1. INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, sorne thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality oC this reproduction is depeadent upon the quality oC the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed~ a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs inc1uded in the original manuscript have be en reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.UMI Unlverslty Microfilms Internatlonal A Bell & Howell Informatlon Company 300 North Zeeb Road. Ann Arbor. MI 48106-1346 USA 313:761-4700 800:521-0600Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

2. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3. Order Number 9317187Words' semantic constitution as a guide to reality: The "Cratylus" reconsidered Levin, Susan Barbara, Ph.D. Stanford University, 1993Copyright @1993 by Levin, Susan Barbara. AH rights reserved.UMI 300 N. Zeeb Rd.Ann Arbor, MI 48106Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5. WORDS' SEMANTICCONSTITUTION AS A GUIDE TOREALITY: THE CRATYLUS RECONSIDEREDA DISSERTAnON SUBMITIED TO THE DEPARTMENT OF PHILOSOPHY, THE GRADUATE PROGRAM IN HUMANITIES, AND THE COMMlTfEE ON GRADUATE STUDIES OF STANFORD UNIVERSITYIN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY INPHILOSOPHY AND HUMANITIESBy Susan LevinMarch 1993Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6. Copyright 1993 by Susan Levin AH Rights ReservediiReproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7. 1 certify that 1 haye read this dissertation and that in rny opinion it is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.kCh k ~,~::9 l.,Julius M. Moravcsik (Principal Adyiser)1 certify that 1 haye read this dissertation and that in rny opinion it is fulIy adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.Alan Code1 certify lhat 1 haye read this dissertation and that in 111y opinion it is fulIy adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.1 certify that 1 haye read this dissertation and that in 111y opinion it is fulIy adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.~UL "~'--~Andrea Ni;;4ngale (Graduate Program in Humanities)Approved for the University COI11miUee on Graduate Studies:iiiReproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8. Abstract Accounts of the Cratylus' historical sources, which focus typically on philosophers and sophists, have yet to explain why etymology has such a pivotal role in the dialogue. Based on extant evidence one may treat Plato's discussion of etymology as a critical response to techniques and assumptions that were of central importan ce to a non-philosopbical, literary tradition. Plato's initial positing of 'tXVll status for naming, based on criteria advanced in the Gorgias, is followed by a sustained challenge thereto. Recognition of tbis dynamic goes a long way toward addressing scholars' persistent concem with the Cratylus' apparent lack of cohesion. Though the dialogue's emphasis is negative, at the end Plato offers hints of bis own metaphysical theory and associated view of appropriateness in naming. He develops these elues in the Phaedo, where the notions of naturalness and appropriateness are tied directly to Platonic metaphysics. The literary tradition's handling of eponyrny constitutes a precedent for Plato's use of it to treat questions of appropriateness. Having rejected etymology in the Cratylus, in the Phaedo Plato revises eponymy based on his metaphysical theory. The most fundamental and closest links between the two dialogues do not rest on the use of Fonns per se; they center instead on the notions of naturalness and appropriateness, whose treatment in the Phaedo is closely tied to Forms. In the Sophist and Politicus, Plato concentrates on mutual relations between Fonns. Connections between the Cratylus and these dialogues do not rest on a shared interest in OlupEO'l 8p,OEt ~prov, OUOE A~Etat 1tplv rov' oc; O' E7tElt' ECPU, tplaKtTl po~ OtXEtat tUXcv' Zflva () tt~ 7tpo 9vte;ta 1tv'ta 1tpy~a'taleal. 1tcra; vo~a; dxov (Hist. 2.52). As mentioned, there is another set of instances in which considerations involving the namer constitute the basis on which v~a'ta are assigned. Such v~a'ta ffiight be given based on various features or aspects of the namer. This category divides in tum into various subcategories. First, the assignment could be grounded on a physicaI or related aspect of the namer. An example of this type of case is found at the outset of Euripides'Helen, where the poet says that when she was born Proteus and Psamathe named their daughter Eim, 'to ~T'tpo; Ct.yAtcrlla (8-11). Second, a name might be given based on the namer' s emotional state. One finds a prominent instance of this in an Odyssey passage in which Odysseus' grandfather Autolycus indicates the basis on which he assigned his grandson's ovoJl.a: 1tOAAOtcrtV yap EYro ye ooucrcrJl.evo; 'to' Kvro ... 't!> o'Cry aloud without fear the victory of Zeus, you wilJ not have failed the truth. (tr. by Lattimore) 67 "Etymologica," 147. 68 "Etymologica," 148. 69 "Etymologica," 148.43Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 48. 'OOUOEU~OVOJ.1' EOtCO noovuJ.1ov (19.407-9).70 In another case, which centers00theacquisition of a nickname, CIeopatra's parents came to caJI their daughter "Alcyone" due to the mother Marpessa's grief at being snatched away by Apollo: 'AAK')VTV KaAemcov EnOOVUJ.10v, OVEK' ap' autii~ l.lltllP aAKuvo~ nOAunEv9o~ ottov Exouoa KAalEvo J.1tvKEpyO~avTtpnaoeol~o~'AnMcov (!l. 9.561-4).1 1Tbird, a name rnight be given based on an attitude or character trait of the namer. For example, Herodotus says that Cleisthenes ehanged the names of the Dorian tribes so as to make fools of the Sieyonians, Le., based on bis eontemptuous attitude toward them. The names were derived from the words for "donkey" and "pig," with the endings changed. While he named his owo tribe the 'ApXAao1.,ano 'tii~ COUtou apxii~, he named theothers Yi't(l1., 'OvEa.tat, and XmpEitat. (Hist. 5.68). In another instanee Euripides says that the son of Proteus and Psamathe was named eEOlCAJlEVO~ based on his father' s piety (He/. 8-10).72The Furies, as depicted by the tragedians, evoked great fear in mortaIs; this is evident from their approaeh to the naming of these deities. In Oedipus at Colonus Oedipus is toId, regarding the spot where he sits, that ai. yap EI.UPO~Ot. 9Ea mp' XOU01., rii~ 'te Kal ~K'tOUKpa,1. (39-40). Having asked under what name he shouId invoke them, he isinformed that the people of CoIonus prefer to address them as ta~ 1tv9' poooa~ EUJ.1EVoa~(42).1 3 The mortaIs in question assign them the nameEUJ.1EVOE~based on70 There are other passages in which the same verb is used with reference to Odysseus. These cases involve others' anger toward Odysseus hirnself: Odysseus is singled out as the object of Zeus' wrath (t v o taov roMaao, ZEU;) (1.62); Ino asks Odysseus, t7ttE tOl ilOE noaEloCllv EvoaxSCIlv rooaat' EK7tyI..CIl;, 'tOl KaKa 7tol..l..a CPUtEE1; (5.339-40); Odysseus indicates that he knows ro; IlOl ooc>oUatal ICA.Utoe; Evvoc:ryalO; (5.423); Odysseus tells Penelope (who do es not as yet know his true identity) that ooaavto yap ati) ZE; tE Kal. 'HD..lOe; because his cornrades had slain he kine uf Helios 09.275-6; om. in sorne MSS). 71 Strictly speaking the mother, whose grief leads to the additional assignrnent, is of course one of two namers. On this interpretation atl; refers to the daughter, and JllV to her mother Marpessa. Murray prints atr, rather than atfi;, and takes the JllV as referring to Cleopatra rather than to Marpessa, as is evident from his translation: 'The mother herself... wept because Apollo that worketh afar had snatched her child away." One point in support of the former reading is that the halcyon is known to cry when separated from its mate; 00 this reading Horner has a direct analogy in view. (1 owe this observation to Andl'ea Nightingale.) 72 The bracketed words oi SEOUe; a~CIlv ~ov Ol1veylC' are rejected by Nauck (concerning the approach to narning evinced by this passage cf. Plato, erar. 397a-c). The case of Theoclyrnenus is actually rather more cornplex than Euripides' remark suggests since presurnably the name was not given merely based on the father's attitude, but a1so based on the parents' hopes that their son, once mature, would display that same piety. Insofar as the name is given based on a wish regarding the son's nature, it could be accommodated in the first of the two basic categories as well, i.e., that in which appellations are given based on important considerations involving the bearer. 73 Lines 122ff. a1so show rnortals' fear of these deities: one trembles to name them due to their power to harmonon44Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 49. their fear, that is, based on the attitude they actually want these divinities to have toward them. This comes out clearly when the Chorus in effect asks them to be true to this assigned name with positive descriptive content: it teIls Oedipus to repeat the prayer that "as we call them EUIlEvioac;, which means the gentle of heart, may they accept with gentleness (~ eUIlEvrov) the suppliant and his wish" (486-7).74 One sees the same trepidation and reluctance to name the deities in Euripides' plays: In Orestes Electra says ovollsEtv yap aiooUllat 8Eac; eIlEvloac;, a't 'tvo' ~alltA.A.rov'tat

9apEv (34-6). (Nisetich's rendering of OallLrov EtEPO~ as "an evil power" does nol convey the fear of naming involved here, but rather masks it (Pindar's Victory Songs, 170).) In connection with the giving of euphemistic VlLata it is also worth noting authors' substitution of E'){>VU!1o~ for PlO"'tEPO~ as the adjective meaning "Ieft," presumably on similar grounds, namely, out of fear of the referent's power. In the case of Ec>VU!10~ this fear would be generated specifically based on what is characterized as being posirioned on the left: for instance, in Prometheus Bound Prometheus says that ya!11jfrovxrov 'tE 1ttfjO"lV oicovrov O"Ke9p~ 5lc>plO"', OltlV; 'tE 5E~lOl. q>O"lV EUrovlLou~ tE (48890). The context makes clear that it is what occupies that position is considered unfavorable or threatening. Conversely, the positive character of what is on the right is mentioned also by Horner: as Telemachus, taking leave of Menelaus, expresses his fervent desire that Odysseus be at home when retums to Ithaca, E1t1ttato 5E~lO~ opVt~, aiEtO~ pyTv xlva q>pcov VXEO"O"l1tl..ropov, ~ILEpOV e~ aUI..f~ o 5' i~OV'tE~ E1tOVtO VPE~ ~OE yuvailCE~ 5 O"q>tO"lV iyy9EV iA900v OE~lO; ~'~E 7tpcr9' l1t7troV o OE iOOVtE; yf9r,O"av, Kal. 1taO"lV eVl q>PEO"l9u!1o~ iv9J, (Od. 15.160-5). a bird flew by on the right, an eagle, bearing in his taIons a great, white goose, a tame fowl from the yard, and men and women followed shouting. But the eagle drew near to them, and darted off to the right in front of the horses; and they were glad as they saw it, and the hearts in the breasts of all were cheered. (tr. by Murray) Similarly, as Telemachus raises the issue of Zeus' punishment of the suitors, a hawk later flies by on the right; this leads TheocIymenus to infonn him that oi) tOl aVEU geou E1ttato 5E~tO; opVt~ eyvrov yp ILtv eO":vta iooov oirovov evta. !1E'tpou 5' OUK EO"n yVEO~ ~aO"tl..EtEpoV aUo EV 5~1Lq 'I9KT1~, U' !1e"i~ KaptEpol /lid (Od. 15.531-4). At the c10se of the poem, Odysseus-before identifying himself as such to Laertes-telIs him that he last saw Odysseus several years ago; moreover, ~ 't o ecr91..o1 EO"av opvl9E~. In addition, one votes, for example, with the right hand, as when Danaus telIs his daughters that the Argives acted in a decisive fashion, 1tav5T]lLlat yap XEPO"l 5E~lCOV!10l~ Eq>pt~EV aieT,p45Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50. In addition, anovo~amight be assigned based on a chance intersection of pathsbetween the namer and the recipient individual. 77 In Ion, Xuthus tells "Irov that he is receiving this name because he was the first one Xuthus encountered as he left (E~tV'tt ~Ot)Apollo's shrine (661-3). This analysis is later reiterated when the Chorus leader statesthat Xuthus caBed him "Irov', E1tE1tEp 1tpYco~ T1vtTVUJ,10~ to mean "Ieft" in the sense of "west" (see for instance 1.72). EuC>vuJ,lo~ is also employed in the literal sen se of "well-named," as for example in Pindar' s mention of feet in praising the speed of two victors (Nem. 8.47), and his reference to justice as "named in loveliness" (EUroVUJ,10v OKIlV) (Nem. 7.48). ~uarovuJ.1o~ is also used, as at Ajax 914 where the phrase oucrC>vuJ,10t; Atll~ appears; this reference assumes special relevance given thal earlier in the play Sophocles highlights the descriptive conlent of the hero's name. See in addition O.e. 528, where the poel refers lo Oedipus' oUcrrovUJ,l1l I..Ktpll. With regard to the issue of a reluctance lo name, it is interesting thal three times in the Odyssey Penelope makes reference lo "ill-fated Troy, not lO be named" (KIlICO'iAlOV OUK OVOJ,1llcrtiv) (19.260, 19.597, and 23.19). Finally, Herodotus declines to mention Osiris' name in connection with a certain activity performed in his honor (see 2.61). As concerns the omission of Osiris' OVOJ.11l see also 2.86; there Herodolus, discussing the Egyptian practice of murnmification, says that embalmers first produce samples in wood, of which the best and most expensive is said dVllt tO) OUK ocrlOv 1tOlEJ,1llt to ovoJ.11l Enl. tOtotCJl7tpiYJ.1lltl OVOJ.1~ElV. 77 Hirzel, somewhal more generally, mentions "ein zutaIliger Umstand (tX11)" as "namenschtipferisch" in this instance (87).46Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51. said that KalVOV aE 'toVOfl' aya xpvov 1I:E1I:Aaaflvov "1 rov , iv'tl afi8EV on auvtv'tE'tO (830-1).1 8 Third and finaIly, an individual might receive his or her ovofla based on sorne object associated with the namer. In this connection Sophocles has Ajax tell his son Epuaxrc; that he was named for the shield (aICoc;) which Ajax now gives to him (Aj. 574-6). Over and aboye the wide range of cases faIling in those categories and subcategories treated in what precedes, it is interesting to note that sometimes a particular ovofla will give rise to a set of statements or story which takes on alife of its own. For instance, in the Bacchae Euripides offers an explanation of how the tale arose that Zeus bore Dionysus by way of his thigh. Zeus gave an ether-formed Dionysus as hostage (oJ,Lllpov) to Hera. Over time "men garbled the word (ovoJ,La fJ.E'tao..t~aav'tE~) and said that Dionysus had be en sewn into the thigh of Zeus (EV fll1P!> Lll~)."79 Whereas, what actuaIly happened was that Zeus gave Hera a dummy as hostage for her son (ro/l~pEuaE) (288ff.). In another case Herodotus makes reference to a well-known story bult around the descriptive content of the name Kvro, which belonged to the woman who, with her husband, tookcare of the boy who was to become Cyrus. Once retumed to Cambyses' paIace he sang her praises: oi OE 'tOlCEC; 1I:apaAa~v'tE~ 'to ovofla 'tolho, 1va eEto'tpro~ 001(n 'tOtal I1pancrt 1I:EptEtva a'tlC; alhr KEXroPlllCE (1.122).80In still another instance, also found in Herodotus, it appears as though the writer draws conclusions about the nature of the hippopotamus based on its name, which he gives as '{1I:1I:0C;1I:0'tlltoe; (2.71): in this passage Herodotus claims that it 'tE'tp1l:0UV Ea't,aXllAov, 01l:Aat 130e;, atllV, AO OE oqaYlaa{}vn "tOtcp ovoJla ~v Arov' t:xa o' a.v "t1 "al "tOU ouvJlatOe; EltapOltO (Hisl. 7.180). (In another case of aman bearing the name of an animal Herodotus tells how Cleomenes expresses his hostility by exploiting the descriptive content of the name Kple;, which belonged 10 an especially conspicuous Aeginelan opponent. As he was leaving the island, Cleomenes, having discovered the name, offers the following response: "Holl vuv lCatax;aAlCou, 6>lCPl, ta. lCpEa, Clc; O'JVOlOJlEVOe; JlEy:Acp lCalCi> (6.50).) 83 On the subject of ties between elements of language and reality, il is wOlth noting that Euripides draws several contrasts between them (ofien using the tenninology ovo.ta vs. EpyOV), which might be considered in sorne very Iimited sense a precursor of Plato's exploration of ways in which the connection between fragments of language and reality can be problematic or discrepant. For instance, in The Trojan Women Hecuba cornments that she functions as Astyanax's healer, OVOJl' Exouoa, ta.pya o' oi) (1233); in The Phoenician Women Eteocles remarks that Ei 7t(XOl tautov lCaAov EqU aoqv a' a.ta, OUlC ~v av aJlqlAElCtOe; av9pmole; Eple;' VUV o' oe' OJlOlOV OUOEV Ot' lO"OV PPOto"iA.OC; 't 't' c1ta 'tv 'tE vouv 't 't' olllla't'd (O.T. 370-1).94Although these writers do not enunciate views on links between phonemes and elements of reality, several of the aforementioned passages indicate sorne tendency to correlate certain consonants with external phenomena based on connections between the way in which the sounds in question are vocalized and important features of, or factors involving, those entities under discussion. As mentioned, there are many passages in which guttural consonants are associated with expressions of hostility and references to various evils; if their vocalization is considered to involve a certain harshness, or other disagreeable features, then Itere too a similar connection may perhaps be envisioned. In addition, it is worth noting another case treated by Pindar, in which an individual originally narned Aristoteles is said to have later received the name B't'to~ (i.e., "Stammerer") based on his suffering from that particular speech impediment. In Pythian 5 Pindar says that the sound of this individual's voice, with Apollo's aid, put to flight a pack of lions, which were frightened by his exotic accent. In Pythian 4 Pindar mentions that Battus went to Delphi to ask the god to remedy his starnmer (ou0"8pou q>rova~ ... 1tOtv.) (63-4); it is interesting that certain of the term ~a't'toc;' constituents, namely ~ and 't, might be considered of special relevan ce to the malady whose sufferer it designates. A sensitivity to sound, and henee repetition of various phonemes, occurs on a wide range of levels in the literary tradition from Homer through Euripides. In addition to cases in which a range of individual letters appear with great frequency-for reasons ranging from the ostensibly aesthetie to the substantive-authors many times utilize the technique of repetition with regard to clusters of two or more letters. These groupings oceur in a minimum of two words, and involve no apparent interest in etyrnology: see for instanee 94 These Iines were drawn to rny attention in this connection by Sir Kenneth Dover, who suggests that one could think of the letrer 't in this context as an expression of harnmer blows.52Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 57. aAAE~JlEV aAA~A01.(nV (JI. 3.9) and .! ooC; liOE'

'tEAEcr9at (663-4). 96 On the frrst two occasions Helen herself is speaking. 97 Here the EA. combination is also present in Menelaus' name. 53Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 58. axaptv (I.T. 566); ~'ltTlP ~'ltrop (Soph. El. 1154);98 1tVOC; aU1tVOC; (Phi!. 847); and OEcrJ.lOV aOEcrJ.lOV (Eur. Supp. 32). There are also several cases of complete repetition: 1t~ 1t~ 1t~ (Aj.867-8; see also 912); ail..tvov ail..lVov (Aj. 627); El..aXEV EAaXEv (He/.214); E ZE, ta.n; once lived with Peleus apart from men; for this reason the Thessalians caIl it E>etOEwv (Eur. Andr. 17-20). Herodotus too notes several cases of the relevant type. He mentions the 1tt~ called Archandropolis whieh seems to him to be named after Arehander son of Phthius (Hist. 2.98). In addition, Herodotus indicates his puzzlement as to why three different women'snames were given to what is actuaIly a single land mass. He notes the prevailing Greek assumption that Libya was named after a native wornan, and that Asia was named after 107 Tr. by de Slincourt 108 For additional cases falling in this subcategory see Eur. Ion 1575-8, 1590-4 (on balance it seems preferable to place these two passages here rather than in thal subcategory in which parcels of land are in the recipient position); Rer. Hist. 1.7 (cf. 7.74),1.94, 1.171, l.l73 (cf. 7.92), 4.149, 5.68, 7.1 I (a people and their land named after the conqueror Pelops), 7.62, 7.90-1, and 8.44 (in this instance Cecrops, not Ion, is in the primary role). There is a noteworthy parallel between those Iiterary-tradition cases involving individual s giving tbeir names to groups and Plato's own handling of eponymy in tbe Phaedo: the former involve a single individual giving his name to a group of individuals, while in Plato's case it is preeisely a single entity, namely a Form, that gives its name to a group of individual sensibles. That being said, the differences are numerous and striking. Central among them is the faet that in Plato's framework tbe two parties are native to fundamentally different planes of reality; whereas, in tbe Iiterary tradition the individual s and groups in question are viewed as belonging to, or improperly aligned with, the empirical world. 109 For "1A.l0~ as a synonym for Troy see, e.g., /l. 1.71,15.66,15.71, and 20.216. 60Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 65. Prometheus' wife; as for Europe, he says that one cannot detennine where it got its name unless one cIaims that it comes from the Tyrian woman Europa (4.45) (with regard to the threefold division ofthe earth into Europe, Asia, and Libya, cf. Pindar, Pyth. 9.5-8).1 10 As concems cases in wbich a deity is in the primary role, Bacchus is said to have given his name to a parcel of land ('ta.cro' E1tClVUJ.l.OV yat;) (Soph. O.T. 209-11). Moreover,on several occasions the goddess Athena is associated with Athens as primary name-bearer tonominatum. Regarding tbis connection between Athena and Athens, Burkert notes that "whether the goddess is named after the dty or the city after the goddess is an ancient dispute. Since -ene is a typical place-name suffix ... the goddess most probably takes her name from the city."Ill Such lioguistic arguments notwithstanding, writers in the literary tradition claim on several occasions that Athena gave her name to the city of Athens. For instance, Oedipus addresses Athens as the "city named for great Athena, Honored aboye all cities in the world!" (Soph.O.e.107-8).112 EIsewhere Hermes refers to "the famousGreek city named for Pallas of the golden spear" (Eur. Ion 8-9, cf. 29-30), and later in the same play Euripides has Athena herself state that the city of Athens got its name from her (1555-6). In such cases one canDot help but concIude that cIear implications for-and expectations of-greatness follow from the assignment. 113 Fourth, natural inanimate entities may be named afier individuals. In this connection, Pindar remarks that HeracIes "called [the hill at Olympia] Kronos' Hill, for in former times .. .it had lain beneath deep drifts of snow, without a name (vc>VUJ.l.Ot;)" (01. 10.49-51, cf. 01. 5.19). Moreover, drawing 00 lo's name, Aeschylus foreteIls that the inlet ofthe sea which lo is driven to will be caBed '1 VlOt; ... 'tlt; crlt; 7tOpEUt; J.l.vlJ.l.U 'tOlt; 7ta.cr1.V ~pO'tOlt;(Prom. 840-1).114 Fifih and lastIy, individual s may give their names to certainhuman constructions, as when Pindar reports that Apollo named a temple of his the "Ismenion" after the son given him by the nymph Melea (Pyth. 11.5-6), and Herodotus 110 For additional relevant cases see Pind. 01.7.73-6; Her. Hist. 4.148, 6.47, and 7.178. Moreover, Pindar indicates on severa! occasions that parcels of land have their eponyrnous nyrnphs. For example, he treats Aigina both as a nyrnph (01. 9) and mother of Aiakos (Nem. 7-8, Isth. 8), and as the island to which she gave her name (Nem. 3-5). In Pythian 8, the poet depicts her explicitly as the nyrnph who gave the island its name: he addresses her as Atytva !pAa ~fm:p (98) and asks her to help Aigina gain its freedorn. Theba and Thebes receive the same treatrnent: Theba is rnentioned as Aigina's sister (lsth. 8), and Thebes as lhe Boiotian city (lsth. 4). Pindar also depicts Theba explicitly as the nyrnph who gave Thebes its name (lsth. 3 and 7). In Pythian 9, Pindar depicts !he nyrnph Kyrana as the one who gave !he city of Cyrene its name; in !hat same ode, Libya is treated both as a parcel of land and as that land's eponyrnous nyrnph (in Pythian 4 she is depicted as Epaphus' daughter). With regard to Rhodes see 01. 7. 111 Greek Religion, 139. 112 Tr. by Fitzgerald. 113 Of course, strictly speaking these can only be attached to the people of the city thus named. 114 Also falling in this subcategory are Aesch. Eum. 689-90 and Prom. 299-300.61Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 66. comments that "the Delphians call this gold and silver which Gyges sent the Gygean Treasure, after the donor's name (E1tl tOl> ava.8vto~ E1tcovuJ.!iT)v) (Hist. 1.14) (with regard to human constructions see also Eur. Hipp. 32-3).1 15 As concems human practices, a feast is named for Agamemnon (Soph. El. 282-5), and a banquet for Thyestes (tCx E1tCVUJ.!a. Oel1tva. 9uO"tou) (Eur. Oro 1008). The second category of assignments cornprises those instances in which natural inanimate entities give their vJ.!a.ta. to various types of entity. These might be individual s, as when Horner notes that Telarnonian Ajax killed l:tJ.!oEO"toV, ov 1tOtE J.!1tT)P "IoT)9EV KattOl>O"a. 1ta.p' OX9UO"lV l:tJ.!EVtoC; yEva.t', E1tE pa. tOKE>O"tV aJ.!' E0"7tEtO J.!l1Aa. iaa8a.t TOVEKeX. J.!lV KAEOV l:tJ.!oEO"tov (ll. 4.474-7). On other occasions, the recipient entities are peoples. In this connection, Herodotus reports that the people living around the rnountain called "Atlas" were narned "Atlantes" after it (Hist. 4.184). Sirnilarly, following their migration to Asia the Thracians becarne known asBithynians, but say they were previously caBed Stryrnonians after the river Stryrnon on which they lived(oiKovn:~E1tl l:tPUJ.!V1) (7.75). In addition, there are several cases inwhich natural inanimate entities function both as primary and recipient ovoJ.!a-bearers. To give just one example, Herodotus mentions the river Aegae on the Crathis, which is never dry ovoll EO"nN~O"a.tov.tOUe; 'to (Her. Hist. 7.40). In addition, the recipient entities might be natural inanimate entities; in this connection, the115 Herodotus mentions Ismenian Apollo at 5.60. 116 For other relevant passages see Hist. 4.52. 7.58, and 9.51.62Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 67. town Therma on the gulf is said to have given that gulf its name (Hist. 7.121),117 Moreover, individuals might receive their appelIations from places associated with noteworthy events, as when a man named Archias reports to Herodotus that 't> 1t

J.LEvO~ 1tAilv ~tC;, Ei to J.L,'tav a1tO 'PPov'tooC; iix90~ xpil ~otAeiv E'tT'tJ.LCO~ ... Z~va o n~ 1tpO'PPvco~ E1ttVlCUX 1CA.~roV 'tE~E'tat'Ppevrov 'to 1ta.v. (Ag. 160-75)121Euripides' tone suggests that he is somewhat less awestruck: oO''tt~ 1tOt' d O', ouO't1taO'to~Elovat, ZE~, Et't' av'J'K1l

,.U1t o' e:yro). While the first several words hearken back to the aforementioned Agamemnon passage, what follows suggests a marked difference in attitude andorientation. It is perhaps not coincidental that a Euripidean character, having observed Hecuba' s plight and condition, juxtaposes Zec; and 'tXTJ and muses about the possible preeminence of the latter: J, ZEU, 'ti A,~ro; 7tn:p o' avSpro7tOUC; piv ~ o~av&Uc.oc; 't1VOE KE1CTT10Sat J.L'tTJv ['I'EUOl1, OOKouv'tac; OatJ,lvrov dvat yvoe;], 'tXTv OE 7tv'ta 'tav~po'tOtC;E7ttOK01tEtV; (Hec. 488-91).1 22 (In this connection see alsoCycl. 599-607 and Ion 1512-15.)1232. Writers sometimes focus explicitly on descriptive content in their use of divine epithets that are relevant in particular contexts. EIsewhere they draw attention to the multiple appellations of particular deities; they may also use forms of identification or address-some more elaborate than others-which are tied to divine aid either hoped for received. 124 Examples inelude: a) Aeschylus. Eum. 90-1, Apollo to Hermes: 'EpJ.Ll1, q>A,aOOE, lCp'ta o' rov E7trovUJ,lOC; 7toJ,l7ta'ioe; tO'St. b) Aeschylus. Th. 8-9, Eteoeles: From this disaster ZEUe;aA,E~TJnptoC;E7trovUJ.LOC;yVOt'to KaoJ.Leirov 7tA.et. c) Herodotus. Hist. 7.192: Having discovered that the Persian ships had been destroyed, the Greeks offered prayers and libations nOO'EtOroVt oro'tl1 pl; nOOElorovoe; oro-riipoc; E7trovuJ,liTJv a7tO 'to'tou En lCal Ee; 'tOEvOJ,li~ov'tEc;.d) Herodotus. Hist. 1.44: Croesus invokes Zeus under several epithets on a single occasion, because all appear to him to be relevant to the goals he has on that particular occasion. In his profound grief at his son' s death, Croesus ElCA.EE J.lev ,1 ia. lCa9pO'lov, J.lap'tu pJ,lEVOC; 'tel 1mo 'tOl> ~ei vou 7tE7tOvSroe; erTJ, h:A.EE OE E7ttO'nv 'tE Ka!. halp~lOv, 'tov a'tov 'to>'tov ovoJ.l~rov SEV, 'tOV J,lEV E7tcr'tlov KaA.rov, Oln on OiK01.O'l imOOE~J.lEVOe; 'tov ~E'iVOV 7talOOe; EA.v8ave ~O'Krov, 'tov122o Zeus, what can 1 say? That you look on roan and care? Or do we, holding that the gods exist, deceive ourselves with unsubstantial dreams and Hes, while random careless change and change alone control the world? (tr. by Arrowsmith) Line 490 is bracketed in the ocr (del. by Nauck). 123 Cf. Burkert, who mentions these last three passages in sorne remarks about TXll's "rise to farne" (Greek Religion, 186, with fn. 29). 124 Or perhaps received ami hoped for (in future), as in the third case cited below. 66Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 71. OE btalp~tOV, C>~ 1tOA,E/llc.l'ta'tov. 125crU/l1tI.Ll'a~(j)'A,aKaa'tovEPlKOle) Pindar. Isth. 5.1: The mother of the sun, Theia, is addressed as 1tOA,Uc.lvU/lE. f) Aeschylus. Prom. 209-10: Prometheus says of his mother that she is called bothThemis and Earth, 1tOMrov vo/l'tcov/lOp~ /la.g) Sophocles. Ant. 1115-16: The choros addresses Dionysus as 1tOA, Uc.lVU/lE, Kao/lEa~ v/la~a:yaA/lCl ... h) Euripides. Hipp. 1-2, where Aphrodite describes herself as follows: ~po'tOtO'lnOA,A,~/lEV EVKOUK aVc.lvu/lo~ geeX KKAT/lal K1tpt~ opavou 't' ecrco.i) Euripides. Bacc. 274-6: Teiresias says that mankind possesses two supreme blessings, one of which is LlTl/ll'tTlP ge-rf o' Ecr'tv, OVO/lCl o' 1t'tEPOV ~OA,n KAEt. j) For elaborate forms of address, see the priest Chryses' prayer 'A1tAACOVtavaK'tt, 'tov ~'KO/lO~ 'tKE Ar'tc.l: KA,u9 /lEU, apyup'tu~', OC; XpcrTv cl/lq>t~~TlKa~ KAA.av 'tE sa9rv TEVtOot 'tE tt avcrcrEt~, L/ltv9EU, El 1to't 'tOt XClpEV't' E1tl. VTlOV epE'I'a, 11 d ol 1tOt 'tOl Kata 1tova /lTlp' eKTJa tCl pCOV 1]0' a i yrov, tOE /lOl 1CPlTlVOV EAOCOp' tEcrEtaV Llavaol. E/la oKpua crOtO'l ~A.ecrcrlv. (11. 1.35-42)126 Regarding Apollo see also Rhesus 224-32. AIso noteworthy for the degree of its elaboration is the way in which Hippolytus and his attendant huntsmen address Artemis in Euripides' Hippolytus (62-71); see also Hippolytus' prayer in what follows (73ff.). As previously discussed, there are numerous cases in which the descriptive content of the gods' names themselves is highlighted, e.g., vacious writers connect Zeus' name with Ot. andlor s1v. Epithets depicting vacious functions of deities are also plentiful. In these instances, rather than the god's name itself being the linguistic entity on which authrs125126In the violence of his grief Croesus prayed to Zeus, calling on him as God of Purification to witness what he had suffered at the hands of his guest; he invoked him again under his titIe of Protector of the Hearth, because he had unwittingly entertained his son' s murderer in his own house; and yet again as God of Friendship, because the man he had sent to guard his son had tumed out to be his bitterest enemy. (tr. by de Slincourt) Hear me, lord of the sil ver bow who set your power about Chryse and Killa the sacrosanct, who are lord in strength over Tenedos, Smintheus, if ever it pleased your heart that 1 built your temple, if ever it pleased you that I bumed a11 the rich thigh pieces of bulls, of goats, then bring to pass this wish I pray for: let your arrows make the Danaans pay for my tears shed. (tr. by Laltimore)67Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72. concentrate, attention is turned to the epithet as what does the describing, i.e., as what designates the capacity in which the god is being invoked in a particular context. This vast multiplication of di vine appellations lies in sharp contrast to the situation with regard to human names, where a single individual OVOIla. is both expected and generally adequate. 127 In contrast, there are so many specific duties-for instance-which humans assign to various deities that it was simply not felt to be sufficient to rest content with a single name. 128 One important factor motivating the process must surely have been the profoundly asymmetrical relation between gods and human beings, i.e., the felt dependence of the latter on the former; this, in tum, generated the need for a diversity of names to correlate with the manifold and diverse functions which hurnans hoped andJor expected the gods to fulfill. In fact, with regard to di vine narnes, it was not the case that one arrived at a fixed, closed systern. Burkert conveys the fluidity and dynamism of the process with great vividness: These detached figures are linked ... to specific dornains and functions in which their influence can be obtained and experienced. This link is guaranteed in two ways, by the epithets and by the personified abstractions in their retinue. Hyrnnic poetry, doubtless following aneient tradition, loves to heap divine epithets one upon another .. .in the eult it is the task of the officiant who speaks the prayer to encircle the god as it were with epithets and to discover the just and fitting name. In an established cult there will always be a fixed, well proven Dame, but this does not inhibit the search for further epithets. The epithets in tum are complexo Sorne are uninteIligible and for that very reason have an aura of rnystery; others result frorn the fusion of gods who at first were independent.. .. Many are taken from sanetuaries ... or from ritual, as if the god himself were performing the ritual aet. ... Many are formed spontaneously to denote the dornain in whieh divine intervention is hoped for; in this way each god is set about with a host of epithets which draw a eornplex pieture of his activity. (Greek Religion, 184)127 This being said, it is true, of course, that writers mention certain cases involving name changes, or the acquisition of additional appellations, i.e., niclmames. Alcinous puts Ihe point about Ihe universal presence of human proper names aplly when he asks Odysseus lo identify himself by name: ett' OVOJl' Ottl crE KElfh KAEOV JlTt'tTlP tE 1taTlp tE, aAAol 9' o'i Kata acrtu Kal. o'i 1tEplvalEtoucrtv. o .LEv yp tls 1tJl1tav avrovuJlS Ecrt' av9pro1tCllv (Od. 8.550-2). Elsewhere Herodotus distinguishes a people called Ihe "Ataran tes" as Ihe only ones whose constituent members lack individual appeIlations: they avrovuI.1Oi Eicrt JlOUVOl av9pro7wlv twv iJlElS lOJlEV' vI.1(X.~EV 6)0' ~ 'to 7tnv 't11't/l~, /l~ 'tle; ovnv' ox pOO/lEV 7tpovoatO'l 'tOl> 7tE7tpc.o/lvou yAroO'O'av EV 'tXal v/lc.ov, 'tav oopya/lJ3pov cl/l 1tE1tProIlvOU belong only to a god or a daimon."134 1 agree with this final statement about gods' and daimons' capacities; however, what Fraenkel does not consider here is whether Zeus himself-as Helen's male parent-might be envisioned as namer. 1 would c1aim that Aeschylus' phrasing lends at least as much support to the c1aim that the Chorus is here a1luding to his possible role as namer. Any concerns about ambiguity notwithstanding, what is ultimately of interest is the reasoning process which one can trace in the Chorus' observations. Success (tXTl) is said to be achieved in this instance because Helen's effect on mortals has turned out to be precisely that predicted by her name. In fact, in the passage's reflective sequence-versus that of presentation-it is Ihis fact that leads to the Chorus' initial musing about the name's source. This becomes evident in what follows the long opening question, where the Chorus indicates its underlying motivation in making the inquiry at ail; the phrase 1tet 1tPE1tVtroo~' E1td o' EUKvat, 'AAeXO''l:'OpOC; /lEV 1tPOOtOV, etta oE 8vou !>vou tE 8avtou S' oO'a tE rii tp(j>Et KaK. o 'Yp 1tOt' axoo Zilv 'Y' E1C(j>UO'a. 0" E"{ro, 1tOAAolO't Kilpa ~apppOl.C; "EUllO'. tE. (766-71)70In the aforementioned Agamenuzon passage Aeschylus does not raise the issue of patemity, being content to use etymology to get his point across; in faet, far from questioning Helen' s status as Zeus' daughter, one might view the Chorus as a1luding to his possible role as namer of his progeny.7 1 Here, in contrast, Andromache insists that Zeus was not actually Helen's father, and her remarks at least imply that Helen's dreadful conduet makes her69 In one noteworthy passage Apollo makes some dogmatic cornments about the basis of an individual's entitlement to be called "parent" (tolCe;) (Eum. 657-61). In his view it is categorically and only Ihe man who deserves to be called a child's "parent"; for Ihe same sentiment expressed wilhout raising the malter of appropriateness see Orestes 551-6 (cf. the Loeb, which refers to Oro 552 in a note on the Aeschylus passage). 70 O flowering of Ihe house ofTyndareus! Not his, not Zeus' daughter, never that, but child of many falhers 1 say; Ihe daughter of Vindictiveness, of Hate, of Blood, Dealh; o" a11 wickedness that swarms on earth. I cry it a1oud: Zeus never was your father, but you were bom a pestilence to a11 Greeks and the world beside. (tr. by Lattimore, slightly modified)71 It seems c1ear that in Ihe Agamemnon passage Aeschylus' raising the issue of Helen' s namer is not equivalent to his raising the malter of patemity. If one assumes that an individual's progenitor serves as namer one would conclude that Zeus' status as Helen's father is in question just because the Choros expresses uncertainty about the name's source; however, this interpretation would go far beyond anylhing one finds in the tex!.103Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 108. utterIy unworthy of the appellation "daughter of ZeuS."72 Nevertheless, while the two tragedians adopt different approaches, they accord in their emphasis on the destructive character of her activity, and in their inclination to make the point, at least in sorne sense, by appeal to linguistic considerations.13 Finally, writers raise the issue of a particular individual's right to be called "son." In a striking case of this type Reracles informs Hyllus that he has a duty to ease his father' s plight by building a pyre and buming the body, lest he forfeit his entitlernent to be called Hecacles' "son": you OE IlTlOEV eiaoro OellCpU, aAA' aatvalCtO~ lCaOellCpUtO~, El1tEp d touO' avop~, ep;ov' El OE Il~, Jl,Evro a' eyro !Cal VpeEV rov ap(llo~ EiaaEl /3ap~. YA. OlllOt, 1teln:p, t Ei1ta~; Orel Il' Elpyaaat. Hp. 1tOla opaat' eativ El OE Il~, 1tatpo~ aAAou YEVOU tOU IlTlO' ellO~ !CATleft~ En. (Soph. Tr. 1199-1205)74 In a somewhat analogous case Haemon, having tried unsuccessfully to dissuade Creon from killing Antigone, states that El 1l1l1tatllP ~ae', Et1tOV av a' OUlC El) "(UvatlCEC;, E~ tO aUJ,l1tOVEtV J,lEt~o' 1t' aJJ..ou lC01>lC EJ,lo 1tElCatOv (1367-9).Funetional entena eombined with theproposed etymology of Antigone' s name would seem to grant Sophocles' treatment the status of ajudgment of deserto More generally, this case, along with those involving Ion's adopted "mother" and "father," indicates a degree of flexibility present in assessments involving funetional tenns: sueh eases provide the clearest evidence that the duties one perfonns, rather than biological or gender considerations, constitute the deeisive criterion.76 As previously noted, av't has the sen se of "like" or "as good as" a1ready in Homer; for references see rny treatment of Antigone's name in ch. 1.105Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 110. Appendix B Terminology Writers Use to Assess Appropriateness 1 provide here sorne analysis of the terminology which writers use to raise questions of appropriateness in the three categories treated in what precedes. When one examines the relevant passages, what is most striking is the marked diversity of means which writers employ; these range from single words to judgments which an author offers through multiple features of his remarks.CATEGORY-ONE CASES In several instances, adverbs introduce writers' assessments; elsewhere verbs or participles combined with adverbs or adverbial accusatives convey their judgments. In still other cases an author uses a verb alone to malee the point, or, more frequentIy, combinations of adjective and adverb. One adverb that writers employ is ap9ro~. In Euripides' TrojanWomen one finds a passage in which ap9roe; plays a central role: 'tel Jlropa YelP 1tv't' EcntV 'Apoavl1~ apXEl 8Eae; (989-90). Here it is this single word that clearly introduces the dimension of appropriateness. In another instance ap9roc; forms the core of a judgment of desert, when an Aeschylian chorus, referring to Eteocles and Polyneices, says that Ol 011't' ap9rne; K(n' E1toovuJlav ... Kal 1tOA.UVEt1Cel~ roA,oV't' mE~E'i otavoat (Th. 829-31). Here the poet extends the descriptive content of Polyneices' name to cover Eteocles as well;77 while in the absence of ap9roc; the evaluation may be implied, this adverb once again malees it explicit. 'AA,l1eroC; is another adverb used to express judgments of appropriateness, as for example in Aeschylus' Suppliants: "E1t(l(poC;, aA,119roc; pucroov E1tCVU/lOC; (315). In this instance the adverb alone makes what would otherwise be merely an etymology"E1taq>oc; ... puatoov E1tC.OvuJloc;-into a judgment of veridicality; aA,119roc; here modifies the adjective E1tC.OVUJlOC; which by itself here introduces only the basis on which this 77 As previously discussed, Sophocles does the same with regard to Antigone and Ismene, with the descriptive content of the former individual's name being applied to both.106Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 111. individual received bis name. This adverb is also of central relevance in another case: aA:r9&; o' ovo,"I.(X nOAuvE,1CTJ 1ta't~p e8E't erOl 8E.~ 1tpovo~ VEllCc.oV E1tCVUJlOv(Eur. Phoen. 636-7). Here aA1l8&; is one of two factors making what would otherwise remain on the plane of etymology into a judgment of deserto In fact, one can break the statement down into three distinct levels: the words OVOIl constitutes a judgment of appropriateness when Prometheus, giving lo a foretaste of things to come, says that f~ElC;O' 'Ypptcr"C~v 1tO'tllJlOV Ol> 'JIEuOrovullov (Prom. 717).84 WhiIe these two wordsneed not in principIe signal that a judgment of appropriateness is at issue, in this ease they do: the idea is that there is an extremeIy cIose fit between the river's appelIation and the eharaeter of its activity. In yet another instanee, 'JIEUOroVUJlOC; combines with the modifying adverb rcavolCroC; to raise the issue most emphatiealIy, when EteocIes asserts that neither heretofore nor at present has Dike taken Polyneices' side: ~ 0~"C' av 81 Moreover, he proceeds to dwell on the strife resulting from the course of action which the bearer has undertaken (580-3). Strictly speaking of course, tOVO!Ul is the direct object of the participle, while the remainder goes with lC!lAEi. 82 The fonner is Wyckoffs, the lalter thal provided by Liddell and Scott, A Greek-English Lexicon. 83 See Eum. 90-1 for Aeschylus' use of these same two words in an expressed wish with reference to Hennes. 84 More generally see 717-24.109Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 114. 1tavBlccoc; 'l'EuBc.vuJ.1oe; ~1CTl, ~UVOu()'(X pa{av yeXJ.1COV, KaA.OUCJtv au'tllv 8EoVllv' tex BEta ri!Q 'teX 't' ov'ta Ka. J.1A.A.ovta 1teXv't' in{CJ'ta'to. (He/. 10-14)87 Here the relative adverb of timeMe is used to fix the perlod when the original appellationwas used; then, the o in the phrase ene. o rnarks this name and perlod off from those that follow, while the temporal conjunction e1te.-as counterpart to &rE-highlights the point at which the shift occurred. Finally, in lines 13-14 Euripides provides the reason for the change (as the presence of yeXP indicates). One finds similar elues where nicknames are assigned, as for example in the case of "Alcyone"; here a combination of temporal marker ('tte) and grounds of shift-introduced by ovEK'-signal Cleopatra's acquisition ofthis additional ovoJ.1a (JI. 9.561-4).88 In future-directed cases involving divine appellations wrlters often employ the imperative mood of the verb to help indicate that the speaker is expressing a wish that the85 Specifically, the adverb is what lends the remark its emphatic quality. Note Eteocles' use of 1tavoilCCOATJJ.l.Ot J.l.1l'tEp, o 'tEKO-Ucr 1tEP (Ion 1324). Ion conveys hisassessment through this form of address combined with the participle that follows; this participle has concessive force, which is highlighted by the presence of 1tEp.9997 Moreover, it is possible that Sophocles hints at a judgment of non-desert via a juxtaposition of noun and adjeetive, when he has Eleetra speak of Clytaemestra as LlTI1P clLltrop (El. 1154); he here uses the noun LltT1P, but promptly retraets it by introducing a eorresponding privative adjective. Strictly speaking, of eourse, what results is far more than a mere retraction. 98 In the former instanee Orestes supplies the alternate appellation, aceompanied by the explanation that Llltp' aiOou.Lat A'yEtv. 99 Subsequently the priestess evinces her agreement with Ion's view that she can fittingly be called his mother on non-biological grounds: laov y

~ Ill't1lP analysis tied to the Derveni Papyrus specifically, or Orphic "scholars" more generally (32. While Baxter is quite right to recognize that the Iiterary tradition is important (though he concentrates only on Homer and Aeschylus), systematic study of the extant evidence justifies the c1airn that it is not merely one source among nurnerous others. (Invoking the literary tradition in a very different context, Seth Benardete asserts that the dialogue sheds Iight on Plato's understanding of Greek tragedy. He has in mind nOI their conunon use of etyrnology, but the general issue of tragedy's stance toward human Jife and possibility, as when, in discussing the etyrnologies of "A pollo" and "Pan," he c1aims that "A poli o is tragedy's own view of itself. Pan is Socrates' .... The goatishness ofPan is the laughably human about which tragedy sings its rnyths and lies. Pan is the root of the tragic Apollo"; in what follows he identifies the idea that "the individual loses nothing of himself in attaining significance" as "the essence of tragedy, [which] was shown to obtain per impossibile only in the realm of the arch-sophist Hades" ("Physics and Tragedy: On Plato's Craty/us," 137). The interpretation advanced in this artiele is highly speculative, and in any case does not speak to the issue of the tragedians and other poets as Platonic opponents in the Cralylus due to their own use of etyrnology specifically.) 75 This is especially so of the Phaedo. The Cratylus and Phaedo are related largely as negative to positive, with the Cratylus clearing the ground for Plato's use of eponymy, also prominent in the Iiterary tradition, to treat appropriateness in the Phaedo. Rosenstock rightly highlights the existence of a cornplementary160Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 165. are thus not completely wrong, but simply have not properly identified the fono of that response. Key here is the fact that both target the Iiterary tradition directly: When Plato considers the Iiterary tradition he is not concemed to refute the sophists; in faet, he has no deep or abiding interest in their procedure in this regard. Plato is not responding to the Iiterary tradition via the sophists, but to the poets themselves, hence only indirectly to the sophists insofar as they too used that material, albeit wrongly to bis way of thinking. 4. 391d4-392b2: Discussed frst are Homeric instances of different ovJlo:ta used by gods and humans with reference to the same entities; one can assume that those appellations the gods employ are naturally-correct. Examples: l. riverXanthus and Scamander; 2. bird-Chalcis and Cymindis; 3. mound before Troy-Batieia and Myrine.Socrates notes that speaking directly of gods'assignments is Iikely beyond their resources. 76 The issue of adivine component involved in matters of naming frst arises here, as Plato makes the transition from sophists to poets in grounding his treatment of op9trc;. Plato would oot want to embrace this approach to correctness since a suffcient explanation of appropriateness in such cases is simply to invoke appellations' divine source. To claim that particular names are superior simply because they are associated with these superior beings lea ves no room for a sustained investigation of the sort that Plato is interested in pursuing. Theserelationship between the two dialogues; however, he wrongly views the Cratylus as "the perfect complement to the Phaedo"--rather than vice versa-privileging what he views as the Cratylus' challenge to "the Phaedo's reliance upon simple logo; as a medium oftruth" ("Fathers and Sons," 410). 76 There are severa! Iliad passages in which Homer says that men give an entity one name, while the gods employ another: I. l/. 1.403-4: Homer refers to that son of Poseidon, having a hundred hands, "whom the gods can Briareus, but all men Aegaeon"; yap an: ~TlV o{) 1tatpO~ aJ,LEvrov. 2. l/. 2.811-14: There is a mound before Troy, which men call Batieia, "but the immortals can it the barrow (crllla) of Myrine, Iight of step." (In this and Ihe preceding passage I employ Murray's translations.) 3. JI. 14.290-1: There is a mountain bil'd, which the gods call Chalcis, and men Cyrnindis. 4./1. 20.73-4: There is a river, which the gods can Xanthus, and men can Scamander. In his remarks here Plato mentions a11 except the frst. In contrast, it is worth noting Hesiod's focus on parallels between gods and men with regard lo naming: l. Th. 195-7: Gods and men call the maiden Aphrodite. 2. Th. 270-2: There is no distinction belween what men and gods can the progeny of Ceto and Phorcys, Le., the Graiae. Moreover, Hesiod emphasizes such parallels in connection with other matlers: I. Th. 220: The Fates pursue the transgressions (1tapa1~acria~) of both men and gods. 2. Th. 406-7: Lelo is characterized as ~1tlOV lo men and the irnmortal gods. 3. Th. 766: Death is characterized as hateful even 10 Ihe irnmortal gods.161Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 166. instan ces serve to introduce Homer-and by extension the literary tradition in general-as having treated correctness; however, it is the ensuing treatment of the names given to Hector's son, which involves a focus on semantic constitution, that typifies the brand of Op9tlle; al issue for Plato in the Cratylus. The matter of names' possible divine source will be a recurrent theme in the dialogue. 77 5.392b3-393b6:Etyrnology and the issue of appropriateness:Scamandrius/Astyanax and Hector. a. 392b3-e8: The fitness of Hector' s son' s name is explored. Socrates notes that Homer gave him two oV/lata: Astyanax and Scamandrius. The more insightful individuals, here identified as the men, are said to have utilized the former appellation; the less insightful, here the women, are supposed to have employed the latter:06lCOl>V otcr9aon"O/ll1POe; tO nalBov tO tOl>i>1to toov Tpcrov v here closely to the particular notions of life and death in the way required by tbis interpretation. 117116 One might say that there is a significant c1ash between those expectations generated by the OVOIUX and the kind of impact which its bearer has on reality. While use of the bow results in death, such ties as tbereare between the bow itseIf and Jife are indirect, i.e., they exist insofar as Jife and death are themseIves coincident opposites. 1be fragment does not suggest a way in which the bow ilself couId represent Jife or a coincidence of opposites invoIving Jife and death; with Kahn (202), I do not find it IikeIy that Heraclitus had in mind the obvious fact that hunters' activity of killing secures the continuation of their own lives. Cf. Fr. 67: aEO~ f~PTJ EepPVTJ, XE1ILCI)V apo~, 1C1..E~O~ Eip~vTJ, lCpO~ 1..1Il~ ('tQvav'ta a7tav'ta' o{'to~ vou~), Ql..1..olOu'tal oE OICCOO'7tEp (7tUp), 7t'tav c:ru~.Ll'yll auro.Lat:Jlv, Vo~~E'tal lCae' fOovTv K:cJ'tou, with Kirk's coroment, which seems to me right, that "God .. .is said to be the coromon connecting eIement in alI extremes, just as tire is the common eIement of different vapours (because these were conceived as a compound of tire with different kinds of incense). Change from one to another brings about a total change of name, which is misleading, because only a superficial component has aItered and the most important constituent remains" (The Presocratic Philosophers, 191, italics mine; this represents a modification of his earlier, somewhat more positive stance (The Cosmic Fragments, 118. The proliferation of names would seem to reinforce the common and mistaken human tendency to greatly magnify the importance of perceptible changes and contrasts. For the contrary view see Marcovich who, in ,his comments on ~O~-f31~, claims that DK 67 does not contradict the view that for Heraclitus an ovo~a "reveaIs a greal deal of the true CPt:Jl~ of its object" (192). In his remarks on Fr. 67 itself, he emphasizes a supposedly close tit between the content of this fragment and the view that "names reflect a certain pan of the very essence of things" (416; italics mine in both cases); even this claim, which is notably weaker than that on p. 192, is unsubstantiated by the text of the fragment. For a clash between reputation and reality see Fr. 28: OOlCov'ta yap oOlClllro'ta'to~ YlVromm, epUl..crcrEl; despite a high reputation, which would ostensibly correlate with superiority to ordinary human beings in terms of knowledge, for Heraclitus such individuals emphaticaIly do not have this elevated status. While there may well be inlenlional word-play here, there is no suggestion that this play takes the form of a proffered etymology; also, one would be hard pressed to find a clear explanatory relation going in either direction. 117 Contra Kahn (The Arr and Thoughl of Heraclilus, 270), who does see a tie to ~lv here, and with Kirk (The Cosmic Fragments, 392) and Marcovich (445), both of whom tind the link unlikely. (In The Presocratic Philosophers, ch. 6, Kirk says nothing about a possible etymology or other natural link between name and referent.) Marcovich takes the phrases di1CT; ovolla (Fr. 23) and ZllV~ 5vo~a (Fr. 32) to "imply that the name corresponds to the very essence of its holder" (193) (regarding the former, Marcovich notes (230) that in the former case "ovolla seems to imply 'idea'"). One might view the latter178Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 183. The surviving fragments indicate a sensitivity to and use of language quite similar in nurnerous ways to that displayed in poetic creations from Horner through Euripides. While Heraclitus certainIy evinces interest in word-play in general, what one does not find is a rnarked interest in etyrnology specifically, as treated in theCratylus and by Plato's non-philosophical predeeessors. To the extent that HeracIitus may be viewed as offering etymologies and/or granting an explanatory role to certain parallels in descriptive content (as with J.l.pO~-I.lO'ipa in DK 25, perhaps ~uv vcp-~uvcp in Fr. 114, along with the substantially negative import and indirect ties in~t~-~o~,Fr. 48), he is foIlowing a procedure initiated anddeveloped far more by his literary predecessors, contemporaries, and successors;1I8 in fact, in those two ofthe three aforementioned eases which make sorne use (however lirnited) of etymology, the same parallels are found in poets themselves (J.l.po~-J.l.o'ipa in Aeschylus, and ~t~-po~ in SophocIes). Even if one were to lend more positive import to a given example, notably, the ~t.~-po~ parallel, the faet remains that insofar as HeracIitus does raise the issue of etymology he is drawing on a non-philosophical, literary tradition to which he-like Plato-is heavily indebted for the initial formulation of this and related problems. Writers in that tradition, unlike HeracIitus, make extensive use of etymology, grant their analyses clear explanatory value, and use etymology on numerous occasions to treat explicitIy the question of names' appropriateness. Finally, mitigating stronglyas indicating that Heraclitus' Ev 'to (Jo

'tO A,f:yEl~; 1:0. A'YEl 1tOU 'HpKA,El'tO~ on "1tv'ta xropet Kal ouah Jlvel," Kal 1to'taJlol> poft a1tEllcl;rov 'tCx ov'ta A,yel ro~ "Ol~ E~ 'tov au'tov 1to'taJl.ov OUK av EJl.Patll SEO>, roe; tt OE1VOV 111lVOVtOe; (404e1-2). In what follows he offers four derivations correlating with the god's four OuVIiE1e;. 127 Hesiod speaks of the goddess whom gods and men call 'A 9pcp9r (Th. 195-8). Euripides offers an etymology of the name 'AK a~lOv 'HcrlOcp aVtlI..yE1V, aAI..a crUyxropEtV en Ola ri1v tO} acppou yvECJtV '''AcppoSfll'' E1CI..i9r (ElC added by Hermann). 128 There is a notable Odyssey passage in which Athena's name may be associated with vrlla and voue; (voe;). Odysseus, upon his anival on Ithaca, asks his divine protector Athena-there disguised as a mortal-whether he has in fact reached his island home. The goddess responds by saying that aid tOl tOlO)tOV ivl crtf9Ecrcrl v1)ua' ti) crE lCal o> ovallal ItpOA17tElV Ocrfllvov EVta, OlJVElC' E7tTt1e; Emnano183Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 188. lCai arxvoos lCal ExcpPOOV (13.330-2). This passage assumes heightened inlereSI based on Plalo's own similar treatmenl of the name in the Cratylus, combined with the facI thal a reference 10 the interprelers of Horner begins his analysis: wlCaal oit !Cal o naAalOl n,v 'A9rv&v VO!l~ElV OOaltEp o vuv ltEpi "J.1rpov OElVOl. lCal. yap tOtOOV o ltoiJ..oi ~rYOJ!EVOl tOV ltOlrtV cpaal -riv 'A9rv&v autov vouv tE IClXi olvOlav ltEltOlr!Cval, !Cal. ta OV!llXta ltOlroV EOllCE tOlOUtV tl ltEPl. aurii~ olavoEla9lXl, Etl OE J.lEl~VOO~ 'Ayoov 9EOU vralV cOOltEP 'AYEl Otl ". 9eova" Eat1.V atTI, t!! a1..cplX ~EVl!Cro~ aVtl tou ~ta xpraJ.lEvo~ lCal. tO irota Kal. tO alYJ.la acpE1..c>v. raoo~ OE OUOE tatn, U' l~ ta 9Eta vooar~ lXutilc; olacpEpVtOO~ trov aUoov "9EOVrV" E!C1..Eaev. OUOEV OE c:iltXEl 1C1x n,v EV t!! ~9El vralv ooe; oaav n,v geov tatTIv '''H90vrv" J,LEV ~O1..ea91X1 ltpoaEutElv' ltaplXyaywv OE ii auto~ ~ tlVEC; ;)atEpOv Eltl tO lCU10V lc; ci>OvtO, ." A9rvav" ElC1..eaav (407 a8-c2). The men of old seem lo have viewed Athena just as do the contemporary interpreters of Homer. For many of these, interpreting the poet, say Ihat he represented Athena as nous and dianoia. The one consrructing names appears lO have believed sorne such thing about her. Moreover, the name-giver appears 10 have had a still more elevated conception of her as divine intelligence (Iheou noesis) hence the name he Iheonoa, in which alpha appears in Ihe foreign way instead of eta, and the iota and sigma are omitted. Maybe, however, on account of her superlative knowledge of divine malters (la lheia noouses) he called her Theonoe. And one would nol be far wrong in believing Ihat the constructor of names conceived of her as moral intelligence (en 10i ethei noesis), hence the name Elhonoe which either he himself or others afler him modified lO whal they believed a superior form, calling her Athena. AIso worth mentioning here are Hesiod's cornments on Athena's binh, specifically, his attribution of superlative intelligence both lO her mother Metis and lo Athena herself. In the Theogony, the poet remarks that Zeus first wed Melis, ltAelata 9EroV eiou'iav iOE 9vrtrov av6pc>ltoov. c:iU' OtE ol p' ~J.lE1..1..e 9EaV YAaUlCronlV 'A9ivrv t~Ea91X1, tt' EltElta 001..q> cppvac; E~a1tatfaa~ aJ.lU1..l01al AyolalV ilv EalCt9EtO vrOv ... EIC yap rii~ elJ.lapto ltEplcppova tlCVa YEVa9al, 1tpc>tTlV J.lEv ICOprV YAaulCc>molX T pltoyvElav 'laov Exouaav ltatpl. J.lvo~ lCal. ltlIPPOVa ~ouMv, autap EltElt' apa ltaloa 9EroV plXcrtAila Kal. c:ivoprov ~J.leJJ..ev t~Ea9al, il1tp~lOV ~tOP exovta. (887-98) and she was wisest among gods and mortal men. But when she was aboul 10 bring forth the goddess bright-eyed Athene, Zeus craftily deceived her with cunning words and put her in his own belly ... for very wise children were destined to be bom of her, first the maiden brighl-eyed Trilogeneia, equal to her father in strength and in wise understanding; bUI afterwards she was lo bear a son of overbearing spirit, kind of gods and men. (tr. by Evelyn-White) (For the binh of Athena from Zeus's head see, e.g., Th. 924 and Pindar, al. 7.35-7.) 129 Aesehylus offers an etymology of the name "Apr~, linking it with aplX: nllcp~ OE xprJ.ltoov !CalCoc; olXtTJta~ "AprC; c:ipav ltatp'lav n9de; 'Aae~ (Th. 944-6). In a possible, different attempt at etymology with regard to Ares' nante, Homer has Hephaestus bemoan the faet that Aphrodite seoms him due to his physieal deformity, cpl1..El o' a 'tor'Aov "Apra, ove;( !lEv !CaA~ 'tE lCal. aptlto~ (Od. 8.309-10). Rather than follow either of the aboye routes, Plato's analysis moves along different lines: OUICOUV, Ei !lEV ~O1..el, lCata. 'tO appEV tE lCal. lCata tO avopEiov "" APl1~" av Elr' ti o' a.o lCata 'to184Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 189. 408b7: Hermes (including cornment on appropriateness of Hermogenes' name); q. 408b8-d5: Pan (tragedy is mentioned).l30 C. 408d6-410e5: Plato analyzes a range of terms denoting natural bodies, elements, and cyeles. At issue here, generally speaking, are ov'tcx which Presocratic philosophers take tobe fundamental-whether as primary substances (apxcx) or in sorne other way. Plato here moves permanentIy away from proper narnes for the duration of the Cratylus; I3l this shift coincides to a large extent with the dawn of philosophy, insofar as the latter tradition rejects the notion of personal, divine beings as explanantes with regard to the formation and operation of the cosmos. Notable in this portion of the Cratylus are those v)J.cx'tcx designating elernents; such entities are granted vacious types of priority by Presocratics, and are central to explanations of what they take to be problernatic. 1321. 408d6-409a5: fAtoC; (natural body); 2. 409a6-c3: O'EA~vll (natural body; Anaxagoras rnentioned);133 3. 409c4-6: !lEC; (natural cyele); 4. 409c6-9: aO''tpcx (natural bodies); 5, 6. 409clO-41Obl: 1tUp, ooop (elements; foreign origin); 7. 41Obl-6: alp (element); 8. 41Ob6-8: cxi9lp (element); 9. 41Ob8-c3: yii/yatcx (elernent; Horner rnentioned); 10, 11. 41Oc3-e5: Vtcx'l)'tc;!e'toc; (natural cycle; this use of a dual etyrnology recaIls the earlier twofold analysis ofZeus' name). D. 411al-421c2: At issue here are the so-called KcxAa v!lcx'ta (41la2, a8-bl). This class neludes a wide range ofterms whose referents are of great importance fromoO'KAl1PV tE Kal aIlEt.O'tPOIjlOV, o~ "appatov" KaAE'ital, Kal tatn av 7tavtax 7tOAElllKi!> 9Ei!> 1tp1tO ''''Apl1'' KaA.:'icr9a (407dl-4). 130 Notably, Plato's analyses of the final two names center on language. Perhaps, in sorne way, the remarks on Pan help bring the focus back to nature and the natural world. 131 This is so with the exception ofHermogenes' name, which comes underdiscussion again in 429b-e. 132 To give several examples ofelements as apxai: Thales privileged water. While Anaximander's apX1, tO a7tElpOV, was not identical with any natural substance in the developed world, his primary cosmogonic opposites, the hOI and the coId, were none olher than "flame or fire and mist or air" (The Presocratic Philosophers, 120). Anaximenes gave pride of place to airo Heraclitus does not have an apxiJ in the standard Milesian sense; notably, he does no! believe thal the cosmos had a genesis (DK 30). However, Heraclitus views fire as the most fundamental material substance, indeed, as crucial to the cosmos' operation, and seems a! times to link c10sely or even identify tire with the AyoC;. Notwithstanding important differences between Heraclitus' tire and previous candidates for primary substance, it resembles its predecessors in a crucial way: namely, to the extent that it is a kind of "slufr' which is, on the one hand, unified (the worId order as 7>p aEi~(j)ov, DK 30), yet, on the other hand, distributed in those entities which constitute the cosmos. Empedocles, in turn, in his concem lO address the argumenls and objections of Parmenides, posits multipIe primary substances or "roots": earth, air, frre, and water. 133 In his etymology of this term Plato comments that !:EAavaiav o yE lCaAoucnv atTv 1toAAo. and proceeds lo identify Ihe original form which, once hammered togelher (cruYKEKPOt1lIlVOV), yieIds this appellation. The form !:EAavaia is found in Euripides (Phoen. 176).185Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 190. Plato's own point of view, in the areas of ethics, psychology, epistemology, and metaphysics. Key here is the fact that Plato ineludes tenns denoting entities which he takes to have the highest ontological status, and to enjoy natural, logical, and explanatory priority as well (ya9v, ~1ca.10V, and lCaA.v).l34 Rather than mentioning every term anaIyzed in this section of the dialogue, 1 select a range of thev~a'taat issue.Terms designating key elements of Plato'sphilosophical system are subject to the same etymological analysis as other VIlpoO'vll (411e-412a); E1tlO't1~l1 (412a and 437a); O'oq>a (412b); ya.9v (412c); ~llCalOO'vr (412c); ~tlCalov (412e-413d); v~pEa (413d414a); 'txvr (414b); pe't1 (415c-d); lWA.v (416c); EProC; (420a9-b4); A.1geta. (421bl-3); av and oUO'ta. (42Ib-c).135 Plato also treats the term 11oov1 (419b7-el), and subjects the termavo~aitself to analysis (421a7-bl), which is noteworthy giventhe fact that it is precisely v~a.'ta whose status as "naturally correet" is here in question. From within this framework Plato cannot privilege the aforementioned terms and their referents in the way he believes they deserve. Moreover, this structure provides no way for Plato to draw fitting distinctions between eertain contrasting terms-specifically, of course, their referents-mentioned in the Cratylus, contrasts which play important roles in his metaphysics, epistemology, ethics, and moral psychology. 1 have in mind here, for instanee, the dichotomy between arolla and 'l'ux1 (399dl-40Oc9) of which he makes so much elsewhere, as 134 The literary tradition had trealed justiee as a deity. Aesehylus observes !hat .1Ka received her name based on !he fact !hat she is Zeus' daughter (CIw. 949), and in faet once compres sed .1l~ Kpa yields .1lKa. Elsewhere Aeschylus refers to her as 'l.1l~ 7tate; xap9voe; .1ilCTl (Th. 662). Hesiod before him had spoken of the maiden as 'l 7tCxp9voe; .1ilCTl, Lil0e; EKYEyauia (W.D. 256) Given Ihe similarities between the formulations in Seven against Thebes and Works and Days, it is possible Ihal Aeschylus intentionally recalls Hesiod's treatment. Writers treated her as one whose aid mortal s might hope for and enlist. In contrast to !he literary tradition, Plalo views justice as a Form rather than a deity; as emerges most strikingly in !he Republic (Books 2-9), for him it is something at which one can and should aim. The differences in treatment are marked though Ihe two entities, as trealed by poets and Plato, have certain attributes in eornmon. (Among the Presocratics, for .1ilCT as what might be tenned "the personification of normality and therefore regularily" (Kirk, The Presocratic Philosophers, 202) see Heraclilus, Fr. 94: "HA,10e; ya.p ox. ')7tEp~TaEtal ~tpa' El oE ~T, 'Eplvee; ~1V .11crle; E7tKOUpOl E~E')piGOu(nv. In the proem to his hexarneler poem Parmenides too indulges in personification, here closely folIowing Hesiod; !he mention of "avenging Justice" (.1ilCT 7tOA.lt01VOe;) makes the paralleI especially vivid (for .1ilC1l as avenger see Hesiod, Works and Days 213-73; Aeschylus, Libation Bearers 948-51, Seven Against Thebes 662ff.; and Sophocles, Ajax 1389-92).) 135 A key measure of Ihe futility, even danger of this approach is found in the etymology conflating etval and ival ("10 be" and "to go"), which runs as follows: to oE "ov" Kal ~ "ouaia" ~ol..oyei t!> aA,11gei, 'to i&ta a7tOAa~v' iov ya.p GTlj.lalVE1, Kal. tO "OUK ov" au, roe; 'tlVEe; Kal. OVOll~oUG1V at, "OUK iv" (421b7-c2). Cf., e.g., !he etymologies of CIlPVl1Gle; (CIlopae; ... Kai po VT]Gle;, 411d4) and aya9v (to 900 ... t!> ayaG't!> atr 'l E7trovullia EGtiv, "taya9v," 412c4-5).186Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 191. well as that between apen and KaKa (415a9-el). AIso relevant here is the contrast between lCaAv and aiaxpv (416alO-dll), and that involving aA~gela and 'l'euOOC; (421bl-7).In addition, Plato includes the terms 1tl9Ulla and 9UllC; in hisetymological discussion (419d8-e2). Even more centrally, Plato provides the tenns of the contrast between 7tUJ't1llT, VTO'lC;, qpVTcrtC;, and aoqa, on the one hand, and o~a,on the other, a dichotomy which is essential to his epistemological reflections.Key here are tenns denoting concepts and distinctions which play crucial roles in the Phaedo, Republic, and Symposium, and which, in many cases, are subject there to sustained philosophical analysis. 136 One finds analyzed here tenns designating entities with the highest ontological status from Plato's perspective, in contrast to those entities to which poets and Presocratic philosophers give pride of place. Ihis is not to say that all three cIasses of entities share no common properties; for example, neither gods nor natural elements nor Platonic Fonns are subject ultimately to perishing, and all are invisible, at least in part. 137 Notwithstanding this Iimited sharing of attributes, for Plato it is the differences between the types of entity that are ultimately decisive. On the issue of explanatory priority in particular, Plato challenges the approaches taken both by poets and by philosophers. As concems poets, this comes out notably in the Cratylus itself; there, Plato disputes repeatedly the paradigm of explanation in tenns of origin, insisting instead that merely identifying a god as the source of sorne phenomenon or product, e.g., an ovolla, constitutes merely an excuse for having no genuine account to offer. As concems his philosophical predecessors, Plato criticizes them sharply in the Phaedo for privileging the notion of material causation (96aff.).138 He accuses them-including Anaxagoras, whose approach Plato describes as initially more promising-of conflating a merely necessary condition (lCEtVO aveu oi}) with a136 The folIowing chapter contains additional remarks on the aforementioned terms and referents.137 Among the Milesians, though what was construed as fundamental was something material, it was thought of as being only partially visible, as in the case of water, or invisible yet substantial, as in the case of airo Thales appeared to view water not just as the original state from which the cosmos developed, but also as the underlying constituent of entities in the developed world; since it is not perceived as uniformIy distributed, sorne ofthe water constituting entities must be invisible. Anaximander's apXl, is not identical with any natural substance in the developed world; it is qualitatively indefinite, hence not as such visible. With regard to Anaximenes, Kirk notes that for him "aTp was not mist, but ... the invisible atmospheric air.. Jt looks ... as though Anaximenes simply assumed that sorne part, at least, of the atmospheric air was substantial. and indeed the basic forro of substance" (The Presocratic Philosophers. 146). 138 For present purposes 1 need not address the question of precisely how much of what is said regarding natural science there can be attributed to the historical Socrates (for a discussion of this issue see Hackforth,Phaedo. 127ff.).187Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 192. genuine explanans, i.e., that on account of which something happens or is accomplished (99b). Plato notes the explanatory priority of Forms at the end of theCratylus, and treats it at length in the Phaedo, Symposium, and Republic. This section shows most vividly of all how wrong one goes by seeking insight via the analysis of words' constitution: because it contains the most important tenns treated, consideration of them and their referents merely by etymology casts into sharpest relief the superficiality of that approach to correctness and insight. 411b-c further extends the scope of ~~rov~EVta OlEAoGal, E1tElta trov tProv lCata do" ta tE acprova lCal aq>80yya-otrocrl yp 1tO) AyOUcrlV Ol OElVOl 1tEPl tOtmv-lCal ta a' q>rov1EVta J,1EV O, OU J,1VtOl yE aq>80yya; lCal autrov trov q>rov"Vtrov o(m olq>opa do" EXEl aAAlArov; lCal E1tElOaV 'Cauta OlEAwJ,tE8a [ta ovtaJ El) 1tavta a;) ot; OEl ov.ata E1tlBE1VIX1, Ei EcrtlV Ei~ avaq>pEtal 1tvta roO'1tEP ta crtolXEla, E~ 6>V EO'tlV iOElv at tE lCaL Ei EV atoi~ EVEO'tlV do" lCata tov autov tP1tOV roO'1tEP EV tOl~ crtOlXEOlC;' 'Cauta 1tavta lCaAro~ olaBEacraJ,1vou~ E1tO'ta0'8al E1tlq>PElV ElCaO'tov lCata t~v /lOltTJta, EvtE EV Vl o!1 E1tlq>pElV, EvtE (JU'Y!CEpavvvta 1tolla [v], rocr1tEp Ol 'roypljlOl !X>UM/lEVOl aq>o.owv EvOtE /lEv OOtPEOV J,1vov E1tlVE)'ICav, EvotE oe tlOUV aAAo trov q>ap/lalCrov, EO'tl Oe otE 1tOAAa O'uylCEpcraVtE;, otov otav avopElcEAov O'lCEua~roO'lV ~ aAAo tl trov tOlOtrov-lC; a.v ol.al &1CfI lcaGtTJ i Eilc:wv oEi0'9al lCacrtou cpap/lalCou-otro ol lCal i'Ei~ ta O'tOlXEla E1tl ta 1tpaY/lata E1toO'O/lEV, lCal EV E1tl EV, o{) a.v ooKi OEiv, lCal O"/l1tOAAa, 1tOlOUvtE~ ol O'UAAaia; lCaAOUO'lV, lCal crUAAaia; a' cruVtlaVtE~, E~ 6>V ta tE V/lata leal ta pilllata O'JVti9EVtal' lCal 1tAlV ElC trov VO/lcltroV lCal p,,/latrov /lya To" tl lCal !CaAov !Cal OAoV ov tii ypaq>lKi, Evtau8a toV Ayov tii vo/lacrm:n i1 p"topt1Cfl i1 ittc; EO'tlv i tXV". (424c5-425a5) Then must we not begin first of aIl by dividing vowels, then a1so consonants and mutes, into c1asses-for this is how experts in these matters speak-and in tum those letters which are neither vowels nor mutes? And must we not also distinguish the different c1asses in the case of vowels? And when we have thoroughly distinguished a11 the existent things to which, in tum, one must apply onomata, we will proceed analogously on the side of ta onta, which inquiry would allow one to see existents as they truly are and whether there they have c1asses within them in the same way as do the basic elements of language. After we have considered these things thoroughly we will know how to apply letters based on the principIe of Iikeness, whether there is need of a one-to-one correlation or whether many letters are lo be combined, just as painters, wanting to create a likeness, sometimes uses only purple or some other single color, and other times mixes many colors together, as for instance when creating a flesh-colored pigment or sorne otber such thing. As the artist uses colors based in each case on what the image requires, so too do we apply elements to things, making use of one or many as required. From these elements we construct syllables, and in tum put syllables logether lo get onomata and rhemata. Combining these we arrive at language, large and fine and whole-just as happens with the figure in painling-by Ihe art of grammar or rhetoric, or whatever techne one wishes to invoke. An interest in basic Iinguistic elements is evidenl in several later dialogues. With this treatment of sounds compare Phi/. 18b6-d2, where the Egyptian Theuth is Ihe one supposed to have engaged in the inquiry there described (for Theuth as the inventor of writing see Plulr. 274c5-275b2). For remarks on the combinalion of ypa/l/lata see Soph. 252e9-253a12, and for extended discussion of syllables and their crtOlXEia see Th. 202eff. Notably, these discussions in the Theaetetus, Sophist, and Philebus a11 post-date Plalo's introduction of the Method of Division, which occurs in the Phaedrus (266b-c). For a treatment of olapEcn; that privileges the markedly innovative character of the Sophist and PoliticlIs investigations by contrast with earlier methodologies see Stenzel, Plato's Method of Dialectic.ao189Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 194. Here one finds an instanee of division. However, one is not entitled to conclude that the Method of Division as such is at issue, any more than one would justified in so concluding with regard to the Gorgias based on its employment, in a speeific context, of a classifieatory schema.l 42 In this passage, the 'txval of rhetoric and naming are mentioned together (425a4-5), which is noteworthy given the faet that the 'txvTl status of each is challenged in the Gorgias and Cratylus, respeetively. It is possible, at any rate, that Plato here hearkens back to the earHer Gorgias diseussion and the fate of rhetoric in that context. While supposedly describing the original procedure of formation, Socrates slips into the fiest person, then purports to correet himself: Jl&AAOV O oux nJlEte;, aAAa. AyOlV ;TlvX~v. O'uvSEO'av JlEv ya.p O'tOle; n1tEp O'"(KEt'tal o 1taAalOt nJl&e; OE oE'l, Et1tEp 'tEXV1KOOe; 1tlO''tTlO'JlESa O'K01tE10'9al au'ta. 1tv'ta, O'tOl 01EAO/lVOUe;, et'tE Ka'ta. 'tp1tOV 't 'tE 1tpoo'ta ov/la'ta KEt'tat Kal. 'ta. O''tEpa Et'tE /ll, O'tOl SE&O'Sal" aAAOlC; OE O'uvEPE1V /l~ J.1l] i8Aovn Ayov BtBvat 7tEpl trov 7tPc.trov ovoJ.1'trov roe; p8roe; KEttat {426al-3).l48 These are precisely the sorts of factors that were invoked throughout the investigation to explain the genesis of tenns presently in use, and Plato's remarks here are intended to showthat 424c-425a describes no approach yet taken, and that in faet one cannot supply a rationa! account of "proeedures" followed to date. It is not just su eh "evasions" that Plato takes to be uoacceptable, but any falling short with regard to the procedure outlined aboye. This comes out in the lines direetIy followiog the aboye remark (io fact, Plato highIights the continuity by use of KatOt): KatOt Otrp ne; tp7trp trov 7tPCtrov voJ.1trov tl]v op8tT)'ta J.1l] OiOEV, aOva'tv 7tOU trov yE -uatprov eioVat, a i~ KEVrov avYKT oT)Aoua8at Chv nc; 7tpt J.1T)OEV OiOEV aAAa. OfiAoV o'tt,''''''"i' ,-. , tov epaaKOV'ta 7tEpt autrov tEXVtKov Etvat 7tEpt trov 7tprotrov ovo,k1trov J.1A1.C1t tE Kal Ka8apc.>tata OEt EXEtV a7tOOE'i.~at, 11 E eiOvat on t rE a'tEpa iOT) qlAUaplaEt. 11 aOl aAAroc; OOKEt; (426a3-b3)149146 This is one of the two requirements mentioned in rhetoric's exclusion from the c1ass of 1:XVat at the oulSet of Socrates' conversaton with Polus: KOAaKElaV .lEY ot)v aU1:o KaA.ro, Kal aiOXpv ltPOO

OelV, EltE 7tap' E9pel,OTJAOtbLClKpcxn:~, 'to J.l.OtClJ.l.CX'tl OTJAOV O'tlav'tl~aMa. J.l.Tt 't) E1tl'tUxvn, 434al-2, ef. 433d8-e2); and that resemblance157 Once again, the or.toupj'Oc; ovo.'tCov and vo.09trc; are identified (431eI-4), and naming's 'tXVll status is assumed (e6-7, 11) though no Ionger with any conviction. 158 Contrast Cratylus' remark on the ov.ata of letters in 43Ie-432a. AIso, contrast the observation that something can be designated when this "general character" is present, ICa.V .il nv'ta 'ta npocrtlCov'ta eXl1 (43334-5), with the more restricted acknowIedgment in 432e3. 198Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 203. between VJ.UX'tCl and their referents is only possible if a resemblance relation holds on the most basic level, namely, that of letters. The elaim that representation by likeness is infinitely superior (oA.cp KCll1tClv'tl BlCl(j)pEl) will be evaluated in remarks made in 434b-440e. By posing the choice here as one between resemblance and randomness (434al-2), Plato wishes once again for choice of a nature-based view to appear temporarily "obvious." Here, as in Hermogenes' early remarks, an extreme position and a genuine convention-based approach are conflated (see 433e2-434a2). In what follows, however, Plato leaves the extreme stance behind, and assesses the latter as an independent ground of correctness when terms' constitution is at issue. C. 434b9-435d 1: The resemblance theory is challenged.1. 434b9-435b6: Plato has Socrates consider the noun aKA,l1P't11~ ("hardness").In addition to the letters rho and sigma, which are expressive of rapidity, motion, and hardness( VOJlOou "(E ~ o 01ClVOOJ.l.EVO~ (j)8yyOJ.l.Cll, Elm:p 'to A,elPoCl aVJ.l.olV E (JJeA,l1P'tT'!n El BE 'toho o'twe; EXE1, 't aUo ~ ClU'tOe; ao:mep auv80u leCl VJ.l.Cl'tOe; auv8fKTl, E1tElOf "(E 011A,Ot !Cal. 'ta. OJlOlCl KCll 'ta. aVJ.l.olCl "(pelJ.l.J.l.Cl'tIl, t80ue; 'tE KCll auv8fKTle; 'tUXV'tCl; El o' on J.l.elA,lV iO"~ av'tt 'tOl> A.~aapro ael iyetv (434d12), this is rejeeted as whollyunnecessary since thepresence of letters signifying opposites in no way inhibits communication. The possibility of suceessful communication rests ultimately on the faet that certain combinations of letters are sanctioned by custom and convention as denoting particular entities; likeness is not the basis of signification, but custom, whieh signifies by like and unlike letters as such combinations happen10be established.In the case of O"1CA.TlPtl1~, Plato focuses on the presence of letters signifying opposites beeause this brings into sharpest relief the faet that likeness is inessential. Additional support for this conclusion les in the faet that the word O"KATlPtTl (1555).224Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 229. sprung from gold" (80); the poet refers here to the hero Perseus, who was conceived following Zeus' approach to Danae in a shower of gold. 26 Strictly speaking, any suggested judgment of desert pertains not to the Persians as a group, but to an individual Persian; nevertheless, Xerxes is no ordinary Persian, but his people's supreme leader and representative. My research shows that the literary tradition was quite con cerned with etyrnology and eponymy, but never distinguished clearly between them; in fact, authors use E1t>VUJlO~ and related terms indiscriminately with reference both to etyrnology and to eponyrny. In contrast, Plato distinguishes sharply between etymology and eponymy; although he devotes much of the Cratylus to contesting an etymology-based foundation of appropriateness, he does not, there or elsewhere, argue similarly against eponymy. Instead, I suggest, he taps its unexploited potential. In its use of eponymy, the literary tradition typically rests content with identifying the entity giving rise to an assignment. In contrast, Plato maintains that a satisfactory theory must be grounded on a determination of the nature and status of the primary entity; this, in tum, yields a set of conditions under which that entity's OVOI.HI can properIy be applied to a certain class of recipients. Wbile Plato manifests a limited interest in individual sensibles, his main concem is with the Forms; therefore, the eponymy relation's ultimate focus on the primary entity, rather than on the recipient, is quite in keeping with his own orientation. In the Phaedo, Plato embraces eponyrny as the optimal framework for presenting the semantics of the Fonnparticular relation; his transformed version of eponymy incorporates those revisions necessitated by the structure and content of his middle-period metaphysics.THE CRATYLUS In the Cratylus, Plato explores rival conceptions of the ground of words' appropriateness or correctness (oP8tllO'e1.~ of specific mortal or divine individuals as the objects of etymological analysis). More generally, Plato's approach exemplifies writers' conflation of proper names and general terms, and of judgments of appropriateness involving veridicality and desert;38 it also manifests the literary tradition' s invocation of a range of different criteria, most notably characteristics and extensions of bearers, as the basis of its "deep" structural analyses. While the literary tradition operated unreflectiveIy and unsystematically, Plato employs that tradition's assumptions and analytic techniques quite consciously, with the ultimate goal of discrediting them. As previously noted, the literary tradition's etymology-based judgments of appropriateness do not involve clear-cut and pervasive normative criteria. In fact, Plato mounts a two-pronged attack on those judgments' status as norm-govemed. First, he concentrates on the literary tradition' s own assumptions, according to wlch one attributes "natures" to mortal and divine individuals. Study of this tradition suggests that namers' hopes for bearers' futures underlie a large class of assignments, which may successfully disclose mature bearers'

O'El~ or oUO'lal. In his view, any ta!k of so-called individual natures would be, by definition. exclusively a matter of O~a; one simply cannot have E1tlO't1'T) of the entities in question. which would be the facuIty required to sustain c1aims of the process' elevated status. As noled in the preceding chapter, having q>O'El~ or oUO'lal as subject matter is the linchpin of the tXvn enterprise: they are pivota! to specifications of a given txv1l' s tA.O~; they are what expert practitioners understand; they are the uItimate sources of a tXV1l' s benefit to human beings; and they are pivota! to any rational account of a tXV1l' s procedures. 48 On the dialogue's negative emphasis cf. Mridier (38). As noted in what precedes. severa! commentators see in the dialogue a great deal ofPlato's own positive view. 1 readily acknowledge that the Cratylus c1ears the ground for subsequent developments by disposing of rival notions of the import of words and the basis of their appropriateness; although Plato offers c1ues to his positive stance, their development occurs elsewhere.235Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 240. number examples (434c-435c) shows that he opts for the convention-based prong of the nature-convention dichotomy; that is, he maintains that one judges correctness by appeal to custom (eGoC;) and convention (O'uv9TK11).49 Plato's view is not simply that custom and convention contribute to words' appropriateness, but rather that ultimately they determineit. Notably, the relevant norms are wholly detached from words' constitution; this, Plato insists, neither refiects nor prescribes standards. The conclusioo reached here applies to every OVOIla., whether proper name or general term: regarding any word's constitu