Skip to: Site Menu | Main content

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

(page 2)
Scroll jars at Qumran

If Qumran-type scroll jars are not known as late as the 1st century CE at Jericho, what about at Qumran? During the first two seasons of excavation at Qumran, de Vaux claimed that all scroll jars at Qumran were from the 1st century CE, that is, Period II, and none earlier. In his Schweich lectures of 1959, however, de Vaux corrected that—though without explaining this correction or what items he had in mind—saying that scroll jars were in use at Qumran in both Period Ib and Period II.

Today no one disputes that scroll jars were used at Qumran in Period Ib. (Magness: "cylindrical [scroll] jars are represented [at Qumran] in the post-31 B.C.E. phase of Period Ib" [Archaeology of Qumran, 80].) Was de Vaux correct in claiming scroll jars were used also in Period II?

I made the following table from de Vaux’s reports, listing all known scroll jars from the buildings at Qumran and the published information available for these jars. These tables are being published in a forthcoming volume of papers from the Brown University conference on Qumran archaeology held in 2002, edited by Katharina Galor and Jurgen Zangenberg. The column to the far right gives dating conclusions based on de Vaux’s published information. This last column is from me, based on the known information for each jar.


Table 1

Scroll jars found at the buildings of Qumran

(HC = Humbert and Chambon 1994; F=de Vaux’s "Fouilles" preliminary reports in Revue Biblique)

Locus Inventory number

Type of scroll jar

Photo Notes and comments Dating
2 27 Classic tall cylindrical HC #142 F1953, Fig. 2:2. HC 292. Buried in a Ib floor that was reused in II. In 1952, de Vaux announced that both the jar and the floor were 1st century CE before realizing that there was a Period Ib and that the locus 2 floor was built in Ib. Uncertain
13 768 Classic tall cylindrical (+ 2 others) HC #104-107 F1954, Fig. 5:4. HC 297. The authors of HC (p. 48) draw these jars in locus 13 in both Ib and II, indicating they understand these jars were or could have been installed in Ib. Uncertain
17 939 "Squat" type F1954, Plate 11b at right. F1954, Fig. 5:7. HC 299. De Vaux assigned this jar to "Niveau II" (Period II), but no stratification information is given. Since this period assignment has no known justification and is among other period assignments of F1954, Fig. 5 is known to be based on flawed and later-corrected assumptions (see comments at locus 44 and 45c); this Period II assignment is similarly untrustworthy. Uncertain
34 621 "Squat" type HC #68-70; F1954, Plate 12a HC 303-304. Identified as Period Ib by de Vaux, who appears to say it was installed in one of several Ib basins ("dans l’un d’eux"; F1954: 208). Magness at one point argues this jar is Period II (Archaeology of Qumran, 123), but elsewhere in the same work agrees with de Vaux that this jar was Period Ib (pp. 52, 200). Ib
44 917


"Squat" type F1954, Plate 11b at left. F1954, Fig. 5:9. Southeastern pottery annex. Considered "Niveau II" by de Vaux before realizing the southeastern pottery annex was in use in Ib. The mistake concerning the southeastern pottery annex was corrected by de Vaux (F1956: 541). There is no evidence this jar is Period II and the surrounding loci jars weigh in favor of a Ib dating for this jar. Uncertain
45a 799 or 800 Classic tall cylindrical None HC 307. This jar is probably Ib because HC #356 shows it with what looks like a Jericho type 2A jar (= F1954, Fig. 5:8) which at Jericho is attested exclusively 1st century BCE (Bar-Nathan, Hasmonean and Herodian Palaces at Jericho, 27, 150). Probably Ib
45c 908 "Squat" type HC #356 F1954, Fig. 5:3. HC 307-308. Southeastern pottery annex. As with the other F1954, Fig. 5 jars, de Vaux claimed this one was "Niveau II" at a time when he thought all of the southeastern pottery annex was only Period II. When de Vaux corrected the dating of the southeastern pottery annex to both Ib and II (F1956: 541), the basis for this jar’s attribution to II was removed. This jar’s dating is therefore uncertain, but in light of the locus 45a analysis Ib seems likely. Probably Ib
61 1474 Classic tall cylindrical HC #353 HC 312. Not enough information to know dating. Uncertain


The second table (below) lists additional possible examples of scroll jars. In these cases, I could not tell from de Vaux’s information whether these jars are scroll jars or not, but it is likely that at least some of these are.


Table 2

Additional possible cases of scroll jars at the buildings of Qumran (but identifications as scroll jars are uncertain)

Locus Inventory number Type of jar Photo Notes and comments Dating
1 2 "jarre cylindrique" None HC 291. No information. Uncertain
8 192 "jarre cylindrique" None HC 294-95. Uncertain
43 885 "jarre cylindrique" None HC 306. No information. Uncertain
80 1465 "une petite jarre" (cylindrical) HC #359 HC 317. Southeastern pottery annex. "Sous ce sol," the lower level, which is Ib. HC have this jar as Ib in their drawings (p. 168). Ib
81 Unknown Small cylindrical jar HC #316-318 HC 317. De Vaux describes this jar as found at the lowest of 3 levels: "inférieur … peut-être la période Ib?" Ib (?)
101 Unknown Either a "squat" type scroll jar or Jericho 2A ovoid-type HC #286 HC 323. HC photo #286 is identified as a Ib photo in the caption. Ib (?)
114 2504 "jarre cylindrique" None HC 327-28. Two "jarres à large ouverture." All of the locus 114 finds are identified as Ib by de Vaux at Archaeology, 5 n. 1, and HC 327-28. Ib
120 Unknown "une jarre cylindrique" None HC 330. No photo or drawing of jar available; uncertain dating. Uncertain


Several of these scroll jars are dateable to Period Ib. Others are uncertain due to a lack of adequate information. Not one of these scroll jars is dateable with confidence, on the basis of present information, to Period II. De Vaux himself identified some of these jars as Period II in his early publications from the first two excavation seasons. But in each case, any basis for confidence in those Period II identifications of de Vaux is removed when later corrections of de Vaux are considered, as noted in the table.

For example, de Vaux found a buried "scroll jar" in locus 45, in the southeastern pottery annex, in the second excavation season at Qumran in 1953. De Vaux published this jar in 1954 identified as from Period II. But de Vaux mistakenly assumed at that point that the southeastern pottery annex was in operation only in Period II. Later de Vaux came to realize the southeastern pottery annex was in operation in both Periods Ib and II, and he published a correction on that point in 1956. The 1956 correction means the scroll jar from locus 45—or any of the other jars from the southeastern pottery annex published in 1954—cannot be associated with Period II with confidence.

Similarly, de Vaux found a "scroll jar" in locus 17 in the second (1953) excavation season, which de Vaux published in 1954 identified as a Period II jar. But this was at a time when de Vaux thought all cylindrical scroll jars were Period II, in principle. Since locus 17 is close to locus 2, the date of the buried jar of locus 17 may be the same as that of the buried jar of locus 2 (due to the similarity in jar burials and proximity of the loci). Was the perception that all scroll jars must be Period II the reason de Vaux identified the locus 17 scroll jar as Period II? There is no other known reason. Nothing in de Vaux’s notes in Humbert and Chambon gives a stratification reason to know that the locus 17 jar is Period II, and de Vaux otherwise was making period assignments subjectively, not from stratification. De Vaux’s later correction acknowledging that scroll jars indeed existed at Qumran in Period Ib—but without ever identifying which jars, presumably among his previously published ones, he meant—means the scroll jar of locus 17 cannot be associated with Period II with confidence.

And so on, case by case, right down the list. Not one of these scroll jars at Qumran is confirmed as manufactured or installed in Period II on the basis of currently available information.


Bowl-lids on scroll jars

Another potential method for dating the scroll jars is by dating the bowl-lids found associated with these jars. The most common type of bowl lid of the Qumran scroll jars appears in Bar-Nathan’s volume as Jericho type LD ("the bowl-lid J-LD found at Jericho is identical to the one most common at Qumran," Bar-Nathan, Hasmonean, and Herodian Palaces at Jericho, 27). I collected all instances of this type of bowl-lid in Bar-Nathan’s 2002 volume, listed in this table.


Table 3

J-LD bowl-lids at Jericho of the kind used with scroll jars in the caves at Qumran

Bar-Nathan ref. (2002) Jericho location Inventory number Date (from excavators)
p. 229

Room AE16, Western Mansion, Twin Palaces.

#246 85/75-31 BCE
p. 239

Storeroom F165, Building FB2, Industrial Area.

#247 31-15 BCE
p. 240

Pool F182, Ritual Bath (together with Pool F176), Building FB1, Industrial Area.

Uncatalogued 31-15 BCE
p. 238

Locus F123, East of Winepress F113-F188, Industrial Area.

Uncatalogued 15 BCE-6 CE


The dates in the right column of this table are from the Jericho excavators. As can be seen in this right column, all of these bowl-lids are either contemporary or close to contemporary with Qumran Period Ib in the 1st century BCE. Not one of these bowl-lids is known to be as late as Qumran’s Period II. In light of this data, what is the archaeological basis for claiming that scroll jars in the caves, or the scroll deposits, are as late as Period II?


Follow the logic here

Furthermore, if these jars or jar-lids were to be shown at some point to have been used at Qumran or in the caves in Qumran’s Period II, that would not establish that scroll deposits occurred in Period II. De Vaux thought this way at first—as some Qumran specialists do today—because he thought these jars were used only for holding scrolls. But de Vaux later realized that the main uses of the "scroll jars" were unrelated to holding scrolls. At first, de Vaux thought the large, cylindrical jars in the caves—the so-called "scroll jars"—had all held scrolls. Later he corrected this to saying that these jars in the caves were instead used for storage of food or supplies.

Harding (1955): "our first idea, that these [jars of Cave 1] must all have contained manuscripts, is no longer tenable"

De Vaux (1962): "ces jarres … ont servi normalement, à garder des provisions" ("the jars normally served to hold supplies")

Magness today also argues that the primary use of the so-called Qumran "scroll jars" was "hoard[ing] stores of pure food and drink in the caves, and that "it is not clear whether scrolls were similarly hoarded in the caves" (Magness, Archaeology of Qumran, 86).

If the primary purpose of these jars is unrelated to scroll deposits—according to both de Vaux and Magness—then the mere use of these jars at Qumran in the 1st century CE, in Period II—if that were to be established—would not confirm a dating of scroll deposits in Period II. The jars might have been used instead, say, for their primary purpose, and not had scrolls put in them at all in Period II. Unless it is known scroll deposits were associated with these jars every time they were used at Qumran, it is not known.

But there is little point in arguing over the interpretation of what hypothetically might be established in the future. The point here is as follows: nothing now known or published has established on archaeological grounds that Qumran scroll jars, let alone scroll deposits, postdate Qumran’s Period Ib.

| Page 1 | Page 3 | Page 4 |