TABULATION. 2 I! THE POETRY OF THE GREEK BOOK OF PROVERBS.

Transcription

1 46 THE JOURNAL OF THEOLOGICAL STUDIES TABULATION. V VI XXII XXV l & I! C ><S><T. No Com ~ 0 C><S><Com. s& l! I! I& I! 0 C><S= T. No Com 2 0 C><S- Com C= T><S. NoCom C= Com ><5. NoCom Com shows error 2 I! of C. w. H. WORRELL. THE POETRY OF THE GREEK BOOK OF PROVERBS. THE lae Dr Blass sigmaized as a wase of ime he search for verses and fragmens of verse in he New Tesamen. 1 He admied, however, ha in his respec one book sood apar from he res-he Episle o he Hebrews. There no only did he noe (in xii 13 ff) a faulless hexameer, followed almos immediaely by wo faulless rimeers, bu (a far more essenial maer) he discovered running hrough he Episle ' a carefully execued muual assimilaion of he beginnings and endings of senences and clauses'. The general endency is for he utlxo o run in couples, he concluding (and someimes he opening) syllables of he second line having he same scansion as he corresponding syllables in he previous line. In he aenion paid o l).e balance of final syllables he sysem is a sor of approach o rhyme. In he Greek Old Tesamen wo books mus be exemped from he general sricure above menioned. I have previously poined ou in his JouRNAL 2 ha he couple sysem is illusraed, perhaps even more srikingly han in Hebrews, in he Book of Wisdom. In he Greek version of he book of Proverbs ha rhyhmical device is no, so far as my observaions have gone, represened o any appreciable exen. On he oher hand, he number of complee or fragmenary hexameers and iambic lines in ha book is far oo numerous o be he resul of acciden. My aenion was firs 1 Gramm. of N. T. Greek p vi 232.

2 NOTES AND STUDIES 47 drawn o his many years ago by my friend he lae Dr Redpah. He did no, however, I believe, place his observaions on record, nor, o my knowledge, has he subjec been handled by ohers. I may, herefore, be worh while seing ou he evidence in exenso. The maer clearly has a raher imporan bearing on he reconsrucion of he ex of he ' LXX', which in his book has suffered a good deal of conaminaion.' One insance will suffice o illusrae he sor of criical problem which arises. The 'faulless hexameer' which Blass found in Hebrews (xii 13) is a quoaion from Proverbs (iv 26) :- Kal. Tpoxa<; op0ar;?to~o"at 11 TO<;?TOO"V vp.wv. Bu in he MSS of Proverbs s he line is differen, and, unless i can be classed as anapaesic, unmerical :- opoo.s Tpoxas?To{n uo<;?toalv. Which of he wo exs is he older? I is no surprising ha his paricular book of he Greek Bible should be he one conspicuous insance of an aemp a merical arrangemen. Verse was he naural vehicle for proverbs. The merical form helped o impress hese household sayings on he memory. The grammarian Hephaesion, who wroe a manual on Greek meres, ells us ha he meres employed for proverbs were he hexameer, he iambic, and he versus paroemiacus. A glance a he oldes collecion which has come down o us (ha of Zenobius, who lived under Hadrian 6 ) will verify he saemen. These meres are all largely represened in he Greek book of Proverbs. Tha Hellenisic Jews produced paraphrases of Scripure in Greek verse is well known. We have an insance of a hexameer collecion of yvwp.a exending o upwards of 200 lines, largely based on 0. T. language, wrien in he Polemaic age by a Jew who sough o pass off his work as ha of a much older composer of apophhegms, Phocylides of Mileus. We have a specimen shewing how a Jewish wrier of he same period handled a Biblical subjec in iambics in he considerable fragmens preserved in Clemen of Alexandria and Eusebius of he ragedy eniled 'he Exodus' (~ 'Eaywy~), wrien by he Alexandrian Ezekiel. 7 I See Lagarde's Anmerkungen zur gr. Ob. der Proverbim (Leipzig 1863) passim. I am indebed o Professor Burki for he loan of his valuable book, now unobainable. The Briish Museum does no possess a copy. 2 ro<ein ~*P 17, W.-H."' 8 Holmes and Parsons cie Thd. iii 627 for he N.T. form, bu Theodore is merely quoing from Hebrews. Teubner ex p. 26. r> Ed. Leusch and Schneidewin in Paroemiographi Graeci, 2 vols., G<lingen e I have used he ex of ps.-phocylides conained in Bergk's Poeae Lyrici Gram (1866) p. ii p. 4fi6. Cf. Schilrer H.]. P. div. ii vol. iii Schiirer ib. 2 25, Swee bzrod. o 0. T. in Greek 369 ff.

3 48 THE JOURNAL OF THEOLOGICAL STUDIES The majoriy of he fragmens of verse colleced below require no aleraion or ransposiion of words o give hem a poeical form. Bu i may be regarded as cerain ha in he course of ransmission of he ex scribes have oblieraed he rhyhm in oher passages which now read like prose. The commones error exhibied by he copyis of Greek poery consiss in ' arranging [he words] according o he order which hey would have in prose ; according o heir grammaical consrucion',' in bringing ogeher aricle and subsanive, adjecive and subsanive, giving paricles an early posiion in he senence, and so on. If his happened in ranscribing wha was known o be poery, sill more liable o similar corrupion would be he ex of a work mainly in prose wih an underlying poeical elemen which has hihero generally eluded deecion. Transposiion of he MS order is herefore, under cerain condiions, quie a legiimae procedure in he reconsrucion of he original ex. Lagarde, who as a rule does no seem o have noiced he versificaion in Proverbs, remarks on ix I I?roAvv {~rr Lc; XPovov: 'den sicher beabsichigen jambischen onfall verniche 103 2?roAAovc; XP6vovc; {~rr c;.' Apar from ransposiion, I have no ofen indulged in he precarious ask of conjecural emendaion, which, I believe, is rarely called for. In wha follows ( ) denoe a conjecural addiion or correcion, ( ) unmerical words sanding in a merical conex, ha words have been ransposed. Figures below he words indicae he order in which hey sand in he MSS. HEXAMETERS Complee or nearly complee hexamelers. The complee hexameers are few and rugged, hough no rougher han many in pseudo-phocylides or he old Greek proverb-wriers. A high sandard of versificaion would indeed be surprising. Greaer regard seems o be paid o accen han o quaniy. In paricular, w and TJ may be reaed as shor vowels. These licences are no peculiar o he ranslaor, bu are shared by him wih oher wriers of his species of verse. (I) ii I 5 Jiv ai Tp{{3o O"KoAaL Ka Kap:~n)Aa ai TpoxaL a1rrwv. 1 I quoe from he lae Dr Waier Headlam's learned and copiously illusraed aricle' On Transposiion of Words in MSS' in he Class. Review xvi ~43 ff. 'The order of he words,' he wries, 'is he very hing which [he Scholiass] mos ofen hink requires elucidaion ; here is no form of noe in scholia so common as TO ~ ij~ olitcus, "he consecuion is as follows''.' ~ On xvi 28 he wries: 'dass 1..ap.TrTfipa 861..ov 11"VpUEVEI KaKol~ nach einem ragiker klinge, filhle Jager.' I have no had access o Jager's eigheenh~cenury work.

4 NOTES AND STUDIES 49 (2) iii 13 (p.akapws d.vbpw1ro<;) ~ }..,1.,. ' ll. ' 6." 1,1., ( O<; f.vp V UO''LaV Ka UV'YfTOS O<; f.of. ''POV'Y/ULV. (3) XXV 25 ijju7rf.p.08wp lfro')(p(jv lfrox:u 8ufr6Jcrn 7rpofT'Y!vls, ", \., 2, c. ' " '() (~ ovrw<; ayyf.aa \ ayav.1 K YYJ'> p.akpo f.v 'Y/Kf.1 M!l'l1 W!l) ~31 C',i' 0'~ Cold waer o a hirsy soul. So runs he original of he firs line of his couple, and he poin o noe is ha i has no equivalen for ijju7rf.p and 7rpOfT'Y!vls, which seem clearly o be insered merely o round off he verse. For he shor 'Y/ in 8u{I6Juy (which is acually wrien 8lfwu in cod. 297) we may compare Ps.-Phoc P.'Y/8' lp.1rlvp d.v8xa vf.ikos, and, for a = d in he same posiion, he proverb ( = our ' A Rome do as Rome does ') in Zenob. i 24 d.aaotf. ~.u.aoiov Tf.AEBnv Kal xijp l7rf.uba. (4) xxvii I olj y"o.p ywijukf.stl Tl~na.q 1rwvua would be complee were -r replaced by 6 T. The line has previously occurred (wih ol8as for yvijukf.'>) in iii 28, and finds a close parallel in Ps.-Phoc. n6 omf.ls yvwukf. Tl JLf.Tavpwv ~ T p.ef i:jpav. Then we find lines jus falling shor of compleeness by a syllable or wo a he beginning. Here is pracically a complee hexameer preceded by he end of an iambic line :- (S} ii r6 f vu, p.~ ue KaTaA.aPn KaKTJ PovA.~,.q 'roa7rovua 3 88auKaA.av veon,tos. ( 6) Vi 2 I ( /J.fl )alfa ~ awov<; ml crii lfvx'fi 8u). 1rav-r6s.. The 8 is an inserion, meri grai'a, as in he parallel passage /J.flalfa ~ avta<; 1r u<e Tpax~AIJ! This las appears o be an example of wha I call ' Heads and ails' (see below). (7) XV 6 (8) xix 4 (7roAA~,) oi 8' due{3 'is oa6p{o K y~s d7roaovvta., ', ~...,~' f.vvol ayav'l TOL<; f.woulv avt'y/v, ' (' f.yyf. -, 1 Transposiion, wih in some cases a sligh aleraion, of words produces he following furher insances. (9) vi II a.q 8' Zv8na 8pop. v<; KaKo<; ijju7r p cl1ravrop.oa.~un ( E)l3Ev MSS. 1 Or.;rrE.\i~ ( ~.' ). 3 a71"o.\nroiiaa should be read wih ~&c., no, pace Lagarde, a7i"oaei7i"ovi14 of B &c. The endency of correcors was o aler he o forms of he second aor. which were becoming obsolee ; cod. A consanly alers -l.\71"ov 9 -l.\e111"ov, Gramm. o/ 0. T. in Greek p. 234 VOL. XIII. E

5 50 THE JOURNAL OF THEOLOGICAL STUDIES ' &:n-os~un I A, (') '\ 7rVp TL~ V L KOA.'Tr'f! 7'0. OE ~' e LJLO.TL ' OV, KO.TO.KO.VUEL, ; 2 I (u) vii 9 ~v{k' /lv ~uvx{a. KO. VVKTEpVO<; ~ocpo~ UTa. 2 I So he Armenian Version according o Holmes and Parsons : he Greek MSS end he line wih vvktepvfi Ka -yvocpws7j~ For furher verses in his dramaic episode see below. (I 2) viii 8 mfvta. SKa.LOUVV1J'> p.ita MJL«Ta TOV utop.a.to<; p.ov, 3 2 I ol!s v lv() ukoaov (To!To<; oll) utpa.yya.awses. ( ) 8 I 3 IX l a > \ \ > > I~ 1 ~ ( 'f\ ' ' ~ 11 ' > ' 0.1\A. 0.7r07r7J07JUOV, JL1JO O.V I TO UOV OJLJLOo 7rpo~ O.VT'r]V ~<T'!1~s a>..a.6tpwv yap ilswp oiltw~ Sa./3~0"'! I 3 1 (14) xiv 4 ov S ye~p.a.ta 7roA.Aa (ru)([l), cpa.vepa 4 {3oo~ luxvs. 2 I (I5) xxi I4 cpnsop.evo~ Swpwv luxvpov 9vplw lyelpn. 2 I Heads and ails: hexameers incomplee in he middle. In hese insances he utlxos, designedly as i seems, begins and ends like a hexameer. The middle of he hexameer is unrepresened. Someimes he hexamerical exremiies are unied, and here are no inervening words. Thus :- (I 6) iv I 7 ose yap utolivta u"it' &.ue{3ela.s. (I 7) xxv 3 ollpa.vo~ fji/n/ms, yij S f3a9i'a. In he former of hese insances he pronoun has no equivalen in he Hebrew, being apparenly insered, jus as i is in (I8) below, men' graia. The laer insance finds an echo in Job xi 8, where he A ex appends o vl/n/a.os b f om. o (CJ ollpa.vo~ he words yij S f3a.9la. (sic). These merical ags were easily remembered, and have a way of repeaing hemselves. A oher imes he hexamerical exremiies are separaed by unmerical words inervening. Thus :- (x8} j 5 TwvSE yap (d.kovum uocpo~) uocpwtepo<; UTa.. The firs wo words, unrepresened in he Hebrew, recall (16) above; he las wo similarly end a a-tlxos in ix 9, cf. uocpo<; la-ta x 4a. The whole line was probably, in view of his laxiy in he maer of long and shor vowels, inended by he wrier as a rough hexameer. 1 p~ xpoviupr <v.t/il TO"'!' may be a gloss on d7romj3'1uov. 2 /lvopa B( is an obvious error (cf. xxiii 5). s So 109; <... unjups (ce.) may have come from xxiii 5 No Heb. equivalen. The word is insered Hf#ri graia, as is e. g. avbpau in x 10, which easily falls ino verse: (oppauv) ivvevaw uvva'yfl (Toil) dvbpau i\v1rar.

6 NOTES AND STUDIES SI ( 19} xiii r 2 Kpiiuuov 1 vapxo,...ivw; (/3o7JOwv Kap8{'!- Tov -rrane..\..\o-, '),,, '"', f-'wov KaL (L~ EI\1TLO ayovto~ (20} xxiii 20 1-'iJ '{q(), olv01l"ottj~, (1-'718 lkte{vov crof-'fjo..\als,) KpEwv &.yopau f-'o'i~. Lagarde has shewn ha uvl-'fjo..\a'i~ is a duplicae rendering of he words ranslaed by olv01l"ottj<>; he line may herefore once have been a complee hexameer. ( 2 r) xxix I KpE'iuuov 1 &.vi]p (..\~v} &.v8p{>~ CTKA7Jpo-rpax~..\ov,, /: ' ' (..1. l ' ~},, 2 E~a71"LVTJ'> yap 'f'"~of-' "0V avtov OVK ECTTLV LaCTL~. ( 22 ) XXlX. 42 CTLVOOVaS ~' (' E71"0LTJCTfV ' fcal ' a71"foot0 '~ } ~' ~ X 1 7TfpL'oWf-'aTa TOL<; ava vaeol~. Hexameer endings : versus paroemiad. The insances of hexameer endings. are very numerous. I will place in he forefron hose consising of he second half of a hexameer, he porion following he caesura, This half of he hexameer is merically equivalen o he line which in anapaesic mere is known as he versus paroemiacus, he 1 proverb verse'. The proverb verse seems, in fac, iri is origin o have been an incomplee hexameer, and o have had no connexion wih he anapaesic sysem. I is noorious ha he paroemiac in Greek ragedy rarely, if ever, conains a proverbial saying. Greek proverbs, on he oher hand, if merical (as a large proporion of hem are), are wrien in one or oher of he wo mos familiar meres, hexameers and iambics ; anapaess are unrepresened, unless he ' paroemiac ' is o be regarded as such. We find proverbs consising of complee hexameers or complee iambics ; he more pihy of hem are, however, compressed ino a few words forming he beginning or he end of eiher of hese lines. In he circumsances i is reasonable o regard he 1 paroemiac ', which is very common, as he laer half of a hexameer. The name was aken over as he designaion of he concluding line in he anapaesic sanza, he scansion of which accidenally coincided wih ha of he semi-hexameer of he proverb-maker. This may be common knowledge, bu I have failed o find any auhoriaive saemen on he subjec. In he Greek version of Proverbs here is a similar large use of hexameer endings and beginnings, bu an almos complee absence of any approach o anapaesic rhyhm. I have herefore classed he paroemiacs under hexameers. The ranslaor in his fondness for his form of ending is cerainly conscious ha i had inheried a name which marked i as par excellence he mos suiable ' MpEiaa0111 MSS, excep 103 in (19). In (21) he word may be insered merigraia. 2 The c in iiiallac may be shor in lae poery (L. and S.). Cf. he conjuncion of i[a1fi1'71s (a form more suiable for hexameers han he usual LXX i[a'ifcva or i[alpv'ijs) and dv[atos in vi 15. E 2

7 52 THE JOURNAL OF THEOLOGICAL STUDIES medium fon1. wrier of maxims. 1 The fac ha one phrase has for him become sereoyped, recurring as a sor of refrain in four passages, pus ou of he quesion he possibiliy of undesigned coincidence. (23) x 3 (w~v 8' &.a' {3wv &.vatpil{m. X 28} xi 23 (.\1rl.~ /)' &.cn{3wv d1roa Ta. xv 6 Kap1rol. I)' 6.u {3wv d1ro.\ovvm. In ii 2 2 ~8ol. /)' 2 &.u {3wv lk y71~ 6.\ovVTa probably he ex is a faul (read &.r.o.\ovvta}. Cf. Job viii 13 l.\?tl.q ap &.u {3ov~ &.1ro.\ i:ta ( 6.\ i:ta A). Oher examples of his ending are:- (24) i 26 = 27 PX'YJTa VJLLII 3 6.\ 0po~. () b ''"'''()' '' I 25 IV 27 avtos u op as?to'y}ct Tas Tpoxa~ uov. The las insance, pracically a complee line, recalls he hexameer form of iv 26 as cied by he aucor ad Hebraeos (xii 13). The Greek version of Proverbs has, besides TpoxW., an alernaive word for ' (car-) rack' (;Jlll), namely /J.wv. This is no necessarily an indicaion of a pluraliy of ranslaors. The choice of words made i possible o keep he hexameer rhyhm when Tpoxul. was useless for he purpose :- ii 18 (26) vi 10 1rapa ' ( T'' ~) aoy ""' 4 Tl>V~ ' a~ova> "l: a VT'Y}>. ~ { OA.yov JL V V'ITI!O~, oa.yov o Ka 'YJCTa. >\ I ' ) '< ~ >\I <'' I() 5 The refrain is repeaed wih a sligh variaion in xxiv 48 6.\{yov vvctt~w, IJ)..lyov 8 Ka8v'ITI!w, and is followed in boh passages by an imperfec senarius. ( 2 7) X 11 'IT'YJY~ (w1ls' (v xnp 8Ka{ov. (28) xi 20?Tpou8 KTO 8' a~<? mfvt S 11.JLWJLOL IlavT > is insered meri graia; we should follow codd. 109, 147, 157 in rejecing b Ta~ ~8os- avtwv a he end of he line as a Hexaplaric gloss. The refrain recurs in I xxii II (29) xxiv 14 (3o) xxiv 35 8 KTol. 3' a~<??tavt S.IJ.JLWJLW [ X {A u?tap.a{vn {3auA '1k (.\?T{s u' ovk (ykataa ll{, ~~~ /)' Eo3ov 8 ovk d?tvl{ v. Hephaesion (foe. ci.) objecs o he name because i was no he exclusive mer,e for proverbs ; his Scholias defends i. 2 Inser a< wih A 68, 103, ro6, 109, &c. For accen see Monro Homeric Grammar p pnd T.O.v n-y vrw is a double ; Lagarde considers 1rapd To/ iibv o be he older rendering. 5.pucpov Ill PVO'Ta'm is a double, omied by V ( =' 23 '), 109and he Armenian VS. 6 avroil is doubless a laer inserion.

8 NOTES AND STUDIES 53 ( 3 I) XXV o Kvpws «ivta1to8~1t(!. 1 (32) xxvii 27 ~v ~w~v uwv (1 pam5vtwv. To hese insances we should probably add (33) xiv 9?Jcp).~uovu-w 2 Ka0apup.6v. 3 The following are examples of semi-hexameers (3 fee), jus falling shor of he full versus paroemiacus. (34) vi 8' c Tp..fJuaua 1rpo~xOrJ (35) vi 18 (l7rlpwoovt S KaK01TO! i.'v. (36) x 6 KMvY,n 1Tlv0os /J.wpov. (3 7) xi I 6 yvv1] p.luowa &KaLa. (38) Xi 22 yvvakl KaKOcppovL K~QS. (39) xiv 10 oljk wp.lyvwril ~PP ' (4o) xxii 29 1rap 1TTlva d.vopcf.u vwopoi.'s. ( )..., \_. ' 4I XXlV 50 OVK (V 1rLJ1.1Tw.CTaV avttjv ( 42 ) XXlV 54 ( Ka ') 0~ '~ -~-- O''EWS ".I.. E1TL ' 7rETpaS ' ( Ka ' Tp!,..OVS 'R ) VI]OS ' 1TOVT01TOpOVcrYJS, ( Kal OOOVS avopos) W V 0T7}T!. Here he ranslaor's apparen inenion is o se ou, as i were, in abular form he 'hree hings which are oo wonderful for me', by giving each of hem a separae hexameer ending ; here are ' yea four ', bu he fourh has defied his effors: In he middle half-line concerning he ship 'in h~ hear of he ~e~' (C' :l~:l) we have a Homeric reminiscence:- Od. xi. I I rijs 8~ 1TaV'f/Jl-Ep.,.. Ti-raU wtla,rovt01topova:1js (43) xxvii 19 Jp.oa 1TpOCTw1Ta 1rpouW:n-oLs (cf. verse I 7). The foregoing insances amply suffice o esablish ha he hexameer endings a he close of he ITTLXOL are he resul of design. If all he examples falling shor of hree fee were added, he oal would be brough well up o a hundred. A he risk of wearying he reader and for compleeness I will add hose which I have noed amouning o a leas wo fee. The following are beween wo and hree fee :- (44) vi 8a ws lpyans l<tt{v. (47) ix 3 Kparijpa Ayovu-a. (45) vi 24 yvvak(js v1ravopov. (48) X 3. ov A!Jl-OKTOV.qiT! (46) viii 2I aya1twu!v V1Tap~!V. (49) X 4 7T v{a /J.vopa Ta1T voi'. 1 The added words uo a-ya8a may be due o a scribe's unwillingness o leave he direc and indirec objec unexpressed ; a-ya8a has no Heb. equivalen and is omied in cod. V. 2 So cod ; O<pEA~uovu"' ce. Also xiv 21 Elo.Ewv (llo, om. 295).,.,.~xovr ij4kaput6s. MaHapuTs is wrien meri causa a he end of a O'Tixor (xvi 20, xxix 18) 2 po.ccipos (he usual LXX word) in an earlier posiion. The las word insered merigralia: cf. i JI.

9 54 THE JOURNAL OF THEOLOGICAL STUDIES (so) xiii 2o (sr) xi~ I2} =XVI 25 (52)xiv23 (53) xiv 32 XV 10 { (54) cf. ix IS (XVi 7 (ss) 1 cf. xxi 7 ITO/>O~ ITo/>0~ littal 1, d~ 7rVOp.lva ~aov. ~ ' C:VC:ITTL 7rC:pLITITOV (hit }ui'ty]'tl UKaw~. v1ro rwv 7rapL6vrwv. roir~ 7rapL6vra~. 7rOL W ra UKaLa. 7rpQ.ITIT LV 'T,)_ akala. The following are wo fee only :- ( 64) jy I 2 olj KO'IraiTC:L~.? (65) jy 24 JL«Kp!V a7rwital. (66) vii 26 o~~ 1rc:J>6vwKc:v. ( 67) Viii I ITOL 1nraKOVO"ff ( 68) xiv 15 c:l~ p.eravolav. (69) xvi I2 ( rolp.~c:}tal Op6vo~ dpx7j~.5 {70)lxix 7 a/>povl TpvJ>~. ( 71) XX Vi I a/>povl 'TLJJ.~ (72) XXVi 8 aj>povl OO~aV. (56) xviii 13.\oyov 1rp~v dkovita. (57) xxii 2I dya~v 1nraKovnv. 2 (58) xxiii 27 da.\6-rp~ olko~. {59) xxiv 64 () Ka.\w~ ala{3avn. (6o) xxiv 77 Kal Kp'ivc: OLKa{w~. 3 (61) XXV 18 r6gwp.' dklow'tov. ( 62) XXViii 28 ITTvOVITL a{kaol.' (63) xxix 45 ITo/>o~ vop.oolitp.w~. ( 73) { xix I 3 ( ha)wv d'l!'oac:rra. (74) xxiii 28 (ITVvr6p.}w~ d'l!'oac:'i- TaL. (75) xxviii 8 Ka~ 7rAc:ovaiTp.wv.?(76) xxix 7 voii~ bryvi!jp.wv. (77) xxix 29 oljk d7rop~it L. (78) xxix 33 ra~ Oc:pa1ral va~. (79) xxix 46 j}vc:itc:v av~v, Hexameer beginnings. These are far less numerous han he endings. The ITrxo~ opens wih he firs half of a hexameer in i Io (omiing vu), v I6, 20 p.~ 1ro.\ir~ ito 1rpo~ d.\aorp{av, vi 4b, 20, zsc, xi 29b, xxiii 2c, xxviii 12b, Consecuive lzexameers. The dramaic scene of he seducion of he young man by he harlo abounds in fragmens of poery. Mer wha looks like a senarius in he earlier par of her address f Vii 1 5 'T~ ITOV 7rp61TW7rOV (l/c'pro(}oijit' c:iifyyika ITC: I here follow fragmens of several consecuive hexameers :- 16 dp./>l'ta7rol~ a EITTpwKa 'TO~ 11'11'' 'EyV'II'TOV/ I 7 (Ka~ OJ EppayKa 7 KpOK. 'TOV a olkoji p.ov KLVap.I!JP-'.B I uo<f>os <uv of B, which Lagarde adops, is an accommodaion of he ex o he M. T. For U11p:ropE110f1EVor (2 ) we should read U11ppEf1/3&f1EVor wih 68, 109, 147, &c.: cf. vii I 2 pifi{jeta. 1 Perhaps a paroemiac : "'(VWUIV (T'} d-y. inr. s A paroemiac, if wih N we read BuiepwE. The nex line, wihou ~eelvow (cf. cod. V), is a rough hexameer, The las word insered meri graia: cf. i I I, s A11'i,.Tol1 MSS. One hesiaes o fill he blank wih eafv'l" J-1011 from he previous line. 7 1Mppa"'(ea (lppa-,ea Chrys.).,.~, eoltj]v J-1011 MSS. 8 ewaflwfl'i' MSS. For he spelling see L. and S.

10 NOTES AND STUDIES 55 IS A.O'/ d.ro>..a-vuwp.fv cpv..{as lws op8p(os /C'I]'Ta), 2 ~wpo s ( bkil)>..u8wp.fv lpwt. e '.!J~') 1 e ~' ' 1 19 av7jp o"\o rap OTLV, ooov p.akpav 7r' 7r'Opvnm. 2 3 I The resul of her blandishmens (21 f) is described in iambic rhyhm (see ex. (23) below). For furher examples of consecuive fragmens of hexameers see (3) and (28) above. IAMBICS Complee or nearly complee iambic lz"nes. A fair number of complee lines are o be found in he received ex as i sands. The number may be very largely increased by minor addiions or aleraions, or by ransposiion of words on he principle saed above. ( 1) xii 26 },, 5.-, l::,.., ap.apravovtu'> KUTQOLW~ TQL KaKa. = Xlll 21 (2) xiv x6 (3) xiv 24 (4) xvi 22 (5) XX I (6) xxii 24 (7) xxvi I (8) xxvi 26 (9) xxvii 13 uocp6s cpof3718ds ElKALV V d.r6 KaKov. OT cpavos uocpwv ravovpyos, ij 8E 8aTp{3~... '~('\...6 ~, 7r7Jyr] ~ 0 /'/'> fwola TOL'> KfKT'YJJI.fVOL'). d.koaautov olvos Kal 7 ij{3putlk6v p. 87J. p.~ 'lu8 -ra'ipos &.v8pl 8vp.i:.8, cp{>..'l! 8p6uos w d.p.fr'l! XffJU7r p va6s w 8 p L /x.p.apr{as royvwuto'> 8 lv crvv 8plos. ilf3purry,s 60TL'> -rd.a>..o-rp1a >..vp.alvaa. For 7J reaed as a shor vowel compare hexameer (3) above and he secular proverbs quoed laer in his paper. The iambic endings o he lines preceding he las insance (... rovn8{utovs.myovs, &.r Kpvf37J, 1rar"JA8 y&p) should be noed. A very sligh aleraion or addiion complees he line in he following passages:- (xo) xi 12 p.vkt7jp(u)n ro>..l-ras v8~s cpp vwv. MSS p.vkt7jpl, which should perhaps sand : ;; and r, a and a are used indiscriminaely in his species of poery. These iambic fragmens 1 + KU MSS. 2 ~s op8pov MSS. 3 So 109: 3Eiipo 3E V, perhaps righly: BEiipo Kal he res. 4 ou "(ip ~rap. cl dv. JJOV iv oikrp MSS. Omi he BE afer,.nr. wih 68, 109, &c. 5 The BE in he firs passage is omied by he cursives 68, 103, 106, 109, &c. e The consan form in he Tragedians; 'a>ijs MSS. 7 The wan of elision is allowed in hese apophhegms: cf. ObK lut,.evfas ou3e 2v jl<i'(ov Jmll6v, Menander rvwjl«l Moi'OITTIXOI 436 ap. Meineke Fmg. Comic. Graec. iv p s Noin he Heb.: insered apparenly merigraia. Cf. (17) below.

11 S6 THE JOURNAL OF THEOLOGICAL STUDIES ofen come in clusers. Noe he merical endings o he preceding crtxo : KaTJJp6wcr v 'II"OAs-, dcr {3wv KancrK&.cp7J, and in he nex verse &.v0p o{yawctctos' (d'll"okaam ) {3ovAdS' lv CTVV 0p{1f!, 1 and, wih ransposiion, "" ~\ \ 1 S 'II"VOT/ 0 'II"LCTTOS' 'll"paf'jmlt (' ) ' f' KPV'II"T L The convenien phrase va ~> cpp&wv recurs a he end of a crt{xos- in vii 7, xii 11, xv 2I, xvm 2, xxiv 45 The verb in (Io} suggess ha he same hand is a work in he nex insance :- ( Il) xii 8 ( b)zvw6pokapoos' a ij.vktyjp{~ Ta. Traces of iambics underlie he lines immediaely following (in v. Io dv A ~p.ova should be pronounced or wrien dvyjae~p.ova, for which here is auhoriy), and hen, by ransposiion of one word, we have ( 12} xii 11 a ozvwv 6s- ECTTLV ~OVS' v 8aTpf3aiS". Cf. (3 ). (I 3) xiii I 0 ( a~p) KaKOS' p. 6' i'j{3p WS' 7rpMCT L KaK&., he nex line ending.. 7ryvwp.ov S' crocpo{. (14) xxiv 45 (x>>) 3 &.p.7r Awv iv6pw'ii"os' lv8~s- cpp&wv. Transposiion of words, wih minor aleraions, produces a large number of insances, exending someimes o several consecuive lines. I feel coiwinced ha a good deal of ransposiion has aken place in he MSS on he principle saed in Dr W. Headlam's paper. I do no Jay sress on any furher emendaions suggesed, and hink i may be los labour o aemp o complee he verses. I seems unquesionable ha he ranslaor consciously impared an iambic ring o porions of his version, bu he impression produced is ha he was conen wih a parial approximaion o poery, and did no always rouble o produce finished lines. We mus, however, allow for he possibiliy ha his work is based on an older poeical source. I will rever o his laer. (IS} i 3 (o acr6a} crtpocpas Aaywv n Ka AVCTL S' alvyp.&.twv. 2 3 I The las hree words are added in a group of cursives ( 68, 109, r 47, &c.). Their source is presumably Wis. viii 8 (unless here is an older source behind boh passages), bu ha would no absoluely preclude heir having sood in he original ex of he Greek Proverbs, which is, I believe, he laer book of he wo. (r6} i 23 'II"VO~S' (8() p~crv (~>) lp.~> 1rpo~op.a. (I 7) iii I 5 vyvwcrtos' ECTTL 'll"iicrv lyy{~ovcr{v ( o)! 1 iv avvellp 9' a free rendering, meri graia. The large use of avvollpcov in his book is noeworhy. 2 The addiion is perhaps unnecessary. For an iambic proverb lacking he iniial syllable cf. Zfus K«Tfi'3f xp/jvos flstds 1Juf>8lpas, Paroem. Graecii p MSS allfla'llfp. MSS 'lliiaw Toi's f"y'yl(owv cwtj1. Cf. (8) above.

12 (18) iii 34 (19) iv 3 ff (2o) iv 27 f NOTES AND STUDIES Wr p7jcpallour{!ll) a!itltooct TaL ((}djs). 1, ' ',, (' ) 2 ~, ' Kayw yap VWS 7/ll V7nfKOOS TraTpL Kall ' ( T!f> ~) 'TrpoCTW'Tr'f> ' JL1JTpoS ' ( 1JYa7r7JJL lios ' ), s ;A "f0ll (T ) Kd8{8auKOll o (yoll is)- A6yos ~JLET po'> ds ~ Kap8{all pn8&w ( Tas)!1To.\as cp-6a.auu, JL~ (' TrLAall8allov )!.. ~ ~.., ( ' ) ' ' ~r,, KaK7JS 0 TOll CTOll a7rocttp~'' 011 7rOoa"... ~' ', ( ~ ).. l: ~,.. () ' OOOVS yap K TWll O ~LWll OW 11 OS, (8) CTTpaJLJLEvaL 8 duw ;.e dpctt pwll. (2r)v 20 f. dykaaals njs JL~ l8as llijnrwll yap duw ocp8aa.jlwll 6 ov. 2 I (22) vi 13 f CT7JJLalli L 6 Tro8,. 8L8auKu 6 8aKroAwll W VJLaCTLII, ( 8 ) CTTpaJLJLEvy ~( 7 Kap8{'/- T KTWli TaL CTVII{CTT1JCTLII 7r0Au. ( 2 3) vii 2 r Trolifi 8' 0.Tr 7rAali1JCT ( TOll8') 8 ojlla.f.:.,. atro X LAwll {3p6xou() T' l[wma.l (llll) 9 E7r7JKoAov6YJCT ll 8 K 7rcpw6 is. Cf. on his passage wha is said above under Consecuive hexameers. (24) ix 18 b... Kd?To ~s JL~ Trlns allotpas,!8 c 2 3 I lla TroAw ~uns xrolioll w1}s s ;T1J CTOL 7rpOCTT (}~CT TaL 10, (25) x 27 cp6{3os (8 ov To) 11 TrpoCTT ()1JCTLll ~p.lpas. ( 2 6) XXV 6 JL7J8' Ell 8waCTTWll ( TOLS) T07rOLS vcp CTTaCTO. (27) XXV 10 a (xaps Ka2 cpa{a) laro6 pol. ' ) 12 ~ ' 1S,.. ' ( TYJp! <T avt!f> JL1J 7r0liUO!CTTOS Y 11'[/, all'(~) cpvaa[oll WCTVllaAAUKTWS o8ovs. No ransposiion or emendaion is made in he remaining insances. The following are examples of wha I have called Heads and ails. (28) viii 5 lio~ctat' 11KaKOL Trallovpylall. (29) x 18 o 8' Kcppo!IT S (A.o8op{as) O.cppoliWTaTOL. 14 (3o) xiii I vias Trallovpyos ~Koos TraTp. ( 31 } XXIX 2 7 J0 1\.V"fJLa R..'' 0 "' alio/l!f> ' KaT V (} VliOVCT ' ' OOOS... ' 1 MSS Kvp&as, M. T. N11'1, o 8<6s in he N. T. ciaions. The ranslaor presumably read 1W1', which he consanly renders by B os ; see an ar. by Dr Red pah in ]. T. S. vii 6oS. 2 MSS E'"(fi'OflfJ" ~ MSS d'"(a1twf1<vos. 4 As in iii x. MSS here have 1-'il lriaobu. SoN: or read all wih he res of he MSS. 6 +lifmss. 7 Inser lle wih NA V, 68, &c. 8 MSS aurov. 9 MSS avrov. 1 So codd. 109, 147, 157, 254, 260 for 1rpo11r<B~: cf. ix 11. u MSS Kvplov: see noe on (IS) above. 1 2 MSS as rf!ptj110v. 18 MSS lvaf1i). 14 Perhaps inended for a rough line. 57

13 S8 THE JOURNAL OF THEOLOGICAL STUDIES Iambic endings. An iniial foo or more is waned in- (3 2) i 2 I, l1r' IJ.Kpwv 1 T LXfWV KT/P"V!T!TETa. (33) iii 22 a... E7rp.(>..na. Toi:s!Toi:s du(os. (34) V 10. amotpo o-ljs l!r)(llos. (35) xvii 16 (!va. T) v1rij~e XP~p.a.T' II.cppov; (36) xviii 3... E1s f30.8os Ka.Kwv, Ka.Ta.cppovEi. (37) XX 4. dkv7jpos o-flk a1o-xrjv(ta.. (38) xxii {3a.!T>..Ev!T BEi 7ra.pEITTava. (39) xxvii 22.. J.cppov' lv p.(!t'f! crove8p{'f! ' (40) XXViii r0pEvop.EVOS p.7raak~!teta. The following consis of he laer half of he line, he 3! fee following he caesura, and are herefore comparable o he paroemiacs. I place firs hree recurren insances, one of which is imporan. (41) vi 29 3 } XV~ OVK ab'l?w()~!teta! XVll 5 (42) xi 21 l, 5 ~,, 6 j xnp xnpa.s Ep.{3a.>..wv.. XVI (43) XX } ~ (, ~) )..' cf. xxvi 2 2 v Ta.p.na. or E> Ta.p.na. Ko a.s.? ( 44) i 11. KoVWV7J!TOV a.p.a.to<;. ( 45) ii I 4.. (7r). 8a.!TTpocpjj Ka.Kjj. (46) vi >..vxvos lvtoa~ vop.ov. ( 47) ix I 2.. p.ovos ~v avta~!tns Ka.Ka. (48) XV 7 acppavwv o-flk d.ucpa.aei<;. (49) XVi 18.. (1rp0 8 ) 'Tf'Ti:Jp.a.TOS Ka.Kocppo!T6V7J. 8 (so) xxiv 6... Ka.p8{a.s f3ov>..evtkijs. ( 5 I) xxiv Evo8ws 7rOpEvETa.. (52) xxvi II a = Sir. iv 21.. l.1r&yov!t' ap.a.p7 a.v. The phrase in example (43) E1s Ta.p.lia. Ko>..La.s has an imporan bearing on he dae of he ranslaion. The shorer form Ta.p.E'iov is unaesed in he papyri before he firs cenury A. n.; he correc Ta.p.Eiov is invariable in he hird cenury B. c., and occurs once in he second. For he firs cenury B. c. evidence is waning ; ha is he earlies dae a which 1 Omi a< wih V, uvv 3p{Qv Bl< &c. 3 Wih he words immediaely preceding we ge anoher complee line : "'(VVa'i«' Vrravapov, ovi d8pc&1671aeta. 4 An alernaive rendering, possibly by anoher hand, is ou«&.r,prijptjtos iura, xi 21, xix l, 6, xxviii The 3~ in he second passage is omied by nine cursives (68, 1o6, 1491 &c.). 8 Following a 'head and ail': 7rp0 uvvrp /3qs (J.IE'V) >)"'(ErrOI v/3p s.

14 NOTES AND STUDIES 59 Tap. wv is likely o have been wrien.' Tap.{ )1:a in hese passages has he suppor of BN and of some cursives which elsewhere prove hemselves rusworhy. 2 The mere also favours he ex. If he reading is righ, i goes o confirm he inference, which I have elsewhere drawn from anoher orhographical deail, 3 ha he Greek version of Proverbs is no older han Ioo B. c. This insance furher suggess ha he ranslaor himself is responsible for he parial versificaion, and ha he is no culling excerps from an older collecion enirely in verse. I would be edious and is unnecessary o caalogue he examples of ut{xo wih an iambic ending of 3 fee and under, which would bring he oal number of lines and fragmens in his mere well over loo. Iambic openings of lines, like hexameer openings, are fewer han endings. Among oher insances we have :- ( )... ' ~' 53 Vll I I av 7r'T fjwji- VTI (' ), 0 U'TW KaUWTO<; (54) viii 30 KaO' ~p.ipav 8' vcppavop.7jv. ( 55 ) IX. I 6 0<;' ft'tv ~ VJI-WV ~ a'i'pov U'TaTO<; '..1.. '.. Cf. (56) XXiV 2 5 acppoveu'tato'!> yap ip. (57) xiv I uocpal. yvvalk <;!fko(80p.7jrrav oikov<;). (58) xxiv 67 iav 'Tf'Poll umvtov (d<> vcpporrvv7jv).' (59) xxvi 19 OTaV oe: cpwpaowuv 5 Consecuive lines. To hose produced by ransposiion, &c., quoed above we may add hese fragmens:- { 6o) v 4... ~KoV7Jp.ivov I p.a.>.aov p.axa{pa<; 8U'Top.ov.. ( 6 I) xix 4 f... ov uw(}-qrr Ta. I b KTiJp.&o<; cppov'yjo'v In boh he hexameer and he iambic porions one small grammaical poin-he use or omission of he aricle before a possessive pronoun or before an adjecive used subsanivally "-is governed by merical consideraions. Conras he following :- ix 18a XXV 2 TO uov op.p.a 'Tf'pO<; avt~v. Kal. p.~ Ta ua X {A7J. Hexameers. 11 I o i<; u~v 7 8ufvoav. vi 2 1 i'tf'l. ail lf!v;ci oa 'Tf'aVTo<;. xxvii 2 7 ~v ~w~v uwv 0 pa1f'ovtwv. 1 Gramm. ofo. T. in Greek i 63 ff. 2 In passage (1) 157, in (2) 109, 157, 295, in (3) 147, 149, 157, 159, 295 s Gramm. of 0. T. i 61. OuaEi> is used hroughou he book o he exclusion of oums, which was almos universal in he Polemaic age unil abou 130 B. c., when i began o go ou of fashion. 4 So A~ V, 68, 106, 109, 147, &c. Their ex, as producing an approximae (or raher, since o and r» are used inerchangeably, a complee) senarius, is preferable o <v Ev</>pouvvp of B &c. 5 So A 68, Io6, 149. &c. ; opa8wuiv of B &c. is obviously inferior. " Noe, oo, he omission of he aricle in Hex. ( 17) above. 7 TT)v male B.

15 6o THE JOURNAL OF THEOLOGICAL STUDIES xvi 7, xxi 7?rO 'iv (1rpacr:mv) Ta xi I6 yvvq p.wovcra UKaa. UKaa. Iambics. iii 22 a bnp.l.>..na To<; cro'i<; &crteo<;. iv 4 l<; crqv Kap8{av. v I o JA>..6-rpw ~,. lcrxvor;. The posiion of he dependen personal pronoun before or afer is governing word is, in he hexameers, affeced by he same consideraions; evidence in he iambic porions is waning. Conras:- Vlll I cro v?rakoijc17j. xxu 2 r 1rpof3a>..Aop.evo<; cro. xxiv I4 (A?r{<; er' o~k lykataa yi L. viii 8 Tov crtop.ato<; p.ov. xxii 28 ol?ratep <; crov. xxiv 49 7J?r V{a crov. Final a~tov, a~t1jr;, &c., are very useful in hexameer endings:- e. g. ii I 8 TOV<; l.~ova<; a~<;, vii I I oi m)8 <; a~<;,.xxv 5 8p6vor; a~ov, xxvi 1 o KcrTacr<; a~wv. ' Wha explanaion are we o give of he phenomena? Were he half-verses ever complee? Are hey he disieca membra of an original work or works wrien enirely in verse? The lae Dr Redpah, if I remember righ, was inclined o adop some such view, hough I do no hink ha he had formulaed any definie heory. One hypohesis may, I hink, a once be se aside. namely ha he version which has come down o us was ever wholly in verse. Large porions of i are unmerical, and he ex of some of hese prosaic porions is aesed in he earlies known ciaions from he Greek Proverbs in he pages of Philo and he N. T. I is difficul o suppose ha he ranslaion, which we have seen reason for hinking was no made before roo B. c., had wihin abou a cenury afer is producion undergone such radical change. And if he bulk of he version was in prose, i is improbable ha he ranslaor (like Dane in La Via Nuova) on occasions alogeher abandoned prose for poery, inerspersing large paches of he laer in wo differen meres ; ha he aemped wih varying success o impar a poeical colouring o he whole seems more naural. Anoher possible explanaion has more o recommend i. I migh be hough ha our ranslaor made use of an older verse ranslaion or paraphrase of selec passages from he book of Proverbs, or perhaps raher wo ranslaions, one in hexameers, he oher in iambics, and ha he incorporaed phrases from one and he oher in urn. Some warran for a belief in he exisence of a los collecion of proverbs, parly Biblical, parly unscripural, wrien in iambic mere, has been found in a passage in he N. T. I refer o he allusion o he uncleanly habis of he dog and he sow in 2 Pe. ii 2 2 crvp.f3 f3'1jk V a~to'ir; To T17<> JA'IJ8ov<>

16 NOTES AND STUDIES 6[ 7rapopla<; Kyw,.. ETTICTp '!'<~.C bri ni ih.ion!ipama 1 Ka ~Y<; >..ovctap.vyj :l<; KVACTJWV (v. 1. KVACTp.a} f3opf36pov. The reference o he dog seems clearly derived from Prov. xxvi II, hough no from he LXX, which has here a quie differen and purely prosaic rendering : i:j!t7r:p Kvwv 6rav i1rl>..by 1r rov avrov ~p.nov Ka p.cttfto<; yvtfta. The sow has no equivalen in he Hebrew or Greek book of Proverbs; is origin has wih much probabiliy been raced o a parable in he Sory of Ahikar abou a pig which wen o he bah wih people of qualiy and on coming ou wen and rolled in a muddy dich. 1 I has ofen been poined ou 2 ha he pair of proverbs in 2 Pe. runs easily ino iambic rimeers ;- e7r' illwv Ulpap.' E1f'CTTp!f!a<; KVWV A:Aovp.VYJ o' ~<; E<; KVACTp.a f3opf36pov. Here hen, i migh be hough, is a relic of a los iambic collecion of miscellaneous proverbs, in which he Biblical dog was associaed wih he unbiblical sow, jus as in Proverbs LXX he canonical parable of he an is reinforced by ha of he bee (vi 8 a). Again, we have o accoun for Blass's 'faulless hexameer' (no from LXX) in He b. xii 13: Ka rpoxos opba<; 7ro~CTaT TOL<; 1f'OCTLV vp.wv, which migh be regarded as a survival from a los hexameer collecion. Bu he aucor ad Hebraeos has jus before (xii 5) quoed wo verses from Proverbs LXX verbaim, and i is herefore probable ha he is quoing from i again, only more freely. The hexameer is produced by conversion of he singular verb (which in he 0. T. occurs in he address o 'my son') ino he plural, and by ransposiion of wo words. The plural is necessary o he N. T. wrier in order o adap he ciaion boh o his readers and o he immediaely preceding ciaion from Isaiah. Tha he hrew he line, unconsciously perhaps, ino hexameer form is in keeping wih his procliviy for rhyhm. A ' endency o fall ino iambic rhyhm' has likewise been noed as a 'feaure of he syle of 2 Peer', 3 ogeher wih a preference for grandiose language. The iambic ring' of 2 Pe. ii 2 2 and he rare words ~lpap.a and KvACTp.a may herefore be explained wihou recourse o he hypohesis of a los collecion of proverbs in iambic mere made by a Jew of Alexandria. Such a collecion may very well have exised ; bu he poin o be I Tie Sory of Ahikar, ed. Conybeare, Rendel Harris, and Mrs. Lewis (Camb. Univ. Press 1898) lxv f. As Rendel Harris poins ou, he sory of he pig 'going o he bah' explains and jusifies he middle voice ("A.ovuaplvTJ) in 2 Pe. 2 See in paricular he ineresing remarks of Dr Bigg in his inroducion o he Ep. in he ln. Cn~. Comm f. " Bigg in I. C. C This, as was poined ou o me by he Rev. E. D. Sone, exends o he inroducory words which wih a sligh change migh be wrien Tu T1js cl"a.7j8ovs l{e/3tj 11'apo1pias; bu obviously his inroducion could no have formed par of he hypoheical poem.

17 62 THE JOURNAL OF THEOLOGICAL STUDIES emphasized is ha he heory of a los poem or poems lying a he back of he Greek book of Proverbs does no help o accoun for. he phenomena which i presens. For ( 1) inernal evidence proves ha he hexameer and he iambic fragmens in Proverbs LXX are he producion of a single hand. The wo meres have a common vocabulary and he same phrases recur in boh. Compare he hexameer endingsx I I 1MJY~ w~!> lv X p~ 8Kalov xix 4... lvvo~ &:yab~ 'TO~ d86cnv a~~v wih he iambic linexvi 2 2 1MJY~ ( o)1j~> EVVO~a 'TO'> K KTTJp. vw;. "Evvoa occurs welve imes in his book, bu only once again m he 'LXX' proper. Compare again:- Hex. xxii 29. -rrap CTTava &.v8pacn vw9poi:~> Iamb. xii 8... vw9pokc p8~o!; S p.vkryjpl, Ta. Nw9poKap8w<> is a J:rr. A (' ; vwbpo<> occurs only wice again, in Sirach. Again:-? Hex. xxix 7... vov<> 1hnyvwp.wv? Iamb. xiii hnyvwp.ov!; uorpo{. The adjecive (4 exx.) wih he subsanive -rryvwp.om'wq is peculiar in LXX o his book. Again :- Hex. xxix 23 Tov<> S Ta-rrnv&rppova<> 86b.7 (B 0'> allto<>) lpel8e~ Iamb. iv 4... d., ~v Kap8lav lpe~8!tw. 'Epn8 v occurs nine imes in Proverbs, only once elsewhere in LXX. The same conclusion is suggesed by anoher line of reasoning. The hexameer fragmens no infrequenly come in clusers, and he iambics similarly end o fall ino groups in consecuive CTTlxo. Were his invariable, i would lend some suppor o he heory ha he ranslaor worked wih wo older poeical collecions before him, using firs one and hen he oher. Bu no seldom we find a fragmenary hexameer and a fragmenary senarius in he parallel members of one and he same Hebrew verse. This again suggess ha a single hand is responsible for he wo meres ; a piecing ogeher of disinc sources wihin a single senence is highly improbable. Thus we have :- v 20 f (H.) p.~ -rroav<> Zu9 -rrpo<> laaotp{av, (JL'Y}S uvvlpxov) (I.) &.ykaaal<> Tf}<> p.~ l8{a<; lvw-rrwv yap duv orp9a.\p.wv Bwv. xxii 29 (I.)... {3auA VCTL 8 1: -rrapecttava, (H.) ( Ka~ JL~) -rrap CT'TaVaL lv8pactl I'WfJpo<;. xxiv 48 (H.)... o.\lyov VVCT'Ta,w, o.\lyov S Ka9v7rVw, (I.) o.\lyov 8' lvaykaal,op.al X PCTLV ( CTTq()'Y} ).1 1 The firs 'I is probably, as elsewhere, reaed as a shor vowel.

18 NOTES AND STUDIES 63 xxiv 64 (I.) (Tpla 8llcTTLv 8.) :MSw~ '1'opru Ta, (H.) (Kal TlTaPTov) & KaAw~ Sa{3a{v. Parallel insances of he collocaion of he wo meres in he balancing clauses of a senence occur in he collecions of Greek secular proverbs. Thus:- 1' I I 1\ \ ~ > ~~ (H) OLVO~ TOL xap VTL '11' 1\ 1 Ta)(ll~ 111"11"0~ aolof1 1 (I.) vowp ~ o "' 7Tvwv I XPTJ<TTOV \ ovow "'" av T KOL~. I 1 Is i hen possible ha he ranslaor had before him, no wo poems, bu a single collecion wrien in he wo meres? No; inernal evidence suggess furher ( 2) ha he hand responsible for he merical porions is ha of he ranslaor of he whole (or he bulk) of he book. For he characerisic vocabulary of he merical porions reappears in passages which are, and probably always have been, unmerical. An examinaion of he use made hroughou he book of such words as lp:ls:w, crkoao~, T:KTalv:w, -ik-ap ~ will illusrae his. Again, in numerous passages wih a merical ring which is obviously inenional i would be difficul or impossible o complee he broken lines. For insance, he hree consecuive hexameer endings in xxiv 54 serve a definie purpose (see above on Hex. (42)), bu i is exremely improbable ha he lines were ever spun ou o heir full lengh. The book, as we have i, doubless conains many laer glosses and accreions, bu he versificaion is fairly evenly disribued over he whole of i, so ha i is probable ha he bulk of our ex goes back o he original version. The versificaion, i should be added, exends o hose porions which are peculiar o he Greek ex. We are driven herefore o he conclusion ha he ranslaor, sporadically, in places where he could readily do so wihou deparing oo widely from his original, impared a merical colouring o his work. He was mainly concerned o give he <TTLXo a merical ending ; more rarely he gave hem a merical opening; on occasions he wroe a complee line or couple ; in passages where he sring of deached proverbs was replaced by a conneced and dramaic narraive, such as ha of 'he srange woman' (chap. vii), here may originally have been several consecuive lines of poery. His procedure in fac seems closely o resemble ha of he old Greek proverb-wriers. We can race in he Paroemiographi Graecz he sages in he growh of he merical proverb : firs he purely prosaic maxims, hen he rugged jingles aping poery, he fauly or faulless half of a hexameer or senarius, usually he laer half (i. e. he paroemiac or he porion of he senarius following he caesura), and las, he complee line or couple, no always immaculae. For he purpose of corn- 1 Zenobii Cen. vi 22 ap. Paroem. Graeci i p. 167.

19 64 THE JOURNAL OF THEOLOGICAL STUDIES parison I have roughly analysed he proverbs in he oldes collecion preserved, he six 'cenuries ' of Zenobius. Zenobius quoes in all SS2 proverbs, of which a leas I6o are merical and perhaps unmerical ; some of he approximaions o mere may have been overlooked. The merical proverbs may be divided as follows :- Hexameers. Iambics. Perfec (or approximaely) 21 Perfec (or approximaely) 48 Endings. Paroemiacs (perfec) 1 IQ Endings. Over 3-! fee 4 4 Paroemiacs (approx.) IQ Beween 3 and 2 fee ll IQ JFee 5 (or approx.) 6 IS Beween 3and2!fee 7 2Q Beginnings. 2! fee 3 6 Beginnings. Over 2! fee 8 7 ' Head and ail ' (?)I 2! fee" 8 ss The mos ineresing of hese are he crude and illierae aemps a verse, which beray heir plebeian origin. A jingle wih a merical ring is a sufficien subsiue for mere. The vowels '7 and w, and he diphhongs n and m may be reaed as shor ; " and o may be long. Similarly, in our language, 'A sich in ime saves nine' fails o achieve rhyme where 'There's many a slip', &c., succeeds. The following are approximaions o complee lines:- Hex.... Aiye ovte Tp{To oilt TTapTo. KAa{n 0 JI K~uas, 0 a~ JI K'l(J "is U11"0AWA JI. Iamb. linv77s avayk'y/s' oviiw luxvpotepov. Ills 7rpOS' Tov aln-ov aiuxpo" 7rpouKpovnv AOov. The following appears o be a 'head and ail':...:._ ' '... '... ajit KaKOV KVJIO'O VJI Bu he numerous approximaions o he paroemiac furnish he quaines insances:- «<A>..os {3Lo<>, «<A.\1J /i{ata, 1 e. g. d yafh) Hal pqca. P,fT' O.pTov : E'/>V"'fOV a~6v, ~pov O.JlELI'ov. 2 e. g. d.u' oiuc aj9cs Manr1Jf: 'Pa/Japav9vos 15plos: 11u1ov l<f>' 'Eppfi. 3 e. g. 11'av.-a Ai9ov "'"f' : aliv BE 9Eol pa~<aps. ' e. g. a-ypo[lov p:q lata<f>p/wei /lf/ropos : 43fiS IJJI111'Ef' fls 4ijJ..ov 11'Af()JV. 5 e. g. lav9apov 110</>WTEpos : OVK UlffV 'Yf 81JI1f()JS. 6 e. g. {3arpaxocs olvoxof~ : Alvllcoc.-:;, lhlalav. 7 e. g.-aj..avta TavTaJ..ov: 'IJ..uh 1a1wv. 8. e. g. apltov 11'apOVI11JS rxv"ij pf, (~Tfl (1J l as in exx. in Prov. LXX above): El pf, lllvaco {3ovv, ~}..avv' 6vov : b-yeittepos lp<itoll!os (KpoT. ). g e. g. -r"-av( ( -y}..av~<') Els 'A9~vas.: 4u)s Kopcv9os. 10 Examples of 'Heads and ails' in iambic mere occur in Menander's -yvwpa1 ( op. ci. ), e. g. p:q 11'pO> TO I<Epllos a El 11'flpW {3AE11'fiV : VEJ.LEITIII cpvj..ai1110v J.L1JilEII inrep<f>povwv. IQ2

20 NOTES AND STUDIES 6s IDoL Kap.ov, IDoL wvavto. 1 d)v/nryj~ ofl owpoook Ta. {Jow lcp' lavtii_) Kov 'ita (?ranspose lcp'. p. K.). ALop.~OELOS avaykyj lkt?js 1M7Aov 1r6oas xns. Avo?Js T1}v 6upav lkamrev (?) AUKOS 7r pl cflplup xopw (?) o-of)' V Ta ol'j()' ~Awwa. All he feaures which are found in he collecion of Zenobius reappear in fac in he Greek book of Proverbs. The ranslaor employs he wo meres which by long radiion had been considered appropriae for hese homely maxims. 2 He shews he same parialiy as he old proverb-wriers for half-lines, beginning or ending a he caesura, and in paricular for merical endings ; he same disregard for nice disincions beween long and shor vowels. On he whole he uses he iambic mere slighly more ofen han he hexameer. We may be cerain ha he was quie familiar wih many of he old secular proverbs ; i is sufficien o quoe vii 2 2 Kvrov 1rl 8Hrp.ovr;/ xxiii 3 r yvp.v6t por; il'li"epov.' As we have seen, he probably produced his version in he firs cenury B. c. and in he firs half of he cenury, abou he ime when Tarrhaeus of Cree and Didymus of Alexandria were engaged in puing ogeher heir collecions of proverbs, upon which Zenobius subsequenly.drew." If we had any reason o doub ha Alexandria was his home, we could infer ha he was a ciy-dweller from he fac ha, in common wih he ranslaor of he laer half of Jeremiah, a ' neighbour' (lr.\) is for him a 7rcAlrYJs, a 'fellow-ciillen '. 6 The parial versificaion pervading he Greek version serves a pracical purpose of some imporance in exual criicism, hough is uiliy in his respec is limied by he fac ha i is only parial. Absence of mere is obviously no infallible crierion for deecing laer inerpolaions and orrecions in a work in which he prosaic elemen predominaed from he firs. Bu (I will conclude by aemping o sum up some of he uses o which his merical es may legiimaely be pu) : ( 1) absence l Cf. Jo. iv 38 laj..oc K<KOJT<aKaaw KTll.., following close upon wo oher proverbs, 35 TETpipqvos iiiti xc:. fj<pllf!os lpx<ta, and 37 MAOS ~11T1v <) 117r lpojv Ka1 MAOS 0 fj<picojv = laao p~v 111r ipovuv, llaao BE &.p-ijaovta Diogen. ii. 62 (Par. Gr. i p. 205). This lile group of hree Greek proverbs aribued o Chris wihin he compass of four verses is curious. 11 An occasional anapaesic line like viii 15 a,' ~poii {3aaAEs {3aaJ..Evovacv may be accidenal ; anoher rendering would hardly be possible. Cf. KVOJV l1rl B<apa, Zenob. iv 73 "f"fi"otepovs ll1rlpov J\6-yo~ 1raJ..auSs <f>tj11; Eusah. (ap. Parvem. Gr. i p. us) ; cf. "f"fivot<pos AE/31JpiBos ec. Zenob. ii 95 5 Paroem. Graeci i xii If. 6 In xi 9, 12, xxiv 43; bu elsewhere in Prov. <f>[j..o>. VOL. XIII. F

21 66 THE JOURNAL OF THEOLOGICAL STUDIES of mere does become a crierion where a choice has o be made beween wo 'doubles', one of which has a merical ring and he oher has no. Thus, in i 14, he firs of he doubles, Kowov 8~ {3aA>..rivTLov Krrw 6Jp. 8a 'lrrivus-, is, on accoun of he hexameer ending, besides oher reasons, o be preferred o he second, Kal p.apu{'lr'lrwv ~v " VYJ8frrw ~p:iv; we should perhaps go furher and adop he order of words in cod. V, which gives he senence an iambic opening as well, {3a'A'Aavrwv 8~ Kowov KTA. For, (2) where here is a diversiy of readings affecing he order of words, ha reading which produces rhyhm, especially a rhyhmical conclusion o a senence or sense-line, is o be preferred o a varian which lacks rhyhm and places he words in heir simples prosaic order or in he order in which hey sand in he Hebrew original. (3) Where here is no varian reading, bu he language is poeical and ransposiion of words prosaically arranged will produce a complee or fragmenary verse, ransposiion is generally legiimae, nowihsanding he lack of MS auhoriy. (4) Where a hexameer or iambic ending occurs near lze dose of a senence or sense-line, here is some ground for suspecing ha any appended unmerical words are an inerpolaion. Judging by merical ess, I should infer ha he minuscules 109, 147, 157 possess a high value in his book; he group 68, 161 &c., and cod. V are also imporan. Occasionally he original ex seems o be preserved in he Armenian Version; i is perhaps significan ha Proverbs was he firs book of he Greek Bible o be ranslaed ino ha language. 1 H. ST J. THACKERAV. OF words in he Old Tesamen which have apparenly a fixed and seled meaning, one which occurs prey frequenly is he word l"~.,, generally rendered by 'palace' or 'casle'. In he Auhorized Version i is ranslaed' palace' in 31 places ou of 32 in which i occurs(' casle' once only). And he Revised Version is almos equally uniform, giving 'palace' 28 imes and 'casle' 4 imes(wih margin 'palace' 3 imes). Bu here is no like uniformiy in he mos ancien versions. The renderings of he LXX comprise 8 p.l'awv (10 imes), xwpa (6), {3ap> (5), /J.p.cpo8ov ( 2 ), wih o!ko>, 'lr6a>, y11, 'Aa6s-, /J.vrpov, 7rVpy6{3ap>, and lvavr{ov once each, while in wo passages here is no word which cerainly answers o i. In he Vulgae here is no quie he same variey. Turris occurs 4 imes, emplum once, urbes once; more ofen i falls 1 Swee lnrod. o O.T. lis.