Bad omens for Cornell?

In September 2019, US law enforcement agents seized the so-called Gilgamesh Dream Tablet from the possession of the Museum of the Bible in Washington DC. The Gilgamesh Dream Tablet is a cuneiform tablet whose text records part of the Epic of Gilgamesh. On 18 May 2020, the US Attorney’s Office of the Eastern District of New York filed a civil action to forfeit the tablet, alleging it had originated in Iraq and entered the USA in contravention of US law. The forfeiture complaint makes for interesting reading. It claims that in March or April 2003, after first viewing the material prior to 2001, an unnamed US “Antiquities Dealer” bought a group of cuneiform tablets from the Jordanian dealer Ghassan Rihani in London. The “Antiquities Dealer” was accompanied by what the complaint terms a “Cuneiform Expert” (since suggested to be independent scholar Renee Kovacs). The Cuneiform Expert recognised that the tablets did not bear ordinary administrative texts but were instead “potentially of a literary nature”. The Antiquities Dealer paid $50,350 for the tablets, the Gilgamesh Dream Tablet among them, and arranged shipment to the USA.

Once in the USA, the Cuneiform Expert arranged for Andrew George to study the Gilgamesh Dream Tablet in 2005. His study was published in 2007 (George 2007). In his publication, George stated that the tablet was published with “the permission of the owner, who wishes to remain anonymous” (George 2007: 59). The owner was presumably the Antiquities Dealer. Importantly, George also noted that:

The tablet was reported to be part of a group of unpublished tablets that included omen and liturgical compositions, some mentioning Pešgaldaramaš (peš-gal-dàra-meš) and Ayadaragalamma (a-a-dàra-galam-ma), kings of the First Sealand Dynasty, and to share with them aspects of physical appearance and ductus. I was able to confirm this report from photographs of the tablets in question. In particular, the present piece closely resembles a tablet of lung omens dated to Pešgaldaramaš. Close resemblance is not an infallible criterion for attributing provenance, but it is enough to permit a provisional hypothesis that the tablet published here derives from the same source as the Sealand tablets (George 2007: 63).

So, someone, presumably the Cuneiform Expert, had reported to George that the Gilgamesh tablet had been (or was still) part of a group including omen and liturgical tablets dating to the First Sealand Dynasty and supplied photographs of them. George then commented upon the presence in the group of a lung omens tablet dated to Pešgaldarameš.

Tablets attributable to the First Sealand Dynasty were unknown before 1999 when Stephanie Dalley started studying 474 administrative tablets in the Schøyen Collection, subsequently published in 2009 (Dalley 2009). Kovacs was associated with the Schøyen Collection until 2005, when she handed over responsibility for publishing the collection’s cuneiform holdings to George. George had first met Schøyen in 2001 (George 2009: xi). So, presumably, both Kovacs and George would have been aware of the Sealand tablets in the Schøyen Collection – George definitely was (George 2007: 63). By 2005, if Kovacs was the Cuneiform Expert, she would have been able to recognize the Sealand kings from their names and advise George accordingly, which is what he indicated in his 2007 paper.

In 2013, George went on to publish 10 omen tablets dating to the First Sealand Dynasty from an anonymous private collection (George 2013: tablet nos 22-29, 31-32). In his introduction, he stated that he had studied the tablets sometime between the years 2005 and 2012 and that:

Images of most of the tablets in the anonymous collection were made at the Rosen Seminar, Cornell University, and are published here by generous leave of David I. Owen, Curator of the Tablet Collections (George 2013: xi).

The texts have been made available on the Cuneiform Digital Library Initiative (CDLI) database.

CDLI omen texts published by George (2013)

The entry for George’s tablet no. 28 (CDLI P431306) is particularly interesting. It describes the text as:

Lung omens, middle lobe; 1st Sealand period script, dated by colophon to Pešgaldarameš.

So, another Sealand lung omens text dated to the reign of Pešgaldarameš. A startling coincidence, especially remembering that no First Sealand Dynasty texts at all were known before the late 1990s. The CDLI lists 77 Old Babylonian omen texts, but none in private collections and none dating to the First Sealand Dynasty except those published by George.

What does it mean? There are two possible explanations. The first is that in 2003 the US Antiquities Dealer bought a group of First Sealand Dynasty tablets including the Gilgamesh Dream tablet, some omen tablets, and particularly a lung omens tablet. He sold the omen tablets and they were lost from view. Meanwhile, an unknown private collector had acquired a different group of First Sealand Dynasty omen tablets, including another lung omens, which George was subsequently able to study and publish. Stranger things have happened, but the second and more parsimonious explanation is that the tablets bought by the Antiquities Dealer and the tablets studied by George were in fact the same ones. They were bought by the Antiquities Dealer in 2003, sold to a private collector, and sometime later studied and published by George. (In his 2007 publication, George does not mention the Pešgaldarameš colophon on the lung omens tablet, but then perhaps it hadn’t been visible on the photographs supplied to him).

CDLI entry for lung omens text

But there is more. When George published the tablets in 2013 he stated that they were with an anonymous private collector. Today, on the CDLI database, they are listed as the property of the Department of Near Eastern Studies at Cornell University. Has the private collector sold or donated them to Cornell? Remembering that George acknowledged publishing the tablets with the permission of David Owen, who is Professor of Ancient Near Eastern Studies at Cornell, it seems a likely eventuality

To some extent, this is all speculation. I cannot prove that the tablets sold by Rihani to the Dealer in 2003 are the ones published by George in 2013 or that they are now in the possession of Cornell. I don’t know for sure. Cornell might though.


Dalley, Stephanie. 2009. Babylonian Tablets from the First Sealand Dynasty in the Schøyen Collection. Bethesda: CDL.

George, Andrew. 2007. The civilizing of Ea-Enkidu: An unusual tablet of the Babylonian Gilgameš epic, Revue d’Assyriologie et d’archéologie orientale 101: 59-80.

George, Andrew. 2009. Babylonian Literary Texts in the Schøyen Collection. Bethesda: CDL.   George, Andrew. 2013. Babylonian Divinatory Texts Chiefly in the Schøyen Collection. Bethesda: CDL.

Cornell cuneiform

I have just noticed that in June 2019 Cornell University’s Jonathan and Jeannette Rosen Cuneiform Tablet Collection (CUNES Collection) was closed permanently. At the time of its closure, the collection was said to consist of approximately 10,000 cuneiform-inscribed objects. The announcement stated that “in the 20 years the cuneiform tablet collection was at Cornell, it was studied by dozens of scholars and has led to over 3 dozen books and articles, with more in press”. It didn’t say why the CUNES Collection was being closed.

The CUNES Collection was established in the early 2000s (or, if the announcement is accurate, in 1999) by a donation of possibly 1,500 cuneiform objects made by the collector Jonathan Rosen, with further donations in view (D’Arcy 2003). Many had been in the possession of Rosen since at least 1997 (Mayr 2007: ix; Owen 2007: vii), and Cornell’s acceptance of the tablets was dependent upon Rosen’s assurance of legal acquisition (Gottlieb and Meier 2003). Cornell used the donation to create the Jonathan and Jeannette Rosen Ancient Near Eastern Studies Seminar in the Department of Near Eastern Studies under the curatorship of David I. Owen, Professor of Ancient Near Eastern and Judaic Studies. Its mission was to house the tablets and facilitate their study. Rosen financed the creation of research and technical support positions (Owen 2007: viii). The CUNES Collection continued to grow in size after its establishment, as shown in the table below from its website statements. By 2013, it was reported to comprise approximately 10,000 objects (Felch 2013), so it must have reached its 2019 size sometime between 2011 and 2013.

Website dateNumber of objects
December 2009c. 6,500
October 2011c. 9,000
November 2015c. 10,000

The CUNES Collection has been entered into the database of the Cuneiform Digital Library Initiative (CDLI). The CDLI database records 10,435 CUNES cuneiform objects, the overwhelming majority comprising clay tablets. Just under half that number (4,973) have been published. Of the remainder, 670 are marked as “unpublished assigned”, meaning that their study and publication is planned or ongoing, and 4,792 are marked as “unpublished unassigned”, meaning that there are no immediate plans for their publication.

The CDLI database provides “provenience” information for 4,061 objects, though the term “provenience” is used in relation to the name of an ancient settlement, not always its modern location. Many of the ancient settlements are known only from references in the cuneiform texts themselves and their actual locations remain a mystery. As this next table shows, 15 settlements are named as provenience. Most of them are believed to be situated within the boundaries of Iraq. One obvious exception is the Syrian site of Ebla.

SiteNumber of objects
Adab (modern Bismaya)1,168
Du-Enlila (location unknown)1
Dūr-Abī-ēšuḫ (location unknown)349
Ebla (modern Tell Mardikh)1
Ešnunna (modern Tell Asmar)1
Garšana (location unknown)1,485
Girsu (modern Tello)4
Irisaĝrig (location unknown)92
Isin (modern Bahriyat)9
KI.AN (location unknown)4
Larsa (modern Tell as-Senkereh)4
Nippur (modern Nuffar)197
Puzriš-Dagan (modern Drehem)14
Šuruppak (modern Fara)19
Umma (modern Tell Jokha)713

There is no record of how the objects reached Rosen or when they left Iraq. Assyriologists working on CUNES objects have occasionally claimed that they “derive from clandestine collections”, were “illicitly excavated”, or “come from illicit excavations”. I could give references but I am trying to spare scholarly blushes. David Owen is usually more circumspect, suggesting for example that the tablets have an unclear or suspect pedigree (Owen 2013a: 336). Sometimes he is less guarded, writing that the tablets “were evidently the product of clandestine excavations in Iraq sometime during the last [i.e. twentieth] century” (Owen and Mayr 2007: 1). Other knowledgeable Assyriologists have expressed their opinions that many of the CUNES objects were “looted after 1991” or “discovered through illicit excavations”. I cannot verify any of these allegations as I am not privy to any kind of provenance documentation. Other than the fact that the tablets were acquired from Jonathan Rosen, Cornell has consistently refused to release any information about their ownership histories. This active censorship of important evidence relating to the provenance and ultimately the legitimacy of the CUNES objects is surprising and regrettable in an academic context.

In 2001, US Customs in Newark intercepted and confiscated a shipment containing 302 cuneiform tablets en route to the Sumer Gallery of New York from Bin Jassim Transglobal of Dubai (Studevent-Hickman 2018). In 2010 the tablets were returned to the possession of Iraq. After study and publication (with the permission of the Director General of Iraqi Museums), 145 of the tablets were shown to be part of an Ur III (2100-2000 BC) archive belonging to a man named Aradmu, discovered at a presently unknown site somewhere in the vicinity of Nippur. These 145 tablets confiscated by US Customs comprise only one part of a larger archive, which includes a further 224 tablets in the CUNES Collection (Studevent-Hickman 2018: 4). Of the 197 CUNES Nippur objects registered on the CDLI database, only eight have been published, and a further 16 are marked “unpublished unassigned”, mainly dating to the Middle Babylonian (1400-1100 BC) period. The remaining 173 objects, all dating to the Ur III period, are marked “unpublished assigned”. Presumably they constitute the matching part of the Aradmu archive confiscated by US Customs in 2001.

The gradual increase in size of the CUNES Collection from the date of its establishment through to 2013 is evidence that Rosen continued to donate material over a prolonged period of 14 years. Perhaps he was making intermittent donations for his own financial reasons, or perhaps he was continuing to acquire material on the market and passing it on to Cornell. Behind the numbers, the evidence suggests that Rosen was indeed continuing to be active on the market. By 2013, the CUNES Collection had acquired at least 89 tablets from the site of Irisaĝrig (Owen 2013b: 3), and in 2020 the CDLI listed at least 92 Irisaĝrig tablets in its possession. Jordanian customs confiscated 167 Irisaĝrig tablets in 2003, but tablets from Irisaĝrig did not start appearing on the international market until 2004 (Viano 2019: 49; Molina 2013: 72). Thus, it seems likely that Rosen and then Cornell could only have acquired Irisaĝrig tablets sometime after 2003, at least four years after the establishment of the CUNES Collection. Owen himself has confirmed this fact, writing in 2013 that the Irisaĝrig tablets he was publishing had only appeared over the preceding eight years (Owen 2013a: 28).

Rosen’s initial donation to Cornell was said to be associated with a significant tax break (D’Arcy 2003). Customs documents obtained by Benjamin Studevent-Hickman under the Freedom of Information Act dating to 2004 but reporting an investigation opened in 2001 recorded an import of 1,679 tablets valued in total at less than $50,000, which secured a $900,000 tax deduction when they were donated to Cornell (Studevent-Hickman 2018: 212-213). The name of the donor was redacted but subsequently revealed to be Rosen (Felch 2013). The investigation concluded that any allegations of wrongful import were unsubstantiated. Based on the evidence of the customs documents, a rough calculation suggests that Rosen purchased each tablet for $30 and donated it to Cornell for an appraised value of $536. Multiplying up, he could have bought the entire CUNES collection of 10,435 objects for about $313,000 and donated them to Cornell for a tax deduction worth $5.6 million. Writing about the Green family’s Hobby Lobby tax deduction schemes, Candida Moss and Joel Baden (2017: 24) state that “the magic ratio for the Greens was 1:3: for a given investment to be financially viable, they had to be able to write it off at three times the amount that they purchased it for”. If the figures suggested for his donations to Cornell are anywhere near correct, Rosen was working to a ratio of 1:18. Some of his profit was ploughed back into Cornell for research support, though presumably that too would be tax deductible.

In November 2013, Rosen’s attorney was reported as saying that Rosen had always intended that the tablets should reside permanently in a public institution for scholarly research (Felch 2013). But it seems that Cornell was planning on returning the CUNES objects to Iraq (Felch 2013). At about the same time, David Owen wrote in apparent contradiction of Rosen’s attorney that the plan had always been for the tablets to be returned (he used the word donated) to Iraq once they had been conserved and published (Owen 2013: 352). Not surprising really. Once Rosen had extracted what money he could from the tax system and Owen and his colleagues had extracted the historical information of the texts, why would Cornell want to pay in perpetuity for the curation and conservation of the tablets? Better to give them back to Iraq and let the Iraqi taxpayer pay. But where is the justice in that? The return of the tablets alone, shorn as they are of their monetary and informational values, cannot compensate Iraq for the losses caused by their removal (Brodie 2020). Surely the beneficiaries should be prepared to share some of their profits with Iraq, which is after all the ultimate loser in this dubious business. If Cornell really is planning to return the tablets, at the very least it should make a financial contribution towards the long-term costs of storage once the tablets are back in Iraq. But what would be an appropriate sum of money?

Negotiations are currently under way between Hobby Lobby and Iraq over the return of 6,500 cuneiform objects (MOTB 2020; Brodie 2020). In a welcome precedent, part of the plan is for Hobby Lobby to provide financial support for the ongoing study and conservation of cultural objects, rumoured to be in the region of $15 million (Moss 2020). How exactly that money might be spent is yet to be determined. My own opinion is that it should be used to establish a trust fund for the future financial support of Iraqi cuneiform scholarship. But the point is that for 6,500 tablets, Hobby Lobby has judged appropriate recompense for losses suffered by Iraq to be $15 million, or about $2,300 for each tablet. Applying the same valuation to the 10,435 CUNES tablets, it would imply that Cornell might choose to pay Iraq $24 million. That is a lot of money. Perhaps Rosen might want to help out.


Brodie, Neil. 2020. Restorative justice? Questions arising out of the Hobby Lobby return of cuneiform tablets to Iraq, Revista Memória em Rede 12: 87-109.  

D’Arcy, David. 2003. Collector gets tax break for donating cylinder seals to university, Art Newspaper, no. 139, September: 5.

Felch, Jason. 2013. Cornell to return 10,000 ancient tablets to Iraq, Los Angeles Times, 3 November.

Gottlieb, Martin and Barry Meier. 2003. Ancient art at Met raises old ethical questions, New York Times, 2 August.

Mayr, Rudolf. 2007. Acknowledgements, in David I. Owen and Rudolf H. Mayr, The Garšana Archives, (Cornell University Studies in Assyriology and Sumerology. Volume 3), ix-x. Bethesda: CDL Press.

Molina, Manuel. 2013. On the location of Irisaĝrig, in Steven Garfinkle and Manuel Molina(eds), From the 21st Century BC to the 21st Century AD, 59-88. Winona Lake: Eisenbrauns.

Moss, Candida. 2020. Is Iraq getting screwed in a looted treasures deal with Hobby Lobby? Daily Beast, 27 August

Moss, Candida and Joel Baden. 2017. Bible Nation: The United States of Hobby Lobby. Princeton: Princeton University Press.

MOTB. 2020. Statement on past acquisitions, Museum of the Bible, press release, 26 March.

Owen, David. 2007. Acknowledgments, in D.I. Owen and R.H. Mayr, The Garšana Archives, (Cornell University Studies in Assyriology and Sumerology. Volume 3), vii-ix. Bethesda: CDL Press.

Owen, David. 2013a. Cuneiform Texts Primarily from Iri-Saĝrig/Āl-Šarrākī and the History of the Ur III Period. Volume 1: Commentary and Indexes, (NISABA 15). Bethesda: CDL Press.

Owen, David. 2013b. Cuneiform Texts Primarily from Iri-Saĝrig/Āl-Šarrākī and the History of the Ur III Period. Volume 2: Catalogues and Texts, (NISABA 15). Bethesda: CDL Press.

Owen, David I. and Rudolf H. Mayr. 2007. The Garšana Archives, (Cornell University Studies in Assyriology and Sumerology. Volume 3). Bethesda: CDL Press.

Studevent-Hickman, Benjamin. 2018. Sumerian Texts from Ancient Iraq: From Ur III to 9/11. Atlanta: Lockwood Press.

Viano, Maurizio. 2019. On the location of Irisaĝrig once again, Journal of Cuneiform Studies 71, 35-52

Capacity degrading

There is a lot of public money being spent these days on capacity building projects designed to help protect cultural heritage in the Middle East. I am not too convinced personally that professional training of this type translates well into cultural heritage protection, but at least some people in the area are benefiting from opportunities that were not previously available, so I can’t complain too much. What I do want to complain about is the opposite of capacity building, what I propose to call ‘capacity degrading’. What is capacity degrading? I intend it to mean reducing a national fund of professional expertise or competence in such a way as to diminish the public good – the opposite of capacity building in fact.

Let us look at Hobby Lobby again. I came across a comment published in 2014 by someone in a position to know that the Hobby Lobby collection contains an ‘enormous collection’ of cuneiform tablets, an observation that chimes well with Hobby Lobby’s own claim to possess ‘One of the largest collections of cuneiform tablets in North America’. So even after the US Customs seizures and returns Hobby Lobby will still retain a large holding of cuneiform tablets that are destined to be studied and published by members of the Green Scholars Initiative. The Green Scholars Initiative comprises ‘Scholars from 60 participating colleges, universities and seminaries around the globe’, but there is no evidence that any of them are Iraqi scholars from Iraqi universities. Similarly, we can look at the Cuneiform Library at Cornell University that holds approximately 10,000 cuneiform tablets formerly in the possession of Jonathan and Jeanette Rosen. This material is being studied and published with commendable alacrity, but again without the visible participation of any Iraqi scholars or universities.

No satisfactory account has ever been offered as to the source of all these tablets. They are widely believed to have been moved illegally out of Iraq in the years following 1990, and are now unavailable to Iraqi scholarship. Furthermore, they are being used to further the careers in Europe and North America of the next generation of cuneiform specialists, none of whom are Iraqi. So Iraq has suffered a double loss, first of the tablets themselves, and then of the intellectual or cultural capital that the tablets engender. Hopefully the next generation of Iraqi specialists is being trained elsewhere. I don’t know. Otherwise, going forward, Iraqi universities might struggle to re-establish themselves as international centres of excellence in the field of cuneiform studies, which is after all the study of Iraq’s history. There will be a long-term loss to the cultural and intellectual life of Iraq, a diminishment of the same public good that capacity building projects are intended to enhance. Thus while projects such as the British Museum’s Iraq Emergency Heritage Management Training Scheme, run in collaboration with the Iraq State Board of Antiquities and Heritage, are busy building capacity, other institutions are just as busily degrading it. Governments and their taxpayers might be excused for asking why their capacity building efforts are being undermined in this way.