In his paper, “Un motivo per cui le tavolette amministrative neo-sumeriche sono così numerose”, Francesco Pomponio (1992) drew attention to several groups of forged casts of tablets. These forgeries are relatively unusual because they have been so well documented. The aim of this note is to take that study further for the forged Ur III tablets from Girsu that were considered in that paper. These were most probably the work of a single forger and it is interesting to note the way that they are now spread around Western collections.

These forged copies of tablets are itemised in Table 1, which extends the list noted by Pomponio. [The groups of copies are identified here using their CDLI numbers and all references to CDLI are with respect to downloads in May 2015.] Table 1 shows that the copies are mostly clustered together within three collections.

In the case of the relevant tablets in the British Museum, the majority have registration numbers of the form 1901-02-09, nnnn indicating that they were all registered on 9^th February 1901. The copied tablets from the University of Pennsylvania are all numbered within the range UM 29-15-951 to 962 (except possibly NATN 682).

There are also numerous duplicates amongst the tablets published by Legrain (1913) from the Cugnin collection. Legrain was very much aware of the presence of these casts and states that they are very skilfully made and notes that in one example there are 9 copies of the same tablet. [It is convenient here to use the numbering of the texts in Legrain's paper, so that Cugnin 21 refers to the tablet with the text given by Legrain as no. 21, and so on. This note is concentrating on Ur III texts from Girsu so Cugnin tablets 28, 30-41, 57, 82, 88, 102 & 107 will not be considered here.]

The clustering of these forged Ur III Girsu tablets within these collections strongly suggests that they are the work of the same forger.

Cuneiform tablets can be forged either by writing by hand on clay, attempting to copy ancient tablets, or by creating casts of ancient tablets. In the present case, the forged tablets have been created by separately making copies of the obverse and reverse using moulds and then joining together the two halves. Pomponio notes that the forged tablets have two characteristics. Firstly, the signs are formed in a shallower way than normal; and, secondly, there is usually evidence of a seam where the obverse and reverse were joined. For some forgeries, this seam is disguised by partially covering it with seal impressions (Leichty 1970). In others, the edges are smoothed and polished to remove signs of the join (Hilgert 1997). However, in the examples shown by Pomponio, neither of these has been done.

With the exception of P127604, all of the tablets listed in Table 1 are small (typically, 27 × 31 × 15 mm) with a few lines of text on each side. One might assume that the copies of larger tablets in the Cugnin collection, with more lines of text, were either the work of a different forger or that the same forger had produced them later when his skills and confidence had developed.

As already noted, all but one of the British Museum tablets listed in Table 1 have registration numbers of the form 1901-02-09, nnnn, indicating that they were all registered on the same day. These were purchased from Élias Géjou, a well-known antiquities dealer who supplied cuneiform tablets to the British Museum. [These include the 184 tablets with registration numbers beginning 1897-05-13 (BM 23103-23286); and the 787 tablets with registration numbers beginning 1899-04-15 (BM 85194-85980).]

Eight of these forged tablets have registration numbers that are clustered together within the short range 1901-02-09, 1101 to 1117. There is probably a strong relationship between the BM registration numbers and the individual containers that held the tablets when they were initially delivered to the museum. Therefore, it seems likely that most of these eight tablets were actually in the same container when they are sold to the British Museum.

This supports similar findings for this group of copies in the Pennsylvania museum and the Cugnin collection. The implication is that the forger released batches of copies into the antiquities market and, in these three instances, these forgeries remained within those batches when they were eventually sold to their current owners. It is not clear how long the chain was between the forger and the international dealers who sold tablets to the museums. However, it seems unlikely that a dealer, such as Géjou, would want to risk his reputation by incorporating a small number of forgeries within a sale of a much larger number of tablets. Furthermore, it also seems unlikely that a dealer would knowingly leave copies of the same tablet within the same batch, since this increases the likelihood of the copies being spotted.

On the basis of the available evidence, the forged tablets in Table 1 from Chicago (A), Couvent Saint-Etienne, Jerusalem (SE) and the John Rylands Library, Manchester (JRL) appear to be singletons rather than indicators of a larger group of copies.

For Ur III specialists, the warning is to be particularly watchful for the possibility of forgeries within British Museum tablets with registration numbers beginning 1901-02-09, as well as tablets from the University of Pennsylvania within the range UM 29-15-951 to 962 and tablets from the Cugnin collection.

However, forgeries in the British Museum are not by any means confined to those with registration numbers beginning 1901-02-09. In their catalogue, Sigrist et al. (2006: 306) note 60 forgeries (45 made from moulds and 15 hand-written). These are scattered around several groups of purchases, although a relatively large number (i.e., 38) were from the same batch, purchased from William T. Burbush, Castle Bromwich, Birmingham (with registration numbers of the form 1898-02-15, nnnn). These forgeries currently remain unpublished and so it is not possible to determine how far such copies have found their way into other collections.

Table 1