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.

A Woman’s Seal and a Man’s Seal from First Temple Jerusalem Excavations

7 March 2016

A Woman’s Seal and a Man’s Seal from First Temple Jerusalem Excavations

By Dr. Christopher A. Rollston, George Washington University (rollston@gwu.edu)

The announcement of two Iron Age seals from Jerusalem is most welcome. These were found on scientific excavations that have been conducted in Jerusalem by Drs. Doron Ben-Ami, Yana Tchekhanovets, and Salome Doron. The photos of the seals that have been released are those of Clara Amit, courtesy of the Israel Antiquities Authority.

I. The Broader Context of the Jerusalem Seals

Seals were part and parcel of the economic and legal activities of the people of the ancient Near Eastern world, particularly the elites. Seals would be used in cases, for example, of the purchase or sale of something of substantial value (e.g., land, precious metals), or in the case of a marriage, or divorce, or adoption (etc.). Within Iron Age Israel, Judah, Moab, Ammon, Edom, Phoenicia, Philistia, and Syria, stamps seals were very commonly used. Many seals have been found on scientific excavations in these regions. Most of these seals are quite round and have about the same diameter as a small coin (although they are thicker than coins). Most were made of attractive, and sometimes rare, stones (incised with a sharp incising tool, made of metal). Most seals have holes drilled through them so that a string (“cord”) could be attached to them. Some were attached in antiquity to a ring, and on rare occasions (such as tomb contexts), the seal and ring are found together. Seals with words inscribed on them are called “Epigraphic Seals.” Seals without words inscribed on them are called “Anepigraphic Seals.” Seals with imagery (e.g., animals, people) on them are called “Iconic Seals.” Seals without imagery are called “Aniconic Seals.” Some seals have words and imagery, some just imagery, some just words. Seals are often divided into “lines.” Each line is referred to as a “register.”

There are some particularly nice references to seals and sealing practices in the Hebrew Bible. Among the most detailed descriptions is the one contained in the book of Jeremiah, a prophet of the late First Temple and early Exilic Periods. Within this biblical text (Jeremiah 32), the prophet is said to have purchased a field from a kinsman of his, in the tenth year of Judean King Zedekiah (ca. 587 BCE, just as Jerusalem was about to fall to King Nebuchadnezzar II of Babylon). Jeremiah is said to have signed a deed of purchase, in the presence of witnesses. There were two copies of this deed (both probably written on papyrus), one the “open copy” and one the “sealed copy.” The open copy would normally be retained for rapid reference and would often have been kept by the purchaser (or a close associate). The sealed copy, however, would normally be archived, often in the house of a scribe, or in an archive of a palace or temple. The sealed copy was the binding legal copy and would only be opened if and when there was some reason to verify in a decisive legal fashion the fact or the nature of the agreement (e.g., purchase, or sale, or marriage, or adoption, or divorce, etc.). In any case, the sealed copy would be rolled up or folded up, and a string would be wrapped around it and then a small clump of wet clay would be attached carefully and precisely to the string, and then the parties to the agreement would press their seals into the clumps of wet clay (the result would be an impression of the seal in the clay, the impressed clumps of clay are referred to as “bullae,” sing: “bulla”). The seals would serve as proof of the event (e.g., purchase, sale, marriage, divorce, adoption). Note that seals are normally incised (i.e., made) in mirror image, so that when they are impressed into the clay, the resulting image is positive (i.e., in the correct orientation). Within the narrative of Jeremiah, the prophet Jeremiah gives both the sealed copy and the open copy to Baruch ben Neriah for safe keeping.

II. The Readings of the Two New Jerusalem Seals

The first seal reads (Register 1): L‘lyhnḥ; (Register 2): bt. g’l. Translation: (Register 1): Belonging to ‘Elyhnḥ; (Register 2): daughter of Go’el. As is customary for seals, this seal (and the other one) begin with the letter lamed, that is, the “Lamed of Ownership.” The first personal name on this seal (i.e., on the first register) consists of two morphemes: ‘ly and hnḥ. Note that it is also arguably possible that the yod (y) is to be associated with the second morpheme (as a marker of the imperfect tense), rather than the first. In any case, the first morpheme is from the root meaning “to be high” and is associated with a number of personal names and divine names (e.g., the personal name ‘Eli in Samuel and the divine name ‘Elyon, for example, in Deut 32:8-9, et passim). The second morpheme is arguably that of the root nwḥ, in the Hiphil stem. From this verbal root come various names, including the personal name Noah of Genesis (Gen 6, et passim) and Samson’s father (Manoah, Judg 13). At this time, I would suggest that an acceptable translation of this personal name on the seal is something such as “God Most High has brought rest.” The second register begins with the standard word for daughter, that is, bt. Although this is not the first seal of a woman to be found, such seals are always of substantial importance because they were, and thus are, quite rare. The second personal name (G’l) is the standard word for “Redeemer” (e.g, a kinsman redeemer). This root is well attested in the Hebrew Bible. For example, Boaz is referred to as a kinsman redeemer (e.g., Ruth 3:13, et passim), because he functioned as a kinsman redeemer for the land and family of Elimelech. In the case of this seal from Jerusalem, that same root word is used as the basis of a personal name. Of course, the putative linguistic symmetry of the two verbal roots for the personal names on this seal is quite nice, with the first personal name referring to “bringing rest” and the second root revolving around the semantic realm of “kinsman redeemer,” (cf. Ruth 1:9, with Naomi’s statement that she hopes her daughters-in-law could find “rest” (nwḥ) in the house of a different husband, after the death of Naomi’s two sons).

The second seal can be read (Register 1): LS‘ryhw. b (or LS‘dyhw b); (Register 2): n.Šbnyhw. Translation: (Register 1): Belonging to Sa‘aryahu (or Sa‘adyahu), so- (Register 2): -n of Shebnayahu (with the word for “son” beginning at the end of the first register and concluding at the beginning of the second register, as is fairly typical at times in seals). The more difficult aspect of this seal is the proper reading of the fourth letter of the first register. The apparent morphology of the head of that letter in the published photo is strongly suggestive of a dalet, rather than a resh (because of what appears to be a slight overlap over of the head). But the length of the stem of that letter is more suggestive of a resh. To be absolutely certain of the reading (i.e., of the morphology of this letter), therefore, I’d prefer to be able to look at this under a microscope myself. Perhaps I can do that soon, permissions permitting. In any case, for the sake of argument, I will discuss both potential morphemes (i.e., of the reading dalet and of the reading resh) for the personal name on the first register. For S‘r, the lexeme is arguably the one that has the basic meaning “to visit, inspect, conduct affairs,” or the lexeme “heavy gale, high wind, to drive away, to blow away.” I slightly prefer the former lexeme, if the reading resh is accepted. Conversely, for the reading S‘d, the lexeme is arguably the one meaning “support, strengthen.” The presence of the yahwistic theophoric is also to be noted, as this is a nice marker of a Judean personal name, something that makes sense in a Jerusalem context, of course. The personal name of the second register (Shebnyahu) is attested in the Hebrew Bible and in the epigraphic record. For example, within Isaiah and Kings (Is 36:3; 2 Kgs 18:18, etc.), there is reference to an official state scribe during the time of Hezekiah (r. ca. 715-687 BCE) as Shebna (a shortened form of the same name on the seal; for additional references to this personal name, see Neh 9:4; 10:5, 12:14, etc.). Epigraphic references include the Old Hebrew Arad Inscriptions (e.g., Arad 27, late 7th/early 6th centuries BCE) and arguably the Royal Steward Inscription of Jerusalem (late 8th century BCE). In terms of meaning, some have suggested that the tri-literal root of this personal name means something such as “to come near, close.” Perhaps.

III. Script, Language, and Date

The script of both of these seals is the standard Old Hebrew script. The script of both seals is nicely done, certainly the work of a trained seal maker (for an important reference in Second Temple Jewish Literature to a seal-maker, see especially Ben Sira 38:27). I have seen some suggestions by some scholars that the script of the ‘Elyhnḥ Seal might be Ammonite. This is definitely not the case. The script of both of these seals is Old Hebrew. Note in particular the stance of the bet (on both seals), reflective of the standard recumbent stance of the Old Hebrew, rather than Ammonite. Regarding the date, I would contend that a date in the late 8th century or early 7th century BCE is the best palaeographic date. Regarding language, it can be said that the language of both of these seals is Old Hebrew, and, of course, the presence of the (Judean) Yahwistic theophoric on the Ben Shebnyahu Seal argues for this as well. Finally, I should also like to emphasize that the reason it was quite rare for women to have seals was not because, as some have apparently suggested, because of the “generally inferior economic status of women.” Rather, I would contend that it was because ancient Near Eastern societies (including that of the Iron Age Levant) were patriarchal. For this reason, men were normally responsible for most of the agreements that would require the sealing of documents. There were certainly some exceptions, as reflected in the lofty narratives about Job’s daughters (Job 42), the daughters of Zelophehad (Num 27), and the Noble Wife (Prov 31). But the biblical and epigraphic evidence converge to suggest that men were the normal brokers of agreements requiring sealed documents.

In sum, these two seals are Old Hebrew. They are both well done, the work of a well-trained seal maker. The palaeographic date that I would assign to them is that of the chronological horizon that spans from the late 8th century to the early 7th century BCE. The Yahwistic theophorics are predictable, but still important. The fact that one of these seals is that of a woman demonstrates that she was a very prominent woman indeed, someone who must have engaged in business and legal activities that necessitated her owning a seal. This is most impressive and certainly the most important component of these new finds.
Christopher A. Rollston, George Washington University (rollston@gwu.edu)


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.

Wife of Jesus Coptic Papyrus: Brief Methodological Musings

10 April 2014


The first time that I saw a good image of this papyrus (early fall of 2012), I was suspicious about its authenticity and told my wisdom lit class (which happened to ask about it one evening, during a class in the early fall of 2012) that I believed it to be a probable modern forgery….the ink and the script just didn’t look right to me. That’s still the way that I feel. I have worked in Sahidic Coptic for ca. twenty years, having studied it formally during my doctoral program and having taught it recently at the graduate level. Moreover, I’ve read pretty heavily in the Nag Hammadi Corpus as well. And Hellenistic Greek is also in my wheelhouse, as are ancient inscriptions in general.

As for the laboratory tests, the carbon tests on the papyrus demonstrate that the papyrus is ancient. That’s no surprise. Just as modern forgers of ostraca use ancient pottery sherds for their logia, so also a modern forger of a papyrus inscription would use some ancient papyrus (which, although certainly not as readily available as pottery, is still available….with or without ancient ink). It should be emphasized that papyrus was often reused (note the phenomenon of palimpsests) and so the putative date for the papyrus itself (prior to the Common Era) is not an important issue at all, neither for authenticity, nor against authenticity. It is an absolute non-issue. Also, the fact that the “ink” used on this is consistent with the chemical composition of ancient ink is also not necessarily evidence for antiquity. After all, the chemical composition of ancient “ink” has been known for some time and the chemicals available in antiquity are certainly still available today. Thus, the chemical composition of the ink is not necessarily an argument in favor of authenticity. Also, it is also possible for someone to scrape off (e.g., from a papyrus) ancient ink from the words of some mundane ancient inscription….and then add a little water to the dried ink which had been scrapped off and then resuse the ink. Some people (including some scholars) assume that modern forgers are not all that bright (and thus would not be that clever in forging something). In contrast, I believe that modern forgers (at least from the final quarter of the 20th century and on) are quite sharp…..and for good reason they try to be very clever: after all, there is much money to be made and modern forgers knows this….so, as for this piece, I remain very suspicious of its authenticity. Perhaps it’s ancient….but I doubt it.

I should also note that if it turns out that Jesus of Nazareth was married (i.e., if some good, credible, first century evidence comes to light at some point in the future and demonstrates this), that sounds fine to me. In other words, I have no theological objections to this sort of thing. My concerns about this papyrus (which is later than the first century CE anyway), are only epigraphic in nature. I have no vested interest one way or the other.

Sincerely, Christopher Rollston

The James (Ya’akov) Ossuary: The Kalman Interview at the Albright Institute of Archaeological Research in Jerusalem (Video)

6 January 2014


The broadcast in its original airing may be viewed .

The Ninth Century ‘Moabite Pedestal Inscription’ from King Mesha’s Ataruz: Preliminary Synopsis of an Excavated Epigraphic Text and its Biblical Connections

17 December 2013

Rollston Collating Khirbet Ataruz Moabite Pedestal Inscription.

Rollston Collating Khirbet Ataruz Moabite Pedestal Inscription.

The Ninth Century ‘Moabite Pedestal Inscription’ from King Mesha’s Ataruz: Preliminary Synopsis of an Excavated Epigraphic Text and its Biblical Connections

Christopher A. Rollston
National Endowment for the Humanities Research Scholar
Albright Institute of Archaeological Research, Jerusalem
e-mail: Christopher.Rollston@gmail.com


The site of Khirbet Ataruz is located in modern Jordan, on the ridge of Jebel Hamida, with Wadi Zarqa Ma‘in to the north and Wadi Wala to the south. Khirbet Ataruz is located some fourteen kilometers to the northwest of Dhiban. The site has been known for some time (e.g., Glueck 1939; Schottroff 1966; Timm 1980; Niemann 1985), but Chang-Ho Ji is the first to conduct full scale excavations at the site (Ji 2012). During the process of these ongoing excavations, Ji discovered a pedestal (arguably of an incense altar) with an inscription on it (cf. the inscribed incense altar from Mudeyineh, Dion and Daviau 2000). Significantly, the archaeological context for this inscription from Ataruz was an Iron Age II Temple, with a striking assemblage of cultic objects (Ji 2012; cf. also Finkelstein and Lipschits 2010; Finkelstein and Lipschits 2011).

Shortly after the initial discovery, Chang-Ho Ji requested that I analyze and publish the inscription and I accepted this gracious invitation. Moreover, because of the complex nature of this inscribed pedestal, particularly, the presence of several sets of hieratic numerals, I have since brought Stefan Wimmer and P.Kyle McCarter in to assist with the publication. The editio princeps will be completed during 2014, co-authored by the three of us, with formal publication probably appearing during late 2014 or early 2015 (e.g., ADAJ and also something such as Levant, ZDPV, BASOR, or Maarav). This preliminary synopsis, however, is something I have written, based primarily on the document that I previously submitted to the Department of Antiquities of Jordan when I agreed to publish it. I wish to express again my gratitude to the Department of Antiquities of Jordan and to Excavator Chang-Ho Ji for permission to publish it.

Prologue to the Discussion of the Ataruz Moabite Pedestal Inscription: The Mesha Stele and the Bible

The Moabite site of Atarot (Ataruz) is prominently mentioned in the Mesha Stele, along with several additional sites (Dearman 1989). Namely, after four lines of introductory material (including reference to King Mesha’s dedication in Qarḥoh of a high place for Kemosh, the national God of Moab; cf. 1 Kgs 11:7), King Mesha, son of King Kemoshyat, states that because Kemosh was angry with his land (Moab), he (Kemosh) gave the land of Moab into the hand of King Omri of Israel (r. 876-869). Mesha then continues, noting that Omri’s son (bn) succeeded him (Omri) and wished to continue to maintain hegemony over Moab, but Mesha rebelled. Several sites are selected for particular emphasis in this section of the Mesha Stele, notably Madaba, Atarot, Nebo, and Yahaṣ. Regarding the land of Madaba, Mesha states that “Omri had taken possession of all [the land] of Madaba, and he dwelt in it during his days and half of the days of his son, (around) forty years. But Kemosh returned it in my days.” Regarding the land of Atarot, Mesha states that “Now the people of Gad had dwelt in the region of Atarot for a long time, as the king of Israel had built Atarot for them [cf. Num 32]. But I fought against the city and I took it and I killed all the people of the city, (as) it was for Kemosh and for Moab. And I brought back from there the altar hearth…and I [drag]ged it before Kemosh in Qiryat. And I settled the Sharonites and the Maḥarites in it.” At that point, the narrative continues with Mesha’s conquest of Nebo, including its “seven thousand [Israelite] warriors, as well as male sojourners (grn), women, female sojourners, and young maidens.” All of these Mesha, “devoted to ‘Aštar of Kemosh.” Moreover, Mesha took from Nebo “the vessels of Yahweh and dragged them before Kemosh.” Regarding Yahaṣ, Mesha states that “Now the King of Israel had built Yahaṣ and he dwelt in it while he was fighting against me. But Kemosh drove him out from before me”(Rollston 2015). Much of the remainder of the Mesha Stele is devoted to the public works of King Mesha of Moab, coming on the heels of his military successes against the Northern Kingdom of Israel.

It should be remembered that the language of the Mesha Stele is Moabite, with the consistent use of the nun throughout this inscription to mark the masculine plural, thus, distinguishing the Moabite language from the Old Hebrew language (and also from Phoenician, etc.). Furthermore, the phonology of the Mesha Stele distinguishes it from the Aramaic language and from the Deir Alla Dialect (Garr 1985; Huehnergard 1989). That is, the Mesha Stele is written in the Moabite language and this is not something that can be disputed on linguistic grounds (pace Segert 1961).

Within the Hebrew Bible, there is a similar account of Omride hegemony over Moab and Mesha’s subsequent rebellion, with reference to (an annual) tribute from Moab to Israel of 100,000 lambs and the wool of 100,000 rams (2 Kgs 3:4-27). The Israelite account in Kings and the Moabite account in the Mesha Stele are different in some respects, though. (1) For example, the Israelite account states that Mesha rebelled after the death of King Ahab of Israel (r. 869-850 BCE), namely, during the reign of Ahab’s son, King Jehoram of Israel (r. 849-842 BCE), the brief reign of King Ahaziah of Israel (r. 850-849 BCE) not part of the equation. The Moabite account, however, suggests that this rebellion occurred after Omri’s death and, thus, during the time of Omri’s son (Ahab). (2) Furthermore, the Israelite account in Kings states that King Jehoram of Israel was successful in crushing the rebellion of Mesha (but with the human sacrifice of his [Mesha’s] son and the ensuing wrath of Yahweh or Kemosh precipitating an Israelite military withdrawal). However, the Moabite account in the Mesha Stele declares that Mesha was very successful in the rebellion against Israel and won major victories against Israel, regaining lost Moabite territory, and even gaining new territory. (1) The first difference (i.e., son of Omri in the Mesha Stele over against son of Ahab in Kings) is not necessarily much of a tension because the term “son” could be used in a broader sense to refer to a son or grandson of Omri (cf. also Rollston 2010, 52-55). Nevertheless, it is also possible that there is a real tension and that the Israelite account or the Moabite account got it wrong. (2) The second difference (the Israelite account claiming victory over Moab and the Moabite account claiming victory over Israel) could be understood as a demonstration of the fact that in the ancient Near East kings preferred to proclaim victories, regardless of the precise facts and actual outcomes. Note, for example, that both Ramesses II and the Muwatalli II claimed victory against the other in the Battle of Kadesh in 1274 BCE. Conversely, it could be that the Moabite account and the Israelite account are of different military episodes or different victories in the longer conflict. It should also be mentioned here that the events described in the Israelite account would be dated to ca. 849 BCE (i.e., the time when Jehoram was on the throne of Israel and Jehoshaphat was on the throne of Judah).

Finally, it should also be emphasized that the repeated campaigns of King Shalmaneser III (r. 859-825 BCE) of Assyria into the Levant during the middle of the 9th century (cf. Kurukh Monolith, Black Obelisk) arguably provided the Moabites with some relief from Israelite hegemony, probably contributing to some of Mesha’s military successes and public works. Of course, the Aramaic (Tel Dan; Biran and Naveh 1995) and Hebrew (2 Kings 8-9; cf. 1 Kgs 19:17) narratives about the deaths of King Jehoram of Israel and King Ahaziah of Judah around 842 BCE (and thus the usurpations of King Jehu of Israel and Hazael of Damascus) are also part of the broader tableau as well.

The Tell Ataruz Moabite Pedestal Inscription

The inscribed pedestal contains seven lines of text incised into the stone (of the pedestal). Four lines of the inscription are written along the vertical axis and three lines are written along the horizontal axis. This inscription was not written by one hand. The inscription is abraded in some places. Based on the morphology, stance, and ductus of the script, I date the writing comfortably to the 9th century BCE. It would be most difficult to date the script of this pedestal to the 8th century. After conveying my palaeographic date of this inscribed pedestal to Ji, he indicated to me that he dated the temple and its associated finds to the 9th century as well. Also of substantial import, Finkelstein and Lipschits also have indicated that the architecture of this component of this site dates to the 9th century (2010; 2011), and they have indicated to me (personal conversation) that the pottery and objects can be dated to this chronological horizon as well.

It should be emphasized that during the 9th century BCE, the script used to write inscriptions in the Moabite language was the Old Hebrew script, arguably a fact related to the hegemony of Northern Kingdom of Israel during the Omride period (Naveh 1987, 65; Rollston 2010, 54). Of course, this hegemony is something that King Mesha of Moab discusses in his stele. Nevertheless, the Moabite script developed further during succeeding chronological horizons into an independent national script, that is, the distinctive Moabite national script, something demonstrated very nicely by the script of the Mudeyineh Incense Altar Inscription (Dion and Daviau 2000; Rollston 2010, 62-63).

The Khirbet Ataruz Moabite Pedestal Inscription is written in the Moabite language. The linguistic markers in the Ataruz inscription correspond with those of the Mesha Stele. Most prominently and importantly, the masculine plural marker used in the Ataruz inscription is the nun, rather than the mem of Phoenician and Hebrew (and congeners). The fact that this inscription is written in the Moabite language is not surprising, based on the location of this site and the historical details about this site that are contained in the Mesha Stele (namely, Mesha of Moab conquered this territory, establishing Moabite hegemony). Of substantial importance in this inscription is the presence of some rather impressive hieratic numerals, some of which are particularly high in numeric value (in the thousands). Word dividers do occur in this inscription, as expected, but not with absolute consistency (this too is quite customary).

The structure of the initial line of the inscription is a hieratic numeral, followed by /mn/ and then an additional lexeme. A similar structure is present in line three and arguably (a truncated version) in line two as well. There are a finite number of possibilities for the semantic domains of the words that are present, but the brevity of the inscription does complicate the matter. In any case, the structure is numeric + mn + lexeme. Because this inscription was found in a temple context, I believe that it is reasonable to propose that the numeric refers to the amount of a commodity offered. The word /mn/ is arguably the prepositional morpheme “from” (I am less inclined to think that it is a reference to a “mina,” that is, a unit of measure). The prepositional /mn/ does not occur in all cases (it is clearly present in lines one and three). The lexeme following /mn/ can be understood to refer to the source of the offering. It is plausible to understand the source (i.e., the lexeme after the word “from”) as a person, a category of persons, a place (and there is some semantic bleeding and overlap with regard to these enumerated categories). In any case, the first line seems paradigmatic in this respect. The hieratic numeral is followed by “from” (mn) and after the word divider the consonants present are /‘brn/, with nun the masculine plural marker in Moabite. The lexeme after /mn/ in the first line could refer, for example, to a category of people “those across” (e.g., a wadi, or some geographic feature or territory), or a people group (e.g., “Hebrews”), or something such as an eponymous ancestor of a clan. Lexical and textual support can be found for all of these (e.g., Deut 32:49; Num 27:12 for the mountainous district in northwestern Moab and associated with Nebo; a Gadite chief mentioned in 1 Chr 5:13; and the common gentilic “Hebrews” attested in the Pentateuch, Deuteronomistic Literature, and the Prophets). As for the second line, there is a hieratic numeral and then the consonants /grn/, with nun arguably as the masculine plural marker (although /grn/ “threshing floor” must also be considered). It should be mentioned that /prn/ is also a possible reading (palaeographically), which could, of course, refer to a category of offering, “bull” or the like. I should note that reading a dalet here (instead of a resh) is not really plausible. The absence of the /mn/ “from” in such a terse text as this does not necessitate the supposition that the structure of this line is entirely different from the first line and third line, but that possibility must be considered. The fact that /grn/ (“sojourners”) occurs in the Mesha Stele is of some importance, of course. Also of import, line two is rather long and arguably contains the lexemes gdl (reading a /p/ for the first letter rather than a /g/ is palaeographically possible; note that I am disinclined to read the second letter as a possible /r/ because of the shortness of the stem) and šlm (with the various lexical possibilities for this sequence of letters, including those with the semantic domain of “pay,” and those with the semantic domain of “peace offering,” etc.). As for the third line, the preposition “from” recurs. The fourth line is difficult, but reasonably legible in the best of the photographs at our disposal. The final three lines of the inscription (or, one could argue, the first three lines of the inscription) are different in terms of formula, and perhaps most interesting mathematically. More precisely, in lines five and six, the formula is: hieratic numerals + /sh/ and /L/, arguably a shortened form of the word for sheqel. The seventh line contains /kl/ and then after this a hieratic numeral as well.

At this juncture, we now have access to additional (and better) photographic images of this important inscription and we will begin working in earnest on this text in the coming months. Kyle McCarter and I have most (probably all, I suppose) of the lexical and syntactic options on the table (not all of which are included here, of course) and Stefan Wimmer has reasonable and cogent readings for the hieratic numerals (and he is also a fine scholar of Northwest Semitic in his own right). Because this inscription hails from the period very shortly after Mesha’s conquest of the site (remember that the inscription is written in the Moabite language, a fairly diagnostic indicator of those controlling the site at the time of the production of the inscription, I believe), the inscription provides useful and important data for attempts to reconstruct the nexus of power in the second half of the 9th century. In addition, this inscription lends credence to the claims Mesha makes in his stele (e.g., his conquering of Atarot, etc.), something that may suggest that the materials in Kings about the victory of Jehoram over Moab were either legendary or short-lived, or about a slightly different chronological horizon. The reflections detailed here will not necessarily be those embraced in the final publication, but they do convey the interpretive contours that have been part of my preliminary reflections and will probably be formative for the final publication.

Basic Bibliography

Biran, Avraham and Naveh, Joseph.
1995 “The Tel Dan Inscription: A New Fragment.” IEJ 45:1-18.

Dearman, Andrew, ed.
1989 Studies in the Mesha Inscription and Moab. Atlanta: Scholars Press.

Dion, Paul E. and Daviau, P.M. Michèle
2000 “An Inscribed Incense Altar of Iron Age II at Ḫirbet el-Mudēyine (Jordan).” ZDPV 116: 1-13.

Finkelstein, Israel and Lipschits, Oded.
2010 “Omride Architecture in Moab: Jahaz and Ataroth.” ZDPV 126: 29-42.

Finkelstein, Israel and Lipschits, Oded.
2011 “The Genesis of Moab: A Proposal.” Levant 43: 139-152.

Garr, Randall.

1985 Dialect Geography of Syria-Palestine, 1000-586 B.C.E. Philadelphia: University of Pennsylvania.

Huehnergard, John.

1989 Remarks on the Classification of the Northwest Semitic Languages. Pages 282-93 in The Balaam Text from Deir ‘‘Alla Re-evaluated. Proceedings of the International 84 Symposium held at Leiden 21–24 August 1989. Ed. J. Hoftijzer and G. van der Kooij; Leiden: Brill.

Ji, Chang-Ho
2012 “The Early Iron Age II Temple at Khirbet Atarus and Its Architecture and Selected Cult Objects.” Pp. 203-221 and Plates 44-49 in Temple Building and Temple Cult: Architecture and Cultic Paraphernalia of Temples in the Levant (2m-1. Mill. BCE), ed. Jens Kamlah. Wiesbaden: Harrassowitz.

Naveh, Joseph.
1987 Early History of the Alphabet: An Introduction to West Semitic Epigraphy and Palaeography. 2nd ed. Jerusalem: Magnes.

Niemann, Hermann Michael
1985 “Ein Statuettentorso von der Ḫirbert Aṭārūs.” ZDPV 101: 169-177.

Rollston, Christopher A.
2010 Writing and Literacy in the World of Ancient Israel: Epigraphic
Evidence from the Iron Age. Atlanta: Society of Biblical Literature.

Rollston, Christopher A.
2015 Northwest Semitic Royal Inscriptions: Writings from the Ancient
World. ed., Theodore Lewis. Atlanta: Society of Biblical Literature.

Schottroff, Willy.
1965 “Horonaim, Nimrim, Luhith und der Westrand des „Landes Ataroth“:
Ein Beitrag zur historischen Topographie des Landes Moab.” ZDPV 81: 163-208.

Segert, Stanislav
1961 “Die Sprache Der Moabitischen Königsinschrift.” Archiv Orientální:
29: 197-267.

Timm, Stefan
“Die territoriale Ausdehnung des Staates Isael zur Zeit der Omriden.” ZDPV 96: 20-40.

Wimmer, Stefan
2008 Palästinisches Hieratisch die Zahl- und Sonderzeichen in der althebräischen Schrift. Wiesbaden: Harrassowitz.

Chapters in the History of Modern Forgery: Professors Leopold Messerschmidt and Charles Torrey on a Hebrew Inscription

2 November 2013

This post is a draft of a selection from my (forthcoming) volume entitled, Forging History in the Biblical World: Textual Forgeries from the Ancient and Modern Middle East, Medieval Europe and the New World


Professor Leopold Messerschmidt authored an article for Berlin’s Orientalistische Litteratur-Zeitung, a particularly prestigious German journal. The year was 1903. The article was about an inscribed piece of clay, about two inches in diameter, and round. The script of the inscription was written in ancient Hebrew, but poorly formed. Messerschmidt, however, could read it. He was a distinguished Semitist, well-known for his work in Assyriology, and with important contributions to the field of Hittite as well. The inscription was owned by Paul Mauersberger. He had purchased the inscription near a Jerusalem train, reportedly for a mere trifle. Quite understandably, he had brought the inscription to Leopold Messerschmidt because he wished for someone to read it.

The preceding decades had witnessed the production of hundreds of forged inscriptions, most from the antiquities market, but scores had actually been salted in archaeological tells. Shapira’s “Moabite Forgeries” were among the most notorious, but there were countless additional forgeries during the final decades of the 19th century. Therefore, caution was the modus operandi for most scholars, especially for those working on inscriptions in the languages such as Hebrew, Phoenician, and Aramaic, as well as for those working in Greek and Latin. Vigilance was mandatory. Messerschmidt had some concerns about the authenticity of this inscription, but he reasoned that it might just be ancient. He entitled his article “Fälschung?” (Messerschmidt 1903).

Because Messerschmidt was not a Hebraist, he contacted one of the most esteemed and prolific scholars of Northwest Semitic at the time, Mark Lidzbarski (1868-1928). Although still quite a young scholar, Lidzbarski had published the most authoritative collection of Northwest Semitic inscriptions, Handbuch der Nordsemitischen Epigraphik (1898), completed during his time in Kiel and still a useful compendium today. Lidzbarski had been born into an Hasidic Jewish family. His given name was Abraham Mordechai Lidzbarski, but he had converted to Christianity and so changed his given name. Lidzbarski’s verdict on Messerschmidt’s inscription was quite decisive: “Das Plättchen is nach meiner Ansicht sicher eine Fälschung,” a forgery modeled on the coinage of Hasmonean King John Hyrcannus (r. 134-104 BCE).

Although the inscription was brief, just five or six words on four short lines, Lidzbarski noted both palaeographic and orthographic problems. In fact, on this inscription the name “Yohanan” and “Jews” were both misspelled. Letters were grossly malformed. But Messerschmidt was not convinced that it was a forgery. After all, he wondered “why, if it were a forgery, would someone ever attempt to forge a coin out of clay?” And he also noted that this inscription was bought for a trifle. And in addition, its previous owner, the poor workman from whom Mauersberger purchased it, was not actually attempting to sell it. It was Mauersberger who initiated the sale. Do not these things, Messerschmidt asked, “gegen eine Fälschung zu sprechen?” That is, “Do these things not speak against it being a forgery” (Messerschmidt 1903, 241).

But even F.E. Peiser, the editor of the journal Orientalistische Litteratur-Zeitung in which Messerschmidt’s article appeared, appended a footnote at the bottom of the page on which Messerschmidt’s article concludes. Peiser succinctly and decisively states that “This clay disk is a part of a forger’s apparatus” basically, a modern mold which had been formed to produce forged metal coins. Messerschmidt, however, was not at all certain that Lidzbarski and Peiser were correct. Presumably his concern was that he did not wish to declare prematurely an important ancient inscription to be a modern forgery. In certain ways, some might reasonably suggest that his caution is commendable.

Charles C. Torrey of Yale University, however, was astounded that Messerschmidt might even attempt to consider this inscription ancient. Writing in the lofty Journal of the American Oriental Society, Torrey entitled his article “On a Palestinian ‘Forgery’” (Torrey 1903). He began his article with these words: “It is a disk of baked clay, about two inches in diameter, reproducing very clumsily and on a much enlarged scale a well-known coin of John Hyrcannus. Nothing more be said, of course, as to the value of this ‘antique’; it does not even deserve to be taken so seriously as would be implied in giving it the name ‘forgery.’”

Torrey then continues, however, writing that “The fact is, this is one of a class of objects not infrequently hawked about the streets of Jerusalem by certain vagabonds of a familiar type—half beggar, half rascal. The things are made by pressing clay into forms which some idler has amused himself by fashioning. The conditions which produce such works of art as this are a little spare time, a sense of humor, and the remote possibility of gulling some brother rascal, or perhaps even a tourist. It would take perhaps an hour to whittle out of wood such a form as the one from which this ‘coin’ was made. I have frequently been offered just such discs in Jerusalem, the would-be vender always accompanying his offer with a broad grin. One of these objects now in my possession (a clay disc, about two inches in diameter, pressed from a form) bears a representation of Eve and a serpent, with a few meaningless letters appended. Apparently there was never a thought of getting more than a few paras each for these ‘inscriptions.’ It is not surprising, then, that the native workman mentioned in this case did not show any great eagerness to turn his property into money.” Then Torrey concludes his article with this sentence, “As for this worthless Palestinian trinket, it is certainly a misuse of language to call it a ‘forgery’” (Torrey 1903, 209-210).

It would be permissible to suggest that Torrey’s retort is harsh. Indeed it is. But I suppose that it would also be permissible to suggest that Messerschmidt should have listened to Lidzbarski’s reasoned and detailed demonstration (all provided in a letter from Lidzbarski to Messerschmidt) that this inscription is a forgery of particularly modest quality. Messerschmidt, however, persisted. Often in the long history of forgery, Messerschmidt would have found good company.

Christopher A. Rollston
National Endowment for the Humanities Research Scholar
Albright Institute of Archaeological Research

The Decipherment of the New ‘Incised Jerusalem Pithos’

11 July 2013

The Decipherment of the New ‘Incised Jerusalem Pithos’


Christopher Rollston


During recent excavations in Jerusalem, Eilat Mazar of Hebrew University discovered a large pithos with incised linear alphabetic Northwest Semitic letters. According to press reports, the discovery will be published “in a paper by Dr. Mazar, Prof. Shmuel Ahituv of Ben-Gurion University of the Negev, and Dr. David Ben-Shlomo of the Hebrew University, following their extensive research on the artifact. Prof. Ahituv studied the inscription and Dr. Ben-Shlomo studied the composition of the ceramic materials. The paper, “An Inscribed Pithos From the Ophel,” appears in the Israel Exploration Journal 63/1 (2013).”

“According to Prof. Ahituv, the inscription is not complete and probably wound around the jar’s shoulder, while the remaining portion is just the end of the inscription and one letter from the beginning. The inscription is engraved in a proto-Canaanite / early Canaanite script of the eleventh-to-tenth centuries BCE, which pre-dates the Israelite rule and the prevalence of Hebrew script.”

“Reading from left to right, the text contains a combination of letters approximately 2.5 cm tall, which translate to m, q, p, h, n, (possibly) l, and n. Since this combination of letters has no meaning in known west-Semitic languages, the inscription’s meaning is unknown.”

“The archaeologists suspect the inscription specifies the jar’s contents or the name of its owner. Because the inscription is not in Hebrew, it is likely to have been written by one of the non-Israeli residents of Jerusalem, perhaps Jebusites, who were part of the city population in the time of Kings David and Solomon.”
End of Citation of Press Reports


A. Palaeographic Assessment.

The script of this inscription has been accurately described by my friend Shmuel Ahituv as “Proto-Canaanite.” Sometimes this script is referred to as Proto-Sinaitic (because of the prominence of the Serabit el-Hadm Inscriptions) and sometimes this script is referred to as Early Alphabetic. I prefer the term Early Alphabetic, following my teacher, P. Kyle McCarter, Jr. In any case, suffice it to say that I am in complete agreement with Ahituv.

Some scholars have stated in blogs that this inscription is perhaps written in the Phoenician script, not the Early Alphabetic Script. Technically speaking, however, the term “Phoenician” would be problematic for the script of this Incised Jerusalem Pithos. Note the following: (1) The Early Alphabetic Script could be written sinistrograde (right-to-left), dextrograde (left-to-right), boustrophedon (i.e., consecutive lines written from left-to-right then right-to-left, etc.), and in columns (i.e., vertically); (2) The stance of the letters in Early Alphabetic writing was not fixed. That is, the letters could “face” different ways and the letters could be written with dramatically different degrees of rotation, even within the same inscription; (3) Within the Early Alphabetic script, the number of consonsantal graphemes (“letters”) could be, and often was, higher than twenty-two graphemes.

However, during the terminal horizons of the second millennium several developments occurred in linear alphabetic Northwest Semitic. Namely, (4) the direction of writing became fixed: it was consistently sinistrograde (right-to-left); (5) The stance of the letters became more stabilized and standardized, with trained scribes normally “facing” the letters in the same direction, and without dramatic variations in the degrees of rotation; (6) the number of consonantal graphemes was reduced to twenty-two. For many decades, the established convention within the field of palaeography has been to refer to this script (reflecting these three changes), at this time period, as Phoenician. Joseph Naveh’s statement continues to reflect the consensus of the field, namely, the transition from Early Alphabetic to Phoenician “took place in the mid-eleventh century B.C. (Naveh, Early History of the Alphabet, Jerusalem: Magnes Press, 1987, 42; Rollston, Writing and Literacy in the World of Ancient Israel: Epigraphic Evidence from the Iron Age, Atlanta: SBL, 2010, 19).

Significantly, (1) the script of this new “Iron Age Incised Jerusalem Pithos” reflects the (varied) stance of letters in the Early Alphabetic script (e.g., the stance of nun is precisely the reverse of the normal stance in the Phoenician alphabet and its congeners in the National Scripts from later periods). (2) Moreover, this inscription is decipherable and it is written dextrograde, that is, from left-to-right. As noted, Phoenician was consistently written sinistrograde. Thus, there is is sufficient evidence to state that the script of this inscription should be classified as Early Alphabetic (i.e., “Proto-Canaanite”), not Phoenician. In short, Ahituv’s classification is certainly correct.

In terms of the palaeographic date for this inscription, I would be most comfortable with the 11th century BCE, rather than the 10th century BCE. The primary reasons for this dating are (1) the direction of writing (dextrograde); and (2) the fact that the five strokes of mem are of approximate equal length (the fifth stroke is elongated already in Ahirman, but note Azarba‘al, which is slightly earlier than Ahiram). I wish there were more letters in this inscription, but there are not. In any case, based on the palaeographic evidence at hand, a date in the 11th century seems to me to be reasonable. For detailed discussion of the script of this period, see McCarter, The Antiquity of the Greek Alphabet and the Early Phoenician Scripts, Missoula, Scholars Press, HSM 9, 29-63; Rollston, “The Dating of the Early Royal Byblian Phoenician Inscriptions,” Maarav 15 (2008): 57-93 and the bibliography cited in these two sources. In other words, I date this inscription prior to the rise of David and Solomon. Thus, again, I find myself in happy agreement with Professor Ahituv.

B. Readings

Inscriptions in Early Alphabetic are often (as noted) written in dextrograde and I would contend that this new Incised Jerusalem Pithos is written in this fashion, that is, from left-to-right. Here are my readings: mem, qop, lamed, het, nun, [re]sh, sh[in]. My readings, therefore, differ to some degree, with those of the authors of the editio princeps. Also, I would emphasize additional matters of ductus, namely, the horizontals of the het were made after the verticals (evidenced by the pattern which I term “damming”).

For the morphology and stance of the letters, I would draw the reader’s attention especially to W.F. Albright’s chart of Early Alphabetic in his volume entitled The Proto-Sinaitic Inscriptions and Their Decipherment, HTS 22, Cambridge, Harvard University Press, 1966. This chart is conveniently reproduced in Joseph Naveh’s Early History of the Alphabet, 25. In addition, very useful is the chart in an article by Frank Moore Cross entitled “Newly Found Inscriptions in Old Canaanite and Early Phoenician Scripts, BASOR 238 (1980): 1-20, conveniently reproduced in the collection by Cross entitled “Leaves from an Epigrapher’s Notebook: Collected Papers in Hebrew and West Semitic Palaeography and Epigraphy, HSS 51, Winona Lake: Eisenbrauns, 2003, 213-230. See also the drawing of the (alphabetic) Tell Fekhariyeh inscription in Ali Abou-Assaf, Pierre Bordreuil, and Alan R. Millard, La Statue de Tell Fekherye et son inscription bilingue assyro-arameenne, Paris, Editions Recherche sur les civilizations, 1982.

Perhaps most important is the fact that the letter which the authors of the editio princeps read as a peh, I read as a lamed. I would draw attention in particular to the forms of lamed in Albright’s chart. In addition, I would draw the reader’s attention to the form of lamed in Tell Fakhariyeh. The form in this new Incised Jerusalem Pithos is very similar to that of Tell Fekhariyeh, but facing in the opposite direction. This is quite understandable, as Fakhariyeh is written sinistrograde and the Jerusalem Incised Pithos is written dextrograde. That is, in both cases the “hook” of the letter is facing in the direction of writing. Notice, of course, that the nun is written in the same fashion (i.e., “facing” in the opposite direction from that which we would expect from the later, standardized Phoenician script). I can understand the reasons for the desire on the part of the authors of the editio princeps to wish to read a peh (or even a gimel), but reading a lamed here is much preferable, from my perspective, especially since this reading yields a good lexeme, that is, a comprehensible reading…and in addition, it is palaeographically permissible and reasonable (some have attempted to read this letter as a sade and do not find that reading cogent). I am happy to note that shortly after I mentioned that I differed with the authors of the editio princeps with regard to some of the readings, my dear friend Anat Mendel mentioned (independently) that she felt this letter could be read as a lamed. Because I believe Mendel has a very good “eye for form” (to use the term Frank Cross coined for palaeographers), I take this as a good sign that the readings herein proposed is reasonable.

C. Translation

The root present in this inscription is qop,lamed,het, that is the word “pot,” or “cauldron” (cf. 1 Sam 2:14; Micah 3:3, with this noun attested in the Bible with the feminine marker tav of Hebrew). I must admit that “pot” is not the most sensational of readings, but I do believe it to be the correct reading. Significantly, Thomas Lambdin discussed this word in some detail in “Egyptian Loan Words in the Old Testament,” JAOS 73 (1953): 149, noting that it was attested in Egyptian “since the Old Kingdom.” Someone might counter that it would not make sense to have the word “pot” written on a pot, as it would be a tautology, of sorts. However, I would note (for example) that the word “nbl” (“pot,” “vessel”) is very well attested in the Reisner Samaria Ostraca (e.g., prior to the reference to the commodity, such as “wine” or “olive oil”). As for the mem which precedes the qop, lamed, het on the Incised Jerusalem Pithos, I would propose that this could be part of the formation of this noun (as many nouns are formed with a prefixed mem, of course) or I suppose that it could be the preposition min with an assimilated nun. I have a slight preference for the former, I suppose. The nun that follows the two-bar het could conceivably be a plural marker (as in certain dialects of Northwest Semitic), but because of the context, I prefer to consider this noun to be singular, “pot,” rather than “pots,” and thus I consider the nun to be the first letter of the next word, perhaps followed by a resh (thus yielding, “Ner”). Of course, Ner is an attested personal name (cf. even 1 Sam 14:50, Abner son of Ner, the commander of Saul’s army, although I would certainly not propose an identification of these biblical and epigraphic figures on the basis of the evidence at hand, especially since the reading “Ner” is not certain), and so in this case (if this reading and rendering are accepted), this portion of this inscription would read “Pot belonging to Ner” (nb: someone might propose that there was a dalet in the lacuna after the proposed resh, thus, the inscription would refer to a “pot of nard.” I do not find that impossible, although someone might counter that a pithos would be a rather large amount of this precious commodity). There is a preserved portion of a letter at the far right side of this potsherd. Someone could propose that it should be read as a nun. I do not think so, however. Rather, I would propose that it is probably a shin, turned on its side, as it were, something which is common enough in Early Alphabetic.


This inscription is written in the Early Alphabetic script. The inscription is written dextrograde. I am most comfortable with a date in the 11th century BCE. The extant lexeme on this inscribed pithos is arguably the word “pot.” The word “pot” may have been followed by a personal name, such as “Ner” or (perhaps) a commodity such as “nard.” I prefer the assumption that it is a personal name. In terms of the language of this inscription, the language is certainly Northwest Semitic, and I would suggest that it is methodologically safest to posit that the language is Canaanite (as the script is Early Alphabetic, that is, Proto-Canaanite). However, I think that someone could propose that the language is Phoenician. In fact, I also believe that it is linguistically possible (but perhaps slightly more difficult) to argue that the language is Old Hebrew. Ultimately, however, as is often the case, there is no diagnostic element present which allows us to draw a firm conclusion. For this reason, the matter of the precise dialect of Northwest Semitic must be left open for this inscription, but my belief is that “Canaanite” is the best way to refer to the language of this incised pithos.

Finally, I should like to conclude by stating that I believe this is a nice inscription, important in various ways. Of course, I personally would be very disinclined to “build a kingdom upon this potsherd.” But I would wish to state that this is an inscription that fits nicely into, and augments, the totality of our epigraphic evidence for the early Iron Age, and I congratulate Eilat Mazar on this find and I congratulate Shmuel Ahituv and David Ben-Shlomo and Eilat Mazar on its publication in the pages of Israel Exploration Journal.


ADDENDUM I: Gershon Galil has indicated to me that he reads this inscription as sinistrograde, yielding, of course, the “portion,” “part” root. This is a plausible reading, but because reading this inscription as dextrograde accounts so nicely for the direction of both the lamed and the nun (i.e., the way which these letters are facing), and because the word “pot” followed by a personal name or a commodity works so nicely, I continue to prefer to read this inscription in this fashion. I thank Gerson Galil for bringing his reading to my attention. Sincerely, Chris Rollston


Addendum II, Sunday, July 14.

Quick notes, replying to some queries and comments.

(1) The reason for my preference for the 11th century is the fact that the script is Early Alphabetic, not Phoenician. The Phoenician script, as we know it, was developed in the 11th century, and from that point on, that script (i.e., Phoenician script) spread rapidly and it became the MutterSchrift of the subsequent (daughter) national scripts: Old Hebrew and Aramaic. It is not impossible that there were two parallel script traditions and that the Early Alphabetic script and the Phoenician script lived side-by-side for some time. At this point, however, the evidence need not be understood in this way. I suspect that there were pockets where conversion from Early Alphabetic to the Phoenician script took longer. (2) The fact that the find may have been, or was, a 10th century context does not preclude the possibility that the pot itself was from the 11th century. Usage periods can be longer than we might presuppose, even for pots. (3) Furthermore, I think that even if a pottery form is well attested in a particular horizon, there can be, and often are precursors to it. Within the field of palaeography we call these harbinger forms. The same thing can happen with pots..that is, a form that is predominant in one period can have very similar, or even identical ancestors in a preceding horizon. (4) As for the notion of heirloom pots…I did not indicate that this is the way that I personally felt. Rather I stated that it was possible. From my perspective, possible and probable are two very different things. (5) Finally, with regard to the 11th century vs the 10th century….we’re talking about a few decades. That’s it. I don’t think that pottery typology (and the bases for it) or palaeographic typology (or the bases for it) are so precise (in terms of absolute dates) that there is much to argue about here. We’re talking about a few decades. That’s it. (6) Final comment….someone queried about whether or not I knew the Serabit script well (an amusing comment to say the least). Suffice it to say, that, yes, I have known it well for a very long time now.

With all best wishes,



9 April 2013


Christopher A. Rollston
Visiting Professor of Northwest Semitic Languages and Literatures
George Washington University
Washington, D.C.


One of the most auspicious developments in “forgery detection” during recent years has been the development and usage of laboratory methods to attempt to determine deauthenticity. Most welcome, therefore, is the recent report (e.g., the article by Stephanie Pappas on the Huffington Post) which discusses the laboratory tests on the Gospel of Judas. It seems to me that this is very useful and I think that these results bode well for that particular document (especially because National Geographic’s “Standards and Practices” people are among the very best, and they clearly maintained strict protocols with regard to the laboratory analyses of the Gospel of Judas). Moreover, it has also been reported that the “Jesus Wife Papyrus” has been sent off for laboratory tests. I applaud this as well (I should hasten to add, as discussed below, that it would be particularly useful if the actual lab reports, rather than just the results, would be published as well. For example, it would be very useful to see the graphs generated by the spectroscopy).

I would also wish to emphasize at this time, though, that laboratory reports also require an interpreter. Those of us in the humanities sometimes seem to forget this (at least in practice). Here are a few scenarios, most of which are now attested in the laboratory analysis of actual epigraphic objects from the antiquities market (see section II below). Scenario 1: an epigraphic object is sent for analysis and the laboratory finds modern contaminants under the patina. At that point, the hard scientist has to make a decision. She or he could suggest that the piece is a modern forgery. This is entirely rational and I think quite convincing in most cases. However, it would be possible for him or her to propose that the modern contaminants under the patina could be the result of storage and handling practices (and thus they might suggest that the piece is actually ancient). In other words, the hard scientist must make a decision about the best way to understand the laboratory results. Scenario 2: Or again, let us say that someone has a potsherd from the antiquities market with an ink inscription on it and they send it off to a laboratory for thermoluminescence testing. The results for the thermoluminescence test would probably suggest that the pottery was indeed fired in antiquity. The hard scientist might conclude that this inscription is indeed ancient, based on the results of the thermoluminescence testing. However, this could be wrong. After all, a savvy forger would surely use an ancient potsherd and then forge an inscription on it, in a script and language that corresponds very nicely with the region and time period from which the potsherd came. That is, just because the potsherd is ancient doesn’t mean that the inscription is ancient. Scenario 3: Similarly, someone might send a piece of papyrus off for Accelerator Mass Spectrometry testing (an advanced form of carbon dating) in order to determine if the writing on it is ancient. The AMS test might very well indicate that the papyrus is ancient. However, it would not be prudent for a hard scientist automatically to conclude that the text was authentic, as ancient papyrus is something that can be purchased on the antiquities market and a modern forger worth his or her salt would be wise enough to use that as the medium. Scenario 4: Someone might propose that analyzing the ink should be considered decisive. This is difficult, as finding enough carbon in ink is very difficult (i.e., even for an AMS test, which requires less Carbon 14). But someone might suggest that spectroscopic analysis of the ink could be used, so as to determine the chemical composition of the ink. I believe that this could be useful, at least at times. But I would also hasten to add that enough is known now about the production of ink in antiquity (and the chemical makeup of many inks) that a savvy forger could also find some ancient carbonized remains (e.g., a piece of carbonized timber found on an excavation) and use that as the basis for the production of ink that appears ancient indeed, employing also data known about the rest of the chemicals often found in actual ancient inks (e.g., the right amount of iron, etc.).

For these sorts of reasons, I believe that caution should be the modus operandi. Of course, it is often suggested that forgers are not really that gifted, that savvy, and it is also suggested at times that no one who actually had such knowledge and talent would ever produce modern forgeries (i.e., someone that smart would not do it because of ethical reservations about such activities). Actually, I am convinced that the best modern forgers have formidable intellects and abilities and I think that the recent past demonstrates this rather convincingly, as some modern forgers have fooled some very gifted scholars. Moreover, I find it to be sweet that someone would assume that no one capable of producing really good modern forgeries would do so, but I am not convinced that brains and ethics necessarily always go hand in hand. Thus, I continue to be convinced that the future will see epigraphic forgeries that are absolutely undetectable. Indeed, I have already seen some inscriptions (and analyzed them microscopically) that are virtually perfect, from the script and language to the medium and the patina. I suspect that this will only increase in the coming years.

In addition, I must also admit that I think the field of epigraphy should be very careful about determining the credentials for the laboratory and the hard scientists. Not all labs are equal and not all hard scientists are equal. And there is always an interpretive component. Furthermore, I think that it should go without saying that those with a vested interest in the results should have no part in the sending of pieces to laboratories.

The remainder of this blog post focuses on protocols for laboratory testing as well as a discussion of some laboratory tests which have been performed on epigraphic objects during recent years, including some of the problems with some of these tests. Most of the discussion that follows first appeared in print in an article of mine published a decade ago in the journal Maarav (2003) [footnote 1]. In any case, the focus will be on (1) useful protocols for laboratory testing, (2) followed by concrete discussion of lab testing, based on published lab reports (and some of the problems with these). The hope of all this is that the field can continue to become more and more conscious of the need for the employment of strict protocols for laboratory testing and the importance of understanding the interpretive component that is part of laboratory analysis.


Owners, agents of owners, buyers, agents of buyers, dealers of antiquity, and epigraphists are often interested in having laboratory tests performed on non-provenanced epigraphic objects. In addition, these groups are often required to assess the results and significance of laboratory tests. For methodological reasons, I would argue that certain protocols should be adhered to so that the laboratory tests can actually function as valuable tools.

(1) First, laboratories must be demonstrably legitimate institutions, with the capability of performing the appropriate laboratory tests, and with credentialed individuals performing the tests. Because most epigraphists are not au courante with regard to the best scientific laboratories, there is the potential for substandard laboratories to “authenticate” objects without sufficient scientific bases, and for epigraphists to “accept” such results as authoritative. Furthermore, laboratories that have “authenticated” forged objects (e.g., because of incompetence or collusion) are not to be considered the most reliable and authoritative for future tests [footnote 2].

(2) Second, it is prudent for multiple laboratories, performing the same basic tests, on the same objects, to be used; moreover, it is also imperative that said labs do not communicate together regarding the objects. It is preferable for at least two labs to be used, but cases of conflicting results will necessitate the use of additional labs [footnote 3].

(3) It is of fundamental importance that the laboratory tests be double blind. That is, those requesting the test should not know the names of the laboratories or scientists performing the laboratory tests, and the laboratories should not know the names of the owners, potential buyers, dealers, or (ideally) even the name(s) of the epigraphist(s) studying the epigraphs. Laboratory tests that are not double blind are of minimal value, or perhaps of no value at all, because violation of this protocol ultimately compromises the process, either in theory, or in fact [footnote 4].

(4) Generally, no information about the epigraphic object’s potential or probable antiquity (and thus authenticity) should be given. The reason for this is simply that the scientist performing the laboratory tests, if he or she knows the expected or desired results of the test, can factor in various data in particular ways in order to achieve the expected or desired result [footnote 5].

(5) The full report of all tests should be made available in complete form in the formal publication of the inscription. Laboratory tests that violate these protocols, rigorous though the lab tests may appear, are to be considered of modest value, or of no value at all [footnote 6].

(6) With high profile epigraphs or corpora of epigraphs, when it is readily apparent that there are problems with the laboratory tests performed previously, museums, departments of antiquity, and universities will sometimes need to intervene and appoint disinterested specialists to analyze said epigraphs or corpora. Of course, in circumstances such as this it will not be possible for the tests to be double-blind. Nonetheless, it is expected that a full publication of these lab results will be satisfactory [footnote 7].


Laboratory tests can be of substantial usefulness in “deauthenticating” epigraphic objects [footnote 8]. Because laboratory tests have been widely used in recent years, and because “laboratory authentication” is often considered definitive, the usefulness of laboratory tests, especially the necessary protocols for laboratory tests and the necessary caveats regarding their decisiveness, merits discussion. For the sake of clarity, various specific laboratory analyses will be discussed; however, it should be noted that this discussion is not exhaustive, but rather illustrative. Finally, although I have attempted to make this discussion accessible, it is inevitable that some specialized vocabulary will be used [footnote 9].


The “Moussaieff Ostraca” have been subjected to certain laboratory tests [footnote 10]. The McCrone affirmed that the soil on the ostraca was Middle Eastern and that the chemical composition of the ink was very similar to that of some of the Dead Sea Scrolls. This data is, of course, of limited usefulness, as carbon based ink is not difficult to produce, and the presence of Middle Eastern soil on an object ultimately proves absolutely nothing about the antiquity of an object [footnote 11].

Mikrofokus Oy Lab focused its analysis on a white patina covering the surface of the ostraca. Using a SEM-EDS (i.e., a scanning electron microscope equipped with an EDS), the McCrone lab had determined that there was a micrometeorite or pyrophoric igniter powder (e.g., from a cigarette lighter) contaminating a sample. Moreover, Mikrofokus Oy affirmed the presence of “spherical particles found which seem to be inside the lower layer of the white material” (i.e., under the patina). The lab then noted that the particles were composed of silicon, and that they were identical in form, size, and chemical composition to those commonly found among flyash particles in industrial environments. Mikrofokus Oy suggested that these were indeed modern contaminants, but ultimately proposed that such particles need not mean that the ostraca were forgeries, even though they were found under the patina. Variable handling and storage practices were assumed to be the cause of this anomaly. Moreover, because producing such a patina today would require a special environment, access to calcite crystals of the right size, specialized knowledge, and a long time, (e.g., “years”), the lab concluded that the patina on the ostraca was a natural precipitation product, and that “what is under the white material must then be genuine.” Note that the Mikrofokus Oy lab, however, was able to duplicate the patina on the ostraca quite closely, using sodium silicate, calcium carbonate, and water.

Obviously, Mikrofokus Oy’s interpretation is a possible interpretation of the data generated by their SEM-EDS test, but it is my opinion that this test does not actually generate much confidence. That is, a number of their interpretations of these test results are potentially troubling. For example, the modern contaminants under the patina could justifiably be understood as being there because the patina was produced during the modern period. Indeed, this is perfectly plausible. Also, the Mikrofokus Oy laboratory assumes that the patina is probably not modern because few people have the appropriate chemicals and the ability to produce such a patina. However, within their own report, Mikrofokus Oy stated that they were able to produce a similar patina, within a very short period. The position of Mikrofokus Oy seems to be that if the patina is not easy to reproduce, then it is safe to assume that it was not done in this case on these ostraca. This is not, however, a necessary conclusion, and I would argue that it is not even necessarily a logical conclusion. In short, it is my opinion that although these tests do not necessarily demonstrate that the Moussaieff Ostraca must be disqualified (i.e., considered forgeries), neither is it possible to state in any way that these tests are non-problematic, and authenticate the ostraca.

The Geological Survey of Israel performed SEM-EDS analyses on the patina of the “Ya‘akov Ossuary” [footnote 12] 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 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 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 [footnote 13].

The Geological Survey of Israel’s official report for the Jehoash Inscription is much more detailed than the one published (initially) for the ossuary, but this official report is replete with some problems, alas [footnote 14]. For example, (1) the report assumes that if the patina on the medium (in this case, stone) essentially “matches” that which is in the grooves of the inscription, the inscription must be ancient. This is problematic, as it is possible to argue that a modern forger had carefully chosen a stone, cleaned it with non-treated water (e.g., water from a pure spring), incised it, and then the stone was buried and a patina “cultivated,” either passively (just letting it develop naturally with the compounds in the soil and rainfall) or actively (i.e., “watering” the soil at times, and adding some “safe” compounds, such as lime, iron, etc.) that might assist in the development of the patina. The result would be a patina on the surface of the rock that reflected a chemical composition that was similar to that of the inscribed area. Of course, Goren has now demonstrated conclusively that the precise patina described by the Geological Survey of Israel could be produced quite readily in a lab [footnote 15]. (2) The Geological Survey of Israel put enormous credence in the fact that within the patina were organic materials that could be dated (on the basis of 14C) to antiquity. Again, this is an interpretation, but because ancient organic materials are readily available on the market (as well as to those associated with excavations), this cannot be considered the interpretation. A clever forger is certainly capable of attempting to dupe a lab technician by augmenting a fabricated patina with some organic material. Moreover, with regard to the Jehoash Inscription in particular, this is now the most convincing interpretation [footnote 16].

During the past fifty years, of course, Carbon 14 tests have indeed been very helpful in assessing the antiquity of organic material at times. For example, Carbon 14 tests (AMS) were performed on various Dead Sea Scrolls, and the results were decisive, demonstrating the antiquity of the scrolls and providing a control on palaeographic analysis [footnote 17]. Moreover, several decades ago, Carbon 14 tests were performed on the “Hebron Documents” and the tests yielded a modern date, a date that has stood, pace Mendenhall [footnote 18]. Nevertheless, several caveats must be kept in mind even with Carbon 14. (1) Variable handling and storage conditions can have an impact on laboratory tests such as Carbon 14. (2) A Carbon 14 test of organic material (e.g., leather, papyrus) does not necessarily demonstrate the antiquity of the inscription, as pieces of ancient leather and papyrus are sometimes found without visible writing, and can be readily used by a modern forger as the “medium” (i.e., material) for a forgery.


Thermoluminescence testing is a potentially important means of determining the age of an object that has suffered in some sort of a conflagration. In essence, very high temperatures reset the thermoluminescence “clock.” Hence, fired ceramics (and “burned” bullae) can be subjected to a thermoluminescence test, and an approximate date of firing can be determined. Thermoluminescence testing on non-provenanced objects, however, is generally not nearly as precise as radiocarbon dating, because there are numerous (quite imprecise) variables to calculate. For example, the chemical content of the soil in which an object has been deposited, the proximity of rocks, and even the types of rocks, all impact the thermoluminescence signal. Therefore, non-provenanced objects, and even provenanced objects not accompanied by samples of the soil surrounding the objects, yield only very approximate dates, with a plus or minus of centuries.

Thermoluminescence texts were recently performed on Hebrew bullae that had ostensibly been fired in some sort of conflagration. Namely, during January 1997, a thermoluminescence test on a bulla was performed by Doreen Stoneham at the Research Laboratory for Archaeology and the History of Art in Oxford. The report is important for a number of reasons. For example, it states that the sample was obtained in “powdered form on 8 January 1997” by Doreen Stoneham, and that it was “presumed” to have been taken from a “small fragment of terracotta.” Stoneham concludes that “using standard methods and techniques it is estimated that the material of the sample was “last fired between 1100 and 1800 years ago” and that this is consistent with the “suggested period of manufacture of the object concerned.” The report concludes by noting that the “result is given in good faith; however the laboratory takes no responsibility for financial loss incurred through an erroneous report being given,” that is, with a certain amount of customary legalese [footnote 19]. Obviously, this thermoluminescence test suggests that the “powder” analyzed was ancient [footnote 20].

Ralf Kotalla of the Ralf Kotalla “Laboratory” (the quotation marks are part of the official document’s report) performed a thermoluminescence test on an Iron Age bulla as well. Kotalla’s report states that the bulla was “2700 years old +/- 20% of the overall age,” and even notes that it originated in the “Orient.” In addition, Kotalla notes that the “assumed age on the basis of stylistic characteristics [is] appr. 2700 years” [footnote 21]. These thermoluminescence test results are especially interesting, because Kotalla was told the precise date of this bulla (i.e., based on the script it dates to ca. 700 BCE). For this reason, he was able to factor in numerous variables and produce a date that corresponded perfectly with the desired result. Stoneham’s test yielded an honest result, but Kotalla’s reflects severe bias. Note that I am not necessarily questioning the antiquity of these bullae. Indeed, for “fired” bullae, a thermoluminescence test is potentially helpful [footnote 22].


Spectrographic analyses can be very useful as well, for example, in determining the composition of a metal. Two inscribed arrowheads were analyzed at Conservation Laboratory of the Freer Gallery of Art in Washington, D.C. several years ago. Based on the script, the arrowheads appeared to derive from the Late Bronze Age. The precise results of the spectrographic analyses were as follows: Arrowhead 1: CU 78.925%; ZN – ; PB 7.166%; FE 1.878%; SN 11.940%; and AS 0.091%. Arrowhead 2: CU 73.731%; ZN 0.035%; PB 5.055%; FE 0.981%; SN 20.198%; and AS – ; The absence of zinc in the first arrowhead and the minimal amount in the second arrowhead is important, since copper alloyed with zinc (brass) is rarely found in ancient artifacts. However, McCarter has astutely noted that the “antiquity of the arrowhead is no guarantee of the antiquity of the inscription,” and he states further that “there is now reason to believe that forged arrowheads have begun to appear on the antiquities market” [footnote 23]. The Freer also subjected these arrowheads to microscopic examination as well, and this revealed that the corrosion on the surface of the blades is also present in the incised grooves of their inscriptions. It has been argued that this “mitigates strongly against the possibility that either inscription is modern” [footnote 24]. I concur that this information is helpful, and I am not questioning the antiquity of the inscriptions on these arrowheads. Nonetheless, I would note that creating a corrosion on the surface of a metal object (either before or after cleaning the object) is not difficult for the right person, with access to the right chemicals [footnote 25].

Ultimately, because some forgers are quite savvy about laboratory tests (and because lab scientists have sometimes failed to consider the skills of forgers and their associates), such tests are often helpful, but cannot normally be considered decisive in determining authenticity. Obviously, laboratory tests can disqualify an inscription, but this is not the same as authentication.


At this juncture, I should also like to note another type of problematic data that is used to argue for the antiquity of an inscription, namely, that of the tools employed to produce it. For example, regarding the Ya‘akov Ossuary, it was noted that there were “no signs of the use of a modern tool or instrument found” [footnote 26]. Again, this is useful information, but it could never be considered decisive. A shrewd forger could certainly have used an ancient tool, ancient worked metal, or modern bronze or iron consisting of the same composition as ancient tools, so as to have produced the entire inscription, or a portion thereof. Methodologically, it is simply not sage to assume that a modern forger would not take such precautions. Indeed, I would argue that a good modern forger would definitely take such precautions, for modest precautions yield large dividends.

In sum, the use of laboratory tests for inscriptions from the market is auspicious, but the labs conducting the tests must be vetted, protocols for the testing must be put in place in every case, and the results of the laboratory tests must be fully published so that they can be scrutinized as well. In short, there is much to be hopeful about, but methodological doubt must be maintained as well.

[footnote 1]. Rollston, “Non-Provenanced Epigraphs I: Pillaged Antiquities, Northwest Semitic Forgeries, and Protocols for Laboratory Tests.” MAARAV 10 (2003): 135-193.

[footnote 2]. P. Kyle McCarter, Jr. has suggested (personal communication) that it could be very useful to submit three epigraphs to the labs: the epigraph of interest, another that is definitely a fake, and a third that is definitely ancient. That is, he is suggesting the use of “control groups,” as it were.

[footnote 3]. Note, for example, that excavators routinely send samples of organic materials to multiple (often two) labs. The point is that using multiple labs can function as a “control” of sorts.

[footnote 4]. The point here is that recently owners, dealers, or agents thereof, have been taking objects to the lab of their choice, verbalizing expected results, and paying for the tests. This is hardly the ideal.

[footnote 5]. With regard to this issue, note the problems with the thermoluminescence tests discussed earlier.

[footnote 6]. Note that the initial report on the Ya‘akov Ossuary published by the Geological Survey of Israel did not contain much of the raw data or summaries that are to be expected of a SEM-EDS analysis. See A. Rosenfeld and S. Ilani, “SEM-EDS Analyses of Patina,” 29.

[footnote 7]. In my opinion, the IAA should be applauded for doing precisely this during 2003 (for the Ya‘akov Ossuary and the Jehoash Inscription).

[footnote 8] The (2003) report from the commission convened by the IAA to do laboratory analyses of the Ya‘akov Ossuary and the Jehoash Inscription demonstrates this point. Needless to say, I am not convinced by those who have criticized the conclusions of the IAA committee. Indeed, some of the attempted “rebuttals of the report,” read like apologias. In any case, for a brief, but cautious and sound discussion of laboratory testing, see also N. S. Fox, In the Service of the King, 30-31.

[footnote 9]. For a cursory review of the methods discussed here, see W. Ashmore and R. J. Sharer, Discovering our Past: A Brief Introduction to Archaeology, 3rd ed (New York: McGraw-Hill, 2000), esp. 159-168. For analysis using a scanning electron microscope, see also my discussion: C.A. Rollston, “Laboratory Analysis of the Moussaeff Ostraca using the Scanning Electron Microscope (SEM) with an Energy Dispersive X-Ray Microanalyzer (EDS),” 8.

[footnote 10]. For a summary of the lab reports, see C.A. Rollston, “Laboratory Analysis of the Moussaïeff Ostraca using the Scanning Electron Microscope (SEM) with an Energy Dispersive X-Ray Microanalyzer (EDS),” 8-9.

[footnote 11]. Regarding the composition of the ink of ancient inscriptions, see A. Lewis, “Report on the Lachish Letters with Remarks upon the Use of Iron Inks in Antiquity,” in The Lachish Letters, ed. H. Torczyner (Tur-Sinai), 138-193. Of course, carbon (of various sorts) is often found on excavations. The production of carbon ink that appears ancient (e.g., on a 14C test) should not be problematic in the modern period.

[footnote 12]. A Rosenfeld and S. Ilani, “SEM-EDS Analyses of Patina Samples,” BARev 28 (Nov/Dec 2002): 29.

[footnote 13]. Note that there are various other problems with this report as well. For example, the report states that the gray patina found on the surface of the ossuary was “found also within some of the letters, although the inscription was cleaned and the patina is therefore absent from several letters.” However, there is no statement of precisely which letters had this patina and which lacked it (an important issue for various reasons, but especially because of the suggestion that the second half of the inscription was from a later hand). Of course, forgers can use the issue of “cleaning” to their advantage, in various ways.

[footnote 14]. S. Ilani, A. Rosenfeld, and M. Dvorachek, “A Stone Tablet with an Ancient Hebrew Inscription Attributed to Yehoash, King of Judea: Archaeometry and Epigraphy,” GSI Current Research 13 (2003): 109-13.

[footnote 15]. Y. Goren, “An Alternative Interpretation of the Stone Tablet.”

[footnote 16]. Goren proposed this in his article “An Alternative Interpretation of the Stone Tablet,” 7. E. Boaretto’s final report for the IAA also states that the carbon material in the patina of the Jehoash Inscription does not indicate authenticity.

[footnote 17]. See G. Doudna, “Dating the Scrolls on the Basis of Radiocarbon Analysis,” in The Dead Sea Scrolls after Fifty Years, ed. P.W. Flint and J. C. VanderKam (Leiden: Brill, 1998), vol. 1: 430-471.

[footnote 18]. G.E. Mendenhall, “Philistine Documents,” 99.

[footnote 19]. This report is printed in the editio princeps of various non-provenanced inscriptions, namely: R. Deutsch, Messages from the Past: Hebrew Bullae from the Time of Isaiah through the Destruction of the First Temple: Shlomo Moussaieff Collection and an Up to Date Corpus (Tel Aviv: Archaeological Center Publications, 1997), 169. Recently, R. Deutsch also edited a Festschrift in honor of S. Moussaieff, a collector of antiquities, and published a number of additional bullae. See Shlomo: Studies in Epigraphy, Iconography, History, and Archaeology in Honor of Shlomo Moussaieff (Tel Aviv: Archaeological Center Publications, 2003).

[footnote 20]. The date yielded by this test are very broad, and considerably later than might have been expected for an Iron Age bulla. Among the reasons for this are (1) the large number of variables that must be calculated, but cannot be done with precision without data regarding the context in which it was ostensibly found; (2) the theoretical possibility that it suffered a conflagration centuries after its date of manufacture; (3) the theoretical possibility that it has been exposed to some sort of irradiation.

[footnote 21]. For the report, see R. Deutsch, Messages from the Past, 168.

[footnote 22]. I should like to propose that epigraphists and laboratories should anticipate that some forgers may attempt to carve bullae (in relief) from ceramics fired in antiquity, and then to patinate the “epigraph.” Of course, it should be noted that for ostraca, thermoluminescence tests are useless (i.e., a test performed on a fired potsherd) as a forger will find ample sherds in the Middle East upon which to pen his or her logia.

[footnote 23]. P. K. McCarter, Jr., “Two Bronze Arrowheads with Archaic Alphabetic Inscriptions,” in Eretz-Israel 26: Frank Moore Cross Volume, ed. B.A. Levine, et al., (Jerusalem: Israel Exploration Society, 1999), 126-127.

[footnote 24]. P. K. McCarter, Jr., “Two Bronze Arrowheads,” 127.

[footnote 25]. That is, a forger could take a “blank” ancient bronze arrowhead, prepare the surface, forge an inscription, and then create a corrosion (or some type of patina). David A. Scott, senior scientist at the Getty Conservation Institute, has stated (personal communication) that in this case, “the patina under the inscription would be different than the patina over the rest of the surface.” That is, Scott argues that it would be detectable. However, he affirms that the arrowhead “would have to be cut” (and then it should be possible “to tell by metallography that the inscription was added and not original to the object).” If the arrowhead was not cut, Scott suggests that “it should still be possible” to make determinations regarding the differences in the patinas (and thus the antiquity of the inscription itself), but he stated that one would then “have trouble describing the process easily.” For more data about these subjects, see “D.A. Scott, Copper and Bronze in Art: Corrosion, Colorants, Conservation (Getty Conservation Institute, 2002). Ultimately, the point is that even a corrosion should not automatically be assumed to be indicative of antiquity. Certain types of corrosions can be created; therefore, sophisticated laboratory tests (sometimes destructive) are needed to determine the depth and nature of the corrosion in order to assess the antiquity of the inscription.

[footnote 26]. A. Rosenfeld and S. Ilani, “SEM-EDS Analyses of Patina Samples,” BARev 28 (Nov/Dec 2002): 29.

Nebuchadnezzar’s Destruction of Jerusalem, The Cyrus Cylinder, and the Building of the Second Temple

27 March 2013

Nebuchadnezzar’s Destruction of Jerusalem, The Cyrus Cylinder, the Building of the Second Temple:
A Brief Historical Synopsis of Salient Elements

Christopher A. Rollston, Visiting Professor of Northwest Semitic Languages and Literatures,
George Washington University

Cyrus the Great of Persia is called “Meshiah” (that is, “Anointed One,” “Messiah”) in the Hebrew text of Isaiah 45:1 and Yahweh’s “Shepherd” in Isaiah 44:28. This sort of grandiose language may seem striking to some. It should, as it is striking. But the backstory provides the basic rationale for this lofty verbiage. Namely, several decades before Second Isaiah referred to Cyrus as “Meshiah“ and “Shepherd,” Judah had suffered mightily at the hands of the Babylonians. It all began in ca. 597 BCE. The gold and silver of the Jerusalem Temple and Royal Palace had been plundered, but both buildings still stood. King Nebuchadnezzar the Great of Babylon was marching victoriously back to Babylon, not only with these precious metals but also with several thousand Judean prisoners of war. Among them were King Jehoiachin and much of the Judean royal family (2 Kings 24). Things were bad, but they would get worse, as Nebuchadnezzar would return to Jerusalem some ten years later to avenge and to destroy. Nebuchadnezzar’s rationale was this: Zedekiah had become king of Judah after Jehoiachin was exiled but he had not been the loyal vassal for whom Nebuchadnezzar had hoped. Nebuchadnezzar was angry, he came to Jerusalem and besieged it for some eighteen to twenty months, beginning around 587 BCE (2 Kings 25).

Conditions inside Jerusalem soon became desperate. The book of Kings laconically states that during the terminal portion of the siege “the famine became so severe that there was no food for the people” (2 Kings 25:3). But the poet of Lamentations limns the picture more poignantly, “the hands of compassionate women had boiled their children, they became food for them” (Lam 4:10). Desperation reigned. Then things deteriorated further. The walls were breached and the Temple and Palace were burned to the ground, along with all the houses of Jerusalem (2 Kings 25:9). And brutality of a different sort began as hand-to-hand combat concluded: “women in Zion were raped, virgins (raped) in the cities of Judah” and “princes were hung by their hands” (Lam 5:11, 12). Words could not adequately describe the horror.

Zedekiah had abandoned Jerusalem shortly before its fall. But he and his young sons were captured near Jericho, deserted by the armed Judean soldiers who had pledged to protect them. Nebuchadnezzar decreed that Zedekiah and his sons be brought forward. They were, and then Zedekiah’s young sons were brutally slaughtered before their loving father’s eyes. At that point, a Babylonian soldier gouged out the Judean king’s eyes, his last visual memory now a haunting one. Zedekiah was led away in chains to Babylon (2 Kings 25:7). The year was 586 BCE. This was the nadir of nadirs. From the hill of Zion profound sorrow and anger flowed. Raw human emotion is reflected in the words of a Psalmist: “O daughter of Babylon, you destroyer. Happy are the ones who seize your little ones and dash them against the rock!” (Psalm 137:8-9). Nebuchadnezzar’s campaigns against Judah had brought bloodshed, starvation, and destruction. Judah remembered Nebuchadnezzar as a brutal conqueror, and this he was.

But history is made of reversals and the tables were soon turned. East of the Tigris River, Cyrus the Great had begun to reign in Persia (around 559 BCE), and he soon began to weld together a full-fledged empire, defeating the Kingdom of the Medes and the Kingdom of Lydia. King Nabonidus (r. 556-539 BCE) was on the throne of Babylon, as one of the successors of Nebuchadnezzar the Great. But he would be Babylon’s last king. He had already spent around a decade of his reign at Tayma, an oasis in the Arabian Desert. There is actually an allusion to this tradition in an Aramaic Dead Sea Scroll called “The Prayer of Nabonidus.” In any case, based on the Mesopotamian texts at our disposal, there seem to have been some rumblings against Nabonidus even during his decade at the oasis, especially within the Babylonian priesthood. After all, he was said to have been most devoted to the Moon God Sîn rather than Marduk, the head of the Babylonian pantheon. Nabonidus was an apostate, or so it seemed to some. He returned from the oasis, disaster now looming across the Tigris River. Then Cyrus began to march, and the prize he wanted most was the kingdom of Nabonidus.

The ancient historical sources are not all in agreement about the battles that were fought between the Babylonians and the Persians. Cyrus himself boasts that he entered Babylon without a battle, hailed (he says) as a liberator even by the Babylonians themselves. But the full story was certainly bloodier, and the Babylonian supporters of Cyrus fewer (Herodotus suggests as much). Nevertheless, Cyrus gained his prize, Babylon was his in 539 BCE. The Persian Empire Period had begun. Babylon had fallen. The Judeans who had felt the brunt of Babylon’s war machine fifty years earlier probably shed few tears at this news.

But there is more. Cyrus not only brought the Babylonian Empire to its knees, he also decreed (according to the book of Ezra) that the exiled Judeans in his realm be permitted to return to Judah and to rebuild the Temple. According to the book of Ezra, he also allowed the exiles to take with them (some of) the sacred vessels which had been pillaged from the First Temple by Nebuchadnezzar. And the book of Ezra states that there was an edict of Cyrus that said: “Yahweh, the God of Heaven, has given me all the kingdoms of the earth, and he commissioned me to build for him the Temple in Jerusalem, which is in Judah” (Ezra 1 and 6). In due time, work on the Second Temple would begin, and it would be completed by around 516 BCE, during the reign of Darius the Great of Persia (r. 522-486 BCE). It is within this context that the words of Isaiah 44 and 45 are best understood. Cyrus of Persia had destroyed the (Babylonian) destroyers of Jerusalem, freed the Judean Exiles, and decreed that the Jerusalem Temple be rebuilt. I suspect that Second Isaiah was not alone in his jubilation about Cyrus.

Cyrus was certainly famous in antiquity (and in the modern period) for his benevolence, even among the Greeks, due in part to Xenophon’s lengthy work entitled “Cyropaidia” (literally, the ‘Education of Cyrus’). But during excavations in Babylon in 1879, the now famous “Cyrus Cylinder” was found, galvanizing further the reputation of Cyrus. Certain salient facts about this cuneiform text are worth mentioning at the outset: (a) In terms of size, it is quite small, about ten inches by four inches, and cylindrical in shape. (b) In terms of language, although Cyrus was a Persian, the Cyrus Cylinder is written in the Akkadian language (i.e., not in Persian, the native language of Cyrus). Of course, this makes sense, as the target audience for this inscription was Babylonian, not Persian. (c) In terms of the amount of textual content, the Cylinder is relatively short, just a few hundred words long, preserved in some forty to fifty lines of cuneiform text. (d) In terms of date, it arguably hails from the very first years of the reign of Cyrus. (e) In terms of archaeological context, it was found as a “foundation deposit” in an ancient Babylonian building.

The content of this text is priceless, and it is laced with some very savvy royal apologia. It is most impressive. Here is a synopsis of the content of the Cyrus Cylinder, using the translations of Irving Finkel of the British Museum. The text begins with a narrative in the third person (rather than the first person, that is, “he” not “I”) which condemns the Babylonian King Nabonidus (whom Cyrus had just vanquished, of course), along with statements impugning Nabonidus for not being a pious worshipper of Marduk. The Cyrus Cylinder says that because of Marduk’s anger for Nabonidus, He (Marduk) raised up Cyrus the Persian, “an upright king,” taking him “by the hand” and ordering him (Cyrus) to go to Babylon and remove Nabonidus from power. Moreover, Marduk was “like a friend and a companion” to Cyrus. Then, at line 20 of the Cyrus Cylinder, the grammatical first person begins to be used. “I am Cyrus, king of the world!” Cyrus himself then declares that he is the king “whom Divine Marduk and Divine Nabu love.” He also states that upon his arrival in Babylon, the Babylonian people welcomed him with joy as he entered. He affirms that they viewed him as a liberator. After he became nicely ensconced in Babylon, Cyrus states that many kings from various regions “brought me weighty tribute” and “kissed my feet.” In return, he decrees that the people from various regions that had come under his dominion (especially because he had just vanquished Babylon) should be allowed to return to their homelands and to rebuild their temples. In addition, he requests the following: “May all the Gods that I returned to their sanctuaries, every day before Marduk and Nabu, ask for a long life for me, and mention my good deeds.” Finally, he also affirms that he has “enabled all the lands to live in peace.” The Cyrus Cylinder is a stunning archaeological artifact.

We do not know much at all about the personal religion of Cyrus the Great, but it is most reasonable to contend that he worshipped some of the Persian Gods, perhaps especially the God Ahuramazda. This was, after all, the case for several of the Persian kings who succeeded Cyrus. Therefore, it is all the more interesting that that Cyrus declares in the Cyrus Cylinder (written for a Babylonian audience) that he vanquished Babylon because the Babylonian God Marduk told him to do so! Of course, kings in the ancient Near East normally declared that they had divine patronage, but normally of their own Gods. In this case, however, Cyrus declares that the Babylonian God Marduk transferred His support from the Babylonian King Nabonidus and gave it to the Persian King Cyrus. Moreover, it is important to remember in this connection that the book of Ezra states that Cyrus had said something similar to the Judeans, namely, “Yahweh, the God of Heaven, has given me (Cyrus) all the kingdoms of the earth, and he commissioned me to build for him the temple in Jerusalem, which is in Judah” (see Ezra 1 and 6). That is, according to these texts, Cyrus told the Babylonians that the Babylonian God Marduk told him to do what he did, and Cyrus told the Judeans that the Judean God Yahweh told him to do what he did. And I think that it is entirely reasonable to suppose that Cyrus told the Persian people that the Persian God Ahurzmazda told him to do what he did. I should note in this connection that this sort of brilliant royal apologia is not confined to Cyrus. After all, during King Sennacherib of Assyria’s siege of Jerusalem in 701 BCE, the Assyrian Rab Shakeh uses (at least according to 2 Kings 18:25) the same sort of rhetoric, arguing that it was Yahweh the God of Judah who summoned him (Sennacherib) to attack Judah. And the Neo-Assyrian Kings Esarhaddon and Ashurbanipal used similar rhetoric as well.

Of course, much has been made, especially during the past few decades, of the religious and political tolerance and generous diplomacy of Cyrus the Great. In fact, the Cyrus Cylinder itself has been referred to at times as an “Icon of Freedom” and even as “The First Bill of Human Rights,” oft repeated slogans as it is now in the midst of museum travels in the United States. Some thirty years ago, however, Amelie Kuhrt argued quite cogently that these sorts of appellatives might just be too grandiose. And most scholars within the field have concurred with Kuhrt’s corrective (demonstated again very nicely by Jacob Wright’s fine article on the Huffington Post several months ago). After all, the text of the Cyrus Cylinder is rather brief and much of the language contained in it can be found in earlier ancient Near Eastern Royal Inscriptions. And there is no grand affirmation of human rights within the Cyrus Cylinder, per se. And although Cyrus allowed the various people-groups (e.g., those who had been captured and exiled by the Babylonians) to return to their homelands, these people would certainly remain under Persian hegemony, and fealty to Persia would be demanded (including tribute). In short, there were some strings attached, big strings.

But I should also wish to emphasize that, at least for me, I remain very impressed with the words and actions of Cyrus. After all, not all conquerors in the ancient Near East were so kind to the conquered as Cyrus arguably was. Nebuchadnezzar’s treatment of the Judeans is an obvious demonstration of that. And I must also affirm that the basic deference of Cyrus to the religious sensibilities of the conquered is most admirable. True, Cyrus was not the only suzerain to be tolerant of the religious practices of a vassal (for discussion, see especially Beaulieu). But I would counter that not all suzerains were so tolerant, thus, I consider this to be a benevolent act. Someone might retort that his actions were more “savvy diplomacy” than “religious tolerance.” Perhaps so, but I admire his actions still. And, of course, it is both striking and important that a Judean writer of the late 6th century understood the actions of Cyrus to be good and noble, so much so as to cause him to refer to Cyrus as Yahweh’s “Shepherd” and his “Meshiah.” I take this as pretty good evidence (because it is close to being contemporary with the actual decree of Cyrus) that Cyrus was viewed by many in antiquity as a benevolent monarch, with regard to both politics and religion.

Some final musings: Within the contemporary world, people often attempt to mine ancient texts for models, virtues to be embraced or vices to be shunned. This can certainly be a useful thing, but it can also be a precarious venture, as it is all too easy to read too much into these ancient texts. But in days such as ours, full of many political and religious tensions across much of the globe, I must admit that some of the words and deeds of Cyrus resonate with me. I think something can be learned from these words. They deserve to be studied as important diplomatic and religious statements, as potent words from some two and a half millennia ago that were moving, at least in part, in good directions. And as for me, I’m happy to see movement in the direction of more tolerance, regardless of the ancient or modern texts in which it can be found, and regardless of whether the form is fledgling or fully developed.

Next Page »