The following paper is the result of some studies prosecuted at intervals during the last winter. The views advanced are presented cautiously and not dogmatically, as the out come of those studies. They represent a comparison of the Greek and Hebrew texts, verse by verse. But when one treads through a well-nigh trackless forest, he must needs go slowly. The textual criticism of the Old Testament is largely an undeveloped field. The burning discussions that now rage concerning the contents and authorship of the various books of the Old Testament render imperative the determination, as far as possible, of the actual text of the various books.
Let it be understood that the remarks made in this article are based on and apply to the Book of Joshua simply. The editions of the Septuagint and Hebrew used were Hahn's Hebrew and Swete's Septuagint.
In order to understand rightly the condition of the facts before us, so far as known, let us go back as far as possible and follow the history of the text. For Joshua must have fared like the MSS. of the New Testament in many things.
1. There was, of course, the original Hebrew text, before the Septuagint was translated. This text must have been comparatively pure, subject to the usual errors at the hands of copyists, with the exception that the great sacredness accorded to the MS. may have secured an unusual degree of immunity from such errors by the scribes. Whether there were divergent texts that had sprung up during the centuries, we cannot say. The probabilities are against any very great differences, however.
2. Then, probably in the third century B. C., there was the translation of the Hebrew text existing then by the traditional Seventy-two. Now, this translation, not of our text now, but of the text. of that time, would transfer any errors that had crept into the Hebrew text, if it were done faithfully and without alteration.
And it may be said, whatever may be true of other portions of the Septuagint, the translation of Joshua seems to be an honest piece of work. The following statements can be made as to the character of this work.
(1). It is, as a rule, closely literal, even in the form of the sentences which often follow the Hebrew order. Greek idioms are often sacrificed to the Hebrew. It is manifestly a painstaking effort to follow slowly the Hebrew and is remarkably true to the original.
(2). Slight explanations are sometimes added that give a different coloring. In 2:1 νεκίοκους is added, reading Joshua sent two young men.
(3). There are frequent differences of a quite natural kind such as occur in New Testament MSS.; for instance in 2:10 the Hebrew has "what ye did unto the two kings of the Amorites," but the Septuagint has "what he did, etc., "referring to God. Such differences in person are common in Joshua. It would be tedious to enumerate like variations in number 10:32, voice 6:24, tense 9:21, subject 9:3, names 9:1, pronouns 3:9, and numbers 15:44. These are examples of frequent phenomena.
3. Let us go on to the Septuagint text of the fourth and fifth centuries, our text. It is simply reasonable to suppose that this text had suffered variations of the kind common to all MSS. The Septuagint may even have been more liable to such errors from careless transcribing than the Hebrew, because it received less reverence at the hands of copyists. It was not, perhaps, held so sacred. And so the usual errors will doubtless be found here as compared with the original Septuagint translation. Liberty might be taken to revise a mere translation. Of course, the Septuagint MSS. have suffered like changes since the fourth centuries until the age of printing.
(1). Hence the copyists may have added explanatory glosses not found in the original Septuagint, or the translators may have inserted such glosses not found in the Hebrew. Many additions may be explained by one or the other of these hypotheses. Compare the marginal notes that so often crept into the text of the New Testament. There are eight such additions that are of some importance, but only three of much length. Four verses are appended to 21:42 giving an account of Joshua's portion and the preservation by Joshua of the stone - knives with which he had circumcised the children of Israel. To 24:30 there is added a verse stating that these same stones were put upon Joshua's tomb. At the close of the book are added two verses about the priesthood of Phinehas and the worship of Astarte and1 Ashtaroth by the children of Israel and the subjection to Eglom, king of Moab. Now, these may all have been later traditions that crept into the text by the fourth century.
(2). There may likewise have been an occasional trans position of a passage for some reason. Two examples occur of this kind; 19:48 precedes 19:47, and 24:31 precedes 24:29. B likewise places 9:12 before 8:30.
(3). Omissions may have been made at the time of the translation or subsequently by the copyists because of the pleonasm of the Hebrew. The words, "to possess it, " in 1:11 are omitted in the Septuagint. So in 2:4, "I went not whence they were." These omissions belong to B and usually to A, but not to F, as will be shown presently.
4. But let us pass on to the Hebrew text of the fifth and sixth centuries, the text of the Massoretes. Although we have no Hebrew MSS. that reach back so far, we may presume that the present text is substantially what the Massoretes left it, when they fixed the vowel points. It is no doubt true that the very great reverence paid to the very words of the text by these scholars preserved the Hebrew from many errors incidental to other MSS. And yet the numerous traditions that had accumulated around this text and that were often regarded as of equal authority with the text it self, perhaps made the Hebrew liable to additions from the Talmud and Targums. A marginal note made by a rabbi could possibly slip into the text as we know was so often the case in the New Testament MSS. These scribes might not take anything from the text, but might add what they regarded as of equal force. There are many examples of this nature that thus admit of a forcible explanation, cases where the Hebrew has more than is in the Septuagint. A striking example is 10:13, where the sentence, "Is not this written in the book of Joshua?" is omitted in the Septuagint. In II. Sam, 1:18 Joshua is translated τοῦ εὐσοῦς. Now, may not some scribe have inserted this clause from a marginal note of inquiry? In other words, is it not possible that the Hebrew has sometimes had the same treatment as the New Testament Greek at the hands of copyists?
5. Now that we have pushed our way to the two texts as we now have them, let us institute a brief comparison between the Hebrew and the Septuagint of the present day.
(1). Both are substantially the same text, for the translator follows closely and minutely the Hebrew. And the differences can hardly be due to carelessness or wilful correction as Graf contends. If such handling of the text be true of the other parts of the Septuagint, it is not so here. And nothing seems clearer than that the work of translation in the different books of the Septuagint represents varying degrees of excellence. The slightest comparison will show that the Greek is an effort faithfully to render the Hebrew of Joshua.
(2). The Septuagint text of Joshua represents the shorter text. Apart from the eight passages mentioned above, the additions are few and unimportant. But the additions in the Hebrew as compared with the Septuagint are more numerous and stand about in the proportion of three to two. These additions are mostly words or phrases with an occasional sentence.
(3). The changes in words are less frequent, yet interesting. Some of these alterations are easily explained. In 2:11 B has ὅς for θεός. Compare I. Tim., 3:16 where θεός has displaced ὅς in many MSS. because of similarity of abbreviation. In 2:19 the phrase, "his blood shall be upon his head," is rendered by ἔνοχος ἑαιτῷ, which is more like a paraphrase than a translation.
(4). The three cases of transposition have already been mentioned. This would indicate that some liberty was taken in the order by the translator or copyist, unless it were done by the Hebrew copyist. On this matter I have no opinion.
6. Some conclusions as to the text of Joshua. It may be repeated that these remarks apply simply to Joshua and may or may not be true of other portions of the Septuagint. There were certainly many translators of the Septuagint, and one can even detect differences in style in the Septuagint text of Joshua. From the middle onward there is a less accurate rendering of the Hebrew tenses and a constant repetition of set phrases.
(1.) The MSS. of the Septuagint text of Joshua do not all represent the same type of text. The three chief documents that Swete relies on for his text are B, A, and F. B shows the evident divergence from the Hebrew and seems the most independent, and hence the most valuable for text - criticism. F is the fuller text of all, and has all that the Hebrew has and all that B has that is not in the Hebrew, with one exception of no importance. F then appears to be a complete document comprising the Hebrew and B. A is a kind of mum, agreeing now with B and now with F, and may rep resent the transition from B to F; but A agrees with B oftener than with F. So then, strictly speaking, there is no Septuagint text, but texts. The Septuagint as a whole cannot be pitted against the Hebrew. Unfortunately F stops at 12:12, for the rest of it is lost. So then the Hebrew and Septuagint present a threefold text, division (a) the Hebrew, (b) B and usually A, (c) F, and sometimes A.
May it not possibly be true that both the Hebrew and B, the most divergent texts, are somewhat too full, and that a shorter critical reading would represent the true text? If the additions in B are explanatory glosses, it is possible that some of the additions in the Hebrew, as compared with B, may be likewise due to the scribes.
(2.) Neither do the other versions agree as to the text of Joshua.
(a) The Hexapla of Origen. Here we have the He brew text of the second century which usually agrees with the Hebrew of the Massoretes in the consonants, but not always in the vowel points. Likewise the translations of Aguila and Symmachus are from the Hebrew of the second century and agree with the Hebrew of the Massoretes very closely. But the translation of Theodotion is a revision of the Septuagint of the second century with additions from the Hebrew to supply the omissions in the Septuagint. Compare F.
(b) Other versions in Walton's Polyglott and else where. The Vulgate and Syriac are from the Hebrew and represent the Hebrew of the fourth century. The Arabic and the Targum of Jonathan likewise present the Hebrew text. The Septuagint text given in Walton is that of B A. The Egyptian, Ethiopic and Armenian versions are from the Septuagint text of B A.2
(3.) We observe then the same peculiarities of text in the other versions as a whole as we found in the Septuagint and the Hebrew. Here, also, there seems to be a threefold text division. Let us now group the various authorities in question:
(a) On the one hand the Hebrew of the Massoretes, the Hebrew of Origen, the translations of Aguila and Sym machus, the Vulgate, the Arabic and Syriac versions, and the Targum of Jonathan, represent one, and on the whole the best type of text.
(b) On the other hand B A of the Septuagint the Egyptian, the Ethiopic, and the Armenian versions represent another type of text, the most divergent from the He brew. This type is most useful for purposes of text criticism.
(c) And then there is a mixed text or conflate type, represented by F of the Septuagint and the translation of Theodotion. It may be interesting to state that the authorities say that the Samaritan recension of the Pentateuch is also conflate, agreeing now with the Septuagint of B A, and now with the Hebrew.
I beg leave to submit a few remarks suggested by these studies.
1. A critical edition of the Septuagint is very much needed. The Cambridge scholars propose a larger edition with fuller critical apparatus. The Septuagint needs a new Zirchendorf to arise before any great amount of work can be safely done along these lines.
2. A critical edition of each of the different versions would add much to the working apparatus for restoring the original text.
3. It will be found, I think, that the Septuagint is not of uniform value in the matter of text criticisms. It may be of more value in Joshua than in the Psalms. Each book will need to be examined on its own merits. The sweeping
claim of Professor Workman3 that the Septuagint always represents the best texts will not stand.
4. The results of such investigation will not be startling, but will serve to restore and establish the purity of the He brew text, if any glosses have crept into it. If this critical work on the text be first done, some of the occasions that Higher Criticism claims to find for the later date of the Pentateuch and other portions of the Old Testament may be removed. Before Higher Criticism can work safely, Lower Criticism needs to find out as nearly as possible what the original text was. And then destructive criticism may find its task more difficult.
5. When the critical apparatus is ready, most of the canons for the textual criticism of the New Testament will be found of service here also, modified to suit changed conditions.
6. There will be need of patience to avoid wholesale generalizations, as if a mare's nest were found on every page, and deep reverence for the Old Testament as God's inspired word, and honesty of purpose to find out the truth free from any pet theory. What we want first is facts, and theories afterwards. Is not the textual criticism of the Old Testament of sufficient importance to justify the Christian scholars of the world in making great efforts to give it that critical completeness to which the New Testament has by laborious toil in great measure attained?
1) Gerenius supposes Ashtaroth to be images of Astarte.
2) It may be proper to state that I have reached these conclusions from personal examination of the Hebrew and Septuagint, the Vulgate, and the Hexapla of Origen and the Targum of Jonathan. For the Arabic and Syriac I had to rely on the Latin translations in Walton's Polyglott. I had to rely on the authority of the encyclopędias for the portion of the Armenian, Ethiopian, and Egyptian versions. I had the Coptic Pentateuch at hand which shows that it agrees in the main with BA. I have not compared these versions at every point with the Septuagint and Hebrew, but only enough to show the drift in each case.
3) The Text of Jeremiah. Hence I am not surprised that Professor H. P. Smith (Journal of Biblical Literature, Vol. IX.) has found so many inaccuracies in the work of Professor Workman.