BOUGHT WITH THE INCOME FROM THE SAGE ENDOWMENT FUND THE GIFT OF ( Hcnrg W. Sage 1891 £.Mffi ~~~T , ,.. S)/ljfi$L Cornell University Library BS1140 .G79 General introduction to the OffJ-.Testamen olin 3 1924 029 277 667 The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924029277667 GENERAL INTRODUCTION TO THE OLD TESTAMENT THE TEXT By the Same Author IN UNIFORM BINDING GENERAL INTRODUCTION TO THE OLD TESTAMENT— THE CANON. Crown 8vo, $1.50 THE HIGHER CRITICISM OF THE PENTATEUCH. Crown 8vo, $1.50 THE UNITY OF THE BOOK OF GENESIS. Crown 8vo, $3.00 GENERAL INTRODUCTION TO THE OLD TESTAMENT THE TEXT BY WILLIAM HENEY GREEN, D.D., LL.D. PROFESSOR OF ORIENTAL AND OLD TESTAMENT LITERATURE LN PRINCETOK THEOLOGICAL SEMINARY NEW YORK CHAELES SCEIBNEE'S SONS 1899 A. 1^6 lyi Copyright, 1899, by CHARLES SCRIBNER'S SONS TROW DIRECTORY PRINTING. AND BOOKBINDING COMPANY NEW YORK PREFACE The subject of the Canon of the Old Testament was discussed in a previous volume. The history of its formation was traced, the determining principle of its collection was considered, and the books contained in it were identified. One who is entering upon the study of these books will further desire to inquire into the character and condition of their text. The first thing to engage attention is the language in which the Old Testament was originally written, in its relation to other forms of human speech, and its special adaptation to be the vehicle of this preliminary revelation. The history of Hebrew as a living language will bring to light diver- sities of usage in different styles of composition and in different periods of time ; and it is a matter of interest and importance to inquire whether any facts ascertained tend to discredit the genuineness of the books of Moses in whole or in part. And the history of its study since Hebrew ceased to be spoken will show what reason there is to believe that it is correctly understood by modern scholars. The changes which have taken place in its written characters naturally suggest the inquiry whether they have in any way proved detrimental to the accuracy with which the text has been preserved, or have injuriously affected its interpretation. And in par- ticular the origin and authority of the vowel points must be investigated in order to ascertain whether they VI PREFACE can be confidently relied upon as a trustworthy guide to the meaning of the sacred text. The question then arises, how has this original text of the Old Testament been transmitted to us, and what guarantee is there of the fidelity and care with which this has been done? This leads to the consideration of manuscripts, their various classes, the oversight ex- tended over them, the rigid rules prescribed for their transcription, their age, their wide dispersion and gen- eral character. The Old Testament was besides early translated into various languages, and these ancient versions still exist. It is important to know something of the character and history of these versions, that some judgment may be formed of the value of the testimony which they render respecting the primitive text. Man- uscripts, Versions, the quotations of Scriptural passages in early writers, and their statements about them, includ- ing that vast body of critical annotations known as the Massora, comprise the apparatus available for tracing the history of the text from age to age during the long interval which has elapsed from the time of the sacred writers to the present day. They constitute likewise the material for what is technically called textual criti- cism, the object of which is to ascertain with the utmost possible precision the exact words of the sacred pen- men, as determined by a minute and painstaking exam- ination of all external authorities. Princeton, N. J., November 1, 1899. TABLE OF CONTENTS THE TEXT OP THE OLD TESTAMENT Its External Form 1 Written mainly in Hebrew, a few sections in Ara- mean, 1. II. The Semitic Family op Languages 2 Early attempts to discover the primitive language, 2 ; wrong methods, 3 ; eight families, three groups, isolating, agglutinative, inflective, 4 ; unity approximated, if not de- monstrated, Indo-European and Semitic families differ as to external or internal flexion, 5 ; constitution of roots, alpha- bets, 6 ; trilateral roots, derivative forms, 7 ; comparative richness in inflections, number of branches, fluctuating or stationary, 8 ; respective character of races, adaptation to the Old Testament revelation, 9 ; pictorial or reflective, 10 ; definite or indefinite, 11 ; fitted for preliminary revelation, different names, 12 ; home of the Semitic family, principal branches, 13 ; their mutual i elation, 14. III. The Hebrew Language 16 Derivation of the name, 16 ; Hebrew, Israel, Jew, Grecian, Hebrew language, 17 ; dialects or provincialisms, 18 ; prose and poetry, 19 ; Arabisms, Aramseisms, pro- phetic style, 20 ; periods, 21 ; deterioration, objection to the antiquity of the Pentateuch answered, 22-24 ; alleged Ara- maic forms and words of late date in the document P, 25, V1U CONTENTS 26 ; answered by Dr. Driver, 27-29 ; copiousness of Hebrew, 30 ; rich in synonyms, 31 ; economy of roots and words, 32 ; lost roots, relation to cognate tongues, 33 ; primitive mean- ings, 34; religious terms, Egyptian words, 35; Sanscrit, Persian, Greek, 36 ; Semitic words in western tongues, 37 ; when Hebrew ceased to be spoken, 38 ; supplanted by Ara- mean, its peculiarities, 39 ; Jewish and Christian Aramean, 40; history of the study of Hebrew among the Jews, 41 ; early Christians, 42 ; revival of letters, the reformers, 43 ; traditional school, comparative, 44 ; idiomatic, 45 ; compre- hensive, 46 ; Dr. Driver shows that ^K in P is no proof of late date, 47-54 ; nor is "rtlil, 55, 56. IV. Hebrew Letters and Vowels 57 Two forms of Hebrew letters, their mutual relation, later rabbis, earlier traditions, 57 ; Origen, Jerome, 58 ; Buxtorf's hypothesis, 59 ; Gesenius, Kopp, accepted view, 60 ; causes of change, when it took place, 61 ; undue importance at- tributed to it, 62 ; antiquity and authority of vowels and accents, Aben Ezra, Elias Levita, 63 ; Cappellus, Buxtorf, 64 ; not coeval with the letters, 65 ; Origen, Jerome, 66 ; Targums, 67 ; Talmud, 68 ; when vowel-signs were intro- duced, 71 ; two systems, 72 ; value of the vowel signs, 73. Hebrew Manuscripts 75 Two classes of manuscripts, 75 ; Synagogue MSS., 76 ; those for private use, 77 ; their age, how determined, Tam and Velsh character, 78 ; subscriptions, Massoretic and non- Massoretic, 79 ; ancient codices in high repute, dates of others, 80. VI. Versions : the septtjagint 82 Versions ancient and immediate, mediate in whole or in part, 82 ; origin of the Septuagint, Aristeas' account, 83 ; Aristobuhis, Josephus, Philo, Justin Martyr, Irenseus, the Talmud, 84; Epiphanius, Aristeas discredited, 85; when and why the law was translated, 86 ; the rest of the Old Tes- CONTENTS IX FAOK tament, diversity of translators, 87 ; freedom in translating, how regarded by Jews, 88 ; and Christians, other Greek ver- sions, Aquila, 89 ; Theodotion, Symmachus, Origen's Hex- apla, 91 ; Quinta, Sexta, Septima, 92 ; aim and method of the Hexapla, 93 ; Octapla, Tetrapla, 94 ; what became of this great work, 95 ; its effect upon the text of the Septuagint, revisions by Pamphilus, Lucian and Hesychius, 96 ; princi- pal manuscripts, printed editions, 97 ; Lagarde's scheme for restoring the text, the Septuagint Daniel, 99 ; other sources cited by Greek fathers, versions made from the Septuagint, Ethiopic, 100; Egyptian, Gothic, Armenian, Georgic, Slavic, 101 ; Arabic, 102. THE TARGUM8 102 Their origin, 102 ; earliest mention, 103 ; ten Targums, Onkelos, a misnomer, confused with Aquila, 104 ; accepted in Babylon, 105 ; general character, printed editions, 106 ; Jonathan (or Joseph) on the Prophets, 107 ; general char- acter, Pseudo-Jonathan and Jerusalem on the Pentateuch, 108 ; other Targums, 110. THE SYRIAC PESHITO Ill Why so called, made by Jews or Christians ? Ill ; general character, relation to the Septuagint, traditions of its origin, 112 ; its probable date, the Syro-Hexaplaric version, 113. THE LATIN VULGATE 113 Early Latin versions, the Itala, revised by Jerome, 114 ; Roman and Gallican psalter, Jerome's own translation, va- riously regarded, 115 ; his defence, progress of the work, 116 ; displaced the older version, 117 ; use of the term Vul- gate, confusion of texts, 118 ; attempts at correction, Cassi- odorus, Alcuin, Nicolaus, Cistercians, Correctoria Biblica, 119 ; University of Paris, Dominicans, Franciscans, theSor- bonne, 120 ; Roger Bacon's testimony, printed editions, 121 ; revision of the text, corrected translations, the Council of Trent, 122 ; diversity of opinions, final action, 123 ; differ- ently interpreted, 124 ; commission appointed to correct the text, checked by the Pope, 125 ; editions prepared by schol- ars and publishers, 126 ; that of Sixtus V., 127; recalled and corrected, 128 ; that of Clement VIII., 129. X CONTENTS PAGE THE SAMARITAN PENTATEUCH 129 First copy obtained by Delia Valle, 129 ; published in polyglots, 130 ; different opinions as to its origin, arguments tracing it back to the schism of Jeroboam, 131, 132 ; shown to be invalid, its probable date, divergence from the Masso- retic text, 133, 134 ; different views as to its critical value, investigated by G-esenius, grammatical emendations, 135 ; glosses, conjectural emendations, conformed to parallel pas- sages, corrections of imagined difficulties, 136, 137 ; Samari- tanisms, Samaritan ideas, 138 ; relation to the Septuagint, 139 ; how explained, 140 ; Samaritan version, Arabic ver- sion, 141. VII. The History of the Text . 142 Four periods, 142 ; graving on stone and metal, books, 143 ; liability to error in transcription, standard copies, ven- eration for sacred books, words perhaps separated in early Hebrew writing, 144 ; some evident errors, others alleged, but differently explained, 145 ; alleged changes in canoniza- tion unfounded, second period, 146 ; Massora, change in the Hebrew character, vowels, 147 ; established text shown by the K'ris, verses, 148 ; sections, 149 ; different views re- specting the text, 150 ; not wilfully altered by the Jews, critical decisions attributed to the scribes, 151 ; extraordi- nary points, the Talmud, third period, 152 ; vowel signs, enlargement of the Massora, 153 ; tables of various readings, Parashas, Haphtaras, 154; Sedarim, fourth period, division into chapters, enumeration of verses, 155 ; printed editions, rabbinical Bibles, critical editions, 156 ; polyglots, Complu- tensian, 157 ; Antwerp, Parisian, London, 158, 159. VIII. The Criticism of the Text 160 \ H igher c riticism, 160 ; Epistles of Phalaris, Wisdom of I Solomon, ad ditions to Est her and Daniel, 161 ; naturalistic prepossessions, textual criticism, 162 ; sources of error in manuscripts, 163 ; intentional alterations, value of manu- scripts, how estimated, all represent the Massoretie text, 164 ; established before the Massorites, early quotations, 165 ; CONTENTS XI PAGE Talmud, Origen, Jerome, 166; versions, their critical, ex- egetical and hermeneutical value, 167 ; caution required in their use for criticism, 168 ; character of the version, 169 ; state of its text, Targums, Aquila, Symmachus, Theodo- tion, Jerome, correspond mostly with the Massoretic text, 170 ; so the Peshito, but Septuagint and Samaritan differ, 171 ; how explained, 172, 173 ; superiority of the Massoretic text, internal evidence as a test of various readings, 174; wrong use of duplicate passages, 175 ; acrostics, 176 ; deca- logues, poetic structure, critical conjecture, 177 ; hypothesis of one primary manuscript, 178, 179 ; general conclusion, , 180, 181. I .Index 183 TREATISES CONSULTED ON THE TEXT In addition to works mentioned elsewhere, the follow- ing have been consulted in regard to the text : J. Buxtobf : Tiberias sive Commentarius Masorethicus, 1620. J. Buxtorf, Fil. - Tractatus de Punctorum Origine, Antiquitate et Authoritate, 1648. B. Walton : Prolegomena in Biblia Sacra Polyglotta, Vol. I. 1657. H. Hodt : Contra Historiam Aristese de LXX Interpretibus Disser- tatio, 1684. H. Hodt : De Bibliorum Textibus Originalibus, Versionibus Graecis et Latina Vulgata, 1705. H. Prideatjx : The Old and New Testament Connected, 1715- 1717. B. Kennicott : Dissertatio super Ratione Textus Hebraici V. T. , 1756. Dissertatio Secunda, 1765. B. Kennicott : The Ten Annual Accounts of the Collation of He- brew Manuscripts of the Old Testament, 1770. O. G. Ttchsen : Tentamen de Variis Codicum Heb. Generibus, 1772. Befreyetes Tentamen, 1774. J. B. de Rossi : Varia3 Lectiones Veteris Testamenti, 1784. W. Gesenitjs : Geschichte der Hebraeischen Sprache und Schrift, 1815. W. Gesenius : De Pentateuchi Samaritani Origine, Indole et Auctoritate, 1815. G. B. "Winer : De Versionis Pentateuchi Samaritanae Indole, 1817. G. Riegler : Kritische Geschichte der Vulgata, 1820. XIV TREATISES CONSULTED ON THE TEXT U. F. Kopp : Bilder und Schriften der Vorzeit, Vol. II., 1821. L. Van Ess : Pragmatisch - kritische Geschichte der Vulgata, 1825. L. Hikzel : De Pentateuchi Versionis Syriacse quam Peschito vo- cant Indole, 1825. A. T. Hoffmann : Hebraeische Schrift, in Ersch und Gruber, All- gemeine Encyklopaedie, 1828. H. Hupfbld : Beleuchtung dunkler Stellen d. A. T. Textgeschichte in Studien und Kritiken, 1830. Zunz : Die Gottesdienstlichen Vortrage der Juden, 1832. Z. Fkankbl : Vorstudien zu der Septuaginta, 1841. Z. Frankbl : Ueber den Einfluss der Palaestinischen Exegese auf die Alexandrinische Hermeneutik, 1851. A. Dillmann : Bibeltextdes A. T., Herzog Encyklopaedie, II., 1854. Second Edition, 1878. E. Bertheau : Hebraeische Sprache in Herzog Encyklopaedie, V. 1856. A. Geiger : Urschrift und Uebersetzungen der Bibel, 1857. G. P. Oehler : Hebraeische Sprache, in Encyklopaedie d. Erzie- hungs-und Unterrichts-Wesen, III., 1862. S. Pinskek : Babylonisch-Hebraeische Punktationssystem, 1863. F. Kaulen : Geschichte der Vulgata, 1868. H. Ewald : Lehrbuch der Hebraeischen Sprache, 8th Edition, 1870. Z. Fkankel : Zu dem Targum der Propheten, 1872. E. Kiehm : Die Sogenannte Grundschrift d. Pentateuchs, in Studien und Kritiken, 1872. F. Field : Originis Hexaplorum quse supersunt, 1875. A. Kuenen : Les Origines du Texte Masorethique de l'Ancien Tes- tament, traduit par A. Carri&re, 1875. Fkitzsche : Alexandrinische Uebersetzung d. A. T., in Herzog- Plitt Encyklopaedie, I., 1878. V. Ryssel : De Elohistse Pent. Sermone, 1878. E. Koenig : De Critics Sacra? Argumento a Linguae Legibus Rep- etito, 1879. E. Koenig : Lehrgebaude d. Hebraeischen Sprache, 1881. Giesebrecht: Der Sprachgebrauch des Hexateuchischen Elohis- ten, in Zeitschrift fur die A. T. Wissenschaft, 1881. S. R. Driver : Some Alleged Linguistic Affinities of the Elohist, in the Journal of Philology, XL, 1882. TREATISES CONSULTED ON THE TEXT XV A. Bbelinbk : Targum Onkelos Herausgegeben una Erlautert, 1884. A. Dillmann: Sprache der Priesterschrift A (P), Num.-Deut., pp. 663-665, 1886. B. Stadb : Geschichte des Volkes Israel, 1887. C. H. Cornill : Einleitung in das Alte Testament, 1891. C. A. Bbiggs : General Introduction to the Study of Holy Scripture, 1899. THE TEXT OP THE OLD TESTAMENT I ITS EXTERNAL FORM The books which compose the Canon of the Old Tes- tament having been ascertained, we proceed to consider their original form, or the language in which they were written, which was that of the covenant people, the peo- ple of Israel, to whom they were primarily addressed. Accordingly, the original language of the Old Testament is the Hebrew, with the exception of a few chapters in Daniel (ii. 4-vii. 28) and Ezra (iv. 8-vi. 19 ; vii. 12-27) and a verse in Jeremiah (x. 11), which were written in Aramean at a time when the language was in a transition state. 1 In order to understand the significance of the language of the Old Testament in its bearing upon its contents and its interpretation, it will be necessary to inquire first into its character as related to the other languages of mankind, and then to give a more particular account of the language itself and of its history, both as a living language and since it ceased to be spoken. ' Two Aramean words occur, Gen. xxxi. 47, in the name given by Laban to a heap of stones, to which Jacob gave an equivalent Hebrew name. II THE SEMITIC FAMILY OF LANGUAGES We are told in Gen. xi. 1 that the whole earth was originally of one language and of the same words. This verse has been to philology what the chimera of the philosopher's stone was to chemical science, a stimulus to earnest and long -continued endeavors, which, though unsuccessful as far as the prime object of their search was concerned, nevertheless opened the way to the most unexpected and brilliant discoveries of a different de- scription. Starting from the statement of revelation that there was a time when but one language was spoken on earth, it was conjectured that this primitive tongue might still be in existence and might be recognized among the multiplicity of dialects which divide man- kind, the proof of its primitive character lying in its being made to appear that all other languages might have been derived from it. This mother tongue of all was the object of a most zealous search, extended in- vestigations were made, lists of words were gathered by enterprising travellers from the remotest parts, the vocabularies of different languages were examined and compared, and relations real or imaginary were pointed out. Ardent enthusiasts claimed, each for his favorite tongue, the honor of having been the primal fount of human speech. But as the investigations proceeded and materials were accumulated, it became more and more apparent that no satisfactory result could thus be reached. In fact, the methods adopted at the outset 2 THE SEMITIC FAMILY OP LANGUAGES 3 were seriously at fault, and the conclusions arrived at were as a matter of course unreliable. 1. It was assumed that the bare similarity of sound between words of like sense, taken at random, was suf- ficient to establish an identity of origin; and this when a slight knowledge of the history of words would show that the agreement was wholly casual. 2. The simple presence of the same or related words in two languages was thought sufficient to establish an organic connection between those languages, without inquiring first whether these words may not have been borrowed by one from the other, and so form no part of its original and native stock. 3. The affinity of languages was rested solely upon etymologies, to the disregard of grammatical structure, which is a far truer test of kinship between tongues. 4. It was conceived that if two tongues were related, one must have been directly derived from the other, when the proper inference might be that both were alike descended from some common source. It needed but the correction of these errors and the adoption of a method based upon sounder principles to bring order into the vast mass of discordant materials upon which students of language had hitherto em- ployed themselves in vain. As soon as they began to penetrate beneath the surface, the most astonishing analogies and most remarkable and pervading similari- ties of structure revealed themselves in widely separated tongues. Languages arranged themselves spontane- ously, as it were, into families and groups. And now philology seems to be working its way back to the point from which it set out, viz., the original unity of lan- guage, though by a very different and unexpected route. The immense multitude and variety of human tongues are in the judgment of scholars 1 reducible to eight 4 GENERAL INTRODUCTION families, the several members of each of which are more or less closely related. These families may again be classified in respect to certain marked peculiarities of structure into three grand groups, which have been re- spectively denominated : 1. The isolating languages or languages of undevel- oped roots, in which there is no inflection, no parts of speech properly so called, no modification in the forms of words to express relations of person, number, gen- der, tense, mood, case, etc., and no derivation of words one from the other, but the ultimate roots are thrown together 'loosely like stones in a heap, hard, angular, void of all affinity or coherence. 2. The agglutinative languages, which are one step removed from the unyielding stiffness and rigidity of those just spoken of. They possess all the various parts of speech, which are subject to regular modifica- tions of form to express the different relations of ideas. The various roots are conglomerated with certain syl- lables in the formation of words and their inflection. These formative syllables are, however, only cemented to the root, not organically attached to it, but preserv- ing their independent character like the separate courses of hewn stone in a building. 3. The inflected languages are the most highly de- veloped and the most perfect of all the forms of human speech. The root and the formative or inflective sylla- bles are in them so intimately imited, that to the popular consciousness they have become one and inseparable. They resemble not a building constructed by adding one layer to another till the whole is finished, but a growth, whose branches are not merely cemented to the trunk, but indissolubly joined to its very substance. 1 See Max Miiller's Lectures on the Science of Language, and Whit- ney's Language and the Study of Language. THE SEMITIC FAMILY OP LANGUAGES 5 It is scarcely to be expected that it will ever be possi- ble by any scientific process to demonstrate the original unity of these families and groups of tongues. In as- scending from the existing forms of languages and tracing them backward toward their source, the last results that can be reached by the most subtle and searching analysis are too far removed from the beginnings of human speech to warrant the hope that the long accretions of ages can be eliminated so as to detect those primal and original germs from which all languages have sprung. This, however, philology can do. It can enable us to under- stand how the different tongues and dialects spoken on the earth might be as various and as widely sundered as they now are, though all descended from one aboriginal stock, by pointing out influences that have been at work potent enough to bring about all the diversity which now exists. The inflective group, which is the highest type of language, embraces two families, spoken alike by the white race and by the nations most influential in his- tory and foremost in civilization and enlightenment, and as a consequence the best known and most carefully studied, viz., the Indo-European and the Semitic. The language of the New Testament belongs to the former, that of the Old Testament to the latter. While these two leading families stand thus together, possessing a common type which distinguishes them from all other families of tongues and places them at the summit of all forms of human speech as respects their structure, they differ remarkably from one another in various particulars. The first fundamental diversity is that Indo-Euro- pean tongues form and inflect their words by means of external additions to the root. Semitic tongues form their words and to some extent inflect them by 6 GENERAL INTRODUCTION vowels inserted in the body of the root or directly attached to it, the only exception being that of pro- nominal prefixes and appendages. 1 With this is allied a peculiar constitution of the roots of words in these two families respectively. The Indo-European root is a syllable, the ultimate unit of articulate speech ; this may consist of a vowel alone, or of a vowel with one or more associated consonants, and suffers no change in any of its combinations except as the laws of euphony may interfere to modify it, the vowel of the root being as essentia] and inalienable a part of it as its consonants. The Semi tic root, on the contrary, consists solely of con- sonants, which constitute the skeleton or framework, and by the addition of vowels is converted into a word in actual, living use. The consonants of a word, there- fore, determine its radical signification, while the vowels are subsidiary, suggesting its modifications and sub- ordinate shades of meaning. The consonants are fixed and unchanging, the vowels fluctuating and unstable, altering with every derivative and every grammatical form. Hence a remarkable diversity in the alphabets of these two families of tongues. The Semitic alphabet consists of consonants exclusively ; the vowels form no part of the essential structure of the word. And the 1 In English the root "love" has as its derivatives lover, loving, be- loved, loveable, lovely, loveliness, unloveliness. From the Latin root " am " of the same sense come the verb amo, the noun amor, and the further derivatives amator, amatorius, amasius, amabilis, amabilitas, amita, amicus, amicitia, inimicus, inimicitia. In Hebrew, on the other hand, from the root "gdhl" are derived the words 5^3 gadhal to be great or large, 5^3 giddel to make large, 5 ,l; l5n higdil to make great, b*|3 gadhel growing great, 3"n3 gadhol great, 3^3 giddel too great, 3"13 godhel greatness, <15"IjI g'dhulla magnificence, '"''IS gadhil tassel (threads combined into a body of some size), s'li.'H migdal tower (some- thing great or strong). THE SEMITIC FAMILY OP LANGUAGES 7 modifications of its meaning are sufficiently suggested by the connection in which it stands. But Indo-European alphabets include vowels as well as consonants, since both are equally essential in determining the meaning. The number of letters in an Indo-European root might vary from one to six, the only restriction being that they must be compacted into a single syllable. But the fundamental peculiarity of Semitic inflection natu- rally required that its roots should consist of a uniform number of consonants. The production of a regular system of significant forms from its several roots by the simple addition of vowels could only be effected if its roots were themselves of one invariable pattern. Bi- literal roots would be too short to admit of the needed variety to serve for the various classes of words and forms. Triliteral roots were the briefest that would answer the purpose, and hence Semitic roots are as a rule triliteral. Even those which appear to have been originally biliteral are enlarged by an added letter to bring them up to the customary length. And those which exceed the normal number, the quadriliterals and quinqueliterals, are later formations. This constitution of the Semitic root made it very easy for the verb to have a regular and pervading system of derivative forms. Thus in Hebrew from the root Ml are formed the simple active, Icatal, to kill ; passive, niktal, to be killed : the intensive active, kittel, to mas- sacre ; passive, Jcuttal, to be massacred : the causative active, hiktll, to cause to kill; passive, lioktal, to be caused to kill : the reflexive active, hithkattel, to kill one's self. The Indo-European has its causatives, fre- quentatives, inceptives, desideratives, etc., formed by significant syllables appended to the root, but these have no such prevalence and simplicity of structure as the Semitic system. 8 GENERAL INTRODUCTION Another difference resulting from the fundamental diversity in the mode of word-formation and inflection is the comparative richness of the Indo-European in forms and inflections and the comparative poverty of the Semitic. The possibility of multiplying forms by external flexion is unlimited. For every shade of mean- ing that a people apprehend and desire to express, some word or particle can be found to convey it, and by attaching this to the root or word that is to be modified, a new inflexion or a new significant form will be created. But the possible changes of vowels within the compass of three consonants necessarily limits the number of significant forms that can be thus produced. Hence, too, the Indo-European family includes a considerable num- ber of principal branches, the Celtic, Germanic, Italic, Slavonic, Lithuanic, Greek, Iranian, and Indian, each of which has its various subdivisions. The Semitic has but four main branches with their subordinate varieties, the Assyrian, Aramean, Hebrew, and Arabic, and these exhibit no greater differences than the languages be- longing to any one of the subdivisions of the Indo- European family, e. g., French, Italian, and Spanish, or English, German, and Danish. It may be added that compound words are almost unknown in Semitic, while they abound in Indo-European, and form an important part of the riches of these tongues. A further contrast connected with that radical diver- sity which has thus far engaged attention, is the sta- tionary character of the Semitic toDgues and the mobil- ity of the Indo-European. The readiness with which new forms are produced as old ones are dropped or worn out, keeps the latter in a state of constant flux. There is incessant change for the better or the worse, as the case may be. With the Semitic tongues it is different. Their structure does not admit of this facility for in- THE SEMITIC FAMILY OF LANGUAGES 9 definite change. There is little of that attrition of con- sonants, that wearing out of the beginning or end of words by rapid or defective or repeated utterance, which is one of its most fruitful causes. The conso- nantal base of the word is its ultimate root, and this cannot be abridged without violating the constant law of triliterals, and cannot be modified materially without destroying its identity and rendering its recog- nition difficult or impossible. This diverse character of the two families of languages both grew out of and reacted upon the characters of the races by which they were respectively spoken. The Semite abides substan- tially unaltered from age to age. Travellers find the same dress, the same manners, habits, and modes of life in Palestine at the present day that existed in the days of Abraham. With Europeans there is constant change ; fashions in dress vary from season to season, the con- veniences and comforts of life, customs, and laws un- dergo perpetual alteration. The Semite abides on the same spot on which he was born. With the exception of the trading colonies of the Phoenicians and the fanat- ical conquests of the Saracens, the Semite populations have remained fixed in the same territory from the dawn of history, and that one of very limited extent. The restless, moving Indo-European population has spread itself across both the continents of Europe and Asia, occupying a broad belt of territory from Great Britain to the peninsula of Hindostan, and is now fill- ing the new continent of America and settling on the islands of the ocean, ever the same energetic, progres- sive race. May not this suggest a reason why the Old Testament was given to a Semitic people, a steadfast adherence to what was delivered being the chief quality demanded in that traditionary dispensation ? But when the period was come for aggression, for breaking over 10 GENERAL INTRODUCTION the old boundaries, giving up ancient usages, and carry- ing the gospel to the ends of the earth, the work was given in charge to Europeans. The distinction between the Semitic and the Indo- European tongues thus far considered has relation to the outward forms of words. A second characteristic difference respects their meaning or significant con- tents. It is that the Semitic languages are, so to speak, more pictorial, and the Indo-European more reflective. Words expressive of abstract ideas and of spiritual con- ceptions are in all languages, for the most part, based upon roots primarily relating to external objects or im- pressions made upon the senses. In the Indo-Euro- pean languages, however, the origin of these words no longer remains in the popular consciousness. The metaphor which first suggested their employment is lost sight of as they are currently used, and is only discov- erable by a careful investigation and analysis, or his- torically tracing them back to their origin. When a man is spoken of as a sincere friend, no one thinks of the figure suggested by the etymology sine cerd — pure honey freed from wax. And tribulation no longer car- ries the thought back to the tribulum, or threshing in- strument, though we still use, semi-consciously, at least, the kindred image of harrowing up the feelings. The term agony is used without recalling the desperate struggles of the palsestra, from which it was first bor- rowed ; and insult without thinking of the victor leap- ing on the body of his prostrate foe. Inculcate does not bring before the mind the image of treading in the grain, so pressing the soil upon the seeds that they shall grow, nor implicate the folds of a garment in which something is enclosed. But in Semitic words, the original metaphor still remains as palpable in many cases as it was to those who first made use of them. THE SEMITIC FAMILY OF LANGUAGES 11 The terms are not only proved to be figurative by. ety- mological research, but continue so in the conscious- ness of all who speak the language ; the primary sensu- ous signification and the secondary intellectual or spir- itual signification co-exist side by side. Thus anger is variously denominated, e.g., qs from excited breathing, nan heat, linn burning, -py boiling, ran breaking asun- der with violence, D5H roaring. Patient is denoted by DiSS -pis slow breathing ; impatient D^SS "lip short or quick breathing. Discouragement or despair is 0013 or JffiO melting of the heart, reins, or knees. Desire is S532 thirst or 5|D3 growing pale. To pardon is "B3 to cover or nD3 to hide. To be proud is csn 81B3 to lift up the head, oil to be of lofty stature, or "I33PPI to vaunt one's strength. Truth is nasi that which is firm, or ]3 stable ; beautiful "PSiB that which shines ; right is ~I12£ straight ; wrong is niy, biy or b'nbnB curved or crooked, or H3S3 of ill odor. To form 32y or to create sna is to cut out or carve. To decide anything is "ITS to cut it off. The essence or substance of anything is D3?y its bone. Connected with this quality of Semitic speech is a lack of that precision and definiteness or exactness of expression which the Indo-European labors to attain. The thought is simply suggested in outline or in sub- stance, and it is left to the intelligence or imagination of the hearer to fill it up and complete it. Vividness and force are aimed at rather than minuteness of de- tail. The Indo-European seeks to express his meaning with exactness, and leaves less to be supplied by the hearer. Thus while the Hebrew has but two tenses and but scanty provision for the different modes of action, the Greek has nine tenses and an ample variety of moods. The former utters his sentences in succes- sion, but without concatenation or indicating their rela- 12 GENERAL INTRODUCTION tion to one another. For the most part they are simply strung together by the copulative "and." The more logical Indo-European cannot satisfactorily express his style of thinking without calling to his aid a number of particles which shall distinctly exhibit the relation of clause to clause, thus enabling him to build up his long and complicated periods, each part of which is held in its due relation and proper subordination by the appro- priate conjunction. Here again we can see how a Semitic language was best fitted for the preliminary revelation of the Old Testament, which was so largely figurative and symbolic in its character, which dealt in outlines and in shadows. But in the exactness necessary for the final stage of divine revelation, in which the truth was no longer to be set forth in symbols or hid under a veil, but in which the doctrines of religion were to be exhibited with great plainness of speech and in their final form and set in their logical relations and based on convincing argu- ments, an Indo-European language was employed, and the mind of a Paul, who was not only trained in Jewish lore but thoroughly educated likewise in the learning and philosophy of the Greeks. The family of languages kindred to the Hebrew has received several different names. Thus Jerome called them the oriental languages ; but in our more extended knowledge of the East this designation has ceased to be distinctive. The name Syro-Arabian has been pro- posed, formed after the analogy of Indo-European by combining the extreme limits of the territory occupied by this family. The name most commonly applied to it, however, is Shemitish, or in its Greek form Semitic, derived from the name of the patriarch Shem, the son of Noah. Its use is justified by the fact that according to Genesis x. the principal members of this family were THE SEMITIC FAMILY OP LANGUAGES 13 descended from Shem. The Elamites, however, though sprung from Shem, spoke an Indo-European language ; and the Canaanites and Phoenicians spoke a Semitic tongue, though descended from Ham. The proper home of the Semitic family embraces all the territory between the Red Sea and the Mediterra- nean on the west and the Persian Gulf and the Tigris on the east, and stretching from the Taurus Mountains on the north to the extreme south of the Arabian Peninsula. This includes AraMa, Palestine, Syria, Mesopotamia, and Babylonia. Here these languages have existed as far back as they can be traced. They exist there still. From this possession they have never been driven, though they have at times overrun these limits, especially as carried by Phoenician commerce and Mohammedan in- vasion. These were the languages of the great empires that were anciently founded in this region, or at least of their capitals, the mightiest and most renowned of the ancient world, Babylon and Nineveh, also of the great commercial metropolis of Tyre together with her colonies. Besides being the languages of civilization and of trade they have been consecrated as the lan- guages of religion, and have thus wielded a still more important and extensive influence upon human history and the destinies of mankind. The religion of Mo- hammed came forth from Arabia ; the revelations of the Old Testament were made to the people of Israel ; the divine Founder of Christianity lived and taught in Palestine. The four principal branches of the Semitic family are the Assyrian, the Aramean, the Hebrew, and the Arabic. The Assyrian is found on the cuneiform monu- ments. The Aramean includes the Jewish Aramean, the Christian Aramean or Syriac, the dialect of the Samaritans, and some other minor varieties. With the 14 GENERAL INTRODUCTION' Hebrew are closely allied the Phoenician and the lan- guage of the Canaanites. The Arabic is closely allied with the Ethiopic, the language of ancient Abyssinia. Hebrew occupies an intermediate position among the main divisions of the family, both geographically and philologically ; the Aramean lay to the north and east of Palestine, and Arabia is on the south. The Arabic is the softest, most flexible, and most copious; the Hebrew next ; the Aramean least. The historical order in which they appeared and flourished is Babylonish or Assyrian, Hebrew, Aramean, Arabic. The Assyrian was the language of an extensive literature inscribed chiefly on clay tablets, which have only lately been ex- humed and deciphered. Hebrew literature began with Moses ; when this language finally ceased to be spoken, it was absorbed into the Aramean, which, in its turn, became the language of a flourishing literature, both Jewish and Christian, then passed into decline and was itself absorbed into the Arabic, which is now the sole living representative of the Semitic family, with the ex- ception of some trifling remnants of other branches spoken by inconsiderable communities. Though the Arabic is the most recent of the Semitic literatures, it must not be inferred from this that the language is the most recent in its origin and structure of the Semitic tongues. An analysis of Arabic forms establishes the surprising fact that instead of being the latest and most modern development, it is really in some re- spects the most primitive, and preserves the original forms and inflections with less change than any of its sisters. For the purpose of comparison with the Hebrew the Arabic has the advantage of the greatest copiousness and of being still a living language. The Aramean seems to be the most closely related to the Hebrew. The Ethiopic preserves quite a number of THE SEMITIC FAMILY OF LANGUAGES 15 analogies which have been lost in other tongues. And much may be hoped from the Assyrian, when it has been more fully investigated, particularly in aiding to determine the primitive meaning of obscure Hebrew roots. Ill THE HEBEEW LANGUAGE The Hebrew language received this name because it was spoken by the Hebrew people. Two different ex- planations have been given of the term as applied to them. One that it is derived from "Qy in its appellative sense beyond, as in the phrase "in^n -qy beyond the river ; accordingly "nay Hebreiu would denote one be- longing to the region beyond the Euphrates. The word first occurs in Gen. xiv. 13, in application to Abram who had recently removed to Canaan from Ha- ran ; it is there rendered ireparr)3S5 some 130 times is certainly remarkable. It will be worth while, however, to ex- amine the instances in detail. It is clear in the first place that though ultimately "CK superseded "03X, both forms were in use together in the earliest periods of the language ; the examples from J, Judg., Sam., are suf- ficient to establish this. It was competent, therefore, for any writer, whatever his date, to use i3X, if for some reason it seemed to him to be preferable to i3DK. Now two differences are noticeable between the two forms. One is slightly fuller and more emphatic than the other ; and they are not rhythmically equivalent. Hence, though doubtless many cases would occur allowing equally of either form, we should not expect the usage of the best writers, where it fluctuates, to be determined entirely by accident or caprice, but rather by a delicate, instinctive appreciation of the form best adapted to the structure and rhythm of particular sentences. And indeed this is exactty what takes place. Sometimes the writer's choice is evidently determined by the posi- tion which the word occupies in the sentence, sometimes by a feeling that the sense which he desires to convey will be better brought out by one particular form ; and there are besides individual phrases of frequent occur- rence, in which one form is all but uniformly preferred to the other. How heavy, for example, in 2 Sam. xv. 20, ^b"in i3X "VEJX b»' "jbin iS81 would be the repetition of "03K : in the differently constructed sentence Judg. xvii. 9, aSEX 1S383 Tftb ibin 132JO, on the contrary, it is perfectly suitable. iS8 suits the rapid movement of Deut. xxxii. 39, Sin ij« "OS ^ nns> 1»1 ; ^38 the state- lier style of Isa. xliii. 25, "p»»B nrrra Sin r53X 1D3N "Wttb (contrast similarly 2 Sam. xix. 21 and Josh. vii. 20 j Isa. xlv. 7b and xliv. 24b ; xlvi. 4= and xlix. 25b ; and THE HEBREW LANGUAGE 49 even xliii. 10b and 11). Gen. xxvii. 19, *p133 lift? 1338, Jacob not unnaturally lays a slight emphasis on his as- sumed personality; ver. 32, Toy "p133 *]33 138, Esau takes his own for granted. Analogously, when the sub- ject of a verb is to be expressed separately, "08 empha- sizes it slightly, 1338 is used where a rather stronger emphasis is desired ; contrast, e.g., 1 Sam. xxvi. 6b with Gen. xlvi. 4 ; 2 Sam. vii. 8 with xii. 7b. If, further, in- stances of 13381 be compared with those of 138I, it will be seen that often the latter could not be substituted for the former without a distinct loss in meaning ; idsi implies a mere contrast, while in 133X1 a real stress lies upon the pronoun. "What has been said above will ex- plain the use of 138 in passages such as Gen. xiv. 23, Isa. xxxvii. 24 f., xxxviii. 10, and the much greater fre- quency of 138 and 1381 above 133 8 and 13381 in the Psalms. Lastly, cases in which the shorter form as a rule is decidedly preferred, are when the pronoun is appended to the verb for the sake of emphasis, whether with or without 03 (Judg. i. 3, lb"n33 ins 138 D3 ilrobm ; 2 Sam. xvii. 15, 13 X insy nKT3l) ; and when it follows the participle (1 Sam. iii. 13, irvo fix 138 T3B11B 13 ; Judg. xv. 3) ; on the contrary, before the participle (especially if nsn precede) 133s is more common (1 Sam. iii. 11, "I3T !"I1B11> 1338 mtl). Further examples will be referred to below. So we find almost universally ipK ifi, i?8 ^5718 and 138 "Vn8>1 (2 Sam. ii. 20 1338 "03811) but ^B5 1338 and 1338 ij&tt?, i?38 "»?. We are now in a position to consider the use of 138 in P. In the great majority of cases it forms part of the formula njm 138 (sometimes with additions), and about one-half occur in the section Lev. xvii.-xxvi. Ex- amples of mm 138 and cognate expressions (mm 138 D3inb8 or D31»-[pT3, mm 138 13 13>T\ etc.) are found A 50 GENERAL INTRODUCTION in Ex. vi. 2, 6-8, 29, vii. 5, xii. 12 ; xvi. 12, etc. (similarly "H© btf ijj*, Gen. xvii. 1, xxxv. 11) ; and beyond the limits of P, Deut. xxix. 5 ; Judg. vi. 10 ; 1 Kings xx. 13, 28 ; Isa. xl. ff. ; Joel iv. 17 ; Jer. xxiv. 7, xxxii. 27, and often in Ezekiel. It is clear then that mm i3X was a standing type of expression, which, though used most frequently by the Elohist and Ezekiel, was undoubtedly in use long before the exile. The example Deut. xxix. 5 is in this respect peculiarly instructive ; for, while the two " margins " of the Deuteronomic legislation agree elsewhere with the body of the work (ch. xii.-xxvi.) in exclusively preferring i3)3S?, in this single formula "OS* is employed. Had the author of Deut. xxix. been the first to use the expression, he would surely have em- ployed "038 ; his use of 13 s? shows that it had been in use before, and was merely borrowed by him ; and when the type had once been formed with 138*, it is nat- ural that it should be adhered to uniformly. Indeed it may be traced back much earlier than to D ; to Gen. xv. 7, xxviii. 13 J (the latter passage vindicating the originality of ts in the former) ; Ex. xiv. 4, 18 E, even if vii. 17, x. 2, xv. 26 be disallowed. In face of these facts, whatever weight may be attached to the corre- spondency of P and Ezekiel, or to the frequency with which P employs the expression, the mere circumstance that he uses 13s and not 133s is no decisive indication of the age at which he lived. There is, however, another formula in which iD3K is employed, occurring in both recensions of the Deca- logue (Ex. xx. 2, 5 = Deut. v. 6, 9), and occasionally be- sides (Hos. xii. 10, xiii. 4 ; Isa. xliii. 11, xliv. 24, li. 15 ; Ps. lxxxi. 11 ; cf. Gen. xxvi. 24 J omiS ifibx iD3s5 T3B, xxxi. 13 E ; xlvi. 3 E ; Ex. iii. 6 E ; Isa. xlvi. 9 ; Ps. xlvi. 11). But this is much rarer than 13s ; and looking at JE alone, it is noticeable that, while neither THE HEBREW LANGUAGE 51 formula is there very frequent, ">3SC occurs not less than ''338. In fact it is tolerably clear that whoever first formulated the expressions in question (and both are met with early) was determined in his preference for 135? or "028 by considerations of rhythm ; when the predicate is mm alone, i33£ never occurs, the lighter 13s being uni- formly employed ; when the predicate is weightier (as amnsc Tibs, or b« rm ban), then 133s is found more ap- propriate to balance it ; mm 13s is also retained when some further defining clause is appended (as DDTibx DDiaipa or o"nsB •psa nsnx rkyafi). So also Lev. xix. 2, xx. 26, xxi. 8 (where the predicate tJYip precedes). With xi. 44, 45b 13s ©lip 13 compare Ex. xxii. 26 J 13 ^Si? fWI. mm i33S is confined (except Isa. xliii., xliv.) to the instances .where "pnbtf (with the light suffix) follows (Ex. xx. ; Hos. xii. 10, xiii. 4 ; Ps. lxxxi. 11, the alter- nating types being CDTlbs mm ^Stf and T>nbss mm ^33!* ; for *pnb» mm ">3tf will scarcely be met with beyond Isa. xli. 13, xliii. 3, xlviii. 17). Let us examine some of the other examples of 1355 in P, and consider whether any principle can be estab- lished for their use, irrespective of date. 1. Tpn "OK Tl»« Gen. ix. 12 ; Lev. xiv. 34, xxiii. 10, xxv. 2 ; Num. xiii. 2, xv. 2 ; Deut. xxxii. 49, 52 ; and similarly after the relative and before a participle, Ex. xxv. 9 ; Lev. xviii. 3, xx. 22, and Num. xv. 18 (i:k -iflJK not) D3ns SPiVa) ; Lev. xviii. 24, xx. 23 ; Num. v. 3, xxxv. 34. D in similar sentences uses always 133s ; if D's preference for "02&5 does not constitute him one of the earliest writers of the 0. T., 1 we should argue with caution from P's preference for 13 X that he is one of the latest. Both forms were in use early ; J uses 13s, for instance, Gen. xxvii. 8 (xxiv. 3, 37, 42, xxviii. 20 "ojk) though it must be admitted that the early historians gen- 1 l.e.i in the opinion of the critics. W. H. G. 52 GENERAL INTRODUCTION erally prefer iD38 in this connection. It will be noticed, however, that the instances in P are mostly cases of stand- ing expressions ; and it is quite possible that ]ri3 138 "1IB8 may have been as regularly in use (133 8 except in Deut. only Josh. i. 2) as m»3> 138 TID8, which is met with some eight times in writers of all ages from Gen. xviii. 17 J to Mai. iii. 17, 21, while niBi> i:>38 "il»8 occurs only twice, in one verse Jer. xxxiii. 9, where the rhythm strongly demands it. How tenaciously phrases, when once formed, were adhered to by the Hebrew writers, may be illustrated from the phrase 8i3E "OH, which from Ex. x. 4 J occurs nearly twenty times (especially in Jeremiah with njn following) ; only once do we find run 8-1373 "DDK ren Jer. vi. 19. 2. Gen. xvii. 4, H3H 138 ; and n3H "OKI (Ex. xxxi. 6 ; Num. iii. 12, xviii. 6, 8 ; Jer. i. 18), or i33H ""SSI (Gen. vi. 17, ix. 9 ; Ex. xiv. 17 E ; Jer. xxvi. 14, xl. 10, but not Ezekiel). This is the usual form when the pronoun precedes H3n ; when it follows it, we have on the other hand "CCS n3!"l, Gen. xxiv. 13, 43, xxv. 32, xxviii. 15, xlviii. 21 ; Ex. iii. 13, iv. 23, vii. 17, etc. ; the same idiom in other books, as Josh, xxiii. 14 ; Judg. vi. 37 (to denote a provisional occurrence), vii. 17, and often ; Mai. iii. 23 ; 1 Chron. xvii. 1 ; whereas "OX ran is very rare even in the books which, as a rule, favor 138 most decidedly, 2 Kings x. 9 ; Jer. xlxii. 27 ; Ez. xxxvii. 5, 12, 19, 21 ; 2 Chrori. ii. 3 are exceptional. The two forms H3n 138 and 1338 nsn are obviously, however, not equiv- alent. And inasmuch as in the passages from P the sense requires that the pronoun should occupy the first place, there was no option but to employ 138 ; for the collocation ran 1338 (occurring, I believe, once only, in the singularly worded clause Jer. vii. lib) does not ap- pear to have been generally in use. 3. Lev. xx. 5 1 8 VWl, xxvi. 32, after the verb. 138 THE HEBREW LANGUAGE 53 here is in accordance with the usual custom, 2 Sam. xii. 28, xvii. 15 ; Judg. viii. 23. 4. Lev. xxvi. 16, ni»»8 138 5|» ; and following the verb, xxvi. 24, 138 S]8 irobiT, (also with Eft), 28, inicn 138 5|8 DDJlS. So Judg. i. 3; 2 Sam. xviii. 2, 22, 138 D5 SS-PffillS ; Hos. iv. 6, 138 m fM nD»8, and else- where. In neither of these cases (as said above) would 1338 be expected ; an early writer would avoid it as too heavy, not less than a later one (exceptions as Gen. xxx. 3, 30b are very rare) ; and even D, abandoning his usual iD38, writes as it were instinctively (xii. 30) DJl p !"il»»8i 138. In addition to the passages cited, there are some eight or ten places besides, where the occurrence of 13s is not readily reducible to principle (Num. xx. 19, ijpiaT 138 follows the type of VPSl i38 Gen. xxxiv. 30, xxxvii. 10, xli. 11 ; 1 Sam. xiv. 40, etc., which is a good deal com- moner than that of vrcm 1338 Judg. xi. 37, cf. vii. 18 ; 2 Sam. iii. 28 ; Josh. xxiv. 15). The result of our investi- gation appears to be that, while the predominance of 138 in P is marked and undeniable, it is not so certain that this predominance is to be attributed to the late- ness of date. Though there are naturally many oc- casions on which either form might be employed with- out serious detriment to sense or rhythm, the best writers do not use them entirely without discrimina- tion ; at one time 1338 is preferred, at another 138. The majority of instances in P consist of the formula mni 138, which was certainly in use long before the date at which Giesebrecht would place the Elohist ; the formula being fixed, the frequency of its occurrence renders it characteristic of P — or of P and the Holiness Law Lev. xvii.-xxvi. — but constitutes no criterion of the period at which P was composed. And several of the other examples occur in phrases in which analogy 54 GENERAL INTRODUCTION would lead us to suppose that even an early writer would prefer 13s to "038. iDDS in Deut. occurs almost entirely in two or three fixed formulae, as "JIJZK "WK, asnx msa 133s. I do not deny that the preponder- ance of h 3S in P has some significance, but it is far less than the mere statement of the numerical ratio 130 : 1 might be imagined to imply. It is unnecessary to add anything to Dr. Driver's lucid and satisfactory discussion of a very complicated question, in which the superficial method of merely counting the words instead of taking account of the reasons of their occurrence very naturally leads to wholly deceptive results. It may be worthy of con- sideration whether in addition the general character and design of the writing, in which they are found, may not in this instance have exerted an appreciable influ- ence upon its selection of words. The quiet, unim- passioned and purely objective tenor of the ritual law does not, like the fervid and energetic style of Deuter- onomy embodying the farewell charge of the great leg- islator, require the use of the emphatic form of the pronoun. For the same reason, while Deuteronomy abounds in emphatic forms of the verb, these do not occur in the ritual law. Giesebreeht maintains, p. 235 f. , that "pbin to beget, characterizes P in distinction from J and all the older historical literature ; that this form is only to be found in the later books, while ibi originally denoted both to beget and to bring forth, but in the course of time grad- ually lost the former signification, and came to be used no longer of the father, but only of the mother^ And Professor Konig in the Theolog. Literaturblatt, October 8, 1897, and December 6, 1895, cf. also his "Einleitung ins. A. T.," p. 229, insists that P's use of -pbin and i;» is an irrefragable proof of the late date of this docu- THE HEBREW LANGUAGE 55 ment. Its use of i5X has been vindicated above. The facts in regard to lb" 1 and h^bin are these. The use of lb 1 in application to the father is limited to a very few passages scattered sparsely over the Old Testament. It is found only in genealogies in Genesis not relating to the chosen race, viz., iv. 18, x. 8, 13, 15, 24, 1 26 (copied 1 Chron. i. 10-20), xxii. 23, xxv. 3, and in Deut. xxxii. 18 ; Ps. ii. 7 ; Prov. xvii. 21, xxiii. 22, 24 ; Job xxxviii. 29 (but in ver. 28 Tbin) ; Isa. xlix. 21 ; Zech. xiii. 3 (twice of both parents) ; Dan. xi. 6. "With these exceptions, lb"* is uniformly and with great frequency used of the mother ; and Tbin is uniformly used of the father both in the generations of Genesis relating to the chosen race, and everywhere else in the Old Testament when paternity is spoken of. So that Jer. xxx. 6 using "ibi in its ordinary sense says, as of something incred- ible,. Ask and see whether a man doth bring forth (A. V., travail with child). That this does not mark a limit in time, after which "lb" 1 could not be used of the father, is plain from the examples above given from Zechariah and Daniel, not to speak of the passage in Isaiah, which the critics would transfer to the latter part of the Babylonish exile. The only conclusion that can be drawn from the facts as they lie before us, is that of Dillmann that l^b-in is " the more precise form of expression," or as Driver phrases it, is " adopted for greater distinctness " in preference to a term which might be "used indifferently of either parent." The more exact term is used in tracing the line of descent of the chosen people ; the less explicit sufficed for other populations and for the occasional uses of poetiy. 1 This verse is no exception to the statement above made ; Arpach- shad, Shelah, Eber, are here introduced as ancestors of the various Arab tribes. In xi 12, 14 they appear as ancestors of the chosen race, and T'bin is used. 56 GENERAL INTRODUCTION Both "ib^ and Tbin occur in relation to the father in Deuteronomy, Isaiah, Job, and Chronicles. They are accordingly in use at one and the same time, and for all that appears may have been both in use in the time of Moses. So that the occurrence of one of these words in one class of the genealogies in Genesis, and of the other in another class, does not justify the conclusion that these must have been separated by centuries. rv HEBREW LETTEES AND VOWELS The letters used in modern Hebrew Bibles are the same that are found in all existing manuscripts with scarcely any variation of form. But Jewish coins from the time of Simon Maccabseus, 140 B.C., to that of Bar Cochab, a.d. 135, have a different character, which is evidently related to the Phoenician and closely re- sembles that in use among the Samaritans. This same character somewhat modified occurs on the inscription discovered in 1880 in the tunnel leading to the pool of Siloam, which is supposed to belong to the reign of king Hezekiah, 2 Kin. xx. 20 ; also upon some ancient Hebrew seals and gems. The question hence arises as to the relation of these two forms of the Hebrew letters, that in ordinary use at present, which is called the square letter, and that which is found upon the coins and in- scriptions. The later Jewish rabbis held the square letter in great veneration, and regarded it as the primi- tive character in which the Scriptures were originally written. In the earlier traditions, however, the coin letter is called Hebrew writing, and the square character is called Assyrian writing and is said to have been in- troduced by Ezra. Thus in the Talmud : 1 "In the be- ginning the law was given to Israel in Hebrew writing and in the holy tongue ; but it was given to them again in the days of Ezra in Assyrian writing and in the Ara- mean tongue ; Israel chose the Assyrian writing and 1 Quoted by Geaenius, GescMchte d. heb Spraehe u. Schrift, p. 150. 57 58 GENEKAL INTRODUCTION the holy tongue, and left to the idiots {i.e., the Samari- tans) the Hebrew writing and the Aramean tongue." And again, "Although the law was not given by the hand of Ezra, the writing was changed by him, and is hence called Assyrian l because it came up with them from Assyria." Origen records the same tradition : " They say that Ezra used different letters after the captivity." He further states that in the ancient character Tav had the form of a cross, and that the Hebrew word Jehovah was retained in the ancient character 2 in some manu- scripts of the Septuagint, and was read pipi by persons ignorant of Hebrew. In like manner Jerome says : " It is certain that Ezra the scribe and doctor of the law, after Jerusalem was taken and the temple restored under Zerubbabel, found other letters which we now use, since up to that time the Samaritans and Hebrews had the same characters." To which it may be added that the Samaritans call their letter Hebrew writing in contrast with the square character which they denominate the writing of Ezra. 3 The relation of the square character to that upon the coins was vehemently disputed in the seventeenth cen- tury, particularly between Buxtorf, 4 Professor in the University of Basle, and Cappellus, 5 Professor of Orien- tal Languages in the University at Saumur in France. The former affirmed, the latter denied, the antiquity and 1 Assyria is here, as in Ezra vi. 22, inclusive of Babylonia. Others understand fWllDX as applied to the modern Hebrew character to mean not "Assyrian," but " firm " or " erect " and equivalent to the " square letter; " so Hupfeld in Stud. u. Krit., 1830, p. 296 : Havernick, Einl. I., p. 293. 2 Origen is here plainly in error in saying that this was the old He- brew character ; it is only the square letter in which HIPP could be mistaken for mm. 3 Gesenius, ubi supra, p. 144. 4 De literarum hebraicarum genuina antiquitate, Basileae, 1662. 'Diatribe de veris et antiquis Hebraeorum literis, Amstelod., 1645. HEBREW LETTERS AND VOWELS 59 originality of the present Hebrew character. Each brought an immense amount of rabbinical learning to confirm the position which he had taken. Buxtorf held that there were two characters simulta- neously in use among the ancient Hebrews, a sacred letter in which the tables of the covenant and the Scriptures were written, and a common letter employed in business transactions and for secular purposes gener- ally. The square letter found in all the manuscript copies of the Scriptures was the sacred one. That upon the coins was the secular or common character. During the Babylonish exile the priests and learned men among the Jews, being principally engaged with the transcrip- tion and study of the law, preserved the sacred character and allowed the other to fall gradually into, disuse. The poorer and prof aner classes of the people, who had been left behind, neglected the law, and gave up the use of the sacred character, preserving only that appropriated to secular purposes. From them it passed to the Sa- maritans ; consequently it is the secular character only which is found among them. When Ezra and the other captives returned from Babylon, they brought the old sacred character with them ; and this accounts for the tradition which ascribes its introduction to Ezra, and speaks of it as brought from Babylon. This hypothesis of two co-existing alphabets was sup- ported, 1. By the analogy of other nations, e.g., the an- cient Egyptians, who had a threefold character, the hiero- glyphic engraved on monuments, the hieratic or sacred, and the demotic or popular. 2. By Isa. viii. 1, where " write with the pen of a man " was supposed to mean write not in the sacred but in the common character which everybody employs and can understand. 3. A passage in Irenseus which speaks of sacerdotal letters among the Hebrews. In Irenseus' ignorance of the He- 60 GENERAL INTRODUCTION brew, however, his authority cannot be of much weight, especially as in the very sentence from which the proof is brought there are other palpable errors. The verse in Isaiah means no more than write plainly and legibly. And the analogy adduced might illustrate a similar fact among the Jews, if proved, but cannot of course prove it. Accordingly, this hypothesis, though widely adopted at the time, has long since been abandoned. Gesenius in his earlier writings accepted the old Jewish tradition, and accordingly maintained that the Samaritan or coin letter was the older, and was in use in both the kingdoms of Judah and of Israel until their respective captivities. It was preserved by the remnant of the ten tribes, and through them came into the pos- session of the Samaritans. But the Jews in their sev- enty years' captivity at Babylon, gave up alike the use of their language and of its written character, adopting both the letter and language which they found at Baby- lon. This they brought back with them from the exile and have since retained. But the difficulties remain that the square letter was not in use at Babylon, and that the coin letter was employed by the Jews as late as the time of the Maccabees, and even later. This vexed question was, however, at last set at rest by the researches of Kopp 1 and others in comparative Semitic palaeography. It has been demonstrated that all the Semitic alphabets are connected by affiliation or derivation. The common parent of the whole is the Phoenician, to which the Hebrew coin letter is so nearly allied that it must have been one of its early modifica- tions. This was accordingly the original Hebrew letter, and its adoption in the first Jewish coinage shows that it must have continued in ordinary use as late as the time of the Maccabees. To this succeeded the square 1 Ulrich Kopp, Bilder und Schriften der Vorzeit, 1821. HEBREW" LETTERS AND TOWELS 61 character, not by a sudden change, not by being intro- duced from abroad, whether from Babylon or elsewhere, but by successive and gradual modifications through a considerable space of time ; just as the Greek letter of modern times was formed upon the uncial originally in use. The connecting links can still be traced in Ara- mean inscriptions found in Palmyra and in Egypt, in which the alphabet was in a transition state. The causes which were mainly operative in produc- ing these changes can also be pointed out. There was first a cursive tendency ; in rapid writing the letters are formed in the quickest and easiest way, minor de- tails are neglected, and only the principal strokes nec- essary for their recognition are retained. And instead of the pen being lifted after each letter is formed, it is linked to its successor by a horizontal connecting stroke, which at the end of words naturally runs into a terminal flourish, thus giving a peculiar form to certain final letters. This was followed by a calligraphic ten- dency, which again separates the letters, and aims at regularity and evenness of form, which is so remarkable in the square Hebrew character. Careful comparisons of extant inscriptions have made it possible to note the precise changes which have taken place from time to time in each letter of the alphabet. At what precise time the square character was fully formed, as we now see it, cannot be accurately deter- mined. It must have been before the fourth or even the third century of the Christian era, as is plain from the testimony of Origen and Jerome. Their statements re- specting the Hebrew letters existing in their day iden- tify them beyond question with those that we now have. And it is highly probable that these same letters were in use in the time of Christ from an expression in Matt. v. 18, where Xodh is referred to as the smallest of the al- 62 GENERAL INTRODUCTION" phabet ; this is true of the present Hebrew character, but not of the coin letter nor of the Samaritan. The attempt has been made to determine which letter was in use when the Septuagint version was made in a very ingenious manner, by examining all the places in which an error seems to have arisen from the translator mis- tal»ng one letter for another, and determining in the use of which alphabet these mistakes could most easily have occurred. It has been plausibly maintained that the majority of instances appear to favor the idea that the translation was made from manuscripts in the square character or in one that closely resembled it. Still the uncertainties attaching to this style of argument are so great that it cannot be said to have led to any settled or satisfactory result. Perhaps the safest conclusion is that the square character came into use between the time of Simon Maccabseus and that of Christ. The old character was, however, not wholly displaced by it immediately, as appears from its being still retained in the coinage as late as a.d. 135. Undue importance was attached to this controversy at first in consequence of its being complicated with other questions which were thought to depend upon it. It was treated as though it involved the integrity of the Hebrew text. Cappellus insisted that this was ex- tremely faulty and stood in need of almost constant cor- rection. Buxtorf, as the stanch defender of the integ- rity of the sacred text, was induced to take extreme ground upon the opposite side, and to maintain that the Hebrew Scriptures have been preserved in their original form, even to the shape of the letters in which they were written. To admit that the Jews had de- parted from the primitive Hebrew character, while the Samaritans had retained it, seemed to be yielding some- thing to the prejudice of that steadfast adherence to HEBREW LETTERS AND VOWELS 63 antiquity on the part of the Jews, which gives assurance of the correctness of the Scriptures which have been in their keeping. It seemed to be admitting a superiority in one point at least of the Samaritan over the Jewish Pentateuch, whence a presumption might arise in favor of the greater accuracy of the former. But it is evident that the purity and authority of the Old Testament are not in the slightest degree affected by the shape of the letters in which it is written. It is no detriment to the New Testament to be printed in the ordinary Greek let- ter instead of the old uncial character. The controversy respecting the primitive form of the letters was subsidiary to another of greater importance, which also concerned not the matter of the text but its form, viz., that relating to the antiquity and authority of the Hebrew vowels and accents. The Jewish rabbis of the Middle Ages were almost universally of the opin- ion that the vowels were an integral part of the original text, or else that they were appended to it by Ezra and received the sanction of his inspiration. Aben Ezra, however, a Spanish rabbi of distinction in the twelfth century, is quoted as holding that they were the work of the Jewish scholars at Tiberias. 1 Elias Levita in the sixteenth century was the first to discuss the subject at any length, which he did in the preface to his work en- titled " Massoreth Hammassoreth," published in 1538, where he claims that the vowels were added to the text by the Massoretic doctors at Tiberias. The elder Bux- torf 2 replied to the arguments of Levita, and sought to 1 Aben Ezra says " we receive all the points from the wise men of Tiberias." The younger Buxtorf claims that his meaning in this pas- sage is not that they invented the points, but by a careful collation prepared an accurate edition of the pointed text, which should be ac- cepted. — De Punctorum Origine, p. 20. 2 Father and predecessor of the Buxtorf before mentioned as Pro- fessor at Basle, in his Tiberias sive Commentarius Masorethicus, 1620. 64 GENERAL INTRODUCTION refute his position. Levita's opinion, however, found favor with Cappellus, who wrote a defence of it, repeat- ing, confirming, aud augmenting the arguments of Le- vita, and sent the manuscript to Buxtorf for his judg- ment. Buxtorf returned it to the author, noting his objections to the various positions, and replying to the arguments, confessing frankly the difficulty of the sub- ject but advising against the publication of the views contained in the manuscript as of dangerous tendency. Cappellus then sent his treatise to Leyden to a distin- guished Dutch scholar, Van Erpe, more commonly known by his Latin name Erpenius. He had previously arrived at the same conclusion respecting the Hebrew points with Cappellus, and at once committed his man- uscript to the press. 1 All who held contrary views turned their eyes to Buxtorf for a reply, and he received numerous and urgent solicitations to that effect, among others from Archbishop Ussher. His other labors, however, prevented his preparation of it, though he had projected one and sketched its outline. After his death his son and successor published, in 1648, his treatise on the Origin, Antiquity, and Authority of the vowel points and accents, arguing that the points, if not in existence prior to Ezra, were at least introduced by him. To this work, which contained not a few asperities and person- alities, Cappellus replied in a tone of equal acrimony, defending his former position and endeavoring to in- validate that of his adversary ; this was not published until 1689, more than thirty years after his death. The views of Buxtorf upon this subject were, at the time, very generally adopted by the orthodox party in the Church, both in England and on the continent, while those of Cappellus, though approved by many, were severely reprobated by others as dangerous and i Under the title, Arcwurp Puoctuationis Revelatum, 1624. HEBREW LETTERS AND VOWELS 65 heretical. It was made an article of faith in the " For- mula Consensus Helvetica '' that the Old Testament is inspired " as to its vowels or points, or at least as to the power of its points." And the learned Puritan di- vine, John Owen,- wrote earnestly against the views of Cappellus as adopted by Bishop Walton in his "Prole- gomena to the London Polyglot." He evidently argues under the impression that to deny the inspiration of the points is to destroy all certainty in the interpretation of the Old Testament, and thus undermine its authority. It is now universally conceded, however, that the vowel-points were not coeval with the letters of the text for the following reasons : 1. The cumbrous minuteness of their notation ren- ders it extremely improbable that they were in use so long as the Hebrew was a living tongue. 2. The analogy of the kindred dialects. Syriac and Arabic were originally written without vowel signs ; these are a later invention. The Samaritan uses sub- stantially the old form of the Hebrew letters, but has no vowel signs ; neither have the ancient Hebrew in- scriptions, nor the Phoenician monuments. 3. The synagogue rolls, to which special sacredness is attached, never have the vowel points ; this can only be accounted for, if the points are not an original con- stituent of the sacred text, but a subsequent innovation. 4 The early Jewish tradition that the vowels were -preserved orally from the time of Moses, and first com- mitted to writing by Ezra, is like the similar tradition respecting the altered form of the letters, accepted as an admission that the vowel signs were not coeval with the letters ; their introduction is supposed to be attrib- uted to Ezra for the sake of gaining for them greater credit. 5. The present vowel system could not have been in 5 66 GENERAL INTRODUCTION existence when the Septuagint version was made ; for it deviates from it considerably in its manner of trans- literating proper names, and repeatedly translates words as the letters would admit, but the vowels would not. 6. Origen in the third century a.d. in his Hexapla gives a pronunciation of the Hebrew words in Greek letters, which does not agree with the vowel points. 7. It has been a disputed question whether Jerome in the fourth century was acquainted with the present vowel system; but it is now well established that he was not. He generally adheres to that pronunciation and understanding of the text which is yielded by the vowels ; but he often speaks of the ambiguity of words, which are only ambiguous when written without the points. In such cases he frequently chides the LXX. for departing from "the Hebrew verity," which he him- self follows. And though it might seem in some in- stances that he was governed by something in the text, additional to the letters, in deciding upon its meaning, a careful scrutiny of his language shows that this was not the case, but that his decision rests upon the sense required by the context, the rendering of later Greek versions, or, which is his main dependence, the instruc- tion received from his Hebrew teacher. 1 He speaks of 1 Jerome on Hos. xi. 10 : " The word MJM, which is written with three letters, Mem, Jod, Mem, if it be read Majim means water, if mijam from the sea." On Hos. xiii. 3: "Locust and chimney are written with the same letters, Aleph, Res, Beth, He, which, if read arbe, means locust, if aruba chimney," He adds that whoever follows the LXX. in reading " locust " " refuses to receive the Hebrew veri- ty." On Hab. iii. 5: "In Hebrew there are three letters, Daleth, Beth, Res, without any vowel, which if read dabar signify word, if deber pestilence. . . . This word can be variously read, and this ambiguity is the cause of various readings and interpretations," i.e., in different versions. Quoted from Buxtorf, De Punctorum Origine, p. 160 ff . Hupfeld in Stud. u. Krit. , 1830, p. 584, gives the following from Jerome, Epist. to Damasus, 125, Question 2 on Ex. xiii 18, where the HEBREW LETTERS AND VOWELS 67 vowels, it is true, and of accents; but by vowels he means the vowel letters, and by accents the pronuncia- tion of a word, or what is added to its written form in giving it vocal expression. 1 8. The Targums, particularly those of Onkelos and Jonathan, adhere very remarkably to the sense as given by the points. But, as in the case of Jerome, this proves not an acquaintance with the points, but with the traditional interpretation of the text which has for the most part faithfully preserved its meaning, and with which the points are also in accord. Some in- stances in which they depart from the reading yielded by the points can be accounted for by their attributing LXX. render " in the fifth generation " and Aquila " armed." " I open the Hebrew roll, and examining the characters themselves more carefully, I find it written Tahamisim. [The pointed text is O^lCTSrp, not mi but mu. ] All the contest is about the word hamisim, which is written in these letters, Heth, Mem, Sin, Jod, Mem, whether it means five or armed. And indeed we cannot deny that five is expressed by this word, but in the plural number, not quinta as they interpret in the singular, but quinti. All Judea, however, exclaims in unison that Aquila, as in other places, so particularly in this, has translated properly, and the seats in the synagogues universally agree therewith. For the same word and written with the same letters may have with them different sounds and meanings." This passage is partially quoted by Buxtorf, p. 151, who lays great stress upon Jerome's careful inspection of the Hebrew text in order to decide between discordant versions, where the letters are ambiguous and the vowels alone determine the meaning. But the fact is that here, as in some other cases, Jerome fails to give the vowels correctly, while holding to the sense yielded by the vowel points as commonly interpreted. And he puts his decision not upon the ground of what he discovers in the text, but the accurate rendering of Aquila vouched for by universal Jewish testimony. 1 He says on lea. xxxviii. 14 : If the letter Vau be placed as a middle vowel between two Samaehs, it is read Sus and means a horse ; if Jod, it is read "Sis" and denotes "a swallow." And in Epist. 73 to Evagrius : " It matters not whether it be called Salim or Salem, since the Hebrews very rarely use vowel letters in the middle of a word, and the same words are pronounced with different sounds and accents at the pleasure of the readers and in different regions." 68 GENERAL INTRODUCTION an Aramean sense to Hebrew words and forms, or adopting in preference an allegorizing interpretation. 1 9. It is more difficult to determine whether the Tal- mud contains any allusion to the vowel points, but a thorough examination of the case has shown that it does not. It nowhere mentions the name of any vowel or accent. It uses the word ff'ttSB accents to de- note the vocal expression rather than any written sign ; and D^plCB verses, but whether these did or did not cor- respond with the verses at present in use, there is no proof of the existence of the accentual system subse- quently built upon them. The phrase is of frequent occurrence " do not read so, but so," 2 where the word 1 Thus Onkelos renders Deut. xxvi. 5 : " Laban the Aramean sought to destroy my father," instead of "An Aramean ready to perish was my father," not because he read "73 N for 1355, but he treated the lat- ter as the Aramean Aphel "DISS. Allegorizers were also fond of availing themselves of the ambiguity of the letters apart from the vowels in order to discover a multiple sense in Scripture, and thus en- hance its richness. 1 Buxtorf , pp. 96, 101, gives these among other illustrations of this punning upon Scripture language. On Isa. liv. 13, All thy children shall be taught of Jehovah, the remark is made : Do not read "thy children " (:p!fi) but " thy builders " (^32). On Ps. i. 23, And prepareth a way that I may show him the salvation of God, it is writ- ten : Do not read " prepareth " {QIC) but "there is" (D©). Rab says he that salutes his fellow before he has prayed is as if he made of him a high place, and quotes in proof Isa. ii. 22 ; do not read " wherein " (HB2) but " a high place " (!T!22) is he to be accounted. He adds, p. 98 : " This change of words and variant reading is not made by them from any doubt or contention or dissension, as if they either hesitated about the true reading and wished to introduce an- other, or contended with one another about a different reading in these places ; nor are they conjectures respecting various readings of any word, but as my father correctly and truly wrote in Tiberias ch ix., " because to the literal and genuine reading and explanation of any word they wished to add another, not to deliver the true sense of the word but to elicit some other allegorical srmse.'' They take this liberty because they believe that there is an infinite amplitude of the word of God, and hold this principle : " In order to elicit various HEBREW LETTEBS AND VOWELS 69 in question is capable of different senses, that are only distinguishable by the vowels. The sense intended in each case is indicated, however, not by inserting written signs for the vowels, but by the connection, or else it is supplied by the teacher, as this was primarily designed for oral instruction. Such expressions do not imply that there was any doubt as to the proper reading, or that the common reading required correction. They are simply an ingenious play upon words in order to con- nect ideas with a passage which are conceived to lie hid in its letters, however remote they may be from the sense required by the connection. Sometimes indeed these multiple senses can only be obtained by changing the letters as well as the vowels of the word. Of a similar import is the singular phrase " There is a mother of Scripture,, and there is a mother of tradi- tion." Different opinions are supported by a passage containing a word which written without vowels is ambiguous. Of one it is said: "There is a mother of Scripture," i.e., its source is the text in its established meaning, as commonly and properly read. Of the other: " There is a mother of tradition," i.e., its source is a pronunciation different from that currently accepted, but which some ingenious rabbi has devised, and others have received it from him. It is then added, " There is a mother of Scripture, and there is a mother of tradition," i.e., both views are admissible and included in the mul- tiple sense of Scripture. The ambiguity arising from the absence of vowels opens the way for different con- clusions from the same words, and a certain degree of senses from it, it is allowable in this way by the change of Towels, consonants, and sounds to bend and twist the words of Scripture, proyided no sense, is produced contrary to the word of God and the analogy of faith." 70 GENERAL INTRODUCTION validity is attached to each, although they are not held to be of equal authority. 1 1 In regard to the significance of these Talmndic phrases Hupfeld says, ubi supra, p. 554 : "I must here protest in advance against the utterly perverted point of view under which these expressions are com- monly put, as if they contained real, i.e., critical doubts and contro- versies about the true reading of the ambiguous text ; and, as Gesenius expresses himself, ' transported us to a time when they began to feel more pressingly the ambiguity of an unpointed text. ' This has been long since contradicted in the case of the first formula ' do not read so, but so ; ' and it has been shown as well from the internal character of these readings as from the explanations of the Talmud and the later rabbis, that we have here no critical emendations before us, but simply a play upon particular words in the text, and twisting them in order to con- nect with them certain dogmas, fancies, witty applications, and the like ; a procedure that in the Hermeneutics of the Talmud itself is not reckoned a valid hermeneutical proof, but a spurious or second- ary proof, and merely has for its object to fix those fancies in the mem- ory by a memorial word, i.e., by attaching them to a word in the text of the Bible. Not only vowels, but also consonants, are often changed in a manner so contrary to all the sense and connection, that they can- not for this reason be regarded as real readings. " And p. 561 : " The readings designated by the term 'tradition' are consequently in every instance not critical, but simply theologico- juridical variants; ie., in the Talmudic mode of teaching used as means of proving controverted dogmas by transmitted arbitrary modi- fications of the church reading in words where the original as written — a priori considered, i.e., apart from the connection, grammar, and tradition — would naturally lead to another reading." And p. 564 : " The text of the Bible is thus a basis of infinitely manifold meaning (the forty-nine-fold or seventy-fold face of the law), and therein precisely consists its divine character and pre-eminence. This pre-eminence is principally based upon the absence of all vowel signs ; so that the naked text was the only admissible and legitimate form, not only for use in worship (for synagogue manuscripts), but also for learned use (Talmudic and Cabbalistic argumentations). So little, therefore, did they feel the need of reading signs, that on the contrary the ambiguity of the text was an indispensable need for their aims in teaching, and an essential condition of the lauded manifoldness of its sense. The true reading, however, was in no wise doubtful on this account, but stands already so firm in tradition, that it agrees through- out with our modern text." HEBBEW LETTERS AND VOWELS 71 It thus appears that Hebrew was written without the vowel points when the Septuagint was translated, in the time of Origen in the third century A.D., in that of Je- rome in the fourth century, and down to the comple- tion of the Talmud at -the close of the fifth century. There is documentary evidence of their existence at the beginning of the tenth century. A manuscript of the latter prophets written a.d. 916, and discovered by Firkowitch in 1839, is pointed with the vowels through- out, though with a different system of notation from that now in use. That the writer was likewise familiar with the present system is shown by his frequent use of it in the marginal notes. Aaron ben Moses ben Asher, who lived in the first half of the tenth century, and his contemporary Moses ben David ben Naphtali, each prepared a copy of the Hebrew Bible with vowels and accents ; that of the former was particularly prized for its accuracy, and was widely accepted as the stand- ard. Ben Asher (as he is commonly called) is said to have belonged to a noted family of punctuators, his an- cestors being traced back for five generations, 1 which must certainly reach back into the eighth century. The Massora, a vast collection of critical notes on the Hebrew text, is largely concerned with the vowels. Some of its earlier portions, such as the K'ri and K'thibh, which relate merely to the consonants, are al- ready noticed in Jerome and the Talmud. But of the subsequent additions made to it in the course of cen- turies there are unfortunately no means of fixing the date. It can only be said on the basis of the facts stated above that the vowel signs must have been intro- duced between the sixth and eighth centuries. This was probably in imitation of the Syrians, whose ac- quaintance with Greek learning led them to feel the de- 1 Baer and Strack's Dikduke Hat'amin, p. x. 72 GENERAL INTRODUCTION fects of a purely consonantal alphabet and to supple- ment it by vowel signs. The precise relation of the two systems of points above mentioned has not yet been ascertained. In both of them a and o are represented by the same sign, i by a single dot, e by two dots placed horizontally ; Daghesh has the same double office and is indicated in the same way, and vowels are abbreviated by being combined with the symbol denoting the absence of a vowel. The differences are equally remarkable and are suggestive of distinct modes of conceiving and dealing with the subject. 1 They must consequently have had to some extent a common origin, and then been sepa- rately developed from that point. The system now in use was wrought out in the schools at Tiberias. The other has been called the Babylonish system on the supposition that it was the product of the schools in Babylonia ; this has in its favor that the principal man- uscript containing it generally follows the Babylonish 1 In the recently discovered system u is represented by a vertical stroke which may be intended for a diminutive Vav, a by a figure re- sembling the letter Ayin, and a or 6 by what may possibly be a dimin- utive Aleph. There is no sign corresponding to Seghol ; its place is taken by Pattahh, Hhirik or Tsere, according as it is derived from one or the other. O, which is represented by two dots placed vertically, never occurs in a closed or dagheshed syllable without the accent. All the other vowels are used in three forms according to the nature of the syllable in which they are found : 1. In open syllables, as well as in closed or Dagheshed syllables with the accent or half -accent (Methegh), the vowels appear in their simple form with no added sign of shortening. 2. In unaccented closed syllables, or in place of a compound Sh'va (which is reckoned a, distinct syllable), a horizontal stroke answering to Sh'va is drawn under the vowel sign. 3. In Dagheshed syllables the horizontal stroke is drawn above the vowel sign. The accents are less numerous than in the other system, and their consecution not so uniform. See Pinsker, Einleitung in das Babylonische-Hebraische Punktationseystem. HEBREW LETTE&S AfrD VOWELS 73 readings where they differ from those of Palestine, and that the Babylonish doctors developed their Talmud and Targums after a method of their own, and may have done the same in respect to these accessories of the sacred text. Others prefer a designation which in- volves no theory of its origin, and call it the superlinear system from one of its obvious and striking peculiari- ties, that all the vowel signs are placed above the letters. The Tiberian system, which was adopted in Palestine, was the most minutely elaborated, and finally super- seded its rival altogether, though both were for a time used together in certain places, as is shown by manu- scripts containing the Hebrew text with a Targum, the former having the Tiberian punctuation and the latter the superlinear. The superlinear is found as late as the thirteenth century in a treatise of Maimonides. When the opinion so strenuously maintained of the antiquity and inspiration of the vowel points was set aside by the evidences of their real origin, .many swung to the opposite extreme of entirely rejecting and disre- garding them as not only a human and unauthorized, but a very erroneous, addition to the inspired text. And to this the silly trifling of the rabbis generally and the follies of the Talmud gave no little color. If the vowel points were from the same hands, it was contended that they must be utterly unreliable and worthless. Bibles were accordingly printed without the points ; grammars, lexicons, and commentaries were prepared with no refer- ence to them, and no regard for their authority. A careful and extended examination of the whole matter has, however, led scholars to the adoption of a medium course. The points are not inspired, but they are substantially correct. The signs were the invention of the Massorites, but the pronunciation which they yield was not. By a most careful and minute system of 74 GENERAL INTRODUCTION notation they have recorded the precise sounds of all the words as a steadfast tradition had conveyed it to them. There is nothing in the system like rabbinical trifling or cabbalistic mysteries. The text itself is left un- touched, and is wholly unaffected by the addition of these accessories, which furnish the key to the pronun- ciation and remove the ambiguity consequent upon the use of a purely consonantal alphabet. The comparison of the kindred dialects lends a strong confirmation to the accuracy of the Massoretic points. The pronuncia- tion of Jerome agrees with that yielded by the points. That of Origen and the Septuagint differ from it, but they do so by system and by rule, as it might be ex- pected that a provincial pronunciation would. And Jerome informs us that the Hebrew was differently pro- nounced in different countries. The Septuagint and Origen give the Alexandrian pronunciation ; the Mas- soretic points that of Palestine, which doubtless best represents the true original sounds. It would be in the last degree unwise to refuse such an invaluable aid be- cause it is of later origin than the letters themselves. The points form what may be called a traditional com- mentary upon the text, conscientious!}' noted down by learned Jewish scholars under circumstances peculiarly favorable for a correct understanding of it. They are a most important help, which ought not to be slighted ; and though they may be departed from in cases of evi- dent necessity, they should be adhered to unless there are very good reasons for not doing so. The accents obviously belong to the same system with the vowels, and must have been introduced at the same time, however the strife may be decided as to their original design, whether they were intended to guide in cantillation or to serve as marks of interpunction. HEBEEW MANUSCEIPTS The Scriptures in their original form are preserved in manuscripts. Hebrew manuscripts of the Old Testa- ment are neither so numerous nor so ancient as the Greek manuscripts of the Scriptures, but there is much less divergence among them owing to the extreme care with which they have been transcribed. Hebrew manuscripts are of two classes, those for synagogue use and those for the use of private persons. Synagogue manuscripts contain the portions of the Old Testament selected for public reading in the worship of the synagogue. The law, which is read in regular course, is upon one manuscript. The prophets, both the former and the latter, or the historical books of the second division of the Hebrew canon, as well as those which are strictly prophetical, are not read in course. But lessons are selected from them for each Sabbath, called Haphtaroth, 1 corresponding with the lessons of 1 According to Frankel, Vorstudien zu der Septuaginta, pp. 50, 51, the Haphtara is so called from "fOB to open, since it was originally the opening part of the service. The ancient custom was to deliver an explanatory discourse as introductory to the reading of the law. In this discourse the speaker included some passage selected at will from other parts of Scripture, commonly from the prophets, which was germane to the lesson of the day. This crystallized into a regular series of lessons from the prophets read prior to the section from the law. Subsequently this order was reversed, but the name Haphtara was retained. Zunz, Gottesdienstliche Vortrage, p. 5, gives a pre- cisely opposite explanation of the term. He holds the common opin- ion that the lesson from the prophets was called Haphtara (cessation 75 76 GENERAL INTRODUCTION the law called Parashoth ; these are written upon a sep- arate manuscript. Then the Book of Esther, which is read at the feast of Purim, and the rest of the five small books, called Megilloth, or Rolls, the Song of Solomon, Ecclesiastes, Ruth, and Lamentations, which are severally assigned to fast or feast days, are written separately. Synagogue manuscripts are upon rolls, written with the greatest care and conformably to rules prescribed in the Talmud, which are needlessly and superstitiously minute. They must be written upon parchment pre- pared from the skin of a clean animal, and for this spe- cial purpose, in the square letter, without vowels or ac- cents, 1 with black ink, in columns, or in the two Songs of Moses, Ex. xv., Deut. xxxii., in parallel clauses. No word nor even letter must be written without the scribe first looking each time at the original which he is copy- ing. The extraordinary points and the letters of un- usual size, position, or form were carefully copied. The manuscript must be corrected within thirty days after it was written, and should four errors be found on any page it was condemned. Very few of these rolls have come into the possession of Christians, since they are buried or destroyed by the Jews when they become old to save them from the danger of desecration. The place of sepulture for manuscripts and other sacred objects which are no longer fit for use is called the Gheniza (itra). or dismission), because with it the services of the day were concluded, and then the congregation was dismissed. 1 It is stated by Eiehhorn, Einleitung, 4te Ausg., p. 462, that some few manuscripts have a double text, one pointed and one unpointed side by side; the unpointed as prescribed for the scribe of the law, the pointed in order to aid the public reader in difficult places which he did not understand, and, therefore, did not know how to read and pronounce correctly. HEBKEW MANUSCRIPTS 77 The private manuscripts contain sometimes, though rarely, the entire Old Testament, more commonly only a part, sometimes only a single book. These are occa- sionally written upon rolls, but for the most part they are in the shape of books of various sizes, folio, 4to, 8vo, and 12mo. Those that are written in the square character are mostly pointed with the vowels and ac- cents. The letters were all written first and the points added subsequently, sometimes by a different person and generally with a different pen and ink. A large proportion' of manuscripts passed through several hands in the course of their preparation. One wrote the consonants, another appended the vowels and K'ris, a third corrected it, a fourth added the Massora and scholia, and perhaps a fifth retouched it after it became defaced by age or use. In some cases several of these offices were performed by the same person, and it is so recorded in the manuscript. They are nearly all writ- ten with black ink, the initial words or letters being frequently ornamented and the margin decorated with figures of flowers, trees, or animals. The material is in many cases parchment or prepared skins, more fre- quently cotton or linen paper. The prose portions are mostly in columns, the poetic portions in clauses or par- allel members. The Hebrew text sometimes stands alone, but is often associated with a translation, a Tar- gum, or an Arabic or some other version, disposed in alternate columns, or written with the text in alternate verses or lines, or in the margin in a smaller character. The upper and lower margin often contain the great Massora, a body of critical traditions concerning the text, sometimes a rabbinical commentary, prayers, psalms, and the like. The outer margin is for correc- tions, scholia, various readings, notices of the legal and prophetic sections, commentaries of the rabbis, etc. The 78 GENERAL INTRODUCTION inner margin, i.e., the space between the columns, is de- voted to the K'ris and the little Massora. A space is left between different books, but not between the two books of Samuel, the two of Kings, the two of Chroni- cles, nor between Ezra and Nehemiah. In the arrange- ment of the books a few manuscripts follow the order of the Talmud, others, and particularly the Spanish, adopt that of the Massora, but there is great diversity, especially in the Hagiographa. 1 There is often a sub- scription at the end of the manuscript, stating the name of the transcriber and that of the punctuator, the date of transcription and sundry other matters pertaining to it. Some private manuscripts are written in the rabbini- cal character, mostly upon paper, without either the points or the Massora, and with many abbreviations ; they are of little value and of comparatively modern date. The determination of the age of Hebrew manuscripts is attended with great, and in many cases with insuper- able, difficulty. In other palseographic investigations, as in Greek and Latin, the shape of the letters, which varied at different periods, supplies an important crite- rion. But in Hebrew no aid can be derived from this quarter. With unimportant variations the square letter is the same in all existing manuscripts. The Jews in- deed distinguish what they call the Tarn and the Velsh letter. The Tam is so called from a person of that name, or more probably in its appellative sense "the perfect letter," and is found on the rolls of the German and Polish Jews. The Velsh character is more common in those of the Spanish and Oriental Jews. But the letters are no sure guide even as to the country in which the manuscripts were prepared. The most reliable 1 See the volume on the Canon, pp. 206, 207. HEBREW MANUSCRIPTS 79 help in settling the age of manuscripts is found in the subscriptions, which have been before alluded to. But unfortunately these are wanting in the majority of cases ; and when they do exist, it is not always easy to under- stand them. It is often uncertain from which era they are to be reckoned, whether from the creation, the era of the Seleucidse, the destruction of the second temple, or some other point of time. There is an ambiguity also in the mode of writing numbers ; it is usual to omit the thousands, and sometimes even the hundreds. And in some cases the statements are misleading, whether from an unintentional error in recording the number, or from deliberate falsification with a view of enhancing the value of the manuscript by making it appear older than it actually was. Some scholars have proposed a division of MSS. into Massoretic and non-Massoretic, meaning by the former those which follow the text as corrected and fixed by the revisions and labors of the Massorites; and by the latter such as contain a text which did not pass through their hands, but has come down independently of them. The suggestion, however, is futile. All the MSS., without ex- ception, represent the Massoretic text with greater or less accuracy. This text was not the creation of the Masso- rites ; they simply perpetuated what they found already in existence, and placed such guards about it as might forever preserve it from corruption or even the minutest change. Hebrew MSS. have been obtained from Jews in India and in China, and when collated have been found to yield the same text as that in our copies of the Bible. Mention is made in Jewish writings of the twelfth and thirteenth centuries and in the margins of MSS. of cer- tain ancient codices which were held in high repute for their accuracy, and regarded as standards from which others were transcribed, or by which they were 80 GENERAL INTRODUCTION corrected. Such were the codex of Hillel, that of Ben Asher, that of Ben Naphtali, that of Jericho, of Sinai, of Israel, Sanbuki and Taggin. These have all perished, with the possible exception of the codex of Ben Asher, and little is known of them now. Great pains have been taken, however, to gather from all available sources the various readings which are attributed to them. The codex Ben Asher, so called because punctuated by Aaron ben Moses ben Asher, who has before been mentioned as flourishing in the former half of the tenth century, is reported to be still preserved at Aleppo. And a copy of the prophets written without vowels by Moses ben Asher (the father of Aaron), a.d. 895, is said to be in the Karaite synagogue at Cairo. One of the oldest manuscripts whose date is certainly known is that of the latter prophets, now in the library at St. Petersburg, formerly belonging to the Society of History and Antiquities at Odessa, first described by Pinner 1 in 1845 and subsequently published in fac-sim- ile under the direction of Professor H. L. Strack. It is remarkable as having the superlinear punctuation ; part of the last chapter of Zechariah and of the first of Malachi are left unpointed. It is written on parchment and in small folio ; there are two columns on each page, with the Massora in the margins. The subscription is dated a.d. 956, and states that the MS. was written forty years before, i.e., a.d. 916. Pinner further describes in the Odessa collection a leathern roll of the Pentateuch, intended for synagogue use and therefore without the vowels, accents, or Masso- ra. It was brought to Odessa from Derbend in Da- ghestan, and consists of 45 pieces, is 56 ells long and an ell broad. According to the subscription it was cor- 1 Prospectus der Odessaer Gesellschaft gehorenden Heb. Manu- scripte. HEBREW MANUSCRIPTS 81 rected a.d. 580, with the implication that it was written still earlier. Two others containing fragments of the Pentateuch are dated respectively a.d. 843 and 881. Of all the manuscripts collated by Kennicott that which he esteemed the oldest is numbered 590, and contains the prophets and Hagiographa. It was written, according to its subscription, a.d. 1018 or 1019, and is in the Imperial Library at Vienna. That which he numbered 154 is in Carlsruhe. It is called the Reuch- lin Manuscript, is dated a.d. 1106, and contains the prophets. No. 126, a manuscript of the latter prophets, which Kennicott thought should be assigned to the be- ginning of the fifteenth century, is by Dr. Margoliouth referred to the sixth, and by Heidenheim between the sixth and the eighth, and by Strack not earlier than the latter part of the eighth century. 1 The manuscripts which De Rossi thought the oldest in his collection were some fragments of the Pentateuch, which he rescued from the Gheniza at Lucca ; one of these he at first attributed to the eighth century, but subsequently said that it might perhaps belong to the ninth or at least be quite ancient. In a collection made by Firkowitch, a Karaite Jew, and sold to the library at St. Petersburg, one fragment of the Pentateuch was dated A.D. 489, and another a.d. 639, while other manuscripts are attributed to the eighth, ninth, and tenth centuries. 2 'Dr. Margoliouth reports having seen a MS. in the possession of a Jewish family in Damascus, which according to a note upon its title-page belonged to the third century ; and another in the neighborhood of the city, which was reputed to have been written in the time of the Macca- bees, and the oldest sheets of which he was disposed to refer to the sixth century a.d. 2 A very complete account both of the lost standard codices and of the oldest extant MSS is given by Strack in his Prolegomena Critica, 1873, pp. 14-55. 6 VI VERSIONS The Septuagint The Old Testament has been preserved to us not only in its original language, but in other languages likewise into which it was early translated. There are four ver- sions of the Old Testament, which are both ancient and immediate, or in other words were made before the period of the Massorites and from the original text. These are the Greek Septuagint, the Aramean Targums, the Syriac Peshito, and the Latin Vulgate. They may be regarded as severally representing distinct tradi- tions of the sacred text, the Septuagint that of the Alex- andrine Jews, the Targums that of the Jews of Palestine, the Peshito that of the Oriental Church, and the Vul- gate that of the Western Church. Of these, two extend likewise over the New Testament, viz., the Peshito and the Vulgate. The Septuagint and the Targums are con- fined to the Old Testament. A version which is made not from the original, but from a previous version, is said to be mediate. There are several versions which are immediate in the New Testament and mediate in the Old, as the Latin Itala and the Hexaplaric Syriac version, both of which are made throughout from the Greek Bible, that is, from the original in the New Testament and from the Septuagint version in the Old. Others are mediate in both Testaments, as the Anglo- Saxon, which was made from the Latin Vulgate. The first language into which the Old Testament was THE SEPTUAGINT 83 translated wa,s the Greek, and the first translation made of it, or indeed of any work whatever, was the Septua- gint. 1 Much obscurity clouds the question of its origin. The earliest tradition on the subject is found in a letter still extant, which purports to have been written by a certain Aristeas, 2 occupying an important position at the court of King Ptolemy Philadelphus, to his brother Philocrates. In this it is related that Ptolemy Phila- delphus was anxious to obtain for his library at Alexan- dria copies of the laws of all nations, and was advised by his librarian, Demetrius Phalereus, to procure the Jewish law. Whereupon the king ordered the release of all the Jews that were in bondage in his dominions, and despatched an embassy with costly presents for the temple at Jerusalem and a large sum for sacrifices, to- gether with a letter to the high-priest, Eleazar, asking that a copy of the law should be sent him with six men from each tribe competent to translate it. In compliance with this request the law was splendidly written in gold letters and sent along with seventy- two men well skilled in Hebrew and Greek. They arrived on the same day that a victory was gained over Antigonus by the king's fleet, and were re- ceived with great ceremony, lodged sumptuously in the palace, and a feast of seven days was held in their honor, during which they severally astonished the king by their wise answers to his questions. A splendid 1 According to a statement by Aristobulus, a Jewish writer, who is said to have flourished in the reign of Ptolemy Philometor, 181-146 b.c, which is preserved by Clemens Alexandrinus andEusebius, a pre- vious translation of the law was made before the Persian domination, and Pythagoras, Socrates, and Plato borrowed some of their ideas from it. But this has no other support, and has little intrinsic proba- bility. 2 JElian, in his history, speaks of an Aristseus, coeval with Ptolemy Philadelphus, who was one of the chief nobles that governed Syria under King Antiochus. — Hody, De Bibliorum Textibus, p. 2. 84 GENERAL INTRODUCTION house on an island was then assigned them for their labors, where they completed the translation in joint session in seventy-two days. Their finished work was ■ then read to the assembled Jews, and elicited from them the highest applause and unanimous approval ; and a solemn curse was pronounced upon anyone who should ever alter what was so accurately and sacredly done. The king was highly delighted with the issue, and filled with admiration of the law, and dismissed the trans- lators with rich presents for themselves and for the high-priest. Aristobulus speaks of the law as translated under the direction of Demetrius Phalereus in the reign of Ptol- emy Philadelphus, but gives no further details of the matter- It is disputed whether he is to be regarded as an independent witness to the existence of such a trans- action, or simply repeats it on the authority of the letter above referred to. Josephus and several of the Chris- tian fathers refer to Aristeas and credit his story. Philo believed that the translators were inspired in the execution of their task, and says that a festival was annually observed on the Island of Pharos in commem- oration of their work. The story was subsequently em- bellished with miraculous features. According to Justin Martyr the seventy-two translators were shut up in as many separate cells with no communication with each other, and severally made independent translations, which, when compared, were found to agree perfectly in every word ; and he says that he saw the remains of these cells when visiting Alexandria. This story is also found in Irenseus and several of the Christian fathers as well as in the Babylonish (but not the Jerusalem) Talmud, 1 and contributed not a little to the veneration 1 Both of the Talmuds speak of the translators and specify thirteen (Jerus. Tal.) or fifteen (Bab. Tal.) alterations made by them in the text. THE SEPTTTAGlNT 85 with which the Septuagint was regarded. Epiphanius repeats the same, with the modification that the trans- lators were put in thirty-six cells, two in each. Justin Martyr and several of the fathers say that the original copy of the Septuagint was preserved in the Serapeum in their day. But the Alexandrine library was burned in the war of Julius Caesar. Cleopatra founded a new library in the Serapeum, which may have contained a Greek copy of the Scriptures that was mistaken for the autograph of the LXX. The letter of Aristeas, how- ever, contains historical mistakes and is encumbered with other difficulties which prove it to be a fabrication. Demetrius was never in charge of the Alexandrian library. He was a distinguished statesman and in high favor with Ptolemy Soter, the father of Philadelphus, but incurred the displeasure of the latter by endeavor- ing to prevent his succession to the throne in the inter- est of an older brother. He was consequently thrown into prison upon the death of Soter, where he shortly died. Philadelphus gained no naval victory over Anti- gonus, such as is reported in this letter, and the letter itself is not mentioned for more than three hundred years after it purports to have been written. The Egyp- tian words occurring in the translation show that it was the work not of Palestine but Egyptian Jews. In sev- eral instances Hebrew words are simply transferred, not translated, showing that the version was not made for only four of which are found in the Septuagint, viz., Gen. ii. 2; Ex. iv. 20, xii. i0; Num. xvi. 15. The post-talmudic tract Sopherim says that five wise men translated the law for Ptolemy. This has led to the opinion that the work was done by five and then approved by either the whole number of the translators or by the Sanhedrim, or else that two separate translations were made at different times, one by five translators, another by sev- enty-two. But Berliner has shown that the number five is a textual error. — Targum Onkelos, II., p. 77 ff. 86 GENERAL INTRODUCTION the king, but for Egyptian Jews, in whose ordinary par- lance these words were familiarly retained. And the version is of unequal merit in different parts of the Pentateuch, proving that it was not prepared by one body of translators. The story plainly originated in the desire to exalt the dignity of the Jewish law by representing Ptolemy Philadelphia as so solicitous to procure it, and to conciliate greater favor for the ver- sion as prepared by royal command and having the sanction of the high-priest. The version was doubtless prepared to meet the wants of the Jewish community at Alexandria. Jews emi- grated to Egypt shortly after the destruction of Jeru- salem by Nebuchadnezzar, Jer. xliii. 5-7, xliv. 1. Large numbers established themselves in Alexandria under Alexander the Great and Ptolemy Soter, who gave them all the privileges of the city and invited them freely to settle there. As they gradually lost the knowl- edge of Hebrew, the law ceased to be understood when read in the original, and the necessity arose of having it interpreted in a language with which they were famil- iar. The law was doubtless translated into Greek in the reign of Ptolemy Philadelphus, 1 285-247 B.C. This fact preserved by tradition forms the basis of the story of Aristeas, and is necessary to account for its ready acceptance and wide prevalence. The name Septuagint was given to it from the supposed number of trans- lators, though some have sought to explain it from an imaginary sanction given to the version by the Jewish Sanhedrim or a similar body in Egypt composed of sev- enty members. 1 As some of the fathers say that it was translated in the reign of Ptolemy Soter, Hody, in order to reconcile these different statements, concludes that the version was made in the last two years of his life, when his son Philadelphus was associated with him on the throne. THE SEPTUAGINT 87 The rest of the books of the Old Testament were trans- lated subsequently at different times and by different hands, but we have no definite information respecting them. It would seem natural to suppose that there would be a demand for the prophets in the first in- stance, as they came to be included with the law in the regular worship of the synagogue. From a note ap- pended to the Greek version of the Book of Esther, it appears that it was translated by a certain Lysimachus of Jerusalem, and brought to Egypt in the fourth year of Ptolemy Philometor, 185 B.C. The prologue to the Book of Ecclesiasticus speaks of the whole Old Testa- ment, " the law and the prophets and the rest of the books " as already translated into Greek before 132 B.C. The entire version received the name given to the part first translated, the Septuagint ; which led some of the fathers into the mistake of supposing that the whole Old Testament was translated at one time and by one body of translators. The different books indicate a great diversity in the character and ability of the translators. The Penta- teuch, particularly in Leviticus and Deuteronomy, is the best for general fidelity. The version of Daniel was so incorrect that it was in ecclesiastical usage laid aside, and another version by Theodotion substituted in its place. Ecclesiastes is slavishly literal, to the disregard of the plainest rules of Greek construction. In Jere- miah and Proverbs especially, and sometimes else- where, verses, and even whole chapters are transposed from their proper order. The liberties which the translators allowed themselves with the text are best explained by remembering the purpose which they had in view. This was not to make a scholarly version with the rigorous accuracy that would be demanded in modern times, but to make the meaning intelligible to plain and 88 GENEBAL INTRODUCTION ordinary hearers and readers. Hence they are more concerned to reproduce the sense than the form. They do not hesitate to substitute literal for tropical expres- sions 1 (Gen. xxxi. 20 ; Num. xxiv. 17; Isa. i. 25, ix. 13), or to insert words or clauses by way of explanation (Gen. iv. 8, xliv. 4, 5 ; Ex. iii. 8 ; 2 Sam. vi. 10), to change rhetorical interrogations into the affirmation or negation implied (Gen. xxix. 15 ; Ex. viii. 22 ; Deut. v. 22, E. V. 25), to alter the person and number without prejudice to the sense (Gen. ix. 2, 6, 16, xxxiii. 13), or change the form of expression (Gen. vi. 5 ; Ex. iv. 13, xi. 8), to simplify anthropomorphisms in order to guard against conceptions unworthy of the deity (Gen. xviii. 32 ; Ex. xxiv. 10 ; Num. xii. 8), or to render a passage in accord with current interpretations (Ex. xii. 40). There are also numerous instances of mistranslation due to ignorance or negligence, or perhaps inaccuracies or defacements of the manuscript used by the trans- lator. One very remarkable variation between the LXX. and the Hebrew is the systematic alteration of the ages of the patriarchs in Gen. v. and xi. This version was at first held in the highest venera- tion by the Jews, not only of Alexandria but even of Palestine, as is shown by the tales which gained currency regarding its origin, and the belief which many enter- tained of its inspiration. It seems to have been read in the synagogues of the Greek-speaking Jews even in Judea itself. 2 It was used by Philo exclusively, and by Jose- phus more than the Hebrew. The apostles and evan- ' Erankel, Vorstudien, pp. 163-79. '' According to Frankel, Vorstudien, pp. 56, 58 note, the Greek ver- sion did not supersede the reading of the law in Hebrew, but was used in connection with it by the interpreter. It is expressly said of the Book of Esther that it might be read in any language, but there is no intimation in Palestine sources that the Pentateuch was ever read in any other than the original. THE SEPTTJAGIOT? 89 gelists, in citing from the Old Testament, draw from it as well as from the Hebrew. The Christians, into whose hands it passed from the Jews, received it with the same veneration that was felt for it by the latter. But as the Christians drew their weapons from this ver- sion in their controversies with the Jews, the latter fell back upon the original Hebrew, and insisted upon the discrepancies between it and the LXX. ; and their for- mer favorable opinion of this version was changed into a settled detestation, which is thus expressed in the Talmud : " The law was written in Greek in the days of King Ptolemy, and darkness was upon the whole earth for three days." " That was a hard day for Israel like the day in which the calf was made." This gave rise to mutual recriminations. The Christians charged the Jews with corrupting the text of Scripture, because they did not receive the rendering of the LXX.;' the Jews retorted the charge upon the Christians, because they did. Hence originated several new translations with the design of giving a more faithful rendering of the Hebrew text. None of them, however, attained ecclesiastical sanction or came into general use like the Septuagint. They are consequently now extant only in a fragmentary state. The principal of these versions were by Aquila, Theodotion, and Symma- chus. Aquila seems to have been a Jewish proselyte from Sinope in Pontus, who flourished, according to Epi- phanius, 1 about a.d. 128. His version is known to have ' Epiphanius gives a wholly unreliable account of Aquila. He says that he was brother-in-law of the Emperor Hadrian, and was by him put in charge of the work of restoring Jerusalem ; while there he be- came a Christian, but on account of his pertinacious adherence to his heathenish ideas and practices, was excluded from the church. "Where- upon he abjured the Christian religion and became a Jew. It has even been suspected that the statement that he was from Fontus 90 GENERAL INTRODUCTION 4 been in existence before a.d. 177, when it is spoken of by Irenseus. It is slavishly literal, aiming to render every word and particle, regardless of elegance and of the proprieties of the Greek language, and often abso- lutely unintelligible. It retains the primary sense of words where derived senses are intended, follows He- brew idioms in violation of Greek usage, manufactures words in the endeavor to preserve the etymology, di- vides words into syllables and translates each separate- ly, and introduces Hebrew words in a Greek form ; and this though there are indications that he was not igno- rant of good style and was familiar with classic authors. This version is accordingly useful in questions of lexi- cography and in the determination of the text, but use- less in hermeneutics and in the explanation of difficult places. A revised edition was issued by the author, in which the same principles were more minutely carried out. 1 He has been charged with perverting Scripture in order to obscure its testimony to Jesus Christ, but Jerome acquits him of any such design. Aquila ren- dered the Hebrew particle ns by aiiv even when it is simply the sign of the definite object : thus in Gen. i. 1, , Ev KetyaXaicp eKTiaev 6 debs aiiv rbv ovpavbv ical aiiv ttjv yrjv. Jerome, who often praises his rigorous accuracy, yet says of his attempt to reproduce the etymology of words : " Quis enim pro frumento et vino et oleo possit vel legere vel intelligere %ev/«a, ottos piapov, aTiXwvoTqra, quod nos possumus dicere fusionem, pomationemque et splmdentiam ? " According to Jerome he was a pupil of Eabbi Akiba, who taught from a.d. 95 to 135, which originated in his being confounded with the Aquila spoken of in Acts XTiii. 2. According to the Jerusalem Talmud his translation was ap- prored by R. Eliezer and R. Joshua, who lived near the close of the first century, or by their pupil R. Akiba. 1 See Field's Hexapla, p. xvi. ff. THE SEPTtfAOtNT 91 may account for his strenuous adherence to the letter of the text. There are conflicting statements regarding Theodo- tion, whether he was a Jew or an Ebionite, and whether he was from Pontus or from Ephesus. His version is attributed to the reign of Commodus, a.d. 180-192, and resembled in style and character the Septuagint, whose errors and deficiencies it was its principal aim to correct. Symmachus was an Ebionite, and is assigned to the reign of Severiis, a.d. 193-211. Two editions of his version are spoken of. He sought to give a free translation, and to express the sense in pure and elegant Greek. In Gen. v. he agrees with the Samaritan Pentateuch in the ages assigned to the patriarchs. Jerome character- izes these three versions by saying that Aquila strives to render word for word, Symmachus prefers to follow the sense, and Theodotion does not differ much from the Septuagint. These versions contained nothing apocryphal, except that of Theodotion, which had the postscript to the Book of Job, and the additions to Daniel, viz.: the story of Susannah, the song of the three children, and Bel and the Dragon. In the course of repeated transcription the text of the Septuagint suffered greatly, until in the early part of the third century Origen complains that every different manuscript contained a distinct text. To remedy this evil, and at the same time furnish aid to Christians in their interpretations of Scripture and in their contro- versies, he undertook the labor of removing the dis- crepancies in the copies of the Septuagint by a com- parison of the best and most accurate manuscripts and at the same time of exhibiting its agreement with or divergence from both the original Hebrew and other existing versions. With this view he planned and exe- cuted his Hexapla, upon which he is said to have spent 92 GENERAL INTRODUCTION twenty-eight years of his life. 1 The Hexapla (six-fold) was so called from its being arranged in six parallel columns ; the first contained the original text in Hebrew characters; the second the same in Greek letters to facilitate its pronunciation ; the remaining columns the versions of Aquila, Symmachus, the Septuagint and Theodotion respectively. Aquila and Symmachus stood next to the Hebrew as most closely allied to it, one in form, the other in sense, and Theodotion followed the Septuagint, to which it was most nearly related. In addition to these columns there were in various parts of the work one, two, or three more for the sake of in- troducing three other partial versions in the various books which they respectively contained. These ver- sions are only known from their connection with this work, and the few scattered fragments of them which have been preserved. Their authors are unknown. They are named from the position which they respec- tively occupied among the versions in the Hexapla, the Quinta, Sexta, and Septima. The Quinta is said to have been found by Origen at Nicopolis near Actium ; and the Sexta was discovered by him in a cask at Jericho ; its author is known to have been a Christian from his translation of Heb. iii. 13, where " thine anointed " is rendered " Jesus thy Christ." There are no certain references to the Septima except one in the Psalms and one in Habakkuk, and these nearly repeat Theodotion. Jerome says that the Quinta, Sexta, and Septima were chiefly used in the poetical books, that is, 1 He thus describes his labors in a letter to a friend : " We are col- lating manuscripts and cannot sup, nor walk after supper, nor rest our bodies ; but even at these times we are compelled to pursue our literary work and correct codices. Nor can we sleep the whole night to refresh the body. I say nothing of our toil from early morning to the ninth or tenth hour. " THE SEPTUAGINT 93 as he explains elsewhere, Job, Psalms, Lamentations, and the Song of Songs. The aim of Origen in this work was not so much" critical as exegetical and polemic. His purpose was not to bring the text of the Septuagint back to its pristine state, but to make it adequately represent the original. With this view he adopted the following plan in its correction. Where the manuscripts of the Septuagint differed he chose as the preferable reading that which was most nearly conformed to the Hebrew and the other versions. The spelling of proper names was in many cases corrected by the Hebrew. When any words oc- curred in the Hebrew to which there was nothing equiv- alent in the Septuagint, he inserted them from one of the other versions, generally from Theodotion, as the one most nearly approaching the style of the Septuagint, and an asterisk was prefixed to them to indicate the fact. If, on the other hand, any words occurred in the Septuagint to which there was nothing answering in the Hebrew, he prefixed an obelos or horizontal line to show that this was the case, but did not expunge them. When a brief passage or clause in the Septuagint was trans- posed from its proper place, as shown by the Hebrew and the other versions, it was allowed to remain where it was, but this was indicated by prefixing an asterisk and obelos combined. When long passages were found transposed, as in Ex. xxxvi.-xxxix. and Jer. xxv.-li., the true order was restored as in the Hebrew for the sake of more convenient comparison. A figure called the metobelos was placed at the end of a passage or clause, to which critical marks had been prefixed, to show how far their influence extended. 1 The lemnisk, a horizon- 1 These various marks, which were in familiar use by the classical critics of the time, were borrowed from them by Origen with a slight modification in the mode of their employment. Verses in Homer, 94 GENERAL INTRODUCTION tal line with a dot both above and below it, and the hypolemnisk, a horizontal line with a single dot beneath it, were merely varied forms of the obelos, and used for the same purpose. 1 Besides the Hexapla of Origen, mention is made by early writers of an Octapla and a Tetrapla. The Octa- pla was not a separate publication, but identical with the Hexapla, which was so called in those parts in which two additional versions were used, and conse- quently eight columns were required. Some modern scholars have used the word Enneapla (nine-fold) to des- ignate those parts of this work in which the three ad- ditional versions were used together ; but this term is not found in any ancient authority. 2 The Tetrapla was a distinct issue of the four principal versions with- out the Hebrew. Hody, Ussher, and others contended that this was prepared in the first instance ; and that at a later time, when the three minor versions, Quinta, Sexta, and Septima had been discovered, the Hexapla was produced as an enlargement upon the original plan. Field, however, agrees with those who maintain, which were found duplicated in exact terms, were marked with an asterisk in that place in which they seemed best suited to the connection, and with an asterisk and obelos combined where they appear to be less appropriate, and were therefore thought to be out of place. The obelos was prefixed to verses which were regarded as spurious. ' Epiphanius invented a special sense for these two marks based upon his conceit that the seventy-two interpreters did their work indepen- dently in thirty-six cells, and that their translations agreed perfectly in sense, with unimportant differences in the form of expression. A lem- nisk with its two dots prefixed to a word or clause indicated a reading suggested by two pairs of interpreters ; a hypolemnisk with its single dot one proposed by a single pair. ! Pentapla occurs in a single instance in a marginal note appended to an ancient manuscript, and denotes the four principal Greek versions together with the Hebrew in Greek letters. Heptapla is found in two notes attached to the Syro-Hexaplaric version and means the Hexapla when the Quinta version is used. THE SEPTTTAGINT 95 on the authority of Eusebius, 1 that the Tetrapla was a subsequent abridgment of the more comprehensive work in order to bring it within a more manageable compass. In a.d. 232 Origen was driven out of his native city of Alexandria, and retired to Csesarea Palestina, where, with few exceptions, he passed the remainder of his days. He died at Tyre in the reign of the emperor Gallus, a.d. 251-254, at the age of seventy. His great work was in fifty volumes and was too cumbrous ever to come into general use, and probably was never completely transcribed. It is not certainly known where it was deposited at the time of his death ; but after lying in obscurity for fifty years it is mentioned as belonging to the library at Caesarea, whether put there originally or brought thither from Tyre. It was still there at a later time when consulted by Jerome. There is positive evidence that the library was in existence in the sixth century ; but it was destroyed shortly afterward in some manner now unknown, and this invaluable treasure perished with it. It has been conjectured that it was burned when Caesarea was taken by the Saracens, a.d. 638 ; the city, however, was not captured by assault, but surrendered by the citizens, who paid a heavy ran- som for its protection. The Hexapla now only exists in fragments scattered through the works of ancient writers or noted on the margin of manuscripts. These have been diligently collected and published at various times, most completely by Field in 1875. 2 1 The whole question turns upon the word used by Eusebius, Eccles. Hist., VI., 16, whether it is ImKaTaffRtviaas or iirurKevdiras, which is a less attested reading, and whether the proper force of the preposition eVl is to be insisted on. 3 Previous collections were issued by Drusius, 1622, Lambert Bos, 1709, and Montfaucon, 1713. The authority of Field in the Prole- gomena to his edition of the Hexapla has been followed in all that is stated above respecting the contents of this work and all that is con- nected with it. 96 GENERAL INTRODUCTION The labors of Origen, instead of remedying the diver- sity already existing in the copies of the Septuagint, tended indirectly to increase them, as not infrequently transcribers preserved the additions made to the text, but neglected the marks by which they were designated as such. Great confusion was thus caused by the min- gling of different versions. Three martyrs, Pamphilus (+309), a presbyter of Csesarea Palestina, Lucian (+312), a presbyter of Autioch, and Hesychius, an Egyptian bishop who suffered death in the same persecution, successively issued carefully prepared editions of the Septuagint. Pamphilus, assisted by Eusebius, issued copies of the Septuagint column of the Hexapla, retain- ing its critical marks. This became exceedingly popu- lar and was used to the exclusion of every other text throughout Palestine. The Emperor Constantine direct- ed that fifty copies of this Palestine edition should be prepared for use in the churches. Lucian revised the text of the Septuagint, supplying its omissions from the other Greek versions, whose expressions he slightly modified, duplicating the translation where it departed from the Hebrew by adding to the existing text of the Septuagint a corrected version of the passage, occasion- ally inserting explanatory clauses to make the meaning plainer, and sometimes substituting synonymous words for those contained in the Septuagint. This recension was adopted at Antioch and Constantinople as well as in the intervening region. The revision of Hesy- chius was adopted in Alexandria and Egypt ; little is known of its peculiar character. 1 These labors certainly 1 Jerome says of these different editions : " Alexandria et Aegyptus in Septuaginta suis Hesychium laudat auctorem; Constantinopolis usque ad Antiochiam Luciani Martyris exemplaria probat ; mediae in- ter has provincise Palestinos codices legunt, quos ab Origene elabora- tes Eusebius et Pamphilus vulgaverunt ; totusque orbis hac inter se trifaria Tarietate compugnat." THE SEPTUAGINT 97 did not result in establishing a uniform text ; on the con- trary, they seriously increased the divergence already existing. Jerome complains of great diversity in the copies and corruption in the text in his day. And the errors which have since arisen in the course of tran- scription during so many centuries have immensely ag- gravated the difficulty. A great number of manuscripts of the Septuagint pre- served in the libraries of Europe and the East have been more or less thoroughly examined by scholars with the view of obtaining the best text. Three princi- pal manuscripts, which have been published in fac- simile, are familiarly known as the Codex Alexandrinus in the British Museum at London, the Codex Vaticanus in the Vatican Library at Rome, and the Codex Sinait- icus in the Imperial Library at St. Petersburg, a por- tion of which is in Berlin and was previously published under the title of the Codex Friderico-Augustanus. The first portion of the Septuagint ever printed was the Psalter, two editions of which (1481 and 1486) appeared before the entire Old Testament was printed. The Greek Old Testament as well as the Greek New Testa- ment was first printed in the Complutensian Polyglot (1514-1517), though in both instances the publication was preceded by the appearance of a complete edition in another quarter. The Aldine Edition of the LXX., as it is called after Aldus Manutius, from whose press it was issued, though printed after the Complutensian, was pub- lished four years before it in 1518. Both these editions claim to have followed ancient and good manuscripts, though no particular account is given of them and noth- ing is now known respecting them. They have been sus- pected of altering the text in order to bring it into closer conformity with the Hebrew or the Vulgate, and of intro- ducing readings from the other ancient Greek versions, 7 98 GENEEAL INTRODUCTION On this account the text of these editions is less valued in a critical point of view than it would otherwise have been. The text of the Complutensian was followed in two others of the principal Polyglots, that of Antwerp and that of Paris. The London Polyglot followed what is known as the Sixtine Edition (1587), so named from Pope Sixtus V., at whose instance it was prepared and published. The work was urged by him, while still a cardinal, upon Pope Gregory XIII., and after his own ascension to the pontifical chair was carried by him to completion. A large number of manuscripts were col- lated for it, and extracts from the Septuagint were gathered from the writings of the fathers. The result was a conviction of the general superiority of the Vati- can manuscript, the text of which was accordingly fol- lowed in the main in this edition, while a copious body of various readings obtained from other sources was added. This was the best and most correct edition of the Septuagint which had yet been issued, though it did not quite deserve the exalted commendation bestowed upon it by the Pope in announcing its appearance, who strictly prohibited any change in it in the future, whether by addition or subtraction, on the penalty of in- curring the wrath of Almighty God and of the blessed apostles Peter and Paul. An edition in which the text of the Alexandrine manuscript was followed, was published at Oxford in 1707. It was principally pre- pared under the direction of Grabe, a Prussian then resi- dent in England, whose name it bears, and who under- took the work under the auspices of Queen Anne, though he did not live to finish it. It was completed by others after his death. A more accurate reprint of this manuscript was published by Field in 1859. The several editions of Tischendorf (the first in 1850, the fifth in 1875) and that of Swete (the first two volumes THE SEPTTJAGINT 99 in 1887 and 1891) follow in the main the Vatican MS., with various readings from other valuable cod- ices. Lagarde has projected a scheme for restoring the original text of the LXX., by a series of approximations, which if it shall be found practicable, may ultimately issue in reaching the end desired with a tolerable degree of certainty. In the bewildering multitude of MSS. and multiplicity of texts he considered it a hopeless task to seek to attain the primitive form of the LXX. by means of individual codices. He therefore proposed a classifi- cation of all MSS., according to their affinities, into three divisions, which shall represent respectively the three recensions of Pamphilus, Lucian, and Hesychius. Each of these divisions may be relied upon to restore, as far as this can be done, the recension from which it has been derived. This will put us measurably in posses- sion of the three different texts of the LXX. which were in circulation in the time of Jerome. And by a careful comparison of these it is hoped to attain the best form now available of the primitive text from which they were all derived. Lagarde himself undertook the recon- struction of the recension of Lucian, and published his results as far as he had proceeded. It has already been said that the Book of Daniel in general use in the Greek Bible was the translation of Theodotion, which was substituted for that of the LXX. on account of the great inaccuracy of the latter. The Septuagint version of Daniel was long supposed to be lost, but was at length discovered in a manuscript in Rome in the library of Cardinal Chigi, and was identi- fied by its agreement with the citations made by Jerome and others of the early fathers from this version, and by the fact that those passages which they remarked upon as wanting in the LXX. were missing in this manu- 100 GENERAL INTRODUCTION script. It was first printed in Eome in 1772, and has been repeatedly published since. Among the authorities cited by Greek fathers or noted on the margin of manuscripts, mention is made of the 'Hebrew,' the 'Syriac,' and the 'Samaritan.' What these severally denote can only be conjecturally determined. The ' Hebrew ' sometimes means the first or second column of the Hexapla, i.e., the Hebrew text in its own proper form or transliterated into Greek characters ; and sometimes a Greek version of the Hebrew by some unknown author, chiefly referred to in Genesis, Job, and Ezekiel. The 'Syr- iac ' was not the Peshito, but a Greek version based upon it, whose author is unknown. The ' Samaritan ' de- notes either the Samaritan Pentateuch or the Samaritan version of the Pentateuch, with which the great major- ity of the references made to it correspond. A large proportion of the early versions of the Old Testament were made from the Septuagint, as Hebrew was at that time understood by few except Jews. Among these were the Latin Itala, and the Syro-Hexa- plaric, which will be considered hereafter. The Bthi- opic version is attributed to Frumentius, Bishop of Axum, in the fourth century, one of the founders of the Christian Church in that region, who is known in Ethi- opic tradition as Abba Salama. This is doubtless the version referred to by Chrysostom (a.d. 354-407), who speaks of the Scriptures having been translated into the language of Ethiopia. It contains, besides the ca- nonical books, several apocryphal writings in addition to those extant in the Greek, such as the Book of Enoch, Pourth Esdras, the Ascension of Isaiah, etc. Accord- ing to Cornill (Ezekiel, p. 37) this version appears in different manuscripts in two forms, the older made di- rectly from the Greek, and another, a later revision of THE SEPTUAGINT 101 the same, altered in various passages into correspond- ence with the Hebrew. Portions of it only have been printed ; the Psalms and Song of Solomon in 1513 and several times since. The Pentateuch and Former Prophets, together with several apocryphal books, were published by Dillman, 1853-1894 ; some of the Minor Prophets and the Lamentations by Bachmann, 1892, 1893. The Egyptian versions belong probably to the latter part of the third or the beginning of the fourth century. They are in three different dialects, that of Upper Egypt, the Sahidic or Thebaic ; that of Central Egypt, the Coptic or Memphitic, and that of Lower Egypt, the Bo- hairic. The last of these is complete in manuscripts ; the Minor and Major Prophets and the Book of Job were published by Tattam in Oxford (1836-1852), and the Book of Daniel by Bardelli in Pisa (1849). Cornill has shown that, while it closely follows the Septuagint, there are occasional indications of its having been influ- enced by the Hebrew. The others only exist in frag- ments. The Gothic version by Ulphilas in the fourth cen- tury, including the whole Bible except the Books of Kings, was made from the Greek. None of the Old Testament is now extant. The Armenian version was made from the Greek, and is attributed to Miesrob, the inventor of the Armenian character, in the early part of the fifth century. It is said to have been since interpolated from the Peshito and the Vulgate. Several editions of it have been pub- lished. The Georgic version was made in the sixth century by an unknown author. The Slavic version of Methodius and Cyril in the ninth century is commonly supposed to have sprung from the 102 GENERAL INTRODUCTION Septuagint, though some have thought that it was made from the Itala. Some Arabic versions of different books of the Old Testament, which are printed in the Paris and London Polyglots, also follow the Septuagint. The Taegums. The versions or paraphrases of the Old Testament in Jewish Aramean are called Targums, from an Aramean root signifying to explain or translate, and which is still preserved in the " dragoman," or interpreter of the Le- vant. It occurs once in the Bible, in Ezra iv. 7. The traditional account of their origin is that, when the Ara- mean had become the language of the people, and the Scriptures in Hebrew were no longer intelligible to them, an interpreter was appointed, as well as a reader, in each synagogue, the office of the former being to ren- der each passage in the language of the people as it was read in the original by the latter. Explicit directions were given as to the method to be observed. Each verse of the law was read singly and then translated ; in the Prophets three verses might be read together unless the connection forbade it. A trace of this usage is still found in many manuscripts, in which the Targum fol- lows the original verse by verse. The interpreter must not use a written translation, nor must he look upon the book used by the reader, lest it should be imagined that he was reading his translation from that volume. The private use of aids to translation was not, however, prohibited. These oral renderings, in the course of time, assumed a somewhat fixed and conventional form ; the best renderings by the most skilful interpreters, particularly in obscure and difficult passages, were re- garded as normative and came into general use. These THE TAKGTTMS 103 traditional interpretations, preserved orally and to some extent perhaps in writing, ultimately formed the basis of written Targums, which had the advantage of greater accuracy and certainty, and gradually superseded the use of extemporaneous translations. The Targums are thus not strictly the work of the person by whom they were finally compiled. They were, on the contrary, the growth of long periods of time, and contained materials gradually accumulated through successive generations. They represent, not so much individual opinions, as the prevalent interpretations and generally accepted ideas of the times when they were produced. From the mode of their preparation it will further appear how variant readings might be embodied in them from the start. As synagogue usage would not be absolutely uniform, alternate renderings of particular words or phrases might be introduced where the compiler did not feel competent to decide between them, so that one would find place in one copy and another in another, or both be put together in the same text. The earliest mention of a written Targum is the statement that one on the Book of Job 1 was shown to Gamaliel, and this must have been before the destruc- tion of the second temple. As it is not probable that this was the first book translated, and especially as a Greek version of the Scriptures had long been in use among the Greek-speaking Jews, it maybe assumed that the law and perhaps other portions of the Bible had already been rendered into the language of the Jews of Palestine, if not for public, yet for private, use. The Targums do not form one continuous version of the Old Testament, but are distinct works compiled by different 1 The subscription to the Book of Job in the LXX. speaks of it as translated from the Syriac, which some understand to refer to an Ara- mean Targum, others to the Hebrew. 104 General introduction persons at different times, each containing one or more of the sacred books. Ten Targums are known to be in existence in whole or in part. There are three on the Pentateuch commonly called that of Onkelos, the Pseudo- Jonathan, and the Jerusalem ; two on the Prophets, one attributed to Jonathan ben Uzziel, and another the Jerusalem ; one on the Hagiographa, containing Job, Psalms, and Proverbs; one on the five Megilloth, Buth, Lamentations, Ecclesiastes, Esther, and the Song of Solomon ; two additional on Esther ; and one on Chronicles. The name of Onkelos has been attached by mistake to the oldest and best of the Targums on the Pentateuch. The Jerusalem Talmud speaks repeatedly of Akilas (D^ipy) the proselyte, whose translation of the Script- ures received the approval of E. Eliezer and B. Joshua, evidently meaning the Greek version of Aquila. The Babylonish Talmud repeats the same identical things of Onkelos (obpDIx), and erroneously makes him the author of the Targum on the Pentateuch. This has led some scholars to suppose that Aquila was the author both of a Greek and an Aramean version, or else that the Targum was based upon his Greek version, and so re- ceived his name. Neither of these, however, was the case. Aquila's strictly literal version for Greek-speak- ing Jews may have suggested the desirableness of hav- ing something similar in the vernacular of Palestine. And some other pupil of the distinguished E. Akiba, whose principle of rigorous adherence to the text of Scripture is well known, may have prepared the Targum. This is in fact the opinion of Berliner, who after a thorough investigation assigns its composition to the second half of the second century a.d., as there is nothing in the Targum to point to a later date. He attributes no weight to the consideration, which some THE TAEGITMS 105 have urged, that its dialect is that of Babylon rather than that of Palestine, since the distinction is mainly in the vowels, and these are a later addition to the text ; and as far as it affects the consonants, it was developed at a subsequent period. The Targum, though prepared in Palestine, did not gain authoritative sanction there, since the official interpreters of the Synagogues were unwilling to Lave their liberty of translation abridged or superseded by its introduction in public worship. For this reason it is never mentioned in the Jerusalem Talmud, which only occasionally cites Targumic render- ings for the purpose of censuring them. When and by whom it was taken to Babylonia is not known. Its re- daction did not take place there, but was completed in Palestine. It speedily rose to great favor, however ; is often cited as authoritative in the Talmud, and is there spoken of as "our Targum" and the "Babylonish Targum." For this reason, coupled with the absence of any allusion to it in the Jerusalem Talmud, and from a notion that its dialect is different from that of Palestine, Frankel 1 claims that it had its origin in Babylonia, and he ascribes it to a pupil of Bab, the founder of the school at Sura and of Jewish learning in that region. It is only in writers of the ninth and tenth centuries and thenceforward that the name of Onkelos is attached to the Targum on the Pentateuch in conformity with the anonymous passage in the Babylonish Talmud. Azariah de' Rossi in the sixteenth century was the first to point out that the Akilas of the Jerusalem Talmud was the Greek translator Aquila, and not the author of the Tar- gum. This Targum adheres closely to the original text, of which it gives for the most part a simple literal trans- lation. Sometimes, to render the meaning clearer, an 1 Targum der Propheten, pp. 6-9. 106 GENERAL INTRODUCTION explanatory word or brief clause is inserted, figurative expressions are resolved into literal terms or plainer figures, the interrogative form is replaced by the affir- mation or negation which it implies, etc. In cases where there were opposing interpretations the Targum follows the accepted view of the passage, or that which in his exegetical understanding of it was the better founded. While aiming to give precise expression to the thought conveyed by tradition, it is never suffered to be overrun, as in other Targums, by legendary or supplementary mat- ter. Only in poetical passages it deals more freely with the text, but never so arbitrarily that it does not at least serve as a basis for the paraphrase. It systematically avoids anthropomorphisms and anthropopathies, and any forms of expression which might seem dishonoring to, or unworthy of, the infinite God ; the word or glory or shekinah of Jehovah is often substituted for the divine name ; and as Elohim may be used of false deities as well as of the true God, it is commonly re- placed by Jehovah in the latter case, and in the former a word is substituted for it that is indicative of false worship. This Targum was first printed without the vowels in Bologna, 1482, together with the Hebrew Pentateuch. It was first printed with the vowels in 1491. The name of Onkelos was not connected with it in either of these editions. The Babbinical Bible of Venice, 1517, was the first publication in which it was entitled the Tar- gum of Onkelos ; and this name has been given to it in most subsequent editions. The oldest Targum on the Prophets commonly bears the name of Jonathan ben Uzziel, 1 a pupil of Hillel, in 1 " Jonathan ben Uzziel translated the Prophets into Aramean from the mouth of Haggai, Zechariah, and Malachi. When he did this, the land of Israel was shaken for forty parasangs, and a voice was heard saying THE TAKGtTMS 107 the first half of the first century a.d. It is nowhere re- ferred to in the Jerusalem Talmud. In tbe Babylon- ish Talmud it is called the Targum, not of Jonathan, but of B. Joseph, who presided over the academy of Pumpeditha in Babylonia in the beginning of the fourth century. Frankel * supposes that he may have made use of some pre-existing renderings of Jonathan in its preparation, and so the name of the latter came to be connected with the Targum, or else the abbreviation "Ti may have been misunderstood to mean Targum of Jon- athan 0|!"i3T> mnn) instead of Targum of Joseph (main SpT). It has been conjectured that, as the name of Onkelos was given to the Targum on the Pentateuch from its being confused with the Greek version of Aquila, a like confusion of the Targum on the Prophets with the version of Theodotion may have given rise to the name of Jonathan, which has the same signification. But this surmise is without foundation, since Theodo- tion is never referred to in either of the Talmuds, nor in any ancient Jewish writing. The Targum on the Prophets borrows several verses and clauses from that on the Pentateuch, where the pas- sages are of like tenor, agrees with it in avoiding anthro- pomorphisms, and renders many words and expressions in the same way where later Targums use different terms and forms of speech, though in some instances agreeing preferably with the latter. It is much more free in dealing with the text than the so-called Onkelos. This Who is this that has dared to disclose my secrets to men ? Jonathan stood up and replied : It is I who have revealed thy secrets to men ; and thou knowest that I did it not for my own nor my parent's glory, but for thine, that dissensions might not be multiplied in Israel. But when he proposed to interpret the K'thubhim, the roice was heard again admonishing him to refrain ; for it is enough. " — The Talmudic tract Masseketh Megilla quoted in Hody, p. 172. 1 Targum der Propheten, p. 11. 108 GENERAL INTRODUCTION is partly attributable to the character of the prophetic writings, and the need of paraphrasing to make their meaning clear and exhibit their application to the expe- rience of the past and the hopes of the future in a manner conformable to the ideas of the time. In the former prophets, which are historical, and in portions of the latter prophets, it often translates strictly. In numerous instances, however, it makes additions which are not required for perspicuity, and introduces legen- dary matter which quite obscures the text, and has no obvious connection with it. It not merely substitutes literal for metaphorical expressions, but undertakes to expound figurative passages in detail by giving a sup- posed meaning to each item in the description. Thus the parable in Isa. v. 1, 2, is not translated but inter- preted throughout. The beloved is Israel the seed of Abraham, the vineyard is the land given them for an in- heritance, the horn of fatness is a high mountain, the fencing is the betrothal, the tower is the temple, the wine- vat is the altar, the grapes expected are good works, those actually produced are iniquities. It was first printed in 1494 with the Hebrew text and a Rabbinical commentary. Of the 'two remaining Targums on the Pentateuch, one is commonly called the Pseudo-Jonathan, from its having been erroneously attributed to the author of the Targum on the Prophets. The other is only ex- tant in a fragmentary condition, and is known as the Jerusalem Targum. Both are seriously defaced with legendary additions, contain numerous foreign words, and are written in the degenerate Aramean of the Je- rusalem Talmud. Neither the Talmud nor any early Jewish writer ever speaks of the Targum of Jona- than on the Pentateuch. Mention is made of a Pales- tine or Jerusalem Targum, the quotations from which THE TARGUMS 109 show that it embraced the entire Pentateuch. These quotations sometimes correspond with the Pseudo-Jon- athan, sometimes with the Jerusalem, sometimes with both, and occasionally with neither. These two Tar- gums are strikingly similar, and in certain portions ab- solutely identical. The only conclusion possible from these facts is that the Jerusalem Targum formerly ex- isted in several different editions, two of which still re- main, one complete, the other in fragments. Azariah de' Rossi speaks of having seen two manuscript Tar- gums on the Pentateuch, which were exactly alike, word for word, one of which was entitled the Targum of Jonathan ben Uzziel, and the other the Jerusalem Targum. It is not improbable that the divergent names may have arisen from a misunderstanding of' the abbreviation i"n, which was read Targum of Jonathan Circi" 1 msin) instead of Targum of Jerusalem (DlSiri iTablDTT 1 ). The inflated character of this Targum made it readily susceptible of manifold additions and altera- tions. Its possessors, desirous of having it as com- plete as possible, inserted in the margin whatever they found in other copies that was not in their own ; this was by future copyists added to the text, which thus grew in variety and extent. The Pseudo-Jonathan now only exists in printed form, based on the edition of Venice, 1591. The manuscript from which this was taken is lost, and no other has yet been discovered. There is one manuscript of the fragmentary Targum in the library of the Vatican, from which it was first printed in the Rabbinical Bible of 1518. Some extracts from it are also found in a codex in Paris, as a sequel to the corresponding passages in Onkelos. This two- fold Jerusalem Targum is based on the simple transla- tion of Onkelos, from which it has been developed by the gradual accretions of later times, and when these 110 GENERAL INTRODUCTION are removed, its primary form is still in a measure discernible. Geiger proposed a very improbable theory reversing this order, and regarding Onkelos as an abridgment of the Jerusalem Targum, whereas the consistent principle on which the former is constructed throughout clearly marks it as the original. The Jeru- salem Targum is assigned by Zunz to the second half of the seventh century. From numerous quotations and allusions it is plain that there was once a Jerusalem Targum on the Proph- ets, which has now almost entirely perished. A brief fragment of it has been found on the margin of the man- uscript numbered 154 by Kennicott. 1 The Targum on Psalms, Proverbs, and Job has been falsely ascribed to E. Joseph (+ a.d. 325). It belongs to a much later date. Proverbs differs from Psalms and Job in being free from legendary additions; and it seems to have been made not from the Hebrew, but from the Syriac version. 2 The Targum on the Megilloth is not so much a translation as a paraphrastic exposition, which runs to the greatest excess in the treatment of the Song of Solomon. These Targums, as well as that on Chron- icles and the so-called second Targum on Esther, probably belonged to a comprehensive Jerusalem Tar- gum on the Hagiographa, which was of late origin, and the work of different men, but of the same general character. So far as is known there is no Targum on Daniel, Ezra, or Nehemiah. The Talmud assigns as the reason why Daniel might not be translated that it reveals the time of Messiah's advent. But as these are the books in which the Aramean sections occur (Ezra and Nehemiah being reckoned one), it is probable that they were left 1 Eichhorn, Einleitung, Ed. 4, II., p. 99, ? Ibid., II., p. 103. THE SYRIAC PESHITO 111 untranslated from a reluctance to mingle an uninspired version with the sacred text. The Sybiac Peshtto. The old Syriac version was called the Peshito or simple, which has been variously explained as denoting that it was the one in common use, or indicating its literal character as a translation, or its adherence to the literal as opposed to allegorical interpretations, or, which is probably the true solution, its simplicity as one single translation in contrast with the composite character of the Syro-hexaplaric version, into which extracts from different translations were incorporated. Bphraem Syrus speaks of it as the work of several translators ; and his statement is confirmed by the dif- ference which is observable in the style and character of the translation in different books. It has been queried whether the version was made by Jews or Christians. In favor of the former it has been urged that it was made by evidently competent translators directly from the Hebrew, and not from the Septuagint, like most of the early Christian translations ; and that it agrees in general with the traditional Jewish interpretation. But, on the other hand, the close affinity of the Syriac with the Hebrew would account for an acquaintance with the latter language on the part of Syrian Christians ; there are none of the arbitrary paraphrastic additions to the text characteristic of the Jewish thought of the period which abound in the Targums ; the rendering of Messi- anic passages agrees better with Christian than Jewish ideas ; and the lack of exactness in rendering the lists of clean and unclean animals in Lev. xi. and Deut. xiv. in- dicates a degree of indifference to, or a want of acquaint- ance with, the details of this ritual observance which is 112 GENERAL INTRODUCTION scarcely supposable in a Jew. It has been plausibly suggested that it may have been the work of Christian Jews. It is one of the best of the ancient versions in accu- racy and general excellence. It adheres closely to the Hebrew text with few variations ; these are more fre- quent in the Book of Psalms than elsewhere, whose liturgical use and often-repeated transcription might easily introduce textual errors. Its agreement in many particulars with the rendering of the Septuagint has led some scholars to suppose that the translators were aided by this version in their work. But its divergence from the Septuagint is much greater than its corre- spondence with it, especially in difficult and important passages, where, if anywhere, dependence upon it might have been expected. Apart from such coincidences as might occur between two independent translations of the same work, the agreement of the Peshito with the Sep- tuagint seems to be largely due to its having been sub- sequently altered into conformity with it in consequence of the high esteem in which the Septuagint was held. That such changes were freely made is apparent from comparing the quotations from the Peshito in Ephraem Syrus with the text yielded by manuscripts. The origin of this version is wrapped in obscurity ; it was made in a period of which we have do written record. It is itself the basis of the Syrian literature and called it forth. The Syrian Church is known to have existed from quite early times, and its necessity would require, even if its existence did not presuppose, such a version. Syrian writers record a tradition that it was translated in the time of the Apostle Thaddeus and Abgarus, king of Edessa, and under their direction. And the still more extravagant claim was made that it was in part prepared in the time of Solomon at the re- THE LATIN VULGATE 113 quest of Hiram, king of Tyre. It is positively known to have been in existence in the fourth century, for Eph- raem Syrus (+ 378) made it the basis of his commenta- ries, and spoke of it as then in common use in the Syrian Church. Many of its words and phrases seem to have already become obscure to him, or at least to have re- quired explanation to make them intelligible to his readers. The probability is that it belongs to the mid- dle of the second century A.D., and emanated from Edessa. This version originally contained all the ca- nonical books of the Old Testament with the exception of Chronicles, but none of the Apocrypha ; these were, however, at an early period rendered into Syriac. The Peshito continued to be the received version throughout the whole of the Syrian Church until the separation between the Monophysites and the Nestori- ans. The Monophysite bishop, Paul of Telia, in a.d. 618 prepared a new translation called the Syro-hex- aplaric version, because it was made from the Septuagint as found in the Hexapla of Origen, retaining all its critical marks. It is slavishly literal, to the disregard of the proprieties of the Syriac language. It is ex- tremely accurate in its rendering of Greek words, and where the precise meaning could not be otherwise ex- pressed, the Greek word itself is often inserted. It is comparatively easy, therefore, by its aid to reproduce the text from which it was translated, and Field has de- rived immense advantage from it in his attempt to re- store the Hexapla. The Latin Vulgate. The necessities of the Western Church early led to the preparation of Latin versions of the Scriptures. Augustin informs us that these were very numerous. 8 114 GENERAL INTRODUCTION He says : " Those who have rendered the Scriptures from Hebrew into Greek can be numbered, but the Latin translators cannot, for every one into whose hands a Greek manuscript came in the first periods of the Chris- tian faith, and who fancied that he had some skill in both languages, ventured to translate." He adds that of these translations the Itala, 1 so called probably be- cause it was made in Italy or was in general use there, was to be preferred on account of its superior accuracy and perspicuity. All these versions were based upon the Greek throughout, upon the Septuagint in the Old Testament and the original in the New. The variety in the translations, aggravated by erroneous and negli- gent transcription, was at length productive of such con- fusion and so many discrepancies, that the complaint was made that there were as many different texts as there were manuscripts. Such endless diversity was natu- rally destructive of all confidence as to the true text in quoting Scripture or arguing from it. Repeated solici- tations were made of Jerome, one of the most learned men of his time, and equally skilled in Hebrew, Greek, and Latin, to undertake the revision and correction of the existing versions. Accordingly, during a visit at Rome, a.d. 382-384, at the urgent request of Damasus, bishop of that city, he began a revision of the gospels, then proceeded to the rest of the New Testament, and after this to the Psalms, which he hastily revised in the first instance. Damasus dying near the end of 384, 1 The attempt has been made to reproduce the Itala by collecting all the citations in the writings of the early Latin fathers. The method was highly ingenious, and was very diligently and laboriously carried into effect, but it did not lead to a satisfactory result. For the fathers made use of different versions, and these not always correctly copied ; and they frequently quote from memory, giving the sense but not the exact words, so that all that could be obtained by the compilation was a mixture of different versions inaccurately quoted. THE LATIN VULGATE 115 Jerome left Eome, and after spending some time in Alexandria took up his abode in Bethlehem. Here he revised the Psalms more carefully in connection with the Hexaplaric text, employing the same critical marks that had been used by Origen. The former of these revisions was adopted at Rome, and was known as the Eoman Psalter; the latter came into use among the churches in Gaul, and received the name of the Gallican Psalter. Jerome continued his labor of correction, un- til he had gone over successively the books of Job, Proverbs, Ecclesiastes, Song of Solomon, and Chron- icles, of each of which books he has left a double preface in his works ; he himself speaks at a later time of having thus corrected the whole of the Old Testa- ment. While thus engaged, however, he became satisfied that no simple revision of previously existing versions would meet the necessities of the case ; and that a new and independent version was required. They were made from the Septuagint, which in many cases departed from or obscured the true sense of the inspired writers. He resolved, therefore, to retranslate the entire Old Testament from the original Hebrew, procuring for that purpose, at considerable expense, the assistance of learned Jews, and making a diligent use of pre-existing Greek versions. In this he was encouraged by many warm friends who shared his views, and earnestly re- quested him to prosecute the work. Great prejudice was, however, awakened by it in the minds of others. Such was the veneration with which the Septuagint was generally regarded on account of the supposed inspira- tion of those who prepared it, that every departure from it was regarded as a falsification of the word of God. Even Augustin begged him to desist on account of the offence given by the alterations which he was making,. 116 GENERAL INTKODUCTION and to limit himself to correcting the existing versions into accordance with the Septuagint. Jerome perse- vered, nevertheless, but was led by these clamors to keep as near the version in common use as possible, and even to retain some things he did not approve, so that the rendering which he prefers in his commentaries often differs from that which he adopted in his transla- tion. In defending himself against the censures and reproaches which were heaped upon him from various quarters, and which he very keenly felt, he lays great stress upon " the Hebrew verity," and the importance of supplying the Christian Church with a weapon which they could confidently use in their controversies with the Jews ; he appeals to the confirmation derived from the later Greek versions of Aquila, Theodotipn, and Sym- machus, and puts the blame of the divergence of the Septuagint from the original not upon the translators so much as upon faulty transcription. He began his trans- lation in 390, but did not complete it until 405. It was not all as elaborately prepared, however, as this length of time might seem to indicate. Some parts of it, at least, were hastily performed. Thus he speaks at one time of translating a thousand verses a day, and of hav- ing completed the translation of Proverbs, Ecclesiastes, and the Song of Solomon in three days. His occupa- tion with the Scriptures for so many years gave him a facility that may account for rapid work in portions with which he was especially familiar. Notwithstanding some marks of undue haste, however, the limitations to which he subjected himself in order to conciliate op- posers and the absence of those philological aids which are within the reach of modern interpreters, 1 his version 1 He seems to have been guided by the importunity of friends ad- dressed to him from time to time in the remarkable order in which he successively took up the books of the Old Testament. " He first THE LATIN VULGATE 11 7 4. must be accorded the precedence above all others of ancient date. The new version, though it was at first viewed with much distrust, and its introduction into the public wor- ship of the Church was impeded by the strong attach- ment of the people to the translation with which they were familiar, slowly but surely worked its way to general favor. Not only its superior accuracy and greater clearness, but its better style and purer Latin gave it a decided advantage over the old version, which it gradually displaced. 1 Augustin's early scruples were so far overcome that he used it in his commentaries, though he seems to have adhered to the old version in translated the books of Samuel and Kings, and published them with his often mentioned Helmed Preface (Prologus Galeatus). In the latter he defended himself in advance against the assaults which he ex- pected on account of his new method of translation. Then followed the Book of Job, supposably because he had just finished revising it in accordance with the Septuagint. Next came all the prophets in their order ; and after them the Psalms. A long sickness here inter- rupted his literary labors, until he again resumed them toward the close of the year 393, with the translation of the three books of Solomon. In the years 394-396 appeared Ezra, Nehemiah, Chronicles, and Gen- esis ; and from that time until the beginning of 404 Exodus, Leviticus, Numbers, Deuteronomy; and finally in this and the following year Joshua, Judges, Ruth, Esther, together with the deuterocanonical con- stituents of Daniel and Esther, as well as the books of Tobit and Ju^ dith." — Kaulen, p. 168. He translated no other of the apocryphal books than those just named. 1 The early Latin versions were not in the classic Latin of the best writers, but in the language of the people, which chiefly prevailed in Italy outside of Rome and in the provinces ; it was archaic in char- acter and unaffected by Greek culture. This shows that they were made at a time when Christianity was principally confined to the humbler strata of the population. The language of Jerome's version is that of cultivated Romans of the fourth century, not the classic style of the golden age of Roman literature. Only occasionally his familiarity with the Itala led him to retain some expressions of the popular dialect. — Kaulen, pp. 130, 181. US GENERAL INTRODUCTION his preaching and the services of the Church. In some places the old version was used, in others the new, ac- cording to the individual preferences of pastors and people. This diversity of usage is clearly reflected in the scriptural quotations found in the writings of the fathers down to the close of the sixth century. Thence- forward Jerome's version may be said to have been universally adopted, and to be entitled to be called the Vulgata, or the translation in common use. This term was in the time of Origen and Jerome applied to the Septuagint in contrast with other Greek versions, and sometimes in a more special sense to the inaccurate copies of the Septuagint which were extensively cir- culated in contrast with the more limited number of corrected copies. It is a matter of dispute whether it was applied also to the Latin version made from the Septuagint. 1 It became the established designation of Jerome's version from the thirteenth century onward. The simultaneous use of Jerome's version and the Itala gave occasion to the frequent correction of one by the other ; and this combined with the errors usually attendant upon transcription introduced a diversity of readings as great as that which existed before Jerome began his labors. The familiarity of transcribers with the old version often led them in copying that of Jerome inadvertently to mingle the texts. There is a manu- script of the eighth century in which individual verses and even whole sections are taken from the old version, showing that the copyist had this before him as well as Jerome's translation. And from the eighth century on- 1 Van Ess, p. 25, contends that the expression " vulgata, communis editio," as used by the Fathers before, in, or after the time of Jerome, never denotes a Latin version. Kaulen, p. 10, maintains the opposite. Roger Bacon, about 1266, uses the expression " exemplar vulgatum,'' which is the first time that the word vulgate is found in its modern application. — Kaulen, p. 251. THE LATIN VULGATE 119 ward there is a manifest disposition to retain all that was in the Itala additional to Jerome's strict rendering of the Hebrew. This led sometimes to duplicate ren- derings of the same clause or sentence, as found first in one version, then in the other. Various attempts were made to remedy the disorder thus arising. Cassiodorus, in the sixth century, with the help of some friends, undertook to correct the text of the Psalter and the Prophets by a comparison of old and valued manuscripts, and gave explicit directions to the monks of his cloister for their guidance in copying the Scriptures. The learned Alcuin was commissioned by Charlemagne to revise the text of the Latin Bible. He completed his work in 801, and it was adopted as the standard throughout the kingdom. A large number of manuscripts from the ninth to the thirteenth cen- turies still exist, which were conformed to this revision. Successive revisions were made by Lanfranc, Archbishop of Canterbury in the eleventh century, and by Cardinal- deacon Nicolaus at Home in the twelfth. This process of individual correction did not restore unanimity in the copies of the Latin Bible. Accord- ingly a new method was inaugurated by Stephen II. of Citeaux, that of correction by corporations. He sought to determine the text not only by a collation of manu- scripts, but by a comparison of the original Hebrew, in which he had the aid of, a learned Jew. His emenda- tions were declared the accepted standard of the Cis- tercian order ; and subsequently other learned bodies adopted a definite form of the text as their own. With this view, collections of various readings were made called Epanorthotee or Correctoria Biblica, in which critical judgments were passed upon these readings, and that which was best accredited. These were pre- pared in two ways. At first a manuscript of the Bible 120 GENERAL INTRODUCTION with a broad margin was taken, and all corrections or remarks that were thought necessary were written in the margin or between the lines. The whole manu- script was then copied as a critical edition, or the crit- ical remarks were transferred to a pre-existing copy of the Bible. At a later time, to make this apparatus more widely useful, the remarks were copied separately without the text, and the body of various readings put in circulation for the correction of Bibles. Their in- fluence was greatly impaired, however, by the circum- stance that they were rarely copied verbatim. Fre- quently only extracts were made at the pleasure of the transcriber, and additions of his own were inserted if other aids were not at his command. The sources of these critical remarks were various ; they were based upon the original Hebrew, ancient manuscripts, the com- mentaries of Jerome, the writings of other Fathers, the Itala, and other early versions and later authorities. 1 The first Correctorium of which we have any knowledge is that of the University of Paris about 1226, which was formally approved by the Archbishop of Sens and had a very wide circulation. Another was prepared under the direction of Hugo St. Clair, the Provincial of the Dominican Order in 1236, of which an improved edition was subsequently issued. One of less importance was published by the Franciscans. But the most valuable of all the works of this description was that which was preserved in the Sorbonne at Paris in the fourteenth and fifteenth centuries, and was long supposed to have been prepared by the Sorbonne, but was really made up by combining that of the Dominicans with another Correctorium of unknown origin. These critical labors, however valuable, did not result in unifying the text, for each represented a separate lit- 1 Kaulen, p. 248. THE LATIN VXTIGATE 121 erary corporation, and transcribers continued to accept or reject readings at their own pleasure, or to introduce suggestions from other sources. Boger Bacon ( + 1284) writes to Pope Clement : " The text is for the greater part horribly corrupted, and where there is no corrup- tion there is yet much doubt. And this doubt arises from the conflict of correctors, for there are as many correctors, or rather corrupters, as there are readers in the world, because every one presumes to change what he does not understand, which is not allowed to be done in the books of poets. For antiquated words and figu- rative expressions are not changed when the poets and books in other departments are read ; but here every reader makes changes out of his own head." 1 Never- theless, it is observable that there is a greater measure of uniformity in the manuscripts of the fourteenth and fifteenth centuries than there had been previously, so that the works above mentioned were not altogether fruitless. It is a singular fact that versions in modern languages made at this time were not translated from contemporary manuscripts, but from those of the ninth and tenth centuries, or even older, showing that these still continued to be used by pious people. The invention of printing opened a new era in the history of the Vulgate. Instead of each separate copy being laboriously written out by the pen, with a fresh accession of errors at each transcription, a whole edition could now be issued from the press identical in every word and letter. The Latin Bible was the first book ever printed, and for a century no other was so fre- quently and largely published. The first edition does not specify either the time or place of its issue, but it is known that it appeared at Mayence in 1450 ; the first that is dated was issued at the same place in 1462. ' Hody, p. 420. 122 GENERAL INTRODUCTION While the copies of the Latin Bible were thus im- mensely multiplied and their cost greatly reduced, no special care was taken at first in respect to the text, which was printed from any manuscript that came to hand or from any edition previously published. Cardinal Ximenes, in 1502, with a number of able scholars whom he had summoned to Alcala for the pur- pose, began a careful revision of the Vulgate on the basis of exemplary ancient manuscripts and the original Hebrew, for the Complutensian Polyglot, which was published some years afterward. It nevertheless gave great offence, and Bishop Nicolaus Ramus compared the Vulgate column, which was interposed between the Hebrew on one side and the Greek on the other, to Jesus Christ crucified between two thieves. 1 Erasmus sought to improve the Latin of the Vulgate. Various attempts were made by others, both Catholics and Prot- estants, to correct the translation by the original He- brew, others still prepared new and independent trans- lations. All this, however, served to increase the confusion already existing instead of relieving it. At length the Council of Trent undertook to remove, by ecclesiastical authority, the discord which was the accumulation of ages, and with which the labors of a long succession of scholars, both individually and con- jointly, with all available external helps, had proved unable to deal effectually. Accordingly, in its fourth session (April 8, 1546), when the subject of the Script- ures engaged its attention, it decreed that "the Vul- gate, which had been approved in the Church by the long use of so many centuries, should be held authentic 1 Cardinal Ximenes had used this simile in his preface with a some- what different application. The three columns were so placed to repre- sent the Latin or Roman Church between the Synagogue and the Ori- ental Church as Jesus between two thieves. The Latin vuxgate 123 in public reading, controversy, preaching, and exposi- tion, and that no one should dare or presume to reject it on any pretext whatever." This, however, would have been of little avail amidst the diversity then exist- ing in the copies of the Vulgate if no measures had been taken to establish some one standard edition. Accordingly, the decree of the Council contained an order for printing it with the utmost possible correct- ness. 1 There was great diversity of opinion in the Council on the subject of this decree. A considerable minority were in favor of taking no action at present, urging that if the translation were corrected into accordance with the original Hebrew and Greek it might be pronounced au- thentic. But this would be the work of years, and could not be undertaken by the Council. It would, conse- quently, be safer to leave things as they had been for the past fifteen hundred years, and let the Latin trans- lations be examined and tested by the original texts. But the majority insisted that this would be to disturb the peace of the whole Christian world, to play into the hands of the Lutherans, and to open the door for all sorts of heresies. If everyone was at liberty to raise the question whether the translation was correct, and to take refuge in the originals and other translations, no 1 This decree of the Council of Trent follows the one fixing the canon of Scripture, and is in the following terms : " Insuper eadem sacrosancta synodus considerans non parum utilitatis accedere posse ec- clesise Dei, si ex omnibus latinis editionibus, qua? cireumferuntur, sac- rorum librorum, qusenam pro authentica habenda sit innotescat, statuit et declarat, ut haec ipsa vetus et yulgata editio, qua; longo tot ssecu- lorum usu in ipsa ecclesia probata est, et in publicis lectionibus, dis- putationibus, prasdicationibus et expositionibus pro authentica habea- tur, et ut nemo illam rejicere quovis prsetextu audeat vel praesumat . . . decrevit et statuit, at posthac S. Scriptura, potissimum vero haec ipsa vetus et vulgata editio, quam emendatissime imprimatur." 124 GENERAL INTRODUCTION one would know what to believe. If, in the providence of God, the Synagogue had an authentic Old Testa- ment, and the Greeks an authentic New Testament, surely the Roman Church, which is dearer to Him than all besides, must be provided with an authentic Bible. There was a wide discrepancy also in respect to the sense in which the Vulgate should be pronounced au- thentic. Some maintained its inspiration, and that Jerome was guided in his translation by the same Spirit that was in the prophets, or, as others preferred to ex- press it, a Spirit like that of the prophets ; others held that the Council was under divine guidance, and its sanction would make the version infallible ; and others still that however the Vulgate might err in trivial matters and in the rendering of words, it could not be charged with errors affecting Christian faith or morals, as the past experience of the Church had shown. 1 If this decree meant no more than to give ecclesiasti- cal sanction to the Vulgate in preference to any other Latin version, there would be little in it to object to. It is observable that no anathema is attached to dis- obedience, as in the decree concerning the canon. The Vulgate is not in explicit terms put on a par with the originals or exalted above them. There is g no direct prohibition of versions in the languages of the people, or denunciation of Protestant versions, as some members of the Council desired. The Vulgate is simply pro- nounced authentic as opposed to other Latin editions then in circulation. And this is limited to public eccle- siastical use ; no restraint is put upon private use of the Bible in any form whatever. It has accordingly been maintained that this was not a doctrinal, but a disci- plinary, decree. It is difficult to believe, however, that in prohibiting its rejection " on any pretext whatever " 1 Hody, p. 489 f . ; Van Ess, p. 211 f. THE LATIN VULGATE 125 there is not a tacit reference to the originals. The well- known sentiment of the majority of the Council makes it clear that the intention was to declare that the author- ity of the Vulgate must not be disputed, even where it clearly departs from them. It certainly was so under- stood at the outset. And this has been defended by alleging that the originals, as we possess them, are cor- rupt and untrustworthy, and that the Hebrew in partic- ular is read in accordance with the vowel points, which are a merely human addition, so that it is better to trust a version made by a competent and reliable translator before the originals became corrupt, and before the He- brew was marred by the addition of the vowel points. Some zealous advocates have even maintained that the originals should be corrected by the Vulgate, and not vice versa. All this is palpably contradicted by the real facts in the case. There are ample means for removing any errors that may have crept into the original texts since the days of Jerome. And it is no disparagement to his version to say that even in its primitive form it could not be put on a par with the text from which it was made. And that version has suffered unspeakably more in the course of transcription during the ages that have intervened since its preparation than the Hebrew of the Old Testament or even the Greek of the New Testament. In order to give full effect to their decree the Council resolved to appoint a commission to correct the text of the Vulgate and have it printed. Accordingly, certain persons were deputed by the Council for this purpose, and the work* was begun. But unexpectedly a mandate came from Pope Paul III. to his legates who presided over the Council, ordering them to proceed no further in the correction of the Bible until the method of pro- cedure had been determined by the Roman assemblage of Cardinals to which was committed the province of 126 GENERAL INTRODUCTION governing the Council. On the receipt of this mandate the Legates ordered that what had already been done in the emendation of the Bible should be brought to them, and that the commission should proceed no further until they received fresh orders. At length the Council was terminated in 1563 after a long intermission, and its de- crees were confirmed by Pius IV. 1 Kaulen, p. 427, gives the following account of the matter from a Catholic point of view : " A commission of scholars was appointed by the Council after its fourth session to revise the text of the Vulgate, and this set itself at once zealously to work. Little was, however, to be expected from their labours, since there was a lack at Trent, not indeed of intelligence and experience, but of manuscripts and old printed copies of the Vulgate. As soon, therefore, as the labours undertaken for the aforesaid end became known in Rome, Paul III., through the Cardinal legates, had the Trent commission stopped until further orders, and gave command that the material already acquired should be sent to Rome. Here the in- vestigations which had been begun were continued with- out interruption. Of their further course little more is known from the years next ensuing than that Cardinal Sirlet was one of the most zealous members of the com- mission appointed for this purpose." At any rate noth- ing came of it. As the preparation of an official standard copy of the Vulgate lagged, repeated attempts were made by schol- ars as the professors at Louvain, and by publishers as Stephens in Paris, Plautinus in Antwerp,|and others, to supply the need by fresh issues carefully corrected, which were offered to the public as answering the re- quirements of the Council, and entitled to be regarded as authentic under the terms of its decree. 1 Hody, p. 493, THE LATIN VULGATE 127 At length Pope Sixtus V., after publishing his edition of the Septuagint in 1587, addressed himself to the task. For this purpose he invited the co-operation of a large body of distinguished scholars, who collated a great number of the best ancient manuscripts of the Vulgate, which he had gathered with much pains and ex- pense. They likewise compared the Scriptural quota- tions in the writings of the Fathers and occasionally resorted to the originals, but cautiously and rarely, as Sixtus himself explains, so as not to disturb what had the sanction of long-continued use. Sixtus reserved to himself the final judgment upon all the readings pro- posed ; and to secure perfect accuracy in the publication he carefully read the whole, and corrected the errors of the press with his own hand, sometimes by the pen, sometimes by printed slips pasted over the mistakes which were discovered. The prefatory bull accompanying the publication contains this announcement : " In this our perpetu- ally valid constitution ... we resolve and de- clare from our certain knowledge and from the plen- itude of apostolical authority that that Vulgate Latin edition of. the sacred page of the Old and New Tes- tament, which was received as authentic by the Coun- cil of Trent is without any doubt or controversy to be reckoned that very one which we now publish, cor- rected as best may be, and printed in the printing office of the Vatican, to be read in the universal republic of Christendom and in all the Churches of the Christian world, decreeing that it, approved as it is, first by the universal consent of the holy Church and of the holy fathers, then by the decree of the general Council of Trent, and now also by the apostolical authority deliv- ered to us by the Lord, is to be received and held as true, legitimate, authentic, and undoubted in all public 128 GENERAL INTRODUCTION and private controversies, readings, preachings, and ex- positions." i The same instrument goes on to prohibit the future publication of any edition of the Vulgate, unless con- formed to this in every particular ; and to require that all previous editions should be corrected into accord- ance with it, and that the same thing should be done with all missals, breviaries, and other church books con- taining passages of Scripture. Any disregard of these regulations, it is affirmed at the close, would incur " the wrath of Almighty God and the blessed apostles Peter and Paul." This edition is dated 1590. The distribution of copies had scarcely begun, when it was arrested by the death of the Pope on the 27th of August in that same year. The scholars who had been engaged in the prep- aration of this edition were much displeased with the changes which Sixtus had at his own discretion made in their work. The stringent regulations in regard to mis- sals and other church books created wide-spread dissat- isfaction, as well as the restrictions laid upon private use, though the Council had made no limitation except in regard to public use. Sixtus' successor, Urban VII., died thirteen days after his elevation to the pontificate, and was succeeded by Gregory XIV. Urgent represen- tations were made to him that the use of Sixtus' edition should be publicly prohibited. Bellarmin, however, ad- vised that this should not be done, but that the credit of Pope Sixtus should be saved by correcting what had been improperly changed, and publishing it anew under his name, with a preface stating that some errors either of the printers or others had crept into the first edition from undue haste. This was accordingly done. Gregory appointed a fresh commission to revise the work, who i Van Ebs, p. 279, THE SAMARITAN PENTATEUCH 129 finished their task with great expedition in nineteen days, and presented the result to the Pope in October, 1591. Gregory XIV. died on the fifteenth of this month and was succeeded by Innocent IX., who died December 30th. His successor, Clement VIII., published this re- vised edition in 1592 under the name of Sixtus V. The preface was written by Bellarmin, in which he says, in flat contradiction to the fact and to his own statement elsewhere, that Sixtus V., when his book had been print- ed and was about to be given to the public, discovered that a few errors had arisen in the printing, and gave direction that the whole work should in consequence be recommitted to the press. His death prevented the ex- ecution of his purpose, and now, in the beginning of the pontificate of Clement VIII., the work which Sixtus V. designed has been completed. This edition of 1592 is the officially recognized standard copy of the Vulgate. 1 The Samakitan Pentateuch. The Samaritan Pentateuch is the Hebrew Pentateuch in Samaritan letters. The existence of the Pentateuch among the Samaritans seems to have been known or at least suspected by European scholars in the latter part of the sixteenth century ; for Joseph Scaliger at that time speaks of the importance of procuring copies of it, and complains of the negligence of Christians travelling in Palestine in not securing what might prove to be of such value to sacred studies. The first copy that was ever brought to Europe was obtained by the Italian Peter della Valle. This celebrated traveller spent twelve years in the East, visiting Turkey, Egypt, Persia, and 1 This conflict of papal authority gave rise to numerous publications at the time, among others to the Bellum Papale, sive Concordia Dis- cors Sixti V. et dementis VIII. circa hieronymianam editionem, by Thomas James, London, 1600. , 9 130 GENERAL INTRODUCTION India, and published upon his return to Eome the best account of those countries that had then appeared. The Samaritans, who are now confined to a few inconsidera- ble families in Nablus, seem to have had at that time small communities likewise in Cairo, Gaza, and Damas- cus. Delia Valle, at the instance of De Sancy, then French ambassador at Constantinople, undertook to visit them and procure a copy of their law. After unsuccess- ful efforts at three of these places he at last purchased two manuscripts from them in Damascus in 1616. One contained the Hebrew text of the Pentateuch in the Sa- maritan character on parchment, which he sent to the ambassador, who deposited it in the library of the Ora- toire in Paris ; the other, on paper, was a Samaritan ver- sion of the same, which he retained himself. After be- ing described by John Morinus in the preface to his edition of the Septuagint in 1628, they were seventeen years later, in 1645, published for the first time in the Parisian Polyglot. They were again printed in the London Polyglot in 1657, corrected somewhat by the •aid of additional manuscripts which had meanwhile been procured. Kennicott collated sixteen manuscripts for his edition of the Hebrew Bible. Their respective ages it is difficult, perhaps impossible, to determine with certainty. Of those which are dated none is more ancient than the thirteenth century, and one belongs to the sixteenth. 1 1 Rev. W. Scott Watson a few years since succeeded in obtaining a copy which was written in a.d. 1232, and another whose oldest portion is dated a.d. 656. The former is in the New York public library. The most sacred copy of the law, which is sedulously guarded in the Synagogue at Nablus, has this subscription, " I Abishua, the son of Phinehas, the son of Eleazar, the son of Aaron the priest, wrote this copy in the court of the tabernacle, on Mount Gerizim in the thirteenth year of the settlement of the children of Israel in the land of Canaan." Of course, no confidence is to be placed in this statement. THE SAMAKITAN PENTATEUCH 131 The origin of the Samaritan Pentateuch has been a subject of vehement dispute. John Morinus, Kennicott, and others claimed that it was derived in the line of direct transcription from the Pentateuch existing in the kingdom of the ten tribes at the time of the schism of Jeroboam. Prideaux, Gesenius, and others connect it with a fact narrated by Josephus (Ant., xi. 7, 8) of the reign of Darius Codomannus. This is that Manasses, brother of the high-priest at Jerusalem, had married the daughter of Sanballat, prince of the Samaritans. Threatened by his brother and the other priests with exclusion from the priesthood unless he put away his foreign wife, he fled to his father-in-law, by whom a temple was built on Mount Gerizim as the rival of that at Jerusalem, and Manasses made high-priest. This is in all probability the same event that is alluded to in Neh. xiii. 28, in which case Josephus has made the mis- take of attributing it to the reign of Darius Codomannus instead of Darius Nothus. The hypothesis is that Ma- nasses carried the Pentateuch with him, and the Sa- maritan copies are derived from it. Le Clerc proposed an intermediate hypothesis, which has met no favor, that the Pentateuch was brought to the Samaritans by the priest sent to teach the heathen colonists the man- ner of the God of the land, 2 Kin. xvii. 28. The following arguments have been urged in favor of the first-named hypothesis : 1. As both the tenor of the history and the language of prophets, who like Hosea and Amos were sent exclu- sively to the ten tribes, prove the existence of the Pen- tateuch in the northern kingdom, there is no need of supposing any other origin for the Samaritan copy than it furnishes. 2. The hostility between the rival kingdoms of Israel and Judah, and subsequently between the Samaritans 132 GENERAL INTRODUCTION and the Jews, would have prevented the former from ac- cepting the books of Moses from the latter, so that the Samaritan Pentateuch cannot be traced to any origin later than the schism of Jeroboam. 3. The Samaritans receive of all the books of the Old Testament only the Pentateuch ; if other sacred books had been in existence at the time when they borrowed the Pentateuch, it is natural to suppose that they would have taken them likewise. 4. The Samaritans would not have desired to assist the Jews after the exile in rebuilding their temple, Ezra iv. 1 ff., unless they were in possession of the Pentateuch. 5. The Samaritan Pentateuch is written in the old Hebrew letter in use before the captivity, and not in the square character which the Jews subsequently adopt- ed, and which the Samaritans might have been expected to employ if it was so written when they obtained it. Upon these grounds, which certainly have a plausible appearance, it was confidently affirmed that the Samar- itan Pentateuch must have sprung from copies exist- ing in Israel at the date of the schism. And this was thought to furnish a strong point in defence of the genuineness of the Books of Moses that they could thus be traced back by two independent lines to such a re- mote period. Indeed, so stringent was it felt to be, that some opposers of the Mosaic authorship were con- strained by it to admit that the reduction of the Penta- teuch to its present form could not be later than the time of Solomon. The reasoning above recited is neverthe- less invalid. Although the Pentateuch did exist in the kingdom of Israel at every psriod of its history, the Sa- maritans are descended, not from the ten tribes, but from the heathen settlers who were imported into the land after it had been desolated and Israel carried into captivity. THE SAMARITAN PENTATETTOH 133 That the enmity subsisting between the Jews and the Samaritans was no bar to the latter borrowing the re- ligious books of the former will appear from considering the causes and the nature of that enmity. The Samari- tans claimed to be children of Israel and brethren of the Jews as often as it was their interest to appear so ; though receding from their claim when it would involve them in trouble to be regarded as Jews. Hence the de- sire of the Samaritans to assist the Jews in rebuilding their temple. Hence, too, their proffers of assistance were persistently refused by the Jews. The claim of the Samaritans to be Israelites, when they were not, was the reason of the hatred which the Jews felt toward them. The refusal of the Jews to admit their unfound- ed claim was the ground of the enmity felt by the Samar- itans. Hence the eagerness of the Samaritans to grasp whatever would support their pretence. They aped the Jews in everything ; their doctrines, their temple at Gerizim, their worship, their very fables were borrowed from the Jews. The Pentateuch was coveted by them because its possession might seem to evidence their Israelitish origin. Their exclusive reverence for the Pentateuch was not because they were not aware of the existence of other sacred books, but it arose out of the nature of their re- ligious system. It was a reason of the same kind that led some heretics in the early periods of the Christian Church to reject the Epistles of Paul, and others to reject everything but those Epistles. They refused to acknowledge what did not suit their creed. It was an arti- cle of faith with the Samaritans that on Mount Gerizim was the place where men ought to worship. According- ly, they disavowed all those Scriptures which recognized worship at Shiloh or on Mount Zion. The Pentateuch itself was altered in more than one place to give sacred- 134 GENERAL INTRODUCTION ness to Gerizim. Joshua is connected in their traditions with the building of the temple on that mountain. Moses and Joshua were held in great esteem. David and Solomon, from their connection with Jerusalem, were objects of extreme aversion. From their point of view they could accept no portion of the Scriptures ex- cept the work of the great legislator. Their proposal to assist the Jews in rebuilding the temple after the exile is a proof, not of their possession of the Pentateuch, but of their desire to be counted a part of the Jewish nation. And while the Samaritan character is older than that in use among the JeAvs, this is no criterion of the time at which they received the Pentateuch, since it is now universally admitted that the square character was not introduced by Ezra. And even if the Samaritans had found the Pentateuch in the square letter, they would have copied it in the character in use among themselves, as they do at present in writing Arabic. The opinion that the Samaritan Pentateuch is derived from copies existing in the kingdom of Israel at the time of the schism is now abandoned by scholars. And in the absence of any definite information as to the time when the Pentateuch was introduced among the Samaritans, the defection of Manasses and the erection of the temple on Gerizim suggest the most probable occasion. On comparing the Samaritan with the Jewish copies it was found that, while agreeing in the main, they yet differ in several thousand readings. A large proportion of these consists of insertions of the vowel letters, the insertion or omission of the copulative conjunction, and other variations which have no effect upon the sense. Quite a number, however, are of greater consequence. In upward of a thousand of its characteristic readings THE SAMARITAN PENTATEUCH 136 it agrees with the .Septuagint against the Massoretic text. Here, then, arises the question as to the source of these variations and their critical value. There has been the greatest diversity of views upon this subject. On its first appearance many were disposed to entertain the most exalted opinion of the Samaritan text, and to regard it as much superior to that of the Jewish copies. Others have held that the Samaritan should be esteemed authoritative at least in those passages in which it agrees with the Septuagint. Others still have put the Samari- tan and Jewish copies on a par as different recensions of equal antiquity and equal claim to authority. But the thorough examination of the subject by Gesenius has shown it to be of no critical value whatever. The manuscripts are not written with the same care as those of the Jews, and differ considerably from each other ; many errors are found in them arising from the inter- change of similar letters, the transposition of letters, and inaccurate orthography. Yet in many peculiar readings they all concur, and some of these are known from the citations of Origen and Jerome to have existed in the Samaritan copies in their day. Apart from the errors of negligence, however, the investigations of Gesenius have shown that the great body of the Samaritan char- acteristic readings are intentional alterations of the text, the reasons for which can still be assigned. These are divided by him into eight classes, under each of which he gives a large number of examples. 1. The first is that of grammatical emendations. Un- usual and anomalous forms are exchanged for those in common use, archaisms are avoided, lack of formal agreement in gender and number is corrected, and the vowel letters are supplied where the original omits them. In many cases these agree with the K'ris of the Jewish text, which had a similar origin. 136 GENERAL INTRODUCTION 2. Explanatory glosses are added to the text. 3. Conjectural emendations are introduced, mostly by the change of a letter or two, to improve the sense or remove some fancied verbal difficulty. Thus in the blessing of Jacob, Gen. xlix. 10, " The sceptre shall not depart from Judah, nor the ruler's staff from between his feet," libjfl his feet, by the change of 1 into % is converted into "pbin Ms standards — " nor the ruler's staff from amidst his standards." 4. Corrections or additions for the sake of conformity with parallel passages. Thus in Ex. iv. 18, the father- in-law of Moses is called in the Hebrew text " Jether ; " the Samaritan has " Jethro," which is his name else- where. In the genealogy of Gen. xi. 10, " and he died " is added after what is said of each patriarch as in ch. v. And whenever a partial list is given of the Canaanitish nations, the Samaritan copies insert the full enumera- tion as found in other passages. 5. Larger interpolations of sentences and even several successive verses from parallel passages. In numerous instances Exodus is thus interpolated from later pas- sages in the same book or from Deuteronomy, in order that when anything is referred to as having been said or done by Moses it may always be stated in identical terms, or when any command of God is repeated or obeyed by Moses it may be expressed with the same fulness of statement as when first given. 6. Corrections with a view of removing some sup- posed historical or other difficulty. Thus the four hundred and thirty years preceding the departure of Israel from Egypt, Ex. xii. 40, are made to cover the peregrinations in Canaan as well as the settlement in Egypt by changing the sentence, " Now the sojourning of the children of Israel which they sojourned in Egypt was four hundred and thirty years," so as to read "in THE SAMARITAN PENTATEUCH 137 the land of Canaan and in the land of Egypt." In this addition it agrees with the Septuagint. The most re- markable variation of this sort is in the genealogies of the antediluvian and postdiluvian patriarchs, found re- spectively in Gen. v. and xi. Here the Septuagint and the Samaritan differ both from the Hebrew and from each other ; and it is easy to discover that both have been altered from the Hebrew with different ends in view. The Septuagint has corrected the antediluvian line on the presumption that at that age of the world no one was less than one hundred and fifty years old at the birth of his first son ; and when anyone is stated to have been a father at an earlier age than this, the Sep- tuagint corrects it by adding one hundred years to the term before the birth of the son and subtracting as many from the subsequent years of his life, so that his entire age remains the same. On the contrary, the Samaritan assumes that no one would be more than one hundred and fifty at the birth of his first son ; when, therefore, this term was exceeded, as in the case of Jared, Methuselah, and Lamech, one hundred years or more are taken from it. Then, if the remaining years were left unchanged they would seem to have survived the flood ; accordingly, these are so altered as to make them all die in the very year of the flood. In the case of the postdiluvian patriarchs, Gen. xi., the Septuagint adds one hundred years whenever anyone is said to have had a son before his fiftieth year, except in the case of Nahor, the last instance of the sort, whe.re fifty years are added ; and the remaining years of their lives are so altered that no father may outlive his son, and that there may in each successive generation be a diminution in the term of human life. Under the same class Gesenius also puts alterations like that in Ex. xxiv. 10, where it is said of the seventy 138 GENERAL INTRODOCTION elders, that " they saw the God of Israel ; " here the Samaritan by the insertion of X changes "lTrVH, " and they saw," into WWI, " and they clave to the God of Israel," that it might not seem to contradict those pas- sages like Ex. xxxiii. 20, which declare that no man can see God and live. The Septuagint evades the same difficulty in another manner by reading " they saw the place where the God of Israel stood." 7. Samaritanisms in words, constructions, inflections, or orthography. Samaritan copyists might very easily slide into their native forms, terms, and idioms in the course of transcription. 8. Alterations for the sake of conforming to Samari- tan ideas. This includes the removal of anthropomor- phisms and anthropopathies and the use of euphemisms. Jacob's reproof of Simeon and Levi in his last words to his sons, Gen. xlix. 7, is converted into a commenda- tion by changing Ills, " cursed be their anger," into "P"Iit, " noble was their anger." The principal passage of this kind is Deut. xxvii. 4, where Moses directs that stones should be set up on Mount Ebal, and the words of the law be written upon them. The Samaritan has here changed " Ebal " to " Gerizim," where their temple was built ; and the passage so changed has been twice inter- polated elsewhere, viz., after Ex. xx. 17, Deut. v. 21. There are only four passages in which Gesenius pre- fers the Samaritan to the Jewish reading, viz., the in- sertion of " let us go into the field " in Gen. iv. 6 ; " numbered " for " led forth " his trained men, Gen. xiv. 14 ; " one ram " for " behind him a ram," Gen. xxii. 13 ; " a bony ass " for " an ass of bone," Gen. xlix. 14. The general meaning would not be affected in even a single instance if these changes were accepted; but other critics of note do not favor them. The coincidence between the Samaritan and the Sep- THE SAMARITAN PENTATEUCH 139 tuagint in so many characteristic readings cannot be casual, and has been variously explained. Some have held that the Septuagint was made from a Samaritan codex ; others that the Samaritan has been altered into conformity with the Septuagint, or vice versa; others still that both alike were made from a recension of the Hebrew differing from the Massoretic text. Gesenius states the facts of the case as follows : 1. The agreement is largely in those readings which smack of a gloss added to the text or a conjectural emendation of difficult places. 2. They agree in trivial things which do not affect the sense, and in the transposition of words or letters, and in arbitrary permutations. Thus the copulative conjunc- tion is inserted two hundred times in the Samaritan, and omitted in about half that number ; and with few exceptions the Septuagint does the same. 3. In added glosses the Samaritan goes far beyond the Septuagint, which agrees with the Hebrew against the Samaritan almost as often as it coincides with the latter. The Septuagint nowhere follows the Samaritan in its larger interpolations, nor in Samaritanisms. 4. In smoothing difficult passages the Septuagint sometimes adopts one conjecture, and the Samaritan another, as was shown in their modes of dealing with the ages of the patriarchs in Gen. v. and xi. 5. The Septuagint sometimes differs from the Hebrew where the Samaritan does not. This mostly concerns the permutation or transposition of letters, or more frequently still in supplementing passages from their parallels. Gesenius very properly rejects the notion that the Septuagint was made from a Samaritan original, since the Alexandrian Jews would not have accepted a manu- script of the law from those whom they hated so cor- 140 GENERAL IHTRODTJCTIOlir dially. He thinks that the most satisfactory hypothe- sis, and the one that is freest from difficulties, is that the Samaritan and the Septuagint were both made from a recension of the Hebrew differing from the Mas- soretic text. He supposes that there was a recension con- taining glosses and conjectural emendations, individual copies of which, agreeing in most but not all of their read- ings, were in circulation both in Alexandria and among the Samaritans. And that there was another recension which scrupulously sought to preserve the primitive read- ing even in places where it was difficult or obscure ; and that this, though not absolutely faultless, was in vogue among the Jews, particularly in Jerusalem. This would account for the agreement of the Septuagint and Sa- maritan in so many trivial matters, while their differ- ences could be explained by the various readings in different codices and by the freedom used by translators and transcribers, and yet more by the fact that the Sa- maritans continue to reform the text in the various ways spoken of above. On the other hand, Grotius and Archbishop Ussher were of the opinion that the Samaritan has been con- formed to the Septuagint. And when we consider the de- pendence of the Samaritans upon the Jews, from whom they borrowed their law, their religious rites, their modes of interpretation, many of their doctrines, and their legends ; and when we remember the veneration with which the Septuagint was regarded, it can scarcely be doubted that their Pentateuch was modified under Alexandrian influence. The occasional agreement of the Syriac Peshito or Jerome with the Samaritan can be similarly explained. The Syriac is known to have been corrected in numerous instances into conformity with the Septuagint ; and Jerome states that he frequently retained the rendering of the Septuagint where it THE SAMARITAN PENTATEUCH 141 differed from the Hebrew. It is very rarely the case that the Targums, which adhere strictly to the Massoret- ic text, agree with the Samaritan in its variations from it. When they do, this does not imply that they had a Hebrew text differing from the present ; they only ex- press in their rendering the same traditional gloss which the Samaritan puts in the text. There are two versions made from the Samaritan Pen- tateuch. One is in the Samaritan language, and is thought by Winer ' to date from the second century a.d. It appears to be the source of the quotations from the " Samaritan " in the writings of the Fathers of the third and fourth centuries, with which it almost uniformly corresponds. The other is an Arabic version attributed to Abu Said of the eleventh or twelfth century. 1 De Versionis Pentateuchi Samaritanas Indole. vn THE HISTOEY OF THE TEXT "We have now considered the primitive form of the Old Testament Scriptures, or the languages in which they were originally written ; the mode of their preser- vation by manuscripts ; and the alternate forms in which they have been perpetuated, or the ancient ver- sions in different languages. We now proceed to the contents of the Old Testament, and shall first consider the history of the sacred text. By the text is techni- cally meant the precise words of the inspired writers. The momentous question here arises, Is there good reason to believe that they have been faithfully trans- mitted to us? Through what vicissitudes have they passed in the long ages that have elapsed since their first appearance ? Have the requisite pains been taken in their preservation to protect them from wilful mutila- tion or negligent transcription ? and with what result ? The history of the text of the Old Testament may be* most conveniently divided into four periods ; the first extending to the cessation of inspiration and the collec- tion of the canon under Ezra and Nehemiah ; second the period of the Scribes ; third that of the Massorites ; and fourth the post-massoretic period reaching to the present time. Of the first period but little is known. We have only some incidental hints and a few facts from which con- clusions can be drawn as to the course of things. Men- tion is made of letters carved on solid materials. The 143 THE HISTORY OF THE TEXT 143 ten commandments were engraved upon tables of stone, Ex. xxxi. 18, xxxii. 15, 16, xxxiv. 1. The precious stones of the high-priest's breastplate had graven upon them the names of the twelve tribes of Israel, Ex. xxxix. 14, and a plate of gold attached to his mitre bore the inscription "Holiness to Jehovah," ver. 30. A copy of the law was written by Joshua on plastered stones set up on Mount Ebal, Josh. viii. 32, agreeably to the direc- tion given by Moses, Deut. xxvii. 2-4. Job wishes that his attestation of his innocence might be graven in the rock with an iron pen, and filled in with lead, that it might endure forever, Job xix. 24. When Jeremiah would describe the conspicuous and indelible character of the sin of Judah, he says that it is written with a pen of iron and the point of a diamond, Jer. xviii. 1. The primary signification of the Hebrew words aro to write, and pn statute, imply that hard materials were first used for writing and for recording laws But apart from monumental inscriptions and signet engravings, there is no indication that materials of this description were in common use in Old Testament times. Books Were used for writing whether for some imme- diate purpose, Ex. xxiv. 7 ; Num. v. 23 ; Josh, xviii. 9, or for permanent preservation, Ex. xvii. 14 ; Deut. xxxi. 24, 26 ; Jer. xxxii. 10, 12, 14. The original word (iSD) means something scraped or smoothed, and prob- ably indicates that they were made of skin or leather, as among the early Greeks and other ancient nations. The same thing is implied, Num. v. 23, where the writ- ing was washed off with water without injuring the ma- terial. Perhaps linen or paper from the bark of trees, or the papyrus may also have been employed. The book was usually in the form of a roll, Ps. xl. 7 ; Jer. xxxvi. 2, 23 ; Ezek. ii. 9 ; Zech. v. 2, which when folded together was fastened by a seal, Dan. xii. 4. It was 144 GENERAL INTRODUCTION written in columns, Jer. xxxvi. 23, with ink, Jer. xxxvi. 18, Ezek. ix. 2, and a pen, Judg. v. 14 (?) ; Ps. xlv. 1 ; Isa. viii. 1, probably of reed cut into the proper shape, Jer. xxxvi. 23. The sacred books were liable to the same casualties which have befallen all the literary products of an- tiquity. More or less errors were inevitable in the course of repeated transcription through long periods of time. A standard copy of the Books of Moses was preserved in the temple, with which other copies could be compared and corrected, and thus guarded from er- ror. The originals of some other books may have lasted for a considerable time, and so have been available for this purpose ; but of this we have no definite informa- tion. The veneration with which the sacred writings were regarded as the product of inspiration, and in- vested with divine authority, has effectually operated in preserving them from destruction, while all other writ- ings belonging to this period have been suffered to perish; and it doubtless led to special care in their transcription, though it is probable that the excessive scrupulosity of later times was not brought into requi- sition until actual experience of the existence of diver- gent copies had demonstrated its necessity. In the inscription of king Mesha (who is spoken of in 2 Kings iv. 4, 5) and that of Siloah (probably dating from the reign of Hezekiah, 2 Kings xx. 20), and in some Phoenician inscriptions as well as in the Samaritan Pentateuch, each word is separated from that next to it by a dot. This makes it not improbable that in the early Hebrew writing the words were not run together, but distinguished either by a dot or by spacing. The Talmudic rules for copying manuscripts required that a space equal to the width of a letter should be left after each word. It is not necessary to suppose that this was THE HISTORY OF THE TEXT 145 an innovation ; it is quite as likely that it was simply an adherence to ancient custom. The few instances in which words are improperly divided, or the Septuagint divides differently from the Massoretic text, may easily be accounted for on the same principle as other errors. The old form of the Hebrew letter was in use through- out this period, and words were written without the vowels, except as these were scantily supplied by the vowel letters. A comparison of duplicate passages shows the exist- ence of occasional errors, particularly in unfamiliar proper names, as Dodanim Gen. x. 4, but Rodanim 1 Chron. i. 7 ; Eiphath Gen. x. 3, Diphath 1 Chron. i. 6 ; Hadar Gen. xxv. 15, Hadad 1 Chron. i. 30; Aram 2 Sam. viii. 13, Edom 1 Chron. xviii. 16 ; or numbers, as seven hundred 2 Sam. viii. 4, but seven thousand 1 Chron. xviii. 4 ; seven 2 Sam. xxiv. 13, three 1 Chron. xxi. 12 ; forty thousand 1 Kings iv. 26, four thousand 2 Chron. ix. 25 ; twenty-two 2 Kings viii. 26, forty-two 2 Chron. xxii. 2. Josh. xv. 32 and elsewhere numbers are given at the end of lists of cities which are not equal to the particulars contained in them. These may, how- ever, be explained otherwise than as errors of transcrip- tion. Villages may be included in the lists which are not counted as cities in the enumeration ; or cities which subsequently grew up in the districts described, may have been inserted to complete the lists without a corresponding change of the numbers. The differences occurring in the duplicate Psalms, such as Ps. xviii. compared with 2 Sam. xxii., may be in part attributable to the mistakes of copyists, but in the main they are better explained as the result of a revision by the author himself or by others, or as Ps. xiv. and liii., an adapta- tion to another occasion. The inference sometimes drawn from such passages of a lack of care in transcrib- 10 146 GENERAL INTRODUCTION ing the sacred books during this period is wholly un- warranted. Stade i has the conceit that the collection of the Canon was accompanied by a wholesale falsification of Israel- ifcish history and religious life ; that the books excluded from the Canon gave an entirely different version of affairs from those which were received ; that those which were admitted to the Canon were carefully re- vised in order to bring them into harmony with the views of the collectors; everything opposed to the ideas then prevalent was expunged, whatever seemed wanting to their full and adequate expression was in- serted ; and thus they were made to represent a stage of religious development remote from that in which they were actually written, and to express ideas foreign to those contained in them in their original and genuine form. All this is spun out of his own brain. It is ab- solutely baseless ; and is' simply a conclusion drawn from a critical hypothesis at variance with the facts of the Old Testament, and which requires to be bolstered up by a thoroughgoing perversion of those facts. There is no reason to suspect that any wilful changes were made in the text of the Scriptures by the collectors of the Canon for any cause whatever. The second period in the history of the text extends from the collection of the Canon under Ezra and Nehe- miah to the completion of the Talmud in the fifth century a.d. With Ezra began the race of scribes who were devoted to the study of the Scriptures and were the custodians of the sacred text. It was their func- tion, as they understood and expressed it, " to put a hedge about the law," i.e., to ascertain, defend, and per- petuate the true interpretation of Scripture, and to pre- serve it from any possible error in transmission. With 1 Geschichte dea Volkes Israel, I., p. 14 ff. THE HISTORY OF THE TEXT 147 this view they began the formation of that body of critical observations upon the text known as the Massora (tradition), which was continued and enlarged by their successors in the following period. They even counted the number of letters, words, verses, and sections in each book, and noted the middle letter and word, which were marked in some cases at least by a letter of un- usual size. Other letters above or under size, or out of the ordinary form or position, were used to call attention to some point of interpretation or hidden sense or usage of words or mode of writing them or some other matter which teachers desired to inculcate upon their pupils, but whose meaning is now unknown. All this is spoken of in the Talmud as ancient and the work of the early scribes. The minute directions in the post-talmudic tract Sopherim to be observed in copying the sacred books were now formulated at least in part, and show the rigid supervision exercised and the extreme care used to guard against the intrusion of errors as far as that was possible. In this period occurred the change from the old to the more modern form of the Hebrew letter, not by the sudden introduction of a new character from abroad, but by gradual modification largely induced by the effort after regularity and symmetry of form and an elegant calligraphy befitting the sacred task in which the copyists were engaged. There is no reason to believe that this gradual alteration in the shape of the letters had any effect whatever upon the substance of the text. The vowels were not indicated by written signs, but the pronunciation was fixed by a steadfast tradition, even in the case of words whose written form is am- biguous. This is shown by explicit statements of the Talmud and Jerome. The different pronunciation rep- resented in the Septuagint and the Hexapla of Origen 148 GENERAL INTRODTJCTIOH is that of Egyptian Jews, as distinguished from that of Palestine. The definiteness with which the text was established is shown by the existence of K'ris, of which the Talmud makes mention. It had become usual in reading the Scriptures to substitute Adhonai (Lord) for the divine name Jehovah, which was regarded with superstitious awe ; to employ customary forms and constructions for those which were unusual, and euphemistic expressions for those which seemed indelicate ; and to omit certain words that were deemed superfluous, and introduce others in places where they appeared to be lacking. This was done, however, without any change in the written text, which was considered fixed and unalterable. The K'ri (that which is read) was distinguished from the K'thibh (that which is written) ; the latter remained in the text, notwithstanding the fact that traditional usage had given the preference in reading to the former, which was for the present only preserved orally, and was at a later time noted in the margin. The division into verses was already ancient in the time of the Talmud. It was then marked by the dou- ble point Soph Pasuk ( : ), which was long anterior to the written accents by which clauses and sentences were subsequently indicated. Possibly a space may have been left at the end of a sentence from the first in He- brew writing ; and this may have been the origin of the verses. At any rate exegetical study and the public reading of the Scriptures would early suggest the need of such divisions and establish some uniformity in them. In the poetical writings the parallelisms would of themselves determine the clauses. And direction is given in the Talmud that in Ex. xv., Deut. xxxii., Judg. v., and 2 Sam. xxii. each clause should constitute a line. Synagogue manuscripts as a rule are without the divis- THE HISTORY OB" THE TEXT 149 ion into verses. The Samaritan Pentateuch marks the sentences, but not the Massoretic verses ; and the early versions differ considerably from the Hebrew text in the division of sentences. The Talmudic enumeration of verses is not quite identical with the number in the Massoretic text, but does not vary materially from it. They are spoken of as known to Ezra, and even referred to a tradition from Moses at Mt. Sinai, which implies that they originated in an unknown antiquity. The division into sections was also pre-talmudic. They are mentioned in the Mishna, and often in the Gremara, where they are referred to a tradition from Moses. Direction is there given in copying the law to preserve the Parashas; accordingly, they are found in the Synagogue rolls. The number of these sections in the Pentateuch is 669. They are of two kinds, respec- tively denominated ninins open, and niaino closed. After the principal sections, which indicated the main divisions of the subject, the rest of the line was left open, the following section beginning a new line. Af- ter the minor sections or subdivisions of the matter a moderate space was left, which was closed by the follow- ing section commencing in the same line. In printed Bibles the rule respecting open and closed lines is not observed, instead of which these sections are designated in the Pentateuch by the initials B or D respectively. Similar sections in the other books are simply indicated by spaces without the letters B or D. These sections are for the most part appropriate and evidence a cor- rect understanding of the text on the part of those by whom the division was made. They are quite distinct both from the capitula of Jerome and the sections (pip) in the Samaritan Pentateuch ; in the latter Genesis is divided into 250 sections, and the entire Pentateuch into 966. 150 GENERAL INTRODUCTION When and how did the text become fixed and unal- terable as we find it in the time of the Talmud ? Here we are without definite information. Did it come down as a determinate text from the time of the collection of the Canon ? or was it settled by a gradual process through the agency of the scribes ? We have no posi- tive means of knowing ; and opinions are formed largely by the preconceptions with which the question is ap- proached. The only facts of importance bearing upon the case are that the Septuagint and the Samaritan Pentateuch depart considerably from the Massoretic text, while the later Greek versions, the Targums, the Tal- mud, and Jerome, adhere closely to it with only minor variations. Does this mean that the former authorities were inaccurate and the latter more rigorously exact, the Hebrew text meanwhile remaining substantially stationary ? Or that these various authorities faithfully represented the Hebrew text at the time to which they severally belong, and that the latter underwent through the labors of the scribes that measure of transformation which the differences between the former would indi- cate ? The real significance of the facts above recited, and the inferences properly deducible from them, will come up for consideration in discussing the criticism of the text. Of course it may be presumed that no amount of care could prevent the occurrence of occasional errors in the course of frequent transcription. That the scribes were on the alert to correct such errors as far as possible by comparison with other trustworthy copies is certified by a statement in the Talmud that three manuscripts of the law were collated on a particular occasion, and the tes- timony of two against one was accepted as decisive of the true reading. In our ignorance of the number and character of the manuscripts available for a particular THE HISTORY OF THE TEXT 151 book at any given time, and the care with which colla- tions were made, and the critical skill displayed in dis- tinguishing the true from the false, we are not in a situation to revise their work or pass judgment upon it. But it is natural to suppose that where so much pains was taken to insure correctness by men devoted to the study of Scripture and possessed of an almost incredible familiarity with its letter in its minutest details, the text, though not free from minor defects, would on the whole be safely guarded. The Jews were charged by the early Christians with wilfully altering the text of Scripture to the prejudice of Christianity. But it is universally admitted that this was a mistake, arising from the fact that in their contro- versies the Jews refused to acknowledge the authority of the Septuagint, and appealed from it to the Hebrew. Certain critical decisions are attributed to the scribes, which have led some to apprehend that they meddled improperly with the text, and even made changes on theological grounds. Thus the Talmud speaks of five instances of removal by the scribes (D'lBID "I1B5). This, however, was simply a declaration on their part that a conjunction l and, which was not in the text, and did not belong there, should not be inserted in reading. Gen. xviii. 5, xxiv. 55 ; Num. xxxi. 2 ; Ps. xxxvi. 7 (A. V., 6), lxviii. 26 (A. V., 25). In the Massora mention is made of eighteen instances of correction by the scribes (D"nsio 1*lBPi). According to Buxtorf they are passages in which one might sup- pose from the connection that the Writers meant to ex- press themselves differently from the way in which they actually did; but in which the scribes adhere to the correct reading. 1 ■The passages in question are Gen xviii. 22; Num. xi. 15, xii. 12; 1 Sam. iii. 13 ; 2 Sam. xvi. 12, xx. 1 ; 1 Kings xii. 16 ; 2 Chron. x. 16 ; 152 GENERAL INTRODUCTION The extraordinary points over certain words or letters are supposed to suggest a doubt as to their genuineness ; though it seems that this was not sufficient to lead to their erasure. The Talmuds represent, not the critical but the her- meneutical side of the function of the scribes of this period. Their laboriously minute interpretations of the law in its application to every conceivable case, which were elaborated generation after generation, grew into a vast body of jurisprudence. This was at first preserved orally, but at length swelled to such dimensions as to overtask the most retentive memory, so that it became necessary to commit it to writing, if it was to be perpetuated and enforced. The necessity was made more urgent by the destruction of Jerusalem and the final overthrow of the Jewish state. Hence the Mishna of E. Judah ha-Kadosh in the second cen- tury a.d., and the Jerusalem and Babylonish Gemaras or comments of later Rabbis in the fourth and fifth centuries. The third period is that of the Massorites in the more restricted sense, and extends from the sixth to the eleventh century. The same necessity which produced the Talmud, now led to recording the critical material, which had hitherto been perpetuated only by oral in- struction. One most important addition to the text was the introduction of written signs to aid in its pronun- Jer. ii. 11 ; Ezek. viii 17 ; Hoa. iv. 7 ; Hab. i. 12 ; Zech. ii. 12 ; Mai. i. 13 ; Pa cvi. 20 ; Job vii. 20, xxxii. 3 ; Lam. iii. 20. Aa apecimena it ia said that in Gen. xviii 22 they changed "The Lord atood yet be- fore Abraham " to " Abraham atood yet before the Lord ; " 2 Sam, xx. 1, "Everyman to hia goda " ("PTOH) to "Every man to his tents" (l^WlS) ; Hos. iv. 7, "They have changed my glory into shame " to " I will change their glory into shame.'' All which looks like frivolous punning upon the text by ingenious alterations of its meaning, and casta no suspicion upon the correctness of the received text. THE HISTORY OF THE TEXT 153 ciation. As the knowledge of Hebrew was diminishing by the lapse of time, and the schools were waning, it was of prime consequence that the imperfections of a purely consonantal alphabet should be relieved, and its ambiguity in respect to the pronunciation and the meaning should be effectually removed. It had an- swered fairly well while Hebrew was a living language, and the reader familiar with his native tongue could mentally supply what was defective in the notation. But this could no longer be counted upon. Accordingly, while the consonantal text was left intact, by means of diacritical points, vowel signs, and accents the exact sounds of the words were represented, the signification of those which were previously doubtful was determined, and the limits of clauses and the mutual relation of the words composing them was indicated. And this not in an arbitrary manner, but in accord with a steadfast and reliable tradition. It is obvious that the minuteness and complexity of this system of notation greatly increased the liability to error in transcription. Hence, renewed pains were taken to guard against it as effectually as possible by extend- ing the critical observations of the Massora, which had previously concerned merely the consonants, to this new system of points. The number of times that particular words occur, or that they are written in a particular way, or that unusual or anomalous forms are found, and where, and much more of the like sort are noted with the utmost care. These observations are so extended and precise that it is possible by means of them to re- construct in a large measure the exact text upon which they were based. These critical notes were at first written on the margin (the marginal Massora) or at the close of manuscripts (the terminal Massora) ; but as they increased to an enormous extent, they were subse- 154 GENERAL IHTBODTJCTION' quently written in separate -volumes. The entire collec- tion is called the great Massora ; the little Massora is an abridgment of it. A table of various readings affecting the vowels ex- clusively, prepared early in the eleventh century, notes the differences between the standard authorities ben Asher of Palestine and ben Naphtali of Babylon. An earlier list of differences in the consonantal text between the eastern (Babylonish) and the western (Palestinian) is printed at the end of the second edition of Bomberg's Rabbinical Bible. The Massoretic verses are substantially the same as those noted in the Talmud; the differences between them may be accidental or may arise from the correction of what was esteemed an improper division. 1 Two different modes of reading the Pentateuch were observed in the Synagogues of different localities. One, in which it was completed in one year, led to its di- vision into fifty-four Parashas, 2 which are marked in manuscripts and printed editions with three B's or D's of large size according as they begin with an open or a closed section. In one exceptional instance, Gen. xlvii. 28, a Parasha begins in the midst of a section. Corre- spondent with this division of the law are the lessons selected from the prophets called Haphtaras. The Jewish story respecting their origin, which is not very credible, is that Antiochus Epiphanes having prohibited the reading of the law in the Synagogues, an equal num- ber of sections from the prophets was substituted in their place ; which, when the persecution ceased, were 1 The Talmud reckon8 5,888 verses in the Pentateuch, with Lev. xiii. 3 as the middle verse; the Massora has 5,845 verses and Lev. viii. 8 the middle verse. — Ginsburg, Introduction to the Hebrew Bible, p. 70. 2 These Parashas were finally fixed in their present authorized form in the fourteenth century. — Buhl, Kanon und Text, p. 227. THE HISTORY OE THE TEXT 155 still retained and read as they are at present in connec- tion wiih the law. The other mode of reading the law was to finish it in three years. This led to its division into 154 or 167 Sedarim, as they are variously num- bered ; x the number in the entire Bible is 452. The Talmud states that the annual method was customary in Babylon, and the triennial in Palestine. As the former became the general practice in later times the Sedarim are not commonly marked in manuscripts, and are not indicated in printed Bibles. The fourth and last period in the history of the text is the post-Massoretic. The main function of this period is the faithful transmission of the Massoretic text with its accessories. The division into chapters is not of Jewish, but of Christian, origin. It was first in- troduced into the Latin Bible in the thirteenth century, and is attributed to Stephen Langton. It was used to facilitate reference in the concordances to the Vulgate ; and was adopted for the same purpose by Isaac Nathan, about 1440, in his Hebrew concordance. R. Solomon ben Ismael, about a.d. 1330, was the first to note the numbers of the chapters in the margin of the Hebrew Bible. 2 The first printed edition of the Hebrew Bible in which they were thus noted was the Complutensian Polyglot, in 1517. The edition of Arias Montanus, in 1571, was the first in which the Hebrew text was broken up into chapters, and the Hebrew numerals placed in the body of the text itself. The enumeration of the verses was first introduced in the Hebrew Bible in Bomberg's edition of 1547, in which the number was noted in the margin oppo- site every fifth verse by the appropriate Hebrew let- ters numerically used. Arias ' Montanus, in the Ant- werp Polyglot, 1571, attached Arabic numerals in the 1 Ginsburg, pp. 32-65. * Ibid., p. 25. 156 GENERAL INTRODUCTION margin to every verse throughout the whole Hebrew Bible. 1 The first portion of the Hebrew Bible ever printed was the Psalms with the commentary of Kimchi, at Bologna, in 1477. The Hebrew Bible was first printed entire, at Soncino in the Duchy of Milan in 1488. Only nine copies of it are now known to be in existence in Europe. A second edition, undated, is supposed to have been printed at Naples somewhere between 1491 and 1493. The third complete edition printed at Bres- cia in 1494 is interesting from the circumstance that Luther made use of it in translating the Bible into German. The edition of Athias, in 1661, was based on very old manuscripts, and has been generally followed in subsequent editions. A Rabbinical Bible is one which in addition to the original Hebrew contains the Targums, the Massoras, and commentaries of the Rabbis. That of Daniel Bomberg was published in Venice ; three successive editions were issued, the first in 1517 ; the second in 1525, which is particularly famous as containing the Massora collected by Jacob ben Chayim and a text con- formed to the Massora. Buxtorf's Babbinical Bible was published at Basle in 1618. That of Amsterdam in 1724. By a critical edition of the Old Testament is meant one which in addition to the received text of the orig- inal contains a critical apparatus, or a collection of various readings gathered from manuscripts and ver- sions. The most noted critical editions of the Hebrew Bible are that of Houbigant published at Paris in 1753, and that of Kennicott, in two volumes, published at Ox- ford, the first in 1776, the second in 1780, and contain- 1 Ginsburg, p. 167. Prof. G. F. Moore in the Journal of Biblical Literature, toI. xii., p. 76. THE HISTORY OF THE TEXT 157 ing the readings from 694 manuscripts. De Eossi pub- lished a few years later at Parma, in 1784, the various readings obtained from several hundred more manu- scripts, besides early printed editions and ancient ver- sions. These were issued without the text. By a Polyglot is technically meant an edition of the Scriptures exhibiting at one comparative view the originals and one or more ancient versions possessing critical authority. There are four principal Polyglots, the Complutensian, the Antwerp, the Parisian, and the London. The Complutensian Polyglot was so called from Complutum (Alcala in Spain), where it was pre- pared and published by Cardinal Francis Ximenes, Archbishop of Toledo, with the assistance of several learned men connected with the university of that place. He was allowed the use of several manuscripts from the Yatican Library by Pope Leo X. ; and others were purchased by him at a vast expense. The work is said to have cost him 50,000 ducats. It consists of six volumes, the first four of which are occupied by the Old Testament ; the Hebrew, Vulgate, and Septuagint being arranged in parallel columns, to which in the Pentateuch is added the Targum of Onkelos at the bottom of the page. The fifth volume contains the Greek New Testa- ment and the Vulgate in parallel columns, and the sixth among other things a Hebrew Grammar and Lexicon. The volume containing the New Testament was the first printed in 1514; the last was printed in 1517, but in consequence of the death of Ximenes the work was not published until 1522. It is now exceedingly rare, as only six hundred copies were printed ; three copies were struck off on vellum. The story was for some time in circulation that the manuscripts upon which this publication was based had been sold by an illiterate li- brarian to a rocket-maker as useless parchments ; this is 158 GENERAL INTRODUCTION now known to be a fabrication, since the manuscripts belonging to Cardinal Ximenes, and which were pre- served in the library at Alcala, are in the library of the University at Madrid. Nevertheless, the Hebrew manuscripts and the printed editions, from which the text of the Complutensian was drawn, have not been definitely identified. The Antwerp, or as it is also called, the Royal Poly- glot (Biblia Regia), was printed at Antwerp in eight volumes, folio, under the patronage of Philip II. of Spain, in 1569-1572. Its text was based on the Complutensian and the Bible of Bomberg, and was followed in the Pa- risian and London Polyglots. It contained, in addition to what was to be found in the Complutensian, the Targum of Jonathan on the Prophets and a Targum on the Hagiographa. The last three volumes contain grammars, lexicons, tables, and treatises on various sub- jects, together with a Latin version of the whole Bible. Only five hundred copies of it were printed, and a large number of these were lost in a voyage to Spain. The Parisian Polyglot was published at Paris in 1645 in ten volumes, folio, at a vast expense, and the publisher was ruined by the undertaking. It contained, in addition to all that was in the Biblia Regia, the Samaritan Pen- tateuch with the Samaritan version, the Peshito of the Old Testament, and an Arabic version. The London Polyglot, by Bishop Brian Walton, pub- lished in 1656 in six volumes, folio, made several additions to those which had preceded it. The first three volumes contain the Old Testament, exhibiting in separate col- umns the Hebrew, Vulgate, Septuagint, old Latin Itala as restored from the extracts in the Fathers, the Pesh- ito, Targums, and an Arabic version ; to which are added, in their proper place, the Samaritan Pentateuch and version, and an Ethiopic version of the Psalms and THE HISTORY OF THE TEXT 159 Canticles. The Targum on Chronicles was not dis- covered until the work was in the press, and conse- quently does not appear in it. The fourth volume ex- hibits the whole of the Apocrypha in Greek, Latin, Syriac, and Arabic, together with two separate editions of the Book of Tobit in Hebrew. Then follows the Pentateuch according to the Targum of the Pseudo- Jonathan, the Jerusalem Targum, and the Persian ver- sion of Tawos. The New Testament occupies the fifth volume. The sixth contains various readings and criti- cal remarks; and prefixed to the whole in the first volume are learned and valuable prolegomena. Ed- mund Castell published, as an appendix to this work, his Heptaglot Lexicon in 1669, in two volumes, upon which he spent seventeen years and the whole of his fortune. " This work was published by subscription under the patronage of Oliver Cromwell, who permitted the paper to be imported free of duty. But the Pro- tector dying before it was finished, Bishop Walton can- celled two leaves of the preface, in which he had made honorable mention of his patron, and others were printed containing compliments to Charles II. and some pretty severe invectives against republicans." i 1 Home's Introduction, Bibliographical Appendix. vni THE CRITICISM OF THE TEXT The word criticism is derived from the Greek Kplvm to judge, and denotes an act of judging, or in its techni- cal sense the art of judging. There are two principal branches of Biblical criticism, as of literary criticism generally, which are respectively denominated textual criticism and higher criticism. 1 Higher criticism is oc- cupied with the questions of the genuineness, integrity, and trustworthiness of the books of the Bible. It in- quires whether they were written by their reputed authors, whether they are complete and unadulterated in all their parts, free from mutilations, alterations, or interpolations, and whether they are a reliable and truthful representation of the mind of their inspired authors, and of the times and circumstances under which they purport to have been written. Its office is to ascertain the truth in regard to these various mat- ters ; if false views have been entertained, to refute and dispel them ; if these writings have suffered any ma 1 terial injury, to detect and correct it, discriminating the genuine from the spurious, and the original from what has been subsequently added. This work is to be per- formed not arbitrarily nor capriciously at the mere pleasure of the critic, but soberly and cautiously after 1 Various other terms more or less descriptive and appropriate hare been used to designate these two kinds of criticism, such as book criti- cism and word criticism, or internal and external, or rational and me- chanical, or a priori criticism and a posteriori criticism. 160 THE CRITICISM OF THE TEXT 161 carefully examining and duly estimating all the facts of the case and all the considerations bearing upon it. Literary forgeries in ancient and in modern times have been detected by appropriate tests. A noted in- stance in classical criticism is found in the so-called Epistles of Phalaris, tyrant of Agrigentum in Sicily, which long enjoyed a high reputation, and were re- garded both by ancient and modern authorities as the genuine productions of their reputed author. But in his celebrated controversy with Boyle, who had pub- lished an edition of these epistles in 1695 and was backed by all the learning of Oxford University, Bentley utterly demolished his antagonist and his cause, prov- ing incontestably that these epistles were the fabrication of some sophist belonging to a much later period. This was shown by the mention of the names of cities which were not built until long after the time of Phalaris, al- lusions to tragedies and comedies as things well known and of ordinary occurrence, the introduction of senti- ments and expressions manifestly derived from later writers, and by the dialect of the epistles themselves, which is the later Attic, such as was the language of the learned in the latter ages of the Boman empire. 1 In like manner it can be shown that the book en- titled the "Wisdom of Solomon could not possibly have been written by Solomon himself; and that the ad- ditions to Esther and Daniel did not belong to these books in their original form. There is no real objec- tion to the just and impartial application of the higher criticism to the canonical books of Scripture. Criticism legitimately employed, so far from betraying a want of reverence for Holy Scripture, is the offspring of a sacred regard for the word of God, which cherishes that word too highly to suffer anything that is purely human to 1 Smith's Dictionary of Greek and Roman Biography, Art. Phalaris. 11 162 GENERAL INTRODUCTION remain mingled with it. But a style of criticism which is warped by naturalistic prepossessions, to which every prophetic disclosure of the future is an anachronism, and every miracle is a legendary exaggeration, and rev- elations of truth must be pared down to fit in with some scheme of progressive natural development, is in its principles and results antagonistic to the Bible, and necessarily leads to false conclusions corresponding to the false principles on which it is based. Such a method of treatment must as a matter of course issue in a denial of the genuineness of many of the books of Scripture. And the literary grounds which are mar- shalled in support of conclusions thus reached, do not alter, even though they may partially conceal, the ani- mus of the whole proceeding. Nor does the fact that professedly evangelical men strangely enough are will- ing to accept conclusions wrought out by those whose principles they reject, relieve the vice inherent in the scheme itself. The other branch of criticism, and that with which we are now more immediately concerned, is textual criticism. Its function is to determine by a careful ex- amination of all the evidence bearing upon the case the condition of the sacred text, the measure of its corre- spondence with or divergence from the exact language of the inspired penmen, and by means of all available helps to remove the errors which may have gained ad- mission to it from whatever cause, and to restore the text to its pristine purity as it came from the hands of the original writers. Its office is not, as the excesses of some Biblical critics have led many to imagine, to sit in judgment upon the word of God, and to take from it or add to it at pleasure. It is not an arbitrary, but a judicial, process, based on fixed and intelligible prin- ciples, and conducted in a determinate manner, in which THE CRITICISM OF THE TEXT 163 all the evidence is diligently collected, thoroughly sifted, and accurately weighed, and the decision given in ac- cordance with the ascertained facts. The sources from which evidence may be derived that is available in textual criticism are either external, as manuscripts, quotations in early writers, and ancient versions ; or internal, considerations drawn from the text itself ; and when everything else fails, use must be made of critical conjecture. If manuscripts were not liable to errors in transcrip- tion, their testimony in every case would be final and conclusive. But errors may arise from accident or de- sign. The liability to accidental errors was immensely greater when every copy had to be written separately by the pen, than now when from a single form of types any number of copies exactly corresponding can be struck off. But with all the care that is taken in revis- ing the proof, few works issue from the press without more or less errata. A second transcription would not only perpetuate the errors previously made, but intro- duce new ones ; and thus they would go on increasing in arithmetical progression with every fresh copy that was made. These errors have for greater distinctness been further classified into those arising, 1. From the eye, such as confounding similar letters, transposing let- ters or words ; omitting letters, words, or sentences, es- pecially when two sentences end alike, and the second is in consequence mistaken for the first. 2. From the ear, where one reads and another writes, and letters or words of similar sound are mistaken for each other. 3. From memory, where a writer in undertaking to re- produce a clause or sentence omits or transposes a word, or substitutes a synonym, or conforms the sentence to some familiar parallel passage. 4. From defect of judg- ment in erroneously dividing words, misunderstanding 164 GENERAL INTRODUCTION abbreviations, mistaking letters inserted to fill a void, space at the end of a line for a separate word, or some marginal remark for a part of the text. Intentional alterations have been made in the text of manuscripts without any evil design, from a mistaken desire to correct imagined errors ; thus, an easier read- ing is substituted for one more difficult, supposed slips of the pen are corrected, apparent omissions supplied, sentences made more classical or elegant, or assimilated to parallel passages. As the probability is that errors multiply with each successive transcription, those manuscripts which stand nearest to the original autograph may be expected to have fewer errors than those which are at a greater remove from it. Other things being equal, the oldest manuscript is to be preferred. Another consideration affecting the value of a manuscript is the care with which it has been written. If there are no slips of the pen and no indications of negligence, it is fair to presume that it is in general an accurate copy of that from which it was made. Its worth may also be esti- mated by the general agreement of its text with that of other valuable codices ; its correctness in the main thus ascertained gives weight to its authority where it stands alone. In order to obtain a correct text of the Old Testa- ment many hundreds of manuscripts have been collated, and their various readings noted, particularly by Kenni- cott and De Eossi. The result is the discovery that they are throughout in substantial agreement. The various readings are for the most part of a trivial character, not materially affecting the sense. All Hebrew manuscripts from the time of the Massorites have been conformed to what is known as the Massoretic text, which has been regarded as the standard authority. All deviations from THE CRITICISM OF THE TEXT 165 it in existing manuscripts are due to unintentional er- rors of transcription, or in a few instances possibly to corrections by copyists on the basis of some ancient version. The genuine Massoretic text is so fenced about by the multitudinous critical notes of the Mas- sora that it can be reproduced with remarkable accuracy by the aid thus furnished. 1 As the stream cannot rise higher than the fountain, it is impossible by means of manuscripts to rise above this text so as to form an in- dependent estimate of it or to undertake its correction. This can only be done by the assistance of the other sources of criticism. The Massoretic text was not the creation of the Mas- sorites. They found an established text already in existence, and their labors were directed to protect it from any possible deterioration, and secure its faithful preservation and accurate transmission. How invio- lable the text, as they possessed it, was felt to be is ap- parent from the K'ris and K'thibhs. Where a different reading was preferred to that in the text, no change was made in the latter, but the preferable reading was put in the margin. A second source of criticism is found in quotations in early writers, or in remarks made by them from which a safe conclusion can be drawn as to the text existing in their day. The value of such quotations for critical purposes is mainly dependent upon the question whether they give the precise words of the passage cited, or whether in quoting from memory they are only careful to preserve the sense without regard to the ex- act language. Only a probable answer can be given to this question, which must be judged of by the circum- stances of each individual case. If stress is laid upon 1 The most successful attempt is that of Baer in his edition of the fleveral books of the Hebrew Bible. 166 GENERAL IMRODtTCTIOff the words, or the form of expression is important in the matter spoken of, or is set in contrast with the language of some version which varies from it, one may con- fidently conclude that the passage has been accurately quoted. In exhortations or practical discourses it is less likely that the writer has concerned himself about the precise words of any passage of Scripture referred to than in commentaries or controversial writings where greater exactness would be necessary. From the quotations in the Talmud, whether in the Gemara of the fifth century a.d., or the Mishna of the second, and in the frequent appeals by Jerome in the fourth century to " the Hebrew verity," as well as the testimony of Origen in the third century in his Hexapla and in his numerous commentaries, it is abun- dantly evident that there was at this early period a fixed and authoritative Hebrew text, identical in the main with the Massoretic text as we possess it at pres- ent, and this was interpreted and understood in accord- ance with the sense yielded by the Massoretic vowels. The existing text can thus be determinately traced not only through the manuscripts, but far beyond any ex- tant manuscripts to the Massorites, and beyond # them through the Talmud on the one hand and Jerome and Origen on the other, century by century, until we reach the second century of the Christian era, where we find it in sole and undoubted authority, and regarded as handed down in its purity from the time of Ezra and that of Moses himself. Such a conviction implies that the text as they knew it and had received it, was undis- puted and of long standing. The remaining source of criticism is the ancient ver- sions. They have a critical, an exegetical, and a herme- neutical value, which should be clearly distinguished. By their critical value is meant the aid which they THE CRITICISM OF THE TEXT 167 furnish in determining and restoring the true text of Scripture. By translating the version back again into the language from which it was made, the original text may be obtained which the translators had before them. Their exegetical value is the aid which they furnish in rendering difficult words and expressions. The herme- neutical value arises from their exhibiting the princi- ples, methods, and results of the style of interpretation adopted by the translators, which it is reasonable to infer was that of their contemporaries likewise. They thus reveal the state of the text and the current mode of interpreting it at the time when they were prepared, and render important service in the explanation of what is obscure and in determining the meaning of words and phrases of rare occurrence. Different versions are of unequal value in these respects. A version may be of great hermeneutical importance by shedding light upon the history of interpretation and yet be worthless critically or exegetically. These various uses are inde- pendent of each other, and result from different and to some extent opposite qualities of the versions in ques- tion. In order to have any critical value whatever a ver- sidh must be ancient, and it must be immediate. Only those versions of the Old Testament are held to be ancient in this technical sense which preceded the period of the Massorites. No version made since could be an independent witness to the text, for it could only represent more or less perfectly the text which we actually have before us. An immediate version is one that is made directly from the original. Those made from previously existing versions are called mediate. A mediate version may be useful in the criticism of the version from which it was derived and aid in restoring its primary text, but it is no direct witness to the original 168 GENERAL INTRODUCTION text and cannot be employed in its criticism or restora- tion. But whether a version is mediate or immediate does not affect its hermeneutical value. It must be im- mediate, but not necessarily ancient, to be useful exe- getically ; its value in this respect depends solely upon the knowledge and skill of the translator. Versions do not represent the original text as directly as manuscripts or quotations in early writers. The latter exhibit it in its proper form ; in versions the form has been changed by the transfer into a different language. Hence there is a double liability to error in their employment for critical purposes. The version may be defectively rendered and so represent the orig- inal inadequately ; and in reversing the translation in order to obtain the original from which it was made fresh errors may be committed. Great skill and caution are requisite to a proper use of versions in the criticism of the text. Every deviation of a version from the Massoretic text does not justify the assumption that the translator had a different text before him. From negli- gence or want of knowledge on the part of the trans- lator, the passage may be carelessly or blunderingly rendered. A conjectural sense may be given to words or phrases that were imperfectly understood ; though even the mistakes of a version may sometimes afford a clew to the text from which it must have been made that such mistakes should be possible. The translator may moreover have taken considerable liberty with the text with which he was dealing. Being more concerned to make the version useful to readers than to preserve the precise form of the original, he may give a free rather than a literal translation. He may aim to give the general sense as he understands it rather than to render it word for word. He may sim- plify passages that seem obscure, may omit what seems THE CRITICISM OF THE TEXT 169 redundant, or amplify where a fuller statement would be more perspicuous. He may resolve figurative expres- sions by substituting what they signify. He may avoid forms of speech peculiar to the original language, or which he thinks liable to be misunderstood. And not only in the substance of the text but in its arrangement changes may be made with the idea of introducing a more desirable order or improving the connection. In these and other ways a version may vary considerably from the text without implying that a different text lay before the translator. A version so prepared might be more valuable for the use of contemporaries and for exegetical purposes, but would yield comparatively little aid in criticism. For this latter purpose it cannot be too slavishly literal, or adhere too strictly to every word and particle, or follow too closely every idiomatic ex- pression, however foreign from the modes of thought and speech of those for whom the version was designed ; it might even attempt to reproduce the etymology and composition of words, however unintelligible this would make it, because then the critic can with greater facility and certainty determine the precise form of the original from which the translation was made. If, on the other hand, the version is a paraphrase rather than a translation, giving not an exact rendering of the original but the translator's understanding of it, with remarks inserted by way of explanation or illustra- tion, and interweaving his sentiments with the text, its hermeneutical value will be thereby increased, but it will be of little service in the way of criticism. It thus becomes a valuable authority in the history of opinion and of modes of interpretation, but valueless for the determination or restoration of the original text. Before any practical use can be made of a version in the criticism of the original, a careful inquiry must be 1?0 ftENERAL INTRODUCTION instituted into the condition of the version itself and the purity of its text. Its value as a critical aid depends upon the accuracy with which it represents that copy of the original from which it was made. But if the text of the version has itself been corrupted in the course of repeated transcription, this coincidence no longer exists. The manuscripts of versions are liable to the same sources of corruption as those of the original, and re- quire the same means of correction. Besides this they have another fruitful source of corruption peculiar to themselves, especially when there is more than one ver- sion in the same language, viz., the interpolation or cor- rection of one from another or from the original. This was very natural for transcribers, who were not so much concerned to preserve the primitive form of the transla- tion with exactness as to furnish it to the reader as much improved by such comparisons and alterations as possible. It was very unfortunate, however, for their critical value. What, now, is the testimony of the ancient versions re- specting the state of the Hebrew text at the time that they were prepared? The Jewish Targums, which cur- rently bear the names of Onkelos and Jonathan, and received their present form in the fourth century, though based on much older materials, presuppose the Masso- retic text with very slight variation. The same is true of the Greek versions of the second century, Aquila, Symmachus, and Theodotion, which were of Jewish or Ebionite (Jewish Christian) origin and were prepared with the distinct purpose of giving a more adequate rep- resentation of the Hebrew than was to be found in the Septuagint. The Latin version of Jerome in the fourth century corresponds in general with the Massoretic text ; and any variations from it are sufficiently explained by his own confession that he did occasionally depart from THE CRITICISM OF THE TEXT 171 the Hebrew in his translation, contrary to his own judg- ment, as a concession to the high esteem in which the Septuagint was held, and the clamor which was raised against any deviation from it. The Syriac Peshito, which may most probably be referred to the middle of the second century, is in accord for the most part with the Massoretic text. Where it agrees with the Septua- gint in deviating from it, as it does in a number of in- stances, the probability is that it has in these particulars been altered into correspondence with that version in consequence of the great repute in which it was uni- versally held among Christians. We are thus led by the witness of the versions to the same conclusion that was reached by means of the quotations in early writers, that so far back as the second century of the Christian era the Hebrew text was substantially identical with what is now known as the Massoretic. When we pass, however, to prechristian authorities, the Greek Septuagint of the third century B.C., and the Samaritan Pentateuch earlier still, we find a consider- able divergence from the Massoretic text. The ac- knowledged character of the Samaritan Pentateuch, as this has already been exhibited, deprives it of all weight as a critical authority where it differs from the received Hebrew text. The critical value of the Septuagint has been very variously estimated by scholars, some rating it far above the Massoretic text, others regarding it as entitled to no consideration whatever. The extreme ex- altation of the Septuagint was vehemently urged in the seventeenth century in the interest of a depreciation of the Hebrew text as exceedingly corrupt and altogether untrustworthy. After the contest had been opened by John Morinus affirming, and Simeon de Muis denying, the superiority of the Septuagint text, Cappellus en- tered the lists as its champion. He undertook to show, 172 GENERAL INTRODUCTION in his " Critica Sacra," as the result of his study for thirty- six years, that the readings of the Septuagint were in a multitude of instances to be preferred to those of the Hebrew. He sought to prove this by the discrepancies in parallel passages in the Old Testament, by citations in the New Testament and in the early Fathers, and by a comparison of other ancient versions. This work of Cappellus was regarded at the time as a dangerous attack upon the integrity and authority of the orig- inal Scriptures, and its publication was prevented for ten years after its preparation. At last his son, who had meanwhile gone over to the Roman Catholic Church, obtained through the influence of Morinus and others royal leave for its publication at Paris in 1659. In accounting for the divergence of the Septuagint from the present Hebrew text it has been held on the one hand that the Hebrew text of that period was in a very unsettled state ; that the Septuagint fairly repre- sents the manuscripts from which it was made ; and that the unified and established form of the Hebrew text, as it appears in the second century a.d., was the result of critical labors expended upon it in the meantime by the scribes, the effort to obtain a uniform and uni- versally authorized text being intensified and the issue accelerated by the destruction of Jerusalem and the overthrow of the Jewish state, which turned attention more earnestly than ever to the Scriptures as the sole surviving bond of union. This view of the case, it must be said, is largely conjectural. No record sur- vives in Jewish writings or in any others of the critical reconstruction of the text which is here assumed, and from which such important consequences are deduced. Origen and Jerome always attribute the deviations of the Septuagint from the Hebrew text, as they knew THE CRITICISM OF THE TEXT 173 it, to errors on the part of translators or transcribers ; and they never intimate a suspicion that the Hebrew had itself undergone a change. It is of course quite possible that there may have been very inaccurate and carelessly written copies of the Old Testament Scriptures, and that these may have gained considerable circulation particularly among Egyptian Jews and others outside of the Holy Land, and may have been used by the ■ Septuagint translators. But if this were so, it would not warrant the inference that there was no settled and authorized Hebrew text at the time, and no standard copies in which it was to be found, and from which it was transmitted. Rever- ence for the Scriptures and regard for the purity of the sacred text did not first originate after the fall of Je- rusalem. On the other hand, these divergences are as easily and more naturally explicable if attributed to the transla- tors than to copyists of the original. The same causes which lead to a modification of the text in transcription would be operative in a translation in an aggravated form. A freedom might be used in rendering the Script- ures into another language which would not be thought of in transcribing the original. A measure of discretion must be allowed in a translator for which a copyist has no occasion, and which would not be permissible in him. And in this first attempt at making a work of such mag- nitude intelligible to those of a different tongue, no such rigorous rendering could be expected as would be de- manded from a modern translator. The sacredness and authority of the original would not attach to an unin- spired version. Accordingly, accurate precision was not aimed at so much as conveying the general sense, and in this the translators allowed themselves a large meas- ure of liberty. When to this is added an imperfect 174 GENERAL INTRODUCTION knowledge of Hebrew, conjectural renderings or para- phrases of words and passages not understood, slips arising from want of care and the like, it is easy to un- derstand how the general correctness of the Septuagint might consist with very considerable deviations from the original text. Some critics still inordinately exalt the critical value of the Septuagint, and are disposed to make frequent changes in the Hebrew text on its authority. But there is a general agreement among careful scholars that, while this version is to be highly esteemed for its an- tiquity, and the general testimony which it renders to the integrity of the existing text, and the aid which it furnishes in the rendering of obscure and doubtful pas- sages, the Massoretic text is on the whole vastly superior to it, and should not be corrected by it, except where there are stringent reasons for so doing ; and that in the great majority of cases where a divergence exists, the presumption is strongly in favor of the correctness of the Hebrew and against the Septuagint. Neither the original character of the latter, nor the history of its preservation, nor the present state of its text entitles it to the precedence. Only in cases where there are inde- pendent reasons for suspecting the accuracy of the Hebrew, can emendations by the Septuagint be reason- ably admitted. In estimating the separate value of the various read- ings gathered from the different sources which have now been reviewed, besides the weight of external evi- dence attaching to them severally, considerations may be drawn from internal grounds in the nature of these readings themselves. The most general rule is that the reading which will most satisfactorily account for the others is the true one. And here recourse must be had to the various modes in which errors arise, as previously THE CRITICISM OP THE TEXT 175 exhibited. For this reason the more difficult reading, or that which contains unusual forms, is often to be re- garded as the original one, since transcribers would be naturally tempted by the difficulty to substitute an easier reading or a more customary expression. Further, that which yields the best sense, and agrees best with the context and the scope of the writer, has a claim to be regarded as the true reading. Tbe style of the author may also furnish a presumption in favor of one reading and against another. This is a criterion, how- ever, which has often been abused. False conclusions have in many cases been reached by judging of the genuineness of passages by rhetorical maxims or fancied characteristics of style. There is no collateral source of information about the text prior to the Septuagint. Its condition previously can only be inferred from an examination of the text itself. An improper use has been made of duplicate passages on the assumption that they must originally have been identical in every word and phrase, and that every deviation of one from the other is a textual error requiring correction. Thus Num. xxiv. 17b, iritfB ynm no "\32rbl "ij?"lj?'1 3XT0, ' shall smite through the corners of Moab and break down all the sons of tumult,' is re- peated with variations in Jer. xlviii. 45b, nxs bssrfl Ti81» isa Ip'lp) 2Sitt, ' hath devoured the corner of Moab and the crown of the head of the sons of tumult;' but these variations are not errors of transcription. One inspired writer in adopting the language of another did not feel bound to repeat it verbatim, but in the con- fidence of his equal inspiration modified the form at pleasure to suit his immediate purpose. So the Psalms that occur more than once with some change in the ex- pressions by no means warrant the conclusion that only one of them has been accurately preserved, or that 176 GENERAL INTRODUCTION neither has, and the true original must be elicited by a comparison and correction of both. Both copies are authentic ; and their very discrepancies are proof of their careful preservation, and the conscientious pains both of the collectors of the Canon and of subsequent transcribers in retaining each in its integrity and keep- ing them from being assimilated to each other. Ps. liii. is not an erroneous copy of Ps. xiv., nor vice versa; but an adaptation of an earlier Psalm to a new situa- tion. As Delitzsch correctly remarks, "a later poet, perhaps in the time of Jehoshaphat or Hezekiah, has given to David's Psalm a reference to the most recently experienced catastrophe of judgment." Ps. xviii. and 2 Sam. xxii. are two different forms of the same Psalm, the former as it was sung in the sanctuary, the latter most probably as it was current in the mouths of the people when the Books of Samuel were written. Wrong inferences have also been drawn from the im- perfect structure of certain alphabetic acrostics. Thus in Ps. ix. the letters from Aleph to Kaph mark the ini- tials of verses, though with some irregularities. Ps. x. begins with the next letter Lamedh, and toward its close the four last letters of the alphabet occur in regu- lar order ; but in the intervening verses the alphabetic structure is entirely disregarded, although their number corresponds with that of the letters omitted. This is not due to an erroneous text or a reshaping of the Psalm ; but, as Delitzsch properly insists, the Psalmist did not allow himself to be fettered by regularity of form when it interfered with the free expression of his thought. He also refers to the fact that the Syriac presents simi- lar irregularities in alphabetic poems. The failure of the text to correspond with the de- mands of certain hypotheses, which have been obtruded upon it, has also led to the unfounded charge of textual THE CRITICISM OF THE TEXT 177 errors. Bertheau devised a very ingenious scheme of dividing the Mosaic laws into seven groups, each group containing seven decalogues; and he carried it through by assuming dislocations and interpolations sufficient for his purpose. As some of the laws in Ex. xxi.-xxiii. contain groups of ten, it has been inferred that the entire section was originally a series of deca- logues, and that these have been defaced or mutilated by errors in the text, and the attempt has been repeat- edly made to remove these errors, and thus restore the laws to their primitive form. But the wide divergence in the results reached shows that no satisfactory con- clusion has yet been attained, and the primary assump- tion still lacks confirmation. Textual errors have been charged upon poetical passages because the lines con- travene certain rules which have been formulated for Hebrew verse ; but the question arises whether the rules may not be discredited rather than the text, when these are not in harmony. The errors assumed to exist on insufficient grounds, such as have now been referred to, have been thought to be so numerous as to throw discredit upon the care with which the text was preserved during this early period, and to indicate that it was very uncertain and inaccurate. There is no good reason, however, for such an opinion. Nevertheless, there are occasional errors which are obvious; and as they appear in the versions as well, they must have antedated them. Where there are no external helps for their correction, we can only have recourse to critical conjecture. This should be only sparingly used, apd should be restricted to cases of actual necessity. The unlimited use made of it by some critics converts the text into what they themselves would have written instead of what the author actually wrote. 13 178 GENERAL INTRODUCTION Justus Olshausen 1 asserted that "the most palpa- ble errors and most manifest deficiencies of the text were not recognized as such, or at least were left un- touched, when it was authoritatively established, and this evidently rests not upon a comparison of manu- scripts, but in every part of the collection upon only a single authority, upon a single, often seriously damaged manuscript which was followed with slavish fidelity." Lagarde 2 subsequently propounded the same hypothesis that all Hebrew manuscripts are traceable to one faulty source, but based it on a somewhat different reason. He assumes that the extraordinary points over certain words or letters indicate according to Greek and Syrian usage that they should be expunged, that letters written above the line were a later addition, and that open spaces in the lines mark a hole in the parchment, or show that the skin was imperfectly tanned and could not be written upon, or that the transcriber could not read the copy before him, or was not at the moment provided with the red ink needed for headings. He then proceeds " if now puncta extraordinaria and literce suspensce of the Hebrew text prove that the copyists have made mistakes, and if the Piska implies that some casualty had befallen the scribe or the skin on which he was writing, all manu- scripts which have these points, letters floating in air, and open spaces must necessarily be slavishly faithful copies of the same original. It would be possible, though surprising, that all copyists should have the same correct idea in the same place, but that all should independently of one another and of the copy before them have made the same mistake in the same spot, and have corrected it in the same way, is unthinkable." 1 Die Psalmen Erklart, 1853, p. 18. » Anmerkungen zttr Griechischeu Uebersetzung der ProYerbien, 1863, pp. 1, 2. THE CRITICISM OP THE TEXT 179 If, however, the Hebrew text was already traditionally settled and could not be disturbed, when these points and letters of unusual size, form, and position were ap- pended, whether they expressed the critical doubts above suggested, or were intended as reminders of other teachings of the schools orally given but now for- gotten, it is quite conceivable that copyists should value and perpetuate them. Besides, there were minor diver- sities in the text that prevailed at different influential centres notwithstanding the agreement in general, as appears from the variant readings of the Babylonish and Palestine Jews before spoken of, and from incon- sistencies in the Massora itself indicating some slight disagreement among leading authorities at that early period. There is not that absolute unanimity, there- fore, in the earliest obtainable form of the text, which the theory of a single manuscript source implies. In the language of Dr. Dillmann, 1 "The assertion of P. de Lagafde that all Hebrew manuscripts are descended from one parent manuscript with all its errors is so improbable in itself, and so poorly supported by the alleged testimony of a late Christian author, that no one else will probably assent to it." The sum of the whole matter is this. The Hebrew manuscripts cannot compare with those of the New Testament either in antiquity or number, but they have been written with greater care and exhibit fewer various readings. In fact the various readings obtained from an extensive collation of the Hebrew manuscripts are of little importance, for they all represent substantially what is known as the Massoretic text. This is so minutely guarded by the Massora, that it can by its aid be accurately determined, and traced back long prior to ' Bibelteyt des A. T. in Herzog-Plitt Encyklopadie, II., p. 388. 180 GENERAL INTRODUCTION existing manuscripts. It is further shown by the Tal- mud, as well as by the testimony of Origen and Jerome, to have been regarded as a settled and inviolable text in their day, and can thus be traced back century by century from the fifth to the second century of the Christian era. The same is freshly confirmed by the Latin version of Jerome, the Syriac Peshito, the Jewish Targums and the Greek versions of the second century, all which agree substantially with the text as we possess it at present. Their deviations are of minor consequence, though affording material for its correction in some in- dividual cases. The Septuagint in the third century B.C., and the Samaritan Pentateuch diverge from it much more widely ; but neither of them offers a text that is now regarded by scholars as comparable in accu- racy with the Hebrew. And their divergence, whether laid to the account of inaccurate Hebrew sources or to liberties taken by Greek translators and Samaritan copyists, does not prove that there was not at the same time an authorized and reliable text represented in standard copies. The Septuagint may be of service in correcting the Hebrew text in certain cases ; but the Samaritan Pentateuch is valueless for purposes of criticism. Prior to those which have just been mentioned there are no external authorities with which to compare the Hebrew text. But an examination of the text itself does not reveal the numerous errors which some have thought to find there. And there is no reason to doubt that the text was even then guarded with sedulous care by the scribes who had special charge of it from the time of Ezra. There are indeed some manifest errors which may in part be corrected by parallel passages ; the rest must be left to critical conjecture. The reten- tion of the former class, however, though the remedy THE CRITICISM OF THE TEXT 181 was so close at hand, is a fresh evidence of that rigid adherence to the letter, which has so remarkably safe- guarded the Old Testament. It may be safely said that no other work of antiquity has been so accurately trans- mitted. INDEX Abba Salama, 100 Aben Ezra on Hebrew vowel signs, 63 Age of Hebrew manuscripts, how determined, 78 Agglutinative languages, 4 Alcuin, revision of the Latin text, 119 Aldine edition of the Septuagint, 97 Alexander the Great, 86 Alexandrinus codex, 97 Alphabetic acrostics, critical use of, 176 Alting, 45 Anglo-Saxon version mediate in both Testaments, 82 Antiochus Epiphanes, 154 Antiquity and authority of the vowel signs, 63 ff Antwerp polyglot, 98, 158 Aquila, 89 and note ; his version, 90, 92, 116; agreement with the Massoretic text, 170 Arabic language and literature, 13, 14 ; relation to Hebrew, 33 ; originally without vowels, 65 Arabic versions, 102, 141 Arabisms in Hebrew, 20 Aramaeisms, proper and improper, 20,22 Aramean in the Old Testament, 1 ; in the New Testament, 40 Aramean inscriptions, 61 Aramean language and literature, 13, 14; displaced Hebrew, 38; Jewish and Christian, common characteristics, 39 ; differences, 40 ; extant writings, 40 Arias Montanus, 155 Aristeas' account of the Septua- gint, 83 ; discredited, 85 Aristobulus, IE note, 84 Armenian version, 101 Assyrian language and literature, 13, 14, 15 Asterisk in the Hexapla, 93 Augustin's attitude toward Je- rome's version, 115, 117 Babtlonish vowel signs, 72 Bachmann published parts of the Ethiopic version, 101 Bacon, Roger, state of the Latin text, 121 Bardelli edited Daniel in Coptic, 101 Bellarmin, the Vulgate, 128, 129 Ben Asher, 71; codex of, 80; table of various readings, 154 Ben Naphtali, 71 ; codex of, 80 ; table of various readings, 154 Berliner on Onkelos, 104 Biblia Regia, 158 Bohairic version, 101 Bomberg, enumeration of verses in the Hebrew Bible, 155; his rabbinical Bible, 156 Books for writing, 143 183 184 INDEX Bos, Lambert, 95, note Buxtorf (father), 62 ; answers Levita, 63 ; his rabbinical Bible, 156 Buxtorf (son), 58; his hypothesis concerning Hebrew letters, 59 ; vowel points, 64, 66, note, 67, note ; on expressions in the Tal- mud, 68, note Cappellus, 58, 62 ; vowel points, 64 ; the Septuagint and Hebrew text, 171, 172 Carving on solid materials, 142, 143 Cassiodorus, correction of the Latin text, 119 Castell, Edmund, 44, 159 Chaldee, 39 ; see Aramean Chapters, when introduced, 155 Charlemagne, revival of Hebrew study, 42 ; revision of the Latin text, 119 Charles II., 159 Chigi, Cardinal, the Septuagint Daniel, 99 Chrysostom, 100 Cistercian revision of the Latin text, 109 Clement VIII., the Vulgate, 129 Cleopatra, 85 Codex Alexandrinus, Vaticanus, Sinaiticus, Friderico-Augus- tanus, 97 Codices, standard, 79 Comparative school of Hebrew study, 44 Complutensian polyglot, 97, 155, 157 Comprehensive school of Hebrew study, 46 Constantine, 76 Copiousness of Hebrew, 30, 31 Coptic version, 101 Cornill, the Ethiopia version, 100 ; Coptic version, 101 Correctoria Biblica, 119, 120 Council of Trent, the Vulgate, 122-126 Council of Vienna, appointment of Hebrew Professors, 42 Critical conjecture, 177 Critical editions of the Hebrew Bible, 156 Criticism, higher, 160-162; text- ual, 162; sources of, 163; in- ternal grounds, 174, 175; sum- mary of results, 179-181 Cromwell, Oliver, 159 Cyril, Slavic version, 101 Damasbs, Bishop of Borne, 114 Daniel, Theodotion substituted for the Septuagint, 87, 99; Septu- agint version of, 99 ; Coptic version of, 101 Danz, 45 Decalogues, critical use of, 177 Delia Valle, the Samaritan Pen- tateuch, 129, 130 Demetrius Phalereus, 83, 84, 85 De Rossi, 81, 157, 164 De Rossi, Azariah, 105 Deuteronomy in the Septuagint, 87 Dialects of Hebrew, 18 Dillmann, the document P, 29 note ; edited parts of the Ethio- pic version, 101 ; the hypothesis of a single manuscript, 179 Drusius, 95, note Duplicate passages, evidence of errors, 145 ; wrong use in crit- icism, 175 Ecclesiastes in the Septuagint, 87 Ecclesiasticus, prologue to, 87 INDEX 185 Egyptian versions, 101 Egyptian words in Hebrew, 35 Eliaa Levita, 42, 63 Enneapla, 94 Epanorthots, 119, 120 Epiphanius, the Septuagint, 85 ; Aquila, 89, note ; Lemnisk, 94, note Erasmus, correction of the Latin of the Vulgate, 121 Errors of transcription guarded against, 144; shown by duplicate passages, 145 ; not made by col- lectors of the canon, 146 ; how they arise, 163; intentional al- terations, 164 Ethiopic language, 14 Ethiopia version, 100 Eusebius, 95 and note, 96 Ewald, language of the document F, 26, note ; his grammar, 46 Extraordinary points, 152 Families of languages, 4 Field, 90, 94, 95, 113 Firkowitch, 71, 81 Franciscan correctorium, 120 Frankel, Targum of Onkelos, 105 ; of Jonathan, 107 Friderico-Augustanus, codex, 97 Frumentius, 101 Gallican Psalter, 115 Geiger, Jerusalem Targum, 110 Gemara, 41, 149, 152, 166 Georgic version, 101 Gesenius, 46 ; Hebrew letters, 60 ; age of the Samaritan Penta- teuch, 131; its critical worth, 135-138 ; its relation to the Sep- tuagint, 139, 140 Gheniza, 76, 81 Giesebrecht, the document P, 25, 26 Gothic version, 101 Grabe, edition of the Septuagint, 98 Grammatical period of Hebrew study, 41 Grecian in the New Testament, 17 Greek words in Hebrew 36 Gregory XIII., 98 Gregory XIV., 128, 129 Grotius, the Samaritan Penta- teuch, 140 Groups of languages, 4 Gussetius, 45 Haphtaboth, 75, ITE Hebrew accents, 74 Hebrew Bible, early editions of, 156 Hebrew, the, cited by Greek fathers, 100 Hebrew coin and square letter, 57 ff., causes of change, time of transition, 61, 147; not af- fect the integrity of the text, 62 Hebrew inscriptions without vow- els, 65 Hebrew manuscripts, 75 ff. ; for Synagogue use, 75, 76 ; private, 77 ; in rabbinical character, 78 ; determination of their age, 78 ; raassoretic and non-massoretic, 79 ; standard codices, 79 ; oldest MSS., 80, 81 Hebrew, the language of the Old Testament mainly, 1 ; derivation of the word, 16 ; its usage, 17 ; its application to a language, 17; the language of Palestine, no trace of polytheistic origin, its dialects, provincialisms, 18 ; in prose and poetry, 19 ; in differ- ent periods, 21 ; in the later 186 INDEX books of the Old Testament, 22 ; its stationary character no ob- jection to the antiquity of the Pentateuch, 22-24 ; its copious- ness, 30, 31 ; synonyms, 31 ; economy of roots and words, 32 ; lost roots, 33 ; relation to the other Semitic tongues, 33, 34 ; Egyptian words, 35 ; Sanscrit, Persian, Greek words, 36; its words in western tongues, 37; when it ceased to be spoken, 37, 38 ; succeeded by Aramean, 38, 39 ; periods of its study, 41 ; vowel signs, 41, 63 ff. ; when in- troduced, 71 ; two systems, 72 ; their correctness, 73, 74 ; words anciently separated in writing, 144; pronunciation orally pre- served, 147 Heidenheim, 81 Hellenist, 17 Heptaglot lexicon of Edmund Castell, 44, 159 Heptapla, 94, note Hesychius, revision of the Septu- agint text, 96, 99 Hexapla of Origen, 91-93 ; at Cajsarea, 95; effect upon the Septuagint text, 95 Hexaplaric, Syriac version me- diate in the Old Testament, 82 ; made from the Septuagint, 100 Hillel, codex of, 80 History of the text, four periods, 142 Hody, 94 Houbigant, 156 Hugo St. Clair's correctorium, 120 Hupfeld, vowel points, 66, note ; Talmudic phrases, 70, note Hypolemnisk in the Hexapla, 94 Idiomatic school of Hebrewstudy, 45 Indo-European family differs from the Semitic in external flexion, 5 ; constitution of roots, alpha- bet, 6, 7; richness in forms, number of branches, constant change, 8 ; less pictorial, 10 ; greater precision, 11 Indo-European races, their char- acteristics, 9 ; put in charge of the New Testament, 10, 12 Inflected languages, 4 ; embracing the Indo-European and Semitic families, 5 Inscriptions of Mesa and Siloam, words separated, 144 Irenseus, 84, 90 Isaac Nathan, 155 Isolating languages, i Israel, codex of, 80 Itala, mediate in the Old Testa- ment, 82 ; made from the Sep- tuagint, 100, 114; revised by Jerome, 114; reproduced from quotations, 114, note Jehbmiah much transposed in the Septuagint, 87 Jericho, codex of, 80 Jerome, 12, 42, 58, 61, 95 ; prior to Hebrew vowel signs, 66; K'ri andK'thibh, 71 ; pronuncia- tion of Hebrew, 74 ; Aquila, 90 ; Theodotion, Symmachus, 91 ; and other Greek versions, 92; revisions of the Septuagint, 96, note ; revised the Latin version, 114, 115; his own version, 115, 116, and note; superseded pre- vious Latin versions, 117 ; his capitula, 149 ; fixed Hebrew text, 166 ; agreement with Mas- soretic, 170, 172 INDEX 187 Jerusalem Targum, 108-110 Jew, 17 Jewish coin letter, 57; tradition respecting vowels, 65 Jews wrongly charged with cor- rupting the text, 151 Jews' language, 17 Job, Targum on, 103 Jonathan, Targum of, 106-108 Josephus, 84 Judah ha-Kadosh, 152 Justin Martyr, 81, 85 Kennicott, 81, 130, 131, 156, 164 Kimchi, David, 42, 156 Kimchi, Joseph, 42 Kimchi, Moses, 42 Kopp, 60 K'ri and K'thibh, 71, 148, 165 Lagarjde, restoration of the Sep- tuagint text, 99 ; hypothesis of a single manuscript, 178, 179 Lanfranc, revision of the Latin text, 119 Langton, Stephen, introduced chapters, 155 Language, unity of, 2, 5 ; families, isolating agglutinative, inflected groups, 4; Semitic and Indo- European contrasted, 5-12 LeClerc, age of the Samaritan Pentateuch, 131 Lemnisk in the Hexapla, 93 Leo X., 157 Levita, Elias, 42 ; Hebrew vow- els, 63 Leviticus in the Septuagint, 87 London polyglot, 98, 158, 159 Lucian, revision of the Septua- gint text, 96, 99 Luther's Hebrew Bible, 156 Lyra, Nieolaus de, 43 Lysimachus, 87 Manuscripts, Hebrew, 75 ff. ; for Synagogue use, 75, 76 ; private, 77 ; in rabbinical character, 78 ; age how determined, 78 ; masso- retic and non-massoretic, 79 ; standard codices, 79; oldest MSS., 80, 81; worth of, how estimated, 164; represent the massoretic text, 164 Margoliouth, 81, and note Massora, 41, 71, 77, 78, 151, 153 ; marginal, terminal, 153; great, little, 154 Massoretic, period of Hebrew study, 41; verses, 154; text guarded by the Massora, 165 ; not the creation of the Masso- rites, 165; superior to that of the Septuagint, 174 Massorites, 152 ; signs for vowels and accents, 153 Memphitic version, 101 Methodius, Slavic version, 101 Metobelos in the Hexapla, 93 Miesrob, Armenian version, 101 Mishna, 41, 149, 152, 166 Montfaucon, 95, note Morinus, John, the Samaritan Pentateuch, 130, 131 ; the Sep- tuagint, 171 Muis, Simeon de, the Septuagint, 171 Nebuchadnezzar, 86 Nieolaus, revision of the Latin text, 119 Non-massoretic manuscripts do not exist, 79 Obelos in the Hexapla, 93 Octapla, 94 Odessa manuscripts, 80 Olshausen, Justus, hypothesis of a single manuscript, 178 188 INDEX Onkelos, 68, note ; Targum of, 104-106 Oriental languages, 12 Origen, 42, 58, 61 ; prior to He- brew vowel signs, 66 ; pronun- ciation of Hebrew, 74 ; his Hex- apla, 91, 92 ; its aim and meth- od, 93 ; his expulsion and death, 95 ; agreement with the masso- retic text, 166, 172 Owen, John, 65 Pamfhiltjs, revision of the Sep- tuagint text, 96, 99 Parashoth, 76 ; pretalmudic, 149 ; fifty-four, 154, and note Parisian polyglot, 158 Paul of Telia, 113 Pellican, Conrad, 43 Pentapla, 94 Pentateuch, antiquity of, not dis- credited by the character of the language, 22-24 ; in the Septua- gint, 87 Pentateuchal document P alleged to be of late date, 24; argued by Wellhausen and Giesebrecht, 25, 26 ; answered by Kiehm and Byssel, 25, note; Ewald, 26, note ; Dr. Driver, 27-29 ; Dill- mann, 29, note ; not proved by the use of 1JS, 47-53, or T>blPI, 54-56 Persian words in Hebrew, 36 Feshito, 82; why so called, by whom made, 111 ; general char- acter, relation to the Septua- gint, 112, 171 ; its date, 112, 113 ; agrees in general with the mas- soretic text, 171 Phalaris, epistles of, 161 Phehician monuments without vowels, 65 ; sometimes the words separate, 144 Fhilo, 84 Pinner, 80 Points, extraordinary, 152 Polyglots, 97, 98, 157-159 Post-massoretic period, 155 Prideaux, the age of the Samari- tan Pentateuch, 131 Primitive language, search for the, 2, 3 Prophetic style in Hebrew, 20 Proverbs, transpositions in the Septuagint, 87 Provincialisms in Hebrew, 18 Ptolemy Philadelphus, 83, 84, 85, 86 Ptolemy Philometor, 87 Ptolemy Soter, 85, 86, and note Quinta version, 92 Quotations in early writers, 165 Rabbinical Bible, 108, 156 Raymund Martini, 43 Reuchlin, John, 43 Eeuchlin manuscript, 81 Kiehm, the language of the docu- ment P, 25, and note Roman Psalter, 115 Roots in the Hebrew Bible, the number of, 30 Ryssel, the language of the Pen- tateuchal Elohist, 25, note Sahidic version, 101 Samaritan dialect, 13 ; letters, 59 ; no vowels, 65 Samaritan Pentateuch first brought to Europe, 129; first printed, 130, its origin, 131- 134; its text, 134; its critical value, 135 ; shown by Gesenius to be worthless, 135-138 ; rela- tion to the Septuagint, 139, 140 ; verses and sections, 149 Samaritan, the, cited by Greek fathers, 100 INDEX 189 Samaritan version, 141 Sanbuki, codex of, 80 Sanscrit words in Hebrew, 36 Scaliger, Joseph, 129 Schools for Jewish learning, 41 ; of Hebrew study among Chris- tians, 44-46 Schultens, the number of Hebrew roots and words, 31 ; compara- tive study of Hebrew, 44 Scribes date from Ezra, 146; their function, 146, 147, 150'; critical corrections, 151 Sections pretalmudic, 149; open and closed, 149 Sedarim, 157 Semitic languages differ f romlndo- European in internal flexion, 5 ; constitution of roots, alphabet, 6 ; triliteral roots, verbal spe- cies, 7 ; less rich in inflections, fewer branches, more station- ary, 8 ; more pictorial, 10 ; less definite, 11 ; fitted for the Old Testament revelation, 12; spoken mainly by descendants of Shem, 12, 13 ; their proper home, 13; languages of relig- ion, 13 ; principal branches, 13 Semitic races, characteristics of, 9 ; suited for the preliminary revelation, 9 Semitic words in western tongues, 37 Septima version, 92 Septuagint, which letter in use when made, 62 ; prior to vowel signs, 66 ; transliteration of He- brew, 74 ; ancient and im- mediate, 82; Aristeas's account of its origin, 83 ; Aristobulus, Josephus, Philo, Justin Mar- tyr, Irenaeus, the Talmud, 84; probable origin, 86, 87; dif- ferent translators of various ability, 87; liberties taken in translation, 88 ; how regarded by the Jews, 88, 89 ; by Chris- tians, 89; corruption of the text, 91 ; the Hexapla, 92, 93, 95 ; re- visions of its text, 96 ; manu- scripts, and printed editions, 97, 98 ; Lagarde's plan of restoring its text, 99 ; Daniel, 99 ; ver- sions made from it, 100-102 ; divergence from the massoretic text, 171-173 Sexta version, 92 Shemitish languages, 12 ; see Semitic Siloam inscription, 57 Sinai, codex of, 80 Sinaiticus codex, 97 Sixtine edition of the Septuagint, 98; of the Vulgate, 127, 128 Sixtus V., 98, 127 Slavic version, 101 Solomon ben Ishmael, 155 Sopherim, post-talmudic tract, 85, note, 147 Sorbonne, correctorium, 120 Stade alleges falsification at the collection of the canon, 146 Strack, H. L., 80, 81, and note Summary of the results of textual criticism, 179-181 Superlinear vowel system, 72, 73 S wete, edition of the Septuagint, 98 Symmachus, version of, 91, 92, 116, 170 Synagogue manuscripts without vowels, 65 ; described, 75, 76 Synonyms in Hebrew, 31 Syriae, the, cited by Greek fa- thers, 100 Syriae language and literature, 13, 14 ; without vowels originally, 65 Syro-Arabian languages, 12 Syro-Hexaplaric version, 82, 100 ; its date and character, 113 190 INDEX Taggin, codex of, 80 Talmud, 41, 57; prior to vowel points, 68 ; ita phrases ex- plained, 68, 69 ; notices K'ri and K'thibh, 71 ; on the Sep- tuagint, 84, and note ; rules for copyists, 144 ; teits already fixed, 148 ; when committed to writing, 152 ; number of verses, 154, note Talmud of Babylon confused Aquila and Onkelos, 104 ; Tar- gum of Joseph, 107 Talmud of Jerusalem, Aquila, 90, note, 104 Tam letter, 78 Targnms, 40, 41 ; prior to vowel signs, 67 ; ancient and imme- diate, 82; origin of, 102, 103; how many, 104 ; Onkelos, 104- 106 ; Jonathan, 106-108 ; Pseu- do-Jonathan, Jerusalem, 108, 109 ; Hagiographa, Megilloth, none on Daniel, Ezra, and Ne- hemiah, 110; corresponds with the massoretic text, 170 Tattam edited parts of the Coptic version, 101 Tetrapla, 94 Text, when fixed, 150, 166 Thebaic version, 101 Theodotion, Daniel, 87, 99 ; his life and version, 91, 116; in the Hexapla, 92; agreement with the massoretic text, 170 Tiberias, doctors of, authors of the vowel signs, 63 Tischendorf, editions of the Sep- tuagint, 98 Traditional school of Hebrew study, 44 Trent, Council of, the Vulgate, 122-126 ; its decree, 123, note ; differently understood, 124 ; commission to correct the text, 125, 126 Ulphilas, Gothic version, 101 University of Paris, correctorium, 120 Urban VII. ,,128 Ussher, Archbishop, 64, 95, 140 Various readings, tables of, 154 Vaticanus, codex, 97 Velsh letter, 78 Verses, pretalmudic, U8- Versification, rules of, in criti- cism, 177 Versions, ancient, immediate, me- diate, 82 ; their critical, exeget- ical and hermeneutical value, 166, 167; caution requisite in criticism, 168-170; their testi- mony regarding the massoretic text, 170, 171; divergence of the Septuagint and Samaritan Pentateuch, 171 ; how explained, 172, 173 Vowel points, 41 ; their origin, 63 ff. ; introduced by Massorites, 153 Vulgate, ancient and immediate, 82 ; Jerome's version, so called, previous usage of the term, 118 and note ; corruption of the text, correction by individuals and fraternities, 119 ; first print- ed editions, 121 ; in the Council of Trent, 122-126; papal edi- tions, 127-129 Walton, prolegomena, 65; dedi- cation, 159 Watson, W. Scott, 130, note Wellhausen, document P, 25 Words in Hebrew Bible, number of, 30 Ximenbs, Cardinal, revision of the Vulgate, 122 ; Complutencian polyglot, 157 Zunz, Jerusalem Targum, 110 Other Works by Prof. William Henry Green D.D., LL.D. CHARLES SCRIBNER'S SONS, Publishers « « « General Introduction to the Old Testament The Canon 8vo. v^/.jo Contents: History of Introduction to the Old Testament — General Introduction to the Old Testament— The Canon of the Old Testament — Testimony of the Bible in re- gard to the Formation of the Canon — The Critical Theory of the Formation of the Canon — The Determining Principle in the Formation of the Canon — The Completion of the Canon — The Threefold Division of the Canon — When and by Whom Col- lected — The Extent of the Canon — The Canon of the Jews — The Canon of Christ and his Apostles — The Canon of the Christian Church — The Apocrypha Condemned by Internal Evidence— Order and Number of the Canonical Books. "This -volume satisfactorily meets what had come to be a pressing, not to say oppressive, want with the English-speaking public. In the voluminous and rapidly succeeding publications of the destructive critics the fundamental beliefs of the church seemed to be in danger of being lost by default. We have in this learned and yet popular volume a most able and satisfactory defence of the more generally accepted views con- cerning the canon of the Old Testament." — Bibliotheca Sacra. "Critics of all schools will be glad to have Dr. Green's compact pres- entation of one possible theory of the canon. . . It is to be com- mended for giving so frankly and fully the views which he opposes in the language of their advocates. He has a keen eye for the weak points in the armor of ' the critics.' " — The Congregationalist. "A very timely book. Naturally Dr. Green could not overlook the, fact that his views are not those of many of the more recent writers on the Old Testament. He has taken especial note of the views of Driver OTHER WORKS BY PROFESSOR DR. GREEtf Ryle, and Wildeboer. His arguments for the traditional date of Daniel, Ezra, Nehemiah, Esther, and Chronicles deserve the most careful atten- tion." — Hartford Seminary Review. " Dr. Green, as his grammars and other works testify, is beyond sus- picion in the matter of Biblical, and especially of Old Testament scholar- ship. In the works of Briggs, Ryle, and Wildeboer the canon of the Old Testament has been ably dealt with from the point of view of the ' higher criticism. ' It is well that the other side should be represented, and it would be hard to find one more competent to represent it than Dr. Green, who might be called the Nestor of Old Testament critics." — American Journal of Theology. * * « The Higher Criticism of the Pentateuch 8vo. $1.50 Contents : The Old Testament and its Structure— The Plan and Contents of the Pen- tateuch—Moses the Author of the Pentateuch— The Unity of the Pentateuch- Genuineness of the Laws— The Bearing of the Divisive Criticism on the Credibility of the Pentateuch and on Supernatural Religion. " This book is peculiarly welcome for its grasp of the main question, and the precision and logical force of its reasoning. It is a calm, cour- teous, and intensely reasonable discussion of all the points at issue by one who is familiar with all the extensive literature about them and who is able to state his position with surpassing clearness." — The Christian Intelligencer. " Delitzsch said that there were not five men in the world qualified by mind and attainments to understand the whole Pentateuchal controversy and give a well-considered opinion on it in all its bearings. If that is true, here is one of the four. The work is an honor to America, the best compendium of both sides of the great debate that has appeared in Europe or in America. All those in America who believe the Bible to be God's Word have in Dr. Green a champion, a leader, whose fairness to opponents and clear perception of the impregnable rock abundantly qualify him for that position. "— The Presbyterian and Reformed Review. "No clergyman should be without this volume. . . . We com- mend it to every student of the Bible. The author has fully carried out OTHER WORKS BY PROFESSOR DR. GREEN' his purpose, namely, to show, as briefly and compactly as possible, that the faith of all past ages in respect to the Pentateuch has not been mis- taken." — The Churchman. "A most valuable contribution to the literature of what we may call Scriptural apologetics. It is a strong, learned, and logical argument for the acceptance of the Pentateuch as Christ accepted it." — Boston Tran- script. " He has stanchly defended the unity and the Mosaic authorship of the entire Pentateuch. There is nothing, therefore, in his latest work which he has not repeatedly said before ; but he has put together in small compass the best that can be said in support of the traditional opinion by its ablest surviving defender." — The Nation. « « * The Unity of the Book of Genesis 8vo. $3.00 Contents : The Book of Genesis — The Generations of the Heavens and the Earth — The Generations of Adam — The Generations of Noah — The Generations of the Sons of Noah — The Generations of Shem — The Generations of Terah — The Generations of Ishmael — The Generations of Isaac — The Generations of Esau— The Generations of Jacob — Conclusion. " No stronger or more interesting statement of the conservative posi- tion on the question of Pentateuchal criticism can be found than this volume." — The Advance. " In ten chapters, covering the whole of Genesis, he considers calmly all that is said by these leaders (Graf, Kuenen, Wellhausen, Dillmann, Reuss, and their followers), and as calmly and with great point and force states the opposing considerations. His treatment of the second chapter of Genesis and of the sixth to the eighth chapter, the asserted duplicate accounts of creation and of the flood, furnishes excellent speci- mens of the masterly simplicity which is born of comprehensive knowl- edge of the facts, of clear induction, and of » pellucid style." — Presby- terian and Reformed Review. " This work will greatly enhance Professor Green's already high repu- tation as one of the foremost living scholars in the Old Testament field. OTHER WORKS BY PROFESSOR DR. GREEN It is an exhaustive criticism of the documentary hypothesis, so far as the book of Genesis is concerned, and aims to prove that this hypothesis has no adequate foundation upon which to rest. . . . The work is one which no one interested in Old Testament study can afford to ignore. It marks an epoch in the formation of English-speaking opinion on the subject." — The Churchman. "Professor Green's analysis has failed to shake our conviction that the Book of Genesis was composed, like the Diatessaron of Tatian, of preexisting accounts, which have been skilfully woven together, not without leaving marks of the joiner's art. But he who desires to know what the ablest and most thorough scholarship in either England or America can say in opposition to this opinion will find it in Dr. Green's volume." — The Outlook. « « « Celebration of the Fiftieth Anniversary of the Appointment of Professor William Henry Green as an Instructor in Princeton Theological Seminary May fifth, 1896 With Portrait. 8vo. Net, $1.50 "The volume forms a fitting memento of a rare and impressive occa- sion." — The Interior. "As the addresses were by eminent speakers, the book has a value of its own in addition to the interest of the honored subject." — Cincinnati Herald and Presbyter. CHARLES SCRIBNER'S SONS, Publishers 153-157 FIFTH AVENUE, NEW YORK