THE CLASSICAL QUARTERLY JULY—OCTOBER, 1927. TECHNICAL TERMS IN ARISTOPHANES. Every living science, especially in its early stages, is compelled to devise fresh terms, either by coining new words or by giving new meanings to old ones. Unless and until these fresh terms become absorbed in the vocabulary of everyday speech, their unfamiliarity makes them a target for the shafts of the humourist. There can be no doubt that in the late fifth century B.c. literary criticism (using the expression in its widest sense, to include all methodical investigation of literary technique) was still a new science. We can trace its beginnings in the treatises of the Sophists, many titles of which have been handed down to us. Strepsiades’ lesson in metric, though of itself amusing enough, would certainly gain in topical appropriateness if enacted at a time when such investigations were not only much in the air, but were still novel. And the whole ‘ Agon’ of the Frogs, the character of which is forecasted in lines 796-802, depicts in the strongest colours the contrasted views of technician and inspirationist. We should therefore naturally expect a play of such a kind, written at such a time, to be full of technical jargon, barely understood by the ‘ man in the street,’ and forming the object of his half-con- temptuous amusement. That is, I believe, exactly what we do find, to an extent insufficiently recognized. Professor Radermacher, in his recent edition of the Frogs, has rendered valuable service by pointing out the frequent occurrence in that play of technical terms which meet us later in Dionysius of Halicarnassus and other critics. But I believe that technical language lurks unsuspected in many other passages, though the precise meaning may often be beyond recovery. Let us take the points sertatum. In 98, yoveyos is a word used literally of a fertilized egg, as opposed toa wind-egg (dveyiaios). Both words are used metaphorically by Plato in Theaet. 150C to 151E. As applied to a poet, yovtuos clearly means ‘ full of pro- ductive energy.’ Theword is palpably unfamiliar to the audience. The Philistine Herakles repeats it with a puzzled air, and Dionysus explains, repeating the word yet again, so that no one may miss it. We come across yévipos later, in its literary sense, in Lucian’s Orator’s Instructor, § 25, in a context which suggests that in his day, too, the expression had something comic about it: mdvra abth y’ (sc. } yAaTTa) ériatdcOw Kai youpwrépayevécOw. (In Long. 31. I yovtpov probably governs a lost genitive.) idt@tas, in 459, is admittedly difficult. It can hardly, as most editors believe, equal woAditas. Tucker ingeniously suggests that the chorus are NOS, III. AND IV., VOL. XXI. I 116 ]. D. DENNISTON means either euphony in the literary sense, or the melodious recitation of lines, and that Aristophanes drags the word in by the heels because it is a novelty. He also uses the comparative adverb in an odd context in Eccl. 807 : TOXD yap éuperéorepov mpotepov yécas THeEiv } TpLtaKov®” npuépas That passage, too, looks like parody. adwwdnOpa in 904 is explained by the lexicographers as meaning a rolling- place for horses, or the sanded floor of the wrestling ring. It is to the second sense that we must look in the present passage. Wrestling has, it is true, been mentioned in the strophe (878): but dAcvdy@pa is a curious metaphor in itself, and clashes rather violently with the picture of Aeschylus as a giant, rooting up words like tree-trunks. Here, as in 819, I believe that we have an inten- tional confusion of metaphor, and that ddAwd7nOpa was already known to the audience in a technical sense. It is exactly the sort of metaphor which might well be used to denote a peXérn, or exercise—a training ground, as it were, on - which to practice for bigger things. In 942 émvaAdrious is usually explained as denoting the metrically lighter line, with frequent resolved feet, employed by Euripides, particularly in his later period. This view is supported by the verse-weighing scene (1378-1410). But in Athenaeus (654) é7vAXLov Means a miniature epic: and it is conceivable that in Aristophanes it may be used to describe the elaborate messenger- speeches which are so prominent a feature in Euripides’ plays. But whatever its sense, the word is closely connected with Euripides’ art. It occurs thrice in classical Greek, and every time in reference tohim. Dicaeopolis, on calling at the poet’s house, is informed that— 0 vous pev Ew EvrAXEyor erTVANA ovK évodov, avtos 8 évdov avaBadny trove? diav. sae ua Ach. 398-400. And, in the Peace (532), Trygaeus enumerates émvA\a Evpiridov among the blessings of peace, along with roast thrush and Sophoclean lyrics. All this suggests a definite connexion between Euripides and the word émvAdov. No doubt the diminutive form also contributes to the jest. We may compare pnuatiov, used of the epigrams of demagogic orators in Knights 216 and Wasps 668, and of Euripides in Ach. 444, 447, the second time with emphatic repetition ed x’ olov 75n pnwatiov éuripmrapat, and Peace 534 rom pnuatiov Sixavixav. Cp. also the string of diminutives in Knights 100 BovAevpartiowr Kai yvoptdiov Kal voidiwv. mepimators in 942 has usually been taken to mean philosophizing. (Hesychius may be thinking of our passage when he writes: qwepiratoc: ai ioropiat. Kal oi AOyoL. 7 Tomo. Siaxivncewv.) But editors who offer this explanation give no grounds for supposing any connexion between walking and philosophizing, before the foundation of the Peripatetic School. That Plato al (Prot. 3 convers can finc Raderm explains reasonal Orator’s down th Eustath: ing siler emotion. been, as At any r his actor did not i have ma and desc not this « was topi And Ast: (another é€a Bo (e.g. 38. 2 to our pul used for t climax or was com Euripides TO ‘yévos ° opening | resourcefu etc.) as fa The \ as Many [ Xpna moro: The ’Ezio 1 Fr. 7. occurrence yf lines, ovelty. rolling- second 1e, been n itself, rooting 1 inten- to the h might vere, On » lighter 7 in his 8-1410). ceivable ssenger- rhatever s thrice | calling ong the All this ov. No compare 216 and mphatic aT iov Tiwy Kab phizing. aToL* at ffer this cing and at Plato TECHNICAL TERMS IN ARISTOPHANES 117 (Prot. 314E) represents Protagoras as év t@ mpocte@ mepirarovvra, while he converses with his friends, is not very remarkable: and the earliest passage I can find implying a connexion between the two activities is Alexis, Fr. 147: lS KALPOV HKELS, WS Eywy’ aTropoupervn dvw KaTw Te TepiTaTovo’ wotep IIXatwv copov ovdey eiipnn’, GAXA KoTL@ TA OKENN. Radermacher regards epimdtots as a ‘rhetorischer Kunstausdruck,’ and explains it as allied to Aadid, and opposed to 1d civtopov.' This is a reasonable conjecture, but we may look for alternatives. Lucian, in the Orator’s Instructor (§§ 21-2), recommends the aspiring orator to walk up and down the platform while speaking: and the word he uses is wepizratos. Now Eustathius (pp. 1343, 62: 1941, 1) tells us that Aeschylus was fond of present- ing silent, veiled, motionless characters in an attitude expressing some form of emotion. Aristophanes himself refers to this in g11-20. Aeschylus may have been, as Eustathius suggests in one case, influenced by paintings he had seen. At any rate, he appears to have had a liking for a sort of tableau treatment of his actors, while devising elaborate movements for his chorus. Euripides, who did not intend his characters to be mere rrpocyijyata tis tpaypdias, may well have made them move about the stage with greater freedom and naturalness, and described this by the more or less technical term trepirratos. Whether or not this explanation is correct, there can be no doubt that the word vepitaros was topical. It is dragged in by the heels, with obvious emphasis, in 953. And Astydamas (whether the elder or the younger) combines it with yA@rra (another topical word: see below) in the remarkable line— yAooons tepitraros éotiv adoreryia.! ésBoras (956, 1104) is thrice used by Longinus in the sense of opening (e.g. 38. 2 €v TH eta BorAR Tod Llavnyvpixod). But earlier in date, and more apposite to our purpose, is a fragment from Antiphanes’ Poets (Fr. 191), where eto Sonn is used for the opening of a play, as opposed to cataotpody, the carrying of it to its climax or conclusion. It seems likely that the literary sense of this word, too, was comparatively new in the year 405, and was possibly originated by Euripides. He was clearly much preoccupied with the problem of expounding TO yevos Tov Spduaros (946) without relapsing into the stereotyped form of opening parodied in the Anx«vvov episode: the extant plays illustrate the resourcefulness with which he devised openings (a wish, a prayer, a gnome, etc.) as far as possible removed from formal narrative. The word §:8réov had a special significance for an Aristophanic audience, as many passages show. There is a plague of books in NegeXoxoxxvyia. The Xpnaoporoyos enters with one, which provides the main joke of the episode. The ’Ezicxoros attributes his arrival to a dadrov BiBriov Tedéov (1024), which tragedy, when the mixture of genres was not un- common. 1 Fr.7. There is nothing remarkable in the occurrence of such a line in fourth-century 118 j. D. DENNISTON seems a rather unnatural expression.’ And when on the top of this the WVndicpatorearns enters with yet another volume open in his hand, Pisthetaerus can stand it no longer. rovti ti éoTw ad xaxov To BiBriov; (1036). The cumulative effect of these episodes is remarkable. Even more striking is a fragment from the Tagenistae (Fr. 490) : Tovtov tov avdp’ } BiBrLov drépOopev } Upodixos ) trav adoreoyxar els yé TUS. We have no context, but clearly a book is represented as a dangerous object in itself. In the light of these passages, we should be chary of interpreting 8tBdiov in Frogs 1114 as a work on literary criticism, a book of the words, or any other definite book. What, then, is the point of the ‘book’ joke? It would be simple enough if books had been inordinately rare in the Athens of Aristophanes, as rare, say, as peacocks. But this can hardly have been so. Books were a normal article of export in Xenophon’s day (Az. VII. v. 14), and Socrates in the Apology talks of picking up the works of Anaxagoras at a drachma apiece. Booksellers are often mentioned in the comic fragments. The truth seems rather to be that, while books were not excessively rare, they were still rare enough to be the hall-mark of a type. The fact that an individual in fifth-century Athens was always seen with a book marked him out, even if you did not know what the book was, as a member of the ‘ reading set.’ Euripides certainly belonged to the set. He was one of the first Greeks to amass a considerable library (Ath. 3a). His plays were ‘concentrated essence of books’ (943). In the weighing scene he is invited to throw his books into the scale as a makeweight (1409). Huppolytus 451-2 ef \ < 4 a / Ocot EV OVY yYpahas TE TWY TAXALTEpa@V éyouvolv, auto T’ eialy év povoais del seems to imply interest in reading as a pursuit (as Nauck observes in his Vita Euripidis. ypadds in the sense of ‘ paintings’ is less probable). Still more significant is Fr. 369, where the chorus expresses a desire to hang its armour on the wall and enjoy a tranquil old age— déATev 7 avaTTiacoiw ypu av codoi KréovTat. It must be remembered, too, that in the fourth century appear the ava- yvwortxot, who wrote plays for the study: and it may not be fortuitous that Euripides’ devoted admirer Dionysus read the Andromeda on board ship. 1 In spite of Dittenberger, Sy//. 117, quoted by Linguistic conventions are so capricious that in van Leeuwen : 7d dé BiSdiov [7d Wndlouaros wapa- Oxford to-day if you talk to a ‘Greats’ man of dévac abr ]ax Tov ypauparéa rijs Bods abrixa wddka. Herodotus and Thucydides as ‘texts,’ he under- Birds 1288 is also curious: xarjpay és rd BiBXia. stands you: whereas no ‘Mods’ man would We scarcely know enough of Greek idiom to say understand the word as applied to the Poetics and with certainty whether the word f:8Nov, without Ars Poetica. qualification, is used naturally in these passages. I to su used ¢ in odc insiste struck Tpaye pay} THS TE Cratir avTav C descril groun Clouds Me but oc pvOwv (295: : and F satiric: almost fact. (J at the NET TOL dpevi : Kal Tru NOyos : Anpwv compo doKols porte especia indicat Here t / pe 1 Euri Anaxago interest this the sthetaerus 36). The king is a object in terpreting words, or ioke? It Athens of been so. . 14), and was at a fragments. rare, they that an rked him - ‘reading st Greeks centrated hrow his 1 his Vita till more S$ armour the ava- tous that ip. ous that in ats’ man of ” he under- man would 2 Poetics and TECHNICAL TERMS IN ARISTOPHANES 11g In all the passages quoted above there is, I think, something in the context to suggest that technical jargon is being satirized. Either the expressions used are odd in themselves, or they are odd in their surroundings, or combined in odd ways, or else they are repeated at a close interval with an unmistakable insistence. But besides these there are many other expressions which probably struck the audience as new-fangled, pompous, or pedantic: 862 ra veidpa tis tpaywdias : 899 axivnro: (see Radermacher): 1122 doadys ev tH dpdoe TaV mpayyatev (cf. Phrynichus, Fr. 55, 77 SvaOéces trav emav): 1120 TO mpaToV THS Tpaypoias wépos: 1178 ororByv: 1281 otdaow perov: Clouds 1367 a€vorartov: Cratin. Fr. 85 dav un cvorpéwn ta mpdypata: Fr. 427 perexBory: Fr. 386 avravayvevas (collate). Closely allied to these literary terms are certain others which I will describe as ‘ intellectualist.’ Here Euripides and Socrates are on common ground, and the vocabulary of the Frogs often coincides with that of the Clouds. Aeros in the intellectual sense is never found in Aeschylus or Sophocles, but occurs three times in Euripides: Med. 529 voids Xer rds: 1b. 1082 ANerrroTépwv wvOwv: Fr. Inc. 924 XerTav pvOwv. Elsewhere we find it in Hippocrates (295- 25 és Tas Téyvas twaxées ov AeTTTOL), Antiphon (3. 8. 2 Aeros Kal axpsBns), and Fr. Iamb. Adesp. 16 Xerrtov vdov. Otherwise it only seems to be found in satirical passages, where it often appears, sometimes coupled with v«vds, almost invariably with reference to the philosophers or Euripides. Plato, in fact. (Rep. 607C), quotes AerTas peptsvavTes as a typical sneer of the poets at the philosophers; and Alexis (Fr. 221) speaks of the philosophers’ Ad-yor NeTrToL SveopsrAevpévae te Hpovtides. Cf. also Ach. 445 wuKvyn yap AeTTA pnxXava gpevi: Eccl. 571 muxvnv ppéva xai dirocogov éyeipey: Amphis, Fr. 33, Ae wTAS Kal muKveas é&erdfew : Hermipp. Fr. 22 AertoAoyiay: Cratin. Fr. 307 broXetT0- Noyos: Anaxandr. Fr. 36 Aemrrotépws. In the Clouds Socrates is AeTwToTAaTwY Anpwv iepevs, and Strepsiades is all agog AerroAoyeiv. The word, with its compounds, is, in fact, strewn freely over the Clouds. Svaderroroyotpar tais Soxois HS oixias, Strepsiades observes maliciously as he sets fire to the Ppovticrnpiov. In the Frogs, too, it is well to the fore: e.g. 876, 956, and especially 1108-11, with the pointed repetition of the word. But the clearest indication of the amusement it afforded is Lys. 28-9: KA. 4 ov tt Nerrov eats TOUppiTTTAacpEVoOY. AT. ob tw ye XNerrtov dof SANs THs “EAAdSos év tais yuvatkiv éoti 7 owrnpia. Here the repetition, as in Frogs 96-8, throws the whole stress on the adjective. péptuva, with its verb wepysvav, seems to have been a poetical expression 1 Euripides is said to have been a pupil of and in the doctrines of Diogenes of Apollonia in Anaxagoras and a friend of Socrates, His particular, can be freely illustrated from his interest in philosophy and science in general, poetry. 120 J. D. DENNISTON brought into prose use by the philosophers. Aristophanes, who never uses péptpva elsewhere, has it thrice in the Clouds, and the compounds pepipvo- gpovtiatns and dmepipepipyws once each. Fr. 672 ds ta péev adavy pepypva ra Se yapa0ev éoOie clearly refers to a philosopher. A™ertas pepyuvavtes I have mentioned above. gpovtis and gpovritew are scarcely used by Aristophanes, in a positive sense, except of the philosophers. Eccl. 571-2 wuxvynv dpéva nal dirocodov gdpovtida: Nub. 723 obtos ti moveis ; ovyt dpovrifeis: Epicr. Fr. 11. 4 tota gpovtis, motos Sé Aoyos .. .; id. Fr. 11. 22 Suvehpovtifov. gpovtiorns and, of course, dpovtiornpiov are peculiar to the Clouds. éxdpovtica: (695) and xkatateppovtica (857) are comic variants on the simple verb. The humorous use of yA@rra is a stock cliché of comedy. Cratinus (Fr. 293) addresses Pericles as ® pevyiorn yA@TTa tov “EXAnvidwrv. Strepsiades is compelled to swear allegiance to ‘Chaos, Clouds, and Tongue.’ The victim’s tongue is taken out at sacrifices, in compliment to the éyyAwrroydo- topes (Birds 1705). Under the care of the “Ad:xos Adéyos Pheidippides’ tongue will grow in inverse ratio to his zruyy (Clouds 1013-8). Its ‘double edge’ will ‘ glitter’ (7b. 1160). A character in Plato’s Zeds xaxovpevos, singing the praises of the tongue, reminds his hearers that they wear tongues in their shoes, and vmoyhwttidas—whatever that may mean—on their heads at dinner. I have already alluded to yAwoons repizraros in Astydamas. In the Frogs Aristophanes makes the tongue uncoil itself (827 yAa@oo"’ avedtocopevn), and lose its temper (898 yAdooa pev yap jypiwra). The hinge on which it wags, yAwoons otpoguyé, is one of the deities to whom Euripides prays. Satyrus tells us that Aristophanes expressed a desire to measure this wonderful tongue of Euripides : du’ Hs Ta AerrTa pyyat’ éLeopnxeTo. Cf. also Eccl. 573-4 xaw7 . . . yAwrTrns érivoa. What is the point of all this? It seems possible that yAwrta was a popular sobriquet for an intellectual of any kind. Cratin. Fr. 293 and Bekk. Anecd. 31. 17 (for what it is worth—rov yropais roddais Kal cvvexéor Tovs axpowpévous Siauadratrovta. TAdrwv) tend to support this explanation. Or else yA@tra was a term used by intellectuals in some kind of esoteric significance, to represent, say, the power of expressing thought. Cf. Eur. Supp. 203-4: mpa@tov pev évOels avvecwy, eita 8’ ayyeXov yA@oocav Noywv Sovs. Or, again, yA@Tra may have been primarily an esoteric term, and, secondarily, a nickname for a type. But, whatever the reason, there can be no doubt that the word was a source of amusement to an Athenian audience. (It may be objected to all this that, so far as we can discover from Plato and other sources, Greek philosophers, at any rate before the time of Aristotle, used a on whi the styl I have to talk have be In as it i usually I cann have gt must b in wor at the 1 directic categor What t fer uses MEP t{LVo- HEptpva VOVTES positive Loropov 4 twoia ¢ and, 5) and ratinus psiades . The TToydao- tongue re’ will praises es, and I have phanes temper NOTONS ells us gue of was a - Bekk. tL TOUS n. Or soteric . Eur. darily, ot that TECHNICAL TERMS IN ARISTOPHANES 12 used a language noticeably free from technicality. That is a point, however, on which it is unsafe to dogmatize. How much, after all, do we know of the style and diction of many of these philosophers ? In any case, the evidence I have given above shows clearly that the philosophers were popularly believed to talk a language of their own, whatever the grounds for this belief may have been.) In an investigation of this kind it is as necessary to use the imagination as it is perilous to abuse it. Anything in the nature of demonstration is usually out of the question. The last word is with personal feeling, and I cannot expect universal acceptance of judgments which, for my own mind, have grown more confident as I have re-read Aristophanes. But the evidence must be taken cumulatively. And if it be granted that Aristophanes abounds in words of which the esoteric or technical character can be detected, but at the meaning of which we can only guess (any attempts I have made in this direction are purely tentative), we should be chary of explaining words of this category, and still more chary of athetizing them without strong reason. What the text of Aristophanes wants is, not obeli, but inverted commas. J. D. DENNIsSTON. MAGIC SPEARS. § 1. AMONG the conundrums relating to Roman ritual which Plutarch set himself to solve was this: ‘ Why do they part the hair of brides with the point of a javelin?’ To this question Plutarch offers a number of answers, all based on the assumption that the spear was symbolical. Professor Rose in his discussion of the passage makes short work of these fumbling guesses, and points out that the original purpose of the spear was magical.? We are here concerned with several other parallel cases : I. It is now commonly held that the Roman practice of passing prisoners of war under the yoke, though eventually it no doubt became a mere token of defeat,” probably originated in precautionary magic, its primitive purpose being to intercept the evil spirits which the captured enemies might otherwise disseminate among their captors. This theory, which was first put forward by Warde Fowler* and by Sir James Frazer,* has been strikingly confirmed by Professor Halliday, who has supplied an apt analogy from Medieval Tartary, a yoke of spears under which the relatives of a dead man passed for the avowed purpose of purifying themselves.5 But why was a yoke of spears deemed to have magic in it? Partly no doubt because it was a gate, in which evil influences could be caught in much the same way as the wrong sort of electricity is caught on the grid of a radio valve. But another example from Tartary suggests that this is not the complete explanation: ‘quando aliquis eorum infirmatur ad mortem, ponitur in statione eius una hasta, et circa illam filtrum circumvolvitur nigrum; et ex tunc nullus audet alienus terminos stationum eius intrare.’* The original object of this rite was no doubt to intercept the pollution spread by the dying man, and this function was deemed to be fulfilled by a single spear. We may therefore conclude that while the joining up of several spears to form a gate reinforced the magical effect in the Roman ritual of the yoke, yet that effect was primarily produced by the spears as such.’ This conclusion is of course hypothetical: it may be that the nigrum filtrum is the important point, as a symbol of death, and that among 10Q.R. 87: H. J. Rose, Roman Questions of from the story of the Friar Carpini, 1246 a.D.). Plutarch, 205. ?7 Two spears not merely make a gate with its 2 Livy III. 28. 10. The tigillum sorovium was special value but also give you twice as much so explained by Roscher, Lex. Il. 21; porta spear-magic; cf. Deonna, Rev. Arch. Ve Série, triumphalis by A, von Domaszewski, Abh. z. rém. XX. (1924) 88 sq., on this sort of duplication. Religion, 222 sq. For the gate of purification cf. Ditt. Syll.3 982. 9 3 Classical Review, 1913, 48-51; Roman Essays (Pergamon), xal dvedOdvres rhv widny, Kal’ hv ra and Interpretations, 70-75. 4 The Golden Bough (3rd ed.) XI. 193 sq. 5 Folklore, 1924, 94. 6 Hakluyt, Principal Navigations I. 63 (extract ayworhpa ridera, xaBapoi tcrwoay avOnuepdv ; for the yoke as positive as well as negative cf. A. D. Nock, C.Q., 1926, 101 sqq. the Ta the filt; 2. held. of an 4 origina auctior That s also by evidenc 3. follows T poonk spear v the Att into th been a the to! murder ghost. commo § : commo explan: old, are or repu Sp in Gree P- 5978 Rome | to the « in Sar explain 13. K passim, e (p. 44) th of the vi earlier sa 2 [Den defixiones, 3 S.v. Possibly Cephalus Fab, 189, to me). * The : and the earlier by pleting 1 arch set he point ll based > in his ses, and risoners ‘token of se being -herwise ward by med by rtary, a avowed med to ch evil sort of le from aliquis a illam +rminos oubt to deemed ile the in the spears at the among 6 A.D.). > with its as much rer Série, lication. 1.3 982. 9 10” tv ra pov ; for cf. A. D. M. CARY anp A. D. NOCK 123 the Tartars a spear would be the most readily available pole to which to attach the filtrum or with which to make a yoke. 7 2. It was Roman practice to set up a spear where an auction was being held. This custom might perhaps be interpreted as the primitive equivalent of an advertisement: ‘Cheap bargains in enemy booty!’ But if the yoke was originally intended to purge evil influences from the foemen’s bodies, the auctioneer’s spear might have served the same purpose for their possessions. That some such disinfection of enemy property was desired is perhaps indicated also by another custom indicated by the phrase sub corona uendere; there is evidence for the apotropaic use of wreaths.* 3. Attic custom required the funeral of a murdered person to be ordered as follows: ézreveyxeiv Sopu él rH éxpopa, cal mpoayopevey eri TO pvHpaTt, ei TLS Tpocnkwy éott THS avOpwrou, ETEerTAa TO uyHua HuraTTEL eri TpEIs Nuépas.2 The spear was probably driven into the tomb: Harpocration quotes a fragment of the Attic historian Istros, referring to the story that Erechtheus drove a spear into the tomb of Procris and Cephalus.* The spear in this instance may have been a symbol of a vendetta; on the other hand, the three days’ watch over the tomb suggests that trouble was apprehended from the ghost of the murdered person, and that the spear was a device for trapping and laying the ghost. Other objects could serve similar purposes; a wooden stake has commonly been driven through suicides in modern times. § 2. In each of the cases the use of the spear can be explained in terms of commonsense. Cumulatively these instances perhaps point to some common explanation and suggest that their value is not purely symbolic; the usages, if old, are much more likely to rest on concrete beliefs and on magic of sympathy or repulsion. It is on these lines that we seek an explanation.‘ Spears gua weapons can be sacred. It has frequently been observed that in Greece there are various traces of this sanctity. Plut. De genio Socratis 31, p- 5978, relates that the Theban magistrates carried a spear with them.’ At Rome the same is true, as we see from the veneration paid to Mars hasta and to the ancilia. A spear used in war is fuller of power than one not used; thus in Samoa special veneration is paid to battered war clubs.° This perhaps explains the fact mentioned by Ovid that a bent spear was used to part the 1 J. Kéchling, Relg. Vers. u. Vorarb. XIV. ii. passim, e.g. 81. He follows Wilken in holding (p. 44) that the custom came from the crowning of the victim in sacrifice, captives having been earlier sacrificed (But ?). 2 [Dem.] 47, § 69. Cf. the use of iron nails in defixiones, and Gruppe, Griech. Myth. 895,. 3 S.v. éweveyxeiv Sépu x.7.d., Pp. 77- 5, Bekker. Possibly the spear is the magic spear with which Cephalus killed Procris according to Hyginus, Fab, 189, (as Dr, R. B. Onians has suggested to me). ¢ The apotropaic nature of the hasta caelibaris and the spear in auctions has been recognized earlier by S. Eitrem, Opferritus, 388. After com- pleting this paper we found that F. J. M. de Waele, The Magic Staff or Rod in Graeco-Roman Antiquity, 1927, 177, had anticipated us in ex- plaining the double value of the hasta caelibaris. 5 For other evidence cf. Fr. Schwenn, Arch. f. Rel, XX. 299 sqq. Pausan. IX. 4o., 11 tells of the worship of Agamemnon's sceptre by the Chaero- neans, ddpu édvoudfovres; but may mean simply ‘stick’ and sceptre worship is known, cf. A. B. Cook, Zeus II. 1132,; F. J. M. de Waele, op. cit., 107 sqq. ; and a Lydian confession inscription (Jahvesh. XXIII, Beidl. 7), 1. 15, éwecrdOn obv rijs Geoh rd oxiwrpov kal rod Kuplov rot Tidyov, with Zingerle’s full note, Kol. 13 sqq.; ib., p. 16, No. 2, 1. 9, 4 dé Taria éwéornocev cxirrpov Kal dpas €Onxev év TO vaw. 6 Frazer, Pausanias V, 211. 124 M. CARY anp A. D. NOCK Roman bride’s hair.!. Hasta recurva suggests a spear with its point bent twice in a crook; here it may be one bent with use inwar. The idea that a weapon which has hurt or killed makes a powerful charm is confirmed by Plin. N.H. XXVIII. 33: ferunt difficiles partus statim solui, cum quis tectum, in quo sit grauida, transmisertt lapide uel missilt ex eis qui tria animalia singulis ictibus inter- fecertnt, hominem aprum ursum. probabilius id facit hasta velitaris euolsa corport hominis, si terram non attigerit ; eosdem enim inlata effectus habet. sic et sagittas corport eductas, st terram non attigerint, subiectas cubantibus amatorium esse Orpheus et Archelaus scribunt; and XXXIV. 151: pungique (sc. prodest) leuiter muerone, percussus homo sit, contra dolores laterum pectorumque subitos, qui punctionem adferant. We have further the statement of Festus s.v., p. 55. 3. Lindsay: caelibart hasta caput nubentium comebatur, quae in corpore gladiatoris stetisset abtectt occisique, ut quemadmodum illa coniuncta fuerit cum corpore gladiatoris, sic ipsa cum utro sit. The explanation is on a par with Plutarch’s; the fact is valuable. Gladiatorial games were not endemic at Rome, but imported from Etruria; a spear used in them cannot have been used in early Roman times. Then all the men of the community would fight in war fairly often, and spears which had killed there would be readily available. Later, such spears would most easily be got from the arena.” § 3. This sanctity makes the spear an object of power and usefulness. It has perhaps additional efficacy because its head is made of iron, a metal which holds a peculiar position in ancient superstition.* In numerous instances iron was tabooed, and it is noteworthy that a general ban was placed upon it at Greek sanctuaries.* This taboo is Italian also, as we learn from the expiatory sacrifices offered by the Arval Brothers ob ferrum inlatum in aedem scripturae causa. On the other hand, iron was frequently prescribed as a specific in magical papyri;° and an interesting case is recorded in which the Christian community at Miletus exorcized the evil demons of a sacred stone by nailing upon it a cross made of iron.® Other 1 Ovid, Fasti II. 560. * They would perhaps have additional value from contact with Biawdvara on the virtues ascribed to the blood of gladiators; cf. F. J. Délger, Vorir, Bibl. Warburg, 1923/4 (1926), 196 sqq. The witch in Lucan VI. 547 takes insertum manibus chalybem; nails from crosses which had been occupied by malefactors were employed in magic (L. Fahz, R.G.V.V. II. ii. 44 [=150]). De Waele, op. cit. 177, compares the primitive idea that a youth who would marry must first kill a man, so that his soul passes into him; is the motive clearly that, and not rather the desire to accomplish an exploit proving manhood ? 3 Naturally we do not deny that these prac- tices may well go back to the Bronze Age; but it is of the essence of such things to gain cumulative effect as they go on. 4 Plutarch, Praec. Rei, Publ. Ger. 26, p. 819k: xpvoov peév els Evia Tv iepGy elordvres Ew xarahel- wovet, Gidnpov 8’ ws awrXGs eiwetv eis ovdév cuvecogépovorv. Other instances of taboos are given by Frazer, Golden Bough (3rd ed.), III. 230; T. Wachter, Reinheitsvorschriften im griechischen Kult (R.G.V.V. IX, i. 115-7). Cf, (Callisth.] Hist. Alex. III. 17, 31, p. 113, 8 Kroll (Alex. on visiting an Indian temple is told Saccdted, cldnpov ob xa@jxe eis rd lepdy eiaed- Geir). 5 P. Osloenses I., col. IV. 108, and note by S. Eitrem (p. 56). To the parallel instances given by Eitrem add the Paris papyrus (in Wessely, Griechische Zauberpapyri), ll. 2130, 2153, 2306. Some earlier Oriental uses are discussed later, p. [5], n. 8. 6 Usener, Kleine Schriften, IV. 198. instanc have be It magical Iron is chains, the peo fangled propert Is | and of magnet on Ida, worked magnet Mediter that the of iron EDeNKET historic conscio enjoins Irc to poss propert 1 For s III, 176, prescribec Eitrem, F for cuttir M=p. 33 26.B.3 African uw Testament 3 III. 2 4 Bron: to iron fo bronze gt 1902, 5). was hari tempering that iron its gener: dev Apulei 5 For i 147, sola | magnete) L adprendens dum, Ou by W. R how far h 6 Pliny Pugh kin t twice veapon y Plin. quo sit is inter- corport sagittas Dr pheus Merone, ctionem ndsay : stetisset OY1S, SIC fact is 1 from times. spears would ss. It which that a [talian ‘others m was ig case 1e evil Other D. SIQE: KaTaXel- civ elds ances of igh (3rd ‘schriften 115-7). p. 113, imple is by eiged- note by nstances yrus (in 0, 2153, iscussed MAGIC SPEARS 125 instances in which the use of iron is specially enjoined or specially forbidden have been collected from many other parts of the world.' It remains to ask why iron should have played a prominent part in magical ritual. Sir James Frazer has thrown out some tentative suggestions. Iron is prescribed because (i.) it is strong and hard; (ii.) when made into chains, clappers, etc., it is noisy ;? (iii.) to the users of stone or bronze tools— the people among whom belief in magic is generally strongest—it is a new- fangled and therefore uncanny material. But the only one of the aforesaid properties of iron which is peculiar to that metal is the last-named one.‘ Is not the true explanation to be found in iron’s power of being attracted and of being magnetized by the magnet?*® In the Mediterranean lands magnetic stone is well known ;° it is said by Nicander to have been first found on Ida, which is also the spot where the Idaean Dactyls are said first to have worked iron, and since this stone is one of the principal iron ores, iron magnetized by contact with the stone must have been a familiar object to the Mediterranean peoples from the outset of their iron age. Actual evidence that the Greeks were from early days acquainted with the attractive properties of iron may perhaps be found in Homer's well-known saying, ‘avtos yap épérxetat dvdpa aidnpos,” though it would be fanciful to press the point. In historic times it is a commonplace; and the fact that magnetic force was consciously used by magicians is proved by a recipe in a papyrus which enjoins the use of a magnetic stone.® Iron is peculiar in the first instance as a new metal—a new metal found to possess these peculiar characteristics is something uncanny ;° its magnetic properties interested the Greeks and seemed supernatural.”° It is significant 1 For specimens of taboos, cf. Frazer, G.B.* III. 176, 205, 232; for instances of iron being prescribed, ib. I. 159; III. 232-6, 315; VIII. 51; Eitrem, P. Osloenses 1. 56. The Vestals used it for cutting up salt (Varro ap. Non. 222, 20. M =p. 330, Lindsay). 2 G.B.31. 159. For iron bells cf. a Central African use noted by Frazer, Folklore in the Old Testament III. 479. 3 III. 235, 230. # Bronze was often prescribed in preference to iron for making an apotropaic noise ; e.g. the bronze gong at Dodona (A. B. Cook, J.H.S., 1902, 5). It may be questioned whether iron was harder than bronze before processes of tempering were known, A. Abt’s suggestion that iron is supposed to break spells because of its general superiority to bronze (Die Apologie dev Apuleius 87) hardly goes deep enough. 5 For its being magnetized cf. Pliny, Nat. 34. 147, sola haec materia (sc. ferrum) uirus ab eo (sc. magnete) lapide accipit vetinetque longo tempore, aliud adprendens ferrum ut anulorum catena spectetur inter- dum. Our explanation has been given earlier by W. Ridgeway, Proc. Brit. Ass., 1903, 816; how far he developed it we do not know. 6 Pliny 36. 127, passim inuentuntur. Professor Pugh kindly informs us that it is found in Elba, in Piedmont, in the Austrian Tyrol. In antiquity it was found in Cantabria (Plin. N.H. 34, 148). L, de Launay, Dar.-Saglio II. 1083, rightly points out that Pliny mentions magnetite, but not as an iron ore, 7 Odyssey 16, 2943 19, 13. 8 A. D, Nock, Journ, Eg. Arch. XI. 154; also Pliny 36. 130; J. Rohr, Philol. Suppl. XVII. i. IO, 92-5. 9 It might be argued that this magical value is also due to its having been first obtained from aerolites, as is inferred from its description in the Pyramid texts as ‘the heavenly metal.’ Such objects are, however, rare in workable conditions (Roeder in Ebert’s Reallexikon dey Vorgeschichte III. 66; L. de Launay, Dar.. Saglio II. 1077b); and there is no trace of the idea in Greece, or reason to believe that the Greeks would in this point be influenced by the Egyptian idea. 10 Note the story of the iron image of Sarapis in the temple at Alexandria, made to rise through mid-air thanks to the attractive power of magnetite in the roof (W. Weber, Drei Unter- suchungen zur dgyptisch-gvriechischen Religion, Heidel- berg, 1911, 9 sqq.); Plin. NH. 34. 147 for a similar unfinished temple of Arsinoe. M. CARY anp A. D. NOCK 126 that Lucretius devotes 150 lines to proving that magnetism is a purely natural phenomenon (VI. 906-1064); St. Augustine speaks in strong terms of the impression which it made upon him. Commonly enough the magnet was held to have a soul or life, a view which is at least as old as Thales; and this implies something of the sort for iron,’ as does also the belief that iron in mines was continually replenished, more coming into being to replace what was taken out.2, Nowa substance with uncanny potencies, while generally taboo, may well be useful for certain purposes; thus contact with the ground, while forbidden in many magical operations, is ordered in others.* The peculiar properties of iron make it a powerful Justramen, perhaps the more useful because it is capable of exercising itself properties of attraction. In general it is held to be effective as being disliked by evil spirits: xow7 yap Tis Tapa avOpwrois bTOAN Wis StL vexpoi Kai Saipoves cidnpov doBodvrar ; 4 hence the Sibyl’s command to Aeneas, uaginaque eripe ferrum (as Norden explains) ; hence its common use for rings, or contra noxia medicamenta and aduersus nocturnas lymphationes, and, again, the placing of it on vessels of wine to prevent the wine from turning ;° the iron lampstand in a divination recipe in P. Lond. 121. 550 W. will keep off daemones not desired. So further east a Hindoo bridegroom in the Punjab is said to carry an iron weapon into the bridal chamber to drive away the evil spirits that haunt him.’ It is possible that the Greeks acquired these superstitions from the Hittite iron-workers in Asia Minor. Mayvns may well be connected with the Magnesia in Asia Minor rather than with that on the Greek mainland.® 1 J. Eg. Arch., lc. Orph. Argon. 360, N@ov dvridéov mdavodaiuovos; Arnob. IV. 12; Th. avdjevra... &uyvyov. Nemes. de natura hominis1, Hopfner, Griechisch.-dgyptische Offenbarungszauber, p. 40, Matthaei (referred to Posidonius by I. 202 sq., §§ 781-4. C. Reinhardt, Posidonius, 105, 350). But in 7 Frazer, Folklore in the O.T., I. 521. On carmen pavaeniticum, § 15, 114 sqq., ed. Lund- survivals in Mohammedanism of Semitic belief strom, Anecd. Byz. 1. 8 (twelfth century according in the protective virtues of iron cf. I. Goldziher, to Krumbacher? 803), the magnet éuyvxyot... Arch. X. 41 sqq. The evil spirit Werzelya in ovyxiwel Thy &Wvuxor obciav. the Sisinnius story has a name possibly derived 2 Strabo V. 2. 6, p. 224c (cf. Cumont, Ztudes from a word for iron (E. Petersen, EIZ OEOD, Syriennes, 197 sqq.); Callimach. Coma Beren., fr. 118}; what conclusions, if any, are to be drawn 35c, Schneider (=4 Mair), ya:ddev dv7é\\ovra ~=60wwee do not know. According to Manetho ap. Kaxov purdv of puv Ednvar. Plut. /s. e¢ Os. 62, p. 376B, iron was Called in 3 A. D. Nock, J.H.S., 1926, 49 sq. (cf. alsothe Egypt ‘the boneof Typhon’ (the power of evil) ; need for women as agents in some rites, their cf. Roeder in Roscher’s Lex. IV. 777. exclusion as ‘dangerous magic’ from others ; 8 A text at Boghaz-kéi, giving directions for the use of menses for some purposes, their avoid- the averting of plague, says: ‘I take a chain for ance at other times; the use of agnus castus the neck and an iron ring witha[ ] stone as a help to ritual continence, also apparently and bind them on the neck and horns of the as an aphrodisiac, as discussed by E. Fehrle, ram. This is done at night before the tents (?) R.G.V.V. VI. 139 sqq.; and perhaps to be’ with the following formula, ‘‘ The god who runs explained on the line suggested by H.J. Rose, away in anger, the god who has caused this Folklore, 1925, 322 sqq., of the working up of dying, I have bound on this ram: now content emotion and its diversion into other channels’ thyself with that. The next morning I release as a feature of ritual). it’’’ (J. Friedrich, Aus den hethitischen Schriften II. 4 Schol. Q in Od. XI. 48. [Der alte Orient XXV. 2], 11). Salmanassar I. 5 Plin. N.H. 34. 151; Eitrem, Offerritus 61, put a plate of iron in the foundations of the 9 temple he restored at Assur (Br. Meissner, 339- 6 Cf. the prayer P. Lond. 121, 1. 100, Wess., Babylonien und Assyrien I. 265); this implies méuwov po Tov adnOwiv 'Aokdnmiv dixd Twos apotropaic qualities. An On ow: hypoth out evil microbe disinfec emergir 1 For 1 smearing and duri: Hochzeit T Primitive | spiritual : tended to smell of p: cf, the use natural of the let was nd this 1 mines lat was r taboo, l, while yeculiar useful : KOLVN byras ; 4 Norden ta and of wine recipe er east nto the ossible kers in Minor 12; Th. eszauber, 321. On ic belief yldziher, zelya in derived , GEO, e drawn etho ap. ‘alled in of evil) ; ions for ‘hain for ] stone s of the tents (?) ho runs sed this content | release viften II. assar I. ; of the [eissner, implies MAGIC SPEARS 127 An iron spear had a double value, as being a spear and as being of iron. On our hypothesis, for which we cannot claim that it is more than a hypothesis, the spear drawn through the bride’s hair combed out or frightened out evil spirits; the yoke of spears drew into itself or dispelled the uncanny microbes carried by the captive enemies, and the auctioneer’s spear similarly disinfected their kit ;! the spear at the tomb of the murdered person fixed the emerging ghost or rendered it powerless.” 1 For this idea of catching ghosts, cf. the smearing of the doorposts at the time of birth and during the Anthesteria (Samter, Geburs Hochzeit Tod, 299), well explained by H. J. Rose, Primitive Culture in Greece, 113, 131, as a sort of spiritual flypaper; it may also have been in- tended to frighten away ghosts with the evil smell of pitch (C. Clemen, Neue Jahrb., 1926, 21 ; cf, the use of garlic discussed by Bertholet, Rei, M. Cary. A. D. Nock. Gesch. Gegen.* I. 315). 2 It should be remarked that nowhere is magnetized iron prescribed in the recipes con- sidered. Our case rests on the fact that iron is susceptible to magnetic influence and capable of magnetization. Weare indebted to Professor H. J. Rose and to Dr. A. B. Cook for valuable criticisms and suggestions. THE BACCHANALIAN CULT OF 186 B.C. THERE is no little division of opinion regarding the provenance of the Bacchanalian rites which were suppressed with much cruelty by the Senate? in 186 B.c. Since the Dionysiac orgies were native to Phrygia, and since Livy tells the story in question immediately after describing the immoral practices that were brought back from Asia by the returning army of Manlius Vulso in 187, it has frequently been assumed that Anatolia was the source of these rites. Reitzenstein? and Cichorius,’ in discussing a recently-found decree of Ptolemy IV. regarding an inquisition into Dionysiac rites in Egypt, both assume that the Roman cult had come from the Orient. Carcopino, in a recent dis- cussion of the ‘ Underground Basilica’* at Rome, suggests that the objection- able element of the cult was the Pythagorean club, which might become a political menace. Other scholars,> remembering Livy’s statement that one of the priests of the cult had operated in Etruria before coming to Rome, have tried to find traces of Dionysiac rites on Etruscan monuments. For the historian of Roman society who must attempt to comprehend how the mystery religions won their way in Italy, the question of provenance and route is of no little importance. In the case of Mithraism and the Syrian cults, it is now usually recognized® that people of Western stock were not to any great extent converted to the mystery religions, and that the devotees of these cults were largely natives of the East who had carried the rites with them in their migrations westward. To me it seems probable that the Bacchanalian worship also illustrates a movement of peoples rather than of a creed carried by proselytizing, and that the people in question were the captives taken in Southern Italy during the last years of the Punic War. Had the cult spread from Etruria we should expect to find some trace outside of a few objects of art testifying to the vitality of the cult in Etruria. Nor will the suggestion that it came in the train of a philosophizing group account for its rapid spread among the lower classes; for Pythagoreanism appealed only to the learned, and these were few enough in Rome in 186 B.c. Lastly, Manlius’ army can hardly be looked upon as the vehicle. Dionysiac able incident to win favour. It has been sug- gested that Cato’s oration de conjuvatione, Festus (ed. L.) 280, was delivered on this occasion. 1 Livy, 39, 8-19. 2 Archiv. Rel. XIX. 191; cf. Weissenborn- Miiller, on Livy, 39, 8, 5: ‘ Thracien und Klein- asien.’ 3 Rémische Stud. 1922, p. 21. One need not go to Ptolemaic laws to explain the acts of the Senate. The paralellism in phraseology is only such as must occur in legal documents, and Rome’s attack upon an immoral foreign cult is sufficiently explained by the fact that Cato’s nationalistic reform party was now gaining the ascendency and was making use of every favour- * La SBasilique Pythagoricienne, Paris, 1927, p. 179. Carcopino’s suggestion implies that Tarentum was the source of the Bacchic cult, and with that we must agree. & See Warde Fowler, Relig. Exp., p. 346. L. R. Taylor, Local Cults of Etruria, p. 207, finds Dionysus only in Populonia. 6 Toutain, Cultes Paiens, passim; Frank, Roman History, p. 510 sqq. cults w that re would | though these fe year or It | in 208 } upon as store of been pu hear mt arts, cu. actual i found a that the chthoni had alre he tells Dionysi: Loc and tho stances were.® out, the then fot Tarentu largely very im] usually | these cu were pr servative cities we mostly a 1 Livy, 2 2 Evans, wangler, S; interpretati supported some inscr p. 689. See Gr., p. 25. Greece, see 3 Livy, 321 Sqq. e of the Senate? ace Livy yractices Vulso in of these ecree of | assume sent dis- jection- scome a hat one ne, have 2nd how nce and » Syrian > not to otees of es with hat the than of vere the r. ne trace Etruria. 5 group reanism 186 B.C. ionysiac been sug- née, Festus $10n. is, 1927, lies that chic cult, , 346. L. 207, finds nk, Roman TENNEY FRANK 129 cults were not popular in Asia Minor as late as the second century B.c., for in that region Attis had absorbed the functions of Dionysus. Manlius’ army would hardly absorb much mysticism in a rapid raid through the country, and though captives were taken in Galatia, few were brought from Phrygia, and these few could not have spread the cult over a large part of Italy within a year or two of their return. It is my belief that the 30,000 prisoners taken in the capture of Tarentum? in 208 B.c.—some twenty years before the date in question—should be looked upon as the chief source of the Bacchic cult in Central Italy. The splendid store of votive objects that can be seen in the museum of Tarentum has never been published. When scholars become better acquainted with these we shall hear much more than we have of the influence of Magna Graecia upon Roman arts, cults, and customs. There has been some difference of opinion as to the actual interpretation of the many thousands of votive offerings which Viola found at Tarentum, but at present scholars agree with Evans’ early surmises? that the heroized dead represented upon them are closely associated with the chthonic Dionysus. That the Dionysus cult was very important at Tarentum had already been conjectured from Plato’s casual reference (Laws, 637), in which he tells that the whole city of Tarentum would give way to drunkenness at the Dionysiac festival. Locri was another city which the Romans captured during the same war, and though we are not told explicitly that captives were taken, the circum- stances of the capture and treatment of the city leave little doubt that they were.*> Furthermore, as Oldfather, following a suggestion of Schmidt, points out, the names Lucris and Lucrio in Plautus seem to refer to slaves from Locri then found at Rome and doubtless in large numbers. Now at Locri‘ as at Tarentum—not to mention other cities like Paestum and Medma—Dionysus largely displaced Hades as the associate of Persephone-Kora, and became a very important chthonic divinity in this region, where the cult of the dead was usually prominent. With Pythagorism and Orphism giving new content to these cults, there can be no doubt that mysteries and impressive initiations were practised through most of the cities of the coast. It would be a con- servative conjecture that at least a hundred thousand inhabitants of these cities were brought by the exigencies of the Punic War into Central Italy— mostly as slaves, but also in part as free refugees from the raids of Hannibal 1 Livy, 27, 15-16. 2 Evans, Jour. Hell. St., 1886,i p. 1 sqq. ; Furt- wangler, Sitz. Minch. Akad. 1897, 134 n. This interpretation has been questioned, but it is now supported by the discoveries at Locri, and by some inscriptions of Tarentum. See /.G. XIV., p. 689. See also Gianelli, Culti e Miti della Magna Gr., p. 25. On the chthonic rites of Dionysus in Greece, see Farnell, Cults, Vol. V. 3 Livy, 29, 6-9; Oldfather, Philologus, 1912, 321 sqq. # Orsi’s numerous reports of his excavations in Not. Sc. 1902, 1909, 1911, 1912, 1913, I917, and Boll, ad’ Arte, 1909; Quagliati, Ausonia, 1909, 142 sqq.; Oldfather, Pauly-Wissowa, XIII. 2 (preprint); Gianelli, Culti e Miti, 218 sqq. See also Gianelli, pp.. 45 (a temple of Dionysus at Heraclea), 82 (a temple at Metapontum), 28 (Orphic charms at Thurii and Petelia), 136 (Dionysus on coins of Laos), 146 (Kore and Dionysus at Posidonia), 253 (Kore and Dionysus at Medma). K 130 TENNEY FRANK and the Roman generals. Surely these hordes carried their most sacred creeds to their new homes. It is unfortunate that Livy’s long account of how the Bacchanalian rites were first discovered at Rome is so fantastic. The story of Aebutius, his unnatural mother and his mistress, has been pronounced a ‘romance’ by Heitland,! and with that we must unfortunately agree. The annalists of the early second century did not have space for full details of this kind. But we must not be too ready to reject the whole account. Cato’s speech delivered on the occasion seems to have survived,” and though it doubtless blackened the canvas as such speeches usually do, it preserved some accurate details. At any rate, Livy’s summary of the ‘senatus consultum’ shows a commendable resemblance to the original which we have. Livy has no explicit theory as to whence the rites had come. He tells the story immediately after describing the vicious practices that were introduced from Asia by returning soldiers, but this does not mean that he thinks Asia the source, for he implies time and again that the rites had been at Rome for several years. He first says (39, 8) that the cult had been spread in Etruria by a Greek priest and soothsayer of low station: Graecus ignobilis . . . sacri- ficulus et uates. Many of the war captives must have been purchased by plantation owners of Etruria, and whether or not the soothsayer in question had been a captive who had gained his freedom, any devotee of: Dionysus would presumably find a welcome where large groups of captives from Tarentum or Locri might be found. At Rome the rites were carried on in the quarters of the poor near the docks behind the Aventine,® and the chief priestess was said to be a Campanian woman‘ (L. 39, 13), who was assisted by two men de plebe Romana (ch. 17). After the first raid we hear no more of the affair at Rome, but two years later (39, 40) the propraetor stationed at Tarentum was engaged in suppressing magnas pastorum conjurationes et reliqguias Bacchanalium (conjuratio is the word Livy uses throughout his account of the Bacchic associa- tions). The owners of the large ranches of the south were, of course, among the purchasers of slaves captured in these wars. And finally, in 181 (L. 40, 19), the Senate ordered the praetor of Apulia to carry on the prosecutions in his province, ne serperet iterum latius. By that time it would seem the Romans had discovered the real source of the cult in Southern Italy, from which it might ‘spread once more.’ It is, therefore, clear from the localities mentioned that Magna Graecia was the source, and we remember that the fragment of the repressive edict which we possess—Dessau, J.L.S. 18—was found in Bruttium between Locri and Croton. Livy’s account is unfortunately too deeply coloured by the bitterness of his source to give us a reliable picture of the practices of the association, but at Naples (Peterson, Cults of Campania, p. 194), citizens of Campania may of course have been initiated in that city. 1 Roman Rep. I1., p. 229. 2 De Conjuratione, Festus (Lindsay), p. 280. 3 In Luco Stimulae, see Huelsen, Top. III. 171. # Since the cult of Dionysus was prominent we lea source and th could | we ne Tarent associa one wi rites v: and ref own pr of the « Livy s: poisoni misund a time we may} by the used w pederas Magna tion wit of the r since tl were ve charges devotee of the \ were nm removec WI! after th the nat of Magr where t Apulia ; were us! the war the Pur 1 Cf, E iapa Arovurc nelli, Cuiti evidence < (discussed and Beloc! ed creeds lian rites utius, his ance’ by ts of the But we delivered ylackened tails. At mendable » tells the troduced inks Asia Rome for 1 Etruria . + Sacrt- hased by question Dionysus Carentum : quarters stess was Oo men de affair at itum was shanalium C associa- e, among -. 40, 19), ms in his mans had it might oned that nt of the Bruttium erness of ation, but nid, Pp. 194), > have been THE BACCHANALIAN CULT OF 186.B.C. 131 we learn enough to see that they indicate South Italy rather than Asia as the source. Livy believed that the secret clubs at first consisted only of women and that later men were included, and finally the rule was adopted that initiates could not be over twenty years of age. If our theory of the source is correct, we need not assume that these changes were first made at Rome. In Tarentum’ Dionysus had his own temple and the Persephone cult was associated with his, while at Locri Persephone’s temple was the pre-eminent one with which the cult of Dionysus was associated. And so the chthonic rites varied from city to city. We must suppose that the slaves, freedmen, and refugees from these cities would in their new homes attempt to revive their own practices, and that the results would depend largely upon the provenance of the dominating group in each case. We need, of course, not believe all that Livy says about the stupra, the nocturnal rioting over wine, the practice of poisoning those who betrayed the rites, and all the rest. The Romans later misunderstood the nature of the Christian eucharistic meal, and that too at a time when the mystery religions were fairly well known. But in this case we may believe that the charges had some foundation if the rites were brought by the captives from Magna Graecia rather than from Phrygia. Wine was used without stint at the Dionysiac festivals of Tarentum, as Plato remarks; pederasty, as yet unknown at Rome, had not been unusual in the cities of Magna Graecia; sacred prostitution seems to be well vouched for in connec- tion with Persephone’s famous shrine at Locri,? and if the curse upon betrayal of the rites was at times impressed with human aid, we can hardly be surprised, since the initiates must have known that the perils resulting from betrayal were very real at Rome, especially to slaves and freedmen. In a word, the charges, even if they are overstated, point to south Italian Greeks as the devotees. That the practices should have combined the vicious tendencies of the various cults of the various cities is not surprising, since the devotees were now to a great extent in miserable slavery in a foreign country and removed from the restraining influence of their old social and political ties. When, therefore, we consider the date of the inquisition—twenty years after the arrival of the large hordes of captives from Tarentum and Locri— the nature of the practices—which resemble, though in aggravated form, those of Magna Graecia rather than those of Greece and Asia—and finally the places where the practices were found and suppressed—Tarentum, the ranches of Apulia and the estates of Etruria, Campania, and Rome, where war captives were usually sold—there can be little doubt that the devotees were in the main the war captives and refugees that came northward during the later years of the Punic War. The severity of the suppressive measures is partly explained 1 Cf, Evans, J.H.S. 1886; the inscriptions numerous at Rome it would help to account for iapa Acoviow in Ins, Gr. XIV., p. 689; and Gia- the unusual prominence of women in Livy’s story nelli, Cuéti e Miti, p. 25sqq. Thereisalsosome of the Bacchiccult. For women in the Dionysiac evidence of matriarchy at Locri, Polyb. XII. 5 cult in Greece, see Farnell, V., p. 160. (discussed by De Sanctis, Mon. Ant. 1914, p. 688, 2 Oldfather, Pauly-Wiss. XIII. 2, 31, Perhaps and Beloch, 12, 1, 84). If Locrian captives were at Temesa also; cf. Gianelli, p. 272. 132 THE BACCHANALIAN CULT OF 186 B.C. by the new dominance of the nationalists, who were eager to make a case against foreign influences, and partly by the fact that so large a number of the devotees were Greek slaves and non-citizens who had no standing in Roman courts. We have no right to assume that Roman citizens had to any extent been converted to those wholly non-Roman mysteries. It is more than probable that the strange discovery of the Pythagorean books five years later near the supposed tomb of Numa should be associated with this event. No one now believes that the tomb was Numa’s or that the books had been used by him. Indeed, a careful reading of Livy’s account* reveals the fact that he did not believe it (‘recentissima specie’). There were two stone boxes sealed with lead found below the Janiculan by the gardener digging in the plot of a scriba—presumably a freedman. The inscriptions on the boxes indicated that one contained the body of Numa and the other his books. The first was empty; the second contained some rolls of pontifical rules and seven rolls of Pythagorean writings, all on papyrus, according to Pliny. The probability is that some Greek refugee from Tarentum had brought his Pythagorean books with him, and that when in 186 the Senate began its inquisition into the foreign rites and punished the leading devotees of Dionysus with imprisonment and death, the owner of the books hid them in a manner that might save him and possibly his books as well. Numa’s name would presumably be respected as sufficiently Roman even by Catonian bigots. The books were burned, and the scribe escaped punishment, but he did not dare acknowledge ownership by accepting the indemnity that the Senate was ready to offer. Finally, it was long ago suggested? that the Ava Ditis in Tarento, where the rites of the saecular celebration were performed, must, because of the name, be traced to the Greek Tarentum. The date of the introduction of the cult was 249 B.c. The accounts are too unsatisfactory to permit of explanation. I only wish to suggest that this date is about twenty-three years after the first capture of Tarentum, when many captives (including the poet Livius Andronicus) were brought to Rome. Perhaps captive Tarentines like Livius were on this occasion also the vehicles of a new cult. TENNEY FRANK. 1] follow Livy’s account (XL. 29), which the first capture of Tarentum in 272 B.c. Since seems to be based upou Piso’s Annals, Pliny this cult was in charge of a magistrate cum im- (XIII. 84), who quotes some phrases from Cas- perio, it is likely that Rome had first introduced sius Hemina, differs in some minor details. See it for the benefit of her Tarentine captives (a also Varro in St. Aug. de Civ. Dei, 7, 34; Val. prudential measure which we often find in the Max. 1, 1, 12: Plut. Numa, 12. early days), and had placed it under strict military 2 See bibliography in Wissowa, Rel. und Kult?, _ supervision to prevent orgiastic practices. In the p. 309 sqq. Bona dea is generally identified following century Rome was less generous in the with Dameia of Tarentum, and Wissowa(P.W. treatment of the cults of her war captives, III. 690) assumes that her cult came to Rome at { Fr. 4 (D TOU’ TOS aU Kal “pol wuyny é KATET KEL TavrToia, pev ovr | avXa Kal SIM have bee interpret passages 1827, pp He 1 refer to. EADY } idéas’ WC p- 35, Il. KaTacTac diaxoo un he would Zelle refers pr the refere Gom worlds. Burr innumera of the in goras am cannot be be conter innumera But | Anaxagor as Burnet explaining ‘a case r of the Roman y extent agorean sociated that the .ccount* sre were rardener ‘ions on ther his ontifical ‘ding to brought egan its ionysus manner > would s. The iot dare is ready , where e name, the cult ination. the first ronicus) on this NK. >. Since e cum im- itroduced iptives (a nd in the t military 2s. In the dus in the es. THE HOMOIOMERIES OF ANAXAGORAS. Fr. 4 (Diels, Fragmente der Vorsokrattker) : Tovtwv é obtws éyovTwv ypr Soxeiv éveivat TOAAG TE Kal TavToia év Tact Tois cuyKpivopuévots Kal oTépuata TavTwv ypnudTwy Kal idéas TavToias éxovTa Kat “pods Kat ndovas. Kal avOpa@rovs te cupTayiva Kal Ta adda Coa boa wuxyny exer. Kal trois ye avOpmmoow elvat nal modes cuveKnpévas Kal Epya KaTecKevacpéva, @oTEep Tap’ Huiv, Kal THY Yhv avToiot PvEew TOAAG TE Kai TAVTOLA, MY EKELVOL TA OVHLOTA TUVEVEyKapEVOL cis THY OlKnoLWW Yp@vTat. TavTa pev ovv por AéAEKTAL TEpl THS ATroKpictos, STL OvK av Tap Hyuiv pwovoy atoxpLOein, ara Kal adXAn. SIMPLICIUS is very much puzzled by this passage as other commentators have been after him. He discusses it in several places, and after giving his interpretation says: dAAad TavTa per cite ob Tws eiTe GAXwS Exet, CnTeiv aEsov (see passages collected in E. Schaubach, A naxagorae Clazomentt Fragmenta, Lipsiae, 1827, pp. 89 sqq.). He is definite in his negative conclusions. He says the passage does not refer to romovs adXous THs ys, on the ground that ‘#Asov avroiow elvar Kai cednvnv Kal Ta GdXa woTrep Trap’ Huiv’ and ‘ orépyata Sé wavTev Ypnuatav Kai idéas’ would be inappropriate (see H. Diels, Simplicit in Arist. Phys. Comm., p- 35, ll. g sqq.). Again it is not said car’ ddXas Tas oiKHoEs Opoias OvaNS Kataotacews TH Tap’ jpiv (id., p. 57, 1. 20). Nor is it about an aic@nrn divaxoopnots preceding ours in time, T@ ypov@ tavTns Tponynoapévn. For then he would have said éypa@vro and not xpavraz (ll. 18 sqq.). Zeller (English trans., Vol. II., p. 359) says it is an obscure passage and refers probably to a part of our world, most probably the moon. He denies the reference to other worlds. Gomperz and Windelband-Goedeckemeyer refer the fragment to endless worlds. Burnet (£.G.P., third edition, pp. 269-270) also takes it as describing innumerable worlds and points out the improbability of its being a description of the inhabitants of the moon. ‘ Aetios,’ he says, ‘certainly includes Anaxa- goras among those who held there was only one world; but this testimony cannot be considered of the same weight as that of the fragments.’ We might be content to set aside Aetios’ testimony if this fragment clearly spoke about innumerable worlds. : But (1) coming as it does at the beginning, or quite near the beginning, of Anaxagoras’ book, a reference in it to other worlds would not be quite in place, as Burnet says, but on the contrary would be rather sudden. Anaxagoras is explaining in general what is meant by wdvta év mavti. The theory of 134 P. LEON innumerable worlds is not so obviously connected with this, whether as ground or conclusion, that it can be introduced without further preliminaries. We should expect rather to find it mentioned after the general treatment, when Anaxagoras comes to the more detailed cosmogony. (2) It is curious that Simplicius, who is anxious to consider and refute all interpretations of this passage other than his own, should not mention the theory of other worlds. On the contrary (ib., p. 178, 1. 25), along with Empedocles he places him amongst those dco: éva Tov koopov daci. (3) The quotation in Simplicius is given as one continuous passage and accepted as such by Diels and Burnet. Whether we punctuate with a full-stop after xpoas xai ndovas as Diels does in Fragmente der Vorsokratiker, or with a comma as he does in Simplicius, it is obvious that cal dvOp@ovs te cuprraynvat, etc., must go closely together with the preceding sentence, in sense as it does in construction. Now if the meaning is ‘we must suppose that there are contained many things, and of all sorts, in the things that are uniting, seeds of all things, with all sorts of shapes and colours and savours, and that men have been formed in them (in other worlds ?), and that these men possess inhabited cities and cultivated fields as with us,’ etc., there must be either a saltus in the thought or an omission in the quotation. Anaxagoras is speaking of what we must think exists in the things that are uniting, i.e. the cold, the wet, the dry, or the small particles of these or anything, and not of what exists in other worlds. He might indeed be saying ‘and thus men were formed’ (i.e. because Ta cuyxptvoueva contain seeds of everything). So Diels translates, ‘ Und dass sich so auch Menschen zusammenfiigen,’ the so not being in the Greek. But then the expression would be very elliptical and there would be little point in what follows about the civilization and agriculture, etc., of these men, and certainly no point in ®c7ep tap’ nyiv. Interpreting the passage as one, we must take cuprayjvar like Soxetv which precedes it and the other infinitives which follow it with év mace tots ovyxpivopévots. In all entities that combine to form things (i.e. in all entities that on some theory other than that of Anaxagoras may be called ‘simples’ or ‘elements’) there must be o7épyara of all things, etc., and there must have been formed therein men having cities, a sun and moon, men carrying on agriculture, etc., as we do. The xdopos is reflected with all its details in everyone of its parts, aspects, or attributes. Since in the xoowos there are men who have cities and carry on agriculture, and since there is in it a moon and sun and stars, all these details and all this organization will be found in miniature in a drop of rain, in a piece of bread, in the air, etc. Before the separation and work of organization took place (piv 5€ atoxpiOjvat tadra) the whole was an indeterminate mass and there was no single determinate thing. But once voids has performed its organization, the latter is pervasive, going right through in every conceivable part ad infinitum. It is present not merely on the surface, in that which we perceive as a complicated world of men, cities, etc., but also in that which to our perception appears merely as a grain of sand, a drop What th are to en Un) EOTL, is such present, in his Sf Thu main pri ever, aS and app. YyVv auto THY OlKN It is he write termino] goras th TO CUpT diaKpice under d Kal els T (3) he s merely » Demokr of Anax this par: This is, extendi! of any percepti rate not whole s' presents incompl mind, tl say, the aicOnrn relation meanin; presenc howeve: or that useless. It 1 ; ground 2s. We it, when sfute all tion the ig with age and ull-stop ‘with a raynvat, it does ere are seeds of en have habited s in the vhat we he dry, n other because nd dass « But oint in 2n, and Soxeip ot TOES entities les’ or st have ing on ails in re men yn and und in re the ra) the thing. going merely cities, ' sand, THE HOMOIOMERIES OF ANAXAGORAS 135 a drop of water, etc. (ov« dv map’ jpiv povov amoxpiOein, GAXA Kai GAX7). What the whole is, that ‘the flower on the crannied wall’ is. Indeed if we are to emphasize fr. 12 (Diels) «ai omota Euedrev EvecOar Kai orroia Hy, dooa viv wn EoTL, Kal oTrola éoTt, TavTa Suexdapuyoe voids, the universe as a series of events is such that every event past, present, or future contains every other past, present, or future, an idea analogous to that worked out by Professor Alexander in his Space, Time, and Deity. Thus Anaxagoras at the beginning of his work, as is natural, gives us the main principle of his theory in an emphatic and uncompromising form, howso- ever, as again is natural with most people; he may afterwards modify, hedge, and approximate to others. The otherwise quaint insistence on detail «al rnv yiv avtoiot pvew TOANG Te Kal TavToia, dv éxeivot TA OVHLETA GUVEVEYKApLEVOL ELS THY oiKnoWw XpavTat, is, of course, appropriate as a vivid illustration of an idea. It is in some such way that Simplicius seems to take the passage, though he writes under the influence of contemporary ideas and in contemporary terminology. On p. 34, ll. 18 sqq. (Diels, Simplicius) he attributes to Anaxa- goras three moments: (1) vontn évwors or the original undifferentiated state of To oupmav which he identifies with the Parmenidean One; (2) the voepa didxpiots Or suvovaiwars (elsewhere dsaxdcpnors) to which he says the passage under discussion refers; (3) the aicOn7?) ctpmvora and % é« Tav adTav yéveots kal eis Ta avTA avddvars, of the perceptible world (the wap’ jyiv of Anaxagoras), (3) he says is made like(2)ddwyoiwra. If by voepa dvaxpiors Simplicius means merely that which is not perceived but is inferred (as e.g. the atoms of Demokritos or the Parmenidean One), then his is an adequate interpretation of Anaxagoras. He is anxious to emphasize (p. 157, 1. 18; p. 461, 1. 12) that this parallel d:dxprois or Suaxdopnocs is not a happening preceding ours in time. This is, of course, true. A system of events, e.g. the Peloponnesian War extending over several years, is somehow contained in any and every moment of any and every year (this in the voepa didxpiows corresponding to the perceptible system) ; but it is not preceded by its moments or parts, or at any rate not in the same way in which the whole system is preceded by another whole system, e.g. by the Persian War. In other words, every series of events presents itself to perception as extending over a length of time and as incomplete at any one moment. But actually and as apprehended by the mind, the whole series is present as a whole at anysingle moment. That is to say, the phenomenal temporal series has a noumenal timeless duplicate, the aicOnr? Staxoopnors has a voepa Siaxoopnors which is od« éyxpovos, and the relation between the two is, of course, not one of time. If this is Simplicius’ meaning, it is a correct deduction from the Anaxagorean principle of the presence of the «dcyos in any of its parts, when applied to events. (I do not, however, mean that Anaxagoras himself probably drew this conclusion clearly or that Simplicius had his ipsissima verba. The testimony is, however, not useless. ) It is an idea like this that must have suggested Lucretius’ otherwise too 136 P. LEON surprising veductio ad absurdum, I. 915 sqq. If, he says, that from which things are derived must be of like nature to the derived, then— fiet uti risu tremulo concussa cachinnent et lacrimis salsis umectent ora genasque. And in II. 978 (Giussani’s edition) he adds: multaque de rerum mixtura dicere callent et sibi proporro quae sint primordia quaerunt. We must suppose that the atoms laugh and weep and discuss the atomic theory. Anaxagoras does say something very much like this, but with this difference: Anaxagoras would argue—What you call an atom or an element (in his own system there is no room for these) does not indeed weep and laugh and philosophize. For vois is not present in everything (fr. 11). But in this so-called atom or element are to be found men or intelligent beings who do weep and laugh and philosophize. The general impression one gets of ancient satire, whether in Aristophanes, Plato, or Horace, is that it keeps close to fact. Lucretius would be doing this if we give the meaning suggested to fr. 4. Otherwise he seems to be taking rather a surprising leap. This interpretation will also explain more clearly what Anaxagoras meant by his theory of ‘ homoiomeries,’ whether he used that word or not—the theory that the part is like the whole, or that the whole is present in the part, and that everything is in everything. That theory generally suffers in respect of clearness through its being explained in terms of elements, or at any rate under the influence of the theory of elementary or simple substances. Aristotle emphasizes in his illustrations organic substances. But it is agreed generally that this selection is due to Aristotle’s own interests and is not to be attributed to Anaxagoras (cf. Windelband-Goedeckemeyer, p. 61). Simplicius (p. 174, ll. 19 sqq.; p. 176, 1. 27; p. 178, ll. 30 sqq.) tries to state the theory in relation to the opposites. Anaxagoras, he says, does not derive everything from everything absolutely, but only the opposites from each other. But this statement, it must be admitted, is made tentatively. Elsewhere he says the derivation is not of everything from everything ov« duéows adda Kata taEw (p. 460). Burnet (Z.G.P., pp. 263 sqq.) rejects the interpretation that ‘ wheat, for instance, contains small particles of flesh, blood, bones, and the like.’ According to him Anaxagoras means that any particle of matter, howsoever small, contains all the ‘ opposites.’ But this would not be a very great advance on the theory of Heracleitos, and obviously Anaxagoras did not make merely a return. Again, are these opposites simple? If so, what of the assertion that everything contains everything? Better to say with Simplicius that the opposites not merely inhere together, but contain each other. Is the number of these opposites finite? If it is, what of the assertion that the homoiomeries are infinite? From Ar. Phys. 187a 25 sqq. and Simplicius’ lengthy com- mentary thereon, one gathers that Anaxagoras meant more than that matter is divisible finite nu that the | simple. in the m: of matte left indet number c abstractic others we This cer Tannery between to him a it than w this distir much mo evolving. superfluot is because simple, pt substance really mo Arist whole in , of divisio whole. fF into part: Anaxagor: geneous s the whole while oth Aristotle’s that Anax there is or vous. Th nothing bi Only of m each othe with thing every part yet so tha the whole vous) ovdév which atomic ‘th this ement laugh n this ho do ncient 0 fact. fr. 4. meant sheory t, and espect y rate istotle erally buted - 174, ry in thing t this s the Taew theat, like.’ oever vance erely ‘rtion it the mber leries com- ter is THE HOMOIOMERIES OF ANAXAGORAS 137 divisible into an infinite number of particles, each of which contains all of a finite number of qualities. But perhaps Burnet does not intend to suggest that the number of the opposites is finite, or that each opposite is ultimate or simple. P. Tannery (Pour l’Histoire de la Science Helléne), with whom Burnet in the main agrees, maintains that, according to Anaxagoras, the constituents of matter are ‘des qualités élémentaires,’ the number of which Anaxagoras left indeterminate, as indeed it is (p. 287). Further (p. 285), he says the number of qualities may be infinite, and any we should choose would be pure abstractions. This I take to mean that to select some as more primitive than others would be adopting an arbitrary, but perhaps scientifically useful, method. This certainly does more justice to the idea of Anaxagoras. But although Tannery admits (p. 288) that Anaxagoras did not establish a clear distinction between quality and substance, yet by his interpretation he is attributing to him a firmer grasp of this distinction and a greater preoccupation with it than were possible at tlie time. Anaxagoras did indeed help the making of this distinction, but only by turning against the trend of thought (represented much more by Empedocles than by himself) from which this distinction was evolving. In his own system the distinction between quality and substance is superfluous. (He made use of the opposites in illustrating his theory, but that is because he had to use current conceptions.) Soin the main is that between simple, pure, or elementary substances, and complex or compound or composite substances. He even seems to have held that what is classified as simple is really more complex. Aristotle illustrates and criticizes the ‘homoiomerous’ relation of part to whole in Anaxagoras’ system from his own point of view, which is a recognition of division of the whole into (a) parts like the whole, (b) parts unlike the whole. Flesh is divided into parts which are flesh, but the heart is divided into parts which are not hearts. We may from this get the idea that Anaxagoras also made this distinction, and recognized certain pure or homo- geneous substances, the peculiarity of which is that their parts are always like the whole and presumably like each other—e.g. particles of gold are all gold— while other substances—e.g. a man, a house—he held were different. But Aristotle’s criticism (Phys. 188a 13 sqq.) and Simplicius’ commentary show that Anaxagoras neglected or did not admit this distinction. For Anaxagoras there is only one thing corresponding to what Aristotle calls ouoesdys—namely, vous. The latter is always nothing but voids. Hence parts of it are always nothing but voids. Fr. 12 voids Sé was Guords date wal o peifwv Kai o éXaTTOY. Only of mind can it be said that the parts are always like the whole and like each other, and this constitutes its uniqueness. Everything else is on a line with things like a man, a house, etc., not oyuoesdys. The whole is indeed in every part and any part in any other in some way (possibly like a plan), but yet so that every part remains different from every other and presumably from the whole. This we may gather from fr. 12, érepoy 5é (anything other than vous) ovdév éotiv Sworoy ovdevi, and fr. 4: in the primordial state of the cvyav i bie | 138 P. LEON or whole there were orepuadtwv areipwv mrAR00s ovdév eouxdTwv arAdAHAoLG. OVdE yap Tov adXwv ovdev Eoixe TO Erepov Te Erépw. Some of these ovéppata must have been such as go to the making of one substance or one quality. Yet they did not resemble each other. Of the particles of gold we cannot say that they are just gold, for the simple reason that gold itself is not just gold, but an infinite number of other things also. If we wish to understand the ‘ homoiomereity ’ of Anaxagoras we must take not a whole, a substance, or a quality, but the whole, 7o cvurayv, the koopos, as Anaxagoras does in the passage we are discussing, where he takes the world with all its organized detail of men and cities, etc., as Aristotle does in Phys. I. 4. 203a 19, and Simplicius in his commentary, p. 460, especially ll. 26 sqq. Every part of the cosmos (whether we call it part, substance, quality, event, aspect, collection, any entity other than the ovyay) is homoiomerous to this cosmos, being itself a cosmos, and somehow containing the detail and organization of the xdopos and of every other part. otdyv Tov popiwv .. . peiywa opuoiws T@ Tayi (Aristotle). Further, the same can be said of a part of a part of a part, ad infinitum, as is pointed out by Simplicius, p- 172, ll. 20 sqq. This, and only this, is what Anaxagoras meant by the ouotoueph or what others called ouovoyepj. And the fragment we are discussing is intended, I think, to make this theory clear. We may conceive the latter as a direct riposte to Empedocles. ‘ Why,’ Anaxagoras may have asked, ‘fix on some things as more elementary than others? and why only on four? Everything is like the cosmos itself, and nothing is simple but vods. Further, admitting the partial use of your classification of things in an order from the simpler to the less simple, what you call simpler is really more complex. The simplest thing (next to vods) is really the complete xocyos, for nothing else arises out of it, and therefore it need not be anything other than itself. But from your fire, air, earth, and water arise plants, animals, men, etc. Since being cannot arise from non- being, water, for example, must therefore have not only its own being, but the being of a man, a plant, an animal.’ This is indicated by Ar. Phys. T. 3. 302a 28 sqq., dépa Sé xal wip peiypata TovTwy Kal TOV ado oTepudtav wavTwy* elvar yap éExdtepov aitav é€ dopdTwy omolopep@v TavT@V 4 nOpocpévov. S10 Kal yiyveoOat wav’ éx TovTeY. We may illustrate this from a similar objection made in modern biology : f ‘ ... If there is a real mechanism of heredity, the machine must, upon the theory of biological evolution, have been most intricate in the earliest living cells.... Now if C, and C, be variants born of B, and if C, produce variants D, and D,, and C, produce variants D, and D,, not only must C, and | C, each have contained mechanical provision for two variants, but B must have contained such provision for four; so that the further we recede the more complex must be the heredity-determining machinery in the organisms ; we reach (H. W. B. Joseph, The Concept of Evolution, pp. 21, 22). We may with the help of the present passage also explain the saying about tl (sc. TO peyarou KOO LoS) contains equal to the whc contains It 1 Anaxagc Aristotle héyovTas Socrates of nature grateful. Kat Kabe mind as: Plato us states. sometim: Yet they not his \ to have compoun Mind, o1 avUTOKpaT: QuTov 6v7 dua Wave part of it. principle. mistaken it is not n according which is \ mind for One and the whole infinite nt by the pre Parmenid His | already in He p Democritt DUS. OVOE Ta must Yet they nat they , but an ve must trav, the he takes tle does specially bstance, yrav) 1S ntaining \OUY TOV can be nplicius, - by the scussing ‘Why,’ ury than self, and of your le, what ) vous) 1S -herefore irth, and »m non- n being, | by Ar. yp addX\ov | TavT@v biology : f ipon the st living variants C, and | B must cede the rganisms e saying THE HOMOIOMERIES OF ANAXAGORAS 139 about the numerical equality of the small and the great. Fr. 3, cai icov éotl (sc. TO péya) TO outxp@ TAHBos. Fr. 6, wai Ste S€ ioas potpai eiot tod Te peyadrou Kal tod cuixpod TARVOS, Kai ovTws av ein ev Twavtl mdvta. If the Koopos contains 2a, 3b, 4c, etc., then any of its parts, however small, also contains 2a, 3b, 4c, though on a reduced scale. Thus the part is numerically equal to the whole. The smallest part contains the infinite whole, and just as the whole contains an infinite number of parts, so the smallest part too contains an infinite number of parts. It is perhaps permissible to make here a few remarks on the value of Anaxagoras’ contribution and on its place in the history of Greek philosophy. Aristotle’s tribute to his recognition of voids, ‘ olov vydwv épavn tap’ eixa héyovtas Tovs mpotepov,’ may be heartily endorsed. That he did not, as Socrates or Plato would have had him do, drag in vods to explain the details of nature, we may with Gomperz consider something for which we should be grateful. The description in fr. 12 of voids as AewTOTaTOV Te TavTwY ypnudTeV kat kaBapwtatoy is not enough to make us charge him with having conceived mind as something material. Plans, Adyor, a Aoyiorns, are spoken of as AéwTor. Plato uses AewTa@s peptuvav. Kalapds is, of course, a word applied to mental states. Besides, as has been pointed out, Piato and Aristotle themselves sometimes slip into language of the crassest materialism when talking of mind. Yet they had surely a grasp of the nature of the immaterial. If we examine, not his words, but the réle allotted to voids in his system, we must allow him to have had a quite creditable conception of mind. Everything else is a compound, a piyyua, like the whole, and is a reproduction of it in miniature. Mind, on the other hand, is alone dmdois, dysyns, xaOapos (Arist.). It is avToxpatwp and orders everything. Plato’s Cratylus 413c, avtoxpdtopa yap avTov dvTa Kai ovdevi peperypévov Tavta dnoiv avtov Kocpeiy Ta Tpdypata dia TavTwy tovra, is a fair summary of fr. 12. Mind is in the xcoopos, but not part of it. It is the principle of its organization, and is an active and intelligent principle. Such are the attributes allotted to it that it cannot possibly be mistaken for matter. What, after all, can one say about mind but that it is not matter, showing also what its relation to the latter is? Its function, according to Anaxagoras, is to be the principle of organization in a xdopos which is wholly present in an infinity of parts, and which therefore demands a mind for this infinite complexity. In the question as to the relation of the One and the Many, Anaxagoras recognized the claims of the latter by making the whole consist of an infinity of parts, each part also consisting of an infinite number of parts. The Parmenidean One is represented in his system by the presence of the whole in every part and by voids, which, like the One of Parmenides, is absolutely homogeneous (vois 5é was duotos, fr. 12). His probable relation to Empedocles, that of contradiction, we have already indicated. He probably influenced the atomic theory at least in its elaboration by Democritus. In that theory the constituent parts of the xooyos, the atoms, 140 P. LEON are themselves xooyot, like the homoiomeries of Anaxagoras. Only, perhaps through reaction against Anaxagoras, instead of possessing all the qualities which are present in the world of perception, they are deprived of most of them. Also the atomists, with Anaximander, believed in the existence of innumerable worlds besides our own. Anaxagoras’ multiplication of the «oopos in its parts is of course different from this. Anaxagoras’ bold and uncompromising conception of an infinite whole, present as a whole in an infinity of parts, in a part of a part of a part, ad infinitum, cannot but have had something to do with the difficulties raised by Zeno and Melissus about the One and the Many: infinite magnitude and infinite divisibility. One cannot help being tempted to find traces of the effects of this same idea, through Gorgias, in Antisthenes’ denial of the possibility of any judgment other than identical judgment. According to Anaxagoras’ theory we can in some sense say that everything is everything else. Water is not merely water, it is also a plant, a man, etc., anything which arises from it. A natural reaction against this is to maintain that ‘ water is just water’ is the only judgment we can make about it. We can renew the attempt made by Schleiermacher (Geschichte der Philosophie I., p. 104) and by Tannery (Pour l’Histoire de la Sctence Helléene, p- 291) to find in Anaxagoras a source for the theory of Ideas, but on somewhat different grounds from those urged by them. The problem of the presence of an infinite whole in an infinity of parts, even if the whole is material, is par excellence the problem of the universal, and was bound soon to lead to metaphysical and logical enquiries, and away from the merely ‘ physical’ philosophy. But even in details we can see a remarkable similarity between Plato and Anaxagoras. The latter presents us with an infinite multiplication apprehensible by the mind only, and not by the senses, of every entity. The whole is multiplied in every part and every part in every other. Now it is precisely such a multiplication of the perceptible in a conceptual region that Aristotle attributes to the ‘friends of the ideas.’ Met. ggob 1, Enrovvres Tavde Tav bvTwY AaBely Tas aitias Erepa TovTaIs ica Tov apiOpov Eexomicay, adotep el Tis apLOunoat BovrAopevos, éXaTTOVvwY pev yTwV oloTO pwndéev SuYnoEerOaL, Treiw 5é tmoinoas apiOuoin. Tannery considers the edn as though they were always qualities. But there are also eién of substances, of horse, man, etc. Just so in Anaxagoras there is multiplication of substances as well as qualities. Of course it is only a partial resemblance. In Anaxagoras’ system there must be a multiplication of every rode, whereas the eidos, the tovdvde, is not the rode. But it can be argued that every rode is potentially a universal (we speak of a Napoleon), and in any case the resemblance is striking enough. Another resemblance we may find in what Aristotle tells us of the Platonic TO éya Kal TO usxpov which is identified with ywpa, and is the basic constituent of the material. The name at least cannot but recall Anaxagoras’ material universe, of which he says one of the fundamental characteristics is that OUTE . petyadXou Ana culminat for a si simplicit logical : transforr about th: other ha position to whict parts as to subor there are To call unworthy) UNIVE! perhaps qualities most of tence of @ KOT LOS » whole, a part, Ss raised ude and is same idgment can in merely natural he only hte der Helléne, mewhat yresence erial, is lead to hysical ’ yetween lication y. The Now it region ITOUVTES OMLOAY, rer Oan, ey were an, etc. ialities. re must he ToOde. ak of a -latonic stituent 1aterial is that THE HOMOIOMERIES OF ANAXAGORAS 141 OUTE . . . TOU GyuKpOv eoTL TO ye EAXdYLoTOY, GAN’ EXacaoY de>... Kal TOD peydrou ae éore peifov* Kai cov éoti TO cpexp@ TAOS ° Anaxagoras does indeed, as he has been said to do, represent the culminating point of the enquiry into the one ¢voi. That simple enquiry for a simple unity becomes curiously complex, just because of the very simplicity and the thorough-going and uncompromising nature of Anaxagoras’ logical mind. It has with him reached a stage where it must become transformed and pass on the one hand into logic in Plato, into the enquiry about the nature of predication through Gorgias and Antisthenes, and on the other hand into metaphysics, the theory of ideas, also in Plato. This central position of Anaxagoras is made clear by the passage discussed, according to which, I think, in considering the ‘homoiomeries,’ we should look upon parts as ‘homoiomerous’ primarily to the whole «écyos, and only secondarily to subordinate wholes. Indeed, it is implied in Anaxagoras’ principle that there are only two entities which are properly wholes, the «oopos and vois. To call anything else a whole is more or less arbitrary, a principle not unworthy of the most thorough-going of modern absolutists. P. LEON. UNIVERSITY COLLEGE, LEICESTER. THE ATHENIAN HOPLITE FORCE IN 431 B.C. , ‘ ~ Py ya > , e . ‘ , , Xpnuace pev ovv ovTws EOdpavvey avTovs: oTAiTas Sé TpLaXLALOUS Kal “UpioUsS ? bd “ > a , \ “ >, ¥ e , \ , elvat avev Tay év Tois dpoupios Kai T@Y Trap’ éTarkiv éEaxioxiNiwy Kai pupior. a 4 > 4 \ A e / e / > 4 > , A TocovToL yap épvAaccoY TO TPwTOV oOTFOTE Of TrOAgpLoL EoBdroLeY, ATO TE TOY mpecBuratwv Kal Tov vewTadTwY Kal peToixwy Soo. oTAiTaL Hoav. Tov TE yap Darnpixod teiyous otddvot Hoav mévte Kal TpidKovTa, KT... . imméas é / amépaive Staxociovs Kal xtrdiovs Edw immotokdtais, éEaxociovs S€ Kal xLdious rofotas, Kal Tpinpets TAS TAWipwoUs Tplaxocias. Tadta yap iTIpxev ’AOnvaiois \ > 4 ¢ , e e > 4 \ A y , Kal ovn é\doow Exacta TovTwY, te 7 éoBorn TO TpwTov Euerre IleXoTTOvYNCiwY ésecOar kai és Tov Todemov KabioctravTo.—Thuc. II. 13. 6-9 e . 7 > , 4 € \ ~ 7 \ oi ’"A@nvaio. améotetkay tas éxatov vais tepi IleXomévyncov . . . Kai , c , — > «A \ , , ViAious oTALTas ET AUTa@Y Kal ToEOTAaS TeTPAaKOTIOUS.—23. 2. ’AOnvaios twravinpet, adroit Kal oi pérouxor, éoéBarov és thv Meyapida... \ e \ / > n = a e \ \ e yf Kal-ot twept IleXotrovynacov ’AOnvaio oi év Tais éxatov vavoiv ... ws HaGovTo \ > fol , sa] > / yw v b > \ \ Tovs €x THS TWodews TavotpaTia év Meydpois évtas, Errevoay trap’ avTovs Kal / fa /, 56 , 57; A 10 Ul "AO , S Ff EvveyeiyOnoav. otparotredov te péytotov bn TovTO aOpoov ’AOnvaiwy éyéveTo... , 4 e ww > / 9 > \ e "A@ A \ 5’ ’ ~ et > pupiwv yap oTAtTa@V ovK éEXdaaous Hoav avTol ot AOnvaio: (xwpis 8 avtois o1 év , , * t \ , ’ o. / , Tloreéata tpioxidior Hoav), pétoxor S€ EvveréBadov ove éXdaaovs TpicxtNiwv oTAiTaV, Ywpis 5é 6 GAXOS Gutrdos Yida@v ovK OAL'yOs.—31. I-2. THERE is still something to be said about these figures for the Athenian hoplite force, the more so as the most reasonable discussion of them, Meyer’s, is spoilt by some unsound inferences and has in consequence not found support. Their difficulty is apparent: a muster mavdquet in 338 meant calling up all classes up to the age of 50 (Lycurg. c. Leocr. 39), and since Socrates fought at Delion and Amphipolis when he was in his late forties, and not at Mantineia when he was over 50, we may assume that it meant the same in the fifth century ; we also assume (though this is by no means proved) that ‘the youngest’ are those in their nineteenth and twentieth years, as in the fourth century, certainly after the reform of the Ephebeia, perhaps earlier (Aeschin. II. 167) ; military service ceased at 60. But how could the number of men in these twelve classes, 19-20 and 51-60, stand in the proportion of 13 : 17 (16,000 less 3,000 metics, 13,000 plus 1,000 cavalry and 3,000 metics) to the men between 21 and 50? They could not be more than a third, and might be less. (Beloch, 1923, tries to make the problem more difficult by the arbitrary assumption that the 13,000 citizen hoplites are all the men of 21-60, instead of 21-50. Meyer, who accepts Thucydides’ figures, forgets that there 1 Beloch, Bevilkerung (1885), 60 sqq.; Gr. Geschichte III. 2. (1904) 878 sqq. Fawcus, Aufgebote, Klio V. (1905) 341 sqq.; Gr. Geschichte J.H.S. XXIX. (1909) 23 sqq., is very unsys- III. 2. (1923) 386 sqq. Meyer, Wehrkraft, etc., tematic. in Forschungen (1899) II. 149-195. Busolt, Gr. must ha into Mes suffered | classes suffered as cleru younger In 1 making 1 would th and 50, gave this conseque editor, v statemen the solut It is if one mi a margil (* 13,000 figure for But, (Diod. X early in Thuc. I. ExT@ ETEL in Thucy probable things w appearan affecting last degre and a prc own day, Athens of for his ar the reserv for this is 1 It is un Wilamowitz Thucydides tion that old That a man Suggest this Organized st ypaumareia, ; C. , 1 uptous | puplov. TE TOV Te yap , \ rméas O€ , 9 D ' yedLous LAnvaious ovynotwv \ - Kat Nea naOovto ITOUS Kal EVETO .«. 7 ¢ , ots OL ev toxiriov Athenian Meyer's, ot found 8 meant nd since ‘ties, and the same ved) that is in the s earlier » number ortion of 1etics) to uird, and It by the of 21-60, hat there Fawcus, yery unsys- A. W. GOMME 143 must have been ‘ oldest and youngest’ metics, besides the 3,000 who marched into Megara. Busolt and Meyer also argue that the classes 51-60 will have suffered specially heavy losses in the battles of 459-445, while the eight youngest classes would have seen no fighting; but the classes 29-50 will equally have suffered between 459 and 439, and in addition many Zeugitae had left Athens as cleruchs between 445 and 431, and these will have been mostly younger men.) In 1885 Beloch solved the difficulty by deleting «ai wuvpiwy from II. 13, 6, making the number of the reserve 6,000, of whom 3,000 were metics; there would thus be 3,000 oldest and youngest citizens against 14,000 between 21 and 50, a satisfactory proportion. He was followed by Busolt. In 1905 he gave this up for a new solution, that the figures é£axioyiAiwv Kai pupiwr, and in consequence the next two sentences as well, were inserted by Thucydides’ editor, without however explaining why the editor should make so silly a statement any more than the historian. In 1923, still positive, he returned to the solution of 1885. It is tempting to alter the text ; numbers are easily corrupted. Though, if one must alter it, I prefer the suggestion that the figure 16,000 comes from a marginal note to the effect that the total active hoplite force was 16,000 (‘13,000 citizens plus 3,000 metics—see 31. 2’) and has ousted some other figure for the reserve. But, unfortunately, Ephorus had these figures, or something like them (Diod. XII. 40. 4). It is answered that the worst corruptions can occur quite early in the history of a text; did not Ephorus read dexdtm ére in Thuc. I. 103. 1, which may be corrupt for rerapr@ or (as I think more likely) But the possibility that one corruption (of such a kind) had occurred in Thucydides’ text and was accepted by Ephorus, makes it not more but less probable that there was a second; and though it is probable that some improbable things will happen, we must not strain the principle. Ephorus, who to all appearance was no great chronologer, may easily have been misled by an error affecting only the history of Sparta a hundred years before his time; it is to the last degree unlikely that he would have accepted figures so apparently absurd, and a proportion that, if false, would have been as false for the Athens of his own day, which he knew, and for any other city or group of men, as for the Athens of 431. Moreover, he did not simply accept the figures in Thucydides, for his are somewhat different : 12,000 hoplites of military age and 17,000 in the reserve, oldest and youngest and metics—the same total, though the figure for this is not in Thucydides or Diodorus, but different items; just as in this sd 4 > EXT@ ETEL | deme that he could almost have known by name every member of it—old, young, and of military age; and who calculated that the oldest and youngest of the Spartan army at Mantineia formed only one-sixth of the whole—V. 64. 3), this, together with Beloch’s vacillation, shows the difficulties that arise from attempting to dismiss, instead of solving, the problem. 1 It is unnecessary to do more than refer to Wilamowitz’ view (Ar. u. Ath. II. 201 sqq.) that Thucydides simply made the superficial calcula- tion that oldest and youngest=men in their prime. That a man of Wilamowitz’ understanding could suggest this of Thucydides (who lived in an Organized state with a xarddoyos and Anéiapxixa ypaupareta, and himself belonged to so small a 144 A. W. GOMME same chapter (which is in the main based on Thucydides II. 13) he corrects the latter’s figures for the tribute of the Allies. Whether Ephorus made this correction as the result of his own researches, or borrowed it from another,? it is clear that Thucydides’ figures were tested, and an alteration made not by the simple and obvious deletion of the awkward ten thousand, but by making an apparently extravagant proportion still more extravagant. In fact, this reading é€axicyiriwy xai wupiwv is one of the best attested in Thucydides and must be kept, ei rs xai dAdo, at almost anycost. Thisis not the same thing as accepting it as historically true; all we have to show is that a statement by Thucydides is not in itself absurd, nor inconsistent with what he says elsewhere. What does he mean by his 13,000 hoplites and 1,000 cavalry: does he mean strength on paper, or effective strength? The total number of men of hoplite census and over, between 21 and 50, or the number who actually joined up and, in the autumn of 431, marched into the Megarid?? Beloch takes it to mean paper strength only (and indeed of men up to the age of 60) ; yet when he deals with the figures for the battle of Delion, although Thucy- dides there uses the same terms, zravdnyet and wavortpatia (IV. go. I, 94. 1), he assumes that they denote effective strength. Or rather the effective strength reasonably expected by the generals: ‘ Unter “ Sollstarke”’ verstehe ich die Reiter und die Hopliten von 20-60 Jahren, die in den Listen gefuhrt wurden, unter “‘ Starke des Gesamtaufgebotes” die Reiter und Hopliten von 20-50 Jahren, die nach der Schatzung kompetenter Offiziere bei Feldziigen zur Verfiigung standen, ohne Riicksicht darauf, ob sie sich bei dem betreffenden Aufgebot auch wirklich gestellt haben’ (Kiio, p. 374). But he does not explain why Thucydides should mean one thing in 431 and another in 424 (and why, for the fourth century, Xenophon follows the latter and not the former practice). He does not tackle the question of paper strength. (I use the term ‘paper strength,’ as Beloch, of all men of hoplite census between the relevant ages, and ‘total muster’ for his ‘Starke des Gesamtaufgebotes’ ; ‘ effective strength ’ means the number of men actually in the field.) To get at the effective strength from a given paper strength we must deduct (1) the malingerers, (2) the permanent margin of temporary sick and absent, (3) the exempt, (4) the permanently unfit. For the campaign of 431 we may safely ignore the malingerers as a negligible quantity : the war had but just begun (yAv«ds aeipois 6 1oXepuos), hopes were high, the campaign would be short and was in the autumn, the discomforts and dangers trifling. The Beloch the latter view. Nor is it certain, though it is always assumed, that the roféra:, horse and foot, were thetes. It may be noted that after the reform of ¢. 336 B.c. the epheboi (whom Beloch now—1923—believes to have been men of the hoplite-census only, though formerly— 1905—he thought they included thetes) were trained in the use of the bow, the javelin, and the sling, as well as in rd érdopayeiv (AO. 42. 3). But I omit these from my argument, as well as (for the present) oi év rots ppovplas. 1 Beloch says from Androtion, who is also the authority for ’A@. 24 (Gr. Gesch, 387). If Andro- tion is anything more than a name to us, this last suggestion is very wide of the mark. 2 To avoid a clumsy elaboration I shall use the terms ‘hoplite-strength’ and ‘hoplite-census’ to include the cavalry, the men of the two richest classes—all oi drda wapexduevor. It is not certain whether the epibatai of the fleet were men of hoplite-census or thetes armed by the state. Meyer maintains the former, second have be for we | conditi a certa more d see Wil least th: of them form pz were th have re as a clas given cz 100 trier The meant n his attac unfit to 1 in difficu regiment bayonet A.O.C. of 50 is < for Rome that mor as hoplite four-fifths an except: tions,! ou between 4 tion of G p. 163: w not know) mean 2,70 Pericles in its affairs ¢ unfitness | (AOr, 49. lost hands 1 It is extre such analogie: State, whose degree that v and emigrati ‘ects the ade this ther, it t by the king an ; reading must be ccepting ucydides does he of men actually Beloch - of 60) ; 1 Thucy- 94.1), he strength > ich die wurden, on 20-50 igen zur reffenden does not rr in 424 not the . (I use between sebotes ’ ; we must sick and xn of 431 r had but gn would ng. The tain, though ., horse and d that after eboi (whom e been men . formerly— ‘hetes) were javelin, and (AOm. 42. 3). t, as well as THE ATHENIAN HOPLITE FORCE IN 481 B.C. 145 second class were certainly ignored by Thucydides in c. 31, for they must have been by Pericles in c. 13; how many there were we cannot possibly sa for we do not know what the average of sickness in Athens was (under ae conditions of peace; the army had not been in the field). But there would be a certain number, especially of those between 40 and 50. Exemption is a more difficult matter. There were 500 bouleutai and 700 officials (’A@r. 24: see Wilamowitz /.c.), the great majority of the latter civilians. Of shebe a least three-quarters will have been of the hoplite-census, and some two-thirds of them between 30 and 50: say 500-550 men exempt, who did not in fact form part of the effective strength of the army that invaded Megara. But were they included by Thucydides and Pericles? As individuals they must have remained on the xarddoyos, for they changed from year to year; but as a class they were constant, and competent officers would know that fee any given campaign some 500 men on the register would be unavailable. The 100 trierarchs (c. 23. 2) would be similarly unavailable. The last class, the permanently unfit, is the most important; by these are meant not ‘ weaklings and semi-invalids,’ as Busolt so scornfully calls them in his attack on Meyer (who alone has drawn attention to them), but all men unfit to march and fight in heavy armour, to bear the fatigues of active service in difficult country, all who would be a hindrance rather than a help in the regiment. Remember that all hoplites belong, as we should say, to the bayonet strength of the army; none are orderlies or batmen or AS.C. or A.O.C. Remember also that mobilization for service abroad up to the age of 50 is a very severe measure according to both ancient and modern ideas; for Rome, as for us, 40-42 was the normal limit. It is unreasonable to idtde that more than one-quarter of the men of 41-50 were fit for active service as hoplites, even for so gentle a campaign as that to Megara; nor more than four-fifths of the men between 21 and 40 (such a proportion would argue an exceptionally healthy population). On the analogy of present-day popula- tions,! out of 14,000 men between 21 and 50, some 3,600-4,000 would be between 41 and 50 (the smaller number corresponding to the growing popula- tion of Germany, the larger to the stationary population of France—Meyer p. 163: whether Athens had a—naturally—increasing population or not, we pr not know). If we take the lower figure—to be on the safe sivisdibe would mean 2,700 unfit men of 41-50, and about 2,000 21-50. Were they included by Pericles in his figures? I cannot believe it. No state could possibly manage its affairs on such a system. How a man secured exemption on the ground of unfitness we do not know; for the cavalry apparently his oath was taken (A@r, 49. 2); but exemption must have taken place—or were the men who lost hands and feet and eyes in the plague (II. 49. 8) kept on the regimental 1 It is extremely unsatisfactory to have to use else, and we are not seeking exact figure such analogies, especially in the case of asmall (Beloch’s great service has been to insist “A th : state, whose population had been effected to a necessity of making some calculation, on in degree that we cannot estimate by recent wars falsity of doing nothing because we cann t be and emigrations. But we cannot do anything exact.) : L 2os= ~~ ‘ = : 146 A. W. GOMME strength and regularly called up for every campaign and as regularly failed, to the disappointment of the commanders?! But even if the unfit remained on the strength, they could not have marched; which is my point at the moment. Taking then 300, say, for the first two classes (the malingerers and the tem- porary sick), and 600 for the third, the exempt, we must deduct over 5,000 from the 10,000 paper-strength of the army of Megara. Or, if this campaign was intended chiefly for display, and many really unfit men marched with the rest, say half of them (with the consequence that the 13,000 of c. 13 is only taken from c. 31 and is misleading), we must still deduct 3,500; and all from the 10,000 of the Megarid army, none from the 3,000 at Potidaea, who must have been counted (for they were paid) and have been picked troops chosen for a long and arduous campaign. Fewer than 7,000 then really marched into the Megarid? But Thucydides says, quite categorically, that there were as many as 10,000. Did he make so gross an error?? Beloch notes (Kilo, p. 344) that Thucydides (who had been strategos) is careful not to give figures for armies when his information is only that a mobilization is wavénpei, knowing that effective strength is a different thing from paper strength (cf. his caution about Delion and Mantineia). Why, then, on this one occasion (when he was prob- ably present, and for an army which he knew) does he forget his caution, or rather every grain of common sense ? We are compelled to the view that, according to Thucydides, the ‘ total muster’ of the Athenian hoplite force in 431 (ignoring, that is, the malingerers and the temporary sick and absent) was 14,000 men (including the cavalry). If there were some 5,000 men between 20 and 50, exempt and unfit, relegated to the reserve, as suggested above, then to get the relative numbers of the classes 21-50 and of the oldest and youngest we subtract this from the reserve of 13,000 (16,000 less 3,000 metics) and add it to the 14,000 of the active army, making a ratio of 8,000 to 19,000 (which latter figure gives us of course a lower propor- tion of permanently unfit than that suggested above; but I will keep on the safe side). Let us now take the metics—3,000 of them took part in the Megara campaign; and it is generally assumed that this means the whole number of metic hoplites between 20 to 50. (Meyer even ignores oldest and youngest metics, and so gets an exaggerated estimate of the total number of citizens of hoplite-census). I see no reason to suppose this. As far as we know, this was the first occasion on which metics were employed as hoplites for service abroad ; at least they are not mentioned in the crisis of 458 when the oldest and youngest defeated the Corinthians (though we must not build much on omissions in the meagre narration of Pentecontaétia), and Thucydides (c. 13) reckons them all in the reserve. There were none at Potidaea. It is obviously the contrary (év rw xaradéyp : Aristogeiton wanted his name struck off the active list). 2 We need not work out the figures of absentees on Beloch’s assumption that the 13,000 are all hoplites up to 6o. 1 Beloch does maintain that unfitness was only tested on mobilization for each campaign (Geschichte, p. 388), and bases his view on an anecdote in Plutarch (Phoc. 10), which is worth- less for the purpose, and suggests, if anything, proba the c once reser alreac only t marcl active doing and 5 were } If other | Over 5 list, an tractec must | hpoupi: identif, danger nothin; contra! the acz Thucy¢ subject have bi perhaps J.ALLS. - active li Bu smaller assumec sword a when T means a countrie 1 Fawcu ‘the men i the recrui giving peac (II. 18-19), recruits in 2 Assum not include prises oi me failed, to ained on moment. the tem- yer 5,000 ampaign with the 3 is only all from vho must hosen for 1 into the .as many 344) that yr armies ving that ion about was prob- 1ution, or the ‘ total alingerers valry). If legated to he classes 2 of 13,000 making a er propor- ep on the e Megara 1umber of _ youngest citizens of y, this was se abroad ; dest and much on les (c. 13) ; obviously peiton wanted e figures of hat the 13,000 THE ATHENIAN HOPLITE FORCE IN 481 B.C. 147 probable that the metics of hoplite-census were not so thoroughly organized as the citizens; that men who settled in Athens at the ages of 35-50 would not at once be enrolled as in the active army, though all could be counted in the reserve ; that only men who came to Athens very young, and the sons of those already settled there, were trained soldiers of regular battalions. We have only to assume that there were 2,000 metics of hoplite-census who did not march to Megara—who were between 51 and 60, or unfit, or not enrolled in the active army, who were importing corn or making arms or in some other way doing business as usual—and we have left 6,000 citizen hoplites of 19-20 and 51-60, which is (on modern analogy) about the number required if there were 19,000 of 21-50. If anyone thinks the above figures for the unfit exaggerated, there are still other men who were certainly available for the reserve: for example, epibatai over 50 or unfit. There must have been at least 2,000 epibatai on the active list, and so about 700-800 in the reserve. If they were thetes, 700 must be sub- tracted from our figure for the hoplite reserve. If they were hoplites, 2,000 must be added to hoplite active list. Secondly, there are the men éy rois dpoupios (II. 13. 6). We do not know their numbers. Meyer and Busolt identified them with the 2,500 hoplites of AO. 24. 3. But, apart from the danger of relying on the statements of that very rhetorical chapter, there is nothing to show that these 2,500 were engaged in garrison duty;! on the contrary, they are the orpatevoyevor of 24. 1, as the garrison of the arsenal and the acropolis and the 20 guard-ships are the ¢pouvpotvtes. The dpovpa in Thucydides may be the frontier posts of Attica only, or may include as well subject cities which were garrisoned. In any case the number—which cannot have been less than 500, or they would not be worth mentioning, and was perhaps 1,000, for there were many ¢povpia in Attica itself (see L. Chandler in J.H.S. XLVI., 1926, 1 sqq.), is to be added to that of the hoplites on the active list.” But I do not regard the number above suggested for the unfit (which is smaller than Meyer’s) as at all exaggerated. It is extraordinary how often it is assumed that every citizen of a Greek city between 20 and 50 was fit to wielda sword and carry shield, cuirass, and helmet (or to row in a trireme)—or that when Thucydides or Xenophon says so many men went on a campaign, he means about two-thirds of the number. They forget how many men in modern countries are relegated to the reserve or are fit only for the auxiliary services. taken strictly, should mean (rogofra yap... dwré re Trav mpeoBurdrwr Kal Trav v. . . . TOU Te yap Pax. relxous, xrX.). If they are included in the 16,000 their number has to be deducted from that of the men of 19-20 and 51-60 and added to the active list, thus reducing still further the proportion of the former to the latter, making it at most 5,500 : 19,500. 1 Fawcus (p. 23), relying on 'A@. wr. 42, says ‘the men in the garrisons’ were ‘the youngest,’ the recruits of 18 and 19. But Aristotle is giving peace-time conditions. Oinoe, for example (II. 18-19), was certainly not garrisoned by raw recruits in 431. 2 Assuming, that is, that oi éy rots ¢povplos are not included in the 16,000 ; that that figure com- prises oi rap’ radi only, as Thucydides’ words, 148 A. W. GOMME The number 13,000 for the Athenian garrison reserve will of course represent paper strength only.’ All that I assume (I am in this only following Meyer) is that the whole of the reserve, including the exempt and the unfit of 21-50, are described by the term ‘oldest and youngest ’—not a very extravagant assumption. Thucydides says that the army of Myronides in 458 consisted of the oldest and youngest ; but no one supposes that only men of 19-20 and 51-60 went on this campaign, that men left behind from the Egyptian and Aeginetan campaigns for one reason or another did not join Myronides’ force. We may note that the words hopos aro tav Evypdyov are also, apparently, loosely used in this chapter. It is certainly not so extravagant as to suppose at once that figures so well confirmed as those of II. 13 are corrupt and that Thucydides included unfit men in the active forces of Athens. In general we are liable to underestimate the degree of military organization at Athens. To beable, at the right time, to man a fleet of 100 or 150 triremes with fit and trained men and provide them with supplies, especially if most of them were foreigners, argues a high degree of or- ganization. There is no reason to suppose that the army was any worse managed. Beloch, however, objects to these figures on two other grounds as well; that they are inconsistent with the figures for Delion (and with Xenophon’s and others’ for the fourth century), and with Pericles’ strategy. The latter argu- ment is easily answered. Why should Pericles, says Beloch, have refused battle with the Peloponnesians if he had 29,000 hoplites at his disposal? But, whatever figures we adopt for the total number of the hoplite-census, it is agreed that Pericles had no more than 14,000 men (no more than 11,000 in 431 and 430, with 3,000 away at Potidaea) who were (more or less) equal, man for man, to the enemy. Would any sane general have encumbered that army with the unfit and the old in order to face the picked troops of the Peloponnese and Boeotia, who numbered at least 27,000 (Beloch, Klio VI., 1906, p. 77) ? The reservists had won, or helped to win, a victory over the Corinthians in 458; but that was at a moment of exceptional energy and elation of spirits; it was a hazardous experiment, the result a piece of good fortune not likely to be repeated. 1 I leave out of consideration the thetic reserve (men unfit to row in the fleet and act as rotéra), though these would certainly be available for the defence of the walls. Thucydides’ figures are in fact too low for the total reserve, not too high. 2 Beloch (Klio, p. 363) notes that Thucydides’ language in I. 105. 4 is not very happy: rév 5’ éx rhs wédews Urodolrwy of re mpecBiTara Kai ol vewrara, as though it was only the oldest and youngest of those that remained in the city who joined up, and the fittest men stayed behind. He also argues that the whole of the Athenian active army was not engaged in Aegina and Egypt, and that the greater part of Myronides’ force consisted of first-line troops. This may be so, but if we assert it we are contradicting Thucydides. He bases his argument on the fact that Tanagra also was fought while the: Athenians were still engaged in Aegina and Egypt, and to that battle they sent an army of 14,000 men, most of whom must have been Citizen troops. It is puzzling. Butitis probable that after the investment of Aegina, and perhaps also of Memphis (and after Myronides’ victory), troops were withdrawn and returned to Athens; just as Pericles withdrew part of the forces from Samos (I. 116. 3), Phormio from Potidaea (I. 64-65, II. 58. 2), and the Peloponnesians from Plataea (II. 78. 1). Cf. E. M. Walker in Camb. Anc. Hist. V. 81. (1 cannot believe, in view of Pericles’ expedition to Sicyon and Oeniadae and Cimon’s to Cyprus in 449, that the Athenians lost 250 ships and 50,000 men in Egypt—id., p. 84; cf. p. 78, mn. 1.) conti horse engag Thra the r woulc (citize died « aveker the sa hoplit nearly citize1 expres census the n 16,000 not be contac remair Sparto hoplite additic Aegina hoplite subtrac metics we refi (II. 49. metics, Thucy¢ Hi content plague ITI. 87. army. 1 Beloc and so th that the 1 are hoplit 2 Beloc have had before an will inclu during th It m epresent vhole of by the icydides ungest ; mpaign, for one 1e words ter. It so well ed unfit mate the , to man 2m with ee of or- lanaged. as well; on’s and er argu- refused ? But, us, it is O in 431 man for my with ponnese p. 77)? hians in irits ; it likely to while the: egina and n army of 1ave been s probable id perhaps s’ victory), o Athens; the forces 1 Potidaea yonnesians Walker in believe, in cyon and 449, that oo men in THE ATHENIAN HOPLITE FORCE IN 481 B.C. 149 At Delion the Athenians were present in full force, with the metic contingent, and could only muster about 7,000 hoplites and perhaps 1,000 horse (IV. go. 1, 93. 3, 94. 1). We do not hear of many Athenian troops engaged elsewhere at the time—4o00 with Demosthenes (101. 3),1 some in Thrace (82), some doubtless with the forces in the Hellespont (75), as well as the regular garrison troops in Attica and some of the subject cities. There would not be more than c. 10,000 active troops, as against 17,000-18,000 in 431 (citizen and metic hoplites, cavalry, and oi év tois dpovpious); 4,400 hoplites died of the plague, é« trav radfewv, and 300 of the cavalry, rod 5¢ dAXov dyAov avefevperos apiOuos (III. 87. 2-3).2 Since it is probable that there was roughly the same proportion of losses in the hoplites as in the cavalry, we may put the hoplite deaths at 30 per cent. of their number; this would mean a total of nearly 15,000 before the plague, which is not much more than the number of citizen hoplites in 431 (13,000 +0i év rots dpovpiows); so that é« tay tafewr is expressly written to show that the 4,400 are not simply men of the hoplite- census, but hoplites of the active army.* They include, as likely as not, the metic losses as well; in which case the hoplite losses (4,400 out of 16,000-17,000) will not have been so severe as those of the cavalry, which would not be surprising, as some of the former in the garrisons did not come into contact with the plague, and those at Potidaea not so thoroughly as men remaining in Athens. Besides these losses, there were not many from battles, Spartolus, where 430 fell (II. 79), being the only serious conflict in which hoplites were engaged. We could put these losses at 600 or 7oo in all.* In addition some zeugitae had doubtless left Athens as cleruchs to Potidaea, Aegina, and Mytilene. From all causes there would be a net loss to Athens of hoplites of the active list of about 5,700. Add the 300 cavalry losses, and subtract from the 17,000-18,000 of 431 (13,000 hoplites + 1,000 cavalry + 3,000 metics +0i €y Tois hpovpiows: say 17,500), and we are left with 11,500. When we reflect that a large number of men who survived the plague were crippled (II. 49. 8), it is not surprising if there were not more than 10,000, citizens and metics, fit to fight in the ranks in 424. There is no inconsistency in Thucydides’ two sets of figures. How falsely this matter can be argued will be seen by Beloch’s main contention (Kiio, p. 361). He believes (against Meyer) that the figures of the plague losses refer to all adults of the hoplite-census (making é« rav rafewy of III. 87. 3 superfluous) ; but he will assume that they refer only to the active army. There would then be, he says, about 10,000 deaths among the hoplites 1 Beloch (Klio 360) says these were epibatai cydides’ language—rod d¢ &ddov bxAov dvetevperos and so thetes; but on the same page he agrees dpOuds (cf. II. 31. 2, IV. 94. 1)—does not suggest that the 1,000 on board the fleet of 431 (II. 23. 2) _ that hoplites outside the active army are included are hoplites. The cases are exactly parallel. in ‘the rest.” But note that is particularly the 2 Beloch argues that Thucydides could only dtvaus, the military force, of Athens with which have had comparative figures of the hoplite force he is concerned (III. 87. 2). before and after the plague, so that this number 4 Beloch (Klio 359) says these war losses would will include also any who died a natural death have been more than made up by 424 by ‘ natural during these years. increase.’ But many boys between 14 and 20 3 It must be confessed, however, that Thu- must have died during the plague. 150 THE ATHENIAN HOPLITE FORCE IN 481 B.C. of all ages (29,000 : 10,000 = 13,000 : 4,400); add a loss of 1,500 for war and cleruchies, and in 424 we should have 17,500 hoplites, of whom 4,500 would be of 19-20 and 51-60—i.e. 13,000 of military age or nearly twice as many as at Delion. ‘So one or other of Thucydides’ figures must be false.’ On the contrary; taking into account the abnormal invalidity following the plague, it is nearly the proportion that we should be compelled, 4 priori, to assume. Beloch himself (pp. 372-3), by rejecting the figures of II. 13, comes to the con- clusion that there were 8,000 men of hoplite-census, citizen and metic, between 21 and 50, and no more, and that of these 7,000 fought at Delion. It is fantastic. Even to get this margin of 1,000 he has to ignore the cavalry (the gaps in whose ranks he agrees were made good from the hoplites—Ar. Equit. 225) and to assume that not another Athenian hoplite was engaged elsewhere during the Delion campaign.! | These figures affect the further problem of the population of Attica in the fifth and fourth centuries. With that I am not at present concerned. But before being used, ignored, or altered, every important piece of evidence, in such a problem, must be examined to see if it is (1) well attested, (2) reasonable in itself, and (3) consistent with other statements of the same author. If it is, it must stand, as the evidence of that author. Thucydides’ figures for 431 B.c. fulfil all three conditions. They are not only reasonable and consistent; they are necessary, and alone consistent. It is the modern, assumed figures which will not work. If they are inconsistent with the evidence of Lysias, Xenophon, Diodorus, Ktesikles, then only such weight can be attached to them as Thucydides’ name and other considerations deserve (for example, the MS. tradition of the later figures is not so well attested). But fortunately they are not inconsistent with the fourth century evidence. That,as Beloch has shown, is consistent with Thucydides’ figures for 424; it is therefore consistent also with those of 431. They give a total citizen population of hoplite-census and over, between 21 and 60, of 24,000-25,000, therefore a total of adult hoplite of 27,000-28,000.2, We have no figures, even approximate, for the number of thetes. Even if we knew for certain the number of Athenian ships in 479 (when the hoplites were at Plataea), and the number of Athenian thetes on them, we should still have no evidence for 431; for we do not know how their numbers were affected in the interval by natural increase or the reverse, by war losses, by emigration as cleruchs, by the increase of wealth and their transfer- ence to the hoplite class. I cannot go into the question now; but we are not compelled to suppose that there were ‘more than 12,000 to 15,000 of them in 431; which would make the total citizen population about 40,o00—a figure that Beloch would accept. A. W. GOMME. 1 1.G.2 I. 949 alone makes this assumption would give a proportion of hoplite population to highly improbable. Acharnae roughly corresponding to its represen- 2 I cannot accept the 3,000 hoplitesofAcharnae, tation—22—in the boule) ; but it is useless to even if they include all hoplites over 20 or over emend. The figure is not confirmed by Ephorus. 18. XXX might be a corruption of XHH (which M of 154 proven some t from t! folio I Thoms pretati Tl handb« deities numer the rev most 1 Vatica Third connec tenth accura mytho; Serviu: indirec work t and w phrase book 1 heroes is inch Pi such c on th Lactar value, thougk first a medie\ the gt author Isidor¢ war and ould be ny as at On the lague, it assume. the con- between It is Iry (the uit, 225) » during a in the d. But “nce, in sonable If it is, 13I B.C. t; they ; which 1ophon, 1em as 1e MS. hey are shown, nt also us and plite of iber of in 479 ates on w their by war ansfer- ire not hem in figure MME. lation to epresen- seless to ‘phorus. AN ENGLISH COMMENTARY ON OVID. MS. 124 in the library of St. John Baptist College, Oxford, is a quarto of 154 pages, written in two or more hands of the fifteenth century. Its provenance is unknown, and, as it has been carefully and ruthlessly rebound at some time in the nineteenth century, it is impossible to derive any information from the binding. After an unusually elaborate alphabetical index it bears on folio 11 an inscription in a seventeenth-century hand as follows: ‘ Libellus Thomae de Walsingham De generacione naturaque deorum simul cum inter- pretatione Ouidii Metamorphoseon Simoni Priori Sancti Albani dedicatus.’ The inscription is roughly accurate. The first treatise is an elementary handbook to mythology, and describes the origin of the more important deities and their various emblems and symbols. It is compiled from the numerous handbooks to mythology which were in vogue in Europe soon after the revival of interest in classical literature under Charlemagne. Three of the most important of these were edited by Bode in 1834 from MSS. in the Vatican. They are usually quoted for convenience as the First, Second, and Third Vatican Mythographers. The present compilation is most closely connected with the third of these treatises, which belongs to the ninth or tenth century ; but it is a patchwork composition, not nearly so complete or so accurate as the parent treatise, and borrows occasionally from the second mythographer. These mythographers derived their information largely from Servius, Fulgentius, Martianus Capella, Isidore of Seville, and directly or indirectly from the first book of Cicero, De Natura Deorum. In the present work the fable is generally followed by an ‘ exposicio’ explaining its meaning and where possible its moral. The tract on Mythology is followed by a para- phrase of the text of the fifteen books of the Metamorphoses. At the end of the book is a series of genealogical tables, tracing the descent of the gods and heroes from Celius, the father of Saturn, down to the Emperor Nero, who is included as being the last descendant of Aeneas. The ancient scholars rather neglected Ovid. They certainly wrote no such commentaries on him as Servius wrote on Vergil. The earliest work on the Metamorphoses which survives is the ‘narrationes,’ attributed to Lactantius Placidus. This is a prose epitome of the poem. It is of little value, and the name and date of its author are quite uncertain. Heinsius thought it was later than the Gothic wars, and it must be earlier than the first and second Vatican mythographers, since they quote from it. The medieval scholars, however, wrote commentaries on the Metamorphoses with the greatest zest from the eleventh century and onwards. The saintly authorities of the Early Church had shaken their heads over the sinner Ovid. Isidore of Seville thought him particularly noxious. But the poet was 152 F. W. HALL undeniably attractive to read—even Isidore ends by quoting about twenty pages from him—and parts of him, such as the Metamorphoses, were too useful in education to be neglected. Like an oyster enamelling into a pearl some foreign body within its shell, the Church overlaid the Metamorphoses with a surface of moralized allegorical interpretation. This process began abroad, especially in France, where Ovid was a particular favourite. Perhaps the most popular of all the commentaries of this type was one written by Petrus Bercherius or Pierre Bersuire at Avignon circa 1330, printed by Regnault, 1515. His treatise is not badly written, but one specimen of his art will suffice. In the story of Phaethon and the chariot of the sun, the Sun, he says, typifies the Pope; Phaethon is a young and imprudent prelate whom the Pope has set over a province of the Church, who perishes from his own wicked ambitions. He cannot rule the chariot, which is the Church, nor discipline the horses, who are his subordinates, who accordingly become inflamed with wickedness and burn up the earth. The commentary in this MS. is written on similar lines, though, as far as I can see, it is not influenced directly by Bersuire’s work. The author, however, as he admits, has sought for material wherever he could find it. In particular it is to be noticed that he uses the curious [ntegumenta Ovidi— rough elegiac distichs or quatrains summarizing the stories in Ovid’s work. These are generally attributed to a certain Iohannes Anglicus, and belong, according to Traube, to the thirteenth century. The MS. is, of course, quite useless for the exegesis of the text of Ovid. My object, however, is to call attention to the compiler Thomas of Walsingham, and to the hitherto unpub- lished preface with which he introduces his work. He is, of course, one of the best known of the monkish historians, and is our chief authority for the reigns of Richard II., Henry IV., and Henry V. Anything that he wrote must be of interest to us. None of the modern authorities on Walsingham seem to have troubled to enquire about this manu- script. It was inaccurately described by Tanner in his Bibliotheca ; and Riley, the editor of Walsingham’s works in the Rolls Series, only repeats the information that he found in Tanner. The author of the account of Walsing- ham in the D.N.B., however, wrote after the publication of Coxe’s accurate description of the manuscript, and might even, since he was a fellow of a neighbouring college, have consulted the MS. itself. Yet he is content merely to repeat Tanner’s account, adding loftily that ‘the title of the MS. suggests remoteness from W.’s literary pursuits.’ The Preface runs as follows: In Christo Patri venerabsli Domino Simoni priors monastertt Sancti Albani: sacre pagine professori> Thomas de Walsingham suorum minimus subditorum veuerentias debitas et honores. {| libellum de genevacione naturaque Deorum simul cum interpretacione fabularum libri Metamorphoseos ex opusculis plurimorum magna collectum diligencia parens vestris tussts edidt, tanto quidem libencius quanto nout vos omnes res poeticas apprehendtsse profundius § Ea videlicet considevacione ut si quid secus, st quid minus quam ves postulat inseruero. lima vestve discrectonis et sciencte quam nichil latet ommino poeticum corrigatuy | Dsstenxt vero presens numerum pboeticum non transgrediens in libros quindecitm anteponens et connectens a es tvactatum libvorum | parte toll libvorum mouere di eadem tai omnia mt expletum deorum f extitisse. quast mar nostvis sa intvare VC ad intvar vacionabsl UNIUEYS? 1 dantuy 4 venevande Deorum 1 Thi Benedic St. Alba voted at on Octo and The He reca ad eoru! Gesta Al —whicl not see ham in assert tl of Simc visiting of some ‘In poe (Amunc in the I posed | Walsin; at St. permitt Th is impo Southe it twenty 00 useful arl some Ss with a _ abroad, haps the y Petrus cegnault, art will he says, she Pope 1 wicked iscipline ied with is far as author, find it. Ovtdi— S work. belong, e, quite to call unpub- , and is enry V. modern /manu- | Riley, its the alsing- ccurate vw of a merely iggests 2 pagine honoves. m libre S tUssts fundius 0. lima xt VEO inectens | AN ENGLISH COMMENTARY ON OVID 153 tractatum de natura picturaque deorum primo libro Nasonis methamorphoseos cauens plures libvorum numeros ponere quam poeta dictauerat et stbi deferens in hac parte 4 Ex alia uero parte tollens occastonem lectortbus superficialibus conquerendt ne si multiplicassem presencium libvorum numerum primum librum poete uel posutssem tercium uel secundum | Nec lectores moueve debet, licet in multis locss velut eadem repetita reperiant que etst videantur in multis eadem tamen nusquam itevantur vel sine addttamento vel sine causa, et, precipue quia non omnia mihi que volebam pariter occurrebant, sed per successiua et varia interualla. | Post expletum librum quintumdecimum pater honorabilis tabulam de generacionibus heroum vel deorum posui a primo celio patre saturnt usque ad neronem quem de stirpe enee legimus extitisse. In qua lectores minus prousdt non solum ves ymaginan poetscas sed eas ut sta dicam quast manu poterunt contrectare 4 Et hunc quidem libencius assumpss laborem ut grammaticis nostvis satisfacevem qui dicunt res poeticas figmenta friuola siue vana et non (sic) nonnullos intvave volentes ad cognoscendum poemata non permittunt, cum spst nec intrent nec sint habiles ad intrandum % Horum causa pater amabilis laboraui diligencius ad assignandum tam vacionabiles causas quam utiles singulis poete fabulis ut videant confundantur zelantes popult } uniuersi videant quidem fructum quem prius non gustauerant tmesse vebus poeticts, et, confun- dantuy inuidt qui nobis detrahebant igneque ruboris et vevecundte conflagrentur, Vestre venevande paternitati doctor et pater eximie me commendo, Explicit prologus in archana Deorum Incipit liber primus Ourdit Metamorp hoseos. The Prior here mentioned was Simon Southerey, formerly head of the Benedictine hostelry in Oxford, and at a later date Prior of the cell of St. Albans at Belvoir. Both he and Thomas of Walsingham were present and voted at the election of John de la Moote to be thirty-first Abbot of St. Albans on October 9, 1396. Soon after his installation the new Abbot recalled Simon and Thomas Walsingham, then Prior of the cell of Wymondham, to St. Albans. He recalled them at their own earnest request. ‘ Mundialibus curis lacessitos ad eorum preces et instantiam praecipuam ad monasterium reuocauit ’ (Riley, Gesta A bbatum 3. 436). If Abbot John recalled them soon after his installation —which I suppose must have followed close upon his election in 1396—I do not see why Riley should say that Walsingham ceased to be Prior of Wymond- ham in 1400; nor do I see why the article in the D.N.B. should go further and assert that he held the office till 1409. The new Abbot secured the election of Simon as Prior of St. Albans, and is said to have used him as his deputy in visiting the dependent monasteries in the See of Lincoln. Simon was a man of some distinction. He was especially skilled in the writing of Latin verse : ‘In poetria doctissimus inter cunctos regnicolas nostris temporibus habebatur’ (Amundesham, 2. App. E., p. 305). He studied astronomy, and MS. Digby 98 in the Bodleian contains a fragment of an Almanack of the Fixed Stars com- posed by him. He was also one of the staunchest opponents of Wiclyf. Walsingham, who had formerly been Precentor and head of the Scriptorium at St. Albans, and had been appointed Prior of Wymondham in 1394, he permitted to enjoy the quietude of the cloister, ‘ claustrali quieti uacare.’ The present compilation must of course be subsequent to 1396, but it is impossible to fix the date accurately, since we do not know when Simon Southerey ceased to be Prior. He still held the office in 1401, when the new 1 Isaiah 26. 11. 156 J. A. SPRANGER the sacrifice. The former of the two suggestions has met with the almost unanimous approval of modern commentators, and may be said to be now generally accepted; but it is not without its difficulties. The appearance of the shade of Achilles, the cause of all the trouble, must have taken place over his actual grave, and the Thracian cenotaph must be supposed to have been set up for the express purpose of the sacrifice, for, though cenotaphs were common enough, there was no other reason to erect one so near his actual place of burial and so soon after his death. But if the ghost in question appeared over the grave at Sigeum, then the problem at once arises, Why did the Greeks wait to perform the sacrifice until they had shifted their camp across the Hellespont ? There is not a single indication, in this first part of the action from 1. 69 to 1. 628, that the scene is laid anywhere but at Troy; 1. 74, as the Scholiast points out, is spoken as by one who was not in Thrace at the time. The Scholiast on 1. 521r/ also points out that if the sacrifice took place at Achilles’ tomb, the whole army, which is understood to have been then encamped in the Chersonese, could hardly have been present, as Talthybius says it was. The fact is that, as far as can be judged from internal evidence, the whole Polyxena tragedy takes place in the Grecian camp before Troy, and not in the Thracian Chersonese at all. The same cannot of course be said of the Prologue (cf. ll. 32-36) nor of the Polymestor tragedy, in which three verses (ll. 770-771 and 1. 1142) and the word évrai@a in |. 772 clearly point to the Thracian Chersonese as the scene of events. Certain slight inconsistencies have been pointed out? which do not fit in very easily to this Thracian scenario, but the verses quoted above are quite clear on the matter. Let us now see if any explanation can be found for all this confusion. We may begin by noticing that there was no a priori necessity for placing the scene of the Polymestor dramain Thrace. Homer tells us* that a Thracian army was present at Troy on the Trojan side, or at least so far sympathizing with the Trojans as to be considered and treated as an enemy by the Greeks. But this was while the fortune of Troy yet hung in the balance: once the fall of the city appeared inevitable, such a king as Polymestor would assuredly change his attitude ;4 nor would it be in Agamemnon’s interest to renew open hostilities with him on Hecuba’s behalf.® We might therefore assume that Polymestor, together with the Thracian warriors whom he calls upon in the moment of his dire need,® was encamped beside the Greek army, which he also calls upon for help in the same breath.’ This would explain how Hecuba comes to send a handmaid to fetch him in such a casual manner,® merely asking Agamemnon that she be given a safe-conduct 8:a orpatod.® 1 Schwartz, Scholia in Eur., Vol. I., p. 50. 4 Cf. ll. 1132-1144 and 1175-1177. 2 Ll. 823, 1020, 1111-1113, and Schol. ad loc. 5 Cf, ll, 850-863. 6 L. 1088. in Schwartz, Vol. I., p. 83. 7 L, 1091. 8 L. 890. 3 K. 4348qq. Cf. also the Rhesus, passim. 9 L, 889. Thi which cl] To Hecuba of actio coast; f the inte would b camp w In f and 114 remains suffer at such. On exclude To omit while th sense, as But no textu unneces: requisite Mo! together Separate first part Polyxen: have bee Per! drama t Verrall : as the H Suc. a drama short fo sentatio1 Thi: SUCCESS, | stage. |] it had t further bk for the si 1 Li, ¢ e almost » be now urance of lace over ave been phs were is actual question Why did sir camp 1 1. 69 to Scholiast ok place sen then althybius »vidence, ‘roy, and 6) nor of ) and the he scene ‘+h do not bove are ion. r placing Thracian yathizing - Greeks. the fall issuredly ew open Thracian ncamped breath.’ tch him given a 088. THE PROBLEM OF THE HECUBA 157 This we might assume were it not for the verses we have already noted, which clearly state the contrary. To continue. Polymestor’s explanation of his delay in coming to see Hecuba? is by no means incompatible with the supposition that the scene of action is to be understood as being in the Grecian camp on the Dardanian coast; for if we were to have imagined him residing at the time somewhere in the interior of his kingdom, a delay in coming to visit the widow of his old ally would be easier to account for and less likely to call for comment than if his camp were quite close at hand. In fact, if we take the whole play as it now Is, excluding Il. 32-36, 770-771, and 1142, and replace the word évtav@a in |. 772 by évOévde, no difficulty remains in assigning the scene of action of the play, nor does the play itself suffer any perceptible distortion. The present inconsistencies disappear as such. On the other hand, to keep the verses we have thus excluded, and instead exclude the inconsistencies on the other side, would be a far harder matter. To omit 1. 140, for instance, would involve us in grave metrical difficulties ; while the exclusion of ll. 540 and 1020 would interfere seriously with the sense,.as would also that of Il. 1111-1113. But the verses we have excluded are perfectly good and clear, and there is no textual evidence of anything in the nature of such a large and apparently unnecessary piece of interpolation and alteration: something more than this is requisite by way of a solution to the problem. Moreover, we still have to explain why Euripides should have joined together the Polyxena and Polymestor tragedies instead of making two Separate and longer dramas, as he well might have done. The subject of the first part of the Hecuba forms, so far as we can tell, the whole substance of the Polyxena of Sophocles; and the story of Polymestor could, one would think, have been worked up into a full-sized independent play without much difficulty. Perhaps we may find an explanation of all this in an aspect of Greek drama that has been brought to notice and discussed by Professor A. W. Verrall: I refer to the plays originally composed for private performance, such as the Helen, Ion, Medea, and others.” Such a play we may assume the Polyxena of Euripides to have been: a drama complete in itself, full of pathos and tragedy, but brief—far too short for public performance, though eminently suitable for private repre- sentation. This playlet we may imagine as having had a considerable measure of success, so much so that its author was induced to bring it out on the public stage. But this he could not do while it remained in its original form: either it had to be spun out to at least double its actual length, or else something further had to be added to bring it to the normal length of a tragedy intended for the stage of the theatre of Dionysos. 1 LI. 962-967. 2 Verrall, Four Plays of Euripides, Cambridge, 1905, pp. 123 sqq. 158 THE PROBLEM OF THE HECUBA This addition may have been the Polymestor tragedy, with the scene of action laid at Troy as for the Polyxena. Certain subsequent considerations may have led the author to alter the scene to Thrace. The first of these is that the airiov of the whole play, the transformation of Hecuba herself into a hell-hound and her death from the mast, was to be referred to a locality in the Thracian Chersonese,’ and the point of this would be emphasized by laying the scene of the play in the immediate neigh- bourhood of the Kuvoconyua promontory. The second is that the arrival of the Greek army on Thracian soil would give Polymestor, cowardly and cruel ruler of that land, the final stimulus toward the desperate step of murdering Polydorus.* A third consideration is that after the fall of Troy it would have been only natural for the Thracian contingent to withdraw to its own side of the Hellespont, if only to sever more completely any connexion with the defeated Trojans. These motives may well have led the author of the Hecuba to introduce the slight alterations we have noticed into his play when it was finally prepared for public performance. J. A. SPRANGER. 2 Schol., 1. 32. Schwartz, Vol. I., p. 16. 1 L, 1273. IN of the speaks | only ins as exer Thebes where v Th Plutarc who see be abse apparer they ca harmos Heripp! confirm ppoupas names. Tov éy ATEKTEL there w laid on particip fate fro the acc: harmost of strar has littl No usually 1 Comp gen. Socr. éxelvov (FP tenor of V 2 De ger in the text 3 Dinar the scene . alter the formation st, was to nt of this ate neigh- soil would stimulus lave been n side of with the introduce as finally ANGER. HERIPPIDAS, HARMOST AT THEBES. A. THE THREE HARMOSTS IN THEBES. In Plutarch’s two narratives (Vita Pelopidae XIII. and De genio Socratis) of the recapture of the Cadmea by the Thebans, 379/8 B.c. (winter), he speaks of three harmosts as in command of the Spartan garrison. This is the only instance in Spartan history where more than one harmost is mentioned as exercising authority in the same city, and it suggests the question: Was Thebes for some reason receiving different treatment from the other cities where we hear of harmosts in residence ? The names of these harmosts and their respective fates are told by Plutarch. They were Lysanoridas, Herippidas, and Arcissus. Lysanoridas, who seems to have chief authority in Thebes itself at least, had happened to be absent from Thebes at the time of the coup. The other two harmosts apparently took command and waited for Lysanoridas’ return. But later they capitulated and were allowed to retire. In consequence of this all three harmosts were court-martialled. Lysanoridas was fined very severely ; Herippidas and Arcissus were executed. Diodorus Siculus (Bk. XV. xxvii.) confirms some particulars. He knows of the three commanders (jyeyoves THs g¢povpas) and their fates, but he does not call them harmosts nor give their names. Xenophon, our only other authority, merely says tov pév adppoorny Tov éykaTadiTovTa THY axpéTokw Kal ovx dvapetvavta tiv Bonbeav atéxtewav (Xen. Hell. V. iv. 13). This by itself might be taken to imply that there was really only one harmost in Thebes, if it were not for the emphasis laid on the identity of the particular harmost by the use of the article and participle. This suggests that Xenophon was distinguishing one harmost’s fate from that of the others; and this interpretation would harmonize with the accounts in the other authorities, except for the omission of the second harmost who was executed. But in any case Xenophon’s Hellenica is so full of strange omissions that the argument from silence when applied to him has little or no value. ? B. HERIPPIDAS—THERIPPIDAS. Nothing further is known of Lysanoridas or Arcissus.* Herippidas is usually identified with the only bearer of that name “known to us elsewhere: 1 Compare the form of statement (Plut. De gen. Socr.1.) Avoavopliday 6¢ rpirov abrov dvr’ éxelvov (Phoebidas) méuavres, and the whole tenor of V. 2 De gen. Socr. XXXIV. ; but there is a lacuna in the text. 3 Dinarchus (I. 40) is very rhetorical, and has been carried away by patriotic feeling. In many respects his account is historically untrue. * Unless, as Mr. Wade-Gery has suggested to me, Lysanoridas is to be identified with the Lysandridas of Theopompus (Hunt, frag. 233). Athenaeus (XII. 609b) quotes from the Fifty-sixth Book of the Philippica a reference to the beauty H. W. PARKE a Spartan general of some note. We first hear of him in Diodorus (XIV. xxxviii. 3). He came from Sparta, in 399 B.c. (probably as harmost : so Meyer, Theopompus, p. 118), to Heraclea, in Trachis, where he successfully settled a oraovs and subdued the neighbouring tribes. In 396 he accompanied Agesilaus on his expedition to Asia, and led an embassy to Tissaphernes (Xen. Hell. III. iv. 6). In 395 he was chief of the ctdpBovra (oi epi ‘Hpimidav), who relieved the earlier body (0: mept Avcavdpov) on Agesilaus’ staff (Xen. Hell. III. iv. 20). He is mentioned on several other occasions in the campaign, and commanded the Cyreans at the battle of Coronea (Xen. Hell. IV. iii. 15). He was later appointed navarch, probably in 391 (Xen. Hell. IV. viii. 11). After such a brilliant career it is remarkable to find him grouped as harmost with two unknown Spartans, not even as chief of the three. But another passage in Diodorus may cast light on more recent operations in which he had taken part. Diodorus (XV. xxx.), in an excursus explaining why Histiaea-Oreus, unlike the rest of Euboea, was hostile to Chabrias on his campaign of 377 B.c. (summer), tells that shortly before (@payd mpo trovTay tay ypovwyv) Neagenes had made himself tyrant of Histiaea by the help of Jason of Pherae. In consequence of his misrule the Spartans sent ‘ Therippidas,’ who after unsuc- cessful negotiations expelled the tyrant by force. Casaubon conjectured that this name, Therippidas, is a mistaken reading in our MSS. for Herippidas, whom he identified with the Herippidas mentioned elsewhere in Diodorus as commander at Heraclea. He does not appear to have connected him with the harmost at Thebes, but based his conjecture solely on the similarity of the names, apart from the historical relations of persons identified.’ Casaubon’s conjecture has been generally rejected ; but the identification of Therippidas with the harmost at Thebes has some historical probability. For the Spartan expedition to Histiaea must have taken place before the fall of the Cadmea; Sparta could not later have sent a force through hostile Boeotia, and there was no appreciable Spartan navy till just before the battle of Naxos.” This and the fact that it was ‘ shortly before’ 377 would point to 379 (summer) as a probable date for ‘ Therippidas’’ expedition: that is, at the close of the Olynthian war and before the fall of the Cadmea. Such an identification does not imply the necessity of emending our MSS. Diodorus is apt to vary greatly from our other authorities in the forms of his proper names. So the fault may lie with his authorities or in a slip of Xenopeitheia, the mother of Lysandridas, and adds that the Spartans killed her and her sister Chryse, Sre xai tov Avoavipliay éxOpdv bvra ’AvyeoiNaos 6 Baciteds xaracracidcas puyadevO jvac érolncev id Aaxedaiporlwy. This passage must, then, have occurred in one of Theopompus'’ many digressions. If the identification is correct it provides another reference to Lysanoridas’ condemnation, and goes beyond our other accounts in attributing it to Agesilaus. This is very probable in view of the setback which the 160 surrender of the Cadmea had inflicted on Agesilaus’ foreign policy. ' Casaubon, Polyaeni Stvategemata, Notae in Libr. II., Herippidas. Editio princeps, Lug- duni Batavorum, 1589. 2 In 376 the allies demanded the formation of a Spartan navy: Xen. Hell, V. iv. 60. 3 E.g. Philodicus (XV. xxvii.) = Ecdicus (Xen. Hell. IV. viii. 20); Hegelochus (XV, 84) = Hege- silaus (Xen. de Vect. III. 7), and many minor instances, of his is four thenes probab from * harmo: Pr Pepare they wv having Sparta vaguely (e.g. X can ac This c of Hist (Xen. J expand W!) answer applied the offi authori develop is a sec applied Fiv Xenoph he appli one ma} commal specific 1 Comp: that Herir to explain Plutarch ; pidas. Ty Hyper-Lax mos), neve inscription than the c! 2 Polyae his expuls of another Plut. Amai lodorus ost : so essfully panied »hernes ot Tept esilaus’ ions in a (Xen. | (Xen. nd him of the recent -Oreus, 77 B.C. sagenes ae. In unsuc- d that ppidas, orus as n with r of the ication ability. re the hostile . battle oint to at the ng our > forms a slip icted on Notae in os, Lug- 1ation of us (Xen. | = Hege- ly minor HERIPPIDAS, HARMOST AT THEBES 161 of his pen. Therippidas is not known elsewhere as a Spartan name.’ It is found once as an Athenian name in the form Therippides (one of Demos- thenes’ guardians, Dem. XXIII. iv.). Assuming the identity, then, it is probable that the Spartans on hearing of trouble in Histiaea sent Herippidas from Thebes to settle the disturbance. He appears to have installed a harmost there—at least, we hear of one in 377 (autumn) called Alcetas.? Probably at the same time Herippidas took for Sparta the islands of Peparethus and Sciathus. For in the expedition of Chabrias mentioned above they were retaken by Athens, and Diodorus (XV. xxxi. 5) speaks of them as having been reraypévas tro Aaxedatpovious (compare also E. von Stern, Spartanische Hegemonie, p. 80). It is true that Diodorus is apt to speak vaguely of Greek cities as subject to Sparta when they certainly were not (e.g. XV. xxviii..3 : of Byzantium, Chios, etc., in 378 B.c.). But probably we can accept his statement in this case, while there is no evidence against it. This conquest might, however, have been achieved by Alcetas, the harmost of Histiaea, after the departure of Herippidas, for he later had three triremes (Xen. Hell. V. iv. 56). In any case it shows that Sparta was prepared to expand her empire somewhat in this direction. C. HARMOSTS. What, then, does this suggest as to Herippidas’ position in Thebes? To answer this one must consider the meaning of the term ‘ harmost’ when applied to a Spartan official. Its most familiar application is when it describes the office of a resident governor, usually with a garrison, exercising some authority in a city outside Sparta. The harmost in this sense was especially developed to hold down the subject cities of the Spartan empire. But there is a second and quite distinguishable extension of meaning when the term is applied to the commander of an army operating abroad.® Five examples of this use can be quoted from Xenophon’s Hellenica, and Xenophon (who had seen service under Thibron, the very commander to whom he applies the title) is not likely to write untechnically. From these instances one may deduce that in its second significance ‘ harmost’ implied a military commander (not one of the kings), with a force operating by land in some specific area‘ and with a definite objective assigned by higher authorities. 1 Compare Poralla’s Prosopographia. The fact tion of date, except that the incident preceded that Herippidas is unique as a name would help the battle of Leuctra. He was probably there to explain its corruption ; similarly in MSS. of sometime between 405 and 376 B.c. Plutarch it has occasionally become Hermip- 3 E.g. Xen. Hell. Il. iv. 28 (Lysander), pidas. Two hundred and fifty years later the III. i. 3 (Thibron), IV. ii. 5 (Euxenus), V. ii. 37 Hyper-Laconized form ZHPIINMOZ (=Onpix- (Teleutias), V. iii. 20 (Polybiadas) ; Aeneas Tac- mos), never SHPIIIIIIAAS, occurs in Spartan ticus XX VII. 7 (Eudamidas). inscriptions (cf. C.J.G. V. 210, etc. ; not earlier 4 So Aeneas Tacticus, in the passage quoted than the close of the second century B.c.). above, speaks of Eudamidas (when leading a 2 Polyaenus II. 7; Frontinus IV. vii. 19, and force of 2,000 men, Xen. Hell, V.ii. 24) as 6 Aaxw- his expulsion, Xen. Hell. V. iv. 56. We hear vwv dpyoorhs éwmi Opdxns. He also speaks as of another harmost of Histiaea, Aristodamus, in a contemporary and a technical writer. Plut. Amat. Narrat, III, ; but there is no indica- M = SS Ee hey Se | nap Semen Sem nee — ee ~ aoe. SNEED * Sete Ci ere TELE ie aS «es I ee seasons SAAR WS CER ES Tia - ~~ + — a nn : : _ ne, saison vs = 162 H. W. PARKE This restriction of command corresponds to the localization of the first type of harmost in a particular city. In accordance with this we find Thibron transferred to another sphere of operations by the ephors, and in three out of the five remaining instances the harmost is mentioned in connexion with a special district.! D. HERIPPIDAS’ COMMAND. Was Herippidas sent to Thebes as a harmost of the second type? Plutarch seems to imply that he was sent along with Lysanoridas immediately after the original capture of the Cadmea in 382 (summer), but this may be due merely to the compression of his narrative. Herippidas’ liberation of Histiaea has been dated conjecturally to 379. If this is assumed as correct, one may suppose that he was appointed after the success of Polybiadas (who was a similar harmost appointed to crush Olynthus) had freed Sparta from commit- ments further north than Thebes. For Olynthus was made a subject ally, but we do not hear of a resident harmost there. So Herippidas will have been appointed in the spring of 379 to watch Spartan interests in Northern Greece. His experience of the neighbourhood in 399 at Heraclea and at the battle of Coronea may have helped to prompt the choice. His chief activity that year was at Histiaea, as told by Diodorus, and he retired to winter in Thebes towards the close of 379. It follows that one would expect to find in the garrison of the Cadmea some indications of the presence of this additional force of men in winter quarters; and, as it happens, some indications are forthcoming. For both Diodorus and Plutarch give a number for the Spartan garrison 1,500 men. This agreement assures us that there has been no recent corruption such as often happens to the numbers in our manuscripts. Also, as it is most unlikely that Plutarch was copying from Diodorus whom he does not quote in the Lives among his many authorities, we may assume that the figure goes back to an early common source (probably of the fourth century) and therefore is worthy of some credence.’ If, then, we take the 1,500 as resting on good authority, let us compare this with another instance of a Spartan garrison, so as to decide whether the total is typical of a garrison alone. The nearest parallel is offered by the 1 On harmosts compare also Kahrstedt, Griechisches Staatsrecht, p. 179, etc. Compare for a similar pair of meanings ¢povpd = garrison, and gpoupd (in ppovpday Epavov)= expeditionary force. In its former sense dpmoors is also applied to officials appointed by Athens (Xen. Heli, IV. iv. 8), Thebes (id. VII. i. 43), and Sinope (Xen. Anab, V. v. 19). There seems to be no instance of dpuorrhs in the second sense applied to other than Spartans. 2 The choice of likely sources lies between Ephorus and Callisthenes. Dionysodorus and Anaxis are hardly ever quoted, and were prob- ably little read. If Ephorus was the originator of the figure 1,500, he might have obtained it by reckoning a mora to each harmost (for he estimated the mora at 500, Plut. Pelop. XVII.). But this method of calculation is improbable, as the garrison included a large majority of allied troops (Diod. S. XV. xxvii. 2) who would never have been brigaded in Spartan morae. In any case Spartan morae were only kept abroad in very serious wars during this period. Yet Von Stern (Spartanische Hegemonie, p. 58, note 2) explains the three harmosts wrongly as mora commanders. The ultimate common source for both Diodorus and Plutarch is probably Ephorus, while Plutarch has also used Callisthenes. Com- pare Fried Bloch, Untersuchungen zur Plutarchus de genio Socratis, Miinchen, 1g10. garri indu garri: ravag city t like | twent dispa force I smalle finds » fact th for ex augme It govern separa might Chalci render Bi route if Histiae power chronol Pharsal implied the sy describe also sul some of during t leader acquaint ’lacovos from T disprove mention 1 Comp Unfortuna instance y harmost’s in war w it type hibron ee out with a type ? diately be due listiaea 1e may was a »mmit- lly, but e been Greece. attle of at year Thebes ~admea winter or both 90 men. such as unlikely . in the es back refore is >ompare ther the by the st (for he p. XVIL.). ‘-obable, as y of allied nuld never e. Inany abroad in Yet Von 3, note 2) y as mora source for y Ephorus, nes, Com- + Plutarchus HERIPPIDAS, HARMOST AT THEBES 163 garrison in Athens in 404 B.c. In both cases an oligarchic minority had induced Sparta to put them in power by military force, and had obtained a garrison from Sparta to reinforce their position. But Athens, in spite of the ravages of the Peloponnesian war, must always have remained a more populous city than Thebes, and the Thirty can never have had the high standing of men like Leontiadas, who had led one of the two parties in Boeotian politics for twenty years.’ Also the ’A@. od. (xxxvii. 2) is perhaps right in dating the dispatch of Callibius and his force after the occupation of Phyle when a strong force was seriously needed. However, apart from considerations which would lead one to expect a smaller garrison at Thebes than at Athens, it is surely remarkable when one finds the ’A@. vox. giving the number of the garrison at Athens as 700. The fact that Thebes had more than twice the Athenian garrison should in itself call for explanation, and the explanation here offered is, that it had been largely augmented by another force in winter quarters. E. SPARTA AND JASON. It remains to enquire what particular motive actuated the Spartan government to take such special precautions as to station a force under a separate commander to watch the country north of Thebes. One motive might have been that they wished to keep in touch with the land-route to Chalcidice. For though Olynthus had capitulated it might not have been rendered completely harmless.? But it is more likely that the danger lay nearer and threatened the land- route itself. The mention of Jason of Pherae in conjunction with Neagenes of Histiaea suggests the nature of the peril. Unfortunately the rise of Jason to power is shrouded in obscurity. This reference in Diodorus is the ftrst in chronological sequence, and the next is not till the embassy of Polydamas of Pharsalus to Sparta (Xen. Hell. VI.i. 2 seq.). The date of the embassy implied by Xenophon is 374 B.c., but Beloch has proposed to transfer it (with the synchronous expedition to Phocis) to the year 371 B.c. Polydamas describes Jason as holding ‘ the most and greatest cities of Thessaly,’ besides also subject-peoples like the Dolopes and Epirus. It is not impossible that some of this empire was already in his power by 379, and that the Spartans during their expeditions to Olynthus had occasion to notice the rise of this new leader in Thessaly. Polydamas assumes that the Spartans are already acquainted with Jason’s name at least—dxovete pév ody ed old’ Gre Kal ipeis ‘lacovos dvoya—and says it is his duty as proxenus to warn Sparta of danger from Thessaly. The fact that Xenophon does not mention them does not disprove earlier relations between Jason and Sparta, since he does not even mention the expedition to Histiaea. Further collision between the two powers 1 Compare Hellenica Oxyrhynchia XII. i, 2. considerations for which we cannot make due Unfortunately Athens seems to be the only other allowance. instance where we know the numbers of a 2 In 377 the Olynthians were able to send a harmost’s garrison in peace times. Instances forceof cavalry to help Thebes against Agesilaus in war were influenced by strategic and other (Xen. Hell. V. iv. 54). y if it 4i an it in} i Hy ut ' wis 4 ai 11 vig § ae i ig 4 ne ni By A a} ait a + & } % ; : § - aig at ota ii a} i 4 . i 4] ae a ie if af heh } iy 4 ua a " + By Jot Sheet 2 es Se RE Ag hk per i ae ee ag te 164 | H. W. PARKE had been prevented in the interval, because the revolution in Thebes cut communications and the consequent war occupied Sparta’s forces so much that she could not even help Polydamas. Jason, for his part, was otherwise engaged in developing his empire. The action of Jason in securing a confederate as tyrant of Histiaea suggests an interesting historical parallel to the expansion of Macedon. Philip appears to have learnt much from the career of Jason, and to have followed a similar line of development. We find that he also got possession of the northern part of Euboea by the use of tyrants as his tools. Hence in 349 Athens, which had taken the place of Sparta in relations with Northern Greece, was compelled to send Phocion on a similar expedition to that of Herippidas. Geographically Histiaea, since it commands the mouth of the Pagasean Gulf, was of more vital and earlier importance to Jason than to Philip, especially as Jason had an ambition for sea-power (Xen. Heil. VI.i. 11). It seems as though he expanded first north-west and south-east, and after his failure at Histiaea he concentrated chiefly on reaching north-west to the Adriatic before he com- pleted his control of Thessaly by threatening Pharsalus. F. CONCLUSION. To sum up, the presence of three harmosts in Thebes seems possible of explanation by assuming that one (Lysanoridas) was the regular harmost of the garrison; a second (Herippidas) was a harmost in command of a field force: as for the third (Arcissus), we have no information. Perhaps he was another leader of a field force or, more probably, he was the acting harmost of the garrison in Lysanoridas’ temporary absence. By adopting the latter explanation of Arcissus’ position one can obtain an interpretation of Xenophon’s narrative consistent with our other suggestions. Xenophon mentions only one harmost, and says that he suffered the death-penalty. This, then, cannot be Lysanoridas. If we suppose it to be Arcissus the omission of Herippidas might be explained thus. Xenophon had served under Herippidas (who had been commander of the Cyreans), hence he may have chosen to omit from the Hellenica all reference to his old commander’s last campaign which ended in disgrace and death. Many of the gaps in the Hellenica are best explained on the assumption that Xenophon chose his subjects and method of treatment from personal and not historical motives; one of these was to omit what it would pain him to record—e.g. probably the battle of Tegyra as a Spartan defeat. Further, he suppresses the name of the harmost executed just as he suppresses the name of the leader of the mora annihilated at Lechaeum (6 éxel hpovpav moréuapyos, Xen. Hell. IV. v. 11).? This account seems preferable rather than to explain the harmosts as forming a collegium of three, for this is without any certain parallel among 1 Compare Plut. Phocion XII., rapadvopévov eis same district compare again Athens in 403: riv Ed3oav rod Pidlwov . . . kal rds wéders while Callibius was still harmost of Athens (’A@. oixecoupévou dia Tupdvywy. wor. XXXVIII. 2), Lysander was sent as harmost 2 For the presence of a harmost with a and his brother as navarch to help the Thirty garrison and a harmost with a field forceinthe (Xen, Hell. II. iv. 28). Spa (Gr. this only milit who wher elect incap and | isolat But i ment to the was € seems a cup limiti comm case o B Arciss accord have s (comp: Also tl unjust the res Sparta Arcisst togethe TI for thi Agesila Wat 1 The stances i: was supr but from be safel navarchy Baoirela. So called ( difficulty ss cut h that erwise istiaea cedon. » have sion of in 349 rreece, ypidas. 1 Gulf, ally as hough istiaea e com- ible of 1ost of a field 1e was armost - latter phon’s s only cannot ippidas ho had t from ended plained atment t what ypartan t as he ‘+haeum osts as among in 403: ens (’A@. ; harmost 1e Thirty HERIPPIDAS, HARMOST AT THEBES 165 Spartan military officials and quite unknown among harmosts. Kahrstedt (Gr. St., p. 166) says that two are merely cip~Bovro to the real harmost; but this contradicts the only authority who gives them a technical title, and the only known instances of cipBovroz, explicitly so called, are men acting on the military staff of youthful or inexperienced Spartan kings, and with navarchs who had proved unreliable.! But Kahrstedt’s theory of c¥uSovro is founded on Thuc. IV. xxxviii. 1, where we learn that there was a Spartan voyos that three Spartan officers were elected so as to command successively, in case the first or the second was incapacitated in turn (compare for another actual instance Thuc. III. c. 1, and cix. 1). This procedure may have been adopted on all occasions of isolated commands though Thucydides does not make this perfectly clear. But it does not seem to bear any relation to the office of c¥pBovdo elsewhere mentioned, for there is no indication that the junior officers acted as advisers to their commander. On the other hand, there is no proof that a cvpBovdos was expected to succeed—e.g. on the death of a navarch. The ésrictoXevs seems then to have taken up the command automatically: but he was not a avuBSovros, at any rate in the three definite instances cited above. For the limiting commission of c¥p~BovrAo were not sent out till the navarch had held command for some time, while his émicrodXeds (his prospective successor in case of his death) must have been appointed at once. But even apart from Kahrstedt’s theory about svp~Bovros Herippidas and Arcissus cannot be explained as the junior officers of Lysanoridas, appointed according to the voyos referred to above. For then only one of them should have succeeded till he was incapacitated. But they both commanded together (compare the use of the plural continually in Diodorus’ and Plutarch’s account). Also the Spartans considered both deserving of the same penalty. This seems unjust if one was only a subordinate officer to the other. It may have been the result of a momentary exasperation which distorted the judgment of the Spartan authorities. But it is much easier to understand if Herippidas and Arcissus were each the commander of a distinct force temporarily placed together in Thebes. The suggested explanation fits in constructively with the data available for this period and provides another instance of the imperialistic policy of Agesilaus. H. W. PARKE. WADHAM COLLEGE, OXFORD. 1 The reason for cjuBovko in these circum- year of office had proved himself unreliable. stances is clear. A king on the field of battle Probably to depose a navarch was difficult, so was supreme by Spartan véyos (Thuc. VI. 66, 3), he was restricted by one or more cipBoviu. The but from youth or inexperience he might not essential characteristic of a harmost’s office was be safely trusted to his own discretion. The that he was already limited: if he proved incom- navarchy also was, as Aristotle says, cxeddv érépa + petent he could be superseded easily (e.g. Bacirela. In all these instances of otywBovko. Thibron, Xen. Hell. III. i. 7). cdpuBovdoe were so called (Thuc. II. 85, III. 69, VIII. 39) the same only attached to limit those with absolute difficulty had arisen. The navarchearly in his command. —— = wr Sans porany SS eae { | ( 3 # i # fi 4 K d | | ty 4} i} Hae ti ay it ey ane = po THE TOLEDO MS. OF PLUTARCH’S MORALIA. It seems worth while giving some account of this MS., because in recent years it has been said to contain certain pieces of the Movralia which it does not, and not to contain others which it does.! At the foot of the first page of the text the MS. carries the pontifical shield with the arms of the Medici, and from this it is reasonable to suppose that there was a time when it belonged either to Leo X., who was Pope from 1513 to 1521, or to Clement VII., who was Pope from 1523 to 1534. This being so, 1534 is established as the teyminus ante quem of its writing. Towards the end of this century, about 1598, it was acquired by the Spanish Cardinal Saverio de Zelada. In 1801 the Cardinal Archbishop Antonio de Lorenzana bought in Rome a certain number of Greek and Oriental manuscripts which had formerly been in the collection of Cardinal Saverio de Zelada, and placed them in the Chapter Library at Toledo.? Included with these was this MS. of the Moralia, which remains there to this day. Seven years later, in 1808, was completed the manuscript catalogue of the Chapter Library by Padre Lorenzo Frias, in which the Plutarch MS. is entered thus: ‘ Plutarco: sus obras morales en griego: un tomo fol. de vitela. Siglo XIV. Z-51-5.’ But it had to wait twenty-two more years before Gustave Haenel gave it record in print,* and he dismissed it with the simple entry ‘Caj6én 51. No. 5 Plutarco, obras morales en griego; saec. XIV. Membr. fol.’ Indeed, he gave no more than a reproduction of the earlier catalogue, and no one had any fuller knowledge of the MS. until forty-five years later M. Ch. Emile Ruelle published a description‘ of his explorations in certain Spanish libraries, and included an account of the Chapter Library at Toledo. He draws up a list of the Greek MSS., in which the Plutarch MS. stands twentieth. In the course of his observations he rightly rejects as ‘de tout point inadmissible’ the dating of the MS. in the fourteenth century by the early catalogue, but himself dates it ‘vers la fin du XVI° siécle,’ which is made impossible, as already explained, by the evidence of the pontifical shield with the arms of the Medici. M. Ruelle’s list of the contents of the MS. is correct, except that he says that the IoA:rixa rapayyéAuara come first instead of fourth. The scribe had wrongly written at the beginning of his index [loActixa rapayyéApara as the title of the first treatise, but he saw his mistake afterwards and marked it with a 6’, which seems to have escaped M. Ruelle’s notice. It seems also to have escaped M. Ruelle’s notice—for he does not record the fact—that the scribe noted in the index that of the IlAarwvixa (ntipara ‘ovx eipéOn 4 dpyy.’ The first three (yrjpara are not given, and the piece begins with the first words of the fourth (jrnya, Ti Syrore, ryv Yvxnv dei rperBurépay aropaivwy Tov cwparos. .. . 1 I take this opportunity of thanking Canon Eduardo Estella, the Archivist of the Chapter Library of the Cathedral, for such help as he gave me when I was at Toledo. I am also very grateful to Dr. E. H. Minns for much kind assistance. 2 The most recent reference to the coming to Toledo of these MSS. gives the date as ‘ towards the end of the eighteenth century.’ But there is evidence for the precise date given above. 3 Catalogi Librorum Manuscriptorum, qui in Biblio- thecis Galliae, Heluetiae, Belgit, Britanniae M., His- paniae, Lusitaniae asseruantur, nenc primum editi a D. Gustavo Haenel, Lipsiae, 1830, col. 993. 4 Rapports suy une Mission littévaive et philo- logique en Espagne, par M. Ch. Emile Ruelle, in Archives des Missions scientifiques et littéraives. Troisiéme Série, Tome II., Deuxiéme Livraison. Paris, 1875. 5 Op. cit., p. 586 and pp. 590-592. M. I give with th I ma es Bw MR He He ef. lhU6e.l CU CURD OR OC ee fee ee ee cs am Cad Cad @2rs €ads @€ad Pan 023 C2 Add fn tr e- er rd Th index a are ent ex (nT npc T Pog wre t years not to d with re was , or to blished t 1598, irdinal ak and averio 1 these of the ntered » XIV. | gave No. 5 ive no fuller shed a ccount which ‘ightly ‘teenth which d with except scribe e title which uelle’s of the n, and poxiy | Biblio- {., His- editi a t philo- elle, in évaives. raison, G. B. A. FLETCHER 167 M. Ruelle gives the titles of the various pieces of the Movalia in French, and I give here the Greek titles as they are written in the index of the MS., together with the pages on which the treatises occur. Page. 23 3 g 18 63 83 128 157 162 167 170 180 193 216 217 220 223 251 255 258 265 268 269 275 290 3°95 315 323 333 344 358 360 376 416 437 462 465 501 515 There The index is on pages one and two: Title. Number in Index. ‘ 4 5’ ToATiKG TapayyeApata IN7 ~ , TapapvOntiKds mpds Loiav yuvaixa a. éx Tov mérepov AOnvaior kara wodepdy 7) Kata copiav évdo£drepor & mepi Too Ste paduorra Tois Hyeudor Se Tov prrdcogov SiareyerOae = -¥ , , ; , et tperButépw TroActrevTeov € , . ee ? mept tridos Kai ooiptdos s a , yuvatkov aperai ¢ , Tpds nyewova araidevTov n tTepi capkopayias Adyos a” g s , , 4 , Tept capxopayias Aoyos B U va TEpl eiwappevns 2 , rp > es eS ‘ traTwviKka éex(nripata Gv ovx evpeOn 7 apx7 Tepi fOVvo“KNS civoyis Tov dt Tapadofdrepa of oTwikol TMV Trountav A€yovrtv eriTopn THS TVyKpliorews dpurtopavous Kat pevavdpou €TLTOpMy) TOU TEpl THS EV TO Tipaiw Pvxoyovias oO > 4 ~ ” € , - 4 Ste ovde (hv Eorw ydews Kat’ EriKoUpoV ei adds eipnras To, AdDe Budoras 5 > , e 4 . , ei avTadpkys 7) Kakia mpos Kakodatmoviav TEpt THS eis TA ExYOva pirorropyias wept POdvov Kai pious 4 ’ 4 , , . > , Tepi povapxias Kal Syuoxparias Kai oAvyapxias epwtixat Sunynoess > , 4 aiTias PvTgTiKat Two-thirds of page 288 blank. Page 289 is blank. Tept Tov dperkdvrwy THs (sic) prrAordpas BiBriov «a Ere wept Tov avrov BiBXriov PB’ ért BuBXAiov vy’ ért BiBXriov 5 €rt wept Tov avTov BiBriov €° Tept ToU TpwTov Wuxpov Ore Sidaxrdv 7) apery Tepi mapadAnAwv EAANViKGY Kal pwyatKaV keparaiwy kataypapy pwpacky EAAnvixa Page 421 is blank except for ov8ev deires written across tt, , ~ 4 .4 4 Biou trav Séxa pnrépwv / c ‘ 4 Ynpicpara vréep SnporGevovs TTWLKOV EVAYTLWPBATWY rept Tov ec év SeAqois cvprociaxav BiBria 6’ & - & Sez oO a, ~~ “ ~ SA Tf re’ As’ Ag An’ AY are no divergencies between the titles of the treatises as given in the index and as given at the beginning of each one separately, except that the following are entitled thus in the text: mapapvOnrixds mpds tHv idiav yuvaixa, tAaTwveKxa ex(ntihpara Sv ody ebpéOn 4} dpxy ({yTjpara is written in the margin), ra rod diaAdyov mpdcwra dvnoipdrns, swrnpixos Avoias wept povorkys, wept TwV dperKdvTwy Tois dtAo- 168 G. B. A. FLETCHER godos BiBAiov mp@rov, epi Tav aperkdvTwv Tois pirogddas, Pvotxov Soypatwv éxiTopAs, BiBriov B’, rept trav dpecxdvrwv rois diroaddots y’, wept TOV aperKdvTwr Tots PiAowdgors BiBAiov 8, repi tv dperxdvtwv Tots piroaddors €’, keparaiwy xataypadiy}, rept Tov e TOD ev SeAgois, tAovrdpxov ouprociaxav BiBréia 0 év tO a’, TAOVTApXoU Gvproc.aKkav SedTEpov (P. 539), tAovrapxov cvpmrociakav y’ (p. 562), tAovrTdpxov ovprociaxav & (p. 579), Tade eveoTivy €v TO TéurTw TOV TupTogiuKov (p. 604), Tade EoTLV ev TO s TOY TUPTOTLAKOY (p. 623), rade eveorw ev ro ( trav cvprociaxev (p. 640), cvprociaxdv dydoov (p. 668), wXovtapxov cvprociaxav 6 (p. 697). The last page of the MS. is no. 715. M. Ruelle records the note at the foot of the first page of the index, A€tz[ovcx] Aoyo. pf’, ‘il manque 47 discours.’ He continues, ‘cette note prouve que son auteur portait 4 81 le chiffre des traités divers attribués 4 Plutarque (81-34=47). Les derniéres éditions de Plutarque n’admettent que 79 articles.1 I] n’y a pas d’autre conjecture a tirer de cette divergence, sinon que l’auteur a considéré comme authentique deux écrits refusés aujourd’hui 4 Plutarque...’ There is no reason for saying this. The index, as can be seen above, has 39 entries. If this is the number which the scribe subtracted from the total number of the pieces of the Movala known to him, then that total was 86; if he subtracted the actual number of pieces in the MS., it was 79—unless, in both cases, he made a mistake in his arithmetic. To say that he subtracted 34— i.e., that he thought he had two more entries in his index than there were pieces in the MS., rather than one, three, four, five, six, or, as was the fact, seven—is to say something for which there is no evidence. M. Ruelle brings his comments on this MS. to an end with the words, ‘il est a souhaiter qu’on n’entreprenne pas une nouvelle édition de Plutarque moraliste . . . sans l’avoir mis a contribution.’2 In 1878 M. Charles Graux made a brief reference to M. Ruelle’s article published three years previously, and to the Chapter Library at Toledo in a Rapport suv une Misston en Espagne,® which calls for no comment. Fourteen years later another work of M. Charles Graux was published with a detailed account of the Toledo MS. of Plutarch’s Movalia, 51. 5.4 The conclusion he draws from the note at the foot of the first page of the index is that ‘dans l’édition a laquelle se rettache le MS., on compte donc quatre-vingt-six livres de Moralia; ce chiffre se réduit en réalité 4 quatre-vingt-un, puisque le traité de esu caynium est compté comme deux livres, et le de placitis philosophorum comme cing.’ The number should be 79, because the treatise Aetia Romana et Graeca and the treatise decem ovatorum wuitae are both counted as two, as can be seen from M. Graux’s own list, where they are numbers 33-34 and 35-36 respectively. In 1905 Victor Hahn published his extremely useful De Plutarcht Moralium codicibus quaestiones selectae at Cracow, in which he gives brief accounts of 240 MSS. of the Movalia.’5 This is what he says of the Toledo MS.:* ‘Codex Toledensis, numero 20 a Ruellio significatus, s. XIV. secundum Haenelium, s. XVI. secundum Ruellium, continet 32 Moralia hoc ordine: 30. 24. 6. 49. 35- 27. 33. 28. 36 I. 37. 36 IT. 44. 66. 76. 38. 44. 73. 46. 13. 12. 58. 71. 50. 51. 53- 69. 61. 62. 65. 63. 74. 32. 25- 1 The important point is that the number in ordre et complétées par Albert Martin, Paris, Parisinus 1672 is 78 by our reckoning. 1892, pp. 264-266. 2? Bernadakis made no use of it. It still 5 Op. cit., pp. 43-128. The divergence of my remains to be seen whether the new Teubner numbers of the page-references to Hahn’s work editors will do so, because the MS. happened from those given by some other writers is not to contain any of the pieces included in their explained by their depending on whether they first volume published in 1925. are taken from the work as a separate piece, or 2 In Archives des Missions scientifiques et litté- as bound in one volume with other works pub- vaives. Troisitme Série, Tome V., Premiére lished during the same year under the same Livraison, Paris, 1878, pp. 111-136. auspices at Cracow, having its pages numbered 4 Notices sommaires des Manuscrits grecs d' Espagne accordingly. et de Portugal par Charles Graux, mises en 6 Of. cit., p. 71. Ruelliu: A of my ’s work iters is er they iece, or ks pub- e same mbered THE TOLEDO MS. OF PLUTARCH'S MORALIA 169 Ruellius Morala tralaticio ordine editionum enumerat.! Fol. 1. exstat notitia : Acir[ over] Adyou pf’ (= 47), erant igitur codicis librario 79 moralia nota (32+47)... . Adhuc non collatus.’ In the first place it is not easy to see why Hahn troubles to mention that the MS. happened to come twentieth on M. Ruelle’s list of Greek MSS. in the Chapter Library at Toledo, more especially because he does not give the library reference mark at all, although Haenel, Ruelle, and Graux had all done so before him. But more inexplicable is the fact that the numbers given by Hahn do not accurately represent the contents of the MS. One would suppose them to be the numbers by which scholars, including Hahn himself, ordinarily refer to the Moralia— that is to say, the numbers formed from the order in which they are given in the MS. known as E (Parisinus 1672). And it seems that this really must be the case, since they come no nearer to giving correctly the contents of the MS., whether they are taken to be the numbers given by Lamprias, those in the order of the Didot edition, or those in the order of Wyttenbach’s edition.? It simply has to be accepted that, for some reason unknown, the list is inaccurate, though the error is astonishing when it includes, as it does, the statement that the MS. contains one piece (44) twice, which it actually contains once.2 That Hahn’s statements about this MS. are confused can be illustrated by one example. : It is well known that 76 is one of the pieces of the Movalia which are preserved only in B (Parisinus 1675) and in E (Parisinus 1672). Now 76 is one of the pieces said by Hahn to be contained in the Toledo MS. If it were contained in it, he would be justified in writing, as he does,‘ of E, ‘In s. 70-76 constituendis hic fere unus codex fundamentum textus praebet ’; and he is equally justified in so writing when, as happens to be the truth, it is not contained in it. But having said on one page® that the Toledo MS. contains 76, he is guilty of contradicting himself when he says on another,® ‘ praeter codicem E solus B s. 70-77 exhibet,’ and on another,? ‘commodissime res in s. 70-76 comparata est, quorum tantum duo fontes Parisini B et E exstant, qui codices quamvis unius archetypi apographa sint, tamen inter se minime consentiunt.’® In May, 1924, Dr. J. B. Titchener, of the University of Michigan, having obtained photographs of the Toledo MS. for 62, was the first scholar to publish any of its readings. He was concerned with the relationship of the extant MSS. of 62, and came to the conclusion that the Toledo MS. was a descendant of the MS. he called R2,!° which is Vaticanus 139, and designated y by the new Teubner editors. He shows no acquaintance with the works of Ruelle and Graux, and, apart from his photographs, his knowledge of the MS. rests only on a brief note sent to him by Canon Estella and on the account given by Hahn, part of which he quotes without doubting the truth of it. It was fortunate for him that 62 chanced to be both in Hahn’s list of numbers, and contained in the MS, He refers to the MS. in his 1 M. Ruelle adds to the title of each treatise Hahn twenty are correct, eleven are wrong. He its number in the ‘ordre de Lamprias’ and its _has only thirty-one numbers, because, as stated, number in the ‘ ordre de |’édit. Didot.’ he gives one number twice. 2 It looks very much as if this is true, because 4 OP. cit., p. 51. 5 Op. cit., p. 71. in Hahn’s list there occurs a piece in two 6 OP. cit., p. 48. 7 Op. cit., pp. 91-2. — ee on ~ Se divisions; its number is given as 36, which is Ilepi capxopayias in the order of E, and this piece is actually one of those included in the Toledo MS. But the argument is not decisive. For there is one other piece of the Moralia in two divisions—57, Ilepi ris "AXetdvdpouv rixns 4 dperjis. However, argument is hardly needed when there is no reason whatever for supposing that Hahn’s numbers are not meant to refer to the order of E. * It may be said that of the numbers given by ® To save confusion it should perhaps be pointed out that in the following paragraph on p. 92 ‘Toledensi’ appears to be an error for ‘ Escorialensi.’ 9 J. B. Titchener, The Manuscript Tradition of Plutarch’s Aetia Graeca and Aetia Romana in University of Illinois Studies in Language and Literature, Vol. 1X., No.2. University of Illinois Press, May, 1924. 10 See op. cit., pp. 25-7, and the stemma on p. 64. aa or ~- 170 G. B. A. FLETCHER description and in his table of Sigla as Toletanus 20, which means that he understood from Hahn’s words ‘Codex Toledensis, numero 20 a Ruellio significatus,’ that Codex Toledensis (or Toletanus) 20 was its proper reference mark. From what has been already quoted above, it is obvious that the proper title of the MS. is Toletanus 51. 5, which means, of course, that it is the fifth book on the fifty-first shelf. Toletanus 20 _ would be a title without any meaning. In November, 1924, Dr. C. G. Lowe, of the University of Yale, did for 63 what Dr. Titchener had done for 62, and, as part of his examination of the extant MSS. of 63, published from photographs readings of the MS. at Toledo.t His knowledge of the MS. is no greater than Dr. Titchener’s, except for what he learns from his photographs, and shows no acquaintance with the works of Ruelle and Graux or Suspicion of Hahn’s accuracy. And, like Dr. Titchener, he wrongly gives the reference .ark of the MS. as Toletanus 20. The conclusion he reaches is that for 63 the MS. has the same parent as Dr. Titchener holds that it has for 62, and for both pieces it has two brother MSS. Now these three MSS.—the parent and the two descendants—all give their pieces of the Morvalia in the order of the Planudean Corpus. But Dr. Titchener, having been misled by Hahn, says* ‘T has thirty- two... essays in a different order’; and Dr. Lowe, similarly misled, says* ‘ T has changed the order of the essays found in its archetype without any reason, since there is no indication of any attempt of systematic arrangement.’ Both these statements are untrue, because the actual order in T is 23. 27. 28. 58. 30. 32. 33. 35- 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 53. 55- 61. 62. 63. 66. 68. 78. and 58 may be presumed to be out of place through some accident in copying. T has been variously dated. The fourteenth century and the end of the sixteenth century are impossible, the former by reason of the character of the writing, the latter by reason of the evidence of the pontifical shield with the Medici arms. M. Graux dates the MS. as ‘du XVI siécle.’ Dr. Titchener accepts in one place* Canon Estella’s date of the end of the fifteenth or beginning of the sixteenth century, but in his table of Sigla® puts ‘s. XV.’ Although Canon Estella’s date is perhaps as exact as one has any right to venture upon, I am inclined to follow Dr. C. G. Lowe in dating the MS. in the latter part of the fifteenth century. It seems to me rash to say, as Hahn does, that, because Aéiz[ovor] Adyor pf" is written at the foot of page one, ‘erant igitur codicis librario 79 moralia nota.’ In such cases scribes are very prone to make a mistake in their arithmetic, and since in the index 36 is entered as two pieces, 51 as five, 62 as two, and 63 as two,a mistake would have been easy for a scribe who was trying to get the right number of pieces by allowing for these facts, but ended in allowing one too many. But it has to be remembered that scribes had no fixed way of deciding what was and what was not an independent treatise in the Moralia, as is shown, for example, by the variation in the numbering of the works in Parisinus 1672 and in Vati- canus 139, a fact which makes the statement in Toletanus 51. 5 valueless, but supplies another possible explanation why the scribe thought 47, not 46, treatises were missing in his MS. Vaticanus 139 is the probable ancestor of Toletanus 51. 5,° 1C. G. Lowe, The Manuscript Tradition of manuscript. He finds that T is in 66 derived Pseudo-Plutarch’s Vitae Decem Ovatorum in Uni- from Vaticanus 139, and that any intermediaries versity of Illinois Studies in Language and Litera- there may have been do not include any of the tuve, Vol. IX., No.4. University of Illinois Press, other MSS. known by him to be descended from November, 1924. that MS. This agrees with the results of Dr. * Of. cit., p. 25. J. B. Titchener and Dr. C. G. Lowe, and the 4 Op. cit., p. 15. three results together, combined with the Planu- 5 Op. cit., p. 67. dean order, make it as near certain as it is 3 OP. cit., p. 33. 6 I have to thank Mr. F. H. Sandbach for possible to be that T is descended from Vati- very kindly reading this article when still in canus 139 throughout. and in 36, 57, correct] 60 wron number In | have all and, wh knowled MS. is | difficulty The pag Ile Gat.—l obv, 14 p 17 erie) om., 7 € 12 péxpt II mavAa 5. 5 ov é1 10 Toray 24 mpood IO vorovc 0 avrd ai 4 kal o1 10. 5 emcee et 7 18 peta 23. I ayy eira THV O 21 no la 6 d€ om., 21 reve, Ig 70, 23 13 aAvy, | om., 2 « Ig yevao 16 idtoy, 1 om., 6 &v. 21 dexope 18 €iou, 2 23 avomrre 16 avrois, om., 24 € Aeipew + 39. 7 € x6, 13 €vyéav 42. 13 tBu te (Bach 44.5 pave yiver Oar, 2 aT pas, 21 « derstood at Codex has been US 51. 5, tanus 20 63 what MSS. of owledge from his yraux or ives the ; is that and for and the anudean s thirty- 3*'T has yn, since th these + 33- 35: . 66. 68. ying. of the r of the > Medici 'S in one ixteenth s date is »9 follow you pt’ is ita.’ In id since 5s two,a number y- But hat was xample, in Vati- ess, but treatises iS 51. 5,° 5 derived mediaries ny of the ded from ts of Dr. and the he Planu- ) as it is om Vati- THE TOLEDO MS. OF PLUTARCH’S MORALIA 171 and in that MS. 66 is numbered 70, the works counted as two being very likely 36, 57, 60, 62. The scribe of Toletanus 51. 5 may have counted his treatises correctly, but, misled by a list he had which rightly gave 57 as one work and 60 wrongly as two, have arrived at the number 47 through supposing that the total number of the Muralia was 79. In conclusion I give my collations of 28. 30. 35. 36. 40. 41. 44. 48. and 66. They have all been made with Bernadakis’ text, the references are to his pages and lines, and, where I have noted the coincidence of the MS. reading with a conjecture, my knowledge of the conjecture and its author comes from his apparatus criticus, The MS. is finely written in a neat and regular hand, which hardly ever presents any difficulty to read. Occasionally it has a note written in the margin by a later hand. The page measures 31°9 x 21°7 cm. Ilepit rov dre parcora trots nyepooe Sei rdv PirScoghov HudrA€Eye- Gat.—l. 4 re percévae (yp, TYsav Kai pervevas in marg.), 7 xal before piAoKdAwv, 13 «t for otv, 14 wy for por. 2, 3 dherdrev eireiv, 6 yiverOat, 7 éAuvvdovra, 8 Evrawra, 10 érctiOnon, 17 eripeAnOnoerat, 19 pidocdpew for cwdhpdve, 24 ‘cvBotixyv. 3. 1 devvdous, 3 peyaArov om., 7 €yyevouévns (Duebner), 9 drodatvew, 10 7d éptyywov td Bordviov, II mpwrov, 12 pEXpls, 13 dawrdppiat, 24 Swwve twAdTwv. 4. 1 oTpateias oxyriwv, 6 Kdétwv, 10 péev OM., II mavAov Om., 12 oxynriwvos Tov om., 13 Siadé~wpar, 14 no lac,, 15 evduaGero, 19 av OM. 5. 5 ov evaiowos, II Kai om., 16 mporrdAov, 18 xareyéety, 20 év dtadexopevors. 6. 8 Ev O:m., 10 tovavrns om., 11 BovAopévois, 13 yap OM., 14 py} OM., IQ ovTos, 23 SuwKewv hevyew, 24 mpoodvra petpiw. 7. 2 Kal wpatoys Koopious, 7 pirooddev for directv PoBnOyncera, 10 vorovew ayaba ypnyévors, 14 axovev, 18 oreipwv, 20 prdoyewpyos, 21 Hv OM., 23 Hv for 0 avrd ab mddw dvypykas ci oe 5. 8. 19 A€youow, 20 od eipye. 9. 3 arecrupravifor, 4 kat om., 8 mporvveis, 11 ws SiorKdrrar, 12 ayaa, 20 dvnp for dv, 22 OnBaiov. 10. 5 érvecxns, 17 madivnorov, 20 Spopaiovs. ci rpeoBurépw worAtrevréov.—2Zl. 4 é@ avrg, 6 eis, 16 drodeiyy, 17 Kowny, 18 peraSdrnrar, 22.7 xaOds, 12 éyxaraBwoas, 14 adrdv, 21 éEirnAdrepa, 24 Tov OM. 23. I dynpw, 3 twror’ om., 4 oldev, g idias OM., 10 THs OM., 14 KaAapyrtpias, 20 ovr’ av for elra tiv oupBeByxviav, 24. 1 ope p’, 4 od piv, 5 petadrAAdrrev, 8 dydvwv, 11 S0fy Karas, 21 no lac. 25. 5 éOopvByOncav eirev, 12 aye, 19 cuppdxous, 21 Td Téppa, 26. 4 Kat, 6 d€ om., 8 roAAGy, 10 KoAwvod, 12 éx’ avAG, 13 EvOa, 14 puvvdperas, 20 Td Errypappariov, 21 revfe, 22 wévre cal, 27. 2 récoapow, 6 eLavurrapévovs, 12 9, 18 ayopas woAcTeias, Ig Td, 23 Twa, 25 tpodas. 28. 3 ddpodiriav aris, 8 otrws, 12 AEvKoyAAOv jv cizev, 13 dAvv, 16 dtAoypyyariav (yp. in marg. Pirotysiav), 22 é£noev, 24 cwopadtwv. 29.1 Tas om., 2 dmrnuBdrvpevas, 3 ériOryyovor corrected from émnOyyovor, 8 ynpwPookeirat, Ig ywwoKes, 27 attra ot orpades. 30. 2 ofov, 8 érapivivdas, 12 KaPypas, 15 / > ayKaAas 16 avrov, 20 5€ om. dv, 17 Tw ov pavupa. ov avdpds. cAevTa OM. > ~ ee oe EKEL OUKET , 20. 5 kai, Bpvxopeva, cat before IT PATOVVTLs x DV, 7 popa, Hv, 18 Tas, 3 wpas for }7av Tapo- ;, 6 amroKo- y evTovoia, > , ameAavety, 5 KaXdv ov , pewpay lat, var eret 4 om., 7 Kai ) for zepi. “~ 7 tov dovov’ 15 €avrovs, \axrivovres, , evov Tie, ” e 5 ovuTo-— Ig Te OM. 0 lac. aAAa 116. 1 xa vyyiver Gas, '. 10 Hvvov, Tws mpos TH THE TOLEDO MS. OF PLUTARCH’S MORALIA 173 cvvowes TOU Ott mapadosdstepa of gTTwikol TOY TOLNTaYV AEyovG YD. —280. 19 ov@év. 281. 1 ddpavots, 3 after yeyovev from 280. 13 6 d€ rav Lrwcxey to this point is repeated with the same readings as before, 4 €x@urros, 6 cal om., 7 xai ddvvoomdbos, 13 mpoeritiOevros, 14 mpoertywoavTos, 17 amodureiv, 18 ovTws, 20 peTa- BardrAwow. 282. 11 edfar et rT, 17 dvayK@v, 22 dveivar, 23 pera Twv for perarTor. émiTouyn THS TVyKpicews apitoTopavovs kati pevadvdpov.—203. 3 no lac., 10 rovrov, 16 xakias, 18 yaorpi (7 KwAos, 20 Tov om. Td yeAGv. 204. 2 dé, 5 dpa, 6 aypious toiot, 7 Kai om., 8 adAX’ ai tprxoBoorpuxes, I1 yvpdvwrov. 205. 1 ovrws, II mpdcwrov, 12 yvvaiki pecpaxiw, 14 ovTw dee, IQ Tas. 206. 5 wepimevovorv, 10 Kaddv mapexov, 16 7» Kwpekov, 17 Seybevto, 19 diAomwovois, 20 Kat for eri, 21 dxpoarar, 207. 1 no lac., 5 eAxwrivors, 6 Snxrixdy, 7 OpvAAovpévn, 8 kat Om. pemeypeva, 10 aadharés, 13 peTpiws, 14 aAnberrépors. | ei KaA@s etpyntat Tb AGGe Bidoas.—A479. 10 evapvrretOar, 15 tavTd, 16 Tiv Kata mpwpav oppnv. 480. 5 e«apev om. dé om., 7 wy ayvopev, 12 avontw for dyvwpove, 13 ovdev, 14 ppevntifwv, 16 Tov, 21 mpowexov, 22 EXEL, 24 Tvvepyafomernv. 481. 3 éxacrov, 8 Te, g erapivwvda, 20 TvvTaTTOpevat, 22 7) OM. dpeT)v. 482. 2 AavOdvover, 10 av for dAAa, 16 éXevPepwOn, 20 erapivevdas, 22 dgas. 483. 2 duapyjoas nuvve, 3 Ovpdv av, 6 cwpara, 7 Wuxas, IO Exwot pi) aroppeovTwy pnde Tivopévwv, 15 avTdv, 16 puxpa, 17 troonpaivwr, 19 5¢ oreporjas arectoinoev. 484. 1 tpépovtes, 2 ovvrdvws trAacbEvtes, 5 avOpwruv, 7 TOAAG, 8 avTw, 12 yivomevwv, 19 adnv, 23 ywwoKxer Oar, 24 yoke. 485. 4 Ta, 7 pnd? drXAo tr, 8 GeAev for xaipew, 10 ydiocpa trAapdv, 11 avTdv, 12 wEeprapmrexopevos, 17 Toler pev om. evGevde. 486. 4 no lac., 12 ovde, 16 airep cioiv, 18 Os OM., IQ KatarovTifwr (Reiske), 21 «ai after ampagiav om. Tepi povapxias kai Synpoxpartias kai dAtyapxias.—I130. 3 no lac., 5 7, 8 6 Om. mpotpyvopevos Siaipov, g €& aTTLK@ pev, II mapadederypévw, 17 inrnp, 18 rAéoves, 19 7 0m. 131. 14 evoroxos. 132. 1 4 for Kati before trepBoAnv, 3 dmrexAn- pwraro, 4 crapridtas, Q Kai 7} 0Om., 11 8’ before aperpiav om., 18 tpiBwva, 21 Aecxovpyetov. 133. 10 rovrov, 14 no lac. TTWLKOV EVavVTLWPaTwWY,-—214, 4 ovTWs, 5 TO adTd, II Adyous, 15 Tov before dpyev. 215. 4 opodAoyovpevois, 20 ovGev, 21 TH 7Siovt, 22 4 , > Ff = . / 21 orov, 248. 2 xapirdpevov, 3 otovel, 4 yuvopevov erixAnow, 5 emixkAnow, 6 yiverat, 10 efvas for dvra, 12 of for ei, 15 Sé ovdepiay for Seo. Sé piav adévra, 16 Kat’ aAdAor, 17 avra €deye: Ett, 19 Ws OM., 20 TO avev: Tav Stadhopwy. 249. 3 of om., 5 GeAnoece, 16 pev Om., 20 atta, 21 dv before dviwapxtov, 23 Ttav—yxaipe. Om., 27 Ta OM. 250. 1 rijv ércxatpexaxiav, 2 yiverat, 4 yiveras, 6 Tov before POdvov om., 13 ovdev KaPijKes, 14 Sumrapéevns for Scepyopevns, 21 ovTws, 24 yevouéevys, 26 73 dyabdv. 251. 3 xa” avo, 5 aperiv, 8 tHv for rdv, 10 yap om. daci, 14 gaciv, 15 déov év, 19 TovTd, 22 avdpeiov. 252. 2 cai om., 3 awd for Kami, 5 ordoes, 7 Kal mepittds for rept rdv Adyov, 10 avadias, 16 ein, 20 dv—iva Om., 23 tvevpova, 253. 5 avrov, 6 7d dAnfés, 10 Tdv, 16 Teroapa- xovra évvéa. (Bernadakis), 17 revrnxovradvo, 23 Suddhopov, 25 pact. 254. 1 xai for dre de, 3 épydfov, 9 rodro, 13 éoTw om. Kaka, 14 ait Aéyerv: Td Se, Kaxdv+ 7d Se Kakdv Emi tavta, 16 Suarimtovra, 17 GAAws, 19 TH ayaOy~, 23 TH OM. 255. 1 te before THY, 2 vonoat, 4 dAAdom., 12 rovom. 256. I re, 3 édAous no lac., 6 avrovs for rovs Geors, 20 kai ovx. 257. 3 yeyovacw, 8 ovrws, 12 araipovor, 16 evexa for xdpiv, 17 Tov after tAnGovs OM. yévovs OM., 21 Kal ptAavOpwrovs dei, 24 €ipydoatro, 27 KTifopevas woAcs. 258. 14 odv om. Kkdxeivo: dé om., 22 yiverOar, 24 av yévorro, 26 yivopévns. 259. 3 «ivi, I2 mpwrw OM. aidoiov, 14 ywopévwv, 15 tporayovons, 16 Exouev, 22 arexvixov, 23 ovley, 24 pepovs. 260. 4 ro om. SvereAciero for & ereAciero, g yiver Oat, 10 yevopevors, 11 Syrov Oedv, 15 6 8& Xpvourros om., 16 reromnpévn, 21 yevopevov Adyw, 25 7) Om. 261. 2 Kai om., 8 yiverOar, 9 pyre for pnd after ddixetv, 14 ovpBaiver, 19 érixoAd(er, 21 cat before Touro Om. yiver Oar, 22 Tov OM., 25 iiov Exee Tiva Adyov yiverat, 26 Kai before avr# om., 27 yiverat. 262.1 ovre yap tayaGa, 2 ra évavria, 7 7d, 8 atrov yivdpevov, 14 Eviorrapéevors éviwv, 16 GAA’ —apOjvac om., 18 aPeArnpiav 7d (nreiv, 23 THs OM., 25 Str. 263. I &v Tm} OM., ytvovras, 5 ypagop evepyevay, 3 arapxi 20 7d for 8 avrny €, Ig A€éywr 267. 5 av €£ avrou | that in < not omit! eavTov, in for év ad ae 18 dpedre; 5 mepexe BadAwv. HEVS “yevy TO TPWTYW oKXAnportat 18 puyvip 6 reAevtai om. 272. 27 €ipnKev 18 teGein (ntav are, peony, II before ris. om., 6 Trav 23 ye om. TPOKATApK’ ‘evepyeray, cvyKkaTadec 279. 2 car It oug and the si; are not alv but record fter ovde, 2 deiLevev, trd Geov, tayabdv, 18 ovcns 7 TAUTAS, 3 dé om., Tiva OvV, 38. 12 7d pnot om. ; , »y OM. Kat for gyre fl eziriac) pPaiverv, dvTos TOU ovv om., £ oias 37 pov, II ev line om. LparkEeva- : Si8ouct, OnAvdovv, T om., nTpartv 7 13 pyvas, 7. I yevo- vats OM., ) yiverat, 7 addXov, GerAno ere, 1 ra OM. » KaOHKE, a?’ avrd, avdpeiov. acapias, Tevoapa- or ore de, cakov él fore TV, us Geos, rov after is woXes. 9. 3 eiot, 23 ovlev, 11 dyrov 61. 2 Kal at before UT OM., OTApLeEvols 263. I ev THE TOLEDO MS. OF PLUTARCH’S MORALIA 175 T) OM., 2 mepatvopévwv, 5 Kai dyaOois, 8 wool, 10 tév dvTe (« of dvre on erasure) yivovrat, 12 peuixOae, 16 rupavos for ’Avripavros, 17 éri before reripors. 264. 2 évriav, 5 ypapopéevwy kai om. 6. xai 7d: od mdvras after Xpyorovs, 10 yevvntdv, 13 mpds évepyetav, 16 eb mountixdy, 17 ovTws, Ig ovdév. 265. 1 yevynroi, 2 Aéyovrac om. dyévvyTor, 3 arapxis, 5 yevvynrot corr, 6 ppoveiv, 7 yevverOar, 11 ed rounrixdy, 12 mpoBdddAovor, 20 7d for rov. 266. 1 yevouévwv odv om., 4 péxpis adrdv, 6 Yvyijs xwpiopds, 7 péxprs, 8 avrnv eLavadwon, 10 vov—éyovros OM., 14 avTG, 16 KaratarTépevov, 17 drodatvouévur, 19 A€ywv—xKdcpov om., 21 atbrov, 22 avrov, 24 yiverOar, 25 vopuov for xédcpov. 267. 5 avrav for doder oi 5, 7 éavrd for év aire, 8 after Setrac the words from tpéperat «£ avrov (266, 15) to this point are repeated with the same readings as before, except that in 266. 17 tpepopevn is given for tpepopévous, in 266. 19 Aéywv—xdopov are not omitted, but given with the variation from Bernadakis’ text only of airdv for cavrov, in 266. 25 xdapov is given for the previous vépov, and in 267. 7 év avira is given for év air instead of the previous éavrd, 14 yvxovpevov, 16 yiverOar, 17 Yu, at 18 dpeorepov, 19 Aerropepiorepov av’TO, 21 dpaidv, 22 yiverOar, 268.1 tpitw for rpudtw, 5 mTepexerat, 10 wapayvéw, 12 8’ dAov Tov KOTpov OM., 15 ov OM., 17 Te Td, IQ peTa- BdrdrAwv. 269. 4 yiverOar, 6 poveleaten, 8 rw OM., II yevvouernv, 15 pyoi, 16 éyywo- pevns yevvapevat, 18 ovtws ayevvntov, 270. 6 tiv Aqurporynta Oepydrynra tov rupéds, 7 év TO TPOTHW TOV Proikov (nTnpaTwv, 8 ovdév Tas GAAO Tas ELets, II aitLos 6 TvvéeXwv, 12 ds oxAnpotata, 15 7’ for y’, 16 yiverar, 17 oxAnpdv for oxAnpy ty before pi) ruKvG om., 18 pvyvipevos, 19 €ori, 26 ovs S€ wai for ovaas Kai, 29 aitw. 271. 3 ais (Emperius), 6 reAXevtaiov, 7 TovTO, G av’THs, II yuwopevov Svvardv, 13 tSpicarGa, 15 oidpevov, 16 Kat om. 272. 1 xat om., 2 avrd om., 8 djA0c (o written over v), 15 7, 22 7d before zpis, 27 eipnxev. 273. 5 avrdv, 10 avrTd for avrov before pérov, 13 ovtws, 17 6 8 ef for ov 5H, 18 tefein 6 koopos for reBevta Tov Kdcpov, 19 POopay ovtws, 20 THS TOV ameipwv for (ntav areipy, 21 tas before ovvoxas om., 24 dtapopys, 25 eavtdv. 274. 5 atrdv, 7 7d Om. peonv, Il pepopevors, 17 7d for rov before rv ovciav yvwpéevnv, 21 SyAovots, 24 Tijv before rijs, 25 paxopevov. 275. 1 Suddopov eridexrdv, 3 S€ om., 5 7) OM. eipappevnv: THv om., 6 ravrys olov emidexrov, 12 Gu’ om. 276. 3 yiverOat, 10 SnAovdrs, 17 7d before pév, > et 23 yeom. 277. I rov om.,, 3 Karapxtixoyv, II et for 7 A€yomev, 14 avroreAH: TO before TPOKATAPKTLKHV, 21 yiverOar, 24 avTH, 25 Tov avOpwrov, 27 dv ai om, 278. I eiperias ‘evepyevav, 3 yivovrae no lac., 6 yiverOar, 11 mpoorimrovras, 12 5), 14 mapadoynv Kal cvyxataGerw pavracias, 15 «i for 7, 19 byy, 20 Adyos om., 21 wepi before tivos om. 279. 2 xara for kal, g rais pavraviaus, 12 ywwwoKwv, 15 mpoorimrey, It ought perhaps to be mentioned that the gatherings of the MS. are irregular ; and the signatures, which are written at the foot of the last page of each gathering, are not always correct, as the following table shows. I have no explanation to offer, but record them so that scholars may have the evidence before them: Page, _Sctibe's —. Page. = Number. 42 e 294 ty 514 Ke 82 B 314 ub 534 K¢ 102 y 334 Le 554 kK?) 122 5 354 7 574 Kd 142 5 374 u¢ 594 «¢ 162 ¢ 394 in «614 KN 182 n 414 1] 634 K@ 198 0 434 Ka 654 r 218 ‘ 454 KB 674 da 238 Te 474 Ky 694 Ay 254 iB 494 Ko hae as 274 (8 176 THE TOLEDO MS. OF PLUTARCH’S MORALIA Nearly all through the MS. the title of the particular treatise is written in red ink at the top of the page. But these titles are not always correct, since the title of one treatise is sometimes carried on at the top of the first few pages of the next. It seems to me that, until recently at all events, this MS. has been very nearly in the position of having been condemned unread. It may have thoroughly deserved the treatment it has received, and, if so, it should return to neglect. But, since the justification of its neglect has not yet been made certain, my purpose has been to correct and supplement the existing accounts of the MS., and to provide others with further material for forming their judgment of its value than has been hitherto available. G. B. A. FLetTcuHer. NOTES ON SOPHOCLES AND THUCYDIDES. SOPHOCLES, Antigone 853-856, Jebb’s text: mpoBaa’ éx’ érxatrov Opacovs bynrdv és Aixas BaOpov , * , x , T POVETTET ES, @ TEKVOV, TOAUV. mpocereres L, rec.: mpowémauras rec.: woAv rec.: woAvv L rec.: roAv mpocvéraias, & réxvov Pearson, which is rather displeasing in view of the unbroken proceleusmatic rapafardv in the antistrophe 874, rapaBardv ovdapy rede, Jebb’s way of dealing with the passage seems far-fetched ; and although he quotes Agam. 383, Aaxrioavti péyav Aixas | Bwpdov, and Eumen. 539, Bwpov aiserar Aixas, pnde viv Képdos dav abews modi Aaé arionis, he does not see that this passage points to the same action. The alteration of zoAv to 7odi will give the sense required. Schneidewin was on the right track with rodoiv, but wodi is generally found in this expletive use, and sometimes mooi, but not wodoiv; just as Horace always uses pede in this sense, not pedzbus. Illustrations of the confusion of the letters in Lare: Oed. Tyr. 101, where it originally had Svcavras, which has been changed to Avoavras; Oed. Col. 1597 «duce, a mistake for éAvoe. The mistake raépmroA:s for rayrodv is not far off at Ant. 614, and also at Aesch, Sept. 624, yupvw6év dprdcas Sopi (Romahn) should almost certainly be accepted for ddpv. Thucydides VI. 4.1: TpwrAdv te 6vopa xwpiov oixioas. The order of the words is exceedingly awkward: all that is required is to transpose TpwriAov and xwpiov. Thucydides VII. 27. 4: 67é pév cai rrcdvwv eriovrwy, ore & €€ avdyxns THs tons dpoupas katabeovons Te THY xwpav kal Anurteias TovovpErys, ot APnvaior peyddra €BAdrrovTo: Everyone from the Scholiast downwards has struggled with ions, and some impossible meanings have been forced into it. It may be suggested that it is merely a dittography of the last syllable of avayxns (ICHC), and may therefore be deleted. J. U. Powe tt. —~ TE CoNTEN ITT. Vie IN c of Aristc of the M On by some reference Euclid d which wi will be d thought Aristotle the ‘ bool The bool course, w important There is, thought ; the divisi TOTWOUTOV € Anot roll; and end of th Aristoteliz of a parti common i: dé K.T.A, thought in additions in the N.E N.E. pattern, re OTL pev ovr of discussi at the beg between cc words 6d « passage is for this pla as to insert but also to ends of sed } Jaeger, des Aristoteles n in red = title of ext. ry nearly deserved since the been to 1ers with hitherto TCHER. WET ALT AS, leusmatic ling with LVTL peyav (Oéwe moot on. The the right ometimes t pedibus. originally 1 mistake d also at accepted he words »piov. ; THS bons AarrovrTo’ ind some is merely leleted. OWELL. THE COMPOSITION OF ARISTOTLE’S POZ/TICS. ConTENTS: I. General Considerations, p. 177.—II. The Case of the Politics, p. 180.— III. The Parts of the Politics and their Interrelations, p. 180.—IV. Jaeger’s View, p. 183.—V. Von Arnim’s View, p. 185.—VI. Conclusion, p. 187. I, GENERAL CONSIDERATIONS. In considering the question as to the order of composition of different portions of Aristotle’s works it is necessary to start (as Jaeger did in his study of the genesis of the Metaphysics) with some idea as to his method of composition. On looking at the surviving works one sees at a glance that at some date and by some hand they have been carefully arranged as a continuous series. Internal references forward and backward are frequent. The author refrains as carefully as Euclid does from anticipating in an ‘earlier’ discussion the answer to a question which will arise ‘ later.’ The forward references are merely promises that a question will be discussed. These multitudinous cross-references are so interwoven with the thought and the argument that there is little doubt that in the main they are due to Aristotle himself. On the other hand, the short transitional statements with which the ‘ books’ as we have them close must always be accepted with some reservations. The book is a device of the ancient bookseller, not the unit of composition. Of course, where they could, the editors have made the ends of books correspond with important breaks in the argument; but wholly artificial book-endings do occur. There is, e.g., the end of N.E. 0, which corresponds to no important stage in the thought ; and here the editor or bookseller has merely emphasized the artificiality of the division by inserting the wholly inappropriate clause, wept pev obv rovtrwy émi TorouTov €ipjcOw, Another reason for distrusting the ends of books is this—that the book filled a roll; and if anything was to be added, clearly the easiest place to insert it was the end of the roll. It would be natural to add a short sentence of transition in the Aristotelian manner, indicating which roll came next. There are several instances of a partial quotation from the first sentence of the roll next following. More common is a Janus-faced sentence of the form epi péev ody x.7.A. wept 5€ or Aourdv dé x.7.4, And such a sentence may occur at the end of a book, even when the same thought in a fuller form begins the next. But,as Jaeger has shown, more important additions than this are often easily to be detected at the ends of rolls. There is one in the N.£. which, I think, Jaeger has not pointed out. N.E. B g opens with an elaborate formula of transition of the regular Aristotelian pattern, resuming the discussion of virtue up to the point reached. Starting with Oru pev odv, we expect to meet a balancing antithesis with de stating the next subject of discussion. The balancing é¢ has its obvious place a page and a half further on, at the beginning of N.E. IT (where Bywater follows our MSS. in printing 5)). In between come some miscellaneous practical precepts rather lamely introduced by the words 8:3 kai épyov éoti orovdaiov eivat, There is no reason to deny that the inserted passage is Aristotelian, but there is every reason to doubt whether it was intended for this place. It was no doubt inserted here to fill up a roll. These considerations as to insertions at the end of rolls apply of course not merely to posthumous editions, but also to Aristotle himself ; and not merely to the ends of books, but also to the ends of sections of the argument which fall within the existing books. Aristotle is 1 Jaeger, Entstehungsgeschichte dey Metaphysik by the abbreviation Enist. des Avistoteles (1912), referred to subsequently 178 J. L. STOCKS rather apt to end a discussion with some loosely related paragraphs of the nature of footnotes or addenda, The units of Aristotle’s composition are the péBodos and the Adyos. The péOodos (in its concrete, not its abstract, sense) is the larger unit, often hardly distinguishable from the zpayyareca—i.e. often approximating to the modern ‘ science,’ as the name for an independent department of enquiry. The bounds of the péOodos in the Aristotelian architectonic do not correspond with the bounds of the treatises as we know them. Taking the several groups of his works in order and ignoring minor works, which are either spurious or fall outside the scheme, we find the following péBod0. established : 1, Concerning nature (or material nature—Met. 1076a 9). This includes Phys., D.C., G.C., Meteor. Its subdivisions are not correctly rendered by the present arrangement. The last two books of the D.C. belong to the G.C. 2. The set of short treatises following De An. is called (476b 9) 7) epi trav Cowv peGodos. 3. The relations of the several zoological treatises is not quite clear, but in Gen. An. V. the discussion of the Parts of Animals (which includes Gen. An.) seems to be divided into (a) a general explanation, and (b) the explanation of the various forms which a given part assumes. Gen. An. V. in short (called by Zeller a ‘sort of appendix ’) is referred to as a separate peOodos. 4. The Metaphysics, Rhet., and Poet. are each referred to as constituting a separate peGodos. : 5. The conception of the connexion of the Logical Treatises is not very clear, but the Topics is apparently regarded as a separate sé Oodos. ? 6. Inside the Politics different pé8odo are it seems distinguished : (a) AE seem to constitute a pefodos, which is described (1293b 30: cp. 1295a 2) as mepi woXTeias. (b) Z refers to itself at the beginning (1317a 19) as a new u., and refers to a passage in A as ‘ in the preceding p.’ (1317b 34). (c) A refers to a passage in I’ as ‘in the first p.’ (1289a 26). (d) ‘This p.’ in H (1324a 3 and 22) seems to refer to the construction of the ideal state, with which HO are occupied ; but the conception might be more general. With this use of pw. in the Politics may be compared N.E. 1094b 11, where the p. (sc. the ethical enquiry as a whole) is said to have the aims stated, and to be a branch or subdivision of roActixy (roX. tis otea). The Politics seems to be the only one of our Treatises which is referred to as containing a number of péfodo: ; but it is perhaps a mere accident that the Physics and the De Caelo are not similarly broken up. it will be clear from these instances that pé0odos is a very fluctuating conception. There is no fixed unit ;, but the guiding idea is that of a continuous argument, which may be divided and subdivided into constituent arguments. Hence, e.g., a reference to what was said ‘at the beginning’ is at almost any point of any of Aristotle’s works equivocal, if not more closely defined. In the Meteorologica, for instance, it might mean the beginning of that work, or the beginning of the De Caelo, or almost any passage in the Physics. Jaeger adopts for convenience the small-scale notion of a peption. , which ference istotle’s ance, it almost 1 in the ‘itation, 1. any of ut only ompared LyMaTELG : Hy THE COMPOSITION OF ARISTOTLE’S POLITICS 179 It is clear that this elaborate architectonic in Aristotle’s works is in a certain sense artificial. It is impossible to suppose that he actually wrote each component member of the series in the order in which the scheme puts them.! The elaborate ordering was rendered possible presumably by the fact that none of the components were published ; a new member, therefore, could always be inserted, and suitable adjustments in related sections previously composed were easily made. But if Aristotle’s later compositions were directed to filling gaps in this elaborate scheme or rewriting sections of it, it is difficult to attach any great weight to arguments from awkwardness of transition or from the occasional incoherence of cross-references.? It is only natural that such maladjustments should occasionally appear. The wonder is that they do not appear more often. In fact, the rarity of such phenomena makes one think that Aristotle’s method of composition was rather more continuous than Jaeger seems to suggest. Jaeger® is no doubt right, however, in saying that in regarding Aristotle’s works we tend to reverse Aristotle’s emphasis. He thought primarily of the smaller unit (call it the Methodos), and of the science as a group of related Methodoi. We tend to think first of the science which, under a certain mode of treatment, falls into certain divisions. He got to his notion of science, one might say, inductively from the methodos; we get to the methodos deductively from the science. But then Aristotle was engaged in creating that organization of knowledge into departments which our thought presupposes. But however this may be, it is beyond doubt that there is a definite order established by what I have called Aristotle’s architectonic, and this order is of course to be taken, not as the order in which the treatises were written, but as the order in which he wished them to be read. Whether he was to read them aloud himself to the students of the Lyceum, or intended merely to put them in the library to serve as textbooks for the use either of the fellows or of the students, there seems to be no way of deciding, Jaeger suggests that a quasi-publication for such writings was effected by reading them aloud, in the way Zeno read his discourse on the One and the Many at the beginning of Plato’s Parmenides. In the house of Pythodorus, as in the School of Plato or Aristotle, the reading was followed by a discussion. Socrates demanded a repetition of the first Adyos, and thus began the argument retailed in the Dialogue. The parallel is interesting, and very likely describes correctly the procedure followed in the Academy and Lyceum. But it is worth noting that it does not account for Aristotle’s architectonic. Aristotle could hardly rely on keeping a class together long enough to enable him to go through the whole of his didactic works in this way. And one would expect more regularity in the length of the smaller units if all alike were to be treated thus. Jaeger‘ calls attention to the passage at the end of the Sophtstics Elencht, which is of course a ‘ book’ of the Topics, for corroboration of the view that Aristotle’s Adyoa were intended for the lecture-room. ‘In connexion with rhetoric,’ says Aristotle, ‘there was an abundance of old material at hand; but on syllogistic inference we had absolutely nothing to say beyond what our own long-continued labour could discover. If your observation convinces you that from this original foundation the enquiry has now been put in a sound state as compared with other disciplines which tradition has matured, it remains as a duty for you all, or for those who have listened to me, to excuse any omissions in the treatment, and to be very grateful for its achieve- ments’ (184a 9 sqq.). He takes this to imply, not that the Topics consists of lecture notes in our sense of that word, but that it was intended to be recited word for word 1 Cf. Zeller, Aristotle I., p. 105. Incessu proved spurious by its concluding transi- 2 E.g. the De Motu and De Incessu Animalium tion to the De An, (or Parva Naturalia—reading are not proved spurious by the fact that the end doubtful). of Part. An. (preceding M.) promises an imme- 3 Entst,, p. 160. diate investigation of Gen. An. Nor is the De 4 Entst., p. 144. 180 J. L. STOCKS before the School; and he adds that, in the case of writings thus recited, the members of the School would often possess copies which they had made themselves. Similarly, in Plato, Phaedrus had a copy of a discourse which he had heard Lysias give ; and Socrates in the Phaedo had to rely on the word of one reading from such a copy for the fact that the principle of the rule of reason came from Anaxagoras. It seems questionable how far it is legitimate to build on this passage and a few corroborative hints. The architectonic seems to me to point rather to the library than to the lecture-room. II. THe Case OF THE POLITICS. The case of the Politics is clearly very different from that of the Ethics, with which it is most closely connected. Of the Ethics we have three different versions preserved in the corpus. These are written on a plan which is practically identical: they are pretty well as nearly verbally coincident as the Synoptic gospels, and the most diligent search has failed to find in any one of them conclusive evidence of contamination by later (e.g. Stoic) doctrine or terminology. We have only one Politics, but this one contains evident inconsequences ; and while its parts seem to have changed their order at some time, no rearrangement produces a satisfactory sequence. The two cases are different; but in each the undeniable evidence of doubt and disorder affords an excellent starting-point for the critic who wishes to discriminate the various strata in Aristotle’s writings. In both cases clearly the architectonic has not completed its work, and the secrets of the workshop may be revealed. : The two problems are tied together by the close relation of the topics discussed in the two treatises, involving fairly frequent echoes and cross-references, Further, the alleged fact of the dependence of part of the Politics on the Eudemian rather than the Nicomachean Ethics is one of the grounds given for dating that part of the Politics, and the body of doctrine thus disengaged becomes cardinal in assessing the line of Aristotle’s philosophical development. Other questions enter here. Thus Professor Jaeger in his Avistoteles claims that the results he has obtained by a similar analysis of other works, especially the Metaphysics, show a development in a similar sense, and thus corroborate, as they are corroborated by, his findings in the field of Ethics and Politics. As no particular line of argument in a question of this kind is ever quite cogent, a good deal depends on constructing such a convergence of probabilities; and Jaeger’s case is certainly seen at its best when it is stated asa whole. For that, however, a book would be needed. Here we must ignore the wider issues and concentrate on the narrower. In this paper I shall confine myself to the Polttics. Ill. THe Parts oF THE POLITICS AND THEIR INTERRELATIONS. The major components of the Politics! are the following péOodo: : I, Tepi oiKovopias = Bk. A. 2. epi TOV TpdTEpov aTodynvapevwy Tept ToALTEias = Bk. B. 3. wept woditeiwy Kai rapexPacewv = Bk. Yr. 4. mept tov GAAGV ToAiTEiwv Kai Tepi POopas Kai GwTypias avtov = Bks. AE, 5. Wept woAtTElwy KaTagTaTeEws = Bk. Z. 6. wepi ras apiorns ToXrTeias = Bks. HO. The six components may in effect be reduced to five, since AEZ clearly belong together, and there is no good reason for separating them. The case of the Politics is, I think, unique among the longer treatises of Aristotle in that each of these five components begins with an introduction, which 1 This list of components, with their titles, is adopted from Jaeger, Entst., p. 156. might stage in careful stateme introduc and Tof of the | paragray where tl the pev- retrospe statemer to make As | I=, an edito! seem to. 2=] beginnin which cc some MS 3= I and also dispute t! 4-5= of H, wh It is Back outside it M. 3 which is 1 the openi In M is referrec aristocrac as Tpwror Tape Pace. specific (e In th several pa expected. In M. to the disc reference - accepted i discussed ¢ doubt a ve Politics ; b evxnv is i reference Case) and id, the selves. Lysias such a S. 1 a few library 1 which eserved hey are e most nce of ily one eem to factory ance of shes to rly the may be scussed ‘urther, er than Politics, > line of rofessor analysis r sense, f Ethics | 1s ever bilities; ‘or that, ues and atics. - 4. >. 7 belong itises of 1, which THE COMPOSITION OF ARISTOTLE’S POLITICS 181 might well stand as the introduction to the whole. The beginning of each main stage in Aristotle’s progress through his immense programme is usually marked by a careful introduction, and each subsection by a more or less elaborate transitional statement. The great groups of related topics which form our main treatises are introduced practically without exception (the two exceptions are, I think, Hisé. An. and Top.) by a general statement, which is made to determine the starting-point of the discussion. The subordinate divisions begin normally with a sentence or paragraph which is partly retrospective and partly prospective. (In such cases, where the Greek allowed of it, the editors have often divided the sentence, so that the pev-clause ends one book and the 4dé-clause begins another.) There is no retrospect at the beginning of any of these sections of the Polttics; there is no statement that such and such questions remain to be dealt with. Each seems to make a completely independent start. As for the ends of the sections: 1=A ends with a rather lame but elaborate apology (too elaborate to be due to an editor) for not completing the programme which the preceding discussion would seem to foreshadow, and announces the subject of B. 2=B ends with a peév-clause, and should therefore be followed by a book beginning with a dé-clause. It is worth noting that the only one of the other sections which could fulfil this requirement is the last. At the beginning of H epi 5¢ has some MS. authority. 3 =I ends by announcing that its sequel will be the discussion of the ideal state and also with a garbled fragment of the first sentence of H. This proves beyond dispute that at some date H followed I. 4-5 = AEZ ends once more with a pev-clause, which without a 6 at the beginning of H, which follows, has no justification. It is necessary, further, to note the internal cross-references. Back References (1 include only references within a Methodos to a passage outside it): M.1=A has none. M.2=B has none. M.3=T' refers twice to the zpwro: Adyo. One reference is clearly to A, which is referred to as epi oixovopias kai Seororeias. The other is equally clearly to the opening discussion of I‘ concerning the identity of the good man and good citizen. In M. 4-5=AEZ there are eleven more or less certain references to I. This is referred to in A (89a 26) as tpwrn p. ©. TOV TodiTeiwv, (89a 31) as the discussion of aristocracy and kingship, (goa 1) as the discussion of aristocracy (doubtful), (93b 2) as mparoe Adyo.—the subject again aristocracy, (93b 27) as Ta Kar’ dpxnv (ref. mapexBdoeis), (95a 4) aS mporot Adyo (kingship). The other references are less specific (e.g. eiprras mpdrepov or roAAdkis). In this portion of the Pol. there is no clear reference to B, though there are several passages (e.g. Sparta, Carthage) in which such a reference might have been expected. There is also no reference to A. In M. 6=HO there are six back references in all. There is a definite reference to the discussion of slavery in A (as mpa@ro: Adyor, 25a 30). Another unmistakeable reference to I’, also called wrparo. Adyo. (33a 4). A reference to B is generally accepted in the phase with which the main topic of H is introduced. ‘I have discussed other forms of government,’ says Aristotle, ‘ now for 7) car’ evynv modus.’ No doubt a very general statement, which might apply to any or every other part of our Politics ; but the phrase is almost identical with the opening words of B, where xar’ edx7v is in similar opposition to ai adAAa woAitera. We may call this, then, a reference to B. The other three references are all vague (eipnrac woAAdxis in every case) and will bear no weight. The subjects, however, are such that, while B may 182 J. L. STOCKS be partly in mind, none of our other four (or three) méfod0 can be. There is no shadow of a reference in HO to AEZ. On the face of it these references (a) closely connect AEZ with I’; but since by one clear reference I" presupposes A, we get the sequence AT'AEZ. (b) Secondly, they exhibit a rather weaker connexion between HO and all the first three books. (c) Thirdly, they reveal complete disconnexion between HO and AEZ, It must be remembered that the references do not in any case settle the order of composition ; they only reveal the plan on the basis of which the books were composed. Taken in this sense, these references show that the books of our Politics embody no single clearly defined plan, but oscillate between different plans. So far we have considered only the back references. There are also the forward references to be taken into account and the external evidence. Forward References.—1 find eighteen forward references? in the Politics. These are all of course promises to treat a certain topic. Of these eighteen promises the editors find only three to be-fulfilled. Of the three fulfilled promises two are in B, and are said to be fulfilled in I’ (monarchy) and A (mixture of monarchy and democracy) respectively. The other is in A, and since it is fulfilled in the next book (i.e. within the same pé@odos) it should not really count. There remain fifteen unfulfilled promises distributed between my six divisions as follows: 1 (A) 1, 2 (B) 3, 3 (fT) 1, 4 amd 5 (AEZ) 2, 6 (HO) 8. I think it is pretty certain that practically all of these promises were to be fulfilled in the treatise on the ideal city, of which we possess only the beginning (HO). This is obviously to be assumed in the case of the eight promises made in books H@; for it is clear that the outline of the ideal city was to end the political series. But it is significant that there are only two promises in the section AEZ, and that neither of these could very well have been fulfilled in a treatise on the ideal city. These are the following: A 1300b 5, ‘ These are the different modes of constituting magistrates, and these correspond to different forms of government—which are proper to which, or how they ought to be established, will be evident when we determine the nature of their powers. Z 1316b 35: At the beginning of Z Aristotle says he will consider the modes of organization proper to each kind of democracy. This is actually the subject of the early part of Z. Then he adds: ‘ Moreover, we ought to consider the various combinations of these modes themselves: for such combinations make constitutions overlap one another, so that aristocracies have an oligarchical character, and constitutional governments incline to democracies.’ Clearly these two promises might have been fulfilled in close connexion with one another, but hardly in a sketch of the ideal constitution, The two passages are clear evidence that Aristotle had in mind a continuation of AEZ which was not identical with HO. External Evidence.x—References to the Polstics in other works of Aristotle are scanty. In the architectonic order its place is so late that this is almost inevitable. Only three passages need be mentioned, and only one is of any importance: 1. There is a rather vague reference back to the Politics in Rhet. A (1366a 21). The reader is referred to it for a detailed account of ‘the moral qualities and institutions peculiar to the various forms of government.’ The whole middle portion of the Politics from I’ to Z might be held to be intended here. 2. Secondly, there is the passage in N.E. E (1130b 26), in which the relation of politics to education is reserved for subsequent discussion. Here the primary reference seems to be to the discussion at the end of N.E. K, but in view of the 1 In the number eighteen I have included all 71a 20, 72a 25; I’ 1276a 31; A 1296a6, 1300b 7; forward references which seem to point beyond Z 1317a 4; H 1324a 2, 26b 33, 30a 5, 30a 32, the book in which they occur. The passages 35b 3, 36b 25 ; © 1338a 33, 41b 38. are: A 1260b 10; B 1265b 17, 66a 24, 69a 29, concluc referen forms 1 which ° drrAococ worth conside and de investi; the bes laws ar In with A an app doubtft charact possibi EKATTY words fulfilms for our or of \ explan: did no editor, their p Ge referen gives 0 was thi basis < that it Le of the he brin I. already Platon the po means clearly of the the etl Nicoma older v 1 Wh had the is no fr Probabl but wha ere is no since by Secondly, ee books. t must be nposition ; Taken in no single e forward s. These mises the )are in B, rchy and next book in fifteen 1 (A) I, ttain that ideal city, sumed in 1e outline there are well have and these how they r powers. modes of sct of the > various stitutions ster, and (Eun. sec And I na merely ‘q1 But | extracts 11 excerpt ca Therefore Jz like a house Ratton, I he may be’ to1 is necessary nately to wri of the MSS mention of copy.) Neit be imputed 1 The sy course (as Preface), a consensus © that 46-read At any ar in D. Jonatum,’ ‘act: pro 'riscian’s Donatus’ question i? The SS. (the oned the in Early he older us many In the 1uch less tor : semed to e extant nily. It -used by ow been Jeauvais, Donatus- s of the TL have nily) into (I repeat §-reading ‘med our ; history. ’s recent 1ann and ain taken now and he variant, es MS. (0) ance hand. DONATVS-EXTRACTS IN THE CODEX VICTORIANVS (D) 1o1 henceforth a sheer waste of time to mention these discussions of the 8-text which are previous to the publication of the readings of this important member of the 6-Family, this Paris MS. ~. The ‘8-text’ of ‘ces autres’ is not the real Stext. The Oxford edition for the first time gives us a truer view of the 6-text.1 Similarly the ‘ Donatus’ of ‘ces autres’ is not the real Donatus. They regard the lemmas in the extant form of the excerpt commentary as genuine, and cite them as Donatus- readings—e.g. NIHIL EST QUOD DICAS MIHI at Eun. 1064. We now know such lemmas to be mere interpolations by Carolingian or (I suspect) Renaissance editors (see Class. Quart. XIX. 28). And when the formula legitur e¢ occurs in the extant excerpt commentary, ‘ces autres’ regard it as proof that Donatus had the lemma- reading in the MS. of Terence which he used, and had found the variant reading in another MS. of his time—e.g. PATERETUR and futiretur (N.B. poti-, not pote-). at Phorm. 469. We now know that (often or usually) the formula is a mere interpolation by Carolingian or Renaissance editors (see Class. Quart. XX. 104). Therefore not merely is the ‘6-text’ of ‘ces autres’ a phantom, but their ‘ Donatus’ is a phantom too. On the strength of these phantom ‘ Donatus-readings’ they believe that the riff-raff variants of our minuscule Mar-texts and (I suspect) of Renaissance printers were known to Donatus (fourth century), and that the edition perpetrated by the pupil of Professor Calliopius (FELICITER CALLIOPIO BONO SCHOLASTICO) was far earlier than the usual period of these student-editions—viz. the fifth century. So we may say that their ‘ Calliopius’ is a phantom too.? So is their ‘ Terence of Nonius’ (see my Nonius Marcellus, Oxford, 1901). How can one fight with phantoms? To return to the Donatus-extracts in D, Professor Wessner has now consoled us for the non-appearance of Part II. of Vol. III. of his edition (Part I. appeared in 1908 !) by publishing a full list of them (Phil. Wochenschr. XLVII. 443). If this list had been available for the Oxford text, I would have added? to the apparatus criticus (where the symbol Don. indicates a reading mentioned—not necessarily approved— by Donatus) : (Andy. 608) nullius Don. (teste schol. D ‘“nullius” secundum Donatum’) (scil. disyll.), (Andr. 816) licet Don. (teste schol. D ‘uel * licet’” secundum Donatum’), (Eun, 211) quod mihi carior Don. (teste schol. D ‘aliter “quod mihi carior”’ secundum Donatum’), And I now realize that wel uelis should have been omitted at Andry. 826, leaving merely ‘quod iubet’ as the words assigned to Donatus. But Wessner, after giving a full list (now first published) of these Donatus- extracts in D, goes on to argue, from the similarity of most of them to the extant excerpt commentary, that they were taken from it, and not from a full commentary symbol. Often the real 5-reading is attested by p alone or by D alone or by G alone or by L Therefore Jachmann’s whole argument collapses like a house of cards. But since, with Daddie Ratton, I hold it to be ‘the most uncivil thing may be’ to mention names, at least when censure is necessary, and since Jachmann had unfortu- nately to write his book before Kauer’s collations of the MSS. were available, I made no express mention of Mr. Jachmann. (Cambridge please copy.) Neither discourtesy nor ignorance should be imputed to us. 1 The symbol 6 in the apparatus criticus is, of course (as explained on the last page of the Preface), a mere shorthand expression for ‘ the consensus of DGL}.’ It is not to be supposed that d-readings are only such as have this alone (most often by the pair Dp), having been supplanted in the other MSS. of the 6-Family by the y-reading. 2 If Professor Fraenkel will read Mr. Craig’s Jovialis and the Calliopian Text of Terence, I think he will regret his assent (in his excellent review of Professor Housman’s excellent edition of Lucan) to Jachmann’s theory that Calliopius was (like Jovialis) a mere corrector of a friend’s (or patron’s) MS. of Terence. 3 Let me correct a misprint in it. At Andr. 513 read non habet * t.’ 192 W. M. LINDSAY no longer extant. This argument is hardly convincing. Does Wessner think that the excerpt commentary did not excerpt the ipsa uerba of Donatus? What reason has he for so thinking? The excerpt by Paulus Diaconus of Festus, De Significatu Verborum, usually excerpts the ipsa uerba of Festus; indeed, often reproduces a paragraph entire, without any excerption. In Class. Quart. X. 106 I published Festus-extracts (discovered by Anspach) in eleventh-century annotation of Isidore’s Etymologiae by a North Italian bishop, Grauso. (They have recently been published, with the rest of the annotation, more fully and accurately by Professor Whatmough in the Bulletin Du Cange.) It is clear that Grauso had before him a full Festus. What reason then should we have to argue thus: ‘In this or that note Grauso’s words are identical with the words of Paulus; therefore Grauso must, in this or that note, have used Paulus, not Festus’? And yet that is precisely the argument used by Wessner. Surely it is plain as a pikestaff that the only admissible evidence is—(1) Donatus- extracts which can be proved to be genuine Donatus, and yet have not been excerpted in the excerpt commentary, e.g. neuter tum praesensevant (Andr. 839); (2) Donatus- extracts which can be proved not to be genuine Donatus, but wrong-headed extracts from the excerpt commentary. For this second class all that Wessner can produce is: (Andr. 825) legitur et ‘quod iubet’ secundum Donatum. What Donatus wrote was quod lubet. Since / and initial ¢ are often indistinguishable in minuscule, the fact that twbet appears also in the excerpt commentary has very little weight. (Eun. 24) pro ‘ existimauit ’ secundum Donatum, What Donatus wrote was Extstimauit pro extstimat. The archetype of the excerpt commentary had, in error, Extstimat pro existemautt. Why may not the same error have been in a MS. of the full commentary, a MS. used both for the excerpt commentary and for these extracts in D? (Eun, 211) aliter ‘quod mihi carior’ secundum Donatum. Wessner’s attempt to connect this with the sentence in the excerpt commentary pro ‘gui miht sum carior’ is not convincing. And for the first class may be adduced also: (Andr. 720) aliter ‘ laborem’ secundum Donatum. The excerpt commentary has excerpted merely ‘dolorem’ distinxit probe et post intulit sepavatim quod sequitur. For probe Umpfenbach suggested Probus (i.e. Probus read dolovem and put a full-stop after it). (Andy. 839) secundum Donatum ‘tum’ in versu non scribitur sed subauditur. This, added to what is preserved in the excerpt commentary, suggests that Donatus’ text of Terence presented the line in this form : uero uoltu, ibi me adesse neuter cum praesenserant. Wessner concludes with a mention of Lupus’ letter (Ep. 103) to Pope Benedict III. (855-858), but seems not to know that Professor Beeson (whose book on Lupus is long over-due) has discussed this letter recently (Class. Phil. XVII. [1922] 283, The Text Tradition of Donatus’ Commentary on Terence). Beeson says: ‘In this letter Lupus asks for a MS. of Jerome’s commentary on Jeremiah “ post sextum librum,” a work which, left unfinished by Jerome, comprised only six books; but Lupus evidently thought his own MS. was defective. Lupus asks further for MSS. of Cicero’s De Ovatove and the twelve books of Quintilian ; DON: of tk requ Tere is pre he w; Therefore mentary \ great boc mentary as a patc. Easter) is we have r an expert commentary italics, sin is not now was early. Wess But in hi: [1921], 168 based on t form of the scholia toc compound persons, I theory of - Silver Age Servius. ( the second to believe of aliquid i Juvenal’s si is preparing But th by Donatu: has preserv positive tha Rome. (H with a veng reads Profe And—] not seem t everything alien additi commenta editors. O include one merely Ang Paris 7920. usually mer editor’s cap Also nk that reason nificatis luces a blished sidore’s lished, fmough Festus. Tauso’s or that nt used onatus- cerpted onatus- xtracts ishable iS very >xcerpt e error >xcerpt ary pro tntulit is read ur. ynatus’ _ Pope e book XVII. ¥ remiah sd only Lupus itilian ; DONATVS-EXTRACTS IN THE CODEX VICTORIANVS (D) 193 of these authors he has only parts, and he desires complete texts. His final request is for a MS. of Donatus: “pari intentione Donati commentum in Terentio flagitamus.” The reason why Lupus is asking for a MS. of Donatus is probably the same as in the other cases—the copy that he has is defective and he wants a complete one.’ Therefore we are to understand from the letter that Lupus had a Donatus-com- mentary which lacked the Hautontimorumenos scholia. These he seeks at Rome, the great book-market. Whether that commentary was the full or the excerpt com- mentary does not appear; but my own inclination is to regard the excerpt commentary as a patchwork compilation later than Lupus. Since our earliest MS. (I saw it at Easter) is undoubtedly of the eleventh century and written in the F leury scriptorium, we have no right to assign the compilation to a date before (say) 1000 a.p. Beeson, an expert palaeographer, argues persuasively that the archetype (whether a full commentary or in excerpt form) was Insular ; but his excellent article lacks the words in italics, since he has clearly not read Léfstedt, and therefore is not aware that there is not now the slightest reason for believing that the excerpt form of the commentary was early. Wessner in this article says nothing of the date of the excerpt commentary. But in his review of recent works on Terence scholia (Bursian’s Jahvesbericht 188 [1921], 168) he seems to accept Léfstedt’s demolition of the theory which Gerstenberg based on the Bembinus scholia. These scholia do not come from the extant excerpt form of the commentary. Rather they come from a full commentary (with the Hawt. scholia too, from which comes, e.g., the right reading in Haut. 143 exercivent, a compound of ex and sarcio). Wessner, however, finds a difficulty where most persons, I fancy, will find none. He still clings with strange persistence to the theory of R. Klotz (De Scholits Stattants Commentatio I., Treptow, 1895), that the Silver Age writers were tabooed by all commentators and grammarians before Servius. Common sense allows us to believe that they may have been tabooed in the second century, the time of the Archaic Revival. But common sense forbids us to believe that in the fourth century Donatus, if he were discussing (say) the use of aliquid in the sense ‘something of importance,’ would not allow himself to quote Juvenal’s si wis esse aliquid. However, since I understand that Professor H. J. Thomson is preparing a paper against this theory, I will not discuss it here at greater length. But though he strains at the gnat—quotations of Lucan, Statius, Juvenal, etc., by Donatus—Wessner swallows the camel. For him the mere fact that accident has preserved a letter of Lupus with a request for a Donatus-commentary is proof positive that the commentary was unknown north of Italy till Lupus got a MS. from Rome. (How does he know that Benedict sent one?) This is Combinationsforschung with a vengeance, and I am sure that Professor Wessner will repent of it when he reads Professor Beeson’s sober arguments. And—I must add—Wessner’s presentation of the Donatus-extracts in D does not seem to be quite fair. He often includes alien additions, He should exclude everything except the actual words which are labelled ‘secundum Donatum.’ These alien additions he connects with the lemmas which appear in our extant excerpt commentary, lemmas which I am inclined to assign sometimes to Renaissance editors. Our only pre-Renaissance MS. of the excerpt commentary (unless we include one which is ascribed to the end of the thirteenth century, and contains merely Andy. 321-Eun. 454; Hec. 430-773) is the eleventh-century Fleury MS., Paris 7920. And it is a mere fragment (Andr., Ad. 1-121), while its lemmas are usually mere initial letters, which would be subsequently expanded at a scribe’s or editor’s caprice. Also Wessner misuses the word ‘lemma.’ In the excerpt commentary at O 194 DONATVS-EXTRACTS IN THE CODEX VICTORIANVS (D) Andy, 781, au interiectio est conturbatae mulieris, why does he call awa lemma? When / we discuss the genuineness of the lemmas in our extant excerpt commentary, what we mean is those which are used merely as a peg on which to hang the note, like NIHIL EST QUOD DICAS MIHI, the lemma to Donatus’ scholium on Terence’s phrase quod dicas in Eun. 1064 (see Class. Quart. XX. 103). Well, I have not spent as much time over Donatus’ commentary and Terence annotation as Professor Wessner has. He is the acknowledged authority on this subject. But I must say that he does not seem to me to have proved his case about these Donatus-extracts in the Codex Victorianus (D). And I long for a true presentation of the extant commentary without these misleading lemmas and without the interpolated ‘ /egitur ct.” The Teubner edition (though we are all grateful for it) has wrought a deal of mischief. W. M. Linpsay. PALATAL } IN GREEK DIALECTS. To the instances of non-syllabic « after -cA- or -vA- which were given in Class. Quart. XVII. 196 (cf. Schwyzer, Exempla 2518 17, note), two more may be added: (1) YAewvias, Kaibel, Epigrammata Graeca 783 (Cnidos), occurs in a trochaic line, and is scanned Y-u-. Kaibel calls it a ‘momen suspectum’; but in the re-edition in Anc. Gr. Inscr.in the British Museum, No. 796, the reading is defended by a comparison with 2cAdmos in C.I.G. 1577. 3=1.G. 7. 2429 (Thebes). (2) 6 trav GAAav proOwodpevos oxorav tay exit vavtidew, Sylloge? 1000. 11 (Cos), contains vavriAew, the Coan spelling of vavriAw or NavriAyw, probably a place-name. It was correctly accentuated by Kontoleon, Ath. Mitt. 16. 406, and the identity with vavtiAw seems also to have been suspected by Th. Reinach, Rev. Et. Gr. 4. 370, note 1 (‘peut-il s’agir du poisson vavridos ?’). R. McKenzie. THE In the | White sougl more versior the formatio Works and D conflator. The infe result of a d version; tha nothing be lc The for critical canot even in the fi change in the was probabl; the accusatio Fr. 105), whe Theseus, if t! the Hesiodic together ”’ int least a sixth- But an. relies will sca on this confla 1924). Evel: far from bein; marks of earl. But it we the passages | 1. Li. 1-: not containin; interpolation, Pp. 36-39; al: the existence 2. Li. 7: | replace 1. 72 ( ) which he is sé te Bi having cut ou parallel to B, and is only v so-called B ve B and not A i 4. Li. 16 B, ll. 169, 16¢ (D) When y, what te, like phrase [erence on this e about a true without | for it) DSAY. n Class. ded : zic line, ition in parison | (Cos), e-name. ity with , note I NZIE. THE SO-CALLED PEISISTRATEAN EDITION OF HESIOD. In the Classical Quarterly for July-October, 1924, the late Mr. H. G. Evelyn White sought (1) to establish that the vulgate of Hesiod is a conflation of two or more versions ; (2) to point out the inference to be drawn, and to suggest a date for the formation of the received text. To establish his first point in respect of the Works and Days he gives a list of eighteen passages where he suspects the hand of a conflator. The inference that he draws is that the vulgate text can be nothing else than the result of a deliberate conflation by an editor whose aim was to produce a standard version; that the editor’s working principle was ‘gather up the fragments that nothing be lost.’ The formation of this text he would put in the time of Peisistratus, for its critical canons clearly put it outside Alexandrian times or the fourth century; and even in the fifth the knowledge of Hesiod was so widespread that so revolutionary a change in the text was then impossible. It must have been in the sixth century, and was probably done at Athens under Peisistratus. For what force would there be in the accusation of Heraius of Megara, third century B.c. (af. Plut. Thes. XX. = Rzach, Fr. 105), who charged Peisistratus with suppressing a line reflecting adversely upon Theseus, if there had not been a tradition that Peisistratus had somehow influenced the Hesiodic text? ‘ But the fact [?] that we have two (or more) versions “ stitched together”’ into one’—that is the grand argument in favour of a Peisistratean, or at least a sixth-century, edition. But an examination of the eighteen annotated passages on which the author relies will scarcely convince the reader of the truth of the conflation theory, and it is on this conflation theory that the whole structure of the argument rests (p. 149, C.Q., 1924). Evelyn White’s first error seems to me to be that he forgot that repetition, far from being evidence of the existence of a dual version, is one of the characteristic marks of early verse (especially verse of a didactic kind—e.g., The Book of Proverbs). But it would be unfair not to examine his argument in detail, and see whether the passages he gives will really bear the inference which he draws from them. 1. Ll. 1-10. This can hardly be counted relevant. The existence of a version not containing these lines at the very most could only prove that the lines were an interpolation, and in this case does not even prove that (see Mazon’s edition, pp. 36-39; also F. Leo, Hesiodea, Gottingen, 1894). It cannot be adduced to prove the existence of a parallel version. 2. Ll. 72-76. Evelyn White admits that 1. 76 is probably interpolated to replace |. 72 (which = Theog. 573), and they may not be an example of the doublets which he is seeking. 3. Ll. 116-120. All. 116 fin. érOda S¢ wravra—118 d¢Oovov—the conflator having cut out -uu--, which would have completed the verse and made it exactly parallel to B, of & €@cAnuoi—120 The trouble is that 1. 120 has no MS. authority, and is only vouched for by Diodorus (see Rzach, ed. ma., ad loc.). Moreover, the so-called B version of 8’ éOeAnpoi «ri. is not a mere repetition of A. If we had only B and not A it would be nowhere stated that the earth of its own accord bore crops. 4. Ll. 169, 169a-b-173. The alternative versions are given thus: A, ll. 172 sq.; B, ll. 169, 169a-b, reading these B lines after 1. 173 with Rzach, not after |. 168. T. A. SINCLAIR 196 But of the lines assigned to the B version, 1. 169 is only introduced from Proclus into certain late MSS., and cannot be said to be part of the received text, much less the fragmentary lines 169a-b, which are found only in the Geneva papyrus. In fact the B lines, far from showing that conflation has occurred, are strong evidence that it has not, for they are not even intelligibly incorporated in the text at all. Those copyists who did add the line that is called 169 did not know where to put it, seeing that the Jemma of Proclus is to be found at 1. 158 (not 1. 169 as Rzach says; it is on p. 115 in Gaisford, Vol. III.) ; but the scholium refers to 1. 174, and one MS. has it after 1. 173 (so Paley, crit. note). That the Geneva lines should be added to 1. 169 (J. Nicole, Rev. de Philologie XII. 113) is probably quite right, but there seems to be no reason why any of the lines should appear in the text at all. Interpolation was attempted, perhaps in the interests of Orphism,! but failed to succeed, and the misplaced lemma of Proclus and the mutilated Geneva papyrus are interesting traces of the attempt. The Alexandrian critics were not deceived. 5. Ll. 169c-d, 174 sqq. Az=ll. 160c-d, 176-178; B=ll. 174-177, 182 sq. Ll. 169c-d stand or fall with 169a-b; but even if they were genuine the existence of the doublet would depend on the way in which the fragments were restored. There is no repetition in Weil’s first restoration (Rev. de Philologie XII. 113). For the rest there is no difference, except that A is said to have 1. 178 and not B. 6. Ll. 178-181 are said to belong only to A, apparently because ll. 179-181 were | bracketed by Lehrs, and because they would make a very suitable ending to a description of one of the ages. But Lehrs’s bracketing is most unreasonable, it merely takes the words out of the poet’s mouth just when he was making his prophetic announcement ; and there is no reason why ll. 178-181 should end the prophecy. 7. Ll. 182-201. ‘The variants distinguished in this section are sub-variants within the second of the two main versions of the “ Iron Age.” ’ LI. (a) 185-186 are said to be a pair of doublets to ll. 187-188. But they do not look like alternatives. The second follows and supplements the other quite naturally. L. 189, which is clearly only an inept expansion of the gloss on 1. 192 yetpudixar, is divided ; it sends xecpodixa: to take the place of ov« éora: in 1. 193. The rest of it is left in the air. And ov« éora, which pace Evelyn White, is needed to make 1. 192 intelligible, disappears, because no place can be found for it in the a-version to which it is asserted 1]. 191-192 belong. 8. Ll. 225-237. A, ll. 225-229; B, ll. 230-237. There is no exact correspon- dence in the subject-matter in the two passages, so the author relies on the verbal correspondence of the opening Sixas .. . i@eias followed by roto., 1. 227, and the i@vdixnot and a similarly placed roto. in the B version. But these are merely two examples of a very common mode of expansion in Hesiod and in all gnomic poetry. The type is,‘ He that diggeth a pit, himself shall fall therein.’ Compare in this poem ll. 238-239, 282-284, 359-360, and many others. 9. LI. 238-247. A, 1. 238411. 239-241; B, 1. 238411. 242-247. It is remarked that the lot of the unrighteous is twice stated, but again, since there is no similarity between the two statements, the author relied on the repetition—rois dé (1. 239), roicw §’ (1. 242)—which has no real significance. LI. 242-247 may be said to supple- ment Il. 239-241, but there is no reason to regard them as alternatives. 10. Ll. 249-255. A, ll. 249 éyyis yép—251; B, ll. 252 rpis ydp—255. But 1. 248 cannot be left without rjvde Sixnv (1. 249), which would have to go if B were sub- stituted for A. It is true that much the same thing is said in ll. 249-251 as in 1]. 252-255. But in the former passage it is said of the gods themselves, in the latter of Zeus’ ten thousand spirits; these are added by way of explanation—explanation 1 Mazon, p. 75. THE S$ very necess everywhere than could | zx. LL to work wit! ll. 397-398. we were to: gone. This method. L 12. LI. éridoow, Or | be a conflat l. 393 with | that ll. 399-4 from that is slight. 13. LiL To. Epyov ode repetition to unnatural ; i after -uu- carried on fi necessary, si naturally enc in different v the words ar simply for re itself a comp 14. LI.. the solstice,’ Support. Bur nothing coul at the solstic offer you this be preceded | 15. LI. B, ll. 504-523 work of ar winter’; C, ‘ is adduced f given are not to be looked (b) both pass Hesiod rever had occasion deduced from winter was cz vindicated by 1 Perhaps thi his earlier arti p. 209, on ‘H where ll, 504-52 in tone and colc lus into 1 less the fact the nce that Those it, seeing ; it is on MS. has to l. 169 seems to rpolation and the ng traces 182 sq. stence of There the rest (8x were mg toa nable, it king his end the -variants -186 are rnatives. 3, which rided ; it ft in the elligible, ich it is rrespon- ie verbal and the rely two > poetry. > in this emarked imilarity (1. 239); ) supple- ut 1. 248 ere sub- st as m he latter lanation THE SO-CALLED PEISISTRATEAN EDITION OF HESIOD 1097 very necessary. The anthropomorphic gods of Olympus could not themselves be everywhere at once. Hesiod had to find a wider sanction for his plea for justice than could be found in them. 11. LI. 298-307. A, ll. 298-301; B, ll. 33-307. Both passages are exhortations to work with warnings to those who will not. So are ll. 311-316, to say nothing of ll. 397-398. In fact the whole poem is so permeated with this gospel of work that if we were to restrict the poet to one reference only to work half his theme would be gone. This would be ridiculous, but it would be a logical extension of the author’s method. LI. 298-307 must stand as they are. 12. Ll. 392-404. After yupvdv 8 dudev (1. 392) one may read Aci y’ dpa... eridoow, Or B épydfev xr. (1. 397), leaving the beginning of 1. 397 (008 érperpjow) to be a conflator’s addition. This is supported (a) by the supposed correspondence of l. 393 with 1. 398—a correspondence so slight as to be negligible; and (0) by the fact that ll. 399-400 express much the same as ll. 394-395. The most that can be inferred -_ that is that lL 399-400 may be interpolated, but the evidence against them is slight. 13. Ll. 410-413. A, ll. 410 and 411; B, Il. 410 pnd? dvaBdAAer Oar + 412 pedern Se ToL épyov opékAe +413. The correspondence in this case is more striking and the repetition to us unnecessary. But the division of the passage here given is most unnatural; in B pnd dvaBdAAeo Gar would be left as a complete sentence; a pause after -uu--- or ---uv-— is of course very common, but only when the sense is carried on from the previous line (e.g., ll. 781-782, 602-603, etc.). If any division is necessary, surely it is A=l. 410, B=Il. 411-413. For (1) érwovoepyds avjp follows naturally enough after 1. 409 ; (2) there is no reason to put érwowoepyds and duBoArepyds in different versions; their meaning is not exactly the same, and if we exclude 1. 410 the words are in their natural places. All this is evidence, not for any conflation, but simply for regarding 1. 410 as an interpolation. This is not improbable since it is in itself a complete and familiar proverb, 14. Ll. 479-490. A, Il. 479-484; B, ll. 485-490. A begins ‘ But if you plough at the solstice,’ and B, ‘But if you plough late,’ and on this the author relies for support. But a reading of the whole passage will serve to convince anyone that nothing could be more natural than the repetition «i 5¢ xev 6y dpdoys. ‘If you plough at the solstice you will have a poor crop, etc., but if you should plough late, I can offer you this remedy.’ It is almost essential that dy in order to be intelligible should be preceded by neAlovo rpomys. 15. Ll. 493-535. Here three versions are distinguished: A, Il. 493-499; B, ll. 504-523; C, ll. 493, 524-535. Evelyn White thinks A is the original ; B, ‘the work of a rhapsode who saw the opportunity to introduce a striking picture of winter’; C, ‘the work of a poet who sought to produce a rival version.’ No evidence is adduced for the separation of A and B.' For separating B and C the reasons given are not weighty, especially in a poem where a high degree of artistic skill is not to be looked for. Thus (a) not to stop at 1. 523 would make a very long sentence ; (b) both passages present pictures essentially the same. This is only partly true; Hesiod reverts in 1. 529 sqq. to the picture of shivering animals which he had already had occasion to mention in connexion with the North Wind; but there is little to be deduced from the fact. It is true that tae genuineness of much of this description of winter was called in question by the older editors (e.g., Lehrs), but it has been ably vindicated by P. Mazon,? as well as by E. de Franco. The third reason given is 2 Les Travaux et les Jours. Edition nouvelle, Paul Mazon. Hachette, 1914. 3 L’Inverno Esiodeo. E. de Franco. Catania, 1926. 1 Perhaps the author intended us to refer to his earlier article in Class. Rev. XXX., 1916, Pp. 209, on ‘Hesiod’s Description of Winter,’ where Il, 504-523 are said to ‘stand out strongly in tone and colour,’ 198 THE SO-CALLED PEISISTRATEAN EDITION OF HESIOD that ‘rpoyaAdv . . . yépovra in B and tpirod: (yépovre) in C cannot have been reached independently ; the poet of C has deliberately set himself to cap his rival’s conceit (or vice versa). But in the first place it is not suggested that they were reached separately: secondly, there is no difficulty in understanding how the poet, pleased with his own conceit, liked to repeat it. The ingenuity of the author’s explanation is quite misplaced. 16. LI. 618-662. Again three versions are said to be distinguishable: A, ll. 618-640; B, ll. 641-645; C, ll. 618, 646-662. It is stated that in all three groups the initial lines are to the same effect. This is not true of B and a great exaggeration in respect of Aand C. (Note that 1. 618 is included in both A and C.) The question is then asked: ‘Is it likely that a single poet would thus begin his subject thrice over?’ It is only necessary to read the whole passage to see that the question is futile. ‘If you are thinking of voyaging at the setting of the Pleiades, don’t do it; put away your gear and wait for the proper time, then begin, as our father did (il. 618-640). In everything you should observe the proper time, especially in seafaring, and you should choose your vessel carefully (Il. 641-645). Then when you are ready to begin I will tell you all about it—though my experience is limited to sailing on the Euripus—you remember the time,’ etc. (ll. 646-662). Hesiod then proceeds (Il. 665 sqq.) to give the advice promised. The continuity is perfect. The worst that can be said is that the poet rambles a little before coming to the point. 17. Ll. 678-694. A, ll. 678-681, 689-694; B, ll. 682-688. The author was doubtful about this one, and rightly so since it seems to depend on the similarity of ll. 687 and 691. 18. LL 765-768. A, ll. 765 sqq.; B, Il. 769, 768 (sic). The mention of jpara 8 éx AvoGev in 1. 765 and aide yap pyépac eiot in 1. 769 is the supposed reason. But 1. 769 with its yép could hardly follow 1. 764. L. 765 is the only one that could introduce the subject. The net result of the foregoing examination seems to be that in passage (7) I. 192 may be an interpolation as also 1. 410 in passage (13). There is some suspicion of interpolation also in passages (12) and (15). The grounds are scarcely sufficient to build a theory of conflation on; but indeed the author does not wait for the evidence before assuming the truth of his own theory. All through the first part of his paper —the part we have just been examining—we find statements of this kind: ‘ The con- fiated versions, then, formed part of the Hesiodic text current in the fourth century.’ Such statements, whether about the fourth or fifth century, are backed up by good evidence. But surely it is a flagrant petitio principit to speak of the ‘conflated versions,’ There is abundant evidence, as his quotations show, for the existence of a text of the Works and Days in the fourth and fifth centuries B.c., and if it was a conflation where is the evidence for a pre-conflation text or texts? Evelyn White’s quotations only go to show that the so-called conflation probably never took place. The inferences which the author drew in the second half of his article were summarized at the beginning of this chapter, and if I have succeeded in disposing of the base on which his main inference rested, it will be unnecessary to examine the remainder of the article in detail. But in respect of the existence of a so-called Peisistratean edition, I still preserve an open mind ; it is quite possible the poems of Hesiod shared in the boom in literature. But of the particular form and content of that ‘edition’ we know nothing, and the dissection of the Works and Days leads to no remarkable discoveries. T. A. SINCLAIR. F. PROF these com Singular | From has come of Glossan FR 205 RU 77 Whic (s.v. fernif the first | Glossarun fervifodina process of Plural : apparentls unknown (L.L. VII ferrum fe silver-min silver-min qui ‘the etym The c Certainly Abstrusa ‘ But this § offers: laf preface to was turne Indic. Act gloss was: Besides, a quite unsa of Glossars a gloss re SIOD n reached onceit (or + reached , pleased planation yuishable : all three d a great A and C.) begin his » that the Pleiades, in, as our especially nen when s limited siod then ct. The point. thor was uilarity of para. 8 Sut 1. 769 ntroduce (7) 1. 192 icion of ficient to evidence 1iS paper The con- century.’ by good ersions,’ xt of the mn where ns only Sle were osing of nine the so-called oems of ntent of ds to no -LAIR. FERRIFODINAE AND SIMILAR COMPOUNDS. PRoFEssoR Linpsay has called my attention to the need for an examination of these compounds, with a view to ascertaining whether they are ever used in the Singular by Latin writers, or always in the Plural. From a scholium, probably by Donatus, on Aen. X. 173: (Ilua) insula inexhaustis Chalybum generosa metallis, has come a gloss which appears in the Ansileubus Glossary (just published in Vol. I. of Glossana Latina, Paris, Les Belles Lettres, 1926) in two forms: FR 205 Frofra (!): insula Tirreno mari in quo (qua) ferrifodina exercentur, RU 77 Rufa: insula Tirreni maris in qua ferrifodina exercetur. Which form of the interpretation is right? The Latin Thesaurus cites (s.v. ferrifodina) the second form only. But that is merely because it did not know of the first form, since Goetz in his excerpts (Corp. Gloss. Lat. V.) from the Liber Glossarum (another name for the Ansileubus Glossary) has not mentioned it. Can ferrifodina be used in the Singular, whether in the sense of ‘an iron-mine’ or of ‘ the process of getting iron from amine’? Probus in his scholium on Geo. I. 56 has the Plural : Chalybes] natio Pontica, in qua sunt ferrifodinae, apparently in the sense of ‘iron-mines’ (not of ‘an iron-mine’). The word was unknown in Varro’s time, for Varro, talking of Latin inconsistencies, says (L.L. VIII. 62): ‘ubi fodiatur argentum argentifodinas dici, neque ubi fodiatur ferrum ferrifodinas.’ By argentifodinas here it is not clear whether Varro means ‘a silver-mine’ or ‘silver-mines’; but in another passage he undoubtedly means ‘a silver-mine’ (L.L. V. 7): quis enim non uidet unde aretofodine (/ege argentofodinae) et uiocurus ? ‘the etymology of “silver-mine” and “ road-surveyor”’ is as plain as a pikestaff.’ The dictionary of Georges cites /apicidina (in the Singular) only from a glossary. Certainly the Singular is found in more than one glossary. For example, the Abstrusa Glossary (published in Vol. III. of Glossaria Latina) has \!-A 24): Lapicedina: locus ubi lapides caedunt. But this gloss is referred by the editor to (Itala?) 4 Reg. 22, 6, where the Vulgate offers: lapides de lapicidinis ad instauvandum templum Domins, and it is stated in the preface to the edition that in the course of time nearly every noun-lemma of. Abstrusa was turned into a Nom. Sing. and nearly every verb-lemma into a 3rd Sing. Pres. Indic. Act. We have therefore no warrant for denying that the original form of the gloss was: Lapicedinis: loco ubi lapides caedunt. Besides, as Lindsay has already pointed out (Class. Quart. XII. 120), glossaries are quite unsafe guides in such matters. The Philoxenus Glossary is edited in Vol. II. of Glossaria Latina, It has (LA 58): Lapidicina: Aaroyia, a gloss referred by the editor to (Paul.) Fest. 105, 22: Lapidicinae ubi exciduntur lapides. iM : hy 200 J. D. CRAIG Either the compiler of the Philoxenus Glossary or, more likely, a subsequent reviser or scribe turned, for the convenience of readers, the lemma into the favourite lemma- form in glossaries (as in our dictionaries), the Nom. Sing. Now undoubtedly the Singular of compounds with facto was current, e.g. officina, which of course is not derived from the verb officio but represents ofificina. These facto-compounds some suppose to be adjectives (-mus, -na, -num), agreeing with taberna : opificina, the taberna of an opifex; aurificina, of an aurifex; carnificina, of a carnifex, a man who makes caro bouina out of an ox, caro uitulina out of a calf, etc. Similarly tonstrina, the taberna of a tonsor. Everyone admits the currency of these Singulars. But what of the compounds with fodio and caedo? I have collected this evidence. As regards the former, we are concerned mainly with the word fodinae, because it is very difficult to decide in some cases whether we are dealing with a compound noun (aurifodinae, ferrifodinae, etc.) or with two words (auri fodinae, fervi fodinae, etc.). Usage varies. While harenae fodinae and aurs fodinae are definitely implied in, e.g. Ulpian, Dig. 24, 3, 7, 14, the compounds haventfodinae and cretifodinae are equally clear in Ulpian, Dig. 7, 1, 13, 5 (cretae fodinae never occurs). Putting aside glossary forms as suspect, we find that in every recorded instance but one the Plural (fodin-ae, etc.) is used. It is always gold-‘ diggings,’ silver- ‘diggings,’ and soon. Doubtless the reference is mostly to more than one mine, pit, or quarry, but it is hardly a coincidence that only the Plural has survived. Ammi- anus Marcellinus (24, 4, 21) writes: ‘milites quibus cuniculorum erant fodinae mandatae,’ which, if it means (as in the context it must mean, for the tunnels are being dug by these soldiers) ‘the task of digging the tunnels,’ indicates that fodina was not admissible. The one exception is Pliny, N.H. 33, 78: ‘Exstat lex censoria Victumularum aurifodinae in Vercellensi agro,’ where Vict. is an adjective (derived from a place- name) variously transmitted in the MSS. It can scarcely be questioned that the unanimous evidence of usage elsewhere points to aurifodinarum as the true reading. The combination Victumularum aurifodinae is suspicious; usage suggests -fodimarum ; we might go further and say that, even supposing fodina to be possible, it is unlikely that gold-mining in this particular spot was carried out in one shaft only. The cause of the false reading could be palaeographical: -fodinarum written -fodinay (with a cross-stroke down through the arm of 7) and copied out wrongly by the scribe of the archetype from which the MSS. of Pliny are derived. It is noteworthy that salinae (‘salt-pans’ or ‘ salt-quarries’) never occurs in the Singular, that jig(u)iimae (Plural) appears to be the only allowable number when the reference is to the place from which the potter digs his material, and that /a(u)tumiae (‘stone-quarry’ or ‘-quarries’) is used always, except once,! in the Plural. Equally to the point is a group of adjectival forms in -ariae (havenaniae, ferrariae, nitvartiae, avgentariae, sulp(h)uvariae) which mostly imply fodinae. The first four show no variation from the Plural. Avrgentaria is employed only when ¢aberna or ars is the complementary noun. But there is a solitary recalcitrant example of the Singular 1 See Neue, Formeniehre 1.*, p. 7oo: ‘ Lapici- dinae ist Plurale tantum; desgleichen /atomiae oder latumiae, dies nach Charis. Exc. art. gramm. S. 99 (I. 549, 17), doch hat Varr. L.L. 5, 32, 151 auch den Sing.’ The former reference, in Gramm, Lat. ed. Keil, runs: ‘Feminina semper pluralia ... hae lautumiae \aroula.” But Barwick in the new edition (small Teubner) of Charis. (p. 36, 1. 4) omits this example. The Varro passage (new edition by Goetz and Schoell) is: ‘quod Syracusis, ubi de causa custodiuntur uocantur latomiae, etde (inde?) Lautumia translatum, quod hic quoque in eo loco lapicidinae fuerunt.’ Verrius Flaccus perhaps drew on Varro. Cf. Paul. Fest. 104, 20: *‘Lautumias ex Graeco et maxime a Syracusanis qui Aaroulas et appellant et habent ad instar carceris : ex quibus excisi sunt lapides ad extruendam urbem.’ Lautumia in the Varro passage consorts very oddly with latomiace and Japicidinae. Is it associated with AXaréma the Plural of Aaréucov (more correctly Aaroueior) ? sulpuvaria uel sulpurs as a femin: one calcaric sulpurarta | originally | nitvartae, @ lapictdinae | For ci look up thi Fodine Pliny, N.f Dig. 3; 4y | -aviae. Corp. Insc Caesar, B. Lapics Cicero, De Lat. 2, 51! eviser 2mma- officina, These y with 1, of a lf, etc. oounds we are 1 some tc.) or fodinae younds fodinae stance silver- 1e, pit, \mmi- odinae ils are fodina larum place- at the ading. avUm ; likely cause vith a of the in the n the tumtiae yaviae, show is the gular latum, erunt.’ ». eco et pellant si sunt in the stomiac wa the )? FERRIFODINAE AND SIMILAR COMPOUNDS 201 sulpuvavia which must be noted (Ulpian, Dig. 48, 19, 8, § 10): ‘in calcariam quoque uel sulpurariam damnari solent.’ First calcartam. The Thesaurus recognizes calcaria as afeminine noun. The Plural in Ammianus, 27, 3, 4 has reference to more than one calcaria. Similarly carbonaria is used in the Singular, and the possibility is that sulpuvaria is of the same type, and that not fodina, but officina is the noun that was originally understood. So sulpurarta would not fall in the same group as harenariae, nitravtae, etc., and would not be evidence for a Singular fodina. The evidence for lapicidinae can be dismissed briefly. Apart from glossaries the Plural is invariable. For convenience I append a list of relevant passages. It is quite bewildering to look up the references in the Thesaurus or in the dictionaries of Georges and others. Fodinae. Cato, Orig. (Gell. 2, 22, 28): Varro, R.R. 1, 2, 22: Vitruvius 7, 7, I: Pliny, N.H. 33, 78; 33, 98: Ulpian, Dig. 7, 1, 13, 5; 24, 3,7, 14; 27, 9, 3, 6: Gaius, Dig. 3, 4, 3 39, 4, 13: Paulus, Dig. 50, 16, 77: Ammianus 22, 15, 30; 24, 4, 21. -aviae. Cicero, Pro Cluent. 13, 3: Varro, R.R. 1, 2, 23: Tacitus, Amn. 6, 19; Corp. Inscr. Lat. 3, 1088: Pliny, N.H. 31, 109 sqq.: Cato, Org. (Gell. 2, 22, 9): Caesar, B.G. 7, 22, 2: Livy 26, 27,2; 34, 21, 7: Ulpian, Dig. 47, 2, 52,8; 48, 19, 8, ro. Lapicidinae (or -dicinae), Plautus, Capt.g44: Varro, L.L. 5,151; RAR. 1, 2, 22-3: Cicero, De Div. 1, 23: Vitruvius 8, 3, 9: Pliny, N.H. 3, 30; 7, 195: Corp. Inscr. Lat. 2, 5181, 54 and 3, 75 (-caedtnae). J. D. Crate. ey ere ~ ~ie ee w S00 oy eae EB, 2 —_ ae THEOCRITEA. I. 95. 96 vOe ye pav ndcia xai & Kirpis yeAdouwa, AdOpy pev yeAdowa, Bapiv 8 ava Oupdv éxowa. MoDERN editors take AdOpn (or P.’s AdOpia)' yeAdowwa as=‘smiling secretly ’— i.e. hiding the smile she felt inclined to show. This involves the two difficulties of (1) the yeAdowa of 1. 95; (2) the expression Bapiv 8 dva Ovpdv éxowa. To take (1) first: If Aphrodite came smiling it is absurd to say in the next line that as a matter of fact she was hiding her smiles. This has led to Ahlwardt’s quite unjustifiable aééa, accepted by Hiller. (z) The pev . . . & suggests an antithesis. The antithesis to ‘ hiding her smile’ is ‘ showing her anger.’ So Cholmeley and Legrand. This involves an impossible use of dvéxew, and no light is thrown on it by Cholmeley’s citation of Eur. Med. 482 dverxov aot ddos awrypiv. ‘* Upholding’ in the sense of ‘remaining faithful to’ is scarcely justified by Eur. Hec. 121 dvéxwv Aextpa, (Pearson follows Jebb in rejecting a similar dvéxecv in Soph. O.C. 674.) Nor is the situation saved by taking a. =‘ keep in check.’ One could not have ‘hiding pév her smiles, keeping in check &¢ her anger.’ Besides, the whole sense is against this rendering. A. was really angry with Daphnis; she was not merely pretending. It is true that (in 1. 138) 70d’ avopGeco:, but this is a mere death-bed repentance. Is not the real explanation this: ‘A. came smiling, smiling tveacherously, but continuing her anger’? For this use of Ad@py cf. Hom. Od. 17. 80 AdOpa xteivovres. Or, if AdOpia, Call. Hymn to Ap. 105 AdOpwos efre. For 4. =‘continue’ cf. Thuc. 1. 141 a. rod€pous, and Ar. Thes. 948 a. dpyia. 3. 18 @ 7d KaXrdv rofopeioa, 73 wav AiBos. Editors have long disputed whether the MSS. Ai@os or the Schol. v./. Adzros should be accepted. Of the moderns, Wilamowitz, Cholmeley, and Edmonds stick to the MSS., Hiller and Legrand accept the v./, That this is right can be gathered from the context. The Cyclops is praising, not disparaging, Galatea; and though one cannot demand consistency from a lover, he does definitely call her (1. 39) ovk adapyavriva. Conclusively in favour of Aéos is Call. frag. 121, where the Muses are asked to place their Aurwoas xeipas in blessing on the poet’s work. The ancient Greeks did not wash as much as the modern English, and they used oil. I am indebted to Mr. E. A. Barber, of Exeter College, for the quotation. 13. 69, 70 ioria & Hpibeor perovixtiov eLexdGarpov “HpaxAja pevovres. In my notes in C.R., November, 1925, I suggested that “HpaxAéa peivayres should be read, and the passage translated, ‘they cleared away the sails at midnight, having waited for H.’ It has been suggested to me that pévovres can be kept if pecovixrwv is taken, against the order of words, with “HpoxAja. The translation will then be, ‘they were making ready the sails while waiting for the arrival of H. at midnight ’—lit., ‘ waiting for the midnight H.’ éxxa@aipew is used not of ‘clearing away,’ but of ‘getting ready,’ as it is in e.g. Plat. Rep. 361D exxaOaipev aorep avipiavra =‘ put the finishing touches to’ (L. and S.). This agrees with the schol. who paraphrases éxx. by éfarAovv. For this typically Theocritean persona. peOOVUKT Thi (1) 1. 12 (2) apa | possibili trav. The of much their wo same re 1, 127 by If a out—e.g ETTPWTAL would n: Of 2 compres kAivyn bu of «xAXAiva ‘lui, app it almost and mus hung for Fev KataBoor ‘ Scarlet the aAAo oberhalb Stiitzen vengean Ifa (repunctt ‘and the tly ’*— culties o take hat as ; quite smile ’ ossible d. 482 to’ is ecting ‘keep dé her angry nen’ y, but 17. 80 tinue’ Xitros , stick hered 10ugh I. 39) (uses cient 10uld ving n be The r the used 361D yrees itean M. PLATNAUER 203 personal use of a temporal adjective cf. 16. 93 oxvdaiov . . . ddirav, and for perovixrios itself so used Eur. Hec. 914 pevovixtios AAvpayv. 15. 123-8 4 éBevos, & xpuods, & ex AevKd EAéavros aterot otvoydov Kpovida: Ari raida hépovres, 125 mophipeo: S€ tarnres dvw paraxwdrepos trvw. a. MiAnros épe? xo tav Laplay KataBdokwv ‘éeotpwra krAiva Twdwvids THe KaAGE apd.’ trav pev Kimpis éxet trav 8 6 poddmayus “Adwus. This passage as it stands in the Oxford text seems open to four objections: (1) 1. 127 is put into the mouth of Miletus and Samos in defiance of the scholiast ; (2) ayuda is an emendation of Ahrens (with little to recommend it save palaeographical possibility) for the MSS. aAAa; (3) only one xAivy is mentioned, yet trav pév.. . tav 6. . .; (4) what does 4vw mean? That the scholiast attributes /. 125 to Miletus and Samos is not a matter of much moment. A more important consideration is that these places, famed for their wool, should lay claim rather to the woollen rdamnres than to the xAivy. For the same reason Legrand’s full stop at épet, whereby 1. 125 is spoken by Miletus and 1, 127 by Samos will not commend itself to many. If dAAa is to be kept, we must suppose with Kreussler that a line has dropped out—e.g. éorpwra: xXriva ra Kimrpidc ra xadg aira (after 1. 126), Hiller; or, better eotpwra: §’ ddAa ra Kimpidts ra xaptéoos (after 1. 127), Mr. Barber. Homoeocatarcta would naturally account for this. Of rav pev .. . trav &’ . . . Cholmeley says ‘compression.’ But surely this is compression run mad, Better Edmonds’ tiv pév... rov & .. ., supposing one kXivn but two tdrnres. But it is difficult to refer back to 1. 125 over the head of xAiva in 1. 127. Still less likely is Legrand’s rdv pév (Rossbach) ... tav 6’. . ., ‘lui, appartient 4 Cypris; elle, 4 Adonis.’ L. 128, one would have thought, makes it almost certain that a line containing a reference to a second xAivn has dropped out and must be supplied. And once one tampers with the MSS. it is as well to be hung for Kreussler’s sheep as Ahrens’ lamb. Few editors face the difficulty of évw, Valkenaer, bracketing padaxwrepa .. . katafdoxwv, and putting a colon at xAiva, took rdrynres as in apposition to «diva. ‘Scarlet rugs on top, a couch is laid . . .; another for... .’ This gets over the dAAa difficulty, but it is, I submit, impossible Greek. Hiller comments: ‘dv, oberhalb der v. 123 f. beschriebenen Schnitzereien, Bildwerke, u.s.w., welche die Stiitzen des Lagers bildeten.’ This is the doctrine of hidden ornament with a vengeance. One does not cover up carving and statues with scarlet ramnres. If a line has dropped out may we not suppose further confusion and rearrange (repunctuating) as follows ? 123 & €Bevos, & xpvads, & éx AevKG EAEHavTos 124 aierot oivoydov Kpovida Avi raida deportes. <éorpwra kriva ra Kirpidc ta xapeeoog, > 127 éoTpwrat kAXiva rwHdovids TY KarAw arAa° 128 trav pev Kirpis exer trav 8’ 6 poddmaxus “Adwris* 125 ‘moppvpeo: St rdryres dvw padrakwrepor Urvw’ 126 a MéiAnros épet yo trav Dapiav xataBdoxwv. ‘and there are upon the x Xivac rugs... . 24. 13-16 rapos ap’ aiva réAwpa Siw rodvpHxavos “Hp kvavéas ppiccovras trd oreipawre Spdxovtas Spoev eri rAariv ovddv, 60 orabpa KotAa Ovpdwy oikov, dreAnoaca hayeiv Bpépos “HpaxAja. 204 THEOCRITEA The objections urged against 1. 16 are two: (1) That ofxov is pointless ; (2) that areAjcaca payev is bad Greek. Hiller (following Paley) heroically—and perhaps wisely—excised the whole line. Stadtmiiller’s eZxev with a predicative xotAa (‘where the gate posts yielded inwards’) is not justified by the so-called parallel in Soph. O.T. 1262 éxAwwe Kotha xApOpa. Folding doors may ‘yield inwards,’ not door posts. Briggs’s tentative suggestion of devdv, gpixrdv, or otAov taken adverbially with areAjoaca is weak. Kreussler’s ofyev with oraOud as subject and xoiAa Ovpdwv as object is very far-fetched and its meaning far from clear. Legrand reads épxw; ‘avec des menaces accompagnées de serments elle leur avait enjoint de dévorer....’ This gives admirable sense, but it is impossible Greek. Legrand cites in support of that construction Ap. Rhod. 3. 607 ameiXee va 7’ épvcGa:; but this does not prove that the subject of dmecAciv can be different from that of the infinitive it governs. Valkenaer saw the difficulty when he suggested are:Ajoavra, referring back to réAwpa. ‘Sed nimis distat’ was Briggs’s sage comment. A possibility (as has been suggested to me) would be to interchange époev and oixov, constructing dpaeev with ¢ayetv—as in Hom. Jl. 13. 794—and taking azretAy- gaca absolutely. For this aorist, where one might expect present participle, cf. Hom. J/. 23. 186 as dar’ dradjoas. 24. 89,90 Kdyxava § doraddov EvXr’ erotpdoar’ 7 Tadvovpov 7 Parov 7 avepwe Sedovnpevov atov &xepdov. L. and S.: dxepdos, 4} (less frequent 6, Theoc. 24.90). I can find no other case in which &yxepdos is masculine,’ nor is such a thing in itself likely. In face of the other genitives may we not suppose that Theocritus wrote dxépdov? For atov as a sub- stantive cf. Hes. Opera 743, where it is used of the ‘dead’ of the finger-nail. For Theocritus’ use of the adjective as substantive see 1. 49 trav tpw£ipov (with Hiller’s note) ; for this ellipse with a genitive cf. 5. 61 rav wavr@ sc. yav. 23.51 .. . Bpdxov & éBadXc rpaxnAw. The MSS. ¢Badde will not scan and must be corrected. Either the Juntine éuBaXdXe or the Aldine évéBaAXe would do, but in face of Hom. J]. 24. 272 emi 5€ xpixov éxropt BaddXov, ‘they thrust the pin through the ring,’ I imagine the author here wrote eréBad Xe, M. PLaTNAUER. 1 In Schol. MV Hom. Od. 11. 287 V has r& showing that r& is a mere slip. dxépdur, but M has rh, and both have adriu, PR both o! first pri has bee lines of contrib study o other V mentar’ to S an who ha other s« tion.‘ ‘ Here i comes { are dirt have bi Profess supplen and tha the stat this thi There | Arabia. Triphy. for fror Pancha and D | preferri find D not see difficult compile Servius here re; a critic’ the sam was he quae di imperat legimus Bitiam 1 The . 165, in XXXVI. Notes on of Virgil, 2 Cf. « 286), and 3 See! ) that rhaps where Soph. posts. with wy as : leur ssible yn. v’ | that ested ment, y and “etAn- e, cf. se in other sub- For ller’s itine pikov yrote SERVIVS AVCTVS AND DONATVS. PROBABLY few scholars would now doubt that at least the bulk of the matter both of Servius’ commentary on Virgil (S) and of the additions to it which were first printed by Daniel (D) descends from Donatus. The problem of these additions has been approached by a number of writers from different directions, and different lines of evidence have been found to converge on one conclusion. An important contribution to the discussion has recently been made by J. J. Savage in a thorough study of the multifarious scholia contained in the well-known Bernensis 165 and some other Virgil manuscripts. He shows that some of these scholia go back to a com- mentary which was fuller than that of Servius, and which moreover was related both toS and to D.2 Other fragments of such a commentary are to be found by those who have the patience to sift the rubbish of the Latin glossaries,*? and also, among other sources, in Isidore, who is all the more useful for his habit of verbal transcrip- tions When material of this kind, is found, there is a strong temptation to say ‘Here is Donatus,’ and one can hardly doubt that that is where much of it ultimately comes from; but the question how far we may assume that the words of Donatus are directly reproduced can hardly yet be regarded as satisfactorily answered. I have before ventured to doubt whether Donatus was the tmmedtate source of D.° Professor Rand and Mr. Savage think that he was, that the compiler of SD supplemented S with material which S had not already taken from the same quarry and that where he found Donatus corrected by S he did not reproduce from Donatus the statement to which S had objected. But there are passages where, prima facie, this theory does not seem to suit the facts. They are not all of the same kind. There is no real difficulty, for instance, at Geo. Il. 139, where S says ‘ Panchaia: Arabia, ut diximus supra,’ and D adds ‘ Panchaia: regio Arabiae, ubi et templum Triphylii Iouis,’ etc. The latter note is inconsistent with the opinion of Donatus, for from the Liber Glossarum (C.G.L., vol. V. 230, 22) we know that he described Panchaia as ‘ipsa Arabia,’ not ‘regio Arabiae’; so that S here agrees with Donatus and D differs. But it is probable enough that Donatus mentioned both views, while preferring the former. At Aen. IX.675 (672 in Thilo), as I have before pointed out, we find D duplicating a criticism of Donatus which already exists in S. This it does not seem so easy to explain away, and Mr. Savage, I rather think, minimizes the difficulty. He argues that my doubts would only be justified if we supposed that the compiler of SD ‘worked on rigid scientific lines, never repeating what he found in Servius.” What I find difficult to explain on Mr. Rand’s theory is not that the compiler here repeats something from Donatus which S has already taken, but that he records a criticism of Donatus, and that, too, where S has already criticized him on exactly the same lines. Where did he get this criticism? If he learned it only from S, why was he not content with what S had written? The S note (on the words ‘ portam quae ducis imperio commissa’) is this: ‘scimus quod Mnestheo et aliis ducibus imperauerat Aeneas, ad petenda auxilia proficiscens, ne portas aperirent .. . item legimus Mnesthea et Serestum ductores fuisse et rerum magistros, non Pandarum et Bitiam ; unde apparet quia hunc locum male inteilexit Donatus dicens commissam 1 The Scholia in the Virgil of Tours, Bernensis 165, in Harvard Studies in Class. Phil., vol. XXXVI., pp. 91-164. Cf. the same author's Notes on Some Unpublished Scholia in a Paris MS. of Virgil, in T.A.P.A., vol. LVI., p. 229. 2 Cf. e.g., Harvard Studies, p. 147 (on Aen. II. 286), and pp. 156 sqq. 3 See my chapter on Virgil Scholia in Ancient Love in Mediaeval Latin Glossaries, pp. 46 sqq., and my edition of the Abstrusa Glossary in Glossaria Latina, vol. III. (Société Anonyme d’Kdition * Les Belles Lettres,’ 1936. ‘Cf. Philipp, Die histovisch - geographischen Quellen in den Etymologiae des Isidorus von Sevilla, I, (Berlin, 1912). 5 Ancient Lore, pp. 57 Sqq. 206 SERVIVS AVCTVS AND DONATVS portam, id est creditam Pandaro et Bitiae ; qui duces non erant. Cornutus uere et melius sensit dicens portam quae ducis imperio commissa fuerat, hoc est clausa, eam aperuerunt. commissura enim dicitur tabularum coniunctio, sicut Cicero in Fun- daniana meminit. ergo melius est ut commissam dicamus clausam, quam creditam Pandaro et Bitiae.’ D says ‘ Ducis imperio commissa: aut quae eis fuerat commissa imperio absentis Aeneae ; aut, quod est melius, quae imperio Aeneae fuerat com- missa, id est clausa; unde et commissurae dicuntur coniunctiones tabularum. secundum superiorem sensum contrarium est quod dixit ‘ rectores iuuenum et rerum dedit esse magistros’ (i.e. Mnestheus and Serestus, cf. line 171). Aeneas enim abscedens principibus omnium rerum commiserat curam.’ D, like S, prefers the interpretation of Cornutus, and for the same reason. One may think it very likely that Donatus, while expressing his own opinion, quoted Cornutus also. But if the compiler of SD had Donatus actually before him and was proceeding as Mr. Rand suggests, why does he here duplicate at length the very criticism which he found inS? Mr. Savage says that D has ‘accommodated himself to his basic Servian text’ by inserting the words ‘quod est melius’; but he has really done more than that, since he goes on in the next sentence to explain why it is better. If we suppose that he found this criticism of Donatus in his source, it is easy to account for the duplication ; it was due to inadvertence. On the other view he must have deliber- ately remodelled the text of Donatus to make itSay just what he saw that S had said already. (I do not know whether the fact that the D note on this line stands before the S note has any significance ; if it means that here he consulted his non-Servian source first, then he must have found the criticism there.) I add one more instance. At Geo. I. 260 D says ‘Cato properare dicit eos qui prima quaeque ordine suo mature transigunt, festinare autem illos qui multa incipiunt eaque adnectendo nec terminando praepediant.’ Professor Lindsay has called my attention to the fact that Isidore (Diff. I. 440) also quotes Cato for the distinction between these two verbs. Now Isidore’s statement is made in terms which are different from and briefer than those of D, and he gives a whole extra sentence of quotation which D does not give: ‘Marcus Cato sic distinguit dicens “qui unumquodque mature transigit, is properat; qui multa simul incipit neque perficit, is festinat. ego unumquodque quod adortus eram transigebam.”’ Where did Isidore get this quotation? Not from Cato himself; nor from Festus (268, 2, Lind.), for Festus does not give the additional sentence; nor, apparently, from Verrius Flaccus (even supposing that Isidore had access to him, which is very unlikely), for Gellius at XVI. 14, where he makes the same distinction, is clearly using Verrius Flaccus,! yet he too, like Festus, gives the quotation from Cato with- out the extra sentence; nor from Nonius (441, 18 M), who got the quotation from Gellius. Isidore seems generally to get information of this kind either from Festus or from commentaries on Virgil. This particular quotation is not in Servius: where is he more likely to have found it than in Donatus? Now compare the terms in which D and Isidore describe the distinction drawn by Cato, and see how hard it is to assume that in D we have the very voice of Donatus. Isidore’s version rather than D’s gives the impression that it is a quotation of Cato’s actual words. And so it is, for it agrees verbally with Gellius’ version as well as Festus’ (as far as they go), and Gellius is careful to say that these are Cato’s own words: ‘ uerba sunt ipsius ex oratione quam de suis uirtutibus habuit.’ If, then, Isidore got the quotation from Donatus, D’s note not only omits part of it (which would not be surprising), but represents a rehandling of Donatus. Was this done by the compiler of SD? At least it was not done for the sake of abbreviation, which we should expect to be his motive. H. J. THomson. BanGor, N. WALEs. 1 Cf, Gell., l.c., Verrius Flaccus rationem dicere uolens differentiae huius, etc. IN tl assem va know by nam ’—al fortunam meter, wi and Lore the popul The before At A. L (followed fuit. Tt a translat case the « dheidov Kai shown by based on B. T C. T ‘tanti qu habetis’ ( Marx is v Marx) ha Latin: ¢s. [ Juv. But 3 apparent! figure sim form— Stran poétrques a find it in correct fo ed. Haup p. 873), al Aeneas S quoted b The Mayor, in 1 The E metre, not re et eam Fun- litam issa com- rum. erum enim s the ikely f the Rand ‘ound rvian than pose r the iber- said efore rvian ; qui nulta Y has - the erms xtra cens eque here 8, 2; from very arly vith- rom ‘stus here arms rd it ther d so go), S ex rom but At . his THE HISTORY OF A PROVERB. In the Classical Review (1925, p. 118) I quoted, for Petronius 77. 6 ‘assem habeas assem valeas,’ a proverb (from Mayor on Juv. 3. 143, addenda) unnoticed as far as I know by other scholars—‘ quantum habebis tantus eris; frange lunam et fac fortu- nam ‘—and suggested that we should invert and correct—‘ frange lunam [et] fac fortunam ; quantum habebis tanti eris’—thus getting an accentual trochaic tetra- . meter, with rhyme in the first half (the type dealt with by Usener, K/. Schriften 2.256, and Lorenz, preface to his edition of the Pseudolus, p. 39), which could be added to the popular trochaics collected in Baehrens’ Poet. Lat. Fragmenta. The proverb had different forms, but I can find no trace of the frange lunam part before Augustine (Discipl. Christ. 12). There seem to be three forms: A. Literary, as quoted by Seneca, Ep. 115. 14, in a passage of tragic iambics (followed by a translation from Euripides), ‘ubique tanti quisque quantum habuit fuit.’ The iambics are clearly Senecan, not republican, and therefore unlikely to be a translation from the Greek (unless by Seneca himself for the occasion).! In any case the closely parallel passage quoted by Plutarch zepi ¢sAorAoveias § 7: xépdacve [xa | peiSov kal rorovtrov (rocvov?) vopsle cavrdv a£cov doov av éxys (mutilated iambics ?) is shown by the context not to be Seneca’s original. The literary use is obviously based on a popular saying. B. There is no exact parallel to the Petronian form ‘assem habeas assem valeas.’ C. The commonest form is like the first half of Augustine’s quotation—Horace ‘tanti quantum habeas sis’ (Sat. I. i. 62); Apuleius ‘tanti re vera estis quantum habetis’ (Afol. 23). These support my change in Augustine’s proverb, and show that Marx is wrong in rejecting F. Dousa’s conjecture in Lucil. 1120 ‘quantum (tantum, Marx) habeas, tantum (? tanti) ipse sies, tantique habearis’ (tantum sies is doubtful Latin: esse aliquid scarcely justifies it.) [Juv. 3. 143 has only a slight resemblance. | But in the Middle Ages the second half comes into its own. It was hit upon, first apparently by Marbodus (Bishop of Rennes, died 1123), to illustrate the rhetorical figure similitey cadens in his treatise de Oynamentts Verborum. It here appears in the form— fac tibi fortunam, festina frangere lunam, et contra factum (1. fatum) faciat (-e¢) te cura beatum. Strangely enough it does not appear in the authors printed by Faral in Les Arts poétiques du XII¢ et du XIII Siecle, though they make free use of Marbodus, but we find it in the famous Graecismus of Eberhard of Bethune (thirteenth century) in correct form (in Wrobel’s edition), and hence in a ‘verbreitete,’ mediaeval Ars Poetica, ed. Haupt, Ber. Sachs. Ges. Wiss., 1848, 53 ff. (see Norden, Antske Kunstprosa, p. 873), and finally at the end of the Middle Ages in the Artis Rhetoricae Praecepta of Aeneas Sylvius Piccolomini, the future pope (1456). fac tibi fortunam festinam (/eg. -a) fingere (frang-) lunam et contra fatum faciet te cura beatum quoted by Polheim, Lat. Retmprosa, p. 468). The proverb does not figure in Werner’s Sprichworter des M.A., nor, according to Mayor, in Erasmus. W. B. SEDGwICK, 1 The Euripidean quotation is Senecan in tion by Seneca. Our passage is Senecan in metre metre, not in style—almost certainly a transla- and in style. I doubt if it is a translation at all. SUMMARIES OF PERIODICALS LITERATURE AND GENERAL. American Journal of Philology. XLVII.4. October, November, Decem- ber, 1926. P. Haupt, Etymological and Critical Notes. Five notes, of which four are very brief, dealing with the ‘ bdellium and onyx’ of Genesis, the ‘sesame’ (Arabic simsim) of Ali Baba, the meaning of Latin bucina and Greek Bvxdvn, the pronunciation of Ethiopic, and the interpretation of the allegory of the two shepherds in Zechariah xi. K. Malone, Agelmund and Lamicho. An examination of the story told by Paulus Diaconus (Hist. Lang. I., §§ 14-18), with a comparison between his version and the earlier account in the Origto Gentis Langobardorum, and an attempt to reconstruct the primitive Scandian version of the wars of Hunding and Wulfing. G. Hirst, The Significance of Augustior as Applied to Hercules and Romulus: A Note on Livy I. 7.9 and 1.8.9. Argues that in both these passages the word augustus is applied to a hero frequently taken as the prototype of the Emperor, and its use therefore covers a reference to his newly assumed title. Examines all the instances of the word in Livy and compares them with those in other writers, especially Vergil. W. A. Old- father, The Alleged Avarice of Sophocles. Explains the bitterness of the reference to the ‘prosperity’ of Sophocles in Pax 695-6, by supposing that he was himself a paxarporro.os in succession to his father Sophillus, and that his interests were there- fore diametrically opposed to those of the landowning class to which Aristophanes belonged. H. H. Law, Hyperbole in Mythological Companions. Collects a number of instances to show that the form of comparison in which a mortal is declared to be superior to a god, was common in Greek poetry no less than in Latin; and argues that there is therefore nothing distinctively ‘Plautine’ in such passages as Miles 11-12, etc., but that Plautus may well have found these comparisons in the Greek plays very much in the form in which he presented them. XLVIII. 1. January, February, March, 1927. J. C. Rolfe, Marks of Quantity in the Monumentum Antiochum. Gives statistics of the use of the afex and J longa in this inscription, and argues that these marks were intended chiefly as guides to the correct pronunciation of difficult or doubtful words. H. C. Nutting, Quid Hoc Homine Facias? Protests against ‘the careless appeal to comparative syntax to settle questions that should be approached in another way,’ and argues that the only right method of determining the exact force of the ablative in such phrases with facere, fieri, and esse is by careful examination of all known instances. A. W. Van Buren, Inscriptions from Rome. A brief description of sixteen hitherto unpublished inscriptions (one in Greek iambics) from the collection of the late Dr. G. N. Olcott of the American School of Classical Studies. D. M. Robinson, Inscriptions at Dineiy. Transcribes, and briefly annotates six inscriptions from Dineir, the ancient Apamea, and gives an account of a fragment of a sar- cophagus discovered in the same region. W. Kirkconnell, Linguistic Lacontsms. Gives a table of ‘the number of syllables required in rendering the same text’ in forty different Indo-European languages, and concludes that ‘the language with the smallest number of syllables tends to be the one with the fewest inflections.’ E. G. Sihler, Polybius of Megalopolis. A study divided into several sections: (1) The life of Polybius and his reasons for writing his history as and when he did; (2) his attitude towards t position : personali Her! B. S forwardly doctrine, Ekloge V1 insolence. Atticus un scarcely t sellers ar friend, an: prices, et developm round abc views abo Em neues | lished by ep. suppl. 2 the law of theatre, tl Misz TaXElaVv En imaTLov Xenophons Gorg. Pal. tion of J.C kritos’ in Pythagore Evethimios. 1, 734- Mnen L. Ra R. mainta in Ter. An Here D. h town’ (W. sctlicet iror language o Lindeman: dotus, dati: copies. 17 orthograph found in h J. D. Meer to the pas cupmaice ¢ pevos—arg wédv£, the his notes Decem- are very > simsim) ation of ariah xi. Paulus and the ruct the rst, The y 1.7.9 o a hero overs a word in A. Old- rence to imself a e there- ophanes mber of d to be argues wes as ; in the tatistics . marks loubtful careless shed in force of n of all tion of llection D. M. riptions @ Sar- Gives n forty ith the E. G. e life of attitude SUMMARIES OF PERIODICALS 209 towards the Roman state; (3) the audience for whom his work was intended; (4) his position as a political thinker; (5) his criticisms of other historians; and (6) his personality. Hermes. LXI.4. 1926. B. Snell, Die Sprache Hevakits. The fragments must be interpreted straight- forwardly as descriptions of human experience, The Adyos, besides being H.’s doctrine, is the meaning implicit in the visible world. W. Baehrens, Zu der siebenten Ekloge Virgils. The victory of Corydon over Thyrsis is the victory of modesty over insolence. The poem illustrates Vergil’s character. R. Sommer, 7. Pompontus Atticus und die Verbreitung von Cicevos Wevken. In republican Rome a book-trade can scarcely be said to have existed, and attempts to distinguish publishers from book- sellers are futile. Atticus was neither; his services to Cicero were only those of a friend, and his propagation of C.’s works was informal. Various smaller matters— prices, etc. O. Schroeder, IPOKNH. The different forms of the legend and its development. W. Kranz, Diotima von Mantineia. D. is historical, and was in Athens round about the beginning of the second Peloponnesian War. An examination of the views about Eros attributed to her by Plato shows their consistency. L. Wickert, Ein neues lateintsches Grabgedicht. Commentary on the obscure inscription first pub- lished by Bendinelli in Not. d. Scav., 1923, pp. 357 sqq. (= Lommatzsch, Carm. Lat. ep. suppl. 2121). E. Bethe, Programm und Festzug dev grossen Dionysien. Exegesis of the law of Euegoros (Dem. In Meid. 10). The xaos is the great procession to the theatre, the zouzy the reception of the statue in the Academy on the evening before. MiszELLen.—A. Wilhelm, Zu Appian und Dion von Prusa, In App. Syr.1 for «is TAaXelav emiTetxioparos Opuynv read oixodouyv: in Dion’ Dig. 14 for éav tis imatrov amrodg KiBinrov read kepparstov. J. Morr, Des Gorgias Palamedes und Xenophons Apologie. X.in Ap, 27 and Mem. lV. 2. 33 shows verbal reminiscences of Gorg. Pal. E. Preuner, Zum attischen Gesetz iiber die Speisung im Prytaneion. Restora- tion of J.G. 1.2, No. 77. M. Wellmann, Zu Demokrit. The quotation from ‘ Demo- kritos’ in Clem. Al. Stvom. I. 15, 69 is not from the Abderite but from the neo- Pythagorean of the third century. F. Hiller von Gaertringen, Nochmals Apollon Evethimios. On I.G. XII. 1, 730, referred to 63/2-38/7 B.c. Also restores J.G. XII. 1, 734+ Mnemosyne. LV. 1. 1927. L. Rank continues his Donatea from Vol. LII. Among other passages in which R. maintains that D. has preserved the right explanation is the famous conversation in Ter. Andra (185), ‘ Meum gnatum rumor est amare.’ ‘Id populus cuvat scilicet.’ Here D. has two notes pointing to the true explanation, ‘ Why, it’s the talk of the town’ (Wagner, Parry, Ruhnken, Magin wrongly follow Eugraphius in making sctlicet ironical). A.G. Roos, De Avriani Indicae Dialecto Ionica. This study of the language of the Jndica is supplementary to the works of Allinson (A./.P., 1886, p. 203), Lindemann, and H. W. Smyth. Recently discovered papyrus fragments of Hero- dotus, dating from the time of Arrian, show the same variety of forms as our own copies. Therefore, with the exception of forms arising from itacism or false orthography, it is to be assumed that Arrian is imitating the variety he himself found in his copy of Herodotus rather than that Cod. A of the Jndica is at fault. J. D. Meerwaldt, De duobus puertlis sermonts exemplis apud Theophrastum seruatis, referring to the passage (Char. V. 5), discussed in Vol. LIII., p. 350—xai rois pév (asdiors), cupraifev avtds A€ywv* aoKos, TéAeKus+ Ta Se emi THs yaorpds éav Kabevdew dpa OAL Bd- pevos—argues that doxos, réAexus, represents a child’s way of pronouncing dpricxos, wéAv£, the latter word being akin to wéAAa and Lat. pelvis. C. Brakman continues his notes on Livy. A. J. Kronenberg contributes notes on Plutarch’s Lives. P 210 SUMMARIES OF PERIODICALS J. M. Nap, Ad Catonis Librum de Re Msitan, attempts to trace the influence of Cato’s lost work on later authors. The most noteworthy instance is perhaps the military oath against theft, which was administered in the consulships of Laelius and Cornelius Scipio Asiaticus (Gellius XVI. 4, 2, quoting Cincius), i.e. 190 B.c. The | oath was probably instituted in 191, and in that year Cato was military tribune. Hence the probability that Cincius’ information about the oath is derived from Cato, De Re Militar. P.Groeneboom has notes on Aeschylus’ Prometheus. He observes that whereas the Prometheus has been unjustly under suspicion as having 640 words not found in the remaining six Aeschylean plays, 596 such words are found in the Persae. W.E. J. Kuyper, [povwris, argues that this word, found thrice in Eur, and once in Aesch., means not ‘ pronus’ but ‘imbecillus.’ W. Voligraff, [lvpacrpov, maintains that this word, the true form of which is rvpavorpov (the letter v being written above the a in the papyrus of Herodas), means a pair of ‘tongs.’ L. van Liempt, De Testimonio Flaviano, shows that there is no reason on grounds of style to dispute the authenticity of the reference to Christ in Josephus (Ant. Jud. XVIII. 63). P. H. Damsté, De duobus locis Taciteis (Hist. V. 3 and 15), proposes in the former passage to read cum grex asinorum agvestium PASTUM tm rupem nemove opacam concessit. Pastum became pastu, then was corrected to ¢ pastn. In the latter passage for neque ut in pedestri acie cominus [Med, cominus minus] certabatuy he would read neque... communs nisu certabatur. Philological Quarterly (Iowa). VI. 3. 1927. : Roy C. Flickinger, A Study of Tevence’s Prologues. Exhaustive and valuable. Philologus. LXXXII. 2. 1926. S. Weinstock, Die platonische Homerknitik und thre Nachwirkung. Plato’s criticism of Homer (and of poetry in general), the starting-point of all later theories of poetry— e.g. in Aristotle, Plutarch, the Stoics, Epicureans, and Neoplatonists. Stoic allegory and anti-platonic spirit of rhetoric discredited Plato’s views. L. Weber, Tellos, Kleobis, und Biton. Herodotus (I. 30) owes the stories to written sources (emanating from Athens and Delphi respectively) as well as epigraphic. P. Thielscher, Zur Maniliusiiberlieferung. Maintains that G is a copy of L as against Garrod’s view that the two MSS. are descended independently from a common archetype. Contests Garrod’s views on several other points. Manilius as the poet’s name has little authority ; better to call him Fronto. O. Schissel, Lollsanos aus Ephesos. An account of this sophist’s life and of his system of rhetoric. J. Balogh, Voces Paginarum II. Continues article in previous number, illustrating thesis widely from late antiquity and Middle Ages. Attributes decay of reading aloud in part to Christian dislike for pagan cult of rhetoric. Miscellaneous: T.Stettner, Evidence of MSS. (a lacuna), proves that Isidore used Cassiodorus in Onigines II. and III. LXXXII. 3. 1927. F. Zucker, Zur Textherstellung und Evklivung von Philodems V. Buch Iepi Ilotnpadrwv. Corrections and interpretations of Jensen’s edition, many having been approved by him. Also excursus on eipnorAoyeiv, -Aoyia, -Aoyos. W. Ruppel, Politeuma. Bedeutungsgeschichte eines staatsvechtlichen Terminus, I. Traces history of this word in literary and epigraphic texts, with emphasis on Hellenistic use. W. Ensslin, Die Demokratie und Rom, Discusses democratic movements in the second century B.c,, with special reference to the Gracchi. O. Gléckner, Die Gottes- und Weltanschauung des Celsus. Emphasizes the Platonic element. Miscellaneous: A. Gudeman replies to G. Meyer’s criticism (Philol. LXXXII., pp. 67 sqq.) of his dating of Minucius Felix. P. Pfister argues that Gordium was world-centre on Early Ionian maps; Alexander knew this. The Editor: A note on the Birchner- Dorpfeld controversy about Leucas-Ithaca. Su Dr Grossen Su} H. Athenisc des Arts Rh Ch: act of t papyrus The ger work in Lampri V. The College Vol. IT] of punc' enyprxar, ed. Nau emendat C. Cich« his Romi the Ale: of Thra passages Riv A, | ancient | G. Pasgq expansio La pace « that the | years. | offered | mente’ de made by P. Maas éoti Ovye Pi AadEerAd (€vdrviov) V. regar N.S. IV L’ argome: Necrolog Cazioni r N.S A. R from Ri > of Cato’s e military elius and B.c. The y tribune. rom Cato, © » observes 540 words nd in the e in Eur. lvpacrpov, r v being > L. van f style to VIII. 63). he former n concesstt. for neque neque ..-+ uable. 5 criticism f poetry— c allegory er, Tellos, emanating cher, Zur view that Contests has little n account inavum II. antiquity dislike for a lacuna), uch ILepi 1y having Fy Ruppel, history of listic use. ts in the Die Gottes- sllaneous : q-) of his centre on Biirchner- SUMMARIES OF PERIODICALS 211 Supplementband XIX. Heft 1. 1927. Dr. V. Tscherikower, Die Hellenistischen Stidtgniindungen von Alexander dem Grossen bis auf die Romerzeit. Pp. xi+216. Supplementband XIX. Heft 2, 1927. H. Hommel, Heliaia: Untersuchungen zur Verfassung und Prozessordnung des Athenischen Volksgerichts, insbesondere zum Schlussteil dey "AOnvaiwv ITLoAtreia des Amsstoteles. Pp. viii+149. Rheinisches Museum fiir Philologie. Heft. Chr. Jensen, Menanderstudien. 1. J. gives the text of the opening of the fourth act of the Epitrepontes, accepting D. S. Robertson’s view that the pois of the Didot papyrus is Pamphile’s speech to her father. R. Vélpel, Zum Schauspielerproblem. The general limitation to three actors is due to the necessity of giving them equal work in the three trilogies in view of the agon. K. Ziegler, Plutarchstudien 1V. The Lamprias letter was composed by someone who knew Plin. Epp. III. 5. 1. V. The Seitenstetten MS. comes from a monastic library, probably from the Jesuit College in Vienna. VI. Notes on a number of passages in the Lives, edited in Vol. III., fasc. ii., of the new Teubner text; they conclude with remarks on methods of punctuation. F. Thedinga, Die Paraenese in des Porphyrios Schrift Tept droyijs euyvxov, argues that I. 27-57; II. 37-46, 49-52; IV., p. 20, pp. 264. 2-266. 13, ed. Nauck; III. 26, from p. 222. 25, and 27 (all of which he translates, giving emendations), are taken from Numenius, and discusses N.’s dualistic philosophy. C. Cichoines, Der Astrologe, Ti. Claudius Balbillus, Sohn des Thrasyllus. Supplements his Rémische Studien, 390 sq., by identifying the Balbillus in the letter of Claudius to the Alexandrians, and of an Ephesian inscription with the B., who was the son of Thrasyllus. J. Weidgen, Zum Thukydidestext I. Emendations of a number of passages in Bk. [., including zpooddo:s for rporddwv in 13. I. Rivista di Filologia. N.S. V. Fasc. 1. 1927. A. Rostagni, Poesia ed estetica classica. An inaugural address on the drift of ancient poetry away from mythological subjects to the varieties of human experience. G. Pasquali, Un monelogo de Captivi. Lines 516 sqq. are a rather inappropriate expansion of something found by Plautus in one of his models. G. De Sanctis, La pace di Nicia. A criticism of the view, adopted by Ferrabino in L’tmpero atentese, that the peace was a successful move which postponed the fate of Athens for seventeen years. In reality it lost Athens the greatest of all her opportunities—the opportunity offered by the end of the Argive-Spartan agreement. L. Pareti, J supposts ‘ sdoppia- menti’ delle guerre servili in Sicilsa. A defence of the authorities against attempts made by Ciaceri and Giacobbe to find in the first war retrojections from the second. P. Maas, ’Avreyévas Ovydrnp. In Schol. ad Theocr. 17, 61, read 7» yap Bepevixn éotit Ovyarnp ’Avriydvns tis Kaoodvdpov trot ’Avtiratpov dbeAPov éorovdaxdtos «.7.4, A. Vogliano, Ancora [ VIII mimiambo ds Heroda (€virviov). Reply to R. Herzog (Philologus, 1926) on the sex of the narrator, which V. regards as female. A. Vogliano, Ji frammento tragico fiorentino (vide Rivista, N.S. IV., pp. 206 sqq.). Suggestions by v. Wilamowitz and Maas. S. Luria, L’argomentazione di Antifonte in Ox. Pap. XV.1797. Recensioni. Note bibliografiche. Necrologo: Ettore De Ruggiero, by G. Cardinali. Cronache e commenti. Pubbli- cazioni ricevute. Neue Folge. LXXXVI. erstes N.S. V. Fasc. 2. 1927. A. Rostagni, I/ dialogo aristotelico Tlept rouprav, III.-IV. A long article concluded from Rivista, N.S. IV. Book I. justified poetry and dealt with the business of the 212 SUMMARIES OF PERIODICALS poet. Book II. sketched the ideal poet. And the subject of Book III. was poetic inspiration. The historical context of the work. V. Costanzi, La costituzione di Clistene. The four tribes had ceased to be of practical importance before C.’s time, and the demes were already part of the constitution. Beloch goes too far in transferring C.’s work to Peisistratus; but all that C. did was to regularize the results of a long development. G. De Sanctis, Le decretali di Civene. A running commentary on the Delphic laws published by S. Ferri in the Notiziario del Ministero delle Colome, IV. (1927). De S. does a good deal to improve the text and adds a translation. The date is not long before 324 B.c. A. Barriera, I] codice fiorentino naztonale II°. VI.-14 ¢ la tvadizione manoscnitta del De Oratore. This is another of the integyy not derived from the Laudensis. A. Solari, Sulla preistoria di Trento. The growth of the ¢ervitorium of Tridentum, its communications and its early inhabitants, among whom were Etruscans. S. Ferri, L’ iscrizione greco-latina di Lavello. Con- jectures on the rustic text first published in Riv. Ind.-Grec.-Ital. VIII. (1924), p. 151. Miscellanea: I. G. Pasquali, Summum tus summa iniuria. Remarks suggested by Stroux’s article, now published separately by Teubner, on the history and significance of the phrase, and by Riccobono’s essays on the development of Roman law in Mélanges Corml. I1. G. De Sanctis, Minucio Felice e Lucio Vero. The proper conclusion to be drawn from Oct, 18, 5-6, is that the work was written either before 161 or a considerable time after 169 a.p. II1. G. Lugli, A proposito della *‘ Forma Italiae.’” A reply to Calderini’s criticisms. IV. A. Maiuri, La rvipresa degli scavi di Evcolano. Recensioni. Note bibliografiche. Cronache e commenti. Pubblica- zioni ricevute, Wiener Studien. XLV. 1. 1926. K. Bielohlawek, MeAreo@a: und podrry. Concluded. J. Mesk, Zu den Troevinnen des Euripides. Deals with a number of isolated passages. F.Schupp, Zur Geschichte dev Beweistopth in dev dlteren griechischen Gerichtsvede, 1. An analysis of the topics of ‘ proof’ employed by the orators from Gorgias to Isokrates. The earliest handbooks in Rhetoric were severely practical. A. Perkmann, Stveitszenen tm der griechisch- vomischen Komédie, 1. In this P. analyzes the stage wrangles in Aristophanes, and shows how they lead up to the main aywv. J. Morr, Posetdontos vom Rhodos iiber Dichtung und Redekunst. A discussion of the Stoic origin of Cato’s definition of the orator as ‘ Vir bonus dicendi peritus.’ K. Mras, Zu Luktan. Shows from his own collation of the older MSS. how the text of L. has been doctored into smoothness. A. Bojkowitsch, Hirtius als offizier und als stylist. R. Holland, Bettrage zum Verstandniss dev Maecenaselegien. K. Prinz, Martialerkldrungen. Discussion of I. 13, 21, 42. E, Groag identifies the poet Porfirius Optatianus in a recently discovered inscription at Rome. F. Walter has a number of emendations chiefly on the minor Latin poets. MIszELLEN.—Kern holds that of rapa pexpdv Oeot at Sparta were nameless gods of daily life. Adler comments on Philo wepi néOns. Schuster finds a similarity between Agathias (Anth. P. XI. 382) and Plin. Epp. II. 20. 2-6. Weymann on passages in later poets illustrating the Eclogues. Laekenbacher on Sen. Apocol. 8. Hauler on Fronto, p. 127, 12-14 (Naber). LANGUAGE. Glotta. XV. 3-4. Consists chiefly of reviews of publications in1924. (Paul Kretschmer on Greek; W. Kroll on Latin Syntax, History of Language and Metric; E. Vetter on Etruscan ; A. Nehring on Latin Grammar.) Bell’s Latin Dual, Lodge’s Plautine Lexicon, Meillet and Vendryes’ Grammaire Comparée, receive favourable mention ; also Leumann’s The Latin f- and b-Tenses (IF. 42). Ipsen’s comparison of f>h in Latin with Armenian po*foh and Jur strongly rather 2 Passive P. Kret: Macedor a weak not, as I are not : XV. j. 3 botanica We may horse) is origin, al a Lat. t pound, 1 regarded sublimem: evidence Vulg. L kept its | different variatior W. Ha language psycholo Artomant (if the -a@ argues tl suggests On Kaki Phryg. o represen Prell wit: connects meant o1 Ind: A. § de Sauss (or -eevs), resumes Bliimel 1 which he not ana construc Save on examine in the ‘ books. S poetic zione dt .’s time, o far in brize the running Ministero nd adds fioventino sr of the fo. The abitants, Con. ), p. 151. ested by nificance 1 law in } proper r before ‘ Forma scavt dt ubblica- voevinnen eschichte opics of ndbooks nechisch- nes, and dos tiber . of the his own »thness. tandntss 42. &, tion at ets. gods of yetween ages in uler on Greek; VUSCQN ; Meillet mann’s menian SUMMARIES OF PERIODICALS 213 p>*fohk is questioned. Nehring disputes Ribezzo’s gremium (<*gurebh-mi-om) : Bpédos, and Juret’s windex€-*uimen-dic-s. W. Kroll, writing on The Language of Sallust, strongly attacks the theory that Sallust is influenced by the vulgar idiom; he is rather a conscious archaist—witness his employment of Asyndeton, Parataxis, Passive Constructions, etc., parallels to which are quoted from earlier Latin writers. P. Kretschmer connects 4Ac¢a : alnus : Aliso (Old German place-name); in the Macedonian word the intervocalic s has become voiced. Inscriptional 7 pus shows a weak form of the stem as against 7pws; cf. xyekovn: xéeAVs. KudaPyvacrov not, as Debrunner maintains, ¢*K via @ynvés€-Kvéaov; ethnic names in -nvos are not Attic, and are found only outside Greece. XVI. 1-2. J. Sofer, Lexical Studies in the Etymologies of Isidove of Seville, treats several botanical, zoological and anatomical expressions, names of colour, nautical words, etc. We may note ola : #Aévn; cinnibar : Goth. hinnu-bards, ‘ Kinnbart; guaranen (a kind of horse) is a contamination of late Lat. wavanio=‘equus admissarius,’ of Teutonic origin, and warantia, ‘ madder,’ a colour-word ; bata not necessarily : Sp. bahta; stringes a Lat. by-form of *stvinga. H. Jacobsohn regards sublsmis as a ‘ Bahuvrihi’ com- pound, meaning ‘he under whose feet the threshold is... The word comes to be regarded as adverbial through its constant use in the accusative case, and sublimem—>sublimen by dissimilation. M. Boas examines in detail the supposed evidence for Vulg. Lat. subj. wolim, and rejects it. K. Rocher attacks the theory of Vulg. Lat. au>o0. The evidence of the Romance languages shows that, e.g. aurum kept its diphthong (Roum. aur, Old Prov. aur) ; cdda-cauda, olla-aula belong to quite a different category (Roum. coadd, Old Prov. coza: cdda), and display I.-G. grade- variation. ©. Danielsson writes on Etruscan Inscriptions from the District of Comacchio. W. Havers classifies the instances of ‘unconstrued nominatives’ in the I.-G. languages (Nominative of Denomination, of Enumeration, etc.) and discusses the psychological basis of each class. J. Schnetz connects the (presumably) Celtic Aviomanus with (1) *man- (if the -a- is short), and so meaning ‘ well-disposed ’ ; (2) *mdn- (if the -a- is long) : Old Lat. manus, and so meaning ‘friendly and good.’ M. Schuster argues that concinnat in Naeuius, f. 37, means ‘gathers,’ ‘ collects.’ L. Radermacher suggests I Buxrys: iB (‘cry’). R.Wimmerer writes on ¢rcovoros; H. Lewry on xaxxaByn; L. Spitzer on mentula. J. Zingerle recants his former theory that Phryg. aidpw: dwpw shows the insertion of spirant -v- between the vowels; a v- represents a diphthongal pronunciation. ata, e.g. xape is also discussed. W. Prellwitz on dv@pwros and the Gk. words in -wy. With the root *ves-, ‘pursue,’ he connects dpoiios (6uoFios) and also I.-G. vivds, ‘man.’ €@eXAovripas (Od. 2. 292) meant originally ‘ volunteer oarsmen.’ Indogermanische Forschungen. XLIV. 3-4. 1927. A. Sechehaye writes an account of the Geneva school of linguistics founded by de Saussure. J. Wackernagel points out (cf. 136) that the Dem. form is Kvéa@nvaceis (or -eevs), not -aios, and that as to Kvéafov the accent is uncertain. P. Meriggi resumes (cf. 1 sqq.) his study of nasals in association with heterorganic plosives. R. Bliimel reviews recent views of the origin of the so-called accusative of specification, which he derives from a ‘ free’ local accusative (e.g. edpos); in évoya, Skt. ndma he sees not an accusative but a predicative nominative without copula. B. adds statistics of constructions in which both the whole and the part are named (always in that order, save only once whole—part) in the same case (acc. gen., dat.) in Homer. F. Slotty examines the use of the first person plural in Latin in Caesar, Tacitus, and Catullus, in the ‘sociative’ and ‘affective’ or in both functions together. Reviews of new books. INDICES IL.—GENERAL INDEX. Agar (T. L.), the (Homeric) hymn to Hermes, 45 ff. Albanian borrowings from Greek, 56 Anaxagoras, the homoiomeries of, 133 ff.; A.’s view of Empedocles, 138 archaism in Terence, go ff. Aristophanes, technical terms in, 113 ff. Aristotle's verses in praise of Plato, 13 ff. ; the composition of the Politics, 177 ff. von Arnim’s view of Aristotle's Politics, 185 f. ‘assem habeas assem valeas,’ 207 assonance and rhyme (in Sibylline Books and Virgil’s fourth Eclogue), 100 ff. Athenian hoplite force in 431 B.c., 142 ft. Austin (R. G.), Virgil and the Sibyl, 100 ff. Bacchanalian cult of 186 B.c., 128 ff. Beloch on Athenian hoplite force, 142 ff., 148 ff. Birt on Latin prosody, 1 ff. Caesar’s position, 58-56 B.c., 30 ff. Carbachius and Gelenius, their work on MSS, of Livy, 68 ff. Cary (M.) and Nock (A. D.), magic spears, 122 ff. Cicero de Amicitia, ‘lost’ MS., 53 Clodius, policy of (58-56 B.c,), 30 ff. Cook, A. B., on Tenes, 37 ff. Craig (J. D.), archaism in Terence, 90 ff. ferrifodinae and similar compounds, 199 ff. Crassus’ position, 58-56 B.c., 30 ff. Denniston (J. D.), technical terms in Aristo- phanes, 113 ff. Dionysus-cult, in Tarentum and Magna Graecia, 129 ff. Donatus, as authority for readings in Terence, go ff. extracts in the Codex Victorianus (D.) of Terence, 188 ff. Servius Auctus and, 205 f. Dorieus, 50 double axe, on coins of Tenedos, 41 f. Empedoclean doctrine as viewed by Anaxagoras, 138 Ennius, Ann., 503, 81 ethical theory in Gorgias and Hippias Maior, $7 Eudemus, 14 Euripides, the Hecuba, 155 ff. the Hipfolytus, 18 ff. Evelyn White on the text of Hesiod, 194 ff. facio, noun-compounds, 200 Farnell on Temes-Hemithea legend, 38 f., 40 ferrifodinae, 199 ff. Fletcher (G. B. A.), the Toledo MS, of Plutarch's Moralia, 166 ff. ‘frange lunam fac fortunam,’ 207 Frank (Tenney), the Bacchanalian cult of 186 B,c., 128 ff. C. Gracchus, Appian and Plutarch as authorities for, 106 f. Gomme (A. W.), the Athenian hoplite force in 431 B.C., 142 ff. Good, its divinity held by Plato and Aristotle, 16 f. Greek history, ‘ trivialities ' of, 50 f. Grube (G. M. A.), marriage laws in Plato’s Republic, 95 ff. discussion of views on Hippias Maior, 82 ff. Hall (F. W.), an English commentary on Ovid, 151 ff. Halliday (W. R.), Tenes, 36 ff. ‘harmosts,’ their powers, 161 f. Hebrew poetry, parallels with Virgil’s fourth Eclogue, 100, 104 f. Hemithea, cult in Tenedos, 37 ff. Herippidas, harmost at Thebes, 159 ff. Hermes, the hymn to, 45 ff. Herondas, R. Herzog on, 109 Herostratus, 50 f. Hesiod, the ‘ Peisistratean’ edition of, 195 ff. Hittite use of iron, 126 Housman (A. E.), prosody and method, 1 ff. Ideal theory, in Hippias Maior and other dia- logues, 86 f. Anaxagoras a possible source, 140 f. ‘intellectualist’ (technical) terms in Aristo- phanes, 1109 f. iron, its use in magical ritual, 124 ff, Jaeger (W.), Aristotle’s verses in praise of Plato, 13 ff. discussion of views on the Politics, 177 ff., 183 ff. Jason of Pherae and Sparta, 163 f. Johnson (8. K.), Livy’s fourth Decade, 67 ff. xécpos and voids in Anaxagoras and Empedocles, 138 Kyknos, legend of, 37 ff. Latin, medieval English pronunciation, 154 Leon (P.), the Aomoiomeries of Anaxagoras, 133 ff. Leukophrys and Tenedos, 37 ff. 3 Lindsay (W. A.), Ennius Amn., 503, 81 Donatus-extracts in the Codex Victori- anus (D) of Terence, 188 ff. literary (technical) terms in Aristophanes, 113 ff. Livy's fourth Decade, 67 ff. Lobel (E.), trivialities of Greek history, 50 f. Aéyos and uéGodos in Aristotle's composition, 178 magic spears, 122 ff. Magna Graecia, the source of Bacchanalian rites at Rome, 128 ff. magnetic properties of iron, their magical value, 125 f. MSS. available for Livy’s fourth Decade, 67 ff. marriage | Marsh (F. 30 ff. McKenzie pale medieval | *mox,’ 57 Nock (A. | ‘Numa’s Olympioc Orphic re Ovid,an E palatai Parke (H. 159 ff. parody ir Platnaue: Plato, Ar the mé Plautus, Plutarch P. Pompey’ Powell (: Thucy: prosody proverb, Punic W Pythago: rationali 18 ff. Roma (P Rose (H. Royds’ | Sedgwic. histor Servius Sibyllin«e Sinclair of He Socrate: Sprange of E Hecub AELIAN Vai Aemiliv (ap Aeneas xx! Aeschir ii. | Aeschy A g' Eu Sep authorities te force in istotle, 16 f. in Plato’s faior, 82 ff. on Ovid, l’s fourth 195 ff. 1 ff, ther dia- > f, Aristo- of Plato, ° 177 ff., 7 ff. edocles, 54 , 133 ff. Victori- 113 ff. f. 1, 178 n rites value, 7 ff. ft GENERAL INDEX marriage laws in Plato’s Republic, 95 ff. Marsh (F. B.), the policy of Clodius (58-56 B.c.), 30 ff. McKenzie (R.), Varia, 79 f. palatal A in Greek dialects, 194 medieval Church’s treatment of Ovid, 151 ff. ‘mox,’ 57 ff. Nock (A. D.), see Cary (M.) * Numa’s tomb,’ 132 Olympiodorus, 14 f. Orphic religion and Plato, 17 Ovid, an English (medieval) commentary on, 151 ff. palatai \ in Greek dialects, 194 Parke (H. W.), Herippidas, harmost at Thebes, 159 ff. parody in Plautus, 88 f. Platnauer (M.), Theocritea, 202 ff. Plato, Aristotle’s verses in praise of, 14 ff. the authorship of the Hippias Maior, 82 ff. marriage laws in the Republic, 95 ff. Plautus, parody in, 88 f. Plutarch’s Moralia, the Toledo MS., 166 ff. P. as an authority for C. Gracchus, 106 f. Pompey’s position (58-56 B.c.), 30 ff. Powell (J. U.), textual notes on Sophocles and Thucydides, 176 prosody and method, 1 ff. proverb, the history of a, 207 Punic War and Bacchanalian rites at Rome, 131 f. Pythagorean books, discovery at Rome, 132 rationalist interpretation of Euripidean plays, 18 ff. Roma (‘Pdéun), ancient theories of the name, 81 Rose (H. J.), ‘mox,’ 57 ff. Royds’ Virgil and Isaiah quoted, 100, 104 Sedgwick (W. B.), parody in Plautus, 88 f. ; history of a proverb, 207 Servius Auctus and Donatus, 205 f. Sibylline Books, and Virgil’s fourth Eclogue, 100 ff, Sinclair (T. A.), the so-called Peisistratean edition of Hesiod, 195 ff. Socrates contrasted with Plato, 15 f. Spranger (J. A.), the meaning of the Hippolytus of Euripides, 18 ff.; the problem of the Hecuba, 155 ff. I]1.—INDEX AELIAN :— Var. Hist. xiii. (24), 42 Aemilius Macer :— (ap. Isid, Orig. xii. 4, 24), 3 Aeneas Tacticus :-— XXvii. (7), 161 Aeschines :— li, (167), 142 Aeschylus :— Agam. (383), 176 Eum. (539), 176 Sept. (624), 176 215 Stocks (J. L.), the composition of Aristotle's Politics, 177 ff. cbpBovra (Spartan officials), 165 SUMMARIES OF PERIODICALS : Language, 55 ff., 112, 214 f. Literature and General, 52 ff., 106 ff., 208 ff. American Journal of Philology, 106, 208 f. Athenaeum, 106 f. Classical Philology, 52 f. Glotta, 55 f., 212 f. Hermes, 209 Indogermanische Forschungen, 56, 112, 213 Mnemosyne, 53 f., 107 f., 209 Neue Jahrbiicher, 108 Philological Quarterly (Iowa), 55, 109, 210 Philologus, 109 f., 210 f. Revue de Philologie, 110 Rheinisches Museum, rio f., 211 Rivista di Filologia, 211 f. Wiener Studien, 212 f. Tarrant (D.), the authorship of the HMippias Maior, 82 ff. Tartar yoke of spears, 122 f. technical terms in Aristophanes, 113 ff. Tenes, 36 ff. Terence, archaism in, go ff. Donatus-extracts in the Codex Victoria- nus (D), 188 ff. Theocritus, notes on some passages in, 202 ff. Thomas of Walsingham’s commentary on Ovid, 151 ff. iio (H. J.), Servius Auctus and Donatus, 205 f. Triumvirate, its position during 58-56 B.c., 30 ff. Two Brothers, the (Egyptian folk-tale), 41 uterque, with plural verb, 190 Verrall’s interpretation of Euripidean plays, 18 Virgil’s fourth Eclogue, 53 influence of Sibylline books on, roo ff. Zoroaster compared with Plato, 16 ws Exacra, 79 f. LOCORVM. Aetna :— (6), 2.4; (290), 4 Alcaeus :— (P. Oxy. 1788, fr. 15, ii. 18), 79 Alexis :— (fr. 147), 117; (fr. 221), 114, 119 Ammianus Marcellinus :-— (24. 4. 21), 200 ; (27. 3. 4), 201 Anaxagoras :— [Diels] (fr. 3), 139; (fr. 4), 133, 137 (fr. 5), 139; (fr. 11), 136; (fr. 12), 135, 137, 139 216 INDEX LOCORVM Anth, Lat. Ries, :— 423 (3), 7; 439 (1) 33 458 (7 8q),7 Antigonus of Carystus :— (p. 30, Keller), 51 Antiphanes :— (fr. 113), 115; Poets (fr. 191), 117 Antiphon :— (3. 8. 2), 119 Apollodorus :— iii. 15 (7), 39 Apollonius Rhodius :— Ang. iii. (607), 204; iv. (611), 24 Appuleius :— Apol. (9), 7; (23), 207 Aristophanes :— ee 444), 116; (445), 119; (447), II Av. (1288), 118 ; (1705), 120 — 119; (571-2, 573-4), 120; (807), II Equit. (100), 116; (197), 103; (216), 116; (225), 150; (461-71), 114 Lys. (28-9), 119 Nub. (130), 114; (723, 1013-18, 1160), 120; (1367), 119; (fr. 376), 115 Pax (532, 534), 116 Ran. (91), 115; (98), 113 ; (102), 26; (459), 113; (799-802, 800, 814-29, 819), 114; (827), 114, 120; (837, 839, 854), 115; (876), 119; (881, 891), 114; (897), 115; (898), 120; (902), 114; (904), 116; (gI1- 20), 117; (942), 1163 (943), 118; (946, 953), 117; (954), 115; (956), 117, 119; (963), 43 ; (1020), 115 ; (1024), 117 ; (1034), 16; (1036), 118; (1104), 117; (1108-11), IIg; (1114), 118; (1378-1410), 1163 (1409), 118 ; (1471), 26 Tagenistae (fr. 490), 118 Thesm. (948), 202 Vesp. (668), 116 Aristotle :— Constitution of the Tenedians, 39, 42 ff. Met, (25a 30; 33a 4; 89a 26, 31; goal; 93b 2, 27; 95a 4), 181; (g990b 1), 140: (1025a 6), 82 Nic. Eth. 15, 17; (1130b 26), 182; (last chapter), 183 wept dirocodlas, 15, 16 Phys. (187a 25 sqq.), 136; (188a 13 sqq.), 1373 (i. 4, 2034 19; i. 3, 302a 28 sqq.), 138 Pol, (1261b 11), 99; (1270b 12, 1272b 20), 186 ; (1218a 35), 187; (1300b 5, 1316b 35), 182 ; (1332b 26), 185 Rhet. (iii. 15), 26; (1366a 21), 182 Top. (146a 21 sqq.), 82 [Ath. Pol.] (24), 145; (37. 2), 163; (38. 2), 164 ; (49. 2), 145 Augustine :— Discipl. Christ. (12), 207 Aulus Gellius :— xvi, (14), 206 Bekk. Anecd. (31. 17), 120 Callimachus :— Hymn to Ap. (105), 202 fr, 121, 202 Carm. epigr. Buech, : 331 (3), 2,3; 331 (4), 25 (422), 3; 986 (8); 1136 (6), 6 Cassius Hemina :— ap. Pliny xiii. (84), 132 Catullus :— iv. (9), 7 Cicero :— Acad, prior., ii. (66, 87), 60 ad Att., i. 14 (4), 57; i. (16), 31; ii. (22, 23), 31; iv. (1), 343; xii. 29 (2), ro ad Fam., i. (1), 343 i. (9), 35, 36 ad Q. Fr., ii. (3), 353 ii. (6), 33, 35 de Am., (40), 10 de Divin., i. (47), 60 de Domo, (15), 30 de Fin., i. (35); iv. (5), 58; iv. (8), 60; v. (60), 58 de Har. Resp., (23), 30 de Leg., i. (54), 60 de Nat. Deor., iii. (37), 58, 65 de Prov. Cons., (11), 35 de Re Pub,, i. (20), 58; ii. (38, 65) 60 in Pis., (38), 58 in Verr., I1. ii. (174), 11 Phil. iv. (12), 58 Tusc, Disp., ii. (26), 58 Ciris, (189), 2, 3.93; (374, 481), 9 Claudian :— Gigant., 2 III. coms. Hon. (117), 4 cons. Liv. (76), 6 Cratinus :— (fr. 293), 120; (fr. 307), 119 Demosthenes :— 23 (§ 4), 161 ; [47 (§$ 69)], 123 Diodorus :— i. 6 (1), 45 (1), 16; v. (55, 63), 40; xiv. (38, 3), 160; xv. (27), 159, 160; xv. (27, 2), 162; xv. (28, 3), 161; xv. (30), 160; xv. (31, 5), 161; xv. (38, 3), 160; xv. (84), 160 Dionysius :— Dem. (50), 115 Ditt. Syll.3 :— (982. 9), 122; (1000. 11), 194 El. in Maec. i. (139), 2, 8 Ennius :— Ann. iii., 2, 3, 11 ; (503), 81 [frag. spur.], 8% Trag. (81), 88 Ephorus :— ap. Diod. xii. (40, 4), 143 ap. Plut. Pelop, (xvii), 162 Epigy. Graec, (Kaibel, 783), 194 Epigr. Plaut., ap. Gell. i. (24, 3), 2, 3 Euripides :— Alcestis, 18, 27 Bacchae, 18 Euripides Electr: Hecub 157 (77: 67, 157 158 Herac Hippe 61-' 103 (45 898 21 | 13] 23 Ion. 1 Iph. « Iph. Mede Suppl Telep. [ fr. 1 Eustathit Odyss (1) Festus :— [Lin 33! Fr. Iamb. Frontinu IV, ° G.L.K. :- Vi., | Harpocr: [Bek Heraius ap. I Herodot i, (20 Hesiod : Hippocr (295 Homer : Ilia ; 986 (8) ; . (22, 23), O; xiv. (27, 2), 0; XV. 54), 160 INDEX LOCORVM 217 Euripides (continued) : Electra (1347 sqq.), 25 Hecuba (32-36, 74), 156; (121), 202; (140), 157; (521), 156; (540), 157; (770-2), 156; (772), 157; (850-63, 889, 890), 156; (962- 67, 1020), 157; (1088, 1091), 156; (1111-13), 157; (1132-44, 1142, 1175-77), 156; (1273), 158 Hevacleidae, 19 Hippolytus, 18 ff.; (5, 28), 19; (48-50, 58-60, 61-72, 73-87, 88), 20 ; (88-89), 27; (93, 98, 103), 20; (113, 120), 21; (176), 21, 22; (451-2), 118; (612), 21 ; (786-7), 26 (887- 898), 22; (887), 21; (890), 22; (803-98), 21; (1041-1044, 1060-63, IIOI), 22; (1285- 1312), 26; (1312), 233 (1331), 24; (1391), 23; (1396, 1411, 1437), 24 Ion. 18; (666), 21 Iph. A. (394), 273; (542), 22 Iph. T. (1063), 21 Medea (263), 21; (482), 202; (529, 1082), 119 Suppl., 19; (203-4), 120 Telephos [fr. 706], 42 [ fr. inc. 924], 119 Eustathius :— Odyssey, 1343 (62), 117; 1697 (54), 43; 1941 (1), 117 Festus :-— [Lindsay] (55, 3), 124; (314), 64; (326, 28- 330, 18; 362, 10; 386, 32), 81 Fr. lamb. Adesp. (16), 119 Frontinus :— IV, vii. (19), 161 G.L.K. :— vi., p. 86 (10), 7 Harpocration :— [Bekker] (p. 77, 5), 123 Heraius of Megara :— ap. Plut. Thes, XX, (=Rzach, fr. 105), 194 Herodotus :— i, (29), 80; i. (132), 395 Vv. (42-43), 50 Hesiod :— Op. (1-10, 72-76, 116-120), 195; (169, 169a-b- 173), 195 f.; (169c-d, 174 sqq., 178-181, 182-201, 225-37, 238-47), 196; (249-55), 196 f. ; (298-307, 392-404, 410-13, 479-90), 197; (493-535), 197 £.; (618-62, 678-94), 198 ; (743), 204; (765-68), 198 Hippocrates :— (295. 25), 119 Homer :— Iliad x. (434 8qq.), 156; xiii. (794), 204; xix. (400), 11; xxiii, (186), 204; xxiv. (7), II; xxiv. (272), 204 Odyssey ix, (109), I1; xvi. (294), 125; xvii. (80), 202; xviii. (196), 10; xix. (13), 125; xix. (564), Io Horace :— Ars Poet. (168, 183, 220), 62 Carm, i. 1 (17), 4 (20), 18 (76), ii, 1 (11), iii. 6 (23, 47), 62; iii. 6 (25), 62, 65; ili. IO (5 8qq.), 9; iii. 27 (66, 69), iv. 4 (5 Sqq.), iv. 7 (11), 14 (14), 62 Horace (contiuued) : Epist, i. 1 (12), 17 (55), ii. 1 (126, 191), 2 (95, 161), 62; ii. 2 (160), 65 Sat. i. 1 (62), 207; ii. 3 (192), 62, 65 Hymn to Apollo :— (177), 46 Hymn to Hermes :— (126), 47; (313-318), 45 ff.; (319, 320) 45; (324), 46, 47; (325-26, 331, 334), 473 (336-38), 48; (340, 342, 344, 346), 49; (522), 46; (577), 47 Ilias Latina :— (551, 1017), 4 Isidore :— Diff. i. (440), 206 Juvenal :— lii, (143), 207; viii. (105), 4 f.; x. (54), 2, 4 f.; xiv. (1, 2), 4 Livy :— xxi. 19 (6), 81; xxvii. (15-16), xxix. (6-9), 129; [list of passages from fourth Decade}, 76 ff.; xxxiv. I1 (12), 73; xxxiv. (44), XXXV. 3 (3), 743 XXxv. 4 (2), 73; xxxv. 16 (6), 72; xxxv. 36 (3), 75; xxxvi. 10 (12), 13 (17), 71; xxxvi. 43 (1), 72; xxxvii. 2 (12), 69; xxxvii. 4 (1), 5 (6), 60 (3), 74; xxxviii. 4 (9), 73; xxxix, (8-19), 128; xxxix. (13, 17, 39, 40), xl. (19), 130; xl. (29), 132 Longinus :— (38. 2), 117 Lucian :— Ovat. Instr. (§§ 21-2), 117; (§ 25), 113 Lucilius :— (1120), 207 Lucretius :— i. (915 $qq-), 136; il. (27), 3; ii. (272), 2; ii. (978), 136; vi. (906-1064), 126 Lycurgus :— in Leocr. (39), 142 Macrobius :— Sat. iii, 3 (6), 10. Manetho : — ap. Plut. Is, et Os. 62 (p. 3768), 126 Manilius :— i. (90), 4; iv. (478), 3; iv. (634), 4 Martial :— i, 14 (I 8q.), ii. 30 (5), 75 iii. 3 (4), v. 69 (3), ix. 94 (2), 6; xiv. (145), 7 Spect. 5 (3 8q-), 9 (1 8q-), 21 (t Sq), 25 (2, 7), 7; 28 (10), 2, 3, 6 Maximianus :— El. i. (95), 2; 3 Mela :— ii, (14), 5 Ovid :— Amor, ili. 7 (2, 53-6), 11; (55), 2, 3, 12 Met. ii. (621), 24; vii. (569), viii. (118), xiii, (423, 549, 556), 4 Ex pont. iv. 2 (9), 7 Trist, i. 7 (36), 9; ili. 1 (78), 7; iv. 6 (42), 6 218 INDEX LOCORVM Pap. Lond. 121 (1. 100) [Wess.], 126 Pausanias :— ii. 2 (1), 40; ii, 38 (6), iii. 16 (4), 50; ili. 24 (3), 41; iii. 26 (10), 39; Vili. (49), 41 Persius :— i. (77), 88; vi. (26), 9 Petronius :— (77. 6), 207 Pherecrates :— (fr. 79], 114 Pindar :-— Pyth. xii. (4), 47 Plato :— Cratylus (413C), 139 Crito (50a sqq.), 83; (53D), 85 Epist. (3444), 109 Euthyd, (3014), 86 Euthyphro (5A), 84; (6D, £), 86; (154, B), 83 Gorgias, 15 f.; (452A, 455D), 83; (4978, 498D, E), 86; (506c, D), 83 [Hippias Maior] (passim), 84-87 ; (2858, 286D), 84; (287C, 289D), 86: (291E), 84; (292D), 86 ; (2988), 84; (299C sqq.), 82 Leg. (8938), 83 Lysis (217B sqq.), 86 Parm., (131B-132C), 87; (140A, 148), 85 Phaedo (78D), 86; (89C, 96c), 84; (100D, E), 86 Philebus (514 sqq.). 87 Protag. (3148), 117 Republic (i.), 15; (3544), 15; (423E, 439D), 99; (457D, 458D), 95; (459A-B), 96; (459D-E), 95; (459E, 460A), 95, 98; (460B), 96; (460), 95; (460C-D), 96 ; (4608), 95 ; (461A), 95, 96; (461B-c, B-E), 95; (461C, D), 96; (463C-D), 99; (4648, 466a), 95; (468c), 96; (4794 sqq.), 83; (5188), 84; (540C), 17; (5464), 98; (587D), 187; (595A, 6034), 99; (607¢), 119; (615B), 98 Symp., 16; (201D sqq,), 83, 84; (211D), 86 Theaet. (165a-168c), 83 ; (184E sqq.), 87 Timaeus (18D), 99 Plautus :— Amph. (550), 60 Bacch. iv. (9), (932), 88 Capt. (192), 59 Cas. (84), 59; (621 sqq.), 88 Cist. (43), 61 Curc, (378), 130 Men. (47, 430), 59; (779, 785), 190 Merc. (781), 60 ; (802), 61, 65 Mil, Gl. (150), 59 ; (1168), 59, 65 Poen. (404, 596), 59 Pseud. (573), 60; (702, 772), 88; (833), 89 Rud. (111), 59, 65; (417), 59; (942), 89 Trin, (864), 59, 65 Pliny :— Nat. Hist. vii, (12), 5; xxviii. (33), 124; XXXiv, (147), 125 ; Xxxiv. (151), 124, 125 Plutarch :— Amat, Narrat, (3), 161 Aristoph. et Menand, comp. (854), 115 de Aud, Poet. 11 (16C), 39 Caesar (11), 31 Cato minor (39), 33 Plutarch (continued) :— Moralia [texts collated], 171 ff. F non. poss. Suav. Viv, 21, 8 (I102A), 39 Numa (12), 132 Pelopidas (13), 159 Phocion (12), 164 Praec. vei publ. ger. (26=819E), 124 Praec, san, (19=132E), 39 de Pyth, ovac. (12=399F), 42 mwepi dito dovelas (7), 207 Quaest, Graec. (27), 39 de vect. rat. aud. (7; 458), 115 de genio Socratis (1, 5, 35), 159; (31=5975), 123 Polyaenus :— li. (7), 161 Polybius :— xii. (5), 131 [Priapea] 75 (1), 4 Propertius :-— i, 3(17), 13 i. 5 (20), 8; i. 17 (10), 10; ii. 1 (10), 6; ii, 13 (25), 2, 3, 5, 9, 10; il. 13 (189), 3; ii. 29 (39), 2, 3,9; iii. 3 (9 S4.), 7; iii. 8 (37), 93 iii. 11 (46),2; iv. I (10r), 3,5; iv. 3 (44), 5; iv. § (64), 2, 5, 6; iv. 5 (66), 6; iv. 11 (81), Io | Quint. Ser. :— (28), 4 Schol. Q in Od. xi. (48), 126 Seneca :— Epist. 115 (14), 207 Sibyliine Books [Rzach] :— ii, (9-10), 102; ii. (210-11), iii, (105 sqq.), 103; iii. (275-7), 102 ; iii. (378, 394-5), 103; iii. (780-83, 788-90, 827-29), 102; v. (281 sqq:), Vi. (9), vii. (9), viii. (209 sqq.), ix. (348), 103 Sidonius :— vii. (13, 237, 276, 277, 376), 3 Simplicius [Diels} :— in Arist. Phys. Comm, [pages and lines] (34, 18 sqq.), 1355 (35, 9 $qq-) (57, 18 sqq., 20), 133; (172, 20 sqq.), 138; (174, 19 sqq.), (176, 27), 136; (178, 25), 134; (178, 30 $qq-), 136; (460, 26 sqq.), 138 Sophocles :— Ant. (614, 853-56), 176 O.C. (674), 202; (1597), 176 O.T. (101), 176; (1262), 204 ; (1264), 80 Statius :— Silvae, v. 5 (83), 9 Strabo :— v. 2 (6=224C), 126; xiii. (615), 41 Suetonius :— Aug. (2), 65 f. ; (6), 64 ; (8), 63; (19, 38, 47, 50, 62), 64 ; (94), 63 Aug. Solil. ii. (36), 64 Cal, (12, 35), 63 Claud, (6, 26), 64 ; (41, 45), 63 Dom. (1), 63 Galb. (4), 64 de Gramm, (9), 64; (23), 63 Suetonius Ann, i' Terence :— A delph Andy. (720 IgI 192 Eun. | (106 Haut. (596 Phorm Theocritus i, (34), 204 xxii (89-< Theopomp Philip} fr. 2 Thucydide i, (103 145, 142 ; 145: 149: iii. ( (38, 144, 149 ; (66, 165 = 5978), 10; li. I >; ii. 13 (9 sq.); 1 (101), 5, 6; Iv. 5 sqq-), 5), 103 ; v. (281 qq.), ix. es] (34, qq., 20), 9 sqq.), 178, 30 38, 47, INDEX LOCORVM 219 Suetonius (continued) :— Iul, (28, 71, 94), 63 Ner, (2, 6, 35), 63 Oth. (10), 63 de Rhet. (1), 63 Tid. (53, 61), 64; (67), 63 Vesp. (2), 64; (3), 63 Vit. (6), 64 Tacitus :— Ann, iv. (74), 14 Terence :— Adelph, (204), 60; (774), 94 Andr. (483, 507), 60 ; (536), 190; (608), 191 ; (720), 192; (745), 190; (781), 194; (816), IgI ; (825), 192; (826), 191; (839), 190, 192 Eun. (24), 192; (211), 191, 192; (219), 58; (1064), 191 Haut. (143), 193; (473), 943 (589), 93; (596), 94 ; 90 ff. passim Phorm. (469), 191 Theocritus :— i, (34), 204 ; i. (95, 96), iii. (18), 202; v. (61), 204; xili. (69, 70), 202; xv. (123-8), 203; xxiii. (51), 204; xxiv. (13-16), 203 ; xxiv. (89-90), 204 Theopompus :— Philipp. LVI. (ap. Athenaeus, xii., 6o09b) ; fr. 233 [Hunt], 159 Thucydides ;— i, (103, 1), 143; i. (141), 202; ii, (13), 144, 145, 146, 150; li. (13, 6), 143; ii. (13, 6-9), 142; li, (23, 2), 142, 145, 149; ii. (31), 1453 li. (31, 1-2), 142; ii. (49, 8), 145, 149; ii. (79), 149; ii. (85), iii. (69), 165; ili. (87, 2-3), 149; iii. (100, 1), (109, 1), iv. (38, 1), 165; iv. (75), (82), 149; iv. (90, 1), 144, 149; iv. (93, 3), 149; iv. (94, 1), 144, 149; iv. (101, 3), 149; vi. (4, 1), 176; vi. (66, 3), 165; vii. (27, 4), 176; viii. (39), 165 Tibullus :— i. 6 (30), 8; i. 7 (61), 2, 3, 9 f.; ii. 4 (21), 10; iii. 6 (3), 9; iv. 6 (19), 2 Ulpian :— Dig. (7, 1, 13, 55 24: 3s 7, 14), 200 Valerius Maximus :— i, I (12), 1323 vi. 4 (3), 42 Varro :— L. L. v. (7), viii, (62), 199 ap. Aug. de Civ. Dei, vii. (34), 132 Vegetius :— Mulomedicina Chironis (§§ 220, 669, 707), 5 Virgil :— Aen, i. (501),2; iii. (464), 2, 3, 10; vi. (152), 10; x. (408), 7; xii. (648), 2, 3, 10 f. Catal. 9 (60), 2 Ecl, iv. (1-10), (20-23), 101 ; (27 sqq.), 103 ; (28-31), 101; (32), 103; (32-3), (37-40), IOI; (38-41), 103; (48-52), tor; (53-4), 103; (58-9), 101 Georg. i, (260), 206 ; ii. (139), 205; iv. (453), II [Priap.] 2 (2), 8 [Schol., (? Donatus) in Aen., x. (173)], 199 [Schol., (Probus) in Georg. i. (56)}, 199 Xenophon :— Anab. V. v. (19), 162; VII. v. (14), 118 Hell. II. iv. (28), 161, 164; III. i. (3) 161; III. i. (7), 165; III. iv. (6, 20), 160; IV. ii. (5), 161; IV. iii. (15), 160; IV. iv. (8), 162; IV. v. (11), 164; IV. viii. (11, 20), 160; V. ii. (24, 37), iii. (20), 161; V. iv. (13), 159; V. iv (54), 163; V. iv. (56), 161; VI. i. (2 sq.), 163; VI. i. (rr), 164; VII. i. (43), 162 de Vect. iii. (7), 160 Zonaras :— (1555), 114 II1l.—INDEX VERBORVM. I. GREEK. dufa, 214 arlvdnOpa, 116 "AvOnvéwv, 50 &v@pwros, 56, 215 dEoos, 56 diwy, 114 dpyedXa (Maced.), 56 avéaddcrouov, av0dbws, 115 avbwpw: dwpw, 215 BiBrlov, 117 f. Bpépos, 214 y\Grra, 120 ybvipos, 113 deapphdnv, 45 dpayudrns, Spdgwy » 55 édeXovT7pas, 215 elaBorh, 117 &xehoa, 55 éupéreca, 115 f. éuraddocoua, éumetrAnyuevnv, 80 émovewos, 215 éwtd\uov, 116 e000: lOus, 100, 48 evroyla, 56 evpnoidoyeiv, -Aovyia, -Aovyos, 212 -nvos (ethnic), 214 hplov, 50 pus : Hpws, 214 'IBucrjs: (8%, 215 ‘ ldaorns, 113 fi Ur7os, 56 Acruvium, 112 alnus, 214 anxia, 56 Anxurus, 56 -aviae, 201 Ariomanus, 214 aurum, 214 INDEX VERBORVM KaxkaBn, 215 xarédecéev, 16 Kedddaiorv, 115 Kvda0nvaceds, 215 xvda0jvasov, 112, 214 NaBetv, 17 Aahid, Adros, II5 Aewrés, NewroNoyelv, I19 payyavory, 56 pavdparyopas, 56 wéxresOa, porwr}, 213 pépipva, mepiuvar, 119 f. puvOcqra, 50 f. vauTirey, vauTirey, 194 Edpmrrukra, 114 oloméXos, 45 f. omoe.dys, 137 opuduos (6u6Fios), 215 bvopa, 215 wepmdveyv, 112 weplrara, 116 f, wralovov, 114 II. LATIN. faber : vafer, 56 facio-compounds, 200 ferrifodinae, 199 ff. jfodinae, 201 gremium, 214 janitor, 56 baia : (Span.) bahia, 214 cinnibay, 214 coda : cauda, 214 lapicidinae (-dicinae), 201 lavix, 112 la(u)tumiae, 200 concinnat, 214 dens, 56 domuitio, 56 manus, 214 mentula, 215 mox, 63 f. exercivent, 193 adhaéman, 112 Fés (Cretan), 80 *gurebh-mi-om (Indo-Eur.), 214 Hipta (Lycian), 55 Wpovwr7s, 211 wuyh, 112 wipactpov, mUpaverpov, 211 odpt, 56 Zirewvlas, 194 oréua, 56 ctppaxos, 56 cuvaydypeTa, 79 oxwiddrapos, 114 Tapxtw, 55 Tevédios (dvOpwros, Evviyopos), 42 vrorinTw, 79 girlas Bamos, 14 pita, 14 dpovtis, ppovrifey, 120 xaruwol, 114 f. XGpe, 215 XeAwvn: xéAvs, 214 xpuvcd@povos, 47 wrévn, 214 ws (? > 8s), 79 f. “wy, 215 ola, 214 olla : aula, 214 praetor, 56 prandium, 56 *stringa : stvinges, 214 sublimen, 55, 214 Tarquinius, 55 tenus, -tinus, 56 uavantia, 214 uindex, 214 uirgo, 112 uolim, 214 waranio : (Span.) guavanen, 214 III. SANSKRIT AND PRE-HISTORIC. maksubhi$, 63 nama, 215 *oketéu (Indo-Eur.), 112 piga-s, 112 Kakasbos (Lycian), 55 Aliso (O.G.), 214 IV. GERMANIC. morixles (O.H.G.), 112 kinnu-bards (Goth.), 214 *svos (Indo-Eur.), 80 tanah, 56 *uimen-dic-s (Indo-Eur.), 214 *virds (Indo-Eur.), 215 viike (f.) (M.L.G.), 112 avanen, 214 10 ur.), 214 215