Playing the Pythagorean: Ion's Triagmos

ION OF CHIOS Fragments of a Polymath EDITED BY VICTORIA JENNINGS University of Adelaide ANDREA KATSAROS University of Adelaide © 2005 ii CONTENTS List of contributors Acknowledgments Abbreviations Notes on citation 1. INTRODUCTION Victoria Jennings and Andrea Katsaros SECTION I: SURVIVAL 2. THE HOCUS OF A HEDGEHOG: ION’S VERSATILITY John Henderson 3. SHOT FROM THE CANON: SOURCES, SELECTION AND SURVIVAL Guy Olding 4. THE POET AND THE PLACE: A MODERN CHIAN PERSPECTIVE ON ION OF CHIOS AND HIS HOME ISLAND Nikos K. Haviaras SECTION II: ION SUNGRAPHEOS 5. ION’S EPIDEMIAI AND PLUTARCH’S ION Christopher Pelling 6. ION OF CHIOS AND POLITICS Anne Geddes 7. ION THE WINEMAN: THE MANIPULATION OF MYTH Guy Olding 8. TRAPPED BETWEEN ATHENS AND CHIOS: A RELATIONSHIP IN FRAGMENTS Alastair Blanshard SECTION III: ION THE POET 9. ION OF CHIOS AND THE POLITICS OF POLYCHORDIA Timothy Power 10. SNOWY HELEN AND BULL-FACED WINE: ION AND THE LOGIC OF POETIC LANGUAGE Michael Clarke 11. STAGING EMPIRE AND OTHER IN ION’S SYMPOTICA Andrea Katsaros iii SECTION IV: ION THE TRAGEDIAN 12. ION OF CHIOS: TRAGEDY AS COMMODITY AT THE ATHENIAN EXCHANGE Alexander Stevens 13. ION OF CHIOS, SOPHOCLES, AND MYTH Judith Maitland 14. LOOKING FOR OMPHALE Pat Easterling SECTION V: ION THE PHILOSOPHER? 15. PLAYING THE PYTHAGOREAN: ION’S TRIAGMOS Han Baltussen 16. LEGWORK: ION’S SOCRATES Richard Fletcher 17. ION’S HYMN TO KAIROS Victoria Jennings Bibliography Concordance to the fragments of Ion of Chios Index of passages cited Greek index General index iv ACKNOWLEDGMENTS ‘Each friend that comes annoys, that goes affronts, as Ion has it’. We are fortunate that we have not suffered such discontentment as Plutarch describes (71* Leurini = De tranquillitate animi 466d). A number of colleagues and friends have assisted and encouraged this project; in particular, we would like to thank the original participants in our panel on Ion of Chios at the Australian Society for Classical Studies conference in 2003. They justified our faith in the project from its conception. In addition, for their practical help or personal encouragement, we would like to thank Pat Easterling; John Henderson; Polly Low; Robert Ussher; Peter Wilson; Ian Worthington; Noriko Yasumura. Alastair Blanshard and Alexander Stevens stepped manfully into the debate on volume structure. Victoria would also like to thank her tolerant colleagues Yeng Chiam and Natalie Cahill. The editors are grateful to the Faculty of Humanities of the University of Adelaide for financial assistance towards a (second) Colloquium on Ion of Chios at the University in September 2004, which enabled Australian contributors to present ideas and works in progress. This workshop would not have been possible without the administrative nous, not to mention unflagging efforts, of Dr Han Baltussen. The editors have benefited from nomination as Visiting Research Fellows to the University of Adelaide. v ABOUT THE CONTRIBUTORS Han Baltussen is Lecturer in Classics at the University of Adelaide, South Australia. Alastair Blanshard is Lecturer in Greek History at the University of Sydney. Michael Clarke is Lecturer in Ancient Classics at the National University of Ireland, Maynooth. Pat Easterling is Regius Professor Emerita of Greek at the University of Cambridge. Richard Fletcher is a graduate student in Classics at the University of Cambridge. Anne Geddes is Senior Lecturer in Classics at the University of Adelaide. Nikos K. Haviaras teaches Greek and History at the First Lycaeum of Chios. John Henderson is Professor of Classics at the University of Cambridge. Victoria Jennings is a Visiting Research Fellow at the University of Adelaide. Andrea Katsaros is a Visiting Research Fellow at the University of Adelaide. Judith Maitland is Senior Lecturer in Classics and Ancient History at the University of Western Australia. Guy Olding is affiliated to the Australian National University. Christopher Pelling is Regius Professor of Greek at the University of Oxford. Timothy Power is Assistant Professor of Classics at the University of Washington at Seattle. Alexander Stevens teaches Latin and Greek at Sydney Grammar School and is a Honorary Research Fellow at the University of Sydney. vi ABBREVIATIONS Abbreviations of ancient authors follow The Oxford Classical Dictionary (3rd edn). Abbreviations of periodicals follow L’Année philologique. Other abbreviations found in the text: ABV ARV2 CEG CHCL CPG DK EGF FGrH IEG2 IG Leurini LCS LCS Suppl. LIMC LSJ OCD3 PMG PCG SEG SH SIG TrGF West Beazley, J. D. (1956) Attic Black-figure Vase-painters. Oxford Beazley, J. D. (1963) Attic Red-figure Vase-painters, 3 vols, 2nd edn. Oxford Hansen, P. A. (1983-89) Carmina epigraphica graeca, 2 vols. Berlin Easterling, P. E. and Knox, B. M. W. (eds.) (1985) Cambridge History of Classical Literature I: Greek Literature. Cambridge Leutsch, E. L. von and Schneidewin, F. G. (1839-51) Corpus paroemiographorum graecorum, 2 vols. Göttingen Diels, H. and Kranz, W. (1951) Die Fragmente der Vorsokratiker, 3 vols, 6th edn. Dublin and Zürich Davies, M. (1988) Epicorum Graecorum Fragmenta. Göttingen Jacoby, F. (1923- ) Die Fragmente der griechischen Historiker. Berlin and Leiden see West Inscriptiones Graecae (1873- ) Leurini, L. (1992, 2000a) Ionis Chii Testimonia et Fragmenta (Classical and Byzantine Monographs 23). Amsterdam Trendall, A. D. (1967) The Red-Figured Vases of Lucania, Campania and Sicily. Oxford Trendall, A. D. (1983) The Red-Figured Vases of Lucania, Campania and Sicily. Third Supplement (Consolidated) (BICS Supplement 41). London Lexicon Iconographicum Mythologiae Classicae (1981- ). Zürich Liddell, H. G., Scott, R., Stuart Jones, H. and McKenzie, R. (1940) Greek-English Lexicon with a Supplement, 9th edn. Oxford Hornblower, S. and Spawforth, A. (eds.) (1996) The Oxford Classical Dictionary, 3rd edn. Oxford and New York Page, D. L. (1962) Poetae Melici Graeci. Oxford Kassel, R. and Austin, C. (1983- ) Poetae Comici Graeci. Berlin and New York Supplementum epigraphicum Graecum (1923- ) Lloyd-Jones, H., and Parsons, P. (1983) Supplementum Hellenisticum. Berlin and New York Dittenberger, W. (1915-24) Sylloge Inscriptionum Graecarum, 3rd edn, 4 vols. Leipzig Snell, B., Kannicht, R. and Radt, S. (1971-85) Tragicorum Graecorum Fragmenta, 4 vols. Göttingen West, M. L. (1992) Iambi et Elegi Graeci ante Alexandrum cantati, vol. 2, 2nd edn. Oxford [IEG2] vii NOTES ON CITATION Ion’s fragments are cited by reference to Leurini’s editions (1992, 2000a), a standard reference (e.g., FGrH, TrGF, DK, West IEG2, Page PMG) and the ancient authority. Ascription follows Leurini’s model: * ** *** the work to which a fragment is ascribed is uncertain; a dubious ascription; a spurious ascription. In general, ‘T’ indicates ‘Testimonium’ and ‘F’ ‘Fragment’. A concordance to the fragments cited in this volume may be found at the end. In addition, the reader is directed to the concordances of Leurini 2000a and 2000b. Transliteration follows general conventions. Baltussen 1 CHAPTER 15 Han Baltussen PLAYING THE PYTHAGOREAN: ION’S TRIAGMOS 1. INTRODUCTION Was Ion of Chios a philosopher? As so often when one asks a seemingly simple question, the answer turns out to be complicated and - given that in this case the evidence is exiguous - inconclusive on most issues. This explains why my discussion regarding Ion as a philosopher is to a large extent exploratory and speculative. I intend to use other people’s work in the process - in particular the articles by Huxley (1965), West (1985) and Dover (1986), of which the latter two appeared so close in time to each other, that they are independent treatments. Nevertheless, my approach differs from all these. I examine Ion’s status as a philosopher, an approach hitherto unexplored, by bringing into play all the evidence and providing a context. How one defines ‘philosopher’, in particular with reference to the early fifth century, where the term is still fairly fluid, is a natural corollary to my investigation. My answer to the question first posed above will be a qualified ‘yes’, since the evidence suggests that Ion was knowledgeable about the subject, yet perhaps not completely serious while writing about it. Research into the topic revealed that some of the earlier work (seeking answers to different questions) did not attempt to look for coherence in the surviving philosophical material. I believe that some of Huxley’s speculations are overconfident and potentially misleading; West focuses more on the historical evidence; Dover’s comments are mostly from a literary perspective, analyzing the stylistic features which characterize Ion’s work. Regarding the Triagmos, Dover’s analysis is beyond doubt the most helpful. However, the valuable comments of all three are taken into account.1 Regarding any search for coherence, this will be the first attempt to study the intellectual context as well as assess the ‘merit’ (if any) of Ion’s extant philosophical fragments. 1 Dover’s analysis of the style and literary context remains unchallenged. My arguments are intended to be complementary to Dover and West, both still well worth consulting. The earliest literature has faded from the scholarly debate (see Diehl 1916), which is not always for the better. For a good introductory summary of sources, and Ion’s role in it, see von Blumenthal (1939) 3-4. Baltussen 2 The importance of a synoptic discussion of all philosophical fragments which provides a meaningful context to, and exegesis of, Ion’s contribution in this area is immediately obvious, particularly when we appreciate the extent of his philosophical output.2 While there is very little left, what remains provokes many stimulating questions. In particular, I focus on the work with the intriguing and puzzling title Triagmos (‘Triad’?), printed in Diels and Kranz’s Die Fragmente der Vorsokratiker (= 36A-B DK).3 It has been suggested that the role of the number three indicates that Ion had Pythagorean leanings. After looking at the evidence more closely (§2), I shall explore the Pythagorean connection by discussing what we know about early Pythagoreanism (§3). Next, on the basis of the fragmentary evidence, I shall consider the extent to which Ion can be linked to Pythagoras or his followers (§4). It is, in other words, our task to see if there is any philosophical meat in the meagre remains of his work. I shall concentrate on two questions in a deliberately discursive treatment of the fragmentary evidence: (1) What is Ion’s connection with philosophy, in particular Pythagoreanism? (2) What can we say about the work Triagmos? The specific focus on this work arises simply because it is his only philosophical work on record. Here, the meaning of the title deserves further scrutiny (§5). Both these questions will also draw us towards other relevant issues, and, collectively, they should be of some help in clarifying Ion’s contribution to philosophy. 2. THE EVIDENCE FOR HIS PHILOSOPHICAL WRITING(S) Our knowledge of Ion’s philosophical output amounts to four B-numbered fragments in Diels-Kranz (DK),4 but this paucity of evidence does not render our scrutiny futile: there are actually ten short texts and a further six A-fragments 2 Brisson (2000) goes some way to synthesize, but is restricted by the format of the encyclopaedic lemma. 3 114-119 Leurini = 20-25 von Blumenthal. It can also be found in a one page section in Barnes’ overview of early Greek philosophy in the Penguin Classics (1987): 223, cf. pp. 82-3 (in the section on Pythagoras). 4 In order not to overcomplicate the discussion with side-issues, I will use the terms ‘fragments’ [F] and ‘testimonia’ [T] in the sense of Diels-Kranz where ‘fragment’ (label B) is an alleged direct quotation, while ‘testimonia’ (label A) are indirect reports on the author, sometimes providing paraphrase of his words. For comments on why this way of editing fragments is no longer viable, see Laks (1998), Burkert (1999) and Baltussen (1999), (2000). For a problem in the current DK division, see below TEXT 4 and conclusion. Baltussen 3 (testimonia), apart from other connections one might make with the ‘literary’ evidence.5 At least one source is happy to call Ion a philosopher (HKNQUQHQL: T4 Leurini = 36A3 DK = Suda s.v. Ion Chios, K 487; see also TEXT 1, below), so we are invited to take his status in this area seriously, even if he is more often labelled a ‘tragedian’. Isocrates mentions him next to Empedocles in a list of RCNCKYP UQHKUVYP, ‘experts of old’, as his exemplars of those who explain the world through a limited number of elements (114 Leurini = 36A6 DK [first text] = Antid. 268). The use of the term sophistes is rather broad here - it was used of musicians, poets and the Seven Sages (for example, Hdt. 1.29, who, interestingly, also uses it of Pythagoras at 4.95), and does not necessarily have the pejorative meaning it acquired after both Plato and Aristotle had put their spin on the history of rhetoric and education. Moreover, Isocrates is in fact counting him among the natural philosophers, so sophistes may be no more than the equivalent of an ‘expert’ in some area. The opening lines in fragment 36B1 illustrate Ion’s reputation as a versatile author. I shall quote this passage almost in full, because of its considerable importance (hereafter referred to as TEXT 1): ¹,YPG¢ITC[GFG? MCK? OGNJRQNNC? MCK¼ VTCIFKžCLMCK¼ HKNQUQHQP VK GRKITCHQOGPQP UWIITCOOC QŸRGT VQ?P .CNNKžOCZQL 7TKCIOQ?P CPVKNGIGUSCKž HJUKPYˆL™(RKIGPQWLGPGPKžQKLFG? MCK¼ RNJSWPVKMYL GRKITCHGVCK 7TKCIOQKž MCSC? 'JOJVTKQL Q 6MJ[KQL MCK¼  ™$RQNNYPKžFJL Q 1KMCGWL CPCITCHQWUK FG? GP CWV VCFG NGIGK CTZJ OQW VQW NQIQW RCPVC VTKžC MCK¼ RNGQP VQWFG RNGKY QWFG?P QWFG? GNCUUY VQWVYP 5 I shall make some comments along the way in order to connect his literary and philosophical activities, although Diels is probably right that Ion’s tragedies have no philosophical content (‘enthalten… nichts Philosophisches’: p. 380 DK, ll. 6-8); in the lyric poems he discerns some verbal parallels with Empedocles (discussed further below). Therefore, it is worth asking to what extent the poetic output expresses views (implicitly or explicitly) that coincide with his philosophical outlook. See West (1985) 76. Baltussen 4 VYP VTKYP GPQL GMCUVQW CTGVJ? VTKC?L UWPGUKL MCK¼ MTCVQLMCK¼VWZJ Ion: he composed many lyric poems and tragedies and some kind of philosophical treatise entitled Triad (‘triagmos’). Callimachus says that its authorship is disputed, and in some copies it is entitled Triads, in the plural (according to Demetrius of Scepsis and Apollonides of Nicaea6). They record in it the following: he says ‘This is the beginning of my account. All things are three (VTKžC), and there is nothing more or less than these three. Of each one thing the excellence (CTGVJ) is a triad (VTKCL), intelligence and power and fortune’. (T9a, 114 Leurini = DK 36A1, B1 = FGrH 392 T3, F24a = Harpocration s.v. Ion)7 Four points are significant in this passage. First, Harpocration tells us how prolific a writer Ion was. He organizes the writings in a peculiar order: lyric poems (or songs), tragedies, ‘a philosophical treatise’.8 Secondly, for this last work he uses the (rather technical) term sungramma which means a prose work with systematic content, and what is more important, he provides a title.9 We may here observe that the adjective HKNQUQHQPindicates that sungramma needs a specification, implying that it could be used for other types of works. Moreover, the indefinite pronoun (VK) leaves us with the impression that Harpocration (or his source) did not quite know what it was about. Other titles exist (Kosmologikos, Peri meteoron), and these probably refer to one and the same work (see §5).10 Thirdly, we are immediately made aware of the 6 Apollonides, a grammarian in the time of Tiberius, wrote a commentary on Timon’s Silloi (D. L. 9.109), a ‘Spurious Inquiry’, and on proverbs: see RE 2 (1896), no. 29, cols. 120-1. 7 Tr. Barnes (1987), slightly modified. Harpocration was a grammarian working in Alexandria (dates unknown, perhaps Imperial period). He wrote a Lexicon on rhetorical authors ‘designed as an aid to reading, not to composition’ (OCD2 s.v., 488). The Greek text, which is rather corrupt, is that of Huxley (1965) 41, who emends Diels’ QWFGPRNGQPJ? GNCUUQP into QWFGPQWFG GNCUUY - accepted by, for example, Brisson (2000). 8 Contrast T8 Leurini = 36A2 DK = Schol. Ar. Pax 835: dithyrambs, tragedy and lyric poems; and the T4 Leurini = 36A3 DK = Suda s.v. Ion of Chios, K 487: tragedian, lyric poet, philosopher. 9 I emphasize this because Harpocration seems to offer the earliest reference to this title: he quotes Demetrius of Scepsis (born c. 214 BC) and Apollonides. This ‘title’ is found in several other sources, on which see my comments in §5. For sungramma see LSJ s.v., but compare Dorandi (1991) who (in another context) distinguishes it from hypomnema (a work-form also attributed to Ion, in T8 Leurini = 36A2 DK = Schol. Ar. Pax 835. 10 Thus already Diehl (1916) 1862: ‘Kosmologikos identisch mit Triagmos’. That Ion was in good company by writing a peri meteoron can be gathered from precedents: see Thales 11A2 DK (Suda); Baltussen 5 importance of the number three - an obvious Pythagorean trademark (see TEXT 2a below). However, on the basis of this text, we have very little to go on to understand Ion’s motive for choosing this numerical theory, nor is it clear how the number three would actually function. One might perhaps paraphrase further the phrase ‘is a triad’ (last line): for example, by saying that the excellence (arete: virtue? quality?) of all things consists in or requires three aspects; but one would still want to know how this informs us about the scope of the number three in his overall theory.11 Ion’s word selection - ‘no less and no more than three’ - may seem redundant, yet within a more oral and competitive culture it is the kind of emphatic statement which further underscores the position: it is not mere repetition, but reinforcement and sharpening up of the main point. In other words, it gives the number three a universal role by emphatically excluding other options. I shall say a bit more on this point in §3. Fourthly, as Dover rightly points out,12 this kind of opening statement, with its generalizing and dogmatic tone, fits in well with some of the earliest prose works known from the Presocratics; here we may compare some of the other rare prose authors like Heraclitus, Diogenes of Apollonia and the Pythagorean Pherecydes. These parallels, also mentioned by Dover for their contrasting importance, seem even more revealing by their differences rather than their similarities. Heraclitus famously makes a bold general statement: ‘of this account (logos) which holds forever men prove uncomprehending...’ (22B1 DK). But Diogenes of Apollonia (late fifth century BC) is closer in style and wording to Ion. It is reported that Diogenes wrote a prose treatise (UWIITCOOC) which started thus:  NQIQWRCPVQ?LCTZQOGPQPFQMGK¨ OQKZTGYµPGK»PCKVJ?P CTZJ?PCPCOHKUDJVJVQPRCTGZGUSCKVJ?PFGTOJPGKžCP CRNJPMCK¼UGOPJP Epicurus Ep. Hdt. (D. L. 10.78); Diogenes of Apollonia 64A4 DK. For example, is ‘each thing’ then another way of saying ‘all things’, or perhaps even an intensifier? This particular use of ‘to be’ is unclear and, as we now tend to say, typically pre-Aristotelian in that it fails to distinguish between its possible usages as a copula (for predication: ‘X is such and such’), as truth statement (‘X is true’), or as existential (‘X is’ = ‘X exists’). See Kahn (1973). 12 Dover (1986) 28. 11 Baltussen 6 It seems to me that a person who begins an account must needs supply a starting point which is indisputable, the expression [of it] simple and dignified. (64B1 DK = D. L. 9.57) Diogenes of Apollonia here provides a methodological preface which stands out because it makes a claim about the starting point (itself not yet stated), insisting it be securely grounded (content) and appropriately formulated (form). Only then does he proceed to make the actual statement of his position, in which a similarly universal claim is made about the world and everything in it (VQ? OG?P EWORCP RCPVC VC? Q¢PVC, ‘all existing things… are the same thing’:64B2 DK = Simpl. in Phys. 151.31; cf. similar in Pherecydes13). Since this is a more sophisticated and stylized way of presenting a theoretical view - showing an awareness of audience and public debate - I take it to be of a later date than Ion.14 In sum, what we have for Ion is just one philosophical prose work, which fits the style of early cosmological explanations, and which for its title, format and content depends on a handful of brief references and remarks. As I shall consider the title (if it is that) last, let us briefly focus on the other two points. Regarding format, we are dealing with a very early prose work,15 intent on positioning itself within the Presocratic milieu, with a strong Pythagorean flavour. It is of considerable interest to note (a) the combination of cosmology and qualitative (ethical?) concerns; (b) a number used as an explanatory factor; (c) the link between the human domain and fortune (see TEXT 1 above). These features move his position closer to Pythagoreanism. It is, after all, not such an bizarre idea to try to establish mankind’s role in the world by analyzing both humans and the world as well as their interaction. Early Greek thought deals with that complexity of issues within the traditional religious framework, and it is clearly a Pythagorean innovation to 13 Pherecydes: see the useful discussion in Kirk, Raven and Schofield (1983) 51-70. Cf. Dover (1986) 29. Note, however, that the quoted sentence is not the actual opening of the work, if we are to believe Simplicius who introduces it with the remark, ‘immediately after the prooimion he writes the following...’ (in Phys. 151.28). In the apparatus to 36B1 (p. 379) DK compare Hippoc. Virg. 1 and Art. 4, which also emphasize the undisputed nature of the claims to come. 15 It was one among several prose works. Cf. his Chiou Ktisis, which West (1985: 74) claims ‘must at all events count as the earliest prose work of its kind that has left any record’, and the Epidemiai. For early philosophical prose that survives, we know of works by Heraclitus c. 500 BC (D. L. 9.5), Anaximander c. 550 BC (D. L. 2.1-2), and Pherecydes c. 600-540 BC - possibly the earliest prose book of Greek philosophy (D. L. 1.116). 14 Baltussen 7 produce a more abstract causal account (for example, harmony of the spheres) to bring them even closer together. Therefore, we are certainly justified in looking in that direction for further clarification of Ion’s position, and some context will be crucial for a better understanding of the meagre evidence. 3. EARLY PYTHAGOREANISM No ancient source refers to Ion as a Pythagorean: labels used are philosopher, tragedian, sophist, poet. Yet the evidence points to such a link, as we have already seen, and Ion talks about Pythagoras twice (92 Leurini = 36B4 DK = 30 West = D. L. 1.120; 116 Leurini = 36B2 [texts 1-2] DK = D. L. 8.8, Clem. Al. Str. 1.131). These scraps of evidence give some indication of where his interests lie. How much of Pythagoreanism was known to Ion as a citizen of Chios, an island close to Samos (Pythagoras’ birthplace), is impossible to know. We think Pythagoras (c. 570-480) probably left for South Italy around 530, and it is there that his ideas first made a significant impact. Because our knowledge of early Pythagoreanism of the sixth and early fifth centuries is insubstantial,16 let us consider two authors who exhibit the wellestablished influence of the Pythagoreans in the late fifth and early fourth centuries, namely Plato and Aristotle. Late in his life, Plato wrote his Timaeus, a cosmological masterpiece named after a Pythagorean astronomer from Italy. It begins thus: ‘One, two, three’ - a count of those present, but also a playful way of introducing one of the main themes of the work, the orderly structure of the universe in which numbers (especially the number three) play a major role. In it we find the harmony of the spheres in mathematical detail, the ‘triangular relation’ of elements,17 and the socalled ‘third factor’ or Receptacle, a spatial entity which mediates between (ideal) forms and (concrete) objects, allowing them to have a presence in the visible world.18 16 See especially Huffman (1999) for further literature. The number three occurs in several places, but among the most significant are: (1) its symbolic role in the binding of world soul and the elements, in which (significantly in Timaeus) two things require a third to be connected (Tim. 31bc, 35ab, 36c); (2) in the triangles that constitute the ‘building blocks’ (surfaces) of the basic elements (triangles and squares) to form air, fire, water and earth (Tim. 52-58). 18 Belief in the special and symbolic value of numbers seem to be perennial. We still say things like ‘good things (or bad) come in threes’, ‘third time lucky’. There is something about the number three which is intriguing: for instance, counting up to three is the minimum required to enable a hearer to 17 Baltussen 8 Aristotle usually criticizes the number theory as applied to the physical world, but his understanding of the theory has proved to be a distortion of the view of Philolaus, not Pythagoras or the Pythagoreans in general.19 Pythagoras himself had, in fact, little to say on cosmology, and is known best for his vast knowledge, religious outlook, shamanism (transmigration of souls) and practical ethics.20 Therefore, when we focus on the specific clue to Ion’s link with Pythagoreanism, the number three, we should make a connection with his contemporary Philolaus (born c. 470). Aristotle gives us a good indication of the importance of three in Pythagorean philosophy (TEXT 2a = Arist. Cael. 268a10-12): ...as the Pythagoreans say as well, the universe and all things in it (VQ? RCP MCK? VC? RCPVC) are determined by the number three (VQKLVTKUK?P), since end, middle and beginning give the number of the universe, and the number they give is that of the triad (VQ?PVJL VTKCFQL). (tr. Barnes, modified) As so often with selective quotation and fragments, the context, which is left unexplored by Diels or Huxley, offers further information and thus deserves to be taken into account. Aristotle adds a further clarification of the Pythagorean position, which arises while he is discussing the subject matter of the science of nature bodies and magnitude. Elaborating on the latter, he takes the Pythagoreans to mean that the ‘three’ refers to ‘the three dimensions’, and he elucidates how this is a means of giving a comprehensive account of the physical world (TEXT 2b = Arist. Cael. 268a7-9): Magnitude divisible in one direction is a line, in two directions a surface, in three directions a body. There is no magnitude not included in these; for three are ‘all’, and ‘in three ways’ is the same calculate fairly accurately when something will happen, and thus quantify his or her level of expectation. 19 See Kahn (1974) 163ff. and Huffman (1999) 82-3; cf. p. 69 where he notes that we have in fact ‘more precise evidence for Philolaus than for Pythagoras himself.’ 20 Huffman (1999) 68, who, with reference to Burkert’s analysis, stresses the importance of distinguishing Pythagoras from the Pythagorean tradition. Baltussen 9 as ‘in all ways’. It is just as the Pythagoreans say [TEXT 2a follows]... (tr. Guthrie, with my italics) Remarkably, Aristotle seems here to agree with this position,21 with a pre-emptory statement which anticipates a reference to the Pythagoreans, perhaps even playing on words when he uses the word ‘is defined’ (Y©TKUVCK, 268a11), which obviously also means ‘circumscribes’, ‘includes’.22 On the basis of this strong link between ‘three’ and ‘all’, the role of the number ‘three’ is almost given the status of ‘laws of nature’ (cf. Y©URGT PQOQW?L GMGKžPX >sc. HWUX@, 268a14), to be found also, Aristotle reminds his audience, in religious rituals.23 Therefore, it is the use of three in relation to spatial dimensions, which make it ‘allencompassing’, since there are no dimensions over and above the three listed. Though a perfectly acceptable interpretation of the Pythagorean view, one may well wonder if this is what the Pythagoreans had in mind. This leaves us with the remark about ‘the beginning, middle and end’ (TEXT 2a above). Simplicius (c. AD 530) gives us further help in his commentary on De caelo by reporting a Pythagorean argument to clarify the point on the three dimensions (TEXT 2c = Simpl. in Cael. 7.8.23-27): ...this the Pythagoreans showed as follows: the totality has beginning, middle and end, and as such is delimited by the triad. And perhaps we say of the totality that it is complete on account of its having 21 Simplicius already expressed a similar remark in the early sixth century AD: ‘It is worth noting that Aristotle has uncharacteristically made use of Pythagorean proofs (endeixis) in the service of demonstration (apodeixis)’ (On Aristotle On the Heavens 1, p. 9.10-11 Hankinson). His point is thus twofold, as Hankinson (2002) ad loc., n48 clarifies: ‘The argument is uncharacteristic, not only because Aristotle rarely has much favorable to say for the Pythagoreans, but rather [sic] because this sort of appeal to plausibility has no place in a properly-organized science (see Top. 1.1).’ 22 In this context, it could be related to Philolaus’ view that the universe started with things that are ‘unlimited’ and things that are ‘limiting’, which he says apply to ‘both the world-order as a whole and all the things in it’ (44B1 DK). 23 Guthrie (1939) 5 (note f) mentions Stengel (1920) for more on the religious practice. In De caelo Aristotle also brings in common usages of linguistic expression to support the connection between ‘three’ and ‘all’: ‘Our language too shows the same tendency, for of two things or people we say ‘both’, not ‘all’. This latter term we first employ when there are three in question; and in behaving thus, as I have said, we are accepting nature herself as our guide’ (268a16-18, tr. Guthrie). Baltussen 10 beginning, middle, and end. For what is not a totality lacks something by comparison with the totality, and it is incomplete. This information is not found in Aristotle, and one suspects Simplicius may have used Alexander of Aphrodisias (AD 200) on this issue, since he goes on to discuss the latter’s argument (7.8.28ff.). Whatever the origin, it is clear how the Pythagorean position can be interpreted, if one wonders exactly how the all-encompassing role of the triad is to function: ‘beginning, middle and end’ is one way of saying ‘everything’, and within the context of physics Aristotle re-interprets it, plausibly and more concretely, as the three dimensions. Returning to Ion now, it is obvious that he is less detailed in his claim, though not less ambitious in scope: by stating that ‘all things are three, and nothing is either more or less than those three’, he makes the number three all important and omnipresent. As astutely recognized by Dover (1986: 29), the second sentence (‘of each one thing the excellence is a triad, intelligence and power and fortune’) creates a problem for an ancient reader. Should the first part be taken to refer also to basic elements of the universe, as is common in cosmological works? Here we might ponder whether there is a connection with another fragment in which Ion is said to have chosen fire, earth and water as principles of nature (115* Leurini = 36A6 DK [second text] = Philop. De gen. et corr., p. 207, 16-20 Vitelli). This is, however, not quite clear from the statement. The claim would also imply that both the animate and inanimate are included. If correct, this creates the further puzzle (also in Dover 1986: 29) of how to grasp the implication that fortune, skill and understanding (sunesis) are present in both these spheres of existence. Dover rejects this possibility outright (p. 29): ‘although there are circumstances in which inanimates could be regarded as having power (MTCVQL) and fortune (VWZJ), they do not have understanding (UWPGUKL).’ This idea - that understanding has a universal presence in the world - may not be as far-fetched as he assumes. If we consider the kind of view held by Empedocles, an admirer and follower of Pythagoras, or when we think of Anaxagoras and Diogenes, who made ‘mind’ all-pervasive in the universe, there is some room for manoeuvring Baltussen 11 (see West 1984: 76, and my note 24). Empedocles’ view that all things in the world had sensation and understanding in them comes from a reliable source (Simpl. in Phys. 331.10 = 31B103 DK): ‘and one may find many such statements to bring in from Empedocles’ Physics such as this: “all things are capable of thought through the will of Tyche”.’24 Given the close links of both authors to Pythagoreanism25 and note tyche - Ion could have held a similar view. Nevertheless, to say with Dover (1986: 29) that it is ‘possible to reconcile the opening of the Triagmos with the attribution to Ion of a three-element theory’, requires some qualification. Given that the second part of fragment 36B1 DK (TEXT 1 above) shifts the attention to evaluative terms and ethics (excellence, intelligence), it is not likely that he was referring to three elements in this particular statement. This is not to say that he did not have a three-element theory: rather, the first fragment presents the overall philosophical position, in the style of Presocratic thinkers, whereas the information on the three elements must reflect the core of his ‘Physics’. Isocrates and Philoponus, most likely following their source, would only be reproducing an existing list. Separate from physics is ethics, no doubt the more important level of his ‘system’, in which arete is described as something arranged structurally, which could also ‘be predicated of entities other than persons’ (Dover 1986: 29). Therefore, I would prefer to think of the triad and the three elements as manifestations of the same fundamental idea of ‘threeness’, the one a global, overarching law, the other a concrete and area-specific feature of the physical world.26 As mentioned above, this would concur with the Pythagorean perspective of making an effort to describe - crudely put - the world (physics) and mankind’s role within it 24 For sensation see also 31B102 DK (= Theophr. Sens. 21): ‘thus everything has received breathing and smells.’ Ion’s proximity to Empedocles may be further supported by verbal echoes found in 68* Leurini = TrGF 19 F53f = 36B3b DK = Photius C 3262 and an elegiac poem (see next note). 25 See, for example, 31A7 DK (= Simpl. in Phys. 25.19, quoting Theophrastus; tr. Kirk, Raven and Schofield (1983) 281 [no. 335]): ‘[Empedocles] was an emulator and associate of Parmenides, and even more of the Pythagoreans...’; and 31A11 DK (= Athenaeus 3e [DK wrongly 5e], tr. Gulick): ‘a Pythagorean and abstainer from living beings.’ Dover (1986) 31 notes that the phrase used by Ion for Dionysus - MCNYP GRKJTCPG GTIYP, ‘‘achiever’... of actions which we find attractive or admirable’ (89 Leurini = 26 West = Ath. 447d; tr. Dover) - strongly resembles Empedocles’ description of Pythagoras as UQHYP GRKJTCPQL GTIYP, ‘‘achiever’... of actions which reveal wisdom, skill, understanding’ (31B129.3; tr. Dover). Cf. note 33. 26 The emphasis on ‘each one thing’ might imply there is no room for a global, over-arching law, but on balance it would seem that Ion expressed the universality of his view via each individual constituent of the world. Baltussen 12 (ethics). In sum, Dover’s statement only holds in the qualified sense that Ion was talking about the three elements indirectly. 4. ION’S PYTHAGOREANISM: REAL, PARTIAL OR IMAGINED? Thus far, we have found a number of indications which position Ion sufficiently close to Pythagoreanism. What else can we learn from the sources to assess Ion’s allegiance to the Pythagorean perspective? Pythagoras crops up in several places. We know Ion spoke highly of Pythagoras (92 Leurini = 36B4 DK = 30 West = D. L. 1.120) and it is a plausible suggestion27 that he was attempting to rehabilitate Pythagoras after some of the criticism from, for example, Heraclitus, who disparagingly refers to his knowledge as polymathia and kakotechnia (22B129 DK = D. L. 8.6). Yet the only other fragment which is explicitly allocated to the Triagmos (116(I) Leurini = 36B2 DK [second text] = Clem. Al. Str. 1.131) mentions Pythagoras as having written certain works which (Ion claims) he then attributed to Orpheus. Such a remark is far from unambiguous, and, as Dover suggests (1986: 31), could be a stab at the great man’s credibility (disputed authorship). In 92 Leurini = 36B4 DK = 30 West = D. L. 1.120 a further doubt is cast on his intentions, when he is reported to have said that Pherecydes must be living a good life after death, if (GK™RGT) Pythagoras was right. My paraphrase of this passage is rather crude, because the text is far more ambiguous, and again, Dover is illuminating. He emphasizes that GK™RGT has sceptical overtones, veering towards ‘if indeed’ rather than having the force of a neutral conditional. Callimachus’ report on the disputed authorship for the Triagmos (see TEXT 1) is but another twist of fate that has befallen Ion. Another piece of information may be relevant. Aristotle mentions a man by the name of Xuthus (Phys. 4.9; see 33 DK), while discussing whether there is void in the universe. From the context, we can infer that he is talking about the view of ‘some thinkers’, and Xuthus is mentioned as an example. He seems to have argued that if there were no void (empty space) that the universe would either not have any movement or it would ‘bulge’ or ‘rise in waves’ (MWOCPGKVQ? QNQP) as a result of the processes of rarification and condensation (expansion and contraction of 27 Huxley (1965) 40. Baltussen 13 water and gases). This is not a clear-cut Pythagorean view. Huxley (1965: 38) considers whether this might be Ion’s father, who could have been a Pythagorean. The evidence for this is scant, but cannot readily be dismissed.28 It is likely that natural philosophy and cosmology were known to him from his youth: Chios and Samos are relatively close geographically, and we know of other famous Chians working in the sciences.29 More important still is Ion’s ethical stance, since ethics is the area in which Pythagoras gained his reputation. In 114 Leurini = 36A6 DK = Antid. 268, Isocrates refers to Ion in the context of presenting Pythagoras as an example of high morals. Pythagoras is famous for establishing a community which adhered to high standards of moral behaviour. He outlined rules for proper diet, right actions, clothing, etc., through which the Pythagoreans came to be regarded as a sect. Such a strong interest in ethical matters, in particular in the structure of goodness, is present in our TEXT 1 (36B1 DK), and confirmed in Plutarch (TEXT 3 = 118* Leurini = 36B3 DK = Plut. De fort. Rom. 316d [cf. Quaest. conv. 717b]): Ion the poet, in the work he wrote without metre and in prose, says that fortune, although a thing most dissimilar to wisdom (UQHKC), produces very similar effects. (tr. Barnes) Here we find an interesting observation which compares ‘fortune’ and ‘wisdom’ (Barnes 1987: 223) or ‘skill’ (Dover 1986: 29): the focus of this comment, stating that fortune and wisdom may be different in some ways, but produce similar effects, points to the kind of reflection upon the relation between cause and effect, and the fact that similar outcomes may have different causes. In this case, the spheres of fortune and skill/wisdom can almost be read as ‘destiny’ and ‘human endeavour’ respectively, or perhaps as ‘accidental events’ versus ‘intentional (willed) results’. Whether this means Ion wants to convey a sense of resignation, or whether he 28 One source is Harpocration (TEXT 1); the other source is the fourth century AD Neoplatonist Iamblichus (VP 267) who names Buthus in a list of Pythagoreans from Croton (South Italy) - a possible corruption of Xuthus? Huxley (1965) 38-9 leaves it at that. 29 Huxley (1965) 39: the mathematician Hippocrates and the astronomer Eunopides of the fifth century BC. Baltussen 14 intends his audience to think harder about what humans are capable of, is not easily determined. We should not overlook the influence of early Greek (‘Ionian school’) philosophy on Pythagoras’ approach to nature. Kahn (1974: 170) is no doubt right that Pythagoras was familiar with the Ionian thinkers before he left Samos. He has become known as the first Greek thinker to combine ‘mathematical abstraction’ to a high degree in analyzing the world.30 In short, these three components - ethics, cosmology and number-theory - can be said to have produced Pythagorean philosophy. Significant in the ethical guidelines, which are mostly found in short maxims called akousmata, is that they often fail to provide the rationale for their existence - which is perhaps why some categorize them as riddles (Ath. 452d). They were probably committed to memory, contained mostly rules of abstinence and prohibitions, ranging from the injunction not to eat beans, cocks and certain fish to such prohibitions as ‘do not poke the fire with a sword’ (that is, do not provoke an angry man) or ‘do not sit on the corn ration’ (live not in idleness: see Kirk, Raven and Schofield 1983, no. 276). No doubt the gnomic style is partly due to oral culture since it contributes to easy memorization. Next, there is the importance of number theory and the harmony of the heavenly spheres which are probably well-known. I note especially four specific maxims to illustrate the number symbolism:31 • the source of wisdom is the tetractys; • the wisest is number, second is the man who assigned names to things; • the wisest thing in our power is medicine; • the most powerful is knowledge. These alert us to certain concepts which were highly valued, and how they came to be expressed in number theory. This theory takes its starting point in the tetractys, the set of four first digits (1, 2, 3, 4) which, when added up, produce ten, the number 30 A move which he allegedly made after noticing that metal pieces of different length produce different sounds (Xenocrates fr. 9 Heinze). 31 Iambl. VP 82 = 58C4 DK. Translated in Kirk, Raven and Schofield (1983) no. 277. Baltussen 15 referred to as ‘perfect’ or ‘complete’ (both senses expressed in the Greek word VGNGKQL), since it ‘includes’ all numbers. Here we are back at the general claim of Ion’s philosophy as discussed above (TEXT 1). It is easy to see how the aspect of comprehensiveness and perfection implied in the number 10 became associated in Pythagorean cosmology with the whole cosmos, the All. These are roughly the main features which may account for the potential appeal of this system. It was ambitiously comprehensive, ethically sound, and (in some sense) wonderfully explanatory. Ion might have found appeal in such a system, given his cosmological and ethical statements, but doubts remain as to how successful his foray into philosophical territory is. One fragment related to cosmology might seem somewhat naïve (TEXT 4 = 117* Leurini = 36A7 DK = Dox. Gr. 356b21 = Aët. Plac. 2.25.11): [About the essence of the moon]. §11. Ion: a body (UYOC) partly of a translucent shiny nature (WGNQGKFG?L FKCWIGL), partly dark (‘lightless’, CHGIIGL). The doxographer, in his list of opinions of natural philosophers, includes Ion in a lemma on the ‘essence of the moon’: these philosophers were arranged in such a way that they range from viewing the moon as ‘pure fire’ to ‘an earthlike substance which can be hot or cool’ in a more or less logical progression.32 Ion has been incorporated among the Ionians who regard it as bodily, after Thales (§8: geode), Anaxagoras and Democritus (§9: stereoma), and Diogenes (§10: anamma, ‘ignited mass’). Ion’s observation on the moon as partly ‘translucent shiny’33 and partly 32 Runia (1989) 254. Runia’s analysis is particularly useful for understanding the way in which Diels has reconstructed the Aetian lemmata from two texts: Pseudo-Plutarch (date unknown) and Stobaeus (c. AD 500). Ion is only to be found in the Stobaeus part of the text: Diels (1879) 356, ll. 21-22. 33 There is a temptation to translate WGNQGKFG?L as ‘glasslike’ (for example, LSJ referring to Galen, second century AD), but this could be misleading (as my colleague Dr. Margaret O’Hea, an expert in ancient glass, points out to me), because glass before the Hellenistic period was never fully transparent in the way that Roman or modern glass can be. She writes, ‘The Neo-Assyrians and the Achaemenid Persians after them did produce transparent and colourless glass, but most Greeks would never have seen one of these, not even a Perikles - even Alkibiades didn’t have glass vessels in the forced sale of his confiscated goods, and Persian bowls, though they appear in Greek graves, are scarce as hens’ teeth (a couple over two centuries). Greek glass was normally so strongly coloured (and small-scale) that even though translucent, it would appear opaque, but not matte. The glass vessels dedicated on the Akropolis were probably such as these. My point would be that the term Baltussen 16 ‘dark’ only makes sense when explaining a full moon, if we suppose that it is like a marble, half shiny, half opaque, so that by rotating it would seem to us to change its shape progressively, thus producing the four quarters. At any rate, a far more striking point is that the word for ‘lightless’ (CHGIIGL) is also found in Sophocles (OC 1481, 1549), whom he knew personally,34 and that ‘shiny’ (FKCWIGL) is a term used by the Pythagorean Philolaus to describe the sun (Aët. Plac. 2.20.12 = 44A19 DK). The latter occurrence speaks in favour of the term as authentic, and may imply that the text ought to be a ‘B’ fragment, not an ‘A’ testimonium. These are further examples of Ion’s allusive and creative style which associates him with both tragedy and philosophy. Such a creative use of language can be further illustrated by his interest in inventing compound words reminiscent of Homeric language.35 His view on the physical principles of the universe, referred to by Philoponus, is straightforward, and earned him a place in a convenient scheme of an increasing number of elements.36 But this seems the extent of his - for us observable - fame as a natural philosopher: Philoponus refers to him by the (no doubt traditional) title ‘Ion the tragedian’. ‘glass-like’ to a classical Greek would have been more likely a reference to something very shiny and bright - no coincidence that the eyes of cult statues were normally made of coloured glass at this time, and gods’ eyes were flashing/bright’ (personal communication, January 2004). See also Weinberg and McClellan (1992) 21f. and especially Stern (1997) [I owe these references to Dr. O’Hea]. 34 See the anecdote from his Epidemiai quoted in Athenaeus (104 Leurini = FGrH 392 F6 = Ath. 603e-604d). I note that Ion, in Omphale (27 Leurini = TrGF 19 F24 = Ath. 690b), also shared a word for a certain perfume (DCMMCTKL) with Sophocles (and Aeschylus), and he seems to echo Euripides’ CGTUKIWKQP (‘limb-lifter’) by hisCGTUKPQQL (‘lifting the spirit’, said of wine: 86* Leurini = 744 PMG = Ath. 35de; see Gulick ad Ath. 446b). Less convincing is Athenaeus’ suggestion that, in Omphale, Ion borrowed the descriptive elements used of a glutton from Pindar (fr. 168b SnellMaehler): 31 Leurini = TrGF 19 F29 = Ath. 411b. 35 On compound words see, for example, UWPSGVQWLNQIQWL (‘combination words’: T4 Leurini = 36A3 DK = Suda s.v. Ion Chios); UGNJPQ>RG@VJ, ‘moon-fallen’,in the lyric fragment discussed by West (1983a) 46 (95* Leurini = TrGF 19 F66a = Philodem. De piet. p. 13 Gomperz; tr. Campbell 1992 = fr. eleg. 30A); FCMVWNYVQP, ‘with finger-like handles’ (of wine cups: see LSJ s.v.), from Ion’s Agamemnon (1 Leurini = TrGF 19 F1 = Ath. 468c); and see previous note. On his Homeric interest, compare his use of certain Homeric words (again for wine cups): HKCNJ (25 Leurini = TrGF 19 F20 = Ath. 501f) and MQVWNQP (61* Leurini = TrGF 19 F51 = Ath. 478b), for which he was criticized by Epigenes as having misunderstood its meaning - see Leurini (1983) and (1984). On other instances of Homeric language see Clarke and Stevens in this volume. 36 115*(I) Leurini = 36A6 DK [second text] = Philop. De gen. et corr., p. 207, 16-20 Vitelli, tr. Williams (1999): ‘Having discussed those who suppose that some one of the four [elements] is the principle of bodies, he now sets out [the views of] those who suppose more than one principle, either some of the four [or all: e.g.] fire and earth were suggested by Parmenides, these together with air by Ion of Chios, the tragedian, and four by Empedocles.’ The omission of water seems odd. Note that Philoponus reports a different set of principles later on (fire-earth-water: 115*(II) Leurini = 277, 14 Vitelli); it is impossible to say whether he nodded off, had different sources at his disposal, or whether it is simply an error. Baltussen 17 So let us return to our main question, now in more specific form: is Ion a Pythagorean in his use of the number three? The way in which he uses it (structural analysis of virtue) and the abstract level at which it operates (applied to the animate and inanimate as well as concepts) are good indicators for such a view. One can sympathize with (and perhaps marvel at) the ingenious intuition that number can somehow be used to reach an abstract understanding of certain concrete aspects of reality, but it is pressing it too far to say that numbers can generate physical things (Aristotle), given that they themselves are not physical. There is present, however, a palpably distinct position which allows us to say that, for Ion, Pythagoreanism is more a source of inspiration than a matter of strict allegiance. One aspect may still justify drawing him closer to the Pythagorean doctrine, and that is to do with the way in which the Pythagoreans represented numbers. Here it will help to visualize the way in which numbers were constructed: their idea of numbers is very much geometric, viewed in a spatial way as an arrangement of dots or pebbles. For example: 3= 9= 4= 10 = This shows how numbers were given a geometrical nature, thus making ‘shape’ a factor in how regularity could be associated with even numbers, and irregularity with odd numbers (with nine it only works in two rows, not three).37 It also makes explicit how the number three returns as a triangular shape associated with all numbers (tetractys: 1 + 2 + 3 + 4 = 10; and musical ratios: 1:2, 3:2, 3:4) - the triangle being the first geometrical figure. Thus, three acquires special meaning by 37 Compare Simpl. in Phys. 457.16-17: ‘it is the habit of Pythagoreans to illustrate by [drawing] figures (UZJOCVQITCHGK¨P)’. Baltussen 18 signifying (almost magically) something which is finite and yet all-encompassing. In short, it represents confinement and comprehensiveness. 5. THE TITLE Let me offer a discussion of an issue related to the second problem broached at the start of this paper, namely What does the title of the work mean? Despite the considerable number of articles on Ion, little analysis has gone into this work’s form and meaning (exceptions are von Blumenthal 1939: 18 and Janko 2001: 7).38 Given that a title could provide a valuable hint about the intended aims and content of a work, this would seem an omission. But we should tread with considerable caution. Giving a title to a philosophical prose work was not common until the Hellenistic period. As a rule, titles originated in first words (incipit) or thematic tags, and these were typically used as a title for a scholarly environment in which labelling and organizing writings are important, such as Aristotle’s Lyceum or the Alexandrian library. And this is not all: I note that the verb used to indicate ‘title of a book’ (GRKITCHGKP) is sparse before the Hellenistic period.39 Moreover, a study of the title commonly used for works of early natural philosophers, On Nature (RGTK? HWUGYL), has shown that, as a rule, this is absent from the original works and those of much later origin.40 This particular title was subjected to a curious process of historical distortion whereby the attribution of titles seems to become more confident and certain as time progresses (that is, as the sources become more distant from works at issue). Therefore, it is not inappropriate to be somewhat sceptical about the question of whether it was Ion’s own.41 The evidence for a title is (how could it not be?) threefold: we find VTKCIOQL, .QUOQNQIKMQL and RGTK? OGVGYTYP (latter two both in T8 Leurini = 36A2 I cannot agree with Janko (2001) that it is connected to VTKC\Y, ‘throw down dice, win’ (wrestling), as it is disconnected from an assessment of the context. See further below. 39 A search of TLG CD-ROM (E) for the noun and the middle verb form third person singular indicates that the stem occurs once in a small number of fifth century authors, including Thucydides and Plato, and then starts to proliferate after Aristotle (twelve instances), until it abounds in authors of the second century AD. Even here some distortion of the evidence may have occurred, since often the word is due to the (later) source author who mentions the (supposed) title of a work. 40 Schmalzriedt (1970). 41 Note similar doubt in Dover’s crisp parenthetical remark (1985: 28): ‘a philosophical work which was called (not necessarily by him, and even after him not always by everyone) 7TKCIOQL or 7TKCIOQK.’ 38 Baltussen 19 DK = Schol. Ar. Pax 835). The first one is the most quoted form and is found in three variations: (1) VTKCIOQL in the singular. This is found in Harpocration (post-Hellenistic).42 (2) VTKCIOQK in the plural, as in Demetrius of Scepsis and Apollonides [both quoted by Harpocration]; Clement of Alexandria (c. AD 200), Diogenes Laertius (c. AD 200), and the Suda (c. AD 950).43 How such a variation from singular to plural might occur is not obvious. (I am reminded of MQUOQL/MQUOQK, a term related to cosmological context, which could be meaningfully employed in singular or plural in fact, much depends on it in the case of Democritus, who is supposed to have talked about multiple worlds, a claim denied by others. Pl. Tim. 55cd is usually taken as a reference to this view). (3) VTKCUOQK, according to Diehl (1916: 1864.2-3) an error: ‘VTKCUOQW?L irrtümlich’ (but see my comments below). In terms of word-formation, VTKCIOQL seems unique: I note that few cognates exist; and the noun suffix OQL seems to have no specific meaning except to create a noun (Buck 1933: 319-20). There are plenty of parallel formations for either IOQL or UOQL, but so far I have found only one which resembles ours in both ending (OQL) and in variation of the root ending: the word VTKIOQL / VTKUOQL - meaning ‘screech of a partridge’ - occurs in Aristotle (Hist. an. 614a22) and Theophrastus (Caus. pl. 5.10.5), or as ‘the grinding of teeth’ in Hippocrates (Acut. 6).44 Riedweg (2002: 82) has unearthed a semantic parallel of some kind: a work by Andron of Ephesus (date unknown) is called 7TKRQW?L, 42 Burkert (1972) 129 and n50 claims that Isocrates’ testimony is a guarantee for the authenticity, but for that Schol. Ar. Pax would suffice (T8 Leurini = 36A2 DK [second text] = FGrH 392 T2). 43 Diels does not report that the plural of the title also occurs in the Suda s.v. Orpheus (T9b Leurini = 36B2 DK = FGrH 392 F25c = Q 654), with the gloss that the work is said to be by ‘Ion the tragedian’ - a misunderstanding, according to von Blumenthal (1939) 20. 44 The formation which creates the variation of \ and I in stem-ending is related to the IndoEuropean suffix -yo- found in, for example, CIKQL and C\QOCK: Buck (1933) 317-8. Cf. Frisk (1960-72) 930, where the title is given, translated as ‘Triade(n)’, and said to be a denominative formation from VTGKL,VTKC (from which VTKCLFQLderives: p. 922, sub 4). Baltussen 20 ‘Tripod’ (see Euseb. Praep. evang. 10.3.6 = Porph. fr. 408 Smith; cf. D. L. 1.30; Clem. Al. Strom. 1.21.129). With these basic elements in mind, we can probe a little further into the morphology. There are, I believe, two distinct problems: for its meaning and translation, we need to decide how to parse VTKCIOQL; in addition, we should ask whether it is feasible that this word could fulfill the role of a title. As to the question of parsing the title, we have at least two options: (a) it could be a creative variation on VTKCL, VTKCFQL, the term for the number three associated with Pythagoreanism (e.g. Arist. Cael. 268a13; Metaph. 1081a34 and b12); or (b) we might think of it as constructed from the stem VTKCI/ VTKCU + OQL. Now, instead of linking it to CIPWOK, ‘breaking’ (von Blumenthal, Huxley), we may consider another possibility: the verb VTKC\Y / VTKCUUY can mean ‘throw down three times’ in wrestling (an unlikely context for producing a relevant link), or ‘multiply by three’. The latter meaning is found only late, once in Iamblichus, the fourth century Neoplatonist with Neopythagorean leanings whom we encountered before in the discussion of Ion’s father, Xuthus. Despite this link with Pythagoreanism, such a late occurrence can hardly be imposed upon a noun from the fifth (?) century BC. But how should we decide between these seemingly opposite options - ‘break or divide into three’ or ‘multiply by three’? ‘Division’ may fit the idea of structure slightly better (see above, on ethical concepts), but ‘multiplying’ could also stand for the nature of the universe as characterized by endless triadic sets. This part of the puzzle, it seems, resists further elucidation. All translations obviously attempt to find a term that can convey ‘threeness’ - and at the same time a notion which can represent multitude as well as a unit (in other words, a collective abstract noun). Unfortunately, few proposals really succeed: TREBLINGS (Kirk, Raven and Schofield 1983, no. 262) is rather uninformative; TRIPLE DIVISION or TRIPLES (Huxley 1963: 39 following von Blumenthal’s DREITEILUNG) hangs on a tentative connection with CIPWOK, ‘to break’; Diels’ DREIKAMPF (DK 36, p. 379) is not explained and not particularly plausible (does he mean that VTKCIOQL derives from VTKCIYP?); Dover’s THREE or Baltussen 21 THREENESS, though simple, does nothing to advance our understanding of the content of the work. Perhaps Barnes’ TRIAD, the most recent and modern rendering, fulfils both criteria of multitude and unit. At this point, we should return briefly to the question of whether this title could have been Ion’s. First some facts: as far as we can tell, Ion only uses the numeral three and the abstract noun VTKCL. This shift from ‘three’ to ‘triad’ is an interesting move to a more abstract level. When the title first occurs in Harpocration (postHellenistic), uncertainty about its form (singular or plural) was already apparent. Early titles usually grow out of the prepositional phrase ‘RGTK? X’ as a contentindicator, so a singular noun is also the more unusual choice (Andron’s Tripod is probably fourth century BC). As noted, the content might not have been known to Harpocration, given his qualifier: the indefinite pronoun VK. Again, titles are a bookish affair, and often serve a practical function in the referencing or storing of books, and the need for such a function may not have arisen before Aristotle. Therefore, even Clement’s seemingly confident attribution of a passage to the Triagmos (116(I) Leurini = 36B2 DK [second text] = Clem. Al. Str. 1.131) may not be as secure as would appear. On the other hand, this title is unique, as is the Kosmologikos, so we do have a possible argument in favour of its authenticity. For the second title we may have a small but tenuous point in favour, because the scholiast (mostly ninth-tenth centuries AD, but some go back to Didymus) uses the verb HGTGVCK, which in such contexts means ‘is preserved, has been transmitted’. Can we argue either way with absolute certainty for a hapax legomenon? Until we have a text from the work in which either is used, I suspect we cannot. Nevertheless, all things considered, in particular Ion’s original use of language (on which see, for example, Clarke in this volume), he was no doubt capable of creating such a word, if only to claim a niche within the Pythagorean philosophical tradition. 6. CONCLUDING REMARKS In any discussion about Ion of Chios, the challenge of speculation will always be exacerbated by the persistent gaps and uncertainties of the evidence. However, not all speculation is fruitless. I have presented a new account of the extant fragments that constitute the ‘philosophical thought’ of Ion of Chios. I have included all the Baltussen 22 fragments (suggesting the B fragments might be augmented by one, viz, 36A7 DK = TEXT 4),45 and provided more context to Ion’s thought, without imposing too rigorous a coherence upon these scattered bits and pieces - a temptation which always presents itself when dealing with fragments. Ion is one of those figures who lost out in the struggle for literary survival. In ancient times Ion’s literary output, though at some stage part of the canon, was relegated early on to the ‘second eleven’,46 the justification for which is now lost.47 It is no surprise, therefore, that this fate has been perpetuated even in modern perspective, given that even his standing as a lyric poet has been eclipsed by others.48 Except for some hints and brief references of near-contemporaries (Plato, Aristophanes, Isocrates), most of our sources are much later, and these all seem to come from a learned environment, its pivotal centre being Alexandria:49 the important sources such as Clement, Diogenes Laertius, Harpocration (possibly based on Callimachus50), the Suda (often based on D. L.), Philoponus and the Scholia on Aristophanes all point in this direction. Plutarch might seem an exception, but he was a voracious reader and punctilious maker of notes, basing his writing on a wide range of sources.51 One senses that Ion’s somewhat idiosyncratic and eclectic writings were only appreciated by learned readers, and that soon after his death the survival of his work depended on the specialist interests of scholars such as grammarians, historians, anthologists and philosophers. He did have one author write a book about him, the otherwise unknown Baton of Sinope mentioned by 45 TEXT 4 = 117* Leurini = 36A7 DK = Dox. Gr. 356b21 = Aët. Plac. 2.25.11. Diehl (1916) suspects that soon after the third century BC his original works had disappeared. Contrast West (1985) 71, who thinks that they were still available up to the third century AD. 47 Surely Pseudo-Longinus’ comment is a later verdict: T17 Leurini = 36A4 DK = TrGF 19 T6 = Subl. 33.5. 48 It may just be the result of practical considerations, but his marginal status today is summed up by Campbell’s Loeb volume of Greek Lyric, vol. 4, which bears the title ‘Bacchylides, Corinna, and Others’ [my emphasis] among whom Ion is listed as ‘also included’. However, as noted above, Barnes (1987) gave him a separate (one page) section. 49 I note Diehl’s comment (1916: col. 1865, ll. 29-33): ‘vor allem Ions Tragödien im Altertum Gegenstand philologischer Studien von seiten der Lexikographen und Fundgruben für Verfertiger von Anthologien and [sic] Exzerptenliteratur’. 50 As suggested by Huxley (1965) 39n48, presumably based on T9a, 114 Leurini = 36A1 DK = Harpocr. s.v. Ion, where Callimachus is mentioned. The problem of polyeideia and categorization was already raised in antiquity, as T15b Leurini = Diegesis to Callim. Iamb 13 suggests: see Pfeiffer (1949) 205-7 (fr. 203); cf. TT22a, 24 [=T9a] Leurini = fr. 449 Pfeiffer (p. 348) on the attribution of the Triagmos in Callimachus’ Pinakes. See also Henderson in this volume. 51 He even knew Latin (pp. 19-21): see Lamberton (2001) 13 (with further references). 46 Baltussen 23 Athenaeus,52 and one Epigenes is said to have interpreted his expressions (see note 35). It is a reminder of how little is needed to become marginal or a curiosity, irretrievably lost for posterity. Ion will remain an intriguing and elusive figure. It is unclear what purpose this philosophical prose work may have served, and its fragmentary state will not allow a full answer. Dover speculates that it might be a creative and playful ‘exploration of the implications of an idea’ in the manner of poets, and he compares Pindar’s Second Olympian with its unusual allusion to afterlife and a cycle of incarnation, which also links him to Empedocles.53 This seems quite plausible. What is clear is that by the time of Harpocration (TEXT 1) the content of the Triagmos was something of a mystery for the grammarian (or his source): to me, the indefinite pronoun in VKUWIITCOOCHKNQUQHQP signals no direct knowledge of the work itself. Ion was a resourceful author, willing to try his hand at anything; maybe he was joining in with the fashion of the day, which was to consider and present an explanation of the physical world in prose. This is testimony to his versatility, as Callimachus confirms, but also, inadvertently, it may have meant that he was spreading himself too thinly. As a result, a remarkable reputation in any specific area was never his. My guess is that Isaiah Berlin would probably have categorized Ion as a fox - a type of author who knows many things - as opposed to the hedgehog, which ‘knows one big thing’.54 If he had, it would be quite appropriate, since in one fragment Ion expresses a dislike of hedgehogs (surely a stab at Archilochus?), preferring the bravery of lions: the hedgehog, when the enemy appears ‘winds its spiny body in a ball, and is invincible against bite and touch’ (44 Leurini = TrGF 19 F38 = Ath. 91d; tr. Gulick). %CVYPFsQ6KPYRGW?L[see FGrH 268]GPVQK¨LRGTK¼¹,YPQLVQWRQKJVQW, ‘Baton of Sinope in his work On Ion the Poet...’ (tr. Campbell 1992; 94 Leurini = FGrH 392 T8 = 31 West = Ath. 436f; Ael. VH 2.41). Jacoby (Commentary on FGrH 268, p. 211) thinks that the book might have included interpretations of poems, but admits that it is unclear why Baton would have been interested in Ion. 53 Dover (1986) 31 after Sandbach (1958-59). Pindar’s Ode was in honour of Theron, tyrant of Acragas, native city of Empedocles. Dover (pp. 31-2) points out in passing that the word kruphos (concealment) only occurs at Pind. Ol. 2.97, where he clearly uses Pythagorean ideas of reincarnation, and in Empedocles 31B27.3 DK (see also notes 24 and 25, above) 54 See Berlin (1953), where he explains how he came to use these labels for characterizing the great minds in history: they originate in Archilochus fr. 201 West (c. 650 BC). 52 Baltussen 24 Perhaps we may comfort ourselves with the thought that more of the work might not necessarily have given us more of his ‘philosophy’. Like the Presocratics with whom he wished to be affiliated, he may have been giving an outline, which represented ‘little more than a provocative preview’55 in written form. This might have been built upon further during oral exchange between teacher and pupil in the more formalized philosophical seminar. I somehow doubt, however, whether it was the objective of this polymath to acquire pupils and teach Pythagoreanism. His virtuosity in many areas of literary expression evokes the image of a brilliant raconteur and writer who contributed to, and possibly ‘pioneered’ other ‘genres’. His contribution to philosophy is hardest to assess, since it has barely survived: we see a versatile mind capable of handling abstract notions at the juncture between philosophy and literature.56 However, for a better grasp of content and argumentative structure, we would need more than a grand opening statement and some ‘mental sound bites’. I therefore agree with Dover (1986: 30) that Ion wore his Pythagoreanism rather lightly, trying his hand at yet another mode of expression.57 His interest in Homer, tragedy and lyric poetry indicates his position as a ‘public intellectual’ providing comments on social and moral issues (like Xenophanes), which could mean that the cosmology he outlines was a more perfunctory part of the overall account. We should not forget that the split between philosophy and literature had not yet reached its full measure. But, as we saw, Isocrates does give him a place among ‘experts’, which must indicate some kind of recognition for his (perhaps) one and only attempt to philosophize. If he did anticipate Plato’s dialogue format, as Dover plausibly suggests (1986: 34-5), it should be added that it was no doubt Plato who outshone Ion to such an extent that he was not remembered for this contribution. The dialogues by other authors prior to, and contemporary with, Plato have almost vanished. Ion may thus be regarded as a kind of catalyst, whose often original forays into the different fields of the literary landscape caused him to be scattered like light through a prism. All that remains for us to see is the glimmer of a long extinguished star. 55 Tarrant (1996) 135 on the writings of the Presocratics. See by West (1985) 76 on this point. 57 Cf. Diehl (1916) 1864, 46-7, who thinks the Triagmos was written ‘in Anlehnung an die Pythagoreer’ [my emphasis]. 56