the different degrees of accuracy with which the two texts have respectively been preserved. The Hebrew MSS. may not be earlier than the ninth or tenth century, but the copious quotations from the Old Testament in the Talmud, the evidence of the Vulgate (translated from the Hebrew about A.D. 400), and the Greek versions of Aquila, Symmachus, and Theodotion (so far as they are known to us), testify to the text having come down without substantial corruption from a a much earlier period. The Targums, or paraphrases of the old Hebrew into the Aramaic dialect of every-day life, show a text, dating from about the beginning of the Christian era, which differs but slightly from the Massoretic. In later times we know that scrupulous care was taken to ensure the minutest accuracy in the transcription of the sacred books; and this evidence tends to show that the same care was exercised at least from about the period of the fall of Jerusalem, and possibly much earlier. On the other hand, the MSS. of the Septuagint exhibit differences among themselves of the most marked kind. The Codex Vaticanus evidently represents a distinct type of text from the Alexandrinus; the Codex Sarravianus, which is a Hexaplar MS., differs from both. Some copies belong to definitely-marked editions of the Septuagint, as described above; others, and the greater number, cannot be classified on any such principle, but exhibit every degree of individual variation. would appear as if the Greek-speaking Jews had been less careful to maintain an unvarying standard of accuracy in their transcriptions of the Scriptures, and that consequently the presumption, in any case of divergence, is in favour of the Hebrew.

It

This consideration is strengthened by an examination of the Canon of the Septuagint and the style of the translations of the several books. Sir H. Howorth would have us believe that the Canon of the Septuagint is that which really represents the accepted Hebrew Bible before the fall of Jerusalem; but his assertion has yet to be made good. Neither in the New Testament nor in Josephus do we find the apocryphal books cited on the same footing as the canonical. It is, no doubt, difficult to draw the line between Ecclesiastes and the Song of Songs on the one hand, and Ecclesiasticus and Wisdom on the other; but these are the points of nearest contact between the two groups of books, and few will deny that the standard of the canonical books-and therewith the significance of the doctrine of Inspiration-would be lowered by the inclusion of Tobit and Judith, or the incorporation with Daniel of the stories of Susanna and Bel and the Dragon. Sir H. Howorth prefers the narrative of 1 Esdras to that of the Hebrew Ezra ; but it is a bold thing to

affirm that the episode of the three youths who contended before Darius for the palm of wisdom (1 Esdr. iii. iv.) is a part of the book as originally written. In short, most of the contents of the Apocrypha have the appearance of being late compilations, in which historical truth was subordinated to ethical and literary considerations, and fall far short of the standard which we find in the books universally accepted.

Now if the Alexandrian Jews, among whom the Septuagint originated, were indifferent as to the Canon of their Bible, and readily accepted additional books and interpolated passages, it will not be altogether surprising if they were also lax in the matter of the text. An examination of it in detail goes far to show that this was actually the case. The Pentateuch is, no doubt, fairly well and accurately translated, but many of the later books are full of errors, proceeding sometimes from actual mistranslation, and sometimes from a tendency to paraphrase. The Book of Isaiah, in particular, is so ignorantly translated as to be almost unintelligible; and a large part of the variations in Job, of which a specimen has been given above, must be set down to the translator, and not to differences in the text. The version of Daniel was found so unsatisfactory, even in earlier times, that it was superseded by that of Theodotion, and has only survived in a single Greek manuscript and the Hexaplar Syriac translation. Hence, in comparing the Septuagint with the Hebrew, it is very necessary to consider whether any given variation in the former may not be either a mistranslation or a loose paraphrase of the received Hebrew text, instead of indicating a different original.

There are, therefore, several weighty considerations which tell in favour of the Hebrew text as against the Greek, and which should, at least, prevent a hasty decision to the advantage of the latter. They are not, however, to be taken as decisive of the whole matter. There are other arguments which tend to show that, in some passages at least, the Septuagint has preserved a truer text than the Hebrew. In the first place, there is the a priori probability that the Hebrew text should have suffered some corruption in the long interval between the original composition of the several books and the earliest date at which we have evidence of the existence of our present text, especially when it is considered that, during that interval, the character of Hebrew writing had completely changed; and it is extremely improbable that the Septuagint, which branches off at a point in the middle of this interval, should never have preserved the right reading where the Hebrew has lost it. It is also certain that the Targums, the earliest parts of which date back to the beginning

beginning of the Christian era, testify to a text which, though resembling the Massoretic, is not identical with it. Further, there are several passages in which the Hebrew text simply makes no sense, and where, if the versions did not help us, it would be necessary to resort to conjecture. Additional evidence to the same effect is furnished by the Samaritan Pentateuch. The Samaritan community, having finally broken with the Jews of Jerusalem, and having set up a worship of their own on Mount Gerizim, in the days of Nehemiah, carried with them the Pentateuch; and it is practically certain that no alteration subsequently introduced into the Hebrew text would consciously be adopted by their rivals. Here, therefore, is a perfectly independent witness of very early date; and as there is no authority for supposing any connexion or reciprocal influence between it and the Septuagint, any reading in which these two agree, as against the Hebrew, has a strong claim to be considered original. Unfortunately the Samaritan Bible consisted of the Pentateuch alone, and in these books the variation between the Hebrew and the Septuagint is at its least. Most of the divergencies between the Hebrew and the Samaritan, too, are of slight consequence, often implying no real difference of text; so that the number of test passages is very small. There are, however, several of these in which the Septuagint and Samaritan combine against the Hebrew; and in such passages it is difficult not to believe that they are right.

There is, therefore, both a priori probability and actual evidence that in some passages where the Massoretic text is wrong, the Septuagint text is right. In the large majority of instances, however, direct evidence of this kind cannot be adduced. What is to be done in these cases? Can either text be said to have so far made good its superiority over the other, as to be implicitly obeyed in all doubtful passages? The older scholars and translators, from Jerome downwards, for the most part said, 'Yes. The Septuagint is full of inaccuracies, and is, moreover, only a translation. The Hebrew is the original language, and has been scrupulously preserved. The Hebrew text is the one standard to be adopted throughout, except in cases of the most obvious necessity.' Sir H. Howorth, and those who think with him, would likewise answer, 'Yes,' but in an opposite sense. They say, 'Yes. The Hebrew text is the production of a group of Jewish scholars, fired with intense animosity against Christianity, and willing to pervert their Scriptures rather than admit that the form in which the Christians quoted them was the true one. The Septuagint, on the other hand, comes down from a much earlier date, and has

preserved

preserved a purer tradition of the text. When, therefore, the Greek plainly indicates a reading different from that of the Hebrew, it is the Greek that, as a rule, is to be accepted as genuine.'

It would, no doubt, lead to a more sensational conclusion if we were to adopt the latter view, and to declare that the Old Testament used by the Christian Church for the last sixteen hundred years has been an Old Testament garbled and corrupted by the enemies of Christianity. Such a conclusion cannot, however, be adopted without very much stronger cause than has yet been shown for it. The charge of deliberate falsification, in particular, breaks down completely; or, to speak more precisely, it has never fairly stood up. If the Jewish synod at Jamnia (to which Sir H. Howorth attributes this work of fraud) aimed at cutting the ground from under the feet of the Christians by removing the Old Testament testimony to Christ, they did their work with singular inefficiency. According to him, they omitted a verse in Joshua (after ch. xv. 59), because it contained a mention of Ephratah, which is Bethlehem'; but they left untouched the prophecy of Emmanuel, and the fifty-third chapter of Isaiah, and Micah's prophecy concerning Bethlehem, and Zechariah's concerning the coming of the King, and all the other passages on which the Christians relied to show that in Jesus were fulfilled the prophecies of the Messiah. The more important variations between the Septuagint and the Hebrew touch upon no point of controversy. It would not help the Jewish doctors in their arguments to omit the inconsistent parts of the narrative of David and Goliath, or to invert the order of the prophecies of Jeremiah; and there is nothing that would tell against the Christians in the additional passages which are found in Job. Nor does this theory explain the narrower limits of the Hebrew canon. It is true that part of Wisdom might be interpreted as Messianic; but the Christian Church gave a like interpretation of the Song of Solomon, the inclusion of which in the Canon is put forth as the great achievement of the synod of Jamnia. Ecclesiasticus and Tobit were of no controversial importance; and the Jews, rather than the Christians, might be expected to magnify the claims of Judith and the history of the Maccabees. In short, the charge of fraud against the compilers of the Hebrew text and canon rests upon the flimsiest foundations, and only weakens a position which, by itself, is far more tenable.

For if the question of fraud be set aside, and the matter discussed in a temperate spirit and with the proper provisos, there is, in fact, a good case to be made out for giving greater

The

weight to the readings of the Septuagint than has hitherto been admitted. Deducting the many cases in which the variation between the Greek and the Hebrew is apparent and not real, there remains a considerable residuum in which the Septuagint is plainly based upon a text different from that which is preserved in the Massoretic revision; and in many of these passages the Septuagint reading is primâ facie preferable. Most modern scholars, though by no means all, regard the Septuagint version of Jeremiah as superior to the Hebrew, and this is perhaps the most prominent example; but it is rather in respect of single verses or passages through the whole Testament that the value of the Greek tradition is best seen. For these it is necessary to look in the commentaries on the several books, such as Professor Driver's work on Samuel, or Prof. Cornill's on Ezekiel, or in the more popular editions which summarise the conclusions of the leading scholars. It is a question, in each case, of detail, and can be settled by no sweeping measures or wide assumption of the universal superiority of one or the other text. Many possibilities must be taken into consideration. Greek translator may be loosely paraphrasing the existing Hebrew text; or he may have supplied a different set of vowelpoints to the consonants, which were, of course, alone transcribed in the MSS. of his day; or he may be removing a difficulty in the Hebrew by the use of conjecture; or he may really have had a different text before him. Each of these possibilities has to be taken into account; and this must be the work of trained scholars. The work is now going forward, and many scholars in many lands are engaged in promoting it in the only way in which it can at present be promoted, by minute study of the individual books. For the results, whatever may be our private expectations, we must possess our souls in patience. When many of the preliminary studies have been completed, and when the textual evidence has been ascertained and marshalled in a satisfactory form, then it will be time for the appearance of the great synthetic scholar, who will summarise the results, and formularise them into accepted principles. But for such conclusions the time is not yet.

One conclusion, however, and that of a reassuring nature, may be indicated in advance. Whatever may be ascertained as to the comparative originality of the Canon and text of the Septuagint and the Massoretic Hebrew respectively, there can be no ground for demanding extensive changes in the accepted Old Testament. Let it be supposed, for instance, that scholars agree that the shortened version of Job is the original one, and that the Septuagint is right in omitting several large sections