Skip to: Site Menu | Main content

Redating the Dead Sea Scroll Deposits at Qumran: the Legacy of an error in Archaeological Interpretation

(page 4)
[End of paper. Questions and answers are reconstructed from memory and lightly edited for clarity.]

Now I hope there is time for some questions—the tougher the better.

Q. What about the inkwells found at Qumran?

A. About five inkwells have been identified as coming from Qumran. Since this is more than have been found at any other site, this has been cited to argue that scrolls were produced at Qumran. And if some of these inkwells were from Period II, this would suggest, so the argument goes, that scrolls were produced at Qumran in Period II, which then ended up in the caves. But first of all, it is not controversial that most, or at least a large number of the texts, were brought to the site, imported. The thinking is that a large number of the texts were imported, while some others were produced at the site. But citing the inkwells to show this runs into some problems. Two of these five inkwells appear to be from Period III, post-68 CE. One of the inkwells was reported by de Vaux found in a Period III context. This is in Humbert and Chambon 1994 (locus 31). Another of the Qumran inkwells, this one from the antiquities market, was argued to be Period III in an article by Torleif Elgvin and Stephen Pfann a couple of years ago. [T. Elgvin and S. Pfann, "An Incense altar from Qumran?" Dead Sea Discoveries 9, 2002: 20-33.] Elgvin and Pfann do not claim full certainty that that one is Period III; they say it could be Period II, but they argue that it is Period III. They give reasons why it looks like Period III to them. But Period III is later than anyone says the scrolls were deposited. This is when de Vaux says Romans were at the site. Now some are saying that Period III was maybe Jewish inhabited—different Jews resettling Qumran after the destruction of 68 CE, but Jews. In any case, Period III is later than anyone has the scrolls deposited, yet two of the five inkwells—one for sure, according to de Vaux’s record of its find-spot, and the other argued to be—are from Period III. What is one to make of this? Of course, one could say maybe de Vaux’s Period III inkwell was found out of context, but nothing in the published information says that—that would be simply ad hoc. If there are inkwells in Period III which have nothing to do with the scrolls, what can be concluded from inkwells before Period III? It just isn’t known. The inkwells confirm that there was writing happening at Qumran, but that is already known from ostraca found at the site, so that is nothing new. None of the writing found at the site has anything to do with the scrolls found in the caves, either—no distinct wording, no identity of a scribe, etc. If anyone can get better information out of these inkwells, I’d like to know.


Q. Was there a motivation to have a first-century CE dating of the scrolls in the early years of the scroll discoveries since people were asking whether these texts said anything about Jesus?

A. This raises an interesting point. There was a lot of looking for Jesus and John the Baptist, etc. in the scrolls in the early years. But that does not seem to have been a factor in the scroll deposit dating. Scholars today know that none of the texts are written as late as the first century CE. Everyone agrees on that. Well, Eisenman and Thiering think there are first-century CE references in the texts, but their views are not accepted by other Qumran scholars. In fact, ironically, Eisenman uses the argument (paraphrased): "why would the people of the Qumran scrolls spend all of their time writing peshers and other texts about events and persons of a century or more earlier, and never write about what was happening contemporary to themselves, in the first century CE?" Given the premise of CE deposits, Eisenman’s argument has a point. The problem is the premise. The Qumran field already knows the text compositions are not later than 1st century BCE. The proposal here is simply to move the text deposits earlier in agreement with where Qumran scholars already have the end of text compositions.


Q. But were there underlying motivations behind the strength of the First Revolt, first-century CE deposit dating?

A. I don’t think anything was going on other than honest mistakes based on responses to information as it became available, in terms of how it originated. De Vaux’s interpretation of Qumran’s archaeology is filled with early mistakes which later are corrected or seen to be somewhat different, in many areas of interpretation. If de Vaux got the scroll deposit datings mistaken, that is only in keeping with what has happened in other details and phenomena associated with the site. As for the tenacity with which the First Revolt dating is held after it got started, this would get into unverifiable speculation, into matters of psychology and so on, that are probably not fruitful (because unverifiable). For example, it could be argued that the 1st century CE is the century everyone knows about—it has Josephus, the New Testament, Jesus, the fall of the temple, etc.—whereas the first century BCE is (relatively speaking) a dark age. Who knows anything about or cares about the 1st century BCE?—apart from people like us who make our living studying this stuff. The 1st century CE is the century of interest. Did this enter into the appeal of the story of the Qumran scrolls deposited in caves at the First Revolt in the 1st century CE? It makes such a good story. There was also the story of Essenes at the site in the 1st century CE, which came to an end at the First Revolt. This story was linked to the First Revolt dating of the text deposits from the start. When the earlier, 1st-century BCE habitation period at Qumran was discovered in the second excavation season, the start of the Essene habitation was simply moved back, like stretching a rubber band back to make it longer, without questioning anything else. Basically, the dating of the deposits in the caves started as a mistake in de Vaux’s early archaeological interpretation, which seemed like a good idea at the time, and which just sort of took on momentum, never got corrected, and here we are.


Q. What are the consequences to scholarship of the Qumran texts if the deposits are earlier than thought?

A. This again is an interesting point because the consequences are less than they might seem. Scholars already have all of the Qumran texts written not later than 1st BCE, so nothing is changed in terms of when the texts were composed. In fact, such a correction of archaeological interpretation has no major consequences that I can see apart from the correction itself. It is simply moving the date of the deposits earlier to be in agreement with where scholars already have the end of composition of the texts. Let me correct that—I just said nothing of substance changes, but maybe that is not correct. With First Revolt deposits, the texts look like the work of a sect or something outside the mainstream since they seem "different" than the mainstream. But if all of the text deposits are from 1st century BCE, the possibility is opened up that the texts could represent an earlier mainstream. Of course earlier text deposits could still also be marginal or sectarian, but there is a greater range of possibility open. That is the main possible larger consequence that I can see.


Q. What about the "Herodian" lamps found in Cave 1?

A. The "Herodian" lamps found in some of the caves do not prove 1st century CE or Period II scroll jars. De Vaux found three "Herodian" lamps in a Period Ib context in locus 114, according to his Schweich lectures volume. Also, I noticed in Bar-Nathan’s Jericho volume that what de Vaux called the "hellenistic" lamps—which are exclusively Period Ib at Qumran and in Cave 1—were found at Jericho with "Herodian" lamps, contemporary in the same location, in a context dated by the excavators 15-6 BCE. If these lamps were found together at Jericho, it cannot be assumed the "Herodian" lamps in Cave 1 are necessarily later than the "hellenistic" lamps in Cave 1 (since the two types of lamps appear together at Jericho). And the "hellenistic" lamps in Cave 1 are definitely Period Ib. The "Herodian" lamps in Cave 1 could be from later, but it isn’t certain. Of course the big question with "Herodian" lamps as well as the other pottery found in the caves where scrolls were found is association: what can be known to be associated with the scrolls, as distinguished from possible later intrusions.


Q. Were the "scroll jars" ever used for holding human bones?

A. No report of anything like this have I ever heard. There was a "scroll jar" reported found in a cemetery way up in Quailibah, or Abila, in Transjordan far to the north of the Dead Sea, supposedly in a 2nd century CE context. The jar was never drawn or illustrated, and its present whereabouts is unknown. But nothing about human bones in it.


Q. Were all the scrolls put in the caves at the same time? Why were the scrolls put in the caves?

A. This question goes to the heart of one of the most important, and unfortunately unresolved, issues in understanding this whole thing. There seem to be at least three basic possibilities for the text deposits in the caves. They could be permanent disposals with no intent to recover them, stored and hidden away carefully because they are sacred texts so no one would find them and defile them. Or a second possibility could be that they could be depositories in active use that were interrupted and abandoned at some point. Or third, they could be hidden away in the caves in an emergency or time of crisis and were never recovered because the people who did this were killed or deported, and unable to get back to recover their scrolls. And it is possible that different explanations among these three could apply to different caves. It is interesting that the "outlying" caves seem to have the scrolls in jars, whereas the texts in the "inlying" caves adjacent to the site—which is most of the Qumran texts—seem not to have been stored in jars, but rather put in those caves loosely. Yet the texts are the same in kind in all of the caves. There seems to be no basis for a distinction of different kinds of texts, or systematically different dates of texts, in particular caves. It is the same kind of texts of apparently the same (range of) dates in the caves, except for the inlying/outlying jar/non-jar association distinction—at least it seems to be something like this. How is this to be interpreted? I don’t know for sure, and I don’t know that most Qumran scholars claim certainty in answer to these questions either. Interesting questions, but without secure knowledge, it is difficult to see how these issues can be argued to support any dating of the text deposits over some other dating.


Q. Were "scroll jars" found in caves other than with scrolls?

A. Yes, de Vaux and Harding found large numbers of "scroll jars" in many caves which had nothing to do with any sign of scroll activity. This was the main reason de Vaux and Harding changed from their earlier view that the jars’ primary purpose was for holding scrolls, to instead, according to de Vaux, being used primarily for holding food or supplies.


Q. Were "scroll jars" found in locus 86 (where the Period Ib bowl was found with what Cross said was "first century A.D." writing)?

A. No, no "scroll jars" were found in that locus in de Vaux’s reports.


Q. Has Cross’s palaeographical dating of the locus 86 bowl ever been challenged?

A. No—the dating comes from Cross, and Cross still sticks to it, in his 1995 third edition of The Ancient Library of Qumran. No one has challenged Cross’ palaeographical dating on this particular item to my knowledge. Yet the bowl is from what is unanimously agreed to be a Period Ib context at Qumran, the writing was done before the bowl was fired, and the bowls are exclusively 1st century BCE at Jericho, according to Bar-Nathan. So the archaeological dating of the locus 86 bowl is pretty airtight. But Cross has not budged. In 1995, he wrote that he saw no reason to alter his palaeographical dates in the Hasmonean or Herodian periods from what he published in 1961. Cross still insists the writing on that Period Ib bowl from Qumran locus 86 is "first century A.D." But the writing on that locus 86 bowl is only 1st century CE in his system, not in reality.


Q. What is your source on saying that Masada texts are later (palaeographically) than the latest Qumran texts?

A. A very alert question. Unlike the first two points—the dates of internal allusions and non-Masoretic versus Masoretic biblical texts—both of those observations are well recognized (only the implications have been missed)—this third point about Masada Shir Shabbat being later palaeographically than the latest Qumran text deposits has not yet been noticed by Qumran scholars on the level of observation. I found it, I have the data on it, and I will be publishing on this. So on this third point, unlike the first two, I don’t have a secondary source to cite—the footnote on this one is "Doudna, forthcoming." The reason this palaeography has not been noticed is because of the First Revolt deposit date’s filtering effect on perception—it is there, but it has not been "seen."


Q. Again, what does Bar-Nathan say about scroll jars in First Revolt contexts at Masada?

A. I wrote Bar-Nathan and asked about this point concerning scroll jars at Masada. Either because of a misprint or a possible language difficulty, I’m not sure which, anyway, Bar-Nathan was very courteous and wrote me back with an answer, but unfortunately it did not answer my question. The "scroll jars" are Bar-Nathan’s Type SJ2B. I wrote Bar-Nathan an email and asked if she knew of any SJ2B "scroll jars" at Masada in First Revolt contexts. (I noted I could find no reference to any SJ2B jars at Masada in her 2002 Jericho volume.) Bar-Nathan replied that she had "SJ2" jars at Masada which seemed to be in First Revolt contexts, from the Yadin excavations. But SJ2 is a whole larger category or class of jars which includes many non-Qumran scroll jar types. So that did not answer the question. Only SJ2B are Qumran-type "scroll jars," and that was the question. I wrote Bar-Nathan back and asked her to clarify if there were any SJ2B—2B—jars, specifically, found at Masada in a First Revolt context. But I never got an answer back from Bar-Nathan to my followup question. So that is all I could find out about that. There is no verification of these alleged Masada First Revolt scroll jars. Of course, it is always possible that real evidence could be shown at some point that scroll jars were used in Period II at Qumran or that scroll deposits occurred in Period II. I am saying there is no evidence for this in terms of what is presently known, and it sure looks like it has been a mistake all this time.



(1) "Hellenistic" lamps: 1st century CE to Period Ib

After the first excavation of Qumran in 1951, de Vaux dated two "hellenistic" lamps which had been found in Cave 1 to the 1st century CE (de Vaux 1955 [DJD I]: 11). Later, de Vaux corrected the dating of the "hellenistic" lamps from both Cave 1 and the site to Period Ib exclusively (R. de Vaux, Archaeology and the Ded Sea Scrolls. The Schweich Lectures of the British Academy 1959 [Oxford: Oxford University Press, 1973], 18, 49-50). The Qumran Period Ib dating of these lamps is corroborated by the Jericho finds (Bar-Nathan, Hasmonean and Herodian Palaces at Jericho, 111). In 1998, Robert Donceel demonstrated that the two "hellenistic" lamps found in Cave 1 were made by the same workshop which produced two "hellenistic" lamps found in debris from Period Ib in Trench A outside the north wall of Qumran, and that clay testing suggested that this workshop was located at Qumran itself. Donceel commented: "Si ces observations renforcent l’hypothèse, dictée d’ailleurs par la simple vraisemblance, de relations entre le site et les grottes toutes proches ; elles attesteent néanmoins qu’elles étaient réelles à une époque bien antérieure à celle des événements qui conduisirent à la prise du site par les Romains puis à celle de Jérusalem et à la ruine du temple" ("If these observations reinforce the hypothesis, dictated otherwise by simple likelihood, of a relationship between the site and the nearby caves, they nevertheless attest that this relationship was at a time well before the events which resulted in the capture of the site by the Romans and then Jerusalem and the destruction of the temple") (R. Donceel, "Poursuite des travaux de publication du matériel archéologique de Khirbet Qumrân. Les lampes en terre-cuite," in Z.J. Kapera [ed.], Mogilany 1995. Papers on the Dead Sea Scrolls offered in memory of Aleksy Klawek [Qumranica Mogilanensia 15; Cracow: Enigma Press, 1998], 87-104 at 104).

(2) Animal bone deposits—Ib/II to Ib

De Vaux wrote: "The clearest proof of all [that the occupiers of Period II belonged to the same group which had left Qumran in Period Ib] is, perhaps, the evidence that so special a rite as the burying of the [animal] bones was observed at both periods" (Archaeology, 120). But after examining de Vaux’s unpublished materials in preparation for publication of de Vaux’s excavations, Robert Donceel reported in 1998: "nous … sommes arrivés à la conclusion qu’aucune de ces dépositions n’y est postérieure à la phase Ib du R.P. de Vaux" ("we have arrived at the conclusion that none of these [animal bone] deposits are later than de Vaux’s Period Ib") (Donceel 1998: 99).

(3) The most common bowl at Qumran—Ib/II to Ib

Bar-Nathan notes that there are no 1st-century CE attestations at Jericho of Jericho type BL5, the most common type of bowl found at Qumran (708 of this bowl were found at locus 86 alone at Qumran) and suggests that de Vaux may have erred in attributing these bowls to both Period Ib and Period II. Bar-Nathan: "In view of the absence of this bowl [J-BL5] from first-century C.E. contexts at Jericho, the dating of the material from Qumran Period II might have to be revised" (Hasmonean and Herodian Palaces at Jericho, 89).

(4) Scroll deposits in caves—Ib/II to Ib …


For further on the subject of this paper, and references, see:

Greg Doudna, "The Legacy of an Error in Archaeological Interpretation: the Dating of the Qumran Cave Scroll Deposits", in K. Galor and J. Zangenburg (eds), Qumran. The Site of the Dead Sea Scrolls. Archaeological Interpretation and Debate. (Volume of papers of Brown University conference, November 2002.) Leiden: Brill. Forthcoming.

Greg Doudna, 4Q Pesher Nahum: A Critical Edition. Sheffield: Sheffield Academic Press, 2001. Appendices A and B (pp. 675-754).

Greg Doudna, "Dating the Scrolls on the Basis of Radiocarbon Analysis", in P. Flint and J. VanderKam (eds), The Dead Sea Scrolls after Fifty Years. Vol. I; Leiden: Brill, 1998. Pp. 430-471.

| Page 1 | Page 2 | Page 3 |