Please Support the Bible Translation Work of the Updated American Standard Version (UASV)
Texts and versions provide the raw materials for the discipline known as textual criticism. The ultimate aim is to provide a text in the form intended by its author. Generally speaking, the greater the age of a document, the greater is its authority. There may be cases, however, where this does not hold; for instance, of two MSS, the older may have been copied from a recent and poor exemplar, while the other goes back to a very much earlier and better one. The history of a document must be taken into consideration before a verdict can be given on readings. Documents are exposed to the ravages of time and the frailty of human nature, and the latter gives rise to most of our problems. The errors of scribes, however, seem to run in well-defined channels. Among common errors are:
haplography (failure to repeat a letter or word); 2. dittography (repeating what occurs only once); 3. false recollection (of a similar passage or of another ms); 4. homoeoteleuton (omission of a passage between identical words); 5. line omission (sometimes through homoeoteleuton); 6. confusion of letters of similar form; 7. insertion into body of text of marginal notes. The comparative study of texts can help towards the elimination of corruptions. Here numerical preponderance is not decisive: several representatives of the same archetype count as only one witness. The form of textual transmission is best depicted as a genealogical tree, and the facts of the genealogical relations can be applied to the assessment of evidence for any given reading.
OLD TESTAMENT: HEBREW
The documentary evidence for the OT text consists of Heb. MSS from 3rd century bc to 12th century ad, and ancient versions in Aramaic, Greek, Syriac and Latin.
From the earliest times, the Jews had at their disposal the means of producing written records. The Semitic alphabet was in existence long before the time of Moses (*Writing). Moses would have been familiar with Egyp. writing and literary methods. He may, too, have been acquainted with cuneiform, for the el-*Amarna and other letters show that Akkadian was widely used during the 15th to 13th centuries bc as a diplomatic language. If the Bible did not expressly state that Moses was literate (Nu. 33:2 and passim), we should be compelled to infer it from collateral evidence. There is, therefore, no need to postulate a period of oral tradition. Analogies drawn from peoples of disparate culture, even if contemporary, are irrelevant. The fact is, that the peoples of the same cultural background as the Hebrews were literate from the 4th millennium bc onwards, and from the 3rd millennium men were being trained not merely as scribes but as expert copyists. It is unlikely that under Moses the Hebrews were less advanced than their contemporaries or that they were less scrupulous in the transmission of their texts than the Egyptians and Babylonians (cf. W. J. Martin, Dead Sea Scroll of Isaiah, 1954, pp. 18f.).
Before describing the sources at our disposal for the restoration of the text of the OT, it is important to recall the attitude of the Jews to their Scriptures. It can best be summed up in the statement by Josephus: ‘We have given practical proof of our reverence for our own Scriptures. For, although such long ages have now passed, no one has ventured either to add, or to remove, or to alter a syllable; and it is an instinct with every Jew, from the day of his birth, to regard them as the decrees of God, to abide by them, and, if need be, cheerfully to die for them. Time and again ere now the sight has been witnessed of prisoners enduring tortures and death in every form in the theatres, rather than utter a single word against the laws and the allied documents’ (Against Apion 1. 42f.).
That Josephus is merely expressing the attitude of the biblical writers themselves is clear from such passages as Dt. 4:2 (‘You shall not add to the word which I command you, nor take from it, that you may keep the commandments of the Lord your God which I command you’) or Jer. 26:2 (‘… all the words which I command you to speak to them; do not hold back a word’). There is no reason to suppose that the Jews ever abandoned these principles. Many of the divergences in texts may be due to the practice of employing the same scribes to copy both biblical texts and Targums. As the latter are frequently paraphrastic in their treatment of the text, this laxity could subconsciously easily affect the copyists.
The Transmission of the Text
Measures for the preservation of the text were already in use in the pre-Christian era, for in the Dead Sea Scroll of Isaiah (e.g. plate XXIX, lines 3 and 10) dots are put over doubtful words, just as the Massoretes did later. In NT times the scribes were too well established to be a recent innovation. It was doubtless due to their activity that terms such as *‘jot’ and ‘tittle’ owed their currency. The Talmud states that these scribes were called sōp̱e rîm because they counted the letters in the Torah (Qiddushin 30a). Since their intensive preoccupation with the text of Scripture qualified them as exegetes and educationalists, the transmission of the text ceased to be regarded as their prime responsibility.
The writing of the consonants only was sufficient as long as Heb. remained a spoken language. Where a word might be ambiguous ‘vowel-letters’ could be used to make the reading clear. These ‘vowel-indicators’ were in origin residual: they arose through ‘waw’ (w) and ‘yod’ (y) amalgamating with a preceding vowel and losing their consonantal identity, but they continued to be written, and in time came to be treated as representing long vowels. Their use was then extended to other words, where etymologically they were intrusive. Their inscription or omission was largely discretionary. Consequent variants have no significance. It was not until about the 7th century of our era that the Massoretes introduced a complete system of vowel-signs.
The Massoretes (lit. ‘transmitters’) succeeded the old scribes (sōp̱e rîm) as the custodians of the sacred text. They were active from about ad 500 to 1000. The textual apparatus introduced by them is probably the most complete of its kind ever to be used. Long before their time, of course, others had given much thought to the preservation of the purity of the text. Rabbi Akiba, who died about ad 135, was credited with the saying, ‘The (accurate) transmission is a fence for the Torah.’ He stressed the importance of preserving even the smallest letter. In this he was by no means the first, as the statement in Mt. 5:18 shows: ‘Till heaven and earth pass away, not an iota, not a dot, will pass from the law, until all is accomplished.’ The Massoretes introduced vowel-signs and punctuation or accentual marks into the consonantal text. Three systems of vocalization had been developed: two supralinear (Babylonian and Palestinian) and one infralinear, except for one sign. This system, called the Tiberian, supplanted the other two, and is the one now used in Heb. texts.
As it was the resolute purpose of the Massoretes to hand on the text as they had received it, they left the consonantal text unchanged. Where they felt that corrections or improvements should be made, they placed these in the margin. Here the word preferred and which they intended to be read (called the Qerē’, ‘that which is to be read’) was placed in the margin, but its vowels were placed under the consonants of the word in the inviolable text (called the Keṯîḇ, ‘the written’). It is possible that a form given in the margin (Qerē’) was sometimes a variant reading. The view held in some quarters that the scribes or Massoretes boggled at giving variant readings, and in fact deliberately suppressed them, is contrary to what we know of the actual practice of the copyists.
The Massoretes retained, for instance, certain marks or the earlier scribes relating to doubtful words and listed certain of their conjectures (seḇîrîn). They used every imaginable safeguard, no matter how cumbersome or laborious, to ensure the accurate transmission of the text. The number of letters in a book was counted and its middle letter was given. Similarly with the words, and again the middle word of the book was noted. They collected any peculiarities in spelling or in the forms or positions of letters. They recorded the number of times a particular word or phrase occurred. Among the many lists they drew up is one containing the words that occur only twice in OT. Their lists finally included all orthographic peculiarities of the text.
The textual notes supplied by the Massoretes are called the Massorah. The shorter notes placed in the margin of the codices are referred to as the Massorah Parva. They were later enlarged and arranged into lists and placed at the top or bottom of the page. In this form, they were called Massorah Magna. This fuller form may give, for instance, the references to the passages where a certain form occurs, whereas the shorter would give only the number of the occurrences. The notes provide the results of their analysis or textual peculiarities. They give variant readings from recognized codices, such as the Mugah and Hilleli (both now lost).
Among the names of Massoretes known to us is that of Aaron ben Asher, who was active in the first half of the 10th century ad. Five generations of his family seem to have worked on the Heb. text, and under Aaron the work reached a definitive stage. The best codex of this school is thought to be the one formerly in Aleppo, now in Israel. Another noted family of Massoretes was that of ben Naphtali, one of whom was apparently contemporary with Aaron ben Asher. The differences between them in their treatment of the text was largely confined to matters of Vocalization. The ‘Reuchlin’ codex in Karlsruhe is a representative of the ben Naphtali approach.
The text edited by Jacob ben Chayyim for the second rabbinic Bible published by Daniel Bomberg in Venice in 1524–5 came to be accepted practically as a standard text. The text was eclectic in character, and scholars have been aware for some 250 years that it could be improved. It is significant, however, that M. D. Cassuto, a scholar who probably had a finer sense for Heb. than any other in this field, and who had an unrivalled knowledge at first hand of the Aleppo Ben Asher codex, evidently saw no reason for preferring this to the Ben Chayyim text, which he retained for his fine edition of the Hebrew Bible (Jerusalem, 1953). The non-expert might easily be misled by the somewhat hyperbolic language used of the extent of the differences to be found in the various MSS. They relate mostly to matters of vocalization, a not altogether indispensable aid in Semitic languages. Linguistically considered they are largely irrelevant minutiae, at the most of diachronistic interest. Belief in the golden age of the phoneme dies hard; it ranks with the naïveti that believes ‘honour’ is a better spelling than ‘honor’. Vocalization in a Semitic language belongs primarily to orthography and grammar, and to exegesis, and only to a limited extent to textual criticism. There never was an original vocalized text to restore. It is clear that the Massoretic text is a single type which became recognized as authoritative after the Fall of Jerusalem in ad 70. All Hebrew Bible fragments found with relics of the Second Revolt (ad 132–135) in caves near the Dead Sea belong to it, in contrast to the situation at Qumran before ad 70 (see III, below).
The Dead Sea Scrolls
The discovery of biblical MSS, in caves to the W of the Dead Sea, has revolutionized the approach to the OT text by going some 800 years behind the Massoretic apparatus. It has also been a salutary reminder that the purpose of the discipline is the restoration of a consonantal text. The original find included one complete ms of Isaiah and another containing about one-third of the book. The later discoveries brought to light fragments of every book of the Bible, with the exception of Esther, as well as Bible commentaries and works of a religious nature.
The Dead Sea biblical MSS give us for the first time examples of Heb. texts from pre-Christian times, about 1,000 years earlier than our oldest MSS; thus they take us behind the alleged suppression of all divergent texts in ad 100. According to the Talmud, an attempt was made to provide a standard text with the help of three Scrolls formerly belonging to the Temple, by taking in cases of disagreement the reading that had the support of two (TJ, Ta‘anith 4. 2; Soferim 6. 4; Sifre 356). The finds have helped to relegate questions of vocalization to their proper sphere, that of orthography and grammar, and have deprived of much of its pertinency the work done in the field of Massoretic studies by providing us with MSS much older than any hitherto at our disposal.
The Isaiah MSS provide us with a great variety of scribal errors, but all of them familiar to textual criticism. We find examples of haplography, dittography, harmonization (i.e. alteration to something more familiar), confusion of letters, homoeoteleuton, line omission and introduction into the text of marginal notes.
The great significance of these MSS is that they constitute an independent witness to the reliability of the transmission of our accepted text. There is no reason whatever to believe that the Qumran community would collaborate with the leaders in Jerusalem in adhering to any particular recension. They carry us back to an earlier point on the line of transmission, to the common ancestor of the great Temple scrolls and the unsophisticated scrolls from Qumran. Beside MSS close to the MT, fragments of others display Heb. texts that differ. Until all the material is published, it is hard to evaluate them; by their nature these have attracted most attention (see *Dead Sea Scrolls for details). That any are generally superior to MT or represent an older text is questionable; each passage has to be considered separately in the light of known scribal customs.
The Cairo Genizah
The MSS discovered from 1890 onwards in the Genizah of the Old Synagogue in Cairo are of considerable importance for the vocalized text. (A Genizah was the depository for scrolls no longer considered fit for use.) The lack of uniformity in vocalization and the virtual absence of variations from the consonantal text show that the vocalization was secondary. Among the fragments of biblical MSS from this Genizah are some with supralinear vowel-signs. In the collection were also quantities of fragments of Targum and of rabbinic literature. Some of the MSS may be older than The 9th century.
The Hebrew Pentateuch of the Samaritans
The Heb. Pentateuch preserved by the *Samaritans is unquestionably derived from a very ancient text. The Samaritans, probably the descendants of the mixed population of Samaria, the result of a partial deportation of Jews by Sargon in 721 bc, followed by the plantation of foreigners (cf. 2 Ki. 17:24; 24:15–16), were refused a share in the rebuilding of the Temple by the Jews returning under Ezra and Nehemiah. The breach which followed (probably in the time of Nehemiah, c. 445 bc) led to the establishment of a separate Samaritan cultic centre at Mt *Gerizim, near Shechem. Contacts between the two communities virtually ceased during The 2nd century bc, and it is to this period that the distinctive Samaritan text form is assigned. It is probably a revision of a form current much earlier. All copies are written in a derivative of the ‘Phoenician’ alphabet akin to that on Jewish coins of The 2nd century bc, not the Aramaic ‘square’ script used for Hebrew after the Exile.
The oldest ms is in all probability the one traditionally accredited to Abishua, the great-grandson of Aaron (1 Ch. 6:3f.). The ms itself, written on thin vellum, is not uniformly old; the oldest part seems to be that from the end of Nu. onwards. Expert opinion would assign this scroll to the 13th century ad, or not much earlier than its alleged discovery by the high priest Phinehas in 1355.
The first copy of the Samaritan Pentateuch reached Europe in 1616 through Pietro della Valle, and in 1628 an account of it was published by J. Morinus, who claimed it to be far superior to the Massoretic text. This seems to be the case with every new discovery of documents, prompted either by a preference for the Greek Septuagint (LXX, a Greek translation of the Hebrew OT made about 280-150 BCE) or an innate hostility to the traditional Jewish text. There was in this instance another motive at work: the desire on the part of certain scholars to weaken the position of the Reformers in their stand for the authority of the Bible. Gesenius, probably Germany’s greatest Hebrew scholar, brought this barren controversy to an end and demonstrated the superiority of the Massoretic text (1815). We are witnessing in our day an attempt to reinstate the Samaritan Pentateuch. Some of its protagonists betray by their faith in the trustworthiness of the Samaritan transmission an ingenuousness never surpassed by the most extreme conservatives. It is true that in some 1,600 places the Samaritan agrees with the LXX, but the disagreements are equally numerous. It is not easy to account for the agreements; one possibility is that when corrections had to be made in the Samantan Hebrew Pentateuch an Aram. *targum was used (the Samaritan dialect and Aram. are practically identical, and the Samaritan version, that is, the translation of the Pentateuch into Samaritan, in places agrees verbatim with the Targum of Onkelos). There are numerous traces of the influence of the targums in the LXX.
For many of the variants, a simple explanation can be given: the attempt to show that God had chosen Gerizim. After the Ten Commandments in Ex. 20 and in Dt. 5, the Samaritan inserts the passage Dt. 27:2–7 with ‘Mount Ebal’ replaced by ‘Mount Gerizim’, and Dt. 11:30 changes ‘over against Gilgal’ into ‘over against Shechem’.
Many of the variants are due to a misunderstanding of grammatical forms or syntactical constructions. Others consist of gratuitous additions from parallel passages. Some stem from dialect influence. Many arise from their effort to remove all anthropomorphic expressions.
There is no evidence that the Samaritans ever had a body of trained scribes, and the absence of any proper collations of MSS, as attested by the numerous variations, is not compatible with any serious textual knowledge. Neither do the deliberate changes or superfluous additions distinguish them as conscientious custodians of the sacred text. Therefore, its variants must be treated with extreme caution. See the important survey by B. K. Waltke, in J. B. Payne (ed.), New Perspectives on the Old Testament, 1970, pp. 212–239.
By W.J. Martin and A. R. Millard
Bibliography. W. J. Martin and A. R. Millard, “The Hebrew Pentateuch of the Samaritans,” ed. D. R. W. Wood et al., New Bible Dictionary (Leicester, England; Downers Grove, IL: InterVarsity Press, 1996); C. D. Ginsburg, Hebrew Bible, 1926-; R. Kittel, Biblia Hebraica, 1952; C. D. Ginsburg, 5, 1897; F. Buhl, Kanon und Text, 1891; F. Delitzsch, 5, 1920; O. Eissfeldt, 5, 1965; P. E. Kahle, The Cairo Geniza2, 1959; F. G. Kenyon, Our Bible and the Ancient Manuscripts, 1939 (new edn., 1958); B. J. Roberts, The Old Testament Text and Versions, 1951; E. Würthwein, The Text of the Old Testament, 1979; M. Burrows, Dead Sea Scrolls of St. Mark’s Monastery, 1950; W. J. Martin, Dead Sea Scrolls of Isaiah, 1954; F. M. Cross, The Ancient Library of Qumran and Modern Biblical Studies, 1958; P. E. Kahle, Der hebräische Bibeltext seit Franz Delitzsch, 1958; F. M. Cross, S. Talmon, Qumran and the History of the Biblical Text, 1975.