At the conclusion of the Campbell Morgan Lecture for 1953 - a lecture of supreme significance for the present time - Professor R. V. G. Tasker referred to our appreciation of the light archaeological, linguistic and textual studies can throw upon the Old Testament. It is with the last of these, namely textual studies, that this lecture has to do. The greatest advance in our knowledge of the text of the Old Testament and its transmission has been brought about, not by the work of scholars, but by a chance find by Arab shepherds in 1947 of a collection of manuscripts in a cave near the Dead Sea in Palestine. The manuscript, which forms the subject of this lecture, is the one that has rightly attracted the most attention. Throughout I shall refer to it simply as the Dead Sea Scroll of Isaiah, although I am well aware that some scholars are beginning to refer to it as the Qumran scroll, from the name of an ancient settlement near the cave, but the unrivalled importance of our manuscript and the fact that it alone contains the whole book of Isaiah makes it well able to maintain its identity under any designation. I retain the title for yet another reason: out of my high regard for those American scholars who first used it and who gave the academic world the means of studying the scroll with a promptness and in an exemplary and magnanimous manner all too rare in the world of scholarship.
The nature of the find was so sensational and the claimed date of the manuscripts so incredibly early that it is now easily understandable why some scholars felt that there must somewhere be a discrepancy in the evidence and that a minute examination of the documents would bring to light some facts to lower the date to a period with some already familiar landmarks. True we had documents from pre-Christian times, but, apart from the fragmentary Greek
papyrus of Deuteronomy in the Rylands Library from the second century B.C., none of the Old Testament. The earliest manuscript of any note belonged to the tenth century. Moreover the early documents came almost exclusively from one country, Egypt, which enjoys a climate conducive to the preservation of perishable material such as papyrus and leather. The provenance of the Dead Sea Scrolls was not likely in itself to indicate for the documents a high antiquity. Nevertheless it must not be forgotten that the climate of Palestine is far from uniform, in fact, one might say there are within the confines of this small country three climatic zones: that of the coastal or maritime plain, that of the plateau or highland region reaching a maximum height of some 3,000 feet, and that of the Jordan valley, 1,300 feet below sea-level at its greatest depth. The first has a typical Mediterranean climate, the second with temperatures comparable with our own but having practically two seasons only, a dry and a rainy. The southern part of the Jordan valley, however, is semi-tropical; here it is mild in winter, hot in summer, tropical fruits, such as bananas, flourish, and here hot springs are found. It is only here that conditions, comparable with those in Egypt, obtain.
We should not lose sight of the fact that the supporters of a late date have performed an essential and salutary task. The cross-examination undertaken by them was necessitated by the fact both of the credulity and depravity of human nature. Men are prone to believe on inadequate or objectionable evidence, and the fact that on more than one occasion even the academic world has been the victim of a hoax is ample proof of sporadic depravity. The impostures of Abraham Firkovich in the 19th century in the very field of the Old Testament manuscripts stood out as an effective warning. It was therefore imperative that some scholars should act as a devil's advocate and raise every conceivable objection to the genuineness of the documents. It is fortunate that among these scholars, who constitute, as it were, the counsel for the prosecution, are men whose scholarship is riot a whit inferior to that of their opponents and whose
width of knowledge is such as to enable them to detect any flaw in the evidence. That the case has gone against them is no reflection on their skill and competence as advocates, it is rather due to the fact that they accepted a brief that was inherently indefensible. We can rest all the more assured that nothing has been left unsaid and that there has been no miscarriage of justice. Most documents whose antecedents are obscure have had to run the same gauntlet, even the Elephantine papyri did not escape. L. Belleli wrote a book of over 200 pages to prove that these documents were spurious. He based his arguments on such things as the advanced form of the script, the use of two systems of dating, resemblances in language and matter to Hebrew. Professor D. S. Margoliouth questioned the genuineness of one of the most important of the papyri. G. Jahn viewed them as an ancient forgery since they clashed with his conclusions as to the non-genuineness of the Book of Ezra, as well as justifying his rejection on linguistic grounds. The latter argument was a curious one but not without a parallel at the present time in the attitude of some scholars to the Masoretic recension of the Old Testament: that since many of the linguistic phenomena in the documents were familiar to us from a much later period, hence these papyri must be late. Whereas the obvious and logical deduction should have been that the linguistic characteristics in question must, like the papyri, be also old.
Among the objections raised to the early date of the Scrolls was the fact that guiding lines were used above the letters. From the little we know of the methods of scribes in antiquity we have no grounds for assuming that this is incompatible with an early date. It is hardly likely that the scribe in the interest both of symmetry and the best utilisation of space would fail to avail himself of the services of the ruler or some other lining device such as was used as early at least
as the third millenium. Caroline R. Williams describing the tomb of Per-Neb (c. 2700 B.C.) speaks of "lines mapping out the composition by defining the height of the dado and marking out the borders, the different registers, and the spaces for the long vertically written inscription," and again (p. 7) "That a ruler was generally used for the vertical and horizontal lines seems from their appearance unquestionable." The Kilamuwa inscription (c. 825 B.C.) has lines above the letters. A fragment of Leviticus, probably much older than the Dead Sea Scrolls, has guiding lines. Although we could not assign with certainty the meaning linear "ruler" to any word in the Old Testament, there are references in the Talmud both to the instrument and to the practice of ruling. Rabbi Minjamin said: "The ruling of the Mezuza (the parchment scroll containing Deut. 6, 4-9 and 11, 13-21 attached to the upper part of the right-hand door-post of Jewish homes) is a decree of Moses from Sinai." This is, of course, hyperbole, but it indicates that even in brief compositions ruling was regarded as a pre-requisite (s. Talmud, Menahoth 32[b]). The antiquity of ruling is vouched for also by some of the oldest Greek inscriptions which have their letters spaced with absolute accuracy, while the disc of lead used by Greek scribes for ruling lines must surely have been an indispensable part of the scribe's equipment from the earliest times.
A greater objection brought against the antiquity of the Scroll is the liberal use of vowel letters. That vowel letters have been used from ancient times is abundantly clear from the oldest inscriptions, including the Moabite stone. To deny the use of vowel letters would be tantamount to saying that at one period it was impossible to distinguish in Hebrew writing between "Give me the money" and "Give him the money," or, for that matter, between "to him" and "to her." We shall have something more to say on this point later when we come to discuss the matter of orthographical variants.
A third objection was the similarity to the Masoretic text, which is assumed by many scholars not to have acquired its present form until well into the Christian era. One scholar, Professor P. Kahle, believes the injunction in the Talmud prohibiting the retention of an uncorrected scroll was an "ad hoc" injunction, which resulted in the final elimination of divergent texts. This is not borne out by the context or by Talmudic usage. In fact there is no ground whatever for assuming that it differs from hundreds of other similarly phrased injunctions which had not a specific but a general application, valid in all periods. If this doubtful interpretation is his only evidence for the extraordinary statement that there existed circles which endeavoured to retain the literature proscribed by the Jews, "at the time of this great Jewish auto-da-fé, one of the greatest known to world history," he could not more effectively evoke our scepticism.
The existence of papyrus fragments written on both sides found in the cave led the same scholar to the conclusion that these must have been parts of codices, which he holds, did not appear in the East before the first or second centuries A.D. This assumption is by no means certain, but even if correct, there are still no cogent reasons for holding that the fragments must have come originally from a codex. On the contrary, the tenuous nature of this deduction makes it hardly acceptable on inferential grounds. The fact that the Babylonians from earliest times wrote on both sides of their tablets, and that the Egyptians used both sides of the papyrus as early as the Middle Kingdom (c. 2000 B.C.) should have made anyone wary of dubbing such a practice part of a late innovation. A practice so highly commendable on economic grounds must in literary as well as non-literary documents have appeared early. It would, moreover, be a little too much to ask anyone to believe that Ezekiel's reference (chap. 2, 10)
to a roll with writing on both sides, as the original means, was merely a subjective metaphor.
Of the many grounds for assigning an early date to the scrolls the archæological evidence is perhaps the one most difficult to refute. The jars were from the first described as Hellenistic, and although the date has been modified and the experts would now put them not in the 1st century B.C. but the 1st century A. D., since a Roman coin of this period was found later along with similar jars, this change does not affect materially the age of the scrolls, for the date of the deposit is a different problem from the date of the scrolls. Professor G. R. Driver tried to eliminate the evidence of the jars by saying the jars may have been much older than the manuscripts. There are two reasons that make this most unlikely: the jars were obviously made specifically for the purpose, and from the care in storing and sealing it is evident that a high extrinsic value was placed on the contents, hence objects of value would not readily be entrusted to old containers. In this connection it is of interest that we have here too the main reason for the rejection of Professor Sukenik's theory of the find being the contents of a Geniza (a place in or at a synagogue for the depositing of defective scrolls). When documents were relegated to the Geniza they were finally discarded and consigned to ultimate disintegration. The storing of documents in jars must have been a fairly common practice judging by the references in literature: Jeremiah 32, 14; Assumption of Moses, 1, 17, eighth column of Origen's Hexapla was found together with other Hebrew and Greek books in a jar near Jericho (A.D. 211-17); letter of Timotheus, Nestorian bishop (8th century) mentions finds of jars with manuscripts. Eight Aramaic papyri (similar in style to those of 5th century) were found in a jar at Hermopolis Magna in Egypt in 1945, and Greek papyri were found in a jar under the floor of late Coptic buildings at Elephantine.
The palaeographical evidence is of the greatest importance. True there is a paucity of epigraphical material from Palestine. The earliest examples of the use of the North Semitic Script are on a statue of a squatting figure from the Hyksos period about the 17th century B.C., and on the Gezer potsherd. The first inscription of any considerable size is the Moabite stone (King Mesha c. 842 B.C.). Then comes the Siloam inscription at the end of the 8th century. Possibly earlier than either of these but not so important is the Gezer calendar, apparently a school exercise (late 10th century?). Earlier than the Siloam inscription are the Samaria Ostraka (c. 773-736), which provide our earliest examples in Hebrew of what may be properly called "handwriting", done probably with a brush. The script of the Lachish Ostraka, nearly two centuries later (c. 597-588), has in the interval changed but little.
Of greater relevance for our present purpose are the papyri, the oldest of which dates from 515 B.C. The best known are the Aramaic papyri (5th century) from Elephantine emanating from a Jewish military colony. The Nash biblical papyrus (only a fragment) is very probably pre-Christian (according to Albright not later than first half of the 1st century). The script used is not very different from that of the Isaiah Scroll; in fact, anyone who could read the one would find little difficulty in reading the other. The epigraphical material, though meagre, is sufficient to provide unequivocal evidence of the long history, wide dispersal, and the modificatory trends of the North Semitic script, while the papyri make it abundantly clear that scripts closely resembling that of our Scroll were in use over a long period in pre-Christian times. A new impetus and importance have been given to the study of Hebrew palaeography and, though when compared with the achievements of scholars in the field of Greek studies,, much still remains to be done, its pronouncements, thanks largely to the quantities of new materials that have recently been recovered, are being received with growing respect.
Much of this evidence is not easily demonstrable, as it consists of the comparison of scripts, assigned to their respective periods, either on morphological grounds or from the evidence of collateral archæological objects, hence it is the outcome of long preoccupation and familiarity with minutiæ. Of the details that lend themselves more readily to exposition, two must suffice.
There are in Hebrew five letters which acquired later a secondary form when final in any word. In this manuscript these are still in a transitional stage, in fact of the five, two show no evidence of possessing double representation. Now there are numerous misreadings in the Septuagint that can only be explained on the assumption that the Hebrew text from which the translation was made had no special form for these letters when final. There are too in the Masorah (the compilation of critical notes by Jewish scholars on the text of the Hebrew Bible) variants in the form of wrong division of words containing one of these letters, which could not 'have occurred if final forms had existed in the early manuscripts. The copy of the Law captured in the Temple at Jerusalem, which, Josephus tells us, was carried in the triumphal procession of Vespasian and Titus, was later returned to the Jews about A.D. 220 by the Emperor Severus. About its ultimate fate we know nothing, but there is extant in rabbinical literature some information about certain of its peculiarities, thirty-two in all, about a quarter of which consisted in the use of medial for final forms. Thus at the time of the composition of this manuscript the transition to consistency in the use of final forms was virtually complete. From a remark in the Jerusalem Talmud about the three copies of the Law in the Temple we gather that they, too, did not employ final forms. The introduction of these forms must have taken place in the period between the Greek translation (3rd century B.C.) and the composition of the Severus copy of the Law (considerably earlier than A.D. 70). This evidence alone would be sufficient for assigning a pre-Christian date to our Scroll.
The second piece of evidence of early date is the shape and size of the letter yodh. When our Lord said in Matt. 5, 18, "For verily I say unto you, Till heaven and earth pass, one jot or one tittle shall in no wise pass from the law, till all be fulfilled," the indisputable inference is that the jot was then the smallest letter of the Hebrew alphabet, otherwise the statement would have little force. That this was not always so we might have inferred from its name ("hand"), implying that the original form was a pictograph of the hand, which was far from being the smallest object among the prototypes of the signs, viz, mouth, eye, and tooth. Epigraphical material, however, leaves us in no doubt that jot in size formerly rivalled a number of its fellow letters. Now in this manuscript it is still often of line height, and as large as the letter "waw", which it closely resembles, with the result that it is often hard to distinguish the one from the other. This would seem to indicate unmistakably a time prior to the new Testament era for the composition of our Scroll. Already on the Uzziah plaque (37 B.C.) the reduction of yodh to its proverbially small size has taken place.
In ancient times the human and material elements in transmission were each on occasion adverse to accurate transmission. The human element, inherently incapable of perfectibility, was often prone to fall short of accuracy in the making of copies, and the material was exposed to the ravages of time and the accidents which are the lot of all perishable things. While the latter factor has often caused major disasters involving total loss, it is probable that, taking all in all, the human has wrought the greater havoc. Certainly in the field of Old Testament transmission the fact of the preservation and existence of such a large corpus of writings would seem to vouch for the lesser evil of the material factor. While the forms that material defectiveness may take are unpredictable, human aberrancy seems to run in certain well-
defined channels, so that there is the recurrence of a group of errors irrespective of the language and script in transmission. Among the causes of these errors are the following: the writing once of a letter or word that should have been written twice (haplography); or its converse (dittography); or false recollection on the part of the scribe, as when a passage suggests something and he writes not what is visibly present but what is mentally present; or the double occurrence of the same word in a passage may cause the scribe's eye to wander. from the first to the second, resulting in the omission of the intervening words (homoteleuton); the omission of a whole line or a number of consecutive lines (sometimes in conjunction with homoteleuton); confusion of letters similar in form; change in the order of words (often through re-insertion into the text of a word which stood as a correction above the line or in the margin of the model). It might seem that texts were exposed to progressive deterioration at a rate of say two per cent in the first copy and so on, and that the text would soon be beyond repair. As a matter of fact there is no such thing as compound deterioration, for certain preservative qualities counteracted the tendency: grammar, context of the passage, and the common sense of the scribe (In the transcription of Hebrew texts it is very improbable that there ever was mechanical copying by scribes unfamiliar with the language). The Scroll has not been spared the fate common to all transmitted texts, and all these transcriptional errors are present to a limited extent in our manuscript. The existence of errors in texts and the need to rectify them gave rise to the discipline we call Textual Criticism, which has as its object the reconstruction of a text that approximates most closely to the original or autograph. Such a critical recension aims at providing a text in the form intended by the author. To this end emendations may, on occasion, be necessary, but they should be confined strictly to their proper sphere: "the removal of the absolutely vicious."
Large numbers of the variants in our Scroll (it would be misleading to call them variant readings) are by their nature void of significance. The mere counting of variants is the unmistakable badge of the tiro in the field of textual criticism. Variants go by weight not by number, they are evaluated not enumerated. To the class of insignificant variants belong in the first place the orthographical, that is, those that involve differences in spelling only. Such variants surprise only print-conscious readers, prone to forget the vicissitudes of their own spelling until the printing-presses imposed on them the present mechanical uniformity. In a modern text there is not much grist of this kind left for our above mentioned tyro (his identity has remained unchanged, despite the change, deliberate but still correct, in spelling). Both "Ihoauerd" and "louerd" would seem to us now outlandish modes of spelling "lord", but they evidently did not perplex a man of the thirteenth century. Of all such variants textual criticism takes little or no cognizance. The lavish use of vowel-letters (consonants used to indicate vowels; Hebrew script originally possessed no special signs for these) contributes largely to the multiplication of such variants. On this point some scholars seem to have completely lost sight of the fact that vowel-letters are in origin not intrusive but residual: they arose in the first instance through certain of these letters losing their consonantal value, this in turn leading to changes in the vowel-pattern; for instance, through the crasis of the vowels thus brought into contact, disyllables emerged as monosyllables. The spelling with the retention of the "extinct" consonant was hence the product of etymology and not of phonetics. Thus began a system that later could acquire, often disregarding the dictates of philology, the dimensions we now see in our present manuscript.
We can eliminate on the score that they too are devoid of significance, those variant forms that give practically, if not precisely, the same meaning as the forms which they replace. Under this heading come in the first place synonyms or near synonyms. In Chap. 37, 32, for instance, Zion stands for
Jerusalem. In 13, 10 the verb "to give light", the more familiar one, derived from the noun "light", replaces one not so common. Other examples taken at random are (for the benefit of those who do not read Hebrew I give in the brackets the synonyms found in the standard text, in the A.V. translation): 13, 10, lighten (give light); 24, 1, ground (earth); 39, 2, kingdom (dominion); 42, 20, unclose (open, either verb is unusual with ears, but the Scroll agrees with the choice of 50, 5); 43, 21, say (show forth, lit. recount); 43, 23, make (serve); 45, 7, good (peace); 51, 2, make fruitful (bless); 62, 1, be silent (hold peace); 65, 5, touch (approach). Related to this class are what might be called synonymous idioms. These occur through the exchange of convertible idiomatic forms of expression of the type, "he says", "they say", or "it is said", all identical in meaning. Not differing greatly from these are what we might designate idiomatic variants. Under this head would come the use of a singular for a plural or vice versa (either with nouns or verbal forms), for Hebrew in common with other Semitic languages made wide use of the singular as a collective, known in philological jargon as "generic singulars".
The additions to the text do not seem to raise any insoluble problems. Again the overwhelming majority are of a minor character, consisting in such things as the insertion of the conjunction "and" (one letter in Hebrew), and other particles. Some of the larger ones are explicable on mnemonic grounds, as for instance, the addition at the end of verse 15 of chapter 1, "your fingers with iniquity", a phrase familiar to the scribe from chap. 59, 3. In chap. 4, 2 after "Israel" is added "and Judah". The phrase, "and for the sake of thy servant David", is inserted at the end of 38, 6. To chap. 51, 3 is added "sorrow and sighing will flee", obviously anticipating verse 11. In chap. 21, 16 "year" is in the plural preceded by "three". In chap. 30, 6 between the words "trouble" and "anguish" appears the word "drought". In chap. 53, 11 the word "light" is added after "he shall see". The longest addition of all is that in chap. 38, 20. This was caused by the scribe's eye wandering back after he had almost finished a line
to the same position on the line above, thus leading to a repetition of the whole line. Frequently where the word of the scroll actually differs from that of the Masoretic text it involves only the change, addition, or omission of a single consonant. When such changes come to be translated into English they may easily give an exaggerated impression as to the real extent of the discrepancy. In most of the cases it is impossible to decide whether the change is merely a slip (say, lapsus calami) or intentional. In view of the fact that often the similarity in form of the letters renders confusion easy, most of the changes are probably not deliberate but purely accidental. How slight the difference between certain pairs of letters really is, might best be conveyed to a non-Hebraist by saying that one looks something like the sanserif form of the other. In 29, 3 "forts" seems to be written as "nets" through the confusion of two letters possessing even closer similarity. The same two letters are involved in the change in verse 5 of the same chapter, where the reading "godless" for "strangers" occurs. It is very doubtful, however, whether in handwriting such divergences would have been registered by a reader whose mother tongue was Hebrew. There are instances where the scroll may have preserved the correct form: in 33, 8, the reading is "witnesses" as against "cities". Examples of the effect of the omission of a consonant are: chap. 2!, i word (desert); in 41, 5 the verbal form "were afraid" is reduced to the adverb "together". As an example of the effect of an additional jot, "water" in chap. 40, 12 becomes "waters of the sea".
Omissions consisting of one or more words and constituting discrepancies on a considerable scale, create, at first sight, what seems to be the greatest gap between the text of the Scroll and the textus receptus. Doublets, that is, the repetition of a word or phrase as a stylistic device or for some other reason, are usually lopped of one member. It would seem as if our copyist had a rooted aversion to them. In chap. 6, 2 where the Hebrew expresses distribution by the simple expedient of repetition, as "six wings, six wings", the Scroll dispenses with the repetition. In the next verse (3), the copyist is ready to
write "holy" twice, but seems to demur at doing it a third time. In chap. 9, 9 "gird yourselves and be ye broken" stands only once. In chap, 35, 8, "and a way", which immediately precedes in the Masoretic text an identical form, occurs but once. In chap. 37, 29 "thy rage against me" is written only once although the phrases are not adjacent, and the same treatment is meted out, under similar circumstances, to the phrase "saith the Lord" in chap. 59, 21.
Some of the shorter omissions are clearly cases of homoeoteleuton, in chap. 2, 3 where the copyist's eye skipped from the first "to" to the second, thus excising the phrase "to the mountain of the Lord". The absence of the words "therefore thy raging against me" was almost certainly due to the double occurrence of the preposition "to" with pronominal suffix "me". The omissions in chap. 45, 8 and 51, 6 occur at blanks in the manuscript, where either the text was defaced or illegible. The copyist has attempted apparently to remedy the defect by inserting in the latter passage: "And see, who has created all these", but he has left in his text evidence that this was not the original occupant, for the "in like manner" demands the presence of a preceding simile.
The small group of major omissions, so forbidding at first sight, proves on better acquaintance to be reasonably amenable. The first omission is in chapter 2, part of verses 9 and 10 (therefore, forgive them not. Enter into the rock, and hide thee in the dust, for fear of the Lord, and for the glory of his majesty). There is no evidence here that homoeoteleuton has been responsible. The only phrase in the immediate context and its repetition (with a slight variation) that might have distracted the scribe remains in the Scroll. Was this wording then actually in the model used by the copyist? The fact that the Septuagint gives the passage in full would not in itself be decisive if the context and sense did not betray incompleteness and provide some tangible explanation of its absence. When we study the extent of the gap we find that in all the text has lost 45 letters as compared with the Masoretic. From a rough estimate on the basis of the intact lines in Plate II, we find the average number of letters per line to be 43 hence 45 letters
represent a normal line. This then must be a case of "line omission", a phenonenon familiar to us from classical manuscripts. The simple explanation of such occurrences would seem to be that whereas the copyist retained the place horizontally in the manuscript, he lost it vertically. This would in our case involve a drop of as little as a third of an inch. Such a move laterally would have meant the loss of not more than three letters, but the same displacement vertically meant the loss of a complete line. The incontrovertible proof that this is no mere supposition, is provided by the example from chap. 38, 20, mentioned above (p. 12 f.). Here we have the same phenomenon involving identical steps but in this case in reverse. This particular variety is rare for it indicates greater negligence. A scribe who could commit the one, could a fortiori do the other. It will be seen that in chap. 38, 20, one and the same line is not involved but two parts of consecutive lines, the normal form of the error. The same example also shows that these may occur conjointly with the vertical slip a small lateral movement. A. C. Clark describes it as a looking forward and passing from one line to another. Among the numerous examples collected by him is one where the point of departure is right in the middle of a word (p. 9, no. 44).
The next example comes from chap. 4, part of verses 5 and 6 (and smoke, and the shining of a flaming fire by night: for upon all the glory shall be a defence. And there shall be a tabernacle for a shadow in the daytime). The Septuagint gives the passage with one minor omission. Here quite clearly is a case of homoeoteleuton, the copyist's eye sprang from "by day" to "by day" (A.V. "by day" and "daytime"). The length of the omission, however, is unusually great to have been caused solely in this way, and one suspects a contributory factor. This could have been line omission, since the number of letters in question is 46. This page's average for the lines which are complete is a little under 4s. We now come to a lacuna, larger than either of the preceding examples, in chap. 16, part of verse 8 and 9 ("the lords" to "vine of Sibmah").
The Septuagint shows no evidence of diminution in quantity but the translation seems to have gone awry. Clearly here, too, homoeoteleuton is present in the recurrence of the phrase "vine of Sibmah". Again, the excessive extent of the gap suggests the presence of a secondary cause, and line omission is the one that lies at hand, but here not one line but two were dropped, for a minimum of 75 letters is missing, say two lines of 38 and 37 letters, feasible quantities in a context with an average of just under 41 In the next example the order of the words in Hebrew do not lend themselves to homoeoteleuton. This example occurs in chap. 23, part of verse 15, (Tyre shall be forgotten seventy years, according to the days of one king, after the end of seventy years shall be). The Septuagint agrees almost verbatim with the Masoretic text. The number of missing letters (maximum 41) fits in well in a page averaging about 36 letters per line and possessing eight lines with 40 or a little over. The next example (chap. 40, part of verse 7 and 8), is again clearly a case of both homoeoteleuton and line omission. The number of letters missing is 40, and the average per line being 43. The line is on the short side, but it has only one letter less than line 10. This fortuitously well-laid trap must have caused more than one copyist to stumble, the same passage is missing from the Septuagint: either it was absent from the Hebrew text used, or the translator himself fell into the trap. Among these major omissions the one in chap. 63, part 3, is unusually small (34 or possibly allowing 'for alternative spellings, 36). Nevertheless line omission is the simplest explanation. The passage was certainly in the text used for the Septuagint. When, however, we distribute these letters, say i6 to complete line 29, and the remaining 18 letters or so joining the now displaced second half of line 29 to form line 30, we have two of 42 and 46 letters respectively, not out of the
question in a context where long lines predominate. (In chap. 65, 16 where some 34 letters are missing it looks as if a whole line had been obliterated from the manuscript and the copyist met with no success in trying to decipher a few of the words. In chap. 54, 17 the manuscript must have suffered in a similar fashion but here the first word of the line remained legible, the rest is left blank). Two things should by now be reasonably clear. First, in showing that the discrepancies bear all the marks of a familiar form of transcriptional error, we must draw the inevitable conclusion that they were not a permanent feature of the texts that constituted the line of transmission from which the text of the Scroll sprang. This is of the highest importance for our knowledge of the characteristics of the family to which our text belongs. Second, less important but nevertheless of considerable interest, the text used as a model had lines of equivalent length to those of our text.
Before we proceed to the matter of the descent of our manuscript it might be apposite to say something about copying in general. About the rules and methods that governed scribes we know little and that only through references in the Talmud. That the method was ocular, that is, from another text and not from dictation by a fellow scribe, seems certain from the concession that the Tefillim and Mezuzoth might be written without consulting a copy (Men. 32b). It was, however, forbidden to write the Law from memory without an exemplar. This would imply that there were Jewish scribes from early times who, like the hafiz of Islam of a much later period, knew the Law by heart.
While we know comparatively little about copying among the Jews, there is considerable information about the
standards that obtained in the copying of texts in the Ancient East generally. It is now axiomatic to say that from ancient times men could write, but with that goes a corollary the full significance of which I for one have only recently realized: men from ancient times could not only write, they could also copy. Assurbanipal (7th century) tells us he set up, for the purpose of his private study, a library in his palace. He maintained a large staff of scribes whose responsibility it was to make copies of documents of every kind and from many places (learning to copy was an important part of a scribe's training, which lasted from early youth to manhood). That the servants of such a scholarly master would not have dared take liberties with their texts, goes without saying. The correctness of this conclusion is attested by the high standard of the copies of parts of the law code of Hammurabi which have come down us with many of the other documents from his famous Iibrary at Nineveh. The art of copying had, of course, been perfected much earlier than this, in fact the authoritative text of life Code of Hammurabi on the famous stele is itself self-evidently a copy, made not by a ready scribe but by a lapidary from a written document, in this case, a clay tablet The virtual absence even of grammatical mistakes bears eloquent testimony to the skill of the creator of this almost perfect copy from 1750 B.C.
Ancient Egyptians not only practised copying but were able early to enunciate clearly the essentials of accurate copying. These have never anywhere been better formulated than in a text from the tomb of Iuya (c. 1400 B.C.). It reads: "[The book] is completed from its beginning to its
end as it was found written, having been copied, revised, compared, and verified sign by sign."
It would indeed be passing strange if there never arose in Israel men who loved letters and had a solicitude for the preservation of their literary heritage and that Hebrew scribes alone were without conscience in the matter of transmitting texts. While paganism could produce an Assurbanipal, patron of learning and of letters, devoted to his "precious tablets" as he called them, and a Shabaka with his enthusiasm to perpetuate in a worthy and lasting form the revered literary work of his ancestors, the Jews despite their high ethical principles are alleged to have allowed their sacred writings to be mutilated, chopped and changed at the whim of every unscrupulous, nondescript scribe, Bowdler, or Verballhorn. Duhm, in his day considered by many the greatest authority on Isaiah, believed the collecting of the "bogus bones" of the prophet did not finally cease until long after the time when, as we now know, our manuscript lay complete.
The class of manuscript about which we know most is the liturgical or hieratic, that is those intended for use in Temple or synagogue. That there must have existed demotic or lay copies we could have surmised but, hitherto it has been felt that none was likely to have survived. The line of transmission to which they belonged would have represented an early embranchment, and any representative of this line would have given us an independent witness to the reliability or otherwise of our transmitted text. Our Scroll is such a witness, and herein lies its unrivalled importance. It may come as a surprise to a layman to learn that this unimpressive looking document can make a contribution to our knowledge greater far than that of the finest illuminated liturgical specimen. Not even the discovery of the Mugah scroll, or the Hilleli, or the Great Machsor, or even the Severus, prized as booty by Titus, all of them, doubtless, examples of superb craftsmanship
and rich beauty, whose names are mentioned with respectful awe by the rabbis, could have taken the place of our Scroll as a crucial witness. True our witness is a little shabbily and carelessly dressed and he stutters somewhat in his speech, but he has not been briefed for his part, as would have been affirmed of any of the others, being products of a hieratic guild. When our Manuscript was completed it was given no imprimatur. It never heard the words "Turn or burn", It was a layman's copy, the cost of which must have been trivial. It would seem to have been the work of two scribes (there is a break exactly half-way through. representing. in all probability, the place of the equal distribution of labour). The work would not entail more than 24 hours copying. two days' work, and as scribes would certainly not be paid more than workers in the vineyard, the cost of labour would come to 2 denarii. The skins, in a pastoral country, would be cheap, say another denarius, making in all three shillings and four-pence, or fifty-one cents.
The Scroll carries with it certain information about its antecedents, enabling us to affirm that it bore neither to the autograph nor to the liturgical line of transmission an immediate relationship. The evidence for this are the inversions or transpositions of words, in all in some 18 places. Such changes in word-order are, as we said earlier, due to the wrong re-insertion of omitted words which stood above the line or in the margin of the model. Hence we have the evidence of the existence of another manuscript holding an intermediate position between our text and the autograph, or, what is more likely, between our text and the archetype which was the common ancestor of both the liturgical and lay families. It is now impossible even to guess at the age of the manuscript which served our Scroll as its immediate model, but in this matter it would surely not be unreasonable to suppose that an older rather than a later manuscript would be chosen. Beyond this again stood the archetype at which the lines divided. Where this position, was relative to the autograph, and how many stages intervened between them will never be known. It may seem to the uninitiated that material for corroboratory
evidence limited to two streams is somewhat scanty. As a matter of fact, extraordinary as it may appear, bifurcation seems to dominate the field of transmission of classical texts almost exclusively, so that the acquisition of one single manuscript has now raised the textual criticism of the Old Testament to equality.
The very discussion of such questions indicates how complete is the re-orientation which has taken place. We no longer ask what is the relation of our Manuscript to a recension of 100 A.D. Our Scroll takes us so far beyond this point, that questions, which a short while ago held a central place in the problem of transmission, have now little more than antiquarian interest. This manuscript has added not only one new and earlier point, on the line of transmission, it has indirectly provided still another two: that of its model, and that of the archetype from which came the liturgical and the lay families of manuscripts. There is now nothing to prevent anyone who feels so inclined from believing that if this line were projected backwards it would end in an autograph similar in all essentials to the text that has been transmitted to us. Sir Frederick Kenyon, that great scholar, whose range of vision in the field of manuscripts was unequalled, indicated unerringly the central problem when he said: "The great, indeed all-important, question which now meets us is this - Does this Hebrew text, which we call Massoretic, and which we have shown to descend from a text drawn up about A.D. 100 faithfully represent the Hebrew text as originally written by the authors of the Old Testament books?" He believed even then that an affirmative answer was possible. Little did he or anyone dream that a day would come when a witness of such ancient lineage and high credentials would appear with evidence to convince many that his question will no longer brook the answer no.
 M. Burrows, J. C. Trever and W. H. Brownlee. The Dead Sea Scrolls of St. Mark's Monastery. Vol. 1. New Haven, 1950.
 L. Belleli. An Independent Examination of the Assuan and Elephantine Aramaic Papyri, London 1909.
 P. Kahle, Die Hebræischen Handschriften aus der Hhle, Stuttgart, 1951, p. 58.
 S. Franz Rosenthal. Die Aramaistische Forschung. Leiden, 1939, p. 31f.
 G. R. Driver. Letter to The Times, 30th August, 1949.
 C. R. Williams. The Decoration of the Tomb of Per-Neb. New York, 1932, p. 6.
 Bulletin of the American Schools of Oriental Research, 1950, p. 20f.
 Kahle, op. cit. p. 27.
 Kahle, op. cit. p. 60.
 Kahle, op. cit. p. 58.
 J. E. Sandys (ed.), Companion to Latin Studies, 1913, p. 238.
 J..Cerny. Paper and Books in Ancient Egypt. London, 1912, p. 34, note 96.
 The Times, 23rd August, 1949.
 In Akkadian the name for such a jar (gerginakku) seems later to have been given to the whole library.
 Emil G. Kraeling. The Brooklyn Museum Aramaic Papyri, 1953, pp. 18., 66.
 H. Gressmann. Altorientalische Bilder zum Alten Testament, 1927, p. 194 and Plate cclix.
 For relevant literature see F. M. Cross and D. N. Freedman, Early Hebrew Orthography, 1952, p. 8.
 Bulletin of American Schools of Oriental Research, 1949, p. 19.
 Ginsburg. Introduction to the Massoretico-Critical Edition of the Hebrew Bible (London 1897), 411ff.
 E. L. Sukenik, Megilloth Genuzoth, Jerusalem 5948, Plate IV (opp. p. 15).
 J. E. Sandys (ed.), Companion to Latin Studies, Cambridge, 1913, p. 792.
 A. C. Clark, The Descent of Manuscripts, Oxford 1918, p. 9.
 Op. cit. p. 18ff.
 On the potential fecundity of homoteleuton s. P: Maas, Textkritik, Leipzig, 1950, p. 29.
 s. A. Rahlfs, Septuaginta, Stuttgart 1935, p. XXIX (s. also critical notes on passage, p. 659: Origen in his Hexapla, c. 275 A.D. indicates its omission from his edition of the LXX, but found in the Hebrew; similarly, the presbyter Lucian in his revision, c. last quarter of third century).
 Chap. 34, 17 to chap. 35, 2 ("they shall possess" up to "excellency of our God") is written in a much smaller hand, as if the scribe had found himself with three short lines but only space for a little more than one and got out of the difficulty by compressing his writing and telescoping the lines. Chap. 37, 6 and 7 may be another instance of the same procedure, in this case two longish lines followed by a very short one; the last two would then have been telescoped. In chap. 38, 21 a normal line (say 40) and a short one (say 25) had to be inserted in the space left for one line. All these cases may have originated in material defect in the model, the scribe leaving a blank until he could consult another manuscript that was intact and easily legible at this point.
 W. Eilers. Die Gesetzesstele Chammurabis, Leipzig, 5932, p. 5, n. 2.
 "King Shabaka (c. 710) in his preamble to his copy on stone of the document of Memphitic Theology, (possibly from the 4th or 5th dynasty) says: "His Majesty copied this book anew in the house of his father Ptah, south of his wall. His Majesty had found (it) as a work of his ancestors having been eaten by worms. It was not known from beginning to end. Then His Majesty copied it anew, so that (it) is finer than it was earlier, so that his name might endure and his monument might be established in the house of his father Ptah, south of his wall." (Kurt Sethe. Dramatische Texte zu Altaegyptischen Mysterienspielen, Leipzig, 1928, p. 20).
 E. Naville, The Funeral Papyrus of Iouiya, London, 1908. Pl. XXXIII.
 J. Cerny, op. cit. p. 25 and ref. 131.
 Cuneiform Texts from Babylonian Tablets in the British Museum. (the King's name is not given but the context indicates Assurbanipal) 1906. Part XXII. Plate 1.
 Carl Cornhill. Introduction to the Canonical Books of the Old Testament, London, 1907. p. 265.
 J. Bedier. La tradition manuscrite du Lai de l'Ombre. Romania 54, 1928. 161ff, 321ff.
 On transmission in general see Paul Maas, Textkritik, Leipzig 1950, and on bifurcation in particular, op. cit. p. 30.
 Sir Frederick Kenyon. Our Bible and the Ancient Manuscripts. London, 1939, p. 47.