Carbon C tests on samples of the Dead Sea Scrolls have substantially confirmed the previous date of the scrolls based on paleography the shape of the letters , according to two recent reports.
As one of the reports recognizes, the C14 tests were undertaken as a result of prodding by outside scholars, principally Robert H. Eisenman of California State University, Long Beach, who question the paleographic dating of the scrolls.
In C tests were performed on the linen in which one of the scrolls had been wrapped, but no tests were done on the scrolls themselves. Since then, C testing has become far more refined and only a minuscule amount of material is needed from a scroll or a fragment to perform the tests.
Fourteen samples were taken from different texts—eight from caves in the Wadi Qumran and six from nearby sites often included in scholarly understanding of the term Dead Sea Scrolls. Each text sample was subdivided into at least three subsamples, which were then exposed to different ultrasonic and chemical cleaning procedures. Ten of the 14 documents had been dated paleographically, but these dates were not made known to the scientists performing the C procedures.
The paleographic dates varied from the mid-second century B. Four of the scrolls tested—from the non-Qumran sites bore dates within the document, ranging from the mid-fourth century B. This information was also withheld from the scientists performing the tests. The results of the C tests are expressed in terms of a probability that the date of the document falls within a particular range. In the table above, the range is indicated by a black bar and sometimes two black bars.
The probability that the document falls within the black bar or bars is 68 percent one sigma. The five cases in which there are two black bars results from variations in what is called the calibration curve; if this curve is relatively flat, it produces one black bar that is, one solid range of dates.
If the calibration curve has peaks and curves, this will result in two bars with two different ranges. The probability that the date falls within one of these two ranges is still 68 percent. The relative likelihood as between the two ranges is indicated by the varying widths of the two black bars. If one of the two is very narrow, it may be for all practical purposes ignored. The hatched bars represent the date range as estimated by paleography. A vertical line indicates the precise date of documents that bear internal dates.
For eight out of ten documents dated by paleography, that date and the C date fall within the same range. In document 7 the difference is about 50 years. In document 2, however, the Testament of Kohath, the C date is approximately years earlier than the paleographic date. We will return to this aberrant case.
Of the documents with internal dates, the C date range included the internal date in three of the four cases. In the fourth, the C date missed by about 10 years. In general the C tests appear substantially to validate the paleographic dating. Professor Eisenman dates a number of Qumran documents later than the paleographers do, consistent with his theory that they reflect conditions during the last decades before the Roman destruction of Jerusalem in 70 C.
Eisenman objects that neither he nor any other outsider was included in the group managing and monitoring the tests. The only tested document that he says is critical to his theory is the Testament of Kohath, which he dates to the mid-first century C.
The paleographers date it to about B. Eisenman emphasizes, on this basis, the unreliability of the other C dates produced in these tests. The Swiss lab recognizes a problem with the C test on the Testament of Kohath.
They point out that C dating from the Kohath sample cleaned ultrasonically differed by about years from the date provided by the sample cleaned both ultrasonically and chemically. In no other case did they get this difference between samples subjected to differing cleaning procedures.
So something must be wrong, the lab concludes, with the Kohath test. Eisenman insists that new tests be performed on this and other documents that he dates later than the paleographers and that he be a member of the managing and monitoring team. It is unlikely, however, that additional C tests will be conducted in the foreseeable future.
I thank George Bonani of Zurich for freely providing BAR with a prepublication copy of this article and explaining some of the testing procedures.