Bible (Texts)

views updated



The received Hebrew text of the OT as it appears in modern printed Bibles includes a basic consonantal text in "square-letter" Aramaic characters that was stabilized with entirely minimal variations by about a.d. 100. This basic text was already provided with its own verse and paragraph divisions, indicated exclusively by intervals of varying width within the text itself. No other markings, headings, colophons, or numberings of any kind are a part of this text. Though a spacing arrangement was known that would set off visually the hemistichs (half-line units) of Hebrew verse, most OT poetry is transmitted in the same format as that used for prose; exceptions are made always for Ex 15.118 and Dt 32.143 and often for Job, Psalms, and Proverbs.

This basic text is accompanied in modern Bibles by a traditional apparatus for its pronunciation and public reading, which reached its standard form in the days of Aaron ben Moses Ben Asher of Tiberias in Palestine, c. a.d. 930. Other Masoretic systems were developed both in Palestine and in Babylonia between the 8th and 10th centuries, but these are now mainly of historical interest. The few parts of the OT transmitted in Aramaic (in Gn 31.47; Jer 10.11; Ezr 4.86.18; 7.1226; Dn 2.47.28) share in all respects the textual history of the Hebrew books. The OT books composed in Greek (Wisdom and 2 Maccabees) or preserved complete primarily in that language (Sirach, Tobit, Judith, Baruch, 1 Maccabees, and parts of Esther and Daniel), share in the distinctive history of the Septuagint. The Semitic evidence for Sirach and Tobit will be mentioned below; see also the articles on the books named, individually. What follows traces back the Hebrew text through the various stages of transmission for which evidence is available, namely (1) printed editions of the OT; (2) collations of manuscript materials; (3) medieval manuscripts and Origen's second column; (4) Sirach and Tobit; (5) the Samaritan Pentateuch; (6) the oldest MSS, from the 3rd century b.c. to the 2nd Christian century.

(1) Printed Editions of the Old Testament. The first Hebrew Biblical book to be printed was the Psalms, with D. imchi's commentary (Bologna 1477); the first complete printed OT in Hebrew was that from Soncino (1488). The text of the Alcalá Polyglot of 1521 (see poly glot bibles), somewhat marred by typographical errors, was nevertheless based in part on two excellent 13th-century Spanish MSS and on another MS now lost that seems to have had Babylonian connections. The prototype for most editions of the Hebrew Masoretic text (MT) is the second rabbinical bible published by Daniel bomberg in Venice (152425); its editor was the Jewish

scholar Jacob ben Chayyim. Separated by six centuries from the fixing of the Ben Asher tradition, he dealt in eclectic fashion with the Masoretic data available to him, accepting, from Ashkenazi manuscript sources, a number of over-refinements and inconsistencies in details. Fine control of the Ben Asher system is reflected in the critical apparatus minhat šay of Shlomo Yedidiah de Norzi (d.1626) printed in an OT from Mantua (1742). The later undertakings of S. Baer, sponsored by Franz Delitzsch, between 1869 and 1895, and of C. D. Ginsburg in OT editions (1894, 190826) failed to provide a sounder basic MT than the Ben Chayyim form of it.

Two current editions deserve notice: the Biblia hebraica, third and later editions (Stuttgart 192937 and later dates) with the text prepared under the supervision of Kahle and a critical apparatus by various scholars under the leadership of R. kittel; and the 1958 edition by N. H. Snaith for the British and Foreign Bible Society of London. The critical apparatus of the Kittel-Kahle edition has been roundly criticized, with a good deal of reason,

for its treatment of Septuagint (LXX) evidence in particular; its actual Hebrew text, based on a Leningrad MS [see (3) (i) (c) below] is quite successful in recovering the Ben Asher Masoretic tradition in a consistent form close to the source. A fully revised edition is actively being prepared. In its disposition of the text on the printed page, the Stuttgart OT abandons the traditional prose arrangement for the modern editors' judgment of poetic structure; this can be, and often is, a valuable aid, but it is also sometimes quite misleading. The Snaith edition, taking its start from Norzi's results, follows a carefully selected but much later MS [see (3) (i) (f) below] and presents, on its editor's testimony, a text very close to the Kahle text, in the standard prose arrangement with Psalms, Proverbs, and Job printed as verse. An undertaking now in progress at the Hebrew University in Israeli Jerusalem proposes to issue an OT text based on the Aleppo Codex [see (3) (i) (a) below] and other good MSS, with an apparatus of variants from all pertinent sources. An earlier Israeli edition bearing the name of M.D. Cassuto was issued by others after that scholar's death and has little to recommend it.

(2) Collations of Manuscript Materials. Three systematic compilations of some size for variants within the MT tradition exist, besides narrower collations from smaller MS groups (e.g., by J. H. Michaelis, 1720). The earliest, Vetus Testamentum hebraicum cum variis lectionibus, ed. B. Kennicott (2 v., Oxford 177680) concerns the consonantal text only. Its collating base is derived through E. van der Hooght's 1705 OT from the Ben Chayyim text of 152425; it provides variants from more than 600 MSS and 50 editions of the OT or its parts. In the Pentateuch it supplies also the Samaritan text [see (5) below] from the London Polyglot, with a collation of 16 Samaritan MSS. The next, Variae lectiones Veteris Testamenti ed. Giovanni Bernardo de Rossi (4 v. Parma 178488), with a supplement, Scholia critica in V.T. libros (Parma 1798), presumes, but does not print, the same collating base as Kennicott. De Rossi controlled a collection of some 800 MSS not included in the Kennicott collation. He presented not an exhaustive, but a selective listing of variants. For those that he did take into account he repeated Kennicott's evidence, added his own, and supplemented the Hebrew collation with data from the versions both supporting and differing from the received MT. Variants bearing not on the consonants, but on the vowel pointing, are also selectively cited. Though the versional evidence always needs rechecking in the light of later critical study, this is the most instructive compilation of variants antedating the discovery of the dead sea scrolls [see (6) below]. Ginsburg's collation of more than 70 MSS, largely from the British Museum's collection, and of 19 early printed editions of the MT, in The Old Testament Diligently Revised (3 v. in 4 London 190826) goes over some of the same ground as the earlier compilations and is generally disappointing in its presentation and in its results.

(3) Medieval Manuscripts and Origen's Second Column. Here are included (i) the basic witnesses to the Ben Asher tradition; (ii) MSS with divergent vocalization from the Tiberian; and (iii) Origen's transcription of the OT Hebrew text into Greek letters.

(i ) Basic Witnesses to the Ben Asher Tradition. Noteworthy MSS that contain the MT with the standard Ben Asher Tiberian vocalization are the following:

(a) The Aleppo Codex (known as A), originally a complete OT furnished with its vowel pointing and accents by Aaron ben Moses ben asher (c. a.d. 930). It was donated to the Karaite Jewish community in Jerusalem and subsequently endorsed for its accuracy by mai monides; it is known to have been in Aleppo at least as early as 1478. During the Arab-Jewish hostilities in 1947 it disappeared for a time and was thought destroyed; the recovery of the MS in a badly truncated state was announced in Israel in 1958. It now lacks all of the Pentateuch to Dt 28.17; 2 Kgs 14.2118.13; Jer 29.931.33;32.24, 911, 2124; Am 8.12-Mi 5.1; So 3.20Za 9.17; 2 Chr 26.1935.7; Ps 15.125.2 (MT enumeration); Sg 3.11 to the end, and all of Ecclesiastes, Lamentations, Esther, Daniel, and Ezra-Nehemiah. Never before available for systematic collation, it is under intensive study as part of the Textus project of the Hebrew University in Jerusalem and is to be employed, when possible, as the foundation for a new critical edition of the MT, as stated above [see (1)].

(b) The Cairo Prophets (known as C), the oldest dated Hebrew MS, written and pointed by Moses ben Asher in 895. Originally, like A, the property of the Karaite community in Jerusalem, it was seized during the First Crusade, then restored by King Baldwin at the instance of the Karaites of Cairo, among whom it is still preserved. It contains the prophetic portion of the Jewish canon, hence the so-called Earlier Prophets (Joshua, Judges, 1 and 2 Samuel, 1 and 2 Kings) in addition to the so-called Later Prophets (Isaiah, Jeremiah, Ezekiel, and the 12 minor prophets; see prophetic books of the old testament). This MS was collated for the Kittel-Kahle OT apparatus and also by Cassuto. It is now alleged to conform rather to Ben Naphtali readings than to those of Aaron ben Moses ben Asher in the subsequent generation.

(c) The Leningrad Codex (known as L), dated 1008, MS B 19a of the Russian Public Library in Leningrad, brought originally from the Crimea by A. Firkowitsh in 1839. A colophon to this MS affirms that it was equipped with vowels and Masora from books corrected and annotated by Aaron ben Moses ben Asher. The pointing shows evidence of some reworking in the direction of conformity with what is otherwise known of Ben Asher practice. It was chosen as the best available base for the Kittel-Kahle edition, and its claim to transmit Ben Asher readings was cross-checked with the 10th-or 11th-century treatise of Mishael ben Uzziel on the differences between the Ben Asher and Ben Naphtali traditions; by this criterion it is trustworthy, but Mishael's list is of uncertain date.

(d) British Museum or. 4445, in London, a Pentateuch of which Gn 39.20Dt 1.33 survives with brief lacunae in Nm 7.4773 and 9.1210.18; from the first half of the 10th century, referring in its margin to the scholar (Aaron) ben Asher in a manner that supposes he was still alive. This codex was used by Ginsburg, who dated its consonantal text a century earlier than the pointing; according to Kahle, text and pointing are contemporaneous.

(e) An OT in Parma, copied in Toledo in 1277, used by de Norzi for his critical work and later collated by De Rossi (his number 782).

(f) British Museum or. 262628, a complete and richly illuminated OT copied in Lisbon in 1483; the foundation, along with De Norzi's treatise and some supplementary

MSS, for Snaith's edition. Like most good Sephardic MSS, it has been subsequently reworked to bring its pointing into agreement with the Ben Chayyim text; it is the unrevised readings of the first punctator that Snaith has followed.

(g) The second Firkowitsh collection, in Leningrad, contains 10th-century MT materials, notably a Pentateuch from the year 930.

(ii ) Manuscripts with Divergent Vocalization from the Tiberian. Not all, but a large part of what is known about medieval Hebrew MSS outside the Ben Asher tradition is derived from the contents of the geniza (repository for disused religious texts) of the Ezra synagogue in Old Cairo (which before a.d. 969 was the Melchite church of St. Michael). The Biblical MSS from this source, scattered among libraries at Cambridge, Oxford, Paris, New York, and elsewhere, were studied especially by Kahle and his pupils, and more recently by A. Díez Macho. They include:

(a) Fragments with a Palestinian vowel-pointing older than that of the competing schools of Tiberias; the MSS that contain it often provide consonantal variations also.

(b) Manuscripts ascribed by Kahle to the Ben Naphtali school, rivals of the Ben Asher family. A number of these MSS are now seen by Díez Macho and others as transitional between the Palestinian and the full-fledged Tiberian systems. To this category seem to belong, in addition to various geniza fragments, the Codex Reuchlinianus of 1105, now in Karlsruhe; also a Pentateuch and a complete OT in Parma (De Rossi's codices 668 and 2, respectively). Díez Macho distinguishes three stages: a tentative proto-Tiberian form, the elaborated divergent form in the codices mentioned, and a later accommodation to the victorious Tiberian system in most features of the text. The difference between the Ben Asher and Ben Naphtali schools is narrowed according to this interpretation to some 900 small details, largely in the use of a single accent mark (the meteg ).

(c)Manuscripts with a Babylonian vowel-apparatus written above the consonants of the text, hence called supralinear. The Cairo geniza yielded texts of this class in some profusion: the introduction to the Kittel-Kahle edition enumerates more than 120 such MSS, and Díz Macho has since enlarged the count. These fall into two classes, one with an early, simpler, and the other with a later, more developed, vowel system; the range in time is from about the 8th to the 10th century. Parallel to the geniza materials in this class is the firsthand vocalization in MS Berlin or. qu. 680, from the Yemen, of which (including seven leaves in New York) some 101 leaves are wholly or partially preserved. The St. Petersburg codex of the Later Prophets (known as P), dating from 916, already shows the use of the Babylonian symbols to record what is in fact an accommodation to the Tiberian Masoretic system.

(iii ) Origen's Second Column. The principal current interest in the divergent vowel systems so far described lies in the opportunity they give for testing the Ben Asher vocalization, late and in many respects artificial, against other traditions and tendencies reaching back closer to the period of spoken Hebrew. The endeavor has also been made to exploit for this purpose the traditional Hebrew pronunciation among the Samaritans [see (5) below]; and the fuller consonantal orthography of some Qumran texts [see (6) below] is pertinent evidence on certain points. Transliterations of Biblical proper names into Greek or Latin letters are of interest in the same regard; and in the so-called Theodotionic recension of the LXX, for reasons not fully understood, there is a sprinkling of transcriptions into Greek letters of ordinary Hebrew words. The most notable single source of this kind is, however, the preserved evidence, mainly from the Ambrosian Library's palimpsest Psalter published by Cardinal G. mer cati, for the second column of origen's Hexapla. This systematic transposition of the Hebrew text into Greek letters presents, within the limitations of the Greek alphabet, a sampling of the way the text was pronounced in the first half of the 3rd century at the latest. On the basis of the uniformity of this transcription and its variance from proper name forms in the LXX, Mercati sees it as contemporary with Origen; Kahle would make it Jewish in origin, like everything else in the Hexapla, and therefore, presumably, earlier. In any case it reflects the standardized Hebrew (consonantal) text subsequent to c. a.d. 100.

The materials listed up to this point pertain strictly to the Jewish canon of the OT and to the consonantal text as stabilized for the future by about the end of the 1st Christian century. Although that text has authentic roots in pre-Christian Judaism, the evidence of the LXX, the NT, Josephus, the Samaritan Pentateuch, and the Qumran and other discoveries combine to indicate that both the scope and the form of OT literature as it circulated among the Jews was somewhat more fluid and varied before that time. The textual evidence for this is discussed in what follows.

(4) Sirach and Tobit. The Cairo geniza contained not only MSS of the OT books received in the Jewish canon, but also five fragmentary Hebrew MSS of Sirach, dating from the 10th to the 12th century. Of these, four were published between 1897 and 1901, the first direct evidence for the original text of the book apart from dubious and undependable scattered citations in rabbinic literature. The fifth MS was brought to light in 1931, and again in 1958 and 1960 additional leaves of two of the known MSS appeared in print. Controversy over the authenticity of these materials sprang up with their initial publication, and skepticism on the part of Jewish scholars in particular has been somewhat widespread in recent years. There can, however, be no doubt, either of the basic authenticity of the text or of the fact that a certain amount of retroversion from the Syriac, done in the period when these copies were made, has been introduced. An added anomaly in the history of this book is that, although the citations of the earlier rabbis are nearly all vague and inaccurate, Gaon saadia ben joseph al-Fayyumi (d. 942) quotes Sirach in Hebrew quite exactly in 25 cases out of 26. A clue to this situation seems to be afforded by the Qumran discoveries and related research. From cave two at Qumran come late 1st-century b.c. fragments of Sir 6.2031 (2Q18), published by M. Baillet, which are just large enough to show a coincidence of wording and a similarity of stichometric arrangement with the geniza copies (though the wording relates to geniza MSS A and C, while it is MS B, not extant for this portion, that is stichometric in the Cairo group). Also in 11QPsa cols. 2122, edited by J. A. Sanders, copied in the 1st Christian century, stand the first half and the last two words of the acrostic poem in Sir 51.1330, this time in an authentic text where the geniza form has long been recognized as secondary to the Syriac. The 196364 excavations at Masada (near the southwest end of the Dead Sea) yielded fragments of 13 columns of a scroll of Sirach in a Hebrew script of the first half of the 1st century b.c. They contain portions of Sir 39.2744.17 written stichometrically, two hemistichs to a line. It is reported by Y. Yadin [Yediot 29 (1965) 120122, in Hebrew] to be in general agreement with the text of MS B from the Cairo geniza and to put the authenticity of the medieval copies beyond dispute. When one combines these facts with the indications from Christian, Jewish, and Muslim sources that MSS from a "cave" sect turned up near Jericho shortly before a.d. 800, it seems possible to identify both the occasion for recovery of an incomplete text of Sirach before Saadia and a part

of the impetus to textual study among the Karaites that accompanied the activity of the several schools of Masoretesthe more so as the Damascus Document of the Qumran group also first came to light in the Cairo geniza.

In this connection may be mentioned the Qumran cave four fragments of Tobit, from four MSS in Aramaic, the original language, and one in Hebrew. J. T. Milik, who is publishing these, affirms that they support in all cases the longest available form of the book, usually represented by the Greek Codex Sinaiticus and by the Old Latin Version. The medieval Aramaic and Hebrew texts of this book, however, are all entirely secondary; none has appeared from the geniza. Of Baruch ch. six (the "Letter of Jeremia"), which was certainly composed in Hebrew, only a Greek fragment (7Q2, published by Baillet) is known from Qumran.

(5) The Samaritan Pentateuch. This is a pre-Christian Palestinian Hebrew recension of the Mosaic books, transcribed in an archaic script derived from the paleo-Hebrew form of the Canaanite alphabet. The earliest

copy of it to reach western Europe was secured in Damascus by Pietro della Valle in 1616. It was published in the Paris and London polyglots, and its critical significance became the focal point of controversy. It is an expanded, repetitious form of the text, with a limited number of specifically sectarian details. The fact that in a large number of individual readings it coincides with the LXX against the MT has made it a continuing stimulus to text-critical study of the OT. It is now represented in European libraries, notably the John Rylands Library in Manchester, England, and the Russian Public Library in Leningrad, by a large number of copies, some dating from the 12th and 13th centuries. The famous "scroll of Abisha," kept by the Samaritans at Nablus and ascribed by them to the 13th year after the conquest of Canaan by Josue, is in its oldest part a MS of the 11th Christian century. Having been twice photographed and its oldest part having been published by F. Perez Castro, it has proved to be a factitious piecing together of materials of varying ages. Kennicott was able to collate 16 MSS of the Samaritan text; a hand edition with variants was published byB. Blayney in 1790. A. von Gall issued (191418) from Berlin an edition with ambitions to be critical; it describes and collates a number of significant MSS; but since others of equal or greater importance were not available to the editor, a definitive edition (now promised by Perez Castro) remains to be produced.

In the light of the new evidences from Qumran [see (6) below], it is clear that the point of departure of the specifically Samaritan text from the earlier Palestinian recension on which it depends is to be sought in about the days of John Hyrcanus (134104 b.c.). The Samaritan text and script, as well as history, converge on this result. Critical evaluation of this text will now be in a new setting, since it is henceforth only one of several witnesses to the state of the text in Palestine at the end of the 2nd century b.c. Transmission of this consonantal text in its older copies has, however, been remarkably faithful, as is proved by comparison with 4Qpaleo Exm [see (6) below]. In general, the expanded, transposed, and reworked features of this text are of no great moment from the standpoint of the textual critic, though the Palestinian recension represented is of historical importance; but its witness to specific ancient Palestinian readings divergent from those of the MT continues to be instructive and significant.

Study of the Samaritan pronunciation of Hebrew has been carried forward by several scholars from a variety of sources: oral dictation by Samaritans of portions of the Pentateuch transcribed into a Western phonetic orthography by Europeans; a vocalization contained in four Samaritan Pentateuch MSS; and grammatical and lexical treatises of Samaritan authors published in Hebrew and Arabic. Whether the evidence from these sources can be integrated with other (Qumran, Hexaplaric) materials to furnish a coherent impression of earlier pronunciation of Biblical Hebrew remains to be seen.

There is a Samaritan Aramaic Targum to the Pentateuch, not to be confused with the Hebrew text just described. This Targum has origins going back to the 4th Christian century and varies greatly from one MS to the next. Published editions of it are inadequate; but when fully known, it seems likely to be of much greater interest for the history of Palestinian Aramaic than for Biblical studies as such. On the other hand, the survival, mostly from the Hexapla, of a number of passages out of the Samaritikon, or Greek translation of the Samaritan Pentateuch, forms a useful link between the earliest Samaritan MSS and the older Palestinian recension from which they were ultimately derived.

(6) The Earliest Manuscripts. These come from the 3rd century b.c. to the 2nd Christian century. For the period before Origen, direct evidence of the Hebrew OT text was almost nonexistent up to 1947; the unique exception was the Nash papyrus, c. 150 b.c., from Egypt, containing Dt 6.46 and the Ten Commandments. Since that time, distinct discoveries of 2nd-Christian-century materials from the Wadi Murabbaāt and of still earlier texts from Khirbet Qumran and other, thus far less productive sites (the wadies west of Engeddi, and Masada) have yielded copies of some OT books (Ecclesiastes and Daniel) scarcely more than a century later than the composition of the books themselves. By the end of 1964 the number of separate OT MSS of which at least some fragments are extant from these various sources stood at about 180; ten are from 2nd-Christian-century contexts (Wadi Murabbaā, Wadi Khabra) and the rest all antedate a.d. 68 (Qumran) or a.d. 73 (Masada) at the latest.

All these MSS, on leather or papyrus, were written in columns on one side only of the material; no Hebrew text of the 2nd century or earlier in codex (book) form is known. The complete scroll of Isaiah (1QIsa) from Qumran is made up of 17 strips of carefully prepared leather sewn end to end to a length of 24½ feet, 10½ inches high, meant to be kept rolled up when not in use. In it the text is disposed in 54 vertical columns, with an intentional main division after col. 27, the end of the present ch. 33 (of 66 ch.). This arrangement is suggestive for the format of any large biblical book at this period; nevertheless, from Qumran there are MSS with as few as nine lines of text to the column, and others with more than 60, whereas 1QIsa averages 30 lines to the column.

To take first the 10 MSS left by refugees after the Second Jewish Revolt (a.d. 132135), the evidence includes fragments of Genesis, Exodus, and Numbers by one scribe; of Genesis in a different hand; of three other MSS of Numbers, two of Deuteronomy, and one each of Psalms, Isaiah, and the Minor Prophets. Only the MS of Psalms from the Wadi Khabra (fragments in Jordan, some few in Israel) is actually of 1st century date and exhibits some variation from the MT. The rest show the fixity of script, format, orthography, and content that constitutes the basic MT text. This might be expected for the Pentateuch and Isaia. The Minor Prophets (Mur 88), however, is preserved in very substantial portions representing ten of the 12 books; there are only three meaningful variants in it from the MT consonants, and only one of these is notable, though not an improvement.

Quite different are about 170 MSS from the 1st century and before. They include some archaic texts (4QExrf, Samb, Jera) dated from c. 250 to 200 b.c. by F.M. Cross on paleographical grounds. A somewhat larger number of 2nd-century b.c. texts is followed by the bulk of the MSS, dating from the 1st century b.c., with texts of the 1st Christian century also present in quantity. All books of the full Catholic OT canon are somehow represented (although 1 and 2 Chronicles by one isolated fragment with about five incomplete lines of text), except Esther, Judith, Wisdom, 1 and 2 Maccabees, Baruch ch. 15, and the LXX additions to Daniel. A limited number of these early MSS are in the paleo-Hebrew script descended from preexilic forms; besides Pentateuch MSS, surprisingly, there is one of Job. The orthography is not so consistent as in the MT. Although some of the oldest MSS have narrowly consonantal spelling, perhaps a third of the Biblical MSS show in varying degrees a much fuller orthography with lavish use of the weak consonants h, w, y and ' (aleph ) to mark the place of vowels in the word. This usage, common also in extra-Biblical texts from Qumran, parallels that of Syriac and differs from medieval Hebrew practice, in that all "o" and "u" vowels are represented by w, regardless of their length, whereas only long "î" or "ê" vowels are indicated by y.

Unknown individual readings that are not mere vagaries of the particular copyist are on the whole somewhat rare. But textual tendencies in Palestine that could be envisaged only doubtfully and obscurely from the LXX and Samaritan evidence can now be studied directly in these texts. Far from proving the overall superiority of the LXX, these Hebrew MSS help to endow that version with a continuous history of development that makes the jumbled evidence in extant Greek MSS more adequately subject to control.

At the present stage of investigation, the incidence of fuller Palestinian readings coinciding with the LXX or the Samaritan, and the identifiable Palestinian tendency to an expansionist technique in copying and editing Biblical texts, from an early postexilic date until the reaction that is represented by the MT, have led Cross to posit for the Pentateuch (at least Exodus through Deuteronomy) and Samuel in the received text a Babylonian origin that would have kept them apart from the development in Palestine. For Samuel in particular, where the MT represents a surprisingly truncated and defective recension, some such explanation is surely called for. The MT of the Pentateuch is a sound, tightly organized, unexpanded text of a quite different character; but again, it is doubtful that such a text can be directly filiated to the fuller and less stabilized forms evidenced for Palestine from the proto-Lucianic LXX, the Samaritan, and now the Qumran sources.

Of individual MSS thus far published, only brief mention can be made. The complete Isaiah scroll (1QIsa) dates from c. 100 to 75 b.c. It is a reworked text of Isaiah, disclosingbeneath the very full orthography of its second half especially and beneath its harmonizations of related passages, simplified readings, and borrowings from other OT booksa basic text quite close to the MT tradition, with which all other (at least 14) Qumran MSS of Isaiah coincide more closely still; however, the degree of nearness of 1QIsb, a later and more fragmentary MS, to the MT has in fact been overstated in the literature. An early 2nd-century b.c. copy of Exodus in the old script (4Qpaleo Exm) contains all the expansions known previously from the Samaritan Pentateuch [see (5) above], except that about the unhewn altar on Mt. Garizim after Ex 20.17. It proves the Samaritan recension quite faithful to a pre-Christian Palestinian form of text; but there are now a number of Qumran MSS that evidence, in varying degrees, these same Palestinian tendencies to expansion in Exodus through Deuteronomy, of which the Samaritan text is no longer the prime witness. In general, Qumran MSS of the historical books tend to coincide with the LXX evidence, especially that of a proto-Lucianic type; very remarkable in this regard is 4QSama (1st century b.c.), in which a notable amount of the text of 1 and 2 Samuel is preserved. The short recension of Jeremiah hitherto known only from the LXX is present in 4QJerb, one of four MSS of that Prophet at Qumran. The compilation in 11QPsa (1st Christian century), which combines 35 canonical Psalms in an irregular order with eight other compositions, seems to show special interest in David as a person and as author; various considerations suggest that the standard canonical order of the Psalms is presupposed by this unique collection.

In addition to the strictly Biblical MSS, the several hundred extra-biblical texts from the same sources will have to be studied extensively for biblical lemmata (formal citations of biblical verses as a basis for commentary in the pesharim ), incidental quotations and allusions, before the full contribution of the discoveries since 1947 to an understanding of the history of the OT text can be assessed.

Bibliography: o. eissfeldt, The Old Testament: An Introduction, tr. p. ackroyd (New York 1965) sections 115118, 126. p. e. kahle, The Cairo Geniza (2nd ed. New York 1960); Der masoretische Text des A. T. nach der Überlieferung der babylonischen Juden (Leipzig 1902); Masoreten des Ostens (Leipzig 1913); Masoreten des Westens, 2 v. (Stuttgart 192730); Der hebräische Bibeltext seit Franz Delitzsch (Stuttgart 1961). m. greenberg, "The Stabilization of the Text of the Hebrew Bible, Reviewed in the Light of the Biblical Materials from the Judean Desert," The Journal of the American Oriental Society 76 (1956) 157167. g. e. weil, "La Nouvelle édition de la Massorah (BHK iv) et l'histoire de la Massorah," (Vetus Testamentum Suppl 9; 1963) 266284. Textus: Annual of the Hebrew University Bible Project, ed. c. rabin (196063), v.13. a. i. katsh, Ginze Russiyah (New York 1958), fac. of Heb. MSS preserved in the U.S.S.R. a. dÍez macho, "A New List of So-Called Ben Naftali Manuscripts , "Hebrew and Semitic Studies presented to G. R. Driver , ed. d. w. thomas and w. d. mchardy (Oxford 1963). g. mercati, Psalterii hexapli reliquiae (Vatican City 1958); "Il problema della colonna II dell' Esapla," Biblica 28 (1947) 130, 173215. f. perez castro, Séfer Abiša' (Madrid 1959), see the important review by e. robertson, Vetus Testamentum 12 (1962) 228235. n. h. snaith, "New Edition of the Hebrew Bible," ibid. 7 (1957) 207208. m. baillet, "La Récitation de la Loi chez les Samaritains d'après Z. Ben-Hayyim," Revue biblique 69 (1962) 570587. b. j. roberts, "The Hebrew Bible since 1937," Journal of Theological Studies 15 (1964) 253264. d. barthÉlemy and j. t. milik, Qumrân Cave I (Discoveries in the Judean Desert 1; 1955). p. benoit et al., Les Grottes de Murabbaāt (ibid. 2; 1961). m. baillet et al., Les "Petites Grottes" de Qumran (ibid. 3; 1962). j. a. sanders, ed., 11Q Ps a (ibid. 4; 1965). m. burrows, The Dead Sea Scrolls of St. Mark's Monastery, 2 v. (New Haven 195051). e. l. sukenik, The Dead Sea Scrolls of the Hebrew University (Jerusalem 1955), in Heb. y. yadin, "The Expedition to the Judaean Desert, 1960: Expedition D," Israel Exploration Journal 11 (1961), 40 and plate 20D, a MS of Psalms. j. t. milik, Ten Years of Discovery in the Wilderness of Judaea, tr. j. strugnell (Studies in Biblical Theology 26; Naperville, Ill. 1959). a. a. di lella, The Hebrew Text of Sirach: A Text-Critical and Historical Study (The Hague 1965). f.m. cross, jr., The Ancient Library of Qumran and Modern Biblical Studies (rev. ed. 1961); "The Development of the Jewish Scripts," The Bible and the Ancient Near East, ed. g. e. wright (Garden City 1961); "The History of the Biblical Text in the Light of Discoveries in the Judaean Desert," Harvard Theological Review 57 (1964) 281299. p. w. skehan, "Qumran and the Present State of O.T. Studies: The Masoretic Text," Journal of Biblical Literature 78(1959) 2125; "Exodus in the Samaritan Recension from Qumran," ibid. 74 (1955) 182187; "The Qumran Manuscripts and Textual Criticism," (Vetus Testamentum Suppl 4; 1957) 148160; "A Psalm Manuscript from Qumran (4Q Psb)," The Catholic Biblical Quarterly 26 (1964) 313322; "The Biblical Scrolls from Qumran and the Text of the Old Testament," The Biblical Archaeologist 28 (1965) 87100. y. yadin, The Ben Sira Scroll from Masada (Jerusalem 1965).

[p. w. skehan]


This article will treat in chronological order the forms in which the Greek text of the NT has appeared from the earliest extant manuscripts (MSS), discussing printed editions and indicating projects to reproduce the text more adequately.

The autographs of the 27 canonical books of the NT, written or dictated by several inspired authors over a period of two generations, were lost before almost any extant manuscript (MS) was penned. After having been produced on papyrus scrolls, the autographs had been copied by hand, and these MSS had circulated among individual Christian communities until they in turn were replaced. In the course of transmission both scribal errors and conscious alterations modified the form of the original text. Short clarifications, modification of unfamiliar words, omissions, and harmonizations appeared in MSS. Those in ancient languages into which the NT was translated for Christians who did not speak Greek indicate some modifications not found in any extant Greek MS. In addition, homilies and commentaries of early ecclesiastical writers at times present other textual variations. The Greek text of the NT, as it appears in modern printed editions, is reconstructed on the basis of study and evaluation of all such witnesses, which are only a fraction of its many forms in history. This article deals with the Greek witnesses and modern critical presentations.

Extant Greek New Testament Manuscripts. The first attempt at a complete listing of all extant Greek NT MSS was made by C. R. Gregory (18471917) in the Prolegomena of the great 8th edition of C. von Tischendorf's Novum Testamentum Graece (Leipzig 1894). Gregory later completed this list in what is accepted as the official list and method of identifying NT Greek MSS (Die griechischen Handschriften des Neuen Testaments. [Leipzig 1908]). Supported by authority of the Kommission für spätantike Religionsgeschichte of the German Academy of Science in preference to proposals of Hermann von Soden, Gregory's list has been continued by E. von Dobschütz, J. Schmidt, and K. Aland through notices in Zeitschrift für die neutestamentliche Wissenschaft und die Kunde der älteren Kirche. The notice by Aland in 1957 brought the number to 67 papyri, 241 uncials, 2,533 cursives, and 1,838 lectionaries, although many of these 4,689 MSS contain only small parts of the Greek text. To keep information accurate, Aland founded the Institute for New Testament Textual Research at Münster in Westphalia in 1959 and inaugurated a series called the Arbeiten zur neutestamentlichen Textforschung in 1963. Of the large number of MSS, Gregory could list only one uncial, the Codex Sinaiticus, and about 35 cursives as having the entire NT.

With regard to the material on which the text is written, NT manuscripts are of papyrus, vellum or parchment, and paper. Since paper MSS are late and relatively unimportant, the official list divides the MSS into papyri, uncial vellum, and cursive vellum MSS. The use of vellum became common only from the time of Constantine, who ordered 50 copies on vellum for the churches of his empire. Since it was expensive and limited in quantity, usable parts of worn vellum codices were salvaged to be used again. Thus parts of the NT have been preserved in later writing on such palimpsest MSS. The following is a brief description of the most important NT MSS in each of the three groups. Since the dates proposed depend upon paleographic evidence, they are at times only tentative (see paleography, greek).

Principal NT Papyri. Significant progress has been made in knowledge of early forms of the NT text because of discoveries of papyri, some of which were written as early as 150 years before the oldest vellum MS (see papy rology). In the official list they are designated by a capital P followed by the Arabic numeral indicating the order in which their discovery was reported. By 1964 some 77 papyri containing parts of the Greek text had been announced, dating from the 2nd to the 8th century. Although most of these are fragmentary, about half of the Greek NT is now extant on papyrus, parts of every book except 1 and 2 Timothy. Four of these fragments are from scrolls, the rest from codices (MS books). The oldest (P 52, John Rylands Library Gr. 457) contains parts of Jn 18.3134, 3738, written during the first half of the 2nd century in a text type like that of the Codex Vaticanus.

The most complete descriptive list of Greek NT papyri appears in the article "Papyrus Biblique" by B. Botte [Dictionnaire de la Bible, suppl. ed. 6 (1960): 110920]. It gives content, date, location, publication, and text type of each papyrus to P 72, except for P 38, erroneously omitted. P 38 (University of Michigan C.1571) is a 3rd-century fragment of a popular, i.e., unrevised, text of Acts 18.2719.6 and 19.1216. Botte's article fails to give information on photographic facsimiles. The papyri most important for NT study have been found in this century and form part of the Chester Beatty and Bodmer collections.

Chester Beatty Papyri. In 1930 an Irish businessman, A. Chester Beatty, purchased in Egypt a collection of 11 Biblical papyri, including three of the NT. All these were edited by Sir Frederic George Kenyon with photographic facsimiles as The Chester Beatty Biblical Papyri (7 v. and pl. London 193337). From the NT are: (1) P 45 (Chester Beatty I). This consists of 30 mutilated leaves of a 3rd-century codex about 10 by 8 inches. Extant are fragments of all the Gospels and Acts in a popular text with no particular Western characteristics. This papyrus revolutionized understanding of the NT text by showing so-called Caesarean readings at a date much earlier than had previously been suspected. (2) P 46 (Chester Beatty II plus University of Michigan Inv. 6238). This early 3rd-century single-quire codex measures about 9 by 5½ inches. The 86 extant leaves of the original 104 contain parts of almost all the Pauline Epistles and the Epistles to the Hebrews, which follows immediately after Romans. About half of Romans, most of 1 Thessalonians, and all of 2 Thessalonians are missing, and Ephesians precedes Galatians. G. Kuntz has shown that this is an extremely valuable witness of the proto-Alexandrian text type, despite its many scribal errors. The University of Michigan owns 40 leaves but permitted Kenyon to edit them with the rest of the codex. Independently H. A. Sanders also studied the Michigan leaves in A Third-Century Papyrus Codex of the Epistles of Paul (Ann Arbor 1935). (3) P 47 (Chester Beatty III). This 3rd-century codex, consisting of ten leaves of Rv 9.1017.2 with lacunae, is the earliest MS of this book and presents a text similar to that of the Codex Sinaiticus.

Bodmer NT Greek Papyri. These form part of the collection of classical, Biblical, and apocryphal texts in Greek and Coptic acquired by the Swiss industrialist whose name they bear, for his private library in Cologny near Geneva. Of the 19 published, the following six are Greek NT texts: (1) P 66 (Bodmer II). This is the extant 108 pages in a codex of five quires about 6½ by 5½ inches containing most of Jn 1.114.26, except for a lacuna of 6.1135, and fragments of the remainder of John V. Martin edited the first 14 chapters in 1956 and part of the fragments in 1958. Shortcomings of this edition and the lack of photographic reproductions were remedied in a second edition of the fragments, which includes a facsimile of the entire papyrus [V. Martin and J. W. B. Barnes, Papyrus Bodmer II. Supplément. Evangile de Jean XIVXXI (Bibliothèque Bodmer, Cologny-Geneva 1962), 54 pp. and 154 plates]. The first editor dated this codex earlier than a.d. 200, and H. Hunger says that it is no later than a.d. 150. However, J. Duplacy refers to two unnamed papyrologists who place it in the 4th century [Recherches de science religieuse 50 (1962) 251]. It omits the pericope of the woman taken in adultery (Jn 7.538.11) and of the moving of the waters at the pool of Bethesda (Jn 5.4). Carelessly written, it is often corrected by the original and by later scribes. The text, which fluctuates in its agreements with classical text-types, shows clear resemblances to the Old Latin. (2) P 72 (Bodmer VII and VIII). This 3rd-century codex contains the Epistles of Jude and 1 and 2 Peter. M. Testuz, curator of the Bodmer library, edited these in 1959 and found the text much like that of Codex B and the Bohairic version, especially for 2 Peter. A complete collation of 1 Peter by E. Massaux indicates that it bears greatest similarity to the cursives 104, 424, 326, and 81 and reveals one of the many popular texts of the 2nd and 3rd centuries [Ephemerides theologicae Lovanienses 39 (1963) 61671]. (3) P 74 (Bodmer XVII). This is a 6th-or 7th-century codex containing all of Acts and fragments of all the Catholic Epistles. Since the back part of the codex was severely damaged, the fragments decrease rapidly in size. R. Kassar edited this papyrus for the Bodmer library in 1961.(4) P 75 (Bodmer XIV and XV). This is the extant part of a 144-page codex from about a.d. 300 originally containing Luke and John. Extant are 25 full pages, 26 pages almost complete, and fragments of others. These were edited for the Bodmer library in two volumes with 98 plates by V. Martin and R. Kassar in 1961. Bodmer XIV contains Luke 317 and 2224 in a text similar to Codex B. It contains the long text in Lk 22.1920; 24.12, 40, 51b and omits the bloody sweat of Christ (Lk 22.4344), also omitted by the Vaticanus and the first hand of Sinaiticus. Bodmer XV is the text of John 115, similar to that of Codex B. (5) P 73 is a fragment of Mt 24.43 and 26.23 that was found between leaves of P 74 and is unedited.

Principal NT Uncials. Most NT vellum MSS copied between the 4th and 10th centuries were without separation of words, had little punctuation, and were written in large, unconnected letters called uncials, a Latin word meaning "one-twelfth," i.e., of a line of letters. In general these are the most highly esteemed witnesses of the text because of their careful composition. They are identified by a capital letter or by 0 plus an Arabic number, known as a siglum. Uncials fundamental for the study of the NT text are listed below.

Codex Vaticanus. The Vatican codex Gr. 1209, siglum B or 03, is a 4th-century MS originally containing the entire Bible in Greek, but Gn 1.146.27, Ps 106(107)138(139) and Heb 9.14 to the end of the NT are now missing. Since the Catholic Epistles are before Paul, they are extant, but the Pastoral Letters and the Apocalypse are lost. Although this codex was in the Vatican Library when it was first catalogued in 1475, it was published completely only in 1857. The splendid photographic edition of the Vatican Library appeared in 188990 in seven volumes. Although the entire NT seems to have been copied by one scribe, this is uncertain because the letters were inked over by a monk in the 12th century. Codex B offers the best example of the refined text existing in Egypt in the early 3rd century and, except in Paul, is free from the readings of the widely diffused popular 2nd-century texts. Codex B is the chief witness for the text type called "neutral" by Hort, "Hesychian recension" by von Soden, "text B" by Lagrange, "proto-Alexandrian" by Zuntz, and "Beta" by many recent critics. Westcott and Hort used it as the fundamental text for their edition, and through them it has played the decisive role in many manual editions. Its chief defects are mechanical, such as doubling or omission of letters, syllables, or lines.

Codex Sinaiticus. This is the Codex Frederico-Augustanus plus British Museum Add. MS 43725, siglum S or Hebrew aleph. K. von Tischendorf found this MS on two of his expeditions to the monastery of St. Catherine at Mt. Sinai, part of the OT in 1844 and the NT plus the Letter of Barnabas and the Shepherd of hermas in 1859. His second find of 199 leaves was sold by the Soviet government to the British Museum in 1933. In their study, Scribes and Correctors of the Codex Sinaiticus (London 1938), H. J. M. Milne and T. C. Skeat date the codex from the 2nd half of the 4th century and attribute it to three scribes writing from dictation. It has many mistakes, especially in the part executed by the third scribe, and was corrected in different ages by nine hands. Its place of origin seems to have been Caesarea. Kirsopp Lake and Helen Lake edited a photographic facsimile (Oxford 1911).

Codex Alexandrianus. This is an early 5th-century MS (British Museum, Royal MS 1 D VVIII; siglum A or 02) that once contained the entire Bible. Now the NT begins with Mt 25.7 and lacks Jn 6.508.22 and 2 Cor 4.1312.6. A photographic edition begun for the museum by K. Lake in 1909 was completed only in 1957 by T.C. Skeat in five volumes. This codex, noteworthy for its frequent substitution of synonyms, presents a text of unequal quality. The Gospels belong to the inferior Byzantine type, Acts and the Epistles to the Alexandrian type.

Codex Ephraemi Rescriptus. This is in the Paris National Library, Codex Gr. 9; siglum C or 04. The extant 209 leaves of this 5th-century palimpsest were employed in the 12th century for a Greek translation of the sermons of Ephraim. The 145 leaves from the NT preserve passages from all books except 2 Thessalonians and 2 John. In 1845 Tischendorf published as much as he could read of it. Fundamentally the text is Alexandrian, but it contains mixed readings. After Codex C the list of Gregory uses the same letter for more than one uncial because the rest contain only limited parts of the NT. Four of these demand mention here.

Codex Bezae. A MS of only Gospels and Acts; siglum D or 05. It is commonly known by the name of its 16th-century owner, Theodore beza, and is now preserved at Cambridge University. This 5th-or 6th-century uncial is the oldest extant Greek and Latin codex. The order of the Gospels is that of most Latin MSS, Matthew, John, Luke, and Mark. Its text, which is complex, is the leading witness of the Western text and the oldest MS with the pericope of the woman taken in adultery.

Codex Claromontanus. This 6th-century Greek and Latin Codex (siglum D or 06) of the National Library, Paris, contains 533 leaves written in sense lines of unequal length. The text type is also Western.

Codex Washingtonianus. This MS (siglum W or 032) of 187 leaves, now in the Freer Gallery, Washington, D.C., presents a modified uncial script of the 5th century except for the quire containing Jn 1.15.11, written by another scribe and offering an early Egyptian form of text. Mark often agrees with P 45, but other parts exhibit a variety of types. After Mk 16.14 is found the so-called Freer logion, a 16-line addition partly quoted by St. Jerome as being found in many MSS.

Codex Koridethi. This codex (siglum Θ or 038), found in a monastery on the Black Sea, was published by Gregory and Beerman in 1913. It cannot be dated exactly because its writing is unique, but it is placed between the 7th and the 9th centuries. It is related to families 1 and 13 mentioned below, and is a witness of the Caesarean text of the Gospels.

Principal NT Cursives. During the 9th century a cursive or miniscule book-hand began to replace the uncial style of writing, and this prevailed until the introduction of printing. Cursive MSS are generally of lesser value as witnesses, but some of them preserve early readings otherwise lost. They are identified simply by an Arabic number. The "queen of the cursives" is 33, a 9th-century codex containing most of the NT except Revelation in an Alexandrian text. The large number of cursives has enabled critics to trace relationships between "families," that is, groups of MSS originating from the same archetype. Two of these are described below.

Family 13. The collection is known also as the Ferrar family, after an Irish clergyman, W. H. Ferrar, who first established a relationship between four cursives in 1868. After his death a collation of these, 13, 69, 124, 346, was published by his collaborator, T. K. Abbot, in 1877. Further research, especially by Von Soden and K. Lake, enlarged this family to 13 medieval codices including one lectionary, all containing only the Gospels except 69, which includes the entire NT. The most striking feature of this family is the position of the pericope of the woman taken in adultery (Jn 7.538.11) after Lk 21.38. They preserve the Caesarean text-type and were copied in monasteries in southern Italy. J. Geerlings has continued research on this family in the series Texts and Studies. Inv. 1921 he published the hypothetical archetype of the Ferrar family text for Matthew, Luke and John.

Family 1. This is a designation for a group known also as the Lake family. Kirsopp Lake, in Codex 1 and its Allies (Cambridge 1902), identified four cursives, 1, 118, 131, and 209, as members of the same family. After relationship had been established between these, the Ferrar family, and the Codex Koridethi, B. H. Streeter, in The Four Gospels (London 1924; rev. 1930; repr. 1953), postulated the existence of a local, so-called Caesarean text of the Gospels. K. Lake, R. P. Blake, and S. New made a brilliant corporate effort to recover this text in "The Caesarean Text of the Gospel of Mark" [Harvard Theological Review 21 (1928) 207404], in which they published a reconstruction of Mark ch. 1, 6, and 11 in this form. However, the discovery of P 45 showed that the "Caesarean" text had roots in Egypt. In projected studies on Family II, Geerlings hoped to shed more light on this intricate phase of the history of the NT text.

Greek NT Lectionaries. For the convenience of monks and clerics, volumes of liturgical readings from the Gospels and Epistles, called lectionaries, were compiled as early as the 6th century. Of the 1,838 of these in Gregory's list, fewer than 200 are uncials, and more than 1,200 contain readings from the Gospels only. Since most of them have not been investigated critically, their value as witnesses to the NT text is still unknown. To remedy this neglect, E. C. Colwell and D. W. Riddle inaugurated the series Studies in the Lectionary Text of the New Testament with their Prolegomena to the Study of the Lectionary Text of the Gospels (Chicago 1933). This series reached its fifth study with the monograph of William D. Bray, The Weekday Lessons from Luke in the Greek Gospel Lectionary (Chicago 1959). Since these researches have reached only the preliminary stages, results are not conclusive. Indications are that the lectionaries may belong predominantly to the Byzantine text-type but may at times support ancient readings.

Patristic Citations. Additional information about the text of the NT can be gleaned also from citations of ecclesiastical writers and Fathers of the Church, especially those who wrote before the widely diffused Byzantine text began to prevail in the 7th century. Gathering and evaluating this information is extremely difficult because of the lack of reliable critical editions and uncertainty about the accuracy of citations. Up to the 1960s, evidence had often been inconclusive. In his introduction to the pioneering study of P. M. Barnard, The Biblical Text of Clement of Alexandria in the Four Gospels (Texts and Studies 5; Cambridge 1899), F. C. Burkitt weakened Hort's theory of a neutral text by concluding that Clement was a witness for the Western text-type, related to the Sinaitic Syriac version of the Gospels. In the Pauline Epistles G. Zuntz cites Clement as having made use of a proto-Alexandrian text.

In three articles in the Journal of Theological Studies from 1935 to 1937, R. V. G. Tasker presented evidence to show that the text of origen (d. c. 254) usually follows the Alexandrian type but offers Caesarean readings in parts of Matthew and John. Although no critical edition of St. John Chrysostom is available, those who have investigated NT citations in his writings are agreed that he is not the father of the Byzantine text and that his text differed from that of any extant MS. He evidently combined readings from more than one source for greater clarity. In his study The Gospel Text of Cyril of Jerusalem (Copenhagen 1955) J. G. Greenlee indicated that Cyril's text was "pre-Caesarean" with similarities to Sinaiticus for the synoptics but a popular type for John. Among contemporary critics, M. E. Boismard places great stress upon the testimony of early ecclesiastical writers, and in his work on John he has shown that at times they witness to readings that may be original, although not found in any extant Greek MS.


The use of printing gradually brought to an end the multiplication of textual variants of the NT.

Early Editions. Desiderius erasmus published the first printed edition of the NT in a hurried and faulty Greek and Latin edition at Basel in 1516. The Greek text had already been printed in 1514 in the text of D. L. Stunica for the Complutensis polyglot bible of Cardinal ximÉnez, but this edition did not actually appear until 1520. Erasmus used it to improve his 4th edition of 1527.

Critical notations of alternate readings were first added in the margins of the text published by R. estienne, who edited the text four times from 1546 to 1551. His final edition, which introduced the present division of the text into verses, was often reprinted. Another edition destined to have marked influence upon the diffusion of the NT text for 200 years was produced by the brothers B. and A. Elzevir (Leiden 1624). They introduced their second edition of 1633 as the textus receptus, "the text received" by all. Although this represents the official text of the Greek Church and is based on the largest number of uncials, e.g., N, Y, V, K, II, Ω for the NT as a whole and A, E, F, G, H for the Gospels, and the vast majority of cursive manuscripts, this text type is of an inferior critical quality.

Preliminary research to improve the quality of printed editions was made by Richard simon, who pointed out the insufficiency of the textus receptus in his Critical History of the NT Text (London 1689), translated from the French edition of the same year. In 1734 J. A. Bengel made a positive contribution toward an improvement of printed editions by dividing MSS into "families, tribes and nations," thus facilitating a more accurate critical evaluation of their text [Novum Testamentum graecum (Tubingen 1734)]. A major improvement in the presentation of MSS in the critical apparatus was introduced by J. J. Wettstein, whose edition was based on 330 MSS. He was the first to designate uncials by capital letters and cursives by Arabic numerals in his sigla, anticipating the system of Gregory [Novum Testamentum graecum, 2 v. (Amsterdam 175152; photographic reproduction Graz, Austria 1961)]. For his edition of the received text, J. J. Griesbach divided the Gospel MSS into the three classes that were to be commonly accepted, to which he gave the misleading names Western, Alexandrian, and Constantinopolitan recensions (Halle 1777; 2d ed., Leipzig 17961806).

Later Improved Editions. Only in 1830 did a NT editor depart from the custom of editing the received text. This was the distinction of the small but revolutionary edition of Karl Lachmann, who published also a larger edition (Berlin 184250). His goal was to reproduce the text current in the 4th century, and he limited his edition to a small number of witnesses. The English editor S. P. Tregelles, whose edition was completed after his death by Hort in 1879, likewise concentrated on a limited number of older witnesses. Most famous of all NT editors was Konstantin von tischendorf, who collated and published more than 40 Biblical MSS and produced eight critical editions of the NT with four widely divergent texts. His final text, the 8th editio critica major (Leipzig 186572), was marred by his excessive preference for the readings of Codex Sinaiticus, which he had recently discovered. The lengthy introductory volume, which his pupil C. R. Gregory worked 22 years to write, is still a fundamental source of information about NT MSS, although surpassed and antiquated in many ways.

The 19th-century text that proved most decisive for NT studies was the result of 28 years of collaboration between Brooke Foss westcott, later an Anglican bishop, and Fenton John Anthony Hort [The New Testament in Greek, 2 v. (London 1881; 2nd ed. 1898)]. In contrast to Tischendorf, who refused to rely on the classification of MSS, this edition insisted that the history of the text must be considered in order to arrange the MSS according to their exact critical value. Hort's introductory volume, which is a treatise on NT textual criticism, explains that their text is based on "the best documentary evidence." To find this, MSS were divided into four classes, neutral, Western, Alexandrian, and Syrian. Internal criticism was to remain supplementary (see paras. 76 and 82). Critics found fault with the narrow basis for choosing the text, and the editor of the second edition, Francis Crawford burkitt, acknowledged this fault (additional note to para. 170), because the discovery of the Old Syrian palimpsest of the Gospels by Mrs. Agnes Lewis in 1892 had emphasized the possibility that other combinations of witnesses, such as agreement of versions, could command more critical reliability than the primary Greek witnesses. As early as 1904, K. Lake called the edition by Westcott and Hort "a failure, though a splendid one." A generation later, comparing the work of Westcott-Hort with that of Marie Joseph lagrange, he commented that perhaps no thesis more subjective than that of Hort ever existed [Revue biblique 48 (1939) 498]. The edition was attacked also because of its hypothesis of a neutral or uncontaminated text in Codex B and because the editors used internal criticism in a way incompatible with their stated principles. An effort to make greater use of all MS witnesses including the ancient versions appears in the edition of Bernard Weiss (Leipzig 18941900; 2nd ed.1905), but it also depends chiefly upon Codex B.

The fruits of the critical researches on the NT text by Tischendorf, Westcott-Hort, and Weiss have been widely diffused chiefly through the frequently edited manual edition of Eberhard Nestle [Novum Testamentum graece (Stuttgart 1898)]. Its fourth edition was used also as the text of the first edition of the NT by the British and Foreign Bible Society. During its 24 editions the Nestle publication has incorporated new information on the text as far as possible in its present format. K. Aland, who along with Erwin Nestle is the current editor, is preparing an entirely new edition in a new format that will represent a major revision.

The last complete critical edition was prepared by a staff of 40 under the direction of H. von Soden [Die Schriften des neuen Testaments in ihren ältesten erreichbaren Textgestalt, 2 v. (Berlin-Göttingen 190213)]. After a thorough examination of the Greek MSS, he divided them into three recensions: H (Eta) by Hesychius in Egypt; I (Iota or Jerusalem) by Pamphilius of Caesarea, and K (Kappa) by Lucian of Antioch. Although Von Soden's text and new system of sigla were severely criticized, his introductory studies on the MSS contain information of great value. His critical researches influenced the manual editions of three Catholic editors. These were Heinrich Joseph Vogels, Novum Testamentum graece (Düsseldorf 1920; 3rd ed. Herder, Freiburg 1950); J. M. Bover, Novi Testamenti Biblia (Madrid 1943; 3rd ed. 1953), and Augustin Merk, Novum Testamentum graece et latine (Rome 1933; 8th ed. by J. P. Smith 1957). A revised edition of the last mentioned has been announced.

In an attempt to provide an edition of the NT with a more complete and accurate critical apparatus than was available, a committee of English scholars undertook the project and entrusted editorship to S. C. E. Legg. Severe criticism of the two volumes that appeared, Mark in 1935 and Matthew in 1940 (Oxford), caused a modification of plans. To prepare this complete critical apparatus an international committee was set up with M. M. Parvis as American secretary; the British secretary, G. D. Kilpatrick, was also editor of the second edition of the Greek NT of the British and Foreign Bible Society (London 1958). Another smaller international NT project was begun in 1956 under the direction of the American Bible Society and similar groups in other countries. The editorial committee undertook to prepare a new critical text with a limited apparatus of variants having theological and exegetical importance, and an accompanying supplement to explain the choice of readings adopted.

Bibliography: v. taylor, The Text of the NT: A Short Introduction (New York 1961). f. g. kenyon and a. w. adams, Our Bible and the Ancient Manuscripts (5th ed. rev. New York 1958). b. botte, "Manuscrits grecs du NT," Dictionnaire de la Bible, suppl. ed. l. pirot et al. (Paris 1928) 5:819835. g. zuntz, The Text of the Epistles (New York 1953). w. h. p. hatch, The Principal Uncial Manuscripts of the NT, with 76 plates (Chicago 1939). k. and s. lake, eds., Dated Greek Minuscule MSS to the Year 1200, with 150 tables (Boston 193438). m. m. parvis and a. p. wikgren, eds., NT Manuscript Studies (Chicago 1950). b. m. metzger, The Text of the New Testament (New York 1964). l. vaganay, An Introduction to the Textual Criticism of the NT, tr. b. v. miller (St. Louis 1937). Encyclopedic Dictionary of the Bible, tr. and adap. by l. hartman (New York 1963) 241922.

[j. m. reese]