BS 16 65 A8 B 388267 UNIVERSITY OF MICHIGAN SIOURRIS.PENINSULAM.AMCENAMI 1817 W SCIENTIA ARTES VERITAS LIBRARY OF THE FOLURIBUS LYRIAUS UN TCERON CURCUMSPICE $1.000$... : د/ع (HIIT! MINIHILINITE RECEIVED IN EXCHANGE FROM Unix.of Chicago Librari THE HEBREW TEXT OF ZECHARIAH 1-8 COMPARED WITH THE DIFFERENT ANCIENT VERSIONS BY EIJI ASADA 1 TOKYO, JAPAN CHICAGO The University of Chicago Press 1896 BS 1665 5 A8 THE HEBREW TEXT OF ZECHARIAH 1-8 : COMPARED WITH THE DIFFERENT ANCIENT VERSIONS BY EIJI ASADA TOKYO, JAPAN CHICAGO The University of Cbicago Press 1896 A DISSERTATION PRESENTED TO THE FACULTY OF THE GRADUATE SCHOOL OF ARTS AND LITERATURE OF THE UNIVERSITY OF CHICAGO, MAY 1, 1893, IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY BS 1665 A8 (Reprinted from THE AMERICAN JOURNAL OF SEMITIC LANGUAGES AND LITERATURES, Vol. XII., Nos. 3 and 4. Chicago, Ill.] EXCIANTES RIVERSITY OF LORARY MAN 4 33 ( TABLE OF CONTENTS, INTRODUCTION. 1 - 5 LITERATURE. . 6 VARIATIONS IN GENERAL. 6 VARIATIONS DUE PRESUMABLY TO THE TRANSLATORS. 8 VARIATIONS DUE PRESUMABLY TO THE MSS. 19 VARIATIONS OF DOUBTFUL ORIGIN. 25 THE HEBREW TEXT OF ZECHARIAH 1-8, COMPARED WITH THE DIFFERENT ANCIENT VERSIONS. By EIJI ASADA, Ph.D., Professor of Old Testament Literature in the Aoyama Methodist Seminary, Tokyo, Japan, INTRODUCTION. The purpose of this thesis is to compare the Hebrew text of the first eight chapters of Zechariah with the ancient versions, and to examine the variations presented in the versions. In the presentation of the results, I have received suggestions from Workman's The Text of Jeremiah, and from Patterson's The Septuagint Text of Hosea. But I have tried, as far as possible, to consider the nature of every variation more carefully than Workman did, and to classify the variations more logically than Patterson. It is not the purpose to write a commentary on the book or notes upon the text, but simply and concisely to pre- sent the variations in the different versions and classify them according to their probable origin. Consequently there is no attempt made to explain all technical names and expressions com- mon in the works of textual criticism. The most important of all the versions is the Septuagint, and I have examined it more carefully than any other version. The LXX. of Zech. 1-8 seems to be the work of one man, per- haps different from the translator of the remaining chapters of the book. The translation is a very careful and excellent presen- tation of the original. But it is less literal than the LXX, trans- lation of other portions of the Old Testament, and presents many interesting variations. There are cases of suggestive additions, of careless omissions, of free paraphrase, and of unintelligible translation. The next in importance is the Vulgate, which gives a very accurate and faithful translation of Zech. 1-8, and con- 5 6 THE TEXT OF ZECHARIAH 1-8 tains fewer variations than the LXX., the Peshitto or the Targum. Therefore, it seems that the MSS. used by the Latin translator were not much different from the MSS. underlying the present Massoretic text. The Targum Jonathan of these chapters, like all other Targumim, is full of paraphrases and interpretations. But it furnishes many important suggestions, and, in a few cases, gives a better reading than that of the Massoretic text. The Peshitto of Zech. 1-8 is also useful for textual criticism. True it is that the Syriac translation is, in general, free, obscure, and inaccurate; but many of its variations are to be accepted in preference to the Massoretic text. Besides these four chief ver- sions the Arabic version has been consulted, which differs but little from the LXX., and the valuable translations by literal Aquila, cautious Theodotion and clever Symmachus. LITERATURE. For the constitution of the text the following books and editions have been used and consulted: Baer and Delitzsch's edition of the Hebrew text, Tischendorf's sixth edition of the Septuagint, and the texts of the other versions as found in the London Polyglot, Origen's Hexapla, and Stier-Theile's Polyglot. Some of the works constantly consulted are: Driver's Notes on the Hebrew Text of the Books of Samuel. C. H. H. Wright's Zechariah and his Prophecies. W. H. Lowe's “Zechariah” in Ellicott's 0. T. Commentary for English Readers. A. Köhler, Die Weissagungen Sacharjas, chap. 9.-14, Erlangen, 1861-2 Hitzig-Steiner's Die zwölf Kleinen Propheten. T. W. Chambers' "Zechariah" in Lange's Commentary. Maurer's Cominentarius in Vetus Testamentum. Driver's Introduction to the Literature of the 0. T. Keil's Minor Prophets. Briggs' Messianic Prophecy, etc. For the sake of convenience and simplicity, Syriac and Arabic words are written in ordinary Hebrew characters. VARIATIONS IN GENERAL. Variations are numerous, interesting and, in some cases, extremely peculiar. There are many cases in which the readings differ in respect to the tense of a verb. For instance, the trans- THE TEXT OF ZECHARIAH 1-8 7 lators give the present tense for the past (1:6 in LXX.), the past for the future (8:3, in Vulg.), the future for the past (7:13; 8:10 in LXX.), the future for the present (1:5 in LXX.), the present for 37 with participle (8:7 in LXX.), etc. The ver- sions present also a few changes in regard to the person and number of a verbal form; e.g., plural for singular (2:17 in Tar- gum), 3d pers. for 1st pers. (2:15 in Pesh.), 1st pers. sing. for 3d pers. plur. 8:8 in LXX.), 3d pers. plur. for 1st pers. sing. (2:15 in LXX.), etc. It may be noted also that a finite verb is given for a participle (1:8; 2:7 in LXX.), a participle for a finite verb (2:17 in Pesh.), a finite verb for an indefinite (1:14, 17; 8:21 in Pesh.) an imperative for an infinitive (3:4 in LXX.), etc. Not infrequently the translators change the form or construc- tion of a noun, violating etymological or syntactical principles or disregarding the sense of the passage and its relation to the con- text. The genitive is translated by the accusative (1:17 in LXX.), the nominative by the accusative (7:2 in LXX., Vulg., Targ., Pesh.), the accusative by the nominative (7:7 in LXX., Vulg., Pesh.), the vocative by the accusative (2:11 in LXX.), etc. The plural is given in translation for the singular in the Hebrew (4:12; 7:5 in LXX.), the absolute state for the construct state (7:9; 8:16 in LXX.), a proper noun for a common noun where it was difficult to translate (6:14 in Vulg., Targ.), a com- mon noun for a proper noun not familiar to the translator (7:2 in LXX., Pesh.), and a proper noun for another (5:11 in LXX., Targ., Pesh.; 7:2 in LXX.). A proper noun is sometimes mis- taken for a verbal form (6:10, 14 in LXX.), and in one case an untranslatable foreign word is translated, and that of course inex- actly (5:6 in LXX.). The pronoun also suffers from various changes. For instance, 2d pers. is given for 3d pers. (3:8 in Pesh.), plural for singular (5:5 in Targ.), an interrogative pronoun for another (5:5 in Targ.), etc. In one instance a cardinal number is rendered by an ordinal (1:12 in LXX. and Vulg.) In some cases one part of speech is given for another, e. g., a finite verb for a noun (1:3 in LXX.), an infinitive for a noun (4:7 in LXX.), a noun for a verb (7:3 in LXX.), etc. The form of a sentence is often 8 THE TEXT OF ZECHARIAH 1-8 changed, e.g., the Hebrew declarative is rendered as an interrog- ative (1:6 in Vulg.; 8:6 in LXX., Vulg., Eng., Pesh. [?]), and vice versa the interrogative translated as a declarative (1:12 in Vulg.), the interrogative is turned into the imperative (1:6 in LXX.), the declarative into the imperative (6:8 in Targ.), etc. Besides these, there are a great many more difficult and per- plexing variations. The addition and omission of letters, words, phrases, and sentences is very common; and their causes are various. We find also a few inadequate substitutions, and, in some cases, unnecessary repetitions. The arrangement of letters and words is often changed, and a new construction is given. Inaccurate or free translations are occasionally given, and the readings in the original text are obscured. All these variations may be classified in two groups: (1) Vari- ations due to the translators, and (2) variations due to the manuscripts. In the first division, I include those additions, omissions and variations of every other kind, for which the trans- lators are responsible; and under the second I classify those variations which existed in the MSS. used by the translators, those which are due to the condition of the MSS., and those which had their origin after the work of translation had been done; (3) variations of doubtful origin. In respect to some variations, I have found it extremely difficult to determine to which class they properly belong. It seems to be better to leave such variations unclassified than to attempt to theorize concerning their origin on the basis of mere conjecture. Therefore, I group them together under a third head as doubtful cases. I. 1. VARIATIONS DUE PRESUMABLY TO THE TRANSLATORS. 1. Variations arising from a different pointing.-- For 739299 (1:8) LXX. seems to have read b37a2 with Dâgesh in the 5, and renders των κατασκίων. . Pesh, follows this and translates 7,352227. Keil says that 539 is the form for “shady place.” Fürst compares the word with 780. Böttcher would read 17539. But Baer's reading 43342 (after Kimchi) seems to be best, and is supported by the Vulg. “in profundo.” — For niza" (2:4) * Wellhausen, Kleine Propheten, p. 173: "Die Bedeutung des Wortes ist unbekannt.” THE TEXT OF ZECHARIAH 1-8 9 T: ;7:9) בלשפּט For LXX. reads niz and renders eis xelpas. This makes the passage meaningless.—1912 (2:13). This reading is supported by the LXX., toîs douteúovotv aŭtos; Pesh, reads 177723. But Baer gives the correct reading 07 7235.- For 27 (4:13; 6:4) LXX. reads , but the reading accepted by the Mass. Text, Vulg., Targ. and Pesh. is to be preferred.— For 7377 (7:3) LXX. reads 1977 and gives tò dyiaoua “the holy place.' (7:9; 8:16) LXX. reads beva.— For Tee 75 (8:20) Vulg. reads 7 nux and renders “usquequo.” 2. Variations arising from a different grouping or transpo- sition of words. Some of the variations in this class are inten- tional changes made by the translators, and a few are due to the corruption of the text. But most of them seem to be due to the careless and hasty work of the translators. In 1:5 Pesh. connects to X271 with the preceding sentence, and destroys the beauty of the Hebrew parallelism.—Pesh. places 1022" (1:11) immediately after 39, but the Massoretic order is to be preferred.-773 (at the beginning of 1:17) is connected by LXX. with the preceding verse.—In 2:6, Pesh, transposes the words an and 22x, but other versions agree with the Hebrew. – Pesh. places (3:2) at the beginning of the address, i. e., immediately after 10 2017 8.--- For 21751 (6:11) Vulg. reads 5051 211.-In 8:13 Pesh. places 1870 5x at the end of the verse, and spoils the rhetorical beauty of the whole passage (8:9-13) which, in the Hebrew, ends as well as begins with the same words, -507-7-pirin.-- In 8:15, Vulg. transposes aw1778 ) . 3. Variations arising from ignorance, disregard, or an unsuccessful presentation of Hebrew idioms, or from a viola- tion of Hebrew syntax.—While some allowance must be made for the difference of idioms and syntax in different languages, one cannot overlook those variations which could have been avoided, if the translators had been more faithful to the original text. LXX. attempts to give the force of the cognate accusative 93D (1:2), by rendering úpyioon ... ópynu merálny, which is somewhat awkward.—- For 07287 (1:3), Pesh. gives the actual impv. form 2x, and fails to present the force of the 1 .את בית יהודה and קצף 10 THE TEXT OF ZECHARIAH 1-8 והלבש followed by ,העברתי .express the peculiar force of the perf אלה שני בני For .והלבשן original_force . Wellhausen reads שני בני היצהר האלה TapeoTeact . This rendering would be for consecutive. But the Heb. is more idiomatic and therefore preferable.—LXX. renders 7223 (1:8) by ciotýkel, which is less vivid than the original.- For 7w by T (1:12), LXX. gives TOŪTO Èßdounkootòv ēros. Vulg. follows LXX., and translates “sep- tuagesimus annus." But in view of Targ., Pesh., and the Heb., we must reject the LXX. reading, which does not suit the context so well.— Targ. renders 721 (1:17) by yun. This is impos- sible, because the verb is not followed by the preposition 3, but by 3.- For 22377 (3:4), LXX. gives kaì èvdúcare, and fails to . , (cf. Harper, Hebrew Syntax, $ 28, 4, a). Targ. and Pesh. pre- sent the sense of these words fairly well, though they weaken the . .-- b7991 77377 (4:14) LXX. gives oroi oi dúo vioi rîs mórytos . I 09729.—For 7738) (5:2), LXX. has jxewv, and Vulg. “cubi- torum," both of which renderings fail to express the force of the preposition 2. Targ. and Pesh. omit the preposition altogether. --For 3 na DRT (6:5), LXX. gives ékfoPEVOvtal mapaotîvai ; Vulg. “egrediuntur ut stent”; Pesh. 47 12-p7. But all these versions utterly fail to give the original meaning.-- For 25 p (6:8), LXX. gives the extremely literal translation yêu Boppa.---- 1723 (6:11), plural in form and singular in sense, is incor- rectly rendered by LXX. otepávovs; Vulg. "coronas”; Arab. Targ. gives the compromising translation 27 3035, but Pesh. has the simple $33. The same word in 6:14 is again taken by Vulg. as plural, but by LXX. as singular. See Well- hausen, 179, on this verse.--In b:a77 x 2757 (7:7) LXX., Vulg. and Pesh. disregard ng and take » 277 as the subject of the verb "to be" understood. Wellhausen reads 7* for nx. -Vulg. renders bas (7:12) by “cor suum,” failing to express the collective idea of the pron. suffix in the original.-LXX. renders 22x7 78 (8:3) by móds åndevń without the article. Wright translates "a city of the truth,” without ascribing the absence of the article to the syntax of the construct state. But Targ. has Roup7 xnp.-27P 7 (8:3) is rendered by LXX. õpos dylov without the article. But Targ. and Pesh. give the correct . .אַכָאליל .טורא קדישא translation ... - THE TEXT OF ZECHARIAH 1-8 11 is (1 : 4) כא-.are better אמר .and the Syr ,אָמר .The Targ .is onmitted by LXX (3 : 9) הנה--.מתתעיר by כעור Pesh . renders לאמר ויען 4. Variations which may be ascribed to carelessness and inaccuracy of the translators.—Under this division may be included many of the omissions and additions of unessential par- ticles, conjunctions, adverbs, pronouns, etc. For instance: The LXX. déye for ox3 (1:3, 4, 16; 2:9, 10, 14;3:9, 10; 5:4;8:6, 11, 17) loses sight of the peculiar force of the original word. . , (1:) omitted both by LXX. and by Pesh., but the general tone of such an earnest request as expressed in the passage favors its presence. For 1 (1:10; 4:5; 6:5), Pesh. gives XY without the conjunc- tion before it.- LXX. omits 8 (1:12) and fails to present the emphatic force in the original. - For x3 (1:14), Pesh. gives 7287, which, of course, is wrong. So also in 1:17.-In 2:17, ,(:9) . —,2777 (4:2) is omitted by Pesh.— In 4:6 and 7283 (twice) are omitted by Pesh.--Pesh. renders 87 (4:11, 12) without the conjunction and destroys the idiomatic Hebrew.-- Op 27 (4:12), which is the noun-predicate of 728, is connected by Pesh. with anos, confusing the gender. Symmachus also presents this error.— 687 (5:7) is omitted by Pesh.— In 6:3 Vulg. read 287 09772.*—The second 2x5 in 7:3 is omitted by Pesh.— For "77" (7:13), LXX. incorrectly gives kai čotal. This error affects the LXX. translation of the following verbs.—-Vulg. transposes 708 in 8:9.- From "nan x37 (8:14), Pesh. omits 25 and renders mp3.787. (So in London Polyglot, but Lee's edition has 5).- From Domy 10 (8:20), Pesh. 7 " .In , , . 5. Obscure rendering and the omission of difficult words.- In many cases, the translators attempt to give the general sense of a passage, in which they find some word or words too difficult to render literally. This brings forth an obscure and sometimes unintelligible translation. It seems to be more common to omit difficult words altogether than to give an uncertain translation of them. . (1:) , * On 6:3 see especially Lagardə, Nominal-übersicht, 29 rm. LXX. y após, Targ. 720p, of ashy-gray color. Aquila káprepos, agreeing with Hebrew; Lagarde proposes 'to read ,8 : 21 In-. עבים as appositive to ישבי ערים and takes ו omits .לאמר for ויאמרן Pesh . Seems to have Tead and hesitates ,אילנא simply by (1 : 8) ההדסים Pesh , renders DX, "of whitish color." 12 THE TEXT OF ZECHARIAH 1-8 ושכנתי is difficult in (8:17) את כל אלה אשר שנאתי - שורא and מתביא צום הרביעי וצום החמישי וצום The meaning of----. אשר the word to express what kind of trees they are. See also the Syr. of 1:10, 11. LXX. translates regularly by õpn (cf. 6:1).- In 6:3, Pesh. seems to feel the difficulty connected with DSN, and omits the word altogether. Cf. 6:7.—In 6:14, Vulg. renders prosa by “et Hem," and Targ. also takes it as a proper name. But Pesh. omits the uncertain word 77, and substitutes *w939 1955 73. Cf. v. 10.-739 (7:2) part of a proper name, is ren- dered by LXX. and Pesh. as a common noun ; Baoileús, 35a. haw (8:3) is differently rendered by translators. Pesh. does not seem to be sure about the tense of these verbs, and avoids the difficulty by rendering both by the participles () construction. LXX. renders tauta návra éuionda, Theodotion adds å before éuionoa. Pesh. follows the LXX. Vulg. and Targ. have tried to translate the wx, but have failed to give the force of nx. On the other hand, LXX, and Pesh, have preserved the original construction of 38 50 Ox, and consequently neglected .-- wyb137 **2017 (8:19) must have been very obscure in the mind of the LXX. translator, for he renders motela ' tempàskal νηστεία η πέμπτη, νηστεία ή εβδόμη και νηστεία ή δεκάτη. But Aquila, Symmachus, and Theodotion understand the correct meaning, and translate rather inexactly νηστεία και του τετάρτου, και η του πέμπτου, Kai + To +85óto, Kai + Too 8ccuro. For 8 79 (8:20), Pesh. has 532, but LXX. omits . To avoid the difficulty, Henderson supplies 7'7" between the two words. 6. Explanatory additions.--When the translators think the original to be too concise, too elliptical, too figurative, too obscure or too anthropomorphic, they supply some words or phrases by way of explanation. After 2017 WH2 (1:1) Pesh. adds the phrase 4972 7h3. This seems to be quite a common phraseology of the prophets (cf. Ezek. 26:1; 31:1; Hag. 1:1, et al.), and it would not be unreasonable to suppose that the phrase may have existed in the original text. Köhler suggests that the word wh means the day of the new moon, i.e., the first day of the month. But it is doubtful that “the first day of the month” should mean more than “the beginning of the month.” Therefore, the phrase seems to THE TEXT OF ZECHARIAH 1-8 13 be an explanatory gloss; and even if it was in the original, we must be grateful to the editor for omitting it.-Before W92 (1:15) Targ. inserts "ay 3y, but this reading is not supported by other versions.— Before 8p (1:17) LXX. adds the extra sentence και είπε πρός με ο άγγελος ο λαλών εν εμοί. But this insertion seems to be out of place.—Vulgate explains 997 (2:4) by the additional phrase, “per singulos viros."— After 158 (2:8) LXX. supplies déywv, which is unnecessary. After 785277 (3:3) Pesh. adds 717", and makes its favorite phrase.--- In 3:4, Pesh. gives X3x32 as the subject of my04.–For 73 (3:5) Targ. and Pesh. seem ,Wellhausen .בזרים טובים or בגדים טהורים to have read either הזהב () כהן , למשמעני by להקשיב In 7:11 Pesh . renders-. חמישיא רשביעיא 176: bei 6732, vermisst man das adj. “rein."— Before 1767 (4:12 end), LXX. supplies tès étapvotpidas.— After 77 (6:13) Targ. adds 27.–For p7777 (6:15) LXX. has kaì oi paspàv år? aútớv. — "yawah "wana (7:5) is rendered by Pesh. 877 .. renders , which does not suit the context. -For Xp WXS (7:13) LXX. öv Tpórov cine. -- For 735 (8:2) LXX. seems to have read bwana3 772337, and renders ry 'lepovoadinu ka tnv Slóv. - Targ. interprets 53277 n8 (8:15) by D3w1na nan5.- For Dºzay (8:20), LXX. reads 0:27 29.- For777 Os opas (8:21, 22), LXX. gives éxĞntîoai tò a pównov kupio, and Targ. ” DTP 12 pa318 yana). 7. Double translation.— The translator gives, side by side, different renderings of single words, when he is not quite sure of the original meaning. For example: In 1:8, for o pow LXX. gives και ψαροι και ποικίλοι, which would show that the translator himself did not know the exact meaning of the word. Cf. 6:3. 8. Variations arising from misunderstanding or misinterpre. tation of a word or passage.- For 170 - (1:1) LXX. gives viòv 'Addù, thus making 7757 and 172 - stand in apposition. The translator seems to have taken Zechariah not as grandson of Iddo, as in Vulg., but simply as his descendant. So also in 1:7. Lowe, however, inclines to take the viòr as a corruption of vioù. — 777 11825 occurs forty-four times in the first eight chapters, and eight times in the remainder of the book of Zechariah (1:3, three times, 4, 6, 12, 14, 16, 17; 2:12, 13, 15; 3:7, 9, 10; 4:6, 9; 6:12, 15; 7:3, 4, 9, 12 twice, 13; 8:1, 2, 3, 4, 6 twice, 7, 9 twice, 11, 14 twice, 18, 19, 20, 21, 22, 23; 9:15; 10:3; 11:5; 12:2, 7; 5:4; : 14 THE TEXT OF ZECHARIAH 1-8 (1:17) תפוצנה. עד מתי sentence in the question introduced by .LXX ; יתמלין .Targ ; נסתרקן .is incorrectly rendered by Pesh 14:16, 21 twice). In all but three places, LXX. renders kúplos Travrokpátwp, and twice kúpios tûv duvánewv (1:9; 7:4). The Syriac translation X375777 corresponds to the LXX. Tavtokpátwp. All the attempted translations fail to give the original meaning and are no better than the mere transliteration caßaue (13:2). Vulg. gives “dominus exercituum,” which is perhaps the meaning in the original. — LXX. renders nie" (1:5) by týcovrat, and this is followed by Vulg. which gives "vivent." But the context requires the present tense, which is well expressed in the Hebrew. Vulg. takes the whole of 1:6 as a question, but LXX. changes the inter- rogative sentence in the verse into an imperative sentence with the verb décobe. --For 3 (1:6) LXX. gives évré douai without any sufficient reason.— For 270" (1:6) LXX. incorrectly gives kal återpiónoav.-That N (1:12) is rendered by LXX. ás útepeldes and by Targ. 15 727059 ann876. But Vulg. and Pesh. agree with the Hebrew. ---Vulg. translates w byw JT (1:12) by “Iste iam septuagesimus annus est," and does not include the -() . ; . "; dla xuońcovrai; but Vulg. gives the correct translation “affluent." Targ. fails to give the original sense of 2:11. — For D'aw97 (3:8) Pesh. has 7"22-27 737, which is not supported by any other version. — 7323 (3:8) is certainly a difficult word. LXX. ren- ders 'Avarolýv; Vulg. "Orientem"; and Pesh. 77. These trans- lators either take the word as an equivalent of Syr. Xa, or read 12; of. Zech. 6:12, Isa. 4:2, Jer. 23:5, 33:15. Aquil. renders the word by αναφυή, and Symm. by βλάστημα. The last two seem to express the original most satisfactorily.— For Thea HAM (3:9), LXX. gives õpúoow Bópov, probably reading 19; Aquil. Äraylúbw ảvolymara aitas; Targ. 8 675 XX; Pesh. 7970 XIX ha. None of these readings seem to be better than the Hebrew, which is followed by Vulg. and Symm.–For "nw (3:9) LXX. gives yniaonow, and this is followed by Pesh.-LXX. seems to regard mwa (4:7) as an Aramaic infinitive from the root w", and renders Toll katopiwoai. But this word is undoubt- edly a noun, as we find $7025 in Targ. and anypa 78.838 in Pesh.; an imperative form of 77 has probably been omitted () . ליחדי .is rendered by Targ (4:10) ושבחו וראו --, למוישר before THE TEXT OF ZECHARIAH 1-8 15 (4:10) האבן הבדיל For-. ונחרון ונחרזון .and by Pesh כד יחזי .לכאפא דקודשנא .Pesh ; אבן בלשקולתא (4:12) שבלי-.is almost untranslatable הבדיל fact the Hebrew " . () LXX. gives ròv libov Tòv kagoiTépivov; Aquil. kagoirépov ; Symm. TÒN kExuplouévov; Theod. ápı@pow; Vulg. “lapidem stanneum;" Targ. ; . None of these trans- lations can express the exact meaning of the original; for, in () is rendered by LXX. kládol, and by Vulg. "spicae.” The former is better than the latter. — For 737 (4:14), LXX. gives this πιότητος; Αquil. στιλπνότητος; Symm. ελαίου; Theod. λαμπρότητος. - LXX. takes 52a (5:1) either as a feminine form of bana or as an equivalent of the Aramaic xbza, and renders Spétavov. In this it is followed by the Pesh., but Aquil. and Theod. render διφθέρα ; Symm, κεφαλίς or είλημα. LΧΧ. is certainly mistaken. - For 228 (6:3) LXX. gives Tolkio. Yapoí; Targ. 717935 795720p; Symm. and Theod. prefer medidvoí to rolkidor, but Aquil. takes the usual meaning of 52x and renders kaptepoí. In 6:7 D'S728,7 is rendered by LXX. and Targ. in the same way. But Theod. suggests toxupoí; Aquil. offers an emendation by giving auppoi; but Symm. strangely gives ouveobıymévol. —-7717 x 137 (6:8) is taken by Targ. as an imperative sentence.-73799 17987. n827 7270 nxzan (6:10) is rendered by LXX. Tapà Tôv αρχόντων, και παρά των χρησίμων αυτής, και παρά των επεγνωκότων αυτήν. The translator was either ignorant that these are proper nouns, or regarded them as symbolical names. A similar case may be noted in 6:14, where 737-57,7210351 D373 is rendered by Tols υπομένουσι και τους χρησίμοις αυτής, και τους επεγνωκόσιν αυτήν. (Codex A: ajrôv). - 777 80 777 (6:13) is variously rendered. LXX translates the word 777 by åperúv; Aquil. émidošórnta; another copy CŮM Pénelav; still another sótav; Vulg. "gloriam;” Targ. 17; Pesh. renders the whole sentence by Xhaw yap3 7777.- LXX., Vulg., Targ. and Pesh. take 5x n'a (7:2) as in the accusative of direc- tion.— For noas (7:3) LXX. gives év to oiko.—*720a7 ""272 (7:5) is rendered by LXX. év tais réurtais ñ év tais eßdójais. But Aquil., Symm., and Theod. translate èv to TÉUTTO kaì év to éßdóuw. The latter is the correct rendering. - For 177227 y 8 (7:14), LXX. incorrectly gives yîv eklektv. "bwh....nau (8:3); Vulg. attaches two different tenses to these verbs, and translates .. et habitabo.” Pesh, avoids the difficulty by "reversus sum # 16 THE TEXT OF ZECHARIAH 1-8 και κατα- .. ואשרי שכנתי in בפתהא by (8:5) ברחבתיה Targ . renders- לאלהים -.original . rendering both by participles. Wright regards the first verb as å present-perfect, and the second as a present. But this does not suit the context so well as the LXX. kai éTLOTPÉWw .... oknvóow, which Targ. practically follows, by rendering 2108 . (: order to distinguish the word from 2017 at the beginning of the verse. --The second half of 8:6 is taken interrogatively by LXX., Vulg., Targ., and uncertain in Pesh. Hitzig, Köhler and others object to it. yawa (8:7) is rendered by LXX. oów, but by Vulg. “salvabo.” The latter seems to be the meaning in the — . 3(8:8) is rendered by Vulg. “in popu- lum . . . . in Deum." This literal and unintelligible rendering shows that the translator did not understand the meaning of the passage.- For 72275 (8:9) LXX. gives åd' où ọkodóuntat. From this, Hitzig concludes that LXX. read 01272. Hitzig does not seem to have read the LXX. translation of the entire verse very carefully.—737 12 (8:10) is understood by Vulg. and Targ. to mean “on account of the affliction," but Pesh. gives the correct translation, *3138 OTP 10.-09burt 37 (8:12) is rendered by . , . Vulg. translation “semen pacis erit" seems to be best. -- For 7972 (8:13) LXX. gives év eúdoyiq and weakens the sense of the original. 9. Free translation or paraphrase.This is very common, biblical student knows, in Targ. and Pesh. The varia- tions in this class may be divided into two groups. (a) Cases in which the original sense is fairly presented. For $p....55P (1:2), Targ. gives by 173727....73977 (1:3) . 135 83085-For " 13"wp7 83 (1:4), LXX. gives où mposéo xov Toll cloakowoaí pov, and Targ. 23 n°3* *3. They seem simply to have paraphrased the same Hebrew text.— Pesh. paraphrases 1:: y no 7.-nupu naw (1:11) is rendered by Targ. 8730 xin Nupw. For "10s (2:10) LXX. has owátw. The meanings of the words are opposite to each other. But this is a clear case of paraphrase, because the scattering of the people from Babylon is practically the same as the gathering of them into Jerusalem. One But the .זרעא כהוא בשלמא ,and by Pesh , זרעא יהי שלם .Targ as every הוחרגז ואתפני במימרי Targ . has (1:3) ואשוב אליכם For .קדמוהי איכא אגון אבהי כון רכבי דלבא לעלם :the whole of 1 : 5 as follows : . THE TEXT OF ZECHARIAH 1-8 17 But this does not agree with the .תימרון נביא לא לעלמא ק־מין would expect something like ansap (Wellhausen, loc. cit., 175). LXX. paraphrases nysna (3:4) by modýpn.- For TIN (4:2), LXX. gives émrávo aütñs. (It is equivalent to the preceding 17958, Wellhausen, 177).—For 1912 (4:6) Targ. has "9227120.- Targ. takes 5173377 (4:7) as referred to Rome, and gives a very full paraphrase of the whole verse. For 73 87 (5:6), LXX. gives Tò mérpov, and Vulg. “amphora," both of which are inferior to Symmachus' transliteration oibi.- For 7377 (7:3), Vulg. gives "vel sanctificare me debeo,” and Targ. 1972352 293 Y5287.-For 1290 726 77 (7:3) LXX. gives mon ikavà črn.--- For 20 baby 720 (7:12), Vulg. has "et cor suum posuerunt ut adamantem.” -- For 22750 by (7:14) Targ. gives 242252.-LXX. ren- ders 877 58 (8:13) by Dapoeite, which is less exact than the Vulg. “nolite timere." Cf. 8:15.--- For wyn en b277,758 (8:16) Pesh. gives the free translation 1723 82709 377. (6) Cases in which the original sense is missed. Targ. paraphrases the second half of 1:5 as follows: 687 . context.- For 9057 17 (2:10), Targ. gives the paraphrase, . ferent from the Hebrew that I am inclined to regard it as a Targumic paraphrase of a different reading. At any rate, the reading is not in harmony with the remainder of the paragraph. In vv. 11-13, 14-16, 17, the commands are first given in the imperative form, and then followed by the reasons or grounds thereof, introduced by . Why should not v. 10 also have the same formula, seeing that its second half is a causal clause intro- duced by n? It is true that 950X introduces an imperative sentence, but it is not part of the divine message, which begins with box.--For 7 x 92 (2:13) Targ. gives no 79 For (3:3) . -For (4:14) Targ. gives N3727.-Targ. seems to disregard the suffix in D3 (5:6) and paraphrases Dp 53 128. — The Tar- gumic paraphrase of v. 7 is extremely free.- For 19 8 1727 (6:8), Targ. gives in the 1722 1773 max. Wellhausen, 179: " (6:12) Targ. gives the strange paraphrase, "200" bon't tuny. .This is so dif .אכלו למבדריא ואמרן להון אתכנשו בארעא כי הין Targ . has (3:3)היה לבש בגדים צואים For-. מהת גבורתי בני היצהר For-. ליה בנין דנסבין להון כשין דלא כשרן לכהונתא ומתחתיו יצמח For -. יניחהו one would expect the imperfect : 18 THE TEXT OF ZECHARIAH 1-8 הלא לאיטבא Targ . has (7:6) הלא אתם האכלים ואתם השתים For– This rendering is inferior to the elliptical .לכון אתון שתן construction in the original, and is favored by no other versions. is incorrectly rendered (8 : 4) ואיש משענתו בידו מרב ימים .Targ .וגבר עובדוהי תקניא יגכון עלוהי מסגי ירביא .by Targ כד תיקר דהלתי בעיני שארא דעמא wrongly renders part of 8 : 6 by .הדין ביומיא דאכרן אם קדמי ריקרון 10. Interpretation rather than translation. -- For 75923 (1:8) Targ. gives 5222, comparing "the shady valley” with Babylon. : : ,בבבל . (1:8) .Targ-.תקנין .is rendered by Targ (1:13) טובים- Targ . gives (2:1) קרנות For --.קרוי עמוי by (1:17) ערי renders which seems to be an interpretation , though the translator , בלכון כל Targ . gives (2:17) כל דשר For--מלכיות may have read (3 : 8) צמח which cannot be accepted . Targ . interprets ,רשיעיא אלין עמא דהוו Targ . has (5:6 ) זאת האיפה For --משיחא by .LXX ,(5:11) ארו שכער For –. נסבין ויהבין במכילתא דשקרא -.ארעא דגביל .and Pesh ; מדינת בכל .gives y Ba3u8vos ; Targ בכסלו For-.תשבחתא Targ . lias (6:14) העטרת In the place of which is no better than a mere ,דהו כברן Pesh . gives ,(7 : 1) הצום transliteration as given in LXX . , Vulg . and Arabic.– For .Pesh----.הצום תעני אתון מתענין קדברי Targ . has (7:5) צבתני אני For ---.ולמוסכנא ולדמתפנא לותי by (7:10) גר ועני interprets .kup8day air @ y dragay drc0a עוד LXX . gives at ,(7:12) ולבם שבו שמיר כאשר קרא For=.פוקדנא .is rendered by Pesh (7:12) הדברים- (8:22) עצומים. כמא דאתנביאו להון כביא Targ . gives ,(7:13) .רברבין .is rendered by LXX . Toad , and by Targ : : . . 11. The translators change the text, so as to avoid difficulties, or to suit their own interpretation. (1:), , For (), . own. But LXX. has kai karaoknvbrovolv, which does not suit the .(. LXX. omits 2017 (3:5), taking the last part of the preceding verse, as well as the first sentence of this verse, as Jahveh's address to the angel attendants. But this omission is quite inconsistent with the LXX. translation of the preceding verse. Wellhausen, 176, adopts reading of the LXX. (12-0). —For 77277 (:), . . or omitted ,אשר LXX . seems to have readl ,(1 : 6) הלוא For Pesh . evidlently read ,(2:15) ושכנתי the word altogether , -For . , .ושכן .is taken by Targ . as plural (2:17) הס-.context very well ,(3 :) ואבר . המה -.() , , . תאנה For-.אתם Pesh . evidently read ,(3 : 8) (3:10) גפן THE TEXT OF ZECHARIAH 1-8 19 but this rendering weakens the figure , גפנו ... תאכר Targ . read .Pesh - הזאת for האלה and ,מה for מי In 5 : 5 , Targ . read in והוא יבנה את היכל יהוה But LXX . seems to regard . . , in the original, which is a characteristic feature of Messianic speech (cf. Mic. 4:4).- For 72 (4:12) LXX. has év tais xepoi.- :, . a , omits 777 557 x 7.27 (6:12), supposing, probably, that the copyist added here by mistake the first part of the following verse. . , " 6:13 as an unnecessary repetition of the last sentence of the preceding verse, and omits it altogether. I think the LXX. reading is more plausible than the Syriac.--For 34 (7:2) Pesh. has 7301.--For D8257 (7:3), Targ. gives ***90.- For baba bauho 5x (7:10), Vulg. has "non cogitet in corde suo,' but the Heb. is more idiomatic and is supported by LXX., Targ., Pesh., and partly the Arabic.--For Xp WXS (7:13), Pesh. gives . correct.—-pua (8:5) is rendered by Targ. qhawn (cf. 2 Sam. 6:5). For 7hanna (8:5), some Greek manuscripts of LXX. (:) reads " nw.--LXX. renders 677,7 (8:12) by roll daoù nov TOÚTOV, which is not correct, containing an addition.-- In 8:15, LXX. adds kai before 172727, because the translator read 'naw (), . . This reading is very smooth and seems to be .על דקרית אכון .LXX (8 : 8) ושכני For - .ברחבתיהם are based on the reading .ודינא ושלמא Pesh . gives ,(8:16) ומבפט שלום For -שבתי for II. VARIATIONS DUE PRESUMABLY TO THE MSS. 1. Errors made by the copyists of the versions.- In this class I include those errors which are due not to the original Hebrew text, or to the translators, but to the copyists of the text of a translation. (a) Addition: For 10" (7:11), Vulg. has “et averterunt, which seems to be, as Wright suggests, a mistake of the copyists for “et verterunt.' (6) Omission: For 02057 8922 (8:7), some Codd. of the LXX. have simply dvopôv, but others add ydiov.— In 8:13, LXX. has ο οίκος Ιούδα και οίκος Ισραήλ. The omission of the article before the second oikos is to be taken as a copyist's error. (c) Repetition: For 77707 (7:12), LXX. has toù vóuov mov. The mov seems to be a repetition by mistake of the latter part of the preceding word, vómov. 20 THE TEXT OF ZECHARIAH 1-8 It is possible .עברי עברי טוב .and Targ ,ערים בטוב and Pesli . read (d) Alteration: For 2017 87577 (1:6), LXX. gives oi katedá- βοσαν. But it seems to be a corruption of oυ κατελάβοσαν. 2. Errors due to the condition of the texts used by the translators.— That the texts used by the translators were in quite bad condition is evident from the existence of those pecu- liar variations which could not have arisen, if the writing had been clear, full, and exact. Some of the causes of these variations are: (a) Omission of the final . “According to Lagarde, the three letters 1, 2, n, when occurring at the end of a word, were not written in the MSS. used by LXX., but represented by the mark of abbreviation () which already appears on Hebrew coins.” (Driver's The Books of Samuel, Introd., p. lxix). In my examination of Zech. 1-8, I have found at least one variation due to the omission of the final 0.- For 3102 mg (1:17) LXX. . 202 , and . . that the 2 of 21072 originally belonged to any, but it is more probable that the final was omitted, as usual, in the original MSS.; and LXX. and Pesh. seem to present the correct reading. (6) Confusion of consonants. Considering the condition of the ancient MSS. used by the translators, and also their method of translation, it is not at all improbable that some consonants were confounded with others. In some cases the confusion seems to have arisen from a similarity in form, and in others, from a similarity in sound. For yomo) (2:4) LXX. reads 7771575 and renders Toll ófüval. Schleusner thought that the LXX. translation has simply given the sense of the passage. But Vulg., Targ., and Pesh., though they do not give exact equivalents of the word, seem to have intended to translate ), which is certainly the correct reading.–The confusion of with 7 is quite common. In 2:6, LXX. fails to give the suffix of nx. In 3:9, Ann is rendered by LXX. without the suffix, In 4:2, LXX. and Pesh. again omit the suffix from 33. In 4:11, LXX. and Targ. do not give the suffix of 1722w. Again in 5:2, LXX. omits the . There is one case in which is taken for F1, viz: in 4:3, LXX. reads 757 for 73207. Wellhausen, : “, ורחבה Suffix from the suffix referring to ,מיבריכה read מימין הגלה for ** :177 M - THE TEXT OF ZECHARIAH 1-8 21 0774327; cf. 11; 5527 is a false paraphrase.”—For 17 (2:8), LXX. reads ans, and renders katakápows; Symmachus åreziotws; Theodotion eis marós. But Vulg., Targ., and Pesh. agree with the Hebrew. Also see Wellhausen, Kleine Propheten, 175.-For 09777 4285 (2:10), LXX. reads nnnn ya782. Several MSS. and Vulg. read anx. This reading seems to be better than the Hebrew, because 21777 278 simply means “the four direc- tions," and not the actual “winds." Wright, however, does not believe yan to have been the reading of Vulg. or Pesh., and holds that the latter, at least, has probably read 73783. The reading of an original MS. > for 2 (both being very much alike; see Riehm-Baethgen, Handwörterbuch, article “Schrift”) is very common. Mention may be made here of Hos. 9:7; Amos 5:8, 17; Mic. 1:2; Zech. 2:10; 6:14. This explains satisfactorily the LXX. translation ( = ) for Heb. 5. Wellhausen, loc. cit. 175, says: “One would expect something like yaxa."-- For 017 (2:17), LXX. has củlaßeíobw; Pesh. 377727, and Targ. 190. These translators seem to have read 7.-For the second bar (3:7), LXX. reads 6x7.— For 17p3 (5:3), Targ. gives "ps, which has perhaps, as Wright suggests, arisen from the confusion of pa (:), . ý ådikia atrwy. Wellhausen, 178, follows LXX. and in addition omits as a gloss X3779187 TXT 2284. Pesh. seems to follow LXX., and gives 4217. Symmachus' suggestion mpòs Toûto åtoßXétrovou is certainly based upon the Heb. by, but it is not an exact translation.--LXX. renders 7970 (8:10) by cotal, probably reading 77". Lowe thinks, however, that the LXX. translator read 772 as an Aramaic future. At any rate, the verb should be taken as a past, as in Vulg., Pesh., and in some MSS. of the LXX. (c) Corruption of the Text. Though the confusion of conso- nants is, in a sense, due to the corruption of the text, yet, under this special heading, I include those strange and remarkable variations which compel me to ascribe them to the fact that the original text was very badly corrupted and obscure. For 27 27 (1:6), Pesh. gives 779508. Perhaps the text was corrupt, but it is possible that the translator read 1220,7.-For "N1257 (1:15), Vulg. has “opulentas;” Targ. 1952 7707; Pesh. and renders d8ikia ערכם LXX . reads ,(5 : 6) עינם For-. ככה with --- 22 THE TEXT OF ZECHARIAH 1-8 ישיבו For=. בבעלה and the latter ,(ב)ברענן the former read -muss die Bedeu מהלכים :176 ,Wellhausen .הלך participle from qwann27; the English Version (both A. and R.) follows the Targ. But LXX. gives the correct rendering, tà ouvenitudéueva, which seems to be for D"82.7, as Schleusner well suggested.* - For $3" (2:7, after "2), LXX. seems to have read 23) and renders ciotýkel. But the Heb. is to be preferred, because it suits the context better.t-LXX. renders the second half of 2:12 by διότι ο απτόμενος υμών ως ο απτόμενος της κόρης του οφθαλμού αυτού, as if the text had read .. naza. This is followed by Targ. But Vulg. and Pesh. give faithful translations of the more vivid original Hebrew.— For 99321 (2:15), LXX. incor- rectly reads 1037, and renders karapeútovtral.-- For 23:22 (2:17), LXX. gives ék vepedôv, and Pesh. 822. It is probable that (2), – (3:5), LXX. has évidere.-0933.72 (3:7), a very difficult word, is rendered by LXX. Åvuotpepouévous, by Vulg. “ambulantes," and by Pesh. ;33772. Hitzig's objection to Gesenius' interpretation of the word does not prove that the form is an Aramaic hiph. , : tung “ Zutritt" haben; cf. Jer. 30:21.-- For 70877 287 (4:7), LXX. probably reads 570779 1787, as Schleusner supposed, and renders tòv lidov rîs kampovopías. Vulg. translates “lapidem pri- marium," and Pesh. also has X- WX9X3. 7027 is rendered by Aquila tòv mpwrejovta, by Symmachus tòv åkpov, and by Theodo- tion tòv a pôrov. Targ. gives the interpretation, 787 17-02 7227p5a 1790. Weighing all these translations, we must reject the LXX. reading and adopt the Heb. as the original, though it is very difficult.--73 you you naon (4:7) is also very difficult. LXX. seems to have derived the word ni8wn from 70, and renders ισότητα χάριτος χάριτα αυτής. This is followed by Aquila's éél óra xápitos, and the Vulg. "exaequabit gratiam gratiae eius." Symmachus gives προς χάριν αυτής; Theodotion offers κατάπαυσις, Katámavois atrị ; Pesh. has 827777 401707. These translations sufficiently testify to the helpless corruption of the Hebrew text. Wellhausen: "Der Sinn der letzten Worte des Verses lässt sich nur muthmassen.”—LXX. renders b227072 (4:10) by oi émi- BAétortes, and this is followed by the Syriac ;74. But better is the Vulg. “discurrunt,” which is adopted in the English Version. * See, however, Wellhausen, loc. cit. 174, and Isa. 37:29. Wellhausen, loc. cit. 174, suggests 79 (cf. 3:5). THE TEXT OF ZECHARIAH 1-8 23 and renders das ,כמות or למות LXX . reads either ,(5 : 3) כמוה .בקרויכון and renders ,בעיריכם Targ . which seems to have read and renders וישבו בערים Vulg . reads ,(8:20) וישבי ערים For - - For 0175ya (4:12), LXX. seems to have read baby2.- For (5:3), . , éws Davátov. This, however, may be due to the omission of the final 17 in the original MSS. Tischendorf's text omits the second 7725. Vulg. has “sicut ibi scriptum est” for the first ,77733 7722, and "ex hoc similiter” for the second. Wellhausen: probably read 72 722 (= jio) “since how long." - For 7805 by (6:13), LXX. gives (kal čo tai iepeùs) é deštûv atrol. Wellhausen, 179, proposes to read 152299.-- For 1987 (7:3), LXX. seems to have read ,7 xa, as Wright suggests, and renders cloendudev úde. For 772727 (8:15), LXX. gives tapatéta yuae kai dave- vonuar. Wright's suggestion that the translator read "naw is plausible.-3yw2 (8:16) is supported by all versions but ., , . - (), . " et habitent in civitatibus." 3. Recensional variations. These are the variations which can be best explained by supposing the translators to have used MSS. more or less different from the MSS. on which our Masso- retic text is based. Some of the variations in this class are to be preferred to the Massoretic text, while others should be rejected. We note the following: (a) Errors made by the copyists of the Hebrew text. These are the deviations from the correct Hebrew text, which are solely due to the copyists of the Hebrew text, and which were adopted by the translators. (a) Addition: For 9770 58 (1:4) LXX. " (, ; ) . nixas, Pesh. following LXX. in 1:16 and 8:17. - Forma (1:16) LXX. reads "D'21.- After 2287 (1:6) LXX. adds apòs aŭròv, and is followed by Pesh. But other similar passages favor the Hebrew reading. (B) Omission: From 0.747235 (2:13) Pesh. omits 3.- In 4:2, the Kethibh 728must be a copyist's error (Wellhausen, 141). The Qerê suits the context better, and is found in many MSS., LXX., Itala, Vulg., Targ. and Pesh. From "nx pytay (6:8) LXX. omits "nx, and in this is followed by Pesh. - For X (6:10) LXX., Pesh. and Targ. (in London Polyglot) read X.-- From 6:12, LXX. and Pesh. omit 3x3 in both cases. -- 1823 (7:4) is omitted in some Codd., Targ. and יהוה LXX . reads (8:17 ;16 ,1:13) יהוה For -. ואל תהיו reads . 24 THE TEXT OF ZECHARIAH 1-8 . and בין ההָרים * LXX . seems to have read (1 : 8) בין ההדסים For * Pesh. -(y) Repetilion: For 7720 20 (4:2) LXX. and Vulg. read simply you. We are either to take these words distributively, or perhaps better to regard the second as a mere repetition by mistake of the first (so Hitzig, Ewald, Henderson). Köhler and Wright conjecture that there are two sets of seven pipes each. Briggs favors this view. But this interpretation does not seem to be more natural than to regard the second 1920 as a copyist's error. (Wellhausen, 176–7).--(8) Explanatory or marginal glosses, which crept into the text: For 1973 (1:1) Vulg. reads 73257 2703, as in Hag. 1:1 and 15.-- After "8075 (1:6) LXX. adds év veţuari uov. This may have been copied from 7:12. --After 7777 na (2:4) LXX. adds kai tòv 'Iopana kat- tatav. -- After D"270 D5 (8:19) LXX. gives kaì cúppuvońceobe. --(e) Changes made by the copyists to avoid difficulties or ambi- guities: For :Ox (1:6) some Codd. and Theodotion have vulv.- (1:8) renders åvà pérov tûv dúo opéwv; and in this is followed by the Arabic. Hitzig thinks that the LXX. translator may have read 97767.- For 7 * 17.(2:11) LXX, has eis Llúv, which does not suit the context. Lowe, however, compares this with a similar mistake in Ezek. 21:15. — For The (3:4) LXX. has aŭróv. Wellhausen, 176, . : (:2) . For 1874 (4:9) some Codd., Vuly., Targ. and Pesh. read On37a9 (plur.; so Wellhausen, 177).- For 9871 (5:7) LXX. has idov, and Vulg. "ecce”; Wellhausen, 178, adopts 777 from LXX. text, or simply 1. --For 65723 (8:23) LXX. reads 723 in both cases. Some copies have metà coll for the first, and metà úpôr for the second, and Pesh. is like this. But Vulg. and Targ. support the Hebrew.-($) Changes which cannot be casily accounted for: () . . For 7355 70p (6:7) LXX. gives kui étéßlenov toll Topeterbai, and other copies και εζήτουν, και επέβλεπον του πορεύεσθαι. It is possible that the translator read 1pany. -- For 093217 37 (8:12) LXX. seems to have had a different text, and gives y deitu eipávny. Wellhausen, , : .--() LΧΧ. gives the strange translation και συνελεύσονται κατοικούντες πέντε πόλεις εις πόλιν μίαν. But other copies have κατοικούντες μίαν εις μίαν. . --. נרות LXX . readls (4 : 2) כרתית For --. אתר proposes to read i -. ונצטבא .and Pesh ויתרעי Targ . gives (2:16) ובחר fo1 : For a (8:21) והלכו יושבי אחת אל אחת For --.כי אזרע השלום :reads ,181 THE TEXT OF ZECHARIAH 1-8 25 * (b) The original readings preserved in the ancient MSS. used by the translators. All the recensional variations are not cor- ruptions and incorrect readings, but some of them are to be pre- ferred to the Massoretic text, and seem to be the original readings. We mention the following: In 1:8, Pesh. correctly omits 72717.-- Before-6-p70 (1:8) LXX. and Pesh. have the conjunction 7.-- For 783917 5728297 (1:9) perhaps to be preferred, in view of the similar formulae in this para- graph (1:10, 11, 12, 13). — For yoxa (1:11) LXX. reads 333 1787.--In 1:13, LXX. adds , before 623 627.--In 2:2 LXX. and Pesh. read 72 7372.-In 2:4, LXX. omits 283 and gives após pé instead. "This reading agrees with the form of the similar passages in 1:9; 2:2, 6, 8, etc., and is probably correct. -- (:) . This suits the context remarkably well, and even adds a rhetorical force, and therefore I am inclined to take it as the original read- ing. - From 1027 (2:10) LXX., Vulg., and Pesh. omit the conjunction 1.-- For (2:15) LXX. and Pesh. read 15.--- In 3:1, LXX, and Vulg. read 5717987).-- For 7779 28" (3:2) Pesh. read 7777 7832 284 (cf. Wellhausen, loc. cit., 175).- For (3:) . 7273 .... as a parenthetical insertion. --For 62:58 (4:9) LXX. reads 738. This suits the context well, and seems to be the correct reading, though all the other ancient versions favor the Hebrew. --- Forbena (7:2) LXX. Targ., Pesh. and Baer read 387"as one word.-- Before 3 (7:10) LXX., Vulg. and Targ. add the conjunction 7.-- Before 283 (8:1) many Codd. and Pesh. read 58. In spite of the objection of the Massorah this seems to be the correct reading in view of 4:8; 6:9; 7:4, 8: .למתבר and renders לזרות Targ.seems to have read (2 : 4) לידות For ויאמר אליו Wellhausen considers .ערכיך LXX . readls (3 : 4) ערכך III. VARIATIONS OF DOUBTFUL ORIGIN. While there are not a few doubtful cases among the variations which have thus far been discussed, it is even more true of the varia- tions under this special heading, variations which are extremely difficult to explain. Their origin may be accounted for as: 1 26 THE TEXT OF ZECHARIAH 1-8 () ציון ,The translator .כנשתא דציון and renders ,בית ציון Targ . reads 1. Recensional, or a change made by the translator. For instance: In 2:4, 6*270727p ng is rendered by LXX. tà réocapa képuta. The Hebrew is to be preferred. - For 27 (7:2), part of a proper name, LXX. has 'Apleoegép. 2. Recensional, or due to the carelessness on the part of the translator. Note the following examples: For 7717 (2:4) LXX. reads 717".-- For 73 na (2:14) . , , however, may hàve been misled by the usual scriptio defectiva in the original MSS. 3. Recensional, or misinterpretation of the translator. Thus: For 287 (3:5) Vulg. and Pesh. read 28"), and, as the result, the former gives the duplicate statement that Joshua was clothed with new garments, and both present an unpleasantly abrupt change from the direct imperative 970077 to the indirect jussive 1210". It is best to follow the Heb. and Targ., and read 287, because it suits the context best and also strengthens the (:) . ý opelvn ; Pesh. seems to follow LXX. and renders X770. 4. Due to the corruption of the text, or an intentional change made by the translator. So we have: In 1:6, LXX. has an additional word dexeobe, which is proba- bly for 17p, as has been suggested; and this reading may have arisen from some confusion connected with the word "PM.--For 09 (8:22) Targ. has 352. The translator may have given his interpretation of the original. 5. Recensional, or due to the condition of the text: For D'X2% (1:5) Pesh. read X3. The absence of the final b in the original text may have misled the Syriac trans- .--(1:) . 1987.--For DTO (1:7) Pesh. gives X32").--In 2:9, Pesh. omits 75 and gives 1792 in its place. LXX . gives (7 : 7) הנגב For -- .בגדים and צנים contrast between i Pesh . has (1 : 6) וישובו For.הנבאי for ינביאי lator to read .ואתרעיו GENERAL OBSERVATIONS. It is undoubtedly true that some of the explanations offered in this thesis are far from satisfactory. But, taking it for granted that most of them are correct or probable, it may not be uninter- THE TEXT OF ZECHARIAH 1-8 27 esting to observe some of the characteristic variations in the dif- ferent versions. Most of the variations in the tense of a verb are found in LXX. The changes from one part of speech into another are found only in LXX. and Pesh. Variations due to a different pointing are characteristic of LXX., but those due to a different grouping of words are rare outside of Pesh. LXX. has many additions, but Pesh. has only a few, and Vulg. none. Omissions are most numerous in Pesh., and half as many in LXX., but very rare in Vulg. and Targ. Variations arising from a violation of the principles of Hebrew syntax are found almost exclusively in LXX. Some explanatory glosses are given in Targ. and Pesh., but more in LXX. Obscure rendering is a characteristic of Pesh., and too literal translation is common in LXX. Paraphrase and interpretation are abundant in Targ., but most of the strange, inexplicable variations are found in LXX. Misinterpretations are quite numerous in all versions, but original readings are pre- served more in LXX. and Pesh, than in the other versions. EMENDATIONS OF THE MASSORETIC TEXT ON THE BASIS OF THE ANCIENT MANUSCRIPTS AND VERSIONS. .with Pesh ויען המלאך הדבר בי ויאמר אלי Read .9 .with LXX ודברים נחמדים Read .13 -fol ,לידות for לזרות following LXX .; and , לאמר for אלי Read .4 1: 8. Omit 777 with Pesh. and read bpw following LXX. and Pesh. . 11. Read 1787 353, following LXX. . 15. Read X297 with LXX. (but LXX. OVVenit idejéva) and see Well- hausen, Kleine Propheten, 174. 17. Read one with LXX. 2: 2. Read -773% 72, following LXX. and Pesh. , , lowing Targ 10. Omit the conjunction from 0, (80 also Wellhausen, loc. cit., 175) following LXX., Vulg. and Pesh.; and read and with several MSS., Vulg. and Pesh. 15. Read 73 for -), following LXX. and Pesh. 3: 1. Insert 17 after 77, following LXX. and Vulg. " 4. Read 70093, following LXX. (so Wellhausen, 175). 4: ; . 9. Read 75% with LXX. .with Pesh ויאמר מלאך יהוה Read .2 ,ושבעה with the Qeré ; and omit ואמר Read .2 :4 אליך 28 THE TEXT OF ZECHARIAH 1-8 5: 9. Read 77995, following one of Baer's MSS. 10. Read 1777 for 2017, following two MSS. 6: 6. Read X** with Ewald. Wright's objection to this emendation is not conclusive. 10. Read 7270 1821 with Baer, following some ancient Hebrew and Greek MSS. 13. Omit 777 570 721 8777, following LXX. 7: 2. Read buna (so Baer-Delitzsch) as one word, following LXX., Targ. and Pesh. 10. Read with LXX., Vulg. and Targ. 13. Read naap wab, following Pesh. 8: 1. Insert 3x before maxb, following many Codd. and Pesh. 9. Omit 5 with Hitzig. --- VITA. I, Eiji Asada, was born on the 22d of May 1865, in a small town called Hanaoka in the southern part of Japan. In the Spring of my sixth year I was sent to a public school at Tokuyama, four miles from my birth- place, and was graduated from the same in July 1879. Then I went to Yamaguchi High School, Hiroshima High School and Kyoto High School, from the last of which I was graduated in June 1883. In the following Spring I entered the First Higher Middle College, Tokyo, and completed the Science course in June 1887. After having spent one year in the department of Mathematics in the Imperial University, Tokyo, I came to the United States of America in order to take some theological studies. In September 1888 I entered the Theological Sem- inary of the Northwestern University, Evanston, Ill., and was graduated from the same in May 1891 with the degree of D.B. While in the school of Theology I became so interested in Semitic studies and Old Testament work that I attended Professor Harper's Summer school for five sessions and studied Old Testament literature and Semitic lan- guages. In 1891–2 I pursued the same studies in New York City under Professors Briggs and Brown at the Union Theological Seminary, and also under Professor Gottheil at Columbia College. In October 1892 I entered the Graduate School of The University of Chicago, and took further studies in Semitics and the Old Testament for one year under Professors W. R. Harper, Hirsch, R. F. Harper, Price and Goodspeed. UNIVERSITY OF MICHIGAN 3 9015 07464 5691