Bryn Mawr Classical Review

Bryn Mawr Classical Review 2009.09.54

Michael A. Tueller, Look Who's Talking: Innovations in Voice and Identity in Hellenistic Epigram. Hellenistica Groningana; 13.   Leuven/Paris/Dudley, MA:  Peeters, 2008.  Pp. xii, 231.  ISBN 978-90-429-2011-8.  €54.00.  

Reviewed by Marco Fantuzzi, Columbia University and Università di Macerata (
Word count: 3151 words

Hellenistic epigram, as M.A. Tueller reminds us, "is a child with more than one parent" (p. 205). This is especially true of the erotic epigram with its long and varied pedigree of short love-poems written in several meters and belonging to different genres. It is less true of the sepulchral and dedicatory epigrams, whose literary development in the Hellenistic age frequently consisted of innovative variation or contamination of their inscriptional archetypes or contemporary parallels, characterized (sometimes) by less ambitious literary concerns and (always) by more compelling real-life tasks. Inscriptional epigrams, both archaic and Hellenistic, had the real life function of communicating standard 'facts' about the dead or the dedicator, so that they often made similarity of expression and broadly shared conventional motifs, rather than preciosity of wording, their characteristic feature. In the Hellenistic age contemporary and past inscriptions remained at the same time both the parents and brothers of the literary form of dedicatory or sepulchral epigram. This relationship helped to keep the epigram shaped according to the formalized standards of content and expression that were characteristic of inscriptions, even after its predominant circulation in books would have exempted it from needing to pursue the standards belonging to inscriptions. Even the practice of variation on a theme or the frequent and flaunted use of intertextuality that characterizes most Hellenistic literary epigrams (not only sepulchral and dedicatory), probably relied on the formal and repetitive language which had featured, and continued to feature, in funerary or dedicatory inscriptions with an actual real life function.

The modern investigation of the dynamics of the Hellenistic literary re-use of inscriptional conventions started, in my opinion, with George B. Walsh's seminal paper of 1990.1 Of course, many previous contributions had spotted connections between Hellenistic epigrams and inscriptions, but in terms of positivistic "Quellenforschung", without highlighting the complexity of the literary phenomenon or the motivations and forms of its poetics. After Doris Meyer's works of 1993 and 2005 on Callimachus,2 Laura Rossi's commentary on Theocritus' epigrams of 2001,3 several contributions by Peter Bing in the last decade, chap. 3 in Kathryn Gutzwiller's Poetic Garlands of 1998, and chap. 7 in Tradition and Innovation in Hellenistic Poetry of 2004,4 the book of M.A. Tueller (henceforth "T.") is the first attempt at categorizing the different forms in which this re-use took place in a systematic way.

Chap. 1 "Before the Hellenistic Period" provides a clear and useful summary of the various works by scholars of archaic culture, religion, and psychology who have touched upon the main linguistic conventions or motifs of inscriptional epigrams (speaking object, apostrophe to the passerby, etc.), and is the necessary preface to the literary analysis of Hellenistic epigram in the following chapters. In particular, systematic categorization also leads T. to demonstrate how intensely conventional the language of funerary or dedicatory inscriptions was. Some of these conventions were already well known (but in every case T. provides excellent new statistical graphs of the evidence): that the inscribed object is usually assumed to be the speaker in the great majority of cases up to the end of the 5th cent. BC,5 or that in dedicatory epigrams the addressee is most likely the receiving divinity (who almost never has the role of a speaker), while in sepulchral epigrams the addressee is most likely the passerby. Furthermore, T. innovatively highlights, e.g., that epigrams featuring two speakers must be a dialogue between the object and the passerby, or that a ξένος/ξεῖνοςis always the passerby within sepulchral epigram. This relatively short chapter appears to be ancillary to the chapters that follow, and it is perhaps for this reason that it sticks to a purely phenomenological level. But for the cultural psychology that underlies the phenomena which T. describes see the recent works of Christiane Survinou-Inwood6 and Christos Tsagalis7.

T.'s arguments are always solid and convincing, although the task of categorization which they pursue -- and the trend to reductio ad unum which results from this -- sometimes leads him to oversimplified formalism. For example, T. maintains that the anonymous first-person external mourner8 must coincide with the passerby (pp. 40-1, 76, 79). T. has to admit, however, that the passerby, who usually does not speak in inscriptional epigrams, would then speak only in his role as an external mourner (p. 40): a strange exception which may in fact point out a flaw in his argument. The belief that the external mourner is a superfluous category, because he and the passerby coincide, is made even more dubious by Asclep. AP 13.23 = HE 962ff., where the speaker, who has the features of an external mourner in 5-6, addresses a passerby in 1-4.9

Another point where T. seems to oversimplify a bit for the sake of clear-cut categories is his observation on the absence of first-person authorial statements in inscriptions. T. examines some epigraphic and literary epigrams that were written to express a message from the poet but do this through the voice of the dead ("Erinna", AP 7.710 = HE 1781ff.) or at least in the third person (such as CEG 819.ii and 819.iii). T. proposes the fixed principle that the traditional inscriptional performers (the dead or the stone) must be the speakers and, for this reason, prevail over the "I" of the composer of the message. T. may be correct, but other readers might also believe that these instances of the third person being used for the authorial voice could also derive from the preference for the third person, which can be found in many archaic sphragides, from Alcman PMG 39, to HHom.Apoll. 172-3, to Theogn. 22b-4, and that, in place of a self-statement, authoritatively acknowledges the operation of the poet from the outside world.10

Chap. 2 "Creating Hellenistic Epigram" examines some epigrams by authors probably belonging to the first generation of Hellenistic epigrammatists (Perses, Nossis) that display hypermarked inscriptional conventions or recover conventions of the archaic inscriptions which were no longer fashionable in the 3rd cent., thus acknowledging their debt to -- but also showing their mastery of -- the models of this new literary form. The most significant of these conventions is the address to the god in the imperative in dedicatory epigrams, which is attested in the 5th cent. but not in the 4th cent., and is taken up by Nossis AP 6.265 = HE 2799ff. (was Nossis showing off, I wonder, that she shared the interest in the history of the inscriptions which moved, e.g., Philochorus to compile his collection "Attic epigrams"?)

Chap. 3 "The passerby in sepulchral epigram" is discussed in several contexts: a) the passerby as addressee, b) the passerby as speaker, c) the unusual passerby, and d) the passerby and the burier. The tombs or the dead of many literary, more than inscriptional, epigrams request behaviors or utterances from their passerby which were not in tune with the standards of pious attention/commemoration which the passerby had to perform according to conventional duty. In fact, the examples listed by T. show that such anomalies in the message entrusted by the dead to the passerby often correspond to the social anomaly, or at least the unconventional nature, of the deceased's character or death.11 Thus, it seems that in most cases it is the particular unconventionality of the deceased's way of life that paves the way for the epigram's original aspects. This fact may in my opinion shed some light on the sense of the title τρόποι -- "characters" or "ways of addressing"? -- that labels the eight epigrams at the end of the Milan papyrus of Posidippus (102-9 A.-B.). They were all sepulchral, and at least the first seven of them included dialogues between the dead and the passerby (the eighth is too lacunose to draw any conclusions): was the ambiguity of the title accidental or purposeful? I would not rule out the possibility that Posidippus, or his anthologizer, was following the practice of contemporary epigram, and his own practice as an epigrammatist, by purposefully adopting this ambiguous title knowing that it was capable of conveying the idea that these dialogues between dead and passerby, with their focus on peculiar reactions to the social standard of "nomoi, i.e. social conventions about how one treats a tomb",12 were excellent case-studies of the psychology of human characters. Besides, the passerby of literary epigram often takes the initiative to focus his attention not entirely on the dead, but also, e.g., on the cursedness of seafaring (Call. AP 7.271=HE 1245ff.), and most frequently on relatives or members of the family, as, e.g., in "Simon." 7.511 (see above concerning the identity of the speaker) or Call. AP 7.522=HE 1227ff. on Timonoe.13

As for the unusual passerby, the polyandrion epigram CEG 467 (late-classical) had already replaced the conventional generic passerby with a 'witness', Chronos, who was more suitably tailored for the task of making the glory of fallen soldiers evident. Literary epigrams of the 3rd cent. multiply the case of CEG 467 which remains for us an isolated experiment, by addressing, e.g., Arete on the tomb of Ajax (e.g. Asclep. AP 7.145=HE 946ff.), the sea for drowned people (e.g. Asclep. AP 7.284 = HE 950ff.), the sea birds in the case of a missing drowned sailor (e.g. Leon.Tar. AP 7.654 = HE 2048ff.), etc. For analogous reasons, the role of the passerby is also innovated by an address to, or a mention of, the burier in some epigrams for shipwrecked (Posid. 94 and 132 A.-B., Call. AP 7.277 = HE 1265ff.). Here again, it seems to me, the reworking of inscriptional conventions is, or pretends to be, triggered by some uncommon feature of the deceased's death: the dead were normally buried by their family or by the state, but this did not usually happen to the shipwrecked; hence there is a need to acknowledge an extra-family burier, who sneaks into the customary role of the passerby.

Chap. 4 "Dedicatory epigrams" investigates the re-use of language and motifs from funerary inscriptions and epigrams in the following works: a few dedicatory epigrams by Moero, Anyte, and Leonidas; two quite unconventional Leonidean dedications by the Cynic Sochares, who, as a believer in the Cynic credo of poverty, had nothing valuable to dedicate; dedications (e.g. by Leonidas and Callimachus) where the god is made to speak and interact with the traditional voice of the dedicator; and Callim. AP 6.121 = HE 1321ff. with its funny joke on the "Cynthiades" addressed in the incipit, who are not nymphs (e.g. of Artemis, mentioned at l. 2), as were, e.g., the "Deliades" of the Homeric Hymn to Apollo, but goats -- only at the end do we come to recognize that the epigram is not for an actual dedication, but instead is a reflection on the safe tranquility which the goats are allowed to enjoy after the past dedication of a hunter's arms to Artemis.

Chap. 5 "Sepulchral Innovations" focuses on the most substantial innovation to the inscriptional conventions, the epitymbia. One of the main conventions of archaic inscriptions was that the monuments could speak instead of the dead, which, in some epigrams of Callimachus (AP 7.317=HE 1269ff., 7.520 = 1199ff., 7.524 = 1187ff., 7.728 = 1255ff.), gave rise to dialogues or monologues where the dead provides information about the quality of life in the underworld or, at any rate, after death. Another convention of archaic and classical inscriptions was that the deceased was very seldom the speaker in cenotaphs (he seems to have had this role only in the 5th cent. CEG 166, from Sicinos). Therefore, Asclep. AP 7.500 = HE 954ff. probably had to rely on "established conventions for non-cenotaphic monuments" in order to create the speech of the deceased in a cenotaph (T., p. 112). In this case, however, T. is unusually elliptical as he does not explicitly state the models or parallels to Asclepiades (Nossis AP 7.718 = HE 2831ff., Call. AP 7.521 = HE 1237f., and Inscr. Métr. Égypte 30 Bernand, of the 3rd cent.). It would have also been useful to highlight that the poetics of 'epitaphic pride' which underlie the cenotaph of Asclepiades are remarkably similar to the message of CEG 166 mentioned above: in both texts the absence of the physical remains of the deceased (the bones inside the tomb) is emphasized and compared to the ability of the tomb to perpetuate his name.

Chap. 6 "Asclepiades and erotic epigram" reexamines some instances of the frequent, and also very well known, re-use of epitaphic language in erotic epigrams, which is seen especially in Asclepiades, but also in Callimachus.

Chap. 7 "Innovations in less common varieties of epigram" is devoted to the re-use of epitaphic and dedicatory conventions in the construction of several varieties of epigram which were most likely newly devised in the Hellenistic age and reveal the adoption of epigram as a form of communication for the practical needs of everyday life, in particular for the detailing of ritual prescriptions and ads for professional services (for instance, we know of money-lenders, interpreters of dreams, and seers). All the epigrams in this section come either from the corpus Theocriteum, a tank into which many non-Theocritean epigrams composed for real life use appear to have been poured, or from the papyrus of the "New Posidippus", an extremely large collection which was compiled during, or soon after, Posidippus' lifetime and was not affected by the literary bias of later anthologies (most of all the one by Meleager) favoring Posidippus' erotic and funerary epigrams.

Chap. 8 " 'Writtenness' and 'spokenness' in epigram". Discussions about the peculiarities and the limits of written communication were not uncommon in the classical age (cf. the most famous Plat. Phaedr. 275d), and the archaic CEG 286 seems to have anticipated this issue. The epigram of every age, T. correctly states (p. 141), appears to have a need "to compensate for its writtenness by designating a speaker, addressee, etc."; hence the epigrams' attention to providing written statements with various forms of dialogic variety. T. deals with several texts by "Erinna" and by Aratus where a reflection on both the greatness and the limits of the medium of writing is most evident.14

Chap. 9-13 are perhaps the most original pages of this thoroughly enjoyable book -- although some of T.'s analyses and results overlap with the two contributions of Irmgard Männlein-Robert in 2007, with which T. was understandably unfamiliar.15 Inscriptional epigrams were attached to an object, which in many cases represented something or someone. Already the archaic CEG 24, on a statue of the girl Phrasicleia which claimed to be the kore Phrasicleia, an eternal replacement of the deceased human girl, had conveyed the statue's ambition to replace representation with complete equation. The epigram on Midas' statue, which is probably archaic (AP 7.153= Pfohl 24) and introduces itself with the phrase "I am a bronze parthenos", proposes in a straightforward and unproblematic way the paradox that a maiden can be "bronze" -- even though maidens are humans, and humans are not bronze. T. devotes chap. 9 to these two archaic texts and to several other epigrams of the 3rd cent. in which the identity of the statue and the person it represents is not problematized (in particular Moero, and horse-monuments from the hippika of the New Posidippus). Other 3rd-cent. epigrams which pursue the motif of equalization without over-emphasizing its paradox are investigated in chap. 11 (in particular epigrams from the corpus Theocriteum which may have been epigrams used by professional versifiers). The emphasis on or the surface annihilation of that paradox, however, becomes the key point in many Hellenistic epigrams dealing with this motif. It comes as no surprise that apart from speaking the eulogies of the dead or dedicator in its own voice, the epigram also gives voice to the autonomous ambition of the statue to be a perfect equivalent of what it represents. What qualifies the Hellenistic innovation to the archaic "Phrasicleia psychology" is the adoption of a different and cunning rhetoric of persuasion -- the prolific variation on a theme by different authors can be seen in the numerous texts composed about the cow of Myron, all of which maintained that this cow was more real than a real cow and had no other raison d'être than competing to explain this supposed matter of fact in new ways. Chap. 10 mainly focuses on Nossis, who has four epigrams developing this motif, and on some other epigrammatists whom T. calls "followers" of Nossis (although a wide appreciation of this motif is more likely than a direct intertextual presupposition of Nossis). Especially intriguing in this chapter is the assimilation of Call. AP 9.507 = HE 1297ff. to this motif, the epigram which maintained that Aratus' poetry was entirely equal to its model of Hesiod. Despite the statement in l. 1, however, that the ἄεισμα and the τρόπος of Aratus are "of Hesiod", the poet's valediction in ll. 3-4 addresses the resulting(?) λεπταὶ ῥῆσιες as "of Aratus": is this not a distinction, I wonder, between the linguistic level of the utterance, which recognizably belongs to Aratus, and the overall Hesiodic mode/genre of the composition?16 Chap. 12 mainly focuses on Callimachus' most adventurous experiments with the same motif. For example, T. discusses the reversal of the ontological relationship between the statue and the person it represents in AP 5.146 = HE 1121ff., where the Graces are now said to be four, because Berenice has to be included - Berenice's existence, however, is not defined in terms of human "being", but in terms of being "molded" (ποτεπλάσθη), as if the statue of her or of the Graces preceded her/their human existence. T. also discusses the bronze rooster dedicated by Euainetos, who in AP 6.149 = HE 1161ff. admits that he did not witness the victory which he had been cast to advertise, but has to trust the man who commissioned him. Chap. 13 highlights another variation on the motif of the 'psychology' of statues: how the dialogue with the statue or its monologue sometimes involved (with a very Hellenistic flavor) the aetion of the statue itself (e.g. Posid. 120 or 142 A.-B., or, outside the epigrammatic genre, the self-confession of Apollo Delios about his own ontology in Call. Aet. 114).

From time to time T. approaches epigrams in a way which seems a bit elliptic and upsets the consistently high level of completeness which usually characterizes his discussions (see especially the case of Asclep. AP 7.500 = HE 954ff. pointed out above). These episodes of unbalanced or sketchy analysis of the texts, however, remain few. On the whole this book is a paradigm of solid and often brilliant scholarship, which should be read by every scholar who is interested in Hellenistic epigram (the rich index of passages also makes quick consultation very easy). It will also be a very educative tool in the hands of all graduate students interested in familiarizing themselves with Hellenistic poetry, and, more generally, of undergraduates who want to make use of an excellent guide to the technique of close reading.


1.   "Surprised by Self: Audible Thought in Hellenistic Poetry", CPh 85, 1990: 1-21.
2.   Respectively "Die Einbeziehung des Lesers in den Epigrammen des Kallimachos", in M.A. Harder, R.F. Regtuit, and G.C. Wakker (eds.), Callimachus (Hellen.Groning. 1), Groningen: 161-75; Inszeniertes Lesevergnügen: Das inschriftliche Epigramm und seine Rezeption bei Kallimachos, Stuttgart 2005 (which T. does not quote).
3.   The Epigrams Ascribed to Theocritus (Hellen.Groning. 5), Leuven 2001.
4.   M. Fantuzzi and R. Hunter, Tradition and Innovation in Hellenistic Poetry, Cambridge 2004.
5.   T.'s sound criticism of the assumption by Svenbro that, since in most cases the demonstrative ὅδε points self-referentially to the first person of the speaking object, it cannot refer to a third person is, however, especially remarkable.
6.   "Reading" Greek Death to the End of the Classical Period, Oxford 1995.
7.   Inscribing Sorrow: Fourth-Century Attic Funerary Epigrams, Berlin 2008.
8.   That is, an anonymous mourner who does not belong to the family of the dead, identified as a type of speaker by David M. Lewis ("Bowie on Elegy: A Footnote", JHS 107, 1987: 188) and Albio C. Cassio ("I distici del polyandrion di Ambracia e l' 'io' anonimo" nell'epigramma greco", SMEA 33, 1994: 101-117) on the basis of CEG 43, 51, 470, and "Simon." AP 7.511 = FGE 1006f.).
9.   T., pp. 153-4, prefers to suppose that there is a change of speaker, suggesting that the tomb addresses a passerby in 1-4 and that the passerby immediately responds with his comment in 5-6. But the same uncommon (and unexpected!) emphasis on the sadness of the dead's father and the lack of direct attention for the dead himself that characterizes Asclep. AP 13.23, however, can also be found in "Simon." AP 7.511 (already mentioned above), which is perhaps the most frequently quoted example of a literary epigram featuring the external mourner. This peculiar emphasis on parental grief might be, in my opinion, a specialized function of the voice of the first-person external mourner, who in this way would act as an 'externalizer' of the family's mourning.
10.   This last case is extremely telling, since Theognis leaves the "I"-centered dialogue with Cyrnos in 19-20 in order to move away from what would have otherwise been a self-presentation and to take advantage of the more objective perspective of a presentation by "everyone".
11.   Because of his/her isolation or disavowal: misanthropes, unburied shipwrecked, the prostitute Philainis; but also due to a more than human blessedness, as in the case of Call. AP 7.451 = HE 1231f., or a more than Greek cosmopolitism, as in the case of Meleag. AP 7.419 = HE 4000ff.).
12.   Cf. D. Obbink, " 'Tropoi' (Posidippus AB 102-103)", in B. Acosta-Hughes, E. Kosmetatou, and M. Baumbach (eds.), Labored in Papyrus Leaves, Cambridge MA and London 2004, 295.
13.   The real point, however, of this epigram of Callimachus for Timonoe is, in my opinion, the exceptional amount of information the passerby possesses. For he seems to criticize the uselessness of the portrait of the deceased on the stone and takes the opportunity to bear his own 'witness' in support of the information on the stone by personally confirming the heaviness of the husband's grief. T. appears to pay no attention to this peculiar focus of AP 7.522.
14.   Arat. AP 11.437 = HE 766f. receives particular interpretive efforts from T., although perhaps not the most successful ones. T. believes that the presentation of Diotimos who sits ἐν πέτραισι, "saying to the boys of Gargara βῆτα and ἄλφα", suggests the ineffectuality of Diotimos' tomb's attempts at speaking (in T.'s interpretation, "BA" would be an unintelligible sequence of letters). I agree with T. that, after a verb such as the initial αἰάζω, the phrase ἐν πέτραισι probably refers to (or at least hints at tombstones, as suggested already by P. Schubert, "A propos d'une épigramme d' Aratos sur Diotimos", Hermes 127, 1999, 501-3), rather than to rocky land. I also agree that here Diotimos is not an unsuccessful living "poet" (according to the lemma of the epigram) compelled to teach boys in this inhospitable land, who was sympathized by Aratus in an epigram which, although it adopts a funerary tone, was not actually funerary (as it has commonly been interpreted). In fact, I find it appealing (in agreement with T.) that instead this epigram reflects upon the tomb of Diotimos and his speaking image on it, but I do not see why his "saying βῆτα and ἄλφα" must be a failure in communication or why we need to get rid of the idea that Diotimos was a school-teacher. παισίν certainly invites an identification with this profession (otherwise why would he only be speaking to "boys"?). Furthermore, "teaching (?) ἄλφα βῆτα " was precisely what characterized the school teacher according to Callim. Iamb 5, fr. 195.3, from which we can also see that the job of the school-teacher was the object of condescension (σε δαίμων, in the same line, stigmatized this job as a misfortune sent by fate).
15.   "Epigrams on Art: Voice and Voicelessness in Hellenistic Epigram", in P. Bing and J. S. Bruss (eds.), A Companion to Greek Epigram, Leiden 2007, 251-271; Stimme, Schrift und Bild: Zum Verhältnis der Künste in der hellenistischen Dichtung. Heidelberg.
16.   In the case of Leon.Tar. AP. 6.355 = HE 2203ff. it would have been useful to highlight that the humor of the paradox that the child Micythos is dedicated by his mother while still "alive", ζωός (l. 1), is increased by the fact that the road to this degenerate dedication is paved by language itself. For in this epigram the syntactical splitting (cf. γραψαμένα at the end of l. 2) of the technical verb ζῶια γράφειν/ζωγραφεῖν (e.g. Herod. 4.88. ζῶια γραψάμενος πᾶσαν τὴν ζεῦξιν κτλ.), which idiomatically means merely "to paint", is combined with the pun, which is etymologically motivated, between ζῶιον "animal"/"picture" and ζωός "alive".

Comment on this review in the BMCR blog
Read Latest
Index for 2009
Change Greek Display
Books Available for Review
Bryn Mawr Classical Commentaries

BMCR, Bryn Mawr College, 101 N. Merion Ave., Bryn Mawr, PA 19010

HTML generated at 12:01:12, Friday, 18 September 2009