The Mount Ebal Lead ‘Curse’ Inscription in Late Bronze Age Hebrew: Some Methodological Caveats

26 March 2022

The Mount Ebal Lead ‘Curse’ Inscription in Late Bronze Age Hebrew: Some Methodological Caveats

Christopher Rollston (

George Washington University, Professor of Northwest Semitic Languages and Literatures, Chair of the Department of Classical and Near Eastern Languages and Civilizations.


Setting the Stage:

Some sensational claims were made in a press conference on March 24, 2022 about a small lead ‘inscription’ that is purported to hail from the Late Bronze Age, to be written in the Ancient Hebrew language, to consist of forty letters, to be full of curses (i.e., with the tri-literal root ’rr occurring ten times), and to twice mention Yahweh. And as part of these claims, it was asserted in the press conference that there are a lot of firsts for this inscription (e.g., oldest Hebrew inscription, earliest reference in a Hebrew text to Yahweh, etc.).  ( ).

These are some mighty sensational claims. However, sensational claims require sensational evidence, that is, evidence that is absolutely overwhelming and entirely compelling.  And in this case, I would suggest that some methodological doubt is probably a very useful thing. Of course, this find does hail from Mount Ebal, the famous site which was excavated by Adam Zertal during the 1980s. Therefore, this find is interesting, and it is ostensibly important. But something else that is normally just as important is methodological caution regarding sensational conclusions! After all, dramatic claims have been so very common during recent years, and the end result is almost always the same: the dramatic and sensational claims crumble under the weight of scrutiny, and then more sober conclusions rise to the fore as the most compelling.

Some Further Details

Here are some of the basic facts.  On March 24, 2022 at Lanier Theological Library (in Houston, Texas), Scott Stripling (Provost of The Bible Seminary in Katy, Texas; and the Director of Excavations for the Associates for Biblical Research at Khirbet el-Maqatir and Shiloh, Israel), along with Pieter van der Veen (Johannes Gutenberg-University, Mainz), and Gershon Galil (University of Haifa) held a press conference to announce the discovery and putative decipherment of a 2 cm x 2 cm folded lead inscription (nota bene: the inscription remains folded, that is, it has not been opened). According to Stripling, Galil, and van der Veen, the forty letters on the inside of this folded lead object are not discernible via the naked eye.  However, via imaging that was conducted in Prague at the Academy of Sciences of the Czech Republic, they (i.e., van der Veen and Galil) believe that forty letters can be seen, that these letters can be read, and the words that result can be deciphered.  Here is their translation: “Cursed, cursed, cursed – cursed by the God Yhw [Yahweh], You will die cursed. Cursed you will surely die. Cursed by Yhw – cursed, cursed, cursed.” Furthermore, these scholars contend that the script of this inscription is “Proto-Alphabetic” (it is perhaps useful to mention that a standard means of describing the alphabetic script at this time period would be Early Alphabetic or Proto-Canaanite, rather than “Proto-Alphabetic).  Stripling, Galil, and van der Veen also state that there are some letters on the outside of this folded lead object, but they do not mention which letters or words they might be reading on the outside.

Significantly, this inscription was not found in a stratified context during excavations at “Mount Ebal.” Note that Adam Zertal directed the excavations at Mount Ebal in the 1980s, and he believed that he had found at this site a “structure” which he (Zertal) believed was probably an altar and could be connected in some fashion with the altar mentioned in Joshua 8:30-31 (on this, see now an update below, as an addendum to this blog post of mine). Rather this inscribed lead object was found in 2019, as part of a process of wet sifting and dry sifting some of the dirt that had been removed as part of the 1980s excavations. Perhaps also useful to mention: the press conference at Lanier Theological Library references some carbon remains that were found during the sifting, but there is no reference, alas, to any carbon dates (e.g., AMS, etc.). Also important to mention is the fact that Stripling, Galil, and van der Veen have not yet finished writing the scholarly article about this find.  They are hoping to complete it in the coming months and then to submit it for publication somewhere.

Some Biblical and Ancient Near Eastern Background

The Claims of Stripling, van der Veen, Galil…and Some Responses

There are some rather striking claims in the press conference about this lead inscription and about its implications.  First and foremost, I would emphasize that reading and deciphering Early Alphabetic inscriptions is difficult. Thus, is hard for me to believe that all of the readings of Stripling, Galil, and van der Veen will stand the test of time. In fact, I would predict that almost all of the readings posited in the press conference will be vigorously contested, once scholars in the field of epigraphy are allowed to see the images of the Academy of Sciences of the Czech Republic. Furthermore, I am certain that the translations of the readings will also be contested.

And it should also be emphasized that at the press conference *no* images from the Academy of Sciences of the Czeck Republic were shown.  Thus, claims were made, but the real evidence was not shown! Normally, even during a press conference about a new inscription, a good image or two of the inscription is shown.  But in this case, None!  Also of import: it is striking that the only drawing presented at the press conference was a single drawing of a single putative occurrence of the divine name Yhw.  And I would emphasize that this drawing struck me as particularly schematic in nature. As a result of all of these sorts of things, my hermeneutic of suspicion is, therefore, quite heightened.

But it’s worth looking even more at some of the dramatic claims.  Stripling stated that: “One can no longer argue with a straight face that the Biblical text was not written until the Persian Period or the Hellenistic Period, as many higher critics have done when we clearly do have the ability to write the entire text [of the Bible] at a much, much earlier date.” Galil makes the same basic statement: “No one can claim the Bible was written in later periods, the Persian Period or the Hellenistic Period.”  Similarly, Galil stated: “the person who wrote this was a genius, not only a scribe, but a theologian!” Stripling also stated that “our friends from the other side of the academic aisle have disparagingly spoken of us [that is, those] who believe that the Bible was written at an early date as this, because that was not [supposed to be] possible because there was no alphabetic script with which to write it.  Clearly this [inscription] flies in the fact of that.” Galil goes on to state that “the scribe who wrote this important text….believe me…he could write every chapter in the Bible.” Galil also goes on and states that this “is the most important inscription ever found in Israel.”

It’s useful to step back for a moment. First, I would emphasize that with all due respect to Striping, most epigraphers believe that the alphabet was invented by the 18th century BCE. Thus, his statement that scholars have contended that there was no alphabet to write with doesn’t make enormous sense. Note also that Ugaritic is also an alphabetic language (and dates mostly to the 13th century BCE).

Second, I was also struck by the “loose language regarding chronology” in the press conference.  For example, I think that what Stripling, Galil, and van der Veen were intending to say is that this inscription (as they understand it) provided evidence that writing of the Pentateuch (or portions of it), or the Hexateuch (i.e., Genesis through Joshua) could have been written prior to the Persian or Hellenistic Period.  The fascinating thing is that normally they did not use very precise language and seemed to speak about the writing of the Bible as a whole. Thus, someone who is not familiar with the field might assume that this inscription proves that the whole Hebrew Bible was written in the second half of the 2nd millennium BCE (their date for this inscription).  I don’t think that’s what they intended to say (as books such as Ezra, Nehemiah, Esther, 1-2 Chronicles, etc. are obviously all Second Temple, and none of the Latter Prophets or the book of the Twelve could have been written so early, based on the material in these books and the chronological reference points contained therein, etc.). 

Also, it is perhaps worth mentioning that people such as I have argued for some time that there was a fair amount of alphabetic writing in the late 2nd millennium BCE (e.g., Rollston, “Inscriptional Evidence for the Writing of the Earliest Texts of the Bible: Intellectual Infrastructure in Tenth- and Ninth-Century Israel, Judah, and the Southern Levant.” Pp. 15-45 in The Formation of the Pentateuch: Bridging the Academic Cultures of Europe, Israel, and North America, eds. Jan C. Gertz, Bernard Levinson, Dalit Rom-Shiloni, and Konrad Schmid. Tübingen: Mohr Siebeck, 2016). And so I found it striking that these authors did not seem to know about the history of the field in this regard.

Also, and perhaps even more importantly, even if we assume everything that Stripling, Galil, and van der Veen state about the readings and translation is correct (and that’s a big assumption), they have told us that there are 40 letters. They have also said that the word curse or accursed occurs 10 times. There are three letters in that root (’rr). So that’s 30 of the 40!  And the remaining 10 letters are used to write “God,” “die,” and “Yhw.” If we do have those four words or roots: namely, “curse,” “God,” “die” and “Yahweh,” I’m happy to say that somebody back then and there could write, and hopefully somebody else back then and there could read it. But to say that based on those four words or roots that somebody could write the whole Bible….well, that’s a bridge (way) too far for me.  After all, there are 8500+ words in the Hebrew Bible (counting verbs, nouns, adjectives, adverbs, particles, common nouns, proper nouns), and four is a pretty small fraction of the whole, therefore!

Furthermore, Galil also stated: “this is not just a curse. It is actually a legal text, not just a legal warning.” Van der Veen also is of the same mind: “It is a legal verdict about an unknown person or group who are addressed in the inscription.” This too is quite a leap.  After all, we have 40 letters, four basic root words, and all of a sudden we have a legal text! I don’t think that my own esteemed master-teacher of law and diplomacy in the ancient Near East (the late Raymond Westbrook) would find these statements by Galil and van der Veen to be the least bit compelling.  And neither do I.

But it gets even more interesting.  When Stripling is asked by someone in the audience if this inscription coincides or correlates with the Covenant Renewal Ceremony in the book of Joshua, he replied: “I believe the answer is ‘yes.’” And when someone in the audience asks if this inscription impacts the way in which we should conceive of the Exodus from Egypt and the Conquest, he replies that this inscription reveals that “the Exodus from Egypt and the Conquest of the land of Canaan would have occurred at an earlier date” than has usually been supposed. He elaborates further than this inscription “tips the scale in favor of an earlier date.”

It is perhaps useful for me to mention in this connection that the consensus view (among those, such as I, who believe that there was some sort of Exodus, and that there was also some sort of entrance into the land of Canaan for at least some of the Proto-Israelites, and that there were at least some battles as part of that) is that the 13th century BCE is the operative century for the Exodus and Conquest (based on the convergence of a fair amount of evidence, biblical and otherwise…in other words, I fall into the same category as Richard Elliott Friedman in his volume entitled Exodus: How it Happened and Why it Matters).  However, a date in the 15th century has been the darling of a few scholars through the years (especially because of the statement in 1 Kings 6:1 that the First Temple was built 480 years after the Exodus. As for me, I mostly feel that 40 is a stock numeral in the ancient Near East and in the Bible, and since 12 is the number of the tribes of Israel, well, 480 sounds like a schematic numeral that plugs in two very common biblical numerals). In any case, Stripling even concludes on the basis of this little inscription that he finds support for the early date.  That’s quite a leap on the basis of a lead inscription that cannot be dated with enormous precision and contains zero personal names (of historical people) and zero references to any historical event!  Again, this is a bridge too far.

It is also worth mentioning that I am not too certain that the divine name Yhw is in this inscription (although I hope that it is). Moreover, I would emphasize that the script of this inscription (if there actually is a legible inscription) is Early Alphabetic, not Hebrew.  Thus, I’d really be very disinclined to call it the earliest Hebrew inscription (in terms of script or language).  In this connection it is worth noting that the words “God,” “cursed,” “die,” “death,” are all Common Semitic.  That is, they are present in several different ancient Semitic languages (e.g., Ugaritic, Phoenician, Aramaic, Hebrew, Akkadian, etc.); thus, they are not “diagnostic” for any particular Semitic language, since they occur across the Semitic languages. Sometimes I hear someone say, “Oh, this text must be Hebrew, because this is a Hebrew word.” Well, the problem with that is that the word may be Hebrew, but it is not *exclusively* Hebrew.  But rather it occurs in several different Semitic languages (hence the term “Common Semitic”). Of course, someone might say in reply to me, “Yes, those other words in this inscription are Common Semitic, but the word Yhw is the name of the God of Israel.” And I would reply that we can talk more about this when good images of this inscription are available.  But even if these three letters are present (yhw) and are to be read in this order (which may not be the case), that’s not the only word that those letters could be.  In short, I tend to believe that it’s best to try to avoid making precarious statements.  And I’ll stick with that methodology for this inscription as well. 


In sum, I would mostly suggest that we step back and let the dust settle on this one.  It seems to me that Stripling, Galil, and van der Veen have made a fair number of big assumptions. Moreover, I am far from convinced of their readings….especially since they have not even provided so much as a single good image!

And it also seems to me that the best predictor of the future is the past, and in the past, time and time again, sensational claims turn to ash in the crucible of serious, philological and epigraphic analysis. So, let’s wait and see how this turns out.  But as for me, I’m afraid that I’m too methodologically cautious to embrace the sensational assumptions of Stripling, Galil, and van der Veen.

Addendum: Prof. Amihai Mazar has emphasized to me (personal communication) that Zertal’s suggestion that he had found an altar at Mount Ebal is “much debated within the archaeological community.” He has also emphasized that in this region, “no one has seen, before or after this discovery, an altar of approximately 8m x 9m in size.” Prof. Mazar has also noted that “the earlier installation could have been used in a cultic context, but does not resemble an altar.” I’m very grateful for Prof. Ami Mazar’s note and for the data contained in it. Sincerely, Chris Rollston

Deja Vu all over Again: The Antiquities Market, the Shapira Strips, Menahem Mansoor, and Idan Dershowitz

10 March 2021

Déjà vu all over Again:

The Antiquities Market, the Shapira Strips, Menahem Mansoor, and Idan Dershowitz

By Christopher Rollston, George Washington University



Idan Dershowitz has authored an article entitled “The Valediction of Moses: New Evidence on the Shapira Deuteronomy Fragments,” ZAW 133 (2021): 1-22. In his article, Dershowitz states that he offers “new evidence and arguments against the prevailing theory that Wilhelm Moses Shapira forged his infamous Deuteronomy fragments.” He considers the Shapira Strips to be authentic, basically, the first of the real Dead Sea Scrolls. That’s quite a claim.  In fact, he also believes that “The Shapira fragments are not only authentic artifacts, but are unprecedented in their significance: They preserve a pre-canonical antecedent of the Book of Deuteronomy” (p. 22 of his article). He also has a forthcoming monograph dealing with this subject, which will include a full transcription of the Shapira Strips, replete with translation and copious notes.

First and foremost, I would emphasize that Idan Dershowitz is doing the scholarly community a great service in producing a very useful transcription of the Shapira Strips, replete with copious annotations. This will be of much value for those interested in the history of modern inscriptional forgeries.

But I believe that his work will convince very few epigraphers (i.e., scholars who specialize in actual, ancient inscriptions) that the Shapira Fragments are authentic, ancient documents. And I do not believe that his work will convince all that many text scholars (i.e., scholars who primarily work not with actual ancient inscriptions, scrolls, papyri, but rather with edited texts in print editions) that the Shapira Strips are ancient…although I suspect some text scholars will find Idan Dershowitz’s proposal alluring, especially since it seems to “confirm” the things some of them have believed about the textual transmission of Deuteronomy in its earliest forms.

Nevertheless, the totality of the extant empirical evidence continues to demonstrate that the Shapira Strips are modern forgeries, and they reflect the same basic tendencies and problems which are present in most modern fakes and forgeries of the past two, three, or four centuries. I will discuss some of the evidence in this blog post, but in a more detailed fashion in a print publication.

Finally, I should also mention that I was an invited participant in the symposium at Harvard Law which was held in 2019 in which Idan Dershowitz presented his views. As I emphasized in that meeting, the evidence against authenticity is compelling: the Shapira Strips are indeed modern forgeries, modeled mostly on the book of Deuteronomy, with the sort of “forger’s flourishes, augmentations, and additions,” that are hallmark features of forgers’ methods…time and time again through the centuries.

NB: In this blog post, I reference a number of my articles.  Most of these (but not all) are available on  I have included bibliographic data for these in the “for further reading” section at the end of this blog post.  I would have liked to have integrated all of these references into this blog post, but since the New York Times article appeared today, and I wanted to get this post up rapidly, I am just including these as an addendum at the bottom of this blog post.

The Setting in the 19th centuy: 1868-1884

The Shapira Strips surfaced on the Antiquities Market in ca. 1883, and were shopped around far and wide by Moses Wilhelm Shapira (1830-1884) who had an antiquities shop in Jerusalem.  If they had been deemed ancient in 1883, they would have been by Shapira’s own statements) worth an absolute fortune. A decade prior to the surfacing of the Shapira Strips, Shapira had been closely connected with the Moabite Forgeries (ca. 1873, 1874, etc.).  The Moabite Forgeries were inscribed terra cotta and stone objects, with inscriptions modeled rather poorly on the great Mesha Stele Inscription, discovered in ca. 1868 (of which Salim al-Kari, an associate of Shapira, had made a squeeze prior to the famous shattering of the Mesha Stele by heating it and pouring water on it).  Although a number of scholars at the time deemed these Moabite Forgeries to be ancient (e.g., the great Semitist Konstantin Schlottmann), the archaeologist and epigrapher Charles Clermont-Ganneau rapidly debunked them as modern forgeries.  Clermont-Ganneau was absolutely correct.  In fact, the Moabite Forgeries would not fool even a beginning student of Northwest Semitic inscriptions today. 

It is also worth noting that many forgeries were surfacing in this era, in the wake of discoveries such as the Mesha Stele and the Temple Mount Inscription.  Clermont-Ganneau debunked many of them, authoring articles and even a monograph on the subject.  And, of course, Clermont-Ganneau was among the first to debunk the Shapira Strips as well, along with Christian David Ginsberg. Notably, Konstantin Schlottman asserted that the Shapira Strips were forgeries!

Also important to mention is that Shapira was very familiar with aged scrolls (e.g., from the Middle Ages), as he sold a number of them to the British Museum, especially from the region of Yemen.  It should also be emphasized that the Shapira Strips are no longer extant. After Shapira himself committed suicide in the Netherlands (after the Shapira Strips were declared to be modern forgeries), the Shapira Strips were later sold at auction and are often presumed to have later been burnt in a tragic house fire. At the very least, they have never surfaced again.

Further notation: During the past three or four centuries, many hundreds of forged inscriptions have appeared on the antiquities market.  These modern forgeries come in all shapes and sizes and are written in a number of languages, including Greek, Hebrew, Aramaic, Phoenician, Latin, Syriac, and Coptic. Some of these modern forgeries were quite poor, some of them were quite good.  But the production of textual forgeries in the modern period is quite a common thing, and it has often been quite lucrative for the forgers and for those who sell forgeries.  It is a major problem in the field and has long been so.


I. Not the first, not the last.

Idan Dershowitz is not the first scholar to attempt to contend that the Shapira Strips are ancient, and he will not be the last. That is, most famously, Menahem Mansoor (University of Wisconsin, 1959 “The Case of Shapira’s Dead Sea [Deuteronomy] Scrolls of 1883) argued at length that the Shapira Strips were not modern forgeries, but actual ancient Dead Sea Scrolls. But the convergence of epigraphic evidence is squarely against the Shapira Strips. However, I would emphasize that when one thinks about the long history of textual forgeries (going back many centuries), the Shapira Strips are quite good, especially for their time (i.e., late 1800s), but not nearly good enough to be considered ancient. That is, they are modern forgeries. I do not know if Shapira himself forged them.  But I am quite certain that they are demonstrably modern forgeries, not at all ancient.

II. A Methodological Imperative: Dramatic Claims Require Dramatic and Compelling Evidence (and we just don’t have dramatic, compelling evidence)

From the outset, I would emphasize that dramatic claims require dramatic and compelling evidence. Phrases such as: “Could it be?” “What if?” or “Might it be the case?” are simply not good enough.  Speculations about possibilities, or specious arguments about what a forger would have known, could not have known, might have done, might not have done, could have forged, could not have forged, could have written in his forgery, could not have written in his forgery…well, these have been demonstrated time and time again to be fruitless speculations, certainly not empirical evidence. And when scholars mount such arguments, time and time again they are demonstrated to have been wrong in their assumptions about all of these things that a forger could or could not have done, or could or could not have known.

Let’s now frame this in a very pragmatic fashion.  If the Shapira Fragments were to surface today, the leather would be subjected to carbon 14 tests; the ink would be subjected to chemical analyses (e.g., using a scanning electron microscope equipped with an EDS); there would be very careful analyses, using magnification, of the script itself, its morphology, the stance of the letters, and the ductus (i.e., the number of strokes forming a letter, the direction of those strokes, and the order of those strokes); the patina on the surface above the ink would be analyzed for modern contaminants in it and under it; there would be analyzes of the ways in which the ink had or had not flowed into the current cracks in the leather itself (much as was recently done with the Museum of the Bible’s Dead Sea Scrolls Forgeries). But the Shapira Fragments are lost to history. They were presumably destroyed. So there is no way to do these sorts of basic, benchmark, empirical analyses. And without these sorts of analyses today, no inscription would be declared ancient by a serious scholar trained in epigraphy.

But, in essence, Idan Dershowitz is essentially asking that we forget about all that, and consider the Shapira Strips to be ancient manuscripts, not modern forgeries.  But since the Shapira Strips have disappeared, and are presumably gone forever: (a) there can be no carbon 14 tests of the leather, (b) there can be no laboratory testing of the chemical composition of the ink (e.g., using a Scanning Electron Microscope equipped with an EDS), (c) there can be no careful palaeographic analysis of the script using magnification of the inscriptions themselves, (d) there can be no laboratory analysis of the patina which is present on the leather or on the ink, (e) and there can be no analyses of the ways in which the ink has adhered to the leather (e.g., when someone attempts to forge a manuscript today, the ink will often “leak” into ancient cracks in the leather…and that is very telling, of course).  Moreover, looking at a photo or a hand-copy of an inscription is absolutely not the same as holding an inscription in your hands. There is just no substitute for being able to look at a manuscript oneself and to collate it oneself.  Thus, for someone to attempt to declare the Shapira Strips ancient or authentic in spite of the fact that none of these analyses (such as those listed above) can be done is an absolute deal breaker.  We simply must be able to analyze the Shapira Strips themselves (i.e., the actual documents) before anyone can make a compelling declaration of antiquity.

To put it differently, if an inscription appeared on the antiquities market today, a smart, methodologically savvy, trained epigrapher (i.e., a scholar trained in the actual ancient media, ancient writing technologies, ancient media, etc.) would not declare an inscription to be ancient without first subjecting the inscription to the examinations mentioned above.  Thus, to ask us today to accept as ancient the Shapira Strips when such analyses cannot be done is a bridge too far, way too far.  And, of course, on top of all this, the evidence (mentioned already back in 1883 and 1884 is quite damning, including, but not limited to, the anomalies with the script).  The Shapira Strips are modern forgeries.

Dramatic claims require dramatic, compelling evidence, and we just don’t have it with regard to the Shapira Strips. Rather we have hypotheticals, and circumstantial evidence, at best. And that’s just not going to make the cut, alas, in light of the insurmountable problems with the script (and the eerie parallels to the Moabite Forgeries).

III. Motives: Economic in part, but also in part to Bolster the Traditional View that Deuteronomy was Ancient, not a Pious Forgery of the late 7th century BCE.

There was a strong economic motive for the production of the Shapira Forgeries.  After all the Shapira Strips were stated (by Shapira, among others) to be worth a fortune. But there was more: The Shapira Strips were intended by the forger to Bolster the Traditional View that Deuteronomy was very ancient, and not a pious forgery of the late 7th century BCE. Of course, the forger also knew that any “find” that could bolster the traditional view would garner much attention, and be worth even more, in all sorts of ways.

Many within both Judaism and Christianity have long believed that the Pentateuch (also known as the Torah, that is, Genesis, Exodus, Leviticus, Numbers, and Deuteronomy) was written by Moses, who lived during the 13th century BCE.

The Shapira Strips are largely taken from the book of Deuteronomy (and modeled on the script of the Mesha Stele, which hailed from the 9th century BCE). Thus, if the Shapira Strips were ancient, they would be dated much closer to the time of Moses than any manuscripts of the Pentateuch available in the 1800s (or even now).  For this reason, some people really wanted (then and now) the Shapira Strips to be ancient and authentic, since that would put them much closer to the time of Moses (again, the script of the Shapira Strips was modeled on the Mesha Stele, which did come from the 9th century BCE).    

Now, let’s put all of that in context, in the 1800s.  “Biblical Criticism” was gaining a great deal of headway during the 1800s and was embraced by many scholars and laypeople, but it was repudiated by many others as heretical.  In terms of Biblical Criticism of the 1800s, note that W.M.L. DeWette had argued in the early 1800s that the book of Deuteronomy was a pious forgery from the 7th century BCE, not from the time of Moses in the 13th century BCE.  Please allow me to emphasize that date: EARLY 1800s (decades prior to the forged Shapira Strips)….in other words, this view that Deuteronomy was a pious forgery had been circulating for several decades…and many traditionalists were still upset about it, and they longed for hard evidence to the contrary.

Similarly, the first edition of Julius Wellhausen’s book entitled “The Composition of the Hexateuch” (note: Hexateuch is a way of referring to the first six books of the Bible, namely, Genesis through Joshua) was published in 1876-1877, and in this volume Wellhausen also argued for a “late date” for the composition of Deuteronomy (among other books).  

Again, many in Christianity and Judaism had long believed that Moses (who lived during the 13th century BCE) was responsible for Deuteronomy as well as the rest of the Pentateuch.  But scholars such as DeWette, along with many other scholars in the 1800s, were dating Deuteronomy to the 7th century BCE, that is, around 500 or 600 years after Moses.  Many traditionalists (even many scholars) believed that to be absolutely heretical. 

Indeed, the traditionalists believed that the Bible itself was under assault from Biblical Criticism and its late-dating of Biblical texts. 

Now….enter the Shapira Strips!! They solved everything and vindicated the Bible….that is, The fact that the Shapira Strips seemed to demonstrate that Deuteronomy could be dated much closer to the time of Moses than scholars such as DeWette had contended was widely hailed as absolutely marvelous. And the Shapira Strips could demonstrate that Deuteronomy was much older than the late 7th century….after all, the script of the Shapira Strips was very similar to the script of the Mesha Stele, and the Mesha Stele was 9th century…more than two hundred years prior to DeWette’s dating of Deuteronomy! 

Of course, it should always be remembered that modern forgers (and ancient forgers) know their market.

And it’s also worth emphasizing again that the Shapira Strips were being touted as priceless…and for this reason the price Shapira was asking for them was indeed a fortune. 


IV. Textual Forgeries: A very old story

(1) There is a long history of talented, unscrupulous people (including scholars, antiquities dealers, and disgruntled students) producing textual forgeries (as well as other types of forgeries, of course, as well).  Thus, we have textual forgeries from the ancient, medieval, and modern world….that is, from ancient Mesopotamia and ancient Egypt, as well as ancient Israel and Early Christianity, from the Middle Ages (e.g, the Donation of Constantine), and from the modern world (I deal with this in a print article on the long history of textual forgeries).  Thus, it’s a very old tactic. (2) During the past few centuries, modern forgeries revolving around the Bible and Biblical World have surfaced on the antiquities market. These have generated (even before, during, and after the time of Shapira) a great deal of money.  These forgeries are in various languages, including Greek, Latin, Hebrew, Aramaic, Coptic.  (3) Forgers have often patinated their works so as to cause them to appear ancient.  And also, the forgers will, when possible, use older or ancient media as well (e.g., ancient pots, ancient papyri, parchment). In short, forgers have long been quite sharp, quite knowledgeable. (4) Forgers and dealers often produce forgeries that will peak the interest of scholars and collectors….that is, forgers know the market…they know their target-audience, “marks” well. They know what people want, and they produce fakes with the desired content. (5) Forgers and dealers often go to “text” scholars, rather than to epigraphers, palaeographers, and papyrologists….they do this because text scholars are accustomed to working more with edited texts than with ancient written artifacts (this was the case with the Jesus’ Wife Papyrus, and with countless other modern forgeries). (6) These text-scholars often take the bait, as it were (in my two Maarav articles, I deal with some of this….and the ways in which various scholars ultimately “authenticated” inscriptions which were actually modern forgeries).  (7) Shapira either produced himself, or commissioned the production of, the Moabite Pottery Forgeries and the Moabite Stone Forgeries. These are pretty bad forgeries…and would not fool anyone in the field today, but they fooled quite a few people in the 1870s.  I’ve written about these some (e.g., see the articles in the Finkelstein volume and the Naveh volume, in which I deal some with Shapira, especially his pottery and stone forgeries).  (8) The Shapira Strips were declared modern forgeries in the 1880s on the basis of strong and compelling evidence.  Thus, Ginsberg and Clermont-Ganneau were quite right (that is, long ago, at the time the Shapira Strips surfaced).  (9) Shapira had means, motive, opportunity. .  


V. Basic Principle: After the discovery of a truly sensational ancient inscription, forgers will often produce modern forgeries which are similar in terms of script or content to the authentic, ancient one. The Mesha Stele (discovered in 1868, and often called the Moabite Stone) was just such an inscription…it dates to the 9th century BCE, is written in the Old Hebrew script (because, as the inscription itself mentions, that King Omri of Israel had subjugated the Moabites and held hegemony over them), it is written in the Moabite language (remember: language and script are two different things), and it contains content which dovetails with the Bible’s description of King Mesha of Moab (e.g., 2 Kings 3). The Mesha Stele is a truly sensational, ancient inscription.  

It has been clear for more than a century that he Moabite Clay forgeries were largely the product of Shapira’s friend and business associate Salim al-Kari (Salim had made a squeeze of the Mesha Stele, hence, he knew what the script looked like), while the Moabite Stone Forgeries were the product of Martin Boulos, but the fact remains Shapira himself is also definitely and deeply connected with these blazing forgeries of the early 1870s. In fact, Shapira or an associate of his would sometimes take would-be buyers to a site, state that some of these clay and stone “inscriptions” had recently been found at this or that site, and then he would invite the would-be buyers to dig around…and lo and behold they would find some in the ground.  Of course, these had been “planted” there so that they could be found, but it was a very effective tool in the toolbox of Shapira.

The Shapira Strips are also modeled on the script of the Mesha Stele.  The script of the Shapira Strips is considerably better than that of the Moabite Forgeries.  This is not surprising, as a decade had passed between the time of the production of the Moabite Forgeries and the Shapira Strips.  Scholars had been quite critical of the script of the Moabite Forgeries, and this was all documented in journal articles of that time.  The forger of the Shapira Strips was reading those articles, especially those published in ZDPV.

Of the many similar stories is this more recent one: The Tel Dan Stele Inscription was discovered on the excavation at Tel Dan in 1993 and 1994. This inscription mentions the “House of David” (i.e., the Dynasty of David), is written in Aramaic, and dates to the 9th century BCE. Less than a decade after the discovery of the Tel Dan Stele, a modern forgery (which also fooled a number of scholars) known as the Jehoash Inscription surfaced on the antiquities market (in ca. 2001).  I collated this inscription in the Rockefeller Museum in Jerusalem, a day or two before my testimony for the prosecution in the Israel Forgery trial in the 2000s.  Again, the point is that in the wake of the discovery of a bona fide sensational inscriptional find, forgers begin to produce very similar fakes…and these can, and sometimes do, sell for vast amounts of money.

There is also a very similar situation with a Greek inscription from the Temple Mount, discovered in 1871.  It is a truly fascinating inscription which contains a warning to any foreigners who might venture too far into the Temple complex in Jerusalem.  And not long after this discovery, a forged Temple Mount inscription appeared on the antiquities market. 

In short, the appearance of the Shapira Strips and the Moabite Forgeries follows an established pattern, a pattern that is attested in the periods before and after Shapira’s time.


VI. A few words about the Script

The script of the Shapira Strips is not the same as the bastardized script of the Moabite Clay and Stone Forgeries sold by Shapira. Indeed, the script of the Shapira Strips is much better than the script of the Moabite Clay and Stone Forgeries, but the script of the Shapira Strips has a handful of eerie similarities to the Moabite Clay and Stone Forgeries. 

Idan wishes for us to dismiss this evidence, or to assume that those producing the hand-copies of the Shapira Strips were utterly inept.  But, with all due respect to Dershowitz, we have enough good hand-copies, and even script charts, to be able to state that the script of the Shapira Strips is flawed, and these flaws are similar to the sorts of flaws often found in modern forgeries through the decades.  This evidence cannot simply be dismissed.

  Also relevant: we can state that the script of the Moabite Clay and Stone Forgeries and that of the Shapira Strips is similar (in a few tell-tale ways) because we have some fairly good hand-copies of the Shapira Strips (hand-copies which were made by scholars after Shapira announced the Shapira Strips in ca. 1883), and, of course, many of the Moabite Clay and Stone Forgeries are still in existence today (especially in England and Israel).  In short, the script of the Shapira Strips is a better than that of the Moabite Clay and Stone Forgeries, but the hallmark features that demonstrate the script is forged are present in both groups (i.e., the Moabite Clay and Stone Forgeries as well as that of the Shapira Strips).

VII. Forgers often model their forgeries on ancient literary texts (e.g., the Bible, or some other ancient literary text), or on some ancient inscription.

Especially relevant for the Shapira Strips: In this connection it is useful for me to emphasize another standard method of forgers: forgers often model their forgeries on the script and words of actual ancient texts.  They do this for a number of reasons, one of which is to attract the attention of scholars and the public (as people will often say about these that “they authenticate the Bible”).  But there is another reason as well: it is hard for someone in the modern period to produce a fake which contains no errors with regard to the script, spelling, syntax, and word-meanings (when compared to actual ancient texts), but if the forgers mimic some of the words, sentences, spelling, or syntax, of a genuine inscription or an ancient literary text, it is much easier to avoid mistakes…and so forgers often mimic or copy the words from ancient texts..  Thus, many forgers borrow quite heavily from genuine ancient texts (either inscriptions or literary texts).  The forger of the Shapira Forgeries is very heavily dependent on the book of Deuteronomy.  The forger of the Shapira Strips picked and chose this text and that text (as forgers often do), but the dependence is crystal clear.  Idan Dershowitz wishes to claim that a forger could not have done that. I have learned, in part the way, that making assumptions about what a forger could or could not do, is perilous. For centuries, they have really been quite good, much smarter and better than we thought.

VIII. Modern Forgeries dismissed early on by a consensus of scholars, with some scholar or scholars coming along later and arguing that these were not forgeries after all: We’ve seen this previously too.

After the discovery of the Dead Sea Scrolls (beginning in 1947), some people began to suggest that the (now lost) Shapira Strips might have been authentic and were, basically, the first of the Dead Sea Scrolls.  As already noted, among those who contended this was Menahem Mansoor of the University of Wisconsin.   

This sort of thing happens from time to time with regard to forgeries.  For example, there is a modern forgery known as the “Brazilian Phoenician Inscription.”  This surfaced during the late 1800s and was rapidly dismissed as a modern forgery, that is, it was deemed not to be ancient Phoenician, but rather a modern forgery.  However, during the 1960s, Cyrus Gordon, then of Brandeis University began to contend (in an article published in 1968) that the Brazilian Phoenician inscription was ancient.  Frank Cross of Harvard rapidly wrote a rejoinder (published also in 1968) demonstrating that the Brazilian Phoenician Inscription was a modern forgery, and not a particularly good one. In short, it’s nothing new for someone to come along and suggest that some modern forgery is actually ancient.  It happens.

In the case of this most recent attempt by Idan Dershowitz to suggest that the Shapira Strips are ancient, I would simply note that this has been attempted in the past, and without success (e.g., Menahem Mansoor).  

In Professor Dershowitz’s case, he is attempting to contend that the contents of the Shapira Strips (the things included as well as the things excluded) corresponds with just what scholars would expect for an early version of the book of Deuteronomy. I would counter that it is always precarious to argue that an inscription from the market must be considered ancient based on what we think a non-extant (!) proto-biblical text might have said! That’s putting the cart before the horse in all sorts of ways.


I will be including a long discussion of the Shapira Strips in my forthcoming volume entitled (tentatively), Pious Forgeries: Forging History in the Ancient World of the Bible & the Modern World of Biblical Studies. Eerdmans Publishing Company. Forthcoming. I look forward to continuing this conversation, and to providing additional, detailed evidence against the Shapira Strips.


Christopher Rollston (

For Further Reading:

“Non-Provenanced Epigraphs I: Pillaged Antiquities, Northwest Semitic Forgeries, and Protocols for Laboratory Tests.” Maarav 10 (2003): 135-193.

“Non-Provenanced Epigraphs II: The Status of Non-Provenanced Epigraphs within the Broader Corpus of Northwest Semitic.” Maarav 11 (2004): 57-79.

“Navigating the Epigraphic Storm: A Palaeographer Reflects on Inscriptions from the Market.”  Near Eastern Archaeology 68 (2005): 69-72.

“The Antiquities Market, Sensationalized Textual Data, and Modern Forgeries.” Co-authored with Andrew Vaughn. Near Eastern Archaeology 68 (2005): 61-69.

“The Public Display of Forgeries: A Desideratum for Museums and Collections.”  Written with Heather Dana Davis Parker. Near Eastern Archaeology 68 (2005): 75.

“Who Wrote the Torah according to the Torah?” (August 2017).

“Forging History: From Antiquity to the Modern Period.” Pp. 176-197 in Archaeologies of Text: Archaeology, Technology, and Ethics, eds. Matthew Rutz and Morag Kersel.  Joukowsky Institute Publication Series of Brown University, Oxbow Books, 2014.

“The Ivory Pomegranate: The Anatomy of a Probable Modern Forgery.” Pp. 238-252 in Epigraphy, Philology and the Hebrew Bible: Methodological Perspectives on Philological and Comparative Study of the Hebrew Bible in Honor of Jo Ann Hackett, eds. Jeremy M. Hutton and Aaron D. Rubin. Atlanta: Society of Biblical Literature, 2015.

“The Bullae of Baruch ben Neriah the Scribe and the Seal of Ma‘adanah Daughter of the King: Epigraphic Forgeries of the 20th Century.” Pp. *79-90 (English) in Eretz Israel 32: The Joseph Naveh Volume, eds. Joseph Aviram, Shmuel Ahituv, et al. Jerusalem: Israel Exploration Society, 2016.

“The Putative Authenticity of the New ‘Jerusalem’ Papyrus Inscription: Methodological Caution as a Desideratum,” Pp. 321-330 in Rethinking Israel: Studies in the History and Archaeology of Ancient Israel in Honor of Israel Finkelstein, ed. Oded Lipschits. Winona Lake: Eisenbrauns, 2017.

The Forger Among Us: The Museum of the Bible Dead Sea Scrolls and the Recent History of Epigraphic Forgeries

15 March 2020

The Forger Among Us:

The Museum of the Bible Dead Sea Scrolls and the Recent History of Epigraphic Forgeries

Prof. Christopher Rollston (Ph.D. Johns Hopkins University)

Dept. of Classical and Near Eastern Languages and Civilizations

George Washington University


On March 13, 2020, the Museum of the Bible held a symposium in Washington, D.C.  The focus of the symposium was the presentation of various laboratory tests (on the basis of physical characteristics, elemental and molecular analysis, chemical analyses; using, for example, FTIR analyses, XRF analyses, SEM-EDS analyses, etc.) performed on sixteen fragments, putatively of the Dead Sea Scrolls, which are part of the Museum’s holdings.  The point-person for the laboratory tests was Colette Loll (of Georgetown University), founder and director of Art Fraud Insights, the firm responsible for the laboratory report (which is 212 pages long).  The hard-science team included (in addition to Colette Loll) Abigail Quandt (of the Walters Art Museum), Aaron Shugar (of the State University of New York), Rebecca Pollak (SAFA Senior Research Conservator), Jennifer Mass (Bard Graduate Center for Decorative Arts, Design History, and Material Culture), and Thomas Kupiec (Founder of the Kupiec Group and CEO of ARL, Bio Pharma and DNA Solutions).  The respondents at the symposium (the symposium was entitled “A Journey for the Truth: Investigating the Recent Dead Sea Scrolls Fragments”) were, in order of presentations: Dr. Greg Bearman (Cal Tech), Dr. Christopher Rollston (George Washington University), Dr. Kipp Davis (Trinity Western University), Dr. Sidnie White Crawford (University of Nebraska and Princeton Theological Seminary), and Dr. Lawrence Schiffman (New York University).  Respondents were given access to the full laboratory report (i.e., all 212 pages) during the weeks prior to the symposium.   

              Dr. Jeff Kloha (Chief Curatorial Officer of the Museum of the Bible; Ph.D. University of Leeds; Kloha joined the Museum of the Bible in 2017) began the symposium with a brief welcome, and Dr. Michael Holmes (Director of the Museum of the Bible Scholars Initiative; Ph.D. Princeton Theological Seminary; Holmes joined the Museum in November 2014) provided a brief history of research on the “Museum of the Bible Dead Sea Scroll Fragments” (including the fact that these “fragments” were said to be, in part or in whole, from the Kando family, that is, the family of Khalil Eskander Shahin, who was a broker and seller of many of the Dead Sea Scrolls discovered during the late 1940s and early to mid-1950s).  Among the things that Kloha and Holmes indicated in their remarks was the fact that they fully accept the verdict of Art Fraud Insights: all sixteen Dead Sea Scroll fragments are actually modern forgeries, not ancient.  Based on the constellation of evidence (palaeographic, laboratory), it is (from my perspective, previously and also now) readily apparent that all sixteen of the Museum’s Dead Sea Scroll fragments are indeed modern forgeries.  This seemed to me to be the clear consensus of all of the respondents.

              The conclusions of Art Fraud Insights were preceded by prior laboratory tests (on a smaller selection of the Museum’s Dead Sea Scrolls fragments, tests which also suggested that these fragmentary materials were modern forgeries) and the assessments of various Biblical Scholars and Epigraphers (e.g., Dr. Kipp Davis, Dr. Arstein Justnes, Dr. Michael Langlois, among others).  Thirteen of these Dead Sea Scroll fragments had been published in a volume entitled Dead Sea Scrolls Fragments in the Museum Collection (eds. Emanuel Tov, Kipp Davis, and Robert Duke; Leiden: Brill, 2016, 236pp).  The morning of the symposium, a National Geographic article authored by Michael Greshko was published online (

              It is important to put the Museum’s sixteen modern forgeries in a broader framework.  (1) First and foremost, it should be emphasized that textual forgeries have a very long history, going back to Ancient and Medieval times.  For example, the Famine Stele was an ancient Egyptian forgery (hailing from the Ptolemaic Period but purporting to come from the time of the Egyptian Old Kingdom), the Manishtushu Cruciform Monument was an ancient forgery (purporting to hail from the time of Sargon the Great’s son in the late 3rd millennium BCE, but in reality hailing from the Neo-Babylonian Period, some fifteen hundred years later), and the Donation of Constantine was a Medieval forgery (and in reality not from Constantine the Great, but a forgery from the Middle Ages which purported to be from Constantine the Great; for discussion of these, see Rollston, “Forging History: From Antiquity to the Modern Period,” 2014, pp. 177-184, and the earlier literature cited there).  (2) Textual forgeries (in Phoenician, Hebrew, Greek, Aramaic, etc.) also abounded during the late 19th century and the early 20th century, with the Shapira Forgeries (some on stone, some on pottery, and some on parchment), the Brazilian Phoenician Forgeries, and “the Messerschmidt Clay Forgeries” ( ).  (3) Furthermore, during the final decades of the 20th century and the early years of the 21st century, epigraphic forgeries continued to abound, with the Hebron Philistine Documents, the Moussaieff Ostraca, the Jehoash Inscription (all dealt with in Rollston 2003, Rollston 2004, Rollston 2005, as well as in some later articles of Rollston), the Ma’adanah Seal, the Baruch Bullae (both dealt with in Rollston 2016), the Ivory Pomegranate (dealt with in Rollston 2015), the Ya‘akov [James] Ossuary, the Jesus Wife Papyrus, the Jerusalem Papyrus (the latter two dealt with in Rollston 2017) being some of the most famous.  (4) And the list could go on.  But the point is that these forged Dead Sea Scroll fragments are part of a much broader phenomenon.  It is also important to mention that all of these forgeries of the past 150 years are connected with the Antiquities Market in some fashion. (5) The methods of forgers are varied, but can be fairly well ascertained without much difficulty.  Forgers have a multitude of tools at their disposal (see Rollston 2003, pp. 137-139, forgers can and do use the best of our lexica and grammars, the best of our epigraphic and palaeographic discussions, the best of our historical and cultural discussions.  (6) They also read our discussions of laboratory analyses (cf. Rollston 2003, pp. 182-191, for discussion of various protocols and procedures for laboratory tests, and the failings of some laboratory scientists as well), including the chemical composition of inks and patinas (and forgers will read this superb analysis by ArtFraudInsights, alas, and they will learn much from it so as to produce better forgeries on a variety of media).  And, of course, forgers can and do use ancient media (e.g., ancient potsherds, stones of Levantine quarry, ancient papyrus, ancient vellum; for fuller discussion, see Rollston 2003, pp. 138-139; Rollston 2004).  Indeed, only the most foolish of forgers would not use some ancient medium (e.g., with the hope that a carbon test on the medium may be performed and thus “authenticate” the inscription itself).  (7) And forgers have also (since at least the time of the Shapira Forgeries) attempted to create fake patinas (see Rollston 2003, pp. 183-186).  (8) Furthermore, modern forgers are also becoming quite adept at duping even some fairly good hard-scientists by salting things such as carbonized remains into their fake patinas (as was the case with the Jehoash Inscription; see Rollston 2003, pp. 183-186 et passim), and by using ancient carbonized remains (e.g., from the remains of beams burnt in antiquity) to make inks which can pass laboratory tests (dealt with especially in Rollston 2017, pp. 322-323).

              The motives of the forgers are varied.   Here are the ones which I have contended (in various publications, including Rollston 2003, pp. 191-193; Rollston 2014, pp. 176-177) are most common during the past 150 years, but many (if not all) are also part of the ancient world in some fashion. Some have suggested (or assumed) that the primary motive for forgers is economic.  True enough, venality is certainly a motive through all time.  However, I am confident that a larger number of motivations can be posited, based on known forgeries: (1) Venality (i.e., greed); (2) Hubris (i.e., the forger’s belief that he or she is too good to make mistakes, detectable or otherwise); (3) “Sour Grapes” (e.g., of a doctoral student purged from a program, or a veteran scholar who feels he or she has been slighted by another scholar); (4) Professional Rivalry (e.g., the production of a forgery which confirms something which said scholar has previously said in print); (5) Pranks (i.e., an outright joke); (6) Professional/Personal Aggrandizement (e.g., fame for being associated with some stunning new inscription from the market); (7) Religion and Politics (especially the desire to prey on the sincere beliefs, hopes, and fears of good people).  I have discussed all these in greater detail in various places through the years (again, Rollston 2003; Rollston 2004; Rollston 2014).

As an ancillary note, and as a modus operandi of forgers: namely, forgers of, dealers of, and owners of forged antiquities often seek out scholars whom they believe will “authenticate” their finds.  Along these lines, forgers, dealers, and owners know which scholars most frequently “authenticate” inscriptions (including some forgeries) from the Antiquities Market and they especially seek out the opinions of those scholars (because they want their own inscriptions to be authenticated). 

Similarly, forgers, dealers, and owners will often approach scholars whom they know do not work with epigraphic remains, but are rather “text scholars” (i.e., scholars who mostly work with edited texts rather than the actual physical objects…and so these scholars do not intimately know the look and feel of something which is actually ancient). 

It is important to mention that although some have suggested that scholars are never the culprits, the facts on the ground demonstrate otherwise.  Note, in this connection, the famous forgery published by Princeton Professor Coleman-Norton (dealt with in Rollston 2003, pp. 192-193), a forger who was outed by his own student, Bruce Metzger.


              The Forger of the Museum of the Bible Dead Sea Scroll Forgeries: With this basic summary of some of the early history of forgeries, I should like to return, in earnest, to the forger of the Museum of the Bible’s forgeries.  I shall be fairly forthright here.  Here is a profile of the forger: I believe that the forger of these Dead Sea Scrolls forged fragments is a trained scholar in our field, with access to actual ancient scrolls.  I believe that the forger forged them during the course of a few months, or more likely, a couple years (this also accounts for some of the variation in the script).  I believe that venality (indeed, outright and blatant greed) is a primary motivation (literally, netting the forger millions of dollars for these Museum of the Bible forgeries), but greed is not the only motivation.  I believe the scholar of these forgeries is particularly hubristic, and assumed he (or she) could fool all other scholars (and also probably delighted in this assumption).  I do not think that these were forged as some sort of a joke (as was the case in the Coleman-Norton forgery and in the case of the Hebron Philistine Documents). Clearly, I believe that the forger is amoral.  Also, I believe that the forger worked primarily alone, but could have included a paid friend or associate who had at least a high-school level knowledge of chemistry (these forgeries are not sophisticated enough to have included the assistance of a trained scholar in chemistry). 

              Also, I believe that a good investigative journalist should be capable, given the resources (e.g., several months of compensated work) of a good newspaper or learned society, should be able to discover the identity of the forger.  I hope that the weight of all of the relevant national and international laws is brought forth against this forger (although, as a realist with regard to conviction rates, I suspect that the most that can be hoped is that the identity of the forger will be discovered).

              Some final reflections: The Museum of the Bible (and its precursors, associated entities, etc.) engaged in some lamentable and egregious actions, as well as lapses of professional ethics, some of which were also breaches of the law (hence the US Dept. of Justice case, of course).  This cannot, and should not, be forgotten (for all sorts of reasons).  It was bad, and it was wrong (I’m thinking here of, among other things, the purchase of many objects, including cuneiform tablets, from conflict zones in the Middle East).  And the Museum has paid a price for those actions, not only in the serious tarnishing of the reputation of the Museum and the surrender of pillaged objects and inscriptions, but also substantial financial penalties from the US Dept. of Justice.  But I have also long said, and long believed, in the possibility and importance of true penitence and the possibility of forgiveness and redemption.  Indeed, I do believe in the possibility of forgiveness and redemption, on a personal scale as well as for organizations, learned societies, and even nations. 

              So, for what it’s worth, here’s what I think in this case of these forged Dead Sea Scroll fragments now in the custody of the Museum of the Bible, and some of the broader implications.  These were purchased with the understanding that they were known prior to the enactment of the “Convention on the Means of Prohibiting and Preventing the Illicit Import, Export and Transfer of Ownership of Cultural Property – 1970” (e.g., that they came from the Kando family and so the chain of custody was secure).  Frankly, I believe those advising the Museum at the time of the purchase were (at the very least) naïve, but that’s another story for another time.  The thing that I wish to emphasize here is that the Museum of the Bible ultimately came clean: They rapidly acknowledged that these scroll fragments might be modern forgeries and that the lore (and documents) associated with their origins might be a fabrication (indeed, I have long said that I put zero credence in any “statement about the antiquity” or “documentation” that is shown to me by an owner, collector, or dealer, as they have so many reasons to prevaricate).  Then, the Museum funded the very extensive laboratory tests (I have no idea what this must have cost, as the laboratory reports were numerous, conducted by first-tier scientists, and are reported in a tome that exceeds 200 pages….the final sum for all this work must be in the five figures, and I would not be surprised if it was in the low six figures).  And then, to top it all off, major people in the current senior leadership at the Museum (e.g., Jeff Kloha and Michael Holmes….and many others, I believe as well) have embraced the results of the laboratory report: All sixteen of these Dead Sea Scroll fragments are modern forgeries.  I would imagine that this was, and is, a tough pill to swallow.  And frankly, I’m impressed that they publically and unambiguously stated that they accept this conclusion.  Indeed, I’m impressed.  I’m not surprised, as I have come to know Jeff Kloha and Michael Holmes some during the past few years.  These are honest people, good and kind people, and smart scholars.  Frankly, I have come to like and appreciate them. 

              One final point: I have believed for a long time (and said in print long ago, after a comment to me from mentor and friend, the late Frank Moore Cross), museums should have forgery exhibits so as to raise the awareness of the public and of scholars about the problem of modern forgeries (see Rollston and Parker 2005).  The Israel Museum did this a number of years ago with regard to the Shapira Forgeries.  It was a marvelous exhibit, drew a lot of attention to the problem, did a lot of good, and my earnest hope is that the Museum of the Bible will allocate space, as a signal of its further desire to right wrongs, for an exhibit of these forged Dead Sea Scroll fragments.  And, along those same lines, since many museums have modern forgeries in their holdings (including the Smithsonian and the Met), perhaps the Museum of the Bible could work with such museums to produce a first-class exhibit about the problem of the antiquities market and the problem of modern forgeries.  That’s an exhibit I would love to see.  It’s unusual, it’s important…and, of course, the Museum does have the forgeries.

Select Bibliography

Rollston, Christopher A. “Non-Provenanced Epigraphs I: Pillaged Antiquities, Northwest Semitic Forgeries, and Protocols for Laboratory Tests.” Maarav 10 (2003): 135-193.

______.  “Non-Provenanced Epigraphs II: The Status of Non-Provenanced Epigraphs within the Broader Corpus of Northwest Semitic.” Maarav 11 (2004): 57-79.

______.  “Navigating the Epigraphic Storm: A Palaeographer Reflects on Inscriptions from the Market.”  Near Eastern Archaeology 68 (2005): 69-72.

Rollston, Christopher; Heather Dana Davis Parker.  “The Public Display of Forgeries: A Desideratum for Museums and Collections.”  Near Eastern Archaeology 68 (2005): 75.

Rollston, Christopher.  “Forging History: From Antiquity to the Modern Period.” Pp. 176-197 in Archaeologies of Text: Archaeology, Technology, and Ethics, eds. Matthew Rutz and Morag Kersel.  Joukowsky Institute Publication Series of Brown University, Oxbow Books, 2014.

______“The Ivory Pomegranate: The Anatomy of a Probable Modern Forgery.” Pp. 238-252 in Epigraphy, Philology and the Hebrew Bible: Methodological Perspectives on Philological and Comparative Study of the Hebrew Bible in Honor of Jo Ann Hackett, eds. Jeremy M. Hutton and Aaron D. Rubin. Atlanta: Society of Biblical Literature, 2015.

______.  “The Bullae of Baruch ben Neriah the Scribe and the Seal of Ma‘adanah Daughter of the King: Epigraphic Forgeries of the 20th Century.” Pp. *79-90 in Eretz Israel 32: The Joseph Naveh Volume, eds. Joseph Aviram, Shmuel Ahituv, et al. Jerusalem: Israel Exploration Society, 2016.

______.  “The Putative Authenticity of the New ‘Jerusalem’ Papyrus Inscription: Methodological Caution as a Desideratum,” Pp. 321-330 in Rethinking Israel: Studies in the History and Archaeology of Ancient Israel in Honor of Israel Finkelstein, ed Oded Lipschits.  Winona Lake: Eisenbrauns, 2017.

The Tel Aviv University PNAS Study: Some Methodological Musings

13 April 2016


Tel Aviv University’s Epigraphic Hebrew Project is among the most innovative and important in the world, with the collaboration of scholars from the hard sciences, epigraphy, and archaeology. During recent years, a number of seminal articles have been published as part of this project. Very recently, an additional multiple-author article has appeared in PNAS (April 2016). This article is entitled “Algorithmic handwriting analysis of Judah’s military correspondence sheds light on composition of biblical texts.” I have served as a consultant on Tel Aviv’s Epigraphic Project and find the project to be particularly important, productive, and auspicious. I am not among the authors of this article, but I find the technology, hard sciences, and mathematics in the article to be especially impressive. At this juncture, I shall summarize the salient components of the PNAS article, some of the conclusions of the article, and then I shall offer some sober reflections.

I. The Tel Aviv University Study, the Postulates, the Conclusions:
The Tel Aviv study is based on an innovative an important algorithmic analysis of sixteen ostraca (i.e., ink inscriptions on broken piece of pottery called potsherds) from the Judean fortress of Arad. The English publication of the Hebrew-language Inscriptions from Arad appeared in print in 1981 (Aramaic inscriptions were also found there, but from a later period). The volume is entitled Arad Inscriptions, authored by Yohanan Aharoni (Jerusalem: The Israel Exploration Society). The Hebrew inscriptions in this volume number 112 (this includes ostraca, inscriptions incised into pottery, and seals).
(1) The Tel Aviv University study published in PNAS analyzed the following ostraca: 1,2,3, 5, 7, 8, 16, 17, 18, 21, 24, 31, 38, 39, 40, and 111. (2) The Tel Aviv study presupposes that all sixteen of these inscriptions were written ca. 600 BCE (pages 2-3). (3) Based on the analysis of the handwriting on these sixteen inscriptions, the authors stated that they “deduce the presence of at least six authors” in this group of sixteen ostraca (page 1). (4) The authors of the PNAS study believe that some of these ostraca were “most likely” composed at Arad (e.g., Ostraca 31 and 39) and some were probably dispatched from other locations (e.g., Ostraca 7, 18, 24, 40). Within the abstract it is stated even more emphatically. Namely, “our algorithmic analysis, complemented by the textual information, reveals a minimum of six authors within the examined inscriptions. The results indicate that in this remote fort literacy had spread throughout the military hierarchy, down to the quartermaster and probably even below that rank” (page 1). (5) Because Arad was a military fortress and there are some named military officials, the authors posit that there must have been “a proliferation of literacy” at this time period (NB: the term “proliferation of literacy” occurs numerous times in the article), that is, around 600 BCE. (6) Part of their reasoning is based on the fact that there were similar fortress sites at places such as Tel Malhata, Lachish, Horvat ‘Uza and so from this they contend that by “extrapolating the minimum of six authors in 16 Arad ostraca to the entire Arad corpus, to the whole military system in the southern Judahite frontier, to military posts in other sectors of the kingdom, to central administration towns such as Lachish, and to the capital, Jerusalem, a significant number of literate individuals can be assumed to have lived in Judah ca. 600 BCE” (page 3-4). (7) The authors then argue that “the spread of literacy in late-monarchic Judah provides a possible stage setting for the compilation of literary works. “ The authors concede that “biblical texts could have been written by a few and kept in seclusion in the Jerusalem Temple, and the illiterate populace could have been informed about them in public readings and verbal messages by these few.” But they contend that “widespread literacy offers a better background for the composition of ambitious works such as the Book of Deuteronomy, and the history of Ancient Israel in the Books of Joshua to Kings” (page 4). (8) Ultimately, the authors conclude that by ca. 600 BCE, therefore, all of this “implies that an educational infrastructure that could support the composition of literary texts in Judah already existed before the destruction of the First Temple” (586 BCE).


II. Rollston’s Reflections
A. There has been sufficient inscriptional evidence for some time from the world of ancient Israel to contend that already by 800 BCE there was sufficient intellectual infrastructure, that is, well-trained scribes, able to produce sophisticated historical and literary texts. Indeed, I argued for this a decade ago in a detailed epigraphic article (Rollston 2006), and I was not the first epigrapher to do so. Moreover, among the most important recent monographs discussing scribal education in ancient Israel, within the broader context of the ancient Near East are those of David Carr (2005), Karel Van der Toorn (2007), Seth L. Sanders (2009). And long before this, the subject of schools and literacy in ancient Israel was the object of much discussion in scholarly literature, with Hermisson (1968), Whybray (1974), Lemaire (1981) being among the most enduring. Reference could also be made to the work of D.W. Jamieson-Drake’s work (1991), although a number of scholars have discussed the serious problems with the data and the assumptions that served as a foundation for that volume. And, of course, I would wish to emphasize the careful and enduring work of the great James Crenshaw on the subject of scribal education in ancient Israel (1985; 1998). In short, the subject of writing and literacy in ancient Israel and Judah has a long history within the field. As for my own work, the main point for which I have contended is that we have sufficient epigraphic evidence to demonstrate that there was detailed, sophisticated, standardized education in the Old Hebrew writing system (script, orthography, hieratic numerals, phonology) in ancient Israel and Judah. And the evidence for that is already present by ca. 800 BCE (Rollston 2006; 2008; 2010; 2012). For this sophisticated scribal apparatus, I used the term “intellectual infrastructure” during a presentation in Jerusalem during May of 2013 (now forthcoming 2016). I believe that the cumulative epigraphic evidence from sites such as Kuntillet Ajrud, Beth Shean, Tel Rehov, Arad, Samaria, Jerusalem, Lachish, Horvat Uza, Tel Ira, and Beer-Sheva is compelling. And epigraphs in Phoenician, Moabite, Aramaic, Ammonite, and even Edomite augment the Old Hebrew data substantially. More recently, I have discussed scribal curriculum in Old Hebrew in some detail (2015; forthcoming 2016). Thus, the first point that I would wish to make regarding the PNAS study is that with its date of ca. 600 BCE, this study is too conservative. I would contend that we have such evidence already two hundred years prior to this. As an ancillary note, I should like to emphasize that I am not arguing here arguing that this or that portion of the Bible hails from a particular place and time (that is a separate, longer discussion, of course, because of the long transmission history of much of the biblical material), although I concur with the consensus of the field that the late 7th and early 6th centuries were periods of substantial literary productivity in Judah. But most importantly, to reiterate, I am contending that the epigraphic evidence at hand demonstrates rather nicely that there were educated scribes in Israel and Judah by the late 9th and early 8th centuries BCE and that these scribes were capable of writing fine historical and literary texts. Thus, in sum, as for the PNAS article, I would say (with some good-natured humor and a turn of phrase), “I see your 600 and raise you 200” (i.e., to ca. 800 BCE).

B. The authors of this study contend that all sixteen of the ostraca they analyzed were written around 600 BCE. However, the original excavator (Yohanan Aharoni) argued that some of these ostraca (i.e., the ones studied in the PNAS article) came from stratum VI (e.g., 1-24, etc.), some from stratum VII (e.g., Arad 31, 38, 39), and at least one (Arad 40) from stratum VIII (Aharoni 1981). In other words, the original excavator argued that the sixteen ostraca used in this study were from three different chronological horizons, not one. That is, the original excavator believed that these ostraca definitely did not come from the same time-frame, but rather during the course of ca. a century. There has been a substantial amount of discussion regarding the stratigraphy of Arad (especially Z. Herzog 2002, 3-109) and even Aharoni noted that some of the inscriptions were found in loci that he viewed as mixed or unclear (see Aharoni 1981, 181-185 for a loci-table and also his discussion in the body of the volume for all the ostraca). I conversed with Professor Ze’ev Herzog about Arad about the stratigraphy of Arad VI and VII a number of years ago (namely, 1998), as I wished to see strata VI and VII as contemporary. At that time, he emphasized to me that these were separate strata, sequential, not contemporary. That is not to suggest that there is a vast expanse of time between these two strata, as there is not, but the point he emphasized was that stratum VII and VI were sequential, not contemporaneous. In terms of absolute dates, Aharoni dated the Stratum-VII destruction to ca. 609 BCE and the Stratum-VI destruction to ca. 586 BCE. As for stratum VIII, the dates for it have been much discussed as well. Aharoni argued that Stratum VIII-destruction was ca. 701 BCE (during the punitive campaign of Sennacherib against Judah). Some would wish to push the date a little later, of course. In any case, Arad Ostracon 40 is one that Aharoni considered to hail from Stratum VIII. Nadav Na’aman (one of the authors of the PNAS study) has argued that Arad Ostracon 40 is to be associated with the ostraca from Arad VI. His reasons are as follows: (a) “No eighth century letter written on a potsherd has been discovered in any site in Palestine….it seems that writing letters on pottery began only in the seventh century BCE.” (b) “the orthography indicates a relatively late date, with internal matres lectionis for ‘yš (lines 7-8) and yhwd[h] (line 13).” (c) “Epigraphically, the letter [Arad 40] has many parallels with both Stratum VIII and VII-VI ostraca…” (d) “The situation described in the ostracon closely matches the reality of the late years of the Judahite monarchy.” (e) In addition, Na’aman suggests that “Malkiyahu, the recipient of the letter [Arad Ostracon 40] is possibly the same officer Malkiyahu the son of Zerabu’ur, who led troops to Ramat-Negev according to Ostracon 24 from Arad” (Na’aman 2003, 199-204). The authors of the PNAS article embrace Na’aman’s view (see footnote on page 3, and note also that the authors of the PNAS study are also very much aware of the difficult stratagraphic history of Arad and the secondary literature discussing it). I would suggest that a fair amount of Na’aman’s reasoning is potentially problematic. (a) Thus, regarding epistolary texts in the ancient Near East, I would emphasize that this is a very old practice, centuries older than even the oldest Levantine alphabetic texts. That is, letters were around in the ancient Near East for a very long time. Moreover, using potsherds as a medium for writing alphabetic texts is also well attested long before ca. 600 BCE, of course. I would be cautious, therefore, about arguing on the basis of an absence of an epistolary ostracon, that we should date Arad 40 to the early 6th century (i.e., Stratum VI). (b) The usage of internal matres lectionis in Old Hebrew inscriptions does increase during the late 7th and early 6th centuries, but the fact of the matter is that we do have the usage of matres lectionis in Old Hebrew inscriptions already by the end of the 8th century, with the Royal Steward Inscription from Jerusalem being a prime example (Rollston 2006, 63-64). (c) As for the script, I find some of the forms in Arad 40 to reflect a time-frame prior to the late 7th or early 6th century. In short, I do not see a compelling palaeographic reason for dropping the date of Arad 40. (d) Judah was in a difficult political situation from the reign of Ahaz (d. ca. 715 BCE) to the assassination of the Judean Governor Gedaliah (sometime shortly after the fall of Judah in 586 BCE). And Edom was a alive and well in the region for much of this period, as the two-volume magnum opus edited by Thomas E. Levy, Mohammad Najjar, and Erez Ben-Yosef demonstrates (2014). Therefore, I find it difficult to assume that the only time during the final century of the First Temple Period that Edom could have been a nemesis for Judah was ca. 600 BCE. (e) As for the personal name Malkiyahu, note that Na’aman used the word “possibly.” That’s important. Moreover, a quick look at the Hebrew Bible reveals that some eight people have names based on this root, and I’m confident that there are more (if I looked harder). In short, this is not a rare name. So to assume that they are the same person is…well…an assumption that might be erroneous. Note in this connection Lawrence Mykytiuk’s foundational work on personal names, a work that reminds us all of the importance of having at least a shared name and a shared patronymic (or some other inscribed feature, such as the same title, etc., etc.) for any attempt to suggest some sort of identification between two individuals (2004). And, of course, in Arad Ostracon 24, we have reference to someone called Malkiyahu son of Zerab’ur, but in Arad Ostracon 40 we have reference simply to Malkiyahu (with no patronymic). That’s a problem for anyone attempting to posit a certain, or near-certain identification. Note, therefore, that because papponymy (naming after a grandfather) was a very common practice in the ancient Near East (including Israel and Judah), one could make a decent case that the Malkiyahu of Arad Ostracon 40 is the grandfather of the Malkiyahu of Arad Ostracon 24, with the former being a well-known patriarch of the family. Although this might (and I repeat, *might*) account for most of the data that we have (including the fact that in the case of Arad 40 we do not have a patronymic but in Arad 24 we do), I am disinclined to speculate. After all, without a patronymic, and with names based on the root mlk being fairly common, it is simply too problematic to argue that these two are definitely the same person. So, I won’t. In this too, therefore, I must differ with the authors of the PNAS study.

C. As for the contention of this Tel Aviv Study that we can posit a “proliferation of literacy” at ca. 600 BCE, based on the assumption that there are at least six different hands in the sixteen ostraca that were analyzed, I would simply suggest that this is a very broad assumption that I would not be inclined to make. After all, the assumption of the PNAS study that all of these ostraca come from ca. 600 BCE is difficult to embrace (see discussion above). Furthermore, in reality we do not know how many of these ostraca might, or might not, have been produced at the site of Arad. Compare, however, some of the assumptions of the Tel Aviv study (e.g., page 3, and footnote). That is, from my perspective, there is nothing in the content of these ostraca that make it at all compelling to state that any were definitely produced at Arad. After all, even lists could have come from elsewhere. That is not to say that there were not readers and writers at Arad. There were. After all, that’s where the ostraca ended up. But the origin-point is not something that can be ascertained on the basis of the data at hand. Thus, rather than arguing on the basis of sixteen ostraca (that ended up at Arad) that we have a “proliferation of literacy,” I would simply conclude that we have some readers and writers of inscriptions at Arad. That’s all we can say. Who were these readers and writers at Arad? I would emphasize that these writers may very well have been scribes associated with the army. After all, we do have references in the Hebrew Bible (e.g., 2 Kings 25:19; Jeremiah 52:25) to the “scribe of the leader of the army” (and I find this title to be something that can be considered a credible, historical thing). In addition, I do think that it is reasonable to contend that some military officials (at various levels of the command structure) could read and write, and I have noted this in print as well. But, I would wish to emphasize that we do not have evidence for enlisted soldiers, that is, the average soldier, reading and writing. We have evidence for some literacy among some of the army officials. That’s what I think we can say. Furthermore, although this study does not suggest this, per se, some have already seized upon it (within hours of the appearance of this study!) and begun to contend that we have people from all walks of life writing and reading in First Temple Judah. That’s quite a leap. I would, therefore, emphasize that we have no evidence for the common folk writing and reading, not from the epigraphic record and not from the Bible (see especially Ian Young’s articles on literacy, 1998a; 1998b). I’d like to be able to state that there were carpenters who built houses by day and read papyri manuscripts at night. And I’d like to say that there were blacksmith’s shaping metal over a furnace during the day and penning contracts at night. And I’d like to be able to say that there were potters turning pots on the wheel by day and writing alphabetic acrostics by the light of olive-oil lamps before dawn. And I’d like to say that there were shepherds guarding their flocks by day and writing out parchment king-lists by night. But I can’t. We have no inscriptions with content that causes us to think that people from these vocations are producing or consuming texts (on some poorly written texts, by people with little training, see below). Some might counter that we seem to have so many more inscriptions from the late First Temple Period and so we must assume the proliferation of literacy. I would counter that we also have a growing population and thus a burgeoning governmental apparatus during the late First Temple Period, and so we have a need for more professional scribes and literate elites. Governments need scribes and literate officials. And as government grows, so does the number of scribes. That’s fine, but it does not mean that we can assume that the general populace (people from all walks of life) is reading and writing texts. Furthermore, since we have a growing population in Judah, even though we have more scribes and literate officials, it could be argued that the percentage of the population that is literate stays about the same. Also perhaps of some consequence regarding readers and writers in antiquity, we actually have a Jerusalem scribe in the early 2nd century BCE who notes that scribes were reading, writing, traveling, and solving riddles….but the average person was simply not able to do such things (Sira 38:24-39:11). It is reasonable to posit that this was even more the case during the First Temple Period.

D. To be sure, most people contemplating the subject of literacy today bring their own experiences to the table, namely, the widespread literacy of the modern world. But in the modern world, we normally have government mandated education of (for all practical purposes) the entire population. Things were very different in antiquity. We have no evidence at all of government mandated education of large portions of the populace in antiquity. So what do we have? We know that scribes and high governmental officials (temple, palace) and military officials read and wrote. Did some of the trades-people sometimes learn to write? I suppose so. But was it common? No, I don’t think so. Furthermore, note that we have several hundred Old Hebrew inscriptions from the Iron Age that are very well done, with a sophisticated knowledge of letter morphology, stance, orthographic conventions, the use of hieratic numerals, knowledge of epistolary conventions, some understanding of phonology, the ability to use effectively different tenses, parts of speech, and to do so very well. I’d like to suggest that everyone could do that, but I can’t. And, in fact, when someone without formal, standardized education attempted to write an Old Hebrew inscription (or any other ancient script), it is painfully obvious. And we do have a few inscriptions to prove that! In short, much as it pains me to say it, the writers in readers of texts in ancient Israel and Judah were elites, not the common person. In essence, the common person could get along just fine in life without learning to read or write (see Rollston 2006, 48-49 regarding the time it takes to learn one’s first writing system, even an alphabetic one).


So, in sum, the Tel Aviv Epigraphic Project is scintillating. The technology and talent that the authors of this PNAS article bring to the table is unmatched anywhere in the world. But the sociological conclusions about the “proliferation of literacy” in Judah is not something that can be posited on the basis of this study. The methodology is stunningly important, but I would wish to see more caution regarding the conclusions.


Aharoni, Y. Arad Inscriptions. Jerusalem: The Israel Exploration Society, 1981.
Carr, D.M. Writing on the Tablet of theHeart: Origins of Scripture and Literature. Oxford: Oxford University Press, 2005.
Crenshaw, J. L. “Education in Ancient Israel.” JBL 104 (1985): 601-615.
_____. Education in Ancient Israel: Across the Deadening Silence. New York: Doubleday, 1998.
Faigenbaum-Golovin, S.; Shaus, A.; Sober, B. ; Levin, D.; Na’aman, N.; Sass, B.; Turkel, E.; Piasetzky, E.; Finkelstein, I. “Algorithmic handwriting analysis of Judah’s military correspondence sheds light on composition of biblical texts.” PNAS Early Edition (April 11, 2016): 1-6.
Jamieson-Drake, D.W. Scribes and Schools in Monarchic Judah: A Socio-Archaeological Approach. JSOTSup 109. Sheffield: Sheffield Academic Press, 1991.
Hermisson, H. J. Studien zur israelitischen Spruchweisheit. WMANT 28. Neukirchen-Vluyn: Neukirchener, 1968.
Herzog, Z. “The Fortress Mount at Tel Arad: An Interim Report.” Tel Aviv 29 (2002): 3-109.
Lemaire, A. Les écoles et la formation de la Bible dans l’ancien Israël. OBO 39. Göttingen: Vandenhoeck & Ruprecht, 1981.
Levy, T. E., Najjar, M., Ben-Yosef, E., eds. New Insights into the Iron Age Archaeology of Edom, Southern Jordan: Volumes 1-2. Los Angeles: Cotsen Institute of Archaeology Press, UCLA, 2014.
Mykytiuk, L. J. Identifying Biblical Persons in Northwest Semitic Inscriptions of 1200-539 BCE. SBLAB 12. Atlanta: Society of Biblical Literature, 2004.
Na’aman, N. “Ostracon 40 from Arad Reconsidered.” Pp. 199-204 in Saxa Loquentur: Studien zur Archäologie Palästinas/Israels, Festschrift für Volkmar Fritz zum 65. Geburtstag, eds. C. G. Den Hertog, U. Hübner, S. Münger. Münster: Ugarit-Verlag 2003.
Rollston, C. A. “Scribal Education in Ancient Israel: The Old Hebrew Epigraphic Evidence.” BASOR 344 (2006): 47-74.
_____. “The Phoenician Script of the Tel Zayit Abecedary and Putative Evidence for Israelite Literacy.” Pp. 61-96 in Literate Culture and Tenth-Century Canaan: The Tel Zayit Abecedary in Context, eds. R. E. Tappy and P. K. McCarter, 2008.
_____.Writing and Literacy in the World of Ancient Israel: Epigraphic Evidence from the Iron Age. SBL Archaeology and Biblical Studies, 11. Atlanta: Society of Biblical Literature, 2010.
_____. “An Old Hebrew Stone Inscription from the City of David: A Trained Hand and a Remedial Hand on the Same Inscription.” Pp. 189-196 in Puzzling Out the Past: Studies in Northwest Semitic Languages and Literatures in Honor of Bruce Zuckerman, eds. M. J. Lundberg, S. Fine, W.T. Pitard. Leiden: Brill, 2012.
_____. “Scribal Curriculum during the First Temple Period: Epigraphic Hebrew and Biblical Evidence.” Pp. 71-101 in Contextualizing Israel’s Sacred Writings: Ancient Literacy, Orality, and Literary Production, ed. Brian B. Schmidt. SBL Ancient Israel and Its Literature, 22. Atlanta: SBL, 2015.
Sanders, S. L. The Invention of Hebrew. Urbana: University of Illinois Press, 2009.
Toorn, K. van der. Scribal Culture and the Making of the Hebrew Bible. Cambridge: Harvard University Press, 2007.
Whybray, R.N. The Intellectual Tradition in the Old Testament. BZAW 135. Berlin: de Gruyter, 1974.
Young, I. “Israelite Literacy: Interpreting the Evidence, Part 1. VT 48 (1998a): 239-253.
_____. “Israelite Literacy: Interpreting the Evidence, Part 2.” VT 48 (1998b): 408-422.


22 March 2015

By Dr. Christopher Rollston, George Washington University (Washington, DC)


Myth Number One: Forgeries are Rare. Actually, forged inscriptions are quite common. In fact, people have been forging inscriptions for a very long time. True, recent decades have witnessed many modern epigraphic forgeries, coming on the heels of many notable forgeries during the late 19th century and early 20th century (Rollston 2003; Rollston 2004; Vaughn and Rollston 2005; Rollston 2005). But epigraphic forgeries are attested not just during the modern period, but also during the Middle Ages, and even earlier, with some of the earliest forgeries hailing from ancient Mesopotamia and ancient Egypt (Rollston 2014), and with ancient Christians producing scores of modern forgeries as well (Ehrman 2012). During the past 140 years, most forged inscriptions were sold on the antiquities market.

Myth Number Two: Forged Inscriptions are Easy to Detect. Some forgeries are of low quality and easy, therefore, to detect, but some are so good that they have fooled some of the best scholars in the world (for discussion, see Rollston 2003; Rollston 2014).

Myth Number Three: There was just one Jesus. Actually, this was a fairly common name in the Second Temple Period (i.e., the time when Jesus of Nazareth lived). In fact, within the New Testament, there are at least five people with the name Yeshua (Jesus): namely, Yeshua (Joshua) the successor of Moses (e.g., Acts 7:45); Jesus son of Eliezer (Luke 3:29), Jesus Barabbas (Col 4:11), a resident of Cyprus named Bar-Jesus (i.e., son of Jesus, Acts 13:6) and Jesus of Nazareth. Second Temple Jewish literature contains references to many additional people with this name. Significantly, therefore, even if the entire “James Ossuary Inscription” were ancient, there would be no guarantee that it belonged to the brother of Jesus of Nazareth. For that to be the case, there would need to be a more specific descriptor, something such as “the brother of Jesus of Nazareth,” or “the brother of Jesus who is called the Messiah.” Since we do not have that sort of data for this inscription caution is required.



The James Ossuary was not found on a scientific expedition. Rather, it was pillaged from the ground and sold on the antiquities market. Thus, there is no reliable chain of custody for this ossuary. There are no photos of its discovery. There is no means of ascertaining the number of bones that were placed in this ossuary. Moreover, it should be remembered that the bones of multiple people were placed in a single ossuary at times (regardless of whether there was no name inscribed on the ossuary, or just one name, or two names, etc., etc.), and so even if some bone fragments were still in this ossuary, it would have been impossible to determine whose they were. Of course, because the James Ossuary was not found on a scientific expedition, and because the chain of custody cannot be determined with any certainty (because it has been in the hands of antiquities dealers and antiquities collectors), and because antiquities dealers sometimes place some bones in one ossuary that came from a different ossuary, to attempt to discern anything from any bone fragments in the James Ossuary would be fruitless. After all, without a clear chain of custody, forensic medicine is not at all decisive.


The script of the first half of this inscription /Y‘qwb br Ywsp/ reflects distinct depth and clarity. In addition, kerning (quite common in this period) is present. However, the second half of the inscription /’hwy d yshw‘/ is not carved with the same depth, clarity, and kerning. Because of camera angle and lighting (as well as shadowing), some photographs capture this rather well. Personal collation confirms that there is indeed a distinct difference in depth, clarity, and kerning between the first and second halves the inscription. Compare also the predominant consistency of the depth, clarity, and kerning of some of the longer ossuary inscriptions (e.g., Rahmani 70 Lid, Plate 11 [deep with kerning, slight reduction in size because of space constraints]; Rahmani 80 Lid, Plate 13; Rahmani 370 Lid, Plate 52 [light incising throughout]; Rahmani 430 Rim, Plate 62 [deep incising and kerning throughout]; Rahmani 560 Lid, Plate 80 [consistent depth and clarity]; Rahmani 796 Front, Plate 116 [light incising throughout]. Cf. Rahmani 12 Lid, Plate 2 [with “signature” not as deep, etc.]; Rahmani 893 Front, Plate 135 [reduction of depth and size, arguably an issue of centering].

Most Tenable Conclusion: Two Hands, with the words “James son of Joseph” written by one person and “Brother of Jesus” (i.e., brother of Yeshua) written by another person. However, based just on the script, it is tenable to suggest that (a) the hands are both ancient; or (b) the second hand is modern; (c) the entire inscription is modern and the forger was not assiduousness enough in forging. Within the fields of biblical studies, hermeneutics of suspicion have been invaluable. In my opinion, it is prudent to retain such hermeneutics with this market inscription as well, as it is the “brother of Yeshua” component (written in a different hand) that arguably makes this ossuary financially valuable. I should note that I have heard it suggested that “there is no logical reason for someone to add ‘brother of Yeshua’ to this inscription.” However, I can think of a million (financial) reasons for a forger to have done so. Of course, someone might retort that this assumes the modern forger would have *known* that this would make it valuable. I would think a forger would certainly know this. Someone might further retort that Golan did not know this reading until he was told and so would not have added it. I would suggest that any forger worth his salt would *want* epigraphers to *believe* that he did *not* know how significant this addition might be. It is a very savvy technique.


The Geological Survey of Israel performed SEM-EDS analyses on the patina of the “Ya‘akov Ossuary” (Rosenfeld and Ilani 2002). The results showed that the patina is composed mainly of CaC03 (93%) and contains Si – 5.0%; A1 – 0.7%; Fe – 0.3%; P – 0.4%; and Mg – 0.2%. The report notes that there are no modern elements (such as modern pigments), and the patina adheres firmly to the stone. Again, this is valuable information, but it is imperative that one not conclude or assume, on the basis of this evidence, that this entire inscription is ancient. Rather, this test simply demonstrates that this object need not be disqualified on the basis of anomalies in the chemical composition of the patina. It certainly does not authenticate the patina. Indeed, the report implicitly concedes this point with the words: “no evidence that might detract from the authenticity of the patina and the inscription was found.” This is an important and honest assessment; however, it must be noted in the strongest possible terms that the absence of certain anomalies in the chemical composition of a patina is not the same as a demonstration of the antiquity of a patina. Note, however, that subsequent laboratory tests were performed and problems with the distribution of the patina were noted (Ayalon, Bar-Matthews, Goren 2004).


Forgers continue to produce some very sophisticated forgeries. These are sold on the antiquities market. High quality forgeries can and do sell for tens of thousands of dollars, sometimes even hundreds of thousands of dollars. The James Ossuary was not found on a scientific expedition, but rather appeared mysteriously and was sold on the antiquities market. The chain of custody for this ossuary and its inscription cannot be known. The second half of the inscription on this ossuary (i.e., the part that says “brother of Jesus”) is the part that would potentially make this ossuary worth some money. Dramatic claims required decisive evidence. In the case of the James Ossuary, we simply do not have that caliber of evidence. Indeed, I consider (at least) the second half of the James Ossuary Inscription to be a probable modern forgery, not an inscription that can be said to be an ancient inscription connected with the brother of Jesus of Nazareth. I wish that this inscription could be said to be entirely ancient and I wish that it could be said that it is certainly to be connected with the brother of Jesus of Nazareth. But the evidence supporting this conclusion is simply too tenuous, alas.


Ayalon, A., Bar-Matthews, M., Goren, Y. “James Ossuary. Journal of Archaeological Science 31 (2004): 1185-1189).

Ehrman, B. Forgery and Counter-Forgery: The Use of Literary Deceit in Early Christian Polemics. Oxford University Press, 2012

Rollston, C. “Non-Provenanced Epigraphs I: Pillaged Antiquities, Northwest Semitic Forgeries, and Protocols for Laboratory Tests.” Maarav 10 (2003): 135-193.

Rollston, C. “Non-Provenanced Epigraphs II: The Status of Non-Provenanced Epigraphs within the Broader Corpus of Northwest Semitic.” Maarav 11 (2004): 57-79.

Rollston, C. “Navigating the Epigraphic Storm: A Palaeographer Reflects on Inscriptions from the Market.” Near Eastern Archaeology 68 (2005): 69-72.

Rollston, C. “Forging History: From Antiquity to the Modern Period.” Pp. 176-197 in Archaeologies of Text: Archaeology, Technology, and Ethics, eds. Matthew Rutz and Morag Kersel. Joukowsky Institute Publication Series of Brown University, Oxbow Books, 2014.

Rosenfeld, A. Ilani, S. “SEM-EDS Analyses of Patina Samples,” BARev 28 (Nov/Dec 2002).

Vaughn, A. and Rollston, C. “The Antiquities Market, Sensationalized Textual Data, and Modern Forgeries.” Written with Andrew Vaughn. Near Eastern Archaeology 68 (2005): 61-69.

Review of The Jesus Discovery: The New Archaeological Find that Reveals the Birth of Christianity, by James D. Tabor and Simcha Jacobovici (New York: Simon and Schuster, 2012).

12 April 2012


Alas, Tabor and Jacobovici’s monograph is replete with a superabundance of problems. Within this review, I will confine my discussion just to (some of) the major problems. Nevertheless, this review is quite long, something demanded by the problematic content of the volume itself. For this reason, readers may wish to scan for highlighted headings within the review so as to find the areas of critique in which they have the most interest. I should like to mention that much of the content of this review was first presented on the Official Blog of the American Schools of Oriental Research. Finally, I should like to emphasize that although I have attempted to document the secondary references quite well, it was not feasible to refer to all of the contributions that all scholars have made. Nevertheless, I have attempted to refer to as many of the primary contributors as possible.


Here are the basic claims of Tabor and Jacobovici. “Talpiyot Tomb B” contained several ossuaries, or bone boxes, one of which had some interesting ornamentation, one of which had a four-line Greek inscription, and one of which had the Aramaic word “mara” written in Greek letters. Taken together, Tabor and Jacobovici argue that the ornamentation and the inscription constitute “the earliest archaeological evidence of faith in Jesus’ resurrection.” They go on and state that these ossuaries “also provide the first evidence in Jerusalem of the people who would later be called ‘Christians.’ In fact, it is possible, maybe even likely, that whoever was buried in this tomb knew Jesus and heard him preach.”

In addition, Tabor and Jacobovici claim that because “Talpiyot Tomb B” is within around two hundred feet from “Talpiyot Tomb A” (the tomb Tabor and Jacobovici have also dubbed the ‘Jesus Family Tomb’) “the new discovery [i.e., “Talpiyot Tomb B”] increases the likelihood that the ‘Jesus Family Tomb’ is, indeed, the real tomb of Jesus of Nazareth.” Tabor and Jacobovici also believe that “Jesus of Nazareth was married to Mary Magdalene and had a son named Judah,” something which they have been proposing for several years now. Tabor and Jacobovici also assume that “both tombs appear to have been part of the property of a wealthy individual possibly Joseph of Arimathea, the man who, according to the gospels, buried Jesus.”

At this juncture, I shall turn to a fairly detailed discussion of both tombs and the contents thereof. Suffice it to say that I am confident most scholars will not consider the grand claims of Tabor and Jacobovici (as enshrined in the title of the book) to be cogent. The reason is quite elementary: the conclusions they draw do not follow from the extant evidence.


This tomb was discovered in 1980 by Yosef Gath during a salvage excavation at a site in the neighborhood of East Talpiot, Jerusalem. I shall refer to it as Talpiyot Tomb 1980 because of the year of its discovery. It contained ten ossuaries, six of them inscribed. These were subsequently published in Rahmani’s A Catalogue of Jewish Inscriptions (1994, nos 701-709). The personal names on the ossuaries of this tomb are as follows: (1) Mariamē kai Mara (Mariam and Mara). (2) Yhwdh br Yšw‘ (Yehudah bar Yeshua‘). (3) Mtyh (Mattiyah). (4) Yšw‘ br Yhwsp (Yeshua‘ bar Yehosep). (5) Ywsh (Yoseh). (6) Mryh (Maryah). The names Yehosep, Yoseh, Yeshua‘, Yehudah, Mattiyah, Maryah, Maryam, Mariamne, Mara and Martha (or the variants thereof) all have multiple attestations in the multilingual corpus of ossuaries and some are very common (Rahmani 1994, 292-297; Ilan 2002). In fact, even the name and patronymic “Yeshua‘ bar Yehosep” (i.e., “Jesus son of Joseph”) is not unique in the epigraphic corpus. After all, some eighty years ago, Sukenik published an ossuary inscribed “Yeshua‘ son of Yehosep” (“Jesus son of Joseph”) and the names Yeshua‘ and Yehosep (“Jesus” and “Joseph”) are predominant in the family of Babatha’s first husband. In fact, the father of Babatha’s first husband was named Yeshua‘ and his father was named “Yehosep,” so this is yet another “Yeshua‘ son of Yehosep” (i.e., “Jesus son of Joseph”; see Sukenik 1931; Lewis 1989, 35-40; cf. Yadin 1971, 233-234; Kraeling 1946, 18-19). Thus, even with the small corpus of epigraphic attestations of personal names, the Talpiyot Tomb 1980 occurrence of “Yeshua‘ bar Yehosep” (“Jesus son of Joseph”) is not even unique.

It is certainly true that filmmaker Simcha Jacobovici, along with Charles Pellegrino, James Tabor, and Andrew Feuerverger attempted to argue that this was the family tomb of Jesus of Nazareth (Jacobovici and Pellegrino 2007; Tabor 2006; Feuerverger 2007). But the epigraphic evidence (such as personal names) from this tomb does not support their contention, neither does the DNA evidence, nor does the statistical evidence. Indeed, a cross-section of scholars (including Eric Meyers, Shimon Gibson, Jodi Magness, Sandra Scham, and I) wrote articles several years ago in the academic journal Near Eastern Archaeology (published by the American Schools of Oriental Research) arguing that the cumulative evidence certainly did not support the view that this Talpiyot tomb was the tomb of Jesus of Nazareth or his family.

In addition, I should also emphasize that Tabor and Jacobovici’s desire to state that the Ya‘akov Ossuary (often called the “James Ossuary,” especially by those who wish to say this ossuary was not tampered with epigraphically and that it is Christian) came from Talpiyot Tomb 1980 is speculative (and I would also note in this connection that the patina of stone ossuaries from the same quarry which were housed in the same basic environment in a Jerusalem tomb will, of course, share certain many chemical features…so even patina evidence is of no great value. I will be happy to talk more about this later, should the need arise).

It is important to remember this dictum: Dramatic claims require dramatic evidence. And ultimately the strong consensus of scholars working in the fields of ancient epigraphy, archaeology, and ancient religion was then, and is now, that Talpiyot Tomb 1980 is not that of the family of Jesus of Nazareth. That is, the dramatic claims of Tabor and Jacobovici could not be embraced previously (i.e., 2007) because the evidence simply was not there. I am happy to resurrect this discussion, but the claims of Tabor and Jadobovici for this tomb are no more convincing now than they were then.


During course of construction work in Jerusalem during the spring of 1981, a tomb with nine kokhim (“burial shafts”) was discovered. There were a total of eight ossuaries in this tomb (originally distributed in four of the kokhim, that is, “carved chambers”), one of which was removed in 1981 (one belonging to a small child or infant). It was noticed then (in 1981) that there were some Greek inscriptions on (at least) two of the ossuaries, but the tomb was not excavated and documented thoroughly because of various exigencies, including religious sensitivities. Ultimately, modern buildings were soon erected at this site. However, rather than destroying this tomb, the modern buildings were built above the tomb. Tabor and Jacobovici have dubbed this tomb “Talpiyot Tomb B,” but I shall refer to it as Talpiyot Tomb 1981, based on the year of its discovery.

During the course of a few days in 2010, James Tabor, Rami Arav, and Simcha Jacobovici (now the primary researchers for this tomb, which ) were able to send a robotic camera into this tomb (through the basement floor of the building which had been built on top of the tomb) and to photograph the tomb itself, the ossuaries in it, and some inscriptional remains. One of these inscriptions, consisting of four very brief lines, has garnered substantial attention, as has some of the ornamentation (which Tabor and Jacobovici refer to as “iconography,” a term that conjures up in the minds of many readers something which is quite “Christian”) on one of the other ossuaries. Indeed, Tabor and Jacobovici have claimed that this four-line inscription on one ossuary, and the ornamentation on another can be understood as referring to a belief in some sort of resurrection and that this inscription and ornamentation are, therefore, Christian. They have also noted that another of the ossuaries in Talpiyot Tomb 1981 has the word “mara” on it, an Aramaic word normally meaning “sir,” or “master” or even “husband” (although it is written in Greek letters in this tomb, as is often the case in epigraphic materials from this region). Frankly, I would find it very interesting if this were a Jewish-Christian tomb, but the evidence simply does not support this view. At this juncture, the focus will turn to historical and epigraphic consideration of the salient aspects of the finds in Talpiyot Tombs 1980 and 1981.


At this juncture, I shall turn to a discussion of the salient aspects of the evidence and a critique of the conclusions of Tabor and Jacobovici. I shall attempt to be as thorough as possible and to refer to some of the most relevant primary and secondary literature.


The Discovery Channel Documentary makes the assertion that the data from Talpiyot Tomb 1981 constitute the earliest attested reference to a resurrection (note the way that Tabor and Jacobovici read, translate and interpret the four-line Greek inscription, discussed below in section D). That is a sensational assertion and it is not accurate. The fact of the matter is that within various segments of Late Second Temple Judaism, the notion of a resurrection was warmly embraced, and many of these statements antedate the rise of Christianity. The locus classicus in the Hebrew Bible is arguably the following text from the mid-2nd century BCE: “Many of those sleeping in the dust of the earth shall awaken, some to everlasting life and some to everlasting peril” (Dan 12:2). Within the Old Testament Apocrypha, the notion of a resurrection is embraced at times as well, with the narrative about the martyrdom of “the mother and her seven sons” being a fine exemplar of this. Thus, according to the narrative, one of the sons said during the torture that preceded his death: “the King of the universe will raise us up to an everlasting renewal of life, because we have died for his laws” (2 Macc 7:9). Similarly, the mother herself says within the narrative, as an exhortation to her martyred sons: “the Creator of the world…will in his mercy give life and breath back to you again” (2 Macc 7:23). 2 Maccabees arguably hails from the first half of the 1st century BCE. Regarding the dead, the Wisdom of Solomon also affirms that the dead “seemed to have died,” but “they are at peace,” and “their hope is full of immortality,” and they will ultimately “shine forth” and “will govern nations and ruler over peoples” (Wisdom 3:2-8 passim, with the Greek future tense being used here). The Wisdom of Solomon arguably hails from the second half of the 1st century BCE. Significantly, all of these texts antedate the rise of Christianity and they all affirm a belief in a resurrection. In short, many Jews believed in a resurrection long before Christianity came along. To be sure, a belief in a resurrection was not universally accepted by Jews in the Second Temple period. Some Jews did not believe in a resurrection. For example, the traditionalist Ben Sira rejected the notion of eternal bliss for the righteous and eternal punishment for the wicked. Thus, he wrote: “Who in the netherworld can glorify the Most High, in place of the living who offer their praise? No more can the dead give praise than those who have never lived; they glorify the Lord who are alive and well” (Sir 17:27-28). In sum, although not all Jews of the Late Second Temple period accepted the notion of a resurrection, there are texts from this period that demonstrate that a fair number did.

Furthermore, the Jewish historian Josephus (lived ca. 37-100 CE) also discusses the subject of the perishability and imperishability of the soul, with regard to some of the major strands of Judaism during the first century of the Common Era. Regarding the Pharisees, therefore, he states that they believe “every soul is imperishable, but the soul of the good alone passes into another body, while the souls of the wicked suffer eternal punishment.” Conversely, regarding the Sadducees he states that “as for the persistence of the soul after death, penalties in the underworld, and rewards; they will have none of them.” Regarding the Essenes, Josephus states that they believe “the body is corruptible and its constituent matter impermanent, but that the soul is immortal and imperishable…sharing the belief of the sons of Greece, they maintain that for virtuous souls there is reserved an abode beyond the ocean, a place which is not oppressed by rain or snow or heat, but is refreshed by the ever gentle breath of the west wind coming in from ocean, while they relegate base souls to a murky and tempestuous dungeon, big with never-ending punishments” (Josephus, Jewish War, II, 11-14; for more discussion, see Nickelsburg 1972, 164-169). Of course, pericopes within the Greek New Testament regarding the Pharisees and Sadducees dovetail nicely with Josephus. The locus classicus for the New Testament is arguably contained within the book of Acts: “The Sadducees say that there is no resurrection, or angel, or spirit; but the Pharisees acknowledge all three” (Acts 23:8; cf. also Matt 22:23).

Naturally, scholars of the Hebrew Bible, Second Temple Judaism, the Greek New Testament, and Early Christianity have for a very long time dealt with these ancient assumptions about the afterlife. Moreover, based on the convergence of the evidence (such as the texts cited above), the consensus of the field has long been that some Jews within the Late Second Temple period embraced a belief in a resurrection and some did not (e.g., DiLella 1966; Collins 1998; Ehrman 1999). To be sure, Christianity (originally a sect of Judaism, with strong apocalyptic tendencies) did embrace a notion of a resurrection, and this is very clear from the documents of the Greek New Testament. But the fact remains that many Jews of the late Second Temple Period believed in a resurrection, not just Jewish Christians. Thus, even if the inscription or ornamental motifs of this tomb provided evidence for a belief in a resurrection, one cannot assume that this must have been a Christian (i.e., Jewish-Christian) tomb. After all, many non-Christian Jews accepted the idea of a resurrection. Again, even if something in Talpiyot Tomb 1980 or Talpiyot Tomb 1981 could be construed as referring to a resurrection, it does not follows that this tomb must have been a “Christian” tomb.


Tabor and Jacobovici have posited that Talpiyot Tomb 1981 is a tomb belonging to Joseph of Arimathea (i.e., the “Joseph of Arimathea” mentioned in the canonical gospels), and that this tomb also contains the actual ossuary of Joseph of Arimathea himself. Here are some citations of Tabor and Jacobovici’s views: Talpiyot Tombs 1980 and 1981“are most likely located on the rural estate of Joseph of Arimathea, a wealthy member of the Sanhedrin who according to all four New Testament gospels took official charge of Jesus’ burial” (Tabor and Jacobovici, 2). But he is framed as wealthy and so they believe they have to account for the modest nature of this ossuary, thus, they suggest that there may have been “something about his faith or piety as part of the Jesus movement” that led him to “prefer such a modest bone box” (Tabor and Jacobovici, 89). Then they conclude that “it is not hard or even overly speculative for us to posit that the Talpiyot Tombs are a tiny but amazing glimpse into the life of Joseph of Arimathea” (Tabor and Jacobovici, 128).

The ossuary in Talpiyot Tomb 1981 which they consider to be that of Joseph of Arimathea is one they also refer to as a “humble ossuary” (Tabor and Jacobovici, 89). Along the same lines, they query: “might Joseph of Arimathea have chosen a…modest ossuary for himself and his most immediate family—but one that boldly proclaimed their faith even in the midst of opposition and conflict?” (Tabor and Jacobovici, 90). It should be noted that the reason they refer to this ossuary as “boldly proclaiming their faith” is because the ossuary they believe to be that of “Joseph of Arimathea” is the one with the ornamentation they understand to be “Jonah and the Whale.” As will be discussed below, the consensus of scholarship is that this ornamentation is that of an amphora or some sort of architectural feature, not “Jonah and the Whale.” In any case, the main point is this: there is not a single inscription from Talpiyot Tomb 1980 or Talpiyot Tomb 1981 that suggests that these tombs are to be associated with Joseph of Arimathea. I would propose that for a historian to make a credible argument that this is the land, tomb, and ossuary of Joseph of Arimathea there must be solid evidence, such as the name “Joseph of Arimathea” inscribed on the ossuary. But, since these words are not there, it is really not convincing to posit that this is the tomb of Joseph of Arimathea. Tabor and Jacobovici may believe that it is not “hard or overly speculative” to say that this is the land, tomb, and ossuary of Joseph of Arimathea, but I think most epigraphers, prosopographers, and historians would find it to be quite speculative.


Introduction to the Issue: The Aramaic word mara (written in the Greek script in these Talpiyot tombs, rather than in the Aramaic script) occurs on an ossuary in Talpiyot Tomb 1980, namely, in the phrase Mariamē kai Mara (i.e., Mariamē and Mara; see Rahmani 1994, # 701). Tabor and Jacobovici assume that the inscription on this ossuary should be understood as referring to one person and so they render it “Mariam called Mara” (Tabor and Jacobovici 2012, 28). They refer to Rahmani’s reading in the editio princeps of this inscription, namely, Mariamēnou Mara, which Rahmani translated “Mariamene, who is (also called) Mara.” Rahmani had originally stated that he believed this name was “in the genitive case” and was “a diminutive of Mariamēne” (Rahmani 1994, 222). Significantly, Stephen Pfann has published a most cogent correction of Rahmani’s reading, noting that there are two words and a very clear kai between them (which Rahmani had unfortunately initially misread). Hence, Pfann renders this ossuary inscription as “Mariame and Mara” (Pfann 2006). As I mentioned above, however, Rahmani has now accepted the corrected reading, that is, Mariame kai Mara. Tabor and Jacobovici do not, however, accept the corrected reading. In any case, in terms of additional occurrences of mara, Tabor and Jacobovici note that mara also occurs on an ossuary in Talpiyot Tomb 1981 (Tabor and Jacobovici, 67).

Here are the statements by Tabor and Jacobovici regarding the Aramaic word mara: “it is the feminine form of Mar, which in Aramac means ‘Lord’ or ‘Master’” (Tabor and Jacobovici, 46). Or again, “Mara is the feminine form of Mar in Aramaic, which means ‘Lord’ or ‘Master,’ as explained in the previous chapter” (Tabor and Jacobovici, 67). They state that “we are convinced that Mara is an honorific title, not a proper name.” They also state that “if you add the feminine ending to Mar you get Mara” (Tabor and Jacobovici, 115). The footnote accompanying that statement is: “The Aramaic name Marta (Martha) is derived from Mar/Mara. Some argued that Mara is just an alternative form of Martha but as we explain chapter 5, such is not the case” (Tabor and Jacobovici, 221). Again they state that “Mara, which comes from the Aramaic masculine Mar, is the absolute feminine, whereas Martha (Martha) is the emphatic feminine. They both come from the same masculine noun and mean the same thing, but Martha evolved into a name and is common (Tabor and Jacobovici, 227; cf also Tabor 2012, 13-14). Of course, it should be mentioned that they also contend that the ossuary with the words Mariame kai Mara (which they believe should be read Mariamēnou Mara) should be understood as the ossuary of Mary Magdalene and that the title Mara is a title that “can potentially refer to her place of leadership and authority in the emerging Christian movement” (Tabor and Jacobovici, 131; cf. 96). It is striking, of course, that the word “Magdala” does not occur in these Talpiyot inscriptions, a weakness of their argument that she is the referent in Talpiyot 1980 (i.e., Rahmani 1994, #701). In any case, as is apparent, these statements from Tabor and Jacobovici assume strongly that mara is a definitive feminine form. There is no discussion of the fact that mara is also a very fine masculine form of this Aramaic word (especially the determined state, but even in the absolute state).

Here are the basic philological data, however. The word mr’ (Mara’) is an Aramaic, masculine, singular noun meaning “sir,” “master,” “lord.” It is well attested (as a masculine noun) in the Aramaic corpus of Northwest Semitic inscriptions, in both Old Aramaic and also in Imperial Aramaic (sometimes with the spelling mry). Note that in the case of the Old Aramaic occurrence in Tell Fakhariyeh (e.g., line eight) the Akkadian text of this Akkadian-Aramaic bilingual uses (the Sumerian logogram to indicate that the Akkadian word should be understood as) bēlum, obviously a masculine form, not a feminine. This word even occurs in Nabataean and Palmyrene (which are later dialects of Aramaic), with the masculine form spelled mr’. The feminine singular is attested in Imperial Aramaic as mr’t, and the feminine singular determined form occurs as mr’t’ (Hoftijzer and Jongeling 1995, 682-689). The masculine form of this word also occurs in the Aramaic of the Hebrew Bible, with the spellings mr’ and mry (see Dan 2:47; 4:16, 21; 5:23; Koehler and Baumgartner 2000, 1921-1922). Moreover, it also occurs in Jewish-Palestinian Aramaic with the spelling mr and mr’. The feminine form of this Aramaic word occurs in Jastrow as well and is martha’ (see Jastrow 1950, 834-835, s.v., Mar IV). It is often stated that (for some of the Late Second Temple occurrences) the word mara can sometimes be a shortened version of the word martha’, and thus can sometimes refer to a woman (either as a personal name, or as a title meaning ‘lordess’ or the like). Thus, Tal Ilan states about the name mr’ (also spelled mrh during the Second Temple period) that “this is one of the rare cases of a name serving for both males and females” (Tal Ilan 2002, 392; cf. also 423-424).

The point that I would emphasize is this: although the name mara (to use the Greek spelling) might sometimes be used as a shortened form of the name or title martha’, the fact remains that it is not methodologically permissible to assume that mara is always a feminine (i.e., a shortened form of martha’). After all, as discussed above, the form mr’ is most readily understood as an Aramaic masculine (cf. also Rahmani 1994, #561). Thus, any historical construct built on the assumption that mara is definitively feminine must be considered a tenuous case indeed. For this reason, I find it to be quite problematic that Tabor and Jacobovici assume that the occurrences of mara on these ossuaries must be feminine. The philological evidence demonstrates decisively that mara can readily be a masculine form and so this certainly merited a discussion byTabor and Jacobovici in these recent publications.

In short, it is plausible to contend that in Talpiyot Tomb 1981, the word mara refers to a man, not a woman. Also, with regard to Talpiyot Tomb 1980, I would suggest that it is entirely plausible to suggest that this is the ossuary of a woman and a man, that is, a woman named mariame and a man known as mara. Someone might suggest that the woman’s name would not come first in this culture. However, I would note that order of death could reasonably account for the ordering of the names. Moreover, we do sometimes find a woman’s name first in literary texts that refer to a woman and a man (e.g., Acts 18:18; 18:26; Romans 16:3; 2 Tim 4:19). In short, it is philologically and historically plausible to suggest (A) that the persons referred to as mara in these two Talpiyot tombs were men, not women; (B) and it is philologically and historically plausible to suggest that one was a man and one was a woman; (C) and it is also permissible to suggest that both were women. But it is imperative that we be candid about the fact that we actually do not know and so it is precarious to assume. The fact that Tabor and Jacobovici simply assumed that the term mara was feminine and even omitted the philological evidence for it as a masculine is particularly unfortunate, and probably quite telling. After all, they wished for this term to be considered feminine because it was important for the edifice they had erected.


There has been some substantial discussion about the four-line inscription, its readings and its renderings. The purpose of this post is to delineate the history of published proposals, summarizing salient points. Most of my posts on this find have been posted on the Official site of the American Schools of Oriental Research. For this reason, and because of subsequent discussion, I am now posting this summation here.

On February 28, 2012, James Tabor’s reading and translation of the four-line inscription was released. Namely, he and Simcha Jacobovici read it as follows: “DIOS IAIO UPSŌ AGB.” They translated their readings as “Divine Jehovah Lift up! Lift up!” Again, they believe Talpiyot Tomb 1981 to be a Christian tomb (in fact, they state that it is arguably the tomb of Joseph of Arimathea himself, although there is no ancient epigraphic evidence to suggest this) and they suggest that this four-line inscription is to be understood as reflective of an early Christian confession of a belief in the resurrection (and they have also argued that some of the ornamentation on a different ossuary from the same tomb is distinctively Christian, something that has been widely criticized as well). Also rather striking (and quite difficult to sustain), Tabor and Jacobovici have argued that the graphemes AGB (line 4) should be understood as the Greek transliteration of an H-stem verbal root gbh, although they had also mentioned (and dismissed) considering it to be a Semitic personal name transliterated into Greek graphemes, namely, “Agabus” (Tabor and Jacobovici, 90-94).

On February 28, 2012, around 1:00 p.m., a statement of mine was posted on the Official Blog of the American Schools of Oriental Research in which I discussed various aspects of Talpiyot Tombs 1980 and 1981 (Rollston 2012a). Among other things, I stated that: “Regarding the reading of line two, I wish to emphasize that I do not consider the reading “Yahweh” (i.e., the Greek form of it) to be convincing at all. Simply put, this reading is wrong. To be sure, the tetragrammaton is attested in ancient Greek (with various spellings) and Iaio can be considered a viable Greek spelling of the tetragrammaton. However, the problem is that the first grapheme of line two is not an iota (and, at the very least, this grapheme would be necessary for reading the tetragrammaton in this line). Of course, Tabor and Jacobovici believe that the first grapheme of this line is an iota, and they are obviously assuming that this grapheme consists of a distinct and deeply incised top horizontal, a bottom horizontal, and a long vertical connector. There is, however, a palaeographic problem with this reading. Here is the reason: for the Greek script(s) of the Late Second Temple period, the morphology of iota is quite consistently a vertical stroke (sometimes with modest curvature), but without distinct top or bottom horizontals. This is the case for Greek texts on soft media (e.g., papyri) and on hard media (e.g., stone). The panoramic Greek script charts of the great Princeton palaeographer Bruce Metzger are reflective of this (e.g., Metzger 1981, 23, figure 2). For further demonstration of this aspect of the morphology of this grapheme, readers might also wish to consult photos of the Greek textual material from this chronological horizon on soft media (e.g., the Greek papyri from the Bar Kokhba Cave of Letters; See Lewis 1989, passim ) and on hard media (e.g., Jerusalem Ossuary inscriptions; see CIIP 1. #64, 65, #98, #134, #189, 199, etc.). I would suggest that the convergence of the cumulative evidence demonstrates in a cogent manner that the first grapheme is simply not an iota. In reality, this grapheme is most readily understood as a tau (i.e., a top horizontal and a vertical) or (alternatively) a zeta. However, it is certainly not an iota. Of course, since there is no iota here, there is no tetragrammaton” (Rollston 2012a).

I did not provide all of my readings at that time, nor did I go into further detail about the palaeography, as my statement on February 28 was certainly not intended to be an editio princeps of that inscription. However, I did state in that initial article that I was “most comfortable with reading the last two graphemes of line one and the first two graphemes of line two as ‘osta,” that is, ‘bones,” a word that certainly does occur in a number of ossuary inscriptions and burial texts. Further, if one were to wish to read hupsō, I would then be inclined to understand this inscription to be stating that the bones of the deceased are not to be removed, that is, ‘lifted up’ from the ossuary” (Rollston 2012a).

On March 8th, 2012, Richard Bauckham responded with an article on the blog of the American Schools of Oriental Research. Bauckham stated that he believed “the inscription is actually very clear.” He also went on to indicate that he accepted all of Tabor’s readings (i.e., the Greek graphemes Tabor believes are present), but he translated the inscription as follows: “Belonging to Zeus IAIO. I, Hagab, exalt (him/you).” In addition, he states that “It is the only ossuary inscription to mention God in any way, let alone to use the divine name.” He also states that “as far as I know, our inscription is the only extant example of an identification of Yahweh with Zeus in a Palestinian Jewish context after the Maccabean period” (Bauckham 2012a). Bauckham also quoted my statement that the first grapheme of line two is definitely not an iota and then said of himself that although he “is not an epigrapher” (these are his words, not mine), he would “venture to say that he [Rollston] is being far too dogmatic.” He then went on to refer to a few examples and said they had “very distinct top and bottom horizontals.” However, it is significant that he goes on to use the term “serifs” (i.e., “apices”) for these strokes. Indeed, (in the examples he cites from CIIP) they are serifs, that is, lightly incised strokes, not full-blown deeply incised strokes. That is, they are not something that an epigrapher would normally consider to be good parallels for the full blown, deeply incised stroke that is at the top of the grapheme in question (i.e., the first grapheme of line two of the Talpiyot inscription). There was another problem as well, however, and Bauckham sensed it: “It does need to be explained why, in our inscription, only this letter is adorned with apices” (i.e., serifs). He’s certainly correct in deducing a serious problem with his view. Namely, the only grapheme in this four-line inscription from Talpiyot with serifs (i.e., what he understands to be serifs) is his and Tabor’s iota at the beginning of line two! After all, in the Greek epigraphic corpus from this period and horizon, when serifs are present, they are normally present on multiple graphemes (see the images I posted on the ASOR web site, Rollston 2012b). Here is Bauckham’s way of accounting for it: “The most likely reason seems to be that it is the first letter of the divine Name, which the writer wished to mark out as special.” He then goes on to state that “It is his equivalent of the various other ways of distinguishing the divine Name when it was written in Hebrew or Greek biblical manuscripts or elsewhere (such as the common practice among Qumran scribes of writing the Name in paleo-Hebrew chaacters” (Bauckham 2012a).

On March 15, 2012, in response to a number of requests, I wrote an article with my readings and some proposed translations (Rollston 2012b), as well as with palaeographic matters Bauckham had decided to discuss, especially the iota (and thus the tetrgrammaton) in line two of the Talpiyot inscription. Here is the essence of my epigraphic reply: “(1) I would note, however, that these inscriptions [the very ones to which Bauckham had referred] have serifs on multiple graphemes and just one, as the Talpiyot inscription allegedly does. (2) Furthermore, I would note that on most inscriptions with serifs, the serifs are not nearly as deeply incised as is (for example) the top horizontal of the Talpiyot grapheme Tabor and Bauckham consider to be an iota with a serifs. That is, the top horizontal of that grapheme does not have the appearance of a serif, but rather a full blown, deeply incised stroke. Bauckham senses the first problem and states that “it does need to be explained why, in our inscription, only this letter is adorned with apices [i.e., serifs].” He then states that “the most likely reason seems to be that it is the first letter of the divine Name, which the writer wished to mark out as special.” He goes on to suggest that this is similar to the way the divine name is treated in some Hebrew and Greek biblical manuscripts. He refers in particular to Qumran practice (Bauckham 2012a). However, I would note that the practice at Qumran is quite dissimilar. At Qumran, Emanuel Tov states that “divine names were written in a special way in many Hebrew Qumran texts” and then he provides the following synopsis: (A) All four graphemes of the tetragrammaton are written in Paleo-Hebrew characters in texts which are written in the square script; (B) Four dots in texts written in the square script; (C) A dicolon (:), followed by a space, placed before the Tetragrammaton (written in the square script); (D) the use of a different color of ink, in the case of 11Q22 (Tov 2004, 219-220, et passim; see also Tov 2001). In other words, there are no cases of the initial grapheme formed in a distinct way, but the remaining graphemes of the Tetragrammaton written in the standard (i.e., non-paleo-Hebrew) script. It is worth noting in this connection that Larry Hurtado has done a great deal of work on the Nomina Sacra in early Christian Greek manuscripts, but even in these manuscripts, there is nothing that parallels the sort of thing that Bauckham is proposing here (Hurtado 2006, 95-134; see also Metzger 1981, 36-37).

In addition to discussing these epigraphic factors, I provided all of my readings then (Rollston 2012b, i.e., March 15), using some of the photographs National Geographic provided me with in May 2011 and those published on the web (NB: I had served as the Epigraphic Consultant for National Geographic on this find for several months). Namely, I reiterated my reading of tau for the first grapheme of line two. Here are my words and I would draw the reader’s attention again to the images in my article, which are posted on ASOR’s Official Blog (Rollston 2012b): “I would ask the reader to look carefully at the image labeled Talpiyot 1, the second line, the first grapheme. At the bottom of the vertical of this grapheme is a pit in the stone (right next to the left oblique stroke of the alpha). I would ask the reader also to look at a different photograph, with a different light angle, namely, the image labeled Talpiyot 2. It is clear from this image that there is no horizontal stroke on the left side. Rather, there is a downward scratch (in fact, it may be that the person inscribing this ossuary made this mark when he was forming the upper part of the head of the upsilon, although it could have happened at almost any time). In any case, the point is that the “marks” Tabor and Bauckham considered the bottom horizontal of an iota are just pitting and scratches. Frankly, this sort of thing is very common in the field of epigraphy. The end result, of course, is that a recognition of the pitting and scratching yields a perfect tau. I should also make an additional notation regarding this line, namely, the grapheme Tabor and Bauckham consider to be the second iota. I draw the reader’s attention again to the image labeled Talpiyot 1, the second line and the third grapheme. It is a very clear epsilon, not an iota.”

Then I said: “Astute readers will have noticed, at this juncture, that the word osta “bones” can now be read (the last two graphemes of line one and the first two graphemes of line two). The normal spelling of this word in the plural is ostea, although the spelling osta is also well attested in the Greek corpus. In this case we have, I believe, either a dialectical variant in the pronunciation of this word (causing it to be spelled ostae, rather than ostea), an actual orthographic variant, or a simple orthographic error (all three of these things occurs in the corpus of ancient funerary inscriptions). In any case, reading “bones” in a funerary context is quite compelling. Moreover, the final grapheme of line two is an omicron and the first grapheme of the following line (line three) is an upsilon. This is, I believe, simply the negative, a lexeme that occurs rather frequently in tomb contexts when there are references to bones and ossuaries.”

I should mention in this connection that within that article of March 15, I discussed in some detail the sorts of statements that we find in Jewish burial contexts from the Late Second Temple and Early Post-Biblical chronological period. Of course, suffice it to say that words such as “bones” and “ossuary” are well attested.

I went on to note that in terms of readings, I would posit the following: DE OSTAE OU PSŌ AGB . Understanding the verbal to be psaō, I stated that “I would posit that it is reasonable to render this inscription: “Here are bones. I touch (them) not. Agabus. “ As such “Agabus” could be the name of the deceased, and thus this could be translated “Here are bones. I touch them not, O Agabus.” Conversely, it could also be that the first person singular is used here of the man who asserts that he does not touch bones. Thus, this could then be translated quite nicely as “Here are bones: I, Agabus, touch (them) not.” I suggested that the intransitive meaning is also viable. Thus, something such as “Here are (my) bones. I, Agabus, crumble not away.”

At that time, and now still, I also consider it possible to read the verb upsoō here (as I mentioned also in Rollston 2012a). In this case, it would read something along these lines: “Here are the bones. I lift not (the bones/ossuary), O Agabus,” or “Here are the bones. I, Agabus, lift (the bones/ossuary) not. I should note in this connection that I consider the proposal astutely suggested by Bauckham, namely, the presence of the personal name “Agabus” (in line four of the Talpiyot inscription) to be satisfying. In any case, the point is that the content of this inscription falls within the traditional sorts of statements that occur in Late Second Temple and Early Post-Biblical tomb contexts….it’s all about bones.

After reading my post, Robert Hull, a colleague of mine, suggested to me that rather than reading a form of ōde at the beginning of this inscription (as I did), he would prefer to read di, that is, a standard shortened form of the preposition dia. I suggested to him that the second grapheme of this inscription was abraded enough that I believed an iota to be a plausible reading for line one, grapheme two, that is, the short form of the preposition was something I considered viable. At my suggestion, he posted his proposal on ASOR’s blog as a comment to my article. Using his understanding of di, the rest of my readings, and one of my proposed possible renderings of the verb psaō, Hull proposed the following: “On account of [the] bones, I, Agabus, do not touch.” At that juncture, Bauckham, using my proposal of an intransitive meaning for psaō and all of my readings (but with Hull’s suggestion of di for dia),suggested (in the comment section of my March 15 ASOR blog post) the following: “Because of (these) bones, I, Hagab, am not crumbling away (disappearing).” He has also argued that he does not believe purity concerns should enter in to the discussion (i.e., he does not think them relevant). I have suggested that I think it is possible that purity or taboos might be part of the equation (i.e., as a possible rationale for the possible translation “I do not touch”).

Obviously, I am pleased with, and comfortable with, these suggested translations of my readings of the inscription. And, of course, I’m certainly comfortable with Hull’s understanding of a short form of di in line one. To be sure, though, I would not be surprised to find entha somewhere on this ossuary (which would then be joined with my initial reading of de at the beginning of line one, thus forming the very commonly attested beginning of funerary inscriptions, namely, enthade “here”). After all, on ossuaries, words, or portions thereof, sometimes begin on one part of an ossuary and then continue on a different part of the ossuary.

In short, in terms of readings for this very brief inscription (just fourteen graphemes!), I continue to contend for the following reading: DE OSTAE OU PSŌ AGB, while also considering viable: DI OSTAE OU PSŌ. In terms of the verb, it could be understood (as I suggested on March 15, Rollston 2012b) as psaō, with either the transitive or intransitive meanings I mentioned then (i.e., “I touch not,” or “I crumble not away”/”I disappear not”). Conversely, because we do see the shortened form of the negative attested epigraphically in Greek (i.e., o for ou; perhaps also compare the phenomenon of crasis in Greek), it is also viable to suggest (as I did in Rollston 2012a) that the verb preceded by the negative is indeed upsoō (i.e., “lift,” “raise up,” “exalt”), especially since a number of ossuary inscriptions refer to the movement or non-movement of ossuaries or bones (see Rollston 2012b for these references). Of course, in the latter case something such as this is tenable: “Because of the bones, I move not (the ossuary), O Agabus,” or “Because of the bones, I Agagus, move not (the ossuary),” with the ossuary being understood, as it is the thing being written upon. Of course, something such as “Here are the bones, I move not (the ossuary/bones), O Agabus,” or “Here are the bones, I Agabus move (the ossuary/bones) not” is also plausible. In sum, I consider this inscription to be about bones, and it is also clear that the tetragrammaton is simply not used in this inscription. I should mention here that Bauckham has recently posted a nice summary of the discussion of the inscription (Bauckham 2012b) and it is clear that he is now very disinclined to read the tetragrammaton in this inscription.


Most recently, James Charlesworth has stated that he believes the name “Jonah” is incised in the Aramaic script on the “Amphora Ossuary” (i.e., the ossuary Tabor and Jacobovici have dubbed the “Jonah” Ossuary”). I have looked closely at the photographs of this ossuary, especially the place where Charlesworth believes he sees letters encrypted. I would contend that there are no letters here, but rather these are just incisings that are part of the amphora. That is, this ossuary has a great deal of incising work on it and the incising that Charlesworth considers to be letters are just horizontal and vertical incising marks that were part of the production of the design of the amphora (see Posner 2012 for a discussion). The name “Jonah” is simply not here. Moreover, James Davila and Antonio Lombatti have stated the same thing, that is, Jonah is not there.


The presence of ornamental designs on an ossuary is fairly standard. Rosettes are among the most common ornamental motifs, but the repertoire is really quite broad. Rahmani has discussed them in great detail. He mentions that, in addition to rosettes, the attested ornamental motifs include things such as depictions of tomb facades, columned porches, lattice gates, nephesh towers, amphorae, menorahs, grapes and grapevines, palm trees, and even a putative fish (Rahmani 1994, 28-52). Of course, the fact that ossuaries would have such rich ornamental diversity should come as no surprise, as Late Second Temple period tombs themselves would sometimes be decorated rather nicely as well (see Berlin 2002, 138-148). The following pericope from Maccabees is also apropos in this connection: “Simon sent and took the bones of his brother Jonathan, and buried him in Modein, the city of his ancestors…and Simon built a monument over the tomb of his father and his brothers…he also erected seven pyramids, opposite one another, for his father and mother and four brothers. For the pyramids he devised an elaborate setting, erecting about them great columns, and on the columns he put suits of armor for a permanent memorial, and beside the suits of armor he carved ships…” (1 Macc 13:25-29).

Predictably, therefore, there are ornamental motifs on the ossuaries of Talpiyot Tomb 1981. One ornamental motif in particular has attracted the attention of Tabor and Jacobovici. Namely, Tabor and Jacobovici have contended that one ornamental motif is to be understood as Jonah and a fish. But they did not stop there. Rather, they have argued that this ornamental motif is not just any fish, they have actually argued that it is the dag gadol (the “big fish”) of the book of Jonah. But they went still further, as they speculated that the etchings at one end of the motif are a graphic depiction of Jonah himself, as he is being spewed from the mouth of the whale. In addition, they have also contended that this symbol should be understood here as the earliest reference to “Jonah as a symbol of the Christian resurrection,” citing the following text from the Greek New Testament: “For just as Jonah was in the belly of the sea monster for three days and three nights, so also shall the son of man be in the heart of the earth for three days and three nights” (Matt 12:40), as well as later Christian usage of the Jonah motif (Tabor and Jacobovici 83-90, et passim).

There has been more discussion about the ornamentation than about any other aspect of Talpiyot Tomb 1981. Joan E. Taylor sent an article to Eric Meyers and me on February 28, 2012 and stated that she believed the ornamentation was simply that of an unguentarium, that is, an amphora used to hold burial unguents such as nard. Taylor’s article was posted on ASOR’s blog within two days (Taylor 2012). At the same time, Antonio Lombatti suggested the same thing independently, as did also Tom Vereena. Moreover, scholars such as Mark Goodacre and Robert Cargill have also argued strongly for the very same thing (i.e., that it is an amphora), and it seems to me that there can be no real doubt about the fact that the ornamentation on the side of the ossuary in question is that of an amphora, something Rahmani listed as among the attested types of ornamentation on ossuaries. Indeed, it seems reasonable to refer to this ossuary as the “Amphora Ossuary,” rather than the “Jonah Ossuary” (the latter being the dubbing of Tabor and Jacobovici). I should note in this connection that Steven Fine (early in March 2012) made a sustained case for considering the ornamentation to be a nephesh tower (Fine 2012), something that seemed at that time quite convincing. Finally, it should be mentioned that already in 1981, an article in the Hebrew newspaper DAVAR reported on Talpiyot 1981 and suggested that the ornamentation on this ossuary was an amphora (Meyers and Rollston 2012).


There is also a serious omission in Tabor and Jacobovici’s publications regarding the DNA. Prior to discussing the DNA evidence, certain things must be remembered: (1) Evidence from many Late Second Temple Jerusalem tombs demonstrates that multiple people were often buried in the same ossuary, and the names of all people whose bones are placed in ossuaries are often not all written on the ossuary. (2) Furthermore, the bones from the ossuaries of Talpiyot 1980 were not even “available to Amos Kloner [in 1996] for study since they had been transferred to the religious authorities for reburial, in accordance with an agreement that was made between the Israeli government and the religious authorities who objected to the storage of human bones within the Antiquities Authority’s storerooms” (Gibson 2006, 120); thus, they were certainly not available to Tabor and Jacobovici. (3) Of course, Tabor and Jacobovici were able to find some small bone fragments, but the problem is that it is not possible to know if the fragments Tabor and Jacobovici sent for analysis are those of someone whose name is on the ossuary! There are, therefore, some “missing links” in any attempt to do DNA analyses and there is no way to overcome these problems.

In any case, Tabor and Jacobovici had DNA tests done on the few bone fragments still present in the “Yeshua’ Ossuary” and the “Mariame kai Mara Ossuary” (i.e., after these ossuaries of Talpiyot Tomb 1980 had been cleaned and the bones reburied long ago). They have stated that it was not possible to recover Nuclear DNA, but it was possible to recover Mitochondrial DNA (Tabor and Jacobovici, 196-202). Here are the statements Tabor and Jacobovici have made: In the “Mariamene” ossuary “we found only tine bone chips.” Or again, “the bone chips we found contained no marrow.” And yet again, “There was no possibility of nuclear or gDNA with these samples due to their degradation.” Or again, “It is unfortunate that we were not able to conduct full DNA tests on all of the bones found in all the ossuaries from the Jesus tomb. Ideally that would have allowed one to construct a kind of provisional ‘family tree,’ at least in terms of the familial genetic relationships between those individuals buried therein. Since the bones themselves were never examined scientifically and no one is even sure what happened to them, that opportunity is forever lost” (Tabor and Jacobovici, 199-202).” Because the mitochondrial DNA from these two ossuaries did not match (i.e., that is, they did not have the same mother, or were not those of a son and mother), Tabor and Jacobovici believed that it was safe to conclude that the people whose bone fragments were removed from these two ossuaries might have been married (in fact, they believe that it is safe to assume that these bones are those of Jesus of Nazareth and Mary Magdalene, whom they propose was married to Jesus of Nazareth).

It should be emphasized that it is not methodologically prudent to assume that there was only one person buried in the “Yeshua’ Ossuary” and only one person buried in the “Mariame kai Mara Ossuary.” After all, multiple people were often buried in the same ossuary and when this occurs it is commonplace for all the names *not* to be listed. Moreover, with the “Mariame kai Mara Ossuary,” one can make a good case for the fact that the inscription itself refers to two people, not one. Furthermore, there is also something else that is of critical importance: from mitochondrial DNA, one cannot determine gender! Thus, the bones from the “Yeshua’ Ossuary” could readily be those of a woman and those of the “Mariame kai Mara Ossuary” could readily be those of a man! Obviously, this is a critically important aspect of the DNA evidence, but rather than mentioning this, Tabor and Jacobovici simply assume (and thereby lead their readers to assume) that the bone fragments from the “Yeshua’ Ossuary” are definitely those of a man, and they assumed that those from the “Mariame kai Mara Ossuary” are definitively those of a woman. Obviously, this serves the purposes of Tabor and Jacobovici very well, but it is hardly a conclusion that follows directly from the evidence. Ultimately, this “omission” of theirs demonstrates decisively that candor about the DNA evidence was not provided by Tabor and Jacobovici, and this is a most unfortunate omission.


Tabor and Jacobovici have argued that these two tombs are definitely connected in some fashion (indeed, their terms “Talpiyot A” and “Talpiyot B” seem to suggest just this, even though there are additional tombs in this part of East Talpiyot). They begin by mentioning that these two tombs are around two hundred feet apart and then they state that “the new discovery [i.e., Talpiyot Tomb 1981] increases the likelihood that the ‘Jesus Family Tomb’ [i.e., Talpiyot Tomb 1980] is, indeed, the real tomb of Jesus of Nazareth.” Tabor and Jacobovici further conclude that “both tombs appear to have been part of the property of a wealthy individual possibly Joseph of Arimathea, the man who, according to the gospels, buried Jesus.” But I would note that there is no necessary connection between these two tombs. That is, ca. 200 feet is not a small distance. Moreover, I would emphasize that there is no epigraphic evidence to connect these two tombs, and there is no archaeological evidence to suggest a close connection either. In short, it is speculative to assume that these two tombs are to be connected.


Tabor and Jacobovici contend that Jesus of Nazareth was married to Mary Magdale and that they had a son named Yehudah. They think the evidence from these tombs proves all of this and they also suggest that some hints of this can be found in the New Testament and Early Christian literature. Bart Ehrman, a premiere scholar of the Greek New Testament and Early Christianity, has stated that the question people most ask him is this: “Were Mary Magdalene and Jesus of Nazareth married?” Here is Ehrman’s answer: “It is not true…that the Dead Sea Scrolls contained Gospels that discussed Mary and Jesus.” “Nor is it true that the marriage of Mary and Jesus is repeatedly discussed in the Gospels that didn’t make it into the New Testament. In fact, it is never discussed at all—never even mentioned, not even once.” He goes on and notes that “It is not true that the Gospel of Philip calls Mary Jesus’ spouse.” Then he queries: “What what does the historical evidence tell us about Mary and Jesus?….it tells us nothing at all—certainly nothing to indicate that Jesus and Mary had a sexual relationship of any kind” (Ehrman 2006, 248). Ehrman’s historical analysis is dead on. I completely concur with him.


In sum, the technology used on Talpiyot Tomb 1981 by Tabor and Jacobovici is interesting and it would have been nice to be able to focus on it. Moreover, it would have been nice to have had a sober discussion by Tabor and Jacobovici about the finds from Talpiyot Tombs 1980 and 1980; nevertheless, this did not occur. Ultimately, therefore, this volume is replete with a superabundance of problems. I have just discussed a few of them. Some of the additional problems have been dealth with by scholars such as James McGrath and Mark Goodacre, who have focused especially on Tabor and Jacobovici’s problematic use of New Testament texts (see the ASOR Blog for these posts as well). I fear that this volume will be used in the field of academics primarily as “a case study of flawed methodologies and flawed practices”. The reason for this is simply that there are so many tenuous assumptions, leaps of logic, and omissions of data. This is all really quite tragic. Fortunately, there has been a cadre of scholars working to correct the assertions and arguments of Tabor and Jacobovici and this has been most useful. Moreover, the American Schools of Oriental Research deserves a great deal of credit for providing a forum for much of the discussion.


Bauckham, R.
2012a “The Four-Line Ossuary Inscription from Talpiyot Tomb B—An Interpretation.” ASOR Blog March 2012.

Bauckham, R.
2012b “The Four-Line Greek Inscription from Talpiyot Tomb B: A Summary of Options for Reading and Interpreting It.” April 2012.

Bauer, W., Arndt, W.F., Gingrich, FW., Danker, F.W.
2000 A Greek-English Lexicon of the New Testament and other Early Christian Literature, 3rd
ed. Chicago: The University of Chicago Press.

Berlin, A. M.
2002 Power and its Afterlife: Tombs in Hellenistic Palestine. NEA 65: 138-148.

2010 Corpus Inscriptionum Iudaeae/Palaestinae: volume I, Jerusalem, Part 1, 1-704. H. Cotton, L. Di Segni, W. Eck, B. Isaac, A. Kushnir-Stein, H. Misgav, J. Price, I. Roll, and A. Yardeni, eds. Berlin: DeGruyter.

Collins, J. J.
1998 The Apocalyptic Imagination: An Introduction to Jewish Apocalyptic Literature, 2nd ed.
Biblical Resource Series. Grand Rapids: Eerdmans.

Di Lella, A. A.
1966 “Conservative and Progressive Theology: Sirach and Wisdom.” CBQ 28: 139-154.

Ehrman, B. D.
1999 Jesus: Apocalyptic Prophet of a New Millennium. New York: Oxford Press.

2006 Peter, Paul, and Mary Magdalene: The Followers of Jesus in History and Legend. New York: Oxford.

Feuerverger, A.
2007 “Open Letter to Statistical Colleagues on the Tomb Computation.” Online:

Fine, Steven. “Comments from Prof. Steven Fine on the Jesus Discovery.”

Friedrich, G.
1972 Theological Dictionary of the New Testament: Volume 8. Grand Rapids: Eerdmans.

Gibson. S.
2006 “Is the Talpiot Tomb Really the Family Tomb of Jesus?” NEA 69: 118-124.

Hoftijzer, J. and Jongeling, K.
1995 Dictionary of the North-West Semitic Inscriptions. Leiden: Brill.

Hurtado, L. W. T
2006 The Earliest Christian Artifacts: Manuscripts and Christian Origins. Grand Rapids: Eerdmans.

Ilan, Tal.
2002 Lexicon of Jewish Names in Late Antiquity. Tübingen: Mohr Siebeck.

Jacobovici, S. and Pellegrino, C.
2007 The Jesus Family Tomb: The Discovery, the Investigation, and the Evidence that Could Change History. New York: Harper Collins.

Jastrow, M.
1950 A Dictionary of the Targumim, the Talmud Babli and Yerushalami, and the Midrashic
Literature. New York: Pardes Publishing House.

Koehler, L. and Baumgartner, W.
1994-2000 The Hebrew and Aramaic Lexicon of the Old Testament. Leiden: Brill.

Kraeling., Carl H.
1946 “Christian Burial Urns.” Biblical Archaeologist 9: 16-20.

Lampe, G. W. H.
1961 A Patristic Greek Lexicon. Oxford: Clarendon Press.

Lewis, N.
1989 The Documents from the Bar Kokhba Period in the Cave of Letters: Greek Papyri. Judean
Desert Studies. Jerusalem: Israel Exploration Studies.

Liddell, H. G. and Scott, R.
1993 A Greek-English Lexicon, rev. and augmented ed. Oxford: Clarendon Press.

Meyers, E. M.
2006 “The Jesus Tomb Controversy: An Overview.” NEA 69: 116-118.

Meyers, E.M. and Rollston, C.A.
2012. “Jonah Ossuary Discussed in Print in 1981.”

Metzger, B. M.
1981 Manuscripts of the Greek Bible: An Introduction to Greek Palaeography. New York:
Oxford University Press.

Nickelsburg, G. W.E., Jr.
1972 Resurrection, Immortality, and Eternal Life in Intertestamental Judaism. Harvard
Theological Studies 26. Cambridge: Harvard University Press.

Oikonomides, Al. N.
1974 Abbreviations in Greek Inscriptions: Papyri Manuscripts and Early Printed Books.
Chicago: Ares Publishers.

Pfann, S. J.
2006 “Mary Magdalene Has Left the Room: A Suggested New Reading of Ossuary CJO 701.”
NEA 69: 130-131.

Posner, Michael,
2012. “Ancient Ossuary Hints at Earliest Reference to Jesus.” April 12, 2012.

Rahmani, L. Y.
1994. A Catalogue of Jewish Ossuaries in the Collections of the State of Israel. Jerusalem: Israel Antiquities Authority.

Rollston, C.A.
2006 “Inscribed Ossuaries: Personal Names, Statistics, and Laboratory Tests.” NEA 69: 125-

2012a “The Four-Line Greek Inscription from a Talpiyot Tomb: Epigraphic Notes and Historical
Discussions.” ASOR Blog March 15, 2012.

2012b “The four-Line Ossuary Inscription from Talpiyot Tomb B (1981): Summary and Restatement.” March 17, 2012.

Scham, S.
2006 “Trial by Statistics.” NEA 69: 124-125.

Sukenik, E.L.
1931 “Nochmals: Die Ossuarien in Palästina.” Monatsschrift für Geschichte und
Wissenschaft des Judentums 75: 462-463.

Tabor, J. D.
2006 “Testing a Hypothesis.” NEA 69: 132-136.

2012 “A Preliminary Report of a Robotic Camera Exploration of a Sealed 1st Century Tomb in East Talpiyot, Jerusalem.” Bible and Interpretation web site, February 28, 2012.

Tabor, J. D. and Jacobovici, S.
2012 The Jesus Discovery: The New Archaeological Find that Reveals the Birth of Christianity. New York: Simon and Schuster.

Taylor, J.E.
2012 “The Talpiyot Ungentarium. March 2, 2012.

Tov, E.
2004 Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (STDJ 54; Leiden/Boston: E. J. Brill.

Yadin, Y.
1971 Bar-Kokhba: The Rediscovery of the Legendary Hero of the Second Jewish Revolt against Rome. New York: R

Modern Epigraphic Forgeries

5 September 2009

The Crisis of Modern Epigraphic Forgeries and the Antiquities Market: A Palaeographer Reflects on the Problem and Proposes Protocols for the Field [1]

Synopsis of The Problem

The number of Northwest Semitic inscriptions appearing on the antiquities market continues unabated. Some of these epigraphic objects are genuine (i.e., ancient) inscriptions, but have appeared on the market as a result of illicit excavations. [2] Some of these epigraphic objects, however, are modern forgeries.
It should be safe to presume that because of the presence of modern forgeries on the antiquities market, vigilance and caution would be the modus operandi of specialists (and non-specialists) within the field. Sometimes, however, credulousness has actually been regnant of late. This suspension of critical judgment has precipitated at least two crises: (1) The dataset of ancient Northwest Semitic has been corrupted with modern forgeries, and (2) the general public has become suspicious about the capacity of the field to produce and convey reliable information.

The purpose of this article is to discuss various aspects of the forgery crisis, including some of the assumptions that foster it, and to propose various protocols for the field so as to protect the dataset of Northwest Semitic.

The Forger’s Toolbox: Available Resources

The field has sometimes had the a priori assumption that modern forgers cannot produce “good forgeries,” that is, forgeries that “appear ancient.” However, I would argue that forgers have all of the resources necessary to produce superb forgeries that “pass all the tests,” or at least pass them to the satisfaction of many. To elucidate this point, it is useful to list some of the primary and secondary sources that would be most useful for a forger with a knowledge of biblical Hebrew attempting to produce an inscription written in Iron Age Hebrew script and language (i.e., Old Hebrew):

(a) a standard dictionary of biblical Hebrew, Hoftijzer and Jongeling’s Dictionary of the North-West Semitic Inscriptions, and Davies Ancient Hebrew Inscriptions; [3]

(b) Cross and Freedman’s Early Hebrew Orthography; [4]

(c) Waltke and O’Connor’s Introduction to Biblical Hebrew Syntax; [5]

(d) Birnbaum’s The Hebrew Scripts, Naveh’s Early History of the Alphabet, and Cross’s seminal articles on the Iron Age Hebrew script in BASOR;[6]

(e) Donner and Röllig’s Kanaanïsche und aramïsche Inschriften;[7]

(f) Freedman’s Anchor Bible Dictionary; [8]

(g) Pardee’s Ancient Hebrew Letters;[9]

(h) and Avigad and Sass’s Corpus of West Semitic Stamp Seals. [10]

Using such sources, a deft modern forger has the essentials regarding script, orthography, vocabulary, syntax, language, and culture to produce a fine Old Hebrew forgery. All of these sources are widely considered standard in the field and are readily available; therefore, knowledge of, and access to, the proper resources is not an issue (and such sources are available for all the Northwest Semitic languages, not just Hebrew). Forgers also now have available software programs (e.g., Adobe Photoshop) that can be used to facilitate accurate “script production.”

Some of the remaining requirements would be adequate time; some knowledge of, or expertise in, chemistry or ancient metallurgy (or an associate with such expertise); access to various materials such as potsherds, ancient metals, stone of Levantine quarry, small pieces of ancient papyrus or vellum, some carbonized remains (for the production of “ancient” ink), and sufficient finances. None of these necessities is problematic.

Because non-provenanced epigraphs often sell for thousands or tens of thousands of dollars, funding is not a major issue; that is, the sale of one forgery could fund the production of several additional forgeries. Moreover, ancient materials of various sorts are readily available to those participating in, or associated with, excavations or those dealing with the market. In addition, the chemical composition of ancient patinas can be replicated. [11]

Specialists and non-specialists in epigraphy and biblical studies must come to terms with the fact that the production of a good forgery in the contemporary period is not facile, but neither is it now as difficult as specialists and non-specialists within the guild would have affirmed in the past. Forgers have all the tools needed to produce a rather impeccable forgery. Fortunately, of course, forgers often make mistakes (and these can be detected), but it is imprudent to assume that this is always the case. The point is that forgers have ample “means.”

Motivations of Forgers

Some have suggested or assumed that the primary motive for forgers is economic, but I am confident that a variety of motivations can with certainty be posited for the production of forgeries. Of course, venality is certainly a component present in the production and sale of forgeries. Non-provenanced inscriptions routinely sell for four, five, and even six figures. Some recent non-provenanced inscriptions have been valued at seven figures.

Some forgeries are arguably the result of sour grapes (e.g., a student purged from a Northwest Semitic epigraphy program) or professional rivalry, with the forger hoping to “dupe” the “offender.” Naturally, a forgery can sometimes be a prank. For example, the forger of the Hebron Documents was probably a prankster or a dolt, or both. Moreover, there is a certain amount of prestige associated with being the person who “collects,” “vets,” or “finds” a significant “ancient epigraph” from the market. Indeed, the public and even scholars within the field can sometimes lionize such people because of “sensational” non-provenanced epigraphs. For this reason, it is my position that forgers may sometimes produce inscriptions so as to be lauded as the one who “found,” “vetted,” or “owns” a sensational epigraph.[12]

Religion and politics are also strong motives for the production of a forgery. For example, there was arguably a strong religious motivation for the production of the Shapira Fragments and the initial aura surrounding them. [13] The fact that the Jehoash Inscription was reported to have been found in the region of the Temple Mount has political and religious overtones. Ultimately, forgers are probably motivated by a combination of such factors, and, of course, with each success, hubris is fostered. The main point is that forgers have substantial “motive.” [14]

Fabrications of Forgers: Selected Modern Forgeries

We should note that forgeries of Northwest Semitic inscriptions have been produced for more than a century. For example, during the late nineteenth century, a Phoenician inscription surfaced purporting to be an account of Sidonians from the region of biblical Ezion-geber, who circled the “land belonging to Ham” (i.e., Africa) during the reign of King Hirom (r. 552-532 B.C.E.), but were blown far off course to a “distant shore.” This inscription was reported to have been found at a place called Pauso Alto near the Paraíba River in Brazil. Although many were jubilant about the inscription’s contents, Lidzbarski declared it a definite forgery, and the inscription was forgotten. [15] Gordon, however, published an article in the late 1960s discussing this inscription, and argued that it was definitely genuine.[16] Cross was not convinced and demonstrated decisively that it was a forgery, based on the fundamental problems with the script, orthography, and lexicon. [17]

During December of 1970, non-provenanced documents reported to have been found in the region of Hebron were announced. Brownlee and Mendenhall considered them ancient, presented them at the 1971 Society of Biblical Literature meeting, and argued that they were “Philistine.”[18] Although Cross declared that the documents were forgeries, noting parallels between the Siloam Tunnel Inscription and the “Hebron Documents,” Mendenhall persisted in affirming the authenticity of the “Hebron Documents.”

During the early 1980s, Naveh did a detailed analysis of one of the “Hebron Documents,” demonstrating at length that it was a modern forgery. In fact, he demonstrated that the forger had in essence simply copied large portions of the Siloam Tunnel Inscription, but had done so (essentially) from left to right, that is “backwards.” [19]

During the late 1990s, Naveh and Eph’al argued that the contents of the non-provenanced “Moussaieff Ostraca” were suspicious, but did not state definitively that they were modern forgeries.[20] I subjected these Moussaieff Ostraca to extensive palaeographic analysis, and the results demonstrated that the script of these ostraca reflected numerous features that deviated in fundamental ways from all provenanced Old Hebrew inscriptions, with some tell-tale features revealing the forger’s mistakes with the script. I concluded that these two ostraca were definitive modern forgeries and the results of my research were presented at annual meetings of the Society of Biblical Literature in 2001 and 2002, with the formal publication appearing in 2003. [21]

During early 2003, the “Jehoash Inscription” surfaced, allegedly having been found in the region of the Temple Mount in Jerusalem and purchased on the antiquities market. Although the first line is not extant, it is readily apparent that the inscription purports to have been commissioned by the late ninth century Judean King Jehoash (Joash). Naveh, McCarter, Cross, and I analyzed this inscription independently and concluded that it was a definite modern forgery. Some scholars, however, concluded that it might indeed be authentic. [22]

Cross published a detailed analysis of the Jehoash inscription, noting some of the severe problems with its orthography and contents, and Eph’al wrote a brief but compelling article detailing the striking similarities between the Jehoash Inscription and the material in Kings and Chronicles about the reforms of Jehoash. [23] Moreover, I marshaled palaeographic evidence against it, demonstrating that the script was a “script mélange” with some letters being Phoenician, some Old Hebrew, and some Aramaic. The conclusion was readily apparent: it is a definitive modern forgery.

I also noted that there are some striking parallels between the Moussaieff Ostraca and the Jehoash Inscription:

(1)Moussaieff Ostracon 1 and the Jehoash Inscription both refer to donations to the temple of Yahweh, under the auspices of the monarchy.

(2) Moussaieff Ostracon 1 and the Jehoash Inscription both arguably refer to the same monarch (Jehoash of Judah).

(3) The Moussaieff Ostraca and the Jehoash Inscription both share certain palaeographic “anomalies”: (e.g., with the Old Hebrew letters šin and samek).

(4) Cumulatively, these are striking connections; therefore, in my opinion, there is a distinct possibility that the same forger produced the Moussaieff Ostraca and the Jehoash Inscription.[24] Thus, history demonstrates that forgers have seized the “opportunity” at times. Moreover, there is every reason to assume that this will continue to be the case.[25]

The Epigrapher’s Toolbox I: Knowledge of the Methods and Mistakes of Forgers

Several methodological points regarding forgeries and forgery detection can be articulated at this juncture. Modern forgeries are often reported to have come from specific locations to increase the credibility of the objects’ authenticity (e.g., Hebron Documents, Jehoash Inscription, Brazilian Phoenician Inscription).[26] Therefore, epigraphers must not consider information about purported sites of discovery for non-provenanced inscriptions to be useful, in and of itself, for making determinations regarding authenticity.
For a number of reasons, modern forgers have traditionally relied heavily upon provenanced epigraphic and biblical materials. Sometimes this information is damning (“Hebron Documents”), but sometimes (“Moussaieff Ostraca”) this information is more suggestive or even of no absolutely necessary consequence. Also of significance in this connection is the fact that although forgers have been predisposed intentionally to use attested words and phrases, they are sometimes ignorant of the semantic evolution of these words (Jehoash Inscription).

Forgers often produce inscriptions with sensational contents (e.g., Moussaieff Ostraca, Jehoash Inscription, Ivory Pomegranate), perhaps because these create enormous interest (and irrational exuberance) and yield high selling prices. Forgers are beginning to produce patinas that appear ancient (Jehoash Inscription, Moussaieff Ostraca). This fact, combined with the fact that some lab testing of epigraphic materials has reflected incompetence and collusion, has created problems. The point is that even lab tests must be scrutinized, and protocols for lab testing must be put in place. [27]

Modern forgers often commit serious palaeographic and orthographic errors (e.g., Phoenician Inscription from Brazil, Moussaieff Ostraca, Jehoash Inscription). Palaeographic and orthographic anomalies and anachronisms are of fundamental importance, and in my opinion egregious violations of attested ancient orthography and palaeography provide sufficient basis for complete rejection of a non-provenanced epigraph. Red flags should be noted, and not easily dismissed, even with the”sample size” argument. [28] The end result is that the field of epigraphy should be capable of eradicating many, but not all forgeries from the dataset.

The Epigrapher’s Toolbox II: Protocols for Treatments of Non-Provenanced Inscriptions

Some specialists might suggest that non-provenanced epigraphs should be eliminated in toto from the Northwest Semitic dataset. I would suggest that such materials can sometimes be used, but (1) they must normally be subjected to the most rigorous epigraphic and laboratory analyses in order to determine with substantial reliability that they are ancient, and (2) they should be separated from the provenanced corpus and also flagged as non-provenanced.

Non-Provenanced Epigraphs in Handbooks and Collections: The Principle of Separation

First and foremost, it is readily apparent that those discussing a specific non-provenanced epigraph should articulate the fact that the source of such an epigraph was not a controlled archaeological excavation. In the past, scholars have sometimes been remiss in this regard. [29] This practice is particularly problematic because readers might reasonably conclude that such inscriptions are definitively provenanced and ancient. Ultimately, I would suggest that those discussing an epigraph should clearly refer to the “circumstances of discovery and recent history” in a precise manner so as to avoid causing readers to make erroneous conclusions about the actual or putative origins.

For some time, there has been a tradition of including non-provenanced epigraphs side-by-side with provenanced materials. For example, the fine collection of Northwest Semitic seals, bullae, and jar sealings published by Avigad and Sass contains numerous divisions (e.g., “Hebrew Seals,” “Hebrew Bullae and other Sealings,” “Hebrew Jar-Handle Impressions,” “Phoenician Seals,” “Aramaic Seals,” etc.); however, provenanced and non-provenanced epigraphs are not separated. [30]

Lindenberger’s superb collection of Northwest Semitic epistolary epigraphs also fails to separate provenanced and non-provenanced epigraphs.[31] Not separating provenanced and non-provenanced materials was a convenient, utilitarian practice in the past, but I would posit that combining the data in this fashion is problematic: it implicitly and erroneously suggests to many readers that the data from non-provenanced materials and provenanced materials are on a par. Therefore, I would argue that at this juncture, for methodological reasons, provenanced and non-provenanced epigraphs should be separated, placed in distinct sections of handbooks and collections, and be given descriptive labels such as “Provenanced Epigraphs” and “Non-Provenanced Epigraphs.” In short, the field must become very intentional about presentation in handbooks and collections.

The Principle of “Flagging”

Within certain types of works (e.g., lexica), it may not be practical to present the provenanced and non-provenanced materials separately (i.e., with completely separate entries of some sort for the provenanced and non-provenanced evidence). Therefore, I would suggest that non-provenanced epigraphs cited in the entry be marked or flagged in some fashion so as to signify their status as non-provenanced. This system will allow the reader to understand immediately that this non-provenanced epigraphic data may need to be weighted differently (i.e., it is not necessarily of the most pristine sort).
Several potential methods of marking are possible. For example, the reference could be preceded (or followed) by the mathematical symbol Ø, signifying in this case the absence of provenance. Hence, “ØMoussaieff Ostracon 1” would convey to the reader that this particular ostracon is non-provenanced, as would something such as “[non-prov]Moussaieff Ostracon 1.” The section on sigla or abbreviations within the volume or article could be used to communicate the author’s system of flagging. [32]

The Principle of Relegation

It is readily apparent that epigraphic materials without secure provenance and without certain antiquity are normally compromised, problematic, and precarious bases for reconstructing the past. Nevertheless, scholars sometimes do continue to base certain conclusions about various aspects of antiquity on non-provenanced materials. For example, Heltzer authored a recent article about property rights of women in ancient Israel, but his article is based predominantly on non-provenanced epigraphic materials, and one of the epigraphs he mines heavily for ancient data is actually a modern forgery. [33] Because the potential for forgery is consistently present, scholars must begin to relegate non-provenanced data to a secondary or tertiary position at the very least and must be disinclined to base conclusions regarding history, religion, language, epigraphy, etc., upon such data. The result will be more accurate constructs of antiquity.

The Principle of Categorization

Although several caveats and provisos must be present, I would suggest that specialists begin to be more intentional about categorizing non-provenanced inscriptions: although it is not pragmatic to ignore non-provenanced inscriptions, neither is it prudent to assume that all non-provenanced inscriptions are of equal status in terms of possible authenticity.

I would propose the following categories of assessment regarding the antiquity or modernity of an inscription:

1. Modern Forgery
2. Probable Modern Forgery
3. Possible Modern Forgery
4. Probable Ancient

Ancient Inscriptions that reflect no real aberrations (in terms of script, orthography, etc.), and for which it is certain that laboratory anomalies are absent, can be considered probable ancient or ancient inscriptions. Inscriptions that reflect serious or egregious problems or deviations from the provenanced corpus are to be considered modern forgeries or probable modern forgeries.

Of course, palaeographers will sometimes differ about the authenticity of an inscription. Regarding this issue, I would note the following: (1) Substantial disagreement of palaeographers in print is not nearly as common as is agreement.

(2) Genuine disagreement in print, when it does occur, can often be attributed to the high quality of a forgery or a genuine inscription with modest aberrations; moreover, it can also be due to the relative competency of a palaeographer with the relevant script series.

(3) Palaeographers are sometimes misled by problematic or erroneous laboratory tests, causing a palaeographer to assume the inscription is genuine and then to account for the anomalies with tenuous or strained arguments.

(4) Sometimes a sensational epigraph will cause such exuberance that critical judgment becomes impaired and declarations of authenticity are made on the basis of tenuous evidence. In any case, the views of specialists should be cited, and an assessment of the possible or probable antiquity should be made. [34]
In sum, modern forgeries have been produced for some time. Forgers have means, motive, and opportunity; however, epigraphers and palaeographers also have a substantial counter-arsenal. At this juncture, methodological doubt and rigorous protocols are desiderata. Caveat Eruditus.


1. This article is a much-condensed version of the following detailed articles: Christopher A. Rollston, “Non- Provenanced Epigraphs I: Pillaged Antiquities, Northwest Semitic Forgeries, and Protocols for Laboratory Tests,” Maarav 10 (2003): 135-193; Christopher A. Rollston, “Non-Provenanced Epigraphs II: The Status of Non-Provenanced Epigraphs within the Broader Corpus of Northwest Semitic,” Maarav 11 (2004) 57-79. I am grateful to my research assistants, Heather Dana Davis Parker and Alan Dyson, for their conscientious work.

2. Although it is readily apparent that ancient (i.e., “genuine”) non-provenanced inscriptions often contain valuable data, significant amounts of data are eradicated because of the absence of secure contexts (i.e., provenance, stratum, locus, etc.). For example, discussions of the history, administrative apparatus, and personnel at specific sites require precise data about provenance. Moreover, the fields of dialect geography and palaeography also require reliable data about provenance. The point is that it is indubitable that non-provenanced epigraphs are (with rare exceptions) compromised, and of diminished significance. For a lengthy analysis of the limitations of non-provenanced inscriptions and the superiority of provenanced inscriptions, see Christopher A. Rollston, “Non-Provenanced Inscriptions II.”, (59-70)

3. J. Hoftijzer and K. Jongeling, Dictionary of the North-West Semitic Inscriptions (Leiden: Brill, 1995). G. I. Davies, Ancient Hebrew Inscriptions: Corpus and Concordance (Cambridge: Cambridge University Press, 1991).

4. F. M. Cross, Jr., and D. N. Freedman, Early Hebrew Orthography: A Study of the Epigraphic Evidence, AOS 35 (New Haven: AOS, 1952).

5. B. K. Waltke and M. O’ Connor, An Introduction to Biblical Hebrew Syntax (Winona Lake: Eisenbrauns, 1990). For epigraphic Hebrew, see especially S. L. Gogel, A Grammar of Epigraphic Hebrew, SBLRBS 23 (Atlanta: Scholars, 1998).

6. S. Birnbaum, The Hebrew Scripts: Vol 1-2 (London: 1954-1971); J. Naveh, Early History of the Alphabet: An Introduction to West Semitic Epigraphy and Palaeography, 2d ed. (Jerusalem: Magnes, 1987); Cross’s articles have been republished by Cross in a volume entitled Leaves From an Epigrapher’s Notebook: Collected Papers in Hebrew and West Semitic Palaeography and Epigraphy, HSS 51 (Winona Lake: Eisenbrauns, 2003), 114-128. For the Old Hebrew script, see also Christopher A. Rollston, “The Script of Hebrew Ostraca of the Iron Age: Eighth – Sixth Centuries B.C.E.” (Ph.D. diss., Johns Hopkins University, 1999). This volume will be published in a revised and augmented form as The Art of the Scribe in Israel and Judah: The Script of Ancient Hebrew Ostraca.

7. H. Donner and W. Röllig, Kanaanïsche und aramïsche Inschriften, 5th ed. (Wiesbaden: Harrassowitz, 2002).

8. D. N. Freedman, The Anchor Bible Dictionary (New York: Doubleday, 1992).

9. D. Pardee, Handbook of Ancient Hebrew Letters, SBLRBS 15 (Chico: Scholars, 1982).

10. N. Avigad, with revisions by B. Sass, Corpus of West Semitic Stamp Seals (Jerusalem: Israel Exploration Society, 1997).

11. This is something that I have argued for some time, but Goren, an archaeologist who uses various geological methods, recently demonstrated this conclusively. See Y. Goren’s, “An Alternative Interpretation of the Stone Tablet with Ancient Hebrew Inscription Attributed to Jehoash King of Judah,”

12. Moses Wilhelm Shapira was of course a Jerusalem antiquities dealer, but he also fancied himself a “scholar” and “archaeologist.” He gained much prominence during the late nineteenth century because of his “finds,” and this resulted in his elevation in various social circles. Naturally, however, with the exposure of the “Moabite Potteries” (often with inscriptions) and the “Shapira Fragments” as forgeries, his status plummeted, and he ultimately committed suicide. It is my opinion that Shapira himself forged the “Shapira Fragments.”

13. On this point, see P. K. McCarter, “Shapira Fragments,” BAR 23 (May/June, 1997): 40.

14. For discussion of, and bibliographic references for, the forgeries referred to in this paragraph, see especially, Christopher A. Rollston, “Non-Provenanced Epigraphs I,” passim.

15. For an early analysis, see K. Schlottmann, “Die sogenannte Inschrift von Parahyba,” ZDMG 28 (1874): 481-487, with the plate published on page 481 (and reproduced in my article). For Lidabarski’s assessment, see M. Lidzbarski, Handbuch der Nordsemitischen Epigraphik (Weimar: Verlag von Emil Felber, 1898), 132.

16. C. H. Gordon, “The Authenticity of the Phoenician Text from Parahyba,” Orientalia 37 (1968): 75-80.

17. F. M. Cross, “The Phoenician Inscription from Brazil: A Nineteenth-Century Forgery,” Orientalia NS 37 (1968): 437-460. This article was republished in Leaves from an Epigrapher’s Notebook, 238-249.

18. See W. H. Brownlee and G. E. Mendenhall, “An Announcement Published by the Department of Antiquities of Jordan and the Archaeologists Dr. William H. Brownlee and Dr. George E. Mendenhall regarding the Decipherment of Carian Leather Manuscripts,” ADAJ 15 (1970): 39-40; G. E. Mendenhall, “The ‘Philistine’ Documents from the Hebron Area: A Supplementary Note,” ADAJ 16 (1971): 99.

19. J. Naveh, “Some Recently Forged Inscriptions,” BASOR 247 (1982): 53-58.

20. I. Eph’al and J. Naveh, “Remarks on the Recently Published Moussaieff Ostraca,” IEJ 48 (1998): 269-273. For the original publication of these ostraca, see P. Bordreuil, F. Israel, and D. Pardee, “Deux Ostraca paléo-hébreux de la collection Sh. Moussaïeff,” Semitica 46 (1996): 49-76; Bordreuil, Israel, and Pardee, “King’s Command and Widow’s Plea: Two New Hebrew Ostraca of the Biblical Period,” NEA 61 (1998): 2-13. Cf. D. Pardee, “Hebrew Letters,” in The Context of Scripture III: Archival Documents from the Biblical World, ed. W. W. Hallo and K. L. Younger, Jr. (Leiden: Brill, 2002), 86.

21. Rollston, “Non-Provenanced Epigraphs I,” 158-173.

22. H. Shanks, “King Jehoash Inscription Captivates the Archaeological World,” BARev 29 (March/April 2003): 22-23.

23. F. M. Cross, “Notes on the Forged Plaque Recording Repairs to the Temple,” IEJ 53 (2003): 119-123. I. Eph’al, “The ‘Jehoash Inscription’: A Forgery,” IEJ 53 (2003): 126.

24. Rollston, “Non-Provenanced Epigraphs I,” 175-180.

25. It should be noted that at this juncture, the Israeli Special Commission has analyzed various inscriptions, including the Moussaieff Ostraca, the Jehoash Inscription, and the Ivory Pomegranate (inscriptions that I argued some time ago were forgeries) and concluded that they are indeed all modern forgeries.

26. It should also be remembered that the Shapira Scrolls were reported to have been found in the region of the Wadi Arnon in Jordan. For a fine analysis of the Shapira Scrolls, see N. A. Silberman, Digging for God and Country: Exploration, Archaeology, and the Secret Struggle for the Holy Land: 1799-1917 (New York: Alfred A. Knopf, 1982), 131-146. See also, F. Reiner, “Tracking the Shapira Case: A Biblical Scandal Revisited,” BAR 23 (May/June 1997): 32-41; 66-67.

27. See Christopher A. Rollston, “Non-Provenanced Epigraphs I,” 182-191.

28. For discussion of the assessment of anomalies and the sample size argument, see Christopher A. Rollston, “Non-Provenanced Epigraphs I,” 180-182.

29. For example, in a recent reference work of superb caliber, there is a brief synopsis of the Ivory Pomegranate, reference to the readings, and a “palaeographic” date; however, there is no clear statement about the fact that this epigraph is non-provenanced. Furthermore, the title of the entry clearly suggests that the provenance was Jerusalem. See, namely, K. L. Younger, Jr., “The Jerusalem Pomegranate,” in The Context of Scripture: Volume II, Monumental Inscriptions from the Biblical World, ed. W.W. Hallo and K. L. Younger, Jr., (Leiden: Brill, 2003), 173. Based on various factors, especially palaeographic issues, Frank M. Cross has considered the Ivory Pomegranate to be a probable forgery for some time, as have I. See Rollston, “Non-Provenanced Epigraphs I,” 182, n. 115.

30. N. Avigad and B. Sass, Corpus of West Semitic Stamp Seals (Jerusalem: Israel Exploration Society, 1997). However, it is imperative to note that the discussion of each epigraph in this volume contains sufficient data to allow the user to determine whether or not the epigraph is provenanced. Furthermore, this volume also contains a list (p. 548) of the provenanced epigraphs published in the corpus as well as a discussion of those epigraphs considered probable forgeries (pp. 453-460).

31. J. M. Lindenberger, Ancient Aramaic and Hebrew Letters, 2nd edition, WAW 14 (Atlanta: Society of Biblical Literature, 2003). Note, for example, that the Moussaieff Ostraca are treated together with the Yavneh-Yam Ostracon (pp. 109-112). However, it is again important to state that Lindenberger is careful to affirm that the Moussaieff Ostraca are non-provenanced and that some epigraphers consider them to be modern forgeries.

32. Obviously, for large corpora of non-provenanced materials, it might be more practical just to discuss the issue on an ad hoc basis in the prologue of the volume (e.g., lexicon), rather than appending a siglum to each separate entry throughout the volume.

33. M. Heltzer, “About the Property Rights of Women in Ancient Israel,” in Shlomo: Studies in Epigraphy, Iconography, History and Archaeology in Honor of Shlomo Moussaieff (ed. R. Deutsch; Tel Aviv: Archaeological Center Publications, 2003), 133-138. For his broader discussion of “women” in the epigraphic corpus, see M. Heltzer, “The Women in the Hebrew Epigraphy of Biblical Times,” Revue Internationale des Droits de l’Antiquité 43 (1996): 11-35. Nevertheless, even with this article, many of his conclusions are based on non-provenanced data, placing this material on a par with the provenanced data. The forgery that I refer to above is Moussaieff Ostracon 2. See C. A. Rollston, “Non-Provenanced Epigraphs I,” 145-146; 158-173 for a discussion of the numerous palaeographic problems and aberrations with this ostracon. See pages 183-184 of my article for a discussion of the serious problems with the laboratory tests performed.

34. It is significant that J. M. Lindenberger is careful to make such a notation. See his Ancient Aramaic and Hebrew Letters, 111-112.

Citation: Christopher A. Rollston, ” The Crisis of Modern Epigraphic Forgeries and the Antiquities Market: A Palaeographer Reflects on the Problem and Proposes Protocols for the Field [1],” SBL Forum , n.p. [cited March 2005]. Online: