refer often enough in oblique fashion
to the scientifictheory according to which the skywas, whatever its appear-
ance, curved in the horizontal plane and.the paths of the celestial bodies
circular (e.g. fr. S93, fr. 919, fr. 941) but there seems to be no clear case in
which they identify appearance and reality. W. S. Barrett should not have
coupled1 Euripides, El. 731 and fr. 114 with Plato, Phaidr.247c and taken
v&na as 'the top of what, viewed from the outside, is the sky's convexity';
thismakes it impossibleto explain why vcZrrais alsoused metaphorically of
the sea (Euripides, Hel. 129, 774 et al.).

XXXIV
The first four words ofVarro's quotation, Acherusiatemplaalta Orci (metri-
cally a paroemiac), head an unattributed quotation of republican tragedy by
Cicero at Tusc. 1 . 48: quaeest anustam delira, quaetimeatista, quaeuosuidelicet,
si physicanon Julicissetis,timeretis,'Acherunsiatemplaalta Ord,pallidaleti,nubila
tmebrisloa,'J nonpudetphilosophumin eo gloriari, quoJ haecnon timeat et quoJ
falsaessecognouerit?
The words of Cicero's quotation form two anapaestic dimeters (for the
fourth-foot dactyl c£ v. 87 o pater o patria o Priamidomus;the enjambement
could be avoided by placing loa, before palliJa). They make good, if incom-
plete, senseapart from the context of Cicero's discoune.Whereas the ordi-
nary appellation of the underworld in republican drama is Acheruns, the
words which Cicero quotes form a grandiose periphrasis based on Roman
officialese:loa, are the places where men go about their ordinary business,
templathose where the senate and magistrateshave dealio~ with the gods or
with men under the protection of the gods.:i Orcus, ruler of the dead,3keeps
to his templa while the dead themselves inhabit the palliJa. .. nubila tenebris
loca.For letum, 'mornri' (manes, umbrae, simulacra, etc.), c£ the set funeral

• Euripides:Hippolytos,p. 186.
a Cf. the cannendeuotionisquoted by Macrobius, Sat. 3 . 9. 7 locatemplasaaa
urbemqueeorum relinquatis. •. nostraqueuobis loca templa sacra urbs acceptior
probatiorque sit. For the same phraseology wed less precisely cf. Accius, Trag.
529-31 Volcaniatemplasub ipsis I collibus,in quos delatuslocosI dicitu, alto ah
liminecaeli,Cicero, Manil. 70 testorqueomnistkos et eosmaximequi huiclocotem-
ploquepraaident.Plautusparodiesit on two occasions:Mil. 413-14 quomme in
locisNeptuniistempli.sque turbulentisI seruauit;Rud. 906--9
Neptunohasegogratias
meopatronoI qui salsislocisincolitpisculmtis,I quomme ex suis locispulcreomatum
expediuitI templisreducem,plurimapraedaonustum.
3 C£ G. P. Shipp, Glotta XXXIX{1961), IS4 ff.

255
 COMMENTARY
formula ollusletodatusest (Varro, Ling. 7. 42, Festus,p. 304. 1 ff.), Lucretius
3 .42 Tartaraleti, Virgil. Georg.4.481-2. The word occurs frequently in
republican epic and tragedy as a synonym of mors; it hasa high-falutin tone
wherever it occurs in comedy. 1
The words which Varro attributes directly to the AnJrom«haform a
slightly irregular iambic trimeter (see above on fr. m). They can also be
takenas a paroemiac followed by an anapaestic manometer. They make quite
good sense,being a grim variant of the returning traveller's conventional
address to his homeland. 2 The speaker would be someone on the point of
death, possibly Astyanax, beholding in a vision the realm of Hades; c£
Euripides, Hipp. 1447 Kai 6t'tvep-ripwv6p&'> m'.,).~.3
Most scholars since Stephanus' time seem to have assumed that Cicero
quotes the very same speech as Varro does, some considering that Varro
omitted pallidaleti nubilatenebrislocaafter Ord, others that Cicero omitted
salueteinferaafterthisword. It is difficult to see how Varro could have omit-
ted pallidaleti nubilatenebrisloca.. If they formed one of the metrical units
recognised by ancient editon of dramatic texts we would have a caseparallel
with the omission of tibi utilisquehabere(an iambic dimeter catalectic) from
the quotation of Plautus, Cist. 8-11 at Ling.7. 99.4 But they do not form such
a unit. It is possible that Cicero omitted salueteinferain order to merge the
tragic piece into his own discourse. However the conglomeration of words
that results from this hypothesis-Acherusia templaalta Ord salueteinfera
pallidaletinubilatenebrisloca-ought not to be foisted on Ennius; inferagoes
tidily neither with templanor with loca.
In difficulties like this one it is profitable to question basic assumptions.
And indeed the assumption that Varro and Cicero are necessarily quoting the
same passage is an arbitrary one. Lucretius repeats the two-word phrase
Acherusiatempla three times in the six booksof his didactic poem ( 1 • 120,
3 .25, 3. 86). In the extant scripts of both Greek and Latin drama there recur
phrases much longer than four-word ones. Cicero does not even say he is
quoting Ennius. What he quotes should be assigned to the 'incertae incer-
torum fabulae •.

1
Plautus, Aul. 661, Mere. 483, Mil. 1241.
:a Cf. Aeschylus, Ag. 508, Sophocles, El. 67 ff., fr. 825. 1, Euripides, Htrakles
523 ff., frs. ss8. 1-2, 696.1-2, 817, Menander, frs. 1, 287, Plautus, Ba«h. 170,
Stich.649.
3 Vahlen compares Euripides, Amir. 413-14 ~ ovµt'i&avt;is I a·,dxoo,rpos
·A16flVand 501-3 xtpcxsalµQ'TT}lpas~p6xo101mv.13µwaI mµ,roµa, K«TO:
yafcxs.These passages arc not strictly parallel. The Euripidean Andromache is
only threatened with death.
4 See Leo, Die plaut. Cant.p. 6.
 ANDROMACHA

XXXV
Vahlen's notion that this fragment refersto resistanceby Andromache to
Ncoptolemus' sexualdcmands1 is highly implausible. Indeed heliumgerert
and proeliumf aart often suggest something very differentin erotic contexts.
The person with whom Andromache has been quarrellingcould only be
Ncoptolemus. The issue would most likely be the proposal to sacrifice
Polyxcnc to the shade of Achilles, Ncoptolcmus' father. Ncoptolcmus is
nowhere represented as concerned in the killing of Astyanax.
Zj)ljnger suggestedathat Cicero had the Ennian Andromache in mind
when, writing to Atticus in 6o (2. 1. s),he said of Clodia, the wife of the
consul Metellus, sedegoillamodimaleconsularem. ea estenimseditiosa,ta cum
uiroheliumgerit; nequtsolumcumMetellosedetiamcumFabio•••• Orclli and
Buecheler3are aslikely to beright in proposingone of the perpetuallyquarrel-
some uxoresdotataeof comedy.4
For the declaration that a man's name suits hischaracter, hisdeeds,or bis
fate cf. Aeschylus,Ag. 681-7 Tfs1TOT'oovoµa~ cI>S" IlsTO,rav mrrvµoos
... Tav I Sopfyaµ~pov ~,1ve1Kf\ I8' •EJJvav;. Hile.315, Sophocles,Ai.
430-1, 6.-.880, Euripides,Ba. 508, Rhes. 15~. Phoin.636-7, 6.-.SI7, Telephos
Pap. Mediol. 1. 11-13. For the rhetorical value of thus referring to the sig-
nificanceof a name cf. Aristotle,Rhet. 2. 23 . 29, Cicero, Inu. 2. 28, Q_uintilian,
Inst. s. 10. 30-1.
XXXVI
Scaliger struck out Troianoas a gloss, substituted iactarierfor what he
imagined to be iactariof the paradosisand joined the resulting trimeter with
the two quoted by Cicero, Tusc.1. 105 (vv. 7~ ). He seemsto have thought
the versesbelonged to Ennius' Hecuba.Columna compared Euripides,And,.
8-1o~T1S1T001VµAv·EK-rop· f~ "Ax~ I8av6vr"rotl6ov, ,raJ6a6" ov
TIKTCA> I ~up8wra Trvpy<A>v
1TOO'El •AaTVOOJax:r· cm·6p8fCA>V and assigned
the versesto the Latin Andromacha. But thismakes senseonly on the assump-
tion that Ennius' play was set at some time well afterthe killing of Astyanax.
DiintzerS proposed that iubentor .flagitantbe supplied with iactariand com-
pared Euripides, Tr. 725~l'fl<Xl6! wpy<A>v&Iv acpeTp<A>1l<G>v 6:rro.

' Vahlen compared Euripides, Tr. 658 ff., 673 ff., Andr. 38 ff., 111 ff.
:a Ciceround dit altromischtnDichter,p. 88.
3 Ind. schol.Grtifswald1868, 18 f. ( = Kl. Sehr.1646 f.).
4 Ribbcck, C.R.F. 1, on Com. inc. 26, compared Axionicus, fr. 6.9 olov
cp().ep(sTfs mt 1<CXl µaxETaf Tf µ01, and Menander, fr. 25i .6 Korte yw1)
KpcrrETff'CXVTCAWffl1T6rrrrt µaxn· &l(. Cf. also Plautus, Mm. 76s ff., Ovid, Ars
2.155.
s Zeitschr.J.d. Alt. 1838, p. 49.
17 257 JTO
 COMMENTARY

100Hectoris natum: substantival gnatus and gnata are normal in the

manuscripts of Plautine and Terentian comedy; 1 even in the indirect and
much modernised tradition of the fragments of tragedy these forms are as
common as natus and nata. It is therefore unlikdy that Ennius wrote natum
here.
Outside the vocative the substantive gnatushad a markedly lofty tone; in
comedy it occurs 131 times as againstfilius 318; in tragedy on the other hand
it occurs 16 times as againstfilius only once.

de Troiano maro: there is no need to suppose that Servius auct. Aen.

3 . 489 (Cakhascednitdeiidendum
ex murisAstyanacta
. .. praedpitauite muro)is
dependent on Ennius; Vahleni ought not to have approved ofLacbrnann's
alteration of de to e.
The use of the adjective Troianoinstead of the genitive Trouiedevates the
style; cf.Ennius, Trag.124 Bacchico. .. modo,144 matemo sanguine,208 nemore
Pelio, 218 domumpatmulm, 237 erilis... corporis,271 humanumgenus, 323
hostilimanu,Plautus, Mil. 13-15 quemneegosnuaui in campisCurculionieis I
ubi Bumbomachides I erat imperatorsummus,Neptuni
Cluromestoridysarchides
nepos.
XXXVII
For doubts about the attribution to Ennius' tragedy see above, p. 234 n. s.
Mussare,'murmur, express displeasure quietly'3 probably stood in some
form among the words quoted.

XXXVIII
Augificareoccurs only here in extant Latin. Lucretius 2. 571 hasthe adjective
auctificus.
The verb is probably a coinage on the model of magnificare. The
great frequency of the latter verb, aedificare,
ludificare
and sacrificare
in comedy
shows that such formations were not in themselves poetic; but the republi-
can tragedians, in their search for diction distinct from that of every day,
concocted several new ones. Pacuvius hasamplificare, Accius orbificare.

XXXIX
This fragment looks like part of a lament either by Andromachc or by a
chorus of Trojan fdlow captives. Mercerus compared Euripides, .Andr.111-
& 6co<pvaµ01 Kcrrif:xxX()OOS,
12 'Tt'OAACX &v(K"lAe1,rov I 60'"t'VTe Kai
8<XAaµovs Kai ir001v w Kovfcns.

1
Sec Kornn,Altlat. Forsch.p. 132. 3
E.P.R. •. p. CCIII n.
3 Possibly the Greek µv3t:1v;cf. Lewnann, in Mllanges]. Marouzeau(Paris,
1948), 381 f. ( = Kl. Sehr.164 f.).
 ANDROMACHA
103 qaantis cam aaumnil: contrastTc:renc::c, Hee.8']6ex quanld oerumtUI
extroxeris.The preposition cum rardy disjoins a noun from its prcc:ediog
dctcnninant in republicandrama; I can find only Plautus,Amph.245, Ann.
37-4.Men. 895, Most.658, 977, Persa54, Caccilius,Com.59, Turpilius,Com.
168, Accius,T,ag.216. In the rclativdy few fragmentsof Ennius' epicon the
other hand there are eight examples (51~ 54, 77, 201, 338, 381, +u, 540).
~antis must therefore have receivedan unusuallyheavy emphasis.

illum e:unclaai diem: the ironic flavourof Ennius'phraseis obscuredby

suchparallds as Euripides,T,. 433 Sa<ayapwrA,')aas ht], Cicero, Cann.
&. 22. 27 tot nosad T,oiambelliexanclabimus ~ n.
annos( Homer, 2. 328 &s
flµetsToaaao·r-rna Tn'OAEµf~oµev cx00t
). Apuleius,Met. 6.11 tetranox
exanclata. In the mindsof Attic theatre-goerswrAEtvet sim. were associated
with sufferingwhen usedmetaphorically; the source of the metaphor was
perhapsa nautical usage,'draw offbilge-water'. 1 To judge by Livius, T,ag.
30Jloremanclabant Cann.fr. 36 camisuinumquequodlibabant
Liberiex carchtsiis,
~
anclabatut( Od. 23. 304-5 ~ Kai lq>1aµ~Aa I laqxllOV,,roAA05&
m&wvfl~ olvos). Plautus, Stich.272-3 ne iste edepoluinumpoculo
pauxilluloI saepeexanclauitsubmerumsdti.ssume" a non-metaphoricaluse of
anclare~t sim., associated
with feastingand pleasure,was the one most familiar
to Ennius' audience.Anclaremay have been a word of the sacral language
borrowed from some South Italiandialectof Greek.3 If this is the case,there
was a marked oxymoron perceptibleto the first hearersof the words quantis
cumaaumnisiliumexanclauidiem.
XL

104-5 longinque a domo Ibellamgerentes: contrastCicero, Manil.32

fuit propriumpopuliRomanilongea domobellare.Theadverb longeis common
doesnot appearagainafter
in republicandramaand classicalprose; longinque
the present passageuntil Fronto, p. 132.2.
105 bellurn .•. summum mmma indaatria: sumnuzindustria and similar
phrases arc common; c£ Plautus, Most. 348 summisopibusatqueindustriis,
Caesar, Ciu. 2.4.1, Cicero, Brut. 165, Att. 2.22.3, Fam.13.10.3. Summum
heliummust be a momentary creationfor the sakeof word play; Caesarand
Cicero use repeatedlythe phrase nuzgnumhelium.
The use of the one adjectivetwice within the one phrase is common in
1
Sec F. Solrosen,Btitriigezurgmchischm Wo,iforschungI (Strasbourg, 1909),
pp. 184 f.
:a Sec also below on v. 144.
3 Cf. anclabrisin the Roman sacrallanguage (Paulus,p. 10. 18) and &v'f ).ttTp{a
in the Athenian (Schol Luc. Dial. tko,. 2. 1, p. 276.4 Rabe).
259 17-2
 COMMENTARY
Attic tragedy (cf.Euripides,El. 337 and the enmples collectedby Denniston
oJI«.), in Romm tragedy (cf. Ennius 128, 264, Accius 134, 2o6 [quodre in
summasummumessearbitror),655) and the more elevated parts of Romm
comedy (cf. Plautus, Cist. 288, 308, Pom. 1270, Terence, Haut. 4-5, Phorm.
587 et al). Vcry frequently one of theadjectivesis usedeither redundantly or
in a fashion removed from common usage.

XLI
Thisfragment appearsto come from a messenger'saccount of thefuneralof
Astyanax.Dead warriors were commonly carriedon theirshields(cf. Virgil.
Atn. 10.506, 841) and the Andromache of Euripides' Tf>Ct>&&s (1133 fI)
asked for Astyanax to be entombed in his father's shield;his corpse, how-
ever, was washed in the waters of the Scamander(1150 fI).

1o6 loamt: for the normal collocant.Plautus hasthe simple verb four
times in militaryor para-militarycontexts (Amph. 351, Cure.25, Potn. 612,
Rud. 474) but elsewherein republicanLatin it isrestrictedto legaland fuian-
cial contexts.

107 clipeo: clypeowould be an impossiblespellingfor Ennius' day. The

manuscripts of Plautus have clipeusfor the most part with an occasional
clupeus.D at Trin. 719 hasclypeus.
The clupeuswas a circularbronze shield which covered everything from
neck to~ used according to Livy (1.43 .2) by the first classof the old
Servian army but long replaced by the oblong scutum. The adapters of
both tragedy (Ennius 189, 370, Trag. inc. 62; c£ Pacuvius 186 clupeat)and
comedy used the word to denote the Greek ao-rr(s. Scutum is absent from
tragedy and comedy hasit only in special contexts: Plautus, Batch. 72
(pun); Plautus, Trin. 1034 and Turpilius, Com.40 (moralisingwith reference
to the interests of the spectators).

XLII
The speaker may be referringto the insincerity of public orators (cf.
Euripides,Helt.130 ff:,254 ££).His metaphor is from the testing of pots for
hidden cracks: cf. Plato, Theait.179D St<XKpOVOVTa dTE vyt~ EiTEaa8pov
f&fyynm, Lucretius3 . 870-3 proindeubiseuideashominemindignarier ipJUm,I
post mortemJore ut aut putescatcorporeposto I autflammis inttrjiatmalisue
ferarum,I scirelicetnonsinarumsonere.
Sonerewas already restrictedto tragedy and epic in the early second cen-
tury. Plautus hassonareseven times.

200
 ANDROMA·CHA

xun
Lipsius' alteration of the corrupt trabemto trahtnsseems to be generally
accepted (c£ Ovid, Met. II • 709 attonitogtmitus a corde tralnmtur et al.).
Vahlcn scannedthe fragment in his second edition as a trochaic pentameter.
Hephaestion,6. 2, cites an example of this vcnc from D.lHrnachus.There is
no evidence of its existencein the extant scripts of Attic tragedy. Bentley
thought he could find three examplesin the comediesof Terence (Eun.293,
Phom,. 194-, 485). Bergk1 and Vahlen2 thought they could add several
examplesfrom republicantragedy. None are convincing nor, in the state of
the evidence, could they be.3
109 sed quasi aut ferram aut lapu: c£ Aeschylus,Prom. .i.p at6flpo-
n
q,p<AW K&Kffhpas elpy<XaµM)S, Euripides,Med.1279-81 ~mv• &p• ws
~ nnpos fl at6alpos,ans•.• KTEVEis.
110 rarenter: once elsewherein tragedy, twice in comedy (Caecilius135,
183); a poetic coinage, perhaps modelled on the antonym.frequenter.R.aro
was the adverb normally used (4 times in Plautus, twice in Terence).
A number of adjectivesin -us have adverbs in -ter recorded fairly often in
both tragedy and comedy but most suchformationsoccur only in tragedy or
at the ends of iambic and trochaic verses in Plautine (not Terentian)
comedy.
XLIV
The other two pieces in Macrobius' ueliuolusarticle do not form metrical
units but a strong tendencyto quote suchunits is observableelsewherein the
discussionof Virgil's allegedlyborrowed words. R4pit ex altonauts ueliuow
can b(- scannedas an anapaesticdimeter if nauesis treated as a monosyllable
(c£ Plautus, Bacch.797, Mm. 344).
111 ueliuolas: this compound occurs elsewhere at Ennius, Ann. 388,
Laevius, fr. II, Lucretius 5.1442, Virgil. Am. 1.224-, Ovid, Pont. 4.5.42,
4. 16. 21. No other compound in-uolus(uolare) seemsto occur in republican
or classicalverse. Pliny hasaltiuolus
twice (Nat. 10.42, 10. 130). -uolus(uelle),
apart from maleuolus and bentuolus,is almost as unproductive (Catullus has
multiuolus(68. 128] and omniuolus(68. 140]). There seem to be no adjectives
in -m-n'is ('Tm'Oµa1)in Ionic epic or Attic tragedy except V\J'l'TmTIS-
1
Zeitschr.J.d. Alt. 1855, 294, Philologusxxxm (1874), 3o8 ( = Kl. phil. Sehr.
I 114, 373 f.).
~ Ind. ledt. Berlin 1888/9, 18 f. (= Op. ac. 1419 f.}, SB Berlin 1901, 348 ff.
(= Ges.phil. Sehr.n 610 ff.), Ind. ledt. Berlin1902/3, 4 ff. ( = Op. ac.n 449 ff.).
3 Cf. Timpanaro, SIFC N.S. XXI (1946), 77 ff.
261
 COMMENTARY

ANDROMEDA

The title Andromeda is given to Livius by Nonius; to Ennius by Vcrrius,

Nonius and Priscian; to Accius by Nonius, Macrobius (Gramm.v 606.38)
and Priscian. Most ofNonius' quotations of the Accian Andromeda come from
Lindsay's list 8 'Accius ii', a few from lexicographical sources. Two of his
quotations of the Ennian tragedy appear to come from list 27 •~ph. verb•
and two from 28 •Alph. adverb• ; that at p. 20. 19 headsa lemma following
one illustrated by an •Accius ii• citation and preceding lemmata illustrated
from the same source; it appears to have displaced the present extra-
citation, which comes from •Accius ii• and illustrates the lemma somewhat
better; one can only guess at its source.
The three sets of fragments can be interpreted so as to dealwith the rescue
by Pcrseus of Andromcda, the daughter of Ccpheus and Cassiepeia.from
a sea-monster. Scaliger's treatment of the text of Varro, Ling. s. 19 (see
above on fr. xxxm) has enforced among all students the belief that Ennius
adapted the famous tragedy produced by Euripides in 412 B.C. 1 The title
•Av6poµt6a is alsorecorded against the namesof Lycophron, Phrynichus
and Sophocles.
There is no clear evidence that any of the threeLatin tragedies were known
outside the ranks of the lexicographers after the second century B.c. At Fin.
2. 10.s Cicero translate,<1
birosc1fa gnomic verse of Euripides'•Av6poµt&x
which frequently turns up in extant Greek literature (fr. 133) and probably
was quoted in his philosophical source.
Ribbeck 3 suggested that the words quoted by Festusat p. 488. 23 probably
as from Ennius (fr. cCVI}-4spe,asaxa tescatuor--were spoken by Perseus
when he arrived on stage and saw Andromeda chained to the rock; that the
words quoted by Nonius at p. 197. 3 as from Ennius (fr. ccxm)---et quisillaec
estquaelugubrisucdnctaeststolo-were spoken by Perseus when Andromeda' s
mother arrived on stage, that the words quoted by the Danieline Scrvius in
his note on Georg.1. 12-13 (fr. cCXIX)--ageroppletusimbriumJremitu-were
spoken by Andromcda to Perseus, describing the floods which Poseidon
had sent upon her land (ApoJlodorus 2.4.3 .2).
Vahlen3 set the tetrameter attributed to Ennius by Varro at Ling. 7. 16 (fr.
OCCI )-ut tibi TitanisTriuiadednitstirpemliberum-in a plea by Andromeda
to Perseus to take her away with him as his consort despite the wishesof her
father.

1
For Alexander's fondness for quoting this play see Athenaeus 12. 537D.
3
~t. seen.p. 261, Die rom. Trag.p. 165.
3 E.P.R. 1, p. 138.
 ANDROMEDA
The content of none of thefour fugmcnts in questionis ,,«i6c mough to
allow either acceptanceor outright rejection of the modem attributions.'

XLV
W dcker interpreted this fragment as from a speech by Andromeda
when Cepheuswent back on hispromise (Apollodorus2. 4. 3. 4. Ovid, Met.
4. 701 f[) to give herin marriage to Peneus. The legalformulae employed
by Enniusseemto me quite inappropriatein themouth of a woman wishing
to cohabit with a man againstherfather's will. They applied properly only
to matrimonium iustum,not to concuhinatus.
I thereforeacceptHartung's vjew3
that the fugment is from the speech in which Cepheusmade his original
promise.

112 liberum qaae.endam cama: Ennius'vocabularywasperhapsexactly

that employed in contemporary Roman betrothals; cf. Plautus, Capt. 889
libtrorumq,uintndorum causaei aetlouxordatast,Varro ap.Macr. Sat. 1. 16. 18
uxoremlibtrumquaertndorum C4USO ducerertligiosumest, Suetonius, Jui. 52,
Gellius4.3 .2, 17 .21.44- Paralld phraseologymight have existedin Ennius'
original; c£ Euripides,Atulr.4 Saµap 6o8etaa ,raiSoiro1bs •EK-rop1, Hera-
lJeidai523-4. Or. 1o8o, DemosthenesS9 .122 TQS µ!v yaphat~ f\6ovfis
MK·fxoµev,TOS SeiraUaKQS Ti\sKa6.f\~ &parreias TOO owµaros,
TClS Se ywatKQS TOV,rai6cnro1Elaecn yvt1aioosKai TOOVwSov cpvAaKa
mcm'iv fxe1v,Mcnandcr, Dyslc.84,2-3&A>.'fyyv&>,ra(&.,v m' aparft>
yvt1aioovI T'l'\v8vyartp• i\611µe1p&K16vao1, Perilt.435-6, fr. 682.
Ennius' morphology must have had an archaic tone even in the second
century. Plautushasthe genitive liberumonly once (Most.120 (baccbiacs])as
against libtrorumfive times (Capt. 889, Cist. m, Men. S9, Most. 121, Poen.
74). Apart from the first person singular present indicative Plautus has
q,u,esereonly at Bacch.179 (parody ofJegal language; c£ 181 auulatum amore
uinctumque ).

familiae matrem taae: cf. Gellius 18. 6. 8 matronamdictamesseproprie

quaein matrimonium cumuiroconuenisset,quoadin eomatrimonioffldntTtt,etiamsi
libtrinorulumnatif orent;dictamqueita essea matrisnominenon adeptoiamsed
cum spe et ominemox adipiscendi:unde ipsum quoquematrimoniumdicitur;
matremautemf amiliasappellauun esseeamsolatnquaein maritinumunumdpioque
1
With fr. cca C. 0. Mueller compared Euripides, Med. 71•-1s; with fr.
ccxm La Pcnna,Maiav(19s2), 9S, compared Aeschylus,Choe. 10-12 (without,
however, asserting that Ennius adapted Aeschylus' play).
2
Dk griech. Trag. p. 661; cf. Vablen, E.P.R. 1, p. 137.
3 EuripidesRestitutus 11, p. 353.
 COM-MENTARY

tsstt, quomamnon in nuurimonium tantumsedinf amiliamtJUOqMe mariti tt in sui

hertdis locumumisstt. Comedy bas thephrase mater f amiliasfive times in very
formal contexts (Plautus, Amph. 831, Mere.405,415, Stich.98, Terence, Ad.
747). It is odd that Enniw wed the formfamiliat in such a context as this but
familias..• tu« may have sounded inelegant to second~tury can.

XLVI
This fragment describesthe sea-monster; cf. Plato's description of Glaucw at
PoliteiaI0.611D ol TOV8CXA6:rnovD.cxOKov 6poovres OVK&, m ~S(oos
<XVTOV t6o1ev'TT}Vapxa(cxvcpv<JlV V'll'OTOVTQ'TI! 1TCXAata TOVaooµ<rroS
~ ... MA~a8m v,ro TOOV KVµaroov, lfXAa & irpcxrneq,Vldvm,
6o-rpea'TI! Kaiq,v1daKairnpas. For the second dcment of the descrip-
profandumI extantematqueumerosinnato
tion cf. Ovid, Met. 1 . 331-3 supraque
muria ttctum I caerukumTritonauocat,4. 725-7 nunc tergacauusuptr obsita
conchisI nunclaterumcostas,nunc,quattnuissimacaudaI desinitinpisctm,falcato
uerberattnse. It must come either from Andromeda's account of events lead-
ing to her encbaiorocnt or from a messenger's repon of Pcrsew' encounter
with the monster.

113 k!i'upeo inuestita axo at:que ottreis tquun excrabrentf: for the
sound play cf. Trag. me.74 per speluncasS4xis structas asperisptndentibus,
Virgil, Atn. 6.471 si Jurasiltx aut stet Marptsiacautes;see above on v. 4,
Scrupeusoccun elsewhere at Pacuviw, Trag. 310 and perhaps at Accius,
Trag.431. Plautus, Capt. 185, Pacuvius, Trag.252, Lucrctiw 4.523 have
scruposus.Neither word seemsto occur in classicalprose. It is an odd epithet
for saxum.Adjectivesin -eus indicating the materialof which a thingis com-
posed occur fairly often in both comedy and tragedy. Plautus wcs the
formation for comic effects (e.g. ocultus,pugneus,stimultus,uerbereus)
but the
remains of tragedy have only one other neologism of the loose type affected
by theclassicaldactylic poets :1 caauleusat Enniw 26.
The compound inuestireoccurs dscwhcrc in Latin only at Maccenas, fr.
ap. Sen. Epist.114.5 and Seneca, Here.0. 381. Comedy has utstirt9 times.
Scabrere seemsto occur only here and at Pacuviw, Trag.314, A number of
similar formations occur in tragedy and arc absent from comedy: e.g.
clarert,jlaccere,ftondere, languere~nigrere,pigrere,putrere,senere,stupere,
tabere.
XLVII
I have found no plausible explanation of this fragment.
1
Cf. in contemporary epic Ennius, Ann. 577 populeafruns.
 ANDROMEDA

XLVIII
Ennius' trochaic tetrameter comes from a messenger's report of Peneus'
encounter with the monster. For the structure and content of the sentencecf.
Euripides, I.A. 1578-9 lspevs & cpaoyavovJ\~v hmi~o:TO,I J\cnµov
T' hreoxomte•, Iva ,rJ\1')~ &v.The Euripidcanmcs.,engcr useduimeters
(cf. fr. 145).
115 corpus: here, properly, 'living body'; soJT1ctimes
in high poetry
'cadauer' (e.g. &mus, Trag.138, Accius,Trag.323, 6ss,667;c£ OWSACX).
contemplatur: the active form of this verb was normally usedin both
tragedy and comedy. The deponent form is guaranteedby the metre here,
Plautus, Poen.1129, Terence, Haut. 617, Phonn.210. It is difficultto see any
semantic, stylisticor chronological pattern in the phenomena.

corporaret: 'that he might tum the beast's corpusinto a aulauer'. Accius,

Trag.604 usesEnnius' neologism as a simple synonym for interfice,e.Words
like iugulare(iugulum)and obtruncare (truncus)would have provided models.
Plautus invented words in similar f.ashionfor comic effect: cf. Capt. 84 dum
ruri ruranthomines,Men. 105 domidomitussum, Mil. 34 ne dentts dentiant.

XLIX
C£ Euripides, Hek. 28-9 t<ETµcn s·trr· aKTalS,l'i>:A<n'h, ,r6VTOVaa>..~, I
,ro~ots St<XVAOIS KVµo:roov~v~. Virgil. Aen. 6. 362 nuncmejluctus
habttuerS11ntquein litoreuenti,Lucan 8. 698-9litoraPomptiumftriunt truncusque
uadosisI hueilluciactaturaquis.The fragment may come from the messenger's
report of the encounter. If so Jeramcan hardly be correct; only part of the
beast would be carried back and forth by the tidal current. Bergk'sf era•is
ludicrous.

116 runus pronus: cf. Terence, Hee. 315 trepidari

sentioet cursarirursum
prorsum.

of tidal movement cf. Pliny, Nat. 2.213, 9.176,

reciprocat: for reciprocus
16.170; for reciprocarecf. Cicero, Nat. deor. 3.24, Livy 28.6.10, Pliny,
Nat. 2.212, 2.219 et al.

tferamt:
comedy hasbtluaand btstu, oftenbut not/era. Tragedy on the
other hand hasf era7 times, btluatwice and btstianot at all.
1
RhM m (1835), 70; cf. Ribbeck, Die rom. Trag. p. 169 n. 191, Vahlen,
E.P.R.1, p. 136.
 COMMENTARY

L
This fragment clearly comes from the messenger'sreport.

117 alia: 'reliqua'; c£ Plautus, Amph. 12 et al. It is difficult to sec the

point of the disjunction from membra.

diifertdiaupat: c£ Lucretius 1 .651 disieaisdisquesipatis,Cicero, Ltg. agr.

1 . 2 disperdat
acJissipet,De orat.1 . 187 dispersa
etdissipata,
Ltg. 2. 42 Jistractiac
Jissipatiiacent,Nat. Jeor. 2.41 Jisturbatac dissipat,Caesar, Gall. 2.24.4
s.58. 3 dispersiacdissipati,Livy 2. 28. 3 Jispersamet Jis-
diuersosJissipatosque,
sipatam.Contrast Accius, Trag. 587 dissipentet disturbent{text dubious).
Tragedy has Jissupare3 times, comedy only once (Plautus, fr. 171).

118 uisceratim: 'with bits of B.eshhanging'. The adverb docs not occur
elsewhere in Latin but is not necessarilya poetic coinage. Guttatimand
articulatim
are the only exactly paralld formations in tragedy. It may have
bdonged to the vocabulary of butchers; c£ Plautus' assulatim(Capt. 832,
Men. 859), frustillatim(Cure.576), offatim (True.613). Caecilius' ossiculatim
(Com.so), Pomponius'.ftustatim(Atell. 177).

maria: a poetic plural. At Plautus, Rud.1, Trin. 1087, Terence, Ad. 790
mariais a genuineplural. Neverthelessthe context hasin eachcasea height-
ened tone.

spumant sanguine: c£ Virgil, Aen. 6. 87 Thybrim multo spumantem

sanguinecerno,9.456 spumantissanguineriuos.

LI
These words were obviously spoken by Andromeda but it is difficult to
imagine a context. W dcker put them in a speechby Andromeda trying to
persuadeher father to reversehis decisionabout Pcrseus,Hartung in a speech
announcing her own decisionto go with Perseusdespiteher parents' wishes.
Webster suggests that they are part of an apostrophe to Cassiepciain
Andromeda's prologue speech ( .. .audelismater).

119 filiis... obiecta sam . .• Nerei: c£ Virgil, Aen. 10. 88-9() nosnetibi
jluxas Phrygiaeresuertere fando I conamur?nos.an miserosqui Troas.khiuis I
obiecit?No cleardistinctionis made between theNereids, cheirwrath, and the
1
BICSL xn (1965), 29. C£ Columna, Q. Ennii Frag.p. 410.

266
 ATHAMAS

monster sent to avenge than. V ahlen interpreted filiis Ntrd as in gratiam

Nereidum,•den Nerciden zu licbc• and compared Seneca, Tro. .248 tuam
natam parensHelenae immolasti.
Plautus hasfiliisfeminine at Pom. 1128 and Stich.s6?. Filiabusoccun no-
where in republican drama.

ATHAMAS

The title Atlutmasis given eight times by Nonius to Accius 1 and once by
Charisius to Ennius.a The five trimetcrs which Charisius quotes were once
regarded with suspicion becauseof the siroHarity of theirstructure to that of
imperialtrimcters. 3 It is worth pointing out that Charisius quotes a large
amount of indisputably republican verse in theircompany.
Cicero refers three times to the madnessof .t\tbamas,4 at least once, it
seems, with the theatre in mind.
It is impossible to make out with certainty the relationship of either
tragedy to the saga of Athamas, his three wives and three sets of twin
children.S Ennius' play concerned in some way the orgiastic wonhip of
Dionysus. One of Athamas' wives, Ino, is reported to have looked after
Dionysus as an infant and on another occasion to have been discovered by
Athamas, aftera long absence from home, consorting with bacchant women
6
on Mount Parnassus.
The title •A8aµasis recorded against thenames of Aeschylus, Sophocles
and Astydamas.
LII
The five trimcters must come from a messenger's speech desaibing a
Bacchic orgy. It is interesting that males are alleged to have takenpart in the
orgy. The historical Boeotian orgiastic cult of Dionysus was exclusively

1
Pp. s6.13,31s.19, 323.34,416.13,470.27, 488.36, soo.3, s~.21. All
these quotations seem to come from Lindsay's list 8 'Accius ii'.
3
The immediate source of pp. 3 u.. 3 ff'. seems to have been the •AfOPl,ICX{
of
Julius Romanus.
3 Sec Bothe, P.S.L. voL v, p. 38, Wclckcr, Die griech. Trag. p. 1374, F. A.

Lange,~- mm.pp. 16, 30, B. Schmidt, RhMXVI (1861), S99· The objec-
tions of these scholan are answered by L. Mueller, Q. Enni Reliquiat,p. 238, and
F. Skutsch, RE v (190s), 2619. S8 ff.
4 Har. resp. 39, Pis. 47, Tusc. 3. n. Cf. Apollodorus 3 .4. 3, Ovid, Met.

4.471, Fast. 6.484.

s Sec Robert, Die griech.HeldenSdgen i, pp. 41 ff.
6 Cf. Hyginus, Fab. 4 (an account of the hypothesis of Euripides' •t~ ).
 COMMENTARY

female except for the priestly leader until very late in antiquity.1 Plautus
regarded the contemporaryItalianform of the cult as a peculiarlyfeminine
bwiness? Such allegationsof male participation as are found elsewherein
literature (e.g. Euripides, Ba. 222ff.; c£ Livy 39.8.5} are ofa hearsay
character.The messengerof Euripides' BaKxcnexpresslyrefutesthe sus-
picions of Pcnthew {680 ff.).

120-1 hiserat in ore Bromius, hisBacchuspater, Iillis Lyaeus: for the

many names of tu6wcros c£ Ovid, Met. 4.11-17 turaquedant Bacchumque
uocant BromiumqueLyaeumque I ignigenamquesatumqueiterum solumque
binuurem;I additurhis Nyseus indetonsusque ThyoneusI et cumLenoeogenialis
consitoruuae I NycteliusqueEleltusqueparenset lacchuset Euhan, I et quae
praeterea per GraiasplurimagentesInomina,Liber,habes,Arrian, Anab.s.2. S.J/.
Attic tragedy has the names Bp6µ1cs(Aeschylus,Bum. 24 et al.) and
Bal<xoS {Sophocles,0. T. 211 et al.); Ava.Jeshowever seemsnot to occur
before Leonidasof Tarentum {A.P. 6. 154 .1 ). Republicandrama elsewhere
hasBromius{Plautus,Men. 83s) and Bacchus{Plautus,Men. 83s [?],Pacuvius,
Trag.310, 422) but not Lyoeus.
For Bacchuspaterc£ Horace, Carm. 1. 18. 5. Pater was part of the cult
title of many very old Roman gods.3Enniw had the Roman Liberpaler
in mind. Zai ,rarep occun often enough in Attic drama but BaKxe
,rarep never.
Foreratin oreBromius cf.Terence,Ad.93in c,rest omniJWPUIO(sc.At"hinus),
Cicero,Rep. 1. 30 in oresempereratilledelphigeniaAchilles,Ltg. 1.6 adeum,
qui tibi semperin oreest,Catonem. .. uenias,Tusc.1. 116 Harmodiusin oreestet
Aristogiton,Livy 9. 10. 3 Postumiusin oreerat,eum laudibusadcaelum ferebant.

121 aitis inuentor sacrae: c£ Euripides,Ba. 279 (:xYrpVoS vypov ,rG,µ'

11vpe,378-86, 421-3, 534-S, Acciw, Trag. 240 Dionyse. .. uitisator,Ovid,
Am. 1. 3. n uitisquerepertor.
The noun inuentoroccurselsewherein republicandrama only at Terence,
Eun. 1034-5 o Parmenomi, o mearumuoluptatumomniumI inuentorinceptor
perfeaor,a highly coloured passage.Tragedy and comedy employ such
formationsfreely,Plautinecomedymuch more freelythan Terentian.Their
presencein the officiallanguage(praetor,imperatoret al.) lent them a certain
dignity of tone.

r SeeM. P. Nilsson, The DionysiacMystmes of the HellenisticandRomanAge

(Lund, 1957), pp. 14 f.
a Cf. Amph. 703, Aul. 408, Batch. 53, Cas. 979-80, Mil. 851-6, 1016; most of
the contexts are Romanising ones.
3 Cf. Gcllius 5. 12. 5, Lactantius, Diu. inst. 4. 3. 11.

268
 ATHAMAS

For uitis•. .Sdaaec£ Euripides,fr.76s

olvav&aT~I TOVlEpOV~pw,
~
Horace, Cann. 1 . 18 .1 nullamVaresacrauitepriusseuerisarborem( Alcaeus,
tt. 342 Lobel and Page µ116"Iv &:M.ocpvmia,Js irpcmpov Sa16p1ov
&µir0.6> ). Ennius' adjectiveseemsto have soJnctbingof the senseof 1€p6s,
'supernaturally powerful' (cf. Hcsiod, Theog.93, Euripides, Hipp. 1206);
it normally referredto thingswhich already belonged to a god.

ID tam puiter feahan eahimnt: the context of Charisius' quotation

demandsthe insertionof the cry euhoe(evol; cf. Sophocles,Tr. 219 avcrra-
paoot1 evolµ' 6 1<1cro6s, Aristophanes,Lys. 1294) in some form. Eulumcould
be taken as a cry (evav; cf. schol.Eur. Plwin.656 ywaielv evio1s... TOevol
evavhrup6eyyoµa,ais) though elsewhere in Latin it occurs only as an
epithetof thegod (Lucretius5 . 743, Ovid, Met. 4 .15 ). Euhiumlikewisecould
be a cry (cf. Ovid, Ars 1.563 parsclamant'euhioneulwe', Persius1.101-2 et
lynam m«tUU jlexuracorymbis I'euhion' ingeminat}as well as an epithet of the
god (cf. Euripides, Ba. 157 e<naTOVe<nov ay<XAAoµEVCX18e6v, Lucretius
5.743, Cicero, Fl«c. 6o, Horace, Cann. 1.18.9, 2.11.17). Vahlen under-
stood Euhan(' euhoeeuhoe') Euhium. .. inibat(Fabricius'supplement}as the
same type of phrase as Aristophanes' TOV'Apµ661ov~O"ETat (Ach.980 ). 1
This leavespariteran extraordinarilylong way ttom its verb. For this reason
.Rucchder's supplementeuhoecantam is worth considering.One could keep
Fabricius'supplementeuhoeeuhoeand alter euhan to euhans(ev~ea>v;cf.
Catullus 64.391, Virgil, Aen. 6.517); the cries euhoeeuhoeeuhiumwould
then stand outsidethe constructionof the sentence (cf. Sophocles,Tr. 219 ).

123 ignotm iuuenam coetm: there is no need to emend here. lgnotus

is to be taken in sensewith iuutnum.For cnallagein republicantragedy see
above on v. 57.
Ignotuscan be taken as 'unknown to the speaker' (c£ Plautus, A.sin.494,
Capt. 344, 542, Rud.1043, 1044) or 'unknown to society at large, of low
socialdegree' (c£ Terence, Plwrm.751).Dionysuswas a notoriously demo-
cratic deity; c£ Euripides,Ba. 421-3, Aristophanes,Batr.405 ff.,Livy 39. 8. 3
Graecusignobilisin Etruriamprimumutnit . •. sacrificulus
et uates,39. 17. 6 capita
autem coniurationisconstabatesse M. et C. Atinios de plebe Romana;E. R.
Dodds, Euripides:Bacc~ (Oxford, 1900),pp. 127 £
luUfflisoccursonly once elsewherein republicandrama (Trag. inc. 210 ).
Comedy has adulescensand adulesc.entulus commonly; neither of these
words appcan in tragedy. Cicero and Caesararc similarly shy of iuuenis.3

• Cf. also Demosthenes 18.260, Catullus 64.2ss,Juvenal 6.636, where the

bacchic cry is treated as the internal object of the verb.
i Sec B. Axelson, in Melangts]. Marouztau,7 ff.
 COMMENTARY
The word perhaps belonged properly to the tecbuical language of the
military levy.
Coetusoccurs only once elsewhere in republican drama: Plautus, Amph.
6s7, of a militaryencounter; it is common in clamcal Latin of an assemblyof
persons (often an illegalone, as opposed to a contio summoned by a
magistrate).

I
123-4 altema uice inibat: 'the members of the malegroup were joining
the female group one replacing another'. Cf. Manilius1.2s8-g quaemedia
obliquopraednguntordinemundumI solemquealtmsisuidbuspertemporaportant.
The accusative uicemis regular in comedy except at Plautus, Amph. 334,
where editors follow Scaliger in altering the transmitted uods . .. uice.
lnireis usedmetaphorically everywhere elsein republican dramaexcept at
Trag. inc. 130.

12.f alacri, • •• illlultam: cf. Cicero, Att. 1 . 16. 7 al«ris exsultatimprobitas

in uictoria,Livy 21 . 42. 3 alacer . ••exultans.The adverb alacriterdocs not
occur before Frontinus.
lnsultareof physical movement occurs elsewherein republican dramaonly
at Terence, Eun. 28s. It seems to be quite absent from clamcal prose. The
simple verb saltareoccurs 13 times in comedy, not at all in tragedy.

Bacchico •.. modo: 'Baccharum modo'. For the dance of the female
bacchants cf. Sophocles, Ant. 11so-4, Euripides, Ba. 166-9, 748, 862-72,
1090-1, fr. 7s2.

CRESPHONTES

The title Cresphontesis given to Ennius six times, once by Macrobius, once
by Gellius,twice by Verrius and twice by Nonius. Five of the fragments
quoted illustrate verbs strange to classicalLatin either in their form or in
their usage. The sixth is allegedby Macrobius to come from a speech on
which Virgil modelled that of the mother of Euryalus at Aen. 9.481-97.
The author of the rhetorical treatiseaddressedto Hcrennius quotes a dialogue
in trimetcrs betweenthe wife of a Crcsphontes and her father and an appeal
by the Crcsphontes (so one part of the tradition) of a play by Ennius for
peace and reconciliation. Ckero gives no sign that he knew any Latin
tragedy about a Cresphontes; at Tusc.1 • 11 s he translateshimselfsome tri-
meters from Euripides' Kpeacp6vnis (fr.449) whichhe found in his source.1
1
Probably Crantor. The trimeters arc quoted frequently in Greek pbilO:..
sophical writing.
270
 CRESPHONTES
In the stories connected with heroic Messeniatwo persons appear bearing
the name Crcsphontcs, one the Heradid who gained the kingdom by deceit
in the famous lottery, the other his son whose recovery of the kingdom
from the usurper Polyphontcs provided the theme of Euripides' tragedy
l<pEc7q,6v,,,s.1
This play was admired by Aristotlc2and still had readers in
the third century A.D.3 The only clear reference to a tragic presentation of
the dder Crcsphontcs seemsto be at Ammianns Marccllinus28.4.27.
Most students have interpreted the Latin remains as bdonging to a version
of Euripides' Kpeacp6V'n}s. 4 RibbcckS and L. Mucller6 on the other hand

reconstructed Ennius' play to deal with the dder Crcsphontcs.

Three fragments causedifficultyfor the orthodox view. Two arc textually
insecure but the third seemsto me to be quite unambiguous and fatal to this
view.
Fragment Lvmwas seenby Columna to refer to the lottery for Mcsse:nia,
the richest part of Hercules' domain; c£ Sophocles, Aj_ 1285-6, Euripides,
fr. 1083 (Strabo 8.366), Plautus, Cas. 39~. Apollodorus 2.8.4, Pausanias
4. 3 . s.The transmittedtext. aninterse sortiunt urbanatqueag,os,suggeststhat
the lottery took place during the very action ofEnnius' play. Hartung7put
the fragment in a separate play from the others. Other scholan have either
emended an or referredthe fragment to soD1etbingelse.The words trans-
mitted arc metrically defective and there may be something wrong with an.
It is difficult to see in what context the younger Cresphontcs could make
the appeal for peaceand reconciliationquoted by the anonymous rhetorician
at Her. 2.39 (fr. CLXII).Wecklcinaccordingly proposed that the hitherto
received text ChrtsphontemEnniusiruluxitshould be rejected in favour of
either Polyphontem Enniusiruluxitor Thesprotum Enniusiruluxit.The latter is
in &et the reading of the better classof manuscripts and the verses can be
accommodated quite well in Ennius' Thyestes;see bdow, pp. 416 f.
1
Apollodorus2. 8. 5 and Pausanias 4. 3 . 8 make his name Aepytus,Hyginus,
Fab. 137 Tclcphontcs. From Schol. Arist. Eth. Nie. 3.2.1111212 and Epigr.
Cyzic. A.P. 3. s it is plain that Euripides gave him his father's name. See
Robert, Die griech.Htldmsagt u ii, pp. 671 ff.
a Cf. Poet. 14.1454a5.
3 See Pap. Oxy. 2458 (Part xxvn [London, 1962], pp. 73 ff.), Mette, Hmnts
xcu (1964), 391 ff. For the possibility that this comes from an actor's text see
Turner, AC xxm (1963), 120 ff.
4 Columna, Q. Ennii Frag. p. 398, L. C. Valckenacr, Diatribe in Euripidis

DranurtumR.tliquias (Leiden, 1767), pp. 181 ff., Welcker, Die gritch. Trag.
pp. 828 ff., Hartung, EuripidesR.tstitutusD, pp. 47 ff., Wilamowitz, An41.Bur.
p. IS4, N. Wccklcin, in Ftstschriftfur L Urlilhs (Wiirzburg, 1880), I ff.,
Vahlen, Ind. lectt.Berlin 1888/9, 16 ff. ( = Op. Ac. I 417 ff.).
s ~- san. pp. 266 f., Die rom. Trag. pp. 186 ff.
6 Enta.R.tl. pp. 238 f. 7 See EuripidesR.tstitutusu, pp. 39 ff.
271
 COMMEN·TARY
Fragment LIII consists of a dialoguebetween two pc:nons whocan only be
Cypsclus the king of the Arcadiansand Merope hisdaughter, marriedat the
moment of speaking to the dder Cresphontcs. 1 Neither Columna nor
Valckenaertriedto explainhow this could be fittedinto what they knew the
action of Euripides' Kpeacp6v"TT\s to be. Hartung put it into his postulated
second play. Wdcker altered the transmitted reading Chttsphontemto
Polyphontemand proposed the following dramatic situation: on hearing
about the price put on the head of his grandson (Hyginus, Fab. 137.2)
Cypsdus came to Messcnia to remove his daughter &om the house of
Polyphontes; Merope in the meantime had recognised the younger Cres-
phontcs and, in concert with him, pretended to be reconciled with Poly-
phontcs; accordingly she had to resist her father if the plans for vengeance
against Polyphontcs were to succeed. Bergk2 approved the alteration Poly-
phontesand offered a somewhat lessincredible context: Cypselus had learnt
of a rebdlion brewing against Polyphontes and triedto persuadeMerope to
return home by revealingto her that Polyphontcs had murdered her first
husband Cresphontes, a fact of which, up until then, shehad been ignorant.
Wilamowitz3 and W eclclein4 avoided the need to provide a context by pro-
posing the names Ctesiphontem and Deiphontemrespectively assubstitutes for
Cresphontem. W ecklein's arguments persuadedVahlen to remove the tri-
meters in question &om hissecond edition of Ennius' remains.
Marx has seemed to many to disposeof the whole problem by treating the
trimeters as the product of the rhetorical schools rather than the theatre. In
his first edition of the anonymous treatiscS he complained of their fugklity
and the hiatus in nam si improbumesseCresphontem existimas.In bis second
edition 6 he alleged significance in the fact that existimareoccun nowhere else
in republican tragedy and is absent from the high genres of clmical poetry.
Marx's argument persuadedRibbeck to expel the trimctcrs &om his third
edition of the remains of tragedy.
As to the alleged fugidity of the trimeters two opinions are possible; ac-
cording to Columna: 'Enniana haec uerba nuda puraque sunt, ucterisque
Romani sermonis candorem et elegantiam prae se ferunt. • Hiatus before the
final diiambus is presented fairly often by the tradition of early second cen-
tury iambic and trochaic verses, much less often by the tradition of those
composed towards the time at which Marx's hypothetical rhetorician would
have been operating. A considerable number of words absent from classical
poetry, including the tragedies of Seneca, occur in the remains of republican
1
Cf. Pausanias 8. 5 . 6.
a RhM xxxv {1880), 244 ff. 3 Anal. Bur. p. 154 n. 5.
4 PhilologusXXXIX (1880), 409, Ftstschr. Urlichs,3 f. Wecklcin appean to
withdraw at SB Munchen1888, n7 and to return at BPhWxxxv (1915), 385.
5 Leipzig, 1894, Prolegomena,
p. 132. 6 Leipzig, 1923, p. 58.

272
 CRESPHONTES
tragedy. Marx's case against the trimeters is obviously a Bimsyone. On the
other hand a good positive case can be made out in favour of their dramatic
ongm.
Along with the controversial trimetcrs there are quoted in the discussion of
uitiosaeargumentationes at 2.31-4-6 trimeters from the Medeaof Ennius, tri-,.
meters from the Trinummusof Plautus, trochaic tetrameters from a tragedy
by Pacuvius concerning Orestes, a tetrameter given by Ribbeck to tragedy
(Trag. inc. 162) but possibly from comedy, two tetrameten from a tragedy
by Ennius, three trimeters on a tragic theme (Trag. inc. 186-8), six tri-
mcten spoken by an Ajax (Trag. inc. 4-9-54-),and five trimeten given by
Ribbeck to tragedy (Trag. inc. 177-81) but again possibly from comedy. An
argument from Pacuvius' Antiopais described.Among the argumentationes
which Cicero refutes in his youthful rhetorical treatise De inutnlioneat
1 . 78--96arc five of the pieces of verse quoted in the anonymous treatise. The
argument which the anonymous treatise attributes to Pacuvius' Antiopais
attributed to the Euripidean tragedy. 1 Hypothetical arguments quoted by
the anonymous treatise without names (2.46) are referred by Cicero to
Orestes and Ulysses (1 . 92). Cicero alsoquotes an argument mentioning the
treachery of Eriphylei which does not appear at all in the anonymous
treatise. There can be no reasonable doubt that behind the two treatises, here
as elsewhere, lies a common Latin source and behind thisLatin source a Greek
treatise which drew examples from the theatre as wcll as the assembly and
the law court.3 For direct quotation of drama the Latin source substituted
quotations of Latin drama, the indirect quotations it merely translated.
There is no evidence that our archetypal rhetorician made verse translations
himself;the verses quoted at Her.2. 34-and 2. 38 belong to a special category.
SinceCicero docs not quote or refer to the verses about Cresphontes4 it is not
certain whether they stood in the common source or were adduced by the
author of the anonymous treatise from his own knowledge of the Latin
theatrc.S Marx's theory would demand that in the original Greek treatise
there stood Greek verses sayingthe same thingsabout Cresphontes as are said
by the Latin trimeters in question. But while the remains of Greek rhetorical
and philosophical writing quote with remarkable frequency the Kpeo-cp6V'TTIS

1
Ikm apudPacuuiumat Inu. 1 .94-is to be deleted as a harmonising gloss.
i 1.94,. Valckenaer argued. Diatribe, p. 151, that we have here a direct
quotation of tragedy (cf. Trag. inc. 14-3-4-).
3 Cf. Matthes, Lustrumm (1958), 81 ff.
4 Cicero's language at 1. 83 is like that of the anonymous treatise at 2. 38 but
his example is the one used by the anonymous treatise at 2. 4-2to illustrate a
slightly .different type of argument.
S The reference to Pacuvius at 2.4-3 must be his own; contrast Cicero
I .94-.
18 273 JTO
 COMMENTARY

of Euripides they contain nothing about any other tragic Cresphontes. 1

Evidence for free composition in verse on tragic themes in Greek rhetorical
schools is scanty. It is difficult to believe that the practice went on in Latin
schools of the Republican period. Common sense requires that the verses be
treated as of the same characteras the others providing examples of faulty
argument at Her. 2.31-46; ie. as coming ultimately from the Latin theatre,
eitherfrom the writer's source or from his own knowledge.
If the verses belong to Ennius, attempts to reconstruct the Kpeaq,6vn)s
of Euripides with the fragments of Ennius' Cresplwntesand vice versa
must cease.There is no room for a discussion of the younger Cresphontes'
virtues as a husband in the plot-structure of the Euripidean tragedy.
Ribbeck's view that the Latin tragedy dealt with the overthrow of the
dder Cresphontes must be reinstated.

LIii
Stephanus, Columna and Scriverius print all the poetic matter quoted,
except the repeated nam si improbumesseCresphontem existimascur me huic
locabasnuptiis,as the words of Ennius. Bothe excludes only the repeated nam
si improbum esseCresphontem existimas.Dclrius 1 and L. Mudler exclude all the
repeatedmatter and duxi probum,maui, postcog,wuietfagio cognitumas being
the work of a rhetorician. The pre-Marx Ribbeck excludes the repeated
matter and nulla. .. incommodis. The only course still untried is to exclude
both replies to the argument of Cresphontes' wife. Something could besaid
in favour of every course. I have followed Dclrius but think it possible that
the rhetorician extracted duxi probum,maui, post cognouiet fugio cognitum
from a following speech and glued it to part of the speech of Cresphontes'
wife already quoted so as to provide an illustration of a second method of
answering the 0-eyxoSa\l<XYJ(OOOV Kara 6£A1u.1µa.At 2. 34 the trimeter
utinamneeramans meadomoeffmet pedemis botched together from different
parts of a tragic speech.
There are paralld situations in Menander's 'E,nTprnov-res(s10 ff.), in a
fourth~tury play whose character is disputed3 and in Plautus' adaptation
of Menander's 'A&Aq,olJ:\', the Stichus(1s ff.). Behind the two Latin argu-
ments lies the assumption that the woman's father had full legalpower to

• It is significant that the 'non-dramatic' pieces of tragic verse in Cicero's

philosophical writings arc all translations of pieces frequently quoted in the
remains of Greek philosophical writing.
1
SyntagmaTragoediaeLatinae(Antwerp, 1593) 1, p. 164.
3 Euripides fr. 953 = Menandcr 1, p. 143 Korte = Anon. p. 180 Page
( G.L.P.). Sec most recently W. Buhler, Htrmts XCI (1963), 34S ff.

274
 CRBSPHONTBS
dissolve the marriage against her will I The Greek arguments leave it quite
unclear whether the father imaginedhe could use anything but persuasion
or brute force. It is possible that theconditions of Roman societyinfluenced
the way Plautus and Ennius adapted their Attic originals. The Twdvc
Tables recognised aform of marriage sinemanirand such marriages could be
legally dissolved by thewoman's father until the time of Pius or Marcus.3
Only one kind of marriage is known at Athens andit is doubtful whether the
father could interfere against theexpressed will of his daughter.•
For the type of argument employed by Crcsphontcs' wife c£ Euripides,
BI. 1015-17 To irpayµa & I 1,1cx86VTas, i\v ~ a~ioos I
lxTJs,
1,11cntv
cr,vy&lv6{Katov• el & 1,11'}, Ti Setcr,vy&lv;, I.A. 1034-5, Hyps.ap.Lydum,
De mens.4.7 (p. 48 Bond), Trag. Lat. inc. 178-Bo, Terence, Hee. 558-9.
For thecontent of the argument cf. Plautus, Stich.130-1 namout olim,nisi
tibiplacebant,non datasoportuit,I manunc non aequumest abdudpain illisce
absentibus.

125 iniaria abs te adficiol!" indigna~ c£ Terence, Phorm.730 uereorerane

ob meumsuasumindignainiuriaadfaiatur(from a highly emotional speech).
Aliqueminiuriaadficeredocsnot occur elsewhere in comedy. The legalism
alicuiiniuriamfacere (c£ Lex m tab. 8.4) is frequently usedboth properly
and metaphorically (c£ Plautus, Stich. 12-16 patremtuommeumqueadeo..•
uirisqui tan/as absentibus
nostris f acit iniuriasinmerito
).

pater: thesimple vocative is regular in both comedy andtragedy; mipain

or pain mi would indicate a warmth evidently lacking here. Sec Kohm,
Altlat. Forsch.pp. 179 ff.

u.6-7Ii improbum eae Czesphontem exiltimu, I car me

huic
locabu nuptii,?: •If you really think Crcsphontcs is a bad man, tell me
why you triedto makehim my husband.' For the unusualarrangement of
tenses in the conditional sentence c£ Plautus, Asin. 452 sed si domi est,

• The suspectedverses Stich. 53-4, uerumpostremo est situm:I

inpatrispotestllte
faciendumid nobisquodparentu impnant, make the assumption quite cxplicidy.
It is implicit in IS£, 69, 128.
a Gaius I. III.
3 Sec H. F. Jolowicz, HistoricalIntroduction to tht Studyof RomanLaw, ed. 2
(Cambridge, 1952), pp. 242 ff., 561 f.
4 The authorities differ. Sec J. H. Lipsius, Das attischtRecht und Redats-
verfahrenuii (Leipzig, 1912), pp. 483 £, E. Levy, in Gediichtnisschriftfor
E. Seckel
(Berlin, 1927), 154,R. Taubcnschlag,ZSavSlXLVI,Rom.Abt. (1926),74ff., U.E.
Paoli, AegyptusXXXII (1952), 284. The passage of Demosthenes always quoted
(41. 3) says nothing about the woman's attitude.

21s
 COMMENTARY
Dtm«netum uoltbam,Terence, Plwrm.1023 iam tum erat senex,sentaus si
tm'tcundosfaat.
Verse 126 as transmitted is metrically odd in two ways. There are few
clearexamplesin Ennius' iambic and trochaicversesof the pure clement of
the mctron beingoccupiedby a long penultimatesyllablcand none of hiatus
before thefinal diiambus.Both phenomena are rare in thetradition of early
comedy but sufficientto deter all but the most extreme analogistsamong
emendcrs; they are very much rarer in that of later comedy.1 The only
elegant emendation of the verse in question so fu offered is Wilamowitz's
namsi inprobumtsse Ctesiplwntem tximnuu. However Wilamowitz's motives
were base and the rationalcof the corruption is hard to explain.I Jcavethe
transmitted verse in the company of: on theone hand, 178 ntcassetquoquis
audatu (Iunius: ntc.assatquosquisaudatur codd.)pnbittrt~ 28o pa1""'muttirt
pltbtio puiculumtst, 388 pltbts in hoe rtgi antestatlocolicet,Pacuvius 27t
attattm adiungas audatumrttiantia,Trag. inc. 17 omnisatqualisuinabatquin-
quertio,and, on the other, Pacuviuss ita satptuosedictioabsk ddtur,248 parittr
k wt ergailiumuulto ut ilium k (ud Vossius)ergasdo, Accius 301 tloqutrt
proptrt (Iunius:propritcodd.) acpauortmhuncmtUm txpectora.
For mt huicloatbas nuptiiscf. Q_uintilian,Dtcl. 247, p. 10.14 uxortst quat
ftmina uiro nuptiiscollocatain sodttattm uitaeutnit. Plautus, Aul. 192, 228,
Terence,Phorm.646 have aliquamlocart;Plautus,Trin. 1S9collocart;Terence,
Phom,. 752 nuptumlocart;Plautus, Trin. 735 nuptumcollocart; Plautus, Trin.
782 in tnatrimonium locart.Ennius must be theone varying the regularlegal
phraseology.

1z8 inaitam inuitum: for the polyptoton cf. Aeschylus,Hilt. 227-8

,r&')s 6' &v ycxµwv &Kovaav mcovrosirapcx I
ayvosywotT' 6v;,Prom.
19, 671, Euripides,Hipp. 319, Seneca, Cltm. 2. 1. 2 inuitus inuitocum. ..
protulisstttradertfiUt,Suetonius,Tit. 7. 2 Bermianstatimaburbtdimisitinuitus
inuitam.

linqaere: the simple verb occurs six times in tragedy, rtlinqutrtonly

three. In comedy on the other hand rtlinqutrtis extremely common whilc
linqutrtoccun only four times (in paratragic contexts: Plautus, Asin. 28o,
Capt.282, Cist. 643, PseuJ.141).
1
For the 'Dipody Law' see J. Draheim, Hmnts xv (1880), 238 tf., W.
Meyer, Abh. Bayer. Ak. xvn {1886), 36 tf., R. Klotz, Grundzugt,p. 315, F. W.
Hall, CQ xv (1921), 99 tf. For the 'LocusJacobsohnianw'see A. Luchs, in G.
Studemund, Studia in PriscosScriptorts Latinos I (Berlin, 1873), 22 tf., R.
Klotz, GrurulzUgt,pp. 132 tf., H. Jacobsohn, Qy_atstiones PlautinatMttricat tt
Grammaticat {Diss. Gottingen, 1904), pp. 2 tf., 26 tf., A. Klotz, Htrmts LX
(192s). 328 ff.
 CRESPHONTES
129 nata: vocativegnataoc.cursseventimesin comedyas againstfilia four.
There is nothing peculiarlytragic about Ennius' languagehere. Outside the
vocativegnatadoeshave a markedlylofty tone. It oc.curs7 timesin tragedy
as againstfilia 4; in comedy S4 as against 183.
Mea gnata is the regular phrase. Simplegnata mwt indicate considerable
lackof warmth (c£ Plautus, Trin. 1 ).
130 Ii probus est, collocaai: 'if as you say he is a good man, I will not
alter the arrangementI made'. The tense of collocaui in contrastwith that of
locabas in v. 127 and the additionof the prefix indicatethe completenessand
fixity of the arrangement.
131 diuortio: the technicallegal term (c£Paulus,Dig. 50.16.191) for the
dissolutionof a marriage;it is usedthus three timesin comedy (Plautus,Aul.
233, Mil. 1167, Stich.204).

incommodis: for the substantiveplural c£ Terence,AJU/r.627, Haut.932,

Hee. 165, 840, Phorm. 248 (Donatus remarks: nimis comicenon cladesout
aerumnasposuit),Acciw, Trag.350. Plautushasonly the singular(Amph.636,
Capt. 146, Mere.773, Most. 418).
LIV
Ribbeck.made two suggestionsabout the context of this verse: first that it
wasspokenby Meropc to her fatheraftcrhe had proposed the secondmar-
riagefor her (c£ fr. I.III),the secondthat it was spokento Polyphontcsafter
the murder (c£ fr. me). The metre counts somewhatagainstthe first sug-
gestion.The versecould come from a prologue speech;the republicanpoets
did not alwaysuse trimcten for these.If the play is about the younger Crcs-
phontcs a number of contextsarc imaginable.
132 ducit me uxorem: doubtlesslylegal phraseology;c£ Plautus, Aul.
170, Cas.69,107, 322 et al.

liberorumsibi qaaesendumgratia: more legalphraseology;see above

on v. 112.
Gratiamodifies the legalistictone; with the genitive, mea etc., ea etc.,
gratiaseems to have had a loftier tone than causa;in tragedy the ratio of
occurrenceis s:3, in comedy so: 142.

LV
Ribbecktook thisverseas spokenby Mcropc to her father in the sceneof the
quarrd (c£fr. LIII); Wecklcinas spokenby the younger Cresphontcsclaim-
ing the reward from Polyphontesfor 'Cresphontcs'' death (Hyginus,Fab.
137.3).
277
 COMMENTARY
133 audi atque auditis: thiskind of epanalepsis is comparativelyuncom-
mon in republicandrama (cf.however Plautus,Amph.278, Aul. 365-6, Trin.
127, True.490, Terence, Amir.298, Eun. 1057, Accius, Trag.365 sistispelli,
pulsumpatimini).It probably had a stylisticweight lackingin its Attic equiva-
lent. The wealth of participlesavailableto them allowed the Attic dramatists
to employ it frequently (cf. Euripides, Hile.743, Hipp. 313 et al.).

hOIUDlentnrn: only here and Plautus, Asin. 172 (parparidatumhostimen-

tumst, operapro pecunuz):hostireis equally rare, occurring only at Ennius,
Trag.149, Pacuvius, Trag.3-46,Plautus, Asin. 377, Laevius,fr. 1; likewise
retlhostire
(Naevius,Praet.6, Acciw, Trag.92, Didasc.fr. 10). Hostimentum is
unlikely to be a poetic coinage; Plocn finds only three formations in
-mentumin tragedy as against 52 in comedy.
adiungito: to judge by Plautinc usage the so-calledfuture imperative
form in -to had a highly formal tone even in the early second century; the
materialis collected by 0. Riemann, RPh x (1886), 161 ff.

LVI
Ribbcck made two suggestionsabout the context of thisfragment: fine that
it was spoken by Polyphontes in reply to somooneadvisinghim not to kill
the elderCresphontcs;secondthat it was spoken by Cresphontesin reply to
Meropc warning him to be careful of his enemies. W elcker compared the
versesof Euripides' l<peacp0V'T1lsquoted at Schol Eur. Med.85: mlvo yap
,rmove•6mp 1r~ • Icp1AC>v
f3pcn'ot 1,1@.1~ lµmrrovoVK atC7)(VV01,100
(fr. 452) and put it in a speech by Polyphontes defending his actions to
Meropc. The tense of deprecer excludesthisinterpretation. Vahlcnmade the
speaker the younger Cresphontcsreplying to a plea by Mcrope to spare the
life of Polyphontcs.
The sentimentexpressedmust have been commonplace; cf. Seneca, Thy.
324-5 maleagis;recedis, ani~; si parcistuis I parceset illis,Plautus, Asin. 177
quaeamantiparceteademsibiparcetparum.
134 ego meae cam uitae pucam: theconjunctioncumhcrcscomsto have
a conditionalforce; cf. Plautus, Capt.892 dubiumhobebisetiamsanctequomego
iuremtibi, Tibullus 2. 3 . 5-6 o egocumaspiceremdominamquamf ortiterillicI
uersarem . . •solum.
letum inimico deprecer: for the sakeof concinnity Ennius avoids what
must have been the normal construction, deprecari ab+ abl
For mortem depre'4ric£ Cicero, Vm. 5. 125, Caesar, Gall.7.40.6, Sallust,
lug. 24.10, Ovid, Fast.2.103, Pont. 1.2.57. For the poetic word letumsec
above on fr. XXXIV.
 CRESPHONTES

LVII
Hartung interpretedthese words as coming from a messenger'sspeechde-
scribing the sacrificeat which the younger Cresphontcsslew Polyphontes
(c£Hyginus,Fab.137 .6). Ribbcckacceptedthis interpretationand extended
the plot of the tragedy accordingly.Vahlcn (reading o pietas,eam [~ ire]
secumaJuocant . .. ) took them as from a speechby Meropc comp)ajnjngof
how shewas not allowedto preparethe corpsesof her husbandand sonsfor
burial (c£ fr. ux). L. Mueller referred them to an account of Polyphontcs
and hisfellowconspiratorspurifyingthemselvesafterthe murder of the dder
Cresphontes.

136 nitidant: only here and Accius,Trag.6o2 in republicandrama;from

elsewhere in Latin the dictionaries quote only the agricultural writers
Columellaand Palladius;probably, therefore, a poetic neologism.

LVIII
For the context of this fragment sec above, p. 271.

137 inter se aortiunt urbetn atqae agros: 'they draw lots to secwho of
themselvesshouldhave the state'. Probablythe regular earlysecondcentury
usage;c£ Plautus,Cas.395,413. Sortirimeant 'to have lots drawn' (Plautus,
Cas.298, 342, Mere.136b).
The phraseurbematqueagroswould have recalledthe officiallanguage:c£
Plautus,Amph. 226 urbemagrumarasfoeossequeuti dederent, Livy I . 38. 2 et al.

LIX

The speakermust be Mcrope, the corpsesmentionedthoseof the elderCres-

phontes and two of her three children.The time of spcakjng may just as
likdy be straight after the murder as some yearslater.
Macrobius,or his source,ought to have provided more of the speechof
Euryalus' mother: e.g. from Aen. 9.483: neete sub tantapericulamissumI
aJfari extremummiseraedatacopiamatri?I heuterraignotaeanibusdatapraeda
LAtinisI alitibusqueiaces.neete, tua fanera, materI produxipressiueoculosaut
uulneralaui, I uestetegenstibi quamnoctesfestin11diesqueI urgebamet telacuras
solabaranilis. I quosequar?aut quaenunc artusauolsaque membraI et fanus
lacerumtellus habet?It is not certain that this is modelled on the piece of
Ennius' Cresphontes quoted. Lamentsat not beingableto give thedue rites to
one'snext of kin must have beencommonin republicantragedy;c£,in Attic,
Sophocles,Ant. 21-30, El. 865-70, 1138-40, Euripides,Hilt. 51-3. Cc:rtainly
both Enniusand Virgil reversethe chronologicalorder of events. But such
279
 COMMENTARY
VO"ltpov1rp6Tepovis frequent in the Aeneid,1 Attic tragedy, and the re-
publican adaptations. It may even have been a particular feature of funeral
descriptions; cf. Euripides, Heroltles136o-1 SosTova& TVJ.lJxt> Kai mpf-
I 6aJ<pvo101TIIJWV.
VEKf)OVS
<TnlAOV
The text transmitted is obviously unmetrical and its Latinity hasbeen sus-
pected.By changing the order mihi corporato corporamihi Bothe produced
two iambic tetrameters. Scholars have beenfurtherworried by the very bold
personification miseraelaaimae"and the departure from the common ar-
rangement of words lacrimislauere.3Hence Buccheler's mihi licuitnequemeae
lauerelacrimae,Usener's mihi licuitnee miserailauerelacrimis,Ribbeck's mihi
licuitmiserae,nequelauerelacrimaesalsae,0. Skutsch's licuitnequemiseraelauere
lacrimae(latice) salsumsanguinem(deleting mihi altogether). 4 In a genre
which can produce the phrases superstitiosis
hariolationibus and curlssuspiranti-
bus(seeabove on v. 33)one ought not to be worried by miseraelacrimae. The
phrase lauerelacrimae is no bolder than auresaucupant(Ennius, Trag.245) or
oculiuescuntur(Accius, Trag. 189).

138 terram inicere: cf. Virgil, Aen. 6.365-6 aut tu mihi tmam inice. I
Ennius and Virgil import into the Greek heroic world the culminating act of
the Roman funeraryritual, varying slightly the traditional phraseology; cf.
Cicero, Leg. 2. 57 neetameneorumantesepulchrum est,quamiustaf actaetporcus
caesusest. et quod nunc communiterin omnibussepultisuenit usu, ut humati
dicantur,id eratpropriumtum in iis, quoshumusiniectacontexerat,
eumquemorem
con.firmat;nam prius quamin os iniectagleba est, locusille, ubi
ius ponti.ficale
crematumestcorpus,nihilhabetreligionis;iniectaglebattumet illisthumatusest,
etglebauocatur,actumdeniquemultareligiosa iuraconplectitur,
Varro, Ling. 5 .23,
Paulus,p. 250. II s.v. PllAJ!CIDANEA.

cnaenta conuestire corpora: the adjective cruentacould be corrupt; the

blood was usually washed from corpses before they were dressed; cf.
Homer, n.18.343 ff., Euripides, Alk. 158 ff., Hile. 765 ff.

1
See Norden, Aen. v1•,p. 379.
11
T L.L. vm no3 .4 can offer no parallel; contrast Terence, Hee. 379
lacrumansmisera,385 laaumemmiser.
3 Ennius, T,ag.276, Plautus, Pstud.10, Afranius, Com.tog.322, Accius, Trag.
420, Ovid, Ars 3 .744, Met. 9.680. Accius, Trag.578 makes a slight variation:
salsiscruorem
guttislacrimarum lauit.
4 The apparatus of Willis adMacr. Sat. 6.2.21 is misleading. Skutsch refen

me to such expressions as Euripides, I. T. 228 olKTp6vT' ~wrc.w 6a,cpvov,

Hik. 96et al. for the number and attribute oflacrumaand points out the parallel-
ism of the two infinitives conuestire
and lauereand the two sets of triple allitera-
tion arising from his emendation. See now HSCPh LXXI (1967), 135.

280
 ERECTHEVS
Thecompound verb ccmuestire occurs elsewhere in republican dramaonly at
Trag. inc. 137. Comedy hasthesimple uestirenine times. Ennius formed the
compound here for the sake of the triple alliteration. Other rare formations
with con-in tragedy are commiseresare {Ennius 162, Pacuvius 391,Trag. inc.
246), configere{Accius 539), consenuire
{Pacuvius 189), consepire {Ennius 252 },
contegere(Accius 540),contremere{Pacuvius 413 ), conuisere
{Accius 598). Plautus
seemsto have produced them freely for comic and rhythmical purposes. 1
139 miserae lauere lacrimaeaal,um sanguinem: the epithet salsum,
3
from a subjective point of view, would fit laaimaebetter than sanguinem.
Ennius and other republican tragedians preferred the effectof alliteration;
c£ Accius, Trag.322, Trag. inc. 77.
Laaima was an old borrowing from Greek which drove the inherited
word out of the ordinary language; the alliterative phrase laaimis'4uert
looks like a creation of the dramatic poets; the only thing parallel in Attic
tragedy which I can find is Euripides, Heraleles 48o-2 µe-raj3<XAo0aa 6' ~
Ivvµq,asµwvµtv Kf\pcxs
TVXTl ,Xe1v,I lµol & SaKpvaAO\TTp<X
oom6CA>K'
6vcm'Jvo1scptpe1v{Bothe: 6vOTT)VoS fpEVOOV codd.).
Emotion rather than strict regard for chronology controls the order of
events in Merope's discourse; c£ Plautus, Men. 509-10 nequeegoErotioI dedi
neepallamsurrupui,Terence, Haut. 779 illi nequedo nequedespondeo. Com-
parativdy speaking,tragedy describes events Ocrrepov irp6TEpovquite often:
e.g. Ennius 148 idegoaecumaciustumfecisseexpediboatqueeloquar,249 hoeego
tibidicoetconiecturaauguro,Pacuvius 91 omniaanimat formatalitaugetcreat,Trag.
inc. 148 uis quaesummas ftangit infirmatque opes.The Latin poets could have
been introducing artificialities from their Attic originals {c£ Euripides, El.
16o ff., Hik. 918 ff., Ion 154 f., Or. 814 f.) but thefrequency of the pheno-
menon in comedy {add to the above quoted casesPlautus, Cist. 675, Men.
643, Mil. 773, Pseud.133, 283, Rud.996) suggeststhat ordinary Rom.ans
sometimes spoke in this way in their more excitedmoments.

ERECTH.EVS

Gellius and Macrobius give the title Ereaheusto Ennius. The pieces quoted
could come ultimatdy from the dictionary ofVerrius Flaccus.Festusquotes
an Ereaheusand the name of Ennius as the author may be restored in the
defective manuscript. Nonius' one quotation comes from Gellius.3
1
Sec Lorenz, Plautus:Pseudolus(Berlin. 1876), p. 38.
2
C£ Bentley, Emendationes,in J. Davies, Cimo: Tusculan« Disputaticmes
(Cambridge, 1709), p. 8.
3 Sec W. M. Lindsay, Nonius Marcellus" Dictionary,p. 69.
281
 COMMENTARY
Euripides produced a tragedy •EpEx&vsconcerning the siegeof Athens
by Eumolpus and a Thracian army and the sacrificeof the ddest daughter of
Ercchtheus and Praxithca on the ordcn of the Delphic oracle.1
Cicero refers three times to the legend (Sest. 48, Fin. s.62, Tusc.1. 116)
without giving any sign that he knew a dramatic treatment of it.
Ribbeck: assumed with Columna that Ennius adapted the Euripidcan
tragedy and assigned 3
a number of piecesof verse quoted with the name of
Ennius to the Erectheus;that quoted by Varro at Ling. s.14 (fr. CLXXXVI) to
an apostrophe by Ewnolpus to the land of Thrace; those quoted by Fcstusat
p. 110. 16 (fr. CCI) and p. 388.25 (fr. CCIV) to a patriotic speechby Ctho-
nia, the princess who was to be sacrificed; that quoted by the Danieline
Scrviuson Aen. 2.62 (fr. CCXXII) to a speechby Praxithca consentingto the
sacrifice.Unnecessaryviolence has to be done to the transmitted text of frs.
CLXXXVI and CCXXII to makethem fit the plot of the •Epex8£vs.3 Fn. CCI and
cav could bdong to any of severalplays.

LX
140 quos DOD miseret nentinis ! for the double negativec£ Plautus, Mil.
1411 iura tenon nociturum... nemini,Lucilius ss1 proprium uero nil neminem
habere.A certain degree of emotion is indicated in the speaker.
Neminis occurs only here and at Plautus, Capt. 764-s in republicandrama.
The desire for word play dictates Plautus' neminisI misererecertumest, quia
mei miseretneminem.
LXI
The speaker is Praxithca. Columna compared Euripides, fr. 360. so-2
)(pfjae'cI>1TOAiTCX1 TOiSAµoisAoxevµacnv,I ac;>3EO&, vtKaT'. mrrl yap
'fNXiiSµ1ijsI OUK roe·c!rrroosovTT)vs·tyw (7(00'(A)1t'OA1v. Ennius' lan-
guage must have recalledthe cant of contemporary politicians:cf. Cato, Or.
fr. 164 Malcovati... ne subsoloimperionostroin seruitutenostraessent.libertatis
suae causain ea sententiafaissearbitror,Cicero, Pis. 1 s c«dem illi ciuium,uos
seruitutemexpetistis.hieuosetiamcrudeliores:huicenimpopuloitdfaeratante uos
consuleslibertasinsita ut emoripotius quamseruireproestaret,Phil. 6.19 aliae
nationesseruitutempati possunt,populi Romani estproprialibertas.

141 libertatemparo: paro rather thanpariois guaranteed by the metre.

The regular phrase in classicalprose is libertatemparere;c£ Rhet. inc. Her.
4.66, Cicero, Leg. agr.2.16, Livy 3.61.6 et al.

1
See Lycurgus, uokr. 98 ff., Apollodorus 3. 15 .4, Hyginus, Fab. 46, 238,
Robert, Die griech.Heldensage n i, pp. 140 ff.
a QHaest.seen.p. 262, Die rom. Trag.p. 185.
3 See Introduction,
p. 62, below, p. 406.
282
 EVMENIDES

LXII
These words probably come from a messenger'sdescription of Eumolpus'
besiegingarmy.
There is an antithesis here between arma'defensive' and tela 'offensive
weapons'. Arriguntgives a feeble effectand, in any case,the only reasonably
near parallelseemsto be Propertius 4.4. 19-20 uiditharenosisTatiumproludere
campisI pictaqueperjlauasarma(frenaPalmer: loraHartman) leuareiubas.
There is much to be said for Bergk's fa(goriunt; 1
n.
cf. Homer, 22. 133-4
aefc...,vTT11A16:6a
µEA{11vKcrra 6e~1ov&µov I &nn'lv• &µcpl~ XCXAKOS
0.6:µm-ro, Virgil, Aen. 7. 525-7 atraquelate I horresdtstrictissegesensibus,
aeraquefa(gentIsolelaassitaet lucemsubnubilaiactant,11 . 6o1-2 tumlateferreus
hastisI horretagercampiquearmissublimibusardent.

143 horrescunt tela: to the observer the lancesappear to be shivering;

cf. Ennius, Ann. 285 densantur
campishorrentia telauirorum,Virgil, Am. 10. 178
milk rapitdensosade atquehorrentibus hastis.The aspect is slightly different at
Homer, n. 4.281-2 cp&Aayyes .. . aaKEaivTE Kai fyxeai mcppnrutai,
13. 339, Euripides, Phoin.1105, Ennius, Ann. 393 horrescit telisexercitusasper
utrimque,Horace, Sat. 2. 1. 13-14, Livy 44.41 .6.

EVMENIDES

Nonius gives the title Eumenides four times to Ennius and sixty-six times to
Varro. The pieces he quotes ofVarro's Eumenidescome from Lindsay's list
31 'Varro ii', a list apparently based on a volume containing eighteen of
Varro's Mcnippean satires. The four pieces of Ennius' Eumenidesillustrate
unclassicalverbs and verbal forms; their source is uncertain but it is possible
that they all come from list 27 'Alph. Verb'. Ennius' work was clearly a
tragedy which dealt with the trial of Orestes at Athens concerningthe kiUing
of his mother.
The Evµa,f6esof Aeschylus was the only work of the three classical
tragedians known at Alexandriato have presented this story2 and there is
thus a strong presumption that thiswas Ennius' model. Scaligcr3interpreted
Nonius' four quotations as versions of specificpassagesof Aeschylus' play.
It is possiblethat the title Eumenidesdocs not come from the theatre but
from a scholar who believed that Ennius' original was the Acschylcan
1
Philologusxxxm {1874),290 {= Kl. phil. Sehr. 1357).
acelTcnft µv8o,roda.
, According to the extant hypothesis: ,rap' o(/&-riP<t>
3 Coniect.Varr. Ling., on 7. 19.
 COMMENTARY
tragedy.1 The story of the transformation of Orestes' persecuton from
'Ep1vvesinto Evµev{6es was of popular interest at Athens and of learned
interest elsewhere.:i It does not seem a likdy theme for the early second
century Roman stage. The final sceneof the Athenian play hasno rdcvancc
to thepreceding action and could have been omitted from a Latin version.3
Aeschylusopened hisplay with thesceneset in Delphi and then shiftedthe
action to Athens. Such changes of scene took place rardy, if at all, in later
Attic drama4 and it is worth pointing out that Ennius could have omitted
the Aeschyleanopening or replaced it with another set in Athens, thescene
of tho principal action.
Orestes was a popular figure on the Roman stage; he appeared quite
certainly in the Electraof Atilius, in the Chryses,Dulorestesand Henniom,of
Pacuvius,in the Erigom,of Acciusand in the Oresusof art unnamed author
(Donatus, Gramm. IV 375. 25); possiblyalsoin the Hennionaand Aegisthusof
Livius, the lphigeniaof Naevius, the Agamemnonidae, Aegisthusand Clyte-
mestraof Accius.
Cicero does not name the title Eumenulesor any pomble variant but in
referring to the trial of Orestesat Mil. 8 he seemsto have the stage in mind.S
His referencesto the friendship of Orestes and Pylades at Fin. 1.65, 2.79,
5.63, and Lael. 24 concern a play by Pacuvius. Those to the madness of
Orestesat Pis. 47 and Tusc.3 .11 concern the stage but one cannot say more.
Three references to stage Furiaepunuing matricides with blazing torches
(S. Rose. 67, Pis. 46, Leg. 1.40) have been thought to concern Ennius'
Alcmeobut since it is almost certain that the Furiaedid not appear on stage
before the audience'seyes in that play (see above, p. 186) one should think
rather of the Eumenides.
At De orat. 1 .199 Cicero refcn halfobliquely, halfdirectly to a speechby
Apollo in a play by Ennius (fr. CUCVI). Ribbcck6 set thewords quoted in the
god's addressto the Athenianjury, comparing Aeschylus,Bum. 576ff. and
614 ff.,but did not go so far asactuallyto print them alongsidethe quotations
of the Eumenides made by Nonius. Vahlen suggestedthat7 they were spoken
by Apollo as he expelled the Furies from his oracular shrine, comparing
Aeschylus, Bum. 179 f., 185, 194 f. It seems to me equally possible that
Apollo spokethe prologue of Ennius' Alcmeoas we now know he did that of
Euripides'•AAKµic..>v 6 6uxKop{v8ov(fr. 73 a Snell),and that Cicero is quot-
1
C£ Introduction,p. 6o n. 3. a C£ Cicero, Nat. deor.3 .467.
3 C£ Tcrzaghi, SIFCN.S. VI (1928), 184 (=Stud.Gr. et Lat. 815).
4 Cf. above on fr. 1.
s The references at Inu. 1. 18 and 92 (c£ Her. 1. 17, 25, and 26) may stem
from Greek rhetorical writing.
6 ~aest. seen.p. 271. Cf. G. Hermann on Acsch. Bum. 6o8 (Leipzig, 1852).

7 E.P.R. 1 , p. 142.
 EVMENIDES
ing this. There are no really striking correspondencesbetween Cicero's
quotation and anything in Aeschylus'Evµa,l&s.
Discussing the temperate zones of the earth at Tusc. I . 68-9 Cicero
describesthat of the northern hemisphereas follows: hieautem,ubihabitamus,
non intermittitsuo tempore•caelumnittsare, arbores frondescere,I uitesl«tificae
pampinispubesme,I ,ami bacarum ubertmeincuruesa,e,I segeteslargirifruges,
jlo,e,e omnia,lfontesscatere,herbisp,ata conuestirie,'.
He is obviouslyquoting
tragedy and Ribbcckimaginedhim quoting Ennius'versionof tho dialogue
betweenAthenaand the Erinycs at Aeschylus,Bum. 902-9: - Ti ow µ'
&,c.:,yasTij6' ~vf\acn x&ov(; I- cmota VfK11sµ11KCXKTlS rn(0'1<01Ta, I
Kai TCXVTa yfi8EV1K-re irOVTlas6p6crov I i~ ovpavov -re· Kaveµu>v
m'Jl,laTCXI EV11A(u>5'RVEOVT' tma-re(xeiv )(66va. I Kapir6v Te yafas Kai
f3ar&vmlpp\/TOVI &crrotaw ev8evowTa µ11Kaµvetvxp6vct>, I Kai Toov
~(C® cmepµa-re®<rCAffllp(av. Hermann prefaced the five trimoters
directly quoted by Cicero with suo non intermittattempo,e.1 Vahlenprinted
them as part of the Latin Eummides.Even if the verbal parallelismwere less
superficialthan it is this would be a rash prnceccling,Furthermore, as I
have argued above,there is no certainty that Enniusretained tho final scene
of Ae.,chylus'play.
At Ling. 7. 19 Varro quotes a pieceof Enniusnow extremdy corrupt but
evidently about the Areopagitae(fr. cxcn). Scaligcrcompared Aeschylus,
Bum. 690and most editors have included it in the Latin Eummides.It could
come from any play set in Athens or referring to .A.tbenianinstitutions.
In Paulus' epitome ofFestus, p. 73. IS, there is quoted a tragic addressto
Night without the author's name: ErebocreatafusciscrinibusNox te inuoco.
RibbcckcomparedAeschylus,Bum. 321 µ&np & µ" rnKTES, w µ&np Nv~
and assignedthe fragment to Ennius' adaptationof the Furies' prayer. The
addressto Night is one of the commonestof tragic motifs (seeabove on fr.
xxxm) and I have thereforenot acceptedRibbeck'sassignation.

LXIII
Scaligcrcompared Ae.,chylus,Bum. 463-7: ft<TetvaTflVTeKoOaav,OVK
6:pVT}aoµcn,I6:vTtKT6vots 1r01vataicptATaroviraTp6s. IKaiT&'>v6e
Kotvij
/\~ias rnafTtc,s,I&A)'TI irpocpu>v&v &vnKEVrpaKap6fq:,Iel µftTt Toovs·
lp~o1µ1TOVS rncxrrfovs.

I.U materno: 'matris': c£ Pacuvius, Trag. 144, Accius, Trag. 23 ;

comedy has maternusonly once and in the sense of 'maternal' (Terence,
Haut. 637).

1
On Acsch. Bum. 894.
 COMMENTARY
sanguinee:undando: c£ Euripides,Rlses.430-1 M' alµaTTlpos°lriAa-
VOS~yaiav IKveTls 1-ftvrAEITO AOY)(1J 8pt~ TE avµµ1~s cpovos, Cicero,
Sest. 54 meperculso admeumsanguinemhauriendum. .. aduolauerunt,Ovid, Met.
7. 333, 13 . 331, Livy 7. 24. 5 et al. The full grim point of the Ennian Ores~•
language is not revealed by these parallels; he thinks of the ki])ing of
Clytemncstra as the offering of her blood to the shade of Agamemnon: he
will draw not the conventionaldraught of wine (c£ Aeschylus,Pers.610 ff.,
Euripides, I. T. 159 ff., Paulus, Fest. p. 319.1 f. lll!SPAllSVM VINVM dixerunt
quia uino sepulcrumspargebaJur) but one of blood. For exandareas a ritual
word see above on v. 103. The heroes of Attic tragedy frequently describe
their deedsof bloodshed as if they were acts of sacrificeto divinities; c£,
where Clytemnestra'sdeath is concerned, Aeschylus,Choe.904 hrov, irpos
CXVTOV Tov6e at aq,a~ooethCA>, Euripides,El. 1221-3 tyc;.,l,1EVrn,~c;.,v
q,aptlKopcxlS lµais I cpaayav<t) KaTTlp~aµavI µ<mp<>S la(A)&pas µe6e{s,
Or. 562 rnl 6' levaa 1,111-ripa, 842 aq,ay1ov 18e-ro µaripa. Sometimes the
audience is merely being presented with a horrific metaphor; sometimes
there is a tone of blasphemy.' The Ennian Orestes is perhaps making an
implicit defenceof his deed.The Romans seem to have conceivedof many
kinds of legal and semi-legal killing in sacrificialterms. i

alci,a,rem; c£ Pacuvius,Trag.169, Accius,Trag.293; to judge by comic

usagethe deponent formwas normal in the common languageof the second
century; the only other examplesof the active in recorded Latin seem to be
Sallust,lug. 31 .8, Livy s.49.3, and ValeriusFlaccus4.753.

LXIV
s· 'OpiOTT)SKav la6'fl11q,os
Scaligercompared Aeschylus,Bum. 741 VlK~
Kpleij, Ribbeck752-3 avflP 66' ~ aiµ<X"ToS • I iaov yap
6h<1'}V
W"TITap{8µf1µaTOOV ,r&ACA>v.
The sevenLatin words must be spoken by Athena, president of the court,
to the Furies, the prosecutorswhose charge has failed. They imply that the
decision is Athena's own and not the automatic result of the deadlock
brought about by her equalisingvote. In trialsof the late republic equality of
votes meant acquittal3 and the president of the court had nothing to do
except announce the jurors' decisionbut it is possiblethat in Ennius' day the
1
See Fracnkcl on Aesch. Ag. 1384 ff., 1432 f.
a Cf. Festw, p. 260.7 ff., Pliny, Nat. 18.12, Servius auct. Aen. 6.609,
Suetonius, Aug. IS, Dionysius Hal. 2. 10.3, 2.74.3, Plutarch, Romul. 22.3,
Dio Cass. 43 .24.4. But see Latte, RE Suppl. VII {1940), s.v. Todesstrafe,
1614 ff., for a different view of Roman conceptions.
3 Cf. Cicero, Clumt. 74, Fam. 8. 8. 3, Plutarch, Mar. S · s.

286
 EVMENIDES

decision in some cases of homicide was purely that of a magistrate exercising

jurisdiction and that Ennius gave hisAthena the words dicouicisseOrestem.uos
al, hotf aassitewith contemporary Roman legal practice in mind. 1

145 dico: c£ Terence, Eun. 961-2 at ne hot nesdatis,Pythias:I dicoedico

uobisnostrumwt iliumtrilemfilium (solemn, semi-legal context).

Orestem: a necessarycorrection; Greekterminations of proper names did

not appear on the stage until the time of Accius.3 The manuscripts of Plautus
have the en termination much le,s frequently than that in ~m; in no case is
-en metrically necessary;in two (Epid.612 [~ 197), True. 514) it is metri-
cally impossible; at Pseud.991, on the other hand (c£ Terence, Hee. 432,
Accius, Trag. 65'71), -em is metrically necessary. See Housman, Journalof
PhilologyXXXI (1910), 236 ft:, Leumann, MusH n (1945), 239 £ (=Kl.
Sehr. 110£).

ah hoe &ceaite: faassereoccun twice elsewhere in tragedy (Pacuvius

326, 343) and only four times in comedy (Plautus, Men. 249 [solemn wam-
ing to insolent slave), Rud.1o61, 1062 [semi-legal argument), Terence, Plwrm.
635 pegal discussion]). The phrase may be an ancient formula of acquittal
rdlecting a state of affairsin which the prosecutor in person brought the
defendant to court and exacted the penaltyin the event of conviction.

LXV
Scaliger compared Aeschylus, Bum.571-3 01yav &pi}ye1
Kai µa8eiv&eoµovs
lµoVSI n6AlVTE 'lfO:O'CX\1~ TOValavfixp6vov I Kai Too6",6irc,.,scxv ro
Karayvc.ooeij6(KT) (Athena's address to the jury) and wrote Ennius' words
as taare opinoesseoptumum,ut pro uiribusI sapereatquefabularitute nouerint.
Bothe compared 276--9 fycl>616ax&ls w K<XKots hrlOTcxµooI noA).ovs
Ka6apµoVS,Kai My£1v 6irov s,KTl I 01yav e· 6µoloos· w & Tc';>&
,rpayµcm I q,c,.,vetv hax&t)v ,rpos aoq,ov 616aaK@.ov(Orestes' speech
on arrival in Athens)and wrote sapereesseopinoroptimum pro uiribus:tacereac
fabularitutenoueris.Ribbeckwrote itasapereopinoesseoptumum,ut prouiribusI
1
Plautus sometimes refers to legal institutions that could have had no equiva-
lent in his Attic original; cf. Aul. 416.
a Ennius has ana (acc.) and crateras in epic {Ann.148, 5n). It is difficult to
give an account of Salm'acida{voc.) at Trag.347. Where comedy uses a proper
name taken from the Greek fint declension, the a of the nominative and
vocative cases seems normally to be scanned as short; but the matter is dis-
puted; cf. RH. Martin, CQ N.S. v (1955), 206 tf., VI (1956), 197, 0. Skutsch,
CQ VI (1956), 90, VD (1957), 52.
 COMMENTARY

tacereacfabularitute noutris,L. Mueller ego sapae opinoe~ optumumpro

uiribus: I tacereetfabularitute noueris.Vahlen kept what is transmitted and
interpreted it rather tortuously as 'qua sum condicionc tacerc optumum essc
censco, et hoe si feceris,sapcre pro uiribus (els 6vvaµ1v,6-n µaA1a-ra)et
fabularitute (aaq,<XAoos) noucris: scd in hac cawa iam mihi loqucndum est,
ut didici'. I think it better to assume that Ennius rewrote Orestes' speech
according to his own ideas of what was rhetorically appropriate and to
interpret the two trimeten transmitted as a generalsententiarather than a
statement about particulan: 'as a rule I think it the bestpolicy to keep one's
mouth shut and to show only such verbal clevernessas one's physicalstrength
allows; but (atque)in some circumstancesone may know how (noueris) to
speakwithou~ risk'. Ennius was given to altering the tone and emphasis of
the general statementshe found in hisoriginals (secbelow on frs.LXXXIV and
cv) and to adding general statements quite foreign to his originals (secbelow
on frs. cv and cvm).1

146--7 tacere opino esse optumum. et pro uiribus I •pere: cf.

Sophocles, 6.-.61.4KOOIJoS1'icny{\TEKai TCX iroOp' fflTI,Euripides, Hera-
kleidai476-7 ywooKl yapcny{\ TEKai TOaoocppovelv I KaAAta-rov.
The active form opinooccun twice in tragedy, 12 times in comedy (not in
Terence); the deponent only once in tragedy but extremely often in comedy.

147 atque: adversative; cf. Plautus, Aul. 287, Terence, Atulr.225 (and
Donatus ad loc.), T.L.L. II 1077. 1s ff:

&bulari. .. noueris: I can find no parallel for this construction in re-

publican drama but cf. Propcrtius 2.25.38, 2.28.13, 3.23.5.

tute: cf. Accius, Trag. 590; tuto is regular in comedy.

LXVI
Scaliger compared Aeschylus, Bum. 463-9 IKTE1va
-n'tvTEKovaav... Kai
ToovSeKOlvij/\~(as hralTioS ... avs· e[ StKOOCA>S
eiTEIJT)KplVOV
6(1<1'1V
(Orestes speaking); Ribbeck 616-20 owooirOT' ehtov IJavTIKolcnvh,
6p6vo1s,I 0\11< mp1, I O IJT)KWVaoo
avSp6s, ov ywatKoS, ov ir6AE<A>S
1
H. Fuchs, HermesLXX (1935), 245, writes t«ere opinoesseoptimum,at pro
uiribusI sapereatquef abularitutenouerisand suggests that the trimeten belonged
to a scene created independently by Ennius in which either Apollo briefs
Orestes on how to behave in court or Orestes in soliloquy repeats advice given
to him by Apollo. Such freedom in adaptation is not unexampled in what we
know of republican drama but the scenes postulated by Fuchs arc difficult to
believein.
288
 .BVM.BNIDES

ira-n'\p. I TO µiv
Zeus'O1'vµir{<A>v S{KcnovTOU6'6aov a6M1 µa&tv, I
~vMj fflcp<XUaK(A)s· (iµµ' tmama6m 1f<XTJX)S, 657 KOOTOVTO ~CA>,
Kooµa8' ws 6p8ws~p&>{Apollo speaking).
The verb fecissemakes it fairly plain that Orestes' deed and not Apollo's·
advice is in question.
The corrupt acciuscries out to be altered to aciusbut thisalteration leaves
hiatus in two places in the trochaic tetrameter (eachcase tolerable in itself'
but suspiciousin company) and a very queer phrase id iusfaare (for which
iusorartis no parallel). Furthermore the normal order of the doublet seems
to have been ius atqueaecum{c£ Ennius, Trag. 155, Plautus, Stich. 423,
Terence, Haut. 642, Cicero, Brut. 145,Tacitus, Ann. 14. 18, Fronto, p. 81 .15,
Claudian 10.313).1 On the other hand the adjectives usually arranged
themselves aequusac iustus(c£ Plautus, Amph. 16, Cicero, Phil. 2.72, Fin.
3 . 71, Seneca, Epist. 95. 52; rhythm dictates the refractory instances at
Cicero, Fin. 3 . 71 and Horace, Sat. 1 . 3 . 98). Accordingly I propose to write
id egoaecumaciustumfecissewith hiatus at the diaeresisof the tetrameter.
The substance of the Eooiao verse, which is plain whatever the detailsof
its restoration, departs a long way from the spirit of Aeschylus' Evµw{5Es.
The Aeschylean Orestes hesitates to claimeven TOS{Kcnov for his act {c£
Choe.1027), the Aeschylean Apollo claimsit for the will of Zeus. Ennius'
penonage pretends confidenceabout not only the justice but also the
equity of his act. The notion of equity {ToffllElt<ES) is absent from the
Evµw{Ses and quite rare elsewhere in Attic drama {c£however Sophocles,
O.K. 1125-7, fr. 703.2, Euripides, fr. 274, fr. 645 .6, Aristophanes, Neph.
1437-8).

148id ego aecam ac imtum fecisse: the subject of the accusative and
infinitive is omitted quite frequently in republican drama; c£ Plautus, Capt.
256, Terence, Haut. 16-17, Phorm.54etal., A. F~ N]bbCXXI (1880), 725-
34-The context always precludes ambiguity.

expediboatque eloqaar: c£ Terence, Phorm.197-8 -ado quidportas,

obseao."'4utid, si potes,uerboexpedi.I -f aciam.- eloquere.
The form expedibooccurs three times elsewhere in tragedy {Pacuvius66,
281, Accius 490), only once in comedy {Plautus, True. 138 [high-&lutin
dialogue]); expediamseems to have been the normal form (Plautus, Amph.
912, Terence, Andr.617, Phorm.238).

1
For hiatus after the second arsis of the tetrameter see Jacobsohn, Q!y,ut.
Plaut. p. 4; for hiatus at the diaeresis see above on v. 17.
2
Vahlen seems to have been dimly aware of this when he suggested id ego
ius atqut aecumfecisst;see Ind. lectt. Berlin 1888/9,7 f. (= Op. ac. 1406 f.).

19 JTO
 COMMENTARY

HECTORIS L YTRA

Vcrrius is represented by the tradition as having given Ennius the title

Hectorislyrat(Fcstus,p. 334. 16); DiomcdcsLustra(Gramm.I345. 3, 387. 30);
Nonius Hectoristlystrat (pp. 467 .39, 472.23, 489.29, 490.6, 504.35,
sn. 10, 518 .24), Hectoris tlistrat (pp. 3SS. 17, 407 .24), Heaoris lytra
(pp. n1.18, 399.8, 469.30) and Lytra (p. 222.27). The Heaoris tlystrat
quotations all come pretty certainly from Lindsay's list 10 'Ennius', a list
basedon a volume or volumes containing this play and the Telephus.About
theothers there arc varying degreesof uncertainty; the Lytra quotation may
come from the work of Flavius Caper which seems to lie behind Nonius'
third book.
Stephanusprinted the sixteenquotations under two titlesHectorislytraand
Lustra but later editon have assumed the existence of only one play. The
quotations can all be interpreted as from a tragedy set in the Myrmidon part
of the Greek.encampment before Troy and climaxed by Priam's ransom of
Hector's body from Achilles.W clckcr made the tragedy beginwith Priam's
arrival at Acbi11cs'
tent and suggestedthat it was an adaptation of Sophocles'
<l>pvyES. ScholPargued that it began even before Patroclus' departure to do
battle with Hector and that Ennius combined the plots of Aeschylus'
Mvpµ166~. NT)petSes and •EKTOpos AVTpaft <I>pvyES. Scholl's view has
won an extraordinary number of adhcrcnts.3
There isnothing absolutelyimprobable, despiteAristotle,Poet.s.1449b 13,
about a tragedy covering the events narrated in books XI-XXIV of the niad.
Aeschylus' 'Aycxµiµvoov and Evµa,(Ses,Sophocles'Tpaxlvun, Euripides'
•Av6poµaxT'I, 'h<nfSesandI~j:x>ta have actionsspreadover long intervals
of time. The notion that Ennius was responsiblefor combining the plots of
three Attic tragediesis harder to swallow, there being no hard evidencethat
the republican poets ever worked as freely as this.
Only one of the sixteen quotations (fr. uaav) hasto be taken as from a
scene set in time before the arrival of Priam at Achilles' tent.
One might allow that Ennius himselfadded to the Acschylean•EKTopos
AVTpaa sceneshowingthearming of Achillesand hisdeparturefor battle but
there remains the difficultythat Aeschylusemployed a chorus of Phrygians,
Ennius one of Myrmidon guards (fr. LXXV).

1
Dit AtschylischeTrilogie,p. 428, Dit griech.Trag. p. 136.
2
Beitr. pp. 472 ff.
3 Bcrgk (N]bb LXXX111[1861], 629 [ = Kl. phil. Sehr.12991), C. Bailey (CR
xvm [1904], 171) and Leo (Gesch.p. 193 n. 1) arc the only scholars I can find
expressing scepticism.
 HECTOIUS LYTRA
The title of the play obviously could not have been mcrous LYlill.
A second-century Roman would have smiled to sec mcrom LVSTRA on a
bill-boaid but many have thought this to be Ennius' title.I The subject-
matter of the play would demand that lustrabe interpreted as theequivalent
of the Greek AVTpa.There is no evidencethat the word was so usedin the
common language of the early second century; neither lune nor any of its
compounds translatedAVE0'9cn; the Latin verb was redimere(Plautus, Asin.
106,107, Capt. 337 et al.). Varro's etymology at Ung. 6.n. lustrumnomi-
ru,tumtempusquinquemu,le a luendo,id est soluendo,quodquintoquoq~ anno
uectigaliaet ultro tribuuiper censorespersoluebantur,proves nothing about
ordinary usage. HEcrous LYnA could not have beenwritten thus by Ennius
birosdf Ribbeck accordinglymade thetitle mcrous LVTRA. However what
lies behind our muddled tradition may be simply a classical
transliterationof
•EK,-oposAVTpa. 2

The title •EKTopos AVTpais recorded againstthe names of Dionysiusand

Timcsitheus as well as Aeschylus.Hyginus, Fab.106, which relates events
from Agamemnon's seizure of Briscis to Achilles' surrender of Hcctor's
body, is headed Hedorislytra.The plot of an Attic tragedy often lies behind
Hyginus' narrative but one can never be sure how much comes from the
prologue and how much from the action.
Our ignorance of the preciseextent of Ennius' plot makes it impossibleto
apportion a number of fragments relating to the siege of Troy betweenthis
play and theAchilles;sec above, pp. 162 ff.

LXVII
Thisfragment appearsto refer to the revenge taken (Scaliger's hostiuite manu)
or to be taken (Vahlen's hostibitismanu, Timpanaro's hostibiseminus)by
Achillesfor Patroclus' death.
For the phraseology of Scaliger's text and texts similarly constituted cf.
Homer. n. 20.-462 TOV µtv 6ovpl ~~v, TOV & axe6ov c!rop1 ~.
Cicero, Cato 19 nequeeminushastisaut comminusgladiisuteretur,Ovid.Met.
3 . 119 comminustnse f erit, iaculocaditnninusipse.

149machaera: only here in republican tragedy, frequent in comedy as a

weapon of the mercenary soldier, twice in the Annalesin obscure contexts
(400, 597). The word is likely to have come from Greek:through thecom-
mon languagerather thanthe scriptsof Attic drama. M&x<X1pa occurs once in
Attic tragedy (Euripides,Hile.1206) and often in comedy but never indicates

' Most recently R. M. Ogilvie,JRS u (1961). 35.

3
See Introduction,pp. 59 f.

291
 COMMENTARY
amilitaryweapon.1 Tragic soldicn wield the e,cpos
or the cpaayavov.comic
the o,ra&t, {Mcnandcr,Ptrik. 165, Sam. 314, 315, Misoum.B t s. C t 25).

huta: hereand twice elsewherein republican tragedy {Ennius165, Trag.

inc. 235).fairly common in comedy of the soldier's spear and the athlete's.

LXVIII
2
Scholl regarded thisfragment as having beenspoken by Achillesbefore the
departure of Patroclus for battle. It could as easily have reference to the
decision to hand over Hector' s body to Priam.
The words transmitted will not form any regular metricalpattern and the
phrase ut hoeconsiliumAchiuisauxili (auxiliicodd.)faatis anomalous.3I have
printed, without much confidence,Lindsay'sarrangement of the text.4 For
the minglingof catalecticand acatalectictetrameters c£ vv. 185-8, 219-21,
296-9.
150 at: makes the request more urgent; c£ Plautus, Most. 38-9 at te
Iuppittr I diqueomrresperdant,Livy 1 . I2. 5 at tu pa.ta deumhominumquehinc
saltemarcehosteset al. •AXJ..a
is similarlyusedin Greek; c£ Aeschylus,Theb.
n6-19 &XA"er,2ev iraTEp ir<XV"IV\ts, I iravi-oos&p,,eovSatoov &Ac..:>a1v
et al.

omnipotem: perhaps a poetic calque on ircxy1<p<X'ffls. In the remains of

archaic and classicalLatin it appears to be restricted to poetry. Its early
adoption by the Christians confirms that it was absent from the sacral
language of the pagans. See Fraenkel on Aeschylus,Ag. 1648. For the form
of the compound epithet sec above on v. 45.
Omnipotensoccurs unaccompanied elsewhere in Latin at Virgil, Aen.
4.220. 10.615.

151 ted: this and other -J forms were obsolescentin the early second
century except in the language of religion and law; in the remains ofEnnius'

1
Cf. Shipp, Glotta XXXIV (1954). 149 f. The diminutive however seems to
have been so used; cf. Menander, fr. 793 &<mi610vhrp1aµ11vTI Kai IJ(X)(afp10v
(~ Plautus, Cure. S74-S ita me macluieratt cliptus. .. btnt iuumt, True. 5o6
macluieram tt cliptumposctbatsibi).
:I D..!.._
.IXUJ'. p. 497.
3 Contrast Plautus, Amph. 92, Aul. 715-16, Cure. 267,Epid. 676, Pom. n37,
1277, Terence, Haut. 992-3. A. Rciffenchcid, Ind. lectt.Brtslau 1885/6, p. 7
(not seen by me), defends auxili.
4 Cf. Timpanaro,SIFC XXI (1946), 79-80, 0. Skutsch. CQ N.S. xm (1963),
90 n. 2; aliterStrzelecki.E.osXI.m (1948/9), 159-61.

292
 HECTORIS L YTRA
writings one caseistransmitted(Var.4S medtgo (Manutius:mt tt tgocodd.]),
in comedy only a few; editon concur in restoring medand tedto avoid hiatus
after the anis of the first foot of iambic and trochaic verses.Sec Mauren-
brecher, Huuusund Vtrschltifang, pp. 107 ff.

ezposco: only here and Plautus, Ptrsa 495 (paratragic context) in

republican drama; the simple form posart is common in comedy.

faat: thisand compound forms do not occur in officialinscriptionsof the

second century; they are relatively common in tragedy (4 instances as
~
against8 of sitm etc., 19 of simetc.)and uncommon in comedy ( 22: 196: 684
in Plautus, 1 (Hee. 610):85:254 in Terc:ncc,2:1:8 in Naevius et al.); they
tend in both genres to occur at the ends of iambic and trochaic versesand in
formulaic expressionslikef orsJuat.See F. Thomas, RechtrchtssurIt subjMKtif
latin(Paris,1938), pp. 10 ff.
LXIX
We have here a periphrasisfor Athtrunsconcerningthe context of which it is
uselessto speculate.
For the periphrasiscf. Seneca, Tro. 178-80 tumsrusa uallisaptritimmmsos
sptcus I tt hiatusErtbi... iter... praebtt, 430-1 Stygis profandaeclaustratt
specusI laxantur,Thy. 105~ grodtreadinfernos
obscuri specusI atnntm(/Ut notum.

152 inferum: probably the genitive plural of inftri, 'beings of the

lower world' (cf.Plautus, Aul. 368, Pacuvius, Trag.289, Accius, Trag.57);
not to be grouped with Ennius, Trag.98, Accius, Trag.479, Trag. inc. 75
(infera,'lower world') or with Seneca, Htrc. 0. 15, 743 (inftri, 'lower
world').

aastos: twice elsewhere in tragedy (along with uastitasonce, uastitudo

4 times, uastesareonce); only once in comedy (Plautus,Bacch.1053).

LXX
Schollreferredthisfragmentto the fighting describedin n;adXI. 1 One might
with equal plausibilityrefer it to any action involving Hector. It could come
from a prologue relating events long past or from a messenger'sspeech
desaibing a recent happening off stage, eitherthe actionin which Patrodus
lost his life or that in which Hector lost his.
There is little talkof the strategicaldeployment of troops in the Iliad; the
1
&itr. p. 491. Cf. Homer, II. II. S~I.

293
 COMMENTARY

early epicpoets concentrate attention on the deedsof individual heroes.1 In

the battle-pieces of Attic tragedy, 3 on the other hand,the heroes are treated
as a-rp<rl"flyo{ rather than ,rp6µaxo,. Ennius' two trimeten gave a Roman
cast to what was an already established tragic moti£

153 Hector fei ~rnmat annatos educit foru: Mercerus' ui summais

universally accepted; the phrase occun ofien in drama (Plautus, Amph. 210,
Cas. 80, Mere.45, Terence, Ad. 493); but one must ask why Hector should
need to exercise uis either in arming or leading out his troops.
Ennius draws on the military language; c£ Plautus, Amph. 216-18
AmphitruocastrisilicoI producitomnemexercitum.contraTeloboaeex oppidoI
ltgionu educuntSU4S nimis pulchrisarmispraeditas,Cato, Orig. 83 tempus
exerdtusex hoelocoeducendihabebis et al.
Forasfrequently in republican drama accompanies verbs compounded
forascf. Plautus, Cas. 798 et al.
with ex-; for educere

154 cutrisque cutra ultro iam ferre occupat: Vossius' alteration of

fme to con/me removes two anomalies. Hiatus after the thesis of the third
foot is rarely transmitted in extant trimeten and scholan prepared to tolerate
hiatus elsewhere deny it here.3 Nevertheless Ennius 249, Pacuvius 293, 364,
Accius 84, to say nothing of the comic examples,4 are difficult to normalise.
Castracastrisconferrewas a regular phrase of the military language; c£
Cicero, Diu. 2. 114 castraenim in Thessaliacastrisconfutaaudiebamus, Caesar,
Ciu. 3. 79. 3, Hirtius 8. 9. 2 et al.SBut castriscastraferre is a possible poetic
variation; for the replacement of the compound form of the verb with the
simple in tragedy see above on v. 14,

1
Cf., nevertheless, 15.281 ff., 16.171 ff. The general charact.er of the
Homeric narrative is well brought out by comparisonof 16. 394 ff. with Virgil's
imitation at Aen. 10.238 ff., 285 ff.; Virgil makes Turnus a o,-pcrr1ly6s. SecL.
Wickert, PhilologusLXXXV (1930), 4SS ff.
a E.g. Euripides, Hile. 707 ff., Herakleidai799 ff., Phoin. 1090 ff: Cf. Plautus,
Amph. 216 ff.
3 E.g. Lindsay, E.L.V. p. 231.
4 These are collected by P. Friedlander, RhM LXD (1907), 73 ff., and A.
Klotz, HermesLX (1925), 332 ff.
s The polyptoton looks poetical (cf. Homer, II. 13. 131 &cmis &p' &crnl6'
fpE16E, 1<6pvs1<6pw,&vipa 6" &vf}p, Euripides, Hile.666-7 hrneOcn6' hrnfis
iljaav &v8CAl11'>.ta~1 I TETpa6po1aiT' &v,;• &pµa8' &pµacnv,Phoin. 109s~
ltiSpous s•l,rn~ ~ l,rn6ToosI ha~. Herakkidai836-71ravs maMcxx&ts
,ro6(, I &vt'IP
6' m· &v6plo,-as) but the employment of the phrase by Cicero and
Caesar fixes it in the regular military language.

294
 HECTORIS L YTRA

LXXI
This is almostcertainlya fragment of a speechaskingor advisingAchillesto
hand over Hcctor's corpse. From a similar speechin Aeschylus'•EKTOpOS
AV"Tpa comes (fr. 266) Kai TOVS 6cxv6vrasel 80.as evepyn"ElvI EIT' ow
KCXKOVpyetv, &µt 1Se~{oos fxa ITc;,µ{rrexa{pav µf\-reAVfflla6mcpfhTOVS. I
fiµG'>vye IJWTOI Niµans lae' \7TT'Effflpa I Kai TOO6av6vTOs It AfKTt
1rpaaae1 KOTOV. The sentiment is more like that in a similar context of
Euripides'·hcETfSes: 594-7 Iv Set µ6vov µ01· TOVS 6eovsfxe1v,6ao1 IS{Kf1V
<n~VTal. TavTa yap~ve· oµov I v{Kf1V S{ScAXJW. ~ s· ovSev
Afyet I ~pOTotcnv,i\v µ'I'\TOV6eov XP~lOVT'fx1J. One may make the
same point about the content of this fragment as about LXVI. Talk about
equity was almost certainlyabsentfrom Ennius' originalif thisoriginalwas
Aeschylus'•EKTopos AV'Tpa,quiteprobably absentif it wassomeother play;
the personagesof Euripides'•11cni6es talk simplyofv6µ01 ~G'>v (378),o
TT<XVEAAf\voov v6µos (5.26,671), v6µos ,ra).mcs Scnµovoov(563) and S{Kfl.
Virtushere docsnot standfor virtue in generalbut rather for the particular
virtue of the warrior; for the contrast between uirtusand justice, etc. cf.
Trag. inc. 116-17 tandemuirtuttmistamuenitt ttmpuscumg,auittrgtmts I si
nequtltgts nequtmorescogunt,Livy 1 • 26. 12 absoluenmtque admirationtmagis
uirtutisquamiure causae;justice, etc. are set alongside uirtusby Cicero at
ManiI. 36 nontnim bellandiuirtussolumin summoacpe,fectoimpe,atore quaerenda
est sedmultaesunt arttseximiathuiusadministrae comittsqutuirtutis.acprimum
quantainnocentia dtbtnt tsse impe,atores,
quantadtindein omnibusrebusttmpt-
rantia,quantafidt •.. and by Augustusat Mon. Ancy,.34 clupeusaurtusin curia
Juliapositusqutm mihistnatumpopulumque Ronumumdartuirtutiscltmtntiaeque
iustitiaeet pittatisCdUSSa (&pE-ri\v
Kai hnebcaav Kai 61KatOOVVT1V Kai EV<Ji-
pe,av) ttstatumest per tius cluptiinsaiptionem.

156iu at.qaeaecam 1e a mallsspernitprocal: an odd phrase; sperne,e

normally takes a simple object in republican drama. The genesis of the
Ennianphrase can be seenin Plautus,Mil. 1232 illt illassptmit, seg,eglllabse
omnisextratt unamand Plautus, Capt. 517 spesopesauxiliaque a me seg,tganl
se (Dousa: tnt cod.).
spernuntque
LXXII
Th.isfragmentis grosslycorrupt and no good purposeis servedby attemptsat
restoration.1 A little can be said about the subject-matter.There is no ques-
tion of taming real horseson the fieldof battle or of curbing real horseswith
1
For the latest and by no means leastintelligent attempt secMariotti, SIFC
XXXI (1959), 233-s: due tt qu4druptdum iugo I inuitamdomaln.frmatt iungt,
ualu/4quomtmacu,I in.frtnatiminis(nequitur).
295
 COMMENTARY
threats. Some person or personshas been acting obstreperously.The word
inuitomsuggeststhat it is a woman; one might parallelAeschylus,Ag. 1066-7
X<XAtvov s· oVKrnfo-rarm q,ipe1vI nplv alµCffllpov!~aq,pf3E(16mµEVOS
and 1071 EIKova'oo,ayK1J Tij& Kafv1aov3VY6vbut hardly Lucilius1041-2
annt egote ""'"""' {Lacbmann'. acuamcodd.) atqueanimosomI Thessalmnut
itulomitom f,enis subigomante domemque(c£ Anacreon417 .4 f. Page). How-
ever it is difficultto imaginewho the woman could be in thisplay. ~orum
suggestsmen; c£ Aeschylus,Ag. 163S>-40TOVSEµ1')m18avopa I 3£V~W
~{ms, Sophocles,Ant. 291-2 ovs· vrrolVY'i> I i\6q,ov6tKOO(t.)S
eixov,
wsCTTtp-yEtV iµt, El. 1460-3 ~ et TIS<XVTWV ~,rfaiv KEVcxTS napos I
!~pe,-· av6pos TOO&,vOv 6p&v VEKpoV I o-r6µ1aSix'lT<XlT&µa, µT)SE
npos ~fav I !µoO 1<0i\ao-roO npoo,vxcl>v q>v0'1J cppwas,Livy 39. 2s. 13
Philippumut etJUum tenaam nonparentemftenis asperioribus a,stigandumesse.
But again it is difficultto identify the men in question.
lnf,enareoccurs elsewherein Latin only at Accius,T,ag. 1 s,Cicero,Pis.44,
Virgil, Am. 12.287, Livy 37.20.4, Pliny, Nat. 9.100. The simple.frenareis
absent from republican drama but quite common in classicalprose and
verse. For ;,._as a tragic prefix see above on v. 26.
Tenaciaoccursonly here in republicandramaand may have been a tragic
neologism.Cicero has tenadtasat Nat. deor.2. 122.

LXXIII
This fragmentcomesfrom an articleput together from listsapparentlycom-
piled by Nonius himselffrom the original texts. ConstititcredoSaimander
arbortsuentouagantis almostcertainlywhat Nonius thought he saw in the
text of the Hectorislytra.yet while uentocould be taken c!nroKOIVOV with
constitit(c£ Virgil, Eel. 2.26 cumplacidumuentisstaret·mare,Georg.4.484
lxionii uentorotaconstititorbis)it makesno sensewith uagant.Nonius must
have misread a uacant.
Schollreferred the fragment to the fight between Achillesand the river-
god narrated in n;aJXXI, Bergk to the fight between Achillesand Hector.
Certain dose parallelsin Greek and Latin literature make it seem likely,
however, that the fragment concerns some miraculous event or divine
epiphany;the parallelsare Euripides,Ba. 1084-s afYTlcn6' cxl8ftp,atya 6'
VAlJ.loSVQffl)I q,vi\i\'
etxe,&T)pwv s·0\/K<XV
i\Kovaas~v, Aristophanes,
Orn. 777-8 ,rri\~ 6Eq,Oi\a TE 1t'OtKii\a
&T)pwv,IKVµaraT' faJ3eaeVT)VEIJOS
af8pT),Thesm.43-4 !xhw SE1TVO<XS cxl8ftp,I t<OµaSEn6VTov
Vf)VEJ.loS
µ1')KEi\a6efTCA>,
Limenius7-9 (Powell 149) nci[s 6e y ]6:8T)cn n6i\oS ovpa-
vtoS [&weq,ei\oS6:yi\a6s,] I [v]f}viµovs 6' laxev al81')p &e[i\i\oov
TaxvmT]ets [6p6µ]ovs, I i\f\~e 6e j3cxpv~p6µCA>v NT)[piCA>S3aµeves
o[t6µa, Alcaeusof Messene,A.P. 7. 412.s-6 f,oov6' fCTTT)cnv ..• •Aa(t.)11'6s
(grief for deathof Pylades), Virgil, Aen. 9. 123-s obstipuereanimisRutuli,
296
 HECT ORIS L YTRA

ipseI turbansMessapus
conterritus equis,cunddt#retamnisIrtiuu sonansrtuOCdll[Ut
pedan Tiberinusabalto,10. 101-3 eodiante deumdomusaltosiksdt I et treme-
factllsonotellus,siletarduusaether,Itum Ztphyriposuae,premitpLidtldaequora
pontus, [Tibullus) 3. 7. 121-34- 1 The word aedo suggests that nothing in the
action of the play is referred to; rather that we have a heanay report of
earlier events (c£ Plautus, Epid. 34 Muldberaedo amu, fecit '[Ude habuit
Stratippocles
).i
LXXIV
The text of this fragment is corrupt and its specific interpretation obscure3
but one can give it a fairly secure context: news of Patroclus' death hascome
and Achilles is without arms, powerless to avenge his beloved. 4

16o canct.ent: once elsewhere in tragedy (Accius 72) and only twice in
comedy (Plautus, Cas.792, Epid. 162); comparatively uncommon in Caesar
and Cicero (who both use it as a deponent).

LXXV
This fragment must come from an address by Priam to the guards at
Achillestent.
Janus PalmeriusS restored it as per uoset nostrumimpaium etfidemMy,mi-
donum,uigiles,commiserescite. Bcrgkand Vahlcn thought they could make
sense of it with only one alteration: per uos et uostrumimperiumet falem,
My,midonumuigiles,conmiserescite. However three grave difficulties can be
made about the text they offer.
First, the words will not fit into any of the recognisedversepatterns of
republican drama. 6

1
One has to guess at the context of Ennius, Var. 9-12 mundusaieli U4Stus
constitit. .. constitere
amnu perennesarbo,esuento uacant.
a 0. Skutsch (London seminar, IS November I9SS) proposes to replace
credowith cursu.See now HSCPh LJOCI (1967), 136 f.
3 Ribbeck {cf. Co,oll. p. xxiv, Die ,om.T,ag. p. 123) wrote in his final
edition ei ipsicunctent. .. and made the speaker some Achaean refusing Achilles'
request for armour after the death of Patroclus. Vahlen (cf. Ind. lectt.Berlin
1888/9, s ff. [ = Op. ac.I 403 ff.]) wrote ut ipsicundent?and made the speaker
Achilles pointing out his lack of armour in reply to Iris' command to go into
battle and recover Patroclus' body.
4 The mood of cupiantmakes it very difficult to accept Welcker's view (Die

griech. T,ag. p. 136) that the words are spoken by Achilles to Priam in an
account of past events.
S Spicilegium {Frankfurt, I 580); repr. in Gruterus, 1.Ampas,
IV 787.
6 For the trochaic pentameter alleged by Bergk see above on fr. XLDI.

297
 COMMENTARY
Secondly, while the appeal per uos can be paralldcd in Cic:cro'soratory
(Plane.103 nolite iudias per uosperfortwuu per liberosuutros inimids mds • •.
darelaetitiam)and in thewritings of the historians (Sallust, lug. 14.25 patru
conscriptiper uosper liberosatqueparentisnostrosper maitstatem populi Romani
subuenitemihi misero,Hist. 2.47.13, Livy 29.18.9, Curtius 9.2.28) and in
classicalpoetry (Lygdamus I . Is) it hasno parallel in republican drama.Here,
in certain very solemn passages,per is disjoincd from its object by the pro-
nominal subject and/or object of the verb of supplication (secabove on v. 3).
The construction was already old-fashioned in tone in the early second cen-
tury and Cicero's noliteper uosetc. seemsto spring from a misunderstanding
of its nature. One should not try to take uos as the object of an unexpressed
verb of supplication.This was a Greek construction (c£ Euripides, Andr.892-
3 npos a& Tc;;.;v& yowarCA>v, I oiKTipovftl!QS)and seemsto appear first in
Latin at Rhet. inc. Her. 4.65 (perte ea quaetibi dulcissimasunt in uita miserere
nostri),in what looks like a rhetorician's literal translation of a Greek story,
and then sporadically in poetry (e.g. Lucan 10. 370, Silius 1.658).
Thirdly, the idea that mere uigilesshould possessa thing like imperiumis
absurd. The word imperiumhad a semi-religioustone lackingin Greek words
for power and authority; it belonged properly to the officiallanguage and
denoted the authority of the chief civil magistrates. Plautus uses it, perhaps
in jest, of the authority of the head of the private Greek household' but
there is no parallel anywhere for the use which Bergk and Vahlcn foist
on Ennius.3
I cannot sec how to restore the fragment without extensive rewriting.
That there stood in Ennius' text a variant of the appeal perfidem (Plautus,
Trin. 153, Terence, Andr.290, Sallust,lug. 10. 3 et al.) isfairly certain; for the
coupling offideJ and imperiumc£ Cann. deu.ap. Macrob. Sat. 3 •9. II ut me
meamquefidem imperiumquelegionu exerdtumque nostrum qui in his rebus
gerundissunt bentsaluossiritisesse.3One should take imperiumetfidan Myrmi-
donumtogether; according to some Roman politicians and political theorists
imperium belonged in the last resort to the whole populus Romanus (c£
Cicero, Phil. 3. 37 cum senatus auctoritatempopulique Romani libertatem

1
Cf. Asin. 87 doteimperiumuendidi.It is possiblethat the juridical language
already used it so; cf. Paulus, Fest. p. 55 .9 nuptiali iure imperiouiri subititur
nubens,Ulpian, Dig. so.17. 4 uelle non crediturqui obsequiturimperiopatris uel
domini.
a 0. Prinz, T.L.L. vni 574.38 ff., puts the fragment at the end of his
examples of imperium,de certarumpersonarum potestate/apud
exteros/demonarchis
but is obviously nonplussed.
3 Ennius, Ann. 107 has fales and regnumtogether, Caesar, Gall. 2.3 .2,

2.13.2, Cicero, ad Q. fr. 1.1.27, Curtius 8.4.21 fales and potest41,Livy

34. 35. 10, 37 .45. 3, 38. 31.6 .fidts and dicio.
298
 HECT ORIS L YTRA
imptriumque Livy 4. s.1, Ulpian, Dig. 1. 4. 1) and theearly tragedians
defendat,
were ready to transfer thiscollective imptriumto the heroic world of Greece
(c£ Accius 231 Argiuomimperium).
Ennius' Priam draws on the solemn formulae of the Roman deditioinfit/em
for his appeal. Something of the characterof Euripides, Mtd. ~II a:>,;A•
6vt-oµaf ereTijaSe ,rposyeve1aSos I yOVo:Toov -n: Tc1>v
acZ>vlJ<Ea{a -n:
y{yvoµoo, I oll<TlpovolKT1p6vµe is likely to have stood in his original.
ITposTc1>v ,rposToO yeve{ov,irposTc1>v
8ec1>v, yova-roovand irposTiis
6e~1asarc the normal csprcssionsin Attic tragedy; appeals to abstract quali-
irpos~{as ••• ). In Roman comedy
ties arc rare (c£ Sophocles, O.K. 515 J.111
the normal appeal is per deosor per geniumor pergenua or per dexteram;
abstracts appear when the tone rises (c£ Plautus, Asin. 18 ted obttstorper
senectutem tuam,Capt.245-6 perf ortunamincertamet per meite ergabonitatem
patris,I perqueconseruitium
commune).

162 comrnis«escite: a high tragic form; twice elsewhere in republican

tragedy, only three times in all comedy (Plautus, Rud.1090, Terence, Hee.
129, Turpilius, Com.211); comedy hasmisertsarethreetimes in very solemn
contexts (Plautus, Epid. 526, Trin. 343, Terence, Haut. 1026); misereriis the
regular form (37 times in comedy, 7 in tragedy).

LXXVI
The noisy arrival is a commonplace of tragedy (c£ Euripides, I.A. 317, I. T.
1307-8, Rhes. 11-16) as well as of comedy (c£ Plautus, Bacch.583-6, 1120,
Cure.277, Trin. 1093). Scholl referred the fragment to the arrival of news of
Hector' success in his attack on the Greek encampment. It could equally
well concern that of news of Patroclus' death (soRibbcck) or the arrival of
Priam in quest ofHcctor's body.

163 quid hoe hie clamoris: c£ Plautus, Aul. 403 sed quid hoeclamoris
oriturhincex proximot, Trin. 1093 quidhoehieclamoris
audioanteatdismtast,
Caccilius, Com. 245 quidhoeclamoris t

· quid tamulti est: c£ Plautus, Poen.207 quidistuctumultiestt, Turpilius,

Com. 154 quid hie tumulti ante/orest, Pomponius, Atell. 121 quid hoe tst
tumultit

nomen qa.i usurpatmeum: contrast Plautus, Bacch.1120 quissonituac

tumultutantonominatmt t, Cure.304 quis nominalmet, Mil. 901, Most. 784,
Rud.98, 678 b, 868,Terence, Phorm.739, 990, Trag. inc. 1 quisenimest qui
meumnomennuncupatt, 97 quismeumnominansnomenatdetxciett
 COMMENTARY
Q!.i is transmitted quite often as an indefiniteor interrogativesubstantive
in both tragedy (Pacuvius 228, Accius 6-4,343,562,625) and comcdy.1 Its
enclitic position here is highly unusual.
For usu,pare,'nominare •, c£ Cicero, Off. 2. 40 Ldelius,is qui Sapiffuusur-
patur.Plautus has the verb six times (&ah. 149, Cas. 631, Cist. sos, Persa
736, Trin. 846, Pseud.135), mostly as a high-&lutin synonym of uti; else-
where in republican drama it occurs only at Trag. inc. 127 and Accius, Trag.
praet.29.
LXXVII
Lindsay's analysis of the structure of Nonius' dictionary tends to confirm
Columna's view that this fragment followed LXXVI directly. Nonius illus-
trated TVMVLTI and STREPm with pieces taken successively from the list based.
on Ennius' Hectorislytraand Telephus. 3

LXXVIII
165aes sonit: Nie. Faber's emendation is almost certain; c£ thecorrup-
tion at Catullus 41. 8. The phrase can hardly refer to theblast of trumpets (so
T.L.L. I 1073 .48). The clashof iron-tipped spears on brazen shieldsand
helmets is meant; c£ Euripides,Herakleidai 832 ,r6aov T1v·avxets ,ra-ra-
yov acm{&>v ~IJElV, Ennius, Ann. 363 tum clipeiresonuntet Jerristridit
I
acumen,Virgil, Aen. 9. 666-?tumscutacauaequedantsonitumjlictugaleoe.3

&anguntarhastae:c£ Euripides, Plwin. 1399-402 &,ro 6' C6pava•

&<povSopv ... ~v s•&KoVT•C6pavaev,Ennius, Ann. 405 semperalnm-
danteshastasfrangitqueq,ustitqut,Hostius, Cann. fr. I percutitatque hastam
pilansproepqrukref,angit, Plautus,Amph. 232 tel4f,angunt.

terra sudat sanguine: a bizarre phrase imitated by Virgil (?) at Am.

2. 582 Dardanium totienssudaritsanguinelitus.Lucretius (s. 1131-2, 6.1147-8)
and Livy (22. 1 . 8, 27. 4. 14) usesanguinesudarein somewhat more appropriate
circumstances. Ennius has conBated two motifs of the traditional battle
narrative, the sweat pouring from the bodies of the contestants (c£ Homer,
n.16. 109-11 atel s· apy~ ~ET· &a6µan, K0:6St ol 16poosncxv-ro8ev I
~ ~v elxev I aµnvevaai,Euripides, Plwin.
(ppm,, oust 11'1J
'lt'OAVS
1
See 0. Seyffert, BPhW xm {1893), 277 ff., Lofstedt, Syntactican (Lund,
1933), pp. 79 ff.
a See Lindsay, Nonius Marcellus'Dictionary, p. 26, RhM LVD (1902), 202.
3 Nevertheless, contrary to epic practice {the simile at n. 18. 219 is the
significant exception}, the battles of Attic tragedy are signalledby the trumpet;
cf. Aeschylus, Pers. 395 (contemporary fighting), Sophocles, Ai. 289 ff.,
Euripides, Heraleleidai830, Phoin. 1102.

300
 HECTORIS L YTRA
1388-9 1TAE{(A)V Se Tots 6plootv lOTc:Uaaa· 16~ Ii\ TOla1 6p6xn, 61a
cp{Ac.>V
6pp<A>6{av, Ennius, Ann. 406 totumsudorhabetcorpus)and the blood
pouring upon the ground {c£Homer, n 8.65 ~ s· 001,laTl yaia, 10.484
lpv6a{wro 6' alµa,-1 yaia, Euripides,Phoin.1152 e,,pav 6' ISevov yaiav
atµcrros (xxxis). The alliterative pair sanguiset sudoris common in Latin
writing {c£ Ennius, Trag.347, Cicero, Leg. agr. 2.16, 2.69, Livy 2.48.2,
7.38.6 et al).
LXXIX
The unexpressedsubjectof cernunthave good fortune aswellastheir weapons
helping them. Fortunanever indicatesimpersonalchance(TVXTl) in republican
drama except at Pacuvius, Trag.366 ff., where the speaker is discoursingon
philosophicaldoctrines; when unqualified it is normally optimistic in tone.
Its nearest equivalent in Attic drama would be 8e6sor Safµc.>v{contrast
Aeschylus,fr. 395 cpwl SeTei'> avcmw&1v 8e6sand Euripides, fr.
K<Xl,lVOVTl
432. 2 Tct>yap irovown xoo 8eosavAAaµ~1 with Terence, Phorm.203
fortisfortunaadiuuat).

166 saeuiter: the regular form in republican drama, 6 times in comedy,

3 times in tragedy; neither saeuiternor saeueseemsto occur in classicalLatin.

fortunaferro: for the asyndeton bimembre c£ the tragic exampleslisted

above on v. 9 and the formulaicfortefortuna (Plautus, Bacch.916, Mil. 287,
Terence, Eun. 134, 568). Neither Columna'sfortunafmi nor Bergk'sfmo ac
f ortunais necessary but the latter emendation has the merit of directing
attention to the fact that the longer word rarely precedesthe shorter in these
asyndeta.
For the combination offortuna andferrum c£ Virgil, Aen. 10.421-2 da
nunc Thybripaterf mo quodmissilelibroIf ortunam.

cemant: HomericKp{voVTa1 (n. 2. 385, 18. 209); a usagehard to explain;1

to be found elsewhere only at Ennius, Trag.232, Ann. 196,555, Pacuvius,
Trag.24, Accius, Trag.326, Lucretius 5. 393, Virgil, Aen. 12. 218, 709. The
compound tkcernereis used with the same sense at Ennius, Ann. 133, fre-
quently in Virgil and Livy, occasionallyin Cicero and Caesar. Comedy has
tkcernere'decide' often and cernerewith the same senseat Plautus, Bacch.399,
Cas. 516, Trin.479. For cernere'perceive' see above on v. 13. Certareseemsto
have been the regular word in the common language for 'fight for a deci-
sion' (Plautus,Mere.345, Mil. 714, Persa238, True.948,950, Terence, Phorm.
20, Pacuvius, Trag.25).

1
SeeEmout, BSL XXIX (1929), 82 ff. ( =Philologica1, pp. 83 ff.), Lcumann,
Gnomon xm (1937), 31, MusHIV (1947), 133 (=Kl.Sehr. 154).
301
 COMMENTARY

LXXX
This fragment seemsto describean encounter between two warrion; while
one is temporarily blinded, the other takes to his heels. The vision of
Homer's warrion is frequently obscuredon the field of battle by mists, etc.
(cf. n. 16.790 ff., 17.644 ff., 20.321 ff., 20.441 ff.) but there is no exact
parallel in the n;aJfor Ennius' Latin. I should guessthat the final encounter
between Achillesand Hector is beingdescribed; Homer's account is some-
what different (cf.n.22. 136-7.EKTOpa 6' oos w61,cm,lAeTp6µos· ov6'
&p•h' hAfl I av61 µhie1v, 6,daCi>Se 'ITVA~ Aim, ~ii Se q>o~118efs)
but so too is his account of several other events described or enacted in
Ennius' tragedy.
The abrupt asyndetonand change of subject do not compel one to postu-
late a lacunabetween abstulitand derepente. If there is something missinghere
it becomesdoubly hard to undentand why Nonius quoted ecce... abstulitin
an article illustratingthe adverb derepente.
167ecce autem caligo oborta est; omnem prospectum abstulit: cf.
Anon. Bell. Aft.52 nisi. .. puluis. .. uentoflatus omnium prospectuo.Jfecisset,
Virgil.Aen. 8. 253-4 inuoluitquedomumcaligine caecaIprospectum eripiensoculis,
Livy 4.33.8, 10. 32.6.
F.cceautemindicateseither the speaker'ssurpriseat an unexpected tum of
the action (c£ Plautus, Mere.748,792), or as here, hisfeelingthat the hearer
will be surprisedat what he is to narrate (cf. Virgil. Aen. 2. 201-5 Laocoon. ..
taurumingentem adaras.Iecceautemgemini. .. anguesIincumbunt
mactabat pelago).
appearselsewherein republicandrama only at Plautus,Mil. 009.
'Prospectus

168 contulit sese in pedes: c£ Plautus, Bacch.374 me continuocontuli

protinamin pedes(paratragic);pedibusse dareseemsto have been the regular
phrase (Naevius, Com. 35, Plautus, fr. 15); Plautus, Capt. 121 si nonestq,wd
dem,meneuis demipse- in pedesis humorously meant. L. Mueller was thus
quite wrong to excisecontulitsesein pedesfrom the Ennian fragment as being
untragic.
LXXXI
Sublimeiter(cf. vv. 190-1) indicatesthat the fragment refers to the horsesof
the sun-god (c£ Pindar, Ol. 7. 71-2 6 yM6AIOS&K-rlvCi>V ircmip, I nOp
,n,e6VTCi>V apxcslmTCA>V, Euripides, I.A. 159 nOp TI: Tl:6p{,mCA>v TOOV
•AeAfov,Virgil, Georg.1. 250 nosqutubiprimusequisOriensa.fllauit anhelis,Aen.
12. 114-15 cumprimumaltosegurgitetolluntI solisequi,lucemque elatisnaribus
ejflant)rather than to those of any of the heroes (c£ Euripides,Alie.493 nOp
TIVEOva1 IJVK'?TlpCA>V&iro, Lucretius 5 •29 et Diomedisequispirantesnaribus
ignem,Virgil. Aen. 7.280-1 geminosqueiugalisI semineah aetheriospirantis
302
 HECVBA
naribusignem).Omitting ut and comparing Ennius, Var.11-12 sol equisiter
repressit Iconstitereamnesperennes,arboresuento uacantVahlen
ungulisuolantibus,
argued that the fragment belonged in the same context as LXXIII. This is an
acute suggestion but I should prefer to leave ut and take the fragment as part
of a simile.

1'9 quadrupedantes:an odd formation, restored at Accius, Trag.6o3,

sccurcly transmitted at Plautus, Capt.814 quiaduehuntu,quad,upedanti crucianti
cantherio (a tragically styled parody of a praetor's edict), Virgil, Aen. 8. 595-6
agminefacto I quadrupedante putremsonituquatitungulacampum,11 .614-15
per.fractaquequadrupedantum Ipectorapectoribusrumpunt; perhaps concocted by
Ennius himselffor the sake of sound-play with halitantes. Tragedy hasquadru-
pes commonly of animals(Naevius 25, Ennius 157, Pacuvius2, Accius 315,
381), comedy only once (Terence, .Atulr.865 [perhaps paratragic]). The
formation quad,upedans is of the same type as undans(Ennius, Trag.179),
uiridans(Accius, T,ag. 244), unanimans(Plautus, True.435 [paratragic]), a
type of formation rare in old Latin and confined to poetry in classicalLatin.

halitantes: only here in Latin; halareitselfis rare and confined to poetry.

LXXXII
Scholl took this fragment as spokenby Achilles to a worried Patroclus after
Ulysses has come with bad news from the battle-field. It could as easily be
spoken by Priam to a companion while on the way to Achilles' tent or by
Achilles to Priam afterthe ransom has been arranged.The Attic tragedians
made Ulysses the opponent of all just and merciful arrangements. 1

170 nomas: probably the regular form in the common language; its
absence from the rest of extant dramais accidental; nouimusdoes not appear
either; the situations of drama leave little scope for the first person plural;
nosti, nossemetc., nosseare common in Plautus and ahnost universal in
Terence.

HECVBA

Aulus Gcllius asserts at NoctesAtticae11 . 4 that the Hecubaof Ennius was a

version of Euripides' homonymous tragedy and quotes three verses of the
Gt-eckscript along with the particular Latin rendering. At 2.23 he makes a
1
Vahlen {E.P.R.1 , p. 145) appears to have taken the fragment as spoken by
Mcnelaus to Ajax after the wounded Ulysses has shouted for help {Homer, 11.
I 1 • 463 ff.); such an interpretation would require a quite incredible change of
scene.
 COMMENTARY

similarbut much more extended comparison between Mcnand.cr' s mOKtov

and Caccilius' vcnion. The extant scholia to Terence's comedies contain
scraps of similar comparisons between Terence's Latin and the Greek of his
originals. We may suppose that Gcllius based his assertion about Ennius'
Hecubaeither on a detailed pcnonalknowledge of the texts of this play and
the Euripidean or on a didascalic notice of the type affixed to the extant
texts of the six comedies of Terence and the Stichusof Plautus. It has a relia-
bility quite lacking in apparently similar assertions by Varro and Cicero
about particularLatin tragedies (secabove, p. 236).
Nonius Marcellus gives Ennius the title Hecubaeleven times. The eight
pieces quoted seem to come from grammatical sources rather than Nonius'
own knowledge of the text. The wording of only two of them (frs.xc and
xcu) parallels Euripides' Greek 1 as closely as does that of Gcllius' quotation
(LXXXIV) but the others can be fitted into free versions of Euripidean speeches
without upsetting at any point the framework of the Greek plot. This
circumstance indicates clearly what others only suggest, namely that the
republican poets rarely followed the exact wording of the Athenian classics
they were hired to adapt. It is accordingly improper to use, as many have
done, the text of Euripides' "E1ea~11 as an aid in restoring that of the Ennian
fragments.
A trimeter, ueterfatorum terminussic iussaat, is attributed by Priscian
(Gramm.II 264. 1 s) to a Hecubaof Accius. Iff atabe interpreted as •oracles' the
trimcter hasno counterpart in Euripides' "E1ealnl.If it be interpreted as •fate'
or •destiny' there arc a number of possible counterparts: c£, for example,
vv. 43 ft 1Tfflp<A>µM), 584 8eoov6vayt<cnow.The imagery of the boundary
stone is absent from Euripides' "EK&fnl but can be found elsewhere in Attic
tragedy (c£ Aeschylus, Ag. 1154-s ir68ev 6po\Js fxe1s8£<,ma(cxs 66oO I
KaJCoppf}µovcxs, Choe. 927 irarpos yap ataa Tov6e aovp(3e1 µ6pov,
Sophocles, 0. T. 723 To1a0Tacpi;µcnµoorr11eal 61oop1aav).Land division
was a frequent source of metaphor in republican dramatic writing (c£ Plautus,
Poen.48-9) and it is possible that the author of ueterf atorumterminussic iussaat
drew upon it to express the rather un-Roman notion of a fixed immutable
destiny. 2 Hermann3 suggested that Priscianwas in error in attributing the
trimeter to Accius' Hecuba rather than to Ennius' Hecuba. Such errors are

1
P. Victorius, VariarumLectionumlibri XXV (Florence, 1553), x 1, com-
pared Euripides, Hek. 8.26ff. and 497 ff.
:a C£ Lucretius, 1. 75-7 unde reftrt nobisuictorquidpossitoriri, I quid nequeat,
.finita potestas denique cuique I quanam sit ratione atque alte terminus haerens
(Epicurean natural law), Virgil, Aen. 4.612-14 si ... neusseest I et sicfata Iouis
poscunt,hieterminushaeret,Horace, Cann. saec.25-7 uosqueueracescecinisse Parcu
I quodsemeldictumest stabilisquererumI terminusseruet.
3 On Eur. Hek. 578 (Leipzig, 1800).

304
 HECVBA
common enough in the grammarians1 and the singularity of the reference to
Accius' Hecubamakes the existence of the play dubious. On the other hand
the titles Alctstis,Eriphylaand Persidae are given by Priscian to Accius eachon
a single occasion and occur nowhere else in our sources; the existence of these
it would be folly to doubt. Furthermore the usage of republican drama
favours the interpretation 'oracle• for terminus f alorum rather than 'fate•.2
At Ling. 7.6 Varro illustrates the use of the word templumfrom the plays
Hecuba,Periboeaand .Andromacha; copious quotations elsewhere give an
Andromacha to Ennius and a Periboeato Pacuvius; but for the doubts about
Priscian, Gramm.II 264.15 one would be tempted to assign Varro's first
quotation to Accius, the third of the dassicaltragic trio. With some doubts
I have followed editorial tradition and printed it alongside the pieces directly
attributed to Ennius' Hecuba(fr. LXXXIII).
There is no unambiguous evidence that Cicero knew Ennius' Hecuba.On
the other hand his works contain many quotations of Ennius' .Andromacha
and Pacuvius' Iliona, tragedies similar in many ways structurally to the
Hecuba.3Accordingly quotations such as those at Tusc.I. 3'7'4 and Orat. 153S
1
which recall structural elements of Euripides' EKcxfnl (vv. 1-2, 837, 1040-4)
need not beattributed to Ennius' adaptation. 6
Where sententiae with parallels in Euripides' •EK&J,11 are concerned even
greater scepticism is justified. 7 The remains of Euripides' tragic production
are full of general statements about society, religion, politics, morality, etc. 8

1
Using a common source Priscian, Gramm.II 26o.4 and Nonius, p. 223 .2
attribute a fragment the one rightly to the Niptraof Pacuvius the other wrongly
to a Niptra of Accius; cf. the clear errors at Nonius, pp. 75. 8, 90. 10, n6. 8,
n7.18, 170.12, 176.12, 209.25, 223.2, 382.12, 479.13, 515.12.
:a Cf. Marx, SB Leipzig LXIII {19n), 62.
3 On niotui see Zielinski, Tragodumenon Libri Tres (Krak6w, 1925), pp. 48,
124.
4 See Columna, Q. Ennii Frag.p. 376.
s See Ribbeck, Die rom. Trag. p. 145, Havet, RPh XXVIII (1904), 219 ff.
6
Scaliger, Coniect.Va". Ling., on 10.70, attributed to Hecubathe canticum
quoted at Tusc. 3 .44 on the grounds that it resembled Euripides, Hele.158-00,
overlooking that later, at 3. 53, part of the canticumis quoted as coming from
Andromacha.
7 Two trimeters quoted by Cicero at Tusc. 2. 13 were assigned by Columna
to Ennius' Hecubabecause of their similarity to Euripides, Hek. 592-5. In RhM
XVI {1861), 443 ff., Mommsen published some Virgilian scholia (Breu. exp.
Georg.1. 1, p. 199. 3 Hagen) quoting the verses as from Accius' Atreus and thus
confirming Muretus' restoration of Cicero's corrupt introductory words as
illud Accii 'probae(illudacimprobeRK: acinprobeGV).
8
See C. W. Friedrich, Die dramatischeFunletionder euripideischmGnomen
(Diss. Frciburg, 1955).
20 305 JTO
 COMMENTARY

often repeated from one play to another. Vahlen printed the words nimium
boni tst cui nihil est mali attributed to Ennius at Fin. 2 .41 (fr. c.xxm) as a
fragmentof Htcubaon the grounds of its similarity to Euripides, Hele.627-8
l<l!lVOS6~lCJrCXTOS,lrrCt>I Kar' ~µap TVY)(avEl µ116tvKCXK6v. 1
The idea
I
recurs at Ba. 910-n TO& Kar· fjµap lrrct>f3{0TOS ev6atµ~ µCXKap{3CA>. 2

V ahlen also printed among the fragments of Htcubathe amicuscertu.s in rt

incertacernitu,attributed to Ennius at Lael.64 (fr. CLXXXV). The substance of
this trimeter appean not only at Htle. 12»-7, wTOlS KCXKOIS yap aya8ol
I
aacparrcrr01 q,0.01,3but also at Or. 454-s, 6voµa yap, lpyov s• oVK
I
fxovcnv ot q,0.01 ot µ,; •irl Tatcn avµq,opats 6VTES 4 Ennius seems
cpiA01.
to have employed smtentiaemuch more frequently than did his succcsson
Pacuvius and Accius. He quite certainly alteredthe emphasis of some of
those which he found in his originals (seebelow on LXXXIV) and added othen
totally absent (seebelow on cvm); it is possible that he omitted some which
did not appeal to him.SIn circumstances like this it seemspointless to try to
localise Sffltfflti« in particularplays; too many possibilities are open.

LXXXIII
The personages of tragedy frequently address the open sky above the stage;
c£ Aeschylus, Prom. 88 w6tcs ale,;p, 1091-2 w
1TavTCA>V I
al&,;pKo1vov
q,<XoS
e~{aaCA>v, Sophocles, El. 86, Euripides, Ion 1445, fr. 443. The heroine
of Euripides' 'EK&f3ri refers sarcastically to the practice at vv. 334-s w
I
8vyarep, ovµol µh, Aoyo1,rposal6tpa cppov601µ&Triv~l~ &µcpl
aoO q,6vov.6 The only near parallel in the play with Ennius' iambic tetra-
meter is the anapaestic address to the lightning bolt of Zeus at vv. 68-70, w
tu6s,er,aKO'Tiavv~,
0"1'1:poira ITl irOT·aipoµoo lwvxcs ovrCA> I&lµacn
q,&aµacnv;But Ennius can be shown quite certainly in other placesto have
alteredthe metrical structure and the substance of his original no lessdrasti-
cally (seebelow on fr. LXXXIV).

1
Murctus (VariarumLedionumlibri VIII [Venice, 1559], v 13) seems to have
been the first to make this identification.
2
C£ also Euripides, Hnalelts503 ff., Hile.953 f., Hipp. I I I I ff., fr. 196, fr. 714.
s Hartung (EuripidesR.estitutusI, p. 529) was the first to make this identifica-
tion.
4 Cf. Or. 665-8.
s Among the fragments of Menander's 'Av6p(a (43) and 'Eavrov
T1'1c..>pOVlllVOS(132) arc smtmtuu which have no equivalents in Terence's
adaptations. The substance of Andria959-00 is said by Donatus to come from
.Mcoander's EvvoOxos but it has no equivalent in the Latin version of this
play.
6
Cf. Euripides, Amir. 91 ff., I. T . .4,2f., Mtd. 57 f. Comedy parodies it; cf.
Menander, Sam. III, Plautus, Trin. 1070, Terence, Ad. 790.
306
 HBCVBA
171 templa caelitam: cf. (a) Accius, Trag.proet.2, Varro, Men. 56, (b)
Ennius, Ann. 5-4-1templummagnumlouis altitonantis,Lucretius 5. 1188 in
aitloquedtumstdestt kmpl4 loa,,unt,(c)Enniw, Ann. -4-9coeliaitrUl4templ4,
Terence, Eun. 590 ttmpl4 aitli summa ('sententia tragica'-Donatw),
Lucretius I . 1014, 1105 et al., (d)Lucretius 5. 1204-5 magniaitlemamundiI
templasuptr skllisqutmicantibusattherafixum, (f) Seneca, Htrc.f. 3 templ4
summi... atthms. The only comparablephrase in extant Attic tragedy seems
to be Aeschylus.Ptrs. 365 -ri~ al~.
The word templumprobably came from the augural language(seeabove on
v. 88). 1 It is difficultto say whether aitles,which occun four times elsewhere
in republican tragedy and only once (Plautw, Rud.2) in comedy, was a
~
saaal word or a poetic creation ( ovpav1c,s,ovpavfCA>V ). It is absent from
classicalprose. Yet its second dement is not common in poetic neologisms.

spleadidis:elsewherein tragedy at Accius678; absentfrom comedy. This

could be accident, for splendertoccun at Iuventius, Com. 6, Plautw, Poen.
314, Rud.3 (splendens sttll4),splenJorat Plautw, A.sin.-4,26,Aul. 602, Mtrc.
880, Mil. 1. However the incidence of adjectivesin -idu.sseemsto be some-
what greater in tragedy than in comedy. Plocn counts -4-0formations in
comedy and 27 in tragedy, 10 of which do not occur in comedy.

LXXXIV
Ennius' opening haec,which has no counterpart in the Greek of vv. 293-._
TOs· a~fooµa,KCX\1 KCXKOOSAfy,J,TO aov
ImfaEl, must refer to the content
of his version of vv. 288-92, ~ a'TfOKTEfveav '86vosI ywatKQS, &sTO
np&'m>voliK krefvare I ~µoov a,roo-rraacxvns, &AA" Cf)KTfpan.I
voµc,s s· Iv vµtv Tots T' ~pots lacs I Kai TOlat 6ovAOISatµcrros
KEITatmpt. One may doubt whether he adapted the substanceof vv. 291-2;
this was an anachronisticreference to the protection which Athenian law
gave to slavesagainstarbitrary acts on the part of their mastcn 3 and would
have created sympathy for the speaker in the Athenian theatre; but in
second-century Rome, where masten enjoyed unrestricted power over the
lives of their slaves, it would have sounded mcrdy paradoxical. At Stich.
~. atqut id nt nosmirtmini,hominesmuolos Ipotart, amart atqut adaruun
condietrt:I licet haecAlhenis nobis,Plautw feels obliged to explain the
behaviour of his Menandrian slaves.

1
Ennius could be employing an ardutectural metaphor (see above on v. 3);
for ttmplum,'roof-beam' cf. Lucretius 2. 28, Paulus. Fest. p. SOS•1 f., Vitruvius
-4,.2.
3
Cf. Isocrates 18. s2, Demosthenes 21. -4-6-9,Ps. Xcnophon, .Atls.1. 10.

20-2
 COMMENTARY

The substance of vv. 294-SA6yoSyap Cl<

1
T a~OWTOOV loov 1(00(TC>v I
OOICOWTOOV avrosov Tcnrrova8we1was a commonplace both in the fifth
µ,;
century and later; c£Euripides, Atulr.186-7 fyooSi Tapj:x:> TO SOVAMlV
I
µi aoi A6yoovCfflCOOTJ -rr6AA1
~ovaav lvS11Ca,&. 327, Trag. Grace.inc.
&. 119, Cicero, S. Rose.2 quiasi qui istorumdixissetquosuuktis adesse,in quibus
summaauctoritasest atqueamplitudo,si uabum tk re publicofedsset,id quodin
lu,ccousafieri neasseest, multoplura dixisseq,u,mdixissetputaretur.egoautnnsi
sunt liberedixeroneq,u,qumn
omniaquaedicenda similiteroratiomeaexireatquein
poterit.for the purely general contrast betweenol OOICOWTES
uolgusemarurre
and ol a~OWTES Ennius substitutes a particular one between opultnti,
'those with opts, with social power and influence', and ignobiles,'those
without a nomen,of obscure and humble family•. This contrast is not apt to
the situation of Ulyssesand the enslaved Hecuba. Ennius must have had in
mind not the text of his original or even its plot but one of the themes of
contemporary political debate. He would have known that critia chid
Euripides for making his heroes and heroines speak like contemporary
Athenian philosophers and politicians. 1 In making his Hecuba utter some-
thing quite incongruous with the dramatic situation he alsohad Euripides as
a forerunner, if not in •EICaJhlcertainly in other plays. a His motive, like that
of Euripides, would have been a straightforward desire for theatrical effect,
something to which the Aristotelian canons of organic unity are not always
relevant.
Euripides' spoken trimeters are replaced with musically accompanied
trochaic tetrameters. These verses bring with them here, as elsewhere, an
amount of figurated language absent from Euripides' trimeters (whose
breuitasis noted by Gellius). Avros
ov Tcnrr6vobviously inspires aequanon
aeque,a type of polyptoton common in Attic tragedy but rare in Roman
drama.3 The phrase facile. . .flexerisis chosen for an alliterative effect quite
rare in Attic tragedy but common in the republican adaptations. 4 /\6yoS ...

• Ennius may have read a commentary on the •EK~T'I containing remarks

like those in the extant scholia; cf. ad 2 S4TCXVTCX els niv KCXT CXVTOV
1
1roA1TEfav
~ Ka8 lCNTOV
1
Afyr1.KailOTITOtoiiTos6 E-Op11rf6T1S, mptarTTC..W TOTS1'p(,)(71
Kal
TO\IS)(p6vovsovyxic..w, 898 ov KCXA&'.,s ffl0'1 TCXVTCX 6 •Ayaµiµ\KAW.fxpt1v yap
CXVTOV &rra~ 66\mx ff\Vxap1vo,"1fri\aat Kai µit ruy~ fflVbrroO yvcbµ11v·
o-0yap~ TOICNTTJOffilCElTO ~0'1M\IS ~.
:a See Lucian, Iou. trag.41.689, the scholiast on Phoin. 388, modem scholan
on Heraleles1340-6.
3 The only exact parallel I can find is Plautw, Amph. 3 SS familiaris,aaipiere
f axo haudfamiliariter.
4 Cf. Naevius 13 tMa manu moriare,28 lino li~t lumina, Ennius 118
spumanlsanguine,165 sudatsanguiM,181 modia moremgerit, 184 g,auitergmums
et al.

308
 HECVBA

av-r6sis turned into the highly elaborate uJem diall t~ oratio;the

doubling of one word or phrase with another of much thesame meaning,
the anaphora, the ha]ancing of one colon with another slightly longer are
cenainly to befound in some trimeter speechesof Attic tragedy and comedy
but they are most at home in the long verses of the Latin adaptations (see
above on vv. 9, 19). The units of sense of the Latin discourse are made to
coincide with those of metre in thenormal manner of republican drama.

173 mm cam opulenti locuntur: Porson's nam opuknti cum locuntur

removes an instance of hiatus after a monosyllablein thethesisof a trochaic
venc, 1 gives thegeneral opukntian emphasiscorrespondingwith that of the
particular tu placed in front of the conjunction ttsi in v. 172and produces an
order of words common in republican drama but rarer later and therefore
likdy to be 'moderoised '. 1 I am not sure however that these are grounds
firm enough for altering the paradosis.
Opts usually stands for socialpower and standing rather than property
(diuitiae)in the speechesof republican drama. Opukntus is sometimes clifti-
cult to distinguishfrom the much more commonly occurring diues(it is
opposed to pauperat Plautus, Aul. 247, 461, 479, Caccilius,Com. 170ff.)but
at Plautus, Cure.284-6, neeusquamquisquamst tam opukntus,qui mi opsistatin
uia I nee straJegusnee tyrannusquisquamnee agoranomusI nee dmutrehusnee
eomarehus, and Rud.713, destnatu Cyrenensiquemuisopukntumuirum,it bean
plainly the meaning 'socially powerful, influential'.3

ignobilN: the meaning herecannot beanything but 'sociallyobscure', as

at Pacuvius, Trag.221 and Terence, Phorm.120. At Livius, Com. 3, Plautus,
Amph. 440, Pseud.s92, 964 it is 'unknown to one individual or another'.
Nobilismeans 'known to one individual or another' at Plautus, Poen.1s8,
Pseud.1112, Rud.619, Trin. 828; 'socially MJ>ineot'at Plautus, Cist. 12s
(suspectedof latenessby some scholars),Terence, Atl. IS, so2, Eun. 204, 9s2,
1
The only other tragic instance of this hiatus seems to be Accius 3,4s. For a
list of the instances transmitted by the codices of PJautus in which cum and
quom stand unclidcd in thai see Maurcnbrechcr, Hiatus und Verschkifung,
pp. 35 ff.
a Cf. Plautus, Amph. 91-2 histrionesanno quom in prosa,eniohie I louem
inuocarunt,uenit, ,427-8 ltgiones quom pugnabantmaxume, I quid in tdbmuiclo
fttisti ?, Cas.7-8 et al. However for the' modem' type of order cf. PJautus, Capt.
77-9 quasimuru stmptr tdimus alimum cibum; I ubi resproUltatsunt, quom rus
hominestunt, I simul prolataerts sunt nostrisdtntibus, Cas. ,417-18, Cure. 105-6
et al.
3 Among the classicalprose writers only the historians Sallust and Livy have
opulmtusat all commonly. On Livy's usagesee Drexler, RhM CD (1959), 55-8.
 COMMENTARY

Haut. 227,009. Plautus usesnobiliw three times (Capt.299, Mil. 1324, Rud.
933) of social eminence.The two setsof ineanin~ must--havebeenalready
coexistentbefore 184 B.C. By the fint century nobiliswas a fully teclmical
term of thepoliticalvocabulary,denoting a member of one of those families
whose names were known as a result of ancestralconsulshipsand which
enjoyed an almostcompletemonopoly of the principalmagistracies.Ennius'
fint patron, Cato, had beenforced to makehis politicalway in a climateof
opinion well desaibedby vv. 173-4-

174 eadem dicta eademque oratio: the two phrasescan be distinguished;

diaaare the words and oratiothe way they are uttered; c£ v. 258 quamtibiex
oreorationem duriterdictisdedit.
LXXXV
Tearsdrip in Attic tragedyrather lessfrequentlythanblood and sweat; but c£
Aeschylus,Choe.185-6, Prom.398-401, Euripides,Heraleles 1354--0,Ion 876,
as well as Euripides,Hek. 100 6p~ VEl<pOV-rovs· ooK<l'TQCTT~OO Sco<pv;
Mercerusthought that fr. LXXXV was a versionof thislatter passage.in which
Hecuba draws Agamemnon's attention to the corpse of Polydorus, and
restored it as uidehuncmeaein quemlacrumae guttatimcadunt.Vossiusrestored
it as uidehancmeaein quametc. suggesting that the words are spoken by
Hecuba to Ulysses as he grasps her daughter Polyxena. This suggestion
involveslessalteration to the paradosisand is none the worse for implying
that Enniusalteredslightlythe form of the EuripideanHecuba's appealat vv.
277-81. Scaliger had let the words transmitted stand and set them in a
version of Talthybius' account of the death of Polyxena (c£ vv. 519-20 vw
TEyapAE)'OOV KaKCXI -riy~oo,66" 6µµa).

175 lacrumaeguttatim cadant: contrast the form of the expressionat

Propertius 4. 1 . 144 guttaquoqueex oculis. .. cadet.The adverbguttatimoccurs
elsewherein republicandrama only at Plautus,Mere.205 and not again until
Apuleius.
LXXXVI
In moments of good fortune the personagesof tragedy and comedy fre-
quently, if not regularly,addressthanksto the gods; c£ Sophocles,Ant. 330-
1, Tr. 200-1, Euripides, Herakleidai869, Menander, Sam. 26<>-70,Plautus,
Capt. ']68 ff., 922 ff., Mere.842 ff., Most. 431 ff., Persa251 ff., 753 ff., Poen.
1274 ff., Rud.9o6ff., Stich. 402 ff., Trin. 820 ff. Ennius' personage gives
ironical thanks for ill fortune. The malignityof the gods is a constant theme
of Euripides' •EKaj311-231-3 Kcfyooy• 6:p' OVKfflvt:1C7KOV oij µ• tx?fiv
eavetv,I ovs·ci.)"A£a#.v µE Zais,-rpiq,e1s·, 6,r005 6pooI KaKOOV 0
KaK &XAa
µef30V ti 'T<XAaiv·
1
tycb, 57-8 <WrlO'TlKcbaas & ae I cp8efpe,8eoov-ns
310
 HBCVBA
Tfis ,rapo1e· MTpa~{as, 199-201 otav otav QV ao1 Aoof:xxvI t)(&{crrav
I oos
apf)T)'Tav'T• &paw 'TlSScrlµoov,721-2cI>'TAf\µov en 1l'OAV11'0\MTCX'l'TlV
~&;v I Sooµoov 16r}t<EV OOTISlcrr{ <701j:kxpvs, 958-60 q,vpovm s·
cnrrol 8eol ,ra).1v T& Kal ,rpoooo I 'Tapayµ0V hrn8wns,00S I
ayvCA>C7ic;r
crijxA>µw CXVTOVS-butthere is nothing paralldwith the Eooiao irony in
this play or clscwhere. 1 I suggest that the Latin words come from an outbunt
by Hecuba on hearing of the discovery of Polydorus' body.~
The words Iuppiterand gratulor(in the sense 'give tbaoks ') belonged
properly to the sacral language; the second was always used in auspicious
circumstances. Of the Attic tragediansAeschylus was particularly fond of
employing sacralwords and phrases in paradoxical contexts (c£ Ag. 645
11'pffl'ElMy£1v ,rooava 'TOVS' 'Ep1woov, 1144-5 <Trivova· aµq,16aAi\
K<XKotsI a116cl:>v~{ov, 1385-7 Kal ,mn-ooK6-n I 'TP''TTI"hrev6{6CA>µ1,
'T0V K<XT<X x6ov6s, I ·A16ov,VEl<f)C>v ac,.n;;pos,EVK'TOOQV xap1v)•
.Enoius indulges in similar paradox at Trag.287: qui iliumdi deaequemagno
nuiaassintmalo.3
Fracokelhas argued 4 that Ennius' tetrameter parodies not just the Roman
sacral language in general but the particular prayer offered to Iuppiter
OptimusMaximusby the triumphing imperator.

176 Iuppiter tibi summe: the disjunction of the epithet by tibimoved to

second place in the sentences was unusual in Ennius' day and could be imita-
tive of the style of sacral formulae.
Summuswas a common epithet of Iuppiterin comedy (c£ Plautus, Amph.
933 et al.), along with supremus and tmJgnus.It seems to be a poetic translation
of V\fllCTToS rather than a borrowing from the sacral language.
or V'IT<XToS
Cicero's public orations always refer to IuppiterOptimusMaximus.

1
Zeus is abused directly to his face at Herakles339-47 ... apnij en VtKw
8vt}TosC::,v8Eovl,Afycxv • I ... 6'.1,1a8i\s
TISEl 8E6s,ft 6hca1osoVKftvs.
a Cf. Dilntzcr, Zeitschr.J.d. Alt. 1838, pp. s2 f., who allegesan oxymoron
and compares Eur. Hek. 232 and 784. Columna's theory, accepted by Ribbeck
and Vahlcn, that the words form Hecuba's cry of triumph after the revenge
taken upon Polymcstor would require a beneregestain the text; cf. the context
of Plautus, Rud. 1178-9 quomistaecrestmJleeuenittibi I Gripeg,atulor.A. Della
Casa, Dwniso XXXVI (1962), 69 f., offers a good criticism of Columna's but no
convincing interpretation of her own.
3 Cf. also the way in which Lucretius expresses the Epicurean detestation of
religion in language redolent of the traditional Roman cult at 1. 82-3: quod
contrasaepiusilia I religiopeperitscelerosa
atqueimpiafacta;Propcrtius' use of the
sacral piareat 1. 1.20 and 3. 19.18.
4 Pl. im Pl. pp. 238 ff. ( = Elementi,pp. 229 ff.), on Aeschylus, Ag. 1387.

s Cf. the disjunction of per and its object discussedabove on v. 3.

311
 COMMENTARY
tandem male re ge,ta: tandemis to be taken closelywith mme; 'really,
positivdy'; cf. Plautus, Cure.7 tandemts odiosusmihi.
The ablative absolute is not common in either republican comedy or
tragedy and here the normal phrase would bea clauseintroduced by quom,
quodor quia.There is someevidencethat theabsoluteconstructionoriginated
in the officiallanguage;the prayer of thanksgivingaddressedby successful
generalstoJupiter may have containedthe phraserebene gtSta.Perhapsat one
time remgererewas proper to the officiallanguagebut it is usedso often in
comedy of the doingsof priuati(even theabsolutephraseoccun eight times:
Plautus,Amph.655, Persa7S._ Stich.'4,02, '4,II, SO'/,Trin.592, 1182, Terence,
Ad. 77S) that we mwt supposeit to have been unremarkablein common
parlancein the early secondcentury.

gratalor: rare in the meaning 'gratias ago' (elsewherein drama only at

Caeciliw,Com. 9, Terence,Haut. 879, Afranius, Com. tog.21) and then only
in addressesto gods.
LXXXVII
The generalsenseof thisfragment is clear despitethe corruption it hassuf-
fered; a person is acldrcssedin whose state the written law provides no
penaltyfor the murderer of a parent or guest. This was notoriouslytrue of
historicalAthens, at least as far as parenticidewas concerned;1 the Roman
Twdve Tablesimposedvariow kinds of death penalty for variouskindsof
crime but seemto have left the punishmentof homicidein generalto private
arrangement.a How the prosecution of L. Hostius, the fint for patricide
known to theRoman aooalisn,was mounted we do not know. At any rate
the case must have been still fresh in the minds of Ennius' audience (see
above on fr. XIV}.The language and style of the fragment recall strongly
the remains of republicanleges.
There appear to be verbal linksbetweenthe fragment and remarks made
by the Euripidean Hccuba as she appealsto Agamemnon to punish Poly-
~
. .. 866v6µ(A)vypacpa{;quisparentmtout hospitemneawet
mestor (scripstis
~ 803-4omves ~ IKTEfvovcnvfi 8Eoovlepa-roAµoocnvcpipeiv).Jacobus
Nicolaw of Loo3 accordinglymade it bdong to a version of vv. 798-8o5.
Anachronistictalkof written law is common in Attic tragedy (c£ Aeschylus,
1
On Solon'slaw sec Cicero, S. Rose. 70, Diogenes Lacrtius I. 59.
a SecLatte, TAPhA LXVU (1936), 24ff., RESuppL vu (1940), s.v. Todesstrafe,
1614, ZSavSt LXVU(1950), SI ff., W. Kunkel, Untersuchungmzur Entwicklung
des romischmKriminalverfahrensin vorsullanischer
Zeit (Munich, 1962), pp. 38 ff.
3 MiscellaneorumEpiphillidumLibri X, VIII 7, apparently fint published inJ.
Grutcrus. Lampas,Tom. v, Suppl. (Frankfurt, 1606), p. 588.On Jacobus sec W.
Clausen,CPh u:x (1964).96.
312
 HBCVBA

Hilt. 707-9, Sophocles.Ant. 454-s. Euripides. Hilt. 433-4, Ion -U2-3) and
Greek. thought commonly associatedoffences against guests with those
against parents {c£ Hesiod, Erg. 327-34, Aeschylus,Bum. ~1. Aristo-
phanes.Batr. 146-50). But it is difficultto associatethe point of the fragment
with that of 798-805 or indeedwith anything in Hecuba's utterancesat this
stage. Vossiusreferred it to Hecuba's reply to Polymestorin the debate before
Agamemnon. The absencehere of any verbal counterpart to Ennius' Latin
does not count against Vossius•theory but again it is difficult to see what
point the Latin could have had. Osann 1 referred the fragment to Agamem-
non's speechrefusing to punish Hecuba and in particular to vv. 124,-S, but
looked at as a whole it is no possibleversion of TCX)(• ow ,rap• vµtv ~61ov
~EVOKTOVEIV • IftµtvSt y• ataxpov Toto,v·ru,,01v T6&. Scholan may of
course be mistaken in scck.ing~ point related to the plot and arguments of
the Euripideanplay. Ennius may be-- making l\is actor speakof contemporary
Rome or contemporary Athens. I must admit. however. that even so the
point eludes me. ·

I
177-8 quis parentem aut hospitem necauet: for relativequisc£Lex XII
tab. I 4 prolttarioiamciuiquisuoltt uindexesto,n 2, Lex luci Lucer. C.I.L. 11
401. S, Lex Sil. ap. Fest. p. 288. 34, Cato, Ag,. 145. 1. 148.2, Plautus, Mere.
991 supplicisibi sumatquiduolt ipseob baneiniuruun;it was clearly a legal
archaism already in theearly second century.
Necare,which occurs twice elsewherein tragedy and nine times in comedy
{not in Terence), probably alsohad an archaic tone, though much lesspro-
nounced than that of quis;comedy, reflecting the common language. pre-
ferred thecompound enicare{28times in Plautus, sin Terence; I include the
metaphorical usage).
For the meaning 'murder' c£ Festus,p. 174.26, Rhct. inc. Her. 1.23.
The relativeclauseprecedesthe principal quite rarely in dramabut doesso
regularly in republican legal texts.i

178 tquos quilt cruciatu perbiteret: if Junius' quo quis cruciatube

accepted we have either a very loose arrangement of relative and principal
clauses,for which there is no parallelin particular but quite a few in general
{c£ Plautus, Aul. 790-1 quihomoculpamadmisitin se, nullusttam paruipreti I
quinpudeatquinpurgetsese,Capt.941 quodbenefecistirefereturgratiaet al.), or
the use of quisas an indefinite after an interrogative/relative pronoun, for
which the only exact parallelI can find is Ennius. Sat. 61-2 (namquisest (sise
Usener)frustrariquemJrustrasentit,I quiJrustraturisJrustraest si non ilk est
frustra).
1
Anal. crit. 127 ff. i See W. Kroll, Glott4 m (1912). 8 ff.
 COMMENTARY

For crucialuperbiterttc£ Plautus, Pseud.776-8 interminatust... si quisnon

sibi,I tum erascruciatumaxumoptrbitert,R.JUJ.
hoditmunusmi.sis.set 494-5, Cas.
300, Epid. 513, Cicero, Nat. deor.3. 81. The verb baetere/bitert
seems to be
evidencedfrom the Twdve Tables (Varro, Mtn. 553); it and its compounds
occur sporadicallyin Plautine comedy as high-falutin variants of irt, etc.
Tragedy appean to have affectedthe verb rather more frequently (ptrbitere,
for example, occurs 3 times againstptrire5; comedy has perbitereonly twice
but ptrireover 200 times) and for longer. The whole phrasecruciatuperbitere
may come from the language of thelaw; the Romans regularlyaccompanied
legal execution with tonure (secabove on fr. XIV).

LXXXVIII
This fragment might come from any of a number of contexts in Ennius'
~ olsµ• &Aos
play; c£ Euripides, Hek. 26-7 KTCXVOOV I 1,&E8iix·.
28-9, 446,
634, 701, 781-2, 797, 938, 1259.

179 undaat4m1 •Jam; c£ Accius, Trag.401 undankinf,eto.

The adjectiveundans,like other similarformations, seemsto be restricted
to poetry in classicalLatin. It occurs in republican comedy only at Plautus,
Epid. 436.
Salus masculineseems to occur only here in Latin. Salum neuter occurs
sporadicallyin classicalprose and verse.It must have beena borrowing from
Greek first made in the nautical language; maredetermined the gender per-
hapsfrom thebeginning, Enniusmay have beenunsure of the gender or have
varied it for poetic reasonswith that of a@.es in mind.

LXXXIX
This fragment must have beenuttered by Hecuba as Ulysses,unpersuadcdby
her arguments, left the stage with Polyxena. If theless well attested reading
heume miserumis correct the speak.erwas Polymestor as the Trojan women
proceeded to slay his children.At neither point in the action has Euripides'
Greek anything parallel
Warrion normally washed in water after battle (cf. Homer, n.23. 35 ft,
Od. 22. 478, Sophocles,Ai. 654-6, Virgil,Aen. 2. 718-20 ). The EnnianHecuba
predicts a ghastly revenal of the normal practice. The murderers of Attic
tragedy were ritually purified by animal blood (c£ Aeschylus.Bum. 283,
452, fr. 327, Euripides,I. T. 1223, Sthen. 17 ft) but thispracticeseemsto have
been unknown at Rome and one cannot assumeany allusionto it in Ennius'
Latin.

18o lauere .anguen saaguine: for the idea of bathing in blood c£

Euripides,Hele.1281 cp6v1a
~oVTpa, Furiusap.Gdl. 18. 11 •4 unguinediluitut

314
 HBCVBA
tellus, Virgil, Georg.3. 221 Lmit ater corporasanguis,Aen. 10.727-8 lauit
improbat«ter I ora auor, 12.721-2 sanguinelargoI callaarmosquelauant,
Propcrtius4. 10.37-8 destctaTolumniI ceruixRomanossanguinelauittquos.
For the polyptoton cf. Sophocles,0. T. 100-1 cp6vct) fOVOV nOAtv I
AVOVTQS, Euripides,El. 857-8 alµa s· alµ<XTOSI ffllepOS6av&taµosM8e
TCf>8avovTi vOv,Htrakles40 WS ~01J fOV0\1,
cp6vft> 1.T. 1223-4WS'6Vct>
fOVOV I µvaapov lKv('fXA), Accius, Trag. 82-3 cum patre paruospatrum
hostificeI sanguinesangutnmiscert{Pas.1Cfatius:
hiscertcodd.) suo, Trag.pratt.
4 lue patrium {Buccheler: ue patrum codd.) hostilifaso sangutnsanguine,
Lucretius3 . 71 caedemcaedeaccumulantes.

XC
This fragment clearlycomes from a version of Euripides,Htle. 824-35Kai
µ1'tv- laCA>S "'1vTOOAoyov 1CEVOV TO&, I Kwpav npo~v· &XA•
6µCA>S elp1'ta£Tat• I npos aolat nMVpols nats lµ'i'i1<01µ(3ffatI Ii f<>tpas,
ftv KCV\OVat Kaaav6pav ci>pvyes. I ,ro0 ~ cpfAass~· evq,p6vas
&(~etS, &va~, I i\ TWVW EUVijcplATCXTCA>V aanaaµaTCA>V I xap1v T(v·
1~, nats lµfi, 1<1:fVT)s s· tyc:,;... TOV6av6vTcxT6vs·~; I TOVTov
KaAC)sspa,v 6vra KTl&oTi\v ai8Ev I 6pacms.
Absentfrom the Greek is the emphasisupon obedienceto the will of the
male during sexualintercourse{morern gtrit)and upon restraintof the female's
desires (uerecundett modice).Even where marriage and concubinage in
general are concerned Attic tragedy does not emphasiseso strongly the
virtue of femaleobedience;cf. the EuripidcanAndromache'saccountof her
relationshipwith Hector at Tr. 645-56, in particular655-6il6115•~ XPfiv
VtKovn&nv, IKEfVct> TE v(KTIV~v fxpfiv mxpm,cn,and Hecuba'sadviceto
Andromacheon how to treat her new master at Tr. 699-7<><>: Tiµa & TOV
napwra 6ean6-nlv aieev, I cp(Aov 616ovaa 60.Eapav6pl oi:>vTpo,tCA>V.
In the society depicted by comedy and its Roman adaptationsobedience
was the chief characteristicwhich distinguishedthe yvv1'1yap£Tfl/matront1
from the hafpa/amiai; cf. Pap. Didot. 1.14-16 fa-r· av6pl Kal yvvatKl
J<Efµa,os v6µcs, I T<t>"'1v61ClnAovs ftv fxe,crripyetv mi, I Tij s·6a• &,
~ Tav6p(, TCXVT. <XVlT\V'TrOElV,Plautus,Men. 787-9 quotiensmonstraui
tibiuirout moremgeras,I quidilk f aciatne id opserues,quo tat, quidrerumgerat
(father to married daughter),Amph. 83S>-42 nonegoillammi dotemducotSSt
quat dosdidturI std pudicitiamtt pudortmtt stdatumcupidinem, Ideummetum,
parentumamorernet cognatumconcordiam, Itibi morigeraatqut ut munifia,sim
bonis,prosimprobis(wife to accusinghusband), Ovid, Ars 3. 585-6 hoe est
uxortsquodnonpatiaturamari: Iconutniuntillascum uoluertuiri. There was
alwayssomething a little paradoxicalabout female obedienceoutside legal
wedlock; cf. Plautus,Mm. 202, Most. 188 ff:, Terence,Andr.285 f[
315
 COMMENTARY
For restraint of female desirecf. Plautus, Amph. 840, Turpilius, Com. 37-9
quaemulieruoletI sibi suumamicum(Acidalius: summamamicamcodd.) ~
indulgentemet diutinum,I modiceatlpleparceeius muiat cupidines-fason's
denunciation ofMedea's lust at Euripides,Med.522 ff.,Clytemnestra'sboast
about her aooq,poaw11 at Euripides,I.A. 1159, and thestory told by Pluwch
at Mor. 140c indicatethat such restraintwas not a peculiarlyRoman marriage
ideal.
The Euripidean Hecuba appeals to Agamemnon not only through the
pleasurehe hashad of her daughter but alsothrough the quasi-legalrelation-
ship which the associationof Agamemnon and Cassandra has in her view
establishedbetween the Greek king and herself (vv. 834-5 TOVTOVK<XA~
6~v wraKT16E<m1V cri8EvI 6paae1s). 1
This would have been a difficult
notion to put straight into Latin. Ennius seemsto have substitutedthenotion
that Cassandra's sexualbehaviour was more like that of a legalspouse than
that of an ordinary concubine. The notion assumeda climate of socialideas
common to Rome and historical Athens. There is nothing specifically
'Roman' about the passageexcept its language.2

181 tibi in concubio ••• morem gerit: for concumbere of thefemale's part
in sexualintercourse cf. Terence, Hee. 393. Concubiumoccurselsewhere in
republican drama only at Plautus, Trin. 886 (concubium.. . noais).The phrase
moremgtrereand the adjectivemorigerus, to judge by Plautus, Capt.966, Cos.
463, 897 and Terence, AJ. 214-15, seem to have served in the common
language as sexualeuphemisms.3

uerecandeet modice: cf. Cicero, Tull. 5 uerecundemodiaque.Livy

26.49.16 uerecunde
ac modeste,Ammianus 16.7.3 uerecunde
et modice.

XCI
Personageswho have sufferedmisfortune frequently expressthe desire for
suicide in both Attic and Roman drama; c£ Euripides, .AIUlr.841 ff.,
Herakles1146 ff.,1247 ff., Tr. 1282 £, Plautus,Asin.6o6 ff., Cist.639 ff..Mil.
1240 ff.,Terence, .AIUlr. 6o6 (nondixit 'gladium'aut'l"i"eum' neessettragicum
-Donatus), Phorm.552, Seneca,Phoen.105 ff., Herc.f. 1221 ff. The Hecuba
of Euripides' •EKa(nlnowhere expressesthis desire but frequently makes
plain her distaste for life and asksothers to kill her: vv. 167-8 oVKhl 1,101
~(cs IayaOTOS tv q,&E1, 386-7 'l')µass·6yovres irpos ,rupav•AxtAA.~ I
1
Cf. Euripides, Tr. 308 ff.
a For a somewhat different view sec G. W. Williams,JRS XLVIII (1958), 20.
3 Cf. the use of nubereat Plautus, Cist. 43-4, nuptiae in the example of
translatio•.. obsunitatisuitatulaecausaat Rhct. inc. Her. 4.45.

316
 HBCVBA
1<EV'n!lTE, 391 vµetsSe µ• &Ma 6vyarpl avµcpovevcrcrre,
µ,i cpe{6eo6', 396
iroAAiJy' &vayKTi6vyarpl avv6avetv Aµi, 505-6 apa 1<6:µ' rn1crcpa~oo
Ta~ I 6oKOW'Axooots 1jA8Es;WSq>{A' aY Afyo1s.The chorus of this
play interpret the speech of the childlessand blinded Polymestor at vv.
1096-106 as motivated by a suicidalurge. It is therefore anybody's guess
where fragment XCI would have stood in the action of Ennius' adaptation.
The text of the fragment is plainlycorrupt. Vahlen printed the paradosis,
perhaps having in mind such passagesas Sophocles,Phil. 1004, Tr. 1089,
Euripides,Alie.837, Herakles268-9, Med.4~, 1244, Tr. 1178, Neophron,
fr. 2. 12, Aristophanes,Thesm.776. But the verb daterulesout an apostrophe
to the speaker'sown hands and the absenceof qualificationan addressto the
hands of others. Scaliger's anuisruns up againstthe difficultythat anus in
republicancomedy (it is absentfrom tragedy)hasa contemptuoustone when
applied to women of the upper orders. It could be that two distinct quota-
tions, miseretemanusdateand ... datef mum qui me animapriuem,have been
run together in Nonius' article as at pp. 467.32, 478.26 and elsewhere;
miseretemanusdatemight then belong to a version of Euripides,Hek. 59 ff.
and ... date f mum qui me animapriuemto some play other than Hecuba.
182 miserete: for the activeform cf.Ennius,Ann. 171 ; the deponentform
is regular in drama except for the impersonalmiseret.

qui me anima priuem: cf.Sophocles,Phil. 1427 vocrq,1Eis j3{ov,Plautus,

Men. 905 animapriuabouirum,Lucretius5 . 991 doneceosuitapriuarantuermina
saeua;contrast Plautus, Trin. 129 tkdisti.•. gladiumquise occideret.

XCII
This fragmentclearlybelongsto a versionof Euripides,Hek. 492-8 OV)(i\6'
6:vacrcraTcZ>v ffOAVXPVcroov <l>pvycZ>v, I ovx i\& Tlp1aµov Tov µfy'
6Aj3iovSaµap; I Kal vvvir6AtSµeviracr· <MCTn11<EV Sop{, I <XVT1l Se
OOVATl ypavs 6:-rroosrnl x8ovl I KElTal,1<6VE1 q,vpovcra 6VO"TI1VOV
Kapa. I q,evq>eO• yepoovµevetµ•, 6µoosSe µ01eavetv I el11irplv at~
irep1mcrelvTVXTJ TIVLThe Talthybiusof the Tp~Ses refers explicitlyto
hisown poverty (v. 415) but the matter of wealth and poverty is a constant
theme of the •E1<a131"1, discussedexplicitlyat vv. 317 ff., 492 £, 1218 ff., and
it is not surprisingthat Enniusshould make it explicitagain in an otherwise
fairly closeversionof vv. 49']-8.
Ennius here for once keeps the spoken iambic verse of his original.
Stylisticallyhe moves away with the nominal phrase mortemobpetam,the
repetition senex. .. senexand the alliterationof grauitergmuun.
183 mortem obpetam: a fairly common phrase in classicalprose but
absent from the rest of republicandrama.
317
 COMMENTARY

euenat: thisform and eue,umt are restored with a fair degreeof certainty
five times in Plautine comedy (Cure.39, Epid.290, 321, Mil. 1010, Trin.41 ), 1
each time at the end of a metrical unit and in a context of some solemnity;
eueniotand eueniont,which occur 19 times in comedy, were clearly the normal
forms in the common language.

13' tenex: this type of pleonasm is hard to parallel in either Greek or

Roman drama but c£ Ennius, Trog.29~ ntqut sepulaum... hobeot,portum
I ubi. .. corpusrequie.scot.
corporis,

grauiter gernarn! c£ Trag. inc. 116, Virgil, Georg.3. 133, Ammianus

14.5.7.

IP HI GENIA

The title Iphigeniois given to Ennius by Verrius, the Virgilian scholiasts,1

Aulus Gellius and Julius :Ru6nianus; and to Naevius by Nonius Marcellus
(p. 370.23). The piece which Nonius quotes as from Naevius' Iphigeniois
corrupt to the point of almost complete obscurity3 but since at least one
piece of verse quoted with hisname and no title comes from a play set on the
BlackSea coast4 we may suppose that he adapted Euripides' I. T. or another
tragedy on the same theme.S The five pieces quoted as from Ennius'
Iphigenioput thisplay without much doubt in Aulis.
Scaliger6 considered that Ennius adapted the extant version of Euripides'
I.A. and most scholars have accepted his opinion. Wilamowitz suggested?
that vv. S90-7 of the extant I.A. is a remnant of a version of the play con-
taining a chorus of soldiers and that Ennius adapted this version; M. Len-
chantinDe Gubernatis8 had earlier suggested that Ennius' immediate model
was a tragedy based on the Euripidean I.A. Certainly more than one acting

1
Cf. Pseud.1030 (aduenat),Rud. 626 (peruenat),Triu. 93 (peruenant).
1
Schol. Veron. Eel. s.88 uses Verriw' lexicon either directly or indirectly.
Serviusauct.Aen. 1. 52probablydrawson the same source; Festus{p. 510. 19 if.,
s.v. uostus)may have omitted the Enniw passagefrom his epitome.
3 Sec the restorations attempted by Mariotti, StudUrb XXIV (1950), 176, n
Bellum Poenicume l'ortt di Nevio (Rome, I9SS), p. 131, and 0. Skutsch, CR
N.S. I (1951), 146--7, vm (1958), 48.
4 Trag. 62 (Cicero, Orat. 152).

s The 'lcp1ywua of Polyidus was set among the Taurians (Aristotle, Pott.
16.14ssa6, 17.14ssb10).
6
Coniea. Va". Ling., on 7. 73.
1 Htrmts uv (1919), SI if. (= Kleine SchriftenIV [Berlin, 1962), 2891f.).
8
MAT, LX1D(1913), 416.
318
 IPHIGENIA
version of the Euripidean I.A. must have existed in theatrical circles and
since Lenchantin wrote there have turned up pieces of what looks like a post-
Euripidean tragedy dealingwith the sacrifice oflphigenia. 1 However the five
pieces attributed to Ennius• lphigeniacan be interpreted as coming from an
adaptation of the Euripidean I.A. no more free than certain other republican
adaptations whose originals are certainly known. The continuing popularity
of the I.A. in Athensand other Greek cities2 and Ennius• well-evidenced
predilection for Euripides give Scaliger•s opinion a certain advantage over
the others. Bcrgk3 tried to explain the divergences from the I.A. in terms of
'contarnioatio • with Sophocles• 'lcp1ya,e1a,a play which certainly dealt with
the sacrificebut was probably set in Argos rather thanAulis.4
A title lphigeniaseems to be referred to by the author of the rhetorical
treatise addressed to Hcrennius (fr. xcm) and by Cicero (fr. xcva). Since in
the one casethepersonages Agamemnon and Menelaus arc named and in the
other Achilles we may suppose that Ennius• tragedy rather thanNacvius' is
meant. Nonius appcan to quote from Achilles' speech in an article of
uncertain origin (fr. xcvd).S
No other republican play about events at Aulis is known but both Ennius
and.Accius wrote a Ttlephusdealing with a later episode in the saga. Behind
Ennius' Telephusthere stood quite probably Euripides' homonymous
tragedy, a tragedy to which the I.A. bore a striking resemblance; in both
plays there was a quarrel between Agamemnon and Menelaus and a speech
by Achilles complaining of the delay in setting out for Troy; the baby
Orestes was used in the I.A. by Iphigenia (on Clytemncstra's orders) to
softenAgamemnon, in theTfiAEcpOS by Telephus (perhaps too on Clytcm-
nestra' s orders) to impel Agamemnon to protect him against the spy-
hunting Greeks; various arguments and sententiaeappeared in both plays.6
A certain caution is therefore requiredin dealing with fragments not attri-

1
Pap. Brit. Mus. 2560; Milne. Cat. p. S1 nr. 78; A. Korte, APFx (1932), 53.
2
There are verbal allusions in Eubulus, fr. 67. 10 (370). Philetaerus, fr. 4
(701), Mcnander,Sam. 329 (1602) and the Alexandrian Machon 21-4 Gow
(22-3). Plautus refers to the story at Epid. 490-1, perhaps echoing his original.
One of Euripides' two plays about Iphigenia was re-performed at the Great
Dionysia of 342/1 (I.G. n• 2320). Euripides' name and scenes from the I.A.
appear on a Megarian bowl (nr. 10 in U. Hausmann, Htllenistische Rtliejbechn
ausattischenundbootischen Wnkstiitttn [Stuttgart, 1959]) of the late third century.
3 Ind. ltctt. Marburg1844. XIV (= Kl. phil. Sehr. 1229). In Commmtatiodt
FragmmtisSophoclis(Leipzig, 1833), p. 15. Bergk had proposed Sophocles'
•1,1yiv11aas Ennius' original.
4 Sec Zicli6ski, Tragodumenon Libri Tres, p. 271.
s Verrius could be the source; cf. Paulus, p. 163. 12.
6
E.g. Ttltphos fr. 714 ~ ~
I.A. 17 ff., 446 ff.; fr. 719 I.A. 1255-75.

319
 COMMENTARY
butcd to any particular play but appearing to concern the events and the
arguments of the I.A.
Shortly after hisspecificquotation of Ennius' IphigeniaIulius F.u6nian1u
quotes (fr. c1) without the name of either author or play three trochaic
tetrameters addressedby an Agamemnon to a Mcnclausand referring to the
projected slaying of Agamemnon's daughter. Columna gave these to the
Iphigeniaon the grounds of similaritybetweenthe Latin phrases and phrases
in I.A. 378-401. More decisiveto my mind arc the proximity of the specific
quotation of the Iphigenia and the lack of any other known republican
tragedy to which the tetrameters could belong.
Some anapaestsquoted by Varro (fr. XCVIa,b) as spoken by an Agamem-
non and attributed by Vcrrius (fr. XCVIc),apparently drawing on the same
source,1 to Ennius were given by Scaligerto the Iphigeniaon the grounds of
their similarity to I.A. 6-8. The similarity is not so very great but since it is
imposable to imagine an Agamemnon speakingthus at the beginningof the
Telephusor any other known Enoiao tragedy we may accept Scaliger's
opinion.
A. F. Naekei argued that in referring to Iphigenia in his catalogue of
patriots at Tusc.1. 116 (fr. xcrv) Cicero had in mind the actual wording of a
passageof Ennius' Iphigeniabasedon I.A. 1475-6, &ym µe -rav •1~.fovIKai
<J>pvyo.'>v OJ-m-0A1v. Diintzer added the paralld I.A. 1484-6 ws lµoiow,
el xpeC.::,v,Iatµa01 8vµao, Ieiaq,crr•l~cv.ef'fl<A).Here asdscwhcre one may
be scepticalabout attempts to extract the poet's phrases from Ciccro's3 but
Vahlen4 goes much too far in as.,crtingthat no reference at all is made to a
poetic treatment of Iphigcoia's death. It is perhapstrue, but nevertheless
irrelevant, that Ciccro's phrasesshow no trace of poetical rhythm. It is true,
but again irrelevant, that the parallelsadduced from the I.A. arc not verbally
close. It is not true that the metaphor in sanguishostiumeliciatur
sanguinesuois
sufficiently mild to be Cicero's own making, that it is no bolder than,
for example, that in oriensincendiumbtlli P&mici suo sanguinerestinxissent.
The words, as I shall argue later, recall the language of sacralpractice and
it is well known how fond Ennius was of importing the Roman sacral
language and the ideas informing this language into his adaptations of Attic
tragedy.
1
SceJ. Kretzschmer,De A. GellifontibusI (Diss. Greifswald,1860), pp. 16,
41, 4S, F. Mentz, Comm. phil. Ien. IV (1890), 51, R. Reitzenstein, M. Terentius
Va"o und]ohannesMauropus,pp. 32-3, G. Goetz, BPhW XXI (1901), 1034, R.
Kricgshammer, Comm. phil. Ien. VII (1903), 80.
~ Ind. kat. Bonn 1822, m ( = Opusc. 1 86).
3 Sec the attempts ofDiintzcr, RhMv (1837), 444, Bcrglc, Ind. lectt.Marburg
=
1844, xm ( Kl. phil. Sehr. I 256), Strzelecki,Bos XLID(1948), 161.
4 Ind. kctt. Berlin 1879/80, 14 ( = Op. ac. I 101).

320
 IPHIGENIA
Study of the organisationof Cicero's argument gives positive grounds for
thinking that he is not speakingwith his own voice but alluding to the words
of another writer. He is discussingthe examples of patriotic self-sacrifice
given by •rhetores'. AJ a rule he brings forward the name of the hero or
heroine (repetunt ab Ereaheo cuius .filiae. .. Codrum [?] ... Menoeausnon
praetermittitur . .. Harmodiusin ore et Aristogiton. .. LacedaemoniusLeonuuu,
ThebanusEpaminondasuiget) and describeshis or her deed with a preterite
verb (mortem expetiuerunt. .. se in mediosinmisit hostes. .. largitusest palriae
suumsanguinem ). Iphigenia's name, on tbe other hand,is not formally intro-
duced like the other names and her deed is describedin the present tense.
Furthermore her death makes the number of mythical examples four,
whereas one would expect three to balance the historical ones. I would sug-
gest therefore that Cicero took:the daughten of Erechtheus, Coclrus and
Menoeceus 6.-omhis philosophicalsource and added Iphigenia as an after-
thought with a speech 6.-oma Latin tragedy in mind.The present tense of
eliciaturwould suit such a speechbut is quite out of place in Cicero's di,..
course. Ennius' Iphigenia,which Cicero knew well, was very probably the
tragedy quoted.
Dobree1 assignedthe trochaic piecesof a stichomythic dispute between an
Agamemnon and a Menelausquoted by Cicero at Tusc.4. 77 to the Iphigenia
on the grounds of their similarity to sentencesin I.A. 317-33. Ribbeck:and
Vahlen both accept Dobree's idea. However the Iphigeniais not the only
republican tragedy known that could have contained such a dispute.
Hartung2 gave the piecesto Ennius' Telephus.
In a letter to Atticus (13 . 47. 1) Cicero alludes unmistakeably to a tragic
speech addressedto an Agamemnon--posteaquamabs te Agamemnonon ut
uenirem(namid quoque fecissemnisi Torquatusesset)std ut scriberemtetigitaures
nuntius,extemploinstitutaomisi;ea quaein manibushabebamabieci,quodiusstras
edolllui.3Cicero's identificationof his addresseewith the mythical Agamem-
non means of itself very little4 but the words extemploSand edolaui'and the
1
Aduersarian (Cambridge, 1833), p. 373; cf. Diintur, RhMv (1837), 444.
Earlierscholan had mistakenly referred the piecesto a quarrel between Atrcus
1
and Thyestes. EuripidesRestitutusI, p. 202.
3 Cf. Varro in epistulaIuli Cauaris (Nonius,p. 263. 3): quemsimulacRomam
uenissemi adtigitaurls nuntiusextemploteastin curriculum eontuliproperepedes.
4 Cf. the passagesof Plautus discussedby Fraenkel, Pl. im Pl. pp. 95 ff.

(= Elementi, pp. 89 ff.), Cicero, Att. 1. 18. 3, S. Rose. 98, Rhet. inc. Her.
4. 46, Forttmatianus,Ars rhet. 3 . 7; cf. also the Atheniannicknameslistedby
Anaxandrides, fr. 34.
s Elsewherein Cicero only at Q. Rose. 8, absentfrom Caesar, common in
drama.
' Elsewhereonly at Varro, Men. S9 (quoting Ennius) and 332.
21 321 JTO
 COMMENTARY

cataleaic trochaic dimeter tttigit aurtsnuntius1 arc dearly of dramatic origin.

lusmas was not a natural word for Cicero to use in correspondence with
Atticus. Lade~ argued that Cicero had in mind Ennius' version of I.A.
607-34, the scenein which Clytcmnestra arrived from Argos with Iphigenia
in obedience to Agamemnon's first letter.3 Other tragic contexts arc
imaginable.
Diintzet4 compared the words 'I'Mlisilli mattr attributed by Varro to
Ennius at Ling. 7. 87 with a passageof the fust choral ode of the I.A .• 207-8
'Axv.fiaI TOV & emsTa<E.Vahlen printed the three words among the
fragments of the lphigenia,comparing I.A. 701 and 708. The Achilles,the
HtctoriJlytraand the Ttlephusarc equally likely to have contained such a
reference to Achilles' parentage.
A. GrilliS argued that Ciccro's words at Tusc. 2.33. sin ttctus Vokaniis
armis,id estfortitudint,rtsiste,refer to Ennius' version of I.A. 1072-3 6,ri\oov
•Hf(XICJ'TOTrOVCA>V I ~ lvSvra. Cicero may refer to a poetic
account of the armour of Acbi11esbut Ennius' Iphigeniais not the only
possible source.
F. Skutsch6 argued that Cicero's words at Tusc.3. S7, neesileturilludpottn-
tissimirtgisatUtpatstum qui l4U"4lsenemetf ortunatumtsst didt quodingloriussit
"'f't ignobilisadsuprtmumdiemperuenturus, refer not to Euripides, I.A, 16-18
but to Ennius' version. He seemsto have convinced A. Klotz.7 He was clearly
right to think that Cicero had a Latin tragedian's verses in mind. Scholars
now assume that in the case of direct verse quotations Cicero always draws
on a Latin poet unlesshe names a Greekas the author8 and the same assump-
tion should be made in the case of indirect quotations. nlud pottntissimirtgis
atlllJ"lffl"mrefersto a Latin versejust as surely as docsqui hoeanapatJto dtantur
at Fin. 2. 18 where the run of the argument allows him to quote directly.
However the Agamemnon of the Iphigeniawas not the only rtx pottntissimus
of republican tragedy and the sentiment in question not by any means an
uncommon one.9
1
Cf. Plautus, Poen. 1375-6 quodutrbumaurismeasI tttigit, Rud. 233 mto uox
mulitbrisouristttigit mtas. a Anal. seen.p. 15.
3 The most recent attempt at restoration is that of Stn:dcclti, in Charisttri12
TluuldMoSinko ... oblat12(Wanaw, 1951), 344. Sec the more sceptical ap-
proaches of F. Sk.utsch, RhM LXI(19()6), 618 ( ==Kl. Sehr. 308), and Dablmaoo,
MusH VII (1950). 216-18.
4 RhM v (1837), 443. 5 Acme x (1957), 75.
6
RhM LXI (19()6), 611 ( ==Kl. Sehr. 302).
7 Scaenicorum RonuuwrumFragment12 I (Munich, 1953), p. 101.
1
Cf. Przychock.i, F.osXXXII (1929), 215 ff.
9 Cf. Euripides, Hipp. 1028 fi T&p•6Ao{µf1v~s Ion 621 ff.,
&vC.:WVl,IOS,
Seneca, Thy. 393 ff. Cicero's philosophical source probably quoted Euripides,
I.A. 16-18 (cf. Plutarch, Mor.467B-F, 471 c, 474D ).
322
 IPHIGEN.IA

. At Diu. 2. s6-7 the notion that the crowing of cocks has a supernatural
divinatory significance (c£ Diu. I • 74) is refuted. Cicero takes from hissource
Democritus' explanation of the well-known fact that cocks crow before
dawn and adds quiquidemsilentionoctis,ut aitEnnius,fauent f aucibus
russiscantu
plausuquepremuntalas(see fr. CLXXIX). It is hard to say whether silentionoais
is meant to repeat or to complement antelucem.In either case Ennius' ana-
paests must be interpreted to refer to the noise of the cocks. A reference to
their silence would be quite absurd. The point is that cocks crow any old
time and do not need divinity to prompt them. However that may be, the
passages from theI.A. prologue compared by Ribbeck 1 and Vahlen,2 9-u
oOKow~6yyos y• oo-r· opv{&wvI oVTE8CXA6:aO'T}s • cnyal s· <MµCA>V I
T6v& KaT• EOp1,rovfxovcnv, I 56-7 AEVKa{ve1 IT66e~ fi6T\).aµ,rova•
T\WS,refer to the particular situation by the Euripus. Ennius' anapaests, unless
Cicero has altered them drastically, seem to have referred to the behaviour
of cocks in general.3
In his letter consolingHeliodorus for the death of Nepotianus (Epist.
60.14.4) St Jerome appears to draw from Cicero's now lost Consolatio 4 an
Ennian sententia-plebesin hoeregiantestat:locolicetj lacrimare
plebi,regihoneste
1
non licet(fr. CCXV}-similar to I.A. 446--9T)6vaywe1a 6 OOS fxe1TI XPficn-
µov. I Kai yap Sal<pvaai~6{oos avrots fxe1,I ooravTaT• elmtv. Tc;'> St
yewa{c+>,vcnv I &\,o).~ Tavra. On the grounds of this similarity
Columna assigned the two trimeten to the Iphigeniaand all editon have
followed him. However such similarities can be treacherous where sententiM
are concerned and no assignations should be based upon them. Homer's
gods and heroes wept freely, likewise those of fifth-century Attic tragedy.S
But the socialetiquette of the Athenian aristocracy of the late fifth century
condemned the public display of emotion. 6 So too that of the Roman repub-
lican aristocracy if we may judge from Pacuvius' alteration of Sophocles'
Nhrrpa, Cicero's approval of this alteration,? and Cicero's own treatment

1
~st. seen.p. 254, Die rom. Trag. p. 95.
2
Ind. ledt. Berlin 1888/9, 10 ( = Op. ac. 1409).
3 Columna, C. F. W. Mueller and F. Skutsch, RhM LXI (1900), 610 ( = Kl.
Sehr. 301), include siltntio nodis in the Ennian fragment. Zillingcr, Ciceround
die altromischenDichter,p. II5 n. 1, gives it to Cicero.
4 Sec A. Luebcck, Hieronymusquos nouerit saiptorts et tx quibus hauserit
(Leipzig, 1872), pp. 105 f., 157, Kunst, De S. Hieronymistudiis Ciaronianis,
p. 142.
s Cf. Sophocles, Tr. 1046-1n, Phil. 785-820.
6 See Plato's attack on the poets for allowing the heroes to moan and beat
their breasts: Polittia 10.6o5c.
7 Tusc.2. <J9•

323 ::n-2
 COMMENTARY

of Sophocles' T pax{v1cn.1 One could imagine the sentiment of pkbts in hoe

regiantes"1t:locolicetloainuireplebi,regihonestetum liat 2 in severalof Ennius'
tragedies.
XCIII
The words which the rhetorician quotes are plainly of tragic origin and
make good sensebut do not form a recognisable metrical unit. 3 It does not
follow that becausethe memoriser is advised to think of actors preparing to
perform a scene ofEnnius' Iphigenia(seeabove, p. 319) the words necessarily
come from this play.•
XCIV
For the rationale of Cicero's discoune see above, p. 321. Sanguisis to be
understood with hostiumfrom the previous sentence. Very probably the
polyptoton sangui.s sanguine(for which see above on v. 180) stood in the
speech of Iphigenia which Cicero has in mind.
Sanguinemelicerewas a common medical term ( T.L.L. v ii. 367. 22 ff.) but
the usage here is clearly metaphorical. The blood of the enemy will be madeto
flow like water from a fountain or rain from the sky ( T.L.L. v ii. 367.71 ff.).
I suggest that Ennius parodied the language usedin prayers to Iuppiterelicius
at the aquaelidum.S
XCV
At Rep. 1 . 30 the tragic verses are directed in the first instance at C. Sulpicius
Gallus, an aristocrat famous for his knowledge of astronomy, and in the
second at those who would prefer to discussa vision of two suns reported as
a prodigiumrather than the sedition threatening the welfue of the state. At
Diu. 2. 30 physid in general are the object of attack.
Donatus (Ter. Ad. 386) appears to have in mind the story of the natural
philosopher Thales and the slave girl recounted by Plato (Theait.174A) and
many others (Hippolytus, Phil. 1. 1, Diogenes Laertius 1 . 34, Aristides, Or.

1
Tusc.2.20-2: the hero's cries of pain at 1081-5 are omitted from an other-
wise fairly close version.
2
Cf. Ovid, Fast.4. 845-7 hateubi rex dididt lacrimasintrorsusoboruuI deuorat
et clausumpectoreuulnushabet.l.flerepalamnon uult, exemplaquefortiaseruat,Met.
13 .474-5 at populus lacrimasquasilia (i.e. Polyxena) tenebatI non tenet.
3 The usual reading iam domum itionemregesAtridae parantproduces a tri-
meter with the highly unusual division of syllables At-ridae. The transmitted
reading makes equally good if not better sense (for the construction cf.
Plautus, Aul. 201-2, Stich. 283) but produces a trochaic tetrameter lacking one
trochee and afflicted with the same unusual division of syllables.
4 Cf. Bergk, Ind. lectt.Marburg1844, XIVf. ( = Kl. phil. Sehr. 1 230 f.), C.
Pascal, RFIC XXVI (1898), 30.
s See Aust, RB vii (1905), 2366-,, s.v. elicius,Latte, Rom. Rei. pp. 78 f.
 IPHIGBNIA
48.85, Tcrtullian, Nat. 2.4, Schol. Lucian, p. 147 Rabe) to illustrate the
common man's distrust of philosophical spcculation. The words in Syrum
could be interpreted as referring to Thales; for the story of his Phoenician
ancestry c£ Diogenes Laertius 1 • 22. Certainly Donatus had Thalesin mind.1
The words which Donatus puts in the slave girl's mouth obviously come in
corrupted form from thetragic Acbi1Jes' speech. I suggest that some earlier
commentary upon Terence's play contained not only the story of Thalesand
the slave girl but alsoa clearly designated quotation from the Iphigenitl'and
that in the course of time and successive epitomisations the two elements of
the original note became conflated.
For Cicero's loose mode of introduction at Rep. 1.30, ille de lphigenia
Achilles'the famous words of Achilles in the Iphigenia',c£ De orat.3 .217
AtreusJeretotus'almost everything said by Atreus ', Fam.7. 6. 2 Medeamcotpi
agere'l have begun to act therole of Medea •, Att. 12. 4S . 3 tu ueroperuolga
~
Hirtium ( 12. 47. 3 Hirti librum. .. diuolga ), Horace, Sat. 2. 3 . II stipare
PuuonaMenandroet al.
The tragic verses have been much emended since the discovery of the
palimpsest fragment of the De republica. 3 The emendations offered arc
extremely unconvincing and one may doubt whether the words and phrases
transmitted are sufficiently anomalous to justify thesearch for replacements.
Both tragedy and comedy mingle acatalectic and catalcctic tetrameters in
speeches of elevated tone; very often, as here, a catalcctic verse concludes a
series of acatalcctic ones (c£ Plautus, Aul. 727-30). I should put a question
mark aft.erobseruationis and interpret the fuse venc as 'what right have
astrologito look for signsin the sky (andgive men like us advice on how to
conduct theaffairs entrusted to us)?' Ennius' Latinity is admittedlypeculiar.
The normal phraseology would be quidastrologis signain caeloobseruatio est?
(c£ Plautus, Amph. 519 quidtibiJwnccuratioestrem,uerbero,aut muttitio?,Asin.
920 quidtibi huncreceptioadte est meumuirum?,Aul• .µ3 quidtibi nostactiost?
et al.).4 The subjunctive verb indicates impatient expectation of the answer
nihil; c£ Plautus, Mil. 615 quishomosit magismeusquamtu's?, Amph. S76-JJ
quidhoesit I hominis?The postponed quidand the locution quid.. .obserua-
1
For Thales as the archetypal sapienscf. Plautus, Baah. 121 ff., Capt. 274 ff.,
Rud. 1001 ff.
3
Cf. the way in which Cicero glosses the story of Anaxagoras' reception of
the news of his son's death with a speech by a tragic Telamo, Tusc.3 .28, 3. 58;
his quotation of both Euripides and Ennius on the real nature of the supreme
god at Nat. deor.2.65.
3 For an account of the emcnders' attempts see K. Ziegler, HermesLXXXV
(1957), 495-501.
4 See T. Bogel, N]bb Suppl. XXVIII (1903), 57 ff., Lofstcdt, Syntattiai 11,
p. 253, G. Pasquali, RAI vu iii (1941), 29 ff.

325
 COMMENTARY

tionisperhapsalso indicate impatience on the speaker's part. Astrologorumis

the most difficult word to defend. In third and second century drama geni-
tives accompanying verbal nouns in -tus arc normally subjective while those
accompanying verbal nouns in -tio (-sio)are objective. 1 Ncvcrthdess the
word cannot be emended and to talceit with either signaor ntmo is even harder
to justify. One finds the pronominal adjective replacing the dative at
Plautus, Cas. 261 mesinascurareancillasquaemeaestcuratio,Persa586 tua mers
est tua indicatioet al. and subjective genitives of nouns all through official
inscriptions of the late second century (e.g. C.1.L. 12 583 .m deea retius petitio
nominisquedelatioesto, VI quaestiotius praetorisesto).
At I.A. 919-74 the Euripidean Achilles denounces the Greek leaders one
after the other, describing Calchas as an unreliable adviser: 956-8 Tis &
as
µav-ns m· &vfip,I 6i\ly" &A118f\, 1r0Ma Se 'flEVSi\Afye1I "TV)(cbv,
OTavs~µ~ TVXTJ, 61o{xETcn. Calcbas' mode of divination is not mentioned
in the I.A. But elsewhere in early Greek epic and tragedy he is talked of as
an oloovo,r6i\05,that is the equivalent of the Roman augur.2 Earlier in the
I.A. (520-1) Agamemnon and Mcnelaus, with Calchas in mind, denounce
diviners as a class.Such general denunciations of diviners arc common in
Attic tragedy (e.g. Sophocles, Ant. 1055, Euripides, Hel. 744 ff., fr. 795).
The anti-intellectual sentiments of the Ennian Achilles are common in
Attic tragedy(e.g. Sophocles, fr. 671 µ1aool,1EV ocrns Tacpavf\mptO'Koml,
Euripides, Hel. 757yvcbµ11s•ap{O'Ttl µav-ns i\ T evj:x>vi\{a,
0
fr. 913 µen-
Ca>p<>i\6yoov airarasoovToi\µ11payi\ooaa·Eh<oj:x>i\ET
... O'KOAlO:S mpl TOOV
&cpavoov, ovSevyvcbµ11s s·
µe-rixovaa, 973 µav-ns a()lO'TOS&rns EIK<X3€l
Kai\C)s). Whether or not astrological divination was known in fifth-century
Athens3 the tragedians certainly excluded it from the heroic world. The
Aeschylean Prometheus docs not reckon it among the types of divination he
discovered for men. 4 He talksof the stars only as guides to the scasons.SThere
is remarkably little talk of the stars in either Ionian epic or Attic tragedy;
they were of interest to farmers rather than to men of state. The only pos-
sible tragic reference to divination by the stars is in Euripides' account of
Melanippe' s learned mother Hippo: fr. 482 ,rpena 1JEV Ta 8ela ,rpovµav-
I XP1lO'J.lOl0'1
TEVO'QTO 6
aacpanv &o-ripoovtrr• &vToi\ats.
1
SecA. W. Blomquist, De genitiui apud Plautum usu (Diss. Helsingfon,
1892), pp. 111 ff., Marouzeau, MSL xvm (1914), 146 ff.
2
Sec Homer, II. 1 .68-72, 13. 70, Euripides, I. T. 662, Propcrtius 4. I .109 ff.
3 SecW. Capelle, HermesLX (1925), 373 ff.
4 Prom.484-99. Cf. Euripides, Hilt. 211-13.

s Prom.454-8. Cf. Sophocles, fr. 399.

6
Wilamowitz, SB Berlin 1921, 74 n. 3 ( = Kl. Sehr. 1 453 n. 2), declaredthat
only arc involved. A. Bouche-Leclcrcq, L' astrologie
610cJTH.1Eia g,ecque(Paris,
1899), p. 37 n. 1, was less dogmatic.
326
 IPHIGENIA
Roman tragedy, on the other band, both republican and imperial,isfull of
talk of astrologcn, the planets and the zodiac:e.g. Pacuvius 407 ,u,m si tfU«
euenturasunt prouideantaequiperent Ioui (i.e. astrologi;see Gcllius 14. 1 . 34),
Accius 331-2 luciferalampadeMettm exurat Iouis, 678-So peruadepolum
spkrulid4mundisiderabinis{Popma: bigiscod.) continuis sex aptisignis(Scaliger:
continuise apit spoliiscod.), Sen~ Thy. 836, 844-66, Here.f. 944 ff.,Otd. 40,
251. Roman epic treats the heroic world similarly: e.g. Virgil, Am. 3. 36o,
10. 176, Statiw, Theb.3. 558.
It is possibleto regard Enni.us'versesas coming from an adaptation of I.A.
919-74- Elsewhere without any doubt (secbelow on fr. cvm)Enni.usinserts
sententioeabsent from his originals or rewrites Sffllentioe(see above on fr.
LXXXIV) with the world of experience of birnsc:lfand his audience in mind.
To have ridiculed Calcbas ~ an augurmight have given offence to his
like all private diviners, was generally
aristocratic patrons1 but the ostrologus,
despised.a

186 cam C.apra aut Nepaaut exoritur nomenaliquod belaarum:for

the seriesof alternatives c£ Plautus, Aul. 24 aut turiaut uinoaut aliquisemper
supplicat,Capt. 382 pater exspectataut me aut aliquemnuntium, True. 53 aut
empt4ancillaaut aliquoduasumargenteum.For the periphrasis with nomenc£
Euripides, I. T. 662-4 T6v T' w olCA>VOlS aocpov I K<XAxavr·'Ax~ T'
6voµa, Kai Tov &6A1ov I 'Ayaµiµvov', C.I.L. 1 581 {186B.c.) neuenominus
1
Latini neue sociumquisquam,Ciczro, Fam. 7. 5. 3 huic ego nequetribunatum
nequepraefecturam nequeulliusbeneficicertum nomenpeto; Vahlen, Ind. kctt.
Berlin 1878, 8 ff. ( = Op. ac. I 58 ff.), Lo&tedt, F.ranosx (1910), 22 ff.,
Coniectanea {Uppsala. 1950), pp. 42 ff.
Capranormally represents the constellation Ate in later Latin poetry.
However at Anth. Lat. 622. 5 and 626. 5 it represents Aly6t<Epoos.3 The
astrological context docs not demand necessarilya zodiacal constellation•
but makesone likely.
Nepa represents IKopirloSin later poetry but Verrius flaccusS seems to
have thought that Enniw could mean KapK(VoS.

187 caeli scrutantur plagu: Varro, Men. 233, Ovid, Met. 11.518,
Seneca, Oed. 972 repeat the phrase caeliplagasas a mere periphrasis for
1
Pacuvius comes close to attacking the state augun at Trag.83 ff.
:a SecCato, Ag,. S4- auguremhariolumChaldaftnnnequemconsuluis#uelit,
Appian, Hisp. 85, Plutarch, Mor. 201 B (on Scipio and~ in the Roman
camp at Numantia).
3 For Caper = Aly61clpc.,scf. Manilius2. 179 and Housman's note.
4 See on the constellation Af~ Manilius s; 128-39.

s Nonius, p. 14-5.12, Paulus, p. 163. 12; see above, p. 319 n. s.

 COMMENTARY

caelum.Ennius' context (exoritur)suggests that he has the circle of the

horizon in mind.Elsewhere in drama plagaedenotes a (hemispherical, circu-
lar?) hunting net. 1 There may be the same link here as between Homer, 11.
5 .487 ~ a\f'l<n A{vo1•&Awreiravaypov and Plato, PhaiJr.247B 6Kpav
v,roTI'}vovpav1ovQ\fll6cx1rOpE\IOVTcn.
It is difficult however to under-
stand within this area of metaphor Ennius, Sat. 65 subuloquondammarinas
propterastabatplagas(circle of 'Cul<Eav6s?).

XCVI
At Ling. 7. 73-s Varro quotes Ennius' anapaests as an illustration of how the
poets represent multanox and turns aside to discussthe Latin names of the
constellation of the Great Bear. It is certain that the latter discussion comes
from the work of another scholar and highly likely that this work is alsothe
source of the anapaests. 1 If so, Varro' s 'interpretation-multam noctemostendere
uult-may be determined by the general structure of his discussion of poetic
accounts of time rather than by the dramatic context of Ennius' anapaests.
Scaliger accepted the manuscript division of speakers at Euripides, I.A. ~8
and made a similar division of speakers in Ennius' anapaests after uidetur.3
Hermann took over Scaliger's basic assumptions and put the division after
clipeoto take account of Ling. s. 19. On any unprejudiced view the phrase in
altisonocaeliclipeogoes with superat. V ahlen 4 understood all the anapaests as
uttered by the one speaker, Agamemnon, to himsel£ The word uideturis
hardly proper to a soliloquy. I therefore take quidnoctisuideturas a question
addressed by Agamemnon to a second party and the other words of the
quotation as information about the state of the sky designed to help the
second party's reply. The relative positions of the stars do not of themselves
£ix the time of night.
The text ofVarro's discussion as well as that of his quotation is corrupt
and there is little hope of certain restoration. Stellaslacks a complementS and
sublimeagenspresents a quite intolerable hiatus.6
1 3
Plautus, Mil. 6o8, 1388, Poen.648, Trin. 237b. See above, p. 320.
3 Even quoting at first hand ancient writers were prone to make no reference
to such divisions; cf. Cicero, Diu. 1 .66 and above, p. 207.
4 Ind. lectt.Berlin 1888/9, 14 ( = Op. ae.I 414).
s Vahlen (Ind. lectt.Berlin 1888/9, 14 [ = Op. ac.1414], HermesXLID [1908],
514) seems to have taken stellasas one of the objects of agens.F. Skutsch (RhM
LXI [1900], 6o5 [ = Kl. Sehr. .296]) took it as the object of superat,comparing
Plautus, Stich.365 commodumradiosussesesolsuperabatex mari.Neither explana-
tion gives a very clear image. Ribbeck (Coroll.p. xxv) saw the difficulty.
6 Leo, Ausg. kl. Sehr.1 196, pointed out that agensmight be treated as having

suffered iambic shortening. But this phenomenon is comparatively rare in

tragedy even in iambo-trochaic verse (for anapaests cf. Accius 290).
328
 IPHIGBNIA
Tumebus' sublimisagensand Vahlen's sublimeagitatUremove the hiatus.
Sublimumagenswould be a simpler change. 1 For the hiatus sublimumagensc£,
in anapaestic verse, Plautus, Ba«h. 1193 mmtem amabo,Cure.137 plordamabo;
in dactylic verse Ennius, Ann. 307 aeuumagtbant,"332 militumocto, Var. 3
Scipioinuide,Cicero, Arat. fr. 24 etesioein uad4ponti (Drat.152), Luaetius
6. 716 etesioeesse.Where hiatus in general is concerned the tradition of
republican dramatic verses hasit much more often aficr words ending in
vowd + m than after words ending in l.3 However this may simply rdlectthe
rdative frequency of words ending in m and e in the lexicon.
Varro's remark etplaustrum apptlldtum,a partetotum,ut multaperhaps con-
tains the clue to what is wrong with Ennius' temosuperatstellas.I take temo
literally as the yoke-pole formed by Alioth, Mizar and Benetnasch 4 and
supply plaustriwith stellas.Ennius would then be describing the rdative
positions of the seven stan when Mizar hasreached its highest point in the
sky.
In the anapaestic dialogue which opens the I.A. Euripides' Agamemnon
either asks his aged servant or muses to himselfabout the appearance of the
sky: 6-8 Tis 'TTO'T' &p' aCMT\p66e nop6µ&VE1I O'Etp1cs &yyvs Tils
hrrarropov I 1The1a6os ~aaoov ht µEO'OT)P11S;S The mention of the
Pleiades at the beginningof the campaigning season as still unset would have
put the time somewhat before dawn. Sophocles describes Palamedcs' dis-
covery of how to divide the watches of the night by thestars at fr. 399. 8 f[
and Euripides twice elsewhere (.Rhes.528 ff., Phaeth.fr. 773 . 19 f[) uses the
Pleiades to fix the time. The Bear is never so used in Attic tragedy. In
classical Roman poetry, on the other hand,whatever the position of the
observer, the position of the Bear seems regularly to fix the time of night
and that of the Pleiades the season of the year. The parallelism of phrase
~aaoov ht µEaO'T)pT)S ~ sublimumagensetiam~ etiamnodisiterconfirms
to some extent the external indications that the Greekand Latin anapaests are
linked. Why Ennius changed the seven stars of the Plciades for theseven of

1
For the form see Nonius, p. 489. 7. A number of adjectives swing in early
Latin between -us -a -um and -is -t ( futtilus, pronus,steriluset al.).
" Cicero, Brut. 58; c£ Timpanaro, SIFC N.S. XXI (1946), 53.
3 See Maurenbrechcr, Hiatus und Verschleifung, pp. 16 ff.
4 Cf. Souda •p 295 ~vµos••• Kai Ti'ls"ApKTOVot Kcrra -n'iv oOpavy' &crripes
wo •Hpcn<MfTov, Cicero, Arat. fr. 16 (Nat. tkor. 2. 109) ArctophylAxuulgoqui
diciturtsstBootesIquodOYABI'I'BMONll .WIJ'NC'I'-'.M praestquatitArctum(~Aratus
92-3 T6v f &v6pEshnw{OW1 BO<m'\Vj owex· &µcx~alflS hra~l,IM)S 1t6nat
"ApKTov), Ovid, Met. 10.446-7 tempuserat quo cunctasiltnt interquetrioMsI
flexerat obliquoplaustrumtemontBootts.
5 For the interpretation of the Greek see Housman, CR xxvm (1914),
267.
 COMMENTARY
the Bear it is hard to say.1 Ancient critics were much divided on the precise
significance of Euripides' words. 1 Rome's different position on the globe
and the different times of the year at which her dramatic festivals were held
may have made astronomical observations comprehensible in Athens at the
time of the Great Dionysia seem ludicrous when literally translated. In any
case the Latin Agamemnon's question is a different one from the Greek.

188 quid noctu uidetur: 'how late do you think it is?'; c£ Plautus,
Amph. 153-4 quimealterestaudacior homo. .. quihoenoaissolusambulem?,164,
292, 310, Cure. 1. For the omission of personalpronouns in high dramatic
style see above on v. 17.

188-9 in altiaono I caeli clipeo: it is not certain that Apuleius, Socr.2

refers to this passage; for repetition of phrases in drama see above on
fr. XXXIV.
For images presenting the visible sky as a hemispherical container and
their absence from Attic tragedy see above on fr. xxxm. The only classical
anceston of Ennius' shield image are the descriptions of Achilles' shield and
its decoration by Homer (n. 18.483-9) and Euripides (El. 464-9).

190-1 sublimum agem I etiam atqae etiam noctis iter: 'making its
nightly way still high in the sky'. Comedy and classical prose employ
regularly the phrases iterfaare, confaere,perficere(c£ Plautus, Cas.9(58,Mere.
913, Persa221); iter agere,etc. are absent. For Ennius' phrase c£ Ovid, Ars
2. 84 altiusegit iter, T.L.L. I 1382. 69 ff.
Etiamatqueetiamnormally means 'again and again' (cf. Plautus, Aul. 614,
Trin. 674) but for Ennius' usage c£ Gellius 2.30.3 a uentoquidemiamdudum
tranquillasunt, sedmareest etiamatqueetiamundabundum.

XCVII
Verrius Flaccuscommonly quoted republican poetry in metrical units and
the words which Festus transmits here will form an iambic trimeter if the
first syllable of Acherontemis treated as short. But Plautus has this name fre-
quently enough for there to be no doubt about the prosody regular on the
early second century stage. We must suppose that Festus' quotation or the
tradition is defective and treat the words as an incomplete trochaic tetrameter.

1
Cf. Virgil's imitation of Od. s.270 ff. at Aen. 3. 512 ff. and Macrobius'
discussion (Sat. 5. II. n).
a See Theon Smym. De astr. XVI, p. 202 Martin; there seems to have been
agreement about what is meant at Rhes. 527 ff. and general condemnation of
Euripides' ignorance.
330
 IPHIGENIA

Columna thought that Ennius was adapting I.A. 1503 8avovaa s•OVK
avcrlvoµcnand later scholan have adducedother passages uttered by the
Euripidean Iphigcnia after her decisionto go willingly to her death: 1375
Ka-r6cxvdv~ µ01 6E6oKTcn,I 506--9AcxµircxSoV)(os ~pcx I tuos TE
q>fyyos,mpov rnpov Icxt&>vcx KOO µotpcxvolKT)aoµEV. , xcxtpiµo1,q,O.ov
cpaos.There are closerverbal parallelsin other plays: e.g. Euripides,Hek.
414 cmuµ1 s,;KaTCA>, Hik. 1022 <I>epcn:q,ove{cxs 6CXAaµovs,
i\~CA> Herakks
6' ls VEl<f)OVS.
1247 elµ1 yi)s vrro, T,. 460i\~CA> From 1368 Iphigcniamakes
no mention of the underworld; at 1437 ff. she forbidsany of the customary
funeral rites and her wishis ahnost instantlyapproved; she is to be the pos-
session of Olympian Artemis, not of Pluto. Before her change of heart
Iphigcnia expressesseveral times (1219, 1250-1, 1281-2) the traditional
Greek horror of the underworld. Atherontemobiboetc. should therefore be
treatedas coming from an adaptation of some utterance by the Euripidcan
Iphigeniabefore v. 1368.1

192 Acherontem obibo: cf.Apuleiw,Met. 4. 20 uitaemetasultimasobiret.

The phrasesmortemobireand diem (suum)obireare frequent in comedy; obire
does not occur in other contexts.
Etruria seems to have been the intermediary between 'Axip<J.>v,one of
the rivers of the heroic underworld, and A.cheruns,
the underworld itselfin
the imaginationof early secondcentury Romans.2

Mortis thetauri: cf. the alleged epitaph of Naeviw, ap. Gell. 1.24.2,
postquamstOrci traditusthesauro.For the personificationof Morscf. Plautus,
Capt. 692 obsutew tuas te Mortimisero,Cist. 640 recipeme adte Morsamicumet
beneuolum.The third and secondcentury dramatists usuallyspeak of Orcus.
Enniw' grandiosephraseologyalludesobliquelyto the name Dis patergiven
to the Greek TlAovrCA>v when he was introduced to the Roman state cult in
249.3 For similar obliquity cf. Sophocles,0. T. 29-30 ~CXS s• I •A16T)S
a-micxyµotsKcxly6o1s irAovr(3ercn; for the explicit etymology Ennius,
Var. 78 Pluto Lotineest Dis pater,alii Orcumuocant,Cicero, Nat. deor.2. 66.
The Greek borrowing thesaurus occun frequently in comedy, usually in
the singular; the plural is always accompaniedby other signs of stylistic
elevation (Plautw, Aul. 240, Mil. 1064, Pseud.628, True.245 ).
1
M. L. Cunningham has interpretedthe fragmentas a questionand com-
pared I.A. 1219 Ta6' v,royiis µfi µ• l&tv &vaylcqs; see 0. Skutsch, HSCPh
LXXI (1967), I.µ.
~ See Pasquali, StudiEtruschi1 (1927), 291 ff. ( = Pagineme,u, stravaganti
di un
filologo~pp. 163 ff.). 'Axif"'>V often indicatesthe underworld in Hellenistic
poetry: cf. Asclepiades,A.P. S . 8 S. 3 et al
3 See Latte, Rom. Rei. pp. 246 ff.

331
 COMMENTARY
obiacent: not elsewhere in republican drama; tragedy has the simple
iaare4 times, comedy 16. Ennius choosesthe unusualcompound for the sake
of word play with obibo;c£ his expetuntat v. 23 and indnct« at v. 26.

XCVIII
The speakerbids someonewhom he esteemseither to come towards him (c£
Plautus, Mil. 828 proade hue,Terence, Eun. 470 procedetu hue,Seneca, Tro.
705 huee latebrisprocedetuis}or to move away (c£ Plautus, Capt.954 agetu
illueprocede);Ennius could be adapting either I.A. 1-2 &">,rprof}v 66µCAW
ToovSe,rapo18a, Iontxe (Vahlen)or 139-40 &XA'18'l~aaCA>V aov ir66a,
yf\pq;I µT}Sh, vm(KCA>V(Scaliger).The 1TlaTaT1')S of Agamemnon's servant
is a constant theme of Euripides' play (vv. 45, 114, 304, 867).

193-4 gradam proferre pedam,I nitere, cesaas: for the imperative in

parenthesisc£ Aeschylus,Theb.435 Tote;,&cpc,.YT{, iriµm, Tfs~O'TI)aeTCXI;,
Choe.779 &yyeXA'loOaa,irpaaae, TarreoTCXAµa,a, Ennius,Ann. 201 Jono,
dudte,Joque,Plautus, Mere.111-12 ex summisopibusuiribusqueusqueexperire,
nitere,I erus ut minoroperatua seruetur,Catullus 14.21-2 uos hinc interea,
ualete,abiteI illucundemalumpedemattulistiset al. Some modem scholarsput
a stop after nitereand make the infinitiveprofme its object-a construction
absent from Caesar and Cicero but common enough in the historians and
dactylic poets and therefore possiblein archaic tragedy.
Plautus, Men. 754 has gradumproferamin bacchiacs.For the periphrasis
gradum... pedum c£ Euripides, Tr. 333-4 iro6oov I ~povaa cp1ATCXTav
j3aaiv, Lucretius 5 .914 trans mariaaltapedumnisusut ponereposset.1 Verrius
must have understoodgradumas an internal object of procede.

194 o ii.de:the doctrine reported at Serviusauct. Aen. 1. 113, quidamuelint

.fidumamicum,.fidelem seruumdid, appears to be refuted by this verse, v. 237,
Ritschl'salmost certain restoration of Plautus, Most.785, and Livy 33. 28. 13.
But it is clear from the comparative rarity of.fidusin comedy (5 occurrences
as against24 of.fidelis)and classicalprose and from the contexts of occurrence
that it conveyed much more emotion than.fidelisand was more likely to
appear in the dialogue of socialequals.Applied to a person of servilestatusit
perhaps indicated an out of the ordinary affectionon the part of the speaker.
Bergk's supplement o.fide(senex) is quite unnecessary;the absenceofjidus as
a substantive from our record of early Latin loses significancewhen one
considershow rarely even the adjective occurs.
1
No real parallel is provided by Pollio ap. Gell. 10.26.4 transgressus
a
transgrediendo
dicitur,idqueipsumab ingressuet a pedumgraduappellatum.

332
 IPHIGBNIA

XCIX
The structure of the soldien' argument is dear despite several corrupt words:
a general statement about the employment of otiumis followed by a reference
to the particular dramatic situation. This structure has many parallels in
republican drama: where stichic iambic trimeters are usedat Plautus, Poer,.
627 ff., Pseud.767 ff.; where iambic tetrameters are used at Plautus, Rud.
290 ff.; where trochaic tetrameters are used at Pacuvius, Trag.366 ff.,Plautus,
Bacch. 540 ff., Poer,. 504 ff., Terence, Ad. 855 ff., Eun. 232 ff.; where dif-
ferent types of verse are mingled at Plautus, Amph. 633 ff., Epid. 166 ff.,
Men. 571 ff., Caecilius, Com. 142 ff. One cannot decidewhat metre Ennius
gave his soldien' words on a priorigrounds. It is, however, significant that no
such argumentative structure beginsin the middle of a metrical unit.
Three kinds of measurement have been tried: one in stichic trochaic
tetrameten, 1 one in corresponding lyric strophes,:aand one in a mixture of
trochaic lengths. 3 The third kind requires least alteration to the words
transmitted.
The first thirteenwords transmitted form a trochaic dimeter and a catalectic
trochaic tetrameter without diaeresisbut with caesura afterthe fourth arsis.4
The extended word play, otio. .. negoti. .. negotium. .. , hasfew analogues in
Attic drama (c£ however Philemon, fr. 23 .3-4 6 Ao16opoov yap, &, 6
Ao16opovµEVOS I µ'I'}irpocnro1fiToo,Ao16opetTCX1 Ao16op&>V) but seems to
have been considered a stylistic ornament in early second century Rome
(c£ Ennius, Sal. sS>-62 nam qui lepidepostulatalterumJ,ustrariI quemJ,ustratur
J,ustra eum dicit.frustraesse.I tutm qui sesef,ustrari quemf,ustra sentit, I qui
.frustraturisJ,ustraest si non ille est.frustra,Plautus, Amph. 33-6, Capt. 255-6,
Pseud. 704-5, Terence, Andr. 251HJ). The interpretation of negotiumin
negotiohasgiven difficulty.Vahlen offered the translation 'sehr viel Arbeit'.
E. H. Warmington 'when he is awork at work'.S There is no parallel how-
ever for this particular kind of expression in republican drama although it is
common in Attic (c£ Euripides, I. T. 197 q,6vashrl cp6v~,&xea &xe<nv
et al.) and appean occasionally in classical Latin dactylic poetry (c£ Ovid,
Ars 1 .244 et Venusin uinis ignisin ignefait). It would be better to translate
the thirteen words as 'the man who has no job to do and does not know

1
Cf., most recently, 0. Skutsch, RhM XCVI (1953), 193 ff.
a Fir$t by Ribbcck, most recently by 0. Crusius, Die Responsion, pp. 114 ff.
(cf. Strzelecki. in TragiC4I, 58).
3 Cf. Vahlen, Hermesxv (1880), 262.
4 For this type of tetrameter, which is comparatively rare, sec W. Meyer,

Abh. &yer. Ak. XVII {1886), 75 ff.

S R.tnu,ins of Old Lalin I (London, 1935), p. 309.

333
 COMMENTARY
how to employ the resulting leisure has more difficulty than when there is
difficulty in a job on hand'. For the two meanings of negotiumc£ Donatus,
Ter. Amir.2 negotiummodopro molestiaet cura,non lahore.For the switch of
meaning in polyptoton c£ above on v. 10.s (adjectives) and Plautus, Boah.
323 uerumuerumnescio,Capt. 741 post mortemin mortenihil est q,u,Jmetuam
mali, Epid. II 3 in re dubiare iuuat, Mil. 4 praestringatoculorumadem in ade
hostibus,Pseud.90 antetenebrastenebras persequi,Terence, Haut. 41 meacausa
causamhanciustamesseanimumindudte,Accius, Trag.109 malaauxerein malis,
422 nequeuita ullipropriain uita est.
Three other treatments of the thirteen words are worth mentioning.
Bothe wrote otioqui nesdt utieras the end of a catalectic tetrameter. There is
however a strong presumption that Ennius' discourse beginsa metrical unit
and Bothe' s reading assumes a tetrameter without either diaeresis or caesura
afterthe fourth arsis.Timpanaro suggested 1 that otioquinesdtuti should either
be kept as the second half of a full tetrameter or changedto qui uti nesdtotio.
This has an order of words more modern in type than the one transmitted
(c£ Plautus, Poen.210 negotisibiquiuoletuimparare,627 uiamquinesdt)and it
would be difficult to explain the corruption. Taking up a suggestion of A. M.
Dale, 0. Skutsch supplemented the thirteen words as follows: otioqui nescit
uti ( quomotioest, in otio) Iplus negotihabetquam,quomest negotium,in negotio.
This is free from metrical anomalies and the mode of expression attributed
to Ennius perhaps no more twisted than that transmittedat Sat. 59-62.
The two word-groups, cui quod agat institutumest in illis negotiumand
otiosoinitioanimus,are defective in sense and metre. Clearly one refers to the
busy and the other to the idle and ahnost certainly the original rhythm was
trochaic. No convincing restoration has been offered.i
Gellius introduces his quotation with the words in eius tragoediae choro
inscriptosessehos uersuslegimus.Similarlanguage is to be found at Varro,
Ling. 6.94, quareuna origineinliciet inlicis,quod in choroProserpinaeest, et
pellexit, and Gloss.Lat. 1 128, apudRomanosquoquePlautuscomoediaechoros
exemploGraecoruminseruit.Varro could be referring to a tragedy Proserpina;
the glossator' s source must have had in mind the fishermen of the Rudens
(290 ff.) or the aduocatiof the Poenulus(504 ff.). Ennius' 'chorus' was com-
posed of soldiers and of such there is no trace in the extant version of
Euripides' I.A. 3 The Attic odeswere sung and dancedby women of Chalcis
come out of mere curiosity to look at the army assembling at Aulis. Their
content was to a quite unusual degree unconnected with the action of the

1
SIFC N.S. XXI (1946), 76 f.
a 0. Skutsch suggested cuiquodagatinstitutumin otioestnegotiumand otiosoin
otio(aeger)animus.
3 Verses 589 ff. have been, however, so interpreted. See above, p. 318.

334
 IPHIGENIA

I.A. There is no signthat Ennius adapted the text of any of them. Students o
Roman tragedy seem all to have assumed that Ennius replaced them with
utterances of the troop of soldiers. The only matter of dispute is the play
from which Ennius extracted his soldier chorus.
There is no need to assume that Ennius made such radical alterations to the
I.A. The soldiers, like the attendants of the Euripidean Hippolytus (desig-
nated xopos by our manuscripts), 1 may have appeared on the stage only
once, accompanying Achilles to the tent of Agamemnon (I.A. 801-1035). In
any caseonly the soldiers of Agamemnon could have been present through-
out the action of the play and the utterance of the Ennian soldiers marks
them plainly as Achilles' Myrmidons; the boredom and impatience of these
were proverbial: c£ Homer, n. 2.778-9 .. .ol s· &pxov &f)TltcptAOV
iroetovns I cpoiTCA>V Ma Kai Ma KaTa OTpaTOVov6e µaxovro,
Euripides,I.A. 812-18yf\v yap i\1,roovCl>apa<XAOV fi6eTl{}AEa I µivoo',rl
Arn'Tals Tatai6' Evp{,rov irvoats, I Mvpµ166vas taxCA>V. ol s· ml
1Tpoo,<E{µevo1 I i\fyova·· 'AxtAAEV, Ti µa,oµa,; ,roaov xp6vov I tr·
IKµE"Tpfiaoo XP~irpos 'li\{ov OT6i\ov; I Spay•, et TI 6paaus, i') mray•
o1Ka6eOTpaT6v, I Ta TOOV 'ATpElOO>V µ~ µa,oov µei\Af}µa,-a,Tekphos
in Pap. Berol 9908n 13-24 T{~·; ov XPTlV fiavxov l<Elo6at,r66a ...
ate{ irar· trn vooxei\etsKai ~ I j?>{}ae1s e· IKaOTosµvp{as Kae{}-
µa,os I Mye1,TOs· lpyov [ov)Saµov ,ropewrat. I K[&y)oo~ ~
6pci[T]e 6pav ho1µos ~v I fi[K]oo,OTpaToS -re M(vp]µ16oov,Kai
1ri\eva[oµa1] I [Ta T)oov •ATpe16a:[vov µa,oov] µei\Af}µa,-a,Accius,
Trag.611-12 iam iam stupidoThessalasomnoI pectoralanguentque senentque.
Terence admits himselfto importing the persons of the parasite and the
soldier into his adaptation of Menander' s EwovxoS(19 ff.) and to adding a
whole scene to the •A6ei\cpo{ (6 ff.). Donatus reports that he imported the
persons of the freedmanSosia, the lover Charinus and the slave Byrria to the
'Av6p{a (advv.14, 301;c£ Ter. AtUlr.9 ff.)and that of the young man Anti-
pho to the EvvooXoS (adv. 539). Terence never speaks of these alterations to
the dramatic structure of Menander's plays as coming from his own creative
imagination; he always claims the authority of a scene in another comedy. It
therefore makes some sense to ask what tragic scene Ennius would have
claimed as authority for the soldiers he imported into his adaptation of the
I.A. Scholars in the past have asked a slightly different question but their
answers could in principle be adapted to this new one.
Bergk 2 tied otioquinescitutiplus negotihabetquamcumestnegotiumin negotio
with a sententiaquoted by Stobaeus (30. 6) from Sophocles' 'lcp1ywe1a-

1
61; cf. Aeschylus, Bum. 1032, Euripides, Phaethonfr. 781 .14.
a Ind. lectt.Marburg1844, XIV( = Kl. phil. Schr.1229); cf. De Frag.Sophoclis,
p. IS, Welcker, Die griech. Trag. p. 110.

335
 COMMENTARY
TIK'Tllyap ov&v ioe).ov
EIKafaaxoAil
(fr. 287), anddeduced
that Ennius
replaced the odes of Euripides' play with versions of those of Sophocles'
besidesmaking other alterations.
The tie is not a closeone. The Sophocleansententiawould be more likely
uttered by a generalthan men of lower rank; it breathes the same spirit as
Euripides, I.A. 1000-1 cnpcrrbsyap a6p6os, apybs ~v Tc.>V oiKO&v,I
Mox~1TOV1l~ Kai KaKOCTT6µovs q,wt. 1 It concerns the man who has
acquired leisure through some chance rather than from his own deliberate
effort and the sociallydeleteriouseffectof such leisure.The Ennian sententia
on the other hand concernsthe man who does not know how to manage the
leisure that comes to him and the personaleffect of this ignorance. Even if
the tic were a closeone it would not show any structurallink betweenEnnius'
play and Sophocles'.Ennius at leastonce, perhapstwice (&s.cv, CVIII ). makes
his Medeautter sententiae quite absent from his Euripidean original without
apparently making :my structural alteration to that original.
Where sententiae arc concernedin Roman tragedy and comedy it may be a
mistakealwaysto look for a dramatic source.At least twice (&s.LXXXIV, cv)
Ennius can be shown altering a sententiaof his original in terms of his own
experienceof Roman life. Here he may be composing with the arguments of
contemporary Greekphilosophicalschools2 in mind rather than any dramatic
context. Such detailed psychologicalanalysis is scarcelyappropriate in the
mouths of soldiers.The mental effectsdescribed-animusnescitquiduelit. ..
inartemat animus,praeterpropteruitamuiuitur--arcthose sufferedby Demo-
aitus' avc>Jlµ<>VES (fr. 202 T<A>Vcrrre6vrCA>V
6piyovrat, Ta & napE6VTa ...
aµCXA6vvovcn), by the love-sick Phaedra(Euripides,Hipp. 181-s) and by
those ignorant of the Epicurean science of living and dying (Lucretius
3.1057~).
I.eol suggestedthat Ennius' soldierswere thechorusof Euripides'T{)~.
His only argument was that some anapaests quoted by Priscian (Gramm.
2.512.21) from Accius' Teltphus-iam iam stupidoThtssala somnopectora
langumtqutsenentqut-prove Euripidesto have employed a chorus of soldiers.
They prove nothing of the sort, even if it were certain.as it is not, that
1
0. Skutsch suggested that Sophocles' trimctcr refers to the particular
situation at Aulia and takes up a generalisationabout 1cisurcmade in an earlier
choral ode.
a On philosophical discussion of axo"Ari/otiumsec F. Boll, 'Vita contem-
platiua ', SB Htidelberg,Phil.-hist.Kl. Abh. vm, 1920 ( = Kl. Sehr. 303 ff.), W.
Jaeger. 'Ober Unprung und Kreislauf des philosophischen Lebensidcals•, SB
Bnlin,Phil.-hist. Kl. 1928,390ff. ( = ScriptaMin.1 3-47ff.),J. L. Stocks,CQxxx
(1936), 177-87, Fraenkel,Horace (Oxford, 1957), pp. 212-13, J.-M. Andre,
Reeherehessur l'otium ronuiin(Paris, 1962), pp. 27 ff.
3 De Trag. Rom. p. 15 ( = Ausg. kl. Sehr. 120-4), Geseh.p. 192 n. 3.

336
 IPHIGBNIA
Acciusadapted Euripides.If soldiersdid form the chorus of the Euripulean
Ti)AEcposthey would have been Agamemnon's Argives, not Achilles'
Myrmidons.One could imaginea sub-chorusof Myrmidonsaccompanying
Achilles to Agamemnon's palac:c.However the fu:t that in the Berlin
papyrusAchille-himselfdescribestheMyrmidondiscontentseemsto ruleout
this possibility.We shallhave to look outsidetheTi)~ for the origin of
Ennius' soldiersbut Leo'smethod oflooking at the dramatic substanceof the
trochaicversesrather than their generalargumentremainsthe one to follow.
Mariotti' seessomethingof an' andaturastilisticasommessae convenativa•
in the discourseof Ennius' soldiers,'persone di categoriainferiore•. It is true
that their manner of speaking does not possessthe finished dcgance of a
Ciceronianoration or philosophicaldialogueand that their vocabularycon-
tains no elementthat we should thinkabsentfrom the common languageof
the early second century. But one could say this of probably most of the
sententiouspassagesof republican tragedy and comedy. Peculiarlypoetic
vocabularymassesin thenarrativeand descriptivepassagesof the spcccbcsof
all socialcategoriesin tragedy. In theelaborationof phrasalpatternsthe dis-
course of Ennius' soldiersmarks itselfoff even from the most ornate of the
paratragicmonodiesof Plautinecomedy; one notes theforward positionof
verbs and relative clause,the frequency of impersonalverbs, the asyndeta.
the personificationof animus,the extendedword play with otio. .. negoti. ..
negotium. .. negotio,the polyptota negotiumin negotioand ad (quoi)'J."04, the
ascendingtricolon with anaphora id agit id studetibi menttm atqut animum
delectatsuum,the ascendingdoublet menttmatqut animum,the etymological
figureuitamuiuitur;all of thesephenomenaoccur in comedy but nowhere in
such massas here.

198 ibi mentem •tque animam delectat1UU1D:for ibi St deltctdtor ibi

delectatur.
The doublet occurs in comedy only at Plautus, Trin. 454 satin tu
sanusmentisaut animitui and is rare in classicalprose. Dactylic verse has it
(Lucretius 1. 74, 3 . 142, Horace, Epist. 1. 14. 8, Virgil, .Atn. 6. II et al.)
perhaps partly on the analogy of Homer's 1<p<XS{f1V Kai 8vµov IKCXVEV
(n. 2. 171). K<X'Ta
cppwa Kal KCXTCX 8vµ6v (4. 163), TOlov& VOOV Kai 8vµov
(4.309).
199 totioso initiot anima,: Lipsius'otiosoin otioanimuscan be taken in
two ways: 'the mindin idle idleness•or 'the mind of a man idle at a time of
idleness'; for the first c£ Cicero, Plane.66 cuifaeritne otiumtfUitkmumquam
otiosum(Nacvius,Trag.35acremaaimoniam,Plautus,Aul. 215 """4 .. •nuJitia,

' Laioni suE.nnio,p. 119. Cf. G. Luck, Obatinigelnlerjdttionentin lattinisdtoa

Umgangsspracht
(Heidelberg, 1964), p. 52.

H 337 JTO
 COMMENTARY

Capt.774 amoenitate amoena,Cas.217 nitoribusnitidis,Epid. 120 pretiopretioso,

Mil. 959 pulchrampulchritudintm,Poen. 134 gratasgratias,Pseud.882 suaui
suauitate,Lucretius 1. 826sonitu... sotuJnti,
3 .993 anxiusangor);for the second
cf. Cicero, Cael. 1 quibusotiosisne in communiquidemotioliceatesse{Plautus,
Cure. 533 iratusiracundia, Mere.191 negotiosieramusnos nostrisnegotiis,844
l«tus laetitia
).

anim111 nescit quid uelit: cf. Lucretius 3. 772 quidueforas sibi uult
membrisexiresmectis,1058 quidsibiquisqueuelitnesdre,Livy 41.20.4 nescire
quid sibi uelletquibusdamuideri.The omission of the pleonastic rdlexive
pronoun perhaps raises the levelof the style.

200 em neque domi nunc nos nee militiae 1111D111:em is rare outside
republican drama; it precedes statements at Plautus, Asin. 336 em ergois
argentumhueremisit,Baah.274 em aaipitrituJhaeenuncerit.Here it seems to
indicate an exasperated tone of voice. In a tragedy of the classicalperiod it
would be an extreme colloquialism; one cannot be sure about its exact status
in early second century Latin.
Domi. .. militiaeis officialphraseology; cf. Terence, Ad. 495-6 UtuJsemper
militiaeet domi Ifuimus, Cicero, Manil.48, Mu,. 87, Pis. 1, Lig. 21 et al

202 incerte errat anim.111: for animiincertierramus.The Attic tragedians

commonly apply &A5:a6oo,'h'A<X3Ea6oo, 1TACXVO:a6at, q,otTavto mental
behaviour. For the Latin phraseology cf. Ennius, Trag.215-16 eraerrans...
animoaegro,ap. Cic. Tuse.3. 5 (fr. CLXXIV) animusaegersempermat, Plautus,
Mere.347 tantuscumcurameosterroranimo,Pacuvius, Trag.302 triplicipertime-
factus maeroreanimiineerteerransuagat,Lucretius 3 . 463-4 morbisin eorporis
auiusmat I saepeanimus,3 . 1052 animi incertojluitans more uagaris,Cicero,
Carm.fr. 47. 1 igtuJris
hominesin uitamentibuserrant.

praeterpropter uitam uiuitur: the opposite of reeteuiuimus(cf.Horace,

Epist.2.2.213); the philosophical notion of living as an exact skill,like,for
example, carpentering, is implied.
Salmasius changed uitam to uita. A scribe could have mistaken praeter
propterfor a preposition. On the other hand the accusativedependent on the
impersonal uiuiturhascompanions in republican dramawhich are difficultto
explain away: Plautus, Cas. 185 pessumisme modisdespieatur domi,Mil. 24
epityraesturinsanumbene,254 inducamusueraut essecredatquaementibitur,
Pseud.817 teritursinapisscelera,1261 ubi mammiamammicula
opprimitur.The
rarity of the construction may simply reflect the rarity of the impersonal
passivein comedy (uiuituroccurs only at Plautus, Persa17, Trin.65, Terence,
Haut. 154).
 IPHIGBNIA

For the etymological figure cf. Plautus, Bpid. 386-7 cogit4rentpostto I

uitamut uixissmtolimin adulescentia,
Mere.473, Mil. 628, 726, Persa346, 49-4.
Turpilius, Com. 143, Cicero, Vm. 2. 118 et al

C
As often in rhetorical treatiscs the illustration refen not only to the text of
1

the dramatic speech but also to the mode of utterance; the words quoted
scarcely of themselvesillustrate ayavaJ<'TT\OlS. The Agamemnon of Euri-
pides' I.A. is at his angriest in the sceneof his quarrel withMenelaus. Accord-
ingly scholarssince Columna's day have wanted to put the Latin words in an
adaptation of this scene, although there are no obvious verbal correspon-
dences.
The words Mene1"usme obiurgatwould be spoken more naturally about
Menelaus in his absencethan to his face directly.2 Regimenshould be inter-
preted as 'my command of the Greek armada' rather than 'Menelaus' domi-
neering behaviour' (c£ Trag. inc. 29 proindemetabste regimenArgosdumest
potemuconsili).The Latin Agamemnon is describingthe opposition that has
arisen betweenhis officeand the welfare of hisfamily. AccordinglyI propose
that Ennius' trochaic tetrameter be treated as coming from an adaptation of
the scenepreceding that of the confrontation of the two brothen. Some sort
of stimulus can be seen in the iambics 84~ Ka~OTp<X'Tilyttvtm"ra
xap1vt I eO.ovrocniyyov6vye. Ta~fooµa6EI ~OS TIS~•
MevO.ec,.,
avr·!µoO A~V T66e. The anapaests2.2-3 Kai TO,rp6nµov (Nauck::
codd.) IyAVKV
cp1A6T1µov µa,,A\11TI:l
si ,rpoa10Taµevovrefer in general
to the troubles that positions of worldly honour can bring.
My interpretation implies that the Latin Agamemnon put up a much
stronger resistance to the proposed sacrifice than either the Euripidean
Agamemnon claimed (97-8 ov Sriµ"a&Acpos,rcxvra,rpoacp~v A6yovI
rne1aeTAf\vcn 6e1va)or the Euripidean Menelaus allowed {36o&aµa,os
8vae1vV'Trio-rr}s,rat6a). Recognition of the force of circumstancesrather
than base ambition would have motivated his decision to acquiesce.The
Euripidean Agamemnon was a most contemptible and unheroic figure.
Ennius, if my interpretation is right, gave him a character more in harmony
with the one which the Athenian philosophical schools and Roman aristo-
cratic tradition thought appropriate to the statesman. Comparable are the
alterationsmade by Pacuvius to the character of Sophocles' Ulysses(Cicero,
Tusc.2.48) and by Terence to Menander's Micio (Donatus, Ad. 938).l
1
Cf. Charisius, p. 364.21 (Trag. inc. 123-4), 374.8 {Trag.inc. 120-2).
2
They could be an angry aside; cf. Plautus, Cure.572 f.
3 For the 'Stoicising' of the heroes in epic poetry sec Norden, Aen. Vl 1 ,
p. 154.
339
 COMMENTARY
203 MeaeJaaame obiurgat; id meia rebaaregimen rest:itat:Bentley's
simple alteration produces a regular trochaic tetrameter. The fact that the
intensive restitare is rare in archaic Latin compared with the positive and even
rarer in classical Latin stands probably to the favour of restitathere; the same
is true of many recorded intensives in -itare.For the predilection of tragedy
for intensive forms seeabove on v. 68.
According to Fcstus, p. 348 .15: llBGIMBNproregimento usurpant poet«. The
extant texts do not seem to have regimentum;regimenoccurs only here in
republican drama, not at all in Caesar, Cicero and Sallust, five times in I.ivy.
For the predilection of tragedy for nominal formations in -men see above on
v. 13.
Restitattakesthe place of expected obstat,obest(c£ Lucretius 1. 110 ratio
nuIlaestrestlltuli,2 • 4soaeraqut quaeclaustrisrestdntiauociferantur
), perhapsfor
the sakeof the triple alliteration at the end of the verse.

Cl
In the Euripidean Agamemnon's attacks on his brother (I.A. 317-412) there
is nothing as rhetorically figured as these three trochaic tetrameten; the
nearest to them in form and substance are verses of Mcnelaus' speech of
TS<VOVj 1.n'}T'
reconciliation: 481-4 Kai ao1 napcnvc1> 1.u'lT'o:rrOKTE(ven,
~ TOVl,K)V. 0V yap I en
lv611(0V TaµCX Ii~
l,l!vC1"fEVCXlE1v, s·
I
fxaiv, 8v(i01<&1v TET0VS aovs, s· iµovs
T0VS 6pav cpaos.
204 ego proiector quod tu peccu: usually interpreted' as a syntactic
graecism: i.e. imTIµc;jµcn, •mihi proicitur quod ... '. Timpanaro 1 pointed out
that proido/proiecto is unexampled in the meaning normal to obicio/obiecto and
offered the quite satisfactory interpretatio~ •io sono messoinnanzi come capro
espiatorio, sono esposto al biasimo pubblico a causadella tua colpa '. For the
type of argument c£ Plautus, Epid. 139-40 men piacularemoportetfieri ob
stultitiamtuam, I ut meumtergumtuae stultitiaesubdassuccidaneum? Ribbeck
punctuated this and the next correctly as positive statements; the future
indicative sometimes occurs in repudiating questions (e.g. Plautus, Men. 198
egosaltabo?),the present never, as far as I can see.

tu tcfeJioquut ego arguor: Stephanus' Jelinquis produces a metrical

anomaly (seeabove on fr. xn) but quite good sense.0. Skutsch suggests3

1
Cf. R. Frobenius, Die Syntax des Ennius (Diss. Tiibingen, 1910), p. 81,
Kroll, Studien, p. 249, Lofstcdt, SyntacticaII, p. 412, J. B. Hofmann and A.
Szantyr, 1.Ateinische
Syntax und Stilistilt (Munich, 1965), p. 33.
:a Maia m (1950), 28. He would now, however, interpret (letter to 0.
Skutsch) 'I am being sacrificed in your interest' (cf. Catullus 64.81-2).
3 London seminar, 29 November 1954; see now HSCPb IJCCI (1967), 133 ff.

340
 IPHIGENIA

that arguoris a displaced gloss on proiectorand replacesit with lumn (c£

Horace, Ca,m. 3 . 6. 1 delictamaiorumimmerituslues).

206 reconcilietur: only once in comedy (Plautus, Capt. 33; reconciliassere

occurs at Capt. 168, 576) as against conciliare14 times.

necetar: a word normally applied to criminal homicide (sec above on

v. 178) and the extermination of unwanted children. slaves, criminalsand
prodigia(sec above on v. 18). The accepted notion that it refers at base to
bl)jng without bloodshed 1 appears to be contradicted by this passage and
Pacuviw, Trag. 329. Both passages have a highly emotional tone and the
speakers may be emphasising the word's brutal associations rather than its
literal meaning.
CII
Ribbeck set these words in an adaptation of the lament uttered by the
Euripidean Iphigenia after the rejection of her plea to Agamemnon (vv.
127!r-33S), comparing in particular her apostrophe to Helen(vv. 1333-s lw
loo,I µey@.a ira6ea, µey@.a s• &xea I ~avat6ms TI&laa Tvv6apls
KOpa). V ahlen set them in an adaptation of the dialogue between Agamem-
non and Clytacmnestra about the proposed marriage (vv.691-741 ), compar-
ing in particular vv. 73S-7 (-ov KaAOVa, 6XA(t>
a• ~oµW:la6<X1 OTpcrroO.
- KaAOVTEKOVO'av
TQµCX y• a, olKct>µft
µ• iK6ow<Xl-riKva. - Kal TCXS
µ6vas eiva1Kopas.
The attempt to find verbal parallels between Greek and Latin, here as so
often elsewhere, falsifies interpretation of the Latin. In a play about the
Trojan expedition one would naturally interpret the words qu« nunc ahste
uiduae et uastaeuirginessunt as part of a complaint uttered by a woman
against the commander: because of his actions the young women of Greece
have beenleft without suitors and the means of bearingchildren. The plight
of the women of Greece during the siege of Troy is a constant theme of the
speeches and odesof tragedy: c£ Euripides, Hele.322-s, Hel. 1125, Or. 1134-
6, Tr. 37!>-81. The Euripidean Clytaemnestra makes no such complaint in
her attack upon Agamemnon (vv. 1146-208} but it would not be out of
place in a free Latin version.

WJ uiduae et uastae uirgines: the uiduaand the uirgoare normally dif-

ferentiated (e.g. Plautw, Cure. 37-8 dum ted abstineasnupta uidiuiuirgine•••

1
Cf. Fcstus, p. 158. 17 NliCI DATVS propriedicitur,quisineuulnereinterfectus
est,
ut uenenoautJame, 190. s OCCISVM a necatodistinguiquidam,quodaltaum a cae-
dmdo ~ ictu Jim ditunt, alterum sine ictu, Priscian, Gramm. n 470 .14 f.,
WolfBin, ALL vn (1892), 278.

341
 COMMENTARY

amaquiJlubet)but cf. Seneca, Ag. 195 uirginesuiduae,Apuleius, Met. 4.32

Psycheuirgouiduadomiresidens. For the tautology cf. Aeschylus, Pers.288-9
oos 1TOAAQS Tiepa{6CA>VI 00-.0'CXV ewt6as fi6' avav6povs.Sophocles. 0. T.
1502 XEf>O'OVS ••• 1<ayo:µovs, Euripides, Hipp. 547 &vcxv6pov To nplv 1<al
awµcpov. Cato, Agr. 141 {prayer to Mars) uti tu morbosuisosinuisosque.
uiduertatem uastitudinemque,calamitates intemperiasque
prohibessisdefendasauer-
runcesque,Plautus, Pseud.~ haruncuoluptatum mi omniumatqueibidemtibiI
distraaiodiscidiumuastitiesuenit.

MEDEA EXVL; MEDEA

The title Medeaexul is given to Ennius three times by Nonius (frs.axe, CXIa
and b)1 and once by a Virgilian commentator, the so-called Probus (fr. cx).i
The Latin of the three pieces quoted is quite close to the Greek of three
passages of Euripides' extant MT)&aa.3The other titles given to republican
tragedians which consist of proper name and adjective may all be treated as
transliterations or translations of the titles of the Attic originals. 4 Such titles
were applied to Attic plays by grammarians wishing to distinguish two plays
about the one hero or heroine by the one author. Among the seventy
tragedies of Euripides known to the scholan of Alexandria there was one
apart from the extant Mt't6e1awhich dealt with the Colchian woman. But
this is regularly quoted in our sources as Alyevs. We should therefore sup-
pose that Medeaexul was a title applied by grammarians to an adaptation of
Euripides' MT)&aaso as to distinguish this script from another by Ennius
about Meclea.S

1
Pp. 39.2, .261.21, 292.20 probably depend on Lindsay's list 27 'Alph.
Verb'.
:a Probus' work exists only in humanist apographs of a lost medieval codex.
Certain things in it could not possibly come from the mouth or pen of M.
Valerius Probus Berytius. Whether or not the core of the work belongs to him
is an insoluble problem; see Marx, C. Luciliicanninum reliquiae,vol. I (Leipzig,
1904), p. lxxii, J. Aistermann, De M. ValeriiProbiBerytii uita et scriptis(Diss.
Bonn, 1910), pp. 72 ff., Norden, Enniusund Vergilius,p. 10.
l Fr. CXI was compared with Euripides vv. 49-51 by Stephanus; fr. CIXwith
Euripides vv. 250-1 by Muretus (Var. Lectt. IV 7); fr. ex with Euripides vv.
1251-4 by Scaliger (Coniect.Varr. Ling., on 7 .9).
4 See Introduction, p. 58.

s Hyginus, Fab. 25 gives the w68Ea1s of Euripides' Mri6e1aunder the title

Medea;Fab. 26 an account of the birth of Medus to Aegew and the exile of
Medea from Athens for witchcraft under the title Medeaexul; the latter could
be but does not look like the w6&<ns of a tragedy.

342
 MEDEA EXVL; MEDEA
There is further and stronger evidence both that Ennius adapted the
MT16E1a of Euripides and that he wrote another tragedy about Medca. This
second tragedy was set in Athens. There is nothing either for or against the
supposition that it adapted the Alyevs of Euripides.
The title Medeais given to Ennius by Cicero (fr. cmd), Varro (fr. cxxna) 1
and Donatus (fr. cmm)once each, by Prisciantwice (frs.cmo, cmp )2 and by
Nonius five times (fr. cmb, cxm, oav, cxv, cxvi);l to Accius by Nonius
sixteen times 4 and Priscian once ;S to Pacuvius in error twice by Macrobius
and once by a Virgilian scholiast. 6 Of the seven pieces quoted with Ennius'
name one (fr. cm) is said by Cicero (Fin. 1.4) to come from an adaptation of
Euripides, and is used by Latin writers on logic and rhetoric to illustrate
a fallacious type of argument (Rhet. inc. Her. 2.34, Cicero, Inu. 1.91,
Top.61, Fat. 34, Q_uintilian, Inst. 5.10.84, Iulius Victor, Rnet. 12) where
Greek writers use the prologue of Euripides' Mft&1a (Clemens Alex.
Strom. 8. 9. 27 ), and in fact follows fairly closely the Greek of Euripides

' The fact that the trimeter quoted by Varro at Ling. 7 .9 docs not form a
sense unit and its mode of introduction {contrasting with the one discussed
above on fr. XI) show that it comes from another grammarian's collection. In
the seventh book Varro refen frequently to Aurelius Opiliw, Serviw Clodius
and 'glosscmatum scriptores '. Reitzcnstein, DabJroaoo and Schroter have
shown {sec Introduction, p. 53 n. 5) how a mass of undigested glossographical
material was jammed into the schematic framework of this book.
:a The quotation at Gramm.D 320. 15 occurs in an article of the sort which
goes back at least to Flavius Caper and perhaps even as far as the source of
Varro, Ling. 1. 33. That at Gramm.m 423. 35 probably comes from an earlier
writer on metric {c£ 418. 10). The view of Marx {Ind. ltctt. Grtifswald 1891,
10 ff.) and Jeep(PhilologusLXVID [1909], 12 and 16 n. 27) that Priscian chew it
from Rhet. inc. Her. 2. 34 is implausible. The quotations of Ennius, Accius
and Turpilius in the treatise on Terence's metres are all lengthy and come from
the beginningsof plays as do those of Plautus and Terence. Accius' plays like
Terence's seem to have been chosen in alphabetical order. Most quite certainly
could not have entered Priscian's treatise via a rhetorician and we should not
treat the Ennius quotation as any different from its companions.
3 The source in all cases is probably Lindsay's list 27 'Alph. Verb'.
4 Pp. 12.6, 16.9, 16.13, 89.5, 90.5, 159.13, 179.25, 235.1 (?), 237.43,

307 .21, 323. 12, 362. S, 422. 30, 467. 37 and 504. 10 from Lindsay's list S
'Accius i'; p. 467.21 from list 27 'Alph. Verb'.
s Gramm.n 336. 18.
6 The error at Sat. 6. 1 • 36 could come from Macrobius misreadinghis source

(on which sec CQ N.S. XIV (1964], 289ff.). That at Sat. 3.8.7 is shared by
Servius auct. Am. II. S43 and must come from Donatus. Varro gives the cor-
rect title (drawing upon the ultimate common glossographical source) at Ling.
7.34-

343
 COMMENTARY
vv. 1-8. 1 Petrus Victorius joined two together to form (fr. cxn) what are
clearly the opening verses of a tragedy set outside the walls of heroic Athens.2
The remaining four pieces have no close verbal parallels in the Greek of
Euripides' M{}Sua but it could be argued that they have a place in an
adaptation of this play no freer than that of the 'EKcx~.3
Varro's quotation at Ling. 7 .9 (fr. cxna) is the plainest of the signs point-
ing to the existence of two plays by Ennius about Medea. Some have tried to
push the words astaatqueAthenas. .. contemplainto adaptations of speeches
and odesof the Euripidean M{}Seaa: Plank4into the speech made by Medea
afterthe departure of Aegeus, vv. 764-810; Ladewig5 into the ode delivered
by the chorus after Medea' s revelation of her plans of vengeance, vv. 824-65 ;
Pascal6 and N. L. Drabkin7 into the speech in which Aegeus promises
Medea refuge in Athens, vv. 719-30. These four scholars all think. that the
speak.er is talking of Medea' s future domicile in Athens and conjuring up
a picture of it in the present. This is the way in which prophets moved
by the spirit of a god talk, e.g. the Cassandra of Ennius' Alexander.It is
not appropriate to any penonage of the freest adaptation of the M{}Seaa
imaginable.
Vahlen8 argued that Cicero and Varro give no signof knowing a second
play by Ennius about Medea and that Nonius quotes pieces of the adaptation
of the M{}Seaa under the title Medea as well as under the title Medea exul. He
explained away asta atque Athenas. .. contemplawith the hypothesis that
Ennius had combined in one monster tragedy the action of Euripides'

1
This was fint pointed out by A. Politianus, MiscellaneorumCenturiaPrima
(Florence, 14,89),cap. 27.
i Var. Lectt. XIV 16. C£ Scaliger, Coniect. Varr. Ling., on 7 .9. It is possible
that a common glossographical source lies behind Varro, Ling. 7 .9 and
Nonius, p. 470.4 (= list 27 'Alph. Verb'): Varro excerpted the second tri-
meter without worrying about the sense while Nonius, who in excerpting
actual texts liked to quote units of sense,excerpted the words astaatqueAthenas
anticum opulentumoppidum contempla.Victorius made Ennius' original a lost
play by Euripides; P. Elmsley, EuripidisMedea(Oxford, 1818},p.66madeitthe
Alyevs.
3 With fr. cxm Stephanus compared Euripides, vv. 1009-70; with fr. oav
Plank, Q. Ennii Medea, p. 96, compared v. 1258, Ribbeck v. 764, Vahlen v.
752; with fr. cxv Ehnsley, Euripidis Medea, p. 151, compared vv. 431-2,
Vahlen vv. 627-34; with fr. CJCVI Columna compared vv. 131-3, Vahlen vv.
772-3, 0. Skutsch, in Nauicula Chiloniensis:studiophilologaF.Jacoby... oblata
(Leiden, 1956), 112, "· 67.
4 Q. Ennii Medea,pp. 97-8. s Anal. sun. p. 16.
6
RFIC XXVII {1899},3.
7 The Medea Exul of Ennius {New York, 1937), pp. 10 ff. {tentatively).
8
E.P.R. 1 , pp. CCVII-ccvm, 162.
344
 MEDEA EXVL; MEDEA
Mfi&1awith that of his Alyevs.V ahlen's view of thingsisnow theonhodox
one. 1 It will not stand up to a rigorous check.
There is admittedly no sign that Cicero knew of a Met/easet in Athens.
But many plays cited in Nonius' dictionary give no certain sign of their
presence in Cicero's writings (e.g. Ennius' .Andromeda, Erectheus,Hectoris
lytra,Hecuba,Nemea,Phoenix,Telephus).Cicero's knowledge was that of the
cultivated amateur, not the scholar.
Varro, although a piece of the Athenian Met/eaappean at Ling. 7. 9, need
not have known this play at first hand as he did the Corinthian Met/ea.His
mode of quotation (seeabove, p. 343 n. 1) suggests that he took the piece
from an earlier scholar's collection.
We can deduce little about the construction of Nonius' source, list 27
'Alph. Verb'. The introductory words Ennius Met/eacan be treated as an
abbreviation of Ennius Met/eaexule, but equally well of something like
Ennius Met/eaAtheniensi.Extant Greek anthologies, lexica and scholia, by
dropping the distinguishing epithet, frequently leave us in doubt asto which
of Euripides' Melanippe, Alcmeoand Phrixus scripts is being quoted.
The Met/eaexul which V ahlen visualises, in which a large and involved
action reaches its climaxin Corinth and is followed by another action set in
another city at at least several years' remove, is not historically credible. No
surviving contemporary adaptation of comedy exceeds by more than a few
verses the length of fourteen hundred:a and we can reason that second-century
audiences did not allow more licence to tragedy. Even considerable pruning
of the choral odes and the omission of a detachable scene like that involving
Aegeus would not have reduced the length of the Corinthian section of
Ennius' alleged tragedy much below a thousand verses. Aeschylus'
'AyaµtµvC1>v and Evµa,{5Es,Sophocles' Tpax{v1aa,Euripides' 'Av6po-
µax11,•h<nl&s and I&evEix>1a have actions spread over long intervalsof
time. Sophocles' Ai~ and Euripides' •EK&jni have actions which can
scarcely be said to possess unity in the Aristotelian sense. But nothing we
know resembles Vahlen's hypothetised monster; nor is there any good
evidence that the Roman adapters of comedy and tragedy ever employed
such a combinatory technique.

1
It seems to have been accepted with varyingdegreesof enthusiasm by F.
Skutsch (RE v [1905], 2594), Leo (Gesch. pp. 187 ff., 193 n. 1), Terzaghi
(SIFC N.S. VI [1928], 191), Drabkin (The MedeaExul, pp. 10 ff.), Warming-
ton, Klotz and Heurgon. C. Bailey (CR :xvm [190-4], 171) was sceptical.
a The Pseudolushas 1335 verses in our texts; the prologist remarks (v. 2)
Plautinalongafabulain sC4tlUJm uenit, the slave Pseudolus (v. 388) nolo bis iterari;
sat sic longae.fiuntfabulae. The Casinohas 1018 verses; the matron Clcustrata
remarks (v. 1006) hancex longalongioremneftlliamusfabu'4m.

34S
 COMMENTARY
One might try to salvage V ahlen' s general theory by postulating only a
small scene at the end of the Medeaexul showing Medea' s arrival in Athens.
Our texts have what looks like a scene based on Sophocles' •Avny6V11added
at the end of Aeschylus' •Eirra hrl 01'\~- Some of our texts have added to
the end of Terence's Andriaa scene in which Charinus and Philumena are
betrothed. 1 Our texts of Plautus' Captiuiand Poenulusshow obvious traces
of similar interference with Plautus' original versions. The prologue of
Terence's Adelphi admits to the insertion of a scene into the action of
Menander' s comedy by the adapter. However the shift of scene from Corinth
to Athens remains very difficult to accept (seeabove, p. 165).
The Medeaexul was well known outside the ranks of lexicographers. We
therefore expect to find among the unattributed quotations of tragedy made
by Cicero, Varro and the rhetoricians quite a number belonging to it. The
searchfor such quotations has not however always been sufficiently critical.
Cicero knew at least two other tragedies about Medea: Pacuvius' Medus,"
which dealt with the reunion in Colchis between Medea and the son she bore
to Aegeus, and Accius' Medea,3whose scene has been set by one scholar or
another in practically every place mentioned in the stories about Medea. 4
Qgotations of tragedy referring to Medea and accompanied by the name of
Ennius should go to the Medeaexul. Where Ennius is not mentioned too

1
Sec 0. Skutsch, RhM c (1957), S3 ff.
2
Off. 1 . 114, Nat. deor.3 . 48.
3 Nat. deor. 2. 89; one of the trimcters quoted by Cicero appears under
Acdus Medeaat Nonius, p. 90. s; Priscian, Gramm.m 424. 9 quotes six as by
Acdus in Argonautis.H. Keil (in app.) argued that Priscian drew on Cicero and
that the title Argonautaewas a mere error. However Priscian goes on to quote
passages unknown elsewhere from Accius' Persidae,Phoenissaeand Telephusin
that order. All four seem to come from prologues and we should suppose that
they came directly from texts to the source of Priscian's metrical treatise (see
above, p. 343 n. 2 ). An even closer alphabetical arrangement appears in the titles
if we suppose that the first play bore the title ArgonautaeuelMedea(cf. Aeschylus'
C!>p(ryEsi\ •EICTopos >.VTpa,Accius' Stasiastaeuel Tropaeumand Aeneadaeuel
Decius). Nonius would have used one, Priscian the other. Stephanus gave
Accius two separate plays Argonautaeand Medea,arguing (Addenda,p. 428)
that Nonius was in error at p. 90. s. Ribbeck (c£ Manutius on Cic. Fam. 7 .6,
Bothe, RhM v [1837], 259) abolished not only the play but even the title
Argonautae.
4 In Colchis at the end of the heroine's wanderings by Welcker (Die griech.

Trag.pp. 1214-16); in Athens by Mercerus (onNonius, p. 237 .43); in Corinth

by Scaliger (Coniect.Va". Ling., on 7 .9), in Scythia by Ribbeck (In Tragicos
RomanorumPoetasConiectanea.SpecimenI, pp; 2sff., comparing Apollonius
Rhod. 4.31 s ff. with the verses quoted by Cicero); in Colchis at the beginning
by Manutius (on Cic. Fam. 7 .6) and Ladewig (Anal. seen.p. 18).
 MEDEA EXVL; MEDEA

many pombilities are open for a firm decision to be made. Similarity with
the text of Euripides' M1')6e1a is a quite treacherous guide. Study of Nonius'
quotations of the Hecubashows that Ennius departed radically from the
wording of his original even more often thanhe adhered to it.
Tumebus 1 assignedthe two trimeten quoted at Tusc.3 .63 (fr. CVI)to the
Medeaexul. Columna printed the Euripidean passage Turne bus had in mind:
Med. 57-8. Succeeding editors have all followed Columna.
Politianusi compared the pieces quoted in Fam. 7. 6 (fr. cv a) with
Euripides, Med. 214 ff. Manutius3 argued that there is no parallel for the
tetrameter qui ipsesibisapiensprodesse nonquit nequiqumn sapitin the extant
Mi)&1abut that it is rather like a Greek trimeter quoted by Cicero at Fam.
I 3 . 15 . 2 as by Euripides4 and assigned it to the second Ennian Medeapostu-
lated by Victorius and Scaliger. Columna accepted Manutius' argument.
Later editors have not. The substance of qui ipsesibisapiensprodesse nonquit
nequiquam sapitis in fact no further removed from that of Euripides, Med.
294-301 thanis the substance of the previous sententiafrom vv. 215-18. In
any case,just as Athenian actors excised and transferredalmost at will the
general statements about religion, morality, society and politics they found
in the scripts of classicaltragedy and comedy,S so too did the Roman
adapters of thesescripts. Donatus, or his source, could not find the substance
of Terence, Atulr.95~ in Menandcr's'Av6p(a but did in the Ewovxas.
There is no trace, however, in Terence's adaptation of the latter play. The
substance of Caecilius, Com. 143 is apparently absent from the speechof
Menandcr's TTh6K1ov which is beingadapted. Ennius was much admired by
6
early first century rhetoricians for the quality of his sententiae. He seems to
have usedthem much more frequently than did succeeding tragedians. They
were an orator's device as much as clausula rhythm, antithesis, the figures of
speech and thought and the rest and the handbooks of rhetoric gave precise
instructions on their proper use.7 It is likely that Ennius handled them with
more art than my analogy with the behaviour of Athenian actors might
suggest.

1
Aduersariorum tomusII {Paris, 1565), XIX 5. a Misall. I 27.
3 On Cic. Fam.7. 6; in volume VI of the edition published by Aldus (Vcnice,
1579). The 'Scholia' of 1540 do not discussthe matter.
4 This had already been pointed out by Stephanus.
S Euripides, &kchai 1028 is plainly inserted in our text from Medeia 54.
Androm«he 330-2, considered objectionable in its present context by Didymus
and many later scholars, is ascribed by Stobaeus (104.14) to Menander. See
C. W. Friedrich, Die dramatische Funletionder euripideischmGnomen,p. 232.
6
See Rhet. inc. Her. 4.7.
7 Cf. Aristotle, Rhet. 2.21.1394a-1395b, Rhet. inc. Her. 4.24, Q..uintilian,
Inst. 8. 5. 1-8.

347
 COMMENTARY
Columna compared the tetrameter, nequetuumumquamingremiumextolLu
liberorum ex tegenus,quotcdby Cicero at Drat. 155 (fr. o.xx), with Euripides,
Med. 803-4 o<iT'!~ {µov yap iratSas &.vn"at irOTEI 3oovras To Ao1ir6v
and editors included it in the Metka exul until they were persuaded by
Elmsley and Bcrgk1 to put it in the Phoenix.It is plainly a curse and, if placed
in the Metka txul, would suit best an adaptation of Mcdea's taunts at Jason
at the end of Euripides' play.
Four piec.csof trochaic verse referring to Mcdeaarc quoted by Cicero at
Nat. tkor. 3 .65-7 (fr. cvm). The first three must come from the one speech
by a tragic Medca. It would be difficult to make this speech concern any
situation in Mcdea's life except that dealt with by Euripides' Mfi6e1a. The
three piecesshould therefore go into the Metkatxul. With the third Columna
comparedEuripides,Med.399-4,001 andPlankvv.371-5. 3WiththcfustOsann
compared vv. 365-9.4 No paralld of any kind can befound for the second in
the Euripidean text but since it forms a sententiathis is not surprising. Ri~
beck's placing ofit among the incertaincertorum is quite unjustified.The fourth
piece gives an account of Mcdea's Bight from Colchis and the slaughter of
her brothcr.S Neither the Euripidcan Mcdca nor any of her friends alludes
outright to thisdeed.6 Jason throws it up at her afterthe slaying of his own
children: 1334-5 KTavovaa yap 61'aov Kaa1v ,rapia,,ov I TO K<XAAf-
'lt'f>Ct>POVelai~s •ApyoOsaxaq,os. This is the version of the story told by
Sophocles(Schol. Apoll. Rhod.4. 228). The Latin tetrameters give a different
one. However in order to put them in the Metka exulone need only make
the assumptionthat Ennius gave one of Euripides' choral odesa new content.
Greater changesthan thisto the Attic classicsarc well evidencedin the Roman
adaptations.7 Osann8 and Bothe put them in the Metkaexul, Ribbcck and
Vahlen9 excluded them. Vahlen's stated reason was the lack of correspon-
dence in Euripides' Greek. A better reason would bethe possibilityof putting
them in other plays known to Cicero. The problem is a difficultone and study
of Cicero's mode of quotation docs not help. Different poets' handlingsof the
one legend arc sometimes indicated ( Tusc.3 . 62, 4. 69), sometimesnot ( Tusc.
3 .28, 39, 58).

1
Sec below, p. 389.
a Q. EnniiFrag.p. 323. He did not include the three tctrameteninhis actual
collection.
3 Q. Ennii Medea,pp. 100-2. 4 Anal. crit.pp. 117-18.
s Cf. Cicero, Manil. 22, Ovid, Epist. 6. 129-30, Trist. 3 .9.27-34.
6
Medea herself comes nearest at v. 257; contrast the nurse at v. 32, the
chorus at vv. 431 ff., Medea at vv. 476 ff.
7 See above, pp. 333 ff.
8
Anal. crit.p. 125.
9 Cf. Ind. lectt.Berlin 1877, 1 ff. ( = Op. ac. I 34).

348
 MEDEA EXVL; MEDEA
The pieces quoted by Cicero at De orat.3 .217 (fr. crv) and Tusc.4.69 (fr.
CVII)could hardly go anywhere except into an adaptation of theencounter
between Medea and Jason, Euripides, Med.446-626. With thefint Stephanus
compared vv. 502-4, with the second vv. 530-1.
In discussingthe relations between his client and Ptolemaeus at R.ab.Post.
28-9Cicero quotes threeroyal utterances from tragedy: ergoaderatuis, ut ait
poetll noster,quae'summasfrangit infirm4tqUeopes•... nemo nostrumignorat
ttiamsi txperti nonsumusc-OnSUttudinem regiam.regumautemsunthateimptria
'anjmaJuerteac diaopare et praeterrogitatumtsit pief et illae minae•si te
secundoluminehit (hoecodd.) offenderomoriere',quaenon ut delectemur solum
debemus,sedut cauereetiametfagerediscmnus.Scaliger assigned
legereet speetllre 1

the second two to Ennius' Medeaexul, comparing with the first Euripides,
Med. 321aXA'~16' cl>s Ta)(lOTa,J.l'll
1'6yovs Afyeand with the second vv.
352-4 el a' Ii 'moOaa Aaµ~ 6'f,ETai 6eoOI 1<alirat6as hrt-c,s 'riia&
npµovoov xeov6s, I8aviJ- Whatever one maythinkof thefirst identification
both pieces must be assigned to the one speech or seriesof speeches. Such a
••. might be recalled in any circumstances1
striking statement as si te secundo
but an unremarkable saap like animaduerte acdicto. .. would hardly be quoted
except in conjunction with something remarkable from the same context.
The behaviour of Ribbed and V ahlen (in his second edition) in assigning
one and not the other to theMedeaexul is hopelessly unsystematic and makes
a demonstrably false assumption about the fidelity of Roman dramatic adap-
tations to the wording of their originals. The Creon of the Medeaexul is a
possible speaker of the two pieces but so too are several other monarchs of
republican tragedy. The first piece quoted by Cicero could, but need not, be
given to the same speaker)
There is nothing to be said either for or against Ladewig's assignation of
the trimeter quoted by Cicero at Deiot.25,pereantamid duminimiciuna inter-
cidant,to Ennius' adaptation of the speech made by Medea after the chorus's
attempt to dissuade her from her plan for vengeance.•
K. 0. Mueller identified the words ut tibi Titanis Triuia dederitstirpem
liberum,quoted by Varro at Ling. 7. 16 as byEnnius (fr. cxa), with Euripides,
Med. 714-15ovroos fpoosaol ,rpos 6eoovrueaq,6pos IytvotTO ,ra{6oov.
Vahlen found an equally if not more plausible context in the Andromeda
(see above, p. 262).
Three anonymous pieces from extant rhetorical writing, miserisunt qui
uxoresducunt--41tu duxistialteramfrom Rhet. inc. Her. 2. 39, quam magis
1
Coniect. Varr.Ling., on 7 .9.
1
& it is by Cicero himself at Alt. 7. 26. 1.
3 Terzaghi (BFC xx:xn [1925), 16) assignedit to the Medeaexul.
4 Anal. seen.p. 17.

349
 COMMENTARY
urgtttam nuigis admolef atinulumuigtt from Q__uintilian,
otrUmtUJ Inst. 9. 3 . 1 S,
non commemoro quodtlatroni statui opprcssi et dom ista uirorii et segetis
armata ~
conis saeui sopiuit impetii non quod domauit uiros et segetis
armataet manusfrom Charisius, pp. 372.19 and 374.1, have been assigned
to the Mtde11aul. Only the third with its fairly certain reference to
the Colchiandragon need be seriously considered. Saiverius printed it
among the fragments of the Mede11 aul with the wording of H. Putschius'
text non memoro,quoddrtU:Onissopiui impetum; non, quoduiros donu,uitt
#getis armat« manus. Plank complained of the lack of a verbal parallel
in Euripides and Welcker 1 supplied vv. 476-82 lact)O'<X a•, ws
tacxcnv
'001'tvoov0001 I TCXVTOV OWE1oil:n)aav•Apyci'>ov 01<aq,os,Imµq,8wra
TCXVf>OOV,rup,rvOO>V hncnari,v I 3NYA1JalKai 0'11'EpOWTa 6avaaiµov
yvt1v· I 6pal<C)V'T<X as
e·, irayxpvaov &llmxoov~ I omfpms la<t>3£
iroAvirA6Ko1s &rnvoswv, IKTElvau'awaxov aol cpaos aCJritp1ov.~The
uncertainty of the text makes further discussion profitless.3

CIII
(a) These nine trimeters are said by Cicero at Nat. Jeor.3. 75 to have been
spoken by iliaanus,'that famous old woman•. i.e. Medea' s nurse. They must
have been the opening verses of Ennius' Medeaexul for at Fin. 1. s Cicero
refers to the whole play as utin11m ne in nemore;ancient writers commonly
referred to poems by means of their opening words. 4 The context of
Priscian's quotation at Gramm. m 423 . 3s ff. suggests the same conclusion;
the examples of comic trimeters in his treatise on Terence's metres all come
from the prologues and opening scenes of plays; those of tragic trimeters
look as if they come from similar places. Thus the theory once propounded
by Fraenkd5 according to which Ennius' tragedy opened with a dialogue
between the paedagogus and the nurse with the paedagogus makingthe first
utterance cannot be seriously entertained.
BehindEnnius' Latin lie the first eight trimeters of Euripides' M11&1a:
d8 1
~• • ApyoOs µ1')61crnTaa8atC7Kat<>S
K6).xwv 's alav l<V<MCJS
Ivµir~11ya6as,
1
Die griech.Trag.p. 1378.
:a The sopiui of the Latin tragedian makes his account closer to that of
Apollonius Rhod. ♦. 123 ff.
3 Cf., most recently, S. Boscherini, SIFC N.S. :xxx (1958), 1~15.
Boscherini's point that Acciw would have written dracontisis unhelpful;
classicalrhetoricians and their scribes would have normalised to draconis.
4 Cf. Theocritus 14.30, Propcrtius 2.24.2, Ovid, Trist. 2.261, S34, Persius
1.96, Martial 4.14.13-14, 14.185.2.
S HermesLXVII(1932), 3SS f.; not reprinted in Kl. BtilT.

350
 MEDEA EXVL; MEDEA
1,1116' Tl11Afov
fv va1TCX1cn 1TEO'Elv
1TOTE
1TE\IK1l,
T1,1118eiaa 1,1116'
lpnµcoo-at XEPCXS
&v6p<r,v &p{aTCi)V,
ot TO irayxpvaov 6epas
Tlu.l~ µm'\A8ov. ov yap av6eairo1v•lµi)
Mf\6e1cxmipyovs yi'is rnAEVa''ICi)Atdas
fp<A>Tl8vµov hrnAayela' 'laaovas.
These were designed to suggest the atmosphere created by Jason's decision to
marry Creon's daughter and by Medea's angry reaction as much as to nar-
rate the background of the plot. The sudden mention of an earlier element in
the traditional story at v. 3 and the anaphoric repetition l,.lfl•.. l,.lfl6'... µ116',
much more unusual and striking in Attic trimeters than a similar pheno-
menon would be in Latin trimeters, indicate the nurse's nervous distraction.
Euripides' verses were much admired in antiquity for their arrangement. 1
As occasionally in his adaptation of the 'EKo:131'1
Ennius preserves the bones
of Euripides' grammatical arrangement (seeabove on frs.LXXXIV and xcn):
the series of negative wishes, the relative clause, the independent apodotic
sentence. He removes some of the Euripidean fleshand adds some of his
own. The dressingis typically Roman with little counterpart in Euripides'
Greek.
The nurse's first wish with its rich geographical detail and the magnificent
image of the Argo swooping like a bird through the Clashing Rocks is
omitted, the second is filled out into two and the content of the third together
with that of the relative clause is put into a relative clause and an adverbial
exp1aining the etymology of the name Argo. Thus the events of the tradi-
tional story are referred to in chronological sequence. Whatever Ennius'
motives were in securingthissequence they were pretty certainly in accord
with the teaching of contemporary rhetoricians. 3 Euripides made the heroes
themselves row the Argo; Ennius abandons this detail, perhaps thinking it
inappropriate for men of the first social rank to perform such a menial task.
Ennius may also have had contemporary naval expeditions in mind when he
had his Argo built of p.r-wood instead of the traditional pine. Ennius'
etymology is hisown but a commentary upon the Mri&1a, as well as hisown
individual love of etymologising, may have provoked him to include it; the
mode of expression-nunc nominaturnomine-suggests the academic com-
mentator with his mind on the present rather than a personage, albeit a
minor one, of the heroic saga.

I Cf. Med. hyp. rncnwiTat 6~ ti el~oAi) 61cx

TO1Ta8T)T1KC)s
cfyav fxetv t<al
ti m~pyc:ca(a· 1,1116'h, vaircna11<alTCXt~s· 6mp ayvof\aas T11,1ax(6asTct>
VaTEf>Ci>cp11crl t<EXPfia8a1,
1T()WTct> ~ •01,111pas.
1
Cf. Cicero, lnu. 1 . 29 aperteautemna"atio potent es.sesi ut quidqueprimum
gestum erit ita primum exponeturet rerumac temporumordoseruabiturut ita nar-
renturut gestaereseruntaut ut potuissegeri uidebuntur.
351
 COMMENTARY

Many of the words and forms used by Euripides in hiseight trimeten were
quite absent from everyday Attic. A few of Ennius' must have beenrare in
contemporary Roman speech but none could be labelled purdy poetic. On
the other hand, whereas Euripides played discreetly with only four sounds
11,P, s, and T, Ennius plays extensively with A, B, c, E, R, s, and T. He adds the
etymological figure nominalurnomine,more striking than the usual type
because of its pleonasm, substitutes periphrases for nouns (obiegna . .. trobes
~ '1'ffllKTl)and verbs (inchoontli txortliumapisset;tjfmtt pttltm ~ rnAEVOe)
and loads hissentences with words conveying more weight than information
. .. adtmom . .. inchoonJi
(stcuribus . .. llrittis. .. imptrioregis. .. mons . .. s«UO ).
The degree of rhetorical embellishment is not however so high as when Attic
trimeten are turned into verses to be accompanied by the tibiot (cf. fr.
LXDCIV).
208-9urioun I
ne••• NCaribaa cuaa accidiaet •• •ac1tanm: ~
ete·
~• ... µ1')6"... fflOtlv iron I Tµll&taa.
It is possible that the readings of our codices reBect the rhetorician's own
belief about the text. Cicero clearly believed the nurse to have said ClltSlle
occitlissent 11biegn11t(see Top. 61 in his trobibus).However the statements of
Varro at Ling. 1. 33 and Priscian at Gromm. u 320. 1 s ££ must go back
ultimatdy to a grammarian's examination of a text and, at least where
Ennius is concerned, should be preferred to those of rhetoricians and literary
amateurs rdying on memory. The readings at Priscian, Gramm.m 423 . 3s f[
may be medieval corruptions.
A«itltrt, 'cadere', occurs frequently in tragedy (4 times elsewhere~ .s
codere);it is relativdy uncommon in comedy (7 times~ 29 codert).
208 in nemore Pelio: ~
a, vamxun ITT)A{ov.
Nemus,'a grove of trees reverenced as belonging to a deity•, was a word of
the sacral language, absent from comedy (where saltusis common) and rare
in classical prose.
The so-called 'poetic singular•is not common in the remains of republican
tragedy but is in those of Ennius' Annales(8s,224, 277, 299, 439, 472, •84,
533).
The use of the adjective instead of the genitive of the noun raises the tone
above the commonplace; cf. Livius, Trog.3S in Ptlio.• . oai and above on
v. 100.

209 abiegna .•. trabes: ~ '1TNKT1·

For the periphrasis cf. Virgil, Aen. 2.112 trabibusconttxtusocernis,6.181
froxineaequetrobes.
Elsewhere the Argo is always said to have beenmade of pine wood
(Herodotus 4. 179, Euripides, Andr. 863, Apollonius 1.386, s2.s, Catullus
352
 MEDEA EXVL; MEDEA

64. 1 ff., Horace, Epod. 16. 57 f., Propertius 3 .22. 11 ff., Ovid, Am. 2. 11 .2,
Met. 1 . 94 f., Anon. Culex 137, Phacdrus 4. 7. 4 ff.). Mere caprice may have
dictated Ennius' altcration 1 but his liking for sound play {better achieved
with abiesin thiscontext than with pinus)and for rewriting the heroic myth
in contemporary terms must be considered. A fu-wood Argo suggested to
Ennius' audience a military expedition whereas a pine-wood one would have
suggested commercial enterprise. 2

210-11 neue inde naais inchoandies:ordiam I cepiaet: for exordium

apisstt there arc good early parallels: Plautus, Poen.2 indemihiprincipium
capiam,Varro, Rust. 3 .1 .10 initiumcapiamhinc,Cicero, Phil. 5. 35 a Bruto. •.
capiamusexordium,Fin. 5 .23, Leg. 1. 8, 2. 7 a at.erisdisinmortalibus
sunt nobis
agendicapienda primordia,T.L.L. m 324 .1 ff.; for exordiumcoepissetthe paral-
lels arc late and in authon ignorant of republican drama: Columella 3 . 6. 4
sol in eandempartemsigniferipereosdtmnumeros,editperquoscursussuiprinci-
pium coeperat,Tacitus, Hist. 2.79 initiumfermdi ad Vespasianumimperii
Alexandriaecoeptum,Ammianus 26.6.19 principiaincauu coeptaet temere,
T.L.L. m 1427. 84-ff.
For the pleonastic gerund inchoandi cf. Plautus, Mil. 63 7 ut apudteexemplum
experiundihabeas,ne titastf oras,Poen.34 domumsmnonesf abulandiconferant.
For nautmincohare cf.Livy 21 •26. 8. The word is rare in republican drama.
According to Servius, Atn. 6.252: est uerbumsaaorum.
. The metaphor from weaving may have been still alive in Ennius'
exordium:cf. Ann. 477 idemcampushabettextrinumnauibuslongis.3The word
occun elsewhere in republican drama only at Trag. inc. 181.

211-12 I
quae nunc nominator nomine Argo: for the etymological
figure cf. Sophocles, Phil. 6o5 6voµa 6' oovoµCX3m>, Euripides, Ion 80-1,
800, Pacuvius, Trag.239 quistu es ,nulierquaeme insuetonuncupastinomine?,
Trag. inc. 97, Plautus, Asin. 780, Terence, Phorm.739. 'OvoµCX3EIV and
nominarearc normally unaccompanied. For etymological figure in general
see above on vv. 6-7.

212-13 qwa Argiai in ea delecti uiri I uecti petebant: ~ x~ I

I av6poov&pfO"'Toov
ol .•. IJE"rilA&ov.
This is the text that must have stood in the archetype of our manuscripts
of the anonymous rhetorician's treatiseand, with the exception of delecti,
in

1
At Euripides, Hel. 229 ff. the ship that took Paris to Sparta is made of pine,
at Hek. 631 ff. of fir.
a Sec Theophrastus, Hist. plant. S. 7. 1, Livy 28. 4S. 18, RE 2 IV i (1932),
2216 ff., s.v. Tonne.
3 C£ also, in a different context, Plautus, Pstud.399-400.

23 3S3 JTO
 COMMENTARY
that of Cicero, Tusc.1 . 4S, where it is confirmed by the context of Cicero' s
discourse. Priscian, Gramm.m 424. s~ offers the possible but muchlessgood
quauectiArgiui dilectiuiri petebantillam. Imitations of Ennius' phraseology
by Lucretius (1.86 ductoresDanaumdelecti,prima uirorum)and Virgil (Eel.
I
4. 34-s alteraquaeueluuArgo delectosheroas)put delectibeyond reasonable
doubL .
The main verbal ideaseemsto be carried in the participle: 'because in her
(N.B.the emphatic position of in ea) were carried the Argiuiwho went in·
search... '. The etymological explanation given by Ennius for thename Argo
lies behind Cicero, Arat. 277 Argolicam . .. nauem(contrast Aratus 504) and
Manilius 1 . 694Argiuumqueratem.It is not to be found in Greek writings; the
nearestis Hegcsander's6-n W •Apye1TI) 1TOAElKCXTE01<EVaaeti. 1
It makes
complete senseand doesnot contradict the traditional story only if Argiuiis
interpreted not as 'men of Peloponncsian Argos' but as 'Graed of the heroic
age'. This was a regular usage of the republican stage; cf. Ennius, Trag.330
eloquereeloquereres Argiuumproeliout se sustinet(the Argives and others
fighting before Troy), Plautus, Amph. 208 abiturosagroArgiuos(the soldiers
of the Thcban king Amphitruo ). Ennius himselfprobably concocted the
ctymology. 1 Plautus was given to introducing his own etymologies in
comedy; cf. the Latinate explanation of Epidamnus at Men. 263.
It is hard to say whether with delectiEnnius is making explicit an element
of the traditional story neglected by Euripides (c£ Apollonius 3 . 347-8 Tij 6'
waye1paµevos ITavaxatSos I
efTl cptp10TOVtip~v. Theocritus 13. 17-18
&p1crrf\es ... ,raaav AJ<1TOA{oov 1Tp0AEAE-yµa10l, Apollodorus I .9. 16) or
merely heightening the tone of the nurse's speech with a technical term of
the Roman military language (cf. T.L.L. vii.452.83 ff.).

213 ~
pellem inauratam arietis: TO ircxyxpvcrov6tpcxs.
Comedy has auratusthree times, inauratusnot at all.

214 Colcbia, imperio regis Peliae, per dolum: ~ TlEA{q:.

The preposition is omitted before Colchisas often in tragic style; sec above
on v. 43.
ImperioregisPeliaereminds modem scholars of Apollonius' ~O'lAf\oS
{q,T)µOOWTJ TlEA(ao(1 . 3) but was probably designed to remind Ennius'
audience of the style of official Roman military reports; cf. C.I.L. 11 626
( 133 B.C., dedication to Hercules) L. MummiL. f. cos.duct.auspidoimperioque
eius.khaia capt.CorintodeletoRomamredieittriumphans.The ductuswas Jason's
and, in the view of the hostile nurse, there existed no valid auspiciumat all.

1
Etym. mag.s.v. 'Apyoo, Tzetzes, Lycophr.883.
1
Cf. B. Bilinski, in Tragica1, 88 ff.

354
 MEDEA EXVL; MEDEA
The designationof Peliasas rex both reflectedthestyle of officialreports,1 in
order to elevatethe tone of the nurse'sdiscourse,and graspedat contemporary
Roman political prejudice,in order to excite sympathy for theplight of the
nurse and her mistress.
Perdolummarked the behaviour of the Argonauts as reprehensibleby both
heroic (cf.Euripides,I.A. 1457 66A<t> 6', ayewG:,s 'AT~ T" OV1< ~{CA>S)
and contemporary Roman standards (cf.Livy 1. 53 .4 minimearte Romana,
lraudeac dolo,42.47.5 ff.).
215 nam nomquam ... domo eft"erretpedem: ~ ov
yap av...
mipyovs yfis frrAEVa"'ICA>AKfas.
Remote past apodoses,both dependent and independent, seem normally
to have been expressedby the imperfect rather than the pluperfectsubjunc-
tive in the Latin of republican drama; see Thomas, R.tcherches, pp. 199 ff.
NuffliU'Zfflis often merdy a strong nonin republican drama; cf. Donatus,
Ter. Andr.384, 410.
Domoabireis the usual phrase in comedy (Plautus,Amph. 502-3, Epid.46,
Mere.12, Stich.29, Trin. 1010, Terence, Eun. 661). Ptdtm ejferrtappears in
elevatedcontextsin Plautine comedy (Ba«h.423, Capt.456-7, Mere.831) as
do similarperiphraseswith pes.The only thingparallelin Terentian comedy
is Andr.808 nampol si id scissemnumquamhuetetulisstmptdtm, a verse uttered
under great emotional stress.
Medea's homelesmessis referred to later in the Euripidcan prologue (vv.
31 ff.) and often during the action of the play (vv. 328,431,441, 64,2,798).

215-16 era emms mea ••• 1 Medeaanimo aegro: ~ &a-rro1v•11,111

I
M116e1a.
The Euripidcan chorus refers to Medea's mental aberration at vv. 431-2:
avs· lK IJEV olKCi.>V
,rcrrp(CA>v frrAEVaasI l,l<XlVOl,IW~ Kp<X6{~.
Only the metrical pause separatesMedea's name from the designation of
her status in the Euripidean prologue. Ennius raisesthe tone by widening
the gap; contrast Ennius, Trag. 291-2 a soau I Oenonu,o,Plautus, Capt. 26
medicusMenarchusemit, Cist. 171 d4l eam puellam meretridMel«nidi and
compare Sophocles,0. T. 826--7, Euripides, El. 763-4, Hipp. 51-3, 581-2,
Tr. 861-2, Plautus, Ba«h.589.
For the phraseology of errans... animoaegroc£ Ennius, Trag. ap. Cic.
Tu.sc.3. 5 (seefr. cuoav) animusaegerstmpererrat,Plautus, Mere.18-31 nam
amortmhaeccunctauitia stctari solent. .. aegritudo ... sed amoriacaduntttiam
hatequaedixi minus. •• error. ..•
Era is the normal appellation in comedy; dominooccun only at Plautus,
Cist. 773, Stich.296, Terence, Haut. 298, 301, 628.
1
Cf. v. 292 OtnomtW rtgt and above on v. 34.
3SS
 COMMENTARY
216 am.ore IUUO •ucia: ~lpwn 8vµov OOl'Aaytta· ·1aaovos.
For the wounding effect of love c£ Euripides, Hipp. 392 me( µ•
lpc:a>ShpCOO'EV, Thcoaitus 11.15, 30.10, Ca11irnachus, A.P. 12.134.1,
Plautus, Cist. 298, PerS424, Luactius 1. 34, 4.1048, Virgil, Atn. 4. 1,
Propcrtius 3.21.32, 3.24.18, Ovid, Ars 1.21-2, 166, 257, Seneca, Ag.
188-9.
'"Epc:a>S,
normally &1v6s (Euripides, Hipp. 28 et al.), never receives an
epithet of the characterof saeuusin Attic tragedy. Ennius' phrase appealed
to later Latin poets; c£ Virgil, Eel. 8 .47-8 saeuus anwr docuit natorum
I
sanguinematrem commaculare manus,Aen. 4.532 saeuit amor, Lygdamus
I
4. 65--6 saeuus amor docuit ualidos temptarelabores, saeuus amor docuit
uerberasaeuapati, Seneca, Med. 849-51 quonamcruentamaenas praeceps I
amoresaeuoI rapitur?
CIV
There arc grounds for thinking that Cicero' s quotation is as defective as those
of Terence, Eun. 46-9 at Nat. dtor. 3 . 72, Eun. 114-15 at Att. 7. 3 . 10, and
Andr.117-28 at l)eorat. 2.327, which would cause some difficultyin the
absence of a direct tradition of Terence's comedies.
Eonius' apparent trimcters must come from an adaptation of the Euripi-
dcan Mcdca's verbal assault on Jason, vv. 465-519. One would expect
Ennius to have turned such a highly emotional utterance into musically
accompanied verse, especially as he did so with Jason's reply (see fr. cvu).
Both Attic tragedy and Roman comedy on occasion insert a series of tri-
mctcrs into musically accompanied scenes but always for good cause:
Aeschylus, Pers.176-214 (narration of dream); Euripides, I.A. 402 to end of
scene (sharpdrop in emotional tone); Plautus, Amph. 1006 to end of scene,
Cist. 747 to end of scene, Cure.635 to end of scene (sharp drop in emotional
tone); Rud.1338 to end of scene (recitation of oath); Ba«h. 997 to end of
scene, PerS4501-12, 520-7, Pseud.998 to end of scene (reading of letter);
Stich.762-8 (piper takes time off for a drink); Terence, Andr.215-24, Eun.
323-51, Haut. 265-311 (narration). In Accius' BrutusTarquin recounts his
dream in trimctcrs (Trag. Praet.17-28) while the coniectores interpret it in
trochaic tetrameters (29-38). Ennius' Andromachc seems to have described
Hector' s death in trimctcrs and expressed her own emotions in musically
accompanied verses (Trag. 78-94). There is no reason why Ennius' Mcdca
should descend to trimctcn for rhetorical questions like quonuncme uortam?
quoJiter incipiamingrtdil In any case such tautological phrasal doublets
normally occur in musically accompanied verses rather than trimctcrs (see
above on v. 19).
One could treat the trimctcrs as 'lyric', like those at Plautus, Baah.
669,Epid. 24, 46-7, 177, Stich. 300 et al. The unusual arrangement of
 MEDEA EXVL; MEDEA
words at the beginningof the verse v-, v- - 1 might be excused by a
musical setting.
After her rhetorical question vvv,rot Tp<m<..>µcn; theEuripidcan Mcdca
poses a dilemma: ,rarepa ,rpos irarpos 66µovs, I oOs aol irpo6ovaa Kai
,rarpav aq,1K6µ11v; I ft ,rpos TaAOO~ TTEA1a6as; KaAOOS y" &v ow I
&~cnVT6µ' olK01s ~v ,rcrripa KaTa(TavoV (502-5). The factthat Ennius'
domumpaternamnelanne ad Peliaefiliasl docs not follow the Euripidcan
dilemma closelywould not of itself besignificant.But sucb diJcrnrnataseem
regularly, in both Greek writingi and Latin,3 to be formed like Euripides,
Med. 502-5. Furthermore the famous dilemma of Caius Gracchus, quome
miser conferamlquo uortaml in Capitoliumnelat J,atris sanguinemadet.an
Jomumlm4lremneut miseram1'unentantem uideamet abitctaml(Cicero, De orat.
3 . 214; c£ Cicero, Mur. 88, Ps. Sall.Inu. 1), must have been uttered with
Ennius' Mcdcain mind.Norden4 thought that the absenceof the customary
answers to each question of the dilemma in the Ennian fragment showed
that Gracchus' inspiration lay elsewhere.Things arc the other way about.
The presence of the answers in the Graccbao fragment may show that
Cicero misquoted Ennius.

217 quod iter incipiam iDgredi: the plconastic ingrediis added to make
the phrase longer than the preceding one (secabove on 11. 19). Plautus has
iter indperethree times (Cas. 164, 817, Mere.913), iter inaptareonce (True.
130), itercapereonce (Bacch.325). For ingrediindperecf. Cicero, Catil. 3 .6,
Phacdrus 5.7.17.

218 domum patemamne: contrast Plautus, Rud.116 ad alienamdomum,

Stich. 5o6 in patriamdomum.But even if the text of the trimcter were above
suspicion there would beno need to understand an adanoKOlVOV ( on the
analogy of Plautus, Asin. 163, Pseud.124 et al.). Tragic stylefrequently dis-
penseswith prepositions (secabove, v. 43). Domuspatmu,, 'home of my
father', occurs only here in republican drama; domuspatriawas the regular
phrase (Plautus, Mere. 831, Stich. 5o6, Ennius, Trag. 84); for paternusas a
synonym of patriuscf. Plautus, Stich. 88 (paratragic), Pacuvius, Trag. 144,
328.

1
For an attempt to emend out of existencethis and other tragic examplesof
the arrangement see Havet, RPh XIV (1890), 33 f.
2
C£ Sophocles,Ai. 4S7 ff., Euripides,Hile. 1094 ff., Hnaltles 1281 ff., I. T.
9(Sff.,Diodorus 13.31.1.
3 C£ Terence, Phomi. 185-6, Catullw 64. 177 ff., Sallwt, lug. 14. 17, Virgil,
Aen. 4. S34 ff., Seneca, Med. 451 ff.
4 Die antilteKunstprosa I, edition of 1915, Nachtriige,
pp. 13 1f.

357
 COMMENTARY

CV

Cicero adviseshis friend Trcbatius to take the same attitude towards being
an expatriate as did Mcdcain Ennius' tragedy. He quotes three trochaic
verses verbatim, the fint two beingrevealed by their rhythm and vocabulary
(gessere,patriaprocul,aetatemagerent),the third by its mode of quotation as
well. It haslong beenrealised that in hisaccount of the context Cicero inter-
weaves Ennius' words and phrases with his own. 1 Q!!ite foreign to the style
of his correspondcncc arc the phrases quaeCorinthumaram altamhabebant, i

and matronaeopulentaeoptumates.3 The superlative form of the verbal parti-

ciple gypsatissimiswould have indicated considerable violence of feeling
even on the stagc.4 It is quite out of keeping with Ciccro's urbane admoni-
tions to Trcbatius. The alliterative phrase uitio. .. uerterentalsostands out. It
is common in dramaS but absent elsewhere in Cicero's letters and formal
writing.6
Sixteenth and seventeenth century scholars gave to Ennius a form of
words like matronaeopul.entae optumatesquae Corinthumaram altam habetis
corresponding with Euripides' Kop{v61<n ywalKES; Bcntlcy7 the trochaic
tetrameter quaeCorinthumaram altam habetismatronaeopul.entae optimates.It
would be in Ennius' manner to expand a simple vocative with adjectives and
a relative clause, especially when turning trimctcrs into musically accom-
panied verses (secabove on v. 34). A post-classical Attic tragedy has Medea
addressing the women of Corinth with a similar relative clause: [q>O.<Xl
I[
y]watl(£S at Kop{v&tovmSov oha:t]n xoopasTfia6e TI'a'Tp<t)OlS
voµo1s.8 However the order of words in Bentley's text with the relative
1
C£ Sest. 121, De orat. 1. 199, Diu. I. 132, Tusc. 1.69.
a Cicero does not have urbem,oppidumetc. habereelsewhere; however such
phrases arc common in the historians (sec T.L.L. VJ iii.2401 .29 ff.).
3 Opulentusdoes not occur elsewhere in Cicero's own letters and is rare in
his formal writings. The asyndeton bimembre has no real parallel among those
collected by C. A. Lehmann, ~stionts Tullianae(Prague and Leipzig, 1886),
pp. 24 ff., and H. Sjogren, EranosXVI (1916), 32 ff. The formulaic certaclaraat
Att. 16. 13 A.2 is the nearest. There arc plenty in drama; see above on v. 9.
4 Cf. Plautus, Asin. 282-3 maxumasopimitates gaudioecfertissimasI suiserisille
una mecumpariet,gnatoqueet patri, Aul. 723 perditissimus egosum omniumin terra,
824-5 egonte emittammanu I scelerumcumulatissime ?, Capt. 17S, Cas. 694, Cure
16, Mm. 698.
s C£ Plautus, Amph. 1142 et al. (not in Terence).
6
Vitio dareoccun at Off. 1.71, 1.112, 2.58.
7 Emendationes Cic. Tusc. p. 26.
8
Pap. Brit. Mus. 186 (Milne, Catal. 77): c£ Euripides, Hipp. 373-4 .
Tpo31'1v1CX1 ywatl<lS, at T66' faxa-rov I olt<EtTExcbpasTle~omcxs ,rpov&mov.
 MEDEA EXVL; MEDEA

clause pr«eding t:he vocative matrow hasno exact parallel that I can find
in republican drama.1 The older texts on the other hand leave it difficult to
explain why Cicero adopted such an unusual order of words in his own
sentence. If a guesshasto be madeat Ennius' exact words I would suggest
uos Corinthumaram altamhabttismatrow opulent« optum4tes.There is no
equivalent of thesewords in Euripides' play but the adaptation of the •EKaf:nl
provides many parallelsfor the freedom I am allowing Ennius.
Petrus Victorius1 took ilia manibus
gypsatissimisto be Ennius' description of
a Mecleawho acquired her white complexion by means of cosmetics.Jacobus
Nicolaus3argued that the phrase was Cicero's own description of the male
actor playing the role of Mecleaand drew attention to Isidore, Etym.10. 119,
nomenautemhypocritae tractumestabspedeeorumquiin spectaculis conttcta Jade
inaduntdistinguentes uultumcaerukominioque coloreet aterispigmentis,habentes
simul«raorislinteagipsataet uariocoloredistincta,
nonnumquam et collaet manus
aeta perungentesut ad persow colorempffllfflirentet populumdum in ludis
agerentfalkrent; modoin spedeuirimodoinf emiw . ... The passage of Isidore is
now neglected4 but Jacobus Nicolaus' explanation of Cicero's discourseseems
to be univenally accepted..S
Jacobus' explanation multiplies hypotheses nnnccemrily. Cicero doesnot
mention the stage, except in a notional way with.quoniam Mttkamcoepiagere
much later in his letter. nla manibusgypsatissimismust reflect Meclea's
description ofherself or her report of someone else'sdescription.There was a
contrast betweenuosmatrow opulentaeoptumatesand egoperegrinamanibus
The normal whitening cosmeticswere, in Attic drama, 'f'll,\V-
gypsatissimis.
81ov{Aristophanes,EJJeks.878, Eubulus, fr. 98 et al.) and, in Roman, cerussa
(Plautus, Most.258) and aeta {Plautus, True.289--94). The only mention of
Y'J'i)OS/gypsum usedfor this purpose isat Rufinus, A.P. s.19, as far as I can see
{contrastA.P. 11. 374 and 408). It may have beenusedin elaborate prepara-
tions for very dark complexions or persons wishingto disparage the use of

1
C£, however, Plautus, Amph. 1o66 exsurgitt. .. qui tmort meoocddistisp,ae
metu.
i Var. Lectt. xx 13. Cf. Tumebus, Adunsariorum libri XII (Paris, 1s64),

m 18 (comparing Valerius Flaccus 2. ISO and Rufinus, A.P. s. 19).

3 Misc. Epiph. 11 8 (in Gruterus, Lampas,v Suppl., 309 ff.).
4 Cf. for example Bliimner, RE VII ii (1912), 2100. 19 ff., s.v. Gypsum,
Warnecke, RE VIIIii (1913), 2118 .40 ff., s.v. Histrio,Bieber, RE XIVii (1930),
2082. S3 ff., s.v. Maske.
s T. B. L. Webster and 0. Skutsch, in NauiculaChiloniensis,110 n. 1, vary
this explanationslightly: ' ... oder konnte etwa Cicero Bur. 30 ,r~ov 6ip1')v
ungcnau als,r~ov xtpa in Erinnerung gehabt Wld dieseschneeweisseHand
scherzhaft, da es sich um Schauspicler handelte, als gypsatissimabezeichnet
haben?.

359
 COMMENTARY
atta may have deliberatelymisdescribedit asgypsum,a substancenormally
used for decoratingwalls and ceilings.1
The substanceof ne sibi uitio ill« umerentquoJabtsseta patriacertainly
appearedin Ennius' play, the expressionuitioumereprobably. Attempts at
complete restoration could not hope to be univenally convincing. 0.
Skutsch2 makesthe plausiblesuggestionne mihi uitiouosuortatisexul a patria
quodabsum.
In Euripides' Mft6e1athe heroine's fust entry is preceded by a lyric
dialoguebetween the nurse and the chorus of Corinthian citizen women.
The latter expressmendsbipand sympathyfor Medea,distastefor what they
considerto be Jason's wrongdoing. Neither here nor elsewherein the play
do they aiticise anything in Medea except her violent reaction to Jason's
behaviour.When Medea entersshe is afraid lestthe Corinthian women re-
gard her as an arrogant person (vv. 214-21 )3and a foreignerhard to get on
with (vv. 222-4 ). 4 She is depressedby the thought of how defencelessher
foreignnessmakes her (vv. 252-8; c£ vv. 591, 801-2). Euripidesmakescon-
siderableplay with thefactthat hisheroineis not a Greek,givingher mostof
the characteristicsassociatedin fifth-century Athens with foreign wom~
proneness to excessivelamentation, servility before men in authority,
inability to understand broken promises,knowledge of magic and witch-
craft.SHe repeatsover and over againthe theme of her absencefrom home
(vv. 32-5, 166, 2ss~.328, 431-8, 502-3, 642-51, 798-9) but allows only
Jason to make it a matter of reproach (vv. 13~32). Medea and the Corin-
thian women treat each other as being of equal digmty.6
Behind Ennius' contrast of the Corinthian matronaeopulentaeoptumates
and the barbarianMedeanumibus gypsatwimismust lieEuripides,Med.252-8:
m· ovyap CXVTCS npos en,c&µ•i\1<11
A6ycs·
aol µw n6A1se· 1)5' ml 1<CXl n<rrpos 66µ01
0
piov T 6VT1atS 1CCXl
cpfAc..>v
awovafa,
ow• vpPf30µa1
tycl>s• lpt1µcs &1"0A1s
npos &v&p6s,1Kyijs ~apf3apov ~1JC7l&M1,
ov µf1-rip',OVK&W.cp6v,ovxl ovyywi\
p&8opµfaaa8at Tfia&• fxovaa avµcpopcis.

1
Sec Bliimner, RE vn ii (1912), 2094 f. Pliny (Nat. 35. 199) speaks of creta
being used to mark the feet of slavesin the market; Tibullus (2. 3 . 6o) and Ovid
{Am.1. 8. 64), both writing emotively rather than descriptively, use the verb
3
gypsarein this connection. In NauiculaChiloniensis,109.
3 For the Athenian dislike of asµv6Tflssee Euripides, Hipp. 91 ff., Alk. 199 ff.
4 For the Athenian attitude to foreigners see Aeschylw, Hik. 200 ff.,
Euripides, Hik. 891 ff.
5 Cf. D. L. Page, Euripides:Medea{Oxford, 1938), pp. xviii ff.
6
Note Mcdea'sform of address Kopfv8tatyvvall<ES at v. 214; cpfAatat vv.
227. 377. 765, 1116, 1236.
 MEDEA EXVL; MEDEA

Significantly Ennius uses the word matron« rather than mulierts. The
Corinthian women are bound to their mates by iustummatrimonium and
hence protected by all the majesty of the city-state's law and cwtom. Medea,
by implication, is only a concubina.Her powdered limbsmake even plainer
her foreignness and friendlessness. A fair complexion was much prized in
Rome and Greek-speaking communities 1 and seemsto have beena common
indicator of class.i Considering the attitude to cosmetic aids in Grccce,3 an
attitude reBected plainly in Plautw' adaptations of Attic comedy, 4 it is un-
likely that any women except foreign-born harlots usedthem in early second
century Rome.
The formal structure of Ennius' sententia

Al Ai
A !multi.nutmrtm bentgtssereat publicampatriaprocul;

B !multiqui domiatlaltm agerentproptereasunt improbati

imitates that of Euripides Med. 214-17

A Iol6a yapiro~ovs 13f>o!wv

~~
cnµvov5YEYWTa5 cSiii.ichoov
&rro,
Ai
B T0\155• W 8vpa(o15.

6\/0'Wtav hm'}aavTo 1<alf!><it8vµlav

However not only are Euripides' trimeters replaced with musically accom-
panied trochaic verses and the triple alliteration of publia,mpatridproculand
the anaphora of multi. .. multiSbut a different accusation is replied to, one of

1
Cf. Catullus 13.4, Propertius 2.13.53, Charlton 2.2.2; for defence of a
dark one cf. Theocritus 10. 26-9, Asclepiades, A.P. 5. 210, Philodcmw, A.P.
5.121, Virgil, Eel. 2.15-18, 10.38~.
a Cf. Cicero, Pis.fr. 8, Propertius 1.4. 13, Horace, Epist.I. 18. 3-4, Lucian,
Parasit.41. 3 Cf. Xenophon, Oik. 10, Lucian, Dial. deor.20. 10.
4 Most.258 ff., True.289 ff.

s Anaphora of adjectives, nouns and verbs is very uncommon in the tri-

meters of Attic tragedy and comedy. It is quite common in the trimeters of
Roman comedy and very common in the long musically accompanied verses
(see Haffter, Vntersuchungen, p. 84).
361
 COMMENTARY
peregrirumorather than O'EIJVOTT\S-Grauitas is probably the nearest Latin
and that of course was no vice at Rome. Absenting
equivalent of O'EIJVOTl)S
himselffrom the family circle left a Roman open to serious rebuke. 1 The
Ennian sententia drew on the experience of Roman magistrates, whose
prouindae had begun in recent times to be commonly assignedabroad, and
on the language with which they boasted of politicaland military success.i
The second sententiaquoted by Cicero probably comes from Ennius'
adaptation of Mcdca's address to Crcon, Euripides, Med. 292-315. It would
have replaced the sententiaon ~{a at vv. 294-301:
xP1is· o<nroe·
6o-ns &fmq,fXA>V
,dq,VK'&vt'\p
iratSas mp1aa&')sIK6166:0'l<E0'8cnaoq,ovs·
XCA>plsyap 6'.AT)sfls fxovcnv &pyfas
q,86vov,rpos &crrwv&).q,avovcnSuaµevij.
cnccnolcn~ yap KCnva aoq,a
irpoaq,ipc..>v
66~1s ~pEloS ICOV a~ mq,v1dvar
TWVs· CXV SOICOWTCA>V elSWCXITI 1TOIK0.ov
1<pdaaoovv0~1a&lsw 1r6AE1 '-vrrpos q,avij.
These verses were full of meaning for the Athenians of the fifth century but
had little for the Romans of the early second, among whom traditional
wisdom had not yet been seriously challenged. Ennius' replacement was a
commonplace of vulgar talk among both Greekand Latin speakers. 3

1
Cf. Naevius, Com. 92-3 primum ad uirtutem ut redeatis(Fabricius: reductis
cod.) abeatisab ig,u,uuiI domi(Ribbeck: domoscod.) patrespatriamut colatispotius
qiuunperegriprobra(Ribbeck: probrocod.), Cicero, Plane. 29 ut uiuat cum suis,
primumcumpa,ente- nam meo iudiciopietasfandamentumest omniumuirtutum-
quemuereturut deum, Cael. 18 reprehendistis apatrequodsemigrarit.quodquukm in
hac aetateminime reprehendendum est. qui cum et ex publica cau.saiam esset mihi
quidemmolestamsibi tamm gloriosamuictoriamconsecutus et per aetatemnuigistratus
petereposset,non modopermittentepatre sed etiamsuadenteab eo semigrauit.
i Cf. Cicero, Cat. 3. IS quaesupplia,tiosi cumceteris suppliwionibusconferatur,
hoeinterestquodceteraebentgesta,hate UM conseruata republia, constitutaest,Pis. 97
sin autemaliquidsperaueras, si cogitarasid quodimperatoris nomen,quod1"ureati f asas,
quodiliatropaeaplenadedecoris et risuste commentatum essedeclarant,quiste miserior,
quis te damnatior,qui nequescriberead senatuma te bent rempublicamessegestam
nequepraesensdicereaususesi Plautus parodies the phraseology at Amph. 195-6
me a portu praemisitdomu,nut haecnuntiem uxori suae I ut gesseritrem publicam
ductuimperioauspiciosuo, 523-4 operamhancsurrupuitibi, I ex meprimo ut prima
sciresrem ut gessissempublicam.
3 C£ Aeschylus,Prom.473-5 KCXKOS s•lcrrpos~TIS ls v6aov Imacl>va8vµETS
KaiO'E<XVTOV 0\11<fxt1s IEVpElV cmofo1sq,apµCXKOIS laa1µcs, Euripides, fr. 6 I µ1aw
aofov {6VT•)w'-6yo1cnv,ls s•6VT')cnv ov aoq,6v, fr. 905 µ1aw aoq,10'T,iv6o-ns
ovxavrci'> aoq,6s (= 'Menander', Monost.457 Jakcl), Plato, Hipp. meiz. 283B
TOVaoq,ov a\JTO\Iavrc;, µ~1CTTaSETO'OfO\Idv<XI,Isocratcs, Ad Nicoclem52 6
36.2
 MEDEA EXVL; MEDEA
219patria procal: the preposition is absent only here in republican
drama; contrast Plautus, Capt. 551 proin tu ab istocproculreced4set al

220 qai domi aetatem agerent proptereaauntimprobati: W esenberg

wrote quod;Vahlen suggestedtentativelyquia,thinking no doubt of palaeo-
graphy and Plautus, Capt. 174-5 quiamihistnatalisdiesIproptereaate uocariad
te adcenamuolo,Mil. 1257 quiame atnat proptereaVenusfedt eamut diuinaret.
But relative and principal clausestend to be loosely arranged in old Latin,
particularly where the relative precedes (seeabove on vv. 177-8, below on
v. 228).

221 neqaiquam: avoidedby Caesarand Cicero, preferred tofrustraby the

classicalpoets but, according to Wolffiin (ALL n [1885], 9) "in der archa-
ischen Literatur... keinen bestimmtcn Charakter tragt •. The situation is in
fact far from clear.Tragedy has nequiquamtwice,frustra not at all. Terence
has nequiquamonly once (Haut. 344),frustraII times. Plautus has nequiquam
24 times andfrustra 24 (including the locutionfrustraesse14): both words
occur preponderantly in musicallyaccompaniedverse,Jrustramore so than
nequiquam.It therefore looks as ifin Ennius' day nequiquambelonged fully to
the common language but very soon after dropped out, to be employed
henceforth only by poets and poeticisinghistorians.

CVI

These two trimeters must come from Enniw' adaptation of the conversation
between the paedagogusand the nurse, Euripides, Med.49-95. They reflect
the wording of vv. 56-8:
fyooyap Is TOUT'~11K' &hY1166vos,
wo6' I~ µ• V'ITfi).8£yij TEKOVp<XVCf>
M~CXI Sevpo&O"tTolVfls
µo>.oVC71J ivxas.
Ennius adds only the word play miseram. .. miserias.

222 cupido cepit •• •me: republican drama has cupidoquite rarely (not
elsewherein tragedy, 5 times in comedy) compared with lubido.For Ennius'
locution c£ Terence, Hee. 88-9 edepolte desideriumAthenarumarbitrorI
Philotiumcepissesaepe.

µ116w~ avTCf> xpficnµcs ov6' &, c».Aovq,p6vtµov,ro11')a&1EV, N.T. Mark,

Euang. 15. 31 c».Aovslac.oow,mvrov ov 6V\la'TCX1
aooaCXt,Plautus, True.49S~
sineuirtuteargutumciuemmihihabeampropraejic4,I quaealioseollaudat,eapsesese
uerononpotest,Pacuviw, Trag.348 odi egohominesignauaoperaetphilosopha sen-
dareI stultumesse,Ovid,
tmtia,Phacdrus 1.9. 1-2 sibinoncauereet aliisconsilium
Ars I. 84 quiquealiiscauitnon cauetipsesibi.
363
 COMMENTARY

proloqui: compare Virgil, Atn. 2. 10-11, 6. 133-4 andcontrastLivy 1 .6. 3

Romulum... cupido cepit... urbis condendae, 32. S. 3, 33. 38. II, 38. 16.4,
40. 21. 1. Grammarians 1 dispute about the extent to which the Ennian type
of expression is a conscious graecism where it occursin Augustan poetry. In
the passage under discussion Ennius is unlikely to have retained the Greek
syntax if it was grossly inconsistent with contemporary Latin usage.
One findsin fact a state of apparent anarchy in comedy where the infinitive
and the gerund arc concerned. Plautus has occasiotst with the infinitive
3 times, with the gerund 4; tempus tst with the infinitive 3 times, with the
gerund 6.

223 Medeaimiseria1: Turnebus' emendation restores the iambic rhythm.

The manuscripts ofCicero's writings frequently modernise archaic forms in
quotations of dramatic verse: e.g. De orat. 2. 193 txtinxisti, Off. 3 .98
percepisset,Tusc. 3 . 26 socero,3. 28 sdebam.The manuscripts of Plautus'
comedies either corrupt the genitive in -ai where it occursor modernise it to
-ae.i
The genitive in -aeappears to have been normal in tragedy. Epic had -ai
{Ennius, Ann. 16, 33 bis, 119, 191, *203, *343, *489) and -as (see Priscian,
Gramm.11 198. 8 ff.)quite frequently as well as -ae. Plautine comedy had -ai
occasionally in formulaic and stylised passages.3

CVII

In her denunciation of Jason (vv. 465-519) the Euripidean Medca speaks of

their relationship as one of cp1A{a(vv. 470, 499) and implies that she had
thought him anaya8bs&vfip (vv.465,488,498,518,586,618). She mentions
in passing the xap1s (v. 508) she had shown him.Jason accepts the definition
of their relationship (vv. 549, 622) but argues that it was not so much the
socially honourable feeling of xap1s that motivated her actions as sexual lust
(vv. 527-31 KVTTptv voµ{300Ti\s !µfis VCM<Aflp{as I ac.:ne1pavelvoo&wv
s·
n Kav6~,roov µ6vT)v.1aoi gOTI1,1EVVO\JSAE1TTC,S-aAA' rn{~vos I
A6yas 61u.8etv, oos •Epc,.>S a' f)vayKaae I T~ots acpVKTOtSTovµov
ti<ac:;,aoo SEµas, SSS, 568-75). Medea interprets Jason's actions similarly
(vv. 623-4). Neither party makes anything of Medea's juridical status4 and
the chorus refers to Medea as Jason's &Aoxas (v. 578).

1
Cf. Hofmann-Szantyr, Lat. Synr. u. Stil. p. 351.
1 , pp. 342 ff.
:a See Leo, Pl. Forsch.
3 See Leumann, MusH n (1945), 253 n. 37, IV (1947), 121 (= Kl. Sehr.
125 n. 2, 143).
4 But cf. ""· 591-2 OUTOVT6a· Elxev,&Ua ~aptxxpovM)(oSI ,rposyi\pas
ovx ev6o~ov i~~a1vt ao1.
 MEDEA EXVL; MEDEA
Ennius' trochaic tetrameter with its antithesisand rhyming metra 1 reflects
the stylistictraditions of the Roman stage and the ethical notions of Roman
society as much as it docs Euripides' trirneters and their substance. For the
fifth-century Athenian xap1s was the attitude of a man volunteering to do
something to which he was not bound.3 This for the Roman was benignitas or
beneuolentia rather than honos.Cicero distinguishesthe two attitudes at Off.
2. 21: quaecumque igiturhomineshominitribuuntadeumaugendum atquehonestan-
dum, aut beneuolentiae gratiaf aciunt,cumaliquade causaquempiamdiligunt,aut
honoris,si cuius uirtuttm suspiciuntquemquedignumfortuna quam amplissima
putant,aut cuifidem habentet benerebussuisconsulerearbitrantur, aut cuiusopes
metuunt,aut contra,a quibusaliquidexspectant,ut cumregespopularesuehomines
largitionesaliquasproponunt,autpostremopretioacmercede ducuntur.Plautus and
Terence normally use the word honorin thissubjectivesensewhere personsof
comparable social status are concerned, e.g. citizen man and wife: Plautus,
Stich.48--9noloegosorormecrediesseinmemorem uiriI nequeilleeoshonoresmihi
quoshabuitperdidit,Terence, Haut.687 quamegoscioessehonorequouisdignam.
A tone of irony or high-falutin banter is to be detected in contexts like that of
Plautus, Mil. 1074 ff., True. 589 ff. The formulae honoriscausaand honoris
gratia3 should not be identified completely with the Attic xap1v TtvOS,els
xapw,xap1Tos 4
WEKa.I suggest therefore that Ennius made his Medea
speak of the honorshe paidJason in adaptation not of Euripides,Med. 508 aol
xap1v~povaa but of the whole tone of vv. 465-519. Ennius' language
assimilatedMedea's status to that of uxor (yvv,'\ycxµE'TT}), whereas in legal
reality she would have been only a concubina.S

224 amoris magis quam honoris ... gratia: cf. Cicero, Pis.65 conuiuium
publicumnon dignitatiscausainibit... sedplane animisui causa,Seneca, Contr.
1 . 6. 9 {Latro) puellam non misericordiamotam sed libidineet ideo non esse
benefidum.
CVIII

Behind Ennius' tetrametcn lie Euripides, Med. 364-75 :

KaKG°>s • Tfs 6:vnpEI;
1rhrp<xKTcn 1TCXVTCXX1J
&XJ..·oCm TCXVT1J '11)6o1<1:tn1TOO,
TCXVTa,
fr• eta• &)-&'>VES
TOISwc.oo-rlwµ,{01s
Kai TOiat KT1ooiaaatv ovaµncpol 1r6vo1.
yap &, µa T6v& 8c..mw<7cx{ 1TO'TE,
001<1:lS

1
See above on v. 8. :a See Fraenkel on Aeschylus, Ag. 182 f.
3 Plautus, Amph. 486, 867, Asin. 191, Aul. 2s, Mil. 620, Poen.638, Stich. 338,
Terence, Phorm.928.
4 & is done by F. Mehmel, Philologusxc (193s), SOSf.
S Cf. Hecuba's description of Cassandra at Enniw, Trag. 181.
 COMMENTARY
El µfiTt 1<lp6afvovaavft TEXVCA>JJM'IV;
ous·&II npocnt,rov ous·&II fi~µ11v XEpotv.
6 s· Is TOCTOVTOV µc...,p{as
&q,{l<ETO,
~· f~ov av-rc;,T&µ•lMlv j3ov>.Evµa-ra
yi\s ~6VT1, T{ivs· &q,i\KEVfiµtpcxv
µElvafµ•, fv f.lTpElSTOO\I fµoovfx8poov VE1<p0VS
efiaoo,ncrripa TE Kal K6p11vn6aw T• fµ6v.
and 399-400:
s· fyw 0'1)1\1 KCXl
fflKpOVS ).vypovs efiaooyaµovs,
mKpOVst Ki\Sos xeovos.
Kai q,vyas t1.1as
Ennius imports the sententia- qui uolt quoduolt ita datse resut operamdabit,
the metaphors mentetrauersaand repagulaquibus. .. recludam,and the elabor-
ately arranged pair of antitheses mihi maeroresilli luctum,exitium illi exilium
mihi.
· Cicero breaks off his quotation of the tetrameters at one point, expecting
hisreaders to fillin the gap for themselves.1 Vahlen restored the beginning of
the third tetrameter with the almost certain ni ob rem, and placed the query
mark immediately afterwards. However heavy pauses are rare after the first
trochee in Ennius' tetrameters and in those of the other third and second
century playwrights. Ennius' mode of expression is usually much fuller.
I suggest therefore that with parumnt ratiocinariuideturet sibi ipsa neJariam
pestemmachinari?Cicero twists the content, if not the wording, of the re-
mainder of the tetrameter into his own argument about the potential evil of
reason.3 In any case Cicero shows no tendency to break quotations at pauses
of sense where he does not break them at pauses of metre.
Language like quiestuersusomniumseminatormalorumusually appears when
Cicero quotes full metrical units (Fam.13. 15 .2, Diu. 1. 17, 2. 12, 2.25, Nat.
deor.3 .79, Rep. 5 .1, Off. 3 .82, Tusc.1. 1o6, 2.26, 4.63, Fin. 2. I05, Cato 16)
or the beginningof one (Att. 2.19. 3, Cato 1). But qui uolt quoduolt ita datse
res ut operamdabitis a trochee short either at the beginning or internally.
Three fifteenth-century codices supply an esseafter the first uolt but this
destroys the figure of speech.A semperbefore se doeslessdamage to Ennius'
rhetoric and its loss in the tradition is more easily explained.4

225 nequaquamiatuc iatac ibit: ~ OVTI "TCXVTIJ Sic abireseems

"TCXVTa.
to have been the normal phrase (c£ Terence, Andr. 175, Catullus 14. 16,
Cicero, Att. 14. 1. 1, Fin. 5. 7, Seneca,Herc.J.27).

1
Cf. frs. xv b, xvn a, xxvn a.
a Ind lectt.Berlin1877, I ff. ( = Op. ac.I 34 ff.).
3 Cf. the manner of quotation at Sest. 121.
• 0. Skutsch, in NauiculaChilonimsis,110 ff., inserts the transmitted six and
a half trochees directly after ni ob rem.
 MBDBA EXVL; MEDEA

magna inest certatio: 'there is plenty of fighting ahead'. Verbal nouns

in-tio are usually accompanied by est; for the use of in- to form poetic com-
pounds see above on v. 26.
nam ut ego illi supplicarem tanta blandiloquentia IDi obrem:
21f>-7
~ 6o1<elsyap&v l,1ET6v6e8CA>1TEVcrcrl
1TOTe, I el µfiTt 1<ep6a(vovcrav
fi
T£XVCl>IJM1V; Ribbeck's illi is enforced not by Euripides' T6v6ebut by the
ilk of v. 229 and the illi . .. illi ... illi of vv. 230-1. Vt frequently introduces
questions about impossible or incredible acts and situations in the dialogue
of Latin drama; cf. Plautus, Baah. 197-8 egonut quodab illocattigissetnuntiusI
non impetratumidaduenientiei redderem?,375, Mil. 962, Rud. 1o63, 1244, True.
441, Terence, .Atulr.618, Phonn.304- The normal mode of introducing such
questions was plainly with egoneut but Ennius perhaps wished to emphasise
illi rather than ego.
Blandiloquentiadoes not occur elsewhere in archaic or classicalLatin.
Plautus, Trin. 239 has blandiloquentulus and Laberius, Mim. 1o6 blandiloquens.
-loquentiatype nouns are rare: Plautus has at Rud.905 uaniloquentia,Trin. 222
stultiloquentia,Novius, Alell.38 tolutiloquentia,Trag. inc. 110 superbiloquentia,
1
Cicero, Orat. 191 and Fam. 13. 15 .2 magniloquentia, Brut. 58 suauiloquentia.
They must have been created originally by the adapters of tragedy.
Forni ob remcf.Terence, Phonn.525-6 nonpudet I uanitatis?--minime dum
ob rem, Sallust, lug. 31. 5 idf,ustra an ob remfadam in uostramanu situm est
~irites.

228 qui uolt quod uolt ita dat semper se res ut operam dabit: cf.
Caccilius, Com. 29o-1 Jae I
uelis: perfaies.
The figuration of qui uolt quod uolt is common in Greek' and Latin3
dramatic writing. For the sentiment cf. Cicero, Att. 14. I . 2 de quoquidemilk
ad quemdeuertiCaesaremsolitumdieere:'magni reJerthie quid uelit sed quicquid
uolt ualdeuoh'.
For ita datse rescf. Terence, Hee. 380 omnibusnobisut resdantseseita magni
atquehumilessumus, Cicero, Att. 3 .23. 5 ut se initia dederint.
Dat has a future reference; cf. Ennius, Ann. 100 nam mi ealidodassanguine
poe,uu.
For the lack of dose syntactic connection between relative and principal
clause c£ Ennius, Trag. 256 ea libertasest qui pectuspurum etfirmum gestitat,
Plautus, Asin. 323 em ista uirtusest quandousustqui malumfert, Terence, Hee.

1
Cf. Seneca ap. Gell. 12. 2. 7.
:aCf. Aeschylus, Ag. 67, 1287, Choe. 780, Bum. 679, Sophocles, 0. T. 1376
O.K. 273, 336, T,. 1234, Euripides, El. 85, 289, Med. 10n, Or. 79, Tr. 630.
3 Cf. Plautus, Epid. S54 meminiid quodmemini,Most. noo quodagasid agas,
Trin. 2+2 qui amatquodamat {cf.Mere.744), Poen.874 qui homoeum noritnorit.
367
 COMMENTARY

608 istucests11pere
qui ubiquomqutopussit onimumpossitjlectere,Com. inc. 76
onusest honosqui sustinetrempublicllffl,Cicero, Leg. 2. 19 qui stcus foxit tku.s
ipseuindexerit.

229 i1le trauersa mente: ~

6 s· fs TOO'OVTOV µ(A){)(cxs&q>h<ETO. For
Ennius' metaphor cf. Cato ap. Gell. 6. 3 . 14 secundaeres laetitiatransuorsum
truderesolent a recteconsulendoatque intellegenJo,Seneca, Epist. 8 .4 coepit
transuersos
agerefelidtas.

229-30 mi hodie tradidit repagalaI quibus ego iram omnem

~
recludam: -n'tvS'aq>f\KEV flµipcxvI µetvat µ•,a,~- ... The Euripidean
Medea refers to her anger later in the speech: Med.395-8 ov yap... xatp<a:>v
TlS avrCJv TOVµov&Aywet dap. It is a major theme of the play (vv. 91.
93 f., 99, 172 f., 176 f., 26o f., 271, 395 ff., 446 f., 520, 589 f., 615, 870, 878 £,
898,909).
Repagulaare regularly the ban placed across the leaves of a door on the
inside (Cicero, Vm. 4.94, Diu. 1. 74, Apuleius, Met. 1.14. 1, Paulus, Fest.
p. 351. 3 f. repagulasunt quaepatefadendig,atia itafiguntur ut e contrarioop-
pangantur.haecet repagesdicuntur).It is difficult to interpret closely Ennius'
metaphor. The same area provides Plautus' quite intelligible metaphor at
True.6o3 meamqueiramex pectoreiampromam.

230-1 atque illi pemiciem dabo, I mihi maerores, illi lactum,

~
exitium illi, exilium mihi: TpeTsTOOV {µoovtxepoovVEKpOVS&fiaoo,
iro:ripa TE Kai K6pr)v ,r6atv T' {µ6v ... fflKpOVS S' fycb aq>tv.... The sur-
face meaning of the Greek isbelied by the action of the play. Only the king's
daughter will be destroyed. Jason will not die; his children will. If Cicero
reports Ennius' Latin correctly, Ennius substituted vagueness and confusion
for dramatic irony. Maeroresmust allude to the plan to kill Jason's children
but this has nothing to do with the grief she is going to imposeon Creon.
One might assume a lacuna1 in Cicero's quotation after daboand take the
next verse as referring to the vengeance to be wreaked on Jason much more
explicitly than do Euripides' verses. The Greek play does not refer to the
killing ofJason's children until v. 792.
The nouns pernidesand permitiesare confused in Cicero's manuscripts as
they often are in those of Plautus. I see no way of sorting out the confusion.
The phrase pernidemdaremust have been based on malumdare,a phrase
which refers regularly in comedy to the punishment of slaves. C£ Seneca's
letumdare(Med. 17-18).
Maerornormally occurs in the singular in both tragedy (5 times) and
1
Cf. the passage discussed above, pp. 356 f..
 MEDBA BXVL; MEDEA

comedy ( 14 times). The plural occun elsewhere at Plautus, Capt. 840, 8-41.
Epid. 105.
Lucius occun 9 times in tragedy and only twice (Plautw, Vid. &. 2,
Terence, Hee. 210) in comedy. This perhaps reflectsa difference of subject-
matter rather than one of style.
CIX
Behind Ennius' trochaic verse1 lie the trimeters, Euripides, Med. 248-51:
~0V01 ws
s• i)µas &Jdv6wovJ3lov
3&>µevK<XT• otKOVS, ol 6~ µapvav-rat Sop(•
KaK&;scppovoOvTEs ws
· Tpls av,rap• aa-rrl6a
80.01µ• av µ~ov ft T&lCElV
O'TT}VCXt arra~.
3J2 111bannis: in thefust century certainly and in thesecond probably in
armisand armiswere the regular expressions of the common language (see
T.L.L. II 597.26 ff.); only the poets and the historians (Anon. Bell. Aft.
42.2, Livy 9.37.-4, 28.15.4) used subarmis.
uitam cemere: a most peculiar phrase, paralleled only at Ennius, Ann.
195-6 non C4Uponantesbellumsedbelligerantes,Ifmo non aurouitamcernamus
utrique.The verbs certllre,decertare, when used in the
cernereand decernere,
present manner (seeabove on v. 166),are regularly accompanied by de and
the ablative; c£ Cicero, Q!inct. 43, Phil. II .21, Att. 10.9.2, Virgil, Aen.
12. 765 et al.
ex
These words come from Ennius' adaptation of the first strophe of the
docbrniac prayer uttered by the Euripidean chorus as Medea leaves the stage
to kill her children: vv. 1251-60:
tcl>ra n Kcxl,raµcpcrl\s
mls ·1wt.,ov.K<XTf6n·t&-rl TCXV
6).oµwav ywcxtm. irplv cpotvfav
nKVOtS,r~cwtv x4'·CXVTOICTOVOV
tacxsyapa,roxpvcmxs
yovcxs
fl3).acrm,, 8eo0 s•cxlµaTt irhvetvt
cp6f3osw· avtpc.w.
d:AAa vtv, w cpaos6toyEVts,1<6:ratp-
y& KaTaTI'CXVC,OV, -~· ott«a>vTaACXt-
vav cpovlavT. •Eptvw VTI'CXACUM'OpOV.
Since thetime ofVossius scholars have usually wanted to arrange Ennius'
words in catalectic trochaic tetrameters. 2 At least two alterations to the
1
0. Skutsch. however, emends to produce trimeters: nam ter sub armis
nuJlimuitamcemereI quammnll modoparire.Sec HSCPh LXXI (1967). 137.
a Osann, Anal. crit. p. 1.23,produced Ji crctic tetrameten.
24 369 JTO
 COMMENTARY

paradosis, in places where sense and Latinity appear quite sound, are needed
to produce the required tetrameten. Furthermore it seems unlikely that a
stage adaptation which turned the relativdy calm trimeters 364-409 into
stichic tetrameters should use thesame form of verse as a replacement of the
excited dochmiacs 1251-70.
Ennius' words can be dividedinto three rhetorical units. The units of
verse in republican drama,particularly in theearly period, tend markedly to
coincide with units of rhetoric. Any attempt to analyse the rhythm of Ennius'
words must take account of this tendency.
The first unit, luppiter tuqueatko summeSol qui res omnis inspicis,1 can be
treated as a cretic trimeter followed by a catalcctic trochaic dimeter. In pro-
posing this scansion Strzelecki3 compared Plautus, Epid. 330 is nummum
nullum habesneesodalituo in te copiast,which appears in a dialogueof para-
tragic characteralong with other types of crctic verse, trochaic verses and
iambic.
The second unit, quiquetuo luminemareterramcaelumcontines,is metrically
obscurebut sound in scnse.3Strzeleckiscanned the words transmitted as two
catalectic trochaic dimeters.
Thefirst part of the third unit, inspieehoef acinuspriusquamfit,4 formsa pos-
sible trochaic dimetcr.S The second part is corrupt. Bothe's prohibessissalus
satisfiesthesense of the whole passage, the stylistic levcl 6 and palaeography.
The metre however remains obscure.
In adapting the Euripidean ode Ennius preserved the salient features of its

1
Vahlen's spicis(RhM XIV(1859], 566 [ = Ges. phil. Sehr.14o6]) destroys a
common type of sentence arrangement in which a verb of the preceding rela-
tive clauseis taken up in the principal (sec above on v. 8). Where co-ordinate
clauses arc concerned simple verbs frequently pick up their compound forms
(sec above on fr. IX); less frcquentlycompowid picks up simple (e.g. in such
plconastic doublets as Plautus, Amph. SSI sequo,subsequor,Poen. 221 poliri
expoliri,223 lauandotluendo,Caecilius, Com. 212 ploroatqueinploro,Terence.
Eun. 962 dicoedico,Ennius, Trag. 337 nequepati nequeptrpeti).
3
'&s XLIIfasc. 2 (1947), 98 n. 52, in Tragica1, SS,
3 Vahlcn's quiquetuo cumluminespoils the sense; likewise Leo's qui igneotuo
lumine(Dt Trag.Rom. p. 14 [ = Ausg. kl. Sehr.12031).
4 The indicative.fit is unexceptional in early Latin; cf. Plautus, Capt. 831-2,
Cure.210, True. IIS, Terence, Phorm.1037.
5 Priusquamis much more often an anapaest than a bacchius in republican
drama; sec Lindsay, E.L. V. p. 212.
6
The form occurs only in apotropaic prayers of great solemnity: Plautus,
Aul. 6n, Pstud. 13-14, Cato, Ag,. 141.2, Cicero, Ltg. 3 .6, 3 .9, Contrast
Terence, Amir. 568 quoddi prohibtant,Haut. 1038, Hee. 207, Ad. 275. The only
other -ss- form of a second conjugation verb in republican drama seems to be
at Plautus. Asin. 6o3 (liassit).
370
 MEDEA EXVL; MEDEA
grammatical framework, the address to the Sun, the imperative verb of
seeing,the adverbial referenceto the aime to come and the imperative verb
of preventing. However, while Euripides made the Corinthian women give
expression to traditional religious ideas still valid for at least some mcmben
of his fifth-century Athenianaudience, Ennius put in their mouths an amal-
gam of traditional Roman religious thought and Hellenisticphilosophical
speculation. 1 For the special poeticvocabulary of Greek choral lyric &mus
substituted the words and formulae of the Roman sacrallanguage.
A major theme of the Athenian tragedy was the puoisbment of Jason for
perjury through the destruction of his children.Hence the constant com-
plaints by the heroine and her friends ccnteringon Jason's breaking of his
oathratherthanhisactoftakinganewconsort (vv.21-2, 161-3, 168-70, 208-
9, 439, 492, 1391-2) and the elaborate desaiption of the binding of Acgeus
(vv. 734-ss). At the culminating point of the drama, as Medea enters the
palace to carry out a just act of vengeance, the Corinthian women beseech
Earth and Sun to prevent her. Theseare the deitiesby wmch not long ago on
the very stage Mcdea made Aegcusswear and by wmch, we arc given
implicitly to understand, on the fatal day in Colchis Jason himself swore.
They are askedto remove an "Ep1vvs from the house. This "Ep1vvs is not
Medea herselfbut a real demon, one of those who in the world of epic and
3

tragic poetry3 (probably too in that of Euripides' audience) were supposed

to punish perjury among other misdemcanours. She is urging Medea on to
destroy Jason's childrenin punishment for Jason's aime rather than waiting
to punish Medea for the act shehasnot yet committed. This act, a aimc as
well as an act of justice, will be dealt with by another 'Ep1vvs. 4
Sophisticated Greeksof the second century, even in Athens, would have
found the ideas underlying the words of Euripides' chorus hard to fathom
and there were no precise parallelsin Roman religious belie£It is not sur-
prising that Ennius did not try to reproduce them. In any caseit is possible
1
Scholarly discussion has conccmcd itself solely with the replacement of
Earth by Iuppitcr. According to Ribbeck (Dit rom. Trag.p. 1s7) 'dcr Name
Iuppitcr war popularer als die uralte Mutter Erde ... •. G. Herzog-Hauser
(Comment. Vindob.1 [193s], 48) addedsome psycho-analytical sophistication,
suggesting that Ennius' preference for the father-god Iuppiter over the
mother-god Earth rdlects the patriarchal structure of Roman society. R.
Goossens (Latomusv (1946), 288-91) approached the same narrow question
from another angle, suggesting that in Ennius' text of the Greek play stood the
word 6a (= ya) and that this was misW1derstood by Ennius, who learnt
the West Doric dialect of Greek in Tarcntum, where Aa would have been the
equivalent ofbv. a So one of the scholiasts; contraPage on v. 1260.
3 Cf. Homer D. 19. .258-65.
4 Cf. Jason's curse, vv. 1389-90 ~6: a' 'Ep1vvsWatlE TacvCAlV j IP()viaTE
Ahcfl.
371
 COMMENTARY
that he had abandoned Euripides' subtly sympatheticview of Medca and
presented her simply as a foreign-born concubina opposing her master's
arrangementsof his own affairs(seeaboveon frs. cv, CVII, below on v. 237).
His chorus appearsto be requestingIuppiterprodigialis(cl Plautus, Amph.
739) and the Al&t'\p in which thisdeity dwellsand whosenature he sharesto
make somesign by which the Corinthiansmight be warned of Medca's evil
intentions and thus enabledto prevent salusaffectingtheir community. In
his epic poem on his own consulshipCicero writes of Iuppitermuch more
explicitly in terms of philosophicaltheory (Cams. fr. 3. 1-s [Diu. 1. 17))
-principio aetherio jlamnumu Iuppiterigni I uatitur et totumconlustratlumine
mundumI mtnttfUt diuinaaielumterrasque ptttssit, I quaepenitussensushominum
uitasfUtretentatI aetherisaeternisaeptaatqut inclusacauernis-beforcdescribing
the prodigiasent to warn Rome of the troubles in store for her.
The substanceof inspia hoef acinuspriusquamfit ( ~ Kcrr{Sn-•16m: TCXV I
6Aoµa,av ywcxtKa, irpiv q,o1v{avI TS(VOtS irpoa~lv x'P•avro-
acr6vov)loob on the surfaceto be absurd. But as early as the time of the
composerof Od.20. 3sI ff.1 gods and human seerswere wont to see future
eventsas if they were happeningin the presentbeforetheir very eyes.Many
philosophersregarded all events,past present and future, as linked together
and capable of being known by an intelligentdivine cosmic principle; in
their view human seers foresaw the future in so far as their minds were
connected with a principle which observed the eternal process unaffected
by temporal divisions.Some philosophersidentifiedthisprinciplewith the
lower atmosphere and their theory lies behind Philemon's prologising-
•Af\p: fr. 91 &, ovS~els MA118Ev ov& Iv lt'OlWV I ovs· cro 1T01T}CTOOVov~
1Trn0lfl~ 1TCXAat, Iovn 8e6sov,-' &vepc,.,,ros,OVTC,S elµ' fy~. I •Af\p,
&, &, TIS 6voµaCJEtEKai ~{a. Other philosophers,perhaps Hcraclitusand
Empedoclesand certainlythe Stoics,made the fiery upper atmospherethe
sourceof propheticknowledge:1 inspicehoef acinuspriusquamfit would be a
natural prayer to addressto the AIEh')p of Stoic theology.
Probus took the relative clause quifU,ttuo lumine mare terramcaelum
continesas co-ordinatewith quiresomnisinspicisand thus qualifyingSol.3 The
verb contineswill hardly apply to the normal activityof the sun4 and for this
1
See aboveon vv. 43-4.
:aCf. Ps.Hippocrates,De camibus2 60K£el6! µ01aKaMOl,IEV 8epµov&eavc:nov
TEelvenKai V08:IVmwra Kai opf\v Kai aKOUEIV Kai el6wat ,rav-ra !6VTaTEKai
la61JW(X. ToOToow To ,r}.eta-rov,6-rehaPCXXEhi arrcnmx,!~wpT\cm, els TI')v
avc.na-roo mp1q,opfiv,KCXlcnrr6 µ01 6o1<£E1
aie!pa Tots ncv.aiols elpf\a8at.
3 Cf. Plautus,Amph. 677-8, Capt. 271, Men. S49, 1133-4, Mil. 1229, Poen.
1189, Rud. 28-9, 128~, Terence, And,. 481, Haut. 444-5.
4 Vahlcn adducesEnnius, Ann. 542-3 quifulmine claroI omniaper sonitus

arcet,ttrrammarecaelum.But the antecedenthere was plainlyIuppiteror aether.

372
 MEDEA EXVL; MEDEA
reason Janus Rutgenius 1 conjectured contuis.On the other hand the clause as
transmitted makes an excdlent description of the al8rtp,the outer dement of
the Empedoclean four. 2 I therefore suggest that tJUique
tuo luminemareterram
caelumcontinesis co-ordinate with Iuppiterand Sol3and refers to the diviniscd
al8rtp in an allwive manner common in ancient prayers; c£ Homer, n.
3. 27~ Zrii ,rarep... 'H°'t<>s ... Kaimrraµol Kaiya'fa Kaio1vrrivep8e
I
K<XIJMCXS&vepoo,rovsT(wa9ov (contrast 19.258), Aristophanes, Thesm.
I
315-19 Ze0 µey'CXA~Wµ& xpvcroAvpa'n ~fiAOV~ fxetslepav. Kai I
av1rayKpcrris ir6AtVoll1<ovaa,
KopayAav!KoomXPV0'6AOYXE mp1µa-
XflTOV, Virgil, Georg.1 . 12-15 tuqueo . •. Neptuneet cultornemorumcuipinguia
Ceae I ter centum niueitondentdumetaiuuenci, Seneca, Med. 1-4 tuquegenialis
tori I Lucina custos, tJUatquedomituramfreta I Tiphyn nouamfrenare docuisti
ratem, I et tu profundisaeuedominatormaris.
Probus' interpretation might be saved by supposing that Ennius identified
in a very bold way•HAtoS and Al8rtp.Varro remarks (Rust. 1.4. 1) that eius
(ie. agriculturae) mundi esseBnnius scribit,aquaterra
principiasunt eademtJUOt
anima et sol. These are plainly the Empedoclean four.4 Stoic theological
1
Variarumledionumlibrisex (Leiden, 1618), VI 1.
a Cf. Empedocles, fr. 38 yai<XTE Kai irwrcs iro~UKVµc..w fi6' \iypos aiip
I Thcxv fi6' al&iip O'f(yyCAlvmpl K\'n().ovOOTCXVTCX, Achilles,Eisag. 5, p. 36. 19
Maass ZftVCAlv yow 6 KtTIEVS .•. ovpcxv6slo-nv al8ipos TOioXaTOV ••• TOVTO
61Kai irlnrTa mp1txE1ir~1'tvcxvrov, Cicero, Nat. tko,. 2. 58 mundiquiomniaeom-
plexu suo eoercetet eontinet,101 restl#ultimuset a domidliisnostrisaltissimusomnia
cingenset coercenscaelicomplexus,qui idemaetheruocatur,extremaoraet tktnminotio
mundi,in quocumadmirabilitate maximaigneae formaecursusordinatosdefiniunt. .. e
quibussol,117 quemcomplexasummaparseaeliquaeadlseriadicitur.Euripides gives
no sign of acquaintance with the theory of four elements but he frequently talks
of the atmosphere as a whole embracing the earth: Ba. 292-3 ~ TI ToOxeov•
fyKvlcAovµa,ovI atetpes. fr. 919 KOpVcp1't 618Eoov6mp1~ x86v' fxCAlV I tcpaMloSt
at8r)p, fr. 9416~ TOV\J\VOV T6v6' &m,pov atetpa IKai yf\v ,rip,~ fxov8' vypats
Iv &yl<~ats; I ToOTovv6µ13EZfiva, T6v6' ftyoO 8E6v.Cf. Aristophanes, Neph.
264 ai'ip ~ fxe1s-n'ivyf\v µenc..>pov.Pacuvius, Trag. 86-7 hoeuidecircumsupra-
que quodeomplexueontinetI tmam, Lucretius 5 •318-19 tuerehoedrcumsupraque
quod omnemI c.ontinetamplexutmam.
3 Cf. Plautus, Cure.473 ibidemenmt seortaexoletaquiquestipularisolent,Stich.
4-s, Terence, Hee. 478. At Plautus, Baah. 1087, Cure. 480 et al. there is one
grammatical antecedent but two notional ones.
4 In a fragment of the Annales-521-2 corporeTartarinoprog,u,tapaluda

uirago,I cuipar imberet ignisspirituset g,auistmo-an Italian underworld demon

appears to be identified with the NetKoSofEmpedocles; sec Norden, Enniusund
Vergilius,pp. 10 ff., E. Bignone, RFIC N.S. VII (1929), 10 ff. ( = Studi sul
pensitroantieo,pp. 327 ff.), H.Frankcl, HermesLXX (1935), 62ff.; contra,W.-H.
Friedrich, Philologusxcm (1948), 291 ff.

373
 COMMENTARY
treatises,whilekeeping •H>.iosand Al&ftpdistinct, neverthelessdisputed as
to which should be identified with the Zeusof traditional cult.1

234 Iuppitertaque adeo 1111D1De Sol: for tuqueintroducing a further

addresseec£ Carm.deu. ap. Macr. Sat. 3 .9.11 Tellusmater tequeluppiter
obtestor,Virgil. Aen. 8. 71-2 nymph«. .• tuque o Thybri,Livy 1. 32. 10 audi
luppiteret tu lane~irine, Seneca, Ph«dr. 959-00 o ""'l"" parensNaturadeumI
tuqueigniferi rectorOlympi.Thesamearro 1(01VOV arrangementis common in
Greek dramatic prayers; c£ Aeschylus, Theb. 1~48, 151~5, Euripides,
Kykl. 350-4, 599-005, Hel. 1093-8, Aristophanes,Thesm.315-26.
For adeoin invocations c£ Turpilius, Com. 118-19 te Apollo sancte• ..
uosqutadeoumti, Virgil, Georg.1. 5-25 uoso clarissim4 mundiI lumiru2
..• tuque
adeo. .. Caesar.
For the epithet of Sol c£ Cicero, Rep. 6. 9.

qui res omnu impicia: for relative clausesin invocations of deity sec
above on v. + For the content of this clausec£ Homer, ll. 3. 277'H~16s e•
~ ,rav-r• A'°"~
rno Ii lEVCt.>V
1<al,rav-r' rn<XKOVElS,Aeschylus,Choe. 985~ 6 ,rav-r•
TaSe 1·H>.1os,Prom.91 TOVirav61miv KVl<AOV 'l't>.{ov
1<cwZ>.
o o
&. 192. 5 irav6irrcxs•H>.1os,Sophocles,O.K. 869 irav-ra MVaaoov
•H>.1os.

235 mare terram caelam: such tricola were probably common in the
formulae of the sacrallanguage; c£ Corm.aug.ap.Varr. Ling.7.8 conregione
conspicione cortumione,Corm. euoc.ap. Macr. Sat. 3 .9. 7-8 dque populo
ciuitatimttumformidinemobliuionem iniciatis,Cann.deu.ap.Macr. Sat.3 .9. 10
illllmurbemCarthaginem exercitumque Livy
. . .fugaf ormidineterrorecompleatis,
8. 2 5. 10 quodbonumf austumftlix Pal«opolitanis populoqueRomanoesset,tradere
se ait moeniastatuisse.For this particular tricolon c£ Ennius,Ann. 542-3 qui
falmineclaroI omniaper sonitusarcet,terrammarecaelum,Plautus,Amph. 1055
uidenturomniamareterracaelumconsequi, Trin. 1070 mareterracaelumdi uostram
fidem,Terence,Ad. 790o caelumo terrao mariaNeptuni,A&anius,Com.tog.9
marecaelumterramruereactremerediceres,Cicero, Fin. s.9 ut nullaparscaelo
mariterra,ut poeticeloquar,praetermissa sit (contrastS. Rose.131 cuiusnutuet
arbitriocaelumterram4riaque reguntur), Tusc.s.105 et in hoeipsomundocaelum
terrasmariacognoscimus, Ac. 2. 105 oculisquibusisteuestera,elumterrammare
intuebitur.

236 impice hoe &cin.111 priuaquam fit: for the singular verb following
a plurality of divine invocationsc£ Aeschylus,Prom.1091-3 & µflTp()S Aµfis
oi(:xxs,& irav-roovI al&i'\p1<01vov cpaoselAfaaoov,I~µ•~ flc611<a
1
Cf. Cicero, Ac. 2. 126.

374
 MEDEA EXVL; MBDEA
,raaxCA>;Sophocles, BI. 86-9 W cpaSo&yvov I Kal )'TlSla6µo1p"m'IP,
oos ... f.loeov,Aristophanes, Thesm.315-19 ZeOJJEYaAcl>wµa )(PVO"OAvpa
TE ••• Kat av... !Aee &opo.
The etymological figure f adnusf aare occun once elsewhere in tragedy
(Trag. inc. 64), is extrcmdy common in Plautus (Aul.220, Baah.6.41,682,
925 et al) but rare in Terence (only at Eun.6.44;
in trochaic tetrametcn).C£
Sophocles, 0. T. 1374 Cpya••• elpyaaµa,a, Euripides, Hile. 1072 Cpyov
l~1pyaaCA>.
prohibeuis acelm: the word scelushere denotes the state of religious un-
healthiness which will arise from the act rather than the act itself; cf. Livy
22.10. s si quisclepsitnepopuloscelusesto~ cui cleptumerit.

CXI
Ribbeck and Vahlen printed Mercerus' restoration of the second ofEnnius'
two trimeters-quid sic te extra aedisexanimataeliminas?F. Skutsch argued
in favour of Stephan.us' and Columna' s restoration-quul sic extra aedis
exanimataeliminas? 1
Lindsay's analysis of the composition of Nonius' lexicon
permits a more systematic restoration.
Nonius' lexicon has two articles on eliminare,one in the fint book at
p. 38.29-p. 39.7 illustrating the lemma extra limenticereand one in the
fourth book at p. 292. 20-33 illustrating two lemmata, exireand excludere.
Four quotations of drama appear in both articles. The source of the fint
article, both lemma and illustrative quotations, is Lindsay's list 27 'Alph.
Verb' (possibly list 26 'Gloss iii'). One should not expect to find disharmony
between lemma and quotations here. The second article appears in a book
whose lemmata are dictated by Nonius' own arrangement of the material,
one designed to show that individual words have a multiplicity of meaning!!,
Nonius twists the material provided by his sources, often in an extremdy
unintelligent manner, to fit his new lemmata. At p. 292.20 ff. he makesthe
quotations of Ennius' Medeaexul and Accius' Meleager, which at p. 38. 29 ff.
illustrated the lemma extralimeneicere,illustrate the quite different lemma
exire along with a quotation of Accius' Ph0tnissae.The latter quotation
comes from lists 'Accius i', the former from list 27.2
When composing p. 292. 20-33 Nonius must have imaginedhe could read
something like antiquaerilisfida custosan-poris quidsic extra aedisexanimata
1
Ciotta m (1912), 387 ( = Kl. Sehr. 487).
a Lindsay seems to me quite wrong in assigning the quotation of Accius'
Meleagerto list 27 at p. 39.5 and to list 8 'Accius ii' at p. 292.25, especially as
similar corruptions appear at both places. No significance need be seen in the
position of the quotation ofEnnius' Medeaexul at the head of the p. 292.20 ff.
article.
37S
 COMMENTARY
eliminatin list 27 and took the verb eliminatas intransitive and the noun
custosas its subject. But the lemma in list 27 was something like EI.IMINARB,
extra limenticereand could not have been effectively illustrated by such a
sentence; in any case the descriptive phraseology attaching to custosrequires
a noun in the vocative rather thanthe nominative case. I therefore propose
quidsicte extra«dis exanimotamtliminatas the second trimeter quoted by the
compiler of list 27, already corrupt in Nonius' copy, transferred along with
the lemma of list 27 at p. 38. 29 and placed under a different and inappro-
priate lemma at p. 292. 20. 1
Ennius' two trimeten have been regarded sincethe sixteenth century as an
adaptation of the addressof the Euripidean paedagogus to the nurse, vv. 49-
51:
irw.aiov l<Tf\~&cmofVl"ls4',fis,
ofKc.,w
Tf ,rposmi>.cnc,1fflw• 4yOVO'•
lpt}µfav
fflTl~, CXV'Tfl
8pEOl,IMlC7CX\ni;i
KaKa;
P. Maasi objected that the word exanimatahasno counterpart in the Greek
and would more aptly apply to the nune's reaction to the catastrophe of
1271-8. There is considerable weight in Maas's second point. Euripides
pictures the nurse as sympathetic with Medea (vv. 54-6, 78--9) and afraid of
where her anger might lead (vv. 37, 90 ff. et al.) but not distraught. Never-
theless Ennius' adaptations of the nurse's utterances employ language of
greater emotional colouring than the Greek {215-16 mans ... animoaegro
amoresaeuosauda, 222-3 miseram... miserias)and the same exaggerating
tendency may be in play here.
The present tense normally indicatesthat the personage referred to enten
the stage at the moment of speaking {e.g. Plautus, Most. 419 sed quid tu
egredereSplUJerio?},the perfectthat he or she has been there for some time
{e.g. Plautus, Amph. 1078--9 sed quid tu foras I egressaes?). Accordingly
Fraenkel argued3that the Ennian paedagogus was already on stage when the
nurse entered. For reasons which I have given above {p. 350) this could not
have been the case. But the link between the Latin trimeters and Euripides,
Med.49-51 need not be abandoned. Eliminas/-tcan be treated as a resultative
present along with Plautus, Amph. 368 immo equidemtunicisconsutishue
aduenio,non dolis,Mil. 1299 a matreilliusuenio,Most.440 trienniopostAegypto
adueniodomum,Turpilius, Com. 52-3 mecuraesomnosegregantIforasquenoctis
exdtantsilentio.4
1
If thetransmitted text of the quotationof Accius'Phoenissae,eg,edereexi ecfer
te eliminaurbe,is correct this quotation docsillustratecorrectlythe lemma exire.
i HermesLXVII (1932), 243 f. 3 HermesLXVII (1932), 355-6.
4 Sec G. Monaco, SIFC N.S. XXIV (1950), 249-53. Neverthelessit must be
admitted that eliminare,an emphaticallyperfective verb, could not have been
used thw in the common language.
376
 MEDEA EXVL; MBDBA
237antiqaa: ~,rcx).cnov;a solemnand honorific word when applied to
persons {Plautus,Baab. 261, Cure. 591, Terence, Phonn.67, Aprissiusap.
Varr. Ling. 6.68), but not markedly poeticallike the Greek fflWXlOS.

erilis6da CIUtol corporil: ~olKCA>V K"ri\µa &crno{Vfls Aµijs.The nurse

came into Jason's possessionalong with Medea. Ennius' use of the word
custosrather thanancillaisperhaps meant to emphasisethat Medeahasnot the
who needswatching.
statusof a matrona,that Jason regardsher as a concubina
Elsewhere in republican drama custosis used of the attendants of unmarried
young men1 and women2 of freestatus,of concubinae3 and meretriasof servile
status,4 never of matronae.
Fiduswas a word of great solemnity, not normally applied to slaves;see
above on v. 194-For its substancecf. Euripides,Med.821 is ,ra,,rra yap 61'
aoi TCX maTa X()OOIJE6a.
For the periphrasiseriliscorporis( ... erae)cf. the Attic tragedians' use of
Stµas {e.g. Aeschylus,Bum.84 KT<XVElv ... µT}TfXt>OV &µas, Euripides,I.A.
417, Med. 388, 531 et al.), Ennius, Trag.241 opti,nacorpora, Naevius, Trag.
21-2 uosquiregaliscorporis custodiasI agitatis,Accius, Trag.547 pinnigeronon
annigeroincorpore, Ennius, Ann. 93-4 ceduntdecaelota quattuorcorpora sancta I
auium,521 corporeTartarino prognatapaludauirago,Lucretius 1. 770-1 tnrae
.. .corpus,2.232 corpusaqu,ae,2.472 Neptunicorpus,Virgil, Atn. 5.318 ante
omniacorpora, 6.21-2 septenaquotannis Icorporanatorum, II .690-1 Orsilochum
et ButenduomaximaTeucrumI corpora, Ovid, Met. 3 . 58fidissirnacorpora.For
the stylistic level of the adjective, see above on v. 100; for that of erilis
see Lofstedt, Syntactica 11 , pp. 116 ff.

~
238quid ••• te ••• e1irnimt: Ti ... IOTT}Kas.Q!!idonly once elsewhere
in republican drama so far as I can see governs a transitive verb (Terence,
Eun. 162). Q_uestionsare normally introduced by quidadverbial or quidest
quod.But the high tragic style much affectsabstractnouns and incorporalia
as the subjectsof transitiveverbs; see above on v. 17.
The noun limenhad much more solemn associationsfor the Roman mindS
thandid ovSoset sim. for fifth-century Athenians.It occun in comedy only
in very formal contexts (Plautus, Cist. 650, Mere.830, Mil. 596, Most.1064,
1
Plautus, Asin. 655, Capt. 708, Mere. 92, Terence, Phorm. 287.
:a Plautus, True. 812.
3 Plautus, Mil. 146, 153, 271, 298, 305, 467, 550.
4 Plautus, Cure. 76, 91, True. 103.
s See K. Meister, 'Die Hausschwelle in Sprache und Religion der Romer',
SB Heidelberg,Phil.-hist. Kl. Abh. m (1924/5). On the verseof the Arval hymn
saturfa JereMars limensalista berberseeNorden, Aus altromisehmPriesterbuchern
{Lund. 1939), pp. 146 ff.

377
 COMMENTARY

Terence, Hee. 378), in tragedy once in an elaborate periphrasis (Accius 531

ablimintc«li).1 The verb elimitu11e occun three times elsewhere in republican
tragedy; according to Q_uintilian,Inst. 8. 3 . 31, inter Pomponiumac Senecam
.. . essetractatuman 'groduseliminat' in tragoedia Elsewherein
did oportui.sset.
extant literature it occun only at Pomponius, Atell.33, Varro, Men.459 and
Horace, Epist. 1. 5 .25. It was clearly a creation of the tragedians.

aic:~ Tfiv6' &yova• ipT}1,dav.

extra aedis: ~ ,rpos 'ITVA<XlOl.
Plautus 1w ex aedibusand extraportam
regularly. Terence's interdieont extuli.sseextraaedispuerumusquamuelis(Hee.
563) apes the style of public edicts. Pomponius' eliminaboextraaediseoniugem
(Atell.33) is plainly paratragic.
iP!:UDirnatam: ~ CXV'TT)
8pEOl,IMl
CTCX\Trij
k<XK<X.

CXII
This fragment comes from a play set in Athens; sec above, p. 344.
Elroslcy2 compared Ennius' two trimcters with the opening of Sophocles'
'HAaCTpa,a play set on the citadel of heroic Argosbeforethe palaceof the
king: 1-10 w TOVaTp<rn))'TlaCXV"t'oS I
w Tpo{qc 1TOTE'Ayaµi~wovos
,rat, wv mtv• (~ea,:{ ao1 I 1rap6VT1 AEVaae1v, wv irp68vµosfia8' &ELI
To yapir<XAcnov "Apy<>s ov,ro8e1sT66e, I Tiis olaTpo,rAf\YoS6:Aaos
s·,
·tvaxov K6pT}S'I CXVTTl'Opma, TOVAVKOkT6vov 8eovI ayopa
AVKEtoS' ov~ aptaTe()QSs· o6e 1·Hpas 6 KAElVoS va6s· ol 6' lKavo~, I
q>aaKEtvMVl<T}vas T~ 1TOAVXPVO'OVS 6pav, I 1TOAvcp8op6v -re Sc;.;µa
TTEA01r1oo.>v T66e.With thiscomparison in mindWilamowitz3 set Ennius'
play before the palace of Acgeusin the Kf\1r01south of the citadel. One would
expect the king's palace to be on the citadel In any case the imperative adsta
suggests that the person addressed is walking across the stage in front of the
audience and the phrase opulentumoppidumthat he or she 1w the entrance to
the citadel facingat the end of the road. Whether we interpret the two tri-
meters in terms of the topography of fifth-century Athens• or according to
1
Meister finds in Attic drama only Aeschylus, Choe. 571 aµe('flCA> ~cv.ov
{p,ce{CA>v,rv).G>v. a EuripidisMedea,p. 66.
3 'Burg und Stadt von Kek.rops bis Periklcs •, Philologisehe Untersuchungen 1
(Berlin, 1880), p. 128 n. 48.
4 The site of the 'E>.Naiv1ov TOvrroTij ai<pon6>.£1 (Clem. Alex. Protr.3. 45;
cf. I.G. n• 1078.14-15, 41) is now established with certainty on the north-west
slope of the •AKparro>.1s,that is on the left of the Panathcnaic Way as one pro-
ceeds from the Dipylon gate; sec T. L. Shear, Hesptriavm (1939), 207-12, IX
(1940), 268, E. Vanderpool, Hesptria xvm (1949), 134~. H. A. Thompson.
HesptriaXXIX (1900), 334-8.
 MBDBA BXVL; MBDBA

thenormal conventionsof theRoman stage1 we must imaginetheaction of

Ennius' play as taking place before the precinct of the underworld deities,
Demeter, Kore and Plouton. A better paralldis provided by the opening of
Sophocles'Ol6hrovs 6 rnl KoACA>V<t>, a play also set outsidethe citadd of
Athensand in front of a precinct of chthonic deities: 14,-16 ircrnp T<XAaf-
Ol6lirovs, 1TVpyo1~ ot I ir6A1v<Triyovatv,wscm·6µµ6:rc,.w,
TI'CA>p1
irp6aCA>·I xc1>pos s· 66' tp6s.
239 ~theaa• anticam opalentum oppidam: thewords oppidumand
urbswere dilfcrentiated in early second century Latin as mcp6,roA1S and
&o-rv.Plautus normally uses oppidumin contexts of real or metaphorical
siegeandoccasionallywhere the city as a whole is conc:emed(Men.73, Mil.
88, Poen.175, s6o,994-. 1403). Terence,however,twice (Amir.342, .Ad.715)
usesit clearlyof the residentialquartersandpublic places,in contextswhere
Plautususes urbs(Epid.195, 197, 719, Mere.175, Sos,Stich. 113).
With the alliterativephrase opukntumoppidumEnniusis perhapsplaying
with an etymology (c£ Varro, Ling. s.14,1 oppidumah opi dictumquoJ
munituropiscausaubisint) as well as with sound.I.ivy, who in other contexts
is as shy of opukntsuas arc Caesar and Cicero,employsthephraserepeatedly
(1 .2.3, 2.63 .6 et al.) along with urbsopuknta.
The epithetantiquuswould rdicct the feelingsthat fifth-century Athenians
had about their city; c£ Euripides, Med. 824, 'Epexeet6cnTO ir<XAcnov
6A~101.

240 templam Cereris: Enniusprobablyrefershere to the nµa,os

rather
than the va6s. Templumis never used unambiguouslyof thegod's atdesin
republicandrama.
~ 1
For templumCereris EAEvalv1ov c£ Plautus' Cererisuigiliae(Aul. 36,
79S ~ October festivalof El('usiniandeities).
CXIII
Of the five pieces quoted by Nonius as from Ennius' Medeathis has the
closestverbal and thematic parallelin Euripides' Mri&1a, Mcdca's farewell
to her childrenat IOO!r73: 66T", et>TEKVa,I 66T &airaaaa6cn µflTpl
1

6e~1avx~a. j Wq>IAT<XTfl xelp, cp{ATCXT0\16e µ01o-r6µa j Kai axi\µa Kai

evyeves
TI'pC>O'(A)TI'0\1 TEKVCA>V,I ev6cnµovolT0\1. There arc similar passages
in other plays (e.g. Euripides,Helt.4,Q!r-10, Tr. 757-63, fr. 362.32-3) but it

1
Travellers from abroad seem regularly to have entered from the audience's
left; sec Plautus, Amph. 333, Mm. SSS, RMd.156, Terence, And,. 734. The Attic
convention may have been different;sec Introduction, p. .20.

379
 COMMENTARY
is difficult to imagine one, especiallyone addressed to a plurality, in a play
about Mcdea set in Athens.
Nevertheless there arc difficultiesin the traditional identification. Salwte
alone is normally a term of greeting and the closestparallelswith the
Ennian fragment in Roman drama arc Plautus, Cure. 305-7 o mea oppor-
tunitasI Curculioexoptatesalue. . .saluomgauMo I tt adutnire.ado tuam mi
dexteram,Epid. 548-59saluasies. •.salue•. .ado manum. -ocdpt, Poen.1259-
61 hiepaterest uoster,datemanus.-salue insperaunobisIpater;tt complecti nos
sine.cupittatqueexspectate Ipatersalue(c£Euripides, I. T. 902-3, Ion 517-19,
Seneca, Thy. 508-9 ). In farewells salueseems always to be accompanied by
uale(Plautus, Asin. 592-3, Capt.744, Cist. 116, Cure.522, 588,Mere.830). If
the Ennian fragment docs come from a greeting and docs belong to the
Athenian Medeait is still difficultto give it a plausible context. The metre is
hard to establish and the words transmittedby the codicesof Nonius'
lexicon may be corrupt.

241 optima corpora:for this periphrasis see above on v. 237.

1f2 cette: 3 times elsewhere in tragedy against dateonce; only once in

comedy (Plautus, fr. 16o) againstdate46 times. There was more life in eedo
(128 times in comedy against Ja70 ).

CXIV
The descriptiverelative clausemakes it likely that we have here an address to
Sol rather than a piece of narrative. For the relative clause in invocations of
deity see above on v. 4; for the use of the third person verb in invocations c£
n.
Homer, 17. 248-50 & f°'OI. .. of. .. irfvovatv KOO Ennius,
CJTl'100VOV01,
Ann. 620 uosqueLarestectumnostrumquifandituscurant,Trag. inc. 35 Danai
quiparentAtridisquamprimumarmasumite.
•HAlOSis invoked thrice in Euripides' M1'\&1aand often elsewhere in
tragedy. He was the grandfather of Medea (Euripides, Med. 404 f[, 954 ft:)
and thus particularly likely to be invoked in any play about thisheroine.

243 candentem in caelo sublimat &cem: for •HAlOS/ Sol strongly per-
sonified as a man with a torch c£ Theodcctes, fr. 10. 1-2 & Kw.Aupeyyf\
Aaµ,ra6' elA(aaoovq>Aoycs1·HA1e,Lucretius 5 .401-2 SolqueCM!enti I obuius
aeternamsuscepitlampadamundi,976 dum rosea/ace Sol in/met luminacaelo,
Seneca,Here.f 37-8 Sol . .. I binospropinquatinguitAethiopusface. The con-
ventional image of 6.fdH:cnturyAttic tragedy was that of a man driving a
four-horse chariot {Sophocles,Ai. 845-6 et al.)but c£ Euripides, I.A. 1505-7
I AaµiraSovxos ~pa I ~16sTE cpfyyos ...•
(C.:.,loo.

380
 MEDEA EXVL; MEDEA
Candereoccun twice elsewhere in republican tragedy, is absent from
comedy and classicalprose.
For the tragic character of sublimissee above on v. 3. Sublinuireoccurs
only here and at Cato, Orig.2. 63 in literature before the time of Apulcius.

CXV
The theme of Mcdea's lustfulnessruns all through Euripides' Mf}6e1abut
the only possibleparallelsfor this fragmentas a whole are the nurse'swords
at vv. ~8 ov yap av 6icrn01v· lµf\ I Mf}6e1amipyoUSyiis l1TAEVa 1

1
I lpCAnt8vµov a<1Ti\ayeia ·taaovoS turned by Ennius as nam
ICA:>A1dQS 1

numquameraerransmeadomoefferretpedemI Medeaanimoaegroamoresaeuo
saudo,1andthechorus' words at vv. 431-2 aus· ~ ~ olKCA:>V ,rarpfCA:>v
mAEVaasI µaivo~ Kpa6fq;.There would be nothing againstplacingthe
fragment in a freeversionof the choral ode vv. 627-62 or the long speech
made by Jason afterthe discoveryof his murdered children(vv. 1323-50),
but a play about Medeaset in Athenscould contain it equallywell; Acgeus
had to banishher from this city afteran attempt on the life of Theseus.i

1',4 tmedet cordia cupidocorde: there is someevidencethat Euphorion

employed a form Mf}61} (fr. 14.3 Powell) but one ought not impose on
republican drama either the termination -e or variation in the form of a
proper name3towards an obscureAlexandrinism.The obvious connection
betwecnEuphorion'sfii\ (fr. 153 a) and thegauandcaelofEnnius' epic poetry
(Ann. 574, 575) has no relevancehere.
Cordis. .. cordeshould probably be allowed to stand; c£ Carm. Sal. ap.
Varr. Ling. 7 .27 diuom deo, Plautus, Cure. 388 ubi reliquiarumreliquias
conderem,Stich.126 edepoluoslepidetemptauiuostrumque ingeniumingeni,Trin.
309 dum uiuit uictoruictorumcluet, True. 24-5 Venus I quampenesamantum
summasumnuirumredit,Seneca, Med. 233 nam ducemtaceoducum,Thy. 912
regumqueregem.
For cupidocordec£ Ennius, Trag.ap.Cic. Tiisc.3 . 5 animusaeger. .. cupere
numquamdesinit(seefr. CLXXIV), Plautus,&ah. 1015 egoanimocupido.. Jui,
Mil. 1215 moderare animo;ne sis cupidus,Terence,Haut. 208 animusubisemel
secupiditatedeuinxitmala,367 ut illiusanimumcupiduminopiaincenderet,
Plumn.
821-2 eiusmodiin animopararecupiditatesI quas. .. mederipossis.Comedy
doublesanimuswith cor(Plautus,Pseud.1321) or replacesit with the physical
organ (Plautus,Epid. 146, Caccilius,Com. 79) in high-falutincontexts.

1
Plank thought that Nonius had misquoted Ennius' prologue.
:a Hyginus, Fab. 26. 1.
3 Enniushas Medeaat v. 216, Medeaiat 223. Cf. Plautus, Pseud.869.
 COMMENTARY

CXVI
If thiswere a comic fragmentone would most naturallyinterpret it as spoken
by an eavesdropperstanding on stage. In a tragedy it is likely to have been
spoken by the chorus hearing somethingoff stage or by a characterwho has
just entered the stage and speaksof what he has heard £rom ofP (resultative
present; secabove on fr. CXI)or by a characterreporting a past act of eaves-
dropping (historic present). As far as aucupantis concerned Columna's
identificationof the fragment with Euripides,Med. 131-2 and 0. Skutsch's
with Med. 67-9 arc possible.Butfructusis damningto them both. Nothing
overheard in the M1'l6aaprofits the h.carcror anyone else.

345 £ractmaerboram: for this type of periphrasissec above on v. 21.

ames aacupmt: the metaphor is £rom fowling. This and related mcta-
phon arc common in early comedy (e.g. Plautus, Mil. 598-9, 007-8,9SS,
990, Most. -473, Stich. 102, Caecilius, Com. 62-4). They arc absent £rom
Terence and from the copious remains of fourth and third century Attic
comedy. It is likely that they originated in Roman tragedy (c£ Pacuvius185
sermonmshie nostrumex occultoclepit,Accius 292) as an extension of such
locutions as Homer, n.I .201 Kai µiv q>(A>VT}U<XS frmx ,rnp6wra 1Tpo0'-
11v6a,Aeschylus,Hile.657-8 he O"l'OµaTc...w 1TOTalo&oo cpaAanµos evxa,
Prom.IIS T{S axe:,
... irpo<mrraµ• acpeyy,'ts, SSS TO 61aµcp{61ovSi µ01
µtAc,sirpo<mrra,Accius, Trag.-4-49 simulacnotauox aJaurisaccidit,Virgil,
Aen. II . 380-1 uerbisI qu« tuto tibimagnauolant.i

MELANIPPA

The title Melanippais given to Enniusthree timesby Nonius usingLindsay's

list 27 'Alph. Verb' and once each by Gcllius, Macrobius and Priscian; to
Acciusonce by Varro (Ling.7.65) and once by Nonius usinglist 28 'Alph.
Adverb' (p. I S4. Is). Melanippusis given to Acciuseight times by Nonius
using list S 'Accius i' (pp. 15.23, 85.4, 219.1, 233.22, 234.23, 485.31,
soo.14, 521. 8) and once using list 28 'Alph. Adverb' (p. 349. 2 ), three
times by Verrius (Fcstus and Paulus, pp. 180.28, 320.27, 34.0.25; c£
p. 256. IS), and once by Cicero (Tusc.3 .20); to Enniusonce by Nonius after
a seriesof entries£romlist s (p. 469.7). The attributionsat Varro, Ling.7. 65
1
Cf. Sophocles,0. T. 634 ff., Euripides,Herakleidai-474ff.
3
Cf. Plautus' parodies: Amph. 325--0 uox mi ad aurisaduolauit(Mere.864,
Rud. 332)-ne egohomoinfelixfui, I qui nonalasinterutlliI uoluaemuocemgestito.
382
 MELANIPPA

and Nonius, pp. 1s4.1s and 469.7 arc clearly crroncous. 1 The Melanippa
referred to by Cicero at Off. 1 • 114 is probably w Eooiao tragedy quoted
by the grammarians,although Cicero gives no other sign of knowing this
play.
Two tragedies about Mclanippe, the daughter of Aeolus, were composed
by Euripides. One, Mu.avhnn1 ft ooq,i),2 dealt with how, when thetwins
the heroine had home to thegod Poscidon and secretly exposedwere dis-
covered beingsuclded by a cow, Aeolus, on theadvice of his father Hellen,
decidedto have them burnt as prodigies. Mclanippc, having received the
task of dressingthem in funeral robes, delivered a speech criticising Hellen
and explaining the apparent prodigies in naturalistic terms. Euripides' other
tragedy, Mu.avhn"l ft &aµi;ms, dealt with an episode in the later life
of the twins.3
The fint fruitful attempt to interpret the fragments cited under the title
Melanippawas made by Bcrgk,4 who assigned frs. CXIX, cxx and CCXI to a
version of Euripides' Mu.avhn"l ft ooq,i) and cxxn and cxxm to one of
the Mu.avhntll ft 6eaµi;ms. Scholan since have usually tried to interpret
all six fragments as coming from a version of the Mu.avhn"l ft aocpf).

CXVIII
The prose hypothesis of Euripides' Mu.avhrn,1 ft aocpf) describesthe
heroine asK<XME,61acpipovcrav.The Enoian passage referred to by Gcllius
must have beenuttered by someone with Mclanippc's illegitimateoffspring
in mind.
For the substance of the Ennian passage cf. Euripides, fr. 928 ovyap
aa~ (Grotius: aq>EMs I
codd.) mpaa~ TO K<lAAOS ft l,lW'OV Aa~tv.S
Ovid, Fast.2. 161/oedera seruassetsi nonfcmnosafaisset,
Am. 3 .4.41-2 quotibi
f ormosam I
si nonnisiuuta placebat?nonpossuntullisistacoiremodis,3 . 14. 1 non
ego ne pecas cum sis f ormoSllrecuso,Epist. 16. 290 lis est cumf ormamagna

1
Delrius appcan to have thought that Accius' tragedy was a Melanippa.
Welckcr, Die Griech. Trag. pp. 854 ff., and Hartung, EuripidesRestitutusII,
pp. 375 ff., pressed the fragments into a version of Euripides' Mu.avhnr11
1'i&o-µwTis. Scaliger, Conied. Varr.Ling., on 7 .65, said all that is necessary.
a For the prose hypothesis see the rhetorical work published by H. Rabe,
RhM LXID (1908), 1-4.sf., and Pap. Oxy. 2455, fr. 2, col. 1.
3 See Hyginw, Fob. 186, Pap. Berol. 5514, 9772, Wilamowitz, SB Berlin
1921, 63 ff. ( = Kl. Sehr.1440 ff.), Pickard-Cambridge, in Ntw Chaptersin the
History of GrttlcLiteraturem (Oxford, 1933), 117 ff.
4 RhM m (183S), 71-3.
s Compared by Hartung, EuripidesRestitutus 1, p. 116, and assigned to
Mu.avhnn1 1'iaoq,1'}.
 COMMENTARY
pudidtiae,Pctronius 94. I raramftcit mixturamcum sapimtiaJ"""", Juvenal
10. 297-8 rarastatkoconcorduJ jormtJeI atquepudidtiae.
It is idle to attempt to restore Ennius' actualvcnc. His descriptive ablative
~
stataf ormais hard to interpret closely ( Plautus, Mere. 13 Jormaeximio
muliertm,Mil. 782Jonnalepidamuliertmet al.); pcrhaps.formahas beengiven
something of its etymological sense (c£ C.I.L. 11 592.2.2 pecunuz ... sig,uua
Jonnap[ublica]P(opuli]R[omani]}and thephrase is to be interpretedas 'of
the legally fixed stamp or mould, not departing from the norm, neither
strikingly beautiful nor strikingly ugly•.

CXIX
The words transmitted will scan as a hypcrmetric trimetcr 1 or a venus
rcizianus.2Acidalius' removal of thesecond -<JUt produces a regular trimetcr.
On Nonius' interpretation thesubject of sospitentsuperstitentque could be
either thetwins Aeolus and Bocotus or the gods. The redundancy is typical
of the language of tragedy (secabove on v. 9) and prayer; cf. Sophocles.
0. T. 1349-516Ao1e· OOTIS ljv ~ ... arr6 TE cp6vovICppvromwacootIJE,
Plautus, Asin. 16-17 sicuttuomuis unicumgnatumtuaeI superesse uitaesospitem
et superstitem, pecuaquesaluaseruassisduisquebonam
Cato, Agr. 141 . 3 pastores
salutemualetudinemque mihi domof amiliaequenostrae,Lucilius 739 sospitat
saluteinpertitplurimaet pknissima.However the only other time the verb
supe,stitareoccurs (Plautus, Persa331 ut mihi supe,sitsuppttatsuperstitat)it
is intransitive. If Ennius' useis parallelwe have a Ocrrepov ,rpcrnpov of the
type discussed above on v. 139 and the verse must come from a prayer by a
member of the king's family to allow the twins to survive.

246so,pitent: the adJective sospesoccurs twice in republican tragedy and

five times in comedy (against saluus18o times); it doubtless belonged to the
sacral language. The dcnominative sospitareoccurs twice in tragedy and
twice in comedy (Plautus, Asin. 683, Aul. s46); it may have been a tragic
neologism.
CXX
With one alteration the words transmitted will scan as a cretic tetrameter.
Something other than iambo-trochaic verse is suggested by the word order.
Auscultamihi seems to have beenregular (Plautus, AuL 237, Most. 586,634,
Persa 574, Poen. 311; but cf. Plautus, Cas. 204 [anapaestic],Stich.6o2,
1
So Vahlen, Ind. lectt.Berlin 1878, 6 (= Op. ac.1 56), comparing Trag. inc.
191-2.
a So Strzelecki, Bos XLII (1947), 26--9,comparing Naeviw, Tr•g. 13 num-
quam hodit ,jfugiesqui,i manu mea moriare(sec Lindsay, E.L. V. p. 179 n. 2).
 MELANIPPA

Cicero, S. Rose. 104 [auscultdre docs not occur elsewhere in Cicero]). The
imperative iubt normally preceded the infinitive (but c£ Plautus, Capt.668,
Mm. 517, Mere.777, Rud. 1095, True. 585 [cretic]). Nevertheless the sub-
stance is of a dramatic importance hard to parallel in the certainly established
cretics of republican drama. Bothe's crtmitdriis the neatest further correc-
tion; the absence of the intensive form from recorded Latin and the apparent
lack of any intensive force do not count greatly against it; for tragic use of
theintensive sec above on vv. 68, 203.
The speaker would be Hellen; c£ the prose hypothesis of Euripides'
tragedy: Ta ~ TI~ Tci>vJx>,nc6Ac..w fVAarr6µa,a ~ wo TOV
& V'Tl'O
Tavpov, 6r}ACX36µa,a ,.nas
T<i>Vf3o&>vl66vns ws~
-ripcrra TCf> · 6 & Tij TOOircrrpos ·ru11vos
f:xxcnMtirpoa,'}veyi<cxv
~'1TJ maa8els 6AoKavTo0vTa ~ftl a<pf~ ...•
247 nate: the use of gnatushas no stylistic significance here; vocative.fili
does not occur in republican drama. Mi gnau and gnau mi are the normal
forms of address. The absence of the pronoun indicates a certain peremptori-
ness on thespeaker's part (c£ Plautus, Trin. 362, Terence, Haut. 843, 1o65).

cremari: not elsewhere in republican drama; probably from the sacral

language; Cicero and Caesar use it of burning corpses and executing
criminals(Gall. 1.4).
CXXI
Hellen is speaking of the twins. Ennius makes him use thet«bokal language
of a Roman augur; c£ the behaviour of Virgil at Atn. 3. 58-oo: dtlectos
populiadproaresprimumqueparentemI monstradtum reJero,et quaesit sentffltia
posco.I omnibusidemanimus,salera.ta
excedtreterra.

248momtram: 'sign of divine displeasure'; a word of the sacrallan-

guage; c£ Plautus, Most. 505 quaehie monstrafiunt anno uix possumeloqui,
Terence, Plwrm.705-8 quotrespostillamonstraeuenerunt I
mihi introiit in aedis
ateralienuscanis;I anguisper inpluuiumdeciditdetegulis;Igallinacecinit,Festus,
" monendo
p. 122. 8 ut AeliusStilo interpretatur dictumest,uelutmonestrum.item
SinniusCapito quodmonstretfaturum et moneatuolu~m deorum,Paulus,
p. 147. 10 monstradicunturnaturaemodumegredientia, ut serpenscum pedibus,
auis cum quattuor alis, homoduobuscapitibus,iecurcumdistabuitin coque,ulo.

liet: the full forms sitm etc. were still employed in official inscriptions
during the second century but had little life in the common language. The
republican poets usedthem in comedy as an ornament of style but much less
frequently thanin tragedy; F. Thomas, Recherches, p. 10, counts 284 examples
in extant comedy as against ')67 of sim etc. and 8 in tragedy as against 19.
25 JTO
 COMMENTARY

249 hoe ego tibi dico et coniectara aaguro: the second statement,
cbronologially prior to the first, is given even greater emphasis by the
hiatusin caesura. I
For coniecturaaugurartc£ Pacuvius, Trag.78, Cicero, De oral.1 .95, Att.
2.9.1. Ennius' Hellen, however, proposes not to make a prediction but
rather to give advice on what thegods desire; his statement comes as a result
of reasoning (conitctura)not of second sight (hariolatio,
uatidnatio
).

CXXII

The words transmitted will scan as a dactylic hexameter. They should be

treated as such rather than divided between incomplete anapaestic dimeters.
There is a strong tendencynoticeable in Macro bi us, Sat. 6. 4-5 for the quota-
tions of pre-Virgilian poetry to form metrical units.2 Hexameters are found
commonly enough in Attic tragedy (Aeschylus, Ag. 104, 111, 119, 123, 129,
137, 155-7, Xantr. Pap. Ox. 2164, fr. 1. 16 ff., Sophocles, Phil. 839-42, Tr.
1010-14, Euripides, Andr.103 ff., 117 ff., Hel. 164-5, Tr. 595-602) and in
imperialRoman (Seneca, Med. 110-15, Oed. 233-8).
Bergk3 referred the words to the restoration of Melanippe's sight after
her imprisonment; Hartung4 to a discussion of natural philosophy by the
chorus; RibbeckS to a version of the Euripidean Melanippe's famous speech
on thissubject (c£ fr. 484.4 TIKTOVOl ,ravra KavE&.>Kav els cpacs).Many,
6
including Wilamowitz, have accepted Ribbeck's idea without explaining
either the presence of the dactyls or how a sentence introduced by sicwould
fit into a narrative of the creation of things.Welcker7 took the words to be
part of a similewhile Ribbeck allowed that it might come from a version of
fr. 486 61KCX100'WCXS I
TO XP'JC7EOVirp6aCA>1TOV ooe· lompos ooe· ~
I
OVTCA>8avµaOT6s (Meineke's arrangement). The dactyls suggest to me an
utterance by the seerHellen (c£ vv. 43-6) and thetalk of light the epiphany

1
For this type of hiatus, frequently disallowed by students of republican
drama, see above on v. I 54. Ribbeck scanned tibi as an iambus and W. Ax, Dt
Hiatu qui in FragmentisPrisetttPoesisRomanaeinuenitur(Diss. Gottingen, 1917),
p. 29, proposed hoetibi egodicoet eoniecturaid auguro.On iambic tibi, extremely
rare in this part of the iambic trimeter, sec P. W. Harsh, Iambic Words and
Rtgardfor Aatnt in Plautus(Stanford University, 1949), pp. 110 ff., 0. Skutsch,
Mnemosyne4.xm (1900), 231 f.
:a See CQ N.S. xv (1965), 129 ff. 3 RhM m (1835), 73.
4 EuripidesRutitutus I, p. 121.
5 Q.!aest.seen. p. 265, Die rom. Trag. p. 179.
6
SB Berlin 1921, 74 n. (=Kl.Sehr. I 453 n.).
7 Die grieeh. Trag. p. 850.
 MELANIPPA
of a god,.probably Poscidon. 1 Poscidon must have appeared at the end o f
the play to rescue Melanippe and the children he had sired.

ZSolarnirie . .. tremulo: c£ Apollonius Rhod. 3. 756 f)U\{ov... ,r~-

i\n'cn afyATt,Luaetius •. ♦04,-S iamquerubrumtremulisiubarignibuserigert
I
alk cumcotpt4tnatura,s.697subtnris ideotrtmulumiubarluitsit4tignis,Virgil,
Aen. 7. 9 splendettremulosublumintpontus,8. 22-3 sicutaquae tremulumlabris
I
ubi lumenahenis solt repercussum, Ovid, Epist. 18. S9 lu,u,f ert tremulum
pratbebatlumeneunti.

caerala: adjectives terminating in-ulus etc. are rare in republican tragedy,

as are similarly terminating diminutive and instrumental nouns. The
employment of two in the one sentenceperhaps sought some specialeffect.

CXXIII
Most scholars refer this fragment in some way to the imprisonment of
Melanippe.2 It is difficult,however, to imagine an imprisonment in an
adaptation of Euripides' MU\avhmTl 1'iaoq,1').I can find no satisfactory
interpretation of the three words transmitted.

NEMEA

The title Nemea is given to Ennius by Nonius and Priscian. Nonius' source
appears to be Lindsay's list 1 'Glou. i', Priscian's the work ofFlavius Caper,
going back through Probus to Pliny and perhaps beyond. One of the two
pieces quoted must have been spoken by a woman whose libertyof move-
ment had been restricted; the other is quite enigmatic.
Columna thought that Ennius' play dealt with the story of the death of
Opheltes while under the care of the enslaved Hypsipyle and with the institu-
tion of the Nemean games in his memory, a story which hassince proved to
be the theme of Euripides' 'Y'fll'Tl'VATt-Columna's view hasoften beenmain-
tained, most recently by Przychocki.3 Ennius' play perhaps referred to the

1
Cf. Homer, n. 4.75 ff., Hymn. 2. 189, 3 .440-4 ••. oO.as s· els 0\/paYOV IKEV
... nao-av & KpfC7T1v KaTSXEV oO.as, Euripides, Ba. 1082-3 npos o<,pavov I Kai
yaTav lcrn'ipt~ ~ O'Eµvo(iirvpos, Ion I 5-49-50,Lucretius I . 2-9, Virgil, Aen.
1.402, 2.589-90, 2.615-16, 3.151-2, -4.358-9, Ovid, Fast. 1.9-4.
:a Welcker, however, referred it to a discussion of miracles by the philo-
sophical heroine.
3 WSt XXXI(1909), 300-5.

2 S·:Z
 COMMENTARY

institution of the Ncmcan games in the aetiological manner of Attic tragedy

but it is scarcely credible that the title should have done so. One would
naturally expect the title to be the name of the daughter of Asopus,in honour
of whose son Archemoros, according to another story, the Ncmcan games
were instituted. The only Attic tragediancredited with a Neµia isAcschylus.1
Whether or not Ennius adapted this play, rio fragments of which survive,
must remain an open question.a

CXXIV

yaptu.iii>v
252 teneor comaepta: c£ Euripides, Herakles83 cpvACXJ<al
yape1Ataa6µ£6airayx<XAt<ots
1<ar~t~66ovs, Or. 444 l<Vl<A<t)
t<pe{aaoves
76o cpvAaaa6µea6acppovpfotatiravrCXXij,Hyps. fr. 20/21 .12
CfflAOtS,
cpvAaaonat yii cppovpfotatvtv l<Vl<A<t).
Consaepiredoes not occur elsewhere in republican drama; the simple verb
occurs 5 times in tragedy and 3 times in comedy. Paulus, Fest. p. 54.22
CONSIPTVM apudEnniumpro conseptum inueniturprobably has nothing to do
with the piece cited by Nonius; Vahlen draws attention to the article at
p. 56. 13 CONSIPTVM clauispraefixum.

undique uenor: for undiquedrcumuenior(c£ Caesar, Gall. 3 .26 hostes

undiquedrcumuentiet al.),if Nonius is to be believed. Veno,for ueniorseems
to be quite unparalleled. Nonius, however, may merely be using a word
with some affinity of sound to gloss the passive use of uenari•hunt' (c£
Priscian, Gramm.II 387. 12 ).

CXXV

Przychocki referred thewords transmitted to the oracle which badeAdrastos

marry his daughters to a lion and a boar.3 This cannot be so. Pecu,pecusetc.
normally refer to domesticated animalsor animalscapable of beingdomesti-
cated (cf. Lucretius 1. 14, Varro, Rust. 2. 1. 12); there is a tone of irony in
Ovid, lb. 455 inquepecussubitoMagnaeuertareparentis,as in Livius, Trag. s
lasciuumNereisimumpecus.Maritowould suggestthat theanimalalready hasa
consort.

1
Sec Mette, Der verloreneAischylos(Berlin, 1963), pp. 38 f.
a Cf. F. Skutsch, RE v (1905), 2594, Leo, Gesch.p. 190.
3 Euripides, Hik. 138 ff., Phoin. 409 f., Hyps. fr. 8/9. 13 ff., Apollodorus
3.6.1.3.1.

388
 COMMENTARY

PHOENIX

The title Phoenixis given to Ennius by Aulus Gcllius, Hdenius Aao an4
Nonius Marcellus.1 The eight pieces quoted are each open to a variety of
interpretations. Cicero mentions on one occasion (Deorat.3. 57) a Phoenix
apudHommunand on another (Mur.6o) quotes a piece of tragic verse which
might have been spoken by AcbiUcs•old tutor Phoenix in a play about the
events of n;aJIX (seeabove, p. I~).
Among the known Greek tragedians, Sophocles, Euripides, Ion, Asty-
damasand a certain ?)nedoror are credited with a <Dot~. Euripides, play
was set before the palaceof Phoenix•sfatherAmyntor in Boeotian Elcon and
dealt with the falseaccusationsmade againstPhoenix by Amyntor•s con-
cubinePhthia and Amyntor•s violent treatment of hisson.3Columna asserted
that none of the fragments preserved prevents us from identifying the
Enoiao :md Euripidean plays, and Va1ckenaer4remarked on the lofty moral
tone of the two sets of fragments.SFr. cxxx loob like the uttcraocc of
Phoenix, and the sort of Phoenix we know Euripides to have portrayed.6
At Orat. 155 Cicero quotes, apparently from a grammatical sourcc,7 a
dramatic tetrameter by Ennius (fr. CLXX):~que tuum umquam in grnnium
extolLu liberorum ex te genus.Columna gave this to the Medea exul but
Elmsley8 spotted a similarity with Amyntor's curse as reportedat Homer, n.
9.453--0 (iratT1p 6' tµ6s... 1TO~a KCffllparo ..• IJfl1r011! youvacnv
olalV ~O'E0'6cxt q>fAovvtovI te tµieevYEYai,;mx) and gave it to the
Phoenix.The assignationis plausiblebut far from ccrwn. vahJcn•salteration
of the transmitted text to nequetu meumumquam etc. cannot be acceptedeven
if Elmsley's assignationis. Two unjustifiedassumptionsare made, fust that
Euripides followed Homer exactly, second that Ennius followed Euripides
exactly.9
1
Nonius' immediate sources seem to have been Lindsay'slist 27 'Alph.
Verb' and list 28 'Alph. Adverb'; at p. 518 .6 list 1 'Gloss i' is a possible source.
Gellius and Aero may depend on Verrius Flaccus.
:a I.G. n• 2363. 18. ' Cf. SoudaA 1842.
4 Diatribe,pp. 262 ff.
s This is not completely without significance; one would certainly expect
this tone in the quotations of a Stobaeus but not neccssaril.y in those of a
Nonius Marcellus.
6 Cf. Schol. A. Hom. n.
9.453 ml Evp1irl611s 6i &,aµapTTtTOV daayn wv
ft{XA)a
Iv T~ C!>o(vt1<1. Sec above, p. 348.
7
1
EuripidisMedea,p. 201. Ribbeck and Vahlcn wrongly credit Bergk (RMI
m [183s], 73) with the idea.
9 In g,tmiumextollertmay be a variant of the Roman formula discussedabove
onvv. 5~.
 COMMENTARY

Two dramatic pieces quoted by Charisius,one to illustrate epanalepsis----

patn i~om hospitesrM lumineorbauitpater,1the other a~gone illmn:
1
pudoresteloqui:quamcomperi,could be interpreted as coming from a version
of Euripides' <l>olVl~.
3
CXXVI
SinceGelliusis discussing olmoxius•dependent'we mightexpect to find this
word andthissenseexemplifiedin theversesof .Enniuswhichbe quotes. But
ancient lexicographen frequently included etymologically related words
under on~ Jeanna.4
.Ennius'fint two versesmakea statement about uirtus,conventionallythe
successfulexerciseof a man's physicalpowers in war&re, statecraftandpro-
acation,S his second two about libertas,conventionallythe state of beinga
free man and not a slave.6 In both casesconventionalnotions are beingcor-
rected or reinterpreted but textualcorruptionsobscurethedramatist'spoint.
The phraseuirtute... animatumhasno exact parallel;c£, however, Plautus,
Trin. 650 capesis uirtutemanimo.The preservation of animatumwould also
provide some link between the fint sententiaand thesecond(ptctuspurumet
~
firmum uerauirtutt. .. animatum. . .f ortiter). Verauirtusmay have meant not
•true, genuine uirtus' but rather •uirtusaccompanied by a sense of fair-
dealing' (c£ Virgil, Am. 12. 694-5rM ueriusunum I pro uobisf oedusluereet
decernere fmo and Servius ad loc.). The alliterative phrase was a clicheof
Roman public moralising (c£ Plautus, Cos. 88, Cist. 198, Cicero, Pis. 57,
I.ivy 4.31.5, 24.14.6, Horace, Cann. 3.5.29, Epist. 1.1.17, 1.18.8).
The sentenceea libertasestquipectuspurumetJirmumgestitatcan stand; for
similar ~tructurcs see above on v. 228.
R.esolmoxios«must bean attempt to representtheabstract idea of depen-
dencein a context where words like seruitusand cliente/4 would sound too
paradoxical. For this manner of dealingwith the abstract c£ Ennius, Trag.
318 res. •. turbidas,
Plautus,Mere.134 ma/« res,Trin.344 in rebusaduorsis.
The
point seems to be that men whose minds are unfree, ie. men with a bad
conscience,keep out of public sight.
1
P. 370.9; cf. Diomcdes,Gramm.1445.29, Sacerdos,Gramm.vi458.11.
~ P. 374.6.
3 So Ribbeclc, Q!!aest.seen.p. 264, Die rom. Trag.pp. 195 f., T.R.F.', p. 61.
Mariotti, SIFC N.S. XXIV (1949), 87 f., replaced hospiteswith hospesin the fint
piece,assuming~changeof scenefrom Amyntor's palaceto that of Pclcw in
Phthia.
• Cf. Fcstns, p. 334. 8 RBDHOSTDlB, Nonius, p. 2ss. 1 CllBPAll.
s See above on fr. LXXI.
6 So alwaysin republican drama exceptat Plantns, Bacch.168, 'free speech',
and Mil. 702, 'bachelorhood'.

390
 PHOENIX.
The substanceof the four Latin verseswould have been something like
that of Euripides,Hipp. 421-30: &XA'~t I irapptla{~ ~ovns
olKotevir6AtvI wt~v 'A8ftv<Z>v, µflTposowe<' EVKMEls. j 6ovAotyap
&v6pa,K&v6paava11'Aayxv6sTtSt, I 6Tav ~tSij µT}Tf)OSi\ mrrpos
KCXKCX.I µ6vov & TOVTOcpaa' ~tAMaem Pfct>,I Y\M"11lV6tKCDCD
Kaya6r)v, 6-Tct>,rapij. I KCXKOUS A~v• ... XfXNOS.
& &vT}TcZ>v

255 aduerarios: more generalthan hostisor inimicos;six times clscwhcrc

in republican drama; comedy has 99 formations in -arius according to
Ploen's reckoning, tragedy only this one.

256pect1U parum et firmam geltitat: the normal phrasewas animum

gerere(c£ Terence, Hee. 311 quieosgubernatanimustum in.firmumgerunt);
pectuswas a fairly common high-falutin substitute for animus in both
tragedy and comedy; gestit4rereplacesgerertonly here and seven times in
Plautus.
For pectuspurum c£ Euripides, Med. 659--61 &xaptOTOS6Aot6', 6-Tct>
irapEOTtvIµ~cpiAOVSTt~Ka6apav&vo{j~<XVTalV\ijSacppevC>v, Lucilius
296 quodg,acilaut, panix, quodpeaorepuro,I quodpuno similis,Horace,
Sat. 1 . 6. 64, Epist.1 . 2. 67-8, Lucretiuss.18, Anon. Culex 68.

257 obnoxioae: 'obnoxiae,dependent'; c£ Plautus,Epid.69s,Trin. 1038.

With obnoxiosusI count 13 formations in -osus in tragedy (amunnosus,
aestuosus,bellia,sus,dusmosus,
globosus,malitiosus,otiosus,religiosus,
sauposus,
saeptuosus,uillosus,unose).They arc common in Plautinc comedy, rare in
Tercntian.

nocte in obscura: an order of wordshighly unusualin republicandrama,

fairly common in Ennius' epic (Ann. 187, 378, 38o); sec Marouzcau, REL
XXV (1947), 321.

CXXVII

Somebody,perhaps the chorus leader, comments on Amyntor's denuncia-

tion of Phoenix.

258 ex ore orationem: for the word play c£ Plautus,Mere.176tu quidem

ex oreorationemmi tripis, Cicero, Phil. s.20 in me abmstmtoratioMmex ore
impurwimoeuomuit.

duriter: not elsewherein tragedy, four times in comedy (alwaysat the

end of a metricalunit); Jureoccun nowhere in republicandramaand is rare
in classicalprose and verse, which appear to prefer durius.

391
i
 COMMENTARY

dmiter dictis dedit: aiple alliteration of D is significantly rare in

tragedy; it occun only here,Ennius. Trag.303 Jeleaattluaat Delphiau,Trag.
inc. I 84-s disparidominaredomino.M and T begin about the same number of
words in the tragic lexicon and produce 1.2 and 8 aiplc alliterationsrespec-
sec above on v. +
tively. On aiple alliteration in general

CXXVIII

Text and context are obscure. Mariotti' commends Ribbcck's stultustqui

cupitacupienscupitntercupit,2 comparing with cupitacuperev. 228 qui uolt
quoduolt.
For the sentiment c£ Plautus, Pseud.683-4 stultilu,u scimus.frustra
ut simus,
quomquodcupientndariI petimusnobis,quasiquidin remsitpossimusnosme,
Lucilius806 cupiditasex hominecupidoet stultonumquamtollitur.
259 capiem cupienter cupit: c£ Plautus, Cas.267 quidistuctam cupide
cupis?;for the figure in general seeabove on v. 63; for the aiplc polyptoton
c£ Aeschylus,Pers.1041 66<nvKaJ<CXV ICCXl<C>vKCXKOlS, Sophocles,Ai. 866
ir6vos ir6v<t>ir6vov cpipEl,Euripides,&. 905-6 ~ s·mposmpov I
6Afxt,Kai 6waµe1 irapi\).8ev,Euripides,KylJ. 120 m<OVII s·ov&v ov&ls
ov&v6s, Plautus, Amph. 34 iusk abiustisiustussum orator"4tus,Capt.774 ita
hie me amoenitateamoenaamoenusonerauitdies, Cas. 826 nu,lamalaemale
monstrat,Poen.1216 bonusbonisbeneJeceris.
Cupienteroccun here, Accius, Trag.S43 and Plautus,Pseud.683 in republi-
can drama; cupitkoccun six times in comedy, not at all in tragedy.

CXXIX

Wilamowitz and Snell (WSt LXIX [1956], 90) accept Haupt's tum tu isti aede
te etc. (HermesII [1867], 216 [ = Op. m 3751) and compare the trimeten
quoted by Stobaeus2. IS .25 and 3. 13. 14 and very plausiblyassignedto the
by 0. Hense (adStob. 3. 13. 14, p. 456): Kai T4'>& 61'\A~aaµ• 6v, El
<l>olvi~
~AOIO av,I TaAfl~, ~ fyCAlyE K<XVTOS &x8oµaa, I OOTISAEyElVµev
ev,rprn&:>shr{OTaraa, I Ta s· Cpya xe{pCAl T6>V>.6yCAlV 1Tcxpa7XETO.

2'o exerce li.aguam: c£ Ovid.Met. 6.374-s turpesI litibusexercent

linguas,Tacitus, Dial. 31 linguammodoet uocemexercerent.
argutarier: Plautus has argutus•talkative in an idle, uselessway' a
number of times, argutaritwice (Amph.349 perginargutarier?,
&. 81 supnabo-
queomnisargutando praeficas
).

In private conversation,8 February 1966.

1

T.R.F. 1, p. ,u; withdrawn in favour of stullus est qui non cupienda

a
cupiens. .. , T .R.F. •, p. 59.
392
 PHOENIX

CXXX
Phoenix replies to advice suggesting that he resist his father with actions as
well as words.

261 pbu miser: miserioris normal in drama; for this form of the com-
I
parative c£ Plautus, Cas.676-7 tibi inftstasolist plusqu11mcuiquam.

scelestam: not elsewhere in tragedy, frequent in comedy of the beha-

viour of slaves.
The manuscript salestim would in itself be acceptable; Vossius drew
attention to the variation diserte/dimtim,U1Ute/U1Utim,
arcte/arctim,
exquisite/
exquisitim.

fayjm: thisform was probably still alive in thecommon language,unlike

other -s- and -ss- subjunctive/optative forms. Plautus hasf axim 10 times
against fecerims; Terence hasthem twice each. See Thomas, R.echercks,
PP· S3 if.
quod dic:amfore: Vahlen pointed out that this was a periphrasis of the
type disamed below on v. 300. It can be objected that dicamis normally so
usedin direct and indirect questions, not in relative clauses.

CXXXI
Someone, perhapsPhoenix, speaksof Amyntor's angry reaction to Phthia's
accusations.

262 aeuiter ... ferre: c£ Afranius, Com. tog. 301.

ferre &J•m futtilum: triple alliteration of F is significantly rare in

tragedy; it occun only here and at Naevius 45jlammisfierijlo,a; the letter
beginsabout thesame number of words in the tragic lexicon as D, M and T
(seeabove on v. 258).
For fattilis/-usof persons c£ Terence, Atulr.609 muon fortutw measme
commisisse fattili,Afranius, Com.tog.3S, Cicero, Fin. 3 . 38, Diu. 1 . 36, Virgil.
Am. II .339, Phaedrus 4. 18 .33.

CXXXII
Phoenix must be speaking of his failure to stand up to Amyntor.
Feratisis an absolutely necessary correction (c£ Plautus, Asin. 323 em ista
uirtusest. .. qui malumfertfortiter)but leaves the trochaic tetrameter with an
oddly divided third foot (seeabove on fr. xn).

393
 COMMENTARY

CXXXIII

It is difficult to fit this tetrameter into a dramatic treatment of the story of

Phoenix. No one has been convinced by Hartung's suggestion' that a
messenger describesPhoenix receiving back his sight on top of Mount
Pclion.

tam: c£Terence, Amir.131,634, Cicero, ~inct. 16, Verr.2. 3. 139,

:&6.4ibi
Caecin.27.

.TELAMO

The title Telanwis given to Ennius six times by Nonius, i once by Verrius
·and once by Diomcdcs. Two of the seven pieces quoted (frs. CXXXV,
CXXXIX) clearly fix the action of the play in Salamis at the time ofTcuccr's
return from Troy. Aeschylus' I<XAaµ(vicn, Sophocles' TM<pOSand Pacu-
vius' Teuceralsodealt with theconfrontation between Tcuccr and his father
but there is no TEAaµchv among those titles which arc recorded against the
names of Attic tragedians.
At Nat. deor.3. 79, Diu. I. 132 and Diu. 2. 104 Cicero quotes a tragic
speech setting out the Epicurean view of gods and divinen; he names Ennius
twice in connection therewith and thehero Tclamo once. With the excep-
tion of Dclrius all editors sin.ccColumna have printed the speech with the
grammatical fragments of the Telamo.
At Tusc. 3 . 28, 3. 39 and 3. 58 Cicero quotes from two tragic speeches
clearly labelled as uttered by a Telamo. At 3 .44 he quotes a third speech
which we can deduce to have come from a tragedy by Ennius (fr. CLXXV;sec
above on fr. xxvnh). Columna printed the remains of the three speeches
under the title Telamoand argued that the action of the play began with the
exile ofTelamo himself from his original home in Aegina.
Scholars have continued to assign the speech quoted at Tusc.3 . 28 and s8-
egocumgenuitum moriturossciuietei reisustuli,Ipraeterea adTroiamcummisiob
defendnulamGraeciam,I scibamme in nwrtif erumhelium,nonin epulasmittere-
to the Telamoalthough it is difficult to see why it could not equally well go
into Pacuvius' Teucer.
The speech quoted at 44-pol mihif orturu,magisnuncde.fitquamgenus. I

1
EuripidesRestitutus1, pp. 75~.
a Nonius' source at p. 172. 19 seems to be Lindsay's list I 'Gloss. i '. At the
other five places there is considerable uncertainty; list 27 'Alph. Verb' could be
the source.

394
 TBLAMO

namque regnumsuppekbatmi ut sdas quantoe loco,Iq,,umtisopibus,quibusderebus

lapsaforturuiaccidat-hasbeen given to the principal pcnonage of several
Boni.antragedies,to Hecuba by Bothe,1 to Telephus by Duntzer,2 and to
Thyestcsby Ribbcck.3A carefullook at Cicero's argument from 39 to 44
enables one to exclude both Telamo and Thycstcs as possible speakers;
Cicerois adaptinga criticismof Epicurus' doctrineon pain and sufferingand
adorns his adaptation with quotations from and references to Roman
tragedy; at 39 he namesThyestcs,Aeetcs (with a backwardglanceat hisactual
quotationsof speechesby them at .26)and Telamo, at 43 Telamo again;at 44
qui ita dicat,whoever he is, and Anclromacheprovide further examplesof
tragicsuffering.
The speechquoted at 3~icine est ille Te1"'""modoquemgloriaadc«lum
extulit,I quemaspectabant, cuiusob os Grai oraobuertebantsua? I simulanimus
cumre concidit-obviouslycannot go in the sameplay either with the speech
quoted at 28 and s8.orwith the grammaticalfragments.Hermann's attempt4
to interpret it as spokenby Teuccr rccaUingthe banishmentof his own father
from Aegina many years previouslyis quite unconvincing.W dckcrS sug-
gestedthat Cicero quotes a play about Telamon's banishmentfrom Aegina
unrecorded by the grammarians.Ribbcck,6 having reconstructed Accius'
Eurysaces so as to make Aegina the scene of action and Telamon, expelled
from Salamisin old age by revolutionaries,one of the personages,included
the quotation therein. Sincethere is nothing in the mythographicaltradition
about Tdamo losing his Salaminiankingdom in old age Wdcker's view is
much to be preferred..
Acceptanceof the view that Ciceroknew threeplaysabout Tdamo would
raise a doubt as to the propriety of placing the verses quoted at Nat. deor.
3. 79, Diu. I. 132 and Diu. 2. 104 together with those quoted by the gram-
marians.NeverthelessCicero's verseslook as if they were spoken by a man
exercisingauthority, not an exile bowed by misery.With some hesitation
I have printed them under the title Telamo(fr. ocoov).

1
Despite Cicero'seum.
2
Zeitschr.f. d. Alt. 1838, S3; cf. Ladewig, Anal. seen.p. 2s.
l Q!aest.seen.p. 268; cf. ~ rom.Trag.p. 201.
4 De Aeschyli Tragoediis FataAiacis et TeucriComplexis,p. 17 ( :i::: Opusc.vn
381).
s Die griech. Trag. pp. 1379--80; cf. Robert, ~ griech.Heldensagem 2 i,
p. 104,4..
6
Conied. pp. 31 ff., Qyrest. scen.pp. 328 f., Die rom. Trag.pp. 419 ff.

395
 COMMENTARY

CXXXIV
(a) G. Hermann's view that Telamo is •peaking after receiving news of
Ajax's death seemsto have prevailed among laterscholars.But theprincipal
point ofTelamo's speechwas surdy the timeliabilityof diviners' advice. No
divineris known to have foretold Ajax's return home, and theutterancesof
Calchasreported at Sophocles,Ai. 746 ff. were justified by events. We must
have in fact a situation like that of Homer, OJ. 2. 178 ff., Sophocles,Ant.
1033 ff., 0. T. 380 ff., Euripides, I.A. 9SS ff. But what adviceTelamo had
been given I cannot guess.
The general prosperity of the wicked and the misery of the righteous arc
common themes in the speechesof the unfortunate in ancient drama: c£
Euripides,fr. 286. 1-12 cp,,aivns elva16i;T"wovpavi;>8eovs;IoVKelaiv,
OVKeta·, et TlS &vepc;nroov EMM1 I µn Ti;, ircxAatci>
µoopos&v XPiia&at
AOY<t>• I~aa& s· cnrro{,µn ml TOTSlµoTsA6yo1sIY"00"11'lV fxoVTE5.
q,11µ•fyw TVpaw{6aIK'TEiVElV TE1TAE{C7TOVS KTIWCXTCA>VT"6:mxm:peTvI
opKovsTE irapaJ3alvoVTasmrop8eTv ir6AE1s · I Kal Tcxvra 6p&'>VTES
µciAA6v eta· ev6a{µovesITCA>V ruaefk,vvToov fiovxfjKa8'""~. j 1r6i\e1s
TE µncpas ol6a Tlµ(OO'QS ~. Iat µe136voovKAVOVal6vaae~v I
AOYXTlS ap1&µc;> irAE{ovos Trag. Grace. inc. fr. 465 TOAµ&'>
1<p(XTOV"1EVQ1,
Kcrre1mTv, µ,;irOT•OVK elalv 8Eo{ • IKCXKOl
yap EVTVXOVVTES 1Kir1-.itaaovat
µe. The gods themselvesand some mortals take a differentview: cf. Euri-
pides. El. 1350-3 TOTS µevµvaapoTs oVKmap11yoµev,I olaiv s· 0010v
KaiTO6{KatOV Iq,{1-.ov
w f316T<t>,TOVTOVS xcxAmoov I ~VOVTESµ6x&oov
aci>30µev,Ion 1619-.22, Plautus, Capt. 313-15 est profectodeusqui qu« nos
gerimusauditqueet uidet. .. bent merentibeneprofaerit,male maenti par erit.
Rud.9-21.
265 nam si curent: the present subjunctivenormally expressesthe unreal
present in the Latin of republican drama.
benebonis sit, malemalls: cf. Plautus,Bacch.66obonussit bonismalussit
malis.
quod nunc abest: 'which is at present not the case'; T.L.L. I 210. 18 ff.
offersno exact parallel.

(b) The substanceofCiccro's argument is complete at the word lunioli.The

four trochaic tetramcten aut inertes. .. reddantceteraare merely decorative;
indeed they repeatsome of the substanceof the foregoing discourse (inertes
~ non. .. scientiaaut arte; insani~ superstitiosi;quibus egesttu imperat~
1
De Aeschyli TragoediisFalaAiaciset TeucriComplexis,pp. 1sf.(= Opusc.
VII 378 f.).
 TELAMO

inpudenus).1 Stephanus and Column.a began the tragic quotation at non

habeodeniquenaua.,but most recent cditon follow Bothe and beginwhere the
transmitted text fallsinto trochaic rhythm, ie. at sed.Many scholan never-
theless seemto believe that Cicero' s sentence conveys the substance if not the
exact wording of thefirst part of the tragic specch.3
A parallel
. ______ · betweentrue prophecy and__q~ery
distincnon ··--1- u• ma~.J_ at

Euripides, El. 399-400: Ao~(ovyapCµm6o1IXP11aµo(, ~oov & µavT1-

ld\vxafpe1vloo. But if thetragic hero had madeanythinglike Cicero' spoint
4
it is difficult to undentand why most of his speech should be paraphrased.
Aut insaniaut inertescovers all classesof diviners, those like the Pythiawho
prophesy in ecstasy and those like CalchasandTeiresias who employ natural
signs. Telamo clearly denounced all classes.Cicero's use of superstitiosus and
hariolusas straight-out terms of insult conforms with thearistocratic usage of
his time. But in the Latin of the second century the two words had very
precise meanings. Superstitiosus meant 'clairvoyant' .sIt couldnot have been
usedat that time as a near synonym of inpudens.The harioluswas a diviner
who used different methods from the haruspex.Plautus (Amph. 1132, Mil.
693,Poen.791) and Terence (Plwrm.708-4))couple them in lists of diviners.
~
Hariolus( hariolari)is nowhere recorded in second-century Latin with its
later sense 'talker of nonsense'. 6 Vatesseems to be recorded only three times
before the first century (Ennius, Ann. 214 uersibusquosolim Fauniuatesque
canebant,38o satinuatesuerantaetatein agunda?,Plautus, Mil. 911 bonusuates
potnasesse;namquaesuntfuturadids)but not even in first-century usage is it
ever unambiguously a term of insult. 7 Second-century aristocrats might well
have denounced superstitio,uatesand harioli,especially as clairvoyancy was
not one of the methods employed by the state diviners, 8 likewise the profes-
sional sortilegi,
augures,haruspia.s,
astrologi,
amiectores and interpretes
somniorum
whom less privileged people consulted. However, we should not expect a
tragic hero using their language to blur the distinction they made between
one diviner and another.

• Cf. Diu. 1.42 (fr. xvm).

3
This phrase occursnowhere elsein Cicero but occasionally in drama; to the
examples quoted by Fcstus, p. 166.11 ff. add Plautus, Batch. 1102.
3 Cf. M. S. Salem,JRS xxvm (1938), 56-9.
4 Cf. Phoin.954-41.

s See above on v. 35.

6
Sec Latte, Rom. Rtl. p. 268.
7 See J)ablmaon, Philologusxcvu (1948), 337 ff., F. Bickel, RhM XCIV
(1951), 257 ff.
8
The Sibylline oracles of course depended historically on clairvoyancy but
their contemporary kccpcn did no more than consult the books in which they
were recorded.

397
 COMMENTARY
Vahlen was right to give sedto Cicero rather than the tragedian but should
have gone further. We might grant that Ennius could have composed the
phrase superstitiosiuatesinpudentesque
haiioliwithout the particular colour that
thecontext ofCiccro's discourse gives it but the rude and undigested pile of
adjectives and adjectival clauses in suptrstitiosiuatts inpudmusque lulrioli. ••
monstrantuiam looks like the work of a hasty quoter rather than that of a
competent dramatist. 1
What particular diviner Ennius' Telamo had in mind it is impossible to
say. It looks as ifTelamon had been given some advice on what course of
action to take (or not to take) and, after rejecting it, had launched into a
general denunciation of divination and diviners. The classof divincn men-
tioned-those who diuitiaspollictntur-d.onot appear in the remains of Attic
tragedy; there is some point in Vossius' remark 'sed mihi ut et amicissimo
Scriuerio omnes hi uersus socci potius quam cothumi esse uidentur '. We
have perhaps yet another instance of Ennius turning aside from thedrama he
is adapting and commenting upon contemporary affairs (see above on
fr. llXXIV).

266 inertes: 'without ars,i.e. skill acquired by training'; cf. Varro, Men.
359 artem. .. expromisinertem,Horace, Epist. 2. 2. 120-8 praetulerimsaiptor
delirusinersqueuideri. .. quamsapere.

quibus egestasimperat: cf. Plautus, Asin. 671 quiduisegestasimperat.

For the bad moral effects of poverty cf. Euripides, El. 375--0aXA' Cxe1
voaov I mv{a, s·
616acn(El &v6pa Tij XPE{~KCXK6v, Terence, And,. 275
coactumegestateingeniuminmutarier. The word egestashas a bad odour in the
public orations of Cicero (cf. Prou.cons.43, Phil.2. 50, 2. 62 et al.); for him
and his aristocratic audiences it denoted one of the prime causes of political
radicalism.

267qui aibi semitam sapiunt alteri momtrant uiam: cf. v. 221

DOD
qui ipsesibisapiensprodessenonquit nequiquamsapitand my notes thereon.
Semitadenotes a side-track, uiathe highway; cf. Plautus, Cas.675sciensde
1-1ia
in semitamdegredere.
Semitam... sapiuntis an odd phrase; its nearest parallel is Plautus, Pseud.
496 recteegomeamremsapio.Desire for concinnity with monstrantuiammay
have been Ennius' motive. On theother hand one could understand a mon-
strarewith sapiunt;cf. Plautus, Asin. 248 nam si mutuasnonpoteroartumst
sumamfaenore,354-5 si erumuis Demaenetum,I quemegonoui, adduce, Capt.

1
Cicero likes the collocation of harioliand uates(cf. Diu. 1. 4, 2. 9, Nat. deor
I· 55),
 -TELAMO.
~38 pol egosi te audeammeumpal(tm nominem,303 trreminiquomdictobaud
audebat;f aaonuncLitddtlictt,Pseud.120 si neminemaliumpoterot140m u,ngam
patrem.
. Monstrarewas perhapsstill redolent of its origin in the sacral language
(~ mon.strum); it occun twice elsewhere in tragedy, 21 times in Plautus but
only once in the rest of comedy (Terence,.Ad.570); it is absent from Caesar
and rare in Cicero. Ostnukreon the other hand seemsto be common in all
genres in all authon both during the second century B.c. and later.

268 quibm diuitiu pollicentar, ah iii dracl.arn•rn ~plipetunt: the

venality of divincn was a commonplace of Ionian epic1and Attic tragedy.2
The Roman aristocracy always drew a fum distinction between the priests,
who advised thesenate and magistrateson the will of thegods and who were
recruited from wealthy famili~ of Latium and Etruria, and itinerant foreign
professionals.3 It is not therefore very surprisingthat the adapten of Attic
tragedy should here and elsewhere4 allow the advisenof the Greek heroes to
beidentified with money-grubbing quacks.
Plautus, Mere.m, Pstud. 85 ff., 8o8 make it dear that for second-century
Roman audiencesthe drachut1U1 was a coin of very little value. Coined metal
was a source of imagery for the Attic tragedians (c£ Euripides, El. 558-9,
Med. 516-19) but the names of particular coins do not seem to occur in the
remains.
For the mcer at the lowness of the diviner's charges c£ Acschylw, Ag.
1273-4 K<XAOVµM}& q>OlTQS 00Sayvp-rp1a I irrooxos TaAatVaA1µo8vr'\s
fiveox6µ11v (Cassandraspeaking),Juvenal 6. 546-7 atreminutoIqualiacumque
uolesIudoeisomniauendunt.

269 sibi dedacant c:lrachamarn : 'let them deduct their drachmafee' ; c£

Cato, Agr. 144. 3 id uiri boni arbitratudeducetur
. .. si non praebueritquanti
conductum
erit aut locatumerit, detlucetur;
tanto minusdebebitur.

:--eddautcetera: 'and hand over the remainder according to the bargain';

c£ Plautus, Bacch.329-30 id signumstcum Theotimo,qui eum illi adferet,I ei
aurum ut reddat.

• Cf. Homer, Od. 2. 178 ff.

2
Cf. Sophocles, Ant. 1033 ff., 0. T. 380 ff., Euripides, &. 2sstf.
3 Cf. Plautus, Mil. 692--4 (a passage of specificallyRoman colouring), Rhet.
inc. Her. 4.62, Cicero, Diu. 1. 88-92, Propcrtius -4.1. 81-2, Livy 25. 1. 8,
Gcllius 1-4.1.2.
4 Accius,Trag.169-70 nil credo auguribusqui aurisunbis diuitantI alimas,suas
ut aurolocupletmtdomos,Ennius, Trag. 3-43 qui sui quaestuscausa.fiaassuscitant
smtentias{Stephanus' attribution [p. 129] of this vcnc to the Telamois not even
plausible).
399
 COMMENTARY
(e) The two propositions •the gods do not exist' and •the gods do not
interfere in human af&in' are different from the propositions •the gods are
not powerful (whatever they may have beenin the pastY and 'the gods do
not exercisetheir power justly'. They remove one of the foundation stones
of the heroic story. Not surprisingly the first appcan only thrice in the
remains of Attic tragedy1 and the second not at all. Plato refers to the second
at Nom. 10. 88sBand it was commonly associated in later times with the name
of Epicurus.i It appearsatMenander,Epitr.726-8 01£1TOO'CX\IT1'}VTOVS 8eovs
&ye1voX,OATJV, I COOTETO KCXKOV Kai Taya6ov Ka8' -fiµipclv I wµeiv
il<acrr(t>;and Plautus may have taken Mere.4-7 Noctiaut Dii I aut Soli aut
Lunae. .. quospol egocredohumanas querimoniasInontantif acerequiduelint,quid
non uelintfrom Philemon's •Eµiropas.It is possiblethat the original of the
Telainowas composed sometime in the fourth century or later and contained
the Epicurean sentiments in question, equally possible that Ennius, with his
fondness for philosophical speculation, introduced them birnsel£

270-1 deumgenus . .. caelitum. .. hvrnaanm genus: for thefirst peri-

elvai yevos,
phrasis c£ Euripides, Hek. 490 6oKOWTQS6<XlµC>V(A)V Virgil,
Aen. 4. 12 credoequidem,nee uanafides, genus essedeorum;for the second
Euripides, fr. 898. 13 ~lOV ••• YEVOS, Plautus, Poen. 1187 luppiterqui
genus eolisalisquehominum;for both Lucretius s. 11s6 diuomgenus huma-
numque.Plautus parodies the way of speakingwith hisgenuslenonium(Cure.
499, Persas82).

270 dixi et dic:am: c£ Bacchylidcs1 .1S9 q,aµl Kai q,aa(A),Plautus, Mil.

1os8 dixi hoetibidudumet nunc dico,Terence, Hee.722 iamdudumdixi ~
nuncdico.The figure appearsoccasionallyin Attic drama3and with consider-
able frequency in Roman.4 The dramatists were probably aping the magis-
trates' manner of speech; c£ the examples collected from early public
inscriptions by 0. Altenburg, N]bb.Suppl XXIV (1898), 494 £

1
Euripides, Helt.488-91, fr. 286, Trag. Grace. inc. 46s.
:a Cf. Cicero, Nat. dtor. 1 . -4-3-s.
3 Cf. Euripides, Tr. ,f.68iraax(I,) TE Kai imrov8cxK&n mfaolJ(XI,499 ot(l.)v
ffllXOV ciwTE TEV~IJCXI.
4 Without searchingsystematicallyI find Pacuvius, Trag.297, Accius, Trag.
227-8, Plautus, Aul. 216, Bacch.1087, Capt. 392, 933--4,,Cist. 12, 43, Cure. 168,
Men. 118, 991, Mere. 539, Mil. IOS8,Persa777, 847, Trin. 56, 106,619, Terence,
Eun. 1009, Hee. 722, 739.

400
 TELAMO

CXXXV

Telamo addressesTeuccr whom he suspectsof having murdered or partici-

pated in the murder of Ajax with a view to usurping the kingship.
:172tcil,u: thenormal form in republican drama.

natmn ingenuum Aiacem: cf. Plautus, RJU/.738 haecest natdAthenis

ingenuisparentibus.The jurists use ingenuumnasd as a technicalterm.
Ajax was bom of Eriboea, daughter of Alcathous who received Telamo
when he fled to Salamis from Aegina (Pindar, lsth. 6.45, Bacchylides
13 .102, Sophocles,Ai. s69); Teuccr on the other hand was thebastard son of
Hesione whom Herades had given Telamo as part of the spoil of Troy
(Sophocles, Ai. 1299 ff.).

cui tu obaidionem paras: language proper to the description of attacks

on walled cities; parodied at Plautus, Asin. 280, Mil. 219, 222, Most. 1048,
RJU/.838; for more elaborate parody of the tragedians'hyperbolic use of
militarylanguage cf. Plautus, Baah. 232, 709 ff., 92s ff., 1094, Epid. 158 ff.,
Pseud.585 ff., 761 ff.; for similarhyperbole in Attic tragedy cf. Euripides,
Or. 762 c:xrrt-Epel
ir6A1sirpos t,cep<l>v a&>µa,rvpyripovµE6a.

CXXXVI
Bcrgk's restoration of the name Aeacus1 seems certain; c£ Pindar, Pyth.
8.98-100 Aly1va cp{Aaµanp,~fXt) aT6Aft>I 1T6AlV Tav& K6µ13E f.l
Kai~ aw Al<XK4'> I TI11Mt'TE K&ya8ci>TEAaµ~VlavvT. ·AxiAMt,
Apollonius Rhod. 3. 363-4 TEAaµwv6' 6ye KV6faT010I AICXKOO ll<ye-
yaws· ZEvs 6. AICXKOV CXVToS rnKTEV, Ovid. Met. 13 .22-8.
Nonius' remark 'uel clareat' is peculiar;after claraest one would expect
'clara sit'. In any case the verb clarareon the few occasionsit occurs in re-
corded literature is, as one would expect, transitive. Clarereleads a shadowy
existencein poetry and is regularly intransitive. The fragment cannot there-
fore be taken as the beginning of an asseverationof the type itd me di ament
ut .. .•2 It may come &om a statement made by Teucer on arrival in Salamis
expressing pride in his ancestry and joy at having survived the voyage &om
the Troad; i.e. at a point in the drama before Telamo's anger breaks out.
1
RhM m (1835), 73 (PACI ~ (A)BACI). TOVIS for IOVIS was another mistake
in capital script. .
a The formula is very occasionallyvaried with ita me di amabuntetc. but such
phrasesdo not justify Vahlen's nam ita mihi... esseest atquehoelumen candidum
clartt mihi (ut egomornshuius in.sonssum) (SB Berlin 1888, 32-40 [ = Ges. phil.
Sehr.II 263-71)).
401 JTO
 COMMENTARY

274 hoe lumen candidnmclaret:cf. Plautus, Aul. 748 luciclllfoderipi-

lucecu,raet amdida,
amus,Amph. S47 inlucescat Com. pall.inc. 70 priorire luci
clarononqueo.Theverb cu,rereseemsto be a tragic neologism.
With hoethe speaker would have pointed to the sky; cf. Ennius, Trag.301
aspicehoesublimecandens,342 hoequodlucet,Pacuvius, Trag. 86-7 hoeuide
circumsupraquequodcomplexucontinetI terram,Plautus, Amph. S43 luascithoe
iam.
CXXXVII
1
Lipsius thought that this was a statement that could be made only by an
Orestes. It could be made by a monarch asked by suppliants for protection
against potential enemies of his city (cf. Aeschylus' 'IKETf6es,
Euripides'
'Hp<XKAEt6<n). None of the dilemmas suggested for Ennius' tragedy (e.g.
whether or not Telamo should banish Teuccr, whether or not Teuccr should
lead a rebellion against Telamo) seemssharp enough.

275 deam... pietu: •the gods whom I revere'; Ennius seeksparallelism

of phrasewith ciuiumpudor.

sentit: •sententiam dicit, censet'; cf. Catullus 64. 21 tum Thetidipateripse

iugandumPeleasensit,Cicero, Orat. 195 egoautemsentioomnesin orationeesse
quMipermixtoset amfusospedes,Fam. II .21.2 cum ego sensissemde iis qui
extrcitushaberentsententiamJerri oportere,Virgil. Aen. 10. 62.2-3 si . .. meque
hoeita poneresentis.

porcet pador:cf. Pacuvius, Trag.6'7mignllteut uereareloquiporcetpudor.

Theverb porcereoccursthree times in republican tragedy but is absent from
comedy.
CXXXVIII

Tecmessa, Ajax's mother Eriboea and Teuccr's mother Hesione have all
beensuggested as the person to whom the speaker refers. Eriboea seems to
me the most likely. Pacuvius, Trag.422-3 jlexanimatamquamlymphataaut
I
Bacchisaais commotain tumulisTeucrumcommemorans suum must refer to
Hesione; two opinions are possible about 313-14 quae desiderioalumnum
I
paenitudine squalesscabresqucincultauastitudine(from Teucer).

276strata terrae: •stretched on the ground'. There seemsto be no com-

parable dative in the rest of republican drama. 1 Accius has at Trag.praet.25
prostratum terra.Tmae may have beenwritten as an artificialvariant of humi
(cf. Cicero, De orat.3. 22, Virgil. Aen. 9. 7S4, 10. 697 et al.).
1
EpistolicarumQH_aestionum
Libri V {Antwerp, 1577), IV 19.
2
See P. Lejay, MSL xu (1903), 85 ff.
402
 TELAMO
sqa•larn! not transmitted elsewherein Latin; restoredat Plautus,Men.838
and True. 934; squalidusoccun three times in comedy, three in tragedy.
Leumann's intcrpretation1 of the phrase as 'squal- et sord-idam', 'ein
sprachlichesKunststiick', is hard to takebut there seemto be no parallelsfor
the formation.
CXXXIX

Teucer complainsthat Telamo suspectshim as well as the Atreids of being

implicated in the death of Ajax. It is impossible to get a reference to the
charge once levelled against Telamo of Jcilljngbis half-brother Phocas into
the words transmitted.

277 eandem me in 11Upicionem: monosyllabicprepositionsapart from

cumdo not normally follow personalpronouns; the only other two recorded
instances,Plautus, Mil. 1265 nesciotu me ex hoeaudierisan non,Cicero, Tusc.
2. 1 s huncpost,are suspect?

partiuit: the deponent form does not occur in republican drama.

CXL
Telamo agreesto hear Teucer's defence.

2,S more antiqao: c£ Plautus, Rud.624-s qui VeneriVeneriaeque anti-

I
stitoe moreontiquoin custodelmnsuomcommistruntcaput,Varro, Men. 303
ontiquomore siliamium amfedmus,Lucretius 2.610-11 lumc uoriaegentes
antiquomoresocrorumj ld4eomuodtantmotrem,Livy7. 2. 11 iuumtushistrionibus
f obellarumaaureliaoipsointersemoreontiquoridiculaintextouersibusioctitore
coepit.

aadibo: only here in tragedy; at Plautus, Capt.619, Poen.310, Ennius,

Com. 4, Caecilius, Com. 24, 113 in comedy; audiometc. seem to have been
the regular forms (twice in tragedy, 9 times in Plautus, s in Terence).

atque aaril tibi contra atend•• dabo: repeats the substance of the
prding 4Ullibo; for the pleonasm c£ Aeschylus,Choe.4-5 TVµ~ 6' hr'
~ T~& 1CTl()Vaac.> ,ra-rpl I~ve1v, &Kovaoo,Euripides,Phoin.919 oVK
llcAvov, OVKf\Kovaa. Contrais to be taken &iro Kotvov with audiboin the
sense of 'w l,ltpe1,3
in turn, having had my say'; c£ Terence, Ad. so.

1
Lat. Gramm.1, p. 225 n. 1.
:a See Marouzeau, REL xxv (1947), 312.
3 Cf. Euripides,Kykl. 253 &lcovCJOV
tv ~pet.
 COMMENTARY

Vahlcn's attempt to interpret the phrase as 'I shallanswer you' 1 clashes

with the common usage of the phrasesalicuialiquiduttndumdare/rogare."
The phrase auresdarewas probably a colloquialism;3 likewise auribusuti.4
The two together doubtlesslyhad a pompous tone; c£ Plautus, Bacch.995
aurium operamtibidi"', Mil. 954 aurismeasprofectodedoin didonemtuam.

CXLI

:179abnaebunt: probably an artificialform modelled on prohibebunt;

in
the rest of republican drama abnuereand adnuerebehaveaccordingto the
pattern of the third conjugation.

TELEPHVS

The title Telephusis given to Ennius six times by Nonius5 and once by
Verrius, to Accius fourteen times by Nonius (pp. 13.17, 136.18, 155.9,
174.14,226.3,227.29,307.31,347.14,352.8,426.22,485.9,488.3,488.10,
503.28), 6 twice by Priscian (Gramm. 11512.4, m424.24) and once by
Macrobius (Sat. 6.1.57).
Telephus, the son of Auge and Hercules. was king of the Mysianswhen
Agamemnon brought hisfirst expedition againstTroy to Asia Minor. He led
the Mysiansto victory over the Greeksbut sufferedhjmsclfa wound which
would not heal. After an oracle had informed him that 6 Tpoo<T<IS Kai
lcxan<XI, he went to Greeceand eventuallyreachedArgos where Agamem-
non was preparing his secondexpedition. He agreedto guide the expedition
to the site of Troy and Achilleshealed his wound by touching it with the
spear that causedit. Both7 Latin tragediesseem to have been set in Argos.

• Cf. SB Berlin 1888, 34 ( = Ges. phil. Sehr.u 265).

a Sec 0. Scyffcrt, BPhW XXIV (190-4),1325.
3 Cf. Plautus, Trin. 11, Cicero, Att. I. s.4, 2. 14. 2, Arch. 26, Persius 4. so.
4 Cf. Plautus, Mil. 799, and, where other people's can are concerned,
Cicero, Fam.7. 33 . 2 namet Cassiustuus et Dolabeluinosteruel potius uter~
noster studiis iisdem tenentu, et meis aequissimisutuntur auribus,Pliny, Epist.
7. 30. 3 qui auribusmeispostlongumtempussuo iureabutuntur,Symmachus, Epist.
1 . 3 . 3 si nobisutendasauresdatis,dicam,quiddiebussuperioribus egerim.
s The origin of the quotations at p. S37.24 ff. is obscure. The rest come from
Lindsay's list 10 'Ennius'.
6
The quotation at p. 155.9 comes from list 27 'Alph. Adverb', the rest
from list s 'Accius i •.
7 For the legends attaching to Telephus' name sec Robert, Die griech.
Heldmsagem 2 i, pp. 1138 ff.
 TELEPHVS

Aeschylus, Sophocles, Euripides, Agathoo, Clcophon, Iophon and

Moschion have thetitle T ,'\M<p<>S recorded against theirnames.Junius andP.
1
Lcopardus emended the fragment quoted by Nonius at p. 537.27 (&.
oa.m) on the assumption that Ennius adapted Euripides' tragedy. Euripides
was famous for having brought the Mysian king to Argos clrcucdin a
beggar's rags.
J. GcePintcrprctcd theother &agments on the same assumption.The quite
considerableamount of Euripides' tragedy which 1w come to light during
the present century contains nothing destructive of this assumption.
Duntzcr3 assignedthe Eooiao verses quoted by Cicero at Tusc. 3 .44 (&.
CLXXV; sec above on &. XXVII h)-pol mihifortutu,magu nunctkfit quamgenus.
I ~ regnumsuppetebatmi ut $CUI$quantoe loco,I qutmtisopibus,quibusde
rebuslapsaf ortun11 accidat-to a speech by Telephus without suggesting a
context. The Euripidcan Tclcphus seemsto have left hiskingdom of hisown
&ccwill and to have assumed thedisguiseof a bcggar.4 However theLatin
words could have beenuttered by a Tclcphus suspectedas a result of his regal
mannerisms of not being a real beggar and trying to prevent hisparticular
identity being discovered; c£ Accius, Trag.619-20 namsi a meregnumFor-
tun11atqueopesI eriperequiuitat uirtutemnee quiit (Ribbcck: nequitcodd.).
HartungS assignedthe piece of dialogue between Agamemnon and
Menclaus quoted by Cicuo at Tusc.4. 77-quis homote exsuperauit umq,u,m
gentium inpudmtid?::quisautem malitia te?-to Ennius' Telephuson the
grounds that there was a quarrel betweenthe two brothcn in Euripides'
play.6
To the alleged quarrel in Ennius' TekphusRibbcck7assignedvcncs quoted
by Seneca at Epist. 80. 8, quodnisi quierisMenelaehacJextraocddes,andan
anonymous metrician at Gramm.VI 613 .9, proinJemet(Lacbroaoo:proinde
et cod.) abste regimenArgosJum est potestdscon.sili(Lacbroaoo· consulendi
cod.).
It is possible that when his identity was revealed the Boni.anTclcphus

1
Emmdationum et Mis«l"1neonanLJbri Viginti. Tomus Prior (Antwerp
1568}, vm 11,.
a IJt Ttlepho EuripidisCommentatio.I have seenthe copy sent by Geel to G.
Hermann and now in the Cambridge University Library. The date of publica-
tion is usually given as 1830.
3 Zeitschr.f. d. A.It. 1838, 53. Cf. Ladewig, A.Ml. seen.p. 25.
4 Cf. Aristophanes,A.eh.44c:r1, which a scholiastalleges to be 1KTT1M'°'1

Evpnrl6ov (fr. 698). Horace, Ars '}6-1 seemsto refer to a tragedy which repre-
sented Telephus as banishedfrom his kingdom.
s EuripidesRestitutus1, p. 202.
6
Cf. Schol Arist. Neph. 891 (fr. 722), Stobaeus3. 39. 9 (fr. 723).
7 ~- sctn. p. 26,4, Die rom. Trag. p. 109.
 COMMENTARY
seizedthebaby Orestesand took refuge at an altar.1 Ribbeck assigned to such
a scene the verses quoted by Cicero at De orat. 3 . 102 nam sapiensuirtuti
lwnorem petit.Istd quiduideolfmo Sdtptus possidet
prtitmiumluiuJpraetlmn sedis
$IIO'OS.
Any play about Telephus set in Argos would have to end with Tdephus
being appointed to guide the Greek army to Troy. Ribbeck assigned to a
speech by Tdephus at thispoint two iambic pieces quoted as from Ennius by
Cicero at Off. I • sI : homoquimanti comitermonstratuiam,I quasilumendesuo
lumineocantkJt fadt; Inihilominusipsilucetcumilliaccenderit(fr. CLXV), and by
Varro at Ling. 7 .89: si uolesaduortere animumcomitermonstrabitur (fr. CXCVI).
The scholia make it plain that Dicaeopolis' denunciation of the warlike
policies of the Athenian government at Aristophanes, Ath. 496-5 56 parodies
an attack by the Euripidean Tdephus on Agamemnon and Mendaus.
Timpanaro has suggestecP that the words quoted as from Ennius by the
DanielincServius on Aen. 2. 62-Ut uos nostrilibtriIJefe,uumt, prouostrouit4
mortioccumbant obuiam(fr. CCXXII )-<:ome from an adaptation of thisspeech.
None of the assignations I have describedis without probability but it
must be remembered that many tragic heroes could have claimed to have
lost their kingdom, that a quarrel betweenAgamemnon and Menclaus took
place in Ennius' Iphigenio,that the justice of the war against Troy was ques-
tioned in this play and that in the Akxondn a suppliant at an altar was
threatened with death. Furthermore there is a lack of hard evidence that
Cicero knew Ennius' Teltphusand a constant possibility that Seneca and the
metricians quote imperialrather than republican tragedy. I have therefore
printed under the title Teltphusonly the pieces quoted with this title by
Vcrrius and Nonius.
CXLII
Geel interpreted thisvene as spoken by Tclephus, whilestill unrccogniscd.to
theGreek princes; he compared theverses of Euripides, '111 µ01 cp8ovfloirr•,
c!n,6pEs·ru11VCA>V 6Kpo1,I El 1M'"'XoS~v mA11K·h, {a6AolcnvAfye1v
(703), parodied by Aristophanes at Ath. 496-8'111 c!n,6pEs
µ01 q>8oVf1011T•, ol
8Eooµevo1, IEl1M'OOXoS ~v trm-r• w •A&,,valo1sAfye1vI ~oo mpl Tfis
1TOM<A)S,3 E. W. Handley makes4 the speaker someone trying to silence
Tclephus. This is much more plausible. The reply of Aristophanes' chorus
to Dicaeopolis-ss1-s &A118Es ooirfTp11tTE Kal µ1apclrrcm; I Tmrri au

1
Cf. Aristophanes' parody of the Euripidean scene at Ach. 326 ff., Thesm.
689 ff.
a SIFC N.S. xm (1947), 73~-
s For parallel sentiments cf. Euripides, Ion 670 ff.
4 BICSL SuppL v (1957), p. 34-

406
 TBLEPHVS

TOAµ~ Tr'l'CA>)(oS
oov-flµ(is)Jy&1v;-may reflectthe Greekthat Ennius was
adapting. Muttireand piaculumare both strongly abusivewords, unlikdy to
be usedby a person. even one apparently destitute, of his own behaviour.

:a8o mattire: •speak out of place, at the wrong time, in the wrong
company'; normally applied to thebehaviour of slavesin comedy.

plebeio: elsewherein republicandrama only at Plautus,Pom.sIs; pkbs is

likewiserare (Ennius, Trag.388, Plautus, Pseud.748, Caecilius,Com. 185);
populusand popularison theother hand are fairly common. Pkbs was prob-
ably alreadyin the early secondcentury cnnfined to certain set phrasesof the
constitutionallawyers.

piacalam •t: cf. Plautus, True.220-3 nosJiuitemistummeminimus otque

istepauperesnos: I uortnunt sesememoriae; stultussit qui id miretur.I si eget,
necessestnospati: amauit,aequomtif actumest. Ipiaculumst misererenoshominum
rti malegerentum.Elsewhere the associationsof the word piaculumare un-
questionablysacral;c£ Varro, Ling. 6.29 diesfastiper quospraetoribus omnia
uerbasinepiaculoliatfari,Gellius10. IS .10 (fi-omFabiusPictor)siquisaduerbe-
randumducatur, si adpedemtius supplu procubuerit, eodieuerberari piaculumest.
The dramatistsmay simply be using a sacralword that had become one of
general abusein the common language (cf.µuxp6s,scelestus) but it is possible
that they are consciouslygiving a Roman cast to their personages'sentiments.
The notion that theinherited classstructure, distribution of politicalpower
and constitutionalforms were divindy sanctionedwas firmly rooted in the
thought of the republican aristocracy.The annalists(c£ Livy 4.1~, 6.39-
41) representedboth sidesin thestruggle of theorders as taking seriouslythe
question of divine approval for projected socialand politicalchange. A suni-
lar ideology appears in Ionian epic (c£ n.2.196-7 8vµc!>s & µfycxsml
61arpecpiwv(3aalAf1CA>V, I Tl'1fls•be lu6s la-n, q,1A&t Si I µ11Tin'aZ&vs,
204~ ElsKofpavosIO'Tw,I els~. 4>C&>KE Kp6voviroosayt<VAo-
µtreoo I O'ld\1TTp6v T' ft&8tµ1cmxstva aq,fm ~VAEVTJO'I) but it is note-
worthy that neither Homer's Odysseus (ll. 2.246--64) nor his quiescent
lower orders (2.271r-7) bring religious arguments against the behaviour of
Thersites. Attic tragedy on the whole represents the power of princes as
resting on popular consentrather thandivine grant. Sophocles'Agamemnon
deniesthe right of the bastard Teucer to be heard (Ai. 1226--63)but defends
his own authority in naturalisticterms. The fragmentssuggestthat the princes
of Euripides' Tn~ spokt'!similarly.1
1
C£, however, Aeschylus,E.um.625 ff., Euripides, Or. u67 ff.

407
 COMMENTARY

CXLIII

The beggar's rags in which Euripides drcs.,cd the Mysian king causedan
immediatesensationand were probably a necessarypart of all later dramatic
presentationsof the story; c£ Aristophanes,.Ach.432 ff. and the scholia,
Diogenes. Epist. 34.2 T~ TE TOV•HpcxK}Jovs flv{Ka els •Apyos
,rapeyM:To, ,rc,Av xefpovt 0-XflJ.laTtTOV flµnipov iµcpavtaefivat
~x• &a.i,f~ATlOTpa(Burges: &a.i,t~A11Tacodd.) O'OOJ.laTOS Aaj36VTa
fxml &A1m1p1aTVXTlS(~ Euripides,fr. 697), Accius, Trag.613-16 quern
ego ubi aspexi,uirum mmwrabilemI intui uidner, ni uestitustlleter uastitudoI
maestitudoproediurenthominemwe, 617 nam etsi opertus~litate est luctuque
lwrrifiubili(Vossius:lwrrificaUcodd.).
The sourcefrom which Nonius drew thesetwo quotationsis obscureand
Lindsayput a questionmark againsthissuggestion•that they arc made in the
order in which they occur in the text of the play. Strzelccki-ihas repeated
Lindsay's suggestion much more dogmatically. However even where his
own exccrptionsare concerned (e.g. in the case ofLucilius) Nonius some-
times seems to quote in reverseorder. 'Lindsay'slaw' can be a treacherous
guide to the student of republicanpoetry. The trimetcr regnumreliquiseptus
merulidst.o'4could come from a speechof Tdephus revealinghimsdfto one
of the Greek.personagesof the play (soVahlen and Strzelecki)but Gccl's
assigruationof it to the prologue seemsmuch more likely.
281 s-fn•Jidasaeptaa stola: cf. Sophocles,0.K. 1597 IAvcn 6vamveis
o-ro).~ (of Oedipus), Euripides, R.hes.503 TITOOXtK'l'\vl)(oovo-roA,;v,712
{>aKoSv-rc+> o-ro).~11'VKao&fs (of Odysseus).The word stolais usedthrice
clscwberein republican tragedy apparently like the Attic o-roA,;but docs
not occur in comedy. In the mid fust century it denoted the particulardress
of thearistocraticmatrona.3
S«pire is properly usedof fences,wallsetc. (cf.Plautus, Cure.36, Ennius,
Trag.88); for Ennius' hyperbolec£ Plautus, Cas.921-2 soepituesteid quiestis
(mulieres)(add. Loman). I ubi iliumsaltumuideoopsoeptum, rogout altao siruJt
ire.
CXLIV

The phrase hateenodariwas taken by Geel to refer to Mcnelaus' desire to

continue the war and by Welcker4 to Telephus' defence of the Mysian
resistance. WcckleinS interpreted it much more plausibly as Telephus'

~ Eos XLII,fasc. 1 (1947), 30-5.

1
RhM LVII(1902), 202.
3 Cf. Horace, Sat. 1.~.69--71 numquid ego a te I '""l"" prognatumdeposco
consultamnum I uelatumqutstolamta cumconftrbuitira,99 ad talossto'4dnnissaet
circumdatapalla.
4 Die gritch. Trag.p. 488. s SB Mundaen1878, vol. 2, 221.
408
 TELEPHVS

explanationof theoracle 6 T~ Kai lave ,en and put the &agmentin a

speech by Achillesrejecting the pleas of the Achaean leaden to help his
enemy Telephus and rcaUing the number of Greeb killed by Tdephus'
Mysiansduring the earlier expedition.

283 leto dati: a phrase from thesacrallanguage, 1

talcenup by Pacuvius
(T,ag. 148), Lucretius(s.1007), Virgil (Am. 5.806, 11.172, 12.328), Ovid
(Met. 1.67oet al.) and othcn; varied by Plautus with mortidare(Ann. 608,
Mere.472; c£ Horace, Sat. 2.3.197) and ad mortemdare(Amph. 809),by
Acciuswith lttoofferre(Trag. 117) and ltto mittere(491; cf. Plautus,Capt.692
te mortimisero).

284aunt: not normally separatedfrom its complementby a heavymetri-

cal pause.Vahlen compared Plautus, Capt. 884-s quidtu per barbariau urbis
iuras?-quuz enim item asperaeI sunt ut tuom uictumautUltUINS use, Tcrencc,
Ad. 330-2 me miseram.quidiamaedas aut quoiaedas?nostrumnekschimun, I
nostramuitam omnium,in quonostraespesopesque omnn sit« I ertlnt?qui sine
hac. .•• An excited tone of voice is perceptiblein both the comic pamgcs.

enodari: three times elsewherein tragedy,once in comedy (Turpilius14).

CXLV

Geel,Scholl,2 W clckerandJahn3reconstructeda sceneof Euripides'Tfl~

in which the hero appealedto Clytcmnestrafor help and was advisedto seize
the baby Orestesand fly to the altarat the moment of discovery.4 Wccklein
however showedSthat none of theevidenceadduced demandsthe presence
of Clytcmncstra at any point in the action and argued that the seizure of
Orestes (parodied by Aristophanesat Adi. 326 f[ and Thesm.689 f[) was
Tclephus' own sudden stratagem.
Thisleavesthe word illamin the Ennianfragment,which Geeltook to refer
to Clytcmncstra,a little difficultto handle on the suppositionthat Enniuswas
adapting Euripides'tragedy. However the collocationaduorsumillammihi at
1
Cf. Varro, Ling. 7. 42 in funeribusindidiuisquo dicitur•ollus leto "4tusest',
Festus, p. 304. I ff.
a Btitr. pp. I 34 ff.
3 Telephosund Troilos.Bin Brief an Hmn F. G. Weldterin Bonn(Kiel, 1841),
pp. 16 ff.
4 Cf. the story of Themistoclesand the wife of Admctus ('Thucydides
I. 136 f.).
s SB Mundten 1878, vol. 2, 201 ff.; cf. Wilamowitz, GriechischeDichter-
fragmentt n (Berlin. 1907), pp. 69 ff., Robert, Dk griech.Htldmsagtm 2 i,
p. IIS7 n. 8.
 COMMENTARY
the beginning of an iambic trimetcr is odd1 if not impossiblci and mihiseems
quite otiose. Textual corruption may therefore bar all possibility of under-
standing the Latin fragment, to say nothing of using it to reconstruct
the original Greek tragedy.

285 _pro6teri: once elsewhere in tragedy (Trag. inc. 52) where the
quantity of thefint syllableis indeterminate; three times in comedy where
the syllable is twice short (Plautus, Capt. 48o, Terence, Eun. 3) and once
long (Plautus, Men. 643).

286 aduonum: 'in the presenceof'; cf. Plautus, Bacch.698 immosi audias
tibiet al.
quaedictadixit meaduorsum

CXLVI
Geel took Ennius' trimeter to be a version of the Euripidean verse quoted by
the scholiaston Aristophanes,Ath. 8-t<CXK&',s 6Ao(ar" (Dobree: 6AotT' &v
codd.) • &~1ovyap•rua61 (720)-and made Telephus the object of a curse
uttered by Achilles.A view of thiskind is much more plausible than that of
Welcker, according to which thedisguisedTelephus is uttering a curse upon
himsel£The verse of Euripides quoted by the scholiast on Aristophanes,
Ath. 446-K<XA&')sfxo1,.n,T11A~ s• ~oo cppovw(fr. 707)-is more
oblique than the Eonian 3 In any case, to the ancient mind curses were
dangerous thingslikely to rebound on the head of the deliverer even when
directed elsewhere. Ennius' Telephus would have had to be an extraordi-
narily enlightened character to make so light of the power of evil words as
to speakthem against himsel£

'1B7qui ilium di deaeque: the regular order of words in cursesand bless-

ings introduced by a particle; cf. P. Langen, RhM XII (1857), 428 f.
~; ablative is restricted in republican drama almost entirely to curse
formulae; even these are much more frequently introduced by ut or have no
particle at all
Di deaequewas a traditional Roman way of referring to deity (seethe
examples collected at T.L.L. v i 909.14 ff.); the Attic dramatists used a
similarformula very occasionally(cf.Aeschylus,Theb.88 looloo8eol8Ea(-re.
93-4, Menander, Sam. 184-5Tots 8eots... Kal Tats 8Eais).
1
See above on "". 1~.
2
Cf. Plautus, Epid. 179 nequesexta aaumna ambior Herculiquamilia mihi
obiectast(iambic tetrameter) and F. Skutsch, Plaut. u. Rom. p. 136 n. 1.
3 At Ach. 509 ff. Dicaeopolis cunes the Lacedaemoniansand at Tham.
466 ff. Mnesilochus says he hates Euripides. A speech by Euripides' devious
hero lies behind both the comic speeches.
410
 TBLBPHVS

magno mactulint malo: c£ Afranius,Com.tog.264 di te nultt4ssint tMlo,

Pomponius, Aull. 137 at te di omnescum consiliocaluem«tas.sintmalo;the
normal dramatic curse however employs the verb perdere.Ennius' phrase is
yet another example of the use of auspiciouswords in inauspiciouscircum-
1taoccs.1Behind mactareseem to lie such phrases as macte hacill«e ddpe
polucendaesto(Cato, Ag,. 132. 1) and macteestouirtute(Pacuvius,Trag.146),
which appealfor an augment of strength.2 The phrasetkumaliquomactare(c£
Cicero, Vat. 14 cumpuerorumextis deosnumismactaresol.eas) probably pre-
ceded in use the regular clamcalaliquiddeomactare.It is possible that the
oxymoron was a coinage of popular slang rather than of tragedy, for
Plautus usesinfortuniomaaareand similar phrases too often for one to be
comfortable in labelliogthem paratragic.

CXLVII
The Aldinc editor changed incendtreto incendierwhile Merccrus declared
incendtreto be used intransitively. T.LL. VD i 870.22 produces only one
dubious example of this use: Ps. Hieronymus, in Joh. p. 58oD maxime
inandentezelo Iuddeorum. Vahlen suggestedthat the subject of incendtre was
expressedin the following verse.
Geel interpreted the fragment as spoken by the chorus after the quarrel
betweenAgamemnon and Mcnclaus.But ciuitatem ... Argiuummust refer to
the whole Argive state, not just its leaden. Vahlcninterpreted it as spoken
by one of the Greek princes after the disguisedTclephus' defence of the
Mysian resistanceand compared Aristophanes' parody of Euripides: Ach.
S76-7 & Aaµo:X:, ovyapovros &vepwtroS,r6.AcnI airaaav ftµC,vniv
ir6A1vKCXKoppo8&1; ~
( Euripides, fr. 712 ). But incendtrein metaphorical
usagewas no equivalentof KCXKoppo8Elv; it denoted the stirring up of passion
(c£ Plautus,Asin. 420 quisemperme iraincendit,Pseud.201 nimissermonehuius
ira inandor,Terence, Phorm.185-6 quodeiusremediuminueniamirocundiae? I
loquarne?incendam). Telcphus' speechwas meant to allay passion.

288 ciuitatem ... Argiumn: 'clues Argiui'; c£ Cicero, Font. 14 quid

coloni Narbonenses?••. quid Massiliensiumduitas? For the type of peri-
phrasisc£ Euripides,Or. 612 ClacAflTOV
•ApyelCA>v6xAov,846 •Apyelov •••
~v.
1
Seeabove on fr. LXXXVI.
3
SeeH.J. Rose and 0. Skutsch, CQ XXXD (1938), 220 ff., XXXVI (1942),
IS ff.

411
 COMMENTARY

CXLVIII
Geeltook thisverse to refer to the journey of theGreeb to Troy. Scarcely a
happy idea. W clckerinterpreted it as spoken to the Greeks by Calebas about
Tdephus' journey to Argos at the behestof Apollo. A view of this type is
likely to prove correct. Handlcy1compared the acolic verses uttered by
Euripides' chorus afterthe decision of Agamemnon and Menelaus to accept
Tclephus as guide on a second journey to Troy {Pap. Bero!. 9908, col 11 7-10
ot yap Te[y]ECXTIS fiµtv 1·ruas, ov(x)lMvafa, T{K"n1I vavravCJWTIVl
6~ &G'>vI Kai mµmiip• &A{oov Ape-rµwv)but the operative phrase of
theseverses, ow TIVl ... &G'>v, seemsto have a much wider reference than
Ennius' deum. .. deconsilio.

289itiner ... conatam: cf. Pacuvius, Trag. 44-sJokt pigetquemagis

""'tisqut mt conatumhoe nequiqumnitiner,Virg~ Am. 10.685 ttr conatus
utramque
uiam.
The form itineroccurs 6 times in tragedy against ittr twice; comedy on the
other hand has ittr IS times, itineronly 3 (Plautus, Mac. 913,929, Turpilius,
Com. 207).

THYESTES

The title Thyestesis given to Ennius by Cicero, Vcrrius and Nonius; to

Pacuvius by the unreliable Fulgentius2 and to Gracchus by Priscian (Gramm.
II 269.8 ). Cicero once (Orat. 184) names the title alone. I have followed all
editon since Columna in giving the verse here quoted to Ennius' Thyestes
(fr. cu). It seems to me possible that the words attributed to Pacuvius'
Thyestesby Fulgentius, nonillicluteisaurorabigis(uulgo: biiugiscodd.) are the
work of neither Pacuvius nor the grammarian but come from Ennius'
tragedy.
The stories told about the feud between the two sonsof Pclops, Atrcus and
Thycstes, over the kingship of Mycenae were many and various.3 Most
students have accepted Welcker' s view4 that .Ennius' tragedy was set in

1
BICSL Suppl. v (1957), 38.
:a Smn. ant. S7 Pacuuius in tragoediaThyestis (uulgo: Tietis (ti&Zis R:
Tiethis D: tegetisBE) codd.). On the character of Fulgcntius' quotations of
republican poetry see Wessner, Comment.phil. lenmsesVI 2 (1899), 135 ff., F.
Skutsch, RE VIIi (1910), 219, Timpanaro, SIFC N.S. XXII (1947), 199 ff.
3 Sec Robert, Die griech.HtldtnSllgt 1, pp. 28s ff., 293 ff., A. Lesky, WSt
XLID(1922/3), 172 ff. (= GesammelteSchriften[Bern, 1966), 519 ff.).
4 Ztitschr.J. d. Alt. 1838, 229 ff. ( = Die griech.Trag. pp. 678 ff.).

412
 THYBSTBS
Mycenae and showed how Atrcus p1misbc,\Thycstcs for an act of adultery
with hiswifeAeropc by fint banishinghim, then recallinghimand offering
him at a feast the limbs of his murdered children to eat.
I propose to argue that these events took place before the action of the
Thytstts and that Eonius' tragedy was set at the court of Thcsprotus in
Epirus. Thyestescursedhis brother and inquired of Apollo how he might be
avenged. Apollo replied that a son sired on his own daughter would be the
avenger. In the meantime drought struck Mycenae and Apollo told the
inquiring Atreus that it would only end when Thyestesretumed. Thycstcs
came to Epirus where his daughter Pelopia was being cared for by king
Thcsprotus,saw Pelopiaunprotected and raped her (doubtlesdy in ignorance
of her true identity).1 Atreus arrived in Epirusjust afterwards,was struckby
Pelopia's beauty and took herin marriage from Thesprotus, believingher
to be Thcsprotus' daughter. The rape of Pelopia and her marriage with
Atreus in my view formed the action of the Thytstn.
Ladewii seems to have included not only these events but also certain
that took place at Myccnac many yean later in his reconstruction of the
Thyestes.Pelopia exposedin the mountains by Myccnac the childshe con-
ceived as a result ofThycstcs' assault.This childwas found being rearedby
hc-rdsmcn and acceptedby Atreus as his own with the name of Aegisthus.In
young manhoodAegisthusdiscoveredwho his real father was and fulfilled
Apollo's prophecy by ki1lingAtreus.
Of the piecesquoted by Nonius as from Ennius' Thyestes four (frs.CIJV,
CLV, CLVI, CJ.IX)have a content relevant to the discussionof the theme of the
tragedy.
Ribbcck,3who accepted Wek:ker's view of the plot, interpreted fr. cuv,
theumeafortun4ut omniain me conglomeras mt1'4,and fr. ewe, qM"1fl mihi
maximehiehoJiecontigtritmt1lum, as spoken by Thyestes after he discovered
what he hadeaten at thefeast. In thelanguageof both fragments (seebelow,
pp. 4-24, 426) there is an element of oxymoron implying that the speaker's
woes are not unmixed. I would therefore suggestthat Thyestes' discovery
that he hadlainwith his daughter was thecause of his remarks; confidence
that Apollo's oracle will be fulfilledtempers horror at the thought of incest.
Welcker interpreted fr. CLV, set me Apollo ipsedekclilltluctatDelphicus,as
referring to an oracle delivered during Thyestes' fint period of exile (after
his intrigue with Aeropc) and advising him to return to Mycenae. This
1
Two versions of the story of Thyestes' incest are con&tcd at Hyginus, Fab.
87-8. Robert (Die gmch. Htldtnsdgt 1, p. 299) rightly treated the reference to
Sicyon at 88. 3 as an error.
:a Atuil. sun. p. 38.
3 Dit rom. Trag. p. 203. The fragments are not mentioned at ~st. sun.
pp. 26'7£
413
 COMMENTARY
oracle is not recorded anywhere in the mythographical tradition. There are
two versions of the story of his return; according to one he returned secretly
in quest of vengeance ;1 according to the other he came back openly on
Atreus' invitation. 2 I should interpret fragment CLV as either from a speech
by Thyestcs about the oracle concerning hisdaught.cr3 or from a speech by
Atreusabout the oracle telling him to recall Thyestcs from hissecond spell of
cxile.4
Ribbeck interpreted fr. Cl.VI, sinjlacalnmt(Gulielmius: injlacabuntcodd.)
condiciones
rtpudiatott redJiu,,as spoken by Thycstcs to supportcn of a plot to
overthrow Atreus. Ladewig,who believed that Thyestcs' recognition of the
grown Aegisthus was Ennius' centraltheme, argued that the words of the
fragment are taken most naturally as those of a father betrothing hisdaughter
and that the only possible marriage offered by the later history of the
Pclopids is that which Atreus contracted with Pclopia in the belief that she
was Thcsprotus' daught.cr.S He neverthclcssfailed to realisethat his acute
observation nccessarily set Ennius' tragedy outside Mycenae and at a time
some years before the discovery of Aegisthus' identity.
Thycstcs was a personage of at least two other republican tragedies, the
Atreusand the Pelopidaeof Accius. The Atreusquite certainly dealt with the
feast while the meagre remains of the Pelopidatcan be very plausibly inter-
preted6 as deaHog with the recognition of Aegisthus. IfThyestcs had any role
in Accius' Chrysippusit could only have been a minor one.
Cicero nowhere names the title Atreus1but frequently quotes from the
Accian tragedy that bore it. 8 Where utterances by Atreus are concerned.
those at Off. 3. 102 and Tusc.4.55 (c£ De orat.3 .217) are quoted along with
Accius' name; those at Sest. 102 (c£ Plane.59), Phil. 1. 34 and Off. 1. 97 are
as.,ociatedwith the ageof Sulla by Seneca at Dial. 3 . 20. 4; those at Nat. deor.
3 . 68, which certainly come from a play about the feast, were given to the
Atreusby Columna9 and have beenleft there by scholars ever since; that at
1
Accius, Trag. 198 ff.; cf. Aeschylus, Ag. 1583 ff.
2
Hyginus, Fab. 88. 1, Seneca, Thy. 288 ff.
J Hyginus, Fab. 87, Apollodorus, Epit. 2. 14, Schol. Stat. Theb. 1.694,
Servius, Vcrg. Aen. 11 .262; cf. Sophocles, fr. 226, Seneca, Ag. 28-36, 48-9,
294.
4 Hyginus, Fab. 88. 5. s Hyginus, Fab. 88 .6.
6
C£ Welcker, Dit griech. Trag. p. 370, Ribbcck, Q!yrest.seen.p. 335, Die
rom. Trag. 457 ff., Robert, Die griech.Heldensage1, p. 298.
7 At De orat.3. 217 the phraseAtreusJeretotusrefers to the speeches made by
a penonage Atreus; see above on fr. xcv.
8 A Virgilian scholium (Breu.Exp. Georg. 1. 1) ties the quotation at Tusc
2 • 13 to the Atreus.
9 Q. Ennii Frag.p. 419.
 THYBSTBS
Tusc.s . 52 was given by W clckcr to theAlreur and restored to theinart4 by
Ribbcck; the trimcter quoted at Pis. 82 is said by Asconius to have been
spoken to Atrcus by Thyestes in a play by Accius.
A verse whichis quoted four times in Nat. deor.(2.4, 2.65, 3. 10, 3 .40),
twice with Ennius' name, is quoted by Fcstusas from the Thytstn (fr. am).
Its context is obscure.
Columna gave thecurse of Thycstcs quoted at Pis. 43 and Tusc.I. 1o6 (at
the latter place with Ennius' name) to the tragedy Thymes (fr. a.). In
W clckcr's view the vcncs were uttered by Thyestcs as he preparedto leave
Mycenae.~The verb habtatsuggests to me that Thycstcs is speaking about
Atrcus to a third party.3 Drowning was thought by the ancients to be a
particularly terrible form of dcath4 and may have beenwishedon Atrcus as
thewont fate Thycstcs,gorged with his own children's B.csh, could think of
at that moment. However, the vcncs would have hadmuch more dramatic
force if at the time of speakingtheseaseparatedThyestcsfrom Mycenae and
Atrcus was known to be coming in search of him. One might compare the
prayer of Aeschylus' Danaids that Zeus should wreck the ships of their
Egyptian pursuers(Hilt.29-39 ).
The lyric dialogue between Thyestcs and a number of 'hospitcs' which
Cicero quotes at De Orat.3 . 164and Tusc.3. 26 was alsogiven by Columna to
the Thymes (fr. cxux). Wclckcr interpreted this dialogue as having taken
place between a chorus of Cretan women. the attendants of Atrcus' wife
Aerope,and Thycstcswhenthe ~ttcr emergedfrom the dininghall.Rib-
bcck, Mueller, Vahlcn and Robert held to Welcker's view, although, in the
mcantime,Sone ofits main supports, Valckcnaer's identificationof Euripides'
Kpf\aam with the 8via-n}s, had collapsed. If one believes that both the
Thyestesandthe Alreusdealt with the feast and acceptsW elckcr's interpreta-
tion of the dialogue there is no reason for assigningthe vcncs to one play
rather than the other. On the other hand my reconstructionof the plot of the
Thyestesand a more carefulinterpretation of the wording of theversesper-
mits a rational decision.
Warmington, whatever one might think of his reconstruction of the
whole play,6 was right to set the scene of the dialogue at the court of
Thcsprotus. Thyestcs has been standing on the stage for some time and

1
Zeitsthr.j. d. Alt. 1838, 221 ff., Die grieda.Trag. 3S7 tf.
:a Cf. Aeschylus, Ag. l~l µ6pov 6' 6:cpEpTOVTTu.cmi6cnsmevxncn, I
AaK"TIO'IJ(X&hrvov ~6h<CAlST18els~. Horace, Epod.s .86 tf.
3 Cicero writes penderesat Pis. ,u to suit his own discourse.
4 Sec Syncsius, P.pist.4, Immisch, RhM LXXX (1931), 98 ff. Sec below on
fr. CL. s Sec below, p. 418.
6
Remainsof Old Latin I, pp. 346 ff.; Warmington proposedtwo scenes,one
at the court of Atrcus in Mycenae,the second at that ofThcsprotus in Epinu.
41s
 COMMENTARY
addresseshis command to the chorus of 'hospites' as they enter. The scene
would be roughly parallelwith that opening the 'Hpal(Ml6at of Euripides,
where a chorus of citizensenter to find the foreigner Iolaosstandingby the
altar of a temple and arc addressedas ~• (78, 84, 93 ). 1 If Bentleywas right
in assigningthe trochaic vcncs Tantdloprog,uztus Pelopenatus qui quondmna
soau (Bentley: soaro codd.) I Otnomaorege Hippodameam raptis nanctus
nuptiisto the same play as the bacchiacs.they must come from the same
context; i.e. they arc spokenby Thycstcs either in a prologue speechor,
more likely, in a reply to a question from the 'hospices'. Wclcker's inter-
prctation2has the absurd result of putting this background information at
the end of the play at a time when the chorus would know quite well who
Thycsteswas.
W clckcrinterpreted the words quoted by Cicero at Oral. 184 as from a
Thyelta (fr. a.I}--qwmnam te essedicamquit4tdain smtctw-45 a question
addrcucd by Atrcus to a servant come to reveal Thycstcs' plans for revolu-
tion. They look to me like thequestionwhose answeris quoted at Tusc.3 .26
(fr. cxuxb).
Ao account by a tragicThycstesof the prnc:rolio~ of the feast is quoted
by Cicero at De oral.3.217 and Tusc.4.77: ipsushort4turmeftaterut meos
I
malismisa mandarannatos.Saivcrius gave thesewords to theThycstcsof
Accius'Alrtus, along with the words of Atrcus quoted in addition at Tusc.
4. 77: maiormihimolesmaiusmisandumstmalum, I qui illiusaarbumcoramtun-
damet comprinuun. The latter clearlybelong to a play whose principaltheme
was the feast. The former however need not come from the same play.
Indeed they look as if they come from an account: of the feast given by
Thycstcssome time aficrward.The immediatereactionof a father so treated
would have beenone of disgustrather than anger. The possibilityshould be
weighed that the words ipsushortaturmeftata ut meosttU1lis misertn411411llm
natosbelong to Eonius' Thyestes.
To illustrate a type of fallaciousargument the anonymous rhetorician
quotes as from a tragedy by Ennius two tctramcte~ho tu Ji quilnu(studiis
M) est potaw (estpotestIll M) motus(motumM) supmun atqueinferum,I
poam (paam enim M) inta sese(se E) conciliant, conferuntconcordiam (Rhet.
inc. Her.2.39 ""' fr. CLXII)andnamesthespeaker.Editon of the rhetorician's
treatise(includingMarx) make thespeakerCrcsphontesand most collccton
of the tragic fragments put the verses in Eonius' homonymous play (see
1
For hospitesin an address by a foreigner to citizenscf. Plautus, Poen.678.
a Hartung (EuripidesRestitutus1, p. 176) gave the trochaic verses to the
Atrtus and interpreted them as spoken by Thycstes on arrival at Mycenae near
the beginning of Accius' play. Ribbcck (~st. sun. p. 268, Die ,om. T,ag.
p. 201) separated them from the bacchiacs and put them at the beginning of the
Tlsyutts.
416
 THYESTES
above, p. 271). In 1888 Vahlcn1 pointed out that the manuscript tradition
(thesprotum M: threspontem [the.spontemd] E) was best interpreted so as to
make Thcsprotus the speaker and assigned the verses to the Thyestes,with-
out, however, abandoning Welcker's view that the feast was the principal
theme of this play. Warmington accepted Vahlcn's assignation of fr. CJ.XII
and interpreted it as spoken by Thesprotus to Atreus when the latter came to
Epirus in search of Thyestes. Nevertheless, in view of the conflict in the
manuscript tradition, I have left fr. CLXD among the incerta.
In the rhetorical section of hiswork Charisius quotes two pieces of tragic
verse addressed to Thesprotus: at p. 364.21 ff. illustrating the trope 'sarcas-
mos' -rite Thesprotumpudet (Grotius: pudeatcod.) Atrei (Bucchcler: rei
cod.) quodipsea Tantaloducatgenus;at p. 374.8 ff. illustrating the figure of
thought 'transmutatio pcrsonarum'-Thesprote si quissanguineexortamtuo I
proleminta tarasst sacramimmolet,I quidmeritushiesit dubiuman cuiquam faat
(Fabricius:fiat cod.)? The distinction between figures of speech and figures
of thought makes its fust appearance in extant literature in the anonymous
treatise addressed to Hercnnius. The source of this treatise must have con-
tained examples of dramatic poetry to illustrate fully the two kindsof figure
but true to hisprinciples3theanonymous rhetorician quotes only illustrations
of vicious figures. At least two of hisillustrations turn up again and again in
work not directly dependent on his4and it is possible that hissource was also
theultimate source of most of the quotations of republican drama in the later
discussion of rhetorical figures, including those at Charisius pp. 364.21 f[
and 374.8 f£
Welcker assignedCharisius' two quotations to Accius' Pelopidae,5 offering
no interpretation of thefust and takingthe second to be addressed to Thes-
protus in hisabsence.Ladcwig6 seems to have accepted W clcker's interpreta-
tion of the second and gave both to the Thyestes.Ribbcck:7argued that the
two quotations must come from speechesdeliveredat the court ofThesprotus
in a play whose theme was the marriage of Atreus and Pelopia. Believing
that the Thyestesand the Pelopidaewere set in Mycenae he had no title to
attach to hispostulated play. If my reconstruction of Ennius' Thyestesis cor-
rect and if Charisius and the anonymous rhetorician draw on a common
source there is a good chance that the Thyesteswas the play quoted.

1
Ind. lectt. Berlin 1888/9, 17 ( = Op. ac. 1417). C£ Wccldein, Festschr.
Urlidas,1 ff.
3
4. 18.
3 4. 18 nam hie nihilprohibetin uitiis alienisexemplisuti.
4 2.34 (Ennius, Trag. 208-16), 4.18 (Ennius, Ann. 103,109).
s Die gritch. Trag. p. 370. See above, p. 414.
6
Anal. seen.p. 40.
7 ~aest. scm. p. 335, Die rom. Trag. pp. 628 £
27 417 JTO
 COMMENTARY

A number ofEnnius' tragic verses quoted by Cicero without the name of

the play or the speaker(Tusc.3 .44 [fr. axxv], Off. 1.26 [fr. CLJCIXb], 2.23
(fr. CLXXXII],3. 104 [fr. CLXXXIV], Rep. 1.49 [cuaxa]) have been more or
1
less tentativdy given to the Thyestes. The quotation at Tusc.3 . 44 can be
2
proved not to come from here. Contexts for the others can be imagined in
severalof Ennius' plays.
Many Greek tragedians have the title ~S recorded against their
names: Sophoclcs,3 Euripides, Apollodorus of Tarsus, Carcinus, Chaeremon,
Clcophon, Diogenes of Sinope and Agathon. V alckenaer4 suggested that in
Euripides' case~S was an alternative title of Kp11crocn,that Ennius
adapted this script, that fr. CUD, aspicehoesublimecandens quemuocantomnes
louem, was part of Ennius' version of the threeEuripidcan trimetcrs 6p~
T6v6" Cffl'Elpovat~a I Kai y;;v mpaefxove·vypaislv
-rovV\f,'OV
ay1<aAcns;I TOVTovv6µ13EZi;va, T6v6" f\yov8E6v(fr. 941) and that
the trimcten therefore bdonged to the Kpiiaaaa/~s.
RibbcckS dismissed the idea of a directlink between Euripides, fr. 941 and
Ennius, fr. CUD on the grounds that Cicero would not have made his own
version for Nat. deor.2.65 if this were the case.But Cicero's philosophical
source may have introduccc:l the Greek trimeters merely with the name of
Euripides; in that case Cicero would have been quite ignorant of their
specific origin. Nevertheless the similarity between Euripides and Ennius is
not as great as Valckenaer thought and the doctrine is such as might have
beenmentioned in many Euripidcan tragedies.
An inscription found in the Piraeus (I.G. n1 2363; c£ C.I.G. m 0047).
revealed that Kpiiaacn and 8vtoTr\s denoted separate scripts6
Most of the citations of 8vtoTr\s occur in anthologies of moral apoph-
thegms and give no information about the plot of Euripides' tragedy. A
scholium on Aristophanes, Ach. 433 enables one to deduce that Thycstcs
appeared on stage at one point dressedin rags. Two fragments of Aristo-
phanes' TTpoayoov (461,462) have been thought to parody a description
given somewhere in eviO"T11S of the cannibal feast.7 Even if this were the

1
SecRibbeck, Q!!aest.sun. p. 267.Vahlcn printed them all under the title
Thytstes in his first edition but only those quoted at Tusc. 3 . 44 in his second.
:a Secabove, p. 395.
3 The distinctive symbols, a,~ and y (Pap. Mus. Brit. 2110) and the epithet
ItKVoovtoSappear in various citations.
4 n·,am'be, pp. 12, 47.
5 Q!!aest.seen.p. 267.
6
SeeWilamowitz, Anal. Bur. p. 153 (cf. ObseruationesCriticaein Comoediam
GraecamStltdae [Diss. Berlin. 1870], p. 12).
7 SeeBergk, AristophanisFragmenta(Berlin, 1840), pp. 248 ff., Wilamowitz,
Anal. Bur. p. ISJ n. 4.
418
 THYESTES
case one should not conclude that Euripides' tragedy was set in Mycenae
and included the feast within its action. It remains a possible model of the
Ennian Thytsta.
CXLIX
(a) For the context of this fragment sec above, p. 415.
Bentley1 altered the abnutasof Cicero's manuscriptsto renutasin order to
procure a complete bacchiactetrameter; Vahlen in bis first edition altered
adirito adirierfor the same purpose. However such 'contracted' tetrameten
commonly occur in the company of full tetrameters; c£ Plautus,Ba«h.1127,
1128, Cas. 658, 662, 67s, 685, 691, 694, 69S,702, 703, Cist. 3S, Men. 763,
763 a, 771, Most. 125, 783, Terence, Andr.483; Pacuvius, Trag.202 and 340
can be scanned similarly.,

290 quidnam --.st• •• quod: a common variant of adverbialquid(quidnam);

cf. Plautus, Cas. 630, Cure.166, Epid. s6o,570, Men. 677,Most.69 (quidest
quidCD), Pseud.9, True.238,295, Terence, Haut.613,Eun. SS8, SS9, Ad.305.
te adiri abnutu: c£ Euripides, Herakles1218 Ti 1,101irpoae{oov xetpa
011µa{ve1s
cp6vov;
The intensive form abnutareoccurs only here and at Plautus, Capt. 611 in
republican drama; perhaps with some differenceof meaning from abnuere
(Plautus, Capt. 481, Mere.so, True.6); 'why do you keep nodding ... ?'

(b) Scriveriushad suspectedthe existenceof poetic rhythm in qualisenimtibi

ille uideturTantaloprognatusPelopenatus qui quondam a soceroOenomaorege
Hippodameamraptis tumetus nuptiis Iouis iste quidtmpronepos,and Bentley
procured trochaic verse from Tantalodown to proneposby altering soceroto
socru.The phrase Iouisistequidempronepos,however, looks like Cicero's own
interjection. In a tragic quotation one would expect the sequenceIuppiter-
Tantalus-Pelops (c£ Trag. inc. 101-3) or Pdops-Tantalus-Iuppiter (cf.
Ennius, Trag.273 ).
To procure three regular baccbiactetrameters Bentley inserted meobefore
tantauisscelerisin eorporehanet. Severalscholarssincehave tried to achievethe.
same purpose with different additions, none nearly so plausible.The words
in question form a unit of verse (cretic dimeter +trochaic metron) occasion-
ally found in Plautine comedy (e.g. Amph. 245, RJul.667).3 Bacchiacs
and creticssometimesoccur in the samemonologue or dialogue but it is diffi-

' Emendationes
Cic. Tusc. pp. 4S ff. a Cf. Strzelecki, in TragicaI, 64.
3 A. Spcngel detected severalPlautine examples in his De VersuumCretieorum
Vsu Pldutino(Diss. Berlin, 1861), pp. 43 f. but later recanted (Rejormvorschl.
pp. 10s ff.). Sec also Ribbcck, Coroll.p. xxxm.Leo, Die plaut. Cant. p. 17,
Strzelccki, in TragicaI, 65.
419
 COMMENTARY

cult to find swit.chesfrom one verse form to the other accompaniedby such
little apparent change of verbal _toneas there is here; Plautus, Men. S79 ff.
and Pseud.1126 ff. are as near to the form which Cicero's manuscripts give
Ennius' canticum as I can find.

291 Tantalo prognatas: 'the grandson of Tantalus'; c£ Plautus, Cas.

398-9 utiruimtuaquidem(ist4) (add. Guyet) sicutHerculeipraedicantI quondam
prognatis,in sortiendo sorsdeliqutrit.The word progru,tusoccursin the epitaph
1
of Scipio Barbatus (C.I.L. 1 7), comparatively often in early epic and
tragedy, 14 times in Plautus (only once however in trimeters [Capt. 170]),
not at all in the rest of comedy.1 It isnormally a high-falutin variant of natus.

IOCnl: c£ Nonius, p.
223 .21 socruset masculinogenereuetaes did posse
uoluerunt,Priscian, Gramm.n 233. 7. The normal forms in republicandrama
are socer(socerus),soari etc.

292 raptis nanctm nuptiis: cf. Livy 30. Lt,. 2 raptaepropt int,e,anna
nuptidt.
There is no need to restore nanctu'st;for this type of nominal phrase in
republican drama see above on v. 78.

293 nolite bospitesad me adire: thismode of expressing the negative

imperativeis very rare in republicmdramacomparedwith others; I count
only 21 instancesin Plautus and eight in Terence. The tone can be gathered
from the edictalformula nequisfedsse uelit?

ilico istic: compare and contrast Plautus, Bacch.1140 ilia, ambaemanete,

Mere.912 atqueistiesttJilieo,Most. 1o64 ilia, intra limenisti astate,Rwl. 836
illie astateilico,Caecilius, Com. 118 numeteilieo,Terence, Ad. 156 ilia, hie
eonsiste,Phorm.195 st4ilia,. A gesture by the actor (c£ v. 29() abnutas)would
have taken the place of the verb.

29-f contagio mea: 'contact with me'; c£ Plautus, Amph. 30-1 atque
egoquoqueetiam,qui Iouissumfilius, I eonttJgione
meipatrismetuomalum.In
4
Attic tragedy the sight.3 voice and physical proximityS of the polluted

1
See Fraenkel,Horace,p. 82 n. 4.
, Cf. D. Daube, Formsof Roman Legislation(Oxford, 1956), pp. 37 ff.
3 Sophocles, 0. T. 1424-8, O.K. 1483-4, Euripides, El. 1195-7, Herakles
1155-6, 1231, Or. 512-15.
4 Aeschylus, Eum. 448, Euripides, El. 1292-4, Herakles1218-19, Or . .µ8, 481.

s Aeschylus, Theb. 597 ff., Euripides, Herakles 1233, Hipp. 946-7, I. T.

1159 ff., 1226 ff., Or. 51.2-15.
420
 THYESTES
penon was generally thought to be dangerous. For the pollution of
touch cf. Seneca, Thy. 104-

umbraue obilit: for the idea that the shadow has personalityand power
cf. Pliny, Nat. 28 .69 magiuetanttius causacontrasokm lunamquenudariaut
umbramcuiusquamah ipso rtspergi,N.T. Act. ap. s.IS c:xrre Kai els TCXS
TO\JS&a&vets... Iva ipxolJM)VTThpovl(QVft 0'1Qa
b:q,4pE1v
'Tl'AarE{(IS
rnt0'1QaC71J
TlVlavTC>V.
CL
For the context of this fragment secabove, p. 415.
Bentley was the fint to recognise its metrical character, i.e. two trochaic
tetrameters followed by catalcctic tetrameters. Garatoni took the words
nauftagiotxpulsus uspiam1 as the end of a catalcctic tetrameter. Suchtetra-
meters, lacking in diaeresisafter the second mctron but with caesura after
the fourth arsis, arc transmitted in the remains of republican drama a number
of times? However, although uspiamoccurs only once elsewhere in the
orations (Fl«c.29),the phrase should be taken as Ciccro's own gloss on the
poetic matter he intended to quote verbatim)
The contents ofThyestes' curse, shipwreck, no grave or rest for the dead
man, are commonplace in Greek poctry4and Roman.S It would, however,
have sounded more horrible to Roman cars than to Greek; the Romans
allowed even criminalsburial. 6

it- mmmi1 suis 6sm uperis, eauceratul, Ilaten peaclem:

296--7
cf. Euripides, T,. 448-50 1<&µiTOl VEl<pOVq,apayyes yvµvas·~
PAT)µM')VI v6an xe1µapf)<t> f,iovacn, wµcp{oviriAas Tacpov,I &,,pal
ooxrov01v6aaaa6at, Virgil, Atn. 8. 668--C)tt le CatilinaminaciIpendtntm,
scopulo.
One might have expected saxisaffixus;cf. Plautus, Ptrsa295 le cruciipsum
adfigmtpropediemalii, &. 48.
Euisctrartoccurs only once elsewhere in republican drama (Pacuvius,
Trag.4); for similarformations secabove on v. 93.
For laterependtnscf. Plautus, Cas. 390 ptdibusptndtas.

~ tabo same et sanguineatro: for the syntax a, I, tt C cf. Plautus,

A.mph.1011, Asin. 571, Mere.548, 678--9
et al.

1
Before nau.fragio
cod. V has the scholium 'uersus Ennii '.
:a Sec above on fr. XCIX.
3 For this mode of quotation sec above on fr. xvm.
4 Cf. Archilochus, fr. 790 Dichl1, Sophocles, Ai. 1177.
S Cf. Catullus 108, Horace, Epod. 5.86 ff., 10.21 ff., Virgil, Atn. 4.620,
6
Propcrtius 4.5.3, Ovid, lb. 165£. Sec Dig. -i,8.24-
421
 COMMENTARY
For tabumand sanitsof the rcccndy dead cf. Virgil, Am. 8 .487, Valerius
Flaccus4. 749.
For sanguisater of the dead cf. Virgil, Georg.3 .221, Am. 3 .28, 33, 622,
Livy 38.21.9 et al.

298neqae sepulcrmnqao recipiat habeat: for redpereintransitive

(perhapsfrom the nauticallanguage)cf. Plautus,&ah. 294 rursumin portum
redpimus.

portam corpori,: the ideaof Hadesas the ~tµT}Vto which the soul passes
after the voyage of life is common in Greek poetry.1 The classicalRoman
writcn frequentlytook it up., Ennius' Thyestesexpressesmuch more primi-
tive sentiments.Whether they come from fifth-centuryAthens or early
second century Rome is hard to say.

299 remi111hamam aita: the phrase uitamredderemight have been

expected; c£ the alternation letodare/mittere
(above on v. 283).

corpus reqaiescat malis: c£ Virgil,Eel. 10. 33 molliterossaquiescant,

A.en.
6.328, 6.371, Tibullus2.4.49-50, 2.6.30, Ovid, Am. 3 .9.67-8. The Greek
poets frequently usedev&1vof the dead; c£ Homer, n.14.482, Sophocles,
O.K. 621 et al.
Requiescatpicks up redpiatand remissa;the compound is quite rare in
republican drama (Plautus, Capt. 505, Epid. 205, True.209, Terence, Eun.
405) compared with the simple form.

CLI

For the context of this fragment see above, p. 416.

The context of Cicero's discoursedemandsiliain Thyesterather than illein
Thyeste.The latter reading cannot be defended on the analogy of those
phrasesdiscussedabove on fr. xcv.
Bothe made Ciccro'squotation into a regular bacchiactetrameter by sub-
stitutingthe archaicform senectafor senectute.
But Cicero'swhole argument is
that the tragedian'swords would soundprosy without the accompanying
pipe music. In the mid first century senectawas a highly poetic word; it
occun only here in Cicero'swritings, not at all in Caesar's,only six times in

1
Cf. Sophocles, Ant. 1284, Euripides, Herakles770, Ba. 1361-2, Trag. inc.
fr. 369,Lconidas, A.P. 7.264.2, 7.4s2.2, 7.472b.1.
i Cf. Cicero, Tusc. 1. 118, S. 117, Virgil, Aen. 7. S98-9, Seneca, Ag.
S9l-2.

422
 THYBSTBS
Livy's and then for special effect.1 :Raa:biaa arc not always arrangedin
OT(xov: cf. Plautus, Men.S7I ft:,RMJ.
tetrameters KaTCX 18s ft:,Varro, Men.
40S,

300 quemnam ~.eeae dicam ,m: a pompous variant of quu tu esqui (cf.
Plautus, Epid.637, fr. 44). The corociliansquite often insert Jicdm+accusa-
tive and infinitive in direct and indirect questions of a solemn or mock-
solemn character;cf. Plautus, Cas.616 'I"" tgo huntamortmmi use auidicam
datum?,Cure.1 quotedhoenoctisdicamproficisdf oras?,Mere.s16 quidMfflffl
tibi dicamtsst?, Most. 1~ quodid tsst dicamuerbumnaud nesdo,Pseud.7#
sedquidnofflffltsst dicamtgo isti seruo?,Trin.2 sedfinemfort qutmJicamnesdo,
True.70-1 quosquitkmquamadrtm diaunin argtntariisI referrthabtrtnisipro
tabulisnesdo, 689 quamtsst Jicambanebtluam?,Terence, Phorm. 6s~
f actrttgo huncan malitiaI dicam. .. inartussum.The-manner of
utrum stultitia
speechmay have originated in tragedy; c£ Euripides, Rlsu.31H) Tf C7Eq,,&;I
wov~1v;, fr. 1 ,ro{av 01 q,,wµev yatav lKMA01ir6TaI ir6M,
~oeai TiJ&;

tarda in senectate: c£ Plautus,P0ttt. so!HJfagitduisenes:scibamattatitar-

diores,Horace, Sat. 2. 2. 87-8 seu I Juraualttudoinddmt seu tardasenectus,
Virgil, Am. 9.61er-11 neetardasenectusI tkbilitatuiru animimutatqutuigortm,
Tibullus 2. 2. 19-20 dumtardasenectusI inducatrugasinficiatquecomas.

CLIII
The context of thisfragment is obscure.
In hissecondedition ofEnnius' remainsVahlcn withdrew hisadherence to
the view that Ennius had written aspicthoesublifflfflcantltns
. .. 1 but argued
that Fcstus' lemma was SVBUMEN. The existence of this word is highly
dubious and Virgil, Georg. 1 .242 does not exemplify it. The reading
svBUMEN of some of Paulus' manusaipts can be explained as arisingfrom
the etymology which Paulus reports. The Fcstus manusaipt's SVBI.IMIIM
could come from a quotation in Verrius' lexicon omitted in Fcstus'epitome
without change of lemma.
For the identificationof thechief demcnt and thechief god cf. Euripides,
T,. 884-6 wyi\s 6xru1a1<&-n-l yi\s (x<A>v av,
16pav,I 6cms irOT·EI 6vcrr6-
TracrrosEl6wai, I Zeus,elT"avayK11cpvcnos el-n vovs~ii>v, fr. 877
6XA"cxl&i;pThml C7E,K6pa,I 2a,s asave~1s6vo1,1&:Je-rai,fr.9416p~
1
Sec R. M. Ogilvie on 2.-40.5 (Oxford, 1965).
~ Cf. Scaligcr, Now ad VarronisLibrostieRe Rustica(Paris, 1573), p. 235,
Ritschl, RhM VII (1850), 556 ff. ( = Op. D 462 ff.); for the contrary view see
W. Heraeus, Philologus1v (1896), 197 ff., Hafficr, Ciotta xxm (1935), 251 ff.
 COMMENTARY

TOVU\fK)V-rov6' &m1pov aleipa I Kai yiiv fflpl~ exove·

vypats w
&yK&Ams;I TOVTOVvo,.nlEZfiva, T6v6' flyov 8e6v, Philcmon, fr. 91.4
mip&, &v TIS6voµacmeKai fl(a, Ennius, Trag.356-7 isticest is luppiter
quemdico,quemGraeduoamt I aerem.For other pantheistic re-interpretations
of divinenames cf.Euripides,&. 275-6, Phoin.685-6, fr. 781. 11-13, fr. 944-
It is noteworthy that at fr. 941 Euripides makes the chief element the
fluid lower aanosphere. 1 The Stoic philosophers nevertheless interpreted
these verses as representing the doctrine they took from Empedocles,
according to which the outer fiery element of the universe was identifiable
with l.Evs.The source of Probus, Verg. F.cl.6.31 corrupted part of Euripides'
Greek into TOVmp!ixoVTa vypcxv&yK&AmsTbv "°'113£Zfjva while
Cicero at Nat. deor.2. 65, adapting a Stoic source, turned vypats w&yK&Acns
into the vague tenerodrcumieau.It is therefore not out of the question that
Ennius had fr. 941 or another Euripidcan passage of similarcharacterbefore
him when he wrote aspicthoesubli~ cantknsqwm uocantomnesIouem.For a
fairly dear case of the importation of philosophical ideas foreign to the
drama beingadapted sec above on fr. ex.

301 hoe sublime candem ~ for the adjective and present participle in
asyndeton cf. Virgil, Georg.2. 377 grauisincumbens,
Aen. 3. 70 lenisaepitans,
8. 559 inexpletuslaaimans.

quem uocant omnes Iouem: for the gender of the relative cf. Lucretius
4. 132 in hoecaeloqui didturaer.
For UOCllfein the sense of nominarecf. Ennius, Trag.64 quapropter Parim
pa.stores
nuncAlexandrumuo<.ant. Cicero may have believed that Ennius wrote
inuocant;the context of his own argument gave support to such a belie£

CLIV
For the context of this fragment sec above, p. 413.
The strongest rebuke to f ortunain the rest of republican drama seems to be
Terence, Hee. 406 o fortuna ut numquamperpetuo's"4ttl.There is a certain
oxymoron in Ennius' expression that could hardly have stood in hisoriginal.
Whereas TVX11 usually denoted a purely neutral chance and 8e6sand 6af µoov
powers that were as likely to be hostile as beneficent, the f ortunaof the
republican dramatistswas a benevolently positive power.:i

302 conglomeru: only here and at Pacuvius, Trag. 200 in republican

drama; the simple verb gl~are doesnot occur at all; in das.ucalLatin it is
rather more frequent than the compound.

1
C£ fr. 839. a Sec above on fr. LXXDt.
 THYESTES

CLV
For thecontext of this fragment see above, pp. 413 ff.
Bergk1 was the fust to sec that the transmitted words could be-scannedas a
cretic dimeter followed by a catalectictrochaic dimeter; cf.Aeschylus,Choe.
s8s~. S94-S, Bum.491-2, 500-1, 956-7, 916-7, Pers.126-7, 133-4, Plautus,
Amph. 223, 233, Cas.628, Rud.677."
303 delectat dactat DelphiCUI:where two nearly synonymouswords
stand together in asyndetonthe shorter normally precedesthe longer but cf.
Plautus, Baah. 934 miseremale,Mere.681 Jisperiiptrii, Mil. 1204 donauiJedi,
Terence, Andr.248 contemptau spretau,855 eonfidenscatus,Haut. 404 Jisptrii
perii, Eun. 377 abJucJue,Pacuvius, Trag.263 retindetenett; in general see
Lindholm, StilistiseheStudim,pp. 8o ff.

CLVI
For the context of thisfragment sec above, p. 414.
304 sin Saccebuntcondiciones repudiato: the metaphorical use of
jlaccereis odd but cf. Cicero, Ad Q.f,. 2.15.4 MeSSllllajlaccet, 'Messalla's
candidature is not standing up straight, support for it is weak•. Ennius may
have been employing a current colloquialism;cf. Terence's use of stllrein
connection with plays that gainedpublic support (PJwrm.10, Hee. 15).
For eondicionem repudiatecf. Plautus, Trln. 454-5, Rhet. inc. Her. 4. 34,
Cicero, Q!!inct.46, Phil. 13. 37 et al.

CLVII
The context of thisfragment is obscure.
305 sonitm auris meu pedampalmincrepat: contrast Plautus, Cure.
203 sonitumet crepitumelaustrorumaudio,Aul. 811, Cist. 543, Cure.229, Mil.
1393, Rud.661, Trin. IC>93;compare Homer, n. 10. 535 hnrwv µ• OOKV-
ir6&>v &µcplK"l'VIT'oS oOcrra ~,. Sophocles, Ant. 1187-8 Ka( µe
cp86yyoS ob<.elovKaKOO I ~ 61' &rwv, Accius, Trag.479-Sosonitau . ••
peruasitaurls,Plautus, Amph. 33 3 hincenim mihi dextrauox aurisut uidetur
uerberat,Virgil, Aen. 8.58~3 g,auiorneu nuntiusaurlsI uolneret,12.618-19
impulitaurisI confusaesonusurbis,Seneca, MeJ. 116 aurespepulithymenaeus
meas.For the alliterativepedumpulsu cf. Virgil, Aen. 12. 334-5 gemitultima
pulsuI Thracapedum,445 pulsuquepedumtremitexcitattllus,Livy 27. 37 .14
1
Philologusxxxm (1874), 274 (= Kl. phil. Sehr.1 342).
:a W. Christ (Mttrlleda GriechenundRoma•, Leipng, 1879, pp. 407 ff.) seems
to have been the fint to set out the Plautincexamples of this unit of verse.
425
 COMMENTARY
sonum uods pulsupedum modulantes.For petle (pedibus)pellere(pulsare)c£
Ennius, Ann. 1 Musae qu« pedibusmagnumpulsatisOlympum, Lucretius
s.1402, Catullus 61. 14.Horacc, Conn. 1.4. 13, 1. 37. 1-2, 3. 18 .15-16, Ovid,
Ars 1 • 112, Seneca, Oed.433 et al
CLVIII
The context of thisfragment is obscure.

3o6 impetrem &cile ah animo ut cernat: c£ Livy 43. 23. 8 temptareeum

destititcumappareretquantumin eopratsidiiessetneetamenimpetrareabanimo
possetut impensamin remmaximiadomniamomentifaceret; Ennius' phrase was
facilea me ut cernam(c£ Caelius ap.
perhaps an devated variant of impetrem
Cic. Fam. 8.12.1, Petronius 52.5).

CLIX

For the context of thisfragment see above, p. 413.

Vahlen 1 was the first to perceive that the words transmitted form a cretic
dimeter followed by a catalectic dimeter; c£ v. 303.
There is in the phrase contigeritmaluman oxymoron parallel with that in
v. 302. Unlike Greek TV)'X<lVEIV Latin contingereis normally used of
fortunate events; c£ Caper, Gramm.vu 98.8 acdderealiquidaduersididto,
contingerealiquidpulchri,a statement which applies quite well to the usage of
republican drama.

307 hie hodie: c£ Plautus, Cist. 16, Most. 1129, Persa710, Rud. 1417,
Terence, Haut. 162, 176, Eun. 230, Soo.

CLX
The context of this fragment is obscure.
The iambic word paratstanding in the fifth foot of the iambic trimeter
offends against the laws established by Bentley and Luchsa regarding word
division in the final metron of thistype of verse and similar ones. Only four
other tragic verses (Enniw 388, Acciw 442,618, Trag. inc. 21) offend in the
same way. Nevertheless there seems to be nothing against paral from the
point of view of usage; c£ Cicero, Verr. 1. 67 id agiatqueidparariut . .. , Fam.
2. 5. 2 eaparameditare cogitaquaeessein eo dui acuirodebent,Accius, Trag.634
atqueid egosempersic mecumagitoet comparo.
308 putat: 'reputat'; c£ Caecilius, Com. 41 non hateputas? non ha« in
cortleuersanturtibi?,Virgil, Aen. 6. 332 multaputanssortemque
animomiseratus
iniquam,8 . 522 multaqueJurasuo tristi cumcordeputabant.
1
Ind. lectt.Berlin 1888/9, 9 n. ( = Op. ac.I 408 n.).
a In G. Studemund, Studia in priscosstriptores LatinosI, 8 ff.
426
 ADDENDA
p. s On performances in foreign languages at Roman ludisee Mariotti,
Belfagorxx (196s). 37.
p. S3 n. s On Cicero's methods of quoting republican poetry see also
E. Malcovati, CimDM t 14poesia(Pavia, 1943 (Ann. d. Jae. di Lttt. t Filos.
d. Univ. di Cagliari,xm]). pp. 89ft:
p. 167 Phrases such as uasus Bnnianusand htmistichiumBnnii in the
Virgilian scholia refer to Virgil; the Ciccronian scholiast's Bnnianum
htmistichium,however, can only refer to Ennius (Mariotti).
p. 169 Treating deum as a monosyllable would produce a normal
diaeresis (Mariotti).
p. 176 Maleuolentes famam tollunt,beneuolentesgloriamcould be Achilles'
reply to an ambassador's summamtu tibipro nuilauitafanuim txtolles tt pro
bonaparatamgloriam(Mariotti).
p. 192 Vahlcn linked the words uitaecrudatumin his •index scrmonis •.
p. 193 F. H. Sandbach, however, reminds me of &TTOTVµ1rav1aµos.
p. 213 Neque. .. inuitamcould be defended as a colloquial lapse by a
speaker intending to say neque. .. non inuitamand being distracted by the
positive attribute dementem.For such lapses in accumulations of negatives
sec Mariotti in Lanx saturaN. Terzaghiobldta(Univ. of Genoa, 1963). 263.
p. 227 Menandcr, Sam. 140, 159, 196 are not strictly relevant: Chrysis
alleges,and Demeas believes, the child to be her own. Plutarch. Mor.489 P
is hard to explainaway (Sandbach).
p. 246 However for molossic words at head of cretic tetrameter see
Plautus, Most. 109 et al. (Mariotti).
p. 294 Mercier must have intended ui as an ablative of quality; c£
Ennius, Ann. 303-4 Corntliussuauiloquenti I ore Cethegus,Accius636 Tereus
indomitomoreatqueanimobarbaro(Mariotti).
p. 362 For the undesirability of those who uolucrisemperspe et cogitatiDM
rapiuntura ~mo longiussee Cicero, Rep.2.7.
p. 372 For Probus' interpretation cf. Rutilus Namarianus 1.57 qui con-
tinet omniaPhoebus.
p. 388 Maritocould be interpreted as "ita ut eius maritus sit' (Mariotti).
p. 40S n. 4 The common version of the two myths to which Horace
refers could be restored by takingpauper et txul with Telephuset Ptltus
as a form of "double zeugma' (Brink).
I serpat.
p. 423 C£ Lucretius 1.414-15 ne tardapriusper mtmbrasenectus
 BIBLIOGRAPHY
I list (a) those collectionsof the fragmentsof Ennius' tragedieswhich have
furthered in some measurethe establishmentand W1dcrstanding of the text
and (b) those editionsof the sourceswhich I have usedin 1.11aking rny own
collection. Work pertaining to particular tragedies or fragments is not
listed here but cited where it has relevance.Serial tides are abbreviated
accordingto the system of L'Annlt philologique.

(a)
STBPRANVS, R. & H. Fragmenta poetarumueterumUltinorumquorumoperanon
extant. •• undiquea Rob. Stephanosummadiligtntiaolimcongesta,nunc
autemab Htnrico Steplumotius filio digesta... {Geneva [?], 1564),
pp. 108-28.
CoLVMNA, H. Q. Ennii poet« uetustissimi
quaesupersuntfiagma,14abHiero-
nymo Columna conquisitddisposil4et txplicata ad Ioannemfilium
(Naples,1590), pp. 305-475.
DELRIVs, M. A. Martini Antonii Delrii ex sociel4ttIesu synl4gmatragoediae
Latinae(Antwerp, 1593), pp. 9{r107, 163-7.
P. PetriScriueriicolleet4nea
SCIUVlllUVs, ueterumtragicorum. . .fiagmtntdet circa
ipsa not« breues.quibusacceduntsingularilibellocastigationes et notae
uberiores GerardiIoannisVossii{Leiden,1620), pp. 8-38, 13-00.
BoTHB,F. H. Poetdt sctnid Ultinorum:recensuitFridericusHtnricusBothe:
vol v:fiagmenta (Leipzig,1834), pp. 23-78.
RmBECK:, 0. 1 Scenic« R.omanorum poesisfiagmenta:recensuitOtto Ribbeck:
vol. 1: tragicorumLatinorumreliquiae(Leipzig, 1852), pp. 13-62,
248-78.
VAHLEN, J.1 Ennianaepoesis reliquiae:recensuitIoannts Vahlen {Leipzig,
1854), pp. 91-150.
RmBEa:, 0. 1 ScaeniCAtR.omanorum poesisfiagmtnta: secundiscurisrecensuit
Otto Ribbeck:vol 1: tragicorum fiagma,ta {Leipzig,1871), pp. xvi-
xxxviii, I 5-75.
MVl!I.LER, L. Q. Enni carminumreliquiae •.. emendauitet adnotauitLudanus
Mueller(St Petersburg,1884), pp. 91-135, 212-4+
RmBECK:, 0. 1 Scaenicae R.omanorum poesisfiagmenl4: tertiiscurisrecognouit
Otto Ribbeck:vol. 1: tragicorum fiagmtntd (Leipzig, 1897), pp. 17-
85.
VAHLEN, J.1 Ennianaepoesisreliquiae:iteratiscurisrecensuitIoannts Vahlen
{Leipzig,1903), pp. cc-ccxi, nS-203.
428
 BIBLIOGRAPHY

W AllMINGTON, E. H. 1 Remainsof Old Latin: newly editedandtranslated

by
E. H. Warmington:voL I: Ennius and Coedlius (London, 1935),
pp. 217-383.
Kl.on, A., SBBL,0., Vorr, L. Sc«nicorumR.omanorum
Jragmmta:voL 1:
tragicorum
Jragmmta:adiuuantibus Ottone Seel et LudouicoVoit edidit
A!fredusKlotz (Munich, 1953), pp. 44-111.
W AllMINGTON, E. H. 1 Remainsof Old Latin: newly edittdandtransl4Udby
E. H. Warmington:voL 1: Ennius and Coedlius: revised edition
(London, 1956), pp. 217-383.
(b)
(Aero]
in carminaHoratiicommentarius, ed. 0. Keller (Leif'rig, 1902).
Apulcius
prose de magialiber(apologia), ed.R. Helm (Leipzig, 1905).
metamorphoseon libri,ed. R. Helm (Leipzig, HJ07).
demundoliber,ed. P. Thomas (Leipzig, 1908).
de deoSocratisliber,ed. P. Thomas (Leipzig,1908).
Aurelius
'ad Frontonem epistulae', ed. M. P.J. van den Hout (in M. C~lii
Frontonistpistuloe[Leiden,1954]).
Catullus
carminumliber,ed. R. A. B. Mynors (Oxford, 1958).
Charisius
artisgrammatia,e libri,ed.C. Barwick (Leipzig, 1925).
Cicero
Acatkmicorum priorumlibersecundus, ed. 0. Plasbcrg (Leipzig,1908).
ad Atticum tpistularumlibri, I-VIII ed. W. S. Watt (Oxford, 1965);
IX-XVI ed. D.R. ShackletonBailey (Oxford, 1961).
pro L c~lio Balbooratio,ed.A. Klotz (Leipzig,1916).
de clarisoratoribus ed. H. Malcovati (Leipzig,
liberqui Brutusinscribitur,
1965).
pro M. Coeliooratio,ed.A. Klotz (Leipzig, 1915).
dediuinationelibri,ed.0. Plasbcrg, W. Ax (Leipzig, 1938).
adfamiliarestpistularumlibri,ed.H. Sjogren (Leipzig,1923-25).
defato liber,ed.0. Plasbcrg, W. Ax (Leipzig,1938).
definibusbonorumet malorumlibri,ed. T. Schiche (Leipzig, 1915).
rhetoridlibri(deinuentione),ed.E. Strobel (Leipzig,1915).
de amidtialiberqui Loeliusinsaibitur,ed. K. Simbeck (Leipzig,1917).
denaturadeorumlibri,ed. 0. Plasbcrg (Leipzig, 1911).
deoffidislibri,ed. C. Atzcrt (Leipzig,1923).
deoptimogenereoratorumliber,ed. A. S. Wilkins (Oxford, 1903).
oratorisliber,ed. P. Reis (Leipzig,1932).
 BIBLIOGRAPHY
Cicero (amt.)
tk oralllrtlibri,ed. A. S. Wilkins (Oxford, 1902).
in L. CalpumiumPisontmoratio,ed. R. G. M. Nisbet (Oxford, 1961 ).
dt republiC4 libri,ed.K. Ziegler' (Leipzig, 1955).
pro Sex. R.osdoAmtrinooratio,ed. A. Klotz (Leipzig,1922).
pro P. Ststio oratio,ed. A. Klotz (Leipzig,1915).
topicorumlibtr,ed. A. S. Wilkins (Oxford, 1903).
Tusculanarum disputdtionum libri,ed.M. Pohlcnz (Leipzig, 1918).
in C. Verremactionissecundaelibri,ed. A. Klotz (Leipzig, 1923).
'scholia Gronouiana', ed. T. Stangl (in Ciceronisorationumscholiastae
[Vienna, 1912), pp. 279-351 ).
Diomcdcs
'artis grammaticac libri•, ed. H. Keil (in GrammaticiLatini, vol. 1
[Leipzig,1857), pp. 297-530).
Donatus
'artis grammaticac libcr •, ed. H. Keil (in GrammaticiLatini, vol. IV
[Leipzig,1864), pp. 367-402).
in comoedias Terentiicommmtarius, ed. P. Wessner (Leipzig, 1902-5).
Fcstus
dt uerborumsignifiC4tUlibri, ed. A. Thcwrcwk de Ponor, W. M.
Lindsay (Leipzig, 1913).
Gcllius
noctiumAltiC41Um libri,ed.M. Hertz, C. Hosius (Leipzig, 1903).
Hieronymus
epistul«, ed.L Hilbcrg (Vic~ 1910-18).
Horatius
tpodonlibtr,ed. F. Klingner (Leipzig, 1939).
Isidorus
uerborumlibri,ed.F. Arevalo (Rome, 1802).
dt di.fferentiis
Iulius Jlufioianus
'de figuris scntcntiarum et clocutionis libcr', ed. C. Halm (in Rhetores
Latini minores[Leipzig,1863), pp. 38-47).
Iulius Victor
'artis rhctoricae libcr •, ed. C. Halm (in RhetortsLatiniminores[Leipzig,
1863), pp. 371-448).
Lucretius
dererumnaturalibri,ed.H. Diels (Berlin, 1923).
Macrobius
conuiuiorum primidiei Saturnaliorum libri,ed. J. Willis (Leipzig, 1963).
Nonius
decompendiosa d«trina libri,ed. J. H. Onions, W. M. Lindsay(Leipzig,
1903). I quote according to Mercier's pagination.

430
 BIBLIOGRAPHY

Orosius
historiarumoJuersumpaganoslibri,ed. C. Zangemeistcr (Leipzig,1889).
Ouidius
artisamatoriatlibri,ed. E. J.Kenney (Oxford, 1961).
heroidumepistulae,ed. A. Palmer (Oxford, 1898).
Paulus
excerptaex librisPomptiFesti
Phaedrus
fabularumlibri,ed. J.P. Postgate (Oxford, 1920).
Plautus
comoediaqu« Bacchidesinscribitur,ed. F. Ritschl, G. Goetz (Leipzig,
1886).
comoediaqu« Poenulusinscribitur,ed. F. Ritschl, G. Goetz, G. Loewe
(Leipzig.1884).
Pocta incertus
Aetnae liber,ed. F. R. D. Goodyear(Cambridge, 1965).
Pompeius
•commentum artis Donati•, ed. H. Keil (in GrammaticiLatini, vol. v
(Leipzig,1868], pp. 95-312).
Priscianus
•institutionum grammaticarum libri •, ed. M. Hertz (in Grammatici
LAtini,voll. n-m i [Leipzig,1855-91).
'de metris fabularum Terentii liber', ed. H. Keil (in GrammaticiLatini
vol. m ii [Leipzig,1860], pp. 418-29).
[Probus]
•in libros Vergilii commentarius•, ed. H. Hagen (in Appendix Seruiana
(Leipzig.1902), pp. 323-87).
Q_uintilianus
institutionisoratoriaelibri,ed. L. Radermacher (Leipzig, 1907-35).
Rhetor incertus
de rationedicendiad C. Herenniumlibri,ed. F. Marx (Leipzig, 1923).
Rutilius Lupus
•de figuriselocutionis libri•, ed. C. Halm (in RhetoresLAtini minores
[Leipzig,1863], pp. 3-21 ).
Sallustius
de hello lugurthino liber,ed. A. W. Ahlberg, A. Kurfess11 (Leipzig,
1954).
Seneca
controuersiarum libri,ed. H.J. Mueller (Vienna, 1887).
Seneca
liberqui Diui Clmulii a,ro8t(A)(71s per saturaminscribitur(Apocolocyn-
tosis),ed. C. F. Russo• (Florence,1961).
431
 BIBLIOGRAPHY

Seruius
et scholiaquibuscommentarius
in libro) ,. (tgilii commenl4rius auctus est.
1-v Ameidos, ed. E. K. Rand et al. (Lancaster-Oxford, 1946-65);
VI-XII Aeneidos,ed. G. Thilo (Leipzig, 1884); I-IV georgicorum,ed.
G. Thilo (Leipzig,1887).
Tercntius
comoediaquae Adelphi inscribitur,ed. R Kauer, W. M. Lindsay,
0. Skutsch (Oxford, 1958).
comoediaquae Eunuchus insaibitur, ed. R. Kauer, W. M. Lindsay,
0. Skutsch (Oxford, 1958).
Tacitus
annaliumlibri, ed. C. Halm, G. Andresen, E. Koestermann (Leipzig,
1936).
Varro
de linguaLatinalibri,ed. G. Goetz, F. Schoell(Leipzig,1910).
Vergilius
Ameidoslibri,ed. R. Sabbadini (Rome, 1930).
eclogarum liber,ed. R. Sabbadini (Rome, 1930).
georgicorum libri,ed. R. Sabbadini (Rome, 1930).
'scholia Veronensia', ed. H. Hagen (in Appendix Seruiana[Leipzig,
1902 ], pp. 393-450 ).

432
 CONCORDANCE I
This tdition Ribbtcltl Vahltn1
i 13-1s 1-3
I 9 IS
2 16 16
3-4 2-3 10-11
s 7-8 4
6-7 4-S 13-14
8 I 6
9 6 s
10-12 10-12 7-9
13 336-7 19
14 18 20
IS 17 17
16-20 20-4 22-6
21 32 34
22-30 25-31 27-33
31 19 21
32-46 39-53 54-68
47-9 54-6 69-71
so-61 Inc. inc.s-16 35-46
62-3 34-S 47-8
64 38 S3
65 62 78
66 36 so
67 33 SI
68 37 49
69-71 51-9 72-s
72-3 6o-1 76-7
74-7 Acdus 357-60
78-9 91-2 100-1
80 89 8s
81-94 75-88 86-99
9S 340-1 376
96-7 95-6 112-13
98 70-1 107-8
99 65 IOS
100 93 82
IOI 94 Ill
102 68 103
103 90 102
104-s 63-4 80-1
28 433 JTO
 CONCORDANCE I
This edition Ribbedt' Vahlen1
106-7 72-3 83-4
108 69 106
1~10 66-'] 104
Ill 74 79
112 97 120
113 100 us
114 102-3 117
ns IOI 114
116 104, 116
117-18 IOS~ 118-19
119 98-9 122
120-4 107-11 123-7
125-31 [114-19]
132 120 129
133 113 133
134 121 134
135~ 124-5 130
137 II2 128
138--9 126-7 131-2
140 130 139
141-2 128--9 137-8
143 131 140
144 134 147
145 136 149
146-7 132-3 145~
1-48 135 148
1-49 1.49 178
150-1 141-2 177
152 ISS 193
153-4 139-40 158-9
ISS~ 100-1 188-9
157-8 156-7 19(r-2
159 151 185
160 148 179
161-2 158--9 186
163 1-43 156
164 144 157
165 ISO 181
166 1-45 180
167-8 1-46-7 182-3
169 152-4 184
170 137-8 16o
171 163 196
172-4 165-7 199-201
175 172 206

434
 CONCORDANCE I
This tdition Rjbbtck' Vahlm'
176 176 209
177-8 173-4 211-12
179 162 195
180 164 202
181 175 207
182 168-9 197-8
183-4 170-1 203-4
xav lphigtnuifr. X
185--7 199-201 24-2-4
188-91 177-80 215-18
192 202 245
193-4 181-2 213-14
195-202 183-90 234-41
203 193 224
204~ 194~ 225--7
207 203 233
2o8-16 205-13 246-54
217-18 231-2 276--7
219-20 220-1 26o-1
221 240 273
222-3 216--17 257-8
224 233 278
225--7 226--7 266--8
228 Inc. inc. 161 269
229-31 228-30 270-2
232-3 222-3 262-3
234~ 237-9 284~
237-8 214-15 255~
239-40 243-4 287-8
241-2 235~ 282-3
243 234 280
244 241 279
245 218 281
cxviii Mtlanippafr. vi 294
246 249-50 295
247 24,6-8 291
248-9 245~ 289-90
250 251 292
251 252 293
252 254 297
253 255 296
254--7 257~ 300-3
258 265 3o6
259 256 298
260 262-4 304-5
435 28-2
 CONCORDANCE I
This edition Ribbeck' Vahlen•
261 261 308
262 266 307
263 267 309
264 268 310
265 271 318
266-9 273--6 320-3
270-1 269-70 316-17
272 278 324
273-4 279-80 325--6
275 282 328
276 283 JII
277 281 327
278 277 315
279 284 329
280 286 331
281 287 330
282 285 339
283-4 289-90 334-5
285--6 293-4 337-8
287 288 333
288 291 332
289 292 336
290 3o6 352
291-2 Inc. inc.108-9 357-8
293-5 303-5 349-51
296-9 309-12 362-5
300 298 348
JOI 302 345
302 307 353
303 295 361
304 JOI 344
305 296 341
3o6 299-300 346-7
307 308 36o
308 297 340
clxi 372 380
309-10 122-3 342-3
3II 400 422
312 313 173
313-15 366-8 398-400
316-18 351-3 142-4
clxvii 398-9 412-13
319 374 381
320 381-2 404-5
321 363 299
436
 CONCORDANCE I
This edition rubbeck' VohJml
322-33 314-2s 161-72
334 332 176
33S 3S4 20s
336-7 36o-1 392-3
338-40 369-71 354-6
341 378 39S
342 37S 401
343 364 394
344-5 356-7 219-21
346 429
347 338 18
348 319 402
349 389 409
350 380 403
351 388 210
352-3 347-8 388-9
354 396 411
355 Vari448
356-6o Vari454-8
361 335 187
362 3S5 420
363 362 121
364 349 ISO
365-6 Antuues9
367 385 396
368 392 232
369 365 397
370 333-4 417-18
371 402 419
372 393 421
373 390 408
374 387 407
375 423
376 330 175
377 386 406
378 428
319 342 430
380 328-9 416
381 339 52
382 407 378
383 408 319
384 404 4IS
385 373 383
386 345-6 386-7
387 331 382
437
 CONCORDANCE I
This edition RibMclt' Vahlm'
381H) 197-8 221H)
390 394 42S
391 Com. s 424
39.2-3 3SIHJ 390-1
394 343-4 384-s
39S 39S ,426
396 391 410
397-8 383-4 l3S--6
399 406 174
400 397 427
CCXXV Htcubafr. xi 194
,401 403 414
402 40s Inc. 3

CONCORDANCE II
Ribbtckl This tdilion Ribbeclt8 This edition
I 8 63-4 10,4-s
2-3 3-4 65 99
4-S 6-7 66-7 109-10
6 9 68 102
7-8 s 69 108
9 I 70-1 98
10-12 10-12 72-3 106-7
13-IS i 74 Ill
16 2 7S-88 81~
17 IS 89 8o
18 14 90 103
19 31 91-2 71H)
20-4 16-20 93 100
2s-31 22-30 94 IOI
32 21 9S--6 96-7
33 67 97 112
34-S 62-3 98--9 119
36 66 100 113
37 68 IOI ns
38 6,4 102-3 114
39-S3 32-46 104 116
S4--6 47--9 IOS--6 117-18
S7-9 69-71 107-11 120-4
6o-1 72-3 112 137
62 6S 113 133

438
 CONCORDANCE II
Ribbedt8 This edition Ribbtck' This edition
[n4-17] 125-8 193 203
[118-19] liii 194~ 204~
120 132 197-8 388-9
121 134 199-201 185-7
122-3 3~10 202 192
12-4-S I3S~ Iphigeniafr. x xciv
126-7 138-9 203 207
128-9 141-2 205-13 208-16
130 140 214-15 237-8
131 143 216-17 222-3
132-3 146-7 218 245
134 144 219 cva
135 148 220-I 219-20
136 145 222-3 232-3
137-8 170 22-4-s seep. 349
139-40 153-4 226-7 225-7
141-2 ISO-I 228-30 229-31
143 163 231-2 217-18
144 164 233 22,4
145 166 234 2-43
146-7 167-8 235~ 2-41-2
148 160 237-9 234~
149 149 240 221
ISO 165 241-2 244
ISI IS9 243-4 239-40
152-4 169 2-4S 2-48-9
ISS 152 2-46-8 2-47
156-7 157-8 249-50 2-46
158-9 161-2 251 250
I6o-I ISS~ 252 251
162 179 Melanippafr. vi CXVll1
163 171 254 252
164 180 2ss 253
165-7 172-4 256 259
168-9 182 257~ 254-7
170-1 183-4 261 261
172 175 262-4 26o
173-4 177-8 265 258
175 181 266 262
176 176 267 263
Hecubafr. xi CCXXV 268 264
177-80 188-91 269-70 270-1
181-2 193-4 271 ~s
183-90 195-202 272 cxxxivb
191-2 seep. J2l 273~ 266-9
439
 CONCORDANCE II
lobbtdt' This edition lobbtdt' This edition
277 278 3-49 364
278 272 350 clxvi
279-80 273--4 351-3 316-18
281 277 354 335
282 275 355 362
283 276 356-7 3#-5
28-4 279 358-9 392-3
285 282 360-1 336-7
286 280 362 363
287 281 363 321
288 287 364 3-43
289-90 283--4 365 369
291 288 366-8 313-15
292 289 369-71 338--40
293-4 285-6 372 clxia
295 303 373 385
296 30s 37-4 319
297 308 37S 342
298 300 378 341
299-300 306 379 348
301 304 380 3S0
302 301 381-2 320
303-5 293-5 383-4 397-8
306 290 38s 367
307 302 386 377
308 307 387 37-4
309-12 296-9 388 3SI
313 312 389 3-49
31-4-25 322-33 390 373
326-7 Stt p. 164 391 396
328-9 380 392 368
330 376 393 372
331 387 394 390
332 33-4 395 395
333-4 370 396 3S4
33S 361 397 ,400
336-7 13 398-9 c1xvii
338 347 -400 311
339 381 ,402 371
340 95 -403 384
342 319 -40s 4()2
343-4 39-4 -406 399
34S-6 386 407 382
3-47-8 3S2-3 408 383

440
 CONCORDANCE III
Vahlml This edition Vahltn1 This edition
1-3 i 107-8 98
4 5 109-10 set pp. z.5.5-6
5 9 III IOI
6 8 n.2-13 96-'1
7-9 10-12 Il-4 n5
10-II 3--4 ns Il3
12 step. 164 n6 n6
13-1-4 6-7 II7 Il-4
IS I n8-19 n7-18
16 2 120 II2
17 IS 121 363
18 3-47 122 Il9
19 13 123-7 120-4
20 14 128 137
21 31 129 132
2.2-6 16-20 130 135-6
27-33 22-30 131-2 138-9
3-4 21 133 133
35-46 5CHS1 13-4 13-4
-47-8 62-3 135-6 397-8
-49 68 137-8 1-41-2
so 66 139 1,40
SI 67 1-40 143
52 381 1-41 clxvi
53 64 1-42--4 316-18
54-68 32-46 1-45-6 146-7
69-71 -47-9 147 I#
1.2-s 69-71 148 1-48
76-7 72-3 1-49 1-45
78 65 150 364
79 III 151-5 seep. 28.5
80-1 10-4-5 156-7 163--4
82 100 158-9 153-4
83--4 Iof,...,/ 160 170
85-99 80-9-4 161-72 322-33
100-1 78-9 173 312
102 103 17-4 399
103 102 175 376
104 109-10 176 334
105 99 177 ISO-I
Io6 108 178 1-49

441
 CONCORDANCE III
Vahlen2 This edition Vahlen• This edition
179 16o ~2 225-31
180 166 273 221
181 165 274-5 SU p. JjO
182-3 167-8 276-'7 217-18
184 169 278 22,4
185 159 279 2#
186 161-2 280 2-43
187 361 281 2-45
188-9 155-6 282-3 241-2
190-2 157-8 284-6 234-6 :i
193 152 287-8 23S>--40
194 CCXXV 289-90 2-48-9
195 179 291 247
196 171 292 250
197-8 182 293 251
l9S>-201 172-4 294 cxviii
202 180 295 246
203-4 183-4 296 2S3
205 33S 297 252
206 175 298 259
207 181 299 321
208 seep. 305 3oer-3 254-7
209 176 304-5 26o
210 351 306 258
211-12 177-8 307 262
213-14 193-4 3o8 261
215-18 188-91 309 263
2IS}-2I 344-5 310 264
222-3 SU p. 321 311 276
224 203 312-14 seep. 394
225-7 204-6 315 278
228-9 388-9 316-17 270-1
230-1 seepp. 321-2 318 265
232 368 319 occxivb
233 207 320-3 266-9
234-41 195-202 324 272
2,42-4 185--7 325-6 273-4
245 192 327 277
246-54 208-16 328 275
255-6 237-8 329 279
257-8 222-3 330 281
259 cva 331 28o
26o-l 21S}-20 332 288
262-3 232-3 333 287
264-5 seep. 349 334-5 283-4

442
 CONCORDANCE III
Vohlml This edition Vohltn1 This edition
336 289 39♦ 3♦3
337-8 2ss~ 39S 3♦1

339 282 396 367

3♦0 308 397 369
3♦1 30S 398-400 313-1s
3♦2-3 309-10 ..,01 3♦2
3# 30.., ♦02 3♦8
3♦S 301 ♦03 3SO
346--7 306 ♦O♦-S 320
3♦8 300 ♦o6 377
3♦9-SI 293-S ♦07 37♦
3S2 290 ♦OS 373
3S3 302 ♦09 349
3S♦~ 338-40 410 396
3S7-8 291-2 411 3S4
3S9 cxlixb 412-13 clxvii
360 307 ♦14 ♦01

361 303 ♦IS 384

362-s 296-9 ♦16 380
l366-7sl 417-18 370
376 9S 419 371
377 xxviiie ♦20 362
378 382 ♦21 372
379 383 422 311
38o clxi41 ♦23 37S
381 319 ♦24 391
382 387 ♦2S 390
383 3ss ,426 39S
38♦-S 39♦ ♦27 ♦00

386-7 386 ♦28 378

388-9 3S2-3 ♦29 3¥>
390-1 392-3 430 379
392-3 336--7

443
 INDEX I

ENNIUS' TRAGIC VOCABULARY

I refer with Arabic numerals to verses actually quoted as Ennian, with Roman
numerals to passages from which Ennian words have been or could beextracted.
My lists attempt to be inclusive rather than critical. Square brackets mark cor-
ruption, italic numerals conjecture, asterisks conjecture which I reject but
considerto be specially interesting. I use the following abbreviations:
a. accusative n. neuter
ah. ablative nom. nominative
adj. adjective p. plural
adv. adverb part. participle
conj. conjunction perf. perfect
d. dative pres. present
f. feminine prp. preposition
g. genitive s. singular
imp. imperative sub. substantive
ind. indicative subj. subjunctive
indef. indefinite vb. verb
intcrr. interrogative voc. vocative
m. masculine
A, AB a, 24, 30, 104, 114, 156, cva, ACCIPIOaccipite, 242, [391]; accep-
*z63,291,clxvii;ab,62,[104),145, tum (m.s.a.), 362; acceptam, *J62
268, 3o6; abs, 125,207 ACER acrem, 333
ABBO abire, [290] ACHERVNSAcherontem, 192
ABIBGNVSabicgna (f.s.nom.), 209; ACHBRVSIVS Achcrusia (n.p.voc.), 98
abiegnae (f.p.nom.), (209) ACHIU.BSAchilli, 16o
ABIBSabiete, 86 ACHIVVS Achiuos (p.a.), 172; Achi-
ABIGO abige, 24 uis (d.), 151; (abl.), 15, 7S
ABNVEOabnuebunt, 279; abnuebant, ACIBS aciem, 333
[279) AD 114, 136, 209, 218, 240, 293, 322,
ABNVTOabnutas, 290 382, 390
ABOJllOR aborta (f.s.nom.), (167] ADDECBTaddecet, *254
ABSVMabsum, cva; abest, 265 ADBO (adv.), 234
AC 148 see ATQVB ADBO adibo, (192]; adire, 293; adiri,
ACCBDOaccedi, 327 290; adirier, * 290
ACCBNDOaccendat, 314; accenderit, ADFICIOadficio, 129; adficior, 12S
3IS ADIACBOadiacent, (192]
ACCIDO ~rccdarn (subj.), 83; accidat, ADICIO adiecit, (254]
340; accidisset, 209 (accedisset) ; ADITO aditauere, 390
accidissent, [209 accedissent] ADIVNGOadiungito, I 33
445
 ENNius' TRAGIC VOCABULARY
ADIVVO adiuuerit. 74 ALTBRalter, [18]; altcri (m.s.d.), 267,
ADSTO asta, 239; astitit, 2; adstante, 328
89; adstantem, [89] ALTBRNVSaltcrna (f.s.ah.), 123
ADSVM adcst, 41, 41; adsunt, 23; ALTISONVSaltisono (m.s.ah.), 88,
adcs (ind.), *70 188
ADVENIO aduenit. [4s]; aduenirm, ALTVS altum (n.s.a.), 264; alt:a.m,
322;aduenict,4s,49 cva; alto (n.s.ah.), III, 264; alti
ADVBNTOaducntant. 68 (m.p.nom.), *85; alta (n.p.voc.), 98
ADVBRSAJUVS aduersarios, 2ss AMBO amho (m.nom.), 170
ADVBRSVM 2SS, 286 (aduonum) AMICVS amicus, 3SI; amici (p.voc.),
ADVBRTOaduortite, i; aduortere, 369 263
ADVOCO aduocant, 136 AMOR amoris, 224; amore, 216
ABAcvs Aeaci, 273 AN [102], 137, 374, [391]; see ANNE
ABDIS aedis (p.a.), 238 ANDROMACHA Andromachae (d.),
ABGBRacger, 336; aegra (f.s.nom.), 99
[216]; acgro (m.s.ah.), 216 ANGVISangui (ah.), *26; angue, 384
ABGRBacgerrumc, 78 ANIMA anima (ah.), [17], 182
ABQVAUS aequalis {p.a.), 37 ANIMO animatum, [254]
AEQVJ! 174 ANIMVS animus, 62, 199, 202, 336;
Al!QVVS aequa (f.s.nom.), 174; animum (a.), i, 198, 369; animo
aequum (n.s.nom.), 170; aecum (ah.), 74, 216, 3o6
(n.s.a.), 148; (n.s.nom.) 156 ANNE 218
ABR acr, 358; aerem, 3S7 ANNVS annos, 42, 104
ABRVMNA acrumna (ah.), 141; aerum- ANTE (adv.), 33; (prp.), 187
nis (ah.), 103 ANTBSTOantestat, 388 (antistat)
ABS aes, 16.5;acre, 2 ANTIQVVS antiqua (f.s.voc.), 237;
AllsCVLAPIVsAcsculapi, 326 (Aescu- anticum (n.s.a.), 239; antiquo
lapii) (m.s.ah.), 278
AETAS aetatem, 220 APouo Apollo, 28, 36, S9, 303 ;
ABVVM acui, 401 Apollincm, 56
AGER agcr, 394; agros, 137 APPBLLO appcllat, 66; appellarc,
AGO agit, 198; agat, 197, 271, 308; [218]
agcns, 190; agerent, 220 APPIJCO appliccm, 83
Awe Aiax, [14]; Aiacem, 272 APTVS apta (f.s.voc.), 350
ALA alas, 34S ARA aram, 94; arae (p.nom.), 84
ALACRIS alacris (m.s.nom.), 124 ARBITROR arhitror, 349
AllxANDBR Alexandrum, 64 ARBOR arborcs (p.nom.), 159 (ar-
AllQVI aliquod (nom.), 186; (a.), boris)
:xic ARCVS arcum, 29
ALJQVISaliquis (m.s.nom. ), -48; ali- ARDBOardentem, so; ardente, clxvii;
quid, [391] ardentihus (ab.), 27, 32
AllQVOT 346 ARDVVSardua (n.p.a.), 73, [379]
ALJVS alii (m.p.nom.), 68, [8s]; ARBOPAGITBS Areopagitae (p.nom.),
aliae (f.p.nom.), [2s7]; alia (n.p.a.), 364
II7 ARcIVVS Argiui (m.p.nom.), 212;
ALMVSa1ma(f.s.voc.), 3SO Argiuum (m.p.g.), 288, 330
 BNNIUS' TRAGIC
Allco Argo, 212
ARGVORarguor, 204
AllGVTORargutarier, 26o
~ arietis (s.g.), 213
ARMA arma (a.), 143, 16o; armis
(ab.), 232
ARMO armatos, 1s3; armatis (ah.),
72
AlllllGO arrigunt, [143]
ARX arccm, cv a; arcc, 83
ASPBCTVSaspectu, 2 l
ASPBll aspera (n.p.a.), 379; asperis
(n.p.d.), 296
ASPICIO aspicc, 240, 301
ASTROLOGVS astrologorum, 185
AT 1so; Stt ATQVB
ATBR atro (m.ah.), 297
ATHENABAthenas, 239
ATQVB,i, 6, 16, 20, 41, 62, 86, II3,
133, 137, 147, 148, 149, 1s6, 173,
[174], 191, 198, 223, 230, 239,
26o, 273, 274, 278, 309, 3S8, 36o,
382
AVCVPO aucupant, 245
AVDil!NTIAaudicntiam, i
AVDIO audis, [133]; audiunt, 284;
audire, i; audi, 133; audibo, 278;
auditis (part.n.p.d.), 133
AVBO auent, 62
AVFBRO ahstulit, 167
AVGIFICOaugificat, 102
AVGVROauguro, 249
AVIDB 63
AVJUS aures (p.nom.), 62,245; auris
(p.a.), 278, 305
AVRO auratum (m.s.a. ), 29
AVRVM auro (ah.), 91
AVSCVLTOausculta, 247
AVT *33, 71, [77], 81, 82, 109, 109,
177, 186, 186,266,266,266, [309],
373
AVTBM 167
AVVS aui (s.g.), *z73
AVXD.JVM awcilium (a.), 24, 322;
awcili, [151 auxilii]; auxilio (d.),
151; (ah.), 82
447
VOCABULARY
BAccmcvs Bacchico (m.ah.), 124
BACCIIVS Bacchus, 120
BALO halantihus (f. ab.), 54
BAllBAJUCVS harharica (f.s.ab.), 89
BBLLVMbellum (a.), 105; hello (ah.),
284
BBL VA beluas, 360; beluarum, 186
Bl!NII 12, 219, 265; see MBUVS
BBNBPACTVM bcnefacta (a.), 349
BBNEPICIVMbcneficia (a.), [349]
BIGAB higis (ah.), 97
BLANDil.OQVBNTIAhlandiloqucntia
(ab.), 226
BLANDVSblandi (m.p.nom.), 108
BONVS bonum (n.s.nom.), i; honi
(n.s.g.), 335; bona (£s.ah.), II;
(n.p.a.),clxvii; bonis (m.p.d.),265;
(n.p.d.), 294; set MBUOR, OPTIMVS
BaoMIVs Bromius, 120
BRVGBS(nom.), 334 (Phryges)
BRVGIVS Brugio (n.ah~), *31.2
CADO cadunt, 175, 385; cadens,
[301]; cecidit, 370; cecidissent,
[209]
CABDBScaede, [23], [23]
CABDO caesa (f.s.nom.), 209; caesae
(f.p.nom.), [209]
CABLEScaelitum, 171, 270
CABLO caelatis (part.n.p.ah.), 90
CABLVMcaelum (a.), 235; caeli, 96,
[171], 187, 189, clxia, 319, 366,
387; caelo (d.), 223; (ah.), 13, 67,
18s, 243, [366]
CABMBNTAcaementae (p.nom.), 385
CAERVLBVScaeruleo (m.ab.), *z6;
(n.ab.), 365; caeruleae (f.p.nom.),
26
CABllVLVScaerula (n.p.nom.), 2SO
CAIJGO caligo, 167
CANDBOcandcnt, 250; candens, 301;
candcntem, 243
CANDIDVScandidum (n.s.nom.), 274
CANIS canis (s.g.), 384
CANTVS cantus, [345]; cantu (ah.),
345
 ENNIUS' TRAGIC VOCABULARY
CAPIO cepit, 222; ccpissct,211 COLLOCO collocaui, 130
CAPRA Capra (nom.), 186 COMBS comiti, [313]
CAPVT caput. [114]; capiti, 77 COMITBll 313, 369
CAJU>O cardine, 88 COMMINVS 149
CASTRA castra (a.), 154; castris (d.), COMMISCBO commixta (n.p.voc.),
154; (ah.), 164 171
CAVBO cauco, [281] COMMISBIIESCO conmisercscite, 162
CAVSA cawa (nom.), [365]; (ah.), COMMVNITBll [313]
112, 343 COMPLEO complebit, *46; compleuit,
CAVVS caua (f.s.nom.), 365; (n.p. 46
nom.), 250; (n.p.a.), 96 COMPOS compotes (m.p.a.), 317
cmo ccdo, [281] CONATVS conatu,*uo
CJIRIIS Ceres (nom.), clxxxviii; CONCBDO concedit, 5
Ccreris, 240 CONCIDO concide, 372
CBRNO cemo, 13 ; cemunt, 166; CONClllO conciliant, 310
ccmitur, 351; ccrnat, 306; cemcre, CONCIPIO conccpta, [252]
232 CONCOllDIA concor~ 310
CBRTAUO ccrtatio, 225, *248 CONCVBIVM concuhio (ah.), 181
CBRTB clxxib coNcvno concutit, clxia
CBRTVS certus, 351; certo (ah.), 74, CONDICIO condiclones (nom.), 304
[248] ; certos, 3 I 7 CONFBRO conferunt, 310; conferre,
Cl!SSO ccssas,194; cessat, 376 *154; contulit, 168
CIITJIRVS cetera (n.p.a.), 269 CONFICIO conficis, 97
Cll1TB 242 CONFIDBN'l1A confidentia (ah.), 19
CIBAJUVM clharia (a.), clxxxviii CONGLOMBRO conglomeras, 302
Cll!O clet, 36; clta (f.s.nom.), 43 CONIBCTURA coniecturam, 55; con-
CIRCVM 114 iectura (ah.), 249
CIRCVMSTO clrcumstant, 27 CONOR conare (ind.), 7; conatur,
CIRCVMVENIO circumuentus, 16 [no]; conatum (n.s.a.), 289
Cisssvs Cwei, ccxxv CONSABPIO consacpta (f.s.nom.), 252
CITVS Stt Cll!O (consepta)
ems clues (voc.), 42; (a.), 8; cluis CONSBNnO consentit, 21
(p.a.), 341 (clues); cluium, 275 CONSILIVM consilium (nom.), 151;
CIVrrAS cluitatem, 288 (a.), clxvi, 382; consili, 317;
CLAMOR clamoris, 163 consilio (ah.), 289; consiliis (ah.),
CLAJlBO claret, 274 *s
Cl.ASSIS classis (s.nom. ), 43 CONSISTO consistere, ·324; constitit,
CIJPBVS clipeus, 370; clipeo (ah.), 159
107, 189 CONTAGIO contagio, 294
COBPI coepisset, [211] coNTBMPLO contempla, 240
COBTVS coetus (s.nom.), 123 CONTBMPLOR contemplatur, I Is
COGITO cogitat, 308; cogitet, [308] CONllNBO contincs, 23 s
COGNOSCO cognoui, liii; cognitum CONTINGO contigcrit,307
(n.s.a. ), liii CONTRA278
COGO cogis, 128 CONVBNIO conueniat, 311
CoLCHVS Colchis (ah.), 214, *z,u CONVBRTO conuertcre, [ I 381
 ENNius' TRAGIC VOCABULARY
CONVJ!STIO conucstire, 138; con- DBFBNDO defendant, 398; defende, 8;
ucstitus, *z81. defender<; 8
coR cor {nom.), 21; cordis, 244; DBPIO dcfit, 338
cordc, 140, 244 DBPI.AGRO cldlagrata (n.p.nom.), 85
ComITHVS Corinthum, cva DBPORMO dcformati (m.p.nom.), 86
CORONA coronam, 381; corona DBGVSTO degustandum (n.s.nom.),
{ab.), 67 xxviiit
COBPORO corporarct, I IS DBI.Bero delcctat, 198, 303
coRPVS corpus {nom.), 299; (a.), DBUGO delecti (m.p.nom.), 212
us; corporis. 237, 298; corpore, DBUNQVO delinquis, *zo4; delin-
70, 295, 365; corpora (voc.), 241; quas, [204,]
{a.), 136, 138, 341 DBI.PHicvs Delphicus, 303 (delficus)
CORTINA cortina {nom.), 366 DBMBNSdementem, 36
· cRASsvs crassa (f.s.nom.), 387 DBNVO 3SS, 358
CREDO credo, IS9, 289; credcre, DBPlll!COR deprecor, 142; deprecer,
[26<>];crcde, *z6o 134,
CRBMITO cremitari, *z47 DBRBPBNTB 32, 168, 264
CRBMO crcmari, 247 DBSBRO deseri, 392
CRESPHONTES
Crcsphontcm, 1.26 DBSINO desinit, 337
crinitus, 28
ClllNITVS DBSPICIOdcspexit, 264
CRISPVScrispa (f.s.ab.), 86 DBVBHOdeuchi, 392
CllVCIATVS cruciatum, 18; cruciatu, DBVITO deuitari, 325
178 DBVS di (nom.), 287 (dii), 309 (dii);
CllVCIOcruciatur, [178] deos, 401; dcum (p.g.), 3, 270,
CllVl!NTVS crucnta {n.p.a.), 138 275 (~), 289 (deum)
CVBrrvs cubitis {ab.), JS.if. DIANA Diana (nom.), 30
CVM {prp.), IS, 21, 27, 67, 70, 7S, DICO dico, 145,249,356,359; dicam
103,136,308; {conj.), 8, 134,, [156], (subj.), 261, 300, 373; (ind.), 270;
173, 186, 196, 201, 251, [281], 315 dices, 172; dixi, 270; dicta (n.p.
CVNCTO cunctct, [ 1601; cunctent, 160 nom.), 174; (n.p.a.), clxvii; dictis
CVPIDO cupido, 222 (n.p.d.), 258; (n.p.ab.), 331
CVPIDVS cupida, [259]; cupido {ab.), DIBS diem, 103, 395
244 DIPfllllO differt, 117
CVPIIINTBR 259 DWGO dilecti, [212]
CVPIO cupit, 259, [348]; cupiant, DIMITTO dimitto, 318
160; cupiens, 259; cupere, 337 DISICIO disicctae {£p.nom.), 84
CVR 127,128 DISSVPO dissupat, I 17
CVRA curls (ab.), S3 DIV JI
cvao curent, 265; curare, 271 DIVINVSdiuina (£s.ab.), 58
CVRRVScurru, 79 (curro) DIVITIABdiuitias, 268 ; diuitiis, 269
CVSTOS custos, 237 DIVORTIVM diuortio (ab.), 131
DO dat, 228, clxxxviii; dant, 382;
DB 67, 100, 166, [200],269,289, 314, date, 182; dare, 160, 253; dabo,
340 230, 278; dabit, 228; dcdit, 258,
DBA ~ (nom.), 287; deas,48 *333; dcdere, 364; dcderit, 363;
DBDVCOdeducant, 269;deduci, [3921 dati (m.p.nom.), 283

29 449 JTO
 BNNius' TRAGIC VOCABULARY
DOCBO doceret, [56] EICIO ciciehantur, I
DOI.BO dolet, 39 EUCIO eliciatur, xciv
DOLVS dolum, 214 BLIMINO eliminas, *238; ~liminat, 238
DOMO doma, 158 ELOQVOR eloquerc (imp.), 330, 330;
DOMVS domus (voc.}, 87; domum, eloquar (ind.), 148
218; domi, 84, .200, 220; domo, 104, EM 200
215 EMINVS *149
DONO dono, [391] ENIM [12], [310]
DRACHMA dracbnmam, i68, 269 BNODO enodari, 284
DVBITO duhitari, [77]; duhitarit, 77 BO imus, .201; cunt, 136; cam, 136
DVBIVS duhia (f.s.ah.), 76 (? seexs); ihit, 225; ire, .201
DVCO ducit, 132; ducite, xciv; ducct, EQVVS equus, 72
[157]; duxi, 1iii BRA era (nom.), 215
DVCTO ductat, 303 BBILIS eri1is(n.s.g.), 237
DVDVM 62 mmo errat, 202, 336; errans, 21 s;
DVllO durat, 110 erranti, 313; erraui, 1iii
DVJUTBJl 2 58 ET i. 4, II, 18, 19, 42, [45], [54], 6-J,
83, 84, 1iii, [141), 146, 158, [161],
13, BX c, *149, 339; CX, [17], 58, 80, 162, [165], [173], 181, 207, 219,
xxviiie, III, 2$8, .264, 316, 321, 240, 249, 250, 256, 270, 273, 276,
358; seeBXTBMPLO [281], 285, 297, 304, clxvi, 322,
BBVll chore, 91 347, 350, 3SS, 357, 362, [372], 377,
ECCB 167 386
EDICO edico, *145; edicta (n.p.a.), i ETIAM 191, 191
EDO edidit, 58 ETSI 172
EDOCBO cdocerct, 56 EVBNIO euenat, 183 (eueniat)
EDVCO cducit, 153 BVHAN [12.2]
EFPBRO efferret, 215 (ecferrct); ex- BVBIVS euhium, [ 122]
tulisses,244 EVHOE see.fr.Iii; euhoe, *122, *122
EFPLO cfflantes (m.nom.), 14 BVISCBRO euisceratus, 296
EFFOR ecfari, 331 BVrrO euitari, 93
EGEO egens, 80 Eurypylus, cbcrib
EVRYPYLVS
EGESTAS egestas, .266 BXALBESCO exalbescat, .20
EGO ego, 3, 89, 134, *145, 148, ISO, BXAMl!N examen, 44
204, 204, 226, 230, 249, 270, 316; BXAMINO examioata, [238]
me (a.), 23, 25, 36, 38, 39, 40 (men), EXANCI.O exanclando (m.s.ah.), 144-;
40, 78, 127, 132, 180, 182, xciv, exanclaui, 103
.203, 217, 222, 224, 275, 277, 293, BXANIMO exa.nimata, *238; exani-
302, 303, 362; mei, 38; mihi, [17], matam, 238; exanimato (m.s.d.}, 17
21, 139, cva, 231, 231, [247), 273, BXCRVCIOcxcruciat, 25
274, [286], 307 (mihi}, 338, [339]; BXERCBO exerce, .26o; exercitum
mi, [17), 24,229,247,339; me (ah.), (m.s.a.), cbcrib
24, *263, 342,374; nos (nom.),200; HXILIVM exilium (a.), 231; exili, 82
(a.), 170; nohis (ah.}, 71 (exilii); exilio (ah.), 16
l!HIIV 47, [180], 302 BXISTIMOexistiroas,126; existimahas,
EBO 309 [126]
450
 ENNius' TRAGIC VOCABULARY
EXITIVM exitium (a.), 61, 231, 328; FERVSfera (f.s.nom. ), 45; feram, [ I 16]
(p.g.), 44 FESTINO festinant, 395
BXITVS exitum, [231], [328) FESTIVVS festiuum, 395
BXORDIVM exordium (a.), 210 FIDBS £ides, 320; Fides (voc.), 350;
BXOIUOR exoritur, 186 fidem, xvb, 162
BXPBCTORO expectorat, 17 FIDVS fide, 194; fida (£voc.), 237
BXPEDIO expedibo, 148 FIGO fixus, 296
BXPELLO expulsus, cla FILlA £ilia (nom.), 2o6, ccxxv; filias,
BXPBTO expetit, 348; expctunt, 23, 218; filiis (d.), II9
clxvi FIUVS £ilium, 362
BXPOSCO exposco, 151 FINCO fictas, 343
BXSACJURCO exsacrificabat, 54 FIO fit, (102], 236, 358
BXSBCROR exsecrabor, 342 FIJlMVS firmum (n.s.a.), 256; finno
BXSBQVOR exequar, 81 (n.s.ab.), 19
BXSPECTO exspectantes (m.nom.), 63 FLACCBO Baccebunt, 304
BXSVRGO exsurge, i FLAGnlVM flagiti, 341 (Bagitii)
BXTBMPLO 362 FLAM.MA Bamma (nom.), 22; Bam-
BXTOLLO extolles, 10; extollas, 321 mam, 169,clxvii; Bamma (ab.), 85
EXTRA (prp.), 39, 238 FI..AMMifEJl Bammiferam, 2 s
FLBCTO fiexeris, 172
FABVLOJl fabulari, 147 FLVCTVS fluctus (s.nom.), n6, 117;
FACBSSO faccssit, [145]; facessite, 145 (p.nom.), 1
FACILE 172, 3o6; facilius, clxvii FOBDO foedati (m.p.nom.), 399
FACINVS facinus (a.), 236 FONS fontem, 136
FACIO facit, 314; facimus, [236], FOR fa.ndis (n.p.ab.), 36; fatis
*
[372] ; fac, i; facito, 37z; facere, (n.p.ab.), 36
275; faciet, [273]; fecisse, 148; FOJlAS 153
fax:im, 261; factum (n.s.nom.), i, FORMA forma (ab.), cxviii
31, 263; facto (n.ab.), 51; factas, FORNIX fornices, 319
[343]; factum (n.p.g. ), 37; factis FORTIS fortem, (136]
(n.p.d.), 331; see BENBFACTVM, FOJl'llTBll 255, 263
MALBFACTVM FOllTVNA fortuna (nom.), 333, 338,
FALSVSfalsam,262 340; (voc.), 302; (ab.), 166
FAMA famam, to, 12 FRANCO franguntur, 165; fractae
FAMILIA familiae {g.), II2 (f.p.nom.), 84
FANVM fanum (a.), 352; fana (nom.), FllEMlTVS fremitu, 394
85 FllBTVS fretum (a.), 387
FA VBO fauent, 344 FRBTVS freta (£s.nom. ), 82
FAVX faucibw (ab.), 344 FIUGVS frigw (nom.), 358
FAX fax, 41 ; facem, 30, SO, 243 FRVCTVS fructus, [n6]; (p.a.), 245
FBMINA feminae (p.nom.), cxviii FRVX fruges (a.), clxxxviii
fBllB cxviii FVGA fugae (g.), 8z; fuga, (ab.), [82]
FBRO feratis, 263; fer, 24; ferte, 42; FVGIO fugio, liii; fugiat, [20]
ferre, 154, 262 FVltIA Furiarum, 49
FBJlJlVM ferrum (nom.), 109; (a.), FVT11IJ! 26 3
182; ferro (ab.), 166, 312, 399 FVTIUVS futtilum (n.s.nom.), 262

451
 ,
ENNIUS TRAGIC VOCABULARY
GBMITVS gemitum, IIO *1.7.5;hoe {n.s.a.), 236, 249, 285,
GBMO gemam, 184 289, 301, 342; huic {m.), 127, hoe
GENA genam, -400; genae (d.), -400 (m.s.ah.), 145, 388; haec {n.p.a.),
GBNS gentis (p.a.), 3SS 92, 172, 284, [359]; his (m.p.d.),
GBNVS genus (nom.), 271, 338; (a.), 1.20, 120; (m.p.ah.), (268), 269
270, 321 HIC (adv.), 163, [248], 307
Gl!llMANVS germane,69 HlNC [175], 201
GBllO gerit, I 81, clxxxviii; gcrentcs, HlPPODAMEA Hippodameam,292
IOS; gesserunt, IS; gessere. 219; HODIB 229, 307
gesta (f.s.ah.), 176 HOMO homo, 313, 375; hominem,
GES'ITl'O gestitat, 256 66, clxxib
GLORIA gloriam, II, 12 HONESTB 389
GNATA nata (voc.), 129 HONOR honoris, 224
GNATVS nate (voc.), 247; natum HORRESCO horrcscunt, 143
(a.), 100 HOSPES hospitem, 177, hospites
GRADVS gradum, 6, 193 (voc.), 293
GRABCVSGraeci (p.nom.), 356 HOSTIA hostiis (ah.), S4
GRATIA gratia (nom.), (27-4]; (ah.), HOSTDJS hostili (f.s.ah.), 323
132,224 HOSTIMENTVM hostimentum (a.). 133
GRATVLOll gratulor, 176 uosno seefr. lxvii; hostihis, *1.49;
GRAVIDVS grauidus, 72; grauida hostiuit, *1.49
(f.s.nom.), so HOSTIS hostes (a.), 8; hostium, xciv
GRAVITBll 184 HVC 201
GJlBMIVM gremium (a.), 321 HVMANVS humanum (n.s.nom.),
GVTrATIM I7S 271; humana (f.s.ah.), 299
GYPSO gypsatissimis (f.p.ah.), cva HVMVS hum.um, 354

HABEO habet, 196, 381; habetis, cva; IACEO iacent, 8-4,399

habeat, 298
HABllEO haeret, 29S
HALITO halitantcs, 169
HARIOLAno hariolationihus (ah.), 3S
HAlUOLVS harioli (p.nom.), cxxxivb
BASTA hasta (voc. ?), 149; hastae
(p.nom.), 16s
HAVD haut, 77; haud, 9S, 376
Hl!cro11 Hector {nom.), 153, 376;
(voc.), 69, 80; Hectorem, 79;
Hectoris, 100, 333
HEcVBA Hecuha (nom.), 51, ccxxv
HBHAE 370
HELENA Helena (nom.), 205
HBV · 180, 371
. me hie, S, clxxib; haec {f.s.nom.),
22; hoe {n.s.nom.), 39, 151, 163,
200, 27-4; hunc, *1.7.5;bane, 25,
IACIO iacit, 30
IACTO iactari, 1.00
1AM 31, -43,62, 154, (208]; seeB11AM
IBI 58, 198, 26-4, 308, 312
IDEM eadem (f.s.nom.), 17-4; idem
(n.s.nom.), 200; ~dem, 277;
eadem (f.s.ah.), [212]; (n.p.nom.),
174
IGNIS igni (ah.), 26
IGNOBWS ignohiles (m.p.nom.), 173
IGNOTVS ignotus, 123
lllCO 293
ILLB ille, 229 (ill'), 380; illa (f.s.nom.),
33; ilium, 103, 287; illam, [213],
[286]; illi (m.s.d.), a26, 230, 231,
231, 315, 368, [391]; illo, [197];
illos, -40,40; illis {f.p.d.), 121, [197],
(226]; Sll POSTILLA

452
 BNNius' TRAGIC VOCABULARY
ILi.iC illic, 375; illacc (f.s.nom.), 386 INCllBPO increpat,30S
ILLINC [201), 201 INDB [4), 210, 367
JLLVC 201, 201, [201), [37S] INDO indidit, 99, 99
IMBER imbcr, ,4, 357; imbre, 358; INDIGNVS indigna (f.s.ah.), 12S, 1.29
imbrium, 394 INDVCO inducta, [396)
IMMOLOJl immolandam,xciv INDVO induta (f.s.nom.), 396
IMPBRATOR imperator, i INDVSTRIA industria (ah.), I0S
JMPl!JlIVM imperium (a.), 162; im- INBO inibat, 1.24
perio (ab.), 214 INBRS inertes (m.p.nom.), .266
IMPl!RO imperat, .266 INFBRVS infera (n.p.voc.), 98; 1n-
IMPIITI.O impetrcm, 3o6 ferum (n.p.g.), 15.2, 309
IMPLORO imploro, xvb INFLAMMO inftammari, 92
IMPRIMO inprimit, 400 INPJlENO infrena, 158; infrenari, 158
IMPROBO improbati (m.p.nom.), 220 INGBNIVM ingenium, [.272); ingenio
IMPROBVS inprobus, 130; inprobum (ah.), 19
(m.s.a.), 126 INGBNS ingentes, 319
IMPVDBNS inpudentes (m.p.nom.), INGBNVVS ingcnuum (m.s.a.), 272
cxxxivb INGUDIOJl ingredi, 217
IN ( +a.), xxviiit, 168, *175, 264, INGVRGITO ingurgitandum (n.s.
277, 302, 321; ( +ab.), 13, (23), nom.), xxviiit
(23), SI, 107, 120, 164, 181, 184, INICIO iniccre, 138
185,188,196, [197), *199,208,212, INIMICVS inimico (m.s.d.), 134
243, 257, 284, 295, 300, clxvii, 342, INITIVM initio, [ 199]
3SI, 365, 374, 388 INIVRIA iniuria (ah.), 125, 129
INAVRO inauratam, 213 INNITOR innixus, 29
INCBDO incedwit, 26; inccde, [23), INNOCBNS innocens, 119, 205
[23) INNOXIVS innoxium, [255)
INCBNDIVM inccndio (ab.), 41 INOPIA inopia (ah.), 16
INCBNDO inccndere, 288; incendier, INORATVS inorata (f.s.ah.), 6
*288 INQVAM inquam, [175)
INCBRTB 20.2 INSANVS insani (m.p.nom.), 266
INCBRTVS incerta (f.s.ab.), 6,351; in- INSPICIO inspicis, 234; inspice, 236
certi (m.p.nom.), clxvi; incertis INSTITVO institutum (n.s.nom.),
(m.p.ab.), 317 197
INCINGO incinctae (f.p.nom.), 26 INSTRVO irutructam, 91
INCIPIO incipiam (suhj.), 217 INSVLTO insultas, [ 124] ; insultans,
INCUNO inclinatam, 333 124
INCLVTVS inclutum (n.s.a.), 352; in- INSVM inest, 22s
clitum (n.s.a.), 48 INTBGBR integros, 401
INCOHO inchoanda, [2101; incho- INTELLEGO intellegit, 66
andum, [210); inchoandi (n.s.g ..), INTBNDO intend.it, 28
210; inchoandae, [.210); incho- INTBll 9, 48, 137, 310, 341
andas, [210] INTBllBA*9
INCOLVMIS incolumi (ah.), cxvili INTBRBO interii, 180
INCOMMODVM incommodis (ah.), INTRODVCO introducta (f.s.nom.),
131 Jo6

453
 ENNIUS' TRAGIC VOCABULARY

INTVS [48] IVVENISiuuenem, [123]; iuucnum,

INVENTORinucntor, 121 123
INVESTIO inuestita, 1I 3 IVVO iuuat, 36o
INVITVS inuitum (m.s.a.), 128; in-
uitam, 36, 128, [157]; inuitis LABO labat, J 85
(m.p.ab.), 367 LABOR labuntur,385;lapsa (£s.nom.),
INVOCO inuocant, [301] 340
loVIS louem, JOI; louis {g.),94,273, LABOR laboris, 332
cxlixb, 350; Stt IVPPIIER LACEDABMONIVS Lacedaemonia (f.s.
IPSB ipse, 52, [160], 221, 296, 303, nom.), 49
370, *371; ipsum, 285; ipsi (m.s.d.), I.ACERO lacerato (n.s.ah.), 70
315; (m.p.nom.), *16o, 268 LACRIMA lacrimae (p.nom.), 139,
IRA iram, 2JO 175 (lacrumae); lacrimis (ah.), 276
IRATVS irati (m.p.nom.), 108 LACRIMO lacrimare, 389; lacrimari,
IS is, [120], J 56, 38 I ; ea (£s.nom. ), [389]
256, [274]; id (nom.), 203; cum, LACTO lactari, [100]

61, [103], [239], 329 (~). 390; LABTVS laeto, [283]

cam, xxviiit, 136 (? stt Bo); id LAEVVS laeuam, 240; laeua (ah.), 30
(a.), 148, 198, 198, *263, 371; eius, LAPIDEVS lapideo (n.s.ah.), 140

[155]; ei, [99], [153]; ea (£s.ab.), LAPIS lapis, 109

[141), 212; eae, cxviii; eos, 271, LAQVEATVS laqueatis (n.p.ab.), 90

(lacuatis)
392 (~); iis (m.p.ab.), 268; set
LASCIVIO lasciuis, *66
INTBREA,POSTBA,PROPTBRBA
LASCIVVS lasciui, [66)
ISTAC 225
LATEO latent, 257; latuit, 42
ISTB iste, cxlixb; isti (m.s.d.), [260];
LATVS latere, 297
ista (n.p.nom.), 359
uvo lauere (prcs. inf.), 180, 276;
1snc istic, 356; istuc, 225
lauerent, 1o6;lauere (perf.ind.), 139
xsnc (adv.), 293
LBTVM letum (a.), 134; leti, xxxiv;
ITA I, *9, 70, 228, 273
leto (d.), 283
ITER iter (a.), 169, 191, 217
UBBNTER lubenter, 284
ITINER itiner (a.), 289
UBER Liberi, 352
IVBAR iubar {a.), 13
LIBERI liberi, 283, 397; liberorum,
IVBBO iubet, i, [201]; iube, 247
132,321,326;liberum, 112,363
IVDICIVM iudicium {a.), 48; iudicio
LIBEJlO libero, [ I 3 I] ; liberabo, I 3 I
(ab.), 49
IlBERT AS libertas, 2 56; libertatem,
IVDICO iudicauit, 48
141; libertate, [141]
IVGVM iugo, [157)
LIBBT lubet, 201
IVNGO iunge, 158
UCBT licet, [342], 388, 389; licuit,
lvPPIIER Iuppiter (nom.), 356 (lu-
piter), 359 (lupiter); (voc.), 176 139
LINGVA linguam, 260; lingua, [26o)
{Iupiter), 234, 361 (lupiter); see
LINQVO linquere, 128
lovis
UTVS litora (p.a.), 46
IVRO iurandum (voc.), 350
LOCO locant, 1o6; locabas, 127;
IVS ius (nom.), 155, 156; (voc.), 350;
locaui, [130], 353; locauit, *353;
(a.), *5, [148)
locata (n.p.a.), 349
IVSTVS iustum (n.s.a.), 148

454
 ENNius' TRAGIC
LOCVS loco (ab.}, 339, 388; loca
(p.voc.), xxxiv
LONGINQVB 10,t.
LOQVOR locuntur, 173
LVCEO lucet, 315, 342; luccat, [315]
LVcrvs luctum,231
LVDVS ludis (ab.}, 62
LVGVBRIS lugubri (f.ab.), 386
LVMBN lumen (nom.), 27,t.; (a.), 13,
31,t.; lumine, 235, 250, 31,t.
LVNA luna (ab.}, 29
LVX lux {voc.), 69
LYAEVS Lyacus, 121
MACHAl!RA machaera (voc. ?}, 149
MACHINOR seeft. cviii
MACTO mactassint, 287
MABROR maerores (a.). 231
MAGIS 224, 325, 338
MAGNVS magna (f.s.nom.), :us;
magno (n.s.ab.), 43, 287; magni
(m.p.nom.), 1; magna (n.p.voc.),
171; maximo (m.s.ab.), 72
MALB 12, 176, 265
MALBPACTVM malcfacta (a.}, 349;
malcfactis (ah.}, 205
MALO malim, 232; malui, 393 ;
maluit, [393]
MALVS malum (n.s.nom.), 307;
malam, 323; mali (n.s.g.), 335; MINITOR minitatur, [ 18]
mala (f.s.ab.), 10; malo (n.s.ah.), MINOR minatur, 18
287; ma1i (m.p.nom.), 155; ma1a MINVS (adv.), 315
(n.p.a.), 302; malls (m.p.d.), 265; MISER miser, 70, 261; miserum,
(m.p.ab.), 156; (n.p.ah.), 299
MANDO mandatam, 323
MANVS manus (nom.), 46; manu,
(149], 323; manus, [182]; (p.a.),
242, 322; manihus (ab.), cva
MARE mare (a.), 235, 26,t.; marl (ah.),
43; maria, [43]; (nom.), 118
MAllITVS marito (d.), 253
Milo Maro, 353
MATER mater (nom.), so,368; (voc.),
34, 38; matrem, 112
MATBRNVS matcmo (m.ah.}, 1-i,4
MATRONA matronac (p.voc.), cva
455
VOCABULARY
MAXIMB 307
Mm>s Mede (nom.), [2«]
Mm>BA Mcdca (nom.), 216, *z,u,
Medcai, 223 (Mcdcac)
MBDlCVSmcdici, (282]
MELIOR mclior (voc.), 34; mclius
(nom.), 155
MEUVS (adv.), 378
MBMBRVM mcmbra (a.), 118
MEND1cvs mcndici (g.}, z8z
M.BNm.Avs Mcnclaus, 203
MENS mcntem, 198; mcntis, 52;
mcntc, 229
MBTVO mctuunt, 348
MBTVS mctus, 374; mctu, 20, S2
MEVS mca (f.s.nom.), 2o6, (206],
215, 294; (f.s.voc.}, 38, 149 (?),
302, 371; mcalil, [142]; mcum
(n.s.a.), 163; mci (m.s.g.), 37; mcac
(f.s.g.), 380; mcae (f.s.d.), 134;
meo (m.s.ab.), xciv, *295; mca
(f.s.ab.), 141 (m~). 142 (m~), 184
(m~). 316; mcae (f.p.nom.), (175],
[175]; m~, 242,305; mca (n.p.a.},
i; m~ (n.p.g.), 37; ~s {f.p.d.},
203
MilII1A militiae (loc. ), 200
MINA minis (ab.}, 158
[180]; miseram, 180, 222; miscrac
(f.p.nom.), 139
MISBllllO miseret, 38, 140; miscrete,
182
MISERIA miseriam, [ 142]; miseria
(ah.), 142; miserias, 223
MITTO missa (f.s.nom.), 35; misso
(m.s.ab.), 14
MODBS11A modestia (nom.), 33
MODlCE 181
MODO 233,289
MODVS modo (ah.), 124; modis
(ab.), 16
 ENNIUS' TRAGIC VOCABULARY
MONSno monstrat, 313; monstrant, NBC 77, 8,4, 2'10, 320
"67, [313]; momtrabitm, 369 NBCESSB [95], 346
MONSDVM monstrum (s.nom.), 248 NBCO necctur, 2o6; necato {imp.),
MORBVS morbo (ah.), 16 J62; necasset, i78
M01t1oa moriar, [183], 374 NBGO neget, 3 II; negctur, [206];
MORO setfr. axvii negato, [362]
Moas mon, 325; mortem, 183; NBGOTIVMnegotium. (nom.), 196,
Mortis, 192; moni. 398 197; negoti, 196 (ncgotii); negotio
MORTAUSmortalea (nom.), 9; mor- {ah.), 196, [197]
talis (p.a.), 360 NBMO ncmo, 19, 187, 378; 11eminis,
MOS morem, 181; more, 278 1,40
MOTVS motum, [309]; motus (g.), NBMVSnemore,208
309 NEPA Nepa (nom.), 186
MVcao mucronem, 402 NBQVAQVAM 225
MVUl!Jl mulicr (nom.), ,49; (voc.), NIIQVB(36], 108, 108, 138, 138, 139,
34; mulierem, 373, 373; mulierum, 200, [278], 298, 321, 324, 337, 337,
3-4 [339]
MVLTO 3,4 NBQVIQVAM221 {nequicquam)
.MVLTVS multi {m.p.nom.), 68, 1,40, Nsasvs Ncrci, 119
219, 220; multos, ,42, 104; multa Nl!SCIO nescit, 195, 199
{n.p.a.), [232]; multis (m.ab.), 16 NBVB 210
MVaVS muro (ab.), 100 NBVTIQVAM21
MVSSOset frs. xxxvii, cxcix; mussct, NBX neccm, 18
*372 NI 227
MVTTIO muttire, 280 NIHIL nihil, 335, [375]; nihili, 375;
MnMID<>NES Mynnidonum, 162 nihilo, 315
NIMIVM nimium, 335
NAM [S], 72, 9S, Io6, 108, 126, ISS, N10BB Nio~ [227); Niobc, [227)
173, 197, 21s. 2"6, 232, "6s, 273; NISI 144
Stt QVIANAM, QVISNAM, VTINAM NII1DO nitidant, 136
NAMQVB*J6, *J7J, 3"6, 339 NITOa nitcrc (imp.), 19,4
NANCISCOananciscuntur, 156; nanc- NOLO nolite, 293
tus, 292 (nactus) NO.MENnomen (nom.), 186; {a.), 99,
NAScoa nascuntur, [156]; natus, 6o, 163; nominc, 2n; nomina {a.),
291; natum (a.), 272; Stt GNATA, 68
GNATVS NOMINO nominatur, 2II
NAVCVMnauci, 375 NON 66, xxviiit, I,40, 17,4, (187), 221,
NAvnAGIVM naufragio {ab.), cla
-
NAVIS nauis (s.nom.), 210; naucs
(p.a.), III, [210]; nauibus {ah.),
'-67.271, 284, 312, 327, 331, 346,
375, 389
Nosco nomus, 170; noucris, 147
46 NOSTBJl nostram, 333; nostrum
-NB 13, ,40 {men), [102], 218; Stt (n.s.a.), *i61, 246; nostri (m.p.
ANNE nom.), 397
NB 208, cva, 244, 294, 318; ut NOX noctis, 188, 191, cbcrix; nocte,
NBMO, NBQVAQV AM, NBQVIQV AM, 257
NBVB, NBVTIQV
AM NVBIISnubes (p. nom.), 357
 BNNIUS' TRAGIC VOCABULARY
NVBILVS nuhila {n.p.voc.), xxxiv 0MNIPO'I1!NS omnipotens {m.s.voc.),
NVDO nudarc, 3-41 ISO
NVGAT0lt nugator. 37S OMNIS omncm {m.), 167, 387; {f.),
NVLLVS nulla {f.s.nom.), [2.1,8], 320, 17, 230; omnes {m.nom.), 301,
374; nullum, [248]; nulla (f.s.ah.), 382; omnis {£p.a.), 234, 3SS, 36o;
129 omnia {a.), 92, 302, 346, 382
NVMEltVS numcrus, [ 102]; numeros, OPBltA opc:ram,228
*JOZ OPINO opino. 1-46
NVMQVAM 177, 2IS, 337 OPINOlt opinor. 271
NVNC 6, 64, [72], 81, I,41, 200, 207, oPOllTBT oportet, 28s, 329.
2II, 217, 222, 26S, 338, [34,6] oPPETo oppeto, 323; obpetam, 183
NVNTIVS nuntium, 63 {oppetam)
NVP11AEnuptiis {d.), 127; {ah.), 292 0PPIDVM oppidum {a.), 239
oPPLEo opplent, 326; oppletus, 394
0 69,87, 87, 87, [13S], 171, 19-4, 322, OPPRIMO opprimi, clxvii
3SO, 3S2 OPS opcm, 42; opis, 80; ope. 89,316;
0B 227; set 0BVIAM opihus (ah.), 80, 340
0BDVC0 obducere. 376 0PTIMATBS optumatcs {voc.). cv4;
0BB0 ohibo, J9Z optumatum, 34
0BIACE0 obiacent, 192 OPilMVS optuma, [34]; optumum
0BICI0 obiccta {f.s.nom.), n9 {n.s.a.), 1-46;optumam, 39; optumi
0BIVllG0 obiurgat, 203 {m.s.g.). 38; optima {n.p.voc.),
0BN0XI0SVS ohnoxiosae {f.p.nom.). 2.1,I ; optumarum, *34
2s7 {obnoxiose) 0PVLENTVS opulentum (n.s.a.). 239;
0BN0XIVS ohnoxio, [2s7] opulenti (m.p.nom.), 173; opu-
0B0ltI0Jl oborta {f.s.nom.), 167 lentae (f.p.voc.), cva
0BltV0 ohrutus, 372 0ltACVLVM oraclo {ah.), s8
0BSCVlt0 obscurat, 68 oRAno oratio, 174; orationcm, 2s8
0BSCVltVS obscura {f.s.ah.). 2S7 0llBVS orba {f.s.nom.), 83
0BSBcao obsecro, 290; obsecrans, s6 Oacvs Orci, 98
0BSEQV0lt obsequi, ,40 OltDO ordo, S
0BSBltVAn0 obseruationis, 18s ORESTESOrestem, 14s (Orcsten)
0BSIDI0 obsidionem. 272. 376 OltIOJl oritur, -4, 22, 387
0BST0 obstare, 40 OS ore, 120, 2$8, clxvii
0BSVM obsit, 294; obesse. -4,0 0STREVM ostteis (ah.), n3
0BV ARO ohuarant, s onosvs otioso, [199)
0BVIAM 398 O'nVM otio (ah.). 19s. *J99
0BV0LV0 ohuoluta {f.s.nom.). 41
OCCVMBO occumbant, 398 . PACTVM pacto (ab.), 324
OCCVPO occupat, 1 S4 PALAM 280
0CVLVS oculorum, 21; ocu1is {ah.). PAWDVS pallida (n.p.voc.), xxxiv
32 PAil parCD1 {f.), 329
0D1 odit, 3-48; oderunt, 348 PARCO parcam (suhj.). 13-4;pepercit,
OEN0MAVS Oenomao {ah.). 292 [ 77] ; pepercerit. 77
OFFBltO offerre, 77 PARBNS parentem, 177, 377
OMNINO 9S PAltIIIS parietes (nom.), 8,
457
 BNNius' TRAGIC VOCABULARY
PAJUO parcre, so,233; pepcristi, 39; PERGOpergunt, 180
peperit, 3Ss PBRICVLVM periculum (nom.), [280]
P A1US Parim, 64 PIIRNICIBS pemiciem, 230 (pcrmi-
PARITBJl 122, 173 tiem?)
PARITOparitat, *3o8 PBRPBnORperpcti, 337
PAllO paro, 141; paras, 272; parat, PBRTINAX pertinaci (f.s.ab.), 383
308, 328; paratum (n.s.a.), *329; PERVERSE 172
paratam, II, 329 PERVICACIA peruicaciam, [38 3]; per-
PAllnCIPO participct, 329 uicacia (ab.), 383
PAllno partiuit, 2n PBRVINCOperuince, 383
PAllTVSpartu, 73 PBRVOLOperuolat, 387
PASTORpastores (nom.), 64 PES pedcm, 2IS, 244; pedes(a.), II4,
PATBfACIOpatefeccrunt, 334 168, 187; pedum, 193, 305
PATBR pater (nom.), 51, 120, 2n; PESTIS pestem, 24, 61, cviii, 323, 329
(voc.), 87, 125; patrem, 144; PETO pcto, 323; petunt, 268 ; pctam
patris (s.g.), 37, 273 (suhj.), 81; pctens, SS; pctebant,
PATIIRNVS patcrnam, 218 213
PAnoR pati, 337; passa(f.s.nom.), 78 PHU.OSOPHIAphilosophia (ab.),
PADIA patria (voc.), 87; (ah.), cvo, xxviiit
219 PHU.OSOPHORpbilosophari, [9S1;
PATRJCOLl!S Patricoles (voc.), 322 philosophandum (n.s.nom.), 9S
PATRIVSpatriae (f.s.g.), 84 PHRYGIVSPhrygio (n.ah.), 312
PAVCVSpaucis (n.p.ab.), 9S PIACVLVMpiaculum (nom.), 280
PAVLO 33 PIE (135]
PAYORpauor, 17 PIBTASpietas, 275
PAVPEJUES pauperie, 184 PIGBT pigct, 38
PAVPBRTAS paupertas, 68 PINNA pinnis (ah.), 350
PAX pacem, SS, 310; pacis, 380 PINSIO pinsibant, 3S4
Pl!CCO peccas, 204 PLACBOplacet, 9S
PBCTVSpectus (a.), 2j6 PLAGAplagas, 187
Pl!CVS pecudi, 2SJ PLA VSTRVMplaustri, 19"
PBDVMsetft. x,viii PLAVSVSplausu (ab.), 345
Pm.IA Peliae (g.), 214, 218 PLl!BBIVS plcbcio (m.d.), 280
PBLivs Pelio (n.s.ah. ), 208 PLBBl!Splcbes, 388 (plebs); plebi, 389
PBLLIS pellem, 2 I 3 PLVS plus (a.); 196, 261
Pm.oPS Pelope, 291 POL 338
PBNDBOpendeos, 297 POLLICEOR pollicentur, 268
PER 3, *161, 214 POPVLVS populo (d.), i; populi
PBRBITOperbiteret, 178 (p.nom.), cb."Vi
PBRCBLLO perculsus, S3 PORCBOporcet, 275
PBRDOpcrdat, 73 POancvs porticus (p.a.), 326
PBRDVBLLIS perd~llibus (ah.), 367 PORTVS portum,298
Pl!RBO pereat, 205 ; pcriisse, 348 POSSVMpotes, 8; potest, [309 ], 325,
(perisse) 327, 331, 337; possis, 260
PBRGAMVMPergamo (d.), 61; Per- POST (adv.), liii, 358
gama (a.), 73 POSTBA357
 ENNius' TRAGIC VOCABULARY
POSTII.LA 6o PVBLLApuella (voc.), 371
POSTVLO postulat, SS PVBR puerum., S9, 1o6; pueros, 247
POTBSTAS potcstas, 309 PVGNO pugnant, 9
POTIOR potiri, 380 PVLSVS pulsu, 30S
Pons potis (m.nom.), 32-4,380 PVL VIS puluis, 387
POTIVS 373 PVRVS purus, 65; purum (n.s.a.), 256
PRABCO praeco (voc.), i PVTO putat, 308
PRABSIDIVM praesidi, 81 (praesidii) PVTVS putus, 65
PRABTBR (adv.), 202
PREMO premunt, 3-4S QVA [36o]
PRIAMVSPriamus, 52; Priami, 87; QVADRIIVGVS quadriiugo (m.ah.),
Priamo (d.), 39, S9, 93 79
PRIMVS primus, S9 QV ADRINGBNn quadringentos, 11-4
PRINCIPIVM principium. (nom.), 3-41 QVADRVPBDANS quadrupedantes, 169
PRIVO priuem, 182 QV ADRVPES quadrupedum., [ I S7]
PRIVSQVAM 183, 236, 323 QVABRO quaesendum (m.p.g.), 112,
PRO (interjection), 361; (prp.). 10, 132
II, 1-46,205, 398 QVABSTVS quaestus (g.), 343
PROA vvs proaui (s.g.), 273 QVAM [113], 196, 224, 233, 307,
PROBVS prohus, 127, 130; prohum. clxvii, 338, 373, 378, 392, 392; Stt
(m.s.a. ), liii NVMQVAM, PRIVSQVAM, QVISQVAM
PROCBDO procede, 193 QVANTVS quantum (n.s.nom.), 331;
PROCVL 156, 219 quanto (m.s.ah.), 339; quantis
PROEilO proeliant, 9 (f.p.ah.), 103, 340
PROEIJVM proelio (ah.), 330 QVAPROPTBR 64
PROFBRO proferrc, 193; prolato QVASI 109, 314
(n.ah.), 2 -QVB i, *36, -43, [70], Io6, IS4, 174,
PROFITBOR profitcri, 28 S 234, 23s, 242, 246, 246, 2ss.
PROFLVO profiuens, 324 cccrivb, .287, 289, 317, 34S, 36o,
PROGBNIBS progeniem, 39 400; Stt NAMQVB, QVISQVB, VNDI-
PROGNATVS prognatus, 291 QVB
PROHIBBO prohibe, [236]; prohi- QVBO quit, 221
bcssis, 236 QVI qui (relative), S9, 73, 74, 99,
PROIECTO proiector, 204 [141], *177, 195, 2.21,.228,234,235,
PROLOQVOR proloqui, 222, 28s 243, 256, [259], 291, 300, clxia,
PRONBPOS pronepos, cxlix b 313, 328, *342, 357, 376; quae
PROPTBR (adv.), 202; (prp.), 119, (f.s.nom.), 25, 96, 181, 211, 386;
[3S91; see PROPTBRBA, QV APROPTBR quod (nom.), 187, 263, 265, 342,
PROPTBRBA 220 342, (391]; quem, *175, 301, 348,
PRORSVS 116 348, 356, 356; quam, *175; quod
PROSPBCTVS prospectum., 167 (a.), 78, 1841 197, 228, 261; cui
PROSVM prodesse, 40, 221 (m.), [141], 197, 272, 33S, [342];
PVBUCVS puhlicam, 74, 219 (f.), 84; qua, [212]; quo (n.), 49,
PVDBT pudet, 37 SI; qui (n.s.ah.), 182; (m.p.nom.),
PVDicmA pudicitia (ah.), cxviii IS, 16o, 220, 267, 343; quae
PVDOR pudor, 275 (f.p.nom.), 207, cva, cxviii; quos,

459
 ENNius' TRAGIC
QVJ (cont.)
140, [178], 316; quae (n.p.a.), 19,
149, 3s9; quorum (m.), 68, 1s8,
283; quibus (m.d.), S, 142, 266,
268, 309; (n.ab.), 230
qui {interr.), 163; quam, 2s8,
[364]; quod (a.), 217 quo (m.ab.),
*178; (n.ab.), 81; quibus (f.ab.),
340
qui (= utinam), 287
QVIA 212
QVIDBMcxlixb, clxxib, 371
QVIBSCOquicsccre, 170
QVIN 20, 248, [300], 367
QVIS quis (intcrr.m.), 312; (f.), 386;
quid (nom.), 163, 163, 164, 18s,
188,238; (a.), 32, 66, 70, 81, [102],
199, 271, [290], 308, 364, 373; qui
(m.p.nom.), 71.
quis (indcf.), 177, [178]; qui
(n.s.ab.), 32s
QVISNAM quidnam (nom.), 290;
quemoarn, 300
QVISQV AM quisquam, 311; quicquam
(a.), 1o8 (quiquam), 3n; qucm-
quam, [3n]
QVISQVBquisquc, 3 II, 348; qucm-
que, 311; quaeque (n.p.a.), 36s
QVISQVISquicquid (nom.), 342
QVO 6, S7, 83, 83, 217, 298
QVOD 204, CV a, 290, clxxxvili
QVONDAM291
RABIO rabere, *3z
RAPIO rapit, 44, II I; rapere, 32;
raptis (f.p.ab.), 292
llAPTO raptaricr, 79
RAllBNTER I IO
BBCl!PI'Oreceptat, 366
BBCIPIOrecipiat, 298
BBCIPROCOrcciprocat, 116
BBCLVDOrecludam, 230
BBCONClllOreconcilictur, 2o6
BBCI'B 99
BEDDO reddant, 269; reddito (imp.),
304
VOCABULARY
BBDEOrcdcat, 20S
BBDVCOreduci, 392
BBfERO rcfcrte, [42]
BBPVGIO rcfugiat, 20
BBGIPICB91
BBGIMBNregimen (nom.), 203
BBGIVSrcgio, [388]
BBGNVMrcgnum (nom.), 339; (a.),
246, 282; rcgni, 320
BBGBBDIOrcgrederc, 7
llBLINQVO relinquere, [128]; relin-
qui, 392; reliqui, 282
BBMlTI'O remissa (£s.ab.), 299
BBPAGVLVM repagula (p.a.), 229
BBPERIO reppcristi, [39]
BBPVDIOrepudiato (imp.), 304
BBQVIBSCO requiescat, 299
RES res, 228, 330; rem, IS, 74, 219,
227; re, 6, 76, 176, 3SI; res (p.
nom.), 2s,; (p.a.), 234,318; rcrum,
cbrvi; rebus (d.), 203; (ab.), 340
BBSPEcro respectantibus (ab.), 71
BBSTINGVO restinguite, 42
RES'IlTOrestitat, 203
RESTO restat, [203]
RESVMOresumit, 3 SS
REX rcx, 52; regis, 214; regi, 388,
389; rege, 292; reges (nom.), clxvi
RISCVS riscus, [n6]
RVRSVSII6
RVSSVSrussis (f.ab. ), 344
SACBRsacrae (f.s.g.), 121
SACRIPICOsacrificabat, [S-4]
SABPBISS
SABPIO saeptus, 281, 282 {septus);
saeptum (n.s.voc.), 88
SABVITl!ll166, 262
SABVVSsaeuo (m.s.ab.), 4,216; saeua
(f.s.ab. ), 396
SALMACIDBS Salmacida(voc.), 347
SALSVS salsum (m.s.a.), 139; salsa
(n.p.nom.), 118
SALTVSsaltu, 72
SALVEOsaluete, 98, 241
SALVSsalum, 179
 ENNIUS' TRAGIC VOCABULARY
SANCTVSsancta (f.s.nom.), 320 SBNBX senex, 183, 184; (voc.), 361
SANGVBN sanguen(nom.), 20; (a.),180 SENTENTIA sententias, 343
SANGVIS sanguis, [20), xciv, 324; SBNTIO sentit, 27S
sanguinem, (94), 139, [180); san- SEPVLCRVMsepulcrum (a.), 298
guine, 14, (20), 41, 94, u8, 144, SERVITVS seruitutem, 142
165, 180, xciv, 297, 347 SERVO serua, 8 ; seruetis, i; seruauisti,
SANJIIS sanie, 297 224 (seruasti)
SAPil!N'IlA sapientiam, 17; sapientia SI 126, 130, 261, 265, 32S, 369; Stt
(ab.), 325 ETSI, QVASI
SAPIO sapit, 221; sapiunt, 267; sa- SIC 71, 2J8, 2S0
pere, 147; sapiens (f.), 33; (m.), SIGNITl!NBNSsignitenentibus (f.ab.),
221; sapienti (m.ab.), clxvii (sa- 97
piente) SIGNVM signa (a.), 185
SAVCIVS saucia (f.s.nom.), 216; SILl!NTIVMsilentio (ab.), clxxix
saucii (m.p.nom.), 326 SILBO silete, i
SAXVM saxum (a.), 251; saxo (ab.), SIN 127, 130, 304
II3; saxa (nom.), 385; (a.), 297, SINCBllB 108
379; wris (d.), 296 SINE 347
SCABRBOstt fr. xlvib; scabrcnt, *uJ SOCIBTASsocietas, 320
Sc.AMANI>BllScamander, 159 SOCBll soccro, [29 I]
SCBLESDM (261) SOCRVS socru, 291
SCBLBSTVSscelestum (n.s.a.), 261 SOL sol, 243; Sol (voc.), 234
scm.vs scelus (a.), 236; sceleris, 277, SOMNIVM somnio (ab.), 52; somnia
295 (p.nom.), 346; somnium (p.g.), S7
SCINDO sciciderit, 2 SI soMNVSsomnis (ab.), SI
SCIO scit, 378; scias, 339; scire, 329; soNITVSsonitus (s.nom.), 305; soni-
scibas, 272 tu, 4, clxia
SCRIBO scripstis, 177 (scripsistis) SONO sonit, 165; sonunt, 108
SCRVPBVSscrupeo (n.s.ab.), II3 SORDIDVSsordidam, 276
SCRVTORscrutantur, 187 soas sortes (nom.), S7
SE se (a.), so, 56, 137, 156, 228, soano sortiunt, 137
(257), [310), 330; sese, 9, S7, II4, SOSPBSsospitem, 377
*137, 168, 310; sibi, 132, 221, 267, SOSPITO sospitent, 246
269,clxvi, 329; se (ab.), 136, 308 SPARGO spargens,297
SECO sectae, (209] SPBCTO spectat, 187
SBCVM set CVM SPBCVSspecus (p.a.), 152
SECVRISsecuribus (ab.), 208 SPBRNO spernit, I S6
SBD sed, 21, 32, 109, 254, cxxxivb, SPBS spe, [371]
271, 288, 305, 346, (391]; set, 177, SPIRITVSspiritu, 4
303 SPLENDIDVSsplendidis (f.ab.), 171
smmo seditio, [ 102] SPOLIVM spolia, 347
SEMEL 233 SPVMO spumant, 118
SEMITA semitam, 267 SQVALIDVSsqualidam, (276); squa-
SEMPBll228, 270, 336 lida (f.ab.), 281
SBNECTAsenecta (ab.), *300 SQVALVSsqualam, 276
SBNECTVSsenectute, 300 SQVAMAsquamae (p.nom.), *113

30 JTO
 ENNius' TRAGIC
STATIM IS
STATVO statuerit, 1S
STATVS stata (f.s.ab.), cxviii
STELLA stcllas, 190; stcllis (ab.), 171
STl!RNO strata (f.s.nom.}, 276
STlllPS stirpem, 363
STO Stant, 84, 85;. stare, *255;
stcterit, 1S
STOLA stola (ah.), 281, 282, 386, 396
STOLIDVS stolide (voc.}, 66
STRBPITVS strcpiti (g.). 164; strcpitu,
[4]
STVDBO studct, 198
STVDIVM studiis, (309]
STVLTVS stultus, [259)
SVB 232
SVBBX subices (a.}, 3
SVBIJMIS sublimis (m.s.nom.}, *190;
sublime (n.s.a.), 169, [190], 301
SVBUMO sublimat, 243
SVBUMVS sublimum (n.s.a.), 190;
sublimas, 3
SVCCJINSBO succemct, 371
svcCINGO succincta (f.s.nom.), 386
SVDO sudat, 16S
SVDOR sudore, 347
SVM sum, 83, 183, [2611; CS, *70; est,
19,95,127, 130,130, ISS, 163, 16..
170, 187, 196, 200, [248), 256,
[259], .262, [274]. 280, 290, 309,
320, 3~ clxxib, 33S, 33S, 341, 342,
342, 346, 356, 3S1, 374, 37S, 386;
sumus, 200; swit, 140, 207, cxviii,
359; omitted 346; sim, 82, 261;sit,
185, (236), [238), 265; siet, 248;
cssc, [i), 61, 126, 146, *z28, [236),
270, 300, [309]; erat, 120; fuat,
151; fore, 261; fueratis, [263);
fuissc,*148
sum (auxiliary verb), 16, 3S,
119; omitted 78; est, 31, 32, SI,
106, 167, 197, 201, 263, *292, 312,
386; omitted 292; SWlt, 220, 284;
foret, S9; fuit, 396
SVMMVS summe, 176, 234; sum-
mam, 10; summum (n.s.a.), 105;
VOCABULARY
summa (f.s.ab.}, 10s,[153]; (n.p.
a.). clxia; summarum, [ clxvi];
summis (n.cl}, 296; (f.ab.), 80
SVMO sumpscrint, 367; sumptus, S3
SVPBRO supcrat, 189; superauit, 72
SVPIIBS1Tl10SVS supentitiosi (m.p.
nom.}, cxxxivb; supentitiosis (f.p.
ab.}, 3S
SVPBRSTITO supentitcnt, 246
SVPBRVS superum (n.p.g.). 309
SVPPBTO suppetit, 331; suppetcbat, 3 39
SVPPUCO supplicarcm, 226
SVSCITO suscitant, 343
SVSPICIO suspicionem. .262, 277
svsPDto suspirantibus (f.p.ab.), S3
SVSTINEO sustinet, 330
-
svvs suum (m.s.a.), 198; suam, 219;
sitl (m.s.g.), 343; suo (m.s.ab.), 73;
(n.s.ab.), 314 (suo); suos, [102);
suarum, clxvi
TABEO tabet, *102
TABES tabcs, [102]
TABVM tabo (ab.), 297
TACEO tacete, i; tacere, 146
TABDA taedis (ab.), 27
TABNIA taeniis (ab.), 67
TAUS talem (m.), 128
TAM S, 19, 380
TAMETSI [172]
TANDEM 176
TANTALVS Tantalo (ab.), 291
TANTVS tanta (f.s.nom.), 295; tan-
tum (n.s.a.), 331; tanta (f.s.ab.), 19,
226; tantae (f.p.nom.), S7
TARDVS tarda (f.s.ab.), 300
TECTVM tectis (ah.), 90
TEI.AMO Telamonis, 273
TELVM tela (nom.), 143
TEMERB 318
TEMO tcmo, I 89
TEMPEJlO tcmperaret, 6o
TEMPLVM templum (voc.), 88; (a.),
240; templa (voc.), 98, 171; (a.),
cl:xia;see BXTEMPLO
TBMPTO tcmptaret, [6o]
 ENNius' TRAGIC VOCABULARY
TBNACIAtenacia (nom.), 158 TVMVLTVS tumulti (g.), 163
TBNBBRAB tcnebris (ab.}, xxxiv TVOR tuor, 379
Tl!Nl!O tcneor, 252; teneat, clxvii TVllBIDVSturbidas, 318
TBPIDVStepido (m.s.ab.), 1-4 TVnO turpari.94
TBR 232 TVTB 147
TBRRA terra (nom.), 165,250; (voc.), TVVS tua (f.s.nom.), 2o6; tuum (n.s.
352; terram, 138, 209, 235; terrae a.), 321; tuae (f.s.g.), 80, 112; tuo
(d.}, 223, 276; terra (ah.}, n-4. (n.s.ab.), 70, 235
[276]; terris (ab.), 3S5
TBRRIBillS terribilem (m.), 18 vAco uacant, 159; uacare, [255]
TBSCVM tesca (a.), 379 VAGO uagant, [159]
TEXO tcxitur, 44 V All!O ualet, 174
Tum.Is Thclis, 368 (Thetis) VALIDVSualida (f.s.nom.), 158
THBSAVllVS thesauri(p.nom.}, 192 VANVSuanum (n.s.a.), 382
TmtABcvs Thraeca (f.s.voc.), 352 VASTVS uastae (f.p.nom.), 207;
(Treca) uastos, 152
TIMIDvs timido (m.s.d.), 20 VATESuates (p.nom.), cxxxivb
TrrANIS Titanis, 363 vm 33, [s8], 1o6, 192, 299, 333, 352
Tou.o tollunt, 12; tollere, 6o -VB 81, 294; seeNEVE
TOPPER378 VBHO uecti (m.p.nom.), 213
TORREOtosti (m.p.nom.), 85 VBUVOLANS ueliuolantibus (f.p.ab.),
TRABEStrabes (s.nom.}, 209 -4S
TRABS trabem, [no]; trabes (p. VEUVOLVSucliuolas, III
nom.), [209] VBNIO uentum ·(n.s.nom.), 201
TRACTOtractcnt, 318; tractauere, 71 VENO uenor, 252
TRADO tradidit, 229 VBNTVSuentus, 357,358; uento (ah.),
TRAHO trahens, *uo 159; uentis, [358]
TRAVBRSVS trauma (f.s.ab.}, 229 VBRBVM uerba (a.), *253; uerborum
TRBMVI.Vs tremulo (m.s.ab.}, 250 24S
Tlll!S tris (a.), 48 VBI.B uerius, 373
TllivIA Triuia (nom.), 363 VBUCVNDB 181
TROIA Troiae (g.), 69; (d.), 61 VBRBORuereor, 37
TROIANVSTroiano (m.s.ab.), 100 VBRO [37]
TV tu, IO, 172, 20,4. 204, 224, 23-4, VBRSOuenat, 402
(260], 272, [321]; (voc.}, 309; te, VBRTO uertite, cva; uortam, 217;
71, 89, II9, 129, *130, 131, [151], uertant, S7

-
238, *260, 285, 290, 300, 361; ted,
151; tui, 38; tibi, 10, 176, 181, 2-49,
258, 278, 359, 363 (tibi), 371; te
(ab.), 125, 207, 321; uos (nom.},
263; (voc.), 145, cva; (a.), i, [161],
322,397
TVEORtueor, 361
TVLUVStullii (p.nom.), 14
TVM 17, [34], SS, 122, [171], [224],
[260], 264
VBRVM283
VBRVSuera (f.s.ab.), 254; (n.p.nom.),
346
VBSTBR uostrum (n.s.a.), (161];
uestra (f.s.ab.), 325; uostra (f.s.ab),
398; uestri (m.p.nom.), [397];
uestras, 242, 322
VESTIOuestitus, [28 I]
VBSTIS uestem, 276
VIA uiam, [77 ]. 267, 313 ; seeOBVIAM

30-2
 ENNIUS' TRAGIC VOCABULARY
VICIS uice, 123 VLIXBS Vlixem, 170
VICTORIA uictoria {f.s.ab.), 166, 381 vu.vs ullo {n.s.ab.), 324
VIDBO uideo, 288; uident, [187); VlTllO 154
uidetur, 188; uide, [175); uidcte, VMBRA umbra (nom.), 294
47; uidere, 78; uidi, 78, 89, 92; uisa VMIDVS umidas, 4
{f.s.nom.), 32, 51 VMQVAM 244, 321
vrovvs uiduae (f.p.nom.), 207 VNDB 4, 22, II5, clxvi
VIGIL uigilcs (voc.), 162 VNDIQVB 252
VILLOSVS uillosi {m.s.g.). 384 VNDO undantem (m.), 179
VINCO uicisse, 145; uictis (m.p.ab.), VNVS una (f.s.nom.), 49
*367 voco uocant, 64, 301, 356; uocare,
VINDICTA uindictam, *367 [255]
VIll uinun (s.a.), 254; uiri (s.g.), 38; VOLO uolant, 14; uolans, 67
(p.nom.), 212 VOLO uolt, 228, 228; uelit, 199;
VIRGINALIS uirginali, [33] uolentes, 12, 12; uolcs, 369
VIRGO uirgo, 205; (voc.), xvb; VOLVPTAS uoluptatem, 382
uirgincs (nom.), 207; (a.), 37 vox uoce, 58; uocibw (ah.), 372
VlllTVS uirtutem, IS S ; uirtute, I SS, VRBS urbem, 137; urbe, 83; urbcs
254 (a.), 36o
. . . *
VIS WS, 29S; mm, 25; W, 93, 1$3, VJlVO uruat, I 14
334; uiribw (ah.), 146 VSPIAM cla
VISCBRA11M II8 VSVRPO usurpat, 163
VITA . {nom. ), *zoz; wtam,
wta . 77, 93, VT i, 56, [77], lo6, 151, [16o), 169,
202, 232; uitae (d.), 18, 134; uita xciv, 226, [228), 228, 246, 260,
(ab.), 10, [232], 299, 398 263, 302, 300, [309), [31s], 318,
VITALIS uitalem, (3o6) 329,330,339,363,397
vms uitis (s.g.), 121 VTI [208)
vmvM uitio (d.)r cva VTINAM 183, 208, [215), 244
VITO uitari, [93] VTIQV AM see NBVTIQV AM
VJTVLOJl uitulans, 381· VTOR uti, 195; utendos, [278);
VIVO uiuitur, 202; uiuam, 374; utendas, 278
uiuere, 254 VVLNEllOuulneratw, 312
VIVVS uiua, [253); uiuam, 253 VVLNVS uulnere, II s
VLCISCO ulciscercm, 144 vxoa uxor, 2o6; uxorem, 132
 INDEX II

ENNIUS' METRES
1-2 trim(etri) iamb(ici) (?) dim. anap. (91--92
3-4 tetr(amctri) troch(aici) cat- regwce. I baec>
{alcctici) (3?) 9S tetr. troch. cat. (?)
{?) cJ6-97 tetr. dact. {96 ?)
trim. iamb. (7 ?) 98-99 trim. iamb. (98 ?)
tetr. troch. cat. (9 ?) 100-102 {?)
10-12 {?) 103-107 trim. iamb. (104, 107 ?)
13 trim. iamb. {?) 108 tetr. troch. cat.
14 tetr. troch. cat. {?) 109-110 trim. iamb. (?)
IS trim. iamb. (?) Ill dim. anap. (?)
tctr. troch. cat. (16 112-113 tctr. troch. cat. (113 ?)
16-21
V V 114 (?)
modis; 17 omnem I cx- IIS tctr. troch. cat.
animato; 21 ?) 116 (?)
22-23 (?) 117-119 tctr. troch. cat. {117,
24-30 dim(etri) anap(aestici) 119 ?)
{28-29 Apollo I arcum) 120-131 trim. iamb. (126 Cres-
31 trim. iamb. ( ?) phontcm I existimas)
32-4,2 tetr. troch. cat. {33, 36 ?; 132 tetr(ameter) iamb(icus)
VV V -
41 adcst adcst) 133 trim. iamb.
43-46 tetr(ametri) dact(ylici) 134 tetr. troch. cat.
47 colon Rci:zianum 135 (?)
48-49 tctr(amctri) troch{aici) 136 tetr. troch. cat.
so-63 trim. iamb. (63 ?) 137 {?)
64 tctr. troch. cat. 138-139 tetr. iamb.
6s {?) 140 tctr. troch. cat.
66 tctr. troch. cat. (?) 141-14,2 trim. iamb. (141 ?)
67 trim. iamb. 143 (?)
68 tetr. troch. cat. 144-145 tetr. troch. cat. (145 ?)
69-73 {?) 146-147 trim. iamb.
74-7S tetr. troch. cat. {?) 148-149 tetr. troch. cat. (148
76 tetr. troch. {?) fecisseI expedibo; 149
77 tetr. troch. cat. machaera I atque ?)
78-80 trim. iamb. ISO tetr. troch. (?)
81-83 tetr(ametri) cret(ici) (82 ISI tetr. troch. cat.
auxilio I exili I aut) 152-154 trim. iamb. (152-153 ?;
84-86 tetr. troch. cat. (84 154 castra I ultro)
V V
155-156 tctr. troch. cat.
domi; 8 5-86 ?) (?)
157-158
465
 ENNius' METRES
159 tctr. troch. cat. ucrsus Reizianus
160-161 (?) tctr. cret.
162 tetr. troch. cat. trim. iamb. (.248? ; 24-9
dico I et)
V V
163-164 tctr. iamb. (163 quid hoe
V V 250 hexameter dactylicus
hie; 164 quid incastris?) 251 trim. iamb. (?)
165-168 tctr. troch. cat. (168 ?) 252 dim. anap.
169 {?) 253 (?)
170 tctr. troch. cat. (?) 254-258 tctr. troch. cat. (254-
171 tctr. iamb. 255, 257 ?; 258
172-174 tetr. troch. cat. (173 V V
tibi ex ore)
cum I opulenti)
175 trim. iamb. (?) 259-260 (?)
176 tetr. troch. cat. 261-278 tctr. troch. cat. (274 ?)
279 (?)
177-179 trim. iamb. (179 ?)
V 280 trim. iamb.
180-181 tctr. troch. cat. (181 tibi 281 tctr. troch. cat. (?)
V
in concubio) 28.2 trim. iamb.
18.2 (?) 283-284 tctr. troch. cat. (283 ?)
183-184 trim. iamb. 285-288 trim. iamb. (286 ?)
185-186 tetr. troch. 289 tctr. troch. cat.
V
290 tetrameter bacchiacus
187 tctr. troch. cat. (quod
contractus
V
est) 291-292 tctr. troch. cat.
188-190 dim. anap. (190 sub- 293-294 tctramctri bacchiaci
I
limum agcns) 295 dim(ctcr} crct(icus) +
191 parocm(iacus) monomcter trochaicus
19.2 tetr. troch. cat. (?) 296-297 tctr. troch.
193 dim. anap. 298-299 tctr. troch. cat.
194 parocm. 300 tetrameter bacchiacus
195 dim(etcr) troch(aicus) hypermctricus
196-.207 tctr. troch. cat. (197, 301-302 tctr. troch. cat.
199, .204, 207 ?) 303 dim. crct. + dim. troch.
208-218 trim. iamb. 304 tctr. troch. cat.
219 tctr. troch. cat. 305 trim. iamb.
2.20 tctr. troch. 306 (?)
221 tctr. troch. cat. 307 dim. crct. + dim. troch.
222-223 trim. iamb. 308 trim. iamb.
2.24,-.233 tctr. troch. cat. (226 309-312 tetr. troch. cat. (3 I 1-
nam I ut; 2.27,232- 312 ?)
233 ?) 313-319 trim. iamb. (316, 319 ?)
234-236 {?) 320 (?)
237-240 trim. iamb. 321 tctr. troch. cat.
24,1-24,2 (?) vv
322-332 tetr. iamb. (323 rnalarn:
243 trim. iamb. vv
(?). 327 potcst ?; 328 ?; 331
. vv
trim. iamb. ( ?) pot.est; 332 ?)
 ENNius' METRES
333 tetr. troch. (?) 369 tetr. troch. cat.
334 trim. iamb. (?) 370 (?)
335 (?) 371-372 trim. iamb. (?)
336 dim. troch. 373 tetr. troch. cat.
337-343 tetr. troch. cat. (337 374 trim. iamb. (?)
vv
potest; 338, 342 ?) 315 (?)
344-345 dim. anap. (344 ?) 376 trim. iamb.
346 (?) 377-380 (?)
347 trim. iamb. 381 trim. iamb.
348 tetr. troch. cat. 382 tctr. troch. cat.
349 trim. iamb. 383-385 trim. iamb. (384 ?)
350 tetr. troch. cat. 386 (?)
351-354 trim. iamb. (353-354 ?) 387 tctr. troch. cat.
3SS-36o tctr. troch. cat. (359- 388-390 trim. iamb. (390 ?)
36o ?) 391 (?)
361-362 trim. iamb. (361 ?) 392-393 tetr. troch. cat. (393 ?)
363 tctr. troch. cat. 394 (?)
364 trim. iamb. (?) 395-396 trim. iamb. (?)
365 dim. anap. 391-398 tetr. troch. cat. (397 ?)
366 parocm. 399 trim. iamb. (?)
367 trim. iamb. 400 (?)
368 (?) 401-402 trim. iamb. (?)
 INDEX
MATTERS DISCUSSED IN
INTRODUCTION AND COMMENTARY
ablative, absolute, 172, 312; of com-
parison, 180; without preposition,
215, 242, 354, 363
abstract notions, expression of, 390
abstract noun, for concrete, 192;
governing transitive verb, 195;
personification of, 21of., 212£,
224
Acciw, Achilles, 161£; Alcmeo, 185;
Alphesibota, 186; Andromeda,262;
Annorum iudicium, 178; Astyanax,
236; Athamas, 267; Atreus, 414tf.;
Epigoni,185f.; Eriphyla,186; Eury-
saces, 240; Hecuba, 304(; Mtdea,
346; Melanippus, 382; Myrmidones,
162 n. 2, 164; Ptlopidae,414; Telt-
phus, 319, 336£, 404£; Tertus, 47
accusative, after impersonal passive,
338; after verbal noun, 325; and
infmitive,214,289
active for deponent, 172, 175, 286,
288, 317, 388
acton, competitions for, 22£; cos-
tume of, 21; interpolation of
scripts by, 240£, 346; jargon of,
166; making of political allusions
by, 238; number of available to
producen, 20; performing in several
plays by, 239tf.; singing by, 21;
spccia]isation by, 20; status of, 22;
use of nwks by, 22, 241
address, of greeting and farewell,
379f.; to gods, 168£, 170; to
homeland, 256; to night, 254,285;
to parts of the body, 317; to sky,
3o6; to social superion, 212,358£;
tone of, 275, 277, 385
III
adjectives, terminating with -ans,303
314; -arius, 391; -bilis, 196; -eus,
201, 264; -fer, 200; -ficus,250; -icus,
249; -idus, 229, 307; -osus, 391;
-sonus,248; -U4'lus,261; -ulus, 387;
terminating with participial form
(-potens,-tenm.s,-uolans),216, 292
wed instead of adverb, 215,270;
genitive,258,285, 352,377
adverbs, terminating with -ter, 261,
391, 392; -tim, 266, 393
after-life, 255, 331, 422
AlcumttUJEuripidi,6, 63
allegory, 193
alliteration, 42, 170£, 214, 215, 216,
280 n. 4, 281, 301, 308, 340, 390,
392,393
anachronism, 173, 218, 249, 351
anacoluthon, 196
anapaestic verse, 35£, 37,244, 328f.
anaphora, 197, 214, 246, 247, 361 n. S
Antiphon, 238
antithesis, 174, 283
appeals, 169, 298f.
Apulciw' quotations, 330
archaeology, 10 n. 1, 17 n. S, son. 3
archaism, 169, 172, 174, 183, 215,
247, 263, 292£, 311, 313, 364, 380,
385
architecture, 169f., 249, 307 n. 1
Aristarchw of Tegea, 9, 44£, 163 f.
assonance, 170£, 211, 247, 250, 264,
303,333,352,353,363,391
astrology, 326£
asyndeton, 169, 174, 175, 182, 247,
301, 358 n. 3, 424, 425
Atcllane farce, 14£
 INDEX TO INTRODUCTION
Atilius, Electra,47, 206, 254
audiences, 5, 7ff., 23, 34, 37, 49(,
253(
bacch.iac verse, 37, 231 f., 419, 422 f.
Bacchic orgies, 267 f.
Bear, 329£.
Caesar, avoidance of tragic vocabu-
lary and phraseology by, 167, 173,
175, 195, 228, 229, 232, 233, 247,
252, 259, 269, 270, 309 n. 3, 310,
321, 332 363, 379
caesura. 333, 421
Caper, 180, 290
Cato's oratory, 42(
Charisius' quotations, 267
cbiasmus, 200
chorus, 18 ff., 30(, 334(
Christian view of translation, 28
Cicero, account of Roman drama of,
25ff.; avoidance of tragic vocabu-
lary and phraseology by, 173, 175,
195, 232, 269, 309 n. 3, 321, 332,
358, 363, 379; mode of referring
to Greek poetry of, 322; quotation
of Roman poetry by, 4, 53 n. 5,
6o n. 3, 61 f., 162ff., 178, 186ff.,
204ff., 209, 220(, 247, 256, 257,
262, 270, 273 f., 284f., 305, 32off.,
325, 344f., 348(, 350, 356(, 358ff.,
366, 395, 396ff., 422 f.; translation
of Greek poetry by, 4, 46,215,222,
225, 254, 262, 270, 274 n. 1, 323 f.,
329 n. 4, 424; use of antique lan-
guage by, 167, 186 n. 9
coffered ceilings, 249
colon reizianum, 217
comparative, used for superlative,
211; formed with plus, 393
comparison, 220
concinnity,227, 278,398,402
conditional sentences, 275(, 278, 35S,
396
'contaminatio', 178, 237, 290, 319,
335, 345(
AND COMMENTARY
corruption, ancient. 57, 296, 352, 37S
cosmetics, 3 59
crctic verse, 37, 243, 246, 370, 384f.
419(, 425, 4.26
cunes, 410 f., 42 I
dactylic verse, 36, 209, 386
dative, 402
Delphic oracle, 222 f.
deponent for active. 265
dtutrbium, 29 n. 1
diaeresis, 37,169,333,421
differentiation of meaning, 177
dilemma.274£, 357
'dipody law', 276
disjunction, 169, 176, 194, 216, 259,
266, 3II, 355
divination, set prophecy
Donatus, set scholiasts
door pivots, 248
double negative, 282
'EMVafvtov, 378
ellipse, of preposition, 215, 242, 363;
of pronominal adjective, 192; of
pronoun, 194, 289, 338; of verb,
231, 243, 420
enjambcmcnt. stt metrical division
and syntax
Ennius, comedies by, 44; contem-
porary allusions in tragedies by, 44,
308, 398; education of, 43 ; patrons
of, 44, 310; predilection for Euri-
pides of, 4S, 189, 238, 319; racial
origin of, 43; relationship with
Homer of, 302; use of scholarly
commentaries by, 46, 308, 351;
AMdlts, 8, 36 n. 6, SSf., 194 n. 5,
381; Euhtmerus, 194 n. 5; Sabinat,
4 n. I
cpanalcpsis,278
'Ep1ws. 191, 201, 218£., 371
Etruscan theatrical vocabulary, 13
etymological figure, 173, 213, 220,
228,242,339,353,375
etymology, 228,257, 331, 354, 379
 INDEX TO INTRODUCTION
Euphorion, 381
Euripides, Alexandrian edition of
tragedies of, 4S, 61, 203, 208, 237,
342; posthumow popularity of,
9f., 238, 319
fairnessof complexion, 361
Fcstw, seeVerriw
.fragmentaincerta:CLXI, 63; CLXII,174
n. 3,231,271, 416£.; CLXIV, 162£.;
CLXV,4o6;CLXVI,284;CL:xvm,170;
CLXIX,418;cucx,348,389;CLXXI,
163; CLXXD,164; CUCXIII, 3o6;
CLXXV,394£., 40S, 418; CLXXVDI,
399 n. 4; CLXXIX,323; CLXXXI,178,
287 n. 2; CUCXXII,418; CLXXXIV,
418; CLXXXV,3o6; CLXXXVI, 62,
28 I ; CXC, 63 ; CXCI, 262, 349;
cxCII,28s;cxcv, 164,322;cxCVI,
4o6; CXCVII, 164; CCI, 282; CCIII,
164; CCIV,282; CCVI,262; CCVIII,
207;ccXII1,262;cCXIV,164;ccxv,
323;cCXIX,262;cCXXII,282,4o6;
cCXXIII,164;ccxxv,203;ccxxvn,
164
Fulgentius' quotations, 412
Furia, 218 f.
Gellius' quotations, 164, 2s2, 303f., 'ictus', 32 n. 3
334
genealogical obsessions, 12
gerutlve, with adjectives and parti-
ciples, 223; with verbal nouns,
326
gnomologies, S4 n. 1
Greek accidence, 236, 287, 3S0 n. 3,
381
Greek language plays performed at
Rome, s
Greek proper names in Latin, 11, 184, Italic vocabulary, 183
184 n. 2, 381
Greek syntax, 247, 248, 340, 364
Greek titles of Latin plays, s8 ff., 'Lange's law', 36 n. 8
283£., 291
Greek vocabulary, 230, 2s2f., 2s9,
291, 314
470
AND COMMENTARY
gods, epiphany of, 296£., 386£.;
naming of and referring to, 199,
268,311, 331,373f.,41o;natureof,
174 n. 3, 269, 396, 400, 423 f., 424
heroes, behaviour appropriate to,
217f., 339, 3Ss; deeds ofin Roman
art, I I ; in Etrwcan art, 14
heroines, behaviour appropriate to,
188 n. I, 213, 221, 31Sf.
hiatw, 194£., 198f., 210, 243, 272,
276,289,293,294,309,329,386
historians, use of tragic vocabulary
and phraseology by, 172, 173, 17s,
309 n. 3, 332, 3S8 n. 2, 363, 369,
379
homoeotdeuton, 17s, 242
Horace, on chorw, 19; on republican
metre, 33 n. 4; references by to
republican tragedy, 162 n. 4, 40s
n. 4
Hyginw, 204, 291, 342, 413 n. 1
hyperbole, 200, 21s, 234, 401, 408
uo-npov irpcmpov, 280,281,384,386
iambic shortening, 36£., 176, 328 n. 6
iambic verse, 29 n. 1, JS, 233,267,276
409£.
imagery, seemetaphor
imperative, 'future' form of, 278;
negative, 420
imperfect in narrative, 224
imperial tragedy, 48£.
indicative in repudiating questions,
340
intransitive for transitive, 183, 246,
411, 422
isocolon, 246
ivory, 249£.
legal language, 42, 220, 222, 244, 245,
246,263,276,277,287,298,312ff.,
401. 407
 INDEX TO INTRODUCTION
'Lindsay's law', 300, 408
Livius Andronicus, date of fint play,
3 n. 4; Achilles, 161f.; Ajax, 177;
Atulromeda,262; Odyma, 46 n. 4,
180; Ttuca (?), 47 n. 5
I.ivy on history of theatre, 3, 13
'locus Jacobsohnianus •. 276
'Luchs' law', 184, 426
ludi, 3, 12
Macrobius' quotations, 261, 279,
343 n. 6, 386
Magna Graccia, drama in, 16f.
Mcnandcr, posthumous popularity
of, JO
messenger speeches, 163, 165, 229£.,
265, 283, 293
metaphor, 170, 180 n. 1, 182, 190,
21of., 212, 215, 216, 220, 223, 225,
232, 233, 242, 259, 260, 283, 296,
304, 3rY7,324, 328, 330, 338, 353,
368,382,425
metre and dramatic content, 31, 34ff.,
40£., 163, 190, 209, 243, 2s3 n. 4.
333, 3S6f., 370, 38s, 386, 419£.
metre and verbal style, 4of., 17s.
176 n. 3, 19S, 197, 308f., 317, 3S2,
3S6, 361f.
metrical division and syntax, 176,
182, 228, 309, 333, 3SS, 366, 370,
409
metrical terminology, 32 n. 3
military language, 173. 270,294. 338,
3S4. 401
morphology, abnormal, 211, 214,
226, 232, 243, 263, 264, 289, 293,
318, 364, 370 n. 6, 385, 403, 412,
420
music, 29ff., 46f., 2S3f.
Naevius, Acontizommus, 6o n. 3 ;
Iphigtnia, 318
neologism, 26s, 279, 296, 3rY7,378,
402,404
nominative for vocative, 247f.
non-Attic Greek drama, 7 f.
471
AND COMMENTARY
Nonius Marcellus, quotations by, 56,
18of., 296, 300, 317, 34S, 37sf.,
389 n. 5, 408; misunderstandings
by. I 72, 296
non-Latin vocabulary, ste Italic voca-
bulary, Greek vocabulary
nouns, terminating with -m, 197;
-mtia/-antia, 196, 367; -ex, 169;
-mm, 182, 340; -mtntum, 278; -or,
195; -tio, 212; -tor, 268; -tus, 199;
-u,a, 22s
oaths, 169
official language, 166, 173, 17s, 184,
226, 241, 246, 249, 2ss. 268, 270,
279, 298, 312, 338, 354£., 400.
420
orchestra, 18
Orcus,2SS, 331
order of words, 172, 228, 233, 289~
296, 300, 309, 332, 334. 3S7, 358£.,
374, 384£., 391, 403, 410; ste also
disjunction
orthography, 52, 184 n. 2, 220, 234,
258, 260, 291
Ovid, Medea,48
oxymoron, 259, 411, 424, 426
Pacuvius, Armorum iudidum, 47, 178,
254; lliona, 305; Medus, 346
irapa'1)(101s. 25
parcchcsis, see assonance
parody, by comedians of tragic lan-
guage, 39, 172, 175, 220, 242, 2ss,
276, 378, 401; by tragedians of
sacral language, 286, 3 II, 324, 4II
participle, perfectwith present force,
183
Paulus, see V errius
periphrasis, 199, 200, 228, 231, 250,
2SS, 293, 327, 332, 352, 3SS, 377,
393, 400, 4n, 423
personification, 195, 210£., 224, 280,
331,380
philosophy, 168£., 174 n. 3, 191, 2s2,
255, 336, 372ff., 400. 424
 INDEX TO INTRODUCTION
Plautus, and Diphilus. 6; and Epi-
charmus, 7 n. 3; comedies asaibed
to, 6; Asinaria, 9
1
Plciad', 8
pleonasm, s« redundancy, repetition
of words, tautology
pluralfor singular,216, 223, 233, 249,
266, 331, 368£.
poets, effect of theatrical conditions
on, 31ff.; sreaaJjsat\on by, 24, 40;
status of, 6, 22 f.
pollution, 251 f., 420£.
polyptoton, 177, 211, 24S, 259£., 276,
294 n. S, 308, 315, 334, 381, 392,
396,400
post-classical Attic drama, 8 f.
1
prcistoria • of dramatic metres, 33 f.
present tense, resultativc, 376; with
future force, 227
Priscian's quotations, 304£., 343 n. 2,
346 n. 3,350
Probus, SS, 342 n. 2
prophecy, 191f:, 205, 207ff., 212, 213,
215, 217, 222£., 326, 372, 396, 397,
399
prosody, 169, 184, 194, 243, ~ 261,
324 n. 3, 370 n. S, 386 n. I, 410;
see alsoiambic shortening
psychological analysis.336
quotation at second hand, S3, 56£.
rati"oa)jsm, 201
redundancy, 196,201,211,220,226,
242, 24S, 294, 318, 337f., 353, 357,
370 n. 1, 384, 394, 403; see also
repetition of words, tautology
relative clauses, 170, 220, 313, 358,
363, 367£., 372£., 38o
repetition of words, 174, 175, 215,
220,243,278,318,333,367,3700.1
revivals, at Athens, 9, 319 n. 2; at
Rome, 47f.
rhetoric, 42£., 347, 351
rhetoricians' quotations, 273, 319,
339, 416f.
472
AND COMMENTARY
rhyme, 174, 214, 225, 252
Roman ideology, 241, 248, 252, 282,
398,407
Roman law: divorce, 274£.; homi-
cide, 193, 312; judicial sentence,
286 f.; position of slaves, 307;
torture before execution, 193 f.
Roman morality, 3SS, 362, 365, 390
Roman religion: blood, 244, 2s1,
341 ; family ritual, 227; funerals,
28o;Jupiter's altar, 2sof.; portents,
214, 221 f., 225; sacrifice, 251
Rufinian's quotations, 320
sacral language, 42, 199, 201, 21 s.
216, 221, 2i., 225, 230, ~ 248,
259, 268, 286, 292, 298, 307, 311,
320, 3i., 352, 374, 385, 399, 407,
409,410
satura,13
scene-change, 165, 284, 303 n. 1,
390 n. 3
scholars, Greek, 10, 32, 4S f., 61, 179,
203, 208, 237, 283, 308, 330, 342;
Roman, 27f., 32f., SI, 283f.
scholiasts' quotations, 162, 167, 203£.,
239£., 324£.
schools, use of dramatic saipts in, so,
52, S4
Seneca, knowledge of republican
tragedy of, S4f. ; tradition of tra-
gedies of, 49; verbal style of, 2 so
senatorial language, 172, 173 f., 187,
211, 282
smtmtuu, S4 n. 1, 253, 288, 305£.,
323, 327, 336£., 347, 361 f.
sequence of tenses, 225, 233
Scrvius, see scholiasts
shadow,421
singular for plural, 352, 37-4-
sky, shape of, 254£.
sound and sense, 170£.
sound play, see assonance
•split anapaest', 182, 198, 210, 233
stage, acton' entries on to, 20, 248,
378 f.; size of, 18, 188 n. 3
 INDEX TO INTRODUCTION
strategy, 293 £
subjunctive, 293, 318, 325, 370 n. 6
suicide, 316
27f.
c:Nyl(p1C11S,
syntactical contamination, 223
syntax of exclamations, 269
tabu on names, 199
tautology, 166, 175, 197, 214, 337,
342, J70 n. I, 425
ffXvlTCXt, 8 n. 6, 15£
Terence, handling of Greek. scripts by,
178; performance of plays of
during Empire, 48; titles of, 6o;
view of translation of, 23 ff.
Theodectes, 189
third penon for second, 166, 380
tragic arias performed alone, 253 £
tragic motifs, 207 n. 8, 248, 254, 256,
279, 285, 294, 299, 3oof .• 306, 310,
316, 379£
tricolon, 247, 374
trochaic verse, 198, 261, 276, 292,
325
Trojan legend, I I
AND COMMENTARY
'unpoctische Worter', 19-4
V arius, Thytstes, 4-8
Varro's quotations, 62£., 178£, 181,
256, 305, 328, 343 n. I, 3# n. 2,
345
verbs, amplified with con-,281, 299;
tx-, 177, 197, 22-4, 225, 293; in-,
201, 264, 270, 296, 367; ob-, 171,
215, 332; rt-, 197,422; inchoative,
198, 299; intensive, 196, 231, 2-43,
3-40,385, -419;omitting prefix, 183,
212, 213, 260, 276, 29-4, JOI, 366;
terminating with -art, 171, 196,
250, 265, 279, 378, 384; -ere, 264;
-ficart, 224, 258
Verrius' quotations, S3 n. 5, 59, 330,
332, 388, 423
venus reizianus, 384
Victori4, 229£
Virgil's imitations of tragedy, 203,
354
Volnius, 1-4
weapons, 168, 260, 291, 292