OTHER COPIES
The numbers are quite impressive, yet the ones listed are not the only witnesses to the scriptural texts found in the Qumran caves. As mentioned above, there are at least five copies of Greek translations of scriptural books: one of Exodus, two of Leviticus, one of Numbers, and one of Deuteronomy — all of them are pieces from Greek copies of pentateuchal books. Other small fragments may come from still more copies, though not enough text has survived to clinch the case (see 7Q3-5 and the discussion in Chapter 4 below). In addition, there are three manuscripts that have been identified as targums: one of Leviticus (4Q156) and two of Job (4Q157,11Q10), the last of which (11Q10) is extensively preserved. Besides these scriptural copies, there are other kinds of works that are valuable for a study of the scriptural text and its history. As is well known, the caves at Qumran have yielded a series of commentaries on prophetic works. The writers of these pesharim cite a passage from a scriptural book (occasionally books) and then explain the meaning of it. Having completed the commentary on that passage, writers of the continuous pesharim then move on to the next or another one found farther along in the book. 7 These citations from scriptural books and the many "biblical" quotations in other works (e.g., the Damascus Document) considerably augment the fund of information about the scriptural text in the Dead 7. The texts and translations with extensive commentary can be found in M. P. Horgan, Pesharim: Qumran Interpretations of Biblical Books (CBQMS 8; Washington: Catholic Biblical Association of America, 1979). Sea Scrolls. It so happens that, while only one tiny piece possibly containing Habakkuk survives from the relevant part of a manuscript of the Twelve Prophets, the text of the first two chapters of the book is extensively preserved in the commentary on it. There are also Tefillin (phylacteries) and Mezuzot, collections of scriptural texts placed in a small container and attached to one's arm (and head; see, for example, Exod 13:9) or doorway (see Deut 6:9; 11:20). Since it is not always possible to distinguish the two types when only fragments are extant, the numbers may not be exact. But 28 texts identified as Tefillin were found at Qumran (21 in cave 4 — 4Q128-48; the others are 1Q13; 5Q8; 8Q3; XQ1-4) 8 and three at other sites; there are eight Mezuzot from Qumran (4Q149-55; 8Q4) 9 and one from Murabba'at. TEXT S FRO M OTHE R JUDEA N DESER T SITE S Several additional places in the Judean Desert have yielded copies of scriptural books. Not nearly as many were found in them as at Qumran, but their contributions are noteworthy nevertheless. Masada (7): The finds at the famous site are securely dated in that they cannot be later than 73 or 74 C.E., the year when the fortress was taken by the Romans. The numbers are markedly lower than for the smaller Qumran site, consistent with the fact that a different kind of community used it. 1 0 8. The rabbinic rules regarding Tefillin or Phylacteries prescribe that four passages be included: Exod 13:1-10 ; 13:11-16 ; Deut 6:4-9; and Deut 11:13-21 . These passages also appear on the Qumran examples, though some of them contain other or rather expanded passages (e.g., the Ten Commandments). See L. H. Schiffman, "Phylacteries and Mezuzot," in Schiffman and J. C. VanderKam, eds., Encyclopedia of the Dead Sea Scrolls (2 vols.; New York: Oxford University Press, 2000), 2:675-77; D- Nakman, "Tefillin and Mezuzot at Qumran," in M. Kister, ed., The Qumran Scrolls and Their World (Between Bible and Mishnah; 2 vols.; Jerusalem: Yad Ben-Zvi, 2009), 1:143-55 (Hebrew). 9. Schiffman, "Phylacteries and Mezuzot," 2:675-77. The prescribed passages are Deut 6:4-9 and 11:13-21 , though the Qumran copies also have extra ones such as the Ten Commandments. 1 0 . The texts were published in S. Talmon, Hebrew Fragments from Masada (Masada VI: Yigael Yadin Excavations 1963-1965: Final Reports; Jerusalem: Israel Exploration Society, Hebrew University of Jerusalem, 1999), 31-97. Genesis ι Leviticus 2 Deuteronomy 1 Ezekiel 1 Psalms 2 Murabba'at (7 [6]) 1 1 Genesis 2 Exodus 1 (on the same manuscript as one of the Genesis copies) Numbers 1 Deuteronomy 1 Isaiah 1 Minor Prophets 1 (a relatively well-preserved scroll) Nahal H ever (3) 1 2 Numbers 1 Minor Prophets 1 (Greek, extensively preserved) 1 3 Psalms 1 Nahal Hever/Se'elim (2) Numbers 1 Deuteronomy 1 Se'elim (1) 1 4 Numbers 1 Sdeir (1) 1 5 Genesis 1 There are also copies of Joshua (1) and Judges (1) from an unknown location. As at Qumran, so at the other sites the books attested are pentateuchal and prophetic works along with Psalms. 1 1 . For the texts, see J. T. Milik in P. Benoit, Milik, and R. de Vaux, eds., Les Grottes de Murabba'ât (DJD 2; Oxford: Clarendon, 1961), 75-80 and 181-205. 1 2. The copies from Nahal Hever (other than the Minor Prophets scroll) and Nahal Hever/Se'elim were edited by P. Flint in J. H. Charlesworth et al., eds., J. VanderKam and M. Brady, consulting eds., Miscellaneous Texts from the Judaean Desert (DJD 38; Oxford: Clarendon, 2000), 133-66,173-82. 1 3. The edition is Ε. Τον, ed., The Greek Minor Prophets Scroll from Nahal Hever (SHevXIIgr) (The Seiyâl Collection 1; DJD 8; Oxford: Clarendon, 1990). 14. See M. Morgenstern in DJD 38:209. 1 5. See C. Murphy in DJD 38:117-24. NATUR E O F TH E TEXT S The scrolls, despite the limits dictated by their fragmentary state of preservation, have made significant contributions to knowledge about the texts of scriptural books and their history. General Comments Before looking at specific examples, it is fitting to reflect on the sum total of the scriptural manuscript evidence. First — and to state the obvious — the copies furnish the oldest original language evidence for the many passages they represent, centuries older than any other witness apart from some Greek papyri from the second and first century B.C.E. — Greek papyri that are contemporary with many of the scrolls. As the scrolls from Qumran were copied in the period between the third century B.C.E. and the first century C.E., they are several hundreds of years older than the most ancient Greek codices of the Bible (fourth century C.E. ), and they, in many cases, antedate by a full millennium the earliest extant copies of the Masoretic Text (MT). In an age when all texts had to be handcopied, the earlier the evidence the less opportunity, one hopes, there was for scribal lapses and other common copying errors to occur. There is no guarantee that older is better, but the ancient copies offer unique comparative evidence, allowing one to test whether the more recent (MT, LXX, etc.) and the more ancient copies (the scrolls from the Qumran caves) are the same, almost the same, or quite different in their readings and to draw conclusions from the results (e.g., are the changes systematic or are they of other kinds). While all of this is familiar enough, it bears repeating because, with the passage of time, it is too easy to forget what an extraordinary find the Qumran scrolls, including the scriptural ones, prove to be — discoveries in a place whose climate was thought to preclude preservation of ancient parchment and papyrus. Second, the manuscripts from the Judean wilderness provide evidence that scriptural texts were transmitted with considerable care by Jewish copyists. The differences between the Judean Desert texts and MT are indeed numerous though frequently very slight, often ones that do not affect the meaning of the text for most purposes (e.g., spelling changes, omission or addition of a conjunction). Statements in rabbinic literature describe the meticulous procedures used later in copying scriptural texts; it seems great care was also taken at an earlier time, as the Judean Desert texts suggest. The scribes were not transmitting only one form of the texts; yet, from whatever scriptural model they were copying, they presumably did the work with care according to prevailing rules of the profession. An interesting question is exactly what the scribes responsible for the Qumran scrolls understood proper transmission of a text to involve. The question will be considered below. Third, despite the more recent finds, only a very limited set of data has survived, and it yields a correspondingly limited perspective on the history and varieties of the scriptural texts. Nevertheless, the admittedly challenged perspective available today is a broader one than was accessible to all those talented text critics whose work preceded the Qumran and other Judean Desert finds. Before 1947, the textual evidence at their disposal was of relatively recent date: the manuscript trail for MT could be traced back no farther than ca. 900 C.E . , 1 6 and that for the Samaritan Pentateuch (SP) goes back to an even more recent date. 1 7 There are many Greek witnesses that are centuries older than the earliest copies of MT and SP, 1 8 yet, however valuable, they are translations, not copies in the original languages. Other than the second-first century Nash Papyrus, 1 9 there was little ancient Hebrew evidence on which to base one's study. Probes were made using scriptural citations in texts such as Jubilees, but the manuscript evidence for it is also relatively late. The Judean Desert discoveries, however fragmentary, are a wonderful supplement to the textual base and a very unexpected one. Among the greatest contributions of the new material is that in a number of cases there is now Hebrew manu16. The Cairo Codex, containing only the Prophets, dates from 895 C.E., while the Aleppo Codex, which once contained the entire Hebrew Bible but from which large parts are missing, comes from the first half of the tenth century. Codex Leningradensis, which underlies the latest editions of the Hebrew Bible, was copied in 1008 C.E. See, e.g., E. Wurthwein, The Text of the Old Testament (2nd ed.; Grand Rapids: Eerdmans, 1995)> 35-37· 17. The earliest copy may have been made in 1150 C.E.; Wurthwein, The Text of the Old Testament, 47. 18. Some papyri fragments date from before the turn of the eras (e.g., John Rylands Library 458, from the second century). The oldest full copy, Vaticanus (B), was copied in the fourth century C.E. 19. The papyrus bears the text of the Ten Commandments, with elements from both Exodus 20 and Deuteronomy 5, and Deut 6:4-5; see Τον, Textual Criticism of the Hebrew Bible, 118. script evidence for readings previously known only from the versions, especially LXX. The Textual Picture As the experts have noted, the Qumran texts permit one to see that at the time when they were penned (third century B.c.E.-first century C.E.) there was, from copy to copy, a degree of fluidity in the wording of scriptural texts — just exactly as one might have expected. There was not a single, completely uniform, accepted wording of a scriptural book such as Genesis or Isaiah — something that would have been virtually impossible when all copying was done by hand. This is not to say that there was free variation in the wording of texts. Rather, within fairly narrow limits (in most cases) there are noticeable differences from manuscript to manuscript. Some examples will illustrate differing measures of variation. To present an overview of the range of evidence, the language employed by Eugene Ulrich for what he calls the "four principal categories of variation detectable through comparison of the Qumran manuscripts, MT, SP, and OG" will be useful. 2 0 Orthography Anyone who has read the scrolls found in the caves of Qumran is aware that the scribes made much more frequent use of consonants to mark the presence of certain vowels (matres lectionis) than one finds in MT. As someone said recently, they were rather more British than American in their spelling. Orthography is a category of textual variation that can easily be dismissed as devoid of significance — as documentation for a phase in Hebrew spelling and pronunciation, nothing more. But, by their very nature, matres lectionis represent a decision regarding the proper parsing of a form whenever the consonantal text is ambiguous or potentially ambiguous. At times the analysis is the one any Hebrew reader would have made, but at others deciding on the preferred reading and marking it by a fuller 20. E. Ulrich, "The Jewish Scriptures: Texts, Versions, Canons," in J. Collins and D. Harlow, eds., The Eerdmans Dictionary of Early Judaism (Grand Rapids: Eerdmans, 2010), 97-119, here 110 . OG = Old Greek. spelling offered more of a challenge. For example, in Isa 40:6 the consonantal text of MT has The Masoretes understood the form to be The text of the verse so analyzed reads: "A voice says, 'Cry out!' And he said, 'What shall I cry?'" In iQIsaa the spelling of the word is: ΓΠΏ1Κ1 . According to this reading the verse is worded: "A voice says, 'Cry out!' And I said, 'What shall I cry?'" The copyist/interpreter of MT saw the verse as a report about a conversation between the voice and a "he"; the spelling in the Isaiah scroll presents it as direct address and first person response. The reading of the Isaiah scroll is the one translated in LXX: και είπα. Individual Textual Variants Many differences in readings between manuscripts concern small items that are not matters of orthography. These populate every manuscript and can arise for various reasons. Here are a few examples. Isa 6:3 MT: Holy, holy, holy iQIsaa Holy, holy For whatever reason, the Qumran copy has only two instances of holy (this is not the only variant in Isa 6:3). 2 1 The absence of one element from the familiar threefold formula is supported by no other ancient copy of Isaiah. Gen 1:9 MT/SP Let the waters be gathered into one place (=mpa ) 4QGenh l LXX Let the waters be gathered into one gathering (- mpB ) In this instance, two Hebrew words looking almost alike were interchanged. The reading of MT/SP shows greater variation in word choice in the clause; the other reading involves using a noun associated with the same root as the verb (the root Dip ) of the sentence. Isa 45:7 MT: I make weal [ÛI^W ] and create woe [VI] iQIsaa I make good and create woe [3Π] 21. See P. W. Flint and E. Ulrich, eds., Qumran Cave 1: II, The Isaiah Scrolls (2 vols.; D J D 32; Oxford: Clarendon, 2010), 2:125. The cave ι manuscript uses an antonym to contrast with 5Π , not the less directly opposed Dl'rW. Isa 40:12 MT: measured the waters [D'Q ] iQIsaa measured the waters of the sea [W "72] The two readings differ only in the presence or absence of a second yod. Arguments could be mounted for the originality of either, though sea could be a more appropriate counterpart to the other elements of nature in the verse. 2 2 Isolated Interpretive Insertions Ulrich says of this category: "Learned scribes occasionally inserted into the text they were copying what they considered an appropriate piece of additional material." 2 3 In the Qumran period at least, scribes, while copying with diligence, still felt some freedom to take a more active role with regard to a scriptural text than simply transcribing it. One well-documented pattern in a series of scrolls is to blend or combine wording from parallel scriptural passages. For example, a person familiar with the Bible knows that the Ten Commandments are preserved in two places — Exodus 20 and Deuteronomy 5. In Exod 20:11 the reason for keeping sabbath is the model set by God in the first week of the world: Exod 20:11 MT For in six days the LORD made heaven and earth, the sea, and all that is in them, but rested the seventh day; therefore the LORD blessed the sabbath day and consecrated it. Deut 5:15 motivates it with the Israelites' experience of slavery in Egypt and the Lord's deliverance of them from it: Deut 5:15 MT Remember that you were a slave in the land of Egypt, and the LORD your God brought you out 22. See the discussion by Z. Talshir, "Biblical Texts from the Judaean Desert," in Kister, ed., The Qumran Scrolls and Their World, 1:118-19 (Hebrew). 23. "The Jewish Scriptures," 111 . from there with a mighty hand and an outstretched arm; therefore the LORD your God commanded you to keep the sabbath day. Near the end of Deut 5:15, one of the Qumran copies (4QDeut n ) has additional words compared to MT: ".. . to keep the sabbath day and to hallow it. For in six days the LORD made heaven and earth, the sea, and all that is in them and rested the seventh day; so the LORD blessed the sabbath day and hallowed it" (4:4-7). The words additional to the form in MT are from the parallel version in Exod 20:11. Combining material from parallel passages is a characteristic not only of some scriptural copies from Qumran but also of SP (and the 4QReworked Pentateuch manuscripts, the Temple Scroll, and other texts), although in Deut 5:15 SP does not add material from Exod 20:11. Psalm 145: In MT, one verse is missing from the acrostic psalm: although each verse begins with a word starting with the successive letters of the Hebrew alphabet, there is none for the letter nun, which should have appeared between the mem sentence in v. 13 and the samekh sentence in v. 14. It is natural to think it was dropped from the text by scribal error, even if the mechanism for the omission is not obvious. At this place where MT lacks the nun verse, other witnesses have: nQPs a LX X Faithful is God [nQPs a ]/the LORD [LXX ] in all his words, and gracious in all his deeds. The Hebrew word for "faithful" CJQN3) begins with nun; the two witnesses thus supply the missing verse. The «un-sentence was not secondarily stitched together from elsewhere in the Psalter as the expression figures only here. In this case one should remember that what looks to be a manifestly superior text is not necessarily the original reading, since the author of the poem could have been the one who carelessly omitted one of the necessary lines. In that case, someone would have come along later and made the face-saving addition. New and Expanded Editions of Biblical Books There are some cases where sizable and systematic variations separate the witnesses for scriptural books, including those from Qumran. Some examples include entire books; the best-known exhibit is Jeremiah. The versions of the book of Jeremiah found in MT and in LXX are of much different lengths. MT Jeremiah is estimated to be some 13 percent longer than the Greek text; also, the two locate some units differently. 2 4 Among the fragmentary Hebrew copies of Jeremiah found at Qumran, two are similar to the longer readings of MT (4QJer a ' c ) and two align closely with the shorter readings of LXX (4QJer b ' d ). The shorter version is generally regarded as textually superior; support from the two Hebrew manuscripts (4QJer b ' d ) shows that the LXX translator(s) did not arbitrarily subtract large amounts of text from their Hebrew model but rather rendered a Hebrew copy with a much shorter text than the one now found in MT Jeremiah. Conversely, the other two copies show that the scribes in the tradition eventuating in MT also reproduced an early form of the text. The only fragment surviving from 4QJer b happens to preserve a section that illustrates some shorter readings and a differing order of verses. MT: Jer 10:3-11: 3For the customs of the peoples are false: a tree from the forest is cut down, and worked with an ax, by the hands of an artisan; 4people deck it with silver and gold; they fasten it with hammer and nails, so that it cannot move. 5Their idols are like scarecrows in a cucumber field, and they cannot speak; they have to be carried, for they cannot walk. Do not be afraid of them, for they cannot do evil, nor is it in them to do good. 6There is none like you, Ο LORD; you are great, and your name is great in might, y Who would not fear you, Ο King of the nations? For that is your due; among all the wise ones of the nations and in all their kingdoms there is none like you. SThey are both stupid and foolish; the instruction given by idols is no better than wood! 9Beaten silver is brought from Tarshish, and gold from Uphaz. They are the work of the artisan and of the hands of the goldsmith; their clothing is blue and purple; they are all the product of skilled workers. îoBut the LORD is the true God; he is the living God and the ever24. See Τον, Textual Criticism of the Hebrew Bible, 320-27. The largest difference in location involves the oracles against the nations that are chs. 46-51 in M T but are found after 25:13 in LXX. For a study of the M T and LX X forms of the book (written before most of the cave 4 Jeremiah copies appeared in print), see J. G. Janzen, Studies in the Text of Jeremiah (HSM 6; Cambridge, Mass.: Harvard University Press, 1973)· lasting King. At his wrath the earth quakes, and the nations cannot endure his indignation. liThus shall you say to them: The gods who did not make the heavens and the earth shall perish from the earth and from under the heavens. The Greek version of the passage lacks w . 6-8, 10 (the parts in italics above). 4QJer b preserves words from w . 4, 9,11 (with words from v. 5 restored in two places), in the Greek order: 4, 5a, 9, 5b. In the column, line 5 has words from v. 4 at the end, while line 6 has words from v. 9 at the end and line 7 has words from v. 11 at the end. Line 5 and with g] old they fasten it with hammers = v. 4 Line 6 ]blue and purple = v. 9 Line 7 ] shall perish from the earth = v. 11 The Hebrew fragment shows that the Greek translator(s) worked with a Hebrew text much shorter than the one in MT. In general, one may say that manuscripts aligning frequently with the textual traditions embodied now in MT, LXX, and/or SP are found at Qumran, but these configurations do not exhaust the data or even represent it properly. As Emanuel Τον has written: If the tripartite division [that is, that there are texts aligning with MT or SP or LXX] is merely a matter of prejudice, attention should now be directed to the actual relation between the textual witnesses. The textual reality of the Qumran texts does not attest to three groups of textual witnesses, but rather to a textual multiplicity, relating to all of Palestine to such an extent that one can almost speak in terms of an unlimited number of texts. 2 5 Some copies do not fall into any of the old, familiar categories — agreeing with either MT, SP, or LXX — and chart a different course textually. A copious number of textual options were available at the time, and indeed many of them are represented at the one site of Qumran. The data at hand 25. Textual Criticism of the Hebrew Bible, 161; he thinks, nevertheless, that a few groups of closely related texts can be detected (see, e.g., p. 163). are limited, but there is no reason to surmise that Jewish experts were concerned about a measure of fluidity in the texts of scripural books until late in the first century C.E., when Josephus wrote a surprisingly strong statement about a fixed scriptural text (Ag. Ap. 1.38-42; see Chapter 3 below). An End to Fluidity Each of the texts found at the other sites, all of them a little later than the Qumran corpus, may fall into the pre-Masoretic category (though the books preserved at Masada are ones for which there were no variant literary editions), 2 6 perhaps suggesting that by the end of the first or beginning of the second century C.E. the textual plurality apparent in the Qumran scrolls had given way to a far greater uniformity. There may have been social and political reasons for this development, in that the people who happened to use and copy a certain type of text became the central or nearly the only element in society engaged in such activity after the destruction of the temple in 70 C.E. Τον comments: "It is not that M [= MT] triumphed over the other texts, but rather, that those who fostered it probably constituted the only organized group which survived the destruction of the Second Temple. Thus, after the first century CE a description of the transmission of the text of the Hebrew Bible amounts to an account of the history of M." 2 7 CONCLUSIONS FROM THE EVIDENCE After surveying the evidence, what conclusions may be drawn from it and which questions are suggested by it? Several experts have crafted comprehensive theories to organize the data now accessible. It has been difficult for textual critics of the Hebrew Bible to move out from under the impress of the older three-text model — that provided by MT, SP, and LXX. Frank Moore Cross, following William Foxwell Albright, postulated three local varieties of Hebrew texts, each of which was represented by a familiar witness: SP was a prime witness to the 26. E. Ulrich, "Two Perspectives on Two Pentateuchal Manuscripts from Masada," in S. Paul et al., eds., Emanuel: Studies in Hebrew Bible, Septuagint, and Dead Sea Scrolls in Honor of Emanuel Τον (VTSup 94; Leiden: Brill, 2003), 461-64. 27. Textual Criticism of the Hebrew Bible, 195. more expansive Palestinian form of the text for the books it includes, LXX to the Egyptian Hebrew text, and MT (apparently by default) to the Babylonian text. 2 8 By the time of the Qumran copies, these local texts were no longer isolated; all were found in the land of Israel — in fact, at the one small site of Qumran, so that each type could influence the others for good or ill. Shemaryahu Talmon focused more on the role of different groups in the survival of the texts associated with them and the corresponding loss of many other kinds: Jews preserved MT, Samaritans continued to copy SP, and Christians transmitted LXX. 2 9 Emanuel Τον broadened the horizon somewhat, although he too has reckoned with the traditional witnesses in naming five categories of texts, though they are a strange combination in some respects. They are: Proto-Masoretic (or Proto-Rabbinic), in which category, on his reckoning, 47 percent of the copies from Qumran fall; Proto-Samaritan (6.5 percent); Close to the Hebrew Source of the LXX (3.3 percent); Non-Aligned texts (47 percent). He also speaks of a fifth category — texts copied in the Qumra n practice (or spelling), though he does not give a percentage for them since they are included in the other four. An obvious criticism of Tov's percentages is that he has inflated the numbers for the MT-like category, placing in it any text that is equally close to MT and SP or LXX. Peter Flint calculates that "of the fifty-seven (47 percent) analyzable Qumran biblical scrolls that supposedly fall into this [proto-MT] category, only twenty-four . . . are strictly close to the traditional text, while the other thirty-three . . . are as close to the Masoretic Text and either the Samaritan Pentateuch or the Hebrew source of the Septuagint." 3 0 Ulrich has focused on the history of texts for 28. See, e.g., Cross's essay, "The Evolution of a Theory of Local Texts," in Cross and S. Talmon, eds., Qumran and the History of the Biblical Text (Cambridge, Mass.: Harvard University Press, 1975), 306-20. 29. For a summary of his approach, see Talmon's essay, "The Textual Study of the Bible — A New Outlook," in Qumran and the History of the Biblical Text, 321-400. In his lengthy study he proposed "that a major problem to be investigated with regard to the history of the Bible text is not so much the existence of a limited plurality of text-types, but rather the loss of other presumably more numerous textual traditions. Thus phrased, the issue of whether a single Urtext broke up into 'three distinct local families' in which subsequently and separately manuscript variants emerged, or whether conversely, primal traditions which varied among themselves to a limited degree progressively lost their lease on life and ultimately crystallized in a restricted number of Gruppentexte should be studied from a new angle" (327). 30. P. Flint, "The Biblical Scrolls and the Text of the Hebrew Bible/Old Testament," in J. VanderKam and Flint, The Meaning of the Dead Sea Scrolls: Their Significance for individual books and thinks the periods in which they were composed and transmitted overlapped. In a numbe r of cases, what he calls variant literary editions arose, with the newer sometimes but not always replacing the older. 3 1 One of the many strengths of his approach is that it is not as constrained as the others are by the older three-witness model. N EW EVIDENCE AN D THE TEXT-CRITICAL QUEST Discovering what the biblical authors and editors actually wrote would seem to be a noble aim, one to which many have devoted great industry. It should be acknowledged, of course, that even the more recently accessible manuscript evidence is far removed in time from the earliest forms of the texts of scriptural books and sections, even if there is dispute aplenty about when the various compositions and sections of them were penned and arranged. If one follows those who think much of the Hebrew Bible reached its ultimate form in the Persian period, the Judean Desert manuscript finds take one back only to a point a few centuries later. That, of course, is much better than the situation confronting earlier scholars, but the chronological gap between the earliest written form(s) and the surviving manuscript evidence remains considerable. While that gap is a fact, it is also a fact that the student of the Hebrew Bible is, comparatively speaking, in a rather advantageous position. For example, the text of Plato's works, apart from some fragmentary second-third century C.E. papyri, is based on fifty-one manuscripts copied in the ninth century and later. 3 2 To see how it is possible to do better than one could before although puzzles remain (with the Qumran evidence supplying some new ones), it is instructive to examine the passage that is perhaps the best-known textual variant in the Qumran scrolls — the longer reading preserved in Understanding the Bible, Judaism, Jesus, and Christianity (San Francisco: HarperSanFrancisco, 2002), 146. 3 1. A number of Ulrich's essays on the subject appeared in his The Dead Sea Scrolls and the Origins of the Bible (SDSSRL; Grand Rapids: Eerdmans, 1999). For a recent statement of his theory, see "The Evolutionary Production and Transmission of the Scriptural Books," in S. Metso, H. Najman, and E. Schuller, eds., The Dead Sea Scrolls: Transmission of Traditions and Production of Texts (STDJ 92; Leiden: Brill, 2010), 209-25. 32. See G. Fine, ed., The Oxford Handbook of Plato (New York: Oxford University Press, 2008), 71; the oldest manuscript is dated to 895 C.E. (the very date of the oldest surviving witness of MT). 4QSama at the transition from ι Samuel 10 to ι Samuel 11 . Though it involves a longer stretch of text, it may be that a simple scribal error led to the omission of an entire paragraph that is found in 4QSama , a genuine copy of the books of Samuel. First the data will be presented followed by a look at plausible explanations for the available manuscript evidence. For the passage in question, ι Sam 10:27-11:1, there are four rather old configurations of the text that confront the reader and that must be considered in interpreting it. As those who have examined the passage have noted, the phrase at the end of the MT form of 1 Sam 10:27 — WlflÖ D TP I — and the Hebrew expression presupposed by the LXX rendering of the beginning of 11:1 — ΐίΠ Π 1ftD TP 1 — have sufficient look-alike qualities that, if they occurred in the same copy, could have caused textual mischief for a scribe who was not particularly alert. Here is how the witnesses handle the section (with the look-alike phrases in italics). MT: 10:27 · · · They despised him [Saul] and brought him no present. But he held his peace. 11:1 Nahash the Ammonite went up. . . . 4QSama (X frg. a 5-9): . . . They despised him and brought him no present, blank [Now Na]hash, king of the Ammonites, had been grievously oppressing the Gadites and the Reubenites. He would gouge out the right [ey]e of e[ach] of them and would not grant [I]srael a [deliv]erer. No one was left of the Israelites ac[ross the Jordan who]se right eye Naha[sh, king of] the [A]mmonites, had not [go]uged out. B[u]t there were seven thousand men who [had escaped from] the Ammonites and had entered Jabesh-gilead. About a month later, Nahash the Ammonite went up . . . 3 3 LXX . . . They despised him and brought him no present. 11:1 About a month later Nahash the Ammonite went up. . . . Josephus, Ant. 6.67-71: he seems to assume the ending of 10:27 as it is in 4QSama and LXX (that is, without But he held his peace), followed by the statement About a month later, after which he offers the additional material in a form resembling but not identical to the 33. The translation of the section is from the NRSV; brackets have been inserted to show where there are gaps in the fragment. extra paragraph in 4QSama . He then resumes with 11:1 without repeating the introductory About a month later. The textual situation may be clearer if the three key items are represented with letters, yielding the following diagram: X = But he held his peace (ΒΡ1Πο3 VPI) Y = the extra paragraph Ζ = About a month later (Win TPI) The individual witnesses contain these elements: MT: X 4QSama : Υ + Ζ LXX: Ζ Josephus: Ζ + Y That is, no surviving text preserves all three items represented by the letters. Yet, if the look-alike readings X and Ζ and the additional paragraph were present in an earlier copy (one that is no longer extant) and X and Ζ were the triggers for the omissions that yielded the varied texts now available, one could say about the versions: MT: the copyist whose work gave rise to the form of the text now represented in MT skipped from the end of X through Z, thus omitting Y and Z. This may not be a standard form of haplography, but it could have happened. 4QSama : the copyist omitted X but transcribed Y and Ζ — a procedure that seems strange. LXX: the copyist skipped X and Y and preserved only Ζ — a standard case of haplography. Josephus: skipped X and placed Ζ before Y (that is, he changed the position of a date in the text, something that he does elsewhere). 3 4 With identifiable mechanisms for haplography, it is reasonable to suppose, given the surviving readings, that omissions of different portions of text occurred. But this does not account for the form of the 4Q copy. 34. E. C. Ulrich, Jr., The Qumran Text of Samuel and Josephus (HSM 19; Missoula: Scholars, 1978), 168 (see 166-70 for his analysis of the text-critical issues). Despite agreement between 4QSama and Josephus in attesting additional material closely related in content and at the same spot, the two are not identical. The parallel section in Josephus is much longer and includes some items not present in the Qumran manuscript, such as the report about the military purpose served by King Nahash's grisly policy of gouging out the right eyes of Israelites across the Jordan. Yet, as Ulrich has shown, the elements found in the Qumran plus appear in Josephus's Antiquities and in the same order. There is one exception to the common order — Josephus places the "About a month later" statement before the plus, not after it as in 4QSama . In the DJD edition of 4QSama , Cross wrote: "It is possible that the phrase ΙΐΠΠ 1Ώ3 VP1 occurred in a Hebrew text both there [i.e., where Josephus has it, before the plus] and in 11:1 , thereby triggering the haplography of the whole paragraph. In any case, the W")nttD TP1 of M is best seen as a corruption of ΙΣΠΠ VP1 after the haplography." 3 5 If he is correct, MT would preserve only a misreading of one trigger for omission which it mistakenly attaches to the end of 10:27. None of the versions seems to offer the earliest form of the text, although, with the extant evidence, one can surmise how it may have read. The most parsimonious explanation may be the one suggested by Cross: the "About a month later" phrase occurred on both sides of the plus. The text represented in LXX is easily explained as a result of haplography, from instance one to instance two of the phrase, with omission of the intervening material. The textual tradition now found in MT did the same but was further corrupted when ΙΖΠΠ 1Ώ3 VP1 was misread as/altered to Φ'ΊΠΏ^ TPI to fit the context. Josephus retained the first instance of the trigger phrase and the extra material, while he lacks the second instance of the phrase, possibly to avoid repetition of a date he had just mentioned. The most difficult textual witness to explain is 4QSama . It lacks the first instance of the trigger but includes the extra material and the second instance. 3 6 35. F. M. Cross, D. W. Parry, R. J. Saley, and E. Ulrich, eds., Qumran Cave 4: XII, 1- 2 Samuel (DJD 17; Oxford: Clarendon, 2005), 66. 36. The passage as represented in the various witnesses has, of course, received extensive analysis, with various solutions proposed. See the lengthy bibliog. in DJD 17:1-2 . A. Rofé has taken a different approach to the plus in 4QSama and Antiquities: he judges it to be neither an original reading nor a textual variant but a midrash on points unclear in the passage and based on material located in other places in the scriptures. See, e.g., his essay, "The Acts of Nahash according to 4QSama , " IE/3 2 (1982): 129-33. Τον is among those who consider the longer form of the text in 4QSama to be original; Textual Criticism of the Hebrew Bible, 342-44. The Qumran copy itself shows a haplography within — but not the same as — the one under discussion. In it the words "About a month later Nahash the Ammonite went up and besieged Jabesh-" are written in a supralinear position, but by the scribe who recorded the text, not by another hand. The cause for omission of these words was the repeated name Another kind of issue that has become more pressing with the availability of the scriptural copies from the Judean Desert is the distinction between textual and exegetical variants. The question arises especially in works that one would not classify as "biblical" manuscripts but in which scriptural texts are adduced in some form. It is a problem one encounters in dealing with the many citations and adaptations of scriptural texts in a work such as the book of Jubilees. The writer very frequently quotes material from Genesis or Exodus, at times whole lines, at other times just phrases. 3 7 The issue is whether differences in wording from, say, MT or LXX reflect readings of Genesis and Exodus manuscripts or whether they are due to the way in which the writer adapted the material to the new contexts in Jubilees. Which are textual variants in distinction from textual interpretations and modifications? A textual variant would be one that arises in the course of manuscript transmission; an interpretive variant would likely arise in a different way. The issue becomes more complicated because, it seems, scribes felt it was within their rights to help the text along a little. Ulrich has defined what he calls "individual textual variants" (see the list above, where they are the second type) as ones differing from the parent text being copied and consisting of unintentional changes — "e.g., numerous types of errors, inadvertent substitution of lectiones faciliores, loss of letters, loss of one or more words through inattention or parablepsis" — and intentional ones — "clarifying insertions, scribal correction (whether correct or not), additional information, linguistic smoothing, euphemistic substitutions, literary flourishes, theological ideas." 3 8 What Ulrich includes under his category "intentional variants" is an intriguing set of differences vis-à-vis the parent text — various kinds of changes such as "clarifying insertions." Some examples occur in the Temple Scroll. Aspects of the relationship between this lengthy composition and the 37. See J. C. VanderKam, Textual and Historical Studies in the Book of Jubilees (HSM 14; Missoula: Scholars, 1977), 103-205, for a listing and study of the material. 38. Ulrich, "The Jewish Scriptures," 111 . legal sections of the Pentateuch are well known: it often collects elements from all texts on a topic (e.g., a festival) and works them into the first mention of the subject in the scriptural text that is being treated. 3 9 But the relationship changes in the latter parts of the scroll — in the Deuteronomic Paraphrase, where Deuteronomy 12-23 serves as the base for cols. 53-58 and 60-62 and where the scroll adheres more closely to the base text. Regarding the Temple Scroll as a whole, Lawrence Schiffman has written: "It is clear that the author/redactor and his sources had before them Vorlagen of the canonical Torah, in its present shape, which demonstrated genuine textual variation when compared with the Masoretic Text (MT). To this textual base, the author(s) added their own interpretations and adaptations. One of the challenges of scholarship is to distinguish these layers." 4 0 From the Deuteronomic Paraphrase Schiffman examined a series of readings in which its text agrees with LXX against MT and the reading has halakhic significance. These are good test cases for distinguishing textual from interpretive variants — for determining what the original wording of the text might have been. Consider, for example, the variants in TS 54:i9-55:i//Deut 13:7 (Eng. v. 6), where MT reads: "If anyone secretly entices you — even if it is your brother . . . or your mother's son.... " The enticement is to idolatry, and in this case, however close the relative or neighbor guilty of the offense maybe, he is to be executed. Presumably those closest to the offending party are to be witnesses against him, something not allowed in other kinds of legal cases. LXX Deut 13:7: έάν δέ παρακάλεση σε ό αδελφός σου εκ πατρός An added bonus in this case is that 4QDeut c frgs. 22-23 line 1 reads the word TDK, suggesting it contained the longer reading, and SP also supports it. The longer reading specifies that both the brother who is the son of the mother and also the brother who is the son of the father are relatives 39. See, e.g., F. Garcia Martinez, "Temple Scroll," in Schiffman and VanderKam, eds., EDSS, 2:927-33. 40. L. H. Schiffman, "The Septuagint and the Temple Scroll: Shared 'Halakhic' Variants," in G. J. Brooke and B. Lindars, eds., Septuagint, Scrolls and Cognate Writings (SBLSCS 33; Atlanta: Scholars, 1992), 278; repr. in The Courtyards of the House of the Lord: Studies on the Temple Scroll (STDJ 75; Leiden: Brill, 2008), 86. TS 54:19: X MT Deut 13:7: -|Ö"1»KK p Ρ ΤΠΧ "Ι^Ο* Ό σου ή εκ μητρός σου to whom the law applies. Perhaps one would have inferred this from the shorter formulation, but the longer reading makes the point explicit. In such an instance it is easy to formulate a case that either reading could be original. In favor of the shorter one in MT, one could maintain that the less specific reading is likely to be earlier and the longer one is an attempt to prevent misunderstandings. In favor of the longer reading one could argue that the shorter reading is haplographic: a scribe's eye jumped from the first instance of the word p to the second and thus skipped over the intervening words. Both readings would qualify as textual variants because they appear in copies of Deuteronomy itself, not only in the Temple Scroll.41 Schiffman does not decide which reading is original but simply notes that the compiler of the Deuteronomic Paraphrase in 11QT found the reading in his scriptural Vorlage. It may be that MT is defective in this place. The ancient "biblical" scrolls (or, rather, in almost all cases, fragments) from the Qumran caves reveal much about the transmission of the scriptural texts in the later Second Temple period. For one, they place before the reader's eyes many examples of how Jewish scribes transmitted texts that later became constituents of the Hebrew Bible; they also document numerous minor differences between copies, the small sorts of variants that beset any work and that are of interest primarily to text critics. In some cases, however, there is evidence for larger variation between copies, as with the manuscripts of Jeremiah. No discussion in the scrolls themselves regarding variant wordings in different copies of scriptural books has been found, but the modern experts who work on these texts agree that the last centuries B.C.E. and the first century C.E. were times of considerable fluidity in the wording of scriptural texts. An appropriate way in which to end the chapter is with a passage found in Pesher Habakkuk. The commentary on the prophetic book, like other texts in this category, quotes a passage and then offers an interpretation of it. Sometimes, in the comment, the expositor demonstrates that he is aware of a wording of the text at variance with the one he had just quoted. His practice was to use both readings, as though it was a bonus to have more text on which to comment. MT Hab 2:16 can be translated as: You will be sated with contempt instead of glory. Drink, you yourself, and be uncircumcised ['ΠΪ/ΓΠ] . 41. The status attributed to the Temple Scroll by those who copied and used it is a debated point. See Garcia Martinez, "Temple Scroll," 930. The text of Hab 2:16 quoted in Pesher Habakkuk before the interpretation reads: You will be sated with contempt instead of glory. Drink, you yourself, and stagger [^ΪΠΠΙ]. (11:8-9) Then, in the commentary to the passage the writer speaks about being uncircumcised (11:13), the reading found in MT, and later mentions the cup of the Lord's wrath — reflecting the reading in his own scriptural text. 4 2 In this case the expositor exploited the two readings; he did not lament their existence. Those who, like the commentator, could read works such as Habakkuk seem to have been comfortable with a level of textual variation in the sacred books, much as readers of the English Bible cope well with the numerous divergent translations available today. There are many other issues that arise in connection with the so-called "biblical" scrolls from Qumran. To this point there has been no discussion of whether there was a canon of scripture, a Bible, at the time of the scrolls, and, if so, which works belonged in it. These and related matters are the subject of the third chapter. The second is devoted to scriptural interpretation in the scrolls. 42. W. H. Brownlee noted and discussed the two readings in The Text of Habakkuk in the Ancient Commentary from Qumran (JBLMS 11 ; Philadelphia: SBL, 1959), 76-78; on 118-23 he list this and four other examples of the phenomenon that he calls "dual readings."
Please share us on Facebook and Twitter & leave a comment.
If you don't see, what you are looking for with Amazon let us know.
All thanks to the creator, life, prosperity & good health to you
Email Neoconda Oshun if you have a question about anything
No comments:
Post a Comment