STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact email@example.com. III.— A COLLATION OF THE OLD ARMENIAN VERSION OF PLATO'S LAWS, BOOK IV. 1 In the remaining books (IV-XII) of the Laws it will not be necessary to point out in detail the many points of identity between the text of the great Paris MS and that of the Armenian translator. In our examination of books I-III we have shown that the Armenian adheres to the text of the Paris Codex 1807 in almost all cases where other MSS, really apographs of it, show deviations. When the Armenian really departs from the Paris text, it is often but to agree with other genuinely independent and old sources, such as Ficino's Latin rendering or the citations of Eusebius and Stobaeus. In book IV of the Laws the following passages may be noticed. Where the Armenian seems to yield a sound reading I have prefixed an asterisk. The collation is with the text of Schanz all through. St. 7°4 B> Sch. 98. 8 7roXir, rjs irepi to vvvbr) Xe^devra fjptv. The Arm. implies ra vvv SiaXex&Vra. Cp. Theat. 158 C a re yap vvvi Sieikeypcda. But the received text is satisfactory, and all that can be here said of the Armenian reading is that it is equally satisfactory. *St. 704 B, Sch. 98. 10. Schanz reads ; W 8«/ Xi/neW ap' elo-h Kara ravra abrrjs, r) to napdjrav dXipevos ; This is the reading of the Paris MS. Eusebius, in citing the passage, has elah kot aCrijs, f/ k. t. X., which is also implied by the Arm. Ficino renders : "Portus uero suntne prope? An mare illud penitus importu- osum " ; which agrees better with the reading of Eusebius than with that of the Paris MS. *St. 704 B, Sch. 98. 12. Sch. reads with the Paris MS: EiXi- pievos pen oZv TavTjj ye o>s dwarov, and does not notice the variant olv avTtj given in Eusebius and, furthermore, implied by the Armenian. The agreement throughout this passage of the Armenian with Eusebius' citation is important, for it proves (1) that these variants are not due to citation, but were in the text itself which Eusebius had, and (2) that a genuinely old ttxt underlies the Armenian. x See A. J. P. XII 193-210. 336 AMERICAN JOURNAL OF PHILOLOGY. St. 704 C, Sch. 98. 15, 16. Perhaps the translator read eVtSeijo-et for emSerjs in both sentences. *St. 704 C, Sch. 98. 21. The Version adds ira^t, so as to read : Ti 8«/ iredlav re Kat opa)i> (tai vAi;s jrdoT/s jraij pepos endoTtov rjpiv efXij^ei// The sense would be "wood of all sorts." As introductory to the exception he means to make of ship-building wood, Plato may well have written n-do-ip here ; and the fact of n&s following after it may explain its absence from the Paris MS. St. 704 D, Sch. 98. 25 ■nam yiv olv\. The Version adds Udjepa, which seems the addition of one who missed the sense of the passage. *St. 704 "> Sch. 98. 29 ptydXov twos ?8« craTtjpos re airy. The Arm. has aunj, which Ficino also implies: "maximo sibi salutis fundatore opus esset." St. 704 D, Sch. 99. 3. The Arm. translator renders as if o-^eSoi/ came after beovTos. Perhaps, however, he merely read the comma after, instead of before, <tx(86v. St. 705 "1 Sch. 99- 4> 5 7rp6<rotKos yap BaKarra x^Pf ™ P* v na p' Uao-TTjv fipe'pav r)8i. This is the reading of the Paris MS. Stobaeus, in citing the passage, reads Bokor-rn x<*>pa, which is also involved by the Armenian. Schanz omits to notice this variant, which has at least as much authority as that of the Paris MS. *St. 705 A, Sch. 99. 7 iprnpiikdva airri)V. The Version has avT/jv, which is also read in early editions of Stobaeus, and must be accepted, if just before BaXdrrn x<*>P a be read. *St. 7°5 "1 Sch. 99- * 2 Tpa\tia 8c ovaa brjkov i>s ovk av iro\v(popos ap.a. Schanz notes that ndpqjopos is read in the Paris MS, but that the letters dp. are written over an erasure of iro\i<popos, while a manus recentior adds before irdp<j>opos and outside the line the Words iro\v<popos re tit] Kai. Eusebius has iro\i<popds t til] Kai ndfupopos, which Stobaeus also gives, only reading rw for r'. The Armenian, like Eusebius, adds koi nap<p., but omits tit] and favors •ye instead of t. Ficino also had the reading of Eusebius, and well brings out its meaning : "Cumque silvosa sit et aspera, licet omnia ferat, non tamen abunde omnia." The reading of Schanz comes to the same thing, but the weight of the evidence is in favor of adding the words koX ndpcpopos. St. 705 B> Sch. 99- * 9 <rvyx«>povpev Tore \tyeii> fjpas opBas Ka\ rd vvv. The Arm. involves vpds, which Ficino perhaps read, for he renders : "et ilia tunc, et nunc ista recte dicta concedimus." *St. 705 D, Sch. 100. 5. Schanz reads : to hi on npbs pepos dXX' ov npds irdo'av o~xth6v t ov irdw ^wtx^>pow. The Arm. places the PLATO'S LAWS, BOOK IV. 337 comma before, not after, cr^So'i/, with a gain to the sense. Ficino renders : " quoniam vero non ad universam, sed ad virtutis partem ferme equidem non valde laudabam," which perhaps points to similar punctuation in his Greek. Stephanus takes it in the same way. *St. 706 A, Sch. 100. 8-1 1. Schanz reads : tovtov yap 817 rideo-Oai tov vopov op8S>s \morl8epai. povov, of av SiKtjv to£6tov emuTore oro^afijTai tovtov oto) av obk^Ss T&v del na\S>v Tt £vv€nt)Tai povov, Ta Se aXKa i-vpivavra n-apaXftVp. In this passage the Paris MS reads toO tS>v after awex&s and fiovai (sic), povov being written in margin in late hand. The Armenian confirms Schanz in omitting tov after rayis, and also omits u.6vov, which Ficino also omits: "qui veluti Sagittarius illuc animum semper intendit, unde continuo aliquid semperque sequa- tur eorum quae bona sunt." St. 706 A, Sch. 100. 1 2 iav re &pa rt to>v aXkav. Here the Version has aXXo instead of Spa. St. 706 C, Sch. 100. 25. The Version unaccountably omits povijiav after 6ttKitS>v in 1. 25, and ptvovras just below after cmodvrio-Kw. The double omission has a purposive air. But povipav has not only the authority of the Paris Codex and of Ficino, but of Plutarch as well (vit. Philopoem., p. 363 F). It is to be remarked that Stephanus read vop.lp.av from some bad MS. St. 707 A, Sch. IOI. 19. Schanz reads : jrpor fie rovrotr al 8m to vavnKa iroKeav 8vvdpets apa crarripias Tipas oil ra koXXioto) toiv iroikepiiecbv aVoStfioao-tc. fita KvftepvrjriK'js yap /cat TrcvrrjKovrapxias (cat epeniefjs [_*ai] ■n-avrodanav (cat oi ndvv o-novbaiav dvOpaimv ytyvope vrjs, Tas ripds eKaarotr ovk av duvaiTO opBats aVofiifidVat rt?. In the above the Paris Codex has ipeTpiKtjs and Spa o-mrripia. Badham conjectured a-coTrjpias. Ast would remove fita before Kvf2epvt]TiKi)s. Schanz brackets na\ before 7ravro8airwv, and the apo- grapha omit it. Ast conjectured fita navToban&v. The Armenian does little to remove the difficulties of the passage. It gives the following sense : " Praeterea et civitatum quae per navalia et vires simul salutis, et honores non optimae parti rerum bellicarum retribuerunt. Et enim per artem gubernatoriam, L virorum principatu, et principatu variorum quae (or ?qui) est hominum et non specialiter optimorum, honores singulis non possit quispiam retribuere recte." From which it appears that the Armenian read *dpxucrjs or riyepoviierjs instead of iperiKtis, and, with the apographa, omitted koi before n-ajrofianw. It is also almost certain that the Armenian translator read koi after o-arijpias. He seems to 338 AMERICAN JOURNAL OF PHILOLOGY. have understood the whole passage as follows: "Moreover, of cities, those which depend on fleets apportion the forces of safety, as well as the honors, not to the best of the warlike elements," etc. St. 708 D, Sch. I02. 7 ov to o-<o£eo-8aL re km oxik elvai novov dvdpmnois TifiiaiTarov ijyoii^ei'ot. The Arm. renders in the sense 1-0 ovra o-a(eadai km ov to elvai iwvov k. t. X. It is more probable that this is an ingenious misreading of the text we have, than that the Armenian actually found the above reading in his MS. ovtw, of course, meant 'by means of political virtue.' *St. 709 C, Sch. 103. 4. The Version adds the words tc km vonoderelo-dat, which in the Paris MS are added in the margin by a second hand, with the remark iv aXXo> evpov. Ficino translates these words, and all the editions add them. Just below St. 709 C, Sch. 103. 8 the Armenian suggests ivbe'x^at instead of dvex^rM. Ficino also renders 'suscipit.' In the next line the Arm. omits 81a <rvvr]3eiav. St. 709 D, Sch. 103. 14 omits /xaWov. The words which follow to Se o-vpnvevo-ai km Kadanep ittttwv £ebyos Ka6 ev els rairov, to \eyopevov, <rvix<t>vo-rjo-at have given rise to some discussion. For the Paris MS reads ko.6' eva. Whether the Armenian translator read K ad' ha or Ka6' iv cannot be decided ; but he removes km after o-v/«r«0o-ai and sets it after ku6' iv. Immediately below the Version retains oVT<»r and TeXea>TaT<»>', which are read in the text of the Paris MS. *St. 708 E, Sch. IO3. 21 'Qyade, eoiKa nepl vopoderSav iiravicav Kal o-KoirS>v ap.a ipeiv n km <f>av\ov' Ficino renders: " Dum et conside- rare legum latores et laudare vellem, ad vile quid dictu videor lapsus." Perhaps we may infer that he read e'nMv&v for e'navtwv. The Version omits km o-kok&v, but retains inaviav. Perhaps km o-Konav was added by one who read inMvav but was puzzled by the construction nep\ vopodermv inawmv. St. 709 A, Sch. 103. 30. The Version gives instead of dvirpajfe the bad reading aveo-rpe^e, also given in Stobaeus. *St. 709 A, Sch. 104. 1. For \01pav the Version has o-eio-pmv, after which it retains tc, but omitting km after ipirmTovTav and reading aKMpia. Assuming that aiauplai really stood in his text, the translator read as follows : o-eio-pav t« ipnunovrmv xpo"°" € ' wl toXvv ivtavrav noXKwv ttoXKokis anaipiai. This is a better reading than the other texts yield, for in them Xot^Si' is superfluous after voW, and t( after \oip&v. After o-eio-pwv, however, re is in place, and need not be any more bracketed, as it is by Schanz. The only word which the Arm. removes is km, which may have been inserted by one PLATO'S LAWS, BOOK LV. 339 who rightly failed to see the connection between human disease (Xot/itos-) and bad harvests (axaipiat)- The latter might ensue as the result of earthquakes. St. 709 A, Sch. 103. 3. a£euv, the reading of the Paris MS, is given in the Version. Stob.: ap^eiev. Just below in 709 B the Arm. agrees with the Paris Codex and Stobaeus in reading Spas, against 6pola>s of the Apogr. Vatic. In the next line Stobaeus has ev \lyovra Myeiv, but the Version agrees with the Paris MS in giving Xe-y. eS Xe-y. *St. 709 C, Sch. 104. 14. The Version, literally rendered = nam in hiemali tempore comprendere gubernatoriam maxima efficit, necne et hoc magnum aliquid lucrum ponam artis ? We see that it at least retains after Kv^epvrfriK^v the words ij prj, given in the Paris Codex, but rejected by Schanz, in spite of similar uses, like Laws 888 B pkyifjrov 8e, 6 vvv ovbev rjyei <rv, to nept tovs 0eovs 6p0a>s ftia- vor)9evra £qv K(i\5>s rj p.rj. *St. 709 C, Sch. 104. 15, 16. The Version gives the words fj nS>s to Klinias and omits ovtws. Badham felt 1) nas to be awkward at the end of the speech of the Athenians, and conjectured <™ be was. Stobaeus confirms the Armenian in rejecting ovras. St. 709 C, Sch. 104. 17. The Version implies * K al iv or <dv, which Ast conjectured. At the end of the speech it omits 8«v in 1. 22. Just before, in line 20, the Arm. has peXKet, with Stobaeus, against /«XXoi of the Paris MS. St. 709 D, Sch. 104. 24-105. 2. Schanz in this difficult passage reads op6s>s n napbv and eVtSeov. His app. crit. summarizes the various conjectures of scholars. Adopting his reading, the sense is as follows: "Then will not one who has art in one of the directions mentioned (viz. as pilot or physician or legislator) be rightly able to pray for something given him by fortune and only requiring art in addition ? " That is, the artist's prayer will be for those conditions to be given by good fortune under which his art will be able to work and produce what is wanted. The paraphrase of the Armenian translator misses the sense : "Will not, then, one who has art in one of the directions mentioned also pray that he may be able to adjust rightly, by means of his art, that which comes by chance ? " *St. 709 D, 105. 5. For rr/v avrav eixnv the Armenian has Ttpi avrriv evxh v < which gives a better sense. Just below the Paris Codex has 81), for which Sch. substitutes hv. The Armenian implies 817. After that, in 1. 10, the Arm. has *<f>epe 8i?, Z> i/o^o&Va 340 AMERICAN JOURNAL OF PHILOLOGY. k. t. X. Here S> is absent from the Paris Codex, but is conjectured by Schanz, Ficino rendering "O legislator." *St. 709 E. In the next line the Paris text has 6 \a$a>v c£«r 2><tt (K ratv Xoiirav avros Tr)v ffdXip inavms SioiKrjtrai ; Here axrr gives offence, for, though examples are to hand of its redundant use with the infinitive after bvvanai, Uavbs, etc., none are adducible of such a use after %x a - The Armenian involves avy, which must be the true reading. *St. 709 E, Sch. 105. 11-16. The passage which follows is also remedied by the Armenian. Schanz' reading departs from the Paris Codex more widely and gives a less satisfactory sense. He reads : AO. . . • ri p€Ta tovt elirelv op8a>s cotiv ; apa tov vopoBiTOv <j>pa£a>pev tovto; KA. Ti yap; A9. fNat] To8e' rvpavvovpevrjv pot Bore rr)v ttoKiv <prj<rei. Here the Paris Codex has i) yap, not «' yap, and also val, which "delevit Stallbaum." The Arm. testifies to the following reading of the passage : A6. . . . A p^rk tovt eiVav opdSis t<TTiv to tov vop.o6(Tov / apa <j>p(i£ap.ev tovto/ rj yap; KA. Nat. AO. To&e k. t. X. "Ath. What shall we rightly set down after this as the answer of the legislator ? Shall we indicate this answer ? Klin. Yes. Ath. This he will say," etc. Thus the Armenian only involves the transposition of apa and the addition of to before tov vopoBerov. To this addition Ficino also testifies, for he renders : "Quid recte deinde dicitur? An legum latoris responsio afferri debet? Clin. Utique," etc. Ast conjectured to before tov VOpoBiTOV. *St. 710 A, Sch. IO5. 19 K.a\ vvv ttj Tvpavvovpivrj \jrvxij tovto £vveitio-8a> is the reading of the Paris MS ; but it is the n6\ie which three lines before was Tvpawovpivrj, and not the faxy of the rvpawos. " Affertur alia lectio ttj tov Tvpawov yjfvxu," notes Stephanus. The Arm. = " et nunc huic quae a tyranno regitur civitati sequatur anima talis." Therefore read \jsvxr)s toUto ' the despotically ruled city must have, along with the rest, this element of the soul (in its ruler), if it is to profit by the other qualities (of youth, memory, etc.) which he has already got." That this is the true sense is clear from the words just below in 7IO B : Tairrfv toivvv r)plv k. t. X. Even if rvpav- vovpJvn be middle, it is still the n6\is, and not the tyvxi), of the ruler which may be said to govern itself despotically or to have despotic institutions. *St. 710 C, Sch. 106. 10. The Armenian includes ivrvx^is in the speech of Klinias, and instead of npoo-des, pr) kot SXko implies irpocr&es ptj tl S\\o. Therefore the Armenian had the same text as Ficino, PLATO'S LAWS, BOOK LV. 34I who renders: "Cli. . . . fortis, magnificus atque felix? Athen. Adde nihil aliud, nisi ut," etc. Stephanus would give emvxns to Clinias. St. 710 E, Sch. 106. 27. The Arm. implies Xeyo/iev 8e ravra where the Greek MSS have 817. *St. 711 A, Sch. I06. 32 TavTT) tai Tore ra^fos (cat pqarwur) k. t. X. Schanz writes: "in verbis kcu tote haereo." The Armenian omits roVe. So does Ficino: "ibi celer et facilis fieri mutatio solet." *St. 711 A, Sch. 107. 2 olx awag aXX' otfuu woXXd/ar. The Arm. involves ovx air. aWa Kai noK. St. 711 B, Sch. 107. II nopeveatiai 8e avrov det Trparov Tairr). The Arm. reads S17 for §«. The same reading is in one of Bekker's codd. a. St. 711 C, Sch. 107. 17. The Arm. agrees with the Paris MS in reading ko.\ n-m? olopcOa without ovk. St. 7II C, Sch. I07- 32 V TrjV N«CTTopof iiiv ■noji Tts iiraveveyKy (f>v<riv. The verb iirava<pipa> is used in three other passages in the Laws, viz. I 631 A, III 680 D, V 742 C, and always in the sense of 'to refer or trace a thing back to its source or centre.' In the Politeia and Cratylus, 434 E and 425 D, it has the same sense ; also in Lysis 219 C. Thus the sense of 'to bring up again or restore what is past,' which the context requires, is a sense in which imtva<f>epa> is never used in Plato. On the other hand, the sense of 'refer back,' in which he always uses it, is here impossible. The Armenian gives this sense: "vel earn quae Nestoris si quando quis renovaverit naturam." Ficino renders: "aut si quando- naturam Nestoris quis reduxerit." It is clear that the Arm. translator read (TravaveoxjrjTat Or avaveaxrrjTai. Cp. Pol. II 358 B ciravaveaMTo/iai tov 0pa<rvp.dxov \6yov. Ast in his lexicon renders etrava<pe'pa> by refero, and gives no reference to this passage in the Laws. The active inavaveixirj would involve less change in the text, but it has a post-classical air. The uncials COC might easily have been confused by a copyist with T K. Then iitavaviyKiyrat would have given way to the more usual form iiravaveyKji. The change would also be explained if we supposed that the termination tjtcu was written as a compendium in some MS of which the Paris MS is a more or less direct copy, and that this compendium was wrongly copied. Badham suggested such a theory in explanation of most of the bad readings of the Paris MS. *St. 711 E, Sch. 107. 34. In the next line the Arm. omits dvdparrrav. As it is not necessary to the sense and is also omitted 342 AMERICAN JOURNAL OF PHILOLOGY. in Ficino's version : "quem . . . multo magis omnibus excelluisse ferunt," we can infer that it was not in the Greek texts used by Ficino and by the Armenian translator. *St. 711 E, Sell. 108. 4. For lovrav \6yav the Arm. has peovraiv \6yav, which Ficino also read, for he renders: "et illi quoque bead sunt, qui fluentia ex ore modesto audiunt verba." St. 712 B, Sch. 108. 14, 15. The Armenian translator renders: "conemur accommodantes te civitate, seniores tanquam pueri oratione confingere leges." He misunderstood 0-01. St. 71 2 D, Sch. 109. 2, 3 Kai Tij e'viore p.01 (paiverat naoSiv to>v irokea>v 8t)iioKpaTovfi.evi) pdXio-T eomevai is the reading of the Paris MS. Stephanus, following Ficino's version, which has "nonnunquam tamen," conjectured Kai rot, which Sch. adopts. The Arm. retains rif, but for &rjfioKpaTov(uvri has rvpawovfievr) or Tvpawlbi. The substi- tution is probably right, since it agrees better with what precedes and renders «? more tolerable. The sense may be ' and in a way (or under some aspects ni) it sometimes appears to me of all states most to resemble a tyranny.' Just below, in 1. 7, the Armenian retains the reading of the Paris Codex : av iparrjdzis. *St. 712 E, Sch. 109. 10. Sch. reads ndya> <paivop.ai where the Paris MS has KaTa<f>alvop.ai. Ficino has "mihi quoque, Megille, perinde ac tibi contigisse videtur." The Arm. translator probably read Kaya, for he renders: "Eadem ista affectione, Megille, videor ego affici." St. 712 E, Sch. 109. 16 to tov fteo-rroTov de e/caari; irpoo-ayopeifTcit Kpdros. The Armenian translator reads e'jcaonjr, a vicious reading, if it was really in his text. St. 712 E. The speech ralrav 0-01 wddos is in the Arm. given to the Athenian. Then the speech '6ma>s yap k. r. X. to Megillus and Ti'r 8' o 6e6s to the Athenian. Then &p' ovv, etc., to Megillus. Then the words ovkovv — bpav to the Athenian, the words ndvv p.iv ovv to Megillus. The Athenian resumes tS>v ydp B!j down to oIkutcu, and Megillus has the words o-<j>68p' &v w, etc. Comp. the editors. St. 713 A, Sch. 109. 17, 18. The Armenian involves xpn" 8' €t7T€p TOIOVTOV Ttvbs rrjp ttoXiv cSft €7rovop,d^€o-0ai, TO TOV TuiV d\y)8a>s vovv ixdvTav 8eo-7r6^opTos k. t. X. Just below the Arm. has ye n, with the Paris Codex, which Sch. corrects to y in. *St. 713 B, Sch. 109. 25 en npoTepa tovtwv. The Arm. omits tovtvv, which is redundant, and not translated by Ficino. St. 713 C, Sch. IIO. 4 Tt)s toiv TOTe fiaxaplas £<»>ijr. The Arm. omits tS>v. Perhaps Ficino did so also, for he renders: "beatae PLATO'S LAWS, BOOK IV. 343 illius vitae." Just below, in 1. 9, the Arm., along with Ficino and the Paris MS, omits tote, which Schanz introduces from Julian's citation. *St. 713 D, Sch. IIO. II ■y/xow deioripov re km afieivovos. For re the Armenian has twos. Ficino and Julian omit re. Just below, 1. 14, the Arm. retains avTolo-i npat, which Ficino omits. St. 713, Sch. no. 15. The Arm. retains dpa after 6*6s, which Sch. omits, following Julian's citation. The Paris Codex has Spa km. In the next line, instead of ro yeW, the Armenian has *ro0 yevovs, which Ficino also no doubt read: "similiter deus, homines amans, genus daemonum generi nostro praestantius nobis prae- fecit." Schanz brackets to, for which Hermann conjectured roYe. *St. 713 E, Sch. no. 18. The Arm. has dpr\vi\v re km al&S, km iXevSeplav km drpdovlav. Schanz adopts the reading of Julian's citation : tip. re k. al8S> km 87 ap8. The Paris Codex has dp. re k. alBS) km fivoplav (yp. i\tvdzpLav) km d<p9oviav. Ficino agrees with the Armenian: "pacemque- et pudicitiam, libertatem et iustitiae copiam proebens." *St. 713 E, Sch. no. 22. The Arm. omits ouSe w6va>v, which Ficino retains, but, with Ficino, renders dvd<pvgis in sense of ' quies.' Both these authorities therefore read dvdtyv^is, which is also Julian's reading. This is the only one of Julian's divergences from the Paris MS which is reflected in the Armenian. *In the same line the Armenian omits oUtm, which is not wanted, and perhaps read Set. In the next line, for t6v eV« rod Kpovov Xtyopevov fiiov the Arm. has *t. <■. r. Kp. yevopevov /3., which Ficino also read : " imitari nos iubet vitam quae sub saturno fuit." *St. 714 B, Sch. in. 10 /117 87 (paiXov irlpi. The Arm. has hi for 8rj. So also Ficino : "neque de re vili," etc. *St. 714 C, Sch. in. 13 ff. oSn yap vpos k. t. X. The Arm. gives the Sense as follows : oure yap . . . row vopovs, dX\' o n av Kadeo-TrjKvia rj iroXtTela £vu(pepoi> 17, O7ro)s ap£ct T€ act Ka\ prj KaTa\v6f]<T€Tai, tov (pvoti Spov, etc. Perhaps the true reading may be KaSeo-TtjKvia fj Kokirelq gvpcpepov, 6V<»r, etc. The sense of the whole clause would be : "for they say the laws ought to regard not war nor excellence all round, but whatsoever is expedient to the established constitution; namely, that it may always prevail and never be dissolved, thus (they say) is the natural definition of the just best stated." The Armenian has rendered the dative (ca^eo-njKui'a 7roXir«'a as if it were nominative and n as if it were §. In a MS in which no iota subscript was given, such a blunder was natural enough. The 344 AMERICAN JOURNAL OF PHILOLOGY. words TavTfi delv, which the Paris Codex adds before to gvpfa'pov, are omitted in the Armenian. Critics have replaced them by all sorts of conjectures, and all are agreed that they are wrong. Ficino's rendering supports in a marvellous way this explanation : "nee enim ad bellum, nee ad omnem virtutem referri leges debere aiunt : sed ad illud potius, quod constitutae reipublicae conferat, quaecunque ilia sit, ut dominetur semper nee pereat : iusti que definitionem sic natura optime se habere contendunt." This would be in Greek : a'XX' o ti av Ka6eo-TT]Kvlq fjns av rj noKiTila 5 |u/*(£cpoii, onws k. t. X. Supposing this to be the true text, we may suppose that a copyist who did not see that KadeaT^Kvla and wdkmia were datives, left out o « ta> and wrote 17ns av instead. Then the introduction of ravry 8elv (or IBttv) would be necessary in order to make sense. St. 714 C, Sch. III. 20 A«y m o-arfricTTepov. Arm. has Xe'ye ti aacp. St. 714 D, Sch. in. 24. The Arm. adds *rj before o%iov, and just below .paraphrases in the following sense : $) vpos ™ <rvp<pepov iavTta (Or eavrov) nal Tr/9 dp)(tjs tov pev€iv. St. 714 E, Sch. 112. i. The Arm. has adiKripdrav, which is read in the Paris Codex. St. 715 A, Sch. 112. 12. The Arm. seems to have omitted fjSt]. Also Ficino : " millies namque in civitatibus nonnullis id contigit." St. 7^5 ■"■> Sch. 112. 15 T« T€ 7rpdyp.aTa Kara Trjv ttoXiv 0VT<ay *<T(p€Te- purav <r<p68pa. There follows no nal to answer re, and so Stephanus read ye conjecturally. The Arm. involves to. re irpdypara km to koto rt)v iroK. ov. 4(T<p. and omits <rcp6dpa. Ficino renders: "adeo ad se solos rem contraxerunt." St. 715 A, Sch. 112. 18. Is it a mere coincidence by which the Arm. and Ficino both render /xiji-e airols pr/re tuyovois as if it were prpe avrwv rots e Kyovois ? Ficino, e. g., has: "ut nullum victis eorutnque posteris magistratum concesserint." St. 715 B, Sch. 112. 23. The Arm. retains <TTa<Tia>Tas . . . noXiras, read in the Paris MS, but for tovtovs, which follows, implies 6 or oiis, which gives no sense. In the next line it adds *ol6pe9a, or a word of the same meaning, before pdrr/v. So also Ficino : "et iura sua frustra sic ab eis vocari censemus." St. 715 C, Sch. 112. 30 vopxav imrjpto-lav. For voparv, which is a conjecture, the Arm. retains 0eS>v or 6eo0 of the Paris MS ; but in the next line the Version implies 6\ ehm -n\v ptyiemp' rrp> peyiotrpi r» npi>ra. Ficino renders: "deorum quoque cultum et ministerium dabimus ; maximum quidem primo." PLATO'S LAWS, BOOK IV. 345 *St. 715 D, Sch. 113. 9 oo-a 6(o\. The Arm. has 8<ra ol fool, which Stobaeus also read. *St. 715 E, Sch. 113. 20. The Arm. has &v8pes, which Sch. adopts. Ficino : " O viri." Codex Paris, has SvSpes. *St. 716 A, Sch. 113. 22. The Arm. has evBeia nepalvei iravra Kara (pio-tv Ttepupeponevos. Schelling, Diar. Antiq. 1843, Nr. 88, p. 704, conjectured irepicpepopevos. The Paris MS has irepmopevonevos and Omits iravra. *St. 716 A, Sch. 113. 25 raiteivos (col KeKoo-pripcvos. Schanz brackets the last two words, which are 'extra versum' in the Paris MS. Arm. has Taiteivorrin KeKoo-pripevos, with which cp. Xenoph. Air. 2, I, 22 KeKoo-ptjpevtjv to pev oaipa Ka$apoTr)n, rd fie oppara aifiol. In the same line the Arm. implies Sony fie, or less probably « fie tk, where the Paris MS has 6 fie nr. Just below the Arm. implies dvoia (pXeyerai, with the Paris MS. St. 716 B, Sch. 114. 2. For iravra Spa the Arm. has, with Theodoret, citing this passage, Spa irdvra. St. 716 B, Sch. 114. 7. The Arm. has as follows : AijW fiij toOto' ■ye, as tS>v ^vvaKo\ov6t)cr6iiTiov tw 6ca, els to pihXov fiei Siavorjdfjvai iravra avSpa. "This much, then, is clear, that we must for the future conceive of every man among those who shall follow along with God." But this is not satisfactory, and 5n, which Schanz conjec- tures before i>s, is still required. Of course, els r6 /xe'XXoi/ may easily have been corrupted into e'o-opevov, but it seems otiose. The phrase used in the Version, and which I render by els to peXlov, admits of being put into Greek quite literally by the phrase els t6v eneiTa xpovov. In the long speech of the Athenian which follows, the Armenian reflects the Paris MS in the following readings, which Schanz or others find unsatisfactory, viz. St. 716 D, Sch. 114. 18 Kal u&ikos. Just below, 1. 22, Sei where Sch. writes del. In St. 717 A, Sch. 114. 30 the Arm. has e'yKaiporaros. Then Sch. 115, 1. I erpeais and \eyopeva, 1. 4 dpioreia or apio~reia. Then in St. 718 A, Sch. 116, 1. 4 Arm. has Sairdvrjs re, in 1. 6 of same page it has av, which Stobaeus omits. Just below, in 1. 10, irpos 8e5>v. In 1. 12, lastly, it has the same lacuna after 8U£o8os as the Paris MS. In the following passages, however, in this speech it varies from the Paris Codex : St. 716 D, Sch. 114. 22. The Arm. has evxals Kal ^vpnao-iv dvadr)- pao-iv Kal depmrela 0eia o Kal KaWiarov. Here 8eia may be right. St. 716 E, Sch. 114. 27 omits fiSpa after piapov. Just below, in 1. 29, it renders toi? dvoo-iois in the sense of tois dbUois, and in the 346 AMERICAN JOURNAL OF PHILOLOGY. next line oVi'oir in the sense of xadapois. *In the same line it has for airaaw the nom. sing, anas, which is certainly right, the sense being that, for the unholy, all their arduous and ample labor is in vain, but for the pure in heart every service (anas), however humble, is in season, even if it be not 6 ttoXvs. Ficino : "quod opportune faciunt omnes sancti " ; so he read cmaviv. *St. 717 A, Sch. 114. 31 fitXt) 8c aZ. Here the Paris MS has |3. Sf airoi, which is impossible. The Arm. has avra or ravra. The Basle edition already had alra. Ficino: "sagittae vero ad illud quaenam ?" St. 717 B, Sch. II5. 6, 7 TO(? 8e Tovrav avadev down tO i/w8ij]. The Arm. omits, but since these words are glanced at in Plutarch, de Iside, p. 361 A, they can hardly be a gloss. Ficino renders them. *St. 717 B, Sch. 115. 13 vonifav it. The Arm. adds 8«. This must have stood in the original Greek, for Ficino also adds it : "putare enim quisque debet omnia quae possidet eorum esse." *St. 717 C, Sch. 115. 17. The Armenian reads *a< after instead of before xmepTrovovvrmv, and in 1. 19 817 waKaiois for be wa\aiols. These very insignificant changes make the Greek text to run more smoothly. Ficino's text also must have transposed the <a\ : " haec omnia videlicet mutuo data persolvens, et pro curis dolo- nbusque priscis parentum, in senecta recentes reddens, quando maxime indigent." St. 717 E, Sch. 115. 29 t&v el6i<rixtvu>v oyKov is Schanz' reading. The Paris MS has 6yK<ov, but t5>v eldio-pivav. Stobaeus has rbv eWte-fievov ojkov. The Arm. has to» eWto-ncvov vofwv. Ficino : " con- suetam magnitudinem." Here vd/wv might be right, but SyKov is more graphic. *St. 717 E, Sch. 115. 30. The Paris Codex and Stobaeus share the vicious reading tow aWav yevvrjras. Schanz conjectures and reads rots avrmv yewr)Tals, which is actually involved by the Arme- nian and was also read by Ficino: "quae maiores genitoribus suis struebant." St. 717 E, Sch. 116. 1. For emueXcias the Armenian Version has fivi)ixas, which is inept. Just below it omits irap^x°^'">'>, and translates the entire clause as if it were t<» (or to) 8e /«? irapakeLnuv fivrjfirjv (P&€\f)(rj avrStv aWa rovro) fxa\ior del 7rp€<r(3*v€iv, SaTrdvrjs re k. r. X. St. 718 B, Sch. 116. 13. For plq ko\ the Arm. has fimla. In reading an old text a translator might confuse <ai with tat. PLATO'S LAWS, BOOK IV. 347 *St. 718 B, Sch. Il6. 18. The Arm. has irapd.8ei.ypa irpoeveynovra, which Ficino also read: "horum exemplum." The Paris MS has 8e7yp.a. St. 718 D, Sch. 116. 30. After Xf^dhra the Arm. introduces this gloss : irepl dvaiav Kal yoveav. In the next line it omits els to, but in the rest of this hard speech it faithfully reflects the readings of the Paris Codex. St. 719 A, Sch. 117. 13. The Armenian read tpepeiv, with the Paris Codex. Cp. Hesiod, op. 213, for a similar use. Why does Schanz in his text adopt the insipid reading neXei of the editions of Hesiod ? Plato certainly read cptpew. *St. 719 A, Sch. 117. 16 ei to fieo-ov avTo Sewai is read in the Paris MS. Badham conjectured a5 for ai™, which Schanz adopts. The Arm. omits al™. In the preceding line, for 5 ye poi it implies & ye poi. Ficino: "verum quo sermonis spectabat progressio, volo in medio vobis proponere." The Armenian also assigns now pep ovv to Clinias. St. 719 B, Sch. 117. 20. The Arm. omits ov before SrjXop, and after it reads S n. St. 719 D, Sch. 118. 8. The Arm. omits 8el after del. *St. 719 D, Sch. 118. 13. The Arm. has Bdirreiv nirriip 'and if it were commanded to bury her in poetry.' This is better than ainjv, read in the Paris MS and implied by Ficino's rendering. *St. 720 A, Sch. 118. 30. After irpaoraTov the Arm. omits avrov, which "delevit Badham." Just before, after dpapippjio-Kwfieda, in 1. 28, the Armenian omits 8e. *St. 720 C, Sch. 119. 16. The Paris MS has Udo-Tov irept voo-t)- p-aTos enao-Tov twv oliceT&v. Schanz notes thus: "unum alterumne eicao-Tov delendum esse videtur." The Arm. has UaaTw nepl voo-ij- fiaros eKaerrov t. oik., which may be right. *St. 720 E, Sch. 120. 2, 3. The Paris MS, which Schanz follows, has tip apa irpoyrov vofiov Be'iT av 6 vopodeTTjs J ap' ov Kara (piaiv Tr\v irepl yeve'o-ems apxi)v npaiTrjv iroheiov nepi KaTaKoo-p.rj(rei rats rdgeaiv ; Badham felt this text to be wrong, and conjectured rw nepl yeveo-iv, o>s dpxrjv irpwTTjii iroXewv, npiv KaTaK.oerp.rjo-ai reus Ta^eai. For irepl Stephanus conjectured napd. The Arm. involves ap' ov top koto Kpiaiv, rfjv irepl yev. dp. irok. irepi irpd>Tt)v kotok. rats irpd£eaip ; which, however, is not quite satisfactory. Ficino adheres to the text of the Paris MS. *St. 721 A, Sch. 120. 7 koi fj Koipcopla is read in the Arm., as also in Codex A Stobaei. Just below the Arm. involves yaptml 8c 8fj 348 AMERICAN JOURNAL OF PHILOLOGY. vofioi. The Paris Codex has 81) and omits 8e. The codices of Stobaeus have some 81? and some 8e, some neither. *St. 721 A, Sch. 120. 12 ?x<h 8' av tt&s; Uas &B(. The Arm. has cxoi 8' av'lo-as &8e. Badham conjectured nas la-as. Ficino : "quae sic se forsan habebit." In next line, for yapziv 8* the Arm. perhaps implies yapeiv ye or y. 817, but hardly y. Se. *St. 721 B, Sch. 120. 15, 16. The Arm. has xphl J - aa ' 1 p* v ™<tok Koi toIs, T$8e 8( Kal ttj dripla. Heindorf conjectured TrjSe 8e Kal ttj dr., of which reading there are traces in the Paris MS and in the MSS of Stobaeus. Ficino : "dedecore autem tali quodam vel tali." St. 721 B, Sch. I20. 19 8tavot]6evTa as etrnv r; k. r. X. The Arm.= cogitantem quia sit veritabiliter humanum genus et natura quadam particeps est immortalitatis, which is a bungle. *St. 721 D, Sch. I20. 31 to'oo) Kal roam. Arm. = too-co koi xf St. 721 D, Sch. 121. 5 irorepov aiirovs SiTrXoOr]. The Arm. adds avTwv. avrav would not be amiss. Just below it retains t<u pfj<€i to OfUKpOTCtTOV. *St. 722 A, Sch. 121. 14. The Arm. has alpolprjv, with the Apographum Marcianum. The Paris Codex has epol^v. Lower down in the same speech, 1. 29, the Arm. reads pdxv, with the Paris MS. *St. 722 C, Sch. 121. 33 «| aiirav hv vvv SieiKeypeda. The Arm.= ex iis etiam nunc a nobis selectis, which corresponds to «£ airmv &v vvvty el\6)ieda (or ^prjptda). Schanz notes : "vvvSt/ scribendum videtur." Just below, Sch. 122. 1, for nva yeyovos the Arm. has *i-t yeyovos, which must be right. Ficino read nva and SieCkiypeda : "quod per ea, de quibus disseruimus, nobis Deo quodam adspi- rante factum est." St. 723 D, Sch. 123. II ov8e nepl acrparos. The Arm. has oi>Se yap aaparos, with the Paris Codex. Ficino seems to have had oiSc yap nepl a<rp., for he renders: "non enim in cantu neque in omni sermone." *St. 723 E, Sch. 123. 25 /iera 8e toSto. The Arm. involves p. S<f rodro 8fj. The second hand adds tjbr) in margin of the Paris Codex. *St. 724 A, Sch. 123. 34. The Arm. omits airav before -^vxas. So also Ficino : " Post ilia vero quae ad animos, quae ad corpora," etc. Just below, in 1. 35, for npoa-ijKov r the Arm. has npoo-rjicov 8'. The above exhausts, for the IVth book of the Laws, the readings in respect of which the Greek text of the Armenian translator differed from that of the celebrated Paris Codex. Hitherto we PLATO'S LAWS, BOOK IV. 349 have had but two witnesses to the text of the Laws, to wit, (1) the Paris Codex, of which all our other MSS are copies, more or less remote, and (2) the Latin Version of Ficino, representing an independent Greek text which must have perished towards the beginning of the XVIth century ; and we may rank the ancient authors, chiefly Eusebius and Stobaeus, who have handed us down citations of the Laws, as a third source of testimony. But the variants either implied by Ficino or given in the Florilegists were always liable to doubt, so long as they stood alone. There was no certainty, for example, that an omission or addition or other change in Ficino's Version was more than a device of translation. In Eusebius or Stobaeus similarly they might be due to careless citation. It may be claimed for a version like the Armenian that in scores of passages it removes this uncertainty, which especially attaches to omissions and additions. For wherever it supports a various reading found either in Ficino or the excerptors, we are at once certain that that variant was actually in the Greek text used. In this IVth book there are many passages in which Ficino and the excerptors win such support. In every such case a variant, if it really mends the text, is doubly confirmed; while, in case it does not mend the text, the mere participation in error proves that we have in the Armenian a text which has pro tanto had the same history and comes from the same hands as that whose errors it shares. When our collation of the whole of the Armenian Version is concluded, we shall try to exhibit in a summary the new evidence which the Armenian and Latin Versions, in conjunction with one another, afford in respect of the families of the texts of the Laws used by the various excerptors of antiquity, by Stobaeus in particular. Fred. C. Conybeare.