UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Stesichorus and the epic tradition Maingon, Alison Dale 1978

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1978_A1 M33.pdf [ 21.12MB ]
JSON: 831-1.0100235.json
JSON-LD: 831-1.0100235-ld.json
RDF/XML (Pretty): 831-1.0100235-rdf.xml
RDF/JSON: 831-1.0100235-rdf.json
Turtle: 831-1.0100235-turtle.txt
N-Triples: 831-1.0100235-rdf-ntriples.txt
Original Record: 831-1.0100235-source.json
Full Text

Full Text

STESICHORUS AND THE EPIC TRADITION by Alison Dale Maingon M.A., University of St. Andrews, 1970 M.A., University of Alberta, 1972 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF THE FACULTY OF GRADUATE STUDIES (Department of Classics) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May 1978. Cc} Alison Dale Maingon, 1978 DOCTOR OF PHILOSOPHY in In presenting th i s thes is in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree ly ava i lab le for reference and study. I fur ther agree that permission for extensive copying of th is thesis for scho lar ly purposes may be granted by the Head of my Department or by his representat ives . It is understood that copying or pub l i ca t ion o f th is thes is fo r f inanc ia l gain sha l l not be allowed without my writ ten permission. Department of d-^SSlO4} The Univers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date i Abstract In antiquity Stesichorus was labelled "Homeric" by the commentators, but his innovation i n myth was also noted. Until the discovery of fragments of his poems among the papyri from Oxyrhynchus there was l i t t l e material from which any conclusions could be reached regarding the poet's treatment of his inheritance from the epic tradition. In this dissertation,therefore , I have examined the new evidence from the papyri with a view to assessing the poet's reliance upon that tradition in both diction and content, and the extent to which he was innovative. The poet's language at a morphological level i s seen to be almost identical to that of epic, whereas at the level of phonology the intrusion of a Western or "Doric" pronunciation has occurred. The poets adaptation of Homeric "formulae" reveals a prevalent tendency to avoid the repetition of phrases commonly found i n the epic corpus by the introduction of new, unprecedented word-associations. In chapter VI a detailed examination of four of the fragments of four poems,(the Nostoi, the Sack of Troy, the Geryoneis and the Suotherae) indicates the ways i n which the poet adapted thematic elements from the epic tradition, amalgamating epic with non-epic , conventional with original material. The second half of this dissertation i s devoted to Stesichorus' treatment of the inherited body of Greek myth. In those poems in which the poet was concerned with the legends of mainland Greece, innovations appear , with the exception of the Palinode, not to alter the basic structure of the myths. However, there i s evidence of the poet's interest i n elaborating upon or inventing legends located i n the Greek i i west, notably in poems relating the exploits of Heracles, but also i n his orestela and Sack of Troy. In so doing the poet would create or give authority to a body of myths specifically relevant to his western audience. i i i Table of Contents Page Abstract i Table of Contents i i i Abbreviations and Texts Av Acknowledgement v Chapter I Stesichorus, Homer's heir i n the Greek west. 1 Chapter II Stesichorus 1 "mixed" dialect. 13 Chapter III Verbatim adaptation of Homeric "formulae" in Stesichorus. 53 Chapter IV Stesichorus' modification of "formulae" from Homer. 82 Chapter V "Formulaic" expressions that are found i n non-Homeric epic. 137 Chapter VI Structural patterns derived from epic. 157 Chapter VII Stesichorus' adaptation of myth from the epic tradition. 234 Chapter VIII Heracles i n the poems of Stesichorus . 265 Chapter IX That infamous Palinode . 300 Chapter X The Sack of Troy_ . 330 Chapter XI Conclusion . 355 Bibliography 361 iv Abbreviations and Texts Throughout this dissertation I have referred to the edition of Stesichorus' fragments made by Page i n Poetae Melici Graeci (Oxford, 1962) in the abbreviated form PMG. The new fragments have been cited according to their number in the Oxyrhynchus Papyri volumes,published by the Egypt Exploration Society,in the form P.Oxy. : thus the fragments of the Geryoneis are designated P.Oxy. 2617. The texts of the I l i a d , Odyssey and the Hymns are derived from the Oxford Classical Texts series, Homerl Opera vols. I-V, edited by Munro and Allen. The texts of Hesiod are derived from Solmsen's edition , Hesiodi Theogania, Opera et Dies, Scutum (Oxford, 1970) while the fragments are cited as in Fragmenta  Hesiodea, edited by Merkelbach and West (Oxford, 1967) and so indicated by the letters M.&W. Other abbreviations employed are as follows: ABFV; J.Boardman, Athenian Black-Figure Vases (London, 1967). ABV : J.D.Beazley, Attic Black-Figure Va3e-Painters (Oxford, 1956). ARV j J. Beazley, Attic Red-Figure Vase-Painters (Oxford, 2nd ed., 1963) . DK : H. Diels and W. Kranz, Die Fragmente der Vorsokratiker (Berlin, 6th ed., 1952). FGH : F. Jacoby, Die Fragmente der griechischen Historiker (Leiden, 2nd ed., 1957). GLP : C.M.Bowra, Greek Lyric Poetry (Oxford, 2nd ed., 1961). LGS : D. Page, Lyrica Graeca Selecta (Oxford, 1968). LSJ : Liddell-Scott-Jones, A Greek - English Lexicon (Oxford, 9th ed., 1940). OCD : N.G.L.Hammond and H.H.Scullard, The Oxford Classical Dictionary (Oxford, 2nd ed.1970). PLF : E.Lobel and D.Page, Poetarum Lesbiorum Fragmenta (Oxford, 1955). SLG s D.Page, Supplementum Lyricls Graecis (Oxford, 1974). A l l Journals are referred to in the conventional abbreviations found i n L'Ann^e Philologjque. V Acknowledgements I wish to thank H.G. E d i n g e r f o r h i s d i l i g e n t guidance and encouragement throughout the w r i t i n g o f t h i s t h e s i s . I am a l s o i n d e b t e d to Mrs E.A.E. Bongie and A . J . P o d l e c k i f o r t h e i r h e l p f u l comments i n the f i n a l s t a g e s o f t h i s work. Chapter I Stesichorus, Homer's heir in the Greek west. E t a a u x o p o v , e c n t X n & e s d u e x p r t t o u OToya M o t i a n s , e x x ^ p t a e v K a x c f v a s a C S a X c J e v 6dne6ov oZ xaxa nudaydpayipwaLxav (pcfxtv a i t p t v ' O u n p o u 4»uxc\ e v t a x £ p v o u s 6 e v 5 x e p o v < i i , x J a a x o . AP VII 75. Stesichorus was the f i r s t eminent li t e r a r y figure to emerge from, the western Greek world and to make such an impact on the cultural centres of the Greek mainland that his works were ensured preservation for posterity.^" The poems that brought him fame were primarily his arrangements of epic themes to be performed to the accompaniment of the 2 lyre: hence epici carminia onera lyra sustinentem. The surviving t i t l e s and fragments give us a f a i r indication of the extent and limitations of the poet's new approach to heroic poetry. Although he may well have been preceded by Terpander (and others unknown) in the invention of musical settings for the traditional epics, his poems on epic themes appear to have been distinctive in their completely " l y r i c a l " form, composed as they were in a triadic structure and adapted to nomoi for the lyre. Whether the number "twenty-six" quoted i n the Suda refers to volumes or t i t l e s known to the author(s) of that lexicon, we known of only thirteen t i t l e s of poems on legends from the epic traditions (Athla, Geryoneis, Helen, Palinode, Eriphyle, Europeia, Ili o u  Persis, Cerberus, Cycnua, Nostoi, Oresteia I and II, Scylla, Suotherae) and, i n addition to these, one or two poems composed on a less lofty plane, closer to the sentimental romance of the early "novel" (Calyce, Daphnis, Rhadine). The poet's reputation in antiquity was wide-spread, i f we can believe Cicero : (Stesichorus) qui frait Himerae, sed et est  et f u i t tota Graecia summo propter ingenium honore et nomine (Verr. II 2 23) . 2 The precise dates of the poet's l i f e are no more readily available to us than they were to our predecessors, the philologoi of the ancient world, nor are they of direct relevance for an appreciation of the poet's a r t i s t i c achievement. West rightly c r i t i c i s e s the all-too-hasty acceptance of the dates assigned by the Suda, Eusebius,et a l . , who appear to follow a tradition that Stesichorus was born i n the 37th Olympiad 5 and died i n the 56th, The comparative latitude of the 4-year Olympiad' has been reduced to a misleadingly exact 632-556 B.C. i n many hand-books of Greek literature,with l i t t l e justification.^ Of the various traditions claiming either that Stesichorus was the son of Hesiod, or that his death occurred i n the same year as Simonides' birth, or that he was Pythagoras' contemporary, none can be divorced from the context of the lives of other 7 literary figures whose dates are equally uncertain. A possible source for the Suda may have been a schematising literary history such as that of Apollodorus, i n which convenient synchronisations of poets' lives 8 were made. Apart from the evidence of the Parian Marble, which claims 9 ' Stesichorus came to Greece in 485/4 B.C., chronological associations point to Stesichorus' having been alive and active i n the major part of the 6th century. We find allusions to the poet's reaction to an eclipse of the sun, to his relationship with Phalaris, tyrant of Acragas, and to a connection with a conflict between Locri and Croton, a l l of which may be firmly placed i n the 6th century.^ These allusions, together with Simonides' reference to Stesichorus in a poem,^ outweigh the 5th-century date suggested by the Parian Marble. The early chronographers themselves produced chronological schemes of events in 12 the Greek west which were widely discrepant. There is,therefore, l i t t l e point in pursuing the matter of the poet's dates further in the present context. 3 Nor is there a single tradition concerning Stesichorus'>place of origin: Mataurus in southern Italy, Himera in S i c i l y , Pallantion in Arcadia.^ The discussions of various scholars have proved that again the evidence i s inconclusive. The tradition of a Stesichorus :,trom 14 Himera appears i n a number of ancient sources. On the other hand, West, in the most recent compilation of the evidence, i s inclined to see a more consistent tradition connecting Stesichorus with Mataurus in 15 southern Italy. In fact, from the scanty remains of biographical information we can assume only that Stesichorus pursued a distinguished career as a poet in the Greek west, associated with Himera in particular, but also known to have travelled and resided i n southern Italy. Just as his predecessors, the wandering bards and rhapsodes, gave recitations aj.1 over the Greek world, so Stesichorus must have carried his talents to the large cultural centres of both S i c i l y and southern Italy, even to Greece itself,and hence there arose a number of traditions from different quarters claiming association with the poet. For our present purpose we need only acknowledge the poet's presence i n the Greek west in the 6th century. By the middle of the 6th century many of the original Greek colonies of southern Italy and S i c i l y were flourishing centres of commercial activity, for example, Syracuse, Zancle, Rhegion, Tarenturn, Locri, Croton, Sybaris.^ Some of these c i t i e s had themselves established colonies, particularly i n the more westerly areas of Si c i l y ; Selinus, Himera and Acragas were colonies of colonies. Exploration had carried the Greeks into the western reaches of the Mediterranean, to the mineral wealth of Spain and to the shores of the 17 Atlantic , so that not only the agricultural richness of the c i t i e s of 4 southern Italy and S i c i l y , but also their position on the sea-routes to the far west made them of crucial importance commercially to the states of the Greek mainland, particularly Athens and Corinth. As a result of this commercial importance, the city-states of southern Italy and S i c i l y prospered and their prosperity i s duly witnessed by the magnificence of their public architecture and sculpture, their locally produced coinage and pottery, together with the profusion of imported pottery of 18 high quality. I t has been noted that the styles prevalent i n the mid to late 6th century may have lagged behind the trends of Athens and Corinth, but this fact i s by no means indicative of inferior quality i n 19 the a r t i s t i c achievement of the western colonies. The particular tastes of the western .market, as much as the distance from the chief centres of a r t i s t i c activity in the Aegean, may account for what appears to be a slower evolution i n forms. It i s important therefore to note that prosperity^ i n the west i n the 6th century fostered a l i v e l y interest in the preservation of the a r t i s t i c heritage of the mother-country . We can assume that western interest in i t s Hellenic inheritance was not restricted to the plastic arts alone. The colonists settling in the west must have imported the traditional religion and myths of their forebears. At any rate, certainly i n the 6th century there i s evidence of strong ties preserved between the colonies in the west 20 and the religious centres of the mainland of Greece. At both Delphi and Olympia treasuries were b u i l t by c i t i e s such as Syracuse, Tarenturn, Sybaris and Gela. From the l i s t s of victors at the Olympic Games we know that athletes came from the west to compete as early as 684 B.C., when a man from Syracuse i s recorded as having won the wrestling event. 5 In 576 B.C. the f i r s t seven men to fi n i s h in the stadion foot-race were from Croton, and Milon of that same ci t y was one of the most remarkable wrestlers seen at Olympia; his victories spanned six successive 21 Olympiads in the second half of the 6th century. The famous temples at Selinus, Faestum and Sir±s , to name but three, were b u i l t in honour of deities from the traditional Greek pantheon, testifying to the 22 perpetuation of Hellenic religion i n the vrest. It seems, therefore, that i n the west in the 6th century there was economic expansion that could have promoted the development of an independent western culture, but, perhaps contrary to expectation, we discover that there i s a distinct tendency to turn to and preserve the Hellenic cultural heritage. Evidence of the importation of the Homeric and other epics into the west i s slight; we must simply assume that as part and parcel of the way of l i f e in mainland Greece of the 8th and following centuries the ubiquitous epics were brought to thennew land by the colonists who continued to transmit them as verbal embodiments of the excellence of panhellenic culture. Although we have no precise record of the travels of the anonymous reciters of the traditional epics, there i s some proof in the western voyage of Arion around the l a s t quarter of the 7th century. Herodotus recounts the story of this poet;, who, having arrived at the court of Periander of Corinth from his native Lesbos, 23 later travelled to Italy and S i c i l y , where he amassed a great fortune. Herodotus' interest in the tale l i e s i n the miraculous incident i n which Arion was aided by a dolphin i n escaping the murderous hands of fchescrew on his return to Corinth. The tale does, however, substantiate our assumption that bards could and did s a i l from Greece to the colonies 6 in the west. Arion appears to have travelled with the express purpose of improving his financial position, carrying with him on the voyage his lyre and professional garb. Another travelling poet was Xenocritus of Locri, who found his way to the mainland and became noted as one of a 24 group:-) of innovative musicians active i n Sparta. According to the 25 Parian Marble, Sappho spent time in exile i n S i c i l y . Ibycus l e f t the west to take up residence in Samos; Pythagoras l e f t Samos to take up residence in the west. Thus, i t was not uncommon for renowned poets or philosophers to travel abroad. The constant sea-traffic between Greece and the colonies made possible the importation of the singers of epic, so that i n the colonies as in Greece i t s e l f , there were, one imagines, professional entertainers either residing in one place or else following a c i r c u i t of performance around the major centres. The lines of Antipater's poem quoted above reflect the general verdict of the^literary c r i t i c s of antiquity upon Stesichorus. The poet was recognised as an imitator of Homer. The publication of fragments of Stesichorus' poems discovered among the papyri from Oxyrhynchus, containing pieces identified as belonging to the Geryoneis, Iliou Persis, Eriphyle, Suotherae and Nostoi, now provides us with far more concrete material than was available to scholars before the 1950's 26 for a detailed study of the poet's diction and style. In this dissertation, therefore, I propose to examine the position of Stesichorus as imitator of the epic tradition and as innovator, bearing in mind the environment in which he composed, namely the times of prosperity i n the west that gave rise to a flourishing interest in the arts. We have indicated that there was, on the one hand, a somewhat conservative tendency encouraging the preservation of the cultural inheritance of 7 the motherland; presumably Stesichorus in his youth was exposed to and influenced by the recitation of the traditional epics. On the other hand, by the 6th century there may also have been some opposition to more conservative attitudes i n poetry. Arion, who had gained a reputation for certain innovations i n l y r i c compositions, made an extremely successful tour of the western provinces. The connection between Xenocritus, known as a member of an innovative school of music at Sparta, and Locri, where there appears to have been some sort 27 of "school" of poetry , again hints at the importation of new ideas i n composition into the west. Thus, i n the examination of the fragments of Stesichorus' poems I shall consider the extent to which the poet in his adaptation of epic material adhered to the precedents of the epic tradition in his use of language and myth derived from that tradition and the extent to which he endeavoured to revitalise an art-form that may have begun to stagnate through the constant repetition of the same material. The poet's choice of a new musical medium was i n i t s e l f a decisive movement away from the intoned delivery of the traditional epic poems. Unfortunately, nothing has survived of the musical accompaniments to the >,poems of Stesichorus apart from comments such as that i n (piutarchj, 28 de Musica 7, about his use of the Harmatian nome , so that there i s l i t t l e to be discussed on the subject of the poet's innovations in musical accompaniment. The poet's innovations in metrical schemes .and the movement away from the purely dactylic measures towards the sophisticated dactylo-epitrite schemes that are to be found in-Pindar's 29 Odes are the subject of an exhaustive study by M. Haslam* . There has not as yet, however, been published a comprehensive study of the poet's 8 treatment of language and myth in relation to the epic tradition. In the f i r s t part of this dissertation, therefore, I shall consider the poet's diction in respect to his "dialect" (chapter II) and his use of "formulaic" expression (chapters III,IV,V) whether imitated from known Homeric phrases, modified from them, or totally alien to the epic convention. By way of a conclusion to the discussion of the poet's use of language, I shall devote a chapter to a detailed examination of four fragments that compares and contrasts their diction and structure with parallel passages in the epic corpus (chapter VI). In the second part of the dissertation I shall consider the poet's adaptation of the traditional myths giving attention to some of the less well-documented poems (chapter VII) and discussing i n detail his treatment of the hero Heracles (chapter VIII), of Helen (chapter IX) and of the legend of the Sack of Troy (chapter X). I shall be concerned in particular to note where the poet i s consciously avoiding the repetition of epic material and to suggest what might be the possible motivation for his innovations. Footnotes to chapter I. 1 The Suda's information that the poet was originally called Teisias, whether or not correct, i s irrelevant for the study of the poet's work since he was known .to antiquity as Stesichorus, e.g. by Plato, Pausanias, Cicero, Quintilian et . a l . Of his predecessors i n the west we know the names of figures such as Xanthiis and Xenocritus, but of their works nothing has survived. Cf. A. Lesky, A History of Greek Literature (translated by J . Wil l i s and C. de Heer,London, 1966) p. 151 and CM. Bowra, GLP p. 82 f f . G. Vallet, Rhegion et Zancle, (Paris, 1958) p. 312, suggests the possibility of a Locrian "school" of poetry, which is not inconceivable, but lacks the support of external evidence. 2 Quintilian, Inst. Orat. X 1 62; cf. Longinus, De. Subl. 13 3, Dionysius of Halicarnassus, Cens. Vet. 2 7, Horace, Odes IV 9 8, a l l alluding to the epic content of Stejsichorus' poems. 9 3 On Terpander's musical setting for the verses of Homer, see §lutarchj, De Mus. 7. *» 1132 c. The citation of Chamaeleon's remarks on the "singing" of Homer , Hesiod, etc., made by Athenaeus XIV 620 c unfortunately does not name the "singer"> Chamaeleon was writing a work On Stesichorus, but this does not exclude, the possibility of his citing Terpander. For a discussion of the nomes see M.L. West,"Stesichorus," C£ 21 (1971) pp. 309-311 ..and T.J. Fleming,"The musical nomoi in Aeschylus' Oresteia," CJ 72 (1976) p. 222f. The Harmateion appears to have been adapted from an auletic acccmp&himent to a l y r i c a l one. The Tpuot mentioned by the Suda i s borne out by the evidence of the papyri. A triadic structure of strophe/antistrophe and epode may be seen in the Geryoneis and the Iliou Persia. There aiready existed a system of strophic responsion before Stesichorus' time, as is witnessed by Alcman's Partheneion, but the Suda may well be correct in" ascribing to Stesichorus the invention of a triadic metrical structure. 4 On the authenticity of the Rhadine etc., there i s l i t t l e evidence to support either case. Page, PMG p. 137 endorses the views of H.J. Rose, i n "Stesichorus and the Rhadine fragment," CQ_ 26(1932)pp.82-89, in which he argues that the fragment in Strabo VIII 3 20 can be no earlier than the Alexandrian era. Vallet, op.cit., p. 284ff., sees in these fragments (=»PMG 277, 278, 279) traces of popular S i c i l i a n tales. 5 West, a r t . c i t . , p. 302. 2 6 For example, Bowra, GLP p. 74, or H.Smyth, Greek Melic Poetry, (New York, 1963) p. 255. Many, such as Rose, i n his Handbook of Greek  Literature (London, 1965)p. 108, qualify the magic numbers 632-556 with an "approximately". 7 On Stesichorus, son of Hesiod, i n Aristotle f r . 565 and Philochorus 328 F 213, see West, a r t . c i t . , pp. 304-305. Some scholars have associated the tradition of Hesiod's being the father of Stesichorus with the Pindaric (?) epigram quoted by the Suda, under TO *Haco6eC*ov YnpoiS, that refers to Hesiod's longevity and rejuvenation. For a f u l l discussion of this see M Kay, "Hesiod's rejuvenation," CQ^  9 (1959) pp. 1-5. According to Thucydides, H i s t . I l l 96, there was a Locrian tradition that Hesiod died there, which West takes together with other allusions to r e f l e c t Stesichorus' special association with Italian Locri (art.cit., pp. 304^305). The father-son relationship between the two poets seems to betray signs of popular legend and the tendencies of later chronographers et al_. to make convenient synchronisations of the lives of poets. Similarly, Tzetzes* declaration (Vit. Hes. 18) that Stesichorus was a contemporary of Pythagoras looks lik e a useful synchronism of eminent figures i n the Greek wes.t, but may i n fact be closer to the truth than is generally believed. 8 See F. Jacoby, Apollodors Chronik (Berlin, 1902) pp. 196-200. 9 It i s recorded on the Parian Marble that Stesichorus arrived i n Athens in the year 485/4 B.C., the year of Aeschylus' f i r s t victory and of Euripides' birth, when Philocrates was Archon.(Note the rather patent synchronisation of important events concerning literary figures.) 10 A plausible suggestion for the later date for Stesichorean chronology is proposed by W. Ferrari, in "Stesicoro Imerese e Stesicoro Locrese," Athenaeum;15( 1937) p. 235 f f . , where he points out that the dichotomies in chronology and place of origin may have resulted from a double tradition of Stesichorus' dates, rather than from there being two poets of the same name , as was argued by Wilamowitz i n Sappho und Simonides (Berlin, 1913)pp. 233,234. 10 Pliny, NjH.II 54 and Plutarch de Fac. in Orbe Lun.19 (=PMG 271) record that Stesichorus was profoundly affected by a total eclipse of the sun which may be identified as that of 19th May, 557 B.C.; cf. West, a r t . c i t . , p. 306. The connection with the tyrant Phalaris i s mentioned i n an anecdote repeated by Aristotle, Rhet.II 1393 b. Phalaris' dates may be placed ca.570/65 and 554/49 B.C. (OCD). On the problems of S i c i l i a n chronography, see note 12 below. It has been argued by some scholars (e.g. F. S i s t i , "Le due Palinodie d i Stesicoro," Studi Urbinati 39 (1965) p. 313 and A.J. Podlecki, "Stesichorea," Athenaeum 49 (1971) p. 316 f.) that the legend appearing in Pausanias III 19 11 and Conon Narr.xviii i n which Stesichorus i s informed of the cause of his blindness by one of the generals of the Crotonians after their defeat by the Locrians may refer to the conflict between these two states known as the battle; of the river Sagra (Strabo V i 261» cf.Justin, Epitome XX 2 3 ff.) This battle i s thought to have occurred between 560 and 540 B.C. : T. Dunbabin, i n The Western Gr,eeks (Oxford, 1948)p. 359 f f . , although P. Bicknell believes i n an earlier date, "The Date of the Battle of the Sagra River," Phoenix 20 (1966) pp. 294-301. 11 PMG 564: 0*5 6oupl T t a V T O S vCnaoe v £ o u s » 6 u v d e v T c i B a X & v "Avaupov ' ' O u e p i t o X u B ^ d T p u o s e £ ' I u i X x o y . OOTOJ yhp "Oynpos i f a e EtaaJxopos S e t a e X a o u s * These lines give a^terminus antebuem, although no real indication as to how much before Simonides Stesichorus lived. The association of Stesichorus with Homer need not mean that Simonides thought he was of the same vintage as Homer. 12 For example;-, Antiochus applied a 36-year generation-count, while Ephonis one of 39 years, i n their calculations of the foundation-date of Syracuse which had been set as 7 generations of Gamoroi before Gel on. Cf. R. van Compemolle, Etudes de chronologie et d'histoxio- graphie S i c i l i o t e s (Bruxelles, 1959) p. 59 f f . and J. Be'rard, La  colonisation grecque (Paris, 1941)p.285."ff.. 13 The Suda gives Himera as the most l i k e l y place of origin, quoting other suggestions^. Pallantion in Arcadia. Stephanus of Byzantion, under McfxaOpos, says that this was a Locrian colony i n Sicily(?) and call s Stesichorus MneToiiptvos. 14 The sources who specify that Stesichorus came from Himera are as follows: Plato, Phaedrus 243 a; Athenaeus XII 512 f.; Pausanias II 22 6, Ix 11 1; Aelian VH 10 18; Cicero In Ver.II 2 34; {Plutarch de_ Mus. 7; the Suda under Stesichorus . Pollux Ix 100 claims that Stesichorus* tomb was a t Himera. A r i s t o t l e , Conon ( as i n note 10 above) and Himerius (Ori29 3) mention Stesichorus as being responsible f o r g i v i n g the people of Himera advice against tyranny, although the connection does not n e c e s s a r i l y imply that Stesichorus came from Himera. F i n a l l y there i s an i n s c r i p t i o n oh a fragment of a Herm found at Tibur (I.G. 14 1231) that reads "Stesichorus, son of Euclides, of Himera." 15 west, a r t . c i t . , pp. 302-304. 16 See, f o r example, A.R.Burn, The L y r i c Age of Greece (London, 1967) p. 143 f f . ; Dunbabin, o p . c i t . , chapters I I , I I I , X ; V a l l e t , o p . c i t . , p. 139 f f . . 17 See Herodotus H i s t . IV 152; c f . Berard, o p . c i t . , p. 303ff. and .. V a l l e t , o p . c i t . , p. 87 f f . , 18 See Burn o p . c i t . , p. 151 and M.I.Finley, Ancient S i c i l y (London, 1968) p. 29 f f . , On the splendid Doric temples of the 6th-century, see f o r example, G. Richter, A Handbook of Greek Ar t (London,1969) p. 28 f f . Imported pottery was mostly Corinthian u n t i l the beginning of the 6th century, when the Athenian work-shops began t o dominate western markets; see M. Robertson, i n "The V i s u a l A r t s i n Greece," i n T.he Greek World, ed. H. Lloyd -Jones (Harmondsworth, 1965)p. 198 f . and f o r a d e t a i l e d account of the importations i n t o Rhegion and Zancle see V a l l e t , o p . c i t . , p. 140ff. . On coinage, see F i n l e y , o p . c i t . , p. 35 'ff... . 19r On the p r o v i n c i a l i s m of western a r t , see, f o r example, Robertson, o p . c i t . , p. 196 arid 199, Burn, o p . c i t . , p. 151. On l o c a l pottery workrshops, see V a l l e t , o p . c i t . , p. 210 f f . . 20 See A.J.Graham, Colony and Mother-city i n Ancient Greece (Manchester, 1964) pp. 25 and 30; on Syracuse*:! and Corinth, see p. 143'iff • 21 On Milon, see Diodorus IV 24 7 and Strab© VI 1 12. On the Crotonians, see Strabo VI 1 12. Cf. H.A.Harris, Greek Athletes and  A t h l e t i c s (London, 1964) p. 110 f f . . 22 F i n l e y , o p . c i t . , p.27. 23 Herodotus, H i s t . I 24. 24 (J>lutarchJ, de Mus.9 (=1134 b) . 25 Cf. D.L.Page, Sappho and Alcaeus (Oxford, 1955) p.223 f f . . 26 The new fragments appear i n the following e d i t i o n s of the Oxyrhynchus Papyri, published by the Egypt Exploration Society i n London edited by E. Lobel: F r . 2359(Suotherae) and f r . 2360 (Nostoi) i n P. Oxy. v. X X I I I ( L o n d o n , 1956)pp. 11-18. F r . 2617 (Geryoneis), f r . 2618 (Eriphyle) and f r . 2619(Iliou,Persis) i n P.Oxy. v. XXXII (London, 1967)pp. 1-55. Fr. 2803 (Illou Persis) in P.Oxy.v. XXXVII(London,1971)pp. 3-11. P.Oxy. 2359 and 2360 appeared before the publication of Page's PMG in 1962 and are therefore included in that collection of the fragments of Stesichorus. The later fragments are collected in Page's SLG (Oxford, 1974)pp. 5-43. In SLG Page assigns P.Oxy. 2735 (published i n P. Oxy. v. XXXV (London, 1968) . .9-32)_to Ibycus, whereas West, "Stesichorus redivivus," ZPE 4 (1969) knd R. Fuhrer, "Zum 'Stesichorus redivivus'," ZPE 5 (1970)p. 15 f. believe that the fragment i s to be assigned to Stesichorus^ as was suggested by Lobel in the editio princepSy possibly deriving from the. Helen; On account of the uncertainty of authorship I have not incorporated examples from P.Oxy. 2735 in my'examination of the language of Stesichorus. 27 Cf. Vallet, op.cit., p. 312. 28 See page 9, note 3. 29 M.W.Haslam, "Stesichorean Metre," QUCC 17 (1974) pp. 5-57. Chapter II Stesichorus' "mixed" dialect. The new papyri of poems by Stesichorus are naturally disappointing in their fragmentary nature. One always hopes for the discovery of a complete poem so that extensive study of a l l aspects of language and structure may be pursued. As i t i s , the linear incomplete-ness of the fragments necessarily curtails the discussion of the poet's language, restricting in particular discussion of those elements of syntax that l i e beyond the structuring of phrases. Moreover, since the linguistic evidence derived from the works of a single poet cannot be considered in a vacuum, but rather must be considered in comparison and contrast with other literary works from the period of composition, a farther limitation i s imposed through the lack of evidence for much of the contemporary or near-contemporary poets who may have influenced or have been influenced by, Stesichorus. As evidence for possible pre-r: cedents from oral epic we have the Ili a d and the Odyssey, together with a f a i r representation of Hesiod's works, but from the works of most other epic poets, continental or otherwise, we possess a mere handful of citations. Early elegiac and l y r i c poetry, although hardly voluminous in i t s state of preservation, i s our only evidence for the linguistic or s t y l i s t i c a f f i l i a t i o n s of the poets of the Archaic period. L i t t l e before the Odes of Pindar can truly satisfy any scholar investigating the techniques of the early l y r i c poets, and hence any examination of the evidence that we do posses must remain open to further elucidation or contradiction upon the discovery of new evidence fromAscroe?iE^9"0^^en hoard of papyri lurking beneath the sands of Egypt. Since the diction of the poets of the Archaic period does not belong to one invariable form of the Greek tongue, but rather to one or more than one of the local dialects of the Greek-speaking aseas of the Mediterranean, any discussion of the language or diction of a poet w i l l revolve around the dialectal features found within his poetry, whatever the limitations imposed by insufficient evidence, the study of the dialects of pre-classical literary figures is justi f i e d because of the use of dialect in relation to the particular "genre" favoured by the poet in question. The study of Greek dialects has concentrated primarily on the evidence of inscriptions, whilst the equally complex problems of literary dialect have been somewhat neglected. The new papyri furnish those interested i n the development of the so-called literary languages with evidence of a more certain nature than that which i s derived from the citations of such later authors as Athenaeus, whose native Attic or Koine has obscured many of the linguistic phenomena belonging to the non-Attic dialects. We ask ourselves What is "dialect". Inscriptional evidence from the Mycenaean tablets onwards provides us with a relatively limited picture of the dialectal variation in the Greek tongue as spoken by the isolated communities of Greece from the 2nd millenium B.C., revealing a few tantalising facts regarding their temporal and spatial relationships through linguistic a f f i l i a t i o n s . The distribution of the dialects i s a complex problem and can only be adequately understood .. : through elaborate systems such as the factorial analysis suggested by Coleman^". The ancients,on the other hand,believed in a highly schematic, t r i p a r t i t e division of dialect into Ionic, Aeolic and Doric,, a division that derived from some misconceptions of t r i b a l division in the prehistoric period and from assumptions based on the approximate divisions of dialectal a f f i l i a t i o n of literary figures in the class i c a l period and later. . I t i s with caution therefore thafc,we 15 must approach statements such as that of the Suda, where i t is asserted 3 that Stesichorus wrote in the Doric dialect , and yet ultimately in the study of' literary dialect in the Archaic period we discover that i t is s t i l l to some extents useful to rely on a schematic dialectal division. Recently i n the study of early epic i t has been suggested that the language of the Homeric epics, whose present form evolved after the migrations to the eastern sea-board of the Aegean,be considered as belonging to a distinct dialect group designated as the Southern dialect group, in contrast to the Northern dia&ect group which encompasses 4 both the continental epics and the dialects known as Aeolic . A third dialect group, the Occidental or Western, only appears as a distinct entity i n the literature of the post-epic era. For Southern dialectal features of-the? 7i/6th centuries iweccan find testimony in the works of Archilochus and Mimnermus, whose diction,although indebted to the Homeric epics, also exhibits features of their native Ionian dialect^. The Northern group is chiefly represented at the literary level by the language of Sappho and Alcaeus . Of the Western group, Alcman''s Laconian vernacular i s the principal^evidence . In his poetry in particular we can observe the admixture of elements from both Southern and Northern linguistic sources with elements of his native dialect, within the structure of a single work. I propose,therefore, to examine the dialectal features of the language of Stesichorus in the f i r s t instance in relation to the features of the Western group, since this corresponds approximately with the "Doric"dialect that was mentioned by the Suda and since one would expect the appearance of certain Western linguistic features in an area that was primarily "Doric"-speaking. I shall then make a detailed examination of linguistic elements that Stesichorus shares in ccmmon with the language of epic, taking into consideration the question of the Northern-group features that have often been called "Doric" u n t i l recent studies have demonstrated their 8 possible origins in continental epic . There are two points to which attention must be drawn in preface to the discussion of dialectal features in literary works. F i r s t , local inscriptions give us evidence for features of a locaL dialect at a particular time. We must note,however, that poets do not necessarily r e s t r i c t their expression to their native dialect alone. i.n the case of choral l y r i c for example, one can sooner find linguistic a f f i l i a t i o n s between the diction of Pindar and Bacchylides than between the diction of the poet and the dialect of his mother-state. It i s , therefore important to distinguish between literary dialect and dialect as te s t i f i e d i n inscriptions. Secondly, the inscriptional evidence available to us dates primarily from the 5 t h century onwards, with the exception of the Linear-B tablets. Evidence from the Archaic period i s scanty, and yet i t i s for the poetry of this period that scholars draw upon inscription&l material for dialectal features. This evidence should therefore fee used with caution. A. Features from the papyri of Stesichorus that are found in the Occidental or Western group. If we compare certain phenomena occurring i n the poems of Stesichorus with the standard features of the Western group of dialects, as seen in the language of Alcman, we discover that the term "Doric" might be legitimately used of certain characteristic features of 9 Stesichorus' language. These features may be li s t e d as follows: i) Phonology a) Original long alpha i s preserved throughout. b) Short alpha occurs where the equivalent i s short epsilon in other dialect groups. c) The contraction of a * e to n , i n contrast to long alpha elsewhere. d) Dentals remain unchanged before iota, especially i n the 3rd person singular,of the present indicative of athematic verbs, and in the 3rd person plural of the present indicative of thematic verbs. The preservation of original long alpha was one of the principal distinctive features of so-called literary Doric i n classical times, presumably in pronunciation and therefore orthography, although i t i s metrically identical to the long eta of Ionic. The new fragments of Stesichorus' poetry contain over 60 examples of this characteristic feature, with only one exception: pnCri/vopa 2619 f r . 1 i 21. Even in citations of Stesichorus i n other authors the Atticisation of the long alpha amounts to less than 15%, which is the lowest percentage for a l l the choral l y r i c poets 1^. It is without doubt, therefore, that antiquity recognised the long alpha as a distinctive t r a i t i n the language of Stesichorus, and accordingly where long alpha does not occur, the reason may be attributed to an error in transmission. Short alpha i s witnessed i n uapav 2359 f r . 1 i i 6 and Qopav 2803 f r . 11 6, ,:.£his form being the equivalent of Cepog , and i n "Aptayts 2619 f r . 18 11 for '^ptepts . Other possible instances, such as axepos^pr Tpanu.jhave not turned up i n the papyri as yet. In the quotation from the G&ryoneis cited by Athenaeus, the MS records: diptxnd' Cepas TIOTL Bevdea VUXTOE epepvas PMG 185 3. One may postulate an original uxpSc., which has not been preserved in the transmission, despite the "Doric" alpha of the genitive singular of epepvas. Another instance of short alpha may be seen in 6'xot 2617 f r . 4 i i 1 5 , corresponding to ^ T e« The alpha caused the original labio-velar to become a palatal, thus producing the suffix -Ha i n the western group of dialects, whereas the alternative form with epsilon in the southern group developed into dental -T,C. In the fragments of Stesichorus we find two examples of contraction of o + e to n : icotctu^n PMG 264 and 3$Vt.V*i. "2803 f r . l l 1 . Otherwise Stesichorus would appear to follow the epic convention of not contracting vowels of unequal length: hence E C X a x u 6 a o 2 3 5 9 f r . 1 i 9 , 'AXxudov 2 6 1 8 f r . 1 i 3 and 'AutpwEpaos PMG 179 (b). It is no doubt significant, however, that the examples of non-contraction occur only in proper names, derived from epic sources. In the f i n a l category of Occidental features l i s t e d above, the retention of the original termination -TU occurs almost certainly in e x o v x b 2617 f r . 6 4 and possibly i n 3\»TV 2619 f r . 1 i 6 , whereas the citations present us with the atticised form Tt-dnai, PMG 22 3 4 . Parallel to the preservation of the original termination in'TU i s the retention of the dental in the second person singular;of the^, personal pronoun: TU as opposed to co . Of this feature we find a solitary TUV in 2617 f r . 11 5 . In view of the lack of further evidence we assume, but cannot assert,that the dental form of the personal pronoun was normal in Stesichorus. The characteristics considered i n the preceding paragraphs are those elements of phonology noted as distinctively Doric by Thumb If, however, we examine more recently compiled tables of dialectal 12 characteristics, such as those which Risch, Coleman or P£vese present , then we discover that the long alpha occurs also in the Northern group, namely in the poetry of Sappho and Alcaeus, as well as i n the inscriptions f r o m T h e s s a l y a n d B o e o t i a . T h e s h o r t a l p h a a l t e r n a t i n g w i t h e p s i l o n a l s o o c c u r s i n t h e L e s b i a n p o e t s , w h i l e t h e c o n t r a c t i o n o f J a + e t o n a n d t h e r e t e n t i o n o f - x c b o t h s u r v i v e i n t h e l a t e r B o e o t i a n i n s c r i p t i o n s a n d m a y t h e r e f o r e b e p o s t u l a t e d a s c h a r a c t e r i s t i c o f t h e N o r t h e r n g r o u p a s > ; w e l l a s o f t h e w e s t e r n g r o u p . O n e m u s t p a u s e , t h e r e f o r e , t o c o n s i d e r t h e N o r t h e r n g r o u p o f d i a l e c t s a s p o s s i b l e s o u r c e f o r t h e a b o v e - m e n t i o n e d p h o n o l o g i c a l f e a t u r e s , b e f o r e a s s u m i n g t h a t t h e i r a f f i l i a t i o n s a r e s o l e l y w i t h t h e W e s t e r n g r o u p . ? ^ i i ) M o r p h o l o g y ^ a) } c - s t e m n o u n s r e t a i n t h e i o t a t h r o u g h o u t t h e i r d e c l e n s i o n . b ) N o u n s i n -ev>s h a v e t h e i r g e n i t i v e s i n g u l a r i n - e o c . c ) P r o n o u n s : TU= a u , d y e s = n y e u s d ) T e r m i n a t i o n o f t h e 1 s t p e r s o n p l u r a l o f t h e a c t i v e v o i c e i s - y e s . e ) T h e s o - c a l l e d D o r i c f u t u r e : i t p a S e w . . f ) F u t u r e p a s s i v e w i t h a c t i v e e n d i n g s . g ) V e r b s i n h a v e g u t t u r a l - s t e m c h a r a c t e r i s t i c s . h ) A t h e m a t i c i n f i n i t i v e i n - y e v . O f t h e s e c h a r a c t e r i s t i c s w e h a v e n o e v i d e n c e f o r c a t e g o r i e s a ) ) , d ) , f ) a n d g ) . I n t h e c a s e o f b ) , t h e e p i c f o r m B a c u X r j o s 2 6 1 9 f r . 1 4 x6 i s p r e f e r r e d t o t h e O c c i d e n t a l g e n i t i v e i n - e o j . I n t h e c a s e o f e ) , t h e e x a m p l e o f epu£u) 2 3 6 0 f r . 1 i 1 0 w o u l d a p p e a r t o c o n f i r m t h e c o m m o n G r e e k f o r m a t i o n o f t h e f u t u r e a s o p p o s e d t o t h e s o - c a l l e d D o r i c f u t u r e i n -5e<ji>.: I n t h e c a s e o f h ) , f o r t h e t e r m i n a t i o n o f t h e a t h e m a t i c i n f i n i t i v e S t e s i c h o r u s s h o w s t h e u n p a r a l l e l e d f o r m s tTv 2 6 1 7 f r . 4 i 7 a n d e C^etV;t ' 2 6 i ; 9 f r . l 3 5 f o r t h e v e r b " t o b e " " ! * ^ . We l a c k e v i d e n c e f o r the in f i n i t i v e s of other athematic verbs. It appears to be only in the case of c) ,2 pronouns, that we find the western or "Doric" formations in the morphology of Stesichorus' language: T U V 2617 f r . 1 i 21 and autv 2360 f r . l i 3. Thus, at the level of morphology, Stesichorus* dialectal a f f i l i a t i o n s with the Occidental group are rather less obvious than at the level of phonology. As we shall see later, Stesichorus shows a greater a f f i n i t y in this area with the language of epic. Unfortunately, an example of the distinctive termination of the f i r s t person plural of the active voice is not found in the fragments. As a rule, later choral l y r i c does not employ the Occidental -yes, but follows the epic and 15 common Greek -yev . Alcman alone gives evidence for the use of -yes in a literary work, for example nctpnaoyes in f r . l 12, but i t is possible that Stesichorus, who like Alcman prefers the Occidental .:. foscmation of the personal pronoun, also used -yes, although the later choral l y r i c poets.do inot. One wonders i f Stesichorus might have been more inclined to employ Western forms in passages of direct speech than in narrative ones on account of his presumed background of spoken "Doric". Of the "Doric" features exemplified by the papyrus text of Alcman, several are notable by their absence in Stesichorus, such as the nominative plural of the a r t i c l e , which in the Occidental group of dialects retains the tau:TOL and xau . Although there are two instances of the spurious diphthong 0 being represented by t o , as in Alcman-, namely (ipovo'dev 2360 f r . l i 3 and Y^vasoyai, 2617 f r . l l 4, the accusative plural of O-stem nouns and of the a r t i c l e i s consistently represented i n the form - 0 0 s . The text of Alcman,on the other hand, { contains forms in omega with complete regularity i n the genitive singular of feminine nouns in -u>, i n the accusative plural of O-stem nouns, in the accusative plural of O-stem nouns, i n cases of metrical lengthening, and i n lexical instances such as M w a a (Stimes) and &pav6$ (twice) Similarly i n Alcman there is consistent appearance of n for spurious diphthong ,e ( e l i ) , for example in the form of the thematic i n f i n i t i v e pa<y6cfvny f r . l 88. Iii the papyri of Stesichorus, however, there i s one such example,(ptiynv 2617 fr.7 i 2, but this i s by no means the norm There are several lexica l items, considered to be of"Doric" or Occidental origin, which i t w i l l be convenient to mention here. a) The conjunction at is a feature of the Western group of dialects, but i n fact also occurs i n the Northern group, in for example Sappho and Alcaeus, as well as in the later inscriptions of the Thessalian and Boeotian areas^. b) Although the partidlel x a does not i t s e l f occur in Stesichomis, the forms i t o x a 2617 fr.42 (b) 4 and o x a 2617 fr.4 i i i 15, suggest that Xo as opposed to x e or dfv might appear. Apart from i t s belonging to the Western group of dialects, - , x a occurs in the Boeotian branch of the 19 Northern group, but hot however m the Lesbian poets,; . c) In both alcman and Stesichorus instances of the "Doric" e y w v with nu occur: even before a consonant: hythv 6*a\!> 2619 fr.13 3 and hybi\> AEYU> 2619 fr.16 8. d) OIUTEU , the "Doric" variant for ctutod occurs in 2619 fr.47 10. Apart from the example of Alcman f r . l 79, the only evidence for this form is to be found i n traditionally ascribed "Doric" inscriptions. e) The,preposition ueSdf occurs in the papyri of Stesichorus at 2619 fr.21 3, while yeTd* appears in the citation PMG 210. The two forms are lin g u i s t i c a l l y unrelated, ne6d replacing the originayyexct in an odd assortment of dialects, as far as the inscriptional evidence goes: Lesbian, Boeotian, Arcadian, Argolic, Cretan and Theran. Some of the Western group of dialects preserve yexcf, for example Corinthian, but in literary works ite6di is well established i n both Western and Northern 20 groups, occurring i n Airman, Sappho and Alcaeus f) The Occidental form itoxt appears to have been preferred to the alternative upds , occurring eleven times as opposed to two. The instances of the latter in 2619 f r , l i i 6 and 2617 fr.4 13 may be explained by the influence of the appearance of both forms in epic. In this case the Lesbian poets used i t p t S s , parallel to the assibilisation of ?Tb to -au, whereas the later inscriptional evidence of the other two major branches of the Northern group give examples of i toxC p a r a l l e l to 21 the usage in the Western groups. . At the level of l e x i c o n in the language of Stesichorus one would perhaps expect some evidence of local influence and the '. - ,, incorporation of vernacular words into poems, as in the case of Alcman. Since, however, Stesichorus' material-Is derived from a more universal tradition, and since his compositions were intended for a more universal audience, thevConstSnt use of vernacular would not be appropriate, although i t s occasional use could have been turned to the poet's advantage in terms of innovation. In any case, the evidence of the papyri does not indicate that Stesichorus inclined towards the embellishment of his poetry with elements of S i c i l i a n vernacular. As far as the so-called Doric accent in concerned , i t should be noted that the accent marks applied to the texts of Stesichorus by Alexandrian scribes were based on assumptions that the language of the poems was comparable to the "Doric" of later times with which the scribes were acquainted. There are no means b y which one may determine i f the peculiar characteristics of the "Doric" pronunciation and the "Doric" intonation reflected by these accent marks are i n fact characteristic of the language of choral l y r i c of the early Archaic period, particularly since the apparent rules of accentuation do not 22 seem to correspond to the apparent metrical phenomena B. Features from the papyri of Stesichorus that are shared with epic features. Cr i t i c s both ancient and modern have commented upon the 23 "Homeric" nature-- of Stesichorus* poetry . I shall attempt, therefore, in this section to assess how far this "Homericness" l i e s within the phonological and morphological structure of Stesichorus' language, and how far this comment i s merely a reflection on the poef's style, developed from the appropriation of words and phrases from the epic poems. The language of the Homeric poems has long been recognised as a composite one, manifesting elements that are a f f i l i a t e d to more than 24. one dialect-group . In terms of modern dialect-geography of the GKeek language, these elements belong primarily to the Southern group of dialects, the traditional Attic-Ionic and Arcado-Cypriot groups, with a secondary admixture from the! Northern group. The fusion of features from different dialectal origins may have taken place at such an aarly period that, although the resultant "language" was not actually spoken i n any one area, i t would have been understood by members of different dialectal areas. The hypothesis of f a i r l y wide comprehensibility can be more easily accepted i f indeed l inguists are correct i n their asBumptions that there was less divergence between the various dialects 25 i n the prehistoric period I shal l begin by l i s t i ng the dist inct ive features of the 26 language of the Homeric poems, as suggested by Palmer , and then proceed to consider those aspects of Stesichorus' language: .recoghisably derived from epic. i) Southern gafoup Characteristics of Attic-Ionic a) Original long oivfco n. b) Movable nu. c) Prepositions not apocopised. d) Athematic in f in i t i ve with termination -vac . e) Secondary ending of the 3rd person plural = -acxv. f) Potential part ic le is 5v. Dist inct ively Ionic characteristics a) Complete change of or ig inal long o to i). b) Absence of contraction : -ea, -eo, -eto. c) ;:Treatment of M with compensatory lengthening : Servos • d) Gen.sing, of masc. A-stem nouns i n -eco. e) Gen. p lur . of masc. A-stem nouns i n -euiv. f) Analogical genitives: 3aoCAeo£ for BaauAe'ios g) fnv for At t ic eav, av. i i ) Northern group a) Labio-velars become labia ls before front vowels: Tturupes . b) Doubling of consonants instead of compensatory lengthening of vowels: auy e c) Patronymic adjective instead of genitive case: TEACIUISVCOS. d) ux for yu*. e) Dative plural of the athematic declension in -eaot. f) Double sigma ,aa from -xu§;»^§ir,-6t-. g) Athematic in f in i t i ve terminations: -yevctt, -yev. h) Potential part ic le is xe. i) -yu inf lect ion of contracting verbs: (puXnyb. j) Perfect part ic ip le with present par t i c ip ia l endings: eXqXtfdiov. Of the six points in the f i r s t category, Attic-Ionic, a) i s contradicted by Stesichorus' use of or iginal long alpha, b) and c) are observed in Stesichorus,movable nu appearing mostlylin the dative p lu ra l . There is no evidence for points d) and e) , while i n view of the occurrence of forms such as 5xa or noxa i t i s m o s t unlikely that we should f ind 5,v rather than xa. Of the seven points in the category of spec i f ica l ly Ionic features only b) is c lear ly in evidencef the absence of confcraction may be observed in xeCxeos2803 f r .5 7, in ito&eio 2619 fr.16 12 and in oorea 2617 fr .4 i i 8. In the case of c ) , two examples from the;-papyri contain long vowels betraying the loss of internal digamma: Tapudva yuivofcoyac 2617 f r . l l -2, with,however, the non-epic omega representing the secondary long vowel, and oXeadvo'Pos atoXodetpou 2617 f r .4 22. Less certain are two further examples from the c i tat ions: xoupCSuxv x' aXoxov PMG 185 A">, and Acos xoupa( gaauXeuauy) PMG 200 2. In both "these examples close paral le ls from epic may be responsible for the -lengthened form, part icular ly in the case of xoupi?6Lav, where a short vowel would be metrically unsuitable in dactyl ic verse. There are two possible occurrences of the lack of compensation in ydvag( ;PMG 223 2 and in xopotts PMG 223 3, which show that there is already the 26 tendency to neglect the original digamma which became prevalent in later choral l y r i c . Points d), f) and g) are contradicted by Western or other non-Ionic forms in Stesichorus, and although there is no evidence for e) , i t i s highly unlikely that this Ionic form wouM occur. In the category of Northern features, we find parallels for two of the ten characteristic points. There is one instance of the use of the patronymic adjective, namely ' O 6 \ 3 a e u o v 2360 f r . l i 2, with another possible example in Exayav6puov 2619 fr.27 4. The patronymic in - u 6 n s is also found in'Yrcepiovi6ctSPMG 185 1. W.8 find two examples of the dative plural in -eaau: yaxapeaau deotau 2617 f r . 15 1 and in -ecrao npypjL 2619 fr.15 (b) 13. The former i s obviously reminiscent of the epic formula which occurs six times i n the dative plural in the Homeric corpus. There are, however, five examples of a normal dative in -cru in Stesichorus, in positions Where the form in -eaau would have 27 \ \ been possible metrically : xnpotv 2617 fr.13 1; itepu Bouauv eyats 2617 fr.13 27i^}pa*ctOLV 2619 f r . l i 13; and from the citations noat? PMG 185 6 and uSat Qeous, PMG 223 2. The characteristic doubling of consonants instead of compensatory lengthening of the preceding vowel in the treatment of consonant+sibilant clusters appears to be confined i n the Homeric poems to forms of the personal pronouns. As we have seen, the personal pronouns in Stesichorus appear to follow the Occidental pattern, ahd the sole example of a double-consonantal treabaeht ©ccurs in the 28 anomalous case of xAesvvos' which M • shall discuss later . T h i s NiOrthern characteristic does not therefore generally appear in Stesichorus. T h e r e i s n o e v i d e n c e f o r p o i n t s a ) , d ) , g ) , h ) , a n d i ) . N u m b e r f ) s e e m s t o b e c o n t r a d i c t e d b y w o r d s s u c h a s oiCav . '2617 f r . 1 3 2 4 e a x t a e 2 6 1 7 f r . 4 i i 8 , x e p a a a s PMG 1 8 1 2 , a n d j ) b y o X u X o x e s 2 6 1 7 f r . 1 8 4 . F u r t h e r r e m a r k s o n a s p e c t s o f m o r p h o l o g y . A l t h o u g h w e c a n n o t p r e s e n t a c o m p l e t e o u t l i n e o f t h e d e c l e n s i o n s a n d c o n j u g a t i o n s f r o m t h e f r a g m e n t s o f S t e s i c h o r u s , t h e g e n e r a l l a c k o f c o n f o r m i t y w i t h t h e O c c i d e n t a l g r o u p o f d i a l e c t s i n d i c a t e s t h a t w h e r e t h e i n f l e c t i o n d o e s n o t b e l o n g t o o n e s & u x c e ' c o m m o n t o a l l G r e e k , w e m u s t l o o k t o t h e e p i c p o e m s f o r p r e c e d e n t s . i ) T h e m a t i c d e c l e n s i o n . 0 - s t e m n o u n s . I n t h e g e n i t i v e s i n g u l a r w e f i n d i n S t e s i c h o r u s i n s t a n c e s o f t h e p e c u l i a r l y e p i c f o r m - o t o , a s i n , A v a £ c f v 6 p o u o r 2 ' Q . 8 f r . l i i 8 , a s w e l l a s e p i c a n d l a t e r - o o . T h e W e s t e r n c o n t r a c t i o n i n - w f o u n d i n A l c m a n d o e s n o t o c c u r . F r o m t h e e v i d e n c e o f t h e f r a g m e n t s S t e s i c h o r u s s e e m s t o h a v e p r e f e r r e d t h e s h o r t e r f o r m i n - o u , w h i c h a c c o u n t s f o r a b o u t 7 0 % o f t h e g e n i t i v e s o f O - s t e m n o u n s a s c o m p a r e d w i t h t h e 2 9 . H o m e r i c u s e o f b o t h f o r m s w i t h e q u a l f r e q u e n c y O n e a s s u m e s t h a t t h e a c c u s a t i v e p l u r a l i n - o u s a l s o d e r i v e s f r o m e p i c p r e c e d e n t , s i n c e t h e r e i i s n o t , a t a n y r a t e , a t r a c e o f t h e O c c i d e n t a l -o)Sr o r o f t h e s h o r t f o r m o f t h e a c c u s a t i v e i n - o s » w h i c h o c c u r s i n H e s i o d a n d t h e H O m e r i c H y m n s . I n t h e c a s e o f t h e d a t i v e p l u r a l , e a r l y G r e e k i n g e n e r a l a d m i t s t h e u s e o f b o t h - o u a u a n d - o t s » a s s e m a n t i c a l l y e q u i v a l e n t 3 0 . I n e p i c o n e t e n d s t o f i n d - o u x t b e f o r e a c o n s o n a n t a n d t h e s h o r t e r -OL<S before a vowel, as i f i t represented an elision of-©tat. In Stesichorus the two examples of the longer form both bear the nu-ephelcusticon: Iv ueYcfpou0uv 2359 f r . l i 3 and 3 6 t x o u a u v 2617 fr.21 10. In the case of the A-stem nouns also, Stesichorus apparently prefers the longer form, with nu to avoid hiatus where necessary. Since the Occidental group of dialects seenfe to have employed the shorter form 31 alone ( at least as far as the later inscriptional evidence indicates) we must attribute the use of the longer form in Stesichorus to epic influence and to the greater convenience of that form in dactylic metres. A-stem nouns. In the genitive singular masculine , we find one example of the relatively rare epic form in -ao. E u X a t t S a o 6cn,'q>povos 2359 f r . l i 9. Since the epithet accompanying the proper name is also derived from 32 epic, the use of the epic genitive i s hardly surprising in i t s context . The Occidental equivalent, which contracts to "3, does not occur in Stesichorus"^. Stesichorus appears to have employed the dative plural - a u a u with more frequency than -otus. One suspects that -auou simply represents the epic - n t a i , with the characteristic long alpha for eta, rather than the possible formation from short alpha. There is no evidence of "Doric" -^aoi,. If we are correct i n supposing that in 2619 f r . l i 16-17, eiSpuoptot belongs to the common epic eop:\I6iit6i j Zed's/ then we may class this as an instance of borrowing from epic the use of the alternative form of the nominative singular of the A-stem "masculine noun in short alpha. The short alpha derives from the vocative singular, but in this phrase has been lengtfened by position. The phrase,commonly placed at the verse-end, must have become well established early in the development of the epic hexameter, since there i s no evidence of the original nominative form, Stesichorus presumably adopted the phrase directly from epic. We note, however, that the original confinement of the phrase to the verse-end is not maintained, and that the length of the alpha is l e f t ambiguous, although s t r i c t l y speaking i t i s no longer lengthened by position. i i ) Athematic declension. Nouns with consonantal stems of the athematic declension possess the same terminationsiiri a l l dialects, with the exception of the dative plural. The papyri fragments contain two instances of the so-called "Aeolic" dative plural in -eacru, which was employed in the Homeric hexameter to accommodate certain words to the metre, such as Muppt,6d*veao"c. The Homeric device would naturally be suitable for a poet composing in dactylic measures,whether or not hexameter, but as Iras mentioned above, there are more instances of the simple dative in -au, 34 which i s equally epic i n origin. . There are a few individual points which should be mentioned in this category of the athematic declension: a) In the S-stem nouns, such as xecxos, Stesichorus follows the epic pattern of not contracting the vowels placed in juxtaposition by the loss of intervocalic sigma: thus one finds xetxeosr.not TEUXOUS. b) For the declension of ndXus we have only the evidence of the accusative singular and hence no evidence of dialectal variants used by Stesichorus. c) 3aai>Xfjos in 2619 fr.14 6 demonstrates again the influence of epic verse. The eta reflects the original stem-ending -E"0&.?np, whose long vowel was preserved in the Northern group of dialects and in epic. In the Occidental group, however, the original long diphthong was shortened to eu ; hence $ctcrcXeK>S and $aabXe)-os . In the dative plural epic verse shows both long and short vowel versions: SaouXeuou and gacrtXTfeaau, the fonner being the precedent for flaauXeSai. PMG 200 2. d) In the case of vnuaiTv in PMG 192 2, one suspects later correction since in the matter of original alpha, i f nctfrexge else, one finds a consistency of form in the papyri. There i s , however, one possible 36 instance of "Doric" vaas supplemented in 2619 fr.33 3 £)) In 2619 f r . l i i 7, the nominative plural of the adjective TioXtis appears as noXe'es, no doubt derived from the athematic declension in epic, which is in fact found side by side with the thematic form. It seems to be the latter that i s preferred in the Western group of dialects, however, as in itoXXots.f in. Alcman. There are one or two instances of special case-endings that derive from epic, namely -Sev in tipavdOev, 2360 f r . 1 i 3, -6e as in -vo6* 2618 f r . l i i 5-6. No instances of -(pt, however, have come to light i n Stesichorus. The old instrumental dase is 37 particularly associated with epic, and is seldom found elsewhere i i i ) Conjugation. From the meagre evidence we possess of the verbal formations and terminations i n the fragments of Stesichorus, there are a few instances where we may point out epic influence, but the picture i s sadly incomplete. For example, we are lacking the type of personal endings for the f i r s t person plural of the active voice, which could have given us a useful indication of dialectal a f f i l i a t i o n . It appears that Stesichorus has imitated the epic convention of omitting the syllabic and temporal augment, almost certainly for metrical reasons. There i s no syllabic augment, for example1, in 2617 fr.4 i i 10, 2619 f r . l i 11 and possibly in the case of 6e uoX* 2359 f r . l i 6 and v}auov 2359 f r . 1 i i I i The temporal augment is omitted in 2617 fr.4 i 5, 2617 fr4 i i 7 and 2617 fr.6 1.Elsewhere in the fragments the augment is retained, there being six examples of the syllabic augment and five of the temporal. iv) Miscellaneous variants derived from epic. a) The variant form of the second aorist of the verb "to go",fiXudov, i s attested i n JnXud' 2617 fr.29 5, and must have originated from the epic corpus, there being no evidence of i t elsewhere. b) itoxe'euitev in 2618 f r . l i 6 ( and also possibly in 2360 f r . l i 2) is precedented in Homer. Examples from Alcman and later Occidental inscriptions indicate that in the Occidental group of dialects the loss of intervocalic digamma resulted i n contraction of the vowels juxtaposed: thus'ApeXuos became "AXtog. The derivation of the second aorist of 38 *eitu) i s complex , but the penultimate stage in i t s prehistoric evolution appears to have been e p e u n o v , with the digamma between the syllabic augment andAthe diphthong of the stem. In epic the loss of digamma i s not followed by contraction, hence the form eeuitov, which has in turn been adopted by Stesichorus i n a passage highly reminiscent of epic. From this comparison of the language of Stesichorus and that of 32 the epic poems we may conclude that there i s l i t t l e ccoincidence of phonological features. Characteristics such as the lack of contraction occur in cases of metrical convenience and where the phraseology i s borrowed almost directly from epic. On the other hand, i n the poet's morphology there is a much greater incidence of usage parallel with epic, so much so that without the evidence of Occidental ^lies and without evidence for the terminations of the athematic i n f i n i t i v e s , which are anomalous, the general impression given by the language of Stesichorus i s of one that i s structured on the same morphological base 39 as epic. f. At the level of morphology a language is more highly 40 structured and less l i k e l y to permit i n f i l t r a t i o n from alien sources In the language of Stesichorus 1 poems therefore one can see that the poet was basically perpetuating the linguistic structures of epic, incorpacatLng very l i t t l e that may be identified as morphological features of his native dialect. At the iame time, however, intrusions at the more susceptible level of phonology may be observed* C. Features of Stesichorus'1 language that do not occur either in the Homeric epics or i n the Occidental dialect group. i) Phonology. There are instances of the treatment <qf nasal+sibilant clusters in Stesichorus" language that d i f f e r from the comparable treatment of such features i n epic or in the Occidental dialect-group . a) In the accusative plural of the feminine A-stem nouns, original-etvg was reduced to sigma, with compensatory lengthening of the precedingj vowel (the original long alpha having become short as a result o£'.0i'$ho£f'*M\:law) :ih"3shV- Southern group and - Western group the result was -as, while in Lesbian-ats Occasionally in Hesiod, the Homeric hymns, Alcman and later Theocritus, a short for,m of the accusative plural of the thematic declension occurs, possibly adopted as a metrical device by analogy with the short alpha of the accusative plural of the athematic declension and extended from the A-stem to the 42 O-stem of the thematic declension . In Stesichorus we find one example of this short accusative in itaycf s PMG 184 2 . Not one of . the other instances of the accusative plural A-or O-stem in the papyri or the citations i s of certain length, since they a l l are followed by words beginning with a consonant or occur at the versef end. The occurrence of such a short-vowel accusative plural i s now recognised as a metrical device employed increasingly in the later 44 epic poets, and not, as was once thought, a feature of "Doric" ;. Pavese considers the device as one that belongs to the poets of the continental", epics, adopted from the Northern dialect ( his Setten-. trionale) at a time subsequent to the migrations to the coast of Asia 45 Minor . The occurrence of the short-vowel accusative in choral l y r i c , however, and particularly i n Stesichorus, may indicate that the practice was found useful i n the composition of dactylic verse, and 46 therefore does not prove any specific dialectal a f f i l i a t i o n s b) The feminine ending of the present participle derives from 47 *-ont- /ia , which became -ovao and i n turn followed the same pattern of developments as the accusative plural termination of the A-stem nouns mentioned above. In epic, Attic and Ionic we find -ouaa, the regular product of compensatory lengthening in the Southern dialect-group. In the Western group we would expect to find the secondary vowel represented as u), as in the case of MSaa in Alcman. > _* Stesichorus and Alcman, however, both used the form-ouaa, for the feminine participle, such as i s attested i n the Lesbian branch of the Northern dialect group. 48 According to Page , the usage in Alcman is to be explained by ortho-graphy and hot by phonology: "We have reason to suppose that Alcman himself spelt exoaa fr.37 3 and evSoaa f r . l 73 etc.; how he pronounced them we have l i t t l e or no idea." The discovery of instances of -ouaa in Stesichorus, so far unsubstantiated by the same type of ccmpen SatSory lengthening i n other words in which liguid+sibilant clusters are involved, seems to suggest that the occurrence of an apparently "Aeolic" form in Alcman was not merely a quirk of orthography but rather some vestige of an early non-epic tradition. The form also makes an occasional appearance in non-Homeric hexameters such as the 49 lines of Eumelus (ca. 730 B.C.) quoted by Pausanias : xcJu y^P 'I^ oiydfxctx xaxadu*yLO£ eltXexo Motaa a xadapat xa\,' eXeu^epa adyflaX' e^ouaa. 50 Another example may .be .seen i n the hexameter line on the Duns cup : Moujct you d y t p l Zxcfyav6pov eii(p)pu)v Spxoy* deu<v>6ev And a third instance i s the dedicatory inscription from the Heraion 51 T r at Perachora (ca. 650 B.C.) : ejyyeveouxa hurto6 le^ou*. The evidence i s hardly sufficient to substantiate any theory properly and the 52 problem of the origin of such a form i s l i k e l y to remain unsolved However, the evidence from Alcman (11 examples) and from Stesichorus 53, (7/8 examples) /does provide grounds for asserting that this particular form of the feminine present participle was the rule ratherathan the exception in early choral l y r i c , and that by the time of Pindar and Bacchylides i t had almost certainly become the traditionally accepted form, which i n turn affected the form of the 54 masculine aorist participle, namely -otug. . In the poetry of Stesichorus i t i s not certain whether this formation of secondary long o u occurred in other words. Later l y r i c poets, Ibycus, Pindar and Bacchylides, present us with examples of M o u a a ^ . In Alcman*, however, we find the normal Laconian M&aa^6, but the papyri of Stesichorus have not preserved the word. Later commentators record certain of Stesichorus' invocations, but these appear to be periphrases, not specifically including Motaa - or 57 Mcniaa , or else epithets that Stesichorus applied to the Muse, as in PMG 240 and 250, without any indication of the form of the word Muse i t s e l f . What is assumed to be.a parody of Stesichorus' introductory lines to his Oresteia in Aristophanes' Peace 755 f f . , naturally employs 58 the Attic Mouca . Thus there are no grounds other than the form of the present participle for assuming that Stesichorus used M ouca , and the influence of epic in other areas, such as in the form of the accusative plural of 0-stem houns and of the masculine a r t i c l e , may have resulted in the use of Mouaa as opposed to M otca. c) Another instance of a so-called Aeolism appearing in both Stesichorus and Alcman, again involves the phonological development of 59 j a nasal+sibilant cluster . The non-epic adjective xXeuvos takes the form of x A e e v v o - from an original root * x A e P e — to which was added the adjectival suffix -avo-, the latter un^ftgoing phonological change consistent with the group of dialects i n question. Thus one would expect to find i n Laconian xXefnvdg , which with the loss of intervocalic digamma would contract to x X n v o * s . This does not, however, occur in the texts of Alcman as they have survived. We find x X e v v d * in f r . l 44 and what is almost certainly xAeefvivje in f r . 10 (b) 12. Page believes that the oddity of the former results from the 7/6th century spelling xXevd* for xXnvtf which was later interpreted 1, as an error foryjxXevvd' w . In fr.10 (b) 12, however, the space in the papyrus allows for the restoration of {yvj,hinting that the double nasal from original nasal+ sibilant could occur i n Alcman. The emergence of the same form xXeevvot-in Stesichorus 2619 fr.32 7 removes some of the doubts cast upon the veracity of the form in Alcman. As in the case of the feminine present participle i n -otaa, so in the case of the word xXeevvdsi. i t s occurrence in early l y r i c points to some a f f i l i a t i o n of the language used by the 61 early l y r i c poets v/ith the Northern dialectal group, where compensatory lengthening by means of a double consonant is a regular feature. The 62 form xXeevvds recurs in Pindar and Bacchylides , and hence one assumes that this particular form of the word had at some point during the development of choral l y r i c become traditional. Just as in the use of - ouoa, the treatment of av to vv seems to be confined to a small linguistic area and i s not universally 63 applied . In the personal pronouns we find in both Alcman and Stesichorus the Occidental form SfrCv (PMG 1 60, and 2360 f r . 1 i 3). It i s precisely in the case of these pronouns that the Homeric epics offer the alternative forms parallel to the "Aeolic" double consonant form, as in ayuu\>. Thus the choice of the termination-otaa and the form of xXeevvd's cannot have been determined by the precedent of Homeric epic, either directly or by analogy, where did the forms originate? It seems unlikely that Stesichorus invented the forms, or was responsible for the introduction of their usage into choral l y r i c in view of their appearance in the poetry of Alcman. On the other hand, ;inasmuch •< as his poetry may have circulated more widely than that of Alcman, /'it does seem probable that Stesichorus was 37 responsible for the survival of the forms as part of the choral l y r i c apparatus^. i i ) Morphology. a) 2617 fr.13 (a) 1 presents us with an apparently isolated formation of the dative plural of the word for "hand". x e t p and i t s variants have remained a philological mystery, which results primarily from an obvious confusion in the development of the consonantal cluster -pa- and the possible existence of two alternative stems: x ^ P " ^ d -65 X e p a - . On the one hand there are certain words in which - p a - is preserved intact, eg.dd*p0os , and by analogy dative plurals such as d n p a c occur from Homer onwards. Assuming the stem x e p - # one would consider the epic version of the dative plural x e p o t as belonging to this category. On the other hand, there also exists a group of words in which the treatment of the original rho and sigma appears to follow the pattern of such a combination in the sigmatic aorist, namely by the doubling of the rho or by the lengthening of the preceding vowel to compansate for the loss of sigma. /Thus we find the Lesbian cfeppat, and Atifcic-Ionic (Jeupai, from'dpepaai,. By the same token, therefore, from the postulated stem * x e p o - one would expect to derive a double consonant in the Morthern dialect group,X£PP"# as exhibited by the 66 Lesbian poets , whilst in the Occidental group, a secondary long vowel,x n p - . The texts of Alcman provide an instance of the genitive singular xnpo's PMG 3 fr.3 i i 80 and PMG 84. We discover, however, 67 that not only in Alcman, but also in Sappho and Alcaeus the epic form X e p a u prevails;. as the dative plural. In Stesichorus we find two occurrences of the word in i t s oblique cases; xepi- C 2 6 1 7 f r « 1 9 i i 8 and xepo>fjki 2617 fr.47 1, where undoubtedly the short vowel formation from the stem x^P" was metrically useful, epic verse providing a precedent for at least the dative singular. It is strange, therefore, in view of Stesichorus'use of the short form in x^P- and the general prevalence of xepat in other poets, that we should find xnpcri,' in the papyrus. Metrically -pa-makes position, thus obviating the need for a special long vowel in the dative plural.. Are we therefore just i f i e d in assuming that the anomalous form i s the hyper-correction of a later scribe? It seems to me that one could argue equally for either case from the meagre evidence w6 have at present.\r\p - appears in texts belonging to the occidental dialect group although not in Stesichorus and not elsewhere in the dative plural. Secondary long vowel n is t e s t i f i e d in the same papyrus in ^uynv 2617 fr.7 i 2, but this word has not escaped the 68 doubts of scholars . Finally, i t i s not inconceivable that Stesichorus made use of anomalous forms and that, since the variation in possible forms of the word apparently existed from earliest times,the poet had a certain amount of latitude in his choice of the alternative forms and may even have had the license to create a form such as xnpcrC to suit the particular context. b) Another confusing picture is. presented by the evidence for the termination of the present infinitive of the thematic verb, both contracted and otherwise. (pOynv 2616 fr.7 i 2 represents one instance of the so-called Laconian i n f i n i t i v e . This type of i n f i n i t i v e derives i t s name from i t s frequent appearance in the papyri of Alcman, where there are, among several ambiguous instances, two incontestable ones of a secondary long vowel in the thematic i n f i n i t i v e of a non-contracting verb: (pauvnv P M G 1 4 3 a n d d v 6 d v n v PMG 1 8 8 . (In the case of the former, the papyrus in fact reads <pdt\>ev# c u t i f c is certain that as soon as the introduction of the Attic alphabet ffi&ade i t possible to differentiate between long ev and short ev , the normal representation of the i n f i n i t i v e was - n v ^ * ) ! Similarly in contracted verbs i n Alcman we find enauvnvPMG 1 4 3 and yautfv PMG 1 1 7 . In the case of contracted verbs the evidence in Stesichorus is not in accord with the examples cited from Alcman. We find TtoAeuefjuv 2 6 1 7 f r . 4 i 8 , whose termination in - e L V» i f correct, might suggest that (puynv was a hyper-Doricism. Conversely, TtoAepe^tv may be erroneous, but we have no othex evidence in support of one case or the other. More puzzling,however, is the occurrence of yapev 2 6 1 8 f r . l i i 9, where the epsilon is incontestably short. An alternative short form of the i n f i n i t i v e , assumed to be derived from the termination - e v added to the zero-grade of the thematic stem, does occur in early Greek. We find examples from, literary texts such as in Hesiod , Works and Days 6 1 1 , otito6pe*TiEV or Pindar Olympian I 3 , y a p d e v , but the only instance in which a contracted verb is found with a short termination occurs i n Argolic rooAev (? century)^ 0. Page admits that the f i n a l syllables of e i t a u v i t v and yaufTv are metrically indifferent, but argues for - n v on the grounds that qxxtvnv and a v 6 & V r i y are definitely long and because of the 7 1 representation of such inf i n i t i v e s i n later inscriptions . Metrically the termination of ya\i£v in Stesichorus is undoubtedly short and hence we are forced again through lack of further evidence >to postulate that metrical expediency must be the explanation for the poet's use of the short form of the f i n a i syllable of the i n f i n i t i v e , whether or not there was a precedent for such a form in previous poets. The other instances of the thematic i n f i n i t i v e occur i n citations where the risk of contamination makes any judgment uncertain. This i s also true of the example of frxXXeuv in 2506 fr.26 i i 24, a quotation from a commentary on the Oresteia of Stesichorus. whether the original Occidental -nv Was lost i n the course of transmission, or the commentary preserves the original form -ei,v, which would betray epic influence, we shall -probably never know. If readings such as Sojrjsuj.'v 2617 fr.13 18 are correct and (puyfiv i s also possible in view of other Occidental features, then we find ourselves proposing the somewhat untenable hypothesis that the poet used different forms on different occasions according to whim. We must ultimately admit that, although the papyri are more reliable as evidence than the citations, they are not i n f a l l i b l e . The evidence as i t stands cannot provide the basis for any definite statement on the form of the thematic i n f i n i t i v e in Stesichorus. c) The two instances that give evidence of the terminations used by Stesichorus for the athematic i n f i n i t i v e are quite unprecedented in literary works and almost unknown in inscriptional material. In 2617 fr.4 i 7 we discover what appears to be i t o X u M ^ p 6 u o v etv. The phrase has epic precedents, although not with the i n f i n i t i v e , and the possibility of epic etvat seems to be precluded by the fact that enjambement would demand the division et|vau. The only parallel for euv as i n f i n i t i v e of the verb"to be" occurs in Euboean Ionic, which 72 seems an improbable source for Stesichorus . One therefpre suspects some sort of conflation with the thematic i n f i n i t i v e in -etv, . and i f this were the case then one would be assuming that the thematic form in-euv did occur elsewhere in Stesichorus. S i m i l a r l y , the form eCyeuv 2619 fr.13 5 suggests the cross-influence of the thematic -euv. In the Occidental d i a l e c t group the athematic: i n f i n i t i v e ends i n -yev, such as i n nyev , Alcman PMG 1 45 (from o r i g i n a l *eayev, with Occidental n showing the compensation f o r the loss of sigma). Archaic i n s c r i p t i o n s do a t t e s t the form -ynv and a » 73 Rhodian form nyeuv a l s o occurs . I t i s possible,therefore, to explain euyetv as a combination of the athematic -yev and thematic -euv. The •w 74 S i c i l i a n comic poet, Epicharmus, a l s o presents us with the form eCyeuv , so that i t i s remotely possible that eCyetv i s i n f a c t a s p e c i a l S i c i l i a n v a r i a n t , whatever i t s l i n g u i s t i c o r i g i n may be. Unfortunately, we have no evidence as to the frequency of the form i n Stesichorus' works. In so f a r as Stesichorus appears to have employed epic morphemes i n preference to Occidental ones, one would suspect that epic terminations f o r both thematic and athematic i n f i n i t i v e s would creep i n t o h i s poems, but the texts thus f a r have remained s i l e n t . Of these miscellaneous o d d i t i e s of language considered above, we f i n d that three appear to have been derived from a source other than feh^VQpjSl^en^al ; d$f3te< & ^ j^£g^43*' the Homeric epics : namely the short form of the accusative p l u r a l , the short form of the thematic i n f i n i t i v e and the feminine p a r t i c i p l e i n -ovaa. Pavese considers a l l three to 75 have ori g i n a t e d i n the Northern d i a l e c t group . The f i r s t two, however, could e a s i l y have emerged s o l e l y from within the sphere of l a t e r e p i c, whose poets were perhaps lacking the f a c i l i t y i n o r a l composition possessed by t h e i r predecessors, and do not ne c e s s a r i l y point to stimulus from an outside d i a l e c t group. There i s a stronger case f o r i n t e r p r e t i n g the emergence of the p a r t i c i p l e i n -ouaa as the r e s u l t of influence from the Northern d i a l e c t group, i n view of the p a r a l l e l forms that have survived i n the Lesbian poets. Pavese questions the relevance of the parallel phonological change witnessed in the Lesbian poets 7 6, but since there i s a certain amount of consistency in their treatment of the -va- cluster as opposed to the more or less isolated instance of the feminine participle in the language of Alcman and Stesichorus, the form in the latter case must have been acquired by way of cross-influence or as a loan-formation, as in the case of xXeevvd's. The latter does not belong to epic at a l l , the regular form being XXEUTO'S or xXoxds. I can conclude only that both of these, peculiar forms may have been absorbed into a tradition on which both Alcman and Stesichorus drew,faestradition not associated with the Homeric poems but having, a f f i l i a t i o n s with the Northern dialect group at some point. Thus, from this examination of the limited evidence for the language of Stesichorus, we see that i t may be described as morphologically close to the language of the epic poems, while in i t s pronunciation, to judge from the representation of the vowels in particular, i t i s akin to the Northern and Western dialect groups. The language of Stesichorus 1 poems may therefore exhibit a mixture of dialects, but i t i s important to note that there i s on the whole consistency of dialectal a f f i l i a t i o n in the := phonology and morphology of that language. In the preceding discussion I have been concerned primarily with distinctive features and linguistic oddities that occur i n Stesichorus, without recounting the standard features of what we might c a l l common Gfeek. Many of the features observable in the epic poems do i n fact remain constant throughout the history of the Greek language, and i t i s therefore not surprising that Stesichorus' language 43 should show some structural similarity to that of the epic poems, representing as i t does a continuation of the Greek language at a literary level in the Archaic period. In the development of the oral tradition i t s e l f one can see the amalgamation of elements from different dialectal sources which constitutesa linguistic creation unrepresentative of any one dialect group or branch of a group. Yet within i t s own special context, that i s , the oral performance of the poetic composition, the mixture of dialect would be neither a r t i f i c i a l to the performer nor incomprehensible to the audience. As a literary creation the epic i s not an a r t i f i c i a l conglomeration of incongruous elements, but a unified r e a l i t y . F5som a hi s t o r i c a l or linguistic point of view, the, scholar may reduce the whole to a number of components, determined by some a r t i f i c i a l frame of reference for his own specific purpose. He must not, however, forget that the audience for whose ears the poem was intended would hardly have questioned whether the language represented to them a linguistic unity or not. Hence the similarities of morphology seen in theilanguage of Stesichorus and that of epic should be thought of only in terms of a continuation of the traditional literary language in which the distinctive elements from Northern or Southern dialect groups were no longer recognised. On the other hand the phonology of Stesichorus' language is distinctly western in i t s a f f i n i t i e s , with an admixture of features identified as belonging to the Northern dialect group. Although i t has been suggested that choral l y r i c has developed out of the continental epic tradition and has absorbed dialectal features from the Northern dialect group that were later misunderstood to be of "Doric" or 77 Western origin,- , there is another, less complicated explanation for the consistent appearance of long alpha and other phonological features that may be either of Northern or Western origin in the poems of Stesichorus. With the importation and dissemination of the Homeric and alsbvthe continental epics into the west through large cultural 78 centres such as Syracuse i t i s hard not to imagine that professional bards would emerge from the western colonies whose native dialect belonged to the Western group and that their pronunciation of the epic poems could have been influenced by their native dialect. Features such as the preservation of long alpha , the equivalent of eta i n the Homeric poems, would make no differences to the structure of the * -'-hexameter li n e . It would be possible, therefore, for a S i c i l i a n version of the epic poems to influence a poet wishing to create a novel form of verse or at least revitalise the time-worn epic tradition with western material that included linguistic elements from the dialects of the western communities. The proximity of Stesichorus' metres to the hexameter also suggest that the hexameter epic was the direct predecessor of his verse. Whether or not certain metrical patterns of 79 choral l y r i c were derived from the hexameter verse is open to debate , but poems such as the Geryoneis , with i t s purely dactylic measures, together with the proximity of the morphological structure of Stesichorus ' language to that of epic, point to a very close relationship. The hypothesis that an epic background with overtones of the Occidental dialect formed the basis of Stesichorus' mixed literary dialect does not exclude the possibility of simultaneous influence from non-epic compositions. Choral odes appear to have some a f f i l i a t i o n with cult-songs composed for performance in religious festivals, and are particularly associated with the "Dorian" communities . There are some indications i n the remarks of later grammarians that Stesichorus 81 also composed Hymns , and the preeminence apparently given to Apollo might be interpreted in the light of the importance of his cult in the 82 Greek west . Evidence for poetic composition in the west prior to Stesichorus' time is meagre. The presence in the west of Xenocritus 83 of Locri, of Xanthus, a S i c i l i a n writer of Dithyramb , and of Arion, gives some indication of the existence of poetic forms other than the epic in the west from the 7th century onwards. Such compositions may have been in part responsible for the superimposition of non-epic elements in the poems of Stesichorus, particularly i f he himself also composed works other than the lyrico-epics that have survived. The suggestion of non-epic influence , particularly as an explanation of some of the odd linguistic forms that have been discovered in Stesichorus runs contrary to Pavese's theory of the relationship between continental epic and the emergence of choral l y r i c . Indeed, while i t is true that the assumption of a continental epic tradition displaying features from the Northern dialect group may account for non-Homeric forms in Hesiod and the Hymns that were hitherto considered "Doric", and may account in part for isolated forms in choral l y r i c i n mainland Greece, the evidence is insufficient to prove conclusively that the whole tradition of 84 choral l y r i c derived from i t . It is interesting to note , for example, that in Alcman the two instances of the feminine participle in -exact that appears to be precedented in the quotation from Eumelus,' dp not occur in dactylic measures. whether one thinks of Eumelus or Arion as possible sources for the introduction of the Northern form -ouaa in Stesichorus, theories cannot at present transcend the boundaries of speculation. We can state, however, that Stesichorus in his creation of lyrico-epic poems was not only well-versed in the structure of epic language in general, but , as we shall see in the ensuing chapters, had also an excellent knowledge of the Ilia d and Odyssey specifically. One cannot consider his poems to have belonged to a tradition that evolved from the continental epics alone. Footnotes to chapter II. 1 R. Coleman, "The Dialect Geography of Ancient Greece," TPhS (1963) pp.. 58-126. In this lengthy a r t i c l e Coleman elaborates upon the theories put forward by W. Porzig in "Sprachegeographische Untersuchungen zu den altgriechischen Dialekten," IF_61 (1954) pp. 147-169) and by E. Risch in "Die Gliederung der griechische Dialekte in nsuer Sicht, " MH 12 (1955) pp. 61-36, regarding the distribution of dialects i n ancient Gfceece, showing the inadequacy of a simple t r i p a r t i t e system, since i t failed to account for the interrelationships of the sub-dialects of one group with the sub-dialects of another group, i n particular the various a f f i l i a t i o n s of the Thessalian/Boeotian/Lesbian group. 2 Certain distinctions hetween Ionic, Doric and Aeolic were recognised as early as Herodotus" time (cf. Hist.I 142 and Thucydides, Hist. I l l 112 and VI 5), although Strabo (VIII i 2) appears to have been largely responsible for the traditional concept of a tr i p a r t i t e division which was generally adopted by scholars u n t i l the investigaions of Porzig and Risch in the 1950's. On the history of Dialectology, see J.B.Hainsworth, "Greek view of Greek Dialectology," TPhS (1968) pp.62-76 and A.Bartonek, "Greek Dialects of the Second Millenium B.C.," Eirene 9 (1971) pp. 49-67. 3 See the Suda s.v. E x n c C x o p o c . • • • x a ^ e o x u v OUTOU x a n o u i f y a x a AuipC*6t StaA.e'xxaJu e v B u 3 A C * o t , s x c " . Thucydides speaks of a "mixed" dialect spoken at Himera, which he attributed to the fact that i t was colonised by Chalchidians from Zancle (Ionic) who were joined by a group of Syracusans (Doric) (Hist. <*VI 5). Possibly the concept of the "mixed" dialect of Himera was Thucydides' solution to the problem raised by the language of a poet, supposedly from a colony of at least partial "Ionian" origin, which was ostensibly "Doric" on account of features such as the long alpha. Thucydides' suggestion has been followed by some scholars (eg. by Holsten, De Stesichori et Ibyci dialecto et copia  verborum , Stralsund, 1884) in theories that the dialectal mixture in the language of Stesichorus and Ibycus resulted from some sort of hybrid vernacular spoken at Himera and Rhegion. However, inscriptional evidence from Himera i s minimal on account of i t s sack in 409 B.C. and in any case we are not even certain Stesichorus came from Himera. The recognisably "literary" nature of the poet's language rendeis such speculations on the nature of the vernacular in Himera or whenever, totally pointless. 4 For this theory, see CO. Pavese, Tradizione e Generi poetici della  Grecia Archaica (Rome, 1972)pp.16-19. Cf. also J.A. Notopoulos, "Studies in early Greek Oral Poetry," HSFh , 68 (1964) pp. 18-64. 5 For the literary dialect of Archilochus a useful summary i s to be found i n A. Bonnard, Archiloque (Pa\ris, 1958) pp. l v i i - l x i . See also A. Scherer,"Die Sprache des Archilochos," in Archiloque, Entretiens  Hardt X (Geneva, 1964) pp. 89-116. 6 For the dialect of the Lesbian poets, the standard work i s E.M. Hamm, Grammatik zu Sappho und A l k a i o s B e r l i n 1957) . A short summary of the main features i s given in D. Page, Sappho and Alcaeus (Oxford, 1955) pp. 327-329. 7 The language of Alcman i s dealt with extensively by Page, i n Alcman:  the Partheneion (Oxford, 1951) pp. 102-163 and subsequently by Risch, "Die Sprache AlkmansJ" MH 11(1954) pp. 20-37. Recently w.F.wyatt in "The Aeolic substrate in the Peloponnese," AJPh 94 (1973) pp.37-46, traces the r e l i c s of Aeolic features back to a .proto-stage of Greek which arrived in the Peloponnese with a migration of Greek-speaking peoples southwards in the second millenium B.C. His theory does not however take into account the controversial appearance of -otcra in the language of Alcman (cf. page 33 ff.) . 8 Cf. note 4,above. 9 See A. Thumb, Handbuch der griechischen DiaiLekte (2nd ed. rev. by E.Ki'eckers, Heidelberg, 1932)p;, 70f. Cf. CD.Buck, The Geeek Dialects (Chicago, 2nd ed., 1955) p. 154 f.. 10 M. Nothiger, Die Sprache des Stesichorus und des Ibycus (Zurich, 1971" p. 72. 11 Thumb, op.cit., p. 75 f f . 12 Risch, a r t . c i t . , pp.61-76, Coleman, a r t . c i t . , p 58 f f . , and Pavese, op.cit., i n the tables at the end of the book. 13 Thumb, op.cit., pp.71-73. 14 See page 40 f. of this chapter. 15 Pavese, op.cit., p.94: choral l y r i c poets, with the exception of Alcman,use -yev. 16 Page, Alcman..., p. 131 f. and Nothiger, op.cit., p.74. 17 Nothiger, op.cit., p. 75 and see page 38 f. of this chapter on the thematic i n f i n i t i v e s . 18 Nothiger, op.cit.,p. 5. 19 Nothiger, op.cit.,p. 47. Buck, op.cit., p. 22, identifies the suffix -xct equivalent to -xe and particle xct equivalent to xe, as exemplifying Occidental short alpha where the other dialeat groups have short epsilon. In the case of the former the original consonant was a labio-velar, while in the latter a pure palatal; hence the change to a dental i n the case of Ionic etc. In "Doric" verse the lengthened xa would appear to stem from metrical expediency/. \ 20 Nothiger, op.cit., p.52. 21 N&thiger, op.cit., p.12. 22 There are four peculiarities of the so-called Doric accent,of which 2 are borne out by the accent-marks occasionally inserted in the papyri of Stesichorus: 1) .Short .accusative plurals ofco- and A-stem nouns seem to have been considered long for purposes of accentuation, eg.Mcfaos Pindar, 01. II 71. There are no certain examples of this short accusative in the papyri, of Stesichorus,and itayctg in PMG 184 is accented "normally" on the ultima. 2) Final diphthongs at and ouin the nominative plural of 0-and A-stem nouns appear to have been counted long in "Doric" accentuation: 6<J)i/yd"vou T E xat doitaaC'loL 2359 f r ; 1 1 2 ; Exouaai, 2617 f r . l . 2v. 3) Final short syllables considered as long: C^cTvov 2359 f r . l i i 2. e a n A d ' d o v 2617 f r . 29 3. 4) certain adverbs and the genitive plurals of nouns receive a circumflex on the ultima: cf.navtSs 2735 f r . l 6» of Ibycus? 23 The author of the treatise On the Sublime 13 3 places Stesichorus together with Archilochus as the "most Homeric" after Plato. Cf. Dio Chrysostora, Or. i i 33, Quintilian X 1 62 and the verse from the Palatine Anthology VII 75. Of the remarks made by modern scholars, we may point for example to Lesky, op.cit., p. 151 or Frankel, op.cit., p. 320. Bowra, GLP pp.78-80 describes Stesichorus' debt owed more to Hesiod and later epic poets than to Homer himself. Cf. however, C. Santini's a r t i c l e "Omerismi in Stesicoro," GIF 22 (1970)pp.71-76. 24 See L.R.Palmer's a r t i c l e , "The language of Homer," in A Companion  to Homer, edd. Wace and Stubbings (London, 1962) p. 97 f f . and P. Mazon, Introduction a l'lliada (Paris, 1948) p. 110 f f . . 25 See Bartonek, art.cit.,p. 58 f f . and Risch, a r t . c i t . , p. 63. 26 Palmer, op.cit., p. 85 f f . . 27 Nothiger, op.cit., pp.33-34. 28 See page 35 f. of this chapter. 29 See Nfithlger, op.cit., p 46, Pavese , op.cit., p.49, for the usage in Stesichorus, and P. Chantraine, Grammaire Homerique I pp. 46, 69 and 200 for the Homeric usage and H. Troxler, Sprache und Wortschatz  Hesiods (Zurich, 1964) p$5B f f . for Hesiodic usage. 30 P. Chantraine, Morphologie historique du Grec (Paris, 1967) p.40. 31 Chantraine, op.cit., pp. 40-41 and Nothiger, op.cit., p. 19 f f . . 32 See page 220, chapter V I . 33 There is no evidence for the form of the genitive plural, which in the Homeric dialect was uncontracted : 'Otov . There are two very doubtful p o s s i b i l i t i e s of the Occidental contracted fcrnm in Stesichorus: AtcocJSctv 2619 fr.28 2 ( but the accusative singular i s more probable) and 6av 2803 f r . 4 5. In P.Oxy. 2735, whose authorship is contror versial, we find the form MotpSv f r . l 14. Pavese (op.cit., p.87,8) is inclined to think that the uncontracted forms are of Northern origin on account of their appearance in Kesiod. 34 See page 26 of this chapter. 35 Chantraine, op.cit., pp. 99-100. The reason for shortening original long diphthongs is uncertain. ,\M. Lejeune, Traite de  phon^tigue grecque (Paris, 1955) p. 196 , believes that this phenomenon is analogous with the shortening of ei, to E L in the f i n a l position. 36 M.L.west, "Stesichorus redivivus," ZPE 4 (1969) p. 141. 37 Chantraine, op.cit., p. 119, according to whom the usage even in Homer had become redundant and a r t i f i c i a l . w 38 A possible derivation of the stag.es from *wek may be stated as follows, beginning with the reduplicated aorist of the wzero-grade of the root : *we-wk- > e - we - wk -> e-we-uk - > e-weik -> e euit-> eetn- . 39 Cf. the statement made by Page in his a r t i c l e , "Ibycus* poem in honour of Polycrates," Aegyptus 31(1951) pp. 162-164, which was written before the publication of P. Oxy. vol- XXXII . Although the limitations of the fragments prevent a f u l l synopsis of the phonology and morphology of Stesichorus' language, we can now see the truth of Page's supposition that n^^£?^ase:''"the:'idUAlec^ is in fact basically that of epic, with a veneer of "Doric" and a slight admixture of "Aeolic"/and that the aame was true for Stesichorus, except that there were no "Aeolisms" in the latter. At the time there was no evidence of the forms -otact and KXeevvos in Stesichorus. 40 Coleman, a r t . c i t . , p. 64 f f . . 41 Lejeune, op.cit.,pp. 112-113 and Chantraine, op.cit., p. 50. 42 For example, Theogony 60, 267, 401, 534; Works and Days 564, 663, 675; Hymn to Mercury 106. See Nothiger, op.cit., pp. lOO-lol. 43 Cf. however, Haslam's suggestion that i f the colometry of PMG 184 were revised, the f i n a l syllable of itctycfs would be without doubt short (art.cit., p. 17). 44 The question of the short accusative plurals in Hesiod i s discussed by A.Morpurgo-Davies in "'Doric 1 features in the language of 50 Hesiod," Glotta 42(1964) pp. 138-165, The old theory that features such as the short accusative and the short form of the thematic i n f i n i t i v e were of "Doric" origin i s dismissed and i t is suggested that metrical expediency of a short form in an altered variation on an old formulaic expression might explain the creation of such features on the possible analogy of the double form of the preposition eg / etc. 45 Pavese, op.cit., p. 38 f f . . 46 In early choral l y r i c the form in fact only makes sporadic appearances; apart from Stesichorus' itaycts in PMG 184, i t occurs in Alcman PMG 17 5. Later examples from Pindar often occur where there are alternative readings, eg. in 01. I 53 or II 71 or Nem. I 24. 47 See Lejeune, op.cit., p. 110 f f . , Nothiger, op.cit., p. 89-95, Page, Alcman... , p. 133-134 and Risclj, "Die Sprache Alkmans," MH 11 (1954) p. 31 f f . . / 48 Page, op.cit., p. 133. 49 Pausanias, IV 33 2. Cf. Pavese,op.cit., p. 104. 50 Risch, a r t . c i t . , p. 37. P.K. Kretschmer, Griechischen  Vaseninschriften (Gutersloh, 1894) p. 104 f., no. 87. 51 SEG XI 233. 52 Page, op.cit., p. 133-134, believes that historical connections between Cyrene and Thera, and between Thera and Laconia make i t possible that the language of Alcman and the Cyrenean dialect had a common ancestor in "old Laconian", which would explain the appearance of the feminine participle in -ouaa in Alcman and in the dialect of Cyrene. However, the inscriptional evidence from Cyrene is much later than Alcman, the earliest being a 4-th century oath. Cf. the objections to Page's theory expressed by Risch, a r t . c i t . , p. 31 f f . There may be some significance that the odd. ; : occurrence of-ovso)rOuaa is confined to the feminine participle. Does i t perhaps stem from some formulaic expression in a choral tradition restricted to female ' performers? ' • " • ' ~ " • ' • " 1 ' r -53 For example, Alcman, PMG 1 61,73; 3 fr.3 i i 5 and 15, and Stesichorus, 2360 f r . l i 1, 11; 2615 f r . l 2; fr.4 11 16,17; f r . l l 3; fr.43. 6. 54 For example, Pindar, Pyth. I l l 50 : Ai5aaus. • 55 Ibycus, PMG 232 (a) 23; Pindar, 0l_. I 112, III 4, XIII 22 etc. Bacchylides, 5 4; f r . 55 2. In Pindar there are 40 examples of Mouaa and not one instance o£'".:M ouaa, whereas Bacchylides prefers Mouaa, of which there are 12 examples, as opposed to one certain and one doubtful example of Mouaa. 56 Alcman PMG 30; 5 f r 2 i 22; 8:; 9; 46. 5 7 P. Oxy. 2506 f r . 26 i 16,11. 58 PMG 210. 59 Lejeune, op.cit., p. 110; Page, op.cit., pp.107-108; Nothiger, op.cit.,pp.41-42. 60 Page, op.cit., p. 107. 61 Pavese, op.cit., p. 93. 62 For example, Pindar, Pyth. V 20, IX 15 and Bacchylides 2 6; 5 182. Also, Simonides, 136 3D. 63 (fyitfivva is recorded in Alcman PMG 62, other than the example of xAeevvds cited above. In Stesichorus only the formxAeevvd's occurs as an example of this treatment of the nasal+sibilant cluster. 64 It appears that the highest percentage of these so-called Aeolisms occurs in Pindar (although the impression may in part be due to the greater amount of his poetry surviving in comparison to that of the other i y r i c poets. The principal examples are: a) the feminine participle in -ooaa; b) the aorist participle in -aus;c)Mouact = Muse; d) 3rd person plural of the present indicative i n - O L C J L ; e) personal pronouns ctuues etc.; f) the i n f i n i t i v e of the verb "to be" : euuevaL g) consonantal doubling in xeAa6£vv6*s, (paevvds , MXeevv6*£. 65 Lejeune, op.cit., p. 106 and Chantraine, op.cit., p. 79. 66 XepPes, Sappho, PLF 90 (1) i i 21;xe*ppa Alcaeus, PLF Cl 21. 67 Sappho,; PLFf-81 (b) 2; 96 29; Alcaeus, PLF B 13 6. 68 On the papyrus, a "corrector" has written E L beneath n(v). since other i n f i n i t i v e s seem to have been written - E L V , eg.itoA.£UEr_LV 2618 fr.4 i 8, one ironders whether this i s a case of correction, or an explanatory gloss. 69 Page, op.cit., p. 121f. 70 See F. Bechtel, Die Griechischen Dialekte II (Berlin, 1923) p. 449. 71 Page, op.cit., p. 122. 72 See.Bechtel, op.cit., vol. I l l p. 180. 73 Chantraine, op.cit., p. 276; Bechtel, op.cit., vol. II p.646. 74 Epicharmus, f r . 99 2 (Kaibel); cf. the form TtE'itoaxct which is found i n both Stesichorus (PMG 261) and Epicharmus ( f r . 11 Kaibel). 75 Pavese,op.cit., p. 88 f f . . 76 Pavese,op.cit., p. 106. 77 See page 33 of this chapter, with notes 44 and 45. 78 Note that the epic poet Euiaelus came from Corinth. Could the source of the feminine participle in Stesichorus have beennthe recitation of Eumelus' poems in the chief Corinthian colony in the Greek west? 79 Cf. M.L.West, "Greek Poetry 2000-700 B.C.," Cg_ 23 (1973) pp. 179-192,,,innwhich he relates the division of the various types of Greek poetry and their modes to their dialectal origins: Ionic, Lesbian and Doric. The Lesbian mode derived from the Northern dialect group in late Mycenaean times, while Ionic and Dorian developed out of the Southern group, the latter being absorbed by the post-Mycenaean i n f i l t r a t i o n of the Occidental dialect group. According to his theory the metrical evolution of the hexameter was late: not much before the 8th century and thus heresy to those who believe in the Mycenaean origin of the Catalogue of Ships-. As West himself admits his hypothesis must remain speculative.: what the poetic forms of non-epic compositions were in the dark ages we shall probably never know. 80 AiMeillet, Apercu d'une histoire de l a langue grecque (Paris, 1948) p. 200. Cf. Vallet, op.cit., p.305. 81 Clement of Alexandria, Strom. I xvi 78 (PMG 276)j cf. Athenaeus VI 250b. 82 Vallet, op.cit., pp. 305-306. Cf. PMGW232. 83 See chapter I, note 1, with page 7 and note 27. 84 Cf. Pavese's own admission of the limitations of his theory, op.cit., p. 108. Chapter III Verbatim adaptation of Homeric "formulae" in Stesichorus. In this and the following chapters I shall examine the nature of Stesichorus' diction in relation to i t s epic precedents, specifically in,terms of that feature considered distinctive to oral epic, namely the "formula". The generic, stylized and repetitive nature of the language of Greek epic is in i t s e l f totally evident and undisputed*. Parry in his original s t a t i s t i c a l examination of the function of the traditional, fixed epithet in the Homeric poems introduced the word formule, with his own specific definition of the word, indicating i t s technical application 2 within the framework of his thesis . His study of the nature of the repeated phrases, their extension and their economy, in the narrowly limited area of noun-epithet expressions gave the i n i t i a l impetus to" the extensive investigations into the integral constituents of oral techniques in the composition of heroic verse. The term "formula", i n i t i a l l y employed as a technical term, whose usefulness is immense; provided that i t s definition i s clearly understood, has in recent years come under a barrage of criticisms and warnings against the misconceptions and erroneous connotations that may be construed with regard to the nature of oral composition. Nonetheless, the word has become sufficiently well established in the jargon of Homeric studies that i t is acceptable as a generic term, on the understanding that the author w i l l and must define his own specific application of the term in his particular area of research. Modern scholarship in Homeric diction has produced two theories of particular relevance to this examination of Stesichorus' adaptation of Homeric "formulae". F i r s t l y , Nagler, i n his generative approach to the study of "formulae" hasia.ttempted to remove the whole issue' of "formulae" from s t a t i s t i c a l counts of repetitions and purely structural a f f i l i a t i o n s 3 of phrases with metrical patterns . Deriving his method of approach from the generative view of speech habits developed i n transformational linguistics, he suggested applying a theory of a pre-verbal Gestalt to oral composition i n which each formulary repetition would be considered as a particular manifestation on a particular occasion of performance, generated out of, or realised from a mental, not verbal "form" *hat i s , 4 as i t were, inherent i n the poet . The theory,subject as i t i s to the criticism of being applicable to poetic composition of a non-oral nature also, has far-reaching implications in terms of the poetic process in general, and especially i n literary traditions in which imitation i s acclaimed and not castigated. In the case of poets such as Stesichorus, whose literary formation has been strongly influenced by the traditional epics from their cultural heritage, we can, by Nagler's theory, interpret the poet's choice of phrase inca given context as i n part generated out of the traditional associations of that context. The second and more important theory is that of Hainsworth i n which he considers the "bond of mutual expectancy" as being integral to the concept of a "formula" as a repetition of content^. "Mutual expectancy admits of i n f i n i t e gradations, words, at f i r s t fortuitously combined, by recreation slowly become regularly associated. One must note that Hainsworth places emphasis on the words, toe: content. Mutual expectancy depends on content, not form, and this i s of particular rele-vance when one considers the treatment of Homeric "formulae" by later poets Since the metrical structure of the phrases, considered of prime importance 7 in the theories of some scholars on the nature of oral i m p r o v i s a t i o n i s not--always functional i n non-hexameter composition, the poet's use of the traditional "formulae" must depend on content. The traditionally associated word-groups may, be imitated by the poet on the basis of content not structure. Just as in oral improvisation the "formula" i s a device that cannot be divorced from i t s end, namely the narrative, so in later, non-epic composition, the adaptation of such "formulae" by the poet is dictated by the content and by the specific association of noun and epithet in a specific context. Thus I define the "formula" as a repeated word-group; that is, a group of words that occur together more than onceln similar contexts in the poems of the epic corpus. By virtue of the repetition of these word-groups the bond of mutual expectancy established between a certain noun and epithet can become firmly set in the subconscious of both a poet and his audiende through continual exposure to the repetitions. Thus, the bond of mutual expectancy w i l l operate outside the specific context of the hexameter poems. The poet may repeat the "formulae" in one of several ways that we shall see i n the course of examining Stesichorus' adaptation of "formulae". It mVs.t be noted that the "formula" i s no longer part of the technique of the oral aoudo's, but rather one of the tools of a itotnxn*s who creates as well as imitates r . the formulaic inheritance fromihis predecessors. I propose to examine the diction of Stesichorus in terms of his use of "formulae" , which w i l l be divided into various categories according to the relation of the phrases to their Homeric precedents; a) word-groups that have recognisable precedents in the Homeric poems, and have been imitated verbatim by Stesichorus; b) those word-groups that have recognisable precedents, but that have been modified by the poet; c) phrases that have precedents in sources other than the I l i a d and Odyssey . In this chapter I shall examine those "formulae" that have acted as precedents for Stesichorus 1 phrases directly from the Iliad and Odyssey. I have subdivided the word-groups into the following categories, and within each category the order is determined according to the alphabetical order of the nouns i i i each group: a) Noun-epithet groups b) Noun + genitive of possession c) Double noun groups, linked by HOC* or xe ... xctt d) Noun-epithet groups in which one element is supplemented e) Miscellaneous f) Word-groups occurring only once in the Homeric corpus. g a) Noun-epithet groups . 1. yXavjQ Situs 'ASdva 2617 fr.3 3. There seems l i t t l e doubt that the supplement f i r s t proposed by Label in P.Oxy. volume XXXII i s correct. In the I l i a d there are over 30 instances of the "formula" in the nominative case, at the verse-end, and in the Odyssey over 50 instances. In l y r i c , however, there are very few examples of the association of Y^cuxSitus with"' AS'dfva. Page believes that the context of this fragment is a council of the gods^, and i f this i s correct, then one imagines that perhaps here Athene was speaking-^on Heracles' behalf, i n a way that recalls her defence of Odysseus i n the f i r s t book of the Odyssey. 2. xoupu6uxv x' ctXoxov PMG 185 4 ( SLG 17 6) We find this word-group in the I l i a d thrice in the, accusative, twice in the genitive case, but i n the Odyssey only once i n the genitive 57 case and once in the dative. The epithet x o u p t 6 u o s i s almost always restricted to ctXoxos in the feminine, ctvnp or Tco*cms;in the masculine. aXoxos,however, possesses a variety of epithets :yvno-xtf, dvxu$e*n, HeSvrf, &uyapns» but particularly the almost insignificant (ptXn. The additional association with children i s best exemplified i n the I l i a d by the lines: ev^ '* aAoxdv xe <pi?Xnv eXticov xal vifntov utdv or EUtpPaveetv aXoxdv XE <ptXnv Kal vtfituov uudv (Iliad V 480, 688; VI 366). The association of "wife" and "children", though natural, i s not i n every case appropriate, as for example when Briseis i s described as x o u p u 6 u f j v aXoxov (Iliad XIX 298). I t would seem, however, that i n this fragment of Stesichorus we have in n a j 6 a s XE (puXous a variation of the formulaic xctt vrfituov utdv, in which the poet employs an alternative word-group, not previously associated with xoupuSuiv ctXoxov or aXoxov xe q)tXnv^. 3. 6€nas ... XPtoeov PMG 185 1,2 (SLG 17 1,2) The "formula" XPUCTEWL 6e*itaC occurs 6 times in the I l i a d and Odyssey, and need hardly surprise us since golden drinking cups and 12 bowls are f i t t i n g i n any heroic society . The image of the colossal cup of gold floating upon the streams of Ocean i s not found in the Homeric corpus. Helios departs to Ocean for the duration of the night but the vehicle of transport goes unmentioned : he seturlns'-V'::e-?' ctxaXappeuxcto gaduppdou * ftxeavoto (Odyssey XIX 433 f f . , cf. XXIV 11,12,). ffrom Athenaeus one discovers that the poet who wrote the Titanomachia was the f i r s t to describessthe Sun's nightly transport as XE*3TIS and that 13 one Theolytus also used this term . In the epic tradition relating to Heracles' exploits, Peisander (7th/6th century) and later Panyassis t e l l of Heracles capturing the Sun's cup,6Enas or (ptctXn, but no further details are given of the cup's composiiton^ In Mimnermus' version of the Sun's journey back to the East, a different, although logical image is found: Tovy£v Y^ P 6ta xoya cpepeu TtoXunpaxos euvii, ^HouXn 'H<paCo"rou x£PCf^v eAnXayevn Xpuaou T t y n e v t o s ... Believing that this "hollow bed" in which the Sun sleeps resembles the hollow of a cup, Athenaeus includes this fragment in his collection of "cups". In this instance the association of gold is explicit: the bed has been fashioned by the blacksmith of the gods, in precious metal appropriate for the furniture of the gods. One might think that the association of "gold" and "sun" was obvious. It is perhaps surprising that the epithets of Helios in the Iliad and Odyssey are few, and those that do occur describe the sun's luminosity: cpae'duiv, <paea£yBpOTOs, Aaynp6*s. Golden,on the other hand, qualifies various material goods - armour, clothing, thrones, cups -and i s occasionally transferred to deities such as Aphrodite. There i s one isolated comparison between the sun and gold; a necklace brought by one of the suitors for Penelope is described as : X P ^ C T £ 0 V J nAexTpoLCtv eepye"vov, n^Xtov (5s. (Odyssey XVIII 296) . From a later date, the Hymn to Helios XXXI (7/6th century) aeeveals a stronger association between the brightness of the god and his golden helmet (lines 9/10) and his golden-yoked chariot (line 15). Even though there is this later association of gold with the Sun, as he shines on the earth, there is no logical reason 7 for the vessel that carries him after he has ceased shining upon the world to be created out of bright gold. I conclude rather that the amazing cup of the Sun was described as golden because cups of the gods and heroes were naturally and traditionally made of gold. Since the dates of Peisander of Rhodes , those of the poet of the Hymn to Helios and those of Stesichorus a l l belong somewhat nebulously to the 7th or 6th centuries, i t is d i f f i c u l t to determine where the image of the golden cup f i r s t originated, and what part, i f any, the golden bed of Mimnermus' poem played i n the creation of the image. The aptness of presenting the Sun's nocturnal transport as a cup, which Heracles could "hijack" for his expedition to the west, i s interestingly 14 explained by Athenaeus as a joke on the part of poets j i n view of Heracles' propensity for cups of wine, what could be more f i t t i n g than a colossal cup in which to traverse the Ocean? The motif did recur later in a context which may not have been concerned with Heracles, namely in a fragment of Aeschylus* Heliades*^. Both Pherecydes and Apollodorus repeat the tale of Heracles borrowing the golden cup of the sun and their accounts may have been derived from Stesichorus whether or not Stesichorus was the inventor of the image of the Sun's unusual vessel, he may have been the f i r s t to explore, the po s s i b i l i t i e s of using the traditional association of 6e"itas and xptiaeov / thus presenting his audience with a far more concrete picture of the Sun.?s nightly voyage than is giveniintthe Odyssey. 4. yaxa{pe]crat def_o]t^t 2617 fr.13 2.5 Of this formulaic association there are in the dative plural alone 6 examples^in'the Iliad and 7 in the Odyssey, of which a total of 7 occur at the verse-end. A further 5 examples may be added from the Hymns and 2 from Hesiod. There i s , however, only one instance that may have acted as a precedent for the sentiment expressed in this fragment of Stesichorus, namely in Odyssey I 82: et yev 6f| v\3v TOUTO <j>C"Aov yaxapeaau deotot . In this same fragment of Stesichorus, 7 lines above, we find a similar association i n $[e]<3v yaxctpwCv • Generally in the case of the genitive plural Seuiv , the epic poets preferred to leave the noun without any epithet attached, presumably on account of metrical convenience. However, regardless of position or frequency of this word-group in particular cases, the association of yctxapes with d e o t i s sufficiently obvious for us to identify Stesichorus' usage as "Homeric". 5. e n 1 d | x p o x d x a v xopuipcfv 2617 fr.4 i i 10,11 This word-group constitutes an imitation of a Homeric "formula" i f one compares i t with dxpoxefxrii, xopu<pnt itoXu6etpd6os 0<JX\5ynoto (Iliad I 499, V 754, VIII 3) and e n ' dxpoxcfxris xopucpris Ea'you (Iliad XIII 12). Outside the actual Homeric corpus , the combination i s found in the Hymn to Pan : dxppxcfxnv xopu<pnv ynXoaxd'itov etcavoiBaCva>v (XIX 12) . In the epic tradition, however, the word xopu<pn is used primarily i n the sense of physical mountains such as Olympus or Ida. The secondary meaning of "head", of an individual creature, does occur once in the Iliad , as VIII 83, also in the context of a combat between heroes. In this instance the epithet axpnv indicates the possibility of i t s application to the word xopuipn in both l i t s senses. The passage in Stesichorus appears to involve the shooting of one of Geryon's three heads by the hero Heracles. Doubtless i t is the head that towers highest above the hero, as is suggested by some of the early representations of the 17 scene on vase-paintings . The poet has employed xopu<pri in i t s less common sense of "head!', and in retaining the epithet most frequently associated with the word in i t s sense "mountain" he has deliberately suggested both potential meanings, magnifying the dimensions of the monster. Knowing the popular epic usage d x p o x d x n t xopuipnu, the audience would automatically compare the height and size of Geryon's head to a mountain peak. They could envisage the hero faced by a massive, grotesque mountain of a monster. Thus we may observe that in the case of e i t ' 61 d x p o x d x a v xopu<pdv, Stesichorus has obviously copied a "formula" from the epic corpus, but the usage can hardly be termed blind imitation. The word-play fa c i l i t a t e d by the double meaning of xoputprf, depends upon the audience's awareness of the traditional, or more„,frequent context of dxpoTCtxn xopuipn and i t s recognition of the implications of the "formula" when applied to Geryon. 6. ye*Xi, x^wpdv PMG 179 2 The association of x^ <*>p6"v and y d X t does occur twice in the 18 Homeric poems, at Il i a d XI 630 and Odyssey X 234 , and hence we may count this as a "formula" precedented in Homer. The epithet x ^ p d s however, i s more often applied to grass and leaves, particularly young shoots, as in x^ wP<*S pii)TtasB(Odyssey XVI 47) and also to luxuriant :under-growth or even to opos# as in the Hymn to Apollo 233. In i t s second sense of "pale" or "pallid", x ^ p d s occurs in a metaphorical context with dx*0*s (Hesiodic Scutum 264 u . ) and with 6eos(Iliad_=VII 479).. It i s assumed by :LSJ that y £ X u qualified by x ^ p d v must be understood as a reference to the "yellow" of the honey, but there seems no reason why "paleness" is not intended as the sense of the epithet . The phrase ye*At x ^ p d v , therefore, demonstrates how Stesichorus may derive a word-group from the Homeric poems, but his choice may represent an association of noun and epithet that is less common than other possible combinations of either the noun or the epithet. 7. na"C6a cpuXov 2619 fr . 1 6 18 toiC*6as t e qiuXous PMG 185 4 Although there are precedents for this word-combination in the Homeric poems, i t seems that the "formula" <pC*Xov xe*xos was much more common, primarily in singular contexts. S t a t i s t i c a l l y , of a total of 57 instances of the singular, there are 5 examples of itau6ce <ptXov/nv in the Ili a d and 3 i n the Odyssey. Indeed, over half of the instances occur without any accompanying epithet. Accusative (pCXov UL6*V on the other hand, appears 21 times in the I l i a d and 8 in the Odyssey, while nominative <ptAos uildg 6 times in the I l i a d and 22 in the Odyssey. In Hesiod also, only one instance of nau6d'<pJXov'stands beside 7 cases of singular itaus'without any epithet and 4 of the plural. It would appear that the group TtatSct ipJXov became popular only, in later poets. 8. <p£Xou itctTQQd 's 1 9 2350 f r . l i 11 The association of (ptXos and itd*Tnp exists in a l l grammatical cases in the Homeric poems, although the genitive singular i s more often avoided on account of metrical awkwardness. This d i f f i c u l t y , however, may be overcome by the insertion of a preposifeion between the epithet and i t s noun, as for example in cpiTXou yeta itaxpos axouifv (Odyssey XVII 43) or by the use of the longer form of the masculine genitive in -o to . Stesichorus' metre allowed him to retain the traditional association without making use of these alternatives. 9. ev vnualv euacre*Xyoi,s PMG 192 (SLG p. 156, corr.) Despite the fact that this "formula" occurs in a quotation from Plato, which may not be totally accurate in representing the poet's words, I include the&bra^ijroup in this category since the association of euaaeXyos with vaus i s well attested in epic and i t s recurrence in this fragment of Stesichorus (in some form) must be considered an imitation pf the Homeric "formula". It is noteworthy, however, that i n the majority of cases of this "formula" i n the epic poems the preposition eitiT is found, ev, i f correct, may reflect a necessary change, metri gratia. 1©. x e * x v a <pJXa 2359 f r . l i i 3 In the plural one frequently finds the epithets dyXotd or v n i t u a qualifying x e x v a in the I l i a d and the Odyssey. x e * x v a <pJXa occurs twice in the Iliad and not at a l l in the Odyssey, but Stesichorus doubtless relied upon the association's being recognisable from the prevalence of the vocative q>£Xs~ire*x$>ov. The most common word-group involving x e * x v o v is in fact x £ x v o v e u o v , i t s v e r s a t i l i t y demonstrated by i t s occurrence at three different positions in the hexameter line. In 2617 f r . 19 7 one reads x e * x v o v followed by a break in the papyrus. There i s every possibility that, since this fragment appears to belong to a personal address or exhortation, the common phrase x e " x v o v e u o v occurred. 11. 4*» e u p e t a v X'&Mva 2260 f r . l i i 18 (PMG 233) In the Homeric :poems this noun-epithet group occurs in the nominative case alone, 4 times in the Ili a d (and thrice in the Hymns) in a l l instances taking the f i n a l position in the lines oaa x p e t p e t e u p e t a X § o J v, (Iliad IV 182 etc.). When the accusative is given an epithet the sole candidate appears to be i c o u X u B o x e t p a v (cf. dative gwxtavetpnt.-.arid y e A a t v n M , b u t the accusative occurs more often without an epithet. If we turn to the scope of e u p e t a , we discover that in the Odyssey, perhaps naturally, the epithet is applied mostly to the sea rather than to the land, and similarly in the I l i a d , with the exception of i t s association with Troy/: e v t T p o t n t e u p e t n i ' • I assume that in Stesichorus' metrical scheme e u p e t a v x§ d * v a was possible, although unprecedented'in the hexameter, and that the word-group arose from the pattern set by the nominative e u p e t a x$"h>» b) Noun+genitive of possession groups. 1. "_\zag ifjraTi, 2619 fr.15 (b) 4 In the Homeric poems this word-group normally takes a less precise form,Se2v- C d x n x L . Barrett's supplement suggests that the 20 goddess who i s exerting her w i l l i s Athena , and I interpret.this: instance of Stesichorus referring to a specific deity , rather than to the gods inqgeneral, as one of a number of characteristic attempts to regenerate phrases that had become meaningless in the Homeric poems on account of continual repetition. In this way the poet could render -episodes in his poems with greater vigour and immediacy, despite their reliance on the epic tradition. 2. J i toxotuoi ) i t a p a Ttoyds PMG 184 3,4 The word i tnytf i s rare in the Homeric poems, but there are three instances of this notion of the "streams of rivers": aZ x f aXaea xaXa veyovxctL J xcu, Ttnyas noxauffiv xaV nuaea nourfevxa (Iliad XX 8,9; Odyssey VI 123,4) . There i i n y i ^ s n ° f c given an epithet, while in this fragment of Stesichorus' Geryonels we discover that the "streams" are diteJpovcts and d p y o p o p u c o u s , and that the identity of the river i s specified. I shall deal with this passage in greater detail in the discussion of word-groups that contain elements from the epic tradition with new juxtapositions 21 % and associations . We may note here that the epic "formula" Ttnyds TtoxctySv was probably influential in the construction of the Stesichorean phrase, but ithat the poet required a specific reference, and so replaces the plural with the singular itoxauou and introduces the locality of Tartessus: another example of the particularisation of a general statement. 3. ucas Abo's PMG 185 6 By calling Heracles "son of Zeus" Stesichorus i s apparently following the Hesiodic tradition, as exemplified in the Theocjony and 22 ^"ne Shield , but of the 11 instances the use of TOILS i s confined to TOILS xe ALOS yeyaXou in the Shield 371. Elsewhere UL<5S occurs. In the Il i a d and Odyssey there are indeed precedents for the combination of TOILS and Aud*s, but there too OLD'S is more prevalent. Heracles i s chronologically too early to make a legitimate appearance in the Trojan cycle, but Odysseus' journey to the Underworld gave the poet an excellent opportunity to incorporate many of the heroes who lived prior to that era. The fact of Heracles' being the son of Zeus is made pertinent to the situation; he has to suffer despite his lineage: Znvos t 23 yev uctbs J\a KpovL*ovos (Odyssey XI 620) . It is perhaps odd however that in both Odyssey XI and Theogony 952 i t . i s Heracles' "companion" H^ ae who is identified as Tta"C6ct ALOS yeydXoLO , and not Heracles. As regards the completion of the line in accordance with Page's proposed colometry, i t is possible that aLYtoXou occurred, ctLYLo'xoLO'/ou and yeYCtAouo/ou being the two most prominent epithets that accompany Atds. It seems, however, from the texts that survive, that for metrical or other reasons, the second,yeYdX^oco,was the epithet employed in phrases containing itd-Cs/^du6a,whilstxauYbdxobo most frequently occurred in the "formula" describing Athena: xoup^v ALOS ctLYLOxc-LO (11 times in the Odyssey) . Alternatively, one might expect that the proper name 'Hpot>tXe"ns would appear i n this nine-line sentence, but for such an assumption one would need a great deal more evidence on Stesichorus' methods of structuring sentences. I conclude , therefore, that in his choice of this "formula" the poet again appears to make use of the less common word-group for the expression "son of Zeus". 4. <pdos deXuoo 2619 fr.13 8 <pa*os neXifoto occurs 18 times in the f i n a l two and a half feet of the hexameter line in the I l i a d and the Odyssey If the colometry of 25 this fragment has been correctly reconstructed , Stesichorus appears to have retained the f i n a l position i n the line, although for metrical ., reasons he was forced to use the shorter form of the genitive. There are three contexts in which this "formula" appears in the Homeric poems: 1) with reference to the setting sun; 2) with reference to an individual being alive, that i s , looking upon the light of the sun; 3) with reference to an individual dying, that i s , departing from the light of the sun. The third" of these contexts seems most lik e l y in this fragment of Stesichorus, in view of xax'otaav two lines below. One is reminded perhaps of Achilles' speech at the beginning of Book XVIII of the Il i a d , i n which the hero reflects upon the fated death of Patroclus: tog itoxe' uou urfxrip SueitdcppaS.e x a i > you eecite MuityuSdvojv xdv apcaxov exu StSavxos eueto xepalv 3tto TpoJwv XeC*4>euv cpdos neXJoto 'Iliad 'XVIII 9.-11 However, although imitation of Homeric "formulae" i s quite evident in this fragment, the precise context i s only a matter for speculation. c) Double noun groups linked by xaC or xe ... xau. In this category there are few examples. Two groups, Btau xe xa\, ailxpab 2619 f r . l i 6 and SoSpaxcf xe xat, 3poxd"evx[a ueXea 2617 fr.4 i i 13, properly belong to the category of new juxtapositions and 26 w i l l be considered in chapter IV 1. ydxotu T ' dvSpoQxxaaCat 2617 f r . 17 6 There are two parallels for this pair of nouns in conjunction to be found in Ili a d VII 237 and XXIV 548: auxctp eydv eu* ot&a yctxaS x' dvSpoxxaau'as xe. aieC TOU uept daxu ycJxctt T* avfipoxxaauxo xe. An extension of this association exists in Odyssey XI 612: uayLVctu* xe uctxcto xe <pd*V0L x1 dvdpoxxaauxu, with general reference to the*>exploits of Heracles, and the same li n e , or "formula" appears in the Theogony i n a personification of these abstractions as the children of Eris (line 228). Thus the association of ucfxca and dvSpoxxacruxL was probably well-established i n the epic corpus, and in this instance Stesichorus has borrowed the phrase directly. Unfortunately the fragment is insufficiently complete , so that i t s content remains obscure. One suspects that the description of the horrors of war may have been applied to Heracles' adventures i n the west in a way similar to line 612 of Odyssey XI. 2. cap xct [xcA] &CaT3?a 2 6 1 7 f r « 4 i i 8 There i s some question as to whether there is sufficient space in the lettering on the papyrus to include the xao between the alpha and omicron, but there seems to be l i t t l e alternative, particularly in view of the two precedents for the conjunction of actpxcts with oaxe'ct i n the Odyssey: thus eyxctxcf xe actpxctg xe xcu, oaxect yueXdevxa (Odyssey IX 293) and ou yap £TU acfpxcts xe xcu, 6axe*a tves exouauv (Odyssey XI 219) . adpB, , however, i s not common i n the Homeric poems, and tends to occur i n the plural, as in the lines quoted above. The only instance of the singular occurs in Odyssey XIX 450>451: itoXXdv 6e 6utt<puae adtpxos 66<JVXL j XuxpupLg dC^cts, ouS' 6axe"ov oxexo (ptoxds. One could seek an epithet agreeing with atfpxct for the mutilated part of the l i n e , but there i s no evidence for an epithet regularly attached to crcfp£. In view of the emphasis placed upon the penetration of the arrow, by means of the repetition of 6to? in lines 8 and 10, i t seems more li k e l y that the bones as well as the skin were pierced. Hence I read a a p x a x a u o a x e a as an imitation of the Homeric d c f p x a s x e x a u o a x e a assuming that the poet employed the singular rather than the plural of a d p 5 on account of the metrical requirements and that the scribe perhaps, compressed the letter-spacing of the x a u . d) Noun-epithet groups in which one element must be supplemented. In this category I have included noun-epithet groups of which one element is missing in the text of the papyri, but of which, for most cases, the traditional "formula" from the epic corpus gives a good indication of a probable supplement in the context. In some cases the association i s restricted in such a way in the Homeric corpus that, given a similar context in Stesichorus, the probability of the same phrase being imitated by Stesichorus is higher. This section , however, i s speculative in content, and i t s only value is that i t incorporates a l l the epithets from the fragments that are precedented in epic. In each case the probability of Stesichorus* imitating the "formula" precisely i s considered in as far as the context may be determined from what survives of the fragments. (It w i l l be more convenient in this section to arrange the phrases by alphabetical order of the epithets rather than the nouns.) 1. d y x u X o x d S o u 2619 f r . l i 9 This epithet occurs only twice in the Homeric poems, in Iliad II 848 and X 428, in both cases qualifying IladoVES and thus suggesting 3 27 o v e s a Y X u A o x d * 5 o u i n this fragment . Elsewhere the epithet appears with Mn|6euou (Pindar, Pyth. I 78) and K u u e p u u i v (Anacreon, PMGH504), neither of which would be particularly relevant to the Tfcojan theme. 2. dXtud'pqJupov 2619 fr.16 7 The word dAuTtd*p<pupos i s not common in the Homeric poems, occurring only thrice, and appears to be associated with the colour of garments. The context of this fragment appears to involve Aphrodite, |K]Uupoyevn 's(line 6), but there is no tangible clue as to the person or thing described as dAi,7td'p<pupov. 3. y a u f o x o s 2619 fr.18 9 In the Homeric poems,yaLrfaxoc. refers specifically to Poseidon, whether in conjunction with IIoaeu6do)V or with the periphrasis 'Evvoauyaoos. In the Ilia d there are six instances of the latter combination, one of the former and three in which the epithet occurs without any substantive There is also one example of a l l three combined: aXXa IIoaeu6d*a)V Y°H'tf0X0S evvoaC*Yatos (Iliad XIII 43) . In the Odyssey there are six instances of the epithet with IIoaeu6da)V and one with 'EvvoaCyauos. noo"£L6du>v or 'EvvoaifYaLOs are therefore possible conjectures in this fragment. Barrett West and Fuhrer join this fragment with 2803 f r . l l which contains the Doric 0 28 4. 6]uaaSvuuos 2619 fr.19 4 An uncommon word, as is indicated by there being only three instances of i t in the Homeric poems, without any perceptible fixed associations. 6 u 0 o J v u u o £ might from the context here refer to Paris (cf. 6d0Ttapts in Iliad;8III "760 .). If this were the case, the application of the epithet may be original. 5. euaaadrepoi, 2803 fr.7 7 The occurrences of this epithet in Greek are almost totally confined to epic, nor are they frequent there. One formulaic precedent might be recognised in itavxcts enaaauTe'pous u€Xaae x^o\)i TtouXugOTetpnt (Iliad XII 1 9 4 , XVI 4 1 8 , VIII 2 7 7 ). The context of the fragment i t s e l f sheds no light as to what the associated noun might have been, but since the fragments of 2 8 0 3 are related to the Trojan cycle, we might consider a possible par a l l e l from Ili a d I. Between lines 5 and 7 of this fragment a later hand has inserted a remark or gloss : "2? o $ p u y C followed by 35TO£OT.[] .. , possibly referring to 6 TO j i n line 5 . The apparent allusion to an "archer" i s reminiscent in this context of the situation at the beginning of Iliad- 1 where Apollo dpyupciToEos in his anger (cf. the possible supplement x e x o X J c S y e v o g i n line 4 of this fragment) spreads fatal disease through the Greek camp by means of a xaxdv 3 £ X o s . The epithet used to describe the mass of people who succumbed to death as a result i s eitaacriJTepos: . . . TOEO 6 ' 'ATtdXXaJV eu £ a y e * v o u fixouaev, eitet y d f X a ot q>C"Xos ?iev ?ixe 6 ' eit' 'Apyeuouat x a x o v 3 e * X o s * ou 6£ v\) X a o \ §vn£o"xov eitctaad'TepoL, . . . ( I l i a d I 3 8 0 ff.) The other word identifiable in the marginal: notefis o g p u y o - . In i t s simple form this epithet i s applied to Ares (Iliad V 8 4 5 ) , * Achilles (XIX 4 0 8 ) and Hector (VIII 4 7 3 ) , but not to Apollo. In the Homeric Hymn VIII, Ares i s called o g p u y d & u u o s (line 2 ) , but he i s tradi-tionally associated with the spear not the bow. 6 . euxTtyeCv- 2 6 1 9 f r . 3 2 7 euxTuyevov i s consistently associated with TtToXt'edpov, 7 times in the I l i a d and 3' times i n the Odyssey. Moreover, there i s ah interesting precedent for the Stesichorean expression in : l X L ' o t i : ^ | ^ ^ ^ ^ S j e u x x i T y e v o v TtToXiTeSpov (Iliad IV 3 3 - VIII 2 8 8 , XXI 4 3 3 ) . We shall see in the later discussion of xXeevva[ in line 6 of this fragment that Stesichorus has taken the familiar line from the epic 71 tradition, and while retaining some of the elements has introduced others A that are foreign to the Homeric poems,such as xAeevvos and even Tpoucts , 29 instead of IXuou, to break the expected formular pattern. . 7. e u p u o [ i t a \ Ze&s 2619 f r . 1 i 16 The frequency of this "formula' i n the Homeric poems (and i n addition in Hesiod and the Hymns) makes the supplement virtually certain, eupuorca ZeO*s i s in' vcontrol of the fate of Troy, as is expressed in the words of Achilles: ud*Xa ydp e d e v euptfonct Zeus I X ^ p o env uitepe*axe (Iliad IX 419,420). 8. e o x p o x [ 2619 fr.41 1 The usage of this epithet i s obviously restricted. Its regular association withapua or aua£a i n the epic tradition would suggest such a context in this fragment. 9. Q n n o M S X e u ^ o v 2617 fr.3 5 This epithet occurs only in the Homeric poems, and i t s application i s severely limited, namely to Patroclus i n I l i a d XVI 126,584 and 839. The subject-matter of the Geryoneis would suggest that Stesichorus did not i imitate • . this specialised use of the epithet, but i n fact made a novel application of i t . ilO. uevexa*puct[\, 2359 f r . l i i 9 Five of the six instances of this epithet i n the Homeric poems are i n the singular, referring to specific individuals. There i s , however, a single case of A u x w X c A , uevexcJpyau (Iliad IX 529) which occurs within the context of the war between the AiixuXoC* and Koup?jTes » * n which Meleager fought. The context of the Stesichorean fragment appears to be related to the legend of the Calydonian Boar hunt 3 0, but whether the poet followed the particular version of Ilia d IX , which has been adapted to suit the situation of Achilles* refusal to fight, cannot be determine!'*-. 11. iteO] x a X u y o [ - 2617 fr.46 i i 5,6 The epithet neuxcfXuyos has survived only in epic sources, there being 4 instances in the I l i a d , a l l in the dative plural qualifying 9 p e 0 t ' . This limitation of scope suggests, therefore, that the conjectured masculine neuxaXiTyou0LV does not belong to an imitation of the Homeric "formula" cppeoL i t e u x a X u y n t O L . 12. xQya i i o X u jjpXo.tfafJou SctXcfaocts 2619 fr.25 5/6 This "formula" i s a combination of categories a) and b) discussed above. In terms of category b), noun+genitive of possession, the unit i t o X u i p X o o ' o g o u o SaXdaans occurs with xuya or xtfyctTct i n I l i a d II 209, VI 247, XIII 798 and also in the Hymn to Aphrodite VI 4 and the Kypria f r . VII 8 (Allen). xuya combined with S a X d a o n s unqualified i s more common, as i n Il i a d IV 422, X 574, XV 381, XVIII 66, 145, XXIV 96; Odyssey XIII »:88. On the other hand, i t o X u t p X o i T o &OLO %a\dooT\£ occurs most frequently with itapa S u v a (Iliad I 34 etc.) . In terms of category a) , noun+epithet groups itoXO*q)Xou03os i s confined to the genitive singular, long form, qualifying ^ a X a o a a 6 times i n the Iliad., 2 in the Odyssey, together with 2 instances in the Hymns and one i n Hesiod. Hence the supplement of daXcfooag i s assumed to be correct. I note therefore that Stesichorus had again probably repeated a "formula" from the epic corpus, although,.as far as one can t e l l , he has chosen a less common grouping. 13. i tov j r o i t o p o o L 2619 fr.25 2 From i t s sense alone, this epithet must refer to a ship, and indeed in the epic corpus i t s application i s restricted to vaus. In the Odyssey there are 4 examples of the "formula" i n the nominative singular, 2 in the genitive, and i n the I l i a d 2 examples,of the genitive and 11 of the dative plural : e v Ttovxc-itopotcn, ye'eaao (III 46 etc.) and i t a p a vfiuat . . . i t o v T O n d p o u o - u v (VII 72 etc.). If Stesichbrus used a word other than v a u s for "ship", we have no evidence of i t . 14. p n S r f v o p a 2619 f r . l i 21 This attribute i s applied solely to the hero Achilles, 4 times in the Iliad,^.once in the Odyssey and once in the Theogony; HOU, yex' 'AxtXXna p r i S r f v o p a duyoXe*ovxa (Iliad VTI 228 = Theogony 1007). In the context of a debate prior to the Trojans* acceptance of the wooden horse into their city, column i of this fragment preserves part of a speech of encouragement from one of-the Trojans who is suspicious of the horse and who advocates reliance on their fighting strength. pnf-Tyvopa in this context could refer to the dead Achilles as hoMonger being a threat to the Trojan victory. Alternatively , the epithet may have been applied to one of the Trojan heroes who has subsumed the Homeric attribute of i ' . Achilles. 15. aTOyerpJou | SavdxoiQo 2617 fr.4 i i 1,2 The supplement in this line i s derived from an interlinear note made by a later hand. The combination of tfxuyepos and Scfvaxog occurs only twice i n the Odyssey; uvno-rtfpwv axoyepov ddvaxov Mat, xnp' eve*Ttouca (Odyssey XXIV 414) and itavxeg yev OTuyepou dcfvaxou 6 e u X o t a i , 3 p o x 5 t a u (Odyssey XII 341) . Thus we may say that the association i s "formulaic", but we find that axoyepd's more often qualifies substantives, such as O-MOXOS, which imply death, or others such as voOaog, axn, ynpas » the precursors of death. In the Il i a d xaxds and ue*Acts are the most common epithets of death. In this fragment,therefore, i t can be seen that Stesichorus has adopted a word-group whose elements are subject to an indirect bond of mutual expectancy, in.as- much as the particular relationship of o"nJYeP°'s and dctvaxog i s infrequent in the epic poetry that has survived. 16. fl epicuxepoi [yv- 2617 ff:49 2 If the poet here follows the traditional"formula", one would expect that the epithet i s applied to Zeus, as i s the case without exception in the epic corpus: Zeus Tepiuxe"pauvos or Aut Tepiaxepctuvaju occurs 8 times i n the Il i a d , 7 times in the Odyssey, 5 times in the Hymn and 3 times i n the Hesiodic corpus. 17. unep£u*uoL 2359 f r . l i i 5 This epithet in the plural i s regularly applied to Trojans i n the Ili a d (7 times) and to depcJnovTes in. the Odyasey (3 times) . In the context of the Calydonian Boar Hunt , the Trojan association i s highly unlikely, nor i s the single instance of i t s application to the Lapiths (Iliad XII 128) a possible precedent. The poet has most probably trans ferred the epithet from i t s customary position with the Trojans to some other group deserving of the t i t l e . I t i s less l i k e l y that the epithet was applied to SepcfitovTes i n the context of a l i s t of heroes. e) Miscellaneous word-groups. 1. aXX 1 aye 6rf 2619 f r . l i 7 2. TOU 6' ditb xpaxSs 2617 fr.4 i 14,15 3. h%L x^ova 2617 f r . l . 3 xax' auaav 2619 fr.13 10 There is l i t t l e to note regarding these "formulae" other than that their Homeric origin was probably recognised by the poet's audience on account of their frequent appearance in the epic poems. Number 2 is particularly close to Ili a d XVI 793 i n context : T O O 6' dub yev xpotxos Muve*nv &d\e fotSos 'ATKJXXOIV . Noteworthy is the parallel use of the ar t i c l e as a relative of connection. Number 4, occurs 4 times in the Iliad, but xata uoupav appears to be the more frequent usage (21 times in the Odyssey and 9 times i n the I l i a d ) . f) Word-groups occurring only once in the Homeric corpus. By Hainsworth's definition a phrase occurring only once in the epic poems does not constitute a "formula", and yet the adaptation of such a phrase by later poets such as Stesichorus would indicate that the said phrase was constructed of "formulaic" elements* that i s , of words and groups of words that were suitable for incorporation into "formulae", but were never required. Alternatively, such a phrase may have occurred'1 more than once in the epic corpus as a whole, but other instances of i t have been lost. The following four word-groups from Stesichorus' poems occur only once each in the Homeric corpus. 1. afyctxo itopqi[upecot 2617 fr.4 i i 12 In the single instance in which itoptpdpeos i s applied to atya (Iliad XVII 360-361) we find a description of the earth stained with the purple-dark.blood around the body of Patr.oclus, over whom a fierce battle has been raging with Ajax the foremost defender of the corpse. Although the epithet itop<pd*peos'"is more frequently employed to describe articles of clothing or blankets of such a colour, or else the sea (which was the original source of the dye), there exists an interesting extension of the idea of darkness in EAActgE i t o p t p u p e o s dcfvctxos xal yotpct vpctTctJn (Iliad XVI 334 = V 83). Epithets conveying the sense of the darkness of blood are generally pe'Aav (Iliad XVI 529), xeAatvdv (Iliad-I 303, VII 329) and xeActLveqJEs (Iliad XVI 667). Thus, Stesichorus has adopted a less common association , precedented only once as far as we can t e l l , which may incorporate the sense of the profusion of the blood pouring from the monster's wound as well as i t s colour on account of the common use of m o p c p u p e o s with n d v x o s . The verb employed by Stesichorus in this fragment,yvctuvw, often occurs in the Homeric poems in the context of both duuct and x o v t r i . For example, the helmet of Patroclus l i e s on the ground, i t s plumes befouled with blood and dust: yuxvdnactv 62 edeupctu a C y a r u xau xovLnwx (Iliad XVI 795,6). In Stesichorus' description, however, the S5Spa£ is s t i l l worn by Geryon, so that the formulaic "with blood and dust" would be inappropriate. 2. UK] n d x o y o s Tp^^d'Aci,' 2617 fr.4 i 14 The phrase i n i t o x d y t o v xpufpotAeuov appears once in the Homeric poems, in a battle-scene in which the noise of clashing shields and plumed helmets reaches the heavens: duxn 6' o u p c t v o v TX E | 3aAAoyE*vujv aaxsuv T E xcu, C i t n o x d y a i v xpucpaAEUOV (Iliad XII 38,39). More frequently associated with xopus and <pd*Aos (Iliad XIII 132 and XVI 216) , u i t i t d x o y o s may without d i f f i c u l t y be extended to qualify TpixpcfAEua, i t s e l f a compound of (pdtAos. TpoqJCtAeux in the epic corpus, for reasons probably metrical, tends to occur without an epithet. 3. q n J A o K L S dpyctA^a 2617 fr.17 4 A s t r i c t l y epic word, q n J A o i t t s occurs only once qualified by the epithet d p Y C t A e a : ; (Iliad XI 278), whereas i t s epithet a u v r i , regularly found in the position at the verse-end, i s repeated not only in the Homeric corpus, but also in Hesiod's Works and Days 161, Shield 200 and in the Hymn to Demeter 267. The l i s t of nouns qualified by dpYaXeos in Homer i s lengthy, but i t s association with, for example, epts or uautvr) in the Il i a d , and in particular with the sound-word CTOVOS, may account for the transference of the epithet to (pd'XoTtLS, the "din of battle". Stesichorus has again chosen an apparently infrequent noun-epithet combination. The fragment belongs to a context of full-scale warfare, as not only this line, but also y d x c x i , T' dv6po [_XTaau*ab (line 6) suggests . Whether part of the Geryoneis incorporated another ergon of Heracles, i n which he participated i n some great battle, or whether these lines f a l l in a simile, we cannot t e l l . Their relation to the encounter between Heracles and Geryon himself i s not entirely obvious. In 2617 f r . 18 3 the papyrus breaks after the alpha in uAo7tbva[.. Did dpyaX^a or a u v d * follow? Brief though the fragment may be, the context again appears to be one of a battle on a large scale, with individuals (plural) perishing,]]OXU)XOTE£S (line 4), possibly with helmets or bodies r o l l i n g i n the dust, .JAL x o v C a c s ^ (line 1), i n a scene perhaps comparable with one of the battle-scenes in I l i a d XVII. According to 32 the structure of the Geryoneis proposed by Barrett and Page , the major part of the poem was concerned with the encounter between Heracles and Geryon. Thus, assuming that fragments 17 and 18 do belong to the same poem as the identifiable fragments of 2617, I conclude that these elements of a battle-scene belong to a simile or digression. 4. 6t' :' ftxeavoCo uepa'aas PMG 185 2 (SLG 17 3) Apart from one example of 6u' " U x e a v o u o i t e p n a n L S (Odyssey X 508), one can cite several indirect parallels for this phrase such as itepctu) with H O V T O S (Iliad II 617, Odyssey XXIV 118) or else Stagcts Ttopov '' ftxeavoto (Hesiod, Theogony 292) . The other sphere in which one finds the verb itepcta) with the preposition 6ux" is that of a missile piercing the breast of a hero (Hymn to Mercury 45) or his forehead (Iliad IV 502): f| 6* £Te"poLO 6uct xpotcfcpoLO Ttepnoev J aCxyn x ^ ^ t e ^ l . Accordingly, although the phrase used by Stesichorus has but one extant precedent, i t i s by no means unusual in i t s structure or associations. It i s noteworthy, however, that the phrase occurs in annon-Homeric 33 context and amid a series of distinctively modified Homeric phrases Three other "formulaic" phrases, 6u' ctu§epo£s ctx]puYETAS 2360 f r , l i 4, 6a6|yovos ducat, 2617 fr.4 8,9 and ituxuvfas)-. <PP DO V A S 2619 f r . 1 i 19 also have single precedents in the Homeric corpus, but w i l l be noted only here and dealt with i n greater detail in chapter V on account of their relationship to phrases in Hesiod and the Hymns. From the collection i n this chapter of word-groups that have recognisable precedents in the Ilia d and Odyssey I conclude that Stesichorus could and did employ regular "formulaic" expressions drawn from the monumental epics. I note, however, that not a l l of the examples cited are particularly frequent, or the most common expression that the poet could have selected. In category a), of the eleven noun+epithet "formulae", only seven appear to be highly frequent in their occurrence. Number 3, 6ends Xpudeov, i s original in i t s usage, although the traditional association of XPUCEOV and 6e'nas plays an important part in the new context. Number 5 i s likewise an example of a "formula" that in the Stesichorean context has a significance additional to that of i t s usage i n epic, through the poet's play on words. Number 6 occurs only twice in the Homeric poems, but by Hainsworth's definition of a"formula", may be considered as an limitation thereof. In number 11 the association of top eta and x ^ v has been established firmly enough for the nominative case to confirm that the Stesichorean phrase in the accusative must rely on the Homeric precedent, o f the other seven word-groups, numbers 7, and 10 are demonstrably less frequent i n their appearance in the Homeric poems than other "formulae" involving one or other of their particular com-ponents . In category b), of the four instances cited, number 2 i s rare in the Homeric poems and number 3 appears to follow the Hesiddic tradition. In category c) neither example i s overtly frequent, as far as our evidence goeS:.% In category d), of the 17 supplemented phrases with epic parallels, 8 are rare. In category e), number 4 i s an example of the poet's choice of an alternative phrase that was less common. The additional 7 examples of Homeric phrases that occur only once in the Homeric poems further support the view that Stesichorus apparently preferred to copy of. imitate "formulae" of a less stereotyped nature. Footnotes to chapter III 1 J.B.Hainsworth, Homer , Greece and Rome Surveys 3 (Oxford,1969) p.19. 2 M. Parry, L'epithete traditionalle chez Homere (Paris, 1928) p. 16, in which he gave the following definition: Dans l a diction des poemes aediques l a formule peut etre definie comme une expression qui est re"g'ulie*rement employee, dans les memes conditions metriques, pour exprimer une certaine idee essentielle. 3 M.N.Nagler,"Towards a Generative View of the Oral Formula," TAPhA 98 (1967) pp. 269-311 and also in Spontaneity and Tradition : a study J J 80 in the oral art of Homer (Berkeley,1974). 4 Nagler, a r t . c i t . , p. 311. 5 Hainsworth, "Structure and content i n epic formulae;the question of unique expression," Cg_ 14 (1964) p. 155. 6 Hainsworth, The F l e x i b i l i t y of the Homeric Formulae (Oxford, 1968) p. 35 f f . and p. 42. 7 For example in the work of E.G.O'Neill Jr., "The localisation of metrical word-types in the Greek Hexameter," YC1S 8 (1942) pp. 103-178, or more (recently in that of J.A.Russo,"The Structural Formulae in Homeric Verse," YC1S 20 (1960) pp. 219-240. 8 In this chapter , by Homeric poems we mean the Ili a d and the Odyssey. Although I shall take separate account of the phrases related to parallels from Hesiod and the Hymns and the so-called epic Cycle, in chapter V, I shall mention where necessary instances of phrases that occur i n both Homeric and non-Homeric epic sources. J9 Note that variation in inflection, aord order etc., is regarded as modification which does not,however,alter the basic bond of mutual expectancy. 10 D.L.Page, "Stesichorus' Geryoneis," JHS 93 (1973) p. 147. 11 On Ttatdas Te q>C*Aous see number 7 below. 12 Cf. N.R. Collinge, "Mycenaean di-pa and 6e'itas," BICS 4 (1954) pp. 55-59. 13 Athenaeus*Deipnosophistae XI 469e-470d. The fragments mentioned i n the following discassion on the Sun's cup a l l derive from Athenaeus' collection of examples of a cup called a 'HpaxAecov. 14 Athenaeus, XI 469d refers to itctu£0VTss OL HOUTITCU, .... 15 The fragment of Aeschylus appears i n Athenaeus XI 469f. 16 Pherecydes , as quoted by Athenaeus, XI 470c and^Apollodorus B i b l . II 5 10 both refer to a x P^eov 6e'itas, ev <Lu T O V 'ftxeavbv oeeitepaae. 17 Cf. the discussion by M. Robertson i n "GSryoneis : Stesichorus and the Vase-painters," CR 19 (1969) p. 208 f f . . 18 Cf. Hymn to Hermes 560. 19 On <pt*Aqu uax[p] bs utov see page 109, chapter IV. 20 Barrett in Page's SLG p. 26. 21 See page 122 chapter IV. 22 Theogony 316; Shield 66, 110, 150, 163, 320, 371, 392, 413, 424, 447. 23 See page 267 f f . , chapter VIII. 24 Page, a r t . c i t . , p. 146 f f . . 25 West, "Stesichorus redivivus*" ZPE 4 (1969) p. 135. 26 See page 115, chapter IV. 27 E. Lobel, P.Oxy.vol.XXXII, p. 37. 28 See West and Fiihrer i n ZPE 1 (197D pp. 262-264 and pp. 265-266 respectively. Page argues against the join i n PCBhS 19 (1973) p. 51ff. 29 See page 106 f., chapter IV. 30 Lobel, P.Oxy. vol. XXIII, P-H# suggested that at least the f i r s t column of 2359 might belong to the Suotherae of Stesichorus. For a f u l l discussion of the fragment see ;215 f f . , chapter VI. 31 Cf. M.M.Wilcock, "Mythological paradeigmata i n the I l i a d , " CQ 14 (1964) pp. 147-153. 32 See Page 1srarticle "Stesichorus' Geryoneis," JHS 93 (1973) p. 148 f. , 33 See page 110 f f . and 119 f f . , chapter IV. Chapter IV Stesichorus* modification of "formulae** from Homer. • This chapter examines word-groups in Stesichorus that have the outward appearance of Homeric "formulae", but are in fact new combinations, unprecedented i n the epic tradition as far as the extant corpus indicates. The five subdivisions into which these word-groups may be categorised are as follows: I Noun+epithet : groups that comprise new combinations of individual elements from Homeric "formulae". II Noun+epithet : groups of which one element is non-Homeric, thus providing evidence for new associations of elements from "formulaic" contexts in the epic tradition with elements from outside the epic tradition. III Longer units that constitute expansions of simple noun+epithet groups. IV Noun+epithet : groups i n which both elements are foreign to the epic tradition. As i n the previous chapter, the basic unit with which we are dealing consists of a noun together with i t s associated epithet, and also of a noun with accompanying genitive of possession, since such word-groups show the highest number of individual connections with the "'formulae" of epic. Again, we are primarily concerned with the precedents observable in the I l i a d and Odyssey, although parallels from the Hymns and the Hesiddic corpus are taken into account where relevant. I New combinations of traditional elements a) Non-supplemented 1. potdtvous ctxoVTCts PMG 243 The epithet pocStvcis i s not common in the epic corpus, occurring once in *-Homer> J^to describe a whip (Iliad XXIII 538) , once in Hesiod (Theogony 195) and once i n the Hymns (Hymn to Demeter 183), both of these Instances describing feet. The accusative singular axovxa regularly attracts the epithet oStfv, whilefcfehe plural sometimes appears with euCeaxot and daue*es, but more frequently without an epithet. The combination p a 6 o v o b s axovxas i s a unique "formula", i f one grants that the epithet hai formulary potential on account of i t s association with itd*6es on two occasions. 'Unfortunately the significance of the epithet i s lost, other than indicating the "slender" appearance of the spears, since the scholiast who gives this example i n his l i s t of the occurrences of pa6i,*>tfs, does not include the context of the Stesichoraan phrase 1. Ibycus, however, according to the scholiast, described oi xov oupovov gaaxasdvxes xudves as pct6uvoC" instead of jeuu eyelets. This,therefore, i s one of several instances in which the same unusual epithet i s attested for both poets, although in this case their application of the epithet is markedly different. Stesichorus* striking use of pa6t,v<$s may have held the same implications of size that one finds in Ibycus1' use, but the emphasis on the t a l l , slender nature of the warriors' spears i s more.likely when one notes that i n Theocritus the same epithet describes the cypress tree: 2 t a l l , dark, slender and, indeed, spear-like . Stesichorus' association of pa6uv<?S with spears may have influenced his S i c i l i a n descendant in the choice of this epithet highly appropriate for the cypress tree. 2. apuo-rov doL6o*v 2618 f r . l i .4 Examples of the use of epithets with docdds come from the Odyssey for the most part 3 and we note that there the most prominent attribute of the bard i s to be deios (10 times), with less common alter-natives uepuxXuxo's and epu'npos. (Sptaxos, on the other hand, i s used predominantly as a substantive (avdpuitos/avriP being omitted), rather than as an epithet, and hence the combination of dpucTOS with dou66*s i s not precedented even by a parallel of a different noun qualified by ctptaxos. Although i t i s easy to assume that ctpooxos i s more or less the equivalent of fcetos, both epithets reflecting the singer's excellence, the latter also has connotations of divine associations, particularly i n view of the belief that the poet was divinely inspired. The mortal nature of the bards is more apparent i n the Hesiodic corpus, which fact i s perhaps indicative of a slightly different attitude to the substance of poets. Inspiration may come from the Muses and Apollo (cf. Hesiod, Theogony 95 and f r . 305 2, M.SVW.), but the bards themselves are mortal. ctptcxos i s d i s t i n c t l y "mortal" i n i t s associations, and hence i t may be that Stesichorus in his choice of this particular epithet wishes to stress the fact that the bard i s a man, granting; him responsibility for his own 4 excellence rather than assuming total reliance upon divine inspiration . ||npes 'Axctuotf 2359 f r . l i i 3,4 eptfnpes i s an epithet found almost exclusively i n conjunction with e T o t p o v,the sole exception being the word-group ep^ n p o v dou6d*v which occurs thrice i n the Odyssey^. The exact meaning of epCnpes i s uncertain, although i t i s generally assumed that i t derivei from dpopt'crxw to " f i x " , with e p i , - , an intensifying prefix, and hence, i n the context of eTdu&oi*, eptfnpes would be sensibly translated "steadfast" or"faithful". Would such connotations be meaningful in the context of 'AxcttoiT in this passage from the Suotherae? The 'Axottou, when they do receive an epithet in fche I l i a d , are qualified most frequently by euxviftaSes or x c t p n -xoudwvxes* The 'Axcaot i n this passage, however, are more l i k e l y to be the people from the area Achaea, i n the northern Peloponnese, since the l i s t of participants in the Boar Hunt appears to be a regional one. 85 Since each of the regional groups l i s t e d receives an epithet appropriate to renowned warriors, namely cttxucttKxi?s yevexcfpyau and uitepddviou, I would suggest that epJnpes i n this passage has the meaning of "steadfast" or "determined i n battle". In view of the frequent association of euxVTfyudes and xapraxoydwvxes with 'Axauotf (=Greeks), the word-group ep tnpes 'Axcttot would have doubtless struck the audience as unusual. 4. it[aYXP3&rea 6i6*|ua] T 1 2617 fr.6 (a) 3,4 The obvious model for the entire phrase itotYXP^ o'e0' 6«Jya'T*SxovTt must be 'OXdyitta 6o5yax' exouau/exovxes (10 times in the I l i a d , 3 in the Odyssey, 5 in the Hymns) and one i s reminded particularly of those instances that refer to the Muses, for example i n I l i a d II 484 and XI 218. Naturally the halls of the Hesperides, situated i n the far west, w i l l require a descriptive epithet other than 'OAuynux. itayxP^o-ea appears tm be in unique combination with 6<iSuctxa, the common groupings being ocJyaxa xXuxcJ or 6uJyaxa xaXdt, or else a phrase such as 6oJyax 1 '06uaonos £euoto or 6uSuaxa Ktfpxns i n which the owner of the house i s indicated. %ayxp\5acos occurs infrequently in the epic poems; there i s but one example i n the I l i a d , at II 448 du*crctvoL itctYXPudeot, and none in the Odyssey. In the Hymns one finds one instance of the epithet qualifying xd*£d (XXVII 5) and one qualifying apya (3X 4) . In fact, compounds with irctv as f i r s t element are not common in the epic corpus. Thus the associa*; tion of %ayxp\Saea and 6t5yoxa is new, as far as the surviving evidence shows. The association of gold with the Hesperides is hardly surprising since, from Hesiod onwards, they are the guardians of th'e golden apples that were presented as a wedding-gift to Zeus and Hera: 'EcuepCdcis $',nls yTy&ct ite'priv xXuxou ' flxectvoLO Xpuaect xctXa yeXouot q^povxa* xe 6e*v6pea xapudv . (Theogony 215,216). In later tradition, possibly beginning with Peisander , and certainly adopted by the Hellenistic poets such as Apollonius of Rhodes, there i s also a guardian snake watching over the apples... This snake must surely be the result of a conflation of the tradition of the Hesperides' duty with that,:r©f the snake in the Theogony 335, who guards Ttayxp\Jo"ea u r j X a of no specific origin. There i s no evidence i n the extant fragments of Stesichorus of any guardian snake. What may be original, however, i s the poet's transference of the expected epithet that qualifies the apples of the Hesperides to their abode, their itaYXP^o"e<* 6oJyaTa. 5. | Javda 6* ' E X e v a 2619 fr.14 5 As an attribute of a person in the I l i a d and Odyssey SavOrfE belongs primarily to Menelaus (16 times in the Iliads and 15 in the Odyssey) while i t i s used in the feminine of Demeter (twice), of Agamede (once) and of Ariadne (once, i n the Theogony). The epithets of Helen tend to be rather uninformative about her physical appearance: 'Apyeu'n gives her place of origin; xaXXticdpnos and xaXXixouos/nuxouos indicate that she has beautiful cheeks and hair, without giving any frame of reference from which one might determine what aspect of these was considered beautiful. One further epithet,TovvJiteiiXos , i s likewise no more distinctive, since most of the Argive. or Trojan women of noble origin presumably wore long flowing robes The epithet S a v S o s denotes a reddish-brown colour when describing h a i r 1 0 , but without more evidence of Stesichorus' depiction of Helen, I cannot claim that his use of this epithet i s significantly more specific than the Homeric ones. Thus, although non-Homeric, the application of Sctvda* to'Helen i s not surprising, but effective enough that later poets such as Sappho and Ibycus also called Helen g a v d d rather than repeating one from the epic corpus"1"-1". I t is possible that the firm association of E,av%6s with Menelaus in the Homeric poems i s significant; Stesichorus probably intended the relationship between Menelaus and Helen to be accentuated (perhaps ironically) by this trans-ference of the epithet regularly expected with Menelaus to his misguided wife. 12 The context of this fragment seems to be a scene in Troy , i n which some reference i s made to the ultimate;'destrttc:tlc» of Troy by f i r e , whether i n prospect, or immediately before the event. This being the case, the presence of Helen shows that the Il i o u Persia and the Palinode were separate poems and that i n the former Helen was depicted i n f a i r l y traditional manner1:3. It i s possible, however, that even in the I l i o u Persis, the poet was perhaps consciously striving to move away from the firmly established vision of Helen by his introduction of the epithet Zav%d in his vision of her. 6. xJpxov xavocui Qepov 2619 f r . l i i 20 Of the four occasions on which a xtpxog appears in the I l i a d and Odyssey, i n two the bird i s described as eAa<ppd*xaxov uexenvwv ,"the swiftest of winged creatures" (Iliad"'XXII 109 and Odyssey XIII 87). xovuaL'Ttxepos is applied to birds i n general i n Odyssey V 56 and to thrushes specifically i n Odyssey XXII 468, but not to the xt'pxos. The 14 xCpxos apparently belongs to the species of Cpn|-~ ,land we note:: that V~ Hesiod ca l l s the CpriS; uxune*xns Cpn£, xavuai/itxepos opvts (Works and Days 212). The species i s in .general remarkable for i t s long slender wings and for i t s speed, and hence i t i s not an unexpected application of the epithet xavuat'itxepos that we find in Stesichorus, although i t i s unprecedented in the Homeric epics. 7. niti,o6(ipou Ku , itpt6os PMG 223 2 The epithet n i tn&wpos occurs only once in the Homeric poems: eVdct ol nnL<j6wpos evavTLti nXude Vifanp (Iliad VI 251) referring to Hecuba. . Otherwise, i t makes one later appearance in Oppian3"^. KO*itpts# as a periphrasis for Aphrodite, i s not common in the Homeric poems, there being only 5 instances of i t in the f i f t h book of the I l i a d , a l l without epithets. The only instance to occur in the Hymns f a l l s in the second line of the Hymn to Aphrodite, as one of the t i t l e s of the goddess. The name 'Aippodtrn i t s e l f is frequently accompanied by the epithets <pt,Xouuei,6Tis, itoXuxptfo-os or A t o s %\)y<£xr\p in this Hymn. If one considers the formation nui.o-6wpos to be a close par a l l e l to the equally rare compound dyXao-Swpos, which qualifies Demeter in the Hymn to that goddess (lines 54 ,192 ,492 ) together with i t s association with Hecuba in the I l i a d , one notices that Stesichorus' use of this particular epithet has connotations not present in any of the other epithets commonly found in descriptions of the goddess of love. The epithet i s suggestive of the image of a gentle, all-giving mother, which appears to be an entirely new attitude towards Aphrodite. 8. nepcxaXXe*{a v^aaov 2617 f r . 6 (a) 2 The epithet i teptxaXXns , referring to overall beauty, is applied in a variety of spheres in the Homeric poems: to inanimate objects 6uppos» xuSapus , Tte*itXos and 3u>u6*s; to women, as for example in I l i a d V 398 or XVI 8 5 . A geographical context of the sea presents i t s e l f in 'He*Xuos 6* dvopouae, Xuituv itepuxaXXea Xu'uvnv (Odyssey III 1 ) . Since among the epithets qualifying vfiaos we find eOxTtuevn, SevSpnecroro, uXneaca and e p n p n » but not nepuxaXXns , I assume that the poet has created yet another new word-group of a "formulaic" nature, possibly hinted at by the Homeric itepcxaXXe*a Xcyvnv . The itepuxaXXea uaab.v. belongs to the gods; i t is an island beyond the limits of human habitation, a magical i s l e . On this island dwell the Hesperides, guardians of the golden apples and their abode, according to Stesichorus, i s all-golden on this account and presumably because gold i s associated with the possessions of the gods (cf.xpdaeov 6e*itas , Iliad VI 220 and XXIV 101) . It appears that, from the earlier sources available to Stesichorus, principally Hesiod, the poet absorbed the notion that the Hesperides lived beyond Ocean, but since the poet in this fragment gives a more detailed description of the island, I assume that he, or some not too distant predecessor, elaborated upon the elusive abode "beyond Ocean" to ••.create a distinctive island, of which we 16 catch a glimpse in this fragment. The only parallel instance of this word-group , itepuxaXXea vaaov, is to be found i n Theognis, 1277: T W O S "Epws itpoXtnuv Kditpov, rceptxaXXEa vrjaov, i n which the poet refers to Cyprus as such. Is the parallel entirely coincidental, or has Theognis derived the phrase directly from Stesichorus? In an a r t i c l e on poetry i n S i c i l y in the Archaic period, A. Garzya reconsiders the theory that Theognis * Megara was in fact Megara 17 Hyblaea i n eastern S i c i l y , and i f this were the case, then linguistic reminiscences of Stesichorean expressions in Theognis would be somewhat easier to explain. 9. %_ xpov oXeSpov 261? fr.4 i 11 The Hoaeric epithets associated with oXeSpog are Xoypds and au to ' s , while nuxpds i s applied almost exclusively to otaxds (.10 times i n the Il i a d , i in the Odyssey) in i t s primary sense of physical "piercing". In Stesichorus* manipulation of the regularly associated words one can observe a transition from the physical to the metaphorical sense of 18 ittxpo"s, as i t i s found in later authors, namely as"causing bitterness" As the oXe^pos of Geryon contemplated by the hero w i l l in fact be perpetrated by an arrow, the choice of the epithet ubxpos cleverly fore-shadows this and i s consequently highly appropriate; i t is both piercing and ultimately grief-causing. A further connotation inherent in Ttxxpd's is the bitterness of the poisonous g a l l of the Hydra in which the arrow had been dipped. That the arrows of Heracles were indeed smeared with this poison i s indicated in the elaborate description of the arrow as ite<popuYPe\>os a"uon:[t ...} T e xoXat, 6Xeactvopb,s ailoXo6eCc*p3ou oduvacauv * 19 Y 6 p a g , i n the second column of fragment 4, lines 5,6 . The transference of the epithet ittxpd's as seen in this fragment is highly significant in that i t demonstrates clearly the poet's i^tehiiionaiselection of an epithet, possibly hackneyed in i t s traditional association with oZax6z, in an imaginative," effective manner that relies on the audience's awareness of i t s original usage in epic. 10. xpuo-o'itxepe itctpSe've 2506 f r . 26 (PMG 193) In the epic tradition the sole recipient of the epithet XPtyao'itTepos appears to be I r i s , i n Il i a d VIII 398, XI 185 and i n the Hymn to Demeter 314. One finds itapde*vos l i t t l e used, i t s epithets being as follows: a 1.6 o i n , once in the Il i a d and twice in the Hymns; d 6 u t f s twice in the Odyssey and twice in the Hymns. From the context of Chamaeleon's remark on the two Palinodes, one would suppose that this vocative address was an introductory invocation, parallel to Sect (puXduoXue. The latter i s undoubtedly a reference to the Muse, in accordance with normal epic practice of calling upon the goddess of inspiration. We have evidence from the citations that elsewhere Stesichorus invoked the Muse by various t i t l e s . According to Atheaaeus ( V 180e = PMG 250) Stesichorus called the Muse ctpxectuoAitov, presumably in an invocation and doubtless derived from the tradition of addressing the Muse at the beginning of a song (although equally possible i s the meaning "queen of sang"). Moreover, Eustathius, on the f i r s t line of the I l i a d (9 43) remarks that not only Hesiod, but also Stesichorus began his poems with an invocation to the Muse: 6e0p'aye? KaAXidneia Auyeux (PMG 240). Stesichorus was apparently aware of the tradition found i n Hesiod, but not in Homer, that distinguished Calliope 20 as chief of the Muses and patroness of epic . None of the fragments-from the papyri, however, contain what may be identified as an exordium of a poem, but there i s season to believe that the lines of Aristophanes' Peace called Stesichorean by the scholiast , namely 775 f f . , were composed as a parody of the invocation from one of Stesichorus' poems, thought by 21 some to be the Oreateia : Motaa au yev itoAe*yous ditobaayeva yet' eyou xXetotaa $euiv te yayous dv6puv xe 6a£xas xal fcaAuxs yctnoipuv .... (as arranged/in PMG,210). It i s noteworthy that there i s no decorative epithet such as Aeyeto or (puAdyoAitos, but the lines do attest the poet's movement away from the traditional invocation of the Homeric poems. The rejection of one topic for another more appropriate one i s more akin to personal l y r i c than to 22 the epic tradition Thus, in view of the above-cited stesichorean invocations, i t would not be surprising for Chamaeleon to have identified the two Palinodes by their invocatory f i r s t lines, nor that there were in these lines unrHomeric features. If xpuco'ittepe napdeve wasa intended as a periphrasis for the Muse of Stesichorus' poetry, then we have a new and unusual vision of the Muse. In the Homeric tradition epithets are rarely applied to the Muses. Their Olympian domicile i s indicated i n 'OAuynua 6oJyaT' e x o v x e s * or else their kinship with Zeus:in xodpas A b b s aiyioxoto . Mouaa Xbyeta occurs once i n the Odyssey and theee times in the Hymns, while a further description of their vocal talents appears i n the phrase MoCaat 6* evvea itSaat dyecBdyevat out xaXnt (pdyssey XXIV 60 and Hymn to Apollo 189). In the Hesiodic tradition the nature of the Muses and their method of inspiration i s enlarged, but the description of them nowhere mentions 23 golden wings . One could perhaps explain, the association of gold through their relationship with Apollo, who is known in l y r i c as xPuo*oxdyns» XPuaoxoSos or xPUo-cxpopytyS. In l y r i c the Muse is x p u a e a (Pindar) and XPuadSpovos (PMG 953 and 1023). Sappho summons the Muses : 6eupo 6nuxe Mouaau xP^'-ov ACnotaai, ... (PLF, 127), where xP^atov possibly qualifies 6dyov. Thus the use of an epithet compounded from X P U 0 0 ~ is not in i t s e l f exceptional, but the symbol of wings i s unprecedented. The M,uses are not responsible for enea itxepd'evxa , but rather for song that lasts 2 -*. Thus i f 25 Xpuao'itxepe i s intended to invoke the Muse i t i s certainly untraditional 11 • X;apbx<j)v ... xaXXixdywv PMG 212 1 The epithet xaXXdxoyos occurs once i n the Odyssey of Helen (XV 58) and once in the Ilia d of a concubine (IX 449), both in the long form of the genitive at the verse-end. The epithet i s also rare in the Hesiodic poems, although i t s application to the^flpat;; > Works and Days 75,is interesting i n that the Tilpau are closely related to the Graces in that passage. The Graces are not unnaturally renowned for their beauty, as comparisons such as Xaptxuv dyapuyyax' exouoav or Xapuxwv dub xaXXos exouai (fr. 70 38, 196, 215) which occur particularly inufehe Eoiai, indicate. Their overall beauty inevitably presupposes fine tresses, and the Gfcaces are described as eunXo'xayot, which is more or less synonymous with x a X X u x o u a i , , i n the Hymn to Apollo 194. Indeed, i n Ili a d XVII 51ff. the 'son of Panthous, who l i e s blood-bespattered in the dust, smitten by the sword of Menelaus, i s described as having had hair like that of the Graces, although i t is now befouled with blood and grime. Thus, the beauty of the hair of the Graces must have been proverbial. The word-group X o t p t T U V xaXXux6*ucov is therefore unprecedented, but not unexpected. 12. x ^ d v a u u p o q x J p ^ o v 2359 f r . l i i 7 The epithet that most commonly accompanies x^ova and; x^ovC* in 26 the Homeric poems i s i t o u X O B b x e C p a v / n t ,although x^dvo in the accusative occurs frequently without an epithet. In the nominative case, as was noted earlier, e u p e t a x^^v i s the expected "formula"* i t u p c x p d ' p o s is rare, appearing in the genitive with nedtouo (Iliad XXI 602), with dpoupns (Iliad XII 314) and i n the nominative plural with a p o u p a t (Iliad XIV 123), while in the Odyssey III 495 the f o r m i t u p r n p o p o s replaces i tupo<pd'pos to suit the metrical requirements! C ? o v 6* e s n e 6 L * o v i c u p n < p d ' p o v — u v / , — _ — < / > / — > . 27 instead of ... xat daotfpns n u p o t p r f p o t o . The limited application of this epithet in the Homeric poems presents no problem to Stesichorus, particularly since an extension of association from ue6C*ov to x ^ v a ? -/ entails no d i f f i c u l t i e s of logic. From the reconstruction of the colometry 28 proposed by Snell , i t would appear that x $ ° " v a i t u p o c p d p o v l i e s in apposition to u a p a v BOLU>TU6»J ~> the dwellers in the "wheat-bearing land" are i n fact the Boeotians. Consequently, i t is a l i t t l e odd that the poet chose the epithet n u p o < p o * p o s , which implies cultivation, when n o u X u B o T e c p a , which implies grazing land, would have been more appropriate, given that the origin of the name B o u i)Tt * a l i e s in the fact of her having cattle-pastures. However, the epithets distributed throughout this small portion of the l i s t of participants i n the hunt on the whole create new word-groups, alien to the traditional "formulae" of the Homeric poems., In I l i a d V 710 the Boeotians are called the earners of nuova 6fiyov,which as a geographical description i s not very specific and.therefore allows a certain amount of latitude to later poets i n their choice of descriptive epithets. That Stesichorus had l i t t l e or no knowledge of the terrain of Bceotia i s possible, but irrelevant from the standpoint of li t e r a r y compositions. b) Noun+epithet groups containing supplemented readings 13. ctyyov a[yctx) ua 2619 f r . l i i 10 The word SyctXyci seldom occurs i n the Homeric poems, i t s frequency being limited to once in the I l i a d and 7 times i n the Odyssey,• and i t s accompanying epithets are not distinctive: v£ya and uepuxctXXe's, and i n the plural, itoXXa xal iaQXd . An ayaXya, cognate with dyaXXw, may be something that brings glory or delight to i t s recipient or owner, or i n general to the gods. Thus, the ivory cheek-plate for a horse i s described as a work of art: ayaXyct, dyipdrepov x day os &#X0b eXctTnpC* te x u 6 o s (Iliad IV 144,145). However, the association of ctyotXya with the gods i s indicated, for example, in the description of the bu l l dedicated by Nestor to Athena: t v A dyctXyct %ea xexctpotTO tdouoct (Odyssey ;III 438). The fact of the bull's being sacred or dedicated to the god seems to have been implicit in the noun dyctXyct i t s e l f . In such a case any epithet such as tepdv or dyvdv would be tautologous. The epithet dyvrf i s generally restricted to the goddesses Artemis and Persephone, i n the Odyssey and the Hymns. In such instances i t has been assumed that the epithet must mean "chaste" and certainly i t s later use in Alcman and Pindar, referring to "maidens", would not contra-dict this view. If, however, in this instance, one interpets the epithet "revered" or "commanding due respect" as a god or as belonging to a god (cf. i t s association with a grove dedicated to a god, or a fes t i v a l in honour of a god) then the combination of dyvdv and S y c i X p a may be con-sidered logi c a l , although as a "formula" i t i s unprecedented. In Aeschylus' Eumenides 55, deuv dyctXpctTct appears to mean " images", and by Herodotus' time the word ayaXpct had come to signify an actual statue of a god, dedicated to the god (cf. Histories I 131 and II 42). At some point therefore, after the age of epic, the word ctyctXpct gained a more concrete meaning i n the Greek language, and i t i s possible that the meaning of"statue? was implicit or understood in the usage of dyvdv ctyctXpct i n Stesichorus' poem with reference to the wooden horse, the latter being an image created by the Greeks at the instigation of Athena i n order to make their f i n a l assault upon the city of Troy by guile, and dedicated to Athena by the unwitting Trojans. Perhaps the epithet dyvdv was deliberately used i n an ironical fashion: what the Trojans believed to be an innocent, sacred g i f t to the goddess was i n fact the source of their destruction. It i s also interesting that Tryphiodorus in his Capture of Troy uses precisely this phrase , i n an almost identical context of the Trojans bringing the horse into their c i t y . Either Tryphiodorus borrowed the phrase directly from Stesichorus, which would be plausible i f Tryphiodorus 29 were writing in the 2nd or 3rd century A.D. in Egypt , at a time when the papyri of Stesichorus' poems were apparently being circulated there, or else both poets had access to a common source. The epithet dyvd's also occurs in P.Oxy. 2619 fr.18 9, 'apparently as a t i t l e of Poseidon, who i s alluded to by the periphrasis 'Evvoo-tSct!? in this usage SEesichorus follows a practice common to both epic and l y r i c poetry which makes dyvdg the attribute of a god or goddess. There i s , however, no surviving precedent for the application of the epithet to Poseidon. Similarly, i n P.Oxy. 2619 fr.16 7, dyvjav appears in a context suggesting that the epithet may be applied to Aphrodite, and i f this be the case, then the accepted translation "chaste" as ." designated to Artemis and Persephone would hardly be appropriate. I t i s lik e l y , therefore, that Stesichorus has reallocated a traditional epithet, with some alteration of meaning, or else with a rather different concept 31 of the attributes of Aphrodite 14. d i t a X o v [ 6e*ycts 2617 fr.4 i i 16 In the Iliad,d i t a X d ' s i s found governing a u x t f v and 6 e t p n , that i s , a specific part of the body, the neck or throat. In the Hymns , d i t a X d ' s i s also applied to feet: n 6 * eoauxo i t o o a 1 drcaXobai, (Hymn to Demeter 287). I f Page i s correct i n supplementing 6 e y a s » which i s acceptable in both context and metrical scheme, the association i s new. 6ducts occurs primarily in the phrase e l 6 o s Te 6 e * u a s xe or else 6e*y<*s d d a v c f r o i o i 6 y o t o s,but there seems to be n o exact parallel for the pro-posed meaning o f a flower spoiling i t s "shape", the closest being xnv aiyv op xaTaco"xd'v« <pvJcrtv i n Sophocles, Electra 609. One of the necks of Gerypri has already slumped to one side (in line 14 of the fragment} and;:thus the closely associated epithet of auxtfv has apparently heen transferred from i t s expected juxtaposition '•„ with ctuxnv (as in Il i a d XVII49 =» XXII 327) into the framework of the simile, where i t describes the object to which Geryon's fallen head i s compared : /the poppy. The possi b i l i t y of ditaXds being applied tp plants / i s supported by an example i n Sappho, i n which she describes chervil as x * S u a X ' / ctv^puaxa (PLF 96 13), no doubt on account of i t s delicate foliage. I shall return to discuss the simile o f the poppy in chapter 15. e u p u ] x t f p [o]u Tpo<t>as 2619 f r . 15 (b) U e u p u * x o p o s describes c i t i e s i n Greece, and even Greece i t s e l f , but i s never associated with Troy, either in epic or in l y r i c poetry. Epithets for Troy i n the I l i a d and the Odyssey include: e u p e C n , eptfluXd?, e o p u c f f u o c t , e u x e t x e o s , e v k u p y o s and v 4 C i c u A o s . The combination of e d p d x o p o s with Tpou'a would appear to be a conscious departure f r o m the traditional epithets applied to the cit y of Troy, without actually losing the expected association of the supu-element, as found in at least two of these tradi-tional epithets. 16. uuxLva[t]s u t e p j t f y e a a t , 2619 f r . l i i 19 There i s one "formula" from the Homeric corpus whose sense is vi r t u a l l y identical to that of the given phrase, although there i s no actual precedent for nuxtvctts itTepdyeaao as such, itrepa nuxvcf occurs in Odyssey II 151, I l i a d XI 454 and XXIII 879, and in the longer form Ttuxtva itTEpd* i n Odyssey V 53, i n a l l of which Ttt)xtv6*s/ituxv6*s has the sense of "crowded together" or "closely overlapping" of plumage. The use of itoxtvds i n place of uuxvos in Stesichorus may simply be a matter of accommodation to the new demands of the metre, since in Odyssey II . 151 TtTepd i s used to signify "wind" rather than "plumage". It is possible, therefore, that one might consider this word-group under the category of Homeric parallels, since the law of mutual expectancy operates in this instance, with only the minor variation of form, a variation that may be explained as metrically necessary. c) Other types of word-group : noun+genitive of possession. 17. itj oXe'uou jje3 XEUTOI 2619 f r . l i 18 If we are correct i n assuming the supplement of the noun TeXeuxct then the word-combination i s a new one. In the I l i a d one finds, Sox example, 0L6*TOI,O xeXeuxn (VII 104, XVI 787) while in the Odyssey an instance without the accompanying genitive: r\ 6'our' dpvetxai, axuyepov Yduov odxe TeXeuxtfv itocfiaac 6u,vaxab (I 249 *» XVI 126). Synonymous with TtoAe\ioi; xeAeuxci would be xeXos TtoXe*you/oi,o, and although i t does occur twice in epic, this phrase i s far less fsequent than xeXos d c t v o f x o L O . Thus although the combination of xeXeOxd* and itoXe"you is unprecedented, i t i s nonetheless predictable in a context in which the end of the fighting i s considered. The expression i s almost prosaic, as i s witnessed by i t s use, for example, in Thucydides' Histories I 13. The phrase as i t appears in Stesichorus would be impossible to incorporate into the metrical scheme of the hexameter, and hence one finds in the Batrachomyomachia 303 the alternative xeXextf ; xcu, uoXe\tou xeXexn yovonye*pov e5exeXe*a$n. II Combinations of traditional and non-traditional elements, a) Non-supplemented 1. BacaXeus IIXeujdev£6as PMG 219 2 There i s no reference to Pleisthenes in Homer. Stesichorus, however,may have derived the tradition of Pleisthenes in the family-tree of the Atreides from the Hesiodic corpus, where one finds him recorded as 32 the eon of Atreus and father of Agamemnon and Menelaus . gaouXeds is generally found in" epic with a genitive of the place or the people over whom he rules: gaatXfio noXuxptfo"Oi,o Muxtfvng (Iliad VII 180). Comparable is the line ev 6e cnpuv "Pffaos gaatXeds, itctCs 'HCovrjos (Iliad X 431) but nowhere i s there a precise precedent for this use of gaoi.Xeu's followed by the patronymic i n apposition. It i s interesting to note, however, that the precise expression gctauXeus IIXei*adevu6as recurs in the poem in honour of Pblycrates attributed to Ibycus (PMG 282 21), a fact that gives some indication that Ibycus may have imitated phrases from his western -33 predecessor 2. uxpctv BouDTC*6la 2359 f r . l i i 6 Lepd*s 3 ^, when applied to place-names, qualifies c i t i e s rather than larger geographical regions, the most frequent being Troy i t s e l f (for example in I l i a d VII 20). Other c i t i e s so described are Pylos (Odyssey XXIV 108), Thebes (Iliad I 366 , Hymn to Apollo 226) and Pergamum (Iliad V 446). However, for metrical reasons, the shorter form tpds/tf i s more commonly found with'lXu'os A21 times i n the I l i a d and twice in the Odyssey). Boeotia in the I l i a d - i s referred to only indirectly by the name of i t s inhabitants (V 710) and the form Bo waft's i s rare, occurring otherwise only in Xenophon, Hellenica V 1 36. Thus the word-group uxp&v BowjiTuSa may be seen as a new formation in which the poet has employed a common Homeric epithet in an entirely new context, with a non-Homeric noun 3 5. 3. xei*Po3pwTt 6eauu>L PMG 180 In the Ili a d and Odyssey there i s no single, distinctive epithet of Secrud's: apyotXe'os occurs twice, as does xpotxepds. The epithet Xetpo&pwti, i s a unique and v i v i d composition. The construction of the compound is unusual in that the majority of compounds whose second element consists of -8pu»T-, "eating" or "eaten", belong to the thematic declension; for example ncti,6d'Bpu>Tos» xeqiaXdgpWTOg or q>$eupo'Bpu>TOs»none of which, however, i s found in epic, being impossible in the metrical scheme of the hexameter. Nor are compounds whose f i r s t element is based on xeipo-found in the epic tradition. Hence the word xei'PoSpSTL i s not only a 100 36 hapax legomenon but is highly unusual in i t s formation According to Zenobius, the phrase refers to boxing thongs, but is better applied to deauctv in the case of an individual bound to a rock. Since the commentator-s remarks are somewhat vague, and his allusion to the beginning of Stesichorus* Athla as l i k e l y to be guesswork as based upon factual evidence, we are l e f t in a state of uncertainty as to the original use of the phrase by Stesichorus. It i s possible that the poet described thongs as "hand/arm-gnawing" and at the same time drew a comparison with the bonds of Prometheus. 4. Sea" ( p t X c f y o X n e 2506 fr.26 i 10 (PMG 193) There has survived only one other instance of the epithet ( p t X d y o X n o s , namely in Pindar's Hemean VII 9, where i t described the island of Aegina. Compounds in -yoXuog are not common, but two others also occur i n Pindar, both referring to a Muse: epaatfyoXnos and < p u X n a t y o X i t o s . From the epic tradition in the Hymns X u y d y o X n o s (Hymn XIX 19) may have served as a model for (ptXo'yoXitos, although the Hymn to Pan i s thought to be later than Stesichorus and hence i t s evidence is of doubtful value in assessing the traditional material available to 37 Stesichorus . As was noted above, epithets for the Muses in the Ili a d and Odyssey are not frequent, whereas in Hesiod, and therefore possibly in a mainland tradition, there existed a greater variety of "formulaic" attributes of the Muses, whatever the epic:*background in this case, the occurrence of a compound epithet such as < p u X d y o X i t o s in choral l y r i c is to be expected in a situation in which the goddess i s invoked to preside not only over poetic creation, but also over musical composition. With the emergence of new poetical and musical forms i n the age of l y r i c , appropriate epithets had to be invented to encompass ttheiwider jurisdiction of the poets' patronesses, the Muses. 5. ditebpeauouo xuvuAayuoto PMG 255 aitetpeatos occurs infrequently in Homer, and seems to be confined to numerical contexts meaning "limitless" or "countless"; for 39 example in Odyssey XIX 174 , of men, or in Ili a d XX 58, of the earth . 6Cco*v i s described as "unending" in Odyssey XI 621, and i t i s in this sense of the word that 6ne must in a l l probability understand diteupeatos in relation to xuvuXctyvos • The word xuvuXayyd*s i s i t s e l f unique, constructed by analogy with oXoXuyvds, although the changes in vowels, xuvo- to xuvb- and -Xuy- to -Xcty-, are inexplicable other than as some peculiar dialectal variation of South Italian or S i c i l i a n Doric. 6. apxeotuoXitov fyoZoav] P_MG 250 The epithet dpxeaC*uoXnos is not attested \ elsewhere in extant Greek literature, and, as was noted in the case of (puXoyoXite above, compounds in -UOAKOS are few and with the possible exception of Xi/fu-40 uoXitos* belong to a non-epic tradition . In the epic tradition i t was customary to invoke the aid of the Muse or Muses at the outset of a poem, occasionally at certain important points in the course of the narrative, or else in the introduction of a poem within a poem, as in the case of the Catalogue of Ships in the second book of the Il i a d . In the Hesiodicv, corpus one finds that the invocation of the Muse had evolved into an elaborate proem such as at the beginning of the : .Eoiai (fr. 1 , M. & W.) and particularly at the beginning of the Theogony, where the poet sings a formal hymn in praise of the Muses before o f f i c i a l l y requesting their aid. While the situation of the Hymns is somewhat different in that the i n i t i a l invodation must be addressed to the particular deity being honoured, as in AnunTP1 ntfxoyov, creyviYv de6\>, opxoy' deC*6ecv (Hymn to Demeter 1), several of the Hymns also incorporate an address to the Muse, as for example in the Hymns to Hermes and to Aphrodite , XIV, XVII etc*. However, the increased frequency in the Hymns and presumably in later epic, of the "formula" opxcy1 deu'6etv seems to hint at a movement away from the assumption that the Muse alone was in control of the song about to be performed. Nonetheless, tradition did exert sufficient influence that the Muses were never ousted from a place i n the f i r s t lines of a song, as we see in Alcman and the later choral l y r i c poets, and i t seems that the epithet dpxeaJyoXitos specifically reflects this tradition. As was noted i n the case of cptAdyoAitoG, the invention of new epithets for the Muses was also determined by their expanded sphere of influence. 2. dvu<|)OAov itau6a PMG 249 This epithet i s not found elsewhere in Greek, and the verb » 41 *uitTouctt i s not common . we noted in the previous chapter that there was a prevailing tendency in the I l i a d and Odyssey for rcaus and i t s oblique cases to appear devoid of epithets,us6 that dvtf<J>aAov %aC6a i s non-Homeric. b) Noun+,iepithet groups containing supplemented material. 8. Xpuacfopos ^|^avd*TOLO 2617 f r . 13 3,4 The conjectured supplement Xpuaaopos'ddavtitTOLO i s reasonable in view of the context in which a second&speaker ( xovjjd' ditay [eb3dyev<osj| KOTeVx line 2f.), whose parentage is given (genitive dSavdfTOUO indicating "son of immortal..."), discusses the relative merits of l i v i n g as a coward, or risking almost certain death and refers to himself as Xpoa [aoj 90 [s uJ tdyin line 24 of the same fragment. In the epic tradition ddcfvaxos, as an epithet, occurs primarily in association with the word deoC*, but more often occurs as a substantive synonymous with %eoC, "the 103 immortal ones". Only occasionally in the epic poems does the epithet qualify an individual: Zeus for example i n Iliad. II 741, or Proteus in Odyssey IV 358. The less common association of d§d*vaxos with a second epithet dYifpods* as a double adjectival phrase, i s occasionally applied to an individual, as for example to Calypso (Odyssey V 218) and particularly to Heracles in the Hesiodic tradition (fr. 25 28 M.s w.). This phrase does i n fact occur in Geryon's argument (line 9 of fr.13 ) and i n view of the importance-' of the immortality or otherwise of Geryon, the use of dSofvaxos as an epithet for his father i s significant. Geryon is apparently unsure whether he i s immortal or not. The poet indicates that his father certainly was, but divine parentage was not enough to guarantee immortality, since the state of being ctddvctxos w\> dyrfpws had 42 to be conferred upon the individual by the gods,. . Heracles himself bewails the fact that despite his being the son of Zeus, he has to suffer with the rest of mankind i n the gloom of the Underworld (Odyssey XI 601ff In the later tradition of the Hesiodic Great Eoiai, however, we find that Heracles has been granted immortality and dwells among the gods. The latter tradition i s the more generally accepted one, and i t would be more logical on the part of the poet in this passage to strike an indirect comparison between Heracles and Geryon; both are of divine parentage at least on one side, and yet by the w i l l of the gods Heracles w i l l succeed and gain immortal glory, while Geryon i s doomed to be defeated, and v presumably destined to l i v e apart from the gods, in the Underworld. 9. dA&Jauyov 5uap 2619 fr.15 (b) 11 Formulaic expressions with the noun fiuap in Homer present a large array of epithets, the most frequent association being observed in the,.wo?d-group voaxtyov' fiuctp, which describes the long-awaited day of return for Odysseus after his wanderings (11 times i n the Odyssey). Of the other word-groups, two are based on an almost identical pattern of vowel-sounds;>! namely atatpov fiyap (3 times in the Il i a d and once in the Odyssey) and uopauuov fiyap (once i n the Odyssey). aXuJauyos i s a post-Homeric epithet, constructed as other adjectives derived from abstract nouns in- -us, for example?:- xP'fa'-Vios from XPfioxs. Thus SXcooxs, which i s also post-epic, produces the epithet aXoJotyos,whose usage i s primarily confined to the dramatists .and later. One early instance of the epithet 43 occurs i n the poem to Polycrates (PMG 282 14) : Tpqjtas U4>UIU"XOLO dXaJai i^o] v J 8y| ap dvwvuyov. The poem constitutes an extended praeteritio: in which the poet rejects a number of epic themes related to the Trojan cycle. One finds a series of expressions based upon Homeric "formulae", and as i n the case of Stesichorus, some in which the word-association 44 i s not Homeric . In common with phrases found i n Stesichorus i s £ a v d a s 'EXevas and a reference to the Pleisthenid dynasty to which 45 Agamemnon belongs . I t is interesting to note that the ancient commentators were sometimes confused in assigning poems or expressions to either Stesichorus or Ibycus, as in the case of the Athla (Athenaeus IV 172d = PMG 179). The epithet atbAo'oetpos is attested for Ibycus (PMG 317 a) and has now appeared in one of the papyri of Stesichorus (2617 fr.4 i i 5). The recurrence of both Stesichorean £ a v d a s 'EXevas and dXaScruJov Syap in PMG 282 suggests that Ibycus, or an imitator of 46 Ibycus borrowed phrases from Stesichorus./ In the Homeric poems the end of the war is contemplated i n terms of Tiep&u)/uepoLS; for example, Achilles in Book I of the I l i a d declares: 6^t1to'T, 'Axauot J TpoJtov exitepawc1 euvacdyevov inroXteSpoy (163,164). Else-where there are two common "formulae" for the end of the siege and the capture of the cit y : Te"Xos itoXe*uouo (although cf. page 98 above) and Texuup ' IXoou ( 9 times in ffehe I l i a d ) . The context of the phrase i n Stesichorus i s a reference to the ultimate plot to capture Troy by means other than open warfare. The man inspired in cunning by Athena (line 6) could be Epeius, the inventor and builder of the wooden horse that was to bring the Trojan war to a close. Such a theme is beyond the temporal scope of the Iliady and, moreover, the sections of the Odyssey describing the f i n a l capture of Troy do not dwell in great detail on the these. I t i s therefore not surprising that Stesichorus did not adapt.any known phrase from the Homeric corpus. One should note, i n addition, that the expression for the city of Troy, eupux<5p°u TpoCag employs a Homeric epithet in an un-Homeric association. Thus Stesichorus either follows another tradition for the "formulaic" phrase that expresses the capture of Troy, or else fabricates a new phrase on the basis of other epic word-groups such as VO0TL.UOV ?|uap, and i n the process introduces a new association of the epithet eupdxopos with TpoLot. The poet of PMG 282, on the other hand, in his use of dXwcruuov 5uap, employs a phrase for Troy that i s found i n the Homeric poems with reference to i t s capture: U(|»utuXov T p o u i v i s found i n such a context i n Ilia d XVI 698 and XXJ 544, which appears to indicate his preference for maintaining certain appropriate associations in an intentionally Homeric passage. In the Stesichorean fragment, from a metrical point of view, the poet could have employed u<J>i.itd'Xou, but> as is found in other examples, appears to have preferred to avoid the expected epithets. The poet of PMG 282 may have been motivated to repeat some of the well-known "formulae" verbatim in order to make his f i n a l rejection of the composition of epic poems 47 more emphatic, or perhaps the repetition was simply unconscious 106 Stesichorus, on the other hand, was consciously striving to avoid repeating "formulae" verbatim. 10. d p n w p t A q j v X o j j x r d ) o p a 2617 fr.25 4 This epithet, whose singular form i s used almost exclusively of 48 Menelaus in the I l i a d and similarly in the Hesiodic corpus , i s in this fragment of Stesichorus given to Geryon's father, Chrysaor. Elsewhere, Chrysaor i s u£yas(Theogony 281) or x p a r e p d s (Theogony 979) and hence this Homeric epithet dpnifqiLAos that belongs to a glorious warrior i s not totally unfitting for one who was "born with* a golden sword", a son of Poseidon. The combination of dpnC*<puAov with X p u c c f o p a i s nevertheless unprecedented, and doubtless the connections of the epithet with Menelaus imprinted i n the minds of the audience would have caused them to make a subconscious comparison between Chrysaor and Menelaus. Unfortunately the context of the fragment does not permit us to determine whether this comparison was intended to be taken seriously or i r o n i c a l l y . 11. T^puiLag xAeeyvot _ 2619 fr.32 7 Both i n this fragment and i n 2619 fr.15 (b) the spelling of 49 Tpwuxgand*Tptoas with omega has been transmitted. West , assuming that the iota has simply been omitted i n the latter.,: -supplements T p u C a s and grants that for his. metrical scheme the omega must be scanned as short before another vowel. Pindar, according to LSJ»used the Doric (?) form Tpuitas scanned as t r i s y l l a b i c with a shortened omega, but in their editions • 50 both Bowra (OCT) and Shell (Teubner) read T p o t o s and i t is this convention that Page has adopted in SLG 89 and 118. It is, however, impossible to ascertain what the original form may have been. Regardless; of the alternative orthography,the text reads as a genitive singular feminine,, of the word for Troy, and i t i s apparently qualified by x A e e y v c t p " Although Stesichorus does not always confine himself to the Homeric practice of allotting a single epithet to every noun, the evidence of the fragments tends to confirm that a noun+ single epithet was the commonest type of grouping i n Stesichorus. In 2617 f r . 32 one finds an aorist participle, plural "having set f i r e t o j . . " followed by the epithet euxxt.ye[y-, which is frequently asscfiiated with UToAtTeSpav. itxoXuedpov, i f i t occurred, could have been preceded by a genitive of the name of the city, as for example in 'iXc'ou exire*paon, euvatdyevov 52 UToAi'eapov (t l l a d II 133) and i t would be natural to expect an epithet with Tp(pCas rather than a second epithet with the noun that is already qualified by euxxcyevds. Hence Tpwucts xXeevvas-53 xXeuvds or xXeevvd's i s a post-Homeric epithet, whose epic equivalent would be xXoxds or xXeuxde. Neither of these adjectives,however i s used to qualify c i t i e s i n the epic tradition. xXuxcf is commonly associated with 6c5yoxo in the Odyssey, with xevJxeot i n the Il i a d , while xXetxds occurs in the Ilia d alone»with exaxdygn or eTttxoupot,. It is only in the later tradition of l y r i c composition that xXuxds and xXetxds are even rarely applied to c i t i e s . xXetvds , on the other hand, may have been introduced to supply an epithet with the meaning of "famed" or "renowned", cognate with xXdos "glory", but without the associations of,for example, xedxea or duyaxa inherited from the epic tradition and s t i l l employed in those particular contexts. 12. xavuicjY|itXou 2359 f r . l i 7 The name of the person who possesses this epithet has not sur-vived the mutilation of the papyrus, but from i t s context, that of the poem the Suotherae, i t is unlikely that the person was either Helen or Thetis, who alone receive this epithet in the Il i a d and the Odyssey. 108 Thus we have here an instance of the use of certain epithets outside the Homeric corpus for prominent figures other than those with whom the epithet i s habitually associated i n the Homeric corpus. The non-Homeric word-group in this case may have been derived by the poet from an epic tradition that was distinct from the Trojan cycle, but may equally have been invented by Stesichorus. I l l Longer units that constitute expansions of simple noun+epithet groups.or noun+genitive groups. a,) Combinations of noun+epithet and noun+genitive. 1. a ^ u a S 1 dXos B/J^ eas 2617 fr.6 (a) 1 xuua/xuuaxa i s found twice in conjunction with dX6*s in the Odyssey (XII 68, XXIII 387) and once in the Ili a d (VI 136). The more common genitival extension of x u u a i s , however, with daXdoans (twice in the Odyssey and 12 jtimes in the Iliad) . Thus the use of dXo's in this fragment appears to be an alternative synonym that i s less stereotyped. Epithets that are expected with dXds are dxpuyexouo (5 times in the Odyssey and 3 in the Iliad) and TtoXtns/ooo (6 times i n the Odyssey and 10 in the Ili a d ). Although BaSetns occurs once as an epithet of dXo's (Iliad XIII 44) the notion of depth is also conveyed by the expressions containing ge*v§os, obviously cognate with Baku's and therefore partially suggestive of i t : K a r a Be*vSos dX6*s (Iliad XVIII 38, 49) and ev ge'vSeaox dXdfs (Iliad I 358, XVIII 36) . Strangely, LSJ think that a X s generally i indicates shallow waters, near the shore. Although Odyssey IV 270 des-cribes a ship being dragged down to the shoreline, the ship would shortly have entered deeper waters (and in any case, the shores of the Greek coast shelve swiftly^' and steeply into quite deep water). Moreover, not only i s there the example of gctdeuns e? dXd*s cited above, but also expressions such as dXds pnyutva Badetav (Odyssey XII 214) in which Badds/eta i s associated with a X s , although not directly qualifying i t . Thus there are precedents for the use of xtfuaS' with dXo's and of Badeuxs with^dXos, neither of which "formula" i s the most frequent combination of these individual words. The resultant grouping of a l l three elements may be considered as an innovative expansion that takes advantage of the expectancy of two particular associations with d X o s ; element b) d X d s 54 generates the expectancy of either a) MO'tiad* or c)8a§euxs . In the epic tradition we find either the one of the other; Stesichorus has combined both. 2. <pt'Xou itaxjpjos "toy 2360 f r . l i 11 (PMG 209) The Homeric a f f i l i a t i o n s of <pu'Xou itaxpds were considered above and i t was also noted that the combination of <pdXov with"'yud*v was more 55 common than the equivalent expression with itatSa . I t appears that in this instance the poet has adopted the elements of two "formulae", <pdXos with nctxrfp and q>L*Xos with u t d s . By conflating the elements of both he has produced a phrase not i t s e l f found in epic, but again with strong epic associations. 3. (ixea xe*xva IIo6apYas PMG 178 1 The epithets most expected with neuter plural xexva are: v i f a t a (11 times i n the I l i a d and 3 in the Odyssey) and dyXact (twice in the I l i a d , 3 times in the Odyssey and 4 times in the Hymns). The epithet dxees occurs in frequent combination with C u i t o t , particularly in the Iliad (28 times) Thus Stesichorus appears to have transferred the epithet commonly found with t n c o t to horses who are identified as the xexva HoScfpyas . The poet.,. has also named these horses as f X . d y e^v and "ApitaYOV,|,which differs from the Homeric version in I l i a d XIX 400, where two horses called \5dvSos and \ 110 Bet ALOS are said to be the xe'xvct ITo6d*pYns. b) Noun+epithet groups with verbal element. 4. es ctAaos ... 6ct<pvctL0"L xaTctjoxtd'ev PMG 185 5 (SLG 17 8,9) xctTctaxto\>,the traditional reading of the MSS of Athenaeus, cannot stand in the revised colometry of the Gfiryoneis^^. Barrett 57 suggested the alternative xctTaaxodev , which could have been mistakenly written as^xaxcfaxLOV by the simple omission of the epsilon in the course of transmission. The epithet xctTctaxuoslis not unknown, though rare. It is applied to bepuct in Hesiod.'s Works and Days 513 and, more interestingly, is the epithet describing the courier in Clytemnestra's speech in the Agamemnon 493: the man is darkly shaded with olive branches,xnpux' ... xotxefoxLOV xAcf6ous eActtcts. The metrical scheme of the Geryoneis ,however, demands an added syllable and hence the lines read: 6 6 ' es ctAaos egct 6ct<pvctLaL.;xctTct-oxLoev uoat itctus Aubs L ~ u u ~ ^ (SLG 17 8,9) The compound form xataaxtdeLS does not occur in Homer, but simple oxi,o*eLS is used in a variety of contexts eeferring to the casting of shade particularly with opect (Iliad I 157, Odyssey VII 268 and also Hymn to Apollo 34, Hymn to Hermes 70, 79). The epithet conveys not only the sense of the mountains being covered by trees with the result that their slopes are shaded from the sunlight, but also the connotation of the shadows cast by the mountains themselves, axedevxa occurs only once with ctAaect, in the Hymn to Aphrodite 20, and hence we might consider the association of xctTCtaxud'ev with aAaos in Stesichorus as being precedented in epic only indirectly, noting however that the extension from the simple to the compound form i s significant and not merely a metrical necessity. In the Ilia d and the Odyssey ctAaos generally attracts the I l l epithets dyAdo'v or xAuxdv , neither specifically indicating the physical aspect of the grove. In the Hymn to Apollo, however, there are six . instances of the phrase a A a e c t 6ev6ptievxa (at for example line 76, 245) where the epithet may seem redundant , but may also convey the, impression of the density of the trees i n the grove. When one turns to Odyssey XX 278 one finds a closer parallel to the Stesichorean phrase: . . . . x o\r 6' dyepovTO x c f p n x o u o u i v x e s 'Axcti/ot c t A a o s u i t o axuepov e x c t x n B o ' A o u ' A i t d A A w v o s • The grove of Apollo is axtepdv, an adjective that i s cognate with axudets, but there is no specific mention of the type of trees that cause the shade. Hence, although the precise phrase used by Stesichorus is not found in the epic corpus, i t is possible that phrases such as the one cited from the Odyssey XX and the one from the Hymn to Aphrodite ^exercised some indirect influence upon his composition*. Intimately connected with xctxctaxtdev i n the Stesichorean fragment is the dative 6d*<pvcti,aL: bay trees were responsible for the gloom in the grove. The laurel or bay was sacred to Apollo primarily in his capacity as god of prophecy, as in Hymn to Apollo 396 and in later literature, 59 but in Hesiod i t belongs to the Muses alone . These goddesses, Hesiod t e l l s us in his hymn to the Muses at the beginning of the Theogony, presented him with a branch of laurel symbolising the g i f t of poetry. Apart from being the symbol of poetry and prophecy, the laurel was early recognised as useful in more mundane matters. In the Hymn to Hermes 109 that god.is discovered using a stick of laurel to make f i r e , a function of the laurel that is mentioned by Theophrastus i n relation to i t s hardness and resistance to wearf^. One assumes that the advice of Hesiod for the construction of a plough with poles of laurel or elm i s related to this same fact of durability (Works and Days 435). The only reference to the laurel in Homer, however, comes from the description of the Cyclops' cave, which i s U(j>nX6\>, 6d<pvni,cru xaxnpe<pe"s, "high and shaded over with laurel" (Odyssey IX 183) . An evergreen, the laurel has dark, luxuriant foliage and hence i t created a natural vaulted canopy at the entrance of the cave. Stesichorus' phrase 6d'<pvabCfL xaxaaxcd'ev conveys the same notion of foliage casting darkness down over the hero's head as is found in Odysseus' impression of the Cyclops'cave. One envisages a similar setting for the l a i r of the Pytho at Delphi, as i t is described by Euripides, i n Iphlgeneia Taurica - 1245 f f . : S S l , UOLXOXOVHOXOS 0L .VOU0S 6pCtXU)V axtepctC xaxd*xaXxos evuptiXXwu 6ctcpvat y a s iceXcSpuov xe*pas . . . Euripides succinctly conveys the image of the murky place through the epithets axcepds and eu<puXXo$ in sharp contrast with the glittering scales of the monster. Although the reference to the laurel is intentional on account of i t s association with Apollo, who vanquished the Pytho and assumed sovereignty over Delphi, the connotations of darkness in the image must also be significant. The traditional association of the laurel and darkness i s also repeated in the epithet ueXctucpoXXos that Anacreon uses to describe the laurel (PMG 443). I conclude,therefores that Stesichorus i n his creation of the word-group aXaos 6d<pvcsLOb x a x a a x t d e v calls to mind the traditional association of groves and shade, but also diverges from that tradition. Groves and laurel are both firmly linked with Apollo in the Hymn to Apollo and the former also in Odyssey XX 278, but Heracles can hardly be entering a grove sacred to Apollo as he descends from the Sun's cup onjto the island of Erytheia i n search of Geryon. The gloom of the grove i s perhaps intentionally set in antithesis to the b r i l l i a n c e of the Sun's golden cup. It is noteworthy also that the force of the prefix x a x a - . suggestive of something that hangs down over one's head, just as in the case of MCtTripe(pe*s describing the foliage that hung over the entrance to the Cyclops' cave, creates a rather sinister atmosphere. The image of the darkness that i s about to envelop.. the hero foreshadows the impending en-counter between the hero and Geryon, as the hero disembarks from the Sun's glowing cup and thus leaves the brightness of day^*. 5. H(OIYXP3 daea S^jua] x' exovxt 2617 fr.6 (a) 3,4 Earlier i n the chapter I considered the originality and 62 appropriateness of itaYXPdaeci as an epithet for 6o Juaxa . S t r i c t l y speaking the verb Ixovxu plays an integral part in the phrase, based as i t is on the "formula" 'OXdyitux 6<duax' exouai.. Thus' in the entire phrase we have a non-Homeric combination of itctYXPdcrea with 6 i o u a x' in juxtaposition with Homeric ... 6a$uctx' exouau, the latter word,however, having been translated as i t were from exouau to exovxt. 6 . 6Jotto)u itupo xctuoueyC 2619 fr.14 8 Unprecedented i s the association of the participle xcadyevos with the noun itup qualified by 6n'tov. MCov as an epithet of nop is much less frequent than the combination with ctxofuctxov or §e critique's, and the word-group i s not found i n the dative case in the Homeric poems, itupt in the dative, without accompanying epithet, regularly appears with the verb eunprfda (cf .Jitpnaavxas in the following line of this fragment) , although i t s association with some form of xctuo is not unprecedented: cpn itopt mediaevos (Iliad XXI 361) or oitdx* d\> Tpodn uaXeput itupt itoaa Sctnxai, xcaouevn (Iliad XX 317 - XXI 376) . West conjectures itdXu after exe'Xeuae in line 7, and supplements xacoyeyLca in agreement with i t . If this were correct, then i t would appear that some individuals have been ordered to set f i r e to the city, and that the city i s Troy, as in the lines of the Iliad quoted i n the previous paragraph, may be deduced from the reference to Helen in line 5 of the fragment. Page, on the other hand, i s perhaps overly cautious in his doubts that the fragment belongs to the same poem as fragment 1 64 and 15 on the grounds that the metrical schemes are not identical Whatever the text may have been, i t is clear that Stesichorus has preserved and combined the double a f f i l i a t i o n of nup with 6tftov and with xacdyevos, in the same manner as example 1 above. The occurrence of eiJJ itpncrdvTCts in the following line suggests that perhaps the poet ,was^  X 65 sometimes guilty of redundant expansion. 7. xcfpa BegpOTwuevqs axpov PMG 219 1 The perfect participle Begpoxwuevos occurs only once in Homer, in a description of the blood-stained armour of the dead souls in Odyssey XI 41. The association of?'5.xpbv*with xctpa is unprecedented, although suggested by the use of the epithet axpoTaTifv. with xoputprfv, which as we saw in the previous chapter was the model for the poet's decription of 66 Geryon being wounded by the arrow of Heracles . In this phrase, however, xcfpa, applied more commonly to the head of a man rather than to an animal, especially in the phrase xtifpn xoudwvxes 'AxatoC, emphasises the human association of the snake, which in the dream of Clytemnestra 67 represents the murdered Agamemnon . Thus, although there are precedents for the appearance of a bloody snake in epic (for example in I l i a d II 3G8 or XII 202), the poet has created a unique expression here, employing conventional associations to underline the double nature of the snake. c) Groups linked by xe ... xat. 8. gt'au xe xal acxuat j . . . neitoi^d'xes 2619 f r . l i 6,7 In the Homeric poems one finds examples of neit0L86*xes appearing with, i) abstract attributes such as dXxt*, xcfpxeuor 0$e*vei ,~ (for instance i n I l i a d XVII 329) , or else with i i ) tangible things: a) parts of the body, uoatv or xe^peaat and b) weapons, which are an extension of parts of the body, as i n xdSotaLV xdu euaxpecpeE olos daJxoju . . . T t E i t o u f r o T e s ;tll±ad XIII 716,717). Although the juxtaposition of type i) and type i i ) with iteitouddxes i s without parallel,in I l i a d III 431 8C*ni, xal xepat xal f y x ^ are grouped together with the comparative c p e p x e p o s , which does provide a precedent for the combination of abstract and concrete i n a similar context of warriors trusting inutheir physical '68 strength as well as their weapons* . 9. dt^paxd xe xal gpoxdeyija ueXea .2617 fr.4 i i 13 As far as the evidence from the epic corpus indicates, SoJpn£ is never found i n conjunction with another noun. The supplement ue*Xea is not based upon any Homeric pa r a l l e l , sinae the epithet Ppoxdevxa in the neuter plural i s regularly associated with evapa (8 times without exception) . Whatever the noun that followed Bpoxdeyxa was, the word-group thus formed must have been un-Homeric. 10. TdSies noAe'es x 1 entx^ou^poc 2619 f r . l i i 7 Since in the Ili a d the word eTiL*xoupoi 1"alliea", i s restricted to a context linked with TpCes, this phrase follows the Homeric pattern of lines such as : xexXuxe ueu, Tpues xat Aapdavot, n6' eitt'xoupou (Iliad III 456 = VII 348 * VII 368 = VIII 497) or TpSes UTtepfl^uoL xnXexXeuxoL x' eitCxoupou (Iliad VI 111 * IX 233 = XI 563). In such cases, however, the most common epithet of eiuxoupot is the compound xnXexXeuxoL (together with one instance each of noX0*xXeoxou and dyaxXeuxoL, and occasionally the simple form xXeuxot). One reason for the change of epithet in Stesichorus 116 was the metre, since the length of his line required a shorter unit than the Homeric "formula". uoXe'es i s seldom used with any great significance in the epic poems, as for example in the phrase noXees itep edvxes (Iliad V 94 etc.) where i t i s only natural that a hero would be pitted against countless odds. Stesichorus therefore in this phrase appears to have modified the stereotyped epic "formula" to suit his metrical scheme and i f there was a special significance in the choice of itoXe*es i t cannot be determined from what remains of the passage. 11. iteipopulylyevos a£yctx[b . . . . . ] . . [ . J v T e XoXou 2617 fr.4 i i 3,4 The association of cpopi5ao*u) with auuctTu, meaning "defiled with blood", occurs only once in the Odyssey: <popO*£cts c t C u o x i , itoXXwu (XVIII 336). There are, however, several instances of the perfect participle of the related verb <pO*po) that may be considered as indirect precedents for this phrase; for example netpupuevos o t C u a x u uoXXaiu (Odyssey IX 397) . One may compare this also with afuaxc xa*l XtfSpuJiv neflotXayye'vov (Odyssey XXII 402, XXIII 48). atyaxL must have been followed by an epithet that qualifies xoXat, but since in the I l i a d and Odyssey the masculineixoX6*s alone occurs, predominantly in the metaphorical sense of "anger" as opposed to the l i t e r a l sense of " g a l l " i n this fragment, i t is unlikely that the traditional epithets associated with x°A°s*namely dpyaXe'os and S y p t o s , 69 occurred in the lacuna . Doubtless the epithet chosen by the poet indicated the impending pain or even death that the X0*" would cause to the person pierced by the fa t a l arrow-tip. None of the epithets later linked with X°Xif w i l l f i t the lacuna, either, in number of letters or metrically, so that, whatever the poet used, the resultant combination must have been untraditional, just as the overall description of the poisoned arrow has no surviving precedent. In this word-group, therefore, one can observe the expansion of a known "formula", ite<pvJpye\>ov atyaxu ito\Au)t,by means of the new juxta-position of atuctTL and x°Aat, together with what I assume to be the introduction of an untraditional epithet to qualify xoAaE • d) Larger units of noun+epithet groups. 1 2 . v' Eapav Bocaixo'dfa v}ao*ov X&dva n u p c x p o p [ov J 2 3 5 9 f r . l i i 6 , 7 (PMG 2 2 2 ) I have already discussed under separate entries the noun+epithet groups uxpdv Bot,wxt'6a and x^ova Ttupo<pd*pov , observing that both constitute new word-groups based upon unprecedented juxtaposition s of elements from Homeric "formulae". As a structural precedent for the entire phrase one may cite I l i a d II 5 3 5 : o u " vct^duou it£pr\v i e p n s 'Eugouris. In this line C e p d s is applied, unusually/ to a geographical area, as i n the Stesichorean li n e . Both passages belong to catalogues that indicate the provenance of the participants i n a contest or battle. Hence iLt i s possible to consider voC*ov as "formulaic". In I l i a d II the expression with' ua £10 occurs 7 times in the Greek l i s t and 3 times i n the Trojan l i s t . It is noteworthy,' however, that forms of exu or vdyw/qyai, in similar contexts are far more numerous. Thus Stesichorus' expression has "formulaic" overtones, but the entire clause, with i t s double noun+epithet group describing Boeotia, is an expaasion of the type of phrase that occurs with vctC*w in the context of a catalogue, as well a new application of traditional epithets. 1 3 . axdcptov ... 6e*itas eyyexpov osz xpLActyuvov PMG 1 8 1 1 The close association of XP^ a e o v with Senas, at least in the dative case, was noted in the previous chapter, although the most common "formulaic" expression i s undoubtedly bims dyipi.xd'TieAAov (8 times in the I l i a d , 6 in the Odyssey). The meaning of ducpLHurceAAovwas ambiguous even to scholars in antiquity"^. I f Stesichorus understood the epithet as meaning "with a double cup", connected with xO*iteAAov,one of the many words for cup,then i t i s possible that ihe coined the adjective ax\5<poov as a play on words, deriving the epithet from another word for "cup",axu*(pos, just as the adjective regularly associated with 6e*itas had been. An interesting explanation is suggested in Collinge's theory 71 that Greek 6e*itas derives from Mycenaean di-pa, . &£%as, he maintains, was originally a large type of vessel for storing liquid, and although the semantic f i e l d of the Word had changed by the time of the Homeric poems, there do remain two vestiges of the older use of the word, not as a cup to be held i n the hand, but as a large mixing bowl or cauldron. The famous cup of Nestor i n I l i a d XI 632 and the witches• cauldron of Circe in Odyssey X 316 are containers of the latter sort, in which one mixed various ingredients and boiled them. Nestor's cup i s used for the pre-paration of a wine-and^cheese concoction, CirceJs for an immortalising potion for Odysseus: both of them are golden, as i s to be expected in the heroic or divine context. In this fragment of Stesichorus, the centaur Photos offers Heracles a drink from the 6e*itas that is described as ajtdtptov. Collinge suggests that^in fact the centaur presented Heracles with a mixing-bowl (comparable to the xpciTTip one finds in the Theocritean version in Idyll VII 150) and that the epithet was necessary to indicate that the bowl was presented in li e u of a cup. As i s not uncommon in Stesichorus' style, the noun deltas in fact receives a second epithet, euuexpov. This epithet, like axdtptov, i s unique, but in perfect keeping with the image of the centaur passing Heracles an enormous drinking-vessel. Heracles' avidity for food and drink later became proverbial, and i t appears that Stesichorus too depicted these : heroic aspects of Heracles' character. The cup that was euuetpov was thus"proportional" to his appetite, measuring as i t did three flagons, <i>S TpbAdyuvov . It i s possible! to interpret euueTpov as "measuring'', taken closely with the measurement ws TpuXdyuvov that follows. Such an interpetation seems prosaic and less effective than the former suggestion. As I understand this phrase, therefore, the poet has diverged quite widely from the -traditional expression 6e*icas ctuipuxd'iteAAov, fabricating a new epithet oxtfipuov and expanding the phrase to incorporate the notion that the vast dimensions of the "cup" were equal to Heracles' vast capacity for wine. 1*+. d<pL*xoL$' capas TCOTL ge'Wea *ju-xtbs epeuvas . . . ' ,PMG 185 3 (SLG 17 4,5) This unit provides another example of the compounding of word-groups in new, non-Homeric associations. The unit basically consists of a noun+genitive group, which may be broken further into a noun(in the genitive) with double epithet group. The combination of Be*v§ea and VOXTO*S does not occur in the epic tradition, where 3e*vdos in singular and plural i s almost totally restricted to the context of the sea, with daAdtians and otAd's (altogether 10 times in the Homeric corpus and once in '. the Hymns). Since the Sun is about to cross Ocean, 6<ppa 6uf 'ft xeavouo 72 nepaaatj line 2 j so that he might reach the place of murky night, which therefore coincides with the place that l i e s beyond the known limits of the world, beyond 't3ie depths of Ocean, the phrase Be'vSea vuxtds incor-porates both factors: the place across the Ocean, beyond the sunset,is one of inevitable darkness, where Ocean and darkness are almost " indistinguishable. uepdf as an epithet of vu£ i s also unprecedented, but the association i s doubtless derived from the use of C e p d v with ?juap 1. Ili a d VIII 66);, with cpdos ( Hesiod, Works and Days 339); and especially with x v d i p a s as in 6dnL T ' n e X t o s x a t eitl x v e i p a s t e p o v e X d n u (Iliad XI 194, 209, XVII 455). The transference of the epithet commonly associated with x v e c p a s to vu£ is quite natural*. The f i n a l epithet, e p e y v a s , which occurs once in the Homeric poems in conjunction with yu? in the dative (Odyssey XI 606), as well as with, for example, you*, is readily associated with v9£ through the frequent combination of the cognate adjective e p e g e v v d s with vu£. e p e y v d s i s in fact the shorter adjectival form of e p e g e v v d s , evolving from the root e-pe3~, to which was added the adjectival suffix - v o - . The resultant juxtaposition of voiced l a b i a l beta and nasal nu caused the assimilation of the former to the latter* hence e p e y v d s . It appears that the longer form of the epithet was more suitable metrically in the hexameter line, and therefore it_prevailed over the shorter form in the Homeric epics. For Stesichorus, however, the shorter form was more convenient. The verb aqjuxous'may also be considered as deriving indirectly from the "formulaic" associations of enX x v e t p a s and e X d n t . There are two "formulae" that may be cited as indirect precedents for the expression: r i y o s 61 n d X u o s x a r e b u x a l iti xve*<pas ? | X d e v jf 6 times in the O^ysaey and once in the Iliad) and with a slight variation on the previous quotation, Sdnu T 1 f | e * X t o s x a X e n t xve*<pag C e p d v fiX^ne (3 times in the Iliad) . The context of the setting aun in these lines may have influenced Stesichorus' choice of the epithet t a p d s to qualify vu£, and he has chosen d i p t x o u T O , a variant whose meaning differs very l i t t l e from eXdnc/fiXdEV. Thus we see in this line of Stesichorus how his description of the Sun's arrival i n the west is related indirectly to "formulaic" lines in the Homeric corpus, but how, through a series of possibly conscious 121 alterations and the introduction of the metaphor "depths of darkness" with i t s double entendre, the poet has revitalised the traditional commonplaces from the epic corpus that describe the setting of the sun, relying on his audience's subconscious or semi-conscious expectation of certain word-associations. 15. axefiov dv-TLite*pcts xAeuvas 'EpudeCas Tapxnaaou noxauou napa icayds T^.L'XT£v> ont.eLpo'vas a p Y u p o p u S o u s 73 ev xeu&uGSvc ite'xpas PMG 184 3,4 (SLG 7 3,4) Although there was, as noted in the previous chapter, a prototype for the phrase TCOTCXUOO Ttap& Kayo's i n TtTyyas noxauuiv, in the Homeric phrase i i n y a C * appears to mean the "source" rather than the waters of the river, but i t is this latter meaning that scholars wish to apply to 74 the Stesichorean phrase. . I t i s generally assumed that "source" would be i l l o g i c a l i f the reference in the previous line to a place "almost opposite" the island of Erytheia implied that the birthplace of Eurytion 75 was on or near the coast. It was suggested by Bergk that the lines could be transposed as follows: TapxnaaoO icoxauou a x e 6 d v d vx tnepas xA.ei.vos ' EpuSe Jas ev xeufcuSvu itexpas Ttapa nayas diceLpovas dpyupopC*Cous and that the word Maya's meant "veins of ore", as i t was used in Aeschylus' Peraae 238 with reference to the silver-mines at Laurion. However, the context of the fragment in Strabo implies that the geographer interpreted the lines as referring to the source of the river Tartessus, since the lines are introduced in his account of the river's source being near a 76 mountain called Castalo, famed for i t s silver-mines . The fragment as i t stands in the text of Strabo does not f i t exactly into the proposed -. metrical scheme of the Geryoneis and i t seems therefore that something may 77 be missing from the text, possibly after 'Epudetcts . By eliminating the 122 immediate juxtaposition of ax£6ov dvTtite*pos xAeuvas 'Epudejfos and the itayas of the river Tartessus i n the following line one might remove the potential i l l o g i c a l i t y of geographical location, but i s there any justification for demanding of a poet the precision of a cartographer? The epithet cViteo*'p«v could be applied to itoycts in either of i t s senses, duetpwv i s regularly found with yctta (8 times i n Homer, 3 in Hesiod). Not inappropriate of a large body of water, the epithet i s also found i n the phrase en1 ctitetpovct Ttovxov (Iliad I 350) and in xccra itovrov dneupovct (Odyssey IV 510). The epithet i s therefore naturally associated with both earth and water, and could presumably be applied to the Haters of a river or the source. In such an application,however, the meaning of the epithet would be "inexhaustible" or "never running dry", rather than "boundless" or "limitless" as in the phrases with yaZa or TCOVTOV. The unique coinage dpyupopuEous suggests that the noun i t qualifies has i t s roots or origins i n silver (cf. y6aTd*pt^os , Parmenides 78 # B 15 a) . Wilamowitz believed that such an epithet must qualify itexpots 79 and not itaycts ; in other words he suggested that Eurytion was born in a cave with s i l v e r ore in i t . Without more of the context and more - " certainty as to the colometry of the fragment, the possibility of reading d p y u p o p i T c o u with itetpas must remain open. Since, however, the juxta%. position of two epithets that qualify the one noun i s found elsewhere i n 80 Stesichorus , I am inclined to accept the reading of the accusative plural. As such/ the epithet may be: understood as an ingenious variation on the expected ctpyupobdvriS, which is commonly employed in epic to describe rivers (for example in Ili a d II 735, XXI 8, 130; Theogony 340j cf. Bacchylides 8 26) and on whose analogy this compound must have been formed. The expected epithet denoting the silvery ripples of the water has been replaced by one which may, b^ analogy, re c a l l that epithet, but which immediately calls attention to the stream's source lying i n s i l v e r -bearing rock. If the use of Kayo's in the sense of "veins of ore" was current in Stesichorus' time , then the entire phrase may have been intended to express both the obvious description of ;:.ever-flowing, silvery streams that at their source resemble the root-work of a tree, and an allusion to the location of the cave in mountains where there was an apparently inexhaustible supply of s i l v e r , i n veins that were similar to the root-work of a tree. In other words the poet may have created an elaborate phrase that may be perfectly understood in two separate ways. The entire fragment demonstrates the poet's technique of combining Homeric with non-Homeric elements. xAeuvas *Epu$etas i s non-81 « Homeric,while xeu&uffivi, itexpas occurs only in the epic tradition outside 82 the I l i a d and Odyssey,. . The location of the birthplace of Eurytion i s identified by an allusion to the area of the Tartessus in Spain where silver had been discovered i n the 7 th century by the Greeks , an allusion referring to that wealth in the epithet dpYupo*pL£os which must have been recognised by at least some of the poet's audience. 16. depQov e£ott<pvas xe*pas 2360 f r . l i 1 In the Homeric poems, expressions such as nop<puper)Vfptv Svnxotai, xavdaanu | Zeus e£ owpavd'dev, xe*pas euuevac ... (Iliad XVII 547, 548) or-iSe'Sv xepcJeaat nt^naas (Iliad IV 398, VI 183) or Auos te*pas atYi>o*xoco (Odyssey XVI 320) reveal the natural assumption that the gods 83 ~ were responsible for the appearance of portents . The epithet Setos is never applied to the word xe*pas, although a parallel situation may be cited in the case of heaven-sent dreams (Odyssey XIV 495 and Iliad-II 22): the god has sent the dream or caused i t to appear before the eyes of the mortal, and hence i t is Seuos ovetpos. T^hW, Stesichorus' use of Seuov to qualify xdpas reflects the traditional view of the origins of portents without relying directly on any "foamulaic" word-group from the Homeric poems. The verb (pcawto is regularly associated with repots ,as in n vfiyv xd6' eipnve 9eds xdpas ?ie aoi adxfii, (Odyssey XV 168) or in nycv yev xdbf e<pnve xepas ye*ya ynxdexa Zeds (Iliad II 324) where the god causes the portent to appear. The use of the adverb e£ctC"<pvns in Homer i s confined to the sudden outburst of flames : xb 6e (pXeyeu ctHctyctxov nopj opyevov efioiupvris (Iliad XXI 14) or nop, xo x' enecadyevov itdXtv ctvbptov ( opyevov e£ad(pvns (pXeye&ei, (Iliad XVII 738). Portents, however, were not con-ceived of as appearing suddenly. The choice of eCad<pvns may therefore result from a play on words, since the adverb is based on the root and i s thus reminiscent of the association of <padv» in the contest of xepas in the Homeric poems. The adverb, which s t r i c t l y speaking modifies the participle ubovaa, is placed within the noun+epithet phrase fceuov ... xepas thus underlining the suddenness of the bird's appearance and the 84 impact of the portent in the minimum amount of words, . 17. '• iteq>opo-y\ ydvos afyaxjC . . . . J . . £, .31. xe xoXai/; oXeactvopos adoXobe£dp] oo obdvauauv "Ybpas" 2617 fr.4 i i 3-6 In the previous section of this chapter the innovative elements 85 of aCyaxt ... xe XOXSL with iteipopoyydvos were considered . Linked to this word-group through the dative 66dvai,CLV, which is also dependent on the participle itefopuyyevos, i s the f i n a l word-group of a series of phrases that describe the arrow piercing Geryon, namely oleadVopos a'tdXobetpou o6dvaccav "Ybpas. If we disregard for the present the dative obdvauacv the genitival phrase may be considered one of the few totally non-Homeric associations of a noun and epithet in Stesichorus. Of the two epithets, which are both foreign to the epic tradition, aXeofivup occurs in Theognis 86 and the late epic of Nonnos . In the former, the epithet has taken on a metaphorical sense in the context of perjury, qualifying vopMOs , and seems far removed from the l i t e r a l sense in Stefichorus. I assume that the epithet in Stesichorus is to be understood l i t e r a l l y ; in any folk-lore i t i s an essential characteristic of a monster to be "man-slaying". It i s possible that the epithet was coined by Stesichorus , since although the individual elements of the compound do have precedents in Homer, the word does not occur i n the early epic tradition at a l l . The form oXea- occurs once only, i n wAeauxapuos (Odyssey X 510) where the lengthened f i r s t syllable i s dicated by the metrical requirements of the line. The second element, -^ nvwp, however, is more frequently found,: ipStcrtfvoea occurs 5 times in the I l i a d , as does pnStfvcop, the epithet 87 of Achilles, which Stesichorus himself employed in fragment 1 of 2619 Indeed, tpduarfvopa may well have acted as a model for the construction of oXeacfvopa in which the poet chose to replace the commonly associated element with another of the same meaning, but less frequently found. For the individual elements of o t t o A o 6 e t p o s » there are again precedents, though rare, in both Homer and Hesiod: caoAo$to'pri5 (Iliad TV 489) , a u o A o u c ' T p r i s (Iliad V 707) , c t C o A d ' u n T c s ( Theogony 511) , 6 o u A t x < 5 6 e i , p o s (Iliad II 460) and uocxuAo'SeLpos (Works and Days 203). . In the case of compounds in - 6 e t p o s , both in the epic poems and in later instances ( itOLXuAdSeopos i n Alcaeus PLF 345 and ouoAddeupos i t s e l f in Ibycus, PMG 317 (a) 2) the contest of the epithets i s restricted to the description of birds; long-necked swans, or nightingales with spotted throats. The simple adjective a C o X o g i s used once of a snake in 88 Il i a d XII 208 and later, in Sophocles',Trachiniae of the Hydra , whxch description may have been influenced by Stesichorus* choice of epithet in this passage. The only instance extant of a description of the Hydra prior to Stesichorus' Geryoneis i s to be found in the Theogony, where the monster i s identified as Aepvacnv after i t s abode at Lema, and as X u y p S LSULCIV (313f.) , neither of which expressions give any physical picture. Stesichorus' description of the monster as cttoXd'6eLpos may therefore be an innovation, the epithet implying both the shiny aspect of the snake" like neck of the monster, and also the mottled effect produced either by scales or by several colours** 9. Since ofiuvcaauv occurs in the epic poems as frequently without as with an epithet, the usage here is not exceptional, and no example specifically suggests i t s e l f as a precedent for the Stesichorean passage. The dative obuvcactv follows neipopuYVevos, linking the a C u a t c and the XO^SL with their origin, namely the Hydra. It was in the blood and g a l l of the dying monster that Heracles dipped his arrows to make them i n f a l l i b l y lethal for future t r i a l s , and so the word 66uvaLO"LV hints at the imminent death-agonies of Geryon as well as referring to the source of the potential death-agonies, the ctuuct and x.o\d of the Hydra. IV Noun+epithet groups in which both elements are foreign to the epic tradition. 1. xXetv&s ''Epydetcts PMG 184 1 90 The epithet xXetvds is post-Homeric , applied frequently to cit i e s in Pindar, although a parallel relevant to Stesichorus 1 use may be cited in Solon, fragment 7 3 (Diehl): xXeuvfis duo vnaou in which he refers to the island of Cyprua. The island Erytheia i s not known in the extant epic corpus, but i t is perhaps significant that Hesiod, according to Servius' commentary on Aeneid IV 484, calls one of the Hesperides Erythea, since the Hesperides l i v e on an island in the far west, beyond the p i l l a r s of Heracles"74". 2. oXeactvopos adoXo6etpou . . . "Yfipas 2617 f r . 4 i i 5,6 As was noted i n the previous section, the epithets oXecravopos and cttoXo6eC*poy are both non-Homeric, although their formation may be 92 traced to similar compounds in the epic corpus . The Hydra's appearance in Greek mythology is confined to the exploits of Heracles and therefore did not occur in the Homeric corpus. In this chapter 46 instances of noun+epithet groups, whose individual elements, i f isolated, _may be cited as Homeric, demonstrate that the poet availed himself ..freely of his linguistic inheritance, the Homeric corpus. However, when these noun+epithet groups are considered as unities and , where possible, within their individual contexts, i t can be seen that the poet has created new juxtapositions of noun and epithet that as unities have no precedent in the Homeric corpus. The poet seems deliberately to have broken establishedword-associations in which many epithets had become t r i t e or ineffective through constant usage in a set position. In such an adaptation of epic diction the poet has succeeded in preserving the character and tone of epic, while at the same time has created something original by his infusion of v i t a l i t y into moribund "formulae", and to do so he appeiars to have relied upon his audience's reaction,whether conscious or subconscious, to the association of a particular epithet i n an unexpected environment or to the application of novel epithet in a traditional environment. In the case of noxpbv oXe^pov for example (2617 fr.4 i 11 = section I, no. 9), the epithet i s reminiscent of i t s habitual association with OLCTOS i n i t s new position with oAedpos. At face value, Geryon's death would be ituxpds, physically painful, because of the poison on the t i p of the arrow. It would also be 93 grief-causing, because Geryon's mother i s close to the scene of combat, . However, Tttxpd"si.;.also recalls the fact that the instrument of death i s an arrow, because of the association of iaxpd*v and otaxov in the Homeric poems. Although the examples of word-groups based on elements that do not appear i n conjunction in Homer, or of which one element is foreign to Homeric epic, outnumber those of direct imitation, there i s l i t t l e evidence of Stesichorus' employing word-groups, a l l the elements of which are alien to epic. I have found only two such cases, noted in section IV above. Both examples contained epithets that do not occur in the Homeric corpus, where there i s no mention of the Hydra or the island of Erytheia. As proper names, however, the two substantive elements of the phrases are less significant than i f they had been common nouns, for the poet's expansion and innovation in myth would inevitably involve the incorporation of new names such as the island of Erytheia. Apart V from the proper names, the only example of a non-epic substantive i s xuvOXaypdj (PMG 255 = section II, no. 4) whose accompanying epithet i s Homeric. Thus the poet tended .to work within the lexical inheritance of the Homeric corpus; he may have created innovatory compound epithets, but, as far as the evidence shows, did not show much preference for non-epic or-newly coined nouns. Footnotes to chapter IV 1 The scholiast commenting on the p a 6 u v n s XE^ p d s of Hfira in Apollonius' Argonautica III 106, gives a l l the instances of the epithet known to him, including examples from Ibycus and Anacreon, but without helpful contexts. 2 Theocritus, Idylls XI 45, XXVII 146. 3 The I l i a d , unlike the OdyssBg, has no occasion to present a picture of the bard at work, apart from the funeral of Hector, in XXIV 720 f f . It i s often supposed that the greater concentration upon the figure of the bard in heroic society i s evidence for later composition of the Odyssey, but this i s questionable (cf. Frankel, Early Greek Poetry  and Philosophy (Oxford, 1975) p. 10). 4 On invocations see page 90 f f . "Inspiration" i s perhaps a misleading term, since the bard requests factual information from the Muses, presupposing either.,that the format came together with the information, or else that he was in that respect auTo6C6ctxTOs , as Phemius once claimed. As far as the bards' own contribution to the composition of epic is concerned, we can glean v i r t u a l l y nothing from the I l i a d or the Odyssey. It was simply a matter to be taken for granted as the g i f t of the c t o i , 6 d ' s . It is only later, in Hesiod, that the poet's personality emerges. Hesiod claims that when he met the Muses, they presented him with the symbols of poetic composition. Once in possession of the rod of laurel and the "breath? of the Muses, the poet becomes in part responsible for the use he makes of the g i f t ; he is not merely a mouth-piece of the Muses. 5 The adjective e p u n p o s i s heteroclite. 6 In the f i r s t publication of 2359 (P.Oxy. vol XXIII, p. 11) Lobel was uncertain as to the relationship of column i i to column i , which he ascribed definitely as a piece of the Suotherae on the grounds of .names appearing therein, such as the sons of Thestius.B.Snell, i n "Stesichorus' Zoodrjpau," Hermes 85 (1957) pp. 248-251, demonstrated that from the metrical structure the two columns belonged to the same poem, and Ftlhrer followed this, with variations of metrical interpetation in "Zur metrische structur von Stesichorus' Suotherae," Hermes 97(1969) pp. 115-116. An alternative identification was suggested by H. Lloyd-Jones in CR 8 (1958) p. 17, namely that the fragment belonged to the Athla on the grounds that I v d e v uev ... evdev 6e... could refer to spectators. Bowra(GLP p. 96 f.) adopted the theory of spectators, but retained the identification of the Suotherae for the fragment. It seems to me,however, that the epithets might be mere 'appropriate of participants rather than spectators. 7. Cf. page 113. 8. Peisander of Rhodes, according to the Suda, lived in the 7th century B.C., but his dates are not verifiable. He appears to have written an epic about Heracles, including the mission to fetch the Qolden Apples of the Hesperides. Pherecydes of Athens, as quoted by the scholiast on Apollonius of Rhodes (Jacoby FGH'.- 3 16 (a) ) records the fact that the apples were guarded by the nymphs (Hesperides) and by a snake whose name was Ladon. It i s likely that Pherecydes was drawing upon Peisander and Stesichorus for his material.(See also note 16). The guests at the wedding of Peleus and Thetis depicted on the Francois vase are dressed i n elaborate long gowns. Ordinary women, as opposed to goddesses, are also depicted on vases wearing flowing robes as for example in figure 85 and 196 of J. Boardman's Athenian Black  Figure Vases (London, 1974). 10 The application of the epithet SavSd's suggests that i t may be understood as referring to a reddish-yellow colour rather than "blonde", which in English tends to be equated with Nordic fairness. 11 Sappho, PLF 23 5 ; Ibycus, PMG 282 5. 12 Cf. West,"Stesichorus Redivivus," ZPE 4 (1969) p. 139 f f . , although there he i s primarily concerned with th e metre , and not the anomalies in structure. Page, in PCPhS 19 (1973) p. 58 expresses some doubt that fragment 14 belongs to the same poem as fragment 1. 13 If the assumption is correct that Helen is i n Troy in this fragment, and that i t belongs to the Iliou Persis, then the Palinode, in which Helen does not go to Troy, must be separate from the Iliou  Persis. On the problems of the Palinode, see page302f,,chapter IX. 14 The CpnS i s a more general word for "hawk"; see D'A. W. Thompson, A Glossary of Greek Birds (Oxford, 1936) pp. 65-67. Cf. Odyssey XIII 87. 15 Oppian, Halieutica IV 7, where the epithet adorns the Muses, in the poet's verbose invocation: TEpTtwAats o C n t a u v EU6V v d o v n n u o o c o p o t Mouactt xoauifaavTO x a l eSe"<rr£<|xsv dou6ris 6uJpu)i, %ea%ea£w\, ... 16 G.Huxley, Greek Epic Poetry (London, 1969) p. 101, believes that Peisander of Rhodes was perhaps the inventor of the labour in which Heracles had to fetch the Hesperides' apples, dating that poet to 648 B.C. from the information given in the Suda. Bowrav on the other hand, in GLP p. 91, believes that Stesichorus preceded Peisander, and was the inventor of the tale. Since the precise dates of both poets are far from certain, I hesitate to make any firm assertions as to the inventor of the legend. The independent expansion: of themes found in Hesiod or the epic corpus i s not impossible. 17 See A. Garzya, "La poesia l i r i c a greca nella Magna Grecia," P&I 10 (1968) pp. 247,248. 18, Cf. i t t x p o X Adyou in Euripides' Helen 481. 19 See page 116 f. of this chapter. 20 In Theogony 79 Hesiod calls Calliope r\ 6e upcxpepECCTdxri. e a r l v 131 21 There are some grounds for believing that this passage in the Peace derives from the Orestela of Stesichorus, as Bergk f i r s t proposed in his PLG II (1843)j there are two other allusions t h a t the same scholiast has recognised as deriving from the Oresteia at lines 799 and 800 of the comedy (= PMG 211 and 212). 22 Cf. Sappho, PLF 16 or Ibycus, PMG 282. 23 In his Hymn to the Muses at the beginning of the Theogony Hesiod describes various^ attributes of the Muses: they have Ttocra' aJtatAoCatv and re*peva x P ° * a ; they sing with iiEpoxaAAea oaaav^cr 6 i t l Aetptoeaant; they attend princes and bring joy to man amid his sorrows and so on, but nowhere are they golden-winged. Later i n Bacchylides one finds an isolated allusion to a golden feather , xpvaiov. 'itxe'pov, dropped by the Muse { f r . 20 B 3,4) but there i s no indication as to whether this came from a Muse's golden wings or not. If so,has; Bacchylides adopted Stesichorus ' image? 24 The motif of the lasting nature of song is found predominantly i n Pindar ?Cevg^ Pythian.;yjy,^18vahdtllg., 113), .rbut there