BMCR 2009.11.31

Metrics and Rhythmics: History of Poetic Forms in Ancient Greece (English translation by E. Christian Kopff of 2003 edition). Studi di metrica classica 12

, , Metrics and Rhythmics: History of Poetic Forms in Ancient Greece (English translation by E. Christian Kopff of 2003 edition). Studi di metrica classica 12. Pisa/Roma: Fabrizio Serra Editore, 2008. 350. ISBN 9788862270595. €120.00 (pb).

[The reviewer dedicated his Teubner edition of the Pindaric metrical scholia (1989) to Bruno Gentili, one of the authors of the book under review.]

In the middle of the last century Bruno Gentili wrote two important works on Greek metrics, Metrica greca arcaica (1950) and La Metrica dei Greci (1952). Despite the titles, neither was a “handbook” in the narrow sense of the term or a comprehensive synthesis of current research. On the contrary, the reader was faced with two works of original scholarship committed to re-thinking the gap between our sources on ancient theory and modern observatio, as practiced by scholars of Greek metrics in the first quarter of the twentieth century, in a way that would be productive for the contemporary reader of Greek lyric poetry. The same commitment inspires Metrics and Rhythmics, written in collaboration with his student, Liana Lomiento, a scholar already known for her important articles on philological and metrical issues.1 This book is a valuable source of information and, like its predecessors, combative and innovative. The successful translation by E. Christian Kopff, who has added his own clear “Foreword to the English Translation,” pp. 13-17, will give to Gentili’s and Lomiento’s positions the visibility they deserve and will provoke what I predict will be a lively debate on the issues they discuss.

The present work is a translation of the entire Italian edition (Milano 2003, reviewed in BMCR 2004.09.09 by J. B. Lidov) and preserves its division into three sections of unequal length: (I) “A Historical Rethinking” (pp. 27-110); (II) “Sung Poetry” (pp. 113-234); (III) “Recited Poetry and Recitative” (pp. 237-267). The volume ends with a rich bibliography and three indices: “Index of Subjects” (prepared by G. Parlato, pp. 291-304); “Index of Proper Names” (by M. Colantonio and L. Bravi, pp. 305-313) and finally “Index of Cited Passages” (also by M. Colantonio and L. Bravi, pp. 315-350).

The first section is divided into four parts, beginning with a brief but rich chapter in which Gentili and Lomiento delineate their theoretical position (“Metrics, Rhythmics and Music,” pp. 27-42), a succinct aperçu on “Prosody” (pp. 43-49), a very useful dictionary of “Metrical and Rhythmical Terminology and Diacritical Signs” (pp. 51-77) and an examination of “Structures of Versification” (79-88). This last section is based in large part on the terminology of Hephaestion and should be used with the dictionary that precedes it. The brief but important chapter that follows (“Modes of Performance,” pp. 89-110) not only gives an informative synthesis of the practice of performance and the instruments of antiquity, but along with the first chapter contributes to delineating Gentili’s and Lomiento’s vision of the relationship of meter, rhythm and performance. This vision, in my opinion, represents one of the book’s most interesting aspects. Gentili and Lomiento base their argumentation on an assumption that they are well aware is debatable (the recent and lively discussion by Prauscello 2006 should be read with L. Lomiento’s review, BMCR 2007.04.07 and Prauscello’s response, BMCR 2007.05.14). For Gentili and Lomiento the Hellenistic “poetry book,” and especially the division “in cola” of the lyric sections (which is structured by brief sequences, essentially in dimeters with occasional trimeters and tetrameters), preserves significant information about the traces of the original performances available to the Alexandrian philologists and is the basis of the slightly later metrical and rhythmic theory that has survived (Gentili and Lomiento p. 31). Therefore the division into cola of the melic sections of drama and choral lyric, the “colometry” (which continued to be found in the printed tradition up to the beginning of the nineteenth century) is not the work of inept “Grammatiker”,2 but preserves otherwise lost information about their rhythmic and musical articulations. Their positions are not “mainstream” in classical metrical theory of the last two centuries, a theory that in my opinion has taken Böckh’s discovery too far by dissolving the Böckhian “verse” into the only “real” melic sequence, the “period”. The distinctive traits of the period are supposed to be “inherited from the IE verse”, and its boundaries may not necessarily be established even by change of persona canens. (West 1982, 4 explained: “the compositional segments can no longer be called verses or lines, because they extend over many lines of the written text; the term period is used”. On p. 8 he continues, “the period is treated as a continuous piece of language, even if the end of a sentence or a change of speaker occurs within it”).3

Of course, Gentili and Lomiento discuss the problem of the relation between the manuscript “colometry” and the Böckhian “stichometry,” that is, Böckh’s re-arrangement of the lyrics into independent verses that are sometimes much longer than the limit of the trimeter. They cite opportunely the recent frontal attack of Willett 2002 on Böckh’s re-arrangements (really, however, on post-Böckhian scholarly excesses based on his definition of the period). Willett argues that sequences exceeding the sixteen syllables that constitute the extreme limit in the Indo-European tradition cannot be retained by “working memory”. Gentili and Lomiento, however, attempt to accommodate “colometry” and “stichometry” on p. 36: “We can, therefore, affirm that the Boeckhian period is a firm point even in the light of the correct ancient colometry” (see also “Colon, Comma, Verse,” pp. 59-60). We wonder if in the last analysis such an effort is really economical, whether it would not have been more logical at this point to take sides toto corde with the value of the colometric diastolai, or at least to denounce the metrisches Gespenst created by the nineteenth century “period”, all the more so because Böckh himself seems to have noticed ante litteram that there was a “working memory” limit to excessively extended melic series (Tessier 2008).

In my opinion, however, it would be unfairly reductionist to describe their position, which is only apparently “reactionary”, as a simple denial of modern classical metrics, soi disant Böckhian, or as a conflict between colon and stichos as the real forms of ancient melic articulation, because we are dealing up to this point with two theoretical assumptions that are in the last instance largely a matter of opinion. The second assumption, beginning with Böckh and continuing with the Fraenkel of Lyrische Daktylen (1918) and Maas (1923, final German version, 1929, English, 1962), cut off every tie with the tradition that accompanied these texts through Classical Antiquity and the Byzantine Middle Ages. As a result it can perhaps be judged, rebus peractis, as the (largely dated) effort to fit a discipline like Greek metrics, which seems proof against the attempt, into the naively revolutionary scientific and artistic Zeitgeist that animated the great hopes of the period that followed World War I. In the absence of probative elements, however, we can at least ask that an analysis of the rare and often ambiguous ancient sources on the subject respect an elementary scholarly logic.4

Beyond these apparently abstract conflicts, the contemporary reader of Greek texts composed for singing has another more pressing question: is the “descriptive” or deductive metrics, that is, the theory that claims to be based on simple observatio, really more respectful of these texts, as it should be on the basis of its premises? In more general terms, how justifiable is modern confidence in the science of metrics, actually privileging its own often self-imposed rules over other branches of philology, in establishing melic texts? It is here that Gentili’s and Lomiento’s approach reveals (at least to this reviewer) its most fruitful consequences, in the sense that the very tight analytic linking between metrical analysis and the superior rhythmic frame (superior because closer to what we know of the lost performance) allows the production of poetic texts without doing violence to the Greek language. I am alluding here, primarily, to the all too influential idolon responsionis, which reduces strophic melic texts to a precise “metrical” responsion at the cost of sometimes Procrustean interventions. For example, see the analysis of Bacchylides Ep. 5.160 (ep. 10) τάδ’ ἔφα θνατοῖσι μὴ φῦναι φέριστον, interpreted by Gentili and Lomiento (204) as ionmi 2 epitrtr in free “anaclastic” responsion with 3epitrtr (=stesich). Again, Snell’s first Teubner Bacchylides (1961) prints (instead of a demonstrative that ” metro non convenit“) an enigmatic τᾶδ‘ that has the advantage of giving the expected long syllable but raises difficult questions about Bacchylides’ poetic language. Wilamowitz suggested τᾷδ‘ and Lomiento 1990, 125 interprets it that way even while showing that it is not right, but if that is what the Teubner intended, it would follow its custom of printing the iota paragegrammenon. Of course, τᾶδε may well be the conventional way of marking a protraction of the short syllable to two beats (see Gentili’s and Lomiento’s important lemma “Monochronon”, pp. 68-69), but this interpretation seems extraneous to Snell’s intentions and later Maehler’s. For example, τᾶδε re-appears in Snell (and Maehler) at 5.191 ep. 1 again to “heal” a problem of strict responsion. The reading of most editors, τᾶδε φώνησεν, would produce, ion ma ia reiz a ~ ion ma chor (prosa) reiza (cf. Gentili and Lomiento 191 n.2). Not even assuming an isolated “anomaly” of responsion supports the validity of the linguistically strange τᾶδε, but an interpretation with “anaclastic” responsion based on rhythmical syngeneia seems to possess strong parallels. Look at Pindar, Pythian 12.8, where the free responsion found in the manuscripts between 3epitr tr (stesich) and cho 2epitr tr can be “healed” only at the cost of writing in v. 24 a linguistically incongruous ευκλεᾶ (Gentili and Lomiento 208 and n. 1, Tessier 1990, 187 ff.). In support of these interpretations one might add precise statements of ancient theory found in the form of excerpta in the scholia; see Schol. metr. Nem. VII ep 5 25, 19-20 Tessier, Schol. Ar. Ach. 1150b 143, 3-4 Wilson. Are we supposed to think that all this evidence are only traces of a system excogitated by ancient scholars at their writing desk to justify liberties of responsion that had insinuated themselves into the tradition and accidentally made rhythmical sense?

It is time to examine in more detail the second section, “Sung Poetry” (pp. 113-234). I give a synthesis of some peculiarities in their analyses, which are already well known to readers of B. Gentili’s fifty years of scholarship.

(1) “Dactyls” (113 ff.). Gentili and Lomiento do well to eschew the well known and popular tabu of Fraenkel 1918 (supported by Maas 1923) against the existence of independent acatalectic dactylic verses. So the Alcmanian (Gentili and Lomiento 117) and Ibycean (Gentili and Lomiento 121) can be found “with final adiaphoros“. Similarly Gentili and Lomiento not only split up into smaller segments the long holodactylic series (“cognitively” unacceptable according to Willett) that a certain style of editing has introduced into the parodos of Aeschylus’ Agamemnon against the transmitted colometry, but they re-introduce metrical “variety” against other editors’ “dactylic uniformity” (Gentili and Lomiento 120).

(2) “Choriambs” (155 ff.) and “Glyconic meters” (161 ff.). These two sections need to be treated together with the discussion of “Polyschematist Structures” (187 ff.). Gentili and Lomiento are not satisfied with the tacit reduction of choriambic and glyconic sequences favored by modern scholars under the influence of Hermann’s hyper-rationalism. On the contrary they vindicate the complete legitimacy of kata metron scansion even of glyconics, which they relate (in the wake of Körte 1922) to the antispast (162: “by no means […] a theoretical invention of the Alexandrian grammarians”). Gentili and Lomiento view the antispast and the choriamb as “metra of the same composition […] which belong to the same metrical family” (162). The same is true of polyschematist structures when they are forced into a kata metron structure. The metrical liberties of polyschematists (even when in responsion with apparently “genuine” ones) are justified “by the phenomenon of the epiploke which is the source of a wide variety of forms interconnected by a relationship of ‘family likeness'” (Gentili and Lomiento 192). In addition, in connection with the so-called “Sapphic hendecasyllable” (“the so-called epichoriambic form of the trimeter”) Gentili and Lomiento 157 rescue from the discussions in ancient theoretical treatises the notion of mixis kat’ antipatheian, which Hermann arbitrarily eliminated in his Elementa. Far from representing an ancient affectation, it needs to be understood in relation to the important lemma “Asynartetic” (Gentili and Lomiento 55 f.) that summarizes Gentili’s well known theory of asynarteta (see, of course, Gentili 1983). In this theory mixis constitutes, so to speak, the rhythmical “microcosm” between metra which inconexio represents between cola. It is worth adding that the recent research on the ancient sources by Savignago 2008 has successfully explored the apparent incoherencies of the remnants of ancient metrical theory.

(3) “Ionic Meters” (173 ff.). Gentili and Lomiento recognize the ionic a maiore as present in classical poetry, a significant change from Gentili 1950, 130, who was inclined to view this meter as a Hellenistic innovation and analyzed putative appearances in older poetry as an “anaclastic form of the first or second four syllables of the dodecasemic dimeter.”

(4) ” Κατ’ ἐνόπλιον Meters” (197 ff.). These sequences are analyzed as a third great rhythmic family in agreement with Damon’s doctrine and constitute for Gentili and Lomiento the key for understanding the so-called “dactylo-epitrites,” discussed in a long sub-section (202 ff.). Of course Gentili and Lomiento protest against the well known abusio of the term “epitrite” by Paul Maas, who analyzed the meter as a unity of five first times. Gentili and Lomiento denounce the danger implicit in a dactylic interpretation that inclines editors dogmatically “to eliminate as much as possible the free responsion present in the Bacchylides papyrus and in the manuscript tradition of Pindar’s odes” (204). Naturally Gentili’s and Lomiento’s interpretation involves freeing us from the imaginary anceps interpositum. Resolutions of the longum of the so-called “epitrite” are listed on p. 210. Dramatic texts too should be investigated from this perspective, e.g., Sophocles, Antigone 584=595, normalized by, of course, Hermann.

(5) “Dochmiacs” (227 ff.). This chapter presents in more detail than earlier formulations how much Gentili’s metrical exegesis differs from the mainstream understanding of dochmiacs, which has been inspired by the research of A. Seidler (1811-12) and limits the definition of this sequence to the “Attic” dochmiac and indeed attributes its invention to the Attic tragedians (with Snell’s well known “ethical and musical” motivations). On the contrary Gentili and Lomiento cite a long series of lyric “precursors” of the “Attic” dochmiac and list 38 attested forms beginning with the “three basic dochmiac schemata” (229 f.). West, on the other hand, accepts 21 forms (out of 32 theoretical ones) that developed from the “Attic” schema. Gentili and Lomiento include in their list forms “with resolution of the so-called ‘irrational long'”, forms linked to the hypodochmiac, and those connected to the Kaibelianum or “dochmiac prosodiac”. There is no reason to be surprised that in this case as well the libido coniectandi that followed Seidler’s dissertation has gone to work to reduce to its own abstract working hypothesis every “deviant” form. A significant victim of this metrics and its attendant editing has been the rejection of “spurious” responsion between “Attic” dochmiacs and “long” dochmiacs, sometimes with significant consequences for the text of the tragedians itself (Andreatta 1999).

I add some final observations. P. 140: the schema of the “resolved forms” of the ithyphallic (given as + + + ++_ – – ) lacks the short in the antepenultimate position. P. 147: Gentili and Lomiento mention Ar. Ach. 490-495 (= 566-571), where the “mesodic structure” composed of two trim ia inserted into the dochmiac system constitutes a “particularly ambiguous case” in respect to the dochmiacs that surround it. It is perhaps worth mentioning that the antistrophe, as found in the manuscripts, presents a difficult (and variously emended) responsion trim ia ~ 2 doch (492~568) and trim ia ~ doch dim ia (493~569). P. 225: (“Palimbacchaei”) Gentili and Lomiento cite as an instance of melic palimbacchiacs Aeschylus Ag. 1072 ὤπολλον ὤπολλον. This interpretation is perfectly plausible, of course, but one cannot exclude a priori that we are dealing with two lugubrious molossi with brevis in longo. P. 230: Soph. TrGF 27-29 ( Inachus) should be moved from the examples of dochmiac n. 23 (+ + – – + + +) to nn. 24c + + – + + – (v. 27 μανία τάδε κλύειν) and 22c + + – – + – (vv. 28 and 29: σὺ γὰρ οὖν Ζεῦ λόγων and κακὸς εἶ πίστεως).

In conclusion Gentili’s and Lomiento’s Metrics and Rhythmics, which is published in English in a particularly attractive typographic and editorial format, is precious for the very rich mass of literary texts and ancient theoretical testimonia presented and interpreted in a thoughtful fashion. As I said earlier, it would be extremely superficial to describe this book by reducing it to a conflict between “historicists” and “descriptivists” in Greek metrics. In fact a reader who observes how Gentili and Lomiento base their theoretical formulations on texts can see that the advantages for the reader compensate amply for the occasional dissent.


Andreatta, L. 1999. “Normalizzazione del docmio lungo ‘strofico’ nel testo sofocleo,” in G. Avezzù (cur.), Διδασκαλίαι Tradizione e interpretazione del dramma attico, Padova (Studi Testi Documenti del Dipartimento di Scienze dell’Antichità dell’Università di Padova 9): 113-162

Andreatta, L. 2008. “[Haud] integros accedere fontis.” Testimonianze sparse sui carmi kata; scevsin,” Paideia 63: 29-57

Böckh, A. 1820-22. Über die kritische Behandlung der Pindarischen Gedichte, [gelesen in der Akademie der Wissenschaften zu Berlin am 3. Februar 1820, 13. Juli 1821 und 7. März 1822], 261-400 (= Gesammelte kleine Schriften, V, hrsg. v. P. Eichholtz, E. Bratuscheck, Leipzig 1871, 248-396

Gentili, B. 1950. Metrica greca arcaica, Messina-Firenze

Gentili, B. 1952. La Metrica dei Greci, Messina-Firenze

Gentili, B. 1983. “L’asinarteto nella teoria metrico-ritmica degli antichi,” in Festschrift für Robert Muth, hrsg. P. Händel – W. Meid, Innsbruck: 135-143

Fraenkel, E. 1918. “Lyrische Daktylen,” Rheinisches Museum 72, 1918: 161-197, 321-352 (= Kleine Beiträge zur klassischen Philologie, I, Roma 1964: 165-233)

Körte, A. 1922. Review of Wilamowitz 1921, Neue Jahrbücher für die klassische Altertum 25, 1922: 313-330

Lomiento, L. 1990. “Bacchilide: una nuova traduzione e ancora un contributo agli studi sull’epinicio,” Quaderni Urbinati di Cultura Classica 35 n. 2, s. c. 64: 121-132

Maas, P. 1923. Griechische Metrik, in A. Gercke – E. Norden, (Hrsg.), Einleitung in die Altertumswissenschaft, 1. Bd / 7. Heft, Leipzig-Berlin

Maas, P. 1929. Griechische Metrik, unveränderter durch Nachträge vermehrter Neudruck, in A. Gercke – E. Norden, (Hrsg.), Einleitung in die Altertumswissenschaft, 1. Bd / 7. Heft, Leipzig-Berlin

Maas, P. 1962. Greek Metre, translated by Hugh Lloyd-Jones, Oxford

Prauscello, L. 2006. Singing Alexandria. Music between Practice and Textual Transmission, Leiden-Boston 2006 – Mnemosyne Suppl. 274

Rossi, L. E. 1966. “La metrica come disciplina filologica,” Rivista di Filologia e Istruzione Classica 94: 185-207

Rossi, L. E. 1975. “Verskunst,” in Der kleine Pauly. Lexikon der Antike, München: Bd. V, 1210-1218

Savignago, L. 2008. ” ‘Epichoriambikón’: diacronia di usi e fraintendimenti,” Paideia 63: 307-331.

Tessier, A. 1999. “La normalizzazione metrica di Pindaro negli strumenti lessicografici (postille a Pitica 12),” Lexis 17, 183-189

Tessier, A. 2008. “‘Sticometria’ e misura del verso melico greco: Böckh,” Quaderni Urbinati di Cultura Classica 88 n. 3, s. c. 117: 121-124

West, M.L. 1982. Greek Metre, Oxford

von Wilamowitz -Moellendorff, U. 1900. Textgeschichte der griechischen Lyriker, Berlin

von Wilamowitz -Moellendorff, U. 1921. Griechische Verskunst, Berlin

Willett, S.J. 2002. “Working Memory and its Constraints on Colometry,” Quaderni Urbinati di Cultura Classica 71 n. 2, s. c. 100: 7-19


1. See the Bibliography, pp. 279-280.

2. Thus Wilamowitz (1900, 7) in a typically apodictic fashion that has proven undeservedly popular. In reality, he takes the term “Grammatiker” (but not the subject) from Böckh, 1820-22, 301, for whom “keine Abtheilung, wie sie überliefert worden, ein geschichtliches Ansehen hat, weil keine ins höhere Alterthum reicht” [“No colometry as transmitted deserves historical attention since none goes back to distant antiquity”]. Wilamowitz, however, should have explained why, if Böckh was right, such diastolai usually conflict with the syntactic articulations and are found sometimes in verbal synaphaea, as Dionysius of Halicarnassus CV 26, 15 clearly pointed out about Simonides’ “Danae’s Lament,” a passage often cited but rarely taken seriously. (Unfortunately Dionysius is often underestimated as a witness to ancient rhythm and music by scholars who seem to think that he started à rebours from imperial metrical theory). So Prauscello 2007, 23 writes: “Dionysius does not specify anywhere in the text the astrophic or antistrophic nature of the lyric song he is quoting”. Less biased readings, like Andreatta 2008 on strophic responsion, reveal how basically unacceptable that vision is.

3. It is worthwhile comparing West’s definition of the “period” with that of Rossi 1966, 190 n.1 (in my opinion, the latter shows more respect for the poetic reality of the archaic and classical ages, and I would even say of every poetic text structured in verses ): “Boeckh…parlava chiaramente di versi, e non di periodi, come sembra credere B. Snell….La confusione deriva dal considerare periodo quello che è costituito da semplici cola : ma normalmente i cola, quando sono veramente tali, costituiscono il verso: e il periodo è l’unità immediatamente superiore al verso….Del resto, l’equivoco verso-periodo è anche sensibile, sul piano pratico, nelle edizioni di Bacchilide e di Pindaro” [“Boeckh […] was clearly talking about verses, not periods, as B. Snell seems to believe. […] The confusion derives from thinking of a period as composed of simple cola: normally cola, when they really are cola, constitute the verse: and the period is the unity immediately above the verse. […] Anyhow, the confusion of verse and period can be felt, on the practical level, in editions of Bacchylides and Pindar”]. See in addition the very clear demarcation between verse and period in Rossi 1975, 1211: “Kolon (Bestandteil einer höheren Einheit)…Vers (kleinste selbständige Einheit)…Periode (vom Rhythm. bedingte Vers- oder Systemgruppe kleiner als die Strophe)…die Periode war durch Musik und Tanz anschaulich gemacht” [” Kolon (part of a higher unity) […] Verse (smallest independent unity) […] Period (a group of verses or systems conditioned by rhythm smaller than the strophe) […] The period was made clear by music and dance”].

4. A fruitful example of insisting on the necessity of this logical exercise seems to this reviewer E. Christian Kopff’s review of Prauscello in Classical World 102 (2008) 82-83.