Non-Roman fonts used: SPTiberian, SPAtlantis, SPIonian
Vol. 10, pp. 1-7
Rediscovery of the Fragment
On January 9, 1992, while we were working together on a photographic documentation project at the Shrine of the Book, Reed came across a small glass negative depicting an ovoid parchment fragment of a Dead Sea Scroll. I The negative itself was unmarked and apparently was not assigned an inventory number, as is normally the case for photographic negatives within the Shrine's collection.
The fragment documented on this negative was easily recognizable as part of the Targum of Job from Qumran Cave 11.2 First, the script hand of 11 QtgJob is easily recognizable as a distinct hand to anyone familiar with that text. The hand on this fragment easily can be classed as that of the targum's scribe. Second, the shape of the fragment itself and the way the lines are ruled and spaced matches the equally distinctive ovoid shapes of the targum's separate fragments as well as the manner in which the targum's lines are ruled and spaced (note especially fragments 10-22 for a particularly close comparison). Third, the pattern of decomposition along the fringes of the fragment closely matches the similar pattern on most of the edges of the Cave 11 Targum. Fourth, and most decisively, the characteristic introductory dialogue formula [rm])w bwy) )n( begins the text. This, coupled with the fact that the language of the text is obviously Aramaic, left no doubt that the negative depicts a column fragment of 11 QtgJob.
It was not until a couple of months later, when we were going through the routine process of searching and cataloguing the negatives copied during the Shrine project, that we turned to a closer consideration of this fragment of the Targum. It was Reed who first realized that this parchment was not included among the fragments published in 1971 in the editio princeps; nor is it mentioned or apparently known by the authors of any subsequent editions or studies of 11QtgJob.
As of the writing of this preliminary report, the questions regarding the history, whereabouts, and condition of the fragmentindeed whether the fragment still exists at allhave yet to be answered with certainty. Still, it is reasonable to assume that the fragment is simply misplaced rather than lost. The fact that a single negative was found in the Shrine's collection suggests the possibility that the vault of the Shrine is the most likely repository for this text.
According to the lists compiled by S. J. Pfann this particular negative was not shot at the same time that the other fragments of 11QtgJob were photographically documented in the "PAM" series (43.796-824), the former being shot on June 8, 1967 and the latter a good deal earlier in March, 1961.4 Since there are no gaps in the consecutive numbering of the PAM negatives, the logical assumption is that this fragment had already been separated from the scroll before the PAM photographs were made.5 As far as we can tell, no other photographic images-including those showing the scroll before it was completely unrolled, show even in part the column fragment under consideration here or the column fragments that would have been contiguous with it.
The June 8, 1967 date of the photographing of the targum fragment may offer a clue to its background. This work was done at the very same date that many of the Temple Scroll images were first made. This suggests that this piece was part of the same lot as the Temple Scroll material and was acquired at the same time and place, during the Six Day War. The fact that both Temple and Targum of Job scrolls come from Cave 11 may lend further credence to this speculation.
Physical description of the column fragment
As for the fragment itself, as noted above, it shows deterioration and carbonization of the skin along most of its edges characteristic of 11QtgJob. Besides this, there is a small hole in the lower middle of the parchment that slightly obscures some text in line 6. There are also some vertical abrasions on the parchment surface that apparently have resulted in the writing surface flecking off. As a result, textual information is partially obscured in lines 3, 5, 8 and possibly 9. In general the letters look more abraded than is the case for other fragments of the Cave Eleven Targum, making it somewhat more difficult to read. We may speculate that this also indicates that the fragment had a "separate life" from the other text fragments very early on, resulting in a more significant pattern of deterioration. The text is horizontally ruled, and, as is normal practice for Dead Sea texts, the letters are "hung" from the horizontal lines. A total of 10 partial lines are preserved on the fragment with the tenth line being vestigiali.e., only the top of a lamed is clearly visible. Since the vertical margins are not preserved on either the left or right, it is difficult to determine how to position the preserved text in reference to column width. It is worth noting that there is at least one other instance in the poetic portions of the Job targum (9:3=MT 25: 1) where the dialogue introductory formula is well indented; hence it is unlikely that the formula in line 1 is close to the right hand margin.
We read the text epigraphically as follows:
We interpret the text as follows:3
]m)w bwy) )n( [ 1.
] yd ytw(# ll+ Nm [ 2.
]k#)w (dn) )wlm ytx[ 3.
]mdq llm) hrdm rt)[ 4.
](dn)w hxkwh )lm) y[ 5.
]yl rm)y )m lkts)w[ 6.
]d( )wlm ym( lw([ 7.
]tdw +#q wr)[ 8.
]dql Nh y[ 9.
2r]m)w bwy) )n( 23:1[ 1.
] yd ytw(# ll+ Nm [ 2.
hnx]k#)w (dn) )wlm 3 ytx[n) 3.
yhw]mdq llm) 4 hrdm rt)[ 4.
](dn)w 5 hxkwh )lm) y[mp 5.
6]yl rm)y )m lkts)w[ 6.
]d( )wlm ym( lw([y 7.
]tdw +#q wr)[7 8.
M]dql Nh 8 y[nyd 9.
1 . ] 23:1Job answered and sai[d2
2. ] because of my speech which[
3. my [groa]ning. 31ndeed, I would know and I would fin[d him.
4. ]the place of his dwelling. 4I would speak before [him
5. ]my [mouth] I would fill (with) argument , 5 and I would know[
6. and I would understand what he would say to me. [ 6
7. would he] deal unjustly with me? Indeed, until[
8. 7] for truth and law [
9. ]my[ judge]. 8If forwa[rd
The text documented in the fragment and its placement vis-à-vis the rest of the manuscript
Since the first line of the text preserves a dialogue introductory formula with Job as the speaker, it is relatively easy to determine what part of Job is reflected in this fragment. The text begins chapter 23 and intermittently preserves textual information through verse 8. This fits well with the general pattern of text preservation in other fragments and running columns of the targum. That is, typically, 6 to 10 verses are preserved in the columns of the preserved portion of the manuscript. The fragment may now be placed between fragment six and seven and between columns 7 and 8, according to the numbering in the editio princeps. We therefore suggest that in all future studies or editions of the 11QtgJob the fragment under consideration here be labeled as fragment 6a and, by the same token, this text column should be cited in all future studies as column 7a.
There is one further significant bit of evidence that supports our proposal to place this fragment as we have suggested. If one measures the gaps between columns in terms of verses (using the Masoretic Text as a guide), the lacunae are usually 69 verses in length. For example, the gap between columns 4 and 5 (MT 21:11 through 21:19) is 9 verses; between columns 5 and 6 (MT 21:28 through 22:2) is 9 verses; and between columns 6 and 7 (MT 22:1015) is 6 verses. If one were to place the next column 8 according to the editio princeps without inclusion of our "missing" column, one would have to account for a gap between MT 22:23 and 24:11, i.e., an otherwise unprecedented gap of 36 verses. If, on the other hand, the new column is inserted, then the gap between 7 and the new column 7a would be MT 22:23 through 22:30 or 8 verses, i.e., an entirely expected lacuna between columns.
But what about the gap between new column 7a and column 8? This would be a lacuna comprising MT 23:9 through 24: 11 - a gap of 20 verses, again an unprecedented interval in the Cave Eleven Job Targum. This leads us to conclude that there is a significant likelihood that yet another column fragment exists (or at least at one time existed) which covers part of this lacuna. Provisionally, we would therefore like to label this postulated column fragment as column "7b." If we assume a six verse gap between new column 7a and the postulated column "7b" and a similar gap between "7b" and column 8, we can suggest that fragment "7b" probably attested something in the range of 7 verses from about 23:15 through 24:5. If we continue tracking lacunae between the column fragments following column 8, a pattern entirely consistent with this assumption emerges. That is, the lacuna between columns 8 and 9 is MT 24:18 through 24:23 or 6 verses; the lacuna between columns 9 and 10 is MT 26:3 through 26:9 or 7 verses; the lacuna between columns 10 and 11 is MT 27:5 through 27: 10 or 6 verses; the lacuna between columns 11 and 12 is MT 27:21 through 28:3 or 6 verses; between columns 12 and 13 is MT 28:14 through 28:19 or 6 verses and between columns 13 and 14 is MT 29:1 through 29:6 or 6 verses.
Actually, if we are correct about postulating the existence of column "7b," there may remain a trace of physical evidence to its existence. The text fragment marked as 7 in the original edition attests the left hand margin of a column before column 8. Presumably, this would be the left margin of "7b." Unfortunately, there appears to be no trace of any text remaining within this vestigial part of the columnonly rule lines. Obviously, a new search must be instituted to determine whether 7a as well as "7b" (or a photograph thereof) actually exist or not. For the moment, we now can reclaim 39 columns of 11QtgJob rather than the previously published 38, with the prospect of an even 40.
Epigraphic description of the text
All letters are clear. While the last letter on the line is only vestigially attested, the remaining traces are entirely compatible with the expected reading of mem.
All letters are clear although a number of letters at the end of the line are abraded.
The first letter of the line is faint but is most easily read as h9et. The last letter on the line is only partially preserved and could be read as kap, nun or mem. Based on context we prefer kap. All other letters are clear.
While the first letter of the line is broken, its shape makes it a certain 'alep. The last letter of the line is partially obscured but is almost certainly a mem. All the other letters are clear.
Only the head of the first letter in the line can be seen. It could be read as waw or yod. Based on context, the latter is preferable. The second 'alep of the form read as )lm) is partially lost in an abraded area of the skin but is certain. The last letter in the line is broken but is certainly an 'ayin. All the other letters are clear.
Only the head of the first letter in the line is visible but the reading of waw is likely. The letter identified as kap in the form lkts) is partially obscured in a surface abrasion, but its identification is likely. The mem of the form )m is partially lost in a hole, but its identification is certain. There may be some trace of a waw or a yod after the last clear letter (lamed), but this is uncertain. All the other letters are clear.
There appear to be the traces of two letters at the beginning of this line. The first might be read as taw, nun or the latter half of a broken 'ayin The second letter might be read as a zayin, waw or yod. Based on context, we read the letters respectively as 'ayin and waw. The third letter in the line is certainly a lamed; however, it has an abnormally long tail-presumably a scribal slip. The last two letters in the line are both heavily abraded, but the likely readings are 'ayin followed by dalet. All the other letters are clear.
The first letter in the line is partially broken and somewhat faint; still, the reading of 'alep is certain. The fourth letter in the line, i.e., the one preceding shin-t@et looks at first like a waw. However, in all probability the head of the letter is lost in an abrasion of the letter on the right. The probable reading is qop. All the other letters are clear.
A trace of the first letter in the line can be seen, and we have restored it on the basis of context as a yod. However, the reading is highly uncertain. Just a trace of the right edge of a letter is visible at the end of the line. The probable reading is dalet although this cannot be considered certain. All the other letters are clear.
Line Ten: Only the top of a lamed is visible.
Only a few philological comments will be made at this time.
The dialogue introductory formula found here follows the same formula as elsewhere in the Cave Eleven Targum. Note in particular that the absence of the conjunction in the Qumran Targum before (nh is consistent with the other dialogue introductions in this text and is also consistent with the formula normally found in Standard Literary Aramaic as attested by Daniel and Ah9iqar.6 There is no reason to assume that the Vorlage here varies from the Masoretic tradition.
The form min t@la4l, "because" is presumably related to the later forms, e.g., mt@wl in JPA,7 myt@wl in CPA12 and met@t@u4l in Syriac.9 Note also the form bt@ll, which occurs in Egyptian Aramaic10 and in 1QapGen 19:16,20.11 As far as we are aware, this particular construction is not attested elsewhere.
The form mlw) occurs here and in line 7, in both cases functioning as an interjection or emphatic particle, "verily, indeed." In the first instance it translates mi4 yitte4n and in the second instance it renders lo4) )ak. The form must be related to CPA lw)12 which is used with precisely the same force, e.g., in the translation of a CPA biblical manuscript of Exodus 16:3, where lw) is utilized to render mi3 yitte4n. 13 Presumably the form is built on the optative particle lu4/lu4) known, e.g., in early Aramaic and in biblical Hebrew. The form mlw in Hermopolis 1:7 may well be related, in which case the latter should not be connected to vml) or, more specifically to Akk. mala and interpreted as an adverb "much, as much." Instead, it should be taken as an interjection "indeed," in accordance with this form in 11QtgJob. This would fit the context in Hermopolis 1:7 rather well: "...indeed, I do for H9arudj as Banit should do for me."14. We assume that the initial mem represents a prefixed interrogative ma-"what?"-vocalize: mallu4).15
The form ho=ka4h9a4h we take to be the haphel infinitive of vykh9 "to admonish." We read the form as a substantive, equivalent to the cognate noun it translates in Hebrew, to=ka4h9o3t. The adverbial force of the form follows the syntax of the original Hebrew.
The initial form is assumed to be a pael or aphel of the root v(wl, "to act, deal unjustly." The verb might be assumed to be a Hebraism in which case the pael may be preferable, but it could just as easily be Aramaic like its Syriac cognate,16 which is attested in the aphel with this meaning. In that case the form should be read ly([y] rather than lw([y]. The verb apparently renders Hebrew ya4ri4b "he (i.e., God) contends," thus considerably sharpening the force of the text in the translation.
If we understand this correctly, the text varies here from the Masoretic tradition. Note however, that the text is certainly reminiscent of the LXX reading: a0lh&qeia ga&r kai/ e!legxoj, "for truth and reproof..." Following a suggestion of Dhorme's, we might postulate a Vorlage something like ki4 ya4s]a4r we5na4koah9, "for uprightness and straightness...17
This text fragment fits well with other parts of the Cave Eleven Targum, and there can be little doubt-based on the evidence surveyed here- that it was an original part of that document that became separated from it at an early point. Although it certainly has its points of interest, the new fragmentary column offers nothing in the way of extraordinary surprises. Of course, the big surprise is that it exists at all and perhaps another column fragment as well! We can only hope that further investigation may reveal other sightings of missing texts from the Targum of Job from Qumran Cave Eleven.
1. We are employing the nomenclature established by W. S. Ginell, Report on Dead Sea Scroll Studies (Marina del Rey - Getty Conservation Institute, 1993). According to Ginell (private communication), it is incorrect to label the media on which the scrolls were written as "leather," since the skins are untanned. The proper description should be "untanned sheepskin parchment" (cf. p. 10) or, more simply, "skin."
2. The editio princeps is J. P. M. van der Ploeg, A. S. van der Woude, B. Jongeling, Le Targum de Job de la Grotte XI de Qumra=n (Leiden: Brill, 1971); cf. also M. Sokoloff, The Targum to Job from Qumran Cave XI (Ramat-Gan: Bar Ilan Univ., 1974). The latest comprehensive treatment of the targum is in K. Beyer, Die aramäischen Texte vom Toten Meer (Göttingen: Vandenhoeck and Ruprecht, 1984), 280-98.
3. In the drawing, the filled-in characters represent clearly inked letters or traces of ink, the outlined parts of letters represent reconstructions. Only letters which are at least partially in evidence have been reconstructed. For convenience, chapter and verse markings have been noted in the drawing.
4. The image is now published in E. Tov with the collaboration of S. J. Pfann, The Dead Sea Scrolls on Microfiche (Leiden: Brill, 1993), see microfiche no. 121, second row, first image on the left. In S. Reed, M. Lundberg, Inventory List of Photographs (Leiden: Brill, 1993),121, it is listed as SHR X15; however, the proper accession number is now SHR 6215. For the dates of various 11QtgJob photographs, cf. Pfann, "Chronological List of the Negatives of the PAM, IAA and Shrine of the Book," in the Companion Volume, pp. 92,94.
5. A. S. van der Woude has confirmed to me in a private communication that neither he nor the other editors of the editio princeps were aware that there were other fragments of 11QtgJob beyond those they published. Tov has also noted that the fragment was unknown to him before we drew it to his attention. We are grateful to Prof. van der Woude for waiving any publication rights to this fragment on behalf of himself and his Dutch colleagues and to Prof. Tov, who granted us permission to publish this fragment on behalf of the Dead Sea Scroll Publication Project of which he is Editor-in-Chief.
6. In Daniel the formula (nh ...w)mr ... occurs 23 times and twice in Ah9iqar (lls. 110, 118). The related formula (nw ... w)mryn occurs 5 times in Daniel while (nyn ... w)mryn and (nt w)mrt each occurs once. Finally,(nyt w)mrt occurs twice in Ah9iqar (11s. 14-15, 45) and (nw...w)mrw occurs once (1. 121). Cf. B. Zuckerman, "The Process of Translation in I I QtgJob; A Preliminary Study" Ph.D. diss. Yale University, 1980 (Ann Arbor: Univ. Microfilms 8235321, 1988): 292, nn. 51, 52.
7. Cf, e.g., M. Sokoloff, A Dictionary of Jewish Palestinian Aramaic (Ramat Gan: Bar Ilan Univ., 1990), 301.
8. Cf., e.g., F. Schulthess, Lexicon Syropalaestinum (Berlin: Reimer, 1903), 109.
9. Cf, e.g., J. Payne Smith, A Compendious Syriac Dictionary (Oxford: Clarendon, 1903; rpt. 1967), 267; also s. v. t@la4l, "shade, protection," p. 175.
10. Cf. Cowley 38:5, bt@ll )lh s]my) "with the help of the God of heaven." For other forms using vt@ll in Egyptian Aramaic, see, e.g., Beyer (N 2), 590.
11. For further discussion, see, e.g., J. A. Fitzmyer, The Genesis Apocryphon of Qumran Cave I (Rome: PBI, 1971),113. The form in line 20 is bt@lyky perhaps read bt1<1>yky. As pointed out to me by M. Luijken Gevirtz, in this particular context there is a play upon t@ll in the sense "shade"; that is, the cedar in Abram's dream was not only spared "with the help of the date palm" (Fitzmyer's translation, p. 6 1) but, quite literally, because it was "in the shade of the date palm." Cf. B. Zuckerman, "'For your sake...': A Case Study in Aramaic Semantics," JANES 15 (1983): 129, n. 40.
12. See Schulthess (N 8), 101, C. Müller-Kessler, Grammatik des Christlich-Palästinisch-Aramäischen; Teil 1, Schriftlehre,Lautlehre,Formenlehre,Texte und Studien zur Orientalistik 6(Hildesheim:Olms, 1991),149 §126.96.36.199. I am grateful to Douglas Gropp, who pointed out to me the connection of this form in 11QtgJob to the CPA form as well as the potential connection to the Egyptian Aramaic form, mlw (see below).
13. See The Bible in the SyroPalestinian Version; Part 1, Pentateuch and Prophets, Hebrew Univ. Bible Proj. Monograph Series 4; M. H. Goshen-Gottstein, ed. (Jerusalem: Magnes, 1973), 25.
14. See E. Bresciani, M. Kamil, Le lettere aramaiche di Hermopoli (Rome: Academia Nazionale dei Lincei, 1966), 372. The passage reads: mlw kzy (bd )nh lh9rws9 kwt t(bd bnt (ly. The original editors read the form as a pl. imv. of vmll, "dite" with the force "tutti voi, voialtri." J. C. Greenfield and B. Porten, "The Aramaic Papyri from Hermopolis," ZAW 80 (1968): 222 and E. Y. Kutscher, "The Hermopolis Papyri," I0S 1 (1971): 113 proposed a connection to vml) and Akk. mala. Hence, the translation of Porten and A. Yardeni: "as much as I am doing for H9arudj thus may Banit do for me"; Textbook of Aramaic Documents from Ancient Egypt 1; Letters (Jerusalem, Hebrew Univ. Dept. of History, 1986), 14. J. T. Milik, "Les papyrus arameens d'Hermoupolis et les cultes syro-phéniciens en Égypte perse," Bib 48 (1967): 549, suggests that the scribe corrected his reading from an original m1w to h1w, a position endorsed by S. A. Kaufman, The Akkadian Influences on Aramaic, Assyriological Studies 19 (Chicago: Univ. of Chicago, 1974), 69. In any case, Kaufman voiced doubt regarding the likelihood of a connection of Egyptian Aramaic mlw to Akk. mala. It is probably most prudent to assume that mlw)/mlw is a deictic particle with a fairly broad semantic range, in consideration of the examples of its use cited here. Whether it could be employed with a specifically optative force =BH mi4 yitte4n is unclear. (Editor's s note: It is clear enough! I translate 'if only' both here and in Hermopolis 1; i.e. "if only DN would do for me as I am doing for PN" [SAK].)
15. The CPA subordinate conjunctive phrase, lmlw d/hyk lmlw d (cf., e.g., Mü1ler-Kessler [N 12], 149) should also be related to this form rather than to vml)/mala, pace Kutscher, ibid.; cf. Kaufman, ibid., n. 190.
16. Cf., e.g., Payne Smith (N 9),404-5.
17. See. E, Dhorme, A Commentary on the Book of Job, H. Knight, trans. (London: Nelson, 1967),346. The LXX reading may reflect a Vorlage in which a form of xkwt was read for xkwn under the influence of twxkwt in verse 4b. Note, in this respect, that e!legxoj is commonly a translation of txkwt /hxkwt in the LXX, cf. E. Hatch, H. Redpath, A Concordance to the Septuagint (Oxford: Clarendon, 1897),449-50. It is also quite possible that wr) here in 11QtgJob and ga&r in the LXX may reflect a doublet/ misreading, yb > yk under the influence of the last form in Job 23:6.