I. The reader of the Septuagint must begin by placing before his mind the conditions under which it was produced, and the relation of the original work to our present texts, Hebrew and Greek.
1. (a) Strictly speaking the Alexandrian Bible is not a single version, but a series of versions produced at various times and by translators whose ideals were not altogether alike. Internal evidence  of this fact may be found in the varying standards of excellence which appear in different books or groups of books. The Pentateuch is on the whole a close and serviceable translation; the Psalms  and more especially the Book of Isaiah shew obvious signs of incompetence. The translator of Job was perhaps more familiar with Greek pagan literature  than with Semitic poetry; the translator of Daniel indulges at times in a Midrashic paraphrase. The version of Judges which appears in our oldest Greek uncial MS. has been suspected by a recent critic  of being a work of the 4th century A.D.; the Greek Ecclesiastes savours of the school of Aquila  . When we come to details, the evidence in favour of a plurality of translators is no less decisive. A comparison of certain passages which occur in separate contexts distinctly reveals the presence of different hands. The reader can readily form a judgement upon this point if he will place side by side in the Hebrew and the Greek 2 Regn. xxii.2 ff. and Ps. xvii. (xviii.) 3 ff., 4 Regn. xviii.17 -- xx.19 and Isa. xxxvi.1 -- xxxix.8, or Mic. iv. and Isa. ii.
A single specimen may be given from Ps. xvii. compared with 2 Regn. xxiii.
Ps. xvii.3 -- 6.2 Regn. xxii.2 -- 6.
One of these versions has doubtless influenced the other, but that they are the work of separate hands seems to be clear from the differences of method which appear e.g. in the renderings of
If further proof is needed it may be found in the diverse renderings of the same Hebrew words in different parts of the Canon. This argument must be used with caution, for (as we shall presently see) such diversities are to be found not only in the same book but in the same context. But after making allowance for variations of this kind, there remain abundant instances in which the diversity can only be attributed to a change of hand. Thus plstym is uniformly represented in the Hexateuch by Phulistieim, but in Judges and the later books by allophuloi; psch is phasek or phasech in Chronicles^(18) and Jeremiah^1, but pescha in all other books; 'vrym is delosis or deloi in the Pentateuch, but in Ezra-Nehemiah photizontes, photison; tmym is aletheia in Exodus, but in Ezra teleion; in Isaiah tsv't is sabaoth more than 50 times, whilst pantokrator, which in other books, is the almost uniform rendering of the word when it is used as a title of Deity, does not once occur; qhl is sunagoge in Gen., Exod., Lev., Num., and again in the Prophets, but ekklesia in Deuteronomy (with one exception) and onwards to the end of the historical books. The singular  phrase ego eimi = 'nky is limited to Judges, Ruth, and 2 -- 4 Regn.; sun = 't of the object occurs in the true LXX. only in Ecclesiastes; amen is peculiar to Chronicles and Ezra, other books which contain the Heb. word (Num., Deut., 1 Regn., Psalms, Jer.) preferring genoito. Similar results may be obtained from a comparison of the forms assumed by the same proper names in different books. Elijah ('lyhv) is Eleiou in the Books of Kings, but Elias in Malachi and Sirach. The lists in Chronicles use the Hebrew form of Gentile names (Thekoei, Anathothei, &c.), where other books adopt the Greek (Thekoeites, Anathotheites, &c.). In Ezra 'chsvdvs becomes Assoueros, but Artaxerxes is substituted by the translator of Esther, and Xerxes by the LXX. translator of Daniel (ix.1)  . It is difficult to resist the force of this cumulative evidence in support of a plurality of translators, especially when it is confirmed by what we know of the external history of the Septuagint.
(b) Further it is clear that the purpose of the version in the later books is not altogether that which the translators of the Pentateuch had in view. The Greek Pentateuch, as we have seen, was intended to supply the wants of the Alexandrian Synagogue. The Book of the Twelve Prophets, and the three major Prophets, were probably translated with the same general purpose, but under a diminished sense of responsibility, since the Prophets, even after their admission to the Canon, were not regarded as sharing the peculiar sanctity of the Law. But the Hagiographa, excepting perhaps the Psalter, stood on a much lower level, and such books as Job, Esther, and Daniel were perhaps viewed by the Alexandrians as national literature  which was not yet classical and might be treated with the freedom allowed by custom in such cases to the interpreter and the scribe. Our estimate of the translator's work must clearly take account of his attitude towards the book upon which he is engaged.
(c) It is important also to bear in mind the peculiar difficulties which beset the translators in their attempts to render the Hebrew Scriptures into Greek. To translate a Semitic book into the language of the West was a new venture when it was undertaken at Alexandria; the Greek Pentateuch "was the work of pioneers and necessarily had the defects of such work  ." No wonder if even in the later books the Hebrew idiom refused to lend itself to the forms even of Hellenistic Greek without losing to some extent its identity, as the translator of Sirach complains  . Moreover the majority of the translators had probably learnt the sacred language in Egypt from imperfectly instructed teachers, and had few opportunities of making themselves acquainted with the traditional interpretation of obscure words and contexts which guided the Palestinian Jew  . The want of a sound tradition is especially manifest in poetical passages and books, and it makes itself felt in the numerous transliterations, and in faulty readings and renderings of the text  . Such things may well make the reader smile at the claim of inspiration which was set up for the LXX., but they ought neither to mislead his judgement, nor to lessen his admiration for the courage and the general success of the Alexandrian translators.
2. The student must also endeavour to realise the condition of the Hebrew text which lay before the Alexandrian translators.
(a) The text of the Hebrew Bible has undergone no material change since the beginning of the second century A.D. A vast store of various readings has been collected from the MSS. by the diligence of Kennicott and De Rossi, but few among them appear to be more than the omissions or corruptions which spring from the accidents of transcription. All existing MSS. belong to one type of text, and it is, in the main, the type which was known to Jerome, to Origen, and to Aquila, and which is reflected in the Targums and the Talmud. But it is not that which was possessed by the Alexandrians of the third and second centuries, B.C. At some time between the age of the LXX. and that of Aquila a thorough revision of the Hebrew Bible must have taken place, probably under official direction; and the evidence seems to point to the Rabbinical school which had its centre at Jamnia in the years that followed the fall of Jerusalem as the source from which this revision proceeded  . The subject, as a whole, will be treated in a later chapter; meanwhile it is sufficient to warn the beginner that in the LXX. he has before him the version of an early text which often differed materially from the text of the printed Hebrew Bible and of all existing Hebrew MSS.
(b) The palaeographical character of the MSS. employed by the translators requires consideration. It will be remembered that the newly discovered fragments of Aquila present the Tetragrammaton in archaic letters  . These letters belong to the old Semitic alphabet which was common to the Hebrew, Moabite, Aramaic, and Phoenician languages, and which appears on the Moabite stone and in the Siloam inscription and, with some modifications, in MSS. of the Samaritan Pentateuch, and on coins of the Maccabean period. The transition from this ancient character to the square letters  which are used in existing Hebrew MSS. and in the printed Bibles must have been practically complete in our Lord's time, since He refers to the yodh as the smallest letter, and to the kereai which are peculiar to the square alphabet (Mt. v.18). That the change had begun in the MSS. employed by the Alexandrian translators  may be gathered from the fact that they repeatedly confuse letters which are similar in the square character but not in the archaic. Professor Driver holds that the alphabet of their MSS. was a transitional one, in which v and v ,y and ch h ,m and m, as well as v and d ,k and r were more or less difficult to distinguish  .
A few examples may be given from Driver's list. (1) 1 Regn. ii.29 ophthalmo (yn, for vn ); xii.3 apokrithete kat' emou (nv vy for yny vv); Ps. xxi. (xxii.) 17 oruxan (k'rv, for k'ry); Isa. xxix.13 maten de sebontai me (vthv yr'tm 'ty, for vthy yr'tm 'ty). ?(2) 1 Regn. vi.20 dielthein (lvr, for lmd); Jer. xxvi. (xlvi.) 25 ton huion autes (vnh for mn')?  ; 1 Regn. iv.10 tagmaton (dgly, for rgly), xxi.7 Doek ho Suros (d'g h'rmy, for d h'dmy).
Another cause of confusion was the scriptio defectiva in the case of v and y where they represent long vowels, e.g.1 Regn. xii.8 kai katokisen autous (vysyvm for vysvvm); Ps. v. tit. huper tes kleronomouses ('l hnchlt, for 'l hnchylvt); Job xix.18 eis ton aiona (lm, for vylym); Jer. vi.23 hos pur (k's, for k'ys). Abbreviations, also, probably gave rise to misunderstandings; see the instances in Driver, op. cit., pp. lxiii. f., lxx. note 2, and others collected from Jeremiah by Streane, Double Text, p.20.
In the case of numerals errors appear to have arisen from the use of similar letters as numerical signs: e.g.2 Regn. xxiv.13 tria ete, 'seven years,' where v has been read for g. Here has the support of the Chronicler (1 Chron. xxi.12): see König in Hastings' D.B., iii. p.562.
Further, in the MSS. used by the LXX. the words seem not to have been separated by any system of punctuation or spacing. On the Moabite stone  and in the Siloam inscription  a point has been used for this purpose, but the Phoenician inscriptions are without punctuation, and so were probably the early Biblical rolls. The division adopted by the LXX. is frequently at variance with that of the Massoretic text, and is sometimes preferable to the latter, sometimes inferior; but the differences witness to the absence of divisions in the Hebrew MSS. and the non-employment of the final letters ts ph n m k.
Thus Gen. xlix.19, 20 auton kata podas. Aser . . . = qv 'sr ?( qv m'sr); Deut. xxvi.5 Surian apebalen = 'rm y'vd ?( 'r#1502;y 'vd); ? 1 Regn. i.1 en Naseib = vnyv ? ( , vn tsvph); Ps. xliii. (xliv.) 5 ho theos mou ho entellomenos = 'lhy mtsvh ?( , tsvh 'lhym); Jer. xxvi. (xlvi.) 15 dia ti ephugen apo sou ho Apis = chph mdv nm ?( , mdv nschph); Zech. xi.7 eis ten Chanaaniten = lknny ?( nyy lkn).
Lastly, almost every page of the LXX. yields evidence that the Hebrew text was as yet unpointed. Vocalisation was in fact only traditional until the days of the Massora, and the tradition which is enshrined in the Massoretic points differs, often very widely, from that which was inherited or originated by the Alexandrian translators  .
A few examples may suffice: Gen. xv.11 kai sunekathisen autois = vysv 'tm ?( , vysv 'tm); Num. xvi.5 epeskeptai = bqr ?( , bqr); ?1 Regn. xii.2 kathesomai = ysvty ?( , vsvty); Nah. iii.8 merida Ammon = mnt 'mvn ?( , mn' 'mvn) Isa. ix.8 thanaton (dvr , ? , dvr) apesteilen Kurios epi Iakob. In proper names the differences of the vocalisation are still more frequent and apparent, e.g. Madiam (mdyn); Balaam (blm), Gomorra (mrh), Chodollogomor (kdrlmr), Phasga (psgh), Sampson (smsvn).
(c) One other preliminary consideration remains. The student must not leave out of sight the present state of the Greek text. A homogeneous text is not to be found even in the oldest of our uncial MSS., and the greater number of Greek codices are more or less influenced by the Hexapla. The Lucianic text is subject to another vice, the Antiochian passion for fulness, which encouraged the blending or the accumulation of various renderings and thus created doublets  . Besides these recensional errors there are the mistakes, itacistic or other, which are incident to the transmission of ancient books. The state of the Greek text has been touched upon already, and will form the subject of a chapter in the third part of this book. Here it is sufficient to notice the presence of mixture and corruption as a factor in the problem which the student of the LXX. must keep in view.
II. We are now prepared to deal with those features of the version which are not incidental but characteristic of the translators' principles and methods.
1. The reader of the Alexandrian Greek Bible is continually reminded that he has before him a translation of a Semitic writing.
(a) As a whole the version aims at fidelity, and often pursues this aim to the extent of sacrificing the Greek idiom. The first chapter of Genesis will supply instances of extreme literalness, e.g. v.4 ana meson tou photos kai ana meson tou skotous; v. ?5? egeneto hespera kai egeneto proi, hemera mia; v.20 herpeta psuchon zoson. As we proceed, we are still conscious of moving in an atmosphere which is Hebrew and not Greek. Hebrew constructions meet us everywhere; such phrases as aphikesthai heos pros tina, parasiopan apo tinos, prostithenai (tou) poiein, lalein en cheiri tinos, echthes kai triten, apo geneon eis geneas (heos geneas kai geneas, eis genean kai genean) may be found in the Prophets and Hagiographa as well as in the Pentateuch. Occasionally the translators set the sense at defiance in their desire to be true to what they conceive to be the meaning of the Hebrew, as when in 1 Regn. i.26 they render by (deomai) by en emoi. In some books, especially perhaps in the Psalms and in Isaiah, entire sentences are unintelligible from this cause. Even when the Alexandrians have rightly understood their original they have generally been content to render it into Greek with little regard for rhythm or style, or the requirements of the Greek tongue.
(b) To the same spirit of loyalty may be ascribed in part the disposition to transliterate words which present unusual difficulty. The number of transliterations other than those of proper names is considerable  , and they are to be found in nearly all the translated books. In some cases they are due to misunderstanding, as in Jud. i.19 Rhechab diesteilato autois where h) vrzl) seems to have been read as hvrzl, and rkv consequently treated as a proper name; in others, the Hebrew form is purposely maintained (e.g. hallelouia, amen). But in the majority of instances transliteration may be taken for a frank confession of ignorance or doubt; it is clearly such, for example, in Jud. viii.7 en tais abarkenein, 4 Regn. ii.14 aphpho ('ph hv'), Jer. xxxviii. (xxxi.) 40 pantes hasaremoth heos nachal Kedron. As in the first and third of these specimens, the article is often included; and when a proper name is transliterated, the name is sometimes for this reason not easily recognised; thus Ramathaim (1 Regn. i.1) becomes Armathaim (hrmtym)?  . Similarly the h local is taken over in the transliteration, as in Gen. xxxv.6 eis Louza = lvzh. Sometimes two words are rolled into one, as in Oulammaus = 'vlm lvz (Gen. xxviii.19)  . A doublet is occasionally created by adding a translation to the transliterated Hebrew, e.g. in 1 Regn. vi.11 , 15, to thema ergab, vii.4 ta alse Astaroth, xxiii.14 en Maserem en tois stenois. In the case of a significant proper name, where it is necessary for the reader to be made aware of its meaning, the LXX. sometimes translate without transliterating, e. g. Gen. iii.20 ekalesen Adam to onoma tes gunaikos autou Zoe (hvh); xi.9 eklethe to onoma autou Sunchusis (bvl); xiv.13 apengeilen Abram to perate (hvry).
2. The Alexandrian translators, however, while loyal to their original, sometimes even to a fault, manifest nothing like the slavish adherence to the letter with which Aquila has been charged. They often amplify and occasionally omit; they interpret, qualify or refine; they render the same Hebrew words by more than one Greek equivalent, even in the same context; they introduce metaphors or grammatical constructions which have no place in the Hebrew text and probably at no time had a place there, or they abandon figures of speech where they exist in the original.
(a) Slight amplifications, which are probably not to be ascribed to a fuller text, occur frequently in all parts of the LXX.; e.g. the insertion of legon before a quotation, or of pronouns which are not expressed in the Hebrew, or of single words added in order to bring out the sense, as in Gen. xxxiv.10 idou he ge plateia enantion humon, xl.17 apo panton ton genematon hon ho basileus Pharao esthiei, Deut. vii.16 phage panta ta skula ton ethnon (Heb. 'thou shalt eat all the nations'). The translators frequently manifest a desire to supply what the original had omitted or to clear up what was ambiguous: they name the subject or object when the Hebrew leaves it to be understood (Gen. xxix.9 aute gar ebosken ta probata tou patros autes, Heb. 'fed them'; xxxiv.14 kai eipan autois Sumeon kai Leui hoi adelphoi Deinas huioi de Leias, Heb. 'and they said unto them'), or they add a clause which seems to follow as a necessary consequence (2 Regn. xii.21 anestes kai ephages arton kai pepokas: xvi.10 kai aphete auton kai houtos katarastho = yqll ?(kh q) ?ky, or they make good an aposiopesis (Exod. xxxii.32 ei men apheis autois ten hamartian auton aphes). Less frequently they insert a whole sentence which is of the nature of a gloss, as in Gen. i.9 kai sunechthe to hudor to hupokato tou ouranou eis tas sunagogas auton kai ophthe which is merely an expansion of kai egeneto houtos in the terms of the preceding command sunachtheto ktl.; or 1 Regn. i.5 hoti ouk en aute paidion a reminiscence of v.2 te Hanna ouk en paidion. On the other hand the LXX. not uncommonly present a shorter text, as compared with M.T., e.g. Gen. xxxi.21 kai diebe ton potamon (Heb. 'he rose up and passed over'), ib.31 eipa gar Me pote ktl. (Heb. 'Because I was afraid, for I said . . . '); 1 Regn. i.9 meta to phagein autous en Selo (Heb. 'after they had eaten in Shiloh and after they had drunk').
(b) The translators frequently interpret words which call for explanation. Hebraisms are converted into Greek phraseology, e.g. bnnkr becomes allogenes (Exod. xii.43), and bnsnh eniausios (Num. vii.15); v'ny rl sphtym is rendered by ego de alogos eimi (Exod. vi.12). A difficult word or phrase is exchanged for one more intelligible to a Greek reader; thus he eremos is used for hngv (Gen. xii.9); 'Urim and Thummim' become he delosis kai he aletheia (Exod. xxviii.26); in the Psalms antilemptor is written for mgn (Ps. iii.4), boethos for tsvr (xvii. = xviii.3), and glossafor kvvd (Ps. xv. = xvi.9); similarly in Jer. ii.23 to poluandrion 'the cemetery' stands for bgy', i.e. the valley of Hinnom  . An effort is made to represent Hebrew money by its nearest Greek equivalent; thus for sql we have didrachmon (Gen. xxiii.15, Deut. xxii.29, 2 Esdr. xv.15) as well as siklos, and for grh obolos. Occasionally a whole clause is interpreted rather than translated; e.g. Gen. i.2, ? he de ge en aoratos kai akataskeuastos, Exod. iii.14 ego eimi ho on, Ps. xl. (xxxix.) 7 soma de katertiso moi. A dogmatic interest has been detected in some of these paraphrastic renderings, chiefly where the LXX. have endeavoured to avoid the anthropomorphisms of the original; examples are most frequent in the Pentateuch, e.g. Gen. xviii.25 medamos su poieseis (Heb. 'that be far from thee'); Exod. iv.16 su de auto ese ta pros ton theon (l'lhym); xxiv.10 eidon ton topon hou heistekei ho theos tou Israel (Heb. 'they saw the God of Israel,' Aq. eidon ton theon Israel; ib.11 ton epilekton tou Israel ou diephonesen oude heis; Num. xii.8 ten doxan (tmnt) Kuriou eiden; Exod. xv.3 Kurios suntribon polemous ('ys mlchmh); Deut. xiv.23 ho topos hon an eklexetai Kurios ho theos sou epiklethenai (lskn) to onoma autou ekei; Jos. iv.24 he dunamis tou kuriou (ydyhvh). Such renderings manifest the same spirit of reverence which led the LXX. to write ho kurios, or the anarthrous Kurios, or not infrequently ho theos for the Tetragrammaton, just as their Palestinian brethren read for it 'dny or 'lhym?  . In other places the LXX. appear to be guided by the Jewish Halacha, e.g. Gen. ii.2 sunetelesen ho theos en te hemera te hekte (hsvyy, Aq. te hebdome); Lev. xxiv.7 epithesete epi to thema libanon katharon kai hala  ; xix.7 ean de brosei brothe te hemera te trite, athuton estin (Heb. 'an abomination')  . Of Haggada also there are clear traces, as in Exod. xii.40 en ge Aigupto kai en ge Chanaan, 1 Regn. i.14 eipen aute to paidarion Elei  ; v.6 kai meson tes choras autes anephuesan mues, kai egeneto sunchusis thanatou megale en te polei.
(c) The LXX. render the same Hebrew word by more than one Greek equivalent, sometimes even in the same context. In some cases the change appears to be either arbitrary, or due to the desire of avoiding monotony; e.g. in Ps. xxxvi. (xxxvii.) rs is translated by hamartolos in vv.10, 12, 14, 16, 17, 20, 21, 32, 40, but by asebes in vv.28, 35, 38. In many others it may be ascribed to the circumstance that certain common Hebrew words take a special colouring from the contexts in which they occur, and must be rendered accordingly. Thus ntn, 'give  ,' which belongs to this class has received in the LXX. more than 30 different renderings; sometimes it is translated by a paraphrase, e.g. Jos. xiv.12 aitoumai se (tnh ly) Deut. xxi.8 hina me genetai ('l ttn); when it is rendered directly, the following Greek verbs (besides didonai and its compounds) are used to represent it: agein, apostellein, apotinein, aphienai, deiknunai, doreisthai, ean, ektithenai, ektinein, ekcheein, elean, emballein, enkataleipein, epairein, epiballein, epitithenai, epicheein, ephistanai, histanai, kataballein, kathistanai, katatassein, kremazein, paratithenai, peritithenai, poiein, proekpherein, prosienai, prostithenai, sterizein, sunagein, pherein. This is a somewhat extreme instance, but a glance at Hatch and Redpath will shew that there are many which do not fall far behind it, and that in the majority of cases the ordinary words of the Hebrew Bible have more than one equivalent in the Greek of the LXX. The Alexandrian translators have evidently made an honest endeavour to distinguish between the several connotations of the Hebrew words. Thus, to take a few examples: qts is variously rendered by akron, arche, klitos, meros, peras, taxis, chronos; among the equivalents of dvr are apokrisis, eperotesis, krima, pragma, tropos, phone; for lv we have not only kardia, psuche, phren, nous, dianoia, stoma, phronesis, but stethos and even sarx; for phqd, apithmein, episkeptesthai, etazein, ekdikein; for tsdqh, dikaiosune, eleemosune, euphrosune. Conversely, the same Greek word often serves for several Hebrew words. Thus diatheke, which is generally the LXX. rendering of bryt, stands also for dvt (Exod. xxvii.21, (xxxi.7), tvrh (Dan. ix.13, LXX.) and even dvr, (Deut. ix.5); exairein, lutroun, rhuesthai are all used to represent g'l; eidolon appears in different contexts for psl ,tsv ,chmn ,chvl ,bl ,bmh ,'lyl ,'lvh ,'l ?trphym ,sqts ,tslm. Even in the same context, or verse this sometimes occurs. Thus in Gen. i. -- iii. ge translates tsphr ,sdh ,'dmh ,'rts; in Exod. xii.23 vr and phsch are both represented by parerchesthai; in Num. xv.4 f. thusia is used both for mnchh and zvch. In such cases it is difficult to acquit the translators of carelessness; but they are far less frequent than instances of the opposite kind. On the whole the LXX. even in the Pentateuch shews no poverty of words, and considerable skill in the handling of synonyms.
(d) In reference to metaphors the Alexandrians allow themselves some discretion. Thus in Gen. vi.2 'the sons of God' become hoi angeloi tou theou; in Num. xxiv.17 'a sceptre (svt) shall rise' is rendered by anastesetai anthropos; in Deut. x.16 'the foreskin of your heart' is turned euphemistically into ten sklerokardian humon; in Isa. ix.14 megan kai mikron represents Heb. 'both branch and rush.' Occasionally the translators indulge in paronomasia, without authority from the Heb., e.g. Gen. xxv.27 oikon oikian = yvsv 'hlym; xxvi.18 kai eponomasen autois onomata vyqr' lhn smvt; Job xxvii.12 kena kenois; xxx.13 exetribesan triboi mou.
(e) Lastly, the reader of the Septuagint must expect to find a large number of actual blunders, due in part perhaps to a faulty archetype, but chiefly to the misreading or misunderstanding of the archetype by the translators. Letters or clauses have often been transposed; omissions occur which may be explained by homoioteleuton; still more frequently the translation has suffered through an insufficient knowledge of Hebrew or a failure to grasp the sense of the context. It follows that the student must be constantly on his guard against errors which may easily result from too ready an acceptance of the evidence offered by the Alexandrian version. Taken as a whole, and judged in the light of the circumstances under which it was produced, it is a monument of the piety, the skill, and the knowledge of the Egyptian Jews who lived under the Ptolemies, and it is an invaluable witness to the pre-Christian text of the Old Testament. But whether for textual or for hermeneutical purposes it must be used with caution and reserve, as the experience of the Ancient Church shews. With this subject we shall deal in a future chapter; it is sufficient to note the fact here.
III. The beginner, for whose use this chapter is chiefly intended, will now be prepared to open his Septuagint and his Hebrew Bible, and to compare the two in some familiar contexts. The following notes may assist him in a first effort to grapple with the problems which present themselves.
GEN. xv.1 -- 6.
1. Ta rhemata . . . rhema, Heb. dvrym . . . dvr. Legon = l'mr; cf. v.4, where, as elsewhere, Aq. renders, to legein. Huperaspizo sou, Heb. 'am a shield to thee'; cf. Deut. xxxiii.29, Prov. ii.7, al. Ho misthos sou polus. Vulg., A.V., R.V. connect Heb. with the foregoing, supplying v. ?2. Despotes = 'dny, as in v.8, and not infrequently in Jer. and Dan. (LXX.). Apoluomai ateknos -- an interpretation rather than a literal rendering of hvlk ryry?  . Huios Masek tes oikogenous mou = vn msq vyty: cf. Hieron. quaest. in Gen. "ubi nos habemus Et filius Masec vernaculae meae, in Hebraeo scriptum est nvn msq vyty, quod Aquila transtulit ho huios tou potizontos oikian mou . . . Theodotio vero kai huios tou epi tes oikias mou." Damaskos Eliezer, a literal rendering of the Heb., leaving the difficulty unsolved.3. Epeide = hn, and so in xviii.31, xix.19; did LXX. read 'm? Oikogenes here = bnhbyt. Kleronomesei me -- a Hebraism, = kleronomos mou estai.4. kaiuthus . . . egeneto = vhnh. Phone = rvr, as in xi.1, but apparently not elsewhere. hos . . . houtos, 'sr . . . hv'. Ek sou, euphemism for Heb. m?myk, unless the LXX. read mmk. ?5. Pros auton, ^ Heb.6. Kai episteusen = vy'mn (cf. Haupt ad loc.). Abram ^ to theo = byhvh. Elogisthe . . . eis dik., Heb. 'he counted it . . . for righteousness'; possibly the LXX. read as in Ps. cvi.31 (M.T.), where they have the same rendering. The N.T. follows LXX. here (Jas. ii.23, Rom. iv.3, Gal. iii.6).
EXOD. xix.16 -- 24.
16. Egeneto de . . . kai egenonto = vyhy. . . vyhy. Genethentos pros orthron = bhyt hbqr. Ep' orous Seina, Heb. 'on the mountain.' Phone, cod. F with pr. kai.17. Hupo to oros S. (om. S. AF), Heb. 'at the nether part (btchtyt) of the mountain:' 18. dia to katabebekenai, an idiomatic rendering of m?pny'sr yrd. Ton theon = yhvh, cf.21. Ho kapnos, Heb. 'the smoke of it.' Exeste, Heb. as v.16 where LXX. renders eptoethe. Ho laos = hm; M.T., hhr. ?19. Probainousai ischuroterai = hvlk vchzq. ?20. Ekalesen . . . Mousen, Heb. lmsh; the l after qr' is dropt in accordance with Greek idiom  .21. Legon, ^ Heb. Engisosin, a softening of the .Heb. 'break forth' (hrs); in the next verse engizein = ngs ni.22. kai, Heb. 'and also' (vgm), usually kai ge, Aq. kai kaige (Burkitt, Aquila, p.13). Kurio to theo a double rendering of 'l yhvh. Apallaxe ap' auton: another instance of euphemism: Heb. 'break forth upon them' (Aq. diakopse en autois).23. Prosanabenai: the double compound occurs six times in Jos. xi. -- xix. Aphorisai: the verb is here as in v.12 the equivalent of gvl hi. 'enclose,' but with the added thought of consecration which is latent in aphorizein, aphorisma, aphorismos (cf. Exod. xxix.26, Ezech. xx.40).24. Apolese, euphemistic, as apallaxe in v.22; Aq. again, diakopse.
NUM. xxiii.7 -- 10.
7. Parabolee: here for the first time = msl. Lyons Pent., parabula. Mesopotamias, i.e. 'rm nhrym (Gen. xxiv.10), or 'rm pdn (Gen. xxv.20): here an interpretation of the simple 'rm . Ap', legon, Heb. Epikatarasai moi, and katarasomai in v.8, represent zm, whilst arasai answers to 'rr, and arasomai (v.8) to nqv, an unusual instance of carelessness or poverty of language on the part of the translator; oreon (v.9) is equally unfortunate as a rendering of tsrym, while on the other hand opsomai, prosnoeso fairly represent the Heb. Prosnoein renders svr again in Job xx.9, xxiv.15.10. Exaribazesthai (Num.^1, Job^1, Dan. LXX..^1), a late form for exakriboun in LXX. and Jos. To sperma, Heb.'the dust': did LXX. read zr, or have they glossed phr? Kai tis exarithmesetai, reading vmy ysphr. Demous Israel, Heb. 'the fourth part of Israel' (Aq. tou teturtou I.). He psuche mou, as Heb., whilst the next word is sacrificed to an alliteration (psuche, psuchais). To sperma mou is a gloss on 'chryty (cf. Brown, Heb. and Eng. Lex., p.31); hos to sperma touton, Heb. 'as he.'
This passage illustrates both the greater freedom which the Greek translators allowed themselves in poetical contexts, and their comparative incompetence to deal with them.
DEUT. vi.1 -- 9.
1. Hautai hai entolai, Heb. 'this is the commandment.' Ho theos hemon, Heb. 'your God.' Houtos ^ Heb. Eisporeuesthe, Heb. 'go over'; the Greek has lost the local reference, as in iv.14, 4 Regn. iv.8.2. Hina phobesthe . . . humon, Heb.2nd pers. sing. Semeron, ^ . Hoi huioi ktl., Heb. 'thy son and thy son's son.' Hina makroemereusete, Heb. 'and that thy days may be prolonged'; makroemereuein (makroemeros ginesthai) represents this or a similar phrase in iv.40, v.30, xi.9, 21, xxxii.47; makrochronios, makrochronizein also occur in iv.40, v.16, xvii.20, xxxii.27. The group is not found elsewhere in the LXX. except in Exod.^1, Jud.^1, and in Sirach.3. Dounai ^ M.T.; perhaps added to complete the sense of the Greek; yet see v.10 (ltt lk). ?4. Kai tauta . . . Aiguptou ^ Heb; perhaps repeated from iv.45 to form an introduction to Akoue ktl..5. Dianoias . . . psuches . . . dunameos. The readings vary; for dianoias AF Luc. read kardias, and the text of B is here super rasuram; for dunameos some texts give ischuos. The N.T. citations (Mt. xxii.37 = Mc. xii.29 ff., Lc. x.27) present much diversity, giving both renderings of lvvk and both of m'dk; cf. Dittmar, V. T. in Novo, p.50 f. 6. kai en te psuche sou, ^ Heb.; for 'in thy heart' Heb. has 'upon,' "as it were imprinted there (Jer. xxxi.33)  ." 7. Probibaseis, Heb. 'shalt impress them upon'; Aq. deueeroseis, as if the root were snh. En autois = bm. Kathemenos ktl., Heb. 'in thy sitting &c.'; en oiko, en hodo are inexact, Heb. 'in thy house,' 'in the way.' 8. Asaleuton (F, asaleuta) = lttpht, 'for frontlets,' circlets or tires for the head: Lyons Pent. (reading saleuta), mobilia. Asaleuton occurs in the same phrase in Exod. xiii.16, Deut. xi.18. Aq. seems to have rendered the Heb. here and in Exod. by nakta, i.e. 'compressed,' 'tight,' which Field (Hexapla, i.103) explains as the "thecas in quas schedulae membraneae . . . inferciebantur." The LXX. rendering may be an Alexandrian name for the phulakterion, but the whole subject is obscure. 9. Phlias = mzzvt, as in Exod. xii.7 ff.
JOS. x.12 -- 14.
12. He hemera paredoken . . . hupocheirion -- idiomatic rendering of byvm tt . . . lphny. The words that follow (henika . . . Israel) seem to be a gloss derived from v.10. Kai eipen Iesous, Heb. 'and he said in the eyes of Israel.' Steto, Heb. 'be still.' Gabaon, 'Gibeon.' Ailon. 'Aijalon' ('ylvn); Cf.2 Chron. xi.10 A, Aialon.13. En stasei = md, which is thus distinguished from the verb represented by este. Ho theos, Heb. gvy, Aq. to ethnos. Unless a primary error is to be suspected here, the LXX. has glossed its original, from motives of piety. After the stanza inserts a reference to the Book of Jashar, which is wanting in non-Hexaplaric texts of the LXX.; cod. G adds, ouchi touto gegrammenon epi bibliou tou euthous . Ou proeporeueto ktl., a loose rendering of Heb.
JUD. v.28 -- 30  .
28. ^B here omits the difficult word vtyvv ( ^A, kai katemanthanen). ektos tou toxikou, 'forth from the loophole'; cf. Symm. in Ezek. xl.16 thurides toxikai: ^A dia tes diktuotes, 'through the lattice' (cf.4 Regn. i.2, Ezek. xli.16). Epiblepousa . . . Sisara in A appears to be a supplementary gloss. Heschunthe (B) confuses vss pOlel with vvs kal; the general sense of the former is given by eschatisen A. For eschatizein cf.1 Macc. v.53; has it been suggested here by its similarity to the word used in B? Podes: A more literally ichne, but pous represents pm elsewhere, e.g. Ps. lvi. (lvii.) 6, Prov. xxix.5.29. Hai sophai archousai: A, again aiming at a literal rendering, ?sophai archouson. On the other hand B's apestrepsen logous autes heaute is close and yet idiomatic, while A's apekrinato en rhemasin autes goes too far afield; the latter appears to be a Hexaplaric correction (Field, ad loc.).30. Ouch heuresousin auton diamerizonta skula; so ^BA; Heb. 'are they not finding, [are they not] dividing booty?' LXX. seem to have read mchlq for ychlqv. Oikteirmon oikteiresei B, philiazon philois A; both, while labouring to keep up the alliteration of the Heb., miss its point through ignorance of a rare use of rchm?  ; for philiazein cf. xiv.20 B, 2 Chron. xix.2.2. Poikilton (A, poikilon) misses the dual 'embroidery on both sides' (R. V.), or 'a couple of pieces,' "precisely as rchmtym above" (Moore). Bathe in A seems to be an error for baphe, which is found in several cursives; see Field, ad loc., and Lagarde's Lucian. To trachelo autou skula = apparently ltsv'ryv sll; M.T. 'for the necks of the spoil.' ^A substitutes the usual anatole for the spirited and literal rendering of B (cf. Ps. xviii. = xix.7), and appears to have read vgvrtyv; cf. Ps. xix. (xx.) 7.
This passage is a severe test of the translator's knowledge and skill, and shews him perhaps at his worst.
1 REGN. xvii.37 -- 43.
37. begins vy'mr dvd, A, Luc. kai eipen D. Ek cheiros tou leontos . . . tes arkou, an exact rendering; cf. Gen. ix.5 ek cheiros panton ton therion. Luc., Th., ek stomatos tou l. kai ek cheiros tes arkou. Tou aperitmetou, repeated from v.36 (^ ).38. manduan (Jud. iii.16, 2 Regn. x.4): + autou, A, with . Perikephalaian ch peri ten kephalen autou: Luc. (A), with , p. ch. epetheken epi ktl., adding, kai enedusen auto thoraka.39. Ezosen ton Daueid sc. Saoul (cf. v.38); Luc., A, follow Heb. in making David the object of the verb ezosato Daueid). Ekopiasen peripatesas (A, peripatesai) hapax kai dis, 'more than once he wearied himself with walking (strove to walk) in them,' reading vyl', as in Gen. xix.11 vyl'v LXX. pareluthesan (Wellhausen, Driver, H. P. Smith). Hapax kai dis occurs also in Deut. ix.13 (where, as here, there is nothing in the Heb. to correspond), and in Neh. xiii.20, where it represents pm vstym. Aphairousin auta ap' autou, reading the verb probably as vysrm, and omitting dvd. ?40. Lithous teleious in B is obviously wrong, and A scarcely mends matters by omitting the adjective. Correct, with Lucian, lithous leious. En to kadio to poimeniko: kadion = kadiskos, here only in LXX., and perhaps unknown elsewhere: poimenikos (hrym) again in Zach. xi.15. Eis sullogen apparently for lylqvt ?( vvylqvt, Aq. kai en analekterio).41 is wanting in ^B, and probably belongs to the same recension of the story which has supplied the great gaps vv.12 -- 31, 55 -- xviii.5.42. Heb. 'looked and saw'; so A, Luc. Kurrakes; cf. xvi.12, Gen. xxv.25.43. Hosei, added by the translators to soften the opprobrious kuon. En rhabdo kai lithois, 'in (with) staves'; kai lithois is probably intended to make the question correspond to the statement of v.40. The next words in the LXX. kai eipen Daueid Ouchi, all' e cheiro[n] kunos are evidently of the same character -- a "singularly vapid reply" (Driver).
4 REGN. ii.11 -- 18.
11. Auton poreuomenon eporeuonto kai elaloun -- an interesting attempt to combine Greek idiom with some reminiscence of the Heb. phrase; Lucian abandons the Heb., and corrects, auton poreuomenon kai lalounton. Hippos puros, Heb. 'horses of fire'; cf. hippeus, Heb. 'horsemen,' v.12. Ana meson (byv), cf. Gen. i.7 diechorisen . . . ana meson. Anelemphthe, Heb. 'went up'; the Greek verb is apparently repeated from vv.9, 10, where it = lqch. >From this passage it has been borrowed by the translator of Sirach (xlviii.9, 14, xlix.14, B), and by two writers in the N.T. (1, Acts i.2, 11); on its symbolical use see the writer's Apostles' Creed, p.70 f. Hos, ^ Heb.; cf.1 Regn. xvii.43 (above).12. Pater pater, Heb. 'my father' bis. Dierrexen . . .rhegmata, after the Heb.: Lucian omits the noun, probably because of the harshness of the assonance.13. Kai hupsosen = vyrm; Luc., kai aneilato. Meloten, 'sheepskin,' an interpretation of 'drt (Vulg. pallium ) wherever it is used of Elijah's characteristic raiment (3 Regn. xix.13, 19, 4 Regn. ii.8 ff.); cf. Heb. xi.37 perielthon en melotais. Epanothen, sc. autou (Heb., Luc.). Eleisaie, ^ Heb.; kai epestrepsen Eleisaie is Hexaplaric, and wanting in B*, but supplied by B^abA Luc.14. Ho theos, yhvh 'lhy. Aphpho, a transliteration answering to 'ph hv' ( ); in x.10 the same form = 'phv', which was perhaps the reading before the LXX. in this place. Aq. kaiper autos, but Symm. kai nun, whence Jerome etiam nunc .15. kai hoi en Iereicho: ^ kai A Luc. with .16. ys is not represented by ^AB; Luc. adds eisi. Huioi dunameos, bnychyl. En to Iordane, Eleisaie, ^ Heb., Luc.18. In A Luc. Aq. Th. the verse begins 'And they returned to him'; cf. v.13.
PS. cix. (cx.) 1 -- 4.
1. [Ho] kurios to kurio mou, yhvh l'dny. Ek dexion, lymyny; in v.5 the same Gr. is used for l ymyny. Hupopodion ton podon sou: hupokato is the reading of the best authorities in Mt. xxii.44, Mc. xii.36, but hupop. keeps its place in Lc.^ev. act., Hebrews.2. kai katakurieue = vrdh apparently.3. Meta sou, mk ?( , mk). He arche seems to point to a reading ndyvh or ndyvt (cf. Job xxx.15, Isa. xxxii.8); ton hagion (sou) = qdsym) qdsyk); Symm. en oresin (vhrry for vhdry) hagiois. Ek gastros pro heosphorou egennesa se, though not quoted in the N.T., had an important place in post-apostolic Christian teaching from Justin onwards (cf. Justin, Tryph. cc.63, 76, 83; Tert. adv. Marc. v.9; Cypr. test.17, ep.63); in the Arian age it was commonly cited on the Catholic side -- see e.g. Cyril. Hierus., catech. vii.2, xi.5; Athan. or. c. Arian. iv.27 sq.; de decr.3, &c.; Hilar. de trin. vi.16, xii.8. The O.L. seems to have rendered uniformly ex utero ante luciferum genui te, with the variant generavi in Tert. l.c.; Jerome's 'Hebrew' Psalter reads with quasi de vulva orietur tibi ros adolescentiae. The LXX. appear to have read their Heb. text as mrchm mschr yldtyk, perhaps dropping lktl as unintelligible.4. Kata ten taxin, l dvrty Aq. Symm. kata logon. Cf. Heb. v.6 ff., vii.11, 15 (kata ten homoioteta). The translator probably had before him the LXX. of Gen. xiv.18; he transliterates the unique name mlkytsdq in the same way.
PROV. viii.22 -- 25, 30 -- 31.
22. Ektisen me. So ^'BA etc. O.L. ( condidit, creavit ); codd.23 = V, 252, with Aq. Symm. Th. Vulg. ( possedit ), give ektesato -- both possible meanings of qnh. The former rendering supplied the Arians with one of their stock arguments (cf. Athan. or. c. Arian. ii.44 sqq.). Eis erga autou, a loose and partial translation, probably a confession of inability to understand the Heb.; Th. pro tes ergasias apo tote.23. Ethemeliosen me, reading apparently ysdny where has nskty; cf. Ps. lxxvii. (lxxviii.) 69. Pro tou ten gen poiesai, a poor rendering of Heb., probably adopted to bring this clause into line with v.24 with which the LXX. seem to have connected it.24. LXX. overlook chvllty and nkvdy, unless they intend to convey the general sense by poiesai and proelthein.25. Panton, ^ 'I was brought forth.' 30. harmozousa = 'mvn, the word being referred by the translator to 'mn; similarly Symm. Th., esterigmene. He prosechairen implies the reading yvm yvm ;ssvyv is connected by LXX. with the next clause.31. Hote . . . suntelesas: Heb. 'rejoicing in the world of his earth.' LXX. seem to have read mschq vtklyt, as Lagarde suggests; had tvl stood in their text, oikoumene would have been ready at hand as a rendering (cf.2 Regn. xxii.16, Ps. ix.9, &c.). Euphraineto, reading ssyv. Huioi anthropon = bny 'dm; cf. huious Adam, Deut. xxxii.8; b 'dm is translated by this phrase in Ps. x. (xi.) 4, and repeatedly in the poetical books.
JOB xix.23 -- 27.
23. Tis gar an doe; See above p.308; the phrase is repeated in the Hebrew, but the translator contents himself with using it once. 'phv is ignored; its usual equivalent in the LXX., is nun or oun, unless it is transliterated (p.324). Eis ton aiona seems to represent ld, which in belongs to the next verse; Th. translates it eis marturion, reading the word as ld. ?24. B* omits en petrais engluphenai which appears to be necessary to the sense; in supplying it B^ab'A prefix e, a manifest gloss.25. Aenaos estin ho ekluein me mellon, a paraphrase of Heb. 'my Goel lives'; aenaos in the LXX. elsewhere = lm, and g'l is anchisteus (Ruth iii.9, etc.), or lutrotes (Ps. xviii.14, lxxvii.35).25 -- 26. Epi ges anastesai or anastesei appears to correspond with l phr ?(yqvm (yqym, and to derma mou to anantloun tauta with vry nqphv z't. ^A points to lchyvt vry mklkl z't (Siegfried in Haupt ad loc.). But the translator perhaps interprets his text in the light of the doctrine of the Resurrection, which was accepted from Maccabean times (cf. Job xlii.17^a, and see Dan. xii.2, 2 Macc. vii.14, xii.43); as cited by Clem. R.1 Cor.26 (anasteseis ten sarka mou tauten ten anantlesasan tauta panta), the words are brought into still nearer agreement with the faith of the Church; see Apostles' Creed, p.89 f. Para gar Kuriou . . . sunetelesthe corresponds in position with words which divides and points as vm?bsry 'chzh 'lvh, but seems to be partly borrowed from the next verse. ^A suggests ?vm'lvh nsv ly 'lh (Siegfried).27. Panta de moi suntetelestai; , klv klyty.
MICAH v.1 (iv.14) -- 4 (3).
1. Emphrachthesetai thugater emphragmo, i.e. ttgdry vt gdr. Tas phulas tou Israel: LXX. read svty ysr'l for spht y. ?2. Bethleem oikos tou Ephratha: did LXX. read ?bytlchm byt 'phrth? Oligostos ei tou einai 'art little to be,' as Heb. The passage is quoted in Mt. ii.6 in a Greek paraphrase  which substitutes oudamos elachiste for 'little to be,' and tois hegemosin ('lphy) for 'thousands' ('lphy). ?3. Heos kairou tiktouses texetai, apparently for heos kairou hou tiktousa texetai or he. k. tiktouses hote texetai.4. Kai opsetai, to poimnion autou were obelised in Hex. and find no place in ; the former has perhaps originated in a misreading of vrh as vr'h, so that kai ops. kai poimanei is in fact a doublet. Kurios, subject; Heb. 'in the strength of J.,' the subject being the same as in v.1. Huparxousin, vysvv; the LXX. read ysvv, connecting the verb with the previous words; for ysv = huparchein cf. Ps. liv. (lv.) 20 ho huparchon pro ton aionon.
JEREM. xxxviii.31 -- 37 (xxxi.30 -- 36).
Vv.31 -- 34 are cited in Heb. viii.8 -- 12, q.v.31. Diathesomai, in Hebrews sunteleso, cf. Jer. xli. (xxxiv.) 8 suntelesai (krt) diatheken, and ib.15. To oiko bis, in Hebrews epi ton oikon.32. Diethemen, in Hebrews epoiesa: the writer appears to dislike the repeated alliteration in diatithesthai diatheken. En hemera epilabomenou mou, for the more usual tou epilabesthai me or hote (he) epelabomen. Hoti ouk enemeinan en . . . Heb. 'which . . . they broke'; emelesa auton, reading glty for vlty. ?33. he diatheke mou, Heb. 'the covenant' Didous doso, a Hebraism not represented in ; in Hebrews didous appears without doso, and so AQ in Jer. Eis ten dianoian auton, Heb. 'in their inward parts.' 34. vd 1° has no equivalent in the Greek; ton politen autou, Heb. 'his neighbours' (cf. Prov. xi.9.12, xxiv.43 = 28), reminds us that we are dealing The paraphrastic character of the reference appears more distinctly in the second stanza ek sou . . . Israel, which blends Mic. v.1^b, 3^a. It will be observed that cod. A reads hegoumenos with Mt. with an Alexandrian version. Apo . . . heos, l . . . vd; adikiais . . . hamartion, 'iniquity,' 'sin.' 35 -- 37. In 36, 37 precede 35.35. Phesin Kurios, Heb. 'thus saith J.' (at the beg. of the verse). Hupsothe, reading yrvmv for ymrv; tapeinothe, Heb. 'be searched.' Ouk apodokimo: apod. is a contracted future (cf. p.305); ouk is inserted, because the drift of the verse has been misunderstood (cf. Streane, p.156 f.). To genos Israel, Heb. 'all the seed of I.'; genos = zr again in v.37.36. Selenen, , 'the ordinances of the moon' (but cf. hchqym in v.35, Heb.). Kraugen, reading perhaps rgs or rgz for rg. ?37. Kurios Pantokrator = ?yhvh tsv'vt, as almost invariably in the Prophets  from Hosea xii.5 (6) onwards, with the exception of Isaiah, who transliterates tsv'vt (Kurios sabaoth, Isa. i.9, a1). See Thackeray, J. Th. St. IV. p.245 ff.; this passage is from his "Jer. b."
DAN. xii.1 -- 4.
1. Choran (LXX.), probably a corruption for horan (cf. Bevan, p.48); pareleusetai (LXX.), reading yvr for ymd (anastesetai, Th.). Ho angelos (LXX.), a gloss; Th. literally, ho archon. Epi tous huious (LXX., Th.), . . . l bny. Ekeine he hemera, LXX., estai kairos Th.; Th. is again more literal than LXX. Thlipsis hoia ou gegonen (cf. Mt. xxiv.21, Mc. xiii.19). Th. repeats the subject with the view of preventing ambiguity; in the sequel LXX. (as handed down to us) overlook gvy, while Th. adds en te ge or epi tes ges. Hupsothesetai LXX.; Bevan suggests a corruption for eksothesetai or some other compound of sothesetai; but hups. may be a gloss upon the tamer word which stood in the original. Th. rightly, sothesetai. Hos an heurethe, hnmts' -- overlooked by Th., unless we accept the reading of AQ, ho heuretheis [ho] gegrammenos.2. en to platei tes ges, LXX.; en ges chomati Th., Heb. 'in the ground of dust' (but see Bevan, p.201 f.). Diasporan kai aischunen, LXX.; diasp. is perhaps a gloss on aisch.; for the word see Deut. xxviii.25.3. Hoi phosteres tou ouranou, LXX. a reminiscence of Gen. i.14 (LXX.); cf. Sap. xiii.2. hoi katischuontes tous logous LXX., reading mchzyqy dvrym for mtsdyqyhrbym; Th. translates mhtsdyqym hrbym. Ta astra tou ouranou (LXX.), the ordinary Biblical phrase, used in iii.36, 63; Heb., Th. have 'the stars.' 4. Apomanosin (LXX.), didachthosin (Th.). Both senses have been found in the Heb.; cf. Bevan, ad loc. Plesthe he ge adikias LXX., reading rh or rt for dt.
The student who has gone through these extracts, or who is able to dispense with help of this kind, is recommended to begin the careful study of some one book or group of books. For several reasons the Books of Samuel (1 -- 2 Regn.) offer a promising field for work of this kind. They are on the whole the part of the Old Testament in which the value of the Septuagint is most manifest and most generally recognised  , and invaluable help in the study of both the Hebrew text and the versions is at hand in the commentaries of Wellhausen, Driver, and H. P. Smith  . But whatever book may be selected, the method and the aims of the reader will be the same. He will read the Greek in the first place as a version, and he will use all the means at his disposal for ascertaining the original text which lay behind it. But he will read it also as a monument of early Hellenistic Greek, and mark with growing interest its use of words and phrases which, originating at Alexandria in connexion with the work of translating the Hebrew Scriptures, eventually became the vehicle of a fuller revelation in the writings of the Apostolic age.
LITERATURE on the general subject of this chapter: Pearsoni praefatio paraenetica (Cambridge, 1665; cum notulis E. Churton, 1865); Hody, De Bibl. textibus originalibus (Oxford, 1705); Dr T. Brett, A Letter showing why our English Bibles differ from the Septuagint, London, 1743 (dated Oct.17, 1729); A Dissertation on the Ancient Versions of the Bible, London, 1760; Thiersch, De Pent. vers. Alexandrina (Erlangen, 1841); Frankel, Vorstudien zu der Septuaginta (Leipzig, 1841); Ueber den Einfluss der palästinischen Exegese auf die alex. Hermeneutik, 1857; Geiger, Nachgelassene Schriften, iv.73 ff. (Berlin, 1875 -- 8); Selwyn, art. Septuagint in Smith's D. B. ii. (London, 1863); Wellhausen, do. in Encyclopaedia Britannica (London, 1886); W. R. Smith, Old Testament in Jewish Church (1881, ed.2, 1892); Hatch, Essays in Biblical Greek (Oxford, 1889); Driver, Notes on the Books of Samuel, Intr. (Oxford, 1890; second ed., 1913); Buhl, Kanon u. Text des O. T. (Leipzig, 1891); Nestle, Marginalien (Tübingen, 1893); Streane, Double Text of Jeremiah (Cambridge, 1896); Kirkpatrick in Expositor, April 1896: Redpath in A. J. Th. VII. (1903); the various Introductions to the Old Testament; Commentaries on particular books, esp. those of Dillmann and Spurrell (Genesis), Driver (Deuteronomy), Moore (Judges), Wellhausen, Driver, and H. P. Smith (Samuel), Burney (Kings), Mozley (Psalter), Toy (Proverbs), Ryssel (Micah), Oesterley (Amos), Ottley (Isaiah), Cornill (Ezekiel). A complete commentary on the LXX., or on any of the groups of books which compose it, is still a desideratum.
On the Semitic style of the LXX. the reader may consult the Eisagoge of Adrianus (Migne, P. G. xcviii. or ed. F. Gössling).
 The external evidence has been briefly stated in Part i. c. i.((p. 23 ff.).  Cf. R. Sinker, Some remarks on the LXX. Version of the Psalms, p. 9 ff.  Cf. e.g. Job 9:9, xlii. 14; from the latter passage Theodore of Mopsuestia argued the pagan origin of the book (D. C. B. iv. p. 939).  Moore, Judges, p. xlvi.  According to McNeile (Introd. to Ecclesiastes) it is the earlier edition of Aquila's version; cf. Thackeray, Gr. 0. T. pp. 13, 60.  On Job 33:31 see Thackeray, Gramm. O. T. p. 55.  Theod. has Aphphouerou in Daniel.  Cf. prol. to Sirach: ton allon patrion biblion.  A. F. Kirkpatrick in Expositor, v. iii. p. 268. Cf. W. R. Smith, O. T. in Jewish Ch., pp. 75 f.  Prol. ou gar isodunamei ktl.  Even in Palestine "before the Christian era . . . the exegetical tradition was still in a rudimentary stage" (Kirkpatrick, Divine Library, p. 69).  Dr Nestle points out that the mistakes of the LXX. are sometimes due to Aramaic or Arabic colloquialisms, and gives the following examples: Aramaic: Numbers 24:7 exeleusetai. Psalm 140:4 prophasizesthai. Hosea 2:23 (25) egapemenen, vi. 5 apetherisa. Isaiah 4:2 epilampsei, liii 10 katharisai. Jeremiah 38.(xxxi.) 13 charesontai. Arabic: Psalm 83:7 dosei. Daniel 7:22 (LXX.) edothe.  See W. R. Smith, O. T. in J. Church, pp. 56 f.; Driver, Samuel, p. xxxix.; Kirkpatrick, Divine Library of the O. T., p. 64. Among the Rabbis of Jamnia were Eleazar, Joshua, and Akiba, the reputed teachers of Aquila; see Edersheim-White, History of the Jewish Nation, pp. 132 ff., 174 f.  See pp. 39 f.  ktv mrb, or, as the Talmud calls it, k 'svryt; see Driver, Samuel, pp. ix. ff.  Except perhaps those which lay before the translators of the Pentateuch; see Driver, l.c.  A specimen of such a script, but of much later date, may be seen in Driver, op. cit., p. lxv.  Cf. Streane ad loc. and on Jeremiah 20:17.  See Driver, op. cit., p. lxxxvi., or Hastings' D.B. iii. art. Moab.  Driver, op. cit., p. xv.  Jerome in the last years of the 4th century knows nothing of a system of vowel points; see Nowack, Die Bedeutung des Hieronymus für die ATliche Textkritik (Göttingen, 1875).  Cf. Driver, op. cit., p. lviii.  Thus Hatch and Redpath take note of 39 transliterations, exclusive of proper names, under A alone. They are thus distributed: Pentateuch, 4; Histories, 26; Psalms &c., 3; Prophets, 6. The principles by which the LXX. appear to have been guided in these transliterations of Hebrew consonants and vowel-sounds are expounded by Frankel, Vorstudien, p. 107 ff.  Unless the a is here prothetic, which is however less probable.  Cf. Hieron. Quaest. hebr. p. 44 (ed. Lagarde), De situ et nom. pp. 106, 158. Pearson (Praef. paraen. p. 6) endeavours to defend the LXX. even here.  Similarly in Proverbs 22:10, where the LXX. read vysv vt dyn, the last two words are rendered en sunedrio.  See W. R. Smith, O. T. in J. Church, p. 77. Aquila, as we gather from Origen and now know from his published fragments (p. 39 f.), wrote the word in archaic Hebrew characters, which however were read as Kurios.  "Because salt as well as frankincense was used in the actual ritual of their period" (W. R. Smith, op. cit., p. 77).  On xxiii. 11 see p. 17.  "An evident attempt to shield the priest from the charge of harshness" (II. P. Smith, Samuel, p. 10).  The example is suggested by Dr Hatch (Essays, p 18), who gives many of the passages at length. The index Hebraeus at the end of Trommius will enable the student to add other instances (besides didonai and its compounds).  Philo has apeleusokai (see below).  Or, as Dr Nestle, suggests, it may have been taken as introducing the acc., as in later Hebrew or in Aramaic.  Driver, ad loc.  In this passage the text of B in O.T. in Greek, i. 489, should be compared with that of A (ed. Brooke and McLean)  "Of the versions only [Vulg.] comes near the true sense" (Moore). Jerome renders pulcherrima feminarum.  The paraphrastic character of the reference appears more distinctly in the second stanza ek sou . . . Israel, which blends Micah 5:1^b, 3^a. It will be observed that cod. A reads hegoumenos with Mt.  W. R. Smith, O. T. in J. Church, p. 83.  If the student prefers to begin with Genesis, he will learn much as to the LXX. version from Spurrell's Notes (ed. 2, 1898). For more advanced study Proverbs will form a suitable subject, and here he may seek help from Lagarde's Anmerkungen, and Professor Toy's commentary in the 'International Critical' series.
 Cf. R. Sinker, Some remarks on the LXX. Version of the Psalms, p. 9 ff.
 Cf. e.g. Job 9:9, xlii. 14; from the latter passage Theodore of Mopsuestia argued the pagan origin of the book (D. C. B. iv. p. 939).
 Moore, Judges, p. xlvi.
 According to McNeile (Introd. to Ecclesiastes) it is the earlier edition of Aquila's version; cf. Thackeray, Gr. 0. T. pp. 13, 60.
 On Job 33:31 see Thackeray, Gramm. O. T. p. 55.
 Theod. has Aphphouerou in Daniel.
 Cf. prol. to Sirach: ton allon patrion biblion.
 A. F. Kirkpatrick in Expositor, v. iii. p. 268. Cf. W. R. Smith, O. T. in Jewish Ch., pp. 75 f.
 Prol. ou gar isodunamei ktl.
 Even in Palestine "before the Christian era . . . the exegetical tradition was still in a rudimentary stage" (Kirkpatrick, Divine Library, p. 69).
 Dr Nestle points out that the mistakes of the LXX. are sometimes due to Aramaic or Arabic colloquialisms, and gives the following examples: Aramaic: Numbers 24:7 exeleusetai. Psalm 140:4 prophasizesthai. Hosea 2:23 (25) egapemenen, vi. 5 apetherisa. Isaiah 4:2 epilampsei, liii 10 katharisai. Jeremiah 38.(xxxi.) 13 charesontai. Arabic: Psalm 83:7 dosei. Daniel 7:22 (LXX.) edothe.
 See W. R. Smith, O. T. in J. Church, pp. 56 f.; Driver, Samuel, p. xxxix.; Kirkpatrick, Divine Library of the O. T., p. 64. Among the Rabbis of Jamnia were Eleazar, Joshua, and Akiba, the reputed teachers of Aquila; see Edersheim-White, History of the Jewish Nation, pp. 132 ff., 174 f.
 See pp. 39 f.
 ktv mrb, or, as the Talmud calls it, k 'svryt; see Driver, Samuel, pp. ix. ff.
 Except perhaps those which lay before the translators of the Pentateuch; see Driver, l.c.
 A specimen of such a script, but of much later date, may be seen in Driver, op. cit., p. lxv.
 Cf. Streane ad loc. and on Jeremiah 20:17.
 See Driver, op. cit., p. lxxxvi., or Hastings' D.B. iii. art. Moab.
 Driver, op. cit., p. xv.
 Jerome in the last years of the 4th century knows nothing of a system of vowel points; see Nowack, Die Bedeutung des Hieronymus für die ATliche Textkritik (Göttingen, 1875).
 Cf. Driver, op. cit., p. lviii.
 Thus Hatch and Redpath take note of 39 transliterations, exclusive of proper names, under A alone. They are thus distributed: Pentateuch, 4; Histories, 26; Psalms &c., 3; Prophets, 6. The principles by which the LXX. appear to have been guided in these transliterations of Hebrew consonants and vowel-sounds are expounded by Frankel, Vorstudien, p. 107 ff.
 Unless the a is here prothetic, which is however less probable.
 Cf. Hieron. Quaest. hebr. p. 44 (ed. Lagarde), De situ et nom. pp. 106, 158. Pearson (Praef. paraen. p. 6) endeavours to defend the LXX. even here.
 Similarly in Proverbs 22:10, where the LXX. read vysv vt dyn, the last two words are rendered en sunedrio.
 See W. R. Smith, O. T. in J. Church, p. 77. Aquila, as we gather from Origen and now know from his published fragments (p. 39 f.), wrote the word in archaic Hebrew characters, which however were read as Kurios.
 "Because salt as well as frankincense was used in the actual ritual of their period" (W. R. Smith, op. cit., p. 77).
 On xxiii. 11 see p. 17.
 "An evident attempt to shield the priest from the charge of harshness" (II. P. Smith, Samuel, p. 10).
 The example is suggested by Dr Hatch (Essays, p 18), who gives many of the passages at length. The index Hebraeus at the end of Trommius will enable the student to add other instances (besides didonai and its compounds).
 Philo has apeleusokai (see below).
 Or, as Dr Nestle, suggests, it may have been taken as introducing the acc., as in later Hebrew or in Aramaic.
 Driver, ad loc.
 In this passage the text of B in O.T. in Greek, i. 489, should be compared with that of A (ed. Brooke and McLean)
 "Of the versions only [Vulg.] comes near the true sense" (Moore). Jerome renders pulcherrima feminarum.
 The paraphrastic character of the reference appears more distinctly in the second stanza ek sou . . . Israel, which blends Micah 5:1^b, 3^a. It will be observed that cod. A reads hegoumenos with Mt.
 W. R. Smith, O. T. in J. Church, p. 83.
 If the student prefers to begin with Genesis, he will learn much as to the LXX. version from Spurrell's Notes (ed. 2, 1898). For more advanced study Proverbs will form a suitable subject, and here he may seek help from Lagarde's Anmerkungen, and Professor Toy's commentary in the 'International Critical' series.