THE E P I G R A M S

OF P H I L O D E M O S
This page intentionally left blan
 THE EPIGRAMS
OF PHILODEMOS
Introduction, Text, and Commentary

DAVID SIDER

New York Oxford

OXFORD UNIVERSITY PRESS
1997
 Oxford University Press
Oxford New York
Athens Auckland Bangkok Bogota Bombay Buenos Aires
Calcutta Cape Town Dar es Salaam Delhi Florence Hong Kong
Istanbul Karachi Kuala Lumpur Madras Madrid Melbourne
Mexico City Nairobi Paris Singapore Taipei Tokyo Toronto Warsaw
and associated companies in
Berlin Ibadan

Copyright © 1997 by Oxford University Press

Published by Oxford University Press, Inc.
198 Madison Avenue, New York, New York 10016
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise,
without the prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication Data

Philodemus, ca. 110-ca. 40 B.C.
[Epigrammata. English]
The epigrams of Philodemos / introduction, [translation of] text,
and commentary by David Sider.
p. cm.
Includes bibliographical references and indexes.
ISBN 0-19-509982-6
1. Epigrams, Greek—Translations into English. I. Sider, David.
II. Title.
PA4271.P3A27 1997
871'.01—dc20 96-21919

1 3 5 7 9 8 6 4 2
Printed in the United States of America
on acid-free paper
 PREFACE

A new edition of Philodemos needs little justification. Among the very best of
the epigrammatists gathered by Philip of Thessalonica, an associate of Vergil and
other Latin poets, and a literary critic, Philodemos has received only one sepa-
rate edition and commentary, that of Kaibel in 1885, a brief Programmschrift of
27 pages, not much longer than the space allotted him by Gow and Page and earlier
commentators of the Greek Anthology. Kaibel furthermore omitted or discussed
only briefly a number of poems whose erotic contents he thought unworthy of
Philodemos, even though Cicero tells us that Philodemos' poems were full of such
themes.
In this edition, on the other hand, not only do I print and comment on all the
poems ascribed to Philodemos (including several about which there are some doubts
and two which are clearly not by him), I have also had the opportunity to treat a
recently published papyrus which contains a list of incipits to many poems known
to belong to Philodemos and, it seems, to many more which may also belong to
him. I have also attempted to do what earlier editors have done only occasionally
(Marcello Gigante being the most noteworthy of exceptions), that is to assess
Philodemos' epigrams in the light of his Epicureanism, and especially his writings
on the nature of poetry.
In brief, Philodemos is not only among the very best epigrammatists of the first
century B.C. (there admittedly being little competition), he is, thanks to the acci-
dent of Vesuvius, now our source of much Hellenistic speculation (some of it his
own) on the nature of poetry. And as both poet and Epicurean he had several famous
followers among the Italians of his day and later, not least among them Vergil and
vi Preface

Horace. It is time, therefore, to take stock of the scholarship of the last 110 years,
and to offer a new text and commentary of this poet.

Mindful of all the flaws which doubtless remain, and remain mine alone, I would
nonetheless like to thank the many people who have removed even more, or who
have provided access to materials: Elizabeth Asmis, Alan Cameron, Tiziano Dorandi,
Clarence Glad, A. H. Griffiths, Dirk Obbink, Peter Parsons, and Anastasia Sum-
mers for letting me see work in advance of publication; Rosario Pintaudi and Dirk
Obbink for answering papyrological questions; Alan Cameron, Diskin Clay, Chris-
topher Faraone, Anthony Grafton, Thomas Hillman, Ludwig Koenen, Nita Krevans,
Dennis Looney, Georg Luck, Myles McDonnell, Richard Mason, Carol Mattusch,
J0rgen Mejer, Dirk Obbink, Matthew Santirocco, Alan Shapiro, and Jacob Stern
for comments on earlier stages of various sections; Gerhard Koeppel, Amy Richlin,
Roger Bagnall, and David Konstan for the opportunity to try out some ideas before
critically receptive audiences in Rome, Lehigh, New York, and Providence; the
librarians of Fordham and Columbia Universities for all the aid that professionals
can and do cheerfully provide; and the several manuscript and rare-book librar-
ians in the United States and Europe who provided me with access to and copies of
their rare and unique material; Fordham University for providing much needed
support and leave for writing and travel to libraries; and my immediate predeces-
sors in the study of Philodemos' epigrams, chiefly A. S. F. Gow, D. L. Page, and
Marcello Gigante, who are cited too much for where I disagree with them and not
enough for where I have learned from them. And over and above the several par-
ticular reasons given above for thanking Dirk Obbink, I am happy to add the many
conversations we have had over the past few years on numerous aspects of Philo-
demos' poetry and poetic theory. His advice, probably to my detriment not always
taken, has helped to give impetus and shape to my work.
The typescript of this book was submitted in the spring of 1994. After the read-
ers for Oxford University Press made many suggestions for improvement, it was
then my extreme good fortune to have as copyeditor the learned Leofranc Holford-
Strevens, whose keen eye caught errors of all sorts, from those of punctuation to
even more embarrassing scholarly gaps and lapses. Although his name is recorded
here only to credit him with some conjectures in Greek texts, there are far more
places where my messy typescript benefited from his care. Nonetheless, for all the
help I have received from him and others, all errors that remain are to be charged
to me.

Bronx, New York D.S.

August 1996
 CONTENTS

Abbreviations ix

INTRODUCTION 3
Life 3
On the Bay of Naples 12
Philodemos and the Epigram 24
Metrics 41
The Greek Anthology: Formation and MSS 45
About This Edition 54

CONCORDANCES 57

THE EPIGRAMS 61
Sigla 61
Text, Translation, and Commentary 62

P.OXY. 3724 203

TESTIMONIA AD PHILODEMUM PERTINENTIA 227

Bibliography 235
General Index 245
Index of the More Important Passages 248
Index of Greek Words 254
This page intentionally left blank
 ABBREVIATIONS

AA Archaologischer Anzeiger
AGAW Abhandlung der Gesellschaft der Wissenschaften zu Gottingen
AlPhO Annuaire de I'Institut de Philologie et d'Histoire Orientale (Brussels)
AJP American Journal of Philology
ANRW Aufstieg und Niedergang der romischen Welt
AP Anthologia Palatina
API Anthologia Planudea
A&R Atene e Roma
Art G. Arrighetti: Epicuro, Opere, 2d ed. (Turin 1973)
AR W Archivfiir Religionswissenschaft
ASNP Annali della Scuola Normale di Pisa
BACAP Boston Area Colloquium in Ancient Philosophy
BAGB Bulletin de I'Association G. Bude
BCH Bulletin de Correspondance Hellenique
BEFAR Bibliotheque des Ecoles Francaises d'Athenes et de Rome
BICS Bulletin of the Institute of Classical Studies
BIEH Boletin del Institute de Estudios Helenisticos
BMCR Bryn Mawr Classical Review
BPhW Berlinerphilologische Wochenschrift
CA Classical Antiquity
CEG P. A. Hansen, Carmina Epigraphica Graeca, 2 vol. (Berlin 1983-1989)
CErc Cronache Ercolanesi
CFC Cuadernos di Filologia Cldsica
x Abbreviations

CIG Corpus Inscriptionum Graecarum

CIL Corpus Inscriptionum Latinarum
CJ Classical Journal
C&M Classica et Mediaevalia
CO Classical Outlook
CP Classical Philology
CPF Corpus dei papiri filosofici greci e latini (Florence 1989- )
CQ Classical Quarterly
CR Classical Review
CSEL Corpus Scriptorum Ecclesiasticorum Latinomm
CVA Corpus Vasorum Antiquorum
DAA A. E. Raubitschek, Dedications from the Athenian Akropolis (Cambridge,
Mass., 1949)
DK H. Diels and W. Kranz, Fragmente der Vorsokratiker, 6th ed.
EH Entretiens Hardt
EV Enciclopedia Virgiliana
FGE D. L. Page, Further Greek Epigrams
FGrHist F. Jacoby, Die Fragmente der griechischen Historiker
GB Grazer Beitrage
GP A. S. F. Gow and D. L. Page, Garland of Philip [not to be confused
with "Gow-Page," the authors of both GP and HE; nor with GP,
Denniston's Greek Particles]
G&R Greece & Rome
GRBS Greek Roman and Byzantine Studies
GVI W. Peek, Griechische Vers-Inschriften (Berlin 1955)
HE A. F. S. Gow and D. L. Page, Hellenistic Epigrams
HSCP Harvard Studies in Classical Philology
HTR Harvard Theological Review
ICS Illinois Classical Studies
IG Inscriptiones Graecae
JAOS Journal of the American Oriental Society
JHS Journal of Hellenic Studies
K-A R. Kassell and C. Austin (eds.), Poetae Comici Graeci
KG R. Kiihner, rev. B. Gerth, Ausfuhrliche Grammatik der griechischen
Sprache, vol. 2 (in 2 parts) (Hanover and Leipzig 1898)
KS Klein e Schriften
LCM Liverpool Classical Monthly
LGPN Lexicon of Greek Personal Names
LIMC Lexicon Iconographicum Mythologiae Graecae
LSCP London Studies in Classical Philology
LSJ H. G. Liddell and H. S. Scott, rev. R. Jones, A Greek-English Lexicon
9th ed.
MAAR Memoirs of the American Academy in Rome
MAMA Monumenta Asiae Minoris Antiqua
MC Mondo classico
MD Materiali e discussioni
 Abbreviations xi

MMR T. R. S. Broughton, Magistrates of the Roman Republic

MPG J.-P. Migne, Patrologia Graeca
MPL Museum Philologicum Londinense
PAAAS Proceedings of the American Academy of Arts and Sciences
PGM Papyri Graecae Magici
PMG D. L. Page, Poetae Melici Graeci (Oxford 1962)
PP Parola del Passato
QUCC Quaderni Urbinati di Cultura Classica
RAAN Rendiconti della Accademia di Archeologia, Napoli
RE Pauly-Wissowa, Real-Encykopadie der classischen Altertumswissenschaft
RF2 M. Gigante, Ricerchefilodemee, 2d ed. (Naples 1983)
RLAC Reallexikon fur Antike und Christentum
RM Rheinisches Museum
RSF Rivista di studi filologici
SEG Supplementum Epigraphicum Graecum
SH H. Lloyd-Jones and P. Parsons, Supplementum Hellenisticum (Berlin
1983)
SIFC Studi italiani difilologia classica
SLG D. L. Page, Supplementum Lyricis Graecis (Oxford 1974)
Tait J. I. M. Tait, Philodemus' Influence on the Latin Poets. Diss. (Bryn Mawr
1941)
TAPA Transactions of the American Philological Association
TrGF Tragicorum Graecorum Fragmenta, ed. B. Snell et al.
U H. Usener, Epicurea (Leipzig 1887)
VH Herculanensium Voluminum quae Supersunt (Naples 1793-1914)
WJ Wurzburgerjahrbucher
WS Wiener Studien
ZPalV Zeitschrift des Deutschen Palastina-Vereins
ZPE Zeitschift fur Papyrologie und Epigraphik
This page intentionally left blank
THE EPIGRAMS
OF PHILODEMOS
This page intentionally left blank
 INTRODUCTION

Life
Philodemos was born ca. 110 B.C. in the Syrian Greek town of Gadara, high above
the plain of Galilee, about six miles southwest of the Sea of Galilee. In terms of
today's map it is the Jordanian town of Um Qeis, just to the east of the border with
Israel.1 Gadara plays a small role in historical accounts, being mentioned more or
less in passing by Polybios, Josephus, Pliny, Strabo, and others.2 But in literary

1. In giving Gadara as the birthplace of Philodemos, Strabo 16.2.29 (T 6) in fact confuses

Gadara (in the Decapolis) with Gazara, a town closer to the coast of the Mediterranean, but
there can be no doubt that the former was the city of his birth; cf. T. Dorandi, "La patria di
Filodemo," Philologus 131 (1987) 254-256. On Philodemos' life, cf. R. Philippson,RE 19 (1938)
2444-2447; J. I. M. Tait, Philodemus' Influence on the Latin Poets (Bryn Mawr 1941) 1-23;
P. H. De Lacy and E. A. De Lacy, Philodemus: On Methods of Inference, 2d ed. (Naples 1978)
145-155; T. Dorandi, "Filodemo: Gli orientamenti della ricerca attuale," ANRW2.16.4 (1990)
2328-2332; M. Gigante, Philodemus in Italy, trans. D. Obbink (Ann Arbor 1995) chs. 3-5.
2. Jos. B] 2.97, A] 17.320 calls Gadara a Greek city, although the name is clearly Semitic;
Meleager 2 (AP 7.417.2) refers to his town as Cf.
V. Tcherikover, Hellenistic Civilization and the Jews (New York 1970) 98. Further details in the
"Gadara" articles in RE and the Jewish Encyclopedia (4.545). Note that the "Gadarene" swine of
Matthew 8.28-34 must in fact (as at Mark 5.1-13, Luke 8.26-32) come from Gerasa, a city close
to the sea whose name is often confused with Gadara. Gadara is currently being excavated under
the auspices of the Deutsches Archaologisches Institut, Berlin; cf. P. Bol et al. "Gadara in der
Dekapolis: deutsche Ausgrabungen bei Umm Qais in Nordjordanien 1986 bis 1988," AA (1990)
193-266; B. Mershen and E. A. Knauf, "From Gadar to Umm Qais," ZPalV 104 (1988) 128-145.

3
4 Introduction

history, Gadara may properly boast of having produced seven figures of note:
Menippos, Meleager, Philodemos, Theodores, Apsines, Oinomaos, and Philo. An
inscriptional epigram of the second or third century A.D. is clearly justified in call-
ing Gadara a delight for the Muses. ?
Menippos of course is known for his jocoserious (James Joyce's word) style,
through which philosophical views of a Cynic sort are made more palatable by a
humorous tone, and Meleager too wrote Menippean satire, although this is now
but a literary footnote,4 his fame now depending entirely on his epigrams and on
the Garland into which he wove them. Philodemos, who also wrote epigrams,
although belonging to another philosophical school, invested his poems with a
spoudogeloios style which may owe something to his Gadarene predecessors.
Indeed, some of the philosophical point of his poetry has been obscured by their
light-hearted tone (as I shall show). If there was anything jocular about Oinomaos,
however, our evidence fails to show it, but he did follow his predecessors in writ-
ing poetry (specifically tragedies),5 and he wrote one work on Homer,
whose title sounds remarkably like one of Philodemos',
.6 The work of Theodoros, the emperor
Tiberius' teacher, also overlaps with that of Philodemos, in that both wrote on rheto-
ric (as did Apsines), but, as is suggested by other technical titles assigned to him,
his interests were probably more practical than theoretical.7
Gadara was large enough to support two theaters, but it was too small to contain
any of its talented sons, all of whom went elsewhere to seek their fortune (for example,
Meleager in Tyre and Cos, and Philodemos in Athens and Italy). Their talent and
ambition alone would have been enough to make them emigrate, but it is also pos-
sible that the local wars between Greek and Jewish armies for control over the region
were even more impelling. In particular, the Hasmonean ruler Alexander Jannaeus

3. Rev. Arch., 3rd ser. 35.49 = Peek GVI 1.1070.3.

4. One provided by Meleager himself: API All 3 f. = 2 HE
7.418, 419 = 3, 4. In the latter, he refers to
Anyone growing up in Gadara, even if of Greek ancestry, would learn
"Syrian," that is, the local Semitic tongue, as Meleager strongly hints in 4. And at 5.160.3 = 26,
the point of the poem depends upon the reader's knowing Jewish marriage customs; cf.
H. Jacobson, "Demo and the Sabbath," Mnemosyne, 4th ser., 30 (1977) 71 f. On Menippean
satire, see now J. Relihan, Ancient Menippean Satire (Baltimore 1993).
5. The little evidence for which is collected at TrGF 188 (1.316); cf. H. J. Mette,
"Oenomaos," RE 17.2 (1937) 2249-2251; J. Hammerstaedt, Die Orakelkritik des Kynikers
Oenomaus, Beitrage zur kl. Philol. 188 (Frankfurt 1988); id., "Der Kyniker Oenomaus von
Gadara," ANRW2.36.4 (1990) 2834-2865.
6. A further link between Philodemos and Oinomaos (who was active in the early second
century A.D.) may be provided by II (= P.Oxy. 3724) ii.2, an incipit of a Pythian oracle which
was cited by Chamaileon ap. Athen. 22e (fr. 11 Wehrli = fr. 13 Giordano). As I suggest in the
commentary, the incipit may very well belong to a poem in which Philodemos quotes the oracle,
or some part of it, only to continue with a criticism of it. Oinomaos, who seems consciously to
have followed Philodemos in other ways, would have taken his lead here too from his Gadarene
predecessor.
7. Cf. FGrHist 850 for testimonia et tituli; G. M. A. Grube, "Theodorus of Gadara," AJP
80 (1959) 337-365. The mathematician Philo of Gadara is mentioned by Eutokios as having cal-
culated the value of n with great accuracy; cf. Heiberg's edition of Archimedes, 3.258.
 Lif 5

(Yannai) captured Gadara when Philodemos was a young man, taking it only after a
ten-month siege (Jos. BJ1.86, AJ13.356), and we can presume that he forced conver-
sion to Judaism on the Gadarenes as he did at Pella (A] 13.397) about twenty-five
miles to the south.8 Although it is usually assumed that Philodemos left on his own to
seek education in Athens, it is equally likely that he was taken from Gadara by his
parents, who may not of course have gone directly to Athens. This tentative recon-
struction of his youth does not, it should be noted, rest on the supposition that
Philodemos was himself Greek; even were he a "Hellenized oriental," as Momigliano
imagines him, there would still be reason or reasons for him and his family to leave
Gadara.9 In Athens he studied with Zeno of Sidon, the head of the Epicurean school
ca. 100-ca. 75 B.C. We do not know whether Philodemos was already a convert to
Epicureanism or chose it after sampling what the other philosophical schools of Athens
had to offer. Several first-person references in his histories of philosophy seem to
indicate that he was acquainted with Academics, Stoics, and Peripatetics.10 What little
evidence we have for Philodemos' life is largely consistent in the general framework
it suggests for times and places. Biographers of Philodemos should be particularly
grateful to L. Calpurnius Piso Caesoninus, not only because he acted as the poet's
patron in Herculaneum,11 but also because his own life provides some landmarks for
that of Philodemos. And for this we are further indebted to Cicero, who found it
convenient to use Philodemos and other Epicureans in order to attack Piso in more

8. Alexander Jannaeus (104-76 B.C.) "was about as bad as a man could be"; W. W. Tarn,
Hellenistic Civilisation, 3d ed. (London 1952) 236. Cf. further Gigante,Philodemus in Italy (above
n. 1) 68; Tcherikover 246 f.
9. See below, n. 23.
10. Cf.E. Asmis, "Philodemus' Epicureanism," ANRW2.36A (1990) 2376 n. 21. On Zeno,
see further below. For summaries of Philodemos' philosophical activity, cf. Asmis; Dorandi;
Gigante, Philodemus in Italy; M. Capasso, Manuale di papirologia ercolanese (Lecce 1991) 163-
192.
11. Piso can be considered a patron because (i) he allowed Philodemos to spend much time
in his company (Cic. Pis. 68-72 = T 2; note esp. 68 amicitiam, a technical term of the patron-
client relationship); (ii) Philodemos clearly invites Piso to provide support (27; n. 1 <j>iXtdTe
Hetacov, with commentary), which he would surely appreciate (Cic. De Fin. 1.65 tells us that
Epicureans were an impoverished lot; see below, pp. 153f.); (iii) Philodemos' Good King seems
designed as much to please as to instruct Piso, to whom it is dedicated, that is, addressed (col. 43.16
f. Dorandi); (iv) The Socration of Catullus 47 (T 11), who may be Philodemos under another
name, is a frequent diner in Piso's home (see the next section of this Introduction); (v) Philodemos
himself says that the best way to make money is to allow others to share in one's philosophical
discourse:

col. 23.23—32 Jensen). Cf. R. Laurenti, Filodemo e il pensiero economico degli epicurei (Milan
1973) ch. 5, "Le fonti di ricchezza per il saggio," esp. 164-166, who points out that Philodemos
is following Epicurus on this point (D.L. 10.120).
W. Allen, Jr., and P. H. De Lacy, "The patrons of Philodemus," CP34 (1939) 59-65, argue
that this evidence proves only that Philodemos tried to obtain Piso's patronage, but their position
has not won wide acceptance. It thus seems safe to apply the term patron to Piso, especially if one
accepts the three standard criteria as outlined by R. P. Sailer, Personal Patronage under the Early
Empire (Cambridge 1982) 1: "First, it involves the reciprocal exchange of goods and services. Sec-
ondly, to distinguish it from a commercial transaction in the marketplace, the relationship must be
6 Introduction

than one oration. Tendentious as many of Cicero's details are, they must serve as the
starting point for any biography of our poet-philosopher.
Piso and Philodemos first became acquainted soon after the latter's arrival
in Italy and in the former's adulescentia,12 a vague term, which would by itself
allow for their having met when Piso was as young as fifteen or as old as his mid-
thirties. In Post Red. in Sen. 14 f.( Cicero paints a picture of the libertine Piso and
his conversion to Epicureanism: Cum vero etiam litteris studere incipit et belua
immanis cum Graeculis philosophari, turn estEpicureus (T 12). It is true that Cicero
goes on immediately to say that it was the one word voluptas that attracted Piso
to Epicureanism rather than the drier and more demanding areas of study, so that
presumably (even if Cicero's slander is only partly true)13 Piso could satisfy him-
self with this shallow Epicurean veneer without having ever to leave Rome; but
nobody of Piso's new faith would stay away from Naples for long, as it was there
that Siro and others were pleasantly employed in the professing of Epicurean
doctrine to Romans of Piso's class.14 It may then very well have been here that
Piso and Philodemos first met (as Philippson 2445 suggests), rather than in Rome,
as Cichorius and many others say; but Cicero's description of Piso's Epicurean
teachers in Post Red. in Sen. is so much more hostile than that of Philodemos in

a personal one of some duration. Thirdly, it must be asymmetrical, in the sense that the two
parties are of unequal status and offer different kinds of goods and services in the exchange."
See further Sailer, "Patronage and friendship in early imperial Rome: Drawing the distinction,"
in A. Wallace-Hadrill (ed.), Patronage in Ancient Society (London 1989) 49-62, where he de-
fends this definition against critical reviewers. Also useful in this context is B. Gold (ed.), 'Lit-
erary and Artistic Patronage in Ancient Rome (Austin 1982); ead., Literary Patronage in Greece
andRome (Chapel Hill 1987); P. White, Promised Verse: Poets in the Society of Augustan Rome
(Cambridge, Mass. 1993).
12. Cic. Pis. 68 (T 2) Est quidam Graecus quicum isto vivit, homo, vere ut dicam—sic enim
cognovi—humanus, sed tam diu quam diu out cum aliis est aut ipse secum. Is cum istum adulescentem
iam turn hac distracta fronte vidisset, non fastidivit eius amicitiam, cum esset praesertim appetitus;
dedit se in consuetudinem sic utprorsus una viveret nee fere ab isto umquam discederet. (As Nisbet
ad loc. notes, viveret cum is not meant to be taken literally as "dwell in the same house as"; rather
it means something more like "was always at his side.") 70 Devenit autem seu potius incidit in
istum [sc. Pisonem} . . . Graecus atque advena. White, Promised Verse (above, n. 11) 273 n. 2
lists the following additional passages where vivere cum or convictus is used of the poet and
patron: Cic. Arch. 6, Hor. Sat. 1.6.47, Ov. Tr. 1.8.29, Suet. V.Ter. 292.9 f. Roth.
13. Cicero himself provides evidence to the contrary in Pis. 17, 56, 66 f.; cf. Tait 10. As
suggested above, and as is obvious to anyone familiar with forensic oratory of any age, nothing
Cicero states about Piso and, by extension, Philodemos, can be accepted uncritically. My account
attempts to seek the truth behind Cicero's exaggeration, but some would argue that his slander
owes nothing whatsoever to the truth.
14. R. G. M. Nisbet, Cicero in L. Calpurnium Pisonem Oratio (Oxford 1961) 187 f. objects
to the view that Piso's family had an association with Campania. J. H. D'Arms, Romans on the
Bay of Naples (Cambridge, Mass., 1970) is the best overall study of life in and around Herculaneum.
Much useful information is also to be found in E. Rawson, Intellectual Life in the Late Roman
Republic (London and Baltimore 1985), esp. ch. 2, "Rome and the Italian background";
M. Griffin, "Philosophy, politics, and politicians at Rome," in M. Griffin and J. Barnes (eds.),
Philosophia Togata (Oxford 1989) 1-37; D. Sedley, "Philosophical allegiance in the Greco-
Roman world," ibid. 97-119 (103-117 on Philodemos).
 Life 1

Pis. that, even within the bounds of Cicero's rhetorical exaggerations, it seems
unlikely that Philodemos was among their number in Rome. Since Piso won the
consulship on his first attempt (sine repulsa, Cic. Pis. 2), in 58 B.C.,15 he was born
probably no later than 101 B.C. Nisbet (p.v), like others, infers, from Pis. 87 videras
enim grandis iam puer bello Italico repleriquaestu vestram domum . . ., that "this
might almost suggest that he was born about 104 or 105," but as there is some
rhetorical point to be made from Piso's being old enough to understand what
was going on during the years 91-87 B.C., there is no strong reason for setting
Piso's birth before 101, since forty-two was the minimum age for the consulship.16
The year 101 therefore seems the most reasonable date for Piso's birth. And fif-
teen years after 101, then, is the earliest year in which Philodemos and Piso could
have met, although the annoying vagueness of Roman adulescentia allows their
meeting to have occurred many years after 86 B.C., and Cicero, although he may
be blurring the chronology here, suggests that Philodemos—an honorable man,
Cicero says, when he is not in Piso's company—hesitated to refuse a Roman sena-
tor,17 which status Piso would have attained ca. 70, when Philodemos need not
have been much older than Piso's thirty. Thus, the evidence for his relationship
with Piso suggests a birth date of ca. 110 B.C., but obviously Philodemos may have
been born several years earlier or later.
This date, moreover, is consistent with Philodemos' having been anthologized
by Philip and not by Meleager, who would surely have included a fellow Gadarene
had he known his epigrams, but whose last author seems to be Archias of Antioch,
born ca. 120. We cannot know when Philodemos began to write or publish his epi-
grams, but if Meleager's Garland dates to the 90s of the first century, as has been
argued most recently by Gow-Page and Cameron, a birth date of ca. 110 B.C. could
not be antedated by many years.18
That Philodemos had already taken up residence in Italy by 70 is shown
by his having dedicated his Rhetoric, in which he refers to his teacher Zeno of
Sidon in the present tense, to C. Vibius Pansa Caetronianus, of known Epicu-

15. Dated in part from his proconsulship in Macedonia, 57-55 B.C., on which see Nisbet
op. cit., app. l,pp. 172-180. For the evidence, cf. T. R. S. Broughton,MMR2.193f., and Supple-
ment (1986) 47, who estimates that "his quaestorship, aedileship and praetorship may therefore
be attributed to the normal years 70, 64, and 61, respectively."
16. Sulla's minimum was occasionally ignored in the late Republic, most notably by
Caesar, Piso's son-in-law, who became consul at age forty. Cf. R. Develin, Patterns in Office-
Holding 366-49 BC, Coll. Latomus 161(Brussels 1979) 96-101. On the other hand, cf.
E. Badian, "Caesar's cursus and the intervals between offices," Studies in Greek and Roman
History (Oxford 1964) 140-156, who argues that Caesar and others known to have held
office below the minimum age were beneficiaries of a special dispensation granted only to
patricians. The plebeian Calpurnii Pisones could not, therefore, have served before the fixe
time.
17. Pis. 70 (T 2) Graecus facilis et valde venustus nimis pugnax contra senatorem populi
Romani esse noluit. For senatorem, ms. V reads imperatorem, that is, when he was proconsul,
but as Nisbet says, this is inconsistent with Piso's having been adulescens.
18. Gow-Page, HE l.xiv-xv; Alan Cameron, The Greek Anthology from Meleager to
Planudes (Oxford 1993) 49-56.
8 Introduction

rean leanings.19 Zeno was succeeded as head of the Garden by Phaidros, who in
turn was succeeded by Patron in 70/69. If we assume that succession occurs
at the death of one's predecessor, and further assume that the present tenses are
not merely a literary convention,20 then Zeno (born ca. 150) must have died
between 79 and 78, when Cicero heard him lecture in Athens (ND 1.59) and,
say, 72, to allow Phaidros at least a year or so as head.21 This would also allow
a similarly short time for Philodemos to be in Italy before meeting Piso. On
the assumption that Piso was indeed a senator and that Philodemos would not
have spent very many years in Italy before meeting him, I would put his arrival
there and the composition, or at least preparation for Roman publication, of the
Rhetoric ca. 74-73,22 We have arrived, therefore, at the following tentative par-
tial scheme:

19. Cf. C. J. Castner, Prosopography of Roman Epicureans (Frankfurt 1991) 80. The cor-
rect reading and identification of the dedicatee of Rhei. IV is owed to T. Dorandi, "Gaio Bam-
bino, " ZPE 111 (1996) 41 f., who in place of the earlier reading (1.223.5
Sudh.), now reads This now expelled "Gaio bambino" was identified
as the dedicatee of Lucretius' poem; cf. Allen, and DeLacy (above, n. 11), 64, who made the (as
it seemed then) reasonable point that this work of the 70s could have been dedicated to C.
Memmius as a potential patron before Philodemos met Piso, with whom he was more success-
ful; cf. further De Lacy and De Lacy, Philodemus (above n. 1), 150. Most scholars, however,
rejected this identification, preferring to see him as C. Calpurnius Piso Frugi, born ca. 88 B.C.,
and later to be Cicero's son-in-law; cf. PhilippsonRE2445, DorandMNRW2381; Asmis/lNRW
2400 n. 70; Gigante, Philodemus in Italy (above, n. 1) 31-32.
20. Tail 3 is properly cautious here; she also points out that Philodemos may have writ-
ten the work earlier, when Zeno was still alive, and that he added the dedication to an Italian
later when he was in Rome looking for a patron. This would allow Philodemos to be in Rome
in the 80s, when Piso attained his adulescentia, but it still seems better, as I argue in the text,
to have Piso older, in accord with Cicero's reference to him as senator, and at a time when
he would be more likely to take on the role of patron. Dorandi, "Gaio Bambino," 42, argues
for the likelihood that Philodemos began writing the Rhetorica in the 60s and that the dedi-
cation to C. Pansa appeared only in the last book, written in the 40s. But, as argued above,
Zeno's being referred to in the present tense suggests that at least these passages were written
no later than the 70s. The references to Zeno are as follows:
II col. 53.10-11 Longo = Suppl. 45.1 Sudh.) and
(48.13 f.).
On Phaidros, cf. A. E. Raubitschek, "Phaidros and his Roman pupils," Hesperia 18 (1949)
96-103, repr. in The School of Hellas (Oxford 1991) 337-344, esp. pp. 97 f. = 338 f.
21. T. Dorandi, Ricerche sulla cronologia dei filosofi ellenistici (Stuttgart 1991) 64, places
Zeno's death ca. 75; so also H. M. Hubbell, The Rhetorica of Philodemus (New Haven 1920)
259. Cf. Dorandi ibid. 52 f. for a brief account of the evidence for the Epicurean scholarchs.
22. Nisbet (above n. 14) 183 puts his arrival "about 75 or 70," but the data cited above
concerning Zeno and the Rhetoric suggest that this range is a little too broad. Similarly, Hubbell
(loc. cit. above, n. 21), arguing only from the Rhetoric, without considering Philodemos' con-
nections with Piso, may be faulted for arbitrarily placing Zeno's death in 75, hereby having
Philodemos in Italy for too long a time before meeting Piso. Tait 2 is initially the most cautious,
setting the termini for the meeting between the two at "the late 80's and the early 60's";
D. Comparetti, "La bibliotheque de Philodeme," in Melanges Chatelain (Paris 1910), 118-129
puts their first meeting ca. 85 B.C.
 Life 9

110 + ca. 5 Philodemos born in Gadara, perhaps to Greek parents;23

?-? Studies with Zeno of Sidon in Athens;24
ca. 74-73 Arrival in Italy;
ca. 73-70 Act. ca. 35-40, meets Piso.

We should also consider the story, credible in itself, but not altogether securely
stitched together from several lemmata in t\\zSouda, that Philodemos was expelled
from Sicilian Himera during a famine and plague, his Epicurean beliefs concern-
ing the gods having been thought to have brought down the wrath of the gods on
the town:

(i.) Souda s.v.

(ii) s.v. (Souda defines the word as "criticize" )

(iii = T 8) s.v.

The first two clearly derive from one passage in Aelian (fr. 40 Hercher); the third is
linked to the others by Hercher because of the common reference to Himera and
because criticism of and contempt for a god could readily have come from an Epi-
curean (all the more so given Aelian's hostility to this sect). These are not very
secure links, especially since the Philodemos named is not specifically identified as
the Epicurean, but, as was said above, they do consist with one another to form an
acceptable narrative: Plague and famine hit Himera; in searching for a scapegoat,
the citizens recall an alien Epicurean's slighting statement(s) concerning an impor-
tant god worshiped locally; Philodemos' property is seized (which suggests an
extended stay) and he is exiled (along with his books, to judge from the remains of

23. Cicero Pis. 68 (T 2) calls him a Graecus, but this need mean nothing more than that h
spoke Greek as though it were his mother tongue. Momigliano,Secondo Contribute (Rome 1950)
382 calls him a "Hellenized Oriental," but this can be no more than a conjecture; similarly, VH
1 (1793) 4: "Gadareni igitur non omnes admodum Graeci erant, sed Syri; quin immo
Hebraei et Hebraicam religionem ritusque profitentes." (Strabo, it is true, if in fact he is refer-
ring to the right Gadara, mentions the large Jewish population, but of course he writes well after
Philodemos' time; for Josephus, cf. n. 2 above.) Compare what was said above in n. 4 about
Meleager, and note the similarities found between Epigram 12 and the Hellenistic Jewish Gen-
esis Apocryphon: S. J. D. Cohen, Helios 8.2 (1981) 41-53. Of course, as Meleager himself points
out, anyone growing up in that area was familiar with Greek and more than one Semitic culture.
24. This is evident from the many times Philodemos, who can be quite polemical even
against other Epicureans, aligns himself with the views of Zeno, whom he calls a true disciple of
Epicurus; cf. e.g. Rhet. 1.,77.26 ff., and esp. 1.89.11 ff., where he refers to his writings having
been confused with those of Zeno. Note also P.Here. 1005, col. 14.6-13 Angeli (T 14) Kai

cf. K. v. Fritz, "Zeno von Sidon," RE 10A (1972) 122-124. The fragments of Zeno are now col-
lected in Angeli-Colaizzo, CErc 9 (1979) 47-133.
10 Introduction

the Villa dei Papiri). This episode would most naturally have occurred after Philo-
demos had received his Epicurean training in Athens. We also imagine that it would
have occurred soon afterward, before Philodemos learned how to temper his exo-
teric statements. It is tempting to draw neat lines on the map from Athens to Himera
and from Himera to the Bay of Naples, but Philodemos need not have been so tidy.
For example, and as an exercise in pure conjecture, he may have spent some time
with the Epicureans of Rhodes, against whom he was to direct much internecine
polemic.25 There is certainly time enough for him to have traveled to several cities
before arriving in mainland Italy at roughly age thirty-seven. He may, for example,
have visited Alexandria, which he mentions along with Rome as having "detained"
philosophers.26 But whether Philodemos actually went to Alexandria (or Rome for
that matter), this sentence coming as it does after a favorable reference to Athens
(see above) certainly suggests that nobody would want to stay long in either Alex-
andria or Rome to practice philosophy. Sticking to what evidence we do have, there-
fore, we can insert his stays in Athens and Himera into the biographical schema. A
very short stay in Alexandria is possible; a stay of longer but still unknown length
in Rome is strongly indicated by Cic. Pis. 68-72 (T 2). It is not, however, indicated,
as Gigante believes, by the epigrams, which he reads as strict autobiographical tes-
timony which must be sorted chronologically and made to conform to a scheme for
which there is, as we have seen, very little evidence. Thus, Gigante infers from
Philodemos' epigrammatic invitation to Piso (Epigram 27) that it must have been
written in Rome because Philodemos invites his Roman patron to his simple house.
The setting cannot be Herculaneum, Gigante argues, because there Philodemos
lived with Piso in the grand Villa dei Papiri, as evidenced by Epigram 29. As I show

25. Sedley, op. cit. (above n. 14) 107—117 for a discussion of Philodemos'arguments in his
Rhetorika with the Epicureans of Rhodes. Note that Philodemos' references to Zeno suggest a
controversy continuing on after the former had heard the latter's lectures on the subject (Sedley
117) and after Philodemos had left Athens. He may of course have been responding only to what
he had read. See further F. Longo Auricchio and A. Tepedino Guerra, " Aspetti e problemi della
dissidenza epicurea," CErc 11 (1981) 25-40.
26. R£rf.2.145,fr.3.8-15

. (Hubbell suggests that the reference to "necessity" may mean "as hos-
tages.") In De Morte he describes people eager to spend years studying philosophy in Athens,
and then further years touring Greece and some non-Greek sites; after which he would (hope
to) spend his remaining years at home in conversations with friends and relatives (

11 Gigante]). Philodemos himself, however, may have had no expectation of returning home
(cf. below, n. 34)—or, as Gigante RF2 206 believes, "home" for Philodemos may have come to
mean Herculaneum. 18 , if by Philodemos, could have been written during
his stay in Alexandria.
For other cities which could have attracted Philodemos on his early travels, cf. W. Cronert,
"Die Epikurcer in Syrien," Jahresb. d. arch. Inst. in Wien 10 (1907) 145-152.
 Life 11

in the commentary, however, another interpretation of the poem is more likely, one
which says nothing about where Philodemos lived.27
It is also worth mentioning the views held by earlier scholars that Philodemos
travelled with Piso to Macedonia during his proconsulship in 57 or with him to Gaul
in 55. He may in fact have done so—we know of other Greeks who accompanied
their Roman patrons during foreign service—but the evidence for Philodemos' hav-
ing done so derives entirely from a misunderstood passage in his poetry.28
Of his death we hear nothing. It used to be believed that his De Dis contained
a contemporary reference to the activities of Antony's political enemies:

(Book 1, 25.35 ff.29 This had been taken to refer to the events at
the end of 44, when Piso and other Caesarians opposed Antony,30 and so provid-
ing us with a terminus post quern for Philodemos' death, which is often given as
shortly after 40 (act. ca. 70), although he could well have lived many years be-
yond this date. It has now been reported,31 however, that Antony's name is not to
be read. We are thus deprived of the only means scholarship thought it had to
approximate the date of Philodemos' death. Piso himself disappears from extant
literary historical sources soon after the events of 44 (Syme op. cit. 97), and was
thought to have died soon afterward.32

27. Gigante argues for an autobiographical interpretation of the epigrams in many passages;
note in particular Philodemus in Italy 79: "The circumstances in the epigram [sc. 27] indicate
that the little dwelling that the poet offers as the location for the meeting is located in Rome and
not on the sea in Campania ... it cannot be the house with the belvedere where Philodemus had
been a guest of his patron when he mourned with Sosylus the death of his two friends Antigenes
and Bacchius.
28. G. W. Bowersock, Augustus and the Greek 'World (Oxford 1965) 3 believes that
Philodemos accompanied Piso (listing the other Greeks and their patrons). Dorandi, ANRW
2332 rejects the possibility. See the commentary on 8.4.
29. H. Diels, Philodemos uber die Cotter. Erstes Buch. (Berlin 1916) 44, 99 f. Cf. Tait 14,
Nisbet op. cit. (above n. 14) 185.
30. For the details of which, cf. R. Syme, TheRoman Revolution (Oxford 1939) 98, 117 f.
Syme's overall opinion of Piso is quite favorable; cf. esp. 135 f.
31. By K. Kleve ap. Dorandi, Buon Re 28. Another Philodemean fragment does mention
Antony without question: De Signis col.ii.18; c.4 De Lacy, (Pygmies whom)
which has been used to date this work; cf. De Lacy and De Lacy,
Philodemus (above n. 1) 163 f. for the details. But the verb's being in the middle voice allows
that Antony merely had the pygmies brought to Rome at his command, without necessarily
having conducted them either from Syria (in 54 or 40) or from Hyria (40). This Antony refer-
ence, then, provides neither a secure date for De Signis nor a firm terminus post quem for
Philodemos' death. Cf. Asmis, ANRW2.)6A.2)72. It remains possible that De Dis was com-
posed in the late 40s; cf. P. G. Woodward, "Star gods in Philodemus, PHerc 152/157," CErc
19(1989)29^47, esp. 31f.
32. Three inscriptions from Pola, however, name L. Calpurnius Piso Caesoninus, one of
them securely dated to 33 B.C.: ILLRP 423, 424, 639, now published in B. Frischer, Shifting
Paradigms: 'New Approaches to Horace's Ars Poetica (Atlanta 1991) 55 f., who uses them as part
of his argument that this Piso was the father among the dedicatees of Horace's Ars Poetica.
12 Introduction

It is possible that the now elderly Philodemos retired from public life,33 but
this is mere conjecture, and, given the difficulty of dating his works (and reading
them in their sorry physical state), it is quite possible that Philodemos continued to
write and to teach well into his eighties. A passage in De Morte on dying in another
land has usually been taken as an oblique personal reference: "Whenever [there is
the expectation of dying] in a foreign land, this naturally stings lovers of discourse,
especially if they leave behind parents and other relatives in their native land. But
it stings only so much as to prick, so as not to bring pain, [to those] involved in the
inconveniences that attend life in a foreign land" (tr. Asmis).34 It seems that Philo-
demos had resigned himself to die in Italy.
But it is not so much the exact dates as what it was that Philodemos did during
his long stay in Italy that is of the greatest interest, both for historians of the late
Republic and even more so for historians of Augustan literature. And it to this that
we now turn.

On the Bay of Naples

Despite the uncertainties outlined in the previous section, we may safely assume
that Philodemos spent at least the three decades of the 60s through the 40s largely,
though probably not entirely, in Italy; more specifically in the environs of Naples
and Herculaneum, the small but elegant coastal town about 10 km to the south
of Naples.1 Herculaneum was covered by the Vesuvian eruption of 79 A.D. and
eventually uncovered along with Pompeii in the mid-eighteenth century, enabling

E. Sacks, however, in his review of Frischer, shows that the Piso of the inscriptions is far more
likely to be a younger member of this family (BMCR 3 [1992] 113 f.). Also critical is D. Armstrong,
"The addressees of the Ars poetica: Herculaneum, the Pisones and Epicurean protreptic," MD
31 (1993) 185-230.
33. Philippson RE 2447 argues that Siro had died well before 42, which would be relevant
here, were there any evidence to support this claim.
34.

(De Morte 4 col.25.37-26.7). This passage would read all the more
poignantly if Gigante, Philodemus in Italy (above n. 1) 44 is right to maintain that De Morte was
written when Philodemos was elderly; similarly, Dorandi /1NRW2358. I tend to agree, but it
must be admitted that the evidence is stylistic and subjective. It may be that the our papyrus of
this work was written toward the end of the first century B.C.—cf. G. Cavallo, Libri scritture scribi
a Ercolano (Naples 1983) 52 f.—but this does not tell us when the work was composed.
* **
1. Literature on the Villa, to say nothing of that on the whole of Herculaneum, is immense.
On the former, let me mention only two recent works: D. Mustilli et al., La Villa dei Papiri,
Secondo suppl. a CErc 13 (Naples 1983);M. R. Wojcik, La Villa dei Papiri ad Ercolano: Contribute
alia ricostruzione dell'ideologia della "nobilitas" tardorepubblicana (Rome 1986); T. Dorandi, "La
'Villa dei Papiri' a Ercolano e la sua biblioteca," CP 90 (1995) 168-182.
 On the Bay of Naples 13

the world to get a near firsthand view of life in a first-century Italian suburb.
Among the many discoveries, which ranged from the very finest statuary and wall
painting to mundane objects of daily life, there was also found in one of the more
prosperous villas the remains of a library containing about 1,100 papyrus rolls
Of the texts that could to any extent be read, almost all were Greek.2 It has been
suggested, however, since not all the rooms in the still subterranean Villa dei Papiri
have been excavated, that the library found was the Greek one, and that a sepa
rate Latin library of the Villa remains to be found, just as Trimalchio segregated
his books by language (Petr. Sat. 48) and just as Julius Caesar intended to estab-
lish two public libraries in Rome, one for each language (Suet. D. jul. 44) 3 At
the very least, one might hope to find more copies of Lucretius and Ennius, to
say nothing of other authors sympathetic to Epicureanism, such as Vergil and
Horace.4 As the tedious (and initially ruinous) process of unrolling the papyri

On the town itself, C. Waldstein and L. Shoobridge, Herculaneum: Past Present and Future
(London 1908) can still be recommended. Cronache Ercolanesi, although given over primarily
to reports on the papyri, often contains archeological articles. See further below, n. 4. See also
L. Franchi dell'Orto (ed.), Ercolano 1738-1988: 250 anni di ricerca archeologica (Rome 1993).
2. The most noteworthy of the 58 Latin texts are: (i) A fragment of a hexameter poem o
the battle of Actium: P.Herc. 817 (ed.pr. in VH II, v-xxvi. Cf. G. Garuti, C. Rabi'rius. Bellum
Actiacum e papyro Herculanensi 817 (Bologna 1958), who has not convinced everyone that the
author is Rabirius; cf. E. J. Kenney's review in CR 10 (1960) 138 f. For bibliography, cf.
M. Gigante, Catalogo deipapiri ercolanesi (Naples 1979) 186-189. An uncritical but useful text
with English translation: H. W. Benario, "The Carmen de Bello Actiaco and early Imperial epic,"
ANRW 2.30.3 (1983) 1656-1662; A. Cozzolino, CErc 5 (1975) 81-86. A historical account:
G. Zecchini, // Carmen de hello Actiaco: Storiografia e lotta politica in eta augustea, Historia
Einzelschr. 51 (Stuttgart 1987); text with commentary in E. Courtney, The Fragmentary Latin
Poets (Oxford 1993) 334-340, who tentatively suggests that it could have come from Cornelius
Severus, Res Romanae; M. Gigante, "Virgilio e i suoi amici tra Napoli e Ercolano," Atti e Mem.
dell'Accad. Naz. Virgilmna59 (1991) 113-117. (ii) Fragments of Lucretius: K. Kleve, "Lucretius
in Herculaneum," CErc 19 (1989) 5-27; now available in M. L. Smith's Loeb edition of Lucretius,
(iii) Fragments of Ennius' Annals: K. Kleve, "Ennius in Herculaneum," CErc 20 (1990) 5-16.
Other Latin papyri are unidentified, the literature on them being purely paleographical; cf. the
list in Gigante, Catalogo 57, supplemented by M. Capasso, "Primo supplemento al catalogo dei
papiri Ercolanesi," CErc 19 (1989) 210. See further M. Capasso,Manuale dipapirologia ercolanes
(Lecce 1991) 54, 82 f.
3. Cf. Capasso, Manuale (above n. 2) 52-56. That this segregation by language was well
established by Petronius' time is shown by Trimalchio's boast: tres bibliothecas habeo, unam
Graecam alteram Eatinam. Presumably, since any ordinary rich person could boast of two libraries,
Trimalchio has to improve upon this by saying "three", but when he immediately proceeds to
enumerate them, he runs out of descriptions after two. Scholars who emend tres to duas have no
sense of humor, but their arithmetic is quite correct; cf. R. J. Starr, "Trimalchio's Libraries."
Hermes 115 (1987) 252 f.
4. Cf. M. Gigante, Philodemus in Italy, trans. D. Obbink (Ann Arbor 1995), chs. 1-2, and
Capasso, Manuale (above, n. 2), both excellent introductions to the subject, summarizing vast
amounts of recent and not-so-recent scholarship. Plans are under way to excavate the remaining
rooms; cf. G. Gullini, progetto di esplorazione della Villa dei Papiri," CErc 14 (1984) 7 f.;
B. Conticello, "Dopo 221 anni si rientra nella Villa dei papiri," CErc 17 (1987) 9-13; A. De
Simone, "La Villa dei Papiri: Rapporto preliminare: gennaio 1986—marzo 1987," CErc 17 (1987
15-36.
14 Introduction

proceeded,5 it became clear that these Greek texts formed a specialized collec-
tion of Epicurean texts, many identifiable by subscript as works of Epicurus him-
self (his occurring in multiple copies) as well as works by Epicure-
ans such as Polystratos, Kolotes, Karneiskos, and Demetrios Lakon.6
But the author represented by the greatest number of works (some in more
than one copy) was Philodemos. It was inevitable, therefore, that this villa would
be identified as the property of L. Calpurnius Piso Caesoninus, Philodemos' patron.7
This may well be the case, although one should note that the only physical evidence
linking Piso with the villa is his name in Philodemos' Good King, which was found
therein; nor should one forget that over a century had passed between Piso and
Philodemos' deaths and the eruption of Vesuvius. The library could have been willed
to someone (whether relative or philosophical friend; but see next paragraph) who
could have moved the collection to another house; it could even have been sold
(more than once). Yet the villa clearly belonged to a man of influence and culture;
if one has to guess at a former owner, Piso certainly remains the best choice because
of the links between poet and patron indicated by Cicero and by Philodemos him-
self in Epigram 27 and in The Good King.
More interesting and more important here than the identity of the villa's owner
during Philodemos' lifetime is the fact that, to judge from the Greek texts found
therein, Vesuvius might as well have erupted on the day of Philodemos' death (al-
though we probably have copies made after that date).8 This suggests a collection

5. And as it still proceeds; over a thousand rolls or pieces of rolls remain unopened. Cf.
M. Gigante, Catalogo (above, n. 2) for a description of physical state, contents, possible titles
and authors, and bibliography of all the Herculaneum papyri; brought more up to date by
Capasso's "Supplemento," (above, n. 2) 193-264. For a survey of the hands of the papyri and of
the library of the Villa, cf. G. Cavallo, Libri scritture scribi a Ercolano Primo suppl. a CErc 13
(Naples 1983). For a description of how the unrolling is accomplished cf. B. Fosse et al., "Un-
rolling the Herculaneum Papyri," CErc 14 (1984) 9—15; and for an overview of the current state
of affairs, R. Janko, "Philodemus resartus: Progress in reconstructing the philosophical papyri
from Herculaneum," BACAP7 (1991) 271-308; Capasso, Manuale (above n. 2) 85-116.
6. Surprisingly, no Epicurean texts of Zeno or Phaidros (who spent time in Rome) have been
found. Some few texts of the Stoic Chrysippos on logic were also found: P.Here. 307, 1038, 1421.
7. An especially strong case for Piso's ownership was made by H. Bloch, "L. Calpurnius
Piso Caesoninus in Samothrace and Herculaneum," AJA 44 (1940) 490-493; M. R. Wojcik, La
Villa deiPapiri (above, n. 1). After Piso, the strongest candidate for ownership is M. Octavius,
put forward first by H. Diels in 1882 and championed more recently by B. Hemmerdinger and
others. Wojcik suggests that the Villa belonged to the Appii Claudii Pulchri; cf. CapassoManual
(above n. 2) 43—64 for a review of this controversy. Capasso himself, like Gigante, believes that
Piso was the owner, for which opinion he was criticized in a review by P. De Lacy,^4/P 114 (1993)
178—180, who rightfully stresses the fact that no positive evidence connects Piso with the Villa
(nor even, it could be added, with the town of Herculaneum).
8. There are a fewauctores incerti, so that the possibility exists that some of the texts assigned
to Philodemos were in fact written by others, some of them perhaps after his death. Thanks to
the work of Cavallo (above, n. 5), there is no doubt that a number of papyri were copied toward
the end of the first century B.C. and in at least the beginning of the first century A.D.; cf. esp. his
comments on what he calls Group R, the most recently written but containing works by some of
the earliest Epicureans (56).
 On the Bay of Naples 15

that was passed down in the family (most likely Piso's) from one generation to the
next, rather than a collection that passed into the hands of philosophically inclined
Romans (let alone a philosophically active group of Greeks), who would surely have
added texts of similar content.
In any case, the important fact remains that documents of the greatest impor-
tance for the history of Epicureanism and Roman intellectual history were discov-
ered in the mid eighteenth century. They also brought to the fore one of the most
interesting Greek authors of the first century B.C. Hitherto, Philodemos was known
solely from his thirty-five or so epigrams preserved in the Greek Anthology and
from a few references to him in ancient literature as both poet and philosopher,
the most famous and extensive of these being the vivid picture painted by Cicero
in his oration In Pisonem? Although one could not know from these references that
Philodemos had written anything other than poetry (although most philosophers
wrote), the significance of his role in Roman cultural life could be assessed from
his connection with Piso on the one hand and, although without the papyri this
would have had to remain hypothetical, with Siro on the other.
Piso we have already touched upon.10 Siro was a leader of the Epicurean circle
in Naples and the teacher mentioned in two poems of the Catalepton\

nos ad beatos vela mittimus portus

magni petentes docta dicta Sironis.
(5.8 f. = Siro F 6 Gigante)

villula, quae Sironis eras, et pauper agelle,

verum illi domino tu quoque divitiae,
me tibi et hos una mecum, quos semper amavi,
siquid de patria tristius audiero,
commendo, in primisque patrem, tu nunc eris illi
Mantua quod fuerat quodque Cremona prius.
(8 = F 7 G ) n

9. See the collection of Testimonia below, pp. 227-234.

10. See further R. G. M. Nisbet, Cicero, In Calpurntum Pisonem Oratio (Oxford 1961) 183-
188; Miinzer, "Calpurnius (90)," RE) (1899) 1387-1390.
11. Cf. the comments of R. E. H. Westendorp Boerma, P. VergiliMaronis Catalepton I
(Assen 1949) ad locc., who regards these two poems as genuine (and those who disagree as
obstinati). On Siro, cf. further W. Cronert, Kolotes undMenedemos [= K&M] (Munich 1906)
126 ff., H. von Arnim and W. Kroll, "Siro," RE 2.3 (1927) 353 f.; M. Gigante, "I frammenti
di Sirone," Paideia 45 (1990) 175-198. Westendorp Boerma 102 argues that Cat. 5 was writ-
ten in 45 B.C., but certainty is impossible. See further on Vergil, below. For the relationship
between Vergil and Philodemos, cf. Tail 48-63; cf. also M. Gigante, "Virgilio fra Ercolano e
Pompei," A&R 28 (1983) 31-50 (repr. in Virgilo e la Campania [Naples 1984] 67-92); B. D.
Frischer, At Tu Aureus Esto: Erne Interpretation von Vergils 7. Ekloge (Bonn 1975), esp. 167-
198; H. Naumann, "War Vergil Epikureer?" Sileno 1 (1975) 245-257; M. Erler, "Der Zorn
des Ilelden: Philodems De Ira und Vergils Konzept des Zorns in dcr Aeneis," GB 18 (1992)
103-126; K. Galinsky, "How to be philosophical about the end oftheAeneid," ICS 19 (1994)
191-201.
16 Introduction

Siro's pauper agellus, located in Naples (P.Here. 312 = T 15) is probably on the
same scale as Philodemos' A,vrf| KcAvdc; (27); a humble dwelling was a hallmark of
the professional Epicurean. Cf. Cic. Fin. 1.20.65 Epicurus una in domo, et ea quidem
angusta, quam magnos quantaque amoris conspiratione consentientes tenuit amicorum
greges. Quod fit etiam nunc ab Epicureis. It is interesting to note how Cicero's attack
on Piso for inelegant living (In Pis. 67 esp. nihil apud hunc lautum, nihil elegans,
nihil exquisitum) leads immediately to the section on his dealings with Philodemos
(68-72). Since the lauta elegans exquisita Villa dei Papiri has been dated on archeo-
logical grounds to sometime before the middle of the first century,12 it seems pos-
sible to conclude from Cicero's comments that (assuming the Villa dei Papiri
in fact to be Piso's—at least eventually) it had not yet been occupied by Piso in
55 B.C., the year of In Pisonem. But Wojcik has described later additions to the Villa
which she dates to the latter half of the first century. Since precise dating is impos-
sible on architectural grounds alone, it may be permissable to conjecture that these
later additions (the grand peristyle and the northeast entrance) were ordered by
Piso sometime after he had taken possession, which was also of course after he had
returned from Macedonia with the usual spoils. There may therefore have been a
period of only some fifteen years (ca. 55^40) during which Philodemos and Siro
lived in their small homes while Piso lived in his far more impressive dwelling.
Precisely where Piso and Philodemos consorted before this is not known. Piso
had duties that kept him in Rome, but there would have been many opportuni-
ties to visit the Naples area, especially if, as Cicero asserts, he was drawn to Epi-
cureanism. Our meager evidence allows for their having met in Rome and for
Philodemos' having spent many years there before taking up residence in Her-
culaneum; it equally well permits Philodemos never to have set foot in Rome
(however unlikely this is), and that Piso and Philodemos saw each other only in
the south, where of course Piso could have been the philosopher's patron even
before the construction of his magnificent villa. There is also a small possibility
that Philodemos accompanied Piso on at least some of his foreign tours in
Macedonia and Gaul (see above).
What was life like in docta Neapolis (Mart. 5.78.14) and Herculaneum? A judi-
cious averaging of Cicero's various descriptions of partying in the company of
Greeks, taken together with Philodemos' own brief but important picture, provides
a preliminary sketch. In his attacks against Piso, of course, Cicero tells us the worst;
not the extremes of human behavior which the later Roman emperors have accus-
tomed us to, but enough to disgust an earlier age. Well before the Philodemos sec-
tion of In Pisonem, we hear of parties in Rome held during Piso's consulship:

Quid ego illorum dierum epulas, quid laetitiam et gratulationem tuam, quid cum tuis
sordidissimis gregibus13 intemperantissimas perpotationes praedicem? quis te illis

12. Cf. Wojcik 35 f., Mustilli 16 f. (opp. citt. above, n. 1).

13. These last two words are sure signs that Epicureans were present: sordidus is simply
the pejorative opposite of lautus etc. (see above); gregibus seems all by itself capable of conjur-
ing up Epicureans: In addition to the well-known Epicuri de grege porcum of Hor. Ep. 1.4.16,
Cicero himself later in this speech refers to Piso as Epicure noster, ex hara producta non ex schola,
 On the Bay of Naples 17

diebus sobrium, quis agentem aliquid quod esset libero dignum, quis denique in publico
vidit? cum conlegae tui domus cantu et cymbalis personaret, cumque ipse nudus in
convivio saltaret; in quo cum ilium suum saltatorium versaret orbem, ne turn quidem
fortunae rotam pertimescat: hie autem non tarn concinnus helluo nee tarn musicus
iacebat in suorum Graecorum foetore atque vino; quod quidem istius in illis rei publicae
luctibus quasi aliquod Lapitharum aut Centaurorum convivium ferebatur; in quo nemo
potest dicere utrum iste plus biberit a ffuderit.14

As with his attack on Piso's Epicureanism in T 12 (see p. 6), Cicero may be refer-
ring to Epicureans in Rome who need not include Philodemos, if he was already
(perhaps from the very time of his arrival in mainland Italy) a full-time resident in
the Naples area. And, whether Philodemos was present or not, who would deny
Cicero the opportunity to exaggerate? On the other hand, Philodemos himself in a
few poems describes similar parties having taken place in the past; and since such
gatherings are now (in these poems) rejected in favor of a more sedate and philo-
sophical life, they are not presented in the most favorable light. Wine, women, and
song all to excess are the hallmarks of these parties (note esp. Epigram 6), which
accords with Cicero's picture. But how accurate and autobiographical are these
poems? They all seem to center on a turning point in the narrator's life, when he
turns from "madness" to a more reflective time that will be characterized by philo-
sophical discourse and marriage, the latter (and hence the former) being dated to
the narrator's 37th year. If Philodemos were the (truthful) narrator, this would occur
ca. 73-70 B.C.
If consistency were demanded, one could have a wild Philodemos meet Piso in
the late 70s (see above) at a time when the latter was converting to Epicureanism;
Philodemos could shortly thereafter remove himself to Naples, where more sober
Epicureans were gathered under Siro's leadership. Association between Piso and
Philodemos could be intermittent, at least at first, so that Piso would be more under
the influence of the hedonistic Epicureans in Rome than that of the more serious ones
in Naples. This would explain the dichotomy of T 12 between good and bad Epicu-

and has an Epicurean \s\De Fin. 1.65 speak fondly oiamicorum greges. Plutarch repeats the porcine
references inMor. 1091c, 1094a;it has also been found in Cat. AlAPora'etSocration, duae sinistrae
I Pisonis; see further below. Nor perhaps should we ignore the beautiful bronze pig found in the
Villa dei Papiri (now in the Naples Museum: Inv. 4893 (no. 27 in the list of statues in
D. Pandermalis, "Sul programma della decorazione scultoria," in D. Mustilli et al. [above, n. 1]
45; cf. Wojcik 119 f., 124 f.) Pandermalis argues that it was placed near a statue of Epicurus
which dominated the entrance to the peristyle, but Wojcik 124 f. is doubtful, as there are too
many examples of sculpted animals in Herculaneum and Pompeii to feel sure that this particu-
lar pig was part of an Epicurean sculptural program. On the other hand, there is the small pig at
the feet of Epicurus on a cup from Boscoreale; cf. M. Gigante, Civilta delle forme letterarie
nell'antica Pompei (Naples 1979) 110; A. T. Summers, Philodemus'He.p\ noir\\iai;(av and Horace's
Ars Poetica (Diss. Urbana 1995) 6-8. See also n viii.7.
14. Pis. 22. Cf. De Fin. 2.23, also aimed at Epicureans: Mundos, elegantes, optimis cods,
pistoribus, piscatu, aucupio, venations, his omnibus exquisitis, vitantes cruditatem, . . . adsint etiam
formosi pueri qui ministrent; respondent his vestis, argentum Corinthium, locus ipse, aedificium;
bos ergo asotos bene quidem vivere aut beate numquam dixerim. Cf. T. P. Wiseman, Catullus and
His World (Cambridge 1985) 43—45, aptly comparing Epicurus, fr. 67 U.
18 Introduction

reans and the uneasy reference to Philodemos in Pis. 68 ff. Consistency can some-
times be bought only at the expense of historical truth, but this picture may not be
too far off the mark: Piso acting one way with one sort of Epicurean crowd in Rome,
and acting another way with Philodemos and his circle in Naples. After his pro-
consulship, he could have built his villa in Herculaneum, primarily to devote more
time to the kind of Epicureanism toward which he was now tending.
Now, from the mid-50s on, we can picture life in Herculaneum as consisting
of days spent in the serious pursuit of Epicurean ideals (again, as described in
T 12, followed by evenings of moderate drinking, eating, philosophical discussion,
and the recitation of poetry, perhaps in the company of like-minded women.15
Philodemos provides the best evidence, but Cicero too can be tapped for a favor-
able picture of what such an evening could be like. In Pro Archia, where his client
is a Greek resident in Italy, we hear of civilized dinners enlivened by cultured
Greeks. Of particular interest is the fact that Archias was adept at the extempora-
neous composition of poetry—epigrams most likely, given the contents. Archias
was no philosopher, and doubtless differed from Philodemos in numerous other
ways, but it is tempting to see in his description of Archias a picture that Philodemos
might be willing to accept as describing his own talents.16 Philodemos too seems to
have composed theme and variations, and Cicero's picture of Archias not only
enables us to imagine a common setting for the recitation of epigrams, but also
provides some evidence for assigning an epigram to Philodemos when attribution
has been questioned. For example, Epigram 21, a dialogue between a streetwalker
and a potential customer, which is ascribed to Philodemos by the Corrector of P
(which ascribes it to Antiphilos) and by PI, is similar to Epigram 20. Gow-Page follow
P, recognizing that "the two may be deliberately contrasted variations on the same
theme" (GP 2.125), but evidently not considering that as evidence in favor of 21's
having been written by Philodemos. But this is now all but confirmed by the pres-
ence of 21 among a list on papyri of incipits almost all of those identifiable among
which belong to Philodemos. See the commentary.
It is interesting to note that two of the incipits start with the word Parthenope,
the older name for Naples and the name of its eponymous Siren (cf. Pliny 3.62): iv.
14 and iv. 15 . One hesitates to construct a
poem from the first word plus a fraction of the second, but it seems reasonable to
imagine Philodemos in Naples preaching to the converted about its charms, first
in one way and then in another, contrasting, way. A suggestive parallel may be
offered by Vergil's sphragis to the Georgics (4.563 f.):

15. For the role played by women in Epic anism, cf. C. J. Castner, "Epicurean hetairai
as dedicants to healing deities?" GRBS 23 (1982) 51-57; B. Frischer, The Sculpted Word (Berke-
ley 1982) 56 f., 61-63.
16. See below, p. 27, n. 13. For the difficulty in assigning any of the 37 epigrams under his
name in the Greek Anthology to Cicero's Archias, cf. GP 2.434 ff. As Gow-Page note, 435 n.4
the four epigrams (AP 6.16, 179-181) on exactly the same theme (three brothers bring three
different offerings to a statue of Pan) would admirably illustrate Archias' ability to supply varia-
tions on the day's events. On the symposium as the setting for the recital of epigrams, cf. Alan
Cameron, Callimachus and His Critics (Princeton 1995), ch. 3.
 On the Bay of Naples 19

illo Vergilium me tempore dulcis alebat

Parthenope studiis florentem ignobilis oti.

It is conceivable that these lines, which may of course have been written long before
the poem was completed by 29 B.C., were inspired by similarly Epicurean sentiments
expressed by Philodemos, probably in one of his epigrams. Indeed, Vergil may have
heard Philodemos himself recite them at one of the dinner parties we have been
trying to recreate.17 Note that ignobilis oti recalls Cat. 5.10 (following upon the lines
quoted above), vitamque ab omni vindicabimus cum, which as we saw is put in a
distinctly Epicurean setting.
P.Here. 312 (= T 15, to be discussed further below) shows us that Philodemos
and Siro associated with others in Naples. Catalepton 5 and 8 which even if not
genuine probably reflect the accurate recollection of someone in Vergil's circle—
links Vergil with Siro. The circle may be completed with three papyri which place
Philodemos firmly in Vergil's literary circle.18 In the two papyri first known, which
are no longer extant and must be edited from drawings now in the Bodleian Library,
the name of Vergil had, unfortunately, to be restored.

P.Herc. 1082, col.xi

All that is sure here that some Romans are being addressed concerning some sort
of philosophical activity: (Korte) seems sure.19 But this
in itself was of no small interest, especially when we realize that Varius is almost
certainly L. Varius Rufus, whose epic poem De Morte, written perhaps as early
as 44 and before 39 B.C., recalls both Lucretius and Philodemos' own
20
Furthermore, two lines of his poem are recalled in turn by two of the

17. Vergil's nickname Parthenias probably means "virginal," playing on his name, but there
could be a further play on the name of either the Siren or the town. Probably not relevant here
is Vergil's epitaph, which he is said to have composed himself: Mantua me genuit, Calabri rapuere,
tenet nunc I Parthenope; cecini pascua rura duces. Cf. further M. Gigante, "La brigata virgiliana
ad Ercolano," in M. Gigante (ed.), Virgilio e gli Augustei (Naples 1990) 7-22.
18. For earlier scholarship on this question, which until recently depended upon restora-
tion, cf. M. Gigante, "Virgilio fra Ercolano e Pompei" (above, n. 11); id. "Atakta III. Plozio o
Orazio in Filodemo?" CErc) (1973) 86 f. For an imaginative portrait of Vergil's years in Naples,
see T. Frank, Vergil (New York 1922) 47-63.
19. Korte made the addressees the philosophers by restoring
20. Cf. W. Wimmel, "Der Augusteer Lucius Varius Rufus," ANRW 2.30.3.1562-1621
(1569-1585 on DC Morte), repr. in Collectanea. Augusteertum und spate Republik (Wiesbaden
1987); H. Dahlmann,Interpretationen zuFragmenten romischer Dichter (Abh.Ak. d. Wiss. Mainz,
Geistes- u. Sozialwiss. Kl. 1982.11) 24 f.; A.S. Hollis, "L. Varius Rufus,De morte (frs. 1-4 Morel),"
CQ, N.S. 27 (1977) 187-190; P. V. Cova, "Lucio Vario Rufo," £75.441^(43.
20 Introduction

Aeneid21 and Probus names Varius among Vergil's Roman associates in Epicurean-
ism: Vixit (sc. Vergilius) pluribus annis liberaliin otio, secutus Epicurisectam, insigni
concordia et familiaritate usus Quintilii, Tuccae et Varii.22 Vergil himself mentions
Varius in E. 9.35 f. (nam neque adhuc Vario videor nee dicere Cinna I digna) and per-
haps, if the poem is his, earlier in Cat. 7, an elegant four-line epigram addressed to
Varius in which the narrator's lament of frustrated love is subordinated to the prob-
lem of whether or not it is permissible to use the Greek word to describe his
plight.23 The restoration of therefore, seemed not at all unreasonable.
The names of Varius and Quintilius (who is also of interest) appear in a sec-
ond list of Roman names found in the Herculaneum papyri: P.Herc. 253, [Oepi]
fr. 12:24

Quintilius is most likely Quintilius Varus, named by Probus above and similarly
by Servius ad Verg. £.6.13 — Siro F 9 Gigante, hortatur Musas ad referenda ea, quae
Silenus cantaverat pueris: nam vult exequi sectam Epicuream, quam didicerant tarn
Vergilius quam Varus docente Sirone. et quasi sub persona Silent Sironem inducit
loquentem, Chromin autem et Mnasylon se et Varum vult accipiP The two papyri
have room for the same four names in the same order, so that [ ] was totally
restored in PHerc 253 between Varius and Quintilius. Who could the first-named
person have been, whose name ends in -ius~t Korte's choice of Horace was long a
leading contender, but Delia Corte made a very good case for Plotius (Tucca), who
along with Varius edited the Aeneid after Virgil's death.26 But since Quintilius is

21. Cf. Varius fr. 1 Morel-Courtney vendidit hie Latium populis agrosque Quiritum I eripuit,
fixit leges pretio atque refixit, with V. A. 6.621 f. vendidit hie auro patriam dominumque potentem
I imposuit, fixit leges pretio atque refixit.
22. Probus, Vit. Verg. 10—12. Similarly, Donatus, V. Verg. 68 audivit a Sirone praecepta
Epicuri, cuius doctrinae socium habuit Varium. On Vergil's Epicureanism, see above, n. 11.
23. Si licet, hoc sine fraude, Vari dulcissime, dicam: \dispereatn, nisi me perdidit iste
I Sin autem praecepta vetant me dicere, sane \ non dicam, sed "me perdidit iste puer." Cf. Tait 59,
Westendorp-Boerma (above, n. 11) 138 ff. here = the son of Aphrodite, as in Phil. 8,
so that the peculiarly Hellenistic wit of the poem now lies in the fact that
puer is only ostensibly a Latin synonym for Pathos, who is usually imagined as a child; what it
really does is transfer attention from the deified abstraction Desire to the more concrete subject,
the specific boy who is driving Vergil to distraction. Reserving the point until the last word is
typical of Hellenistic epigram in general, and of many of Philodemos' in particular.
24. According to G. Cavallo (above, n. 5), 46, both this papyrus and P.Herc. 1082—each
representing separate works of On Virtues and Vices—were written by the same scribe, whom
he designates Anonimo XXV.
25. There remains the possibility that Varus here is Alfenus Varus; cf. Nisbet-Hubbard on
Hor. O. 1.24. On Siro = Silenus, cf. also schol. Veron. ad E. 6.10 (= Siro F 8 Gigante).
26. Donatus V.Verg. 37, Servius V. Verg. 29-11, Hieron. Chron. 166el4 (17 B.C.). Cf.
F. Delia Corte, "Vario e Tucca in Filodemo," Aegyptus 49 (1969) 85-88; repr. in Opuscula 3
(Geneva 1973) 149-152; and again cf. Gigante, "Atakta III" (above, n. 18) for a more detailed
view of modern scholarship. For Tucca and Varius' role in the editing of the Aeneid, cf. Cova
(above, n. 20), H. D. Jocelyn, Sileno 16 (1990) 263-285.
 On the Bay of Naples 21

entioned by Horace in theArs Poetica21 and in 0.1.24.5-12, where Horace speaks

of Vergil's grief for the recently deceased Quintilius, the restoration of Horace's
name in the two papyri could not be ruled out. It is, moreover, Horace himself who
puts himself in the circle of Vergil, Varius, and Plotius.28 The matter has finally
been settled with the publication of P.Here. Paris. 2, seemingly, like P.Here. 1082
a fragment of Philodemos' ,29 which once again, this time leaving
no doubt as to restoration, lists the four Roman addressees:

21

Why did Philodemos address this particular group? A reasonable guess would
be that these four Romans and Philodemos shared a common interest not only in
poetry and philosophy but more particularly in the relationship between poetry and
philosophy (as we know to be the case with Philodemos). Perhaps we can answer a
little more specifically on the basis of Horace, Ars Poetica 438-444, where Quintilius
is described as the true friend who will not flatter a friend's poetry but rather will
offer only honest criticism.

Quintilio siquid hectares, "corrige sodes

hoc" aiebat "et hoc." melius te posse negares
bis terque expertum frustra, delere iubebat
et male tornatos incudi referre versus.
si defendere delictum quam vertere malles,
nullum ultra verbum aut operam insumebat inanem,
quin sine rivali teque et tua solus amares.

Brink ad loc. (p. 412) says that he knows of "no evidence attesting the conjunction
of friendship and criticism in extant Hellenistic writing on literary theory as it is
attested in Horace and seems to be attested in Lucilius" (cf. pp. 400 f.). Conceiv-
ably, however, Philodemos in a work on flattery could be making the same point as
Horace, namely that people truly interested in philosophy would never flatter their
friends. Just as Good King was written with its dedicatee in mind, so too could
Philodemos have exemplified his general points on flattery with a reference to the
proper way for friends who are poets to speak of each other's works: Offer frank
criticism and be willing to receive it in turn (or risk facing Quintilius' silence, which
is further criticism of you as a poet).

27. 43 8 ff. Quintilio siquid recitares, "corrige sodes I hoc" aiebat, "et hoc"; see further below.
28. Plotius et Varius Sinuessae Vergiliusque I occurrunt, animae qualis neque candidiores I
terra tulit, neque quis me sit devinctior alter (Sat. 1.5.40—43). Cf. Sat. 1.10.81 Plotius et Varius
Maecenas Vergiliusque. And note Hieron. Chron. 166el4 (17 B.C.) Varius et Tucca Vergilii et
Horatii contubernales.
29. This work, part of Vices and their Corresponding Virtues, is dated to the middle of the
century by Cavallo (above, n.5) 41, 54 f. Cf. Capasso, Manuale (above, n. 2) 175 f.
30. Cf. M. Gigante and M. Capasso, "II ritorno di Virgilio a Ercolano," SIFC1 (1989) 3-6.
22 Introduction

Another papyrus fragment, untitled, also tantalizingly attests to Philodemos'

life in Campania: P.Herc. 312 = T 15:

... he decided to return

with us to
Naples and to
dearest Siro and his
way of life there
and to engage in active philo-
sophical discourse and to live
31
with others in Herculaneum

For all the uncertainties of the papyri, then, we should be more than satisfied with
what little solid evidence they do supply to supplement the previously known data
derived from scholia, lives of Vergil, and the autobiographical material in Horace and
the Catalepton, all of which uniformly agree on placing a small group of Augustan
literati in a Neapolitan setting which was thoroughly Epicurean. Nor should Horace
be excluded from possible membership; he mentions Philodemos and his circle even
if they do not return the favor. And if B. Frischer is right, a further link between
Philodemos and Horace is to be found in the father among the dedicatees of the latter's
Ars Poetica, a Piso whom Frischer identifies as L. Calpurnius Piso Caesoninus.32
It may even be possible to detect an Epicurean flavor in the choice of Catalepton
as title for the collection of epigrams (and perhaps for the Priapeia as well),33 a word
whose Greek original is disputed, it having been transliterated back into Greek as
The meaning would thus
be something like "leftovers" (from the unlikely KmdXeuiTOv), "fine" (i.e., "ele-
gant"), or "small" (i.e., "modest"). Of these, E. Reitzenstein favors the second, find-
ing in this sense a reference to the Alexandrian ideal of elegance.34 But since the
collection as a whole is not markedly Alexandrian in tone, it is probably better to
side with the majority of scholars (Westendorp Boerma included) who translate the
title in the last-named sense as "Schnitzel" (Ribbeck), "trifles" (Duff), and the like.
This too is understood in an Alexandrian sense, i.e., with reference to the limite

31. For the text as here printed, cf. Gigante, Virgilio e la Campania (above, n. 11) 75 f.;
id., "I frammenti di Sirone" (above, n.ll) 178-180 (F 1), who reports that Cavallo would place
the papyrus ca. 50 B.C. Obviously other conjectures are possible (cf., e.g., Cronert, K&M [above
n. 11] 125-127), but the general sense is clear. I would, though, not go so far as Gigante in sug-
gesting that the subject of e8oKEi is Vergil. See further on T 15.
32. B. Frischer, ShiftingParadigms: New Approaches to Horace's Ars Poetica (Atlanta 1991)
52-68. Frischer deals with problems of chronology and family relationship among the Pisones,
but I suspect that agreement will be given only grudgingly; cf. above, p. 11, n. 32. On the rela-
tionship between Horace and Philodemos, cf. Tail 64—76; M. Gigante, "Cercida, Filodemo e
Orazio," in RP 235-243; Summers (above, n. 13), esp. ch.2; and the articles in the bibliograph
by Cataudella, De Witt, Delia Corte, Hendrickson, Michels, Munoz Valle, Reitzenstein, and
Wright.
33. Cf. Westendorp Boerma (above, n. 11) xx—xxiv, for further details.
34. "Die Entwicklung des Wortes . Zur Stilbezeichnung der Alexandrincr,
Vest serift R. Reitzenstein (Leipzig 1931) 25—31.
 On the Bay of Naples 23

scope demanded of poetry by Callimachos and Aratus. What should probably be

added is that this sense fits in very well with the Epicurean ideal of moderation in
all things (which is not of course exclusive to Epicureanism).
In addition to Vergil, Varius, Quintilius, and Tucca (and perhaps Horace), there
were other poets to be found in Philodemos' company. The otherwise unknown
Antigenes and Bakkhios of Epigram 29 were probably poets, or at least occasional
tossers-off of epigrams (see the commentary). Catullus too may have made
Philodemos' acquaintance, for in two poems he refers to two friends on the staff of
a Piso who may well be Philodemos' patron.35 In one, no. 28, we hear that Fabullus
and Veranius were unhappy with the profit they made while on Piso's provincial
staff. The other, no. 47 (T 11), berates Porcius and Socration for receiving more
favorable treatment from Piso than do Fabullus and Veranius:

Porci et Socration, duae sinistrae

Pisonis, scabies famesque mundi,
vos Veraniolo meo et Fabullo
verpus praeposuit Priapus ille?
vos convivia lauta sumptuose
de die facitis, mei sodales
quaerunt in trivio vocationes?

G. Friedrich identified Socration with Philodemos, partly on the questionable

grounds that he, like Fabullus and Veranius, had accompanied Piso to Macedonia
(see above, "Life").36 Nonetheless, the identification is an attractive one, although
it has not been universally accepted:37 Although the attested name could
conceivably appear in Latin as Socration rather than the expected Socratio, Socration
is more likely to be the transliteration of which is not otherwise attested,

35. For the relationship between Catullus and Philodemos, cf. Tail 36^47, L. Landolfi,
"Tracce filodemee di estetica e di epigrammatica simpotica in Catullo," CErc 12 (1982) 137-
143.
36. Catulli Veronensis Liber (Leipzig 1908) 228: "Wir haben nach dem Wortlauf unseres
Gedichtes keinen Grund, uns den Socration anders vorzustellen als den Philodemus von Gadara,
der auch bei Piso in Macedonien war." Another link, albeit tenuous, between Socration and
Philodemos is Catullus' choice of Fabullus for the addressee of his invitation poem (13), which
is almost certainly modelled on Philodemos' invitation to Piso (27).
37. Accepting it are T. Frank, Catullus and Horace (New York 1928) 82-84; C. L. Neudling,
A Prosopography to Catullus (Oxford 1955) 147. Rejecting the identification are P. Giuffrida,
L'epicureismonellaletteraturalatinanellsec. a.C. (Turin 1950) 2.179 f., F. Delia Corte,Personaggi
Catulliani (Florence 1976) 204-208. Unsure are Quinn ad loc. and Nisbet (above, n. 10) 183.
The question is further complicated in that not all agree that the Piso of Cat. 47 is Philodemus',
because in two other poems of Catullus Fabullus and Veranius are said to have been in Spain
together (9 and 12) and if the four poems form a tight chronological cycle the Piso can be Cn.
Piso, who was quaestor pro praetors in Hispania Citerior in 65764 B.C. (Sallust Cat. 18.4—19.5) or
L. Piso Frugi. The whole problem is nicely analyzed by Tait 39^12, who shows that "there seems
to be a decided difference of tone between the references to the Spanish jour-ney and the refer-
ences connecting Catullus' two friends with Piso" (41). R. Syme, "Piso and Veranius in Catullus,
C&M 17 (1956) 129-134 =Roman Papers 1 (Oxford 1979) 300-304, ignoring differences of tone
between the Spanish poems and the Piso poems, argues that after his praetorship Caesoninus
24 Introduction

probably because it less a real name than a diminutive for Sokrates.38 Anyone called
Little Sokrates is presumably a philosopher; and just as Cicero singles out
Philodemos as the Greek philosopher with the greatest influence on Piso, so too
would Catullus be more likely to address him than any other, less well known,
philosopher. That he calls him by a nickname should cause no surprise; Siro may
have been called Silenos and Vergil was known by his Neapolitan friends as
Parthenias. (See the commentary to v. 19.) Why he would call an Epicurean after
Sokrates, especially when Epicureans in general and Philodemos in particular made
no attempt to disguise their dislike of Sokrates, will be addressed in the next section.
On the basis of these slight links, we are free to imagine, without insisting on
it, that Catullus and Philodemos were aquainted with one another. There is almost
nothing, however, to connect Philodemos with the most famous of Epicurean poets,
Lucretius, about whose life we know so little, especially now that the dedicatees of
Philodemos' Rhetorica and DRN can no longer be assumed to be the same.39 It is,
though, worth noting that traces of what was surely a complete DRN (mentioned
by Cicero in 54 B.C.) have been found in the Herculaneum papyri (see above, n. 2),
but this does not attest to any personal relationship between the two poets; for all
we know the manuscript may even have been added to the library after Philodemos'
death; the paleography suggests a date as late as the end of the first century B.C.
The many passages in one philosopher-poet which sheds light on the other are of
course due to their common dependence on and adherence to Epicurus.40

Philodemos and the Epigram

I
In his prose Philodemos openly declares his debt as to Epicurus;1 his
debt as (his word, see below) is more diffuse. It is, however, not

very likely served in Hispania Citeriot. In either case, Catullus' Piso is Caesoninus. Still rejecting
this identification is Wiseman, Catullus (above, n. 14) 2; accepting it is Frischer, Paradigms (above
n. 32)57.
38. appears only in Galen, Comp. Med. 12.835 Kiihn. Socratio appears only in
GIL 3.948. Neue-Wagener, formenlebre der lat. Sprache (3rd ed., Leipzig 1902) 1.246 ff., list
some few Latin names ending in -on deriving from Greek names in , but all names in
show up in Latin as -zo; cf. Nisbet, op. cit., (above n. 10) 182.
39. See above, p. 8, n. 19.
40. Cf. G. Barra, "Osservazioni sulla 'poetica' di Filodemo e di Lucrezio," Annali d. Fac. d
Lett. eFilos. d. U. di Napoli 20 (1977-1978) 87-104; D. Armstrong, "The impossibility of metathe-
sis: Philodemus and Lucretius on form and content in poetry," in D. Obbink (ed.), Philodemus and
Poetry (New York 1995) 210-232; D. Clay, Lucretius and Epicurus (Ithaca 1983) 24 f. (who briefly,
p. 291 n. 57, lists some who maintained that Lucretius knew Philodemos).

1. Most notably ir fr.45.8-11 Olivieri

"The basic and most important [principle]
 Philodemos and the Epigram 25

difficult to see that he owes most to his immediate predecessors in the genre, the
authors of Meleager's Garland in general, and Asklepiades, Kallimachos, and his
fellow Gadarene Meleager himself in particular—that is, the Hellenistic pioneers
who transformed classical epigram and merged it with elegy.2 Originally, ETtiypauua
(and related words) indicated nothing more—or less—than words, whether in verse
or not, written on a stele or other object, and was probably to be distinguished from
words which came directly from a living speaker, a bard or rhapsode in the case
of poetry.3 In time, however, the word came to be associated almost exclusively with
verse inscriptions, and even more particularly (but at first not exclusively) with
the dactylic-pentameter distich, the elegiac couplet.4 Although this also came to
be the distinctive verse scheme of elegy, modern scholarship finds it easy to keep
the two distinct during the early and high classical period: Epigrams are short
poems5 written down for public display to memorialize victories, temple offerings,
and the dead; elegies, not limited physically by the nature of stonecarving, tend
to be longer poems composed for oral presentation on a particular occasion: one

is that we will obey Epicurus, according to whom we have chosen to live" (trans. Asmis). The
words are frequent synonyms in the Herculaneum texts. In gen-
eral, cf. M. Erler, "Philologia medicans: Wie die Epikureer die Texte ihres Meisters lasen," in
W. Kullmann and J. Althoff (eds.), Vermittlung und Tradierung von Wissen in der griechischen
Kultur (Tubingen 1993) 281-303.
2. Useful discussions of early and Hellenistic epigrams are: H. Beckl>y,Anthologia Graeca,
2d ed. (Munich 1967) 12-67; Alan Cameron, The Greek Anthology from Meleager to Planudes
(Oxford 1993) ch. 1; E. Degani, "L'epigramma," in F. Adorno et al. (eds.), La cultura ellenistica
(Milan 1977) 266—299; id., "L'epigramma," inG. Cambianoet al. (eds.),Lo spazio letterariodella
Grecia antica 1.2 (Rome 1993) 197-233; L'Epigramme grecque = EH 14 (1968); P. M. Eraser,
Ptolemaic Alexandria (Oxford 1972) 1.553-617; G. O. Hutchinson, Hellenistic Poetry (Oxford
1988) 20-24,72-76,264-276; A. Lesky,Hist. Gk. Lit. (New York 1966) 737-743; R. Reitzenstein,
Epigramm und Skolion (Giessen 1893); id. "Epigramm," RE 6 (1907) 71—111; K. Gutzwiller,
Poetic Garlands: Hellenistic Epigrams in Context (Berkeley 1997). More specialized studies will
be found in the bibliography.
3. The word first appears in our texts in Thuc. 6.59.3 (= Simon. 26(a) FGE),
used of a four-line grave inscription in elegiac couplets. But even as late as Herodas an identify-
ing mark on a forehead can be called an (5.79). And cf. Hesych.
P. A. Hansen, "DAA 374-375 and the early elegiac epigram," Glottal 56 (1978) 195-
201, dates the beginning of the popularity of elegiac inscriptions to ca. 560 B.C. Cf. M. B. Wallace,
"The metres of early Greek epigrams," in D. E. Gerber (ed.), Greek Poetry and Philosophy: Studies
. . . L. Woodbury (Chico 1984) 303-317; H. Hommel, "Der Ursprung des Epigramms," KM 88
(1939) 193-206.
4 . O n l y Aelian, VH 1.17 uses the phrase elsewhere the one word
can have this meaning; cf. e.g., Kritias 88 B 4.3 DK, Arist. Poet. 1447bl2, K. J. Dover,
"The poetry of Archilochus," EH 10 (1964) 187 f. For Latin, note Hor./lP 75 versibus impariter
iunctis and Ovid's more elaborate periphrases, Am. 2.17.21 f.,AA 1.264, Pont. 4.16.11,36, Trist.
2.220. The word in the singular can also mean (i) the pentameter line alone; e.g., Arist.
Quint. 1.24; or (ii) an elegiac poem of more than one distich: Ion of Samos (1 D) = Hansen CEG
2.819.9-13. In the plural, can mean hexameters: CEG 2.888.19, [Hdt.] V.Hom. 36.
5. CEG 1 (ca. 475 poems) contains only two 10 vv. or longer; CEG 2 (ca. 600 poems)
contains eight. Note the view of J. W. Day, "Rituals in stone: Early Greek grave epigrams and
monuments," JHS 109 (1989) 16-28, that the poetry of epigrams was originally recited at the
grave site. This dissolves somewhat the barrier set up above between epigraphic texts and ele-
gies, but the general point remains valid.
26 Introduction

man6 speaking in his own voice (however artfully fashioned), whereas early epigram
was anonymous, avoiding all reference to author.7 The Greeks themselves, how-
ever, did not always maintain so nice a distinction: Lykourgos 142, e.g., refers to
sepulchral inscriptions as ; and [Dem.] 59.98 says
of a single distich.8 Any poem comprising one or more elegaic
couplets could thus be summed up as (neut. pi.). Usage, however, reserved
the term (fern, sg.) (first in Arist. Ath. Pol. 5.2) for the poems of Solon,
Mimnermos, Xenophanes, Tyrtaios, et al. By the beginning of the Hellenistic pe-
riod, lengthy (say, longer than twenty-line) elegies continue to be written, and the
term continues to be used with the same latitude as in the classical period.9
Once epigrams were liberated from their stone prisons, however, they were also
free to increase in size; and as they took on new topoi (inprimis erotic and sympotic),
hitherto within the province of elegy (and skolion), whatever line there was between
elegy and epigram should now be regarded as either nonexistent or insignificant.10
But having won the freedom to extend the epigram, Meleager, Kallimachos, and
others, observing the general Hellenistic love of brevity, soon voluntarily imposed

6. "Kleoboulina" and Sappho are the only names of early female elegists to have come down
to us (cf. West IEG 2 for the evidence), but Theogn. 579-582 and 861-864 are written either by
one or two women (so M. L. West, Studies in Greek Elegy and Iambus [Berlin 1974] 156, 160)
or by a man taking a female voice (so Bowie [below, n. 7] 16). And who, man or woman, wrote
Theogn. 257—260, spoken in the persona of a mare? See van Groningen ad loc. for suggested
answers. Moiro (iv-iii cent.) also wrote elegies: frr. 2-3, 6 Powell.
7. There are, e.g., no sphragides in early inscriptional epigram. On elegy, cf. West, Studies
(above, n. 6) 1-21; E. L. Bowie, "Early Greek elegy, symposium, and public festival," JHS 106
(1986) 13-35. West 10-13 lists and discusses the various occasions where elegies were typically
presented (but see Bowie 15—21). On the several nonauthorial voices of early epigrams, cf.
J. Svenbro, Phrasikleia (Paris 1988; Eng. tr. Ithaca 1993), ch. 2. Note that Simon. Elegy 25 W2
(6 w.) and Soph. Elegy 4 W2 (4 w.) are treated as elegies solely because they are personal poems
delivered by pre-Alexandrian authors; if the same stories embodying these poems were told of
Hellenistic authors, they would without question be classified as epigrams.
8. Cf. B. Gentili, "Epigramma ed elegia," L'Epigramme grecque (above, n. 2) 37—90, esp.
40 f.; West.-SWz'ej (above, n. 6) 3 f.; C. M. J. Sicking, Griechische Verslehre (Munich 1993) 83-86.
9. Parthenios, e.g., credited with elegies by the Souda (= SH 605) wrote longer poems and
no short "epigrams." On the other hand, Macrobius 5.20.8 (= SH 100) says that an epigram of
Aratos comes from his liber elegion, i.e., (AP 11.437 = 2 HE). W. Ludwig, "Aratos,"
RE Suppl. 10 (1965) 29 f., says that Macrobius confuses epigram with elegy, but his usage is
consistent with classical authors; see above, n. 8. Similarly, Stephanos Byz. 171.3 Meineke
(= SH 667) calls Phaidimos, whose epigrams appeared in Meleager's Garland, an elegist. For
the view expressed above on the nature of early elegy and epigram, cf. Day (above, n. 5) and
A. E. Raubitschek, "Das Denkmal-Epigramm," L'Epigramme grecque (above, n. 2) 1—36.
10. For this view, here argued briefly and schematically, see Gentili (above, n. 8) and
G. Giangrande, "Sympotic literature and epigram," L'Epigramme grecque (above, n. 2) 37-90,
91-177, both of whom demonstrate several points of contact between early epigram and elegy.
Some scholars, none the less, attempt to maintain a division between the two. For example, Fraser
(above, n. 2) 1.668, on the basis of subject matter, classifies P.Petric II 49(a) = SII 961, at least
26 lines long, as an epigram, which Gow-Page, HE 2.483 expressly and curtly deny. Other works
variously classified as cither elegy or epigram: Kallim. Epigr. 54 HE (AP 7.89; 16 w., said to
come from an elegy by Diog. Laert. 1.79). Asklepiades 16 HE (AP 12.50), Leonidas 85 HE (AP
10.1); cf. Lesky (above, n. 2) 738. See also P. Kagi, Nachwirkung der dlteren griechischen Elegie
in den Epigrammen der Anthologie (Zurich 1917).
 Philodemos and the Epigram 27

upper limits. Meleager is most influential in this regard in that he seems to have shown
a marked preference for the shorter epigrams when assembling his Garland. The
next generation of epigrammatists, although not of course limited to Meleager for
their knowledge of the genre, could not help but be influenced by his choices in
their own compositions. Indeed, Philip's term for epigrams in his introduction is
(Philip 1 =AP 4.2.6). Note, too, Parmenion 11 GP(AP 9.342.1 f.)
11
In HE there are only twenty-
one poems longer than 10 lines, only six longer than 14; in GP only Philip's intro-
ductory epigram to his collection is longer than 10 lines (14, with which contrast the
58 lines of Meleager's prologue). This marked disparity between the Garlands not
only shows the influence which Meleager's preference for shorter epigrams had on
his successors, it also proves that the Byzantine editor Kephalas was not the one re-
sponsible for the rarity of longer epigrams in the Anthology.
This, then, is what Philodemos inherited: the short poem in elegac couplets,
whose subject matter comprised the topoi of earlier epigraphic exemplars and those
of the longer elegies. Philodemos also adopted the early Hellenistic taste, not of
course limited to epigrams, for point or wit, which would be all the more appreci-
ated if it could be reserved for the poem's last word.12
Hellenistic epigram is also comparable to classical sympotic elegy, not only in
content, as Giangrande has amply shown, but also of performance (cf. West 11 f.)
Epigrams were a regular accompaniment to dinner parties, although it may be
doubted whether they, like elegies, were sung to the flutes and harps Philodemos
and other epigrammatists mention; see esp. Phil. Epigram 6, West 13 f. Especially
noteworthy is the apparently improvisatory nature of elegy and epigram. For elegy,
note Athen. 125a- (Sim. 25 W2);
similarly, Athen. 656c (Sim. 26). Ad libitum compositions, furthermore, are implicit
in the nature of the common practice in symposia for poetic challenges to pass
around the company.13 In the Hellenistic age, we hear of Antipater of Sidon's

This problem of classification is by no means limited to the Hellenistic age; note West's
classification "Incertum an ex epigrammatis" in his Simonides section oflEG, one poem of which
was moved into elegy in West's second edition on the basis of its now appearing in a papyrus of
Simonides' elegies (P.Oxy. 3965 fr.5 = 16 W 2 ). Later librarians, as demonstrated by the Souda
article on Simonides, distinguished between his elegies and epigrams. Epigrams could now be as
long as the author wished. Most notably (and admittedly exceptionally), Meleager's introduc-
tion to his collection is 58 lines long. One of the new Posidippos epigrams, not yet published, is
14 lines long, another is 12; cf. G. Bastianini and C. Gallazzi, "II poeta ritrovato," Riv. "Ca de
Sass," n.121, March 1993; Cameron (above, n. 2) 400.
11. Later expressions of this motif: AP 6.327.2 (Leonidas Alex. 6 FGE) and 9.369 (Kyrillos).
Cf. Cameron (above, n. 2) 13.
12. Hutchinson (above, n. 2) 21; G. Luck, "Witz und Sentiment im griechischen
Epigramm," L'Epigramme grecque (above n. 2) 387^111. Much of G. Giangrande's work on the
Greek epigram has been dedicated to elucidating the peculiarly Hellenistic point at the close;
i.e., the end, often the last word alone, not only provides a neat closure but also may cast what
has preceded in a new light.
13. Cf. West (above, n. 6) 16 f., Gentili (above, n. 8) 40-43. Skolia, in contrast, were to be
taken up only by the best of the company, and were not responsatory by the company at large;
Ilesych. s.v. , discussed by Reitzenstein (above, n. 2) 3 ff.
28 Introduction

improvisitory ability;14 and of Archias, Philodemos' contemporary and fellow Greek

sojourner in Italy, Cicero tells us that he would improvise on an event of that day,
after which, acting like a one-man company of skoliasts, he would produce yet
another epigram on the same event.15 Many of Philodemos' epigrams are designed
to give the appearance of having been composed in just such surroundings. Even
the erotic ones, ostensibly addressed in private to one woman, could well have been
composed for a larger, komiastic or sympotic, company. But it is a fool's game to
try to create the circumstances of composition for each poem, for as Cicero indi-
cates (see last note), the published poems, even if some began as ad lib extempori-
zations, would be polished before publication to the best of Philodemos' ability.16
And publish them he certainly did, again following his immediate predecessors such
as Poseidippos, Kallimachos, and Theokritos (cf. Fraser 1.607 f.), although there is
nothing known of the circumstances of publication. Cicero, it is clear, as well as
the several Latin poets who imitated Philodemos, had access to some sort of col-
lection, perhaps copies made for limited circulation, although Cicero attests to their
popularity, presumably in Rome.17 Philip and the compiler of the Oxyrhynchus
incipits (see below, pp. 53,203 f. for text and commentary) may well have had avail-
able a comprehensive volume designed for the book market.
But if the details of publication can never be recovered, it is still possible to
infer that Philodemos was ashamed neither of having written poetry in general nor
of having limited himself to epigrams in particular.18 How does the epigram fit into
Philodemos' overall scheme of poetry? We can begin to answer this question by
noting that at least once in his theoretical writings, whose examples for the most
part are drawn from epic, lyric, and drama, he found the opportunity to make
mention of "writers of epigrams."

37

14. Cicero,DeOr. 3 .194: Quod si Antipater Me Sidonius, quemtuprobe, Catule, meministi,

solitus est versus hexametros aliosque variis modis atque numeris fundere ex tempore, tantumque
hominis ingeniosi ac memoris valuit exercitatio ut cum se mente ac voluntate coniecisset in versum
verba sequerentur, quanta idfacilius in oratione exercitatione et consuetudine adhibita consequemur.
15. Cic. Pro Arch. 8.18 quotiens ego hunc vidi, cum litteram scripsisset nullam, magnum
numerum optimorum versuum de Us ipsis rebus, quae turn agerentur, dicere ex tempore! quotiens
revocatum eandem rem dicere commutatis verbis atque sententiis! It is worth pointing out that
Cicero immediately goes on to distinguish Archias' extemporaneous verse from his more pol-
ished published work: Quae vero accurate cogitateque scripsisset, ea sic vidiprobari, ut ad veterum
scriptorum laudem perveniret. Epigrams, that is, were encountered either as dinnertime diver-
timenti, or in published form.
16. Even Epigram 22, which Philodemos presents in intentionally unpolished form; see
below.
17. In Pis. 71 (= T 2) multa a multis et lecta et audita.
18. There is at any rate no indication that Philodemos wrote poems in any other genre than
the epigram. Not everybody was so restrained: An incomplete list of Hellenistic poets who com-
posed epigrams in addition to other genres includes Kallimachos, Theokritos, Apollonios Rhodios,
Meleager, Rhianos, Asklepiades, Hedylos, Diotimos of Adramyttium, Anyte.
 Philodemos and the Epigram 29

38

6 Hammerstaedt, quod si cum Gigante scriptum putes fit versus longior 17-21
interpunxit Hammerstaedt 29 Gigante:Jensen:
Mangoni 34-1 Pace 1 D y c k

But if someone said that the virtue of a poet is to be able to compose every
poem <i.e., every poem he does compose, whether in only one genre or
in many> beautifully, there would be agreement as to what we are seek-
ing. For when we investigate who is an excellent poet, we in effect judge
how he composes beautifully the poems he composes; but he [sc.
Philodemos' opponent] says <simply> that it is the one "who composes
beautifully" <who is the excellent poet>.
But if he further postulates <that it is a poet who composes> every genre
of poem <beautifully>, he abandons virtue <of a poem> to be not only
unrealized—for nobody has been able to compose every poem beauti-
fully—but (as I think) also impossible>—for no one could <compose
every poem beautifully>. Besides, not even in a single genre has any poet
maintained an even level.
But if <someone says that the excellence of a poet is> to be able to com-
pose poetry containing <poetic> excellence, it would be less strange;

19. Phil. Poem. 5 coll. 37.2-38.15 Mangoni. For the text (and for a more complete appara-
tus) see now C. Mangoni, Filodemo: Ilquinto libra della Poetics, La Scuola di Epicuro 14 (Naples
1993), which supersedes Jensen's (column numbers to Book 5 given below are those of Mangoni,
which are three higher than Jensen's); M. Gigante, "Filodemo e repigramma," CErc22 (1992) 5-
8. Note: (i) Philodemos' term for a writer of epigrams also occurs in D.L. 6.14 (etc.), IG 92 (1). 17 A24
(m'B.c.;usedofPoseidippos), Eustathios on II. 1.439.28 (etc.), V.Hom. Plut.84. (Other terms are
found only as epigrammatista in Apollinaris Sidonius, and )
(ii) The Theognis referred to is probably the elegist, who was known for his commonplaces, but
Gigante 7 f. argues for Theognis of Athens, a minor tragedian (TrGF 128).
For comments on the text and translation I am grateful to Elizabeth Asmis, Andrew Dyck,
Marcello Gigante, Sander Goldberg, Jiirgen Hammerstaedt, Nicola Pace, and Michael Wigodsky
In the translation which follows, angle brackets contain words which fill out the thought of
Philodemos' typically compressed Greek.
30 Introduction

but first we will have to know what the excellence of poetry is, and when
this excellence has been realized, it will be obvious that the one com-
posing it is excellent, and we might say that this is the excellence of a
perfect poet, although excellence is not generally <i.e., uniformly>
present.... It will necessarily be the case, according to this theory, that
we fail to recognize whether those who have written thus <i.e., beauti-
ful poetry> possessed the virtue of a poet. If "poetry" belongs equally
to the epigrammatists and to Sappho, that man will be saying that to be
the composer of beautiful poems is the same as to be a good poet—which
we knew well "since before Theognis was born."

This appears shortly before the end of Book 5, and hence close to the end of the
entire work.20 Philodemos' immediate point is that poetry, which must be
complete—as opposed to a 7toir|ua, which can be a portion of a (col. 14.26-
36)—can be instantiated by a short work such as a composition of Sappho or an
epigram.21 But if Philodemos had not specifically mentioned the epigram in all the
preceding five books,22 his purpose in doing so now might be to inform his readers
where, according to the theories presented, the epigram, the particular genre pro-
duced by the author of the work they are reading, belongs in the grand scheme of
all poetry. Thus, even without a review of the entirety of On Poems,2) it can be seen
that a more personal, less theoretical, analysis of this passage is possible.
Although Philodemos mentions Sappho only two other times in the extant
papyri,24 her name here likely stands for the very best of poets, just as it does in
Epigram 12, where Flora is excused for being unable to sing Sappho's lyrics. The
pair of epigrammatists and Sappho, therefore, not only exemplifies short poetry, it

20. This may be inferred from Philodemos' words as he begins the final section of Book 5:
(col. 29.21-23). Cf. R. Janko, "Reconstructing
Philodemus' On Poems," in D. Obbink (ed.), Philodemus andPoetry (New York, Oxford 1995)
185f., who allows for a very small possibility that there were more than five books.
21. For Philodemos'use of and in On Poems, see Asmis on Neoptolemos
of Parion (below, n. 23), esp. 210 f. The fragments of Neoptolemos have been collected by H. J.
Mette, "Neoptolemos von Parion," KM 123 (1980) 1-24.
22. Since Neoptolemos wrote both epigrams (collected in Powell, Collectanea Alexandrina)
and a work (fr. 7 Mette = Athen. 10.81), Philodemos may have responded
directly to his views on this genre earlier.
23. For a overview of the papyri which can with some degree of certainty be assigned to
On Poems, cf. F. Sbordone, Suipapiri delta Poetica di Filodemo (Naples 1983) 7-43; R. Janko
(above, n. 20). Also useful are E. Asmis, "Philodemus' Epicureanism," ANRW 2.36.4 (1990)
2403-2406; and M. Gigante, Philodemus in Italy, trans. D. Obbink (Ann Arbor 1995) 36-38.
For more detailed studies of Book 5, cf. Mangoni's introduction and the important series of
articles by Asmis: "The poetic theory of the Stoic 'Aristo'," Apeiron 23 (1990) 147-201; "Crates
on poetic criticism, "Phoenix 46 (1992) 138—169; "Neoptolemus and the classification of poetry,"
CP87 (1992)206-231; "An Epicurean survey of poetic theories (Philodemus On Poems 5, cols.
26-36); CQ, N.S. 42 (1992) 395^15.
24. Once only to quote a short passage (De Piet. p. 42 Gomperz = In cert. Auctor 23 Voigt),
the second time to state (De Poematis Tr.B fr. 20 col. i S. 10-11).
 Philodemos and the Epigram 31

also, I suspect, can be taken to span the qualitative limits within this range: Sappho,
one of the greatest of lyric poets at one end; the ad-hoc ad-libitum epigram at the
other. Philodemos' argument that ^oirion; encompasses Homeric epic (as he says
in On Poems 5 col. 14.31-33) as well as the shorter poems of Sappho thus carries
epigram along in its wake. Since he describes as a poem with a continu-
ously woven theme and meaning—that is, a complete composition—he clearly
25
implies that even a two- or four-line epigram can qualify as and that all,
or almost all, that he says about poetry in general applies to the epigram in particular.
Poetry, for example, as he says more than once, insofar as it is poetry, does not
benefit its readers.26 It is not that a poem cannot contain useful facts or a valid
argument; only that these function entirely apart from any poetic virtue contained
therein. "No one derives a benefit through either medicine or wisdom or many other
kinds of knowledge by attaining the extreme together with poetic elaboration" (col.
4.24-31, tr. Asmis). With "extreme," understand "of poetic virtue" (Asmis hesi-
tantly suggests "of utility"). The emphasis is on "with." A poem, however, may be
a good one even if the poet is wrong on the facts or if his argument or morality is
questionable. Cf. col. 5.6-18 "If there is a narration without benefit, nothing pre-
vents a poet from knowing these things and presenting them poetically without
benefiting us at all. He [sc. Philodemos' opponent of the moment], though, oddly
burdens the good poet with exact knowledge of (all) the dialects, although the
choices he (the poet) makes are quite acceptable." And for an expression of th
belief that bad men can produce good literature, cf. Rhet. col. 21.12-15 (2.226
Sudhaus)
27

But if Philodemos absolves the poet of the need to instruct us, he is equally
sure that no good poem can be free of some thought or ideas. His chief opponent
here is Krates of Mallos, whose ideas on the primacy of euphony Philodemos finds
largely objectionable. The excellence of a poem, therefore, lies in its artistic merg-
ing of thought (which need be neither true nor beneficial) and the standard ele-
ments of poetry, i.e., composition, diction, and (to a lesser extent) euphony. But if
a poem does not benefit, neither does it harm; rather, its diction—or as we might
say now, its persona—represents that of a person who is neither immoral nor a wise

25. Which is implicitly denied, e.g., by Varro, Menipp. fr. 398 Epoema est lexis enrythmos,
id est verba plura modice in quandam coniectaformam. itaque etiam distichon epigrammation vacant
poema. poesis est perpetuum argumentum ex rhythmis, ut Ilias Homeri et Annalis Enni. Cf. H. Dahl-
mann, Varros Schrift "de poematis" und die hellenistisch-romische Poetik. Ak. d. Wiss. u.d. Li
Mainz: Abh. d. geistes- u. sozialwiss. Kl. (1953.3) 26,29 f., 34 ff.; G. B. Walsh, "Philodemus o
the terminology of Neoptolemus," Mnemosyne, 4th ser. 40 (1987) 56-68, esp. 65 ff.; Asmis,
"Survey" (above, n. 23) 413 f.
26. On Poems5. col 4. 10-31,25.30-34,32.17-19; cf. E. Asmis "Philodemus'poetic theory
and On the Good King according to Homer" CA 10 (1991) 4-13.
27. The immediate application of this statement is to rhetors, but G. M. A. Grubc, The
Greek and Roman Critics (Toronto 1965) 200, argues that it applies to poets as well. Cf. Poem
5 col.17.32—18.7, where Philodemos criticizes a Stoic (Ariston?) for crediting Homer and
Archilochos (or Aischylos?) with only modified poetic excellence on the grounds that their
thought and educational values are improper.
32 Introduction

man: "Poetical goals have been established: For diction, to imitate diction which
teaches us something beneficial in addition (to itself); for thought, to take a middle
ground between that of the wise and that of the vulgar" (col. 26.1-8).28
In addition to the above considerations in the proper assessment of poetry,
another important criterion requires that hearing or reading the poem in question
provide its audience with pleasure of a correct Epicurean sort.29 In brief, as Asmis
ably demonstrates,30 Epicurus, despite what later detractors said of him, was will-
ing to accept poetry, although with reservations. In particular, the wise man could
be trusted to have the proper attitude, able to listen to the recitation of poetry with-
out succumbing to its Sirenic charms or accepting its claims to do anything more
than provide harmless pleasure. Poetry, that is, can be classified in Epicurean terms
as a natural but unnecessary pleasure. As such it was allowed a place at the ban-
quets attended by Epicureans, where, at least originally, it was listened to but not
subjected to immediate literary criticism, which would detract from the pleasure.31
Presumably, almost any poetry could be recited at these banquets, but, in keep-
ing with Epicurus' dictum that the wise man will not exercize himself overmuch
with the composition of poetry, original compositions would have at least to give
the appearance of not having required any effort. Epigrams meet this requirement
as no other genre (see above, pp. 27 f.). It is thus possible to apply Philodemos'
general view of poetry to the epigram in particular, as the performance of epigrams
at dinner parties (see above) fits perfectly into our picture of the symposia held in
the Epicurean Gardens of Naples and surroundings.32

II

Having gone from poetry in general to epigrams in particular, we must now focus
even further on Philodemos' own epigrams and ask whether they illustrate his views

28. Cf. Asmis, "Crates on poetic criticism" (above, n. 23); ead., "Good King" (above
n. 26) 8—11, esp. 10, "Philodemus's response to Plato is, in turn, indebted to Aristotle, who pro-
posed that tragic characters should be neither outstandingly good nor bad, but
intermediate Like Aristotle, Philodemus demands ordinary human values. Differently
from Aristotle, however, Philodemus clearly distinguishes the "thought" of the poem as a whole,
as presented by the poet, from the thought of the characters."
29. For accounts of Epicurean pleasure, cf. J. Rist, Epicurus: An Introduction (Cambridge
1972) 100-126; P. Mitsis, Epicurus' Ethical Theory (Ithaca 1988) 11-58.
30. E. Asmis, "Epicurean poetics," BACAP1 (1991) 63-93, with my response, ibid. 94-105;
both reprinted in Obbink (above, n. 20) 15—34, 35—41. For the pleasures of poetry in particular,
cf. Asmis, "GoodKing" (above, n. 26) 13-17. See further my "The Epicurean philosopher as
Hellenistic poet," in Obbink 42—57; M. Wigodsky, "The alleged impossibility of philosophical
poetry," ibid. 58-68.
31. Cf. Asmis, "Epicurean poetics" (above, n. 30).
32. Cf. Asmis, "Good King" (above, n. 26) 15 on the intellectual pleasures to be derived
from listening to poems recited at parties; A. Cameron, Callimachus and His Critics (Princeton
1995), ch. 3, "The symposium."
 Life 33

in any special way. Let us begin by simply categorizing more or less as the Anthol-
ogy does the 36 epigrams I consider genuinely or possibly Philodemean.33 In my
numeration they are:

Erotica: By far the largest category, totaling 21 or 22 poems, most coming from AP
5:34 ]_2) 4-26, 36, plus 8 (Book 10), 4, 6,19 (Book 11), 11 and 24 (both from
Book 12). Gow-Page, by omitting the last distich, convert 3 into a love poem;
see the commentary. The erotica, exclusively heterosexual, admit of several
subgroupings (with some overlapping):

(i) Dark-but-comely: 9,12,16,17; cf. ii.27; cf. the commentary to 17.

(ii) Street encounters: 20,21
(iii) Female narrator: 26, 36
(iv) Adulterous love: 15, 25, 26
(v) Impotence: 19, 25
(vi) Marriage/wife: 4-8. These form part of an extended cycle of poems addressed
to or concerning Xanthippe; see below.

Other books of the Anthology contain fewer Philodemean entries.

Dedicatory: 34, 35
Sepulchral: 33
Epideictic: 3, 29
Protreptic: 32. Poems with imperatives usually are assigned here.
Sympotic: 27, 28
Scop tic: 31

Since the early epigrammatists did not write with all these precise terms in mind,
it is not surprising that some epigrams fit uncomfortably into the Anthology's
schema, even were it to be correctly applied throughout. In Philodemos' case, five
epigrams not only resist standard classification, they readily form their own little
group: AP5.112 (5) speaks of love only to turn away from it with a new desire for

33. More will be said on the arrangement of the epigrams below, p. 54. For now it will be
enough merely to outline those books of the Greek Anthology containing classical and Hellenis-
tic epigrams: 4. The proems to the various Garlands; 5* Erotica; 6* Anathematica (dedicatory
7* Sepulchral; 9* Epideictic (Declamatory); 10* Protreptic; 11* Sympotic and satiric (sceptic);
12* Erotica, largely homosexual; 13. In various meters (sc. other than elegiac pentameters); 15.
Miscellaneous; 16* Epigrams from Planudes missing from the Palatine Anthology. (* = con-
taining epigrams of Philodemos) There are of course other ordering schemes, both ancient and
modern; cf., e.g., G. Pfohl, "Die epigrammatische Poesie der Griechen: Entwurf eines Systems
der Ordnung," Helikon 7 (1967) 272—280, who discusses the various ways one can classify in-
scriptional epigrams. P.Mil. Vogl. inv. 1295, the unpublished Poseidippos papyrus, illustrates
ancient arrangements of epigrams; cf. Cameron (above, n. 2) 19, 400.
34. Depending on whether Epigram 24 is erotic or not. Note also that this poem is prob-
ably not homosexual; and that/IP 12.173 (Epigram 11), also gathered with Strata's Musa Puerilis
is certainly heterosexual, as are all the rest of Philodemos' erotica. See on vi.18.
34 Introduction

mature thought. 9.412 (29) and 9.570 (3), "epideictic" only by default, present the
case for avoiding excessive grief at, respectively, the death of friends and the thought
of one's own death. Had they done so less obliquely, they might have shown up
among the protreptica. 11.34 (6) employs two contrasting symposia as metaphors
for contrasting ways of life. 11.41 (4), in obvious parallel with 5 and 6, seeks ane
modus vivendi.
The affinities of this group are clear, even if a name for it is lacking. Each pre-
sents a narrator or main speaker, whom it is easy to see as a mask for Philodemos,
wrestling, sometimes successfully sometimes not, with the excessive passions of love
and the fear of death. I say "Philodemos" because of the evidence adduced below
and in the commentary that the Xanthippe of the poems is to be understood as the
partner, if not wife, of a Epicurean philosopher who can in turn be thought of, as
it seems Philodemos was, as "Sokrates" or "Sokration." Furthermore, even though
epigrammatists often write with no particular person as narrator or even with some-
one clearly not the author as narrator (note especially Philodemos' two epigrams
with female narrators), I believe that unless the author warns the audience, espe-
cially his original listening audience, to look elsewhere, he is willing to accept
being thought of as narrator. Since, however, the "Philodemos" of the epigrams is
designed to overlap only partially with the authorial persona implicit in the prose
treatises, we are not meant automatically to read the epigrams as straightforward
autobiography. Indeed, this disjunction between personae seems rather to warn us
off from regarding the epigrams as factual documents. Even though we, unlike the
original audience, are ignorant of Philodemus' real age, erotic/marital entangle-
ments, and success or failure in adhering to Epicurean standards of behavior, we
can still detect the rift between the serious promoter of Epicurean doctrine in the
prose and the intentionally somewhat comic character of the poetry who needs frank
instruction from another. Poems like this are also more amusing—not an inconsid-
erable point in Hellenistic epigram—when read this way.
Since in the poems listed above, the proper course proposed, however obliquely,
is one espoused by Philodemos in his prose treatises, we may be permitted to refer
to them as Epicurean poems, as long as we recognize, in line with what was said
above, that thought is just one element in Philodemos' idea of the successful poem.
This group, once formed, can attract others epigrams which, although not inap-
propriately placed by the Anthology, display the same affinities.35
The first step in this expansion is to include all the poems mentioning Xanthippe
by name (or by her nicknames Xantharion, Xantho, and Xanthion), since she fig-
ures in two of the philosophical group (3 and 4). This adds 1, 2, 7, and 11 iv.l.
Although some Hellenistic epigrammatists use a woman's name merely as a filler,
not bothering to endow her with an enduring or recognizable personality from poem
to poem (e.g., Meleager's Heliodora), this is not true of Philodemos' Xanthippe.
To begin with, she is at least twice associated with the theme of marriage, although
not unambiguously so: first in 7 (see the commentary for a defense of the reading

35. For what follows, cf. my "Love poetry of Philodcmus," AJP 108 (1987) 310-324.
 Philodemos and the Epigram 35

) and again in 4, where, as I detail in the commentary, marriage best ex-

plains the narrator's goal.36
The question "Why marriage?" arises immediately. One does not, after all,
expect to find a wife as the object of attention in an erotic poem. As Philodemos
himself says, employing the more usual pose, the unattainable, or at any rate the
unattained, is more desirable than that which is near at hand (11). A more particu-
lar objection would raise the issue of Epicurus' strictures against the wise man's
marrying,37 but as both Chilton and Grilli agree, Epicurus does allow his followers
to marry, although only in exceptional circumstances. This view is in line with the
several other less than absolute strictures of Epicurus listed by Diogenes, includ-
ing the general prohibition against writing poetry.38
What these exceptional circumstances are neither Epicurus nor our sources
spell out, but we may imagine that much would depend on the character of the
woman. Since, moreover, women were welcome into the Garden for their intellec-
tual abilities, these fellow Epicureans would seem to be obvious candidates for
wives.39 Since, furthermore, women were appreciated for their bodies as well as their

36. Contra, M. Gigante, "Filodemo tra poesia e prosa (A proposito diP.Oxy. 3724)," SIFC
1 (1989) 130.
37. Dioe. Laert. 10.119

. Other sources state that according to Epicurus the wise man will not marry: Clem.
Alex. Strom. 2.23.138, Epiktetos (Arrian Epic. Disc. 3.7.19), Theodoret. 12.74. Cf. further C. W.
Chilton, "Did Epicurus approve of marriage? A study of Diogenes Laertius 10.119," Phronesis 5
(1960) 71—74, who defends Gassendi's conjecture. Contra, A. Grilli, "Epicuro e il matrimonio
(D. L. 10.119)," RSF 26 (1971) 51-56, who ably defends the MSS; cf. Seneca fr. 45 Haase ram
ditit [sc. Epicurus] sapienti ineunda conugia. Cf. further B. Frischer, The Sculpted Word (Berkele
1982) 61-63, on marriages within the Garden and the favorable Epicurean attitude toward women
in general; M. Gigante (ed.), Diogene Laerzio: Vita dei filosofi (Bari 1987) ad loc. See also
M. Nussbaum, "Beyond obsession and disgust: Lucretius' genealogy of love," Apeiron 22 (1989)
1-59, who demonstrates the high value placed on marriage by Lucretius. See now T. Brennan,
"Epicurus on sex, marriage, and children," CP 91 (1996) 346-352, esp. 348-350.
38. On Epicurus' prohibitions, cf. my response to Asmis (above, n.30). Note that in the
EthicaEpicurea, P.Here. 1251 col. 15.4-14 (ed. Schmid), marriage (among other things), althoug
of little importance for the most important matters of life, can contribute to men's external goods:

In other words, as Philodemos suggests at De Musica 4 col. 5.25-37, marriage is not

since it is possible to be happy without a wife (Phil. Oik. col. 9.1-3
Some men, nonetheless, can obtain wives who
will work with them for their common good (ibid. col. 2.3—5
Philodemos nowhere says that any man, let alone a philosopher,
should not marry. Indeed, he says that the intelligent man ( , a term which in-
cludes the Epicurean philosopher) will grieve most of all if his wife (or any other close relative)
is left in dire straits (De Morte, col. 25.2-10).
39. Such as Leontion, who wrote a work directed against Theophrastos which Cicero praise
for its style (ND 1.93), and who married Epicurus' chief disciple Metrodoros. Cf. C. J. Castner,
36 Introduction

minds, sex being regarded as a providing a natural, albeit unnecessary, pleasure,

sexual passion would not be expected to stop at marriage. As long as it not illegal,
harmful, or otherwise contraindicated, Epicurus urges a friend
(Sent. Vat. 51).
A woman who could satisfy both body and mind would make the ideal wife.
Such precisely is the picture of Xanthippe Philodemos develops over several poems.
Her sexual charms are seen most notably in 7. Her usefulness as an Epicurean
partner is shown in 4, where she is to remain with Philodemos through his years of
mature thought; and even more so in 3, where she, very much like Sosylos in 29,
keeps Philodemos from deviating from the correct Epicurean path. See further my
comments passim to 6.
If, then, the Xanthippe of the poems is Philodemos' lover, wife, and Epicu-
rean friend, what are we to make of her name? Conceivably, of course, it could
actually be the name of the woman who played such a role in Philodemos' life, but
far more likely it, like Neoboule and Heliodora among Greek poets and like Lesbia
and Cynthia among Romans, either represents a complete fiction or stands as a
convenient mask for a woman whose name was not to be presented in public.
Either case—I suspect the latter but each is consistent with my argument40—raises
another question: What significance might there be to the choice of this name?
Two complementary answers present themselves. First, as I have argued in detail
elsewhere, Philodemos, in the course of a the poetry cycle here described, has
developed a persona of a narrator who tries to lead the life of a philosopher.41 In
doing so, the poet has drawn upon topics found in the biographies of philosophers
of all schools. The motif of philosophic conversion, for example, can be traced
back—at least according to the ancient biographical tradition—to a disciple of
Empedokles, and is found in the lives of Plato and Polemon. Philodemos may have
derived all this from Epicurus' extended treatment of this theme in his Letter on
Occupations in which he discusses, among others, two
converts to Epicureanism: Mys and Leontion.42 His choice of the age of thirty-seven
for the narrator to marry echoes Aristotle (ibid.); and several elements are more
specifically Epicurean. That not all the philosophical topoi are Epicurean accords
perfectly with his view that poetry need not be beneficial, that is to say, didactic or
protreptic. Since the wife of his generic philosopher plays an important role in his
life, she has been given the name of the most famous of philosopher's wives,

"Epicurean hetairai as dedicants to healing deities?" GRBS2) (1982) 51-57; 'Slst,Epicurus (above,
n. 29) 10 f.
40. That is, I choose to believe that Philodemos, not giving any signs of favoring celibacy,
has fashioned a literary persona for the woman in his life. But how close the overlap between
"Xanthippe" and his real significant other I do not speculate.
41. A]P (above, n.35) and in Obbink ch. 4 (above, n. 30) . On the poet's persona, cf.
G. Paduano, "Chi dice 'io' neU'epigramma ellenistico?" in G. Arrighetti and F. Montanari (eds.),
La components autobiografica nella poesia greca e latina (Pisa 1993) 129—140.
42. Cf. Athen. 8.354a-d, with D. Sedley, "Epicurus and his professional rivals," in J. Bollack
and A. Laks (eds.), Etudes sur I'epicurisme antique (Lille 1976) 125 f. For the topic in general,
cf. O. Gigon, "Antike Erzahlungen iaber die Berufung zur Philosophic," MH 3 (1946) 1-21.
 Philodem and the Epigram 37

Xanthippe.43 He was, moreover, apparently following the lead of Epicurus; cf.

Alkiphron Epist. 2.2.1-3 (fr. 142 U), where, in one of his Letters of Courtesans,
Leontion complains of the way Epicurus treats her and others:

Alkiphron could hardly have concocted such a strange scenario out of whole cloth,
nor is it likely that he derived it from the epigrams of Philodemos. Whatever his
source, however, he must have derived it from a tradition with which Philodemos
too was familiar.
That this in turn would seem to call for Philodemos' philosopher being regarded
as a kind of Sokrates should not cause us to reject this identification. It is true that
Epicurus was not overly fond of Sokrates, and that he was followed in this regard
by his early disciples;44 Philodemos, however, who displayed less hostility than
Epicurus to Plato, was similarly more disposed to a favorable consideration of
Sokrates, who, better than anyone else, would provide a poetic paradigm for the
philosopher acceptable to all schools.45 Catullus 47, furthermore, offers some evi-
dence that Philodemos was called Sokrates by others (see above, pp. 23-24), but
whether this nickname was applied before or after Philodemos began his Xanthippe
cycle we cannot say. If the latter, the nickname may well have come about as a re-
sult of the poems.
In any case, there might be another, complementary, reason for the choice of
the name Xanthippe. In On Poems 5, Philodemos attacks Krates of Mallos for his
theory that euphony was of primary importance in assessing the worth of a poem.46
Philodemos of course was not deaf to the sonorous qualities of language; his poems

43. On the historical Xanthippe, see my "Love poetry of Philodemus" (above, n. 35) 321 f.
Our most trustworthy source, PI. Phdo 60a, portrays a woman who cared deeply for Sokrates
and respected his relationship with his friends. The notion that Socrates' Xanthippe was a shrew
is a later biographical fiction based on comic and Cynic sources; cf. W. Ludwig, GKBS 4 (1963)
75—77. On the women in erotic poetry, cf. J. G. Randall, "Mistresses' pseudonyms in Latin elegy,"
LCM4 (1979) 27-35; M. Wyke, "Mistress and metaphor in Augustan elegy," Helios 16 (1989)
25^47.
44. P. A. Vander Waerdt, "Colotes and the Epicurean refutation of Skepticism," GRBS30
(1989) 253-259, argues that the Epicurean school's hostility began with Kolotes. Cf. further
M. T. Riley, "The Epicurean criticism of Socrates," Phoenix 34 (1980) 55-68; K. Kleve, "Scurra
Atticus: The Epicurean view of Socrates," in Studi. . . a Marcello Gigante (Naples
1983) 1.227-253; A. A. Long, "Socrates in Hellenistic philosophy," CQ, N.s. 38 (1988) 150-171.
Most of the later accounts charging Epicurus with jealously slandering his philosophical rivals
derive from Metrodoros' brother Timokrates, who had a falling out with the school; cf. Sedley
(above, n. 41).
45. Cf. G. Indelli, "Platone in Filodemo," CErc 16 (1966) 109-112; and, for Sokrates,
D. Obbink, Philodemus on Piety (Oxford 1996) ad w. 701-703, 1358-1363. Note Cicero's
assessement of the place of Sokrates in the history of philosophy: De Or. 3.61 Nam cum essent
plures ortifere a Socrate, quod ex illius variis et diversis et in omnem partem diffusis disputationibus
alms aliud apprehenderat; proseminatae sunt quasi familiae dissentientes inter se et multum
disiunctae et dispares, cum tamen omnes se philosophiSocraticos et did vellent et esse arbitrarentur.
Cf. further K. Doring, Exemplum Socratis, Hermes Einzelschr. 42 (Stuttgart 1979) 8 f.
46. Cf. E. Asmis, "Crates on poetic criticism" (above, n. 23) 138-169.
38 Introduction

alone give ample testimony to this. In De Poem. Tract. I, P.Herc. 994, col. 29 N, fur-
thermore, he lists several names which strike him as cacophonous: ]
47
For Philodemos intentionally to choose an ill-sounding name for
the chief love object of his erotic poems and to use it at least seven times in the extant
poems and incipits (whether or not one believes in the complex Xanthippe cycle I
argue for) serves as a challenge to Krates and any followers he may have had by offer-
ing a counterexample to disprove his theory. For Philodemos' audience of philosophi-
cally inclined poets, implicit in these poems is the message that the thought
of a poem not only may allow for a harsh sounding name, but that in this particular
cycle of poems it almost calls for a name which by itself can be thought lacking in
euphony, as indeed it was by Philodemos himself. The result may still be a good poem.
This not only is Hellenistic Witz of a high order, it also exemplifies the way in
which Philodemos' epigrams, or at least some of them, manifest his theory of poetry.
In the case of Xanthippe, he shows that a cacophonous word can be used to rein-
force the poem's thought. In 22 he can violate several metrical norms in order to
reinforce the crudity of thought. And in a more general way he alludes to ideas and
anecdotes associated with other philosophers, such as Sokrates, Aristotle, and
Polemon,48 especially the last named's conversion to philosophy (see above). Thus,
although presumably any nondidactic topic can appear in a poem designed to give
pleasure and thus can satisfy Philodemos' criteria for good poetry, in a cleverly
urbane way which is fully consistent with Hellenistic poetics Philodemos chooses
manifestly un-Epicurean topics in order to demonstrate in the clearest way pos
sible his Epicurean theory that neither truth nor benefit (both of which can be found
in Epicurean prose treatises) is necessary in poetry.
The poems, then, are in accord with Philodemos' poetic theories in particular
and may, when looked at obliquely, be in accord with broader Epicurean theories.
Even in 27, for example, the invitation poem to Piso and the most overtly "Epicu-
rean" of the epigrams, Philodemos plays with the idea of Epicurean friendship as
it accomodates itself to Roman amicitia (see the commentary). Two other epigrams
3 and 29, are alike in obliquely illustrating Epicurean ideas ofparrhesia. Both are
dialogues between "Philodemos" and a friend, Xanthippe and Sosylos respectively
who curtly and frankly recall him to the proper Epicurean attitude. The influence
of Philodemos' on Horace's Satires has long been recognized,49

47. The text is most easily available in F. Sbordone, "Filodemo e la teorica dell'eufonia,"
RAAN30 (1955) 25-51, repr. in Suipapiri delta poetica diFilodemo (Naples 1983) 125-153 (se
p. 138); and in id. Ricercbe sui Papiri Ercolanesi 2 (Naples 1976) 94 f. R. Janko (above, n. 20)
locates this papyrus in On Poems, Book II. Note also Phil. P.Herc. 460 fr. 22 = Tr. B fr. 7 col.i
S On Philodemos' theory of euphony, see also N.
Pace, Problematiche dipoetica in Filodemo di Gadara (Diss. Milan, 1992) 95-115.
48. See D. Sider, "The Epicurean philosopher as Hellenistic poet," in Obbink (above,
n. 20), 44-57.
4 9 . N . De Witt, "Parresiastic poems of Horace," CP30 (1935) 312-319; A. K. Michels,
and the satire of Horace," CP39 (1944) 173-177. On Epicurean parrhesia in gen-
eral, cf. Gigante, Philodemus in Italy, 24-29; Asmis, "Philodemus' Epicureanism" (above, n. 23)
2393 f.; C. E. Glad, "Frank speech, flattery, and friendship in Philodemus," in J. T. Fitzgerald
(ed.),Friendship, Flattery, and Frankness of Speech (Leiden 1996) 21-59; id.,Paul and Philodemus
(Leiden 1995), csp. ch. 3, "Epicurean communal psychagogy."
 Philodemos and the Epigram 39

but Philodemos' use of it in his own poems has been ignored. In brief, parrhesia or
frankness is important for the philosopher in two ways. First, he or she must apply
it properly in order to improve others; second, the philosopher must learn to accept
frank speech from others so that he too may be taught the better way. Parrhesia is
not simply speaking the truth (nor is in any way concerned with
epistemology); one must, rather, learn when to apply it to whom in order to achieve
the desired moral end. A teacher will need to apply more cautious language in ad-
dressing a student or a ruler, but the situation of the two epigrams falls under the
rubric of philosopher speaking frankly to philosopher and "if the wise recognize
each other they will gladly be admonished by one another just as though they were
reminded by themselves. And they will sting one another the gentlest sting and be
grateful" (col. 8B).50 By playing the admonished one and writing the words of the
admonisher in 3 and 29, Philodemos exemplifies both aspects oiparrhesia. For this
division between poet and persona within the work, there are both poetic and philo-
sophic models. Of the former the most noteworthy example is perhaps Sappho 1,
where the poet clinically portrays a self who, incapable of rational analysis, is dis-
dainfully regarded by Aphrodite.31 For the latter, consider the several passages in
Plato where Socrates shifts what could easily be criticism of his fellow dialogist onto
himself, as, supposedly, delivered by an outside voice: The Laws in Crito, Diotima
in Symposium, an unnamed stranger in Hippias Major, and the Logos in Protagoras.
In other words, Philodemos may be said to have composed epigrammatic diatribes
in which he allows himself to be chided.52
In sum, although, as we concluded above (p. 32), any (good) epigram—such
as the many erotic ones to be found in this edition—can satisfy both Epicurus' and
Philodemos' requirements for poetry, Philodemos, writing with his particular au-
dience in mind, extended the range of epigrammatic topoi to include a number of
philosophical subjects, largely but not exclusively Epicurean.

Ill

When did Philodemos write his epigrams? Let us assume what scholarship can-
not prove—that early on in his education he was introduced to Greek verse com-
position.53His superiority to his contemporaries suggests long practice as well as

50. P.Herc. 1471, ed. Olivieri, col. 8 B.6-13

I pursue this further in "How to commit philosophy obliquely: Philodemos' epigrams in the light
of his Peri Parrhesias," inj. T. Fitzgerald et al. (eds.),Philodemus and the New Testament World
(Leiden, forthcoming).
51. Cf. J. Winkler, The Constraints of Desire (New York 1990) 171.
52. Cf. E. Norden,AntikeKunstpmsa (Leipzig 1909) 129; B. P. Wallach, Lucretius and the
Diatribe Against Death (Leiden 1976) 6 f.
53. So, e.g., L. A. Stella, Cinque poeti dell'Antologia Palatina (Bologna 1949) 248.
40 Introduction

natural talent. Let us further allow for the possibility that he brought some of
these early efforts with him to Italy.54 That he continued to write in Italy is guar-
anteed by 27, the invitation to Piso. For two other epigrams, 28 and 29, a setting
in Herculaneum is likely, as Gigante has shown, and as I take as given in the com-
mentary; but even though some Greeks are named who have been identified
with individuals living in Italy, it has to be confessed that a Greek setting cannot
be absolutely ruled out.55 Setting, moreover, does not guarantee place of com-
position. Since Philodemos varies the narrating persona of his epigrams, any
attempt to extract autobiographical data which could determine date of com-
position must be regarded with extreme caution, despite the interesting picture
which Gigante has developed from just such an attempt. 56 It is also clear that a
significant number of epigrams were available for purchase in Rome by 55 B.C.,
the date of Cicero's In Pisonem (see c. 71, T 2), as we could have inferred in any
case from the several echoes of Philodemos in Catullus, who died about this time.57
The date of one poem, 4, which is written in the persona of a poet who feels
the call of a more cerebral life now in his thirty-seventh year, could, if taken lit-
erally, be assigned to ca. 73 B.C. (see above, pp. 6 f.), for the evidence for setting
ca. 110 B.C. as Philodemos' birth year.) In the commentary I argue that this par-
ticular age was taken over from Aristotle, but Philodemos certainly could have
been of a mind to write such a poem on his thirty-seventh birthday. I neither deny
this nor make anything of it. I cannot, however, accept A. H. Griffiths's reading
of this poem (BICS 19 [1970] 37 f.) in which he understands the koronis men-
tioned on v. 7 to refer to the actual koronis alongside this poem that would mark
this poem as the last epigram in Philodemos' book. Since I find it more likely
that the koronis refers metaphorically to Xanthippe—and that, moreover, the com-
position of more poems are foreseen—it would seem that we have as little idea of
the arrangement of epigrams within Philodemos' book as we do of when they were
written. Note, too, that in 55 B.C. Cicero refers to Philodemos' poetic activity in
the present tense (In. Pi's. 70 = T 2 est. . . perpolitus; poema . . . facit). There is
therefore no evidence to suggest that Philodemos did not compose epigrams
throughout his adult life.

54. Stated with more certainty than the evidence allows by T. Dorandi, "La Villa dei Papiri
a Ercolano e la sua biblioteca," CP 90 (1995) 175, who follows Gigante in assuming that these
first poems predate his "formazione filosofica."
55. Some of the papyrus incipits clearly refer to Italy; see pp. 212—214.
56. Note the title of the third chapter of Philodemus in Italy, "Philodemus' Epigrams as
Autobiography." I am especially critical of his analysis of 34; see the commentary. Among much
recent work on the poet's persona, the following may be profitably consulted: Paduano (above,
n. 41); S. Goldhill, The Poet's Voice: Essays on Poetics and Greek Literature (Cambridge 1991);
W. Rosier, "Persona reale o persona poetica? L'interpretazione delP'io' nella lirica greca
arcaica," QUCC 19 (1985) 131-144; M. Lefkowitz, "Autobiographical fiction in Pindar," HSCP
84 (1980) 29-49 (repr. in ead. First-Person Fictions: Pindar's Poetic T [Oxford 1991] 127-
146).
57. Two of these were 1 (probably) and 27 (almost certainly); see Tait 36-47.
 Metrics 41

Metrics
A profile of Philodemos' metrical practices fits well within Hellenistic limits, which
are often stricter applications of archaic and classical norms. Although reference
is made to all the poems here published, percentage figures come from the 94
hexameters and 94 pentameters (when not affected by editorial changes which
alter the shape of the line) of the twenty-nine poems regarded as Philodemos' own
by Gow-Page (including the final distich of 3, which they relegate to the notes),
since in the first place it was my intention to learn Philodemos' general practice
in order to see whether it would provide any criteria for helping to determine
authorship of the doubly or doubtfully ascribed poems (it does; see below), and
secondly because some figures were already provided for this group by Gow-Page,
Page (in hisRufinus), West, Greek Metre, and M. van Raalte, "Greek elegiac verse
rhythm," Glotta 66 (1988) 145-178. Cf. also M. L. Clarke, "The hexameter in
Greek elegiacs," CR 5 (1955) 18; W. Seelbach, Die Epigramme des Mnasalkes
(Wiesbaden 1964) 135-140; D. Korzeniewski, Griechische Metrik (Darmstadt
1968) 35-40; C. M. J. Sicking, Griechische Verslehre (Munich 1993) 83-87; S. R.
Slings, "Hermesianax and the Tattoo Elegy (P.Brux. inv. E 8934 and P.Sorb. inv.
2254)," ZPE 98 (1993) 29-37.

I. General
Correption
Philodemos is strict in generally allowing this only
(a) in the first dactyl of the hexameter (11.6), and
(b)at the bucolic diaeresis in the hexameter (5.3, 21.5, 22.5, 36.3) and at the
equivalent postion in the pentameter (12.4, 16.4); at 4.6 I prefer elision.
Within a dactyl only 22.1 (see comm.). For correption of Krai, usually
ignored in compiling these statistics, see introduction to [38] and GP l.xxxix f.
Cf. Kaibel iv-vi.

Elision
In nouns, adjectives, and verbs this is usually avoided in elegiac verses: Asklepiades'
ratio of 14 per 100 lines is abnormally high; 5 per 100 is more common. Philodemos'
ratio is 3.2 per 100 for the Gow-Page canon, with some additional exx. from the
doubtful poems: 4.6 (see comm.), 7.5 (see comm.), 9.7,11.4, 12.8, 15.3,18.4,21.4
(see comm.); AP5.145.3.

Plosive + liquid/nasal
Generally this combination makes a short syllable long by position within a word,
with exceptions allowed for otherwise metrically intractable compounds (3.1
42 Introduction

3.3 and for proper names This

combination tends not to make position when it begins a word, but will do so when
the preceding word is an article, preposition, or . Exceptions are 3.3
(Gow-Page print Schneider's unnecessary emendation ), 14.3
23.7 (although Gow-Page, GP l.xxxix n.2, are
willing to consider this last example as forming a word group which would allow
for a long syllable), 32.2 Epigrammatists generally avoid placing
final short vowel before a plosive-liquid combination; Philodemos is freer than most
in allowing this, with 11 examples all told (including those illustrated above).
Cf. GP l.xxxviii f.; Slings 36 f.

N« movable
Philodemos allows this to make position only once: 9.4; GP l.xliv f.; but cf. II vii.16.

II. Hexameter

Masculine vs. feminine caesura

The ratio is 49:45, or 52% for the masculine caesura, which goes against the gen-
eral tendency of Hellenistic authors greatly to favor the feminine over the mascu-
line. Only Theokritos in his bucolics and mimes is close to Philodemos (50%-52%
fern.); cf. Leonidas (56% fern.) Meleager (61%) and Apollonios (67%), Kallimachos
in the epigrams (78%), Theokritos in the epyllia (72%); West 153, van Raalte 164.
Philodemos has no hexameters without a third-foot caesura, as is the near univer-
sal rule in elegiac hexameters, as well as in Hellenistic hexameters in general; West
153, 157, van Raalte 164, Seelbach 137.

Proparoxytone hexameter-ends
Philodemos' 13% is unremarkable; cf. Philip (14%), Meleager (13%), Palladas
(13%). Cf. Page Rufinus 28.

Bucolic caesura
This occurs in 72% of Philodemos' verses; with Homer's 47% contrast Meleager
(58%), Leonidas and Apollonios (63%), Kallimachos' epigrams (89%), and
Theokritos' bucolics (74%); West 154, van Raalte 165. Five verses (5.3%)
with a masculine caesura fail to have bucolic caesura, which is somewhat "lax":
9.1, 12.1, 3, 26.1, 31.3. The average for early Hellenistic elegists is 4.5%, although
Kallimachos has none; GP l.xliii; Clarke, 18. Philodemos also has a higher than
average percentage of lines (42, or 45 %) combining masculine caesura with bucoli
caesura; for the early Hellenistic elegists this is 35% (20% for Kallimachos).
 Metrics 43

Spondaic lines
These are in any case rare in elegiac hexameters, and do not occur in any poem
ascribed to Philodemos; van Raalte 151. Furthermore, elegiac hexameters tend to
avoid spondees after the second foot: 59% of Philodemos' have none, and no line
has spondees in both the third and fourth feet. On the other hand, Philodemos is
freer with spondees in the first two feet (53 and 50%, respectively) than others:
Nearest to him in van Raalte's list are Kallimachos in the epigrams (31 and 53%)
and Aitia (34 and 48%) and Theokritos (38 and 34%); van Raalte 163.

fifth-foot word breaks

Philodemos (like Philip and Krinagoras, and unlike Antiphilos and Argentarius;
GP 1 .xliv, Clarke) is unusually willing to allow a word break after the first syllable
ofthefifthfoot:25.3 31.3 and34. offend against the tendency
to avoid a word ending at position 9 of the shape |-- -1; cf. Maas, GM §97, who
cites Kallim. H. 1.36, 94, 4.311 and Plut. Mor. 747 f, who calls such a verse
31.3, with word end at position 5, also violates Meyer's Third Law.
Note in additon 9.5 1, 19.1 |, 20.1 , (where also
violates Hermann's Bridge), 33.3

The syllable before the masculine caesura

This is usually long by nature: Exceptions: 3.3 (cf. I (iii) above),
15.3 19.5 22.1 Philodemos at 8.5% thus falls
between the average for all Garland authors (10%) andPhilip (2.5%);GP l.xliin.

Wernicke's Law
Philodemos observes this, not allowing the final syllable of foot four where it coin-
cides with word-end to be long by position; West 37,155 n.50. (9.7 and [37].3
are prepositives.)

Meyer's First and Second Laws

The First Law (against word ending X — | or x - -- | in the second foot) goes
unobserved at 7.1, 9.1*, 22.1*, 23.3*, 34.3*, 34.5, [37].3 (the asterisk indicating
that a word of shape | -— | immediately follows, in violation of the Second Law).
For Meyer's Third Law, see on 32. 3-4.

III. Pentameter
Accented pentameter-ends
The figure of 13% places Philodemos closer to the earlier generations of Helle-
nistic epigrammatists; e.g., Kallimachos (17%). Philip's authors, with the excep-
44 Introduction

tion of Philodemos and Krinagoras (7.6%), employ this feature more sparingly;
e.g., Antipater of Thessalonica (3%) and Philip himself (1%); Page Rufinus 30,
West 159.

The syllable before the caesura

LENGTH. Philodemos is accord with most of Philip's authors in tending to keep
this long by nature. Theokritos has 23%, Asklepiades 11%, Kallimachos and
Leonidas each have 12% of such syllables long by position; whereas Meleager,
Apollonides, Bianor, and Philip have none. What Philodemos actually has depends
upon which reading is chosen (in31.6), whether the text is to be emended (in20.2),
and whether the poem is by Philodemos or another ([37].2 and [38].2). I avoid
positional lengthening in the first, keep it in the second, and use it as evidence against
Philodemos' authorship in the third and fourth. Cf. Maas GM §22, Page Rufinus
30f., West 158.

ELISION. This is found only once, in 21.4, which is also ascribed on weaker
authority to Antiphilos; see the comm. Cf. GP l.xliii.

Homoioteleuton between pentameter-halves

The figure for Philodemos is 22%, which is relatively high. Com ative percent-
ages supplied by K. Muller, Die Epigramme von Antiphilos von Byzanz (Berlin 1935)
29 f., average out at 15-16:

Theokritos 8 Simmias 11
Nossis 9 Hedylos 18
Kallimachos 10 Anyte 20
Mnasalkes 10 Leonidas 21
Antiphilos 11 Philodemos 22
Asklepiades 11 Bakchylides 25
Poseidippos 11 Nikias 33

Agreement between pentameter-halves

Philodemos at 31.6% similarly ranks above average against his fellow epigramma
tists (24.7%) in the related phenomenon where the words before the caesura and
at the end of the pentameter are in grammatical agreement as noun + adjective/
participle/pronoun (whether or not the noun comes first, and whether or not rhyme
results). The comparative percentages come from Slings 37:

Nossis 12.5 Philodemos 31.6

Asklepiades 14.7 Mnasalkes 36.8
Kallimachos 16.1 Anyte 38.5
Leonidas 22.6
 The Greek Anthology 45

The Greek Anthology

I. Philodemos in the Anthology

The "Greek Anthology," as regularly constituted in modern printed texts, com-

prises sixteen books. The first fifteen essentially reproduce the Palatine Anthology
(AP) as it is found in one manuscript (P), which is now divided between two
libraries, the larger part in Heidelberg, the smaller one (containing no epigrams of
Philodemos) in Paris. The sixteenth book gathers from a manuscript in Venice (PI)
more specifically from the eleven sections contained therein of another, smaller,
collection of epigrams, the Planudean Anthology (API)—the 388 epigrams which
it alone contains. This combination of the two manuscripts by modern editors makes
sense since AP in large part and/IP/in its entirety derive from the same source, the
even larger collection of epigrams put together by Constantine Kephalas, who was
protopapas of the palace in Constantinople in 917.:
Since Kephalas' gathering was itself an omnibus edition of earlier collections
of epigrams, beginning with those of Meleager and Philip, readers of the conglom-
erated "Greek Anthology" are presented with ca. 4,100 epigrams written by 363
named and an unknown number of anonymous authors ranging in date from
Archilochos to the tenth century, arranged largely by subject matter or topos (e.g.,
erotic, dedicatory, sepulchral).2
Although the more interesting of these authors, such as Kallimachos,
Asklepiades, and Philodemos, have been given separate modern editions and com-
mentaries,3 it remained for A. S. F. Gow and D. L. Page to reconstitute in large
part the collections of Meleager and Philip.4 Although Brunck, Jacobs, and Reiske

1. This section can be but a brief and simplified summary of a complicated subject: See
further Alan Cameron, The Greek Anthology from Meleager to Planudes (Oxford 1993);
C. Preisendanz,AnthologiaPalatina (Leiden 1911) praefatio;P. Waltz,Anthologie grecque 1 (Paris
1960) iii-xxxvii; HE l.xiv-xlv, GP l.xi-xxxii; Stadtmiiller (see below, n. 16) l.iii-xxxiii.
2. Arrangement by subject matter goes back to Meleager's Garland (Cameron 19-33), which
scheme was also followed by Agathias in his cycle, and by Kephalas, whose arrangement is largel
reproduced in AP (Cameron [above, n. 1] 122-126). Planudes' topical arrangement (see below)
is different. Philip, however, arranged the poems in his Garland alphabetically; i.e., by initial
letter (and no further) of the first word; cf. Cameron 33—40, who argues that "Philip's original
Garland comprised one long alphabetical series without regard to subject matter" (35), but that
within the alphabetic arrangement there was some grouping by theme. That is, if two poems
beginning with (say) beta were on the same subject, they would be placed together. This means
that the Anthology offers absolutely no hint as to the disposition of Philodemos' epigrams within
its original publication.
3. See the bibliography for a list of these editions.
4. "In large part" because Meleager's many pre-Hellenistic epigrams were excluded (to be
published later in Page's further Greek Epigrams), since Gow-Page, as their title indicates, were
interested only in Hellenistic epigrams. Further deviating from Meleager, Gow-Page include
Hellenistic authors who wrote before the beginning of the period covered by Philip in his Gar-
land, but who were not among Meleager's authors, e.g., Theokritos; see further HE 1 .xiii £, 2.525.
46 Introduction

had earlier collected the poems of individual authors in their editions of the
Anthology,5 it now became much easier, thanks to Gow-Page's far more compre-
hensive introductions and notes, to familiarize oneself with the individual contribu-
tors, and, furthermore, to get a sense of early (in HE) and late (in GP) Hellenistic
epigrams as a whole. As Gow-Page note, the quality of poetry in Meleager's collec-
tion is significantly higher than that of later collections, although individual poets
such as Philodemos, Antipater of Thessalonica, Argentarius, and Krinagoras rise
above the generally low level.
Of all the problems faced by the editor of a single author from the Anthology,
the most important is the question of ascription and genuineness.6 Most of the poems
in the Greek Anthology are ascribed by both of its constituents to but one author,
and, with few exceptions,7 this agreement is taken by editors to constitute suffi-
cient grounds for trust. Occasionally, however, AP and/IP/ disagree, often because
of the practice of labelling one poem that is, "by the same poet who
wrote the preceding poem." Since poems were reshuffled from earlier collections
to later, this was bound to produce occasional confusion. In addition, a poem could
lose its label and show up with one or another of the terms indicating anonymity in
either AP or APl.&
In their introduction to individual authors, Gow and Page are careful to direct
the reader to poems ascribed to this author but which they feel belong elsewhere.
For Philodemos they print twenty-nine poems, whereas Kaibel in his edition of
Philodemos printed and commented on only twenty-four as genuine, printing with
brief dismissive statements another five which he regarded for various reasons as
unworthy of him. (Kaibel's resulting twenty-nine poems are not coextensive with
Gow-Page's.)
For Philodemos, the situation is as follows: Of the thirty-six poems ascribed to
him by at least one source within the Greek Anthology,

(i) There is no disagreement for twenty-six, that is, either AP and API are in
agreement or one lacks the poem in question;

HE also includes epigrams by Meleagrian authors known from sources (such as papyri) other
than AP and API, although it will never be known whether these poems in fact formed part of
Meleager's selection. (One suspects from the low quality of many of the new Poseidippos epi-
grams that they would never have satisfied Meleager.) It should also be noted that Gow-Page's
alphabetical arrangement by author is not that of either Garland, on which see n. 2.
5. And in a limited way Planudes before them; see below.
6. This was of course important for Gow and Page as well, since decisions on doubly
ascribed and otherwise doubtful poems (especially those marked anonymous) determined place-
ment within or exclusion from one or another of their authors. Hence the necessity Gow felt to
examine the matter in GA - The Greek Anthology: Sources and Ascriptions (London 1958). For
editors of the entire Greek Anthology, who print everything in the order of AP and then those
unique to API as "Book 16," this question is of less importance.
7. Such as AP 5.24 (13), ascribed by both sources to Phil, but usually given to Meleager;
see the commentary.
8. Cf. Gow GA, passim, on problems of ascription due to these errors.
 The Greek Anthology 47

(ii) For two poems there is ascription to Philodemos in both AP and API, but
AP adds a second claimant: 21 P Antiphilos, with the Corrector (see be-
low) addinj 35 P
(iii) Four poems are ascribed to Philodemos by AP but are anonymous in API
(4,23,31,32);
(iv) Four poems are ascribed to Philodemos by one collection and to some-
one else in the other (2 P Plato, PI Philodemos; 18 P Maccius, PI
Philodemos; 36 P Meleager, PI Philodemos, 37 P Marcus Argentarius, PI
Philodemos).

Gow-Page's twenty-nine epigrams comprise all from (i) except 13, which they
give to Meleager, and all from (iii); and none from (ii) or (iv). They also print as a
thirtieth poem a passage from Horace which seems to allude to a poem of
Philodemos; cf. T 4.
My editorial "solution" to the problem of authenticity is to print all poems, if
not as separate epigrams with accompanying commentaries, then at least in some
other appropriate place; that is, all poems (i) which have been ascribed to
Philodemos by either Anthology, (ii) whose incipits appear in the P.Oxy. list
see below), and (iii) of which there are some grounds for believing we have Latin
translations in the Epigrammata Bobiensia, a late fourth-century collection. Also
included are (iv) one anonymous and postclassical (i.e., Renaissance or Baroque)
epigram which was written to supply the lost original of a poem alluded to by Horace,
and (v) the Oxyrhynchus incipits in their entirety, some of which undoubtedly derive
from unknown Philodemean epigrams. This results in forty-one Greek and two Latin
epigrams, none longer than eight lines. Some of these doubtful poems, I shall argue,
are not in fact by Philodemos, but there seemed to be a clear value in gathering
together and assessing all claims for Philodemean authorship, however unlikely some
may be.
Philodemos, like most epigrammatists, was undoubtably prolific. It was in fact
a genre in which facility of production was often as highly regarded as the finished
product.9 Of all that Philodemos wrote an unknown fraction was published and
readily available in Italy during his lifetime and later to Philip.10 From the existence
of many epigrams known from outside the Greek Anthology (cf. e.g. Kallim. frr.
393—402 Pf. and the new epigrams of Poseidippos, totaling over 600 lines11), we
can be sure that Philip exercized editorial choice in gathering his Garland. In the
case of Philodemos, we have the clear hint of far more Philodemean epigrams in
than are now extant, but of course we do not know how many of these were in-
cluded by Philip but excluded by Kephalas, whether intentionally or, more likely,

9. Cf. Cic. Pro Arch. 18 on Archias (cited above p. 28).

10. Cic. In Pis. 71 (T 2). On Philip's see Cameron [above, n. 1] 33-43 and
(arguing for a Neronian date) 56—65; GP l.xi-xlix.
11. Initial notice of the latter: G. Bastianini and C. Gallazzi, "II poeta ritrovato: Scoperti
gli epigrammi di Posidippo in un pettorale di mummia," Riv. "Ca' deSass" n.121, March 1993.
A preliminary text of 24 of the epigrams in eidem, Posidippo. Epigrammi (Milan 1993).
48 Introduction

because, as Cameron Greek Anthology 43-48 argues, he was working from two
incomplete copies of Philip's (and two of Meleager's) Garland.12 As the poems
unique to either AP or API show, Kephalas' collection itself was subject to further
abridgement mAP and, to an even greater extent, in API. Thus at each one of these
stages we probably have lost some of Philodemos' (and of course others') epigrams.

II. The Manuscripts

For most purposes, there are but two manuscripts to be reported, P and PL Copies
made from these are of only occasional value for Philodemos. Diogenes Laertios
cites one epigram, 2, as the work of Plato. For some few poems of Philodemos,
anonymous excerpts in Souda offer significant variants; cf. Cameron ch. 12. The
readings of (see below) are occasionally of interest. See further below. In the
descriptions which follow, an asterisk (*) indicates those manuscripts which I have
not myself examined.

Heidelberg. Palat. 23, containing AP among other texts, compiled towards the
middle of the tenth century, is the Codex Palatinus, so called from its stay in the
Palatine Library in Heidelberg, although its latter part remained in Paris (as Paris.
Suppl. Gr. 384) after the first part was returned to its rightful home (as Palat. 23)
after the Napoleonic depredations. Cf. H. Gorgemanns in E. Mittler et al. Bibliotheca
Palatina (Heidelberg 1986) 1.485^87.
The manuscript contains the work of more than one scribe, one of whom, J,13
may have provided the lemmata to individual poems. This brief description can
dispense with the details of scribal attribution, but one other hand in P deserves
attention, that of C, its "Corrector," who ca. 950 took P in hand after it had already
received additional lemmata and attributions by J (and perhaps others). C, like J,
had access to an independent text, which he identifies as one made by Michael 6
(the archivist). Although the extent of C's dependence on Michael is
unclear in many details (cf. HE l.xxxv), the value of his comments and editorial
alterations is manifest to anyone who notes how often he agrees with PI in the bet-
ter reading against P, or even more so, when he offers the better reading when PI is
lacking, or against PPL14

12. Kephalas, however, must have embodied Meleager, Philip, and Agathias' anthologies
as completely as he could from his imperfect copies; cf. Cameron 121 f.
13. Now identified as Constantine the Rhodian by Cameron 300-307. For the wanderings
of this MS across Europe, cf. Cameron 178-201.
14. C's contributions cease after AP 9.563; i.e., ten poems in P ascribed to Phil., one anony-
mous poem possibly by Phil. (24), and one from the Planudcan anthology assigned to Phil, did
not receive his attention. On C, see further Cameron 108-120, 129-134.
 The Greek Anthology 49

Unclear to the reader of a printed apparatus criticus, however, is how often C

effects his alterations by either erasure or alteration of the letters in P. Altogether
too many of these alterations have made P's original largely or totally illegible or
indecipherable in Preisendanz's facsimile.15 In these cases, only autopsy, prefer-
ably with the aid of an ultraviolet light, can be trusted, and not always then, so thor-
oughly did C sometimes obliterate. For most readings, my autopsy confirmed that
of Stadtmiiller; occasionally I was forced to disagree with his most careful reading
of the manuscript.16 C, that is, in opposition to the practice of modern copy editors,
conceals the original text. He will, for example, with a little erasing and the addi-
tion of another circle, convert an omicron into an omega. Having been alerted to
this by Stadtmuller, one can then see that indeed there seems to be some squeezing
between the omega and the letter to its right, but nobody would dare declare on
the basis of the facsimile alone that this was the work of C. Thus, throughout the
Anthology, whenever the apparatus criticus states that C offers one reading while
the reading of P is doubtful, the reader should assume that C has obliterated P. In
7.5, e.g., ut vid. P" (Gow-Page) derives from Stadtmuller's

PI
Cod. Ven. Marc. 481, containing API among other texts, is the autograph of Maxi-
mus Planudes, who in 1301 (Cameron 75-77) prepared his own collection; although
also derived ultimately from Kephalas', it is arranged differently from P.
Seven books of epigrams (Pla = la-la) are separated by other matter from a
second grouping (with some duplications, none affecting Philodemos, from the first
seven sections) of four additional books of epigrams (Pl^ = \b-4b), which, as
Planudes explicity says, derive from a different source from that of Pla. As was stated
above, any epigram common to P and Pl is printed in modern editions in P's order
(I follow Stadtmuller and Beckby in noting the location of these poems within PI),
while those unique to PI are gathered as "Book 16." Planudes arranged his books
by subject matter, providing a further breakdown within several of the books. Thus,
Beckby's reference (which I follow) for 4, PI 2^.23,14, indicates that the poem is
found in the second book (2) of the second group of epigrams (b), as the fourteenth
poem of the book's twenty-third topical subsection in this case,
17
Only Books 1-4 (a and b combined, as was customary

15. Anthologia Palatina: Codex Palatinus et Codex Parisinus, Codices Graeci et Latini 15
(Leiden 1911).
16. H. Stadtmuller, Anthologia Graeca Epigrammatum Palatina cum Planudea, 3 vols.
(Leipzig 1894-1906), ending at AP 9.563 (where there is a change of scribes). For Gow-Page's
dependence on Stadtmuller for the readings where C has altered P, cf. HE l.xxxvii with n. 2.
Stadtmuller's apparatus is overfussy, giving details of accent, breathing, punctuation, and letter
placement, when, for the most part, there is no doubt about what is intended. Although I have
learned much from it, the models for my own apparatus criticus are the neater ones of Waltz,
Beckby, and Gow-Page, although I report more conjectures of early editors than they.
17. As was noticed by R. Aubrcton,_B/lGB (1967) 349, Beckby missed a heading after no. 9,
with the result that all his heading numbers after nine should be raised by one.
50 Introduction

since soon after PI was written) and 6 have headings, which are arranged alpha-
betically; cf. Beckby 4.560 ff. for an index of and a concordance between
AP and API (these subsections are like those in the new Poseidippos papyrus). Not
surprisingly, more of Philodemos' epigrams are found in Book 7 than in any other,
this book containing Planudes' erotica, or, as he calls them
It has been little noticed, however, although immediately clear to all who ex-
amine the manuscript, that Planudes, or, more likely, his immediate source, made
an attempt in this book to gather together poems by the same author. Among the
longer runs are Meleager (eight poems, fols. 68v-69r), Paulus Silentarius (21, fols.
70r-71r), Agathias (12, fols. 71r-v), Philodemos (16, fols. 72v-73r), Meleager bis
(13, fols. 73r-v), andRufinus (28, fols. 73v-74r); cf. Stadtmuller 1 xxii-xxix. Since
the Philodemos group contains more than its fair share of disputed poems—both
more than a random sampling from Philodemos should have and more in absolute
numbers than the other author groupings—Planudes' accuracy in attribution calls
for examination. He and all compilers who make use of the phrase in-
stead of writing the author's name (see above) are liable to produce error, first if
the phrase is applied carelessly (typically by the compiler jumping over the imme-
diately preceding poem to the one before it), and second if the poem passes to
another collection which is arranged differently from the first now des-
ignating an altogether different poet). There is no doubt that Planudes is guilty of
carelessness in this regard, but he is not to be condemned outright, as Page does in
examining his attributions to Rufinus,18 whose first poem in Planudes' group, la-
beled is followed (as is true of all the author groups) by instances
(twenty-seven in Rufinus' case) of Four of these poems receive variant
attributions in P: two adespota, Marcus Argentarius, and Kallimachos, none of
which Page admits into the Rufinus canon. Here he may be right, but in Meleager
bis, AP 5.82, which is anonymous in P, seems to me to be worthy of Meleager.19
Consider Planudes' run of Philodemos' epigrams in Table 1. Operating on the
principle enunciated by Page in Rufinus, wherever there is a discrepancy Gow-Page
follow P against PL20 We now know, however, that for two of the poems in the
Philodemos group (2 and21) Philodemean authorship is strongly indicated by their
incipits' appearing in This means that the remaining doubtful epigrams should
be judged individually (as they will be in their respective commentaries below), and
not denied to Philodemos automatically.
On the whole, P is a more reliable manuscript., but Planudes with access to
other sources and with some common sense of his own often offers the true reading.
Cf. E. Mioni, Codices Graeci Manuscript! Bibliothecae DiviMarci Venetiamm,
II. Thesaurus Antiquus: Codd. 300-625 (Rome 1985) 276-283; id., "L'Antologia

18. D. L. Page, The Epigrams of Rufinus (Cambridge 1978) 14-18; sim. Stadtmuller
l.xxvii f.
19. Gow-Pagc HE 2.593 do not even bother to include it among the poems they exclude
from the Meleager canon. C. Radinger, Meleagros von Gadara (Innsbruck 1895) 81 at least men-
tions it (but without argument keeps it anonymous).
20. For their reasons for assigning 13 to Meleager, see the commentary.
 The Greek Anthology 51

T.ABLE 1 The Ru n of Epigriims Assigned to Philodemias by Planud es

I>1 ID n Sider

7.86 5.306 v.13 25

87 8 Plato iv.31 2
88 4 iv.10 7
89 8 — 36
90 124 — 16
91 24 iv.17 13
92 25 — 15
93 13 vii.25 9
94 112 v.ll 5
95 113 — [37]
96 114 — 18
97 115 vii.7 10
98 121 ii.19 17
99 131 v.14 1
100 132 v.20 12
101 308 vi.4 21

unless an<other name is given}.

Greca da Massimo Planude a Marco Musuro," in Scritti in onore di Carlo Diano

(Bologna 1975) 263-307; A. Turyn, "Demetrius Triclinius and the Planudean
Anthology," 39-40(1972-1973)403-
450.
Two early copies of PI contain nothing of interest for the text of Philodemos:
(i) London BM Add. 16409, containing corrections in Planudes' hand; and (ii) Paris
gr. 2744, the first manuscript to unify the separate books in Pi on the same subject.
Cf. Cameron, App. 1-2.

Apographs
Apographs, or more precisely selections, of poems found in P (and unknown to PI)
were made by Claude de Saumaise (Salmasius) and other scholars for their own
and for their friends' use before the larger collection of the Anthology was pub-
lished. Some were made from P directly; others are copies of other apographs con-
taining scholarly conjectures. These apographs traveled widely throughout Europe,
picking up further scholarly conjectures along the way. As a result, ascription of
conjectures to early scholars is a hazardous business, especially to Saumaise. Cf.
HE l.xliv £; J. Hutton, The Greek Anthology in France (Ithaca 1946) 8-11; R.
Aubreton, "La tradition de VAnthologie Palatine du xvie au xviie siecle," Rev. d'hist.
des textes 10 (1980) 1-52, 11 (1981) 1-46; E. Mioni, "L'Antologia Greca"

App.B-V The Appendix Barberino-Vaticana contains three of the six lines of one
Philodemean epigram (11). It exists in three manuscripts: (i) Par. suppl. gr. 119
no. 2, written 1480-1500. (ii) Vat. gr. 240," written ca.1560, and (iii) Vat. Barb. gr
52 Introduction

123,* written 1504-1509. See further L. Sternbach,AnfhologiaePlanudeae Appen-

dix Barberino-Vaticana (Leipzig 1890); and Cameron, ch. 8, who shows that the
source for these manuscripts was close to but independent of AP. Since almost all
its epigrams are erotic and missing from API, it was undoubtably intended as a
supplement to the latter (and hence dating from the fourteenth century), since
Planudes confessed to having omitted poems tending
(intro.
to API 7, t. 68v).

Ap. Voss Leiden Vossianus gr. O8, saec. xvi fin., containing 3, 6, 11, 14, 19, 26
(twice), 28. Cf. Hutton 8 f., 252-254; Aubreton (1980) 5-15; K. A. De Meyier,
Codices Manuscript!. VI. Codices Vossiani Graeci (Leiden 1955) 208 f. See further
the introduction to 6. This manuscript contains two sure conjectures: 6.1, 2, 19.3

Ap.B Gottingen philol. 3; Paris Suppl. gr. 557.* Copies associated with Jean
Bouhier (1673-1746), one in Gottingen, others in Paris. The original seems to
have the work of Saumaise. Cf. Verzeichniss der Handschriften im Preussischen
Staate. I.Hannover. 1. Gottingen (Berlin 1893) 2 f.; Hutton, op. cit., 523-526;
Stadtmiiller viii-x. Brunck made good use of its learned notes (some by J. G.
Schneider). This apographon Buherianum contains Phil. 3, 6, 11, 14, 19, 20, 22,
26, 28, 34.

Gottingen Philol. 6 "Epigrammata exscripta (a. 1758) ex codice bibliothecae

Dresdensis, qui sumptus est ex eclogis anthologiae Isaci Vossii. notas hie adscriptas
in codice reperi." Containing 11, 14, 20,22, 23,26. Cf. Verzeichniss (cit. supra) 4 f.

Ap.L Leipzig Rep. I fol. 55. An apograph owed either to I. Voss or Friedrich
Sylburg used by Reiske, who saw it in Leipzig; Hutton, op. cit., 8 f.; Aubreton (1980)
15-20. This manuscript contains Phil. 11, 14, 20, 22, 26, 34, [38]. Cf. G. R.
Naumann, Catalogus Librorum Manuscriptorum qui in Bibliotheca Senatoria Civitati
Lipsiensis asservantur (Grimma 1838) 4 (no. 4). Now housed in the Universitats-
bibliothek Leipzig.

Leipzig Rep. 1.35 "Exempla ex Anthologia Graecorum Epigrammatum quae est

in bibliothecae Is. Vossii." Cf. Naumann (op. cit.) 4 (no. 5). Containing 11,14,23,
26. Now housed in the Universitatsbibliothek Leipzig.

Cr Hamburg philol. 5. (1716). Another apograph used by Reiske, also containing

the spurious [38] as its last poem, as well as 6.7-8,14,19, 26,28; see the commen-
tary to [38]. Cf. H. Omont, Catalogue des manuscrits grecs des bibliotheques des villes
hanseatiques (Leipzig 1890) no. 17, pp. 10 f.

Leiden B.P.G. 34B, saec. xvii inc., in the hand of J. J. Scaliger, containing 14, 20,
22, 26, 34. Cf. De Meyier, op. cit., 51; Aubreton 20-23.
 The Greek Anthology 53

Papyrus
II P.Oxy. 54 (1987) 3724, ed. P. Parsons, written in the later half of the first cen-
tury A.D., containing 175 incipits, probably all of epigrams, at least twenty-seven
of which belong to poems by Philodemos. In the case of epigrams already known,
I print the incipits in the respective apparatus critici, and the entire list is printed
after the complete epigrams along with a commentary.21 See further below,
pp. 203-205.

III. Printed Editions

I have examined the following early printed editions, all of which derive horn API.
(Jacobs's edition of 1813-1817 is the first to print the epigrams of AP.)

1494. Florence, Alopa. Ed. J. Lascaris.

Cf. J. Hutton, The Greek Anthology in Italy (Ithaca 1935) 117 f.; on the
typeface, R. Proctor, The Printing of Greek in the 15th Century (Oxford 1900)
78 f.
1503, 1521, 1550-1556.
Diversorum Epigrammatum etc. The three editions published by the Aldine pres
in Venice, the first of which, edited by Aldus himself, was set up from a copy
of Lascaris' editio princeps now preserved, with Aldus' instructors to the print-
ers, in Paris (Hutton, GA in Italy 148 f.). The second edition was edited by
F. d'Asola, probably from the same annotated edition of Lascaris. Differences
among the three editions (which I collated at the Morgan Library) are incon-
sequential.
1519. Florilegium Diversorum Epigrammatum
Florence. Per heredes Ph. luntae.
1540. In Graecorum Epigrammatum Libros IV Annotationes longe Doctissimae, iam
Primum in Lucem Editae. Basel. Ed. V. Obsopoeus.
1549. Epigrammatum Graecorum Libri VII Annotationibus ]. Erodaei Illustrati.
Basel, S. Gelenius.
1550. Apud P. et J. Nicolinos Sabienses.
1566. Geneva, H. Estienne.
Florilegium Diversorum Epigrammatum Veterum. Cf. F. Schreiber, The
Estiennes (New York 1982) 143 f. Hutton, GA in France 132 notes that
Estienne's "text is probably founded on that of Badius [1531, not seen], but is
treated by Estienne with the utmost freedom."

21. Incipits (Graece serve to identify a work if a more formal title is lacking;
cf. E. Nachmanson, Der griechische Buchtitel. Goteborgs Hogskolas Arsskrift 47.19 (1941) 31-
49 (repr. separately Darmstadt 1969).
54 Introduction

1600. Frankfort, A. Wechel. Epigmmmatum Graecorum Annotationibus Joannis

Brodaei necnon Vincentii Obsopoei et Graecis in pleraque Epigrammata Scholiis
Illustratorum. The text is based on the Stephanus edition. The anonymous
scholia alluded to in the title (Z in the sigla) probably derived from
M. Musurus (Hutton, GA in Italy 155-158).

Because of the great continuity from one printed edition to the next, I record
only the earliest occurrence of peculiar readings. Thus, "edd. vett. (1494)" indi-
cates that Lascaris was followed in this particular reading by later editions. Note in
particular the apparatus to 9. For greater precision in citing these editions, cf.
Stadtmuller's Teubner edition.

About This Edition

The arrangement of the epigrams in this edition
Since it is impossible to know in what order the epigrams were written or arranged
within their first, or indeed any, publication during Philodemos' lifetime (espe-
cially since I find suspect attempts to interpret them autobiographically), a mod-
ern gathering of his or any other epigrammatist's poems must either be arbitrary
or satisfy some useful scheme devised by the editors. Gow-Page reasonably chose
to follow, making allowances for mistakes and reinterpretation, the general scheme
of the Palatine Anthology: Erotic, anathematic, epitymbic, epideictic, protreptic,
sympotic, and scoptic (HE l.xlvii f.). Their numbering quickly became standard,
and were I editing only the same twenty-nine epigrams they published under
Philodemos' name in GP, I would be happy to retain their order. But since there
are thirty-eight epigrams included here, some new order and numeration had to
be devised.
Making no historical claims, then, I print the epigrams in the following order:

1-8 The Xanthippe cycle, the poems either naming Xanthippe or seeming to refer
either to her or to the marriage with her, arranged, for want of a better scheme,
in what can be taken to be a dramatic chronology of the relationship.
9-26 The remaining erotic poems, arranged in no particular order, except that the
two Demo poems (10-11), the two street encounters (20-21) and the two spo
ken by a woman (25-26) are placed together. It is not sure that 24 is erotic.
27-29 The invitation poem to Piso and two other poems which also seem to
reflect on life in Campania.
30-34 Miscellaneous.
35-36 Doubly ascribed; Philodemean authorship can be neither proved nor
disproved.
[37-38] Not by Philodemos.
 About This Edition 55

Transliteration
For the most part I avoid Latin forms for Greek names but find that I remain more
comfortable with Plato, Epicurus, and a few others than with Platon, Epikouros,
etc. My citations of their works may be by Greek (e.g., Aristophanes' Ekkl.), Latin
(Hesiod's Op.), or English (Clouds) forms.

Abbreviations and bibliographical references

References to classical authors and their works should be obvious; they are usually
the same as or fuller than those in LSJ (but see paragraph above). Those to modern
works are either complete and immediately clear, or they can be deciphered with
the aid of the list of abbreviations. The bibliography, it should also be noted, is
limited to works on Philodemos (primarily his epigrams and literary theory), epi-
grams in general, and the Greek Anthology. Other books and articles are cited only
in their appropriate sections. The briefest of references in my book are of the form
"West on Hes. Th. nn.," which entails that someone named West wrote a (prob-
ably separately published) commentary, with or without text, on Hesiod's Theogony.

Emendations
New to this edition, emendations have been suggested as follows: 4.8 6.5
13.3 15.5 1 9 . 3 3 0 . 3 32.2
%op86KoX.a.

The translations
The translations of Philodemos' epigrams appended to the texts come with the usual
academic disclaimer to any esthetic value. Philodemos regularly shows up in trans-
lations of Selected Epigrams, but it is easy to single out the especially attractive ver-
sions by Sandra Sider of six poems: CO 61 (1984) 79 f.
This page intentionally left blank
CONCORDANCES
 Concordance of Printed Editions
AG Sider GP Kaibel Gigante Brunck-Jacobs

5.4 7 1 9 11 17
5.8 36 Meleager 69 — — —
5.13 9 2 16 2 18
5.24 13 Meleager 4 1 — — 11
5.25 15 3 5 7 16
5.46 20 4 1 — 3
5.80 2 Plato 5 FGE — — Plato 4
5.107 23 5 7 4 20
5.112 5 18 19 15 19
5.113* 37 Argentarius 9 GP — — Arg. 15
5.114 18 Maccius 1 GP — — Mace. 4
5.115 10 6 3 8 2
5.120 26 7 17 — 5
5.121 17 8 14 5 10
5.123 14 9 4 6 7
5.124 16 10 6 1 15
5.126 22 25 p. XXV — 8
5.131 1 11 10 10 13
5.132 12 12 15 — 21
5.145* Asklepiades 12 HE — Ask. 4
—
5.150* Asklepiades 10 — — Ask. 14
5.306 25 13 18 — 6
5.308 21 Antiphilus 14 GP p. vii — 4
6.246 35 Argentarius 18 GP p. xxvi — 27
6.349 34 19 24 16 25
7.222 33 26 21 22 31
9.412 29 20 23 23 30
9.570 3 14 12 12 32,34
10.21 8 15 8 3 24
10.103 32 24 p. xxvi 17 29
11.30 19 27 20 — 12
11.34 6 21 13 13 22
11.35 28 22 p. xx vii 19 23
11.41 4 17 11 14 14
11.44 27 23 22 18 33
11.318 31 28 p. xxvi 20 26
12.103 24 Anon. 56 HE p. xii — —
12.173 11 16 2 9 1
16.234 30 29 p. xxvi 21 28
38* 9
*Not by Phi lodemos.
Printed in the coramen tary to see below, pp. 215, <>20.
PP. 144 f. :Brunck (Jacc>bs prints the last distich as a s<;parate poem}.

58
 Concordance of Manuscripts
p PI n
5.4 VII 88* fol. 72v (om. 5-6) iv.10
5.8 VII 89* 72v —
5.13 VII 93* 72v (om. 3-4, 7-8) vii.25
5.24 VII 91* 72v iv.17
5.25 VII 92* 72v —
5.46 — vii.15
5.80 VII 87* 72v iv.31
5.107 VII 184 75v vii.13
5.112 VII 94* 72v v.ll
5.113 VII 95* 72v —
5.114 VII 96* 72v —
5.115 VII 97* 72v vii.7
5.120 — viii.9
5.121 VII 98* 72v ii.19
5.123 — v.3
5.124 VII 90* 72v —
5.126 — ii.18
5.131 VII 99* 73 r v.14
5.132 VII 100* 73 r v.20
[5.145 VII 116 73v vi.18]
[5.150 — iv.28]
5.306 VII 86* 72v v.13
5.308 VII 101* 73 r vi.4
6.246 VI 5 61v —
6.349 — iv.19
7.222 Ilia 11,11 34r iv.18
9.412 la 36,12 lOr vii.21
9.570 — iv.7
10.21 la 30,5 8r viii.2
10.103 la 88,5 20v ii.21
11.30 — iii.7, v.31
11.34 — ii.5
11.35 — vii.17
11.41 lib 22,14 89v (om.3,7-8) ii.14
11.44 — iv.4
11.318 lib 4,1 87 r —
12.103 VII 194 76r ii.28
12.173 —
—
16.234 IVa 8,89 49v

* In the series; VII 86-101, sixteen poems attributed to Ph ilodemos in PI:

+ 15 instances of See above, pp. 50 f.

59
This page intentionally left blank
 THE EPIGRAMS

Sigla

P = Palat. 23 Heidelberg
J = codicis P partim librarius, alibi lemmatista
C = codicis P corrector
PI = Venet. Marc. 481
App. B-V = Appendix Barberino-Vaticana
ac = ante correctionem
pc = post correctionem
s.a.n. = sine auctoris nomine
Ap. = apographum
Voss = Lugd.-Bat. Vossianus gr. O8
B = Getting, philol. 3 et Paris. Suppl. gr. 557
L = Lips. Rep. I fol. 55
Cr = Hamburg, philol. 5

= P.Oxy. 3724

ed(d). vet(t). = editio(nes) vetus/veteres

= scholia recentiora in ed. vet. 1600

61
62 Epigram 1

Text, Translations, and Commentary

AP 5.131 [11 GP, lOKaibel, 10 Gigante]

P PI 7.99, f. 73 14
[C]
;PP1: PI: P 4 CP1: P

The harp playing of Xanthippe and her talk, her expressive eyes and her
song—and the fire within her just now beginning;
these, my soul, will enflame you. The reasons why or whence or how I do
not know; but you will know, ill-fated soul, that you are burning.

Del Re, Epigrammi greet 82, 129.

Stella 271 f.

This would seem to be the first poem in the Xanthippe cycle, at least dramatically;
it need not have been the first to be written. Since Catullus almost certainly echoes
this poem (see on w. 3-4), it was written before his death ca. 55 B.C. Somewhat
similar is AP5.51 (Anon. 8 FGE)

] Very similar language in 4.5, which also, I

argue, pertains to Xanthippe.
The harp, which was plucked with fingers, rather than the kithara
(which was usually struck with a pick), seems to have been the standard accom-
paniment for female singers at dinner parties (cf. 4, 6) and elsewhere (cf. 3). For
the distinction between harp and lyre, cf. M. Maas and J. Snyder, Stringed Instru-
ments of Ancient Greece (New Haven 1989) 219, n. 1; and for the harp in the Clas-
sical period, ibid. 40f., 147-155,181-185. Romans seem to have been of two minds
about the instrument: on the one hand, it could be thought of as the musical accom
paniment to wild if not orgiastic parties (cf. esp. 6); on the other hand, it was the
instrument of Sappho and the Muses (Maas-Snyder 40, 148 f.; cf. Caesius Bassus 1
 Epigram 1 63

Morel Calliope princeps sapientipsallerat ore). Quintilian 1.10.31 has to argue on

its behalf that necpsalteria . . . uirginibus probis recusanda.
The reading of the papyrus cannot be paralleled; the plural should refer to the
playing of the harp by more than one person (Telestes 810.4 PMG, Phryn. 3 F 11
TrGF, Diogenes Athen. 45 F 1.9 TrGF). The more general sense wanted here, "th
sound of the harp," calls for the singular; cf. Aisch. fr. 57.7 ,
Pi. fr. 125.4 Snell (Terpander invented the barbiton)
A carelessly written lunate sigma was read by the scribe of as
an iota. (M. Gigante SFIC1 [1989] 143 n. 55 prefers the reading of , comparing,
irrelevantly, 6.1 .)
Propertius praises his docta puella's ability with the lyre: 2.1.9 lyrae carmen digitis
percussit eburnis, 2.3.19, 1.3.42.
Meleager refers to men who "speak" erotically only through
their eyes: 85 (AP 12.122.4) , 91 (AP 12.63.1), 108 (AP
12.159.3 f.). Cf. also Headlam-Knox on Herodas 1.40, Ov. Am. 2.5.17 non oculi
tacuere tui.
Conversation was highly regarded, especially in sympotic and/or erotic
contexts; cf. 26.3. On the role of conversation (and silence) in Epicurean company,
cf. F. Amoroso, "Filodemo sulla conversazione," CErc5 (1975) 63-76, an edition
of Phil's (P.Herc. 873).

2 For the significance of this name, see Introduction, pp. 36-38.

Heat is a frequent concomitant of erotic sensation; cf. Sappho 31
,Alkman59(a)PMG Soph.fr.
474.2 f., Kallim. Ep. 11.5 HE6 Phil. 10.6
Argentarius 4 GP (AP5.89.5) , Nisbet-Hubbard on
Hor. O. 1.33.6; N. Zink, Griechische Ausdruckweisen fur Warm und Kalt im
seelischen Bereich (Diss. Mainz; Heidelberg 1962) 75-90; W. R. Smyth, CQ 43 (1949)
122 ff. Phil, himself has4.6 and 16.4
10.6, 14.6.
Although this word on occasion has religious or philosophi-
cal overtones (amply documented in LSJ), it is also frequently a synonym for the
simplex in Phil.'s prose, and probably is so here as well. Cf., e.g., Phil.R/?e£. 2.34.14-
16 Sudh. For the thought
in general, cf. Prop. 1.9.18 haec est venturiprima favilla mali.

3 Where Homer and other early poets had people address their thumos,
fifth-century and later poets substituted the psyche (indeed, the scholia to Pi. O.
2.89 "interpret" as : e.g., vi.19, Aisch.(?) Septem 1033, Soph. Tr.
1260, Eur. Alk. 837. (Note that for Simon. Elegy 21.3 W2
West, ZPE 98 [1993] 11, now prefers
Such addresses usually signal a soliloquy (C. Hentze, "Die Monologe in den
homerischen Epen," Philologus 63 [1904] 12-30), which is of obvious importance
in drama; cf. F. Leo, "Der Monolog im Drama," AGAW 10.5 (1908) 94 ff.; W.
Schadewaldt,Mo«o/og undSelbstgesprach. NeuePhilol. Unters. 2 (Berlin 1926) 201,
212-217. Here, however, Phil, alludes to a division between a rational self, the
poem's narrator, who speaks like a philosopher (though not an Epicurean one; see
64 Epigram 1

below) and his emotional soul. The former maintains, or perhaps would like to be
seen to maintain, a sang-froid which should allow it/him to keep its distance from
the passion-ridden soul. The readers, I suspect, are invited to doubt whether the
rational narrator is being honest with himself and with us.

3-4 The cause is unknown; the feelings overwhelm, just as in Catullus 85:

Odi et amo. quare id faciam fortasse requiris.

Nescio, sed fieri sentio et excrucior,

which is a neat distillation of the thought found in Phil, and a far better poem;
cf. Stella. Catullus may also have acknowledged his borrowing by having odi et
amo echo somewhat . For Catullus and Phil, in general, see above,
pp. 23 f. Jacobs also aptly compares Prop. 2.4.9 f.

. . . nee causas nee apertos cernimus ictus:

unde tamen veniant tot mala caeca via est.

4 Taken as circumstantial by Gow-Page ("y°u will learn [sc. the answer

to these questions] . . . when you are afire"), Beckby ("wenn"), and Gigante
("mentre"), but knowledge of these Aristotelian categories and causes will not likely
come in the middle of intense passion. Better is Waltz's rendering as o.o.: "mais ce
que tu verras bien, malheureuse, c'est que tu en es consumee," the grammar of which
is also reflected in Catullus' fieri sentio. That is (to pursue the Aristotelian thread
picked up above), although Phil, and his soul will be incapable of giving a scien-
tific account, they will certainly recognize its essence, which is the most important
of the categories:
(Cat.2all).
And if they do not know they will certainly be aware of (cf., e.g.,
NE 1095b6-7).

^P5.80[Plato5FGE]
PI 7.87, f.72 P
3.32 31

An apple am I. Someone who loves you sends me. Do but give a nod,
Xanthippe. Both you and I are wasting away.
 Epigram 2 65

Ludwig, GRBS 4 (1963) 4 (1963) 75-77.

Mariotti, Studi Urbinati 41 n.s. B 1-2 (1967) 1073-1078.

Although Platonic authorship was disproved by Ludwig, doubt remained about

Philodemean authorship (Page, e.g., preferred to classify it as a work of Pseudo-
Plato). This should be now completely dispelled by inclusion of the poem's incipit
in (see comm.).
There is a companion piece within the Pseudo-Platonic corpus of epigrams (4
FGE=,4P5.79):

This epigram is given to Plato by P and Diog. Laert (3.32), it is anonymous in PI

(f. 76r), and is absent from 1 . Mariotti shows how2 can be read as a concise para
phrase of 5.79, i.e., that the shorter poem was written in response to the longer
one, which is reasonable. He also argues that since it is a better poem it must be
by another hand, which does not necessarily follow; epigrammatists developed
their own themes as well as those of others. We do not know where Aristippos,
the compiler of the "Platonic" corpus of epigrams, found 5.79, but since there is
nothing reminiscent of Plato about it, he may well, as Ludwig suggests, have found
it in the same place as 2, which would make Philodemean authorship of the former
a possibility. The theme of 2, like that of all exx. of the go-lovely-rose topos, fits
in well with the Epicurean idea that one should enjoy the one life we have; see,
e.g., on 29. One does not, however, have to be an avowed Epicurean to profess
this theme. A metrical anomaly, moreover, argues against Phil., namely a viola-
tion of Hermann's Bridge (1 which Phil, scrupulously observes in
the undoubted poems except in 20.1 (a dialogue poem) and in 22.1, where met-
rical roughness is intentional.

1-2: This apple, not having read its Denniston, is sparing of connective particles.
The first sentence can be considered a simple label, and and
can easily be understood as explanatory asyndeta, which we find elsewhere in Phil.
The overall effect is one of artful simplicity. On explanatory asyndeton, cf. KG 2.344
f., Verdenius on Hes. Op. 211.

1 On the as a love-token, cf. A. R. Littlewood, "The symbolism of

the apple in Greek and Roman literature," HSCP 72 (1967) 149-181, esp. 154 f.
(exx. of , to which now add Sappho S260 SLG the ear-
liest example), 167 f. (inscriptions on apples, the earliest example of which may
be Hes. fr. 214 M-W). In addition to the literary sources cited by Littlewood, a
magic charm involves the throwing of an apple: Supp. Mag. 72 (= PGM 122 Betz)
col. 1.5-14, beginning cf. C. Faraone,
CP90(1995)9f.
66 Epigram 2

As Alan Cameron has pointed out (Greek Anthology 385-387 and ap.
Parsons's comm. on ), the Latin translation of this poem (Epigr. Bob. 32) sup-
ports this reading over , the farfacilior reading:

malum ego: mittit me quidam tibi munus amator.

adnue: marcendum est, ut mihi, Flora, tibi.

Another possibility is that there were two'poems by Phil., one whose incipit is in
and translated in Epigr. Bob, the other the one found in P, in which case
should be retained here.
Note the name Flora, which also occurs in 12 and two other Latin erotic
poems: Varro, Menippeae fr. 136 and Juv. 2.49. Mariotti 1078, giving credit to the
Latin translator of 2, notes that Flora is an improvement over Xanthippe in that
now the girl's name is part of the message of the poem: flowers waste away like
apples. But, as suggested just above, any such credit may be due to Phil.
Cf. the story of Akontion's sending a message along with an apple to Kydippe:
Kallim. fr. 67 with Diegesis and Pfeiffer's n. But, as Mariotti points out, Latin mittere
can be the equivalent of , as in, e.g., mittere tela. Note in particular Ovid,
Her. 21.107 mittitur ante pedes malum cum carmine tali; with with contrast Vergil,
Eel. 3.71 aurea mala decem misi, where throwing seems unlikely. Similarly, Ps.-
Petronius (Anthol. Lat. 218) aurea mala mihi, dulcis mea Martia, mittis. See D. A.
Schmidt, CQ, N.s. 37 (1987) 21, who points out that in Pindar and Bakchylides
(a poem) = "bring" or "present." In the epigram, the object of the verb is
an apple, but if the poem accompanied the poem or were inscribed on it, the two
objects would merge into one. Compare the apple thrown by Eris at the wedding
of Peleus and Thetis inscribed with the words "Let the beautiful woman take (it/
me)" (Luc. Dial. Mar. 5); Littlewood 167 f.
The silent nod in answer to the written message on the apple sug-
gests the possibility that a secret assignation is hereby being arranged.

2 Although PI (or a source) could have erroneously ascribed this poem

to Phil, on the basis of this name alone (just as someone found the presence of the
name useful for ascribing it to Plato), this is hardly likely for the compiler of the
largely Philodemean incipits, who presumably had solid evidence for authorship.
It should also be noted that the name Xanthippe is not very common in poetry, all
exx. coming from epigrams:
(i) Xanthippe is one of three Bacchants in Ps.-Anakreon 5 FGE (AP 6.134):

There is no reason to claim this anonymous poem for Phil., but note that as par-
allel for the dislocation of copulative K«i on v. 4, Page can cite only 5.5
 Epigram 3 67

(ii) In an epitaph ascribed to Simonides (36 FGE = AP 13.26), Xanthippe

appears as the wife of Archenautes and daughter of Periandros.
(iii) Epigr. Bob. 35 offers an erotic Xantho, one of Phil.'s nicknames for
Xanthippe:

Musarum Xantho decimast, Cytherea secunda,

quarta Charis: Xantho Musa, Venus, Charis est.

Because of the general rarity of the name, in addition to the fact there is already
another translation from Phil, in Ep. Bob., it is possible that this too is a translation
of a poem by Phil. Cf. Mariotti 1086-1093 for a survey of the topos of the tenth
Muse, fourth Grace, etc. (cf. on IT iv. 26). Ep. Bob. 35 comes closest to to AP 5.95
(Anon.):

Xantho also appears as a nymph in Vergil, G. 1.336, where R. Thomas, "Virgil's

Georgics and the art of reference," HSCP 90 (1986) 190 f., detects an allusion to
Phil.'s Xantho.
The same form and general sense at 19.6, q.v. Mariotti calls this
an atemporal present, saying that it is too subtle to have it mean that we are wasting
away every moment of our lives. This notion, however, was not too subtle for
Herakleitos and appears in Epicurean atomism; cf., e.g., Lucretius' example of the
rings and statues which have imperceptibly worn away over the years, 1.311-319.
Here of course it is the flower of youth which wastes away, for ich poetry offer
many parallels; e.g., Archil. 196a.24-28 W2, where Neoboule rejected beca
her charts is now gone , Aisch. Suppl. 998

5
68 Epigram 3

AP 9.570 [14 GP, 12 K, 12 G]

P iv.7 caret PI
Huschke: P 3 (mare.):
pac. Brunck: Gieante P Schneider 4
Kaibel: P: Schneider 7 Salm,
P Chardon: P: Salm. 8
Chardon: P: Salm

<Man.> Xantho—formed of wax, with skin smelling of perfume, with the

face of a Muse, of splendid voice, a beautiful image of the double-winged
Pothoi—
pluck for me with your delicate hands a fragrant song: "In a solitary rocky
bed made of stone I must surely someday
sleep a deathlessly long time." Yes, yes, Xantharion, sing again for me this
sweet song.
<Xantho. > Don't you understand, man, you accountant you? You must
live forever, you wretch, in a solitary rocky bed!

VH 1 (1793) 4-8 (ed. D. Carlo Rosini).

Chardon de la Rochette, "Deux epigrarnmes," 209—222.
Del Re, MC 6 (1936) 132 f.
Huschke 145-150.
Kaibel, Hermes 15 (1880) 460.
Luck, EH 14 (1968) 406^08.
Merkelbach, RM 115 (1972) 219-222.
Merlan, Z. Philos. Forsch. 21 (1967) 490 f.
W. Schmid,Acta ConventusXIEirene (Warsaw 1971) 201-207; repr. 'mAusg. Philol. Schr. (Berlin
1984) 267-274.
Sider, AJP 108 (1987) 317-319.
de Aries, Mnemosyne 23 (1970) 30 f.

The Xantho of this poem is and accompanies herself on the harp, just as
Xanthippe does in 1. The narrator, on the other hand is a pathetic character: moan-
ing about death and asking Xanthippe to provide an appropriately gloomy song
(which he calls sweet). Xanthippe gives him his song in altered form, exaggerating
the already ridiculous "sleep an immortally long time" to "live forever (in the grave)"
to point up the illogicality of the man's thinking: if he is sleeping he is alive; if he is
alive he is not dead. And so all his maudlin posturing comes to nothing. We need
not dwell on how much better an Epicurean Xanthippe is than the man. Note,
however, how the form of the poem parallels that of 29, where I think that again
the last two lines are spoken by another in order to bring the first speaker back to
his Epicurean senses.
"Among the curiosities of Greek scholarship" (Gow-Page), dating from VH 1
(1793) 6, is the belief of some that Xantho, the "blonde one," was not human but
a bee: Ardua quaestio est, apimne alloquatur poeta, an Xanthonem psaltriam, wrote
Diibner, although Huschke and Jacobs had earlier tried to dismiss the notion.
Ruhnken went so far as to consider altering 1 . Thus
 Epigram 3 69

was taken in an active sense and was under-

stood to apply to Xantho as well as the Pothoi; see further Chardon. Kaibel finally
laid the matter to rest, but introduced the idea that the last two lines were due to
iram rabiemque byzantini hominis, convincing among others Gow-Page, who shunt
the final distich, without apparatus, to their commentary. Schmid ably defended
its authenticity, showing how, as Xantho's reply, it offers an Epicurean corrective
to the man's plaints. Note the parallels between this poem and the diatribe against
the fear of death in Lucretius, book 3, not only the similarity between the chiding
tones taken by Natura and by Xanthippe towards those with the wrong attitude
(cf. Kenney on Lucr. 3.894-899), but also the several rather specific points of com-
parison indicated below in the appropriate lemmata. It should also be noted that
the last two verses are in harmony with Phil.'s regular metrical usage.

1 Huschke's reading has been confirmed by For the adj.,

cf. Hor. 0.1.13.2 f.cerea Telephi \ laudasbracchia, PlinyNH37.33 candidumatque
cerei colons. As Nisbet-Hubbard point out on Horace, the yellowish color of wax
was not normally a sign of beautiful skin (lilies and milk being more usual for the
pale end of the spectrum). The primary emphasis here, however, maybe on Xantho's
doll-like quality; cf. Hist. Alex. Magn. Rec. e 2 (ed. Trumpf) Thus,
PL Ti. 74c refers to the Demiourge as a cf. Euboulos fr. 41.1 f. 6 . . .
(Waltz suggests "douce au toucher comme de
la cire," but wax has an unappealing tacky quality.) Bee fanciers should also note
Soph. fr. 398.5 although for the first word
there are variants, including one missing from TrGF vol. 4: Basil, Epist. 8.12.
Hapax legomenon. Men and women often perfumed their hair (cf.
e.g. Eur. Kyk. 501 (uup6%picno<;, another hapax), Aristoph. Ekkl. 524, PI. Rep. 398a),
but wider application was always possible; cf. Alkaios 50.1-2
Anakr. 18 PMG
means perfume in general rather than myrrh in particular.
Another hapax. If the word has any special meaning beyond
" divinely beautiful," it would be " capable of producing poetry worthy of the Muses "
Chardon 212 f. (Muses could be portrayed in art as white-haired women, but pre-
sumably this is not what Phil, has in mind; cf. LIMC s.v. Mousai.) For the scansion
, see Intro, pp. 41 f.; similarly with 3 below.

2 Elsewhere in poetry only Meleager 33 (AP5.151.2), of mosquitoes.

Calling her a very image of the bipennate Pothoi is of a piece with
the other exaggerated terms applied to Xantho. Reducing the word to
, as Gow-Page do, not only robs it of its intended hyperbole, it makes
little sense with n60<av. The tendency of besotted husbands to treat their wives as
is derided by Hippolytos (Eur. Hipp. 630-633).
For the frequent association between Pothos (or the Pothoi) and Eros,
Himeros, Aphrodite, et sim., cf. Headlam ad Herodas 7.94-95. Pothoi do not have
to be winged, but cf. Sappho 22.11 f. , Phil. 8.2 (withcomm.),
Meleager7.5 (AP5.179)oei) [sc. .wherethe
wings properly belong to Eros. But note the comment of Pausanias 1.43.6: "And
70 Epigram 3

by Skopas there are statues of Eros, Himeros, and Pothos, if indeed their function
is as different as their names," and in fact from the Alexandrian period on there
was little distinction drawn between Pothos and Eros; note also that while Cupido
is the usual name for Eros, semantically it represents Pothos. Cf. Hofer, "Pothos,"
in Roscher, col. 2903. Cornutus, Nat. Dear. 25, p. 48 Lang, allegorically equates
Eros and Pothos. Winged "amorini" were often painted on the walls of Hercu-
laneum villas. Cf. F. Lasserre, La Figure d'Eros dans la poesie grecque (Paris 1946)
60-62, 220-227.

3 Almost certainly the noun is an internal accus. (so Jacobs,

Gow-Page, de Vries 30); cf. Persiuspra/. 14 cantare. . . nectar, and perh. Kallim.
Aitia prol. fr. 1.33 although other interpretations of
this complicated sentence are possible; cf. N. Hopkinson, A Hellenistic Anthol-
ogy (Cambridge 1988) 96 f. Some commentators take p/upov with
(Dindorf, Diibner, Waltz, Del Re): manibus rorantibus unguentum, but this would
be more naturally expressed with which Jacobs and Waltz were tempted
to read. Also unlikely is the possibility that is a vocative; cf. 37.3
(Not pertinent is Aristoph. Daital. fr. 205.1 K-A

could mean "glistening" or "dewy" (Gow-Page); cf. Hesych.

Romanus Mel. 9^16
Gigante's conjecture ("spalmami 1'unguento")—supported by Van Looy,
Rev.belg.philol.hist. 68 (1990) 173—would leave
without point.
Schneider's nu movable is unnecessary; Phil, allows initial plosive and
liquid to make position elsewhere; cf. Introduction, pp. 41 f.

3—4 Two rare adjj. modifying the rare

In epitaphs, whether real or putative, the material of the stele (which is
very much in Phil.'s mind) is often mentioned (Weisshaupl 54 f.), and the stele may
stand for the whole tomb; cf. API.700.1 (Diodoros Grammatikos)
but the chief aim here is to demonstrate his preoccupation with death.
A is a small rocky cave (Aristotle, HA 547b21, fr. 239 Gigon), here of
course in an extended sense. With , a hapax, cf.
was also thought a hapax until the publication of
P.Hibeh 172 (= SH 991.90), a list of compound poetic adjectives. We should ac-
cordingly note the comments of its editor, E. G. Turner: "The proportion of new
words is a reminder of how much Greek poetry is lost. It is interesting that some of
these words are known only because they are used by Hellenistic poets, almost
certainly at a date later than the compilation of this list [ca. 270-230 B.C.]" (p. 2).

4 The best restoration of meter and sense. which de Vries fa-

vors, is also possible, although the usual meanings of this combination—"at last,"
"at one time," "ever"—are not appropriate. If read, could emphasize after
the intervening word; cf. Denniston GP208, 227 f.; Diggle ad Eur. Phaethon 96 f.
Luck argues for Schneider's regarding the other conjectures as padding,
but, despite de Vries' objections, I find the repetition of 8ei in line with the overall
 Epigram 3 71

maudlin tone. Cf. below 6 vcd vai and note esp. Lucr. 3.894 tarn iam non domus
actipiet te laeta, in a very similar context.

5 The adv. (which, as Kaibel observes, must

modify cf. 31.6) is rare; cf. Ps.-Clemens, Homiliae 13.8
Since the author has just described his thorough ground-
ing in Greek thought in general and "godless" Epicurean (and Pyrrhonian) phi-
losophy in particular, it is possible that he is here recalling something he read in
Epicurus. Cf. Del Re 132, "quell' che ha poi qui, sulla bocca dell'epicureo
Filodemo, un saporeironico, e ci fapensare allucrezianomorsimmortalis" (3.869;
cf. 904letosopitus, ^2\aeternum soporem). GiganteRP 164 adduces Amphisfr. 8
K-A
Cf. Cat. 5.6 nox est perpetua una dormienda. For the motif of death as
sleep (which is as old as Homer; e.g., II. 11.241), cf. M.B. Ogle, "The sleep of death,"
MAAR 11 (1933) 81-117; B .P. Wallach, Lucretius and the Diatribe against the Fear
of Death: De Rerum Natura III 830-1094 (Leiden 1976); D. Puliga,
in Epicure," Elenchos 4 (1983) 235-260 (esp. 258 ff.). Cf. Theog. 567

6 Phil, uses both name and nickname in the same poem also in 11
. Cf. Catullus 12 Veranius, Veraniolum (and56Gz/o, Catullum,59Rufa,
Rufulum).
Repeated as at lambica Adesp. 57 West
Glaukos 3 HE CAP 9.341.3), Poseidippos 5 HE (AP 12.45.1), Asklepiades 17 HE
(AP 12.166.5).

7 Cf. Od. 1.298 (same sedes).

Schmid203 ff. offers several parallel exx. of what he calls "popular-
philosophische Mahnrede." Note in particular Lucr. 3.933 f. (Natura loquens) quid
tibi tan to operest, mortalis, quod nimis aegris luctibus indulges?; P.Oxy. 2.215 (=
Epicurus 11 CPF, ed. Obbink) col. 1.17
Polystratos De contemptu (ed. Indelli) col. 27.1. Indelli (and others)
speak of the anonymity and fictitiousness of the addressee (as a hallmark of the
diatribe style), but more important may be the emphasis that this particular address
lays on the addressee's human limits; cf. Simon. 521 PMG
Perhaps the most interesting parallel is Soph. Ai. 1150-
1158, where Teukros clearly invents someone "like me" to criticize Menelaos:
In the Anthology this vocative, in sepulchral of moraliz-
ing epigrams, always points up the mortality of the addressee: Leonidas 21, 33, 77
HE (AP 7.198, 736, 472), Dioskorides 22 HE (7.37), Automedon 12 GP (7.534),
and four of Palladas (10, 45, 77; 11.62, 300). Cf. Missale Romanum, Ash Wednes-
day: memento homo quia puluis es et in pulverem reverteris.
A pejorative term for a money lender, i.e., a usurer, which Pol-
lux 3.85 lists along with some synonyms; for Phil.'s metaphorical usage, cf. Luc.
Vit. Auct. 23 ( a notoriously stingy person)
Plut. Mor. 18e
72 Epigram 3

Hence my not entirely satisfactory "accountant," which is intended to

point up the man's excessive concern with the number of days remaining in his life.
A far more satisfying Englishing of the sentiment of this final distich is to be found
in Shakespeare's Sonnet 4; note especially 7 f.: "Profitles userer why doost thou use
I so great a summe of summes yet can'st not live?"
7-81 In repeating this phrase, Xantho makes the point
that in death he will occupy this and only this physical spot. Note the very similar
and particularly Epicurean point in a Latin sepulchral inscription dating from be-
fore ca. 50 B.C. (on metrical grounds):
Stallius Gaius has sedes Hauranus tuetur
ex Epicureio gaudiuigente choro.
(Courtney, Musa Lapidaria no. 22 = CLE 961 = GIL 10.2971 = ILS 7781).
8 Saumaise's emendation seems pedestrian alongside Chardon's
far neater suggestion. Xanthippe's point is precisely that made by Lucretius, who
takes to task those who express concern for what happens to their bodies after death
as if they were somehow going to be present; cf. 3.870 ff. proinde ubi se videas
hominem indignarier ipsum, \ post mortem fore ut aut putescat corpore posto \ aut
flammis interfiat malisve ferarum etc., 878 £, 885 ff. nee videt in vera nullum fore
morte alium se \ quipossitvivus sibi se lugere peremptum \ stansque iacentem <se>
lacerari urive dolere, 923. once again reminds us of Lucretius' mors
immortalis (3.869).
As Phil, puts it elsewhere, the man who lives his life thinking of death is al-
ready comparable to a dead man (DeMorte col. 38.17
f.). For this sort of paradoxical expression, cf. Epic. Ep. 3.135
.Lying
behind these expressions are the views of Demokritos: 55 B 160 DK

Cf. Lucr. 3.955 (Natura loquens) aufer abhinc lacrimas, baratre, et

compesce querellas. (For the uncertain sense of baratre, see the commentaries; one
possibility—"spendthrift"—suggests a rough contrast to but this can-
not be pressed.)

5
 Epigram 4 73

AP 11.41 [17GP.11K, 14G]

P PI 2b.23,14, f. 89v s.a.n.; om. w. 3, 7-8 :ii.!4
2 PI: P 3 spat. vac. relicto om. PI 4 Salm.: P: PI
6 PI: P.- Jacobs 8 scripsi: P: ' Hecker

Seven years are coming up on thirty; papyrus columns of my life now being
torn off;
now too, Xanthippe, white hairs besprinkle me, announcing the age of
intelligence;
but the harp's voice and revels are still a concern to me, and a fire smol-
ders in my insatiable heart.
Inscribe her immediately as the koronis, Mistress Muses, of this my mad-
ness.

Giangrande, GB 1 (1973) 141-148.

Griffiths, BICS 19 (1970) 37 f.
Sider,4)T 108 (1987) 315 f.

The first three hexameters are metrically unusual:

(i) SSDSDS (3) and DSDSDS (1 and 5) are the only exx. of these shapes in Phil.,
the former to be avoided because of its heavy spondaic quality, the latter
because of the repetition of the DS pattern.
(ii) These same three lines lack a bucolic caesura, the only three consecutive
hexameter lines to do so in Phil. (See above, p. 42.)

This oddness and heaviness mirrors the narrator's despondency, which is dispelled
by the thought that his life and verses will be cheered up by the presence of
Xanthippe: SDDDDS (7) is one of Phil.'s (and others') most common metrical shapes
(> 13 exx.).

1 Hesiod, who comes by the form honestly (cf. West ad Op.

696), and some very few imitators (Kallim. fr. 714.2 and two post-Philodemean exx.,
AP 14.3.9, 123.13) have the inflected genitive (or ); only Phil,
has an inflected dative.
Phil.'s learned audience is alerted to the possibility that marriage is the subject
of this poem, first, by the precise number thirty-seven, at which age Aristotle, Pol.
1335a29 argues a man should marry (and marry an "eighteen-year old woman"—
an incipit at iii.14, q.v.); and, second, by the oddness of the inflected numeral,
for Hesiod loc. cit. employs his inflected form in a passage urging ca. 30 as the best
age for a man to marry (695-697):
74 Epigram 4

Cf. Hor. O. 2.4.22-4 fuge suspicari \ cuius octavum trepidavit aetas \ claudere
lustrum, with Nisbet-Hubbard's n. for other poems in which the author/narrator
mentions his age.
Often of time (LSJ II 1), e.g. Od. 2.107 and
esp. the inscription (Chiron 17 [1987] 178,
col. 1.1).
Here, as usual, = "years," but because of its unknown etymol-
ogy and its significance in the Odyssey in the sense of either "month" or even "day,"
the literature on this word is extensive; cf. e.g., N. Austin, Archery at the Dark of
the Moon (Berkeley 1975) 244-246; D. N. Wigtil, "A note on " AJP 99
(1978) 334 f.

2 Regularly translated (or paraphrased in commentaries)

as a past tense (e.g., "already torn," GP), but the sense of + pres. is rather
that, at age thirty-seven, Phil, has now become aware that time is passing (as be-
comes clearer with v. 3); KG 2.121. —"divide a whole into parts" (LSJ),
e.g., Dsc. 2.70, milk into curds and whey—here in the passive refers to the part(s)
divided/torn.
The metaphor of the book of life, which Phil, uses in De
Morte IV 39.17 f. . Cf. D.L. 10.138
(refering to his own work, which ends with an account of Epicurus) •

cf. Gigante, RF2 233. (See further below on 7 ) Not quite the same is AP
7.21.5 f. (Simias 4 HE), addressed to Sophokles:

The are the columns of a papyrus roll (Giangrande), rather than the
pages of a codex (Gow-Page), the form of which was probably unknown to Phil.
Cf. C. H. Roberts and T. C. Skeat, The Birth of the Codex (London 1983), who show
that "the codex scarcely counted for Greek literature before about A.D. 200" (37).
In the Herculaneum papyri a could be limited to one charta = but
not necessarily; cf. W. Scott, Fragmenta Herculanensia (Oxford 1885) 14.
as often in early poetry. Construe with rather than as
gen. of separation with

3: This line was omitted in Planudes' exemplar owing to homoeoarcton; all that
Planudes could do was leave a space for the missing line, perhaps in the hope that
he would find it in another selection from Kephalas. (Kaibel mistakenly attributes
the error to Planudes himself.)
 Epigram 4 75

"Of two clauses linked by icai, the first sometimes gives the time or cir-
cumstances in which the action of the second takes place" (Denniston, GP293).
Cf.Thuc. 1.50.5

On the motif of white hair, cf. 3-4.

A metaphor from sowing seed; cf. Kratinos fr. 246.2 K-A
(sp ken by the ashes of Solon in his name:
The construction is odd—normal Greek would call for
—but is readily understandable.

4 Or so one hopes:
(Phil. Parrh. col. xxiv a8-
bl). and are common in Phil.'s prose. Cf. Phil. 5.3-4. Kaibel com-
pares Apollonides 27 GP (AP 11.25.6) fur-
ther suggesting that this author is the Stoic Apollonides of Smyrna entioned by
Phil. Index Stoic, col. lii. 3-4 Dorandi. The only connection betwee pigramma-
tist and philosopher, however, is the lemma to API 235, _ _ ,
which edd. emend to A but see GP 2.147.

5-6 Phil, appears at least at first like those old men Kephalos speaks
of in the Republic who bemoan the lost pleasures of youth:

(329a).

6 Raging desire for the Kephalos spoke of. Cf. 1.2 n.

Gow-Page follow Jacobs on the grounds that elision of is rare
in this position, but -cu is elided fourteen times in GP and more often in HE; and
there is elision in this position twenty-six times in GP. Correption, it is true, would
be more in keeping with Phil.'s usual practice, but this is insufficient reason to
condemn an elided here; cf. Intro., p. 41, Kaibel vi, xiv. Ancient grammar
ians were at odds over the accentuation of disyllabic prepositions when they fell
between noun and adj., some arguing that anastrophe always is called for, others
only if the noun comes first, and others only if (as here) the adj. precedes; cf.
McLennan on Kallim. H. 1.48.

7-8: A vexed passage, all the more so when tamTiv (cod.) is read. Either (i)
emphasizes or (ii) it stands for Xanthippe. Gow-Page argue for (i): "Write
this same Finis." First, it will not do to cite Homer's use of cm-cnv =
oamyv, as this usage is extremely rare elsewhere (Hes.5c. 35, 37; Pindar N. 5.1; not
in tragedy; cf. KG 1.630). Second, their understanding of "same" is strained: "here
is the end of my poem; let it mark also the end of my dissipation." Even spelled out
like this it makes little sense (similarly Giangrande 142). Better is the usual construal
of = hanc ipsam = Xanthippen (Diibner, Kaibel, Giangrande): The
Muses are to inscribe Xanthippe as the sign of Phil.'s new sanity. With this general
interpretation I too am in agreement (see below on but suggest the slight
76 Epigram 4

emendation of in order to obviate the hyberbaton of

Holford-Strevens suspects 7 ; perhaps , in which case PPl's in
8 may conceal something altogether different.

7 This particle, frequently preceding imperatives or exhortations (Denniston

GP 13 f.), came to lose its force as an adversative conjunction in their presence; in
the fifth century it could be repeated in oratio obliqua; e.g., Aristoph. Clouds 1364
(with Dover's
n.), Eur. Or. 1562. In later Greek poetry, a significant number of postponed
occur with imperatives or deliberative subjunctives: Kallim. Ep. 14.11 HE, fr. 110.61;
AP 5.17.5 (Gaeticulus). For other situations which allow postponement, cf. Den-
niston 13; in Hellenistic Greek, cf. Kallim. H. 1.18, fr. 260.55, AP5.9.7 (Rufinus 1
Page). Kallim. fr. 10 van Eldick (approb. Pfeiffer) may be
another ex.
The elaborated paragraphus used to mark the end of a poem (esp.
one in a series) or part of a poem (e.g., a choral ode in drama; cf. Hephaist. p. 75
Consb.), the section (e.g., a book) of a longer work, or the end of an entire work.
That is, it may or may not be followed by more poetry. For a description cf. Schol.
in Aristoph. Plut. 253:
See further G. M. Stephen, "The coronis," Scriptorium 13 (1959) 3-14, with plates
1-2; E. G. Turner, Gk. Mss of the Anc. World2, BICS Suppl. 46 (1987) 12 n. 59.
Many exx. of varying designs are found in the Herculaneum papyri (Stephen 8; G.
Ca.vaHo,Libriscritture scribia Ercolano [Naples 1983] 24), a particularly elaborate
one at the end of On Inference (cf. De Lacey and De Lacey's ed. p. 21 for an illus-
tration). Meleager 129 HE (AP 12.257), perhaps the last poem of his collection (see
Gow-Page's pref. to 129), is spoken by the koronis alongside:

(For another "speaking koronis," beginning

[P.Lit.Lond. 11], see T. C. Skeat, "The use of dictation in ancient book-produc-
tion," PEA 42 [1956] 183.)
Cf. Poseidippos 705.5 f. SH, also addressed to Muses:

(On text and interpretation, neither without problems, cf. Lloyd- Jones, Academic
Papers 2.167 ff.) For the metaphorical use of the koronis in Phil., see above on 2
, and cf. P.Herc. 1005, edited asAgli'AmzcidiScuola by A. Angeli (Naples
 Epigram 4 77

1988), fr. 77 "put an end to the earlier dis-

cussions"; fr. 34 P.Herc. 1428 col. 15.20-23
(De Piet.
p. 25 f. Henrichs); P.Herc. 1418Pragmat. col. 13.6£, col. 21.l3;Rhet. 1120.6-10;
etc.; D. Obbink, Philodemus on Piety (Oxford 1996) 89-94. Phil, is clearly fond of
this metaphor. See further P. Bing, The Well-Read Muse, Hypomnemata 90 (Gottin-
gen 1988) 33-35.
Phil.'s point is that Xanthippe is the koronis that marks the end of the manic
stage of his life, and as a living koronis she fits into the book of Phil.'s life, as de-
scribed at the poem's beginning (Giangrande 143). If she is to do this, she cannot
be simply the last of his loves from his earlier life; a koronis that has disappeared
cannot perform its proper function. Phil, certainly is not claiming that henceforth
(at age thirty-seven!) he will lead a celibate life (pace Philippson, RE 19 [1938] 2446
rp. in Studien). Xanthippe should rather be part of his new life in a way that dem-
onstrates that his old ways are over. She can do this, first, by being hispatrona virgo,
like Lesbia, Cynthia, et al. Note especially Hor. O. 4.11.31 ff. age iam, meorum \
finis amorum, \ . . . condisce modos, amanda \ voce quos reddas, Xanthippe can also
help Phil, by acting as his partner/wife in their common pursuit of Epicurean virtues.
If this poem was in fact accompanied by a koronis, it marked the end of the
poem (Gow-Page). For its being the end of a book of Phil.'s poetry (Griffiths) there
is no firm evidence; to the contrary, Phil, would rather seem to be asking the Muses
to help him in the writing of further poems about his new love. (See next lemma.)
Griffiths's detailed comparison with Poseidippos 705 SH (cited in part above)
proves little, as this poem almost certainly introduced rather than closed Poseidip-
pos' book; cf. Lloyd-Jones, Academic Papers 2.171, 190. On poetic closure in gen-
eral, cf. D. P. Fowler, "First thoughts on closure: Problems and prospects," MD 22
(1989) 75-122 (106 f. on Phil).
For the syntax, cf. Hdt. 7.214.3
Pi. O. 3.29 f. Hellenistic Muses have given up singing for writing: cf. Bing, op. cit.,
pp. 10-48 [= ch. 1]. Phil, none the less maintains the fiction that it is the Muses
who supply the words, in which he differs from Hellenistic epigrammatists (if not
Hellenistic epic poets), who tend to regard them as fellow laborers; cf. Giangrande
143 n. 10.

8 Love as madness: Phil. De Dis 3 fr. 76.6 ff. Diels, love is

Sappho 1.17 f.
Alkaios 283.5, Anakr. 428, Theogn. 1231, Fur. Hipp. 1274-1276;
Plato Phdr. 265b, Theokr. 11.10 f. See further on 5.2. Griffiths 42 n. 48 suggested
reading because the phrase takes the da-
tive; but outside of this formula is free to take the genitive; note in par-
ticular Plut. 789a K. (Editors before Kaibel construed the genitive with

8 _ (The accent is wrong in GP.) Female deities, are called

or with some regularity: Persephone of course, but also, e.g., Eirene
(Aristoph. Pax 976), Demeter (Thesm. 286), Athena (Eur. Kykl. 350), Aphrodite
(Phil. 8.7-8, Eur. Hipp. 415 - Xenarchos fr. 4.21 K-A), Hekate (Aisch. fr. 388),
78 Epigram 4

mountain nymph (Aisch. fr. 342), and heroines (Kallim. fr. 602 Pf., Nikainetos 1
GP [AP 6.225]) and minor deities such as Tyche, Eutelia, Eirene, Peitho. See fur-
ther A. Henrichs, "Despoina Kybele: Ein Beitrag zur religiogen Namenkunde,"
HSCP 80 (1976) 253-286.

4P 5.112 [18 GP, 19 K, 15 G]

P PI 7.94, f. 72
[I]
1 P: PI 2 PI: P 3 C P: PI 4
CP1: P? 5 Lumb Herwerden:
F. W. Schmidt Desrousseaux 6 P: PI

I fell in love. Who hasn't? I reveled. Who is not an initiate of revels? But
whose fault is it I went mad? A god's, isn't it?
Let it go, for already grey hair rushes in to take the place of black—grey
hair the proclaimer of the age of wisdom.
And when it was right to play we played; and since it is right no longer, we
shall lay hold of loftier thoughts.
Griffiths, BICS 17 (1970) 38.
Jacoby, RM 60 (1905) 99 f.
Lumb 11.

A complement to the longer and more complex 4, without any further reference to
lovemaking after the first word, and hence with no hints of marriage. Having ar-
gued that 4 was the last poem in Phil.'s poetry book, Griffiths, noting the similari-
ties between the two poems, goes on to suggest that this poem stood first. Since I
regard Griffiths mistaken about the place of 4, there is no reason to follow him on
the original place of 5 in the collection. And as we have noted in the Introduction,
Phil., like other epigrammatists, often wrote variations on the same theme.
As Kaibel noted, this poem must be a conscious answer to Meleager 19 (AP
12. 117):
 Epigram 5 79

Note in particular that the relatively rare form is used to make precisely
opposed points, Meleager rejecting the teachings of philosophy for revelry, Phil,
embracing them. Propertius models 3.5.19 ff. on Phil. Note esp. the poem's Epicu
rean coloring, which may owe something to a lost poem of Phil.:

me iuvat in prima coluisse Helicona iuventa

Musarumque choris implicuisse m a n u s . . . 20
atque ubi iam Venerem gravis interceperit aetas, 23
sparserit et nigras alba senecta comas,
turn mihi naturae libeat perdiscere mores,. . .
an ficta in miseras descendit fabula gentis, 45
et timor haud ultra quam rogus esse potest.

For echoes of Phil, in Propertius, cf. Tait 79-81.

1 Ingressive, as usual; KG 1.155.

1-2 Phil, assimilates the erotic mania of to the telestic

mania of religious rites. Hence it is easy to credit a god for his state of mind. Cf.
Hesych.

2 It is easy to find parallels for love as madness (cf. 4.8 n. and Brown ad
Lucr. 4.1068-1072,1073-1120); and note in particular Phil. DeDis III fr. 76 Diels
Other edd. mark a period after this word
but I prefer the run of thought as shown in the translation, which makes better sense
of and allows tobetakenasingressive. Jacobs's comment, haecpaulo
gravior reprehensio, seems an understatement.
A ready answer. Cf. Phaidra's (Eur.
Hipp. 241.

3-4 Cf. 4.3. In each poem, Phil, has adapted for his own pur
pose the topos of gray hair signaling both the onset of old age and the end of extreme
sexual passion. Cf. Sappho 58.12-14 Voigt:

and H.Aphr. 228-230:

80 Epigram 5

Prob. also Anakr. 420 PMG

Cf. Apollonides27GPC4P11.25.6) end-
ing a poem on the topos "Eat, drink, and be merry, for tomorrow we may die." But
as Anakr. 358 PMG shows, sexual desire or activity does not
necessarily end at this time (cf. 4). (Horace puts an interesting twist on this motif
by pointing out that dark hair alone does not guarantee obtaining the girl of one's
choice: Epist. 1.7.25-28: reddes \ . . . nigros angusta fronte capillos, \ . . . et \ inter
vinafugam Cinarae maerereprotervae.) Cf. also Soph. Ant. 1092 f.

3 Cf. Pi. N. 4.34 where again the reference is to

old age.

5-6: For the general thought, cf. Prop. 1.14.19 ff. (cited above), Hor. Epist. 1.14.36
(in a passage praising simple, Epicurean, pleasures) nee lusissepudet, sed non incidere
ludum, 2.2.211-216 lenior et meliorfis accedente senecta? etc.

5 Lumb's conjecture ("an easy correction") is so unnecessary, one won-

ders how it occurred to him.
For the morphology, cf. Pi. Pai. 6.87 cf. A. L. Sihler,
New Comparative Grammar of Greek and Latin (Oxford 1995) 516.
Despite both Jacobs's explanation of these words as —
and Kaibel's and Stadtmiiller's defense of the MSS., scholars
have tried to eliminate this dislocation of consecutive (Stadtmiiller even rejected
two conjectures of his own.) Cf. Ps.-Anakreon 5 FGE (AP 6.134), quoted at 2.2 n.
For Herwerden's conjecture, cf. Mnemosyne 2nd ser. 2 (1874) 307, on the basis
of which he further emends v. 5 as follows:

5
 Epigram 6 81

ylP11.34[21GP,13K, 13 G]
P ii. 5 caret PI

1,2 Ap.Voss(marg.) P 3 Ap.B: P 5

PP1

Giangrande, EH 14 (1968) 145 f.; GB 1 (1973) 147 f.; QUCC 15 (1973) 13-15; MPLS (1981) 38.
Hendrickson, A]P 39 (1918) 27-43.
Schulze, BPhW36 (1916) 317, 320.
Sider,A/P 108 (1987) 313 f.

To have white-violet wreaths yet again, harp songs and Chian wine again,
and Syrian myrrh yet again;
to revel again, and to enjoy a drunken whore—-this is what I do not want.
I hate these things that lead to madness.
But bind my brow with narcissus and give me a taste of cross-flutes and
anoint my limbs with saffron myrrh
and wet my lungs with wine of Mytilene and wed me to a stay-at-home girl.

As will become clear in the commentary, I follow Giangrande in interpreting this

poem, which has to be understood dynamically, with attention paid to word order
At first a list of features of a coming banquet seems to promise an example of the
invitation/reminder topos, such as we have most straighforwardly in 28, but which
Phil, also rings changes on in 27 and 29. As we first read, and as Phil.'s original
audience first heard, of wreaths, songs, wine, incense, and feasts—I translate ge-
nerically because at first details like "Chian" and "Syrian" might not seem essen-
tial—and as we meet the repeated and i , we expect that these pleasures
are to be enjoyed once again. The drunken whore disturbs this picture, however,
and turns it completely around. (Phil, could be describing the kind of
party attacked by Cic. In Pis. 22.)
As the poem proceeds, we learn (with no tricks of word order to keep us in
suspense) what is now considered desirable. The puzzle, as Gow-Page ably dem-
onstrate, is that we find no significant difference betwen violets and narcissi, be-
tween harps and cross-auloi, between Chian and Mytilenaean wine. Phil.'s audi-
ence would have been listening to the attributes of the second banquet while trying
to recall the details of the first. But when we get to the last line, we are told the one
significant difference, the one between prostitute andparthenos, between drunken
revelry and wedded tranquility. And with this revelation, we also realize that the
seeming insignificance of the other pairs is exactly that: of no account whatsoever
in comparison with the difference that this new woman will make in Phil.'s life.
hroughout this short poem Phil, plays with and and thwarts audience expec-
tatio : (i) What starts as a desirable party is rejected (see above); (ii) only midway
through the poem does it reveal itself as a priamel (see on v. 4), (iii) whose cap seem
to be the narrator (see on v. 5) but is in fact the girl he wishes to marry. Finally, a
poem seemingly addressed to friends turns out to take the form of a prayer to some
unnamed deities (see on v. 8). These alterations of poetic form and content bril-
liantly reflect the striking conversion of the narrator's life.
82 Epigram 6

The apograph of P once in the possession of Isaac Voss (Leidensis Vossianus

gr. O8), which contains this poem, was said to have been transcribed by Friedrich
Sylburg (see above, p. 52); hence the readings assigned here and in GP to "Ap.
Voss." are assigned by Waltz and Beckby to "Sylburg"; cf. HE l.xliv f.

1 Sc. There is nothing contemptuous in the ellipse, as

Diibner said and as is favored by Gow-Page, who also suspect that the white vio-
lets can represent luxuriance by their being the first to blossom in the spring. That
is, just as others prolong the season for expensive floral wreaths (mitte sectari rosa
quo locorum sera moretur, Hor. O. 1.38.3 f.), others would try to get a jump on the
season with white violets. But once the season began, it would cease to have this
connotation, and shows that no one small stretch of time is meant. Cf. Hor.
0. 1.38.2 displicent nexae philyra coronae.

1 Elsewhere only in Greg. Naz. Carmina de se ipso 37.1211.3 MPG,

although "musical interlude," i.e., "inbetween song," is found in the
Septuagint; also Pi. O. 3.11), (Greg. Nyss. In suam ord.
9.339.16 Gebhardt); see further on 1.1.

1, 2 The particle often emphasizes "adverbs expressing frequency"

(Denniston GP 206); cf. Aristoph. Av. 921, where the poet proudly says
and, expressing regret, Soph. Ph. 806
Note also the repeated of Sappho 1, on which cf. S. T. Mace, "Amour, en-
core! The development of in archaic lyric," GRBS54 (1993)335-364; J. C. B.
Petropoulos, "Sappho the sorceress: Another look at fr. 1 (LP)," ZPE 97 (1993)
43-56, esp. 46-48.

1-2 As Gow-Page show, no meaningful distinction can be drawn

in this poem between Chian and Lesbian wine, both of which were plentiful; cf.
Euboulosfr. 121K-A
Hor. Epod. 9.33f. capaciores adfer hue, puer, scyphos \ et Chia vina aut Lesbia.

2 Although Gow-Page are right to point out that "Syrian"

became an ornamental epithet for myrrh and similar exotic items (Schulze 317 gives
many exx.), we should allow that a Syrian like Phil, would use the term in its pre-
cise sense.

3 Not just "have at the feast," as with its occurrence on v. 2, but the fre-
quent "have sexually"; cf. 21.3.
"Drunken" is preferable to LSJ's "thirsty"; cf. Ov. Am. 1.8.2-4 est
quaedamnomineDipsasanus, \ ex re nomen habet: nigrinon iliaparentem Memnonis
in roseis sobria vidit equis, perhaps (so Kaibel) recalling Phil. A hetaira known for
her tipsy ways was nicknamed Hdpowoc;; Athen. 13.583e.
Women of some complaisant variety are to be expected; cf. Aristoph.
Acb. 1090-1092:
 Epigmm 6 83

although the word especially after Phil.'s more poetic list, would seem to
have some shock value. For women as an essential constituent of a full banquent,
see also PI. Rep. 373a
Xen. Mem. 1.5.4
Cat. 13.1 ff. (4, non sine Candida puella}. Note also Asklepiades 26 HE
(AP 5.185), in which the narrator gives a shopping list to a slave for the day's din-
ner, the last line of which is cf.
Giangrande, "Sympotic literature and epigram," EH 14 (1968) 142.

4 These words come as a surprise (see introduction and on 7.3

producing an eccentric priamel. (For the priamel in general cf. Gerber on
Pi. O.I.I, with bibliography.) Note iv.8 (see comm. ad loc.
for its possible relationship with a Kallimachean priamel) and v. 22
Here, as usual the form • (and occurs at the beginnin
of the poems; cf. Hor. O. 1.38.1 Persicos odi, puer, apparatus, AP 12.200 (Strato).
Hendrickson adduces Anthol. Lat. 458 (= 456 Shackleton Bailey) which seems to
echo both Horace and Phil, (note esp. the placement of non amo):

semper munditias, semper Basilissa decores,

semper dispositas arte decente comas,
et comptos semper cultus unguentaque semper.
omnia sollicita compta videre manu,
non amo. neglectam, mihi se quae comit arnica, 5
se det: inornatae simplicitate valent.
vincula nee curet capitis discussa soluti.
et coram faciem me lavet ilia suam.
fingere se semper non est confidere amori.
quid quod saepe decor, cum prohibetur, adest? 10

(Cf. Housman Classical Papers 1121f. on v. 8.) Note also how Archil. 5.2 W2
alters the natural meaning of the first line
which would naturally be taken to mean that the Thracian was re-
joicing with his own shield; cf. A. W. H. Adkins, Poetic Craft in the Early Greek
Elegists (Chicago 1985) 52.

4 I follow Gow-Page and Aubreton in my translation (Beckby is un-

clear), but perhaps Jacobs, Delectus epigrammatum Graecorum (Gotha 1826) 230
is right to take it as an adverbial phrase (cf. LSJ s.v. C III 7). In which case,
render "I hate these things which are done in madness."

5 This slight alteration seems warranted, as an accented pronoun fre-

quently identifies the "cap" of the priamel; cf. E. Bundy, Studia Pindarica (Berke-
84 Epigram 6

ley 1986) 5 n. 18 for the definition of the term. Phil, is still playing with the audience's
expectations, however, for the real cap comes in the last line.
Although its invention and use are credited to Pan (Bion fr. 10.7
Gow), this instrument is not the same as the panpipe or syrinx, which Longus (1.4.3
and 4.26.2) keeps distinct from it. Rather, it had a reed and was "held transversely
and played by blowing across the open end or, as in modern flutes, across a hole
cut in the side (Apul.Met. \l3.6oblicum calamumadauremporrectum dexteram)"
(Gow on Theokr. 20.29); cf. K. Schlesinger, The Greek Aulos (London 1939) 79
Pollux 4.74 says that it is made of lotus and is a Libyan invention. It is also called
(i) Lucian, Ver. Hist. 2.5, Longus locc. citt., and Heliodoros,Aith.
5.14.2 (who, unlike Longus, seems to confuse it with the syrinx); and (ii)
(Athen. 4.175e, 182d-e. (Hesych. s.v has the confusing entry,
Cf. A. A. Howard, "The or tibia," HSCP 4 (1893)
1-60, esp. 14 £; and the pertinent articles inRE: "Aulos," "Photinx, and"Plagiaulos."
A reasonable construal of the ancient references is that any cross-flute could,
by definition, be called while Libyan examples made
of lotus wood (and perhaps thereafter any of this wood from whatever source) were
called Note then that Poseidonios said
(F 54 E-K = Athen. 176c).

5-6 The closest parallel is Anaxippos fr. 1.27

K-A cf.also Herodas 6.11

7 See above on 1—2 The spelling M for M;

begins as early as the fourth century B.C.; cf. L. Threatte, Grammar of Attic Inscrip-
tions 1 (Berlin 1980) 266.
"Drink deep"; cf. Alkaios 347.1
Eratosth. fr. 3 Diehl = 25 Powell
Anon. ap. Souda T 212
K' Hor. O. 4.12 22 f. te. . . tingerepoculis, Petron. 73 tengomenasfaciamus.
Although some scholarly controversy exists concerning the equation of Backhos
and Dionysos in and before the fifth century (cf. S. G. Cole, "New evidence for the
mysteries of Dionysus," GRBS21 [1980] 223-238), well before Phil.'s day the syn-
onymity between the two is complete. For Dionysus = "wine," cf. Eur. Ba. 284
Procl. in Plat. Crat. 406c

(Orpheus fr. 216 Kern), Hesych. The evidence for

this belief in Campania is gathered by H. Herter, "Bacchus am Vesuv," RM 100
(1957) 101-114, esp. 106 ff. See further W. Burkert, Homo Necans (Berkeley 1983)
224 f.; D. Obbink, "Dionysus poured out: Ancient and modern theories of sacri-
fice and cultural formation," in T. H. Carpenter and C. A. Faraone (eds.),Masks of
Dionysus (Ithaca 1993) 65-86, esp. 78-86.
In his more sober moments Phil, condemns this sort of equation; cf. De Pietate,
P.Herc. 1428 fr. 19 Henrichs (HSCP 79 [1975] 107) and col. 3.10-13 Henrichs
(CErc 4 [1974] 14); cf. Epic. fr. 87 U. More fully, Lucr. 2.656 f. hie siquis mare
 Epigram 7 85

NeptunumCeremquevocare \ constituitfrugesetBacchi nomine abuti... .Thisreadi-

ness to deify objects and actions of everyday life is course a regular feature of early
thought; cf. Wackernagel, Vorlesungen 2.62 f.

8 Marriage is meant, as often with both this verb and its simplex (LSJ
II2):cf.e.g.Eur./l/&. 165f.
Soph. Tr. 536 Note also Eur. Hipp.
545 f. (with Barrett's note). See further E. W. Bushala,
Hippolytus 1147," TAPA 100 (1969) 23-29; A. La Penna,
Maia 4 (1951) 206; Nisbet-Hubbard on Hor. O. 1.33.11. The imperatives from
seemed, when first met, to be addressed to friends (his first
audience?); this final imperative could hardly be so addressed and is more the
sort of request addressed to gods; it thus retrospectively converts the entire poem
into a prayer.
A "stay-at-home"; cf. Eustath. ad Od. 1.412 [sc.
cf.Theokr. 1.115
Also qualifying for this term are spiders (Erykios 9 GP = AP 9.233), bookworms
which do their work without being seen (Evenos 1 GP =AP 9.251)
and oysters (Hesych. s.v . Schulze 317 mistakenly com-
pares Hor. O. 2.11.21 f. quis devium scortum eliciet domo Lyden?

AP5.4[1GP,9K,11G]
P: PI 7.88, f. 72 10
Ql 5-6 om. PI
4 tent. Stadtm. (in app. crit.): P: PI: Jacobs: Salm.
P PI P: C C: P Brunck
C: P: J.G.Schneider: Bosch (apud
Huschke) 6 C: P

Philainis, soak with oily dew the lamp, the silent confidant of acts which
are not to be spoken of,
and then leave. For Love alone does not desire living witness. And shut
the door tight, Philainis.
86 Epigram 7

And you, dear Xantho, (to) me—but now, O lover-loving wife, learn what
Aphrodite has left for us.

Huschke, 150-153.
Mariotti, II5° Libra dell'Antologia Palatina (Rome 1966) 127-34.
Sider, AJP 108 (1987) 311-324.
Snyder 348.

A bedroom scene, one of many in the Anthology (Phil, alone has 14, 25, and 26):
The maid Philainis is told to fill (a presumably already lit) lamp and leave, locking
the door behind her, before the lovemaking begins. The woman Xantho is now
addressed, but, although the inanimate lamp is called by its traditional apellation
of witness, there remains one more animate witness to be gotten rid of: the reader.
Xantho will need no further instructions. This poem thus seems like the model for
Marcus Argentarius 13 GP G4P5.128):

Cf. R. Del Re, "Marco Argentario," Maia 1 (1955) 19 f. Outside witnesses are often
held to be undesirable: "Plato" 6 FGE C4P7.100), Dioskorides 1 HE (5.56), Paulus
Sil. 60 Viansino (5.252), Tibullus 1.2.33 Lparciteluminibus.. .. celarivult suafurta
Venus; cf. F. Wilhelm, "Tibulliana," RM59 (1904) 288 £; Lier 41-43.

1-2 For asyndeton of attributive adjec-

tives, cf. KG 1.277.

A common name (23 exx. in LGPN 1-2); cf. the Philainion in 17. Snyder
reasonably suggests that this name was chosen for the maid to resonate with the
more significant use of the stem later in the poem. See further Headlam-Knox
on Herodas 1.5. In erotic contexts the name might also be meant to recall the (real
or imaginary) Philainis who wrote an illustrated treatise on love-making;
cf. P. Maas, "Philainis," RE 19 (1938) 2122; D. W. T. Vessey, "Philaenis,"
Rev.belg.pM.hist.54 (1976) 78-83; J. E. G. Whitehorne, "Filthy Philaenis (P.Oxy.
39.2891): A real ladv?" Pan. Plor. 19 (1990) 529-542: TJsener. Enirurea 4]9.
Since, therefore, the name has strong erotic overtones—cf. Poseidippos 2,
Asklepiades 35 HE (AP5. 186, 202), Luc. Dial.Metr. 6, and the Philainion in 17
the first-time audience for this poem may be forgiven for thinking that Philainis is
its love object. They would be brought up short, therefore, by the abrupt of
line 3 (I owe this observation to Nita Krevans), which word thus fulfills the same
function as 6.4 , q.v. That is, in both poems the audience's expectations
of a erotic poem are dashed or altered when the woman who would seem to be the
typical hetaira of such a poem is rejected for another woman who is or who will be
the narrator's wife. This structure makes the choice of wife all the more pointed.
 Epigram 7 87

Lamps are several times called witnesses to the lovers' tryst:

Aristoph. Ekkl. 1-16, Argentarius 13.4 (see above), Meleager 23.3-4 (AP 5.197),
AP5.5 (Statyllius Flaccus; a particularly witty example spoken by the lamp itself);
Mart. 14.39 dulcis con scia lectuli lucerna: \ quidquid, vis facias licet, tacebo; cf. Lier
43-45. They can also serve as witnesses to oaths by one or the other lover: 36. The
lamp is an inanimate witness (below, v.3-4); cf. Aisch.Ag. 1090
Before Phil, only doubtfully in Stobaios' citation (along with other
deviations from the codd.) of Theog. 422-424 = fr. dub. 6.2 Young, where
are on the tongue of many men; cf. M. L. West, Stud.
Gk. Eleg. Iambus (Berlin 1974) 155. Elsewhere only in Christian writings, twice in
the sense "inarticulate": Ep. Rom. 8.26 and Greg. Nys. 46.25a MPG
. Only Greg. Nys. 44.3lOc in the same sense as here, "ineffable, secret."
Phil, is preparing the way for the aposiopesis to follow, perhaps (pace Mariotti 130
n. 1) with an allusion to the mysteries (cf. Meleager 11 =AP 6.162
, where the usual word is W. Burkert,
Ancient Mystery Cults (Cambridge, Mass., 1987) 137 n. 44.

2 A lamp is obviously an important household object for an action that takes

place behind closed doors (v. 4); cf. Asklepiades 9 HE (AP5.7), where the poet prays
that if his girlfriend should sleep with another it refuse to
shine on them. The development from mere witness (see above on owioTOpa) to
semidivine power is a natural one; cf. M. Marcovich, "A god called Lychnos," RM
114 (1971) 333-339, repr. in id., Studies in Graeco-Roman Religions (Leiden 1988)
1-7; for lamps in general in the Anthology, see Mariotti 93-112, 121-134.
"Dew" is often applied metaphorically to various liq-
uids; e.g., Pi. O. 7.2 cf. 3.3.
Theophrastos CP5.15.3 uses this compound in the (so-to-speak)
watered-down sense "fill with (any) liquid," as is common enough with the sim-
plexes (cf. LSJ s.w.; note esp. II. 17.390 a bull hide
, but to translate it here as "make it drink deep" (Gow-Page) vel sim. is to
lose the point; if the lamp is to be present and, by the conventions of erotic poetry,
to be a witness (see above), at least let its powers of observation be impaired. Hence,
"make it thoroughly drunk." Jokes about "drunken" lamps are common in com-
edy; cf. Aristoph. Clouds 57 Plato Com. 206 K-A, Alkaios
Com. 21 K-A. Phil, seems to have been imitated in turn by Babrios 114 init.
(so Mariotti 133 n. 2).

3 "Evidence, testimony," although translators find that they must ren-

der it as "a witness, temoin, etc.," It is not so much that the word alone equals
but that "living testimony" readily converts to "living witness."
The aorist gives the air of a gnome (so Mariotti 132) to a phrase
probably made up to suit the occasion. Cf. in general J. Labarbe, "Aspects
gnomiques de 1'epigramme grecque," EH 14 (1968) 349-383.

' "Animate"; in his prose, Phil, prefers (e.g., De Dis 112.5), but
either word is at home in poetry. The enjambement has the effect of an afterthought,
88 Epigram 7

all the more so after the quasi-gnomic v. 3, as if Phil, has just remembered that the
lamp has been called ouviaicop, and so can be considered a witness; cf. 36.4.
Tentatively suggested by Stadtmuller in his app. crit., where he rightly
compares the Dios Apate. In this locus classicus for unobserved sex between hus-
band and wife, Hera says of their bedroom that Hephaistos
(//. 14.339). A literary allusion to this passage would obviously
please Phil.'s audience, who would not yet know that they, like Homer's audience,
are to be excluded before the lovemaking starts. For the sense—"tight" rather than
"compact"—cf. LSJ s.v. III.

5 Having given orders to Philainis, he begins to do the same to Xantho;

"and as for you."
Without PI, we have to choose between P and C, each of which offers
an acceptable reading. Following Gow-Page, I print but not for their reasons:
"A completed sentence here detracts from the effect of the abrupt change of ad-
dress and from the significance of . " First, I do not believe that
there is a change of addressee (see below on and ; and, second, the
force of the last words could be seen as being even stronger after we hear of the
first kiss. I still think, however (as I argued inAJP312), that we have here another
Hellenistic example of aposiopesis designed to avoid the specific details of
lovemaking: Meleager 72 (AP 5.184.5), Antipater 52 GP (AP 9.241.5), Theokr.
1.105, 5.149, Herodas 1.84; also Aristoph. V. 1178; for its continuation in Latin
poetry, cf. J. N. Adams, "A type of sexual euphemism in Latin," Phoenix 35 (1981)
120-128. Contra, A. H. Griffiths in his review of GP, who finds the aposiopesis
very harsh and deems (I think wrongly) the lectio difficilior (JHS 90 [1970]
218). Huschke 151 briefly entertains the conjecture but
in fact prefers the reading printed here.
C has in his usual assured way worked his correction into the text
of P, obliterating what was there before; hence the note of doubt in the app. crit. as
to even the number of letters there originally. (Stadtmuller; similarly
Gow-Page) is surely wrong as there would thus be too much space between -ok
and the following <ru, but at least these scholars (unlike Waltz and Beckby) acknowl-
edge that C has been at work here. With Brunck, followed by Jacobs 1794,
read the line as follows: (sc. Philainis),
surely a difficult aposiopesis to complete. The direct object
(ie, on the other hand, immediately suggests several possible ideas, none of which
Phil, cares to spell out.
Foi + impv. after a vocative, cf. e.g., Eur. Hek. 1287
Asklepiades (cited above, on v.2
But Phil.'s more complex example (omission of the first imperative,
addressing Xantho anew) is probably to be explained more simply as resumptive
after the preceding aposiopesis.
"Lover-loving" (Gow-Page) is the basic meaning of this stem,
which, although not common, also appears in several other words can
be applied to inanimate objects, in which case it means little more than "dear to
 Epigram 7 89

lovers"; e.g., Meleager 42 (AP 5.136.5) IG 14.793a = 560

Kaibel (Naples, IA.D.) appears elsewhere
only in Argentarius 29 (AP 10.18), modifying Aphrodite (Huschke's certain
conjecture).
If the word means "lover-loving," whi half is verbal and which is nominal?
Usually • is the verbal element, but so t is as the second element. In
two philosophical passages the answer is clear: (i) PI. Symp. 192b
where, despite the similarity of forma-
tion, context makes it clear that "the former term applies to the pursuer, the latter
to his younger quarry" (Dover ad loc); c£, however, 213d (Sokrates on Alkibiades)
where is
ambiguous since the "erotic" relationship between Alkibiades and Sokrates is re-
ciprocal. E. Fantham, Phoenix 40 (1986) 48 n. 10, translates "passion for the lover,"
not as anteros, comparing where, however, the latter stem is not ver-
bal, (ii) Arist. Rhet. 1371b24 includes with
On the other hand, is the only stem ending
thus where the final element need not be verbal. Thus, without specific context to
determine otherwise, one is drawn to take first as a typical - word
and then as a typical word in order to bring out the tension inherent in its
formation. Xanthippe, therefore, an Epicurean and hence a "friend" in the special
sense used within the Garden (see on 27.1 is now addressed by a term
that can be understood both as "friend to your lover" and "lover of your friend,"
the alternation suggesting the reciprocity of their relationship: Each is lover and
friend of the other. therefore, is an erotically charged equivalent to
which Phil, uses twice in (frr. 50.8, 85.8); note also
27.2 Also interesting is Semonides' comment on his bee-woman:
(7.86 W); Xen. Symp. 8.3, P.Ant. 15.15-17.
C abraded all but the vertical extender of P's which looks like a
roman-letter "h," adding a diaeresis mark (two dots above the extender) further to
indicate its new status as i. Nouns ending in frequently have their nominative
forms serve as vocatives; cf. V. Schmidt, Sprachliche Untersuchungen zu Herondas
(Berlin 1968) ch. Ill, "Gebrauch des Nominativs fur den Vokativ," 89-95, esp. 93f.;
I. Bekker, Homerische Blatter, I (Bonn 1863) 268-271. The omission of the last
distich by PI is further indication that was the original reading, as the
address to the bed is well within Planudes' tolerance level; on Planudes as a
bowdlerizer, see below, p. 221. For Phil's use of elision, see Intro., p. 41.
Although modern editors from Diibner on prefer an address to the bed here
(Schneider's emendation), this would be the only such address in the Anthology.
Renaissance edd., Jacobs (1813), and Huschke retain C's reading, as does H. J. Polak
Mnemosyne, 2d ser. 5 (1877) 434; but Gow-Page regard it as one of his "blunders"
(GP 1.1 n. 5). It is more likely, though, that C and his source Michael had solid M
authority, rather than that they conjectured the less likely "wife" over "bed," ex-
amples of which can be found elsewhere. Indeed, addresses to a bed by lovers were
regarded as a commonplace by Plut. de Garr. 513
90 Epigram 7

" Cf. Ticida fr.l, Prop. 2.15.2, Mart. 10.38.7. Kaibel adduces Ov.
AA 2.703 as evidence for reading conscius ecce duos accepit lectus amantes; \
ad thalami clausas, Musa, resistefoms, which certainly makes the same point as Phil.;
i.e., the poem stops before the lovemaking begins. Ovid's model, however, may have
been Asklepiades25 WP5.181.12) Why, furthermore,
would Phil.'s conscius bed, which presumably has seen similar scenes before, need
his command .. . ?
P. Chantraine, "Les noms du mari et de la femme, du pere et la mere en grec,"
REG 59-60 (1946-1947) 225, notes that ckora<;, far more than ("legiti-
mate wife"), "presente volontiers une valeur affective," and that "on n'est pas surpris
enfin de lire dans les scenes amoureuses." In sum, the MS
evidence combined with the argument presented here and in the introduction favors
the reading "wife" here. See Intro., pp. 34-36 for further reason to accept a love
poem addressed to a wife.
Open expression of erotic feeling such as is found here of a husband for his
wife is extremely rare in Greek literature; erotic poetry deals largely with pursuit
and rejection. It was noteworthy that Kandaules fell in love with his own
wife (Hdt. 1.8.1). TheDios Apate alone, however, provides a sufficient literary model
for Phil, (see above on 4 the aposiopesis of his poem substituting for
Homer's concealing cloud. For some other instances where a husband's erotic pas-
sion for his wife is expressed or alluded to, cf. M. Lefkowitz, "Wives and husbands,"
G&R N.S. 30 (1983) 31-47, esp. 36-38, where she adduces P.Antinoop. 15; K.
Gutzwiller, "Callimachus' Lock of Berenice: Fantasy, romance, and propaganda,"
AJP113 (1992) 359-385. The situation in Latin literature is far more complex, where
mistresses are often spoken of as wives or at least with language more appropriate
to wives than lovers; cf., e.g., Cat. 109.6, Tib. 1.5, Hor. O. 2.12, Prop. 2.6.41 f.;
R. O. A. M. Lyne, The Latin Love Poets: from Catullus to Horace (Oxford 1980) 2-
8, 79 f., and ch. 2 on Catullus passim; D. Konstan, "Two kinds of love in Catullus,
"CJ 68 (1972-1973) 102-106; M. Santirocco, "Strategy and structure in Horace C.
2.12," Latomus 168 (1980) 223-236; S. Commager, A Prolegomenon to Propertius
(Cincinnati 1974). (Arguing against the view that Calvus wrote erotic poetry to his
wife is E. Courtney, The Fragmentary Latin Poets [Oxford 1993] 208.) Note too
Sulpicia, who, according to Martial 10.35, 38, wrote love poetry addressed to her
husband Calenus (Courtney 361).

6 Aphrodite as the "Paphian" is common by Phil.'s time; she returns to

her temple in Paphos before appearing before Anchises (H.Aphr. 5.58-65), and it
is again whither she repairs after being released from Hephaistos' toils (Od. 8.362-
366). Phil, himself uses the adj. of one of his girlfriends named Demo (10.1).
Aristoph. Lys. 556 is first to apply the adj. to the goddess;
occurs several times in the Anthology. One of the Philodemean incipits is jiupia
ii.10), where the goddess must be meant.
Cf. Argentarius 13.3-4 (see above, intro.), Ov. Am. 1.5.25 cetera
quis nescit?
 Epigram 8 91

AP 10.21 [15GP, 8 K , 3 G ]
PP1 la.30,5 viii.2
.PI: P 3 PI: P 5 P: PI
Brunck 8 (i.e., P: PI
Pl(spscr) PI: |P

Unruffled bridegroom-loving Kypris, ally of just men, Kypris, mother of

stormfooted Desires,
Kypris, (rescue) the one halfway dragged from the saffron bridal bed, me,
the one snowed upon by Celtic snowstorms,
Kypris, peaceloving me, the one who says stupid things to nobody, the one
awash on your purplish sea,
Kypris, lover of harborage and lover of (your) rites, preserve me, Kypris,
now, Mistress, to the Naiadic inlets.

Reitzenstein, RE 6 (1907) 98.

Giangrande, MPL 5 (1981) 37.
Cichorius, Romische Studien (Leipzig 1922) 295 f.
Falivene, QUCC 42 (1983) 129-142.
Cavallini, Museum Critkum 15-17 (1980-1982) 164 f.

In the form of a prayer addressed to Aphrodite. But if the reference to Nais is cor-
rectly understood, Aphrodite's help is scarcely necessary for gaining access to a
courtesan. It would make more sense to imagine that the poem is only nominally
addressed to Aphrodite while really intended for the wife's ears, the true message
thus being: Take me back or lose me to a courtesan. Cf. Cat. 36.11-17 for another
mock-serious prayer to Venus: nunc, o caeruleo creata ponto, \ quae sanctum Idalium
Uriosque apertos quaeque Ancona Cnidumque harundinosam \ colis quaeque
Amathunta quaeque Golgos quaeque Durrhachium Adriae tabernam, acceptum
face redditumque votum, si non inlepidum neque invenustumst, where the repeti-
92 Epigram 8

tion oiquaeque is similar to Phil's repetition of the name; cf. H. Kleinknecht, Die
Gebetsparodie in derAntike (Stuttgart and Berlin 1937) 178 £; H. Pelliccia, Mind,
Body, and Speech in Homer and Pindar (Gottingen 1995) 268-271.
Gow-Page show that this poem is almost certainly the lament of a recently
married man now banned from the bedroom, and they quickly dispatch Kaibel's
argument that the narrator has been caught in adultery. All that is required is some
unspecified argument.
Since the narrator describes himself as stormtossed, Aphrodite, also a god of
the sea, is doubly appropriate—and all the more so when we see how the nautical
imagery can be used for erotic effect. Cf. AP5.11 (Anon. 7 FGE):

Also AP 5.17 (Gaetulicus 1 FGE), cited below on 7-8; cf. also Cat. 68.1-6 fo
another metaphorical use of a storm and shipwreck (Kaibel xii); Hor. O. 1.5, with
Nisbet-Hubbard's commentary on v. 16 deae. And for the imagery of the harbor in
general, cf. C. Bonner, "Desired Haven," HTR 34 (1941) 49-67.

1 This vocative appears seven times all told in this short poem, giving an
air of extreme desperation to the narrator. For "cletic" anaphora, cf. K. Keyssner,
Gottesvorstellung . . . gr. Hymnus, Wiirzburger St. zur Altertumsw. 2 (Stuttgart
1932). Unusually for Phil., there is no A-caesura in any of the four hexameter lines
of this poem, which produces a rushed quality; cf. H. Porter, "The Early Greek
Hexameter," YCS 12 (1951) 1-63, esp. 10-12 for the A-caesura at either position
3 (A1) or position 2 (A2).
Properly "with the calm of the sea" (in Aeolic poetry prob. Alkaios
286a.5; cf. Voigt ad loc.), it easily transfers to calm or undisturbed visage, sound,
or thought, although the metaphor can occasionally be revivified; e.g. Aisch. Ag.
740 Probably also here, as the poem continues in a dis-
tinctly nautical tone. Cf. Gow-Page and Cavallini n. 14, who points to Aphrodite's
role as savior of sailors in danger (Roscher 11.402). Her double role is appealed to
by Gaetulicus 1 FGE (AP5.17.6) spo-
ken by someone about to cross the sea to his girlfriend. Cf. AP 5.11, cited above;
Hor. O. 3.26.5 marinae. . . Veneris (an erotic context).
"Friendly to " a hapax legomenon designed to appeal to
that aspect of Aphrodite's power of immediate interest to the person praying; cf.
W. Burkert, Greek Religion (Oxford 1985) 74. Aphrodite's association with newly-
weds is well known and obvious. Cf. e.g. II. 5.429 (Zeus to Aphrodite:)
Sappho 112
Diod.Sic. 5.73.2. Cf. Joann. Damasc. Enc. in St. J. Chrysost. 96.781.14
MPG "loving one's wife."

1-2 The dative is more common with than

the genitive (3x in Aisch. Cho. alone). Cavallini would like to see the influence here
 Epigram 8 93

of Sappho 1, another appeal from a rejected lover to Aphrodite, which has

(19 f.) and (28). This may be overreaching, but
her remarks that often refers to adultery are well taken.

2 As in Pindar fr. 122.3-5

Bakch. 9.73, Rufinus31.5 Page (AP5.87),Babrius 32.2, Hor. O.I. 19.1,4.1.5 mater
Cupidinum. On the see on 3.2.
"Kicking up a storm" is a close English equivalent. It is a fre-
quent epithet for swift horses (H.Aphr. 211, Simonides 515 PMG, Pindar, N. 1.6
etc.), or excited dancers (Eur. Hel. 1314); but Phil, probably uses it because the
Pothoi are winged, as in 3.2; cf. Euphorion 113 Powell
[Homer's (II. 8.409,24.77,159) may have suggested wings to Phil.,
as Iris is depicted with winged boots in early art and with bodily wings in later art;
Homer himself may have meant little more than "swift."]

3 Cf. 5 Article + pronoun is rare, most

prose exx. coming from late Plato (Lys. 203b, Phil. 14d, 20b, 59b; Tht. 166a;
Soph. 239b), none of which is precisely parallel to Phil.; cf. J. Wackernagel,
Vorlesungen uber Syntax 2.138. In his discussion of the phenomenon Apollonios
Dyskolos Pron. 13.16 Schn. offers several verse exx., including Kallim. fr. 28
(but Pfeiffer ad loc., who prints all the relevant ancient testi-
mony, thinks that the article goes with the substantive, the pronoun being inter-
jected more Callimacheo), 114.5 (but Pfeiffer ad loc. thinks
that neither nor goes with alone: "immo post formulam
xov ubi nomen proprium dei vel deae expectatur, pronomen
reflex, ponitur. Deus per se ipse iurat"). Cf. Menander fr. 409 K-Th
and even more pertinent 19.1
. But since there seems to be no parallel with an adj. in attributive
position, it may be that in Phil, we should understand "the half-dragged (one),
me" (likewise for v. 5), where the pronoun is eventually recapitulated in v. 7
Cf. Simias4.1HEG4P7.21) .where
the pronoun is governed by v. 5 and eleager 103.1-2 (AP 12.101)

Cf. E. N. Lane," . Glotta 66 (1988) 100-123, who shows that

denotes a woven fabric which came to be so closely associated with the bridal
bed it covered that it came to stand either for marriage, especially a recent mar-
riage, or for the marriage bed; and that it never referred, as LSJ, Gow-Page, et al.
say, to the bridal chamber. Cf., e.g., Pollux 3.37
. Note also IG 12.8.441.1-2 (= 208 Kaibel)
Lane unfortunately takes Phil.s' appeal to Aphrodite too
literally, understanding the poem as "a prayer to Cypris for safety at sea by a re-
cently married man."

4 This need not be interpreted too liter-

ally. As Kaibel says, non membra sedanimum riguisse. Cf. Philemon fr. 28.1^t K-A
94 Epigram 8

Even if Piso had never campaigned in Gaul with Caesar during the 50s (Cichorius),
the severity of northern winters could be drawn upon for metaphors of this sort; cf.
Petronius 19 frigidior hieme Gallica; Hor. O. 3.26.10, where Memphis in Egypt is
said to lack Sithonian (i.e., Thracian) snow.
Gow-Page do not explain their preference for the form . ,T Stephanus,
Brunck), which, although possible, is contraindicated by the inscriptional evidence;
LSJ s.v., A. L. Sihler, New Comparative Grammar of Greek and Latin (Oxford 1995)
163.

5 A favorite word of Phil, for "stupid," with at least six occurrences in his
prose. There is then no reason to follow Brunck in emending. Cf. also Giangrande
41 for a defense of the MSS.

6 So PP1; Gow-Page's is no doubt a mere slip on their part.

7-8: Cf. AP5.17.3-4 (Gaetulicus 1 FGE; a prayer to Aphrodite):

Just as has a double meaning (see on 23.8 so too here (where again
we seem to have a reference to Nais) note the extended meaning of . . for the
female's genitals, as in Empedokles 31 B 98.3 DK
and Soph. OT1208. (Cf. also Theogn. 460, where the young wife of
an old man, compared in several ways to a ship, strays at night and

7 A hapax, whose meaning is more metaphorical than the literal

"she who loves to bring to harbor" (LSJ); cf. the preceding lemma.
Aphrodite is lover of her own rites; cf. 33.5-6
Elsewhere only IG 22.5021 = AG Append. 1.252 Cougny (iv/v A.D., of Dionysus)
and Nonnos, Par. Eu. lo. 6.9 (of the Jewish Passover).

7-8 Cf. Od. 5.452 f. (LSJ

s.v. 2).

8 Brunck preferred the reading of PI to that of the obviously corrupt

P (then in the Vatican), but Jacobs suspected that the latter concealed the truth. In
1794 he tentatively suggested "of Actium," but in his ed. of 1813 he
elegantly dropped one letter to come up with an epithet for Nais, the woman Phil,
mentions again in 23, again with erotic language punningly appropriate to her name.
(LSJ's "of the Naiads" misses the point.)
See on 4.8
 Epigram 9 95

AP 5.13 [2GP, 16K.2G]

P PI 7.93, f. 72 Fsc. vii. 25 Suda
s.w. (3-4 (3-4) 3-4, 7-8 om. PI [J]

1 CP1: P?: C(spscr): edd.vett.(1494) P:

PI 3 P Suda utrubique 4 ; Suda: P
P: Suda 5 edd.vett. (1531) PP1 CP1: P 6
P: C edd. vett. (1566): PI PP1:
edd.vett. (1494) Kaibel: Salm.: 7
P Brunck Salm.: P

Charito brings sixty years to fulfilment, but she still has her long train of
dark hair,
and on her bosom those white marble cones of her breasts still stand firm
without encircling halter,
and her unwrinkled skin still exudes ambrosia, total seduction, and a
myriad of charms.
So, all lovers unafraid of wanton passions, come hither and forget her years'
decade.

Stella 259 ff.

An older woman is still sexy, as in AP 5.26 (Anon.), 5.48 (Rufinus 19 Page), 5.62
(Rufinus 23), 5.258 (Paulus Sil. 52 Viansino), 5.282 (Agathias 78 Viansino), 7.217
(Asklepiades 41 HE); cf. Ov. Am. 2.4.45 me tangit serior aetas. The general point
of all these poems is that, whether or not there is a falling off from her prior
beauty, a grace remains that keeps her desirable (see on v. 6). This motif of
course counters the usual preference in poetry (if not in real life) for younger women,
whether pubescent or nubile; Phil, himself has 11 and 16. For invective against old
women, cf. A. Richlin, The Garden o/Priapus (New Haven 1983) 109-116.
96 Epigram 9

1 Phil, is the only author in the "older woman" group to specify the
lady's age. The spelling of the papyrus is not of a common phonetic sort; Parsons
suggests simple carelessness. He also notes that "what little remains does not suit
particularly: just offsets?" Parsons prints a question mark, keeping open the
possibility that the incipit may belong to another poem.
On age sixty usually marking the effective end of one's life, cf. Mimnermos 6,
Herodas 10.1 f.
with Headlam-Knox ad loc.
Primarily of time or one's lifetime, "complete, bring to an end" (cf. LSJ
s.v. I 7), but perhaps also in the sense "bring to fulfillment" (15).
Charito will be shown worthy of her name (see below, on 6
cf. Lucr. 4.1162, a woman is called by the man who is too infatuated
to realize that she is too small (parvula, pumilio) to be truly beautiful. Both Lucretius
and Phil, probably knew Meleager 32 (AP5.149); note esp. v. 2
Cf. further R. D. Brown, Lucretius on Love and Sex (Leiden 1987)
287.
Knowing that the last syllable has to be long, C offers both i and, above
the line, It is not clear to me why some early printed editions of the An-
thology changed Charito's name to Chariklo, hereby destroying the obvious pun
on the meaning of her name.
Adj. derived from which Phil, uses in 4. Since
can mean "years," "seasons," or "hours" (of the day), the adj. specifying the
first meaning, which came to be the least common, is not merely ornamental. Cf.
the inscription cited by Gow-Page ad loc.
copca.

2 An appositional genitive; cf. AP 12.190.3 (Strato) crup^a

(a kind of worm) "Mirabile quell'ondeggiar delle trecce
brune in contrasto al candore immobile del seno" (Stella).

3 "Those well-known (?)" breasts. Cf. LSJ s.v. 12; for the sentimental use
of the demonstrative cf. Lucr. 2. 362, Hor. O. 4.73.18, KG 1.650 £
"Marble-white" is the basic sense, and breasts are praised for their
whiteness (when not for the rosiness of their nipples, or both qualities together as
in Herrick's "Upon the nipples of Julia's breast"); cf. Rufinus 19.3
D. E. Gerber, "The female breast in Greek erotic literature," Arethusa 11 (1978)
203 f., who is right to point out that breasts, like marble, are also admired for smooth-
ness, although I am not convinced that this is "the primary significance" here, sim-
ply because, as Gerber argues, Rufinus elsewhere said that a face was smoother than
marble (AP5.28 = 10 Page). below (5) suggests this in any case.
Lucillius 859 f., which Gow-Page cite on 4 , shows that firmness too may
be alluded to: hie corpus solidum invenies, hie stare papillas pectore marmoreo. Note
also Souda "with outstanding breasts" (LSJ), Hor. O. 1.19.5 f. with
Nisbet and Hubbard's n.,urit me Glycerae nitor \ splendentis Pario marmore purius
The diminutive is not only used to express smallness, but is also, as
often, a term of endearment; Gerber 204 f., 208. The word appears elsewhere only
 Epigram 9 97

in Poseidonios F 55a E-K (Athen. 14.649d), of a small pine cone (regarded as spu-
rious or corrupt, though, by Kaibel, Theiler F87 assenting). Following Brunck, I
correct the manuscripts' accent (KCOVKX) in accord with the general rule for dactylic
diminutives in cf. Chandler, Gk. Accentuation §§343, 350.

4 Cf. Gerber 206 on the vocabulary of firm and sagging breasts, and
in particular Philostr. Imag. 2.18.4 Generally, the former are pre-
ferred, but cf. Paulus Sil. 52.2^ Viansino, who admits to going against common
preference.
: Given the right context like and
often = "brassiere"; cf. AP 5.199 (Hedylos 2 HE) fenced,
,. Cf. A. Henrich.s,DiePhoinikikia desLollianos (Bonn 1972) 123
f., for a discussion of the ancient brassiere. Other terms in a brassiere onomastikon:
. With the adj. cf. AP 6.272.2 (Perses 2
HE) both a hapax and a "pe-
culiar feminine of " (LSJ), should prob. be understood as an epithet
designed primarily to amuse. Gerber 208 f. says that this is the only passage he can
find in praise of bralessness, but Phil, does not actually say that Charito goes bra-
less, only, as a lover would know, that she does not need one to keep her breasts
fro sagging. Cf. further RE 6.2007, Daremberg-Saglio s.v. fascia, 3° pectoralis.

5 Other women remain sexy despite their wrinkles; cf. AP 5.258.1

f. (Paulus Sil. 52 Viansino)
Asklepiades 41.2 HE (AP 7.217) A? 5.26
(Anon.).
Cf. Theokr. 15.108 (Aphrodite)

5-6 Cf. Aisch.Ag. 419 , "all charm of love" (Fraenkel).

6 Although is possible (cf. II. 19.39 f

the reading of C (and Stephanus) interrupts the total concentration on Charito by
referring to her effect on others, hereby blunting the point of the last distich.
Generally used of distinct drops (blood, sweat, and tears), but cf. Eur.
Hipp. 525 ff.
Combining Eur. and Phil., we may say that Eros (as both agent and
result) takes the beauty of Charito and distils it into an ambrosial desire which he
then drops into the eyes of the beholder. Cf. also Pi. I. 4.90b
Alkman59(a) Hes. Th. 910
f. with West's n. ad
loc. and M.Davies, Hermes 111 (1983) 496 f., Simon. 22.12 W2
Krinagoras 50.6 f. GP (API 199) " '

There is some confusion in this line and the preceding between nominative
subject and accusative object of (-) , but the text printed here makes best sense
and is closest to the MSS. Certainly, -lr| would produce an intolerable hiatus.
98 Epigram 9

Charito in other words is well named—a common motif in AP as

well as elsewhere; Snyder 347. Cf. 10, where Phil, discovers that he too has been
aptly named. Puns on are particularly easy and common; Meleager
alone has 30.4 (AP5.140) 32.4(5.149,
see on v.l 47 (AP 5.148)
On the nature of erotic charis, cf. B.
MacLachlan, The Age of Grace: Charis in Early Greek Poetry (Princeton 1993),
ch. 4.
For the loss of charis along with youth, cf. Archil. 196a.27 f. W2 (of Neoboule)
Two other older women
praised in the Anthology retain Rufinus 23
Agathias , perhaps also vii.20
(see comm.). The Graces of course are regular attendants of Aphrodite:
Od. 8.364-366, 18.193 f., H.Aphr. 61-67, H.Ap. 194-196—probably so because
they are deified rays of sunlight (and Aphrodite's origin is at least partially that of
the Indo-European sun goddess); cf. P. Friedrich, The Meaning of Aphrodite (Chi-
cago 1978) 196-198; W. A. Borgeaud & B. MacLachan, "LesKharitesetlalumiere,"
Rev. beige Phil. hist. 63 (1985) 5-14. Indo-European origins aside. provides
thefinishing touch to beauty; cf. AP 5.67 (Capito)

7 "sexual desires." Cf. 8.2 with comm.;

Pollux 6.188 and in general for this
sense LSJ s.v. II1. The text is sound; there is no need for Brunck's emendation or
those even worse given serious hearing by Jacobs (1794).

8 Cf. Kallim. fr. 1.6

Leonidas 20 HE (AP 7.295.6) Menekrates 3 HE
G4P9.55.2)

10

5
 Epigram 10 99

AP 15[6GP,3K,8G]
P PI 7.97, f. 72v [sc. vii.7
[J]
1 CP1: P? 3 Salmasius: C ex ?P:
PI: N Sternbach: Chardon: Boissonade: Jacobs:
N Kaibel 6 PP1: edd.vett. (1494)

I fell in love with Demo from Paphos; no surprise. And, second, with Demo
from Samos; no big deal.
And again, and third, with Demo from Hysiai—this is no longer a joke—
and fourth with Demo from Argos.
It must have been the Moirai themselves who named me Philo-demos, so
that burning passion for a Demo would always take hold of me.

L. Sternbach, Meletemata Graeca (Vienna 1886) 85 f.

Philodemus' pandemic eros. He discovers that he has a redender Name. 31 offers a

rough parallel for having one's character determined at birth, but closer in theme
is Meleager 98 (AP 12.165), where the Erotes are credited with the poet's loving
boys both black (usA,-) and white In vocabulary more than theme Phil.'s
poem would seem to be a descendant, perhaps through a chain of theme and varia-
tion, of Meleager 26 CAP 5.160):

1 A meaningful name in Hellenistic erotic poetry; cf. Meleager 26 (see

introduction, above), 23,27,28 C4P5.197,172, 173); Antipater5 HE (AP6.175);
Phil. 11. One expects one so named to sleep with many men; here, however we
find Phil, sleeping with many Demos.
Any girl born in Aphrodite's birthplace could be expected to
be sexy; see on 2 (LGPN 1 records a real Demo of Paphos from the third
cent. B.C.)
"A ready made phrase" (Gow-Page); but see on 18.3.

2 Samos was famous for its red-light district; cf. Klearchos fr. 44 Wehrli
(Athen. 12.540f)
Plut. Mor. 303c, K. Tsantsanoglou, ZPE 12 (1973) 192 f. Other sexy
Samiotes in the Anthology are found in Asklepiades 7 HE (AP5.207) and Rufinus
17 Page (AP 5.44). It is not only the low number two that is unsurprising, there-
fore, but also that they come from Paphos and Samos. There is also the possiblility
that in addition to the pun that is the point of this poem, Phil, alludes to the (ad-
100 Epigram 10

mittedly weaker) link between his name and these cf. Hesych. s.v

3 Preferred by Waltz and Beckby, this is as close as we can get to

the various readings of P, which are in agreement in having followed by a
vowel marked with breathing. Stadtmuller records PPC, but all I can see
in the MS are the traces of the top of either a a (so also Gow-Page) or a sigma-tau
compendium. (Gow-Page are incorrect in giving N to C; he would have
erased the breathing mark.) Pi's reading would be an unnecessary hapax for "Asian,"
when the name of a city is needed. Salmasius's Hysiai is the name of two Greek
cities, one in Boeotia and another in Argolis; cf. Bolte, "Hysiai (1 and 2)," RE 9.1
(1914) 539 f. Beckby unconvincingly constructs the syllabic acrostic
Jede ein Schwein von Argolis, which has somehow to do with the sacri-
fice of a pig to Aphrodite in Argos. Stadtmuller and Gow-Page adopt Sternbach's
reading.

5 One's lot in life is assigned at birth by the Moirai

(Pi. N. 7.1 O. 6.42 f., II. 6.488 f., 24.209 £, Plato,
Symp. 206d, Apollod. Eibl. 1.8.2.1), who see to it that one lives accordingly; cf.
Fraenkel ad Ag. 1535 f.; K. Krikos-Davis, "Moira at birth in Greek tradition,"
Neohellenica 4 (1982) 106-134; and (for their appearance in art at the births of
Athena, Aphrodite, and Dionysus) S. De Angeli, "Moirai," LIMC 6.1 (1992) nos
13-22. For the fatefulness of names, cf. Aisch. Ag. 681 ff. (the chorus on Helen)

Agreeing with Phil, that it is the Fates who truly name people
is Ausonius Ep. 20.4 Protesilae tibi nomen sic Fata dederunt, victima quod Troiae
primafuturus eras.
The verb appears in poetry before Phil, only in Anaxandrides Comic. 35.5
K-A, but is common in philosophical texts (Plato, Aristotle, Anon. Lond. Med.,
et al).
For the fickleness inherent in one so named cf. PI. Gorg. 481de
on Kallikles ' , the lover both of Demos the son of
Pyrilampes and of the Athenian demos. Phil, may also have in mind Archilochos'
= "prostitute" (207 W2).

6 A well-known characteristic of sexual passion; in Phil, alone

cf. 1.2,4.6, 11.1, 16.4.
PP1 are in agreement here; Stadtmuller says " P, corr.Pl," but I
see no sign that Pi "corrected" nor evidently did Brunck or Jacobs.
Subsequent edd., however (Waltz, Beckby, and Gow-Page), follow Stadtmuller,
but is found only in the early printed editions of the Anthology (1494+).
Although an indicative would explain why Phil, said v. 5 and in particular
, an optative can be regarded as appropriate to the way the Fates fix every-
thing in advance, and would have a slightly comic tone when spoken by an
Epicurean.
 Epigram 11 101

11

AP 12.173 [16GP,2K,9G]
P App.Barbero-Vaticana 11 (om. 4-6) caret PI
ac
1 P App.B-VP- App.B-V G6tting.philol.6 2 Sternbach (1890):
Sternbach (1886) PApp.B-V(i.e., Preisendanz):
Gallavotti: ( delete Ap.B: Kaibel 3
Wilam. P: App.B-V 5 Petit

Demo is killing me, and so is Thermion, the one being a hetaira, Demo
not yet knowing Aphrodite.
And one I touch, the other I may not. I swear by you, Kypris; I do not know
which one I should say is more desirable.
I will say it is Demarion the virgin; for I do not want that which is at hand,
but I have a passion for all that is under guard.

Gallavotti, Boll. Class. Lincei 5 (1984) 88-91.

Kaibel, Hermes 15 (1880) 459.
Prinz, WS 34 (1912) 227 K.
Sternbach, Meletemata Graeca (Vienna 1886) 121.
Sternbach, Anthologiae Planudeae Appendix Barberino-Vaticana (Leipzig 1890) 18-20.
White, Corolla Londin. (1981) 175-177.

A double-sided topos of love poetry: Which woman is more desirable, the one who
makes herself easily available (an adulterous wife or prostitute) or the one who is
or plays hard to get? Prinz surveys the topos in the Anthology and Latin poetry; see
below on w. 5-6. Phil, prefers the latter here and the former in a poem alluded to
by Horace, Sat. 1.2.119 (T 4); cf. 38, Meleager 18 (AP 12.86), Argentarios 4 GP
(5.89), Rufinus 5 Page (5.18), Ov. AA 1.717 quod refugit, multae cupiunt; odere quod
instat, Am. 2.19.3 f., Prop. 2.23.12 f£, Martial 9.32 (hancvolo quae facilis etc.).
On the Appendix Barbarino-Vaticana, a collection found in three manuscripts,
cf. A. Cameron, Greek Anthology (Oxford 1993), ch. 8, who argues for its inde-
pendence from P. App. B-V lacks w. 4-6 of Phil's poem; after v. 3 it continues
with the following verse:
102 Epigram 11

Beckby strangely treats this as v. 4 of the Phil, epigram; hence he and (even more
misleadingly since they make no mention of the displaced line about Egypt) Gow-
Page state that App. B-V omit only w. 5-6). Gallavotti found this verse quoted (still
anonymous) by Simeon Metaphrastes, Vita. S. Pafapzill6, col. 337 MPG) with the
comment:
(Gallavotti also unconvincingly argues that the three-line excerpt
in App. B-V was meant to stand as an independent poem.)
This is one of many heterosexual poems mistakenly assigned to AP Book 12,
usually on the basis of a neuter diminutive proper name which can belong to either
sex. There is little excuse for misclassifying this poem, however, which has so many
indications that the names belong to women. Cf. Cameron, Greek Anthology (Ox-
ford 1993) 239-242, who blames Kephalas for these errors.

1 cf. 12.1 Eur.Hipp. 1064,

Asklepiades 8 HE (AP 5.162.3) Nonnos D.
16.297

2 Sternbach's suggestion (uncredited by Gow-Page, who call it "the

simplest suggestion") neatly both rids the text of an unwanted hiatus and properly
distinguishes between the two women. Probably near-homoeoarcton led the scribe
into error. Keeping to the MS here should entail following Petit on v. 5 (as Brunck,
Jacobs, and Diibner do). But whereas Demo can bethenameofa respectable woman
(32 exx. in LGPN 1-2; contrast 10), Thermion is obviously a redender Name ap-
propriate only to a prostitute, although Brunck et al. never consider the meaning
of this name. Gow-Page's second suggestion—that Demo as intrusive gloss ousted
Thermion in v. 2—is not credible: If Thermion had stood in the text there would
have been no need for a gloss in the first place. Nor would . .
r
be up to Phil, 's poetic standard. Similarly unpoetic is Preisendanz's
suggestion, i "Diese geht als Hetare furs Volk" (Beckby), which
produces a counterproductive paronomasia and an unacceptable hiatus. White
defends the MS, but her reading of the poem (which allows, among other things,
for Thermion to be the virgin who is then called Demarion) is not convincing.
C. De Stefani, SIFC 89 (1996) 205-206, now makes a good case for having
ousted a description of Thermion, but his tentative suggestion
is not attractive.
Not so much "knowing who K. is" as a compression of
(vel sim.); cf. Archil. 1

3 "Fondle" (Gow-Page); cf. Silenos at Eur. Kykl. Ill, numbering among

erotic pleasures Cf. 25.2 For tango as a Latin equiva-
lent, cf. Westendoerp Boerma ad Vergil, Cat. 1.4; J. N. Adams, The Latin Sexual
Vocabulary (London 1982) 185-187. Wilamowitz's conjecture (apud Kaibel) misses
the point: Phil, has (indicative) all the sex he wants with Thermion whenever he
wants; his desire for Demo remains nonetheless unabated.

4 a very rare use," as Gow-Page note, but even were the latter
metrically possible it is made unnecessary by the comparative
 Epigram 12 103

5 Samuel Petit's conjectured gen. of comparison was necessary for all

who accepted the MSS .. and who hence thought Thermion the
virgin (Observationes [Paris 1642] 1.94, approb. Brunck, Jacobs, Dubner). H. Keil,
RM 19 (1864) 263, who follows these scholars in calling Demo the courtesan, con-
siders saving the MS by punctuating
The future provides the answer to the preceding question, and reads
more smoothly than a hortatory subjunctive.
: Aparthenosneednotbeintacta (Wilam. onEur. HF 83 4), but that
is clearly the meaning here; see on v. 2.

5-6 Cf.Kallim.Ep. 1.3-6 HE (AP12.102; the hunter

Epikydes ranges over rough terrain):

Similarly, Eur. Hipp. 184 f.

Xen. Hieron 1.30; Theokr. 6.17, 11.75; Hor. Sat. 1.2.105-108 (a rendering of the
Kallimachos passage shortly before Hor. names Phil, in this same context); Qv.Am.
2.9.9 f.; 2.19.1 f., 35 f.; 3.4.17 f., 25 f. In nonerotic contexts, cf. Hes. fr. 61 M-W
, Bakch. 1.176 f., Pi. P. 3.22; Demokr.
B 202, Lucr. 3.957, 1082. See further Kaibel ad loc.

12

/4P5.132 [12 GP, 15 K]

P PI7.100, f. 73 [C]
Suda s.v. (1 +3
v. 20
104 Epigram 12

3 Jacobs: P: PI: Kaibel transpos. Griffiths

4 ' tent. Stadtmuller 5 PI: -P 6
CP1: P P: PI: Ellis: Hecker: Stadtmuller:
Opsopoeus: Waltz: Seidler 7 ppjpc.
pjac Hecker CP1: (?)P 8
pplpc.

O foot, O leg, O (I'm done for) those thighs, O buttocks, O bush, O flanks,
O shoulders, O breasts, O delicate neck, O hands, O (madness!) those eyes,
O wickedly skillful walk, O fabulous kisses, O (slay me!) her speech.
And if she is an Oscan—a mere Flora who does not sing Sappho's verses—
Perseus too fell in love with Indian Andromeda.

Cohen, Helios 8.2 (1981) 41-53.

Courtney, LCM 15 (1990) 177 f.
Geffcken 133.
Giangrande, Maia 25 (1973) 65 f.
Griffiths, SIC? 17 (1970) 36.
Munoz Valle, CFC1 (1974) 87-89.
Seidler, Ber. Verb. Sachs. Ges. d. W:ss. zu Leipzig 1 (1846-47) 128-130.

A description of a beautiful woman, feature by feature. Many parallels exist for this
in later Greek and non-Greek literature, where it is sometimes called by the heral-
dic term blason anatomique. The description of the woman may proceed from the
foot upwards, as here, or from head to toe. In the Anthology, Rufinus offers many
fine exx.; cf. also Dioskorides 1 HE (AP 5.56). The most notable echo of Phil, in
Latin poetry is Ov. Am. 1.5.19-23:

quos umeros, quales vidi tetigique lacertos,

forma papillarum quam fuit apta premi,
quam castigate planus sub pectore venter,
quantum et quale latus, quam iuvenale femur!
singula quid referam? nil non laudabile vidi.

For praise of women in general in Greek poetry, cf. K, Jax, Die weibliche
Schonheit in der griechischen Dichtung (Innsbruck 1933); A. Richlin, The Gardens
ofPriapus (New Haven 1983) 44-56.
As detailed below, most of the body parts itemized are, at least at first, given
their neutral anatomical rather than erotic names. The phrases in parentheses, on
the other hand, reveal a barely contained passion just below the neutral surface
description. By v. 5 the narrator can no longer keep up the facade and begins to list
the beloved's sexy walk and tongue kisses. The dynamic point of the poem is thus
the great difficulty if not impossibility of a man's maintaining his sang froid—perh.
more specifically his Epicurean ataraxia—in the contemplation of a beautiful
woman. The poem ends with the "X but comely" topos, of which Phil, was so fond,
where X here is the girl's low social status; see the comm.
 Epigram 12 105

1 Of surprise or exclamation with the nominative or, less often, genitive; cf.
Hipparchosfr. 3.3 K-A (a certain kind of cup), Theokr.
15.4 ("What a helpless thing I am!" tr. Gow). Generally but
not universally said by later grammarians to be barytone rather than perispomenon
cf. LSJ s.v. II 4, who cite inter alia EM 79.13
Cf. the three os of Prop. 1.10.1-4 and 2.15.1.
The description of Flora's charms starts at ground level and works its
way upwards, although not mechanically so. As Giangrande correctly points out
(contra Griffiths), Phil.'s description is as dynamic as Flora's own motions, which
soon enough end in (at least as far as the poem is concerned) kisses and an em-
brace. Cohen surveys other Classical, but nonerotic, descriptions of bodies which
proceed from foot to head or vice versa (there is an upward description of Odyssey
atOJ.8.135f,
but offers a more interesting parallel from a Dead Sea scroll (Gen-
esis Apocryphon, col. xx) in which the beauty of Abraham's wife Sarai is described
(head downwards) similarly to Flora's. Cohen's thesis is that these two nearly con-
temporary authors from Palestine were each adapting earlier detailed analyses of
beautiful women; cf. esp. Song of Songs 7.1-8 (from feet to hair, incl. the ivory tower
image for the neck). [Later blason literature consistently starts from the head and
stops short, through pointed aposiopesis, of the genitals; it hence does not men-
tion legs. Cf. Mark Taylor, "Voyeurism and aposiopesis in Renaissance poetry,"
Exemplaria 4 (1992) 267-294.]
From Homer on, = either "foot" (from the ankle down) or "leg" (from
thigh down); here, as the next two nouns show, the former. Flora is not like
Archilochos' Neoboule, who is 206 W2).

1-2 The leg divided into its two largest parts; cf. Tyrtaios 11.23
(a tall shield covers) Calves do
not elsewhere figure in erotic descriptions, but Solon 25.2 speaks of the erotikos
aner&s cf. Asklepiades20 HE (AP 12.161),
where Dorkion reveals a Simon, fr.21.5 W2, Song of Songs7.1 (and
on Song of Songs in general, cf. J. M. Sasson, JAOS 106 [1986] 736-738).
Most of Flora's features are listed without the
definite article, the exceptions being thighs, neck (3), eyes (4), and voice (6), wher
also words occur in the attributive position. presents no
problem, but the others contain finite verbs which defy easy analysis. Can one (e.g.)
speak of "the I'm-truly-ruined thighs"? Gow-Page, with some hesitation, take
as = (2, 4) or although they admit that
stretches the normal usage of Uttivopm + dative, "mad as a result of (some activity
or state)"; exx. from LSJ: , We can regard Phil.'s usage as a
special instance of the parenthetical interjection of a verbal phrase into a syntacti-
cally unrelated sentence, of which Wilamowitz on Eur. HF 222 gives several exx.
Cf. esp. Eur. Kykl. 465 (LSJ, approb. Seaford,
punctuate with only the first comma, taking it with the exx. of + dative
given above). I have accordingly, following others (cf. e.g. Griffiths 42 n. 46), set
106 Epigram 12

these phrases within parentheses. These parenthetical phrases disprove Stella's

contention (263 f.) that until the last distich we could be listening to the descrip-
tion of a statue. , "justly," i.e. "with reason."

2 . Not a word found in erotic contexts, where is far more common;

cf. esp. (Athen. 12.554d = Kerkidas 14 Powell), said of women and
of Aphrodite; (Hes. Op. 373), "rigging herself out in a
way that focuses on her arse," West ad loc. on the other hand, is neutral,
almost scientific in tone; //. 5.66 and 13.651 (Meriones' arrow twice strikes some-
one , Hdt. 4.9.1 (a sea monster is a woman from the buttocks
up). For more on the Greek appreciation of fine buttocks in motion, cf. the adjj.
(Alexis 98.11 K= 103 K-A) and (Kerkidas 14). Note also
Semonides' monkey woman: (7.75 f.),with Verdenius' note,
Mnemosyne, 4th ser. 21 (1968) 148. For the epic language of body parts and func-
tions, cf. Wackernagel, Sprachliche Unters. zu Homer (Gottingen 1916) 224-229.
The pubic area: Rufus, Onom. 109
Pollux 2.174

Again, more often neutral ("flanks, sides") than erotic ("waist, hips")
in tone; note e.g. Kallim. H. 5.88 (Chariklo speaking to her son Teiresias in Athena's
hearing) where it serves her interests to de-
scribe what Teiresias saw in clinical rather than erotic terms (a point missed by
Bulloch: "flanks are inappropriate here, being of little sexual significance"). On
the other hand, cf. [Lucian] Am. 14 (in a description of a statue of Aphrodite)
and Chairemon 71 F 14 TrGF (a description of some
dancing girls):

10

With w. 2—4, cf. Eur. Hek. 558—560, which may have served as Chairemon's model.

3 could include both shoulder and upper arm; cf. Rufus, Onom.
142.2ff, where he also notes that can include the shoulder. Synecphonesis with
final as first vowel is not so rare (West, Gk. Metre 13 offers several examples)
that an unparalleled is objectionable. Shoulders alone appear in the dual (un-
like thighs, buttocks, shanks, breasts, arms, or eyes), perh. to avoid the run of ome-
 Epigram 12 107

gas in (PI). Waltz prints P's ("per crasin pro "), perhaps
rightly.
See on 9.3.
Slenderness is a common compliment for women,
either for a part as here, or applied generally as e.g. Sappho
102.2 ,AP5.173.3, 218.6, 220.6,
Lucr. 4.1167 (the besotted lover calls a consumptive woman rhadine). Cf. Song
of Songs, where twice a beautiful neck is compared to a tower (4.4, 7.4); Ana-
creontea 16.27 West , 17.29 Note
that it is the beauty of three of the features listed here which allow Helen to rec-
ognize Aphrodite (//. 3.396 f.):

4 Stadtmiiller's tentative (app. crit. only) is defended by Griffiths,

who also alters the order of shoulders and breasts (see app. crit.) to enforce a strict
upward order; he is properly criticized by Giangrande (above, on 1 ).
appears instead of in the MSS of Euboulos 56.3 K-A [ap. Athen.] and
Semonides 27 W2 [ap. Apoll. Soph.])

5 "Wickedly skillful," as in Automedon 1.1-2 GP (AP 5.129),

is clearly the preferable reading:

Herrick's "brave vibration each way free"; cf. Sappho 16.17 f.

Argentarius 6.1 f. GP (AP5.104)
Rufinus 21.3 Page (AP5.60) (of a woman swimming
Semon. 7.75 f. (quoted above, on v. 2
This wiggling of the hips is often compared to the way a lizard shakes its
tail: Anakreon 458 ,411 Aristoph.
V. 1173 Rufinus 21.4 Page; cf. J. Bremmer, "Walking, standing,
and sitting in ancient Greek culture," in J. Bremmer and H. Roodenburg (eds.),A
Cultural History of Gesture (Ithaca 1991) 21. Cf. also Epicur. Ep. adldom. fr.131
U = 50 Arr. Ov. Am. 2.4.29 f.
ilia placet gestu numerosaque bracchia ducit \ et tenerum molli torquet ab arte latus.
A woman's sexy walk hints at her motion during sexual intercourse; cf. Brown on
Lucr. 4.1268.
This word appears as adj. only here, De Piet. 1773 f. Obbink (as
restored by Phillipson) and in an unpublished
papyrus of Hellenistic hexameters. Everywhere else only the adv. appears.

6 A hapax: "tongue kisses" is most likely; cf. Automedon 1.7 GP

with what may be the same meaning ("fellates" is less likely). According
108 Epigram 12

to Pollux 2.109, the comic poets prefer the compounds (found in

Aristoph. Thesm. 131) and , cf. Clouds 51, where Strepsiades re-
fers to the sophisticated of his wife. For other terms for kisses,
cf. Henderson, Maculate Muse 182. They must have figured in Philainis Samia's
chapter of which we have only the title (P.Oxy. 39 [1972] 2891).
Lucretius perhaps surprisingly does not mention the tongue in his two brief de-
scriptions of lascivious kisses in Bk. 4 (1108 £, 1194), but cf. Tibullus 1.8.37 f.
pugnantibus umida linguis \ oscula; Qv.Am. 2.5.23-28, esp. 23 f. improba. . . oscula
. . . ilia mihilingua nexafuisse liquet; 3.7.9osculaque inseruitcupide luctantia linguis;
3.14.23 purpureis condatur lingua labellis.
This so closely parallels the two earlier expostulations that one won-
ders why scholars have exercised themselves in producing unnecessary conjectures.
For the sense here of "kill, slaughter" (LSJ s.v. 12 b), cf. Hdt. 1.126.2 etc. (6x in
all), Aristoph. Lys. 1062.
Elsewhere only in the singular: Aristoph. fr. 753 K-A
and Klearchos fr. 2 K-A.

7 Neither "barbarous" (LSJ) nor "Italian" in general, but "Oscan" (Kaibel,

Stella 263, Gow-Page), which, assuming a Neapolitan setting for the poem, con-
notes "a local, uncultured, Campanian girl." Even Romans who never went south
knew the Oscae personae who provided the rustic flavor of Atellan farces; cf.
Diomedes 1.482,490 Keil. Cf. A. Landi, "Lingue in contatto e circolazione sociale
a Pompei," in La regione sotterrata dal Vesuvio. Atti del Convegno intern. 11-15
nov. 1979 (Naples 1982) 211-227, esp. 216, "II greco nell'Italia meridionale e una
lingua egemonica che convive con 1'osco senza assobirlo. Inoltre il greco
probabilmente era considerate una lingua dominante anche nella valutazione
intelletuale o estetica." And several sources record general disdain, esp. by Greeks,
towards the Campanian dialect: Juv. 3.207 divina opici rodebant carmina mures, with
schol. ad loc. opizin [i.e., ] Graecidicunt de his, quiimperite loquuntur.
Aliiopicos dicunt eos quifoedam vocetn habent; Eudoxus fr. 321 Lasserre ap. Steph.
Byz. Eth. 494.7 _ (sc. oi '; Joh. Lydus, Mens. 1.13
tent.
Holford-Strevens) Mayor on Juv. 3.207.
Given the greater variety of sibilants in Oscan than in Latin or Greek, there
may be the additional point that the notion of the elegant Sappho sung by a
Campanian is a funny one, esp. if in recital Phil, exaggerated the sibilants of
cf. C. D. Buck, Grammar of Oscan and Umbrian (Boston 1928;
rp. 1974) 8,20, 73 ff.
Exegetical KCUS giving two exx. of her Oscan
rusticity: (i) She bears neither a noble Roman name nor a Greek one that would
connote a certain amount of glamor. Pointing out the similarity of her name to
(Kaibel; Huet, contrary to meter, actually conjectured it) is critical over-
kill, (ii) Probably "the Cynthias and Delias did sing the poems of Sappho to their
lovers" (Gow-Page); cf. Plut. Mor. 711d. Lucian Merc. Cond. 36 speaks of women
who want it to be said of them
 Epigram 12 109

Sappho herself is the tenth Muse

(API.14,17,407; 9.506.2,571.6 f.), singing her own songs to the accompaniment
of harp or lyre. Cf. Catullus 35.16 f. Sapphica puella \ Musa doctior,
Presumably Pompey's mistress for a time, also named Flora and also renowned
for her beauty, as a professional hetaira would know how to sing Sappho (Plut. Pomp.
2.2-4), and hence cannot be the Flora described here (contra Seidler 129, Beckby).
This suggests that this poem was written before Pompey's Flora gained notoriety.
Phil, may have used the name Flora elsewhere; cf. Ep. Bob. 32 (cited in the comm.
to 2.1
Hecker's unnecessary conjecture, which violates Hilberg's Law (Maas Gk.Metre
§94), was accepted by Bignone, Diibner, Stadtmuller, Romagnoli, and Stella,
adheres closely with cf. KG 2.189.

8 Andromeda was but one of several figures in Greek

mythology considered black in color, whether called Ethiopian or Indian—the two
terms are poetic synonyms, as here, since Andromeda is often Ethiopian. (She may
have been originally Greek; cf. M. L. West, Hesiodic Catalogue of Women, Oxford
1985,147 ff.) Cf. Arrian, Indika 6.9
, Others would include Prosymnos and Staphylos. In view of Phil.'s use
elsewhere of the black-but-comely theme (cf. pp. 33, 123 £), Gow-Page (approb.
Courtney) are probably wrong to downplay Andromeda's color here, putting the
emphasis rather on her "outlandish origin." Perseus, like Phil, with Flora, was taken
by Andromeda's appearance. For Andromeda in particular as black (or white; au-
thors differ), cf. F. M. Snowden, Jr., Blacks in Antiquity (Cambridge, Mass., 1970),
153 f., 157-159. Cf. Ovid(?), Her. 15.35 f. (Sappho:) Candida si non sum, placuit
Cepheia Perseo I Andromede, patriaefusca colore suae. This may well have been writ-
ten with Phil, in mind, but, as H. Jacobson points out, whereas Ovid keeps to the
motif of "homely but cultured," Phil, inverts it: Flora's lack of culture is more than
compensated by her great beauty (Ovid's "Heroides" [Princeton 1974] 284 f.). That
tradition had it that Sappho herself was far from beautiful lends further point to the
reference to her here: P.Oxy. ISOOfr. 1 [T252Voigt]

Cf. Ovid's Candida si non sum (cited above).

The topos of mythological loves providing justification for human ones is of
course ancient; cf., e.g., the reference to Helen in Sappho 16, Theog. 1345 ff.

and in general R. Oehler, Mythologische Exempla in der alteren griechischen Dichtu

(Diss. Basel 1925), Gow on Theokr. 8.59 f. Munoz Valle points out that Horace
justifies passion for znancilla with just such paradigms (e.g.,prius insolentem serva
Eriseis niveo colore \ movit Achillem, O. 2.4.2 ff.), suggesting that Phil, may have
provided the model. Cf. also Ov. Am. 2.8.11 f. Thessalus ancillaefacie Briseidos arsit;
\ serva Mycenaeo Phoebas amata dud, Rufinus 5.1 f.,7 f. Page (AP 5.18):
110 Epigram 12

13

AP 5.24 [Meleager 41]

P Pl7.91f.,72v [sc. iv. 11
3 PP1 4 tent. Waltz in app. crit.: P: PI

My soul, knowing my earlier tears and desires, tells me in advance to flee

passion for Heliodora.
It speaks, but I have not the strength to flee; for shamefully indeed the
same (soul) both foretells and, while foretelling, desires.

Uniformly ascribed to Philodemus by P, PI, and, presumably, H; given to Meleager

by Jacobs and most others (Mackail is an honorable exception) on the basis of the
name Heliodora, which appears in sixteen other poems of the latter, one of which
(Meleager 54) is also ascribed to Poseidippos (23 HE = AP5.215). But Loukillios
also addresses a Heliodora (AP 11.256), and Phil, uses the "Meleagrean" names
Kallistion (57 =/lP5.192) and Demo (23 =,4P5.197, etc.) = Demarion (24 =5.198).
The use of the plural points towards Meleager, but not decisively (see
comm.). The poem's context within the Anthology in indecisive; HE 2.631. The
evidence is clearly insufficient to deny the poem to Phil.; so also Cameron, Greek
Anthology 387; E. Havelock, Lyric Genius of Catullus (Oxford 1939) 143.
Once this poem is restored to its rightful place in the Philodemean corpus, it is
possible to see that its point rests upon some Epicurean notions of the soul, which
is (i) corporeal and (ii) unitary, i.e., the body has but one soul, although (iii) it is
not homogeneous. For this last point our chief testimony is Ep. Herod. 63, which
speaks of differing amounts of wind and heat in the soul, and of one part
Woltjer: codd.) of exceeding fineness. Epicurus uses these differences to
explain Much remains unclear in
 Epigram 13 111

this compressed account (cf. G. B. Kerferd, "Epicurus' doctrine of the soul,"

Phronesis 16 [1971] 80-96), but later Epicurean sources speak more explicitly of
the one soul having both rational and irrational parts, which Lucretius in Book 3
designates as animus and anima respectively. Cf. £ ad Epic. Herod. 66-67
(311U),VH 2 7. 17 col. 22

(313 U), Act. 4.6.6

4.5.5
(312 U), Diog. Oin. fr. 37 Smith. Even if (which I doubt) Kerferd is correct in dis-
tinguishing between Epicurus' simple soul and the divided soul of later Epicurean
tradition, since Phil, belongs to the latter, the analysis of this poem along these lines
remains consistent with the author's beliefs (see below, on 4 and
Once again, the persona adopted by Phil, is that of someone who knows Epicu-
rean teaching on the subject and who would like to follow its precepts, but who
finds himself slipping from his ethical model.

1 The poem describes an address of the narrator's soul to the narrator, just
as in earlier poetry other parts of the body engage in internal dialogue; e.g. Odysseus
, and, in the other direction, the chorus of the
(99 Iff.). Note
also 1. For other "split" souls in erotic contexts, cf. Kallim. 4 HE (AP 12.73),
Asklepiades 17 HE (AP 12.166).
To be understood cmo KOVVOV with and
The sense "warn" is but a slight extension of the basic
meaning "say beforehand," which, like the simplex, can have imperative force with
an infinitive; the infinitive here is thus not so odd as Gow-Page say. Cf. Deinarchos
1.71 Eur. fr. 897.9f. N2
Phil, uses the verb in the sense "said before,
warned," at 23.5. (In his prose he uses the verb merely to refer to what he had said
earlier.) Gow-Page's comments are especially overfussy in that Phil, seems to be
playing on the two meanings of the verb on the last line.
is common in erotic contexts but it is also found in much ethical writ-
ing—in Epicurus most notably fr. 163 U
; other exx. s.v. in Glossarium Epicureum.
With an objective genitive, as usual (LSJ I), a favorite word of Sappho
for sexual passion (it is decidedly unsexual in Homer), and exceedingly common
in this sense thereafter.

2 . Men frustrated in love cry: Asklepiades 18.3 HE (AP 12.135).

The plural is largely prosaic and, except for PL Laws 679cl,
postclassical. Poetic passages are: Meleager 64.2 (AP5.190) and 102.4 (AP 12.70),
Philip 70.6 (API 137).
"Knowing for certain," since knowledge is based on perception
which cannot be false; cf. S. Everson, "Epicurus on the truth of the senses," in
S. Everson (ed.), Epistemology (Cambridge 1990) 161-183.
112 Epigram 13

3 For this accentuation at the beginning of a clause, rather than the usual
cf.Tyrannionfr.
9 Haas(Eustath.ll. 1613.16ff.adll.2.350);J. wACKERNAGEL,
KS2.1068; Schwyzer, Gr.Gr. 389;J. Vendryes, Traite d'accentuationgrecque (Paris
1904; rp. 1945) 108-110; M. L. West, Aeschyli Tragoediae (Stuttgart 1990) xxxi.
For the collocation of particles, cf. Denniston, GP 284. goes with the
adj., as usual (ibid. 280). This accentuation seems preferable to in allowing
"the same,"rather than , "the shameful one itself." The point
is that one and the same soul can act in totally contrary ways. See next lemma. For
the predicate word order and adverbial sense of an adj. indicating mental state, cf.
Xen. Kyr. 1.6.2 oi KG 1.275.
Use of this word argues for Philodemean authorship; see on 18.5.

4 The soul (Diibner, Waltz, Gow-Page), not the woman (Paton). The Epi-
curean soul comprises both rational and irrational parts, respectively the animus
and anima described in Lucretius III; cf. . ad Epic. Herod. 68 £, quoted above.
Although the soul may be said to have parts,
it is but one and corporeal (cf. Epic. Herod. 63
Rist, Epicurus 79 f.). While the rational
part of the soul "speaks beforehand" (in warning) of desire for Heliodora, the soul's
irrational part must also be speaking beforehand of this same desire—and so stirs
it up. Lucretius' analysis is somewhat different, but note 4.1048 idquepetit corpus,
mens unde est saucia amore, 1057 voluptatem praesagit muta cupido, 1106 cum
praesagit gaudia corpus. (For Lucretius' peculiar use of this verb, cf. Brown on
4.1057.) The most interesting parallel, however, comes from Epicurus himself:

(200 U).

14

/IP5.123 [ 9 G P , 4 K , 6 G]
P v. 3 caret PI [C]
2 Dilthey: Knaack 4 C: P
G6tting.philol.6, Lips.Rep. I. 35,Cr: P: Ap.B 5 P: Ap.B
(marg.)
 Epigram 14 113

Shine, o nocturnal bicornate lover of allnight revelry, Selene; pass through

the latticed windows and shine.
Illumine golden Kallistion. There is no ill-will directed towards your im-
mortal self when you gaze down upon the actions of lovers.
You count both her and me happy, I know, Selene; for your soul too was
inflamed by Endymion.

Hopkinson 75, 263.

Knaack, "Analecta," Hermes 18 (1883) 31.

In tone the most lyrical of Phil.'s epigrams. The path of light is traced from Selene
through the windows onto the (probably nude) body of Kallistion; and moonlight,
as Selene's visual rays, will continue to be present during the (presumably immi-
nent) lovemaking. Selene's own love for Endymion is then recalled, his name, the
poem's last word, providing the final detail of the narrator's description: Kallistion
is asleep during his address to Selene, and, if the parallel with Endymion contin-
ues, will remain asleep during lovemaking; see on 6
Cf. Prop. 1.3, where a description, again with mythological paradigms, leads
to embraces and kisses of the still sleeping Cynthia. Propertius' references to moon
and windows (see below on w. 1, 2) suggests an allusion to Phil.; so P. Fedeli, S.
Properzio: II primo libra (Florence 1980) ad 31-33, 32; O. Pecere, "Selene e
Endimione (Anth.Lat. 33 R.)," Maia 24 (1972) 304-316, who surveys the Selene-
Endymion motif in literature. Agathias 90 Viansino (AP5.294) also describes the
stroking and kissing of a sleeping woman. On the eroticism of sleeping figures in
art and literature, cf. E. J. Stafford, "Aspects of sleep in Hellenistic sculpture," BICS
38 (1991-1993) 105-120, esp. 109-112.
As Kaibel notes, this poem was read and imitated by the twelfth-century au-
thor Niketas Eugeneianos, Drosilla and Charikles 8.113-115.

Niketas' source was the Anthology, not Philip; cf. A. Cameron, The Greek Anthol-
ogy 128 £, 341.

1 "The other epithets in this couplet are carefully chosen rarities, and
it is surprising to find here the common form instead of the poetic
(Gow-Page). Phil, could have seen the latter in Aratos, Phain. 999,
modifying , and in Maccius 10 GP (AP 9.403), modifying But
is not alien to poetry, occurring in Aristoph., Eupolis, and SH 1090.
It is credible that Phil, wrote the rarer adj., which was altered subconsciously by
one of his learned editors (rather than by a careless scribe), but not enough so to
justify altering the text of the MSS. (It is unfortunate that stops just short of set-
114 Epigram 14

tling the matter, but since its usual custom is to omit more than one letter, it may
be thought to favor the MS reading.)
For the anomalous accentuation of compound adjj. in , cf.
Kiihner-Blass 1.321. Applied elsewhere to Pan (Horn.Hymn 19.2, Agathias 62
Viansino = AP 6.32.1, and AP 9.142.1) or an animal (Aristotle, HA 499bl8), but
note Horace, Carm.Saec. 35 f. bicornis . . . luna and Orph.H. 9.1 f. ZeA,f|vr|

The only compound adj. beginning ; occurring else-

where only at Orph. H. 3.5 (of Night). The piling up of epithets is typical of hymns
(increasingly so in later literature; Orphic Hymns—cf. 9 in particular—contain
vocative epithets of this sort almost exclusively).
Intransitive, as often of celestial phenomena, from Homer on (LSJ s.v.
A II); cf. Theokr. 2.10 f. Meleager 73 (AP 5.191.1) f|

Notethat, according to Schol. ad Theokr. 2.10.

(fr. 104 Sn.-M.). Cf. also Propertius' men-

tion of the moon at 1.3.31 f. (quoted in the next lemma).

2 the singular is more normal in early Greek; cf. Praxilla 754

PMG
vi3u<|)a (but note that R. Renehan, "Praxilla fr. 8 Page," Hermes 115 [1987] 373-
377, argues for , which may be echoed in Asklepiades 3 HE (AP
5.153.2), where the beautiful face of Nikarete is seen
. Phil.'s plural probably reflects a later style of double-gated window;
cf. Ov. Am. 1.5.3 pars adaperta fuit, pars altera clausa fenestrae, Prop. 1.3.31 f.
donee diversas praecurrens luna fenestras, \ luna moraturis sedula lumintbus, and
next n.
For visible rays, cf. Lucr. 2.114-141. It might also be pertinent that according
to Aristotle, de An. 404al, Demokritos compared the atoms of soul to the motes
seen in rays coming through window

The original meaning "well-bored" yielded to "many-holed," here

"latticed." Cf. Varro RR 3.7.3 fenestris reticulatis, Plautus Mil.Glor. 379 fenestra
clatrata, Vergil A 3.151 f. qua se plena per insertas fundebat luna fenestras.
To the exx. of this verb used for the "casting" of light adduced
by Gow-Page, add Asklepiades 3.1, cited above foi
(Dilthey's conjecture is recorded by Knaack.)

3 : For the sense here, "shine on, illumine" (LSJ s.v. II1), used of the su
moon, or stars, cf. Eur. Hek. 635—637 . . .
(but Dodds ad Eur. Ba. 596-599, says that it "may mean
merely'whom the sun sees'"), Ap.Rh. 3.1377 f. Orph.H.
7.10. Another possible parallel is Leonidas 10 HE (AP 7.648.8), where a room bright
with torches may illuminate (i.e., be brighter than) a log burning in the fireplace:
 Epigram 14 115

. For a different interpretation, cf. M. Gigante

L'hedera diLeonida (Naples 1971) 71.
The name occurs in Hedylos 3 HE (Athen. 486a), Kallim. 16
HE (AP6.148), Meleager57 (AP5.192), and Poseidippos 8 (AP 12.131). Probably
all were named after the well-known hetaira who appears in Machon 433 Gow and
mentioned by Athen. 858b.

3—4 The same as the (Hes.

Op. 521 =H. Hymn5.l); sim. ibid. v. 6, Solon26.1,Theokr.Epig.4HE (AP9.437.4),
and4P7. 1.1 (Anon.), 9.416.1 (Philip52), 9.157.7 (Anon. 85 FGE). Cf. 11.5.429

4 Rare in poetry (Soph. Ajax 829, Phil. 124); it is, however, a par-
ticularly apt word for the observing of heavenly phenomena: Phil. De Signis 25.33
33.12
Ps.-Aristotle,DeMundo391alO;cf. Geminus.E/ew. 16.24,33. There
is thus some wit in reversing the normal application of this verb: One of the heav-
enly phenomena observes two humans.
The gender shows that this statement is not general, but applies
specifically to Selene: "positum ex more poetarum pro " (Jacobs). (Some
apographa, keeping P's reading, punctuate as though were vocative, but
this is not likely.)

5 Cf. Fraenkel on Aisch. Ag. 928 f.

Trisyllabic, in accord with Naeke's law.

6 "For . .. also"; Denniston GP 108.

The story of this love affair was told as early as Sappho: Z Ap. Rh.
4.57 f. (p. 264 W) = Sappho 199 Voigt, although testimonia are few and details
vary. The longest account, Apollod. Eibl. 1.7.5 = Zenob. 3.76, says merely that
"Selene fell in love with Endymion. . . . Zeus granted him a wish. He chose to sleep
forever deathless and unaging." Cicero, on the other hand, says that it was Selene
who put Endymion to sleep so that she could kiss him while he slept (Tusc. 1.38.92).
Cicero's oscularetur suggests repeated action, as well as euphemism, which would
further explain the 50 children she had by him according to Paus. 5.1.4. The pic-
ture of Selene viewing the sleeping Endymion becomes a common one in art; cf. H.
Gabelmann, "Endydmion," LIMC 3.1.726-742. Closest in erotic tone to Phil, is
Prop. 2.15.15 f. nudus et Endymion Phoebi cepisse sororem \ dicitur et nudae
concubuisse deae. Cf. also Ov. Her. 18.59-65, where Leander prays to the Moon,
reminding her of Endymion.
As suggested above in the introduction, the mention of Selene's love for
Endymion fills in the picture that the listener has been developing: Kallistion is lyin
asleep alongside the narrator, and he will try not to wake her as he begins to make
love to her.
116 Epigram 15

15

/IP 5.25 PGP, 5 K, 7 G]

P PI 7.92, f. 72 [sc. Suda s.v. (3 -A)

3 PI: P 5 P: PI:
Salmasius: Jacobs: Stadtmiiller: Gow-Page: Lumb:
Brunck PP1: Desrousseaux Gow-Page 6
Brunck Boissonade Gow-Page ed.vet. (1494)

All the times I dare whether by day or evening to come to rest on Kydilla's
bosom,
I know that I cut a narrow path along a precipice, I know that each time I
risk my head on the throw of the dice.
But what's the use? For she is bold, Eros, each time when she drags me
and altogether knows not even the dream of fear.

Lumb 9.
von Prittwitz-Gaffron, Das Sprichwort im griechischen Epigramm (Giessen 1912) 32 f.

Doubts have been expressed over the authorship of this poem since the preceding
poem in P (13) has been given to Meleager; see, e.g., Stadtmiiller (who thought also
of Asklepiades as author) and Waltz. The presence in of the incipit to 13 should
settle the question of authenticity for both poems.
The topos illustrated here is a variation of that of the "Poet Caught by Love"
(S.L. Taran's phrase: The Art of Variation in the Hellenistic Epigram [Leiden 1979]
103), in which we typically see the man willing to risk all; cf., e.g., Anon. 9 HE (AP
12.99.1) In this poem, however, the poet seems a prisoner
less of Eros than of the woman who not only is herself subject to insatiable desires
but who also, most likely because she is married, induces fear in the man as he thinks
of the various punishments meted out to adulterers; see on 22.6. This is particu-
larly applicable to Rome—cf. C. Edwards, The Politics of Immorality in Ancient
Rome (Cambridge 1993) ch. 1, "A moral revolution? The law against adultery,"
esp. p. 56—but this poem also presents a humorous counterexample to the Epicu-
rean view that the sexual pleasures of adultery are more than canceled out by the
 Epigram 15 117

thought of punishment. Note Origen, Contra Celsum 7.63 = Epic. fr. 535 U (Epi-
cureans do not avoid adultery because it is morally wrong or unnatural)

Even though Origen's account has to be viewed within the wider context of anti-
Epicurean criticism (cf. P. A. Vander Waerdt, "The justice of the Epicurean wise
man," CQ 37 [1987] 402-422), his general point nonetheless provides the back-
ground both for this poem and the equally Epicureanly colored Hor. Sat. 1.2 (cf.
Q. Cataudella, "Filodemo nella Satira 12 di Orazio," PP5 [1950] 18-31); i.e., the
Epicurean sage would do no wrong even were he sure to escape detection and
punishment, although Epicurus thought such certainty unlikely.

1 Only thrice in Homer (+ in Simonides), and then not again

found until Kallimachos (H. 4.254, Ep. 34.2 HE = AP 7.80) and Phil.
LGPN 1 records one instance of (ii c. B.C. Delos); prob.
more significant is Herodas' use twice of cf. Headlam-Knox on Herodas
5.9.
Gow-Page understand "whenever I lie on Cydilla's bosom."
It is tempting to prefer their second choice, i.e., to take w. 1-2 —
which seems livelier. More likely, however . is to be taken
. Cf. Od. 9.45 If. (Polyphemus to his ram)
Cf. KG 1.274 a) and b) for
adjj. used as the equivalents of adv. phrases; F. Letoublon,
Predication, attribut, et apposition," in A. Rijksbaron et al. (eds.),
In the Footsteps o/R. Kiihner (Amsterdam 1988) 161-175.
The word in the sense found here seems to be due to Theokritos:
(14.37), although Kallimachos (earlier or later?) used it in
the sense in utero (H. 4.86). Maccius (earlier or later than Phil.?) also uses it in the
erotic sense (4 GP =AP5.130.3); cf. also Automedon 11 GP (AP 12.34.3) elqamou

The name of Petronius' Encolpius is similarly formed, but whereas

is only mildly erotic, Encolpius has vulgar overtones; cf. J. P. Sullivan, TheSatyricon
of Petronius (London 1968) 117; Henderson, Maculate Muse 140 f.

2 "Philodemus as an adulterer at Rome is risking not only his life . . . ,

but indeed a fate far worse than death (Hor. Serm. 1.2.45ff.)," A. H. Griffiths,/HS
90 (1970) 218. The word does not, therefore, apply any more to his evening forays
than to those during the day (sic Kaibel, contra Gow-Page). On adultery in Rome,
cf. Edwards, op. cit.; A. Richlin, The Garden ofPriapus (New Haven 1983) 215-
219.
118 Epigram 15

An iterative clause with indicative, rather than the more usual sub
junctive + av or optative, emphasizes the factual nature of the specific occurrences;
cf. PL Charm. 158a Xen. Mem. 3.43, KG 2.451.

3 Even if Phil, invented this phrase on the spot, as

Gow-Page suspect, they should not therefore reduce it to mere metaphor; it still
has a proverbial character (Prittwitz-Gaffron). Jacobs compares OvidA4 1.381 no
egoperpraeceps etacuta cacumina. vadam, which is certainly pertinent as Ovid a few
lines earlier had written alea grandis inest. And Ovid's concern in this passage with
concealed adultery, which is reminiscent of Phil's, suggests that he probably had
this epigram of Phil, in mind. For = "make one's
way," cf. Aisch. Supp. 545 Trag. Adesp. 668.6 TrGF, Eur. fr
124 N2, Aristoph. PL 69f.

3—4 Phil, uses the simplex for the more usual (LSJ
s.v. II). must mean "bet one's entire stake"; cf. Thuc. 5 . 1
Plut. Fab. 14.2
Brut. 40.3 Demosth. 20.3 (cited below),
Arat. 5.4. Without a parallel, Gow-Page are right to question the equation of
and , but we would also like a parallel for throwing the dice over one's
head, which is how they would understand the phrase. I think that Phil, is in fact
using the word in just such an extended sense; cf. Plut. (who is clearly fond of dic-
ing metaphors) Dem. 20.3

5 I.e., What is the use of knowing, if one continues the danger

ous practice? On in the sense quid prodest?, cf. Kaibel ad loc., Gow on
Theokr. 8.17.
P does not scan, and PI must be faulty because "the ordinar
rules of rhythm are strongly against punctuating before " (Gow-Page ad loc.).
Most editors (Brunck and Jacobs are exceptions) accordingly punctuate afterwards
and offer new beginnings for the sentence, understanding Eros to be the subject.
(Note that calls for EOT', although elision is usually avoided at the
midline caesura; GP l.xlii.) Desrousseaux and Boissonade think that the corrup-
tion goes further. These earlier edd. are properly criticized by Gow-Page, who,
however, proceed to take the greatest liberty with the text:

Although Eros may indeed be a vocative (as often in amatory epigrams and as I
punctuate), in saying that "it is the victim of Eros [rather than Eros himself] who
puts aside all thoughts of danger here" they assume that only the male can be such
a victim. My text can be read with as nominative, but the poem seems more
effective as an appeal to Eros and with Kydilla as the subject of
 Epigram 16 119

It is easy to see, however, that once miscopied as perh.

because of a misunderstood abbreviated ending, could lead someone unsure of
where the preceding sentence ended to transpose with to save the meter. (A
postpositive may occupy the second half of a resolved position 8 without violating
Hermann's Bridge; West, Greek Metre 38 n.18.)
Cf. 27.2 n.

6 Adverbial is rare in poetry and seems always to be found with the

negative (Soph. Ant. 92, El. 439, Ph. 1239).
For many exx. of this phrase, cf. Knox on Herodas 1.11
The reading suggested for the previous line replaces a merely accept-
able but certainly unimaginative close with a witty turn of thought: "I know [twice
stated] that I risk my life + w. 3^0 going to see Kydilla at all times
of the day and night. But what can I say? She is audacious, and when she drags me
to her at any time, Eros, she does not know the meaning of the word fear." That is,
she does not know the meaning otPhilodemos' fear. Cf. Hor. Sat. 1.2.127 ff. (which
follows soon after a reference to Phil, and a similarly adulterous situation) necvereor
ne, dum futuo, vir rure recurrat, and when the husband does in fact return, 131
cruribus haec metuat, doti deprensa, egomet mi. That all three ol8a phrases are pro-
verbial in character (Prittwitz-Gaffron) adds further point: my two proverbial say-
ings count as nothing if she does not know the third.

16

AP 5.124 [10GP.6K, 1 G]
P PI 7.90, f. 72v [sc. [C]
2 P: PI 3 CP1: P? 6 P: PI

Not yet bare of its cover is your summer growth; not yet do you have a
dark grape cluster to shoot forth the first rays of a young girl's charms,
but already the young Erotes are whetting their swift arrows, Lysidike, and
a secret fire smolders within.
120 Epigram 16

Let's flee, unfortunate lovers, while the arrow is off the string. I am a
prophet of a great and imminent blaze.

Macleod, "Horatian imitatio and Odes 2.5," Collected Essays (Oxford 1983) 245-261.
Schulze, BPhW16(m6)3l9.

With this poem cf. the less elaborate Maccius 2 GP (/IP5.117):

Lysidike, like Demo in 11, is sexually immature but none the less desirable. Hor.
O. 2.5 is an obvious borrowing from Phil; see Macleod and below on 1

The metaphorical equation of the human body and vegetation is a poetic com-
monplace;cf. Alk. 119 (adduced by Macleod), Ibyk. 286.3-6

Aisch.^g. 1391 f.
Cf. further C. Segal, "The tragedy of the Hippolytus: The waters of Ocean and the
untouched meadow," HSCP 70 (1965) 117-169; repr. in his Interpreting Greek
Tragedy (Ithaca 1986) 165-221.
Phil.'s authorship of this poem has been unnecessarily doubted by Kaibel and
Stadtmuller. Kaibel thought that the style belonged to an earlier age, but the poem
is solidly embedded in a Philippan context: AP 5.104-133. Stadtmuller suggests
either Argentarios or Bassos (the author of the next poem in P).
The language of this poem, in praise of the charms of a virgin and anticipating
with some dread a great blaze of passion, shows that it is, perhaps at several re-
moves, a variation of an epigram of Asklepiades in praise of an elderly hetaira who
will not enflame the lover (41 HE = AP 7.217):

1 Cf. Hes. WD 521 10.293

Hor. O. 2.5.1 f. nondum subacta ferre
iugum valet \ cervice, which owes a debt to Phil, and to Asklepiades 23 HE (AP
12.162); cf. Macleod 98 ff., Tait 73.
The kalyx here is not the bud of a flower (as transla-
tions seem to suggest), but the sheath of grain; see below on The reference
may be general; cf. 33.8, Aristoph. fr. 483 K-A
Or, more likely, given the next clause, it may be a specific reference to Lysidike's
forthcoming pubic hair; for the sense "hair" cf. Kallim. H.Del. 4.298 f. jralSec;
 Epigram 16 121

Apollonides 26 GP (AP 10.19.1 f.) Cf.

further E. Eyben, "Antiquity's view of puberty," Latomus 31 (1972) 677-697, esp
691 f. on hair.
= "summer fruits" (LSJ s.v. II), and as such suggesting imminent har
vest. Although + gen. of separation is an extension of the word's basic
meaning, the erotic connotation should not be forgotten here.
The negative of exegetical tccd; see next lemma.
This is the only instance of the intransitive use of the active in po-
etry; cf. PL Ti. 83 a, Theophr. De Igne 50. The chromatic range of includes
the dark blue-purple of grapes; cf., e.g., //. 18.562 (on Achilles' shield)
see further M. Platnauer, "Greek colour perception," CQ 15
(1921) 153 f.; A. E. Kober, The Use of Color Terms in the Greek Poets (Diss. Co-
lumbia 1932) 25-36; V. J. Bruno, Form and Color in Greek Painting (New York
1977), esp. 83-85. Cf. Cat. 17.15 f. puella ... I adservanda nigerrimis diligentius
uvis (Tait 47 n. 91), Hor. O. 2.5.9 ff. tolle cupidinem \ immitis uvae: iam tibilividos
\ distinguet autumnus racemos \ purpureo varius colore.
This passage has been adduced by scholars trying to supplement Archil. 196a. 17
W2 (First Cologne Ode) [, some even restoring (I think
that the man here is referring to the coming night rather than the onset of either his
or the girl's puberty.) Cf. S. R. Slings in J. M. Bremer et al., Some Recently Pound
Greek Poems (Leiden 1987) 37.

2 Hair; cf. Agathias 74 Viansino (AP 5.287.6)

Nonnos 2.197 here of course the pubes,
which has somewhat the same triangular shape as grape clusters. Cf. Aristoph. Nu.
978 Ekkl. 13
Henderson Maculate Muse 136.
Cf. Asklepiades 41.3 f. (above).

3 Cf. Hor. 0.2.8.14 ff.ferus et Cupido \ semper ardentis

acuens sagittas \ cote cruenta. For = "arrows," cf. Eur. Ion 524, Meleager 8
(AP 5.180. I f . ) [Note Asklepiades 16 HE =AP
12.50.3 Boissonade et al.) cf.
Giangrande, "Sympotic literature and epigram," EH 14 (1967) 129 for a defense
of the MSS] The missiles of love appear as early as Aisch. PV649 f.
In general, cf. F. Lasserre, La Figure d'Eros dans la
poesie grecque (Lausanne 1946) 90 ff., 155 f.; esp. ch. 7 for a survey of Eros in
Alexandrian epigrams, pp. 150-171.
Should this be taken merely as an ornamental epithet for the tradition-
ally young Cupids (cf. the plates in LIMC s.v. Eros), or as an epithet transferre
from the young Lysidike?
Eros first becomes pluralized in the fifth century (Pi. N. 8.5 ff., fr.
122.4, Bakch. 9.73, Aisch. Supp. 1042, and [Simonides] 1005.2 PMG); cf. T.
Rosenmeyer, "Eros-Erotes," Phoenix5 (1951) 11-22.
122 Epigram 16

4 Not quite the hapax legomenon Gow-Page think; later writers use
the adj. to describe bread cooked "hidden" in ashes, i.e., = classical (sc.
. (And is the Septuagint's word for bread/cake, e.g. Gen. 18.6, Ex.
12.39)

5-6: The asyndeton indicates the haste with which all must act if they are to escape
in time; for ancient discussions of asyndeton, cf. D. A. Russell on Longinus 19.

5 Cf. Prop. 1.1.35hoc, moneo, vitate, malum, l.93Qquisquises, assiduas

afuge blanditias.
"Those whose is obsessive"; Barrett ad Eur. Hipp. 191-197.
plus present indicative (here entailed by ) is rare; cf. Hdt. 3.134.3

6 Cf. Asklepiades 41.4 (above). For the thought in general, cf. Prop.
1.9.17 f. vero nee [=necdum\ tangeris igni: \ haec est venturiprima favilla mali.

17

AP5.121 [8GP, 14K.5G]

P PI 7.98, f. 72v [sc. ii. 19 Suda s.vv.
(1 3 (2 (3 0]

2 PSuda (s.v. PI: Suda

Small and dark is Philainion, but with hair curlier than celery and skin
tenderer than down;
and with a voice sexier than Aphrodite's she offers her all, often forget-
ting to set a price.
May I love such a Philainion until, golden Aphrodite, I find another, more
perfect one.

Prinz, WS)4 (1912)230.

 Epigram 17 123

One of several poems by Phil, praising the sexual charms of a woman not meeting
the standard criteria of beauty; cf. 9 (sixty-year-old Charito), 16 (prepubescent
Lysidike), 12 (rustic Flora); andperh. also ii. 27 iii.
would appear to be a counterpart to this one (so Alan Cameron). There
are several topoi concerning the beloved's faults: (i) The lover who has so lost his
wits that he is unaware of the beloved's flaws is described by a third party, for which
Plato Rep. 474d-475a and Lucretius 4.1160 ff. are the loci classici. (ii) The lover uses
euphemism and other forms of flattery knowingly in order to seduce the woman.
Philainis of Samos' seems to be the literary model for
.. (P.Oxy. 39 [1972] 2891 fr.
col. ii.3-7). Cf. Ov. AA 2.657-662 nominibus mollire licet mala (cont'd below) (iii)
Phil.'s version: The lover, aware of what are normally taken as flaws, tallies them against
her charms for a third party in order to demonstrate that the latter outweigh the former.
This allows for a certain amount of detachment, such as here where Phil., far from
swearing eternal love, would have it last only until he finds a more perfect woman to
love. What Phil, admits piecemeal Ovid accepts and tallies in one poem: Am. 2.4; cf.
esp. 9 non est certa meos quae forma invitet amores, 47 f. denique quas tola quisquam
probet urbe puellas, \ noster in has omnis ambitiosus amor.

1 Cf. Ov. Am. 2.4.35 haec habilis brevitate sua est, AA 2.661 die "habilem"
quaecumque brevis; Lucr. 4.1162, where the woman who isparvula, pumilio is called
totum merum sal\3y her besotted lover. For a discussion of this well-
known passage and its Greek models and Roman parallels, cf. R. D. Brown, Lucretius
on Love and Sex (Leiden 1987), 128 ff., 286 f. C. D. Buck, Greek Dialects (Chicago
1955) 76, compares the form common in Doric and other dialects, to "dou-
bling in hypocoristic proper names, where it originates in the vocative and is due
to the emphatic uttering in calling."
Obviously " (being) black," rather than LSJ's "grow<ing black,"
hich seems lexically unlikely in any case for a verb in Cf. Theophr. de Igne
50 "red and black." Plato's original is
(Rep. 474e). Cf. also Ov.Am. 2.4.40 est etiam infusco grata colore
Venus, AA 2.657 Lfusca vocetur \ nigriorlllyrica cuipice sanguis erit. With the word
black, this poem qualifies for the black-yet-comely genre; cf. M. Gigante, Civilta
delle forme tterarie nell'antica Pompei (Naples 1979) 189 f., who briefly discusses
this motif i connection with an interesting graffito found in Boscotrecase:

quisquis amat nigra(m), nigris carbonibus ardet,

nigra(m) cum video, mora libenter aedeo [=edo].

This motif shows up again in 12, where the narrator's desire for Flora is compared
to that of Perseus for the black Andromeda. in ii. 27 may be another refer-
ence by Phil, to a black woman; cf. commentary ad loc. Men were expected to ex-
plain how it was they were attracted to women who fell outside their society's idea
of what constituted the normal range of charms (cf. Brown 280 ff.); in this case,
women who were black or swarthy (we do not know which is true of Philainion),
either because they worked outside the house or because they were more "Medi-
terranean," i.e. olive-complexioned, than the norm. Thus, Theokr. 10.26 f.
124 Epigram 17

of a woman likened to a Syrian, who, it should be noted, may well have

been of the same skin color as Phil, himself. Cf. further Asklepiades 5 HE (AP 5.210.3)
Vergil EC/. 10.38 f., Ov. Her. 15.35 f. (quoted
on 12.8). For ancient attitudes towards blacks in general, cf. F. M. Snowden, Jr.,
Blacks in Antiquity (Cambridge, Mass. 1970), esp. ch. 8; id. before Color Prejudice
(ibid. 1983), esp. ch.3.
Is Phil, signaling a literary debt to Philainis with this choice of
name? Both names are probably diminutives of cf. K. Tsantsanoglou, ZPE
12 (1973) 192 n. 35. Two other erotic Philainia are found in Asklepiades 8 HE (AP
5.162) and Anon. 40 HE (AP 6.284). See on 1.1.

1-2 Sc. , as in Archil. 238 W and Theokr.

20.23 (a rejected shepherd describing his earlier appearance favorably)
Lucian calls this a literary commonplace:

(Pro Imag. 5). Ps.-Lucian, Amores 26 may contain an echo of Phil.:

African blacks had nota-
bly "woolly" hair (cf. esp. Hdt. 7.70, Snowden [1970] 6 f.), but since this adj. oc-
curs on the "comely" side of the description, it does not share in the general atti-
tude towards black skin. When Athena, e.g., makes Odysseus more handsome, Kcx8
(Od. 6.230 £ = 23.157 f.).

2 "fine, soft down, as on young birds" (LSJ). Cf. Aristoph. Danaids fr. 268
K-A , quoted by Pollux 10.38, who says
Hippocr. Mul. Affect. 1.61 uses this word to describe
the skin over the spleen of a hydropsiacal patient. Although the color of Ethiopi-
ans' skin is the feature most likely to be mentioned in Greek and Latin poetry, there
seems to be no comment elsewhere on how it felt to the touch, perhaps because
there is no signficant difference. The tenderness of Philainion's skin, then, is not
specifically related to her blackness; Phil, is simply commenting as any lover would
about his beloved.

3 A compendious expression combining two related

ideas found together in the locus classicus for the the , Il. 14.214–
217:

The anthropomorphic figures of Love, Desire, and Allurement are embroidered

into the garment; cf. H. A. Shapiro, Personifications in Greek Art (Kilchberg 1993)
19. Similar is the language of Achilles' great shield, where is used of the
images fashioned (e.g., 18.483). But these qualities, including the power of sexy
talk which can cloud men's minds, accompany Aphrodite's breast halter and work
for whoever wears it. Aristot. EN 1149bl5 associates the kestos chiefly with seduc-
tive words (Homer's ), as does Phil, himself, Rhet. 2.289 Sudh. (a refer-
 Epigram 18 125

ence to Hermes' magic wand)

Cf. Phil. De Piet. P.Herc. 1648 fr. 3.11-15 Schoeber

(CErc 18 [1988] 95); C. A. Faraone, "Aphrodite's KEOTOI; and apples for Atalanta:
Aphrodisiacs in early Greek myth and ritual," Phoenix 44 (1990) 219-243. On
KeoToq as a noun, cf. McLennan ad Kallim. H. 1.14. Unpacked, then, Phil.'s phrase
becomes "speaking with more sexy magic than one who wears Aphrodite's kestos."
Cf. API 16.288 and Antiphanes 1 GP (AP 6.88):

On the kestos, see further C. Bonner, and the saltire of Aphrodite,"

AJP 70 (1949) 1-6 (the heraldic term "saltire" represents "a diagonal or St. Andrew's
cross"); W. Burkert, The Orientalizing Revolution (Cambridge, Mass., 1992) 93
who says that the kestos "seems to be oriental in a particular way"; cf. Janko on II.
14.214-217.
occasionally occurs as adj., cf. Philostr. VA 1.2
Aesop 56 Haus. (if not in a substantive in opposition); AP 4.3.71 (St.
Greg.) [For MSS at Phoenix Ninos 1.5 Powel,
Kaibel read the adj.

3-4 "Acquiescing to any sexual desire" on the part of the man;

cf. Aristoph. Lys. 362 f. (chorus of women)
AP 12.232.4 (Skythinos) cf.
Henderson, Maculate Muse 161.

4 For women who charge little or nothing, cf. 22.1-

3; Ov. Am. 1.10, esp. 47 partite, formosae, pretium pro noctepacisci; Mart. 9.32.5 f.

6 Sc. ;cf.Theokr. 11.76

Lucr. 4.1173 nempe aliae quoque sunt (with Brown's n.).

18

5
126 Epigram 18

AP5.U4 [Maccius 1 GP]

P PI 7.96, f. 72v [sc. m
i P: PI: Salm. 4 PI: P 6
PI: P

Philestion, hard in all ways, the one who never tolerated a lover without
money,
seems more tolerant now than before. No source of amazement, this seem-
ing: I do not think that she has changed her nature.
For even the shameless asp in time becomes tamer, but it bites no bite other
than a deadly one.

Maccius (or Maikios; cf. GP 2.310) is known only through his elegant epigrams.
This doubly ascribed epigram is printed among those of Maccius by Gow-Page,
but "doubts remain." One notes in slight favor of Phil, that this epigram mentions,
even dwells on, the price charged by a prostitute, with which cf. 17, 20, 21, 22,
whereas none of Maccius' undisputed epigrams does. (Phil, is not the only epigram-
matist to write on this subject, however; cf. 5.29-34,63,81,101,109,113 (= [Phil.
37]), 125, 217, 240). See also on 5 and , But the close similarity
between Maccius 2 and Phil. 16 (q.v.) suggests the possiblity that he and Phil, were
poetic comrades who wrote variations on the same theme. Note also that Maccius
4 (AP 5.130) has the Philodemean name Philainis and the rare adj. as
in 15. See further on v. 3. Some lexical evidence, presented below, favors Phil, as
author, but certainty is not possible.
Note that Pi's refers to AP 5.113, which in fact, if I am correct, was
written by Argentarius; see on [37]. Planudes, however, seems to have thought that
he was gathering together a string of poems by Phil.

1 According to Stadtmuller (followed by Gow-Page) (wit

punctuation now erased before the sigma) was corrected tc . The spac
before the sigma, however, seems to me to be too large for iota + punctuation; more
likely the scribe dipped his pen after the left vertical stroke of H, giving the appear-
ance of a later correction. All editors follow Saumaise in emending to
(four instances in LGPN 1-2), a diminutive of the commor but P's read-
ing could just as easily represent a by-form of the attested or
perhaps via cf.

1-2 The subject of the woman's price comes up elsewhere;

see above, introduction.

3—4 For repetition with negation, cf. Horn. //. 22.495

; D. Fehling Wiederholungsfiguren
(Berlin 1969) 129.

3 This phrase is used elsewhere in the Anthology by Phil. 10.1,

Meleager26 (AP5.160.3, where the girl's name is also Demo; on the main point of
 Epigram 19 127

the poem, cf. H. Jacobson, Mnemosyne, 4th ser. 30 [1977] 71 f.), and Leonidas 95
HE (AP 6.130.3). (A later instance is AP 8.219, St. Gregory.) Meleager also has ti
(2 =AP7A11.5); and Metrodoros AP 14.126.1 and Anonymous ,4?/251.5
have

4 Explanatory asyndeton; cf. 32.4, 2.1-2 (with comm.)

5 Although the comparative is common in medical writings (11X in the

Hippocratic and Galenic corpora), it and the superlative are uncommon in poetry,
but the latter is used by Phil. (34.4); the comparative elsewhere in poetry only at
Dionysios 1 HE (AP7.78.1). This rarity tends to favor Philodemean authorship for
this poem. On the other hand, the author, whoever he was, may have been drawn
to this rare form here because of its use by Hippokrates.
The asp, the cobra of Egypt, could be made to act tame, or at any rate
sluggish, but never lost its power to kill; cf. Nikandros, Ther. 158-167, Aelian, NA
1.54.
Since this adj. appears in two epigrams doubtfully attributed to Phil.,
it is worth pointing out how few poets between Homer (II. 4.521 = Od. 11.598
and Phil. (13) employ pEEK gvi53=ceg 1.132
("ca. 650?," Hansen) Theogn.
207 , and perhaps Bianor 13 GP CAP 9.278.3); cf. Pi. N. 11.45.13 is
virtually guaranteed for Phil, by n, so that it would seem, especially since Phil, is
one of the very few prose authors to use this adj. (3x in Rhet., once reasonably re-
stored in De Mus.}, that its occurrence here weighs heavily in favor of Phil, as author.

6 This verb is equally appropriate to Philestion; cf. Eur. Hipp. 1303

Kallim.Ep. 27 HE (49 Pf. =AP 6.311)
. PL Rep. 474d, "Aspasia" ap. Athen. 5.219c, Asklepiades 8 HE (AP
5.162) with Borthwick CQ 17 (1967) 250-254.
Hapax legomenon, derived from the common
which is noted by Herodian Part. 187.9-188.1, along with a few of the many other
compounds ir . For the syntax, an ace. of result, cf. Soph. Ai. 55
KG 1.305 ff. For the poisonous nature of the asp's bite, cf. Nik. Ther. 185

19

5
128 Epigram 19

4P11.30[27GP,20K]
P iii.7 v.31 caret PI
3 scnpsi: Ap.Voss in marg.) Cr: P: Reiske P:
Jacobs: Kaibel 4 P: Page: Boissonade:
Jacobs: Jacoby Pauw: P: Brunck
Graef

I, earlier capable of five or nine (acts), now, Aphrodite, with difficulty (man-
age only) one from sunset to sunrise.
Oy oy oy, and this (one act lasts) but a short time; and often already half-
dead this little rammer dies.
Old Age, Old Age, what will you manage if ever you arrive, since we now
so waste away?

Sidet,AJP 103 (1982) 211-213.

Thomas, CQ41 (1991) 130-137.
Wright, AJP 42 (1921) 168 f.

The topos of the segnis (but not altogether impotent) amator; cf. 25 and 26 (both
told from the woman's point of view), Rufinus 18 Page (AP 5.47), Skythinos, AP
12.232, Strato, AP 12.11, 216, and especially 240:

Among Latin poets, cf. Juv. 10.204-206, Mart. 3.79, 12.86. Note how Phil., in con-
trast to Strato, prefers coyness to explicitness: no verb in the first sentence and no
noun for or merely the pronounTOTJTOin the next sentence; and a
euphemism in the next clause. For all this, however, the diminuendo of the impotent
lover remains clear: What he used to do up to nine times before he now does only
once, with difficulty and the thing itself, the act from start to finish, lasts but
a short time and already half-dead his member often dies altogether.
For Ovid's borrowings, see below on w. 1-2, 4, 5-6. Cf. Kay ad Mart. 11.46.

1 For article + pronoun, v. ad 8.3. Cf. further Alk. 130bl Voigt =

130.16 LP 6 Antipater 22 HE (4P 7.172.1) 6
Meleager 99 (AP 12.23.1 f.) , Apollinides
23 GP 04P9.287.1. [A second-century epitaph begins (Peek, GVI609),
but Phil.'s text is secure.] The scansion is sanctioned by tradition; cf. Intro.,
pp. 41 f. Note that and serve as adverbs for their clauses, and do not
modify alone; cf. also 32.1

1-2 _ Gow-Page, comparing

Asklepiades 25 (AP 5.181.11 f.) but an internal accusativ
 Epigram 19 129

may be a better explanation in both poems; cf. Eur. Hipp. 32 (Phaidra)

Phil., unlike Asklepiades, omits the verb, but ellipsis is merely a specific
form of the euphemism frequently found in erotic poetry; cf. J. N. Adams, "A type
of sexual euphemism in Latin," Phoenix 35 (1981) 120-128. Diibner supplies coi-
tus agebam to fill out the sense. Cf. Ov. Am. 3.7.23 ff.

at nuper bis flava Chlide, ter Candida Pitho,

ter Libas officio continuata meo est;
exigere a nobis angusta nocte Corinnam
me memini numeros sustinuisse novem.

Cf. A. Richlin, The Garden ofPriapus (NewHaven 1983) 117 f.; also22.1, Ov.Am.
2.10.27 f., Hor. Epod. 12.14-16, Prop. 2.22.23 f., Mart. 11.97. With cf. also
Cat. 32.8 novem continuas fututiones. Two similar but more modest boasts in
Pompeian graffiti: CIL 4.4029 hie ego bis futui, 4816 Chryseros cum Successo hie
terna futuimus.
Cf. Tibullus \.539deseruit Venus, spoken in similar cicumstances.
Although addressed at first to Aphrodite, this poem is more lamentation than prayer;
cf. K. v. Fritz, "Greek prayers," Rev. o/Rel. 10 (1945-1946) 5-39.

2-3 The two terms are not synonymous, as Gow-Page

think. Sexual satisfaction was difficult to attain and was over with quickly. See in-
troduction a ve, and below on

2 The phrase in the sense "nightfall" is rare: Aratos

41,747, Pollux 1.70. Cf. Soph. OC 477 (Jebb ad loc. says that the
meaning here is only local, "eastwards," but if so the adj. is unnecessary; cf. LSJ
s.v. 4; Pollux 1.68 is clearly temporal). Cf. Hor. O. 3. 7,2Qprima
nocte domum elude.
After the epi formula _ (12x in Homer, once
each in Hesiod and H.Herm.), it was likely if not inevitable that a Hellenistic poet
like Apollonios would produce (1.725), but Phil, is the only one
to have shortened the latter phrase. The sense is clear enough, however; cf. Soph.
El. 424 f. (Klytaimestra) where the sense seems to be "upon
arising at dawn."

3 : The line's defect is most easily made up thus, the exaggerated lament
being in accord with the narrator's entirely personal view of the situation; for the
accent, cf. Apoll. Dysk. Adv. 177.21. Cf. Strata, AP 12.240.4 (cited above, intro.),
Aristoph. P a x 2 4 7 ( b e t t e r : A i s c h . / 4 g . 125
(Kassandra:)
The most likely sense here is "briefly, for a short time," which is
how Phil, seems to use the phrase at Rhet. 1.273.9-11 (fr. 12)
where Phil, contrasts dialectic with rheto-
130 Epigram 19

fr. 11). Thus, Phil, uses KOTOC where others use 01 (cf.
LSJ s.v. . The usual sense of this phrase, as pointed out by Thomas, "gradu-
ally, little by little, slowly," is inappropriate. Hence I continue, with others, to under-
stand the phrase to refer backwards to (contra Thomas, who, placing
in parentheses, has it refer to

4 The various conjectures weaken an intentionally strong statement;

cf. Automedon 2 GP (AP11.29.3 f.)
id. 1 (AP 5.129.8) (i.e., brings
an old man's penis to erection); Ov. Am, 3.7.65 f. nostra tamen iacuere uelut
praemortua membra \ turpiter hesterna languidiora rosa. Thomas argues for the
word's referring to the flaccidity after ejaculation (comparing Cat. 50.14 f. at
defessa labore membra postquam I semimortua lectulo iacebani), but this is far less
embarrassing than the inability to achieve full erection before intercourse. For
thegeneral tone of the lament cf. Aristoph. Nu. 504

In addition to the passages quoted just above, cf. Skythinos' segnis

amatorpoem (AP 12.232.4): , Strato,AP 12.216.2, Mart. 13.34.1
mortua membra, Ov.Am. 3.7.16 etnon exactum corpus an umbra forem. For0vf|OKet
= "dies, is dead," cf. Kallinos 1.19 W, Hdt. 4.190, KG 1. 137).
Later lexica derive the phrase tepjiepta KaKa from
Termerium, where prisoners were kept, defining it generically as ueydXa Kara
(Photios, Souda). Kaibel and others, however, have recognized that Phil, is here
referring to the monstrous Termeros, who used to kill people by butting them with
his head until he had his head broken by Theseus: Kod tov Tepuepov

[sc. Sinis, Skiron, Prokroustes, etc.]

(Plut. Thes. 11). As (loosely
temporal and causal) makes clear, Termerion kakon applies not to the harm suf-
fered by Termeros' victims, but to the nature of his punishment. And the following
sentence suggests that it is not merely the punishment itself so much as its appro-
priateness that is conveyed by the phrase, which is thus equivalent to
(cf. Pausanias 4.17.4), the punishment that fits the crime. This is certainly the
case with Phil., where that which has done the butting has had its "die."
For punning on , cf. Aristoph. frr. 244, 566 (Hermes
1,568, (?)569 (Aristoph. may have punned on the1
K-A; see further Henderson Maculate Muse 112 f.; R. Seaford, LCM 12.9
(1987) 142f. Cf. also Mart. \\A6Aneclevatextinctumsollidtata [sc.mentula] caput.
Opposed to this, Thomas argues that the antecedent of is Termeros and that
the Termerian evil is the evil meted out by this villain, although it is unclear to me
why what must have been the infrequent occurrence of a cracked skull should have
been granted a special name.
Nevertheless, although I disagree with Thomas's interpretation of Plutarch and
his punctuation of w. 3 f. (see on his analysis of this phrase is prefer-
able to mine of 1982. There is, after all, no in this poem, so that
 Epigram 19 131

Tepiispvov is better understood as "this Termerian thing," i.e., "this rammer," who
like Termeros has his "head" softened.

5-6: Cf. Ov.Am. 3.7.17 f.

quae mihi ventura est, siquidem ventura, senectus,

cum desk numeris ipsa iuventa suis?

Old age, horrible thought that it is, could at least provide a somewhat honorable
excuse for failure. Other possibilities are drugs and witchcraft (Ov. Am. 3.7.13,27-
36,79 f.); and then failure to perform itself is a reason: pudoripse nocebat (Ov.Am.
3.7.37).Cf. Mimnermos 1.5 ff. W2

5 The repetition magnifies the complaint; cf. Aisch. Ag. 1538

etc., Herod. 10.2 Hor. 0.2.14.1
Leheufugaces, Postume, Postume, \ labunturanni-,I).Feln\ing,Wiederholungsfiguren
(Berlin 1969) 169, 174 f.
is the child of Night (Hes. Th. 225) and had an altar in
Gadeira: Aelian fr. 19. He is depicted on vases as an old man with a notably flaccid
(though often exaggerated in size) penis; cf. H. A. Shapiro, "Geras," LZMC4.1.180-
182; id. Personifications in Greek Art (Kilchberg 1993) 89-94. Note also Herod.
2.71 f.
Because is personified, Phil, can reverse what must have been
the more normal expression; cf. Aristoph. Av. 606
But is a "dactylic cliche"; M. S. Silk, Interaction in Poetic Im-
agery (Cambridge 1974) 93 n. 15.

6 Prob. notplurale odestatis, as Gow-Page's "I" suggests; rather,

as suggested by some parallel passages, both the "I" so prominently positioned at
the beginning of the poem and this man's penis, which has just been compared to
an old man near death: cf. 2, SkythinosAP 12.232.1 (an address to a
penis), Automedon 1.6, where I think context makes it sure that
refers primarily to the penis; Mart. 11.46.3 pannucea mentula, with Kay's n.; Ovid's
languidiora (v. ad 4 , Cf. 30, where once again there is some point to the
shift to first person plural in the last word.
For the wasting action of time, cf. Soph. Ai. 713
(lapaivei, Sophron 54 Kaibel Isok. 1.6, Aristotle
fr. 881 Gigon = Galen, De Mixt. 2.581 f. K (adduced by Kay). For the reduction
in sex drive in old age, cf. PL Rep. 329a-c, where Kephalas notes that although
usually men he himself agrees with
Sophokles that he no longer is a slave to the passions of Aphrodite; and Ov. Am,
3.7.41 f.

illius ad tactum Pylius iuvenescere possit

Tithonosque annis fortior esse suis.
132 Epigram 20

20

AP 5.46 [4 GP, IK]

P 01 caret PI vii.
15
1 P: Ap.L P: Brunck (v. comm.) 2 P:
Kaibel Diibner: P: Herwerden C: P 4 ppc.
pac P: Ap.L: Reiske 6
C: P

M(AN). Hello there. W(OMAN). And hello to you. M. What should I call you?
W. And what should I call you? M. Don't be eager to learn this so soon.
w. Nor you. M. Do you have someone?
w. Always: the one who loves me. M. Are you willing to dine together with
me today? w. If you are willing. M. Great! How much for your presence?
W. Give me nothing in advance,— M. This is strange.— w. but after lying
with me give what you think right. M. You're very fair.
Where will you be? I'll send for you. w. Take note— M. But when will you
come? W. Whatever hour you wish. M. I'm willing right now. w. Lead o.

Herwerden, Mnemosyne, 2nd ser. (1874) 13.

Hopkinson 79, 270.
Falivene, QUCC 42 (1981) 94f.

A conversation on the street between a p ntial customer and a saucy prostitute

who displays the professional's hauteur and uses it to manipulate her potential
customer. Conversation poems are mimes in miniature, with the easygoing language
of ordinary life combined with the polished literary form of Hellenistic epigram.
Their origin probably lies in actual epitaphs in this form; cf. W. Rasche, De
Anthologiae Graecae epigrammatis quae colloquiiformam habent (Munich 1910) 6-
16; but literary examples become popular by the third century B.C., with Leonidas,
Kallimachos, and Anyte trying their hand. Phil., however, may have been the first
to write an erotic encounter in this form (Rasche).
 Epigram 20 133

For the sake of clarity, I have, following the lead of some papyrus texts of
mimes—e.g., P.Berol. 13876 = no. 12 Cunningham—provided indication of speaker.

1 An unexceptional response; cf. Aristoph. Pax 718 f.

and [Plato] Eryx. 392b
Phil, allows himself to violate both Hermann's Bridge
and "Plutarch's Law" (above, p. 43), probably because of the colloquial nature of
the conversation.
With this the conversation deviates from what we may imagine to be
the ordinary course; cf., e.g., Prop. 2.23.13-24, esp. 16 neesinit esse moram, siquis
adire velit. The woman in Phil.'s poem, however, instead of complaisantly reveal-
ing her name, lets the man know that she will not give hers unless he reveals his;
note esp. 2 . Presumably the usual practice was for the man to learn the
woman's name while withholding his own. Herwerden would read A.

2 A near hapax (again only in Dittenberger, OGI 339.39 [Sestos ii

B.C.] , which Kaibel altered to the
unattested verb approb. Gow-Page: "the isolated nominative . . .
seems unnatural and contrary to the easy conversational flow," but mime (see in-
troduction above) often includes the natural interruptions and inelegant formula-
tions of colloquial speech. With the ellipsis of here, cf. Herodas 5.14, here
is ellipsed (Headlam; contra Cunningham).
provides the only ex. in my edition of the long syllable at the
caesura of the pentameter produced by lengthening (see Intro., p. 44), but einmal
is not always keinmal.
The man hopes for, rather than expects, a negative answer; cf.
Barrett ad Eur. Hipp. 794. Although from the start the ultimate purpose of this
conversation would be clear to both participants, the first sign of this is this coded
oblique question: "Do you have anyone?" = "Are you free at the moment?" On the
use of in this context, see on 21.3.

3 This may not be equivalent to (Gow-Page,

Hopkinson), which is merely "anyone who fancies me" (Gow-Page), "whoever is
my friend/lover at the moment." Her answer rather seems closer to "I always have
someone: the one who loves me," "toujours celui qui m'aime" (Waltz), "eum semper
habeo, qui me habere cupiat" (Jacobs). Hence my punctuation, with explanatory
asyndeton; on which cf. KG 2.344 f., West on Th. 533, and Ve.tdenius,Mnemosyne,
4th ser., 27 (1974) 17 f. Cf. Theogn. 1367 f.

4 Obviously a code word for what is really on the man's mind; cf. the
current use of "massage" and "escort" in advertising. seems to be P's own
alteration rather than a correction of C, as reported by Stadtmuller, who is followed
by Gow-Page.
I find no exact parallel for this phrase, but with verbs
used absolutely is common enough, e.g., (PI. Ap. 20b). The sense
134 Epigram 20

of "be my guest" (cf. Anstoph.Av. 131), main tains the pretense that she is
being asked only to dine with the man.

5 This interruption may be imagined as an aside, of the sort "where

a speech is interrupted and after the interruption the speaker continues without
taking up the interrupter's point"; D. Bain, Actors and Audience: A Study of Asides
and Related Conventions in Greek Drama (Oxford 1977) 87; cf. esp. ch. 7, which
covers these "asides in conversation."

5-6 Cf. Machon 346 Gow • Kaibel>

spoken by a prostitute to a customer reluctant to pay her price. Gulick
(Loeb ed.) translates "you may give anything you like," which is close to Phil.; Gow
however, rejecting Kaibel's supplement, is probably right to interpret the indica-
tive as ". . . the [specific] fee you wish to pay," i.e., the five minai offered by th
man. Cf. 21.2 6

6 A common euphemism; cf. Henderson, Maculate Muse 160f.

If taken literally, these words assigned by the MSS to the man,
would make better sense spoken by the woman, as an explanation of her novel
pricing policy: "You're an honest man" (and hence will not cheat me). And since it
is not dishonest to set a price in advance, she cannot be called honest for refusing
to do so. (Thus for a while I was prepared to assign these words to the woman.)
These words, however, have to be understood in the wider context of Greek
attitudes towards justice,philia, and reciprocity. In brief, anyone helping you is, at
least for the moment, your friend and is acting justly. then, is a collo-
quial equivalent tc cf. K. J. Dover, Greek Popular Morality (Oxford 1974)
18l£, who cites Isok. 18.63 forthe "clear antonymy between a dikein and eu poiein"';
cf. further Falivene.

7 A present used with future sense: "where will you (come to) be"; so
Kaibel, strongly supported by Wackernagel, Syntax 1.161, Miiller Antiphilos 40,
and Gow-Page. Cf. P. Prag. Varcl. N.S. 41.16 f.
For Gow-Page, this is equivalent to asking where she lives, which Kaibel,
Wackernagel, and Miiller deny. Professional "call-girls" may be found at
a favorite location away from home. The same phrase appears in the companion
piece, 21.3 (q.v.), where the woman's address is clearly what is desired. 21.3
spoken entirely by the man, also shows that Kaibel was wrong to
assign to the woman here.
"The word means learn,' not 'ask,' and it is not easy to under-
stand in this context"—Gow-Page ad loc., who, after running through the strained
attempts of earlier scholars, tentatively suggest that the verb "means in effect 'y°ii
can find out,' i.e., you can easily learn my address," which strikes me as notably
less responsive than her immediately previous answers. As my punctuation shows,
however, I think that what we have here is a man so eager that he interrupts the
answer to his first question with another question. This has the further advantage
of mediating dramatically between and
 Epigram 21 135

By chance, the imperative of this verb recurs in a recently published fragmen-

tary mime (P.Oxy. 53 [1986] 3700, ed. M. Haslam;iA.D.). There is enough to sug-
gest that the imperative may have the same colloquial force as is found in Phil. The
scene is at a door:

Haslam suggests either for v. 5. For his supplement

ofv.6,cf.P.Herm.6.17 ,(ETEP = ( i ) o r ( i i )

Phil.'s poem is far livelier if we understand an interruption of the woman's di-

rections. The man starts out pretending, or at any rate trying, to be businesslike
and detached, but manipulated and aroused by the woman's professional guile, he
can no longer wait for a later appointment.

7-8 These words in particular recall what

Kerkidas said of
(5.27-29 Powell). The incorporation of the antecedent into a relative
clause (KG 2.416 ff.) allows the main clause to be dispensed with entirely. (It would
have been simply [cf. 28.6 , with understood.)

8 WhatLSJs.v. B13 say of adv. "rarely used of time," is

absolutely belied by the Anthology, where all 14 exx. are temporal; cf. in particular
AP 12.200.3 (Strato) which has the same erotic sense as here. Note
also 28.6.
Immediately after , the scribe of P wrote which C erased in order
to write out in its proper place the second half of the pentameter.

21

5
136 Epigram 21

AP5308 [Antiphilos 14 GP, 3 Miiller]

P [sc. A [C] PI 7.101, f. 73r [sc.
vi.4
3 PP1: Gow-Page: Scalieer 4 CP1: r 5 CP1
P
Wait for me, my fine lady. What's your pretty name? Where are you to be
found? I'll give you what you want. Won't you even speak?
Where do you live? I'll send someone with you. No one calls you his, I
hope?—Goodbye, Miss Hoity-Toity. Won't you even say Goodbye?
I'll come up to you again and again. I know how to soften women even
tougher than you. But for now, goodbye, woman.

K. M tiller, Die Epigramme des Antiphilos von Byzanz (Berlin 1935) 39-41.

With ascription to Phil, by PI, C (however oddly), and now (almost certainly) II,
there is no reason to deny him this poem. In writing the scribe of P
jumped over the immediately preceding poem by Antiphilos to the second one back,
which is by Phil. (25).
Also arguing in favor of Phil, is the way this poem complements 20: In each a
man confronts a woman in/of the streets who controls the situation, one by using
speech to sharpen his desire, the other by maintaining silence to reduce the man to
near-spluttering impotence. (I doubt, though, whether, as Miiller 40 argues, C would
have taken a poem with a perfectly good attribution and assigned it to Phil, on the
basis of this similarity alone.) The poem is credited to Phil, by Brunck, Jacobs, and
K. P. Schulze, EPhW36 (1916) 319. Another variation on this scene is offered by
AP 5.101 (Anon. 6 FGE). The woman (who may be the maid acting for her mis-
tress; cf. Page) is responsive, as in 20, but the man is disappointed, as here:

1 The "elegance" of this passing (see next lemma) woman, which the man
obviously finds attractive, will, after she ignores him, be interpreted otherwise; see
below on 4 , See on v.15.
We should imagine that the man sees the woman approach in the
street, addresses her, and (see on v. 4 after being ignored, follows after
her as she continues on her way.
Part of his attempt to win her affection: A woman as elegant as she is
must have a fine name.

2 Homoioteleuton of pentameter halves occurs more than twice

as much in Phil, than in Antiphilos; see Introd., p. 44, Miiller 29 f.
 Epigram 21 137

3 See on 20.7.
Twice he has asked where she is to be found and twice she has ignored
him. He now threatens to send a slave along with her to learn her address—but not
so that he can "come to her again and again" (5), as Jacobs, Miiller, and Waltz think;
see below. Scaliger and Gow-Page are wrong to desire an accusative here.
With clear sexual overtones; cf. Skolion 904 PMG, where this
sense is played off against a more neutral meaning:
Cf. further
Poseidippos 2 HE (AP 5.186.3-4)
Asklepiades 41 HE (AP 7.217.1; cf. Fraser, Ptolemaic Alexandria
[Oxford 1972] 2.805 n.97), Kallim. 2 HE (AP 12.43.6), Asklepiades4 G4P5.158.4),
LSJ s.v. A14.

4 The first half of the poem is addressed to the woman as she approaches;
we may imagine that the second half is addressed to her back. The man now drops
the pretense of politeness. The early sense of this word, "impetuous," gave way
to "haughty"; cf. Page, Rufinus 44 ff. Used to describe either a strutting gait
or a haughty carriage, such as raised eyebrows (see below and cf. the adj.
or a neck held high (Rufinus 10 Page =y4P5.28.4), the word here
shows that the woman does more than merely maintain silence; rather, she indi-
cates through body language what she thinks of this pest. Cf. the similar situation
in Rufinus 33 Page W?5.92.1 f.):
This is obviously not a quality a would-
be seducer would likely find attractive; cf. Agathias 77 Viansino (AP 5.280.8)
(Aphrodite)
Epicurus uses the word in a complimentary sense, perhaps "high-spirited": VS
45
C erased what what was probably a final epsilon (though to my eyes
it could just as easily have been an alpha). Elision between pentameter halves is
rare: only 7-9 exx. in GP. Since neither Phil, nor Antiphilos allows this elsewhere,
the question of authorship is unaffected. (Antiphilos elides in this posi-
tion; Phil, elides only at 31.2 [36.4 may not be by Phil.]—but these short words
do not figure in the statistics; cf. GP l.xliii: H. Lloyd-Jones and N. Wilson,Sophoclea
(Oxford 1990) on OT523.) It is doubtless the colloquial tone of this poem which
allows this deviation from the norm. For this verb in leave-taking, cf. P.Oxy.
219(a).24 (a papyrus mime = 4 Cunningham) Aristoph.
Ran. 165,Ekkl. 477; further exx. given by Headlam-Knox on Herodas 6.97. (Jacobs,
following Lucian Pro lapsu 1, thought that was limited to evening,
to mornings, but this may not be valid in Phil.'s lifetime; cf. Ran. 165, where the
latter is answered with the former.)

5 By itself this word c ld indicate "approach", i.e. "accost"

(Gow-Page), in the street, but it is possible that, since the man seems not to have
seen this woman before and hence could not know that he would see her again, he
plans to adopt (see above on 3 the more aggressive tactic of finding out where
138 Epigram 21

she lives so that he can wait for her at her door. (In which case, this poem could be
called a proparaklausithyron.) I think, though, that the man is covering up his de-
feat with a display of braggadocio: If she will not stop now, he will wear her down
in future encounters and so win her over eventually.
A good word to repeat; cf. Phil. 23.1-2, Meleager 42 (AP
5.136.1)

5-6 The contrast is obvious and of long standing;

cf. PI. Rep. 411a-b
Tht. 186b, Symp, 195d-e, Polyb. 4.21.3

6 As indicates, this final imperative means only "goodbye," and

not also "porte-toi bien," i.e., "reflechis bien a ce que je te dis" (Waltz).

22

AP5.126[25G~P, p. xxv K]
P ii.18 caret PI [C]

1 P: Ap.B, Philaras: Ap.B(marg.): Leid.B.P.G. 34B 2 P:

Reiske Philaras: Ap.B: C: P: Leiden
B.P.G. 34B Ap.B, Ap.L, Leid. B.P.G. 34B, Philaras: P

Mr. X gives Mrs. Y five talents for one favor, and he screws, shivering with
fear, one who is, what's more, God knows, no beauty.
I give five—drachmas—to Lysianassa for the twelve favors, and what's more
I screw a finer woman, and openly.
Assuredly, either I'm crazy or, after all this, he should have his balls cut
off with a knife.

Wright, ,1/P 42 (1921) 168 f.

 Epigram 22 139

The vulgarity of the language led Planudes to omit the poem and Kaibel to deny
Philodemean authorship (p. xxv), but authorship is all but guaranteed by its inclu-
sion in n. The various metrical licenses detailed belowpassim are no doubt intended
to maintain the tone.
Wright argues that Horace refers to this poem at Sat. 1.2.119-122 (T 4):

parabilem amo venerem facilemque.

illam "postpaulo," "sedpluris," "si exierit vir,"
Gallis, hanc Philodemus ait sibi, quae neque magno
stet pretio neque cunctetur cum est iussa venire.

But Horace praises the woman who charges little, he does not attack the man who
pays much. Nonetheless, the combination of high/low fees and castration/Galli
in close proximity while discussing two types of women is curious; cf. C. Dessen,
"The sexual and financial mean in Horace's Serm., 12," AJP 89 (1968) 200-208.
Perhaps, given Phil.'s propensity for composing poems in contrasting pairs, we
can infer that Horace refers to a poem which Phil, composed to go along with
this one. See 38, introduction.
Like Horace, Phil, seems to allude to the difference between a married woman
and one with whom intercourse offers no threat of punishment. Phil.'s model may
have been, as Jacobs suggests, Xenarchos, Pentathlos fr. 4 K-A, in which trouble-
free dealings with prostitutes are compared favorably with the dangers posed by
furtive adulteries; see further below, on 2

As Gow-Page note, "the division of words — |

— in the first half of the line is a very rare phenomenon in this genre." Lack of A-
caesura: 4.1, 6.7 (proper name), 8.1,7, 12.5, 35.1 (if by Phil.). For Meyer's First
Law (the rarity of words ending x —- | in the second foot), cf. West, Greek Metre
37 f., 155. is of the same metrical pattern (West 197), but the
Alexandrians as usual were more Homeric than Homer. See above, p. 43. It is
also unusual to have the syllable before the masculine caesura long by position;
elsewhere in Phil. only3.3,15.3 and 19.5; see above, p. 43. As stated in the intro-
duction, I suspect that the metrical irregularities of this poem are intended
to mirror the crudeness of the person and situation described. Cf. Hesiod's
intentionally irregular line describing the Chimaira; J. Solomon, "In defense of
Hesiod's 'schlechtestem Hexameter,'" Hermes 113 (1985) 21-30. Perhaps also
Catullus 116.8; H. Dettmer, "The first and last of Catullus," Syll. Cl. 5 (1993)
32 f.; and for seven possible exx. in Vergil,E. 7, cf. F. M. Sandbach,CR47 (1933)
216-219.
Genitive of cause giving the thing purchased; cf. Xen. Mem. 1.6.11
Smyth 1373a. For the scansion, see above.
This passage apart, only two forms of the dat. fern. sg. of this odd
noun are known: (i) 8elvi (Epiktetos 1.12.28, Philostratos, VA 6.43, 8.5, Job.
Chrysostom, In Gen. 53.300, In Matth. 58.678) and the undeclined (ii)
(Origen, Contra Cels. 5.45 = Philokalia 17.3). Did Phil, either invent or know of a
140 Epigram 22

third form, (iii) (P), or is scribal error at work (e.g. for )?Of the
three possible forms, (i) is the most presentable, being vouched for by the Atticist
Philostratos and least objectionable metrically, since the hiatus of can be
paralleled by similar instances after and vocatives ending in -i in Herodas
(West, Gk. Metre 161), whose tone is comparable to that of this poem; as well as
after datives of the third declension generally (Jacobs). The hiatus produced by (ii)
is much harsher. The correption of (iii), however, falls within a dactyl, a practice
Phil, elsewhere avoids. The scansion of the fourth foot , violates Her-
mann's Bridge in all three readings.
In this poem, however, we cannot simply go with the best attested form and
least objectionable meter, since Phil, has given his narrator a roughness of tone and
meter which may here be reinforced by a morphological irregularity. Hence, tempted
as I am by (i), it seems best to print the reading of P. Cf. Solomon 25 f.
This strange noun means either (i) "someone or other," where al-
most any name will do, or, less commonly, (ii) "you know who," where there is a
reason for not specifying the name; cf. Dover on Aristoph. Frogs 918. Here, for both
fern, and masc. forms, the latter applies, the reason for anonymity, at least ostensi-
bly, being to protect the guilty adulterers, from the embarrassment of being known
for overpaying and overcharging.

The social vulgarity of the man is matched by the vulgar tone of this word,
which is like E,ng.fuck; cf. Henderson, Maculate Muse 151 f. with addenda, p. 249;
D. Bain, "Six Greek verbs of sexual congress
CQ 41 (1991) 51-77, esp. 54 ff. For the tone, cf. H ii.16, where
is a likely restoration.
In fear of being caught, in contrast with the narrator's doing it
Cf. 15. Gow-Page, following Jacobs, aptly cite Xenarchos fr. 4.16 ff.
K-A, which talks of young men able to prostitutes without fear, in opposi-
tion to married women:

Presumably Phil's- is a married woman. Cf. Antipater53 GP (AP5.109.1-

2)

Sc. (LSJs.v, (A) IV); cf. schol. ad Aristoph.

Ran. 1374
Reiske may be right to suggest vai ud, which
frequently occurs elsewhere, with and without the god's being named (LSJ ibid.),
but the repeated is rhetorically superior.

3 The twelve times a night the narrator is known for being capable
of (or for boasting of); cf. on 19.1
 Epigram 22 141

Names with the ending -ctvaooa are rare, the most famous being
Agamemnon's daughter Iphianassa; some others are Archeanassa (a hetaira in
Asklepiades41 HE = API.211), Kallianassa, Kleanassa, andKleitanassa. Lysianass
appears in myth as a daughter of Nereus (Hes. Th. 258), Polybos (Pausanias 2.6.3),
and Epaphos (Apollod. 2.5.11).
Two conjectures, usually credited to Reiske, can be found at least as early as
Leonardos Philaras's sylloge from P (s. xvii, Paris Coislin 352, f. llv). The conjec-
tures may not be original to him, however, as J. Hutton reports that "he had before
him some of Saumaise's corrections"; The Greek Anthology in France (Ithaca 1946)
189.

(as in Hdt.). Over the omega, C wrote •/•, indicating that

he thought the phrase corrupt and could find no MS variant to replace it, but the
phrase, probably colloquial, should be allowed to stand. No doubt, it was meant to
sound odd.

Transposition on Phil.'s part could have obviated correption, with

little change in meaning; but, as was said above, he probably intended some metri-
cal roughness, here two exx. of correption on one line.
From Homer onwards, one could be said to be apart from
or without
Phil, uses the phrase of intelligent people who may not
be philosophers: On Poems 5.22.35 Mangoni, Rhet. 1.201.12-202.14, 240.10 f.
Sudh. For the scansion, cf. 36.3 which may not be by Phil.
Not "in future" (Gow-Page), an altogether too vapid sentiment
for our narrator; rather, "after all this," i.e., "all that remains to do now" (LSJ s.v.

6: For castration as punishment for adultery in Rome, cf. Hor. Sat. 1.2.44 ff. quin
etiam illud \ accidit, ut quidam testis caudamqe salacem \ demeteret ferro. "iure"
omnes, Mart. 2.60.3 dum ludis, castrabere, 3.85, 3.92, 6.2. See further on 15. For
adultery in general, cf. C. Edwards, The Politics of Immorality in Ancient Rome
(Cambridge 1993) 34-62. Phil.'s poem, however, maybe a literary exercise, draw-
ing, as was said, on Xenarchos, who specifically alludes to the harsh penalties meted
out to adulterers in Athens under the laws of Drakon. Cf. A. R. W. Harrison, The
Law of Athens 1 (Oxford 1968) 32-38. And for a comparison with Horace, cf. M.
Gigante, Orazio: Una misuraper I'amore: Lettura della satira seconda delprimo libra
(Venosa 1993) 82 f.
"Twins" = "testicles" as in Herophilos, who also used the term for
ovaries; cf. Gn\en,DeSem. 4.596 f. K (Herophilos T 61 von Staden), Us. Part. 14.11
(2.323 Helmreich = Herophilos T 109), with von Staden's note to the former
(p. 231). The word appears in this sense in Marcus Argentarius 7 GP CAP5.105.4)
Clem. Alex. Protr. 2.15.2, and in LXX Deut. 25.11; and probably also by Phil, him-
self in 31.4, q.v.
142 Epigram 23

23

AP5.107[5GP,7K,4G]
P PI 7.184, f. 75v vii.13 [] LI]

1 P PI Scaliger PP1 2 P PI 3
P PI 4 P PI 5 PI P 7 PI P 8
PI P C N C N PI P

I know, dear, how to return the love to the one who loves me, and I know
how to bite the biter back.
Do not cause me who loves you too much pain, and do not stir up against
yourself the wrath of the Muses of Pieria, fierce in their anger.
—These words I would shout and and give warning, but you heeded them
the way you would the Ionian Sea.
This is why you are now are howling so loudly, while! rest in the bosom of
Naias.

Falivene, QUCC 37 (1981) 87-95.

Giangrande, MPL 5 (1981) 42.
Rossi, Maia 33 (1981) 213 f. [Rossi M]
Rossi, Vichiana 10 (1981) 163-167. [Rossi V]
Sider, AJP 108 (1987) 317 n. 22.

The first half of the poem, as the audience learns only on v. 5, is in fact a complete
poem as it was already recited more than once to the woman, although to no effect.
It was a warning from a rejected lover who is also a poet: be good to me or I will
invoke the Muses against you, i.e., write poems against you; cf. Rossi V 166. The
8-line poem before us is at least partial fulfillment of that threat, as well as being an
especially interesting example of metapoiesis. The woman, however, has shown that
she is invulnerable to words, which is particularly frustrating to a poet. To recipro-
cate, Phil, goes to Naias, which has the desired effect of reducing the woman to the
same state of inarticulateness as she had reduced him. Only thus does Phil, accom-
 Epigram 23 143

plish with the 8-line poem what he failed to do with the 4-line poem. Before she
treated Phil, like the roar of the sea; now she is the inarticulate one, barking inar-
ticulately at one who rests in the of'Naias".
The trick of turning the first half of a poem into a previously recited poem was
imitated by Ovid Am. 2.5, where again there is no warning that what has been heard
or read up to this point is a repetition of what the poet's mistress had heard before:
haec tibi sunt mecum, mihi sunt communia tecum (v. 31); cf. Horace Epode 2, where
it is not until v. 67 that we learn the preceding were the words of Alfius. Ovid, I am
sure, has Phil, as his immediate model, but all may ultimately be drawing upon
Archilochos, who began at least two poems with the undeclared words of another:
19 W2 which turns out to be spoken by Charon the tekton]
and 122 a father to his daughter); but cf. K. J.
Dover, "The poetry of Archilochos," Archiloque = EH 10 (1964) 206 £, 215 (repr.
in id. Greek and the Greeks [Oxford 1987] 111 f., 116f.). Cf. also Sappho 1, where
in v. 13 we learn that the preceding prayer to Aphrodite was in fact answered by
her.
As Falivene 93 points out, Phil.'s chief model for the repetition of a threat made
to a woman almost certainly is Archilochos 23 W2 (P.Oxy. 2310 fr. 1 col.i.7ff. ed.
Lobel):

1C

15

(Cf. M. L. West,Studiesm Greek Elegy and Iambus [Berlin 1974] 118-20.) Lobel's
restoration of 15, suggested by v. 16, is also consistent with Phil.'s
threat.

1-2: For the prevalence of the thought, cf. Archil. 23.14 f. (cited above), 126 (see
next lemma); Hes. Op. 353
Theogn. 337ff., Pi. P. 2.83-85, Aisch. PV 1041f., Sappho 5.6f., Solon 13.5, Soph.
Ant. 643f. (see below on 1 ); Eur. Medea 809f.,Her 585f. Note also Plato,Kriton
49b
for the broader social and political contexts in which this phrase figures (and for
far more parallel passages), cf. M. W. Blundell, Helping Friends and Harming En-
emies (Cambridge 1989), ch. 2. But for us the most interesting parallel is 24.
Archilochos and Phil, are alone in applying the complete form of this political
maxim to the erotic sphere, which has its own code of reciprocity and divine ven-
geance; cf. Falivene 88f. and Rossi V 165f. The reciprocity-of-love half of the for-
144 Epigram 23

mula is of course implicit in the familiar notion of cf.,e.g.,CEG

2.530 (a wife's tombstone, Attic, ca. 365-340)
cf. J. Pircher, Das Lob der Frau im vorchristlichen Grabepigramm der Griechen
(Innsbruck 1979) 39 f.

= Theokr. 11.30: Polyphemos, in circumstances similar to

that of Phil., often (cf. lOff.) sings to the unresponsive Galateia (and also to the sea
which is her home). inf. in the sense "know how to" is rare (and barely
acknowledged by LSJ);cf. Men. Sent. 20 ' - s
myt jj2129.
is "ironical as in, e.g., Aristoph. Ekkl. 794, PI. Rep. 452b" (Gow-Page).
Cf. Clark Gable's "Frankly, my dear, I don't give a damn." It would seem that this
earlier poem (w. 1-4) was far from the first to be addressed to this woman.
Scaliger's emendation is not (pace Gow-Page) "strongly supported by
(to be taken with ) in v. 2." Word order and sense allow to be
taken with Nor does the fact that " is rare in poetry" count for much
in a Hellenistic poet. (It occurs, e.g., thrice in Aisch., twice in Soph., once each in
Eur. and Xenophanes.) is retained by Stadtmiiller, Dubner, Waltz, Falivene
92 n. 13. On the other hand, is an odd expression (in fact, + verb
is rare in prose and poetry) and may easily be seen as Phil.'s variation on the
more usual found in some of the parallel expressions; cf. Archil. 126 W2
Aisch. Cho. 123
Soph. Ant. 643 sim.Pl.
Kriton 49b (see above); Blundell 29 n.17.

2 Metaphorical as in II. 5.493

And Phil, too probably has biting words in mind. Cf. 18.6.

3 With alone, not, as Gow-Page would allow, with as well.

Phil.'s saying that he loves her too much would upset the idea of balance estab-
lished in w. 1-2.
"Seldom of sexual love" (LSJ s.v. I 3, where the exx. from
Xenophon barely qualify for inclusion), but this no longer applies by the Hellenis-
tic age: in addition to Phil, himself at 17.5, cf. Moschos fr. 2.8, Meleager 77, 82, 93
(AP 12.95.1, 54.3, 158.5) and Argentarius 10, 34 GP (AP5.116.3, 11.320.1); also
Sosikrates 4 Kock

3-4 "Do not"; LSJ s.v. 15.

3 Pi's reading may derive from a scribe's transcribing one remembered

line at a time and unconsciously assimilating the infinitive to the preceding imperative

4 This word appears elsewhere only in Souda s.w.

For the sense, cf. Pollux' list of adjj. expressing the anger of a god:
(1.39).
An accented form is more likely for the beginning of the second half of
the pentameter (Gow-Page). Construe as dative of disadvantage with (so
Waltz, Diibner), rather than "do not of your own will" (Gow-Page).
 Epigram 23 145

5-8: Phil, turns away from the woman who has rejected him. In addition to Archil.
23 (above, intro.), cf. Cat. 8.12-19 (vale, puella, etc.); Tait 45 f.

5 : Pi's reading sits better with the following'! era, which in either case may be
taken as adverbial (see the quotations from Homer in the next lemma). Note also
Ovid's haec (above, introduction).

5-6 The sea between Italy and Epirus; cf. Gow-

Page on Diodoros 5 GP (AP 7.624.1). This compendious comparison can be con-
strued in two ways: (i) Talking to you is like talking to the sea, i.e., you were deaf
to my poem. For the thought, cf. //. 16.34, Eur. Med. 28f.
Hipp. 304 f., Aisch. P.V. 1001, Horace, O. 3.7.21f.
For the syntax, cf., e.g., Od. 11.304 Or (ii) You
heeded my words as much as you would the sea. Cf.//. 13.176= 15.551
Translators uniformly adopt (i), but as suggested above in
the introduction, the punishment Phil, works on the woman is more condign if
he reduces her to the same state (inarticulateness, not deafness) she had induced
in him.

7 Waltz, who r e a d s i s forced to translate unnaturally "a ton tour";

see below.
Perhaps to be taken with each of the two following words;
cf. Homer, II. 16.428 .
"Bark," like a dog; cf. Herakl. B 97, Fraenkel ad Aisch. Ag. 449. Phil.
knows how to repay in kind. The woman reduced him to a state of ineffectual
muttering by failing to react to his poems; he now gives her a taste of her own
medicine by turning a deaf ear to her growls now that he has gone off to Naias. Cf.
below on Naias' name, and Lykophron 1452 f.
The optative of wish (P) is out of place after which, since it
"bears a strong logical force" (Denniston GP565), is followed by either an indica-
tive (often a future) or imperative. An indicative here is balanced by in the
corresponding clause. The indicative furthermore fulfills Phil.'s initial threat,
and is far more satisfying than an optative.

The bosom of male or female; cf. 36.6

But in this poem of nautical imagery (as in 8), there is also a pun on
its sense of watery gulf or expanse, especially since the girl involved is named Naias,
i.e.Naiad; Rossi M. Cf. Argentarius 1 GP (AP 5.16.2)

The name Phil, has given elsewhere to a hetaira who is ready to

receive him when he is rejected by another woman (8). On the form of the name,
cf. Giangrande 43. The well-known hetaira Nais (Aristoph. Geryfades fr. 179 K-A,
Athen. 592c-e) could serve as eponym for later practitioners.
The MS of P is extremely difficult to read here because of C's overwriting, but
most likely P had which C converted to the comma
indicating (as elsewhere in C's corrections) the word break.
146 Epigram 24

24

AP 12.103 [Anon. 56 HE]

P P17.194, f. 76rs.a.n. IT ii. 28

I know how to love those who love (me); I know how to hate if someone treats me
unfairly. For in both (love and hate) I am not without experience.

Falivene, QUCC 37 (1981) 88 f.

Taran, Variation 8 n.2.

With no claimant in the Anthology, and with its incipit in the papyrus list, this neat
little poem may well be by Phil, especially as it complements 23. The Palatine An-
thology, on the other hand, places it, along with thirty-two other interspersed
anonyma, well within a Meleagrean context of 136 epigrams: 12.36-171. Gow,
Sources and Ascriptions 21 ff., is surely right to think that in general anonymous
poems belong to the series in which they are embedded (all thirty-three anonym
are printed in HE, one identified as a work of Meleager by PI), but he himself points
out that this is far from an absolute rule (ibid, and 41), for breaks in series may
occur when two poems from different sources are collated because of similar sub-
ject matter (Gow41 f.), and it may be that 12.103 was placed before 12.104 (Anon.
4 HE) for this reason, the narrators of both poems expressing a jealous hate:

AP 12.107 and 108 also deal with jealous love. Planudes placed this poem ninth in
a series of twenty-five anonyma at the end of his seventh book (amatoria). Of this
series, eight are identified as the work of three of Meleager's authors by P (Meleager,
Asklepiades, and Dioskorides), another three come from the same Meleagrean series
as 12.103, and two are identified as Philip's authors by P (Euenos and Archias). It
may be significant that in Planudes' twenty-five, which are generally interspersed
throughout P's two amatory books (5 and 12), AP 12.103 and 104 once again appear
together in the same order (similarly, AP 12.50 + 51 [missing, however, 50.7-8 and
51.1-2], 5.304+305).
In sum, although Phil, cannot be ruled out, the external evidence for author-
ship is inconclusive.
Although "not visibly erotic" (Taran), it could be recognized as such in the
right context, which, if Philodemean, would include 23. M. Lausberg, Das
Einzeldistichon (Munich 1982) 334, compares Euenos 7 GP (AP 12.172)
 Epigram 25 147

noting that neither is specifically erotic.

1-2: For asyndeton throughout, cf. 2.

1 Cf. 23.1

1-2 Here there is greater deviation from 23:

2 Phil, allows an enclitic ecm after the pentameter's caesura at 31.6. Cf. West,
Gk. Metre 26, who notes that these two forms can occur after caesura and even begin
a verse.

25

AP5.306[13GP,18K]
PP1 7.86, f. 72v Hv.13
1 CP1: P 4 Hecker: PI: P(v.comm.)
CC PI P

You cry, you ask for pity, you look me up and down, you are jealous, you
keep touching me and kissing me hard.
These are the deeds of a lover, but whenever I tell you I am ready and you
hold back, you have absolutely nothing of the lover in you.

Huschke 153-157.
Lumb 19.
Salanitro, Studi. . . Cataudella (Catania 1972) 2.498-501.
de Vries, Mnemosyne, 4th ser. 23 (1970) 30; 26 (1973) 179.

The topos of the segnis amator. A would-be lover fails to satisfy; cf. A. Richlin, The
Garden ofPriapus (New Haven 1983) 117 ff. Phil, also has 19 and 26. The former
148 Epigram 25

is spoken in persona feminae, for which the Anthology offers few parallels, the only
early erotic exx. being 36 (probably also by Phil, but ascribed as well to Meleager),
Asklepiades 19 HE (AP 12.153), and probably this poem as well. There are, it is true,
no grammatical markers to identify the gender of the speaker, but with a Philodemean
parallel for a poem of this topos spoken by a woman and with none for a homosexual
poem, I shall proceed on the assumption that this poem is an example of the former.
Since this topos is an exercise in public self-humiliation, there can be no better way
to accomplish this than by allowing the woman to revile the man.
Ovid Am. 3.7, the lament of an impotent lover, draws upon both this poem
and 19; note especially 77-80, where the woman taunts the man, and see below on
w. 3,4.

1-2: Asyndeton throughout these two lines; cf. Plato Phdr. 255e [sc. 6
240d (also in
an erotic context)

1 Used again with a verb of seeing by Strato

(AP 12.175.4), where the sense is "can't keep his eyes off," which works well here.
Note the role of seeing in the Phaidros passages cited above.

2 The first verb in the poem to make it clear that the context is erotic,
although this was strongly hinted at by the preceding phrase. Cf. Pi. Symp. 213d,
where Sokrates describes how Alkibiades has acted since he first fell in love with
him:

Phil, calls a vice at Rhet. 2.139.14 Sudh. (a de-

tached fragment). Cf. E. Fantham, "Zelotypia: A brief excursion into sex, violence,
and literary history," Phoenix 40 (1986) 45-57.
Touching to arouse, as often; cf. Pi. Phdr. 240d and 255e (cited above).
It can also serve as a euphemism for intercourse itself; cf. Aisch. Cho. 71
Ov. Am. 3.7.39 at qualem vidi tantum tetigiquepuellam. See further on 11.3.
Cf. Aisch. fr. 135.2 (v.l.
cf. Dawe, Collations . . . of Aeschylus [Cambridge 1964] 70, who lists sev-
eral other passages where MSS alternate between these two words). Although
adverbial can serve as a synonym for which has just appeared in
this line, here, as in the Aischylos citation above, it probably refers to mouths pressed
together hard and close; cf. II. 12.454
Ov. Am. 3.7.9 oscula luctantia.

3 Here ordered into a reasonable erotic progression

which would naturally end in sexual intercourse. Jacobs adduces Xenophon Ephes.
3.2.4
Note the generic use of the attributive participle (which, being
active, is more appropriate to men than women; cf. Gow-Page, Salanitro 498 n. 67)
without the article, as often in prose and verse (cf. II. 9.318, quoted on v. 4); KG
1.608 f.
 Epigram 25 149

"Whenever" throughout this one night; cf. 19.2. Not only

do we have a rare spondaic fourth foot (Intro., p. 43), it is especially surprising after
three dactyls, for whereas DDDDDS is among the commonest of metrical shapes
(> 12x in Phil.), DDDSDS is one of the rarest (indeed, only here in Phil.). The result
is a heaviness in these words which emphasizes the contrast with the preceding.
The entry in LSJ can lead one to conclude that this verb is used
only of inanimate objects or abstractions in various metaphorical and technical
senses; but the context makes it clear that the verb is here being used as a synonym
for the more usual and cf. PI. Symp. 213b-c, where
Alkibiades (shortly before the passage cited above on 2) first uses it of Sokrates in
the sense "lie in wait" and then again as an erotic synonym for
For
in an erotic sense, cf. e.g. AP5.2 (Anon.).

4 Emended by many (see below) and crucified in GP, the line ha

been defended by de Vries and Salanitro, who independently cite!/. 9.318

where the verb seems to mean "be/remain

inactive." (But in Homer it is very easy to understand = "death" as object.)
Cf. also, with Salanitro, Soph. El. 958 "remain indiffer-
ent." Although none of these passages makes the same demands on the verb as found
here, they are sufficiently close to defend (if not guarantee) its use in Phil. An
interesting parallel may be found in Aristoph. Peace 341 (Trygaios explaining his
idea of freedom to the chorus) where the sec-
ond pair is close to Phil, in sense ("screw or not," i.e. sleep [alone]), but since
may well have its usual erotic sense here (so Henderson, Maculate Muse 164),
would have to assume the precise sense it has in Phil; cf. Toup's conjecture, below.
Salanitro also adduces passages in Latin poetry where the woman expresses im-
patience with her lover, but the uses oiremorari, etc. in these lines are not quite paralle
because none applies to the inactivity of the segnis amator, who (note the present
general condition) has disappointed her more than once. The indicative of P probably
resulted from its position so soon after Lumb would understand
the word as a wrestling term, "can you last out about?" (cf. Mart. 14.201, citedbelow).
Let me keep the apparatus relatively clean by listing here, in what I consider to
be a descending order of probability, the various emendations suggestions for these
words and the next: Stadtmiiller (looked on with favor by
Gow-Page); (my suggestion; cf. Ov. Am. 3.7.77 male sane, Soph.
Aias 726 for apodotic Kcd, cf. Od. 13.79, etc., Denniston, GP 308);
Jacobs (1794); Paton;
Jacobs (1813); Meineke; Schmidt (but this verb
would destroy the contrast of v. 4); Toup (cf. Ov. AA 2.72
ff., esp. 725 veils maioribus usus); Huschke (cf. Mart. 14.201 e
didicit melius Pikkolos. Kaibel.
"Simply," i.e., the actions of w. 1—2 count as nothing when it comes t
defining the Lover if he can't produce the final action. Cf. Phil. De Ira col.28.26f.
(someone enslaved by anger cannot be a good juryman, Council member, ecclesiast,
150 Epigram 25

archon) Common in Plato and Aristotle, where, as Bonitz

(Index. Arist.76bl9).

26

AP5.120[7GP, K17]
P viii. 9 caret PI [J]

1 P: Ap.L, Leiden B.P.G. 34B, Cr 2 P: Ap.L 3 P

Hecker P: Ap.L v.comm. 4
P: Boissonade P: Jacobs

I came having stolen away from my husband in the middle of the night,
and having gotten wet into the bargain in a driving rain.
Was it for this that we (now) sit doing nothing, and talking we do not go
to bed as lovers should?

Hecker 47 f.
Lumb 12.

Another epigram narrated by a woman complaining about her disappointing lover;

cf. 19 and 25. Normally it is the man who complains, often in paraklausithyra, of
being rainsoaked; cf. Asklepiades 14, 42 HE (AP5.167, 189).

1 Although the adj. is common enough in poetry, it is rarely

applied to temporal nouns; before Phil, only Theogn. 998 afterwards only
Oppian Hal. 5.115 (night), Orphic Argonautica 536 (night), 649 (dawn).
Most exx. of this verb in the sense "cozen, cheat" (LSJ s.v. II) have
as object, but cf. Pindar P. 3.29 and//. 1.132, where
is easily supplied from context.
2 Cf. Soph. fr. 636 P-R:
 Epigram 26 151

Since this thought, admittedly a commonplace (cf. Pearson and Radt ad loc.),
is best known to us from Lucr. 2, proem (suave, marimagno etc.], it is possible that
the Sophokles passage was known to Phil, from Epicurean sources rather than from
his own readings in Greek literature. (Ap. Rh. 2.1083-1087 imitated Sophokles;
note especially 1083

3 Looking backwards and used ironically; otherwise no sense can be

made of the woman's complaint (Gow-Page). For the accent, cf. West on Hes.
Th. 88.
Jacobs (1794) correctly discerned the erotic im-
port of this and the following phrases (although he unnecessarily questioned the
soundness of the text). Gow-Page (following Jacobs) are right to take the preposi-
tional phrase as rather than Kaibel's in otiosorum numero. Cf. D.L.
4.7 (Phryne, failing to seduce Xenokrates) LSJ s.v.
3, Ov. Am. 3.7.15 truncus iners iacui. Note also AP 12.240.3 (Strato)
(cited in full at 19, intro.).
Are both participle and finite verb negated, as Gow-Page
argue, adducing Aisch. Ag. 290f., Thuc. 6.33.1, Soph. fr. 88.8? It is true that
occurs a total of seven times in Phil., always in generally erotic contexts,
so that words (of love) could be regarded by this woman as important as the
lovemaking itself. On the other hand, talking in itself is no substitute for
lovemaking, and could be thought of as part of "sitting and doing nothing." In
support of this, Waltz compares 25, also spoken to a segnis amator; in particular
On balance, I favor Waltz's view: (i)
The rarity of epigrams spoken in the persona of a woman suggests that 25 and 26
were intended as companion pieces, (ii) A participle can be excluded from an
initial negation of A. C. Moorhouse, Studies in the Greek Negatives (Cardiff 1959)
107 f., Fraenkel on Aisch. Ag. 1312. And in this case, the meaning would be even
clearer for an audience if they already heard either 25 or another, nonextant,
variation on this theme.
Kaibel, Gow-Page, and Waltz deserve credit for making sense of the text as
transmitted. The irony of the woman's question and the erotic import of
and have been lost on earlier editors, who make various emenda-
tions. (Lumb), (Diibner), (Stadtmiiller),
(Herwerden).

4 A common euphemism; cf. Od. 8.313

(Sappho 168B.4 Voigt, which uses the verb in the
basic sense, gains in poignancy because makes one think of what "sleeping
with someone" means.)
The text is sound: They deserve the description "lovers," because
they have gone to bed together in the past. Once again, Phil, plays upon the way
lovers use language. The various attempts of the editors here and above simply make
explicit what Phil, keeps colloquially euphemistic.
152 Epigram 27

21

API 1.44 [23GP.22K, 18G]

P niv. caret P1

I 2 Salm 3
Brunck:- P ' 6 PPC: Pac

Tomorrow, friend Piso, your musical comrade drags you to his modest digs
at three in the afternoon,
feeding you at your annual visit to the Twentieth. If you will miss udders
and Bromian wine mis en bouteilles in Chios,
yet you will see faithful comrades, yet you will hear things far sweeter than
the land of the Phaeacians.
And if you ever turn an eye to us too, Piso, instead of a modest feast we
shall lead a richer one.

Braga, Catullo e i poeti greti (Messina 1950) 195-199.

Chardon de la Rochette, "Deux epigrammes," 200—209.
Clay, "The cults of Epicurus," CErc 16 (1986) 11-28.
Dettmer, "Catullus 13: A nose is just a nose," Syllecta Classics 1 (1989) 75-85.
Edmunds, AJP 103 (1982) 184-188.
Geffcken 341.
Gigante, Philodemus in Italy 79-90.
Hiltbrunner, "Einladung zum epikureischen Freundesmahl," m Antidosis: Festschrift W. Kraus
(Wien 1972) 168-182.
Landolfi, CErc 12 (1982) 137-143.
Marcovich, QUCC 40 (1982) 131-8; repr. in Stud, in Gk. Poetry 193-199.
Schulze BPhW36 (1916) 318.
SnyderC/68(1973)350f.
Williams, Tradition and Originality (Oxford 1968) 125 f.

In a poem sent on the nineteenth of an unknown month (but quite likely not
Gamelion; see below on 3 ), Phil, invites Piso to attend on the next day the
celebration in honor of Epicurus (and some other early Epicureans) known as the
Twentieth . The evidence for this festivity has been gathered, translated,
 Epigram 27 153

and analyzed by Clay; the "test." numbers appended to ancient sources below are
those of his testimonia, where they can be conveniently consulted. Clay's central
text (P.Herc. 1232 fr. 8 col. 1 - test. 16) is a section from Phil.'s On Epicurus which
discusses Epicurus' own invitation to attend a feast. Several points of contact be-
tween Phil.'s prose discussion and his poem will be given below in the appropriate
lemmata, but Clay's translation of this important fragment should given in full to
provide what context there is: " . . . as concerns those who experience turmoil and
difficulty in their conceptions of natures that are best and most blessed. [But
Epicurus says] that he invites these very people to join in a feast, just as he invites
others—all those who are members of his household and he asks them to exclude
none of the 'outsiders' who are well disposed both to him and to his friends. In
doing this [he says], they will not be engaged in gathering the masses, something
which is a form of meaningless 'demagogy' and unworthy of the natural philoso-
pher; rather, in practicing what is congenial to their nature, they will remember all
those who are well disposed to us so that they can join on their blessed day (?) in
making the sacred offerings that are fitting to ... Of the friends ..."
Phil. De Pietate 812-819 Obbink (which now replaces test. 17) also refers to
an invitation to an Epicurean dinner; see below, on 1 .If the invitation to an
Epicurean celebration appears even these few times in the fragmentary papyri from
Herculaneum, we may imagine that it showed up far more often in the lost writings
of the various Gardens. There may be no need, therefore, to search earlier Greek
literature for the origins of the poetic invitation, as Edmunds 187 f. ably argues.
But rather than reflecting "a Roman social convention" (Edmunds), it probably
derives, at least primarily, from Epicurean conventions and concerns. And as I
suggest in the introduction to 28, Greek invitation poems can take other forms.
That this poem would give rise to a minigenre in Latin never entered Phil.'s mind;
cf. E. Gowers, The Loaded Table: Representations of Food in Roman Literature
(Oxford 1993), ch. 4, "Invitation poems"; Cat. 13 (which for all we know may have
preceded 27); Hor. O. 1.20 (to his patron Maecenas), 4.12, Epist. 1.5 (to celebrate
Augustus' birthday; on Horace's invitation poems, cf. Tait 68-70); Mart. 5.78,10.48,
11.52; Juvenal 11.56-76. (Phil.'s poem does not qualify as a birthday poem, for which
now see K. Burkhard, Das antike Geburtstagsgedicht [Zurich 1991].)

1 Epicureans should not live for tomorrow; more precisely they should
live each day as though it were their last (bearing in mind that death is nothing to
us); cf. (fr. 490
U = 215Arr), ' (VS14). Phil.
alludes to this doctrine inDe Morte IV col. 37.26 f
cf. M. Gigante, RP 181 (text), 193 f. (commen-
tary, with parallels from non-Epicurean literature). Phil, may intentionally be dis-
arming Piso with this un-Epicurean note by ironically looking ahead with lavish
promises to an event which includes Epicurean instruction. For + pres.,
see below on 2
Simplicity and frugality are Epicurean virtues: cf. Ep. Ep. 3.130
Phil, urges
(De Morte IV, col. 30.10 f.) and contrasts
154 Epigram 27

(Oec. 38.7 £); see further Hiltbrunner 169 £; Gigante

82 f. The frequent contrast in Epicurean literature between and
would seem to make all the more pointed Karneades' mocking of Epicurus for
supposedly asking the rhetorical question
(Plut. Non Posse 4.1089c = test. 21). Yet the Epicureans in fact may well have praised
the lavishness of their own celebrations; two fragmentary papyri making mention
of Epicurean cult and feast use the word reasonably restored in both places as forms
of (P.Herc. 176 Fr. 5 XVII1-7 = test. 15, referring to the cult of
Pythokles; Phil. De Piet. col. 29 Obbink = test. 17). Perhaps, then, Phil's refer-
ence to the frugality of the next day's celebration is ironical; cf. Hor. O. 1.20.1 vile
potabis modicis Sabinum \ cantharis,Epist, 1.5.1-5, Mart. 11.52.
Some otherwise unaccountable ink suggests that the scribe of tried to
correct his error (Parsons).
"A depreciatory term, not necessarily to be taken at face-value"
(Gow-Page ad loc.); that is, presumably, Phil, self-effacingly calls his home a hut,
which fits well with the adj. "humble"; cf. Vergil's referring to the house he inher-
ited from Siro as a villula (Catal. 8.1; Hiltbrunner 169); Cic. De Fin. 1.65 Epicurus
una in domo, et ea quidem angusta, quam magnos quantaque amoris conspiratione
consentientes tenuit amicorum greges! quod fit nunc ab Epicureis.
But the word also can mean "shrine" or "chapel" (IG 22.1533.5 [iv c. B.C.],
D.H. 3.70.2, Plu. Num. 8.8, etc.; so also the related Krinagoras 43 GP [AP
6.253.3] ), a sense equally appropriate to a poem in which a friend
of the Muses invites Piso to a near-religious occasion.
Where is this ? Gigante 79, who thinks that Phil, lived in the Villa dei
Papiri, argues for Rome, but Herculaneum or environs seems more likely; see below
on 3 . The fact that Phil.'s library was found in the Villa dei Papiri does
little to prove that Phil, ever lived there. Philippson RE 19.2945 and Hiltbrunner
169 suggest that Phil.'s house was given to him by Piso, but if so would the poet
have used such depreciatory language in describing it to the donor?
One invited to a simple dinner may appropriately be called friend;
cf. Hor. O. 1.20.5 care Maecenas. Since the dinner, furthermore, is in celebration
of Epicurus, whose company regarded themselves bound by a special bond of
the word easily picks up this sense as well. And since Phil, is here addressing a Roman
who will be asked to grant Phil, favors (at first, no more than his presence), the
bond betweenpatronus and cliens known euphemistically to the Romans as amicitia
is also evoked. Cf. Lucretius' words to Memmius: tua . . . sperata voluptas \ suavis
amicitiae(1.14Q{.).
On Epicurean friendship, cf. Cic. De Fin. 1.20.65 (Epic. fr. 539 U) de amicitia
. . . Epicurus quidem ita dicit, omnium rerum quas ad beate vivendum sapientia
comparaverit nihil esse maius amicitia, nihil uberius, nihil iucundius\ frr. 540—546;
Rist, Epicurus (Cambridge 1972) 127-139; B. Gemelli, "L'amicizia in Epicure,"
Sandalion 1 (1978) 59-72; P. Mitsis, "Epicurus on friendship and altruism," OSAP
5 (1987) 127-153.
On amicitia and Roman patronage, and on Piso as Phil.'s patron, see Intro.,
pp. 5, 14; P. White, "Amicitia and the profession of poetry in early imperial
Rome," JRS 68 (1978) 74-92; R. Sailer, "Patronage and friendship in early impe-
 Epigram 27 155

rial Rome," in A. Wallace-Hadrill (ed.), Patronage in Ancient Society (London 1989)

49-62.
This spelling of the Latin name Piso owes nothing to iotacism, but is
rather the standard absorption of the name to a preexisting Greek name built on
the stem ; cf., e.g., the son of (LGPN 1 s.v.). For more
on the Latin name, which may be of Etruscan origin, cf. W. Schulze, Zur Geschicbte
lateinischerEigennamen, (Abh. d. Gott. Ges. d. Wiss., ph. -hist. Kl.2 5 (Berlin 1904)
209-211. For the vocative form cf. comm. ad 38.7.

Sc. cf. Poseidippos 10 HE (AP5.183.6)

The festivities begin ca. 3 p.m., the usual hour; cf. Hor. Ep. 1.7.71
post nonam venies. The dinner in 28 begins an hour later.
A futuristic present, most often found in the company of a word indi-
cating the specific, and usually not too distant, future time; cf. P.Ryl. 233.7 (iiA.D.)
"it is (to be) roofed tomorrow"; other epistolary exx. of
+ pres. areP.FuadIUniv.31.r.l2f.,P.Mil.Vogl. 50.4,50.10,P.Oxy. 1931.7,3758.120.
Cf. Schwyzer, Gr. Gr. 2.273; B. Mandilaras, The Verb in the Greek Non-Literary
Papyri (Athens 1973) 215(a). The Latin invitation poems use the future (cenabis
bene/belle, potabis, etc.); cf. Nisbet-Hubbard ad Hor. 0.1.20.1. The verb, used for
dragging dead bodies on the battlefield (II. 13.383 etc.) and for hauling reluctant
people into court (Aristoph. Nu. 1218 etc.) or elsewhere (cf. 15.5), must presum-
ably here be understood not so much as referring to great reluctance on Piso's part
but rather as an oblique allusion to the humbleness of Phil.'s home, which, Phil,
suggests, would not be entered readily by someone of Piso's standing. The equiva-
lent of test. 16 is in De Piefate 818-
819 Cf. Alkaios 368
M:
"Musical," i.e., a poet; an especially Hellenistic notion derived
fromHes.ra.96f.
Cf. Kallim.yl^. 1.2 37-8 M
Theokr. 1.141
(Daphnis), 7.95 (Lykidas), 11.6
(Nikias), Meleager 1 (AP 4.1.1) Mo Nossis
11 HE (AP 7.718.3; text uncertain).
Phil.'s word, a hapax, is probably passive, "dear to, i.e. loved by, the Muses,"
as with the similarly formed (Hiltbrunner 171), which develops an active
sense only in late authors (Philo, Lucian). Note, though, the active
(Polyzelos 11 K-A, Theokr. 8.60 and which, according to £ Aristoph.
PI. 550 = But since is a reciprocal relationship, especially as one
imagines one's dealing with divinities, the passive almost entails the active. Cf. Phil.
De Dis 3 col. l-17f.
"let him call wise men friends of the gods and the gods friends of the
wise." Thus, is prob. meant to combine the passive sense of Korinna's
and the active sense of used by, among others, Phil. De
Musica col. IB.2 Neubecker (note also "music lovers," ibid,
col. 22.12). In this Epicurean context (see above, on ) the word thus sug-
156 Epigram 27

gests that Phil, identifies himself as someone who is both a poet and a Epicurean.
More specifically, it obliquely suggests that among other entertainments at this
celebration Phil, will recite some of his epigrams; see below on 5
The poetical equivalent to . Epicurus uses only the latter, Phil,
the former perhaps only once (Pragm. col. 25.14 Diano). That this is the word used
repeatedly for Odysseus' companions may not be irrelevant.

Epicurus stipulated in his will that money be allocated from his estate
for (i) sacrificial offerings to himself (as well as to his father, mother, and brothers),
to be made every year on the of the month of Gamelion, a day
already celebrated within the school in his lifetime; and for (ii) the continuance of
the customary meetings held on the of every month, in which those who agree
with his philosophy are to commemorate both himself and Metrodoros (D.L. 10.18
= test. 1,2). (Cf. the festival to Theseus held on 8 Pyanepsion and the lesser sacri-
fices in his honor on the 8th of every other month; Plut. Th. 36.) This relationship
between his birthday and the celebrations on the 20th was badly misunderstood
until quite recently, because it was thought, first, that Epicurus was bom on
7 Gamelion, and then that his birthday celebration was to be held on the tenth of
this month. In an important note, however, D. M. Lewis, CR, N.S. 19 (1969) 271 £,
showed that (nom.), given as the date of his birth at D.L. 10.14 may well be
nothing more than a intrusive gloss identifying Gamelion as the seventh month of
the Athenian year; and, second, that the term unambiguously
refers to the twentieth day of a month rather than, as had been universally assumed
by students of Epicurus, the tenth (on this form of dating cf. further B. D. Meritt,
Athenian Year [Berkeley 1961] 46 n.6; TAPA 95 [1964] 208 n.27). Thus earlier
confusion (cf. Gow-Page ad loc. e.g. for a typical statement of the problem of the
various dates—7th, 10th, and 20th—as it was then known) resolves itself: Epicurus'
birthday, its celebration, and the monthly Epicurean gathering all occur on the
twentieth. See also K. Alpers, "Epikurs Geburtstag," MH25 (1968), 48-51, who
reaches the same conclusion as Lewis.
It is important to note further that during the month of Gamelion the sacrifi-
cial offerings, presumably held at some suitable outdoors site, were considered
distinct from the regular monthly meeting, although they may well have been con-
sidered the most important of the monthly meetings, if only because of the con-
sumption by the celebrants of the sacrificial offerings; cf. Clay 18 f. That the two
events were kept separate seems clear both from Cicero's close translation of
Epicurus' will (Fin. 2.101 = test. 4), which continues to treat annual birthday cele-
bration and monthly meeting as distinct events; and from Pliny, NH 35.5 = test. 10
natali eius sacrificant, feriasque omnimense vicesima luna custodiunt, quas eiKdSag
vacant. Note also Menippus, who wrote
(D.L. 6.101 = test. 3), which, whether one work or two, argues
for funeral rites apart from the monthly celebration. (One work: Wachsmuth,
Sillographi (Leipzig 1885) 82 n.l. Two works: Clay; Paton [D.L. Loeb].)
In Epist. 1.5 Horace invites a friend to a meal in honor of Augustus' birthday:
9 eras nato Caesare festus.
 Epigram 27 157

This verb almost always has a personal object—"to feed (some-

one) dinner"—and when, only rarely, it is used absolutely, the sense remains "to
feed"; cf. e.g. Plut.Mor. 823e . It never means "to celebrate." Here
the oe of v. 1 is the object of both main verb and participle, as often; it is (see above)
further modified by . This is missed in such translations as "giving a din-
ner for the anniversary of the Twentieth" (Gow-Page) and "er feiert das jahrliche
Festmahl des Zwanzigsten" (Hiltbrunner). E here = "feast of the Twenti-
eth" (ferias . . . quas vacant, Pliny NH 35.5), serves as an internal object,
for which cf. Plut. 1089c (quoted above, on 1 , where note that is
intransitive), Matron (iv c. B.C.), Conv. Attic. 2 [sc. ]...
One expects a future participle, as in Aristoph. Clouds 1218
a futuristic present participle dependent upon an already futuristic
present finite form (see on ) is unusual; cf. M. Sanchez Ruiperez, Estructum
del sistema de aspectos y tiempos del verbo griego antiguo (Salamanca 1954) 92 f.:
"la imaginacion presenta como actual un contenido verbal aun perteneciente al
future."
This is universally understood to modify although
"annual twentieth" produces no immediate sense. The idea that the phrase could
signal that Twentieth which also celebrates Epicurus' birthday cannot be right,
since, as has just been shown, the two occasions were kept distinct. Gow-Page
desperately and with little enthusiasm offer the suggestion that these words might
mean "yearlong (not 'annual') Twentieth," i.e. the twentieth of each month
which is celebrated throughout the year. The problem is resolved when we real-
ize that in poetry (Alkaios 130.35, Ion 19 F 21, Eur. Hipp. 37), as in many prose
authors (Herodotos, Hippokrates, Plato, Aristotle, Xenophon et al.), evicruotoc
is an adjective of three terminations. [For exx. of two-termination
cf. W. Kastner, Die gr. Adjektive zweier Endungen auf- OZ (Heidelberg 1967)
88.] And since (see previous lemma) always takes a personal object, the
adjective would be understood to modify Piso, in the way that Greek temporal
adjectives such as (15.2) often do. Cf. Horn. Epig. 15.10 (theEiresione)
If this analysis, which makes
better morphological sense, and which makes no strange demands upon mean-
ing, is accepted, Phil, would be addressing Piso at a time when the latter came to
celebrate a Twentieth only once a year: "Philodemus invites you to your annual
visit," exactly the same sense this adj. has in the Eiresione. Whether this is a
special celebration, to which many "outsiders" (as presumably Piso is at this
point; cf. test. 16 cited above) are invited (here we can consider the possibility
of the occurring in Gamelion), or whether it is a date designed to suit
Piso's schedule (such as during a regularly scheduled visit to Naples) cannot be
determined.
The possibility should also be considered that is designed to recall
the situation of the Eiresione (see above), in which the beggar makes his annual call
to the house of the rich

Here in a playful reversal it is the wealthy man who reluctantly (see above, on
158 Epigram 27

comes to house of the poor, and who will, again in a reversal of the sense found in
the Eiresione, bring at least the promise of future wealth along with him. On beg-
ging songs in general, cf. W. Burkert, Greek Religion 101 f.; Frazer, Golden Bough
(3d. ed.) 8.317 ff.; M. Nilsson, Ges. d. gr. Rel. 1.124. W. Schmid's conjecture is
obviously unnecessary ("Epikur,"R.L4C5 [1962] 749 f. =Ausg. Philol. Schr. [Ber-
lin 1984] 208).

Sow's udders could be boiled and then grilled, and could be stuffed
before cooking (Apicius 7.2). According to Galen they were most appreciated when
full of milk (De rebus boni malique suci 6.774 f. K.; Plut. De esu earn. 997a gives the
disgusting steps taken to attain this gourmet's delight.) Since they were expensive,
they would be out of place at Phil.'s simple table; cf. Plut. Mor. 124f

Martial 11.52.13 also lists sumen among the delicacies that will not be served at a
meal. Furthermore, since Greeks were far less fond of them than Romans (cf. Athen.
9.399c = 14.656e), there may be a touch of humor in this remark, much as if one
were to invite a French friend to dinner with an apology for not offering snails.
Cf. Chardon 204-206.
Epicurus himself is said by Karneades to have boasted of his con-
sumption of Thasian wine (Plut. Non Posse 1089c = Epic. fr. 436 U = test. 21).
Generally Chian wine was most highly praised (see on 6.1 f.),butEuboulos fr. 121
K-A suggests that when
old there would be little to choose between them.
Derived from the nounX areX cf. Eustath. ad//. 1.35), X
cf.. ),andX ("madebyX ";cf.M ).The
artificial lengthening of X is either by false analogy with the preceding words
in XI- or an example of the common epic lengthening of the first of three shorts to
have the word fit the meter. is a close parallel; cf. W. Schulze, Questiones
Epicae (Giitersloh 1892) 8, 140-179.
Not simply "drink" (Gow-Page), but the vinous toasts and pledges
made after dinner and before the symposium (Latinpraebibere, propino); cf. Athen.
675b W. Heraeus, ". ," RM 70 (1915) 1-41; repr. in KS
(Heidelberg 1937) 190-226; esp. 217. Gigante 83 compares Hedylos 2 HE
(AP 5.199.1)

is apodotic after a (usually negated) protasis; Denniston,

GP 11-13: "even though . . . still." The anaphoric second clause produces a
vivid exegesis of the first; cf. Hdt. 7.11.2
[sc. Cf. Cat. 13.9
sed contra accipies meros amores.
"Altogether true"; cf. PL Rep. 583b Hor. Epist.
1.5,24 fidos inter amicos.
Recitations during dinner were standard (cf. Kay on Mart.
11.52.16), but what in fact will Piso hear (and see, apart from true comrades) at
this Epicurean party? Since Phil, does not write purely ornamental epithets,
 Epigram 27 159

suggests some sort of literary activity, as does the reference to the

Phaeacians (see below). Eikades generally may have been more prosaic affairs, the
participants parading about with images of Epicurus (Pliny NH 35.5 = test. 5) and
reciting long passages ( ) on the virtues of Epicurus, Metrodoros,
Aristoboulos, and Chairedemos (Plut. Live Unknown 3.1129a = test. 20); cf. Clay
and M. Capasso, Cameisco: Ilsecondo libra del "Filista" (P.Herc. 1027) (Naples 1988)
37-53. But Cicero's hostile references to these gatherings, however much they have
to be tempered, would seem to suggest that in Piso's Rome and Naples Epicureans
did more than simply praise their predecessors: Quid ego illorum dierum epulas,
quid laetitiam et gratulationem tuam, quid cum tuis sordidissimis gregibus
intemperatissimas perpotiones praedicem? (In Pis. 22); omnia cenarum genera
conviviorumque (ibid. 70, the Phil, passage).

Mention of the Phaeacians recalls Odysseus' stay in Scheria, and per-

haps in particular his praise of good poets. Phil., that is, is here obliquely compar-
ing himself both to Demodokos, who received extra meat for his singing, and to
Odysseus, who received additional gifts for his account (compared to that of a bard
by Alkinoos; 11.367 f.). Piso will no doubt get the hint. In addition, the Phaeacian
were famous in later literature for their luxurious lives and more particularly for
their feasts—S.Eitrem, "Phaiaken," RE 19 (1938) 1532 f.— cf. Horace, Epist.
1.15.22-24:

tractus uter pluris lepores, uter educet apros;

utra magis piscis et echinos aequora celent,
pinguis ut inde domum possim Phaeaxque reverti.

Epicurus himself was called (Herakleitosyl//eg.

Horn. 75 = Epic. fr. 229 U). See further E. Asmis, "Philodemus' poetic theory and
On the Good King according to Homer" CSCA 10 (1991) 1-45; M. Jufresa, "II mito
dei Feaci in Filodemo," La regione sotterrata dal Vesuvio: studi e prospetti. Atti del
Convegno Internazionale, 11-15 novembre 1979 (Naples 1982) 509-518.
A compendious comparison (KG 2.310 f.) standing either for (i)
"you will hear things sweeter than the Phaeacians heard" (Gow-
Page, Hiltbrunner; better would be ), or
(ii) "you will hear sweeter tales than those told about the
Phaeacians" (Kaibel). Cf. Soph. Ph. 680 ff.
where grammatically can = either
or (the case here) s. Sense (i) may well be primary, with Phil, thus compar
ing himself to Odysseus and Demodokos, both of whom sang to the Phaeacians
(see previous lemma); but sense (ii) need not be absent, which would also have
Phil, comparing himself to the poet Homer, who sang o/the Phaeacians, and hence
of the things Epicureans valued. If Phil, were to recite epigrams, both senses would
be satisfied.
The comparison may owe something to PI. Rep. 614b
(sc. Er). Jufresa op. cit. 51
160 Epigram 27

follows Aubreton (Bude) in seeing Piso as the Odysseus figure in this poem, since
it is he who is to come as a guest of the Phaeacians, who, Jufresa 512 f. has shown,
were used by Phil, in Good King as the model of a Utopian Epicurean community.
(Cf. Juv. 11.61, an invitation to Persicus, venies Tirynthius, i.e., Herakles.)

This sentence seems to look beyond the next day's festivities (contra
Gigante 85), although prob. not exclusively to the next event of this sort, as Kaibel
thought. The thought here accords best with the view that Piso, a sympathetic "out-
sider" to the Epicurean community, is being asked to take a greater part in the
future—in the community in general and perh. as a Phil's patron in particular.
Since an eye can be either friendly or hos-
tile, it is often labeled one way or the other. (Aischylos offers several exx. of both
types; c f . e.g. S e . 3 5 9 Cho. 8 1 0 f . ) C f . Meleager 1 0 8 ( A P
12.159.5 f.):

Although I can find no exact parallel for Phil.'s unmodified eye, perhaps we may
compare Alkman 1.55 [sc. ] and the adj. "ad-
mirable" (cf. Rose ad Aisch. Cho. 350). Perhaps Phil.'s phrase translates Lat. respicio,
as would be appropriate when addressing a Roman; cf. Verg. A. 4.275 Ascanium
surgentem respice, OLD s.v. 8a. 'H must be scanned a trisyllabic to avoid vio-
lating Naeke's Law; similarly 5

"Conduct, celebrate," as in P.Herc. 176 fr. 5 col. 27.15 f. = test. 14

and Phil. De Piet. 812-814 (sc. ), and
common enough elsewhere; LSJ s.v. IV 1.
For which thought Epicurus may be thought to
provide justification: (VS 63). For K = "after, as a
change from," cf. Eur. Or. 279

28

5
 Epigram 28 161

API 1.35 [22 GP, 19 G]

P [sc. ] IIvii.17 " caret PI
5X Page: P Reiske: > P 6 Meineke: Kat P: Reiske

Artemidoros has given us cabbage, Aristarchosbaccala, Athenagoras spring

onions,
Philodemos a small liver, and Apollophanes two pounds of pork (three
are left from yesterday).
Slave, get us Chian wine, wreaths, sandals, and myrrh: I want to have them
in at 4 P.M. sharp.

Cichorius, Romische Studien 297 f.

Gigante, Philodemus in Italy 59-61.
Giangrande, QUCC 15 (1973) 17-19.

Like 29, a poem listing the modest ingredients of a meal which would be appropri-
ate for Epicureans, some of which occur in both poems. And once again friends
are named, each of whom in good eranos fashion is expected to show up with his
share of the meal. For other poetic preparations and anticipations of a meal, cf. 27,
Asklepiades 25,26 HE G4P5.181,185), Poseidippos 10 HE (5.183); A. Wifstrand,
Studien zur griechischen Anthologie (Lund 1926) 63 f. for parallels from Greek and
Roman comedy.
The poem contains the line "Phil, has given us a small liver," which allows Gow-
Page to entertain the possibility that the poem, written by another, was assigned to
Phil, solely because his name is among the invited guests. But can easily in-
clude Phil, ("for us all"), and Phil, refers to himself in the third person in the invi-
tation to Piso, and perhaps also ii.12. Cf. Ephippos fr. 15.11 K-A (a master giv-
ing his slave a shopping list) Note that, in contrast with 27,
Phil, here casts himself as an equal among equals.
If each of the (other) persons named received this poem from Phil, early in the
morning, it could serve either as a reminder of an earlier invitation or, as I think more
likely, as the invitation itself (Tait 68 so understands it). The instructions to the slave
in the last distich would thus be essentially a fiction which provides the time of the
meal. If so, we can expand the brief corpus of invitation poems to include those of
this sort. Asklepiades 26 HE (AP5.185) and Poseidippos 10 HE G4P5.183) could
similarly serve as reminders or invitations. Two prose invitations for the same day:
P.Oxy. 1485 (ii/iii A.D.), 1486 (iii/iv A.D.). Perhaps IT v.23 begins a similar
poem, n iv.3 , vii.8 , and vi.10 could also
have been invitation poems. On the (Aelian VH 8.7, = vocatio ad
cenam) as a genre, cf. F. Cairns, Generic Composition in Greek and Latin Poetry
(Edinburgh 1972) 240-245.
Cicero thought that Epicureans spent too much time talking about food:
(Epicurus) ipse quamparvo est contentus! nemo de tenui victu plura dixit (Tusc.5.89
= fr. 472 U).
162 Epigram 28

1 Cf. 29.2. Listing of the food to be enjoyed is a standard feature in

invitation poems: Cat. 13, Mart. 5.78, 10.48, 11.52; Juv. 11.56-76.
The most attractive candidate remains the one identified by
Cichorius 297: The orator A. of Knidos, son of Theopompos and friend of Julius
Caesar, who as Piso's son-in- law provides a link, should one be needed, between
Phil, and Artemidoros. Plut. Caes. 65.1 calls him 'E
"Artemidoros (28)," RE 2 (1896) 1330 f. [G. Hirschfeld, "C. Julius Theupompus
of Cnidus," JHS 7 (1886) 286-290, argues that Artemidoros was the father not the
son of Theopompus.]
The guests are also named in one of Horace's invitation poems: Epist. 1.5.26 f.
(Butra, Septicius, Sabinus).
Masc. or neut.; cf LSJ s.v. ad fin. Salted and dried fish, frequently
mentioned in comedy, was an important part of the ancient diet; cf. R. I. Curtis,
Garum andSalsamenta (Leiden 1991), esp. 6 f., 10 f., 16-19.

2 This diminutive only here. , the bulbous roots of various

plants, usually grown wild, figure in the diet of the second stage of Sokrates' early
state (Rep. 372c); see on 29.1 Reflecting the same attitude towards this
food as Glaukon's, Herakles rejects preferring instead beef, a
real man's food (Euboulosfr. 6 K-A) Cf. Philemon fr. 113 K-A ( s c . )
Cf. Apicius 7.14. for recipes and Athen.
63d-64f for instances of its occurrence in Greek literature.

3 Jacobs thought perhaps that of a goose (cf. Hor. Sat, 2.8.88fictspastum

iecur anseris albae), but lamb, kid, hare, and pork were also eaten. Gigante 60 is
probably right to say that a less luxurious liver than goose would be more appro-
priate for this Epicurean company.
Identified by Cichorius as the freedman of Pompey who served
as admiral and who went over to Octavian's side in 38 B.C.; "Apollophanes (10),"
RE 2 (1896) 165. Cichorius argues not only that he would have lived in Rome after
this (which is likely) but that this date therefore provides a terminus post quern for
the poem's composition (which is unlikely). Gigante 60, however, doubts Cichorius'
identifications of Artemidoros and Apollophanes, questioning whether the people
who will share this meal with Phil, would be such a socially distinguished group.
But if Piso could be invited to share in a simple Epicurean fare, could any Greek
be too socially elevated to be invited?

4 Humble fare, as when Eumaios tells Odysseus:

(Od. 14.80 f.). Giangrande, however, reasons that
since hetairai are regular features of Greek symposia, and hence figure in sympotic
epigrams, this word must haves its well known obscene meaning here. But if Phil,
is inviting or reminding his guests of a meal set along Epicurean lines, they would
not expect to find hetairai present (so Gigante). Furthermore, the sentence would
 Epigram 28 163

read very oddly with Giangrande's meaning. Pork, moreover, was standard fare;
cf. Alkaios71

5 Sc. cf.6.1-2 Although the MS can stand as read,

a singular for plural (see e.g. Gow's index to Theokr., s.v. "number"), eggs are
not so special as to be singled out for a feast (though of course they were eaten:
P.Petr. 3.142 lists among items to be purchased for a household), whereas
wine makes a regular appearance in this sympotic genre; cf., e.g., Hor.Saf. 2.8.15
ferens Alcon Chium, O. 4.12.16 (an invitation ode) nardo vino, merebere. Gow-
Page's suggestion (printed by Page in the text of his OCT) is therefore easy and
attractive; only slightly less so, because of its rarity, is Giangrande's , which
occurs in the sense "cup of wine" only in Antiphanes 85.2 K-A
and Tryphon's Onomatika (cf. Athen. 11.503de). Unlikely are o vov (considered
and rejected by Gow-Page) and (considered and rejected by me), the ear-
shaped shellfish mentioned by Asklepiades 25.8 (cf. LSJ s.v. o II3; v.23
). Gigante, retaining the MS, suggests that an egg-shaped cup is meant, as
mentioned by Deinon 690 F 4 FGrHist but this spe-
cial cup is reserved for the king of Persia and, golden or otherwise, is unparal-
leled elsewhere.
Cf. Hor. O. 3.14.17 ipete unguentum, puer, et
coronas.
These Greeks seem to have adopted the Roman habit of donning
slippers when entering the house (which would be taken off while dining); thus,
Nasidienus' calling for his slippers (soleasposcit, Hor. Sat. 2.8.77) indicates that he
is ready to leave the dinner. See further J. Marquardt, Privatleben derRomer (Leipzig
1879) 1.313; Hug, RE 1A (1920) 2261; Hor. Epist. 1.13.15.
The form , which is said to be Aeolic, is found in Sappho, Anakreon,
Hipponax, Herodas, Kallimachos, et al.; cf. Hesych. = Herodian Orthograph.3.2.578
Cf. Headlam-Nock on Herod. 7.60.

6 The general sense is "get, have ready," which may for some of the
items entail "buy" (a common meaning; cf. Gow on Theokr. 15.19, Asklepiades
25.1, 26.4f.)—wreaths have to be fresh, and perhaps wine (if that is what is to be
read) is running low—but sandals and a cup would be simply have to placed in
position; cf. Alkaios 346 Antiphanes 85 (cited above).
Other poems containing commands for slaves to obtain and prepare items for din-
ner are Anakreon 356,396 and Ephippos fr. 15 K-A; see further Nisbet-Hubbard
on Hor. O. 1.38, p. 421 f.
For other directions to a slave in this genre, cf. Asklepiades 25 (impera-
tives without a vocative, but the slave is slightingly described in the third per-
son), 26 ( ), Poseidippos 10 ( ). Giangrande would retain the
MS' K i, understanding it as explanatory—get (the items listed), because I wish
to begin on time—but this would be a strange instance of explanatory K i, which
usually explains by being more specific than what preceded.
See on 27.2
164 Epigram 29

29

AP9AU [20 GP, 23 K, 23 G]

PPlla. 36.12, f. lOr vii.21 [C] ov
2 PI: P PP1: G o w - P a g e 3 P P 1 :
Dilthey: PP1:
Scaliger: Kaibel PacPl: PP^ 4 ppj: . vel
- Scaliger: - Schneider 6 P: - PI

PHILODEMUS: Already the rose and chickpea and first-cut cabbage-stalks are
at their peak, Sosylos,
and there are sauteed sprats and fresh cheese curds and tender curly let-
tuce leaves.
But we neither go on the shore nor are we on the promontory, Sosylos, as
we always used to.
SOSYLOS: Indeed, Antigenes and Bakkhios were playing yesterday, but
today we carry them out for burial.

Del Re, Epigrammi greci 84, 129 f.

Gigante, Philodemus in Italy 54-59.
Luck, Rev. de Philol. 33 (1959) 46.
Stella 276 ff.

The death a day earlier of two friends reminds Phil, of the meals they will no longer
share. For the sentiment of w. 1-6, cf. Kallim. 44 HE (AP7.519):

A boating accident would account not only for the death of Phil.'s friends together
but also for his aversion to viewing the sea. But if this is the case, Phil, is failing to
 Epigram 29 165

observe the proper emotional detachment from death expected of an Epicurean.

In particular, he may be ignoring (and hence for the reader, alluding to) the
specific passage of Epicurus that contained Lucretius' model for 2.1 f. suave, mari
magno turbantibus aequora ventis, \ e terra magnum alterius spectare laborem (see
below on 7 and on 26.2). Sosylos, to whom Phil, addresses his grief, seems
to reply in the last distich (see on 7 ), reminding him that their deaths have
to be accepted: Well then, they played yesterday, today they are dead—from which
the message obviously to be extrapolated is that we should enjoy today's (simple)
pleasures such as those given in w. 1-4, for tomorrow we too may be dead; cf. Jacobs
(1794) 241, Gigante54. For the contrast between the pleasures of eating with death,
cf. Alkaios 38 Archil. 13 W2
id. 11
(for which thought
Archilochos was criticized by Plut. Quomodo Aud. Poet. 12, 33a-b). Archil. 13
should also be compared for the way it too moves from grief to a desire to put grief
aside; Archilochos seems to be conducting a conversation with himself. On Phil.'s
attitude toward Archilochus cf. V. De Falco, "Archiloco nei papiri ercolanesi,"
Aegyptusl (1922)287-296.
More particularly, Sosylos' response is equivalent to Epicurus' Vat. Sent. 66
cf. Sent. 40

. For a similar dialogue, in which Lucretius gives

himself the voice of reason, cf. Lucr. 3.904-911.
For recent work on Epicurean views of death, cf. D. Puliga, "X
in Epicure," Elenchos 4 (1983) 235-260; D. Furley, "Nothing to us?" in
M. Schofield, and G. Striker (eds.), Norms of Nature (Cambridge 1986) 75-91;
P. Mitsis, "Epicurus on death and the duration of life," BACAP4 (1989) 303-322;
C. Segal, Lucretius on Death and Anxiety (Princeton 1990).

1 : A flower of late spring, when it might be pleasant once again to

spend time along the seaside. The rose thus helps determine the time of year, but
since wreaths were made of roses it also has a place along with the list of edibles to
follow as one of the items to be gathered for the meal alluded to; cf. Asklepiades 25
HE (/IP5.181.2) ,26(5.185.5)
It may also subliminally prepare the way for the joint funeral of Antigenes and
Bakkhios in that it was placed on tombs during the Roman festival of Rosalia held
in May. Thus the judgement of Gow-Page that it would be "macabre" to associate
the rose of this poem with the Rosalia is misconceived.
With Phil.'s listing of the various blooms, which contrast with the now-dead
Antigenes and Backhios, cf. Meleager 31 (AP 5.144.1-4)
166 Epigram 29

Predicate of , in agreement only

with the nearest (Kaibel); cf., e.g., Aristoph. Ran. 36 , KG 1.39 f. This
way of establishing the season is familiar from Hesiod on; note esp. Hes. Op. 582
Alkaios 347.4 remains the
verb to be supplied in w. 3-4.
"No luxury" (Gow-Page), and yet when his Edenic society is criti-
cized by Glaukon for being Sokrates adds, among other things,
chickpeas(Rep. 372c), as well as salt, cheese, and greens. Kod omco, says Sokrate

That this description of an idealized

life could have served as a model for inhabitants of the Epicurean Garden is made
more likely by Glaukon's rejection of even these additions:
(For the association of Epicureans with pigs, cf. Cic. In Pis. 37, with
Nisbet'sn.,Hor.Ep. 1.4.16, Plut.Mor. 1091c, 1094a, Catullus47.1, above, p. 16 n.
13.) But even if this Republic passage is not alluded to, Gigante is correct to point
out that the food mentioned in this poem is entirely appropriate to the modest menu
of Epicureans. Contrast Phil.'s list of blooms with that of Meleager 31 (quoted above
onl ).

2 ... : Pi's epic but unmetrical is puzzling

Cf. Columella, Cult.Hort. 369 sediamprototomos tempus decidere caules, where, as
in Greek as well (LSJ s.v. Ill), the stalk by synecdoche stands for the whole
plant (cf. G. Kohl, Eng. kale). Phil.'s genitive therefore is not strictly necessary, but
BGU 1118.12 (i B.C.; adduced by Gigante 104 n. 54) and Automedon 5 GP (AP
11.319.4) ' ~ show that cabbage could be called
(cf. esp. Euboulos 6 K-A, adduced on 28.2 ),' (cf. 28.1), or
And once written, the genitive is just as likely to receive the adj.
as , since cabbage is cut at the stalk. Thus there is no reason to depart from
the MSS and follow Gow-Page in reading -10(101; cf. Gigante 104 n. 55. At worst,
this would be a very mild ex. of transferred epithet. Athen. 9.369e-370f collects
reference to cabbage in Greek literature. The ref. here is to the tenderness of young
plants; Automedon 7 GP (AP 11.325.1 f.) criticizes a host for serving "ten-day-old
yellow stalk of hemp-like cabbage."
So accented by PP1 here and again on v.6. (To be more precise, Pi
has , the mark over the upsilon indicating, as often in PI, that this is a proper
name.) Jacobs, followed by all subsequent editor, printed , which is (with
who knows what accuracy) found in Lucian Gall. 29. The MSS of Polyb. 3.20.5
record both and . Chandler, Greek Accentuation 280 f. records
numerous exceptions to the norm that trisyllables in - are paroxytone.

3 : £ ad loc. (printed by Stadtmuller) contains many guesses as to the mean-

ing both of this word (unsure whether it is a plant or a fish) and its accompanying
participle—and one that does neither: , which would link it
with the cheese of the preceding line. Gow-Page rightly fault the conjecture for
producing a barely metrical hapax; like Kaibel's conjecture, it looks to Homeric
 Epigram 29 167

, Both Kaibel and offer a milky , which assumes this noun to

stand for some sort of plant which exudes a whitish pith when cut. There seems,
however, to be no evidence for any plant, edible or otherwise, so called; for the fish
with this and related names, on the other hand, there is abundant evidence;
cf. D'A. W. Thompson, Glossary of Greek Fishes (London 1947) 153-155. Gow-
Page's concern that this fish was considered poor fare by the Romans is properly
answered by Gigante 57-58, who shows that this is in fact the point: humble fare
for a humble Epicurean repast.
A variant form of which can mean "shake, toss, roll"
(intrans.), often used of ships at sea. I am not sure what Dilthey meant bydepisce
palpitante—his conjecture may have been intended to describe the motion of the
fish when alive—but I take the phrase to depict the manner of preparation;
cf. "tossed (salad)," "saute." Small fish being fried in a shallow pan have to be
kept in motion to prevent their sticking together (as I have learned from my own
mistake).
Freshly set cheese will either have salt rubbed on it
(Vergil G. 3.403 parco sale contingunf) or be soaked in brine. In addition to im-
proving the flavor, salt decreases the cells' eutectic pressure and so reduces mois-
ture; it also retards bacterial growth. Cf. further Kroll, "Kase," RE 10 (1919) 1489-
1496.

4 Lettuce, also mentioned in BGU 1118-1112 (see above on

,).
This hapax is rightfully retained by Gigante, against most recent edd.
The outer edge of some curly lettuce is whiter and curlier than the rest of the leaf,
and can easily be called "foamlike." For Scaliger's conjectures, see Luck.

At this point in the poem the audience will assume that Phil, and Sosylos
alone are meant. With the next sequence, however, we realize that Phil has also
been thinking of Antigenes and Bakkhios.
Identified by Gigante as the high point and belved-
ere to the west of the Villa dei Papiri, i.e., Piso's villa, where Phil, and his Epicu-
rean friends met. The belvedere would have had a splendid view of the sea, an
as Gigante vividly describes, would have been a pleasant spot to partake of the
simple meal whose ingredients we have listed. This is indeed tempting; note,
however, that Herculaneum was situated on a promontory: Strabo 5.4.8

Similarly,
Seneca QN 6.1 ab altera parte Surrentinum Stabianumque litus, ab altera
Herculanense conveniunt et mare ex aperto reductum amoeno sinu cingunt, Sisenna
fr. 53 Peter quod oppidum tumulo in excelso loco propter mare. Strabo's
'H (perhaps the original full name of the settlement; cf. C. Waldstein
and L. Schoobridge, Herculaneum [London 1908] 89) obviously cannot have its
common sense of fortified military base; it more likely is simply a synonym of
"lookout"; cf. Aisch. Eum. 948 "city's watch-post."
168 Epigram 29

Phil, thus may be referring in to that part of Herculaneum that juts out
most prominently, and in to the view of the sea from that point;
Waldstein and Schoobridge 59 f.
For the pleasures of the seaside, cf. Nikainetos 4 HE, Cic. Gael. 35 accusatores
quidem libidines amores adulteria Baias actas convivia comissationes cantus symphonias
navigia iactant; AdFam. 9.6.4, actis et voluptatibus; Verr. 5.96.

7 : Of all the possible ways this combination is used, the most appropriate
here is "inceptive-responsive," when "a person who has been invited to speak ex-
presses by the particles his acceptance of the invitation" (Denniston, GP355); for
the difficulties involved with other interpretations, cf. Gow-Page 388, who reluc-
tantly settle for an unparalleled causal use. See also Del Re 129 f. Denniston notes
that the inceptive- responsive usage is "common in Aristophanes and Plato, and is
almost confined to them," which is simply another way of saying that a particular
usage is colloquial, as is entirely appropriate here.
The reply could be spoken only by Sosylos, to whom the preceding words have
been directed. Implicit is the message that the goods of the season are indeed to be
enjoyed, and today, before we too are dead; cf. Phil. De Morte IV col.37.23 ff.

Jacobs was on the right track when he suggested (1794)

that, if the preceding distich ended in a question, the last lines would provide the
answer: Ipsa vitae brevitas et rerum vicissitudines nos admonent, ut ne fruendi
opportunitatem nobis patiamur elabi (p. 241).
With Sosylos as the speaker of the last distich, the epigram is formally parallel
to 3, where the final distich is given over to Xanthippe. Sosylos' answer is also some-
what similar in tone to Xanthippe's, in that both give frank (i.e., properly Epicu-
rean parrhesiastic) answers intended to put an end to Phil.'s reverie, and which are
more in accord with Epicurean teaching than the romanticizing thoughts of Phil.
Cf. Epic. Ep. 3.124 f.
. Phil, comments on this and other Epicu-
rean passages concering death in De Morte IV coll. 1-2 et passim; cf. M. Gigante,
"L'Inizio del quarto libro Delia Morte di Filodemo," in RF2 127 ff.
To the lament of the preceding lines, Susylos could well have quoted Phil.'s
own words on the subject:

(DeMusica IV 6.13-18).
'Av ... : Most likely the same Antigenes as in 2.8
Mi v'Av AsGigante,57FC7 (1989) 136 observes, the mention of Muses
there strongly suggests that ov here = "write poetry"; i.e., that Antigenes and
hence Bakkhios too are poets. Cf. Phil. De Piet., P.Herc. 1428 col. 11.9 (Henrichs,
CErc 4 [1974] 21) v, of the poetry of Diagoras; Hedylos 6.4 HE (Athen.
11.473a) ;4P 11.134.1 (Loukillios)
; Cat. 50.1-5 hesterno, Licini, die otiosi . . . scribens versiculos uterque
nostrum ludebat. The verb in this sense is applied self-deprecatingly by poets to
slight examples of their art. Loukillios calls himself a . Presum-
ably, then, Antigenes and Bakkhios, like Phil, himself, regularly recited epigrams
 Epigram 30 169

at dinner. , although strictly contrasting with , allows us to imagine

that they did this (so Diibner; contra Gow-Page).

8 : A strong, largely atemporal, adversative. There is no redundacy with

; cf. Kaibel.
The vox propria for the laying out of the dead, usually associated
with an expression of grief, which is notably lacking here; cf. II. 24.786,
CEG 159 (Thasos, ca. 500B.C.), 795.7 (Thessaly,
ca. 335 B.C.). For bibliography on ekphora, see N. Richardson, The Iliad: A Com-
mentary, 6 (Cambridge 1993) 183.

API 234 [29 GP, p. xxvi K, 21 G]

PI 4a. 8.89, f. 49v caret P
3 scripsi: PI 4 Gow-Page:PI ' t e n t . Gow-Page:
PI 6 P I : H e c k e r

The stone contains a trinity of immortals: The head clearly reveals Pan the
goat-horned,
the chest and belly Herakles, and the rest, thighs and legs, has Hermes the
wing-footed obtained.
Refuse no longer, stranger, to sacrifice, for your one sacrifice will be
received by the three of us.

Beginning as a third-person narrative of the statue putatively above the inscription

(1-4), the poem addresses the notional passer-by with a request for a sacrifice/gift
(5), the one sacrifice being received by the three gods (6), each of whom, presum-
ably, will respond favorably. The wit of the poem lies in the fact that not until the
last word is it revealed that the poem has been spoken by the gods themselves, shame-
170 Epigram 30

lessly cadging sacrifices. It easy to imagine that Phil.'s Epicurean audience would
have appreciated this from their own point of view. Although Epicurus believed
in anthropomorphic gods ( ad KD I, Sext. Emp. AM 9.25 = Epic. fr. 353 U, Phil.
De Piet. 137-144 Obbink) and acquiesced in their public worship (Phil. De Piet.
653-657,737-740), he also argued that they take no part in human affairs (Ep. Her.
76-78, Lucr. 5.1161-1240, Phil. De Piet. 2032-2450), although they do have
the power to benefit us (Phil. De Piet. cols. 46 f. ed. Obbink). It may, however,
be doubted whether Epicurus would allow that a god could come in the form de-
picted here; cf. Cic. ND 1.46, spoken by the Epicurean Velleius,./lc de forma quidem
partim natura nos admonetpartim ratio docet. Nam a natura habemus omnes omnium
gentium speciem nullam aliam nisi humanam deorum. Moreover, the idea of three
gods speaking as one presents an amusing theological problem: "dans cette con-
clusion, n'y a-t-il pas une parodie du syncretisme religieux; economic pour le fidele!"
(Aubreton). Cf. Phil. De Piet., P.Herc. 1428, coll. 13.23-14.2 (Henrichs, CErc 4
[1974] 24) "it occurs to me to apply to them what Timocles said in his play The
Egyptian about the gods of that country: 'When those who commit impieties against
the acknowledged gods do not at once pay the penalties, whom would the altar of
a cat destroy?'" (fr. 1.2-4 K-A, tr. Obbink).
The literature on Epicurus' attitude towards the gods is immense, but several
recent works survey the many ways Epicurus retained traditional customs and
beliefs: B. Frischer, The Sculpted Word (Berkeley 1982); D. Obbink, "The atheism
of Epicurus," GRBS 30 (1989) 187-223, esp. 200 f. on the various religious activi-
ties in which Epicureans took part; J. Mansfeld, "Aspects of Epicurean theology,"
Mnemosyne, 4th ser. 46 (1993) 172-210.
Gow-Page point out that no triple statue of the sort described here is known,
although it is vaguely reminiscent of the Chimaira:
(II. 6.181 = Hes. Th. 323), which was parodied by Ariston of
Chios: (ap. D.L. 4.33 = 204
SH). Very likely Phil.'s figure is a poetic fiction.
Although an exact parallel for the statue is lacking, single individuals may be
similarly described. Cf. II. 2.477-479:

Rufinus 35 Page (AP5.94):

Note also Anacreontea 17, beginning in which

the image of Bathyllos is described detail by detail, some of them with a simple
 Epigram 30 171

reference to a god or hero, e.g., 32 f.

1 A poetic troika (LSJ s.v. II). Applied to gods by Soph. OT164 (Athena,
Artemis, Apollo), Eur. Hek. 645,Hel. 708, and TV., 924 (Athena, Hera, Aphrodite),
Meleager39 04P5.195.1) and74 (9.16.1) (Graces). Leonidas27 HE WP9.316.3)
refers to a Janiform statue as
"Trois immortels en ce marbre" (Aubreton), which is better than "the
stone has room for three immortals" (Gow-Page); my emphasis.

2 Cf. Bakch. fr. 14 That is, just as

the Lydian stone reveals the gold within (cf. LSJ s.v. ), so too the external
shape of the stone reveals the god within.

3 The same pair at II. 13.290

Herakles' front was of course notably muscular, but it cannot be said to be uniquely
recognizable.

3-4 If the MS's combination of plural and singular

("thighs and leg") is objectionable, as Gow-Page rightly point out, so too is their
own new combination of gen. and ace. It would be easy, however, for a scribe who
has not learned the lesson of Doric and , in ionicizing to
alter along with it (an error of homoioteleuton), especially since the error
produces an acceptable form with . It would now be all but inevitable for
which occurs 5x in Homer and once in H.Hemt., to be altered to genitive
plural. Phil, has a dual among plurals at 12.3. (Kaibel's objections, p. xxvi, to Phil.'s
authorship on the basis of the two Doric forms of w. 1 f. would also be met now
that all possible forms are seen as originally Doric.) Cf. P.Oxy. 2624 fr. 1.4 (387
SLG)

4 Elsewhere only Eustath. Od. 2.9 (of Perseus).

6 Cf. Pi.N. 8.22 fsc. 6 , where the violent sense

of the verb may, as Bury says, derive from medical terminology. Hecker, approb.
Diibner & Paton, compared API 253.5 f. (Anon.)
but, as Kaibel saw, the sense is clearly not the same: Artemis runs
to meet her cloud of incense, Phil.'s statue is stationary.
Epicurus prayed to the gods (Phil. De Piet. 790-797, 879-884 Obbink), but
believed that they, in sharp contrast with this triple deity's eagerness to receive
offerings, were entirely unaffected by outside (i.e., human) forces or concerns; see
intro. above and Obbink's nn. ad locc.
 172 Epigram 31

31

AP 11.318 [28 GP, p.xxvi K, 20 G]

P' Pl2b.4.1,f. 87rs.a.n.
2 P : P I 6 P : P I

Antikrates knew astronomy far better than Aratos, but he did not know
his own birth:
He said that he was in doubt whether he had been born under the sign of
Aries, Gemini, or Pisces.
But he has been found under all three, for he is a tupper, an effeminate
sex maniac, and an eater of dainties.
Maxwell-Stuart, Hermes 106 (1978) 253 f.

Stadtmiiller ap. Riess, "Antikrates (8)," RE 1 (1894) 2427 thought this epigram to
have been written by Antiphilos, but there is no reason to deny the ascription to
Phil. Rather, the relatively rare sense of = "testicles" (here in v.4 as it will
be reinterpreted on v. 6, and again in 22.6) argues for Philodemean authorship.
Phil, combines mockery of astrology with an attack against Antikrates, who
does not know his true sexual nature: Does he play the man with women and men
(Aries), does he play the woman with men (Gemini), or does he play the woman
with women (Pisces)? Cf. AP 11.160 (Loukillios):

For the reaction of philosophers to the claims of astrology, cf. A. Bouche-Leclercq,

L'astrologie grecque (Paris 1899) 570-609; A. A.Long, "Astrology: Arguments pro
and contra," in J. Barnes et al. (eds.), Science and Speculation: Studies in Hellenistic
Theory and Practice (Cambridge and Paris 1982) 165-192; W. & H. G. Gundel,
Astrologumena: Die astrologische Literatur in der Antike und ihre Geschichte
(Wiesbaden 1966) 180-189.
 Epigram 31 173

1 Presumably a real person, but otherwise unknown. It is tempting

to imagine that he was a Stoic, as several of this school were sympathetic to the
determinative aspects of astrology; Bouche-Leclercq 28 ff.—not, however, Panaitios;
cf. F.H. Sandbach, The Stoics (London 1975) 80. Epicureanism, of course, would
reject astrology because its theory of the swerve (quod fati foedera rumpat) would
guarantee that there is no fixed linkage between all the motions of the universe;
cf. Lucr. 2.251-262,277-293; Bailey on Lucr. 5.728. Tacitus Ann. 6.22.3 contrasts
Epicurean and Stoic attitudes towards astrology.
Equivalent to the following imperfects: This was the state of affairs
until now, when the truth has been discovered (v. 5).
Both Aratos and Eudoxos (see next note), as well as Euclid,
entitled their works Phil, uses a generic term which he probably knew
from the astronomical Sphairika of Theodosios (ii-i. B.C.). Earlier,
[sc. ] was applied solely to spherical geometry (e.g., Archytos B 1 DK, in a
sentence that refers as well to astromical knowledge). The usual terms,
and would not fit the meter; cf. O. Hultsch, " Astronomic," RE 2 (1896)
1829 f. for the terminology; and F. H. Cramer, Astrology in Roman Law and Poli-
tics (Philadelphia 1954) 3 f.
Earlier exx. of the use of mathematical topoi by Hellenistic poets are
Hermesianax fr. 7.85-88 Powell and Kallimachos, fr. 191.59-63, both mentioning
Pythagoras; cf. P. M. Eraser, Ptolemaic Alexandria (Oxford 1972) 1.407 f. For epi-
grams directed against astrologers, cf. F. J. Brecht, Motiv- und Typengeschichte des
griechischen Spottepigramms (Leipzig 1930) 41-45.
Aratos of course derived all of his star lore from Eudoxos (so
Hipparchos 1.1.5,et al.; cf. the testimonia in G. R. Mair'sLoeb edition, pp. 196 f.)
He was, however, naturally given full credit for knowledge of his material; cf.
Meleager 1 G4P4.1.49) He was also praised by Kallim. 56
HE = 27 Pf. (AP9.507) andLeonidas 101 HE (AP9.25). Ahorologicalmonument
from Tenos offers an interesting parallel to Phil: The star lore of Andronikos, the
builder of the Tower of the Winds in Athens (i B.C. ex.), is compared favorably to
that of Aratos (IG 12[5] 891.1-5):

Note that Aratos himself in one (of only two extant) of his epigrams compares one
man to another (text and interpretation obscure; cf. Gow-Page on Aratos 1 HE =
AP 12.129), as do Phil., Kallimachos (Aratos is compared with Hesiod), and the
stone from Tenos. Even Leonidas of Tarentum compares him to Zeus (
cf. the stone). Could this have been a common feature of Aratos'
epigrams?
On Andronikos, cf. Fabricius, RE 1 (1894) 2167 f. On Arams' reception by
Hellenistic poets, cf. P. Bing, "Aratus and his audiences," MD 31 (1993) 99-109.
174 Epigram 32

2 The configuration of stars and planets at the time of one's birth has
predictive value. Conversely, it would seem, one could try to deduce one's sign from
one's character; cf. Hor. O. 2.17.17 ff.:

seu Libra seu me Scorpios aspicit

formidulosus, pars violentior
natalis horae, seu tyrannus
Hesperiae Capricornus undae. ...

It is not that Horace is unaware of his own birthdate, but that he is (or pretends to
be) unsure how to characterize himself. Cf. R. Scarcia, "Orazio, Mecenate e le stelle,"
in AA.VV. L'astronomia a Roma nell'eta augustea (Galatina 1989) 34-53.

3-4 Parallels can be found for multiple alternate questions

(Soph. El. 539 ff. has three 's after a ), but perhaps here, after
the second f) is meant to come as a surprise in order to emphasize even more the
extent of Antikrates' ignorance. Note that v. 3, containing a masculine caesura
without a bucolic diaeresis, is Phil, 's only violation of Meyer's Third Law. With his
1%, compare 11% for early elegists and 4.5 for Hellenistic epigrammatists (with
none at all in Kallimachos's epigrams); M. L. Clarke, "The hexameter in Greek
elegiacs," CR5 (1955) 18; West, Greek Metre 197.

3 ... " (Is born) under the sign of"—a regular meaning of the
preposition in astrological writings (unnoticed by LSJ). In general, one may con-
sult O. Neugebauer and H.B. van Hoesen, Greek Horoscopes (Philadelphia 1959),
e.g., BGU 957, P.Oxy. 804, PSI1276 verso, etc. (many more examples may be found
through their glossary, p. 193); for a poetic example, cf. Dorotheos fr. 5 Stegemann
(p. 323 Pingree)

In Dorotheos and others the dative alone may appear (as, e.g., in Dorotheos frr. 6
and 79a [p.395]; P.Oxy. 596 col.2.4 f. See also
T. Barton, Ancient Astrology (London 1994), esp. 21-63.

4 The constellation Gemini, but also "testicles" and/or "ovaries"; cf.

22.6 n. Argentarios 7 GP G4P5.105) similarly puns on this word: Menophila's palate
cf. Gow-Page ad loc.) contains both Dog-star (= "penis"; cf. Henderson,
Maculate Muse 127) and Twins; such is her The inherent duality of
contributes to this amphiboly. Bisexuality as such (a dubious ancient category) is
not the point, but that Autikrates will take his sex in any and all ways.
The two fish of Pisces taken together (cf. Homer's
Aratos 548 ... ), and not two separate constella-
tions, as Maxwell-Stuart argues. First, the references are not to constellations a
such, but to the zodiacal signs in the ascendant at one's birth. The two Pisces
 Epigram 32 175

determine one such division; cf. D. R. Dicks, Early Greek Astronomy to Aristotle
(London 1970) 17. Moreover, for the three signs, there are but three correspond-
ing adjectives, not four; see below, on v.6 Note the many erotic puns on
various fishes in Antiphanes fr. 27 K-A; cf. Henderson 142.

5 For what it is worth - is never augmented to r|ij- in the Herculaneum

papyri; W. Cronert, Memoria Graeca Herculanensis (Leipzig 1903) 205.
As befits Antikrates' being born under the sign of the Ram; used
literally of stud animals (cf. LSJs.v.), and metaphorically of humans ([Archil.] 327.7,
328.7) and of Pan by Cornutus ND 27, p. 49; or perhaps not so metaphorically: cf.
the statuette from Herculaneum showing Pan making love to a goat, Naples, Mus.
Naz. Arch., Raccolta Pornografica, Inv. no. 27709. Cf. D. Bain, "Greek verbs for
animal intercourse used of human beings," Sileno 16 (1990) 253-261.

6 Often applied to sexual folly by Euripides: Ba. 644 (with Barrett's n.),
966 (Theseus:)
etc. That the word has this connotation here is guaranteed by the context and by
its accompanying adv.
This must be the right reading, not only because it gives us only
three terms after , as expected, but also because Phil, prefers not to depend
upon position to produce the long syllable before the caesura of the pentameter
(see on 20.2). According to figures provided by P. Maas, Greek Metre (Oxford 1962)
§22 and M. L.West, Greek Metre (Oxford 1982) 158, only Philip of Thessalonica
is as strict on this point as Phil.; cf. Intro., p. 44. For the combination of adv. + adj.,
cf. Phil. Rhet. 1151.7-8
Sexual delights come in various forms: Alexis uses oi|/ov to stand
for both vagina (fr. 168.6 f. K-A) and penis (fr. 50). As an explication of Antikrates'
link with Pisces, the former is to be understood here. For ancient expressions of
disapproval of cunnilingus, cf. Kaster on Suet. De Gramm. et rhet. 23.7.

32

AP 10.103 [24 GP, p. xxvi K, 17 G]

P PI la.88.5, f. 20v s.a.n. ii.21 iv.16
176 Epigram 32

2 PP1: 8pa%|if|v Gigante iam Jacobs) scripsi:

PP1: tent. Gow-Page: Lumb:
Hecker: Giangrande: tent. Diibner
3 Scaliger: P: PI

Don't keep looking into the butchershop [where you bought] before nor
enter it. Withdraw now to good tripe sausages for a drachma.
One fig too goes for a drachma, but if you wait, a thousand do. For
beggars time is a god.

Giangrande, KM 106 (1963) 255-257.

Lumb 87.

Difficulties in some details remain (see comm.), but the general sense is clear: Give
up what you can no longer afford (v. 1); take satisfaction from what you can now
afford (v. 2). Alternately, the object you cannot afford today may be practically free
later (figs; w. 3 f.). All a poor man can do is pray that in time he can get what he
wants (v. 4). There is nothing in these lines to suggest an erotic context (Lumb,
Giangrande), despite the well-known erotic connotation that CTUKOV can assume.
Unlike the case with real figs, the availability of vaginas is not seasonal. H. Herter
ap. Giangrande says that the point about the figs is that as a hetaira ages her price
drops, which is true enough; but who would advise a poor man to wait until a par-
ticular hetaira grows old rather than just, that very day, turn to an older one? (The
courtesan Phryne, on the other hand, charged more as she grew older
Plut. De Tuenda San. Praec. 125ab.)

1 Obelized by Gow-Page, but now all but guaranteed by , as no other

Greek word beginning - can fit, and is secure. Giangrande deduces from
the context that the required sense is "sacrificial cake," as in Pherekrates 247 K-A
= Phryn. Praep. Soph. 74.9 9"
cf. Hesych. s.v.
As Giangrande notes, such a cake, an offering to a god made with
oil and wine, would be expensive. None of the other, far more common, senses of
(altar, stage, theater, performance) fit here. For = song, missing from
LSJ, cf. Herodian, Partitioned 61.1 Boiss.
The problem with Giangrande's interpretation, however, is that, in order for
to be the object of the two verbs, the cake in question, learned allusion
that it already is, must actually refer to an expensive prostitute—a "tart" would be
the perfect English translation. Even if the obscene interpretation of this epigram
had not already been rejected on other grounds, Giangrande's construal of
with the two verbs is very strained and would have to be rejected in any case. Ear-
lier views may be closer to the truth, namely that Phil, is referring to a source of
expensive food (Jacobs), most likely meat. Diibner's explanation ("notare etiam
potuit popinam ubi sacrificiorum reliquiae venum exponebantur") is possible, but
makes better sense as the place where animals are butchered rather than
 Epigram 32 111

cooked; cf. the various euphemisms, especially , for the slaughtering of ani-
mals in sacrifice as detailed by W. Burkert, Homo Necans, ch. 1.1, Gr. Rel., ch. 2.1.1.
This interpretation assumes that = "butchershop," although now unattested,
was in common parlance.
"Enter" rather than "pass by" seems the appropriate meaning here.

2 "Elliptically, retire, withdraw" LSJ s.v. I 2 b, adducing Hdt. 5.126.1

T
M Xen. Hell. 1.1.34 />
Here, "withdraw from the expensive to the cheap."
Gigante reads , construing with "rendi una dracma al
banco del rivendugliolo," but, as the prefix indicates, when this verb means "ren-
der," it suggests something due, a debt of money or honor (LSJ s.v. Ill), which is
not appropriate here.
Editors have included within the corruption, but it goes
easily with , as in the Herodotos passage adduced. Sense demands that the
reference here is to some cheap food, perhaps a cheaper form of meat to be found
in a butchershop, especially since -suggests "intestines," and would
be an easy error for some form of "guts, tripe," which were regularly made
into sausages; cf. A minimal change adopting these leads
would be tripe sausage stuffed into intestines, the final -
of a strange word affecting the initial perhaps w i t h a t an intermediate
stage of the corruption. The point, not made explicit until the next distich, is that
if you cannot afford fresh meat, you should wait until the cheaper, and perhaps
less fresh, cuts are turned into sausage.

3 : A further example of the advantage of waiting, linked to

the first by the price: At the same moment that sausage is considered cheap for a
drachma, this same drachma will purchase only one fig. For another poem on the
variable value of figs, cf. Ananias 3 W2 (figs are more valuable than gold to starving
men); and for another expression of supply and demand, cf. Poseidippos III.3-4
B-G (speaking of common rock crystal):

Holford-Strevens is preferable.)

4 Proverbial in expression (like English "Beggars can't be choos-

ers"), with explanatory asyndeton; for other proverbs involving beggars, cf. Kallim.
fr. 724 Pf. Hes. Op. 26
At least since Pherekydes, on whom cf. H. S. Schibli,
Pherekydes ofSyros (Oxford 1990) 27 ff.; M. L. West, Early Greek Philosophy and
the Orient (Oxford 1971) 10-14. Also in Soph. El. 179
Eur. Hrkld. 900, and in Orphic literature; cf. also M. B. Galan, "Chronos," LIMC
3.1, pp. 276 ff.
178 Epigram 33

33

API.222 [26 GP, 21 K, 22 G]

P[C]Pl3a.ll.ll,f.34r Suda s.w. (1-2T (2) iv.18
U] T [C: P]
2T PP1 PI, i.e. nomen proprium): Reiske CPlSouda:
P 3 C (rasura): P: K Salm.
PP1: Scaliger, Salm. 5 Theiler (haesitanter sed recte):
r
PP1: vel Herwerden 6 CP1: P 8
CP1: P

Here lies the tender body of the delicate girl, here lies Trygonion, devotee
of feeble effeminates,
(she) through whom chapel and duma gained glory, to whom there was
playful chatter, whom the Mother of the Gods loved,
she who in a class by herself cherished the Cyprian rites of those all-around
women, and helped with Lais's love philtres.
O sacred dust, nourish around this philobacchic's stele not prickly shrub-
bery but tender buds of white violets.

K. Buresch, Aus Lydien (Leipzig 1898) 62-65.

Luck, Philologus 100 (1956) 271-285.
Paton, CR 30 (1916) 48.
Sider, AJP 103 (1982) 208-211.
White, LSCP 8 (1981) 173-175.
Wiseman, CQ 32-(1982) 475 f.

A mock grave epigram for Trygonion, a castrated Gallos; see comm. on vv. 1
2 etc. For the subject of castrati in general, cf.
A. D. Nock, "Eunuchs in ancient religion," ARW'23 (1925) 25-33 (= Collected
Papers [Oxford 1972] 7-15); for an survey of Galloi in literature, cf. R. Ellis's
introduction to Catullus 63; and in general H. Graillot, La Culte de Cybele Mere
des dieux a Rome et dans I'empire romain (BEFAR 107; Rome 1912), especially 287-
319; G. M. Sanders, "Gallos," RLAC8 (1972) 984-1034; T. P. Wiseman, Catullus
 Epigram 33 179

and His World (Cambridge 1985) 198-206. Horace, Sat. 1.2.120 f. suggests that
Phil, used the Greek equivalent of illam . . . Gallis, "the hell with her," in one of
his epigrams.
According to Lucian, De Dea Syria, our most extended ancient account, funeral
service for a Gallos had its special character: "His comrades carry him aloft to the
area just outside the city, place him along with his pallet on the ground and cover
all with stones. They then wait seven days before entering the sanctuary" (52). Phil,
refers to none of this, however. The corpse of Attis was occasionally a subject for
artists: LIMC s.v. Attis, nos. 325 f.; cf. S. Karwiese, "Der tote Attis," O.Jh. (1971)
50-62.
No doubt irrelevant to Phil.'s poem, but too not to quote, is the fol-
lowing anecdote told of Arkesilaos: "Someone had inquired why it was that pupils
from all the other schools went over to Epicurus, but converts were never made
from the Epicureans: 'Because men may become eunuchs, but a eunuch never
becomes a man,' was his answer" (D.L. 4.43, trans. Hicks).
On the subject of sepulchral epigrams, see R. Weisshaupl, Die Grabgedichte
der griechischen Anthologie (Vienna 1889; repr. 1987); R. Lattimore, Themes in
Greek and Latin Epitaphs (Urbana 1942); U. Ecker, Grabmal und Epigramm
(Stuttgart 1990).

1 The beginning of many grave epigrams: 16 in AP and 82 in GVI, a com-

mon formula being ( . . . ) proper name ( . . . ) which Peek,
GVI, classifies as Typus 14; see next note. Usually addressed to the passer-by, the
words may be impersonal (as here), those of the grave stone itself, or those of th
deceased. Cf. Weisshaupl, ch. 2, "Graberformen."
Perhaps because a proper name often follows some
editors, despite J's summary, took Tryphere as the name of the deceased and
as "a pet name for a girl" (LSJ). But, although T is indeed a com-
mon name (14 exx. inLGPN !-2,andinMeleager63 =AP5.154 and Asklepiades
26 HE = AP 5.185), in none of the parallels examined does the definite article
accompany the name, and here the name follows the second (as Pi's
indicates), which is repeated to enhance the pathos, as a repeated
does in Agathias 5 Viansino = AP 5.292. Although I do not find an exact parallel
for the phrase it is not unusual to characterize the deceased with an
adjective before s/he is named; e.g., Peek 371 GVI (Rome, ii-iii cent. =IG 14.1589)
M
As Paton has shown, the poem makes far better sense as a mock lament for a
castrato than as a maudlin epitaph for a real (i.e., biological) woman. Wiseman,
however, would keep open the possibility that, like the next poem in AP, this one
is also addressed to a dancing girl.
The semantic leap from "luxury," to "softness, wantoness, effeminacy"
is easy; cf. LSJ s.w. As Luck points out, this aspect of
Trygonion is repeated in and as well as in his/her name;
cf. also on Cf.Aristoph.Lyr.387f.
I Note also the worshiper of Kybele named
T below on 3 [Giangrande, Eranos 65 (1967) 41 f. does not con-
180 Epigram 33

vince me that ii K (AP 5.193 = Dioskorides 4 HE) is also a

pathicus.]
Here obviously "body," a substitute for the more usual or
of epitaphs. (An infrequent synecdoche: AP 7.3, 363.) Thus, II. 16.856
(quoted just below) was used on a tombstone; Kaibel, Ep. Gr. 243.5 f. The lemma
s.v. in LSJ has been almost completely rewritten in LSJ Suppl., although not, I think,
yet satisfactorily (at the very least, for "Lye. 1173 " [omitted in the revised supple-
ment] read "Lye. 1137"). The basic sense would seem to be "cheek" (in Homer
only in plural: II. 16.856 22.68
Aeolic said to be an epithet for t h e b y E A B T a d J / . 22.68
probably = "apple-cheeked"; cf. Theokritos(P)' 26.1). Since in
Classical Greek (Sappho[?], Soph., Eur., plus Theokr., Kallim., Ap.Rh., Lykophron)
the word came to mean "face," there was probably an intermediate stage = "mouth"
(cf. bucca > bocca), the shift from this meaning to "face" being paralleled by Lat.
os. This meaning is in fact given by Z ad II. 22.68b
and "mouth" does indeed make good sense of the Homeric plurals,
but I suspect this is no more than a guess made at a time when the true meaning of
the word had been lost. In post-classical Greek, the meaning "body" is found
(Theokr. 23.39, Phil.), which may derive from a misreading of Homer's usage found
in several scholia (cf. Z ad7/. 16.856
sim. ad 22.68) and lexica (Hesych. s.v. On
the other hand, a semantic shift from "face" to "body" is also possi\Ae;facies went
in the opposite direction (so Frisk). Note that Theokritos uses the word to mean
"face" in an Aeolic poem, "body" in an Ionic-dialect poem, and that "face" is said
by the scholia ad //. 16.856b to be a specifically Aeolic meaning. Cf. B. Snell,
EntdeckungdesGeistes, ch. 1 (pp.lOff. of Engl. tr.), andM. Leumann,Homerische
Worter (Basel 1950) 218-222, for a different analysis.

2 T A diminutive of the name Tpuycov (cf. F. Bechtel Historische

Personennamen 591); for its form cf. the two hetairai X (Lucian
Dial.Mefr. 10) and (Alkiphron 3.5), both of whom are again iden-
tified as hetairai by Eustathios //. 3.662. For the diminutive of "dove," as
an appropriate name for a Gallos, note Apul. Met. 8.26.4, where Galloi are called
palumbulae (pointed out by A. D. Nock apud Luck 274). Two qualities of the turtle
dove may have suggested the association: (i)
(Arist. HA 613al6), and (ii) turturum educatio supervacua est,
quoniam id genus in ornithone nee parit nee excudit (Columella 8.9). Luck points
out that turtle doves were sacred to Aphrodite and Demeter (both of whom have
been associated with Kybele; cf. L. Robert, /. Savants [1971] 91; see on v.5), but
this does not seem as immediately pertinent as the points just raised. (On the amo-
rousness of doves, cf. especially Prop. 2.15.27 f.) The similarity between doves and
pigeons may also be relevant, as the latter were the sacred bird of Galloi; cf. espe-
cially De Dea Syria 54; Thompson, Gk. Birds 244 ff. A Hellenistic terracotta shows
Attis riding a dove (LIMC s.v. Attis, no. 303). In the Near East, (turtle) doves could
stand for promiscuous women; cf. Job 43.14, Song of Songs 2.14, with (e.g.) Pope's
commentaries ad locc. in the Doubleday Bible Commentaries.
 Epigram 33 181

"Nerveless" is Gow-Page's translation, an expecially good one given

veijpov = penis, for this is in fact the sense of the Greek word. As I showed in greater
detail in AJP, the various ancient definitions for can
all be subsumed under the rubric of "smashing, breaking," especially in a Dionysian/
Sabazian context. Thus, Sabaktes is almost certainly the Smasher or Breaker, one
of the demons of the kiln "Homer" threatens to invoke if the potters refuse him
payment (Epigr. Horn. 14.9 = Hes. fr. 302 M-W; cf. M. J. Milne, "The poem en-
titled Kiln," in J. V. Noble, The Technique of Painted Attic Pottery [New York 1965]
102-113). TheSouda, glossing this poem of Phil., defines the word as
whereas Hesychios says that it = among the Chians, and glosses the corre-
sponding adv. as The seeming disparity between the
meanings offered by the two lexica is paralleled by their glosses of the Souda
again associating the word with Dionysos and Sabazios, Hesychios glossing the verb
as
The link between these various meanings may lie in the nature of Dionysos,
the god of sparagmos, who can receive the epithets (Aelian NA
12.34) and which, as M. Astour,He!lenoserm'tica (Leiden 1965) 191, point
out, probably meant Destroyer before the more benign sense of Deliverer came to
be understood. (Astour 188-193 also offers etymologies for Bassareus, Satyros, and
Bakchos which explain these names as Render, Striker, Killer, etc.) then,
recalls both Sabazios' role as an active destroyer and the particular destruction of
a castrated Gallos, a (Phrygian?) word that itself may mean "the cut one," < IE
*gwhol(cf. Lat. calvus); cf. A. H. Sayce, CR 42 (1928) 161 f.
For the association between Sabazios and Kybele, cf. Aristoph. Av. 873
ff. [ = finch (? cf. Dunbar on Av. 763), a pun on
is to be taken since
= ostrich']; Strabo 10.3.15, 18; Apul.Metf. 8.25, Eustath.Jw Od
1431.45 f. Cf. further S. E. Johnson, "The present state of Sabazios research,"
ANRW 17.3 (1984) 1583-1613, especially 1587 f. (S. and Dionysos) and 1600 f
(S. and Kybele).
Morphologically either (i) pi. (tantum) of "flower" (so
Brodaeus, Beckby), or (ii) the syncopated form of which appears in the
same sedes in Kallim. Ep. 14 HE (5.2 Pf.) and Theokr. Ep. 2 HE (13.2 Gow = A
6.340). Although Waltz would like exploit the ambiguity ("fleur des Salmakis
consacrees a Sabazios"), there is no parallel for this metaphorical, let alone plural,
use of although Cat. 63.64 gymnast. . . flos may be meant as a translation
Luck 275. Thus, especially given the Kallimachean and Theokritean parallels, sense
(ii) is far more likely, although the two earlier writers apply the word to inanimate
offerings (shell and statue, respectively) dedicated to gods—as does Phil. DeMus.
IV 19.1. The use of the word for people is common in NT as a translation of
"something sacred to a divinity," where, however, it is almost always found in the
unfavorable sense "accursed" (see Arndt-Gingrich s.v.). Even if he did not know
Hebrew (as his countryman Meleager did), Phil, may have been familiar with no
longer extant Greek religious texts where dvdGeua was used in either a neutral or
positive sense. The only other author who seems to use this word in the sense "devo-
tee" is Christodoros (iv-v c. A.D.) (AP 2.1.13 f.)
182 Epigram 33

A C: PPl. Paton (unlike Stadtmiiller and Beckby) prints

C's reading but translates "flower"; see above.]
Although, as Arndt-Gingrich s.v. report, NT texts often confuse this
word with I do not believe that Gow-Page are right to look to this latter
word for the sense they find in Phil., i.e., "adornment," as, e.g., Eur. fr. 518. 4-5
N2 (children,)
The spring Salmakis near Halikarnassos had the power, it was
s a i d t o emasculate those who drank from it: Strabo 14.656

Cf. also Ennius ap. Cic. Off. 1.61,

Ov. Met. 4.285 ff. et al. (amply quoted by Gow-Page).
then, must be an intentional hyperbole: almost "emasculated castrati." Buresch,
noting the similarity in meaning, thought that one acts as a gloss on the other as a
kind of hendiadys.

3-4 • For the anaphora of relative pronoun, cf. H.Dem. 481,

D. Fehling, Wiederholungsfiguren (Berlin 1969) 205 f. Like Catullus in 63 (but more
consistently), Phil, refers to a Gallos using the feminine gender, following what
probably was normal cult practice, as Hesych. K indicates. Cf. also
1030 PMG (quoted on v.4) and Cat. 63.12, 34 Gallae; AP 6.51.3
"A (a Gallos), Apul.Met. 8.26. But in this and the other Gallos poems referred
to below on v.5 they are not modified by feminine forms. It is, of
course, an insult for a man to refer to another as a woman, the locus classicus being
II. 2.235 'A copied by Vergil Aen. 9.617 O vere Phrygiae,
neque enim Phryges, but with an additional oblique reference to (Phrygian) Galloi;
cf. also Cat. 13.20; E. Maass, "Eunuchos und Verwandtes," RM74 (1925) 455^58;
Cic. ND 1.93, where Phil.'s teacher Zeno Chrysippum numquam nisi Chrysippam
vocabat (Pease ad loc. gives further parallels for calling men women, as does Fraenkel
onAg. 1625). Literary parallels apart, however, eunuchs usually wore women's cloth-
ing and "were commonly regarded as of the feminine gender"; cf. De Dea Syria 15
(after Rhea castrated Attis, he ceased his male way of life)
Souda s.v.
cf. De Dea Syria 27, 51; Nock 26.

3 The lack of definite article suggests that some special meaning is

attached to this word for "hut, cabin" (LSJ); probably a humble structure in or close
to a sacred area, like the inhabited by Polemon and others near the
Academy so that they would not have to live in the city (Phil. Index Acad., col. 14.39
Dorandi . and D.L. 4.19). Cf. GIG 4591 (Palestine)
[?]. Wiseman makes the attractive sug-
gestion that the mentioned by Phil, is the very one said by Josephos (A]
19.75,90) to have been located near the temple of the Great Mother on the Palatine
(although later on, it was assigned to other, more Roman, gods).
"Holy Assembly." Phil, provides the only sure instance of this
obscure word in a literary text. On the principle of einmal ist keinmal, it was often
emended away to (first independently by Scaliger and Salmasius), and hence
 Epigram 33 183

does not appear in LSJ until the Appendix of 1940. Its sense (and genuineness) is
clarified, as Buresch was the first to recognize, by a number of inscriptions from
Asia Minor: (a) No. 34 Buresch
(b) GIG 3439, iii A.D. from Maeonia,
(c)Ath.Mitt.35(1910)
144 (O. Walter), (d) Anat. St. 18 (1968) 75 no. 19 (A. S. Hall),
'E M O (O ?). GIG 3438, an inscrip-
tion related to (a) seems to use the term as a synonym, which sug-
gests that "(sacred) assembly" is an appropriate translation, (e) It almost certainly
appears again in a bilingual Phrygian-Greek inscription from Dorylaeum (Ath.Mitt.
23 [1898] 362 [MAMA 5.183]) as which P. Kretschmer Atb. Mitt. 25
(1900) 446 argues is a mistake for or and, to judge from the Greek
half, equivalent to (cf. GIG 4591, cited above on (f, g) The word
also appears in SEG 28 (1978) nos. 893 and 899 (both from Maeonia), funerary
inscriptions in which joins with relatives in honoring the deceased;
in 893 it would seem that a fellow worshiper in th was
(cf. ? ) , i n 8 9 9 ' A r e m i n d s us of in GIG 3438).
(h) An inscription from Thessalonica has 'A 'E cf. E.
Voutiras, ZPE 90 (1992) 87-96. [ was mistakenly restored to yet another
inscription; cf. SEG 28 (1978) no. 841.] See further O. Masson, "Le mot 8ot)|ioq
'confrerie' dans les textes et les inscriptions," Cahiers F. deSaussure4l (1987) 145-
152, who tentatively suggests that the word is Maeonian; J. Kolendo, Melanges
Leveque IV: Religion (Paris 1990) 245-249; SEG 40 (1990) no. 1737.
is cognate with Gothic doms, "judgment" (cf. Domesday Book) and
Slav, duma, "council" (and also with Lat. ab-domen; cf. Pokorny 2. dhe, p. 1.237).
Although it may (Masson thinks not) be cognate with "heap,"
entered Greek as a loanword from Phrygian; cf. A. Heubeck, Lydiaka (Erlangen
1959) 81 n. 101.1. M. Diakonoff and V. P. Neroznak,Phrygian (Delmar, NJ, 1985)
print the several Phrygian inscriptions to contain the word: A28 [= B-01 Brixhe-
Lejeune], A58 [the bilingual], C48; A24 contains the fern.adj. dumeja, "woman of
the dumas." Contra, O. Haas, Diephrygischen Sprachdenkmaler (Sofia 1966) 97 f.,
who takes as tumulo, which is closer in sense to Greek
Two literary passages have been emended from : Hipponax 30
Masson-West
(em. Masson,Rev.Phil. [1955] 289; approb. West, Degani), and
Hesychios s.v.
(em. Wackernagel), which Masson (1987) 147 argues derives from a
Hellenistic gloss on the Hipponax passage. The Hesychios passage, if rightly
emended, shows how the word can refer both to the material structure and to the
people assembled (cf. Fr. eglise etc. < Gr. Buresch suggests hendiadys
of
Not "simply an inversion" of verb and dative as Gow-Page and
Luck say, which translates weakly as "Trygonion is well suited to ..." or "der Laube
und Dumos wohl anstand Trygonion"; rather, between the praise entailed in
and M understand it rather as in LSJ s.v. 2 "to be con-
spicuous or famous"; cf. especially Pindar Pyth. 8.28 (Aigina)
184 Epigram 33

That is, just as Aegina was famous because of its citizens, Trygonion did not merely
fit into this group in some congenial way, he was its leading light; perhaps an exag-
geration, but in keeping with the usual hyperbole of tombstones. R. Keydell, rev.
of GP, Gnomon 43 (1971) 680, suggests that this is an example of having lost its
force in compounds, as was to be common in later Greek, but this is not necessar-
ily the case here.
Dat. of the possessor; not with (as edd. usually take it).
More than just "fond of play," at least in poetry, where it is
regularly applied to joyful dance and song: Od. 23.134 (the false wedding celebra-
tion after the slaughter of the suitors), Hes. fr. 123 M-W

and Aristoph. Ra. 333 (the chorus calling lakchos to his dance). The earlier sense
would seem to be somewhere between "sportive" and "ecstatic"; cf.
on a statue base of the 3rd or 2nd c. B.C.
from Thasos; G. Daux, BCH 50 (1920) 240. Similarly, Anacreontea 3.3
42. If. Arist. HA
629bl2 secularizes the term, using it of a playful lion (so too Aristoph. Gramm.
Epit. 2.144), but in Phil, the Dionysian overtones should not be lost. Applied to
chatter, the adjective conjures up a lot of arm waving and moving about. [There
may be a distant echo of Phil, in Pollux 5.161, a typically compendious paragraph,
where and other adjectives are said to be used of and other
activities.]

4 " chatter, gossip," whose stem shows up in poetry mostly in Aristophanes

(8x), but also in a dialogue poem of Theokritos (5.79).
The Great or Mountain Mother, Kybele, associated with
(among other things) the emasculated Galloi; two Phrygian inscriptions address
her e&MatarKubeleja. Note that the galliambic measure was also called
As Hephaest. Ench. 12.3 tells us, it was often used by "the newer" poets in works
concerned with the Mother of the Gods; as an example he cites Kallim.(?) fr. 761
Pf. (= 1030 PMG). The literature on
this Phrygian goddess is large; cf. W. Burkert, Greek Religion 176-179, ch. Ill 3.4,
with references to earlier literature;}. Bremmer, "The legend of Cybele's arrival in
Rome," in M. J. Vermaseren (ed.), Studies in Hellenistic Religions (Leiden 1979)
9-22. The Phrygian evidence suggests that Kubeleja originally meant "of (Mt.)
Kubelon" (cf. e.g. i.e., a specific cf. C. Brixhe,
"Le nom de Cybele," Sprache 25 (1979) 40-45.

5 "In a class by herself"; Gow-Page refer to Jebb on OT 299. Perhaps cf.

Parmenides B 8.4
When this word appears in epitaphs it is usually applies to the
deceased, who in life was loved by one or another of his relatives; but cf. 708 GVI
 Epigram 11 185

= 212 Kaibel (Syria, i cent. A.D.)

For goddesses of fertility being worshiped by the infertile (who could be prepu-
bescent youths or the very old), cf. Nock 28 f.
For Aphrodite's connection with Kybele, cf. Charon 262 F 5
FGrHist; Eur. Hel. 1341-1357, where the chorus links the worship of Aphrodite,
Demeter, and the Mountain Mother. Cf. also Phil. De Piet. fr. 3.11-14 Schober (CErc
18 [1988] 109)
•UTtdpxfew]. Cf. further D. Obbink, "A quotation of the Derveni papyrus in
Philodemus' On Piety," CErc 24 (1994) 111-135, esp. 114 ff.
Paleographically an undemanding change from the MSS'
But, although prep. + noun would be unobjectionable in itself (cf. Pi., P.
9.105 f. Aisch.ylg.62f.
it is hard to understand whatorgia "for the sake
of women" could mean: one either partakes in these activities or not; they are not
done for anybody else's sake. The words cannot mean "concerning women" (so White,
alone among recent commentators to retain the MSS). Dilthey, Observationes Criticae
in Anth. Graecam (Gottingen 1878) 11 f. (approb. Kaibel, Luck) takes with
but only rarely does this preposition follow its noun, in all instances the
ace. cf. Hes. Th. 554 with West's n.; see further Gow-Page ad loc. Herwerden,
followed by Paton (approb. Gow-Page), suggested the very adj. applied
to a Gallos by "Simon." 59 FGE (AP 6.217.9; comm. in HE 2.517 £): [sc.
K , but this is unnecessary. It can probably also be shown to be wrong,
for in all the six Gallos poems in the Anthology, the authors seem, in a fashion typical
of the Anthology, to make sure never to copy one another in the way they describe
the Gallos' unusual sexual category. Erykios calls him "cut when young,"
a hapax (6.234). Alkaios: (6.218). Antipater:
codd.), a rare word for eunuch (6.219). Dioskorides: i.e., keeper
of Kybele's underground chamber (where the castrations took place; 6.220). Since
all these authors were in Meleager's collection, Phil, would certainly feel called upon
to maintain the tradition and find yet another descriptive epithet for Galloi. Theiler's
solution, therefore, not even really a conjecture, is hard to resist: Trygonion, as in w.
2 and 3, is singled out from his peers, this time for his activity during the rites of—
i.e., those carried out by—the people (now) women in front and back. And without
Phil.'s poem would lack the reference to castration which all others
include. White's objection that the adjective cannot mean "hinten und vorne
weiblich," on the grounds that "the Gallus was not 'hinten weiblich,' because hu-
man posteriors are common to both sexes" is in effect countered by the Anthology
itself where more than one poet claims that from the rear one sex can substitute for
the other, which is all need mean, e.g., Argentarius 10 (AP5.116.5 f):

Cf. also 5.49 (Gallos), where a man who enters the woman from behind is called
Mart. HAJ>.l2tequeputacunnos, uxor, habere duos (LZ, vagina and anus);
i.e., she is, like Trygonion, on both sides. Note also which of course
186 Epigram 33

means only that for all intents and purposes (but not literally) one has two right
hands. Cf. also Emped. 31 B 61.1 DK
J. Wackernagel, Vorlesungen iiber Syntax (Basel 1926) 2.159.
For two genitives each with its own its own relationship to the same noun,
cf. PL Lg. 665b KGl.337An.4.

6 Cf. GIL VI, 30780 = CCCA III 237 (2nd half iv c. A.D.), an altar honoring
two holy men:
"Having prepared the love charms/potions."
can be potions or verbal spells (cf. Barrett on Eur. Hipp. 509), and could
also apply either to a drink (cf. Od. 10.379, take food or drink for oneself] or to a
primarily verbal activity.
The name of the famous courtesan could be used to stand for sexual
activity in general; best seen in Agathias 54 Viansino = AP 5.302.19 £, where
Diogenes the Cynic, rejecting the possibility of sex with one type of woman after
another, only one of whom is a courtesan, prefers masturbation:

7-8: Praying for the appropriate flowers to grow at the grave is a common Helle-
nistic motif; Luck 279-282 collects literary and inscriptional exx.

7 The upsilon of this word often appears short in verse before a vowel, as in
the inscription quoted in the next lemma (LSJ s.v.), but is properly long in the
present system.
Trygonion is buried in sacred ground, perhaps because Galloi,
serving in some sort of priestly capacity, were considered pure, perfect, sage, etc.;
Graillot 288 f., 294. Unusually, instead of being addressed to the passer-by, the
epitaph turns out to be an appeal to the earth covering the corpse to produce the
proper floral covering. For parallels, cf. Kaibel, EG 569.5 f.
222b, p.x, 11 f. Krinagoras reverses the prayer:
(AP 1.401.7 f. = 41 GP). For other exx.
ofvocative seeAP7.315 (quoted in the next lemma), 632.5 (Diodoros7 GP),
708.1 (Dioskorides24HE).

8 Brambles are obviously inappropriate for the delicate Trygonion (see next
lemma). S. L. Taran, ]HS 105 (1985) 91 f. cites some epigrams where "brambles"
stand for the rough hair of the longer sexually desirable young man. For an inver-
sion of the usual prayer, cf. AP 7.315.1-2 (Zenodotos or Rhianos), spoken by the
corpse of Timon the misanthrope,
Prop. 4.5.1 terra tuum spinis
obducat, lena, sepulcrum.
More than mere synecdoche of course; the ten-
der cups of the flowers are to remind the passerby of the equally tender Trygonion.
Cf. CIG 5759 = Kaibel, EG 547a.l-6, an address to the dead man to produce fine
flowers, especially v.4 see further
Lattimore 129-131, who also cites the inscriptions noted in the preceding lemma.
 Epigram 34 187

For the placing of violets on tombs, cf. A. B. Cook, "Iostephanus,"/HS20 (1900)

1-13, who demonstrates the connection between violets and Attis and Persephone.
His reference to GIG 6789 = 548 Kaibel is particularly pertinent:

(other delicate flowers are listed).

34

AP 6.349 [19 GP, 24 K, 16 G]

P iv.19 caret PI
1 P: Ap.L (mars.) P: Reiske: D'Orville:
Kaibel 3K- P: Gisante: F. W.Schmidt 6
P: D'Orville Kaibel iam D'Orville): P: An.
L(marg.), Leid. B.P.G. 34(marg.) C: P

Melikertes son of Ino, Leukothea the grey ruler of the open sea and divine
averter of troubles,
choruses of Nereids, Waves, and you Poseidon, and Thracian Zephyros
the gentlest of the Winds,
graciously may you bear me safely across a calm sea in my flight to the sweet
shore of Peiraeus.

Giangrande, GB 7 (1978) 77 f.
Gigante, Philodemus in Italy 49-52.
Hopkinson79, 271.
Kaibel, Hermes 15 (1880) 460 f.

Gigante argues that this carefully crafted literary prayer, with its complete lack of
irony in addressing the gods, could have been written only before Phil.'s arrival in
Athens from Gadara, since in Athens he would have learned from Zeno that the
gods are unreceptive to prayers (sim. Dorandi, CP 90 [1995] 175). This may be so
188 Epigram 34

(cf. 30, which reflects a more Epicurean idea towards prayer), but it should also be
said that (i) we do not in fact know Phil.'s philosophical allegiance or leanings
before he arrived in Athens. There were centers of Epicureanism in his part of the
world; cf. W. Cronert, "Die Epikureer in Syrien," Jahresb. d. Arch. Inst. in Wien 10
(1907) 145-152. And (ii) the reference to Thracian Zephyros (v.4) may suggest a
point of origin other than Gadara. In sum, this poem cannot be relied upon to pro-
vide unambiguous autobiographical statements.
One wonders, moreover, why, if Gigante is right about the strictness with which
Phil, would compose a prayer, he would hold on to a poem so inconsistent with his
later views. Since for most practitioners of the art, the epigram is, by design, the
most ephemeral and occasional of poetic genres, an epigram no longer to the author's
liking would be quietly discarded. If Phil, kept this poem over a period of years, he
would also be capable of writing it at any time during that period, especially since
Epicureans in fact did allow themselves to partake in prayers to the gods. There is,
furthermore, no reason to regard this poem as in any way autobiographical; it may,
for all we know, be written in the persona of the sort of people who Phil, in
De Morte says are deserving of criticism for risking their lives in pursuit of profit:

(col. 33.25-30 Kuiper).

Phil.'s poem, in other words, need be no more than an exercise in a common
topos; cf., e.g., Prop. 2.26, wherein is described a similar prayer for safety at sea,
containing invocations to, among others, Neptune, Leucothoe, and Nereids. Even
more telling is Prop. 3.21, another journey to Athens, where echoes from this poem
of Phil, lead to an Epicurean goal (which as we have just said may not be true of
Phil.'s poem): Propertius plans a trip from Rome in order to escape fmmgraviamore:
On board, cogar et undisonos nuncprece adire deos; and, inde ubi Piraei capient me
litora portus, I will seek solace either in the Academy or in hortis, docte Epicure,
tuis. Indeed, so reminiscent is Propertius of our poem that one wonders whether in
it too Phil, was in flight from a gravis amor, looking to Athens (real or metaphori-
cal) for philosophical solace. This could have been clear to Phil.'s audience as they
heard or read this poem in conjunction with others in a series on the same theme,
e.g., 8.

1-2 Ino, maddened by Dionysos, threw her-

self and her son Melikertes into the sea, whereupon they became known as
Leukothea and Palaimon, deities now upon whom storm-tossed sailors could call;
cf. esp. Orphic Hymns 74-75 and Frazer's note to his translation of Apollod. 3.4.3
for other ancient sources. For the Semitic origins of this tale, cf. M. Astour,
Hellenosemitica (Leiden 1965) 204-212. There is no discrepancy in the narrator's
calling the son by his land-name and the mother by her sea-name, and it may be
regarded as a Hellenistic nicety that Melikertes strictly speaking is the son of Ino
rather than Leukothea. Cf. Prop. 2.28.19 f.

Ino etiam prima terris aetate vagata est:

hanc miser implorat navita Leucothoen.
 Epigram 34 18

1 Reiske is followed by, e.g., Page (OCT) and Hopkinson, but too many
sea deities are called for us to deny it to Ino here: In addition to Glaukos
himself there are Thetis (Ap. Rh. fr. 12.15 Powell; Parthenios 2), Galaneia (Eur.
Hel. 1457), Amphitrite (Theokr. 21.55), Nereids (id. 7.59), and Triton (Leonides
Alex. 12 FGE = API.550); cf. NonnosD. 42.108 (aNaiad)
eiske's reading would provide an adj. for the otherwise bare but is not
ompelling.
Cf. Alkman 50(b) PMG The simplex is
common in hymns; cf. K. Keyssner, Gottesvorstellung undLebensauffassung im gr.
Hymnus (Stuttgart 1932) 75 ff.

2 Although their violent deaths disqualify Ino and Melikertes

from inclusion in Hesiod's Golden Race, his description of them at Op. 121 ff.
offers a generally valid attitude towards helpful divinities:

(The text, however, is vexed; cf. West ad loc. I quote from PI. Crat. 397e-398a.)
Cf. further Keyssner 107-113 f.

3 Several of the words of this poem recall the names of Nereids, as given
by Hes. 77>.240ff., along with their power to calm the sea: (243),
(256); (256);
(243) ; and KiJ|ia ~ several names beginning with esp. 252 ff.

; Not a metaphor, as, e.g., Phil. Rhet. 1.236.5 f. Sudh.

but a literal description of the Nereids, who are often described or pictured
as dancers. Cf. Eur. Tro. 2 N
7T427 N Ion 1080-1084, IA 1054-1057, An dr. 1267; Bakch.
17.101-108; Philostr. Imag. 2.8; Nonnos 48.192-194. See further J. M. Barringer,
Divine Escorts: Nereids in Archaic and Classical Greek Art (Ann Arbor 1995), 83-
89, who, in addition to supplying the literary references just given, has identified
several groups of dancing Nereids on Greek vases. (Wilamowitz S&S 169 conjec-
tured in Simon. 579.3 PMG.)
Kept lowercase by modern editors and taken as parallel to with
N , but "waves of Nereids" is unparalled, whereas Waves appear as gods in
Arternidoros 2.34, and, as was said in the introduction, Phil, seems intent on get-
ting as many sea gods as possible into this poem. Note also that Prometheus' invo-
cation of semipersonified cosmic elements includes
PV 89f. Gigante, Epigrammi scelti and Philodemus in Italy 51, conjee-
190 Epigram 34

tures ("cori ed onde delle Nereidi"), comparing Paulus Sil. 32 Viansino

C4P9.663.4), and Eur. IA166 and ar-
guing that the paradosis is suspect in view of the following but an
address to Waves would not be amiss in a prayer for a calm voyage. Schmidt's con-
jecture, designed to meet the same objection, is far less likely. (Neuter deities oc-
cur elsewhere; e.g., in 19 and in PV, both of whom vase painters
freely portray as males.)

4 Gow-Page wonder why the gentle western

wind of Zephyros should be associated with the north, whence blows harsh Boreas.
They suggest a "misapplied" literary reference to //. 9.5
. But, although Zephyros can indeed be the gentlest of winds,
as Theophrastos, De Vends 38 says it is also,
as Theophrastos goes on to spell out (cc. 38-45), associated with destructive cold
weather and storms, especially in early spring and late fall. And a wind can come
from the west wherever one is located; Theophrastos mentions the west wind in
Thessaly, the Malic Gulf, etc. A traveller from Macedonia to Athens (say, Piso) coul
more specifically call upon a zephyros which came from Thrace to start him on his
journey. In this case, invoking the needed wind as gentle may be an instance of
captatio benevolentiae. If this poem were in fact written for Piso's departure from
Macedonia in 55 B.C., Phil, could have been with him (so, e.g., Cichorius, Rom.Stud.
295; G. Bowersock, Augustus and the Greek World [Oxford 1965] 3), or, equally
possible, it could have been included in a letter to Piso from Phil., who had
remained in Italy.
The phrase dveucov rcprimate Z is copied from Dioskorides 11 HE (AP
12.170.2). For the comparative of this adj., cf. 18.5. For Zephyros in art and myth,
cf. K. Neuser, Anemoi: Studien zur Darstellung der Winde und Windgottheiten in
derAntike (Rome 1982) 119-142.

5 For use of this adj. and its related verb in prayers, cf. Keyssner 91-93.
Most applicable to a wind (cf. Od. 10.25 f.
but as the plural shows,
it is meant to apply to all the divinities, who see to it that he is carried safely by his
ship (cf. Od. 16.322 f.).
Etymologically an oxymoron, but here, as part of a prayer for
smooth sailing, probably better taken as a proleptic adj.: "a wave that is to be flat-
tened." Gaetulicusy4P5.17.3, another prayer to Ino, imitative of 34, contains the
phrase

6 Travelers by sea had good reason to appeal to sea deities in their capac-
ity as cf. Hom.H. 22.5 (Poseidon), Orph.H. 74.4 (Leukothea), 75.7
(Palaimon = Melikertes). D'Orville's (approb. Brunck, Reiske) is altogether
unnecessary.
Normally The former appears as in
Aristoph. Pax 145, and for the scansion here, cf. IG II2 12476/7.3 (ca. 150 A.D.)
L. Threatte, Grammar of Attic Inscriptions 1 (Berlin 1980) 213.
 Epigram 35 191

35

AP 6.246 [Argentarius 18 GP = 15 Small]

P PI 6.5, f. 61v Suda s.w. (i),
(2), (5-6)
3 PP1: Salm.: Brodaeus: Jacobs:
Boissonade: Stadtmiiller 3-4 post 6 (iam tent. Stadtmiiller)
traiec. Beckby, Waltz 6 PP1: Suda 7 PI
P

Course-chasing spurs, nostril-hugging muzzle, tooth-bearing breastpiece,

and willow wand(?)—these Charmos has dedicated in your porch,
Poseidon, because of his Isthmian victory—
and the horses' curry comb and whip for their rumps, bold mother of
whirring sound.
So then, O dark-haired one, receive these, and crown the son of Lukinos
also for a great Olympian contest.

Small, YCS 12 (1951) 103, 121 f.

A dedicatory epigram, of which sort Argentarius offers two other examples (17 GP
= AP 6.201, a dedication to Artemis after childbirth; 23 = 6.248, dedication of a
flask to Aphrodite); Phil, offers none other (nor does any incipit in II seem to be-
gin a dedication). The external evidence for authorship is indecisive, the combined
weight of PP1 favoring Phil., and the lack of an incipit in O somewhat telling against
him. It sits well within a Philippan run of epigrams (6.227-261), which of course
does nothing to settle a dispute between two of Philip's authors. Although Gow-
Page "cannot resolve the doubt about 6.246" (GP 2.166), they include it in
Argentarius because "there is nothing similar in Philodemus, but Argentarius
occasionally composes in this style" (GP2.371). The style, however, belongs to the
dedicatory genre rather than to the author, and had Phil, written one such no longer
extant it would doubtless conform to type. Our small Philodemean corpus con-
192 Epigram 35

tains other poems without parallels, such as the mock epitaph of 33 and the prayer
of 34. Beckby merely says "wohl Argentarius." The poem is also claimed for
Argentarius by Small 121 on grounds similar to those of Gow-Page (Small in addi-
tion to the two poems mentioned above adduces Argentarius 24 =AP9.229, a simi-
larly adjective-filled ode to a flagon).
Gow GA 30-40 surveys the epigrams with alternative ascriptions, of which
there are many in Meleager's, Philip's, and Agathias' collections. Of the various
causes for this phenomenon, the most likely at work here is "erudition and con-
jecture"—erudition if Argentarius is the true author, conjecture if Phil. is. Small
and Gow-Page may well be right, but certainty is not possible. (In any case,
Stadtmiiller's unsupported suggestion ?," in app. crit., need not be
seriously entertained.)
The vast majority of dedicatory epigrams in the Anthology are for objects
maritime, martial, cosmetic, agricultural, musical, etc. which, often now decrepit,
are offered up (and gotten rid of) in thanks. Victory offerings (not always the prizes;
cf. below on 1 are 6.49, 100, 149, 213, 233, 259, 292, 311, 350, of which
the closest to Phil./Argentarius is Maccius 8 GP (AP 6.233):

For such offerings in general, see W. H. D. Rouse, Greek Votive Offerings (Cam-
bridge 1902), ch. 4, "Games and contests," pp. 163-186. Dedicatory inscriptions:
M. Lazzarini, Le formule delle dediche votive nella Grecia arcaica. Mem. dell' Ace.
Naz. dei Lincei 19.2 (1976). Inscriptions celebrating athletic victories: J. Ebert,
Griechische Epigramme auf Sieger an gymnischen undhippischenAgonen. Abh. der
Sachs, Ak. der Wiss. zu Leipzig. Phil.-hist. Kl. 63.2 (Berlin 1972). For equine gear
and horsemanship in general, see J. K. Anderson, Ancient Greek Horsemanship
(Berkeley 1961); P. Vigneron, Le Cheval dans I'antiquite greco-romaine (Nancy
1968); S. Georgoudi, Des chevaux et des bceufs dans le monde grec (Athens and Paris
1990); D. G. Kyle, "The Panathenaic games: Sacred and civic athletics," in J. Neils
(ed.), Goddess and Palis (Princeton 1993) 77-101.
An oddity of this poem is that the contest is not named (as is the case elsewhere
in the Anthology and on inscriptions), which suggests that it is an exercise in the
dedicatory topos rather than a poem written for a real occasion. Perhaps an origi-
nal poem in honor of the son of a certain Lykinos of Sparta who won with horses at
Olympia in 384 B.C. (Paus. 6.2.2) provided a model, as it was not uncommon for
victories to run in families; cf., e.g., Paus. 6.1.6 (Kyniska and her family), 6.2.1 f.,
6.2.8, 6.7.1-7 (Diagoras and his family), 6.7.8.
 Epigram 35 193

1 Although this word when applied (literally) to horses usually means either
"crops" or "goads," (LSJ s.v. 1, Anderson 205 n3Q),Souda
and Gow-Page are very likely right to take it here
to mean "spurs." First, two other horse-goads are mentioned below. Second, the
Maccius poem given above, which either imitates or is imitated by Phil./Argentarius,
(i) again distinguishes from t h e a n d (ii) applies to the former th
descriptive phrase which more certainly can refer to Spurs; cf.
Asklepiades 6 HE (y4P5.203.l-2)

and note also Xen. Eq. 8.5, where, Anderson ibid, argues
refers to spurs. Spurs can be seen on the boy jockey from Artemision, a
Hellenistic bronze of the third or second century; c£, e.g., D. Finn and C. Houser
Greek Monumental Bronze Sculpture (New York 1983) 89 f.
Since spurs are of no use to a charioteer, the race (if indeed the author thought
about the matter) must have been of mounted riders (see introduction above,fin.).
Rouse 151 divides athletic offerings into (i) prizes, (ii) instruments, and (iii)
ther commemorative offerings, of which Charmos here offers (ii).
One of several in this poem (Graefe conjec-
tured it at Nonnus D. 5.233); the others are: 1 2
5 A similar accumulation of hapaxes may be found in 3; Small 83 lists
and in Argentarius. This word, Pi. P. 9.4 (etc.) and
Schol. in Oppian. Hal. 1.140 (swiftly moving fish; not in LSJ) are the
only Greek words compounded with and
also occur as proper names.)
"nostrils," whether of humans or animals; often of
horses in Hippiatr. 21
Muzzles were not used during races, but were used during training
and grooming; cf. Xen. Eq. 5.3
Maccius 8.1-2 (above), D. Cahn, in D. von Bothmer (ed.), Glories of the
Past: Ancient Art from the S. White andL. Levy Collection (New York 1990) 12
A can also be a b i t o r hackamore, but its epithet here, "nostril-
loving," fits better with muzzle. Note, however, that bits often have external lateral
branches which extend as far as the nostrils. In general, cf. Vigneron ch. 2, "Le
harnais de tete," pp. 51-79, with pis. 14-28.

2 Waltz, translating "dentelee," followed by Gow-Page, adduces

Daremberg-Saglio s.v. Ephippium, fig. 2686 (third quarter of sixth cent., Clazomenae
CVA Great Britain 13, BM fasc.8 [1954], pi. 585, fig.l = plate 17 Anderson), ahors
wearing a neck collar from which depend objects which I doubt are meant to be
taken as teeth. Anderson more cautiously says "the collar round the horse's neck is
purely ornamental, unless the discs hanging from it are charms to avert bad luck"
(caption to pl.17). R. M. Cook, the author of the CVA fascicle, says merely that the
trappings are "of Oriental—and perhaps specifically Assyrian—style" (pi. 18); sim-
larly decorated horse collars are to be seen on other Clazomenian vases and
sarcophagi illustrated here. Without any surer visual evidence for teeth, we can
194 Epigram 35

probably do no better than follow Jacobs (1799): "Fuisse videtur lorum, dentibus
distinctum, pectora equi ambiens, quale ornamentum et hodie equis adhiberi solet";
sim. LSJ.

3-4: There is no doubt that other dedicatory poems offer unbroken lists of offer-
ings, but this is in itself no reason to follow Stadtmiiller et al. in transposing these
lines in order to make this poem conform to what is often a very tedious type.

3 Since cruivov (sc. appears in three codd. vett. of Xen. An.

4.4.13 (v.l. which edd. prefer), the reference here may well be to a leather
switch (cf., e.g., fig. 42 Neils). Editors, however, aware of the many representation
n art of a driver or rider with willow wand in hand, prefer the easy emendation to
(K) cf., e.g., fig. 41 Neils. This poem, like Xen. Eq. 8.4
distinguishes between the whip and the switch. The latter, like the muzzle and comb,
would be used during training and grooming.
Specially commissioned items, such as statues, would be set up
in their place; smaller items, such as masks, could be hung from the wall; offerings
in between could, as here, be left in the portico of the temple; cf. e.g. AP 5.202
(Asklepiades 35 HE = Poseidippos 24 Page OCT, a hetaira dedicates her leather
gear to Aphrodite), 6.24.2 (Anon.), 114.2 (Samios 2 HE), 143.2 (Ps.Anacreon 14
FGE), 178.2 (Hegesippos 2 HE), 211.7 (Leonidas 2 HE), 254.8 (Murinos 2 GP)
297.5 (Phanias 4 HE). Afterwards, temple attendants would enter the objects in
the temple inventory and then either store or bury them; Rouse ch.13.

4 This pronoun alone may be enough to mark this poem as a literary exercise,
for although the formula is very common (cf. Lazzari
181-207), inscriptions address the (human) reader who is standing before the of-
fering or statue. On the other hand, cf.AP 6.9.1, 16.1,21.9, 36.1 etc., which are o
course literary examples.
Cf. Pi. I. 8.4 Maccius 8.7-
8
21 exx. of this name in LGPN 1-2.

5 " Cf. Eur. Hipp. 1174 Maccius

8.6, Anderson 95 f. See Xen. Eq. 5.5-10 for advice on grooming.
lit. "hair-drawing"; perhaps constructed with the epic
in mind (II. 15.354, 16.370, Hes. Sc. 369).
A single rider swings the whip backwards and to the side, striking a
flank; a charioteer still strikes the horse's hindquarters but can hit either flank or
the back; cf., e.g., figs. 41, 42, 44 Neils. For in this sense, cf. Od. 4.65
(Menelaos prepares for eating)

6 can describe the rushing sound made by an

object as it moves through the air, e.g., arrows, wings, falling trees; Phil, also use
the word to describe the unpleasing sound represented by rho, three instances o
which appear in this phrase: Phil. Poet. P.Here. 994 col. 33.5-8 (text available i
 Epigram 36 195

F. Sbordone, "Filodemo e la teorica dell'eufonia," in Sui papiri della Poetica di

Filodemo [Naples 1983] 142); similarly D.H. Comp. 14 and S.E. M. 1.102.
Stadtmiiller in app.crit. suggested the reading comparing Nonnos 48.
307
That is, a productive source of; LSJ s.v. II. Thus, Pylos, Iton, Phthia,
nd Thrace is each called by Homer. Cf. I. W s e r n , T h e
Kenning in Pre-Christian Poetry (Uppsala 1951), esp. 7 £, 49 £., 108 (where she
compares Phil.'s use of the "messenger kenning" in 5.3 f. and 4.3 £); and West's
index to Hes. Op., s.v. "kenning." Perhaps the author of this victory poem had in
mind Pi. Ol. 8.1
In wishes and prayers; Denniston GP 16.

7 An epic epithet; e.g. II. 20.224 (cf. Edwards ad loc.), Od. 9.536,
Hes. Th. 278, H.Herm. 347 (vocative); and Schmidt's conjecture at 34.3.
Occasionally the request that the god receive the offering is made ex-
plicit; e.g., IG 12 Suppl. p. 86 (CEG 1.345 = Lazzarini 800)
AP 6.18.5 (Julian), 47.3 (Antipater 43 HE) 191.2 (Cornelius
Longus) 250.3 (Antiphilos 1 GP).

8 With this wish for future victories, cf. Inschrift. von

Olympia 174.3 f. (CEG 2.827 = Geffcken 131 = Ebert55)
Such wishes or prayers are also found at the
conclusion of epinicia; cf., e.g., Pi. I. 1.64-68, with E. L. Bundy, Studia Pindarica
(Berkeley 1986) 77-83. Gow-Page notice the oddness of here; it seems to fall
under LSJ s.v. IV, "relation." Since, strictly speaking, Poseidon cannot crown an
Olympic victor, this phrase may be a compression of something like "crown me
(here) for (i.e., looking forward to) a future Olympic contest."
The vox propria in this context, as many exx. in Ebert show. Usually
the subject of the verb is the victor's city (or fellow citizens), but cf. Ebert 79.3 (Chios,
iiA.D.) Alkaios 9 HE CAP 12.64.1-2)

36

5
196 Epigram 36

AP 5.8 [Meleager 69]

P PI 7.89, f. 72v [sc.

2 PP1: fort. 5 PPL fort. PP1: tent. Gow-

Page PP1: Polak, Deheque: tent. Stadtmiiller 6
CP1: P

Sacred night and oil-lamp, we two together chose no other witnesses for
our oaths than you.
We swore: he that he would love me, I that I would never leave him. You
were witnesses to testimony sworn jointly.
But now he says that those oaths are carried on water, and you, lamp, see
him in the bosom of others.

Ludwig, MH 19 (1962) 156 f.

Wifstrand 55 f.

This poem is unworthy of both Meleager and Phil. That the theme is a common
one proves nothing in itself, but this, combined with the several infelicities detailed
below and the lack of any final point, does not point towards Phil. Gow-Page point
out that of the two poets only Phil, elsewhere writes in the persona of a woman, but
this could well have been done by Meleager as well (as Gow-Page acknowledge),
and of course by any later imitator; see below on 6. It is accepted as Philodemean
by K. P. Schulze, BPhW}6 (1916) 319. Wifstrand and Ludwig point out how this
poem offers variants both on Kallim. Ep.11 HE (see below on 5) and, with specific
reference to the poem's being spoken by a woman (Ludwig), on Asklepiades 9 HE
(AP5.1):

1 Since this adj. means that something belongs to or is closely associ-

ated with a god, it does not directly modify a god; here, therefore, as in Aisch. fr.
69.7 and Eur. fr. 114 N2 night is simply part of the daily
cycle controlled by the gods; cf. II. 11.194:
Cf. P. Wiilfing-von Martitz, " 'I bei Homer und in der alteren griechischen
Literatur," Glotta 38 (1960) 272-307; 39 (1961) 24-43, esp 26 f.; J. T. Hooker,
Hieros in Early Greek (Innsbruck 1980), who 25 f. argues for the word's conveying
the meanings "strong," "swift" when applied to night.
Lamps are often witnesses to erotic scenes
in the Anthology (see on 7.1), cf. the last line; here, though, strictly speaking, lam
and night are but witnesses of the oaths sworn by the lovers. Cf. also Asklepiades
13 HE (AP 5.164.1) The word is
 Epigram 36 197

relatively uncommon in early literature: once each in Aischylos, Euripides, and

Thucydides, and twice in Sophokles.

1-2 For after a negative clause, cf. KG 2.283 f.

and e.g. Il.2l.275 f.

2 This is the only instance of a dativus ret with rather than a

genitive, which may well have been the original reading. When a dative occurs with
either the noun or the equivalent verb it refers to the person for whom one acts as
witness; e.g., Phil. DeMus. IV col. 18.13

3 Correption of a disyllabic word is rare at the bucolic diaeresis: once else-

where in Meleager 95.1 (AP 12.60) and once elsewhere in Phil. 22.5, where, prob-
ably, metrical roughness is intended.

4 As Gow-Page note, word order suggests that the

emphasis is on the oaths sworn together by the two lovers. The phrase
('to receive as testimony': cf. Aristot. Pol. 1338a36) suggests that the lovers
swore to, or by, night and the lamp. See on 7.2
Meleager often has elided at the pentameter half (Gow-Page ad
loc.); Phil, at 31.2 and 21.4.

5-6 6 Only the clauses as a whole are properly contrasted; the con-
trast which seems at first glance to be drawn between "you" and "lamp" is awk-
ward; as is perhaps the reference to the lamp without night.

5 Hesiod (fr. 124 M-W) is credited with first expressing this

idea:

And among the epigrammatists, cf. Kallim. Ep. 11 HE (25 Pf. = AP 5.6):
K
The most influential poetic model, however, is Soph, fr.811
which produced this and many other copies; e.g.,
Menander, Sent. 26]., Philonides fr. 7 K-A
Xenarchos fr. 6 K-A
Catullus 70.3 f. mulier cupido quoddicit amanti I in vento et rapida
scribere oportet aqua. Since all these versions and others (Plato, Philostratos, Lucian,
et al.) speak of "writing," it is tempting to emend accordingly. Gow-Page regard
the paradosis as "free from objection" on the grounds that "variation of familiar
phraseology is characteristic of Meleager's style." But all that passes without objec-
tion is that Greek can use the verb for something "carried" or "floating" on
water—a notion which applied to oaths makes little or no sense here, and cannot
198 Epigram 36

serve as a meaningful variation of this well-known phrase. Rather than alter the text,
however (Stadtmiiller attributes to Polak, Waltz to Deheque), I would
regard this inappropriate word as the author's. would not improve matters
with and is otiose with whereas the reference of is easily
understood as

6 Parallels suggest that this phrase refers to a man lying with a woman
(note esp. Phil. 15 and23; see further Gow-Page, intro.), but as Petronius' Encolpius
indicates, could also be meant (so Stadtmuller, Beckby, and Geffcken).
For similar endings, cf. Phil. 23.8 and Meleager 42 G4P5.136)

37

/IP5.113 [Argentarius 9 GP]

IJ] [P] PI 7.95, f. 72v [sc.

2 CP1: P 3 P: pjpc. P1ac 4 P: PI:
Stephanus 5 PI: P C[rasura]Pl: P

You fell in love when you had money, Sosikrates, but, now a poor man,
you no longer love. Hunger has such curative power!
And she, Menophila, who before used to call you Sweety and Darling
Adonis, now asks you your name:
"Whose son art thou? Where pray is thy city?" It wasn't easy for you to
learn this saw, Nobody is a friend to the man who has nothing.

Arguing against Philodemean authorship are:

(A) The absence of its incipit in n ; see below, p. 204.

(B) Metrics: (i) 2 Phil, is among the strictest of epigrammatists
in having the syllable before the caesura in the pentameter long by nature
(see on 20.2, 31.6). Argentarius on the other hand admits two instances:
 Epigram 38 199

(a) (31GP = AP 7.384.6) and (b) (35

GP = ,4P9.221.6: P1 P, the latter rightly rejected by GP). (li) 3
Generally, as here, a spondaic fourth foot occurs before word-
end only with a prospective monosyllable (Naeke's Law); GP l.xliv. But
even with prospective monosyllable such lines are rare. Phil, has only one
such in the undisputed poems (9.7 whereas Argentanus has two
others: 13.3 and 35.5
(C) Argentarius writes of Menophila in two other poems (7,10GP=AP5.105,
116), but cf. Phil.'s use of Heliodora (13, intro.).

The ascription to Phil, probably arose, as Gow-Page note, from its following
immediately upon a genuine poem of his in Kephalas' collection (5), since the two
poems are found in the same order in both P and P1.

38

Reiske 9 Brunck 9

Ap.L Cr
6 : codd. 7 Toup: ' Ao.L Ci Ao.L:
Cr Ap.L' Cr • Ap.LP' Cr Ap.L 8
codd. codd.

In the recesses of my heart I nourish two loves, one for a Roman woman,
the other for a Corinthian.
The first knows how to cherish the ways and the manners of a matrons.
from her hairnet to her anklets.
The other wantonly lends herself to all manner of love, complaisantly
taking on the positions described by Elephantis.
If, Piso, you bid me choose one of the two, I remain in Ephyra. May a Gallos
have the other!
200 Epigram 38

F. Jacobs, "Uber ein dem Philodemus beigelegtes Epigramm," in Vermischte Schriften 5 (Berlin
1834) 264-291.

This epigram was first printed in 1754 by Reiske, who found it as the last poem of
the Leipzig apograph of P, where it is written by a different hand on a separate
piece of paper tipped in at the end of the volume; and at the end of what he refers
to as \he.Schedae La Crosii, i.e., a copy, now in Hamburg (Cr), containing epigrams
purportedly derived from P: Epigrammata graeca inedita, descripta primum a
Friderico Sylburgio e codice Msto. bibliothecae Palatinae, ex cuius apographo quod
eratapud Isaac. Vossium, ea descripsitlll. Ezech. Spanheimius, ex cuius codice ego ea
descripsi Berolini A.C. 1716, and signed Maturinus Veysiere La Croze. Since this
poem is not in fact found in P, someone either carelessly let a poem found else-
where stand at the end of a collection of epigrams from the Anthology, or may have
mistaken his exemplar (see below) as a source of pure Anthology material; for the
extremely complicated relationship among the many late copies of the Anthology,
cf. R. Aubreton, "La tradition del'Anthotogie Palatine duxvie auxviiiesiecle,"Ref.
d'Hist. des Textes 10 (1980) 1-52, esp. 5-14,24-27; J. Mutton, Greek Anthology in
France (Ithaca 1946) 8 ff.
The name Piso and the phrase in the same context as
Horace's Gallis hanc, Philodemus ait (Sat. 1.2.121) led Reiske to ascribe the poem
to Phil. Brunck, followed, with reservations, by Jacobs in his first edition of the
Anthology, did likewise, and it was considered Philodemean by J. Toup, Opuscula
Critica (Leipzig 1780) 1.158f. Itwas regarded as un-Philodemean by Rosini, VH 1
(1793) 1 n.l and as a later forgery ("due a quelque moderne") by Chardon de La
Rochette, Magasin encyclopedique 4.1 (1798) 563, who seems eventually to have
convinced others. A more substantial argument against authenticity was presented
by Jacobs in 1816 (reprinted in his Vermischte Schriften, cited above). As a result,
later editors of the Palatine and Planudean manuscripts, from Jacobs 1813 on, were
under no obligation to print this and the other poems found only in apographa.
This epigram thus dropped from sight, to the extent that Gow-Page, who of course
knew the works of Reiske, Brunck, and Jacobs (note especially HE l.xliv), make
no reference to it in their notes to the Horace passage; nor are recent commenta-
tors on Horace aware of it. (Orelli 1852 ed. of Horace prints the poem, but regards
it as a forgery.)
Is the epigram actually by Phil.? "Elegans tamen epigramma, nec vetere poeta
indignum," was Jacobs's original assessment, and students of both Phil, and Horace
would surely welcome a positive answer to this question. The only sure piece of
evidence for Philodemean authorship of an anonymous poem, however, is lacking,
i.e., the presence of its incipit in II (see below, p. 204, introduction). That is,
although to an unknown extent, its absence from n counts against Philodemean
authorship. (It is possible that the poem Horace refers to is II v.29 TT)V cmo
see ad loc.)
Chardon de La Rochette, loc. cit., rejected ancient authorship for the poem
entirely, it would seem, because of its absence from P and P1. He attributed it to a
"modern" poet who wished to pass it off as ancient. The case against Philodemean
 Epigram 38 201

authorship is certain. In addition to some other considerations mentioned in the

commentary, note two metrical objections:

(a) The scansion of v.2 at the pentameter-half, which Phil.

generally avoids (see Intro., p. 44).
(b) Hiatus at v.3 is unique. Everywhere else Phil, correpts be-
fore a vowel: 1.1, 6.3, 12.7, 14.5, 17.4, 19.1, 27.7, 29.1,3 (Jacobs 274).
(c) No hexameter contains a bucolic diaeresis, in sharp contrast with Phil.'s
marked preference (Jacobs 273).

A likely guess is that it was composed by an eruditus, Renaissance or later,

either (as Chardon de La Rochette thought) as a forgery, or, as I am willing to be-
lieve, with no intention to defraud, but to provide for the amusement of himself
and for his friends a simulacrum of the lost original alluded to by Horace. Although
the topos of comparing and contrasting two disparate types of women is common
in Phil., the contrast is clear enough in Horace, Sat. 1.2. The address to Piso is a
nice touch, but as the morphology hints (see below), the author's knowledge of
Phil, 's friendship with Piso mav derive solely from Cicero's In Pisonem [T 2] rather
than from Epigram 27, where the manuscript has (and certainly not from
the then unknown Good King). Note that 27 is lacking in PI, and so would have
been unknown to all but a few scholars and literati.
A possible author of this poem is Daniel Heinsius, many of whose Greek epi-
grams are written "to" or "by" Greek philosophers; Poemata Graeca (Leiden 1640).
He also published a six-part cycle of erotic epigrams addressed to Demophile (ibid.
DP. 57-60): note esp. the one on p. 59. which begins
and ends with the choosing of the
latter via a geographic reference:
Heinsius also translated one of Phil.'s epigrams (14)
into Latin (ibid. p.143). But perhaps most interesting is that in his notes toSat. 1.2 he
recognized Horace's debt to Phil., first in his borrowing 92 o cms! o bracchia! from
12, and then from Horace's direct reference to Philodemus he inferred that certum
est magnam hums Satyraepartem ad Philodemi epigmmmata amatoria, quorum multum
nondumeditahabemus, respexissepoetam. . . . "GallishancPhilodemus ait." Hie nobis
dubium non est quin ad illas Philodemi exclamationes attuserit (D. Heinsius, Quintus
Horatius Flaccus [Leiden 1629] 2.53). Indeed, as Hutton, op. cit., 255-259, points
out, not only was Heinsius among the first to hear from Saumaise about the Palatine
manuscript, he made made good use of it in his writings, especially his edition of
Horace. An objection to Heinsius' authorship may be the form (see above
and the commentary), but this perhaps may simply be a hypercorrective slip on his
part. There might have been a twinkle in Heinsius' eye when he wrote on Sat. 1.2.121
"locus plus venustatis haberet si quidem ipsa Philodemi exstarent verba."
St. Petersburg ms. Greek no. 148 (The Hague 1637) tantalizingly mentions
"Philodemi apud Horat. lib. I, sat. II, v. 121" on page 2, but does not in fact con-
tain a copy of this epigram. Cf. E. von Muralt, Cat. des mss. grecs de la Bibliotheque
Imperiale Publique (St. Petersburg 1846) 810; Hutton 9. (Its readings of 7, 11, 14,
20, 22, 34 offer nothing new.)
202 Epigram 38

1 Neither this form nor 8 appears in the undisputed poems, but Phil.
surely would have felt free to use these well-established poetic variants.

2 Both are late forms; for the former only Damaskios

ap. Suda s.v. " " Steph. Byz. Eth. 374.2; for the latter only Steph. Byz.
Eth. 548.12. Since Stephanos also (not surprisingly, given the purpose of theEthnika)
discusses Ephyra (see below, on v.8), he may well be the source of these rare adjec-
tives for the author of this poem (ed. princ. Venet. 1502, ed. A. Manutius).

3 Hor. Sat. 1.2.54 matronam nullam ego tango, 62 f. quid inter I est in
matrona, ancilla, peccesne togata?, 78 desine matronassectarier, 94 matronaepraeter
faciem nil cernere possis.

5 This word sits oddly in this erotic context: "gently" or "soothingly"

is precisely what is not wanted from this Corinthian woman. Epicurus in particular
contrasts
and other disturbing feelings (fr. 411 U).

6 Elephantis seems to have written a compendious work (for

women?) containing cosmetic, medical, and erotic material, accompanied by illus-
trations of sexual positions; cf. O. Crusius, RE 5 (1905) 2324 f. Since the genitive
of her name is 'E _ (-idis or -idos in Suetonius, Tib. 43, Martial 12.43.4,
etc.), either the author or the transmission is at fault. If the latter (an error of
homoioteleuton four lines after Ki , a form such as 8.8 N< is likely.
If the former, he was surely not Phil.; and I should not "improve" the text.

7 Cf. 27.1 v , Good King col. 43.16 f. . Our au-

thor thought that the vocative was formed like other names in , e.g.,'
(Jacobs 275).
Having written , the author needed a consonant here to make posi-
tion, doubtless intending it to represent rather than (with epic elision)
The active seems too strong; Reiske accordingly read in the
sense of L. tollere, but this is no better (Jacobs 275).

8 : Another name for Corinth; E ad II. 6.152; Steph. Byz. Eth. 290.9 f., 300.20
f., 374.1.
Reiske identifies the hand of the marginal annotations as that of
Gisbert Cuyper's, one of the several people who had access to the apograph before
it came to rest in Leipzig; Hutton 9,270 f. could just barely stand (cf. Toup),
but even if were not uncommon, if this poem is to be understood as Horace's
model (real or imagined), ; has to be read unless one agrees with Toup that
Horace's Gallis derives from his misreading of the Greek exemplar!
 P. OXY. 3 7 2 4

P.Oxy. 54 (1987) 3724, ed. P. Parsons. Reviewed: W. Luppe, CR 39 (1989) 125-

126 (who thinks they all are by Phil., even the two ascribed to Asklepiades); B. Palme,
Tyche 3 (1988) 306. Cf. also A. Cameron, Greek Anthology, app. 7; M. Gigante,
"Filodemo tra poesia e prosa (A proposito di P.Oxy. 3724)," SIFC 7 (1989) 129-
151 (cited in this chapter as "Gigante" plus page number). What follows incorpo-
rates Sider.ZPE 76 (1989) 229-236, and draws freely on Parsons's edition and notes.
References below to Griffiths are to some unpublished observations, used with his
kind permission; other scholars cited below without further reference, except for
Obbink, are found in Parsons.
P.Oxy. 3724, written (on both sides) in the later first century A.D., comprising
three fragments, consists largely of ca. 175 incipits (i.e., the first few words, the last
of which may be abbreviated), chiefly by hand C in fr. 1, of which 27 are attributed
to Phil, by the Anthology (in two cases not uniformly, however). Frr. 2 and 3, writ-
ten by hand A, also contain what seem to be epigram incipits. Two incipits in fr. 1
belong to poems already known but anonymous (col. ii.2 and 28). Two others be-
long to poems attributed uniformly to a poet other than Phil., namely Asklepiades
(iv.28 and vi.18, the latter also copied out by another hand (A) in full elsewhere in
fr. 1). Thus, only five poems from the Gow-Page canon of Phil, are unaccounted
for: 11,15,16,30,31 (my numeration). Since all but, it seems, one of the incipits to
poems already known appear in the Greek Anthology, it seems safe to assume that
all the unknown poems, at least those among the list written out by hand C, are
epigrams as well. (For the possible exception, see on col. ii.2.)
It is possible that these incipits list (for what purposes we are not sure) the
personal favorites of someone who liked almost everything he found by Phil, in

203
204 P.Oxy. 3724

Philip's Garland, in which case, there would be no reason to suspect that any, or at
any rate many, of the unknown incipits come from poems of Phil. But the fact that
there are so many incipits unknown to the Greek Anthology (over 145) indicates
that the compiler could draw from the complete editions of Phil, and perhaps other
poets as well. Contra Gigante 133, who argues that the compiler knew Philip's
Garland.
Since, then, twenty-seven of thirty-one known poems are by Phil., it is clear
that the papyrus does not contain a random sampling; it follows also that, given
that Phil, wrote more than the thirty plus attributed to him by one source or an-
other (see Intro., pp. 47 ), the likelihood of there being more incipits belonging to
Phil, among the remaining 145 is high. How high, it is impossible to say, given that
two begin poems whose attribution to Asklepiades by the Anthology there is little
reason to doubt (see below, on vi.18). There remains a small possibility, however,
that these two incipits belong to similarly beginning poems of Phil., which would
allow for all the incipits to be Philodemean. (Note how v.28 recalls the begin-
ning of Asklepiades 1.) Attractive as I find this notion, the fact that the Asklepiades
poem which continues vi.18 is copied out in full in the papyrus seems to argue that
this incipit (and if this incipt, then iv.28 as well) is indeed intended to belong to a
poem of Asklepiades. Furthermore, a glance at the complete incipits for the Greek
Anthology (Beckby 4.686-731) shows that inventive epigrammatists over the cen-
turies rarely duplicated beginnings past the second word, even when one includes
common words like articles and particles.
Thus, although I cannot follow Luppe and Cameron, who are willing to have
them all belong to Phil., I believe that most do; contra Gigante, who would assign
to him only ten from among the unknown poems, adding two as doubtful. Parsons
too is cautious but allows that many could come from Phil.
If, then, the proportion of Philodemean to non-Philodemean poems remains
as high among the unknown epigrams as among those already known, the presence
of a doubtful epigram in IT can contribute to our deliberations about authorship.
That is, the presence of an incipit would seem to weigh heavily in favor of
Philodemus as author. The absence, on the other hand, counts much less in weigh-
ing against Phil., since we have no idea from how many published epigrams of Phil,
the compiler of incipits could draw. Thus, I use the existence of an incipit as evi-
dence for authorship in epigrams 2, 21, 24 (the last anonymous in PP1). The ab-
sence of an incipit does not outweigh other evidence deciding for Phil, in epigrams
11, 18, 35, whereas in epigrams 36, 37, absence combines with other negative evi-
dence in helping me to decide against Philodemean authorship.
The main purpose of the commentary will be to consider any piece of evidence
for signs of Philodemean authorship. Any such evidence, no matter how slight, given
the preponderance of already known Philodemean poems among the incipits,
should, it seems to me, point to Phil, as author. I do not claim to prove that any one
incipit is his, although some are surer than others, but the sum total of all that point
to Phil, tends to indicate that even among those showing no signs of Philodemean
authorship there must be a good number written by him. Many of my identifica-
tions have been rejected by Gigante, largely because they do not fit his autobio-
 P.Oxy.3724i 205

graphical scheme for the composition of the epigrams, in which some were written
before and the others after Phil, converted to Epicureanism in Athens—a view I
dispute; cf. above, Intro., pp. 34,40, and my review ofhisFilodemo in Italia, BMCR
2 (1991) 353-355.1 also am not happy with his suggestion that the epigrams repre-
sented by the incipits formed a Garland edited by Phil, himself (134), who, like
Meleager and Philip, included many epigrams written by himself. First, why would
he have included in his own garland Asklepiades, who was already well represented
in Meleager's collection? Second, while it is reasonable (on the assumption that
the incipits are all, or almost all, by Phil.) that Philip would have chosen from among
a larger number of Phil.'s epigrams, leaving the rest unknown to us, it is unlikely
that Phil, (or anyone else for that matter) would have made a selection of poems he
presumably liked, of which Philip chose none but the Philodemean ones.

Fr. 1

col. i (hand A)

contains

(a) the tail end of some hexameter lines (1-14), which may have been either
(like the lines below) a complete hexameter poem, perhaps an oracle
(Cameron) or a hymn (Janko, Tannenbaum), or incipits consisting of first
lines, in which case they could come from epigrams. I agree with Cameron
and others who argue for the former on the grounds that entire lines cop-
ied out immediately above a complete poem suggest that here too there was
a complete poem.
(b) Identifiable line-ends of Asklepiades 12 HE (15-20; for the text, see below,
col. vi.18); and
(c) what may be other complete epigrams (21 ff.), none of which matches any
known epigram, but since Asklepiades' epigram occurs among the incipits
it is possible that these lines likewise presented the full text of an incipit.
Only three lines, however, offer even a few words each:

25
206 P.Oxy. 3724 i-ii.2

All that can be said is that no poem referring to one or more Muses (even in Doric;
cf. 30) and wreaths can be denied to Phil. For details of the readings, here and
throughout, see Parsons.

col. ii (hand C)

1 Most likely an unaugmented plup. of ("I/he/she was still

shrieking/screaming/croaking"), although elsewhere the perfect stem is
Parsons compares

2 _ Identified by Alan Cameron ap. ed. pr. as the begin-

ning of the hexameter "Pythian oracle" derided by the Cynic philosoper Oinomaos
of Gadara (fr. 11 B Hammerstaedt, ap. Eusebios, PE 5.30), who flourished in the
time of Hadrian.

The oracle (414 P-W), which Fontenrose considers unhistoric (L103), was cited
by Chamaileon (ap. Athen. 22e = fr. 11 Werhli = 13 Giordano). Beyond the mere
Gadarene connection between Oinomaos and Phil., note that Oinomaos'
would seem to be similar in methodology to Phil.'
i and that Oinomaos too was a philoso-
pher who wrote poetry (in this case, tragedies [TrGF 188]); cf. H. J. Mette, RE
17.2 (1937) 2249-2251J. Hammerstaedt,Die Orakelkritik desKynikers Oenomaus
(Frankfurt 1988) 48-53. The possibility, admittedly slight, should at least be raised
that Oinomaos knew the oracle from a poem of Phil., with whom he felt some spe-
cial tie. Phil, himself would have seen it in Chamaileon, whom he names twice in
De Musica (frr. 4-5 Wehrli). See above, p. 4.
If Oinomaos derided the oracle , he may have taken
his lead from Phil., who in turn had some statements of Epicurus on the nature of
wine to guide him: cf. fr. 58 U = 20.1 Arr., where Epicurus, in his Symposium, re-
jects general statements about the nature of wine on men. My proposal, then, is
that the incipit here is not, as far as we can see, uniquely in this list that of a hexam-
eter oracle (although we may have a hexameter oracle written out in col. i. 1-14),
but rather that of an epigram of Phil, that begins with a quotation from this oracle
and continues with a criticism or parody of it. (So also Luppe 125 and Griffiths.)
The topos of the best time of year for wine drinking (cf. Hes. WD 588 ff., Alkaios
347.1, Theogn. 1039 f.) would obviously be appropriate for a sympotic epigram
written by any author, but Phil, would be able in addition to allude to the com-
ments Epicurus made on the properties, medical and otherwise, of wine in hisSywz-
posium (frr. 57-65 U). For a parallel to this form of poetic quotation, cf. Simon.
Eleg. 19 W2,
 P.Oxy. 3724 ii.2-ii.10 207

West, Studies in Greek Elegy and Iambus (Berlin 1974) 179 f. cites other examples
of metapoiesis, but a closer parallel may be found in this papyrus, if col. iv. 8
as I suggest, is Phil.' quotation of Kallimachos; see below.
Gigante 137 calls my suggestion "un'ipotesi troppo fantasiosa," but here as
elsewhere he has misread my aim, which is solely to gather evidence that would
explain how an incipit, if by Phil., could continue. On Oinomaos, see now, in
addition to his book cited above, J. Hammerstaedt, "Der Kyniker Oenomaus von
Gadara," ANRWll 36.4 (1990) 2834-2865.
For the equation of Dionysos with wine, see on 6.7

3 (or Doric : Phil,

and most others tend to avoid < , but will do so to accommodate an other-
wise intractable word; cf. Intro., pp. 41 f.

4 Deleted here, copied again on line 24. Griffiths notes that in

the several instances where a duplicate entry is marked for omission, it is the ear-
lier one.

5 Phil. 6.

6 Probably an erotic epigram (although Parsons allows for its

referring to a work of art); cf., e.g., Asklepiades 36 HE (AP5.209), Rufinus 27 Page
(AP 5.73), where the sight of a woman bathing stirs a man's passion.

7 Erotic?

8 Quite possibly the fifth-century

dithyrambic poet of this name who refers to himself in the third person while praising
his poetic skills (Page, FGE 11-15), but the pervasive presence of Phil, among the
incipits makes it more likely that this is the same Antigenes mentioned in 29 (also
below, vii. 21) as being recently deceased, especially since the two references taken
together seem to indicate that he was a poet; cf. Gigante 136 and commentary
to 29.

9 _ The last letter of' is not sure; see Parsons, who also notes
that |-iop<|>ri may refer to a living body or to one represented in painting or sculpture.

10 Most likely , . for Aphrodite, as in Phil. 7.6; note how-

ever its use as the adjective for the city of Paphos (10.1), whence came one of the
many women named Demo in his life; cf. below, col. iv. 13. is either adj. (with
what noun?) or, more likely, adv., as at AP 7.374.2 (Argentarius 19 GP), 7.241.1-
2 (Antipater Sidonius 25 HE), 12.1693-4 (Dioskorides 8 HE).
208 P.Oxy. 3724 ii.ll-ii.20

11 Sc.

12 : Possibly, as Parsons notes, a vocative in a poem written

by a friend of Phil., but more likely another poem by Phil, himself in which he ei-
ther refers to himself in the third person, as he does in 28 (cf. 10 and i.15), or
addresses himself in the vocative (as, e.g., Catullus often does). The verb, "to catch
birds with lime," may be literal, but more probably as an amatory metaphor (cf.
Timotheos Com. fr. 2 K-A (= fr.2 K) Page, FGE
312 f.). One further possibility is suggested by Aristoph. fr. 736 K-A, where the
stickiness of refers to greediness, a quality which Phil, charges himself with in
3 where there is excessive desire for more days, i.e., a longer life, the
poem reflecting a common Epicurean topic of criticism. This incipit could begin a
poem with a similar concern. Cf. Gigante 136.

13 Even when not addressed, lamps are often

mentioned in erotic contexts: see on 7.2, 36.

14 Phil. 4.

15 : A form of or a compound adjective, as Par-

sons notes, is obviously another possibility; cf. above on v.12.

16 The last word, which seems certain to me, was first read by
Rea. Phil. 19 on impotence, adduced by Parsons, uses more refined language (on
the roughness of oruew, cf. Henderson, Maculate Muse 112), but note 22, with
(hvetv (cf. Henderson 152).

17 : Either . A link between wealth and

sobriety? Cf. Phil. Good King col. 19 "For it belongs not only to the sober
, but also to those drinking, to sing the 'glories of men' [7/.9.189]; nor
[does this happen] only among the more severe, but also among the luxurious
Phaeacians" (tr. Asmis).

18 Phil. 22.

19 Phil. 17.

2C Epicureans shun disturbance of the soul, Epic. KD 8

Nat. 131;
Phil, uses this word eight times, ace. to Vooys's Lexicon Philodemeum s.v. The
adjective is at home in both poetry (Aristophanes and Euripides, but not
in the Anthology) and prose, but its stem is prevalent throughout Epicurean writ-
ings (Epic. -KD. 11 etc.) and often in Phil (Epic. ap. Philod.,
fr. 154U = 72Arr.), ,v . i (Epic.Ep. ad Men. 3.127), and< (Epic.Sent.
Vat. 79). Phil, also uses" > (Oik. XIII.10, 29) and (DeMusicap.63.2
Kempke). Gigante 137 rejects a connection between this incipit and Epicureanism.
 P.Oxy. 3724 ii.2l-iii.7 209

21 Phil. 32? This incipit, which belongs to no other known epigram,

appears more fully below, col. iv.16. Conceivably, therefore, this belongs to an
unknown poem, but other duplicates appear in the list, one of which is marked for
deletion, e.g. [iii.7]] = v. 31.

22 or ( v o c . ) : c o u l d be adj or verb.

23

24 ... See above, line 4.

25 : Chian wine of course is a commonplace as a mark of luxury;

in Phil., definitely in 6 and 27 (the invitation to Piso); anc is Page's likely
conjecture in 28.5 for the MS's

26 _ _ , is at home in both poetry (Aristophanes and Euripides, but not

, cf. Secundus 2 GP (AP 9.260.3) APAP 12.2.5 (Strato)
(same sedes).

21 If the last word is not simply Jtpooco, Par-

sons is right to compare other poems (and Plato) where a woman is desired despite
what would be perceived by a dispassionate observer as one or another flaw. Phil,
has three such poems: 9 (Charito is sixty years old), 17 (Philainion is small and black;
the following "but" shows that these qualities are not attractive to all); and 12 (Flora
is an uncultured Oscan girl); and perhaps II iii.15, 17.
Parsons compares Lucr. 4.1169 simula Silena acSaturast, as does Gigante 137,
raising the possibility that Lucretius knew this epigram, or that both draw on an
Epicurean antierotic source.

28 Phil. 24.

29 : For the first word "there are palaeographic difficul-

ties" (Parsons). • which seems certain, would be a hapax. The hith-
erto known prefixed forms of this root are (D.L. 3.34) ;
(Souda s.v.), (Clem. Alex. Paed. 3.4.29.2), and
(B.C. 50.27).

30: There is only the barest, unreadable, trace of another incipit here, and the
possibility that one or two incipits have been lost below.

col. iii
1-6 (hand B): A recipe in prose for cough medicine.

7 Phil. 19, repeated below, v. 31.

210 P. Oxy. 3 724 Hi. 8-iii. 17

8 Repeated below, v. 32, q.v.

9 Griffiths intriguingly compares the beginning of Prop.

Monobiblos, me contactum nullis ante cupidinibus. Either i (Griffiths)
or (Holford-Strevens) is possible.

10 or, less likely If the former, the

answer to this question is anyone (poet or not) who rails against Eros; cf. Alkaios 6
HEW? 5.10.1)

11 : After the iota "a trace like a high point to the right (accidental?)," Par-
sons. This word has been crossed out, although it, unlike other strikeouts, does not
appear elsewhere in II Obbink suggests that the scribe might have begun to write
out the second line of the preceding epigram and then caught himself. This would
account for the shortness of this incipit.

12 ; Parsons compares Argentarius 12 GP (AP 5.127), a man

sleeping with both mother and daughter. Or perhaps Hor. O. 1.16.1 o matre. . .
pulchrior is pertinent, although metrical norms calls for ur|TT|p KCII 0i>ydn;T|p rather
than

13 If erotic, praise of a woman's hair (cf. Phil. 9.2); a dedication

is also possible (cf. Argentarius 17 GP =AP 6.201.2 and, in general, W. H. D. Rouse,
Greek Votive Offerings [Cambridge 1902] 240-245).

14 See on 4.1, where it is pointed out that 18 years is the ideal

age for a woman to marry. This poem, then, could come from the Xanthippe cycle;
contra Gigante 137. On the other hand, epitaphs frequently point up the pathos of
someone's having died young by giving the age at death; e.g., Kaibel, EG 222
Similarly, AP 1.167.5
7.466.3 (Leonidas 71 HE), 7.468.2 (Meleager 125), CEG 2.709.4,
739.2.

15 ; Alan Cameron has pointed out that this poem begins with
an exact contrast to Phil. 17; see above, col. ii.19. Gigante aptly adduces Hor. Sat.
1.2.123 f. (immediately after Phil, has been named) as an imitation of what is very
likely a now lost epigram of Phil.:

Candida rectaque sit; munda hactenus, ut neque longa

nee magis alba velit quam dat natura videri.

16.

17 : The entails a contrast, such as (i) "If she were not beau-
tiful, I would not pay her anything, but (5e) since she is I pay her high price"; or
(ii) "If she were not beautiful, I would not care about her price, and (5e) if she
 P.Oxy. 3724 iii.l8-iv.3 211

were inexpensive, I would not care about her beauty. As it is, she is both beautiful
and cheap, so count me blessed." Cf. 22.

18 ': It is not by wealth that riches are judged,

as Epicurus said; e.g., Sent. Vat. 25, "Poverty, when measured by the natural pur-
pose of life, is great wealth, but unlimited wealth is great poverty" (tr. Bailey); "If
you wish to make Pythokles wealthy, don't profer money but take away the desire"
(fr. 46 Arr. = 135 U). See below on Frr. 2-3.

19 ; Cf. Aisch. Ag. 39 . 1402

Pi. Ol. 2.85 For
verse, cf. M. L. West, The Orphic Poems (Oxford 1983) is less
likely.

2C : Phil. 14 has (as I understand it) the sleeping Kallistion.

Is this poem erotic (cf. Meleager 36 -AP5.174.1 or sepulchral
(cf. Antipater Sid. 16 HE = AP 7.29.1 Either
Kallikrates or Kallikrateia, both names appearing in the Anthology, the latter in
two anonymous sepulchral epigrams (AP 7.224.1, 691.4).

21

22 This is Rea's almost certain reading can represent

as in vi. 11. Three people drank yesterday from a —on which cf. Gow-
Page on Hedylos 5 HE.

23

Two incipits may have been lost below 23.

col. iv
1 Of the possible meanings of Xanthion rightly
considered by Parsons—herb, town, man's name, girl friend of Phil.—the last alone
obviously has the context going for it; cf. Introduction, pp. 34-38, where I present
the case for a Xanthippe-cycle in Phil.' poetry. Each of the two articulations (with
at least four possible construals) seems equally likely. Gigante 136 suggests!
Xanthion, I did not know what love was, until
you revealed it to me—a statement which can be made at the beginning of relation-
ship (as Gigante believes) or toward its end.

2 Verb or (perhaps personified) noun.

3 Preparation for a dinner (Parsons), perhaps in

an invitation poem.
212 P.Oxy. 3724 iv.4-iv.14

4 Phil. 27.

7 Phil. 3. II confirms Huschke's conjecture.

8 .8 : Q u i t e l i k e l y , a s P a r s o n s s a y s , a p o e t i c e l a b o r a t i o n o r r e a c -
tion to Kallimachos' poem with similar beginning (2 HE = 28 Pf.
and Phil, of course was very much concerned with the nature of
poetry. But if by Phil, how would it continue? Perhaps just as Kallimachos' did,
with , the thought now reversed, for Phil, did not regard length as a
r
criterion of a poem's worth. Cf. Bk. V, col. 7.1-6Mangoni(= col.
4 Jensen)

See also his reference to epigrammatists, discussed above, Intro., pp. 28-30, vii. 5. If
indeed modeled on Kallimachos, this poem too may have begun as a (here nega-
tive) priamel which introduces an erotic subject; cf. A. Henrichs, "Kallimachos
Epigram 28: A fastidious priamel," HSCP 83 (1979) 207-212. Of all the poems whose
incipits we have this is the one I would most like to have in full. See now A. Cameron,
Callimachus and His Critics (Princeton 1995) 387-402.

9 ... but you would not listen; probably erotic. Perhaps this
epigram was meant to follow 23.

10 Phil. 7.
:
11 'The doubtful letters look most like ."Parsons,
who suggests a reference here to the casting to the winds; cf. Meleager 125 (AP
7.468.8) (a notably sibilant line), "Bar-
ren of offspring, you have spat affection to the winds" (tr. Gow-Page).

12 Parsons suggests correcting to KpoKoevxa.

13 See on ii.10.

14 The traces are quite difficult to read. The triangu-

lar shape which suggested a broken delta to Parsons might just as well be a broken
r
sigma or (iam Gigante' most readily satisfies the constraints
of traces, morphology, and metrics, but other suggestions will be welcome. The
compound verb could refer to (i) departure (LSJ s.v. Ill; sc. from Naples, for which
see below); or (ii) a breakup into elements (cf. Tim. Lokr. 102d, Phil. De Morte
30.4 f. , which here would more specifically
allude to death (cf. Diog. Oin. 3 II 11 f. Smith
The first word, however, is clear. A reference by Phil, to Naples by its other
name of Parthenope, or to the town's eponymous Siren, would not be surprising,
 P.Oxy. 3724 iv.14-iv.15 213

even if in his prose he seems to have used Neapolis, if P.Herc. 312 (T 15) may be
assigned to him. As Lykophron 718-721 tells the story, the body of the Siren
Parthenope was washed ashore near Naples and there entombed (cf. Ps.-Aristotle
Mirab. 839a32, Eustath. ad Od. 12.167, p. 1709). A Rhodian settlement on this site
was named after her (Herodian, Pros. Kath. 339.18 = Stephen Byz. Ethn. 504.6),
but in time her name could be applied to all of Naples (Herod, op.cit. 388.20
= Stephen 656.20; cf. Pliny, NH 3.62
Neapolis. . . Parthenope a tumulo Sirenis appellata). Cf. further J. Ilberg, "Parthenope,"
Roscher,Lexikon3.1653ff;M. Napoli, Napoli greco-romana (Napoli 1959);B. Capasso,
Napoligreco-romana esposita nella topografia. . . Soc. Nap. di Storia Patria (Napoli
1905); E. Pirovine, Napoli nella visione del golfo delle Sirene (Napoli 1977), esp.
9-28; M. Gras, "Il golfo di Napoli e ilTirreno arcaico," inNeapolis: Atti del venticin-
quesimo convegno di studi sulla Magna Grecia (Taranto 1986) 11-35, esp. 17-19
(where, doubting that there was a Rhodian settlement, Gras argues that Cumaean
Parthenope became Palaiopolis which later merged with Neapolis); F. Cassola,
"Problemi di storia neapolitana," ibid. 37-81, esp. 40-45 (who allows for a Rhodian
Parthenope before the Cumaean settlement).
Naples could have been mentioned for any number of reasons by Phil, or
another poet, but note how Vergil, who studied Epicurean theory there with Siro
and Phil., not only begins a verse in the sphragis to the Georgics with the same word
as that of this incipit, but also recalls the pleasure it afforded him while writing
(4.563 )

illo Vergilium me tempore dulcis alebat

Parthenope studiis florentem ignobilis oti,

with which cf. Ov. Met. 15. 15.711 f. in otia natam \ Parthenopen. Cf. Tait 48-63;
M. Korenjak, "Parthenope und Parthenias: Zur Sphragis der Georgika," Mnemosyne,
4th ser. 48 (1995) 201 f, who argues (I think unconvincingly) that with Parthenope
Vergil meant to recall his own nickname Parthenias, on which see below v. 19. Phil,
may well have referred similarly to the Edenic character of the place that allows
Epicurean life to flourish. Dirk Obbink has reasonably suggested that Vergil's
sphragis, written ca. 30B.C., alludes to this (or to the next incipit's poem). This view
was strongly criticized by Gigante 139, who improbably would reverse the allusion
and have Phil, refer to the Georgics and to Vergil himself. He thus restores
(sc. Vergil), which he strangely understands as "alumnus of
Naples"; cf. also his "LabrigatavirgilianaadErcolano," in M. Gigante (ed.),Virgilio
e gli Augustei (Naples 1990) 13 f.; "Virgilio e i suoi amici tra Napoli e Ercolano,"
AttieMem. dell'Ace. Naz. VirgilianadiScienzeLettereedArtidiMantova59 (1991)
97. But if there is any link between this incipit (and the next) and Vergil, it is the
sphragis to the Georgics, which in all likelihood was written after Phil's death.

15 Parsons would read IT . I can find no an-

cient source attesting to the healing powers of Parthenope, but Norman Douglas's
travel book on the region records that "the siren Parthenope escaped by taking
refuge during mediaeval storms in the narrow confines of an amulet, such Siren-
charms as are still seen in the streets of Naples and credited with with peculiar ef-
214 P.Oxy. 3724 iv. 15-iv.25

ficacy against the evil eye" (Siren Land, London 1911; I quote from pp. 12f.of the
Penguin ed.). It would not be unknown for such modern beliefs to have ancient
origins.

16 : Phil. 32. The papyrus shows that Gow-Page were

too quick to obelize the last word, difficult as it is. Nor should they have been so
dismissive of Giangrande, RM 106 (1963) 255-257, who made a good case for
having here the sense of "sacrificial cake," although this is not the meaning
I adopt.

17 Phil. 13, attributed to Phil, by both P and P1 (the latter

with the common , but generally regarded by editors since Jacobs as the
work of Meleager on the grounds that the girl of this poem, Heliodora, regularly
shows up in his poems. It would now seem that the MSS are correct in their
attribution.

18 Phil. 33.

19 Phil. 34.

20 In his prose Phil, is fond of the construction

+ infinitive; so perhaps "It is the hour for the man who drinks his wine
straight to . . ."?

21 : After is possible. As Parsons notes, this may be another

poem of Phil, in which a woman named Demo is mentioned, especially as he seems
to have had a predilection for women with this name (10, cf. 11).

22 ; "It is not clear whether the final sigma was meant to be

suspended" (Parsons); i.e., the last word may be __ Most likely another
accusative followed: "The women having robbed the temple (of ?)."

23 "The doubtful letter looks like alpha; if so,

'Apod (a deliberate sophistication of the usual trio)?" (Parsons). Connection, if any,
with 24 or 26 below is unclear.

24 : A possible imitation of this poem is AP 6.17 (Loukianos):

(i) It begins . (see above); (ii) One of the three hetairai mentioned, Atthis,
makes a dedication to Aphrodite. The incipit could continue, e.g.
~ - -. Atthis of course may refer to Attica or Athens, but we should not
exclude Gadara, which Meleager 2 (AP 7.417.2) calls the Syrian Atthis.

25 ButKaioap( ) and (from Holford-Strevens) )

are also possible. Gigante 140 f. tends to favor the incipit as printed here, believing
it to begin a poem of Phil., and hence that the Caesar in question is Julius.
 P.Oxy. 3724 iv.26-w.30 215

26 _ Although several epigrams refer to a woman as the fourth

Grace and/or second Aphrodite (and tenth Muse), a topos Phil, probably tried at
least once (Ep. Bob. 35; see on 2.2 and above, 23), the combination of the number
four with Kypris and the neuter gender makes this an improbable topos for this
incipit. I suggest rather an erotic parody modeled on an Empedoclean line similar
to his B 96.3 , i.e., "four parts of fire" as an ingredient in the
composition of bone. If so, this incipit could be like ii.2 and iv.8 in taking another
line from someone else's poem as a starting point. Cf. 2.

27 Beggars occur often in the Anthology; cf. Phil. 32.

28 The incipit to AP5.150 (Asklepiades 10 HE, 17 Knauer) :

P Sudass (1-2) caret PI

1 C: Suda 2N otn.Suda P add. C

Since this poem occurs in a Meleagrian sequence of AP (5.134-215), where there

is no discrepant ascription, and since there is nothing that would in any way sug-
gest Philodemean authorship, it should be allowed to remain as Asklepiades', who
uses the name Niko elsewhere (13 HE -AP5.164; cf. further S. L. Taran, The Art
of Variation in the Hellenistic Epigram [Leiden 1979] 83). The poem is ascribed to
Asklepiades by the scribe of P, not, as Parsons mistakenly says, by C. See on col.
vi. 18. But see the introductory comments above, on the likelihood that this incipit,
as opposed to the poem of Asklepiades, belongs to Phil.
The situation described here is a variant, probably a conscious one, of that in
Sappho 168B Voigi cf. Knauer ad loc., B. Lavagnini, Nuova
antologia dei frammenti della lirica greca (Turin 1932) 184 ff.; P. Maas, "Zum
griechischenWortschatz," Melanges EmileBoisacq 2 =AIPhO6 (1938) 131 . (repr.
in KS 199 f.); W. Ludwig, "Ein Epigrammpaar des Asklepiades (A.P. v, 7/150),"
MH 19 (1964) 191-199.

29

30 If the verb struck its original audience as it strikes us, as

a hapax, it would at the very first moment mean nothing more than "I have lost my
tonos, I am unstrung" (cf. " lose force," Herodotos Medicus, i A.D.
(RM 58 [1903] 99); Tim.
Lokr. 103de). Tonos is practically a technical term of the Stoics, so it is quite pos-
sible that the hapax, together with whatever followed, produced a lighthearted love
poem couched in Stoic terms. If it is by Phil., it may well have been, e.g., a parody
of Stoic views on the place of emotions in poetry, which Phil, criticizes throughout
216 P.Oxy. 3724 iv.30-v.8

his De Musica IV; cf. the statements on Diogenes of Babylon to the effect that music
produces harmony within the soul (coll. 7.22-31,21.23-35 Neubecker).

31 Phil. 2.

32 I do not understand why Parsons compares Phil. 12—

below, col. v.20—since an address to "hammers of (the) head" could not easily begin
a poem parallel to 12. Perhaps we have here a poem on that remarkable man Phil,
refers to in De Signis 4, "the man in Alexandria half a cubit high, with a colossal
head that could be beaten with a hammer who used to
be exhibited by the embalmers" (trans. De Lacy and De Lacy, who note, p. 93 n. 9,
that he is "apparently not mentioned elsewhere").
Gigante 141, rejecting my suggestion, would like to read this an erotic epigram
like 12 he suggest reading 'o ankles, o head, "but seems
certain, and "ankles," does not scan.

33 Two epigrams display the motif of a stone converted by

Pheidias to a statue of Nemesis (APl 221 £, by Theaitetos and Parmenion respec-
tively). If this poem is by Phil., it could have gone on to reflect Epicurean views on
the worship of stones in the shape of gods. Compare Phil. 30, which I interpret
similarly.

col. v

1 , and are the likely supplements. Zeus most

likely, although Poseidon and Hades are also possible (Parsons).

3 Phil. 14. (AP 12.250 [Strato], beginning vuKtepivr|v was of course

written later than our papyrus.)

6 A name beginning Demo-?

7 A form of: ; If the latter, a reference to Mareotic

wine may follow. Cf. Hor. O. 1.37.14, with Nisbet-Hubbard's n. ad loc. For the
form - rather than the usual ), Lloyd-Jones adduces Steph. Byz. s.v.

8
 P.Oxy. 3724 V.9-V.19 217

9 (abounding in doves, II. 2.502, 582) and

suggest possible restorations.

10 On eram, cf. Z A ad //. 5.887, adduced by

Pfeiffer on Kallim. fr. 384.32. Spoken by a Gallos?

11 ; Phil. 5.

12 ., _, _,: Ladles as a measure of wine in a sympotic epigram is likely; cf.

Argentarius 8 GP (AP5.110.1 f.)

Similarly, Anakr. 383 PMG, Alexis 116 K-A.

13 Most likely Phil. 25.

14 .: Phil. 1, where the MSS have , . „, which I think cor-

rect, although Gigante 143 n. 55 defends IT 's reading; cf. the commentary ad loc.

15 .: Parsons notes that occurs in the Anthology only in

a poem (21) attributed both to Antiphilos (by P) and to Phil, (by C, PI, and, it would
seem by II vi.4); see the commentary. The occurrence of in this context
strongly suggests Philodemean authorship for both incipit and 21; but see below,
line 19.

16 _: Probably both here are unguents, although the former

can refer to the flower spikenard. The latter appears in Phil. 6.2

17 i; Perhaps the beginning of a list of rural delights, which

may then be contrasted with more urbane pleasures; cf. the preceding incipit (Par-
sons) and Phil. 6.

18 : To scan, we must read _, an error that usually presup-

poses an original OPOTEOE, a spelling which would of course antedate Phil. Per-
haps here a simple scribal error. It was on the island of Pharos where Menelaos
met Proteus (Od. 4.385, Eur. Hel. 1-37), who was worshipped by the early inhab-
itants of the region; cf. [Kallisth.] 1.32.2
Cf. further Fraser,
Ptolemaic Alexandria 1.11 f.; Poseidippos 11.1 HE

19 _: Another papyrus list of epigrams, P.Vindob. G 40611, begins

but this papyrus, dating from the third cen-
tury B.C., cannot contain anything by Phil.; cf. H. Harrauer, "Epigrammincipit auf
218 P.Oxy. 3724v.19-v.29

einem Papyrus aus dem 3. Jh. v. Chr. Ein Vorbericht," Proc. XVIInt. Congr. Pap.
(Chico 1981) 49-53. This papyrus will be edited in full by P. Parsons and B. Kramer.
(Note that the occurrence of KOUXJ/OI; in this unknown epigram weakens the case
made above, on line 15.)
Parthenios, if a proper name rather than noun or adj., may refer to P. of Nicaea
(or Myrlea), Vergil's teacher, and hence someone known (if only by name) to Phil.
Or it may refer to Vergil himself, who Neapoli Parthenias vulgo appellatus sit
(Donatus Vita Verg. 11); see above, p. 19, n. 17.

20 Phil. 12.

21 N An erotic epigram is likely, as in the other epigrams where

a Nikarete appears: Asklepiades 3 HE (AP5.153), Dioskorides39 HE (AP7.166),
Nikarchos 2 HE (AP 6.285).

22 Parsons is right to compare 6, but Phil, does not there

give up all parties, and he will continue to wear wreaths; see the commentary. If by
Phil, and in line with 6 would presumably receive the proper modifica-
tion.

23 Either (i) "Lend me your ear," as in Hipponax 118.5

West Plato, Rep. 53 la; or (ii) Conceivably directions to a house-
hold slave; cf. Phil. 28. If not ~ , then Holford-Strevens offers as a possibil-
ity

24 In addition to 1, cf. 3.3-5:

25

26.

27

28 ,. _ ., _ _, This almost certainly was meant to recall Asklepiades 1 HE

(AP 5.169.1) and may therefore also form part of
a priamel culminating in something sweeter or sweetest of an erotic nature.

29 "Latin writers . . . associate this Greek garment [= hima-

tion\ with Greek practices—comedy, philosophy, immorality" (Parsons), any one
or more of which can easily be imagined to have been the subject of a Philodemean
epigram. If a self-contained phrase, it could mean "the women with the little
pallium."
Another possibility is that the reference is to the -pallium, referring to bed-
covering (cf. Ov.Am. 1.2.2nequein lectopallia nostrasedent). Thus, e.g., "herfrom
the bedding (having leaped,. . .)" could begin the description of a husband return-
ing home unexpectedly, such as Horace describes soon after refering to Phil. (Sat.
 P.Oxy. 3724 v.29-vi.10 219

1.2.127ff, esp. 129 f. vae pallida lecto \ desiliat mulier). In which case, the poem
begun here could have provided the source for Horace's Gallis, hanc Philodemus
ait. Cf. Gigante 141 f.

30

The deletions in this list seem designed to avoid duplication,

so that this must surely be the beginning of Phil. 19 rather than of Antipater Sid.
22 HE (AP 7.172); see above, on col. iii. 7.

32 A A statue of Aesop is described in Agathias 16 Viansino (AP

16.332). Holford-Strevens suggests A

col. vi

1 Drugs/poisons appear often in the Anthology, five in erotic

epigrams, including 37.2

2 Parsons offers Latin Rulla conceivably an otherwise

unattested woman in the family of P. Servilius Rullus, tr. pl. 63) or bulla
as bare possibilities. Obbink suggests which could refer to a young girl
("chick") aspullus does to a boy (OLD s.v. 2). Note also late Latin rulla, 'plow staff.'

3 This vocative is a late form of address, e.g., to God in the Greek

of Jews and Christians. Phil, could be using here a term which must have been fa-
miliar to him in Gadara. Cf. below, fr. 3.3. With , we could expect a supple-
mentary participle to follow; Parsons allows also for KCCUCTCU and Xj/auoai.

4 Phil. 21. See above on col. v. 15, and below on vii. 15.

5 ' or (perhaps better) ' (Parsons).

7 Repeated below, vii. 14, q.v.

9T Two frigid epigrams of Julian of Egypt addressed to a bronze

statue of Ikaros begin "Ii (API 107 f.; Schulte, pp. 112 ff.). Parsons also raises
the possibility of a reference to the Island in a festive context, as in Tib. 3.7.9 cunctis
Baccho iucundior hospes I Icarus.

10 : Ingredients for a party? is a likely supplement,

as in Stesich. 187 PMG Anakr. 434 PMG, Asklepiades 25
HE(AP5.181.2)
220 P.Oxy. 3724 vi.ll-vi.18

11 Of IT Parsons notes the similar phonetic spelling in the

papyrus of Phil. De Ira (e.g., , coll. xvi. 32, xix.ll, xxiii. 31, xxxii. 25), but
it is not uncommon in koine papyri.

12 See above on col. ii. 12. Gigante 136 suggests,

e.g., i. E. TOV epana, comparing Meleager 59 (AP5.96.1) i.

13

14 Philip 59 (API 1.3 6) begins but Parsons

says tha t es are not consistent with KaA,.

15

16 : A feminine form of (elsewhere only

Antipater Sid. 4 HE =AP6.160.5; restored at Bakch. 1.74) is possible, but Parsons
prefers

17

18 This seems to be the incipit of AP 5.145 [Asklepiades

12HE,9Knauer]:

PP17.116, f. 73v' Hi. 15-20

3 CP1 . P4PP1: ' C 5
Schneidewin: PP1 61 | ITPP1 C :P(IIutvid.):
.PI

Taran, Variation 73—77.

Gigante 134

Copied out in full in n col. i. 15-20 (where a corruption in the MSS is cleared up),
and its incipit included among the rest, this epigram may thus be by Phil, (as
Cameron and Luppe now believe), although it contains nothing in language or
quality to recommend this ascription. Nor are any of Phil.'s undisputed poems
homosexual. Furthermore, whereas Asklepiades sets other erotic poems in front of
houses (3, 13, 14, 42 HE =AP5.153,164, 167,189), Phil, does not. Probably even
more telling, for this andAP5.150, which is also among the incipits (col. iv. 28), is
that AP includes both in a long run of Meleagrean poems (5.134-215), where there
are no double ascriptions.
 P.Oxy. 3724 vi.18-vii.ll 221

There remains the possibility that the two incipits that match two of
Asklepiades' poems may have been from poems of Phil, which began similarly; cf.
Cameron, app. 7.
In v. 5 the papyrus vindicates Schneidewin's conjecture and also shows
that C's and derives from an attempt to heterosexualize a homosexual
poem, such as we occasionally find in Planudes' bowdlerizing, cf. Cameron, Greek
Anthology 353 f.

19 As Parsons points out, if in fact the soul is being addressed

in the vocative (—TI is also possible), there are only two other examples of this in
AP: Phil. 1, where note and 9.411 (Maccius 3 GP), where
note the plural verb

Col. vi ends here, leaving one-third of the column blank, although the papyrus is
undamaged. One also notes that there is room for at least one column to the left of
col. vii on the verso.

col. vii (verso)

1 See Parsons.

2 uoi is also possible. Perhaps a sepulchral epigram; cf. AP 7.601.2

(Julian of Egypt f. Schulte ad loc.), 7.602.5
(Agathias 23 Viansino)

3 ral epigram.

4| : See on col. ii.15.

5. Is the object once again something literary? Cf. iv.8.

6 A similar beginning at iv.25.

7 Phil. 10.

8 The verb (to buy victuals) can be proparoxytone, paroxytone, or

properispomenon. If the last, an order to a slave in an invitation epigram?

9 With this hapax, suggested by Lloyd-Jones, cf. (Max.

337). A less likely articulation is , Hesych.).

10 ii. 26 also begins with a negative oath. On KPOKEOVTO,

see iv.12.

11 Since = "mortal women" occurs in Kirke's

speech to Odysseus, in which she compares unfavorably the beauty of mortals to
222 P. Oxy. 3 724 vii. 1 l-vii. 25

that of goddesse;
5.212 £), it is possible that this epigram continues with a similar comparison.

12 is very tempting" (Parsons, point-

ing out the paleographical ditticultiesj.

13 Phil. 23, as ascribed by P (ci8r|Xov, PI).

14 See above, vi. 7. Probably -coq, although Parsons is doubt-

ful. The word is used of music by Lucian Dial. Deorum 7.4

15 Phil. 20. Parsons notes that "the scribe normally takes no

notice of" punctuation, but the space before falls between "voices" of the dia-
logue, and may well go back to Phil.'s autograph, where they would have served as
a reminder to change tone or pitch in reciting.

16 : Parsons reads , but an enclitic in position 3 jars, and

in the complete poems, Phil, allows movable nu to make position only once (9.4).
Thus, if this is in fact the reading (Obbink declares eoii not secure), either not by
Phil, or intentionally crude, like 22.

17 Phil. 28.

18

19 The last word can be and both gammas can be

taus, although sense argues for gammas. Cf, Phil. 10.1,2.

20 This may belong to the topos of the sexy older woman

(cf. 9): (i) Philinna is the name in Paulus Silentiarus' example of this genre; (ii) in
three of these poems, the woman, despite some falling off of youthful beauty, re-
tains her (see on 9.6).

21 PhU. 29.

22 (first or third person?) 'and

much else" (Parsons). The last character may be a nu or lambda + omicron:
bene iam sciebam (Holford-Strevens) is very likely.

23 ._ _ No known word is possible; perhaps, as Parsons suggests, a patro-

nymic or Greek adjective (such as the Lucretian Memmiadas or Scipiadas for the
Roman name Mucius.

24 Nu and mu are possible for the first doubtful letter.

25 Phil. 9.
 P.Oxy. 3724 viii. l-viii. 12 223

col. viii

1 _ (preferred by Obbink) or although neither adv. nor

adjective sits comfortably with the idea of rubbing.

2 ; Phil. 8.

3 An erotic epigram is very likely.

4 Cf. Phil. 12.7 . For the genitive with

, c£, e.g., Meleager43 (AP5.137.1) is also pos-
sible; least likely in would amuse.

5 Parsons adduces PI. Symp. 198d

6 Hedylion figures in an erotic epigram of Maccius 5 GP (AP

5.133).

7 . Rea suggests which (to me at

any rate) raises the possibility of a reference to Epicureans as porkers; cf. Intro.,
p. 16, n. 13.

8 This could have begun an address to a woman passing in

the street; cf. Phil. 20 and 21. For the syntax, cf. Rufinus 10 Page (AP 5.28.1)
with Page's n. ad loc.

9 : Phil. 26.

11 would be possible" (Parsons), or L (Obbink).

1] On cf. Phil. 27.1 (with commentary) and 17.

' (Parsons).

12 Parsons suggests another dative in the lacuna, perhaps

, which would be a tight fit. A dedicatory poem is very likely.

Frr. 2-3

Written by hand A, this too may contain incipits and, as is suggested by paragraphoi
setting off w. 9-12 (see on Fr. 2.3), at least one entire epigram. The likely equality
between two lines below with incipits in fr. 1 suggests that here too Phil, is the prin-
cipal or sole source.
224 P.Oxy. 3724

Fr.2

l]...o...[

3 There seems to be the tail end of a paragraphos below the alpha,

which would suggest that the next lines, fitting one of the incipits from Fr. 1, con-
tained a complete poem. But only after only three lines (obviously an impossible
number for an epigram) we may have another incipit, where again it is preceded by
a paragraphos. If we are reading the paragraphoi correctly (which is not certain,
especially since some may be lost), this fragment may contain a mixture of incipits
and complete poems, perhaps as follows:

1-3, three incipits or the last three lines of an epigram

4-5, a poem of one distich
6, an incipit
7-8, a poem of one distich
9-12, one poem of four lines or four incipits
13, an incipit

Although such a combination of poems and incipits may well seem unlikely, it re-
ceives some support from the inference that the poem beginning on v. 4 is followed
by the poem beginning on v. 7, just as these two incipits are contiguous in Fr. 1.

4 =Fr.liii. 18?.,.

5 The second line of the poem on w. 4-5?

6 _. _ _ Parsons suggests e.g.

7 Phi seems possible, although Parsons is doubtful; in which case, cf. fr. 1
iii. 19 and see above on v. 3, which reinforces the identification.

8 The first unclear letter may be either lambda or nu; for the latter Par-
sons compares Maccius 10 GP (AP 9.403.1) But as suggested above it
may be the second line of a poem.

9-12: A complete poem? (Parsons's supplements except for 10 < >)

10

13 _: Followed by the foot of the column.

 P.Oxy. 3724 225

Fr.3

3 Obbink compares Fr. 1 vi.3

4
This page intentionally left blank
 TESTIMONIA
ad Philodemum pertinentia

This collection of sources contains texts where Philodemus is clearly (section A) or

quite likely (B) referred to by others. A very brief selection is offered of passages
from the papyri where Philodemus seems to refer to himself (C). These are limited
to those alluded to in the Introduction. Excluded are all epigrams that have been
interpreted autobiographically. Annotation is minimal.

T 1 Cicero De Finibus 2.119

Quae cum dixissem, "Habeo," inquit Torquatus, "ad quos ista referam, et, quamquam
aliquid ipse poteram, tamen invenire malo paratiores." "Familiares vestros, credo,
Sironem dicis et Philodemum, cum optimos viros turn homines doctissimos." "Recte,"
inquit, "intelligis."
When I had finished, Torquatus said "Although I would be able to respond myself, I
prefer to refer these matters [sc. your arguments] to those who are more able than I
am." "You are, I believe, speaking of your colleagues, the finest and most learned Siro
and Philodemus." "You undestand me perfectly."

Composed in 45 B.C., the dramatic date of the dialogue is 50 B.C.; its setting is Cicero's
Cumanum estate near Naples. Cf. D. Delattre, "Philodeme dans la correspondance
de Ciceron," BAGB (1984) 28 f.

227
228 Testimonia

T 2 Cicero In Pisonem 68-72, 74 (ed. Nisbet)

68 Dicet aliquis "uncle haec tibi nota sunt?" Non me hercules contumeliae causa
describam quemquam, praesertim ingeniosum hominem atque eruditum, cui generi esse
ego iratus ne si cupiam quidem possum. Est quidam Graecus [sc. Philodemus; v. T 3]
qui cum isto vivit, homo, vere ut dicam—sic enim cognovi—humanus, sed tarn diu quam
diu aut cum aliis est aut ipse secum. Is cum istum adulescentem iam turn hac distracta
fronte vidisset, non fastidivit eius amicitiam, cum esset praesertim appetitus; dedit se in
consuetudinem sic ut prorsus una viveret nec fere ab isto umquam discederet. Non apud
indoctos sed, ut ego arbitror, in hominum eruditissimorum et humanissimorum coetu
loquor. Audistis profecto dici philosophos Epicureos omnis res quae sint homini
expetendae voluptate metiri; rectene an secus nihil ad nos, aut si ad nos, nihil ad hoc
tempus; sed tamen lubricum genus orationis adulescenti non acriter intellegenti et saepe
praeceps. 69 Itaque admissarius iste, simul atque audivit voluptatem a philosopho tanto
opere laudari, nihil expiscatus est: sic suos sensus voluptarios omnis incitavit, sic ad illius
hanc orationem adhinnivit, ut non magistrum virtutis sed auctorem libidinis a se ilium
inventum arbitraretur. Graecus primo distinguere et dividere, ilia quern ad modum
dicerentur; iste, "claudus" quern ad modum aiunt "pilam," retinere quod acceperat,
testificari, tabellas obsignare velle, Epicurum disertum decernere. Et tamen dicit, ut
opinor, se nullum bonum intellegere posse demptis corporis voluptatibus. 70 Quid
multa? Graecus facilis et valde venustus nimis pugnax contra senatorem populi Romani
esse noluit. Est autem hie de quo loquor non philosophia solum sed etiam ceteris studiis
quae fere ceteros Epicureos neglegere dicunt perpolitus; poema porro facit ita festivum,
ita concinnum, ita elegans, nihil ut fieri possit argutius. In quo reprehendat eum licet si
qui volet, modo leviter, non ut improbum, non ut audacem, non ut impurum, sed ut
Graeculum, ut adsentatorem, ut poetam. Devenit autem seu potius incidit in istum eodem
deceptus supercilio Graecus atque advena quo tot sapientes et tanta civitas; revocare se
non poterat familiaritate implicatus, et simul inconstantiae famam verebatur. Rogatus
invitatus coactus ita multa ad istum de isto quoque scripsit ut omnis hominis libidines,
omnia stupra, omnia cenarum genera conviviorumque, adulteria denique eius
delicatissimis versibus expresserit, 71 in quibus si qui velit possit istius tamquam in
speculo vitam intueri; ex quibus multa a multis et lecta et audita recitarem, ni vererer ne
hoc ipsum genus orationis quo nunc utor ab huius loci more abhorreret; et simul de
ipso qui scripsit detrahi nil volo. Qui si fuisset in discipulo comparando meliore fortuna,
fortasse austerior et gravior esse potuisset; sed eum casus in hanc consuetudinem scribendi
induxit philosopho valde indignam, si quidem philosophia, ut fertur, virtutis continet
et offici et bene vivendi disciplinam; quam qui profitetur gravissimam mihi sustinere
personam videtur. 72 Sed idem casus ilium ignarum quid profiteretur, cum se
philosophum esse diceret, istius impurissimae atque intemperantissimae pecudis caeno
et sordibus inquinavit. . . 74 Quaere ex familiari tuo, Graeco illo poeta: probabit genus
ipsum et agnoscet neque te nihil sapere mirabitur.

68 Someone will no doubt ask, "How do you come to know all this?" Well, I do not
propose to describe any individual in such a manner as to insult him, especially if he
be a man of parts and learning, a class with which I could not be angry, even if I wished.
There is a certain Greek [sc. Philodemos; cf. T 3] who virtually lives with him, a man
whom, to tell the truth, I have found to be a very gentlemanly fellow, at any rate as
long as he is in other company than Piso's, or is by himself. This man met our young
friend Piso who even then went about with eyebrows raised, and was not averse to his
friendship, especially as the other eagerly sought him; he so far gave himself up to his
 Testimonia 229

company that he absolutely lived with him and scarcely ever left his side. I am speak-
ing not to an ignorant audience, but, as I think, in an assembly of learned and accom-
plished gentlemen. You have of course heard it said that Epicurean philosophers as-
sess the desirability of anything by its capacity to give pleasure—whether rightly or
wrongly is no concern of ours, or at any rate not relevant to the present issue—it is,
however, a dangerous argument to put before a young man of only moderate intelli-
gence, and one that often leads to disaster. 69 Accordingly, as soon as that stud heard
pleasure praised so highly by so great a philosopher, he did not pick and choose; he so
stimulated all his pleasurable sensations, and raised such a whinnying to welcome his
friend's arguments, that he plainly thought he had found in the Greek not a professor
of ethics but a master of the art of lust. The Greek at first drew distinctions as to the
meaning of the precepts; but, as the proverb says, "a cripple has got the ball [sc. but
cannot run with it]"; Piso was prepared to bear witness as to what he had received,
and to put a seal on the matter, and would have it that Epicurus was an eloquent fel-
low; and indeed Epicurus does, I believe, assert that he cannot conceive any good apart
from bodily pleasure. 70 To make a long story short, the Greek was far too charming
and complaisant to have any notion of standing up to a Senator of the Roman people.
Now the Greek of whom I speak is polished not only in philosophy but also in
other accomplishments which Epicureans are said commonly to neglect; he further-
more composes poetry so witty, neat, and elegant, that nothing could be cleverer.
Anyone who wishes is at liberty to find fault with him for this; but let him do so gen-
tly, not as though with a low and bare-faced rogue, but as with a poor little Greek, a
parasite, a poet. When he came upon Piso, or rather fell in with him, he was beguiled,
a Greek in a strange land as he was, by the same savage scowl as has beguiled so many
sages and so great a society as our own. Once in the toils of friendship, there was no
drawing back for him, and, what was more, he wished to avoid the reproach of fickle-
ness. In response to request, invitation, pressure, he wrote reams of verse to Piso and
about Piso, sketching to the life in lines of perfect finish all his lusts and immoralities,
all his varied dinners and banquets, all his adulteries; 71 and in these poems anyone
who wishes can see the fellow's life reflected as in a mirror. I would read you a copious
selection from these (they have often been read and listened to before), were it not
that I am afraid that, even as it is, my present subject is out of keeping with the tradi-
tions of this place; and at the same time I do not wish to cast any slur upon the char-
acter of their author. Had he been luckier in the sort of pupil he found, he might have
turned out a steadier and more irreproachable character; but chance led him into a
style of writing which was unworthy of a philosopher, if, that is to say, philosophy is
correctly described as comprising the whole theory of virtue and duty and the good
life; and the man who professes that seems to me to have taken me to have taken upon
himself the most responsible of functions. 72 He did but imperfectly apprehend what
he was professing in calling himself a philosopher, and chance too defiled him with
the mud and filth of that bestial and unbridled monster. ... 74 [Cicero is defending
his own poetry against Piso's criticism of it] Ask your friend the Greek poet; he will
pass my figure of speech and recognize it, and will feel no surprise at your lack of dis-
cernment. (Transl. N. H. Watts, adapted)

In addition to his notes ad locc., seeNisbet's appendices 3 ("Piso andPhilodemus")

and 4 ("Piso and the Villa of the Papyri"): R. G. M. Nisbet, M. Tulli Ciceronis: In
L. Calpurnium Pisonem Oratio (Oxford 1961). Cf. also M. Gigante, "II ritratto di
Filodemo nella Pisoniana" in RF2 35-53.
230 Testimonia

T 3 Asconius ad Cic. In Pis. 68

Philodemum significat qui fuit Epicureus illa aetate nobilissimus, cuius et poemata
sunt lasciva.
Cicero means Philodemos, who was the finest Epicurean of that time, and who wrote
lascivious poems.

T4 Horace Serm. 1.2.119 ff.

parabilem amo venerem facilemque.
illam "post paulo," "sed pluris," "si exierit vir,"
Gallis, hanc Philodemus ait sibi, quae neque magno
stet pretio neque cunctetur cum est iussa venire.
I like a woman who is available and easy. The woman who says "later," "more money,"
or "when my husband has left" ? As Philodemos says, let the Galli have her: that woman
is his who charges little and doesn't delay when called.

This poem is imbued with Epicurean coloring derived from the prose works of
Epicurus and his school. By capping his general, philosophical, point with a refer-
ence to one or more of Phil.'s poems, Horace hints at a relationship between Epi-
curean/Philodemean poetics and the epigrams. Cf. Tait 67; Q. Cataudella,
"Filodemo nella Satira 12 di Orazio," PP5 (1950) 18-31; M. Gigante, Orazio. Una
misuraper I'amore: Lettura della satira seconda delprimo libra (Venosa 1993). For
an attempt to reconstruct the epigram of Phil, alluded to here, cf. [38].

T5 Philip 1(AP 4.2.8 f.)

Philodemos will shine (in my poetic Garland) like marjoram.

Why Phil, should be associated with marjoram, which Meleager in his introduc-
tory poem had linked with Polystratos, is a mystery. See Gow-Page ad loc.

T 6 Strabo 16.2.29

From Gadara come Philodemos the Epicurean, Meleager, Menippos the jocoserious,
and our contemporary Theodores the rhetor.

Cf. Introduction, "Life"; T. Dorandi, "La patria di Filodemo," Philologus 131 (1987)
254-256.
 Testimonia 231

T 7 Diogenes Laertius 10.3

10.24 .
10.3 (Epicurus' three brothers studied philosophy with him,) as Philodemos the Epi-
curean says in the tenth book of his Syntaxis of Philosopher. 10.24 (Epicurus's succes-
sor Polyainos was just and amiable,) as Philodemos and his circle say.

For the similarities between Diogenes' Lives and Philodemos' Syntaxis (both of
which end with a tenth book on the life of Epicurus), cf. J. Mejer, Diogenes Laertius
and his Hellenistic Background (Wiesbaden 1978) 69-74; M. Gigante, "Biografia
e dossografia in Diogene Laerzio," Elenchos 1 (1986) 25-34; id. Philodemus in
Italy 21,

T 8 Souda, s.v.

(= Aelian, fr. 40 Hercher)

(words for "fine," "penalize"). The citizens of Himera penalize Philodemos with exile
in addition to confiscation.

should probably be deleted as a gloss (Holford-Strevens). See Introduc-

tion, "Life," for related texts derived from Aelian. For the expulsion of other Epi-
cureans, cf. Obbink, GRBS30 (1989) 204 n. 59; ibid. Philodemus on Piety (Oxford
1996) 14, n. 4.

T 9 Ambrose, Epist. 14 (63), 13 Zelzer (CSEL 82/83, 241 f. = Epic,

fr. 385a U, p. 356)

Atque hie [sc. Epicurus] quam alienus a vero sit etiam hinc deprehenditur quod
voluptatem in homine deo auctore creatam asserit principaliter, sicut Philodemus
\Maurini: Filominus codd.] eius sectator in epitomis suis disputat et huius allegat Stoicos
esse auctores sententiae.
And how divorced Epicurus is from the truth can be seen from his assertion
that pleasure was created in man by god from the beginning, just as his follower
Philodemos argues in his summaries, alleging that the Stoics are responsible for this
view.

Cf. W. Liebich, "Ein Philodem-Zeugnis bei Ambrosius," Philologus 98 (1954) 116-

131; T. Dorandi, "Filodemo: Orientamenti della ricerca attuale," AN'RW 36.4
(1990) 2354; D. Obbink, Philodemus on Piety (Oxford 1996) 78-80.
232 Testimonia

T 10 Lucreti Vita Borgiana, p. 6 Masson

C. Memmio Epicureo dicavit opus. Romani autem Epicurei hi memorantur praecipui:
. . . L. Calphurnius Piso Frugi qui Philodemum [PolidemumMS] audivit,... Vergilius
Maro Sironis [Scyronis MS] auditor. .. .
Lucretius dedicated his work to Gaius Memmius. The most notable Epicureans among
the Romans:. . . Lucius Calpurnius Piso Frugi, who was a student of Philodemos,. . .
Vergil who was a student of Siro.

This Piso, born ca. 88 B.C., could easily have known Phil., but the value of the Life
has been questioned. A. Rostagni, "Ricerche di biografia lucreziana, 2: La Vita
Borgiana," ScrittiMinori2.2 (Turin 1956) 121-147, argues that it contains material
deriving from Probus; similarly, C. Bailey, Lucretius (Oxford 1947) 1.2 finds that
"it cannot be entirely discarded." On the other hand, M. F. Smith, Lucretius: The
Man and his Mission (thesis, Dublin 1965) 23-26 argues that it is a product of
humanist learning and "of no importance whatsoever."
For the complete text of this life, see J. Masson, Lucretius: Epicurean andPoet.
Complementary Volume (London 1909) 4-6.

B
Til Catullus 47
Porci et Socration, duae sinistrae
Pisonis, scabies famesque mundi,
vos Veraniolo meo et Fabullo
verpus praeposuit Priapus ille?
vos convivia lauta sumptuose
de die facitis, mei sodales
quaerunt in trivio vocationes?
Sokration and Porcius, Piso's two left-hand men, scabs and famine to the world, does
that prickless Priapus prefer you to my buddies Veraniolus and Fabullus? Does he
serve you sumptuous food and drink all day, while my friends wander the streets look-
ing for formal invitations?

Socration = Philodemos? Cf. Introduction, pp. 23 f.; G. Friedrich,Catulli Veronensis

Liber (Leipzig 1908) 228; Nisbet, Cic. In Pis., pp. 180-182.

T 12 Cicero Or post Red. in Sen. 14 f.

Cum vero etiam litteris studere incipit [sc. Piso] et belua immanis cum Graeculis
philosophari, turn est Epicureus, non penitus illi disciplinae quaecumque est deditus,
sed captus uno verbo voluptatis. Habet autem magistros non ex istis ineptis, qui dies
 Testimonia 233

totos de officio ac de virtute disserunt, qui ad laborem, ad industriam, ad pericula pro

patria subeunda adhortantur, sed eos qui disputent horam nullam vacuam voluptate
esse debere: in omni parte corporis semper oportere aliquod gaudium delectionemque
versari. His utitur quasi praefectis libidinum suarum; hi voluptates omnes vestigant
atque odorantur; hi sunt conditores instructoresque convivi; idem expendunt atque
aestimant voluptates sententiamque dicunt et iudicant quantam cuique libidini
tribuendum esse videatur.

But when Piso began to study the liberal arts, when this monster began to philoso-
phize with Greeklings, then he is an Epicurean. Nor was he deeply involved with this
way of life (whatever it is) to which he has devoted himself; rather he was caught up by
the single word "pleasure." He does not, however, have as teachers those "unworldly
fools" who spend their days on the subjects of duties and virtues, those who urge one
on to hard work and facing danger for one's country; rather, he chooses those who
argue that no hour of the day should be free of pleasure, and that joy and delight should
spread through every limb. These are the men he employs as the superintendents of
his libidinous pleasures; these are the ones who track down and smell out every form
of pleasure; these provide the basis and guidance for his feasts. These same men dis-
pense and weigh out his pleasures, and they lay down the law, judging how much should
be allotted to each pleasure.

Written Sept. 57. Cicero distinguishes two classes of Epicureans: the serious teachers
and the strong hedonists. Since the general picture of Piso's adherence to Epicure-
anism is roughly comparable to that found in Pis. (T 2), where Phil, is singled out
as Piso's most notable teacher and flatterer, we must conclude that Phil, belongs to
the former class here, although we are also free to believe that the picture is a greatly
distorted one.

c
T13 fr. 45.8-11 Olivieri

"The basic and most important [principle] is that we will obey Epicurus, according to
whom we have chosen to live" (tr. Asmis).

T 14 P.Herc. 1005 col. 14. 4-13. Angeli = Zenon fr. 11 Angeli-

Colaizzo
234 Testimonia

Epicureans, with whom I [verb] and with whom I was an obedient follower of Zeno
when he was alive and his tireless laudator after his death, especially of all his virtues
(found) in the "vaunts and ecstasy" from/of Epicurus.

Cf. K. v. Fritz, "Zeno von Sidon," RE 10A (1972) 122-124. The fragments of Zeno
are collected in A. Angeli and M. Colaizzo, CErc 9 (1979) 47-133, but the text of
P.Here. 1005 has now been reedited in A. Angeli, Filodemo: AgliAmici di Scuola
(P.Herc. 1005) (Naples 1988). Since ' is found elsewhere only in Pindar,
N 9.7, I suspect that Phil, is here
quoting a poetical tag, which could come, e.g., from another poem of Pindar (and
analyzed as either "- cr cho" or "- e d", with resolution in dactylo-epitrite as in Pi.
N. 10.32).

T 15 P.Herc. 312, col. 14 ed. Gigante

Korte: Cronert Gigante: Cronert Gigante:

Cronert ". S c h w a r t z I r o n e r t
Gigante: Cronert

He decided to return with us to Naples and to dearest Siro and his way
of life there and to engage in active philosophical discourse and to live
with others in Herculaneum.

Cf. Cronert, Kolotes undMenedemos 125-127 for fuller context; Gigante, A&R 28
(1983) 36 £; id., Catalogo 124 £; Capasso, CErc 19 (1989) 221.
 BIBLIOGRAPHY

Reference Works

Berkowitz, L., and K. A. Squitier, Thesaurus Linguae Graecae Canon of Greek Authors. (3rd
ed.) (New York 1990), s.v. Philodemus. Arranged by title, a comprehensive list of
sources of texts.
Capasso, M. "Primo supplemento al catalogo dei papiri ercolanesi," CErc 19 (1989) 193-
264. Supplement to Gigante.
Erler, M. "Philodem aus Gadara," in H. Flashar (ed.), Die Philosophic der Antike4.1 Die
hellenistische Philosophic (Basel 1994) 289-362. A bibliographical survey.
Gigante, M. Catalogo dei Papiri Ercolanesi (Naples 1979). Detailed description of the papyri
(arranged by P.Herc. no.) plus full bibliography, now supplemented by Capasso, above.
Usener, H. Glossarium Epicureum. Ed. M. Gigante and W. Schmid. Rome 1977. An Epicu-
rean dictionary, based on Epicurus and the Herculaneum papyri.
Vooijs, CJ.Lexicon Philodemeum. Vol. 1, Purmerend 1934; vol. 2, Amsterdam 1941. Though
somewhat dated, a still useful list of Phil.'s prose vocabulary, especially since the pa-
pyrus texts of Philodemus have not yet been entered in the TLG data bank.

Editions of Philodemos' Epigrams

Brunck, R. F. P.Analecta Veterum Poetarum Graecorum. Strasburg 1772-1776. Vol. 2.83-

91 (text), 3.141-145 (notes).
Jacobs, C. F. W. Anlhologia Graeca sive Poetarum Graecorum Lusus ex RecensioneBrunckii.
13 voll. Leipzig 1794-1814. Vol. 2.70-79 (text), 7.211-250 (notes).

235
236 Bibliography

Kaibel, G. Philodemi Gadarensis Epigrammata. Greifswald 1885.

Gow, A. S. F., and D. L. Page, The Greek Anthology: The Garland of Philip (Cambridge
1968). Vol. 1.351-369 (text), 2.371-400 (notes). [GP]
Gigante, M. Filodemo. Epigrammi Scelti. 2d ed. Naples 1989.

Editions of Epigrams

(from AP and elsewhere; in addition to those named in the last section)

Beckby, H. Anthologia Graeca. 2d ed. 4 voll. Munich 1967-1968. [Tusculum]
Dubner, J. F. Epigrammatum Anthologia Palatina cum Planudeis. 2 voll. Paris 1864-1872.
Del Re, R. Epigrammi greci. Roma 1970.
Geffcken, J. Griechische Epigramme. Heidelberg 1916. Repr. Hildesheim 1976.
Gow, A. S. F., and D. L. Page, The Greek Anthology: Hellenistic Epigrams. 2 voll. Cam-
bridge 1965. [HE]
Hansen, P. A. Carmina Epigraphica Graeca. 2 voll. Berlin 1983-1989. [CEG]
Huschke, I. G. Analecta Critica in Anthologiam Graecam. Jena and Leipzig 1800.
Jacobs, C. F. ^.Anthologia Graeca. 3 voll. Leipzig 1813-1817. [In AP order plus Planudea]
Kaibel, G. Epigrammata Graeca ex lapidibus collecta. Berlin 1878.
Lumb, T. W. Notes on the Greek Anthology. London 1920.
Page, D. L. further Greek Epigrams. Cambridge 1981. [FGE]
Paton, W. R. The Greek Anthology. 5 voll. London 1916-1918. [Loeb]
Peek, W. Griechische Versinschriften. I. Grahepigramme. Berlin 1955. [GVI]
Preger, T. Inscriptions Graecae tnetricae ex scriptoribus praeter Anthologiam collectae. Leipzig
1891.
Preisendanz, C. Anthologia Palatina. 2 voll. Leiden 1911. [Facsimile of P]
Reiske, J. J. Anthologiae Graecae a Constantino Cephala conditae libri tres. Leipzig 1754.
New ed. Oxford 1766.
Speyer, W. Epigrammata Bobiensia. Leipzig 1963. [BT]
Stadtmiiller, H. Anthologia Graeca. Epigrammatum Palatina cum Planudea. 3 voll. Leipzig
1894-1906. [BT; only to AP 9.563]
Sternbach, L. Anthologiae Planudeae Appendix Barberino-Vaticana. Leipzig 1890.
Waltz, P. et al. Anthologie grecque. Paris 1928- . [Bude]

Separate Texts and Commentaries

or Monographs on AP Authors

AGATHIAS

G. Viansino, Agazia Scolastico: Epigrammi: Testo, traduzione e commento, Milan 1967.

ANTIPATER OF SIDON AND ANTIPATER OF THESSALONICA

Setti, G. Studi sulla Antologia greca: Gli epigrammi degli Antipalri. Turin 1890.
Waltz, P. De Antipatro Sidonio. Bordeaux 1906.
Weigand, G. De Antipatris Sidonio et Thessalonicensipoetis epigrammaticis. Breslau 1840.
 Bibliography 237

ANTIPHILOS
K. Miiller, Die Epigramme des Antiphilos von Byzanz. Berlin 1935.

ANYTE

M. J. Baale, Studia in Anytes poetriae vitam et carminum Reliquias. Diss. Harlem 1903.
D. Geoghegan, Anyte: The Epigrams, A Critical Edition with Commentary. Rome 1979.

ARCHIAS

T. Reinach, De Archia Poeta. Diss. Paris 1890.

M. ARGENTARIUS

S. G. P. Small, "Marcus Argentarius: A poet of the Greek Anthology," YCS 12 (1951)

65-145.

ASKLEPIADES

O. Knauer, Die Epigramme des Asklepiades. Wiirzburg 1935. Repr. in Taran (ed.) 2.

JULIAN

H. Schulte, Julian von Agypten. Trier 1990.

KALLIMACHOS

L. Coco, Callimaco. Epigrammi. Manduria 1988.

L. A. de Cuenca y Prado, Calimaco. Epigramas. Madrid 1974-1975.
G. Zanetto &P. Ferrari, Callimaco. Epigrammi. Milan 1992.

KKINAGORAS

M. Rubensohn, CrinagoraeMytilenaeiEpigrammata. Berlin 1888.

LEONIDAS

E. R. Bevan, The Poems ofLeonidas ofTarentum. Oxford 1931.

J. Geffcken, Leonidas von Tarent. Leipzig 1896.
M. Gigante, L'edera diEeonida. Naples 1971. (Not an ed.)
B. Hansen, DeEeonida Tarentino. Leipzig 1914.
R. Labellarte, Leonida di Taranto: Antologia di epigrammi. Bari 1969.

LUCIAN

G. Setti, Gli epigrammi di Euciano. Turin 1892.

LUCILLIUS

Rozema, B. J. Lucillius the Epigrammatist: Text and Commentary. Diss. Wisconsin 1971.

MELEAGER

J. Clack, Meleager: The Poems. Wauconda, 111. 1992.

G. Graefe, Meleagri Gadareni Epigrammata. Leipzig 1811.
238 Bibliography

G. Guidorizzi, Meleagro. Epigrammi. Milan 1992.

W. Headlam, Fifty Poems ofMeleager. London 1890.
J. C. F. Manso, Meleagri Reliquiae. Jena 1789.
H. Ouvre, Meleagre de Gadara. Paris 1894.
C. Radinger, Meleagros von Gadara. Innsbruck 1895. Repr. in Taran (ed.) 2.

MNASALKES

W. Seelbach, Die Epigramme des Mnasalkes von Sikyon und des Theodoridas von Syrakus.
Wiesbaden 1964.

PALLADAS

A. Franke, De Pallada epigrammatographo. Leipzig 1899.

PAULUS SILENTIARUS

G. Viansino, Paolo Silenziario. Epigrammi. Turin 1963.

POSEIDIPPOS

G. Bastianini and C. Gallazzi, Posidippo. epigrammi. Milan 1993.

P. Schott, Posidippi epigrammata. Diss. Berlin 1905.

RUFINUS

D. L. Page, The Epigrams ofRufinus, Edited with Introduction and Commentary. Cambridge
1978.

SIMIAS

H. Frankel, De Simia Rhodio. Diss. Gottingen 1915.

SIMONIDES

M. Boas, De epigrammatis Simonideis. Groningen 1905

A. Hauvette, De I'authenticite des epigrammes de Simonide. Paris 1896.

THEOKRITOS

R. J. Smutny, The Text History of the Epigrams of Theocritus. Berkeley 1955.

THEODORIDAS

See Mnasalkes.

General Studies
Asmis, E. "Philodemus' Epicureanism," ANRW 236.4 (1990) 2369-2406.
Capasso, M. Manuale di papirologia ercolanese. Lecce 1991.
 Bibliography 239

Cronert, W. Kolotes undMenedemos. Leipzig 1906; repr. Amsterdam 1965.

Dorandi, T. "Filodemo: Gli orientamenti della ricerca attuale,ANRW2.36.4 (1990) 2328-
2368.
Gigante, M. Ricerche Filodemee. 2d ed. Naples 1983. [RF 2 ]
. Filodemo in Italia. Florence 1990. Eng. trans. D. Obbink, Philodemus in Italy (Ann
Arbor 1995).
Jufresa, M. "Sobre Filodemo," BIEH 10 (1976) 15-73.
Philippson, R. "Philodemos," RE 19 (1938) 2444-2482. Repr. in Studien zu Epikur und
den Epikureern (Hildesheim 1983) 229-248.

Poetic Theory

Romeo, C. "Filodemo: La 'poetica'," in Studi sull'epicureismo greco e romano

offerti a Marcello Gigante. Naples 1983. 2.565-583. (This bibliographical survey is the
proper starting point to learn about Philodemus' poetical theory. Listed below are a
very few highlights from earlier literature and a more complete listing of recent con-
tributions. Cf. also Dorandi [above], pp. 2341 ff.)

Asmis, E. "Philodemus' poetic theory and On the Good King According to Homer," CA 10
(1991) 1-45.
. "Epicurean poetics," BACAP7 (1991 [1993]) 63-93. Repr. in Obbink 15-34.
Barra, G. "Osservazioni sulla 'poetica' di Filodemo e di Lucrezio," Annalid. Fac. diLett. e
Filos. d. U. diNapoli20 (1977-1978) 87-104.
Dorandi, T. "Dichtender Philosoph und philosophierender Dichter: Das literarische
Schaffen des Epikureers Philodem von Gadara," W] 18 (1992) 183-193.
Greenberg, N. A. The Poetic Theory of Philodemus. Diss. Harvard 1955. Repr. New York
1990.
Obbink, D. (ed.) Philodemus and Poetry. New York 1995.
Pace, N. Problematiche di poetica in Filodemo di Gadara. Diss. Milan, 1992.
Sbordone, F. "La poetica di Filodemo," Maia 36 (1984) 17-19.
Schenkeveld, D. M. "OIKPITIKOI in Philodemus, "Mnemosyne, 4th ser. 21 (1968) 176-214.
Sider, D. "Commentary on Asmis," BACAP7 (1991 [1993]) 94-105. Revised version in
Obbink 35-41.
. "The Epicurean philosopher as Hellenistic poet," in Obbink 42-57.
l
Summers, A. T .Philodemus andHorace's "Ars Poetica". Diss. Urbana 1995.
Walsh, G. B. "Philodemus on the terminology of Neoptolemus," Mnemosyne, 4th ser. 40
(1987) 56-68.

Philodemos' Poetry and Its Influence

on and Echoes in Latin Poetry

[Numbers in square brackets indicate the epigrams treated.]

Barra, G. "Filodemo di Gadara e le lettere latine," Vichiana, n.s. 2 (1973) 247-260.

Braga, D. Catullo e ipoetigreci (Messina and Florence 1950).
240 Bibliography

Cataudella. Q. "Filodemo nella Satira I 2 di Orazio," PP5 (1950) 18-31.

Cavallini, E. APX21,ls. (Philod.)," Mus.Crit. 15-17 (1980-1982) 164 f. [8]
Chardon de La Rochette, S. "Dissertation sur deux epigrammes grecques de Philodeme,"
in Melanges de critique et de philologie (Paris 1812) 196-222. [3, 27]
Cichorius, C. "Personliches in den Epigrammen des Philodemos von Gadara," inRomische
Studien (Leipzig 1922) 295-298. [8, 28]
Cohen, S. J. D. "The beauty of Flora and the beauty of Sarai," Helios 8.2 (1981) 41-53 [12]
Courtney, E. "Ovid and an epigram of Philodemus," LCM 15 (1990) 117 f. [12]
Delia Corte, F. "Orazio 'desaparecido'," Maia 42 (1990) 41 f.
Del Re, R. "Filodemo, poeta," MC6 (1936) 121-142.
De Stefani, C. "Note a quattro epigrammi deH'Antologia Greca," SIFC 89 (1996) 205-208.
[11]
De Witt, N. "Parresiastic poems of Horace," CP30 (1935) 312-319.
Edmunds, L. "The Latin invitation poem: What is it? Where did it come from?" AJP 103
(1982) 184-188. [27]
Falivene, M. R. "II codice di nella poesia alessandrina (alcuni epigrammi dellaAntologia
Palatina, Callimaco, Teocrito, Filodemo, il Fragmentum Grenfellianum," QUCC 37
(1981) 87-104. [23, 24]
. "Per l'interpretazione diAP 10, 21 (Filodemo)," QUCC 42 (1983) 129-142. [8]
Giangrande, G. "Konjekturen zur Anthologia Palatina," RM 106 (1963) 255-257. [32]
. "Sympotic literature and epigram," in L'epigramme grecque = Ell 14 (1968)
91-174. [6].
. "Erklarungen hellenistischer Stellen," GB 1 (1973) 137-148. [4, 6]
. "An epigram of Philodemus," Maia 25 (1973) 65 [12].
. "Gli epigrammi alessandrini come arte allusiva," QUCC 15 (1973) 7-31. [6, 28]
. "Vierunddreifiig hellenistische Epigramme," GB 1 (1978) 77 f. [34]
. "Quelques aspects de la technique litteraire des epigrammatistes alexandrins," MPL
5(1981)37-46. [6,8,23]
Gigante, M. "Virgilio fra Ercolano e Pompei," A R 28 (1983) 31-50. Repr. in Virgilio e la
Campania (Naples 1984) 67-92
. "Cercida, Filodemo e Orazio," in RF2 235-243.
. "Filodemo e Pisone: Da Ercolano a Roma," ASNP 15 (1985) 855-866. (repr. in
Filodemo in Italia 103-116, Eng. trans. 79-90). [27]
. "Gli epigrammi di Filodemo quali testimonianze autobiografiche. Filodemo a Atene:
La preghieraaglidei(AP 6,349)," in Filologia e forme letterarie. Studi. . . F. Delia Corte,
1 (Urbino 1987) 385-389. Repr. in Filodemo in Italia 63-68; Eng. trans. 49-52) [34]
, and M. Capasso, "II ritorno di Virgilio a Ercolano," SIFC 7 (1989) 3-6.
. "Filodemo tra poesia e prosa (A proposito di POxy. 3724)," SIFC 7 (1989) 129-
151. Also in Proc. of the XlXth Intern. Congress ofPapyrology. Cairo 2-9 September
1989 (Cairo 1992) 1.403-446. [ II ]
. "Filodemo nella Villa di Ercolano: Incontri al Belvedere (A.P. 1X412 e XI35)," in
A. Battitaet al. (edd.), Studi di Filologia in onore di G. Monaco (Palermo 1991) 1.421-
427. Repr. in Philodemus in Italy 69-79; Eng. trans. 53-61. [29, 28]
Griffiths, A. H. "Six passages in Callimachus and the Anthology," BICS 17 (1970) 32-43.
[4,5]
Grilli, A. "Filodemo," A&R 33 (1988) 1-6.
Hecker, A. Commentatio Critics de Anthologia Graeca (Leiden 1843) 37 [20], 47 f.[26] ,
50-52 [12], 93 f. [25], 346 [32],
Hendrickson, G. L. "An epigram of Philodemus and two Latin congeners, "AJP 3 9 (1918)
27-43. [6]
 Bibliography 241

Hiltbrunner, O. "Einladung zum epikureischen Freundesmahl," in Anttdosis: Festschrift

W. Kraus (Vienna 1972) 168-182. [27]
Jacobs, F. "Uber ein dem Philodemus beigelegtes Epigramm," in Vermischte Schriften 5
(Berlin 1834) 264-291. [38]
Kaibel, G. "Sententiarum liber primus," Hermes 15 (1880) 459 f. [3, 11, 34]
Kleve, K. "Lucrece, 1'epicurisme et 1'amour," Actes du VIII" Congres (Paris, 5-10 avril 1968)
(Paris 1969) 376-383.
Klinger, A. "Augusteer bei Philodem," RM 45 (1880) 172-177.
Landolfi, L. "Tracce filodemee di estetica e di epigrammatica simpotica in Catullo," CErc
12 (1982) 137-143. [27]
Lavagnini, B. Erotion: II Libra dell'amore alessandrino. Epigrammi di. . . Filodemo. Torino
1928.
Luck, G. "Trygonions Grabschrift (Philodemus, AP 7.222)," Philologus 100 (1956) 271-
286. [33]
. "Witz und Sentiment im griechischen Epigramm," in L'Epigramme grecque = EH
14 (1968) 387-408. [3]
Marcovich, M. "Catullus 13 and Philodemus 23," QUCC40 (1982) 131-138. Repr. in Studies
in Greek Poetry (Atlanta 1991) 193-199. [27]
Mariotti, S. "Da Platone agli Epigrammi Bobbiesi: Appunti su due temi epigrammatici
antichi," Studi Urbinati 41 N.S. B 1-2 (1967) 1071-1096. [2]
Maxwell-Stuart, P. G. "The nature of Anticrates. Anth. Graec. 11.318 (Philodemus),"
Hermes 106 (1978) 253 f. [31]
Merkelbach, R. "Philodem, Anth. Pal. 1X570 und die epilogischen Geschichten," RM 115
(1972) 219-222. [3]
Michels, A. K. and the satire of Horace," CP 39 (1944) 173-177.
Munoz Valle, I. "Horacio (Od. II 4), imitador de Filodemo (A.G. V 132)," CFC 1 (1974)
87-89. [12]
Nunn, T.P. "The fatal name: An epigram of Philodemus," Intern. J.Psychoan. 11 (1929)
97 f. [10]
Paton, W.R. "Anth. Pal. VII.222 (Philodemus)," CR 30 (1916) 48. [33]
Prinz, K. "Zu Horaz Sat. 12.121 und Martial Epigr. 1X32," WS34 (1912) 227-236. [8,11,17]
Reitzenstein, R. "Horaz und die hellenistische Lyrik," NJahrb 21 (1908) 81-102 (esp. 83).
Rossi, M. "Nota a Filodemo Ep. V G.P. (AP V 107)," Maia 33 (1981) 213 f. [23]
. "Unmotivo arcaico in Filodemo Ep. VGP (AP5 107)," Vickiana 10 (1981) 163-167.
[23]
Salanitro, G. "Varia Graeco-Latina," inStudi classed in onore di Q. Cataudella (Catania 1972)
2.498-501. [25]
Schmid, W. "Philodem als Dichter und als Philosoph: iiber eine Athetese Kaibels in AP
9.570," Acta Conventus XI Eirene, 21-25 Oct. 1968 (Warsaw 1971) 201-207 =
Ausgewahlte philologische Schriften (Berlin 1984) 267-274. [3]
Schulze, K. P. "Zur Erklarung des Horaz," BPbW36 (1916) 317-320.
Seidler, J. F. A. "Uber ein Epigramm des Philodemus," Berichte iiber die Verhandlungen
der koniglich sachsischen Gesellschaft der Wissenschaften zu Leipzig 1 (1846-1847
[1848]) 128-130. [12]
Sider, D. "Notes on two epigrams of Philodemus," AJP 103 (1982) 208-213. [19, 33]
. "The love poetry of Philodemus," AJP 108 (1987) 310-324.
. "Looking for Philodemus in P.Oxy. 54.3724," ZPE 76 (1989) 229-336. [II]
. "How to commit philosophy obliquely: Philodemos' epigrams in the light of his
Peri Parrhesias," in J. T. Fitzgerald et al. (eds.), Philodemus and the New Testament
World (Leiden, forthcoming).
242 Bibliography

Snyder, J. M. "The poetry of Philodemus the Epicurean," CJ 68 (1973) 346-353.

Stella, L. A. Cinque poeti dell'Antologia Palatina (Bologna 1949), 239-307.
Sternbach, L. Meletemata Graeca. Vienna 1886. [3, 10, 11]
Tait, J. I. M. Philodemus' Influence on the Latin Poets. Diss. Bryn Mawr 1941.
Thomas, R. F. " 'Death,' doxography, and the 'Termerian evil' (Philodemus Epigr. 27 Page
= A.P. 11.30)," CQN.s. 41 (1991) 130-137. [19]
Vries, G.J., de, "Notes on Greek Epigrams, "Mnemosyne, 4th ser. 23 (1970)30-32. [3,25]
. "Philodemus' poor lover again," Mnemosyne, 4th ser. 26 (1973) 179. [25]
White, H. "Two epigrams by Philodemus," in G. Giangrande (ed.), Corolla Londinensis
(Amsterdam 1981 = LSCP 8) 173-177. [11, 33]
Wilkinson, L. P. "Philodemus and poetry," G&R 2 (1932-1933) 144-151.
Wiseman, T. P. "Philodemus 26.3 G-P," CQ 32 (1982) 475 f. [33]
Wright, F. A. "Horace and Philodemus," AJP 42 (1921) 168 f. [19, 22]

Epigrams in General

Autore, O. Marziale e I'epigramma greco. Palermo 1937.

Bendorf, O. De Anthologiae Graecae epigrammatis quae ad artes spectant. Bonn 1862.
Bignone, E. L'epigramma greco: studio critico e traduzioni poetiche. Bologna 1921.
Brecht, F. J. Motiv- und Typengeschichte des griechischen Spottepigramms. Philologus
Supplbd. 22.2. Leipzig 1930.
Cameron, Alan. The Greek Anthology from Meleager to Planudes. Oxford 1993.
Degani, E. "L'epigramma," inStoria eciviltadeigreci. La cultura ellenistica:filosofia, scienza,
letteratura (Milan 1977) 266-299.
Degani, E. "L'epigramma," in G. Cambiano et al. (eds.), Lo spazio letterario della Grecia
antica 1.2 (Rome 1992) 197-233.
Dihle, A. (ed.),L'Epigramme grecque. (EH 14) Fondation Hardt. Vandoeuvres-Geneve 1968.
Ebert, J. Griechische Epigramtne auf Sieger an gymnischen und hippischen Agonen. Berlin
1972.
Ecker, U. Grabmal und Epigram. (Palingenesia 29). Stuttgart 1990.
L'Epigramme grecque. Fondation Hardt. See Dihle, A. (ed.).
Gabathuler, M. Hellenistische Epigramme au/Dichter. Diss. Basel, St. Gallen 1937.
Garrison, D. H. Mild frenzy: A Reading of the Hellenistic Love Epigram. Hermes Einzelschr.
41. Wiesbaden 1978.
Geffcken, J. "Studien zum griechischen Epigramm," Neue Jahrbucher fur das klassische
Altertum 20 (1917) 88-107. Repr. in part in G. Pfohl (ed.), Das Epigramm 21-55.
Gow, A. S. F. The Greek Anthology: Sources and Ascriptions. London 1958. [GA]
Gragg, F. A. "A study of the Greek epigram before 300 BC" PAAAS46 (1910-1911) 1-62.
Hartigan, K. The Poets and the Cities. Meisenheim 1979.
Herwerden, H., van, Studia critica in epigrammata graeca. Leiden 1891.
Hezel, O. Catull und das griechische Epigramm. Stuttgart 1932.
Kagi, P. Nachwirkung der alteren griechischen Elegie in den Epigrammen der Anthologie.
Zurich 1917. Repr. in part in Pfohl, Die griechische Elegie (Darmstadt 1972) 54-71.
Kock, B. De epigrammatum Graecorum dialectis. Diss. Gottingen 1910.
Lausberg, M. Das Einzeldistichon: Studien zum antiken Epigramm. Munich 1982.
Lier, B. Ad topica carminum amatorium symbolae. Stettin 1914. Repr. in Taran (ed.) 1.
Longo, V. L'epigramma scoptico greco. Genua 1967.
Mariotti, S. II5° libro dell'Antologia Palatina. Rome 1966.
 Bibliography 243

Pertsch, E. De Valeria Martiale Gmecorum poetarum imitatore. Diss. Berlin 1911.

Pfohl, G. (e.A.),Das Epigramm: ZurGeschichteeinerinschriftlichen undlitemrischen Gattung.
Darmstadt 1969.
Pircher, J. Das Lob der Frau im vorchrhtlichen Grabepigramm der Griechen. Comm.
Aenipontanae 26. Innsbruck 1979.
Prittwitz-Gaffron, E. von, Das Sprichwort im griechischen Epigramm. Giessen 1912. Repr.
in Taran (ed.) 1.
Rasche, W. De Anthologiae Graecae epigramtnatis quae colloquii formam habent. Munich
1910.
Reitzenstein, R. Epigramm undSkolion. Giessen 1893.
. "Epigramm," RE 6 (1907) 71-111.
Stella, L. A. Cinque poeti dell'Antologia Greca. Bologna 1949.
Taran, S. L. The Art of Variation in the Hellenistic Epigram. Leiden 1979.
(ed.). The Greek Anthology. 2 voll. New York 1987.
Tarditi, G. "Per una lettura degli epigrammatisti greci," Aevum Antiquum 1 (1988) 5-75.
Weinreich, O. Epigramm und Pantomimus. Heidelberg 1948.
Weisshaupl, R. Die Grabgedichte der griechischen Anthologie. Abh. d. Archaologisch-
epigraphischen Sem. d. Univ. Wien 7. Vienna 1889. Repr. in Taran (ed.) 1.
Wifstrand, A. Studien zur griechischen Anthologie. Lunds Universitets Arsskrift. N.F. Avd.
1. Bd. 23.3. Lund 1926. Repr. in Taran (ed.) 1.
This page intentionally left blank
 GENERAL INDEX

Actium, epic poem on 13 Asmis, E. 29 n.19 Catullus, and Phil. 24-25,

n.2 Astour, M. 181 37, 62, 64
adultery 116-117,139-141 astrology 172 Cavallini, E. 92-93
anaphora 92 asyndeton 65, 122, 127, 148, Chantraine, P. 90
Andromeda 109 177 chads. See grace, Graces
Andronikos 173 Atthis 214 cheese 167
Antigenes 168, 207 Attis 179-180 Cicero, source for Phil.'s
Antikrates 172-173 auloi 84 life 5-7
ntiphilos 172 authenticity, criteria for 46- codex 74
ntony, M. 11 47 columns, papyrus 74
Aphrodite 91-92, 124-125, conversation 63
185 Bakchios 168 conversion to philosophy 36,
apographs 51-52 Bakchos. See Dionysos 38
Apollophanes 162 bee, Xanthippe as 68-69 Corrector of P 48-49
aposiopesis 88 blason anatomique 104 cunnilingus 175
apples, as love tokens 65-66 book of life metaphor 74,
Apsines 4 76-77 Demokritos 114
Aratos 173 Bouhier, J. 52 dew 70, 87
Archias 18, 28 brassieres 97 in poems 38
Argentarius 191-192, 198- breasts 96-97 Dionysos = wine 84
199 Brink, C. O. 21 Dios Apate 88, 90, 124
Aristippos 65 buttocks 106 Doric forms 171
Arkesilaos 179 Douglas, N. 213
arms 107 Caesar, J. 214 doves 180
Artemidoros 162 Calvus 90 Dyck, A. 29 n.19
ascription, problems of 46, Cameron, Alan 204, 206,
192 210 156-158
Asklepiades 203-204,215, Catalepton, Epicurean f l a v o r E i r e s i o n e 157-158
220-221 of 22 25-26

245
246 General Index

elegy, women writers of 26 gnomes 87 Krates of Mallos 31-32,37-

n.6 Goldberg, S. 29 n.l 38
ellipsis 82, 129 grace 95 Krevans, N. 86
Encolpius 116 Graces 98 Kybele 184-185
Endymion 115 Greek Anthology 45
Ennius, in Herculaneum 13 Griffiths, A. H. 40, 78, 203 Lais 186
Ephyra 202 lamps in bedrooms 86, 196,
Epicureanism hair, grey 79-80 208
papyrus texts of 14 erotic 121, 124 Leontion 35 n.39, 37
pleasure 6,32 Hammerstaedt, J. 29 n.19 Leukothea. See Ino
poetry 32 hapax legomena Lewis, D. M. 156
soul 11 185-186 LSJ, addenda and corrigenda
views on death 1 6 5 107-108 to,s.vv.
ePICUREANS 193 1365
ATTITUDE TOWARD 193 144
Socrates 37 127 82
and food 161-162 96 174
AS PIGS 16N.13,223 221 176
of Rhodes 10 70 7 0
simple life of 16, 153- 69 149
154, 1 6 0 6 9 ' 1 2 3
of Syria 10 n.26 193 93
Epicurus 97 180
birthday 152 ff. 92 144
will 156,233-234 193 194
epigram 177 (by Luck, G. 70-71
as autobiography 10—11, conjecture) Lucretius, in Herculaneum
17, 25, 40, 188, 204-205 harbor, as erotic 13, 24, 71, 209
improvisatory 18, 27, 47 metaphor 92, 94 Lydian inscriptions 183
topoiof 33,38,104,209, harp 62-63
215 Haslam, M. 135 Maccius 126
Eros 69-70 heat, erotic 63, 100 madness, erotic 79
pluralized 121 Heinsius, D. 201 marriage 34-36, 85,210
Ethiopians 109, 124 Herculaneum 12 ff., 167 matrons (Roman) 202
Eudoxos 173 Himera 9-10 Maxwell-Stuart, P. G. 174-
euphemism 133-134,151. Holford-Strevens, L. 76, 175
See also aposiopesis 108,177,210,214, Meleager 27
euphony. See Krates 218-219 f., 222, 231 Meleagrian context 145
eyes, erotic 63 homoeoarcton 101 Melikertes 188
Horace, on flattery 20 Memmius, G. 8 n.19
Fabullus 23, 232 horse tack 193-194 Menippos 4
Flora 66, 108-109 metapoiesis 142, 206-207
Friedrich, G. 23 Ino 188-189 metrics, in general 41^45, 139
Frischer, B. 22 invitation poems 153, 161, caesura 92
221 correption 197,201
Gable, Clark 144 Ionian Sea 145 diaeresis 201
Gadara, 3-4, 187-188, 214, elision 75, 137
230 Jannaeus, 4-5 Hermann's Bridge 65,
Galloi 178 ff., 200, 202-203, Julian of Egypt 219 133,140
217,230 hiatus 140
Giangrande, G. 176 kennings 195 Hilberg's Law 109
Gigante, M. 29 n.19, 40, Kephalas, Konstantine 45 metrical patterns 73
125, 162, 167, 177, 187- kestos 124 Meyer's First Law 139
188,189-190,207-213, kisses 107-108 Meyer's Third Law 174
215, 220 koronis 76 Naeke's Law 160, 199
 General Index 247

Moirai 100 pigs Selene 113

monolog. See soliloquy as food 158, 162 Shakespeare, Wm. 72
myrhh 69, 82 Gadarene 3 n.2 Simonides 27
myths as exempla 109-110 See also Epicureans Siro 15-16,19,22
Piso Caesoninus, L. called Silenos 24,
Nais 91, 145 Calpurnius 4-5, 155, 227
Naples. See Parthenope 157-160,202,228-229, skolia 27 n.13
Nereids 189 232-233 sleep
nicknames 24, 71 birth 7 as death 68, 71
death 11-12 erotic 115
oaths of lovers 197 owner of Villa 14,16-17, Socration 23-24,34,232
Obbink, D. 210,213,222- 20 soliloquy 63
223, 225 as patron 5 n.11 soul 63
Oinomaos, 4,206 travels 11, 94 Stoics 173,215
old age 131 Piso Frugi, C. Calpurnius 8 Sulla 7 n.16
older women, sexy 95-96, n.19 Sulpicia 90
176, 222 Planudean Anthology 45 synecphonesis 106-107
Origen 117 Planudes, Maximus
Oscan 108 as bowdlerizer 221 Taran, S. L. 116,186
Oinomaos 4,206 his arrangement of Termeros 130-131
epigrams 45 n.2, 49-50, testicles 141,174
Pace, N. 29n.l9 74 Theodoras of Gadara 4
Palatine Anthology 45 Plato, as author of time 177
Pansa, Caetronianus, C. epigrams 65 Torquatus 227
Vibius 7-8 Plotius Tucca 20-21 touching, erotic 148
Paphos 90,207 Polemon 38 Tower of the Winds 173
paragraphos, 224. See also Poseidippos 27 n.10, 46 n.4, Trimalchio, faulty arithmetic
koronis 47, 76 of 13 n.3
38-39 Pothos 69-70. See also Eros
Parthenope = Naples 18-19, prayer 91 Um Qeis. See Gadara
212 present tense for future 134,
Parthenios 26 n.9, 217-218 155 Veranius 23, 232
patronage. See Piso priamel 81,83-84,212 Varius Rufus, L. 19-20
persona, of epigrammatists prostitutes 82—83, prices Vergil, and Phil. 11 n.5, 17—
34, 36, 39-^0 charged 125, 177 18, 20, 67, 213
personification 124,131 puns. See redende Namen called Parthenias 24,
Petit, S. 103 218
Phaeacians 159-160 Quintilius Varus 20-21 Villa dei Papiri 10-16, 154,
Phaidros, head of Garden 8 167
Pherekydes of Syros 177 redende Namen 98, 99-100,
Philainis 102 Wigodsky, M. 29 n.19
in Phil. 86 Reiske, J. J. 200 winds 190
ofSamos 86,108,124 Reitzenstein, E. 22 wine 82,158,163,209
Philaras, L. 141 repetition 70-71,92,131, Wiseman, T. P. 182
Philo of Gadara, 4 182 wives, erotic 90
Philodemos
accompanies Piso 16 Sabaktes 181 Xanthippe (Phil.'s)
in Alexandria 9 Sabazios 181 other women named 66-
birth 3 Salmakis 182 67
death 11 sandals 163 poem cycle on 33-38,54,
on epigrams 28-31 Sappho 30-31,62,108-109, 62
expelled from Himera 9 110
his books 9-10 Scaliger, J. J. 52 Zeno of Sidon 4
publication 28 segnis amator 128, head of Garden 8, 182,
Phryne 176 147-148 234
 INDEX OF THE MORE
IMPORTANT PASSAGES

Epigrammatists are quoted by the number of their latest editor, as well as by Greek Anthology
numbers.

Aelian Alkiphron Anonymous

fr. 40 Hercher: 9 Epist. 2.2.1-3: 37 Epigrams
Agathias, ed. Viansino Alkman (PMG) 6 FGE (5.101): 136
74 (5.287.6): 121 59(a): 97 7 FGE (5.11): 92
78 (5.282): 98 Ambrose 8 FGE (5.51): 62
90 (5.294): 113 Epist. 14 (63), 4 HE (12.104): 146
Aischylos 13 = T 9 9 HE (12.99.1): 116
Agamemnon Amphis (K-A) 56 HE = Phil. 24
419: 97 8: 71 AP 5.95: 67
681 ff.: 99 Anacreontea, ed. West API 235.5-6: 171
991 ff.: 111 3.3: 184 lambica, ed. West2
1391-1392: 120 16.27: 107 57: 71
Choephoroi 17: 170 Anthologia Latina
123: 144 17.29: 107 371 (Ps.-Petron.): 66
Prometheus Bound Anakreon 458: 83
89-90: 189 Epigrams (FGE) Antipater (GP)
649-650: 121 5 (6.134): 66 53 (5.109.1-2):
Fragments, ed. Radt Lyrica (PMG) 140
57.7: 63 363: 69 Antiphanes (GP)
Alkaios, ed. Voigt 358: 80 1 (6.88): 125
50.1-2:69 411:107 Antiphilos (GP)
346: 163 420: 80 14 = Phil. 21
347.1: 84 458: 107 Apollinides (GP)
347.4: 166 Anaxippos (K-A) 26 (10.19.1-2): 121
368: 155 1.27: 84 27(11.25.6): 75,80

248
 Index of the More Important Passages 249

Apollonios Rhodios Asconius De Finibus

3.1377-1378:114 ad Cic. la Pis. 68 = T 3 1.65:16,154
Archilochos, ed. West2 Asklepiades (HE) 2.23: 17 n.14
1:102 3 (5.153.1,2): 114 2.101:156
5: 83 5 (5.210.3): 124 2.119 = T1
19:143 6 (5.203.1-2): 193 De Natura Deorum
23: 143 8 (5.162.3): 102 1.46: 170
122: 143 9 (5.7): 196 De Oratore
126: 144 10 (5.150): 215 3.61: 37 n.45
196a.l7: 121 12 (5.145): 220 3.194: 28. n.14
196a.24-28: 67, 98 13 (5.164.1): 196 In Pisonem
206: 105 16 (12.50.3): 121 22: 16-17
Argentarius (GP) 25 (5.181.12): 90 68-72, 74 = T2
1 (5.16.2): 145 41 (7.217): 97, 120 68: 6
6 (5.104.1-2): 107 Ausonius 71: 28 n.17
7 (5.105): 174 Epistulae 20.4: 100 87: 7
8 (5.110.1-2): 217 Automedon (GP) Post Red. in Sen.
9 = Phil. 37 1 (5.129.1-2): 107 14 f. = T 12: 6
10 (5.116.5): 185 1 (5.129.8): 130 Clemens
13 (5.128): 86 2 (11.29.3-4): 130 Homiliae 13.8: 71
18 = Phil. 35 Cornutus
Ariston of Chios (SH) Babrios Natura Deorum 25: 70
204: 170 114.1: 87 Corpus Imcriptionum
Aristophanes Basil Latinarum
Archarnians Epistulae 8.12: 69 4.4029: 129
1090-1092: 82-83 Bassus, ed. Morel 4.4816: 129
Aves 1: 62-63
921: 82 Demarches 1.71: 111
Clouds Carmina Epigraphica Demokritos (DK 55)
504: 130 Graeca B 160: 72
978: 121 1.132: 127 B 205: 72
1364: 76 2.530: 144 Diodoros Grammatikos (AP)
Ekklesiazousai Catullus 7.700.1: 70
13: 121 5.6: 71 Diogenes Laertius
Lysistrata 1: 20 6.101: 156
362-363: 125 8.12-19: 145 10.3 = T 7
387-388: 179 13.1 ff.: 83 10.119: 35 n.37
Peace 17.15-16: 121 10.138: 74
341: 149 35.16-17: 109 Donatus
Ploutos 36.11-17: 91 Vita Verg. 68: 20
69-70:118 47 = 111:23 Dorotheos, ed. Stegemann
253: 76 50.1-5: 168 5.8 f.: 174
Wasps 50.14-15: 130
1173:107 70.3:197 Empedokles (DK 31)
Fragmenta (K-A) 85:64 661.1:186
268: 124 Chairemon (TGrF 71) B 96.3: 215
483: 120 14: 106 B 98.3: 94
753: 108 Christodoros (AP) Epicurus
Aristotle 2.1.13-14: Epistulae
DeAnima 181-182 1.63: 112
404al: 114 Cicero 3.130: 153
Categories Pro Archia 3.135: 72
2all: 64 8.18: 28 n.18 Fragmenta (U)
Arrian Pro Caelio 131: 107
Indika 6.9: 109 35: 168 135: 211
250 Index of the More Important Passages

Epicurus Fragments, ed. Nauck 2 Epistulae

Fragmenta (U) (continued) 897.9-10: 111 1.5: 156
142: 37 1.5.24: 158
163: 111 Gaetulicus (FGE) 1.7.25-28: 80
200: 112 1 (5.17.3-4): 94 1.14.36: 80
215 A: 153 1 (5.17.6): 92 1.15.22-24: 159
229: 159 2.2.211-216: 80
311:111-112 Hedylos(HE) Epodes
312: 111 6.4: 168 2: 143
313: 111 Herodas 9.33-34: 82
411: 202 6.11: 84 Odes
472: 161 10.1-2: 96, 131 1.13.2-3: 69
535: 117 Herodian 1.19.5-6: 96
539: 154 Partitioned 61.1: 176 1.20.1: 154
Fragmenta (CPF) Hesiod 1.38.1: 83
11, col. 1.17: 71 Opera et Dies 1.38.2:81
KyriaiDoxai 252-254: 189 1.38.3-4: 81
8: 208 353: 143 2.4.2-4: 109
40: 165 521: 115, 120 2.4.22-24: 74
Sent. Vat. 695-697: 72-73 2.5.1-2: 120
14: 153 Theogony 2.5.9-12: 121
25: 211 96 f.: 155 2.11.21-22: 85
51: 36 910 f.: 97 2.14.1: 131
66: 165 Fragmenta, edd. 2.17.17-20: 174
Epigrammata Bobiensia Merkelbach-West 2.18.14 ff.: 121
32: 66 61: 102 3.7.20: 129
35: 67 123: 184 3.14.17: 163
Eratosthenes, ed. Powell 124: 197 4.1.5: 93
25:84 Hesychios, s.v. 4.11.31 ff.: 77
Euboulos (K-A) 99 4.12.12-13:84
41.1-2: 69 183 Satires
121: 82, 158 176 1.2.54, 62-63, 78,
Eudoxos, ed. Lasserre 221 94: 202
321: 108 Hipponax, ed. West2 1.2.119-122 = T 4:
Euenos (GP) 30: 183 139
7 (12.172): 146-147 Homer 1.2.123-124: 210
Euphorion, ed. Powell Iliad 1.2.127 ff.: 119,
113: 93 2.235: 182 218-219
Euripides 2.477^479: 170 1.5.40-43: 21 n.28
Alkestis 3.396-397: 107 1.10.81: 21 n.28
165-166: 85 5.493: 144 2.8.15: 163
Backhai 10.293: 120 2.8.88: 162
644: 175 13.390-391: 171
Hekabe 13.394: 141 Ibykos (PMG)
635-637: 114 Odyssey 286.3-6: 120
Hippolytos 5.452-453: 94 Inscriptiones Graecae
525-527: 97 8.135-136: 105 22.5021: 94
545-546: 85 9.451-452: 117 4.358: 127
1064: 102 Epigrams 12(5).891.1-5: 173
1303: 127 15.10: 157 12(8).441.1-2: 93
Kyklops Homeric Hymn to Aphrodite 14.793a: 89
171: 102 228-230: 79 14.1589: 179
465: 105-106 Horace
Troades Ars Poetica Jerome
2-3: 189 438-444: 21 Chron. 166el4: 21 n.28
 Index of the More Important Passages 251

Kallimachos Maccius (GP) 1.9.18: 63

Epigrams (HE) 1 = Phil. 18 1.10:125
I (12.102.3-6): 103 2 (5.117): 120 2.4.29-30: 107
2.1: 212 8 (6.233): 192 2.4.35: 123
II (5.6): 197 Machon, ed. Gow 2.4.40:123
44 (7.519): 164 346: 134 2.4.45: 95
Hymns Macrobius 2.5: 143
4.298-299: 120 f. 5.20.8: 26 n.9 2.5.17:63
ragmenta, ed. Pfeiffer Martial 2.5.23-28: 108
1.6: 98 11.43.12: 185 2.8.11-12: 109
1.33: 70 11.46.4: 130 3.7.9: 108, 148
28: 93 14.39^10: 87 3.7.15: 151
114.5:93 Meleager(HE) 3.7.16:130
761: 184 2(7.417.2): 3 n.2 3.7.17-18: 131
Klearchos, ed. Wehrli 2 (7.417.3): 4 n.4 3.7.23-24:129
44: 99 7 (5.179.5): 69 3.7.39: 148
Kratinos (K-A) 8 (5.180.1-2): 121 3.7.41-^2:131
246.2: 75 11 (6.162): 87 3.7.65-66: 130
Krinagoras (GP) 19 (12.117): 78-79 3.14.23: 108
41 (7.401.7-8): 186 26 (5.160): 99 Ars Amatoria
50 (16.199.6-7): 97 30 (5.140.4): 98 1.381:118
31 (5.144.1-4): 165 1.717: 100
Leonidas (HE) 32 (5.149.2): 96 2.657-658: 123
10 (7.648.8): 114-115 32 (5.149.4): 98 2.661: 123
20 (7.295.6): 98 41 = Phil. 13 2.703: 90
Loukillios (AP) 47 (5.148): 98 Heroides
11.160: 172 69 = Phil. 36 15.35-36: 109
Lucian 73 (5.191.1): 114 18.59-65: 115
Amores 103 (12.101.1-2): 93 21.107: 66
26: 124 108 (12.159.5-6): 160
DeDea Syria 129 (12.257): 76 .Here. (See also
15: 182 Menander Philodemos)
52: 179 Fragments ed. Korte- 1251.15.4-14: 35
De Mercede Conductis Thierfelder n.38
36: 108-109 409: 93 P.Lit.Lond.
Pro Imaginibus Sententiae 11: 76
5: 124 20: 144 P.Oxy.
Lucreti Vita Borgiana, ed. Menekrates (HE) 219(a).24: 137
Masson 3 (9.55.2): 98 3700: 135
p. 6 = T 10 Mimnermos, ed. West2 P.Prag.Varcl.
Lucretius 1.5 ff.: 131 N.S. 41.16-17: 134
1.140-141: 154 Parmenion (GP)
2, proem: 151, 165 Niketas Eugeneianos 11 (9.342.1-2): 27
2.656-657: 84 f. Drosilla and Charikles Paulus Silentarius, ed.
3.869-872: 72 8.113-115: 113 Viansino
3.885-887: 72 Nonnos 52 (5.258.1): 97
3.894: 71 16.297: 102 Pausanias
3.933-934: 71 1.43.6: 69-70
3.955: 72 Oinomaos, ed. Hammerstaedt Persius
4.1048: 112 11 B: 206 pro/. 14: 70
4.1057:112 Ovid Pherekrates (K-A)
4.1106: 112 Amores 247: 176
4.1160 ff.: 123,209 1.2.2:218 Philemon (K-A)
Lydus, Johannes 1.5.19-23: 104 28.1-4: 93-94
DeMensibus 1.13: 108 1.8.2-4: 82 113: 161
252 Index of the More Important Passages

Philip De Signis, ed. De Lacy and Plautus

1 (4.2.8 f.) = T 5 De Lacy Miles Gloriosus 379:
Philodemos 4: 216 114
De Bono Rege, ed. 25.33:115 Pliny
Dorandi 33.12: 115 NH 35.5: 156-157
19: 208 P.Herc. 253 Plutarch
De Dis, ed. Diels fr. 12: 20 Moralia
Bk. 1.25.35 ff.: 11 P.Herc. 312 18e: 71-72
Bk. III.l.17-18: 155 14 = T 15: 22 124f: 158
Bk. Ill, fr. 76: 77,79 P.Herc. 460 513f: 89
De Ira, ed. Wilke fr. 22: 38 n.38 1089c: 154
28.26-27: 149 f. P.Herc. 994 Theseus
DeMorte,IV 29 N: 38 11:130
25.37-26.7 Kuiper: 12 33.5-8: 194 Pollux
30.10-11 Kuiper: 153 P.Herc. 1005, ed. 3.37:93
33.25-30 Kuiper: 188 Angeli 6.188: 98
37.23 ff. Gigante: 168 fr. 77: 77 10.38: 124
37.26-27 Gigante: 153 col. 14.6-13 = T 14: 9 Poseidippos
38.7-11 Gigante: n.24 2 HE (5.186.3-4): 137
10 n.26 P.Herc. 1082 10 HE (5.183.6): 155
38.17-18 Gigante: 72 11:19 11.1 HE: 217
39.17-18 Gigante: 74 P.Herc. 1428, ed. 705.5-6 SH: 76
De Musica IV, ed. Henrichs 3.3-4 Bastianini-
Neubecker 13.23-14.2: 170 Gallazzi: 177
6.13-18: 168 15.20-23: 77 Praxilla (PMG)
Oikonomikos, ed. P.Herc.Paris. 2: 21 754:114
Jensen Philonides(k-A) Proklos
2.3-5:35 n.38 7: 197 in Plat. Crat. 406c: 84
9.1-3: 35 n.38 Pindar Propertius
23.23-32: 5 n.11 Isthmians 1.1.1-2: 210
38.7-8: 153-154 4.90b: 97 1.3: 113
Parrbesia, ed. Olivieri Olympians 1.19.18: 63
fr.8-11: 24 n.l 2.85: 211 2.1.9: 63
fr. 45.8-11: 24 n.l Fragmenta, ed. 2.4.9-10:64
fr. 45.8-11 = T 13 SneU-Maehler 2.15.15-16:115
col. 8B.6-13: 39 104: 114 2.23.13-24: 133
n.50 122.3-5: 93 2.26: 188
col. 24A.8-B.1 125.4: 63 2.28.19-20: 188
De Pietate, ed. Obbink Plato 3.1: 188
818-819: 155 Epigrams (FGE) 3.5.19-46: 79
1773-1774: 107 4 (5.79): 65 4.5.1: 186
De Poematis, ed. Mangoni 5 = Phil. 2
7.1-6: 212 Kriton Quintilian
29.21-23: 30 n.20 49b: 143 1.10.31: 63
37.2-38.15: 28-31 Phaidros
Rhetorika 255e: 148 Rufinus, ed. Page
1.273.9-11 Republic 5 (5.18): 109
Sudhaus: 129-130 329a-c: 75, 131 21 (5.60.3): 107
2.34.14-16 Sudh.: 63 373a: 83 23 (5.62): 98
2.145, fr. 3.8-15 411a-b: 138 33 (5.92.1): 137
Sudh.: 10 n.26 474d-475a: 123 35 (5.94): 170
2.226 Sudh.: 31 614b: 159
2.289 Sudh.: 124 f. Symposium Sappho, ed. Voigt
col. 53.10-11 Longo: 8 192b: 89 1.17-18:77
n.20 213d: 89, 148 16.7-8: 107
 Index of the More Important Passages 253

22.11: 69 Sophokles Tibullus

58.12-14: 79 Aias 1.2.33-34: 86
102.2: 107 713: 131 1.8.37-38: 108
112: 92 1150-1158: 71 3.7.9: 219
168B: 151 Antigone Ticida, ed. Courtney
T252: 109 643: 144 1: 90
Scholia ad Elektra Timotheos Comicus
Aristoph. Ra. 1374: 424-425: 129 (K-A)
140 Trachiniai 2: 208
Theokr. 2.10: 114 536:85 Tyrtaios, ed. West2
Semonides, ed. West2 Fragmenta, ed. Radt 11.23:105
7.86: 89 398.5: 69
Varro
Seneca 636: 150
QN 6.1: 167 811:197 Uenippeae, ed. Buechler
Servius Sophron, ed. Kaibel 136:66
ad Verg. E. 6.13: 20 54: 131 398:31n.25
Simias (HE) Souda, s.v. Vergil
4 (7.21.1): 93 9 Aeneid
4 (7.21.5-6): 74 9 3.151-152: 114
Simeon Metaphrastes = T 8: 9 9.617: 182
VitaS. Patapini6: 102 Strabo Catalepton
Simonides 5.8.4: 167 5.8-9:15
Elegies, ed. West2 14.656: 182 7:20
19: 206 16.2.29 = T 6 8:15
21.3:63 Straw (AP) Eclogues
22.12: 97 12.240: 128 3.71:66
25: 27 12.175.4: 148 9.35-36: 20
26: 27 12.240.3: 151 Georgia
Lyrica (PMG) Supplementum Magicum 1.336: 67
521:71 72 (122 PGM): 65 4.563-564:199,
Siro, ed. Gigante 213
F6:15 Theognis
Xenarchos (K-A)
F7:15 422-424:87
4.16ff.: 140
F9:20 567-569:71 6:197
Sisenna, ed. Peter 1345 ff.: 109 Xenophon Ephesius
• 5 ^ 1167
53: /7 1 ^71^81
1367-1368: ^
133 3.2.4: 148
Skolion(PMG) Theokritos
904: 137 2.10-11: 114 Zeno Sionius, ed. Angeli-
Skythinos (AP) 10.26-27: 123 f. Colaizzo
12.232.4:125,130 11.76:125 11 = T 14
Solon, ed. West2 15.108:97 Zenodotos (AP)
25.2: 105 20.23: 124 7.315.1-2: 186
 INDEX OF GREEK WORDS

Words are listed as in LSJ (or as they would list proper names). Ignored are
articles and pronouns, as well as uncertain words in and those which begin complete
epigrams. (Words are cited by epigram and line; those of the incipits in are cited by column
and line). Thus, 23.3 for an epigram, iv.6 for the incipits. Some words found only in the appara-
tus are in parentheses. The words of the last two epigrams are in square brackets, but those of
whose authorship can only be guessed at, go unbracketed. See also the general index s.v. hapax
legomena.

3.2 29.5 35.4

4.4, 5.4 7.1 12.8
v.17 34.2 18.3
27.8 29.3 . . 34.4, iv.ll
24.2 2.1,4.5,4.7,5.2,9.2,18.2
20.6, 24.1 9.7, 11.6, 13.3, 15.5, 1 6 . 3 , 3 7 . 5 ]
37.3] 17.1,15.5,20.5,23.5,27.5 33.2
3.8, 15.4,20.3,23.5, (bis), 33.8, 35.7, 36.2, 3.7, vii.3
29.6, vii.5 [37.1] 5.3
3.5, 12.7, v.24 18.4 29.7
8.2 17.6,18.6,36.1,iv.6 31.1
3.5,14.4,30.1 29.3 33.8
28.2
28.2 _._../ 20.3 25.4
25.4
30.2 9.5 8.3,28.4, 35.4, [38.4],
36.2, [38.7] 5.1 v.29, vii.3
v.32 33.5 19.4
17.4 24.2,31.4,36.2, 27.3
3.7 v.8 28.3
29.2 6.5 12.1
7.5 13.3, 18.5 15.2

254
 Index of Greek Words 255

29.5 10.1 1.4,3.8

26.3 6.6 22.3
5.6,25.2,30.6,33.6 18.5,20.7,
"A 31.1 21.3,23.1,23.2,29.6, 16.4
31.3 viii.4
10.4
18.2 1.4, [ 3 7 . 5 ] , 4 . 3
v.15 iii.19, v.27 6.4,20.3,20.4,20.8
28.1 34.1 (bis),21.2,23.4,28.6
30.5 vi.7 12.7, 19.5,20.4,24.1,
9.5 12.2 27.4,27.7, [38.7], ii.25,
28.1 34.6, iv.3 iii.17, vii.ll
1.2 3.6 . 27.3,27.8
29.3 12.6 ii.2
15.6 ' 4.7, v.2 9.8
vi.19 6.6 . 15.5,21.2,24.2,25.3,
18.5 9.4,16.1 28.4, 29.1,31.6, [37.1],
iv.24 21.6 [37.5], vii.20
iii.9 11.4, 25.3, iv.9, viii.8
1 4 . 3 3 0 . 6 , 3 4 . 2 14.3,19.2,27.1,27.7,
2 7 . 1 1 8 . 6 , 2 3 . 2 (bis) 35.8, i v . 1 1
16.6 13.2 19.2,22.1,30.5,38.7
13.4, 29.8, 36.6, vi.18 2 5 . 1 1 5 . 1 , 15.2
22.6 22.1 (bis) 1.3,5.2,19.2,27.2,27.8
19.5 20.4 22.6, 36.3, 36.5
'Ai.. . 19.1 27.3 viii.5
29.4 9.8 7.2
17.5 4.8, 8 . 8 e K T o v e c o iv.30
ii.13 9.8 29.8
v.4 7.2
Bi 29.7 10.2 25.1
B .. . 6.7 35.7 38.6]
(2.1), 14.2, ii.ll 3.4 (bis), 3.8, 11.4, 20.1, 15.5, 27.2, v.28
23.4 22.6 v.25
33.8 6.1,6.2 7.4
23.7 11.5 33.3
16.5 10 (passim), 11.1, 3.3,3.7,4.6,9.3,23.8,
22.2,22.4 11.2, iv.21 26.3,29.5,31.3,31.5,
(3.8) 14.2, 34.5 [38.1], [38.8]. vii.ll, viii.3
4.2 22.6,31.4 27.1
3.8 20.6,21.2,22.1, 'E 14.6
23.5 28.2, vi.ll 33.1 (bis)
28.2 8.1,12.1 27.3
16.2 14.1 19.1
11.6 3.2 (ibo) 7.3
19.3 31.3 9.1
27.4 6.3 27.5
35.1 5.3
27.6 38.1 4.1
18.4,20.6 16.5,34.6,35.3
38.8 33.3 29.5
5 . 3 , 7 . 3 , 1 22.3,32.3,32.3 2.1
15.5,18.5,24.2,30.1,30.5, 3.3 11.2,13.2,34.1
30.5,31.3,31.5 7.2 38.7]
20.1, 20.4, 22.5 28.3 37.6]
31.2 16.5 5.1, 10.1, 12.8, 37.1]
256 Index of Greek Words

9.7,18.1 23.8' 3.2, (18.6), 21.1,

25.3, 25.4, [37.2] 19.4 22.2, iii,17,viii.5
14.4 8.3 33.3
29.1 5.5, vi.14 16.1,33.8
23.3 28.3 4.6, [38.1]
30.4 30.3 15.1, 18.1, 19.3,
15.2, 26.2 8.5 33.6
7.3, 15.6, 16.3, 22.5 20.7
38.1,ii.26, iii.10, iv.30 1.2
15.2 18.6 4.3
11.1 29.8 12.5)
27.2, 27.5 35.6 10.5
36.6 10.1, 18.3 14.4
4.5, 9 (passim), 28.4 11.3,26.4 29.2
11.5 5.2, 33.4 33.1
9.8 11.1 38.4]
20.4 10.6 vi.6
3.5, 26.4 (bis), iii.20 16.1, v.28 8.4
20.8, 28.6 25.1 35.1
3.2 16.3 17.3
17.5,31.5 vii.ll 18.4. 30.1
35.5 16.3 35.1
14.2 34.4 3.1
vii.20 15.5 12.5
38.8] 29.4 23.7
28.4, 29.7, iii.22, 5.4 12.6)
vi. 11 iii.12 7.4
6.2. 6.3. 10.6. 20.2. 30.6 26.1
21.3, 22.5, 25.4, 36.4, 7.4 8.6
[37.2.], [37.6], [38.8] 14.2 23.6
iv.27, v.6, vi.12 12.6, 30.5, iv.33 12.1, 30.4
16.5 20.6
31.2 36.4
34.4 33.7,36.1 7.5)
13.2 ; vi.9 23.8, 36.6
25.2 34.5 21.1. v.15
iv.20 12.8 33.7
34.1 38.2
13.3, [37.5] ii.12 4.7
1.3, 18.3, 18.6,22.5,31.4 vi.12 35.2
(bis) 23.6 ii.l
15.5 35.5 28.1, 29.2
4.2, 4.3, 5.4, 7.6, 8.8, 35.4 22.4
16.3, 19.3, 29.1, ii.6, iv.9, 23.5 31.3
v.30, vii.2 9.4 8.3
viii.6 31.4 6.6
ii.26, v.28 v.l
38.3] iii.22 11.1
34.6 26.3 12.2
20.7. iv.27 5.5 v.12
4.4, 5.4 iv.25 9.2
13.1 12.5 35.7
19.2 20.1, [37.3] 18.4
23.8 27.1 15.1
15.1 14.3 34.3, 34.5
 Index of Greek Words 257

8 (passim), 11.2, 35.6 23.8

11.3, 17.6,33.5, iv.26, 9.3, 12.3 iv.22
viii.12 38.2] v.16
vi.3 10.1. 10.2, 16.6, 18.3, 6.5
ii.2 23.7, 35.8, vii.19 (bis) 8.4
5.1,6.3 34.1 16.6
4.5,5.2 11.25 30.3
9.4 16.1 34.3
1.1 17.1 11.17
8.5 5.3 v.21
34.1 35.4
30.4 27.6 31.2
12.2 v.26 14.1
33.6 3.6 5.5, 18.3, 19.1, 19.6,
8.5,21.2,25.1,26.3 4.5 21.6,23.7,29.8,36.5,
1.1 9.6,21.1,25.4 [37.4], vii.8
4.5 26.1 19.2, 26.1, 36.1, viii.3
28.6, 11.7 21.3 35.5
9.8 38.4
11.5,21.4, vi.15 20.5 3.6
7.6. 36.3 18.2 iv.l
34.2 v 30.5 1.2, 2.2, 4.4
6.1,33.8 2.1 3.1
4.3, iii.15 29.7 20.5, 30.5
23.3 37.4
3.4 vii.9 35.2
30.1 20.1 1.4,7.6, 11.4, 14.5,
8.8 12.2, 30.3 15.3 (bis), 15.6,21.5,24.1
[37.2] 8.2, 33.4, 35.6, iii.12 31.1, [38.4], ii.3, iv.25,
27.1,27.8. viii.ll 17.1 vii.6, viii.7
22.5, 30.3 38.8] 19.3
ii.5 6.4, 24.2, iii.10, iv.8, 6.2, vi.10
9.3 v.22, vii.5 37.2
9.1 9.4 111.14
4.1 28.3 14.5
35.7 17.2 vii.3
23.3 m.5 35.8
22.3 19.2, [37.5] 1.1,27.7
16.4 3.3,3.7 12.4
7.2, 36.1, 36.6, ii.13 7.3. 33.5 36.4
5.6 ii.9 iv.28
4.8,11.8, iv.5 15.6
11.3,22.2,11.26 3.1 iv.3
17.3 27.2 21.1, [37.4]
29.3 23.6 12.7
5.2, 12.4, ii.ll 9.6 21.2.27.5,36.6
iii.15 ii.10 9.7
31.6, 33.1 3.3, 6.6, 28.5, [37.3 33.6
21.5 3.1 36.5
31.1 6.7 36.2
4.8, 6.4 38.1 vi.7
16.6 31.6 20.5
30.2 15.1
2.2, 19.6 3.6 (bis) 5.5, 15.6, 19.6, 25.3
7.3. 36.4 8 8
.8 15.3 (bis), vii.6
258 Index of Greek Words

25.4, [37.6] (bis) 38.4 v.18

v.25 12.8 16.2
27.4 29.4 19.2, iv.13
5.6, 10.3, [37.2] 3.4,3.8 29.2
17.2, 29.4 20.7 30.4
36.3 23.4 iv.27
11.2, 16.1, ii.24 viii.l 7.4.25.2. 26.2
v.23 27.8 7.4
36.1 6.5 37.4
31.5 38.6 1.2, 4.6, 16.4
ii.20 34.5 16.6
31.6 15.5 1.3
vii.8 6.7
9.2, iii.13 35.3
10.4 [37.1] 12.3
5.5 (bis), 29.7 11.4 33.1
28.6, v.9 37.5 5.3, 15.3
3.5,6.1 (bis), 6.2, 11.6 vi.10
6.3 (bis), 10.3, 21.4 (bis) 37.5] 29.1
23.1,23.2 3.2, 8.2, 9.7, 10.6 35.6
v.29 13.1 8.8
30.2 19.6 viii.4
27.5 iv.8 38.2
18.6 5.3
22.5 37.5 33.2
15.3 17.4, 19.3, 25.2, 29.3
28.6 v.30 33.2
25.3 3.5, (16.6), 27.6, 31.2 10.1
20.4 23.5.34.2 28.5
17.3, [38.5] 6.3 12.7
6.8 15.3 31.5
16.2, v.19 8.6 14.1, 14.5
iv. 14, iv. 15 34.3,35.3 17.1
11.5 20.4 4.2
18.3, [37.3] 1.3,3.4, 18.5, 20.3, 29.8
9.6, 11.6, 17.4, 18.1, 19.5, 27.7, vii.ll 13.3
18.3, [38.5], vii.5 31.3 7.1
8.3 10.5,20.7,21.1, ii.27
vi.16 21.3 21.6
7.6, 10.1, ii.l0, 12.1 6.2, v.16
iv.13, iv.24 18.5, 34.4 21.4
9.5, v.21 13.2 9.6
34.6 ii.2 17.5, 23.3, 33.5
27.1, 27.7, [38.7] 20.8 36.3, [38.3]
8.6 . 20.5 9.3, 30.3. 35.2
22.6 35.3 28.5, v.22, vi.18
vii.2, vii.16 13.1, 13.4 (bis), 35.8
2.1,20.7,21.3 23.5 33.7
37.1] 27.4 27.7
19.1,22.1,22.3 6.4, 8.8, 22.4 ii.16
35.2 21.5 33.4
12.5 38.5" 6.8
9.4 ii.27 35.3
25.1 29.6, ii.15, ii.16 iv.22
3s.i: ii.7 8.2
 Index of Greek Words 259

4.4, 5.4 21.4 (bis), 21.6 18.4

26.1 36.5 33.7
7.1,36.1 35.8 6.8
6.2 18.5 12.6
9.2 15.1 17.3, iii.19
31.1 10.3
iv.32 19.5 20.1 (bis), viii.8
4.2 18.1
8.7 27.6 23.1
34.6 14.1, 14.2, 18.3, 9.6, 16.2
37.1 18.4 9.1
29.2, 29.6 22.4 35.4
[37.2], vi.l 3.3. 12.4
22.1 v.18 v.5
28.1 17.4 27.4
4.7 34.5, 36.5, v.23 6.1, 28.5, ii.25
6.7, 26.2 9.7, 13.1, 13.3, 16.5, 84
17.6 34.5 28.4
9.1 13.3,31.3 34.3
15.3 14.4 iii.18
17.2 17.1, 17.5 6.6
19.4 7.1,7.4 3.5
37.3] 7.5 14.3, 17.6, iv.29
iv.23, iv.26 2.1, 7.3, 13.4, 14.3, 9.5, 17.2
10.4 20.3,23.1 (bis), 24.1 (bis), 38.5
15.5, 20.1, 20.2, 21.3 25.2, 26.4, 33.4, iv.6, viii.6 30.1
(bis), 24.1, [37.5], iv.6, 18.1 18.2
iv.12, iv.21, vi.6, vii.ll vii.20
37.5 33.7 26.2
23.7 10.5, 28.3, ii.12, 3.3, v.24, vi.8,
17.5 ii.15 vi.9
3.7 8.1 1.1,4.5
16.3 33.3 11.3
26.3 14.1 35.5
30.2 8.7 viii.1
12.3 8.7 1.3, 8.4, 13.1, 14.6,
28.4,31.5, iii.22, v.12 35.1 vi.19
4.1 7.5, 27.1, [37.6],
ii.29 iii.21, iv.2 1.3,3.7, 19.5,21.4,
30.1,30.6 20.2 34.1, ii.13, iv.24, iv.32
10.3, iv.9 33.6 12 (passim)
38.3 1.3, (9.7), 14.6 19.6, 23.7, v.4
33.2 12.7 1.1
33.1 (14.6), 15.6 12.3
7.41 22.5 28.6)
38.6 i 22.2 9.1, 20.8, iv.20
29.3: : 5.6 10.6, 26.4, 29.6
1.4, 4.6, 16.4 11.6 [37.6]