709.H- 3261: A A 6 2 : 1 i 5 \ 7 ! X == i. -n =^ > -^•"' '-»--^«^-- ii^ Bamstein The Targum of Onkelos to Genesis THE TARGUM OF ONKELOS TO GENESIS A CRITICAL ENQUIRY INTO THE VALUE OF THE TEXT EXHIBITED BY YEMEN MS?. COMPARED WITH THAT OF THE EUROPEAN RECENSION TOGETHER WITH SOME SPECIMEN CHAPTERS OF THE ORIENTAL TEXT BY HENRY BARNSTEIN. Ph. D. • LONDON 1S96 DAVID NUTT 270-271 STRAND. W. DRUGULIN LEIPZIG. DEDICATED TO MY REVERED TEACHER AND FRIEND THE Rev. Dr. M. Gaster to whom I owe an everlasting debt of gratitude for the instruction received at his hands and the constant interest he has shown in my welfare. 1187410 PREFACE. i HE purpose of this enquiry is to critically investigate the recension of the Targum of Onkelos preserved in the Yemen Mss. and to establish a comparison between these Mss. and the Western or European texts, which are at present best represented by Berliner's recent edition. Berliner did not avail himself, however, of these Eastern Mss. and his edition is therefore, to a certain extent, inadequate and can hardly be accepted as final. In order to establish a critical edition this enquiry goes into minute details of grammar, differences of vocalization and ortho- graphy, additions and omissions, interchange of particles and variant readings. Part I contains the general outlines of this research and the results arrived at. Part II shows the more detailed comparison and lists of examples, following the text of the Bible, chapter and verse. The investigation concludes with a few specimen chapters of the Targum text, according to the version given by the Yemen Mss. It will probably reflect the original Palestinian form. Instead of making an index I have given a detailed Table of Contents. It is my fervent hope that this humble effort may be appreciated by students of the Bible and may give an incentive to the true appreciation of the work of Onkelos. CONTENTS. PART I. CHAPTER I. History of the Text PP- i— 3- l) Importance of the Translation (l). — 2) Where it originated (2). — Travels (2).— 3) Effects of travels upon text (2).— Present texts (2), CHAPTER II. Ofikelos and his Translation . . pp. 3 — 5. l) Approximate date of work (3). —2) Character of his Tar- gum (4).— 3) Halachic and Haggadic elements (4).— 4) Public recital of Targum (5). CHAPTER III. Importance of the Yemen Mss. pp. 6— 11. 1) Hitherto their linguistic character alone studied (6). — 2) Origin of Superlinear vocalization (6).— 3) Targum never neglected in Palestine (8).— 4) The Babylonian Redaction (9). — 5) Differences of Yemen Mss. (10), 6) They represent the original Palestinian version (lo). — 7) Object of Investigation (10). CHAPTER IV. Description of Mss. . . . .pp. 11 — 13. CHAPTER V. Superiifiear Vocalization . . . pp. 13 — 17. l) Signs employed in this system (13). — 2) Appropriate to Ara- maic (14). — 3) Origin of system 114). — 4) Absence of Segol (15; 45)-^ — 5) Absence of Sewa Compositum (15; 45 — 46). — 6) No sign for Sewa Quiescens (16). — 7) No sign for Dages or Rafe (16).— 8) Approach to Biblical Aramaic and Syriac (17). — 9) Agreement with Biblical Aramaic (17). CHAPTER VI. Differences of Vocalization . . pp. 18 — 25. l) Less frequent use of vowel letters (iS). — 2) Superlinear system ety- mologicallymore correct(i8).-3)Interchange of vowels. Western texts Patah=YemenQames(i8— 1 9; 46); (7. Monosyllabic words and particles. {a) Second number in brackets refers to the pages of the cor- responding portion in Part II. VIII CONTENTS b. 3rcl Per. Sin. Fret. Peal of ^"v. c. 3rd. Per, Sin. Pret. Peal of Mediae Geminatae. «'. Present Participle, e. 3rd Sin. Pret. Feminine. f. Suffix of 2nd Per. Sin. g. Miscellaneous words. Western Qames = Yemen Patah (19; 47). a. Suffix of 3rd Per. Sin. Fem. b. Vowel of 2nd Radical of Verb. c. Little change in Pause. d. Vowel before Suffixes "3 and sj. e. Other instances. — 4) Western Sere = Yemen Hireq (19; 47), Western Hireq = Yemen Sere (19; 47}. — 5) Western Sureq = Yemen Holem (20; 47). Western Holem = Yemen Sureq (20; 48). — 6; The vowel of the Imperfect (20; 4S). — 7) Various interchanges (21 ; 48). a. West. Hireq = Yemen Patah. b. West. Patah = Yem. Hireq. c. West. Sere= Yem. Patah. d. West. Patah ==Yem. Sere. e. West. Patah =Yem. ITolem. f. West. Holem = Yem. Sere. g. West. Sere = Yem. Holem. h. West. Sureq = Yem. Hireq. i. West. Holem = Yem. Qames. j. West. Sureq = Yem. Qames. k. West. Patah =Yem. Sere. /. West. Patah == Yem. Sureq. — 8) Interchange of Vowels and Sewa (22; 48 — 49). a. West. Patah = Yem. Sewa. b. West. Qames = Yem. Sewa. c. West. Sere = Yem. Sewa. d. West. .Sewa = Yem. Sere. e. West. Hireq = Yem. Sewa. — 9) The Yemenite Pronunciation (22). — 10) General results (25). CHAPTER VII. Variations in orthography. pp. 25 — 28 49—53 a. Dialectical Variations, b. Use of vowel change to distinguish similar words. c. 3rd Pers. Plu. of s"!5 in 1. d. Interchange o letters, e. Other changes, f. nt;^n ■jra?. g. •z'-VTi-, Nhp. h. Mar ginal readings, i. Agreement of Mss. with Nehardean Tradition CHAPTER VIII. Grammatical Variations . . pp. 28 — 35 1) The Noun (28—33; 53 — 57). a. State, a. Absolute and Em phatic. /?. Absolute and Construct, b. Number, c. Gender. — 2) The Verb [t^T) — 35; 57 — 60). a. The Conjugations. «. Pea and Pael. /2. Peal and Afel. 7. Peal and Etpeel, Etpaal or Et tafal. 0. Pael and Afel. j. Etpeel and Etpaal. /'. The Tenses c. Perfect and Imperfect. ^i. Perfect and (Participle) Present 7. Imperfect and (Participle) Present, c. Other Differences. CHAPTER IX. Other Variations pp. 35—39. i) How additions and omissions may be accounted for (35 — 36). — 2) Additions (36—37; 60 — 62). — 3) Omissions (37 — 38; 62). — 4j Contractions (38). — 5) Interchange of Prepositions and other Particles (38 — 39; 62 — 65) [Hebraisms 65]. — 6) Variations in Suf- fix (39). CHAPTER X. Exegetical Variations pp. 39 — 42; 66 — 77. CONTENTS IX PART II. CHAPTER I. CHAPTER II. CHAPTER III Introductory XXXI (85) Upon the words . Remarks n^; bs; Onp; i)^3p; n^i.-^H^X . . pp. 43 — 45. Examples of the rules contained in Part J. pp. 45—77- Specimen Chapters .... pp. 7S — end. remarks (78). Genesis XVII (79}. XXVI (81). XLI (92). LIST OF ABBREVIATIONS. A •= Ms. Codex Montefiore. No. 502. B = „ „ „ „ 508. C = „ Brit. Museum. Or. No. 2363. Z? = „ Codex Gaster. No. 2. B. = Berliner's Targum Onkelos (Berlin 1S84). Bbl. = Babylonian. Bib. Aram. = Biblical Aramaic. D. = Dagges. Hag. = Haggadic. Hal. = Halachic. H. = Hireq. Ht. = Hatef. Ho. = Holem. L. = Levy's Chaldaisches Worterbuch (Leipzig 1S66). Luz. = Luzzatto's Philoxenos (Vienna 1830). O. = Onkelos. P. = Patali. Plst. = Palestinian. Q. = Qames. R. = Rafe. S. = Segol. S. = Sureq. S. = Sere. Se. = Sewa. Spl. = Superlinear. SbL = Sublinear. Trg. = Targum. I. HISTORY OF THE TEXT. I. It is universally acknowledged that of all the Trans- lations of the Bible that bearing the name of Onkelos enjoyed the greatest veneration from very ancient times; because it rests upon the Exegesis of the Tanaim, and is considered to represent the original traditional Interpretation. Its sanctity was enhanced when the Amoraim invested it with great authority by making it an official decree that it should be publicly read in the Synagogue side by side with the Hebrew original. The neglect of this Rabbinical decree was in later times greatly blamed by some of the Geonim. This neglect of the Targum was most noticeable among the Jews living in arabic-speaking countries , after the arabic language had supplanted the Aramaic as the Jewish vernacular. Among other authorities who have blamed the Jews for their laxity in carrying out the principle of ai^ir nnsi x^.p^ ^'i'si the names of Jehuda b. Qoreis, Samuel Hanagid and the Geo- nim Natronai and Hai stand out most prominently. Their admonitions had a salutary effect for throughout the middle ages and at the present time this precept is strictly adhered to, and this accounts for the veneration in which it is held by the Jews — a veneration which, it may be added, is shared by all students of Holy Writ, on account of the light which this translation throws upon many obscure passages in the Bible, as well as for its intrinsic merits '. ' Cf. Delitzsch ,,Zur Geschichte der jiidischen Poesie". Leipzig 1836, pp. 27 and 135. A 2. The Targum originated in Palestine, but was sub- sequently transplanted to Babylon where it enjoyed great esteem, coming, as it did from the Holy Land. Similarly we find the study of the Calendar and the Ritual pursued with great zest in Babylon, both of which are products of Palestinian learning. In Babylon , as we shall see more clearly presently, a fresh redaction of the text was made, which differed in many respects, from the original Plst. Version. From Babylon the Trg. travelled to Europe, first to Italy and then to Germany and Northern France, and also to Spain and Provence. 3. In the course of its travels, the text of the Trg. became more and more corrupt. Some time ^ after its arrival in Europe the original Spl. vocalization was exchanged for the Sbl. system which was used for the Hebrew language, a change which had the most pernicious effect upon the text. The evil was aggravated by the ignorance of the European scribes who inserted marginal glosses in the text and heaped errors upon errors until the text of the Trg. became in an all-but-hopeless condition. 4. The result of this is seen in the texts which are in common use to-day. These are far from perfect. Various expedients have been tried from time to time to improve the text with but indifferent success, the reason being that all the would-be amenders started from an incorrect stand- point. They all assumed that the European copies were based upon the original text, whereas they originate from the Bbl. Redaction. Buxtorf wished to remodel the text according to Bibl. Aram., quite ignoring the fact that a long time had elapsed between the Aramaic of the Bible and that of the Trg. and that the Trg. was composed under entirely different circumstances, and for an entirely different purpose. Lagarde thought he would be serving the interests of I At the beginning of the 12th century by Nathan B. Machir of Ancona, according to an epigraph of the Codex I2 of de Rossi (v. Berliner „Targum Onkelos". Berlin 1884, Vol. II, p. 134). — 3 — science in the best way by omitting the vowel-signs and confining his attention to establishing a correct and critical consonantal text, but, it must be remarked, he did not even attain this ideal. Berliner's edition of the Targum *, (B.) which is a reprint of the Sabionetta edition of 1554 suffers for the same reason as all other European Editions. His sources are all Euro- pean MSS. none of which exhibit the original text but the subsequent Babyl. redaction. It is undoubtedly superior to any of its predecessors, besides which the edition is in- valuable containing as it does the products of a life-study and embodying the ripe results of all B.'s predecessors in the same branch of learning. Before proceeding to the in- vestigation how our Trg. texts may be improved by reference to MSS. unknown to B. and hitherto all-but-neglected and which will serve as basis for our own investigations it is necessary to know something about O., his time and the general character of his translation of the Pentateuch. ir. ONKELOS AND HIS TRANSLATION. I. As to the date of this translation much difference of opinion has existed. Frankel- and Geiger^ place its date in the time of the later Talmudists, but by observing certain characteristics of the translation B. has fixed the time as contemporaneous with or immediately following the time of R. 'Aqiba; second half of the second century. He bases this opinion upon the Targumist's consistent avoidance of anthropomorphisms (which is also a characteristic of the LXX), the little necessity felt for aramaicising Greek words — these words being au conrant in the original Greek in 'Aqiba's time — and the Targumic embodiment of 'Aqiba's hal. and hag. principles*. I Berlin: 1884. 2 „Zu dem Targum der Propheten". Breslau 1872, p. 9. 3 Urschrift und die Ubersetzungen der Bibel. Breslau 1857, p. 164. 4 1. c. pp. 102— loS. Cf. also Zunz "Gottesd. Vort." Frankfort 1S92, p. 132. 2. As regards the character of this Translation, this is described with great clearness and minuteness by B. ^ Our Trg. is neither too literal nor too free; it shows a remarkable agreement with the Massoretic Text^; it always follows the Qere leaving the Ketib unnoticed^ a characteristic which is also noticeable in a MS. of Daniel with Spl. vocalization which came under my notice, and the importance of which will be considered in chapter VIII. The Targumist omits certain pleonasms which are to be found in the Hebrew texf*; he occasionally endeavours to imitate the sound of the original text 5, employs a rich store of synonyms, oc- casionally makes use of circumlocution and simplifies figura- tive expressions. Both the Tetragrammaton and u^nbn are rendered "'"'^, which was written "i^"' (until this was taken as a sign for the Trinity), of which various modifications, such as "^T, 2"'"', "'Z'^ are found. The reason of the Targumist's dis- like of the word c-nsx is found in the employment of this word for strange gods''. When d%"i^n "^^ occur together, the Targumist must naturally reproduce them as they occur in the Hebrew text. 3. According to Rapoport^, where we have a law which applies to the administration of justice the Targu- mist renders literally, but where a command occurs which affects each individual member of the community, the tra- ditional explanation is given. But so many exceptions to this rule are to be found, that it must be rejected, and that given by Adler^ substituted for it. This rule runs thus — Whenever the Targumist has to meet the contra- dictory views of Sectarians, or in those cases wherein the people have not yet attained a complete knowledge of 1 pp. 206—224 1. c. 2 1. c. p. 207. 3 1. c. p. 209. 4 1. c. p. 210, 5 1. c. p. 211. 6 1. c. p. 223. 7 That this is not always the case may be seen from Exodus 12, 12 where a'jU'^ "rfTS is rendered "K'^U'D r-yj. ^ In Kerem Eemed V, 223 and VI, 220 also in i-iai n)3Ki tithv (Prague: 1861) p. 11. 9 In the Introduction to his 15"? nm Wilna: 1874. the tradition, or when the Tanaim themselves differ upon the point in question, the Targumist gives the better of the two traditions, but where the tradition is unassailable he renders literally \ Singer, however, says we can lay down no general rule, but O. employs the Halacha in his translation where a literal rendering would remain obscure-. 4. Great care was taken to ensure the correct recital of the Trg., this being especially the case with the Trg. of the Pentateuch ^ A verse was given out by the Reader in Hebrew, which was immediately followed by its Aramaic translation rendered by the Meturgeman. ]More laxity was allowed with the recital of the Prophets, in which three verses were taken together before the Meturgeman recited their translation'*. Both Reader and Meturgeman had to preserve a reverential demeanour, in virtue of the solemn office which they held 5. As, however, a standard official translation had not then yet appeared, opportunity was given to the Meturgemanim to insert their own interpretation of the text. This privilege was occasionally abused and we find some of the Meturgemanim blamed by the Rabbis^ on this account. The writing down of the Trg. — just as that of the Misna — was at first prohibited", but this prohibition only applied to the writing down for the purpose of public reading, but private copies could be held by the scribes for their own use. I E. 1. c. pp. 224 — 225. 2 Singer: „Onkelos und das Verhalt- niss seines Targums zur Halacha". Halle 1881. 3 B. 1. c. p. 84. 4 Cf. Misna. Megillah IV, 4 and Talmud. Bab. Tr. Megillah 24 a. 5 Cf. Tur, Oral} Hayyim ch. 145 and Maimonides rhsr r'ihr: XII, il. In Tal. Bab. Tr. Megillah Cap. III. R. Nissim quotes a Jerusalemi- tan Talmud as the origin of this custom. ^ Cf. Talmud of Baby- lon Tr. Megillah 23 b and 24 a. 7 Cf. Talm. Jer. Tr. Megillah IV, I and Talmud Bab. Sabbath 115 a. v. Zunz. 1. c. p. 65. — 6 III. IMPORTANCE OF THE YEMEN MSS. 1. It is strange that hitherto only European MSS. have been studied in connection with the Trg. Since then how- ever a number of Targum MSS. have come to Europe from South Arabia but have received little attention; scholars having regarded them of linguistic importance only. In this way, one of them (which I shall subsequently call C) has been employed by B. \ Merx^ and Margoliouth j. 2. At the outset of our investigation into the importance of these MSS. we must ask ourselves two questions. Is it correct to call the Spl. system of vocalization the Babyl. system? Does the study of the Trg. in Babylon, or the redaction of a version of the Trg. in that country necessarily imply that its study was neglected in its mother- country, Palestine? Now the Spl. punctuation has been fre- quently described as the Babyl. system'^, a nomenclature which I 1. c. pp. 159 — 160. 2 "Chrestomathia Targumica", Berlin 18SS, pp. 68 — 103. 3 On the Supeilinear vocalization (Transactions of the 9th Congress of Orientalists II, London 1893), pp. 46—56. 4 Thus Neubauer connects the Yemen Jews with Babylon in his article on "The Literature of the Jews in Yemen" (Jewish Quarterly Review III, pp. 604 — 622). Speaking of the epitaphs which have been dis- covered in Yemen, Neubauer says (p. 608) : "These epitaphs, dated according to the era of the contracts point to an acquaintance with the Babylonian schools, for the Italian early epitaphs date from the era of the destruction of the 2nd Temple and later on bear, in addition, the era of the creation. The Jews of Yemen continue up to the present day to date from the era of the contracts, using Aramaic formulae, which point more to Babylonia than to Palestine. More- over, the use of superlinear vowel-points (usually called Assyrian Punc- tuation) in the pointed Hebrew Texts written in Yemen would argue their connection with the Eastern Massoretic Schools rather than with the Palestinian one at Tiberias". But Margoliouth is not at all cert.ain of the Bbl. origin of the Spl. vocalization. "The theory which identifies the originators of the superlinear punctuation with the Masoretic Schools of the "sn:-;'^ or "Easterns" rests partly on an epigraph in he Parmese Codex de Rossi 12 of A. D. 1311, and partly — or rather is also employed by B., but, as far as can be observed, no proofs have yet been given to show its Bbl. origin '. B. ^ says: "it is doubtful when the vowel-signs were first added but probably they were added in Babylon, in which country the peculiar Spl. punctuation was in use". No reason is advanced why we should make this assumption. Then why assume that they were added in Babylon? Is it not just as likely that they were originally written in Palestine and came over to Babylon with the Trg. text? This seems the more probable after Noldeke's assertion ^ that "the authoritative Targum although redacted in Babylon chiefly — on the internal evidence afforded by the readings and marginal rubrics of the Codex Babylonians" (I. c. p. 51). But he proceeds to point out that this is a very flimsy foundation for the origin of the vSpl. system. How can we trust a single scribe of the 14th century whilst the contemporaries of the originators of this system are silent on the subject. Margoliouth especially singles out Saadia and points out how inexplicable his silence would be were the superlinear system of Bbl. origin. As for the internal evidence from the Codex Baby- lonicus it must be remembered that this MS. does not exhibit the primitive Spl. system, but shows the Spl. punctuation in a highly com- posite and developed form which would naturally approach the Eastern or Bbl. forms of that jSIasoretic School. Hence it is hardly fair to argue from this MS. to prove the Bbl. origin. Margoliouth also refutes various other theories which have been set up to prove the origin of this system, but his own theory that this system represents a mixed nestorian-jacobite punctuation appears to be very forced, and is more than improbable if we are to admit this system to be of Palestinian origin. Cf. Seligsohn pp. 12, 19, 32. 1 Likewise Dalman: „Zwar ist die superlineare Vokalisation der von dort neuerlich nach Europa gekonimenen Targumhandschriften keineswegs als aramaisclie \Viedergabe palastinischer Aussprache des Aramiiischen anzusehen. Vjelmehr wird sie einer in Babylonien ge- lehrten schulmassigen Behandhing des Onkelostargums entstammen, der gegeniiber der Konsonantentext der sog. jerusalemischen Targume ofters eine iiltere und urwiichsigere Form des Onkelostextes reprasen- tiert" (Grammatik des jiidisch-palastinischen Aramaisch, Leipzig 1894, I, p. V — VI). 2 1. c. p. 131. j Th. Noldeke, Mandaische Gram- matik. Halle 1S75. Introduction p. V. — 8 — exhibits a dialect whose fundamental characteristics are Palestinian. For the present then, the system should be called the Superlinear (Spl.) system, without defining it further as either Bbl. or Plst. we shall see further on in chapter V how much more appropriate this system is to the Aramaic language than the Sbl. system now in use. 3. B. ' states that for various reasons but principally through the decay of the Plst. schools, the Trg. left its native country and its study was transferred to Babylon. But are we sure that the Trg. ever left Palestine or that its study was neglected by the Jews of the Holy Land? May it not have travelled to Babylon — just as the Calendar and Ritual did — and yet have been continued in Palestine? It is true that the Jews of Palestine were greatly harassed in their studies by political troubles which interfered with their liberty and yet we know that they were always the great leaders of the study of the Bible, a fact which the Massora — that invaluable guide to the correct Text of our Hebrew Bible — and the various Midrasim bear witness to. Now we know that both the Massora and the Midrasim hail from Palestine. Again, the Jerusalem Trg. which has come down to us^ although differing in character from the Trg. of O., shows at least that the study of the Trg. was pursued in Palestine. Now this Jerusalem Trg. is entirely hagadic in character, may we not then assume that side by side with this hag. Trg. there existed a literal Trg., the same version, in fact, which was transferred to Babylon? Just as Palestine produced two such varied studies as those of the Massora and Midras, it is probable that the two Recensions of the Trg. — the literal and the hag. — existed in this country side by side. This will perhaps be seen more clearly after it will be shown that Yemen MSS. appear to contain the original Plst. text. I 1. c. p. 108. — 9 — 4- The natural result of the attention bestowed upon the Trg. in Babylon is seen in that redaction of the text which although adhering to the original version to a great extent must have inevitably become somewhat corrupted in its language in course of time, in spite of the efforts of the Rabbis of the time to preserve the text intact by placing the stamp of their authority upon it. As texts became multiplied more mistakes must have gradually crept in. As the words were also, in all probability, differently pronounced in Babylon, their orthography became changed, and the ori- ginal vocalization must have been at least considerably modified. Another tendency became noticeable in the sub- stitution of hag. explanations in the place of the original literal rendering. In time a reaction set in in favour of the original Plst. Version; and this movement is represented by the so-called Nehardean tradition, in contradistinction to the tradition of Sura, which is substantially represented by the text which we find to-day in our European editions. When the Trg. came to Europe it was then punctuated in the original Spl. vocalization which it had probably received in Palestine. Some time after its arrival', this Spl. vocali- zation was altered in favour of the Sbl. system in use with the Hebrew text. This inevitably led to great confusion ^ for the Trg. text had already undergone considerable modi- fications after its arrival in Babylon, and now another alte- ration taking place w^hich practically involved the rewriting of the text in its entirety the result can easily be imagined. As copies were increased in Europe the evil became magni- fied and was greatly aggravated by the ignorance shown by the scribes, of the Aramaic language; the result of all this we have already seen in the various devices made by scholars to amend the corrupt text. These were all unsuccesful because the scholars depended upon the Babyl. recension I V. B. 1. c. p. 134. 2 ««Our editions are greatly neglected, whilst old editions and MvSS. have superior readings". (A. Geiger: Nachgelassene Schriften, Vol. 4, Berlin 1876, p. 104.) — lO — which must have been a considerable modification of the PIst. original. 5. Of late years a rich store of treasure has been brought to light by the discovery of MSS. of the Trg. from Yemen most of which are punctuated with the Spl. vocalization, a fact which in itself invites the student to their study if only from a linguistic point of view. On examining these MSS. more closely most far-reaching and profound differences are to be found. These variations occur not only in the vocali- zation, pronunciation and orthography but also in the exe- gesis of the text itself, especially in the interpretation of several important hag. and poetical passages, in which cases the usual hag. interpretation is replaced by a literal trans- lation. ' What then do all these peculiar facts point to? It is unlikely that the Jews of Yemen had a peculiar tradition of their own although we know that they settled in Sou- thern Arabia at a very early period. "Certain it is that centuries before Mohammed there were powerful Jewish communities in North Arabia as well as in South Arabia or Yemen" ^ 6. The various differences which are noticeable in these Trg. MSS. induce one to believe that they represent the original Plst. text. The changes in vocalization, pronun- ciation and orthography remind us — as will be shown in chapters V to VIII — of the Bib. Aram, and Syriac grammatical forms and approximate more closely to the Nehardean tra- dition, or we may perhaps call it the Nehardean recension. The preference shown for a literal translation may be ac- counted for by observing that the Plst. Jews had another Trg. which is entirely hag. in tendency. 7. The object, then, of this investigation is to endeavour I Geiger although ascribing a Babylonian origin to the Targum of Onkelos points out that its character is literality, which he regards as a later protest against the fanciful interpretations of the earlier exegetists. 1. c. vol. 4,p. 104. 2 v. Neubauer 1. c. p. 605. — II — to show the true character of the original Trg. of O. and so pave the way for a critical edition of that Trg. I will now deal separately with each of the points enumerated above, and show in how far these characteristics strengthen our views as to the original and Plst. character of the Trg. of these MSS. But before entering into the examination of these MSS., a short description of those I have made use of, may be now given. IV. THE YEMEN MSS. I. Codex Montefiore, No. 502; which I call A is peculiar in having a Sublinear vocalization for the Aramaic as well as for the Hebrew text, which may perhaps be regarded as an unsuccesful attempt on the part of the scribe to bring the two systems of vocalization into harmony with each other. It also differs from the other MSS. in having the Hebrew text in large square characters in the middle of the page, whilst the Trg. and Arabic translation are placed on the sides and the commentary of Rasi below. It con- sists of 133 folios and extends till the end of Exodus. The writing is bold, square and pointed; and the paper stout and dark. It is probably of the i8th century and is written throughout in one hand (except the marginal notes). The quires consist of 7 leaves, but nearly every leaf has the custos at the bottom. The number of lines in each p^ge, both in the Hebrew text, and in the Aramaic and Arabic translations at the sides, naturally varies according to the length of the commentary of Rasi below, but the average number is 19 lines for the Hebrew text, 30 for the trans- lations and about 8 for the commentary. The Hebrew text has an average of 8 word.s to the line. There are no mas- soretic directions except that at the end of every Parasa the number of verses contained in that portion is given with a mnemonic. It is one of the characteristics of Eastern MSS. — 12 — that they have little or no illuminations^ and all these MSS. (except C) have only an ornamental figure on the margin at the end of every Parasa. 2. Codex Montefiore, No. 508. I call this MS. B. This is a remarkably fine and clear MS. and will be frequently found to be valuable in retaining the original literal render- ing, whilst an haggadic interpretation is inserted on the margin. It consists of 159 folios, 4 leaves making a quire; the custos, being given at irregular intervals ^ an average of 24 Hnes to a page and 10 words to a line. Thick, oriental paper and the oldest part in bold and round hand. It is unfortunately in a very imperfect condition and is written in no less than four distinct handwritings, of which the third and fourth appear to be quite modern. The oldest hand extends from f. 3 — 87, 98 — 102 and 104 — no and is Spl. throughout, both Hebrew and Aramaic, it probably dates from the i6th or 17th century; the second hand (f. 78 — 79, 103 and III — 155), has the Sbl. vocalization for the Hebrew and the Spl. for the Trg. It is less round and bold; and is perhaps of the 17th or i8th centuries, the third (f. i — 2, 156 — 159) is much more minute and is Sbl. throughout; probably i8th or 19th century. The fourth hand has added numerous marginal readings as well as Genesis 44, 10 — 15. Many verses in the second hand are left entirely unpunctuated. Again the only Massoretic note is the number of verses contained in each Parasa. The MS. extends till Exodus 24, 2. Each Hebrew verse is followed first by its Targum. and then by its Arabic translation. 3. C is the British Museum MS. Or. 2363 and has been fully described by Dr. Berliner^. 4. Z> was placed at my disposal by the Rev. Dr. Gaster (Codex Heb. Gaster No. 2) and is the most perfect and reliable of all; scribes' errors being few and far between; and no sign being used either for Dag. or R. The paper is thinner I 1. c. pp. 132, 134 (Note 4), 137, 159 and 160. — 13 — than that of A or B, and the writing is beautifully clear and round; probably 17th century. It consists of 159 folios, written throughout in one hand and extends till the end of Exodus. There are about 27 lines to each page and about 14 words to each line. The MS. has been rebound, hence it is impossible to tell the quires, but every page has the custos. There are frequent marginal notes by a later hand; the only Massoretic notes being the number of verses at the end of every Farasa and the number of mmra and n'-awo at the end of each Book. A note at the end informs us that the scribe's name was nnn-^ -p xiirc ■- nao; whilst a notice inside the original cover seems to point that the MS. was in somebody's possession in 1809. It runs vhn mm •p-iirri (ri)x^i "f sbx "pnn raa -(rrxi "nx (? u:"nrt) iu:inb -paii -las (mnt:)ujb pir "im. This points to the year 21 21 of the Seleucidan era, corresponding to 1809 C. E. The last page is adorned with a cabalistic devise with a large figure of the Menorah, each branch of the candlestick having an appropriate scriptural quotation. Each verse is followed by its Trg. and then by its Arabic translation. All the MSS. are Folios and have been very much used. V. THE SUPERLINEAR VOCALIZATION \ I. We will first consider the system of vocalization used in these MSS. In the Spl. vocalization we have six vowel- signs which are written above the consonants. These are Qames" (Q.), Fatah" (P.), Sere" (S.), Hireq' (H.), Holem '(Ho.) and Sureq'^(S). Besides these signs a horizontal line represents the Sewa I This system has only been lately discovered, being brought to light by the Karaite Rabbi Firkowitsh about 50 years ago. It represents the older system, since it is more simple and primitive than the sub- linear system (cf. also the expression hy -ijjj) and it is unusual to regress from a well developed to a primitive vocalisation. - 14 — Mobile and in C a slanting line indicates the Rafe sign. A, B and C also make occasional use of the Dag. point within the letter. 2. This system of punctuation is peculiarly appropriate to the Aramaic language. The inevitable Hebraisms which must occur in a text punctuated with vowel-signs which were taken direct from those in use in the Hebrew language are not noticeable in the Spl. system. As an example, we may take, the absence of any particuliar sign to represent the Hatef. This semi-vowel is unknown in the Aramaic language and in Biblical Aramaic it must have been taken from the Hebrew. It is remarkable that in a Bible MS. which I con- sulted in the British Museum (Or. 2374), the only words in Daniel and Ezra which seemed to have a particular sign for the Sewa compositum are mp and 3"-P which are written thus 2"ip and 5"rp. This appropriateness of the Spl. system to Aramaic leads us to think that it was invented for the Aramaic language in Palestine as was suggested by Dr. Gaster in the course of his lectures at Montefiore College; and that it is, consequently, just as incorrect to apply the Spl. system to the Hebrew language — as some of the Yemen MSS. do — as it is to apply the Sbl. system to the Aramaic language. 3. Various suggestions have been made with respect to the origin of this system, but this is still a matter of con- jecture. Strack^ remarks "The so-called Babylonian or more accurately the Superlinear punctuation the vowel-signs of which are simplified forms of the matres lectionis x, 1 and "^ and the detached accents of which usually have the shape of the letters with which their name begins was in use among the non- Palestinian Jews of Asia". But no proof for this statement is forthcoming. We have already noticed that Margoliouth^ tries to prove that the system is of a mixed Nestorian-Jacobite character. I "Einleitung in das alte Testament". NSrdlingen 1888, p. 74. 2 1. c. p. 47. — 15 - 4- In tlie first place we must at once notice that the vowel-sign Segol is unknown in the Spl. system, other vowel- signs being substituted for it. In the Yemenite MSS. which Derenbourg saw and described in his "Manuel du Lecteur" ' P. was invariably used for S.; but in the MSS. which I used S. is replaced by P., S. and even by B ., the particular vowel being in most cases justifiable by reference to the word's etymology or by comparing the word in the cognate Semitic languages. This shows that the scribes must have been very careful in preserving the correct original orthography and is another argument in favour of the greater antiquity and authenticity of the Trg. as exhibited in the texts of the Yemen MSS.^ A few Instances are given here^, but detailed lists will be found in Part 11. S. is replaced by P. in 'ns'3 2, 8, rrs 11, 7. S. is replaced by S. in "i£x-iy 14,2, xfb-'ED 25,9. Very rarely by Pi.; e. g. in the word ai^nn 28, 17 (Greek iSicoxt];). 5. Besides having no S., the Spl. system of vocalization has no sign for Sewa Compositum. This sign — as has been previously remarked — is characteristic of the Hebrew and not of the Aramaic language'*. — In the MSS. it is usually replaced by Se. Mobile, or more rarely by Se. Quiescens or a Full Vowel. Se. Mobile replaces Ht.-P. in r-n i, 2. — It replaces Ht.-Q. in "^a^p 4, 10 and Ht.-S. in ts^JX 36,11. Se. Quiescens replaces Ht.-P. in ■|"'~n3 2, 10 and Ht.-S. in "•'in^b 17, 7. A full vowel is found in the MSS. where B. writes a Se. Compositum in such words as x^n 29, 34. Strange to say, Se. Compositum is found m a few iso- lated examples which have either crept into the MSS. by I Journal Asiatique. Sixieme S^rie. Tome XVI. No. 61. Paris 1870. 2 Cf. VI, 2. 3 In all examples which will be quoted, Berliner's edition of the Targum is the one referred to for the Euro- pean readings, but the readings quoted are those given by the MSS 4 Cf. also the absence of any sign in Christian Palestinian Aramaic for Se. Compositum, Noldeke, Beitrage zur Kenntniss der aramaiscenh Dialecte. ZDMG. XXII, p. 507. — i6 — error^ or, are exactly taken as they were found in the Hebrew text; the majority of these instances being Proper Names. A and D both write b-ipb^ 33, 1 8 (the other 2 MSS. B and C, more consistently b"-pi^"')^. 6. A third sign which is not represented in the Spl. vo- calization is the Se. Quiescens; which is also wantmg in Syriac. 7. The Spl. punctuation as originally written had no sign either for D. or R. and there is no doubt that originally the Aramaic language had no signs to indicate the harder or softer pronunciation of the letters rSDnsa. Margoliouth remarks ''The oldest-known MSS. only use the D. in the Hebrew, but not in the Trg., a fact which seems to show that the D. is not a part of the Spl. system as such% but that it was adopted into the text from the other system of Hebrew punctuation, which one may fitly call the Sbl. vowel- system. Or. 2363 has a special sign for the "^£"1 over the letters rs:n;2 but in later MSS. in which the Dg. is largely adopted in the Trg. this sign of the R. is dropped as being no more very necessary" j. He adds further"^ 'Tn the more composite developement of this system, D. is not only in- dicated by the form of the preceding vowel-sign but also by the point within the letter. This is really not necessary, and can only be explained by the adoption into the Spl. system of a feature belonging to the Sbl. punctuation". Of our MSS. A and £ unfortunately show a slight corruption in fitfully employing the D. point and C regularly uses the R. sign as Margoliouth has remarked. But D has neither the one nor the other. Many instances of the D. are to be found in A. This MS. although apparently using the Sbl. vocalization transcribes the Spl. into the Sbl. This may account for the frequency of the D. point in this MS. Here are a few exemples ""i^^i i, 7, P'^kI i, 24, ^^'^'"7" i, 26, "i?'n 1,26, xrn- 2,11, b^ir^r. 2,16, '■rr!r\_ 2,18, N^-Cirn 2,19, x^a-is 2, 23. I Cf. VI, 8. 2 Cf. Dalman 1. c. p. 46. 3 1. c. p. 46. 4 1. c. p. 49. — 17 — By the existence of so many examples in two chapters it might perhaps be thought that the sign was very frequently employed throughout the MS.; but the scribe breaks off suddenly and we find whole series of chapters which do not contain a solitary D. The following are a few examples from B. "i3^ i, ii, rn'nf2 i, i\, x-rr^ i, 28, h~r\ \, 29, -1:^'^ i, 31. ^ has none of these! In the specimen chapter appended to this investigation I have followed D which is the most perfect MS. and has neither D. nor R. sign. 8) The absence of all these signs, shows that the Ara- maic of the Yemen MSS. is much more closely akin to the Bb. Aram, and Syriac than is the case with our European editions of the Trg., and it may be seen how appropriate the Spl. vocalization is to the Aramaic language. 9) Our MSS. agree with Bb. Aram, in many grammatical points. a. 3rd. Pers. Sin. Fern. Suffix ends in n^-, Cf. Daniel 2, II nrn^. b. ist Pers. Sin. and PI. Suffixa of Verb are resp. "^33 and its 3 with P. Cf. Dan. 2, 9 ''rsyi'inn. c. 3rd. Pers. Plu. Fern. Perf. of Verb ends in n. Cf. Dan. 5, 5 nj^£3 \ d. 2nd. Pers. Sin. Suffix is punctuated with Q. Cf. Dan. 4, 22 -yhyi. e. Our MSS. punctuate many words with P. corresponding to the Bb. Aram, form with S. whilst B. punctuates with H. e. g. xabn ■p~b5 xitn rn-a. f. Miscellaneous words such as x:;?'>r x^sb^ ist 'I'nsrxi agree in the two dialects. I Or. 2374 writes thus, although Merx's edition of Daniel [Leipzig, 1882) has the Ketib *pE3 but Q'rd npE:. — i8 — VI. VOCALIZATION. 1. Before considering the changes which the pronun- ciation underwent on being transferred from the Spl. to the Sbl. punctuation^ it may be remarked that the Yemen MSS. do not show that superfluity of vowel-letters which is found in our Targum editions'. As B. remarks "this redundancy of vowel-letters points back to a time when no vowels were yet written and when the vowel-signs were subsequently added these letters should have been struck out, their retention giving rise to much confusion"^. As instances we may notice such words as "'37 and "'bli'n^J which B. writes iX3t and "'xVj'iS' where the x merely points out the a sound. 2. In considering the differences shown by the MSS. and the European-Babylonian texts it will be seen by tracing words to their etymologies that the former present a more accurate pronunciation. Now as it is extremely improbable that the scribes were acquainted with the etymologies of the words which they wrote, this shows that they must have been conscientious in preserving the correct original ortho- graphy. 3. B.'s edition of the Trg. has many words punctuated with P. which appear in the MSS. with Q. a. Monosyllabic words and Particles: na (Syriac >a^') 2, 9. nib' (Syriac loi>-) 2, 19. bi5 (Syriac ^) 3, 8. — b. 3rd Person Singular Preterite Peal of the Verbs I'r. n! 2, 3, nnT 6, 6. This corresponds to the Syriac Form. — c. On the other hand, the MSS. differ from the Syriac in the 3rd Person Singular Preterite Peal of the Verba Mediae Geminatae^ e. g. b^i 16, 4. Syriac would here punctuate the 'J with Petaha. But cf. p'l Dan. 2j 3S.^d. In the case of the Present Participle the MSS. I Likewise in Christian Palestinian Aramaic the matres lectionis are less used. (Cf. Noldeke I.e. p. 447.) 2) Berliner 1. c. p. 133. 3 For the contrary process we may compare the Hebrew ta Zechar. 4, 10 (Pret. of T-a) as if it were from TT3 and nu (for ntj) Isaiah 44, 18. — 19 — again show agreement with the corresponding Syriac form, e.g. ■'ir^T 4, 20, "^sn^T 4, 21, 'puVbT 17, 16.— e. The Q, of the 3rd Person Singular Feminine of. the Preterite of Verba K"b is difficult to explain \ Cf. n-ri 3, 20 r~,-- 4, 25. — f. The Spl. System is undoubtedly more correct in punctuat- ing the Suffix of the 2nd Pers. Sin. with Q. Cf- ~\^i 45, 9, ~\m 17, 5, -j-BsJ 19, 15. So also B. Aram. cf. "(i^ Dan. 6, 13. — g. Miscellaneous words, the majority of which agree with the corresponding Syriac forms. iV^x i, 11 (Syriac iJ^i), n:^ 8, 9 (Syriac IZIm). On the other hand B. has the Q. in many instances in which the MSS. write P.: a. Suffix of the 3rd Person Singular Feminine M^'^n 4, 12. This agrees with the B. Aram. form. Cf. fr^^s Daniel 4, 14, re-n-^ Dan. 2, 11^ — b. The vowel of the 2nd radical of a verb is fre- quently p. just as in the corresponding words in Syriac nian^ i? 29, isn^i 3, 7.-0. Since, as will be seen more clearly below, the pause has little influence in the Spl. punctuation herein showing the small influence of the Mas- sorah of the Hebrew text, our MSS. retain the P. in many cases where B. has a pausal Q. "in i, 5, -"^i^sV i, 22. — d. The vowel before the suffixes x: and *: is P. in the MSS. s<:^5^5 1, 26, ''ii'Siip^i 4, 14. — e. The absence of Se. Compo- situm may account for such instances as m£5 3^ 13, xSnIx 13, 8- 4. We have now arrived at the second pair of kindred vowels H. and S. Only a couple of instances are here given of words which are written by B. with S. but are found in the MSS. with H. and vice versa, but it may be 1 But Biblical Aramaic likewise r-j^s Dan. 7, 19 and r""2 ib. 7, 22. 2 Philippi tells us that the original form was probably Qames in Biblical Aramaic, ".Scribitur autem n-^ pro rr— quia, ut Masora Targumo On- kelosiani ostendit, sermo posterior a et a uti pronuntiatione sic etiam scriptione comniiscere consuevit. Vocalem luiius terminationis origine longum esse ex s~ — apparet, quod targumice in locum antiquioris ah successit". (v. Libri Danielis, Ezrae et Nehemiae ed. Baer, 1. c. p. LVIII.) 3 So also in B. Aram. Cf. v/?~5^. Daniel 7, 16. B2 — 20 — added that in the majority of examples which I have noted the reading of the MSS. may be verified by reference to the etymologies of the, words or to their corresponding Syriac forms. B. has S. with such words as in"'i<'i 2, ig, C|Vn 13, 2. But H. with 3^6n 3, 22, x^ipi 8, 22. 5. The 3rd pair of kindred vowels are Ho. and S. These interchanges appear to depend greatly upon the pronunciation of these sounds. Thus B. writes the words tiiw 3, 20, Ci'is jf, 21 with Sureq; and the latter punctuation seems, at first sight, more correct, but, it must be borne in mind that S. and Ho. are represented by one sign only in Jacobite Syriac. The Nestorians pronounced "^o-^oaj as if it were written with a Holem'; whilst, on the other hand, they pronounced vq\ ^ fti as the ordinary u sound". Perhaps the Yemenite Jews pronounced the words ms and fiiTi) — which are written with /Jj:v in Jacobite Syriac — as pom and som, but having a distinct vowel-sign for Ho,, they punctuated it with Ho. For B. Ho. and MSS. S. v. Part II. 6. Before noticing the other numerous interchanges among vowels in the two systems, it must be noticed that the vowel of the Imperfect seems to vary between I . and P. in the MSS. May we not trace the P. to Arabic influence? In that language the vowel of the Servile Prefix is Fatha. We might naturally expect that MSS. hailing from Yemen in South Arabia would present some Arabic characteristics and that this is the case may be seen by the following examples 14, 23 r^N (Cnrx), 15, 8 rnit but Cs.nx In these instances, I Cf. Duval, Grammaire Syriaque, Paris 1 88 1, pp. 47 — 48. 2 So also Noldeke in his article on Christian Palestinian Aramaic 1. c. p. 456 "The \p (On) in >^qa451 "you laugh" disagrees with the pro- nunciation in East Syriac. We should expect vj (un)". Dalman 1. c. (p. 63) remarks "The superlinear vocalization of the Targums, and usually also the Tiberian, has in the form of the Masculine a long as in B';s avj which however becomes u in the derived forms. The Tiberian vocalization clearly stands here in closer relation with the old Palestinian usage". — 21 — we see that C inclines more to the forms which we find in our Trg. editions, a fact which may be often noticed in the instances which will be quoted later on. 7. Variations in punctuation: a) B.H.=MSS. P. n'^Vin' i, 27, i^'crr 6, 20. b) B.P.=MSS. H.; rnni 4, 20 (but C rri5n), "j'=-an 9, 5 (but C and D 'iDrn). I . justifiable in both instances ^ c) B.S.=MSS. P. •^n\xn 4, 3, nbV-.nni 14^ 18. d) B.P.=MSS. S. sb'-E 15, 10. e) B.P.=MSS. Ho. -jiVi-^ 3, 16, np:-2n 16, 4. Both these are given by Levy. f) B.Ho.==MSS. S. qTr^ {A ^^-pt:) 4, 6. Here again the S. is more correct since C)ipr is a Noun; Levy also C)"')?~. g) On the other hand, the MSS. have Ho. where B. reads S. in p'lm^'^ 2, 24 which is as incorrect as B.'s P?T3 the correct form being plir. h) A has the modified ii sounds which is certainly a vulgarism and which, according to Wright, was heard dia- lectically in Old Arabic and is found occasionally in the vulgar dialects'*. For instance, in the word ■i"'i'i:='^i-=c i) B. snisa where the MSS. give xn;= 4, 21. j) MSS. "pnx. B. "-iniiii 4, 25. k) MSS. n^b-^bp {C and D more correctly the Pael rrh-^^p). B. r-^yq^^. 1) Finally, the Yemen MSS. punctuate with Sureq I Cf. S'lVsV Dan. 3, 15. 2 v. Levy, Chaldaisches Worterbuch, s. v. Leipzig 1866. 3 Strangely enough the reverse process is mentioned by Noldeke as taking place in Christian Palestinian Aramaic. "We find u where we should expect i as in ^joxa* (^Hebrew S";;;^)" p. 456. Dalman mentions that Franz Delitzsch asserted that in the time of the Punctatores in Palestine u was pronounced ii (Dalman, 1. c. p. 63). 4 Wright, "Comparative Grammar of the Semitic Languages", Cam- bridge 1890, p. 77. 5 Cf. also the writing of the Arabic sound Plural Nominative in Ina for iina which is common in the Yemenite Jewish- Arabic texts. V. also Duval , 1. c. p. 47. — 22 — where B. has P. ^ in n-.TU'^Js i, lo, SHinis 4, i. In the latter instance Bevan says the change is due to the following labial^; bfjbru-a 4, 12, X:e6 14, 17, &<;730 41, 40. For the two last examples Levy prefers Patah. It can be seen at a glance that the MSS. readings are superior. 8. The Se.Mobile is much more frequently employed in the Spl. system than is the case in the Sbl., since it is also used with the Gutturals^ in which case the Sbl. system replaces it by Se.Compositum. It must now be noticed that interchanges between Se. and the full vowels are fre- quently met with: a) B. P.: MSS. Se. ^^t:xi 3, 8, inS^b" u, 6. The first instance is Etpeel whilst B. reads, perhaps more correctly Etpaal; this will be fully considered when treating of the conjugations. The word "iSS'^b" being Afel of inr has Se. correctly. — b) B. Q. : MSS. Se. ^aS 3, 17, "laS" V 5, I. These being Preterites of Verba, have Se. just as we find in Syriac. — c) B. S. : MSS. Se.; "'ids 3, 15, i<"3:x 3, 20. For the latter cf. B. Aram. CJS* Dan. 2, 10. The Sbl. vocalization as exhibited in B.'s edition of the Trg., clearly shows the influence of the Hebrew language in these examples. — d) B. Se.: MSS. S. "iso 5, i. Levy, s. v. allows either form in this instance. — e) B. H.: MSS. Se. nipT^nS 2, 2, rn=i 4, I. g. That the Yemenite Jews were most careful in pre- serving the correct traditional pronunciation of the Trg. is shown most clearly by Derenbourg in his "Manuel du Lecteur". "They have still preserved the old-fashioned and good custom of translating each verse in public; a little boy nine or ten years of age stands on the platform and recites the Targum of each verse after the Reader has cited the Hebrew". "Further", says M. Derenbourg, „they have pre- served a more exact tradition of the Targum, whereas we know of other countries where the Chaldean Version was already neglected in the iith century" -5. I Cf. Chr. Pal. Aram, "u sometimes stands for a as in aqI J-S-aiut {kvzv), p. 455. 2 Book of Daniel. Cambridge 1892, p. 81. 3 1. c. p-Sog. — 23 — V ^ The use of Se. Mobile for the semi-vowels shows that this Se. must have had a much more distinct pronunciation than we are accustomed to give it, and it is a fact that oriental Jews sound the Sewa just as the occidental Jews pronounce Segol even at the present day\ When the Se.Compositum is represented in the MSS. by Se.Quiescens we may assume that the pronunciation was a more hurried one, whilst the full vowel would indicate a longer or fuller pronunciation. In connection with the pro- nunciation of Hebrew in use among the Jews of Yemen it is instructive to notice Derenbourg's description of the ac- count given by Jacob Sappir^ of the present pronunciation of Hebrew by the Yemenite Jews. "\s regards their vowels they pronounce Q. and P. like the Germans, contracting the lips for the former and opening the mouth wide for the latter, the Ho. they pronounce as the Polish Jews, the S. like the Spaniards and the S. like a very short P., so as to distinguish it from its original P. The Se.Mobile is pro- nounced in different ways; before a guttural it takes the vowel which that letter has, before a Yod it has that of H., everywhere else it resembles a weak a. There are also people at Yemen who speak less correctly, who confound S. and P. and pronounce Se.Mobile with a full vowel and ignorant or neglectful scribes make these errors current in their copies of the Pentateuch or the Prayers" j. In the segoleted forms we know the 2nd P. is only a help-vowel and it was probably pronounced as we pronounce the S.^ That P. was sometimes read as an e sound may I Cf. The Codd. Gaster No. 146, a 14th century Karaite MS. from Nisibis; Is'os. 155, 159, 160 Maliazor Byzant. Rite (XVII and XVIII cent.) and Codd. Montefiore 444, 445 and 446 Mahazor Corfu, which write §e. where we would expect S. and vice versa; em- ploying .'^., S. and Se. indiscriminately. Thus we read n";-: "vv ~~""?3 "?.?■■? and -'p-j immediately followed by t;-:0. 2 In his "jss I'EC (Lyck 1866). 3 Manuel du Lecteur, 1. c, pp. 510 — 511. 4 So also in Christian Palestinian Aramaic in such words as Wjjl (II^a) "Year" >iif. {yi>>.) "with" (Noldeke 1. c. p. 4541 — 24 — perhaps be seen from Proverbs 12, 28 where the Septuagint, Pesitta and Targura all render ri^^-'bx as if it were written ri'2-bit. Cf. also Jeremiah 13, 21 Ti'iins*'^ where we should expect T^'iTiisi. Dr. Gaster has pointed out that Prayer-Books from Corfu give to the Se. the full vowel-sign e or e almost invariably ^ He also points out that the pronunciation of Se. usually heard at the present day is undoubtedly incorrect, according to the opinion of all the old grammarians from the Dikduke ha-te'amim till Kimhi. The ancient LXX transcriptions of nbPa and tr^i^? as Solomon and Gomorrah point to a similar conclusion^. Cf. also the remarks of Margoliouth on the Spl. vocalization 3. b) In connection with the pronunciation it may be noticed that the Yemen scribes made no alteration in the vocalization to indicate the Pause. This pausal influence is a characteristic of the Hebrew and not of the Aramaic language. The rarity of any vocalic alteration under the pausal influence in Bb. Aram, is a sufficient proof of this fact. Even in Bb. Aram, many examples are to be found where H. is employed where no pause occurs and where we might consequently expect S. Cf. b2-i Daniel 6, 21, b^i^a Dan. 7, 8, lbr:3 Ezra 4, 23, "(SiU Ezra 6, 12, ^aair Dan. 4, 11. We may compare its use in our MSS. to that of the Aramaic portions of the Bible t. We there notice only a few instances with Silluq and Sof Pasuq where P. is changed into Q. and S. becomes H. Berliner says in his Massorah^: — "Both schools (/. e. the Babylonian and Palestinian) have the use of the Pause with Athnach and Sof-Pasuq, whilst with the Bureaus the Pause is also used with Sakef. In the case of Genesis 18,30 there seems no difference between the two schools, both schools reading a pausal form at this Sakef". The word Berliner refers to in this passage is ^"'^pt*'; and yet all the I M. Gaster "Die Unterschiedlosigkeit zvvischen Pathah und Segol" in Stade's Zeitschrift, Giessen 1894, p. 62. 2 p. 61. 3 1. c. p. 47. 4 V. Kautzsch, "Grammatik des Biblisch-Aramaischen", Leipzig 1884, p. 39. 5 Leipzig 1877, Introduction, p. XX. — 25 — four MSS. read here bV^xi, Similarly our MSS. read bbr 24, 33, b-^V^n 44, 2. So ='"p" 21, 7. But here and there we notice the Athnah causing a change of vowel e. g. bin 4, 2. C lengthens more frequently e. g. •pr'i 12, 5, -"sr"- 8, 7, at a Sakef; the other MSS. having S. A has -.^^7 at an Athnah, the remaining MSS. again showing a S. Berliner in his Massorah says the form n-^o is only used in Pause when punctuated with H., but our MSS. read n^c in 24, i, 27, I and 35, 29, where there is no pause. On the other hand they have ^h^k 36, 11 at Athnah where B. has a P. We also notice the influence of the Pause in nl'arb'' 38, 13, c^^Bx 41, 52 (but A n-^tiEX) and '^.-"^ 46, 6. Yet the cases where we find a pausal influence in our MSS. are extremely rare, and hence we see that the influence of the Hebrew original upon the text of the Trg. punctuated according to the Spl. vocalization is comparatively insignifi- cant. It may be noted that most of the examples I have quoted are Proper Names which were probably punctuated in these cases exactly as they occur in the Hebrew text. 10. From all these examples it may be seen what great differences are noticeable in the two systems. In general, the Spl. system approaches the Bb. Aram, and Syriac forms, which we should naturally expect in an Aramaic language, whilst the Sbl. system shows considerable Hebrew influences. Further, the forms given in the MSS. are etymologically more correct than those found in our editions. In a few instances the surroundings influence the pronunciation. VII. ORTHOGRAPHY. 1. The ortliographical variations shown by the MSS. are numerous and important. I shall here only indicate a few. Lists will be found in Part II and also an attempt to justify the readings of the MSS. a) The following variations may be regarded as due to the influence exercised by dialect. — 26 — B. reads "pifiit" 3, iS; the rest have 'piox'^. L. confirms^. A. N'-^S3. B, C, D XTj-^ci 6, 16. B. has .'^'i-J'^p^. B. and C-i-^'-jQ'n 45, 22, A and ^ '^VmOn, D vb"'-j:iN. L. ^, although noticing the form given by B., gives that of D as the best Reading. — b) The MSS. occasionally make use of vowel-change to distinguish the different meanings of a word. Thus b» =upon, l3y=he went up; "i3==son, "13 = besides; n5<= thou, rx = a sign. — c) C has the 3rd Person Plural of Verba s'b ending in '' and not in ix. This peculiarity is pointed out by Berliner in his Massorah (p. 92) as a Syriasm. It is also occasionally found in B, but A and D never show it. As an instance we may take the word T'Ennx'i 7, 20. Some variations are also found in Verba i"r. A, B and D read 1"'-ri?:i 19, 14, but B. and C "j^xri'ai. Landauer^ very appropriately refers us here to the Present Participle of the Verba l":' in Syriac in which language >>JL» is pronounced ka-yem. Similarly ""Nri'2- would be pronounced as if it were written ^^'^<^^xa and this latter reading may accordingly be regarded as a popular spelling of ~'xn^2 3. — d) We some- times find interchanges of letters consistently occurring; thus D is used by the MSS. in cases where B. writes io. The latter is not an Aramaic letter at all, but is imported from the Hebrew. A has it incorrectly in two places where a UJ is to be read viz. voj? 9, 13 and X";&'^p 19, 17. — a and :. B and D have Dpi 31, 42. The Rest ip-i. All have I'lS 31, 27. B. tis which the author of the Pathsegen'* has already pointed out as standing for the more usual "fs, as Landauer remarks s. V. in his Massorah. All MSS. read "^n^ where B. and L.5 ^ s. V. 2 "Massorah zum Onkelos nach neuen Quellen", Letter- bode VII &c. 3 In Biblical Aramaic the Ketib is always with s but the Qere with -. Cf. ■j''-,s- (Qere "^"''^ Daniel 2,38; 3,31; 6, 26, '-;vr^ (Qere ■^;;-■^) Daniel 4, 23, "'os|? (Qere ■]"?".5) Daniel 3, 3, -j-js-! iQere ■;-r'^) Ezra 7, 25. Cf. also Dalman's Grammar p. 45 '*Aus Schreibungen wie r:''-5;>3 fiir rsV;'^, D'-p fiir c-sp darf geschlossen wer- den, dass k zwischen zwei Vokalen als stimmhafte Gaumenspirans gesprochen wurde". 4 Published in Adler's edition of the Targum L c. S s. V. — 27 — give xn- i6, 12. A and D srbV 40, \^. B. B and C nrhf\ The n is a Hebraism, although Bb. Aram, likewise has it. e) B writes n^isbms 14, i. A, Cand D divide the word "i^aii'b -n: and this, at a later time, was looked upon as the correct traditional writing ^ The MSS. give ^z-ii if-in^i 3, 16 and ■z^^rsr 12, 13 in all of which B. writes "'z'^-^nr^ &c. — f j The Yemen MSS. do not know of the so-called EiJ-n ',r:::> which gives us such forms as x;^^- and rrir-j-is but write these words as they are found in the Hebrew Text. — g) By niaini xip the Massorah meant that several passages should stand in the Trg. just as they occur in the Hebrew. The MSS. however appear to ignore this Rule. Thus they all read *rix (and not ^n-x) 34, 19, riNrn 27, 46. A and D read ni^aSi x-s'-^ 27, 2 but here £ and C follow the Hebrew Vin n-"'. ^ and C have wS:nbs on the margin but n^iir in the Text 3 26, 14. A and D have S'1^^ but ij" and C "^V:^ 23, 6. A, B and Z) x^=-i C n^'i 39, i. ^ and B read =--:i-2- x-nl-i, C and Z> c—ni-i ■^nj-i 15, 18 and finally B writes the Tetragrammaton instead of n-n5X4 9^ 6. By these examples it may be at once seen that the scribes could not have recognised these rules of ciann x^p, or, at any rate, they were only slightly acquainted with them. — h) Marginal Readings have already been pointed out occasion- ally. These glosses are important since they are all added by a later hand and explain the origin of many words which afterwards became incorporated with the Text itself. These marginal readings also constitute an argument for the greater age and originality of the texts with Superlinear vocalization which have been preserved to us by the Yemenite Jews. — 3, 23 C has "^"^crXT on the margin, the text of which has been I Cf. Ezra 4, 16 where r;:r::s stands for the more usual s:r:s so Or. 2374 has s^;' si'c" (Dan. 2, 47) where Baer reads n^j^ rryy\ and Siyn (Dan. 6, 13) where Baer has r-yr. 2 Cf. Talmud Babli Tr. IluUin 65a. 3 The reading sinVe is quoted by Theodorus Mopsu- estenus (v. Part II, Ch. II, J). ^ cf. B. 1. c. p. 216. — 28 — corrected by a later hand to "'Osnx'i; but as the word '^'lair'iti is given by Berliner in bis Massorah^ under the heading of P"'5<1 which means that it is undoubtedly the correct reading, we must reject the correction ■'ODn^<'^. 7, i6 C has iTi'n5"'in on the margin to be placed between "^"i and '^nibs'. We shall see when speaking of the Additions that A and D have this word already inserted in the text. 25, 8 Z) has 'p^i'^ on the margin to be read after "non. B has the same word in its text, 27, 22 C has "^nii^ on the margin but "^"i^ cor- rectly in the text. This has been noticed above under niS'im x^p. Many more examples of marginal readings will be noticed when we shall treat of the Variant Readings. — i) The Yemen MSS. almost invariably follow the reading given in the Massorah of Berliner under the head of "^siSTnfisVi which re- presents a School in Babyl. in favour of the original Plst. Text and is a great argument for the authenticity of the text which is found in these MSS. To give one instance; B. reads n]5 and >inn 17, 26 and 27 the MSS. have the Itpeel nisnx and i"i1:ni<. Berliner^ places the Peal reading in the column headed '^5<"nDb whilst the Etpeel is placed under ixriinsb. ''Perhaps" — adds Berliner — "whether we should take the Peal or Itpeel form depends upon the old grammatical question whether the root of the Hebrew word is 5T0 or ^"23". Baer3, however, is of a different opinion. He says, speaking of certain forms which occur in B. Aram, and which he treats as Etpeel ''Praeteritum huius conjugationis interdum syllabam praeformativam abjicit velut 0"]Q = onQr!i< Daniel 5, 28. VIII. GRAMMATICAL VARIATIONS. I. THE NOUN. a) State. There are a few changes in the different states of the Substantives. I 1. c. p. 116. 2 Massorah p. 62. 3 Daniel etc. p. LIX. — 29 — a) In the Absolute and Emphatic States, which in the MSS. appear to be used ahnost indiscriminately'. Noldeke informs us that these two states were entirely confused to- gether in the Mandaic dialect of Aramaic^, and from the examples which will be given in Part II, it will be seen that a similar laxness is noticeable in the Targum as represented in the Yemen MSS. A, B, D snxa 4, 16, B. and C i•: "young man" were still of common Gender and used also for 'she' and 'young woman'. Vide Gesenius, Hebrew Grammar, Translated by Conant. London 1839. §§ 2. 3. 2 We find an apparent incon- sistency in Ezekiel 23, 45 where a"Hs refers to a Feminine Subject. In Verse 47 the word is written ^nn^s. 3 Samaritan (as A, B and D) ■jlBlp'^n. In the following pages some readings from the Samaritan Targum which agree with one or another of the Variant Readings under consideration have been inserted. .Since Samaritan has no vowel signs these instances are only of value in confirming the ^Con- sonantal Text. Upon the much-vexed question whether the Sama- ritan Targumist was influenced by the Targum of Onkelos or agree- ment between the two is due to later interpolations into the Text of the Samaritan Targum I refrain to enter; but bearing in mind that the language of Onkelos and Samaritan are two kindred dialects of Aramaic and that they both represent very ancient Translations of the Hebrew Text, the examples given may be of some value in help- ing to confirm a reading. The edition of Dr. Adolf Briill (Frankfort: 1879) has been, in all cases, the text of the Samaritan Targum referred to. — 33 — These are evidently mere anomalies, and of no import- ance. It is unnecessary to give examples here as they occur throughout all the MSS. invariably. Other variations in Gender will be found in Part II. 2. THE VERB. a) The Conjugations. Very numerous interchanges are to be observed in the Conjugations. a) Peal and Pael. B D^=1n. B. A, C, D 0:21' 31, 54. Landauer^ explains and justifies the latter reading in his Massorah. He says that here only xrw: C33 is mentioned, and hence the Peal is the correct Conjugation. The case is different in Numbers 22, 40 in which example the Pael is necessary, since Balaq offered up a large number of animals (',51 •j-'nin). A, B and D Trh^t"^^. B. and C n-'^x-ii 37, 15. L. gives no Pael of this root, but it may be observed that in the Syriac language the Pael has the same signification as the Peal. P) Peal and Afel. B. and C "'nix', A, B, D ■'p^k^^ 4, 3. Similarly throughout Genesis. Now, there is no doubt that A, B and D have the only correct reading of the Afel of xrx cf. Syriac -V'i- B.'s form is really ist Per. Sin. Imperf. Peal. B. "'?rf7?^'!'7 MSS. have ■i:1-'n5iy'^T 4, 14. The Afel, is the more convenient form, according to L.; this form being preferred to avoid the confusion which might arise in the employment of the Aramaic root riDUJ = to find and the Hebrew root nrd = to forget. 7) Peal and Etpaal, Etpeel or Ettafal. MSS. a'nf n (Peal) B. arnn'n (Ettafal) 3, 19. The Hebrew -::Tr being an In- finitive may be either Active or Passive. Here it seems to bear an active signification, which would confirm the reading of the MSS. L.* also gives the Peal form here, but Path- segen^ supports B.'s reading and explains it reflectively. 1 Samaritan D33\ 2 Massorah s. v. 3 Samaritan niNS"!. 4 s. V. 5 To this verse. — 34 — B,C2ind B, p'^bo^^ A and Z> p;"r&xi (Etpaal) 2 6, 23 . The Hebrew text being Isy^i appears to confirm the first-mentioned read- ing. A and Z> nn^b'^'i^ B. B and C h-iB-inw (Etpeel) 46, 22. The Hebrew is ^h"^. The first-mentioned reading, although giving the sense of the original, is less literal than the passive construction. 6) Pael and Afel. B. in-^x MSS. ^li^x 34, 19. Both these forms have the same meaning, but Berliner, in his Massorah^, places the reading '^n^x among the list of instances included under the head of Qiaim xip and we have already seen that the MSS. appear to take no notice of this Massoretic rule. L. gives "in'^x as Pael of irix, but as the Pael of "inx occurs very rarely whilst the Afel is fairly common, the reading of the MSS. appears to be preferable. B. r^^^yj-^. MSS. nNi'iS'i->"'n'' (Passive), B., B and C "rSn (Active) 34, 22. The Hebrew being D"'5a2 the passive construction is superior. B., A and C "^3^^% B and D •'D?3'2b'9 13^ 16. The latter is the literal translation of the Hebrew r^"^3^?. Similarly in 31, i, B., C and D have 'p'^^ST whilst A and B read n^'ab' this being a translation of the Hebrew "tax;. IX. OTHER VARIATIONS, i) We find a large number of additions, mostly of small words or prepositions in our MSS. Many of these additions I Massorah, p. 98. 2 Massorah s. v. 3 ~;'; rrr: to this verse. 4 1. c. p. 203. 5 s. V. ^ To this verse. 7 Samaritan has "j-Vs-i 8 Samaritan likewise ]"'i''75. 9 Samaritan ":"3»:5. - 36 - make the text clearer; others have Halachic reasons, whilst a few are mistakes. We likewise find omissions, that is to say there are less words given in the MSS. than are to be found in B.'s Edition. Some are omitted to save the scribe the trouble of rewriting a string of Proper Names: the verse, as Berliner points out, being read thrice in Hebrew, instead of being followed by its Targum and then by its Arabic trans- lation. Occasionally these omissions are due to the scribe's carelessness. 2) Additions. All the MSS. insert nin t simplify this verse by adding ""a before tD'ip. Some irregularity is observable in the use of 'i to express the construct state and a similar laxity occurs in Syriac. Generally, the construction without 1 is the more classical one and the use of i is the more popular construction. This employment of "1 being a much freer and looser construction is preferred by the Aramaic language and is much more frequently used than the other construction. The MSS. add it in CiT^n 2, 11 (Hebrew nnion) A, B, D naiyn^ C t2D'^ n, 18. According to Berliner's Massorah^ the 1 is incorrect in verse 18 but necessary in verse 19; since 18 is given as an instance of msin i<"ip. The little word T^ is sometimes added by B and C, mostly, however, incorrectly. C has ni before nibp 21, 17. This n^ is copied from the rx of the Hebrew text; but inasmuch as the Trg. employs a Passive construction here, ni is evidently in- correct. - is added by B in "i"'^^^'^ 13, 10, a reading which is rejected by the author of the Pathsegen"*. 1 is added by the I Samaritan •ava. 2 p. ^g. 3 Likewise Samaritan "ij's^. 4 to this verse. — 37 — MSS. in N~"'r3i" which is asiainst the Hebrew text and in •pn^-i " 33, 13 which follows the original. A and B add it in ~r.'. 10, i not as Hebr. ~ is added by A and Z> in •)inn5€ 42, 24 and since the corresponding Hebrew word is Bii'^bi;"? this reading is the correct one. n is added by A, B and C in rp^£* 42, 21 (B. and D rpr). The reading with 1 appears preferable, since it suits the context so admirably. 3. Omissions. I have noticed some scores of omissions, but shall only mention a few which are not due to any over- sight on the part of the scribe. Sometimes verses are omitted in their entirety to save space and the scribe's time; this is particularly noticeable in genealogies. For instance 10, 22 and 23, 27 — 29. 18, 19 A omits i:pn which does not correspond to any word in the Hebrew text but is explained by Adler in his "^sb ns-^rj "!:"T"-i vr-iTr 1-23 sin", 18, 30 B omits x-i-i; which again has no equivalent in the Hebrew text. 19, 7 C omits 1"= and Pathsegen also remarks that there is no occasion for this word in the Targum text since i--£ = n:. 43, 15 --/ and D omit i-fn- which is again represented by no word in the Hebrew text, but Rasi explains the word as necessary, since, in Aramaic, different Verbs are in use to indicate the seizure of things and the seizure of human beings, 203 being used of things and ~" of men. B. in- serts a T before the following words which I write as they appear in the MSS. ~\prr. 3, 9; 7^n3 7, 22; xrVo 18, 6; n'^rnx3 19, 16; b£-J 23, 19 {A and D only); "ir-Ti-s 4I; 39- An initial i is omitted by A and B before ir^rrNf 3, 21. Also before bs'i 6, 20 which corresponds to the Hebrew. The 1 Samaritan ";«r'!<3". 2 Samaritan rpj-;. 2 Samaritan also -,— ; and rrrs but, on the other hand, -.'j-. - 38 - MSS. omit H" before pn:i" 21, 8. In several other passages the omission of this word is due to the scribe's carelessness. h is omitted before tTiOa 35, 27 by A, B and D ; in connection with which we must notice that it does not occur in the Hebrew text. Also before "prisS 43, 27 where it likewise does not appear in the original^ although it does in the following verse. Pathsegen says we must insert the ? in both verses, according to the usage of the Aramaic language. The word xn is omitted by A in 48, 22 and by C in 32, 18; both passages are however faulty in these MSS. 4) Contractions, The scribes regularly contract certain familiar words which constantly recur, especially those in combination with xn and X"2. A has -i"!--? 4, II. B, C, D -fr -p. A, B and Z> xixri C x:x 5 -\k:n, B -s-jn 4, 7. Ber- liner, in his Massorah*, says the suffix is incorrect, but L. reads as the MSS. Likewise, in the variation of the vowel- point L. permits both forms. B. "jj^r!. A, C, D xlrn, B xjnn 19, 12. The Hebrew is irn. Berliner, in his Mas- sorah,5 says the Targumist wished to make this word agree in form with the following words "["^^^^ "J^l both of which have the suffix of the 2nd Person. Pathsegen remarks the same. But this hardly appears to be necessary and the literal translation of the MSS. therefore seems to be super- ior. B has the word incorrectly punctuated. X. EXEGETICAL VARIATIONS. Most of the variations which have been considered hither- to are of importance, showing, as they do, the greater purity of the Aramaic language which the Yemen Trg. Mss. exhibit. I Hebrew and Chaldee Lexicon translated by S. P. Tregelles London: 1885 s. v. 2 .<=. v. ^^ (Leipzig 1836). 3 p. i>l. 4 p. 74. 5 p. 75. — 40 — These Variations give an incentive to the study of these MSS. if for no other reason than their more accurate re- tention of pure Aramaic forms and expressions. The exegetical variations, however, are of far greater importance. They cannot be accounted for on the ground of difference of dialect, nor do they point to any linguistic peculiarities, but give a clear reason for assuming that they represent another distinct recension. The chief characteristic of these differences is the preference frequently shown for a literal translation of passages which are in our texts rendered ac- cording to traditional Hagadic translations. It has already been pointed out ^ that the preference for literal translations shown by the Plst. recension may perhaps be accounted for by the existence in Palestine of the other Trg. which is certainly of Hagadic character; I refer, of course, to the Targum Jeru- shalmij or it may be accounted for as the outcome of a desire to have a translation which should render the Hebrew text in the most faithful and accurate manner, which is also the tendency of i\.quilas and all the old translators, who are severely literal. A few instances are given here, the rest will be found in Part II. 26, 2 2 B. 5<;3''^a'^'i, C and D x"D2"imE3^% A and B la'isir. The Hebrew is S3i"is>i of which A and B give a literal render- ing "God will magnify us and we will multiply ourselves in the land". That this is the valid reading is further proved by the support it receives from Rasi and L. The author of the Pathsegen, however, gives a reading similar to that of B. as the correct text "God will magnify and increase us". 30, 15 B., B and D 'pno'^rV, A ^:iX?oh^, Cis missing. Luzz. in his ^5 anxj points out that those who corrected the original •pn6"'r^ to -6ri?i made a great mistake, and did not realise the spirit of the Trg. rendering. Onkelos — says Luzz. — undoubt- edly wrote '|"^i6"rii but later revisers thought this reading was incorrect since the Infinitive rinj^^"! should require the Infinitive I Chap. IV S 4. 2 Samaritan i-acni. 3 p. 25. — 41 — 3DT2b'i whilst •pao-'ri would be the Targum of Pn~y — 2nd Feminine Preterite. This, however, Luzz. goes on to say, is not the case; since 0. wished to amplify the brevity and clear up the obscurity of the Hebrew passage and accord- ingly paraphrases "fio-'r-. Vet in spite of Luzz.'s arguments, one is again inclined to favour the literal translation 26-:b-. 30, 22 B reads bn- n^ •'^ n-=i-, B., A, C, D x:-=-t :ri ■1 cip bn-T which is evidently a Hagadic rendering (v. Rasi to this verse). B gives the literal translation of the Hebrew text viz bn"! ns D^ribs n-isr;! and therefore appears to be the best rendering. 39, 11 B (text) nT-n-^iS' isr^b'', B (margin), A, C, D and B. n^:5-r-n ^i:r=5 p^a•:b'. Here we have a good instance of the manner in which marginal glosses crept into the Trg. text. The Hebrew original is ypcyivz nib^b of which ^ is a literal translation. 42, 36. In this verse we see another example of the retention of the original text by B. The Hebrew is 15.^^ which B renders ''h-r.^b" whilst B. and the other MSS. give x= x^n xb'' which gives the sense of the original but is not literal. B.'s translation is also given by L. ^ 43, 2 B. and C ip^EO A, B (margin) and D ■x-^:J-'i-, B (text) erased. Hebrew is lbs. Now the two Aramaic roots p2C and "^^Jr have the same signification. The authorities differ which word to prefer here. L. remarks that ^:£'J usually corresponds to the Hebrew '^'^'z'^ and Rasi likewise reads "'sc. He goes on to say that ip'so is an incorrect reading since the root pso is only used when one has eaten to satisfaction although some food may still be left, whilst "X-ii-" is used for a man remaining unsatisfied although his supply of food is already exhausted. But Luzz. in his Philoxenos* pleads for the reading ip'^EO saying that this is the verb employed in Aramaic when the text refers to a 'finishing' with eating, drinking or any other bodily requirement. The I Samaritan r.-r-rzs -I5<:b. 2 s. v. 3 Cf. also Ezra 6, 15 -^-r. "And was finished". 4 p. 45. — 42 — fact that iX'^^'^ir is given by j5 as a marginal reading seems to support Luzz.'s opinion. It must be noticed, however, that the marginal readings are sometimes preferable and appear to correct a faulty reading which is found in the text. Thus in 45, 11 B., C, B (text) read "^^j^r-iiJn whilst A, B (margin) and D give ■|fD^rn. The Hebrew text is 'd"}^. A, B (margin) and D give the best reading since "pD^rr = be impoverished whilst i^rTniJn has rather the sig- nification of 'be annihilated', 'be undone', 'die'. Rasi and Qimhi both read "po^Dnn but Pathsegen and Luzz.^ say we must read '^:i"'ndn for — say they — how can we speak of all one's possessions i. e. one's cattle &c. coming to poverty? But be this as it may the reading "ro^onn corresponds exactly with the Hebrew "r^ir, a word which is correctly rendered "come to poverty" in the Revised Version, although it may be noted that some of the Jewish commentators differ in their translation of this word, thus Ibn Ezra "to perish", R. Samuel b. Meir "to be dispossessed", Mendelssohn "to be ruined". 45, 1.7 B (text) -nx^, B (margin), A, C, D and B. Wi^■»,. Hebrew ^ixs. The first mentioned reading is preferable since it is quite literal. The reason (says Pathsegen) of the Targumic rendering ib'^mx is to avoid the redundancy of the Hebrew text and thus render zaxa auvsjtv. , 46, I B (text) xro^i 0:21, B (margin) and the others •fra-i nan, Hebrew Q-'niT ri2T*i. Although B (text) is in- correct it is strange that the Trg. of Q^raT nmii in 31, 54 is as he gives here. 47, 20 B. '^ii, MSS. xsp^i, Hebrew '^:^. B.'s reading "pti is also quoted by Pathsegen. On the other hand Landauer gives 5<;p'i in his Massorah.3 Perhaps B.'s reason is that in this verse Joseph really bought the land, whilst in verse 23 he acquired the people. I "5 sns. P. 46. 2 Samaritan T-xi. 3 s. v. PART II. CHAPTER I. Before commencing the lists of examples which illustrate the preceding rules, certain words which call for some comment may here be treated. I believe that most of these words will be found to be more correctly punctuated in the MSS. than in our editions of the Trg. As these words are of constant recurrence their correct vocalization should be fixed. First we have the little word ni. This word, although so exceedingly common is of obscure etymology. Our editions punctuate this word with P. and B. invariably writes r\ L. also writes the word with P., although he compares the syriac f^ = nature, of which V; is construct state. The P. is also supported by Nestle ^ Bernstein-Kirsch and Ge- senius in their dictionaries. Merx in his "Bemerkungen iiber die Vocalisation der Targume" ^ mentions a curious rule which Mercier wished to establish. Mercier wrote n^ in the Trg. when the corresponding Hebrew is nit, but r^ when the text has rx. Merx declares most decisively^ that we are assured that r^ is the correct punctuation by Bar-Ali; r:, on the other hand, is invariably incorrect. Bevan also writes n^"*, also Duval s. Undoubtedlj- the greatest authority who supports the Q., is the great Orien- talist, W. Wright. He remarks :° "We get in the first place I Syriac Grammar, Berlin: 18S9. 2 Contained in the "Ver- handlungen des fiinften internationalen Orientalisten-Congresses" Ber- lin 1881. p. 145. 3 p. 185. 4 Bevan: The Book of Daniel p. 38. 5 Grammaire Syriaque p. 277. 6 Wright: Comparative Grammar of the Semitic languages p. 112. — 44 — rT'X which was doubtless pronounced in the earlier stages of the language lyath or lyath; for otherwise the "^ would not have been inserted in writing, as is almost invariably the case in the older inscriptions. In the inscriptions of later date, however, we find rx and Plautus heard the word pro- nounced yth. The Aramaic foim seems to be shortened from the Phoenician viz Syriac K>, Chaldee n;;', less correctly n^". ^ After the opinion of such an authority the punctuation with P. may fairly be assumed to be incorrect, and that with Q., which is invariably used in the MSS., substituted for it. According to Merx^ both the absolute and construct of hj is hb or br, the use of which depends upon the text of the Hebrew original, which is to be attributed to the Mas- soretic schools. But this is not the case with our MSS. as may be seen from the following examples. 2, 5 ba^, where B. and the Hebrew text both read bb% 2, 16 A bsa, the rest have ^Bia. Hebrew is h'^i-o. 2, 19 B. and Hebrew hb\ MSS. bBi. 3, 17 B 53. B., A, C, D and the Hebrew bb. 6, 2 B br^ A, C, D and Hebrew Vsp. These instances suf- fice to show that no rule can be fixed to indicate the em- ployment of the two forms. I have, however, noticed that C has Ho. throughout except in the following instances ^2" 35, 6 bS-T 20, 8 and 45, 8 and 9 and bss 47, 17. Our MSS. always punctuate Q^P and this I consider the correct punctuation since the corresponding Syriac word is Jjo^ TP>. The Bib. Aram, also supports this reading, since it is Dip and not Q^ip. Berliner, in his Massorah^ has dip as t't: t ' t ■' r'r the reading of the Nehardeans and di|5 that of the Suraans, whilst the Massorah to Exodus 6, 5 says that ''T^,. is not the language of the Targum but that of the "Chasdim". In 1 But Schroder justifies either punctuation "Sie wurde urspriinglich mit lautbarem Jod njx gesprochen, spater wurde ijjath, woraus sowohl arab. ijja als zx2M\zl\%c\i jath, joth hervorgegangen sind, wie iath, ieth ge- sprochen und schliesslich mit Unterdriickung auch des a-Lautes, zu ith, yth. Schroder "Die Phonizische Sprache". Halle 1869 p. 213. 2 Page 31, Note 3 of "Chrestomathia Targumica". 3 p. 64. — 45 — spite of this, however, we may safely reject the reading with a full vowel and read t:n~. which is of course onp in the MSS. As to the word b"^ap a difference of punctuation exists which can be best understood by tracing the word to its etymology. B. and C b£p?3 21, 16, A, B and D have b-^np^. B. and C b^3)3;i, A and D b-^npV ! B b^ipH 33, 18, All have ^T^z.'p'i 41, 3. The Syriac forms are ^^«J», « ^«Ajb and ^^u»«^ which may explain the forms "'i'dpS and b'^ip^. Cf. also the Bib. Aram, forms in Daniel 2, 8. 5"^ip b'^iip and b"^2p are the usual forms we meet with in the Trg. but the readings of A, B and D appear preferable. Finally, how are we to read the word 'T'b"irr'i< which oc- curs so frequently? We may take 4, 26 as a type. Here B. reads "'"'rn'^x, A I'^b'^nx B and D T'b'=>rx and C T'^n'^x. The readings of A, B and D are the usual Targumic forms of the Etpeel of ib"^ whilst C is a Syriasm. Cf. tV^i- C invariably has this form. CHAPTER II. Examples of the rules contaified in the preceding Chapters. i) Substitution of other vowels for S. (Cf. Chapter V § 4). (a) P. b£n 4, 2. y^ 4, 19. -rr^ 5, 16. 'ros- 6, 3, ■^ir^i "i^ii 10, 2. "i'^xn baS 10, 10. (b) S. 'r^i 23, 9. 2) Substituiions for Se. Compositum (Cp. V, S 5)- y a) Se. mobile. a. For Hatef P. : xBVoJn- i, 2. xin- i, 4. n^x'^ i, 6. "i25" I, 7. x£-n-,T i^ 28. p. For I; atef S. '^^t\ 24, 60. b) Se. Quiescens: xrnS i, 25. ninn 2, 11. nVsn^ 2, 23. xinb* 4, 5 [^ incorrectly xnnV] x:rV 4, 9. ^nri** 8, 13. x^^ni 8, 22. 'jirnb'nn- 9, 2. nin"! 10, 10. (^i and D n"'ni) ■]3"'Tni trc^-i^nit, C n^nisnx). X2S 36, 2 and D "pinnx 41, 3 and 19. 3) Vowel changes. a) B. P.: = MSS. Q.^ a. Monosyllabic Words and Particles (Cf. Cp. VI, § 3a) ^= 2, 4. ^"^1,-^1 (Syriac \^). ^5 13, 3. ^n 15, 2 (Syriac ^). n=^n 15, 16 (Syr. U?^). 1^'5 15, i6. n^^ 17, 7 (Syriac j^^nN). -jnnn 17, 7 (Syr. yji^). nxH" 17, n (Syr. \\\^). |i< 18, 9 (Hebrew -s). 'In 19, 8 (Syr. 4=»). tD^^r: 19, 8. x^p 19, 26. Concerning those words to which the corresponding Hebrew or Syriac forms are not given 1= and n5:n"a are in- correct, according to L. s. v. "^"5 is a contraction of 5 and ')> (time) and this '5 is really a shortened form of 5tr;s or, as it is also written Nroi5> which latter form may perhaps ac- count for the Q. ^ in the MSS. L. allows i;! side by side with "1^; whilst to justify the Q. of x^p we may compare the Hebrew frap (Exodus 22, 5). [i. Miscellaneous words (VI, 3g) "^t 2, 18. 'pn-JN'^ 3, 18 (Hebrew tjn). 'ipi-i 3, 21 (Syr.: '^0- ^^^'' 9; 12 (Syr.: Jv>Ns). -£; 10, 9 (Hebrew ^isf). 'X'ir\ 18, 7 (Syr. yi.sj). 'f:n"D 27, 23 where, however, L. says we must punctuate TX^"P.- ■^'■"'"'' 42, 13 and 36 &c. where, according to L., the correct reading is "^riin^?, but when it is noticed that the I I write the words as they appear in the MSS. 2 Cf. Syriac ^ 'what' which is also written v**- - 47 — word T\'h is merely a contraction of r^x x; may this fact not justify the Q. which the MSS. read? b) B. Q. = MSS. P. a. Vowel of the 2nd Radical of a Verb (VI, 3b) unrzrn'- 3, 7. ly^'i^'-i 3, 8. rnrsi 4, 11 nVnrnx" 6, 11. ^~t: 8, 19. l^rsn-i- 14, 15. p. Absence of Pausal Influence (VI, 3 c) b=i^3' i, 29. p5uj73b?3 4, 13. '''^fn 2, 7. ns 3, 9. 7. Before the suffixes x: and ■^: (VI, 3d) x:3 n, 4. x:tirinr 5, 29. 0. Absence of Se. Compositum (VI, 3e) ili'^sb' 2, 3. r-ofx'' 3, 12. ^rr-x 3, 13. c) B. S.: = MSS. H. (VI, 4) rniV 3, 22. C'p5 9, 9. •'■ic' 9, 20. 'I'nb-j-'Tii II, 2. =-;•£■ 15, 16. '•r.'z-: 17, 7. njiip 18, 20. "'r-'JDN 18, 27. In each of these examples (taken from the first 20 Chapters) I have compared the punctuation of the MSS. with that of L. with which it always agrees except nb'^np which L, prefers with a S. d) B. H. = MSS. S. T'i"!^" 2, 6. b-^nr 3, 14. -;-Vn- 5, 22. •^tnr^. 6, 14. c-'D-x 8, 21. 'Tir- 9, 13. n1-Tnx= 9, 16. nl^bsrx 10, 25. xr;-zV u, 3. -ssr: n, 7. -iix 12, 13. T^b'^nn 15, 4. 'ppi^T 16, 6. x:-r 16, 7. "'iox 16, 10. iVx 18, 13 (y^ and C here T'b^x). n^zr 18, 17. xr^-s^s 19, 24. ■'nr^ 19, 25. ns-^zir'-i 19, 33. n^rp^r: 19, 34. All these agree with L., with the exception of ";ox ri"^=ai and rr^rpuj: which L. punctuates with H. He admits either vowel with xns-'sb and xn'^nsia but prefers H. ibx is as in- correct as B.'s I'^b'^x since the Imperfect Peal of "ib- is t^'sx and the Imperfect Afel is ^^5•x which A and C read. A has some errors, punctuating the following words with S. instead of H. xrrm 2, 21. Tr\^^ 6, 15. -^rr S, 11 and b-nn 18, 19. e) B. S. = MSS. Ho. (VI, 5) b-r-^r 2, 16. 2-nr 3, 19. nn^'in 5, i. i-inp 8, 4. xz--" 8, 7. rso'^' ' 8, 17. I Likewise Nestorian Syriac has the sound with \'erba \ — ^ e. g. v«»Jai. - 48 - jt'i'ipT 8, 22. x^''n"i 8, 2 2. i^-u;'''^'! 9, 5. n'lxn' 15, 17. 3'nirii 19, 32. Of these L. admits either vowel in the case of x^in rnbin xn-iiS" siip but prefers S. with ii^p miri'^'i sirn. f) B. Ho. = MSS. S. -nil:? 4, 7. xri^T 8, 22. n"J 10, 30. "TSD^. Of these L. only punctua.tes "p^rJ with Ho. g) B. H. = MSS. P. (VI, 7a) x^on 6, 12. n?5n 8, 9. ",iBay 15, 9. ^ has 'pi^^? 7, 13. For the latter we may compare Syriac >i:^ = with. Of the others L. justifies only tiba-i and x^bsa. h) B. P. = MSS. H. (VI, 7b) x52U5b- 4, 15. ns:^^ is the Bib. Aram. form. (Dan. 4, 13). i) B. S. = MSS. P. (VI, 7c) n^in 6, 7. d5^5 9, 6. nn^a 12, 6. nnp 19, 9 (Z> nnp) by"^i5 19, 31. s'-in 21, 6. Of these aip alone appears to be incorrect, since intransi- tive Verbs usually have S.^ j) B. P. = MSS. S. (VI, 7d). A has r?3 16, 2 and "ih"!;)"! 18, 19 both of which are wrong. k) B. Ho. = MSS. S. (VI, 7f.) inn^=5 25, 31, iVso^n 44, 23 (C I'lSO'-.n). ^ has ''"iT 47^ 17 instead of "^^'n. -jn^i'^sn is confirmed by L. ; and all of the examples may tend to con- firm Derenbourg's statement^ about the Yemenite Jews viz. that they pronounce the Ho. in the same manner as the Polish Jews. 4) The Vowel of the Imperfect (VI, 6) n^rn5u:^T 4, 15. iriax 6, 7. r=rix 14, 23 (C "£nx). nl^-rx 13, 15. --frx 16, 2 (C "^'inrx). -''n&'. 17, 8 (C "^nnx^). alrirx 19, 20 (B aitrirs). In those examples with initial x the P. stands for S. which would be the Sbl. punctuation. V 5) Interchange of Se. and full Vowel. a) MSS. Se. = B. P. (VI, 8 a) in^ir^T 12, 5. I^nx 13, 8. rr^-in^^ 14, 22. 'p^xar 19, 7. b) MSS. Se. = B. Q. -iv5 4, 7. -^rx 4, 9. ^:2np 4, 10. -pbi-i'^ 6, 20. iTfi 9, 23. ■'xanis' 10, 13. "ixninx 10, 18. I This is however not always the case "Non solum intransitiva velut 3';p sed etiam transitiva velut 5St« Dan. 2, lo, uVip Dan. 3, 27 in Zere terminari possunt" Baer: (I. c.) P. LIX. 2 Manuel du Lecteur p. 511. — 49 — n'^px 14, 16. wsn5 15, 17. xrxb^ 16, 7. t C!Ti<). Of these iTn and N"!2> being Preterites V Peal of Verbs have Se. correctly. In ttv -^-anp soxb'a and T^nx the Se. mobile corresponds to the Se. compositum with which these words may be punctuated in the Sbl. system; but the Afel forms 2*inft< and Q'^px seem to be incorrect V with Se. c) B. = S.;MSS. = Se. (VI, 8 c) "^x^'i^S 18, 16. |-nli23^BV 10, 20. -Jibs 12, 3. Tniv, 15, 4. b-^irS 18, 31. All these are truer to the character of the Aramaic language whilst B.'s forms are Hebraisms. ""^ribN alone appears rather doubt- V ful, but, according to Merx^ the Se. is justified by Buxtorf in his Dictionary. 6) Orthographical Variations (VII, le) Chas n"iS'5 (with- out final n) 19, 24. Occasionally the punctuation of the MSS. seems to be in an unsettled state. Thus in 13, 12 -5 has y^TT' and 3"^n^ in the same verse; whilst £> has ^''n' and 3'^n'^. The correct reading is 2"^:^=. Elsewhere, C appearing to be in doubt which is the correct punctuation writes ■)i3'>ba^! L.* admits either vowel but prefers H. The same MS. has Tr^ immediately followed by "prrin in 30, 31, but the P. is here correct since it stands for S. in the Sbl. system. We occasionally meet v/ith forms which remind us of the Sbl. vocalization. Thus B has a^rxn 24, 50 and bii'^N^ 27, 25. J, B and C write -[Biiinx^ 28, 15 (but D -jD'^ivx'i) A and D have xryin 3, 19 which we may take as a Hebraism; the others more correctly Nnrnn. The MSS. read xnVnT and xrlna but xrbni and xrisn^ would be more correct forms since in Syriac these words are written with JLil,^o^», MSS. read xs^S. 23, 15. B. has "^rn. L. says that in this word and in xix the suffix of the ist Person Singular is wanting^. In 14, 10 the MSS. imitate » ]. c. s. V. 2 1. c. s. V. 3 So also in Christian Palestinian Aramaic the suffix of the First Person Singular is wanting in the word \^\ which = »*ai "my father' (cf. Noldeke's Article in the ZDMG. 1. c. p. 5i4\ D - so — the Hebrew original reading ■p"'i ^"'3 but Pathsegen reads as B. viz I'^T? and this is also L.'s reading. The word being of common Gender may perhaps assume either the Masculine or Feminine ending in its plural. A, B, C read Nsos but D has ~ED3 which is closer to the Hebrew original. B, C, D read xn'x 18, 6 but A ""Nnix which latter reading is pre- ferred by Berliner', Landauer^, Levy^^, Pathsegen'' but Adler^ prefers xn'N as B, C and D have. In the same verse A has VXG, B and D yxD, C "1x6 whilst B. writes T'Xp which, says L. ^ is the correct form. 18, 15 A has TO'-'Ti, B, Cand D TQ^'f. A's reading, says the author of the Pathsegen is the Trg. of J^pnrj. 18, 21, C -pnNl, B. TT^^, B, A and Z> V'-nx^. Both forms T'nx and I'nx are allowed by L. but the latter corresponds to the Syriac form. 19, g A, B, C 3^p, B. and D 3i"ip. L. 7 gives the latter form, but Merx^ the former. (V. Pt. II, II, 3i, note). Bib. Aram. 3^p Dan. 3, 26. 19, 13. In this verse B. gives the better reading, having "linrb^np which is likewise given by L. All the MSS. read "iinns'^ip here. B. and C read I'-nn-'tcn 50, 20 whilst A, B and D have I'raiin. L. has the former reading. We may compare the Arabic in which w*«D: which is a Hebraism. B and B. have xV?ii 30, 39 but A, D and C "Vpi-. The Hebrew is /il"?F?:. A, B and D have iVrrV; 31, 2 whilst B. and C give "i^sn'i^. The correct form, according to T., is "'ini?. B, C, D rn2±5 31, 26, ^ x'^^^ija, B. x^^tts. The first ment- ioned reading appears preferable since L. punctuates the word with H. and it is in the construct state. A and B ni'in 31, 27, C and D n^^ri. B. xn"^.in. L. writes n■'^n^ B. and B are correct in reading "O-i 31, 49 which reading is also supported by L., although Pathsegen prefers ■]0"' which C reads. A and D (incorrectly) "i6i\ .^4 is again at fault in 'P^ 32, 14, the fuller form "(nXD being preferable. Cf. the Syriac rr'^J^- 33, 12 A, B bVj, B., C and D bbij. Pathsegen also brji3. Our Editions of Rasi's Commentary to the Pentateuch give his reading of this passage as V,::^ but according to Luz. "^Oheb Ger' and L. Rasi also gave the correct reading bli:"i3 which was only altered by ignorant copyists who strove to bring the Trg. into harmony with Rasi's interpretation of the word !^^"C>:. 2)Z} 14 B. xnnn'-s, MSS. '^T-r-'zb . Pathbegen and L. read as B. but it may be remarked that the words are synonyms, and hence either reading is allowable. C's reading "^^it 'my grief 35, 18 is a more literal rendering of the Hebrew "'lis* than is that of B. A, B and D who write v'l %rief'. 35, 17. The MSS. reading "pBnin is more correct than that of B. who has l"'^)j'!ifl. L. gives the same form as that of the MSS. 37, 23 B. n'^rssin'^s, MSS. rr^i'-ni. The Hebrew text is "insna. B. seems to be a Heb- raism. L. gives a similar reading to that of the MSS. 39, 23 B. "^^r;, MSS. "Tn. The latter reading is again preferable and is given by L. We may compare the Syriac form -U. 41, 25 A (incorrectly) T^n?, B, C and D ivs. B. I'n? which is a Hebraism. Cf. Syriac J-.^^. 43, n MSS. nin'iin'na which Berliner, in his Massorah^, following Luz. Lan- I We may also compare the form with final n which occasionally occurs in Biblical Aramaic. Cf. nr-Trt Dan. 2, 41. 2 p. 117. — 53 — dauer and the nmnn nio^^ all declare a false reading. B. nvflans. 45, ^ A, B, C li'.-ip, D ii"^-p, B. nu-p. Z>'s reading is here the best, and is confirmed by reference to the Chrestomathies of Merx and Levy^ 46, 29. The MSS. give the curious reading '^nis'^rin but since the Hebrew has the Singular form ■inss-;-? the reading of B. who gives rr^^'-'r:' appears preferable. 49, 12 B., A, B and D "in"', C '("ini. Both L. and Merx prefer the reading of C. 7) Grammatical Variations. A. The Nominal States. a) Absolute and Emphatic (VIII, la, a). C, D and B. ''St ^'^^n•'b t, 2, A and B vcz^. Hebrew is min-j. A reads snx 8, 3; 12, 6; 13, 6. The rest xrix. The Hebrew in each case has y^xn. B., C and D in^ri, A and B x^nl'j 15, 18. Text is 'iriDia. 18, 14, MSS. '^i. B. x:^^':. The Hebrew is x>bb. 20, 4 B. and C ^aa (ina), A, B, D s^a;. Hebrew r-^xn. 21, 2 ^ ■|'2T5. B., B, Cand D xr^fb'. Hebrew "isia?, 19, 26 B., ^ and C npian, ^ and D xnb^n. Hebrew n3-2. 28, 17 B. x-nrx, MSS. ^i^S. Hebrew ='P"2n. 31, 39 B niini B., ^, ZP and C xn-^nn-i. Hebrew ns^J. n, 20 C nsi^. B., A, B and Z) xnail Hebrew natp. 34, 10 B. and Cx-inp (Pathsegen Hkewise). A, B and D xmino. 37, 31 B. and C x-,^s^. y^, B and Z> "i^s^. Hebrew n^r:;. 38, 2 B. and C ^1515, ^ and Z> '^ar, ^ xnjn. Hebrew "^i^ss. (For variant reading, see Part 11; II, 7J). P) Absolute and Construct (VIII, i a, 3j. 2, 9 and 17 A and ^ rV-x'i. Z> and B. "'b=XT (L., likewise). C missing. (B) Number (VIII, i b). Several variations are notice- able with the verb mn. In i, 14 ^ has x-in: -n- whilst B., B, C and D give ■pn'^ns I'fln. ^'s reading is probably based upon the reading of the Hebrew text nrx-2 ■'n-; but it must be remarked as Luz. has already pointed out in his 1 Contained in Adler's Edition of the Targum. 2 Cf. also Daniel 3, 26 a-)!. — 54 — ia nnx that it is contrary to the custom of the Aramaic language to place a verb in the Singular whose subject is in the Plural, even though the Verb precedes the Subject. In the Hebrew and Arabic languages this construction is permissible and is frequently employed. MSS. "^n^ B. 'I^Tn I, 29; g, 15 and 47, 24. In each of these cases the Hebrew is fi;^?!'?. In 47, 24 B.'s reading is the only cor- rect one, but the reading given by the MSS., may perhaps be justified in the other two examples, through the subjects being collective; indeed i<^i 9; 15 ^^y be taken as Sing- ular. B. ■,"in> MSS. 'n'^'':>' 9, 26. The Hebrew is ib. B.'s read- ing is given in Berliner's Massorah Parva but the other reading is more literal. Some confusion appears to exist whether to use an adjective or the Plural of a noun when describing nations. Thus A, B, D "ixS^IS 10, 18 (B. and C ni<:s1:). A, B and C nN::;l5 24, 3 (B. and D ■'XJytS) B., A, C and D 1x3^35 24, 37 (^ nxbtS). B., A, C and D nxislai 34, 30 {B '^.s::'l3:2). B., A, C and D i6<^in 30, 21 {B riN^in). In 16, 12. A^ D and C (text) have the incorrect reading -''"i ^ir^ B and C (margin) ■pi''"n:i "i^Tr which is also given by B. and confirmed by L. 17, 13, A "^.?'^2ti "^T?"]. B, C, D and B. "par i^b=^. Now, although the sense of the whole passage might justify a plural here, still the form of the original being Singular confirms the latter reading, which is likewise given by L. 21, 12 ^ 'T!)?T'^., B, C, D ')n^i5n\ The Hebrew text xips- is Sin- gular the Subject of the Verb being "'nt but we require a Plural in the Trg. since the Subject is 'pin. 25, 26 B (in- correctly) in",iii A, C, D and B. rt^-r^. The original is i'l^'i. B is again incorrect in reading S^'^n 26, 15 which should be "^nin as the other MSS. and B. give. 27, 15 A (incorrectly) xn'^D'i. B, C, D and B. sn^^n which L. con- firms. 27, 27 B. "iTH MSS. ■'fn. 2-], 2>^ A, B "pnn, C, D and B. xrin. The latter is likewise given both by L. and the author of the Pathsegen; still it is strange why we should read "p"- in 27, 12 the Hebrew of both passages being — 55 — nr-^^. 27, 39 B. laniia which is also given by the Path- segen. MSS. have "(5ri^. Hebrew is TjaiT"*^. BerHner mentions the Variant reading in his Massorah, but offers no comment upon it. 30, 39 B. and C K^-^s^i. A, B and D iV?^', We require the Plur. Fern, here, hence "p"'^"'' is more correct. 31, 26, 43 and 50. B. T^a, MSS. "irTi. Hebrew T^; hence the reading given by the MSS, is correct. 31, 43 B., B and D ^rJ. A and C "^v. The Hebrew being •^:x:;, the first reading is preferable. 34, 5. B. "pn'^n-';. MSS. iri-nij, Pathsegen supports the reading given by B., but since the Hebrew is in:p:: we should e.xpect •^nwa as the MSS. 34, 24 B (incorrectly) "j^nrnp which appears to refer the V suffix both to Sehem and Hamor. The other MSS. have nTTip which corresponds to the Hebrew in'^r. 37, 4 ^ (in- correctly) "^niinx. The other MSS. 'I'ln'iax which corresponds to the Hebrew nn-^ns. The correct Trg. of rs-z 39, 5 is xr5ni which all the MSS. have. B. reads XJ^s"!^ (plural) which is incorrect. 41, 36 B '{^STT-'^y^, A, C, D and B. "^iviy^ Either reading may perhaps be justified since X'B? is a collective Noun, but the Singular is preferable. 46, 20 B. and B •n'^S'^r-^XT. C T'^'^n-'xn. A and D rrr^'n. Hebrew *Tb^ "iiyx. A and D are correct xa-a a-jvejiv. C is wrong. Landauer supports B.'s reading in his Massorah. 47, 30. B., B and C (incorrectly) ^rn£x. A and D "^rnix which is also given by L. The Hebrew text has ■^rsisx. 50, 9 B., B (original reading) and C ipJo'i. A, D and B (corrected reading) p^Vc^. Hebrew is br:;. (C) Gender (VIII, ic). A and B x6n-n B. and D «J^nnn I, 26. The subject is Nrn-^ which being of common gender, perhaps admits of either form; but L. reads as B. B. "pbrp, MSS. pzp 4, 10. Hebrew is ="P?s which how- ever refers to "'■^n whilst the Trg. refers to "rrm which is feminine and hence the reading of the MSS. is more correct. Pathsegen notices this variant reading and seems to expect that of the MSS. B. x^n, A, B, D x-ri, C (incorrectly) ir 14, 7. The Hebrew is X"n which agrees with 7? but x^"^ - 56 - is masculine in Aramaic and hence B.'s reading is the cor- rect one. A, B and D nrh% B. and C nlsn 15, 9. L. has the latter reading but points out that the Pesitta of this passage is IJA,ol lii^o. 16, 2 A (incorrectly) fT^Dp, the rest have n1^. 19, 8 B., A and C rJ-^si^, B )i-v-fi, D l^rin^. The subject being 135 D is of course a mistake. 20, 9 ^ (incorrectly) 17i?3; the rest have 'p^s. A, B and D are incorrect in having yh 20, 13 since -[nia'^-j is feminine. B, and C read xn. B is incorrect in reading st^n 19, 13 since K^nx is masculine. Rest have T^Sn. The Hebrew in each case is ni and this may perhaps account for the error of ^, B and D in 20, 13. 21, 30 B. and C Nnn, A, B and D "pSn. Since x^^^iii is given by L. as of common gender either reading may be justified; but ^'s reading N^n in 24, 58 and ^'s stin in 25, 30 are both evidently incorrect. The fact of t<^"'n being of common gender may again justify both readings in 26, 20 in which verse A reads rr^p^L whilst B., B, C and D give nm 27, 42 A and C (incorrectly) rTiS5, B., B and D n^n. 29, 3 B. I'lpiu^'i^ which L. also reads. MSS. 1P'>a6^. Pathisegen also as B. who gives the proper gender, the subject being the shepherds. 30, 37 A and Z> ')'ii-'6=i, B, C and B. "p-'iin. So also L. and Path.^egen, but since 'pio'in is of common gen- der we may take either reading as correct. 30, 40 all the MSS. have V'Dii^n^- which is the proper reading, B. has T^^^"'']? which is certainly incorrect. 30, 43 A ^^■'^0, B., B, C and D "jxijo. ^'s reading is more correct, "jsr being masculine. 31, 10 B., C and D "ppHon, A and B ipSo^ which is a mistake smce Ni^^n is masculme. A is again incorrect in reading Tl^'?'] Zl^ 7 and "p?"'j^ 4ij 3- In both instances we require the feminine. 39, 5 ^ ninl, B., A, C, D nin\ B is the correct reading; the other imitates the Hebrew Idiom. 41, 20 B and C read 'iirx but the 2nd time the word occurs in the verse •jirx. B., A and D have both times 'j'^ii'iN which is correct since the first one refers to xn^in and the second — 57 — one to i!<"3a'J and both these words are feminine. 48, 4 All the MSS. read I'r^-^snx here, whilst B. has ■,-?'''=''=r'K which reading is defended in Berliner's Massorah * referring, as it does, to the feminine "^n'^'^'^r' Landauer writes to the same effect in liis Massorah ^ But, it may be remarked, that 1;: is also sometimes masculine. Cf. Targum Jonathan to Isaiah 13, 7 ■"'V'P'!'^^'^ x^"?"! '=• (D) The Conjugations of the Verb. a. Peal and Pael (VlII, 2 a, a). A rnJi, B., B, C and D rnti 8, 4. B. p"^w::^, MSS. p^\atn 27, 27; 29, 13; 50, i. A niaiii, B., B, C, D n-i^n- 31, 20. B •:Jir^", B., A, C, D ;i;"£4t 31, 34. A, B, C nns^, D rnti, B. n'^n:- 31, 40- B. -lT-f^ MSS. n^-J^ 35, 4. A and ^ Srh-t, B. and C 'nVr, D inVu; 42, 16. Aramaic requires a Pael here. ^ and B perhaps a Hebraism. 43, 4 ^ (incorrectly) "pitl^ = we shall sell, it should be "^ir^ we shall buy, which is given by A, C, D and B. 43, 7 MSS. ?"^x£". B. bxir just like the Hebrew text. L. has the same orthography as the MSS, but I think b'^X'i; would be a preferable punctuation, since L. gives no Pael of this root. 45, 24 A (incorrectly) n^MJi. B., B, C and D nri;=i. 50, 21 B. anr, MSS. n-in:\ Although L. mentions both these forms as Pael of =~? still the only correct Aramaic form of the Pael is that given by the MSS. p. Peal and Afel (VIII, 2 a, .3). B (incorrectly) ni:^^^ 12, \i, A, C and D n-J-'^n, B. rj-^:"!. The latter reading is supported by the authority of L. B. ""^rx, B, D *i5x, A, C i^-ix 1 8, 13. Of these three readings the Afel is cer- tainly incorrect. Of the tw^o other forms given B.'s is preferable, although the correct form is "I'^'l'S. B nir", B., A, C and D n£:'i<=^ 32, 23. L. as B. Y. Peal and Etpeel, Etpaal or Ettafal (VIII, 2 a, 7). B. -in and i-^ts. MSS. "Isnii and i^nrx 17, 26 and 27. (Vide VII, li). B. rRr-^n- 25, 23 (Peal) whilst the MSS. read 1 1. c. p. 78. 2 S. V. - 58 -- Ciprrii (Etpaal). In this instance^ we may perhaps take either reading, since both give good sense, the Etpaal having a reflective sense. 6) Pael and Afel (VIII, 2a, 0). A, B n^^n. B., C and D ri^sn 3, 18. The latter reading is also given by L. B. ''•^'^xi^, A, C, D ''■'m% B I'iiiJNi 13, 16. Both the Pael and Afel forms give sense here. Pael^I have made num- erous, Afel = I have made equally numerous. B. has an inferior Aramaic form. e) Etpeel and Etpaal (VIII, 2 a, s). 7,11 A, D itinnsnx, B., B and C Ninnsnx. L. also reads Etpaal, but in the Syriac language the Etpeel (and not Etpaal) is the conjugation in use\ 8, 2 MSS. iirfONi, B. ii^sripNi. L. again has Etpaal but according to Bernstein-Kirsch, the Etpeel is the usual Syriac conjugation. 10, 9 and 27, 13 MSS. ^6iii but L. has the reading D^*n which is also supported by Pathsegen, ac- cording to the use of the Aramaic language. 41, 45 ^• in-'T^-j, A, B and Z? "p^j-an, C (just as the Hebrew) ntcs ■niH. L. reads as B. ''a man who reveals secrets". The best reading appears to be that of C, who does not attempt — 6o — to give a rendering of the name. Besides it is by no means certain that these two Egyptian words mean "revealer of secrets" (as the Trg. and PeSitta). Gesenius* explains the title to signify "saviour of the age" in the Egyptian lang- uage ^ 44, 15 B. and C)^V\'^"T,, A, B and D '\^ry•S^. Here the Hebrew nrs^'i'] has the sense of a Perfect and hence the reading of A, B and D is better. 47, 22 A ^l3?N^, B,, B, C and D V^rx^. The Hebrew text is ^ib^xi and the sense ' T T : of the passage appears to require a Past Tense and there- fore the reading of A is preferable. It may be noticed that in the instances quoted above the MSS. follow the Hebrew text literally, but B. gives us the sense of the Hebrew text. T) Imperfect and Present (VIII, 2 b, y) ^j ^ and D 1=13^ (Imperfect), B. and C Vi^ (Present) 48, 10. The latter reading is perhaps better. The Hebrew is ^a^i''. (F) Additions (IX, 2). A and D read rr^n^-^rji before "^fi^^s 7, 16. C inserts the same word on the margin. This version represents the opinion of those who wish to interpret the words p^;2 11 ^'i5G';i figuratively, to denote generally that the Divine protection encompassed and preserved him. 14, 12 C has bi before riijisp probably confusing this verse with the preceding. 18, 2t^ A, B, D have 'p^^n which is pointed out by B. ^ as an incorrect reading and so also the insertion of ^ before ti'^a which A has in the same verse. Adler and Pathsegen also both object to this reading since the Targumist always avoids irreverent language when speaking of the I 1. c. s. V. 2 The LXX has 'Lov9oiJ.cpavrj/, Siegfried and Stade in their "Ilebraisches Worterbuch zum alten Testamente (Leipzig, 1893) give njye nsE".: "Egyptian title of Joseph. Accord- ing to G. Steindcrff, Zeitschr. f. agypt. Sprache u. Altertumskunde XXVII, 41 f. — es spricht der Gott und er lebt". Jerome renders "sal- vator mundi" but according to Rosellini the name means sustentator •vitae, support or sustainer of life. The Targum version seems to be the popular Jewish interpretation. Cf. Josephus Ant. II, 6. V. Keil and Delitzsch, Commentary to the Pentateuch, Vol. I, p. 35. Trans- lated by Martin. (Edinburgh 1874). 3 1. c, p. 221. e — 6i — Deity. But, since it is a more literal translation of the original Hebrew the manuscript reading seems preferable. 19, 34 B. mi-o^, A, B and D "^Tr-an, C "vi^nS. The Hebrew is r.x. L. seems to know of no reading "'•^^"1 but Landauer in his Massorah ^ gives both readings and also explains the - of ■'U;'2"i2 as standing for an accusative of time. The best reading would then be Xii"3"a. I do not know why the suffix of the I St Person Singular is affixed to this word. 23, 8. The MSS. insert n^x before mm which improves the sense and follows the original. 24, 67 C has ni^-^x n-,6n before xirn thus first giving the literal translation before proceeding to the Hag. rendering which follows and which may perhaps with advantage be deleted. 25, 8 ^ has T^'i'' after r£D'. Since D has the same word on the margin, we may perhaps take this word as a gloss which has crept into the text of B. 26, 26 B., C and D ''■Hrarn, A and B Tiirinn-s. L. rejects the latter reading which is against the spirit of the Aramaic idiom. We must then read, either "riirnn or "^rii^nn^ which have exactly the same significations, the "c being part of the word and not a preposition. 31,13 MSS. r\'^iz.. B. r.i?. Hebrew TTO.. The sense justifies the reading of the MSS. and although Pathsegen also reads n"'3 he seemed to have expected the read- ing n^an and hence we may take the reading of the MSS. as correct. 31, 48 ^ incorrectly inserts n-"? before "p"'". 42, 28 A has C!iE2 before "'Jsrj but the scribe was probably looking at the preceding verse. 34, 14 B has n"i before S'sar'^B. 39, 6 B and C insert Tr^ before 5=. 43, 29 B has n-' be- fore "^rTirs. It may be noticed thas these insertions, the majority of which appear to be incorrect are confined to B and C. Finally, I have to mention a few Prepositions which are here and there added. h is added by C in nsV 17, 5 and il^s' 17, 10, A, C and D have TcnV 29, 3. A has X2sb and -p^-^ 45, 8 which reading is rejected by Pathsegen. A, B and D bxr-iV 49, 8 1 1. c. — 62 — which Merx also reads. i is added by the MSS, X^WSa*! 9, lo which is against the Hebrew text and in "nni'or 33, 13 which follows the Hebrew original. '2 is added in iii'ionn^a 26, 26 by A and B but this has already been noticed as a faulty reading. A, D add it'ON before iTn'iii 45^ 8 and this is really the sense of the passage. The same reading is given in Adler's edition of the Trg. The MSS. add an initial 1 in D'^^ii's-' 15, 18 which A and C read. A has nnNn 20, 5 and again inTi 24, 22. B (incorrectly) 'rrr^i'incn 35, 5. All the MSS. read xsb^'i 40, i. A has c:"'V>^l 42, 6. C and Z> have Q^-^a-inT 45, 13. All MSS. read "^fiinKn 46, i. B and C give b^'n 50, 13. a is added by A, B and D in the word n^iin 31, 13. (G) Omissions (IX, 3). An initial i is omitted^ by C before Xi" 8, 21 but this alters the sense of the passage. By A before '^tr^ 20, 13 which makes this remarkable Hag. passage run much more smoothly. By all MSS. before Qx 44, 27 where again an initial i hardly seems necessary. (H) Interchange of Prepositions and other Particles. (IX, 5) B (incorrectly) -(-n'ra^Bn A, C, D and B. -pnsiu^bb 10, 20. C (incorrectly) xrix hv 44, 1 1. B., A, B and Z> x^iKP". Hebrew n:jnx. In connection with this instance T : T and the following we may compare Nehemiah 3, 37 where according to the occidentales (Palaestinenses) we should read br and according to the orientales^ (Babylonii) bs. 20, 2 B and B> ii'w'f, A, C and B. iT=,\£ by 3. Hebrew is bs* to which ? corresponds in Aramaic. 22, 12 MSS. » Or 2374 sometimes omits 1 where it is inserted in Baer's Edition of the Aramaic portions of the Bible. Cf. '••};_ Daniel 2, 43, -prirVa-jS^ Daniel 3, 21, Trt'J'. Daniel 4, 9 in all of which the MS. omits the initial 1. The Scribe however inserted this '. in "r.'izV'. Daniel 4, 13 in which case it is omitted in Baer's Edition. 2 Vide Baer, (]. c), p. 125. 3 "There is a tendency in Hebr. esp. in S, K, Je, Ez, to use Vs in the sense of hv (and vice versa) some- times !;s being used quite exceptionally in a phrase or construction which regularly and in ace. with analogy has h'j, sometimes the 2 preps. - 63 - t K^nsb'; B, C and B. xrnV \c which appears more correct. The Hebrew text is =r;;b. 3, 12 B ■'b A, C, D and B. ^oi'. ^'s reading is inferior, al- though it gives sense. 3, 17 u5 xn. The rest, more cor- rectly ■'Sx. The Hebrew is "^3. 3, 23 B., A, C, D xrnxa, B xr-ix n\ The text being nT2nsn nx I prefer j9's reading and the Pathsegen also seems to expect ^'s reading. 4, 8 B and B. b£nf, A and C b£n5, D b£n r-i . Adler in "iJb nJ'^na justifies the reading of A and C; he says we never find the Verb np construed with bx but with 3 or br. B. and j9's reading would then be a Hebraism; whilst Z>'s reading is inferior, inasmuch as it indicates a slower move- ment than that shown by the Hebrew text. 6, 13 B., A, C, D D--, B '1^. Hebrew is rx = with. The sense of -ff's reading is "from off the face of the earth" but this idea is not contained in the Hebrew nx and hence ^'s reading is again incorrect. 14, 7 B. x^in, C 1^, A, B and D X'^ri. C's reading is incorrect; (as regards the difference of Gender, V. Pt. II, II, 7C). 15, 16 B., A, B, D x=bn, C x=n. Hebrew is nan. Both readings are correct; both are composed of the words xn and X3 but nsbn has a liquid b inserted. Cf, interchanging apparently without discrimination in the same or parallel sentences. It is prob. that this interchange, at least in many cases, is not original but due to transcribers" i^Heb. and Eng. Lexicon of O. Test, by Brown, Briggs & Driver, Oxford 1S92. s. v. Vk) cf. also Jeremiah 27, 19. - 64 - Syriac liioi. i8, 19 B (incorrectly) ii^hi', B., A, C and Z> "^ri'br but, on the other hand, £'s reading Hi^r is preferable to that of the remaining MSS. and B. viz ^'^^x:^p 18, 29. The Hebrew is I'^^N. Perhaps %Ti^"ip was used to avoid the anthropomorphism ''in His presence" and not ''with Him". 19, 34 A (incorrectly) rr, B., B, C, D dsj. Hebrew is nx == with. 24, 7 B '^'^3, A, C, Z> and B. ^3. Hebrew "h. B's reading appears to be preferable since Dp is usually construed with ^r in Aramaic. 20, 40 A, B and C ip^p. D and B. "J^i'. This latter reading is the correct one, the text being ~x, although the other reading may perhaps be justified. 27, 6 B n^, A, C, D and B. 'p. Although Landauer justifies the latter reading in his Massorah, still we must notice that the Hebrew nx is here merely the sign of the Accusative, which would justify ^'s reading. 27, 37 B N=n A, C, D and B. '^. Since it is rather time than place that is here indicated by s<"isx, j5's reading seems faulty. A xa, B, C, D and B. rh 28, 15. The Hebrew is "im n!!< and A gives the sense of this relative more clearly. B (incor- rectly) xni 31, 5, B., A, C, D -IN. Hebrew is ''S. B (in- correctly) '$z. Rest s<=n 31, 37. The latter reading cor- responds to the Hebrew Ks = here. B is again wrong in 31, 43 where he reads "i? exactly as the Hebrew text, but the Aramaic idiom here requires a 1 and hence the reading of B., A, C and D viz "iBin is correct, B is again wrong in 35, I where he has "]^S" corresponding to the Hebrew ^"5N. The other MSS. have "i^* and K^J is usually construed with b. The less emphatic Particles are the correct ones in 19, 38 viz '^-\ (and not as B linii) and in 35, 17 and 20 T^-i (and not "^!^ as A and C have). 37, 35 B (text) br. B (margin), A, C, D n-b'. Hebrew is bx. Rasi says that here bx = b?^ which would justify ^'s original reading, but n^V is cer- tainly more literal. 38, 12 B. has nib; all the MSS. read I V. Note to 20, 2 above. - 6s - hi. The unknown author of the Pathsegen also reads as B. mb is usually the Trg. for 3X whilst the Preposition by is identical in both languages. Hence the text of the MSS. is superior here. 39, 7 B. and C tp^'^z, B ~;0"i"'3', A and D qoi"^ n-V. The Hebrew is bx. Although the MSS. readings may be defended still that of B. and C is here the best. So also Pathsegen. cipt is usually construed with 2. Cf. Trg. to Deut. 32,4. 42, 28 C has X'.n spit which is not such a good rendering of the Hebrew nsn 051 as the other MSS. and B. have, these give xn qx". I do not know why B. has ■jnib as the Trg. of ~\> in 43, 9. All four MSS. read ~^. 43, II B., i^ (te.xt), C and D have xin whilst A and B (margin) read N=n. The Hebrew text is xisx to which word xSn usually corresponds, but in this particular verse X'EX is merely an emphatic particle. 43, 23. The Hebrew text being "^bx the reading of the MSS. "niV appears to be superior to that of B. which has "^5. On the other hand B.'s reading is preferable in 45, 10 where he reads "^r^b and the MSS. ■^^. The Hebrew is again "^sx. 44, 4. In this verse we have another special use of tlie Hebrew word nx, the sense of which is 'from' and hence the MSS. render "li. B. (incorrectly) n;^. For a similar use of rx cf. "r^rn rx "'rxs: (Exodus 9, 29) where, however, all the MSS. have r.^. 47, 10 B (incorrectly) n-b^, A, C, D and B. aip i-i. The Hebrew text is "^^sb-:. (I) Hebraisms (VII, ig). Of actual Hebraisms C presents the most instances. I have previously remarked that this MS. does not differ so greatly from the Trg. text of B. as the other MSS. do. Yet B reads 3?r^ 1"= 27, 17. C has •'rinn 27, 34 and 38 and QuJ 11, 29; 21. 3 whilst B reads the same word in 41, 45. B has p (for -^S) 21, 9. All the MSS. read ■^"'^; 27, 9, when we should expect ■'■is as B. A has ^T^"^"^ 4I; 55 and 47, 15. C and Z> "pb" 4, 15. The rest read "li. C has 2-p"^ 37, iS. It should be s-'T. All have nxbn^ 35, 21 &c. B and D itSTr 26, n- B's Heb- raism ir 38, I and that of all the MSS. "i-rn 43, 6 have ^ 66 — already been noticed above, when speaking of the variations in the use of the Particles. (J) Variant readings (X). A and D niiinn (so also B.) 1, II. B p"'En. As a mere translation of the Hebrew words ^^^. 'O^'^ ^^1*^ ^^^ version of JS would suffice "Let the earth bring forth green". But as the Trg., in most cases^ not only tries to translate the original but actually goes out of its way to obtain an identical form, the reading of A, D and B. seems superior. 2, 8 A "^W^ B and B> "inrx^f. B. "^yii^). Pathsegen ob- jects to A's reading "'^itt;'! xpI DJin i-i^xi" Nevertheless A gives a more literal rendering of the Hebrew diu;;'! than does the word "'■iiUNi — which means "And he made to dwell". 2, 14 A ^i^'^^, B, D and B. n^an. A is probably a Hebraism, the Aramaic name of the Tigris being rts'i. Syriac i^;. Assyrian Tiglat^ Arabic Sl^J. L. says the Hebrew form of the word has a prosthetic n just as rts^n from bs3. 2, 24 B., A and D have here n">aKi ^'^1iN "^SaTiJa trri but B tri-cx n^i "iri-nx n-3. The latter reading being quite literal is superior to the former. The Hebrew text is iax nx'i T^ax mx. The other version has reference to a traditional translation of these words by the Rabbis Eliezer and Akiba4. 3, i B. d^n^, MSS. n-'iti. To understand the reading of the MSS., we may compare 27, 35 where na"!^:: is rendered irasTinS in the Trg.; but according to the Pathsegen t]''2n is only applied to men. The Syriac jAaai. is also used for cunning. ^, 22 B (incorrectly) xnDX The rest have i<^'^n. Heb. n^":nri. 4, 3 B. and Z> i<3a-inp, A xr^si^pr, B xnnnpn, C xrnD?D. Which is correct? The I vSamaritan niiBi. 2 Or Dignat, Tignat, Diqlat (?). Cf. also LXX. TiYpu. 3 So Samaritan nax r«l n-as rr. 4 V. Ber. Mas- sorah p. 117 and Adler "-h nrrj to this verse. S According to Dalman Jewish Aramaic does not emply the root d3n in the simple signification of "to know", this use being confined to Galilean Aramaic. The corresponding Jewish Aramaic root is :>T' (1. c. p. 38). - C-J - Hebrew is f^C?^'- ^- ^"^ Pathsegen both give 5<33i"'p, and if the word nni^a here means 'offering' this is the only correct translation. But it must be borne in mind that the ^ -* *» word can also mean a present (Arabic ^ = to give) and this would justify both A and Cs reading. B is wrong. Luz. says the reading snmipn is bad, as this word is only used when speaking of persons making presents to each other and never in connection with God ^ 4, 2 1 B. IJ'Mn, A 133^1, B •)1'a^, C lt"cn, D ll^n-i. As if we have not al- ready sufficient variations, Path^egen wishes to read "'SSjia'i or •|">D53t)'i. The Hebrew is ben. Pathsegen goes on to say that the reading "p^^i is a mistake, and I can find no sense in B., D, B or C, the readings of the two last-mentioned MSS. not being even correct Aramaic forms (but cf. Berliner p. 128 who justifies his reading). Yet /^'s reading seems to be the original one since A had this word in the text, but it was afterwards erased and ps^i placed on the margin in its place. Probably the whole phrase xbns ois br 'iSs^t should be deleted from the Trg. text as superfluous. For similar double trans- lations cf. 3, 21; 4, 21; 24, 21; 30, 8; 40, 10; 49, 4 and 49, 8 in the latter we have even a treble rendering. 5, 3 B. n-i? f^"'r"7'!, A, C, D n-^s -^-iDnn, B n-^^s-B. The Hebrew is 'i'abss and hence ^'s reading is the most literal. Of course the other readings give the sense of the Hebrew passage. Path^egen reads z.% A, C, D. 5, 24 B. has n^Tax tr^n^, A has n't^3, B., B, C, D Nsnx which is of course the correct Trg. of y^^iji^. A means Vild beasts' which gives sense but is not the correct translation of the Hebrew, lo, 30 B. Tin^ni^, B, C, D i^nnn-itj, A Tin-iia ni^! A\ reading is an absurdity ^ 11, 3 B., A, C, D rvh'^n'ii, B ''ninSb''. Hebrew in?"!. The readings are here equally good. 11, 3 B j\T^t''i'i'', B., A, C, D iO'i-'"ip^3i. Hebrew iiE'^ip:';. B. is incorrect since ip"^ is (according to L.) Intransitive. Similarly Luz. says in his "i5 rsnx* that ^p"^ is only used for entirely consuming and not for merely drying. Pathsegen also justifies the reading Tirs-iasi. 11, 6 C 'ix^nujT B., A, B, D lii-ifln. C has simply confused the end of the verse with the first iK-'-im. II, 31 B., j9 and C "il"i% A, D 3">63i. The first- mentioned reading is more correct 5. There are two Aramaic Verbs corresponding to the Hebrew root npb (i) "la'i when speaking of to lead from one place to another (2) 303 simply to take, or with ttnrx = to marry. But some confusion is noticeable in the use of these two Aramaic roots. B. has nn*! 14, 21 but all the MSS. here 36^. L. here again reads "nm. Again in 24, 67 B has "ilin, A, C, D and B. n-^DsV. Here it means to marry, hence B is wrong. Finally in 30, 9 B n3"'63"^, B., A, C, D nnnin in which case B is again incorrect, 13, 9 B. and C j<3i3"^:iV xrx, A, B, C (margin) and D -Iscix". Hebrew nb'^N^sm'i. The Trg. here renders i) Cf. Bezold's "Schatzhohle" Leipzig 1888, II, p. VI, also Qoran, Sura 19, 57 and 58 where Edris is identified by Abulfeda with Enoch (Cf. Abulfedae Historia Anteisiamica: Ed. Fleischer, Leipzig 1831, p. 13). 2 But i^n"3nm» (in one word) = "their seats" and would therefore be correct. 3 Likewise the Samaritan has ""snV. 4 1. c. p. 33. 5 Although the Samaritan has rosi, cf. also Singer 1. c. p. 21. 6 Samaritan ao. 7 Samaritan 2D3'.. ^ Samaritan raor. - 69 - the Hebrew text ad sensuni. The manuscript reading comes nearer to the Hebrew in form, whilst B. and C (text) make the verse run more smoothly in Aramaic. L. appears to read as B. since he gives no verbal root ")E3S in his Dictionary. 15, II B., ^ (text), Z? n^rii', C avi<^. A, B (margin) nnoii^. The difference in vocalization has already been treated of above. L. reads as A and B (margin) "he drove them away". On the other hand Luz. ^ takes the reading n-iexi as a marginal gloss, which, although incorrect, subsequently crept into the text. Pathsegen^ says the traditional Trg. is a^'nxi and this reading appears preferable, although so great an authority as L. reads otherwise. 15, 18 B. and Cn''in''3 A, B^ D T^PN. Hebrew "'nrj. B.'s reading is preferable, but the sense of the passage is Future, and therefore the manuscript reading, though inferior, may perhaps be justified. As regards the different roots used, L. solves our difficulty by telling us that the only parts of "iPJ in use in Aramaic are the Future and Infinitive. 17, 2 B., A, C, D ''"•a"'^ 'p^, B x^n^. Hebrew "irn. Path>egen supports the former version. This expression is another instance of the Targumist's constant endeavour to avoid anthropomorphic expressions; whilst B translates literally. The suffix of the ist Person Singular is wanting in this word and in xnx according to L. 5 Hence xra = ''ra. 18, 21 B. and C (text) have xV ox- T^i^n csi SJ^EPX, A, B, C (margin) and D read T^i^n cx^ T^i-n xb"* nx snsnx xs". The Hebrew text is simply nrnx xb cx\ All the commentators have remarks to make on this peculiarly dif- ficult passage. Luz., Pathsegen and Adler all justify B.'s Text "I will (make an end with you, come to terms with » 1. c. p. 35. 2 To this verse. 3 Samam.iu rzr'. 4 Sa- maritan 'J-a. 5 Dalman likewise mentions the avoidance of this suffix in his grammar p. 162 "Die Form des .SiifT. der i Pers. Sing. nach vokal. Auslaut soUte wohl T sein. Sie wird aber geflissentlich vermieden und bei as (und ns) durch die det. P'orm ersctzt". But he places k:"3 under those words in which the diphthong ai became weakened to a. Thus Nra = Ta of. also 'i-ys = •>3"""'y.i Sec. — 70 - you) forgive you, if you repent but if not, I shall punish you". The corrected (MS.) reading runs "I shall (make an end with you) destroy you, if you do not repent, but if you repent, I shall not punish you." Both readings, are, how- ever, very obscure and it is very doubtful whether either is correct. 19, 33 B., B and C n^P, A and Z> n^pi. The text is s1a'i3. The only noticeable difference between C and B is the fact that C (as usual) has a form exactly like that of B. whilst B preserves the punctuation of the Superlinear system more distinctly. The Hebrew is nV=\ Both Berliner and Landauer in their Massorahs give this word as an in- stance of QiJ"ini K-ip. Similarly, the author of the Pathsegen. According to these authorities, the first-mentioned reading is the correct one. But it has already been noticed that our MSS. do not follow these Massoretic rules of Qijnn xnp. At the same time i^:^"^ conveys the sense of the original and both Rasi4 and Ibn Ezra explain nbD^ = ~2n\ 24, 10 B I Likewise the Samaritan has rr^^pa'. nas'ija. 2 Samaritan rsva. 3 Samaritan •jvo'^. 4 a. 1. — 71 — •lEiy. B., A, C and D avj which is the same word as that given in the Hebrew text, and hence, perhaps, a Hebraism. 24, 19 B (incorrectly) "^P^a^b', A, B (margin), 6', J) and B. 'ipuj^b"'. 24, 21 B., A, D na ^■^6, B ni- -rri, C n£ ■'rr. Pathsegen accepts the latter reading, but L. writes as B- ami as early as Rami's time the reading of C was already pointed out as incorrect. The Hebrew text is n? nxn-r^. Berliner in his Massorah places nn "^rx^ as a Suraan reading whilst nm "irnu falls under the head of the Nehardean readings. 24, 38 B (text) "'r^ns^b'i ""i-ixb' B (margin) and the remaining texts "^n'^iilVn xrx nii-T. Since the corresponding Hebrew is •inrB'::^ bx' "^ax n-ia bs the last-mentioned text is more literal and hence, perhaps, preferable'. 24, 44 B. "'b-s-'X, A and B (margin) ■^b^x, B (text) vrx, C "'b-x, D "'Vrx. The Hebrew is ax'iiJx "I shall draw". Although none of the readings here mentioned give us a literal translation of the Hebrew still V^^x is undoubtedly incorrect, (for the differences of vocal- ization see above). 24, 59 B. and C ^•^''-la:, B and D ^ri^ai, A "^n-l-iaj. .-/ appears to have taken the Hebrew "^aisc in the sense of 'warriors' but there is nothing in the context which admits of such an interpretation. The vowel-changes have already been noticed. 24, 62. The Hebrew text runs xizti X2 F^^" 1X3 of which the Targum is — according to B '•v.^'i'i-ii ?r ■pnr>^ Knin ns whilst the other versions give xn-^i-2 "^nT-^i br prs" as a "Gafel" form of n^ij. Rasi also reads the word with a prosthetic 5. B has a mistake in the Gender. 30, 6 jff has ^n-b^ b^nj$ Cjx^ "-"^ ^nis5b b^ip. B., A, C and D simply I Edit. Sachau, Leipzig 1879 (page «»). 2 Samaritan nss. 3 1. c. p. 41. — 73 - read '^n'lVa b-^ip p.v which corresponds literally to the Pebrew ■>l?'ip? s^u; Da^. The first part of ^'s rendering is perhaps a marginal gloss which was afterwards inserted in the text by an ignorant scribe. 30, 15 B., D and B (margin) ns, A and B (text) niib'. C is missing. Since the Hebrew text reads <^b, the reading r>^ although giving the sense of the passage appears to be incorrect, 31, 2 B., A and C (text) '^xfr>->\ C (margin), ^rhr-^i , ^ -(^s-^rVJ, Z>"orr^;. The Hebrew is 13.3"'S. The forms given by B and D are incorrect. As the suffix refers to "^sx and not to pb Cs marginal correction is worthless. But '^dp'^V would be a more correct Aramaic form. 31, 13 C (incorrectly) -i^nn nVx. The other MSS. ead xnn i^r-iS which corresponds to the Hebrew r5<.:n y~tprr xiSs'i, C (margin) and the others i<5l3':ba3. The Hebrew being "i=f^ the latter reading is correct. C (text) gives the sense of the original but is not literal. 34, 7 B. ix-'pjrw, A, B, D •i<''6:n"'X% C iCo'dpni. The meaning of the former reading is "And the men were tried" which gives no sense here. On the other hand the reading of C — which is also given by L. — gives us the correct sense "And the men were grieved" which exactly corresponds to the Hebrew wssT'^^ In this verse then I think we should reject the reading given by B. in favour of that shown by C. 34, 7 B (text) '■nx, B (margin) and the rest 'ibs*. Hebrew 1x2. Which is the better reading? In verse 27 of this chapter all render ix:: by "br so that » LXX also Aeia. 2 Samaritan pr-V. 3 Samaritan V't'si. i Samaritan '*5y. — 74 — perhaps this reading should be preferred. But there are many passages in the Trg. in which the Aramaic Verbal Root which corresponds to the Hebrew K12 is xhn. We may perhaps then accept either of the above-mentioned readings. 35, 5 B ""npS 'pii. The rest have ''npsn x^^iS ^ which is the correct Trg. of the Hebrew '^ri'^s 'aV'ri. It is true that B gives us the exact sense of the original, but yet the reading is faulty. 35, 8 B n'lnni. Rest have 'S^h'o^. Hebrew inriF'a. As an instance of the close connection be- tween these two Aramaic words compare XSJip'nb 31^73 i, 7 and again i^'^^m mnn^ i, 9. The words are really synonyms but Pathsegen ■^ tries to make a distinction between them by saying that one refers to a greater depth than the other and that a fine distinction does exist in the use of the two words may be seen by comparing the Heb. of i, 7 and i, 9. I, 7 has h nnna whilst i, 9 has only nnn?3, the Trg. for i, 7 being h S"i^b and for i, 9 ninrra. As the h is also used here i^i^^ seems preferable. 35, 18 D (incorrectly) psn. The other MSS. and B. read ps^ii which is the correct Trg. of the Hebrew nxais. ;^6, 35 B., C and Z> ax^ "^^pf^^, A Dxa'i xbpria, B axo s'^n^. The original has ^Na trim of which A gives the most literal Aramaic. B is decidedly incorrect. 2,T, 2 A linai. The other MSS. and B. read linniD. Author- ities differ which of these readings we should prefer. Ber- liner in his Massorah"* remarks "Many copies have the in- correct reading "pnyi which imitates the word given in the Hebrew text; the Aramaic word a"^!: first bore the meaning of calumniating when joined with XTli'^a". Similarly Luz.s after citing many passages in which the Hebrew na'^i = the Ara- maic a^'u adds that the reading ■ini'i is nothing more nor less than a mistake. On the other hand L.^ remarks; "Since our root (ai;::) hardly ever occurs in the Babylonian Trg. in this signification (viz. that of calumniation) therefore the reading I Samaritan •^npy nra. 2 Samaritan ■s-hti. 3 To this verse. 4 p. 78. 5 Philoxenos p. 43. 6 s. v. — 75 — I'lnai is superior." We may therefore accept A's reading as the correct one. 37, 23 jB (margin) bs ', ^ (text), .-^, C, D and B. xrx. Hebrew X3 (see above 34, j). 37,35 i9 (text) "P'^in •^2, D and B (margin) Nlb-^is i:, B., A and C xb-rx nS. L. although mentioning the latter reading gives as correct the text given by D and B (margin). But how is the version of B (text) to be explained? It appears to be an allusion to the Jewish doctrine of the Immortality of the soul, but is, of course, an interpolation. The Hebrew original is simply bnx. 37, 2 B. ^?3;:^^ A, D "tsn, B xnan, C ":?:=. The Hebresv has "'??;?=. L. and Pathsegen both give the correct reading as ''x;?^?? although L. also mentions the other render- ing. But before we are able to give an opinion which reading is preferable, it is necessary to be quite clear about the signification of the Hebrew original and here the author- ities differ. Rasi, R. Samuel b. Meir, Nachmanides, Gerso- nides and Mendelssohn render "merchant" whilst the Revised English Version, Ibn Ezra, the LXX and Pe'sitta translate "Canaanite". Adler aptly remarks in his -5? nrn3 3 that the rendering "merchant" is given to spare Judah's honour, he also quotes Talmud and Midras in support of this trans- lation. But after all I think the most correct reading is that given by L. which is literal. 1%, 17 B. nV:i-"'•'' whilst the Editio Bomberg writes ■'"'\ A ""•'I and B '•V* \ The dot over the Sin is not written, Sin is written thus 45 but is usually replaced by D. The word Q^p is usually abbreviated in the MSS., which simply write 'p. In such cases the missing letters have been added in brackets. Numerous emendations appear to be necessary in the con- sonantal text, which should be purged from its numerous superfluous elements, but such corrections should be post- poned until we have attained a more accurate vocalization ^ Different chapters to those given by Merx in his "Chresto- mathia Targumica" have purposely been chosen. I Cf. Part I, Cp. II, 2. 2 Merx considers it necessary to apply both systems of vocalization to arrive at a correct and critical vocalized text, but it has been shown that this is possible by means of the Superlinear vocalization alone (cf. Merx "Bemerkungen", p. 188). CHAPTER XVII. DnnxS ''V ib";.ns*= ^-^zt yfrn '^"^^zr 13 z-ns rr.rr: (i) D''-5Tr '•'•im* ''^ip nSt '^^c bit s:n rf^ irxi Kin's '^xinb" "in^ ^-I'iDsn i^nn 'inti''^^''ii •^•c-p V»~kt (2) I'a'ir^ -flr niiir b^5-ci "^niBK bj nnns bi:\ (3) ^72^7 S'JD 2s': ^^rinn ^-cr ^■'li^^p '^ir: i?', C "prx-. t Bom. cbr. « Bom. ri^*?"* ^^ ""^"- •^" ^om. ^■?^^ y Bom. xn. :r Bom. ",iiar W. ,/a Bom. "!>-'?, C "in-is. M Bom. T^^ C" "^nn-. — So — ^•"Si i^D'^nnB xniDi bS f]^'5 w I'^^i'' x^'^D'an ^ai (12) xin isn'a xSi 'j^'atiy ^n b3T3 xsdd "i-inTi xnia ^"oy '^inii "isoS ''iinn in-in ^tS^ nn^ ^b)2 (13) ■'Wiu^i ''nin'5-iyT xnoi n"'> w i^Si iiiiDi bnyn (14) inr '"n'c© n^ inpri xS inri< in© onnnK^ ''V' n-asi (15) rraTiJ nnw "^nx 'nsDnnxT -in iS "^ns^ ^)^r» qx=i °nn^ "T^nxi (16) )Srv nla K^i'a^yn ^'^ti'jiri "I'^Db'a^ ')-wi5^ "inni na5n rr^nia "I'asi ''■'im 'inisi^ b^ annix bsan (17) ""'Tbr 'j^iw 'jiywri nan nnr dxi "iSi ^'^ri^ I'^iw nx^ '^'' D^'-ipn^i "bxy^tj^n '"'"■'■"ib ^Tr (d^)p oninsc nrixi (18) (>p < ■< — '^^ l^ip n'^ i^prii n5 i3 Tbn ^nni? nnr xnwpa iX"^ ^Srxi (19) ''m^na *innDn5 "nSy a^b rri^y ''''^'a^p tv ^^o^piji pns^ n^a© nifi^ iT'S'^na xn ^nibi n-iS^ap bxyiaTB^ bri (20) « A '^T^'?. (^ ^ T3'!^ c B, C XSDD, y^ i. g Bom. obs. /; Bom. n^nW"!. i Bom. nb;x. /^ B '^•u^p. / Bom. ^:m. ;« Bom. t^a. « Bom. Tin^?". o Bom. P^n^ / Bom. •Pix, ^ "prsi and omits rS^, q Bom. w^a. The 3rd Sin. Fem. Suffix is always punctuated with Q. in this Edition, so this variation will riot be noticed again. r Bom. M'^3"inN\ s Bom. -'nn. / Bom. T^t'isV u Bom. T^Vh- x Bom. "i-fn"!. y Bom. "'«n% s Bom. "ibi. a^z ^ T>^n. <5(^ ilH" is a more correct form, cc A, B bj<3)5au3\ ^a^Bom. ^'''!PT'.. pDi 'iinn'!;^r- x-^ci x*^,ci i"t5 -;: r:c 'rfr- r^'i^r nn nn^^x? (24) n, •• • ' - X'^ci r^ nr: "rr i^iir ■'^cj nbr -in rr^^n bxr^r^i (25) '"n'^rb-i'yn .. — !•• < ■— ■<-< -A n-inn bxy^r^- cn-^nx "itj-x ""'^n xr*"' '"^rn (26) ^5 112 XEcS '';"»ir xr^ii •>^"'b^ rr^r-^n "^ir-x bSi (27) nirj ^i-i'Tr-rx x'-c-^r CHAPTER XXVI. '^ah'^i rrf\-, 'nx^^p xibz-c ""in xr->xn x:rr rrn' (0 nn.Vr "^xrrb'E-i xzb-i ■ib'c''2X r-b pns^ 'b'Txi cn^nx "■'■^'c "c^ns^y r^n"*:" xb 'i^x^ "^V rpb' ■'br.rx^ (2) ii ^'-i-c-'X'^ xf-'xi a B, C n^x, ^ c-px. The best form would be n-px. /^ Bom. T'lr '■-\. c Bom. x-n, ^ x^xr. ,/ Bom. xr:-r-x, A xr:-nx, B xf:-nx, C xr:-r-x. e Bom. "'a-'r'. / B xrri;':, Ber. xVr-:5. a Bom. ^r";r-:. // Bom. •'crxn, ,■/ -i':x:. / .-J 7:r5->-n, Bom. vnr;nr-i. k Bom. 5"5r "". / Bom. V'tT^. m nir-D-^rn. « "^I^rx (B. on the other hand has "j: and also '-t:- for ^"Tjrx in verse 27). Bom. ?.-prx. ^ -£ would be a better punctuation. r Bom. nx-cnp. j- Bom. 3*x". / Bom. "-x". // B =~:i":5. .v Bom. ""ttj. y A -r"'Nj(!) — 82 - •ins "TDnnsi ''11^65 in^a^ia ''^nii xin xns5 "iiT (3) bS n^ ^iz^i 'i^rxi s-^r ^55iD5 ""iin ty^ roxi (4) N^^s '""^rsTGy b5 i:n b^in )i3^in^i i^bxn 'xrins in^i^ °mt:^ -ii::i ^n^^'ab nnnax "biip ^ibn (5) '^n^nixi ''^■a^p ^^iips nnsn pnsj^ ' n^n^i (6) ''nns n^si n-^nnx "pcir by xnnx 'iuj-n ibixtDi (7) sinx '"ir;s ^::ibt:p^ sicb-i ^nnsi nr'^'ab ^b'^rin "inx s-in xiri infi nnisir ^ni? ''''npni by T^i^iJ "^^rcxi x^'aT' 'Tcn n^5 '^"ii?\-D -o mm (8) ^^□y '^^-j^'^n^ pnsji xni XTm iiSnn )^ liipirbB-i xSb-a nirni? npn-i x^n -fnrii stn nna ^^-i^xi pn2iV5 Y-^^ix 5<"^pT (9) nby ^''b-'Dpns s^bi ■jns -iiyT5 s<:b ""'m5y "«in^ I'rti'iiN "tcsi (10) xnin N:by '^'^smr^i?T frns f^m ''°x^y5 irnai ""n^iir a Bom. '1-1-,. d Bom. "ins-s::. <; Bom. -;i"^='ii2Si. d Bom. "nx. e Bom. n~>~n. / Bom. ni-jirxi (correctly). ^ C '^i^p. /^ Bom. --Z. / Bom. "iRX". /^ Bom, l"in3. / Bom. xrr-x. ;;/ Bom. "i^^^;. ?i Bom. b^np •>-[. Bom. "'i-j-c. ^ Bom. i-iiip-'S. q B lii-p. r Bom. "r^l'ix':, B ■'ri^n^xi. j Bom. S'^n'^l. / Bom. ■'•r:x. /^ Bom. po'^". „t Bom. "rnx. y Bom. ^Ti. « Bom. ■'d;s. aa Bom. np::n (and not npsn as B.). <^/^ Bom. ^xi;d. ^(T "^iroN- would be a better reading, dd Bom. T'^'^^j C -"'xn-s. ee Bom. c:^' (B. has r,;:). ^ Bom. "ipxi.. gg Bom. r!-;-:x, ^ r^i^x. hh Bom. "rriN. zV Bom. T'lnisx. /'/f' j5{< correctly; also C b-^ipnx. //Bom.xn n-?. ;«w Bom.n-ins;. ii7i Bom. -■'=■::, C 3':"i\ 00 Cf. LXX which has ix tou ylvou; [jioo. ^^ Bom. =:■. ^^ Bom. xr-^r-^x-. The best reading would be xr^r-ix-. - 83 - ''x^ni's ''•p-^ri'^^t ^^^^'i xrr bz r^ ^r^^ix nips^ (i i) ''bi'jprp xrjpnx "^ n-rrxb' -<-.7\ bt- XTiri xrf5 nzcs" X"^nn s~"ix5 pni- 7^t- (12) '^ xin5 ''xSi- nr '^iini '■':c '"t'tx «5tki x-^n: xnii (13) ''io °x:nbET "i^nn °''r'«5=i x:y '"\^^3 n^5' 'inni (14) •'■"XPtr-E rr^d •'S'^:p=i annnx 'i:ci^5 '■>r;inx ^'^nr '^eht ''^n-^ii bri (15) x^E^ "■|-':V5^i ixrrtE ']'<:iT2i; ■'n-zx ^ns'ipn ■'lie x'l^i'TC ^'b'^rx pris^b '7"''c''2X "n-^xi (16) ""^xini x^ri-i pr "a-^rr nnr^-r xSn:3 x^ir^ pns^ ^^'ir-c b^rx' (i;) ''Tcrn "insni ir^-> ^'^^^'^2 rp lEm pnr^ 2ri (18) inpi cn-inx r^r^ "ir5 "ixrrbi ^^•j'^iT'c-^T ^^^r.^nx znnsx "Tninx ''■'•jinb ''''•'np "n^nT pTici "nrr '''''i'nb' "°V3 ]rri """nnzrxT xbn;n pr:i' ^■^nr "-nsn- (19; fl! Bom. p"'T:-n. ^ Bom. x~-;2. r Bom. r.-rrsr". d C b-^zjpn^. e Bom. "i-. / Bom. "^ninrr? ^ Bom. 2js:. // B has this word added on the margin. B. omits it. / Bom. '^;o. /: Bom. xr— ^n. / Bom. I'rr C -•■-'. /« Bom. ■r^?- « Bom. "r-';i. t> Bom. xrrrns?.. A has this word erased in the text but added on the margin; whilst B has it in the text and nn'^r* on the margin. / L' ^sr-'iz. (J Bom. "r-'rj ^ ^"^-- '' Bom. ""En ■-. s Bom. t-is. / Bom. "r-v and Ra^i quotes the word thus. // Bom. ": ■';-•. .V Bom. "•:s". y Bom. '"t-s. z Bom. N~S"pr, C ri'-f. aa Bom. sir;. M Bom. 'r-'^^- <-'c Bom. ::"r-- r/*/ Bom. n— z, ^ xn-^a. ^^ Bom. ^I'Er'-^n. // Bom. ttizx. jf^'- Bom. •,'i:'!':::5. The best reading would be 7:" :::;•. /th Bom. •,n3 (B. "n-). //■ Bom. r^\- — ;. kk C has "^pi instead of ^~p rirrr,. // Bom. ^^sn\ ////// C 'nzrx-, .-i ?r=rs*. //// Bom. N-sr. co Bom. ■":■?"!• PP Bom. r^a-i:. - 84 - ^x:':"'! n^'>"i£'5 pn'sr-* xr^iJn ay nnr^i ^xriji is;i (20) ^n:i:ir n^© xnpi nSy qs iSDi ^^nns ^"T^i insm (21) snpT nri? ix: xSi *'''ini? ^n^^n "lEni "i^rna p-roi^i (22) ynr njini ]^n)2 '"p^noxi (23) °n^n':x x:s "rasn xinn x^-'b-^bS ■'V' n-ib' "iS:nxi (24) 1-in^ onnnx b^i5 f]5 rn lir rp6n£i •'Iri xrri ^bsi 'xn5"!-c "irn ssni (25) sn-ti pn2^ •'■iny ]"an tidt 'n^DDTC'cb ^^ni^nn ''ny^ci nn:\)3 n^r.nb "xrx ■jS^'^aNT (26) n'^b'-'n nn biD^si ■(insi '>ri5 ''^■jin^rx ""^■j"'!^ pns^ linb 'n-cxi (27) •flri x^r'i'G '^nin ^ni? x:">Tn ^'xmia '^'^in'axi (28) "f^^ n^p nn:i "|d^5t x:r5 rt! Bom. xn':-'-. ^ Bom. xrb-in. <; Bom. xis'a. d Bom. T T -; ~ T T ■ T ; sippiTN. e Bom. x"-";;. / A, B, C •'-rh^, Bom. "^"^nix. The Sewa compositum given by D is irregular in the Spl. vocal- ization; "-r.'N would be the best reading. g Bom. n^"'^. h A, B, C --n'x, Bom. ""^nix. / C n'a">nn. k Bom. ipx:. i A, B (more correctly) riE-;- and ais-'b'i. m Bom. p''^^ ^, C (more correctly) p-bo'i. « j5 ^'^:r5<-. ^ Bom. !!<"';n. / Bom. -nzx. ^ Bom. "7"?:. r Bom. -:""i-:is:. j Bom. wxns^n. / Bom. n^Dsd^. ?/ Bom. ^ix, ^4 irs. .r A r"-c\ y Bom. "^n'.^an-^p, A, B '"'rhz'rrco, and the word is thus quoted in Gen. Rabb. c 64 § 9 (in Wilna Edition), z Bom. ~rX.". aa Bom. -|-.n nr. ^^ Bom. '\^rrr^. cc Bom. 7=r'b'3. //^Z Bom. ^"i^x.^ ^' i<^b NTD^3 K:"ci' ^-^arr ex (29) ^ns-^rtn ^ibSxi ^s^^ncia ^in'5 '^inrn (30) 'prni£n ''■•ninxS -13^ nis^^pi xnsin n^-'-^p^i (31) n^'5" is-^in^ pns^ ^^27 'inxT snnn XT2i-'5 nnnn (32) y5c -IS2 xn-ipi ''n^r )^5 by Pnymr nr^ xnpn (33) n-'inn^ n^ xrrx n^o:i riir "ivnix nn nry nini (34) nsn. ^ Bom. n-ixi. p Bom. nri'i: (and not as B. nrri). This is likewise the Bib. Aram. form. Cf. Daniel 4, 13. q Bom. N"2'r. /- Bom. "i;?"'';''. s Bom. nprnn-i (and not as B. np^-;'). t A Bom. i^'^'ca. So also B i-a-'-ab' (which is better than D). u Bom. =d;. .v Bom. N2!i:xb-^"!. y Bom. N:i::i<^ ^-yv^. z Bom. Tiri:rV?, C (text) 'i'":r-'5. On the margin ^"^•^■^;"'. <7^ Bom. -^n. - 88 - ""i^i '^nns tv ""-layi Dpi ^r^'^ii bSi sin '""stsi (21) npy^ bis ^ns nsn-ibn s^ii5 pS'i ^ninnsi (22) ^symrT^n^ iriinnS qini n^'ay ^riins tv nn'ii (23) 's^cbnn ns^ns pi nib il"! (ni)p ]Ta "raiic snsi (24) ni:^ npi?"^ =y b^brn s^bi ^b ''nrnox n'^b' n-asi s'-ibibi n^:5TS^ n"-' ons npy-i npy"^ n^ p5 '"piiisi (25) ";yb5i snitjn ""^ninx n"^ 'i-im pbi sni-jn ■^i^ '^sr^oSi Psriny "sti npy^b pb "i^si (26) "inn 'iSniijD ^^fzz n'^ '■^211 lb 'rriin sbi "liTi ^snx5i br-ab ''srn'a-j s^b (27) ^^p"i:ini 'i^Einn '^'^■jrinTUnni " siinn ^^)\^ '"''^nnbEi ^^snb^sDS p5 \'^:nbi ''lib sp£':b' ^^^inpiTi: sbi (28) ^^inyrb" "snbsi ^''.wi "jiDry "iiy^b" ^i^i sb^^ n^s (29) a Bom. 1:1X1. (^ Bom. rr^^ ^1. ^r Bom. n?^;^,. ^ ^ xnni r^ r-isn which is better. The correct text should be "i^nox. / Bom. T : V : 1-rx. m Bom. '^ninx. n Bom. "i^x\ ^ Bom. rrr. / Bom. mar. ^ Bom. xr-'ori. r Bom. 5nb6K\ /^Z" Bom. adds *|is before Niina. C has the same word here whilst ^ and D insert it on the margin. Vide Barth in ZDMG. XXX, p. 190. cc Bom. x^nnz. dd Bom. ina^ra. ee Bom. 'i-'"r=^% ^ 'r"!r==''- ff ""'i^T-^"^- gg Bom, Xv^?cx. hh Bom. "i?<':l?. ?V Bom. "i?"?^. /^/& Bom. 'J"3. // Bom. xnbxi (B. x^J^xi), ^; B, C sri?x\ - 89 - en 17 2-jt: np?-' 113X rr^i^ ''sni-'^n xiian "^nx ^xrblx bra -rs: (30) 'nin-as ins •'n-'^ini ■'nx 'li'i ^nrxT npj^ 'n\^x-, (31) ■^i^i ir:5 n^ ''oirr srbi ^ ^ — XDns D-p Qi^pn^ S5 ""inim n^ 'noirr^ irs (32) s-rir-ani nsbi xzzirrni np5?''iT xlrrrn pS bii (33) x:5Trrn by=i nsb'i j<:5ti;^^ ps:i nzrx xbi sr^nb )^rnr-> brill sra'rn S'j^ii'n ' iirns^ir^ nHib^ r^ ■"nn^c: bnni (34) nzTTs sn s;5Tr^ b5 n^ pb "t"'c^i i^n^by 'na^n'^^ S5 ins '^ibi iri?5 '^pn^ N5 ^sn^asb "n^iiri (^^) sSi ©bni -^b ^^r: nnis ins i^ip --6 ^^uf.^i "^biD-'s s^^:^b2 ni nscs "irsi npyi '^^ n^rsi pba si-:t npr-^r "qipm (36) ^''■'nrn snsin ins ^^linno sr linn ^^s-2 pbb flf i^ has X"£-:-z 13 -.-^n in the text, but margin as above. d C Text h-oir. x-^b-i. Margin as above. c Bom. nbix. ^ Bom. xrn-:~. e Bom. "^p"'""^. / Bom. -"^rx:. or Bom. "^rX:. /^ Bom. n-'b^nn. / Bom. "--ex. /> Bom. o^:r. / Bom. ■'t rrcn. ?// Bom. -;-;nn. n Bom. c^pr-^. t? Bom. nr. / Bom. i'H": (B. v-fi)- q Bom. "iirrir-c:. ;- Bom. n^ic?. J Bom. "'iST'^'ip':, B •,'"'3''r"'Tr^. /* Bom. ri^ir"i-. u B «£"!:•. .V Bom. rTzx". y Bom. xn^^xb. z Bom. ~'ipr\ (?r\ 'sni.-;T sn npy'^b )2^ 'tcst (51) T-'iT sri ^^r,'i'j2"'psT '"''nays sS s:sc 3s "snrp sinoi ■;''^n 'sni:\i i^rio (52) a Bom. X3r\ <^ Bom. "'■'Jr, yi "^rr, C "^.t: c Bom. "^•n A "^Ti-, B -^tr. ^/ Bom. ■^s"'-!, B ^"i. :. cc B "'iyr. ^V Bom. "r:3 (B. T^s). <-,f Bom. -•:"'X. ff Bom. x~^i^. gg Bom. n^'iT? ""• The vocalization should be r'^-pxn. hh Bom. ~"::'X, C -:J-x. — 92 — ^tiiib" sin "sniBp trn I'^in ^N:r5 ^]iri- Sth-t "n^inSsi onnnsi '^n^nSs (53) b5"'^5 °^riinsS s^npT snrjn i^roD: npi?"^ "oddt (54) ""T^ai n^rrnbi ^niDnS "^p^irDT snsrn pS n^i-piii (55) CHAPTER XLI. ^'D^iip s?ni "nin ii'j^i\ i^iir "i^rnn qio'a nini (i) Sinn: by '""•'"inrb piE"ir ""^pin ^^yar "ipbo sin! p sni (2) Jinn"] "i^a ^^iin^nnS "sp^bo f^sins pin ynir 5(ni (3) by s^niini ]inbnpb "s^^pi i65 ^•"in^om I'tmab -jir^n a Bom. NiJiJ-i. ^J- C "i3i-n. ^ Bom. xnrp. d Bom. n%ib!<. T : T T ' T " T V. e Bom. ""^n^x';, C 'n^rii^^. f Bom. "lin:"! (B. ^"in:), A VT^T'..' g Bom. 7Dn\ y^ y4 x:3i:. i Bom. ^n?N. i Bom. linnsx-. O ' : T- ■ ■• T v: f .. _ / Bom. b-n-i-is. tn Bom. "^n-ax. « Bom. 0"^=?^ B D'^ir. ^ Bom. N^:;^2. z Bom. nr. /C' A i-bts. / Bom. ■E^pii^si (B. ■jET"'^'^). /« Bom. in-:^ (B. |n-::j). n B. rn^-ra. C -.yrx\ / Bom. wX^^n and usually S. C xrbn and always H. q NE-ii^ would be a better vocalization. r Bom. ""J~n. J sr^^in is the correct punctuation. / Bom ■yrrs-. u C iT^ibn. x Bom. nrsn ^ -irsn. y Bom. b-5-:n ^ Bom. -:n-i!iD (B. ■'rn-TC). ,7,7 Bom. n-. /^(^ Bom. r-jr:r. ^f Bom. ■'TPns (B. "'^Wp?). ^Vz' Bom. xj'^'bn^ ee Bom. •^ri. (B. "irr"). /f Bom. ns-r:*, ^ "x-ns;. ^ Bom. Nrrr^x*. /^/^ Bom. i<;":bn but is very inconsistent; thus in verse 1 1 X":bT and i^^'>?^f^ and in verse 12 n'^pbns. — 94 — br ^n\-S •'n^ nin ']'>'i iizi ^nt^DT s?^5 nini (13) n^5 ]-Q '^^ninr.nsi cioi^ n^ xnpi nyns nSioi (14) n^'i" n'lb" ^TCEi '^n'''a';n x^bn t\cS^'5 nins n^ici (15) "'"i^nbic ^ri^Dn )-q xb 'i^i'ab nyns n^ poi^ ^n'lnxi (16) ni^nsi ""N^b^ 'inr,^ "iX"^ (Di)p p by Qij^p x:sn ^'abna jicr oy ni?ns "b^bri (17) ■jn^Eri nc5 "i^^DE pin :>3Ty "ipbc xnn: 'jii xni (18) sinsn fiJni I'tnrb "S^cn "i^n^nnd "^spbo ''pnni? pin s^aia srn (19) i«yns bSn ]inniDT ^n^'tn i?b non pioni sinb 'im^b 'i£"iii iTC^nb D^nsiai snnnn 5>5r n^ snw^ii sn5''6n snnin sbDsi (20) ^smsius ■'sn'^'aip "pn^i-)2b i?by "ini? yi^ni? sbi "^jin^y^b 'sbyn (21) a Bom, ^is-ii. <5 Bom. n^rN. The best vocalization — . . . ^ would be a'^PN. <: Bom. "inrjnnxi. ^ Bom. rraibn. e Bom. ^"rs>i (B. n'is!i). / Bom. "nrB'2!i, ^ "lasn'i. g B. nin;^. /^ Bom. -T5<% We should expect niri/ Bom. n? T ■• :'- ; — 95 — 'nn ''x"-':p5 "ipc '']"^'t:3i:" vzz snn ^rbnn ■'nnrn (22) » < - f - . i ^)rim Dinp 'iD-'pr ''fp': ^li-: ]'^'i2'C 7£r sni (23) 'si^at: vr'i2'c: r£r H xr^p's ar'i^ij sr'raT (24) ■^b ■''in'o': n'^b'-, "x-'Trnnb "" r'^nTzsi ''X'^T n^ s^n in niriDi STcbn ni7"iEb qci'' °n:csi (25) ni-nsb 11 n ''-inra ^T^rp x^'':3'ir yncT '-i-iis s-iir y£r xnni: xnnin r'5r (26) xin in s6bn 'I'^is s^:r r£r sr5-3 "iin-i^nn 'spbo-^ srir^DT xr^-'cn xrnin ySiCT (27) "pni Di-p ■jE-'piri xr^pb x^'bmrj yniri I'^ix 'x^^-c rnc x:s5 "^it y5r iny^b 'T«ri' iVi ^nins ai' ''n^b^b^T xr^ri xin (28) xinx b55 xni xrzc f-rx x^tr yniij xn (29) b5 ■'TTinr ■jirT'^rn x:sr "iiTr y^c '''" )i)2ip"'i (30) "xy-ixi xry n^ x:23 '^i:'>ii•'^ c^'^s^'O'^ x^-ixn '^"^xi:3c xinn x:e5 Dip "jii xynx5 "xirno yi=^ni xbn (31) :ppn 11 X ] xinb xin ^^B^pn ^ix rS ir£ in^i <7 Bom. r'^tn". /; Bom. I"'^'^. c Bom. •);r^o. ^ Bom. N-:p?3. e Bom. n". / Bom. ",:":":. ^ Bom. y.*?. /i Bom. "rrp?. z Bom. ■,S"'~w\ /' Bom. ir-iu (B. "i""?^). / .^ !^~~'-?- '« Bom. ^''"t?-'- « Bom. N^r-r?. Bom. -r:s"i. / .-^ -,'^rv. q Bom. -iss"2't- '' Bom. y:"^N so i? and C p^- J See verse 3. / Bom. V?"^- " Bom. 'r.T>'.- -'^ Bom. r'^V'"^ "^"i- }' ^ in- serts n^ before ^y-. z B n^rr. aa Bom. rrx. ^i!^ A ni?-S3 (omitting n;i). a- Bom. *,n':!ipi'; (]^. T^ip"'"), C '{z-^'-'. dd Bom. Nvrr. dv B (text) sr-x "^-'t but margin as above n:-nt s-:r. ff Bom. NV2-r. ^^r Bom. r-ipr, (B. ciT^i')- - 96 - ^rpn •'^.N T'i^T ■^rnn ni""iE5 si^bn ^''•niJT byi (32) Vp'^ny^i '<;i •'ini'ai •'> (nn)p its siaSns r^ "^nri xrnx b^ i^zr^nr ^:"c^i nins ^-r^ar'^ (34) '"■■'bs iVsT snn6 x^-ir 'niny b5 n"^ ''■jiTC'iD^i (35) s:e5 ■'■'ir yfcb' xirisi x'ci'b T:: sniny '^'^n^i (36) N:sf5 x>-ni?i xti^" '^iiric^ jibi D^'^i)2'i xi?nsn '-^iin-^ ^rii^n^ b5 ^ryai ni-nc ■''i^ya xiaBnD lEffii (37) '^mni nn: 'j'^iS "nfir-n Ti^^nyb r^b•^L n^ci^i (38) n^3 iVi (d"i)p irj nxin: ^Ni b5 n^ ^^^ ■'> :s>nTni in£ cici'^b ni-ns nicsT (39) Tni5 wifi^ ^"nbDo n"'b' bS iirp T^Tc^-a b^i ini5 b^ '^^s:?:^ inn rs (40) bS br iri ir^D^i "lin cici^b nine nrxi (41) "nr3 nr.1T n^^i bt-iz "niripri' n- nins "^'^^■iysi (42) ^ Bom. "r^xn, ^ •'insn. It should be ■■:nx=i. d Bom. '|ipri, f ■^n'^n is the correct punctuation. d Bom. .T'nsy'sVi. e Bom. "^.'n;:. / Bom. "i^':^©, ^ Bom. n-:-;ri-. /; Bom. n—:;: (B. t??!'). / Bom. N:^-2r. /& Bom. ■(i^w-wi-;. / C -^iz-i: fn Bom. 'pVsn. ;^ Bom. "(raa-^i. o Bom. xn". / Bom. nir-£-j. ^ Bom. ----. r A t":. j Bom. t'C!, ""I, <^ ')""n-a / £ -I'li-fc-. « Bom. i-:=-r:-, A r=r:-. ;c Bom. nn-i-i. y Bom. ",-7. 2 Bom. "'^=^0 (B. irtsiion) B, C •fiz^ti'^. aa A X:"2"^- bb Bom. "nx. ^^ Bom. ■•!:. dd A I'l"^!. ns5 '"^b^srbii snw ''\^bj xb ny ii:5 i^nr ""^i^-ib-nN qDi-ibi 50) •j^Mi xnn y'^.E ""tiis n5 rTics n^b' "n'^'^b'i-: s<:£r'=' <7 Bom. x:-:-c. /; Bom. ^■"'■x\ c Bom. N:":r. d Bom. T ■ T ■ ; - 1 T T : ■ tT5 '"i. f Bom. "3^!!. / ">£" would be a better vocalization. ?• A n'^"i\ // Bom. "p-^T. i Bom. s-OJit). ,^ Should it be c''iJ? / Bom, l'^^:?'?" (B. •j-^'cii'T). Cf. Winer: De Onkeloso eiusque Paraphrasi Chaldaica: Leipzig 1S20. p. 28. »i Bom. "3?. // Instead of rr^b '{h'j, -f-^-j-si s-::5, C simply has rirs rlss which is perhaps preferable. Bom. ir:sb. p A has b:^ instead of br. r/ Bom. "'■'■"T /■ Bom. srsr. s Bom. ■r-:i^5 t\bi^ nnsi x^ns ids bS by nin s:&di (56) syisn s':s5 q^^pni 'a^nsrib '^'j-inn xniiy p-jinsi "s^^nsix ■jTD sSiny pT^b D^na^b 'ins K^ns '^n'^^'i bSi (57) s^ns? bSn 5c:b5 "ci^n "^^x :|6ii a Bom. i:'i-iJ2X, ^ ii^ffl3i<. d Bom. "i^^:?, ^ iB^s'. c A xri-^ar. ^ ^ D-'-iEx. e Bom. xy-xa. / Bom. nsi-nij^ s: Bom. "i^x "n. /; Bom. rrE=\ / Bom. nis'-. i ^, C s-anb" b:-. / ^ D'i-i:i-3. - ■ T : - T : ^ . - . ?/^ Bom. Tc^:^'^-!. ft Bom. "inarn (B. "^"las'ri). Bom. s<:;l:jix. / Bom. •[•ina ^-n. ^ Bom. 'par.. ^ Bom. n';n-4^b (B. ^xnu-cb). s Bom. '^■■^^!i. / Bom. >irix. // Bom. ;qT^. ADDENDA ET CORRIGENDA, An italic number signifies that the line is counted from the bottom of the page upwards. Page 3, line 5. Cf. Zunz, 1. c. p. 132. 6, „ ult. Pro he lege the. 8, „ 7. Cf. Seligsohn. 1. c. p. 20. 13; „ 17. Pro devise lege device. 13, „ 1'. Pro S. lege S. 1 5; ;; 3- Cf. Diwan of Jehudah Hallevi; No. 65 in Luzzatto's Edition where tcsi (it should be "i";x"') is rimed with n-sh. 15, „ 6. Pro ■'5np lege "''snp. 16, ,, 4. Cf. Friedlander. '^A third system of sym- bols for the Hebrew vowels and accents". (Jewish Quarterly Review VII, 27. p. 567), \6, „ 2. Pro exemples lege examples. 19, „ 2. Pro "'JPwT lege ]'^-3ir^. 21, „ 6. Cf. nT3-ir"^ bibx (Jeremiah 14, 14) and T;-ri7J (Jeremiah 15, 11). 26, „ 2. Pro XTj-cn lege xtj^d5, 32, „ 7. Cf. Aboth de R. Nathan. Ch. 34. 37, „ 12. Cf. Rasi a. 1. 46, „ 9. c^:P lege D^;p. 47, ,,15. ^^6' lege -"nci. 47, „ 2. Cf. Earth in ZDMG. XXX, p. 193. 48, „ 2. Pro 'Sy:j lege -Vr. 50, ,, 4. Cf. Midras Genesis Rabba ch. 42. § 7 in Wilna Edition. Page 50, line 72 „ 5^> „ s „ 52. => I „ 56. „ 11. „ 56, 55 5 ., 58, „ 6 „ 59; „ 6 „ 59. >. 13 „ 60, j> 5 — 100 — Pro 'V^cn lege ■|:ra;fjft. N"rr-2 lege x^fu;^. Pro ^T^ii lege '^n'^bl Pro •pl'^i'i'^i' lege l^siniSy. Pro rv\'rv^ lege n^nn. 6. Pro ''■^^'^N'] lege '^T^'^^V Pro ^I'J-' lege xli'n^. Pro •■^ft'iy-^n lege ■|Wi:."Tin. V. Ibn Ezra, to this verse^ who explains the Jewish Interpretation from '|S2 = to hide and (j^ = to open. 61, „ 10. Pro TvMh lege rtnizin, 62, „ 2. Pro n-:-3b' et n-no lege n^iaV et fi^-iy, 63, ,, ult. V. Earth in ZDMG. XXX. p. 192. 64, „ 7' >, >, „ „ „ P- 19°- ^7} >} 3- V- Seiigsohn. 1. c. p. 26. 71, „ II. Pro pn-.i;''^ lege pn:i^% 73, „ I. Pro qi<^ lege qxi'. 78; „ g. Vide Friedlander. 1. c. p. 568. A few Hireqs which sprung off during printing are not specified as they are easily recognisable. UNIVERSITY OF CALIFORNIA LIBRARY Los Augeles This book is DUE on the last date stamped below. ^ U)-6K ^PR OCT 1 1^^^ OCT H m Porm L9-75ni-7,'61(C1437s4j444 Printed by W. Drugulin, Leipzig (Germany). ^006 377 372 5 BS 709. I4 B2^t UC SOUTHERN REGIONAL LIBRARY FACILITY AA 000 621 573 5 Unive Soi Lii