Cuneiform Digital Library Journal
ISSN 1540-8779
© Cuneiform Digital Library Initiative

CDLI Publications
Editorial Notes

PDF Version of this Article
Get Acrobat Reader
Download Cuneiform Font


A New Edition of the Proto-Elamite
Text MDP 17, 112

Laura F. Hawkins <>
University of Oxford

Late Uruk, Proto-Elamite, proto-cuneiform, decipherment


§1. Introduction: The Model Contracts
§1.1. MDP 17, 112, like most proto-Elamite tablets, was first published in the early part of the twentieth century when understanding of both the writing system and the society that produced it were still relatively poor. Recent advances in the study of early Iran and the proto-Elamite writing system in particular have prompted a re-evaluation of the corpus (Englund 2004; Dahl 2013).


§1.2. The very difficult text was published first by V. Scheil in 1923. The present article is based on a new collation (Musée du Louvre, July 2011) that corrected several errors and helped illuminate the overall structure of the text. MDP 17, 112, is a small clay tablet from Susa, inscribed on both its obverse and reverse in the proto-Elamite script. Dating of proto-Elamite tablets is problematic; however, based on certain criteria established for determining the relative chronology of proto-Elamite texts (such as the thickness of the tablet, the format of the text, and the sign forms), it is likely a relatively late text (Dahl et al. 2013). MDP 17, 112, is an administrative document—with two exceptions the only genre identified in the proto-Elamite corpus thus far—and most likely records a roster or inventory of dependent workers belonging to a particular household or institution.


§2. The Text

0. M327+M342 (Concerning the institution/household) “M327+M342.”
1. M217h ˹M59˺ M387c M218 M370? (Group 1:) “M218 M370,” (who is the) “M217h M59 M387c ,” (has:)
1a. M317 , 1N1 1 M317,
1b. M54 , 2N2 2 M54,
1c. M72 , 1N1 (and) 1 dependent female worker.
2. M323 M124 M386a M240 (Group 2:) “M386a M240,” (who is the) “M323 M124,” (has:)
2a. M72 , 2N1 2 dependent female workers.
3. M210d M111l M388 M32 M387 M218 (Group 3:) “M32 M387 M218,” (who is the) “M210d M111l M388,” (has:)
3a. M317 , 1N1 1 M317,
3b. M54 , 2N2 (and) 2 M54.
4. M112n M388 M218 M386a (Group 4:) “M218 M386a,” (who is the) “M112n M388,” (has:)
4a. M317 , 1N1 1 M317,
4b. M54 , 2N2 (and) 2 M54.
5. M93a M388 M285ba1 (Group 5:) “M285ba1,” (who is the) “M93a M388,” (has:)
5a. M317 , ˹1N1˺ 1 M317,
5b. M54 , 2N2 2 M54,
5c. M72 , 1N1 (and) 1 dependent female worker.
6. M387l M372a M388 M49c1 M230a1 M371 (Group 6:) “M49c1 M230a1 M371,” (who is the) “M387l M372a M388,” (has:)
6a. M317 , 1N1 1 M317,
6b. M3b , 1N1 1 M3b,
6c. M54 , 1N1 1 M54,
6d. M373 , 1N1 1 M373 ...
6e. M72 , 1N1 (and) 1 dependent female worker [scribal error!—written after l. 7d]
7. M112o M388 M241 M33 M371 (Group 7:) “M241 M33 M371,” (who is the) “M112o M388,” (has:)
7a. M317 , 1N1 1 M317,
7b. M3b , 1N1 1 M3b,
7c. ˹M373˺ M54 , 1N1 1 M373 M54,
7d. M373 , 1N1 1 M373,
7e. M72 , 1N1 (and) 1 dependent female worker.
8. M51a M388 M218 M229e M371 (Group 8:) “M218 M229e M371,” (who is the) “M51a M388,” (has:)
8a. M317 , 1N1 1 M317,
8b. M54 , 1N1 1 M54,
8c. M46 , 1N1 (and) 1 M46.
9. x M388 x M295a? M218 (Group 9:) “x M295a? M218,” (who is the) “x M388,” (has:)
9a. M317 , 1N1 1 M317,
9b. M54 , 1N1 (and) 1 M54.
10. M112p M66a M66a (Group 10:) “M66a M66a,” (who is a) “M112p,” (has:)
10a. M317 , 1N1 1 M317,
10b. M3b , / 1N1 (and) 1 M3b.
10c. M54 ˹M372˺ x “M54 M372 x,” (who is also a) “M112p,” (has:)
10d. M72 , 1N1 1 dependent female worker.
11. M377e+M377e M388 M4 M4 M218 (Group 11:) “M4 M4 M218,” (who is the) “M377e+M377e M388,” (has:)
11a. M317 , 1N1 1 M317,
11b. M3b , 1N1 (and) 1 M3b.
12. M54 / M370 ˹M386a?˺ M386a (Group 12:) “M386a? M386a,” (who is the) “M54 M370,” (has:)
12a. M373 , 1N1 1 M373,
12b. M72 , 1N1 (and) 1 dependent female worker.
13. M51a M388 ˹M24˺ M33 M371 (Group 13:) “M24 M33 M371,” (who is the) “M51a M388,” (has:)
13a. M317 , 1N1 1 M317.
14. ˹M218 M373?˺ / M387 M372a M388 (Group 14:) “M297+M296 M371(?),” (who is the) “M218 M373? M387 M297+M296 M371 M372a M388 (?),” (has:)
14a. M317 , 1N1 1 M317,
14b. M54 , 1N1 1 M54,
14c. M72 , 1N1 (and) 1 dependent female worker.
1. M317 , 1N14 2N1 (Total:) 12 M317,
2. M54 , 1N14 7N1 17 M54,
3. M373 , 3N1 3 M373,
4. M72 , 9N1 9 dependent female workers,
5. M46 , 1N1 (and) 1 young dependent worker.
6. M139? M32 M387 M218 (Postscript:) M139? M32 M387 M218


figure 1a figure 1b

Figure 1: MDP 17, 112
Note that these images reflect the tablet being rotated around its long, vertical axis, not its shorter, horizontal axis.


§3. Text Commentary
§3.1. Header
§3.1.1. Most proto-Elamite tablets begin with a sign that acts as a header for the entire text, and MDP 17, 112, is no exception. Headers are single and, often, complex graphemes that may represent the individual, household, or institution to which the document or entities recorded belong (Englund and Damerow 1989: 13). These signs correspond to similar information recorded at the end of some proto-cuneiform texts (Englund 2004: 106 n. 12).


§3.1.2. The header in MDP 17, 112, is M327+M342[1], which corresponds to sign M327a in Meriggi’s sign list (1974). This sign occurs in at least 35 published proto-Elamite texts (including MDP 17, 112). It is the header in all of these texts except for MDP 17, 132, and MDP 17, 146, in which it occurs in the body of the texts and appears to be modifying an entry that is being counted.


§3.1.3. Only four tablets containing the header M327+M342 are sealed, and they do not all bear the same seal. The seals associated with M327+M342 are: PES 198 (MDP 6, 233), PES 103 (MDP 17, 489), and an unidentified seal (MDP 17, 50; MDP 26, 52) (Legrain 1921; Scheil 1935; Amiet 1986).


§3.2. Entries
§3.2.1. The main entries in this text start on the obverse and continue onto the reverse, rotating the tablet on its vertical axis. MDP 17, 112, is a complex text compared to the standard proto-Elamite corpus because each entry in this text contains at least one, and sometimes several, sub-entries. Each main entry consists of a long string of non-numerical signs that can be analyzed as consisting of signs from two distinct sub-groups of the signary: owner signs and signs for which a syllabic value has been suggested (Englund 2004: 127; Dahl 2013). Almost all of the owner designations mentioned here occur in other proto-Elamite texts, indicating that they might indeed be standard professions or titles that are common among the proto-Elamite households or institutions that used this writing system.


§3.2.2. Many of the sub-entries do not occur in any other proto-Elamite text, which would be unusual if they represented positions or titles. Instead, based on the structure of the strings of signs there is a high likelihood these are syllabically written words or names, which could either designate the individual who fills the position or title preceding the personal name, or indicate an individual who is responsible to the individual holding the position or title preceding the personal name.


§3.3. Total
§3.3.1. Most proto-Elamite texts contain a total of all account entries. This total is always written on the reverse of the tablet, rotated around its vertical axis (Englund 1996: 161). Proto-Elamite totals can be comprehensive and complex, with each product individually totaled, or they can be simplified totals in which some of the entries being counted on the obverse are subsumed into another group of counted entities on the reverse.


§3.3.2. As expected, MDP 17, 112, ends with a total on the reverse of the tablet. In the main body of the text, there are seven groups of entries on the obverse and seven on the reverse. The total on the reverse of this tablet draws attention to a common feature of the proto-Elamite writing system: the sub-grouping of certain signs as other signs in the totals of the texts. The total on the reverse only contains five entries: M317, 1N14 2N1; M54, 1N14 7N1; M373, 3N1; M72, 9N1; M46, 1N1. Therefore only five items (or groups of items) are being totaled. This suggests that some of the items counted in the main body of the text are being grouped with other entries and therefore totaled under a different sign in the summary. All the entries and numbers in the total of the text correspond to the numbers of the entries in the main body of the text except for M54. There are seventeen M54 in the total but only thirteen in the main body of the text. However, there are four M3b in the main body of the text that are not accounted for in the total. If one counts these towards the M54 total then the numbers add up and there are no numerical entries in the main body of the text that are unaccounted for in the summary, and vice versa. Since M3b is subsumed under M54 in the summary we can conclude that these two signs are likely to be semantically related.


§4. General Commentary
MDP 17, 112, is a typical proto-Elamite text in many ways. It begins with a header, ends with a total, and contains a number of entries and sub-entries that include numerical notations. There are, however, several features of this text that make it unusual and that are worth noting. These include the use of the numerical notation N2 (which in this text count the objects or individuals being represented by the sign M54), the presence of a scribal error, and the inclusion of a postscript.


§4.1. N2 in Proto-Cuneiform
§4.1.1. N2 occurs in both proto-Elamite and proto-cuneiform. In the proto-cuneiform corpus, this sign belongs to the derived sexagesimal System S' and has the numerical value of “1.” A quick look at the proto-cuneiform corpus informs us that the System S' was used in a limited number of contexts: lexical lists concerning animals (W 20266,55), fish (W 21916,2; W 14777,b; W 21208,49), geography (W 21208,50), general vocabulary (W 14337,a+), and one lexical list of undetermined type (W 20266,175); texts concerning liquids (Nissen et al. 1993: 28), particularly beer and beer products; animal husbandry texts; texts possibly concerning foodstuffs; and a few texts of unknown content.


figure 2

Figure 2: Sexagesimal System S' in Proto-Cuneiform


§4.1.2. The precise meaning and function of N2 in proto-cuneiform is not known. It has been suggested that N2 in proto-cuneiform is the precursor to the sign BAD, with the later readings uš2/ug7 attested in the Ur III and Old Akkadian periods (Damerow and Englund forthcoming), including at Ebla (Butz 1981: 338), and in these later texts it is clear that it bears the meaning “sacrificed,” or in the case of humans, simply “dead.” In order to evaluate whether or not N2 could mean “dead” or “sacrificed” in the proto-cuneiform texts as well, we must look to the animal husbandry texts; the precise meaning “dead” or “sacrificed” cannot realistically be applied to lexical lists, foodstuffs, or beer products. Even in these texts, the function of N2 is not obvious. Therefore the evidence we have to connect N2 with “dead” animals is suspect. It comes from texts that are 1000 years younger than the proto-cuneiform texts, and from a proposed connection between N2 and later BAD. The proposed sign evolution that lead the trend to associate N2 with BAD—and therefore with the meaning “dead”—is diagrammed below.


§4.1.3. The only possible evidence for the derived sexagesimal System S' being used to denote dead or sacrificed animals in the Uruk corpus comes from Green’s (1980, 7-8) observation on Uruk animal husbandry texts:

The other two subtotals included in the grand total are designated ŠU and X [viz, SZU and X]. Unfortunately, the poor state of preservation of the tablets hinders the interpretation of these entries. The only tablet with both obverse and reverse preserved (no. 48) includes only the subtotal ŠU, which clearly is the sum of the subcases in obv. i. In comparison, text no. 39 has two types of subcase entries, one with additional remarks and one with only a numeral and no remarks, these perhaps corresponding to the subtotals ŠU and X, respectively. A peculiar numeral sign—like the normal impressed circle or semicircle but with a horizontal wedge through it—is used for these entries. In texts with preserved subtotals it is used only for the subtotal X, never for the subtotal ŠU. Comparison of subtotals and totals shows this numeral to be equivalent in value to a numeral without the horizontal wedge. An examination of the bookkeeping remarks associated with the subtotals ŠU and X points to a cultic significance. Where legible they include SIG-dINANNA, dINANNA, and E2-NUN. Perhaps comparable is UDU-ŠU mentioned in two Old Babylonian texts as selected for, or on the occasion of, a “great festival.” If indeed these entries do record animals designated for cultic purposes, a guess for the identification of the sign X is SIZKUR, ‘sacrificial animal.’

§4.1.4. Even if N2 is used to denote dead or sacrificed animals in the Uruk corpora, the evidence for this is weak and the texts themselves are inconclusive. It is more likely that N2 in proto-cuneiform denotes the entities it counts as “other” or “special.” The meaning “dead” may then have been later attributed to it due to the confluence of the N2/“TIL” and “BAD” signs.


§4.2. N2 in Proto-Elamite
§4.2.1. In MDP 17, 112, N2 is used to count M54 in four different entries (lines 1b, 3b, 4b, and 5b), and in each occurrence two N2’s are listed (in other entries, however, M54 is counted with N1). Based on the total on the reverse, each of these N2’s must be equal in value to the sign N1 in order for the total to equate the entries on the obverse. When totals are present on the reverse of tablets, N2’s are always counted towards the summed totals (versus being subtracted from them) and objects counted with N2 are never totaled separately from the same objects being counted by N1. Therefore there is no total in the proto-Elamite corpus that includes the sign N2.


§4.2.2. As demonstrated in MDP 17, 112, N2 and N1 do not occur in exclusive contexts: the object represented by the sign counted with N2 can also be counted with N1, even in the same text. This holds true for all objects counted by N2 in the proto-Elamite corpus, including the following signs in ten texts (for the graphic forms of these signs, cf. figure 5 at the end of the article):


M1062+M288 (MDP 6, 252: 2, 6; 26S, 5212: 2, 4)
M– or M72? (MDP 6, 290: 6)
M72 (MDP 6, 316 + 322 + 234 + 26S, 335 + Sb 15247: 40'; 26S, 336: 35)
M54 (MDP 6, 316 + 322 + 234 + 26S, 335 + Sb 15247: r. 21'; 17, 112: 2b, 3b, 4b, 5b; 26S, 335: r. 11)
M3b (MDP 6, 316 + 322 + 234 + 26S, 335 + Sb 15247: r. 40'; 26S, 335: r. 24)
M57a (MDP 6, 383: 8)
M388 (MDP 6, 290: 2; 17, 120: 16, 30, 36)
M346 (MDP 17, 350: 4)[2]


Every sign listed above most likely represents some kind of dependent worker (except M346, if that is indeed what is being counted in MDP 17, 350). Therefore it appears that N2 is used exclusively, or almost exclusively, to count various types of dependent workers – as opposed products, commodities, or animals as in the proto-cuneiform corpus. If this is true, then proto-Elamite N2 and proto-cuneiform N2 have nothing in common apart from their graphical similarity.


figure 3

Figure 3: The Early Convergence of N2 and BAD.


§4.2.3. If we look at the text MDP 17, 120, we can see that N2 is used to count M388 in line 36, as well as M66 and M371 in lines 16 and 30, respectively. On closer inspection though, the sign M388 also occurs in every entry that N2 is present. There are two possible explanations for this: 1) M66 and M371 are in fact the entities being counted and it is a coincidence that M388 occurs in every entry; 2) M388 is being counted in all instances and the presence of other signs after M388 indicates that these signs are most likely modifiers or personal names, and can occur either before or after M388 possibly according to their linguistic class (i.e. personal names, adjectives, etc.) The second possibility is most likely. In fact, if we look at the “syntactic” structure of proto-Elamite entries, we can see a pattern where signs that are most likely modifiers or “adjectives” occur before the noun they are modifying, and strings of signs that are more likely personal names occur after the noun they are governing. Line 20 of MDP 17, 120, illustrates this pattern when compared to line 4 of the reverse (part of the total):


20) M387i M387i M388 M272 M66 M371 , 1N1
4') M387i M387i M388 , 2N14 5N1


In this text the entity “M387i M387i M388” is being counted, as indicated in the total on the reverse, so those three signs act as a single unit in line 20 on the obverse. Based on other proto-Elamite texts, we know that M388 is a type of dependent worker, so M387i M387i must act as a modifier indicating that the three signs together denote a particular type of M388. The final three signs, which are all likely candidates for signs with syllabic values, might be the syllabically written personal name of this particular “M387i M387i M388.”


§4.2.4. In addition to acting as a numerical sign, N2 also has a non-numerical value (transliterated as M386a); such polyvalence is not uncommon in proto-Elamite (Englund 2004: 113). M386a is classified as one of the 80-120 signs that are likely to carry a syllabic value. It occurs as a non-numerical sign in over 50 texts.


§4.2.5. As is illustrated above, the precise meaning and function of the sign N2 in the proto-Elamite corpus is also not entirely understood. Similar to the proto-cuneiform corpus, N2 carries a numerical value in proto-Elamite and is a graphical variant of N1 with the addition of a single vertical stroke inside the body of the sign. It seems to carry the same numerical value as N1, which represents one unit in both the decimal and sexagesimal numerical and metrological systems. Proto-Elamite N2 on the other hand only exists in the decimal system. It is possible that proto-Elamite N2 is a borrowing from proto-cuneiform, like most other proto-Elamite numerical signs were, and we could therefore look to proto-cuneiform to help deduce how N2 was being used in proto-Elamite. There are several problems with this, however. The first is that the rest of the proto-cuneiform System S' (N15 and N35) is not attested in proto-Elamite. Second, as was already mentioned above, N2 in proto-Elamite only operates in the decimal system so far as is attested, while N2 in proto-cuneiform operates in the sexagesimal system. Third, the function of N2 and the rest of the System S' in proto-cuneiform is not well established, as was illustrated above. Last, the types of entities being counted with N2 in proto-cuneiform (animals, beer jars, foodstuffs, and lexical entries) are incongruent with what N2 is being used to count in the proto-Elamite corpus (dependent workers).


§4.2.6. There is another explanation, however. Derived signs are common in the proto-Elamite script, and can be made by modifying basic signs by the addition of single wedges, dots, lines surrounding the sign, dots surrounding the sign, small wedge-like impressions surrounding the sign, or by other means. The derived signs most likely represent semantic or phonological variants on the original sign. Their prevalence in the writing system indicates that we should not assume that proto-Elamite N2 was necessarily a direct borrowing from proto-cuneiform. I suggest that proto-Elamite N2 was an independent development derived from proto-Elamite N1 (which, along with the other basic numerical signs, was a direct borrowing from proto-cuneiform). Regardless of whether or not it was an independent development in proto-cuneiform and proto-Elamite, it is clear that N2 was used to designate whatever it was used to count as “other,” or distinct from the usual or majority object being counted. In the case of proto-Elamite this could possibly indicate a difference in gender, in worker status, or in type or amount of rations to be received. In proto-cuneiform, perhaps N2 indicated that a different type of grain was used to make the beer or different additives were infused into the beer in the case of beer texts, that certain animals were already designated for a certain purpose or function in the case of animal husbandry texts, and similar designations in lexical lists.


§4.3. A Scribal Error
§4.3.1. The addition of one M72 in the bottom margin of the obverse (in line 6e of the transliteration, actually in the middle of group 7), which Scheil mentions in his initial publication of the text (1923: 16) can only be interpreted as an emendation to the previous group of entries. Since this tablet maintains a fixed order of counted entities in each entry, we can ascertain that M72 was written directly below the position in which it should have been written if it had been originally included in the chain of entries for group 6; that is, as the last entry in the sequence.


figure 4

Figure 4: The Peculiarities in MDP 17, 112


§4.3.2. Scribal errors of this sort are very rare in the proto-cuneiform record. This text therefore provides a rare example of such a situation, and from it we can conclude that one method of correcting scribal errors—at least in the proto-Elamite writing system—is to insert the missing sign(s) in the lower margins of a line, directly beneath where the entry should have been written in the first place.


§4.3.3. Scribal errors occur more commonly in the proto-Elamite corpus than in the proto-cuneiform corpus (for examples of errors in proto-cuneiform, see Nissen, Damerow, and Englund 1993: 42), especially as arithmetical incongruities between the entries and the total.


figure 5

Figure 5: List of Relevant Proto-Elamite Signs


For example, if we look at MDP 6, 361, we can see that there are three entries counting M376 (in the second entry M376 must be implied even though it is not expressly written) on the obverse totaling 39 units; on the reverse, the total was originally written as being 59[3] units of M376. At least one of the N14’s appears to be erased, but there is still one more than we would expect to find based on the entries on the obverse, which leaves the apparent total at 49 units.


§4.4. Postscript
§4.4.1. The presence of postscripts (also called subscripts in other publications, see Dahl 2012; 2013) in the proto-Elamite corpus is rare, found in only about 100 out of over 1500 published proto-Elamite texts. Postscripts range from a single sign to a string of six signs, and it is still unclear what information they contain, what purpose they serve, and why some texts include them and others do not. The postscript in MDP 17, 112, is written after the total and consists of the following four signs: M139? M32 M387 M218. This combination of signs does not occur in any other published proto-Elamite tablet, either as a postscript or otherwise.


§4.4.2. Although at this stage in the decipherment it is not possible to explain the presence of postscripts on certain Proto-Elamite tablets, preliminary investigations suggest that there may be a relationship between the postscript on a tablet and the header. Otherwise, there does not appear to be any relationship between postscripts and the presence or type of seal on the tablet, the entities being recorded, or other such markers. Due to the wide number of different postscripts, it is unlikely that they all represent a calendrical date, though they could reference festivals or other events. Another possibility is that they are a syllabic or pseudo-syllabic rendering of the name of the scribe who wrote the tablet.


§5. Conclusion
In many ways the text MDP 17, 112, is typical of the proto-Elamite corpus; however, it also illustrates a number of peculiarities of the proto-Elamite writing system (see figure 4), such as the use of the numerical value N2, how proto-Elamite scribes accommodated a scribal error, and the presence of a postscript (illustrated in figure 4). While further research is necessary to fully understand how these features operate in the proto-Elamite writing system, I hope this article has demonstrated that these are indeed interesting aspects of the writing system that do warrant future study.





Amiet, Pierre
1986 L’âge des échanges inter-iraniens 3500-1700 avant J.-C. Editions de la Réunion des musées nationaux. Paris: Ministère de la culture et de la communication.
Butz, Killian
1981 “Zur Terminologie der Viehwirtschaft in den Texten aus Ebla.” In Luigi Cagni., ed., La lingua di Ebla. Naples: Istituto Universitario Orientale, pp. 321-353.
Dahl, Jacob L.
2012 “The Marks of Early Writing.” Iran 50, 1-11
2013 “Early Writing in Iran.” In Potts, D. T., ed., Oxford Handbook of Iranian Archaeology. Oxford: Oxford University Press, pp. 233-262.
Dahl, Jacob L., Petrie, Cameron A.
& Potts,Daniel T.
2013 “Chronological Parameters of the Earliest Writing System in Iran.” In Cameron A. Petrie, ed., Ancient Iran and Its Neighbours: Local Developments and Long-range Interactions in the 4th Millennium BC. Oxford: Oxbow Books. pp. 353-378.
Damerow, Peter
& Englund, Robert K.
1989 The Proto-Elamite Texts from Tepe Yahya. American School of Prehistoric Research Bulletin 39. Cambridge, Mass.: Harvard University Press.
forthcoming The Proto-Cuneiform Texts from the Erlenmeyer Collection.
Englund, Robert K.
1996 “The Proto-Elamite Script.” In Peter T. Daniels
& William Bright, eds., The World’s Writing Systems. Oxford: Oxford University Press, pp. 160-164.
2004 “The state of decipherment of proto-Elamite.” In Stephen D. Houston, ed., The First Writing: Script Invention as History and Process. Cambridge: Cambridge University Press, pp. 100-149.
Green, Margaret W.
1980 “Animal Husbandry at Uruk in the Archaic Period.” JNES 39, 1-39.
Legrain, Leon
1921 Empreintes de cachets élamites. MDP 16. Paris: Ernest Leroux.
Meriggi, Piero
1974 La scrittura proto-elamica. Parte IIa: Catalogo dei segni. Rome: Accademia Nazionale dei Lincei.
Nissen, Hans J., Damerow, Peter
& Englund, Robert K.
1993 Archaic Bookkeeping: Early Writing and Techniques of Economic Administration in the Ancient Near East. Chicago: University of Chicago Press.
Scheil, Vincent
1905 Documents archaïques en écriture proto-élamite. MDP 6. Paris: Ernest Leroux.
1923 Textes de compatibilité proto-élamites. MDP 17. Paris: Ernest Leroux.
1935 Textes de compatibilité. MDP 26. Paris: Ernest Leroux.

Version: 2 May 2015