After the Israeli newspaper Ha’aretz got wind that BAR would be publishing the foregoing article, it published its own story in October 2007 about the Jezebel seal and Professor Korpel’s attribution to Queen Jezebel. Korpel’s article, the newspaper wrote, is “scheduled to appear in the highly-respected Biblical Archaeology Review.”
Christopher A. Rollston, a professor at Emmanuel School of Religion and a budding paleographer, quickly filed a response on the Internet to Korpel’s piece.1 Rollston could find Korpel’s argument in a scholarly paper she had published in the Journal for Semitics.2 Rollston concluded that the seal could not be that of Queen Jezebel. Korpel, in turn, will be publishing a scholarly article in the journal Ugarit-Forschungen that addresses Rollston’s arguments, most of which are technical and easily answered— except one. And that one relates to paleography, Rollston’s specialty.
Rollston studied the four letters on the seal and concluded that they cannot date to the ninth century B.C.E., when Queen Jezebel lived. They must date later. Therefore, even if the name on the seal is Jezebel, it cannot refer to Queen Jezebel, wife of King Ahab of Israel. Here are Rollston’s own words:
“I would not consider it tenable to argue that the script of this seal could be ninth century Old Hebrew. It must be later” (emphasis added).3
If true, Rollston’s argument was devastating to Korpel’s identification of the seal: Her argument in favor of the identification is not even “tenable.”
Rollston is nothing if not certain. There is apparently no room in his scholarship for doubt or hesitation. He operates, he says, purely as a scientist. I had seen this certainty before, when I listened to Rollston’s courtroom testimony in the famous ongoing Jerusalem forgery trial. As a witness for the prosecution, Rollston presented himself as an expert in scripts of the Iron Age— the eighth, seventh and sixth centuries B.C.E. He testified that he could identify each “with certainty.”4
In light of Rollston’s damning criticism of Korpel’s not-yet-published article in BAR, I was faced with the question of whether or not we should proceed with publishing it.
I am not a paleographer. I have no idea about the dating of ancient scripts. I could make no judgment of my own as to who was right— Korpel or Rollston. But I did know paleographers on whom I could call.
So I telephoned Israel’s leading paleographer, the universally admired Hebrew University professor Joseph Naveh. I asked Yossi, as he is commonly known, if he could tell on the basis of the paleography of the four letters that have survived whether the “Jezebel” seal dated to the ninth century B.C.E. or to some time later. He replied that he could not. “I cannot tell,” he said, after looking at the seal in the standard catalog. Inscriptions on seals are especially difficult, he told me. “That is why I have never published a seal— with the exception of an Edomite seal from Hatzevah.a
Naveh is not sure. But Rollston is.
Apparently the late great Israeli paleographer Nahman Avigad, who originally published the seal, also felt that it could be from the ninth century, the time of Queen Jezebel. Indeed, he said so. But, according to Rollston, he was clearly wrong: It’s not even “tenable.”
I also spoke with a number of other paleographers.5 Not a single one said that the letters on this seal must be post-ninth century. The consensus was that the four letters on the seal either were, or certainly could be, from the ninth century, although perhaps, according to some, they could also be somewhat later. But that is as far as they would go. As Naveh told me, “Paleography is not a precise science. It is not even a science.”6 (I don’t want to be unfair to Rollston, so if there is anyone out there who knows of a paleographer who shares his view that the letters on this seal must be, from a paleographical viewpoint, post-ninth century B.C.E., please let me know and I will contact him or her.)
Unfortunately, this is not the only instance in which Rollston’s “expertise” appears to be “contaminating the data bank.” Even more important than his paleographic conclusion regarding the date of the Jezebel inscription is his view of the paleography of the famous Gezer calendar, which was discovered in an unstratified context at Tel Gezer in 1908. Its name reflects the fact that it describes various agricultural activities over a 12-month period, beginning in autumn. It may be an apprentice scribe’s exercise tablet. It is generally considered to date from the tenth century B.C.E.7 and, in the words of Dennis Pardee of the University of Chicago’s Oriental Institute, “is a showpiece of early Hebrew inscriptions,”8 perhaps the earliest Hebrew inscription.
This characterization has recently been questioned, however. Some wonder if it is really Hebrew— or if it is Phoenician, South Canaanite or Philistian, all of which have been proposed. The question can be addressed on the basis of linguistic analysis or on the basis of a paleographic analysis (the shape and form of the letters). In a forthcoming article, Pardee himself opts for Phoenician on the basis of a linguistic analysis (however, he disclaims any expertise in paleography9. In his carefully detailed and qualified conclusion, he writes that “in our present state of knowledge, the combination of morphological and syntactic features requires that the identification of the language of the Gezer text as Phoenician is to be preferred … [But] truth be told, the identification as Canaanite cannot be ruled out.”
The question of the language of the Gezer calendar can be addressed also on paleographic grounds: Was the script composed in Phoenician letters or something else? This too, as Pardee notes, is a much-debated question.10 In his response to Korpel, Rollston flatly states his view that the Gezer calendar is written in Phoenician letters rather than Old Hebrew letters: “I believe that it [the Gezer calendar] is written in the Phoenician script.”11 For Rollston, there is no room for debate. At the Jerusalem forgery trial Rollston also testified that the Gezer calendar was written in Phoenician letters. “We have a distinctive Hebrew script; we have a distinctive Phoenician script,” he told the court. Features of one are “not mixed,” he said, with features of the other. “No paleographer would confuse an Old Hebrew inscription with Phoenician script,” he claimed.
On cross-examination, Rollston admitted that there was some discussion among scholars as to whether the Gezer calendar was closer to Hebrew paleography or Phoenician paleography. “Why, if the matter is so clear, was there this discussion?” he was asked. He stated that “very few people in the world” specialized in the paleography of this period; clearly, he was one of them. He dismissed the people (apparently including leading paleographers) who were not up on the latest scholarship. That other scholars might disagree with him did not concern him. As he testified in connection with one of the other inscriptions involved in the trial, “I base my conclusions on evidence, not authority.”
One of the authorities with whom Rollston disagrees regarding the Gezer calendar is the eminent paleographer and Johns Hopkins University professor Kyle McCarter. Indeed, McCarter was Rollston’s teacher and dissertation advisor. (Rollston received his Ph.D. in 1999.) McCarter recently published his analysis of the script of the Gezer calendar. In his view, the script in which the Gezer calendar is written, like the Tel Zayit abecedary (which is the main subject of the article), is “representative of the linear alphabetic script of central and southern Canaan [where Gezer lies] at the beginning of the first millennium B.C.E.” This script belongs “to the South Canaanite script tradition.” The Gezer calendar script, like the Tel Zayit abecedary, did develop from the Phoenician tradition, but it would be wrong to call it simply Phoenician. The Gezer calendar (and the Tel Zayit abecedary) represent “an inland development of the mature Phoenician tradition of the early Iron Age, but in the tenth century [B.C.E.] it already exhibits characteristics that anticipate the distinctive features of the mature Hebrew national script.” Therefore these inscriptions must be considered to be in “the South Canaanite script tradition.” When these inscriptions were written, McCarter tells us, the Phoenician parent script continued to be used on the Phoenician coast, but not inland at a place like Gezer.12
This is a careful, nuanced and incisive paleographical analysis by a mature scholar of the script of the Gezer calendar. It provides a telling contrast to the absolute, confident, assertive, unqualified conclusion of a scholar (1) who has never published anything on the Gezer calendar (except the bald, unjustifiably certain assertion that its script is Phoenician in his response to Korpel), (2) who wrote that the Jezebel inscription cannot be ninth century B.C.E., and (3) who also wrote that “No paleographer would confuse an Old Hebrew inscription with Phoenician script.”
Either Professor Rollston is the world’s greatest and most expert paleographer— or his dating of the Jezebel seal is wrong.— H.S.
1 Christopher A. Rollston, “Precarious Scholarship: Problems with Proposing that the Seal of Yzbl Was Queen Jezebel’s,” Web site of the American Schools of Oriental Research, October 12, 2007, http://www.asor.org/seal.article.html.
2 Vol. 15, no. 2 (2006), pp. 349–371.
3 In subsequent postings, Rollston retains his position: “The palaeographic issues in my ASOR post must remain.” (Chris Rollston, “Rollston’s Response to Korpel’s Iconographic Discussion,” Dr. Jim West’s Blog, October 13, 2007, http://drjimwest.wordpress.com.)
4 See Hershel Shanks, “Jerusalem Forgery Conference” (Washington, DC: Biblical Archaeology Society, 2007), available online at http://bib-arch.org/forgery/forgeryreport.html.
a See Itzhaq Beit-Arieh, “Edomites Advance into Judah,” BAR, November/December 1996.
5 Stephen Pfann, K. Lawson Younger, Anson Rainey, p.Kyle McCarter and André Lemaire.
6 Robert Deutsch in a Web posting specifically disagrees with Rollston’s dating. (Robert Deutsch, comment on “Rollston’s Response to Korpel’s Iconographic Discussion,” Dr. Jim West’s Blog, comment posted October 13, 2007, http://drjimwest.wordpress.com.)
7 This dating has recently been confirmed by the so-called Tel Zayit abecedary, discussed hereinafter. See Ron E. Tappy, p.Kyle McCarter, Marilyn J. Lundberg and Bruce Zuckerman, “An Abecedary of the Mid-Tenth Century B.C.E. from the Judaean Shephelah,” Bulletin of the American Schools of Oriental Research (BASOR), No. 344 (2006), pp. 5–46. The abecedary can be dated stratigraphically no later than the tenth century and possibly earlier. The Gezer calendar cannot be dated stratigraphically, but it is the same script family as the abecedary. As McCarter notes, “This situation underscores the importance of the recovery of the Tel Zayit stone from a secure and well-understood context.”
8 Oxford Encyclopedia of Archaeology in the Near East, vol. 2 (1996), p.400.
9 Pardee writes, “I have no pretension of being an expert in the palaeography of first-millennium inscriptions.”
10 Pardee cites the relevant literature. None of these citations is a contribution by Rollston.
11 In support of this statement, Rollston cites a 1987 article by Joseph Naveh. More carefully, Pardee cites numerous articles by Naveh and refers to “Naveh’s apparently evolving views” on this issue. I spoke to Naveh about his views on this: What did he think about the Gezer calendar script in light of Kyle McCarter’s characterization of it as South Canaanite? Naveh said “McCarter is probably right. The differences [among Phoenician, Hebrew and South Canaanite] are very hard to tell. We have very few inscriptions from this period and the differences are very small.”
12 Harvard’s Frank Cross, probably the most distinguished paleographer in the United States, comes to a conclusion similar to McCarter’s. Cross finds that the script of the Gezer calendar “already exhibits characteristics that anticipate the distinctive features of the mature Hebrew national script.” (Frank M. Cross, “Newly Found Inscriptions in Old Canaanite and Early Phoenician Scripts,” BASOR, No. 238 (1980), p.1, at pp. 27–28).