i Pindarís Prosodia Introduction, text, and commentary to selected fragments Enrico Emanuele Prodi Merton College
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
i
Pindarís Prosodia Introduction, text, and commentary
to selected fragments
Enrico Emanuele Prodi Merton College
iii
In fulfilment of a vow made on 12th May 2009, this dissertation is dedicated to the memory of James Worthen.
May he rest in peace.
i
Table of contents
Acknowledgements iii
Abbreviations v
Introduction 1
1 ñ The book, I : Testimonia 3
2 ñ The book, II : Contents, structure, scholarship 9
3 ñ The genre, I : Preliminary remarks 27
4 ñ The genre, II : ëDirect traditioní 39
5 ñ The genre, III : ëIndirect traditioní 67
6 ñ Genre and book : Pindarís Prosodia 95
7 ñ The manuscripts 119
Text 139
Commentary 171
F1 173
F2 183
F3 205
F5 227
*F6 301
*F7 307
Bibliography 327
iii
Acknowledgements My gratitude goes first of all to my supervisors, Bruno Currie and Dirk Obbink, who have never spared expertise, support, advice, encouragement, good cheer, and an enormous amount of patience for four years and 99,317 words (not counting the Greek). Thanks are also due for their generosity and kindness to Daniela Colomo at the Papyrology Room of the Sackler Library, who has always been at hand to discuss questions and solve problems, Andrea Clarke and the Manuscripts Reading Room staff at the British Library, Rosario Pintaudi and Giovanna Rao at the Biblioteca Medicea Laurenziana, Fabian Reiter at the >gyptisches Museum und Papyrussammlung in Berlin, and the amiable staff at the Section Grecque of the IRHT in Paris, all of whom allowed me access to papyri, photographs, and microfilms with a true spirit of amicitia papyrologorum (and palaeographorum). I am very grateful to the Bodleian and Sackler Libraries and the libraries of Merton College, Christ Church, and Corpus Christi College in Oxford, the ever-hospitable Dipartimento di Filologia Classica e Italianistica (quondam Medievale) of the University of Bologna, the Fondation Hardt pour líBtude de líAntiquitC Classique, and the British Library, in each of which much work was carried out with excellent resources in very charming surroundings. For their help, advice, and support, scholarly or otherwise, during the writing of this thesis, I am proud to thank first Christopher Metcalf, who read and improved several chapters; then (in purely alphabetical order) Chiara Aimi, Ilaria Andolfi, Roberto Batisti, Ewen Bowie, Vanessa Cazzato, Thomas Coward, Paola DíAndrea, Geri Della Rocca de Candal, Barbara Fero, Caterina Franchi, Tristan Franklinos, Ilaria Gianani, Sarah Harden, Theodora Hadjimichael, Jo Heirman, Alberta Lorenzoni, Richard Rawles, Alberto Rigolio, David Sider, Henry Spelman, Nicolas Stone Villani, Alessandro Vatri, and Martin L. West; and Vittoria Mainoldi, Francesco Pichi, and Chris Sheppard, who in spite of geography have never been far. It is the funded studentís burdensome duty to thank the several Institutions, Societies, and Research Councils that allowed them to pursue their doctorate; I am therefore much obliged to the several Institutions, Societies, and Research Councils that took good care not to impose such an onerous incumbence on my unworthy shoulders. However, the Europaeum generously funded the first year of my graduate studies through a Jenkins Scholarship; Merton College, Christ Church, the Faculty of Classics, and the Society for the Promotion of Hellenic Studies funded my attendance at several conferences, and the SPHS also made possible my stay at the Fondation Hardt in September 2012 and the purchase of photographic reproductions of H4 from the British Library; with unfailing encouragement Juliane Kerkhecker and Matthew Leigh trusted me to teach their students over several years; finally, a Junior Research Fellowship at Christ Church allowed me to bring my doctorate to completion with no financial worries and in the best possible conditions. In all this I cannot forget that my studies up to and including my undergraduate degree were almost entirely funded by a State which, despite difficulties of every order, still believes that an excellent education must be made available to all. None of all this would have been possible without my mother Anna Gasperi-Campani, who encouraged and supported me in every sphere of life for twenty-eight years and counting, and without whose encouragement and support, material as well as intellectual, this doctorate (and all of my studies) could never have taken place. The greatest debt is owed to her.
iv
v
Abbreviations
Greek authors are abbreviated as in LSJ, Latin authors as in the Thesaurus linguae Latinae. Periodicals are abbreviated as in LíAnn�e Philologique, papyri other than those of Pindar or Bacchylides as in the Checklist of Greek, Latin, Demotic and Coptic Papyri, Ostraca and Tablets <http://scriptorium.lib.duke.edu/ papyrus/texts/clist.html>, and Greek inscriptions according to the most recent index volume of the SEG. Manuscripts, and the papyri of Pindar and Bacchylides, are cited with the sigla employed by the editor cited or in the edition more commonly used.
Adler Suidae Lexicon edidit A. Adler, I-V, Lipsiae 1928-38
Alt-'gina Alt-'gina herausgegeben von H. Walter und E. Walter-Karydi, Mainz am Rhein 1974ñ
ABV J.D. Beazley, Attic Black-Figure Vase-Painters, Oxford 1956
ARV2 J.D. Beazley, Attic Red-Figure Vase-Painters, I-III, Oxford 1963
Aucher Eusebii Pamphili Caesariensi episcopi Chronicon bipartitum, nunc primum ex Armeniaco textu in Latinum conversum, adnotationibus auctum, Graecis fragmentis exornatum opera J.B. Aucher, I-II, Venetiis 1818
Bekker1 Photii Bibliotheca ex recensione I. Bekkeri, I-II, Berolini 1824-5
Bekker2 Apollonii Sophistae lexicon Homericum ex recensione I. Bekkeri, Berolini 1833
BerlPap BerlPap. Berliner Papyrusdatenbank <http://smb.museum/berlpap/>
BernabC Poetarum epicorum Graecorum testimonia et fragmenta edidit A. BernabC, I, cum appendice iconographica a R. Olmos confecta, Stutgardiae et Lipsiae 19962
Bethe Pollucis Onomasticon e codicibus ab ipso collati denuo edidit et adnotavit E. Bethe, I-III, Lipsiae 1900-37
BNJ Brillís New Jacoby < http://referenceworks.brillonline.com/browse/brill-s-new-jacoby >
Braswell B.K. Braswell, Didymos of Alexandria. Commentary on Pindar (SBA 41), Basel 2013
CA Collectanea Alexandrina. Reliquiae minores poetarum graecorum aetatis
Ptolemaicae, 323-146 a.C., epicorum, elegiacorum, lyricorum, ethicorum cum epimetris et indice nominum edidit I.U. Powell, Oxonii 1925
CAG Commentaria in Aristotelem Graeca, I-XXIII, Berolini 1883-1909
CAH Cambridge Ancient History, I-XIV, Cambridge 19703-20052
vi
cap. Capitula ad praefationem pertinentia, in Dr. pp. III 306-11
Carey Lysiae orationes cum fragmentis recognovit brevique adnotatione critica instruxit C. Carey, Oxonii 2007
CLGP Commentaria et lexica Graeca in papyris reperta ediderunt G. Bastianini M. Haslam H. Maehler F. Montanari C. RNmer adiuvante M. Stroppa, MOnchen / Leipzig 2004ñ
Consbruch Hephaestionis Enchiridion cum commentariis veteribus edidit M. Consbruch. Accedunt variae metricorum Graecorum reliquiae, Lipsiae 1906
Cunningham CPQRSTSU VWXYTQ Z[\]^_TQ. Texts of the original version and of Ms. B, edited by I.C. Cunningham, Berlin 2003
DGE Diccionario griego-espa7ol redactado bajo de la direcci`n de F. R. Adrados, Madrid 1980ñ
DELG P. Chantraine, Dictionnaire �tymologique de la langue grecque. Histoire des mots, achevC par J. Taillardat, O. Masson et J.-L. Perpillou avec, en supplement, les Chroniques dí�tymologie grecque (1-10) rassemblCes par A. Blanc, C. de Lamberterie et J.-L. Perpillou, Paris 20092
Dindorf1 Aristides ex recensione G. Dinforfii, III, Lipsiae 1829
Dindorf2 Scholia Graeca in Homeri Odysseam ex codicibus aucta et emendate edidit G. Dindorfius, I-II, Oxonii 1855
Dr. Scholia vetera in Pindari carmina recensuit A. B. Drachmann, I-III, Lipsiae 1903-27
EDG R. Beekes, with the assistance of Lucien van Beek, Etymological Dictionary of Greek (IEED 10), I-II, Leiden / Boston 2010
EOropou ab Wc^S[RdWe fgP h[icgP Wjk^kgQfRb Pco l.m. HWf[njgP, opqQRb 1997
Erbse Scholia Graeca in Homeri Iliadem (scholia vetera) recensuit H. Erbse, I-VII, Berolini 1969-88
Fabbro Carmina convivalia Attica edidit H. Fabbro, Romae 1995
FGE Further Greek Epigrams. Epigrams before A.D. 50 from the Greek
Anthology and Other Sources not Included in Hellenistic Epigrams or The Garland of Philip, edited by D.L. Page, revised and prepared for publication by R.D. Dawe and J. Diggle, Cambridge 1981
FGrHist Die Fragmente der griechischen Historiker, herausgegeben von F. Jacoby, Leiden 1923ñ
vii
FHG Fragmenta historicorum Graecorum collegit, disposuit, notis et prolegomenis illustravit, indicibus instruxit C. Mullerus, I-V, Parisiis 1841-70
Fowler R.L. Fowler, Early Greek Mythography, I: Texts, Oxford 2000
Gaisford Etymologicum Magnum, seu verius lexicon saepissime vocabulorum origins
indagans ad codd. mss. recensuit et notis variorum instruxit T. Gaisford, Oxonii 1848
GL Grammatici Latini ex recensione H. Keilii, I-VIII, Lipsiae 1857-80
GMAW2 E.G. Turner, Greek Manuscripts of the Ancient World, edited by P.J. Parsons (BICS suppl. 46), London 19872
Goettling rWgkgs^gP S[R__Rfbjgt cW[Ú S[R__Rfbjve. Theodosii Alexandrini grammatica, edidit et notas adiecit C.G. Goettling, Lipsiae 1822
Gostoli Terpander edidit A. Gostoli, Romae 1990
Gow-Page The Greek Anthology. The Garland of Philip, and Some Contemporary Epigrams, edited by A.S.F. Gow and D.L. Page, Cambridge 1968
GP2 J.D. Denniston, Greek Particles, second edition revised by K.J. Dover, Oxford 1954
Hansen Hesychii Alexandrini Lexicon, editionem post K. Latte continuans recensuit et emendavit P.A. Hansen, III, Berlin / New York 2005
Hansen-Cunningham Hesychii Alexandrini Lexicon, editionem post K. Latte continuantes recensuerunt et emendaverunt P.A. Hansen et I.C. Cunningham, IV, Berlin / New York 2009
HE The Greek Anthology. Hellenistic Epigrams edited by A.S.F. Gow and D.L. Page, I-II, Cambridge 1965
Helm Eusebius Werke, siebenter Band: Chronik des Hieronymus. Hieronymi Chronicon, herausgegeben und in 2. Auflage bearbeitet von R. Helm, Berlin 19562
Hilgard Grammatici Graeci recogniti et apparatu critico instructi, III: Scholia in Dionysii Thracis Artem grammaticam, recensuit et apparatum criticum indicesque adiecit A. Hilgard, Lipsiae 1901
Holwerda1 Scholia in Aristophanem edidit edendave curavit D. Holwerda, I: Prolegomena de comoedia. Scholia in Acharnenses, Equites, Nubes, III/1: Scholia vetera in Nubes edidit D. Holwerda, cum duabus appendicibus, quas subministravit W.J.W. Koster, Groningen 1977
viii
Holwerda2 Scholia in Aristophanem edidit edendave curavit D. Holwerda, II: Scholia in Vespas; Pacem; Aves et Lysistratam, 3: Scholia vetera et recentiora in Aristophanis Aves edidit D. Holwerda, Groningen 1991
I.Teos D.F. McCabe and M.A. Plunkett, Teos Inscriptions. Texts and List, Princeton 1985
I.Thesp. P. Roesch, Les Inscriptions de Thespies (IThesp), I-XIII, 20092 <http://www.hisoma.mom.fr/production-scientifique/les-inscriptions-de-thespies>
Jones-Wilson Scholia in Aristophanes edidit edendave curavit W.J.W. Koster, I: Prolegomena de comoedia. Scholia in Acharnenses, Equites, Nubes, 2: Scholia vetera in Aristophanis Equites edidit D.M. Jones et Scholia Tricliniana in Aristophanis Equites edidit N. Wilson, Groningen/Amsterdam 1969
K. Tragicorum Graecorum fragmenta (TrGF), V: Euripides, editor R. Kannicht, I-II, GNttingen 2004
K.-A. Poetae comici Graeci, ediderunt R. Kassel et C. Austin, I-VIII, Berolini et Novi Eboraci 1983ñ
Kambylis Eustathios von Thessalonike. Prooimion zum Pindarkommentar, Einleitung, kritischer Text, Indices besorgt von A. Kambylis, GNttingen 1991
Karst Eusebius Werke, V: Die Chronik, aus dem Armenischen Obersetzt mit textkritischem Kommentar herausgegeben Dr. J. Karst, Leipzig 1911
Keil Aelii Aristidis Smyrnaei quae supersunt omnia edidit B. Keil, I-II, Berolini 1898
Keller Pseudacronis scholia in Horatium vetustiora recensuit O. Keller, I-II, Lipsiae 1902-4
Koster1 Scholia in Aristophanem edidit edendave curavit D. Holwerda, IV: Jo. Tzetzae commentarii in Aristophanem ediderunt L. Massa Positano, D. Holwerda, W.J.W Koster, 3: Commentarium in Ranas et Aves. Argumentum Equitum quae edidit W.J.W. Koster, Groningen / Amsterdam 1962
Koster2 Scholia in Aristophanem edidit edendave curavit D. Holwerda, II: Scholia in Vespas; Pacem; Aves et Lysistratam, 1: Scholia vetera et recentiora in Aristophanis Vespas, Groningen 1978
Lasserre1 Plutarque. De la musique, texte, traduction, commentaire prCcCdCs díune Ctude sur Lí�ducation musicale dans la Gr@ce antique par F. Lasserre, Olten / Lausanne 1954
Lasserre2 Die Fragmente des Eudoxos von Knidos herausgegeben, Obersetzt und kommentiert von F. Lasserre, Berlin 1966
ix
Lasserre-Livadaras Etymologicum magnum genuinum, Symeonis Etymologicum una cum Magna
grammatica, Etymologicum magnum auctum synoptice ediderunt F. Lasserre et N. Livadaras, I-II, Roma 1976ñ
Latte Hesychii Alexandrini Lexicon recensuit et emendavit K. Latte, I-II, Hauniae 1953-66
LDAB Leuven Database of Ancient Books < http://www.trismegistos.org/ldab/>
Lentz Herodiani technici reliquiae collegit, disposuit emendavit explicavit praefatus est A. Lentz, I-II.2, Lipsiae 1867-70
LfgrE Lexikon des frBhgriechischen Epos, begrOndet von B. Snell, GNttingen 1955-2010
LGGA Lessico dei grammatici greci antichi <http://www.aristarchus.unige.it/lgga/index.php>
LIMC Lexicon iconographicum mythologiae classicae (LIMC), I/1-VIII/2, 1981ñ
LSCG F. Sokolowski, Lois sacr�es des cit�s grecques, Paris 1969
LSJ A Greek-English Lexicon, compiled by H.G. Liddell and R. Scott, revised and augmented throughout by Sir H.S. Jones with the assistance of R. McKenzie, Oxford 19409
M. Bacchylides, edidit H. Maehler, Monachi et Lipsiae 200311
M.-W. Fragmenta Hesiodea ediderunt R. Merkelbach et M.L. West, Oxonii 1967
Manieri Agoni poetico-musicali nella Grecia antica. I: Beozia, a cura di A. Manieri, Pisa / Roma 2009
Meyer Pomponii Porphyrionis commentarii in Q. Horatium Flaccum, recensuit G. Meyer, Lipsiae 1874
Montanari Ptolemaios Epithetes, in F. Montanari, I frammenti dei grammatici Agathokles, Hellanikos, Ptolemaios Epithetes. In appendice i grammatici
Theophilos, Anaxagoras, Xenon; Lesbonax HW[Ú ]Z\_nfTQ edited with an introduction and commentary by D.L. Blank; The Fragments of Comanus of Naucratis edited by A.R. Dyck (SGLG 7), Berlin / New York 1988, 75-112
MP3 P. Mertens and R.A. Pack, Catalogue des papyrus litt�raires grecs et latins, <http://promethee. philo.ulg.ac.be/cedopal/index.htm>
Nachtergael Corpus des actes relatifs aux SEt�ria de Delphes, in Nachtergael (1977) below, 391-495.
Nickau Ammonii qui dicitur liber de adfinium vocabulorum differentia edidit K. Nickau, Lipsiae 1966
x
NP Der neue Pauly: EncyclopGdie der Antike, herausgegeben von H. Ciancik und H. Schneider, I-XIX, Stuttgart 1996-2003 (Engl. tr. Brillís New Pauly: Encyclopaedia of the Ancient World, Leiden 2002ñ, also online at <http://referenceworks.brillonline.com/browse/brill-s-new-pauly>)
Pack Artemidori Daldiani Onirocriticon libri V recognovit R.A. Pack, Lipsiae 1963
Pf. Callimachus edidit R. Pfeiffer, I-II, Oxonii 1949
PMG Poetae melici Graeci edidit D.L. Page, Oxford 1962
PMGF Poetarum melicorum Graecorum fragmenta, I: Alcman Stesichorus Ibycus post D.L. Page edidit M. Davies, Oxonii 1991
Poltera O. Poltera, Simonides Lyricus. Testimonia und Fragmente, Einleitung, kritische Ausgabe, xbersetzung und Kommentar (SBA 35), Basel 2008
Pontani Scholia Graeca in Odysseam edidit F. Pontani, Roma 2007ñ
POxy Online The Oxyrhynchus Papyri Project, <http://www.papyrology.ox.ac.uk/POxy/>
PSI Online PSIonline. Papiri della SocietI Italiana, <www.psi-online.it>
R. Rutherford (2001) below
Rabe1 Scholia in Lucianum edidit H. Rabe, Lipsiae 1906
Rabe2 Ioannis Sardiani commentarium in Aphthonii Progymnasmata edidit H. Rabe, Lipsiae 1928
Radt Tragicorum Graecorum fragmenta (TrGF), III: Aeschylus, editor S. Radt, GNttingen 19992
RE Paulys Real-EncyclopGdie der klassischen Altertumswissenschaft, Stuttgart 1894-1978
Regtuit Scholia in Aristophanem edidit edendave curavit D. Holwerda, III: Scholia in Thesmophoriazusas, Ranas, Ecclesiazusas et Plutum, 2/3: Scholia in Aristophanis Thesmophoriazusas et Ecclesiazusas edidit R.F. Regtuit, Groningen 2007
Rose Aristotelis qui ferebantur librorum fragmenta collegit V. Rose, Lipsiae 1886
Russell-Wilson Menander Rhetor, edited with translation and commentary by D.A. Russell and N.G. Wilson, Oxford 1981
Rzach Hesiodi Carmina recensuit A. Rzach, accedit Certamen quod dicitur Homeri et Hesiodi, Lipsiae 19133
Saffrey-Segonds Marinus. Proclus ou Sur le bonheur, texte Ctabli, traduit et annotC par H.D. Saffray et A.-P. Segonds avec la collaboration de C. Luna, Paris 2002
xi
SB Sammelbuch griechischer Urkunden aus 'gypten begrOndet von F. Preisigke, 1915ñ
Scheer Lycophronis Alexandra recensuit E. Scheer, I: Scholia, Berolini 1908
Schmidt Didymi Chalcenteri grammatici Alexandrini fragmenta quae supersunt
omnia collegit et disposuit M. Schmidt, Lipsiae 1854
Schneider Grammatici Graeci recogniti et apparatu critico instructi, I: Apollonii Dyscoli quae supersunt, 1: Apollonii scripta minora a R. Schneidero edita continens, Lipsiae 1878
Schwartz Scholia in Euripidem collegit recensuit edidit E. Schwartz, I-II, Berolini 1887-91
Severyns Recherches sur la Chrestomathie de Proclos, premiyre partie: Le codex 239 de Photius, I: Ltude pal�ographique et critique, II: Texte traduction commentaire, Liyge / Paris 1938
SH Supplementum Hellenisticum ediderunt H. Lloyd-Jones P. Parsons. Indices in hoc Supplementum necnon in Powellii Collectanea Alexandrina confecit H.-G. Nesselrath, Berolini et Novi Eboraci 1983
Slater Aristophanis Byzantii Fragmenta post A. Nauck collegit, testimoniis ornavit, brevi commentario instruxit W. J. Slater, Berolini / Novi Eboraci 1986
SLG Supplementum lyricis Graecis. Poetarum lyricorum Graecorum fragmenta
quae recens innotuerunt edidit D. Page, Oxford 1974
SSH Supplementum Supplementi Hellenistici edidit H. Lloyd-Jones, indices confecit M. Skempis, Berolini et Novi Eboraci 2005
Sn.-M. Pindari carmina cum fragmentis post B. Snell edidit H. Maehler, I-II, Leipzig 19878-98
Stephanis ∏.{. |fWdRQqe, }bgQPsbRjgÚ fWZQ~fRb. |�_�gVWe sfUQ c[gsTcgS[Rd^R fgt pWnf[gP jRÚ fve _gPsbjve f�Q �[ZR^TQ �VVqQTQ, �[njVWbg 1988
Sturz1 Orionis Thebani Etymologicum ex museo F.A. Wolfii primum edidit, annotationes P.H. Larcheri eiusdem Wolfii nonnullas et suas, indicesque locupletissimos adiecit F.G. Sturzius, Lipsiae 1820
Sturz2 Etymologicum Graecae linguae Gudianum et alia grammaticorum scripta e
codicibus manuscriptis nunc primum edita, accedunt notae ad Etymologicon magnum ineditae E. H. Barkeri, I. Bekkeri, L. Kulencampii, A. Peyroni aliorumque quas digessit et una cum suis edidit F.G. Sturzius cum indice locupletissimo, Lipsiae 1818
Tessier Scholia metrica vetera in Pindari carmina edidit A. Tessier, Leipzig 1989
xii
Thalheim Isaei orations cum deperditarum fragmentis edidit T. Thalheim, Stutgardiae 19032
Theodoridis Photii patriarchae lexicon edidit C. Theodoridis, Berlin / Boston 1982ñ
Thilo-Hagen Servii grammatici qui feruntur in Vergilii carmina commentarii, recens. G. Thilo et H. Hagen, I-III/2, Lipsiae 1878-1902
ThGL r\]RP[Ù] fv] �VV\Qbjv] SVi]]\]. Thesaurus Graecae ab Henrico Stephano constructus. Post editionem Anglicam novis additamentis auctum, ordineque alphabetic digestum ediderunt C.B. Hase, G. Dindorfius et L. Dindorfius, I-VIII, Parisiis 1831-65
Tittmann Iohannis Zonarae lexicon ex tribus codicibus manuscriptis nunc primum edidit, observationibus illustravit et indicibus instruxit I.A.H. Tittmann, Lipsiae 1808
V. Sappho et Alcaeus edidit E.-M. Voigt, Amsterdam 1971
VA Vita Pindari Ambrosiana in Dr. pp. I 1-3
van der Valk Eustathii archiepiscopi Thessalonicensi commentarii ad Iliadem pertinentes ad fidem codicis Laurentiani editi, curavit M. van der Valk, I-IV, Ludguni Batavorum 1971-87
Vogel }bgki[gP �b�Vbgpqj\ �]fg[bjq. Diodori biblioteca historica recognovit F. Vogel, I-III, Lipsiae 1888-93
VS Die Fragmente der Vorsokratiker, gr. und deutsch von H. Diels, Wortindex verfasst von W. Kranz, I-II, Berlin 19123
VV Vita Pindari Vaticana (also known as Thomana) in Dr. pp. I 4-8
W. Iambi et elegi Graeci ante Alexandrum cantati, edidit M. L. West, I-II, Oxford 19892-922
Wachsmuth-Hense Ioannis Stobaei Anthologium recensuerunt C. Wachsmuth et O. Hense, I-V, Berlin 1884-1912
Wehrli Die Schule des Aristoteles. Texte und Kommentar von F. Wehrli, I-X, Basel 19672-92
Wendel1 Scholia in Theocritum vetera recensuit C. Wendel, adiecta sunt Scholia in Technopaegnia scripta, Lipsiae 1914
Wendel2 Scholia in Apollonium Rhodium vetera recensuit C. Wendel, Berolini 1935
Wessner Aeli Donati quod fertur commentum Terenti. Accedunt Eugraphi
commentum et scholia Bembina, recensuit P. Wessner, I-III/1, Lipsiae 1902-8
xiii
West The Contest of Homer and Hesiod, in Homeric Hymns. Homeric Apocrypha. Lives of Homer, edited and translated by M.L. West, Cambridge MA / London 2003, 318-53
Westphal Scriptores metrici Graeci, I: Hephaestionis De metris enchiridion et De poemate libellus cum scholiis et Trichae epitomis adiecta Procli Chrestomathia grammatica, Lipsiae 1866
Wilson Scholia in Aristophanes edidit edendave curavit W.J.W. Koster, I: Prolegomena de comoedia. Scholia in Acharnenses, Equites, Nubes, 1B: Scholia in Aristophanis Acharnenses edidit N.G. Wilson, Groningen 1975
1
INTRODUCTION
3
(1)
The book, I : Testimonia
-1 P.Oxy. 2438 col. ii.35-9 (MP3 1351, LDAB 3724) = Ar. Byz. fr. 381 Slater k]�q�b�[\fRb kÓ �Ã��[g]t� �[Ï cgbq_Rf]� [Õcí �[b]fgdn- 35 Q]gP] W∞] �b�V^R b��� kbp[P]�n��TQ ��, [c[g]gk^- T]� ��, cRbnQTQ R�, cR[[]pWQ[Wb]TQ ��, [�cbQbj^- T]� k�, �SjT_^TQ R� �Q [„�b�] jRÚ [ , —- _]�TQ R�, Õ[c]g[Z\_nfTQ R�, p�[qQTQ R� quaecumque non notantur dispexit articulavit supplevit Lobel 1961a || 1sq. [Õcí �[b]fgdn£Q]gP] Lobel : [jRfÏ cgV£V]gˆ] nol. Natalucci 1995 p. 79 || 38 W�jT_bTQ pap. | WQ[ ¥]jR ¥[ ¥ ¥] ¥ ¥[ e.p. : �Q [„�b�] jRÚ []j¦Vbn fbQR vel �]fb(Q) Gallo 1968 p. 30, v. et 1969 : �Q [„�b�] jRÚ [cR[g^QbR DíAlessio 2000b : �QbjRfR_bSYQfTQ vel �QjRfR_bSYQfTQ c. (sic) vel �QjRfRVWb__nfTQ vel �QjRfRjWZT[b]_YQTQ vel §Qbj�Q kÓ fbQ�Q cgb\_. (sic) Natalucci 1995 pp. 84sq., ìil titolo di un libro che (Ö) raccogliesse al suo interno, i carmi singoli, o piª difficili a raggrupparsiî
-2 VA p. 3 Dr. (A) SYS[RdW kÓ �b�V^R §cfRjR^kWjR� —_QgP], cRb¬QR], kbpP[n_�TQ ��, c[g]gk^TQ ��, cR[pWQW^TQ ��� dY[WfRb kÓ jRÚ S�, √ �cbS[ndWfRb jWZT[b]_YQTQ �f�Q� cR[pWQW^TQ� Õcg[Z\_nfTQ ��, �Sji_bR, p[qQgP], �cbQ^jTQ k�. c[g]gk^TQ ex c[g]Tk^TQ correctum A sec. Dr. | cR[pWQ^TQ ® jRÚ S� Boeckh 1811-21 II/2 p. 555 : dY[WfRb kÓ jRÚ cR[pWQ^TQ �� jRÚ S� A : ìSufficiebat ]fbS_U post �� positaî Schneidewin 1837 p. 25 | �cbS[ndWb A : corr. Boeckh | jWZT[b]_YQTQ cR[pWQ^TQ A : jWZT[b]_YQ�gQ f��Q cR[pWQW^TQ Snell 19531 p. 191 : jWZT[b]_YQTQ �f�Q� cR[pWQW^TQ malim, cf. | ap. Lobel 1961a p. 6 adn. 1, | P. 3.139a p. II 81 Dr., | Theoc. 2.10b p. 271 Wendel1
T3 Suid. c 1617 p. IV 133 Adler ¯S[R°W kÓ �Q �b�V^gb] b�� }T[^kb kbRVYjfTb fRtfR� ∆VP_cbgQ^jR], HPpbgQ^jR], c[g]¦kbR, cR[pYQbR, �Qp[gQb]_g�], �RjZbjn, kRdQ\dg[bjn, cRb¬QR], Õcg[Zq_RfR, —_QgP], kbpP[n_�gP], sj¦VbR, �Sji_bR, p[qQgP], k[n_RfR f[RSbjÏ b��, �cbS[n__RfR �cbjÏ jRÚ jRfRVgSnk\Q, cR[RbQY]Wb] fg~] ²VV\]b, jRÚ ³VVR cVW~]fR. �Q �b�V^gb] A : �b�V^gb] GFM : �b�V^R V | fRtfR : fnkW V | HPpbgQ^jR], �´W_WgQ^jR], ∏]p_bgQ^jR]� KOster 1705 p. III 116 adn. 9 | ]jgVbÏ �¢� �Sji_bR Daub 1880 p. 423, obl. Hiller 1886 p. 358 | �cbjn (¶SgPQ cgb\fbjn sscr. V) : ¯c\ ·jk� Bergk 18784 p. I 367 n. 4 : ¯c\ ·jR� Irigoin 1952 p. 41 adn. 2 : cnQfR �cbjn Daub 1880 p. 423 | jRÚ jRfRVgSnk\Q : jR^ om. V
4
T4 Eust. Prooemium commentariorum Pindaricorum 34.1 Kambylis = p. III 303 Dr. cgbq_RfR kÓ cgVVÏ S[n°Rb H^QkR[g] VYSWfRb, „Q jRÚ dY[gQfRb gÃj ¿V^SR, gà _UQ fÏ cnQfR ÕcW_Q\_Rfb]_YQR� W∞]Ú kÓ cRb¬QW], kbp�[R_�gb, c[g]¦kbR, cR[pYQbR, fÏ jRÚ RÃfÏ _Q\_gQWPpYQfR c[Ù �[RZYTQ, Õcg[Zq_RfR, �Sji_bR, p[vQgb jRÚ �cbQ^jbgb, jRfÏ fUQ �]fg[^RQ ›]WÚ fWf[Rjb]Z^Vbgb, gœ] jRÚ �cbQ^jgP] fWf[R]PVVn�T] dR]^Q. jRfÏ fUQ �]fg[^RQ : jRfÏ fUQ ]fbZg_Wf[^RQ Birt 1882 p. 165 n. 1, obl. Kambylis p. 116* n. 145 : _WfÏ f�Q �]fg[b�Q Vitelli 1884 p. 162 n. 1
-5 P.Vindob. Gr. inv. 39966 verso col. i.5f., edd. Sijpesteijn and Worp 1974 Hb]�kn[gP [�cbQbj]^ ¼� k�, �S½i_bR, c[g]gk^¼[Q 5 Õcg]�¾\[_Rf 5 Hb]�kn[gP et c[g]gk^¼[Q suppl. e.p., cetera Puglia 1998
-6 | (H4) F5.2 = | Pae. 6.124 p. 304 R. �Q f�b �� [f�]Q c[g]gk^[T]Q dY[WfRb fr. 108 addito suppl. Rutherford et DíAlessio ap. Rutherford 1997a pp. sq.
-7 | (VE56MLh) Ar. Eq. 1264b p. 258 Jones-Wilson f^ jnVVbgQ �[Zg_YQgb]b� fgtfg �[ZU c[g]gk^gP HbQkn[gP. ¯ZWb kÓ g—fT]� [F1] lemma om. 6Lh | c[g]gk^gP HbQkn[gP �[ZU M | c[g]T(b)k^gP VE5M || cf. P20 (Suid. c 2757 Adler)
-8 | (h) Il. cit. ad 23.361a1 p. V 426 Erbse (GeM1P11U4) H^QkR[g] kÓ }T[bjifW[gQ kbÏ fv] Rb¿ kbdp¦SSgP �Q c[g]Tbk^gb]� [F13] c[g]Tbk^Rb] Ge || similia praebet E.M. col. 1652 Gaisford, operis indice omisso || v. ad F13
-9 Porph. Abst. 3.16.5 H^QkR[g] kÓ �Q c[g]gk^gb] cnQfR] fgˆ] pWgˆ] �cg^\]WQ, ¡c¦fW ÕcÙ ÁPd�Qg] �kbijgQfg, gÃj �Qp[icgb] ¡_gbTpYQfR] �VVÏ fg~] ³VVgb] �ibgb]. [= F14] ¡c¦fW : ≈fW dub. Nauck 1860 p. XXXII, fort. recte | ³VVgb] : �V¦Sgb] Wesseling 1745 p. 97
5
Four of the five extant lists of Pindarís works (T1-T4) mention the Prosodia; only the list
sketched in the late Life in hexameters (p. I 9 Dr.) omits them, as it does several other
books.1 The lists in the so-called Vita Ambrosiana (T2)2 and the biography of Pindar
preserved by P. Oxy. 2438 (T1)3 are generally regarded as the most reliable witnesses to
the Alexandrian edition of Pindarís poetry, traditionally ascribed to Aristophanes of
Byzantium.4 As far as the remains of T1 go, the two lists coincide almost exactly in the
titles and numbers of the books: the sole exception are the Hyporchemes, which are given
two books in the Vita Ambrosiana and only one (probably wrongly) in the Oxyrhynchus
Life.5 The title Prosodia cannot actually be recognised in the preserved section of T1, but
1 On the Vita metrica (pp. I 8f. Dr.) see chiefly Ludwich 1879 (dating it to the fifth century AD) and Gallo 1997: 44f. n. 10 (allowing for an earlier date in the Imperial age, possibly the third century AD). On the lists of Pindarís works in the biographical tradition see Gallo 1968: 27-45, with bibliography. 2 On the VA see Leo 1901: 28, Lefkowitz 20122: 61-6. The text is impossible to date, but a terminus ante quem is Arethas of Caesarea, who read and used it (Kougeas 1913: 105f. n. 7, cited by Gallo 1968: 19 n. 11) in the tenth century AD; von Leutsch 1856: 14f. argues that it goes back to Didymus or shortly after his time. 3 On P.Oxy. 2438 see chiefly Lobel 1961a, Arrighetti 1967, Gallo 1968, and Natalucci 1995, the last three of which devote ample space to its relationship with the VA and the other ancient biographies of Pindar. Arrighettiís suggestion that its authorship is to be traced to Didymus or his circle (144f., endorsed by Natalucci 1995: 87f.) is hard to accept: Didymus must have known the dates of the Pythians (presumably through the Aristotelian HPpbgQ~jRb ), which our biographer does not, as is apparent from the fact that the earliest and latest dates he gives for Pindarís career are a dithyramb performed in 497/6 and O. 4 of 452 rather than P. 10 of 498 and P. 8 of 446 (see already Lobel 1961a: 5). So, the composition of the Oxyrhynchus Life is probably earlier than the time when the association between the Pythian victor-lists and Pindarís Pythians gained common currency, although necessarily not earlier than Aristophanes of Byzantium (see the next note), to whose time it is dated by Gallo 2005: 34; the terminus post quem shifts forward if Aristarchus was cited at l. 21 (a possibility envisaged by Natalucci 1995: 75; Gallo 1968: 70 favours a reference to Aristophanes, Natalucci herself to Dicaearchus). The papyrus itself is dated on palaeographical grounds to the late second or the third century AD (Lobel 1961a: 1); Lefkowitzís assumption that this is also the date of the text (20122: 61) is unwarranted and, as we have seen with reference to Didymus, unlikely. 4 On Aristophanes of Byzantium as the editor of the ëstandardí Pindar see below, p. 53 and n. 37. In the chapters that follow I shall use the phrases ëAristophanes of Byzantiumí, ëthe Alexandrian editioní, and the like interchangeably: as long as the notion of one authoritative edition stands, the editorís personal identity ñ like the notion that ëheí was an individual scholar rather than several ñ ultimately matters little. 5 The question of the Hyporchemes is thorny. On the one hand, a single book as indicated by T1 leaves the grand total one book short from the necessary seventeen, which T2ís two books fill perfectly, thus suggesting a scribal error in T1 (so Gallo 1968: 77). an the other hand, no ëfirstí or ësecondí book of the Hyporchemes is cited by a surviving ancient author (no conclusive argument in itself, see Irigoin 1952: 37f.), and, more importantly, the phrasing of | O. 13.25c p. I 361 Dr. ¡ H^QkR[g] kÓ �Q _ÓQ fg~] Õcg[Zq_R]bQ �Q ´nXTb d\]Ú c[�fgQ WÕ[WpvQRb kbp�[R_�gQ, �Q kÓ f�b c[ifTb f�Q kbpP[n_�TQ �Q rq�Rb] seems to imply that, unlike the Dithyrambs, they were collected in one book. (The solution proposed by Lehnus 1973a: 398 that ìthe first of the Dithyrambsî is the first ode, not the first book, is scarcely paralleled and not altogether
6
unless we are to think that they were absent from the list altogether while the lacuna at
lines 36f. obliterated another, otherwise unattested two-volume work of Pindarís, to supply
c[g]gk^£T]� is the solution that imposes itself. The two Livesí total of seventeen books is
confirmed by the entry on Pindar in the Suda (T3) and by the so-called Vita Vaticana (p. I
6 Dr.), which does not list any titles beside the Epinicians and thus lies outside our
discussion.6 ìThe alarming wonkinesses of the Suda book-list are safest left unconfrontedî
(Nick Lowe);7 the Prosodia are there nonetheless, with pride of place after the first two
books of the Epinicians. The Suda omits the number of volumes that constitute each book;
so does also the list that Eustathius included in the preface to his now lost Pindaric
commentary (T4) following closely the Vita Ambrosiana (bar the Hymns).8 However, T1,
T2, and by implication T5 and T6 make us certain that the Prosodia took up two books.
persuasive: see already Turyn 1948: 290.) Nonetheless, a slight stylistic infelicitousness on the part of the scholiast to O. 13 is more easily conjectured than the simultaneous disappearance of a book from the papyrus Life and erroneous duplication of the Hyporchemes in the VA; that the missing seventeenth book in P.Oxy. 2438 was to be found in the lacuna at l. 38 (as supposed by Lobel 1961a: 6, Bowra 1964: 159, Arrighetti 1967: 138f.) is unlikely given the surviving traces, which Gallo and DíAlessioís supplements suit better. The different order in which T1 and T2 list the books is no evidence for substantial divergence, as the order was probably never fixed (Race 1987). Natalucci 1995: 81f. notes the risk involved in supplementing T1 from T2 but acknowledges their overall consistency and the difficulty of finding other supplements for T1. 6 I adopt Galloís title Vita Vaticana (1977: 50) in order to avoid the misleading chronological implications of the traditional label Vita Thomana, since a version of the text was used by Eustathius (twelfth century AD) and thus must predate Thomas Magister (early fourteenth). 7 2007: 170 n. 14. At least part of the wonkiness must be due to the compiler reading in one or more sources that the books had to total seventeen (a figure he even repeats as the number of the ìtragic playsî) and making a gallant attempt to inflate the defective record he had at hand with whatever other material he could either find in other sources (kRdQ\dg[bjn, ]j¦VbR) or simply make up: Hiller 1886: 366. On the k[n_RfR f[RSbjn see Hiller 362f., Immisch 1889: 553-8. The articulation of the final part of the list is uncertain; I present my own interpretation, but with no claim to definitiveness. Its chief advantage is to preserve the binary pair between verse and prose, of which the Suda is quite fond: cp. R 112, b 467, 706, _ 198, c 893 Adler. (This is not decisive in itself, as the Suda is also fond of listing a poetís prose works at the end of the relevant entry.) This is also how the scribe of V must have understood the text, given his gloss ¶SgPQ cgb\fbjn above �cbjn. For the explicit acknowledgement that epigrams need not be in verse see W 2270 Adler; this is supported by the specification �cbjn in T3, which would be pointless if the assumption were that they must. On T3 see also Irigoin 1952: 36f., Gallo 1968: 31-5, 42-5. 8 On T4 see chiefly Negri 1998, 2000: 265-73. As she saw, the omission of the Hymns is probably due to Eustathiusí misunderstanding of T2ís —_QgPc as a general term rather than a specific title (1998: 224f.). Concerning the lack of book-numbers, Natalucci 1995: 80f. points out that the phrasing of T3 introduces a list of genres, not of books; the same explanation also applies to T4 (Negri 1998: 226). On Eustathiusí
7
T5 is part of a book-list, presumably the catalogue of a library (or two).9 After the
title of each work, the individual volumes held are listed, but the numeral for the Prosodia
is lost in lacuna, and we do not know whether one or both books appeared in the
collection.10 T6 is the marginal note whose discovery triggered Giambattista DíAlessioís
reassessment of Pindarís papyrus fragments and his consequent recognition of the papyri
of the Prosodia (see Chapter 2). The remaining few testimonies (T7-T9), none of which
specifies a book-number, are the sources that transmit the meagre three fragments that the
indirect tradition explicitly attributes to the Prosodia.
Pindaric commentary and its preface see Kambylis 1991, Negri 2000; it was probably composed in the later period of Eustathiosí diaconate at Hagia Sophia, between ca. 1160 and 1174 (Kambylis 1991: 6-9). 9 First published by Sijpesteijn and Worp 1974: 324-31, re-edited by Puglia 1998, Otranto 2000: 9-15, SB cit. No image of the relevant section seems to have been published to date, and I am consequently unable to verify any readings personally. The text, written on the back of two documents dated to 10 and 12 AD, will have been written sometime in the mid-first century (Sijpesteijn and Worp 1974: 325). The ms. probably comes from the Arsinoite nome (Puglia 1998: 78f.). Coll. i and ii represent two different lists, with only some overlap; for speculation on this see Puglia 1998: 83f. 10 Pace Puglia 1998: 80 and Otranto 2000: 11, the fact that the title is in the genitive, implying that one or more numerals followed, does not entail that only one book was listed. The papyrus gives no hint that a multi-volume work held in its entirety would have appeared in the nominative, as one-volume works do: the Encomia (mentioned by Puglia as a counter-example) are so cited precisely because only one book of these existed (see T1, T2). Judging from what survives outside the lacunae, the Iliad and the Odyssey, whose books are listed individually at col. ii.9-11, may well have been held complete (so Puglia 1998: 83, however dubitatively).
9
(2)
The book, II : Contents, structure, scholarship
INDIRECT TRADITION
The Prosodia have enjoyed mixed fortunes in their transmission. As we have just seen
from the testimonia, later authors cite them sparingly: out of their two books, only two
fragments are quoted explicitly, both in scholarly texts (T7, T8 = F1, F13); Porphyry
summarises another episode without reporting Pindarís actual words (T9 = F14). If H34
belongs to the Prosodia, we can add three unattributed quotations to our tally, the two of
*F4.1f. in the Iliad scholia and Artemidorus of Daldi and that of *F4.6f. in the hypomnema
partly preserved by H37.
Several other fragments have been conjecturally attributed to the Prosodia by
modern scholars. The first self-standing collection of Pindarís fragments, edited by Johann
Gottlob Schneider in the late eighteenth century, printed seven pieces under the heading
H[gsikbR.1 His classification as such of frr. 33c-d Sn.-M., which he identified with the
c[g]gkbRjÙ] cRbnQ for the Ceans mentioned by the scholia to Isthmian 1,2 persisted until
the mid-twentieth century, when Bruno Snell noted their metrical coincidence with a
1 Schneider 1776: 29-32 (on the form with omega see pp. 63f., 82f. below). The earlier collections of Pindaric fragments by Stephanus (15601: II 344-77, 15662: II 344-77, 15863: I 388-410, 16004: II 344-77, 457) and Schmidt (1616: 157-68) are ordered by source and make no mention of books or genres. 2 Schneider 1776: 29 with | I. 1 pp. III 196f. Dr.
10
fragment of the first Hymn and re-attributed them accordingly.3 Beside these and the
explicitly attributed F1 and F14, he included Pausaniasí notices of the Aeginetan Ãb]_R to
Aphaia (2.30.3 = fr. *89b Sn.-M.), which is still a Prosodion in the latest Teubner Pindar,
and of the ìproem to Sacadasî (9.30.2 = fr. 269 Sn.-M.).4 He also classed as a Prosodion
what we now know as Paean 6.117f., although the scholia to Nemean 7 attribute it to ìthe
Delphic paeanî.5
August Boeckh remedied his predecessorsí oversight in this last case,6 but
inaugurated a successful prosodiac career for another part of the Paean, namely the
opening, which Aelius Aristides cites with no indication of its provenance; this
classification was disputed with little success by Johann Adam Hartung, and only
abandoned after the publication of H4 in 1908.7 Boeckh also attributed F1 to the song to
Aphaia, on the dubious basis of that deityís identification with Artemis, a view which was
to persist until Otto Schroederís first editio minor, although already questioned by K.
Kleanthis two decades earlier.8 Theodor Bergk was able to add to the tally F13, already
known as a Pindaric fragment thanks to the Etymologicum Magnum but attributed to the
Prosodia only by a scholion to the Iliad (T8) first published a few years earlier,9 and frr.
3 Snell 1946, reprinted with corrections up to 19754: 82-94 (the attribution of fr. 29 Sn.-M. to the first Hymn is guaranteed by | Lucian. Dem.Enc. 19 p. 225 Rabe1). Ascription of frr. 33c-d to a Hymn had been proposed already by Wilamowitz 1922: 328f.; Thiersch 1820: II 242f. n. 7 (who groups Hymns, Paeans and Prosodia together) calls it a ìDelian paeanî; the identification is doubted by Turyn 1948: 341 and Mingarelli 2003. 4 Schneider 1776: 32. 5 Schneider 1776: 31, contra | N. 7.94a p. III 129 Dr. 6 Boeckh 1811-21: II/2 568. 7 Boeckh 1811-21: II/2 589, Hartung 1855-6: IV 170 (see also Thiersch 1820: II 244 n. 10); Aristid. Or. 28.58 Keil. 8 Boeckh 1811-21: II/2 588, Schroeder 19081: 300, Kleanthis 1886-7: IV 232. 9 Bergk 18532: 250 with An. Par. III 292 (published 1841).
11
*92 and *93 Sn.-M. (cited by Strabo 13.4.6), which he assigned to the same composition as
F14 on account of their similarity of theme.10
The Boeckh/Bergk canon remained largely unchanged through Schroederís, Snellís,
and Maehlerís successive Teubner editions, maiores and minores alike, except the
necessary re-classification of Paean 6.1-6 and frr. 33c-d Sn.-M. Thus, of the six fragments
printed under the heading Prosodia in the latest edition of 1989, half are of conjectural
attribution: frr. *89b, *92, and *93 Sn.-M. For each of these a case can be made, although
it falls short of definite proof. First and foremost, an ode for Aphaia, a distinctly local
figure whose legend straddles the heroic as well as the divine sphere, may well have been
classed as a Prosodion alongside compositions for Aeginaís other hallmark character
Aeacus (F3, F5) and for other heroes such as Heracles and the sons of Electryon (F8, *F7),
rather than among the Hymns;11 moreover, Pausaniasí citation suits quite nicely the format
taken by headings of Prosodia (see below). Nonetheless, as Wilamowitz had pointed out,
there is no conclusive proof to warrant a choice between these, or even the Partheneia (or
the shadier Kekhorismena).12 As for the two fragments from Strabo, they may indeed come
from the narrative sketched by Porphyry, but nothing precludes their being quoted from a
different narrative of the same episode, or even from a mere passing reference to Typho or
Aetna: Straboís text, where they follow a quotation of Pythian 1.17-19, does not even
10 Bergk 18532: 249f. 11 On F3, F5, and *F7 see respectively pp. 206-9, 253-5, 308-12 below. On the high proportion of poems dedicated to heroic rather than divine figures in Pindarís Prosodia see also pp. 107-11. 12 Wilamowitz 1922: 274, see also DíAlessio 1997: 39. However, if Pausaniasí W∞] ÄQ (sc. �dR^RQ) jRÚ H^QkR[g] Ãb]_R o∞SbQqfRb] �cg^\]W implies a heading o∞SbQqfRb] W∞] �dR^RQ (but Pausaniasí citations should be handled with much care), one notes that it is uncertain whether headings of Partheneia ñ none of which survives ñ mentioned the deity addressed, and whether those of Hymns always mentioned the commissioning community (see n. 29 below). This may give a slight edge to classification as a Prosodion, but on no better grounds than lack of evidence for the other books.
12
guarantee that they both come from one poem.13 As Giambattista DíAlessio has argued, all
three unattributed fragments ought to be relegated to the fragmenta incertorum librorum.14
Accordingly, they are not considered in this dissertation.
DIRECT TRADITION
Fortunately, DíAlessio compensated for his fifty percent cut in the indirect tradition with a
reassessment of the papyri until then ascribed to the Paeans: a move which provided the
Prosodia with four manuscripts of the Roman era, essential information drawn from two
more, and the vast majority of the almost two hundred fragments that can now confidently
be assigned to the book.
The manuscript known as H7 had been conjecturally assigned to the Paeans by its
first editor on the basis of the Apolline theme of its largest fragment, our F11.15 The later
discovery of a partial overlap with H4, the main witness of the Paeans, in the text of F5
(then unanimously known as the third triad of Paean 6) seemed to confirm the attribution,16
although the different colometry of the corresponding verses between the two papyri
caused some puzzlement. Thus H8 and H29 were also classed as manuscripts of the Paeans,
however hesitatingly, because of their textual overlaps with H7 itself.17 Further overlap
13 DíAlessio 1997: 40; doubt was already expressed by Puech 19312-522: IV 160. 14 DíAlessio 1997: 39f. 15 Hunt 1922b: 87. 16 H7 frr. 16, 15: see respectively Snell 1938: 431 (after Hunt 1922b: 97), Lobel 1961b: 17. 17 See pp. 128, 135f. below. Lobel 1961c: 29 had noted the oddity of a supposed paean addressed to Aeacus, and suggested the attribution of H29 to the Prosodia or, alternatively, to the Hymns. Nonetheless, the overlap with H7 carried greater conviction, and the fragments were regarded as Paeans in most subsequent editions and studies before 1997 (Snell 1959-643: II 52-7, Bona 1988: XXIX-XXXII (although dubitatively), Sn.-M. pp. 57-9, KÅppel 1992: 346 and passim), with the praiseworthy exceptions of Ferrari 1991b: 385 (H8) and Rutherford 1992 (H29); the attribution of H7 itself to the Paeans was doubted by Wilamowitz 1922: 518 and more recently Rutherford 1995: 45 n. 7.
13
with both H4 and H7, as well as with fragments of the Hymns known from indirect
tradition, caused the somewhat incautious inclusion in the Paeans of most of H26 ñ namely,
of all of its fragments that could not be securely attributed to other books.18
The starting point for DíAlessioís reclassification was his and Ian Rutherfordís
discovery on H4 col. xxx of a title and a marginal note next to the opening of F5: the title
labelled the text as an independent ìprosodion for the Aeginetans to Aeacusî, and the
scholion stated �Q f�b �Æ [f]�ÇQ c[g]gk^[T]Q dY[WfRb ì(it) is transmitted in the first book
of the Prosodiaî.19 On this basis, DíAlessio could argue compellingly that the different
colometry that the fragments of F5 display in H4 and H7 ñ a unique occurrence in our
manuscripts of Pindar ñ is to be explained by interpreting the two papyri as witnesses to
the two books in which F5 could be found, the Paeans and the Prosodia; thus, since H4 is
securely attributed to the former, H7 must belong to the latter.20 This turned an interesting
but severely mutilated manuscript of the Paeans into the primary witness of the Prosodia.
Although too little remains of its text to offer us an overview of the book comparable to
that which H4 gives the Paeans, H7 enables us to identify as Prosodia the fragments of H8
and H29, and among those of H26 ñ which, accordingly, must have been a set of volumes of
Pindar in a matching format comprising at least Hymns, Paeans, and Prosodia ñ those with
whose text it overlaps.
This means substantial fragments of one Prosodion from H7 (F11), at least one from
H8 (F12), two from H26 (F8, F9), two from H29 (F2, F3), and of course F5 from H4 and H5;
DíAlessio makes a good case for classing as Prosodia two more fragments of H26, *F6 and
18 See DíAlessio 1997: 24. For details of the overlaps see p. 130 n. 36 below. 19 Title: Rutherford 1997: 3-6. Scholion (| F5.2 = | Pae. 6.124 p. 304 R.): DíAlessio and Rutherford in Rutherford 1997: 6-8. 20 DíAlessio 1997: 25-7, 37f.
14
*F7.21 All other fragments of H26, except those which overlap with text safely attributed to
other books, should be regarded as incertorum librorum; statistically, several of them must
come from the Prosodia, but, as it is impossible to determine which, none is included in
this dissertation.
One fragment of H26 which probably belonged to the Prosodia offers a clue for a
further identification, however tentative. As noted already by DíAlessio, the last two lines
of H26 fr. 86 can plausibly be identified with the title and opening line of F5 as edited in the
Prosodia (not in the Paeans, since the preceding lines certainly do not match those
preceding F5 on H4).22 As I have argued elsewhere, a case can tentatively be made for
recognising metrical compatibility between fr. 86.1-5 of H26 and fr. 1.3-7 of H34 (*F4(a)),
which would therefore become the fifth manuscript of the Prosodia identified to date.23
The evidence is admittedly thin and the argument speculative, and the attribution cannot be
regarded as certain. Nonetheless, I prefer to appear to err on the side of boldness rather
than to leave an interesting possibility unexplored: thus the fragments of H34 are included,
with the appropriate asterisks, in this edition. The papyri that contribute to the Prosodia
(including the two papyri of the Paeans that contribute to F5) are examined in Chapter 7.
It is possible that further fragments of the Prosodia hide among other Pindaric papyri
now regarded as incerti generis, as indeed among the Pindaric incerta more generally. I
have made no attempt to comb through them and speculate which ones might or might not
have come from a Prosodion; fragments for whose attribution to the Prosodia no case can
be made on arguably solid textual or papyrological grounds have been systematically
disregarded. However, for the sake of completeness, one should at least mention the
21 DíAlessio 1997: 41f. 22 DíAlessio 1997: 37 n. 92, see also Prodi 2013: 54. 23 Prodi 2013: 53-6.
15
possibility that H11 (preserving frr. 140a-b Sn.-M.) may come from a roll of the Prosodia:
the textual overlap between col. i.12-15 and H26 fr. 97 restricts the choice to the Paeans,
the Prosodia, or the Hymns, and the simultaneously Apolline and heroic theme of fr. 140a
is compatible with classification among the Prosodia, although the probable evocation of
Xenocritus of Locri ñ a famed composer of paeans ñ in fr. 140b arguably shifts the balance
towards the Paeans.24
Since F5 is explicitly attributed to the first book of the Prosodia, the question arises
whether all of H7 ñ and, consequently, all three of H8, H29, and the Prosodia-fragments of
H26 ñ are to be attributed to that book alone. This is quite certainly the most natural
conclusion in the absence of evidence to the contrary. It cannot be excluded, however, that
H7 was not one but two matching rolls, book one and book two, produced and sold
together: compare, among many, the very case of H26.25 It may indeed seem suspicious that
we have four (or five) papyri of book one and none of book two, but this must be balanced
by the fact that, since no quotations are explicitly attributed to book two, no papyrus
fragment of it can be identified. A possible, although not a conclusive, argument against
ascribing all the fragments to book one arises from circumstances that will be sketched
below. A similar point can be made, perhaps rather more plausibly, with reference to H26:
if one commissions a set of matching rolls of Pindar and includes the first book of the
Prosodia as well as several others, it would seem strange not to include the second book as
well. But on the existing evidence it is not possible to offer more than an educated guess.
24 It should be noted, however, that if the marginal note to the left of the beginning of fr. 140b in col. ii is indeed a title that mentions Apollo, as argued by Rutherford 2001: 382f. after Grenfell and Hunt 1903: 17, the pattern is not that found in the Paeans (where the reference is to a community and a sanctuary), but that found in the Hymns and Prosodia (see the next section): if Rutherfordís argument is correct, then the Prosodia become a very strong candidate, perhaps the strongest, for attribution. Wilamowitz 1922: 500 suggested that the fragment belonged to the Hyporchemes, a book about whose titles very little is known, but no fragments of the Hyporchemes have been identified among the remains of H26. 25 Lobel 1961d: 31; see pp. 130f. below.
16
TITLES
As far as the papyri go, the titles attributed to individual Prosodia in the Alexandrian
edition of Pindar indicated the community that was taken to have commissioned and
subsequently performed the ode, and the deity or hero to whom it was addressed. F3 is
titled o[∞]SbQqfRb] £ W∞[]] o∞Rj¦Q on H29 fr. 1; the same title marks the beginning of F5 on
H4 col. xxx, capped by the generic indication È[g][¦]Éb[g]Q as (presumably) necessitated
by the fact that the book in which it was included was the Paeans.26 The title of *F7 on H26
fr. 27 is damaged, but it certainly began with �][SW^gb], and the traces that follow are
probably a reference to the ìsons of Electryonî again introduced by W∞].27 The connexion
of H29 fr. 3 to the rest of the manuscript, and thus to the Prosodia, is dubious, but if it does
belong to the book, its title fg~] Rãfg~] confirms this practice.28
The indication of the commissioning community is also applied regularly to Paeans
(Pae. 6, 7, 7b, 8, and also 19 ñ perhaps from the same poem as fr.140a Sn.-M. ñ if a Paean;
Simon. PMG 519 fr. 35(b) = 101 Poltera) and perhaps less regularly to Dithyrambs (Dith.
2; B. 19, 20, but not 15, 17, 18) and possibly Hymns.29 The specification of the divine or
heroic addressee introduced by W∞] is paralleled in the Hymns (fr. 37 Sn.-M. in VA p. 2 Dr.
and Paus. 9.23.3f., B. fr. 1B M.), as indeed regularly in hymns from the Homeric Hymns
onward. In the Paeans it is the supposed place of the performance that is introduced by
26 See pp. 244f. below. 27 See pp. 308-10 below. 28 See p. 135 below. The presence of only the indication of the commissioning community need not perplex: when the shorthand is used, it stands for the whole of the preceding title, any difference from which, or addition to which, is explicitly specified. 29 The only piece of evidence for the Hymns is Paus. 9.16.1, according to whom Pindar �cYcW_°W Ö Êb��\] �] �__TQ^gP] his hymn to Ammon. It is not possible to exclude that Pausanias extracted this information not from an explicit statement in the title but from the fact that the hymn was addressed to Ammon, although there is no particular evidence to support this hypothesis. There appears to be no specification of the commissioning community in B. fr. 1B M. on H.
17
W∞]: Pytho in Pae. 6, Delos in 7b, uncertain in 7 and 19 (if a Paean); again Pytho in Simon.
PMG 519 fr. 35(b) = fr. 101 Poltera. Pausanias thrice applies the same pattern to two
compositions he calls c[g]¦kbgQ (Eumelusí at 4.33.2, 5.19.10, Pronomusí at 9.12.5), both
of which are �] }vVgQ. This raises the as yet unanswerable question whether in Pindar too
certain Prosodia for the cult of Apollo ñ especially if his association with the genre was
already current in criticism by the late third century ñ may have been titled after the
sanctuary rather than the god.30
INTERNAL ORGANIZATION
Little can be gleaned of the criteria that governed the division between the two books and
the order of the poems within each.31 T6 testifies that F5 appeared in book 1, while none of
the later citations specifies a book-number; if my earlier argument on H34 as a papyrus of
the Prosodia is correct, *F4 also necessarily belonged to book 1. The papyri guarantee the
sequences F2-F3 (H29 fr. 1, H7 fr. 8), *F6(b)-*F7 (H26 fr. 7, if Prosodia),32 F8-F9 (H26 fr.
32), and probably *F4(d)-F5 (H26 fr. 86).33 Before T6 became known, Rutherford had
noted the sequence F2-F3, two odes apparently concerned with heroes, and suggested that
the first book ìconsisted of poems devoted to the gods, and the second of poems devoted to
heroesî.34 The subsequent discovery that a poem to Aeacus was included in book 1
disproved this hypothesis, although leaving no obvious substitute for it; the sequence F8-
30 This supposition could be shown to be false, or at least not always true, if H11 were a papyrus of the Prosodia and the title (if one) of fr. 140b in col. ii truly contained a reference to Apollo, see n. 24 above. 31 This section is largely based on my earlier treatment of the subject in Prodi 2013: 56-8. 32 See n. 26 above. 33 DíAlessio 1997: 37 n. 92, Prodi 2013: 54. 34 Rutherford 1992: 68; ì[a]lternatively, the Prosodia to heroes might have occupied the latter part of the second bookî.
18
F9 more generally precludes any supposition that odes for gods and odes for heroes were
to be found in separate books.35 The criterion for division (if indeed there was one) remains
unknown.
On the arrangement of the poems within the books, DíAlessio noted the sequence
F8-F9, where a poem probably for Heracles is followed by one quite certainly for Hera,
and suggested an alphabetical ordering by dedicatee, perhaps subordinated to one by the
commissioning community, as the order of the respective constituents in the titles would
suggest.36 The placement of F5 in book 1 suits both possibilities very nicely. Something to
note is that an alphabetic arrangement by commissioning community, if envisaged as
running continuously across the two books of the Prosodia, with the first half of the
sequence in book 1 and the rest in book 2, strongly suggests that H7 contains fragments of
both books, not only of the first: F11 (H7 fr. 1), which was composed with all likelihood for
the citizens of Naxos, is hardly an attractive candidate for inclusion in book 1, given the
position of nu in the alphabet and the relative prominence of cities such as Aegina and
Thebes in Pindarís work.37
35 It is true that F5 explicitly connects itself with the worship of Apollo (see pp. 252f. below), which could allow Rutherfordís view to stand with a slight modification: a book 1 of odes dedicated to Apollo and/or Artemis (F1, *F4-F5, F11, probably F15) and a book 2 of odes dedicated to other deities and to heroes (F2-F3, *F6-*F7, F8-F9, perhaps F12), which would also fit in nicely with the prominence given by Pollux to Apollo and Artemis as dedicatees of prosodia (P30, see pp. 90f. below). However, the title of F5 ñ assuming that it was the same in the Prosodia as it was in the Paeans ñ relates the poem to Aeacus alone: if god vs. hero had been a meaningful opposition in shaping the internal organization of the Prosodia, it would be excessively strange for a poem to have been placed in the opposite category from the one stated by its title (which was probably assigned to it by the same editor). Furthermore, the sequence *F6-*F7 is likely to provide a further example of a poem to Apollo (see p. 302 below) followed by one to a group of heroes. 36 DíAlessio 2004: 114. 37 By way of comparison, if one ordered the Epinicians alphabetically by homeland of the laudandus, a hypothetical Naxian victor would have to be added in thirty-seventh place on a list of forty-seven. This total includes the Western Greek cities which so far have failed to turn up in Pindarís cult poetry, but their presence does not dramatically alter the count. On F11 as a Naxian ode see Snell 19643 p. 49, Rutherford 2001: 365, Furley and Bremer 2001: I 157f. (although surely it is the sacrifices, not Apollo as Furley and Bremer claim, that come from Naxos). However, it is also possible, as Bruno Currie suggests to me, that not
19
An alphabetical arrangement, by first letter only as was the rule throughout the
Hellenistic age, is attested for Bacchylidesí Dithyrambs by title (15 �Qf\Qg[^kRb, 16 lost,
17 ÌÍpWgb, 18 r\]W�], 19 ∏i, 20 ºkR]) and has been argued with some plausibility by
Edgar Lobel for Sappho and Carlo Gallavotti for Alcaeus by first word of each poem.38 So
far, there is no evidence for an alphabetical arrangement in Pindar: the Epinicians are
organised according to the prestige of the competition in which the victory was achieved
(with room for some adjustments),39 and while the criteria that shaped the sequence of the
Paeans are unknown as yet, it is only a hypothesis that alphabetical order may have
applied to a section of the book.40 Nonetheless, the example of Bacchylides ñ whose
Epinicians, unlike the Dithyrambs, were not alphabetically arranged ñ prevents one from
all poems had an indication of the commissioning community in the title, and that the poems which did not were placed in book 2, shifting the others toward the beginning of the sequence as a result. 38 Lobel 1925: XVf. with Liberman 2007: 46-50, Gallavotti 1942: 165f. with Pardini 1991: 280 n.1. An alphabetical arrangement may have been at play also in (some books of) Simonides: DíAlessio 1997: 53 n. 175 suggests that if jYV\fb in the title of PMG 511 = fr. 7 Poltera on P.Oxy. 2431 fr. 1(a) jYV\fb £ fg~] o∞Rf^gP cRb]^Q is a section-heading within the book of equestrian victory odes (on which see Lobel 1959b: 89) then the following line ñ that is, the poem-heading proper ñ would suit quite nicely the first item in an alphabetic series (but note the similar heading of PMG 519 fr. 120(b) = 9 Poltera on P.Oxy. 2430 fr. 120(b), which seems to have been jYV\]�� �p\QR^Tb Ï[ for a Spartan victor called Athenaeus, see Lobel 1959a: 81, Poltera 292); Poltera 169f. makes a similar case for PMG 519 fr. 35(b) = 100-101 Poltera on P.Oxy. 2430 fr. 35(b) (Paeans), where a poem titled �Qk[^gb] £ W∞] HPpi follows one probably composed for the Athenians, (Lobel 1959a: 56, see also Rutherford 1990: 172-6). On arrangement by first letter only see e.g. the list of Euripidesí tragedies on P.Oxy. 2456 and of various poetsí comedies on P.Oxy. 2659, the collected Euripidean hypotheseis on P.Oxy. 2455 (∆[Y]f\] a∞k^cgP] [col. iv], Á beginning with Áq_WQg] [viii] and ending with ÁYQQ\] [xiii], Ð beginning with Ð[^Xg] R� Ðg~QbX [xvii-xviii] and ending with ÐbVgjfqf\] Ð[^Xg] �� Ðg^Qb]]Rb [xix-xxi]), P.Oxy. 2457 (êVj\]fb] oÒgVg]), and PSI 1286 (Óv]g] ÓRkR_nQpP]), and indeed Bacchylidesí Dithyrambs on A, where ∏i precedes ºkR]; see Daly 1967: 18-26 on the earliest evidence for the phenomenon. 39 See Irigoin 1952: 43f., Negri 2004: 27-34. 40 The relevant sequence is Pae. 2-7, which may have been ordered alphabetically by place of performance: Rutherford 1995: 49 n. 24, DíAlessio 1997: 31 n. 45, 2001: 84. However, there remains the oddity of an alphabetical arrangement by second, not first, word of the title; furthermore, Rutherford assumes an order by notional place of performance, different on occasion from the name actually present in the title (k for Delphi, not c for Pytho, for Pae. 6) while DíAlessio avoids this problem but leaves a suspiciously wide gap between k for Delos (quite certainly the place of performance of Pae. 5) and c for Pytho (Pae. 6). (That F5, titled o∞SbQqfRb] W∞] o∞Rj¦Q, follows Pae. 6(a) is no conclusive evidence against an alphabetical arrangement, since its placement will have been due to its metrical identity with the preceding Pae. 6(a), and thus merely interrupted the presumed sequence much in the same way as P. 3 interrupts the sequence of odes for chariot victories P. 1-7 without damage to the overarching principle.)
20
concluding that Pindarís Prosodia could not have been ordered alphabetically if one or
more of his other books were not.
Obviously, if my suggestion that H34 is a papyrus of the Prosodia and *F4
immediately preceded F5 (teste H26 fr. 86) is correct,41 DíAlessioís hypotheses become
unsustainable: *F4 must have been composed for a Delphian chorus to perform, as he was
the first to observe.42 Nonetheless, even if my suggestion is rejected, his hypotheses face a
few problems, at least one of which is of considerable import.
Firstly, a series of items ordered alphabetically by a word other than the first, as
required by DíAlessioís first conjecture of an alphabetic ordering by dedicatee, would be
quite surprising; and on the hypothesis of an overarching order by commissioning
community, since the performers of both F8 and F9 are not known with certainty, the
apparent alphabetical sequence by dedicatee could be accidental ñ we could have, say, a
poem for the Thebans to Heracles followed by one for the Argives to Hera; this would
undermine the inference of an alphabetical order in the first place.
More importantly, an alphabetical order by commissioning community is difficult to
square with DíAlessioís identification of H26 fr. 86.6f. with the title and first line of F5
even if H34 plays no part in the picture. The title of F5 is identical to that of F3: therefore,
an alphabetical arrangement by community would lead us to expect to find both in one
sequence of poems o∞SbQqfRb] W∞] o∞Rj¦Q, whether one after the other or separated by
one or more other compositions with the same title. (Surely the fact that the alphabetical
order would have been by first letter only cannot have meant that poems with identical
41 See p. 14 above. 42 DíAlessio 1991: 115; see also Ferrari 1992c: 228f., Rutherford 2001: 391. Earlier, Snell had suggested that his fr. 215 was a Theban ode, presumably on the strength of his reading SnÔgQ Ö [S¦Qg] at (a).4, which he seems to have understood as a reference to the Sparti.
21
titles could be split.) Crucially, however, just as the title of F3 was written out in full on
H29 fr. 1, so was that of F5 on H26 fr. 86, if it is recognised there; and at least o∞SbQq[fRb]
was certainly written. This constitutes a problem for an alphabetical order by community in
so far as the practice of the ancient editions of ëchoral lyricí, now largely obscured by the
tacit re-titling operated by the most widespread twentieth-century editions of Pindar,43
seems never to have been to assign identical titles to consecutive compositions, but to use
the shorthand f�b RÃf�b (fg~] RÃfg~], etc.) instead when the need arose.
The medieval transmission of the Epinicians is quite consistent on this score. In the
case of O. 3, 5, and 11, only some later manuscripts add the name of the victor to the plain
f�b RÃf�b of the earlier witnesses; for P. 5 readings are somewhat more varied, but always
with f�b RÃf�b coming first;44 P. 2 and 3 show greater variety, but f�b RÃf�b occurs in a
majority of the manuscripts, and can plausibly be regarded as the original form of the
heading.45 The commentary to I. 4 preserved by H25 fr. A 2(a) is indeed titled f]�b RÃf[�b.
H29 fr. 3 fg~] Rãfg~], if it does belong with that manuscript (or to another manuscript of
Pindar), offers the only surviving example in non-epinician Pindar.46 Bacchylides shows
the same tendency: odes 2, 4 (with the addition H�pbR, since 3 was [∆V�]�cbR), and 7 are
43 The reference is chiefly to the successive Teubners of Schroeder, Snell, and Maehler, but Bowraís OCT displays the same flaw. On the contrary, titles are given largely in their transmitted form in Puechís BudCs, Turyn 1948, and the Mondadori texts and commentaries of Gentili et al. 20064 and 2013; Gentili et al. 2013 and Turyn 1948 go the full distance and also provide an apparatus to the titles. 44 Mommsen 1864: 28, 39, 106, 208. 45 Mommsen 1864: 147, 159, Schroeder 19005: 63f. 46 Alessandro Pardini (ap. DíAlessio 1997: 38 n. 94) claims that the hand in which the title is written is different from that which wrote the title of F3 on fr. 1; the same is certainly true of the layout of the title and the shape of the accompanying asteriskos. In the absence of any stronger evidence for it, the ascription of fr. 3 to H29 should be regarded as doubtful. On the other hand, the asteriskos and, given the parallels listed here, the title itself still strongly suggest pertinence to a manuscript of ëchoral lyricí. See p. 135 below.
22
all titled f�b RÃf�b in A.47 No contrary example in either author appears to have been
published to date.
If this was indeed common practice and the ascription of H26 fr. 86 to the Prosodia is
held to be correct, then it follows from their written-out titles that both F3 and F5 opened a
sequence o∞SbQqfRb] W∞] o∞Rj¦Q, or at least did not immediately follow a composition
with the same notional title; which contravenes the one sequence per community which the
hypothesis of an alphabetical arrangement by community necessarily entails.
Thus, we are left with two routes. We may suppose that the sequences *F4-F5 and
F2-F3 were placed in the first and second books of the Prosodia respectively, assuming
that H7 ñ which contributes to both sequences ñ preserved fragments of both books, each
with a distinct alphabetical sequence. (However, one wonders why two odes for the same
community and addressee should be so split if coincidence of precisely these data was an
ordering criterion.) Otherwise, for the hypothesis of an alphabetic order by community to
stand, the coincidence of H26 fr. 86.6f. with the title and opening line of F5 must be
illusory. An ordering by deity does not necessarily face the same problem, if the
arrangement was flexible enough to allow e.g. W∞] o∞Rj¦Q, W∞] �c¦VVTQR, W∞] o∞Rj¦Q,
but the slight oddity of an alphabetic ordering of the titles by a word other than the first
remains intact.
As the identification of H26 fr. 86.6f. with the title and first line of F5 cannot be
proved beyond doubt, to note its contradiction with an alphabetical arrangement by
47 The title of B. 5, which should have fallen in the same category, is missing altogether from A. In that manuscript, interestingly, the title f�b RÃf�b of B. 7, omitted by the first scribe like that of the preceding ode, appears to have been written by the second corrector (A3) over three abraded lines penned by the first (A2) which may have been the odeís full title (so Kenyon 1897: 65; in 2013: 57f. n. 30 I mistakenly ascribed this view to Snell, who merely reprises Kenyon). If correct ñ but there is no clear evidence for it ñ this strengthens our point: A2 had divined the title of 7 from the text itself (Snell 19345: 10*) without also supplying that of 6; once A3 took care of 6, it became necessary to alter the title of 7 accordingly.
23
community brings us no closer to establishing with absolute certainty which of the two (if
either) is correct. An alphabetic arrangement cannot be excluded altogether, although it
lacks conclusive proof and entails the rejection of an otherwise appealing hypothesis on
H26 fr. 86.6f. By the same token, the choice to identify H26 fr. 86.6f. with the title and first
line of F5 and thus reject the alphabetic hypotheses (with no obvious alternative in sight) is
ultimately a matter of personal preference. Ultimately, the internal organisation of the
Prosodia remains frustratingly unknown.
In this edition I have retained the sequences verifiably or arguably offered by the
papyri: F2-F3, *F4-F5, *F6-*F7, F8-F9, and F9-F10. That the order of those sequences
(plus F11) is roughly alphabetical by what can be conjectured to have been the community
that commissioned the respective poems (with the obvious exception of *F4, and noting
that the performers of F2, *F6, and all the more F10 are beyond conjecture) is due purely
to practical convenience and to the fact that any other ordering criterion would have been
equally arbitrary in the present state of our knowledge. F1 opens my collection for no
better reason than that it seemed particularly suited to the role; Aristophanes reportedly did
the same with Olympian 1, as did whoever prefaced fr. 1 V. to the first book of Sappho.48
The two other fragments known from the indirect tradition (F13-F14) are printed after the
major papyrus fragments, and four smaller fragments from papyri (F15-F18) conclude the
collection. For reasons of time and space, I have not included the remaining papyrus
fragments at this stage, nor have I commented on all the fragments that are included, but
only on F1-F3 and F5-*F7.
48 On O. 1 see VV p. 7 Dr. with Negri 2004: 25-34, pace Slater p. 146 (see p. 53 n. 37 below); on Sappho fr. 1 V. see Lobel 1925: XV.
24
The new numeration of the fragments, as always, calls for an apology. Up to now,
anyone wishing to edit, or merely to cite, Pindarís Prosodia from the existing editions,
must choose two of three different numbering systems: for F1, *F4, F13, and F14, the
continuous numeration of all of Pindarís fragments that Maehler took over from Snell,
Snell from Schroeder, and Schroeder from Bergk; for the fragments previously thought to
be Paeans, either the established numeration or Rutherfordís, which is more
comprehensive and respectful of the manuscript evidence but also distinctly more
cumbersome. (Anyone who chooses to use Maehlerís fragment numbers for the Paeans too
does so at the price of utter unintelligibility for all who do not have the edition ready at
hand.) None of these systems was ever meant for the Prosodia, and the result is painfully
clear: even adopting the least unconventional combination (the Paean numbers when
available and Maehlerís fragment numbers for the rest) would produce several wide gaps,
sometimes an absurd order, and a poem that starts with verse 123 and follows one called
ìfr. 215(a)-(b) + Pae. 22(h).1-5 = fr. 52w(h).1-5î. Furthermore, none of these editions is
complete: a reader wishing to view as comprehensive a picture of the manuscript evidence
for the Prosodia as can reasonably be garnered has to supplement Maehler and Rutherford
with no fewer than five separate publications of the respective originals, and the
unfortunate editor is left with the task of sorting together seven different series of numbers
with their respective names, gaps, and overlaps. This being the case, it is for the readerís
convenience no less than for my own that I have renumbered all the fragments into one
sequence, with a view to extending it to all the fragments of the Prosodia once my edition
reaches completion.
25
SCHOLARSHIP
Very little is known of scholarly activity on the Prosodia after the Alexandrian edition was
compiled. The only name we read is Aristonicus, cited once on F5, whether for an
emendation or for an explanation (| 59, mutilated): it is unknown if the source is a work
on the Paeans ñ several traces of which survive in the margins of H4 and of the relevant
portion of H26 ñ or on the Prosodia; economy somewhat favours the former. An
interpretation is ascribed to unnamed fbQW] by | F3.13 and perhaps | F5.8. All other
material is anonymous.
There is a possibility, already envisaged by its first editor, that H37 comes from a
commentary to *F4, whose verses (a).6f. it quotes at lines 4f.49 If this is so, and if my
tentative argument for regarding H34 (which preserves *F4) as a manuscript of the
Prosodia holds true, it follows that H37 is a commentary to the Prosodia, the only one
known to date. However, it is unknown whether the Pindaric quotation was a lemma or a
quotation made by the author of the commentary to illustrate a point. The possibility that it
was introduced by H^Qk]�[¦] d\[]b arguably suggests ñ but does not prove ñ the latter, as
does the fact that a further poetic quotation at line 9 (fr. dub. 348a Sn.-M.) cannot be
recognised in the remaining lines of *F4(a) and seems quite alien to its subject-matter.50
Neither point is conclusive: the second quotation need not itself have been a lemma, or it
may have come several lines later in the poem, outside the preserved fragments and after a
change of topic. Nonetheless, the grounds for identifying H37 as a commentary to *F4 are
relatively slight, and although its text can be found after *F4(a) in this edition (since it
49 Lobel 1961g. 50 ibid.
26
does at any rate offer ancient scholarly material on the poem), I have not reckoned the
papyrus among the manuscript witnesses of the Prosodia in Chapter 7.
27
(3)
The genre, I : Preliminary remarks
Before proceeding to the reading of the individual fragments that constitute our record of
Pindarís Prosodia, our discussion of the books of the Prosodia requires a few words ñ or,
to be more precise, a few dozen pages ñ on the subject of the defining quality of the
compositions that they contain, that is to say, of the prosodion as a ëgenreí.1
There is no doubt that, at least in principle, the distribution of poems into books in
the Alexandrian editions of Pindar and Bacchylides follows a generic criterion. Poems are
assigned to individual books according to what kind of poems they are, and books are
explicitly titled after the kind of poems that they (mainly) contain.2 Even in the one book
that seems to have no explicit generic identity of its own but is defined by (a degree of)
lack of relation to another, the Separate from the Partheneia (jWZT[b]_YQR f�Q
cR[pWQW^TQ) that followed the two books of the Partheneia proper, the title ñ which
seems to have been in the neuter plural, not singular ñ refers this quality to the poems, not
to the book as such.3 When the appropriateness of inclusion in a given book is disputed, the
1 Given the importance of keeping book and genre distinct in the face of the outright identification that one might be tempted to infer from their apparent coincidence in the Alexandrian edition, throughout my discussion I will distinguish as far as possible between (a) genres and individual poems as belonging to such genres, which will be referred to in normal font (ëpaeaní), and (b) books and individual poems as belonging to such books, which will be referred to in the capitalised italics that we routinely use for book-titles (ëPaeaní). For example, Pindarís ëhymn to Paní (frr. 95-100 Sn.-M.), which according to | P. 3.139a p. II 81 Dr. was one of the kekhorismena (on which see n. 3), is a hymn but not a Hymn, while both the Nemeans and the Isthmians end with poems that are neither Isthmian nor Nemean odes (see n. 5 below). 2 Titles of this kind, consisting of a neuter plural, are comparatively common, cp. also the section-headings of the Milan Posidippus (P.Mil.Vogl. 309) with Krevans 2005: 83-8. 3 There is conflicting evidence over the precise title of the book, whether it was jWZT[b]_YQR f�Q cR[pWQW^TQ (| P. 3.139a p. II 81 Dr., | Theoc. 2.10b p. 271 Wendel1, and the papyrus scrap published by
28
debate may revolve around whether the relevant ode ëisí of a certain kind or another, such
as the debate between Callimachus and Aristarchus on whether Bacchylidesí (?)
Cassandra ëisí kbpP[R_�bjq or should be regarded as a paean (**23 M. on B col. i.9-19).4
As the example of the non-Nemean Nemeans 9-11 shows,5 the advertised principle of
generic consistency within a book is not always strictly maintained, but this only highlights
the significance of the commitment to an overall genre-label even if not wholly suited to
the content.
What does it mean, then, to ëbeí a prosodion? Before attempting to give an answer,
two qualifications to the question itself are in order. The first concerns the thin ice on
which one risks treading when talking of generic identity. It is not merely a matter of the
Lobel 1961a: 6 n. 1) or jWZT[b]_YQR cR[pWQW~R (the transmitted reading in T2). Beside the obvious numerical advantage for the former, of the two it is also the latter that can most easily have arisen out of the other; it is also puzzling insofar as, if those poems were also regarded as Partheneia, it is hard to surmise why they should have been regarded as separate from them (note, however, that T1 knows ìthree (books) of Partheneiaî, with no particular separation). Snell 19531: 191 emended the text of T2 so as to have the adjective in the singular, not the plural, but this gives an unattested title; one even wonders if a typographical error intervened, since Snell used the plural in the relevant section-heading at 250 (an inconsistency which was carried over five editions up to Sn.-M. inclusive). In any case, whereas this difference has great import for the generic identity that was attached to the poems contained in that book, it does not affect the fact that, even in the midst of this identity crisis, the title of the book explicitly reflected the nature (whether positively or only negatively identifiable) of the poems that it contained. 4 On the Cassandra see below, pp. 53f. 5 N. 9-11: inscr. N. 9 p. III 150 Dr. Beside those, the extant remains of the Pindaric corpus offer two further examples of generic mismatch between a book and parts of its content. As we know from H25 frr. 14 and 17, the Isthmians contained at least two odes unrelated to Isthmic victories: one for two (or more?) pancratiasts victorious at the Corinthian Hellotia (fr. 6a(i)-(l) Sn.-M. as reinterpreted by DíAlessio 2012: 48-54), and one for an unnamed Athenianís victory in the local Oschophoria (fr. 6c Sn.-M. with Lobel 1961h: 177, see also Rutherford and Irvine 1988, Negri 2002). The Paeans contained at least one poem called a cRbÏQ c[g]gkbRj¦] (Pae. 4?), one plainly labelled a c[g]¦kbgQ (Pae. 6.123-183 = F5), and a third that was probably in the same situation as either of these (Pae. 7, whose title on H5 fr. Vv ends with c[g]×É�[), see pp. 102f. below. The case of the Paeans is arguably different insofar as ëanomalousí poems were not relegated exclusively to the end of the book (Pae. 5 may have been labelled as c[g]gkbRj¦], but Pae. 6 was not, as far as H4 goes), as instead they were in the Nemeans and probably in the Isthmians. (In this connexion I note that nothing in H25 guarantees that the poem for the Hellotia and the oschophorikon were the very last poems of the book: although the number of odes that preceded them ñ at least twelve ñ makes this view likely, they may have been followed by one or two other extravagant compositions.) Nonetheless, the principle is not unusual: items less relevant to the generic headings are also relegated at the end, albeit of sections rather than of books, in another third-century collection, the Posidippus papyrus, P.Mil.Vogl. 309 (Krevans 2005: 94-6). Cases in which the genre of a poem was actually disputed ñ like the Cassandra and like P. 2 (inscr. p. II 31 Dr.) and the other examples cited by Benelli 2012: 83 n. 11 ñ pose a different set of problems, which fall outside the scope of this discussion.
29
practical fuzziness that the boundaries between genres sometimes have, or of the fact that
notions of individual genres change over time. A deeper problem lies rather with assuming,
consciously or otherwise, that genres exist as entities in their own right, each with its own
identity and laws. This assumption risks portraying a historically over-simplified picture,
perhaps nowhere more so than for archaic and early classical Greece. This interpretation
ultimately posits a genre that comes, actually as well as in principle, before the texts that it
defines; it hypostasizes a set of norms which authors allegedly know and to which they
variously respond, whether by acceptance, modification, or rejection.
Luigi Enrico Rossiís seminal paper on classical genres is an excellent example of this
approach. Each genre has its ìlawsî which ever since the archaic age are ìpresent to the
authorsí consciousnessî;6 what is needed (and what Rossi sets out to provide) is ìa
comprehensive research that views the laws of genres from the outside, as an autonomous
historical givenî, ìnot so much a history of genres (Ö) as much as a history of the laws
that discipline the genres themselvesî.7 Rossi famously went on to periodise the
relationship between ancient Greek authors and genre accordingly: the ages of unwritten
generic laws which authors respected (archaic), of written laws which authors likewise
respected (classical), and of written laws which authors did not respect (Hellenistic).8
Rossiís approach has the welcome effect of focussing attention away from genre as a
purely literary phenomenon on to the external factors that influence it; he regards these as
ìa largely spontaneous product of the historical situation, in the widest sense, in which the
author operatesî with particular connexion to the audienceís requirements and
6 Rossi 1971: 73. 7 ibid. 73f. (emphasis original), cp. 71: ìit is the literary historianís province to reconstruct the formal laws which authors obeyed (Ö) the history of our studies is for the most part precisely the laborious recovery of these lawsî. 8 Rossi 1971.
30
expectations, which by and large had a stronger hold on the author and his choices than
poetic traditions, individual models, and even the committente, the person who
commissioned that particular piece, could have.9
This picture, however, can be further nuanced as regards these ìlawsî and their
working in archaic and early classical Greece. First of all, it is almost a platitude to point
out that neither Pindar nor any of his colleagues is likely to have ever met a ëgenreí or its
disembodied laws as such. What each of them must have known was a certain amount ñ
one supposes an extensive amount, although we regrettably know next to nothing for
certain about poetic training in archaic Greece ñ of songs composed for different
occasions; these songs, in turn, were (or could be) regarded as belonging together in what
we can reasonably term ëgenresí and from which they could extrapolate core themes and
forms and whatever elements they may have regarded as characteristic of each particular
genre for re-use in their own poetic undertakings.
This is not idle punctiliousness, pointing out the detail beyond the generalization as
though it had not been known all along; the effect of this discussion, I argue, is to allow for
greater flexibility as well as historical accuracy than the essentialist model. If one
hypostasizes genre and its laws by placing them outside the texts that they informed, that is
to say, if one supposes that authors engaged with the genre itself and its laws more than
with one another, the practical risk is to flatten out the complexity, on the one hand, of
each individual authorís engagement with the literary (generic) tradition in which he or she
was situated, and on the other, of the traditionís change over time, thus painting a
reassuring but ultimately illusory picture of stability, the ëideaí of a genre that gives rise to,
instead of depending on, each of its concrete manifestations.
9 Rossi 1971: 70f. (quotation from 70).
31
If it is poems and poets that we focus on, rather than the abstract genre (in other
words, if we look at the question more as a matter of parole than of langue), then the
picture of the genre itself, of its articulation, and of poetsí relation to it becomes instantly
richer ñ and, of course, messier. Within any given genre, the composition of each new text
(I use this term for the sake of simplicity, but without restricting my perspective to the
written, or even the exclusively verbal, aspect) draws from and simultaneously adds to the
existing body of texts within which it situates itself; thus each addition, in turn, contributes
to shaping the tradition with which future authors will be confronted. By the same token, to
each addition corresponds in principle a relative diminution in the normative value of each
individual text in the existing tradition.
Of course, for a variety of reasons, some texts will be more influential than others,
their value as models being higher and more durable; conversely, as new ones take their
place, older models, some more, some less slowly, may fall out of use, and although they
may not cease to exist altogether, they may progressively cease to offer a living
contribution to the continuing life of that tradition, that is, to influence the new elements
that are composed into it. Especially in a culture where works of literature concretely exist
in a largely oral or aural milieu, an authorís contact with the notional entirety of a tradition
cannot be assumed; and it will have been only this accessible or accessed part of the
tradition, case by case, that was in a position to influence each poetís choices for each of
their compositions and thus, in the long term and with due room allowed for internal
variety and instability, the shape and history of the whole genre. As Thomas G.
Rosenmeyer noted in a provocatively-titled paper a few years after Rossiís: ìThe prestige
of the father and the rivalries within the family account most satisfactorily for what
stability there is in the formal and aesthetic continuities over the years, while also
32
explaining the great variety of creative departures. There is no need for an appeal to a
biological model of growth and decline as long as the model of the family quarrel, or rather
the playful engagement with the parent, is availableî.10
So much for the literary tradition and poetsí response to it; but the audienceís
expectations and poetsí response to those, which Rossi rightly emphasised, can be argued
to follow a similar pattern, whereby they will have been shaped by the actual repeated
contact with several texts and performances associated with the various genres at least as
much as (if not indeed more than) by independent, pre-formed judgements on what a genre
ëisí or ought to be. This requires, once again, allowance for greater flexibility than the
static notion of ëtheí audience: an audienceís taste (or, more properly, the taste of
successive audiences) can change over time, due to a variety of factors which include the
development of the tradition itself and the broadening of the products it offered, so to
speak. Although audience response may be on occasion a constrictive force, it is also a
dynamic one, which can be shaped by authors as much as it shapes them; when its
constrictive force is to be put into effect on an authorís extravagance from what is expected
or deemed appropriate, one supposes that it works more through trial and error than
through some pre-emptive curb on creativity. An unpopular tragedy is not necessarily any
less a tragedy than Sophoclesí Antigone; rather, it will be one that may be treated with
particular caution by other tragic poets when devising their own tragedies, consciously or
less than consciously sidelined as a model, as we have seen above.
This, as we were saying, makes the question ëWhat is a prosodion?í intrinsically
problematic. The problem is fundamentally historical: if the understanding of each genre
10 Rosenmeyer 1985: 82. Calame 1974: 126 had rightly pointed out that this tendency to push genre ënormsí does not begin with the Hellenistic age, as Rossi had surmised, but was already in place in the late Classical period ñ and, we might add, earlier too.
33
changes according to circumstances, both over time and in terms of locale (another
deciding factor in the largely ëoccasionalí cult poetry of Greece), one cannot assume that
there will be one answer that suits all the evidence ñ all the poems that have been called
prosodia at one time or another, in one place of another ñ and gives us the key to the
identity of ëthe prosodioní. What could, in a certain set of circumstances, define a
prosodion? What were the themes that prosodia could treat, the structure and formal
characteristics that they could have? On what occasion could prosodia be performed, and
in what way? How could any given audience recognise a composition as a prosodion?
Although, as we will see in greater detail in the next three chapters, our evidence is patchy,
disconnected, and sometimes confused (and confusing), it is to these questions that our
inquiry will attempt to seek answers.
This leads us to our second, more practical stumbling-block, namely the nature of the
evidence at our disposal. When dealing with any of the genres into which Pindarís poetry
was classified, and indeed with any ancient genre, we are faced by two broad categories of
evidence. On the one hand, we have texts, whole or fragmentary, of odes belonging to that
genre, or information about such odes; we may borrow a metaphor from textual criticism
and term this the direct tradition of the genre. On the other hand, we have other authors and
texts describing or alluding to the traits that, in their view, characterize the genre and
distinguish it from others: we may call this its indirect tradition. As in the transmission of
texts, so in the transmission of genres the two traditions do not necessarily coincide. When
both the texts themselves and the genre to which they are said to belong are so
fragmentarily attested, as is our case, it is necessary to use both traditions to a considerable
extent; this may entail doing so even when it is not possible to compare the two and to use
one to confirm or contradict the other. Furthermore, just as in textual criticism, what for
34
convenienceís sake we may call the indirect tradition is not in fact a tradition, but a
multiplicity of traditions, to varying degrees independent of one another, each of which
must therefore be considered independently as well as in its potential relation to the others.
The picture is further complicated by the mediated status of the very texts onto which
we are primarily directing our enquiry. As far as the evidence goes, Pindarís Prosodia
were not collected as such by Pindar himself, and may not even have been composed as
such by him; the most important and most substantial part of our direct tradition is itself
ultimately indirect.11 What difference there may have been between Pindarís notion of a
prosodion and the generic criteria that may have underlain the Alexandrian edition of the
Prosodia is ultimately unknowable; indeed, there is no solid evidence that he would have
regarded any of his poems as prosodia, or that the term c[gc¦kbgQ would have meant
anything to him as a distinct generic denomination.12 Unlike other genres, which have a
more solid record starting from an earlier date ñ the paean, to name but one, which makes
its first recorded appearance in the Iliad (1.472-4) and is called upon several times in
Pindarís own work, once in the context of an explicit division of genres (fr. 128c.1f. Sn.-
M.) ñ the word c[g]¦kbgQ is first attested in the late fifth century, and the genre as such
seems to have had a rather shady existence until it was canonized as a component of
Pindarís oeuvre by Aristophanes of Byzantium.13
11 On the Alexandrian edition of Pindar see also pp. 5f. above, 53 below. The point that Alexandrian and post-Alexandrian scholarship is not a reliable guide for fifth-century notions of the lyric genres was made by Harvey 1955: 157 in relation to the likes of FÅrber 1936 and Smyth 1900, who tended to disregard the historical development of such notions and retroject later scholarly notions onto the fifth century; for the methodological consequences see Calame 1974, esp. 114. On the classificatory practice of Alexandrian scholars see also Gelzer 1982-84 and SchrNder 1999a: 110-53. The question of the diachronic change of notions of genres and of the terminological and eidographic issues that this implies has been recently applied, among others, by Rutherford 2001: 3-10, 90-108, 152-8 to the paean, by Fearn 2007: 165-225 (esp. 205-12) to the dithyramb. 12 A similar point is made by Ag`cs 2012: 212 n. 133. 13 See p. 53 below.
35
Thus total or partial coincidence between any two sourcesí views of what a
prosodion ëisí beyond what the evidence unambiguously tells cannot be assumed; one must
resist the easy temptation of demanding that all the piecemeal evidence squares into one
picture, whether that of a static identity of ëthe prosodioní or even that of its organic
development from the beginning to the end. If an ideal repository had preserved every text
and performance, generation by generation, that the poets of Greece conjured into
existence, then we might perhaps be able to read the entire history of its traditions as one
reads in its trunk the whole history of a tree, branch after branch; but it was only a few
flowers ñ perhaps the most beautiful, but already dried of their sap of music and dance,
picked apart from the occasion that created and nourished them ñ that the scholars of
Alexandria, at the zenith of that age of anthologies, were able or willing to collect in their
wreaths of ancient poetry.14 And not even these were fortunate enough to rest undisturbed;
to this day we stoop to gather a few withered petals and sew them together into what can
only be a pale impression of that Hellenistic herbarium.
Given this state of things, it is inevitable that rather than using our knowledge of
certain generic conventions to illuminate and nuance our appreciation of the fragments, as
one can do with (say) Old Comedy or Latin elegy, we are only able to use the scanty
primary material to sketch a blurry picture of what the genre could be, the reverse being
next to impossible. Nonetheless, we shall devote the next three chapters to a thorough
14 For the anthology (specifically the anthology of epigrams) as a literary genre that rises to prominence in the Hellenistic age see Rossi 1995: 641f.; see also Gutzwiller 1998 passim esp. 11-46. Aristophanesí Pindar cannot be equated to Meleagerís Garland, obviously, but both of these artefacts, along with their several relations, can be seen as products ñ however different ñ of a context which was exploring in various directions the collection and arrangement of earlier poetry as an intellectual pursuit. As Martin West points out to me, on papyri too the heyday of anthologies ñ not exclusively anthologies of epigrams ñ is the Ptolemaic age, after which they become much rarer; on such papyrus anthologies see Pordomingo 1994, 2007, and 2013. For a stimulating overview of this combination of phenomena see Hutchinson 2008: 4-15; a comparison between the collection of Pindarís Olympians and Hellenistic poetry books had been proposed also by Negri 2004: 30-2.
36
examination of what ancient sources tell about prosodia, investigating ëthe prosodioní as a
historical construct that can be seen to have played a crucial role in the constitution and
transmission of the body of texts that is the primary subject of our inquiry. Chapter 4 will
examine what we have termed the direct tradition, that is, ancient attestations of
compositions called c[g]¦kbR, with the exception of Pindarís; Chapter 5 will focus on the
indirect tradition, namely what various scholarly sources, both Greek and Latin, claim that
a prosodion ëisí; finally, in outlining certain characteristics that can be traced across the
fragments of Pindarís Prosodia, Chapter 6 will attempt to understand some of the
principles that the Alexandrian classifier may have followed in drawing them together.
This is not a virgin field, but nonetheless one that has been ploughed somewhat less
than the evidence can sustain: a complete bibliography of modern scholarship on the
prosodion can easily be brought together within the compass of a single footnote.15 The
present attempt appears to be more extensive than any that has been published to date, with
several previously neglected testimonies added to the tally in Chapter 5 and (I hope) a
more thorough examination of their content and its usefulness for our enquiry.16 I willingly
plead guilty of probably presenting more than anyone ever wanted to know about prosodia,
15 Smyth 1900: XXXIII-VI, FÅrber 1936: I 30f., 48f. (testimonies collected at II 29f.), Muth in RE XXIII/1 (1957) 853-65 s.v. Prosodion, Grandolini 1987-8 (see also 1991), DíAlessio 1997: 25-40, Rutherford in NP X (2001) 447f. s.v. Prosodion and 2003, Pinervi 2010-11. The fullest discussion to date is Grandolini 1987-8, structured along the historical development of the genre, although it predates the reclassification of the Pindaric papyrus fragments effected by DíAlessio (see pp. 12f. above) and thus ignores most of the material that is now available and focuses almost exclusively on the testimonia; the same, rather surprisingly, does Pinervi, although his article was published well over a decade after DíAlessioís. DíAlessio, in turn, provides the fullest discussion of the papyri while not dwelling at too great length on the ëindirect traditioní. Rutherford 2003 is the most recent treatment of the subject in English, with several useful insights although somewhat obscured by a few oversights (for one of which see p. 78 n. 12 below) and some pretty dreadful proof-reading on the publisherís part; his entry in the NP is as handy an epitome of the state of scholarship on the prosodion as one could wish for. 16 The specification ìpublishedî is rendered necessary by a hitherto unpublished doctoral dissertation, which I have not seen but which is mentioned by Manieri 2009: Alessandro Pinervi, Il prosodio: le testimonianze antiche, UniversitØ degli Studi di Urbino, 2003-2004. I can only suppose that Pinervi 2010-11 represents an extremely condensed version of this dissertation.
37
but I hope that some of the evidence laid out in the pages that follow will be useful to the
reader in making up their own mind regardless of whether my conclusions ñ and, in several
cases, my honest inability to offer one ñ are regarded as persuasive.
39
(4)
The genre, II : ëDirect traditioní
EUMELUS
According to a widespread scholarly consensus, the earliest known fragment of lyric
poetry in Greek comes from a prosodion, namely from the Ãb]_R c[g]¦kbgQ that Pausanias
in his description of Messenia (4.4.1) ascribes to Eumelus of Corinth, best known as an
epic poet, who reportedly composed it for the Messeniansí first sending of ìa sacrifice and
a chorus of menî to Delos during the reign of Phintas son of Sybotas, shortly before the
first Messenian War in the eight century.1 Towards the end of the book, Pausanias quotes
two lines from the poem (4.33.2 = PMG 696), and mentions it a third a time at 5.19.10: no
other lyric composition is cited so often in the Description of Greece. Nothing is known of
its content beside the one surviving fragment, transmitted as a dactylic hexameter followed
by a dactylic pentameter in the Aeolicizing ëDoricí of choral lyric:
f�b SÏ[ ∏pT_nfRb jRfRp�_bg] ¯cVWfg Ùg~]R Ú jRpR[Ï jRÚ �VW�pW[R ]n_�RVí ¯Zgb]R
2 post Ú jRpR[Ï lacunam statuit Hitzig 1896-1910 p. II/1 86 : Ú jRpR[Ï�Q j^pR[bQ� Bergk 1871-83 p. II 116 adn. 28 : Û jbpn[RQ Webster 1970 p. 51 | ¯Zgb]R Dindorf 1845 p. 221 : ¯ZgP]R codd.
The reference to the Museís ìfree sandalsî has been understood as a thinly veiled
assertion of Messenian independence, played out in terms of Messeniaís musical identity
1 On Eumelus and ëhisí epics (BernabC pp. 106-14) see Wilisch 1875, Will 1955: 124-9, West 2002, Debiasi 2004: 19-69; on the prosodion Bowra 1963, Pavese 1987, Grandolini 1987-8: 29-33, Debiasi 2004: 39-48, Caprioli 2007, DíAlessio 2009a: 140-5, Tausend 2012.
40
and its highest divine sanction.2 Pausanias interestingly also cites the poem to support the
widespread conjecture (fv] kÓ ÕcgQg^R] fÙ cgV�) that the captions to some of the scenes
depicted on Cypselusí Chest at Olympia were also due to Eumelus.3 He does not elaborate
on what connexion was seen between the two artefacts, but Henry Stuart Jones must be
right that it involved the combination of the captionsí dialect and metre and the Chestís
Corinthian origin: Eumelusí name must have come to mind naturally enough as the only
archaic poet of note to have hailed from Corinth, and unlike his epic production (which
was in unexceptionable Ionic, as one would expect, and whose authenticity moreover was
open to doubt), the ëDoricí veneer of the prosodion provided an especially close parallel to
the otherwise uncommon ëDoricí hexameters that were inscribed on the Chest.4
Nonetheless, there are good reasons for doubting Pausaniasí account of the
composition of the poem, as was independently realised by a handful of scholars over the
past several decades. Already at the end of the nineteenth century Valerian von Schoeffer
had remarked the unlikelihood of Messenian participation in an Ionian festival and the
suspicious linguistic veneer (the Aeolic Ùg~]R, the ëLydianí ]n_�RVR) of PMG 696, and
consequently suggested that both the poem and the story of its composition were fabricated
2 Pohlenz 1955: 190, Bowra 1963: 152, Cassola 1964: 271f.; contra, Pavese 1987: 55f. Whether ¯cVWfg has a past or a present value (Bowra 1963: 151, Caprioli 2007: 32f., DíAlessio 2009a: 139 n. 3), there is no need to suppose that an action described in the past cannot be conceived as continuing into the present: reminding a deity of his previous favour so as to highlight its envisaged continuation is an established topos of religious poetry (Grandolini 1987-8: 31), and an intimation of discontinuity would presumably destroy its point. On the cult of Zeus on Ithome see Otto 1933: 18-27, Robertson 1992: 219-31, Zunino 1997: 85-94, 101-9. 3 5.19.10, concluding the description of the Chest which takes up 5.17.5-19.7; within the description, nine captions are quoted verbatim (5.18.2-4, 19.3-5). See Debiasi 2005, Borg 2010. 4 Stuart Jones 1894: 39f.; Paus. 4.4.1 claims that the prosodion was Eumelusí only genuine work. For other, less persuasive speculation on the connexions between the Chest and the prosodion see Gallavotti 1978: 13, 31, Debiasi 2004: 40 n. 120, 2005 (but no part of Cypselusí Chest seems to bear on Delian mythology: all the hypothetical points of contact between the Chest and Eumelus investigated by Debiasi 2005: 47-55 concern his epic production, not the prosodion, whereas Pausanias states the opposite).
41
by the fifth-century Messenian exiles in Naupactus.5 On similar grounds and noting that
�VW�pW[R intimates a claim to political freedom, Max Pohlenz suspected that the
prosodion was a fourth-century piece aimed at showcasing a putative native poetic culture
that the newly independent Messenia could retroject to an earlier time than her subjection
by Sparta.6 Maurice Bowra also noted the incongruity between such evocation of freedom
and Pausaniasí dating to a period when Messenian freedom was not evidently in dispute,
but opted for minimal intervention and downdated the prosodion by only a few years, to
the early stages of the first Messenian War.7 Conversely, Noel Robertson pointed out that
no element of Pausaniasí internally coherent account needed to be retained once another
was discarded, and that the time of the Messenian settlement at Naupactus ñ sponsored by
Athens and a staunch Athenian ally ñ was the only period in which a Messenian theoria to
Delos was likely to have taken place.8
More recently, Giambattista díAlessio expressed doubts over the possibility of a full-
scale Messenian state mission abroad in the eighth century, remarked on the lack of
5 Schoeffer 1889: 7f. On the linguistic arguments, reportedly suggested by Diels, see Cassio 2005: 22f. n. 32, DíAlessio 2009a: 142-4 and n. 26. When a revolt of Helots and perioeci against Spartan rule resulted in the rebels being besieged on Ithome but protected by an oracle, the Athenians resettled them and their families in Naupactus (ca. 456/5): Thuc. 1.103.1-3, D.S. 11.84.7f., Paus. 4.24.7. The arrangement survived until the end of the Peloponnesian War: DS 14.34.2, Paus. 4.26.2. See chiefly Luraghi 2001, 2008: 188-94, 2009: 111-15. DíAlessio 2009a: 144f. notes the claim laid on Zeus of Ithome by the Messenians of Naupactus, who commissioned his statue to the sculptor (H)ageladas according to Paus. 4.33.2 (see Cook 1914-40 II/1 741-3, Luraghi 2008: 176f.). On the chronological problems that the attribution of this and other works to (H)ageladas implies see among others P. Moreno in Kunstlerlexikon der Antike I 275-80 s.vv. Hageladas (I)-(II) (postulating two sculptors of the same name), Cartledge 1979: 154 somewhat anticipated by Seltman as reported in Cambridge University Reporter LXII/31 (14th April 1932) 799 (see Cook 1914-40: III/2 1153f.: suggesting a statue made for Messenian rebels in the early fifth century), Luraghi 2008: 176f. (a perioecic or even Spartiate dedication from the late sixth or early fifth century). 6 Pohlenz 1955: 190. Messene was ërefoundedí in 370/69 (Paus. 4.27.9) or 369/8 (D.S. 15.61.1, 66.1): see Roebuck 1941: 31-4 (with further sources cited at n. 21), Alcock 2002: 164-75, Luraghi 2002a: 61-4, 69, 2008: 209-48, 2009: 115-31. 7 Bowra 1963 : 152. Of course it is not intrinsically unreasonable to evoke liberty when it is under obvious threat although not lost altogether, see first and foremost the closing lines of P. 8 (Cassola 1964: 271f.). On Andrewesí hypothesis of a theoria sent by supposed Messenian exiles in Chalcis (in Bowra 1963: 152f.) see the opportune qualification by Debiasi 2004: 46 n. 154 and Caprioli 2007: 33-5. 8 Robertson 1992: 224. On the pro-Athenian zeal of the Messenians of Naupactus see especially Paus. 4.25.1-26.1 with Luraghi 2008: 188-91.
42
archaeological evidence for Peloponnesian contact with Delos throughout the Geometric
period, when the poem was composed in Pausaniasí account, and noted the similarity in
metre and style between PMG 696 and certain examples of archaising fifth-century lyric;
accordingly, he cautiously opted for an attribution to the Messenian refugees at Naupactus
over both the archaic Messenians and the fourth-century inhabitants of the newly founded
Messene.9 Klaus Tausend now favours a fourth-century date, but a theoria to Delos ñ and
even more so the series of later theoriai implied by Pausaniasí c[�fgQ (4.4.1) ñ suits
better a period of close relationship with Athens than the years after 369, when such
relationship was occasional at best.10 This does not mean that a sacred mission of this sort
could not possibly have been sent by the citizens of Messene in the fourth century, but it is
fair to say that the fifth-century Naupactians had more and better reasons to do so.11
This argument is predicated upon one important assumption, namely that Pausanias
is correct to ascribe the prosodion to a theoria to Delos, whatever the reliability of the
other information he provides. If he could read the whole poem, this is the natural
supposition: songs destined for cultic ceremonies in a specific locality and occasion tend to
indicate this quite explicitly. If this is so, it is indeed legitimate to conclude that an early
archaic date for the prosodion is comparatively unlikely, and that a dating between the
9 DíAlessio 2009a: 140-5. On eighth-century Messenia see Morgan 1990: 65-79 and now Luraghi 2008: 108-17, 145. Archaeological data from Delos in Rolley 1973: 524, Chankowski 2008: 18; contrast the plentiful evidence for Messenian involvement with Olympia in Morgan 1990: 89f., 93, see also Luraghi 2008: 111). Fifth-century lyric in dactylic metre intended to evoke archaic citharody: Fraenkel 1918: 321-3. Language and metre of the fragment: Cassio 2005: 22f. (the metre is strongly archaising if not archaic), Caprioli 2007: 22-7 (admitting linguistic compatibility with the later lyric tradition at 26). 10 Tausend 2012: 74-7, a view already pre-empted by Robertson 1992: 224 n. 15. Tausendís argument ñ which, like all authors cited in the main text of this paragraph except DíAlessio, ignores all earlier contributions to the matter except Bowraís, and therefore makes no mention of the Naupactian hypothesis altogether ñ is partly predicated on Corinthian authorship of the prosodion; but once Pausaniasí attribution to Eumelus is rejected, there is no reason for a Corinthian connexion to stand. 11 Cp. DíAlessio 2009a: 144. ìThe surviving dossier of documentation attests a virtual blitzkrieg of self-assertion by the Messenians spanning about a generation from c. 460î (Figueira 1999: 215); on other Messenian attempts at pan-Hellenic display immediately before and during their control of Naupactus see Figueira 1999: 214f., Luraghi 2008: 186-8, 191-4.
43
second and the last quarter of the fifth century seems more plausible than any alternative
suggested so far. It is worth mentioning the outside chance that the poem might be a later,
Hellenistic, antiquarian product, never meant for actual performance and thus not grounded
in, nor datable by, the cult politics of any specific time. However, there is no particular
argument in support of this view; a connexion, even an imaginary one, between ëMesseniaí
and Delos is still not accounted for quite as neatly under this hypothesis as it is in the
reality of fifth-century Naupactus.
A final question arises, namely whether the poem was conceived from the start as a
pseudepigraphic work of ëEumelusí or was composed ìwithout intent to deceiveî
(Robertson) only later to become attached to Eumelusí name.12 The metre of the song
resonates with overtly archaising fifth-century dactylic poetry,13 but this need not imply a
conscious attempt at historical forgery as much as evocation of an ancient and authoritative
lyric tradition, as the available parallels imply. Only Occamís Razor gives a slight edge to
the hypothesis of an intentional pseudepigraphon, one which ìmight have presented itself
as the re-performance of an original song projected to a foundational pastî (DíAlessio):14
with the authority of an archaic ancestor of their fifth-century practice, the Naupactian
community will have legitimated both its claim to Messenian identity and (no less
importantly) its potentially problematic stake in the Delian cult.15
12 Robertson 1992: 224; see also DíAlessio 2009a: 144f., who does not address the question explicitly but speaks of a ìforgeryî (145). 13 See n. 9 above. 14 2009a: 145. 15 On Messenian identity in the fifth century see Figueira 1999, Alcock 2002: 134-63, Luraghi 2001, 2002a: 50-61, 67f., 2002b: 592, 2008: 188-208.
44
PRONOMUS
Another c[g]¦kbgQ W∞] }vVgQ, composed for the people of Euboean Chalcis, is the only
known work of the celebrated aulos player Pronomus of Thebes, whose activity may have
spanned the second half of the fifth century and the first decade of the fourth; once again
we owe the notice to Pausanias, who mentions it in a brief excursus triggered by his statue
in the Theban agora (9.12.6 = PMG 767).16 Nothing survives of the text of this prosodion,
and it is unknown whether Pausanias was himself acquainted with it.17
Peter Wilson and, more cautiously, DíAlessio have dated the prosodion to the period
shortly after Chalcis regained its independence from the first Delian League in 411, since
the increasing links between Euboea and Thebes in those years would account nicely for
the choice of commissioning the poem from a Theban.18 However, one wonders if the
Chalcidian theoria to Delos that Pausaniasí statement implies could have taken place in a
period of secession from the Athenian-controlled League and thus, presumably, from its
16 The scanty testimonia on Pronomus are collected by H. von Geisau in RE XXIII/1 (1957) 748 s.v. H[¦Qg_g] 4; see further Roesch 1989: 208f., Berlinzani 2004: 127-9, P. Wilson 2007: 141-4, 2010. A life span ca. 470-390 is suggested by Wilson 2007: 143 (cp. 2010: 186f.) on the basis of the anecdote that he was Alcibiadesí aulos teacher (see n. 20 below), but his birth can easily be postponed by a few decades if the story is spurious (fl. 425-400 according to Cordano 2004: 318). On his prosodion see further Kowalzig 2007: 85, DíAlessio 2009a: 146f., Wilson 2010: 191-4. Smyth 1900: XXXV argues that Pronomus, being a musician, merely composed the melody for another poetís words, but this jars with Pausaniasí g� jRÚ Ãb]_R cWcgb\_YQgQ �]f^: Ãb]_R implies sung words, not music alone, and one can hardly take Ãb]_R cgbW~Q without qualification as ëto compose the music for a songí. 17 Pausaniasí phrasing seems to imply that the composition survived to his day (Wilson 2010: 191), but this need not indicate that he knew its text first-hand. Wilson further speculates (ibid. n. 46) on whether the prosodion may have been inscribed like the Delphic paeans, simply commemorated by an honorific inscription like Cleocharesí and Amphiclesí prosodia (see below), or listed in a catalogue of his works, ìthe latter made the more plausible by the scholarly form of his citationî. Preservation of scripts for reperformance or of book-copies of the text, perhaps in a collection of Deliaca such as that envisaged by Wilamowitz 1900: 38f. (ìunnecessarily in my opinionî DíAlessio 2009a: 144 n. 25), are also possible alternatives; or might Semus of Delosí HW[Ú cRbnQTQ (BNJ 396 F 23-24) have something to do with it? 18 P. Wilson 2007: 141, 2010: 192-4, DíAlessio 2009a: 147. Euboean revolt and independence in 411: Thuc. 8.91.2, 8.95.3-7.
45
central sanctuary.19 A musicianís nationality does not necessarily imply a political stance:
hostility between Athens and Thebes through most of Pindarís life did not prevent him
from receiving and accepting several Athenian commissions, public as well as private, and
Pronomus too seems to have spent part of his career in Athens with considerable success.20
Thus, I would incline towards a date before the rebellion of 411, in a period when
participation in the Delian cult may have been crucial to Chalcisí status as a member of the
League;21 but equally possible is one nearer the end of Pronomusí career, after Athensí
capitulation in 404 and her consequent loss of control over Delos (as per I.D�los 87, dated
to shortly after 402).22 Unless the composition is referred to a local and otherwise
unattested Delian cult, taking W∞] }vVgQ as the notional addressee rather than the concrete
19 Brun 1996: 9f. and Constantakopoulou 2007: 52f. note the scarcity of links to Delos that characterises all of Euboea except Carystus and Histiaea; contrast Kowalzig 2007: 85 n. 74. On Athensí dominance over Delos in the period of the first League see Meiggs 1972: 300-2, Parker 1996: 149-51, 2005: 80-2, Chankowski 2001 (esp. 84-6), 2008: 29-168 (79-125 on the religious festivals), Constantakopoulou 2007: 66-75. 20 Pindar: P. 7 (which, however, was written for an exile), N. 2, fr. 6c Sn.-M., possibly Pae. 5, 7b (see p. 177 n. 15 below), and 7c(c) (Rutherford 2001: 345), a Dirge for one Hippocrates (| P. 7.18a p. 206 Dr., cp. fr. *137 Sn.-M.), and the Dithyrambs frr. 74a, 75 (which could be identical to the former, pace Snell 19531 ad loc.; on the true meaning of 8 kW�fW[gQ see KeyÜner 1932: 11), and especially 76 Sn.-M. with the story (perhaps apocryphal, but not necessarily) related by [Aeschin.] Ep. 4.2f., Paus. 1.8.4, and VV p. 5f. Dr. On Pindarís involvement with Athens and the difficult of setting time-limits to it see Hubbard 2001, Hornblower 2004: 247-61. Pronomus: mainly the ëPronomus Vaseí (ARV2 1336.1, 1704), a possible allusion to him in Ar. Eccl. 102, and the story that he was Alcibiadesí aulos teacher (Duris BNJ 76 F 29 ap. Ath. 4.184d), see P. Wilson 2007: 143f., 2010: 196-204. Paus. 4.27.7 reports that melodies by Pronomus and by Sacadas were played during the building of Messene in 369 (cited in this connexion by DíAlessio 2009a: 147, Wilson 2010: 194-6), but this need not imply that ìthe Thebans did attach political implications to his workî (DíAlessio l.c.): especially given the presence of Sacadas, it rather seems that Pausaniasí source picked the most illustrious composer of aulos music from each of Thebes and Argos, which sponsored the re-foundation of the city (Luraghi 2008: 210, 214-8, 231f.), so as to illustrate the role attributed to those two communities, with no obvious reference to either the content or the implications of their respective oeuvre. 21 Compare the obligation relative to the Great Panathenaea attested by ML 40.2-4, 46.41-3 (dated probably to 447/6, see Hill and Meritt 1944), 69.55-9 (425/4), and | Ar. Nub. 386a.R-� Holwerda1, with Meiggs 1972: 166f., 292-4. There is no need to suppose that any similar obligation towards Delian Apollo must have ceased, officially or otherwise, with the transfer of the treasury to Athens in 455/4 (D.S. 12.38.2, Plu. Per. 12.1), pace Kowalzig 2007: 110f.: Thuc. 3.104.6 says explicitly that Zg[g�] Ö _Wpí �W[�Q were still being sent before the penteteric Delia were (re)instated in 426/5, and that event will have strengthened, if anything, the impulse to take part in the cult, cp. the celebrated magnificence of the Athenian theoria led by Nicias as described by Plu. Nic. 3.5-7 (noting the plural R� c¦VWb] at 5). See Meiggs 1972: 300-2, Parker 1996: 150f., Wilson 2000: 44-6, Rutherford 2004: 82-9, Kowalzig 110-18. 22 See Prost 2001: 253-60, Rhodes and Osborne 2003: 18f., Chankowski 2008: 169-74, 407f. (with a new supplement).
46
destination (which is not impossible, but at least counter-intuitive), then the years from 411
to 404 appear to be the least favourable for a choral theoria of this description.
So we have two compositions which Pausanias terms ìprosodionî. The question
arises whether this denomination is original, that is, whether these songs were associated
with this term already when they were composed and first performed. There is no obvious
reason to think otherwise: the use of c[g]¦kbgQ around the second half of the fifth century
suits well its attestation in Aristophanes (although he may have used it in a different sense,
see below), and there is no evidence that Pronomusí poetic production underwent scholarly
(re-)classification in the Hellenistic age, even though later reinterpretation of the earlier
evidence remains a possibility. The same goes for ëEumelusí, if the prosodion does date
from the fifth century (and even more if from the fourth): Pausaniasí references at 4.4.1
and 5.19.10 intimate the existence of critical activity on it, which is all but natural for a
work attributed to one of the great archaic epicists, but in the light of the other evidence it
may well have been labelled a prosodion upon its composition. Thus, there is a fair chance
that the term ëprosodioní was attached to at least two choral songs for theoriai to Delos by
the second half of the fifth century.
ARISTOPHANES
The earliest documented occurrence of the word c[g]¦kbgQ dates from the City Dionysia
of 414, when Aristophanes produced his Birds. The relevant scene is at the very centre of
the play. Upon Pisetaerus and Euelpidesí advice, the decision has been taken to found
Cloudcuckooland, and the appropriate name has been devised. Pisetaerus then dispatches
Euelpides to supervise the building of the walls and, as anticipated at ll. 810f., sets out to
47
perform the necessary rites, namely a sacrifice and a procession; the Chorus voices its
agreement. The text runs as follows (848-58):
HÝÞCÝÁoÞßaC �Sà kí µQR p�]T fg~]b jRbQg~]bQ pWg~] fÙQ �W[YR cY_°gQfR fUQ cg_cUQ jRV�. cR~ cR~, fÙ jRQgtQ RÒ[W]pW jRÚ fUQ ZW[Q^�R. 850
maßaC ¡_g[[gp�, ]PQpYVT, ]P_cR[RbQY]R] ¯ZT c[g]¦kbR _WSnVR ]W_QÏ c[g]bYQRb pWg~]bQ, â_R kÓ c[g]Yfb Zn[bfg] ãQWjR 855 c[g�nfb¦Q fb p�WbQ. ÒfT ÒfT ÒfT kÓ HPpbÏ] �gn, ]PQRPVW^fT kÓ mR~[b] ‹bk¬b.
pace N.G. Wilson 2007 p. I viii, emendationes quas Blaydes anno 1882 in luce protulit non referuntur || 848 jRbQg~]bQ R2A5EL : jRbQg~]b R1VMU || 851-8 sacerdoti trib. Wieseler 1843 pp. 108sq. || 853 c[g]¦kbR : T sscr. 5 || 854 ]W_QÏ bis U5 || 856 c[g�nfb¦Q Bentley: c[¦�Rf¦Q codd. : c[¦�RfgQ ãQ Wieseler 1843 p. 109 | fb] p�Wb V || 857 �gÏ f�b pW�b codd. : f�b pW�b secl. Dindorf 1835 p. I 347, def. Fraenkel 1962 pp. 96sq. : ÒfT (om. U) et f�b secl. Bentley || 858 ]PQRPVW^fT Hermann in schedis : ]PQRbkYfT codd. : ÕcRPVW^fT Meineke | ‹bk¬b Hermann : om. L : ‹bknQ cett.
The ancient scholia gloss c[g]¦kbR with a definition of prosodia as lyric
compositions for public worship, based on the wordís derivation from c[¦]gkg] (P14).
This interpretation is commonly accepted by modern commentators: the Chorus agrees to
honour the gods with ìgrand solemn processionalsî and furthermore to seek their favour
by sacrificing ñ ìa little sheepî, with assured bathetic effect.
Contrary to this, Simonetta Grandolini, relying on the version of the scholion
presented by manuscript R (and followed by Tzetzes, P16), contends that in our passage
c[g]¦kbgQ is not the usual substantivated adjective implying _YVg] (or the like), but
equivalent in meaning to c[¦]gkg].23 This cannot be accepted without qualification: the
23 Grandolini 1986: 264; Rís text is printed in the apparatus to P14 below.
48
reading of R is corrupt, as we now know more fully from Holwerdaís edition.24 The
purported equivalence thus lacks genuine ancient support, unless c[g]gk^gb] in P12 is
quite arbitrarily given the same meaning. Taken in its own right, however, her
interpretation has something to be said for it. The emphatic expressions of agreement of
851f. suggest a repetition of the concepts expressed by Pisetaerus, namely the cg_cq of
849, rather than a new one, however consequential from these it may have been. Arguably,
also _WSnVR is better suited to processions than to procession-songs. From the syntactical
point of view, Aristophanes may be ìboldly extending, in the lyric manner, the fig. etym.
c[¦]gkgQ c[g]Y[Zg_Rb, make a procession-approach, to the adjectival substantive
c[g]¦kbgQ sc. _YVg], processional hymnî (Dunbar),25 but as an internal object with a verb
of going, a procession-song sits rather less comfortably than a procession. It is also
possible to follow Blaydes in relating c[g]bYQRb to c[g]^\_b instead of c[¦]Wb_b (-å- is
occasionally attested),26 but the active of this verb is infrequent, and refers more usually to
motion allowed than motion caused: not an exceedingly attractive solution. Reading
c[g]¦kbR _WSnVR ]W_QÏ as subject of the infinitive may be less awkward than an internal
accusative, but jars with the seemingly parallel c[g�nfb¦Q fb p�WbQ to a very considerable
extent. In short, there is a constellation of small problems with the communis opinio.
There is of course a major difficulty with Grandoliniís hypothesis too, in that the
meaning she posits, even if arguably better suited to both syntax and context, is otherwise
unattested. Assuming Hermannís emendation, the mention of an ‹bkn at 859 is not
decisive, since it can refer to the ìPythian shoutî of 858 no less than to the c[g]¦kbR if so
much is required of it. If one takes the word in its common usage and accepts that the over-
24 See discussion at p. 86 n. 33 below. 25 1995: 505. 26 Blaydes 1883: 322 (accepted by Zanetto 1987: 251), see Dunbar 1995: 506.
49
enthusiastic Birds are stretching their syntax somewhat by suggesting to ìapproach the
gods in terms of grand solemn processionalsî, one notes the simultaneous presence of
prosodia and a HPpbÏ] �gn, an expression that probably refers to a paean-cry, as the
scholia suggest (875b.R Holwerda2). (We will see more instances of overlap between the
spheres of paean and prosodion as our inquiry proceeds, and discuss them fully in Chapter
6.) However, I remain rather reluctant to deny plausibility to Grandoliniís proposal.
FOURTH-CENTURY SCHOLARSHIP
The next testimony comes from the fourth-century Platonist Heraclides of Pontus, whose
Collection of Distinguished Composers, cited at length in the pseudo-Plutarchean dialogue
On Music, surveyed the alleged protoi heuretai of poetic and musical genres (fr. 157
Wehrli). For prosodia, this function is fulfilled by Clonas: ¡_g^T] kÓ ÁW[cnQk[Tb
æVgQ¬Q, fÙQ c[�fgQ ]P]f\]n_WQgQ fgˆ] RÃVTbkbjgˆ] Q¦_gP] jRÚ fÏ c[g]¦kbR,
�VWSW^TQ fW jRÚ �c�Q cgb\fUQ SWSgQYQRb. According to a chronology that may go back
to Glaucus of Rhegium, Clonas (a native of Tegea or of Thebes according to competing
local traditions) lived between the time of Terpander and that of Archilochus, that is to say,
sometime in the first half of the seventh century ([Plu.] Mus. 5 Lasserre1 = 1133a).27
27 On this shady figure see Abert in RE XI/1 (1921) 875f., Roesch 1989: 206f., Berlinzani 2004: 120f.; see also Paterlini 2001. Expanding on Haldane 1963: 99, it is worth noting a few connexions with his later fellow-countryman Pronomus: beside the auletic profession and an association with prosodia (however defined), according to the so-called ëSicyonian anagrapheí BNJ 550 F 2 ([Plu.] Mus. 8 Lasserre1 = 1134a-b) he invented a nomos called f[b_WVq], which included the Dorian, Phrygian, and Lydian modes within a single, apparently choral composition, while according to Paus. 9.12.5 and Ath. 14.631e Pronomus was the first to construct an aulos that allowed him to play in all three modes on a single instrument. Since neither his Theban origin nor his claim to inventing the nomos trimeles was undisputed, one wonders if, by the late fifth century or perhaps earlier, a Theban school of auletics may have had a stake in claiming him as a founding father of sorts (similarly West 1992: 333); a similar reasoning may also lie behind Pindarís mention (in a c[gg^_bgQ according to Paus. 9.30.2) of Sacadas of Argos, the other alleged inventor of the nomos trimeles (fr. 269 Sn.-M.).
50
What is possibly most significant for our present inquiry in pseudo-Plutarchís
sentence is the definite article: in Heraclidesí view, prosodia must have a distinct and
recognisable identity as a genre of sorts. Unfortunately, however, we are told nothing on
what this identity may have been. Their mention in the same breath as aulodic nomes and
the ascription of their ìestablishingî to a celebrated aulode (it is always in connexion to the
aulodic nome that Clonas is mentioned in the treatise) continues a string of associations
with the aulos that may be first found in the Birds and will later be canonised by Didymus
Chalcenterus into a defining feature of the genre. Noting the parallel that Heraclides draws
between Clonas and Terpander, François Lasserre maintained that ìthe prosodion is to the
aulodic nome what the proem is to the citharodic nomeî,28 but nothing in the text warrants
such a conclusion, or indeed any specific connexion between prosodia and aulodic nomes
beyond Clonas and his musical speciality.29 Given this dearth of information, the
relationship (if any) between Clonasí prosodia and the cultic, choral prosodia to which we
are accustomed from the later lyric tradition remains obscure.
In a discussion of the Dorian mode found in his treatise On Music (fr. 84 Wehrli), as
we know once again from its pseudo-Plutarchean namesake, Heraclidesí younger
contemporary Aristoxenus of Tarentum claimed that Plato surely was aware (gÃj †SQ¦Wb)
of the existence of partheneia composed by Alcman, Pindar, Simonides, and Bacchylides
in the Dorian mode, �VVÏ _UQ jRÚ ¯fb c[g]¦kbR jRÚ cRb¬QW]. It is unclear whether he
meant prosodia and paeans to be associated with all four poets, with only some of them, or
28 Lasserre1 pp. 154f., endorsed by Grandolini 1987-8: 33. 29 Nonetheless, given that Clonas was credited with being an �VWSW^TQ Ö cgb\fqQ (cp. the aulodic nomos called ¯VWSgb), one wonders whether Callimachus was up to something when he chose elegiacs for his ostentatiously processional (or indeed pre-processional) fifth Hymn. A similar argument is being made by Bowie forthcoming with reference to Sacadas of Argos, another ëelegistí according to [Plut.] Mus. 9 Lasserre1 = 1134c, who would not be so strictly connected to prosodia (as far as our very thin evidence goes) but who would presumably make more sense in the dramatic location of the Hymn.
51
with none, an addition to cR[pWQW~R alone; we have independent, albeit later, evidence of
prosodia by Pindar, Simonides, and Bacchylides (see below), but not by Alcman. The only
element that can be extrapolated with certainty from his testimony is that some
compositions he regarded as prosodia were composed in the Dorian mode;30 I doubt
whether it actually tells us anything about Platoís awareness of the term or the concept.
A ROYAL PROSODION
A further, interesting testimony is transmitted by Athenaeus (6.252f-253d) from
Demochares of Leuconoeís account of the Athenian history of his time.31 The subject
under discussion is the Atheniansí notorious habit of flattery towards those in power. The
two prime examples that Athenaeus draws from Demochares are first the cultic honours,
including paean-singing, that they paid to various associates of Demetrius Poliorcetes
during the period of his rule in Athens (BNJ 75 F 8, from book 20), and then the cRb¬QW]
jRÚ c[g]¦kbR ñ Athenaeusí phrase, not part of a quotation ñ which they sang to Demetrius
himself: in either 291 or 290, upon his return to Athens from Leucas and Corcyra,
Demochares reports, the populace welcomed him not only with incense and garlands and
libations, but also ñ in the reading of both the codex unicus A and the Epitome ñ c[g]¦kbR
jRÚ Zg[gÚ jRÚ ∞p�dRVVgb with song and dance (BNJ 75 F 9, from book 21).32 By
30 Grandolini 1987-8: 46, 1991: 127f. contends that all prosodia were composed in the Dorian mode and, accordingly, had those qualities of grandiosity, dignity, and stateliness that were associated with it. This is something of a stretch from Aristoxenusí text, which makes no such generalisation. 31 On Demochares see Dimitriev in BNJ s.v., Marasco 1984 (edition and commentary pp. 129-221), FGrHist II/A 133-6, II/B 114f. 32 Modern editors usually accept Bernhardyís c[g]gkbRjg^ for c[g]¦kbR jR^ (1845: 458), which restricts the use of c[g]¦kbgQ to Athenaeusí citation while preserving a clear (if indirect) testimony of the currency of the term in Athens at the turn of the third century. However, the transmitted text seems unobjectionable
52
mentioning prosodia and other forms of singing, Demochares underlines the explicit
divinization of the sovereign, as he makes clear with his brief summary of the Atheniansí
over-flattering ∞p�dRVVg] (which Athenaeus then quotes verbatim from Duris of Samos,
BNJ 76 F 13) and its exaltation of Demetrius above ìthe other godsî.33 This is the only
certain attestation of a prosodion sung to honour a human being, ìpart of a pattern of
divine forms being applied to humans in the Hellenistic periodî (Rutherford).34
ALEXANDRIAN SCHOLARSHIP AND THE EDITIONS OF THE LYRICISTS
The next landmark in our survey of the word c[g]¦kbgQ is Alexandrian scholarly activity
on Pindar and Bacchylides, which resulted in the standard editions from which both poets
are cited by (so far as one can see) all later writers.35 As we have seen, two books of
Prosodia were ascribed to Pindar; Bacchylides had at least one, of which only three
fragments, probably from two poems, are preserved by Stobaeus (4.44.16, 4.44.46, 4.34.24
Wachsmuth-Hense = frr. 11-13 M.).36 The existence of a book of Prosodia by Simonides
alongside his Hymns and Paeans is a probable inference from a scholion to Aristophanesí
Birds, 919c Holwerda2; no other source mentions it, and it does not seem possible to
attribute any known fragments to it.
(Demochares is very fond of tricola, as a cursory look at his fragments will show), and Bernhardyís emendation brings no visible improvement. 33 See also Muth in RE XXIII/1 (1957) 862, Marasco 1984: 92-5, Grandolini 1987-8: 37, 1991: 129, Asmonti 2004: 36-9. On the worship of Demetrius (and of his father, Antigonus the One-Eyed) in Athens see Mikalson 1998: 75-104 with bibliography; on the ithyphallos see Palumbo Stracca 2000 and Chaniotis 2011, with bibliography. 34 2003: 714 n. 6. 35 By ëeditioní I mean the classification of poems into books and the ordering of the poems within such books; the order of the books was probably not fixed unless they were numbered sequentially (Race 1987). 36 Frr. 11-12 M. belong to one poem, as responsion between frr. 11.3-5 and 12 suggests: Neue 1822, quod non vidi. All three fragments are commented in Maehler 1997: 311-14, frr. 11-12 in 2004: 234-7.
53
The organisation of Pindarís work into books is traditionally ascribed to
Aristophanes of Byzantium (fr. 381 Slater) after the Vita Vaticanaís statement that
Olympian 1 c[gfYfRjfRb ÕcÙ �[b]fgdnQgP] fgt ]PQfnXRQfg] fÏ HbQkR[bjn (p. 7
Dr.), supported by Edgar Lobelís unavoidable supplement at T1 lines 35f.; this attribution
has been called into question, but without much justification on the available evidence.37
In Bacchylidesí case, the only piece of evidence is a fragmentary papyrus
commentary (B col. i.9-19), whose exact implications in this respect are conjectural, as
indeed is its very attribution to Bacchylides.38 According to the anonymous commentator,
Aristarchus said that the ode in question was kbp]P[R_�bjqQ because it treated the story of
Cassandra, and titled it Cassandra accordingly, criticizing Callimachus who had wrongly
classed it (jRfRfnXRb) in the Paeans through failing to understand that the �c^dp]�_R
ñ the refrain ∞q, which this ode evidently had ñ also occurs in dithyrambs.39 The reference
37 Slater ad loc. strongly opposes the identification (and appears to suggest that the arrangement of Pindarís poems was due to the poet himself), but see DíAlessio 1997: 51-5, Negri 2004: 16-27. On his observation that O. 1 ³[b]fgQ _ÓQ —kT[ contains the complete vowel spectrum (but so do P. 1 Z[P]YR d¦[_bSX, as Negri 2004: 23 n. 1 remarks, and O. 2 �QRXbd¦[_bSSW] —_Qgb), ìwhy would Pindar be so interested in beginning with a vowel spectrum?î (Strauss Clay 2011: 339 n. 9); Aristophanesí views on the genre of P. 2 did not need quoting in inscr. p. II 31 Dr. simply because they were held to be obvious from the fact that he had edited it as a Pythian (Negri 2004: 20f.). Indirect confirmation that Aristophanes was associated in antiquity with establishing the standard text of the lyricists arguably comes from D.H. Comp. 22.17, 26.14 (fr. 380 B Slater): while Slater ad loc. and Tessier 1995: 22-4 must be right to the extent that Dionysius is merely conjecturing that it was Aristophanes who articulated the text of Pindar and Simonides into cola, these two passages nonetheless testify that when thinking of a foundational intervention on the text of these poets, it was Aristophanes that would first come to mind (that he is merely ìproverbialî to indicate ìthe philologists of the Museumî, as Tessier 24 n. 35 contends, I do not find particularly attractive). This is easily explained if the Alexandrian Pindar, which carried this colometry, went under his name. On Aristophanes as a Pindarist see also Irigoin 1952: 35-44; on Wilamowitzís alternative hypothesis that the surviving classification may go back to Apollonius ¡ W∞kgS[ndg] (1922: 108, accepted by FÅrber 1936 I 19) see Gallo 1968: 37f. 38 SH 293 = CGLP I.I/4 Bacchylides 4 (Maehler), with bibliography; first published in Lobel 1956b. The author of the commentary is unknown; Snell 19587: 50* suggests Didymus. 39 At 18, Lobelís supplement �c^dp]é��_R (papyrologically and orthographically legitimate, as demonstrated by Ucciardello 1996-97: 79-82 and 2000 respectively) must be correct. The word must indicate something which misled Callimachus into regarding the ode as a paean, but which Aristarchus pointed out could also be found in dithyrambs; that is, we need something that is most naturally associated with paeans but can, upon reflection (note ]PQYQfR, 17), be also associated with dithyrambs. The paeanic �c^dpWS(S)_R fits this description (for its presence in a Dithyramb see fr. **60.37 M. with DíAlessio 2000b), but Luppeís cR[Rkbq]S\_R and _�pgP kbq]S\_R (1987, 1989) do not: a narrative emphasis was a well-known defining quality of dithyrambs (see Pl. Rep. 3.394b-c, [Plu.] Mus. 1134e with KÅppel and Kannicht 1988: 22);
54
to titling (�cbS[ndWb, i.13) suggests that Aristarchus compiled an edition, which is a more
natural place than a commentary to propose a poem-title, although col. i.9-19 implies that
he also offered interpretative material, most probably in an accompanying hypomnema; the
fact that no such title precedes the quotation of the incipit which prefaces the discussion in
B is no evidence that he only composed a commentary, since the treatment of titles in
hypomnematic papyri does not seem to have been very consistent.40 As for Callimachus,
jRfRfnXRb (i.15) is compatible in principle with an edition, but is equally compatible with
his cataloguing work in the Pinakes, and economy favours the latter over an editorial
activity of which the existing evidence shows no trace.41 Thus the natural inference is that
the ëstandardí Alexandrian edition of Bacchylides was the work of Aristarchus, although
conclusive evidence is lacking.42
It is unknown how far each of these editions relied on earlier collections or schemes
of classification. Callimachusí Pinakes classified Pindarís odes book by book, not only
genre by genre, as is probably indicated by his understanding of Pythian 2 as a Nemean ñ
that is, one of the ëextraí epinicians appended to the ërealí Nemeans (fr. 450 Pf.); however,
there is no clear evidence of how far back this model or his specific eidographic choices
went, or of what relationship there was between those of Aristophanes and his. Earlier,
Callimachus can hardly have been unaware of this, and even if he had been, gà ]PQYQfR would not have been an apposite expression for the commentator to use. 40 H25, the only papyrus hypomnema to Pindar to preserve the beginnings of sections relating to individual odes, frequently heads them with the relevant titles (frr. B 2(a).1, 14.i. 2, 27 (with DíAlessio 2012: 40, 48-51), 17.6, possibly 4.ii.4f. with Lobel 1961h: 162), but sometimes not (fr. 1.i.9). The scarcity of other evidence prevents one from assessing how typical this manuscript may have been. 41 Ascription to the Pinakes is favoured by SchrNder 1999a: 111 n. 1. On this work see at least Pfeiffer 1968: 126-34, Blum 1977. 42 One does think of inscr. P. 2 p. II 31 Dr., which cites the eidographic views of six scholars, none of whom is the actual classifier, Aristophanes of Byzantium (see n. 37 above). In our case, however, conjecturing two editions, one by Aristophanes and one by Aristarchus, might be rather too much for Bacchylides, who does not seem to have been a heavy-weight in Classical and Hellenistic criticism.
55
Chamaeleon of Heraclea had called Olympian 13 an �Sji_bgQ and fr. 122 Sn.-M. a
]j¦VbgQ (fr. 31 Wehrli), but it is doubtful whether these generic labels reflect an actual
classification into books: both words occur in the texts themselves (ll. 29 and 14
respectively), which is rather suspicious. Add Aristoxenusí citation of fr. 125 Sn.-M. as a
]j¦VbgQ (fr. 99 Wehrli: here too the word may have occurred in the text, cp. [Plut.] Mus.
28 Lasserre1 = 1140f), and you have the only extant testimonies on pre-Aristophanic
Pindaric eidography. Unfortunately, none sheds any light on the Prosodia or on what role
they may have played in articulating Pindarís poetry before Aristophanesí edition,
although Aristoxenus fr. 82 Wehrli suggests that they did indeed play some role.
The generic evidence that can be garnered from the fragments of Pindarís Prosodia
will be examined in detail in Chapter 6. As for Bacchylides, the sole sourceís bias towards
sententious statements precludes forming even an impression of the specific characteristics
of his Prosodia and the classificatory principles that shaped the collection. Faced with
these maxims on the happiness and toil of mortals, a radical sceptic could legitimately
doubt even the religious nature of these works.43
EPIGRAPHIC TESTIMONIES
The greatest amount of evidence for compositions termed prosodia is epigraphic, covering
different locales in mainland Greece and the islands and extending in time from the mid-
fourth century to well into the Imperial era. On the one hand, this presents us with a more
historically reliable terminology: no inscription relates to anything other than a
contemporary (or almost contemporary) performance, thus eliminating the risk of generic
43 Grandolini 1987-8: connects frr. 11-12 M. to the ìDelphian teachingî that invites to moderation and sees a similarity with compositions concerned with Apollo such as Pae. 4 and Parth. 2.
56
back-labelling of earlier texts according to later concepts. Some of these inscriptions also
give us information on the performance context that we do not have elsewhere, such as
names of performers, overviews of the programme as a part of which prosodia were
performed, and in two remarkable cases (CID III 1-2) an almost complete text. On the other
hand, what the inscriptions call prosodia are just as diverse a bunch as those we have seen
in the literature so far, although certain trends ñ some of which perpetuate aspects that we
have already noted in this chapter ñ can be seen clearly.
Perhaps shortly after 341/0, the city of Eretria voted to institute an �SàQ _gP]bjv] to
accompany the existing festival of the Artemisia.44 According to the relevant decree, all
those who took part in the musical competition were to �STQ^�W]pRb c[g]¦kbgQ fWêb pP]^Wb
�Q fWêb RÃVWêb with the whole attire that they would wear in the agon (IG XII/9 189.12-14 =
LSCG 92). What sort of business this was is unclear. �STQ^�W]pRb seemingly suggests a
competition, with each of the contestants presenting his own prosodion, but several
elements tell against this interpretation: (i) the text distinguishes this ceremony from the
agon (13f.), though it does not specify whether it was to take place before or after it;45 (ii)
it would be odd to expect artists with such diverse specialities as singers, instrumental
musicians, and rhapsodes to compose one prosodion each and perform it in competition
with one another; (iii) the inscription lists prizes for winners and runners-up in each
category (14-20), but makes no mention of a prize for the prosodion.46 Thus, since there is
occasional evidence that �STQ^�g_Rb can refer to non-competitive performance, the
passage is probably best understood as mandating the artists to perform the prosodion
44 Date: Ziehen 1906: 253f. On the location of the Artemision see Knoepfler in Wagner and Theurillat (eds.) 2004: 296f. 45 Jory 1967: 86 assumes that the performance took place after the contest. One might disagree, especially since a proagon is mentioned at l. 22: see Preuner 1903: 342. 46 Jory 1967: 86 (after Mie 1909: 13 but rightly reversing his conclusions).
57
together (which better explains 12 cnQfR[]]) outside the contest proper.47 A point to note
is that, although a procession was also to take place before the sacrifice and all the
competitors were likewise required to take part in it (35-40), the prosodion was to be sung
in the aule during the sacrifice, not during the procession itself.
A Delphic decree datable to the third quarter of the third century (F.Delphes III/2 78)
honours Cleochares of Athens, cgb\fU] _WV�Q, for composing cgpıkbıQ fW jRÚ cRb¬QR
jRÚ —_QgQ for the god (l. 3) to be performed by a chorus of boys ìat the sacrifice of the
Theoxeniaî:48 the same festival at which Pindarís sixth Paean had also been performed
(cp. 60f.). A few lines further down in the inscription, f¦ fW cgp¦kbgQ jRÚ fÙ_ cRb¬QR
jRÚ fÙQ —_QgQ (5f.) indicates that what Cleochares had written was regarded as three
distinct compositions, not a unity taken to be a prosodion, a paean, and a hymn all at
once.49 Along similar lines, an Athenian decree from Delos dated to 165/4 (I.D�los 1497)
honours Amphicles of Rheneia, _gP]bjÙ] jRÚ _WV�Q cg\fq], who wrote a c[g]ıkbgQ Ö
�__WVÔ] ìto the city [i.e. Athens] and hymned the gods who inhabit the island and the
people of the Atheniansî while also acting as a didaskalos for the chorus of boys who
would sing his composition to the lyre.50
47 Jory 1967: 86; Rutherford 2003: 714 poses the alternative between competitive and joint performance but expresses no preference. On �SiQ and �STQ^�g_Rb said of non-competitive performance see E. Reisch in RE I (1891) 856 s.v. Agones, Foucart 1873: 61, add e.g. Pl. Mx. 235d. 48 Stephanis 1454. Grandolini 1987-8: 38 argues that the pieces had been commissioned by Cleocharesí tribe; the text of the decree says nothing to this effect (the syntax of 3f. rather suggests that they were composed in Delphi), and the provisions for reperformance at 5f. are certainly local. 49 SchrNder 1999b: 75 n. 68. 50 Upper part published by Homolle 1886: 35f., lower part added by Fougyres 1889: 244-50, which remains its fullest discussion to date. Amphicles (Stephanis 145) is also known from another honorific decree issued by Oropus (EOropou 211), which styles him as a Delian and does not mention any specific composition of his. See DíAlessio 2009a: 147f., who compares the formulation of the slightly later I.D�los 1506, honouring the epic poet Ariston of Phocaea.
58
ATHENAEUS AND LIMENIUS
The next two items on our list are the most substantial testimonies of prosodia after
Pindarís time, as well as the only ones (beside ëEumelusí, if so dated) whose text survives
in any bulk. These are two inscriptions that were set up, like the decree honouring
Cleochares, on the wall of the Treasury of the Athenians in Delphi, and originally carried
the entire score of two songs called ìpaean and prosodionî composed respectively by
Athenaeus and Limenius (CID III 1-2 = CA pp. 141-59), both members of the Athenian
guild of the }bgQ�]gP fWZQ~fRb, for two celebrations of the Pythaís from Athens to Delphi,
either those of 138/7 and 128/7 or those of 128/7 and 106/5.51 These are the pieces of
ancient Greek non-dramatic lyric about which we know by far the most: where they were
first performed, at which festival, more or less at which date, and by whom, not to mention
the musical notation (vocal for Athenaeus, instrumental for Limenius).
In both cases the pieces are introduced by a laconic one-line heading that indicates
their genre and author. The indication of the genre actually survives (and not whole) only
for Limeniusí poem: cR]�ÏQ kÓ jRÚ c[[g]¦]ÉbgQ W∞] f[ÙQ pWÙQ √ �c¦]î]W[Q jRÚ
c[g]Wjbpn[b]W]Q Êb_qQb[g] r]×^QgP. The extant part ñ the end ñ of the heading of
Athenaeusí poem follows an almost identical pattern: W]∞] fÙQ pWÙQ √ �[c¦\]WQ
�p]qQRbg]. (Unlike Limenius, Athenaeus was not a citharist, whence presumably the
missing second verb).52 Egbert PNhlmann fills the initial lacuna with cRbÏQ jRÚ
Õc¦[Z\_R, another genre linked to Apollo in some ancient sources, but Petronilla Moens
51 Athenaeusí name had long been thought to be an ethnic, and the composition accordingly regarded as anonymous, until BClis 1988: 210-12 re-calculated the width of the lacuna and recognised the correspondence with the list of technitai F.Delphes III/2 47.19 (�pqQRbg] �p\QR^gP). The dates of the two compositions are disputed: see most recently BClis in CID III 133-42, SchrNder 1999b, Furley and Bremer 2001: I 129-31, Rutherford 2001: 34f. The two texts are re-edited with translation and commentary by Furley and Bremer I 129-38, II 84-100; see also BClis 1988 and her notes in CID III, Fantuzzi 2010: 192-4. 52 Limenius (Stephanis 1553) a citharist: FD III/2 47.22.
59
had suggested cRbÏQ jRÚ c[g]¦kbgQ, which takes up the same space.53 The prosodion has
strong Apolline, and specifically paeanic, associations, as we shall see in Chapter 6; if the
first word of the heading was indeed cRbnQ, the identity of the performance context as well
as of the overall formula of the heading makes a supplement parallel to CID III 2
preferable.54
The genres attached to Athenaeus and Limeniusí poems coincide with two of the
three listed in F.Delphes III/2 78, although mentioned in reverse order. But unlike
Cleocharesí three notionally distinct compositions (however connected they may have
been to each other in practice), here the implication seems to be that each ìpaean and
prosodionî was envisaged as a unity ñ perhaps something like the cRbÏQ c[g]gkbRj¦]
mentioned by the scholia to Isthmian 1 ñ more than as a sequence of two separate items:
the singular ≈ in the headings (extant in CID III 1 and thus likelier than â in the lacuna in 2)
leaves no room for doubt.55 This does not mean that no division can be seen: the traditional
view that Limeniusí poem falls neatly into a paean (1-33) and a prosodion (34-46) is
plausible, given the obvious metrical partition of the text. Nonetheless, the two parts are
fundamentally united from the viewpoint of both structure and argument. We cannot tell
whether some such division occurred in Athenaeusí paean-and-prosodion as well, since the
end is lost: there is no obvious change of metre down to l. 25 inclusive, but the strong
impression that this is the beginning of the final section may be illusory after all.56 We are
reminded of the peculiar case of Pindarís ëPaean 6í, which appears to replicate the
combination of a longer section regarded as a paean (1-122) with a shorter one regarded as
53 Moens 1930: 6. The hyporcheme as an Apolline genre: Men. Rhet. p. 331 Russell-Wilson, Io. Sardianus in Aphth. Prog. 21 p. 119 Rabe2, Luc. Salt. 16, see Rutherford 2001: 100f. 54 So also Furley and Bremer 2001: I 129. 55 DíAlessio 1997: 30 n. 39. 56 A change of metre is suggested by Moens 1930: 8.
60
a prosodion (123-63 = F5), although the metre of the two sections appears to be identical
and the interpretation of their mutual relationship is fraught with problems and makes the
use of the poem(s) as a parallel extremely difficult.57
Athenaeusí composition opens with an invocation to the Muses of Helicon, who are
to heed the singer and come to sing of Phoebus (1-3); the relative ≈] introduces a brief
sketch of the godís relation to Delphi (4-7) before the focus shifts abruptly towards Athens
and her piety, depicted at some length through the images of sacrifice, incense-burning,
and song accompanied by the reed and the kithara (8-14). A smooth transition to the
Athenian performers of the present song (15-17) brings the argument back to Apollo, now
addressed in the second person: his foundational mythic feat of killing the snake Python is
paired with his role in repelling the Gaulish invasion of 279/8 (18-24).58 Parts of three
more lines are preserved, but the train of thought cannot be followed beyond the initial
apostrophe and exclamation (25).
Limeniusí paean-and-prosodion presents some close similarities. It also opens with
an invocation to the Muses to come to Parnassus, lead off the song, and sing of Apollo (1-
4); however, what follows from ≈Q is a leisurely narrative of the godís birth in Delos, with
an Athenian twist (at 6 Leto clutches an olive branch instead of the canonical palm tree),
and his subsequent voyage to Attica, where he was welcomed by the strains of ñ once
again ñ the reed and the kithara (5-17). This episode goes on to be interpreted as an
aetiology for the practice of paean-singing by the Athenian population at large and the
Dionysiac technitai specifically (17-20). The god is subsequently called upon to come to
Parnassus (21f.); the narrative pointedly follows the same shift from Athens to Delphi and
57 Discussed at pp. 228-44 below. 58 On the invasion and the ancient sources on it see Nachtergael pp. 15-205.
61
briefly mentions Apolloís slaying of Python (23-7), Tityus (27f.), and a hissing monster
whose name or description are lost in lacuna (29f.), then, just as in Athenaeusí poem, his
defeat of the barbarian invasion, which saved the oracular seat from plunder (31-3). The
closing ëprosodioní is a prayer to Apollo, Artemis, and Leto for the sake of Athens and her
citizens, the Delphians and their progeny, the technitai themselves, and the Roman Empire
(34-47).
AGONISTIC PROSODIA
In the period between the early Hellenistic and the late Imperial ages we also see
compositions called prosodia performed in an agonistic context. Two Delphic inscriptions
(Nachtergael 4, 9) record the names of, respectively, two and three cgb\fRÚ c[g]gk^TQ
among the artists who competed at the festival of the Soteria in the second quarter of the
third century: Philon of Megalopolis and [Py]thonicus of Hermione (4.16f.), Alexion and
Dexinicus of Sicyon and Xenon of Corinth (9.13-15).59 As is shown by their discontinuous
presence in the epigraphic participant-lists, prosodia were not a constant element in the
programme, and the lack of a secure date for any of the inscriptions makes it impossible to
59 Stephanis 2561, 2177, 129, 600, 1918. Nachtergael 4 (F.Delphes III/4 356, cp. CID IV 31): upper part (1-16 = SEG II 339) first published by Pomtow 1915: 295 and re-edited by Roussel 1923: 6, lower part (17-22) joined by Bousquet 1959: 166-8. Philonís ethnic is lost here, but he is known as a Megalopolitan from another catalogue, F.Delphes III/1 478B.8 (Roussel 1923: 3 as revised by Bousquet 1959: 169), and was probably styled as an Arcadian in a fragment published by Flaceliyre 1928: 264f. n. 3, just as he might have been in F.Delphes III/4 356 itself (SEG XVIII 231: Bousquet 1959: 168f., J. Pouilloux in F.Delphes III/4 p. 24). It is unknown in which capacity he was mentioned in either of these other inscriptions, but the headings that follow his name on F.Delphes III/1 478B appear to coincide with those found on III/4 356, thus suggesting that he may have been listed among ìpoets of prosodiaî there too. If this is so, one or more of Philonís neighbours on that inscription ñ Democrines, Nicomenes, Ch[a]ricle[i]des, Onasand[r]os, Demosthenes ñ may also have composed a prosodion for that yearís Soteria. Nachtergael p. 257 conjectures that also the [E]peratus of Megalopolis (and/or, we might add, [N]eon of Stymphalus: Stephanis 851, 1800) mentioned in Nachtergael 3 = F.Delphes III/1 477.7f., cp. CID IV 42 may have been a cgb\fU] c[g]gk^TQ: the position in the list is compatible with this hypothesis, but it is unknown whether any prosodia were performed that year. On the Soteria see Nachtergael 1977: 209-495, esp. 404-34 (inscriptions on the pre-Aetolian Soteria), 241-72 (their relative chronology), 299-313 (nature and programme of the games).
62
ascertain whether their presence and absence followed a regular pattern. Johann Frei
suggests five- or ten-year intervals, but this is no more than interesting speculation: as far
as the inscriptions go, the programme in the years in which prosodia were sung was not
obviously richer than in other years, which prima facie does not suggest a special occasion.
Since sometime in the second half of the second century, compositions called
ìprosodiaî also featured prominently in the pentaeteric Musaea held under the control of
Thespiae in the sanctuary of the Muses at the foot of Mount Helicon.60 A remarkable series
of inscriptions stretching well into the Roman era, but whose exact dates (and thus relative
order) are not known with certainty, records the cgb\fU] c[g]gk^gP first in the list of each
festivalís winners. The names that survive are those of Damaius of Thessalonica, Eudemus
of Thespiae, Demetrius of Tanagra, Athenion of Thespiae, Bacchius of Athens, Eumaron
of Thespiae (twice), Antiphon of Athens, Vipsanius Philoxenus of Thespiae, and
Callitychides of Thebes (I.Thesp. IV 167, 169-72, 177-9 = Manieri Thes. 27, 29-31, 33, 42-
4).61 The role of prosodia in the festival must have been prominent, given the place their
composers hold in the victory catalogues, but this is virtually all we know about them.62
What sort of poetry all these agonistic prosodia may have been is beyond conjecture.
Their position at the head of the victor-lists of the Musaea perhaps suggests that prosodia
60 On the festival see Schachter 1981-94: II 163-79, 2010-11, and Manieri pp. 313-40, with a re-edition and commentaries of all the ancient testimonies (Thes. 1-49) at 347-423. On the cult of the Muses in the area of Mt. Helicon more broadly see Schachter 147-63. 61 Stephanis 940, 632, 74, 512, 963, 226, 2541, 1364 (Damaius does not seem to appear in Stephanisí register). Some of the poets seem to have been remarkably versatile, since Antiphon also carried the prize as comic poet and actor, while Vipsanius and Callitychides ñ joint winners, like Antiphon and Eumaron ñ also won, respectively, with the encomium of the Muses and the poem to the Emperors, and with the encomium of the Emperors and the poem to the Muses. Such versatility is far from uncommon in the Hellenistic and Roman record: see Chaniotis 1994: 93, 103-5, Le Guen 2001: II 125-30. 62 Pappadakisí tentative restoration of c[g]¦kbgQ in lacuna in a similar agonistic catalogue believed to be related to the Basileia at Lebadeia (1923: 243 = SEG III 368.7, third or second century) is unconvincing, since prosodia are not otherwise attested in that festival (rightly Manieri p. 154). The same goes for Preunerís c[g]gk^gP cgb\fq]] (1903: 339, 341f.) in EOropou 520.2 = Manieri Oro. 5.2, the victor-list of the Great Amphiaraea held at Oropus probably in 329/8 (see Knoepfler 1993, Manieri pp. 225-8).
63
performed at that festival may have been somehow concerned with the deities that
patronised it (an argument that can also be made for the collective prosodion of the
Eretrian Artemisia), but even so much is speculation. Nothing is known of their mode of
performance, or of whether the cgb\fq] necessarily took part in it. Reisch suggests that by
this stage they were EinzelvortrGge, solo performances:63 this is quite plausible, but
evidence ñ once again ñ is lacking. One notices that none of the ìpoets of a prosodionî that
were victorious at the Musaea can be positively identified as a musician:64 it may well have
been the case that, at least in the Imperial age, prosodia composed for that contest were
designed exclusively for recitation.
A DELIAN HßaChÞ}Þa´
We must now go back to a much earlier inscription, whose relevance to the subject of our
inquiry is debatable, but whose text is of some import in view of later testimonies. A
Delian list of contestants from 236 BC (IG XI/1 120.49) lists two °nVfRb, Demetrius and
Cleostratus, as victors _WfÏ fgt c[g]Tbk^gP: just so, with omega and iota.65 The form
c[g]i(b)kbgQ, which frequently occurs in Byzantine manuscripts and texts where we
would expect c[g]¦kbgQ (see Chapter 5), is often regarded as a misinterpretation due to
confusion with the much more common word c[g]Tbk^R or to Didymus Chalcenterusí
paretymology of c[g]¦kbgQ itself, which involved the verb ³bkT (P1-P11 below).
Generally speaking, this is correct. Beside an observable tendency to move from omicron
63 RE III (1897) 2438 s.v. Zg[bjgÚ �S�QW], see also Rutherford 2000a: 148. 64 This is an argumentum ex silentio, and accordingly can carry only limited weight; nonetheless, there is no compelling reason to think that Antiphon, in his capacity as comic poet and actor, had much to do with music, and the other victories of both Vipsanius and Callitychides (see n. 61 above) suggest a specialization in recited poetry and prose rather than musical composition or performance. 65 Stephanis 613, 1453. See chiefly Rutherford 2000a; on the meaning of °nVf\] see BClis 1988b: 244f.
64
to omega (and indeed to forms of c[g]Tbk^R itself) the later the text is, the papyri leave no
room for doubt that the genuine title of Pindarís Prosodia had an omicron (T5, T6); in at
least one case, the alphabetical arrangement of a work guarantees that the transmitted
c[g]ikbgQ must also conceal an original c[g]¦kbgQ (P19); and in the Vita Ambrosiana,
the scribe spotted his own mistake and corrected into c[g]gk^TQ his earlier c[g]Tk^TQ
(T2). Except this Delian inscription, the epigraphic record has c[g]¦kbgQ unanimously.
Thus, it is also likely that when the two forms occur as variants, or at different points in the
same work, the one with omicron is genuine, the one with omega derivative.
However, our inscription is well over a millennium earlier than the earliest surviving
manuscript attestations of c[g]ibkbgQ, and predates even Didymusí paretymology by
some two centuries: it would be unmethodical to dismiss our c[g]ibkbgQ as a mere slip. A
word c[g]Tbk^R meaning ìaccompanyingî or ìaccompanied songî is attested in
lexicographical sources:66 since our two musicians were victorious ìwith the prosodionî
(comitative not instrumental), Ian Rutherford is probably right to parallel our c[g]ibkbgQ
to this shady c[g]Tbk^R. What relationship there may have been between it and ëthe
prosodioní is up for grabs, not the least because we know nothing about this c[g]ibkbgQ
save that it was performed to the accompaniment of plucked-string instruments. The
context resembles that of the agonistic prosodia we have just seen, and Delos is a well-
known hotspot for prosodia (ëEumelusí, Pronomus, Amphicles, probably F11, possibly F1
and F15). Rutherford suggests that the term c[g]ibkbgQ may have arisen in response to the
agonistic, non-processional prosodia, which had become quite disconnected from the
66 With the kithara, Critias VS 88 B 67 in Poll. 4.64 Bethe; more generally _Wfí ¿[SnQgP, Hsch. c 3946 Hansen, Phot. c 1390 Theodoridis, and by implication |L D.T. p. 474 Hilgard. Another, possibly related occurrence is Pl. Rep. 399a fUQ Ú[_gQ^RQ Ä ¯Q fW cgVW_bjvb c[nXWb ƒQfg] �Qk[W^gP jRÚ �Q cn]\b �bR^Tb �[SR]^Rb c[Wc¦QfT] ðQ _b_q]Rbfg dp¦SSgP] fW jRÚ c[g]Tbk^R]. On this use of c[g]Tbk^R see also Ciancaglini 1999, although her argument that the literal meaning of the word must have been ìwhat is added to songî or ìwhat concerns songî (p. 126) I do not find very compelling.
65
origin of their name: ì[p]erhaps a feeling arose that ëc[g]ibkbgQí was a more appropriate
term for what had become essentially a quasi-musical genre.î67 However, this feeling
cannot have been very widely felt for very long, since the lead of our inscription does not
seem to have been followed. Moreover, in respect to the supposed greater appropriateness
of the label ìaccompanying songî to agonistic prosodia generally, one notes that no other
agonistic catalogue mentions musical accompaniment for prosodia: it is only the poet who
receives the prize. In comparison to those, the term c[g]ibkbgQ is much more at home in
IG XI/1 120, where the victors are the musicians themselves, and the song along to which
they play can rightly be called an ìaccompanying (or ìaccompaniedî) songî.
PHILO OF ALEXANDRIA
The inscriptional record has now been dealt with, but one further literary testimony
remains to be discussed. In his description of the Jewish congregation of the pW[RcWPfR^
that constitutes his essay On the Contemplative Life, Philo twice mentions prosodia among
the song that they sing before and after their meals (80, 84). In the second occurrence, fÏ
c[g]¦kbR are opposed to fÏ ]fn]b_R: interpretation as ìprocessional and stationary
hymnsî (Conybeare) is unavoidable.68 In the first, Philo gives a whole list of song-types
allegedly represented among the Jewish anthems of ìthe poets of oldî: �c�Q f[b_Yf[TQ
c[g]gk^TQ —_QTQ cR[R]cgQkW^TQ cR[R�T_^TQ ]fR]^_TQ Zg[bj�Q ]f[gdR~]
67 Rutherford 2000a: 148. 68 1895: 245; on these passages see also Grandolini 1987-8: 43-5. The Armenian version of the essay (Conybeare 1895: 178) has զառ ի յԱստուած երգն ìthe songs to Godî and զչափաւորսն ìthe measuredî, ìmetrical (ones)î respectively; Conybeare 1895: 128 claims that, from the former, leuissima si fiat correctio sensus elici poterit c[g]¦kbR, but does not explain what emendation he means. The text of the mss. may well be genuine (cp. the identical implication of P19-P21), but contrast the translation of 80 quoted in the next note; here, it is striking that also ]fn]b_R is translated in a way that destroys the opposition (contrast once again the literal rendition կայականաց at 80).
66
cgVP]f[¦dgb] W“ kbR_W_Wf[\_YQTQ.69 Here we have a monodic performance, with the
assembly joining in whenever refrains (�j[gfWVW�fbR jRÚ �d�_QbR) are to be sung; at 84,
prosodia and stasima are sung by two choruses, either antiphonally or in unison. In the
hodge-podge of Greek verses and poems of 80, some are more, some less well-known
(cR[R]cgQkW~gQ is a hapax, cR[R�i_bgQ attested as a kind of song only in I.Teos 45.8,
11), and the underlying rationale is not easy to see. Moreover, Philoís terminology should
not be taken too literally. While one may imagine that the Therapeutae took over bits and
pieces of the Greek poetic tradition for their own cult songs, Daumas and Miquel remark
that in the Jewish Antiquities Josephus ascribes to King David ìodes in trimeters and
pentametersî (7.12.3), which must be the Psalms, and Simonetta Grandolini adds that in
the same work Moses is twice credited with hexameters (2.16.4, 4.8.44);70 nothing quite
like the trimeters, pentameters, and hexameters that we know from Greece. In our case too
there need not be any deeper connexion between the songs of the Therapeutae and the
traditions of Greek poetry than the names themselves, nor indeed any older a tradition for
this terminological connexion than Philo himself.71
69 I give no punctuation so as to avoid the ultimately unanswerable question of which (if any) of these terms goes with another: —_Qg] is a good candidate (with either c[g]gk^TQ ñ already in PQ ñ or cR[R]cgQkW^TQ, see Conybeare 1895: 122), but each word between �c�Q and Zg[bj�Q can, at least in principle, stand alone. The Armenian translation (Conybeare 1895: 177) takes c[g]gk^TQ —_QTQ together ñ տաղից, եռաչափաց,
առճանապարհական երգոց, փրկաւետաց, նուիրականաց, յարաբագնաց, չափականաց,
կայականաց, պարաւորաց ìverses, trimeters, songs for the way, salvific (songs), consecrated (songs), (songs) by the altar, metrical (songs), standing (songs), dancing (songs)î ñ but this tells us nothing more than the translatorís own interpretation and (possibly) the punctuation he found in his antigraph. Note how the translator rendered c[g]¦kbgQ literally by fabricating a hapax առճանապարհական (առ c[¦], ճանապարհ ¡k¦], -ական adjective ending: Ciakciak 1837: 202 ìcanzone di rogazioneî interprets rather than translates): he must have read the form with omicron, not that with omega (which occurs already in M, cp. BOPQ at 84), and parsed it correctly. 70 Daumas and Miquel 1963: 140 n. 1, Grandolini 1987-8: 43f. 71 Grandolini 1987-8: 44 is right that it would be wrong to attribute to Philo what one reads in Josephus, but one finds little evidence for her claim of a ìtendency of new Jewish poets to fasten onto Greek poetry in its formal aspectî in order to bridge the gap between Jewish and Alexandrian literature: Philoís �[ZR~¦Q fbQR f�Q cnVRb cgb\f�Q (80) does not prima facie support this as much as a reinterpretation in Greek terms of a traditional Jewish heritage. But the evidence is too limited to allow a firm conclusion.
67
(5)
The genre, III : Indirect tradition
We now come to the second stage of our enquiry: the views of ancient scholars on ëthe
prosodioní, its nature and characteristics; what we have termed its indirect tradition.
Chronologically they follow quite neatly from the testimonies we have examined in the
previous chapter, starting with the work of Didymus Chalcenterus, whose lifetime
straddled the collapse of the Roman Republic and the early decades of the Empire. From
his time onward, almost all we hear about prosodia is citation of older texts and various
degrees of critical reflection on the ëgenreí: only Philo, who somewhat stands out from our
record, as we have noted, and the agonistic catalogues from Thespiae tell of prosodia
performed in their own time. This chapter depicts a different realm altogether: the
information we shall see becomes more and more bookish, second- and third-hand, as
actual contact with its subject-matter gradually disappears. No surviving manuscript of
Pindarís Prosodia is later than the third century AD; Porphyryís citation of F14 (T9) ñ
assuming that he was quoting first-hand, which may well not be the case ñ does not carry
us much further in time; how much longer their text may have survived is only a matter for
speculation. Of the sources that we shall see in the next few pages, Didymus and the
Aristophanes scholiast surely knew the Prosodia directly, as perhaps did ëHeliodorusí and
Proclus, but the opposite seems true of the Byzantine texts that form the bulk of the
primary material in this chapter.1
1 See West 2013: 9f. (on Proclus and Didymus).
68
These testimonia are presented in a roughly thematic arrangement and, subordinate
to this, in chronological order: the aim is to follow the development of different critical
traditions on the prosodion in as historically conscientious a fashion as the evidence
allows, according to the principles set out in Chapter 3. Each text is followed by as
complete an apparatus as I could compile, although it should be noted that I have been able
to refer to reproductions of the original manuscripts only in a minority of cases.2 Variants
in punctuation, breathings, and accents are not reported unless they affect the meaning; the
same applies to errors of itacism, the presence or absence of iota mutum, and
abbreviations. As forms of c[g]i(b)kbgQ (and all too often of c[g]Tbk^R) prevail in late
texts and are significant for the history of the word, I retain the form that each witness
presents and normally neither normalise c[g]¦kbgQ without manuscript authority nor
report modern emendations of the former into the latter.3
2 I have consulted reproductions (facsimiles, microfilms, or digital images) of Photiusí AM, Sardianusí VW, Simeonís CE, Theocritusí K, Dionysius Thraxís A, Aristophanesí RVE5M, and Hesychiusí H, and seen in person the EMís O and Aristophanesí Lh; the readings of all other mss. are taken from the editions cited in the headline of each testimonium. These editions also provide the respective sigla, except those of the Byzantine Etymologica, which are all cited after those assigned to them by Reitzenstein 1897. 3 In retaining the mss.í readings I follow Grandolini 1999: 8-12, although her remark that ìthe erroneous spelling c[g]ibkbgQ and c[g]Tbk^R (Ö) is the only one attested in the Middle Agesî (10) is inexact: the form with omega is prevalent (rightly ìwidespreadî, 1987-8: 41) but not exclusive. When the two forms occur as variants, I do print c[g]¦kbgQ as difficilior, see pp. 63f. above; I also print c[g]¦kbgQ without ms. authority in the one case (P19) in which the position of the entry in the alphabetical arrangement of the work strongly suggests that this is the form the author had used.
69
P1 Procl. Chr. 38-40 Severyns in Phot. Bibl. 239 p. 320a.9-20 Bekker1 (AM)
38. jR^ d\]b fÙQ —_QgQ _ÓQ ‹Qg_n]pRb �cÙ fgt Õc¦_gQ¦Q fbQR W∂QRb jRÚ g∑gQ W∞] _Qq_\Q jRÚ Õc¦_Q\]bQ ³SWbQ fÏ] c[nXWb] f�Q Õ_QgP_YQTQ� ¢ �cÙ fgt —kWbQ RÃfn], ≈cW[ �]fÚ VYSWbQ. 39. �jnVgPQ kÓ jRp¦VgP cnQfR fÏ W∞] fgˆ] ÕcW[¦QfR] S[Rd¦_WQR —_QgP]� kbÙ jRÚ fÙ c[g]ikbgQ jRÚ fÏ ³VVR fÏ c[gWb[\_YQR dR^QgQfRb �QfbkbR]fYVVgQfW] f�b —_QTb ›] WÒk\ c[Ù] SYQg]� jRÚ SÏ[ ¯]fbQ RÃf�Q �jg�WbQ S[Rd¦QfTQ —_Qg] c[g]Tk^gP, —_Qg] �SjT_^gP, —_Qg] cRb¬Qg] jRÚ fÏ ≈_gbR. 40. �VYSWfg kÓ fÙ c[g]ikbgQ �cWbkÏQ c[g]^T]b fg~] �T_g~] ¢ QRg~], jRÚ �Q f�b c[g]bYQRb ¶bkWfg c[Ù] RÃV¦Q� ¡ kÓ jP[^T] —_Qg] c[Ù] jbpn[RQ ¶bkWfg §]fifTQ. 38 d\]ÚQ A : dR]b M | _ÓQ om. M | Õc¦_gQ¦Q Sylburg 1590 (ëquamquam aptius fortassis Õc¦_Q\_gQí p. 386) : ÕcÙ _�_gQ A : Õc¦_Q\]^Q M | g∑gQ A : g�gQWÚ M | _Qq_\Q fbQÏ jRÚ M || 39 ÕcW[¦QfR] Severyns : Õc\[YfR] codd. (fort. passivum sec. Schott 1585) : �QR[YfgP] Sylburg 1590 : ÕcÓ[ ô_¬] Bekker | �jg�WbQ RÃf�Q M | —_QgP] c[g]Tk^gP] jRÚ —_QgP] �SjT_^gP jRÚ —_QgP] M || 40 c[g]^T]b Severyns : c[g]^R]b(Q) codd. | §]f�]RQ A P2 Io. Sardianus in Aphth. Prog. 21.8 p. 120.9-11 Rabe2 (VW) fÏ kÓ c[g]¦kbR �VYSgQfg g—fT] kbÏ fÙ c[g]b¦QfR] QRg~] ¢ �T_g~] c[Ù] RÃVÙQ ³bkWbQ� fgˆ] kÓ —_QgP] c[Ù] jbpn[RQ §]f�fW] ³bkgP]b. c[g]ikbR V | kÓ V : SÏ[ W
P3 | Aphth. Prog. 21.8 ap. Rabe 1907 p. 566 (PbPkR) fÏ kÓ c[g]ibkbR �VYSgQfg g—fT] kbÏ fÙ c[g]b¦QfR] QRg~] ¢ �T_g~] c[Ù] RÃVÙQ ³bkWbQ� fgˆ] SÏ[ —_QgP] c[Ù] jbpn[RQ §]f�fW] õbkgQ. g—fT] �VYSgQfg R P4 Orion coll. 155.22-156.7 Sturz1
—_Qg]� 1. g∑gQ Õc¦_gQ¦] fb] öQ, jRpÙ W∞] Õcg_gQUQ jRÚ _Qq_\Q ³SWb fÏ] f�Q �cRbQgP_YQTQ c[nXWb]� 2. jWZi[b]fRb kÓ f�Q �SjT_^TQ jRÚ f�Q c[g]Tbkb�Q jRÚ cRbnQTQ, gÃZ ›] j�jW^QTQ _U ƒQfTQ —_QTQ, �VVí ›] SYQg] �cÙ WÒkgP]. cnQfR SÏ[ W∞] fgˆ] ÕcW[YZgQfR] S[Rd¦_WQR —_QgP] �cgdRbQ¦_WpR, jRÚ �cbVYSg_WQ fÙ W∂kg] f�b SYQWb� —_Qg] c[g]Tbkb�Q, —_Qg] �SjT_^gP, —_Qg] cRb¬Qg]. 3. ]\_R^QWb kÓ fg~] �Q ÕcW[gZR~] fÏ SWQbjÏ jRÚ fÏ W∞kbjÏ jRÚ fÏ j�[bR ¿Q¦_RfR� ·]fW _U c[g]fbpW_YQgP fgt W∞kbjgt, �cÙ fgt SWQbjgt QgW~]pRb fÙ W∞kbjÙQ, ›] �cÚ ø_q[gP� cgb\fUQ SÏ[ RÃfÙQ W∞kbj�] VYSg_WQ jRÚ pW~gQ. ≈fRQ VYST_WQ ∞k^T], fÙQ ÕcW[YZgQfR Qggt_WQ �j fgt jgbQgt kbÏ fUQ ÕcW[gZqQ� g—fT jRÚ —_QgQ ∞k^T] VYSg_WQ kbÏ fUQ ]W_Q¦f\fR jRÚ cW[Ú f�Q ³VVTQ, —_QTQ ƒQfTQ. 4. jWZi[b]fRb kÓ f�Q c[g]Tbkb�Q jRÚ jRfÏ fgtf¦ �]fb ã≈fbõ fÏ _ÓQ c[g]ibkbR, jRpÏ jRÚ jYjV\fRb, c[gÔ¦QfW] QRg~] ¢ �T_g~] c[Ù]
70
RÃVÙQ õbkgQ� fÙQ kÓ —_QgQ, c[Ù] jbpn[RQ. 5. g—fT }^kP_g] �Q f�b HW[Ú VP[bj�Q cgb\f�Q. 3 fg~] �Q ÕcW[gZR~] : �Q fR~] ÕcW[gZR~] vel �Q fg~] ÕcW[¦Zgb] Sturz | W∞kbjÏ Ö W∞kbjÙQ Ö W∞kbj�] Sturz : ter ∞k- cod. | VYST_WQ Sturz : VYSg_WQ cod. || 4 jRfÏ ® c[g]ibkbR corruptum sec. Sturz : ≈fb addidi, ut Gallorum verbis utar, faute de mieux | c[g]b¦QfW] Sturz, fort. recte, sed cf. P5.3.
P5 Et.Gen. B fol. 245r ap. Grandolini 1999 p. 9 —_Qg]� 1. Õc¦_gQ¦] fb] ‡Q, jRpÙ W∞] Õcg_gQUQ jRÚ _Qq_\Q ³SWb fÏ] f�Q �cRbQgP_YQTQ c[nXWb]. 2. jWZi[b]fRb kÓ f�Q �SjT_^TQ jRÚ f�Q c[g]Tkb�Q jRÚ cRbnQTQ, gÃZ ›] j�jW^QTQ _U ƒQfTQ —_QTQ, 3. �VVí≈fb fÏ ³VVR c[g]¦kbR, jRpÏ jYjV\fRb, c[gÔ¦QfW] QRg~] ¢ �T_g~] c[Ù] RÃVÙQ õbkgQ� fÙQ —_QgQ, c[Ù] jbpn[RQ. 4. g—fT] }^kP_g]. 2 cRbnQTQ Grandolini ex P4.2 : �cR^QTQ B P6 EM col. 2175 Gaisford (NOQ) —_Qg]� 1. jRfÏ ]PSjgcqQ, Õc¦_gQ¦] fb] ‡Q, jRpÙ W∞] Õcg_gQUQ jRÚ _Qq_\Q ³SWb fÏ] f�Q �cRbQgP_YQTQ c[nXWb] jRÚ �[Wfn]. 2. jWZi[b]fRb kÓ �SjT_^TQ jRÚ c[g]Tbkb�Q jRÚ �cR^QTQ, gÃZ ›] j�jW^QTQ _U ƒQfTQ —_QTQ� S[ndWfRb kÓ —_Qg] c[g]Tbk^R], —_Qg] �SjT_^gP, —_Qg] cRb¬Qg], jRÚ fÏ ≈_gbR� kbR]fYVVWfRb ›] WÒk\ �cÙ SYQgP]� �VVí ≈fb fÏ ³VVR �QfbkbR]fYVVgQfRb. 3. c[g]ibkbR Sn[, jRpÏ jYjV\fRb, c[g]b¦QfW] QRg~] ¢ �T_g~] c[Ù] RÃVÙQ õbkgQ, fÙQ kÓ —_QgQ c[Ù] jbpn[RQ. 4. g—fT] }^kP_g] �Q f�b HW[Ú VP[bj�Q cgb\f�Q. 1 Õcg_gQUQ : Õc¦_Q\]bQ Schmidt p. 389 || 2 �cR^QTQ : cRbnQTQ Larcher ap. Sturz ad P4.2 ex Orione, obl. Grandolini 1999 p. 10 cl. P5.2 | c[g]Tk( ) O : de ceteris e Gaisfordii apparatu non constat | alterum kÓ : SÏ[ Schmidt p. 389 | kbR]fYVVWfRb ® �QfbkbR]fYVVgQfRb : �VVí ≈fb fRtfR jRÚ fÏ ³VVR �QfbkbR]fYVVgP]bQ ›] WÒk\ c[Ù] SYQg] Gaisford : �VVí �QfbkbR]fYVVgQfRb ›] WÒk\ c[Ù] SYQg] ante S[ndWfRb transp. Schmidt (sed ≈fb fÏ ³VVR genuinum est) | �cÙ SYQgP] : W∞] SYQg] Q | ≈fb fÏ ³VVR om. NO | �QfbkbR]fYVVgQfRb N : �QfbkbR]fgV( ) O : �QfbkbR]fYVVgP]b Q : fort. secludendum una cum 3 Sn[ cl. P5.3 || 3 jRpÏ jYjV\fRb Gaisford : �pÏ jYjV(WfRb) Q : �pR( )Rb (?) O : �p\QR~gb N P7 Magna Grammatica C fol. 188r, D —_Qg]� 1. jRfÏ ]PSjgcqQ, Õc¦_gQ¦] fb] ‡Q, jRpÙ W∞] Õcg_gQUQ jRÚ _Qq_\Q ³SWb fÏ] f�Q �cRbQgP_YQTQ c[nXWb] jRÚ �[Wfn]. 2. jWZi[b]fRb kÓ �SjT_^TQ jRÚ c[g]Tkb�Q jRÚ �cR^QTQ, gÃZ ›] j�jW^QTQ _U ƒQfTQ —_QTQ� S[ndWfRb kÓ —_Qg] c[g]Tbk^R], —_Qg] �SjT_^gP, —_Qg] cRb¬Qg], jRÚ fÏ ≈_gbR� kbR]fYVVWfRb ›] WÒk\ �cÙ SYQgP], �VVí ≈fb fÏ ³VVR �QfbkbR]fYVVgQfRb. 3. c[g]ibkbR Sn[, �pÏ jYjV\fRb, c[g]b¦QfW] QRg~] ¢ �T_g~] c[Ù] RÃVÙQ õbkgQ, fÙQ kÓ —_QgQ c[Ù] jbpn[RQ. 4. g—fT }^kP_g] �Q f�b HW[Ú VP[bj�Q cgb\f�Q. 2 kbR]fYVVWfRb ® �QfbkbR]fYVVgQfRb corruptum, v. ad P6.2 || 3 jRpÏ vel jRpí Û Currie
71
P8 | (K) Theoc. 1.61[b] p. 54.11-16 Wendel1
—_Qg]� 1. Õc¦_gQ¦] fb] ‡Q, jRpÙ W∞] Õcg_gQUQ jRÚ _Qq_\Q ³SWb fÏ] f�Q �cRbQgP_YQTQ c[nXWb]. 2. gÃZ ›] j�jW^QTQ _U ƒQfTQ —_QTQ, 3. �VVí ≈fb fÏ ³VVR c[g]ikbR c[g]b¦QfW] QRg~] ¢ �T_g~] c[Ù] RÃVÙQ õbkgQ, fÙQ kÓ —_QgQ c[Ù] jbpn[RQ. 4. g—fT }^kP_g]. Omnia interpolavit Wendel1 ut partim cum P5, partim cum P6 congruerent, sed cf. P9 P9 [Zonar.] col. 1767 Tittmann —_Q\f¦Q� kgXgVgSbj¦Q. jRÚ —_Qg]� 1. g∑gQ W∞cW~Q Õc¦_WQ¦] fb] öQ, jRpÙ W∞] Õc¦_Q\]bQ ³SWb fÏ] f�Q �cRbQgP_YQTQ c[nXWb]. 2. gÃZ ›] j�jW^QTQ _U ƒQfTQ —_QTQ, 3. �VVí ≈fb fÏ ³VVR c[g]ibkbR c[g]b¦QfW] QRg~] ¢ �T_g~] c[Ù] RÃVÙQ õbkgQ, fÙQ kÓ —_QgQ c[Ù] jbpn[RQ. 4. g—fT] }^kP_g]. P10 EM coll. 1947f. Gaisford (NOQ) c[g]Tbk^Rb� 1. cR[Ï fÙ c[g]b¦QfR] QRg~] ¢ �T_g~] c[Ù] RÃVÙQ ³bkWbQ� kbRã]fYVVgQfRbõ kÓ f�Q —_QTQ ≈fb fgˆ] —_QgP] c[Ù] jbpn[RQ §]f�fW] ³bkgP]bQ. 2. g—fT }^kP_g] �Q f�b HW[Ú VP[bj�Q cgb\f�Q. 3. ¢ cR[Ï fÙ c[Ù] RÃfÏ] ³bkWbQ ô_¬] fR~] dTQR~]� 4. ¢ cR[Ï fÙ c[Ù] RÃfÏ] ³bkW]pRb fÏ cgbq_RfR� ‹bkÏ] SÏ[ g� cRVRbgÚ fÏ cgbq_RfR �jnVgPQ� 5. ¢ cR[Ï fÙ c[g]nbkWbQ jRÚ Ú[_¦�WbQ fvb ÕcgjWb_YQ\b VYXWb. �j fv] c[Ù] c[gpY]WT] jRÚ fgt ‹bkq. 1 kbÏ codd. : �QfbkbR]fYVVWfRb Schmidt p. 390, ex quo kbR]fYVVgQfRb conieci : ∞k^Rb Nuüez ap. Schott 16152 p. 66 : ÒkbR FÅrber 1936 p. II 30
P11 Magna Grammatica C fol. 155v, D
c[g]Tbk^Rb� 1. cR[Ï fÙ c[g]b¦QfR] �T_g~] ¢ QRg~] c[Ù] RÃVÙQ ³bkWbQ� kbÏ f�QkW —_QgP]� fgˆ] SÏ[ —_QgP] c[Ù] jbpn[RQ §]f�fW] õbkgQ. 2. g—fT }^kP_g] �Q f�b HW[Ú VP[bj�Q cgb\f�Q. 3. ¢ cR[Ï fÙ c[Ù] RÃfÏ] ³bkWbQ ô_¬] fR~] dTQR~]� 4. ¢ cR[Ï fÙ c[Ù] RÃfÏ] ³bkW]pRb fÏ cgbq_RfR� ‹bkÏ] SÏ[ g� cRVRbgÚ fÏ cgbq_RfR �jnVgPQ� 5. ¢ cR[Ï fÙ c[g]nbkWbQ jRÚ Ú[_¦�WbQ fvb ÕcgjWb_YQ\b VYXWb. �j fv] c[Ù] c[gpY]WT] jRÚ fgt ‹bkq. 1 f�Q kÓ D | cf. P10.1 P12 Poll. 4.82 p. I 225.6sq. Bethe (MAVCBFS) ¯Qbgb kÓ jRÚ �_�Rf\[^gP] RÃVgˆ] ‹Q¦_R]RQ fgˆ] �cÚ fg~] c[g]gk^gb].
72
P13 |L D.T. p. 451.17-20 Hilgard (AE) c[g]¦kb¦Q �]fb cg^\_R ÕcÙ �[[YQTQ ¢ cR[pYQTQ Zg[gt �Q fvb c[g]¦kTb fvb c[Ù] fÙQ pWÙQ �bk¦_WQgQ. dY[WfRb kÓ �Q fg�fTb f�b SYQWb jRÚ fÙ �cgf[Wcfbj¦Q� ¯]fb kÓ cg^\_R �]cR]fbj¦Q, jRfÏ fÙQ �cÙ f�Q pW�Q ZT[b]_ÙQ �bk¦_WQgQ. �[]YQTQ E | �cgf[Wcfbj¦Q : �cg]fWcfbj¦Q Kaibel 1898 p. 36 ex EM R 1680 Lasserre-Livadaras : �cgcW_cfbj¦Q dub. Hilgard ex Men.Rhet. p. 336.5 Russell-Wilson | ]cR]fbj¦Q codd. : corr. Kaibel 1898 p. 36. P14 | (RVE5M) Ar. Av. 853a.R p. 133 Holwerda2
c[g]¦kbR _WSnVR� �cÙ f�Q c[g]¦kTQ� g—fT kÓ ¯VWSgQ fÏ] c[g]RSg_YQR] fg~] pWg~] pP]^R], jRÚ c[g]¦kbR fÏ W∞] cRQ\S�[Wb] f�Q pW�Q cgbq_RfR cR[Ï f�Q VP[bj�Q VWS¦_WQR. lemma om. VM | �cÙ f�Q c[g]¦kTQ om. M : c[g]¦kbR tamquam alterum lemma R | kÓ om. RM | c[gRSg_YQR] V : corr. Rutherford 1896 | pP]^R] E5pc : cg_cn] RV5acM | jRÚ c[g]¦kbR fÏ VEM : jRÚ fÏ R : jRÚ c[g]¦kbR f�Q 5 : c[g]¦kbR kÓ jRÚ fÏ White 1914 | fg~] VP[bjg~] R
P15 Suid. c 2756 p. IV 229 Adler c[g]¦kbR� �cÙ f�Q c[g]¦kTQ� g—fT kÓ ¯VWSgQ fÏ] c[g]RSg_YQR] fg~] pWg~] cg_cn], jRÚ c[g]¦kbR fÏ W∞] cRQ\S�[Wb] pW�Q cgbq_RfR cR[Ï f�Q VP[bj�Q VWS¦_WQR.
P16 | Tzetz. in Ar. Av. 853 p. 1147 Koster1
(c[g]¦kbR�) g—fT SÏ[ ¯VWSgQ fÏ] c[g]RSg_YQR] fg~] pWg~] cg_cn]. P17 | Triclinii (Lh) in Ar. Av. 853a.� p. 133 Holwerda2
�cÙ f�Q ³VVTQ c[g]¦kTQ� jRÚ fÏ] c[g]RSg_YQR] fg~] pWg~] cg_cÏ] jRÚ pP]^R] c[g]¦kgP] ¯VWSgQ, jRÚ c[g]¦kbR fÏ W∞] cRQ\S�[Wb] f�Q pW�Q cgbq_RfR cR[Ï f�Q VP[bj�Q VWS¦_WQR. P18 | Triclinii (Lh) in Ar. Av. 853b p. 133 Holwerda2
c[g]¦kbR _WSnVR� ¶SgPQ c[g]WQYXWb], pP]^R]. �cÙ fgt c[g]nST. ¢ cRb¬QR].
73
P19 Hsch. c 3842 p. 187 Hansen (H) c[g]ikbgQ� ‹bkq, pWgt —_QgQ cW[bYZgP]R. c[g]¦kbgQ Hsch. scripsisse intell. Salmasius ap. Schrevelius 1668 | ‹bkq H2 (Musurus) : ‹bkv] H1
P20 Suid. c 2757 p. IV 229 Adler c[g]¦kbR: —_Qgb. �[ZU c[g]gk^gP� [F1]
P21 Suid. c 2772 p. IV 230 Adler c[g]Tbk^R� jRÚ c[g]ibkbgQ� ¡ —_Qg] P22 Etymologicum Symeonis F fol. 151r c[g]ibkbgQ� 1. VbfRQW^R _WfÏ —_QTQ� 2. �j fgt c[g]bYQRb. 3. ∞]fYgQ ≈fb f�Q _WV�Q fÏ _ÓQ jRVW~fRb c[g]ibkbR, fÏ kÓ Õcg[Zq_RfR. 4. jRÚ g—fT] WÒ[\QfRb kbÏ fÙ cnQfTQ c[g]b¦QfTQ W∞] fÙQ �T_ÙQ f�Q pP_nfTQ fRtfR VYSW]pRb. 5. Õcg[Zq_RfR kÓ âfbQR ¯VWSgQ ¿[Zg�_WQgb jRÚ f[YZgQfW] W∞] fÙQ �T_¦Q. 2 �j : �cÙ sscr. F P23 EM col. 1948 Gaisford (NOQ)
c[g]ibkbgQ� 1. VbfRQW^R _WfÏ —_QTQ� 2. cR[Ï fÙ c[g]bYQRb _WfÏ fg�fgP fg~] pWg~]. 3. ∞]fYgQ ≈fb f�Q _WV�Q jRÚ f�Q —_QTQ fÏ _ÓQ jRVW~fRb c[g]ibkbR, fÏ kÓ Õcg[Zq_RfR, fÏ kÓ ]fn]b_R. 4. jRÚ c[g]ibkbR _ÓQ fÏ VWS¦_WQR ³b]_RfR W∞]dW[g_YQTQ W∞] fÙQ �T_ÙQ f�Q �W[W^TQ, cR[Ï fÙ c[g]b¦QfTQ W∞] fÙQ �T_ÙQ f�Q pP_nfTQ fRtfR VYSWbQ. 5. Õcg[Zq_RfR kÓ âfbQR cnVbQ ¯VWSgQ ¿[Zg�_WQgb jRÚ f[YZgQfW] j�jVTb fgt �T_gt jRbg_YQTQ f�Q �W[W^TQ. 6. ]fn]b_R kÓ Û §]f�fW] —]fW[gQ ¯VWSgQ �QRcRP¦_WQgb _WfÏ fÙ j�jVTb k[R_W~Q fgt �T_gt. 7. ≈fW kÓ cW[bYf[WZgQ fÙQ �T_ÙQ, �cqbW]RQ c[¦fW[gQ _ÓQ �cÙ fgt �[b]fW[gt _Y[gP] �cÚ fÙ kWXbÙQ jRfÏ _^_\]bQ fgt �TkbRjgt j�jVgP, �cWÚ jRÚ RÃfÙ] fUQ �QRQf^RQ f�b gÃ[RQ�b cgbW~fRb j^Q\]bQ �cÙ kP]_�Q �cÚ �QRfgVÏ] dW[¦_WQg]� —]fW[gQ kÓ cnVbQ �cÙ fgt kWXbgt �cÚ fÙ �[b]fW[ÙQ ¶bW]RQ jRfÏ _^_\]bQ fgt gÃ[RQgt� fWVWPfR~gQ kÓ cnQfR fÙQ �T_ÙQ cW[bYf[WZgQ. 6 fÏ kÓ ]fn]b_R om. O, cf. P22.3 || 7 �[b]fW[gt : �[b]fW[gt] O
74
P24 Magna Grammatica C foll. 155v-156r, D c[g]ibkbgQ� 1. VbfRQW^R _WfÏ —_QTQ� 2. cR[Ï fÙ c[g]bYQRb _Wfí RÃf�Q fg~] pWg~]. 3. ∞]fYgQ kÓ ≈fb f�Q _WV�Q jRÚ f�Q —_QTQ fÏ _ÓQ jRVW~fRb c[g]ibkbR, fÏ kÓ Õcg[Zq_RfR, fÏ kÓ ]fn]b_R. 4. jRÚ c[g]ibkbR _ÓQ fÏ VWS¦_WQR ³b]_RfR W∞]dW[g_YQTQ W∞] fÙQ �T_ÙQ f�Q �W[W^TQ, cR[Ï fÙ c[g]b¦QfTQ W∞] fÙQ �T_ÙQ f�Q pP_nfTQ fRtfR VYSWbQ. 5. Õcg[Zq_RfR kÓ âfbQR ¯VWSgQ ¿[Zg�_WQgb jRÚ f[YZgQfW] j�jVTb fgt �T_gt fWVgP_YQTQ f�Q �W[W^TQ. 6. ]fn]b_R kÓ Û §]f�fW] ¯VWSgQ —]fW[gQ �QRcRP¦_WQgb _WfÏ fÙ j�jVTb k[R_W~Q fgt �T_gt. 7. ≈fW kÓ cW[bYf[WZgQ fÙQ �T_ÙQ, �c^W]RQ c[¦fW[gQ _ÓQ �cÙ fgt �[b]fW[gt _Y[gP] �cÚ fÙ kWXbÙQ jRfÏ _^_\]bQ fgt �TkbRjgt j�jVgP, �cWÚ jRÚ RÃfÙ] fUQ �QRQf^RQ f�b gÃ[RQ�b cgbW~fRb j^Q\]bQ �cÙ kP]_�Q �cÚ �QRfgVÏ] dW[¦_WQg]� —]fW[gQ kÓ cnVbQ �cÙ fgt kWXbgt �cÚ fÙ �[b]fW[ÙQ ÒW]RQ jRfÏ _^_\]bQ fgt gÃ[RQgt� fWVWPfR~gQ kÓ cnQfR fÙQ �T_ÙQ cW[bYf[WZgQ. 2 _Wfí RÃfv] D | fÏ kÓ ]fn]b_R in marg. habet D, cf. P22.3, P23.3 || 4 c[g]b¦QfTQ sscrr. R] C || 5 fWVgP_YQTQ sscr. jRbg_YQTQ C, cf. P23.5 || 7 cnQfR C : cnQfTQ D P25 [Zonar.] p. 1583 Tittmann c[g]ibkbgQ� 1. VbfRQW^R _WfÏ —_QTQ. 2. cR[Ï fÙ c[g]bYQRb. P26 Procl. Chr. 34 Severyns in Phot. Bibl. 239 p. 319b.35-320a.1 Bekker1 (AM) jRÚ W∞] pWgˆ] _ÓQ �QRdY[W]pRb —_QgQ, c[g]ikbgQ, cRb¬QR, kbp�[R_�gQ, Q¦_gQ, �kTQ^kbR, ∞¦�RjZgQ, Õcg[Zq_RfR. Õcg[Zq_RfR A : ÕcW[gZq_RfR M
P27 Io. Sardianus in Aphth. Prog. 21.8 p. 119.19-21 Rabe2 (VW) —_Qgb SÏ[ jRÚ cRb¬QW] jRÚ c[g]¦kbR jRÚ kbp�[R_�gb W∞] pWgˆ] WÒTpW VYSW]pRb, fÏ kÓ �Sji_bR W∞] �Qp[icgP]. SÏ[ W : kÓ V | c[g]ikbR jRÚ kbp�[R_�R V P28 | Aphth. Prog. 21.8 ap. Rabe 1907: 566 (PbPkR) —_Qgb SÏ[ jRÚ cRb¬QW] jRÚ c[g]ikbR jRÚ kbp�[R_�gb W∞] pWgˆ] WÒTpR]b VYSW]pRb, fÏ kÓ �Sji_bR W∞] �Qp[icgP].
75
P29 Porph. in Hor. epist. 2.1.134 p. 319.27-30 Meyer (M) poscit opem chorus et p(raesentia) n(umina) s(entit) : ac per hoc exauditur. hic autem significat carmina quibus dii placantur, hoc est paeanas dithyrambos hymnos prosodia. P30 Poll. 1.38 p. I 11.11-13 Bethe (MAVCBFS) ‹bkRÚ W∞] pWgˆ] jgbQ�] _ÓQ {cRb¬QW]} —_Qgb, ∞k^T] kÓ �[fY_bkg] —_Qg] g–cbSSg], �c¦VVTQg] ¡ cRbnQ, �_dgfY[TQ c[g]¦kbR, }bgQ�]gP kbp�[R_�g], }q_\f[g] ÒgPVg]. cRb¬QW] secl. DíAlessio 1997 p. 29 | —_Qg] AVC : _ÓQ M : om. BFS fort. recte | �c¦VVTQg] ¡ cRbnQ, �[fY_bkg] (—_Qg]) g–cbSSg] SC | g–cbSSg] FS : ¡ cbSS¦] M : ¡ ÕcbQg] B : g–cbSSW] C : om. A | c[g]ibkbR FS P31 Procl. Chr. 41 Severyns in Phot. Bibl. 239 p. 320a.24f. Bekker1 (AM) jRfRZ[\]fbj�] kÓ jRÚ fÏ c[g]ikbn fbQW] cRb¬QR] VYSgP]bQ. P32 | Hor. (ëpseudo-Acroí) sat. 2.1.1 p. II 116.4-8 Keller (5bfVcF) eclogae haec nomina habent: si ad Iouem, hymni; si ad Apollinem aut Dianam aut Latonam, peanes; si ad Liberum aut Semelen, dityrambi; si ad ceteros deos, prosodia; si ad hominem, laudes aut uituperationes aut luctus aut aliquid tale. aut luctus — tale om. cF
P33 Choerob. in Heph. pp. 216f. Consbruch (KU) fbQÓ] kÓ RÃfÙQ c[g]gkbRjÙQ jRVgt]bQ jRÚ cg_cWPfbjÙQ kbÏ fÙ �Q fg~] c[g]gk^gb] —_Qgb] g—fT jRVgP_YQgb] jRÚ �Q fR~] cg_cR~] W∂QRb fR~] }bgQP]bRjR~]. RÃf¦Q scil. palimbacchium | c[g]gkbRjÙQ Hoerschelmann ap. Studemund 1886 : c[gcgkbRjÙQ KU | jRVgt]bQ jRÚ cg_cgfbjÙQ K, corr. Hoerschelmann ex P34 : g� kÓ cg_cbjÙQ jRVgt]b U | c[g]gk^gb] Hoerschelmann : c[gcgk^gb] U : cgk^gb] K | jRVgP_YQgb] K : jRVgP_YQ\Q U | cg_cR~] K : c[gcg_cR~] U | ��cbfqkWbgQ� W∂QRb Egenolff 1887 p. 393 ex P34
76
P34 |B Heph. p. 302 Consbruch (pk) ¡ jRÚ c[g]gkbRjÙ] jRÚ cg_cWPfbjÙ] kbÏ fÙ �Q c[g]gk^gb] —_Qgb] g—fT jRVgP_YQgb] jRÚ �Q fR~] }bgQP]bRjR~] cg_cR~] �cbfqkWbg] W∂QRb. ¡ scil. palimbacchius | c[g]gkbRjÙ] Turnebus 1553 : c[gcgkbRjÙ] pk | c[g]gk^gb] Turnebus : cgk^gb] pk : cgk^\] Þuntina | �cbfqkWbgQ pk
DIDYMUS IN PROCLUS AND THE ETYMOLOGICAL TRADITION
The first twelve testimonies preserve material ultimately derived from Didymusí treatise
On Lyric Poets, recognisable as such by the citation in P4.5, P6.4, P7.4, P10.2, and P11.2
(only the authorís name in P5.4, 8.4, 9.4).4 They can be divided into two branches.
Representatives of the first are Proclusí Chrestomathy, which survives in a partial
summary compiled by the patriarch Photius for his Library, and two brief notes in John of
Sardisí commentary to Aphthoniusí Progymnasmata and in the so-called H-scholia to the
same text (P1-P3).5 The core of the second branch (P4-P11) consists of the late antique
4 See in the first instance Grandolini 1999; fragments collected and annotated also in Schmidt pp. 386-96. 5 The identity of ëProclusí is disputed. Suid. c 2473 Adler lists the Chrestomathy among the works of the fifth-century Neoplatonist, but an otherwise unknown, earlier grammarian has also been postulated since Valesius 1740: 168f. The issue is still unresolved: see most recently Hillgruber 1990, Longo 1995, and West 2013: 7-11. The dramatic date (so to speak) of Photiusí Library is 855/6, when Photius took part in an embassy to the Caliph in Baghdad and recorded his readings for the benefit of his brother Tarasius, who had stayed in Constantinople (Severyns I 1-3). The exact date of John has been complex to establish. A known terminus ante quem was his tenth-century ms. C (Rabe2 IV, XVI); Beck 1959: 510 placed him at the turn of the ninth century, Rabe2 XVII-XVIII no later than the mid-ninth, Foss 1976: 66 in the mid- or late ninth. Alpers 20132: 16-43 has been able to date his tenure of the metropolitan see of Sardis more exactly to between 824/5 and a date in or shortly before 859; Kennedy 2003: 173 had plausibly suggested that the commentary predates his episcopal ordination. Alpersí brief treatment of Johnís excursus on lyric genres (20132: 144-6) insists on his relationship to the Etymologica, only citing Proclus as a parallel (145 n. 429), but Johnís derivation from Proclus is strongly suggested (i) by terms that P2 only shares with P1 (�VYSWfg, §]f�fW]) not with Orion, and (ii) by P27, which closely echoes P26 and finds no parallel in the Etymologica. The fact that John has QRg~] ¢ �T_g~] like the Etymologica, not �T_g~] ¢ QRg~] like Proclus, does not shift the balance considerably: the reading that we ascribe to Proclus may well have been due to Photiusí
77
and Byzantine Etymologica, at the basis of which stands the grammarian Orion of Thebes,
who was active in the fifth century AD.6 It is from his entry on the hymn that the
Etymologicum Genuinum drew its own; 7 the latter then shaped the ones in the
Etymologicum Gudianum,8 P6 in the Etymologicum Magnum (probably with direct
contamination from Orion),9 P7 in the Magna Grammatica,10 and a scholium found in a
twelfth-century manuscript of Theocritus, which in turn the lexicon falsely attributed to
Zonaras replicates.11 The Etymologicum Magnum and the Magna Grammatica also transfer
the relevant section, apparently with a substantial borrowing from P2, to an entry on
condensation of the original. Much of Johnís note to Aphth. Prog. 21.8 (pp. 119f. Rabe2) appears to derive from the original text of Procl. Chr. 37, 40-2 Severyns, conflated with Men. Rhet. pp. 331f. Reynolds-Wilson (Rabe2 ibid.). On Johnís sources in general see Rabe2 XX-XXXI, Alpers 20132: 43-148 (focussing on rhetorical and non-rhetorical sources respectively). On the eleventh-century H-scholiaís (indirect) dependence on John see Rabe2 XII, XXXII. 6 On Orion and the ancient records of his life see Wendel in RE XVIII/1 (1939) 1083f., Kaster 1988: 322-5.
7 On the Et. Gen. see Reitzenstein 1897: 1-69, Lasserre-Livadaras I iv-viii. According to the subscription preserved by ms. B, the lexicon was completed in AD 882 (Reitzenstein 1897: 69).
8 Et.Gud. only presents the etymology and description of the hymn, with no reference to the prosodion: d fol. 133v = col. 540.38-40 Sturz2 —_Qg]� cR[Ï fgt �Õ_YQT� Õcg_YQT, Õc¦_gQ¦] fb] ‡Q, jRÚ �Q ]PSjgcvb —_Qg], jRpÙ W∞] Õcg_gQUQ jRÚ c[¬XbQ ³SWbQ fÏ] f�Q �cRbQgP_YQTQ �[Wfn]. (I report the reading of d ñ from microfilm ñ as it is the archetype of the surviving manuscripts, whereas Sturz used its descendant w: see Reitzenstein 1897: 90-105, Cellerini 1988: 12-27, see now Maleci 1995.) 9 The EM must have some degree of direct dependence on Orion independently of the Et. Gen., as is suggested by (i) the garbled second half of P6.2, which evidently goes back to P4.2 and is not to be found in P5, and (ii) the full citation of the title of Didymusí work, which is exclusive to Orion and the EM (and the Magna Grammatica, which probably draws on the latter). Neither of the surviving mss. of the Et. Gen. is thought to preserve a complete text of the work (Reitzenstein 1897: 52f.), which allows the possibility that P6.2 derives from the fuller Et. Gen.: titles are especially susceptible to omission in the process of excerption, cp. e.g. the epitome of Athenaeusí Deipnosophistai. However, the concurrence of two such cases within one entry does meet with some suspicion. On the EM see Reitzenstein 1897: 212-53, Lasserre-Livadaras I xvii-xxii. 10 On the Etymologicum of Simeon ¡ _YSR] S[R__Rfbj¦] (early twelfth century) and the so-called Magna Grammatica (thirteenth) which derives from it and the EM, see Lasserre-Livadaras I xii-xviii, Baldi 2013: xxiv-liv.
11 P8 and P9 share the omission of jWZi[b]fRb — cRbnQTQ, which produces nonsense and cannot have arisen independently; P9ís Õc¦_Q\]bQ for P8ís Õcg_gQUQ jRÚ _Qq_\Q, which coincides with the earlier testimonies, proves that the immediate source of P8 does not derive from that of P9. On the date of Theocritusí K (the same ms. as Pindarís A, Ambrosianus C 222 inf.) see Mazzucchi 2000: 205-7 and n. 5, 2003: 274f., 2004 esp. 435-7; on that of pseudo-Zonaras, Alpers in RE Suppl. XIX (1972) 736f. On the borrowings from the etymological tradition in the etymological scholia peculiar to K see Wendel1 XI-XII.
78
c[g]Tbk^R, where it is followed by three definitions of prosody proper (P10.3-5, P11.3-5)
which do not concern us here.12
The relationship between the two branches is unclear. On the one hand, their close
resemblance may suggest direct dependence, either of Orion from Proclus if the latter is a
grammarian of the second-century AD, as is now more commonly accepted, or the
opposite if he is the illustrious Neoplatonist, who had been Orionís student (Marin. Procl.
8 Saffrey-Segonds).13 Internal evidence is not decisive. The material peculiar to Orion
(P4.3) may be his own elaboration on the etymology he found in his source, while that
peculiar to Proclus can be either an innovation (e.g. the derivation from —kT in P1.39) or
an easy inference from the context (e.g. §]fifTQ P1.40).14 Furthermore, the fact that both
texts are excerpts from originally fuller works prevents one from securely establishing the
independence of either from the other. On the other hand, it is also possible that the two
branches have no connexion to one another except the common ancestor Didymus. In this
case, given the unlikelihood of largely identical selection by two independent authors, the
overall similarity of their accounts suggests that Didymusí overview of the lyric genres, or
at least of hymn and prosodion, did not treat the subject much more extensively than its
surviving descendants do.
12 Rutherford 2003: 719 takes them for alternative definitions of prosodion as ìaccompanying songî (see p. 64 above), but see already FÅrber 1936: I 49. Compare the almost identical definitions of prosody in other sources: 3-4 cp. |M in Dionys. Thrax p. 305 Hilgard VYSWfRb kÓ c[g]Tbk^R �cÙ fgt c[Ù] RÃfUQ ³bkW]pRb fUQ dTQqQ� ‹bkÏ] kÓ ¯VWSgQ g� �[ZR~gb fÏ cgbq_RfR, jRÚ �gbkgˆ] fgˆ] cgb\fn], 4-5 cp. [Zonar.] coll. 1580f. Tittmann ¢ cR[Ï fÙ c[g]nbkWbQ jRÚ Ú[_¦�WbQ fvb ÕcgjWb_YQ\b VYXWb. ¢ cR[Ï fÙ c[Ù] RÃfUQ ³bkW]pRb fÏ cgbq_RfR. ‹bkÏ] SÏ[ g� cRVRbgÚ fÏ cgbq_RfR �jnVgPQ, 5 cp. [Theodos. Gr.] p. 61 Goettling WÒ[\fRb kÓ ô c[g]Tbk^R cR[Ï fÙ c[g]nbkWbQ jRÚ Ú[_¦�WbQ fvb ÕcgjWb_YQ\b VYXWb, |l in Dionys. Thrax p. 454 Hilgard VYSWfRb kÓ c[g]Tbk^R cR[Ï fÙ c[g]nbkWbQ ¶SgPQ Ú[_¦�W]pRb fvb ÕcgjWb_YQ\b VYXWb.
13 So Kaibel 1898: 34.
14 The etymology from —kT is apparently not derived from Didymus: see FÅrber 1936: I 48, Severyns II 120, Grandolini 1999: 11.
79
Since Photiusí r�sum� never cites the sources of the Chrestomathy, it cannot safely
be determined how much in Proclusí description of each lyric genre goes back to Didymus
beyond what coincides with an attributed quotation in the etymological tradition. However,
whenever there is an overlap in subject, the picture seems to be one of close reliance. Just
as Proclusí discussion of hymn and prosodion, with the possible exception of —kT noted
above, evidently depends on Didymus, so is every element in Proclusí treatment of elegy
(24-6 Severyns) explicitly attributed to Didymus by the etymologists;15 similarly, the
paragraph on the paean is evidently parallel to a passage in the Etymologicum Magnum
(col. 1859 Gaisford) which contains the paretymology that is in turn ascribed to Didymus
by the Etymologicum Gudianum (col. 446 Sturz2). One may further speculate that if P31
ultimately depends (also) on the double classification of F5 and on such cases as Paean 7
and the cRbÏQ c[g]gkbRj¦] mentioned by the scholia to Isthmian 1, a commentator on
Pindar is a good candidate for making such a point.16 This suggests that Proclusí account of
lyric genres depends substantially on Didymus even when correspondence cannot be
ascertained on independent evidence. Nonetheless, decisive evidence is lacking, and some
caution is in order.17
Beside a conjecturally attributed biographical notice on Theognis (fr. IV.9.2
Schmidt), all that survives of Didymusí treatise is brief snippets on the characteristics of
individual genres: hymn and prosodion (IV.9.3-4), elegy (IV.9.1), and paean (IV.9.5).
Consistently with the bias of the quoting sources, etymology plays a conspicuous role in
the extant fragments. If the list of genres in Proclus 34-6 Severyns and its offspring in the
15 See Severyns II 99-102.
16 On the relationship between paean and prosodion see below, pp. 100-3. That Didymusí classification of lyric poetry is related to the Alexandrian edition of the lyricists was also the opinion of Pfeiffer 1968: 184.
17 Rightly Grandolini 1999: 11.
80
commentaries to Apthonius (see P26-P28) also originate wholly or partly with Didymus,
one notes that each of the titles attested for Pindar and Bacchylidesí books are represented
there, thus confirming the intuitive supposition that his theory was based to a large extent
on the Alexandrian editions of the late archaic lyricists.
As far as the evidence goes, Didymus seems to have treated the prosodion in the
context of hymns sensu lato as constituting a minimal pair with the hymn stricto sensu,
distinguished from it by its mode of performance and thus, one might infer, similar to it in
every other significant respect. In his reconstruction, (i) —_Qg] comes from Õc¦_gQg],
since hymns are durable and help preserve the memory of those whom they sing, with a
double etymology from (Õcg)_YQT and from (Õcg)_b_Qq]jT (P1.38, P4.1, P5.1, P6.1,
P7.1, P8.1, P9.1); (ii) in its broader sense the term applies to such songs as prosodia,
encomia and paeans, which are distinct from it as species are from the genus that includes
them, as their use allegedly shows (P1.39, P4.2-3, P6.2, P7.2, and curtailed in P5.2, P8.2,
P9.2);18 (iii) the prosodion was performed to the aulos during a procession towards an altar
or a temple ñ whence the name ñ while the hymn ëin the proper sense of the wordí was
sung to the kithara while stationary (P1.40, P2, P3, P4.4, P5.3, P6.3, P7.3, P8.3, P9.3,
P10.1).19 The word seems to have been the subject of a twofold etymology, concomitantly
from (³bkWbQ) c[g]b¦QfR] and from ÃbkWbQ c[Ù] RÃV¦Q: a prosodion is a ëprocessional
18 On Didymusí use of SYQg] and W∂kg] see Grandolini 1999: 11f. The expressions —_Qg] �SjT_^gP, —_Qg] c[g]gk^gP, and —_Qg] cRb¬Qg] do not seem actually to occur elsewhere in Greek literature up to the end of the first millennium AD; the closest one finds are c[g]¦kbgb —_Qgb (P33-P34) and —_Qg] cRbnQ (| Ar. Ve. 874a Koster2, | Ar. Eq. 1318a Jones-Wilson, | Aristid. 124.20 p. 132 Dindorf, and a dubious ms. reading at P30) as well as —_Qg] g–cbSSg] in some mss. at P30.
19 §]fifTQ (P1.40) and §]f�fW] (P2, P3, P10.1, P11.1) are best taken to imply stationary dance, not complete stillness: Smyth 1900: XXIX n. 1 and Severyns p. 124 (cp. Dale 1950: 36-8 on the stasima of Attic tragedy). Grandolini 1987-8: 47 suggests that the etymology may have been due not to Didymus himself but to the etymologistís misunderstanding of what was simply a statement on the typical mode of performance of prosodia, but in view of the attribution to Didymus of similar etymologies of hymn, paean, and elegy (see the previous page), economy favours the supposition that they were actually to be found in his treatise.
81
songí and at the same time a ësong accompaniedí by the aulos. It is unclear whether the
first of these etymologies involved a reference to c[¦]gkgb at all: it may be implied in
c[g]b¦QfR], but this explanation does not specifically account for the -gk- element in the
word, and a further reference to ³bkWbQ (thus to be taken apo koinou with both occurrences
of c[¦]) is a valid, and possibly a better, contender.
Both the word-play with c[g]bYQRb and the association with the aulos were present
already in the passage of the Birds discussed in the previous chapter:20 the coincidence is
suspicious, but it is uncertain how much of a part this antecedent may have played in
shaping Didymusí theory. A connexion with the aulos is also implicit in Heraclidesí
statement that the first ]P]f\]n_WQg] of prosodia was the aulete Clonas, who also
established the aulodic nomoi (fr. 157 Wehrli).21 Pronomus of Thebes, who composed a
prosodion to Delos for the Euboeans of Chalcis (PMG 767), as we saw, was also an
aulete.22 The visual arts provide several representations of processions accompanied by
aulos-playing, whether alone or together with other instruments,23 and Pollux likewise
remarks that there was a specific kind of aulos that accompanied prosodia (P12). On the
other hand, there is also evidence that the performance of compositions regarded as
prosodia was not limited to accompaniment by the aulos, at least in the Hellenistic age: the
Delphic ìpaean and prosodionî by Limenius (CID III 2) was performed to the kithara, the
Delian prosodion of Amphicles (I.D�los 1497) to the lyre.24 Furthermore, the polar
opposition between aulos and kithara ñ heavily laden with ideological connotations since
20 See pp. 46-9 above; note, however, that the relevant scholion (| Ar. Av. 857a Holwerda2) connects the aulos-playing to sacrifices rather than to processions or to processional songs.
21 See pp. 49f. above.
22 See p. 44 above.
23 See Haldane 1966: 98-101.
24 See pp. 57f. above.
82
the late fifth century ñ may have itself been overplayed to the detriment of historical
accuracy.25 It hardly needs saying that there is no evidence that Didymus (or, for that
matter, Aristophanes of Byzantium) had information on the musical accompaniment of
each and every prosodion he read, or of any significant number of them. His distinction ñ
especially in that it consists of two different distinctions whose outcomes supposedly
coincide, stationary vs. processional and kithara vs. aulos ñ looks suspiciously neat, and at
the price of excessive scepticism it is hard to escape the feeling that his theory, although
probably based to some extent on ascertainable fact, must also have involved a proportion
of over-schematisation.
Before moving on to the non-Didymean testimonies, an orthographic clarification is
needed. Despite the prevalence of forms with omega in the manuscripts of the Etymologica
(but note P5.3, if correctly reported), Didymus must have written c[g]¦kbgQ with omicron.
That was the form that was established as a book-title for Pindar and Bacchylides, and an
actual change in a work devoted to the Greek lyricists and their works would have made
little sense. Further support for this view comes from Proclus, who seems to have used the
form with omicron: the manuscripts of Photiusí Library alternate between omicron and
omega, but the former prevails numerically, and is unlikely to have arisen from
hypercorrection. Fanciful etymology can thrive even unbuttressed by textual variation:
beside —_Qg] from Õc¦_gQg] and Õcg_gQq, Didymus also connects ¯VWSg] with W“
VYSWbQ (fr. IV.9.1 Schmidt) and cRbnQ with cR�WbQ (IV.9.5). It will have been partly the
very paretymology expounded in these passages, partly the inevitable presence of
c[g]Tbk^R at the back of any respectable scribeís mind, that conspired to produce such
25 Wilson 1999, 2003, Martin 2003, see also Iannucci 2011. On the presence of both aulos and kithara in visual depictions of cult see Haldane 1963: 99-101, 104, Maas and Snyder 1980: 68f.; Berlinzani 2004: 124 n. 26 lists several mentions of the joint playing of stringed and wind instruments in Pindarís Epinicians.
83
pervasive alteration in the subsequent transmission. It should also be noted that what the
manuscripts more frequently present are in fact forms of c[g]Tbk^R (P4.2, 4, P5.2, P6.2,
P7.2), not of c[g]i(b)kbgQ, a process of conflation to which P10 and P11 also testify.
THE DIONYSIAN SCHOLIA
The London scholia to Dionysius Thrax (P13), probably going back to the grammarian
Heliodorus,26 unambiguously associate prosodia with processions, although without
specifying whether it is the chorus itself which processes (compare P23.4, P24.4). It also
adds two interesting pieces of information: that their performance was not the prerogative
of either sex, and that it was choral. The latter was all to be expected, and is also implicit in
the plural forms that pepper the testimonies to Didymusí theory. The former is potentially
more useful, if it reflects features present to some degree in the Alexandrian classification
of the ëchoral lyricistsí: it entails that a song performed by a chorus of maidens would not
automatically be counted as a Partheneion. The connexion of this piece of information
with Alexandrian editions remains uncertain: none of the Prosodia that we know of,
whether those of the fifth-century lyricists or the later examples of the genre, is visibly
meant for females to perform, and when the gender of the performers is known for certain,
as it is in several of the Hellenistic and later examples, they are invariably male.27 Nonethe-
less, we do know of processional rites in which women played a central role (the Theban
26 On the |L see Hilgard XXXII-XXXVII; on Heliodorus see id. XVIII, Schultz in RE VIII/1 (1912) 40f. s.v. Heliodorus 17.
27 Male were the cR~kW] that performed Cleocharesí ìprosodion and paian and hymnî and Amphiclesí prosodion (pp. 57 above), the Athenian technitati that performed CID III 1-2, and with all likelihood also the performers at the Eretrian Artemisia (pp. 56-8); if the prosodia performed at the Delphic Soteria and the Thespian Musaea (pp. 61f.) were solo performances given by the respective poets, these were all male too. According to Paus. 4.4.1, the prosodion of ëEumelusí was also composed for a chorus of men.
84
Daphnephoria to name but one), and there is no compelling reason to discount this piece of
information altogether. One will note that Bacchylides 13.83-99 depict a chorus of
Aeginetan maidens performing a song the content of whose mythical narrative, at least at
the beginning, is strikingly similar to that of F5: although the text of F5 gives no explicit
information on the gender of its performers if we treat it separately from Paean 6(a)
(which openly advertises itself as performed by QYgb, 122), and any direct connexion
between the two odes is uncertain, Bacchylides does portray songs akin to F5 as performed
by females, and these too ñ if they were to any degree historical, and not a figment of his
poetic imagination ñ may at some point have been regarded as Prosodia.28
The commentator classes with the prosodion a sub-category of �cgf[Wcfbjn, a kind
of prosodion I rebours, recessional hymns sung while departing ìfrom the godsî, that is,
from the shrine, rather than while approaching it. Alfred Hilgard connects these, whose
transmitted name is a hapax, to the —_Qgb �cgcW_cfbjg^ described by Menander Rhetor
(pp. 333.10-12, 336.6-23 Russell-Wilson): a very rare kind of songs, used only by ìthe
poetsî and notably by Bacchylides, the opposite of cletic hymns (336.6-8), �cgcg_cUQ ›]
�cgk\_^R] fbQ¦] SbSQg_YQ\] ¯ZgQfW] (333.11f.). The connexion with P13, however, fails
to carry conviction: Menanderís apopemptic hymns are reportedly performed upon the
deityís departure from his or her shrine, while with Heliodorusí �cgf[Wcfbjn it is the
worshippers who depart.29 Georg Kaibel suggested amending into �cg]fWcfbj¦Q, a kind
of song mentioned by the Etymologicum Genuinum and Magnum (R 1034, 1680 Lasserre-
Livadaras), allegedly sung �Q fg~] cRb¬]b by those about to sail as they were removing
28 This argument also applies to one interpretation of F3, see pp. 213f. below.
29 On Menanderís examples see Russell-Wilson p. 235. Wilamowitz 1922: 330 saw an apopemptic hymn in a fragment of Bacchylides (now fr. 1a M.), but the connexion is quite tenuous, as Gostoli 2009: 592f. notes. More persuasive is Gostoliís own argument that an example of an apopemptic hymn as described by Menander is Call. Lav. Pall. (593f.).
85
ìthe wreathsî.30 Kaibel connected these with the wreathed state-ship of the yearly Athenian
theoria to Delos (Pl. Phd. 58a-b) and referred both the Etymologicaís testimony and
Heliodorusí to the un-wreathing that will have taken place upon the shipís departure from
the island for the voyage back. However, there is little in the Etymologica, and even less in
P13, that concretely bears out the special connexion of either song-type with Delos, which
remains nothing more than an interesting possibility.31 To amend away one hapax on the
basis of another is dubious method, and in P13 it is �cgf[Ycg_Rb, not �cgcY_cT or
�cg]fYdg_Rb, that best suits the notion expressed by jRfÏ fÙQ �cÙ f�Q pW�Q ZT[b]_¦Q
with no further specification. The transmitted reading should be retained.32
THE ARISTOPHANIC SCHOLIA
The scholia vetera to Aristophanes, on which both the Suda and John Tzetzes draw (P14-
P16, see also P17), start from the etymology of c[g]¦kbgQ from c[¦]gkg]: this is a
procession for the gods, and (accordingly) c[g]¦kbR are the lyricistsí compositions for
religious festivals. Manuscript R skips the derivation and erroneously links the definition
30 LSJís translation ìa bridal chantî, repeated without comment by the DGE, derives from a rather idle supposition in the ThGL that does not seem to take into account the sourcesí _WVV¦QfTQ (D : _YVVgQfW] cett.) �cgcVW~Q. It is of no value and ought to be corrected.
31 The practice of pouring libations and singing paeans upon a shipís departure was by no means confined to the Delian theoria: see for one the departure of the Athenian fleet on the fateful Sicilian expedition as described by Thuc. 6.32.1-2. On the practice of crowning ships and its sacral connotation see Wachsmuth 1967: 90-3 n. 84. More importantly, nothing in the text of either Etymologicum demands that it is the ship, rather than the men on board, that is being un-garlanded. One might also speculate that the unwreathing could have taken place at the end of a voyage, not at the start of the next.
32 See also FÅrber 1936: I 31. Conversely, Kaibelís �]cR]fbj¦Q is integral to his wider reconstruction of P13, but can and should be adopted even in isolation: the manuscriptsí ]cR]fbj¦Q cannot easily be made to give satisfactory sense in context, and Kaibelís correction is neat and effective (for �]cn�g_Rb and cognates used of salutation upon parting as well as upon meeting see DGE s.vv. �]cn�g_Rb II, �]cR]_¦]).
86
that follows to prosodia instead of to processions; Tzetzes follows suit.33 The most
important thing to note in P14 is that, despite acknowledging the etymology, the scholiast
refers the performance of prosodia not exclusively to processions, but to any ìfestival of
the godsî. This may suggest awareness, or perhaps mere suspicion, that the classification
of an ode as a Prosodion did not entail certainty of processional performance, or even
performance in the context of a procession.34
MORE FROM THE BYZANTINE ETYMOLOGICA
Three brief lexicographical testimonies do much the same as the Aristophanes scholia and
simply present the prosodion as a —_Qg] (P19-P21): in the Platonic and post-Platonic sense
of the word, a song in honour of a god, something which most of our sorces, direct or
indirect, either state or imply in some way.
An entry in Simeon, the Etymologicum Magnum, and the Magna Grammatica starts
from a similar notion, augmented with a reference to prayer and the derivation from
c[¦]Wb_b (P22.1-2, P23.1-2, P24.1-2, replicated in P25). What follows, however, is rather
more bizarre. A first section divides _YV\ jRÚ —_Qgb ñ ëhymnsí more narrowly rather than
ësongsí more broadly, as the context shows ñ into three sorts, according to the stage of the
sacrifice during which they are supposedly performed: prosodia sung while the victims are
33 Grandolini 1986: 263f. defends R and P16 in positing for c[g]¦kbgQ (in the Birds no less than in the scholia, see pp. 46-9 above) an additional meaning equivalent to c[¦]gkg], but this faces the following difficulties as far as the scholia are concerned: (i) the sense given by VE5 is unobjectionable and amply paralleled, while that of R is unique in extant scholarly literature; (ii) the reading of R is easily explained as corruption of that of VE5, but not the reverse; (iii) M omits �cÙ f�Q c[g]¦kTQ like R but retains c[g]¦kbR in the following sentence, thus presenting a corruption similar to R but at a less pervasive stage. Whiteís emendation to the text of V is predicated upon that of R, and should likewise be rejected. Whatever the meaning of c[g]¦kbgQ was in Aristophanes, the scholia originally interpreted it as a reference to the genre.
34 See pp. 116f. below.
87
processing towards the altar, hyporchemes sung while dancing or running around the altar
as the victims are burning on it, and stasima sung while stationary once the running is over
(P23.3-6, P24.3-6, curtailed in P22.3-5). Then, the hyporcheme-stage is divided into a
further three parts: a movement from left to right, like that of the Zodiac; one from right to
left, like that of the heavens; and one ìaround the whole altarî, possibly signifying a
complete circumambulation after two semicircular ones (P23.7, P24.7).
This last section appears to be related to a common conception, ascribed by one
source (cap. f p. III 311 Dr.) to Ptolemyís otherwise unknown treatise HW[Ú ]fRfbjv]
cgbq]WT], of how the triads of choral lyric are articulated into strophe, antistrophe, and
epode according to their respective dance-steps around an altar, steps which purportedly
mirror the motion of the heavenly bodies.35 One difference is that in Ptolemyís view,
according to the available sources, the epode was sung while stationary, representing the
motionless Earth: there seems to be no parallel for the final circle (in which direction?) of
section 7.36 One may suppose that this circle was devised as a filler of sorts when the
astronomical interpretation of the triad was adapted to suit the second of the purported
three stages of sacrificial song: these required stationary singing to constitute the third
35 Cap. b, c, f pp. III 306f., 311 Dr., Macr. somn. 2.3.5, ps. Mar. Victorin, gramm. (= Aelius Festus Apthonius) GL VI pp. 58, 60, Fortun. gramm. ibid. p. 295, |B Heph. pp. 172f. Westphal (not in Consbruch), | E. Hec. 640 in Crusius 1888: 10, cp. Olymp. in Grg. 5.5 (similar although with remarkable differences). See Crusius 1888: 9-14, FÅrber 1936: I 20-2, Mullen 1982: 225-30 (who is ready to take the astronomic symbolism as current in Pindarís time), Montanari 1989, and Briand 2009: 99f., whose contention that ìthe cosmic interpretation of these movements appeared since Aristotleî fr. 11 Rose is rather stretched (Aristotle merely calls the motion of the stars a ìwell-ordered danceî); I have not seen Tessier 1999. Our Ptolemy is more probably the astronomer Claudius Ptolemy than the second-century grammarian Ptolemy Epithetes (fr. *6 Montanari): Crusius 1888: 12, Montanari 1989: 153f. against A. Dihle in RE XXIII/2 (1959) 1862f. s.v. Ptolemaios 78. For speculation on the content of Ptolemyís treatise see Montanari 1989: 154. I mention as a curiosity ñ but one that arguably illustrates the role of even relatively obscure scholarship in the reception of poetry by other poets ñ that this conception of the lyric triad was also dramatized by Giovanni Pascoli in his Poemi Conviviali (I vecchi di Ceo IV 16-18, 29f., 38-40).
36 The problem was noted by FÅrber 1936: I 22.
88
stage and thus left a gap at the end of the second. The lack of an astronomical explanation
for the final circle, unlike the two earlier semi-circles, arguably supports this hypothesis.
The broader articulation envisaged in sections 4-6 is even more obscure. The only
parallel I can find is in Euanthiusí essay On comedy, where comedy and tragedy are said to
have been a simplex carmen Ö quod chorus circa aras fumantes nunc spatiatus nunc
consistens nunc reuoluens gyros cum tibicine concinebat (2.1 p. I 15 Wessner) before
actors, props, and act-divisions were introduced.37 The notion of the development of drama
out of cult performances fits in a long tradition that goes at least as far back as Aristotleís
Poetics (1449a), but the detail of the three states (as it were) of choral performance during
a sacrifice ñ albeit apparently in no particular sequence ñ seems to be unique to Euanthius
in the extant evidence. The terms Õc¦[Z\_R and ]fn]b_gQ in P23 and P24 suit the
hypothesis that this section is to be linked to scholarship on drama, since they were
employed as a contrasting pair in Tzetzesí analysis of the constituents of tragedy.38
However, there is no trace of c[g]¦kbgQ used in a similar context elsewhere, although it is
opposed to ]fn]b_gQ in a passage of Philo of Alexandria concerned with divine cult, as we
saw.39 Verbal resonances suggest that Didymusí doctrine of prosodion and hymn is
somewhere in the background (4 c[g]b¦QfTQ, 6 §]f�fW]), but this need not indicate
actual derivation from Didymus so much as an echo of the immediately preceding entries,
P10 and P11, at the back of the compilerís mind.
Given the derivative, composite, and not wholly consequential nature of this
testimony, its relevance to the rest of our evidence for the prosodion as a lyric genre is very
37 The treatise is edited with a detailed introduction and commentary by Cupaiuolo 19922.
38 See Dale 1950: 17-20. 39 De vita contemplativa 84 �cbpWbn�gQfW] fgfÓ _ÓQ fÏ c[g]¦kbR, fgfÓ kÓ fÏ ]fn]b_R, ]f[gdn] fW fÏ] �Q Zg[W^Rb jRÚ �Qfb]f[gdÏ] cgbg�_WQgb. See above, pp. 65f.
89
much open to doubt.40 Nonetheless, we have seen independent evidence for the sequential
performance of cult songs regarded as belonging to different genres, such as Cleocharesí
ìprosodion and paean and hymnî and possibly Limeniusí cR]�nQ Ö jRÚ c[[g]¦]kbgQ in
Delphi (F.Delphes III/2 78, CID III 2):41 although the state of things may not have been so
neat as the etymologist would have it, and there definitely seems to have been a great deal
of creative confusion in his Byzantine mind, we cannot discount out of hand all the
information he provides after section 2 simply because of its dubious pedigree.
One point to be noted is what Ian Rutherford terms the ìvicariously processionalî
performance scenario sketched in P22.4, P23.4, and P24.4. Here, it is the sacrificial
victims, not the chorus, who are said to be processing towards the altar while the prosodion
is sung.42 Rutherford rightly remarks that this allows for the classification as Prosodia of
songs that show no sign of processional performance (compare P14) ñ if, that is, this
notion can be traced as far back as either the classificatory principles of the Alexandrian
editors of lyric or sound scholarship based on the results of their work.43 As we shall see in
the next chapter, this may have been correct, at least in some cases.
ODDS, ENDS, AND LATIN
Under P26 to P29 are collected four lists of lyric genres that simply mention the prosodion
as one type of cult song, starting from Proclus and finishing with one of only two extant
40 See, however, Grandolini 1986: 265, 1987-8: 50, who does not seem to perceive a problem or to question the historical accuracy of this testimony. Contrast e.g. Di Marco 1973-4, who makes no use of it in his investigation of the hyporcheme.
41 See pp. 59f. above.
42 Rutherford 2003: 718f.
43 ibid. 719.
90
references to prosodia in Latin literature, in Pomponius Porphyrioís commentary to
Horaceís Epistles. We have already seen that John of Sardisí commentary to Aphthonius
(and through him the later H-scholia) probably draws on the Chrestomathy; now the
coincidence between the four names that he gives and the first four that we find in Photiusí
summary strengthens this conclusion. There is no particular evidence that the entirety of
this list goes all the way back to Didymus, although it is quite natural to suppose that his
treatment of individual genres may have been preceded by some such list. Whether
Porphyrioís note is also to be related to the Proclean tradition is a matter for speculation:
the four genres he mentions are the same as those John of Sardis picked from Proclus (and
also as the first four listed by P32), but there is room for arguing that they may have been
intuitive enough to be independently chosen as examples by two unrelated sources.44
Nonetheless, there is no evidence on which to deny a link, especially if such an
authoritative figure as Didymus stands behind Proclusí list.
Julius Pollux (P30) is the only extant source that mentions explicitly one of the most
obvious data that the ëdirect traditioní of the genre revealed and which F1, *F4 and F11
will once more display, namely its long-standing connexion to Apollo and Artemis.45 A
connexion with Apollo is also implicit in Proclusí remark that prosodia are sometimes
inappropriately called paeans (P31), a notion similar to one which was later to occur to
Demetrius Triclinius as he was commenting on Aristophanes (P18). There are indeed
several examples of overlap between paean and prosodion in the Alexandrian edition of
Pindar, from the cRbÏQ c[g]gkbRj¦] that he reportedly composed for the Ceans (inscr. I.
44 It goes without saying that if Porphyrioís reliance on the Chrestomathy (or an intermediate source) could be established, a fourth-century date for its author would become untenable. On the date of Porphyrio see Helm in RE XXI/2 (1952) 2412-16 s.v. Pomponius 106.
45 See pp. 104-7 below.
91
1 p. III 197 Dr.) to Paean 7, which may have also been titled c[g]×É�[Rj¦] if not even
c[g]¦�É�[gQ on H5, to the extreme case of F5, which was classed both among the Paeans
(where it followed a paean composed in the same metre) and among the Prosodia.46 As we
have seen, there are examples of close relationship between a paean and a prosodion ñ and
indeed seemingly of outright identification ñ in the epigraphic record from Delphi;47
Athenaeusí report of the ìpaeans and prosodiaî that the Athenians sang to Demetrius
Poliorcetes also hints to the closeness of the two genres.48
Conversely, the catalogue of genres and their respective addressees found in the
scholia to Horace commonly known as the work of ëpseudo-Acroní (P32) strikingly relates
prosodia to any deity other than those to whom hymns, paeans, or dithyrambs pertain.
While this reflects to some extent the variety of dedicatees that one gleans from Pindarís
Prosodia (but those include heroes as much as gods), the explicit exclusion of prosodia for
Apollo and Artemis is highly problematic in view of P30 and of the evidence we shall
review in the next chapter. In fact, the available evidence suffices (for once) to refute the
scholion conclusively, unless it proceeded from a largely different notion of prosodion, and
a largely different body of texts regarded as prosodia, than those we have examined so far.
The latter is not impossible, but there is a simpler explanation: namely, that the
attribution of prosodia to ìthe other godsî is the scholiastís inference (or his sourceís) from
a text that mentioned those same four genres and related hymns to Zeus, paeans to Apollo
and Artemis, and dithyrambs to Dionysus, but did not provide a similar specification for
46 c[g]×É�[Rj¦]: DíAlessio in DíAlessio and Ferrari 1988: 169 n. 29. c[g]¦�É�[gQ: envisaged and dismissed by Wilamowitz 1922: 187, revived by Rutherford 1997: 5 n. 17. See pp. 102f. below. On F5 and its puzzling status in the Pindaric corpus see below, pp. 228-44.
47 See p. 59 above.
48 See pp. 51f. above. For further thoughts on the mutual relationship of prosodia and paeans in cult practice (including Pae. 6) see Rutherford 2003: 721-4.
92
prosodia. This is precisely what we find in the commentaries to Apthonius, and this may
also have been the case (but with additional genres) in the original text of the
Chrestomathy.49 If our scholiastís source was close enough to these, he may have assumed
that prosodia were composed for whatever deity did not already have their own genre
among those listed.
This is merely a hypothesis, and it would be hard to be any more specific. We have
already noted a possible, if tenuous, link between the Proclean tradition and Porphyrio,
who lists the same four genres; he and pseudo-Acron must have shared sources, at least in
terms of earlier commentators of Horace (the real, lost Helenius Acron, to name but one);
thus it is probable that these entries are at least connected to each other, and perhaps to the
Greek material too, even though the details cannot be traced, or the connexion itself firmly
ascertained, on the present state of the evidence.50
The notion that the palimbacchius was also called c[g]gkbRj¦] on account of its
frequent use �Q fg~] c[g]gk^gb] —_Qgb] (if indeed the text has been correctly restored), as
Choeroboschus and the Byzantine ëfifth bookí of the so-called B-scholia to Hephaestionís
Handbook on Metre contend (P33-P34),51 cannot be persuasively confirmed on the
existing evidence for prosodia, which in any case is unlikely to have been known in any
significant quantity at such a late date. There is a possibility that the phrase was meant in a
vaguer sense, ìthe so-called processional hymnsî rather than the specific kind of lyric
49 Photiusí summary of Proclus does not mention any special connexion of hymns to Zeus, but John of Sardis and the later scholia to Aphthonius do (pp. 119f. Rabe2, Rabe 1907: 566), thus creating the suspicion that it may have occurred in the Chrestomathy but was simply neglected by Photius in the process of summarising.
50 Note how Pseudo-Acro also shares with Proclus and his descendants, albeit in a slightly distorted form, the etymology of dityrambus from k^] and p�[R: in carm. 4.2.11 p. I 329.24-6 Keller, cp. Procl. Chrest. 42 Severyns, Joh. Sardianus p. 119 Rabe2, | Aphth. Prog. 21.8 ap. Rabe 1907: 566.
51 On Choeroboscusí commentary see Hoerschelmann 1881: 282-300, in Studemund 1886: 88-96, Consbruch pp. XVII-XIX; on the B-scholia see Hoerschelmann 1881: 265-82, 1888, Consbruch pp. XIV-XVI.
93
compositions which other sources know (and define) as prosodia; the reference to the
Dionysiac processions that follows immediately may be thought to offer some support to
this hypothesis. However, it seems more plausible to regard the entire statement as
autoschediastic, like several other supposed elucidations of metrical terminology in both
Choeroboscus and the B-scholia. It is an easy conjecture that a metre called cg_cWPfbj¦]
must have some kind of connexion to cg_cR^ and, likewise, that one called c[g]gkbRj¦]
must have some kind of connexion to c[g]¦kbgb —_Qgb; the reference to the Dionysiac
processions can perhaps be explained with reference to the other and more common name
of the metre. If this is the case, nonetheless the reason behind the metrical use of the term
c[g]gkbRj¦] ñ whether in this sense or in the more common one of ëprosodiací ñ remains
quite mysterious.
95
(6)
Genre and book : Pindarís Prosodia
The criteria followed by Aristophanes of Byzantium for assigning poems to the Prosodia
are quite unclear, due to the extremely fragmentary state both of the Prosodia themselves
and of their ëcompetitorsí, the other books of Pindarís cult songs (most lamentably the
Hymns) whose own criteria of classification might have helped one to understand those of
the Prosodia, at least negatively. The Alexandrian edition of Pindar, much like literature
more generally, is a system none of whose constituents can properly be studied with no
regard for the others; the absence of a comprehensive analysis of the Alexandrian
classification of archaic and early classical Greek lyric ñ or indeed the impossibility of
such an analysis given the mutilated and disconnected nature of the evidence ñ is a
regrettable but all too concrete obstacle to a serious study of any single book.1 Nonetheless,
a few considerations (however partial and precarious) can be made on the surviving
fragments, and it is with these that the present chapter is concerned.
PROCESSIONS
The very name of the genre ostensibly indicates a connexion to an approach (c[¦]gkg]) to
sacred space, as most of the indirect tradition remarks (P1-P11, P14-P18, P22-P25).
Interestingly, as we have seen, not all these sources refer to a processional performance:
1 This point was made, with reference to genres more broadly, by Calame 1974: 113f., 123.
96
the Aristophanes scholia and the Suda are the only ones that connect the name explicitly
with c[¦]gkg], but they refer to nothing more specific than ìthe festivals of the godsî as
the occasion for the performance of prosodia (P14-P15, P17), while the tripartite scenario
of sacrificial song alleged by Simeon, the Etymologicum Magnum, and the Magna
Grammatica envisages such performance while the offerings ñ not the chorus itself ñ are
approaching the altar (P22.4, P23.4, P24.4). If prosodia are mentioned in the Birds in
anything like their generic sense, they are also connected to a procession, and the
opposition between c[g]¦kbR and ]fn]b_R in Philo demands a similar link.
It must be further noted that, although in practice the meaning and usage of
c[¦]gkg] are indistinguishable from cg_cq in our sense of ëprocessioní, the composition
of c[¦]gkg] arguably entails its possible usage in the broader sense of ëapproachí to a
sacred place, whatever the distance covered, and whether or not in the shape of a
formalized procession.2 This becomes relevant in respect to practices such as theoria, the
inter-state sending of ritual delegations.3 We would not call it a ëprocessioní any more than
a Greek would have called it a c[¦]gkg] or a cg_cq, but several compositions known to
ancient sources as c[g]¦kbR refer to it: such are the Delian prosodia of ëEumelusí for the
Messenians and of Pronomus for Euboean Chalcis, and also the paeans-and-prosodia of
Athenaeus and Limenius, which reached Delphi at the end of a land journey, the Pythaís,
which is easily construed as a long, articulate procession.
2 On c[¦]gkg] and cg_cq see e.g. Dover 1968: 142. Rutherford 2003: 713 n. 3 wonders whether the wordís other meaning ìtributeî (more correctly ëincomeí or ërevenueí, LSJ s.v.) may be in the background, giving c[g]¦kbgQ the additional sense ìcontributoryî (used as a noun in F.Delphes III/4 355.44 + SEG XXVII 79, I.Cret. I xvii 2.B.8). This may be the case, but I wonder whether the common notion that a song is a gift or offering to a god (on which see the seminal study of Depew 2000 after Svenbro 1984 and Pulleyn 1997) can be stretched so far as to call it the godís revenue, which shifts the perspective considerably. 3 On theoria see in the first instance Rutherford 2000c, 2004a, 2005: 325-37, and in NP XII/1 (2002) 398-400.
97
Some of the fragments of Pindarís Prosodia also suggest processional performance
or performance during a procession.4 The opening lines of F3 are hard not to interpret as
describing a procession of mythical figures (3 ³SgQfí, 4 ãcWfR[b, perhaps 9 ¯[ZWfRb),
either actually enacted by people or images or merely called to the mindís eye by the song
itself. A È[¦]gkgQ seems to be mentioned in F15.9:5 in principle, the financial sense of
ërevenueí (perhaps used metaphorically) is as possible as the religious one, but proximity
to 10 Zg[¦Q and 11 c]Y_cWb strongly suggests the latter. A hint of processional
performance might also be found in *F6(b), where the reference to carrying (5 ]dg[b¬Q,
see n.) can be taken to imply that motion was involved, although the lack of a context
precludes certainty as to the capacity in which it was mentioned. Stepping onto much
thinner ice, one could perhaps argue that the prayer for reception at the end of F5 (60f.
kY£Xí) may have been taken to indicate that the chorus had not reached their destination
until that moment, although this seems like a stretch by any account. The performance
scenario that Giambattista DíAlessio envisages for F9 also involves a procession, although
there is a danger of some degree of circularity if the classification of the fragment as a
Prosodion played a part in his conception of this hypothesis.6 If F8 was destined for the
4 In the rest of this chapter I will use the phrase ìprocessional performanceî as a short-hand for ëperformance related to a procession, whether or not the chorus itself was processingí: although presented in a rather dubious frame, the etymologistsí view that a chorus could sing a prosodion to accompany a procession without taking part in it deserves serious consideration (P22.4, P23.4, P24.4, see Rutherford 2003: 718f.). It goes without saying that in the present section I focus on possible traces of processional performance embedded in the texts themselves, not on any external evidence that the Alexandrian classifier may have had on the subject: firstly, it can legitimately be doubted whether he could access such detailed information on the practicalities of an odeís performance (there may be reasons to think that, on occasion, he had information about the provenance of a song, but this is quite a different matter); secondly, if any such information existed, it is beyond our reach in any case. I hope that my discussion will show that the classification of an ode among the Prosodia did not necessarily entail a firm judgement on its mode of performance; see already Rutherford 2003: 716. 5 DíAlessio 1997: 28, see also Rutherford 2001: 372f. 6 DíAlessio 2004: 115-21, partly anticipated at 1997: 36f.
98
Theban cult of Heracles (an attractive if undemonstrable hypothesis),7 it may be worth
recalling Isocratesí statement that the Thebans worshipped him jRÚ fR~] c[g]¦kgb] jRÚ
fR~] pP]^Rb] _¬VVgQ ¢ fgˆ] pWgˆ] fgˆ] ³VVgP] (5.32). Likewise, if F16 is taken to refer
to the cult at the Boeotian Ptoion, it is quite easy to construe the song as processional.8 If
the remit of c[¦]gkg] included theoriai, or if a theoria could be assumed to have entailed
a procession in the narrower sense anyway, one can add F11, which was composed with
some likelihood for a Naxian theoria to Delos, and perhaps F1, although its place of
commission is unknown.9 F2.11 mentions a pWR[^R, which Ian Rutherford relates to such
an inter-polis delegation, but the ambiguity of the passage in which the term occurs does
not put the reference beyond doubt (see n.).
Nonetheless, processional performance cannot have been the primary, non-defeasible
criterion for inclusion among the Prosodia. On the one hand, in Pindarís work there are
several processional poems which were not so treated: Partheneion 2 was explicitly
7 See already DíAlessio 1997: 38 n. 95, 2004a: 114; I develop this hypothesis at greater length in the introduction to F8, which could not be included in the present dissertation due to space constraints. DíAlessio 2004: 114f. ñ largely on the basis of his arguments on the performance context of F9 (see the previous n.) and the alphabetical arrangement of the Prosodia, with the latter of which I am less than comfortable (see 2013: 56-8 and pp. 20-3 above) ñ argues for an Argive context. W∞ Z[WàQ fgtpí Ú_WfY[R] �cÙ SVi]]R] jgbQÙQ W–XR]pRb ¯cg], one hopes that some more information on Theban cults, if not actually on Pindarís relationship to them, might perhaps be garnered from the late-sixth and early-fifth-century bronze tablets that have been recently excavated in Thebes (as reported by Nikolaos Papazarkadas at the colloquium ëThe Boiotian Fourth Centuryí at Corpus Christi College, Oxford on 25th May 2013; see already Aravantinos 2010: 166f., 233) once they reach publication. 8 On F16 and the Ptoion see DíAlessio 1997: 31f. On the two sanctuaries and the cult of Apollo Ptoieus see especially Guillon 1943 (publication of the tripods and bases, with a fulsome discussion of the cults at I 87-174), Ducat 1971 (edition of the kouroi and remaining finds, with a historical prCcis at 439-50), Lauffer in RE XXIII/2 (1959) 1506-78 s.v. Ptoion, Schachter 1981-94: I 52-73, III 11-21, and, with special emphasis on the entry-way into the sanctuary of the hero Ptoios, Giannisi 2006: 37-41. 9 F11: Naxian involvement and a Delian connexion are demanded by the text itself (5-8 and the myth of 8ff.), but there may be an outside chance that F11 was composed for a local cult of Delian Apollo (if there was one on Naxos) rather than for an actual mission to Delos. At any rate, even if they are taken to refer exclusively to the past and have no reflection on the present song, ll. 5-8 do assert a theoric connexion between the two islands. F1: see pp. 176f. below.
99
processional (66-71, cp. the descriptions of the rite in Proclus and Pausanias);10 the same
was probably true of Paean 7 (10-12 _gb Ö ∞¦Qfb fWVRPSYí ðS jg[PdnQ) and perhaps
also of Paeans 2 (3-5) and 4 (6 jRfÏ c¬]RQ ¡k¦Q, 11-13).11 The comparison between the
speaker and Xenocritus (?) of Locri might potentially intimate a similar context for fr.
140b Sn.-M. if the opening word of the poem is rightly read as ∞iQ (not the otherwise un-
Pindaric ºTQ or a cognate) referring to ëXenocritusí, and even more so if the Yale editors
are right to read bT_[ (i.e. ÒT_[WQ?).12 If we extend our scope to include theoria, the results
are similar: Paeans 4 and 5 may have been composed for a theoria sent to Delos by the
respective communities, Ceos and probably Athens, although performance at a local
Delion cannot be excluded.13 On the other hand, not all the fragments of the Prosodia show
clear signs that the respective poems were performed in procession, and it would be unsafe
to blame the fact only on their fragmentary nature: other criteria, one of which we shall see
shortly, must have been in play.14
One suspects that, with regard to processional performance, the Alexandrian
classifier of Pindar may have used the prosodion as a fall-back genre of sorts: not every
poem regarded as performed processionally will have been regarded as a prosodion, and
accordingly edited among the Prosodia, if it could plausibly be attributed to another genre
10 Procl. Chr. 69, 74-78 Severyns, Paus. 9.10.4. 11 The argument for taking Pae. 2.3-5 as a reference to processional performance goes back to Wilamowitz 1913: 247, see also Lefkowitz 1963: 186, Rutherford 2001: 266f. However, it seems to me that the point at stake is not the speakerís envisaged physical progression from place to place, but that of his topic, starting from the local hero Abderus and moving on to Derenian Apollo and Aphrodite (Radt 1958: 26f., and already Wilamowitz l.l.). 12 bT_[: Oates, Samuel, and Welles 1967: 38. However, it cannot be confidently excluded that the incerta 140a-b Sn.-M. (from H11) may be Prosodia, see pp. 14f. above. 13 On Pae. 4 as a Cean commission for performance at Delos see, among others, KÅppel 1992: 141-51, Fearn 2011: 228-31; based on the possible connexion with I. 1 (see n. 24 below) Rutherford 2000b, 2001: 292f., following Grenfell and Hunt 1908a: 18, suggest a performance on Ceos herself. On Pae. 5 see p. 177 n. 15 below. 14 So Rutherford 2003: 716, 719.
100
(such as a paean or a maiden-song); but if there was no such obvious alternative, then
processional performance may have been called upon to warrant inclusion in the Prosodia.
However, this only shifts the terms of the question: even assuming that performance by a
chorus explicitly identified as one of maidens was a primary criterion for classification as a
partheneion, and thus that any song explicitly performed by maidens in procession would
be edited in the Partheneia and not in the Prosodia, what is the tipping point for a song to
be regarded as a paean and thus (on this interpretation) not a prosodion?
The criteria that Aristophanes of Byzantium used for classification of certain poems
as Paeans are not entirely clear. One was probably the presence of the paean-cry ∞U cRbnQ
in its various shapes (Pae. 2, 4, 5, 6(a)),15 which none of the Prosodia can be seen to have
featured: ∞U ∞Y in the refrain of F9 is no counter-example, since it does not contain the
cRbnQ-word that is essential to the paean-cry proper.16 *F6(b).4 mentions paeans, but the
fragmentary context makes the significance of the word impossible to determine; *F5
explicitly calls itself a paean and ends with an invocation to HRbnQ (5, 60), and indeed it
was also ñ extraordinarily ñ double-classed as a Paean.17
PAEANIC INTERFERENCE
The mutual relationship between paean and prosodion is particularly complicated, as
Proclus and, one presumes, already Didymus realised (P31).18 Aristophanesí Birds provide
15 On the paeanic refrain as a generic marker see SchrNder 1999: 50-3, Rutherford 2001: 68-72. 16 See DíAlessio 1997: 37, SchrNder 1999a: 17. 17 On the double classification see pp. 228-44 below. Note, however, that ancient critics need not have taken 5f. g– ]W cRb\¦QTQ £ ³kg[cgQ WÃQnXg_WQ as a reference to F5 itself. 18 It is worth stressing that the double remove at which our testimony stands from Didymus (or whatever other source Proclus used instead of Didymus) makes it hard to pin down the terms in which the original
101
a possible fifth-century example: at 851-8, the chorus proclaims its willingness to send
forth c[g]¦kbR for the gods, and immediately thereafter calls for the HPpbÏ] �gn, which
can easily be construed as a reference to the paean-cry.19 Indeed, in Delphi the two genres
seem to have been closely connected. We know of Cleocharesí ìprosodion and paean and
hymnî, whose mutual relationship is impossible to assess; we actually have most of two
songs by Athenaeus and Limenius called cRbnQ Ö jRÚ c[g]¦kbgQ (CID III 1-2), as we
saw.20 At least in Limeniusí case there is a consensus that the paean and the prosodion
correspond to the two metrically distinct sections of the song.21
Another case, probably the most contentious, of the association of paean and
prosodion is Pindarís Paean 6: H4 presents three triads in an identical metre, the first of
which (1-122) is headed }WVdg~] £ W∞] HPpi, and the third ñ our F5 (123-83), its
beginning marked apart by an asteriskos like that of the first triad ñ bears the title
o∞SbQq[fR]bc £ [W∞]c o∞¥R[jÙ]� £ c[gc[¦]Éb[g]Q. On a unitary interpretation, we have one
poem that presents the same articulation as the paean-and-prosodion of Limenius: a longer
section regarded as a paean (as we must assume given its placement in the book of the
Paeans) followed by a shorter section called a prosodion, although in the same metre as the
first.22 The context of the whole would also have to be Delphic, and more precisely the
festival of the Theoxenia ñ the very one for which Cleochares composed his ìprosodion
source expressed this complicated relationship: Photiusí dry ìsome call prosodia paeansî may go back to Didymusí (?) assertion of a tendency to confuse the two genres, to some scholarís explicit argument that prosodia are only a sub-category of paeans, to the classificatory uncertainty of one or more poems as paeans and/or prosodia (see the rest of the paragraph), or to none of the three. 19 So | 875b.R Holwerda2; see also P18. However, that c[g]¦kbR in this passage refer to songs is not entirely clear: see pp. 46-9 above. 20 See pp. 58-61 above. 21 See p. 59 above. 22 See pp. 228-44 below.
102
and paean and hymnî.23 On the other hand, a non-unitary interpretation such as I will
attempt to develop in the introduction to F5 unpicks the construct of ëPaean 6í and the
ensuing necessity of a strong connexion in composition and performance between what we
might term Paean 6(a) and F5, but nonetheless encourages the question of what connexion
between the (on this view) two poems their metrical identity may indicate, and how far any
ancient notion of a structural bond between paean and prosodion may have played a part in
Aristophanes of Byzantiumís decision to place the (on this view) two pieces consecutively
and in that order in his edition of the Paeans.
The picture is further complicated by the concept of the ìprosodiac paeanî, cRbÏQ
c[g]gkbRj¦]. This label is applied by the Pindaric scholia to the Delian paean composed
for Ceos to which Pindar appears to refer in the opening of Isthmian 1, and which several
modern critics identify with the only known Cean composition of Pindarís, Paean 4.24 As
we know from H5, a qualification that began with c[g]¦¥É�[was also appended to the title
of Paean 7: the word may have been again c[g]×É�[Rj¦] (DíAlessio) or perhaps
c[g]¦�É�[gQ as in the title of F5 (Rutherford),25 but either way it is certain that there was a
reference to prosodia. It may have been the case that all, or several, poems in that section
of the book of the Paeans (the one nearest to the end, section A of H4) were characterised
by some degree of hybridisation with the prosodion, even if not taken to the extreme of
double classification, as F5 was;26 however, the title of Paean 6(a) on H4 shows no such
23 See p. 57 above. 24 inscr. b I. 1, | 3 pp. III 196f. Dr.; identification with Pae. 4 first proposed by Grenfell and Hunt 1908a: 18. 25 c[g]×É�[Rj¦]: DíAlessio in DíAlessio and Ferrari 1988: 169 n. 29, plausibly arguing that the title of Pae. 4 was likewise æW^gb] W∞] }vVgQ c[g]gkbRj¦]. c[g]¦¥É�[gQ: envisaged and dismissed by Wilamowitz 1922: 187, revived by Rutherford 1997: 5 n. 17. The parallels make both of these likelier than c[Ù] �Ï¥ ¥[ indicating the festival, as alternatively suggested by Rutherford 2001: 343 n. 13 after Snell 1938: 427. 26 See DíAlessio 1997: 30f. and p. 19 n. 40 above.
103
qualification.27 What this may have entailed is not entirely clear. If the adjective can be
taken to have meant ëprocessionalí, one only need to recall our earlier observation about
the likely processional performance of Paean 7 and perhaps also of Paean 4 (and 2, which
may have also been c[g]gkbRj¦]) as well as the likely connexion between Paean 5 (and
4?) and an inter-polis theoria.28 Paean 6(a) itself, with the speakerís claim to have come to
the place of the performance (7-14, 60f.), could be a plausible candidate for such a group,
despite the lack of a specification to this effect in the title. If this is correct, one may
suppose with DíAlessio that when a Pindaric poem displayed markers of both paean and
prosodion ñ such as a paean-cry, or the like, and explicit references to processional or
theoric performance ñ it could be regarded by the Alexandrian editor as a ëprocessional
paeaní and classed in the appropriate sub-section of the Paeans.29
One further piece of evidence remains to be discussed: H26 fr. 94. A marginal note,
apparently referring to the line to its right, ends with ] gkbgQ: as the first editor saw, the
supplement c[g]] ¦kbgQ is inevitable.30 However, as DíAlessio noted, Lobel cannot be right
to argue on this basis that fr. 94 belonged to the Prosodia: if anything, the presence of a
generic indication to this effect should prove that the fragment does not come from the
27 The hypothesis of a block of cRb¬QW] c[g]gkbRjg^ (or similar) that included Pae. 6(a) can be squared with its title in two ways: (i) if Aristophanes of Byzantium envisaged two independent compositions, Pae. 6(a) and F5, but wanted to place them next to each other on account of their metrical identity (or of any other connexion which he saw between them), then the fact that he regarded F5 as a c[g]¦kbgQ necessarily entailed the dislocation of Pae. 6(a) among the c[g]gkbRjg^ even though it would not have been regarded as such on its own right; (ii) if Pae. 6 was already regarded as a unity in antiquity (which is dubious, see pp. 228-44 below), then the fact of being for two thirds a paean and for one third a prosodion would surely make the whole qualify as a cRbÏQ c[g]gkbRj¦], an artefact similar to the poem of Limenius (comparison made by Rutherford 2001: 329). 28 See pp. 99 above. 29 DíAlessio 1997: 30f. 30 Lobel 1961d: 69.
104
Prosodia,31 but from another of the books represented among the fragments of this
manuscripts, either the Paeans or the Hymns. DíAlessio also pointed out that the note
cannot be a title, since titles on H26 are written in the column, not in the margin.32 What this
note could be is a bit of a puzzle, as indeed is the whole of this scrap of papyrus. Simply
put, although we can be certain that something in or near fr. 94 had something to do with
prosodia, the evidence of this tantalizing scrap yields no clue as to any details of this
probable generic overlap.33
DEDICATEES: APOLLO AND ARTEMIS
Pollux claims that the typical dedicatees of prosodia are Apollo and Artemis (P31), and
while this is not a complete picture of the truth, the presence of the two deities in poems
called prosodia is indeed notable; it hardly needs saying that the overlap between paean
and prosodion just discussed could hardly have taken place otherwise. The prosodia of
ëEumelusí, Pronomus of Thebes, and Amphicles of Rheneia were destined for Delos;34
Cleocharesí ìprosodion and paean and hymnî were to be performed at the Delphic
31 DíAlessio 1997: 27, cp. the title of F5 on H4; but see H45 fr. 1. 32 ibid. n. 20. 33 The line of the main text (beginning VW[) that stands to the right of this note begins somewhat to the right of the first trace of ink that survives in the line below it, which may have started even considerably further to the left than this trace; the uppermost surviving line of the main text (which is spaced away from the line beginning VW[ by roughly the height of a line and a half) did start considerably further to the left than VW[, by the breadth of at least four letters. Lobelís hypothesis that VW[ was the beginning of a poem (1961d: 69) seems to me untenable in view of this misalignment. Rather, it can perhaps be regarded as the beginning of a title, indented slightly further to the right and more widely spaced away from the end of the preceding poem than the other titles on this papyrus (see p. 132 below); but I see no way to make this hypothesis particularly persuasive. In any case, it leaves us just as puzzled as before: a poem composed for the Le[uctrians? the Le[ontinians? Neither of these seems particularly compelling. Furthermore, the position of the notes ñ there is another, longer one, penned by a different hand, in the blank above c[g]] ¦kbgQ ñ is quite anomalous for this manuscript; and if the upper note could extend so far as to bridge the entire intercolumnium and reach the next column (that is, ours), which seems the likeliest possibility, there is no reason why this should not be the case for the note that ends with c[g]] ¦kbgQ as well. 34 See pp. 39-46, 57 above.
105
Theoxenia, and the paeans-and-prosodia of Limenius and of Athenaeus were composed for
the Pythaís that Athens sent to the same sanctuary.35 The prosodion that the contestants
were required to sing together at the Eretrian Artemisia will also have been connected with
the worship of the festivalís titular deity, and a similar argument can be made regarding the
prosodia that were performed competitively at the Delphic Soteria.36
Among Pindarís Prosodia, F1 opens a song with Artemis and Leto, and it seems
likely that Apollo too was mentioned after the short section that survives.37 F11 was
probably composed for a Naxian theoria to the sanctuary of Apollo on Delos and recounts
the myth of the twinsí birth on the island, just like Paean 7b.38 F15, which mentions
Ê]RfgÔkRb�[ (4) before a reference to a procession and a chorus (9f.), can plausibly be
linked to an Apolline cult, perhaps on Delos herself.39 If F16 truly indicates that the ode
was composed for the cult at the Ptoion, there is a good chance that the cult was that of
Apollo.40 It is likely that the mention of paeans at *F6(b).4 either constituted the songís
explicit self-definition or occurred in the context of a comparison between the song itself
and others (see n.), which in turn suggests that the comparison hinged on their similar
connexion to Apollo. The cultic status of F5 is contested, but there is no question that the
closing prayer (59-61) addresses Apollo as HRbnQ.41 Finally, if *F4 is rightly attributed to
35 See pp. 57-61 above. 36 See pp. 56f., 61f. above. 37 See p. 174 below. 38 See p. 98 n. 11 above. On the myth see Rutherford 1988. 39 DíAlessio 1997: 28. 40 DíAlessio 1997: 31f. 41 See pp. 252f. below.
106
the Prosodia, we have a further example of a Pindaric Prosodion dedicated to Apollo, this
time to his worship at Delphi.42
On this evidence, Polluxís �_dgfY[TQ (sc. of Artemis and Apollo) is probably to be
understood as offering an alternative, with the prosodion encompassing songs dedicated to
one of Artemis or Apollo as well as songs dedicated to both jointly. There is no conclusive
evidence for a Pindaric Prosodion in the honour of Artemis alone: she is probably the
referent of F18.3 �c]cg]¦R pP[Snf\[, but there is no reason to assume that this made her
the (only) dedicatee of the ode. On the other hand, it is only Apollo whom F5.60
addresses; *F4 too, if a Prosodion, appears likely to have been composed for the worship
of Apollo specifically, given Artemisí relative lack of prominence in Delphi,43 and
similarly the Apolline cult at the Ptoion (if it did provide the context of F16) does not seem
to have given much space to the godís sister.
Despite the high proportion of Apolline-related poems in the Prosodia, it is unclear
how much of a genre-defining status an association with Apollo or Artemis had. Not only
can a poem dedicated to Apollo be classed as a Paean, sometimes with a conspicuous
blurring of boundaries, as we have seen; it seems to have been the case that the Hymns too
included compositions for Apollo.44 In the absence of more information concerning the
latter book and the criteria that warranted inclusion in it, it is hard to judge when a song to
42 See p. 20 n. 42 above. 43 See p. 177 and n. 14 below. 44 A Hymn of Pindar to Apollo ñ probably related to his cult at the Ptoion, since it mentioned the parentage of the hero Ptoios ñ is mentioned by | Paus. 9.23.6 (fr. 51c Sn.-M.); on the dubious pertinence to it of frr. 51a-b, d Sn.-M. see Wagman 1986a: 111f., 125f., 1986b: 397 n. 1, DíAlessio 1997: 31f. (arguing that they may conceivably come from the same poem as F16), and, for a reading of the fragments, Wagman 1986a, 1986b, Olivieri 2004, 2011: 205-13 (who ignores DíAlessioís discussion). Also the first Hymn (on which see p. 10 n. 3 above), commonly regarded as a hymn to Zeus since Snell 1946, is likely to have been dedicated to Apollo: DíAlessio 2005, 2007, 2009b (contra, Bernardini 2009).
107
Apollo was deemed to be a hymn and when a prosodion, unless evidence (or conjecture)
on its performance could be called upon to settle the question.
DEDICATEES: HEROES
The other, and more distinctive, group of dedicatees that can be found in Pindarís Prosodia
is heroes.45 According to their titles, F3 and F5 were composed for the Aeginetan cult of
Aeacus, and *F7 probably for the cult of the sons of Electryon in Argos.46 F8 narrates in
great detail an episode from Heraclesí infancy, and while there is a possibility that the
dedicatee was different, the usually close connexion between the subject-matter of the
mythical narrative of Pindarís cult odes and their respective dedicatee strongly suggests
that the composition from which F8 comes was dedicated to Heracles.47 The case of F2 is
quite complex: the ô[T�kg] £ pWR[^R] mentioned at lines ii.11f. can be interpreted in too
many different ways (see n.) to provide a reliable clue towards the specific cultic context of
the ode, but whatever it is taken to mean ñ a theoria for heroes, a theoria of heroes, a
vision of heroes, or something different again, which the song explicitly undertakes to
bring to mind even to one who lives far away (ii.10-12) ñ it is hard to escape the
conclusion that the occasion for F2 involved hero-worship in one form or another. Finally,
although the cultic context of F12 is impossible to determine given the disconnected
45 DíAlessio 1997: 41f., 1999: 19, see already the title of Rutherford 1992. 46 On the title of *F7 see pp. 307-10 below. It goes without saying that what matters for our inquiry on the criteria followed by the Alexandrian editor of Pindar is his own understanding of the commission and dedication of a given ode, not the correctness of his understanding as judged by our standards. 47 The best alternative (if, still, presumably not a likely one) is Hera; on the matter see DíAlessio 2004: 114. A further, more marginal possibility is that the song honoured Alcmena or Amphitryon: the former had a heroon (Pherecyd. fr. 84 Fowler = BNJ 3 F 84) and a cult (D.S. 4.58.6) at Thebes, while in the fifth century the latterís tomb was the focus of the games that later sources call Heracleia (N. 4.19-21 with Didymus fr. 47 Braswell in | ad loc. p. III 69 Dr., | O. 9.148d, l p. I 301 Dr.): see Schachter 1981-94 I 15f., 30f. If this was the case, F8 would nonetheless remain a heroic Prosodion.
108
remains of its text, the mention of an altar for a hero at (a).1 may suggest that the poem
was somehow associated with a heroic cult.48
This strong heroic presence in the Prosodia is thrown into relief by the apparent lack
of poems dedicated to heroic cult in the Hymns.49 Although our knowledge of that book is
scanty in the extreme, whatever information there is on the destination of the poems it
contained points exclusively to divinities, not to heroes: we know of a Hymn to Ammon
reportedly sent to the Ammonians of Libya (fr. 36 Sn.-M.),50 one to Demeter and perhaps
Persephone, which Pausanias tantalizingly presents as a posthumous work in a very literal
sense (fr. 37),51 and probably one to Apollo ñ long thought to have been a hymn to Zeus ñ
which opened the book of the Hymns, as Lucian testifies (frr. 29-*35c).52 One fragment is
concerned with Heraclesí dealings with Athena (fr. 51e), but there is no evidence that it
came from a hymn to Heracles; the mention of the parentage of the hero Ptoios in fr. 51c
suits a hymn to Apollo just as much as a composition dedicated to Ptoios himself.53
48 DíAlessio 1999: 19. For specific hypotheses on the cultic destination of F12 see Zuntz 1935: 291 (the Argive Agrionia), Rutherford and Irvine 1988: 51 (the Athenian Oschophoria, dubitatively), Ferrari 1991b: 385f. (a Dionysiac ritual); on the text and its interpretation see all of these, Snell 1940: 185-91, Bona 1988: 247-9, DíAlessio 1997: 32f. and 1999, and Rutherford 2001: 420-2. 49 Noted by DíAlessio 1997: 41f. and Rutherford 2001: 164. DíAlessio points out that, conversely, |L D.T. explicitly mentions heroes alongside gods as possible objects of �Sji_bR in hymns (n. 113), but the formulation is vague enough to cover also the praise arguably bestowed by the mythical narratives in which heroes appeared, without requiring the respective poems to be concretely dedicated to a hero in cult, if indeed the distinction mattered to the scholiast at all. (But see also n. 58.) 50 Paus. 9.16.1. 51 Persephone: Paus. 9.23.3f., where Pindar is said to have composed the hymn at the goddessí command after his own death and dictated it to a female relation of his in a dream. Demeter: VA p. 2 Dr. (whence Eustathius 27 Kambylis = p. III 299 Dr.). The similar anecdote on the composition of the ode and the coincidence of the word Z[P]nQbgQ (which Pausanias normalizes as Z[P]qQbg]) indicate beyond doubt that the two sources refer to the same poem, which makes the divergence perplexing. Lehnus 1973b: 5-11 plausibly argues that the hymn was addressed to Demeter (since the epithet rW]_gd¦[g] in the singular can only apply to her), that the opening section mentioned Persephone immediately after her, and that Pausanias was misled by the legend of the post-mortem composition into associating the poem specifically with Persephone, whose otherworldly undertones came more readily to mind. 52 See n. 44 above. 53 ibid.
109
If this was truly the case, it becomes possible to suggest that a poem taken to have
been composed for the worship of a hero would have been included in the Prosodia by
default. The double classification of F5 among the Paeans and among the Prosodia might
confirm the notion that, whereas processional performance was only a secondary criterion
for identification as a prosodion, one which could be overruled by some stronger
consideration, pertinence to hero-cult was instead a primary marker of such identification:
faced with a song which openly advertised itself as a paean and at the same time was
understood as being dedicated to a hero, Aristophanes of Byzantium would have found no
better option than to rely on both features equally and edit the ode both as a Paean and as a
Prosodion. However, the double classification of F5 is probably more complex than this:
specific text-historical issues ñ first and foremost its relationship to the Paean ìfor the
Delphians to Pythoî that precedes it on H4, to which I shall refer as Paean 6(a) ñ must
have played a part alongside more general considerations on what makes a prosodion.54
The suggestion that an association with heroic cult had a genre-defining status, however
attractive on the existing evidence, rests substantially on the relative frequency of poems
dedicated to heroic cult in the Prosodia in comparison to the Hymns, and is thus open to
refutation if further evidence for the Hymns themselves comes to light.
The substantial heroic presence in Pindarís Prosodia is made all the more remarkable
by another factor, namely the absence of any suggestion of prosodia dedicated to hero-cult
anywhere else in our documentation. As far as the non-Pindaric direct tradition goes,
54 On the relationship between Pae. 6(a) and F5 ñ or, according to most scholars, between the first two triads of Pae. 6 and the third ñ see the introduction to F5, pp. 128-44 below. The case of F5 arguably also precludes the hypothesis that individual poems for hero-cult were concentrated among the Prosodia only because they could not be classed among the Hymns, which only included poems dedicated to divinities: if this had been the case, since an alternative classification was available in the case of F5, it would only have been classed as a Paean. Whatever it was, there must have been a positive reason for including F5 (and presumably, by consequence, poems for hero-cult as a group) in the Prosodia.
110
whenever the cultic context of a prosodion is identifiable, it is concerned with a deity:
Apollo, Artemis, and perhaps the Muses at the Thespian Musaea. What is at stake in
Demochares BNJ F 9 is patently divine, not heroic, status.55 The indirect tradition similarly
shows no obvious trace of awareness of prosodia for heroes. A distinction, however, needs
to be made. Whereas several testimonies refer specifically to gods (P13-P17, P19, P30,
P32), most of those that go back to Didymusí HW[Ú VP[bj�Q cgb\f�Q do not: what the
singers of a prosodion approach is simply QRg~] ¢ �T_g~] (P1-P11, P22-P24, but note
P26-P29), and nothing in the text precludes that the altars be those of heroes as well as
than those of gods and goddesses. If, as one reasonably suspects, Didymusí overview of
lyric genres was based to a large extent on the Alexandrian editions of the lyricists, and if
the phrase in question is not due to the process of epitomization and compression that his
words underwent before reaching the intermediaries that transmitted them to us, he may
have deliberately used vague language in order to reflect the mix of addressees that he
recognised in his sources.
It is less clear why only the gods are specifically mentioned by most of the other
sources. It may simply be the case that the distinction was not thought to matter much
(which could be especially true of some more derivative testimonies, whose authors may
have had no contact with actual prosodia), or that the mention of the more important group,
the gods, was thought to include automatically the closely related lesser one. Some ancient
critics may also have thought that the ëtrueí prosodia were those sung to divinities, while
those for heroes were to some extent extravagant, less close to the spirit or idea of the
genre: this might be a useful way to account for the specific and exclusive mention of gods
in otherwise apparently well-informed sources such as the scholia to Aristophanes and to
55 See pp. 51f. above.
111
Dionysius Thrax (P14, P13). Furthermore, as we have seen, prosodia for heroes are unique
to the Alexandrian edition of Pindar in the surviving documentation: if the balance of our
evidence reflects that which was available in antiquity, some ancient critics may
legitimately have thought that the notion of prosodia dedicated to heroes was an editorís
quirk which did not represent the broader state of prosodiac affairs.
If the association between prosodia and heroic cult was truly limited to the
Alexandrian edition of Pindar (or to the Alexandrian editions of Pindar, Simonides, and
Bacchylides, on the last two of whom there is practically no evidence for this matter), the
question arises why Aristophanes chose to class those songs which he understood as being
dedicated to heroes in the Prosodia rather than elsewhere. On a minimalist approach, it
may have been an editorial more than a literary-critical problem: to include songs for hero-
cult in (say) the Hymns may have resulted in a text whose length was excessive for one
papyrus roll ñ a ëbookí in the Alexandrian sense ñ and insufficient for two, while including
them in the Prosodia instead may have balanced the books more efficiently. However, the
actual amount of such heroic poems is unknown, and if the proportion in the surviving
fragments mirrors that in the original books, one supposes that heroic Prosodia may have
taken up roughly half of the total length of the work, that is to say, roughly a whole book.
Furthermore, the little evidence that is available on the ordering of poems within the
Prosodia does not support the view that heroic prosodia were an appendage of sorts: odes
for gods and odes for heroes were to be found side by side, with no obvious partition and
no fixed hierarchical order (probably hero-goddess at F8-F9, god-hero at *F6-*F7 and
perhaps at *F4-F5).56 The question cannot be answered on the existing evidence.
56 See respectively n. 9 above and p. 112 below; pp. 302 and 308-10 below; p. 20 n. 42 above and pp. 253-5 below.
112
OTHER DEDICATEES
Apollo, Artemis, and heroes cover most of the addressees that we can see in the extant
fragments of the Prosodia. If we consider the eleven best-preserved fragments (F1-F9 and
F11-F12), five are related to Apollo or Artemis, and at least as many (but possibly six,
counting F12) to heroes, with F5 appearing in both groups. Neither group, however, seems
to be able to include F9. Although the train of thought cannot be followed continuously
across the fragment, its refrain hails a ìQueen of the Olympians, wife of the best of
husbandsî. To all appearances, this appellation must refer to Hera: it cannot realistically
indicate any other of Zeusí partners, including Leto, and even less a heroine.57
The conclusion is inevitable that, despite the very high proportion in the Prosodia of
poems dedicated either to Apollo (with or without Artemis) or to heroes, there was also an
unknown number of odes for other deities.58 The fact that only one is recorded does not
necessarily mean that it was an exception: it is possible that F9 was a relatively extraneous
appendage to the Prosodia much in the same way as the last two (?) Isthmians and the last
three Nemeans,59 but although the large amount of fragments that survive from H7 may
suggest that they were originally distributed quite evenly across the book(s) of the
57 Rutherford 2001: 404 warns that ì�[b]f¦cg[]]bQ could refer to any wife of Zeusî, mentions Demeter, and suggests Leto as ìan outside chance Ö considering the paeanic associations of the refrainî. But surely neither is a ìQueen of the Olympiansî: crucially, Ar. Ran. 384 (Demeter) parallels the title, not the realm. 58 In this connexion, it may be worth recalling Serv. Aen. 10.738 p. II 464 Thilo-Hagen paean proprie Apollinis laus est, sed abusiue etiam aliorum dicitur; unde Pindarus opus suum, quod et hominum et deorum continet laudes, paeanas uocauit. Rutherford 1995: 51f. interprets the second sentence as a reference to the mythical narratives that Pindarís Paeans contained: in the surviving fragments, most of them were concerned with heroic rather than Apolline myths, which will have been contained in the more fragmentary first half of the book. The first sentence definitely echoes Procl. Chr. 41 Severyns, but one wonders whether Servius may represent the confusion between paeans and prosodia also testified by Proclus at P31: whereas laudes does not sit too easily with a reference to the mythical narrative alone (and not to the dedicatee of the ode), and the vague plural deorum is perhaps even more striking than hominum with reference to Pindarís Paeans as we know them, the definition would suit the Prosodia quite nicely. (But see also n. 49.) 59 See p. 28 n. 5 above.
113
Prosodia,60 nonetheless the relative paucity of the evidence overall may be to blame for
offering a skewed picture of the whole. In this specific case, inclusion in the Prosodia may
have been warranted by an explicit reference to processional performance, such as the
statue-carrying and statue-washing ritual cleverly reconstructed by DíAlessio from the
likely mention of a fountain at F9(a).10,61 or the even more speculative scenario of a
performance at the Plataean Daedala;62 but these are, once again, little more than
fascinating hypotheses.
OTHER FEATURES
The fundamental features of Pindaric hymnography in general are themselves a worthy
subject of inquiry, but space constraints prevent a discussion in this context. From the
formal and structural point of view there is no particularly distinctive feature ñ such as the
paean-cry for (some) Paeans and the emphatic self-reference of female performers for
(some?) Partheneia ñ that visibly sets Pindarís Prosodia apart from the rest of his cult
poetry. On the existing evidence, the sectioning actualit�, mythe, actualit� that Alfred
60 Note, however, the considerable number of originally separate fragments that Lobel 1961b was able to put back together: this raises the possibility that many other fragments too, despite the lack of physical joins (which necessarily excluded them from Lobelís rearrangement), were originally to be found in close proximity with one another. 61 DíAlessio 1997: 36f., developed in 2004a: 115-21. As I suggested in 2013: 56f. n. 24, the hypothesis of an Argive commission would find further support in Psycheís prayer to Argive Juno prope ripas Inachi, qui te iam nuptam Tonantis et reginam dearum memorat in Apul. met. 6.4, if the passage alluded specifically to the refrain of the ode (Zimmerman et al. 2004: 390 relate it only to Ov. fast. 6.36, 37); but the twin notions that Hera/Juno was the wife of Zeus/Jupiter and the Queen of the goddesses are obviously quite commonplace, and a specific reference is hard to prove. 62 I develop this highly tentative hypothesis in the introduction to F9, which could not be included in this dissertation for reasons of space. It is worth noting, in any case, that nothing in the surviving text itself of F9 guarantees, or even particularly suggests, that it was performed in a procession.
114
Croiset observed in the Epinicians appears to be generally operative in the Prosodia no
less than in the Paeans, Dithyrambs, and Epinicians as well as several smaller fragments.63
Overall, the metres that can be recognised in the Prosodia are not significantly
different from those of the Paeans. They share with these a comparatively low incidence of
dactylo-epitrites: only three of the eleven most extensive fragments are in dactylo-epitrites
(F1, *F4, F12), a figure more similar to the slightly lower one of the Paeans (only Paean 5
and the conjecturally attributed fr. *249b Sn.-M.) than to the far higher one of the
Epinicians (24 out of 46, a proportion of over one in two). Two fragments, however, stand
out. *F7 appears to be composed largely, if not exclusively, of dactylo-anapaestic
sequences that recall the metres of Stesichorus more than anything else in surviving
Pindar.64 Conversely, the scanty remains of F9 reveal short aeolic stanzas, perhaps based
on polyschematist cola (so far unattested elsewhere in Pindar), with remarkably free
responsion from one stanza to the next:65 so free that, despite the presence of a koronis as
well as paragraphoi in the margin, it is unclear whether the song is monostrophic or
triadic, and one scholar has even doubted whether the stanzas respond to each other at all.66
Given the ostensible connexion between prosodia and processions, it is necessary to
restate a point made cursorily by DíAlessio some time ago on to the relationship between
metre and performance. A long-standing scholarly commonplace, which goes back to the
early nineteenth century, maintains that a monostrophic structure is particularly suited to
63 Croiset 1880: 361-6, a notion already sketched by Dissen 1830: lxxi; see now Hamilton 1976 with his sections X (ìBeginningî), Y (ìMythî), and Z (ìEndî). 64 See p. 315 below. 65 Relaxed responsion: Lobel 1961d: 54. Polyschematist cola: Gentili and Lomiento 2003: 188. 66 Lomiento 1998: 113f.; see p. 133 n. 44 below.
115
processional performance.67 This might be so, although I see no particular evidence in its
favout. But the ostensibly complementary notion that a poem that was not monostrophic
could not have been performed in procession cannot be entertained on the existing
evidence.68 As DíAlessio pointed out, in Pindarís work there is at least one poem that was
certainly processional and had a triadic structure: namely, the second Partheneion.69 A
procession could therefore be accompanied by triadic and monostrophic poems alike.
While the evidence from the Prosodia themselves is inconclusive given the fragmentary
state of the material (but at least F5 was certainly triadic), it would be unwise to assume
that most of the poems they contained were monostrophic, or that whatever non-
monostrophic poems they contained were not composed for processional performance but
had been placed in the book due to other characteristics.
An impression that one may have when comparing the extant fragments of the
Prosodia to those of the Hymns ñ especially fragments from poem-beginnings and poem-
ends ñ is that the former refer to the performance and to its context rather more frequently
than the latter do. Excellent examples are the openings of F3 and F7, where the Aeginetan
pageant of mythological figures and the bright portrayal of Argosí poetic traditions firmly
set the respective compositions in a performance context which could not be exchanged
67 Dissen in Boeckh 1811-21: II/2 380 (cp. 1830: 393) made this connexion specifically with reference to N. 4, based on comparison with N. 9: Suspicor carmen cantatum esse in pompa victoris per urbem. Favet huic opinioni metrica carminis ratio monostrophica, qualem habet etiam carmen Nem. IX vix alibi quam in pompa cantatum, neque adest quidquam in toto carmine quod obstet. Within two decades, his suggestion could be treated like a matter of fact, extended to all the monostrophic Epinicians, its conjectural status ñ never mind Dissenís role ñ quite forgotten (Donaldson 1841; 211, cp. 201, 255, but note the inconsistency with 90, 317); as time went on, it was taken up enthusiastically by the likes of Bury (1890a: XXXVI, XLIX, 29, 62, 159; 1892: 163), Christ (1896: LX), Farnell (1930-2: II 264, cp. 310), Puech (19312-522: I 155, II 101f., III 52, 116, 119), and its popularity continues into the twenty-first century (see e.g. Rutherford 2001: 164 n. 2, 2003: 715). It is unclear how much this notion owes to the complementary one that triadic structure was meant to mirror dance and thus, by a simple inference, was unsuitable for processions (the two ideas are associated by Christ and Rutherford): one suspects that the closer the association, the less plausible the outcome, given the dubious ground on which this theory rests (see p. 87 n. 35 above, but note Mullen 1982: 299f.). 68 Cp. Radt 1958: 27 on Pae. 2, KÅppel 1992 : 99 n. 41 on Pae. 4. 69 DíAlessio 1994: 64, see already Farnell 1930-2: II 310 (citing P. 5).
116
with any other.70 Similarly, on one interpretation, the closure of F2 portrays the Muse as
spreading mindfulness of the cult or ceremony to which the song is dedicated.71 Contrast
the only Hymn of which we have a substantial section of the opening, the first (fr. 29 Sn.-
M.): throughout the fragment, the focus is on the songís topic, not on the world around it;
the Theban context is clearly but only implicitly suggested by the array of Theban myths
that the speaker rehearses as possible themes for his song.72
However, F1 is as general an incipit as can be; whether or not the rest of the opening
or the closure of the song referred to its context in any firmer terms is only a matter for
speculation. By the same token, there is no indication that the priamel of fr. 29 Sn.-M. was
followed immediately by the mythical narrative of choice without some intervening
reference to the context in the verses now lost. The impression that an ëaccessional songí
composed to a deity for a certain community may have shown stronger roots on the
ground, as it were, than a ëhymní to that deity is bound to remain speculative unless further
evidence is discovered to prove or disprove it.73 Consequently, it would be rash to suppose
that a composition whose text emphatically connected it to a specific ceremony or location
would have had a greater likelihood of being included in the Prosodia than in the Hymns.
After this extensive inquiry into the generic criteria that Aristophanes of Byzantium
may have followed in assigning poems to the Prosodia, a final remark must be devoted to
countering the easy temptation to assume that there must have been such a specific reason
to class among the Prosodia every individual poem that was so classed. In Ian
70 See pp. 312-14 below. 71 See ii.11f. n. 72 On the first Hymn see p. 10 n. 3 above. 73 One thing that is especially regrettable is the lack of directly transmitted poem-titles of Hymns, compared to the three that survive from the Prosodia (those of F3, F5, and *F7); on the existing evidence, it is not clear that poem-titles always include an indication of the performing community, see p. 16 and n. 28 above.
117
Rutherfordís words, the prosodion may have been ìa default category, to which editors
assigned poems that they could not easily fit into (Ö) other genresî.74 A label like
c[g]¦kbgQ could have been flexible enough to cover any cult poem, if the need arose: after
all, even if the text of a certain poem shows no sign that it was performed in a procession,
it is hard to demonstrate that it was not. Compared with the somewhat stronger personality
of genres such as the paean, the book of the Hymns and the two of the Prosodia may have
been used to account for compositions that had no specific reason to be classed elsewhere;
the distinction between the two, in the absence of determining criteria such as a connexion
to hero-cult, may have been simply a matter of convenience. The matter would probably be
hard to settle conclusively even if we could read the complete text of all poems that
constituted all three books; on the disconnected evidence that we have, there is little
chance of reaching a satisfactory conclusion.
74 Rutherford 2003: 716.
119
(7)
The manuscripts
As we saw in Chapter 2, Pindarís Prosodia are preserved by at least four papyri, plus one
whose attribution is uncertain (H34) and the two papyri of the Paeans that preserve the
most substantial parts of F5: H4 and H5. Like all Pindaric papyri throughout this
dissertation, these manuscripts are referred to by the sigla assigned to them by Bruno Snell
and subsequently Herwig Maehler, and this chapter presents them in the resulting order.1
G4 = P.Oxy. 841 = P.Lond.Lit. 45 verso
MP3 1361, LDAB 3713
Edition: Grenfell and Hunt 1908a
Location: London, British Library Pap. 1842 2
Images: Grenfell and Hunt 1908a pll. I-III (coll. xxiii, iv-v, frr. 82, 128), Turner 1952 pl.
CXXI (col. xxii), Roberts 1955 pl. 14 (col. xxiii, fr. 128), Schubart 19613: 79 (coll. iv-v).
1 On scholarly papyri in general see Turner 1956 and 1968: 92-4, McNamee 1981b and 2007: 37-48, Johnson 2009: 270-7. I borrow the descriptive terminology for the layout of the mss. from Johnson 2004: x (fig. 1), that for the writing largely from GMAW2.
2 The exact distribution of the fragments into frames ñ an essential piece of information for anyone wishing to consult the original ñ seems not to have been published to date except a summary overview in Rutherford, Bagnall and Frier 1997: 2, 5. I provide it here. Frame 1 contains frr. 1-5 (coll. i-vi, viii-ix); frame 2, frr. 6-10 (coll. xv-xx); frame 3, fr. 11 (coll. xxi-xxiii); frame 4, frr. 12-15 (coll. xxv-xxxiii, xxxiv-xxxv); frame 5, frr. 16-81; frame 6(1), frr. 82-7, 95-6, and 107; frame 6(2), frr. 88-125 except those contained in frame 6(1); frame 7, frr. 126-62; the fragments of frame 8 all have a blank verso.
120
The largest extant papyrus of Pindar was found at Oxyrhynchus by Grenfell and Hunt,
together with several other comparatively well-preserved texts, on the evening of 13th
January 1906.3 It contains substantial portions of Pindarís Paeans, written in the second
century AD on the back of two rolls assembled from the cut-and-paste of two documents: a
census register from Lycopolis compiled in the last decade of the previous century and a
land survey from an as yet unidentified location (P.Oxy. 984).4 Grenfell and Hunt divided
the fragments into four groups headed A, B, C, and D and distinguished on the basis of the
writing of recto and verso and the present condition of the papyrus: the groups contain
respectively the fragments 1-15 (from coll. i-xxxv), 16-81, 82-125,5 and 126-139, with
fragments 140-162 belonging to either C or D. The first editors did not commit themselves
to a definite position as to the order of the sections; subsequent studies of the manuscript
have shown that the original sequence was either CD or DC for the first roll, and BA for
the second.6
3 Grenfell and Hunt 1906: 10-12, cp. eid. in The Times 38019 (24th May 1906) 4. Among other papyri found on that occasion they list Euripidesí Hypsipyle (then published as P.Oxy. 852), the Hellenica Oxyrhynchia (P.Oxy. 842), Platoís Symposium (P.Oxy. 843) and Phaedrus (P. Oxy. 1016, 1017), and Isocratesí Panegyricus (P.Oxy. 844). Cockle 1987: 22 suggests that all these come from one scholarís library, but decisive evidence is lacking (Bona 1988: XVI); a palaeographic study of the annotations across these papyri ñ which, to my knowledge, has not yet been attempted ñ could perhaps confirm or refute this hypothesis.
4 Book divided over two rolls, presumably in order to accommodate the wide interlinear spaces and intercolumnia without exceeding reasonable limits of roll-length: Grenfell and Hunt 1908a: 13, DíAlessio 2001: 81-3. Recto: described in Grenfell and Hunt 1908a: 12f., partially published in eid. 1908b; the first complete edition of the census register (P.Oxy. 984a) is Bagnall, Frier and Rutherford 1997; the land survey (984b) is still unpublished. Date: early second century according to Grenfell and Hunt 1908a: 13 (841), Bagnall in eod., Frier and Rutherford 1997: 20-2 (984a); mid- to late second century according to Cavallo 2008: 95. Provenance of 984a: Montevecchi 1998: 49-54, correcting Bagnall in eod., Frier and Rutherford 1997: 22-6 (Ptolemais), 56 (Ptolemais or Lycopolis). Reconstruction of the roll(s): Bagnall, Frier and Rutherford 1997: 1-18, 27-56, DíAlessio 2001. Cockle 1987: 22 and n. 23 remarks that many of the other papyri found in that occasion have a document on the front and a literary text on the back; for an overview of this phenomenon see Lama 1990.
5 But frr. 108 and 112 belong to section A: see respectively DíAlessio and Rutherford in Rutherford 1997: 7f., DíAlessio 1992: 82f. The same may be true of fr. 111: DíAlessio 2001: 72 n. 15.
6 Grenfell and Hunt 1908a: 12f.; DíAlessio 2001 after Snell 1938: 425f. and Radt 1958: 4f., see also Rutherford 1995: 46f., 2001: 140-3.
121
The two rolls were copied by two scribes in two distinctive hands: a thick medium-
sized informal round capital, confident but quite irregular, which has been somewhat
unfairly attributed to a studentís hand (sections AB), and a smaller and thinner specimen of
the same style, with a great deal of finials but on the whole hardly more regular than the
other (sections CD).7 The text was written in columns of fifteen or sixteen lines ñ regularly
sixteen in the fragments of sections CD, more frequently fifteen in sections AB ñ with only
one of seventeen and perhaps one of fourteen.8 Column-height ranges from 107 to 123 mm
(col. xv and the sixteen-line col. xvii respectively), with no discernible difference between
sections. Line-spacing is quite generous, with a leading of normally 8 mm (but often
considerably less at the bottom of the column: see coll. i, v, and the sixteen-lines coll. xxi,
xxix) for a letter-height of about 3 mm in sections AB; the latter figure is reduced to about
2 mm in sections CD, consistently with the increase in the number of lines per column in
the second tome. Maasís Law is regularly not observed, but coll. vi and ix do seem to lean
slightly forward. The margin above col. i reaches 32 mm, and that below col. xv (the
shortest extant), 42. Where both column and intercolumnium survive in their entirety, the
column-to-column width can be measured at 145 mm between coll. v-vi, and 143 between
coll. xxii-xxiii and xxviii-xxix; this is due to an extraordinarily leisurely intercolumnium,
which is almost as wide as the column of writing itself.9
7 Two hands: Grenfell and Hunt 1908a: 13. At work on one roll each: Rutherford in Rutherford, Bagnall and Frier 1997: 3, DíAlessio 2001: 83. Volume 1 a studentís copy: McNamee 2007: 21f. and passim, but the fact that the ìpenmanship is awkwardî (22) is no sufficient proof; the amount and quality of annotation (especially the text-critical material) to which the layout of the manuscript was tailored strongly suggest that at least the owner(s), if not necessarily the scribe, had considerably more advanced interests than ìchildren just past the elementary stage in their educationî (ibid.).
8 Sixteen-lines columns in section A: visibly xvii, xxii, xxix, and demonstrably xix, either xxiii or xxiv, and xxxiii. Seventeen lines: fr. 16 (section B). Fourteen lines: possibly col. xx (section A). See DíAlessio 2001: 70 n. 7.
9 Data for comparison in Johnson 2004: 110-13, 118f., 175-84 (intercolumnia); 119-25, 132-41, 185-200 (columns and margins); 155f. (letter-height and line-spacing).
122
Lectional signs are liberally supplied, somewhat more so in the first tome than in the
second; all three forms of accent are employed, together with rough and smooth breathings
of the square type, apostrophes, frequent quantity marks, and both organic and inorganic
trema. Punctuation is applied consistently, and takes the shape of high and middle stops,
with one low dot acting as a question mark (col. xviii.12). The end of strophes and
antistrophes is marked by a paragraphos, which is accompanied by a marginal koronis at
triad-end (coll. vi, xvii, xxvi) and at the end of the monostrophic stanzas of Paean 5 (coll.
xx, xxii); at the end of an ode, the two signs are flanked by an asteriskos, below which a
second hand wrote the title of the next poem (coll. xxii, xxx). Stichometric letters in the
same hand as the text mark lines 900 (col. i.3), 1200 (xxiii.4), and 1300 (xxix.14) of the
book (not of the individual roll) in section A. In the same section, but apparently not
elsewhere, several lines are marked with a diple or a Z-sign in the left-hand margin.10
The generous spacing between lines and between columns was probably intended
from the start to accommodate the plentiful critical and exegetical material that a number
of hands provided to the Pindaric text.11 The two scribes of the main text added interlinear
corrections and marginal variants to their respective tomes, with further additions, also in
capitals, by (probably) two hands in sections A-B: some appear on their own, many are
10 Lectional and critical signs are described in greater detail by Grenfell and Hunt 1908a: 13-15, Radt 1958: 6f. The position of the stichometric notes is not always correct (DíAlessio 2001: 71 n. 9): _ and Q on coll. xxiii, xxix are 101 lines apart, not 100. On the meaning of diple and Z see McNamee 1992: 15f, 19-21.
11 Collected and annotated in McNamee 2007: 316-43. Margins: Grenfell and Hunt 1908a: 15f., McNamee 2007: 13f. On the content and quality of these scholia, see Rutherford 2001: 149f. and McNamee 2007: 97. Grenfell and Hunt 1908a: 15f. label the ìuncialî hands that added the variants as H1 (identical to the scribe of sections A-B), H2, and H3, and the cursive ones of the scholia S1 and S2; they give no siglum for the scribe of sections C-D, and warn that S1 may be the same hand as S2, H2 and H3 may coincide with either or both of these, and H2 and S1 ìare not unlikely to be identicalî (ibid. 16). Rutherford 2001: 149 n. 21 introduces k1, k2, k3 and W1, W2 for Grenfell and Huntís H1, H2, H3 and S1, S2, but conflates the two scribes of the main text into a single label k1; McNamee 2007: 316 calls ìhand 1î the scribe of sections C-D while identifying it with Rutherfordís k1 altogether, claims that the scribe of sections A-B either wrote no notes or is identical with ìhand 2î = H2/k2, and terms S1/W1 ìhand 3î, S2/W2 ìhand 4î, and H3/k3 ìhand 5î. My text adopts Rutherfordís sigla.
123
accompanied by S[ (S[ndWfRb) or � or �\ (�\fW~fRb, �qfWb, or the like, indicating a
problem), at least two are attributed to Aristonicus (coll. vi. 10, xxviii.10, xxxiv.10), and a
few more either to him or, less probably, to Aristophanes, Aristarchus, or Aristodemus
(col. v.11, frr. 94.4, 134.9, probably 82 col. ii.16).12 One or two cursive hands, roughly
contemporary with the main text, added a few more variants and abundant exegetical
material throughout the manuscript, as did the main scribe of sections C-D. As well as
several glosses and basic grammatical explanations, annotation covers a broad range of
points of syntax, myth, cult, and history, ranging from the elementary (and not always
quite correct) to the fairly sophisticated.13 Theon, Chrysippus, and another authority,
perhaps again Aristonicus, are named (col. iv.2, frr. 84, 129-31) but most of the material is
anonymous; a Pindaric fragment is quoted, without any indication of author or book (col.
iv.2 = fr. 213.1f. Sn.-M.).
G5 = PSI 147
MP3 1362, LDAB 3711
Edition: Vitelli 1913
Location: Florence, Biblioteca Medicea Laurenziana 10016
12 � / �\ traditionally understood as Zenodotus (Grenfell and Hunt 1908a: 15 and passim, Pfeiffer 1968: 118), an interpretation rejected by Lobel 1961d: 42, McNamee 1977: 124-8, and definitively Ferrari 1992a. RQ / R[Q understood as Aristophanes and R[ / R[b] as Aristarchus by Grenfell and Hunt 85 (dubitatively), reinterpreted by McNamee 1977: 71-5; at 2007: 319 she argues that economy favours referring all these abbreviations to one commentator, and rightly points out that it is recent figures like Aristonicus, rather than the great scholars of the Hellenistic age, who are more frequently cited in papyrus marginalia (see also 47). One may add that since the text itself was, at least in principle, that of Aristophanes of Byzantium (who did not write a commentary so far as we know), it would be odd to find variants attributed to him.
13 McNamee 2007: 97 is right to point out that most of the marginalia are not high scholarship, but their range is somewhat broader and deeper than she allows. Even something apparently elementary like | Pae. 6.87 (¯[bXW) p. 304 R. � �¦ �[b]fg] fg(t) �[^�T employs a technical term rarely found outside scholarship proper: a reader interested only in understanding the text would have been content with writing ¯[b]W.
124
Images: DíAlessio and Ferrari 1988 pll. I-II (frr. III-VI, XI-XII), Del Corso 2002, PSI Online
The remains of this papyrus codex were excavated by Ermenegildo Pistelli at Oxyrhynchus
in March 1910 and published by Girolamo Vitelli, after a transcription by Lorenzo
Cammelli, three years later.14 The writing dates from the second or early third century AD,
thus making H5 the earliest surviving codex of lyric, and one of the earliest surviving
Greek literary codices altogether.15
The surviving fragments come from at least three different leaves: frr. I-II preserve
parts of ll. 61-70 (recto) and 104-11 (verso) of Paean 6 = D6 R.; the reassembled frr. III-VI
+ XI-XII, of ll. 3-34 of F5 (recto) and 47-61 of the same ode followed by the title and ll. 1-
13 of Paean 7 = D7 R. (verso);16 frr. VII-X and XIII are still unplaced, but VII and IX (top of
a column) cannot belong to either of the leaves just mentioned. The contrasting figures of
forty-three and forty-four lines per page (calculated on frr. II, VI and III, XII respectively) are
more probably due to imperfect alignment between verses on recto and verso than to a
colometric mistake. Given the line-count known from H4, if we assume that the Paeans
were either the first or the only work contained in this codex (which admittedly nothing
guarantees) and allow for about a dozen title-lines before Paean 6, frr. I-II will be from its
fourteenth, and frr. II-VI, XI-XII from its fifteenth, leaf.17 This fits well with a slim eight-
14 Vitelli 1913: 74. Provenance: the editio princeps mistakenly indicates Hermoupolis Magna (ibid. 73), corrected by Pintaudi 1995 after a letter by Pistelli later published in Pintaudi (ed.) 1996: 63.
15 Second century: Vitelli 1913: 74, Turner 1977: 89, 113; first half of the second century: Del Corso 2002; second or early third: van Haelst 1989: 25, Cavallo 1989: 172. For figures on surviving codices I rely on Turner 1977: 89-94 and the LDAB. On non-Christian Greek literary codices see Roberts and Skeat 1983: 35-7, 67-74, Cavallo 1989. 16 Fr. III-VI assembled already in the editio princeps, XI-XII added by Snell 1938: 425-8, with corrections in DíAlessio and Ferrari 1988.
17 Implied by Turner 1977: 113. Rutherford 2001: 302, 339 suggests that the leaves are the fifteenth and sixteenth respectively, but his reasoning is left unexplained. How many poems preceded Pae. 6 is unknown (at least ten, see DíAlessio 2001: 84), but since it is likely that Pae. 7 ended the book (ibid. 81), a dozen is a
125
sheet, sixteen-leaf codex containing only the Paeans, but since no evidence survives as to
whether other works followed them, this hypothesis remains unproved.18
The hand is a small informal round capital whose similarity to that of H4 sections A-
B was already noted by Vitelli.19 Letter-height is on average little over 2 mm for a leading
of ca. 5 mm: the entire column (one per page, as is the rule for codices of verse) will have
been between 215 and 220 mm tall.20 The upper margin measured on fr. VIIr is 38 mm: if
this is near the original figure and the lower margin corresponded (but it may have been
considerably broader),21 we can estimate that the page was little short of 300 mm tall. The
outer (left) margin of fr. VIv measures 49 mm: if the column was centred, the proportion of
breadth to height will have been roughly 3/4 (Turnerís ìgroup 3î).22
The text is punctuated by high stops. All three accents are amply supplied; to a lesser
extent breathings, rough and smooth, of both the square and the more rounded form.
Elision seems always to be marked, as is inorganic trema on initial b and P; organic trema
accompanies the only verifiable occurrence of diaeresis. A longum is consistently written
over ëDoricí R. The title of Paean 7 is inset in the column and split over two lines indented
by two and three letter-widths respectively. Since the margin next to it is largely lost, it is
unknown what sign(s) were used to mark poem-end beside the surviving paragraphos;
there is no such paragraphos below IVr 9 (end of a strophe). No annotations are preserved.
reasonable estimate given the fourteen Olympians, twelve Pythians, and eleven Nemeans. However, allowance must be made for any compositions that had a subtitle written in a second line, like Pae. 7 itself.
18 On Pae. 7 as the last of the Paeans see the previous note. The direction of the fibres (horizontal on the recto, vertical on the verso) does suggest, at least, that both groups of identified fragments come from the second half of a quire (data in Turner 1977: 65f.).
19 Vitelli 1913: 74.
20 Single-column pages in codices of verse: Turner 1977: 10, 35. In our case, a two-column page (that is, little less than half as tall and twice as large as estimated here) would result in an impossible format.
21 As lower margins usually are: Turner 1977: 8.
22 Turner 1977: 15. Both he and Del Corso 2002 suggest that the upper and lower margins matched the outer margin at about 50 mm, thus producing a page of ca. 220 x 320 mm.
126
G7 = P.Oxy. 1792 + P.Berol. inv. 21114+11677
MP3 1363, LDAB 3721
Edition: P.Oxy. 1792: Hunt 1922b + Lobel 1961b + Lobel 1970. P.Berol. inv. 21114:
Maehler 1968. P. Berol. inv. 21114+11677: Sn.-M. 68f., Ioannidou 1996a.
Location: P.Oxy. 1792: Oxford, Sackler Library. P.Berol. inv. 21114+11677: Berlin,
>gyptisches Museum.
Images: P.Oxy. 1792: Lobel 1961b pll. XIX-XX, POxy Online. P.Berol. inv. 21114:
Maehler 1968 pl. 3b. 11677: Ioannidou 1996a pl. 8, BerlPap.
The fragments published as P.Oxy. 1792 were excavated in Grenfell and Huntís fifth
season at Oxyrhynchus, in 1905-6.23 A first batch (1-68) was published in 1922. Four
decades later, Lobel added seventy more (69-138), expunged an intruder (58), and
reassembled some of those published by Hunt, joining some with one another and some
with new fragments; the resulting compounds he numbered in bold, 1-16.24 To these must
be added a final scrap, published in 1970 as fr. 139.The fragments now in Berlin were
purchased in Luxor in 1908 along with other papyri of likely Oxyrhynchite provenance.25
23 Hunt 1922b: V, with Grenfell and Hunt 1906.
24 These compounded fragments of P.Oxy. 1792 have enjoyed the dubious honour of being referred to in a different way in each published edition of their text. Sn.-M. cumbersomely indicate them with the numbers they had in Hunt 1922b plus the amount of new fragments added by Lobel: thus, Lobelís fr. 1 becomes ìfr. 1 + fr. 3 + fr. 17 + fr. 52 + 3 nova frr. (coni. Lobel, P. Oxy. 26 p. 13 = fr. 1)î. In Bona 1988 the fragments edited by Hunt are O2, and the successive additions O2a (Lobelís re-edition, specified by each fragmentís composition as in Sn.-M.), O2b (P.Berol. inv. 21114), and so forth. Rutherford 2001 returns to Lobelís straightforward numbering, but uses asterisked rather than bold figures. This last system has the advantage of using single figures while avoiding bold typeface, which is potentially confusing in small fount; the downside is a risk of confusion with the asterisks used to denote fragments of uncertain attribution. Accordingly I prefer to retain Lobelís bold figures. The fragments preserved in Oxford and the corresponding images at POxy Online are now assembled and numbered as per Lobel 1961b and 1970.
25 Maehler 1968: 97.
127
This bookroll is written along the fibres ìin a good-sized, rather heavy, semicursive
hand which may be referred to the first half of the second centuryî AD;26 the verso is
blank. Letter-height is on average 4 mm, with a leading of ca. 7 mm. No entire column or
margin survives; fr. 1 shows an upper margin of 12 mm and an intercolumnium of 17 mm
to the left. Maasís Law does not seem to be observed. Regrettably for the main witness to
the Prosodia, the state of the manuscript is ghastly: only two out of so many fragments
preserve more than a few disconnected syllables, and the only one that yields continuous
sense (fr. 1) is badly stripped as well as mutilated. No speculation can be made about the
overall layout of the volume.
Lectional signs are very sparse: all three accents, rough breathings of the square type,
organic and inorganic trema, apostrophes, and quantity marks, mostly in the same hand as
the main text, as are the few middle and high stops. The only visible critical activity takes
the shape of marginal diplai (frr. 1.14, 5.1, 3) and interlinear corrections, one of which (fr.
1.14) in a hand recognisably different from that of the text. No marginalia are preserved.
The surviving asteriskoi (frr. 8, 47, 84) are inset in the column; the coincidence of fr. 8
with H29 fr. 1 col. ii.12-9 shows that the asteriskos on fr. 8 was indented by around eight
letter-widths, thus presumably leaving too little room for an inset title on either side.27
Consequently, titles (if present at all) will have been written in the margin.
26 Hunt 1922b: 86.
27 Lobel 1961d: 42.
128
G8 = P.Berol. inv. 13411 (+21239?)
MP3 1365, LDAB 3709
Edition: 13411: Zuntz 1935, Bowra 1935. 21239: Sn.-M. 56f., Ioannidou 1996b.
Location: Berlin, >gyptisches Museum
Images: 13411: Cavallo 1967 pl. 24 (frr. b, c, d), BerlPap. 21239: Ioannidou 1996b pl. 62,
BerlPap.
P.Berol. inv. 13411 was acquired by purchase some time in the mid-1910s, and its
provenance is unknown.28 Attributed by its first editor to Simonides rather than to Pindar or
Bacchylides on grounds of language and style, but published at the same time in Maurice
Bowraís OCT of Pindar albeit as incerti auctoris (fr. 343), it was definitively identified as
Pindaric by Bruno Snell, who noted its partial overlap (fr. a.8-10) with H7 fr. 6.29 Herwig
Maehlerís later addition of P.Berol. inv. 21239, based on palaeographical considerations, is
more doubtful: although a definitive judgement cannot be pronounced on so little text,
Giambattista DíAlessioís scepticism is justified.30 Conversely, the unattributed P.Laur.
28 Zuntz 1935: 282, Gunther Poethke in DíAlessio 1999: 15 n. 2. The papyrus had been in the collection per viginti fere annos at the time of publication (Zuntz), and was mentioned by Wilamowitz 1918: 728, who thought it too poorly preserved to be included in his article. It came to Berlin by purchase; nothing is known about its provenance.
29 Zuntz 1935: 292, Bowra 1935 (not paginated) with Zuntz 1935: 282 n. 1, Snell 1938: 431f. with DíAlessio 1999: 16 n. 11.
30 DíAlessio 1999: 25, noting the ligature-like ]P at l. 2, the different shape of R, and the slightly larger format. One may add that in 21239 the nib seems to have been held level with the base-line, while in 13411 (as is regular in ëBiblical majusculeí: Cavallo 1967: 4-6) it stands at an angle; this is easily seen in round letters. Room for variation even within a single manuscript must be allowed, and none of these elements is decisive, but since the similarity of the handwriting is the only basis for the identification, the case for linking fr. k to the others is on the whole rather weak.
129
II/35 is in a very similar hand and format, and there is a slim chance that it may belong
with our manuscript.31
The writing is a fine specimen of formal round (ëBiblical majusculeí): medium-sized,
regular, with a marked contrast between thin and thick strokes and an angle of writing of
about 30°, well-spaced and strictly quadrilinear (with the usual exception of b, [, P, T).
Within the development of the style it can be dated to the first half of the third century
AD.32 The verso of the roll is blank. Two columns are preserved to their full height in frr.
a-b, for a length of twenty-five lines; given the extreme regularity of the writing, one can
assume that these figures were quite constant throughout the roll. The widest extant upper
and lower margins are those of fr. b, the lower slightly wider than the upper. The broad
letters produce a wide column and a conspicuous difference between longer and shorter
cola. Maasís Law is observed (fr. a), but the slant forward is minimal. As I have not seen
the papyrus except in a photograph, I cannot provide any accurate measurements.
Lectional signs are employed very sparingly. All three accents occur, some written
by a different pen; apostrophes and the occasional trema (organic and inorganic) seem to
be due to the original scribe. R is twice surmounted by a longum. High stops � the only
form of punctuation � sometimes take the shape of a short dash just below the top of the
letters. A paragraphos, presumably marking the end of a strophe or antistrophe, is found in
the margin below frr. a.10, e.3; no other metrical signs are preserved.33 It is possible, if not
31 Edition: Livrea 1981. Same scribe and manuscript: discussion in DíAlessio 1999: 15 n. 2, see also Orsini 2005: 77f. The only obstacle to Pindaric authorship (4 •jT) is not insurmountable, but the identification of the hand itself remains doubtful; accordingly, P.Laur. II/35 is not included in this dissertation.
32 Cavallo 1967: 47 and n. 4 (around 230), Orsini 2005: 50.
33 Ioannidou 1996b suggests that the jÆ beside 21239 l. 7 is a stichometric note (=1000), but if this is so, its position to the right of the column is striking. On the other hand, that is no place for jR^, jRfn, or jik\X (McNamee 1981a: 46, 48, 52) either. If jÆ is stichometric and the figure of twenty-five lines to a column was constant throughout the roll and 21239 belongs with H8, then we may speculate that the fact that l. 1000 is
130
likely for any particular reason, that some of the blank line-ends may follow an inset title,
rather than a short verse, in lacuna. At least two semicursive hands (in two different inks)
have frequently intervened with deletions and interlinear corrections; no more extensive
notes survive.
G26 = P.Oxy. 2442
MP3 1360, LDAB 3747
Edition: Lobel 1961d.
Location: Oxford, Sackler Library.
Images: Lobel 1961d pll. III-VIII, POxy Online.
The fragments grouped under this heading are written in one hand and an almost identical
format, but probably constitute the remains of (at least) three individual rolls,34 as is
indicated by the overlap of parts of their text with fragments whose classification is
otherwise known: Hymns and Paeans as well as the Prosodia.35 Another roll in the same
hand and format, containing the Pythians, was not published for another fifty years: it is
now P.Oxy. 5039.36 All these and P.Oxy. 1787 (Sappho), written by the same scribe in a
not the last of a column suggests that titles in H8 were inset rather than marginal, thus upsetting the line-count. But none of the premisses can be proved.
34 Lobel 1961d: 31. Johnson 2004: 26 alternatively suggests one anthological volume, perhaps also including P.Oxy. 1787 (on which see below and n. 37); but surely the fragments of all these, plus those now published as P.Oxy. 5039, represent too many poems to have come from a single roll.
35 Hymns: fr. 1 col. ii.1-4 ~ fr. 33d.7-10 Sn.-M. Paeans: fr. 14a col. i.10-22 ~ H4 fr. 16.2-15; fr. 14b-c ~ H4 frr. 17+18+16.11f.; fr. 22.1f., 6-9, 29f. ~ H4 frr. 90, 87, 143; fr. 22.1-6 ~ H6 ll. 5-10. Prosodia: fr. 32 col. i.6-10, 16-19 ~ H7 frr. 31, 139; the refrain at fr. 32 col. ii.3f., 10f., 19f. ~ H7 frr. 24.2f., 55, and possibly 83, 84.1; probably fr. 86.6f. ~ H4 col. xxx.13 with title. See Lobel 1961d passim, DíAlessio 1997: 35-7, 40f.
36 Maehler 2010b with Lobel 1961d: 31.
131
closely similar format,37 were found in one lump in the excavations of 1905/6. Sets of rolls
in an identical format are amply attested in Graeco-Roman Egypt: compare P.Oxy. 2430 of
Simonides (Paeans and one or more books of Epinicians), A of Bacchylides (Epinicians
and Dithyrambs), and on a much greater scale the (at least) ten matching rolls of Aeschylus
excavated at Oxyrhynchus by the British and Italian expeditions in the early twentieth
century.38 Frr. 86 and 87 of our papyrus were first published as a part of P.Oxy. 1787 (frr. 8
and 9 respectively) and reassigned by Lobel;39 a scholion published by Lobel together with
P.Oxy. 2438 but written in a hand similar to that of H26 may also belong with it.40
The hand is a rapid, sloping example of the formal mixed (ësevereí) style datable to
the third century AD.41 On average, letter-height is 3 mm for a leading of 6 mm, but both
these and the width and spacing of individual letters show sometimes considerable
variation, thus precluding an accurate estimate of lacunae other than the very shortest.42 Fr.
1 preserves an upper margin of 22 mm and an intercolumnium of the same width at its
37 Lobel 1961d: 31 after Hunt 1922a: 26f., Johnson 2004: 26f. (ìscribe #A30î). Funghi and Savorelli 1992b: 45-54 attractively suggest that the same scribe copied other Pindaric papyri ñ H9/H30 (probably one roll: Lobel 1961e: 86) and H31 ñ as well as P.Oxy. 1788 (Sappho or Alcaeus) and 2443 (Alcman); the identification is denied by Johnson 2004: 26 (ìscribe #A20î), but the resemblance is very considerable indeed, and I cannot exclude it. 38 PSI 1208 (Niobe), 1209 + P.Oxy. 2161 (Dictyulci), PSI 1210 + P.Oxy. 2160 (Glaucus Potnieus), P.Oxy. 2159 (Glaucus Pontius), 2162 (Theori), 2163 + PSI 1472 (Myrmidones), P.Oxy. 2164 (Xantriae or Semele), 2178 (Ag.), 2179 (Th.), 2245 (a Prometheus), and the miscellaneous 2246-55 (from an undetermined number of rolls), plus perhaps P.Gen. inv. 98 (Funghi and Martinelli 1996-97: 7-9): see Radt pp. 10f., Johnson 2004: 18f., 61 (ìscribe #A3î).
39 First published in Hunt 1922a: 33; fr. 86/8 was conjecturally assigned to Pindar by Lobel 1922: 290 (whose tentative identification with I. 9.1 cannot stand, see Snell 1938: 438), and both to H26 in Lobel 1961d: 67f.
40 MP3 1351.1: text in Lobel 1961a: 6 n. 1; image ibid. pl. I, Gallo 1968: 52, POxy Online (s.v. P.Oxy. 2438). Attribution to P.Oxy. 2442: Funghi and Messeri Savorelli 1991. The content of the scholion does resemble | P. 3.139a p. II 81 Dr. (Porciani 1991); Funghi and Messeri Savorelliís hypothesis (1991: 103) that the note may have been related to P. 3.78 on P.Oxy. 5039 ñ whose text was then unpublished ñ cannot be verified, since only ll. 101-3 of P. 3 are extant on the papyrus, but deserves consideration. Elsewhere in H26 the main scribe seems only to have contributed short notes, not such extensive discussions as we find here, but this need not be decisive. 41 See Hunt 1922a: 26f. (on P.Oxy. 1787) with Lobel 1961d: 31.
42 Maehler 2010b: 69.
132
narrowest; the former figure is confirmed by fr. 64 (19 mm in frr. 86-7), the latter by fr.
14a (but 26 mm in fr. 32). The lower margin of fr. 39 is a generous 42 mm, 38 mm in fr.
107b. In fr. 32 the column-to-column width can be estimated to ca. 155 mm. Maasí Law is
observed. Lectional signs are liberally supplied, mostly by the main scribe:43 all three
accents, square rough and smooth breathings, quantity marks, both organic and inorganic
trema, apostrophes, hyphen. Punctuation consists of high stops.
Poem-titles (frr. 7.7, 14a.3a, 86.6, possibly 16.3a) are written by the same hand as the
verses, inset in the column, and inconsistently indented: the title at fr. 7.7 started only some
three letter-widths to the right of the margin if the likely supplements to the first lines of
the text are correct (as did that at 86.6 if correctly identified with that of F5), but that at
14a.3a must have ended only a short space away from the column to its right. The title at
fr. 86.6 has blank spaces of about 6 mm above and below it, while that at 7.7 is not spaced
away from the rest of the column, and that at 16.3a is not spaced away from the preceding
line but is followed by a blank of ca. 5 mm. Poem-end was marked by a simple koronis:
there is no asteriskos visible beside it below fr. 14(a) col. ii.20, after which an unwritten
space (eisthesis or line-spacing) indicates that a title followed. Fr. 32 col. ii shows the
expected use of koronis marking triadic, and paragraphoi strophic, divisions (the latter
also at frr. 38.3, 41a.2, 105.2, 107a col. ii.13); the absence of a context makes it impossible
to tell if the koronides below frr. 19 col. ii.1, and 65.4 marked poem- or triad-end. Two
diplai and a Z in a lighter ink than the text are found (fr. 1 col. ii.1f., 107a col. ii.11).
The layout of the rolls can only be the subject of speculation, but some data are
available. As far as the Prosodia are concerned, the triadic structure that the manuscript
attributes to F9 (fr. 32 col. ii) demands at least eighteen lines of text � eight lines each for
43 Lobel 1961d: 31, Funghi and Savorelli 1992a: 53 and n. 53, Johnson 2004: 26f.; see also Maehler 2010b: 69.
133
ëstropheí and ëantistropheí, plus at least two of the minimum six of the ëepodeí � and one
for the title, plus any remaining verses of F8, between the last extant line of fr. 32 col. i
and the first of col. ii, which is the fourth-to-last of an ëepodeí.44 Combined with the
twenty-two lines that are actually preserved across the two columns between any two
aligned verses, this gives forty-one lines as the minimum height of the column (or forty-
three if F9 was actually monostrophic), although the irregular writing and erratic spacing
noted above makes any results difficult to project onto the entire roll.45 For the Paeans,
Bruno Snell had calculated a column of thirty-nine lines on the basis of a possible overlap
between fr. 14 col. ii.11-14 and H4 fr. 19 col. ii.11-14. The overlap is minimal and the
identification highly doubtful, but the resulting figure would be compatible with one
44 The koronis under l. 4 and the paragraphoi under l. 12 and 20 can hardly be explained otherwise than in their regular function of marking the end of (respectively) an epode, a strophe, and an antistrophe, each ending with the refrain: DíAlessio 1997: 36 n. 82. This gives eight lines as the length of strophe and antistrophe, and at least six as the length of the epode (four verses visible at the bottom of fr. 32 col. ii plus the two of the refrain). Given the remarkably relaxed responsion between the stanzas so delimited (see p. 114 above), Lomiento 1998: 113f. took up a hesitating suggestion by DíAlessio 1997: 36 and argued that the poem is actually astrophic, a sequence of non-responding stanzas linked solely by the overall coherence of the metre and by the repeated refrain. The allegedly parallel case of O. 14 (astrophic according to | metr. O. 14 p. I 388f. Dr. = pp. 12f. Tessier with Lomiento 109-11 and in Gentili et al. 2013: 337f., but see already Dawes 1745: 66-8 and now Irigoin 1958: 30f., Tessier 1995: 125-7) is unconvincing: the system of paragraphoi and paragraphos with koronis � not the ones or the others � employed by H26 suggests not the mere division of a text into sections, as the scholia divide O. 14, but an explicitly triadic articulation which, in ancient no less than in modern metrics, is hardly compatible with an acknowledged lack of responsion. However, for (at best) very relaxed responsion between metrical units explicitly articulated by a papyrus as responding one can compare Simon. SLG 320, 323 = frr. 36, 39 Poltera on P.Oxy. 2623 frr. 2, 5 with Lobel 1967a: 70f. (1981: 21 on authorship). My calculation here is predicated upon the assumption that the ëtriadí that ended at the top of fr. 32 col. ii was the first of the ode: if it was the second, the minimum height of the column would be fifty-nine lines and the roll well over 40 cm broad, a surprisingly large format.
45 The hypothesis of a monostrophic F9 (Lobel 1961d: 54, Bona 1988: 297, DíAlessio 1997: 36 n. 82, apparently now endorsed by Gentili and Lomiento 2003: 188) stems from the consideration that the remains of the ëepodeí are not significantly less compatible with those of ëstropheí and ëantistropheí than these are with each other. The triadic articulation construed by the koronis and paragraphoi is not necessarily an obstacle, if the ode had a number of responding stanzas that was a multiple of three: Aristophanes of Byzantium could have chosen to interpret them as triads with epodes identical to the strophes and antistrophes. The possibility of such competing articulations is explicitly mentioned by Heph. p. 62 Consbruch, and there is an exact Pindaric parallel in N. 4, which the scholia regard as triadic with an epode identical to strophe and antistrophe (metr. N. 4 p. III 63 Dr. = p. 23 Tessier); cp. metr. b P. 6 p. II 192 Dr. (not in Tessier), which describes monostrophic odes as having strophes, antistrophes, and epodes that are identical to each other.
134
slightly higher in one of the companion rolls; the higher figure that rejection of Snellís
hypothesis entails (at least forty-five lines, and presumably not many more) would sit
equally comfortably with the minimum tally extracted from fr. 32.46
The text is liberally annotated by at least two hands: the first may be that of the main
scribe, and is responsible for corrections, variants, and short exegetical notes, while the
second � a tiny cursive � adds interpretative material, often with remarkable largesse.47
Two queries are introduced by �\. The second annotator exhibits a penchant for expanding
on the mythological background of a point in the text (frr. 29, 32 col. i.19, 39.8) and other
erudite details (96B(a)), but more directly relevant material is not excluded. The content of
one note partly matches one found next to the same text in H4 (fr. 29 ~ H4 fr. 82.i), and the
learning that underlies our papyrusí annotations is made conspicuous by references to such
authorities as Hellanicus and Epimenidesí Genealogies (29),48 Callimachus (96B(a)),
Didymus (97, perhaps 39), possibly Istrosí Eliaka and Autesion or Antesionís Lydiaka
(39), and either Nicanor or Aristonicus (99).49 This wide-ranging roll-call of names
(whether known first- or second-hand) stands out in comparison to H4, where only Pindaric
commentators are cited.
Fr. 32 is in an exceptionally poor state, ìfrayed, liable to flake, warped, and stainedî
(Lobel) as well as heavily soiled in some places.50 I am deeply indebted to Daniela Colomo
for a judicious attempt at cleaning the worst-affected area at the left edge of col. ii, which
46 Overlap: Snell 1962: 4f., but see Rutherford 2001: 252 n. 33 and DíAlessio 2005: 139f. (the overlap works for ll. 11-13 but probably not for l. 14). DíAlessio 140f. calculates a column-height of 44 to 53 lines for H9 (Dithyrambs), which may have been copied by the same scribe as H26 (see n. 37).
47 Hands: Lobel 1961d: 31, Johnson 2004: 27. Edition and notes: McNamee 2007: 308-15 (313 on the hand of the variants).
48 Lobel 1961d ad locc. Fowler 1993 had also read the name of Euphorion (consequently edited as SSH 454a) on fr. 29, but it is not compatible with the traces: DíAlessio in Magnelli 2002: 129 n. 8.
49 ] |́ : Aristonicus if oß] |́ (see Lobel 1957: 12), Nicanor if simple |́ : McNamee 1981a: 10, 63 and 2007: 39.
50 Lobel 1961d: 52.
135
allowed for the recovery of two hitherto hidden traces and greater visibility of some
already spotted by the first editor.
G29 = P.Oxy. 2441
MP3 1370, LDAB 3719
Edition: Lobel 1961c.
Location: Oxford, Sackler Library.
Images: Lobel 1961c pl. II, POxy Online, Leone 1976: 144f. (fr. 1), GMAW2 pl. 22 (fr. 1).
The writing ñ a ìmedium to small, upright but flattened round capital of informal type Ö
markedly bilinearî ñ can be dated to the middle of the second century AD; the hand
appears to be the same as that of P.Oxy. 3674 (Plato, Laws book 9).51 Fr. 3 only contains an
asteriskos and a title fg~] Rãfg~]. Its connexion to the other fragments is uncertain: the
shape of the asteriskos, the layout of the title, and the hand in which it is written all seem
to be different from those of fr. 1.52
Part of the upper margin of the roll is preserved (18 mm in fr. 1), and the one extant
intercolumnium measures 22 mm at its narrowest; letter-height is regularly 2 mm, with a
leading of 4 to 5 mm. Maasís Law is observed. Nothing can be made out concerning the
layout of the roll, save that the minimum height of the column was a respectable twenty-six
lines. Lectional signs are generously supplied, mostly by a second hand:53 acute and
circumflex accents, rough breathings of the square form, organic and inorganic trema, apo-
51 Writing: GMAW2 50, with Lobel 1961c: 25 and Leone 1976. Date: Lobel 1961c: 25 (but mid- to late second century according to Cavallo 2008: 95). Hand: Cockle 1984: 99; ëscribe #A29í in Johnson 2004: 63.
52 Different hand: already Pardini in DíAlessio 1997: 38 n. 94.
53 Lobel 1961c: 25.
136
strophes, quantity marks, high stops. Poem-end is marked by paragraphos, koronis and
asteriskos, and the title of the following poem is written by the main scribe in the margin
just below the asteriskos; the margins show two more paragraphoi (frr. 1 col. ii.23, 2.3),
two diplai (fr. 1 col. ii.2, 25), and two Z (frr. 1 col. ii.20, 2.27).
The four marginalia to the right of col. I appear to be due to a second hand.54 One is a
variant and two are exegetical notes, one of which quotes ìsomeî as an authority.55
G34 = P.Oxy. 2448
MP3 = 1377, LDAB 3726
Edition: Lobel 1961f.
Location: Oxford, Sackler Library.
Images: Lobel 1961f pl. XIV; POxy Online.
This bookroll is written in a spiky, slightly sloping and highly contrasting formal mixed
hand somewhat similar (but certainly not identical) to that of H26 and the other rolls
assigned to the so-called ìScriba di Pindaroî.56 Lobel dates it to the late second or early
third century.57 The average letter-height is just over 2 mm, with a leading of just over 5
mm. No upper or lower margins are recognisably preserved, and no fragment preserves
54 Lobel 1961c: 26 and GMAW2 50 maintain that the scholia are the work of two different hands, one of which penned |15 and the other the rest. |15 is written in a lighter ink and with a thinner pen, but I would agree with McNamee 2007: 348 that the hand is the same.
55 McNamee 2007: 348 notes the rarity of ìg—(fT) being used to justify an interpretation, not a textual pointî.
56 On this figure and the disputed attribution of ëhisí papyri see Funghi and Messeri Savorelli 1992a, Johnson 2004: 24, 26, and n. 37 above.
57 Lobel 1961f: 131.
137
remains of more than a single column; even an impression of the original layout is beyond
reach. Maasís Law is not obviously observed.
The text is equipped with acute and circumflex accents, rough breathings of the
square form, frequent quantity marks and apostrophes, tremas, and two hyphens; ì[a]t least
two pens may be distinguished in the lection signs, of which the thicker may be in many
cases � I am not sure whether in all � that of the writer of the textî (Lobel).58 Punctuation
takes the shape of high and middle dots, all seemingly due to the first scribe. Twice a letter
or group of letters is unobtrusively deleted with a suprascript dot (fr. 1.4, 2(a).10), and in
one case a correction made immediately above; other traces of correction are an interlinear
addition at fr. 14.6 and (presumably) a variant reading at 3(a).3. Perhaps due to the almost
complete loss of all margins, no other annotation survives except a single letter at the edge
of fr. 7. The only preserved marginal sign appears to be a koronis,59 marking stanza-end,
visible beside fr. 1.6 and probably centred one line below, under 7.
58 ibid.
59 ìNot a normal coronisî according to Lobel 1961f: 132, but in that place it can hardly be any other sign we know: for the admittedly anomalous shape compare e.g. the upper stroke of the koronis that separates B. 17 and 18 on A.
139
TEXT
141
F1
(fr. 89a Sn.-M.) f^ jnVVbgQ �[Zg_YQgb]bQ ¢ jRfRcRPg_YQgb]bQ ¢ �Rp��TQ¦Q fW ÊRfi jRÚ pg¬Q µccTQ �VnfWb[RQ �W~]Rb; [I] = | (VE56MLh) Ar. Eq. 1264b Jones-Wilson [= T1] f^ jnVVbgQ �[Zg_YQgb]b� fgtfg �[Zq c[g]gk^gP (c[g]Tbk^gP VE5M) HbQkn[gP. ¯ZWb kÓ g—fT]� f^ jnVVbgQ ® �W~]Rb; unde [II] = Suid. c 2757 Adler [= P20] c[g]¦kbR� —_Qgb. �[ZU c[g]gk^gP� f^ jnVVbgQ ® �W^kWbQ ; 1sq. �[Zg_YQgb]bQ ¢ jRfRcRPg_YQgb]bQ [I] EM [II] : �[Zg_YQgb]bQ ¢ jRfRcRPg_YQ(gb]) [I] 5, def. Thiersch 1820 p. II 246 : �[Zg_YQ(gb]) ¢ jRfRcRPg_YQ(gb]) [I] 6 : �[Zg_YQgb] ¢ jRfRcRPg_YQgb]bQ dub. Bergk 18784, hanc lectionem 6 tribuens : de codd. lectionibus v. Jones ap. Fraenkel 1962 p. 205 adn. 2 || 2 jRfÏ cRPg_YQgb]bQ Hartung 1855-56 || 3 ÊRfà jRÚ om. [I] M || 4 pg¬Q [I] Lh, [II] : pgÏQ [I] 56 : pWÏQ [I] VEM | µccTQ om. [II] V | �VnfWb[RQ [I] : �VRfv[R] [II] | �W~]Rb [I] : �W^kWbQ [II] : S[. jRÚ �W~]Rb [II] AVM
142
F2
(Pae. *14 = fr. 52o Sn.-M. = S3 R.) p. i . . . ] ] ][kWb� ] ] � 10 ]ZWbQ � ] � � � � �¾�[Ï ]fn]b] � ] ] ] 15 ] ] ] ] ��Q �gbkR~] � ] 20 ] ] §]fR^[gP] ] ] 25 ] . . . H29
fr. 1 p. i quaecumque non notantur dispexit articulavit supplevit Lobel 1961c || 8 ìperhaps a middle stopî Lobel || 11 fin. ìnot apparently a letterî Lobel || 13 ×–¥�í �¥¾�[Ï ]fn]b] suppl. Snell 19643 ex | || 15 kbn incautius suppl. Snell ex | || 20 �Q �gbk¬b suppl. Snell ex |, ìfort. �Q �gbkR~] in textu erat (Lobel)î Scholia : 13 � � � �¾�[Ï ]fn]b]� gà £ �b R^ �. g—(fT]) fbQ(W]) 15 ] �UQ kbn c[¦p(W]bQ), µQí õ�b� £[ 16 ]VgQ 20 �Q �×bk¬b 13 init. ×Ã¥½ nol., ×– ¥�í dub. Lobel : res incertissima | -W Ú��[n nol. Lobel, recte || 15 articulavit et supplevit Snell duce Lobel | jRfÏ] Rutherford 2001 : �VVW^cWb vel sim. malim || 20 k¬b pot. qu. k�b H29
143
p. ii margo k[ ] ¥ ¥PcgVb� ¾��j[Y]R [ é[ ]]¥ Ô[ ]��¥ ¥[ ] ¥� ¥[ ¥] ¥[ ¥] ¥[ �[ ] ¥RpP[ �] �V� ¥ �R ¥[ � �] �� ¥[ ] �[ 5 (30) WÃ¥kg�^R] kí �c^ZWb[R kW[ pW� V^SébR _ÓQ Ùg~]í, �dí Ú�[Qñ _TQ fWVWPfR~] ¿R[^�W[b V¦SgQ fW[cQ�Q �cYT� [ _Qn]Wb kÓ jR^ fbQR QR^g[Q- 10 (35) pí §jÏ] ô[T�kg] pWR[^R]� �R]�RQb- ]pYQfb kÓ Z�[P]�b f�YVg] ¥[ SQi_R] kÓ f�RZW^R] �]PQ[ ]gd^Rb SÏ[ �W^[W�fRb c�VW�[]fñ 15 (40) 1-15 H29
fr. 1 p. ii.1-15 || 12-15 �H7 fr. 8 .1-4 �
quaecumque non notantur dispexit articulavit supplevit Lobel 1961c || 1 ìperhaps ]QgP c¦VbQ ZRVj[Y]Rî Lobel | init. vel ]�]¥P Snell 19643 | é–cgVb� possis | fort. � esse potest �� | j[� ]R H29 || 2 vel �[ Lobel | ]�¦¥ ¥[ Lobel | hunc v. nota > notat H29 || 3 ]é� Lobel, fort. recte || 5 ]� vel ]� | ] ¥� ¥[ Lobel : ]�Sg[ vel ]�fW[ Snell || 6 ^�] H29 | c^Z H29 | fin. ìsome form of kYjg_Rbî Lobel : kY[XRb Snell : kY[Xg Snell sec. Rutherford 1992 p. 60 : fort. de kWbjQ�QRb vel kWkg[jYQRb agebatur || 6sq. f\V¦]£pW Snell : õV]£pW Currie, fort. recte || 7 �V^S H29 | g~]í� H29 | �[ vel È[ pot. qu. ]¥[ || 7sq. �dí Ú�[Qñ (Ö) �Q Sn]£_TQ Ferrari 1992b p. 152 : ³dR� [�Q ji]£_TQ e.g. Snell, quod vestigiis parum congruit : �dí Ú�[Q�Q (Ö) k¦]£_TQ possis, fort. coniunctione interposita || 8 f¬b H29 | [^� H29 || 9 fin. dY[gb]R Snell : j¦]_gQ Ferrari || 10 jR^ H29 | QR^ H29 || 11 pí§ H29 | T�k H29 || 12 �[^R]� H29 || 13 �[ pot. qu. �[, �[, ¾[ : �[RQ�WfRb Snell : �[b_n van Groningen 1963 p. 128, quod vestigiis minime congruit || 14 ]�Q[W]b] Lobel : ]PQ[Wf�Q j[^]b] dub. Snell || 15 b�b H29 | cVW~¥[]fR vel cVW~ []fí �[Wfn dub. Snell : cVW�[]fnjb van Groningen || finem carminis paragrapho coronide asterisco notat H29, asterisco H7
144
F3
(Pae. *15 = fr. 52p Sn.-M. = S4 R.) o[∞]SbQqfRb] W∞[]] o∞Rj¦Q Á�bkí �Q ³_�Rfb fW[c�Q�b µccgb _ÓQ ��pnQRf�Rb Hg]Wbk¬Q�g] ³SgQ�fí o∞R½[ ´\[Wˆ] kí ¡ �SY[T�Q ãcWfR[b� cRfU[ kÓ æ[gQ^TQ _gV[ 5 c[Ù] ƒ__R �R�Và�Q ZW[Ú [ f[ncW�RQ pW��Q �c�í �_�[[gñ µQR g� jYZPfRb cbW~Q QY¥[jfR[ ® ¯[ZWfRb kí �QbRPf�b ÕcW[fnfRQ [ ¥ ¥]gQR- 10 ¥[ . . . tit., 1-11 H29
fr. 1 p. ii.16-26 || 1-4 �H7 fr. 8.5-8� || 6sq. �H
7 fr. 69� || ex eadem parte papyri H7 fr. 45 exortum esse ci. Lobel 1961b p. 15 quaecumque non notantur dispexit articulavit supplevit Lobel 1961c || finem carminis prioris (F2) paragrapho coronide asterisco notat H29, asterisco H7 || tit. in margine praebet H29 || 2 pnQ H29 | Rb̧ H29 || 3 nSg H7 | Qfí H29 | ìo∞Rj[¦Q vel o∞Rj[�b �R_np(Wb)RQ vel oÒSbQRQ vel o∞Rj[^kR rYfbQ ?î Snell 19643, Aeginae obl. DíAlessio 1991 p. 103 || 5 _gV[gt]b Lobel : _gV[gt]bQ W“d[gQ Snell : _¦V[gb Currie | hunc v. nota Z notat H29 || 6 [d^VRb kYZWfRb e.g. Snell : [kW^jQPWQ e.g. Werner 1967 p. 537 : ìprima facie completeî Lobel || 7 �_�[[g]^RQ vel �_�[[¦fRQ pot. qu. �_�[[¦fTQ, quae omnia ci. Lobel | fin. cWVn�Wb e.g. Lobel || 8 ëÍQRgb � H29 | WbêQ H29 || finem strophae paragrapho notat H29 || 10 �c H29 | fnf� H29 | gQ� H29 | [¡k]¦Q possis | hunc v. nota > notat H29 || 10sq. [�c]Ù QR£È[¬Q e.g. Rutherford 1992 p. 62 contra metrum : [Õc]Ù ´R£� [kR dub. Rutherford 2001 : [³S]gQ Ú£�[Y[RQ Ferrari 1996 p. 133, quod vestigiis parum congruere videtur || 11 ìthere is a trace of ink at a lower level to its left in the marginî Lobel
145
*F4 (fr. 215(a)-(b) Sn.-M. = G10-11 R. + Pae. 22(h).1-5 = fr. 52w(h).1-5 Sn.-M. = Z23 R.)
(a) . . . ¥] ¥[ ¥ ¥ ¥] ¥[ ¥ ¥ ¥] ¥ ¥[ ³VV�R�⸥kí ³�V�gb�]bQ Q�¦_b_R, ]dWfY[RQ kí �∞Qé~¥ k^jRQ �Qk[�Q 㥽�R]fg]� ¥̄R] g�, ‚� fnQ, _q _W jW[f¦_[Wb ¯]fb _gb 5 cRf[ �̂kí �[ZR^R �Q jfWQÚ �bW[^k[TQ ·]]fW Z�R^fRQ cR[pYQ�g� X�Q�[Rñ õ® ¥]Q[ ¥ ¥� ]WQ Sn[, êcgVVgQ [ V�][�� fW j�Ú Õ[_Qñ ]Rb _WV[�¥ ¥ ¥]TQ �SVR�R�] [ 10 ]]j[ ¥ ¥]f×Qé[ ] ¥[ ] ]�QWfg~¥[]] ¥[ ] ¥f[ ¥ ¥]°é�] � ãcg�[Rb ] ¥[ ¥]� ¥[ . . . 11-14 H34 fr. 1 || 2sq. [I] = | (h) Il. cit. ad 2.400 p. I 270 Erbse R∞WbSWQWfnTQ� kbÏ cRQfÙ] ƒQfTQ, �pRQnfTQ. jRÚ g� (jRÚ g� om. Ag) ºTQW] fÙ W∂QRb SWQY]pRb dR]^. jRÚ H^QkR[g]� ³VVR ® ãjR]fg]� fÙ SÏ[ pW~gQ d�]Wb _ÓQ ãQ, pY]Wb kÓ cgV�. | [II] = Artem. 4.2 p. 243 Pack jgbQvb _ÓQ g“Q §RPfg~] ›_gVgSqjR]b _P]fq[bR jRÚ fWVWfÏ] jRÚ cRQ\S�[Wb] jRÚ �S�QR] jRÚ ]f[RfW^RQ jRÚ SWT[S^RQ jRÚ c¦VWTQ ]PQgbjq]Wb] jRÚ Sn_TQ jRÚ cR^kTQ �QRf[gdÏ] jRÚ ≈]R ³VVR fg�fgb] ≈_gbR, §RPf�b kÓ ãjR]fg] ¡_gVgSq]R] �Q]fn]Wb (Hercher 1864 : �Q ]fn]Wb L : �cÚ c¬]b V) Z[vfRb ∞k^Rb jRÚ (add. Hercher) �]pvfb jRÚ ÕcgkY]Wb (Hercher : Õcgkq]Wb LV) jRÚ f[gdvb jRÚ jgP[¬b jRÚ ³VVTb f�b cW[Ú fÙ ]�_R j¦]_Tb jRÚ �cbf\kW�_Rfb (V : �cbf\kW�_R]b c¬]b L) jRÚ c[gRb[Y]Wb, ÄQ ðQ _nVb]fR �cRbQvb, ³VVR ® ãjR]fg] d\]ÚQ ¡ H^QkR[g]. || 6sq. [III] = H37 3-6, v. infra quaecumque non notantur dispexit articulavit supplevit Lobel 1961f || 1 ] ¥é[ vel ] ¥]¥[ || 2 ³VVR [I] Ge [II] : ³VVgb [I] Ag : nVV[ H34 | kíR�V[ ¥ ¥]]bQ H34 : kí ³VVgb]b [I] : kÓ ³VVgb] [II] | ]dWfY[RQ [I] [II] L : ]dYfW[R [II] V || 3 kí R∞QW~ k^jRQ [I] : kí ¥bQé ¥kbjRQ H34 : kYkWbjfRb [II] | �Qk[�Q om. [I] [II] | ãjR]fg] [I] [II] L : §jn]fgb] [II] V : ¥ ¥[ H34 || 4 ¯R]gQ Coles et Ferrari ap. eund. 1992c pp. 230sq. : SnÔgQ Snell 1962 p. 6 (vel S¬bgQ 19643) : ZnÔgQ ex SnÔgQ correctum Lobel ap. Snell 1962 : k¬bgQ Kambylis 1966 p. 240 : primam litteram ° esse censuerat Lobel 1961f | é!R" H34 fort. synizesin indicans (Y� Ferrari : ¥¬ Lobel) | f#Q H34 (sec. van Groningen 1963 p. 129) | $_W% H34 : deletionem recipi non posse monuit Snell | jW[f¦_[Wb S¦QgQ Snell : SYQg] vel jW[fg_[�Q VYSW Kambylis | ìThe connexion of thought might, therefore, be: Do not, then, ridicule (me?) asóold fogey, boor, savage?î Lobel || 5 Y] H34 || 6 bkíR H34 | cbW['k H34 : ìit must have been written over the wrong bî Lobel || 7 nbf� H34 | XRQp[¬] pot. qu. XRQp[ÏQ sec. Lobel, at XRQp[ÏQ mihi quidem elegantius videtur | ìThe lost verb presumably had the sense of jg]_W~Qî Lobel : �SnVVWbQ van Groningen, fort. recte || finem epodi (?) coronide cuius vestigia iuxta v. 6 exstant (at ìnot a normal coronisî Lobel) notat H34 || 8 ìpropter metrum expectes c]Q[ vel j]Q[; c]Q[Yg_]WQ longius spatio, c]Q[W�]]WQ lacunam expleret; sed fort. ~ (b) [hic (c)], 9î Snell | [nc H34 || 9 ìV�][Rb fW jRÚ —[_Q-gb, Tb, gb]î Lobel | �b� H34 || 10 _WV[^d[]TQ pot. qu. _WV[^�]TQ Snell || 11 ìj[RÚ] lacunam
146
expleretî Snell || 12 suppl. Snell | fgêb H34 || 13 ì]f[[Y]°Wb], f[[Y]°Wb], f[Y[]°Wb] sc. õfg[î Snell | �ãc H34 | suppl. Snell Commentarium ap. G37 servatum : . . . ] ¥WQ[ ]Q ¥[ H^Qk]�[¦] d\[]b cRf[^]Éí �[ZR^R[Q Z]R^fRQ cR��Y ��[gP 5 fg]t�fg kÓ kbpP[R[_��kW] ] ¥b jRÚ �cÚ fgt p[ ]k\] fnffW]p�[b ]WQRVjRQWgb]dbV[ ]×Q ¯]ZRfgQ f�Q [ 10 margo quae ratio H37 cum *F4(a) habeat incertum est, v. supra, pp. 25sq. || quaecumque non notantur dispexit articulavit supplevit Lobel 1961g || 1 fort. jfWQ^ Lobel cl. *F4(a).6 || 4sq. citantur *F4(a).6sq.; incertum utrum lemma fuerint an commemorati sint a viro docto de alio carmine disserenti || 6 fg]t¥fg kÓ pot. qu. ] ¥ f¦kW | kbpP[R[_��kW] Lobel : kbpP[R_[�bj¦Q, kbpP[R[_�bj�], kbpP[R[_�Tk�] possis, v. Prodi 2013 p. 53 adn. 4 || 9 ìanother lemma or quotationî Lobel, nunc fr. dub. 348a Sn.-M. | de �Vj¬b agi videtur; utrum �VjÏ QYgb] d^V[R an �VjÏQ §g~] d^V[gb] an aliud fuerit in dubio manet || 10 sic articulavi Scholium : inf. Zg[kÏ] W∂cWQ fgˆ] d�[¦SSgP] jRÚ fÏ] Ú[]�gQ^R] sic articulavit Snell 19643 duce Lobel : ¢ pot. qu. jRÚ McNamee 2007 p. 348 ut vid. | Zg[kR^ nunc fr. dub. 384b Sn.-M., at de casu nihil constat
147
(b) . . . ]T�[ ] ¥× ¥��ÉTQ[ ]] jRÚ V�[�][� ¥[ ] ¥_R�T�[ ] ¥[ 5 perierunt 11 v. ] ¥ ¥[ ¥] ¥ ¥[ ] ¥Q dWPS¼[ ] ¥Q ¥Q ¥ ¥ ¥[ ] ¥× ¥kí Õ ����] ¥b R� �[ (20) ] ¥_ ¥ ¥� QPjfÙ¥] —c ¥[ . . . 1-5 H34 frr. 2(c) || (17)-(21) H34 fr. 2(d) || quo in ordine haec frr. ponenda sint ostendit Lobel 1961f p. 135, qui et ex eodem jgVVq_Rfb ac frr. 2(a)-(b) (hic fr. (c) pars laeva) ea esse vidit. siquidem frr. 3(a)-(b) (hic fr. (c) pars dextra), quae in alio jgVVq_Rfb fuerunt, iuxta frr. 2(a)-(b) ponenda sunt jgVVq]Wb interposita, ut ipse verisimiliter coniecit, necesse est haec frr. ante (c) stetisse, ut ostendit DíAlessio 1991 p. 116. si (c) ante (a) stabat (de qua re nihil scimus) et fere initium carminis praebet, nihil obstat quin haec frr. (vel 2(c) solum) ad carmen prius spectent quaecumque non notantur dispexit articulavit supplevit Lobel 1961f || 3 ]] jRÚ articulavit Snell | bV� H34 | extrema vestigia litterae f dispici expectes, at nisi accentus erratus est, quid aliud suppleri potest? || (18) articulavit Snell || (19) ]��×� dub. Snell || (20) kí� H34 | Õ�[b]Rb ( Snell duce Lobel, at k pot. qu. R ( videtur cl. (c).11 ÈWk^TQ | Úb H34 || (21) articulavit Snell | ]�c H34
148
(c) . . . � � �] �� � �[ � �]cRQf � �[ � �] �R] ³VV×b� �[ �]k �[ � � �] �RQ� ¡ Éí �c�n��[W ] �] �[ �]f[R �[ 5 ] [ ] � � �]�RbSbQ ��gQ� n[ Z]n �[bQ �_]dYcTQ Z[P] [g]c[V¦jgb] ¯cV]é�R )g^]Rb] [ õ®] QY]_g_Rb cR�Ï [ ] HR[]QR]]^É� �[ �] �×[ �j[g]f¦_gb[]] 10 cY]f[Rb]b *b�[R[ ] ¥ � � �� ÈWk^TQ � � �] � �Q WÃjn[È[gP ZpgQ]Ù¥] ¿[_]+R�¦�Q� g–pí µ c[- cgb]b]� � �Sn��g�[ ] �[ �]P[ �]_RQ[ ] � �[ ] � �T �[ ] � �[ �]R� �[ ]é[×[ ] �pé]¥ �[ 15 ] � �̀ [ ] � �[ ] ]�� � ��[ ]]¦_WQg[ ] � �É×[ ]jfWRQ¼[ ] ¥[ ] ¥ ¥f ¥[ . . . 1-16 (pars laeva) H34 fr. 2(a) || 6-12 (pars dextra) H34 fr. 3(a) || 13-19 (pars media) H34 fr. 2(b) || 13-19 (pars dextra) H34 fr. 3(b) || quae ratio fr. 2(a) cum 2(b) atque fr. 3(a) cum 3(b) habuisset ostendit Lobel 1961f p. 135, qui etiam probabiliter coniecit sic frr. 3(a)-(b) iuxta 2(a)-(b) stetisse jgVVq]Wb interposita. quantum frr. 3(a)-(b) a 2(a)-(b) distiterint parum constat; haec dispositio e supplementis quae Snell proposuit pendet. quo in ordine (b)-(c) et (a) ponenda sint non liquet quaecumque non notantur dispexit articulavit supplevit Lobel 1961f || 3 articulavit Snell 19643 | ×b� H34 | °[WP]k dub. Snell, longius spatio || 4 Q�¡ H34 | -� H34 | ¡ ÉÓ c�n��[W possis, pot. qu. ≈ÉW c�n��[W vel ≈Éí �c�nP�[W || 5 ]�]� []]f[R�[ dub. Snell || 7 jPn]�RbSbQ pot. qu. _WVn]�RbSbQ Lobel | Zp¦Q(R) Lobel : vestigia R similiora quam Z mihi videntur | gQ�n H34 (ìanomalous both in shape and position, but it does not look like the apostropheî Lobel) | ³[SRV_R vel ³[pP[_R Ferrari ap. DíAlessio 1991 p. 116 : ³[]fWT] Snell, obl. DíAlessio | j^p]�[bQ nol. Snell | versus aliter divisos esse atque in (a).5sq. suspicatus est DíAlessio (i.e. [ . £ ˙ � ] � � ˙ ˙ [ � ˙ ] ˙ � ) : corruptelam in medio versu (i.e. . ˙ ] � � Ü˙Ü ˙ [ ˙ � ˙ ] ˙ � ) spatiis aptiorem putes || 8 [V¦jgb] Snell 1962 p. 6 : fort. [fY[gb] possis | ¯cV]éXR nol. Snell 19643, def. Ferrari : W–k]×�R Snell 1962, quod vestigiis parum congruit : ¯[]éXR ¯V]éXR ‡[]éXR nol. Snell 19643 || 9 fin. ìcYf[Rb or the likeî Lobel : j[nQRb Snell 1962 || 10 ]b$×%b sscr. É pot. qu. � H34 | ]¾×[ vel ]�×[ | �j[g]f¦_gb[]] Snell 19643 || 11 cY]�[Rb]b Snell : �[g]�¥[Rb]b Lobel | æ^[[R[] �j vel æb[[R[^TQ Snell || 12 ¯cWb]]¥�Q dub. Snell | jn[ H34 | [gP ZpgQ]Ù¥] Snell | �¦¥Ppí H34 || 13 Sn� H34 nisi fallor : ��SR��¦�[WQg] Snell || 14 cRf[]iÇT� [ possis (¼T iam Snell) || 17 ��é� Lobel | ]¦_ H34 Scholium : 9 Wg_� ¥ [Q]Yg_R[b] Lobel
149
(d) . . . ¥ ¥]�pRQR[ ¥ ¥]W[R ]W[ ¥ ¥�¥]VgQ ¥ ¥ ¥]Wkgb] ¥[ ¥ ¥] ¥pWb]WQ[ 5 H26 fr. 86.1-5 (huc traxi 2013 pp. 54sq., v. supra p. 14) 2 ]W[R,]W[ H26 : ]W[R ]W[ Snell 19643 || 4 kRc]Ykgb] Rutherford 2001 || 5 ]� pot. qu. ]�
150
F5
(Pae. 6.123-83 = fr. 52f.123-83 Sn.-M. = D6.123-83 R.) o∞S 1bQq[fR]bc [W∞]c o∞¥R[jÙ]� c[gc[¦]Éb[g]Q ¿Qg2
�_Rj�V�fR Sn[ �]]b }T[bW~ _[W]kYgb]� [c¦]�fT[b] Q¬]g] , [‚] 3}4b3Ù] �4V- (125) VRQ^g4P dRWQQÙ3Q ³]f[gQ� g—4QWjWQ g– ]W c3Rb\¦QTQ 5 ³kg[4cgQ WÃ�QnX
3g_�WQ, �VVí �gbk¬Q 5¦pb4R kW½g_YQR3 jRfW[W~] c¦4pWQ�
4¯VR�W] Q3RPc[�fRQbQ (130) kR^_gQR 4jRÚ fÏQ pW3_^X4WQg�Q �
3[Wf[nQ. ¡ cnQfR fgb f4n fW jR3Ú fÏ f4W�ZTQ 10 ]ÙQ �SSPnVb4XWQ ƒV3�gQ WÃ[�g[cR] æ[4¦QgP c3R~]� 4Õknf�W]]�b kí �3cí ��]T- c43gt� c[gfí �]c
4Ù3 c4[gp�3[TQ 4�Rp�3jgV- (135) c43gQ �[Q]é[�Y°Rfg c�R[pYQgQ o3Ò4SbQ3RQ� f¦fW Z[�]WRb �- 15 Y4[g3] ¯j[P°RQ j¦�[R]b 4�cbZi[bgQ jR3fn]jbgQ Q�fgQ Õ_Yf4W[gQ, µQ3R VWZYTQ �cí �_�4[¦fTQ (140) d[ ] ¥ Rb3 ¥[ ]�� 20 ÙP[[_bkgQñ ® fg ¥[ ]Td[]×QY]fRfgQ }bÙ1] [ ] (145) cW1�[ ] cg[ ]gQ 25 cR[ ]TQ� XW[Qñ⸥ ]Y__WQ Ú�V^Tb� jP[ ] � �[ ]WkW- (150) �R[ ] � ���[×4] ̂ R1b Q×[ WÃ]nQp4W_gQ � �1 �QRbÉTQ� 30 f[[ ] 67jW�4Qg~g Q¬1]gb ]é[ ] ¥[ � �]×tQ� g1µ �î[ ] c[Ú1Q CfPSÙ] ≈[jb×Q �X W–- (155) ] Ébjn �]Rb perierunt vv. 35-46 (157-68) ]� � �[ ¥] �[ ] jVPf¬] ÒkT ¥[ (170)
151
ZR]VjgZn[_�� [ ]kR jW2Zg1VT_23YQg] 50 ]¼b ]�[⸥⸥ �¥]W] ¥ ¥ ¥ ¥]� _P[^R[b] V]¦Sgb f�23Q SW 4k×- (175) [bjf]�cTQ [jVY]WbQ �c23W^[g4QR]3 �[Wfn4] o∞Rj]�k¬Q� d[bVW~]�W 55 ¥ ¥ ¥ ¥]� 4c¦VbQ3 4cR2f[i3bRQ, db- VW~fW] Éí W–+[[g]�� VR¦3Q f¦QÉé [jRÚ] ]3fWdn2Qgb]^ �QbQ� cRQ- (180) pRVYg] ÕSbW[^R]] ]jbn�WfW� Ùgb]¬4Q �¥cR�gVY¥×Q�[R] cgVVnjb, HRbnQ, kY- 60 Xí] �QQ¦_TQ �¥[Qgc]¬Q. tit., 1-33 MH4 pp. xxx.13-xxxii 1 || tit., 1 �H
26 fr. 86.6sq. � (huc traxit DíAlessio 1997 p. 37 adn. 92) || 1 [I] = | (bT) Il. 22.51e p. V 274 Erbse ¿Qg_njVPfg]� �[^]fR[Zg] Õdí 8Q ›] cR]b_YVgP]R (Od. 12.70) (c¬]b _YVgP]R T). �Q kÓ ∆kP]]W^Rb ƒQg_R jVPfÙ] oÒpTQ (19.183) jRfÏ cR[npW]bQ. W∞ kY dR]bQ� gÃj ¯]fbQ (om. T) �X gÃkWfY[gP jRÚ �[]WQbjgt ]�QpWfgQ, f^ �]fb fÙ �]fPnQRX jRÚ cgb\_RfgS[ndg]; jRfÏ ]�QpW]bQ g“Q �]fbQ, ›] fgX¦jVPfg]� ¯]fb SgtQ cR[Ï HbQkn[Tb fÙ p\VPjÙQ RÃfgt �Q cR[TQ�_Tb ZR[Rjfv[b� ¿Qg_njVPfR Sn[ �]fbQ (fgX¦jVPfg] ® �]fbQ T : fÙ fgX¦jVPfg] cR[Ï HbQkn[Tb jRÚ cW[^jVPfg] b) || 3-13 4H5 fr. IIIr 3 || 6-9 �H
7 fr. 15 � || 9-15 (pars laeva) 4H5 fr. XIIr 3 (huc traxit Snell pp.
425sq.) || 9-13 (pars dextra) 4H5 fr. XIr 3 (huc traxit Snell) || 12-14 �H7 fr. 16 � (huc traxit Snell p. 431 duce
Hunt 1922b) || 13-25 (pars laeva) 4H5 fr. IVr 3 || 16-20 (pars dextra) 4H5 fr. Vr 3 || 22-34 (pars dextra) 4H5 fr. VIr 3 (huc traxit Snell pp. 425sq.) || 28-33 (pars dextra) MH4 fr. 46 1 (huc traxit Henry 2003b) || 47-57 4H5 fr. IIIv 3 || 50, 53-61 MH4 pp. xxxiv-xxxv.12 1 || 52-7 (pars laeva) 4H5 fr. XIv 3 (huc traxit Snell pp. 426sq.) || 53sq. (pars media) 4H5 fr. XIIv 3 (huc traxit Snell) || 54, 59 (pars dextra) 4H5 fr. IVv 3 || 58-61 (pars laeva) 4H5 fr. Vv 3 : de recta positione huius fr. et initiorum vv. 58-61 v. DíAlessio ap. eund. et Ferrari 1988 p. 163 || 59sq. fort. inveniuntur in H7 fr. 99 quaecumque non notantur dispexerunt articulaverunt suppleverunt Grenfell et Hunt 1908a (GH). non noto (a) ea quae a viris doctis lecta vel coniecta a fragmentis postea additis sunt refutata, (b) quae hariolatus est Bornemann 1928 || finem carminis prioris (Pae. 6(a) sive Pae. 6.1-122) coronide et asterisco notat H4
(asteriscum dispexit Rutherford 1997a p. 4) || tit. in marg. habet H4 (titulum id esse agnovit Rutherford), in p. H26 | o∞S[bQqfR]b] Diehl 19081 ducibus GH p. 98, cetera Rutherford | dubium utrum c[gc¦kbgQ H26 quoque praebuerit || 1 ]_RjV�f� H4 praeeunte Bekker 1825 p. 589 : ¿Qg_njVPfR [I] T : ¿QP_RjV�fR Schroeder 19005 p. 494, fort. recte | Sn[ �]fb [I] T : Sí ¯�W]]b GH : Sn �� �]]b ci. Radt 1958 p. 173 ex [I], in H4 dispexit Rutherford p. 1 : ìI would not rule out the possibility that the situation of the papyrus may be explained as an attempt to correct an original Sn[ by partly erasing itî DíAlessio 1997 p. 57 adn. 193 || 3sq. é[VVR£Qbg]P H5 || 4 [gQ� H4 || 5 g�cW H4H5 : gà ]Y Currie || 6 nkg H4 | Qn H5 | _WQ� H4 | k¬Q H4 || 7 ¦pb# H4 | kW½g_YQR H4 (kWZg_YQR sec. GH) : ¾g_YQ� H5 (kéjg_YQ� sec. Vitelli 1913) | kWZg_YQRQ || 8 c¦p H4 | �¥ ¥�� H5 | ÏPc[� H4 || 9 f�Q H5 | Ó_^ H4 || 10 nQfn H4 | fnfW H5 | é�Z H5 || 11 WSSPRVb4XWQ H4H5 : �g� sscr. H5 || 12 WP[� H4 : Ó¥P[� H5 | ¦QgP H5 | Rb]� H4 | PkRfbkíWc H4 : �kRfbkíWc H5 : corr. GH : ÕkRf\ZYg] Maas per litt. ad Hunt missas 19.X.13 quae nunc in bibliotheca Sackleriana asservantur 304 P.71/3 (ìein Beiwort von �]Tcgt oder von æ[¦QgP cR~], das ich nicht raten kannî 1913-21 p. I 26) : pPSRf[�Q �cW^ Sitzler 1915 p. 387 || 13 cgt H4 | ìor perhaps c[gpí Õ]c¦î GH || 13 �Ïp H5 || 13sq. �RpPj¦VcTQ Maas, obl. Koster 19532 p. 54 adn. 1 || 14 �¥[Q]é[Y°Rfg Maas : �¥[Q]�[Y°Rfg GH, def. Koster || 15 QRQ� H4 | 9]WRb : H4 | hunc v. nota > notat H4 || 15sq. RW]£[g] H5 || 16 Wj[P°RfRQ H4 : corr. GH | �[ ;¥]b H4 | ¾i[ H5 | hunc v. nota > notat H4 || 17 fnc H4 | Ú_YfW[gQ Calder 1977 || 18 ëÔQR H4 | [¦f H5 | hunc v. nota > notat H4 || 19-21 d[^VgQ S¦QgQ ¯fWjWQ]£ R∞][^Rb ]ˆQ WÃfPZ^Rb]£ ÙP[[_bk¦QTQ ³QRjfR e.g. Snell 19531 || 19 fort. de dPfW�WbQ agitur || 20 �[, é[, ×[, ]¥[ | ]�� Vitelli 1913 : fort. ]� || 21 ÙP[[_bk¦QWc Wilamowitz 1922 p. 134 : _P[[^ñ dub. Turyn 1948 || finem strophae paragrapho notant H4H5 || 22-5 fÙQ [�QÚ �[gfg~cb cTd[]gQYcfRfgQ £ }bÙc [P�ÙQ W∂QRb]£ cW�[pg_Rb,
152
c[�fRQbQ]£ c¦[QfbgQ e.g. Snell || 22 ì]Td[g] vel WPSW]î Snell 1938 p. 426 : ]Rgd[]×QY]fRfgQ possis || 23 kbÙ] H5 || 24 cW� H5 || 25 ita H4 : littera altera in H5 × non videtur esse || 26 àQ� H5 dispexit Snell || 27 init. suppl. Snell | Y_ H5 | ¯__WQ probabiliter Radt, sed et compositum possis | ÚV^Tb� H5 || 28 jˆ H4 | jP[RQ¦cVgjg] vel jP[RQ¦cVgjgQ Turyn : jP[RQWgñ dub. Snell 19531 | fin. dispexit Radt : ]�kW Vitelli : ]]ˆ dub. Snell || 28sq. ìkWXn_WQg], kYXRfg vel sim.î Snell 1938 p. 426 : fort. �k- Radt p. 186 || 29 fin. GH, Henry | ^R' H5 : b�[ H4 || 30 Q¦¥[]fgQ Turyn | WÃ]nQpW_gQ pot. qu. ³QpW_gQ Henry : �QpY_TQ GH | ¼�Q×[ sscr. H4 (GH : �gbQ×[ Henry : ¼[ possis) : i.e. �QpY_Tb Q×[ Rutherford 2001 | postea lectio incertissima : � �[ H4 : fort. ¥� à �� sscr. � H5 : fin. RbkTQ corr. in RWbkTQ vel R\kTQ sec. Norsa ap. Snell : ìparrebbe W_gQiQR[kgQ, che non so cosa possa essereî Vitelli : �_¦Q cgfí �W^kTQ dub. Snell, at cgf vestigiis minime congruit | ÉTQ� H5 || 31 f[[ndgQfb kí] e.g. Henry | fin. Henry : ]× jY¥�î] ∞gQ �[ (fort. ∏¦Q�[g]) GH | Q<bgQ�ê H5 (¬ Henry : = Vitelli : Û Snell) || 32 ]W[ Snell : ]p[ GH : ]�[YQg] Turyn : ]Y[�R] Henry | SWQW¬] kbcV]gtQ nol. Henry | ×PQ�¡ H4 : gêPQ[ ]¡ H5 spatium ad distinctionem aptum praebens | gµ Snell || 33 >\[Q Snell | SÙ]≈[ H5 | X¯P H5 (an XíYP apostropho divisionem verborum indicante?) || �X W–[Qgbí ¿_¦]RQfR d[WQÙ] Snell 19531 || 34 dispexit Norsa ap. Snell 1938 || 48 ìjVPfÏ], non jVPf¬]î perperam Snell | �ê]Ô H5 (� Vitelli) | �[ Snell : î[ possis, duce Snell || 49 suppl. Vitelli | Zn[ H5 | _�b H5
? || 50 init. cgbQ? suppl. Snell 19531 e | | k� H5 | suppl. Vitelli || 51sq. fines versuum agnovit Norsa ap. Snell 1938, contra Vitelli | –__b _ÓQ g“Q, p]é×^, £[SY[R] f¦kí ¿]�[P_WV]Y] e.g. Snell 19531 || 51 ]¼b Snell : ]gb Vitelli || 52 ]Y] Vitelli : recte dispexit Snell || 53 _P[^R[b] supplevi | V]¦Sgb Vitelli | f�Q Radt | [^R H5 | ¦¥Sgbf� H5 | SW sscr. �f� H4 || 53sq. kg£[[bjf]�cTQ Ferrari ap. DíAlessio et eund. 1988 pp. 172sq. | VP[¬]� _P[^R[Q dV]¦Sí ¿c�Q fW k¦£[_gb]] jf�cgb Q[Y_]WbQ Snell || 54 �cT H5 | [jVY]WbQ vel (longius spatio) [dVYS]WbQ Radt : [VYS]WbQ DíAlessio p. 173 | ]W^[gQR] H4 : ]Qg] (scil. �[Wf¬]?) H5 | �[Wfn] : ¿[Sn] dub. Maas 19233 p. 11 || 55 o∞Rj]�k¬Q Turyn | k¬Q� H5 | d[bVW~]�W Snell : d[bVW~Q] �W dub. Radt, obl. Ferrari || 56 kY ]d]� Henry 2003a p. 12 : ]dYT]� Ferrari p. 175 : ]�Q ]d]� DíAlessio p. 175, obl. Ferrari : pWg]Ú Battezzato ap. Ferrari 1999 p. 154 adn. 17, obl. Henry : c[Ù]]¥ Turyn, obl. Radt pp. 93sq. : fort. kí –__]� possis | ]�[TÔRQ H4 : cRf[¼[ ¥ ¥] ¥c¦VbQ (commutatis verbis contra metrum) H5 : cRf[^RQ Snell 1938 || 56sq. db£[VW~fW] Ferrari ap. DíAlessio et eund. pp. 163, 175 : d^£[VTQ] Snell || 57 ¥íW� H5 | ��d[[gQ]� Snell 19531, corr. Ferrari duce Radt p. 190 : éÃ¥n �[Qg[]� Turyn | ��Ï� H5 (��Ù¥ sec. Vitelli, sed primum vestigium huius lineae � videtur esse non �) || 58 f¦QÉé [jRÚ] DíAlessio p. 163 : jRÚ] SYQg] Snell 1938, obl. Erbse ap. eund. 19531 : ¯j]SgQ×[Q] vel ]�S]SgQ×[Q] nol. Snell | ]fWdnQgb]b suppl. GH e | | QbQ suppl. DíAlessio ap. eund. et Ferrari 1988 p. 164 e | | c¬Q GH : cnQ Schroeder 19236 || 58sq. cRQ£pRVYg] DíAlessio p. 164 : cRQ£[WP]pRVYg] vel cnQ£[fí WÃ]pRVYg] nol. Snell || 59 [WÃ]pRVYg] Vitelli | ÕSbW[^R]] Vitelli | Yg]� H5 | fW� H4 | ]¬Q H4 || 60 kí] Snell 1938 (delendum sec. Turyn) : fí] dub. Snell 19531 : ]dí] Hoekstra 1962 p. 7, longius spatio | �¥cR�g�YgQ�[b] Snell 1938 : �cR�gVYgQf[R] Ferrari p. 177 : � �cR�gVY �¼Q �[Ù] Hoekstra : [�]cR�gV[^R]Q Wilamowitz | ��gV� H5 | jb�c H4 | cRb¬Q H4 (fort. servandum) : corr. GH | kY H4 || 60sq. kY£[Xí] Snell, kYXRb praebuisse H5 ratus : kY £[Sí] Diehl 19173 (kY iam 19081 ducibus GH) || 61 ¯QQg_gQ Turyn : �Qk^jTQ e | Puech 19221-231 | Q¦_ H5 | ¬Q H4 | �¥[Qgc]¬Q Ferrari p. 178 : �[RVb]¬Q vel �[P]b]¬Q Rutherford : �[RV^]RQ Snell 1938 (vel �[P]^]RQ, 19531) : �¥[P]RQ Puech, brevius spatio || finem carminis paragrapho notat H5 Scholia (G4) : tit. (?) ]$ ¥�× ¥[ ¥] ¥½��%[ £ ]$ ¥��[ ¥] ¥[ ¥]b ¥%[ £ ]$È�×]¥[ ¥] ¥%[ 2 (N2) �Q f�b �Æ [f]�ÇQ c[g]gk^[T]Q dY[WfRb 3 (N1) �W[ÙQ }bÙ] ��[V]\Q^gP [�]Q o∞[S]^Q\b ≈cgP ]PQWVp¦QfW] ×[� ²VV\QW]]£ W–��[Q]fg cW[Ú fgt RÃZ_gt 7 (N2) ×kW[ 8 (O1) \f[ ± 18 ]\� £ RÈ[ ± 18 ]Q fbQW] £ ¥[ ± 18 ]]fbîR £ [ ± 13 ? W∞kT]�gcgbW~ 12 (O2) ± 9 ? kb]�f[^�gP£[]RQ ± 6 ? c[g]��¥[TQ ��£[pàQ ± 8 ? ñf]g 16 f[ 50 (N1) cÙbQ# 53 (O2) c[g]fRjfbj�] 58 (O1) ]fWdnQgb]^ QbQ È[RQ 59a (N2) �[(b]fg)Q^(jg]) ½[ 59b (N) gb�[ 61a (O2) f�Q �cÙ f�Q ��[k]^ jTQ 61b (N2) �\(fW~fRb) WbVbg_RQ
153
tit. sic GH : equidem nihil dispicere potui | de Aristarcho vel Aristodemo de genere huius carminis disserentibus verisimiliter cogitavit Currie 2005 p. 324 adn. 155 : vel de Aristonico agi potest, v. infra p. 228 adn. 3 || 2 fr. 108 addito suppl. DíAlessio et Rutherford ap. eund. 1997a p. 7 : �Q f�b �[[b(]fg)]Q(^jgP) nol. Rutherford : �Q f�b c[¦Qf]Tb Diehl 19081 || 3 ×[� ²VV\QW]] McNamee 2007 p. 339 duce Radt p. 175 || 7 ¡kW[ GH | ex eodem scholio atque 8 sec. McNamee 2007 p. 340 || 8 ¶f[gb dub. GH | �c[ GH | ]] fbQR GH : fort. ]] fb ¢� R- || 12 fUQ oÒSbQRQ kb]�f[^�gP£[]RQ �cÚ f�Q c[g]��¥[TQ ��£[pàQ �QW]cn]Rf]g GH || 16 f[UQ Qv]gQ, f[Ù �VVqQbgQ ƒ[g], sim. possis || 50 sic H4 || 58 È[RQ DíAlessio p. 164 : ìperhaps the initial of the name of the critic who supported the readingî GH || 59a �[(b]f¦)Qb(jg]) McNamee 1977 p. 75 : �[(b]fgdn)Q(\]) GH || 61a ��[k] ̂jTQ suppl. Vitelli p. X : �½[k]^ jTQ Diehl || 62b �\(fW~fRb) McNamee pp. 124-8 (sed ìRutherfordî ap. eand. 2007 p. 341), v. et Lobel 1961d p. 42 : �q(fWb) ead. 1981a p. 35 : >\(Q¦kgfg]) GH, def. Pfeiffer 1968 p. 118 | W∞c¦_RQ Diehl : �QQ¦_TQ Puech 19221-231
154
*F6
(Pae. 17 = fr. 52r Sn.-M. = S6 R.) (a) margo ] �pWQb �[ ]�QP[ ]_¦Qí ∆�P_[cñ ] �fgQ gà 5\f[Ù]� [ kbT]�PS^Rb] dPfWPg[ 5 gÃ][RQg_n½WR [ ]Q f×tfg �RVVW�[ ]ÏQ �Rp�[jg]VÈ[gñ ]]P[ . . .
(b) . . . ]Y¥T[ ]Tb �[ ]Q �Y¥��[ ]� �é cRb\¦QTQ [ ]+g[b¬Q cWfR�[ 5 ]é[ (a) H26 fr. 6 || (b) H26 fr. 7.1-6 || haec frr. ad eandem paginam spectare ostendit Lobel 1961d. hoc ordine inter se ponenda mihi videntur propter situm jgVVq]WT]. culmen et fundum paginae ea esse fere 12 vv. intervenientibus coniecit Snell 19643, sed non certus sum fr. 7 in fine paginae stetisse, nec puto minus ac 14 vv. deesse (fort. 16) si ibi fuerit, v. supra pp. 132-4. quaecumque non notantur dispexit articulavit supplevit Lobel || (a) 2 Q�AB H26 | QP⟦�⟧ vel QP$È% H26 || 3 ÚSW]_¦Qí ∆V�_[cbgQ ìor the likeî Lobel | ¦Qíg H26 || 5 ‹]�PS^Rb] Rutherford 2001, quod vestigiis parum congruit | ìthe missing noun might be e.g. k¦XRQ or fb_n] (Ö) Then (kb)TVPS^Rb] might qualify e.g. �gbkR~]î Lobel || 6 _-½ H26 || 7 �RVVY_[WQ dub. Lobel, fort. recte || 8 ÏQ H26 | p� H26 || (b) 1 Y�T H26 ut vid. (Ó¥T Lobel) || 2 �[ Snell : vel È[ || 3 �� [ H26 | articulavit Snell 19643 || 4 ]� H26 | �é dispexit Snell | \¦Q H26 || 5 ]fWdRQR]+g[b¬Q DíAlessio 1991 p. 103 : QbjR]+g[b¬Q Lobel : jRÚ kRdQR]+g[b¬Q Snell : ‹]Zg]+g[b¬Q Rutherford (at esset ‹]Zgdg[^TQ)
155
*F7
(Pae. 18 = fr. 52s Sn.-M. = S7 R.) �][SW^gb] é∞¥] �[gˆ]] ÌVWjf[�¼[Qg] cR~kR] �Q ÁPQ]kR[bk¬Q �W[�b fW_Y]QWb cWdPfWP_YQgQ ³[V]g] �Qk][Ú ]gd�b ÈR[YZWb _YVg] [ ] ¥Qí �_dÚ c¦VbQ dVWSW[ ]Q —_QTQ ]YVR] �X �jR���[fgñ 5 ] ¥ ¥[ ¥]� ¥P QR_WQg] g– jWQ �] ��_�cVRj[^RQ c]W[Ú [}]R�kRQ^Rb �]¼b g∑n cgfW rq�Rb ]fW jRÚ Ú�[^]jR QR�VgZgb ]qVR]R� [�]QQ�ZbgQ j[PdR[ 10 ]VWj ¥[ ]� ¥[.] margo ? H26 fr. 7.7-18 quaecumque non notantur dispexit articulavit supplevit Lobel 1961d || tit. ex ][SWbgP] corr. H7 (2) | é∞¥] [fgˆ]] ÌVWjf[�¼[Qg] cR~kR] vel ÌVWjf[P¼[Q^kR] DíAlessio 1997 p. 41 : é∞ ]¥ [fÏ]] ÌVWjf[�T[Qg] pP]^R] vel §g[fn] Snell 19643 : �Ç]¥[cÚ]] ÌVWjf[�TQg] Rutherford 2001 duce Lobel (ìI suppose ] to be the end of a noun and ÌVWjf[PT[ to represent a genitiveî) || 1 �Q suppl. Snell | k¬QÔ H26 || 3 dVYSW[fRb (dub. Gerber 2002 p. 31) pot. qu. dVYSW[b || 4 W–d[]×Qí Rutherford | QíR H26 | c¦V H26 || 5 RjR_RQ ([ H26, compositum indicans : ìperhaps ëfrom untiring mouth(s)íî Lobel || 6-9 _P]�î[]]¦¥_WQg] g– jWQ �] �cVRj[^RQ cY]gb_b £ ›] ¯[b]RQ c]W[Ú [}]R[kRQ^Rb £[¢ (Ö)] g∑n cgfW rq�Rb £[�SYQWfg] e.g. Snell (cnVRb add. ap. Maehler 19754 p. 217) || 6 ]�[ Snell | ]¦P H26 | RcVRj[ H26 : corr. Whittle ap. Friis Johansen et eund. 1980 p. II 185 || 7 ^Rb" H26 || 8 ¡~n H26 : gµR nol. DíAlessio p. 42 adn. 120 | q�Rb" H26 || 9 RbC H26 (disp. DíAlessio 1991 p. 103 : RbÚ Lobel) | R�V H26 || 10 Á\VW�¦W] �c]qVR]RQ sc. �¦R] Snell ap. Maehler 19754 p. 217 : Á\VW�¦Rb rectius Pavese 1993 p. 154, cf. id. 2006 p. 587 | qV H26 | QPZ�b�gQ H26 | ìapparently j[Pd¬b or a form of j[PdR~g]î Lobel : fort. j[�dR possis | ìabove R[ a dot, presumably part of a lection-signî Lobel || 11 Ì]VWj�[[�TQ dub. Lobel, fort. recte : �] possis | ]�é[ vel ]�]¥[ | �]�nQfW] (�]�n]RQfW] Pavese 1993 p. 155) Ì]VWj�[[�TQg]] jf[q_RfR vel jf[YRfR e.g. Snell ap. Maehler 19754, quod vestigiis parum convenit
156
F8
(Pae. 20 = fr. 52u Sn.-M. = S1 R.) . . . ]¼ ¥_R ¥[ ] ¥[ ] ¥ ¥[ ¥ ¥]]¥f[ ] ¥ ¥gQfTQ �b�[ ¥[ ¥ ¥]R[ ] ¥ �VjR�kR ¥[ ½ ¥ ¥¾ ¥[ ] 5 �c¬S×� [ _g[]�×��¥XbR] ¥ ¥�WQ ¥ ¥[ ] ¥[ ¥] É[b]Ï �P[�¬Q �cWb�k ¥[ ƒ¥+�é] pW¦¥cg_È[gb ] ��~¥D×[Q] �cÚ �[Ydg] gÃ�RQ �̂gP }b¦�] �]]W]�¥[g]Qpí, ¡ kí �Qf^gQ �QÏ �jn[�R fí ³Wb�[W 10 ] ¾W��Ú �WVYTQ ³cg cgbj^V×� ]cn]��R�gQ ¯[[b°WQ §nQ fí ¯dR�WQ +�nQ ] ¿¥[__]nfTQ ³cg ]YVR] �k^QR]WQ. ] ³ÈWcVg] �j VWZY¼� QWgf¦½TQ RÃf]¦�[W]Q ƒ[gP]W cW[Ú d¦�Tb� 15 ] ¥ �×∂¥�jgQ �_dbf[�TQg] kW^]��Rfb ]�Z¦_WQRb d�SgQ ] ¥�R c¬�]Rb �]��+^c�gV[gb] æé+[RV]VR�[ ] ¥R ¥[ ]\[�[ 20 ]W]q[ ]��[ ] ¥ ¥� . . . 1-23 H26 fr. 32 p. i || 6-10 �H
7 fr. 31� || 16-19 �H7 fr. 139 �
quaecumque non notantur dispexit articulavit supplevit Lobel 1961d || 1 ]¼� dub. Lobel || 2 atramen post ] ¥ fort. id est quod Anglice offset ink vocatur || 3 fin. vel �[ dub. Snell : at quid quaeso �bV- ? || 4 ]È Lobel : vel ]]¥ | R�k H26 || 5 ]�, ]� Lobel : ½[R]�¾�[ dub. Snell || 6 �cRS×�[ Lobel (ìHercules puer non timuit �cRSg_[YQR] _ _ ] _g[_.?î Snell 19643) : �[ pot. qu. �[ mihi quidem videtur, quapropter �c¬S×� [ conieci | �¥Xå# H26 || 7 fort. WQ�[ | [¬QÓ H26 | Wcb H7 || 8 ¦db H26 : ìnow less clear that at an earlier stageî Lobel || 9 ��~¥D×� Snell 1962 p. 6 | Q^g H26 || 10 Qf¡ sscr. p¡¥ H26 (2) (disp. Colomo : Qfí¡ Lobel : Qf¡ sscr. pí Snell) | �]]W]�¥[g]Qpí Snell | F[�fín H26 || 11 QYRb fW] dub. Snell 19643 | j^V H26 || 12 Y[[Wb°WQ H26 : corr. Lobel | §�QfíY H26 | �Q� H26 || 13 f�Q fí] vel _YSR kí] dub. Snell | nfTQÏ H26 | kGQn H26 atramento amplius addito super accentum acutum, circumflexo simili | Q� H26 || 14 �Vj_qQR kí] vel Ú kÓ _nf\[] dub. Snell | ]n H26 | f¦ ¥ H26 || 15 RÃf]¦�[W]Q ƒ[gP]W Ferrari 1992b p. 152 : ]¦�[W]Q ƒ[gP]W vel ]Ù] [�]Q¦[gP]W Lobel : RÃf]¦�í [�]Q¦[gP]W Snell | Q¦[ H26 | cW[bd¦�Tb vel cW[Ú d¦�Tb Lobel : cW[Ú d¦�Tb def. Ferrari | Tb� H26 || 16sqq. ìThe Cephallenian maids ran about the house in a panicî Lobel || 16 VW^cgb]Rb] Éí vel fRÚ kÓ jR]�í e.g. Snell : pP_�b� fRÚ] Éí Lobel sec. Werner 1967 p. 533 || 17 d�S H26 || 18 ]� (Snell) vel ]É | c¬] H26 || 19 ¥̂ c H26 | V� ¥ H26 || 23 ] ¥ ¥� H26 | ] ¥]¥ Lobel : ] ¥é possis
157
Scholia (G26) : 6 ¥]¥j ¥[ 18 ô æWdRVVq(Q\) c[¦fW[[gQ fgt] �H_db£f[�T(Qg]) }gPV^Zbg(Q) �jRVW~�×, õQ k(Ó) Õ£cÙ fÙQ
HfW�YVRgQ� �[cÙ] k(Ó) æWdnV(gP) £ fUQ c[g]\Sg[^RQ ¯]¾[W]� 21 ]gVg[ 23 RQ(fÚ) Õ_Q\[ 6 ìI can neither verify nor rule out Wjdg[î Lobel || 18 suppl. Snell 19643 || 23 �Q(fÚ fgt) Õ_Q\[ Snell : Õ_Qq[p\ dub. Rutherford 2001 ut apparet e translatione
158
F9
(Pae. 21 = fr. 52v Sn.-M. = S2 R.) (a) . . . ¥ ¥ ¥ ¥] ¥ ¥_× ¥[ ¥ ¥ ¥ ¥]gP[RQ�[ ∞U ∞�Ó �R]^VWb�RQ ∆VP[_]c^T[Q Q��d�Q �[��]f¦cg]bQ õ® fgt¥ ¥WQRP ¥ ¥ ¥ ¥[ 5 VbcW~Q g� ¥ ¥ ¥[ ¥� ¥]XTQ fbe WkR ¥[ ]¥�[ ¥ ¥] ¥_RjR[× ¥ ¥[ �VjÏQ �ZWVT�gP j[RQ^gQ fgtfg �n�é[gQ 10 ∞U ∞Ó �R]^VWbRQ �∆Ç���[_]c^T[Q Q�_dRQ �[b]f¦¥È�g]bQ ® ¯]]WfRb SÏ[ ÚÉ�[ �¥YQRg] T]¥g ¥[ ³]fWÔ jfWn�[ 15 QR�fRb] kí R ¥[ ]¾q]Wb cgVb ¥[ ³Qp[Tc[gñ ∞U ∞Ó �R]^Vé�bRQ ∆VP[_]c^T[Q Q�_dRQ �¥��]¥�¦¥�cg]bQ 20 ® ¯fb kí �¥�k[ ¥ ¥[ ¥ ¥] ¥[ fgtf ¥Q c× ¥ ¥ ¥[ ¥� ¥[ \ ¥[ . . . �H
26 fr. 32 p. ii � || gratias ago Danielae Colomo quae papyrum nimis pulverulentam e theca extraxit et denuo mihi mundavit || de 3sq. 11sq. 19sq. v. etiam (b), (c), (d) quaecumque non notantur dispexit articulavit supplevit Lobel 1961d || desunt ab initio carminis non minus ac 18 vv. (fort. 20), v. supra pp. 132-4 || 1 num jR]��_× ¥[ ? || finem epodi (?) coronide notat H26 || 3 ìNon y sicuro che il testo del primo verso del ritornello non proseguisse dopo ∆VP_c^TQ, ad esempio con un imperativo, come �W^kWfWî DíAlessio 2004 p. 115 adn. 37 || 5 f<P H26 | f×t¥�í nol. Lobel, rec. Snell 19643 || 6 V$Wb%c sscr. � H26 | W~Q H26 | ¡f H26 sec. Lobel, sed fort. pes P est pot. qu. spiritus || 7 ì�Y]XTQ seems likely, though not the only available choiceî Lobel | cetera articulavit Snell | �¥]XT H26 | �knî[ Rutherford 2001 || 8 [× dub. Lobel : [é possis, i.e. _njR[é]¥ vel _Rjn[é]¥]¥[b || 9 articulavit Snell || 10 �n�é[gQ Snell duce Lobel || 11 Ô\ÔW H26 || 12 d�Q H26 || finem strophae paragrapho notat H26 || 13-17 articulavit
159
Snell || 13 [C H26 || 14 � ¥Y H26 || 15 WÔj H26 || Wn� H26 || 16 kíÏ H26 | �[ pot. qu. � Lobel || 18 nQ H26 | suppl. Snell || 19 Ô\ÔW H26 || 20 d�Q H26 (d¬Q Lobel) || finem antistrophae (?) paragrapho notat H26 || 21 articulavit Snell || 22 f<P H26 (b) margo ∞U ∞Ó �R] �̂VWbRQ ∆VP[_]c^T[Q Q�_dR�Q �[b]f¦c×�]bQ ® ] ]RfgkR_[ . . . (c) . . . ∞U ∞Ó �R�]^V�WbRQ ∆VP[_]c^T[Q Q�_d�RQ ��[b]f¦cg]bQ . . . (d) . . . ∞U ∞Ó �R]^VWbRQ ∆VP[_]È^T[Q Q�_dRQ �[b]f¦cg�]¥�Q� . . . (e) . . . Q�_d�RQ �¥�[b]f¦cg]bQ. (b) H7 fr. 24 || (c) H7 fr. 55 || (d) H7 fr. 83 || (e) H7 fr. 84.1 || omnia haec frr. huc traxit Lobel 1961b p. 13, (d) (e) dubitanter || incertum est utrum (b) (c) (d) ante an post (a) sint ponenda (duo �d�_QbR hoc carmen ante (a) habebat, non minus atque unum post). nihil obstat quin (c) et (d) ad (a).3sq. vel 11sq. vel 19sq. spectent; frr. unius �dP_Q^gP ea esse non apparet || utrum (e) ad F9 spectet non est certum, nam extremus v. huius carminis easdem litteras in eodem fere situ habere debebat, sed situs asterisci non est idem atque in fr. 8 (DíAlessio 1997 pp. 35sq. adn. 79). finem Pae. 2 id esse, ut coniecit Snell ap. Maehler 19754 p. 20, parum verisimile videtur, nam H7 Prosodia non Paeanas praebet, nec F10 lacunae in primo v. Pae. 3 aptabis nisi vv. divisionem in fine Pae. 2 mutabis (Ö ∞U ∞W� £ HRb�ÏQ É�Ó _qcgfW VW^cgb) (b) 3 ]Rfg kR_[ Hunt 1922b || (c) 2 �Q H7 || (d) 2 ì]gQ not ]bQ is the reading (Ö) that first suggests itselfî Lobel 1961d p. 54 : ]]¥�Q malim || finem carminis asterisco notat H7
160
F10 ¥ ¥ ¥] ¥[Rc[ . . . H7 fr. 84.3 finem carminis prioris asterisco notat H7 || kW]t¥[í êc[gVV]×Q dub. Snell ut Pae. 3.1 aptaret, brevius spatio siquidem F9(e) postremum v. F9 est
161
F11
(Pae. 12 = fr. 52m Sn.-M. = G1 R.) . . . ]_W[ ]TQbg[ P QR ¥ ¥] ¥gb]bQ ��QY[R Ùg^]]¥Rb] P QR ] ¥RVR kí �[fY_bÉ[ ] ¥iÔgQ R][ ¥ ¥]Zg] �_dYcg[b]í ³Q]pWR fgbn[- kí] Õ¥_Qq]bg] k[YÈîb� pR_Ï kí ¯�[ZgQfRb 5 ´R]X¦pWQ VbcR[g��¦dTQ pP]^[Rb _q]�TQ mR[^fW]]b _^SkRQ æ�]�pbgQ cR[Ï j[\_Q¦Q, ¯QpR [ jWVRbQWdYí �[Sb�[YQfRQ VYSg[Qfb >vQR jRpW�¦_WQgQ 10 jg[PdR~]bQ —cW[pW dPVnX�b È[ ¥ ¥]gQ ¥[ ÚQ^jí �SRQ¦d[TQ æg^gP pPSnf\[ V�Wfg fW[cQ¬] ‹k~Qg]� ¯VR_°RQ kí �WV^gP kY_R] ƒc¼[] �SVRÙ� �] dng] ∞¦QfW] k^kP_gb 15 c�~¥kW], cgVˆ� 5 �¦¥�[g]� µ¥é]R� �¥cÙ ]fg_[nfTQ {]�W^pPbn fé jRÚ Ên¥[Z]é]b]� fYVW[b]Rb kí ¿V[gVPñ jR]fWVn_�RQgQ ¥[ �c]WdpYSXRQfg kí �¥�¾i¥[bR� [ �S]�RÙ] Û] ��í §[j ¥[ ¥] ¥[] ¥[] ¥[ 20 ¥ ¥ ¥] ¥R[RQfg�R[Rj� ¥�Q ¥ ¥fg ¥[ ] ¥[ ] �[ ] ¥�d ¥�gQRQk[R[ ]î[W ¥g[ ¥[ ] ¥ ¥ ¥[ ]dR�[ (25) ]�QWP ¥W ¥[ ] ]_gVgb ¥R ¥[ ]ÈgVVnjb]¥ [ ] ¥ ¥ ¥ ¥[ ] ]c× ¥[ ] ] ¥bQ (30) ] . . . 1-24 H7 fr. 1 || (25)-(29) (pars laeva) H7 fr. 2 || (26)-(31) (pars dextra) H7 fr. 4 (huc traxit Snell 19643 duce Lobel 1961b p. 14) || quantum fr. 4 a fr. 2 distiterit non constat, nec certum est utrum vere sub fr. 2 ponendum sit quaecumque non notantur dispexit articulavit supplevit Hunt 1922b. non noto ea quae a viris doctis lecta vel coniecta a frustulis quae Lobel addidit ad fr. 1 restituendum sunt refutata || 1 �c¦VV]TQ Lobel sec. Werner 1967 p. 537 : ›] _nVR] _Ó[Q W–d[]TQ Ferrari 1996 p. 132 | 1f. ∞g£[cV¦]½gb]bQ Snell 19643, fort. recte || 2 ��QY[R vel �� QW[ vel ��QW[cñ Hunt | Ùg^]]¥Rb] Lobel || 3 Úc]RVÏ (longius spatio ut vid.) pot. qu. _]nVR vel j]RVÏ (def. Furley et Bremer 2001 p. II 109sq., sed brevius spatio) Hunt | �[fY_bÉ[ Hunt : �[fY_bÉ[b
162
Ferrari p. 131 : �[fY_bÉ[g] Rutherford 2001 (at �[fY_bÉ[g] ÊR]�iÔgQ longius spatio vid.) : �[fW_^� [ Lobel | TÔg H7 | ÊR]�iÔgQ Lobel, fort. recte | �][fW[^R vel ³][f[gQ Lobel || 4 VY]Zg] Hunt : [_W^£Vb]Zg] Lobel, brevius spatio ut vid. : Vn]Zg] possis | �_dYcg[b]í Lobel : �_dWc¦[VWb Hunt | ³Q]pWR Lobel | fÙbn H7 || 4sq. fgbn£[kí] scripsi : fgbR[�f Hunt : fgbR�[fR] Lobel, ullum spatium fuisse ante Õ¥_Qq]bg] negans (perperam ut vid. : Sí] suppl. Ferrari) : fgb¬[]£kí] Rutherford || 5 ìWhether the word preceding k[Yc[ is an adjective (IP_Qq]bg], c[P_Qq]bg]) or a substantive (—_Q\]b], S�_Q\]b] (?), _Qv]b]) is not clearî Hunt : Õ¥_Qq]bg] edd. omnes, recte | k[YÈîb disp. Lobel | b�pR H7 | kí ¯�[ZgQfRb pot. qu. kí ¯�[ZWfRb Lobel : kí ¯�[Zg_Rb vel kí ��[Z¦_WpR Rutherford : kí ¯�[ZWRb dub. Furley et Bremer : kÓ +[Y[ Hunt, quod vestigiis aeque congruit : kÓ +[Y[WQ dub. Schroeder 19236 || 6 pP]^[Rb Lobel (ut dat. intellexit Rutherford, credo autem ut nom. ipse) : vel pP]^[R Snell : pP]^[R] Hunt || 8 gQ�W H7 || 9 VYSg[P]b Hunt, corr. Schroeder : et VYSg[b]b possis || 11 È[[]gQg^[Rb Lobel, longius spatio ut vid. || 12 bjíR H7 || 14 hunc v. nota > notat H7 | TkWbQg]� sscr. JRJ H7 (1) scil. fW[cQn] 13, obl. Hunt | WVR_°$W% sscr. RQ H7 (2) | kíRW H7 | ≈c¼[] Maas ap. Schroeder : glossema vocis kY_R] esse susp. est Wilamowitz 1922 p. 519 adn. 2 : ¡c¦¥[fí Hunt, cui obstat divisio syllabarum (¡c¦£fí oportuit) || 17 b]�f H7 | fYVW[b]Rb vel fYVWRb (brevius spatio) Hunt : fWVW[f]R^ Lobel : fWVY]¥Rb Snell 19531, longius spatio | ¿V[gVPSR^ Wilamowitz : ƒV[�gQ pYVgP]Rb Snell : ≈V[ Lobel ut vid. || 17sq. ìëfestal celebrations filled allí the island, or the likeî Lobel || 19 �c]WdpYSXRfg Lobel : ] �dpYSXRQfg vel �Q]WdpYSXRQfg Hunt | [bRb: H7 || 20 Û] Hunt : =] Wilamowitz || 21 ¾R[ÏQ f¦�í ³[í �jfR^QgQfg (vel ³jfRbQgQ ap. Snell 19643) tempt. Lobel : fort. ³[RQfg �Ï[ || 22 d�SgQ ³Qk[R Snell, at quis est vir iste? (si quis est, cRV]�^¥d��gQ vestigiis aptissimum; de Homero autem agi parum verisimile videtur) : fort. SgQÏQ || 23 î[R sscr. W H7 (1) || (27) ]¥R ¥[ vel �R ¥[ Snell || (28) ]��b� Snell || (29) vel ]cר[ Snell || (30) ]� vel �
163
F12
(Pae. 13 = fr. 52n Sn.-M. = S5 Rutherford)
(a) margo •[T� fW �T[_ñ fTb fRQkW[ kTjWQ� gÈ[ ȦQfgbg [ HRVVnkR [ 5 ¥]Ycfgb f[ ]T]R� �W�È[ ]ˆQ cRQf[ ]TfW[gQ R[ —_�QTQ W[[ ]�g[bRb] �n�VVg[Q]fb [ ]�¼[b] c�g][ cg�V�î�Rfg[ ]W½R[ 10 ® QtQ ¥ ¥�Q�[ QP�+¬¥Q ]�[ pPbR^Sbkí R_[ �]fn_WQRb fWV[ ] kTQ fRQPR[ ¥ ¥ ¥ ¥] ]¥̂ Q¥ jf�cTb [ ⸥�Q- 15 k\]n_WQRb cV[g]jn_gP] _�[fTQ —È[g pY]]cbQ ¯SWb[gQ [ R∞pY[b, §Vbj[ ¥ ¥ ¥] kÓ cg[dP- [YRb ]ˆQ j[¦j[Rb ¥ ¥ ¥] ¥�� ¥Wb c[[ WÃn_cPjb WQ[ ¥ ¥ ¥ ¥ ¥]_Tb ]WV�[ 20 �Q kRbf^ fW cR[ ¥ ¥ ¥ ¥ ¥]Wb _RjR[[ ¯QpWQ _ÓQ RÈ[ ¥ ¥ ¥ ¥]RfTQ ∆VP[_cñ fRQ cgVY_gP[ ¥ ¥ ¥ ¥ ¥]Q ]R_nQfg[b [ �Q Z�[ 25 margo 1-25 H8 fr. a1 || 6-10 (pars dextra) H8 fr. a2 || 8-11 �H
7 fr. 6 � (huc traxit Snell 1938 pp. 431sq.) || 14-23 (pars dextra) H8 fr. a3 (huc traxit Snell ap. Zuntz 1935 p. 283) || quantum a fr. a1 distiterit fr. a2 parum constat; quantum fr. a3 e Zuntzii supplemento cV[g]jn_gP] (16), quod certum videtur, apparet || [Plu.] Mus. 15 1136c (Pi. fr. 64 Schroeder 19005, 19236) huic fr. pertinere censuit Snell 1940 pp. 190, at illud fr. �Q cRb¬]bQ non in Prosodiis erat, neque apparet in (a) de Niobae nuptiis ñ immo de cuiusdam ñ agi, v. DíAlessio 1997 p. 33 adn. 60 quaecumque non notantur dispexit articulavit supplevit Zuntz || 1 TÔf H8 (dispexi) || 2 f�b fnQkí §[g[fÏQ (Ö) �cY-] Zuntz || 3 WQ�g H8 | ≈È[gP Zuntz || 4 [_YkgQfR (Ö) jR^] Zuntz || 5-7 [fb_¬ (Ö) RÃfÙ] W–£]]Wcfg^ f[W VRg^� ]fWdRQ]�]Rb kÓ H[�VbgQ _nQfbQ]£ ]ˆQ cnQf[W]]b pWg~]b QW]ifW[gQ R[ e.g. Zuntz || 5sq. �]£[[]Ycfgb Snell p. 185 || 6 Ycf H8 | �W�È[ Snell || 8 �_Q H8 (dispexi) | —_QTQ �[[Rfg~] �Q WÃR]�g[^Rb] Zuntz || 9 c[g][ Snell : cg][ (Hg][WbknTQb dub.) Zuntz || 10 cgVP\[nfg[P _RQfg]�QR]
164
ãQ]W½R Zuntz || post hunc v. finem strophae paragrapho notat H8, nulla nota H7 || 11 tres non duo litteras evanuisse censuit Snell : [pW]ÏQ r[YfbQ vel [�_]ÏQ p[ vel [�cí] �Qp[ Zuntz | fin. �W^]g_WQ suppl. Zuntz || 12 QP$Q% sscr. � H8 (2) (DíAlessio 1999 p. 17 : (1) sec. Zuntz) | + ¥êQ H8 (disp. Snell) | QP_d¬Q Ferrari 1991b p. 387 : Q�_dRQ Zuntz | ]�[QWPQ¦Q fW H\VYR Zuntz || 13 pP$g%bRbSbkí sscr. é (supra p) H8 (2) (é disp. DíAlessio) : [�W�[b]£ptbí R∞S^kí Zuntz : pPbRbS^kí Lobel ap. Snell 19531 : vel WÃbRbS^kí DíAlessio p. 18 || 13-15 Ùg~]Rb kÓ Zg[ÙQ]£ �]fn_WQRb fYV[W]]RQ Zg[]£k�Q fRQPn[ZWÔ] (ìscriptum fuerit -ZWWbî) ]ˆQ jf�cTb[� ]fWdnQgb] kí �Q-] e.g. Zuntz || 14 fort. Ô]f H8 || 14sq. fYV[W]RQ cg]£k�Q Snell 1940 : fWV[WfÏQ Ferrari, fort. recte || 15 P�$W% H8 || 16 ]W_ sscr. � H8 (2) || 17 —c[g jRÚ] Snell, prob. recte : —[_QgQ] Zuntz | ]dbQ sscr. c H8 (2) (i.e. pY]]cbQ, Ferrari pp. 388sq.) : ]dbQ Zuntz : utrum pY]cbQ pro ]dbQ an pro [jRÚ] ]dbQ substituere voluerit corrector incertum est, ut monuit DíAlessio p. 20 adn. 32 | fin. �Q suppl. Zuntz : dpYS_RfR Snell : ì—_QgQ (vel S¬[PQ) �] (vel �gÏQ c[¦])î pot. qu. �n]bQ c[¦] DíAlessio, qui �gbkÏQ jRÚ �n]bQ c[¦] nol. || 18 RbpW[b sscr. � vel É H8 (2) (disp. DíAlessio p. 19) : R∞pY[R pot. qu. R∞pY[bR vel R∞pY[b(R) (non R∞pY[b kí) significare censuit DíAlessio : R∞pY[b Zuntz, def. Ferrari : R∞pY[bí Snell | §V^j[W]]b] vel �Vbj[�QR] Zuntz (ìnon ita �Vbj[�Qb] neque §Vbj[fRÚ], -[fR~]] sim.î) : ìqualche forma obliqua di §Vbjf¦], magari §Vbj[fg~](b)î DíAlessio : �Vbj[iQbRb] nol. Snell p. 188 propter spatium || 18sq. cg[dP£[YRb�]� ]ˆQ j[¦j[Rb] rY]�bQ �¥WÚ Zuntz : cg[£dP[YRb�]b� nol. DíAlessio p. 20 adn. 43 || 19 j[¦j[Rb Snell | ]� ]È ]]¥ ]�, post �� � vel É | cg]]¥Ú¥� DíAlessio || 20 WPRQcWjb sscr. P H8 (2) : WÃnQcPjb Zuntz, corr. Ferrari : �PnQcPjí �WQ[ vel WPRQ c�jí �WQ[ nol. Snell p. 189 : ìAn Baccheum WÃnQ intellegendum?î Turyn 1948 | �Q[pí �Q Sn]_Tb ]YVR[] �QY]ZWQ ≠[R dub. Zuntz | ìPro ultima � possis Vî Zuntz || 21 cn[QfW] Zuntz, ìspatium excedit cn[QfW] �j]W~; non sufficit cn[QfW] �]WÚ neque cR[cfR^Q]Wb; possis cn[QfTQ �]WÚî | _njR[[W] (Ö) WÃTZgtQfg Zuntz : _Rjn[[TQ Ferrari p. 388 || 22 �[[XR_YQ]R f�Q ∆VP[_c^TQ ¯[b] Zuntz : ³[[XRfg Bowra : RÈ[ pot. qu. R�[, non R�[ || 22sq. ¯[b] d�fWP]WQ g∞jf[gfn]£fRQ cgVY_gP [ ¥ ¥ 5^�R]Q e.g. Zuntz || 25 WQ pot. qu. W[b]Q (Snell)
165
(b) margo ]] �Zp[�Q ¡_bVq]WbW[ ¥] ¥ ]î[RfgP cnf[R] §jn] ] ¥Y[R] [ ] Z]é~[R] �[Rbn] d]Y[fRfg] �Qp[icTQ d[ 5 ]P jí R∞]Z[ÙQ cnpgb] ]]b] �QRQ�fgb] W”[WQ ]bRQ R∂]RQ� ] ]�p�Èg[q]RQ[ ] 10 ]]¥[ ¥]d[ ¥ ¥]Q c[g½[^Qg� [ ]]\ jRVW~Q ]Wb fb] ³fW[pWQ [ ]½R][ ] ] ¥� kÓ p�_�b 15 ] ¥WQWb jRfW� ¥[ ] ] ]¼Q ¡c¦¥fW ]�gQ[f]Y] QbQ �j 20 ]¾RQ _YQWbQ ] ]]¥�Q ]TQ j�[bgb ] ¥ ¥ 25 margo H8 fr. b || de ordine (a) (b) (c) parum constat. nihil obstat quin (b) ad paginam primam post (a) spectet, ut coniecit Snell 1940 pp. 189sq., (b).1-5 respondere (a).6-10 (i.e. hoc finem strophae, illud antistrophae esse) ratus, sed v. etiam DíAlessio 1997 p. 32. (b) non spectare ad paginam primam ante (a) ostendit metrum. (b) ante (c) et (c) ante (a) posuit Zuntz 1935 pp. 290sq. indiciis tenuissimis fretus. non est certum (a) (b) (c) frr. unius carminis esse, quamquam ex iisdem metris quae dactyloepitritica vocantur sunt composita || 3 huc pertinere fr. 294 Sn.-M. �VY[R] P�¦Q censuit Turyn 1948, prob. DíAlessio 1999 pp. 22-4. �]�Y[R] vestigiis optime congruit, sed Tityi fabulae vv. 1sq. parum convenire videntur; res in dubio est quaecumque non notantur dispexit articulavit supplevit Zuntz || 1 WjZp[TQ g_WbV\]WbW H8 | ]½, ]¾, ]R� | ]f[Rf�b SÏ[ ≈]fb]] �Zp[�Q ¡_bVq]Wb �¥[ ¥]j e.g. Zuntz || 2 WÃ]î[nfgP Zuntz : cgVP]î[nfgP Lloyd-Jones ap. Rutherford 2001 : �c]î[nfgP possis || 3 Y[R] H8 || 4 Z]é~[R] Bowra 1935 | [Ïb H8 | gÃj ðQ d�Sgb _]×^[R] �[R^R] e.g. Zuntz contra accentum | Ú[- DíAlessio, duce Herodiano || 5 jí WÒcW[ W∂ d]Y[fRfg] �Qp[icTQ e.g. Zuntz | ìLiteram d in initio vocis dY[fRfg] omissam in margine supplevisse videtur scribaî Zuntz, quod nimis improbabile mihi vid. || 6 init. �Q kÓ fW¬b cnf[Rb _YQTQ ]�Sí e.g. suppl. Zuntz | PjíRb H8 : g]Ãj Zuntz : apostrophus fort. est falsa (ut ait Schubart ap. Zuntz) vel divisionem verborum indicat, at cur non genuinum jí (i.e. jW) esset ? : g]– jí possis | Z[g fort. corr. e Zbg H8 | cRpgb] ¥ ¥, fort. cRpg~]¥� aliquo vestigio supra � addito sec. DíAlessio : mihi quidem ea quae gb] sequuntur vid. esse quod Anglice casual smudge vocatur || 7 �cWÚ pWgt fgb k�QR]]b] Zuntz | �QRQ�fgb]�bQ� dub. Snell p. 189 adn. 1 || 8 RQ� H8 (dispexi) | ¿V�]^RQ e.g. Rutherford : �VjnQ fW jRÚ fYVW]bRQ Zuntz || 10 [q] H8 | g]–pí [Û]
166
cg[q]RQ£[fW]] dub. Zuntz : W]Ãp[P]cg[q]RQ[�] Bowra falso accentu || 11 ]]¥[ pot. qu. ]�[ (Zuntz) vel ]é[ : ] ]¥[g]d[^R]Q spatium expleret : ¯Q]�[R] d[[U]Q Zuntz | [$W%bQ H8 | vel -� Zuntz || 13 ] WÒ Zuntz || 15 ]�� ]]¥� ]�� : §j¦Q]�� Zuntz, fort. recte || 16 �[ vel �[ | �jW~]�WQ W∞ jRfY�� Zuntz : WÃ]�WQW~ jRfY��[Q Qg�b (sc. Q¦Tb) e.g. Rutherford || 20 V]�gQfW] Zuntz : e.g. cg[W]�gQfW] Snell p. 186 || 21 QW_WbQ sscr. _ (supra Q) Q (supra _) H8 (disp. DíAlessio p. 24) | cWQpYTQ, ›]fí �]dRVv f�]ZRQ QY_WbQ e.g. Zuntz || 23 ]]¥ (Snell) ]� ]� | ¯kg]]RQ Zuntz ap. Bowra || 24 pWgÚ (Ö) ÚcnQf]TQ j�[bgb e.g. Zuntz | fort. bgb� H8 || 25 ]é� Zuntz (c) . . . ]�� fgbR fb] �_[ ]]Rb ]fRpW~¥]Rb� [ ]� ] ¥��] ¿VgVPSR~] [ ] ] ¥ ¯_RpgQ kí ≈¥fb Ùgb[¬� [ 5 ]�VRb �cVR[ ¥ ¥] ¥[ ¥]cRpR[ ]]¥ ãk[Rb p ¥[ ¥ ¥ ¥ ¥]�gQ[ ]é]ZgQfgÉ[ ¥ ¥ ¥] ¥ ] ¥Wb jRÚ QtQ fY�R] k�[ ]SgQgPfgPQ �_cW�[ 10 ]�RfTQ� gÃkí �ZWVTbg[ ] ]bQ¬Q _Q¬_í ¯f� �gP[ ]¦QfW]]b cbpY]p ¥[ ¥ ¥ ¥ ¥]Rb [ ]]¥kí ��d[ 15 ]W]cgf[ ]�fW�[ ]Tb QR ¥[ ] ¥ ¥[ . . . H8 fr. c || quo in ordine frr. (a) (b) (c) inter se ponenda sint incertum est. Zuntz 1935 p. 290 (c) paulo spatio ante (a) posuit, indiciis tenuissimis de carminis argumento fretus. quaecumque non notantur dispexit articulavit supplevit Zuntz || 1 ³fR (Ö) ¯]ZW]Q fg^R fb] �_[cWk¦_gZpg] e.g. Zuntz || 2 ]Rb� H8 (dispexi) || 4 ]� ]� ]+ | g∞jf[gfn]���] Zuntz || 5 [¬� H8 (disp. DíAlessio 1999 p. 24) : _g~[RQ Zuntz || 6 V�b$Rb% sscr. W (2) W (3) H8 (DíAlessio p. 17 : ]W Zuntz) | cVW sscr. R H8 | gÃj ¯]fb dPSW~Q (Ö) ³]VVRb] �cVn[SZp\]RQ] cnpR[b] e.g. Zuntz | ]cí �pR[QRfñ dub. Snell 1938 p. 187 || 7 gÃ]pí ãk[Rb p×[ ¥ ¥ ¥]�gQ Zuntz || 8 Y]]ZgQ fg�[ ¥ ¥ ¥]� Zuntz || 9 f�b] ÉW~ Zuntz, vestigiis parum aptum | ìapparently fNW[R]î Rutherford 2001 | }�[Ù] Zuntz || 10 init. jRÚ CW_YVR]] suppl. Zuntz | ] S¦QgP fg“Q Zuntz (fgt íQ Bowra 1935) : ]SgQ g–fí g“Q (i.e. ‚Q) Lobel ap. Snell 19643 : res in incerto manet | �_cYV[Tb Zuntz : �_cYV[gb] Bowra | fin. kPQRfgt dub. suppl. Zuntz || 11 Õk]nfTQ Zuntz, vestigiis minime aptum | TQ� H8 | kíR H8 || 13 Q¬Q H8 | cg]bQ¬Q Zuntz, fort. recte | _íW H8 | fgt]fg Zuntz || 14 ]gQfW]$kR_+%[ sscr. ]bcbpW]� ¥[ ¥ ¥ ¥ ¥ ¥]Rb (Zuntz, prob. DíAlessio pp. 24sq. : ]gQfW]]R_+[ sscr. gcbpW]¥ ¥[ Snell) H8 | jRÚ —]fW[gQ] ƒQfW]]b cbpY]pRb [ Zuntz : SR[�]gQfW] ¿cÚ pW]¥È[W]^]Rb Erbse ap. Snell 19531 : cbpY]�[Rb pot. qu. cbpY]�[W DíAlessio || 15 -gQfW]] Zuntz | kíR H8 || 16 k]W]cgf[R Zuntz || 18 ×[ Zuntz : é[ Snell : littera non tam rotunda mihi vid. fuisse | QRÙ[Q }bTQ�]gP (Ö) fWtZWQ dub. Zuntz || 19 non vidit Zuntz
167
F13
(fr. 94 Sn.-M.)
_W_QnbRfí �gbk¬]
[I] = | (h) Il. cit. ad 23.361a1 p. V 426 Erbse (_W_QYTbfg�) c[gcR[gXPf¦QT] (c[gcR[gX�fgQgQ Ge) jRÚ _WfÏ fgt b O. PWQgd�Q kÓ _Y_QTbfg ³QWP fgt W O (fgt b O Ge), æ[Rf~Qg] kÓ _W_Qg~fg, fÙ kÓ �QnVgSgQ _W_Qvbfg (fÙ ® _W_Qvbfg M1P11U4 : fÙ kÓ �QRVgSg�_WQgQ õ f�b Ge) kbÏ fgt \ O c[gjWb_YQgP fgt b O� (jRÚ add. M1P11U4) �[b]fgdnQ\] �Q (om. Ge) HVg�fTb� µQR Üfg]gtfgQ (fÙQ Ge) �_nfbgQ dg[�Q _W_Qvbf¦ _gP. H^QkR[g] kÓ }T[bjifW[gQ kbÏ fv] Rb¿ kbdp¦SSgP �Q c[g]Tbk^gb] (c[g]Tbk^Rb] Ge)� _W_QR^Rfí �gbk¬] unde [II] = EM p. 1652 Gaisford c[gcR[gX�QWfRb jRÚ ]ˆQ f�b b O S[ndWfRb� �cÙ fgt _W_QYgbfg, RÃXq]Wb fgt g O W∞] TO , _W_QYTbfg. PWQgd�Q kÓ _W_Q�bfg ³QWP fgt W O, æ[nf\] (æ[Rf( ) O : æ[Rf~Qg] Q) kÓ _W_Qg~fg, fÙ kÓ �QnVgSgQ _W_Qvbfg kbÏ fgt \ O c[g]jWb_YQgP fgt b O, H^QkR[g] kÓ }T[bjifW[gQ kbÏ fv] Rb¿ _W_QR^Rfg. _W_QnbRfg Boeckh 1811-21 p. II/1 683, def. Wackernagel 1916 p. 90 : _W_QR^Rf(g) [I] [II] : _W_QR~fg Sylburg 1594 [not.] p. 34 : _W_Q¬bfg Schroeder 19005 | �gbk¬] Ge : �gbkv] M1P11U4
F14
(fr. 91 Sn.-M.) H^QkR[g] kÓ �Q c[g]gk^gb] cnQfR] fgˆ] pWgˆ] �cg^\]WQ, ¡c¦fW ÕcÙ ÁPd�Qg] �kbijgQfg, gÃj �Qp[icgb] ¡_gbTpYQfR] �VVÏ fg~] ³VVgb] �ibgb]. [I] = Porph. Abst. 3.16.5 (T9) || v. DíAlessio 1997 p. 40 ¡c¦fW [I] : ≈fW dub. Nauck 1860 p. XXXII, fort. recte | ³VVgb] [I] : �V¦Sgb] Wesseling 1745 p. 97
168
F15
(Pae. 12(a) = fr. 52m(a) Sn.-M. = G2 R.) . . . ] ¥]¥gd[ ] ¥ ¥ ¥�bRQ È[ ]¬SW c[gdR[ Ê]�fg�kRb�[ ]Q� ]WXbR[ ¥ ¥ ¥] ¥[ ] 5 ]É�È[ ¥ ¥ ¥] ¥ ¥T ¥ ¥[ ]_g]T c×VP]Ycf[ ]WbQgfgbfWjWÈR�[ ] È[¦]gkgQ � ¥ ¥ ¥ ¥ ¥ ¥[ ]�W Zg[ÙQ ÕcW[fR�[ 10 ]ZR[bQ V[ ¥] ¥ ¥ ¥�Wj[ c]Y�cWb� _ ¥�gb jR[ ] ¥ ¥QfR] gÒSW�Q[ . . . 1-13 H7 fr. 11 || 4sq. (pars dextra) H7 fr. 74 (huc dub. traxit Lobel 1961b) || H7 fr. 7 ex eadem parte papyri exortum esse censuit Lobel || cave ne nimis credas amplitudini lacunarum quam Snell 19643 computavit, ab initio nihil deesse in vv. 2sq. ratus, unam litteram in vv. 4-9; v. DíAlessio 1997 p. 18 adn. 27 quaecumque non notantur dispexit articulavit supplevit Hunt 1922b || 2 ]é pot. qu. ]]¥ Lobel | ×Ã¥��Q^RQ vel kP]]¥��Q^RQ dub. Snell 19643 | articulavit Snell || 3 ¬SW H7 | articulavit Snell : ìbut for the accent (Ö) Zg[]RSY would be attractiveî Rutherford 2001 || 4 suppl. Snell | gÔk H7 | accentum addidi : Ê]�fg�kRb� [ vel Ê]�fg�kRb �[ pot. qu. Ê]�fg�k�b �[ vel Ê]�fg�k� b�[ || 5 k]WXbR[ Hunt || 6 ]É�È[ Lobel : ]É�� ¥[ Hunt | ]� Snell | � ¥[, È ¥[, ¥×[ Lobel | ]kí �c[¦V]VTQ dub. Snell, brevius spatio || 7 articulavit Snell : ¿]_¦]T vel �]_Ù] ‚ dub. Snell : Ú[]_¦]T possis | ]Yc H7 || 8 j]W~Qg dub. Snell | fgb fYjW Rutherford || post hunc v. finem strophae vel antistrophae vel epodi paragrapho notat H7 sec. Lobel, obl. DíAlessio || 9 È[¦]gkgQ pot. qu. È[Ù] ¡k¦Q DíAlessio : �]È[¦]gkgQ vel WÃ]È[¦]gkgQ dub. Rutherford || 10 articulavit Snell || 11 articulavit Snell | Zn[bQ Hunt | ]é vel ]]¥ | ]Wj[ Hunt || 12 Wb�_ H7 | _é�gb vel _×�gb Lobel, quae vestigiis parum congruunt : _��gb pot. vid. || 13 ]�é vel ]�× | articulavit Snell | gÒSW�Q [ vel gÒSW� Q[
169
F16
(Pae. 7(f).1f. = fr. 52g(f).1f. Sn.-M. = Z9 R.) . . . ∆[] ¥f][bRbQR[ ]WQ HfibT[b H7 fr. 47.1sq. || fort. ex eodem carmine sunt frr. 51a, b, d Sn.-M. (e Str. 9.22.33) et H7 fr. 49, v. DíAlessio 1997 pp. 31sq. 1 ∆[]gf][bRbQR Hunt 1922b : ∆[]bf][bRbQR dub. Snell 19643 : ∆[]g]f[^RbQR Rutherford 2001, at nihil obstat quin genitivum sit | cl. positione asterisci in H7 fr. 84 fort. ∆[£]bf][^RbQR expectes || 2 �Q Hunt | HfT�T[b Wilamowitz 1922 p. 520 || finem carminis asterisco notat H7
F17
(Pae. 7(f).3 = fr. 52g(f).3 Sn.-M. = Z10 R.) ]WQ ]gd[ . . . H7 fr. 47.4 finem carminis prioris asterisco notat H7 || 1 articulavit Hunt 1922b
170
F18
(Z7 R.) . . . ]� ¥ ¥ ¥[ ¥] ¥[ �]�dbcgVW~¥[ �c]Èg]¦R p�[Snf\[ . . . H7 fr. 51 2 suppl. Hunt : �]�dÚ c¦VW� [ minus probabile vid. || 3 suppl. Lobel ap. Hunt cl. O. 3.26 | ]¦R H7
171
COMMENTARY
173
F1
Introduction
F1 owes its survival to Aristophanes, who parodied it in the second parabasis of his
Knights (1265-7):
f^ jnVVbgQ �[Zg_YQgb]bQ 1265 ¢ jRfRcRPg_YQgb]bQ ¢ pg¬Q µccTQ �VRfv[R] �W^kWbQ _\kÓQ W∞] ÊP]^]f[RfgQ Ö
A commentator noted the fact, and the quotation survives in the scholia, on which the
Suda depends.1 The scholia cite author and work, not the book-number; the Suda even
omits Pindarís name. The fragment was the beginning of the poem, as both witnesses
remark. The metre is clearly dactylo-epitritic, although variant readings blur some details
of the opening line.2 No colometric division is preserved in the manuscript witnesses; it is
tentatively reconstructed here on the basis of the layout that its nearest Pindaric relative,
Paean 5, has in H4 (coll. xx-xxii), and the transmitted colometry of Aristophanesí parody.3
1 On Aristophanesí parody of the fr. see Wilamowitz 1919, Fraenkel 1962: 203-7. 2 See n. 3 Pae. 5.37-9 = 43-5: ˙ � ˙ ˙ � ˙ ˙ � � £ � ˙ ˙ � ˙ ˙ � X £ � ˙ � � � ˙ ˙ � ˙ ˙ � X . Ar. Eq. 1265-7: �̊ � ˙ ˙ � ˙ ˙ � ˙ £ � ˙ ˙ � ˙ ˙ � � ̊£ � ˙ � � � ˙ ˙ � ˙ ˙ � � , an articulation confirmed by Heliodorus for the first two cola (| Ar. Eq. 1264a Jones; the metrical analysis of the third colon is lost): see Bravi 2002 with Prato 1962: 58f., Parker 1997: 180-2. Both passages lack an epitritic element corresponding to our l. 3, but there is little ground for supposing that the rest would have been laid out significantly differently.
174
The fragment consists of a simple rhetorical question: What is a finer thing, for those
who begin and those who end, than to sing of Leto and ìthe driver of swift maresî,
Artemis? Two inceptive commonplaces are combined, that of ëbeginning to singí and the
aporetic question of what to sing. The result takes the shape of an ultra-compressed
summary priamel,4 which implicitly denies the existence of the foil at the same time as it
envisages it. Beginning to sing from Leto and Artemis is precisely what the chorus does by
posing the question: performative self-consciousness is at the same time affirmed and
deflected. By switching the function of the accusative from object to subject, Aristophanes
captures and redoubles, as it were, this performative self-consciousness of the fragment:
not only do the chorus refer to their performance, as the persona loquens does in F1, but
also to themselves as the performers, which in Pindarís extant lines is only implied.
What may have followed is unknown. There is a reasonable chance that Apollo
featured immediately after his mother and sister,5 who are not usually found as a self-
standing pair: it is with all three that the Deliades begin their performance in the Homeric
Hymn to Apollo (ll. 158f.), and it is Apollo who is hymned at the beginning and at the end
in the first elegy of the Theognidea (1-4) and in the shorter Homeric Hymn to Apollo
(21.3f.). The parody of F1 in the Knights is followed by an address and prayer to Apollo
(1270-3), which may constitute a hint in the same direction if the Prosodion echoed
through Aristophanesí ode beyond the close reference to its opening lines.6
4 On Pindarís use of the priamel see Dornseiff 1921: 97-102 (who inaugurated the use of the term in criticism of Classical literature), Bundy 19862: 4-19, and Race 1982b: 73-86; on the history of the term see Race 1982b: 1-7. On the concept of ësummary priamelí see Bundy 19862: 7f., Race 1982b: 10-17. 5 So Wilamowitz 1919: 54. 6 Notably, the passage includes the Pindaric phrase HPp�Qb k^Rb (1273 = P. 7.11: Wilamowitz 1919: 55f.), although direct allusion cannot safely be assumed in this case. Wilamowitzís further suggestion that allusion to Pindar went on in the antistrophe (ibid. 56), thus suggesting a Pythian context for the Prosodion with
175
F1 seems to have been widely known in Athens in the second half of the fifth
century. Aristophanesí parody would have little point if the lines were not presumed to be
known at least to a portion of his audience;7 it is worth noting how he reproduces the
fragmentís first two lines literally, at the cost of creating a slight syntactical inconsistency
with his variation on the fourth (two datives and one accusative with the infinitive).
Another clear reference to these lines occurs in a fragment of the fifth-century elegist
Dionysius Chalcus (fr. 6 W.2), where the finest thing at the beginning and at the end is fÙ
cgpWbQ¦fRfgQ.8 The picture of the fragmentís Nachleben is concluded with an oblique but
perspicuous allusion in the opening of Callimachusí Hymn to Zeus (ll. 1-3).9
It has been suggested that the ode was part of the ëschool curriculumí in Athens.10
This is not the only possible explanation for the popularity of the fragment: continuing
sympotic re-performance is a valid, if just as speculative, alternative.11 Some closer
reference to the Theoxenia, is rather less substantiated: Leto and Artemis are sometimes found at Delphi with their son and brother (see n. 14 below), but this is not prima facie what the text suggests. 7 See Franco 1988, Swift 2010: 39-42, and the important qualifications of Wright 2012: 146-50. 8 Quoted by Ath. 15.702c, through whom it reached Eustathius (ad Il. 18.570 p. IV 260 van der Valk; on his dependence on Athenaeus, probably in the epitomized form, see id. I 79-85). To suppose an echo of µccTQ �VnfWb[R in A. Pers. 32 µccTQ fí �VRfq[ would be rather more far-fetched, although it would usefully date F1 to before 472. 9 Ziegler 1913: 352f., Smiley 1914: 48-50, Hunter and Fuhrer 2001: 169-71. The adaptation of the conceit ëWhat is better than to singÖ?í, which on its own is not conclusive, is immediately flagged as a Pindaric imitation by the re-use of �VRfv[R. In Smileyís view, the epithet is given a different meaning, ërouterí (49), but KNhnken 1984: 438-43 argues that the meaning is still the Pindaric (and Homeric) ëdriverí. 10 Irigoin 1952: 16, Fraenkel 1962: 206 n. 1. On various kinds of lyric as the subject of musical instruction see e.g. Pl. Prt. 326a-b, Leg. 2.654a-b, Ar. Nub. 966-8, with Marrou 19656: 80-2. There is little reason to envisage a fixed syllabus such as present-day schools have, but the passage in the Clouds does entail that certain songs could be portrayed and acknowledged as enjoying a privileged role in formal education. 11 Performance of older lyric at the symposium ñ a habit reportedly on the wane by the end of the fifth century ñ is referred to by Ar. Nub. 1355f. (Simonides), Eup. fr. 395 K.-A., Hsch. f 1343 Hansen-Cunningham (Stesichorus), probably also Ar. fr. 235 (Alcaeus, Anacreon), Eup. frr. 148.1f. (Stesichorus, Alcman, Simonides), 398 K.-A. (Pindar): see Reitzenstein 1893: 30-2, Herington 1985: 195-8, Nagy 1990: 107-15, Fabbro 1992, Currie 2004: 56-61, Swift 2010: 44-55, and Cingano forthcoming. PMG 887 = Carm. Conv. 4 Fabbro may represent a sympotic adaptation of (an excerpt from) a Pindaric original, fr. 95 Sn.-M.: see Reitzenstein 1893: 16, Lehnus 1979: 94f., and more sceptically Fabbro pp. 98f. On comic echoes of
176
connexion with Athens may also have existed. Beside F1, Aristophanesí comedies refer to
one Athenian and one non-Athenian ode of Pindar (the Dithyramb fr. 76 Sn.-M. and the
Hyporcheme for Hieron fr. 105 Sn.-M.) as well as to Simonidesí Aeginetan ode for Krios
and an unattributed fragment from his epinicians for chariot-victors (PMG 509, 512 = frr.
16, 1 Poltera):12 this bars a straightforward equation between quotation in an Athenian
comic context and an Athenian origin of the text quoted. The fact that the Athenian fr. 76
Sn.-M. is referred to more often than any of the others is probably due to its particular
popularity rather than to any broader tendency to remember locally-commissioned odes.
F1 itself says nothing about a possible context for its performance: the notion of
beginning with Leto and Artemis, especially if Apollo followed suit, has distinct Delian
sympotic lyric see Kugelmeier 1996: 44-82. A crater by Euphronios shows a named symposiast singing an address to Apollo and probably either Artemis or Leto (SLG S317: Vermeule 1965 with Herington 1985: 197), and it seems to have been customary to start off a drinking session with a paean: A. Ag. 245-7, Xen. Smp. 2.1, Plut. Mor. 615b with Xenoph. fr. 1.13f. W.2, Pl. Smp. 176a, and Timae. BNJ 566 F 32 with Cingano forthcoming (cp. Alcm. PMGF 98, Thgn. 757-64 with Ford 2002: 61f.), see Reitzenstein 1893: 40-3, Fabbro 1986, KÅppel 1992: 51-4, 317-22, Rutherford 2001: 50-2. Two elegiac pieces for Apollo and one for Artemis open the Theognidea, and their placement may be related to this custom (similarly Meyer 1933: 45, van Groningen 1966: 10); the same may have been the case for Aristotleís ¯c\ and �VWSW~R, which begun with an address to Apollo and (probably) Artemis respectively (D.L. 5.27, frr. 671 Rose, 672 Rose = W.; pace Ford 2011: 108f., the hexameters cannot also have begun with an address to Artemis, since the first two epithets in the opening verse are incontrovertibly masculine). F1 would of course be particularly apt for use as an opener. The school-room and the drinking-hall do not exclude each other, nor do they exhaust the range of possibilities in which such a song as this could have been performed. These include informal contexts ñ of which we know next to nothing ñ alongside the more formalised ones: regrettably, Socrates does not elaborate on Tynnichusí paean, √Q cnQfW] ³bkgP]b (Pl. Ion 534d), but the phrase must refer to something broader than performance in the context of official cult. It goes without saying that the popularity of F1 does not necessarily entail that of the whole ode, whether or not these lines truly had a proverbial flavour, as Eustathius claims: Nagy 1990: 108 notes that all three Aristophanic adaptations of Pindar adapt Pindaric openings (cp. also PMG 887 cited above), and the same is true of Simonides (Rawles 2013: 199). 12 Fr. 76 Sn.-M.: Eq. 1323, 1329, Ach. 637-40 (Nub. 299 is doubtful, see Rawles 2013: 177 n. 6; | (V) 299b Holwerda cites the Pindaric passage as a parallel more than a model). Fr. 105 Sn.-M.: Av. 926-45. Simon. PMG 507: Nub. 1356 (Rawles 2013: 183-90). PMG 512: Eq. 406 (Bravi 1999, Rawles 2013: 190-4). One may also add that the most obvious allusion to Pindar in extant Attic tragedy ñ S. Ant. 100, see Bagordo 2003: 201f. with Garner 1990: 80 ñ brings in not an Athenian but, appropriately, a Theban poem. On Pindar in Aristophanes see Irigoin 1952: 13-16; in tragedy, Irigoin 1952: 12f. and more extensively Bagordo 2003: 166-218. On lyric in comedy see also Kugelmeier 1996, Zimmermann 2000, TrCdC 2003, Rawles 2013.
177
overtones,13 thus potentially suggesting performance either in a theoria to Delos or at a
local Delion (or both), but an unrelated cult of Apollo cannot be ruled out.14 If the Delian
hypothesis holds true, this in turn indicates a commission from the catchment area of the
Delian sanctuary: this might suggest Athens herself, whose involvement with it is well
known throughout the fifth century (Paean 5 may be one such example, and 7c(c) = D9 R.
another),15 but nothing can be proved on the existing evidence. Conversely, Boeckhís
hypothesis that F1 comes from the Aeginetan song to Aphaia cited by Pausanias (2.30.3 =
fr. 89b Sn.-M.) rests on the highly dubious premise of the identification of Aphaia and
Artemis, for which no early evidence exists, and is probably to be rejected.16
13 So also Rawles 2013: 178. 14 Artemisís presence in the Delphic cult seems to have been slight (Farnell 1896-1909: II 466f., Mastronarde 1994: 222) and Letoís even more so. Nonetheless, the Delian myth of the birth of Apollo and Artemis is narrated by Limenius in his Athenian paean for Delphi CID III 2, which concludes with a prayer to both and to Leto (see pp. 60f. above): a Delphic context for F1 cannot be definitively excluded. 15 On Pae. 5 as an Athenian commission see Grenfell and Hunt 1908: 20, Bona 1988: 81, Rutherford 2001: 296-8, 2004: 83-5, Kowalzig 2007: 83-6; a Euboean commission according to Wilamowitz 1922: 327f. The text is definitely flattering towards the Athenians, but this does not necessarily imply that it was sung by a group of Athenians intent on flattering their own city: for Athens ìcalling the tuneî over her alliesí choral song on Delos see Fearn 2011: 210-17 on B. 17 (cit. from 211), and if later on at the Panathenaea part of the deal was playing out the role of an Athenian colony (see p. 44 n. 21 above), it cannot be excluded that this may have been the case also a few decades earlier on Delos. If so, Pae. 5 may conceivably have been commissioned and performed by one such ally, perhaps Euboea or one of the Cyclades (cp. 38), although the parallel established between the speaker and the earlier Athenians does suggest an Athenian chorus. Pae. 7b may be another example of an Athenian commission for performance at Delos, if the first word of its title is correctly restored as H[RVV]î�^ [f]Rb] (Snell 19643: 34). On Athensí choral involvement with Delos see also Parker 1996: 150f., Wilson 2000: 44-6, Rutherford 2004: 82-9, Kowalzig 2007: 110-18. 16 Boeckh 1811-21: II/2 588 with Wilamowitz 1919: 54f., 1922: 274-6, DíAlessio 1997: 39.
178
Commentary
1 TU VWXXYZ[ : the question on what to sing is a common opening device, sometimes
introducing a list of possible answers structured as a priamel (I. 7.1-15, fr. 29 Sn.-M.),
sometimes without one (O. 2.1f.), in a form that goes back to epic precedents such as H.Ap.
19 = 207: see Des Places 1947: 103, Race 1982a: 6-8, 1990: 104f., and more examples in
Pavese 1997: 345, 416. ì[A] more distant relativeî of our fragment (Race 1990: 105 n. 54)
is P. 7.5-8:
�cWÚ f^QR cnf[RQ, f^QR g∂jgQ QR^TQ ¿QP_nXWRb �cbdRQY]fW[gQ �VVnkb cPpY]pRb;
The rhetorical question is similar to ours, although the answer has already been given in
the opening lines (1-4) rather than remaining implicit, as one might presume happened in
the incipitary F1. A question of this kind is a functionally equivalent alternative to the
plain voluntative form of hymnic opening (ìI shall singÖî); other such variations,
likewise projecting the impulse to sing somewhat outside the poet, can refer to his pP_¦]
(Alc. fr. 308.1f. Voigt, cp. Od. 8.45) or d[qQ ([Terpand.] PMG 697 = fr. 2 Gostoli, adesp.
PMG 955, cp. Od. 22.347f.) (see Metcalf 2012: 132-4).
\]^Z_`[ZYaY[ ¢ VcTcdceZ_`[ZYaY[ : Schroeder 19005: 419 regarded the
homoeoteleuton as unacceptable (homoeoteleuta satis molesta Ö putidula illa clausularum
monotonia) but obligingly listed the several parallels for its acceptability (O. 6.102f.,
179
13.40, N. 9.51); he further objected to the metrical identity of the transmitted text of 1 with
Archil. fr. 168.1 W., only to be contradicted eight years later by Pae. 5.1, which is also
identical. I follow the majority of the mss. and Aristophanesí parody: if anything, the
rhyming effect gracefully highlights the polarity of the expression, cp. Thgn. 1f. g–cgfW
]W~g £ Vq]g_Rb �[Z¦_WQg] gÃkí �cgcRP¦_WQg] (facilitated by the metre in both cases).
Beginning from a god is a topical conceit of hexameter hymns, foregrounding both the
deity whom the song addresses and the song itself (KeyÜner 1932: 9-12, West 1966: 150f.,
Metcalf 2012: 114 n. 267). Formulaic ³[Zg_í �W^kWbQ plus accusative is found at H.Cer. 1,
H.Hom. 9.8, 11.1, 13.1 (cp. 3 ³[ZW kí �gbkv]), 16.1, 22.1, 26.1, 28.1, cp. the much-
jumbled hexameter on the Douris cup (PMG 938e); the Muse herself is asked to begin
from the god at H.Hom. 31.1. Without an explicit reference to song, forms of ³[Zg_Rb plus
genitive open or close hymnic passages at H.Hom. 9.9 = 18.11, 25.1, 31.18, 32.18, H.Ven.
293, Hes. Th. 36 (on the two proems of the Theogony as hymns see FriedlÅnder 1914,
Minton 1970; Furley 2011: 211f. now construes the whole poem as a hymn), A.R. 1, Arat.
Phaen. 1, Theocr. 17.1, 22.25, and parodically Batr. 1; Hes. Th. 1 and H.Hom. 9.8 conflate
the two tropes (Metcalf). Alcman freely deploys this commonplace (PMGF 14(a)-(b),
27.1f., 29), but it becomes sidelined in later lyric: it occurs in [Terpand.] PMG 698.2, but
Pindar pointedly presents it as a rhapsodic characteristic at N. 2.1-3 (cp. Pae. 7b.8, in an
explicitly Homeric context) and of venerable mythical antiquity at 5.25, although he uses it
in a less epically marked way at 3.10f. and fr. 75.7-10 Sn.-M. (as understood by KeyÜner
11 and, rather too hesitantly, Race 1997: II 311).
180
The polar expression with the beginning and the end, and sometimes the middle to boot, is
also topical (KeyÜner 1932: 13f., West 1966: 166f.). Beside Thgn. 1-4 (Apollo: 1f.
beginning and end, 3f. beginning, end, and middle), one finds Hes. Th. 34 and [48] (the
Muses), fr. 305 M.-W. (Linus), H.Bacch. 17f. (Dionysus), H.Hom. 21.4 (Apollo), down to
Theocr. 17.1-4 (1f. beginning and end: Zeus, 3f. beginning, end, and middle: Ptolemy),
Arat. 14, and in a different context Dionysius Chalcus (cited above, p. 175), eleg. adesp. fr.
27.1f. W. Observably ì[t]he principle is one more honoured in theory than in practiceî
(West 167): it refers, of course, to singing of the god ëalwaysí (van Groningen 1966: 9),
and its assertion is performative at least as much as it is descriptive. In a self-standing cult
song the difference between the two interpretations might be minimal, insofar as both the
opening and the close will typically be concerned with the deity addressed, and in any case
the entire song is performed in his or her honour; there is little reason to assume that it was
not so for F1.
3 BcfgFh[i[ TO jcTk : cp. B. 11.16 �RpP�iQg[bg] ÊRfgt]. The parody at Ar. Eq.
1265f. skips 3 altogether, resulting in a metrical pattern identical to Pae. 5.1-3.
4 fZl[ µddh[ nXWTOY]c[ : surely Artemis, ÊRfgt] �ccg]¦R pPSnf\[ at O. 3.26 and
probably also at F18.3 (see n.); Hummelís assumption that the referent is still Leto (2005:
348, 548) stretches fW Ö jR^ to an almost unacceptable degree. Relying on the Sudaís
reading �VRfv[R], patently derived from Aristophanesí parody of Pindarís fragment,
Heyne 1798-99: III/1 47 (following the lead of Schneider 1776: 32) understood the passage
as a reference to the Dioscuri; given the context, this is hardly a possibility, as Boeckh saw
when restoring the genuine reading of the scholia (1811-21: II/1 588). Presumably Artemisí
181
name or a different designation (e.g. pPSRfY[R) followed somewhere after �W~]Rb.
�VnfWb[RQ must mean ëchariot-driverí, as �VRfq[ regularly does in Homer (LfgrE s.v. 1),
not ërustlerí as in H.Merc. 14, 265, 377: Pindar combines the sense of the former with the
genitive construction of the latter, cp. A. Pers. 32 µccTQ Ö �VRfq[.
Why Artemis should be called µccTQ �VnfWb[R is unclear (see Farnell 1896-1909: II 450,
but add Artemis HTVi with Furley and Bremer 2001: II 389). She helps Hieron harness his
team at P. 2.7-12, but on that occasion she may be introduced more as a patron goddess of
Syracuse than because of any specific connexion to chariot-driving (KNhnken 1983: 62 n.
52). She is labelled �ccbjU pW¦] by ësomeí according to inscr. b N. 1 p. III 7 Dr. in an
attempt to justify the address to Ortygia in the opening of the ode: this may either derive
from passages such as ours and O. 3.26, or be autoschediastic, and the scholiastís citation
of two perfectly irrelevant parallels (Sophr. fr. 166 K.-A. �f[Y]f\Q, Il. 6.205 Z[P]qQbgQ)
instead of the Pindaric material suggests the latter. Homerís Z[P]qQbgQ only confirms the
notion that she, like other goddesses, could be imagined to travel on a chariot (Verdenius
1987: 26), which poets variously represent as drawn by horses (H.Hom. 9.3f.), lions (Dith.
2.19-21 with Lavecchia 2000: 165f. contra van der Weiden 1991: 76), or deer (Call. Dian.
98-112); this explains pg¬Q µccTQ �VnfWb[RQ no better than it does �ccg]¦R at O. 3.26.
Rose 1943 suggests that the passage in O. 3 rests on the identification of Artemis with the
Spartan Orthia (cp. 30), whose connexion with horses is archaeologically attested. The link
seems quite tenuous (see KNhnken 62f., whose own explanation that ìthe only justification
Ö seems to be the odeís commemorating a victory in the horse-racesî is not particularly
more satisfactory): with respect to F1 one notes how Paus. 3.11.9 places statues of Apollo
Pythaeus, Artemis, and Leto in the Spartan agora, an area called mg[¦] because there, at
182
the Gymnopaediae, g� ¯d\�gb Zg[gˆ] �]f¬]b f�b �c¦VVTQb (cp. Hsch. S 1002 Latte, EM
col. 701 Gaisford: Pettersson 1992: 42-56, esp. 42f.), but this is hardly specific enough to
pinpoint a definite connexion.
pg¬Q µccTQ is Pindarís variation (also at P. 4.17, cp. 11.48 with hypallage) of the
formulaic epic joint of µccg] with ‹j�] or ‹j�cgP] (Il. 2.383, 3.236, 4.500, etc.), which
he does use at Parth. 2.44; see further examples at F2.7 n. Together with �Rp��TQgQ, the
further epic variation �VnfWb[RQ, and the hexameter-hymnic conceit of ëbeginningí the
song (see 1 n.), the phrase lends to the opening a distinctly Homeric flavour. If Apollo was
mentioned third after Leto and Artemis, and especially if the destination of the ode was
Delos, another morsel of Homeric qualification is arguably the order of the three deities,
reversed from the one they explicitly have at H.Ap. 158f. c[�fgQ _ÓQ �c¦VVTQí
Õ_Qq]T]bQ £ R“fb] kí R“ Ê\fi fW jRÚ ê[fW_bQ ∞gZYRb[RQ.
183
F2
Introduction
The surviving text of F2 and F3 is preserved for the most part by a large fragment of H29,
and to a lesser extent by two scraps of H7 (only the first of which contributes to F2), which
identify the pieces as Prosodia.1 No metrical markers are preserved before the end of F2,
which is signalled by asteriskos (also on H7), koronis, and the tiniest of paragraphoi. No
responsion can be ascertained from its extant verses. There may have been a sign in the
damaged left margin under any line between ii.2 and ii.5, a sign which could only have
been a paragraphos (marking the end of an antistrophe) given the size of the surviving
koronis, the like of which cannot have been entirely obliterated by the lacuna. Accordingly,
if one can assume that metrical signs were consistently and correctly placed throughout
H29, the poem was either monostrophic with a stanza of at least fifteen verses (comparable
1 See pp. 12f. above. Since the height of the columns is unknown due to the loss of the lower section of the papyrus, and thus the number of lines missing between the last line that is visible in col. i and the top of col. ii, I number the lines column by column, following Lobel 1961c: the margin above col. ii is preserved and so provides an accurate term of reference for col. i. Sn.-M. 58, Bona 1988: 258, and Rutherford 2001: 406 number the lines of F2 continuously on the assumption that col. i had twenty-five lines (but see Sn.-M. ibid. ìdesunt vv. non minus 3î, my italics) � oddly, since col. ii visibly had at least twenty-six. For ease of reference I also print their line-numbers between brackets. As the essay on F2 in Rutherford 2001: 407-10 is largely identical to his earlier treatment of the fragment in Rutherford 1992: 60-2, I cite the former only when it offers new material, and only the latter when they coincide.
184
in length to that of Pythian 12), or triadic with an epode counting more than ten, and not
fourteen, lines.2
Of the first column only a few letters at the end of the longer verses are preserved, to
which little can be added from the marginalia to their right. The absence of civil strife
seems to be mentioned (| 13, but other interpretations are possible), song (| 20), and
comrades (22). It is unknown whether the former two are to be connected: song features
prominently in Bacchylidesí description of the blessings of peace at fr. 4.61-80 M., but the
considerable distance between the two verses and the lack of a context makes this
impossible to ascertain. In turn, §fR~[gb in close proximity to �gbkn may conceivably
refer to members of a chorus (see 22 n.), but this link too is open to doubt. The overall train
of thought cannot be recovered.
The second column opens with the mention of a city and (presumably) bronze, then a
few illegible lines are capped by a reference to rewards consisting of good repute (6)
before a string of statements brings the song to a close. Lines 7-12 are concerned with
song, arguably expanding on the previous verse: a Muse is ìclear-voicedî (7), speaks ìa
speech of pleasant wordsî or ìversesî (8f.), and ìwill make even one who lives far off
mindful of the ô[TR] pWR[^Rî (10-12). The meaning of the last expression is doubtful,
given the ambiguity of both words (see 11f. n.); it is uncertain even if it refers to the cult or
to a myth, although it does suggest that the cultic setting of the ode involved hero-worship
in some form. Three maxims on perfection and touchstone-tested gold (12f.), swift
judgement or understanding (14, if SQi_R] and fRZW^R] are to be taken together), and
2 The suggested length of a monostrophic stanza is due to the practice, observable in Pae. 5 on H4, of separating monostrophic stanzas with a koronis as well as a paragraphos. If this practice was not consistent (which is possible, since Pae. 5 is the only certain Pindaric example preserved on papyrus), then F2 may have been a monostrophic composition with stanzas of eleven, twelve, thirteen, or fifteen or more verses.
185
lifting through ]gd^R (15), each mutilated of its ending, bring the song to a close. The loss
of the predicate at lines 12f. and of the subject at 15 (where �W^[WfRb can be passive or
middle) makes interpretation arduous: the context suggests that the referent is still song
(either F2 itself or more generally), but it is uncertain whether it is likened to the basanos
or to the gold that is put to the test (see n.). Along the same lines, it is doubtful whether the
last verse refers to the fame that poetry confers on excellence (or the like), continuing the
thought of lines 6f. and 10-2, or to the evaluation of that poetry by its audience.
If it is legitimate to assume that lines 7-12 refer to (or at least include) F2 itself, the
closing passage constitutes one of the most explicit visions of reperformance in Pindarís
extant work, almost on a par with Nemean 4.13-16 and Isthmian 2.44-6.3 The Museís
agency links into a continuum the eternal present (¿R[^�Wb 8) of each performance with a
future (_Qn]Wb 10) which is ever distant in space as well as in time. As for the scope of the
commemoration, one is reminded of the opening of Nemean 5 (or of Olympian 9.23-5 or
Pythian 2.67f.), but here the focus is on the destination of the message rather than on its
starting-point or its journey, with personal reception highlighted alongside purely spatial
diffusion; one can parallel the closing line of Bacchylides 13 cRQfÚ jR[�XgQfb VR[�]�, ìa
straightforward text-external future Ö referring to the immediate and long-term reception
of this ode by its publicî (Douglas Cairns).4
As we saw, the Museís memorialisation of the ô[TÍ] pWR[^R (whatever the exact
meaning of the phrase) suggests that the song was somehow connected to hero-worship.
This may also have been the reason why F2 was classed among the Prosodia; the reference
3 Hubbard 2011: 348f. On other such visions of reperformance in Pindarís poetry and their possible scenarios see Loscalzo 2000a, 2003: 85-119, Currie 2004, Hubbard 2004, 2011. 4 2010: 331.
186
to pWR[^R, if taken to mean an inter-polis religious embassy, may have been another.
Given the state of the text, it is not possible to speculate further on the context of the
performance. Franco Ferrari supplements Sn]£_TQ at lines 7f. and understands it as a
reference to a hieros gamos, as had been argued by Ian Rutherford for F3 and Giambattista
DíAlessio for F8.5 The supplement is not beyond doubt, and nothing in it specifically
demands that the gamos it envisages must have been played out in the ritual as well as (or
instead of) in the mythical narrative that probably preceded. Even if both Ferrariís
conjecture and his interpretation of it are accepted, the reader will be no closer to
identifying a specific location or honorand.
The only piece of information that is known on the subject is known only negatively.
According to the title written on H29, F3 was composed for the Aeginetans and honoured
Aeacus. As we have seen, according to the common practice of these papyri, the fact that
the title of F3 was written out in full means that F2 was not concerned with the same
community: if it were, fg~] RÃfg~] would have been used instead.6 If one accepts
DíAlessioís suggestion that the Prosodia were arranged alphabetically by ethnic of the
performers,7 then F2 will have been commissioned by a city other than Aegina but whose
name also began with alpha. If, however, the argument laid out in Chapter 2 against an
alphabetical ordering of the Prosodia holds true, our choice is not even restricted to these:
all we know is that F2 was not an Aeginetan ode.
5 Respectively by Rutherford 1992: 64 (more tentatively than Ferrari allows) and DíAlessio 1997: 36f. 6 See pp. 20-2 above. 7 See pp. 18-23 above.
187
Commentary
Col. i
13 : the first word is illegible (×Ã¥½ no less possible than anything else, pace Lobel 1961c:
28), �¾�[n almost certain. Lobel parallels Parth. 2.63 ��¾�[ÏQ [¯][�Q, cp. also fr. 109.4f.
Sn.-M. ]fn]bQ Ö �Zp[ÏQ jgP[gf[¦dgQ. He rightly questions ìwhether �Zp[n would
have been thought to require an interpretation at allî, but the point of the note may be that,
according to ìsomeî, its meaning in this passage is active (ëhostileí) not passive (ëhatedí).
Bruno Currie points out to me that gà �^ R�� could be glossing Lobelís Ú ��[n only (without
×Ã¥j, or whatever stood in its place), with which Pindar would be marking ]fn]b] as having
a meaning other than ëseditioní: possibly one related to choral song and dance, given the
frequent use of µ]f\_b and compounds in this sphere (see Pae. 2.98-100, E. El. 178-80,
Alc. 1154f., IA 676, Ar. Nub. 271, etc.). I wonder whether Ú ��[n is an acceptable reading
(alpha possible, beta improbable, and the preceding epsilon unexplained), but a similar
interpretation can be envisaged with ×Ã¥j ��¾�[n, a ënon-violent stasisí. However, both of
these hypotheses would require the annotator to have made a fairly simple point in a very
roundabout way by glossing the adjective instead of the noun.
o 15 : parallels such as | A Il. 4.244b Erbse ô kbcVv kbÏ fÙ �VVW^cWbQ fUQ kbn c[¦pW]bQ,
µQí õb kbÏ cWk^gP ñ explaining the sense of a passage by alleging the ellipsis of kbn and
offering the ëcompleteí construction ñ show, contra Snell, that the preposition need not
have stood in the text. I cannot tell that a note of this kind is not what we have here. In any
188
case, µQí õb will have introduced a form of paraphrase, crammed in the following line
either above or to the left of | 16; see Calvani 1996: 300-4.
o 20 must be a variant (Lobel 1961c: 28): a gloss would have used ‹bkvb, as the scholia
vetera always do; an accent is unexpected in a marginal note otherwise. If the variation
only concerned the noun, the text may have had �Q �gbkR~] (Lobel): other isometric
synonyms are not easily found. Of the two, Pindaric usage distinctly favours the plural
(data in Slater 1969 s.v.). Otherwise, it may have been some equivalent phrase such as ]ˆQ
�gbk¬b (N. 9.49), ]ˆQ �gbkR~] (P. 10.57), Õcí �gbk¬] (O. 4.3).
22 : ìor possibly a compoundî (Lobel 1961c: 28). §fR~[gb are called the members of a
chorus at O. 6.87 (and of a victory komos at O. 9.4), but nothing guarantees that the word
was connected to the ësongí of 20.
Col. ii
1 : editors may be right to follow Lobel 1961c: 28 and separate c¦Vb�, but é–cgVb� ñ
attested in Poll. 9.27 and as a personal name, presumably presupposing the existence of the
adjective, in the fifth century (data in Storey 2003: 53) ñ would not be beyond Pindaric
inventiveness; for a list of his compounds in -cgVb] see Hummel 1997: 65, 1999: 313.
Beside the intriguing but scarcely substantiated hypothesis of a ìbronze(-walled?) cityî
aired by Bona 1988: 260, the proximity of the two terms ñ which are not necessarily in
agreement ñ suggests an image of battle, possibly borne out by the mention of heroes at 11;
189
however, the voice of the singing Delphic maidens at Pae. 2.100f. is also ìbrazenî, so a
reference to praise cannot be excluded (for song praising cities see e.g. P. 8.25, N. 10.1f.,
cp. O. 5.10-12).
6 OÃqNZrUca Ní ndU^OY]c : if WÃ¥kg�^R] is a genitive, it is appositional, ìreward consisting
of good renownî (Rutherford 1992: 60 after Lobel 1961c: 28) rather than objective,
ìreward for gloryî (Bona 1988: 260): fame as promoted by song usually is the reward
(³cgbQR O. 7.16, P. 2.14, N. 7.16, I. 3.7, 8.4; _b]p¦] N. 7.63, I. 1.47-51; etc.) of valour and
achievement, not something that invites reward in turn ñ although N. 6.45-9 is a possible
counter-example if k(Y) (48) relates 48f. to the �cW^-clause of 46f. and not to the main
clause of 45f. (cp. F5.1-9). But it can also be accusative plural with �c^ZWb[R predicative,
ìgood renown as rewardî (for the plural cp. I. 3.3, 6.21 WÃVgS^Rb], Pae. 2.67 WÃRSg[^Rb]b,
etc.). Whose renown it was is up for grabs; 10-12 may suggest the rite or festivalís, but that
of its heroic honorand(s) (11) is a valid alternative, and would suit well the transition
between myth and actualit�, if this is what we have here.
NO[ : possibly a form of kYjg_Rb (Lobel 1961c: 28), cp. I. 1.50f. √] kí Ö ³[\fRb jtkg]
Ú�[¦Q £ WÃRSg[\pWÚ] jY[kg] —°b]fgQ kYjWfRb. If the word was a verb-form, two obvious
alternatives are a form of kW^jQP_b if what precedes is an accusative, or of kYkg[jR if a
nominative: cp. respectively I. 8.47f. QWR[ÏQ ¯kWbXRQ ]gd�Q £ ]f¦_Rfí �cW^[gb]bQ
�[WfÏQ �ZbVYg] and O. 1.93f. fÙ kÓ jVYg] f\V¦pWQ kYkg[jW, N. 3.83f. f^Q SW _YQ Ö
´W_YR] {cbkRP[¦pWQ fí ³cg jRÚ ÙWSn[TQ kYkg[jWQ dng], 9.41f. kYkg[jWQ £ cRbkÚ
fgtpí �S\]bkn_gP dYSSg]. The latter is quite attractive, since fame regularly shines, and
often does so from somewhere (cp. -pW 7): see Gerber 1982: 146f. and the next note.
190
7 -fO : Lobel excludes a verbal ending -]pW or -ZpW, which would not have been divided
before the p (1961c: 28). Actually, as Daniela Colomo informs me, the rule does not seem
to be followed with great consistency in the papyri, and indeed S.E. M. 1.173f. mentions
precisely the treatment of pre-consonantic sigma as a matter of pointless grammatical
debate; but one supposes that in lyric papyri ñ where colon-breaks were transmitted
together with the text, unlike in prose ñ the rule would have been more strictly observed. In
the context of good repute, the local suffix (Snellís f\V¦]£pW or a precise location)
indicating its diffusion seems a likelier option than e.g. ¯QW[]£pW referring to its survival
after death (cp. P. 1.92-4, N. 4.79-83, 6.29f., fr. 121 Sn.-M.): cp. O. 1.93f. and N. 3.83f.
cited above, N. 3.64 (RÃf¦pWQ, but see Slater 1969 s.v.), 5.3 (�cí o∞S^QR]), 6.48f.
(f\V¦pWQ), 7.50-2, I. 4.11f. (gÒjgpWQ), etc. Currieís õV]£pW is another interesting
possibility, cp. O. 3.39 (jtkg]), Pae. 2.68 (dYSSg]); in this case, kW[ will not be part of a
verb-form.
XUtOYc _Ó[ vZwaí : after Ùg~]í a high stop above the line, next to the apostrophe. If not
stray ink (and it does not seem so, pace Lobel 1961c: 27, cp. 12 pWR[^R]�), a reader must
have taken these words as a noun phrase. Whether punctuation in papyri was sometimes
part of the transmitted text is debatable (certainly it was not always, see GMAW2 9f. and n.
41), and in H29 it was probably not penned by the original scribe: the annotator may simply
have been wrong. However, we should hesitate before charging him with error in a
fragmentary context, given the possibility that his choice was dictated by good evidence
(e.g. kY) now lost in lacuna. Noun phrases are a hallmark of Pindaric style, from ³[b]fgQ
_ÓQ —kT[ onward (Benveniste 1950: 29-31, Guiraud 1962: 33, 73f., 220f., Hummel 1993:
297-304, see also LanCrys 1996, 1997), and the short, paratactic sentences of this fragment
191
are compatible with this interpretation; I see no compelling reason to assume a scribal
mistake. For elision over syntactical pause see e.g. P. 4.94, 133, 241, 278, etc.
VbS�] and cognates routinely refer to aspects of music and song since Homer (see
Nordheider in LfgrE s.v. VbS�_gVcg], VbSP[¦] B.4-6, VbS�] B.1c-d, 2-4, VbS�dTQg], cp.
Thgn. 241f.) and are a favourite of Bacchylidesí in this capacity (10.10, 14.14, frr. 4.57,
20b.2, 20c.1, 23.4 M.); Pindar too has k¦XRQ Ö �cÚ SVi]]Rb VbSP[¬] �j¦QR] (O. 6.82)
and �cYTQ Ö g∂_gQ VbS�Q (9.47) (see Kaimio 1977: 231-3 and passim). Formulaic Ùgt]R
V^SWbR occurs in Od. 24.62, h.Hom. 14.2, 17.1, 20.1, Alcm. PMGF 14(a).1 (cp. 30),
[Stesich.] PMGF 278.1 (cp. 240), lyr. adesp. PMG 1045, and well into the Hellenistic age
(Massimilla 1996: 238 on Call. Aet. fr. 1.42 Pf.). The inversion of an epic nexus is amply
paralleled in Pindar, see examples in Galletta 1988: 78-105, Sotiriou 1998: 6-10, 70-5.
However traditional, the adjective finds particular significance in the context of fame and
its proclamation (6, 10-12), cp. the Homeric j\[�jW]]b VbSPdp¦SSgb]b (Il. 2.50, 442, 9.10,
23.39, Od. 2.6) with Kaimio 1977: 43.
\xí yz[[ñ : the only one-word solution to cover the end of 7 and the beginning of 8 is
�dR�[Qb]]£_�Q, which clearly will not do (Lobel 1961c: 28). A part of �dRSQ^�T seems
just as out of place, and Snellís ³dR� does not suit the traces (so already Lobel): a vertical
and a horizontal that protrudes slightly to the left (gamma, pi, or just about possibly sigma,
cp. that at the end of Hg]Wbk¬Qg] and ´\[W�] at F3.3f.). Christopher Metcalf reminds me
of Hes. Op. 3 ³dRfg^ fW dRfg^ fW: here tau is excluded, but the reference could be to the
�dR]^R (not a Pindaric word, but cp. Od. 4.704) that the Muse would counter just as the
speaker counters �_RZRQ^R at Pae. 6.10 or °¦Sg] at N. 7.61. However, much would have
192
to be conjectured in the missing part of 7 for the syntax to flow smoothly into 8. (Against
the otherwise potentially attractive ³dRQfg], which Pindar refers to lack of fame at P.
11.30 and N. 8.34, I point out that nu is an impossible reading.) Perhaps an easier solution
is to articulate �dí Ú �[. The natural supposition (if punctuation after Ùg~]í is correct) that
kY may hide in lacuna is no obstacle: Pindar often postpones it after a preposition (Slater
1969 s.v. 4.�). Ferrariís �dí Ú�[Qñ is a sensible conjecture, especially if ñ as it seems
likely ñ the subject is still the Muse; �dí ÚÈ[RV- is another possibility, but the adjective is
unattested so far in Pindar.
8 TOXOeTcwa : ìThe basic meaning of fWVWPfq is ëendí, the ëextreme partí of an action,
event, or process Ö In poetry, however, fWVWPfq also occurs as an equivalent of fYVg]
ëperformance, execution; realisationí (without contrast with �[Zq or the like)î (Waanders
1983: 250; discussion at 241-50, Pindar at 243f.) ñ not Rutherfordís ìriteî, an unattested
meaning so far as I can see (1992: 60 and n. 8; Waanders 1983: 241, whom he cites, says
nothing of the sort). If one wants to see a reference to a rite, a genitive is needed (cp. -_TQ)
to specify the actual referent while taking fWVWPfR~] as ëcompletioní ëperformanceí; such
periphrases are well attested, and that with Sn_TQ especially so (Ferrari 1992b: 151f., with
parallels). Ferrariís restoration is quite persuasive on its own terms (better paralleled and
more meaningful, at any rate, than Snellís [�Q ji]£_TQ), but there is a slight degree of
arbitrariness in conjecturing a word which invests the text with a wholly new connotation
(in this case, performance in a hierogamic ritual, see DíAlessio ibid. 152) without any clue
to the same effect elsewhere in the fragment.
193
fWVWPfq has such a broad range of meanings that it is virtually impossible to pin it down
without context. It can conceivably stand alone and mean ëat the endí (cp. P. 1.35
fWVWPf¬b); the plural, pace Lobel 1961c: 28, is not too troubling. If so, the lacuna at the
end of 7 can perhaps be filled with a minimalist solution such as e.g. �dí Ú�[Q�Q (kÓ)
k¦]£_TQ: for the Musesí house(s) cp. Il. 2.484 = 11.218 = 14.508 = 16.112, Hes. Th. 75, 114,
E. HF 791, etc. (noting the metapoetic context of the epic examples), for ÚSQ¦] said of
locations related to deities cp. h.Merc. 186f., [Hes.] Sc. 203, Alcm. PMGF 14(b), Sapph. fr.
2.1f. Voigt, O. 5.10, P. 4.204, B. 10.29f. (Broger 1996: 32). Otherwise, fWVWPfq can be
specified by -_TQ (if a genitive at all) and be either adverbial or governed by a preposition
(�Q, �c^?) to be supplied at in lacuna at 7, with a meaning ranging from ëendí to ëoutcomeí
to ëperformanceí to ëcomplete courseí (O. 13.75 c¬]RQ fWVWPfÏQ c[nS_Rfg] with
Waanders 1983: 243 contra Slater 1969 s.v. c): Stonemanís ìat the conclusion [of the
festival]î (in id. and Conway 1997: 347) is one possibility among many. This extends the
lacuna considerably, but since the length of the verses is unknown, it cannot be ruled out.
As something that may not be obvious, I mention that a reference to the death of a group of
people would be compatible with the references to heroes (11) and glory (8) and with an
apparent emphasis on commemoration; for the explicit association between death and the
spread of renown cp. N. 7.31f.
¿c]UFO[Y : ìƒR[g] and cognates belong to epic vocabulary and have the same spread of
meaning as English ëassociationí, ëintercourseíî (Bulloch 1985: 174); pace Dyer 1964: 130
and EDG s.v. ƒR[, speech is nowhere in evidence in the epic occurrences (Il. 6.516,
22.127f., h.Merc. 58, 170, h.Ven. 248, h.Hom. 23.3, Hes. Th. 205) or in several later
194
attestations (e.g. PMG 813.9), although it does become one clear focus of the word-family
from Pindar onward, with or without amorous connotations (see Bulloch, but note the
scarcity of such connotations in the two greatest addicts of ¿R[^�T, Quintus and Nonnus).
The amorous element is absent altogether in the Pindaric occurrences (P. 1.98, 4.137, N.
3.11, 7.69; no judgement can be formed on H26 fr. 93.2); Slater 1969 s.v. has ìlow, soft
voice(s)î, but if a particular aspect is emphasised, it is intonation more than volume (see
DELG s.v. ƒR[, Calame 1977: II 151). ¿R[^�W[b will thus mean ëuttersí ëspeaksí in the
charming fashion one would expect of a Muse. Despite the connexion with �[R[^]jT
alleged by Orion col. 114 Sturz1, Et.Gud. col. 417 Sturz2, EM col. 1740 Gaisford (cp.
Apollon. Lex. p. 118 Bekker2, Hsch. g 16 Latte), the meaning ìattachesî suggested by van
Groningen 1963: 128 is a phantom.
9 : not really ìan odd phraseî (Lobel 1961c: 28): to N. 9.3 �cYTQ SVPjˆQ —_QgQ (id.) add
7.16 jVPfR~] �cYTQ �gbkR~], 9.7 pW]cW]^R kí �cYTQ Ö �gbkn. Consistently with the
argument of the passage, one feels that the use of V¦Sg] emphasises content and its
effective communication (see Slater s.v. 1.a) rather than form. Snell and Ferrari suppose
that a word is lost in lacuna (the former a participle governing the accusative, the latter an
apposition governing the genitive), but there is nothing obviously incomplete with the
verse as we read it: the genitive can be appositional, governed by V¦SgQ directly; ¿R[^�T
takes an internal accusative at h.Hom. 23.3 (cPjbQgˆ] ¿n[gP]), a construction that Pindar
perhaps replicates while dispensing with the figura etymologica (Rutherford 1992: 60 n. 7).
10 _[WaOY : ìthere is nothing to show that a compound did not occur. Pindar has �_QR- as
well as the simple verbî (Lobel 1961c: 28).
195
ìThe future can be used to make open ended categorical assertions of a type suitable for
effecting closure, hence it is often found at the end of poems, not exclusively in
encomiastic contexts. When the context is encomiastic, however, the basic idea is to extend
the present occasion, and the praise it embodies, into a limitless (hence grammatically
indefinite) future populated by generations unborn (hence again tending to the
grammatically indefinite) Ö the encomiastic paradox is that the ode itself points beyond
itself by referring to its own future receptionî, thus defying the distinction between text-
internal and text-external futures (Pelliccia 1995: 331, citing Ibyc. PMGF S151.46-8, B.
3.94-8, 11.123-6, 13.228-31): futures such as these refer to a moment beyond the first
performance of an ode, but never truly beyond the ode itself, since it is the reperformance
and remembrance of that very ode that fulfil the action described.
It is also possible that _Qn]Wb refers to mindfulness of the future, not (only) of the past,
with the Muse portrayed as a spondophoros-like figure announcing the pWR[^R (if taken as
a state-pilgrimage, see ii.11f. n.) abroad: the Pindaric speaker claims to fulfil a similar
function by the Musesí mandate at Dith. 2.23-5 �XR^[Wfg[Q] £ jn[PjR ]gd�Q �cYTQ Ö
�VVnkb jR[V]�[bZ¦[Tb. Interestingly, such figures are routinely called theoroi in inscrip-
tions from the fourth century onward, and their delegation may have been termed theoria
in an honorary decree from Olympia as early as the mid-fifth century (Inv. B6970, still
unpublished; Perlman 2000: 19, 63f.). On epangelia and its terminology see Boesch 1908,
Perlman 2000; on poetry construed as a message see the Pindaric examples in Braswell
1988: 379, add N. 4.73-5, B. 2.1-3, 13.230f., and cp. the conceit of ësendingí the ode at P.
2.67f., N. 3.76-80, fr. *124.1f. Sn.-M., B. 5.9-12, 195-7, frr. *20B.3f., *20C.6f. M.,
[Terpand.] PMG 698.2 = fr. 3.2 Gostoli, with Tedeschi 1985.
196
The locus classicus for the association Ùgt]R ~ _b_Qqb]jT is Il. 2.491f., quite possibly
interpolated (Heyne 1802: IV 285) but early enough to be known to Ibycus, PMGF
S151.23-31 (West 1998: 66). Pindar makes this association also at O. 6.90-2 and N. 1.11f.;
Hellenistic literature seems to neglect it, except A.R. 1.22f., while the equivalent Musa (or
Camena) ~ memorare is common in Latin poetry (Hardie 2007: 559 and nn. 44f., add
Verg. ecl. 7.18, Prop. 2.34.31, Sil. 1.3, Stat. Theb. 7.228).
TY[c : Snell suggests identifying this ìsomeoneî with Pindar, qui ipse non adest in hac
ëtheoriaí (1959-643: II 56). This entails construing _Qn]Wb with the accusative alone and
leaving ô[TÍkg] £ pWR[^R] to be governed solely by §jn] (11) (ìpossible, though I think
very improbableî Lobel 1961c: 28): ëand will also cause to be remembered one who dwells
far offÖí (cp. P. 11.13f. ¯_QR]WQ �]f^RQ Ö cRf[ibRQ). This would turn 10ff. into a
claim to future poetic fame comparable to H.Ap. 165-78, Ibyc. PMGF S151.46-8, and B.
3.96-8: poetry produces remembrance not only of those praised (cp. 6), but also of the poet
who praises them. This admittedly suits well the cluster of seemingly meta-poetic gnomai
at 12-15. However, the phrasing of 10-12 is a very opaque way of referring to Pindar
himself: contrast on the one hand Pae. 5.44-8 and F5.59-61, where the potential reference
to the poet as well as the chorus is expressed obliquely but transparently (see F5.60 n.),
and on the other hand P. 4.298f., where the personal self-reference is very explicit indeed
(cp. B. 3.97f., 10.10, 19.11: in each case an evocation of song is specified by a geogra-
phical clue). But how could an audience be expected to understand that ìone who dwells
far off from the ô[TÍ] pWR[^]î is Pindar specifically? More probably, fbQR will indicate
the person reminded rather than the thing recalled, and refer ìgenerally to absenteesî rather
than ìto a particular personî (Lobel) towards whose identity the phrase and its context give
197
no clue. Rutherford 2001: 409 sees a parallel with the homecoming of the theoroi, but
nothing in the text limits the reference only to that.
fbQR has a distinct echo of the anonymous fb] that typically personifies projected reception
in archaic Greek poetry, from the Iliad (6.459, 462, 7.87, 91, etc.: see Kirk 1990: 222, de
Jong 1987: 76-80, 82f.) down to Sapph. fr. 88.13 Voigt, Thgn. 21f., B. 3.97, N. 7.68, with a
particularly close parallel for our passage in Sapph. fr. 147 _Qn]W]pR^ fbQR dR~_b Ü jRÚ
ãfW[gQ Ü ³__WTQ (DíAlessio 1997: 34 n. 69): on this device see Wilson 1979, noting the
tendency of the post-Homeric examples to express praise, and de Jong 1987. Differently
from some of the epic occurrences, we hear no comment from the fb], who is portrayed
only as a recipient rather than also as a respondent; peculiar to our passage is also the
emphasis on distance in space (cp. Snellís f\V¦]£pW at 6f.: Ferrari 1992b: 150) rather than
in time, which is left for the future tense to express (cp. again Sapph. fr. 147). jR^ is
another standard element where distant reception is at stake: cp. the formula jRÚ
�]]g_YQgb]b(Q) (Il. 2.119, 3.287, etc., Od. 3.204, 8.580, etc., Hes. fr. 212b.6 M.-W., Thgn.
251), Il. 3.353 = 7.87 jRÚ ¿°bS¦QTQ �Qp[icTQ, 6.356 jRÚ ¿c^]]T, etc., and again Sapph.
fr. 147. jR^ fbQR QR^g[Q]£pí §jn] closely recalls Alc. fr. 328 Voigt jR^ fb] �cí
�]ZRf^Rb]bQ gÒjWb], but the absence of a context precludes assessing any significance of
the parallel. For the notion of ëputting in mind of one who dwells far awayí by means of
poetry cp. I. 6.22-7, 8.47f. with Privitera 1982b: 236.
One remaining question is whether the genitives at 11f. are governed by _Qn]Wb, §jn] (the
thing recalled being inferred from the lines up to 7), or both. The second option makes
acceptable sense, but the construction with the accusative of the person reminded but no
198
genitive of the thing recalled is unparalleled in Pindar and scarce elsewhere (e.g. Od. 12.38
_Qq]Wb kY ]W jRÚ pWÙ] RÃf¦]); on balance, taking ô[TÍkg] pWR[^R] with _Qn]Wb ñ with or
without §jn]: the difference in meaning is minimal ñ is the safer choice.
[cUZ[[]£fí : Lobel 1961c: 28 notes the Sophoclean mannerism of QR^T said of temporary,
not permanent or habitual, location (Tr. 99, OC 117, 137), but in each of these occurrences
inopportune lingering is implied: Heracles not coming back to his longing household,
Oedipus hiding away from the anxious chorusí sight. I doubt whether such a connotation
(ìeven one who is missing out on the ô[TÍ] pWR[^Rî) can be seen here.
11f. }]h~NZa £ fOc]Uca : the major question-mark of the fragment. In theory, each word
can be adjective, noun, or proper noun, producing six possible combinations: (a) ëthe
heroic theoriaí, (b) ëthe theoria of the heroineí, (c) ëthe theoria Heroísí, (d) ëthe theoric
heroineí, (e) ëthe heroine of the theoriaí, (f) ëthe heroine Theariaí. More acrobatic
solutions that involve separating ô[TÍkg] from pWR[^R] (ìwill remind of the theoria also
one who lives far away from the heroineî, or the reverse) are theoretically possible, but
make awkward word-order (Lobel 1961c: 28), and have little to recommend them as far as
sense is concerned. Furthermore, the noun theoria itself has a broad semantic spectrum,
covering the contemplation of an entity with the eyes or the intellect, the visitation of a
festival abroad or (less frequently) at home whether in an official or personal capacity, and
(as we have seen) a delegation sent abroad to announce such festival (Rutherford 2000c:
133-42, bibl. at 133 n. 3).
(d) and (f) require of pWR[^R a new form and meaning (only the masculine pWn[bg] is
attested, an epithet of Apollo at Trozen: IG IV 748.16, Paus. 2.31.6), with less than
199
impressive results; thus, pWR[^R] is probably a noun (Rutherford 1992: 60f.). (e) is roughly
equivalent to (d), and makes dubious sense. (c) evokes Plut. Mor. 293c-d (Slater 1969 s.v.
ô[TÍ]): an enneateric festival Heroís was held at Delphi after the Septeria, largely with a
_P]fbj¦Q Ö V¦SgQ but connected to Semele as far as an informed observer could guess
from the rituals. While nothing in the text forbids a connexion with Delphi or Semele,
there is no particular reason to see one either, and �[TÍkg] pWR[^R] would be a rather
blunt designation by Pindarís standards (contrast e.g. Pae. 6.61 �Q pW�Q XWQ^Rb to indicate
the Theoxenia); it is not too attractive a solution.
Choice between (a) and (b) hinges on whether ô[TÍ] is adjective or noun. Ferrari 1992b:
151 makes a case for (b): ô[TÍ] is a noun in its one other Pindaric occurrence (P. 11.7f.
�c^Qg_gQ ô[TÍkTQ £ ]f[Rf¦Q), while for ëheroicí Pindar could have used the isometric
and unambiguous ô[TÍR], as he does at O. 13.51, N. 7.46 (for the objective genitive cp.
e.g. Ar. Nub. 307 c[¦]gkgb _Rjn[TQ �W[ifRfRb, Pl. Leg. 650a _WfÏ fv] fgt }bgQ�]gP
pWT[^R]). But one occurrence is a small sample for defining usage: ô[TÍ] is an adjective
at A.R. 1.1048, and Pindar may well have so used it if he wished. If pWR[^R designates an
inter-polis festival, an objection to (b) is the lack of evidence for such festivals celebrating
heroines: this may well be due to our piecemeal knowledge of ancient Greek cults, but the
absence of a whole type from our records demands caution (on heroine-cult see Larson
1995, Lyons 1997). Conversely, inter-polis delegations for hero-cult are attested, albeit in
works of ëfictioní: Philostr. Her. 236 recounts the vicissitudes of a Thessalian theoria to
Achillesí tomb at Sigeum, and Hld. 2.34.2-3.6.1 describes one sent by the Aenianes to
Neoptolemusí grave at Delphi at the Pythian Games (see Rutherford 2009), both
embedding the text of a choral hymn purportedly performed on that occasion. Also the
200
]W_Q¦Q Ö HPp^gP rWn[bgQ on Aegina (N. 3.69f., see pp. 242f. below) suggests a theoric
connexion between the island and Delphi, on whose basis it has been argued that an
Aeginetan theoria to Neoptolemusí tomb was the performance context of Pae. 6 (Currie,
Rutherford). On balance, (a) has the slightest of edges.
But our difficulties are not over yet. If ô[TÍ] is an adjective, is its meaning objective or
subjective? Are we dealing with a theoria conducted for a hero or by a hero? If the latter,
does our visitation take place in cult or in myth? (On the twin meaning of •[T] and its
Pindaric implications see Currie 2005: 60-70.) A comparandum for this question is N. 7.46
ô[TÍRb] Ö cg_cR~], which can indicate ëprocessions such as befit a heroí, ëprocessions
for heroesí, or ëprocessions of heroesí. The first is the likelier (Currie 297-301 with bibl.),
but | 68a pp. III 125f. Dr. suggests the third by explaining S^QWfRb �Q }WVdg~] •[T]b
XYQbR, �Q g∑] kgjW~ ¡ pWÙ] �cÚ XYQbR jRVW~Q fgˆ] •[TR] and paraphrasing fR~]
cRQ\S�[W]bQ �Q R∑] �jq[P]]gQ fgˆ] •[TR] (see also Currie 303). The note may be
autoschediastic, combining extrapolation from the text with information such as the xenia-
type heroic honours that Pindar reportedly enjoyed at Delphi (Paus. 9.23.3, 10.24.5, VA p.
2 Dr., Vita metrica p. 9 Dr., see Clay 2004: 77), but there is no compelling reason to think
so. Similar rites are attested for Heracles at Cos (LSCG 177.59-62), probably Ajax at
Athens (| P. 2.19 p. III 37 Dr.), the Antenorids at Cyrene (P. 5.85-8 with Krummen 1990:
120-3), the anonymous founder of Zancle (Call. Aet. fr. 43.55, 80-3 Pf.), and others; see
more generally Deneken 1881, Jameson 1994. The idea may be applied to Aeacus or an
Aeacid on one interpretation of F3.3 (see n.), cp. also fr. 187 Sn.-M. Thus, reference to a
cult ceremony involving the attendance of one or more heroes conceived as ëcomingí
cannot be excluded; nonetheless, one doubts whether such an act would be called a theoria.
201
Cultic visitations to heroes are attested, however exceptionally (see above), but mythical
visitations by heroes even more so: Theseusí archetypal Athenian delegation to Delos
springs to mind, but the gathering of the ìneighbouring heroesî at Aeacusí court (N. 8.8-
12) is also easily construed in this way (note VbfnQWPgQ ∞kW~Q and 13f.); in cases such as
these, the word pWR[^R is fully at home.
The preservation and diffusion of memory is a long-standing commonplace of encomiastic
poetry (e.g. O. 11.4-6, N. 6.28-34, B. 3.90-2), and is applied to mythical as well as
contemporary figures: P. 3.112-5, I. 5.26-9, 8.47f., etc. (cp. retrospectively Il. 6.358).
Reference to this memorialising power of poetry in relation to a specific mythical deed,
paralleling the ëmythologicalí interpretation of 10-12, is less common, but Pindar exploits
it on a grand scale in the opening of I. 7 and frequently applies it to ëdeedsí more generally
(e.g. N. 6.29f., 7.12-16, I. 3.7f., fr. 227 Sn.-M.). In the rest of his surviving works Pindar
never explicitly envisages the memorialisation of a ceremony (the only thing he claims to
memorialise beside men, gods, their qualities, or their deeds is the city of Thebes, fr.
194.4-6 Sn.-M. with Aristid. Or. 29.57 Keil), but an illustrious parallel for the ëculticí
interpretation of 10-12 is H.Ap. 174-6.
There remains one further possibility: namely, that pWR[^R indicates not a cultic visitation
abroad, but simply a ìspectacleî (Race 1997: II 289), with no intimation of a long journey.
In the fifth century, pWT[^R usually means ëgoing and seeingí rather than simply ëseeingí
(Rutherford 2000c: 134), but the compass of the implied diversion from oneís routine can
be as small as going to the theatre (Ar. Ve. 1005 with MacDowell 1971: 260), cp. S. OT
1492 and the character rWT[^R ëHolidayí in Aristophanesí Peace. Given the emphasis on
202
spectacle and watching that accompanies descriptions of festivals from H.Ap. 151-5
onward, it cannot be excluded that what the Muse is said to do is to put in mind of the
ëspectacle of heroesí even one who dwells far away and thus cannot have seen it for
himself (parallels at 10 n.). In this case, what is foregrounded is communication rather than
memorialisation for the future. Given the loss of the mythical section and our utter
ignorance of the cultic context of the song, there is little chance to pin down an answer.
12f. Bcac[Y£af`[TY NÓ ^]ea�Y : in Pindar and Bacchylides, the touchstone metaphor
occurs with reference to poetry in either of two capacities: (a) the touchstone stands to gold
as poetry stands to the deeds or person it celebrates, like B. fr. 14 Sn.-M., cp. N. 4.82-5 (a
different procedure to the same effect, CannatØ Fera 1995: 416 n. 15), or (b) the touchstone
stands to gold as the audienceís reception stands to the poetry itself, like N. 8.20f. and
probably fr. 122.16 Sn.-M. (Imperio 2000 contra van Groningen 1960: 38f., CannatØ Fera
419-22), cp. also Ar. Ran. 803f., 1198f. with Taillardat 19652: 455f. It has a broader, non-
metapoetic scope in P. 10.67f. (and frequently in Theognis: 417f. ~ 1164g-h, 449f.,
1105f.), cp. O. 4.18, N. 3.70f.; a different shade of meaning in CEG 509.4, where the
aulete Potamon ñ apparently the son of a disciple of Pindarís, | P. 3.137b ñ is himself
]gdg~] �n]RQgQ (see Wilson 2007: 145-8), and another in a sepulchral epigram for Hesiod
(Certamen 14 West, Paus. 9.38.4), whose jVYg] is greatest �Qk[�Q j[bQg_YQTQ �Q
�R]nQTb ]gd^\]. Lobel suggested ìthat something is said to be the fYVg] of tested gold
and ëunderstandingí, ]�Q[W]b], of the swift witî (1961c: 29), thus making fYVg], rather
than the action of assaying, the pivot of the parallel between 12f. and 14, but the point of
comparison in the touchstone metaphor is regularly the test itself and, by implication, the
value it guarantees; there is little point in the image otherwise. Although the train of
203
thought cannot be followed too closely (Rutherford 1992: 62, 2001: 410 is somewhat over-
confident), the text gives no indication that the topic after 7-13 is no longer poetry, and the
frequent metapoetic value of the basanos and the reference to ]gd^R (15) suggest
otherwise (CannatØ Fera 421). I cannot decide between (a) and (b) on the surviving text;
following CEG 509 seems less advisable than either.
T`XZa : due to the lacuna mutilating the end of the verse, it may be adverb, subject (Snell,
Rutherford 1992: 62), object, or a predicate (Lobel 1961c: 29, van Groningen, Bona 1988:
262). The third is the least likely: while ëperfectioní can plausibly pertain to gold once its
purity has been established (cp. N. 3.70f. �Q kÓ cW^[Rb fYVg] £ kbRdR^QWfRb), it is hard to
see what else can be said to be the fYVg] of assayed gold. On the sense of the word see the
classic Waanders 1983: 1-147 (80-4 on Pindar).
14 t[k_ca NÓ Tc^OUca : CannatØ Fera 1995: 421 compares the audienceís fRZW^R]
�Vc^kR] in P. 1.83, where the adjective may refer to the expectation of new topics
(Wilamowitz 1922: 302 n. 2, aptly comparing Ar. Eccl. 582), and suggests that our phrase
may have the same sense. SQi_R can indeed indicate expectations (Slater 1969 s.v. b.�),
but the reference to its ëquicknessí in P. 1 is both clearer and more fitted to a line of
thought than one can discern here; on the other hand, Rutherfordís ì[s]wift judgement (is a
bad thing?)î (1992: 62) lends fRZ�] a negative overtone that I cannot find elsewhere. The
phrase might be genitive or accusative, and nothing strictly demands adjective and noun to
go together. It is worth recalling that SQi_R occurs in the context of assaying gold in B. fr.
33 M., although that sentence is mutilated and without context (see Jebb 1905: 422).
204
ae[[ : Lobelís ]�Q[W]b] depended on his interpretation of 13, which I have criticised
above; the intended meaning of Snellís ]PQ[Wf�Q j[^]b] is unperspicuous (ìof swift
opinion is the judgement of the understandingî?); CannatØ Fera 1995: 421 retains
]PQ[Wf�Q alone, governed by SQi_R] and referring to the audience (cp. O. 2.85). ]ˆQ [
followed by a dative cannot be excluded.
15 aZxUcY : ëskillí, in Pindar often (but not always) the poetís: see Gladigow 1965 esp.
39-55, Gianotti 1975: 88-107, and F7.3 n. Given the content of 7-12 and the touchstone
metaphor at 12f., it seems natural to assume that here too poetic capability is meant,
although one cannot exclude that the referent is rather the audienceís literary taste applied
to evaluating the song.
\OU]OTcY : passive (ëis extolledí) or middle (ëachievesí)? Snellís supplement and CannatØ
Fera 1995: 421 imply the former, citing Pindarís usage at N. 8.40f. R–XWfRb kí �[Wfn Ö
��Q� ]gdg~] �Qk[�Q �W[pW~]í, I. 1.64f. WÃdiQTQ cfW[�SW]]bQ �W[pYQfí �SVRR~] £
HbW[^kTQ, and (probably) fr. 227.2f. Sn.-M. Vn_cWb kÓ Z[¦QTb £ ¯[SR _Wfí R∞pY[í
�W[pYQfR (Boeckh 1811-21: II/2 678: R∞pY[R VR_cWPpYQfR codd. at Clem.Al. Strom.
4.7.49.1). Conversely, the middle ñ presumably with ëfameí, or some such noun, as an
object ñ could be supported by O. 9.10 (�j[Tfq[bgQ êVbkg]), I. 6.60f. (Q^jR]), B. 2.5
(Q^jRQ), and closer to our passage CEG 785.2 (Q^jW] hR�[ÙQ Ö jtkg]), A.R. 1.466f.
(cW[bi]bgQ ³VVTQ £ jtkg]).
dXO�[aT : a part, or a compound, of cVW~]fg]: Pindar uses cVYTQ not cVW^TQ, and
navigation and the Pleiads are (I hope) out of the question. Van Groningenís cVW�[]fnjb(])
is so far unattested in Pindar, as is Snellís adverbial cVW~¥[]f(R).
205
F3
Introduction
Below and to the left of the conspicuous asteriskos, koronis and paragraphos that mark the
end of F2, H29 introduces F3 with the marginal title o[∞]SbQqfRb] | W∞]� o∞Rj¦Q, identical
to that plausibly reconstructed for F5.1
Aeginetan individuals had the lionís share in commissioning Pindar for victory odes,
with eleven extant in their entirety (Olympian 8, Pythian 8, Nemeans 3-8, Isthmians 5, 6
and 8) and at least one fragmentary (Isthmian 9):2 this accounts for over a quarter of
Pindarís known epinician output. Evidence for cultic commissions is more limited, with
significantly fewer extant or known poems than Pindar composed for his fatherland
Thebes, but at least another Aeginetan Prosodion to Aeacus � or rather, another poem so
interpreted in the Alexandrian edition � partly survives (F5),3 and Pausanias cites an Ãb]_R
to Aphaia that Pindar likewise composed for the Aeginetans (fr. 89b Sn.-M.).4
1 See pp. 227f. below. On the title-pattern of the Prosodia see pp. 16f. above. As the essay on F3 in Rutherford 2001: 410-18 is largely identical to his earlier treatment of the fragment in Rutherford 1992: 62-8, I cite the former only when it offers new material, and only the latter when they coincide. 2 On Pindar and Aegina see Hornblower 2004: 207-35, 2007, Fearn 2011a, with the respective bibliographies; on the fragmentary Aeginetan Isthmians, DíAlessio 2012: 31-5. 3 See pp. 228-44 below. If H26 fr. 87 (*F4 below) does not contain the beginning of F5 after all, then its title is evidence of one more Aeginetan ode (Paean, Prosodion, or Hymn: see pp. 13f., 130 above). 4 Paus. 2.30.3. Its place in the Alexandrian edition is unknown, see p. 11 above.
206
Aeacus is known as a son of Zeus as early as the Iliad; the Ehoiai add that his mother
was Aegina, one of the daughters of the river Asopus and eponymous of the island.5
Although not autochthonous in a strict sense, Aeacus was in some versions of the legend
the first man on Aegina, and the islandís human population was created by Zeus at his
bidding, either ex nihilo or from a colony of ants, whence the name ÙP[_bk¦QW].6 Aeacusí
special relationship with his father was also at the core of Aeginaís claim to mythical pan-
Hellenic relevance: when all of Greece was oppressed by a drought (or a famine in some
sources), the Greek cities sent envoys to Aegina to entreat Aeacus to pray to Zeus for their
deliverance; his prayer on the islandís highest peak achieved the desired goal, and there he
founded the sanctuary of Zeus Hellanios to commemorate the event.7 Nothing is known
from myth concerning Aeacusí death, but in Athens he is connected to the underworld as
early as the turn of the fifth century, as either its judge or its gate-keeper.8
Despite Aeacusí status as Aeginaís main hero, little is known concerning his cult. He
was worshipped with an Aiakeion, which Pausanias describes as a ìquadrangular enclosure
of white stoneî situated ìin a most conspicuous place in the cityî (2.29.6); the exact
location of the shrine has not yet been established with certainty.9 The Greek supplication
to Aeacus at the time of the great drought was portrayed in relief by the entrance, and the
5 Zeus: Il. 21.189. Aegina: [Hes.] Cat. fr. 205 M.-W. For ancient sources on characters and episodes of Aeginetan mythology see the exhaustive collection in Weilhartner 2010, easily consulted through its indexes (363-84); on the Aeacid saga see also Zunker 1988 (58-62 on Aegina and Zeus, 63-89 on Aeacus). 6 Zunker 1988: 64-7, Weilhartner 2010: index s.vv. Myrmidonen, Zeus. 7 For the sources see | F5.3 n., noting expecially Isocr. 9.15 (ìa temple communal to the Greeks just where he had made his prayerî); see also Zunker 1988: 67-9, 72-8. 8 See Rein 1906, Zunker 1988: 86-8, Rutherford 2001: 416 n. 15. 9 The remains of a building once thought to be the Aiakeion (e.g. Welter 1938: 39 pl. 36) near the temple of Apollo on Kolonna Hill have now been identified as the temple of Artemis mentioned by Paus. 2.30.1. Walter in Alt-'gina I/1 (1974) 6, 1993: 54-6 and pl. 48 tentatively situates the shrine on another hill closer to the centre of the modern town, a more plausible area given Pausaniasí itinerary from his landing-place in the commercial harbour to the north-west and Kolonna Hill (2.29.6-30.2). See also Zunker 1988: 69-72.
207
low altar in the edifice was said ìas a secretî to be the heroís _Qv_R (2.29.7f.). The shrine
maintained concrete links to supplication into historical times: still in the late fourth
century it was known as a place of asylum, where Demosthenes briefly took refuge before
fleeing to his death in Calauria and Hyperides vainly sought safety from a capital
sentence.10 The scholia to Pindar mention games called Aiakeia,11 which it is hard not to
consider as part of a wider festival in the heroís honour; one may speculate that the
presence of athletes from (ideally) all over Greece at the Aiakeia was construed as a re-
enactment of either the pan-Hellenic supplication embassy or the service that ìthe prime of
the heroes that dwelt aroundî willingly underwent under his command according to
Nemean 8.12 (or both).12 Assuming that Aristophanes of Byzantium was correct in his
inference that the poem was dedicated to Aeacus, the Aiakeia are, from what we know, the
most natural home for a Prosodion in his honour, although the existence of different
celebrations of the hero cannot be excluded.13
Aeacus and his immediate descendants were a concrete presence in Aeginaís
diplomatic landscape. According to Herodotus, when sometime in the late sixth century the
Thebans asked the Aeginetans for help against Athens, they responded by ìsending the
Aeacidsî (fgˆ] o∞Rj^kR] ]P_cY_cWbQ), a kind of help which could be called a ]P__RZ^R
(5.80.2-81.1). Much the same, and with greater success, happened at Salamis in 480:
immediately before the battle, after a tremor had shaken the island and the sea around it,
the Aeacids were formally called on by the Greek fleet as allies (again ]P__nZgP]) and a
10 Demosthenes: [Plu.] Mor. 846e (Lives of the Ten Orators). Hyperides et al.: Plu. Dem. 28.2. 11 | O. 7.156c, 13.155, N. 5.78c pp. I 233, 386, III 97 Dr.; see further below, pp. 253-5. 12 See also Fearn 2011a: 184-6. 13 Rutherford 1992: 63, Fearn 2011a: 181-90.
208
ship was sent to Aegina to fetch ìAeacus and the other Aeacidsî (o∞RjÙQ jRÚ fgˆ]
³VVgP] o∞Rj^kR], 8.64.2), with the battle starting upon her return and possibly � the
Aeginetan claim � initiated by her (8.83.2, 84.2).14
There is no firm evidence for a cult of the Aeacids on Aegina, but their role in her
international relations intimates that some form of domestic cult must have existed too; the
same is suggested by prayers to the Aeacids in Pindarís Aeginetan odes (F5.55-9, P. 8.98-
100), although these do not strictly require a formal cult. It is hard to tell if the ìwell-
fenced precinct of the Aeacidsî of Olympian 13.109 indicates the entire city or a specific
14 See also Plu. Them. 15.1, Philostr. Her. 53.16. Erskine 2001: 62-8 sees the dispatching of the Aeacids as an element of Aeginaís appropriation of the emerging parallel between the Trojan War and the Persian Wars, but Hdt. 5.79-81 (cited ibid. 63 n. 6), if genuine, indicates that the Aeacids could be so conscripted earlier than, and independently of, such comparison. There is some debate over what these ëAeacidsí actually were: images, relics, or other. Scholarly opinion seems by and large to favour the idea of images (presumably statues: Mueller 1817: 115, 123); Nagy 1990: 177f. suggests that they were ìthe bones of Aiakos, possibly accompanied by living representatives of the current lineage of Aiakidaiî, but, considering Greek practices concerning hero-burials, a high frequency of transfers of heroic relics would be quite surprising. (On such transfers � amply attested, but each as a unique occurrence � see Blomart 2004, Coppola 2008, Neri 2010.) Nagyís more recent suggestion (2011: 77) that the Aeacids were ìan ensemble of contemporary Aeginetan aristocrats who were re-enacting, in stylized choral poses, the presence of their notional ancestors, the Aiakidaiî is not notably more attractive: if nothing else, the Thebansí sarcasm at Hdt. 5.81.1 would be pointless if these Aeacids were men. Contrary to Nagyís continuing assumption of an ìexclusive lineage who trace[d] themselves back to the hero Aiakosî (1990: 178 n. 136), there is virtually no testimony of an Aeginetan individual or clan staking such a claim: the one exception of | P. 5.78d p. III 97 Dr. (| P. 3.113a p. III 97 Dr. is less clear), where kinship with the Aeacids is said to apply either to the victorsí family or to the Aeginetans as a whole, is only an interpretation (and not a particularly attractive one, pace Pfeijffer 1999: 172) of Pindarís ambiguous text. While it is possible that Pindar calls ìAeacidsî the ìpeople of Aeginaî (so Slater 1969 s.v. o∞Rj^kR] C, albeit none of the passages he quotes is beyond doubt), thus expressing a notional ancestry for the entire nation in the same way as the Athenians can be called æWj[gc^kRb, an individualís claim to actual descent from Aeacus would be a different business altogether. No source mentions an Aeacidís issue on Aegina, and Paus. 2.29.2 specifies that Aeacus was the only king that the island ever had; the amount of claimant Aeacids elsewhere in the Greek world � from the ruling dynasties of Cypriot Salamis and Epirus to the Philiads of Athens � only highlights the contrast. Indeed, precisely the absence of a local Aeacid blood-line in historical times may have contributed to the clanís appeal as a pan-Aeginetan myth, one that could bind together all of Aeginaís possibly competing clans (on which see Fearn 2011a, esp. 211-26).
209
shrine;15 if the latter, the context � the mention of a victory gained (presumably) at the
Aiakeia � arguably points towards its identity with the Aiakeion.16
A marginal paragraphos under line 8 shows that the strophe had eight lines; it is
unknown whether an epode followed the antistrophe whose beginning survives at lines 9f.,
or the ode was monostrophic. The former gains some advantage from the presence of the
paragraphos alone, since H4 separates the stanzas of the monostrophic Paean 5 = D5 R.
with koronides as well as paragraphoi, but Hephaestion notes the use of the paragraphos
alone for articulating monostrophic compositions, and the use of metrical markers is
sometimes inconsistent across different papyri.17
The ode opens with a deictic reference to ìthisî day, that in which the performance is
taking place (1). Then present ritual action of some sort is narrated, with Poseidonís
immortal mares leading o∞Rj[ (2f.) � a lacuna whose consequences for the interpretation
of the passage are dire, as will be discussed below. It is unclear if the action described is
envisaged as taking place ad oculos in the ceremony itself, perhaps also represented by a
procession of effigies,18 or on the properly divine plane only: the opening of Partheneion 2
parallels the former, that of Dithyramb 2 the latter. The outline of what follows is similarly
blurred. Lines 5-8 seem to form one sentence: Zeus (cRfq[ Ö æ[gQ^TQ, 5) is the subject,
15 | 156b p. I 386 Dr. sees a reference to the entire city. The context (106-12) is indeed a list of cities, but the specific mention of the �T_¦] of Apollo at 108 allows for a specific reference of o∞Rjbk¬Q fí WÃW[jÓ] ³V]g] too; these are the only two elements in the list in which the host city is not named explicitly. 16 Identity assumed by Toepffer in RE I (1893) 921 s.v. 17 Heph. HW[Ú ]\_W^TQ p. 73 Consbruch, cp. metr. b P. 6 p. II 192 Dr. (not in Tessier) which says explicitly kbÙ gÃkÓ jg[TQ^]b Z[�QfRb �cí RÃf�Q kbÏ _Y]gP. Hephaestionís remark could be squared with the practice of H4 if one referred it only to poems found in wholly monostrophic books such as the works of Sappho or Alcaeus rather than to all monostrophic compositions, but the Pindaric scholiast all but contradicts this view. One simply must not press oneís craving for consistency too far. The likelihood of processional performance is no evidence either way, see p. 115 above. 18 So Rutherford 1992: 67, with a detailed hypothesis on the characteristics and itinerary of the procession.
210
and gazes at someone or something, possibly with a gesture of his hand (6), but the loss of
the main verb does not enable the modern reader to determine if he is welcoming someone,
or being welcomed himself, ìto the immortal table of the gods Ö where for him nectar is
poured to drinkî (7f.). If the events described are envisaged as occurring in the rite, a
theoxeny � a ceremony in which one or more gods are ritually welcomed and feasted � is
an attractive scenario,19 with Zeus either coming inviting someone to come to the banquet
if _gV[ (5) is a part of ¯_gVgQ. At the start of the antistrophe a second indication of time
picks up the one that opened the strophe; the absence of an intelligible context precludes
certainty as to whether it too referred to the present performance, thus describing it as a
periodically recurring event.
The interpretation of the strophe largely hinges on the lacuna at the end of line 3, for
which several supplements can be suggested. If Ian Rutherford is right to believe that an
object, direct or indirect, is the likeliest option,20 Poseidonís mares can be conveying (i)
o∞Rj[¦Q, (ii) someone o∞Rj[�b, (iii) o∞Rj[^kRQ, (iv) someone o∞Rj[^kRb, (v) o∞Rj[^kR],
or (vi) someone o∞Rj[^kRb]. (vi) is unparalleled, and thus comparatively unlikely; for and
against each of the others a case can be made, as follows.
(i) � perhaps the most obvious supplement, given the title � can be related to two
known mythical episodes. One is Aeacusí return to Aegina on Poseidonís chariot after
building the walls of Troy with him and Apollo, attested by Olympian 8.48-51 (composed
for a victory in 460) in terms strikingly similar to F3.2f.:
19 Rutherford 1992: 64f., Fearn 2007: 89f. 20 1992: 63.
211
∆[]gf[^RbQR kí �cí ∏]p_�b cgQf^Rb â[_R pgÙQ fnQPWQ, �cgcY_cTQ o∞Rj¦Q kWt[í �Qí µccgb] Z[P]YRb] |||
The other is Aeacusí supplication to Zeus on Mount Panhellenios when Greece was
beset by, according to different sources, either famine or a drought.21 A reference to this
myth would explain nicely Zeusí prominence in the second half of the stanza.
While the first event is not known to have been linked to Aeginetan cult, nothing
precludes such a link,22 and the fleeting reference to a seemingly irrelevant detail does
suggest a broader local relevance of the episode; Didymusí claim that Aeacusí
involvement with building the walls of Troy is not attested before Pindar naturally rests on
written sources available to him and neglects the possibility of local folk-lore.23
Aeacusí supplication, on the other hand, was unquestionably central to the Aeginetan
cult of Aeacus, so much that the Greek envoys who supplicated him to intercede on that
occasion were the subject of reliefs on the walls of the Aiakeion, as we saw. If our ode was
connected with this episode, the connexion with the reliefs would provide a fascinating
picture of the interplay of visual and poetic art in the Aeginetan cult of Aeacus: the reliefs
and F3 would narrate the two stages of the story, and the effect will have been even more
forceful if the ode was sung in procession from the Aiakeion itself � where Aeacusí
worshippers in flesh and blood doubled, as it were, those portrayed in the reliefs � to the
temple of Zeus Hellanios along the same route as the mythical procession.24 A similar
21 For the sources see | F5.3 n. 22 Rutherford 1992: 66, Fearn 2011a: 181-3. 23 Fr. 16 Braswell in | O. 8.41a p. I 247 Dr.; see Carnes 1995: 24 n. 44 and Braswell ad loc. 24 Rutherford 1992: 66.
212
intimation of the cross-mirroring between present worship and mythical visitation of
Aeacus is found in Nemean 8.8-16.25 No source links Poseidon or his horses to this
supplication, but given the parallel of Olympian 8.48-51 (which on this interpretation
would recall F3, likelier than vice-versa) Aeacus may have been depicted as using
Poseidonís chariot also for his most important feat. Casting Aeacus as the protagonist
would also give special significance to cRfU[ Ö æ[gQ^TQ (5), who was his father, but the
frequent general sense of the apposition does not particularly require this solution.26
Neither scenario, however, is free from problems. Perhaps because only known from
one fleeting reference, the first leaves Nereus and Zeus unexplained; Nereus has no place
in the second episode either, as far as all other sources go. Rutherford argues that his
presence can ìbe explained as a symbol of Aeginetan sea-powerî,27 but the transferred use
of Nereus is not found before Callimachus, and even then it has a metonymic value (ëseaí)
that is plainly impossible here.28 Also curious, if not conclusive, is the reference to the
godís immortal table (7) and nectar, if Lobelís supplement is correct (8): not only a
meeting between Aeacus and Zeus atop Mount Hellanion, but a kind of reversed theoxeny,
with the god(s) feasting the mortal rather than the opposite. Rutherford alternatively
suggests that Aeacusí journey may be his periodic return from his place as judge of the
Underworld, but this does not remove the objections concerning Nereus, Poseidon, and
Zeus; moreover, no source mentions his (arguably counterintuitive) periodic return to
25 Rutherford ibid. n. 32 connects the N. 8 passage to the reliefs, but Carnes 1995: 19f. is right to argue that it does not refer to the embassy myth. 26 See Lobel 1961c: 29 and n. 27 1992: 65; on Aeginaís ostentatious claim to seamanship and sea-power see pp. 257-60 below. 28 Call. Jov. 40, later examples in McLennan 1977: 72.
213
earth, and nothing in our relatively ample Aeginetan evidence suggests that his
otherworldly role had any currency in local mythology.29
With (ii), the direct object to be supplied for ³SgQf(b) is presumably either of
Aeacusí partners, his mortal wife Endaís or the Nereid Psamathe(ia).30 Like (i), both
options would tie in nicely with the odeís title. Psamatheiaís presence would also nicely
explain that of her father Nereusí (4);31 however, several sources explicitly oppose Aeacusí
rather casual mating with her to his marriage to Endaís,32 adding the detail of Psamatheiaís
metamorphosis into a seal kbÏ fÙ _U �g�VW]pRb ]PQWVpW~Q, and Pindar himself may hint to
a similar tradition by stating in Nemean 5.13 that (unsurprisingly in view of her name,
ìSandyî) she gave birth �cÚ 5\S_~Qb c¦QfgP. All this makes a proper wedding and
wedding procession, with none other than Zeus in attendance (5), somewhat unlikely.
One argument for seeing a reference to Aeacusí wedding to Endaís could be
Bacchylides 13.96-9, where Aeginetan maidens sing
{Q]ÉRÍkR fW 5gk¦[cRZPQ, Û fÙ¥[Q ∞]]¦¥[pW]×Q ¯fb[jfW H\VYR jRÚ SWVR�[�]�R [jg]��[]fnQ o∞Rj�b _�W�bZpW~]í �Q WÃ¥[Q¬b.
Commentators have argued that this passage recalls actual cult practice at the Aiakeion in
the context of the Aiakeia.33 Depending on the restoration of 94 (fYjg]] Housman, VYZg]]
29 Rutherford 1992: 66f., Carnes 1995: 39 n. 93. 30 N. 5.13 with | 12a, 21a pp. III 90-2 Dr., Philosteph. fr. 35 FHG III 33, | E. Andr. 687 p. II 295 Schwartz, Apollod. 3.12.6, etc., see Zunker 1988: 83-5. Snellís oÒSbQRQ (the heroís mother!) is inexplicable (DíAlessio 1991: 103): Men. Rhet. p. 409 Russell-Wilson, who ostensibly supports this view, must be confused. 31 Bona 1988: 267. 32 So do Philostephanus, Apollodorus, and the Euripides scholion. 33 Power 2000: 80f., Fearn 2007: 116-20.
214
Barrett),34 the union of Aeacus and Endaís is either the only subject of the maidensí song,
or the second after Aeginaís union with Zeus; in either case, the parallel with this
interpretation of F3 is almost as striking as that with F5.35 One objection, if not a
particularly forceful one, against this scenario is the unlikelihood of a mortal bride being
conveyed to her wedding by nothing less than Poseidonís ìimmortalî team.36 Nereusí
presence is perplexing, but the same solution could be proposed as for (i), however
questionably. That of Zeus � usually no great traveller unless impelled by feminine charms
� is also striking, but he does come down to another son in person at Nemean 10.79,
although in rather extreme circumstances.
Both (iii) and (iv) suggest one scenario: the wedding of Peleus and Thetis.37 Between
Homer, whose o∞Rj^k\] is Achilles in all but three instances, and the Hellenistic age, the
patronymic is mostly applied to Peleus.38 If it applies to him here too, his wedding is
arguably the likeliest occasion for the chariot-ride of lines 2f. to have taken place. The
wedding is depicted by several vases, four of which show the couple standing on, or
stepping onto, a chariot.39 It is a suitable environment for Nereus (4), the brideís father: a
mid-sixth-century amphora depicts him bidding farewell to the couple, who stand on their
chariot ready to leave, and on the slightly earlier François Vase his wife and he follow the
34 Housman 1898: 140, Barrett ap. Maehler 1982: 267. 35 See also p. 84 above. 36 This also tells against any supposition that it is the Greek envoys at the time of the drought who are said to be coming to Aeacus; and they will hardly have come to Aegina by chariot anyway. 37 Pontani 1976: 409, Danielewicz in Rutherford 1992: 63 n. 20. On the saga see Zunker 1988: 104-20. 38 o∞Rj^k\] not referring to Achilles in Homer: Il. 16.15, 18.433, 21.189 (in all three cases significantly referring to Peleus). Peleus o∞Rj^k\] after Homer: Hes. Cat. fr. 211.3, 7 M.-W., Alc. fr. 42.5 V., P. 3.87, I. 8.39, E. Andr. 789, IA 1045, etc.; his brother Telamon only once, I. 6.35. See West 1985a: 162. 39 On the wedding of Peleus and Thetis on vases see Brommer 19733: 318-20, Zunker 1988: 116-20, Vollkommer in LIMC VII/1 s.v. Peleus VIII.I, VII/2 pll. 198-212. Chariot: ABV 260.30, Athens NM 12076, ARV2 17.18, 1038.1.
215
chariot-procession towards the wedding on foot.40 Poseidon, if his involvement is rightly
inferred from line 3, and Zeus (5) are also frequently depicted.41 The presence of the gods
at the wedding of Peleus and Thetis is known as early as the Iliad and the Cypria, and
recurs time and again in Greek (and Latin) poetry.42 On that occasion, Poseidon gave
Peleus two immortal steeds, Balius and Xanthus, as his wedding gift.43 One also notes that
in Quintus of Smyrna, Apollo is said to have celebrated the wedding by singing ≈cT]
rYfbQ �[SP[¦cW�RQ | H\VWˆ] ¶SWfí ³jgbfbQ ÚVÙ] _YSR VR~f_R Vbcgt]RQ (3.101f.),
which would have made a useful mythical template for subsequent celebrations in which
the same story was sung.
Between (iii) and (iv), the former is the likelier. Consistently with Greek nuptial
practices, the vases variously portray both Thetis and Peleus on the chariot, her on the
chariot and him stepping on to it, or him helping her on to it from the ground:44 this
40 Athens NM 12076; ABV 76.1. 41 Zeus and Poseidon: ABV 76.1 (the François Vase), 39.15 (the so-called Sophilus Vase). Poseidon: ABV 260.30. Zeus: ARV2 17.18. 42 Il. 24.62f., Cypr. test. ap. Procl. fr. b 86f. Severyns, fr. 3 BernabC, Alc. fr. 42.6 V. (cnQfR] Ö _n½[R[R]), P. 3.86-95, N. 4.66-8, E. IA 707, AR 4.807-9, Catull. 64.298-383, QS 3.99f., 4.131-43, etc., see Stoneman 1977. Several of our sources mention divine song taking place at the wedding: P. 3 (the Muses), N. 5.22-5 (the Muses and Apollo), Euripides (the Muses and Apollo), Pl. Rep. 383a-b citing A. fr. 350 Radt (Apollo), Isocr. 9.16 (the gods), Catullus (the Fates), Quintus (Apollo, the Muses), Apollod. 3.13.5 (the gods); compare also the kithara-holding Apollo on ABV 39.15, 260.30, ARV2 1038.1, and the Muses on ABV 39.15, 76.1. Isocrates explicitly calls the song a Õ_YQRbg], and relates the tradition (dR]^) that the feast was the only occasion on which the gods sang at a mortalís wedding. 43 See 2 n. 44 Peleus and Thetis on the chariot: ABV 260.30, Athens NM 12076. Thetis on the chariot and Peleus stepping onto it: ARV2 1038.1. Peleus helping her onto the chariot: ARV2 17.18. On Greek nuptial practices see: Sticotti 1898, Oakley and Sinos 1993: 26-32, 44f., pll. 62-81, Laxander 2000: 59-72, pll. 30-9.
216
suggests e.g. o∞Rj[^kRQ jRÚ rYfbQ more than Snellís o∞Rj[^kRb rYfbQ, which would have
the bridegroom wait for the bride to arrive, a Roman rather than a Greek practice.45
Rutherford rejected the idea of a reference to the wedding of Peleus and Thetis
ìbecause that took place on Pelionî.46 Indeed, when a rite is construed as celebrating a
specific mythic event, it usually takes place in what is held to be the same location(s): the
Plataean Daedala, the Delphic Septeria, the Theban Daphnephoria, and many more.47
However, there are counter-examples. For instance, the hieros gamos that took place at
Athens on the second night of the Anthesteria between the wife of the archon basileus and
ëDionysusí was envisaged as the re-enactment of Dionysusí union with Ariadne, whom
Theseus (the archonís predecessor) had left behind for the god: the mythical event took
place on Naxos, but the ceremony took place in a building in the Athenian market-square.48
Especially considering how much of the Aeacid saga is Thessalian-related myth grafted on
to Aegina,49 such long-distance connexions in Aeginetan cult should not be too surprising.
Corroboration, however tenuous, may come from Nemean 5, an ode which links itself to
the Aiakeion and the Aiakeia and telescopes a narrative of Peleusí merits into his wedding
with Thetis (22-6).50 QY¥[jfR[, if rightly supplemented at line 8, is no obstacle: already in
the Iliad Thetis uses liberally of the substance, and Quintus mentions libations of nectar at
45 Roman custom of the bridegroom waiting for the bride: Oakley and Sinos 27 and n. 30. E.g. o∞Rj[^kRQ rYfbkb would be equally exceptional, see ibid. 27f. and n. 31 (where the first reference should presumably be to E. Cyc. 513-15, quite irrelevant in any case). 46 1992: 63. Pindar himself locates the episode on Pelion at P. 3.89f. and N. 5.23f.; the latter is an Aeginetan ode, which makes the hypothesis of a different local version untenable. 47 On the Daphnephoria see specifically Kurke 2007: 80-3, 97-9. 48 [Dem.] 59.73, 76, Arist. Ath. 3.5, Hsch. k 1890 Latte: see Burkert 1985: 109, 239f. 49 See West 1985a: 162-4. 50 On the links between N. 5 and the Aiakeion and Aiakeia see Pfeijffer 1999: 193, Fearn 2007: 115, 119.
217
her wedding; Euripides presents the marriage feast as given by the gods, and Haspels
remarks that on the François and Sophilus vases it is they who bring the ìwineî.51
(v) parallels to some extent the practice of sending ëthe Aeacidsí: a distinct
advantage over (i) and (iii)-(iv), whose myths are not independently attested to have been
related to ritual practice. On the existing evidence this practice is attested only abroad and
on extraordinary occasions, but especially given its recurrence � almost the standard
response to a request for military help, it seems � it is relatively easy, although not risk-
free, to suppose that those extraordinary missions were an extension of a similar, regular
rite that took place on Aegina herself, if the Aeacids as a group did have a cult there. (The
ode can hardly have been composed for such an extraordinary sending, if �QbRPf�b (8)
indicates a periodic ritual.) However, this scenario accounts neither for Nereus nor for
Poseidonís mares, unless Balius and Xanthus are called in to explain the latter (but not the
former); even harder to justify is Zeusí presence at the f[ncW�R.
These are not the only possibilities. Notably, o∞Rj[ may have been a genitive, with a
meaning close to any of the supplements cited (o∞Rj[^kR kn_R[fR, o∞Rj[gt kY_R], or
the like), or possibly part of a circumlocution for Aegina herself or a part of it, if a
prepositional phrase followed;52 under the same conditions it could even be o∞nj[WbgQ. But
all of these are rather more far-fetched, and require imagining a great deal in one lacuna. It
is also possible to think that the procession envisaged by F3 is purely ritual, a theoxeny
with no mythological strings attached,53 but if so, where is Aeacus coming from?
51 Il. 19.38f.; Q.S. 3.108; E. IA 707; Haspels 1930: 431. 52 Cp. O. 13.109 o∞Rjbk¬Q fí WÃW[jÓ] ³V]g], N. 4.11f. o∞Rjbk¬Q £ †�cP[SgQ ãkg], 7.9f. c¦VbQ Ö o∞Rjbk¬Q, fr. 242 Sn.-M. Ú _ÓQ c¦Vb] o∞Rjbk¬Q. 53 However, note QY¥[jfR[: if Lobelís supplement is correct, the feast is primarily divine, not earthly.
218
To conclude: each of (i)-(v) is possible; (iv) has little to recommend it, (v) is only
marginally better, and the objections against both legs of (ii) are probably stronger than
against either (i) or (iii). Of these two, the impression is that (iii) suits the text best, and (i)
what we know of the Aeginetan context. The wealth of literary and visual parallels one
finds for so many aspects of (iii) is unrivalled, but the mere reference to a myth known in
lesser detail should not detract from any credit to be accorded to (i), which several
elements in the text can suit seamlessly. The title suggests (i) over (iii), but is not
incompatible with the latter: nothing precludes the hypothesis that such a reward for noble
Peleus reflected on his father and was celebrated accordingly, if the title draws on accurate
records of the odeís commission and performance; and if it was conjectural, awareness of
an Aeginetan cult of Aeacus but not of Peleus (which is, after all, the same state in which
we find ourselves in the twenty-first century AD) may have warranted the supposition. Of
(i) and (iii), the one the case against which seems weaker is (iii): the celebration of an
event set in a faraway land, although admittedly rare, I regard as slightly less problematic
than casting Nereus as a mere symbol of sea-power in a sort of allegorical pageant. Thus, I
hesitantly submit that the procession depicted by the first stanza of F3 is imagined to be
that of Peleus and Thetisí wedding, although the grounds for disqualifying Aeacusí
processsion up Mount Panhellenios are comparatively slight.
219
Commentary
1 -�YNí n[ �_cTY : Pindarís only extant occurrence of ≈kW referring to time; for similar
determinations of time cp. Pae. 6.5 �Q �RpYTb Ö Z[¦QTb, 61 �Q pW�Q XWQ^Rb, Philod.
Scarph. 4f. CA p. 166 †[bQR~] Ö [fR~]kí] �W[R~] �Q ·[Rb] (suppl. Weil 1895: 400), and
such deictics as QtQ in the specific sense (P. 11.7, Pae. 1.5) or ]n_W[gQ (O. 6.28, P. 4.2,
Theocl. 1 CA p. 173, see Rutherford 1992: 63, 2001: 412). It is probably a stretch to hear
an echo of Il. 18.85 ¶_Rfb f�b ≈fW ]Ó (sc. Thetis) �[gfgt ³QW[g] ¯_�RVgQ WÃQvb.
2 µddZY _Ó[ \fW[cTcY : chariots can be pulled by teams of mares as well as stallions,
and lyric (like tragedy) distinctly favours the former: Barrett 1964: 204f., McDevitt 1994,
Finglass 2007a: 313 (aptly citing | E. Ph. 3 p. I 246 Schwartz). Mares pull Poseidonís
chariot also at O. 1.41 and 8.51; two immortal stallions, Balius and Xanthus, were
Poseidonís wedding gift to Peleus and later pulled Achillesí chariot at Troy, Il. 16.148-54,
19.399f., Apollod. 3.13.5 (from the Cypria?), Tzetz. in Lyc. 178 p. II 88 Scheer, but in
view of their gender (note, however, Alc. fr. 42.14 V. XnQpRQ �Vnf\[[R ciVTQ) and the
slight time warp that would otherwise ensue, it seems inadvisable to see a reference to
them here. �pnQRfRb need not be taken literally: anything that lies within a deityís sphere
of action can be called immortal, see 7 n.
3 GZaOYNl[Za : ìall horses are Poseidonís Ö but presumably here the god himself is
depicted as present with his team, whence �pnQRfRbî (Lobel 1961c: 29, but see 2 n. on the
epithet). Poseidon does have a long-standing association with horses (cp. P. 4.45
220
�ccn[ZgP Hg]WbknTQg] and see Schachermeyr 1950 passim), but the only parallel for
such general reference of a horseís pertinence to Poseidon is O. 5.21 Hg]WbkRQ^gb]bQ
(-Rb]bQ | 48 p. I 151 Dr.) µccgb], and even there the interpretation is a scholiastís guess
based on the notion that the god is fv] Ö �ccbjv] ¯dg[g] and d^Vbccg] (| 46, 48 cit.);
Moschopulus ap. Triclinius p. 196 Abel glosses with �QfÚ fgt �[^]fgb], which is not
clearly inferior. There is little doubt that our mares are Poseidonís in a more concrete
sense, whether or not he is depicted as present at the scene.
�tZ[Tí : the present ãcWfR[b (4) suggests ³SgQf(b) (so Bona 1988: 267) or ³SgQf(Rb) (on
elision of -Rb in Pindar see Hermann 1809: XI), not imperfect ³SgQf(g), pace Pontani
1976: 312; with the mares as subject, the active is inevitable. It is a reasonable inference,
but an inference nonetheless, that the context in which one or more persons are driven in a
chariot in an obviously festive surrounding is that of a wedding: Bona cites fr. *30.1-4 Sn.-
M., where the Moirae ÃSgQ Themis to Olympus Z[P]YRb]bQ µccgb], see also Stesich.
PMGF 187, Sapph. fr. 44.13-17 V.
�∞c�[ : on the possible supplements and their consequences on the interpretation of the
fragment see the Introduction.
4 ��]Oˆa Ní ¡ t`]h[ : in our record it is Hesiod who first calls Nereus the Homeric
(âVbg]) SY[TQ (Il. 1.358, 538, 556, etc., Od. 24.58, cp. P. 9.94): Th. 233f. (note jRVYgP]b),
263, 1003; the Nereids, however, are already such at Il. 18.38, 49, 52. Presumably a
flattering passage for the poor sea-god: this is the only Pindaric occurrence of his name in
which he does more than specify one of his daughters in the genitive, contrast O. 2.29, P.
221
3.92 (W�g�VgP), N. 3.57, I. 8.42, B. 1.8, 17.102, and indeed most occurrences of his name
down to the fifth century and beyond.
�dOTc[Y : used of persons on foot following chariots in Patroclusí funeral procession (Il.
23.132-4), but the meaning need not be so specific here, despite the parallel of the François
Vase. Although sense and metre are arguably complete, it is worth noting that the line need
not have ended here.
5 dcT�] NÓ �]Z[Uh[ : both nouns are standard epithets of Zeus, and the nexus occurs
with some frequency in early hexameter (Od. 8.289, Cypr. fr. 9.5 BernabC, Hes. Op. 259,
Cat. fr. 211.11 M.-W., H.Merc. 323, H.Cer. 396) before being reprised by Pindar (P. 4.23,
N. 10.76) and later epicists. In view of o∞Rj[ (3), one notes that in these examples cRfq[
is often literal, referring to Zeus as a characterís father, but sometimes generic (Cypr. fr. 9,
Hes. Cat. fr. 211, P. 4), see OíSullivan in LfgrE B.II.1-2.
_ZX[ : lead and song-dance seem out of place, so a verb-form from the root of ¯_gVgQ is
the only concrete possibility. Given the present tense that dominates the passage (a string
of historical presents being a rather less attractive interpretation), an unaugmented aorist
indicative is out of the question. The obvious choice is between an aorist participle (Lobel
1961c: 29 plausibly suggests a dative plural, but a singular is possible if o∞Rj[¦Q is the
true reading at 3). So, as Bruno Currie suggests, is an aorist optative _¦V[gb (which
dispenses with the need to accommodate another main verb in the sentence), entreating
Zeus to gaze onto the scene and come to it where he will be well received; on this
interpretation, µQR at 8 would be paralleled by ¯QpR at N. 5.38. A future is also possible,
222
although this form is rare, with only three recorded occurrences BC (the earliest are [A.]
PV 689, S. OC 1742).
6 d]Ùa ƒ__c BcXk[ : in tmesi (Lobel 1961c: 29); either used absolutely as at P. 4.271, if
the object of Zeusí gaze could be inferred from the context, or completed by a dative like
at I. 5.28f., E. Med. 860 , etc. (_gV[gt]b 5, Lobel). Zeus watching over an earthly scene is a
familiar sight in early Greek poetry (see also F11), but if g� (8) refers to him and not e.g. to
Aeacus or Peleus, he is inside the scene, not looking in from outside. If the scene was that
represented also by the reliefs of the Aiakeion (which need not have been the case), a
performance there would have construed the viewersí gaze onto the sculpted figures as
mirroring that of Zeus onto their ërealí counterparts in its mythical archetype: compare the
interrelationship between viewer and sculpture brought out by Osborne 1987: 98-103 on
the Parthenon frieze. On sight and gaze in Pindar see now AdorjUni 2011, esp. 112-17.
^O]U : possibly a gesture of welcome (Rutherford 1992: 66) or enthusiastic salutation, cp.
P. 4.239f.: pictorial evidence for such gestures in Neumann 1965: 41-8, 179-81, Bogen
1969 passim. Lobel 1961c: 29 suggests taking it ìwith some verb meaning ëapproachesí,
say, cWVn�Wb, to be supplied (since l. 21 [6] is prima facie complete) in l. 22 [7]î, but the
addition seems unnecessary.
7 \_B]Z[ : accusative with f[ncW�RQ more probably than genitive with pW�Q, which
would make a rather awkward word-order. For this sort of hypallage cp. F5.18 n., P. 4.11
�pRQnfgP ]f¦_Rfg] (Medeaís, with Braswell 1988: 76), I. 8.30 ³_�[gfgb Ö pW�Q
c[Rc^kW], 45af. SVWdn[gb] Ö �pRQnfgb]bQ (Zeus and Poseidonís), Pae. 9.35 VYZWb Ö
�_�[g]^Tb ÙWV^R]. On hypallage in Pindar see Hummel 1999: 344-8 with bibliography.
223
8 [ q̀[VTc] : ìor some case or derivativeî (Lobel 1961c: 29), but it is most logically the
subject of jYZPfRb: if g� refers to Zeus, there are not many other things that a god can be
poured to drink. However, it could also refer to a mortal to whom divine prerogatives are
being extended; this would give more point to Pindarís insistence on divine attributes in
the passage (2, 7). The traces of the last letter are very upright for epsilon, but cp. e.g. 5 kW.
9 �]^OTcY: the subject is presumably either (i) the feast-day itself, which comes every
year (�QbRPf�b), (ii) a deity who is conceived of as visiting the relevant shrine, or
performing some other ritual journey, at that time, or (iii) the delegation of worshippers
that comes to the shrine on the same occasion. An obvious parallel for (iii) is F11.5-8, cp.
Pae. 6.1-15, 7.1-4 (see Galiano 1950-51: 311, DíAlessio in id. and Ferrari 1988: 170 n.
32), Parth. 2.39. For (ii) one can compare Parth. 2.3 •jW]b SÏ[ ¡ [ÊgX]^R], Alc. fr. 307c
V., B. 16.8-10, Call. Ap. 1-8, P.Oxy. 2636 ii.9-15 (Lobel 1967d: 136; Pindar? but see
Ucciardello 2001), and the whole ëgenreí of the cletic hymn. (i) is supported by several
attestations of ¯[Zg_Rb or a compound said of a point in time: Od. 1.16, 2.107, 7.261, etc.
(¯fg]), 10.175, 17.606 (õ_R[), Il. 2.387, 14.77, etc., Od. 4.429, 574, etc. (Q�X), Semon. fr.
1.7f. W. (ô_Y[\Q, �fYTQ Ö cW[bf[gcn]), S. El. 201f. (Ú_Y[R), E. IA 717 (]WVYQ\] Ö
�QfWVU] j�jVg]), and most closely Pae. 1.5-7, where ¡ cRQfWVU] {QbRPf¦] comes to
Thebes bringing a banquet for Apollo, and Catull. 64.387-9, which evokes Jupiterís
visitation of his temple annua cum festis uenissent sacra diebus (387) (cp. 1 here) in a
contrast between mythical and later time which follows from a narrative of Peleus and
Thetisí wedding.
224
n[YceT�Y : probably ëon the anniversaryí: for this meaning of �QbRPf¦] cp. Leg.Gort. (IC
IV 72) i.36, iv.4f., ix.29 with Buck 1955: 323, 357 (ìat yearís endî Rutherford 1992: 63,
cp. Willetts 1967: 55), CID I 9C.47, probably also Pae. 1.5, Hymn. Curet. 5 etc. with
Furley and Bremer 2001: II 10 (see LSJ s.v., DGE s.v. A.2). The attested senses of the
simple dative of the more common meaning ëyearí without an ordinal or specifications
such as f�b RÃf�b, f�b �Xv], etc. are ëwithin one yearí (Hipp. Int. 10, Nat.Hom. 12, Arist.
Pr. 923b, Thphr. HP 9.8.8), ëby one yearí (Arist. HA 571a, Thphr. CP 3.4.1, 4.3.2, 4.6.7,
Teos 41.18, etc.), ëduring the yearí (IvP II 374B.2, C.2, D.2), or ëin one yearís timeí
(Thphr. HP 3.4.6, 3.16.4, 9.2.6f., Eudox. fr. 124 Lasserre2, etc.), none of which suits our
context ñ especially the present tense ñ particularly well.
The temporal determination forms a neatly balanced pair with f�bkí �Q ³_Rfb (1), opening
each of strophe and antistrophe and gradually specifying the occasion which the ode
commemorates. It also intimates to the god (and to the audience, contemporary as well as
future) the worshipping communityís constant devotion in celebrating the festival year
after year: cp. F5.60 cgVVnjb with n., F11.5 pR_n, *F7.4f. with pp. 313f. below, Pae. 2.98
pR_n, 6.16 pR_bQn, etc. If such an occasion celebrates the wedding of Peleus and Thetis
there may be an echo of Alcae. fr. 42.12 V. W∞] kí �Q^RPfgQ, although there the noun refers
not to a solar year but to the period of gestation (cp. Od. 11.248, Hes. Th. 58 with West
1966: 175f., H.Ap. 343 contra Page 1955: 279f.); any suggestion that such is the meaning
of �QbRPf¦] here (for a verb of going in that context cp. F11.15 ∞¦QfW]) is precluded by
the continuing use of the present tense.
225
10 : the lacuna has obliterated two medium-sized letters constituting one short syllable.
Rutherfordís [Õc]¦ and accusative (to go with ÕcW[fnfRQ) is a good shot; a preposition
might also be in tmesi with a participle, in which case [�c]¦ is another possibility. There is
no guarantee that the preposition has anything to do with ÕcW[fnfRQ, so a supplement
along the syntactical lines of Rutherfordís earlier (and metrically impossible) [�c]Ù
QR£c¬Q could be near the truth. The longum on alpha also tells somewhat against his later
[Õc]Ù ´R£ÍkR, since one would expect it either to disambiguate homographs (cp. F2 ii.15
]gdb�b) or to mark ëDoricí alpha (cp. 10 ÕcW[fnf�Q, F2 ii.12 pW�[^R]); moreover, the
pair with ÕcW[fnfRQ is rather odd. The longum would be fully justified by Ferrariís
conjecture (ìsomeone or something, e.g. the chariot of the Hours, comes bringing to the
annual cycle its highest day, that of the present ceremonyî 1996: 133); however, the
pointed apex at the start of 11 seems too far left to be from mu (cp. F2 ii.8 _TQ, 10
_QR]Wb), although this could account for the speck of ink at mid-height next to it in the
margin (the tip of the right leg). In principle, ÕcW[fnfR Q- is just as plausible, and is
discouraged only by the scarcity of viable supplements for Q[ ¥]gQ or Q[ ¥]gQ�. If the scene
portrays Aeacusí reception by Zeus, ÕcW[fnfRQ [¡k]¦Q may be found attractive: beside
the obvious relevance to a mountain path, ÕcY[fRfg] in a divine context occurs at O.
2.77, 4.1 (cp. fr. 169a.4 Sn.-M.), and the epithet ŸcRfg] is applied to Zeus worshipped
on mountain-tops (Parker 1996: 30f. n. 6); for the resulting construction cp. P. 8.41 ¶VPpgQ
£ kWPfY[RQ ¡k¦Q. It seems unlikely that the final alpha ended a word; the regular word-end
marked by Snell in the metrical scheme (19593-643: 56, retained in Sn.-M.) is somewhat
incautious.
227
F5
Introduction
The status of F5 is arguably the most peculiar of the entire Pindaric corpus. Preserved by
H4 and, to a lesser extent, by H5 (whose contribution is essential in the second half of the
piece) as a part of the book of the Paeans in a virtually identical metre to that of the main
body of Paean 6, it was unanimously considered as the third and final triad of that ode for
almost ninety years after its publication; however, its unperspicuous thematic relation to
the preceding two triads, the coherence of the ode as a whole, and the identity of its
original performers (the Delphians according to the title of Paean 6) were the subject of
substantial debate.1
The terms of the question changed dramatically in 1997: Ian Rutherford and
Giambattista DíAlessio used the previously unplaced fr. 108 of H4 to complete a scholium
to the right of F5.2 (H4 col. xxx), which was thus understood to say that the portion of text
it referred to was ëtransmitted in the first book of the Prosodiaí.2 The title and asteriskos to
1 ìThe main problem of Pindarís sixth paean concerns its unityî (opening sentence of Hoekstra 1962): see also Bowra 1933: 44f., Radt 1958: 88-90, Bona 1988: 100f. Farnell 1930-2: I 313, II 408 went so far as to argue that the third triad was a later addition to an original two-triad composition, a view which met with little approval throughout the twentieth century (but see Bowra 1933: 45, Des Places 1949: 24f.) until it was revived by Rutherford 1997: 16-18. 2 | 2 = | Pae. 6.124 p. 304 R. with DíAlessio and Rutherford in Rutherford 1997: 6-8 (parallels for dY[WfRb used of textual transmission conveniently rehearsed ibid. 9-12). The scholiumís slightly surprising position next to the second, rather than the first, line of the piece is arguably explained by alignment with È[g][¦]ÉbgQ
228
the left of the column, which seemingly mark the end of the Delphic paean and the begin-
ning of a ìprosodion for the Aeginetans to Aeacusî, were also first recognised on that
occasion.3 It was thus possible to argue for the identification of H7, whose fragments 16
and 15 contribute to the text of F5 with a different colometry from that of H4 and H5, as a
manuscript of the Prosodia, which triggered the identification of the other papyri of that
book.4 It was again DíAlessio who argued for recognising a fraction of the title and first
line of F5 on H26 fr. 86.6f., following *F4(d).5
The new evidence raises a number of questions on the status of F5 and of the ëwholeí
Paean 6, which scholarship is still struggling to answer. They are only tangential to the
purpose of this dissertation, as they do not impact on the notion that, in the book of the
Prosodia, F5 was (as far as the present evidence goes) an independent ode. They do,
however, contribute useful material towards the history of the ode and its interpretation.
Consequently, it is to these questions that the first few paragraphs of this introduction will
be devoted; readers who wish to proceed to a discussion of F5 in its own right are advised
to skip to page 244 without further inconvenience.
in the title to the left of the column (see below); a scholion to the relevant portion of the title, as it were, like inscrr. P. 2, 10 pp. II 31, 241 Dr., etc. 3 Rutherford 1997: 3-6. The title is followed by three erased lines, probably written by a different hand; their content is now for the most part illegible (Grenfell and Hunt 1908: 46f.). Currie 2005: 324 n. 155 (after Rutherford 1997: 6) plausibly reads a form of c[g]¦kbgQ in its last line, perhaps implying discussion of the pieceís status by some scholar, and suggests that l. 2 conceals an abbreviation for Aristarchus or Aristodemus, who may have treated the status of F5 in connexion to the putative relationship between Pae. 6 and N. 7 (cp. | N. 7.48, 103 pp. III 126, 137 Dr.); a possible alternative is Aristonicus, whose name recurs elsewhere in the Pindar papyri (see p. 123 above) and who dealt with the other certain case of double transmission known to us in lyric, Alcm. PMGF 3 (see n. 24 below). The train of thought, however, is irrecoverable, and the reason for the erasure itself remains a mystery. It would make sense if it was the title of F5, mistakenly written too far down the column and subsequently rectified, but it does not seem to me to be what was written there. 4 Overlap noted by Snell 1938: 431 after Hunt 1922b: 97 (fr. 16) and Lobel 1961b: 17 (fr. *15), conclusion drawn by DíAlessio 1997: 27, 37f.; see p. 12 n. 16 above. 5 DíAlessio 1997: 37 n. 92; see further p. 14 above.
229
The fundamental question is whether Paean 6 was an organic three-triad composition
from which F5 somehow became detached, or conversely whether there were two
independent poems which were at one point rolled together into one: that is, whether the
F5 that we read in the Prosodia is actually a poem on its own right, or a mere excerpt from
a larger whole. The necessary starting point is a reconsideration of our sole manuscript
witness for this question, H4. After this, the argument will be twofold: on the one hand
regarding the history of the text and of the critical activity that shaped it, so as to evaluate
competing hypotheses on the formation of the evidence in its present state; on the other
hand, close scrutiny of the texts themselves, both the ëwholeí Paean 6 and the separate
two-triad Paean � which for the sake of economy I shall refer to as Paean 6(a) � and F5,
so as to assess whether and how far they support or contradict possible views on their
original unity or independence.
As far as the manuscript evidence is concerned, one fact is of great import: namely,
that not only the presence of a title beside the opening of F5 distinctly suggests that it was
understood as the beginning of a new ode, but also that the presence of an asteriskos as
well as a koronis (and presumably a paragraphos, lost in lacuna) between Paean 6(a) and
F5 means that, in an editorís judgement, a poem ended there.6 The signs that mark the
6 Rutherford 1997: 4, 19, 2001: 329f., DíAlessio 1997: 57, Currie 2005: 324. The asteriskos is not visible in its entirety, but its presence is virtually certain: its one surviving stroke had been understood by Grenfell and Hunt 1908: 46 as a part of the clearly recognisable koronis to its right, but its shape and position greatly favour the hypothesis that it belonged to an asteriskos (Rutherford 1997: 4 with drawing).
The use of metrical markers in triadic lyric is described by Heph. pp. 73f. Consbruch: ô cR[nS[Rdg] �cÚ _ÓQ f�b fYVWb fv] ]f[gdv] jRÚ �Qfb]f[¦dgP jW~fRb, �cÚ kÓ fvb �cTbk�b ô jg[TQ^] Ö �cÚ _YQfgb f�b fYVWb ¡ �]fW[^]jg] f^pWfRb, SQi[b]_R fgt fWfWVY]pRb fÙ Ãb]_R. Not all papyri of Pindar follow this system. An asteriskos marks poem-end in H7 (frr. 8, 47, 84), H10 (see Lehnus 1984: 65, 78), H29 (frr. 1, 3), of course H4 itself (col. xxii), and H9 (fr. 1, mistaken for the first line of the title by all the editors of the papyrus to date). H26 must have used only the koronis, since there is no asteriskos next to it in fr. 14(a).ii (end of Pae. 7b; frr. 23 and 65 are indecisive); it seems unlikely to me that there was an asteriskos in H11 fr. 3, and practically impossible that there was one below the title of P. 11 in P.Oxy. 5042. H32 is less clear: there is an
230
break between Paean 6(a) and F5 are the same as separate Paeans 5 and 6(a); with all
likelihood, here too an ancient reader of H4 would have seen two odes, not one.7 Crucially,
then, if Paean 6 was ever regarded as one three-triad poem before AD 1908, this has left
no trace in our manuscript evidence. This should have imparted a more radical shift in our
perspective than it has so far: the unity of the poem is now shown to be a modern
construction, albeit not deliberately such, and ought to be critically evaluated accordingly.
The question first arises whether H4 here faithfully represents the book of the Paeans
in its standard shape. No definitive answer can be given: it is our only witness for this
portion of the text, and its testimony cannot be verified; here as in the greater part of the
Paeans, its reliability is a postulate. Such a postulate does not, however, seem to be too far
odd chance that an asteriskos is lost in the lacuna next to the koronis below fr. 4b.8 (l. 9 is probably the beginning of an ode: Schneidewin 1834: 112f., DíAlessio 1991: 110), and no compelling reason to assume poem-end after fr. 4a.9, where only a koronis is present (DíAlessio ibid.). H28 is genuinely ambiguous, due to a lacuna next to the surviving koronis in fr. 1. Thus, while not all the Pindar papyri agree with Hephaestion in using an asteriskos at poem-break, there is no occurrence in a papyrus of Pindar of an asteriskos signifying anything other than poem-break. (On the asteriskos generally see now Nocchi Macedo 2011.) Bacchylides has a simple koronis � sometimes as plain as a forked paragraphos � marking poem-end on A and Q fr. 3, but an asteriskos at the end of the monostrophic odes 6 and 8 (but not 4 and 18) on A and on the conjecturally attributed PSI 1181 (frr. **60-61 M.); H and L are uncertain due to a lacuna next to the surviving koronis, and one wonders whether the early P had any metrical signs at all (Grenfell and Hunt 1915: 65 remark the absence of paragraphoi marking strophic divisions; the sign at the beginning of fr. *20C M. in fr. 4 may have been a forked paragraphos, but there is none atop fr. *20B M. on fr. 1). Positive papyrological confirmation of Hephaestionís contention that the asteriskos marked specifically the end of a poem, not a division between two or the beginning of the next, is still wanting, but the paragraphos that marks the end of the second antistrophe of Pae. 4 is at the bottom of H4 col. xviii not at the top of the following one The forked paragraphos (if it is one at all) at the top of the column on P fr. 4 could be one piece of evidence to the contrary, but the erratic use of metrical signs on that papyrus (see above) prevents one from drawing any firm conclusions on its basis. 7 The papyrus lends no obvious support to the hypothesis that F5 was marked out as a subsection of Paean 6 rather than as a new poem (Rutherford 1997: 19, 2001: 301, 330). There is nothing to distinguish the metrical signs employed here with those regularly marking poem-end elsewhere: while the uniqueness of this scenario forbids certainty as to which signs would have been used in such a case (and there is no evidence that such a case is possible in the first place), it cannot be lightly assumed that the choice would have fallen on the asteriskos, endowed as it was with a different, established meaning. This also tells against the supposition that asteriskos and title were meant to signal the start of a new poem � which was also an independent Prosodion � without implying the simultaneous end of the (three-triad) Paean: a reader would have had no way to determine that this was, quite exceptionally, the case.
231
from the truth when verification is possible; the manuscript shows no trace of blatant
interpolation elsewhere.8 That the annotator chose to chop off one triad and fabricate a title
for it suo Marte is conceivable, but hard to substantiate: one seeks in vain a similar display
of intellectual audacity in the much-scribbled margins of H4.9 A broader point ought also to
be made, which tells against the hypothesis that H4 took over a conjecture from another
scholar: evidence for ancient scholarly practice in respect of Pindarís text clearly suggests
that it favoured proposing any alterations in the margins of a copy of the standard edition,
or in a separate commentary, over their actual incorporation in the text.10 The distinction
between text and margin, between the realm of tradition and the criticís, was to a great
extent carefully respected: Jean Irigoinís comparison with the modern-day practice of art
conservation � ìto do nothing irreparableî � is appropriate indeed.11 Pindar does not seem
to have been handled any less respectfully than other authors of a similar status, as has
been observed by modern commentators.12 While this obviously does not guarantee that the
text of any given manuscript is free from casual error, nonetheless it suggests that it should
be regarded as free from deliberate alteration until proved otherwise.
8 Irigoin 1952: 89f. Notably, in one case of important corruption (F5.12 —kRfb, one syllable missing) the error is shared with H5, thus intimating that the blame rests with a common ancestor, possibly even with Aristophanesí edition (see n.): on his preservation of known unmetricalities in his text see p. 237 and n. 27 below. 9 On the annotations in H4 see pp. 122f. above. 10 See Irigoin 1994: 41-5, 54, 1998: 408-11, 413, Jacob 1998: 31f. This may not have applied to those authors of whom multiple editions, in the modern sense, existed (such as Alcaeus, Heph. p. 74 Consbruch), but pace Rutherford 1997: 19f. there is no evidence for such re-editions of Pindar after the text established by Aristophanes of Byzantium (see p. 53 and n. 37 above): Irigoin 1958: 51f., 68, DíAlessio 1997: 52. 11 Irigoin 1998: 413. 12 See Irigoin 1958: 49-53, 58f., 68.
232
Significantly, this is not restricted to Pindarís actual words, but also applies to certain
elements that we could consider ancillary. The evidence for metrical markers is not
decisive,13 but titles � despite being an Alexandrian creation, or at least systematisation �
seem to have been treated as a part of the transmitted text after Aristophanesí edition: their
transmission is by and large consistent,14 they are sometimes commented upon by the
scholia,15 and when disagreement occurs, they are not merely discarded to make room for
13 See the marginalia in P.Oxy. 2291 col. ii.3, 2430 fr. 1 col. ii. On the one hand these notes provide a parallel for the conjectural supplement of a missing metrical sign; on the other, they confirm that the usual process was copying from the exemplar, or there would have been nothing worth remarking. 14 The status of titles in the papyri of Pindar is complex, and does not seem to have received a full treatment to date. In some cases, titles are inset in the column, therefore written at the same time as the surrounding text and by the same hand (H1 fol. iV, H5 frr. VV/VIV, H26 frr. 7, 14.i, 16, 65, 86, H28 fr. 1, H45 frr. 1, 5, P.Oxy. 5043 fr. 15); in others, they were added in the margin, by the original scribe (H29 fr. 1, P.Oxy. 5042 with Maehler 2010c: 77, and H32 fr. 4b.8, where the title should be the ]Q to the left of the koronis not the tiny note to its right, pace DíAlessio 1991: 110) or by a corrector (H4 col. xxii, H9, possibly H11); sometimes the margin does not survive, thus precluding verification (H7 frr. 8, 84, H22, H43 (a), P.Oxy. 5038 with Maehler 2010a: 66f.). Even in those instances where the title was added by a different hand, the addition seems to be virtually contemporary to the main text (on H9 see Grenfell and Hunt 1919: 29). Bacchylidesí situation is mostly the same: titles added in the margin by two different hands in A (see Jebb 1905: 127-35), by the same hand as the text in L (if a title at all, see Lobel 1956a: 29), and are inset in the column in H, probably Q fr. 3 (the koronis in fr. 6 marks triad- not poem-end, see DíAlessio forthcoming), and the conjecturally attributed PSI 1181. P fr. 4 stands out, where the title was penned by a hand which I would place in the second or early third century AD while that of the text probably dates from the first century BC or very early first AD (Cavallo 1974: 36 nn. 19f. correcting Grenfell and Hunt 1915: 65). The extant Simonides has three titles, all inset: P.Oxy. 2430 frr. 35(b) and 120(b), and 2431 fr. 1(a), even though I do not quite know what to make of the absence of titles and even of an interlinear space at 2430 fr. 25, where the asteriskos would lead us to expect poem-break. Lobel 1961d: 42 seems to suggest that poem-titles could normally be absent as well as present on individual mss. (which in itself would not refute their authority when they were present), but no demonstrably title-less papyrus of Pindar, Bacchylides, or Simonides has been published to date: while the title of B. 5 is missing from A col. x, all other odes in that ms. are titled, albeit none by the main scribe, which suggests that the absence is an ordinary scribal omission like several others which the second and third hand subsequently rectified. Rutherford 2001: 150 rightly remarks that the alphabetic arrangement of Bacchylidesí Dithyrambs by title also entails that such titles were established in his Alexandrian edition; the same applies to Simonides, if some of his works too were organised alphabetically (see p. 19 n. 38 above). Scribal omission must also be responsible for those Pindaric epinicians whose title is not found on the medieval mss.: P.Oxy. 5043 fr. 15 and H25 A fr. 2(a) show that the lost titles of N. 10 and I. 4 were transmitted at least as late as the Roman era; the similarly lost title of I. 3 is mentioned by inscr. p. III 223 Dr., and whoever wrote inscrr. b-c I. 5 p. III 241 Dr. must also have had the title of the ode (mentioning Pytheas as well as Phylacidas) in front of him. 15 See e.g. inscrr. O. 3, I. 3 pp. I 105, III 223 Dr.
233
the ëbetterí reading.16 This implies, a fortiori, that the poem-divisions that titles and
metrical signs denoted were also received and transmitted together with them.17
All this goes some way towards suggesting that the separation between Paean 6(a)
and F5 presented by H4 goes back to the Alexandrian edition of the Paeans, which
consequently will have presented no less an independent F5 than did (as far as we can
judge) the Prosodia.18 The notion that this was the state of affairs receives further support
from the scholiaís notorious contention that the treatment of Neoptolemus in Paean 6(a)
had offended the Aeginetans, whereupon Pindar sought to excuse himself with a more
ëpolitically correctí narration of the same myth in Nemean 7 (the so-called ëapology
theoryí).19 Several commentators have noted how poorly the offence allegedly felt by the
Aeginetans tallies with the praise that Pindar lavishes on Aegina in F5, the supposed final
16 In the mss. the title of P. 10 presents the name of the honorand in the Attic form πccgjVW~ (see Mommsen 1864: 268) while inscr. p. II 241f. Dr., which also reads it so, points out that the genuine form (which Pindar himself uses, 5) is πccgjVYRb. 17 A witness to the authority of Aristophanesí metrical signs is | metr. N. 4 p. 23 Tessier, which analyses the (monostrophic) ode as triadic while explicitly recognising that the supposed epode has the same metre as the supposed strophe and antistrophe. Rutherford 1997: 19f. presents an alternative hypothesis in which a heading originally not accompanied by an asteriskos may have been ìa mere sub-title, not indicating the beginning of a new poemî, and the asteriskos was added later. However, there is no other evidence for such segmentations of poems or any accompanying indications in Hellenistic or Roman scholarship on earlier lyric (surely the traditional division of the nomos is no parallel); here, their position guarantees that the asteriskos was written before the title, or that the title was written so as to leave a space for the asteriskos. 18 The likely presence of the connective Sn¥� at the very beginning of the piece does not preclude this interpretation. Implausible as this sounds, some quarters of Alexandrian scholarship had a notion that ìbeginning from Sn[î was a typical feature of poetic style: see 1 n. 19 | N. 7.48 (naming Aristarchus), 64, 103 (Aristodemus), pp. III 126, 129, 137 Dr., all conveniently accessible in Rutherford 2001: 321 n. 64. The citation of Pae. 6.117-19 in | 64 establishes beyond question that the text referred to is indeed Pae. 6(a). On Aristarchus and Aristodemusí work on Pindar see Irigoin 1952: 51-6, 59f., Razzetti 2000. A full bibliography on the ëapology theoryí and its modern receptions would take up a chapter by itself: see lists at Segal 1967: 431f. n. 1 and Arrighetti 1987: 75 n. 132, with a handy summary at Currie 2005: 321 n. 133 (more discursively Loscalzo 2000b: 7-24). The plausibility of the ëapology theoryí has been refuted more than once, and flogging dead scholiasts is beyond the remit of this dissertation: the reader is referred to the most recent and detailed argument to this end, Currie 2005: 326-30, whose conclusions I share and will assume throughout this discussion.
234
triad of the offending poem.20 This consideration should be taken one step further. If the
ëapology theoryí is a construction that proceeds from no other source than the poemsí
texts, then the idea that a scholar could conceive it on the basis of the three-triad Paean 6
requires us to assume that he flatly ignored one third of the ode in order to focus on
tendentiously interpreting one line. This may perhaps be blamed on the mild myopia,
absent-mindedness, or lunacy that sometimes accompany prolonged study, but a simpler
explanation lies to hand. If what was felt to need an apology was not the entire Paean 6 but
the independent Paean 6(a) that we see on H4, where a potential slight to Neoptolemus is
not offset by a flattering coda, the reasoning behind the ëapology theoryí becomes
distinctly less absurd. While this may not amount to proof that the ëapology theoryí is a
terminus ante quem for the division between Paean 6(a) and F5,21 nonetheless it is fair to
say that it sits considerably better with an existing division between Paean 6(a) and F5
than with one, three-triad Paean 6. This substantiates the papyrusí testimony that the
Alexandrian Pindar presented not one but two odes in place of what we generally regard as
Paean 6.
This leaves us with three interlacing problems: (i) the presence of two professedly
distinct odes in an identical metre, (ii) the transmission of one poem in two different books,
and (iii) the different colometry with which that poem was articulated in the two books. Of
(i) there is another well-known Pindaric example, Isthmians 3 and 4. It is now quite
20 E.g. Farnell 1930-2: I 313, II 408, Bowra 1933: 45, SuUrez de la Torre 1997: 171f., Rutherford 1997: 14, Currie 2005: 328. 21 The ëdate of birthí of the apology theory is itself unknown, apart from the terminus ante quem constituted by Aristarchus. The phrasing in | N. 7.70 p. III 126 Dr. does not necessarily imply that it was he who first devised it: compare its attribution to Aristodemus in | 150a p. III 136f. Dr. A similar story about the Corinthians taking offence at Simonides for PMG 572 = fr. 290 Poltera is attested as early as Aristotle (Rhet. 1.6 1363a14 with CAG XXI/2 pp. 24f.): Currie 2005: 322.
235
uncontroversial that the Alexandrian edition regarded them as two separate compositions,22
and some consensus has been building up to the effect that they are not merely one poem
that was split in two as a result of ill-advised editorial choice; but the fact remains that the
two odes were composed for the same individual, which makes the coincidence hardly
casual, although positive evidence of a specific ëspecial relationshipí between them � and
of what it may have concretely entailed � is lacking.23 While different compositions in the
same metre are unsurprising on Lesbos, there is no safe example of two wholly unrelated
lyric poems sharing their entire metrical structure elsewhere in Greece; this admittedly
favours the assumption that Paean 6 was indeed one ode. For (ii) the parallel is Alcman
PMGF 3, which, according to a scholion on P.Oxy. 2387 fr. 1, also appeared in the fifth
22 They are marked apart in ms. B, which also prefaces the scholia to I. 4 with a metrical scholium of its own (p. III 225 Dr. = 30 Tessier); this cannot be the result of an accident of transmission, whereas the lack of a division in ms. D is easily so explained (Bury 1890b: 276, 1892: 167f.). | I. 3.24, 29 pp. III 224f. Dr. refer to passages in I. 4 as �Q fvb �Xv] ‹bkvb (ibid.); further confirmation comes from H25 A fr. 2(a), which heads the commentary to I. 4 with a title of its own, f]�b RÃf[�b (Privitera 1982a, see also Benelli 2012: 79-84). 23 For a compact survey of critical positions in respect to the matter see Cole 2003: 241f. One theory of ìspecial relationî (so phrased by Bury 1892: 170) was put forward by Bulle 1871: 585-9 (I have not seen Bulle 1869), envisaging an originally complete I. 4 to which I. 3 was then added as a preface after Melissos won his Nemean victory: thus, while I. 4 will have existed as an independent composition for some part of its history, I. 3 never did. This view has won considerable acceptance: among others, Bury 1892: 169-72, Wilamowitz 1922: 335f., Puech 19312-522: IV 37-9, Farnell 1930-2: I 256, Bowra 1964: 317, 408. A unitarian perspective is still favoured by Sn.-M. 144-8 and Thummer 1968-9: II 55-7. Against both of these hypotheses, one cannot escape the feeling that I. 3 as well as I. 4 has all it needs to be a self-standing epinician, whence the alternative view of two independent poems, the more recent of which will perhaps have been meant to recall the earlier one through their metrical identity: so Schroeder 1878: 1-3 and 19005: 71f., Bury 1890b, Barrett 1954: 149 n. 1 and 2007b: 162-7, KNhnken 1971: 87-94, Privitera 1982b: 43f., Willcock 1995: 69-72; Currie 2004: 62f. attractively suggests that regular re-performances of I. 4 may have provided a context for the first as well as subsequent performances of I. 3 and prompted the re-use of the same metre. Lidov 1974: 178-85 ingenuously submits that the odes were indeed two, but contemporary and meant from the start for a joint performance. Yet differently, and indeed tantalisingly, Cole 2003: 246-52 suggests that I. 3 was meant to replace, not to precede, the first triad of I. 4: on this interpretation, the awkward status of the resulting pieces would go some way towards justifying the Alexandrian editorís misunderstanding of their relationship, although this does not solve all the problems in the text.
236
book of his works, not without scholarly disagreement on which placement was correct.24
As in our case, the scholion gives no explanation for the double transmission.
Finally, no parallel for (iii) is known. Lyric colometry seems to have been
remarkably stable during the late Hellenistic and Roman ages, with only very occasional,
minor divergences in individual manuscripts: nothing like the radical difference that we see
between H4 and H7 in the fragments of F5.25 (ii) may go some way towards explaining it,
since the known examples of lesser variation naturally concern the single transmission of
each ode; but why two different colometries of the same piece should have arisen in the
first place remains a puzzle. The significance (if any) of the fact that one of them coincides
with that of Paean 6(a) is also unclear. If one assumes that either placement of F5 was
somehow secondary to the other, it is easier to interpret as secondary the classification of
the ode among the Paeans with the attendant colometry � both of which could have been
prompted by those of Paean 6(a) � than the equivalent in the Prosodia, where the re-
colisation cannot be accounted for in any obvious way. However, the assumption of a
primary and a secondary classification is not strictly required by the evidence; moreover,
24 c]�[WQS[n(dWfRb) �� [fg]~�] �QfbS[n(dgb]) R—f\ | �Q (or j�Q) f�b] ÈY_cfTb (vacat) j�Ú� �Q �jW^�¼� | �Q f�b] �[(b]fg)Q^(jgP) cW[bWSYS[R(cfg), �Q kÓ f�b HfgV(W_R^gP) | �cW[[^]S[R(cfg]) õQ : see Lobel 1957: 11f., Hutchinson 2001: 104-6, and a full apparatus in Calame 1983: 78. Thus, in the (available) manuscripts the ode also appeared in the fifth book, where Aristonicusí copy marked it as misplaced, and Ptolemyís did not; the annotator agrees with Aristonicus, as cR[WSS[ndT makes clear. Which book the papyrus is a copy of is unknown. Calame 1983: 394f. must be wrong to understand that it is itself a copy of book 5: it makes precious little sense to warn readers of book 5 that the ode they are reading in book 5 can be found (however wrongly) in book 5, and �Q �jW^QTb is an all but unnatural way to refer to the book that both annotator and reader are holding in their hands (rightly Hutchinson 2001: 105). On the other hand, I am inclined to disagree with Hutchinsonís suggestion that in the �Qf^S[Rdgb ìthe poem appeared only in the fifth bookî: the notion that scholars in the late Hellenistic or early Imperial age could move poems around at will goes some way beyond what we know of their practice as far as lyric is concerned (see pp. 230-3 above), and even if common practice had so allowed, it may indeed have been odd of Aristonicus to have the poem transcribed in both books and mark it as misplaced in one of them (so Hutchinson), but arguably no more odd than to have it transcribed only in book 5 and mark it as misplaced there. 25 Irigoin 1994: 80f., DíAlessio 1997: 46-9 contra Tessier 1995: 35-54.
237
the different articulation of Paean 6(a) and the Prosodia-version of F5 � which still have
the same metre, although not also the same text � remains unexplained.26
The question now is whether the mutually independent Paean 6(a) and F5 that the
Alexandrian edition arguably presented are Pindarís own design or the result of subsequent
modification; and if the latter, at what stage and in what form such change may have
occurred. One point is worth making first: that the editor of the Alexandrian Pindar is
relatively unlikely to have split Paean 6 into two on his own initiative. His successorsí
faithfulness to his text seems to have matched his own respect for his sources. Upon
recognising that the forty-eighth colon of Olympian 2 was metrically intrusive and
therefore certainly spurious, he athetised it accordingly, but did not eliminate it from the
text altogether.27 If the �kndbR in which Olympian 5 was regarded as not genuine are to be
identified with his edition,28 we have a further example of Aristophanes not allowing his
critical views to impact on the actual text: the ode, although suspected, was included in the
volume of the Olympians, where it was copied and commented on like its next-page
26 DíAlessio 1997: 59 suggests that the colometrical division may have taken the music into account. If the two pieces reached the editor independently, each with music of its own (whether original or not), it is indeed conceivable that it is this difference that the different colometries reflect, but the survival of archaic and classical music into Hellenistic time is too controversial a matter to be relied on. 27 | O. 2.48c, f p. I 73 Dr. with Irigoin 1958: 45. The colon is commented upon by | b, e, g ibid. and was first physically expunged from the text by Demetrius Triclinius in the fourteenth century AD: see Mommsen 1864: 18, Irigoin 1958: 346. On Aristophanesí text-critical attitude in respect to Pindar see further Irigoin 1994: 43f., 1998: 410f. 28 Inscr. a O. 5 p. I 138 Dr. with the compelling interpretation of gÃj õQ offered by Ruffa 2001: 42-4 (the ode ìwas notî regarded as genuine, a use of the phrase that can be paralleled in the Homeric scholia). For the earlier view that the expression referred to a previous edition by Zenodotus in which O. 5 actually ìwas not thereî (a view that relies substantially on the �/�\ signs on H4, see n. 31 below) see Irigoin 1958: 32f. A further alternative is now offered by Ferrari 2006.
238
neighbours.29 If Aristophanesí text had Paean 6(a) and F5 as two separate compositions, as
H4 and the ëapology theoryí suggest, some form of the division is likely to predate him.
Beyond the simple possibility of the original independence of the two pieces, a wide
array of scenarios presents itself, as Rutherford has expounded in detail.30 A split may have
been due to an unscrupulous scholar, although it is unclear who this dodgy figure may have
been,31 or caused by a tradition of performances of F5 on its own by a community which
had no interest for the rest of Paean 6.32 It may even have been due to Pindar himself, if he
re-used a part of an earlier composition as a self-standing poem.33 On the other hand, the
two pieces may have been originally separate but related in some way, perhaps to be
performed separately in the same context,34 or consecutively by two choruses of Delphians
and Aeginetans:35 this would vindicate the titles of both pieces on H4 , but lacks any secure
parallel. Another ëspecial relationí scenario can be provided by a revised version of the
ëapology theoryí, if Paean 6(a) had offended the Aeginetans and Pindar composed F5 as a
29 Irigoin 1958: 32f.: the scholia record Aristarchusí interpretations in several places (| 1b, 20e, 27b, 29e, 54b pp. I 141f., 145-7, 151 Dr.), not to mention Didymusí pronouncement for the authenticity of the ode and his other comments on its text (Inscr. a, | 20e, 27b pp. I 138, 145, 147 Dr.). Possible evidence that Aristophanesí attitude was not exclusively conservative is offered by | Ar. Th. 162a Regtuit, which (citing Didymus) attributes a textual emendation to him. However, Pasquali 19522: 199 rightly suggests that Didymus cannot have known that Aristophanes had no manuscript authority for his reading; one might also doubt whether Didymus truly had access to manuscripts earlier than Aristophanesí edition, as he appears to claim. (Pfeiffer 1968: 189 accepts the information offered by the scholion.) 30 Rutherford 1997: 14-19. 31 Rutherford 1997: 14. On one view this could be Zenodotus, who compiled an edition of Pindar before Aristophanes according to Irigoin 1958: 32f.; however, after the recent re-evaluation of the marginal signs �/�\ on H4 (see p. 123 n. 12 and above, n. 28) and of inscr. a O. 5 p. I 138 Dr. (Ruffa 2001: 39-44) there remains very tenuous evidence (two variant readings preserved by | O. 2.7a, 6.92b pp. I 60, 174 Dr.) that such an edition existed. 32 Rutherford 1997: 14, cited with some approval by Fearn 2007: 90, 2011a: 202f. 33 Deborah Boedeker in Rutherford ibid., Loscalzo 2003: 98f. A degree of autonomy already in Pindarís conception is also suggested by DíAlessio 1997: 58f., 2005: 136 n. 73. 34 Rutherford 1997: 15: ìBut such a recherch� hypothesis is perhaps best resorted to if all others failî (ibid.). 35 Rutherford 1997: 17-19, espoused by Kurke 2005 (119-25 for a detailed hypothesis on reperformances of F5 on its own in Aegina).
239
ìcompensatory supplementî as well as defending himself in Nemean 7.36 This hypothesis
suits well the attested views of Aristarchus and Aristonicus, and I would be very surprised
if F5 had played no part at all in their views, but its reliance on the veracity of the ëapology
theoryí all but disqualifies it. By now we are in the realm of full-fledged speculation: it
would be hard to favour any one of these hypotheses on any more objective grounds than
oneís own sense of plausibility, and one is left with the feeling that none of them compares
too favourably to that of an unimpeded tradition of two separate odes from Pindarís
conception to the Museum.
If we could be certain that the titles given to Pindarís compositions drew on external
evidence for their commission and performance, then the different titles of Paean 6(a) and
F5 would testify to their mutual independence, unless the scenario of split performance
was accepted. The possibility that the origin of individual odes may have been recorded
and transmitted cannot be discounted altogether, and indeed the notion that lyric poetry
reached the Library in an absolute vacuum is not particularly attractive,37 but a non-textual
origin cannot safely be assumed in every specific case.38
At this point we must turn to the evidence of the texts themselves. As is to be
expected, it will prove ambiguous and often contradictory, providing plenty of ammunition
for competing views. If the preceding paragraphs had given an undisputable answer to the
36 Rutherford 1997: 15-17 after Farnell 1930-2: I 313, II 408, endorsed by Furley and Bremer 2001: I 111f. 37 So DíAlessio 1997: 58, cp. Lowe 2007: 176. 38 Note, however, that the allegation of autoschediasm levelled at the title of I. 3 by inscr. p. III 223 Dr. is unjust: whereas Olympians and Pythians distinguish between the various equestrian contests, µccgb] is the standard tag for all equestrian victories in the Isthmians and Nemeans. Indeed, if this depended on the fact that the Olympic and Pythian victor-lists provided external evidence for the contest won, while for the Isthmians and Nemeans the editor had to rely on the text itself, then the reliability of the indications which titles do provide would be enhanced; cp. the title of Pae. 8b(a) on H45 fr. 1. But the matter is highly uncertain.
240
question whether the separation between Paean 6(a) and F5 is Pindaric or not, there would
not be excessive difficulty in squaring the text with either view, giving greater weight to
this or that element as the external evidence indicated; as things stand, however, different
clues appear to suggest different answers, several can be understood from two
diametrically opposite angles, and no single one can be relied on as decisive.
One obvious argument for the unity of the ode is the metrical identity of the two
pieces.39 The fact that F5 almost certainly began with Sn¥� is another.40 A parallel for the
former has been discussed above; as for the latter, one must note that several texts, Greek
and Latin, in poetry and prose, begin with a floating connective (�VVn, kY, et, at), but no
secure example of a text beginning with Sn[ seems to be found earlier than Dio
Chrysostomís speech On Retirement.41 The suggestion that the text of F5 in the Prosodia
may have had a different reading, say _YQ, amounts to little more than special pleading,
and does not solve the difficulty that the piece still began with Sn[ in the Paeans.42
Aside from Sn[, however, the elements that open F5 are well suited to an incipit: a
second-person invocation enriched by a string of epithets (1-4), the statement that the ëIí is
going to sing (5f.), the posing of a question (8f.) whose answer will develop into a lengthy
mythical narrative (10ff.). On the other hand, none of these is such as cannot occur with a
transitional, rather than a prooemial, function, especially after the strong break that the
unitarian view must envisage between second and third triad; the concurrence of all three
39 DíAlessio 1997: 58, Currie 2005: 325, Kurke 2005: 92 n. 34, contra Ferrari 1999: 152 n. 11. 40 Rutherford 1997: 13, Currie 2005: 325. 41 D.Chr. 20 (also 67, but that is a dialogue); for other examples and discussion see 1 n. 42 Suggested by Rutherford 1997: 13, DíAlessio 1997: 56f. Crucially, the underlying assumption is that F5 in the Paeans was a part of Pae. 6, not an independent poem that followed 6(a), whence the need to dispose of Sn[ only in the Prosodia.
241
elements is perhaps suspicious but not patently impossible, especially if an effect of ëre-
openingí is what Pindar was seeking (compare fr. 33c Sn.-M., also at the start of a metrical
unit). Likewise, the dramatised paean-cry at Paean 6(a).212f. is prima facie a strong
closural element:43 although Pindarís two other known instances of the paean-cry recur at
the end of each triad of the ode (Pae. 2, 4) and the similar refrain of Paean 5 is also
repeated at regular intervals, other specimens of the genre show a clear tendency to have
any non-recurrent paean-cries at the end of the composition, while a one-off occurrence in
the middle of a song would be highly anomalous. Nonetheless, the presence of a major
transition can legitimately be thought to justify, or even to invite, ëfalse closureí; this is not
Pindarís usual practice so far as one can judge, but it cannot be discounted altogether,
especially since we have no comparable examples in the extant Pindar of a major transition
rounding off a metrical system, where such a device naturally has its fullest force. Thus,
although both of these considerations sit somewhat more comfortably with the idea that the
end of Paean 6(a) is in fact an ending and the opening of F5 a beginning, they are easily
accommodated to one Paean 6 if other evidence so requires.
From the thematic point of view, the hundred years in which the actual unity of
Paean 6 was assumed resulted in several arguments being put forward to justify its
integrity and coherence in face of the obvious dissimilarity of subject between the first two
triads and the third. The main connexion was found in the myth of the famine from which
Greece had been delivered by the prayers of the people of Delphi (Pae. 6(a).63f.), echoed
in F5 by the mention of Zeus Hellanios (3f.), whose epithet was reportedly due to a similar
43 Rutherford 1991: 4 n. 15, 1997a: 20 n. 59, 2001: 71f. with n. 12.
242
episode in which a drought had ceased thanks to Aeacusí supplication.44 The
straightforward identification of the two episodes cannot be assumed, since they differ in
the fundamental question of whose the credit was for the cessation of the draught:45 the
simultaneous presentation of both claims may have sounded awkward, especially in the
context of the Delphic Theoxenia, which (as it seems) explicitly celebrated one of them.
On the other hand, it may be argued that the subtle reference to the Aeginetan episode was
simply meant to recall its similarity to the Delphic episode without superimposing the two.
If the speaker of Paean 6(a).1-18 is a choral first person and one that does not suit
the Delphic chorus of the title, then it can profitably be identified with the Aeginetan
chorus of F5;46 but both premisses are open to doubt, and they require the assumption (a
perfectly possible one, to be sure) that the title of Paean 6(a) is simply wrong. Similarly,
the evident if somewhat unsettling parallel between the first person and Neoptolemus, both
of whom come to Delphi (17~120) out of concern for their fb_R^ (11~118), has a special
point if the chorus is one of Aeginetans, Neoptolemusí putative fellow-citizens, but it is no
less pointed with a Delphic chorus, which would emphasise the concurrent parallel
between the Delphians of old through whom Apollo slew the hero and those of the present
who sacrifice him in (and through) their song.47 Bruno Currie has thoroughly explored the
cultic links between Aegina and Delphi that may have been centred on the HPp^gP
44 See Tosi 1908: 203f., Radt 1958: 89f., 132f., Rutherford 1997: 2, 2001: 331f., Currie 2005: 332f., Kowalzig 2007: 183. Sources for the Aeginetan episode: | F5.3 n. The Delphic version is unattested elsewhere, although the oracle plays an advisory role in Pausanias and Clement. 45 So Radt 1958: 133, Rutherford 2001: 332, Furley and Bremer 2001: I 111. 46 Currie 2005: 325, with earlier bibliography at 323 n. 43. 47 The parallel was noted by Rutherford 2001: 315; for an interpretation of the sacrificial imagery in ëPae. 6í and of its relevance to the construction of the persona loquens see Kurke 2005: 102-17.
243
rWn[bgQ that Pindar mentions at Nemean 3.70,48 but such ties are not necessarily relevant
to Paean 6(a) unless one assumes the unity of the ode. Verbal echoes of various sorts have
been detected across Paean 6,49 but while they may usefully contribute to the appreciation
of a poemís cohesion once its status as a single poem is established, they should be treated
with some caution when used to buttress the unity that they ultimately assume.50
In short, while several elements show quite persuasively that the two supposed
halves of Paean 6 can go nicely together if so much is required of them, there is no very
compelling reason to hold that either half requires the other in order to be a complete
poem: from the thematic and structural point of view, each can function excellently on its
own. If by an accident of transmission F5 had not been preserved and the end of Paean
6(a) was followed immediately by Paean 7, it is hard to imagine any modern commentator
seriously entertaining the possibility that the text, with its leisurely proem (1-63), myth
(63-120), and tensely abrupt closure (121f.), was mutilated at the end. That the same
applies to F5 I will attempt to show in the coming paragraphs, which will revert to the
Prosodia and to the independent piece as was transmitted there.
To conclude, we are left not too far past where we started. The argument from
tradition favours the hypothesis of original independence, although alternative possibilities
cannot be excluded. One internal element � Sn[ � tells especially strongly against original
independence, but perhaps it is not insurmountable; the metrical identity of the (on this
view) two pieces is another possible problem. However, to solve these problems by
48 Currie 2005: 333-9; see also Walter-Karydi 2000, Fearn 2011a: 194-204, Rutherford 2011, and pp. 253-5 below on the Aeginetan cult of Apollo Delphinios. 49 See SuUrez de la Torre 1997: 171, Burnett 1998: 515 n. 80, 517 with nn. 85, 87, Kurke 2005 passim, Currie 2005: 325. 50 Compare the ìthought experimentî of Strauss Clay 2011 (quotation from 338).
244
throwing the papyrusí testimony overboard creates in turn a series of other problems
whose import in terms of theme and structure seriously leads one to wonder whether the
intervention is really worth the cost. Up to 1997, economy doubtlessly favoured attempting
to keep the pieces together and make sense of them as much as possible over chopping the
poem in two against what was then regarded as the manuscript evidence; now the situation
is, or should be, the opposite. On the balance of the evidence as it is currently known, I
believe that a decisive argument for discarding the testimony of H4 is lacking. Pending any
new discovery, and fully acknowledging the precarious nature of the evidence, if I were
forced to take sides at all costs I would think it safer to side with the ZT[^�gQfW] (and H4);
in any case, given the balance of scholarship so far, I believe it was useful to present as full
a case for the minority position as I could assemble, and let the reader decide.51 At this
point we can finally move on to discuss F5 itself.
The first two lines of its title on H4 seem to coincide exactly with that of F3 on H29,
which in turn was instrumental to the restoration of the title on H4 itself.52 The addition of a
specification of the genre at the end is paralleled by the title of the oschophorikon
transmitted with the Isthmians as preserved by H25 fr. B 17.6 �p]î�R^Tb ‹]Zg+[g[bj¦Q (if
Lobelís supplement is correct).53 I suppose that c[g]¦kbgQ will not have occurred in the
51 The only committed ëseparatistí in print so far is Ferrari 1999; Furley and Bremer 2001: I 106 print Pae. 6(a) alone but without ìa firm conviction Ö that the third triad is a separate poemî. 52 Rutherford 1997: 4f. 53 Lobel 1961h: 177. It is also conceivable that the word was not ‹]Zg+[g[bj¦Q but ‹]Zg+[¦[Tb along the same lines as the (other) epinicians, name + origin + athletic class, all in the dative (�[b]fg_YQWb o∞SbQqf\b cRVRb]fvb etc.). If so, the anomaly in relation to the rest of the book would have been rather less highlighted than it is by Lobelís conjecture. If the titles of the jWZT[b]_YQRb N. 9-11 had been preserved by the manuscript tradition we could have a better term of comparison for titles in similarly ëirregularí placements; as things now stand, the parallel with F5 itself seems to give ‹]Zg+[g[bj¦Q the edge. The same applies to the ode for a victor in the Corinthian Hellotia which DíAlessio has recently recognised in fr. B 14.i-26ff. of
245
title that headed F5 in the Prosodia themselves, but certainty is impossible;54 there is also a
possibility that the Prosodia may have presented a mirror image, as it were, and entitled
the piece o∞SbQqfRb] W∞] o∞RjÙQ cRbnQ.55 The overlap between prosodion and paean,
discussed in the general introduction, reminds one of the cRbÏQ c[g]gkbRj¦] that the
Pindar scholia mention in relation to the composition of Isthmian 1.56 Even closer to our
case could be Paean 7, whose title in H5 reads c[g]×É�[: whether it was c[g]×É�[Rj¦] as
in the scholia to Isthmian 1 or c[g]¦�É�[gQ as in F5 is debatable, but the title certainly
displays some degree of avowed intersection between paean and prosodion.57
The classification as a Prosodion of a piece that explicitly advertises itself as a paean
is a puzzle. The likeliest explanation is its destination for the worship of Aeacus, whether
known from independent evidence or extrapolated from the text itself: as we have seen,
songs for heroes were generously represented among the Prosodia (at least F2, F3, *F7,
F8), while ancient generic theory does not seem to have contemplated heroes as legitimate
dedicatees of paeans.
the same commentary (2012: 48-54): the papyrus presents the nonsensical abbreviation WVTQbg, which can be expanded into either �VVifbR or �VVTfbgQ^jRb] (52). 54 One notes how H45 fr. 1 seems to contain a title starting with cRbÏQ W∞] [ (Pae. 8b(a); Lobel in Maehler 1989: 5) in a book which preserves fragments of another Paean, 8. The status of that manuscript, however, is quite unclear (Rutherford 2001: 151f.); the lack of an indication of the performers before that of the addressee makes things even more obscure. (The presence of dots above and below the pi and above the nu of cRbnQ and above the epsilon of W∞] does not indicate deletion of the title, as might seem at first sight, but is simply an unusual kind of ornamentation: there are dots also above and below the sigma of }WV]+g~c in the title of Pae. 8 on fr. 5, which surely there is no reason to question.) 55 This is somewhat less likely in view of ancient doctrine on the addressees of paeans (SchrNder 1999a: 10-49, Rutherford 2001: 23-36), but the fact remains that F5 was classed among the Paeans and openly advertised itself as one (5, cp. 60). 56 Pp. 102f. above. 57 See p. 102 n. 28 above. This does not imply, of course, that Pae. 7 too will have enjoyed a double transmission even if entitled ìprosodionî.
246
F5 is the only Pindaric Prosodion for which a complete metrical scheme can be
compiled, with the help of the corresponding sections of Paean 6(a). Since the main
manuscript witnesses for F5 are in fact the two papyri of the Paeans, to which those of the
Prosodia (H7 and arguably H26) add no text, the editor of the Prosodia must make a great
but altogether unavoidable leap of faith, and assume that the text of the piece in the two
books was identical: while there is no positive evidence to the contrary, and indeed the
very notion of a double transmission reassures us that the two versions must have
substantially coincided, the uncertain basis of our text in each individual instance ought to
be kept in mind throughout the discussion.58 Another consequence of the peculiar condition
of our manuscript evidence is that, although the only thing that we know of the text of F5
in the Prosodia is that its colometry was not the same as in the Paeans, the editor is
compelled to adopt the latter out of ignorance of the former, which cannot be safely
established even for the smallest stretches of text.59 Despite these problems, F5 is by far the
best preserved of the Prosodia, and the only one whose beginning, end, and a respectable
proportion of the intervening mythical narrative can be read; only twelve lines, or about
one-fifth of the piece, are wholly lost at the turn of the epode.
The piece opens with a second-person address to the island Aegina, whose name is
famous (1); she is the Lady of the Dorian sea (1-3) and the star of Zeus Hellanios (3f.), her
particular and at the same time professedly pan-Hellenic Zeus. The speaker then proceeds
to announce first a banquet of paeans � a plural which leaves the door open to a wider
58 Rightly Rutherford 1997: 13 n. 41. 59 DíAlessio 1997: 49-51. Conjecture on either or both of the surviving fragments has been variously attempted by Snell 1938: 431, Radt 1958: 12, Alessandro Pardini in his regrettably unpublished Colometrie e tradizioni dei lirici (cited by DíAlessio 1997: 50f.), and DíAlessio 1997: 51 n. 167, but the evidence is too thin to allow definite conclusions even on the few lines preserved by H7, even less on the rest of the piece.
247
referent than F5 itself � which Aegina is to taste (5f.), then a ìmurmur of songsî that she
will receive while recounting the origin of her ìship-ruling divinityî and ìvirtue of rightful
hospitalityî (6-9).
The answer is immediately forthcoming: it was Zeus who gave Aegina her prosperity
(10-12). A nonchalant kY starts the mythical narrative proper, whose first episode openly
motivates the preceding claim: the god snatched up the maiden Aegina from the threshold
of her fatherís house and made love to her under a golden cloud on the ënative backí of the
island (12ff.). For the next fifteen lines, all reconstruction is speculative: the main body of
H4 preserves only the first few letters of lines 19-33, to which its fr. 46 adds a further
portion of lines 28-33,60 while H5 provides slightly more generous line-ends. The probable
mention of Myrmidons and of someone ìmost prudentî (21f.) followed by Zeus in the
genitive suggests that, as in Isthmian 8.21-6 and Nemean 8.6-12, Pindar proceeded from
the union of god and maiden to the birth and deeds of their son, Aeacus (Snell);61 this is
consistent with the expectations that arise from the title. A few lines further, the topic has
seemingly moved on to hospitality, something or someone connected to the sea, and
perhaps an epithet of the sea or of a sea-deity, capped by reception and immortality (27-
30). The next sentence mentions islands and the Ocean, which Benjamin Henry relates to
the Isles of the Blest, where Aeacus and the Aeacids may have been translated after death.
This is an attractive possibility, but not the only one: the clan also had maritime
connexions through Peleusí marriage to Thetis, which Pindar connects to Aeacus and his
virtue in Isthmian 8.26a-48, and the end of the world (represented by the Ocean) can also
60 Correctly placed by Henry 2005: 7f. 61 All references for the rest of this and the next paragraph will be found in the respective nn.
248
denote the extreme boundary of fame. Then Zeus, Styx, an oath, and judgement can be
discerned (33f.): they may be connected to Aeacus, √ jRÚ £ kRb_¦QW]]b k^jR] �cW^[RbQW (I.
8.23f.), as Snell suggests, although this is, again, not the only conceivable option.
At this point the text breaks off, to resume a dozen lines later with a first-person ÒkT
(48): the unlikelihood of an intervention by the primary narrator half-way through the myth
suggests direct speech, perhaps a prophecy, as Rutherford suggests. (Where such a speech
may have ended can only be guessed at.) ìBronze-rejoicingî, ìpenaltyî, and ìangeredî
(49f.) would suit Achillesí wrath in the Iliad, but also, among other candidates, Aeacusí
fury at Peleus and Telamon after the murder of Phocus, which occasioned the spread of the
Aeacid lineage and its glory throughout the world. The Aeacidsí deeds of prowess are
indeed what follows (54f.), presumably seen through the lens of the countless V¦Sgb that
arise from them or that are at hand for those wishing to praise them.
This moment of poetic projection telescopes the myth and the present performance,
and leads seamlessly into the double prayer that constitutes the final section. First, a plural
addressee � presumably the Aeacids � is asked to love their (and the first personís) native
city and ìthis cheerful peopleî, shading it with the bloom of good health (55-9); then Paean
is called on to receive ìone who frequently has a share in the voices of the Musesî (59-61),
neatly recalling the earlier, programmatic occurrence of cRbqTQ and kYjg_Rb in the
opening (5, 7) while bringing the song to a close.
The suggested slippage of the poetic persona in the opening and closing sections
deserves comment. The speaker presents himself with the plural WÃQnXg_WQ (6), which
nothing in the text suggests we should identify with anyone other than the chorus who will
have been performing the piece. But no sooner have they established their status as
249
performers, than their task of narrating is unexpectedly shifted onto Aegina, the addressee
(7): she will receive the chorusí song and at the same time take up their role as speakers in
response to their question. The answer, whose beginning is sharply marked out by
asyndeton, effectively conflates the notional and the actual speaker while keeping up the
impression of personal interaction with fgb (10); the question-and-answer format and the
blurring of voices that inevitably ensues are a well-known prooemial device whose most
distinguished ancestor is the opening of the Iliad (1.8f.), but jRfW[W~] plays it out to an
unprecendentedly overt degree. The ambiguity is not immediately resolved, since the
possessive in ]ÙQ Ö ƒV�gQ (11) can conceivably be taken with either of the personae
involved. One effect of the overlap between sung object and singing subject is the
suggestion of a close identification of the speaker with Aegina herself, island and
(crucially) polis: the chorus is made to stand in for the community, its landscape, and its
emphatically numinous eponymous being, all merged into each other.
Likewise at the end of the piece. The myth culminates in ì(telling) the boundless acts
of valour of the spear-clashing Aeacidsî (54f.), apparently with a wider referent than the
performance of F5 alone. In this act, the singers subsume their own song under the heading
of the countless songs that are (or can be) devoted to Aeginaís distinctive heroes, whereby
the voices thus evoked reverberate, as it were, on to the prayers that follow, overlapping
with the singersí own and reinforcing them. It is also worth noting that the generality and
singular number of ��cR�gVY�gQ�[R] allows the final prayer for reception to refer to the poet
as well as to the performers, both of whom can expect to find the godís favour for the
songs that both have ìoftenî (60) contributed to create.
250
The myth of Aeginaís rape by Zeus is known from several other sources. Hesiod
must have told it, but his one surviving verse on Aeacusí birth (fr. 205.1 M.-W.) mentions
neither the motherís nor the fatherís name.62 Pindar narrates the episode in Nemean 8.6f.
and, in some more detail, Isthmian 8.21-3, both Aeginetan odes. Bacchylides does the
same at 9.55-9, for a victor from Phlius; Pausanias relates how the Phliasians also
dedicated statues of Zeus and Aegina at both Olympia and Delphi (5.22.6, 10.13.6).63 The
story was known to Pherecydes, who seems to have related it in connexion with Sisyphus
(fr. 119 Fowler = BNJ 3 F 119). Scenes depicting Zeus engaged in the pursuit of a female
occur on several fifth-century vases, two of which (ARV2 484.21, 536.5) label the object of
the godís attention as Aegina; Sir John Beazley also identified her in numerous other
scenes of the same type.64 On Aegina herself the episode was represented in a sculptural
group found in the area of the temple of Aphaia and roughly contemporary with the
construction of the temple, around 500. In F5 the myth serves as the aition of the islandís
status, as is made explicit at lines 8-12, and of her name, which is strikingly withheld from
the opening � an effect highlighted by ¿Qg_RjV�fR at the very start of the piece � and first
applied to the maiden, not the island, at the start of the mythical section.
os Barbara Kowalzig notes, the opening advertises the song as a hymn to Aegina.65
Her sovereignty over the ìDorian Seaî is expressed by _WkYgb]R (l. 2), a highly marked
62 On Aegina and her Asopid family in the Catalogue see West 1985a: 100-2, 162-4, Cardin 2010. 63 On Phliusí claim to the Asopids see Fearn 2003: 358-62. 64 Brommer 1980: 45; see also Kaempf-Dimitriadou 1979: 22. Lists in Kaempf-Dimitriadou 1979: 93-7, Brommer 1980: 45-7, LIMC I/1 s.v. Aigina (Kaempf-Dimitriadou) with pll. at I/2 s.v. 65 Kowalzig 2007: 201.
251
term that archaic and classical Greek literature reserves for the gods.66 Such status is
further highlighted by her acknowledged role as the receiver of the performance (5-7),
which is likewise typical of divine addressees although also applied to personified places
in the epinicians (see for instance Olympians 5 and 8). Also noteworthy is her syntactical
function as the subject of an essentielle PrGdikation with the copula ëto beí, an uncommon
construction in extant Greek hymnography:67 this arguably highlights that the pieceís
central concern is what Aegina is, and it is explicitly the reason for this that the myth goes
on to explain.
Accordingly, Zeus looms large over the ode. Aegina is his shining star, and the
epithet Hellanios (3f.) specifically recalls his benefaction to all the Greeks at Aeacusí
behest as well as the pan-Hellenic relevance that the episode (supposedly) gave the
island.68 It is to him that Aegina owes her trademark qualities of seamanship and
hospitality, and his love-making with Aegina-parthenos on Aegina-nasos is both the
reason of, and a clear testimony to, his favour towards the latter; the token of such favour
is the birth of Aiakos, the hero-archegete, to whose worship the piece was allegedly
dedicated. If Zeusí creation of the Myrmidons, again at Aeacusí behest, was also
mentioned (21), the first dozen lines of the narrative will have sketched all the main points
of Aeginaís earliest mythical history, testifying to its divine foundation through the agency
that Zeus deployed in each of these.69
66 Similarly Rutherford 2001: 324; see 2 n. 67 A point made by Norden 19232: 182f., kindly brought to my attention by Christopher Metcalf. 68 See p. 206 above. 69 According to the extant sources, the mythical Myrmidones were created by turning ants, _�[_\jW], into men: see p. 206 n. 5 above.
252
Another deity with whom F5 engages, although on a less conspicuous level, is
Apollo. The songs that Aegina will taste are explicitly defined as paeans (l. 5), and Apollo
is quite certainly the most natural addressee for paeans in the first half of the fifth
century.70 The closural prayer for reception (59-61) is addressed to him in his capacity as
HRbnQ, thus construing him as a recipient of the song � and a close associate of the earlier
addressee, Aegina � with the resumption of cRbqTQ and kYjg_Rb from the opening (5, 7).
His was a major Aeginetan cult in archaic and classical times. A scholion states that
fb_¬fRb Ö ]d¦k[R �Q o∞S^Q\b �c¦VVTQ jRÚ ê[fW_b] (| P. 8.94 p. II 215 Dr.), although
it is hard to relate this statement to a particular period. The sanctuary on Kolonna Hill, the
ancient acropolis immediately to the west of the modern city centre, saw three temples of
Apollo built in quick succession around 600, 570-560, and again 520, each bigger than its
predecessor; the building of the third temple coincided with a reorganisation of the whole
temenos, and Pindarís contemporaries will have seen the area as reshaped in those years.71
Just outside the wall of the sanctuary stood an edifice, dating from the same period, which
has been identified with the ìThearion of the Pythian Oneî that Pindar mentions in
Nemean 3.70: although the function of the thearoi to whom it was connected is not entirely
clear, its name suggests that among its concerns there will have been theoriai pertaining to
70 See n. 55 above. 71 The third temple was published by Wolfgang Wurster in Alt-'gina I/1 (1974) with pll. 26-34, updated by Klaus Hoffelner in Alt-'gina I/3 (1999) 65-100 with pll. 61f.; the first and second temples by Hoffelner ibid. 15-64 with pll. 59f. A map of the three temples can be found ibid. pl. 68, one of the larger sanctuary area in Walter 1993: 35 and Walter-Karydi 1994: 126, 2000: 89. The temple of Artemis that stood next to that of Apollo (Paus. 2.30.1), once mistakenly identified with the Aiakeion (see p. 206 n. 9 above), is published by Hoffelner in Alt-'gina I/3 (1999) 101-15. Paus. 2.30.2 testifies to particular Aeginetan devotion to Hecate, whose cult must date from at least as far back as Pindarís old age if Pausanias is right to attribute the cult statue in her temple to Myron, who floruit in the middle of the fifth century (see G. Lippold in RE XVI/1 (1933) 1124); on the cultic association of Apollo Delphinios and Hekate see Herda 2006: 275-9.
253
Delphi.72 Further connexions between Aegina and Delphi are of course to be found in
Paean 6, if it is taken as a single composition. The Aeginetan calendar had a month
Delphinios (| N. 5.81a-b p. III 97 Dr.), in which the city sacrificed to Apollo as g∞jb]fU]
jRÚ kT_Rf^f\] (ibid. a, citing Pythaenetus BNJ 299 F 6) and a characteristic foot-race
called Hydrophoria, also known from other sources, took place in his honour (ibid. b);73 the
scholia also mention the Delphinia, likewise games for Apollo, which it is tempting to
connect to the former (| O. 13.155, P. 8.88, 91 pp. I 386, II 215 Dr.).74
The pieceís explicit self-characterisation as a paean vis-Ø-vis the information
conveyed by the title � whether it relied on an independent source or was extrapolated
from the myth, as F5 lacks an address to Aeacus at the beginning or at the end � raise the
question of what connexions, if any, there may have been between Apollo and Aeacus on
Aegina, or between their respective cults. Mythical connexions are scarce: beside the
abstract notion of god and hero being half-brothers, they are only mentioned together in
Olympian 8.31-43 as the builders, with Poseidon, of the walls of Troy, and that may have
been a Pindaric innovation.75
Cult-related activity, however, presents us with more interesting material. | O.
13.155 p. I 386 Dr. �Q o∞S^Q\b fÏ }WVd^QbR jRÚ o∞njWbR suggests, but does not prove
conclusively, that Delphinia and Aiakeia were alternative names for one contest, which
72 On the Thearion see Hoffelner 1994 and in Alt-'gina I/3 (1999) 135-72 with pll. 71f., 77, Walter-Karydi 1994 and 2000: 92-5, Currie 2005: 333-7, Rutherford 2011; a plan of the Thearion can be found in Walter-Karydi 2000: 91. 73 See also Privitera 1988. 74 On Apollo Delphinios and his cults see Bourboulis 1949, Graf 1979 (with bibliography at p. 2 n. 2), Herda 2006: 272-7. On his Aeginetan cult see Bourboulis 1949: 70-7, reasonably arguing for the identification with the Hydrophoria (74), but see n. 82 below. 75 So claimed Didymus fr. 16 Braswell: | O. 8.41a p. I 247 Dr.
254
would presumably imply a cultic relationship of sorts between Aeacus and (Delphic)
Apollo.76 Another link is | O. 7.156b p. I 232 Dr. �Q kí o∞S^Q\b fÏ o∞njWbR� g� kÓ Ügb�
�_dg[^f\] �SiQ, g” æRVV^_RZg] _Y_Q\fRb �Q fg~] ∏n_�gb]. The amphorites agon,
whose name is lost on the manuscripts, must be the Hydrophoria held in Apolloís honour,
as the diegesis to Callimachusí eighth Iambus testifies.77 The phrasing of the note,
however, leaves some room for doubt over the relationship between the agon and the
Aiakeia, insofar as the elliptic g� kY may conceivably refer to disagreement as to which of
two games Pindar intended as well as to alternative names for one.78
While Pindar and Bacchylides cumulatively mention seven athletes as victorious on
Aegina,79 in only one occasion are the games in which the victory was gained identified
with certainty, as though there was no risk of ambiguity elsewhere; in that one certain
example (Pythian 8.65f.), the specification has a special point in the Apolline connexion
between Aristomenesí victory at Delphi and the Delphinia he had won earlier.80 This may
lead one to suppose that there was only one major Aeginetan agon, to which the three
names that we find in the sources all refer. However, the seven athletes noted above are a
76 This, of course, if the absence of the definite article before o∞njWbR is not an accident of transmission. Drachmann brackets the clause as an interpolation from | O. 7.156b p. I 232, but that scholion � at least in its present form � makes no mention of the Delphinia. In any case, the genuineness (if not the veracity) of the information it contains is not open to question. On the Aiakeia see p. 207 above. 77 Fr. 198 Pf. with PRIMI I 18. Pfeiffer ad loc. (I 195) plausibly restores g� kÓ ’k[gd¦[bR in the scholion; Drachmannís g� kÓ a∞QiQRbR conjures up an unattested name and has little to recommend it. See also Privitera 1988. 78 Controversy concerning which contest Diagoras had won would entail that at least one scholar supposed the fifth centuryís most celebrated boxer to have won a foot-race six times, see the next note. This is considerably more difficult to admit than divergence on the contestís name, but perhaps not wholly beyond an ancient criticís reach. 79 Diagoras of Rhodes (O. 7.86), Xenophon of Corinth (O. 13.109), Aristomenes of Aegina (P. 8.65f.), Telesicrates of Cyrene (P. 9.90f.), Pytheas and Euthymenes of Aegina (N. 5.41-5), and an unnamed Athenian (B. 10.34). 80 O. 13.109 is potentially ambiguous, insofar as o∞Rjbk¬Q fí WÃW[jÓ] ³V]g] can refer to a shrine of the Aeacids (thus, presumably, indicating a victory at the Aiakeia) as well as to Aegina in general.
255
boxer, a wrestler, and two pancratiasts as well as three runners,81 which makes a distinctly
odd record for a foot-race such as the Hydrophoria. This prevents the outright
identification of the latter with either the Delphinia or the Aiakeia.82 Conversely, wholly to
separate Delphinia and Hydrophoria would result in two agones for the same deity
celebrated with all likelihood in the same month of Delphinios. Perhaps, as Photeine
Bourboulis suggested, the Hydrophoria were one contest within, or associated with, the
Delphinia, thus leaving room for victories in the latter that were unconnected to the
former.83 If this was so, the hypothesis of one set of games that honoured both Apollo and
Aeacus and could be named after either seems to suit best the entire spectrum of the
available evidence, although it cannot be proved conclusively. If a festival or rite of such a
sort existed, the complex interaction of divine levels in F5 would suit it quite nicely.
As is the case for most of the cultic Pindar, the date of the ode cannot be established
beyond extremely broad limits, which nonetheless deserve a brief treatment. A convenient
starting-point is Wilamowitzís date of 490, which he supports with (i) a presumed allusion
to the opening of Paean 6a in that of Pythian 6, (ii) the alleged connexion between Paean
6 and Nemean 7, and (iii) the relationship of lines 1f. of F5 to Aeginetan history.84 (i) has
81 Respectively Diagoras, Aristomenes, and Pytheas and Euthymenes (the latterís speciality is indicated at I. 6.60). The point was first raised by Bourboulis 1949: 70f. against the identification of Hydrophoria and Aiakeia, but it also applies to the Delphinia (see the next note). 82 The Hydrophoria will not have coincided with the Delphinia, since the latter could be won by the pancratiast Aristomenes, albeit in the pentathlon (P. 8.66 with | P. 8.88, 91 p. II 215 Dr); nor can they have coincided with the Aiakeia, if the scholion is correct in claiming that these were won by the pancratiast Pytheas (| N. 5.78c p. III 97 Dr., but note k\VRkq, which suggests conjecture rather than testimony). 83 Bourboulis 1949: 72-4. 84 Wilamowitz 1901: 1286f. (i), 1908: 349 (ii), 349f. (iii). Other dates for Pae. 6 have been proposed over the course of the last century, but on no more persuasive grounds than his: Farnell suggests ìa date before 480î but later than 490, when Pindar was famous enough to be commissioned such a grand Delphic ode (1930-2: II 402); Theiler assumes 469 after Hermannís date for Nemean 7 and backs his dating with supposed stylistic
256
been shown to be illusory,85 and (ii) is inconclusive, due to serious doubts over the
existence of the purported relationship as well as to the lack of a secure date for Nemean 7
itself.86 (Both arguments also presuppose the unity of Paean 6, without which no argument
for the date of 6a bears on that of F5.).
and metrical criteria (1941: 16, 36); Finley proposes ìthe period, roughly, from 478 to 460î on the basis of thematic affinity with odes dated or datable to that period (1951, quot. from p. 71); Radt follows Finley, mentioning 467 after Hermann (1958: 92f.); Hoekstra dates the ode to 490 on the basis of a rather implausible reconstruction of its composition and performance scenario (1962: 13f.); Bowra suggests the 470s or 460s, possibly 467 with Hermann (1964: 410f.); Fogelmark ìabout 467î after Theiler and Finley, on the dubious authority of the presence of colour epithets in Pae. 6 and N. 7 (1972: 43-8, quot. from p. 43); Ferrari hypothetically dates the consecration of the Aiakeion to around 490, and F5 accordingly (1999: 157f.); Kurke connects Pae. 6 with the Aeginetan dedication of the �j[gp^QbR of Salamis at Delphi described by Herodotus (8.121.2-122), therefore not long after 480 (2005: 117f.); Kowalzig (2007: 181-223 passim) assumes that it was composed after the Persian Wars, with no further specification. 85 Wilamowitz 1901 predates the publication of H4, when all that was known of Pae. 6(a) was ll. 1-6 quoted by Aristid. Or. 28.58 Keil. In his view, P. 6.1-3 §Vbjicbkg] �d[gk^fR] £ ³[gP[RQ ¢ mR[^fTQ £ �QRcgV^�g_WQ would have constituted an explicit reference to the invocation to Aphrodite and the Graces in Pae. 6(a).3f. V^]]g_Rb mR[^fW]£]^ fW jRÚ ]ˆQ �d[gk^fRb, which the audience would have heard just a few days earlier at the same festival, the Pythia; inscr. P. 6 p. II 192 Dr. dates the Pythian victory to the twenty-fourth Pythiad (that is, probably, to 490: the starting date for the reckoning of Pythic games is debated, but see Mosshammer 1982, Currie 2005: 25f. and Finglass 2007b: 19-27), thereby providing a secure date for the fragment as well as for P. 6. However, it was soon pointed out that �QR- in �QRcgV^�WbQ indicates thorough ploughing, not ploughing anew (Farnell 1930-2: II 184, Finley 1951: 61f., Radt 1958: 91 n. 1); Segal 1967: 432f. n. 4 suggests that the joint occurrence of the Graces and Aphrodite is too topical to indicate a specific reference. Moreover, the fuller text of the ode on H4 suggests that Pae. 6(a) was performed at the Theoxenia (61f.), which took place well before the Pythia (Wilamowitz 1908: 345, 347, Radt 1958: 83, 90-2), thus undermining the identity of performance context that Wilamowitz had assumed. In any case, an intertext can provide at best a terminus ante quem, not a firm date, for the Paean. 86 The scholia mention � a unique occurrence for a Nemean � the fourteenth or twenty-fourth Nemead, 547 or 527 (B and D respectively at inscr. N. 7 p. III 116 Dr.): both dates are far too early for Pindarís career and must therefore be corrupt, unless one accepts Corsiniís intriguing but unsubstantiated hypothesis that the numbering of the Nemean games started anew after the Persian wars (1747: 52f. after hyp. N. c p. III 3 Dr.), which would presumably give 453 or 451 on the figure transmitted by B (that transmitted by D would be too late for Pindar on this reckoning). Emendation has been variously attempted: Hermann reads Qk� = 467 (in Boeckh 1811-21: II/2 416), Gaspar _R� = 493 (1900: 40), Wilamowitz _k� = 487 (1908: 344f., misprinted ì485î, see Theiler 1941: 16 and n. 1), Bornemann Q� = 475 (1928: 148), Loscalzo _p� = 477 (2000b: 30-2, misprinted ì475î), Rutherford QR� = 473 (2001: 331 n. 95). MacYa 1976: 200-4 supports Hermannís dating with questionable metrical arguments; Sbardella 2007: 76-9 sees in N. 7.32-4 a reference to the coup attempted by Nicodromus (Hdt. 6.88-93) and dates the ode ëclose to 485í, but evidence for this historical allusion is very tenuous indeed, and the coup itself is not securely dated (ibid. 79 n. 36). On the ëapology theoryí see p. 233 n. 19 above.
257
(iii) proved to be the most influential part of Wilamowitzís argument;87 for this
reason, and because it arguably entails a degree of misrepresentation of the text, it is
expedient that it be discussed at greater length. In his view, }T[bW~ £ _WkYgb]R c¦QfTb
(lines 1f.) and QRPc[�fRQbQ £ kR^_gQR (8f.) could not belong after the battle of Salamis, by
which time Aeginaís naval control of the Saronic gulf (the ëDorian seaí) had already
started shifting towards Athens. Two considerations go against this. Firstly, it requires
distorting the meaning of _WkYgb]R: as noted above, the connotation of the word is not
political or military, but markedly sacral, and there is no reason to suppose that this
passage be an exception. Quite the contrary: Pindar portrays Aegina as a Lady of the Sea,
whose relationship with the Saronic gulf is a quasi-divine sovereignty of the same
description as that of Hermes over Cyllene or Aphrodite over Cyprus. Although the image
clearly rests on the actual or assumed might of Aeginaís fleet, nothing marks such might as
military, and its representation is on a different scale altogether from a straightforward
historical record. The same goes for lines 8f.: the attribution of the adjective to Aeginaís
kR^_TQ rather than to herself arguably shifts the viewpoint from that of hard facts to a
higher sphere, one presumably more significant in a cultic context and rather less
dependent on strict factual verification. Pindarís words clearly phrase an Aeginetan claim
to a god-like sea-lordship � one that Aegina need not have actually held by any objective
reckoning when the ode was composed and performed.88
87 It is accepted by WOst 1967: 119-21, Bernardini 1968: 138, and with remarkable enthusiasm by Bruno Gentili (1979: 7 n. 1 = 1981: 103f., 20064: 214 n. 61, and in Gentili et al. 20064: 184 n. 1). 88 MacYa 1976: 193f. makes the same point rather more radically (ìPindarís indifference to objective reality is a well-known factî) but not altogether wrongly as far as our case is concerned.
258
Secondly, it is hardly legitimate to set such a strict end-point to Aeginaís claim to
maritime power as 480.89 Whatever the increase of Athensí naval power after Salamis, a
relative diminution of Aeginaís control of the Saronic gulf does not mean its complete loss,
or the end of her claim to it in such terms as Pindar uses in the proem. At Salamis she had
obtained a resounding success: with thirty triremes, hers was the third-largest Greek
contingent after those of Athens and Corinth;90 Aeacus and the Aeacids, her patron heroes
(as it were), had been summoned from the island and presided over the victory;91 in the
fight it was the Aeginetans who ¶jgP]RQ ³[b]fR (Hdt. 8.93.1).92 On the level of
international recognition, their �[b]fW^R was sanctioned by none other than Apollo, with
his oracleís demand for a further, specifically Aeginetan dedication beyond the communal
�j[gp^QbR of the booty.93 As a result, Aeginetan propaganda could easily claim Salamis as
89 So Finley 1951: 65f., Puech 19312-512: IV 116f., Radt 1958: 93. The issue may have arisen from Eusebius, on Diodorusí authority, giving the ten years until Xerxesí invasion as the time during which the Aeginetans were thalassokratores (D.S. 7 fr. 11 Vogel in Eus. Chronicon pp. I 321, II 206 Aucher = pp. 107, 191 Karst, cp. Hier. Chron. p. 107 Helm, Syncellus p. 296 Mosshammer). Origin and date of his ëthalassocracy listí are disputed, as are many of its details (Myres 1906 with Fotheringham 1907 and Myres 1907, Helm 1926, Forrest 1969: 95-106, Miller 1971, Jeffery 1976: 252f., Huxley 1982: 193-6; Miller 1971: 43f. and Figueira 1993: 48f. on the Aeginetan section); since the count goes no further than the second Persian war, it is a reasonable assumption that overall sea-power in the Aegean was implied to have been in Athenian hands afterwards (so Jeffery 1976: 252f.), as is suggested also by Plu. Them. 4.1. Diodorus does refer to the Athenians as pRVRffgj[RfgtQfW] at various stages of the Peloponnesian war (11.70.5, 12.3.3, 4.1, etc.), but it is to all the Greeks jointly that the term applies in the immediate aftermath of Salamis (9.19.7). It goes without saying that a historiographical model based on exclusive power held for definite periods and changing hands at specific dates � whether such model is attributed to Diodorus, or rather to Eusebiusí need to present schematically the information he drew from him � can hardly be a reliable representation of the complex political and military situation of the Aegean during the archaic and early Classical period. 90 Hdt. 8.46.1, see Figueira 1981: 32f. 91 See p. 207f. above. 92 Similarly Ephor. BNJ 70 F 188 in | I. 5.63 p. III 247 Dr., D.S. 11.27.2, Ael. VH 12.10. Most of the later Attic sources claim the prize for Athens: Asheri (ed.) 2003: 323. 93 Hdt. 8.121.2-122. The veracity of this tale is disputed as maliciously anti-Athenian by Plu. De malignitate Herodoti 40 = Mor. 871c-d and has been questioned by Asheri (ed.) 2003: 323, but see Fontenrose 1978: 320, Crahay 1956: 331f. A similar effort on Aeginaís part to memorialise her role in the Persian Wars abroad is testified by Hdt. 9.85.3: in 469 the Aeginetans had their proxenos at Plataea build a cenotaph for their soldiers who had allagedly perished in the battle ten years earlier (see Figueira 1993: 212 and n. 51).
259
an Aeginetan victory, at least on home ground, as in Isthmian 5.47-50.94 Far from flagging
the beginning of the end, Salamis will rather have boosted Aeginaís claim to seamanship,
whether or not her navyís subsequent exploits could match those of the past.
Despite Athensí tightening grip on the Aegean in the following two decades, the
period saw no major setback that could make a serious case against Aeginetan sea-power
and prevent Pindar from focusing his praise of the island through the customary lens of
seafaring: ìsuch praise of Aeginaís, to Pindar, righteous power is not out of place even
when, it is now clear, that power was waningî (Finley).95 On the contrary, consciousness of
the previous successes of their fleet and confidence in its present condition are repeatedly
noted by Diodorus (11.70.2, 78.3f.) as a key feature of Aeginetan attitude in those years.96
When open war with Athens broke out in 459, Aegina could indeed muster a massive fleet:
at the decisive sea-battle the Athenians captured seventy of her warships, more than twice
as many as she had deployed at Salamis; the total originally at her disposal must have been
considerably higher, since several ships must have been sunk in the battle and others still
survived it in Aeginetan hands.97 It was not until 457 that Aegina was blockaded, besieged,
and subjected: the city walls were torn down, what remained of her fleet was surrendered,
94 Rightly referred to Salamis by | 60a-c, 63a p. III 247 Dr.; the view put forward in 60a that the expression refers to the Peloponnesian War sounds harder to entertain. 95 1951: 65, emphasis added. Figueira 1981: 178-80 plausibly argues that an economic downturn may have struck the island, but one must note that the thirty-talent tribute paid by Aegina after her submission � an enormous sum by any account � was the largest in the Delian League (Hornblower 2007: 290). This suggests that the islandís finances cannot have been in too dire straits at the time: Athens may well have been particularly overbearing towards her former sworn enemy, but she will hardly have demanded and obtained what Aegina could not afford to give. 96 See Figueira 1981: 167, 1991: 106-9. 97 Figueira 1981: 29-31, 1991: 113. Date: Figueira 1991: 107f. n. 9, Hornblower 2004: 222 and n. 373 after Lewis in CAH v2 (1992) 500f. Seventy warships: Thuc. 1.105.2, Lys. 2.48, DS 11.108.4.
260
and a hefty tribute was imposed on her.98 Only then, with no fortifications and no navy, and
after a humiliating act of submission to Athens, Aegina presumably had to resign for good
her claims to sea-power in theory as well as in practice.
This being the case, a terminus ante quem for the imagery of maritime dominion that
shines through the opening of F5 cannot confidently be set any earlier than 459: the text of
the ode as it stands is equally compatible with the ëthalassocracyí of the 480s, with the
glorious days after Salamis, or with the boastful and ultimately misguided self-assurance
that preceded the downfall.99
98 Th. 1.105.1-4, 108.4 with Figueira 1991: 104-13. Knoblauch 1972: 83f. argues from archaeological evidence that the harbour and arsenal too were demolished. 99 Bona 1988: 138 contends that the poem could be even later, with Pindar ìrecalling as still present [Aeginaís] past splendourî, but this does not sound particularly likely: although lines 1f. and 8f. do not constitute an actual account of the islandís power, they unquestionably represent an explicit claim to it � one which would have been all but grotesque in the grim servitude of the 450s or 440s.
261
Commentary
1 ¿[Z�_cVXgTc : ìfame is both the premiss and the result of the present encomiumî
(Rutherford 2001: 324); Aeginaís name is so famous that, as it turns out, it need not be
given at all. The adjective occurs sporadically in epic and other early poetry (Il. 22.51,
H.Ven. 111, 146 with Faulkner 2008: 189, H.Merc. 59, Semon. fr. 7.87 W.2, Ibyc. PMGF
306, etc.), and can be traced to an Indo-European formula, see Schmitt 1967: 60-3. The
feminine form was rare enough to warrant a quotation in | bT Il. 22.51e Erbse as an
example of the composition of a neuter and a feminine constituent, against those who
objected to taking the masculine form as one compound (]�QpWfgQ) on the grounds that a
neuter and a masculine would not go together. Grenfell and Hunt 1908a: 99 compare
QRP]bjV�fRQ at N. 5.9 (the masculine termination for the feminine occurs at I. 9.1f.).
The accent in H4 is consistent with Herodianís doctrine that compounds of a ëdeclinableí
(cfTfbj¦Q) and jVPf¦] are proparoxytone (pp. I 229, II/1 69f., 119 Lentz, cp. EM col. 616
Gaisford), whereas the medieval mss. of the Qpinicians regularly write the compounds of
-jVPf¦] as juxtaposed (O. 8.52, N. 5.9, data from Mommsen 1866 and therefore
unavailable for I. 9.1); Bergkís kRb��½���[¦Q (18784: III 716) is not the true reading at fr.
dub. 333a.7 Sn.-M., see DíAlessio 2000a: 243. Likewise, it is only at Il. 22.52 that the mss.
(except h) show this accentuation: in all the other epic occurrences they are split between
the oxytone ¿Qg_RjVPf¦] and the juxtaposed ƒQg_R jVPf¦].
262
Schroederís ¿QP�_RjV�fR is attractive. Pindaric usage by and large favours the form with
upsilon in ƒQP_R and compounds: all the mss. at P. 2.44, most at 7.6 and 11.6, and a
majority at O. 6.57; omicron and upsilon appear in equal balance at P. 1.38, while omicron
prevails at O. 9.46 and is the only reading at P. 12.23: see Mommsen 1866: 126. Hermann
1809: xxi relates this variance to the metre, Aeolic (where upsilon prevails) or dactylo-
epitrite (where omicron does), but even if right, this gives little help here. Radt 1958: 174
defends the transmitted reading as an epic borrowing (which it is in the dactylo-epitrite
odes if Hermann is correct), which may also be correct, perhaps with a view to raising the
tone of the opening, as Thomas Coward suggests to me.
t��� : Rutherfordís reading is unavoidable: the third letter is genuinely uncertain (only a
vertical remains), but the bottom-left curve that survives from the second belonged more
probably to alpha than to epsilon, and the acute visible above it leaves little room for
alternatives. Furthermore, the corrupt Sn[ �]fb in | bT Il. 22.51e Erbse is better explained
by Sn[ �]]b than by the awkward Sí ¯QW]]b that Grenfell and Hunt initially read (Radt
1958: 173f., 28*, DíAlessio 1997: 57 n. 193) , the less familiar form of the second person
being corrupted out of context into the more familiar �]fb.
The presence of Sn[ in this position is the biggest obstacle to proponents of the original
independence of F5 from Pae. 6(a). There is no example of a Greek text beginning with
Sn[ earlier than D.Chr. 20, 67 (the latter is a dialogue); the elegiac parallels alleged by
Ferrari 1999: 152 n. 11 are all from fragmentary texts (Xen. fr. 1 W., Thgn. 441, Tyrt. fr.
10.1 W.), where nothing guarantees that the beginning of the surviving part was also the
beginning of the poem, although he may be right to suggest that Sn[ here is primarily
263
anticipatory (GP2 70-2; cp. 5 g—QWjWQ). If one broadens the scope of inquiry to other
particles, however, some parallels can arguably be found. The existence of inceptive kY in
archaic sympotic elegy is disputed (see Verdenius 1955: 17 for some possible examples),
and in any case it pertains to an essentially dialogic context which can hardly be applied to
F5. However, the newly published P.Mich. inv. 3250a recto col. iii (see Borges and
Sampson 2012) seems to indicate that a poem by Simonides began f\Q^jR kí ¯R[, again
with inceptive kY. This is the only relatively certain example of a floating connective at
poem-beginning known to me in archaic or early classical choral lyric, but other genres of
literature in later periods provide more plentiful material. Xenophon seems to have been
singularly fond of starting off a book with a connective (Oec. Ap. kY, Smp. Lac. �VV(n), kY
also at [Xen.] Ath.); much later, beginning with a connective became a relatively frequent
styleme in Latin poetry, cp. Prop. 1.17 (et), Hor. epod. 5 (at), Ov. epist. 12 (at), am. 3.7
(at), Priap. 80 (at). How much of a parallel any of these can be for a fifth-century cult
poem is very dubious indeed. It is not possible to exclude with absolute certainty that the
performance of F5 was started off by an introitus of sorts uttered before the song, not
unlike the Gloria in the sung Roman Catholic liturgy, and that Pindar took advantage of
this format to construe the song itself as a response to it, but this is little more than wild
speculation, and one must observe that the rest of choral lyric (cultic or non-cultic) fails to
provide a single parallel for this scenario. Still, the new Simonidean incipit provides one
straw to clutch at to argue for the possibility that, for reasons beyond recovery, a piece of
lyric poetry could begin with a floating connective, especially if the anticipatory use of
Sn[ weakens its normal backward connective force.
264
It must be noted that whatever its relevance to the question of the original relationship of
F5 to Pae. 6(a), the particle constitutes less of a problem for the independence of F5 within
the Alexandrian edition itself. Hellenistic criticism had a notion that poets, especially
Homer, could �cÙ fgt Sn[ ³[ZW]pRb: | A Il. 2.284a, 7.328b, 17.221, 18.182c, 23.627a, b
2.284b, 7.328c2 Erbse, | M Od. 1.337a1, H 4.722c Pontani, 10.174, 189, 501, 14.496
Dindorf, | E. Ph. 886 Schwartz, etc. True, none of the occurrences so glossed is a real
parallel to this Pindaric line: most of them follow a vocative and introduce an explanation
of the address (GP2 69, cp. A.D. Conj. in GG I/1 pp. 239f. Schneider, Eust. ad Il. 17.221 p.
IV 43 van der Valk). However, | Od. 14.496 alleges this usage to justify the athetesis of
verse 495, which produces an inceptive Sn[ identical to ours. Together with the
consistently general phrasing of the principle, this anonymous conjecture suggests that Sn[
at the start of a piece would have been of no excessive trouble to (some) cultivated readers.
�h]YOw : a ìDorian seaî is not attested elsewhere. Wilamowitzís inference that it indicates
the Saronic Gulf (1908: 349) is reasonable, although the epithet significantly eschews
geographical precision in favour of a broader point (ìa sea that is Dorianî). The
assumption that ëDorianí implies ëunder Aeginetan naval controlí (Wilamowitz, WOst
1967: 121) is unwarranted and over-restrictive: the Dorianness of the gulf is predicated on
that of the inhabitants of much of its coast, Aegina first and foremost, not of the warships
that sailed its waters. The historical significance of the phrase lies rather in its highlighting
the ethnic connexions (and hierarchy, see 2 n.) between Aegina and her mainland
neighbours. The Dorian ethnicity of Aegina and her inhabitants is a commonplace of
Pindarís Aeginetan odes: O. 8.25-30 fnQkí ÚVbW[jYR Zi[RQ Ö }T[bW~ VR�b
fR_bWPg_YQRQ �X o∞Rjgt, N. 3.3 }T[^kR Q¬]gQ oÒSbQRQ (with Pfeijffer 1999: 246f.), I.
265
9.2-4 QbQ Ö }T[bWˆ] �VpàQ ]f[Rf¦] £ �jf^]]Rfg, cp. P. 8.20. Especially worthy of
comparison is O. 8.30, where �X o∞Rjgt (prima facie ëfromí as much as ëafter the time of
Aeacusí, see DGE s.v. �j A.VI.a) appears to cast Aegina as the archetypal Dorian territory
by explicitly associating Dorian identity with her traditionally autochthonous founding
father in a seemingly unbroken continuum (Pfeijffer 1999: 246; contrast the reference to
the Return of the Heraclidae at I. 9.2f.), although fR_bWPg_YQRQ may fall short of outright
identification between Aeacid and Doric heritage (Figueira 1981: 173). Other sources also
attest the islandís Dorian or, more specifically, Argive pedigree: Hdt. 8.46, Paus. 2.29.5,
Strab. 8.6.16, | O. 8.39a-b, P. 8.29a-b, N. 3.5 pp. I 246f., II 209, III 43 Dr. On Aeginaís
early history see Welter 1938: 7-30, Figueira 1981: 170-92, 1983, Walter 1983; on the
Aeginetansí ethnic self-consciousness as Dorians see also Walter-Karydi 2006: 82-4.
2 _[O]N`ZYa� : a typical hymnic motif, signposting a special relationship of concern and
authority between a deity and a place, usually a centre of his or her cult (see KeyÜner 1932:
75-7; Benveniste 1969: II 130 is led somewhat astray by his neglect of the post-Homeric
examples). _YkTQ refers to human rulership in the epic formula ôSqfg[W] †kÓ _YkgQfW]
(cp. personal names such as ÙYkTQ, ÙYkgP]R, and their compounds), but throughout
Greek literature the form with one more epsilon is regularly applied to gods and goddesses
alone: its only recorded reference to a human being at Call. Jov. 86 (Ptolemy Philadelphus)
carries a potent intimation of divinity, one of several in that hymn (McLennan 1977: 122).
Such an intimation is therefore to be perceived here too, see p. 257 and 9 n. Against the
overtly political and military interpretations of this passage that have been put forward in
the past (see p. 257), one should also note that the seamanship that Pindar brings to the fore
time and again in his expression of Aeginetan ëselfí-portraiture (see 9 n.) is never explicitly
266
phrased in terms of dominion. The (possible) oblique military allusions in O. 8.20
kgVbZq[Wf_gQ and N. 5.9 W–RQk[gQ (Hornblower 2004: 212f., 226f., 2007: 295, 291),
when seen in the context of Aeginaís undisputable military might (see pp. 258-60 above),
make this fact all the more noteworthy, and demand that a similar obliquity is seen here:
_WkYgb]R only intimates the islandís naval power by expressing her God-given, quasi-
divine sovereignty over the ìDorian seaî.
In the Iliad only Zeus is _WkYTQ, once over Dodona (16.234) and four times from Ida
(3.276, 320, 7.202, 24.308), see Lejeune 1939: 75f. Later authors extend the usage to other
deities and to spheres of action beyond the purely geographical (Lehnus 1979: 115 n. 44),
while only the specification of the place over which rule is exerted remains productive:
Homerís ºk\pWQ itself is so interpreted in some quarters of Hellenistic scholarship (see
Lejeune, add | A Il. 3.276a, 320, 7.202a Erbse). Phrases with _WkYTQ or _WkYgP]R also
occur in inscriptions, and may have been part of certain deitiesí official cult titles (e.g.
�p\Q¬ �p\Q�Q _WkYgP]R): Schlageter 1909: 15, Rumpf 1936, Barron 1964: 35-7, 41. The
participle is usually construed with a genitive of command (fr. 95.1 Sn.-M. with Lehnus
114f. and n. 40, Od. 1.72, H.Merc. 2 = H.Hom. 18.2, H.Ven. 292 ~ H.Hom. 10.4, Alcae. fr.
308.1, 354 V., lyr. adesp. PMG 887.1, etc.), and less frequently overall � but in two of
Pindarís three occurrences of the participle � with a dative of place such as we find here
(Radt 1958: 174 n. 1, Lehnus 115f.): O. 7.87f. Qifgb]bQ �fR�P[^gP £ _WkYTQ, Hes. Th. 54,
Maced. AP 6.30.8.
267
o 2 : on the restoration of the scholion see Rutherford 1997a: 6-8; parallels for dY[WfRb
indicating textual transmission, ibid. 9-12; on the implications of the scholion for F5 and
its status in the Pindaric editions, pp. 12f. and 228-44 above.
3 [laZaQ : intriguingly, the passage corresponding to Apollod. 3.12.6 in ms. S of the
Epitome relates that Aegina was called Nasos until she was renamed after the daughter of
Asopus (text in Papadopulos-Kerameus 1891: 184). Most sources, including the fuller text
of Apollodorus, give the earlier name as Oenone, some Oenopia (a list in Escher, RE I 968
(1893) s.v. Aigina 2); Pindar has the former at N. 4.46, 5.16, 8.7, I. 5.34, the latter at I.
8.21. Nasos seems to be unique and, given the Doric form, can hardly be blamed on
corruption. One might wonder if the variant goes back to a tradition which Pindar knew
and on which he is playing here, with the amusing twist of an island called ìIslandî being
famous in name. But it is also possible that its ultimate source is an interpretation (not
necessarily a plausible one) of our very passage.
�YÙa �X£Xc[UZe : Zeus Hellanios is the Zeus of Aegina and of Aeacus: inseparably tied to
both and therefore quintessentially epichoric at the same time as he claims pan-Hellenic
status and asserts theirs (Kowalzig 2007: 219). His temple was founded after Aeacusí
successful intercession for rain on behalf of all the Greeks, who had sent envoys to Aegina
to supplicate him (see o 3 n.); unsurprisingly for an early cult of Zeus, it was located atop
the islandís highest mountain, Mount Hellanion (HRQWVVqQbgQ in Paus. 2.30.4), now Oros,
which folk lore linked to weather phenomena by the late fourth century (Thphr. Sign.
1.24); some remains of the complex survive (Kowalzig 204-6 with bibl.). On the cult see
Cook 1914-40: II/2 894f. with III/2 1164f., Welter 1938: 91f., Walter 1993: 84-7, Kowalzig
268
2007: 181f., 201-5 (on mountain cults of Zeus in general, Cook II/2 868-987 with III/2
1161-77). The specific mention of this epithet highlights from the very beginning of the
ode Zeusí favour towards Aegina and, with special significance if F5 was indeed employed
in the cult of Aeacus, towards the hero-archegete himself.
o 3 : the myth is narrated more fully in | N. 5.17b p. III 91 Dr., | Ar. Eq. 1253a-b Jones-
Wilson, D.S. 4.61.1, Paus. 1.44.9 and 2.29.7f., Clem.Al. 6.3.28.4-6, cp. | N. 8.19a p. III
142 Dr., Apollod. 3.12.6, and Isocr. 9.14f. (who, however, links it to the foundation of the
Aiakeion, not of the temple of Zeus Hellanios). The pan-Hellenic embassy to Aeacus for
his intercession with Zeus was portrayed on the reliefs of the Aeginetan Aiakeion, see pp.
206f.
Radt 1958: 175 argues that the definite article before RÃZ_gt implies an earlier mention of
it, either in the text (that is, in the three-triad Pae. 6) or in another scholion, but this is not
necessarily the case. The article only implies that the drought was taken as known by the
person who penned the note; marginalia typically stem from separate exegetic material
summarised in the margin for ease of reference, and the easiest explanation of fgt is the
narration of the myth in the hypomnema that served as the source of the note (cp. fgt
RÃZ_gt at | Ar. Eq. 1253a Jones-Wilson, where no other mention of the drought appears).
Radt is also puzzled by W–X�[Q]fg, since in the other sources only Aeacus prays: the plural
may be due to a misunderstanding of the source (id.), to an account of the episode later in
the poem, to conflation of this prayer with those at N. 5.9-13 and/or 8.8-12 (Rutherford
2001: 332 n. 99) or with the similar one found at Pae. 6(a).63-5 � or, perhaps more
269
probably, to the simple inference that when Aeacus prayed on their behalf the Greeks too
were praying with him.
4 xcO[[Ù[ �aT]Z[ : Pindarís use of metaphors drawn from the sphere of light and
brilliance is well served in secondary literature: one must mention Gundert 1935: 11-13,
Bowra 1964: 35f., Bremer 1976: 231-310 esp. 280-310 (311-13 on Bacchylides), AdorjUni
2011: 78-80, to all of which the reader should refer for parallels and for a more detailed
treatment of the various nuances that the individual images convey. What is relevant for
this passage is that both brightness and stars, individually or combined, signify impressive
visibility, with a particular emphasis on visibility from afar. Henry Spelman points out to
me that comparison with a star carries an additional implication of stability and
permanence, such as the otherwise similar comparison to fire (see *F7.4 n.) need not have;
arguably it also suggests the extent to which people can (and do) rely on it for guidance.
Radt 1958: 175 notes the physical implications of our phrase: Aegina is a white
Mediterranean island shining on the background of the dark-blue sea just as Delos is a
f\VYdRQfgQ jPRQYR] ZpgQÙ] ³]f[gQ at fr. 33c.6 Sn.-M. (Wilamowitz 1913: 130f, Snell
19754: 89), where the image is strengthened by a pun with the islandís two names
(f\VYdRQfgQ ~ }vVg], ³]f[gQ ~ �]fW[^R). If a similar view from above is to be
perceived here too (so Bremer 241), one notes the possible re-interpretation of the genitive
}bÙ] �VVRQ^gP as subjective and the ensuing focus on divine perspective over Aegina and
her affairs.
5 Z—[OVO[ : resumes Sn[, cp. P. 9.90-3 (Radt 1958: 173 n. 2): ìbecause of the islandís
fame, the singers will praise itî (Rutherford 2001: 324).
270
dcY�i[h[ : on the form see *F6(b).4 n.; on the possible implications of calling this song
a paean, pp. 252-5.
6 �NZ]dZ[ : Farnell 1930-2: II 408, cited with approval by Rutherford 2001: 324, relates
the striking use of this verb to the custom of singing a paean at banquets (see Rutherford
50-2), but this is quite irrelevant. It is more probably to be connected to the sacrificial
aspect of cultic song, rightly emphasised by Rutherford 170, 324f. n. 75 (see also Depew
2000). Radtís remark (1958: 175f. and n. 3) that the equation between song and food,
unlike that between song and drink, is unparalleled is only true, so far as I can tell, of epic
and lyric (but see Alcm. PMGF 17, 95(b) with Nannini 1988: 26-35): in comedy, works of
poetry are frequently compared to a meal (Taillardat 19652: 439-41, Pellegrino 2000: 12f.
n. 11, Imperio 2004: 216-18), and Aeschylus allegedly compared his tragedies to ìslices of
the great dinners of Homerî (Ath. 7.347e; West 1985b: 78 n. 25 plausibly suggests that the
anecdote goes back to Ion of Chios, but a comic origin is not impossible).
OÃ[WxZ_O[ : one is strongly tempted to relate the verb not to sleep but to a different
activity, likewise traditionally associated with the WÃQq, to which the mythical narrative at
12ff. unmistakeably refers (so WOst 1967: 135 and Rutherford 2001: 324f. n. 75, who very
decently speak of marriage). If so, gà will go with ³kg[cgQ, ìwe shall put you to bed with
a dinner of paeansî (cp. Sandys 19192: 543). But while the middle WÃQn�g_Rb regularly
implies intercourse after Homer (see LSJ s.v., already at Od. 5.119, H.Ven. 190, cp. P.
3.25), the active never does: it refers either concretely to sleep, or metaphorically to the
cessation of things. The former is unlikely, however much favour it may have enjoyed with
earlier translators (Grenfell and Hunt 1908a: 99, Sandys, Farnell 1930-2: I 308, Bowra
271
1964: 149): the banquet metaphor alone hardly justifies the idea that Aegina will go to
sleep after the hymn, and to suppose that the performance would have taken place at
nightfall (so Radt 1958: 176) looks like begging the question. (Callimachus, however, may
have taken it this way if some connexion exists between F5.5-7 and Del. 302f. g–fW
]bTc\VUQ g–fí ³°gdgQ g“Vg] �pW^[Rb] £ ²]cW[g], �VVí R∞W^ ]W jRfR�VYcWb
�_db�¦\fgQ.) For amusement more than instruction I mention Fraccaroliís charmingly
bathetic ìsenza cena di canti a letto mettertiî (1913: II 443 with n. 2 ìQui Pindaro tocca
terraî: either he or his criticsÖ). Puechís ìne permettrons nous pas quíØ ce festin tu
assistesî, glossed with a reference to theoxenies, lectisternia, and a ìlit de paradeî (19312-
522: IV 124 and n. 3), is unlikely: an WÃQq is not a sympotic jV^Q\. Therefore, unless Pindar
is using the active with a connotation that recalls that of the passive (which is possible),
Race 1990: 61f., 69f. may be close to the truth in taking gà with the verb and comparing
such passages as I. 4.20-2 f¦QkW cg[àQ Ö —_QgQ £ �j VWZYTQ �QnSWb dn_RQ cRVRbnQ
Ö �Q —cQTb SÏ[ cY]WQ and 7.16f. cRVRbÏ Sn[ £ W—kWb Zn[b], where sleep signifies the
loss of fame due to lack of continuing praise, to be remedied by the present song: ìthe
hymnist assures Aegina that she will not be abandoned in silence, nor will she be without
her share of celebrationî (62). From a unitarian standpoint, as Bruno Currie points out to
me, the image may have signalled the change of subject and addressee between the second
and third triads of Pae. 6: ìI wonít end the song without giving you your shareî.
7 �ifYc : 5¦pg] and related terms can denote waves (A. Th. 362, S. Ph. 688, E. Cyc. 17,
etc.), their dashing noise or that of oars (Od. 5.412, [A.] PV 1048, E. IT 1133, etc.), or the
sound of several people speaking all at once (Hes. OD 220 and parallels cited by West
1978: 211, add F11.16, A. Pers. 406, 462): see PCron 1974: 169, 239f. Which, if any, is the
272
original meaning is debatable; the otherwise logical progression from the first to the third
is contradicted to some extent by the relative chronology of their surviving attestations.
Here the sense is probably the third, which Pindar arguably tinges with a nuance of the
other two by evoking the image of ìwave-washed Aeginaî (Grenfell and Hunt 1908a: 99,
see Renehan 1969: 225f.). I doubt whether the ëstream of songí and related images (N.
7.11f., 62f., I. 7.19; irrigation, P. 5.99f., 8.57, I. 6.21, fr. 6b(f) Sn.-M.; etc.) truly come into
the picture, as Race 1990: 71 suggests: in all those cases the vehicle of the metaphor is
fresh water, with all its nurturing associations, not the sea, which indeed can be explicitly
opposed to it (Steiner 1986: 72f.; O. 10.9f. is no counter-example, see Viljoen 1955: 49-51
cited by Verdenius 1988: 61).
NOVZ_`[c : the medieval tradition of the Epinicia is split almost evenly between kYjg_Rb
and kYZg_Rb: the former is transmitted by all mss. at N. 58, all but one at O. 5.3, almost
half at P. 5.80, and is clearly the true reading at O. 2.63 (most mss. have kY[jgQfRb, none
kYZgQfRb); the Attic form predominates at O. 13.68, 92, P. 1.98, and is the only reading at
I. 1.51 (data from Mommsen 1864 ad locc.). Evidently difficilior, the form with kappa
must be the original reading wherever it is found.
Receiving a song is what its divine dedicatee does or is asked to do, see Depew 2000: 62-5.
By overlooking this convention, Hoekstra 1962 over-interpreted the image (Aegina
receiving the song instead of sending it) and built upon it a whole background story of how
the Aeginetans would have traditionally sent a paean-chorus to Delphi, but on the occasion
of ëPaean 6í were somehow unable to do so, and therefore are praised in the ëthird triadí
by the chorus of Delphians that stepped in for them; Wilamowitz 1908: 345, 1922: 134f.
273
had imagined the exact opposite, that the supposed Aeginetans of ëPaean 6í were making
up for the absence of the traditional local chorus.
VcTO]Owa : not ìproclaimî (Slater 1969 s.v. jRfW[YT) as much as ìtell at length and in
orderî (LSJ s.v. jRfRVYST, cp. LfgrE s.v. VYST B.II.4), starting from the true origin of
Aeginaís naval lordship and tracing its connexion to the present down the mythical
genealogy, as probably happened at 12ff. There may be more than a nod towards Hesiod,
especially since the mythical narrative may ultimately rely on him to an unknown extent
(see p. 250 and n. 52): the title of the Hesiodic Catalogue of Women is not attested until the
late third century (Nicaenetusí IPQRbj�Q jRfnVgSg], CA p. 2, requires the Hesiodic
precedent), but the Homeric Catalogue of Ships is called so by Thuc. 1.10.4, and it would
be no surprise if the former title too had already been current in the fifth century. The use
of the verb followed by indirect question is Homeric, although the form is not, see LfgrE
s.v. VYST B.II.4.a.R (Homerís future is jRfRVYXT). On the destabilization of roles and the
blurring of singing voices that arise from this second person see pp. 248f.
8 difO[ : the assertion of the islandís glory and status (1-4) and the consequent
undertaking to honour her in song (5-7) seamlessly lead to the question of where such
glory and status derive from, the answer to which will justify the first step and fulfil the
second. Parallels for such questions, incipitary or climactic, are easily found in epic (direct
question, e.g. Il. 1.8, 5.703; indirect, e.g. 2.484-7, 14.508-10), but Pindar too makes uses of
this stylistic device to introduce a catalogue at O. 10.60-3: see Bundy 1972: 78f. The more
common aporetic question on what to sing of, which Race 1990: 104 links to it, is quite a
different business (see chiefly Bundy 58-70).
274
[ced]gTc[Y[ : hapax, ëruling with shipsí or ëruler of shipsí; if the alternative was
meaningful for Pindar and his audience, the construction of c[�fRQb] with a genitive at P.
2.58, 6.24, Stesich. PMGF 58, etc. and the unusual (for Aegina) emphasis on power that
would otherwise result favour the latter. According to Hes. Cat. fr. 205.6f. M.-W. it was
the Aeginetans who first sailed the seas, and Pindar too attests that seamanship was an
important part of the portrayal of Aegina in the years of her independence: N. 5.9
QRP]bjVPfnQ, I. 9.1 QRP]bjVPf¦].
9 NcU_Z[c : ìno personal god, but an indefinite divine potency, the divine inspiration of
sea-supremacy that was infused into Aiginaî (Farnell 1930-2: II 408), the destiny of naval
prowess that was Aeginaís lot.
fO_UXO[Z[ : ìhospitable in accordance with Rightî (Currie 2005: 299 n. 13), another
hapax. The article portrays Aeginaís hospitality as well known (Race 1990: 113 n. 73), as
indeed it is from several passages in Pindar and Bacchylidesí Aeginetan odes: O. 8.21-3
¯QpR ]ifWb[R }bÙ] XWQ^gP £ cn[Wk[g] �]jW~fRb rY_b] £ ¯XgZí �Qp[icTQ, N. 4.12f.
k^jRb XWQR[jYÔ jgbQ¦Q £ dYSSg], 5.8 d^VRQ XYQTQ ³[gP[RQ, I. 9.3f. gà pY_bQ gÃkÓ k^jRQ
£ XW^QTQ ÕcW[�R^QgQfW] (Grenfell and Hunt 1908a: 99), cp. O. 8.27f. cRQfgkRcg~]bQ Ö
XYQgb] £ j^gQR kRb_gQ^RQ, N. 3.2f. fÏQ cgV�XWQRQ Ö }T[^kR Q¬]gQ oÒSbQRQ, B. 12.4-6
¿V�^RQ £ XW^Qgb]b Ö Q¬]gQ o∞S^QR], 13.95 kY]cgbQR cRSXW^QgP ZpgQ¦]. Aeacusí
hospitable reception of the Greek envoys, depicted on the reliefs of the Aiakeion (see p.
206f.), showcases a similar idea, which is also implied by N. 8.8-12 (Fearn 2007: 144). On
Aeginaís claim to hospitality and its ideological implications see Figueira 1981: 324-30,
Pfeijffer 1999: 111f., Hornblower 2007: 297-300.
275
\]OTW[ : intimating an inherent, inborn quality (Bona 1988: 139). Here as with kR^_gQR,
ìthe adjective expresses the essential power of the nounî (Farnell 1930-2: II 408), cp. P.
1.94 dbV¦d[TQ �[Wfn.
10 : cp. I. 5.52f. >Wˆ] fn fW jRÚ fÏ QY_Wb, £ >Wˆ] ¡ cnQfTQ j�[bg], Od. 9.552 = 13.25,
Semon. fr. 1.1f., Sol. fr. 13.17 W., [Terp.] PMG 698.1, B. 17.65f., etc. Pindar is rather fond
of fn (fW) jRÚ fn in its various shapes (Radt 1958: 178), see Slater 1969 s.v. ¡, ≈, ≈] B.4.
Often the scholia read into it a reference to a specific pair of things (goods and evils, | P.
5.74a, 7.23 pp. II 180, 204f. Dr.; deeds and understanding, | N. 1.43 p. III 18 Dr.; etc.), but
this is an unnecessarily restrictive interpretation and should not be imported into F5, pace
Radt 179: the phrase merely signifies variety, Bundy 19862: 74 n. 100 (not ìa
universalizing doubletî of polar opposites as he claims at 1972: 81). Here it is not
appositional, as it is usually taken, but predicative: ìhe who makes all things such and
suchî made you prosperous (similarly Fehr 1936: 42; for this use of fW�ZT see Slater s.v.
b). fgb emphatically introduces an answer also at O 2.90, Pae. 4.40 (Slater s.v. 1.e).
11 ntteWXYxO[ : the imperative offered by H5, also present as an interlinear correction in
H4, is meaningless in this context: Aeginaís prosperity is a given, not a request, cp. 8f. (see
also Grenfell and Hunt 1908a: 99). �SSPRV^�T is often used of divine concessions, see LSJ
s.v.; on its slightly unusual construction in our passage see Radt 1958: 179.
ƒXBZ[ : Aeginaís sea-lordship and hospitality are subsumed within the broader category of
prosperity, with a shift from the particular to the general, from active to passive, and from
earthly to divine agency: the individual qualities which Aegina ìtookî at 8 are now an all-
encompassing fortune that Zeus ìhandedî to her. This is consistent with Pindarís habitual
276
portrayal of ƒV�g] as something that is bestowed on, or offers itself to, its partakers more
frequently than as actively acquired by them (cp. O. 2.21f., 36, 6.72, P. 5.14, 55, etc.,
contrast O. 5.23, P. 3.88f.).
12 OÃ]gZdc : cp. Pae. 8a.15 WÃ]£[[�]gcR æ[gQ^TQ (unless it was �R]£[[�]gcR, Maas ap.
Snell 1959-643: II 45). One can only speculate what this notoriously ambiguous epic epithet
of Zeus (see DELG s.v., LfgrE s.v., Schmitt 1967: 157-9, Kaimio 1977: 70f.) may have
meant for Pindar, if considerations of meaning played a part at all beside its traditional
status: Slater 1969 s.v. gives ìfar seeingî, but P. 6.24 �R[�gcR � where sight is out of the
question � suggests rather ìfar-sounding, i.e. thunderingî (LSJ s.v.), cp. lyr. adesp. PMG
1008.1 WÃ[�gcR jYVRkgQ.
�]i[Ze dcwa : cp. O. 1.10, 4.6, 7.67, Il. 2.205, a variation on the Homeric formulae
WÃ[�gcR æ[gQ^k\(Q) (Il. 1.498, 15.152, 24.98, H.Hom. 23.4) and WÃ[�gcR >W�] / >vQ (Il.
5.265, 8.206, 442 etc., Od. 4.173, 11.436 etc., Hes. Th. 884, H.Cer. 334, 441 etc.). On such
Pindaric variations of epic diction see F2.7 n.
ÕNWT�Oaa�Y : Grenfell and Huntís emendation produces a form unattested before the
Hellenistic age (Arat. 943, A.R. 2.939, 3.860, etc.), but a long syllable is needed to achieve
responsion with 33 CfPSÙ] ≈[jb×Q �X W–-, and Pindar is no stranger to ëamplifiedí dative
plurals of this sort. The shared error in the two papyri is a useful indicator of common
ancestry, but given Aristophanes of Byzantiumís conservative tendencies there is a chance
that the error was already in his edition, see Radt 1958: 12.
277
The mention of Asopusí water is no mere descriptive detail. It seems certain that in
Pindarís time in the city of Aegina there was a fountain called Asopis, which may have
been located near the Aiakeion and near which choral performances could take place; in
this context, its name concretely brought out the connexion between Aegina and the
mythical homeland of her eponymous nymph (see Fearn 2007: 102-5, 115f., 119f.). Then,
especially if F5 was � originally or otherwise, with or without Pae. 6(a) � performed at or
near the Aiakeion, Õknf�W]]�b kí [�]cí �][T]£cgt (cp. N. 3.3f. —kRfb Ö �cí �]Tc^Tb, i.e.
at the Asopis: Privitera 1988) must have presented to the audience in a particularly vivid
fashion the links between the mythical event and the odeís present setting, or in any case
its civic and religious context. Fearn further associates the origin of the fountainís water
from the area of Mount Hellanion, from which it was conveyed artificially to the city, with
the myth of Aeacusí supplication to Zeus Hellanios to end the great drought (see | 3 n.), a
myth which Isocr. 9.14f. connects to the foundation of the Aiakeion. If this association was
indeed felt in the early fifth century, and even more if it entailed a belief that the Asopis
herself resulted from such supplication, then the mention of Asopusí waters also has a
direct relevance to the worship of Aeacus, if F5 was performed in such context; it also
adds a further touch to the picture of Zeusí role as the dominus of the ode (see p. 250).
12f. �a[h]£dZ� : apo koinou with the waters (of the river) and the portals (of the river-
god). On this kind of meaningful syntactical ambiguity see Gerber 1985: 145f.; for the
slippage between a feature of the landscape and its personal hypostasis cp. e.g. O. 6.28-30
(Pitane), 7.13f. with | 24c p. I 203 Dr. (Rhodes), and indeed 7 and 17 here (with nn.), see
Radt 1958: 183.
278
13 �[ZTí : not ìan empty fillerî (Maas 1914-21: I 26) but the most natural way to start off
a mythical narrative, as a score of Pindaric parallels confirms: O. 3.13, 6.13, 13.55, P. 3.5,
4.4, 6.21, 8.39, 9.5, 79, 10.31 (reprised at 45), 12.6, N. 4.25, 5.9, I. 4.52 (a compressed
one-sentence myth in *F7.8, O. 9.9, P. 1.16, N. 8.18, 10.7), cp. B. 3.23, 5.56, 11.40, 20.1.
13f. BcfgVZX£dZ[ : on the metrical lengthening of the last syllable see Radt 1958: 94f.
(parallels listed at n. 4). Maasí emendation brings no improvement: a deep-girt atrium may
not be entirely beyond Pindarís reach, but to conjecture one when an unexceptionably
deep-girt maiden is at hand (cp. Il. 18.122, 339, 24.215, H.Cer. 5, H.Ven. 257, P. 1.12) is
an unprofitable solution to a non-existent problem.
14 \��]`�cTZ : the form with two epsilon is found in mss. at A.R. 1.214, 4.918, cp.
Hsch. R 4944 Latte, Synagoge B R 1355 Cunningham. The manuscript tradition of Homer
consistently has �Q\[W^°- (Il. 20.234, Od. 1.241, 4.727, 14.371, 20.77), but the situation is
rather chaotic at Hes. Th. 990 (�QW[Wb°- bv, �QR[Wb°- nQ, �QR[b°- k, �QR[W°- S); for a
complete picture of the spellings of this elusive verb and a judicious non liquet on its ëtrueí
form see West 1966: 428 (see also Radt 1958: 180f.). Further early evidence for the myth
has already been discussed (p. 250); some later sources add that Zeus metamorphosed into
an eagle (| T Il. 21.189 Erbse, Ath. 13.566d, Nonn. D. 7.121, 210-14, 13.201-3, etc.,
[Clem.R.] Hom. 5.13.3), much as he had done to carry off Ganymede, but the origin of this
detail � an aetiology of the constellation of the Eagle in Nonn. D. 16.58f. � is unknown.
15 TiTO : the correlation with 11 c[gfí collapses the moment of the abduction into that of
its consummation (which must be what 136-9 are about, pace Duchemin 1955: 227),
effacing the intervening change of place: contrast the narrative of the same events at I.
279
8.21f. ]Ó kí �] Q¬]gQ a∞Qgc^RQ �QWSjàQ jgb_¬fg, k~gQ ¯QpR fYjW] £ o∞Rj¦Q
�R[P]dR[nSTb cRf[^.
^]gaOcY : the scene recalls the union of Zeus and Hera in Iliad 14, which Zeus shields
from indiscrete viewers with a golden cloud (343f., 350f.): Norwood 1945: 218 n. 5.
However, while Homer with pointed irony defeats the purpose of the supernatural
concealment by telling in quite some detail what went on beneath it (Il. 14.346-50,
graphically sandwiched between the two references to the cloud), Pindarís golden mist
screens the divine intercourse from his own audience no less effectively than from any
onlookers in the mythical scene. At the same time, of course, it flags the union (which the
established myth and the Homeric model itself make all but obvious) as a key event
sanctioned by the gods: gold is the natural stuff of whatever pertains to divinities, see
Lorimer 1936: 14-29, Slater 1969 s.v. Z[�]Wg] a, cp. fr. 222 Sn.-M.
15f. \£`]Za : the ìhazeî with which the Homeric gods habitually conceal themselves
when they do not wish to be seen (LfgrE s.v. B.1), see Radt 1958: 182, rightly arguing
against the more extravagant interpretations of Dornseiff 1921: 47 (flames), Schmid 1929:
I/1 608 n. 2 (thunderbolt), and Fehr 1936: 42 (gold rain I la Danae). Pindar seems to
emulate Homer with Homer, as it were, by replacing the Iliadic ëcloudí (14.343 QYdg],
350 QWdYV\) with a more markedly Homeric word.
16 �V]e�c[ : the dual �j[P°nfRQ presented by H4 is both unmetrical, if we assume
complete identity with Pae. 6(a), and syntactically impossible: Z[P]YRb and j¦�[R]b might
just conceivably be read as datives (although the breve marked on both endings is more
naturally taken as indicating a nominative than correption of a dative, especially since the
280
final syllable of j¦_Rb must be long), but then 15 Õ_YfW[gQ would be left without a
referent, and one does not expect Aegina to play a part in the creation of the golden haze.
The error may have been a mere scribal slip (ìa dittography of °Rî, Grenfell and Hunt
1908a: 100), but the diple in the margin may also signal an acknowledged locus criticus.
Vi�[c]Y : an arresting double image, centred around an apparently unique metaphorical
use of j¦_R (but cp. Call. Del. 302f. g“Vg] �pW^[Rb] £ ²]cW[g], see 6 n.) and the
established double meaning of 17 Q�fgQ (q.v.): the golden haze shades the islandís ridge
just as golden hair shades a maidenís back (Radt 1958: 183). The golden locks may
themselves have erotic connotations (Rutherford 2001: 325 n. 79), highlighted here by the
suggestion of nudity that the image conveys.
ndY^k]YZ[ : a favourite of Pindarís (also at Pae. 4.46 and ten attestations in the
Epinicians), the earliest recorded author to use it together with Bacchylides (13.92f., in a
passage which may echo F5, see p. 84). It is not entirely true that, as Radt claims (1958:
184), the epithet can only refer to the ëridgeí of Aegina the island, not to the ëbackí of
Aegina the maiden: one would be at pains to find something more epichoric than an
eponymous nymph at the very moment of her implantation into her homonymous island,
and the slight hypallage of ìyour (Aeginaís) back of the landî for ëyour back, who are of
the landí is not significantly more troubling than ìyour (the Aeginetansí) ridge of the landî
for ëthe ridge of the land that is yoursí, as Radt would have it.
17 VcTWaVYZ[ : Radt 1958: 182 rightly remarks that the epithet expresses incidental, not
intrinsic, quality, ëshadedí not ëshadyí: cp. Hes. Op. 513, A. Ag. 493, Supp. 343, 354, etc.
(see also Stesich. PMGF S17.8f.). This entails that it should be taken proleptically, ìda
281
hBllten goldne Nebelhaare den ROcken eures Landes in Schattenî (ibid., emphasis
original). Nonetheless, given �cbZi[bgQ Ö Q�fgQ it is hard not to perceive an echo of the
Homeric ƒ[WR ]jb¦WQfR (Od. 5.279, 7.268, H.Ap. 34, etc., cp. Il. 1.157, P. 9.34).
[�TZ[ : for the metaphorical use cp. O. 7.87 Qifgb]bQ �fR�P[^gP (Radt 1958: 182), P.
4.26 QifTQ —cW[ SR^R] �[q_TQ, the reference to land being only a small departure from
the epic WÃ[YR Q�fR pRVn]]\] (Il. 2.159, 8.511, Od. 3.14, 4.313, etc.). Instead of the
genitive that we find in these examples, the specifying function is fulfilled by 16
�cbZi[bgQ. The singular is, to my knowledge, nowhere else used in this sense: if the
difference is meaningful, there may be a slight emphasis on the image of the maidenís back
over that of the islandís mountainous ridge. As far as the latter is concerned, one supposes
that the exact location may have been understood to be Mount Hellanion (so Radt 1958:
182f.), but the phrase seems rather to indicate the entire island.
Õ_`TO]Z[ : if it is truly a plural for the singular (Slater 1969 s.v., Rutherford 2001: 325) it
is rare but not unique, pace Calder 1977: cp. H.Merc. 310, Call. Del. 203f. with Mineur
1984: 185, 226f., Nonn. D. 5.341, and perhaps Pae. 9.37 with Hope 1986: 173 (but it could
refer to Apollo and Melia), Sol. fr. 19.2 W., see Rutherford 2001: 196 and n. 20.
Alternatively, it can easily be referred either to both Aegina and Zeus (assuming a slight
shift in the primary second-person address from Aegina-island to Aegina-maiden, all but
justified by the overt conflation of the two at 15-17), or to all three of Aegina-island,
Aegina-nymph, and Zeus, in the respective ways that the kaleidoscopic metaphor requires.
I find it rather more strained to refer the plural to the chorus, or to the Aeginetans more
broadly (so Radt 1958: 184f.), unless we are meant to understand that after 6 jRfW[W~] the
282
speaker is indeed the island: it is true that the first person addresses the chorus at the end of
Pae. 6(a) (whether or not F5 is regarded as a different text is immaterial for the purpose of
the parallel), but there is a distinction to be made between seeing such a reference to the
performance in an opening, closure, or transition, and importing one into a mythical
narrative, where it would normally be out of place.
Calderís emendation neatly does away with this difficulty, but at the same time undercuts
the double image of island and maiden that j¦�[R]b, and to a lesser extent Q�fgQ, seem to
require. But it may well be that the ëhairí is a self-contained image not meant to spill over
into the rest of the sentence, that the singular Q�fgQ is of no particular import, and that
what Pindar is describing is solely the golden haze shrouding the island: in that case,
Ú_YfW[gQ is undeniably attractive (ìthe ridge of our landî, id.).
18 µ[c : ìëwhereí more likely than ëin order thatíî (Kirkwood 1982: 318), a meaning that
µQR never has in Pindar (Radt 1958: 185); cp. I. 8.20 ¯QpR in an identical context (quoted
at 15 n.). Here too it seems likely that 18f. move on to narrate the birth or conception of
Aeacus (so Radt 1958: 185) rather than expand on the divine union pointedly not
mentioned at 15-17.
\_B]iTh[ : hypallage seems to be particularly frequent with adjectives pertaining to the
divine sphere, see F3.7 n. The bed is immortal because visited by an immortal, but the
description simultaneously intimates a status more than earthly for the island Aegina, to
which the phrase concretely refers (see 2 n.): cp. Pae. 9.35 VYZWb cYVR] �_�[g]^Tb
ÙWV^R] (the Ismenion, where Melia lay with Apollo).
283
19 : a part of dPfW�T would not be out of place, referring either concretely to Aeacus or
his lineage (cp. N. 5.7, 7.84) or to some quality pertaining to them or Aegina (cp. P. 4.69
fb_R^, N. 8.17 ƒV�g], I. 6.12 k¦XRQ).
21 : a part of ÙP[_bkiQ is likely, whether or not it constituted a reference to Zeusí
creation of the human inhabitants of Aegina out of ants (see p. 206 and n. 6). It seems less
easy to find a suitable context for Turynís supplement.
22 : if Snellís supplement hits the mark as far as sense is concerned, it is worth
remembering that Pindar also uses ]R¦d[TQ at Pae. 1.10 (DíAlessio 1988: 1445) and
9.46, cp. B. 13.186: ]Rgd[]×QY]fRfgQ, which is equally compatible with the metre, is a
valid alternative. It is usually maintained that the person ìmost prudentî is Aeacus (Snell,
Radt 1958: 186, Rutherford 2001: 326), who must have been mentioned somewhere in 19-
21, but another possible candidate is Peleus: Themis terms him WÃ]W�Y]fRfgQ at I. 8.40,
where a few lines earlier the Aeacids collectively are ]id[gQW] (26), which the gods ìhad
in mindî (26a) when they decided to wed Thetis to him; his virtue as described at N. 5.26-
34 is easily construed as ]Td[g]�Q\. But if the relative f¦Q at the start of the verse refers
to Aeacus (so Radt) there is little room for a change of referent.
23 : the accent written in H5 looks odd, but the metre leaves no alternative to }bÙ]. It may
be a reference to Aeacus as a son of Zeus (Snell) or to one or more of his descendants (e.g.
¯jSgQgQ, �jS¦QTQ, cp. N. 7.50), but Zeus may be relevant to the Aeacid myth in countless
other ways.
284
24 : the combination of accent and metre (a short syllable follows, and verse-end probably
precedes) rules out everything but a form of cW�pg_Rb. cW�[pg_Rb itself (Snell) is a nice
possibility, an intrusion of the primary speaker like VYSgQfb F11.9, N. 7.84, or Ò]fW I.
4.35; this seems at any rate more likely than character-speech at this point in the ode.
27 : as Radt 1958: 186 points out, responsion with 6 �VVí �gbk¬Q demands that the end of
the line scan � ˙ ˙˙ � , thus with ÚV^Tb meaning ìmarineî (less probably ìfruitlessî, an epic
epithet unattested so far in Pindar), not ìsunî. Two Aeacids who have an established
connexion with the sea (see also 28 n.) are Peleus and Achilles through their respective
wife and mother Thetis; it is tempting to suppose that these lines deal with one of them
(Peleus?). However, Aeacus himself had been involved with another daughter of Nereus,
Psamathe(ia) (see p. 213), and one may wish to see a reference to the pair of them here. If
so, SY[gQfb (or ´\[vÔ, ´\[YÔ) may hide somewhere in the neighbouring lacunae, cp. P.
9.94 ÚV^gbg SY[gQfg], F3.4 n.; but âVbg] is used as a two-termination adjective at E.
Heracl. 82 (also plausibly conjectured metri causa by Hermann 1796: 440 at S. Aj. 357),
which allows for a feminine referent. If the epithet referred to Psamatheia herself, one
might have perceived a word-play on the Homeric °R_npgb](bQ) Ö ÚV^\b]bQ (Od. 3.38,
4.438, H.Merc. 79). Of course, it may have been simply a reference to an event that
occurred near or on the sea, in which case the mythological train of thought is even harder
to pin down. Radt must be right that ]Y__WQ is the infinitive of W∞_^, but a compound
cannot be excluded.
28 : Turyn and Snellís conjecture of an epithet in jP[RQg- or jP[RQWg-, perhaps applying
to the sea or a sea-creature, is borne out by the context (Radt 1958: 186). It was more
285
probably a compound than the simple adjective, which Pindar elsewhere uses with a long
upsilon (O. 6.40, fr. 33c.6 Sn.-M.). kW£XR[ must represent the aorist of kYZg_Rb, as Snell
saw, so the epsilon that precedes it can be either its augment or the end of the preceding
word (Radt ibid.).
29 � ���[×] ̂ Rb : ìmore likely to be a form of the adjective than of the nounî (Henry 2003b:
9).
30 : the section of this line that H5 preserves must be corrupt, even though it is difficult to
pinpoint where. The noticeable smudging which the ink has undergone after ]W_gQ may
have resulted from an attempt at deletion and correction, and the text has definitely been
tampered with. The first trace is an illegible letter that ends with a vertical (or a narrow
letter followed by iota), then what looks like omega, and probably nu. Over omega there is
a grave or circumflex surmounted by a tiny triangular sign which I read as alpha, a variant
reading or a correction: palaeographically it could also be a circumflex, but this would
entail reading the descender below it as a (rather wonky) longum, which surely it would be
absurd to write on an omega (if this is what it was). If it was indeed alpha correcting
omega, it can be supposed that the letters represented the ending of a genitive plural, but
after ]nQpW_gQ (which I cannot divide in any other plausible way) it would have to have
been a very short word indeed. Henryís WÃ]nQpW_gQ is a likely supplement; perhaps
Z[P]]nQpW_gQ (admittedly unattested so far as an epithet) is another possibility, since
Pindar assigns flowers of gold to the Isles of the Blest at O. 2.72. At the end of the line,
Medea Norsa, whose microscopic sight I cannot match even with mechanical aid,
suggested that RbkTQ was corrected into RWbkTQ or R\kTQ: Pindar never uses ³bkT, so
286
either the division was -R ∞kiQ (with digamma retained as at O. 9.62, 14.16, P. 5.84, Pae.
1.3) or Norsaís corrector was right to write �W^kTQ or �\kiQ. Furthermore, the verse must
scan � ˙ ˙ � ˙ � ˙ ˙ ��̊ ̊� : even if the final nu of ]nQpW_gQ is taken to lengthen the syllable
with no need for another consonant to follow (see 13f. n. but note the variant in H4), it is
hard to construe the subsequent three or four letters � one of which seems to be omega � in
such a way as to provide the one or two short syllables that are necessary to complete the
verse.
H4 fr. 46 gives little help. The beta and low arc (alpha?) after ]nQpW_×Q (whose omicron
does not really look like one) are certainly not what H5 had, although I cannot exclude that
the second part of the interlinear variant may have been meant to represent the same
notional reading; even on their own, the traces are hard to turn into a viable supplement,
especially since the extent of the divergence between the two papyri is unknown.
31 : a reference to the Isles of the Blest (Henry 2003b: 9) may not be inapposite here,
although the context does not quite confirm it. (Not only Achilles ended up there, but also
Peleus according to O. 2.78.) However, I doubt whether ‰jWRQg~g Q¬]gb can truly mean
ëislands in the Oceaní and, if not, whether the two words go together at all; the Isles of the
Blest themselves may have been recalled through a more descriptive phrase, ìislands amid
the waves of the Oceanî, ìislands round which the breezes of the Ocean blowî (cp. O.
2.71f.), or the like.
32 : ìthe contracted forms of adjectives in -¦g] are not found elsewhere in Pindar, though
since he has ÊRfgt] (O. 3.32, etc.) beside ÊRf¦g] (Pae. 5.44) and �gt] (O. 2.83) beside
287
�¦g] (N. 6.52 cj.), I suppose they cannot be ruled outî (Henry 2003b: 9). At the end of the
line the relative gµ is certain, see Snell 1938: 426.
33f. : ad Aeacum iudicem deorum (non mortuorum) et eius ius iurandum videtur referenda
(Snell 19531 ad loc., see also Rutherford 2001: 326, Fearn 2007: 90), an episode that
Pindar mentions obliquely at I. 8.24 √ (sc. o∞RjÙ]) jR^ £ kRb_¦QW]]b k^jR] �cW^[RbQW.
However, if Hubbard is right to argue that those lines ìneed not mean that Aeacus actually
acted as a judge or mediator who settled disputes between the gods, only that he somehow
ëprovided an endí (cY[R]) to their dispute(s)î by fathering Peleus, who married Thetis and
thus ended Zeus and Poseidonís rivalry for her hand (I. 8.27-47; Hubbard 1987: 6-13,
quotation from p. 8, although one might question the equivalence between k^jRb and
disputes), we are lost in the dark again. Moreover, even assuming that Aeacusí arbitration
was indeed thought to have happened, the relevance to it of the godsí oath � not any oath,
and presumably not an oath Aeacus could have sworn, pace Maehler � is unclear. In what
we know of the Aeacid saga there is one oath that might fits that description: the Cypria
and the Hesiodic Catalogue (frr. 2(I) BernabC, 210 M.-W.) reportedly recounted how Zeus
swore that Thetis would marry a mortal after she refused him. Nothing precludes thinking
that this is what Pindar narrated in this section of the ode, but equally nothing indicates that
it is the only divine oath he could have mentioned; for a different possibility see 48 n.
Given the pitiable state of 19ff., it is impossible even to be sure of which Aeacid
generation we are dealing with at this stage. If 31 does refer to the Isles of the Blest, the
natural implication is that at least one has completed its course, but even so, the uncertainty
over the subject-matter of 27f. precludes estimating whether we are now with the second or
the third.
288
48 ]VXeTla : simple or compounded. Over alpha there are a longum and a flattened
accent that could in principle be a circumflex as well as a grave; one would expect a
longum to mark the Doric alpha of the genitive singular rather than the expected long
alpha of the accusative plural, and the (for us) odd combination of circumflex and longum
is paralleled in this papyrus at frr. VIIr.216 (Pae. 7(a).7) and very probably VIr.206 (F5.31).
�Nh : inevitable, since the metre demands a short syllable, so a vowel must follow (both
eta and iota satisfy the traces) and thus a new word begin. Rutherford 2001: 326 suggests
that it ìcould represent a first-person statement by one of the characters in the songî,
namely a prophecy containing an account of the Aeacid genealogy. Recalling FOhrerís
remark (1967: 127) on Pindarís tendency to narrate episodes of the Aeacid saga by flash-
forwarding from an earlier point in the myth (O. 8.42-6, I. 6.52-4, 8.35a-45), he speculates
that the speaker ìhas to be a god, probably Zeus or Apolloî (327). Both of these can be
true: Zeus may be portending the glorious future of the Aeacid line after a solemn oath to
that effect (cp. 32), or Apollo may be prophesysing about Achilles at the wedding of
Peleus and Thetis (so Rutherford, cp. A. fr. 350 R.1). However, given the proximity to the
final section of the ode, it cannot be excluded that the primary speaker steps in before
recalling some further episodes in a more summary fashion at 49-55.
50 : the first surviving word may have been a first-declension genitive: other solutions that
satisfy the metre are difficult to find. Elsewhere Pindar uses the aorist ZgVTpW^] (O. 7.30,
N. 7.25, 10.60), construed with a genitive of cause at N. 7.25 (≈cVTQ), cp. the Homeric
examples listed in LfgrE s.v. G.2.b.R. Considering 49 ZR]VjgZn[_Rb and the Aeacid
context, the supplement that first springs to mind is �f[W�]kR, the subject of the participle
289
being Achilles (Rutherford 2001: 326 n. 83) or perhaps Apollo, cp. Il. 1.9. But alternatives
are easily thought of: H\VW�]kR, the subject being Zeus (cp. Il. 24.114), o∞Rj^]kR, the
subject being Aeacus (cp. I. 5.34f., but with a jarring onomastic short-circuit), etc. The
target of anger is normally expressed by a dative (see LfgrE s.v. G.2.a), but even given
ZR]VjgZn[_Rb (if I am correct to see a longum on the diphthongue) there is little excuse
for conjecturing ]kR�b� in this fragmentary context. The function of the genitive, if it did go
with jWZgVT_YQg], will have been to present the relevant person as the originator, as well
as the target, of the subjectís anger.
o 50 : ìcgbQn is an alternative reading. The mark of short quantity rather suggests cgbQR^
as a variant on cgbQ¬b, but a final b was certainly not writtenî (Grenfell and Hunt 1908a:
100). Radt 1958: 187 suggests that the breve was meant to represent the metrical
shortening of alpha, but it is unclear to me why the annotator should have chosen to write
out a marginal variant rather than simply adding a breve to the text itself. At any rate,
given the rarity of such shortening of long final alpha (id.), it seems inadvisable to adopt
the variant into the text, as Snell does. One cannot tell what stood in its place, but the fact
that only cgbQn is reported in the margin suggests that it was another disyllabic word
scanned as a trochee (unless the grave unnecessarily written on omicron indicates that the
two syllables in the main text were independent words).
53 : Vitelliís articulation of the traces on H5 fr. IIIv is right, see Radt 1958: 187. The
interval between frr. XIv and IIIv, which can be estimated with some precision from 56 and
suits the restorations presented at 55 and 57, is not wide enough for a full three letters
before the omicron whose right end survives, but seems slightly too wide for only two: two
290
broader letters and one iota would fill it perfectly. Vitelliís V]¦Sgb can hardly be avoided,
so _P[^R[b] V]¦Sgb suggests itself as a likely solution; the iota after the lacuna could be a
dative ending. For _P[^g] in the closely related contexts of achievement and renown cp. N.
10.2f. dVYSWfRb (sc. Argos) kí �[WfR~] _P[^Rb] £ ¯[STQ p[R]YTQ ãQWjWQ, 45 the ZRVjÙQ
_P[^gQ which the victorís family won at local games, I. 4.1f. ¯]fb _gb Ö _P[^R cRQf¬b
jYVWPpg] Ö Õ_WfY[R] �[WfÏ] —_QTb kbijWbQ, 5.27f. jVYgQfRb Ö _P[^gQ Z[¦QgQ, 6.22
_P[^Rb kí ¯[STQ jRV�Q fYf_RQpí §jRf¦_cWkgb Ö jYVWPpgb, fr. 172.1-3 Sn.-M. H\VYg]
�QfbpYgP £ _¦Zpgb] QW¦fR] �cYVR_°WQ £ _P[^gb], B. 5.31-3, 9.47-9, 11.126, 13.196-8,
19.1-4; see also 54 n. on �cW^[gQR].
53f. N £[]YVT]gdh[ : an identical nexus at N. 7.9f. kg[bjf�cTQ £ o∞Rjbk¬Q, cp. N. 3.60
kg[^jfPcgQ �VRVÏQ ÊPj^TQ (Ferrari in DíAlessio and id. 1988: 173). This epithet is not
known to occur outside Pindar and may be his own invention (id.). It is presumably
modelled on the Homeric kgtcgQ �j¦QfTQ (Il. 11.364, 16.361, 20.451, Callin. fr. 1.14 W.,
cp. Archil. fr. 139.6 W.), a metonymy for the thick of battle; see also Fennell 18992: 34.
54 [ q q q ]OY[ : probably an infinitive expanding on 53 V]¦�Sgb with epexegetic SW (on which
see Denniston 1929, GP2 138-40). The variant fW in H4 could also fulfil the same function
(GP2 502), but Radt 1958: 163 n. 3 rightly points out that most of the interlinear variants to
ëPae. 6í found on H4 are inferior to the main text. If V]¦ �Sgb is correct, the etymological
short-circuit that would otherwise result gives a slight edge to Radtís [jVY]WbQ � [dVYS]WbQ
is too long � over DíAlessioís [VYS]WbQ: for the verb cp. I. 5.27, in a passage which bears
several resemblances to ours.
291
\dOU]Z[ca : denoting both the variety and greatness of the Aeacidsí deeds of valour and,
proleptically, the boundless reach of their fame (cp. P. 2.64 fÏQ �cW^[gQR k¦XRQ WÕ[W~Q).
55 x[YXOw]¡O : a prayer, as one would expect at the end of a hymn; on its probable
addressees see 56 n. Given this traditional structure, it is distinctly less likely that these
imperatives are addressed to the Aeginetans, as Bona 1988: 140 suggests; his view
becomes outright impossible with the third imperative at 58f. (surely no human agent can
bestow health on a community). For a deityís love for a person or place cp. O. 2.26f.
(paralleling the repetition of the verb), 4.16, P. 1.39, 2.16f., B. 4.1f. and, in the context of a
prayer, O. 6.101f.; similar sentiments are expressed among others at P. 5.117, Parth. 2.3-5
(c[¦d[TQ), P. 8.18-20 (WÃ_WQW~) and, in prayers, O. 2.14, 4.12f. (W–d[TQ), Pae. 5.44-6
(WÃ_WQW~).
o 55 : ìin the imperativeî not ìwith an imperative senseî (Ferrari in DíAlessio and id.
1988: 173), pointing out that d[bVW~]�W is not the identical form of the indicative. All three
main verbs at 55-9 could theoretically be taken as present indicatives, with the closing
prayer encompassing only 59-61 (for a prayer involving only the speakerís favourable
reception see Pae. 5.44-8 and 60f. n.), but the unprepared introduction of the second
person at 55 is better understood in connexion with a switch from narration and description
to direct address.
56 q q q q ]� : Ferrari in DíAlessio and id. 1988: 174f. makes a good case for restoring a
specification of the ìancestral cityî which the addressees are requested to love. His ]dYT]�
is not a Pindaric form; DíAlessioís ]�Q ]d]� is too long for the lacuna, Battezzatoís pWg]Ú
too short; Henryís kY ]d]� suits both Pindaric diction and the available space, but, like the
292
others, he assumes that the imperative is addressed to beings other than the Aeacids (ìthe
godsî, DíAlessio in id. and Ferrari 1988: 164; see also Rutherford 2001: 328), the latter
being referred to in the third person, as they were at 54f. However, it is more advisable to
read d[bVW~]�W as addressed to the Aeacids themselves, like the prayer that concludes P. 8
(98-100): this dispenses with the need to import further figures into the picture and makes
cRf[ibRQ more straightforward, since the adjective will apply also to the grammatical
subject and not only to the primary speaker. If this is correct, one can supplement e.g. kí
–__]� (for the form cp. P. 4.259).
dcT]kYc[ : assuming exact responsion with Pae. 6(a).56, 117 � ˙ ˙ � ˙ ˙ � ˙ (see Henry
2003a: 12 on 117), the dieresis cR]�[TÔRQ on H4 is impossible (Radt 1958: 189). Radt
also argues against taking Tb as a diphthong abbreviated in hiatu, which would give the
required metre while preserving the transmitted text, by pointing out the seven Pindaric
instances of trisyllabic cRf[ibg] with un-shortened Tb (P. 4.290, 6.45, 11.14, N. 4.48, I.
1.35, 2.44) and explaining away the two cases where short Tb is required by the metre, N.
2.6 and 9.14 (189f.). (Pae. 7b.29 cRf]�ibRQ, which he cites at n. 4, is inconclusive, since
it responds with 10 WÃcYcVTb whose own middle syllable is of undeterminable length.) At
N. 2.6, V reads cRf[iRQ while all other mss. have the facilior cRf[^RQ or similar; at 9.14,
cRf[iTQ is the unanimous reading of the veteres, cRf[^TQ Schmidtís emendation (1616:
210). Since the case for Snellís cRf[^RQ rests on a variant and a conjecture confirming
each other and presupposes that a counter-intuitive error arose independently in three
different places at a comparatively early stage, I prefer � however hesitatingly � to retain
the reading of the papyri. (So, for a different reason, also Ferrari 1999: 155.)
293
57 : Ferrari rightly recognises in W–+[[gQ]R ìmerryî (Slater 1969 s.v.) a reference to
festivity and compares N. 5.38, P. 10.38-40, Thgn. 765-7, Xenoph. fr. 4 W., citing
Fraenkelís treatment of the adjective at 1950: II 365-7 (DíAlessio and Ferrari 1988: 176).
At N. 5.38, as here and at Parth. 2.67, the context is a religious rite, cp. F3.1 f�bkí �Q
³_Rfb fW[cQ�b with n. For the pairing of c¦Vb] and VR¦] cp. O. 9.66 (cited by Ferrari),
Thgn. 53; more established is the equivalent pairing of c¦Vb] and kv_g], Il. 3.50, 24.706,
Od. 6.3, 8.555, 11.14, 14.43, Tyrt. fr. 12.15 W., Thgn. 1005, etc.
58 Ti[¢� : ì[t]he use of the deictic ≈kW is frequent in prayers to indicate the object on
which favours are invokedî (DíAlessio in id. and Ferrari 1988: 164 n. 13), cp. O. 13.27,
14.16, P. 8.99, 12.5, I. 6.46.
aTOxW[ZYaU : wreaths are used as metaphors for the bestowal or achievement of excellent
things also at P. 1.48-50 fb_nQ £ gµRQ g–fb] �VVnQTQ k[YcWb £ cVg�fgP ]fWdnQT_í
�SY[TZgQ, Pae. 1.9f. (a closing prayer like ours) HRbÏ]Q kÓ fÏQ SWQWÏQ kR[ÙQ �[Ycfgb £
]R¦]d[gQg] ³QpW]bQ WÃQg_^R] (suppl. DíAlessio 1988), and probably fr. 333a.4-7 Sn.-M.
The realisation that QbQ in the marginal note is not a variant of cRQ followed by the name
of its proposer (Grenfell and Hunt 1908a: 100) but an insertion after ]fWdnQgb]b followed
by È[RQ itself is due to DíAlessio in id. and Ferrari 1988: 164 and n. 15. See ibid. 164-8
(conveniently summarised by Rutherford 2001: 327 n. 87) also on responsion with the
(other) two epodes of Pae. 6(a) and the marginal variants on those verses.
58f. dc[£fcX`Za : appropriate to the constituent of a wreath, however metaphorical: cp.
B. 13.69f. cRQpRVYTQ ]fWdnQgb]bQ £ [�Qp]Y¥[TQ. The third and last occurrence of the
294
epithet in Greek literature is also in the same ode, 228f. æVWbi £ cRQpRVq], with the same
causal overtone as we have here: both health and Fame make their subjects flourish in
every way (see Radt 1958: 192 on P. 9.72 WÃpRVW~ ìGedeihen bringendî). On Pindarís
pervasive use of floral and vegetation imagery see Steiner 1986: 28-39. On the prosodic
instability of the stem (cRQp#V- here and B. 13.228 but cRQp�V- B. 13.69, cp. WÃp�V- P.
9.72) see DíAlessio in id. and Ferrari 1988: 164 n. 14.
59 ÕtYO[Uca] : a request for good health in a hymnís closure comes at Isyll. 60f., Pae.
Erythr. 22-4, Maced. 25f., CA pp. 134, 136f., 139; the word also occurs in the fragmentary
end of Philod. Scarph. 153, which may have likewise been a prayer (Radt 1958: 192).
Here, the reference to health arguably anticipates the transition to Apollo, its divine
bestower, in the final prayer (59-61). Radt also raises the possibility of a reference to the
goddess Hygieia (ibid. n. 4): although not strictly necessary and ultimately impossible to
prove, this interpretation may be borne out by the only other Pindaric occurrence of the
noun at P. 3.73 ÕS^WbRQ Ö Z[P]YRQ, where the epithet arguably carries a hint of divinity,
cp. Simon. PMG 604 = 312 Poltera (Gentili in id. et al. 20064: 416f.).
aVYWFOTO : on the image of shading see chiefly Bona 1988: 10f. on Pae. 1.2f. c[^Q fb]
WÃpP_^Rb ]jbR�YfT £ Q¦\_í ³jgfgQ �cÚ _Yf[R. Both he and Ferrari (in DíAlessio and id.
1988: 176) read the element of covering that it implies as signifying protection, but there is
no hint in either passage that protection is needed against anything, nor a parallel
elsewhere for this connotation of ]jbn�T. ëShadingí those who wear them is simply what
ornaments such as wreaths do, with a frequent suggestion of pleasantness to sight (Archil.
fr. 31 W. ô kY g� j¦_\ £ ‡_gP] jRfW]j^R�W jRÚ _Wfnd[WQR, Semon. fr. 7.65f. W. ZR^f\Q
295
Ö �RpW~RQ �QpY_gb]bQ �]jbR]_YQ\Q, Anacr. PMG 347 fr. 1.1f. jRÚ j[¦_\]], • fgb jRfí
Ú�[¦Q £ �]j^R[�]WQ RÃZYQR, Antigenes FGE 35f. _^f[Rb]b kÓ jRÚ 5¦kTQ �ifgb] £ ]gd�Q
�gbk�Q �]j^R]RQ VbcR[ÏQ ¯pWb[RQ, etc.); the floral imagery in the previous two verses
may also connect the verb with the pleasant shade cast by a tree or bower, cp. Sapph. fr.
2.6f. V. �[¦kgb]b kÓ cR~] ¿ Z�[g] £ �]j^R]fí. The transferral of the verb from the
ornament itself to the person who bestows it is hardly problematic. By echoing 17
jRfn]jbgQ, the reference to shade arguably presents this act of favour requested of the
Aeacids as a mirror image of that bestowed by Zeus on the maiden Aegina at 10-17, with a
hint of the prayer type according to which the recall of favour accorded in the past is
instrumental to the request for its reiteration (da quia dedisti, see Pulleyn 1997: 33, 35f.).
o 59 : on the abbreviation for Aristonicusí name see p. 134 n. 49.
60 n£dcBZX £̀Z[¡[c] : the verb is attested only twice more, and always in a fragmentary
context: Sapph. fr. 21.2 V. WcR�gV\][, iamb. adesp. fr. 38.11 W. fRtfí �c\�¦V\[]]W
(Ferrari in DíAlessio and id. 1988: 177). In the iambic adespoton it seems to mean
ëachievedí, which is consistent with the meaning of the noun �cq�gVg] (see LSJ s.v.) but
unlikely here. Ferrari explains the different construction that the verb has in the two
passages with the common alternation between genitive and accusative that some verbs
show when the object is a neuter pronoun or adjective; I wonder, however, whether the
reason is rather to be sought in the equally common alternation that distinguishes
attainment of the whole and of a part, which would give our occurrence the required sense
ìhaving a shareî (Race 1997: II 271) and parallel the meaning of the noun �c\�gVq� _Y[g]
(Hsch. W 4525 Latte, also at Leg.Gort. (IC IV 72) v.50f., SEG XXII 407). For the thought cp.
296
lyr. adesp. PMG 1001 dR_Ú ∞gcVgjn_TQ Ùgb]¬Q W“ VRZW~Q, see also Pl. Leg. 666d,
Synes. Dion 4 (Rutherford 2001: 328 and n. 89).
dZXXWVY : following a cue by di Benedetto, Ferrari takes the adverb with kYX(Rb) as a
prayer for continuing favourable reception in the future, quoting Ar. Th. 286-8, 949-52, E.
El. 805 (in DíAlessio and id. 1988: 177 and n. 42). Requests for the continuing ability to
carry out an act of worship are a commonplace of Greek prayer, cp. H.Hom. 26.12f.,
Maced. 25f., Aristonous 2.14-17 CA pp. 139, 165, CEG 227.2, etc. (see Pulleyn 1997: 60
n.). Significantly, however, these prayers never involve reception as such, but only the
deityís response to it: reception is invariably restricted to the present occasion. The
irregularity disappears if one recognises another established topos and takes cgVVnjb with
the participle. In the context of a prayer for favourable reception, it is expedient for the
speaker to present the ground on which the addressee should grant it. On the one hand, this
may be simply the fact of offering a performance with which the addressee should be
pleased, as the use of kYZg_Rb entails (see Depew 2000: 62-5); on the other, the speaker
sometimes ostentates his devotion by asserting a connexion with song or worship beyond
the individual ode. This can be effected either generally, by calling oneself such things as
pW[ncgQfR £ Õ_YfW[gQ (sc. of Apollo and Artemis, Pae. 4.45f.), �g^kb_gQ HbW[^kTQ
c[gdnfRQ (6.6, likewise in a prayer for reception), or an �XR^[Wfg[Q £ jn[PjR ]gd�Q
�cYTQ appointed by the Muse (Dith. 2.23-6) (note also Pae. 2.102f., where it is unclear if
the speaker refers to the performance of Pae. 2 itself or to his singing more broadly); or
specifically, by mentioning the acts of worship that one has performed or is in the habit of
performing, cp. *F7.4f. with pp. 313f., F11.5-8 (note pR_n), perhaps F3.9 (see n.), and the
description of such acts performed by others at P. 3.78f. (note pR_n), Pae. 2.97-102
297
(pR_n), 6.15ff. (pR_bQn, in an evident comparison to the speaker), fr. 122.2-4 Sn.-M.
(cgVVnjb, not necessarily to be taken only with the clause that follows, pace van
Groningen 1960: 26), B. 13.86 (fR[dYT]) with Power 2000: 81. Our expression belongs
with the latter group, where cgVVnjb sits comfortably. As Radt 1958: 193 n. 4 recognised,
the speaker is referring not to F5 specifically but to ìregular cult singingî: namely, the
speakerís regular singing that makes him a worthy worshipper and, by implication, F5 a
worthy offering. It must also be noted that the ��cR�gVY�gQ�[R] can indicate both the
performers and the poet, both of which can legitimately lay claim to a long-standing
involvement with choral singing and rely on it to invoke a favourable reception for the
musical artefact that they are jointly producing.
GcYW[ : the circumflex on H4 contravenes the doctrine later propounded by Herodian,
which mandated an acute (pp. III/1 12, III/2 642, 717 Lentz, cp. Choerob. in Theod. pp.
261f. Hilgard, etc.): Radt 1958: 193 n. 3, Forssman 1966: 151 n. 6. ìThe oxytone
accentuation is analogical: [given the original *HRbnTQ] one would expect *HRb�Q, but on
the basis of HRb�Qg] etc. HRbiQ was created (type �jRfg_�RbiQ : -�Qg])î (Ruijgh 1968:
119). It is hard to judge whether the scribe followed a different doctrine of accentuation,
one that kept closer to the wordís etymology and was later drowned out by Herodianís
authority (see Radt 198f. for further examples), or was simply led by the non-contracted
cRb\¦QTQ that he had just copied at Pae. 6(a).121f. and F5.5 to recognise the original
contraction in the nominative. No other instance of this form in the papyrus is accented
(Pae. 2.35, 71, 107; the relevant portion of the word at 4.31 is missing); Mommsen 1866
reports no variant at P. 4.270, the only occurrence of the form in the Epinicians.
298
61 n[[i_h[ : contrary to the annotatorís claim, the adjective indicates that cult singing on
Aegina was established and regularly held (cp. O. 7.84f. �S�QY] fí ¯QQg_gb lgbTf^TQ
with Verdenius 1987: 82f.; A. Ch. 485 kR~fW] ¯QQg_gb �[gf�Q), offering a suitable frame
for the speakerís continuing involvement with it (see 60 n.); see also Ferrari in DíAlessio
and id. 1988: 179, although I am not persuaded that the ìestablished conventions of the
genreî have much relevance here. On a similar note, the Aeginetans are called fR_^Rb kÓ
]gdg^ £ Ùgb]¬Q at I. 9.7f.: the reference cannot be only to song gained with the athletic
victories mentioned immediately afterwards.
nq[[Zd]l[ : I am not as confident as Ferrari is (in DíAlessio and id. 1988: 178) that the last
letter preserved by H5 must be epsilon not theta (for the rounded shape cp. 13 c][gpP[[TQ
on fr. IIIr), so palaeography alone cannot confirm his supplement �¥[Qgc]¬Q beyond doubt:
in principle, Rutherfordís �[RVb]¬Q is equally plausible (said of religious festivities with
choral song-dance at Pae. 6(a).14) and may plausibly be thought to make better sense of 59
Ùgb]¬Q. However, both marginal notes to this verse offer indirect support to Ferrariís
reading: �c¦ in (a) suggests that the referent of �QQ¦_TQ was something envisaged as
coming from the just (not simply ëthe X of the justí), which suits a voice rather more than a
festivity; and if the vox nihili WbVbg_RQ in (b) is a variant that arose from purely graphic
corruption, it seems likelier to be a variant on �Qgc¬Q than on �QQ¦_TQ (Ferrrari 2000:
315, partly anticipated by Radt 1958: 194; Snell 19531: 217 apparently supposed a two-
stage process whereby a variant �QQg_¬Q � itself hardly plausible � was corrupted into
WbVbg_RQ, but this hypothesis is too uneconomical to be attractive.) Ferrari lists several
parallels for the use of �Qgcq with reference to song and music, the closest being Simon.
PMG 519 fr. 35(b).9f. = 100 Poltera (the end of a Paean) where the speakers address
299
Apollo emphasising the �QgcnQ Ö W–dR_gQ that they send forth: there, the reference
seems to be solely to the present occasion rather than to a more general habit, but Ferrari
rightly notes the functional affinity between W–dR_gQ and �QQ¦_TQ.
301
*F6
Introduction
*F6(a) and (b) are preserved by two fragments of H26, a set of matching rolls originally
containing (at least) Hymns, Prosodia, and Paeans.1 Consequently, their identification as a
Prosodion is wholly conjectural, and largely relies on the likelihood of such identification
for *F7, which follows *F6(b) on the same fragment; the occurrence of cRb\¦QTQ at (b).4
may � if it referred to the song itself � suggest that Pindar envisaged it as a paean, but F5.5
confirms that embedded generic clues, however explicit, could on occasion be overruled by
the Alexandrian classifier.2 The provenance of (a) and (b) from one composition relies on
the likelihood that the respective fragments come from the top and bottom of one column,
as argued by the first editor.3 If so, little more than a dozen verses will have been lost
between them.4 The text is too fragmentary for the metre to be recognised; the two
fragments show no sign of responsion, but it is impossible to ascertain beyond mere statis-
tical probability that (b) belongs to an epode.
1 See pp. 13f. 2 See p. 100 above. 3 Lobel 1961d: 35. That fr. 7.18 (= *F7.11) was the last line of a column is possible, but not certain: the blank space underneath can be due to a short verse as much as to the margin, cp. (b).6. 4 ìvv. fere 12î Snell 1959-643: II 58, relying on his figure of thirty-nine verses per column on H26. See pp. 131-3 above.
302
Little can be said about the text. (a) has Olympus or an Olympian (3), something
unspeakable (4), planting (5), something as high as heaven (6), throwing (7), and possibly
the epithet �Rp�jgVcg] (8). Given the frequent metaphorical use of dPfW�T, it cannot be
excluded that 6f. may refer to song, fame, or the like; on the other hand, one notes that the
remains of the passage � including kbT]�PS^Rb], however suspect � would also suit the
description of a mountain covered by woods (possibly, but not necessarily, Olympus), with
gÃ][RQg_n½WR referring either to the trees or to the mountain itself. If so, proximity to
�RVVY�[WQ may suggest that the context is the myth of the Aloadae, but this is only one
possibility among many.
(b) preserves one of the few surviving poem-ends among the Prosodia, although the
last line is wholly lost except the bottom of one letter. What to supply at the beginning of l.
5 is anybodyís guess. Lobelís QbjR]+g[b¬Q can be related to the well-known association
between paeans and victory;5 Snellís kRdQR]+g[b¬Q to a Theban ritual in the honour of
Apollo for which we know that Pindar composed at least two songs (frr. 94b-c) which,
however, were not classified as Prosodia). Rutherfordís ‹]Zg]+g[b¬Q is unattractive for
linguistic reasons; several other possibilities can be thought of, and the matter cannot be
settled at the present state of the evidence.
The community which commissioned the ode is unknown. Whatever the exact
referent of cRb\¦QTQ, it seems to be a plausible inference that the song was addressed to
Apollo, either alone or with Artemis and/or Leto, like F1, *F4, F11, and several other
prosodia beside Pindarís.6
5 Rutherford 2001: 45-7. 6 See pp. 104-6 above.
303
Commentary
(a)
2 : the acute on upsilon is deleted with two dots (Lobel 1961d: 35), but I am unsure
whether the dot level with the base-line of the last letter truly had the same function (ibid.).
4 : ìgà 5\f¦Q not elsewhere in Pindar; gà dRf¦Q Ol. vi 37, Isth. vii 37î (Lobel 1961d:
35).
5 : the simple ‹V�Sbg] is only attested in lexica (Hsch. T 172 Hansen-Cunningham, Suid.
T 78 Adler); Hansen and Cunninghamís doubts about its existence are probably misplaced
(the lemma is in the genitive and thus must come from a context of some sort: for the
principle see Bossi 2005: 317f., cp. Hollis 1998: 63 n. 22), but it would be a hazardous
supplement here. kbT]�PS^Rb] is slightly better, although it is not attested earlier than Is. fr.
27a Thalheim, Pl. Tht. 162a, Lg. 890e, and in poetry not until Call. fr. 713 Pf., AR 4.1258.
As far as the evidence goes, at that stage the adjective means ëimmenseí (LSJ s.v.), which
is not easy to fit to a supposed �gbkR~], pace Lobel 1961d: 35; the sense ëloudí appears to
be later, first in Antiphil. 17.4 Gow-Page, [Orph.] A. 408. Little can be made of the
alternative glosses ³QPk[g] f¦cg] (Synagoge A k 309 Cunningham, Phot. k 632 The-
odoridis) and fÙ _U �cgVV�_WQgQ, an allegedly Aeolic usage (ibid. 683). ‹]�PS^Rb] would
make a Pindaric word (N. 6.44, Rutherford 2001: 423 n. 3), but the first letter visibly ends
with a descender: the only alternatives to lambda are alpha and, just about possibly, mu.
Articulation as a dative plural is not the only possibility, but one would expect something
304
as ambiguous as kbT]��SbR Ò] (cp. Il. 5.245, 7.269, Od. 9.538, 12.175, etc.) to be marked
with a trema.
dPfW�T is more frequently metaphorical (ëbegetí ësow the seeds ofí) than literal (O. 3.33f.,
probably F7.2), see Slater 1969 s.v.; Lobel notes its use with fb_R^ (P. 4.69) and k¦XRQ (I.
6.12).
6 : gÃ[RQg_qj\] occurs once before Pindar, at Od. 5.239 (a fir-tree, cp. Hdt. 2.138.4).
Xerxes allegedly called Mount Athos so (Plut. Mor. 455d). It is said of a voice at Ar. Nu.
357, and becomes something of a commonplace with jVYg]: ibid. 459, AP 7.84 =DL 1.39,
epigram ap. Polem.Hist. fr. 12 FHG III 122, cp. Simon. fr. 11.27 W., Lys. fr. 47 Carey. In
the plural I should think it was used concretely, e.g. of trees (Lobel 1961d: 35) or
mountains, cp. 3 ∆��_[c, 5 dPfWPg[, but the absence of a context precludes certainty.
7 : the ìink not accounted for between W�î (Lobel 1961d: 35) resembles more a casual
smudge than an ìabnormally lowî acute (ibid.).
8 : if the preceding lines refer to landscape, it may be relevant to recall P. 9.101f. �RpP-
j¦VcgP £ I¬].
(b)
4 : the Homeric form, which in epic equally indicates the song (Il. 1.473, 22.391, H.Ap.
518) and the god (Il. 5.899f., Od. 4.232); Pindar restricts it to the former (F5.5, Pae.
6.121f., fr. 140b.9 Sn.-M., cp. B. 16.8, **25.3) while using the ëDoricí HRbnQ indifferently
305
for both (Forssman 1966: 152). Rutherford 2001: 424 sees a reference to cRbnQ-cries
rather than paeans proper (for the difference see ibid. 18-23), but although this is probably
the meaning at Pae. 6.121f. (ibid. 315), the context is insufficient to decide the case here.
Either way, the generic marker is at home in the return to present and song that regularly
marks Pindaric closures. It may refer to *F6 itself or to other ëpaeansí: Pae. 5 ends with an
emphatic self-reference (46-8), but in the identical context of a prayer for reception F5
recalls the speakerís earlier relation to the Muses (60 n.); Pae. 2.96-103 combines other
Apolline performances with the chorusí own in a climactic sequence leading up to the
concluding prayer, and fr. 140b Sn.-M. begins with the similar conceit of the speakerís
comparison of himself and his song to Xenocritus (?) of Locri and his Êg[j[]ÙQ cRbq×[QR
(so supplemented by Ferrari 1990: 233f., but see also Rutherford 2001: 385 n. 12).
5 : Lobelís QbjR]+g[b¬Q is a Pindaric word (O. 10.59, P. 1.59, N. 2.4, 9.49, 10.41) and, in
this context, could evoke the established association between paeans and victory (see
Rutherford 2001: 45-7). Snellís kRdQR]+g[b¬Q would relate *F6 to a Theban processional
ritual in the honour of Apollo for which we know Pindar composed at least two songs (frr.
94b-c), although Procl. Chr. 69 Severyns intimates that they were classified as Partheneia
not Prosodia (see also the unassigned fragment published by Lobel 1961a: 6 n. 1).
kWbcQg]+g[b¬Q, with reference to a rite known in Attica (Is. fr. XLVI Thalheim, Philoch.
FGrHist 328 F 183) and at Ephesus (Men. Kith. 95 with Gomme and Sandbach 1973:
415), is also possible, if not likely on any particular account. However, the plural suggests
something less specific, such as QbjR]+g[b¬Q or DíAlessioís ]fWdRQR]+g[b¬Q (cp. O.
8.10). The ending and accent on the papyrus, if correct, should rule out Rutherfordís
‹]Zg]+g[b¬Q: according to the sources, the rite is the ‹]Zgd¦[bR, neuter plural (Plut. Th.
306
22.4, 23.2, Hdn. Gr. p. II/1 610 Lentz, Alciphr. 1.4.2, 1.11.2), not the feminine singular
‹]Zgdg[^R. The same applies to the Aeginetan Õk[gd¦[bR, so accented by | N. 5.81b p.
III 97 Dr.
cYfRVR is used metaphorically at I. 8.42 (QWbjYTQ) like cYfRVgQ at B. 5.186
(WÃ]kRb_gQ^R], Lobel 1961d: 36), but if the context is QbjRdg[^R or ]fWdRQRdg[^R the
concrete sense may be meant to shine through.
307
*F7
Introduction
The text of *F7 is preserved by one fragment of H26, at the bottom of its only column,
following *F6(b) and arguably *F6(a).1 H26 contained Hymns and Paeans as well as at
least book one of the Prosodia, to which it also contributes most of what survives of F8
and F9; these two are securely identified thanks to partial overlap with H7, which only
contained Prosodia, but such independent confirmation is lacking in the case of *F7. As a
result, the identification of *F7 is necessarily conjectural, and can only be based on
internal criteria. *F6, which must have belonged to the same book, offers little help;
something more can, however, be inferred from the title of *F7 itself and from the
mutilated remains of its text.
The title begins with �][SW^gb], corrected over a clearly mistaken accusative
�][SW^gP]. This makes *F7 the second known Pindaric composition for Argos: no ancient
manuscript preserves the relevant part of the title of Nemean 10,2 but its honourand is
clearly styled as an Argive in the text. Additionally, several other odes of Pindar have been
1 That H26 fr. 7 comes from the bottom of a column is probable (so Lobel 1961d: 36) but not certain: there is a blank under the last surviving line of the fragment, but it may conceivably be due to a short verse in that position (cp. *F6(b).6). On the relation between *F6(a) and *F6(b) see p. 301 above. 2 The title is missing from all the medieval manuscripts, to (re)appear only as a conjecture in Demetrius Tricliniusí Byzantine text: Mommsen 1864: 374. P.Oxy. 5043 fr. 15 confirms that it had a title in Roman times, but only the end c]RVRb]fv�[b is preserved: Maehler 2010d: 83.
308
read as Argive commissions, with varying degrees of likelihood (Dithyrambs 1, 4; F8, F9;
fr. 179 Sn.-M.).3 According to a tradition reported by P.Oxy. 2438 (anonymous Life of
Pindar)4 and the unattributed, late epigram that concludes the Vita Ambrosiana (p. I 3 Dr.),
it was in Argos that Pindar died. What grounds there were for this notion is unknown, but
there is no positive reason to dismiss it as fiction;5 however, it is no reliable ground on
which to assign any individual Argive ode of Pindar, and the fragmentary *F7 least of all,
to a late period in his poetic activity.
The restoration and interpretation of what follows �][SW^gb] is problematical. The
surface of the papyrus is first so badly damaged as to be almost illegible for the space of
one or two letters, then missing altogether for three or four more;6 the text then resumes
with ]] ÌVWjf[� �[, quite certainly the name of the hero Electryon.7 The lacuna is too
wide for Kowalzigís �][SW^gb] é∞�] ÌVWjf[�¼[QR (a form which would suit a Prosodion),
3 F8, F9, Dith. 1 (Snell 19531 with Zimmermann 20082: 41-3, Lavecchia 2000: 93-5, 105, but see van der Weiden 1991: 38f.), 4, fr. 179 Sn.-M. (Wilamowitz 1922: 327); see DíAlessio 2004. On Pindarís involvement with Argos see also Hornblower 2004: 204-6, Morgan 2007 esp. 249-61. 4 See Lobel 1961a: 6, Gallo 1968: 71, DíAlessio 2004: 107f. 5 While favouring an interpretation as a sound historical datum, DíAlessio 2004: 108 alternatively suggests that the author (or his source) may have read one or more Argive compositions by Pindar which they could refer, whether for internal or external reasons, sound or otherwise, to the poetís old age. That the place of a famous personís death may have been recorded and disseminated by a historical source is in itself unproblematic, but one notes that the Life knows no firm date for the event, since it disputes the existing view that Pindar died in 458/7 solely on the basis of O. 4 composed after 452/1 (Lobel 1961a: 5). 6 The extremely uneven writing of H26 does not allow precise estimate of lacunae: see Maehler 2010b: 69 on the companion roll P. Oxy. 5039, and p. 131 above with DíAlessio 2004: 109. 7 Rutherford 2001: 425 n. 1 rightly dismisses the alternative option of the Rhodian heroine Electryone (on whom see TOmpel in RE V (1905) 2318 s.v. 2): she does not seem to have any Argolic connexions, and it is to the family of Electryon that what can be gleaned of the myth (ll. 9ff.) pertains. Electryone is also Hesiodís name for Electryonís daughter Alcmena, who has unquestionable connexions with the saga from which the myth of this ode was drawn, but there survives no indication that she had a cult at or near Argos, and it is hard to see why the editor of the ode should have chosen this rare poetic form instead of her current name (contrast the Attic form πccgjVvb for the textís πccgjVYRb in the title of P. 10).
309
unless deletion and correction in the line is postulated.8 Lobel supposed ì] to be the end of
a noun and ÌVWjf[�T[ to represent a genitiveî, whence Rutherfordís �][SW^gb] ��]�[cÚ]]
ÌVWjf[�¼[Qg].9 As Rutherford acknowledges, however, among Pindarís known works
only the Dithyrambs are usually titled after their subject-matter, and these are an unlikely
candidate for any fragments of H26. Moreover, all extant headings of Pindarís and
Bacchylidesí Dithyrambs have the title proper before the ethnic of the performers, when
the latter is present: Dithyramb 2 �[RjVv]� £ ¢ æY[�W[g] £ r\�R^gb] on H9 fr. 1 col. ii,10
Bacchylides 19 ∏i £ �p\QR^gb], 20 ºkR] £ ÊRjWkRb_gQ^gb] on A coll. xxxviii-xxxix.
Following a different line of thought, Snell had proposed é∞�]� [fÏ]] ÌVWjf[�¼[Qg]
§g[fn] or pP]^R], presumably in order to suit the title-pattern of the Paeans, to which he
attributed the fragment:11 all known Paean-titles refer to sanctuaries, not to festivals, but
there is a parallel in the title of Callimachusí fifth Hymn, W∞] VgPf[Ï fv] HRVVnkg] �
also an Argive composition, whether actually or fictitiously. Differently, again, DíAlessio
supplements é∞�]� [fgˆ]] ÌVWjf[�¼[Qg] cR~kR] or ÌVWjf[P¼[Q^kR], both of which bring
8 Kowalzig 2007: 170, a possibility already denied by Rutherford 2001: 427. Deletion and correction in scribendo would, however, be a plausible way to account for the abnormally long title of Pae. 7b on H26 fr. 14(a) (where only the right-hand margin is preserved: see p. 132 above). On titles of Prosodia see pp. 16f. above. 9 Lobel 1961d: 36, Rutherford 2001: 424, 427. DíAlessio 2004: 109 rightly remarks that no shield of Electryon is known either from the text of *F7 or elsewhere, but it should be noted that Rutherfordís suggestion is purely exempli gratia. 10 Van der Weiden 1991: 61, rightly rejecting Snellís æ]��n�[�R]b] | �[RjVY�×�[] | ¢ æY[�W[g] | r\�R^gb] (Snell 19531: 236). Ferrari 1991a: 3f. (followed by Lavecchia 2000: 30, 125) suggests that the traces on the line above �[RjVv]� be read [r\]|��^g[b], which the scribe would have mistakenly duplicated after the second title. However, the reasons Ferrari envisages for the duplication do not seem altogether compelling; in Bacchylidesí Dithyrambs the indication of the performers, whenever present, always comes after the title proper; and, most importantly, inspection of the papyrus suggests the supposed first line of the title is actually not a line of writing but the bottom half of the asteriskos that separated Dith. 1 and 2 (see pp. 229f. n. 6). 11 Snell 19593-643: II 59 with pp. 12-14, 16f. above on the confusion between Paeans and Prosodia and on their respective titles.
310
the title nearer to those attested for Prosodia.12 Although titles of archaic lyric tend to
avoid the use of the article, DíAlessioís supplement finds an exact parallel in Simonidesí
epinician jYV\fb £ fg~] o∞Rf^gP cRb]^Q (PMG 511 = fr.7a Poltera on P.Oxy. 2431 fr.
1(a)); moreover, an accusative in -gP] at the end of the title would account nicely for
�][SW^gP] erroneously written before it. Although some degree of doubt may persist,
DíAlessioís supplements are the most persuasive. As for the genre, the apparent connexion
of the title with heroic cult independently suggests a Prosodion, granted the presence of
heroes in titles of Prosodia and, on the existing evidence, their absence from the Hymns �
however little we know about these � and the Paeans.13
Known since the Hesiodic Catalogue, Electryon was a son of Perseus and
Andromeda, and Heraclesí grandfather through his daughter Alcmena.14 Different sources
locate him variously in the Eastern Argolid, as a king of Midea or Mycenae.15 His sons,
who always appear as a group variously numbering from four ìand othersî to eight, are
one of those heroic families which, in Martin Westís words, ìexist only to be
slaughteredî:16 their claim to mythical fame seems to rest solely on their fight with, and
reciprocal killing of, their distant relatives the Teleboans from Taphos, sons of PterelaOs,
12 DíAlessio 1997: 41, 2004a: 110. On the sons of Electryon see below. 13 DíAlessio 1997: 41f., Rutherford 2001: 426f. On Prosodia for heroes see pp. 107-11 above. 14 Fr. 195.10 M.-W. = Scut. 3; his name at frr. 135.7, 193.10 M.-W. (P.Cair. 45624, PSI 131) is wholly restored, but guaranteed by the context. Son of Perseus and Andromeda: DS 4.9.1, Apollod. 2.3.5, FD III/1 76 p. 46, etc. (son of Alcaeus and thus Amphitryonís brother in the dubious argg. k-W [Hes.] Scut. pp. 103-7 Rzach). Father of Alcmena: Hes. frr. 193.10f., 19f., 195.10 M.-W. = Scut. 3, E. Alc. 838f., fr. 228a.14f. K., Apollod. 2.4.5, FD III/1 77 p. 46, Plaut. Amph. 99, Tzetz. | Lyc. 839, 1442 pp. 301, 393 Scheer, etc. On Electryon and his family see further TOmpel in RE V (1905) 2317f., DíAlessio 1997: 42. 15 Midea: Paus. 2.25.9. Mycenae: Apollod. 2.4.6 (implying that his possessions included Tiryns and Midea, cp. 2.3.4), whence Tzetz. | Lycophr. 932 p. 301 Scheer. Tiryns is also cited by argg. k-W [Hes.] Scut. pp. 103-7 Rzach. See DíAlessio 2004: 110 n. 15. 16 West 1985a: 29 n. 91; less colourfully DíAlessio 2004: 110.
311
as they were intent on plundering Electryonís cattle.17 In the several versions of the myth
Electryonís death was then caused, wilfully or otherwise, either by his illegitimate (and
only surviving) son Licymnius, or by his prospective son-in-law Amphitryon;18 the latter
went on to conquer Taphos and avenge his brothers-in-law-to-have-been on PterelaOs
himself.19
The only probable attestation of a cult of Electryon or his kin in Argos comes from
the late Hellenistic inscription SEG XLI 282,20 which honours a man for enforcing the
restitution of �W[Ï Zi[R to ìHera and Heracles and Pythaeus and �V\jf[�TQî.21 The
tenure of sacred land strongly suggest that the hero was the object of public cult, for whose
foundation the inscription provides a terminus ante quem. The allotment had been
reportedly made ÕcÙ f�Q �[ZR^TQ, but it is unclear how much earlier a date this entails:
the editor of the inscription plausibly suggests the period shortly after Argosí conquest of
the Eastern Argolid, the area where the myth situates Electryon and his family, in the late
17 Different name-lists are given by Hes. fr. 193.13-5 M.-W. (at least eight), Apollod. 2.4.5 (eight), Tzetz. | Lyc. 932 p. 301 Scheer (four jRÚ ãfW[gb): see Harrauer 1988: 103-7. The cattle raid is described or alluded to in Hes. fr. 193.16-8, 195.20-3 M.-W. = Scut. 17-9 (cp. argg. �-W pp. 102-7 Rzach), Herodor. fr. 15 Fowler = BNJ 31 F 15, A.R. 1.747-51 with | b ad loc. p. 63 Wendel2, P.Vindob. G 23058 I.1-4 with Rabbie and Sijpesteijn 1988: 86-90, Harrauer 1988: 99-101. The problems in Apollodorusí account of the kinship between the Teleboans and Electryon are pointed out by Frazer ad loc., 1921: I 167f.; on the myth see also Gantz 1993: 311, 374, 376. 18 Licymnius, intentionally: | O. 7.49a p. I 212 Dr. Involuntarily: Agias and Dercylus fr. 9 Fowler = BNJ 305 F 9 ibid. Amphitryon, intentionally: Hes. fr. 195.18f. = Scut. 11f., [Hes.] Scut. 82, argg. k-W [Hes.] Scut. pp. 103-7 Rzach (with major variants from unidentified sources). Involuntarily: Pherecyd. fr. 13b/c Fowler = BNJ 3 F 13b-c, Apollod. 2.4.6 (whence Tzetz. | Lyc. 932 pp. 301f. Scheer). Non-committal: E. HF 17, Paus. 9.11.1. Electryon died during the raid itself according to | AR 1.747-51a p. 63 Wendel2; DíAlessio 1997: 42 charges the scholiast with misattributing the story to the Shield, but ô �]fg[^R Ö �Q fvb �]c^kb cR[í �]b¦kTb may also refer solely to the final part of the scholion, which does follow the Shield. 19 Apollod. 2.4.6f. 20 First published (only in part) in Kritzas 1992: 237, connected to *F7 by DíAlessio 1997: 42. 21 The spelling with \ for Attic and epic W is also attested (twice) on the pedestal of the Argive Monument des Rois at Delphi (cited below), which account for all the known epigraphic attestations of his name. The alternative reading �VWjf[P�Qb suggested by Kritzas 1992: 239 as a form of ÌVWjf[PiQ, a purported doublet of ÌVWjf[PiQ\ (Hesiodís name for Alcmena: fr. 195.19, 38 M.-W. = Scut.16, 35, Scut. 86), would appear to be a phantom.
312
460s.22 A statue of Electryon, with his royal ancestors and his progeny through Alcmena,
featured on the monument that the Argives dedicated at Delphi after taking part in the
foundation of Messene in the fourth century (F.Delphes III/1 76, 77 with Pausanias
10.10.5). As for his sons, there is no attestation of their cult in Argos, or indeed elsewhere,
but on a more general level the local worship of groups of dead men (often young and
sometimes warriors) with few to no individual traits but joined by a violent death and
common burial is well known throughout Greece: at Thebes the sons of Heracles
(described by Pindar at Isthmian 4.61-8),23 at Corinth those of Medea,24 at Olympia the
suitors of Hippodamia,25 and many others.
The sacred grove planted in the temenos of the Tyndarids � whose relationship, if
any, with the cult of Electryon or his sons is unclear26 � offers the poet ìsongî (1-3); the
speaker then elaborates on the wealth of songs that surround the city (4f.) in terms that
emphasise their splendour (dVYSW[4, ]YVR] 5) and plenty (�X �jR_��[fg- 5). A reference
to not falling, or not being led, into error (6) presumably marks a shift to the subject-matter
of such songs, instantiated by two references to ì(those who fought) over Troyî (7) and
ìsuch things as once [occurred] in Thebesî (8) in quick succession: a compact priamel that
evokes the two most glorious deeds of war in Argosí past, the Trojan War and the
22 Kritzas 1992: 240. 23 See Schachter 1972: 21f. (with other Boeotian parallels), 1981-94: II 11, Krummen 1990: 59-75. 24 E. Med. 1378-83, Parmeniscus and Didymus ap. | E. Med. 264 pp. II 159f. Schwartz, | O. 13.74g pp. I 373f. Dr., cp. Paus. 2.3.7: see Pfister 1909-12: I 313f., Brelich 1959, 1969: 355-65. 25 Paus. 6.21.9-11. 26 ìPerhaps Electryon had his ô[�bgQ there, or there may have been a sacred grove in the precinct that Electryon was supposed to have foundedî Rutherford 2001: 425. Along the same lines DíAlessio 2004: 111f. infers that both the cult of Electryon and the performance of *F7 took place in the grove. While all this is possible, it is not strictly demanded by the text.
313
expedition of the Seven,27 to serve as foil for the raid of the Teleboans (9 QR�VgZgb) and
the fight over Electryonís cattle (10 ]qVR]RQ, 11 possibly Ì]VWj�[[�TQ-), with which the
mythical narrative proper appears to begin.
The entire opening is played on the two consonant strings of myth and poetry: poets
(l. 3, probably 6), song (4f.), and themes for song (3, 7ff.) are firmly in the foreground, and
it is through these that the selection of the mythical core of the ode is carried out. The
tireless songs that blaze about Argos, the themes that these songs may treat, and the myth
that the ode goes on to narrate form a sequence that projects *F7 onto the cityís poetic, as
well as mythical, background: the story of Electryon is one of the many that a poet can
draw from Argosí cultic present (1-3) or mythical past (7f.) without fear of error, and the
ode itself implicitly finds its place in the beacon of poetry that the city untiringly feeds.
The vagueness of �Qk[Ú ]gd�b and the ambiguity of the present cR[YZWb (3), both of
which can be related to Pindar and the present performance as well as to the broader
scenario of Argive-themed song, add to this combination. Although the priamel is not quite
as grand as the comparable openings of Isthmian 7, Nemean 10,28 or the first Hymn (fr. 29
27 Priamel: Sevieri 1999: 198, Rutherford 2001: 426. Seven: Kowalzig 2007: 172. DíAlessio 1997: 42 connects l. 8 with Amphitryonís Theban exile, which would have been followed by a ìflash-backî on the cattle raid, but a priamel seems to be a neater way to account for the train of thought of the entire passage, and the succinct g∑n cgfW rq�Rb is likelier to point an Argive audience to the expedition of the Seven than to anything else. Along similar lines, the boundary-stone of the enclosure of the Seven in the agora of Argos (SEG 37.283 with Pariente 1987: 595-7, 1992) was inscribed W[ggQ (ô[iTQ rather than ô[�bgQ: Pariente 1992: 200-2) £ fgQ WQ pW£�Rbc. Monument and inscription are datable to the mid-sixth century (ibid. 198-200; before 461, Daumas 1992: 257), and it can be reasonably supposed that the phrase was current in Pindarís time. On the monument see also Boehringer 2001: 142-4. Of possible interest for Pindarís text is also Pausaniasí testimony on a cenotaph of the Argive dead of the Trojan War near the monument of the Seven (2.20.6): if this monument � or some predecessor relating to the same tradition � goes back to the archaic or early Classical age, is Pindar recalling the familiar landscape of the Argive agora when conjuring up the two sequential foils for the theme of his song? On the interrelation of the Trojan and Theban myths in Argos and the Argolid see Cingano 2004, Kowalzig 2007: 165-8. 28 DíAlessio 2004: 112 notes several similarities between the openings of *F7 and N. 10: the ìcatalogicî structure and the use of dVYST (N. 10.2), c¦Vb] (1), and fY_WQg] (19). Noteworthy, however, are also the differences in perspective, most evidently in that N. 10 only touches on the Trojan and Theban expeditions
314
Sn.-M.), which respectively evoke seven, nine, and (at least) eight possible themes to
highlight their final choice, the card of conscious self-positioning within a poetic tradition
is played much more overtly here: of the three parallels just cited, only dVYSWfRb at
Nemean 10.2 (another Argive ode) moves, and not quite as explicitly, along similar lines.
If the words now lost at lines 4f. allowed the singing portrayed there to be understood more
broadly as performed in Argos or by Argives, as well as more narrowly concerning Argos
from a purely thematic viewpoint (see 4 n.), then the sentence has a particular edge in that
it reactivates, as it were, the capital of favour built up for the city by such performances
and add it to the present one so as to enhance its standing in the eyes of the worshipped
being, like the functionally identical cgVVnjb in the finale of F5 (see 60 n.), and indeed in
the eyes of the admiring human audience too.
The choice of the Trojan and Theban myths for the foil of lines 7f. sets a fittingly
warlike tone for the legendary battle that presumably followed in the narrative, and extols
this otherwise (apparently) obscure episode to the loftiest heights of Argive gallantry; it is
unknown whether the cult of the Electryonids (if one existed in Argos) had military under-
tones of some sort, but if true this would tally quite nicely with this sequence of myths and
with the mood that it expresses. How the narrative may have unfolded after the fragment
breaks off is anybodyís guess: it seems plausible that it will have included at least the death
of Electryonís sons, but even that much is conjecture. Ian Rutherfordís view that the story
may have proceeded to include the birth and first heroic feat of Heracles one generation
indirectly, through the fate of Diomedes and AmphiaraOs respectively (ll. 7, 8f.). On the other hand, the role of the Dioscuri in N. 10 � theirs is the myth narrated in an ode which lacks a concluding ëZí section entirely, and to them and their cult the victorís family can boast a very special connexion � can perhaps be connected to the presence of their sanctuary in the opening of *F7, whether or not DíAlessioís hypothesis of the introduction of a cult of Electryon to the very temenos of the Tyndarids in Argos (2004a: 113) is accepted. (On the sanctuary see 1 n.)
315
further (with F8 belonging to the final part of *F7) is possible, but not immediately
persuasive given the considerable metrical difference between the two fragments.29
The surviving text gives little help in determining the cultic situation of the song. No
addressee seems to be mentioned, directly or otherwise, in the opening lines: the priamel
structure, which naturally resists the early disclosure of its climactic term, and the
insistence on poetry and myth rather than on worship per se, conspire to bridge the opening
immediately to the central mythical narrative. It is possible that the ode had a
predominantly narrative focus, and that its relationship to the cult to which it was linked by
the Alexandrian editor was not described at any great length in the text itself; a narrative
emphasis would go well with the (almost) wholly dactylo-anapaestic metre of the
fragment,30 which recalls the austere, epic-sounding stanzas of Alcman and Stesichorus
more than anything else in the known Pindar. (If g∑R can indeed be scanned as two short
syllables, nothing precludes responsion between 1-5 and 7-11, but the text is too mutilated
to offer any acceptable degree of proof.) An explicit reference to the reality of the
performance may have been made only in the concluding section: this is unusual for
Pindar, but not inconceivable (compare Bacchylides 17). Much more surprising would be a
poem with no frame for the myth at all, which one supposes would rather have been
classed as a Dithyramb.31
29 Rutherford 2001: 402, 426 with DíAlessio 1997: 42, 2004a: 114. On the other hand, regardless of any connexion between *F7 and F8, it is not unlikely that an emphasis on Electryon and his family may have been connected to an Argive attempt to appropriate Heracles by highlighting his Argive ancestry, as patently happens in N. 10.11, 13-8. For a similar, later attempt see the ëMonument of the Kingsí just mentioned (FD III/1 41-54, Paus. 10.10.5); Dalviat 1965 (esp. 309 and n. 3) argues that Argosí claim to Heraclesí parentage can be read as representing the bond between it and Thebes on that occasion. 30 See 8 n. 31 See pp. 53f. n. 39 above.
316
As is the case with most of the non-epinician Pindar, the date of *F7 is unknown.
However, if a scenario that has been envisaged by Giambattista DíAlessio for its cultic
background and composition is correct, a tentative dating within reasonably specific limits
becomes possible, if necessarily very speculative. Surviving testimonies to the myth
intimate that Electryon and the Electryonids had a distinctly Eastern Argolic identity,
centred around Midea and Tiryns but � as far as the evidence goes � with few direct links
to Argos.32 On the other hand, Pindarís choice of myths for the foil at lines 7f. corroborates
the indication found in the title that the cult to which this ode pertained was either Argive,
or firmly set within Argosí sphere of influence: while several places through the Argolid
can boast a connexion with the Trojan War, the expedition of the Seven does not seem to
belong to the mythical landscape of the eastern part of the region, and its presence in a
strictly local Eastern Argolic cult-song wholly unconnected to Argos would be quite
surprising. Given these seemingly contradictory elements, it is tentatively possible to link
the ode to Argosí incorporation of its neighbours to the east of the Inachus (and,
accordingly, of their cults) in the second quarter of the fifth century:33 ìthe song looks as if
it is celebrating something like [Electryonís] temporary appropriation by Argos perhaps in
the context of the destruction of whichever city hosted his cult in the first place, Tiryns or
Midea, both plausible candidatesî (Barbara Kowalzig).34 Diodorus dates the fall of
Mycenae to 468/7; that of Tiryns, and probably the incorporation of Midea, will have
32 DíAlessio 2004: 110f. 33 DíAlessio 2004: 109-13f., Kowalzig 2007: 170-8. 34 2007: 171.
317
happened shortly afterwards.35 If all this holds together � and especially if it is legitimate to
exclude that Electryon and his family had connexions to Argos before the conquest � then
*F7 will have been commissioned within the last two decades of Pindarís career, between
the mid-460s and the mid-440s.
35 Mycenae: D.S. 11.65, cp. Paus. 7.25.5f., 8.27.1, Str. 8.6.19, see Moggi 1974: 1251-3. Tiryns: Paus. 2.25.8, 27.5, 8.27.1, 46.3, Str. 8.6.11, a terminus post quem being the Olympic festival of 468, where a Tirynthian competitor is registered as such (P.Oxy. 222 col. i.42 with Grenfell and Hunt 1899: 93). Midea: Str. ibid., St.Byz. p. 451 Meineke, cp. | Theoc. 13.20 Wendel1. See Kiechle 1960, Tomlinson 1972: 102-5, Moggi 1974: 1251-6, Hendriks 1982: 87-98, PiCrart 1997: 329-31 (it is doubtful whether Midea was incorporated through military conquest or synoecism: Moggi 1974: 1259).
318
Commentary
1 -e[]Nc]YNl[ : to call both brothers Tyndarids is no more exceptionable than to call
both Dioscuri, without genealogical implications in either case: Pollux is a Tyndarid at N.
10.73, Castor a son of Zeus at P. 4.171f. (pace | N. 10.150a p. III 182 Dr., over-simplistic).
The [nQRjW (always so called) were widely worshipped in archaic and classical Argos
(evidence collected by Moretti 1998: 237-9, see also Marchetti and Kolokotsas 1995:
218f.). There were at least two temples dedicated to them in its territory, and possibly a
third (see 2 n.). Six dedications to them have been found in the city (IG IV 561, 564, 566,
Roux 1956: 388, SEG XXVI 428, XLIV 418), and arguably one more originates from there
(IG V/1 231 =CEG 363: Moretti 238 and n. 29). They are portrayed in two reliefs, one
found in the area of the Hellenistic theatre (Roux 488, Moretti 237f.), one in the Eastern
part of the city (Moretti 238). Vatin 1982 and Faure 1985 argue that the two Delphic
kouroi commonly known as Cleobis and Biton, the work of Po[l]ymedes and another
Argive sculptor, are connected by their inscription rather to the ìLordsî ([RQR\gQ),
although on that hypothesis there is no evidence that they were an Argive dedication as
well as an Argive product (as Hdt. 1.31.5 would prove if they were his Cleobis and Biton;
a suggestion in that direction in Faure 63 n. 14). In Pausaniasí time the [nQRjW had at least
two sanctuaries in the cityís territory, and there may have been one more at an earlier date
(see 2 n.). DíAlessio 2004: 112f. argues that the long-standing connexion between the
Tirynthian maternal family of the Argive Theaios and the Dioscuri (ÁPQkR[^kRb]) asserted
at N. 10.37f., 49-51 suggests a hereditary priesthood of sorts, which in turn may imply the
319
continuation in Argos of an originally Tirynthian cult of the Dioscuri. In any case, Pindar
did dedicate to them the mythical narrative of N. 10, a narrative emphasised by the
anomalous closure of the ode without a return to the actualit� at the end; on the religious
politics that this choice may express see Hornblower 2004: 205f. Later in time, an
honorary inscription from the Roman era styles a man as a descendant of Perseus and the
Dioscuri (IG IV 590), and an Argive issue of Septimius Severus shows the Dioscuri on the
reverse (SNG Cop. Argolis 100). Plu. Mor. 296f relates an Argive claim (whose date
cannot be determined) that Castor, called _bXR[ZRSYfR], was buried in the city; he and his
brother had indeed relocated to Argos in a tradition related by | P. 1.127b p. II 22 Dr., cp.
P.Vindob. Gr. 29817 (MPER NS I 23) with McNamee 2007: 308.
2 TO_`][OY : �W[ÙQ fY_WQg] designates the Athenian acropolis in Ar. Lys. 482; cp. P.
4.204 ÚSQ¦Q Ö fY_WQg] (Bona 1988: 281), B. fr. 4.53 M. fY_WQg] �n�p¥WgQ.
In his description of the Argive territory, Pausanias mentions two archaic sanctuaries of the
Dioscuri, whom he calls both }b¦]jgP[gb and êQRjfW] in both cases: a temple in the city
(2.22.5) with ebony and ivory �SnV_RfR by Dipoenus and Scyllis (early sixth century
according to Plin. nat. 36.9), and a sanctuary off the road to Lerna (2.36.6) which had
X¦RQR of a similar making. PiCrart 1982: 146 links the city temple to SEG XXVI 428 (dated
by Kritzas 1972: 205 n. 20 to the turn of the fourth century), which mentions a dedication,
a fY_WQg], and the [nQRjW (possibly the dedication of the fY_WQg] itself?), although
uncertainty about the inscriptionís archaeological context prevents one from locating the
building precisely (PiCrart 1998: 341, correcting 1982: 145 pl. 2). Moretti 1998: 239 argues
from the relief mentioned at 1 n. that another sanctuary stood in the area subsequently
320
taken up by the theatre. DíAlessio 2004: 112 plausibly suggests that the reference here is to
the first of these.
The further (if somewhat less likely) possibility must be mentioned that ìthe holy precinct
of the Tyndaridsî is not an actual sanctuary but Argos itself: fY_WQg] is so used, with a
deity in the genitive, also at P. 2.1f. (Syracuse), 4.56 (Libya), Il. 2.696 (Pyrasos), cp. ³V]g]
at Il. 2.506 ~H.Ap. 230 (Onchestus) and perhaps O. 13.109 with | 156b p. I 386 Dr.
(Aegina); Argos herself is �[SW~gQ Ö fY_WQg] ìthe Argive plotî at N. 10.19 (CannatØ
Fera 2004: 99f.), see also Lobel 1956b: 54 on B. **23.2 M. The associations noted above
between the city and the twins might justify the metonymy; the common use of �W[¦] with
reference to cities, originally an epic trait (LfgrE s.v. 2, Slater 1969 s.v. 2.a, note especially
O. 2.9), could also suit this view. One wonders whether Argos contained only one ëplanted
groveí that was instantly identifiable as such; the answer may well be yes.
2 dOxeTOe_`[Z[ �[XaZa : Lobelís supplement is brilliant but speculative, even more
than fW_Y]QWb. If it is correct, we are faced with either a real ë(sacred) groveí or more
generally a ësanctuaryí, with or without trees: for ancient criticsí awareness of this
ambiguity see | O. 3.31a-b, 10.53d pp. I 113, 323 Dr., Str. 9.2.33, cp. | fr. 140a(a) col.
i.13 p. 379 R. (a usage that goes as far back as Il. 2.506 ~ H.Ap. 230). Trees are certainly
meant at O. 8.9 H^]R] W–kWQk[gQ Ö ³V]g] (Slater 1969 s.v.) and, proleptically, 3.17f.
(Aristonicus in | 31a p. I 113 Dr.); a grove in a sacred precinct is probably the most natural
way of understanding the phrase here. Given the common metaphorical use of dPfW�T
ësow the seeds ofí (see Slater s.v. c), it cannot be excluded altogether that Pindar may be
playing on the double meaning of both words and referring to a constructed building as a
321
ìplanted groveî, but elsewhere in his work the verb never has a concrete sense (ëbuildí or
the like) when used metaphorically. There is, I believe, a more concrete possibility of an
intentional ambiguity in the syntax of cWdPfWP_YQgQ, with the intuitive ³[V]g] and with
_YVg] in the next line: given what is said from 4 onward on the abundance and splendour
of Argive myth and song-making, it would be only fair to claim that the (matter for) song
that the sanctuary (?) provides to the poet is ëalready plantedí for him.
3 \[N]]Ú aZx�Y : signally the poet (Lobel 1961d: 37), but without losing sight of the
broader skilfulness that the term specifically implies: Gianotti 1975: 96f. and n. 45 (see
also F2.ii.15 n.).
dc]`^OY _`XZa : ìsupplies the poet with a themeî (Lobel 1961d: 37), cp. N. 6.32
cR[YZWbQ cgVˆQ —_QgQ, perhaps Pae. 4.24 Ùg~]RQ cR[YZTQ âVb] with Cassio 1972 (but
I am not fully persuaded that the reference is not to the illustrious Cean poets, see Carey
1991: 15), E. Tr. 1244f., Su. 1225 (cited by Cassio). The vague phrase, however, leaves
room for a less specific interpretation, e.g. ìprovides an occasion for songî (a natural
statement in reference to the sanctuary in which the ode itself was performed, if this is
truly the implication of the opening statement), or indeed ìgives inspiration for songî: for a
fiction of improvisation dictated by external circumstances cp. I. 1.12, 2.1-5, Pae. 6.7-11.
4 \_xÚ diXY[ : both ëthroughout the cityí (cp. Il. 11.706, A. Th. 843 with Klotz 1917-18:
619f., Hutchinson 1985: 187) and ëabout, regarding the cityí (cp. Od. 11.510); given the
focus on Argos as both subject and object of song (4f., 6-8, see pp. 313f.), the ambiguity is
probably deliberate.
322
xX`tO[ : praise is often construed as flame, just as glory is often construed as light (see
F5.4 n.), see Bowra 1964: 34. For the blaze of song cp. O. 9.22 _RVW[R~] �cbdVYSTQ
�gbkR~], I. 4.43 â°Rb cP[]ÙQ —_QTQ (very close to our —_QTQ ]YVR], 5), B. fr. 4.80 M.
—_Qgb dVYSgQfRb (Lobel 1961d: 37), P. 5.45 †]jg_gb dVYSgQfb mn[bfW], I. 7.23
dVYSWfRb kÓ ∞gcV¦jgb]b Ùg^]Rb] (Bona 1988: 281), Pae. 2.67 WÃRSg[^Rb]b dVYSWb. If
the metre was consistently dactylo-anapaestic, the middle dVYSW[fRb (cp. N. 10.2f.
dVYSWfRb kí �[WfR~] £ _P[^Rb], I. 7.23, B. fr. 4.80 M.) is likelier than the intransitive
dVYSW[b (cp. N. 6.37f. mR[^fTQ Ö ¡_nkTb dVYSWb, Pae. 2.67), the foot of whose iota,
moreover, I would expect to see immediately before the papyrus breaks off.
5 ][ : if the predicate is dVYSW[, probably an adjective qualifying either ]YVR] or —_QTQ.
The metre suggests the latter.
—_[h[ : ìused in a quite general senseî (Lobel 1961d: 36), as always in Pindar, see Slater
1969 s.v.
\Vc���[TZ- : the hyphen in H26 indicates that the word is a compound; very little is left of
the last two letters, but enough to exclude a compound of ³jR_cfg], which is also
metrically unlikely. The sense may have been ìfrom untiring mouthsî (Lobel 1961d: 37),
cp. Hes. Th. 39 �jn_Rfg] Ö RÃkq (the Musesí), [Sapph.] FGE 673 dTQÏQ �jR_nfRQ,
and the ëuntiring stoneí that preserves and repeats the sepulchral epigram CEG 108 (5f.). If
it referred to the sum total of Argive performances, or performances about Argos, rather
than to that of *F7 specifically, the idea is not as outlandish as Rutherford 2001: 425f.
suggests, cp. Pae. 7b.21f. with id. 249f. Otherwise one can think of the distant, rather than
the immediate, source of the ëblaze of songsí and understand ìfrom untiringly-fighting
323
heroesî or similar (�jR_RQfgV¦SZR] I. 7.10, -_nZR] P. 4.171, Pae. 22(f).6, -Zn[_R] fr.
184 Sn.-M.), which would be quite appropriate in view of the military emphasis of the
following priamel (see pp. 312f.); however, I can find no examples in Pindar of �j so used.
Arguably, an echo of the flame imagery evoked by dVYSW[ and ]YVR] still reverberates
thanks to the epic �jn_RfgQ ct[ (Il. 5.4, 15,731, etc., Od. 20.123, 21.181, and beyond).
6 ] �£P QR_O[Za : I am perplexed by Snellís choice of a future participle. The acute may have
stood on the second as well as the first letter preceding mu, if neither was very wide; if so,
the letter immediately before mu need not have been a vowel (rho or sigma are valid
alternatives). Both Snell and Rutherford 2001: 425 assume a first-person subject, but it
may also be an indefinite third person (cp. I. 4.41), ëa poetí (cp. 3), or the like.
\¥_¦dXcV[Uc[ : Pindar has the form with mu six times in the Epinicians; the form without
mu is only found in Attic drama, and always metrically required when present. Its
occurrence in H26 may be further evidence that *F7 was not entirely dactylo-anapaestic, or
a mere scribal oversight of a rather common sort, see Friis Johansen and Whittle 1980: II
185; despite Youtieís Law (see Merkelbach 1980, 2003, Gonis 2005), I incline towards the
latter and accept Whittleís emendation.
Much like Snell and Rutherford 2001: 425, I suppose the sense to have been along the lines
of ìI would not (wander) into error (if I told how the heroes strove) over Troy (or) Öî,
with g– jWQ jfV. essentially a periphrasis for g– jWQ Ú_n[fgb_b. Other restorations with a
similar focussing effect can be found, e.g. ìI would not (wander) into error (and tell how
the heroes strove) over Troy (or) Ö (but rather I shall tell) of when the pirates Öî (Bruno
Currie, in discussion): this would suit Pindarís usual way of dismissing all the elements
324
evoked in a priamel while picking a new one (cp. the openings of Nemean 10 and Isthmian
7), but expansion on the last element of the priamel itself is paralleled at the opening of
Olympian 2 and possibly fr. 29 Sn.M.; here, jR^ (9) arguably suggests a further element in
a list rather than a fresh start.
7 [�]c�Nc[UcY : simply ìTroyî as at Pae. 6.90, Il. Parv. fr. 28 BernabC, without the
historical stratigraphy envisaged by Il. 20.216. With the dative, cW[^ more probably
indicates the object at stake (Slater 1969 s.v. c.R) than a purely local determination, pace
Rutherford 2001: 425.
8 Z∑W : breathing and both accents are written on the papyrus: the acute on alpha is almost
horizontal (cp. 6 ¥]� ¥P QR_WQg]), but impossible to interpret as a longum given the circumflex
on iota. gµR (feminine singular) would give a more regular metrical sequence, but
reinterpretation in such a fragmentary context is unsafe (DíAlessio 1997: 42 n. 120), and
the probable sense of the passage supports a neuter plural. (For g∑g] practically equivalent
to ≈] cp. P. 1.73.) This is the only clear obstacle to a wholly dactylo-anapaestic interpreta-
tion of the fragment, but it can perhaps be obviated by also scanning gb as short, a Homeric
quirk (Il. 13.275, 18.105, Od. 7.312, 20.89 with Chantraine 19866-86: I 168): resolution of
the princeps in a dactylic sequence, although rare, is attested at I. 3.63, Ibyc. PMGF 285.3,
and Attic drama (see Fraenkel 1918: 178f., Natale 2011), and is no more shocking in an
anapaestic environment (see West 1982: 95). If this is too much oddity for a single word,
one will regard the metre as only ëlargelyí dactylo-anapaestic (or Aeolic with Snell 1959-
643: II 59, although such description is not free from problems either).
325
6¨BcY : Pindar uses the singular with reference to the city also, at least, at P. 4.299 and I.
3.12 (pace Slater 1969 s.v. 1, I doubt whether she, and not the nymph Theba, is meant at fr.
195.4 Sn.-M.); the plural is some ten times more frequent (see Slater s.v.). DíAlessio 1997:
42 connects this line with Amphitryonís Theban exile, which would have been followed by
a ìflash-backî on the cattle raid; however, a priamel on Argosí heroic glories as outlined
above (p. 312-14) accounts more smoothly for the train of thought of the passage, and the
pithy phrase is likelier to have pointed an Argive audience to the expedition of the Seven
(Kowalzig 2007: 172), or perhaps both the Seven and the Epigoni, than to anything else.
For g∑n cgfW rq�Rb cp. the sixth-century inscription on the boundary-stone of the
enclosure of the Seven in the Argive agora, SEG XXXVII 283 W[ggQ £ fgQ WQ rW£�Rbc
(Pariente 1987: 595-7, 1992, Boehringer 2001: 142-4), Aeschylusí �cfÏ �cÚ rq�Rc ñ
whose title was current at least by the late fifth century (Ar. Ran. 1021), ñ Isocr. 4.55 fvc
cf[RfW^Rc fvc �cÚ rq�Rc, etc.
9 [cgXZ^ZY : ìattested only of places where ships can ride, not of persons lurking in ships.
(The accent would then I presume be QRPV¦Zgb.) The verb QRPVgZW~Q, however, is used of
both persons and ships lying in waitî (Lobel 1961d: 37, see also Bona 1988: 282, Pavese
1993: 155). The accent is no great problem: if you are ZRVj¦cPVgc when your c�VRb are
ZRVj¦c, so you can be QR�VgZgc when your V¦Zgc (the ambush-place as much as the
ambush itself, see LSJ s.v. 1) is a QRtc. The Taphiansí naval prowess is stressed by Hes.
Cat. fr. 193.16 M.-W. QRPc^jVPfgb, which our QR�VgZgb seems to be specifically
revisiting, perhaps with a hint of Od. 15.427 V\Ícfg[Wc, cp. E. fr. *87b.8 K. V\bcfR^ (see
Stoneman 1977: 479).
326
10 [n][[g^YZ[ V]exc[ : in this metrical context, j[�dR (cp. N. 9.33) appears likelier
than j[Pd¬[b or a form of j[PdR~gc (the lectional sign which was written on the alpha is
too damaged to be decisive). Thomas Coward points out to me the stark contrast between
the blazing imagery of the opening section (4f.) and the suddenly dark and hushed
ambience at the start of the mythical narrative.
ìThe robbery may have been carried out at night. It was not carried out secretly Ö in any
version of the story that we haveî (Lobel 1961d: 37). If an armed attack on a kingís cattle
and its custodians is hard to carry out secretly, its preparations and the approach to its
target can, and this is what j[PdR[ may have indicated. However, Greek mythical cattle
thefts are sometimes carried out wholly in secret and only discovered later (e.g. the
Dioscuri and Apharetids in Cypr. ap. Procl. fr. b 106f. Severyns, cp. the Homeric Hymn to
Hermes): although in AR 1.745-51 and Apollod. 2.4.5 the fight breaks out immediately
upon the Teleboansí attempt to steal the cattle, ]qVRcRQ suggests that in Pindarís account
the theft did take place initially, with the fight and mutual slaughter occurring during an
attempt by the brothers to recover the herd. This would also avoid casting an entire
generation of princes as cattle-herds, a perfectly respectable occupation for younger royalty
in the world of epic (cp. Il. 11.104-6, 20.90-2, 188-90, Od. 13.222f., H.Ven. 53-5, etc.) but
one on which the enlightened fifth century may have frowned somewhat. However, the
verb may also have had a different object from the cattle, cp. e.g. O. 10.66.
327
BIBLIOGRAPHY
329
AdorjUni, Z. (2011): Auge und Sehen in Pindars Dichtung (Spudasmata 139), Hildesheim/ZOrich/New York
Ag`cs, P. (2012): ëPerformance and genre. Reading Pindarís j�_gbí, in Ag`cs, Carey and Rawles (eds.) (2012): 191-223
®, Carey, C. and Rawles, R. (eds.) (2012): Reading the Victory Ode, edited by ®, Cambridge
Alcock, S.E. (2002): Archaeologies of the Greek Past. Landscape, Monuments, and Memories, Cambridge
Alpers, K. (20132): Untersuchungen zu Johannes Sardianos und seinem Kommentar zu den Progymnasmata des Aphthonios (ABWG 62), Braunschweig
Aravantinos, V. (2010): The Archaeological Museum of Thebes, Athens; also online at <http://www.latsis-foundation.org/megazine/publish/ebook.php?book=64>
Argoud, G. and Guillaumin, J.-Y. (eds.) (1998): Sciences exactes et sciences appliqu�es I Alexandrie (IIIe si@cle av. J.-C. - Ier si@cle ap. J.-C.). Actes du Colloque International de Saint-Ltienne (6-8 juin 1996), Saint-Btienne
Arrighetti, G. (1967): ëLa biografia di Pindaro del Papiro di Ossirinco XXVI 2438í, SCO 16, 129-48
® (1987): Poeti, eruditi e biografi. Momenti della riflessione dei Greci sulla letteratura, Pisa
Asheri, D. (2003): Erodoto. Le storie, VIII: Libro VIII: La vittoria di Temistocle, a cura di ®, commento aggiornato da P. Vannicelli, testo critico di A. Corcella, traduzione di A. Fraschetti, Milano
Asmonti, L. (2004): ëIl retore e il gabelliere. Il ruolo di Democare di Leuconoe nella trasmissione dellíideale democraticoí, Acme 57/3, 25-42
Athanassaki, L. and Bowie, E. (eds.) (2011): Archaic and Classical Choral Song. Performance, Politics and Dissemination (TiC suppl. 10), Berlin / Boston
Bagnall, R.S., Frier, B.W. and Rutherford, I.C. (1997): The Census Register P.Oxy. 984: The Reverse of Pindarís Paeans (Papyrologica Bruxellensia 29), Bruxelles
Bagordo, A. (2003): Reminiszenzen frBher Lyrik bei den attischen Tragikern. BeitrGge zur Anspielungstechnik und poetischen Tradition (Zetemata 118), MOnchen
Baldi, D. (2013): Etymologicum Simeonis S-W, recensuit et prolegomena adiecit ® (Corpus Christianorum. Series Graeca 79), Turnhout
Barrett, W.S. (1956): ëDactylo-epitrites in Bacchylidesí, Hermes 84, 248-53 = (2007a), 314-21
® (1962): Euripides. Hippolytos, edited with introduction and commentary by ®, Oxford
® (2007a): Greek Lyric, Tragedy, and Textual Criticism. Collected Papers assembled and edited by M.L. West, Oxford
® (2007b): ëTwo studies in Pindaric metreí, in (2007a), 118-206
Barron, J.P. (1964): ëReligious propaganda of the Delian Leagueí, JHS 84, 35-48
Beck, H.-G. (1959): Kirche und theologische Literatur im byzantinischen Reich, MOnchen
Bekker, I. (1825): Scholia in Homeri Iliadem ex recensione ®, tomus prior, Berolini
330
BClis, A. (1988a): ëA proposito degli «inni delfici» ad Apolloí, in B. Gentili and R. Pretagostini (eds.), La musica in Grecia, Roma/Bari, 205-18
® (1988b): ëLes termes grecs et latins dCsignant des spCcialitCs musicalesí, RPh III 62, 227-50
Benedetti, F. and Grandolini, S. (eds.) (2003): Studi di filologia e tradizione greca in memoria di Aristide Colonna, I-II, Napoli
Benelli, L. (2012): ëZu P. Oxy. 2451 A frr. 1-2í, RFIC 140, 62-86
Benveniste, B. (1950): ëLa phrase nominaleí, BSL 46, 19-36 = Probl@mes de linguistique g�n�rale I, Paris 1966, 151-67
® (1969): Le vocabulaire des institutions indo-europ�ennes, sommaires, tableau et index Ctablis par J. Lallot, I: Lconomie, parent�, soci�t�, II: Pouvoir, droit, r�ligion, Paris
Bergk, T. (18431) (18532) (18784): Poetae lyrici Graeci recensuit ®, Lipsiae
® (1871-84): Griechische Literaturgeschichte, I-III. Berlin
Berlinzani, F. (2004): La musica a Tebe di Beozia. Tra storia e mito, Milano
Bernardini, P. Angeli (1968): ëPindaro e la storiaí, QUCC 6, 136-40
® (ed.) (2004): La cittI di Argo. Mito, storia, tradizioni poetiche. Atti del convegno internazionale (Urbino, 13-15 giugno 2002), Roma
® (2009): ëLíInno primo di Pindaro e la sua destinazione cultualeí, Paideia 44, 73-89
Bernhardy, G. (1845): GrundriS der Griechischen Litteratur; mit einem vergleichenden Ueberblick der RTmischen, II: Geschichte der Griechischen Poesie, Halle
Birt, T. (1882): Das antike Buchwesen in seinem VerhGltniss zur Litteratur, Berlin
Blanchard, A. (ed.) (1989): Les d�buts du codex. Actes de la journ�e dí�tude organis�e I Paris les 3 et 4 juillet 1985 par líInstitut de Papyrologie de la Sorbonne et líInstitut de Recherche et díHistoire des Textes (Bibliologia 9), Turnhout
Blaydes, F.H.M. (1883): Aristophanis Aves annotatione critica, commentario exegetico, et scholiis Graecis instruxit ®, Halis Saxonum
Blomart, A. (2004): ëTransferring the cults of heroes in ancient Greece: A political and religious actí, in E. Bradshaw Aitken and J. Berenson Maclean (eds.), Philostratusí Heroikos. Religion and Cultural Identity in the Third Century C.E., Atlanta, 85-98
Blum, R. (1977): Kallimachos und die Literaturverzeichnung bei den Griechen, Frankfurt am Main (Engl. tr. by H.H. Wellisch, Kallimachos: The Alexandrian Library and the Origins of Bibliography, Madison WI, 1991)
Boeckh, A. (1811-21): HbQkn[gP fÏ ]T�¦_WQR. Pindari opera quae supersunt, textum in genuina metra restituit et ex fide librorum manuscriptorum doctorumque coniecturis recensuit, annotationem criticam scholia integra interpretationem latinam commentarium perpetuum et indices adiecit ®, I/1-II/2, Lipsiae
Boehringer, D. (2001): Heroenkulte in Griechenland von der geometrischen bis zur klassischen Zeit (Klio Beihefte n.F. 3), Berlin
Boesch, P. (1908): rWT[¦]. Untersuchung zur Epangelie griechischer Feste, GNttingen
Bogen, K. (1969): Gesten in BegrBSungsszenen auf attischen Vasen, Bonn
Bona, G. (1988): Pindaro. I peani. Testo, traduzione, scoli e commento, a cura di ®, Cuneo
331
Borg, B.E. (2010): ëEpigrams in archaic art: the ëChest of Kypselosíí, in M. Baumbach, A. Petrovic and I. Petrovic (eds.), Archaic and Classical Greek Epigram, Cambridge, 81-99
Borges, C. and Sampson, C.M. (2012): New Literary Papyri from the Michigan Collection. Mythographic Lyric and a Catalogue of Poetic First Lines, Ann Arbor
Bornemann, L. (1928): ëPindar (1903-1927) und Bakchylides (1908 ff.)í, JAW 216, 131-86
Bossi, F. (2005): ëRicognizione di eventuali adespota in Esichioí, Eikasmos 16, 317-23
Bourboulis, P.P. (1949): Apollo Delphinios, Thessalonike
Bousquet, J. (1959): ëInscriptions de Delphes (1)í, BCH 83, 146-92
Bowie, E. (forthcoming): ëRediscovering Sacadasí
Bowman, A.K. et al. (eds.) (2007): Oxyrhynchus. A City and its Texts, London
Bowra, C.M. (1933): ëEarly lyric and elegiac poetryí, in J.U. Powell (ed.), New Chapters in the History of Greek Literature III, Oxford, 1-67
® (1935): Pindari carmina cum fragmentis, recognovit brevique adnotatione critica instruxit ®, Oxonii
® (1963): ëTwo Lines of Eumelusí, CQ NS 13, 145-53 = On Greek Margins, Oxford 1970, 46-58
® (1964): Pindar, Oxford
Braswell, B.K. (1988): A Commentary on the Fourth Pythian Ode of Pindar (Texte und Kommentare 14), Berlin/New York
Bravi, L. (1999): ëLíombra di Simonide in Aristofane, Cavalieri 402-406í, SemRom 2, 235-41
® (2002): ëIl canto della seconda parabasi dei Cavalieri di Aristofane: manoscritti e scol` metrici (Equ. 1264-1273=1290-1299)í, QUCC NS 70, 41-6
Brelich, A. (1959): ëI figli di Medeiaí, SMSR 30, 213-54
® (1969): Paides e parthenoi, Roma
Bremer, D. (1976): Licht und Dunkel in der frBhgriechischen Dichtung. Interpretationed zur Vorgeschichte der Lichtmetaphysik, Bonn
Briand, M. (2009): ëLa danse et la philologie : Ø partir du mouvement strophique dans les scholies anciennes Ø Pindareí, in S. David et al. (edd.), Traduire les scholies de Pindare I: De la traduction au commentaire: probl@mes de m�thode (DHA suppl. 2), Besançon, 93-106
Broger, A. (1996): Das Epitheton bei Sappho und Alkaios. Eine sprachwissenschaftliche Untersuchung (Innsbrucker BeitrGge zur Sprachwissenschaft 88), Innsbruch
Brommer, F. (19733): Vasenlisten zur griechischen Heldensage, Marburg
® (1980): GTttersagen in Vasenlisten, Marburg
Brun, P. (1996): Les archipels �g�ens dans líantiquit� grecque (Ve - IIe si@cles av. notre @re), Paris
Buck, C.D. (1955): The Greek Dialects. Grammar. Selected Inscriptions. Glossary, Chicago
Bulle, K. (1869): Pindars dritter (und vierter) isthmischer Siegsgesang, Bremen
® (1871): ëPindars dritte isthmische und elfte pythische Odeí, JKPh 103, 585-96
332
Bulloch, A.W. (1985): Callimachus. The Fifth Hymn, edited with introduction and commentary by ®, Cambridge
Bundy, E. (19862): Studia Pindarica, Berkeley / Los Angeles
® (1972): ëThe ìquarrel between Kallimachos and Apolloniosî I: The epilogue of Kallimachosí ìHymn to Apolloîí, CSCA 5, 39-94
Burkert, W. (1985): Greek Religion (tr. J. Raffan), Oxford
Burnett, A. P. (1998): ëSpontaneity, Savaging, and Praise in Pindarís Sixth Paeaní, AJPh 119, pp. 493-520
Bury, J.B. (1890a): HbQkn[gP �c^Qbjgb ´W_WgQ^jRb]. The Nemean Odes of Pindar, edited with introductions and commentary by ®, London and New York
® (1890b): ëThe ìthird Isthmianî í, Hermathena 7, 276-80
® (1892): HbQkn[gP �c^Qbjgb ∏]p_bgQ^jRb]. The Isthmian Odes of Pindar, edited with introduction and commentary by ®, London and New York
Cairns, D.L. (2010): Bacchylides. Five Epinician Odes (3, 5, 9, 11, 13), text, introductory essays, and interpretative commentary by ®, translations by ® and J.G.Howie, Cambridge
Calame, C. (1974): ëReflexions sur les genres littCraires en Gryce archaíqueí, QUCC 17, 113-28
® (1977): Les chúurs de jeunes filles en Gr@ce archaVque, I: Morphologie, fonction r�ligieuse et sociale (Engl. tr. by D. Collins and J. Orion, Choruses of Young Women in Ancient Greece. Their Morphology, Religious Role and Social Function, Lanham MD 20012), II: Alcman, Roma
® (1983): Alcman. Fragmenta edidit, veterum testimonia collegit ®, Romae
Calder, W.M., III (1977): ëPindar, Paean 6.102 (=139)í, AJPh 98, 350
Calvani, G. (1996): ëModi e fini delle parafrasi negli scholia vetera a Pindaroí, SCO 46, 269-329
CannatØ Fera, M. (1995): ëPindaro e la pietra del paragoneí, in L. Belloni, G. Milanese and A. Porro (eds.), Studia classica Iohanni Tarditi oblata, Milano, I 413-22
® (2004): ëPoesia e statuaria: gli eroi argivi di Pindaro e di Antifaneí, in Bernardini (ed.) (2004), 95-106
® and DíAlessio, G.B. (eds.) (2001): I lirici greci. Forme della comunicazione e storia del testo. Atti dellíincontro di studi, Messina 5-6 novembre 1999, Messina
® and Grandolini, S. (eds.) (2000): Poesia e religione in Grecia. Studi in onore di G. Aurelio Privitera, Napoli
Caprioli, M. (2007): ëConsiderazioni sul prosodo a Delo di Eumelo di Corintoí, ARF 9, 19-38
Cardin, M. (2010): ëLíehoia di Egina e le Asopidi nel Catalogo delle donne esiodeoí, in E. Cingano (ed.), Tra panellenismo e tradizioni locali. Generi letterari e storiografia, Alessandria 2010, 151-210
Carey, C. (1991): ëAn Edition of Pindarís Paeansí, review of Bona (1988), CR NS 41, 14f.
Carnes, J.S. (1995): ëWhy should I mention Aiakos? Myth and politics in Pindarís Nemean 8í part I, QUCC NS 51, 7-48
Cartledge, P. (1979): Sparta and Lakonia. A Regional History 1300-362 BC, London / Boston / Henley
333
Cassio, A.C. (1972): ëPind. Pae. 4 (fr. 52d Sn.), 21-24í in ASNP 2, 469-71
® (2005): ëI dialetti eolici e la lingua della lirica coraleí, in F. Bertolini and F. Gasti (eds.), Dialetti e lingue letterarie nella Grecia arcaica, atti della IV Giornata ghisleriana di Filologia Classica (Pavia, 1-2 aprile 2004), Pavia, 13-44
Cassola, F. (1964): ë {VW�pW[g] - ereuteroí, in A. Guarino and L. Labruna (eds.), Synteleia Vincenzo Arangio-Ruiz, Napoli, I 269-79
Cavallo, G. (1967): Ricerche sulla maiuscola biblica, Firenze
® (1974): ëLo stile di scrittura ëepsilon-thetaí nei papiri letterari: dallíEgitto ad Ercolanoí, CErc 4, 33-6 = Il calamo e il papiro. La scrittura greca dallíetI ellenistica ai primi secoli di Bisanzio (Papyrologica Florentina 36), Firenze 2005, 123-8
® (1989): ëCodice e storia dei testi greci antichi. Qualche riflessione sulla fase primitiva del fenomenoí, in Blanchard (ed.) (1989), 169-80
® (2008): La scrittura greca e latina dei papiri. Una introduzione, Pisa/Roma
Cazzato, V., Obbink, D., and Prodi, E.E. (forthcoming): The Cup of Song. Studies on Poetry and the Symposion, Oxford 2015
Cellerini, A. (1988): Introduzione allí Etymologicum Gudianum (BollClass suppl. 6), Roma
Chaniotis, A. (1994): ëZum Frage der Spezialisierung im griechischen Theater des Hellenismus und der Kaiserzeit auf der Grundlage der neuen Prosopographie der dionysischen Technitení, Ktema 15, 89-108
® (2011): ëThe ithyphallic hymn for Demetrius Poliorcetes and Hellenistic religious mentalityí, in I.P. Panagiotis, A.S. Chankowski, and C.C. Lorber (eds.), More than Men, Less than Gods. Studies on Royal Cult and Imperial Worship, proceedings of the international colloquium organized by the Belgian School at Athens (November 1-2, 2007) (Studia Hellenistica 51), Leuven / Paris / Walpole MA, 157-95
Chankowski, V. (2001): ëAthynes, DClos et les Cyclades Ø líCpoque classique : un rCseau Cconomique ?í, REA 103, 83-102
® (2008): Ath@nes et D�los I lí�poque classique. Recherches sur líadministration du sanctuaire díApollon d�lien (BEFAR 331), Athynes
Chantraine, P. (19866-86): Grammaire hom�rique
Christ, W. von (1896): Pindari carmina prolegomenis et commentariis instructa edidit ®, Lipsiae
Ciakciak, E. (1837): Dizionario Armeno-Italiano composto dal P. ®, Venezia
Ciancaglini, C.A. (1999): ëLe valenze linguistiche del termine c[g]ak^R in rapporto a f¦Qg] e Ú[_gQ^Rí, SemRom 2, 117-43
Cingano, E. (2004): ëTradizioni epiche intorno ad Argo da Omero al VI sec. a.C.í, in Bernardini (ed.) (2004), 59-78
® (forthcoming): ëExploring sympotic settings: Stesichorus, Ibycus, Simonidesí, in Cazzato, Obbink, and Prodi (eds.) (forthcoming)
Cockle, H.M. (1984): ë3674. Plato, Laws 9. 854 c-dí, in The Oxyrhynchus Papyri part LII, 99f.
Cockle, W.E.H. (1987): Euripides. Hypsipyle, text and annotation based on a re-examination of the papyri by ®, Roma
334
Cole, T. (2003): ëPindarís two odes for Melissos of Thebesí, in Benedetti and Grandolini (eds.) (2003), I 241-52
Constantakopoulou, C. (2007): The Dance of the Islands. Insularity, Networks, the Athenian Empire, and the Aegean World, Oxford
Conybeare, F.C. (1895): Philo. About the Contemplative Life or the fourth book of the treatise Concerning Virtues, critically edited with a defence of its genuineness by ®, Oxford
Cook, A.B. (1914-40): Zeus. A Study in Ancient Religion, I-III/2, Cambridge
Coppola, A. (2008): Líeroe ritrovato. Il mito del corpo nella Grecia classica, Venezia
Cordano, F. (2004): ëLa musica e la politica, ovvero gli auloW ad Ateneí, in V. de Angelis (ed.), Sviluppi recenti nellíantichistica. Nuovi contributi, Milano, 309-25
Corsini, E. (1747): Dissertationes IV agonisticae quibus Olympiorum, Pythiorum, Nemeorum atque Isthmiorum tempus inquiritur ac demonstratur. Accedit hieronicarum catalogus, Florentiae
Crahay, R. (1956): La litt�rature oraculaire chez H�rodote, Paris
Croiset, A. (1880): La po�sie de Pindare et les lois du lyrisme grec, Paris
Crusius, O. (1888): ëStesichoros und die epodische Composition in der griechischen Lyrikí, in Commentationes philologae quibus Ottoni Ribbeckio praeceptori inlustri sexagensimum aetatis, magisterii Lipsiensis decimum annum exactum congratulantur discipuli Lipsienses, Lipsiae, 1-22
Cupaiuolo, G. (19922): Evanzio: De fabula. Introduzione, testo critico, traduzione e note di commento a cura di ®, Napoli
Currie, B. (2004): ëReperformance scenarios for Pindarís odesí, in Mackie (ed.) (2004), 49-69
® (2005): Pindar and the Cult of Heroes, Oxford
Dale, A.M. (1950): ëStasimon and hyporchemeí, Eranos 48, 14-20 = Collected Papers, Cambridge 1969, 34-40
DíAlessio, G.B. (1988): ëPindaro, Peana 1 (fr. 52a Sn.-M.), 9-10í, ASNP III 18,1441-6
® (1991): ëOsservazioni e paralipomeni ad una nuova edizione dei frammenti di Pindaroí, RFIC 119, 91-117
® (1992): ëDue note su P.Oxy. 841 (Pindaro, Peani)í, ZPE 92, 81-3
® (1994): ëThe Greek paeaní, review of KÅppel (1992), CQ NS 44, 62-5
® (1997): ëPindarís Prosodia and the classification of Pindaric papyrus fragmentsí, ZPE 118, 23-60
® (1999): ëSu di un papiro berlinese di lirica corale (P. Berol. inv. 13411 = Pind. fr. 52n S.M.), ZPE 126, 15-25
® (2000a): ëTra gli dyi ad Apollo, e tra gli uomini ad Echecrate: P. Louvre E 7734 + 7733 (Pind. fr. dub. 333 S.-M.)í in CannatØ Fera and Grandolini (eds.) (2000), 233-62
® (2000b): review of SchrNder (1999), BMCR 2000/01/24
® (2001): ëSulla struttura del libro dei Peani di Pindaroí, in CannatØ Fera and id. (eds.) (2001), 69-86
335
® (2004): ëArgo e líArgolide nei canti cultuali di Pindaroí, in Bernardini (ed.) (2004), 107-25
® (2005): ëIl primo Inno di Pindaroí, in S. Grandolini (ed.), Lirica e teatro in Grecia: il testo e la sua ricezione. Atti del II incontro di studi, Perugia 23-24 gennaio 2003, Napoli 113-49
® (2007): ëPer una ricostruzione del Primo Inno di Pindaro: la ìTeogoniaî tebana e la nascita di Apolloí, SemRom 10, 101-17 [=2009b]
® (2009a): ëDefining local identities in Greek lyric poetryí, in R. Hunter and I. Rutherford (eds.), Wandering Poets in Ancient Greek Culture. Travel, Locality and Pan-Hellenism, Cambridge, 136-67
® (2009b): ëRe-constructing Pindarís First Hymn: the Theban ìtheogonyî and the birth of Apolloí, in L. Athanassaki, R.P. Martin, and J.F. Miller (eds.) (2009): Apolline Politics and Poetics, Athens, 129-47 [=2007]
® (2012): ëThe lost Isthmian odes of Pindarí, in Ag`cs, Carey and Rawles (eds.) (2012): 28-57
® (forthcoming): ëBacchylides at banquetí, in Cazzato, Obbink, and Prodi (eds.) (forthcoming)
® and Ferrari, F. (1988): ëPindaro, Peana 6, 175-183: una ricostruzioneí, SCO 28, 159-80
Daly, L.W. (1967): Contributions to a History of Alphabetization in Antiquity and the Middle Ages (Coll. Latomus 90), Buxelles
Daub, A. (1880): ëDe Suidae biographicorum origine et fideí, JKPh Suppl. n.F. 11, 401-90
Daumas, F. and Miquel, P. (1963): Les úuvres de Philon díAlexandrie, XXIX: De vita contemplativa, Paris
Daumas, M. (1992): ëArgos et les Septí, in PiCrart (ed.) (1992), 253-63
Dawes, R. (1745): Miscellanea critica in sectiones quinque dispertita scripsit ®, Cantabrigiae
Debiasi, A. (2004): Líepica perduta. Eumelo, il Ciclo, líoccidente (HesperWa 20), Roma
® (2005): ëEumeli Corinthii fragmenta neglecta?í, ZPE 153, 43-58
de Jong, I.J.F. (1987): ëThe voice of anonymity: tis-speeches in the Iliadí, Eranos 85, 69-84
Del Corso, L. (2002): file on PSI II 147 in Papiri letterari della Biblioteca Medicea Laurenziana (CD-ROM), Cassino
Denniston, J.D. (1929): ëEpexegetic SWí, CR 43, 59f.
Depew, M. (2000): ëEnacted and represented dedications: genre and Greek hymní, in ead. and Obbink (eds.), Matrices of Genre. Authors, Canons, and Society, Cambridge MA 2000, 59-79
Des Places, B. (1947): Le pronom chez Pindare. Recherches philologiques et critiques, Paris
Diehl, E. (19081) (19173): Supplementum lyricum: neue BruchstBcke von Archilochus, Alcaeus, Sappho, Corinna, Pindar ausgewÅhlt und erklÅrt von ®, Bonn
Di Marco, M. (1973-4): ëOsservazioni sullíiporchemaí, Helikon 13-14, 326-48
Dindorf, G. (1835): �[b]fgdnQ\]. Aristophanis comúdiX, accedunt Perditarum fabularum fragmenta, ex recensione ®, I-IV, Oxonii
Dindorf, L. (1845): HRP]RQ^gP �VVnkg] cW[bqS\]b]. Pausaniae descriptio GrXciX. Recognovit et prbfatus est ®. Grbce et Latine cum indice locupletissimo, Parisiis
336
Dissen, L.G. (1830): Pindari carmina quae supersunt cum deperditorum fragmentis selectis ex recensione Boeckhii, commentario perpetuo illustravit ®, Gothae et Erfordiae
Donaldson, J.W. (1841): HbQkn[gP fÏ cT�¦_WQR. Pindarís Epinician or Triumphal Odes, in Four Books, together with the Fragments of his Lost Compositions, revised and explained, by ®, Cambridge
Dornseiff, F. (1921): Pindars Stil, Berlin
Dover, K.J. (1968): Aristophanes. Clouds, edited with introduction and commentary by ®, Oxford
Ducat, J. (1971): Les kouroi du Ptoion. Le sanctuaire díApollon Ptoieus a lí�poque archaVque (BEFAR 219), Paris
Duchemin, J. (1955): Pindare po@te et proph@te, Paris
Dunbar, N. (1995): Aristophanes. Birds, Oxford
Dyer, R.R. (1964): ëOn describing some Homeric glossesí, Glotta 42, 121-31
Egenolff, P. (1887): ëZur geschichte der Oberlieferung griechischer metrikerí, JCPh 33 (135/1), 389-408
Erskine, A. (2001): Troy between Greece and Rome. Local Tradition and Imperial Power, Oxford
Fabbro, E. (1986): ëConsiderazioni sul peana simposialeí, QFC 5, 25-35 = in K. Fabian, E. Pellizer and G. Tedeschi (eds.), a∞Q\[Ï fW�Z\. Studi triestini di poesia simposiale, Alessandria 1991, 73-83
® (1992): ëSul riuso di carmi díautore nei simposi attici (Carm. conv. 8 P. e Alc. fr. 249 V.)í, QUCC 41, 29-38
Fantuzzi, M. (2010): ëSung poetry: the case of inscribed paeansí, in J.J. Clauss and M. Cuypers (eds.), A Companion to Hellenistic Literature, Malden MA, 181-96
FÅrber, H. (1936): Die Lyrik in der Kunsttheorie der Antike, I-II, MOnchen
Farnell, L.R. (1896-1909): The Cults of the Greek States, I-V, Oxford
® (1930-2): The Works of Pindar translated with literary and critical commentaries by ®, I: Translation in rhythmical prose with literary comments, II: Critical commentary, III: Text, London
Faulkner, A. (2008): The Homeric Hymn to Aphrodite: Introduction, Text and Commentary, Oxford
Faure, P. (1985): ëLes Dioscures Ø Delphesí, AC 54, 56-65
Fearn, D. (2003): ëMapping Phleious: Politics and myth-making in Bacchylides 9í, CQ NS 53, 347-67
® (2007): Bacchylides: Politics, Performance, Poetic Tradition, Oxford
® (2011a): ëAeginetan epinician culture: Naming, ritual, and politicsí, in id. (ed.) (2011), 175-226
® (2011b): ëThe Ceians and their choral lyric: Athenian, epichoric and pan-Hellenic perspectivesí, in Athanassaki and Bowie (eds.) (2011), 207-34
® (ed.) (2011): Aegina: Contexts for Choral Lyric Poetry. Myth, History and Identity in the Fifth Century BC, Oxford
337
Fehr, K. (1936): Die Mythen bei Pindar, ZOrich
Fennell, C.A.M. (18992): Pindar: Nemean and Isthmian Odes, with notes explanatory and critical, introductions, and introductory essays, by ®, Cambridge
Ferrari, F. (1990): ëRestauri testuali a P. Yale 18 (Pindaro, fr. 140a e fr. 140b Maehler)í, Maia 44, 229-34
® (1991a): ëContributi al testo dei Ditirambi di Pindaroí, SIFC III 9, 3-8.
® (1991b): ëTre papiri pindarici. In margine ai frr. 52n (a), 94a, 94b, 169a Maehlerí, RFIC 119, 385-407
® (1992a): ëLa sigla �\/� nei papiri pindaricií, SCO 42, 273-6
® (1992b): ëPer il testo dei Peani di Pindaroí, MD 28, 143-52
® (1992c): ëPindaro, fr. 215 (a), 4í, Maia 44, 227-31
® (1996): ëMinima Pindaricaí, RFIC 124, 129-36
® (1999): ëSul prosodio pindarico per gli Egineti (= ëPeanaí 6, 123-183)í, SCO 47, 149-58
® (2000): ëMinima Pindarica IIí, in M. CannatØ Fera and S. Grandolini (eds.) (2000), 311-20
® (2006): ëMolto rumore per nulla? Líinscriptio R allíOlimpica 5 e líautenticitØ dellíodeí, Eikasmos 17, 75-8
Figueira, T.J. (1981): Aegina. Society and Politics, Salem NH
® (1991): Athens and Aigina in the Age of Imperial Colonization, Baltimore and London
® (1993): Excursions in Epichoric History. Aeginetan Essays, Lanham MD
® (1999): ëThe evolution of Messenian identityí, in S. Hodkinson and A. Powell (eds.), Sparta. New Perspectives, London
Finglass, P.J. (2007a): Sophocles: Electra, edited with introduction and commentary by ®, Cambridge
® (2007b): Pindar: Pythian Eleven, edited with introduction, translation, and commentary by ®, Cambridge
Finley, J.H. (1951): ëThe date of Paean 6 and Nemean 7í, HSPh 60, 61-80
Flaceliyre, R. (1928): ëRemarques sur les SctCria de Delphesí, BCH 52, 256-91
Fogelmark, S. (1972): Studies in Pindar, with Particular Reference to Paean VI and Nemean VII, Lund
Fontenrose, J. (1978): The Delphic Oracle. Its Responses and Operations. With a Catalogue of Responses, Berkeley/Los Angeles/London
Ford, A. (2002): The Origins of Criticism. Literary Culture and Poetic Theory in Classical Greece, Princeton / Oxford
® (2011): Aristotle as Poet. The Song for Hermias and its Contexts, Oxford
Forrest, W.G. (1969): ëTwo chronographic notesí, CQ NS 19, 95-110
Forssman, B. (1966): Untersuchungen zur Sprache Pindars, Wiesbaden
Foss, C. (1976): Byzantine and Turkish Sardis, Cambridge MA/London
338
Fotheringham, J.K. (1907): ëOn the ìList of Thalassocraciesî in Eusebiusí, JHS 27, 75-89
Foucart, P. (1883): De collegiis scenicorum artificum apud graecos, Lutetiae Parisiorum
Fougyres, G. (1889): ëFouilles de DClosí, BCH 13, 229-52
Fowler, R.L. (1993): ëP.Oxy. XXVI 2442 fr. 29: A citation of Euphorion?í, ZPE 96, 15f.
Fraccaroli, G. [1913]: Pindaro. Le odi e i frammenti, traduzione con prolegomeni e commento di ®, I-II, Milano
Fraenkel, E. (1918): ëLyrische Daktylení, RhM n.F. 72, 161-97, 321-52 = Kleine BeitrGge zur Klassischen Philologie, I: Zur Sprache. Zur Griechischen Litteratur, Roma 1964, 165-233
® (1950): Aeschylus. Agamemnon, edited with a commentary by ®, I-III, Oxford
® (1962): Beobachtungen zu Aristophanes, Roma
Franco, C. (1988): ëLa competenza del destinatario nella parodia tragica aristofaneaí, in E. Corsini (ed.), La polis e il suo teatro /2, Padova, 213-32
Frazer, J.G. (1921): Apollodorus. The Library, with an English translation by Sir ®, I-II, London / New York
Frei, I. (1900): De certaminibus thymelicis, Basileae
FriedlÅnder, P. (1914): ëDas ProNmium der Theogonieí, Hermes 49, 1-16
Friis Johansen, H. and Whittle, E.W. (1980): Aeschylus. The Suppliants, I-III, Kdbenhavn
FOhrer, R. (1967): Formproblem-Untersuchungen zu den Reden in der frBhgriechischen Lyrik, MOnchen
Furley, W.D. (2011): ëHomeric and un-Homeric hexameter hymns. A question of typeí, in A. Faulkner (ed.), The Homeric Hymns. Interpretative Essays, Oxford, 206-31
® and Bremer, J.M. (2001): Greek Hymns. Selected Cultic Songs from the Archaic to the Hellenistic Period, I: The Texts in Translation, II: Greek Texts and Commentary, TObingen
Funghi, M.S. and Martinelli, M.C. (1996-97): ëP.Gen. inv. 98: Eschilo?í in APapyrol 8-9, 7-17
® and Messeri Savorelli, G. (1991): ëConsiderazioni paleografiche e papirologiche su P. Oxy. 2438aí, SCO 41, 101-4
® and ® (1992a): ëLo ëscriba di Pindaroí e le biblioteche di Ossirincoí, SCO 42, 43-62
® and ® (1992b): ëNote papirologiche e paleografiche I: lo ëscriba quintoí di Ossirinco e gli scavi al kcm Ali El Gammaní, Tyche 7, 75-88
Galiano, M.F. (1950-51): ëVaria Graecaí, Humanitas 3, 301-22
Gallavotti, C. (1942): ëStudi sulla lirica greca. 5. Nuovi carmi di Alceo da Ossirincoí, RFIC 70, 161-81
® (1978): ëLe copie di Pausania e gli originali di alcune iscrizioni di Olimpiaí, BPEC II 26, 3-27 (followed by collectively-authored ëPostille alle iscrizioni del quinto libro di Pausaniaí, 28-38)
Galletta, G. (1988): ëEpicismi in Pindaroí, in S. Costanza (ed.), Poesia epica greca e latina, Soveria Mannelli, 77-111
339
Gallo, I. (1968): Una nuova biografia di Pindaro (POxy. 2438). Introduzione, testo critico e commentario, Salerno; pp. 14-45 = (1997), 41-69
® (1969): ëGli ]j¦VbR di Pindaro. Nota critica al catalogo pindarico del Papiro di Ossirinco XXVI 2438í, QUCC 8, 105-15 = (1997), 71-8
® (1977): ëEustazio commentatore di Pindaroí, QUCC 24, 43-51 = (1997), 137-46
® (1997): Studi sulla biografia greca, Napoli
® (2005): La biografia greca. Profilo storico e breve antologia di testi, Soveria Mannelli
Gantz, T. (1993): Early Greek Myth. A Guide to Literary and Artistic Sources, Baltimore and London
Garner, R. (1990): From Homer to Tragedy. The Art of Allusion in Greek Poetry, London / New York
Gaspar, C. (1900): Essai de chronologie pindarique, Bruxelles
Gelzer, T. (1982-84): ëDie Alexandriner und der griechischen Lyriker. Reflexionen zur Rezeption und Interpretationení, AAntHung 30, 129-47
Gentili, B. (1979): ëTrittico pindaricoí, QUCC NS 2, 7-33
® (1981): ëLíeffigie bifronte di Neottolemo nel sesto Peana e nella settima Nemea di Pindaroí, in Letterature comparate: problemi e metodo. Studi in memoria di E. Paratore, Bologna, 103-9
® and Lomiento, L. (2003): Metrica e ritmica. Storia delle forme poetiche nella Grecia antica, Milano (Engl. tr. by E. Kopff, Metrics and Rhythmics. History of Poetic Forms in Ancient Greece, Pisa 2008)
® et al. (20064): Pindaro. Le Pitiche, a cura di ®, P. Angeli Bernardini, E. Cingano e P. Giannini, Milano
® et al. (2013): Pindaro. Le Olimpiche, a cura di ®, C. Catenacci, P. Giannini e L. Lomiento, Milano
Gerber, D.E. (1982): Pindarís Olympian One: A Commentary (Phoenix suppl. 15), Toronto
® (2002): A Commentary on Pindar Olympian Nine (Hermes Einzelschriften 87), Stuttgart
Giannisi, P. (2006): R�cits des voies. Chant et cheminement en Gr@ce archaVque, Grenoble
Gianotti, G.F. (1975): Per una poetica pindarica, Torino
Gladigow, B. (1965): Sophia und Kosmos. Untersuchungen zur FrBhgeschichte von ]gd¦] und ]gd^R (Spudasmata 1), Hildesheim
Gomme, A.W. and Sandbach, F.H. (1973): Menander. A Commentary, Oxford
Gonis, N. (2005): ëA. S. Hunt and ëYoutieís Lawíí, ZPE 151, 166
Gostoli, A. (2009): ëLíinno apopemptico in Menandro Retore e nella prassi poetica greca (Bacchilide e Callimaco)í, Paideia 64, 589-96
Graf, F. (1979): ëApollon Delphiniosí, MH 36, 2-22
Grandolini, S. (1986): ëIn margine ad Aristoph. «Av.» 851-57í, GIF 38, 263-6
® (1987-8): ëOsservazioni sul prosodioí, AFLPer 25, 27-52
340
® (1991): ëCanto processionale e culto nellíantica Greciaí, AION(filol) 13, 125-40
® (1999): ëDidimo e la classificazione della poesia liricaí, GIF 51, 1-22
Grenfell, B.P. and Hunt, A.S. (1898): ëXXVI. Demosthenes, H[gg^_bR k\_\Sg[bjn, 26-29í, in The Oxyrhynchus Papyri part I, 53-6
® (1899): ëCCXXII. List of Olympian victorsí, in The Oxyrhynchus Papyri part II, 85-95
® (1903): ë408. Odes of Pindarí, in The Oxyrhynchus Papyri part III, 13-17
® (1906): ëExcavations at Oxyrhynchusí in Egypt Exploration Fund. Archaeological Report 1905-1906, 8-16 = ëExcavations at Oxyrhynchus (1896-1907)í, in A.K. Bowman et al. (edd.) (2007): 345-68: 359-65
® (1908a): ë841. Pindar, Paeansí, in The Oxyrhynchus Papyri part V, 11-110
® (1908b): ë984í, in The Oxyrhynchus Papyri part VI, 321f.
® (1915): ë1361. Bacchylides, Scoliaí, in The Oxyrhynchus Papyri part XI, 65-83
® (1919): ë1604. Pindar, Dithyrambsí, in The Oxyrhynchus Papyri part XIII, 27-45
van Groningen, B.A. (1960): Pindare au banquet. Les fragments des scolies CditCs avec un commentaire critique et explicatif par ®, Leyde
® (1963): review of The Oxyrhynchus Papyri part XXVI, Gnomon 35, 127-30
® (1966): Th�ognis. Le premier livre, CditC avec un commentaire, Amsterdam
Guillon, P. (1943): Les tr�pieds du Ptoion, I-II (BEFAR 153), Paris
Guiraud, C. (1962): La phrase nominale en grec díHom@re I Euripide (Ltudes et commentaires 42), Paris
Gundert, H. (1935): Pindar und sein Dichterberuf, Frankfurt am Main
Gutzwiller, K.J. (1998): Poetic Garlands. Hellenistic Epigrams in Context, Berkeley/London
van Haelst, J. (1989): ëLes origins du codexí, in Blanchard (ed.) (1989), 13-25
Haldane, J.A. (1966): ëMusical instruments in Greek worshipí, G&R NS 13, 98-107
Hamilton, R. (1976): Epinikion. General Form in the Odes of Pindar, The Hague
Hardie, A. (2007): ëJuno, Hercules, and the Muses at Romeí, AJPh 128, 551-92
Harrauer, C. (1988): ëLikymnios und seine Familie in P.Vindob. G 23058í, WS 101, 97-126
Hartung, J.A. (1855-6): Pindarís Werke. Griechisch mit metrischer Uebersetzung und prBfenden und erklGrenden Anmerkungen von ®, I-IV, Leipzig
Harvey, A.E. (1955): ëThe classification of Greek lyric poetryí, CQ NS 5, 157-75
Haspels, C.H.E. (1930): ëDeux fragments díune coupe díEuphroniosí, BCH 54, 422-51 and pll. XX-XXIV
Helm, R. (1926): ëDie Liste der Thalassokratien in der Chronik des Eusebiusí, Hermes 61, 241-62
Hendriks, I.H.M. (1982): De interpolitieke en internationale betrekkingen van Argos in de vijfde eeuw v.Chr. gezien tegen de achtergrond van de intra-politieke ontwikkelingen, Groningen
Henry, W.B. (2003a): ëContracted biceps in Pindarí, ZPE 143, 11-16
341
® (2003b): ëNotes on Pindarís Paeans (P.Oxy. 841)í, ZPE 145, 7-10
Hercher, R. (1864): Artemidori Daldiani Onirocriticon libri V ex recensione ®, Lipsiae
Herda, A. (2006): Milesische Forschungen IV: Der Apollon-Delphinios-Kult in Milet und die Neujahrsprozession nach Didyma. Ein neuer Kommentar der sog. Molpoi-Satzung von ®, Mainz am Rhein
Hermann, G. (1796): De metris poetarum graecorum et romanorum libri III, Lipsiae
® (1809): De dialecto Pindari observationes, Lipsiae = Pindari carmina cum lectionis varietate et adnotationibus iterum curavit C.G. Heyne, III/1: Carminum Pindaricorum fragmenta et Godofredi Hermanni dissertationes Pindaricae, nova editio correctior et ex schedis Heynianis aucta, Lipsiae 1817, 250-75 = Godofredi Hermanni opuscula I, Lipsiae 1827, 245-68
Heyne, C.G. (1798-99): Pindari Carmina cum lectionis varietate et annotationibus iterum curavit ®, I-III/2, Gottingae
® (1802): Homeri carmina cum brevi annotatione, accedunt variae lectiones et observationes veterum grammaticorum cum nostrae aetatis critica curante ®, I-IX, Lipsiae / Londinii
Hill, B.H. and Meritt, B.D. (1944): ëAn early decree concerning tributeí, Hesperia 13, 1-15
Hiller, E. (1886): ëDie antiken Verzeichnisse der pindarischen Dichtungení, Hermes 21, 357-71
Hillgruber, M. (1990): ëZur Zeitbestimmung der Chrestomathie des Proklosí, RhM n.F. 133, 397-404
Hitzig, H. (1896-1910): Pausaniae Graeciae descriptio. Edidit, Graeca emendavit, apparatum criticum adiecit ®, commentarium Germanice scriptum cum tabulis topographicis et numismaticis addiderunt ® et H. Bluemner, I/1-III/2, Lipsiae
Hoekstra, A. (1962): ëThe absence of the Aeginetansí, Mnemosyne IV 15, 1-14
Hoerschelmann, W. (1881): ëUntersuchungen zur Geschichte der griechischen Metriker. Die Composition der Hephaestio-Scholiení, RhM n.F. 36, 260-301
® (1888): Ein griechisches Lehrbuch des Metrik. Literarhistorische Studien, Dorpat
Hoffelner, K. (1994): ëDas Thearion von >gina. Der Thearion-Bauí, AA 1994, 138-48
Hollis, A.S. (1998): ëSome neglected verse citations in Hesychiusí, ZPE 123: 61-71
Homolle, T. (1886): ëNote sur la chronologie des archontes athCniens de la seconde moitiC du IIme siycle av. J.-C.í, BCH 10, 6-68
Hope, N. (1986): A Commentary on the Fourth and Ninth Paeans of Pindar, M.Litt. Oxford
Hornblower, S. (2004): Thucydides and Pindar. Historical Narrative and the World of Epinikian Poetry, Oxford
® (2007): ëëDolphins in the seaí (Isthmian 9.7): Pindar and the Aeginetansí, in id. and Morgan (eds.) (2007), 287-308
® and Morgan, C. (eds.) (2007): Pindarís Poetry, Patrons, and Festivals from Archaic Greece to the Roman Empire, Oxford
Housman, A.E.H. (1898): ëOde XVIIí, in R.C. Jebb, F.G. Kenyon, A. Platt, H. Richards, ®, and J.E. Harrison, ëNotes on Bacchylidesí, CQ 12, 123-41: 134-40
Hubbard, T.K. (1987): ëTwo notes on the myth of Aiakos in Pindarí, GRBS 28, 5-22
342
® (2001): ëPindar and Athens after the Persian Warsí, in PapenfuÜ and Strocka (eds.) (2001), 387-97
® (2004): ëThe dissemination of epinician lyric: Pan-Hellenism, reperformance, written textsí, in Mackie (ed.) (2004), 71-93
® (2011): ëThe dissemination of Pindarís non-epinician choral lyricí, in Athanassaki and Bowie (eds.) (2011), 347-63
Hummel, P. (1993): La syntaxe de Pindare (BIG 24), Louvain
® (1997): ëConnivence binaire et crCativitC sCrielle : Les composCs Ø premier terme superlatif chez Pindare et Bacchylideí, QUCC NS 56, 61-6
® (1999): Lí�pith@te pindarique. Ltude historique et philologique, Bern
Hunt, A.S. (1922a): ë1787. Sappho, Book IVí, in The Oxyrhynchus papyri part XV, 26-46
® (1922b): ë1792. Pindar, Paean?í, ibid. 86-98
Hunter, R. and Fuhrer, T. (2001): ëImaginary gods? Poetic theology in the Hymns of Callimachusí, in F. Montanari and L. Lehnus (eds.), Callimaque (Entretiens Hardt 48), Vandúuvres / Genyve, 143-75
Hutchinson, G.O. (1985): Aeschylus. Seven against Thebes, edited with introduction and commentary by ®, Oxford
® (2001): Greek Lyric Poetry. A Commentary on Selected Larger Pieces, Oxford
® (2008): Talking Books. Readings in Hellenistic and Roman Books of Poetry, Oxford
Huxley, G.L. (1982): ëProblems in the Chronography of Eusebiusí, PRIA 82C, 183-96
Iannucci, A. (2011): ëStrumenti musicali tra generi letterari e performance poetica. Líopposizione tra aulos e barbiton in Crizia (1 D.-K. = 8 Gent.-Pr.), Anacreonte e Teleste (806 PMG)í, AOFL 6, 75-95
Immisch, O. (1889): ëZur Geschichte der griechischen Lyrikí, RhM 44, 553-
Imperio, O. (2000): ëPindaro e la pietra del paragone: una metafora poetologicaí, Eikasmos 9, 59-70
® (2004): Parabasi di Aristofane. Acarnesi, Cavalieri, Vespe, Uccelli, Bari
Ioannidou, G. (1996a): ë14. P.21114 + P.11677í, in Catalogue of Greek and Latin Literary Papyri in Berlin (P. Berol. inv. 21101-21299, 21911) (BKT Heft IX), 23f.
® (1996b): ë143. P.21239í, ibid. 181
Irigoin, J. (1952): Histoire du texte de Pindare, Paris
® (1958): Les scholies m�triques de Pindare (Bibl. Bcole des Hautes Btudes 310), Paris
® (1994): ëLes Cditions de textesí, in F. Montanari (ed.), La philologie grecque I lí�poque hell�nistique et romaine (Entretiens Hardt 40), Vandúuvres / Genyve, 39-82
® (1998): ëLes Cditions de poytes Ø Alexandrieí, in Argoud and Guillaumin (eds.) (1998), 405-13
Jacob, C. (1998): ëLa bibliothyque, la carte et le traitC. Les formes de líaccumulation du savoir Ø Alexandrieí, in Argoud and Guillaumin (eds.) (1998), 19-37
343
Jameson, M.H. (1994): ëTheoxeniaí, in R. HÅgg (ed.) Ancient Greek Cult Practice from the Epigraphical Evidence. Proceedings of the Second International Seminar on Ancient Greek Cult, organized by the Swedish Institute at Athens, 22-24 November 1991 (AIARS 13), Stockholm 1994, 35-57
Jebb, R.C. (1905): Bacchylides. The Poems and Fragments edited with introduction, notes, and prose translation by Sir. ®, Cambridge
Jeffery, L.H. (1976): Archaic Greece. The City-States c. 700-500 B.C., London / Tonbridge
Johnson, W.A. (2004): Bookrolls and Scribes in Oxyrhynchus, Toronto
® (2009): ëThe ancient bookí, in R.S. Bagnall (ed.), The Oxford Handbook of Papyrology, Oxford
Jory, E.J. (1967): ëA cR~] jT_Tbk¦] and the kbÏ cnQfTQ. Some problems of festival competitionsí, BICS 14, 84-90
Kaibel, G. (1898): Die Prolegomena HW[Ú jT_Tbk^R] (AGG n.F. 2/4), Berlin
Kaimio, M. (1977): Characterization of Sound in Early Greek Literature (Commentationes Humanarum Litterarum 53), Helsinki / Helsingfors
Kambylis, A. (1966): ëBemerkungen zu Pind. Fr. 215(a), 4 (Snell)í, Hermes 94, 238-43
® (1991): Eustathios Bber Pindars Epinikiendichtung. Ein Kapitel der klassischen Philologie in Byzanz, Hamburg/GNttingen
Kaempf-Dimitriadou, S. (1979): Die Liebe der GTtter in der attischen Kunst des 5. Jahrhunderts v. Chr. (Beih. zur Halbjahresschrift AK 11), Bern
KÅppel, L. (1992): Paian. Studien zur Geschichte einer Gattung, Berlin
® and Kannicht, R. (1988): ëNoch einmal zur Frage ëDithyrambos oder Paianí im Bakchylideskommentar P.Oxy. 23. 2368í, ZPE 73, 19-24
Kaster, R.A. (1988): Guardians of Language: The Grammarian and Society in Late Antiquity, Berkeley / Los Angeles / London
Kennedy, G.A. (2003): Progymnasmata. Greek Textbooks of Prose Composition and Rhetoric, translated with introductions and notes by ®, Leiden / Boston
Kenyon, F.G. (1897): The Poems of Bacchylides from a Papyrus in the British Museum, edited by ®, Oxford
KeyÜner, K. (1932): Gottesvorstellung und Lebensauffassung im griechischen Hymnus, Stuttgart
Kiechle, F. (1960): ëArgos und Tiryns nach der Schlacht bei Sepeiaí, Philologus 104, 181-200
Kirk, G.S. (1990): The Iliad. A Commentary, general editor ®, II: books 5-8, Cambridge
Kirkwood, G. (1982): Selections from Pindar, Chico CA
Kleanthis, K. (1886-7): HbQkn[gP fÏ sT�¦_WQR _WfÏ _WfRd[nsWTQ, s\_WbisWTQ jRÚ c^QRjge f�Q VWXYTQ W∞e f¦_gPe W� ÕcÙ ®, I-V, �Q ÁW[SYsff
Knoblauch, P. (1972): ëDie Hafenanlagen der Stadt Aeginaí, AD 27, 50-85
Knoepfler, D. (1993): ëAdolf Wilhelm et la pent�t@ris des Amphiaraia díOropos. RCexamen de A.P. LIV 7 Ø la lumiyre du catalogue IG VII 414 + SEG I 126í, in M. PiCrart (ed.), Aristote et Ath@nes. Aristoteles and Athens, Fribourg (Suisse) 23-25 mai 1991, Paris, 279-302
344
KNhnken, A. (1971): Die Funktion des Mythos bei Pindar. Interpretationen zu sechs Pindargedichten, Berlin
® (1983): ëMythical chronology and thematic coherence in Pindarís third Olympian odeí, HSPh 87, 49-63
® (1984): ëH\VgS¦QTQ �VRfq[. Kallimachos, Zeushymnos v.3í, Hermes 112, 438-45
Koster, W.J.W. (19532): Trait� de m�trique grecque, suivi díun pr�cis de m�trique latine, Leyde
Kougeas, S.B. (1913): ø æRbsR[W^Re �[YpRe jRÚ fÙ ¯[SgQ RÃfgt. |P_�gVU W∞e fUQ �sfg[^RQ fve c[if\e �QRSWQQqsWTe f�Q �VV\Qbj�Q S[R__nfTQ �Q lP�RQf^Tb, �Q �pqQRbe
Kowalzig, B. (2007): Singing for the Gods. Performances of Myth and Ritual in Archaic and Classical Greece, Oxford
Krevans, N. (2005): ëThe editorís toolbox: Strategies for selection and presentation in the Milan papyrusí, in K. Gutzwiller (ed.), The New Posidippus. A Hellenistic Poetry Book, 81-96
Kritzas, C.B. (1972): ë�[ZRb¦f\fWe jRb _Q\_W~R �[SgVbkgjg[bQp^Reí, AD 17, 192-219
® (1992); ëAspects de la vie politique et Cconomique díArgos au Ve siycle avant J.-C.í, in PiCrart (ed.) (1992), 231-40
Krummen, E. (1990): Pyrsos hymnon. Festliche Gegenwart und mythisch-rituelle Tradition als Voraussetzung einer Pinderinterpretation (Isthmie 4, Pythie 5, Olympie 1 und 3), Berlin / New York
Kugelmeier, C. (1996): Reflexe frBher und zeitgenTssischer Lyrik in den alten attischen KomTdie (Beitr. zur Altertumskunde 80), Stuttgart
Kurke, L. (2005): ëChoral lyric as ìritualizationî: Poetic sacrifice and poetic ego in Pindarís sixth Paianí, CA 24, 81-130
® (2007): ëVisualizing the choral: Epichoric poetry, ritual, and elite negotiation in fifth-century Thebesí, in C. Kraus, S. Goldhill, H.P. Foley and J. Elsner (edd.), Visualizing the Tragic. Drama, Myth, and Ritual in Greek Art and Literature. Essays in honour of Froma Zeitlin, Oxford, 63-102
KOster, L. (1705): CgP~kR]. Suidae lexicon, GrXce et Latine. Textum Grbcum cum manuscriptis codicibus collatum a quamplurimis mendis purgavit, notisque perpetuis illustravit, versionem Latinam gmilii Porti innumeris in locis correxit, indicesque auctorum & rerum adiecit ®, Cantabrigib
Lama, M. (1990): ëAspetti di tecnica libraria ad Ossirinco: copie letterarie su rotoli documentarií, Aegyptus 71, 55-120
Landucci Gattinoni, F. (1981): ëLa divinizzazione di Demetrio e la coscienza atenieseí, in M. Sordi (ed.), Religione e politica nel mondo antico, Milano, 115-23
LanCrys, N. (1996): ëPindare, Pythique IV. Les phrases nominales, contribution Ø une Ctude linguistiqueí, RPh 70, 255-80
® (1997): ëLes emplois de la phrase nominale chez Pindare, Pythique IXí, Lalies 17, 217-27
Larson, J. (1995): Greek Heroine Cults, Madison WI
Lavecchia, S. (2000): Pindari Dithyramborum fragmenta edidit ®, Romae / Pisis
Laxander, H. (2000): Individuum und Gemeinschaft im Fest. Untersuchungen zu attischen Darstellungen von Festgeschehen im 6. und frBhen 5. Jahrhundert v. Chr., MOnster
345
Lefkowitz, M.R. (1963): ëÁ� jRÚ �Si: The first person in Pindarí, HSCPh 67, 177-253 = Firs-Person Fictions. Pindarís Poetic ëIí, Oxford 1991, 1-71.
® (20122): The Lives of the Greek Poets, London
Le Guen, B. (2001): Les associations de technites dionysiaques I lí�poque hell�nistique, I-II, Nancy
Lehnus, L. (1973a): ëUna glossa pindarica in Corinna di Tanagraí, RIL 107: 393-422
® (1973b): ëContributo a due frammenti pindarici (frr. 37 e 168 Snell3)í, SCO 22, 5-18
® (1979): Líinno a Pan di Pindaro, Milano
® (1984): ëPindaro: il Dafneforico per Agasicle (Fr. 94b Sn.-M.)í, BICS 31, 61-92
Lejeune, M. (1939): Les adverbes grecs en -pWQ, Bordeaux
Leo, F. (1901): Die griechisch-rTmische Biographie nach ihrer litterarischen Form, Leipzig
Leone, A. (1976): ëLa «onciale ogivale» dalle origini alla decadenzaí, StudPap 15, 141-9
Leumann, M. (1950): Homerische WTrter, Basel
Liberman, G. (2007): ëLíCdition alexandrine de Sapphoí, in G. Bastianini and A. Casanova (eds.), I papiri di Saffo e di Alceo. Atti del convegno internazionale di studi, Firenze, 8-9 giugno 2006, Firenze 2007, 41-65
Lidov, J.B. (1974): ëThe poems and performance of Isthmians 3 and 4í, CSCA 7, 175-85
Livrea, E. (1981): ëP.Laur. II/35í, ZPE 42, 35f. and pl. 1a
Lobel, E. (1922): ëTwo fragments of papyrusí, BQR III/36, 289f.
® (1925): CRcdgt] _YV\. The Fragments of the Lyrical Poems of Sappho edited by ®, Oxford
® (1956a): ë2363. Bacchylides, Epiniciansí, in The Oxyrhynchus Papyri part XXIII, 28-30
® (1956b): ë2368. Commentary on Bacchylides?, Dithyrambs or Paeansí, ibid. 51-4
® (1957): ë2387. Alcman, HR[pYQbRí, in The Oxyrhynchus Papyri part XXIV, 8-22
® (1959a): ë2430. Choral lyric in the Doric dialect (?Simonides)í, in The Oxyrhynchus Papyri part XXV, 45-87
® (1959b): ë2431. Simonides, Epinicians?í, ibid. 87-91
® (1961a): ë2438. Life of Pindarí, in The Oxyrhynchus Papyri part XXVI, 1-7
® (1961b): ë1792. Pindar, Paeans; ?other booksí, ibid. 13-25
® (1961c): ë2441. Pindar, Uncertain Category (?c[g]¦kbR)í, ibid. 25-30
® (1961d): ë2442. Pindar, Ÿ_Qgb, HRb¬QW], ?other piecesí, ibid. 31-78
® (1961e): ë2445. Pindar, Dithyrambs (?) (and other categories?)í, ibid. 86-101
® (1961f): ë2448. Pindar, Uncertain categoryí, ibid. 131-9
® (1961g): ë2449. Commentary on Pindar?í, ibid. 140
® (1961h): ë2451. Commentary on Pindar, Isthmians, etc.í, ibid. 154-77
® (1967a): ë2622. Lyric versesí, in The Oxyrhynchus Papyri part XXXII, 63-5
346
® (1967b): ë2623. Choral lyricí, ibid. 66-90
® (1967c): ë2625. Choral lyricí, ibid. 114-9
® (1967d): ë2636. Commentary on Choral Lyricí, ibid. 133-7
® (1970): ëAddendum to 1792í, in The Oxyrhynchus Papyri part XXXVII, 104
® (1981): ë3. Simonidesí, in Papyri Greek & Egyptian edited by various hands in honour of E.G. Turner on the occasion of his seventieth birthday, London, 21-3
Lomiento, L. (1998): ëInterpretazione metrica di Pindaro, Ol. 14í, QUCC NS 60, 109-31
Longo, A. (1995): ëSullíattribuzione della Crestomazia a Proclo neoplatonicoí, SIFC III 13, 109-24
Lorimer, H.L. (1936): ëGold and ivory in Greek mythologyí, in Greek Poetry and Life. Essays presented to Gilbert Murray on his seventieth birthday January 2, 1936, Oxford, 14-33
Loscalzo, D. (2000a): ëConsiderazioni sulle riedizioni di un epinicioí, Prometheus 26, 1-18
® (2000b): La Nemea settima di Pindaro, Viterbo
® (2003): La parola inestinguibile. Studi sullíepinicio pindarico, Roma
Lowe, N.J. (2007): ëEpinikian eidographyí, in Hornblower and Morgan (eds.) (2007), 167-76
Ludwich, A. (1879): ëDie metrische Lebensskizze Pindarsí, RhM 34, 357-69
Luppe, W. (1987): ëDithyrambos oder Paian? - Zu Bakchylides Carm. 23 Sn./M.í, ZPE 69, 9-12
® (1989): ëNochmals zu Kallimachosí Gattungszuordnung bakchylideischer Liederí, APapyrol 1, 23-9
Luraghi, N. (2001): ëDer Erdbebenaufstand und die Entstehung der messenischen IdentitÅtí, in PapenfuÜ and Strocka (eds.) (2001), 279-301
® (2002a): ëBecoming Messenianí, JHS 122, 45-69
® (2002b): ëHelots called Messenians? A note on Thuc. 1.101.2í, CQ NS 588-92
® (2008): The Ancient Messenians. Constructions of Ethnicity and Memory, Cambridge
® (2009): ëMessenian ethnicity and the free Messeniansí, in P. Funke and ® (eds.), The Politics of Ethnicity and the Crisis of the Peloponnesian League, Washington DC, 110-34
Lyons, D. (1997): Gender and Immortality. Heroines in Ancient Greek Myth and Cult, Princeton
Maas, M. and Snyder, J. McIntosh (1980): Stringed Instruments of Ancient Greece, New Haven / London
Maas, P. (1913-21): Die neue Responsionsfreiheiten bei Bakchylides und Pindar, I: JPhV 39, 289-320, II: 47, 13-31 [cited from the offprints (Berlin 1914, 1921) with their self-standing pagination]
® (19233): Griechische Metrik, Teubner
MacDowell, M.D. (1971): Aristophanes. Wasps, edited with introduction and commentary by ®, Oxford
MacYa, L.M. (1976): ëCronologYa de dos poemas pindUricos: Nemea VII y PYtica XIí, ECl\s 20, 191-206
347
Mackie, C.J. (ed.) (2004): Oral Performance and its Contexts (Mnemosyne suppl. 248), Leiden / Boston
Maehler, H. (1968): ëWartetext 5: Ein Pindarfragmentí, ZPE 3, 97-9
® (19754): Pindari carmina cum fragmentis pars II: Fragmenta. Indices post B. Snell edidit ®, Leipzig
® (1982): Die Lieder des Bakchylides, I: Die Siegeslieder, I-II (Mnemosyne suppl. 62), Leiden
® (1989): ë3822. Pindar, Paeansí, in The Oxyrhynchus Papyri vol. LVI, 1-6
® (1997): Die Lieder des Bakchylides, II: Die Dithyramben und Fragmente, Text, xbersetzung und Kommentar von ® (Mnemosyne suppl. 167), Leiden / Boston
® (2004): Bacchylides: A Selection, edited by ®, Cambridge
® (2010a): ë5038. Pindar, Olympian Odes X 1-11, 24-35í in The Oxyrhynchus Papyri part LXXV, 66-8
® (2010b): ë5039. Pindar, Pythian Odes I 6-9, 32-5, III 101-3, IV 39-43, 58-66, 72-80, 134-7, 256-7, VI 9-16í, ibid. 68-72
® (2010c): ë5042. Pindar, Pythian Odes X 69-72, XI titleí, ibid. 76f.
® (2010d): ë5043. Pindar, Nemean Odesí, ibid. 77-87
Magnelli, E. (2002): Studi su Euforione, Roma
Maleci, S. (1995): Il codice Barberinianus Graecus 70 dellíEtymologicum Gudianum (Boll.Class. suppl. 15), Roma
Marasco, G. (1984): Democare di Leuconoe. Politica e cultura in Atene fra 4. e 3. sec. a.C., Firenze
Marchetti, P. and Kolokotsas, K. (1995): Le nymph�e de líagora díArgos. Fouille, �tude architecturale et historique par ® et ® avec une contribution de C. Abadie-Reynal (Ltudes P�loponnesiennes 11), Paris
Marrou, H.-I. (19656): Histoire de lí�ducation dans líantiquit�, Paris
Martin, R.P. (2003): ëThe pipes are brawling. Conceptualising musical performance in Athensí, in C. Dougherty and L. Kurke (eds.), The Cultures within Ancient Greek Culture. Contact, Conflict, Collaboration, Cambridge, 153-80
Massimilla, G. (1996): Callimaco. Aitia, libri primo e secondo. Introduzione, testo critico, traduzione e commento a cura di ®, Pisa
Mastronarde, D.J. (1994): Euripides. Phoenissae, edited with introduction and commentary by ®, Cambridge
Mazzucchi, C.M. (2000): ëUna curiosa legatura epsilon-iota nel codice ambrosiano C 222 inf.í, BBGG NS 54, 203-7
® (2003): ëAmbrosianus C 222 inf. (Graecus 886): il codice e il suo autore. Parte prima: il codiceí, Aevum 77, 263-75
® (2004): ëAmbrosianus C 222 inf. (Graecus 886): il codice e il suo autore. Parte seconda: líautoreí, Aevum 78, 411-40
McDevitt, A. (1994): ëHorses for courses. A note on Bacchylides 3.3-4í, Hermes 122, 502f.
348
McLennan, G.R. (1977): Callimachus. Hymn to Zeus, introduction and commentary by ®, Roma
McNamee, K. (1977): Marginalia and Commentaries in Greek Literary Papyri, Ph.D. Duke
® (1981a): Abbreviations in Greek Literary papyri and Ostraca (BASP suppl. 3), Atlanta
® (1981b): ëGreek literary papyri revised by two or more handsí, in Proceedings of the Sixteenth International Congress of Papyrology, New York, 24-31 July 1980, Chico 1981, 79-91
® (1992): Sigla and Select Marginalia in Greek Literary Papyri (Papyrologica Bruxellensia 26), Bruxelles
® (2007): Annotations in Greek and Latin Texts from Egypt (ASP 45)
Meiggs, R. (1972): The Athenian Empire, Oxford
Merkelbach, R. (1980): ëLex Youtieí, ZPE 38, 294
® (2003): ëIuxta lacunam ne mutaverisí, ZPE 142, 34
Metcalf, C. (2012): Aspects of Early Greek and Babylonian Hymnic Poetry, D.Phil. Oxford
Meyer, H. (1933): Hymnische Stilelemente in der frBhgriechischen Dichtung, WOrzburg
Mie, F. (1909): ëxber kbÏ cRQf�Q und ¡ �cbQ^jbg] in agonistischen Inschriftení, AM 34, 1-22
Miller, S.G. (1978): ëThe date of the first Pythiadí, CSCA 11, 127-58
Mikalson, J.D. (1998): Religion in Hellenistic Athens, Berkeley / Los Angeles / London
Mineur, W.H. (1984): Callimachus. Hymn to Delos. Introduction and commentary (Mnemosyne suppl. 83), Leiden
Mingarelli, S. (2003): Eracle a Cos: una lettura del fr. 33a Sn.-M. di Pindaroí, in R. Nicolai (ed.), ÓP]_¦]. Studi di poesia, metrica e musica greca offerti dagli allievi a Luigi Enrico Rossi per i suoi settantíanni, Roma, 117-132
Minton, W.W. (1970): ëThe proem-hymn of Hesiodís Theogonyí, TAPhA 101, 357-77
Moens, P.W. (1930): De twee Delphische hymnen met muzieknoten, Utrecht
Moggi, M. (1974): ëI sinecismi e le annessioni territoriali di Argo nel V secolo a.C.í, ASNP 4, 1249-63
Mommsen, C.I.T. (1866): Pindari carmina ad fidem optimorum codicum recensuit, integram scripturae diversitatem subiecit, annotationem criticam addidit ®, Berolini
Montanari, F. (1989) ëEvoluzioni del coro e movimenti celestií, in L. de Finis (ed.), Scena e spettacolo nellíantichitI. Atti del convegno internazionale di studio, Trento, 28-30 marzo 1988, Firenze, 149-63
Montevecchi, O. (1998): ëLa provenienza di P.Oxy. 984í, Aegyptus 78, 49-76
Moretti, J.-C. (1989): ëRapport sur les travaux de líBcole Française en Gryce en 1988. Argos 4. � Le thCitreí, BCH 113, 717-22
® (1998): ëLíimplantation du thCitre díArgos dans un lieu plein de sanctuairesí, in Pariente and Touchais (eds.) (1998), 233-48
Morgan, C. (1990): Athletes and Oracles. The Transformation of Olympia and Delphi in the Eighth Century BC, Cambridge
349
® (2007): ëDebating patronage: the cases of Argos and Corinthí, in Hornblower and Morgan (eds.) 2007, 213-63
Mosshammer, A.A. (1982): ëThe date of the first Pythiad � againí, GRBS 23, 15-30
Mueller, C. (1817): Aegineticorum liber scripsit ®, Berolini
Myres, J.L. (1906): ëOn the ìList of Thalassocraciesî in Eusebiusí, JHS 26, 84-130
® (1907): ëThe ìList of Thalassocraciesî in Eusebius: A replyí, JHS 27, 123-30
Nachtergael, G. (1977): Les Galates en Gr@ce et les SEt�ria de Delphes. Recherches díhistoire et dí�pigraphie hell�nistiques (AcadCmie Royale de Belgique Lettres II 63/1), Bruxelles
Nagy, G. (1990): Pindarís Homer. The Lyric Possession of an Epic Past, Baltimore / London
® (2011): ëAsopos and his multiple daughters: Traces of preclassical epic in the Aeginetan odes of Pindarí, in Fearn (ed.) (2011), 41-78
Nannini, S. (1988): Simboli e metafore nella poesia simposiale greca, Roma
Natale, A. (2011): ëSu alcune testimonianze di dattilo solutoí, RCCM 53, 107-45
Natalucci, N. (1995): ëIl P.Oxy 2438 e la tradizione biografica di Pindaro. Una nuova proposta di letturaí, RCCM 37, 57-88
Nauck, A. (1860): Porphyrii philosophi Platonici opuscula tria recognovit ®, Lipsiae
Negri, M. (1998): ëIl catalogo delle opere di Pindaro in Eustazio (Introd. Pind. 34.1-2)í, Aevum(Ant) 11, 217-31
® (2000): Eustazio di Tessalonica. Introduzione al commentario a Pindaro, a cura di ®, Brescia
® (2002): ëLíoschophorikon di POxy 2451 B fr. 17.6 e la sua posizione nellíedizione di Pindaroí, ZPE 138, 31-9
® (2004): Pindaro ad Alessandria. Le edizioni e gli editori, Brescia
Neri, F. (2010): Reliquie eroiche nella Grecia arcaica e classica (VI-IV sec. a.C.), Bologna
Neue, C.F. (1822): Bacchylidis Cei fragmenta, Berolini
Neumann, G. (1965): Gesten und GebGrden in der griechischen Kunst, Berlin
Nocchi Macedo, G. (2011): ëFormes et fonctions de líastCrisque dans les papyrus littCraires grecs et latinsí, S&T 9, 3-33
Norden, E. (19232): Agnostos Theos. Untersuchungen zur Formengeschichte religiTser Rede, Leipzig / Berlin
Norwood, G. (1945): Pindar (Sather Classical Lectures 19), Berkeley/Los Angeles
Oakley, J.H. and Sinos, R.H. (1993): The Wedding in Ancient Athens, Madison WI
Oates, J.F., Samuel, A.L. and Welles, C.B. (1967): í18. Pindar, Unidentified fragmentí, in Yale Papyri in the Beinecke Rare Book and Manuscript Library I (ASP 2), New Haven / Toronto, 37f.
Olivieri, O. (2004): ëLíinno ad Apollo Ptoios di Pindaro (Hymn. frr. 51a-d Maehl.)í, QUCC NS 76, 55-69
® (2011): Miti e culti tebani nella poesia di Pindaro, Pisa/Roma
350
Orsini, P. (2005): Manoscritti in maiuscola biblica. Materiali per un aggiornamento, Cassino
Osborne, R. (1987): ëThe viewing and obscuring of the Parthenon friezeí, JHS 107, 98-105
Otranto, R. (2000): Antiche liste di libri su papiro, Roma
Otto, W. (1933): De sacris Messeniorum, diss. Halle
Page, D. (1955): Sappho and Alcaeus. An Introduction to the Study of Ancient Lesbian Poetry, Oxford
Palumbo Stracca, B.M. (2000): ëLíinno itifallico per Demetrio Poliorceteí, in CannatØ Fera and Grandolini (eds.) (2000), II 503-12
Papadopulos-Kerameus, A. (1891): ëApollodori bibliothecae fragmenta Sabbaiticaí, RhM n.F. 46, 161-92
PapenfuÜ, D. and Strocka, V.M. (2001): Gab es das Griechische Wunder? Griechenland zwischen dem Ende des 6. und der Mitte des 5. Jahrhunderts v.Chr. Tagungs beitrGge des 16. Fachsymposiums der Alexander von Humboldt-Stiftung veranstaltet vom 5. bis 9. April 1999 in Freiburg im Breisgau, Mainz
Pappadakis, N.G. (1923): ë{j lgbTf^Reí, AD 8, 182-256
Pardini, A. (1991): ëLa ripartizione in libri dellíopera di Alceoí, RFIC 119: 257-84
Pariente, A. (1987): ëRapport sur les travaux de líBcole Française en Gryce 1986. Argos 3. � Terrain Karmoyannisí, BCH 111, 591-7
® (1992): ëLe monument argien des «Sept contre Thybes»í, in PiCrart (ed.) (1992), 192-225 and pll. 34-7
® and G. Touchais (eds.) (1998): j[Sge jRb o[SgV^kR. ÁgcgS[Rd^R jRb cgVWgkg_^R / Argos et líArgolide. Topographie et urbanisme, H[Rjfbjn kbWpQg�e |PQWk[^gP, opqQR-j[Sge 28/4-1/5/1990, Paris
Parker, L.P.E. (1997): The Songs of Aristophanes, Oxford
Parker, R. (1996): Athenian Religion. A History, Oxford
® (2005): Polytheism and Society at Athens, Oxford
Pasquali, G. (19522): Storia della tradizione e critica del testo, Firenze
Paterlini, M. (2001): ëI nomoi di Clonaí, RCCM 43, 105-8
Pavese, C.O. (1987): ëIl piª antico frammento di lirica corale grecaí, in Filologia e forme letterarie. Studi offerti a Francesco Della Corte, Urbino, I 53-7
® (1993): ëOn Pindar, fr. 169í, HSCPh 95, 143-57
® (1997): I temi e i motivi della lirica corale ellenica. Introduzione, analisi e indice semantematici: Alcmane Simonide Pindaro Bacchilide, Pisa / Roma
® (2006): ëëTeleboesí and others: On some mistaken proper namesí, CQ NS 56, 587-90
Pellegrino, M. (2000): Utopie e immagini gastronomiche nei frammenti dellíarchaia, Bologna
Pelliccia, H.N. (1995): Mind, Body and Speech in Homer and Pindar (Hypomnemata 107), GNttingen
Perlman, P. (2000): City and Sanctuary in Ancient Greece. The Theorodokia in the Peloponnese (Hypomnemata 121), GNttingen
351
PCron, J. (1974): Les images maritimes de Pindare, Paris
Pfeiffer, R. (1968): History of Classical Scholarship from the beginnings to the End of the Hellenistic Age, Oxford
Pfeijffer, I.L. (1999): Three Aeginetan Odes of Pindar. A Commentary on Nemean V, Nemean III, & Pythian VIII (Mnemosyne suppl. 197), Leiden
PiCrart, M. (1982): ëDeux notes sur líitinCraire argien de Pausaniasí, BCH 106, 139-52
® (ed.) (1992): Polydipsion Argos. Argos de la fin des palais myc�niens I la constitution de líLtat classique. Fribourg (Suisse) 7-9 mai 1987 (BCH suppl. 22), Athynes/Fribourg
® (1997): ëLíattitude díArgos Ø líCgard des autres citCs de líArgolideí, in M.H. Hansen (ed.), The Polis as an Urban Centre and as a Political Community, symposium August, 29-31 1996, 321-51
® (1998): ëLíitinCraire de Pausanias Ø Argosí, in Pariente and Touchais (eds.) (1998), 337-54
Pinervi, A. (2010-11): ëIl prosodio e gli agoni musicalií, Rudiae 22-23, I 415-29
Pintaudi, R. (1995): ëPSI II 147: provenienza e statisticheí, APapyrol 7, 31-3
® (ed.) (1996): Gli archivi della memoria. Biblioteche, filologi e papirologi nei carteggi della Biblioteca Medicea Laurenziana, Firenze
Pohlenz, Max (1955): Griechische Freiheit. Wesen und Werden eines Lebensideals, Heidelberg (Engl. tr. by C. Lofmark, Freedom in Greek Life and Thought: The History of an Ideal, Dordrecht 1966)
Pontani, F.M. (1976): I lirici corali greci. EtI classica, traduzione di ®, Torino
Porciani, L. (1991): ëNote critiche a P. Oxy. 2438aí, SCO 41, 95-100
Pordomingo, F. (1994): ëSur les premiyres anthologies díCpigrammes sur papyrusí, in A. BOlow-Jacobsen (ed.), Proceedings of the 20th International Congress of Papyrologists. Copenhagen, 23-29 August, 1992, Copenhagen, 326-31
® (2007): ëVers une charactCrisation des anthologies sur papyrusí, in B. Palme (ed.), Akten des 23. Internationalen Papyrologen-Kongress, Wien, 22.-28. Juli 2001 (Pap. Vind. 1), Wien, 549-57
® (2013): AntologWas de �poca helenWstica en papiro, Firenze
Power, T. (2000): ëThe parthenoi of Bacchylides 13í, HSCPh 100, 67-81
Prato, C. (1962): I canti di Aristofane. Analisi, commento, scol^ metrici, Roma
Preuner, E. (1903): ëGriechische Siegerlistení, AM 28, 338-82
Privitera, G.A. (1982a): ëLe Istmiche III e IV nel P.Oxy. 2451 e nellíedizione alessandrinaí, in Studi in onore di Aristide Colonna, Perugia, 255f.
® (1982b): Pindaro. Le Istmiche, a cura di ®, Milano
® (1988): ëPindaro, Nem. III 1-5, e líacqua di Eginaí, QUCC NS 29, 63-70
Prodi, E.E. (2013): ëP.Oxy. 2448 (Pi. fr. 215 Sn.-M.) and Pindarís Prosodiaí, ZPE 185, 53-9
Prost, F. (2001): ëSparte et les Cyclades au Ve siycle : Ø propos díID 87í, REA 103, 241-60
Puech, A. (19221-231), (19312-522): Pindare, texte Ctabli et traduit par ®, I-IV, Paris
352
Puglia, E. (1998): ëGli inventari librari di PVindob. Gr. 39966í, ZPE 123, 78-86
Pulleyn, S. (1997): Prayer in Greek Religion, Oxford
Rabbie, E. and Sijpesteijn, P.J. (1988): ëEine neue Perseiden-Genealogie in P.Vindob. G 23058?í, WS 101, 85-95
Rabe, H. (1907): ëAus Rhetoren-Handschriften, 3. Die Quellen des Doxapatres in den Homilien zu Aphthoniosí, RhM n.F. 62, 559-90
Race, W.H. (1982a): ëAspects of rhetoric and form in Greek hymnsí, GRBS 32, 5-14
® (1982b): The Classical Priamel from Homer to Boethius (Mnemosyne suppl. 74), Leiden
® (1987): ëP.Oxy. 2438 and the order of Pindarís worksí, RhM n.F. 130, 407-10
® (1990): Style and Rhetoric in Pindarís Odes, Atlanta
® (1997): Pindar, edited and translated by ®, I-II, Cambridge (MA) / London
Radt, S.L. (1958): Pindars zweiter und sechster Paian. Text, Scholien und Kommentar, Amsterdam
Rawles, R. (2013): ëAristophanesí Simonides. Lyric models for praise and blameí, in E. Bakola, L. Prauscello and M. Telk (eds.), Greek Comedy and the Discourse of Genres, Cambridge, 175-201
Razzetti, F. (2000): ëAristonico e Pindaroí, in G. Arrighetti, Letteratura e riflessione sulla letteratura nella cultura classica. Atti del convegno, Pisa, 7-9 giugno 1999, Pisa, 253-64
Reed, J.D. (1997): Bion of Smyrna. The Fragments and the Adonis, edited with introduction and commentary by ®, Cambridge
Rein, E. (1906): ëAiakos in der Unterweltí, Acta Societatis Scientiarum Fennicae 32 n. 7
Reitzenstein, R. (1893): Epigramm und Skolion. Ein Beitrag zur Geschichte der alexandrinischen Dichtung, GieÜen
® (1897): Geschichte der griechischen Etymologika. Ein Beitrag zur Geschichte der Philologie in Alexandria und Byzanz, Leipzig
Renehan, R.F. (1969): ëConscious ambiguity in Pindar and Bacchylidesí, GRBS 10, 217-28
Roberts, C.H. (1955): Greek Literary Hands 350 B.C.ñA.D. 400, Oxford
® and Skeat, T.C. (1983): The Birth of the Codex, London
Robertson, N. (1992): Festivals and Legends: The Formation of Greek Cities in the Light of Public Ritual (Phoenix suppl. 31), Toronto / Buffalo / London
Roebuck, C.A. (1941): A History of Messenia from 369 to 146 B.C., Ph.D. Chicago
Roesch, P. (1989): ëLíaulos et les aulytes en BCotieí, in H. Beister and J. Buckler (eds.), Boiotika. Vortrage vom 5. Internationalen BTotien-Kolloquium zu Ehren von Professor Dr. Siegfried Lauffer. Institut fBr Alte Geschichte, Ludwig-Maximilians-UniversitGt MBnchen 13.-17. Juni 1986, MOnchen, 203-14
Rolley, C. (1973): ëBronzes gComCtriques et orientaux Ø DClosí, in Ltudes d�liennes publi�es I líoccasion du centi@me anniversaire du d�but des fouilles de líLcole fran`aise díAth@nes I D�los (BCH suppl. 1), 491-524
353
Rose, H.J. (1943): ëPindar, Olymp. iii. 26í, CQ 57, 13
Rosenmeyer, T.G. (1985): ëAncient literary genres: A mirage?í, Yearbook of Comparative and General Literature 34, 74-84 = A. Laird (ed.), Oxford Readings in Literary Criticism, Oxford 2006, 421-39
Rossi, L.E. (1971): ëI generi letterari e le loro leggi scritte e non scritte nelle letterature classicheí, BICS 18, 69-94
® (1995): Letteratura greca, con la collaborazione di R. Nicolai, L.M. Segoloni, E. Tagliaferro e C. Tartaglini, Firenze
Roussel, P. (1923): ëRemarques sur la chronologie des archontes de Delphes au IIIe siycle av. J.-C.í, BCH 47, 1-48
Roux, G. (1956): ëChronique des fouilles et dCcouvertes archCologiques en Gryce en 1955. Argos, IV. Le thCatreí, BCH 80, 376-95
Ruffa, M. (2001): ëLa questione dellíautenticitØ dellíOlimpica 5 di Pindaroí, in CannatØ Fera and DíAlessio (eds.) (2001), 27-45
Ruijgh, C.J. (1968): ëLes noms en -won- (-āwon-, -īwon-), -uon- en grec alphabCtique et en mycCniení, Minos 9: 109-55
Rumpf, A. (1936): ë�p\Q¬ �p\Q�Q _WkYgP]Rí, JDAI 51, 65-71
Rutherford, I. (1988): ëPindar on the birth of Apolloí, CQ NS 38, 65-75
® (1990): ëPaeans by Simonidesí, HSCPh 93, 169-209
® (1991): ëNeoptolemus and the paean-cry: An echo of a sacred aetiology in Pindarí, ZPE 88, 1-10
® (1992): ëTwo heroic Prosodia. A study of Pindar, ìPaeans XIV-Vî í, ZPE 92, 59-72
® (1995): ëEt hominum et deorum Ö laudes (?): A hypothesis about the organisation of Pindarís Paean-bookí, ibid. 44-52
® (1997): ëFor the Aeginetans to Aiakos a prosodion: an unnoticed title at Pindar, Paean 6, 123, and its significance for the poemí, ZPE 118, 1-21
® (2000a): ëHßaChÞ}Þa´. A musical term in a Delian inscription (IG XI 120.49)í, ZPE 130, 147f.
® (2000b): ëKeos or Delos? State-pilgrimage and the performance of Paean 4í, in CannatØ Fera and Grandolini (eds.) (2000), II 605-12
® (2000c): ëTheoria and darsaan: Pilgrimage and vision in Greece and Indiaí, CQ NS 50, 133-46
® (2001): Pindarís Paeans. A Reading of the Fragments with a Survey of the Genre, Oxford
® (2003): ëThe prosodion. Approaches to a lyric genreí, in Benedetti and Grandolini (eds.) (2003), II 713-26
® (2004): ëmg[Ù] W∑] �j fv]kW fv] c¦VWT] Ö (Xen. Mem. 3.3.12): Song-dance and state-pilgrimage at Athensí, in P. Murray and P. Wilson, Music and the Muses. The Culture of ëMousikeí in the Classical Athenian City, Oxford, 67-90
354
® (2005): ëIn a virtual wild space: Pilgrimage and rite de passage from Delphi to Sabarimalaií, in D. Yatromanolakis and P. Roilos (eds.), Greek Ritual Poetics, Cambridge MA, 323-37
® (2009): ëBlack sails to Achilles: the Thessalian pilgrimage in Philostratusí Heroicusí, in E. Bowie and J. Elsner (eds.), Philostratus, Cambridge, 230-47
® (2011): ëThe Thea lrion of the Pythian Oneí: The Aeginetan theacroi in contextí, in Fearn (ed.) (2001), 114-28
Rutherford, W.G. (1896): Scholia Aristophanica, being such comments adscript to the text of Aristophanes as have been preserved in the Codex Ravennas, arranged, emended, and translated by ®, I-II, London/New York
Sandys, J. (19192): The Odes of Pindar including the Principal Fragments, with an introduction and an English translation by Sir ®, London/Cambridge MA
Sbardella, L. (2007): ëPindaro, Omero e la democrazia ateniese: un contributo esegetico alla Nemea VIIí, SemRom 10, 69-99
Schachermeyr, F. (1950): Poseidon und die Entstehung des griechischen GTtterglaubens, Bern
Schachter, A. (1972): ëSome underlying cult patterns in Boeotiaí, in Proceedings of the First International Conference on Boiotian Antiquities (McGill University, Montr�al, 18.3.1972) (Teiresias suppl. 1), MontrCal, 17-30
® (1981-94): Cults of Boiotia (BICS suppl. 38), I-IV, London
® (2010-11): ëThe Mouseia of Thespiai: organization and developmentí, Rudiae 22-23, I 31-61
Schlageter, J. (1909): Der Wortschatz der auSerhalb Attikas gefundenen attischen Inschriften. Ein Beitrag zur Entstehung der Koine, Konstanz
Schmid, W. (1929): ® and O. StÅhlin, Geschichte der griechischen Literatur, I: Die klassische Periode der griechischen Literatur, 1: Die griechische Literatur vor der attischen Hegemonie, MOnchen
Schmidt, E. (1616): HbQkn[gP cW[^gkg], hoc est Pindari lyricorum principis, plus quam sexcentis in locis emaculati, ut jam legi & intelligi possit, ∆VP_cbgQ~jRb. HPpbgQ~jRb. ´W_WgQ~jRb. ∏]p_bgQ~jRb. Illustrati Versione nova fideli. Rationis metricX indicatione certa. Dispositione textus genuine. Commentario sufficiente. Cum fragmentis aliquot diligenter collectis. Indice locuplete, Victorum, Autorum, [sic] Rerum & Verborum. Discursu duplici; uno de Dithyrambis: altero de Insula Atlantica ultra Columnas Herculis, quae America hodi@ dicitur. opera ®, [Wittemberg]
Schmitt, R. (1967): Dichtung und Dichtersprache in indogermanischer Zeit, Wiesbaden
Schneider, J.G. (1776): Carminum Pindaricorum fragmenta curavit ®, Argentorati
Schneidewin, F.W. (1834): ëUeber ein neuentdecktes BruchstOck eines Pindarischen Threnosí, RhM 2, 110-21
® (1837): Eustathii proúmium commentariorum Pindaricorum edidit ®. Accesserunt: I. Supplementa corporis fragmentorum Pindaricorum Boeckhiani. II. Exercitationum Criticarum in Poetas Graecos capita VI. VII. VIII. IX. X. XI, Gottingae
Schoeffer, V. de (1889): De Deli insulae rebus, Berlin
355
Schott, A. (1585): {j fv] H[¦jVgP Z[\]fg_Rp^R] S[R__Rfbjv] �jVgSR^. Photii selecta ex locis laudabilib[us]. quos Proclus ex grammaticis excerpserat, nunc primum edita ex biblioteca A. Augustini Tarraconensium Archiepiscopi interprete ®, Tarracone
® (16152): Procli Chrestomathia poetica interprete & scholiaste ® Antuerp., accessere & notb P.I. Nunnesii Valentini. Item Fragmentum scriptoris incerti anteØ Censorino tributum, & cum eius De die natali libro continenter impressum, nunc correctius & aliquot capitibus ex veteri libro additis auctius editum a L. Carrione Brugensi, cum notis eiusdem Nunnesii, Hanouib
Schrevelius, C. (1668): �]PZ^gP VWXbj¦Q cum Variis Doctorum virorum Notis vel editis antera vel ineditis, Junii, Scaligeri, Salmasii, Palmerii, Guieti, Sopingii, Cocceii, Gronovii, Tanaq. Fabri, &c. accurante ®. Accessit J. Pricbi Index Auctorum, qui ab Hesychio laudantur, Lugd[uni] Batav[orum] et Roterod[ami]
SchrNder, S. (1999a): Geschichte und Theorie der Gattung Paian. Eine kritische Untersuchung mit einem Ausblick auf Behandlung und Auffassung der lyrischen Gattungen bei den alexandrinischen Philologen (BeitrÅge zur Altertumskunde 121), Stuttgart/Leipzig
® (1999b): ëZwei xberlegungen zu den Liedern vom Athenerschatzhaus in Delphií, ZPE 128, 65-75
Schroeder, O. (1878): Studia Pindarica, Berlin
® (19005) (19236): Pindari carmina recensuit ®, Lipsiae/Berolini
® (19081): Pindari carmina cum fragmentis selectis edidit ®, Lipsiae
Schubart, W. (19613): Das Buch bei den Griechen und RTmern, hrsg. von E. Paul, Leipzig
Segal, C.P. (1967): ëPindarís seventh Nemeaní, TAPhA 98, 431-80
Sevieri, R. (1999): Pindaro. Frammenti, Milano
Sijpesteijn, P.J. and Worp, K.A. (1974): ëLiterary and semi-literary papyri from the Vienna Papyrus Collectioní, CE 49, 309-31
Sitzler, J. (1915): review of Maas (1913-21) I, WKPh 32, 385-8
Slater, W.J. (1969): Lexicon to Pindar, edited by ®, Berlin
Smiley, M.T. (1914): ëCallimachusí debt to Pindar and othersí, Hermathena 40, 46-72
Smyth, H. Weir (1900): Greek Melic Poets, London / New York
Snell, B. (19345), (19587): Bacchylidis carmina cum fragmentis edidit ®, Lipsiae
® (1938): ëIdentifikationen von PindarbruchstOckení, Hermes 73, 424-39
® (1940): ëDrei berliner Papyri mit StOcken alter Chorlyrikí, Hermes 75, 177-91
® (1946): ëPindars Hymnus auf Zeusí, A&A 2, 180-92
® (19531): Pindari carmina cum fragmentis edidit ®, Leipzig
® (19643): Pindari carmina cum fragmentis edidit ®, pars altera: Fragmenta. Indices, Lipsiae
® (1962): ëPindars 8. Paian Ober die Tempel von Delphií, Hermes 90, 1-6
® (19754): Die Entdeckung des Geistes. Studien zu Entstehung des europGischen Denkens bei den Griechen, GNttingen (Engl. tr. of 19553 by T. Rosenmeyer, The Discovery of the
356
Mind: The Greek Origins of European Thought, New York 1960 = The Discovery of the Mind in Greek Philosophy and Literature, London 1982)
Steiner, D. (1986): The Crown of Song: Metaphor in Pindar, London
Stephanus, H. (15601) (15662) (15863) (16004): Pindari Olympia, Pythia, Nemea, Isthmia. CXterorum octo lyricorum carmina, Alcaei, Sapphus, Stesichori, Ibyci, Anacreontis, Bacchylidis, Simonidis, Alcmanis, nonnulla etiam aliorum. I-II, [Paris] [Geneva]
Sticotti, P. (1898): ëZu griechischen HochzeitsgebrÅuchení, in Festschrift fBr Otto Benndorf zu seinem 60. Geburtstage gewidmet von SchBlern, Freunden und Fachgenossen, Wien, 181-8
Stoneman, R. (1977): A Commentary on Six Nemean Odes of Pindar, D.Phil. Oxford
® and Conway, G.S. (1997): Pindar. The Odes and Selected Fragments, London/Rutland VE
Storey, I.C. (2003): Eupolis. Poet of Old Comedy, Oxford
Strauss Clay, J. (2011): ëOlympians 1-3: A song cycle?í, in Athanassaki and Bowie (eds.) (2011), 337-45
Stuart Jones, H. (1894): ëThe chest of Kypselosí, JHS 14, 30-80
Studemund, G. (1886): Anecdota varia Graeca et Latina ediderunt R. Schoell et ®, I: Anecdota varia Graeca metrica musica grammatical edidit ®, Berolini
SuUrez de la Torre, E. (1997): ëNeoptolemos at Delphií, Kernos 10, 153-76
Svenbro, J. (1984): ëLa dCcoupe du poyme. Notes sur les origines sacrificielles de la poCtique grecqueí, Po�tique 58, 215-32 = ëIl taglio della poesia. Note sulle origini sacrificali della poetica grecaí, StudStor 25, 925-44
Swift, L. (2010): The Hidden Chorus. Echoes of Genre in Tragic Lyric, Oxford
Sylburg, F. (1590): Apollonii Alexandrini de syntaxi seu constructione orationis libri IIII: a F. Porto ante aliquot annos e manuscripto codice passim & correcti & suppleti; tum Latine redditi, & notationibus illustrati: nunc denuo a ® cum bonis exemplaribus collati, & notationibus aucti. Addita e Procli Chrestomathia grammatica, Photii patriarchae CP. Electa; ab A. Schotto primum edita, Latinitate donata, & scholiis illustrata; nunc iterum a ® m.s. exemplaris collatione passim emendata, suppleta. notationibus aucta, Francofurti
® (1594): {fP_gVgSbjÙQ fÙ _YSR, ¶SgPQ ô _WSnV\ S[R__Rfbjq. Etymologicum magnum; seu magnum grammaticX penu: In quo & Originum & Analogib doctrina ex Veterum sententia copiosissime proponitur: Historib item & Antiquitatis monumenta passim attinguntur: Superiorum editionum variorumq(ue) auctorum collazione a multis ac fúdis mendis repurgatum, perpetuis notis illustratum, tribusq(ue) utiliÜimis Indicibus, Verborum, Rerum, atq(ue) Auctorum numero paene infinito rum, nunc recens adauctum: Opera ® Veter., [Heidelberg]
Taillardat, J. (19652): Les images díAristophane. Ltudes de langue et de style, Paris
Tausend, K. (2012): ëDas Prosodion des Eumelos. Literarische Propaganda und politische IdentitÅtsstiftung im 4. Jh. v. Chr.í, Historia 61, 66-77
Tedeschi, A. (1985): ëLíinvio del carme nella poesia lirica arcaica: Pindaro e Bacchilideí, SIFC III 3, 29-54
Tessier, A. (1995): Tradizione metrica di Pindaro, Padova
357
® (1999): ëLíanalogia cosmo-triade strofica e la sua fortuna bizantinaí, in G. Avezzª (ed.), }bkR]jRV^Rb. Tradizione e interpretazione del dramma attico, Padova, 201-10
Theiler, W. (1941): Die zwei Zeitstufen in Pindars Stil und Vers, Halle (Saale) = Untersuchungen zur antiken Literatur, Berlin 1970, 148-91
Thiersch, F. (1820): HbQkn[gP fÏ ]T�¦_WQR. Pindarus Werke, Urschrift, Uebersetzung in den pindarischen VersmaaÜen und ErlÅuterungen von ®, I-II, Leipzig
Thummer, E. (1968-9): Pindar. Die isthmischen Gedichte, textkritisch herausgegeben, Obersetzt und kommentiert mit einer Analyse der pindarischen Epinikien von ®, I: Analyse der pindarischen Epinikien, Text und dbersetzung der isthmischen Gedichte, II: Kommentar, Heidelberg
Tomlinson, R.A. (1972): Argos and the Argolid. From the end of the Bronze Age to the Roman Occupation, London
Too, Y.L. (2010): The Idea of the Library in the Ancient World, Oxford
Tosi, T. (1908): ëSul sesto peana di Pindaroí, A&R 11, 201-20 = Scritti di filologia e di archeologia, a cura di N. Terzaghi, Firenze 1957, 99-114
TrCdC, M. (2003): ëPrCsence et images des poytes lyriques dans le thCitre díAristophaneí, in J. Jouanna and J. Leclant (eds.), Colloque: La po�sie grecque antique. Actes, Paris, 169-83
Turnebus, A. (1553): ÌdRb]f^TQg] �VWXRQk[YT] {SZWb[^kbgQ cW[Ú _Yf[TQ jRÚ cgb\_nfTQ. Ý∞] fÙ RÃfÙ ]Z¦VbR, Parisiis
Turyn, A. (1948): Pindari carmina cum fragmentis, Cracoviae
Turner, E.G. (1952): ëOxyrhynchus and its papyrií, G&R 21, 127-37, 143f. and pll. CXXI-IV
® (1956): ëScribes and Scholars of Oxyrhynchusí, in Akten des VIII. internationalen Kongresses fBr Papyrologie, Wien 1955 (Mitteilungen aus der Papyrussammlung der Tsterreichischen Nationalbibliothek NS 5), Wien, 141-6 = A.K. Bowman et al. (edd.) (2007), 256-61
® (1968): Greek Papyri. An Introduction, Oxford
® (1977): The Typology of the Early Codex, University of Pennsylvania
Turyn, A. (1948): Pindari carmina cum fragmentis, Cracoviae
Ucciardello, G. (1996-97): ëRiesame di P. Oxy. 2368: alcuni problemi di lettura e di interpretazioneí, APapyrol 8-9, 61-88
® (2000): ëdpYS_R/dpYSS_R ed altre sequenze -SS_-: teoria grammaticale e prassi graficaí, APapyrol 12, 63-91
® (2001): ëPOxy. XXXII 2636: commentario a Pindaro o a Ibico?í in CannatØ Fera and DíAlessio (eds.) (2001), 87-116
Valesius, H. (1740): Henrici Valesii emendationum libri quinque et de critica libri duo nunquam antehac typis vulgatis. Ejusdem ut et N. Rigaltii et I. Bullialdi Dissertationes de Populis fundis. Accedunt H. Valesii Orationes variae junctim excusae. & H. Valesii Oratio de Laudibus Ludovici XIV. & Carmina nonnulla inedita. Praefixa est H. Valesii Vita, ab Auctore H. Valesio recognita & emendata, Amstelaedami
Vatin, C. (1982): ëMonuments votifs de Delphes, V: Les couroi díArgosí, BCH 106, 509-25
358
Verdenius, W.J. (1955): ëNotes on Hippocrates Airs Waters Placesí, Mnemosyne IV 8, 14-18
® (1987): Commentaries on Pindar I (Mnemosyne suppl. 97), Leiden
® (1988): Commentaries on Pindar II (Mnemosyne suppl. 101), Leiden
Vermeule, E. (1965): ëFragments of a symposion by Euphroniosí, in AK 8, 36-9 and pll. 11-14
Viljoen, G. van Niekerk (1955): Pindaros se tiende en elfde Olympiese odes (with a summary in English), Leiden
Vitelli, G. (1884): ëSpicilegio fiorentino XI. Osservazioni sulla sticometria delle opere poetiche (Pindaro, Sofocle, Euripide, Gregorio Nazianzeno etc.í, Museo italiano di antichitI classica 1, 160-4
® (1913): ë147. Frammenti di Peani di Pindaroí, in Papiri Greci e Latini vol. II, 73-9, see also p. X
von Leutsch (1856): ëPindarische Studien. I: Die quellen fOr die biographie des Pindarosí, Philologus 11, 1-35
Waanders, F.M.J. (1983): The History of fYVg] and fWVYT in Ancient Greek, Amsterdam
Wachsmuth, D. (1967): H¦_cb_g] ¡ kR^_TQ. Untersuchung zu den antiken Sakralhandlungen bei Seereisen, Berlin
Wackernagel, J. (1916): Sprachliche Untersuchungen zu Homer, GNttingen
Wagman, R.S. (1986a): ëLe citazioni pindariche in Strab. IX, 2, 33-34í, Athenaeum NS 64, 111-26
® (1986b): ëNote su un frammento di Pindaro (51a Snell)í, ibid. 397-406
Wagner, D. and Theurillat, T. (eds.) (2004): Eretria. A Guide to the Ancient City, Gollion
Walter, H. (1983): Die Leute im alten 'gina 3000-1000 v.Chr., Stuttgart
® (1993): 'gina. Die archGologische Geschichte einer griechischen Insel, MOnchen
Walter-Karydi, E. (1994): ëDas Thearion von >gina. Zum Apollonkult auf >ginaí, AA 1994: 125-38
® (2000): ëBgine et Delphesí, in A. Jacquemin (ed.), Delphes cent ans apr@s la Grande fouille. Essai de bilan. Actes du colloque international organise par líLcole Fran`aise díAth@nes, Ath@nes-Delphes 17-20 septembre 1992 (BCH suppl. 36), Athynes, 87-98
® (2006): How the Aiginetans formed their identity, tr. by J. Clough, Athens
van der Weiden, M.J.H. (1991): The Dithyrambs of Pindar. Introduction, text and commentary, Amsterdam
Webster, T.B.L. (1970): The Greek Chorus, London
Weil, H. (1895): ëUn pCan delphique Ø Dionysosí, BCH 19, 393-418
Weilhartner, J. (2010): 'gina-Kolonna. Forschungen und Ergebnisse, III: Testimonia. Die literarischen Zeugnisse Bber das antike Aigina von Homer bis in byzantinische Zeit, Wien
Welter, G. (1938): Aigina, Berlin
Werner, O. [1967]: Pindar. SiegesgesGnge und Fragmente. Griechisch und deutsch herausgegeben und Obersetzt von ®, MOnchen
359
Wesseling, P. (1745): }bgki[gP fgt CbjWVbifgP �b�Vbgpqj\] �]fg[bjv] fÏ ]T�¦_WQR. Diodori Siculi bibliothecae historicae libri qui supersunt, interprete L. Rhodomano. Ad fidem mss. recensuit ® atque H. Stephani, L. Rhodomani, F. Ursini, H. Valesii, J. Palmerii & suas adnotationes, cum indicibus locupletissimis, adjecit. Tomus I, Amstelodami
West, M.L. (1966): Hesiod. Theogony, edited with prolegomena and commentary by ®, Oxford
® (1978): Hesiod. Works & Days, edited with prolegomena and commentary by ®, Oxford
® (1982): Greek Metre, Oxford
® (1985a): The Hesiodic Catalogue of Women. Its Nature, Structure and Origins, Oxford
® (1985b): ëIon of Chiosí, BICS 32, 71-8
® (1992): Ancient Greek Music, Oxford
® (1998): Homeri Ilias recensuit / testimonia congressit ®, I: rhapsodias I-XII continens, Stutgardiae et Lipsiae
® (2002): ëëEumelosí: a Corinthian epic cycle?í, JHS 122, 109-33
® (2013): The Epic Cycle. A Commentary on the Lost Troy Epics, Oxford
White, J.W. (1914): The Scholia on the Aves of Aristophanes with an introduction on the origin, development, transmission, and extant sources of the old Greek commentary on his comedies, collected and edited by ®, Boston / London
Wieseler, F. (1843): Adversaria in Aeschyli Prometheum vinctum et Aristophanis Aves philologica atque archaeologica scripsit ®, Gottingae
Wilamowitz-Moellendorff, U. von (1900): Textgeschichte der griechischen Lyriker, Berlin
® (1901): ëHieron und Pindarosí, SPAW 1901, 1273-318 = Kleine Schriften VI: Philologiegeschichte, PGdagogik und Verschiedenes, Nachlese zu den BGnden I und II, NachtrGge zur Bibliographie, Berlin 1972, 234-85
® (1908): ëPindars siebentes nemeisches Gedichtí, SPAW 1908, 328-52 = Kleine Schriften VI, 286-313
® (1918): ëDichterfragmente aus der Papyrussammlung der Kgl. Museení, SPAW 1918, 728-51
® (1913): Sappho und Simonides. Untersuchungen Bber griechische Lyriker, Berlin
® (1919): ëLesefrOchteí CLV, Hermes 54, 54-7 = Kleine Schriften IV: LesefrBchte und Verwandtes, Berlin 1962, 292-5
® (1922): Pindaros, Berlin
Will, B. (1955): Korinthiaka. Recherches sur líhistoire et la civilisation de Corinthe des origines aux guerres m�diques, Paris
Willcock, M.M. (1995): Pindar. Victory Odes. Olympians 2, 7, 11; Nemean 4; Isthmians 3, 4, 7, edited by ®, Cambridge
Willetts, R.F. (1967): The Law Code of Gortyn, edited with introduction, translation and a commentary by ® (Kadmos suppl. 1), Berlin
Willisch, E.G. (1875): Ueber die Fragmente des Epikers Eumelos, Zittau 1875
Wilson, J.R. (1979): ëæR^ jY fb] ‚kí �[WW~: An Homeric device in Greek literatureí, ICS 4, 1-15
Wilson, N.G. (1967): ëA chapter in the history of scholiaí, CQ NS 17, 244-56
360
® (1983): Scholars of Byzantium, London
® (2007): Aristophanis fabulae. Recognovit brevique adnotatione critica instruxit ®, I-II, Oxonii
Wilson, P. (1999): ëThe aulos in Athensí, in S. Goldhill and R. Osborne, Performance Culture and Athenian Democracy, Cambridge, 58-95
® (2000): The Athenian Institution of the Khoregia. The Chorus, the City and the Stage, Cambridge
® (2003): ëThe sound of cultural conflict: Kritias and the culture of mousikē in Athensí, in C. Dougherty and L. Kurke (eds.), Cultural Poetics in Archaic Greece. Cult, Performance, Politics, Cambridge, 181-206
® (2007): ëPronomos and Potamon: Two pipers and two epigramsí, JHS 127, 141-9
® (2010): ëThe man and the music (and the choregos?)í, in O. Taplin and R. Wyles (eds.), The Pronomos Vase and its Context, Oxford, 181-212
Wright, M. (2012): The Comedian as Critic. Greek Old Comedy and Poetics, London
WOst, E. (1967): Pindar als geschichtschreibender Dichter. Interpretationen der 12 vorsizilischen Sieges-lieder, des sechsten Paians und der zehnten olympischen Ode, diss. TObingen
Zanetto, G. (1987): Aristofane. Gli uccelli, a cura di ®, introduzione e traduzione di D. Del Corno, Milano
Ziehen, L. (1906): Leges Graecorum sacrae e titulis collectae ediderunt et explanaverunt I. de Prott et ®, II/1: Leges Graeciae et insularum edidit ®, Lipsiae
Zimmermann, B. (2000): ëLyric in the fragments of Old Comedyí, in D. Harvey and J. Wilkins (eds.), The Rivals of Aristophanes. Studies in Athenian Old Comedy, London/Swansea, 273-80
® (20082): Dithyrambos. Geschichte einer Gattung, Berlin
Zimmerman, M. et al. (2004): Apuleius Madaurensis. Metamorphoses. Books IV 28-35, V and VI 1-24: The Tale of Cupid and Psyche, text, introduction and commentary, Groningen
Zunker, A. (1988): Untersuchungen zur Aiakidensage auf Aigina, St. Ottilien
Zunino, M.L. (1997): Hiera Messeniaka. La storia religiosa della Messenia dallíetI micenea allíetI ellenistica, Udine
Zuntz, G. (1935): ëDe Pap. Berol. 13411í, Aegyptus 15, 282-96
Related Documents
