Non-Roman fonts used: SPIonic, SPTiberian
1. The translator(s)2 of the Septuagint of Proverbs had a unique approach towards their parent text. Their method of dealing with the Vorlage is apparent on both a micro-level and a macro-level. As far as the first goes, some individual lexical items are rendered consistently, while many are varied. The translation-technical approach can be described as one of both diversity and unity. This method of translation does not prevent the translator from interpreting in individual instances, even where a specific Hebrew word is normally translated consistently with a single Greek word. On the macro-level the Greek translation of Proverbs also exhibits unique features. The order of some chapters towards the end of the book that differ from MT (and other versions) should be ascribed to its translator(s), as should the removal of the names of Agur and Lemuel, who are mentioned in the Hebrew text as authors of some of the material. In the final analysis, this interpretative approach can be ascribed to the translator's ideology, which is characterised by a fundamentally conservative Jewish religious attitude. This inference is based, inter alia, upon the prominent role of the law of Moses in LXX Proverbs. I commence by dealing with aspects of the translation technique followed by the translator on the level of the lexical item.
2. The translator of Proverbs was clearly a stylist with an exceptional knowledge of Jewish and Greek culture, as well as an extensive vocabulary. In comparison to the rest of the LXX, he makes ample use of hapax legomena, words which he borrowed from the Greek world. There are a number of neologisms as well. Moreover, he uses a rather large number of lexemes in ways not typical in the rest of the LXX. As might be expected, he interprets extensively in some instances.
3. Definitions of hapax legomenon vary. Technically a hapax legomenon constitutes a word that occurs only once in any given corpus, in this case the whole of the LXX. Wagner differentiates between this technical definition and what he calls LXXhapax legomena. He expresses his position as follows: "Entscheidend für die Erhebung eines LXXHplg ist nicht die Einmaligkeit eines Wortes innerhalb der Septuaginta (einziger Beleg), sondern ausschliesslich die Zugehörigkeit zu einem LXX-Buch. Von daher kann ein LXXHplg durchaus öfter (in einer Schrift) vorkommen" (Wagner 1999: 86). In this contribution I use the narrower definition of a hapax legomenon as a Greek word that is used only once in the LXX. One of the main reasons that I opt for this definition is that the OG of this translation unit has not been determined (cf. section 3, below). Consequently, it is appropriate, for the interim at least, to be conservative in this regard. That this is a rather difficult position will soon become clear.
4. I found the following list of hapaxes in LXX Proverbs3 (following the definition is a list of the word's occurrences in classical literature; abbreviations are from LSJ and Berkowitz and Squitier 1990; the plus sign '+' indicates that the Greek word has no Hebrew equivalent):
5. According to this list there are a total of 160 hapax legomena in the Greek version of Proverbs, including examples quoted by Hatch and Redpath (HR). These examples need to be individually evaluated, because some are problematic. For example, in Prov 30:16 HR refer to the reading a[|rhj that appears only in ms A. The other mss all read a[|dhj. That a reading error could have taken place is possible. However, there are many examples of a[|rhj in classical Greek literature. In Ar. Fr. 558, for example, the word refers to the god of destruction. It is also possible that some scribe interchanged these readings. The god of destruction does after all fit into the picture of this verse where Tartaros is mentioned in the LXX. Nevertheless, it should be removed from the list, since it is very unlikely to be the original LXX reading. The fact that only one ms testifies to this reading suggests that a later scribe is actually responsible for it.
6. Other similar examples illustrate the critical importance of scrutinizing the manuscript evidence4 before drawing conclusions. As far as e0kbu/zein (Prov 3:10) "to gush out", goes, HR refer to the original hand of Sinaiticus. Another case is fluari/a, which according to HR occurs in Prov 23:29 in ms S6. I could find no evidence of this ms. It seems that they either used an unknown ms here or made a mistake.
7. There are a number of other Greek words that appear only in a small number of mss in the LXX. For example, a)ndrogu/naioj is the generally attested reading in Prov 19:15. However, the original scribe of S also used the word in Prov 18:8. Thus, there are really two examples of this adjective in LXX witnesses. There is clearly some textual confusion between this word and a)ndro/gunoj, which also occurs solely in Prov 18:8 and 19:15, in other witnesses. However, whereas a)ndrogu/naioj is used only in these LXX passages and in Athanasius 109 [De sacta trinitate] 28.1213.33; 109 28.1213.38, a)ndro/gunoj, to the contrary, appears widespread in Greek sources (Pythag. 007 [Fragmenta astrologica] 11.2.142.14, 16; Hdt. 4.67.7; Pl. Smp. 189.e.2). (A variant form, a)ndrogu/nia, occurs in Iambl. Theol.Ar. 41.14.)
8. This is a difficult example when one operates with a minimalist definition of hapax legomenon as I do. It seems as if scribes were responsible for the textual confusion in these two verses. The fact that Athanasius uses the word a)ndrogu/naioj in the 3rd century CE suggests that he probably read it in his copy of Proverbs, so the word should be treated as a hapax, even though it is attested in only a few witnesses. Because scribal activity is probably the cause of the textual confusion in these verses, I think that a)ndro/gunoj should be treated as a hapax legomenon as well, as LEH correctly suggests.
9. a)pobia/zomai is another example of textual uncertainty related to hapaxes. The word occurs in Prov 22:22 in Rahlfs and HP. However, it is attested to also in Prov 28:24 in mss A, 23, 68, 106, (161 marg), 248, 253, 260, 261, 296, and 297. These are important mss and cannot be discarded. This verb clearly appears in two contexts in significant LXX witnesses. The question remains whether it should be taken as a hapax.
10. e0mbiba/zw also occurs in two passages--Prov 4:11 and 4 Kgdms 9:28--in the latter of which the reading is attested only in ms A. Again this is a rather important manuscript.
11. The same could be argued in connection with la/qrioj, *a, *on, which is used in Prov 21:14 and in one ms, Bc, in Wis 1:11. However, the fact that a corrector was at work here could tip the scales in the favour of a decision to take it as a hapax.
12. o)cu/qumoj, *on too is used in two passages: Prov 14:17 and 26:20. In the latter case, although B S read di/qumoj, mss A, S2, 109, 147, 157, 159, 252, 254, 260, 297--all significant witnesses--read o)cu/qumoj.
13. A special case are the words straggaliw/dhj and straggalw/dhj. They both appear in Prov 8:8, with the same connotation ("knotted"), but in different mss. The first is used in ms S2. straggalw/dhj appears in mss A, B, S*, III, 109, 157, 253, 254, 260, and Ald. Both words are found only in later works, inter alia, straggalw/dhj in Origenes 019 [Philocalia] 6.1.2; 019 6.1.4 and straggaliw/dhj in Origenes 031 [Fragmenta ex commentariis in evangelium Matthaei] 3.2, 2. Again scribal activity could be the reason for these orthographical differences.
14. Manuscript evidence is also determinative for the apparent hapax stomi/j (Prov 30:14). It appears only in mss A and S, whereas ms B reads tomi/j "knife". The meaning "hard-mouthed horse" (see Poll. 10.56) does not make sense in the current context and could be the result of some scribal activity.
15. Several readings occur in only a small number of manuscripts. a!sofoj (Prov 9:8 "unwise") appears only in a few LXX mss (161 marg; 248, 252, 253). kro/kinoj (Prov 7:17) is present only in mss B, R, S*. kako/thj (Prov 24:19 "evildoer") is another applicable example. It appears only in this one passage, and moreover it is found in only one LXX ms, namely S*. The same is true of pelidno/j, *h/, *o/n, too.
16. Another uncertain case is summe/nw A, Alex., ms 252 sub * (S2 summei/gnumi) Prov 20:1 "hold together" (Arist. Metaph. 1077a24; Hdt.1.74). Similarly, summeri/zomai Prov 29:24 appears only in S2, mss 149, 260, 297; and u(peu/qunew Prov 1:23 appears only in ms S*. It is interesting that u(peu/qunoj, which also appears in this verse, is likewise a hapax! However, several LXX mss, as well as other Greek literature (Hdt. 3.80; A. Pr. 326) employ this word.
17. The crucial issue at stake in all these deliberations is the question of what weight should be allotted to individual mss. This question is particularly difficult in regard to LXX Proverbs, since a critical edition has not yet been published in the Göttingen series. In the light of the discussion above, therefore, I prefer to stay on the conservative side and argue that a[|rhj, e0kbu/zein, e0mbiba/zw, stomi/j, and fluari/a should be discarded from consideration as hapaxes. The verbal adjective h(ghte/on is a borderline case, since it is uncertain whether such forms should be taken as hapaxes. These adjustments bring the total to 155 hapax legomena for LXX Proverbs.
18. This total differs from that suggested by Wagner (1997:5), but as I stated earlier, the discrepancy is the result of us working with different definitions. The subjectivity of my position becomes clear from my treatment of the verbal form e0neufrai/nomai (Prov 8:31 "to rejoice"). It is used twice in this one verse, and these are the sole examples in the LXX. Surely the word should be taken as a hapax legomenon. I would therefore have to concede that the rule of thumb of one occurrence in the Septuagint as a whole can only act as a general guideline. Many of the cases which I discussed above could certainly be considered hapax legomena. I, nevertheless, prefer to remain on the conservative side of this issue, dealing with each instance individually, as I demonstrated above.
19. This rather large number of hapaxes (and neologisms) is striking. Even though one has to be wary of overestimating the value of this linguistic phenomenon as an indicator of translation technique (Wagner 1999: 6), it is nevertheless significant that a high percentage of hapax legomena (and neologisms) is found in those LXX books that are also rendered rather freely. According to Wagner's analysis (Wagner 1999: 5), Proverbs and Job have relatively high numbers of hapaxes. The same is true as well of some other wisdom books such as the Wisdom of Solomon and Ben Sira. Moreover, according to Wagner's calculations, 2 Maccabees has 416, 3 Maccabees 205, and 4 Maccabees 396 occurrences of hapaxes.
20. The individual hapaxes in LXX Proverbs make interesting reading in and of themselves. By far the majority of these hapaxes are found as well in Classical Greek writings. However, a number of these words are used exclusively in the LXX or related later literature (glwssoxarite/w, e0gkloio/omai, nohtw/j, nomoqe/smwj, tomi/j, u(peu/qunew). The fact that the translator(s) used ai)poli/on to describe a flock of goats in Prov 30:31 is perhaps not that important; after all, he might not have had another word at hand. However, this Greek word appears widely in Classical Greek sources, inter alia, in Homer Il. 11.679; in Hdt. 1.126, S. Aj. 375, and 10 times in Philo Judaeus. This word appears not only in earlier Greek writings but also in later writings such as the New Testament and the patristic writers, for example, in Matt 26:24, in Origen, and even later in John Chrysostom.
21. The fact that this specific verse in LXX Proverbs contains two hapax legomena, the second being a)le/ktwr, is also conspicuous. Prov 30:31 is an interpretative passage:
Prov 30:31#$yItf-wO) MyInat;mf ryzir:zA
wOm@(i Mw@ql;)a K7lemew@the strutting cock, the he-goat,
and a king striding before his people.
kai\ a)le/ktwr e0mperipatw~n qhlei/aij eu1yuxoj
kai\ tra&goj h9gou/menoj ai0poli/ou
kai\ basileu\j dhmhgorw~n e0n e1qneiAlso a cock strutting courageously among the hens;5
and the he-goat leading the herd;
as well as a king addressing the nation.
22. This verse in the LXX contains three strophes, whereas MT has only two. Lagarde's rules of thumb (1863: 3) are not helpful in this instance. Those responsible for this translation clearly interpreted their Hebrew Vorlage, elaborating on the cock, the he-goat, and the king. I do not think that there existed a deviating Hebrew parent text; this is the translator at his best, interpreting! In my view this rendering demonstrates the competence in Greek literature that the translator had. He embroiders within the framework of the text in a creative manner. Strutting courageously among the hens is typical of a cock, as is leading a herd for a he-goat and addressing a nation for a king. The Hebrew of these phrases does not help us to evaluate the Greek. ryzir:zA is a hapax legomenon, and ai0poli/ou has no Hebrew counterpart. In both cases the translator used words from Greek culture known to him. The least we can say is that he was steeped in Greek culture. This fact is borne out by the large number of neologisms in LXX Proverbs.
23. Neologisms are newly formed Greek words that appear for the first time in the Septuagint and often appear as well in the literature based upon it.6 To determine which words are actually neologisms is a rather difficult task. Dating naturally plays an important role in defining such words.7 In drawing up my list of neologisms, I first compiled a list from LEH and HR. According to Lust (LEH: v), the abbreviation "neol?" indicates that, while the word is found in later Greek secular writers (i.e., those unlikely to have been dependent on readings of the LXX), it does not occur before Polybius in the 2nd century BCE. I have challenged LEH's conclusion with regard to several words. Next, I worked through the text itself. As a result of my investigation, I have compiled the following list of neologisms. Words that are also hapaxes in the LXX are indicated by "(hapax)".
24. As would be expected, a number of these 74 neologisms are at the same time also hapax legomena. In most cases the translator used classical Greek words that already existed; however, there are also examples that occur only in the Septuagint and in the literature based upon it. A number of words indicated by "neol?" in LEH actually pre-date Polybius in the 2nd century BCE and so cannot be considered neologisms.
25. When dealing with neologisms, it is necessary to take into account manuscript evidence. a)galli/ama is used in a number of books in the Septuagint, but in Proverbs it appears only in 11:10, and for that matter only in mss 23, 103, 295, 296, 297, A, B, S2. Thus, it cannot be considered without qualification as a neologism of Proverbs. This applies to a)khli/dwtoj, *on; a)kribasmo/j and makru/nw as well.
26. Nevertheless, this large number of neologisms, together with the hapax legomena, witness to the proficiency of the person(s) responsible for the translation of Proverbs. This is underscored by the way specific individual lexical items are rendered.
27. In this section I will deal with a selection of lexical items in order to demonstrate the varied approach of the translator (cf. my forthcoming contribution to the LXX congress that took place in Helsinki in 1999: Cook 2001).
28. A number of Greek words render more than one Hebrew word. parano/moj, for example, is used 73 times in the LXX, and of these 25 appear in Proverbs. The following distribution obtains: 1:18 (+); 2:22 (dgb); 3:32 (zwl); 4:14 ((r); 4:17 (smx); 6:12 (Nw)); 10:5 (#$wb); 11:6 (dgb); 11:30 (+); 12:2 (hmzm); 13:2 (dgb); 14:9 (+); 16:29 (smx); 17:4 (Nw)); 19:11 ((#$p); 21:24 (+); 22:12 (dgb); 22:14 (rwz); 23:28 (dgb); 25:19 (dgb?); 26:3 (lysk); 28:17 (+); 29:4 (+), 12 ((#$r); 29:18 (+). The translator(s) translated 11 different Hebrew words with this one Greek word!
29. The distribution of e!nnoia is as follows: 1:4 (hmzm); 2:11 (hnwbt); 3:21 (hmzm); 4:1 (hnyb); 5:2 (hmzm); 8:12 (hmzm); 16:22 (lk#&); 18:15 (t(d); 19:7 (+); 23:4 (hnyb); 23:19 (Krd); 24:7 (+). Six different Hebrew words are rendered by means of this lexeme.
30. skolio/j is used in Prov 2:15 (#$q(); 4:24 (tw#$q(); 8:8 (ltp); 16:26 (+), 28 (hkpht); 21:8 (+, and Kpkph); 22:5 (#$q() and 14 (+); 23:33 (hkpht); 28:18 (#$q(). All these Hebrew lexemes express some aspect of "being crooked". Of these 11 examples, 4 render #$q( and 3 hkpht, with 3 textual uncertainties. This distribution is less diverse than the previous examples; however, it gives an indication of how the translator(s) varied their approach.
31. Of the 50 cases in LXX Proverbs of sofi/a, it renders hmkx in 37 instances. This ranges on stereotyping, especially if one takes into account that in 6 cases the Greek has no Hebrew equivalent text. The only other Hebrew readings to which this Greek word attests in Proverbs are hnwbt (1:7, 29), hnyb (2:3; 3:5; 18:2) and rswm (8:33). Similarly, hmkx has been rendered fairly consistently by this Greek translator. It occurs 25 times in Proverbs, and in the large majority of the cases sofi/a is used as equivalent (1:7; 2:2, 6, 10; 3:13, 19; 4:5, 11; 5:1; 7:4; 8:1, 11, 12; 9:1, 10; 10:31; 11:2; 14:6, 8; 15:33; 16:16; 17:16; 21:30; 24:3, 7, 14; 28:26; 29:3, 15; 30:3). Other Greek equivalents are gnw/sij (2:6); e0ntolh/ (4:5); phgh\ zwh=j (18:4), and prose/xw (31:26[25]). In Prov 4:7 and 23:23 Rahlfs has no text.
32. fro/nhsij occurs 15 times in Proverbs primarily in connection with the root Nyb, either as hnyb (as in this verse) or hnwbt. In Prov 9:16, however, it acts as the equivalent for bl-rsx, expressing the nuance of "(those without) common sense", but in combination with e0ndee/si. In Prov 24:5 it attests to t(d.
33. The lexeme paidei/a is another typical sapiental concept. It is used 28 times in Proverbs. In practically all passages in Proverbs it has rswm as the underlying Hebrew reading. An exception is 1:29, where this Greek lexeme appears in mss 296, 311, 733, and codex A with the underlying Hebrew t(d. In Prov 10:17 it probably attests to txkwt.
34. e0ntolh/ occurs 9 times in this book: in Prov 2:1 (hwcm); 4:4(5) (hwcm); 6:23 (hwcm); 7:1 (hwcm); 7:2 (hwcm); 10:8 (hwcm); 13:13 (hwcm); 15:5 (txkt); and 19:16 (hwcm). It is rendered stereotypically.
35. a!frwn appears abundantly (75x) in Proverbs, and in 40 cases it attests to the Hebrew noun lysk, slightly more than 50% of the examples. This is rather consistent, even though other Hebrew words--Cwl (1x), l(ylb (3x), ytp (6x), lyw)/tlyw) (13x), r(b (2x), lc( (1x), bl-rsx (1x), and lbn (2x)--are also rendered in the process.
36. Not all the words used rather consistently by the translator(s) are stereotypes. In a number of instances the translator tends to associate one Greek lexeme with one Greek lexeme, but he sometimes prefers other words when the needs of the translation demand it. One case in point is di/kaioj which appears 435 times in the Septuagint, of which 105 examples come from Proverbs. The following distribution obtains:
37. Thus out of 105 occurrences of this lexeme it is used 66 times as equivalent for the root (qdc). There is a certain amount of consistency; however, he is by no means stereotyping and in some instances is actually interpreting.
38. A similar pattern is found in regard to the noun dikaiosu/nh, which according to LEH occurs 351 times in the LXX and 36 times in Proverbs. In by far the largest number of cases (24x) in Proverbs it is rendered by means of the root (qdc). However, there are a number of textual uncertainties--which are primarily pluses compared to MT--whilst the roots ytp (1:22), +p#$m (8:20, 16:11; 17:23), Nwdm (17:14), dsx (20:28), and lk#& are also rendered.
39. a)sebh/j occurs 90 times in Proverbs, in most cases as the translation for (#$r, but in chapter 1 it is used as an equivalent for three Hebrew lexemes: Mylyw) (1:7), )+x (1:10), and lysk (1:22, 32). In this chapter the translator is interpreting the subject matter in religious terms (Cook 1997a: 110).
40. The variety of Greek lexemes used to express "ways" is interesting. According to LEH the noun o(do/j occurs 891 times in the LXX, of which about 100 examples are found in Proverbs. By far the majority of these occurrences attest to the Hebrew noun Krd. tri/boj occurs 70 times in the LXX, 11 of which are in Proverbs. These 11 instances of the word are used to render 4 different Hebrew words: 1:15 (bytn); 2:15, 19 (xr)), 20 (2x) (xr) and Krd); 3:17 (bytn); 8:2, 20 (bytn); 15:21 (+); 16:17 (hlsm); 30:19 (Krd). From the Greek perspective, o(do/j is used fairly consistently to render Krd. How complicated the picture really is, however, becomes clear when one looks at it from the perspective of the Hebrew. In Prov 30:19 the noun Krd appears 4 times, and the LXX uses 3 different words (i1xnoj, o(do/j [2x], tri/boj) to render the same Hebrew word. This is indeed an example of unity and diversity.
41. Another word that seems to be used relatively consistently is the adjective yeu/dhj, which is used in the LXX 109 times. In Proverbs it appears in 6:19 (bzk); 8:7 (+); 12:22 (rq#$); 14:5 (bzk); 14:25 (bzk); 17:4 (+), 7 (rq#$); 19:5, 9 (rq#$); 19:22 (bzk); 21:6 (rq#$); 21:28 (rq#$); 23:3 (bzk); 24:2 (+), 28 (+); 30:6 (bzk Ni); 30:8 (bzk); 30:9 (#$xb); 25:14 (rq#$); 25:18 (rq#$); 26:28 (rq#$); 28:6 (+); 31:30 (rq#$). It thus attests 9 times to rq#$, 7 times to bzk, once to #$xb, and there are 5 textual uncertainties which seem to be pluses compared to MT. The addition of yeudh/ to Prov 8:7--the same phrase occurs in 17:4, too--is the result of a moralizing tendency related in the translation. In Prov 17:4 the moralizing is expressed by means of a contrast (transgressors versus a righteous person). In Prov 24:2 the Greek reads "falsehood" for "violence," and in verse 28 the adjective yeudh\j is added to ma/rtuj. This adjective is also used in Prov 28:6 without a direct antecedent in the Hebrew.
42. a)lh/qeia in 8:7; 11:18; 14:22; 20:28; 22:21; and 29:14 all correspond to tm) in the MT. Two examples differ: 26:28 (+) and 28:6 (Mt). Dick (1990: 23) argues that the LXX Proverbs has a tendency to add some roots, such as a)lh/q* and yeud*, in specific instances. His argument is somewhat weak, for there is only one example of an addition of a)lh/q*, namely in 26:28. He is nevertheless correct in arguing that it is the result of an antithesis (glw/ssa yeudh\j and a)lh/qeia).
43. a!nomoj occurs in 1:19 ((cb); 10:2 ((#$r); 12:3 ((#$r); 14:16 (+); 21:18 ((#$r); 27:21 (+); 28:10 (+); 29:8 (Nwcl); 29:27 ((#$r). According to LEH this adjective is used 106 times in the LXX. Out of 9 occurrences in Proverbs, 4 rendered (#$r. Although this number seems somewhat insignificant, only two other Hebrew words are so rendered (once each in 1:19 and 29:8), whereas 3 cases are textual uncertainties.
44. It is clear that the translator had a preference for the lexeme kak*. Of the 384 examples in the LXX, 87 are found in Proverbs. Of these the largest number of cases (40), just under 50%, act as equivalent for h(r. This rather large number of equivalent renderings stresses the aspect of consistency in translation. However, the additions in comparison to MT make for interesting reading. In Prov 1:18 two added examples occur. This verse is part of an intentional endeavour by the translator to picture the evil realm in chapters 1 and 2 (Cook 1997a: 407). The noun kakoi/ in verse 28 is another example, as is the adjective kakh\ in the phrase kakh\ boulh\ in Prov 2:17, which reveal the same intention on the part of the translator. Also in Prov 6:3 the translator interprets in a rather free manner, for the friend that the pupil was warned against in verse 1 in the LXX has now become a stumbling block. In the process K1(er" is interpreted as h(frF (evil). This is the sole example in Proverbs where this lexeme is interpreted in this way. The rendering is probably a reference to verse 1, where the friend is called an enemy (see also the connections to verse 2).
45. Prov 6:11 in the LXX has a number of additions. In one case the adjective kako\j appears. The negative nuance expressed by kako\j o(doipo\roj in verse 11 of the LXX is not part of the MT. The verb Klh does not have such a sense as part of its semantic field, even though the RSV interprets the piel participle as "vagabond". The same applies also to the positive nuance expressed in the reference to the "excellent" runner (a)gaqo\j dromeu/j). It is clear what has happened in these instances: the translator has rendered his parent text in terms of the dualism of good versus bad. It does seem rather strange to render the want that will overcome the sluggard as an excellent runner. However, the Greek translator probably related the noun to the verb Ngm "to deliver". This does not, of course, provide an explanation for the adjective. I think the translator simply contrasted the two categories by the addition of appropriate adjectives as he often does. I interpret the additions to Prov 9:12 in a similar manner (Cook 1997a: 266f).
46. The nuanced addition of kaka/ in connection with Prov 13:10 is yet another example of the way this translator goes to work. The evil and the wise are contrasted, and strife (hc=fma) is interpreted as evil. This is actually the sole example of this Hebrew noun being related to kako/j in Proverbs.
47. It is rather difficult to relate the Greek version of Prov 15:15 with the Hebrew. The added noun kaka/ can, nevertheless, be understood in terms of the tendency which I have illustrated above. This also applies to the interpretation of #$y) as an evil person, o( kako\j, in verse 23. An interpretation similar to that found in Prov 13:10, where a person causing strife is called evil, occurs also in Prov 16:28. The translator has also added a phrase to stress the idea of wickedness. Verse 30 also has a similar addition (he is a furnace of wickedness). It may nevertheless be the case that these two pluses are the result of later hexaplaric activity.
48. Clearly some interpretation also took place in the rendering of Prov 18:3:
zw%b@-MgA )b@f (#$frF-)wObb@;
hp%fr:xe NwOlqf-M(iw:When wickedness comes, contempt comes also;
and with dishonor comes disgrace.o[tan e!lqh| a)sebh\j ei)j ba/qoj kakw/n katafronei/
e0pe/rxetai de\ au)tw/| a)timi/a kai\ o!neidojWhen the impious comes into a depth of evils, he shows contempt;
but disgrace and reproach come upon him.
49. The phrasing of Prov 19:6 is partly based upon the interpretation of (Ar"hf (friend) as coming from the root (r (evil). However, the second strophe is another example of how this translator "drives the point home".
bydInF-yn"p; w%l@xay; Myb@irA
Nt@fma #$y)il; (Ar"hf-lkfw:Many seek the favor of the generous,
and every one is a friend to a giver of gifts.polloi\ qerapeu/ousin pro/swpa basile/wn
pa/j de\ o( kako\j gi/netai o!neidoj a)ndri/Many court the favour of kings;
but every bad person becomes a reproach to another.
50. The same interpretation is seen in 19:27. The contents of this verse are interpreted in terms of evil ideas.
rsfw%m (Amo#$;li ynIb@;-ldAxa
t(adF-yr"m;)im' twOg#$;liCease straying, my child, from the words of knowledge,
in order that you may hear instruction.ui(o\j a)poleipo/menoj fula/cai paidei/an patro\j
meleth/sei r(h/seij kaka/jThe son who stops keeping the instruction of his father,
will contemplate evil ideas.
51. In Prov 21:26 the contrast between the impious and the righteous is more evident in the Greek than in the Hebrew.
hwF)jta hw%F)at;hi MwOy,ha-lk@f
K7#&ox;yA )low: Nt@'yI qyd@Icaw:All day long the wicked covet,
but the righteous give and do not hold back.a)sebh\j e0piqumei/ o[lhn th\n h(me/ran e0piqumi/aj kaka/j
o( de\ di/kaioj e0lea/| kai\ oi)kti/rei a)feidw/jThe impious reflects over wicked desires all day long;
but the righteous unsparingly compassion and pity.
52. The additions in Prov 22:14 could also be attributed to the translator, assuming that the pluses are not the result of later scribal activity. The interpretation of "the mouth of the lawless" instead of "a loose woman" is probably based upon the understanding of twrz coming from rwz and not from hrz.
twOrzF yp@i hq@fmu(j hxfw%#$
M#$f-lp@fyF hwFhy: Mw%(z: (Q; K:lwpy)The mouth of a loose woman is a deep pit;
he with whom the Lord is angry will fall into it.bo/qroj baqu\j sto/ma parano/mou
o( de\ mishqei\j u(po\ kuri/ou e0mpesei/tai ei)j au)to/n
14a ei)si\n o(doi\ kakai\ e0nw/pion a)ndro/j
kai\ ou)k a)gapa/| tou/ a)postre/yai a)p' au)tw/n
a)postre/fein de\ dei/ a)po\ o(dou/ skolia/j kai\ kakh/jThe mouth of the lawless is a deep pit;
and he who is hated by the Lord will fall in it.
14a There are bad ways before a man,
and he prefers not to avoid them;
even though he should avoid a crooked and bad way.
53. The interpretation of the Hebrew noun rwOd@ as a wicked progeny (e!kgonon kako\n) appears in Prov 30:11-14 and in my view is to be ascribed to the translator. Verse 11 (MT) reads as follows:
ll@'qay: wybi)f rwOd@
K7r"bfy: )lo wOm@)i-t)ew:There are those who curse their fathers
and do not bless their mothers.e!kgonon kako\n pate/ra katara/tai
th\n de\ mhte/ra ou)k eu)logei/A wicked progeny curse their father
and do not bless their mother.
54. Prov 25:19 can only be understood as the interpretive activity of the translator.
tdE(fw%m lgErEw: h(fro N#$'
hrFcf MwOyb@; dg"wOb@ x+fb;miTrust in a faithless man in time of trouble
is like a bad tooth or a foot that slips.o)dou\j kakou/ kai\ pou\j parano/mou
o)lei/tai e0n h(me/ra| kakh/|The ways of the evil and the foot of a transgressor
will be destroyed in a catastrophic day.
55. The additions to Prov 27:21 contain conspicuous contrasts and could therefore very well be the work of the original translator.
bhfz%Fla rw%kw: Psek@ela Pr"c;ma
wOllfhjma ypil; #$y)iw:The crucible is for silver, and the furnace is for gold,
so a person is tested by being praised.doki/mion a)rgu/rw| kai\ xrusw/| pu/rwsij
a)nh\r de\ dokima/zetai dia\ sto/matoj e0gkwmiazo/ntwn au)to/n
21a kardi/a a)no/mou e0kzhtei/ kaka/
kardi/a de\ eu)qh\j e0kzhtei/ gnw/sinBurning is a test for silver and gold;
but a person is tested by the mouth of them who praise him.
21a The heart of the lawless seeks evil;
but a righteous heart searches for knowledge.
56. Finally, the plus in Prov 28:20 seems to be in line with what I have demonstrated above. In the MT the faithful and the rich are contrasted:
twOkrFb@;-brA twOnw%m)v #$y)iwhereas in the Greek it is the faithful and the evil:
hqen%FyI )lo ry#$I(jhal; C)fw:A trustworthy man will have many blessings,
but the one who hurries after riches will not be unpunished.
a)nh\r a)cio/pistoj polla\ eu)loghqh/setai
o( de\ kako\j ou)k a)timw/rhtoj e!staiA trustworthy man will be greatly blessed, but the wicked will not be unpunished
57. The overwhelming impression that one gathers from this discussion, as well as in my monograph mentioned earlier (Cook 1997a), namely that the Septuagint Proverbs contains a rather large number of words and concepts that either occur only in this book or have been used in a restricted number of instances elsewhere in the LXX. Many of the words are encountered primarily in sapiential literature.
58. The distinctive approach of the translator is clearly observed in a final example taken from Prov 1:26-27. On the one hand, he renders dxp differently: in verse 26 with o!leqroj and in verse 27 with qo/ruboj. The same applies to dy), which in verse 26 has a)pw/leia and in verse 27 katastrofh\ as equivalents. On the other hand, he does in some instances use a single Greek lexeme in order to translate a number of semantically related Hebrew lexemes (e.g., )wb - erxomai). He is evidently a stylist, making abundant use of literary diversity and unity. Because of this nuanced approach it is extremely difficult to reconstruct the underlying Hebrew he used in each individual instance.
59. What does seem significant is that even in cases where the translator seems to be consistent, he still interprets. In the case of the root kak* I demonstrated that the translator actually added and interpreted extensively. It is often not easy to determine what has indeed happened, so each individual instance must be evaluated as objectively as possible. However, as a basic presupposition, one can accept as a working hypothesis, when encountering a deviation in comparison with MT or other prominent textual witnesses, that this translator would have been apt to interpret. The same seems to be even more true in the wake of the adaptations which the person(s) who were responsible for this unit brought about on a macro-level.
60. One of the characteristics of LXX Proverbs is the different order of some of its chapters vis-à-vis the MT and other textual witnesses. I have argued in many contexts that some of the differences in the order of chapters towards the end of LXX Proverbs (chapters 24-31) are the result of the translator interpreting his parent text (cf. Cook 1997a: 310-315; as well as my forthcoming contribution to the proceedings of IOSCS of Oslo, 1998 [Cook forthcoming]). A determinative feature of the way the Proverbs of Israel are described in the LXX is that all the proverbs are seen as originating exclusively from king Solomon, as is stated in the first verse of this collection. Consequently all references to other authors who wrote Proverbs, such as Agur (Prov 30:1 MT) and Lemuel (Prov 31:1, 4 MT) are simply erased. Also in Prov 10:1 the phrase hml#$ yl#$m is not rendered, for it was seen as a tautological statement by the translator(s).
61. The differences between the order of chapters in the MT and the LXX are described in the table below. The sections are ordered according to the LXX sequence (Tov 1992: 337).
| MT section | MT reference |
|---|---|
| I-III | 1:1-24:1-22 |
| VI, part 1 | 30:1-14 (The words of Agur, 1st part) |
| IV | 24:23-34 (Words of the wise) |
| VI, part 2 | 30:15-33 (The words of Agur, 2nd part) |
| VII | 31:1-9 (The words of Lemuel, 1st part) |
| V | 25-29 |
| VIII | 31:10-31 (The words of Lemuel, 2nd part) |
62. In an earlier publication, I demonstrated that the acrostic in Prov 31:10-32 was deliberately linked to Prov 29. The contrast between these passages can be seen in the discussion found in Cook 1997a: 310-315.
Prov 29:27lwE(f #$y)I Myqiyd@Ica tba(jwOt@
K7rEd@F-r#$ay: (#$frF tba(jwOtw:An unjust man is an abomination to the righteous,
but he whose way is straight is an abomination to the wicked.bde/lugma dikai/oij a)nh\r a!dikoj
bde/lugma de\ a)no/mw| kateuqu/nousa o(do/jAn unrighteous man is an abomination to the righteous,
and the direct way is an abomination to the wicked.
63. The contrast present in the Hebrew is also expressed in the LXX, a contrast that appears in Prov 31:10, too, and is consequently linked to this passage, Prov 29:27.
Prov 31:10There is clearly a relationship between this verse and the last verse in the previous chapter of the LXX (i.e., 29:27).)cfm;yI ymi lyIxa-t#$e)'
h@rFk;mi MynIynIp%;mi qxorFw:A good wife who can find?
She is far more precious than jewels.gunai/ka a)ndrei/an ti/j eu(rh/sei
timiwte/ra de/ e0stin li/qwn polutelw/n h( toiau/thWho shall find a virtuous woman?
For such a one is more valuable than precious stones.
64. Now if my arguments thus far hold water, then it would be logical to believe that the translator(s) actually followed some ideology. Surely a translator who approaches his parent text as creatively as was demonstrated above on all levels,8 must have had "ideological/theological" reasons to do so. However, before addressing the issue of ideology it would perhaps be wise first of all not to forget the problematic nature of this LXX text. As I stated at the beginning of this article, the Old Greek of Proverbs has not yet been determined.
65. That Proverbs LXX is riddled with textual problems is well-known. I will use chapter 20 as a demonstration (cf. Cook 2000: 163f):
66. LXX Proverbs has many more plusses than minuses; this is true of 20:9.
These additional phrases in the LXX may not be simple pluses, for they seem to be based on inner-textual corruptions.yb@ili ytiyk@izI rma)yo-ymi
yti)+%fxam' yt@ir:ha+fWho can say, "I have made my heart clean;
I am pure from my sin"?9 ti/j kauxh/setai a(gnh\n e!xein th\n kardi/an
h0 ti/j parrhsia/setai kaqaro\j ei=nai a)po\ a(martiw/n
9a kakologou/ntoj pate/ra h2 mhte/ra sbesqh/setai lampth/r
ai( de\ ko/rai tw/n o)fqalmw/n au)tou/ o!yontai sko/toj
9b meri\j e0pispoudazome/nh e0n prw/toij
e0n toi/j teleutai/oij ou)k eu)loghqh/setai
9c mh\ ei1ph|j tei/somai to\n e0xqro/n
a)lla\ u(po/meinon to\n ku/rion i3na soi bohqh/sh|Who can boast that he has a holy heart?
or who can declare confidently that he is pure from sin?
9a Whoever curses father or mother let his lamp be extinguished
and his eyeballs see darkness.
9b a portion quickly acquired in the beginning
will not be blessed in the end.
9c Do not say: I will repay hatred;
but wait for the Lord that he can save you.
67. When comparing the LXX with the MT, two aspects of Prov 20 are conspicuous. First, the LXX, according to Rahlfs, has no equivalent for Prov 20:14-22. Second, Prov 20:9a-c corresponds to MT verses 20-22.
Prov 20:20
wOm@)iw: wybi)f ll@'qam;
K7#$exo Nw%#$y)vb@e wOrn" K7(ad:yiIf one curses his father or his mother,
his lamp will be put out in utter darkness.20:9a LXX
9a kakologou/ntoj pate/ra h2 mhte/ra sbesqh/setai lampth/r
ai( de\ ko/rai tw/n o)fqalmw/n au)tou/ o!yontai sko/tojWhoever curses father or mother let his lamp be extinguished
and his eyeballs see darkness.
Prov 20:21
hnF#$o)rIb@f tlexebom; htfxjnA
K7rFbot; )lo h@tfyrIxj)aw:An estate quickly acquired in the beginning
will not be blessed in the end.20:9b LXX
meri\j e0pispoudazome/nh e0n prw/toij
e0n toi/j teleutai/oij ou)k eu)loghqh/setaia portion quickly acquired in the beginning
will not be blessed in the end.
Prov 20:22
(rF-hmfl@;#$a)j rma)t@o-l)a
K7lf (#$ayOw: hwFhyla hw%"qaDo not say, "I will repay evil";
wait for the Lord and he will help you.20:9c LXX
mh\ ei1ph|j tei/somai to\n e0xqro/n
a)lla\ u(po/meinon to\n ku/rion i3na soi bohqh/sh|Do not say: I will repay hatred;
but wait for the Lord that he can save you.
68. Surely these are serious textual problems that will have to be addressed before the Rahlfs pocket edition can be used fruitfully. I have, nevertheless, endeavoured to construct the OG for those passages that I have analysed. This naturally applies to the final issue that I want to discuss, the "theology/ideology" of this translator (Cook 1997a and 1999).
69. I have argued extensively elsewhere that the background to the "ideology" of this translation unit should not be looked for in the idea world of Hellenism, be it in the thought world of the Stoics (Gerleman 1950, 1956), or of Plato (Hengel 1973, Deist 1988), or of the mystery religions (Sandelin 1986). I am of the opinion that the translator(s) of this book were conservative Jewish scribes (Cook 1997a: 318f). Dick seems to have a similar point of view when he refers to "a more conservative Greek-speaking Jewish school perhaps resident in Palestine" (Dick 1990: 20). I also agree with many of Dick's other conclusions. As he (cf. also Gammie 1987), I have found no evidence in LXX Proverbs of Stoic influence (Dick 1990: 49).9 As I demonstrated above the person(s) responsible for this unit clearly moralized to a great extent. The question to be answered is what the content of this moralizing indeed was. As I suggested in the beginning the key to this issue is found in the position and role of the torah. This is also where I have to take issue with Dick, who, for example, takes the addition to Prov 13:5 as a reference "to general sapiental teaching and not the specific 'Law of Moses'" (Dick 1990: 26). I would, to the contrary, argue that the law of Moses actually plays a far more basic role in LXX Proverbs than he has been willing to accept. As a matter of fact, there is a concerted effort in this unit to underscore the torah in the face of an opposing movement in Judaism10 to devalue the law (Cook 1999: 457). Consequently, the translator added significant stichs in Prov 9:10a and 13:15 in order to promote the value of the law:
to\ ga\r/de\ gnw/nai no/mon dianoi/aj e0sti\n a)gaqh/jand to know the law is the sign of a good mind.
70. An additional example occurs in connection with the law in Prov 28:4. The Greek represents a significant interpretation of a central Jewish religious concept. MT has:
The Greek reads:(#$frF w%ll;hay: hrFwOt yb'z:(o
Mbf w%rg%Ft;yI hrFwOt yr"m;#$ow:"Those who forsake the law praise the wicked,
but those who keep the law struggle against them".
This is a direct reference to a central Jewish exegetical tradition concerning the law of Moses which plays a fundamental role in this translated unit. At the same time it exhibits significant correspondences and differences with the Letter of Aristeas (Cook 1999: 459).ou[twj oi( e0gkatalei/pontej to\n no/mon e0gkwmia/zousin a)se/beian
oi( de\ a)gapw/ntej to\n no/mon periba/llousin e(autoi/j tei/xojlikewise those who forsake the law and praise impious deeds;
however, they who love the law build a wall around them.
71. The Greek version of Proverbs is an intriguing unit. That the person(s) who were responsible for this unit were creative interpreters is apparent on various levels. As far as lexical issues are concerned, he often exhibits a free approach to the translation task exemplified by variety of lexical choice. However, in fewer instances he chooses to be consistent. Therefore I have defined his lexical approach as one of diversity and unity. He makes abundant use of hapax legomena and neologisms, a clear sign of his proficiency in the Greek culture, at least as far as the form is concerned. He clearly was steeped in the Greek, as well as the Jewish, culture!
72. As far as the translation technique goes, it has become clear that it is imperative to examine the manuscript evidence in connection with individual variants. I demonstrated above that some readings indeed should be taken as hapax legomena and/or neologisms on account of specific manuscripts. Existing lexica/concordances, especially HR, have erred in this regard.11 In the light of some of the textual problems that I have referred to above, it would be wise to be somewhat reserved with our conclusions before the text of this unit has been fixed.12 However, as this undertaking could take many more years to complete, it should not prevent us from venturing into researching this interesting translation unit. The prominent role of the Mosaic law in this unit attests to its Jewishness, as was argued also by Baumgartner (1890) and others. The moralizing present in LXX Proverbs must therefore be understood as stemming from a group of conservative Jews, most probably at the time of translation.
In addition to the abbreviations found in Liddel, Scott, Jones, the following additional abbreviations are used:
© TC: A Journal of Biblical Textual Criticism, 2002.
1This article was completed during my stay at the Katholieke Universiteit Leuven as a senior fellow of Professors Marc Vervenne and Johan Lust. The research was carried out within the "Centrum voor Septuaginta en Tekstkritiek", directed by Johan Lust.
2In this contribution I will generally refer to the translator in the singular, even though it is possible that more than one person was involved (Cook 1997a: 322).
3For this article I have used the Thesaurus Linguae Graecae CD-ROM. I am aware of its problems; however, it makes large amounts of data easily accessible. I have also used various lexica and concordances, notably LEH, LSJ, and HR. I add some basic lexical information concerning each hapax without being exhaustive.
4Cf. J. Lust's treatment of this issue in connection with Symmachus in TC 5 (Lust 2000 link to article).
5Underlining indicates additions compared to MT, and italics, interpretation.
6Cf. LEH 1992: v and Hauspie 2001 ("Neologisms in the Septuagint of Ezekiel").
7In the discussion of neologisms I indicate possible dates. Again, it is not my intention to be exhaustive.
8I did not deal with syntactical issues. Cf. Cook 1997a for a discussion of unique syntactical interpretations followed by this translator.
9Cf. my analysis of the passage in Prov 8:30-31 (Cook 1997a: 229-234), contrary to Gerleman and Hengel. Silva and Jobes (2000: 334) have grasped my intention in this regard excellently!
10Cf. van der Kooij 1997: 522. The Hellenizers of 1 Macc 1 could fit this picture (Goldstein 1976).
11LEH omits the Greek noun kro/kinoj, *h, *on (Prov 7:17), "of saffron", even though it occurs in mss B, R, S*. This applies also to pelidno/j, *h/, *o/n "pale" (Prov 23:29); summeri/zomai (Prov 29:24); surre/mbomai (Prov 13:20) and perhaps u(peu/qunew (Prov 1:23).
12LXX Proverbs was recently assigned to Peter Gentry.
Cook, J. 1987. "Hellenistic Influence in the Book of Proverbs (Septuagint)?" BIOSCS 20: 30-43.
Cook, J. 1999. "The Law of Moses in the Septuagint Proverbs." VT 49: 448-461.
Cook, J. 2000. "Textual Problems in the Septuagint of Proverbs." JNSL 26/1: 163-173.
Deist, F. E. 1988. Witnesses to the Old Testament. Pretoria: NGK.
Gerleman, G. 1950. "The Septuagint Proverbs as a Hellenistic Document." OTS 8: 15-27.
Hauspie, K. 2001. "Neologisms in the Septuagint of Ezekiel." JNSL 27, no. 1, 17-37.
Lust, J. 2000. "A Lexicon of Symmachus' Special Vocabulary in His Translation of the Psalms." TC: A Journal of Biblical Textual Criticism 5.
McKane, W. 1970. Proverbs: A New Approach. Old Testament Library. London: SCM Press, 1970.
Rahlfs, Alfred, ed. 1979. Septuaginta. Stuttgart: Deutsche Bibelgesellschaft. [Cited as Rahlfs]
Silva, M., and Jobes, K. 2000. Invitation to the Septuagint. Grand Rapids: Baker.
Skehan, P. W., and di Lella, A.A. 1987. The Wisdom of Ben Sira. Anchor Bible. New York: Doubleday.
Thesaurus Linguae Graecae 2000. CD-ROM. University of California, Irvine. [Cited as TLG]
Tov, E. 1992. Textual Criticism of the Hebrew Bible. Assen: Van Gorcum.
Wagner, C. 1999. Die Septuaginta-Hapaxlegomena im Buch Jesus Sirach. BZAW 283. Berlin: De Gruyter.