INTRODUCTION

The Sumerian epic Enmerkar and the Lord of Aratta, composed sometime in the second half of the third millennium B.C., provides a famous etiology of the cuneiform writing sys­tem.

bar kin-gi₄-a ka-ni dugud su nu-mu-un-da-an-gi₄-gi₄-da-ka en kul-ab₄ ^kl-a-ke₄ im-e su bi-in-ra inim dub-gin₇ Tbi-inl-gub

Because the messenger’s mouth was too heavy, and he could not repeat it (the message),

The lord of Kulab (Enmerkar) patted some clay and put the words on it as on a tablet (Vanstiphout 2004: 84-85, lines 502-03).

In the view of the author of these lines, Enmerkar, whose alleged impact on (scribal) culture, if not on writing itself, remained part of Mesopotamia’s cultural memory until very late times,^[47] had created the cuneiform writing system for one main reason: because it had the potential to serve as a far more reliable medium for communication over large distances of space and time than the human memory.

and most important apkallu-sage, Adapa. The first­millennium “Catalogue of Texts and Authors” makes the even more remarkable claim that Enmerkar was the author of Sumerian poetic texts (Lambert 1962: 64-65 [III 3-5], 74). Given his association with writ­ing and scholarship, it is somewhat ironic that the Cuthean Legend of Naram-Sin blames Enmerkar for having failed to compose a monumental inscription (naru) addressed to posterity (Westenholz 1997: 264) — or does this story reflect, as suggested to me by Kathryn Slanski, that according to tradition Enmerkar invented writing on clay but not on stone? For a discussion of some other texts dealing with the origins of Uruk’s association with scribal learning, see now George 2009: 110-11.

It is obvious that a script suitable for such a purpose should have been, ideally, both simple and precise.

The Mesopotamian literati were clearly aware of the possibility of drastically simplifying their writing system, at least with regard to Akkadian texts. In fact, during the Old Babylonian period, Assyrian and Babylonian letter writers made do with a repertoire of no more than 68­82 syllabic signs, all of them representing a very restricted number of different values — and even though this meant that they used less than 10 percent of the 954 graphemes constituting the repertoire of cuneiform signs from all ages,⁵ the clarity of their messages was not in the least compromised (Charpin 2008: 39, 53). Scribes who composed administrative texts dur­ing the same time employed a higher percentage of logograms, but the number of different signs used by them was small as well. Akkadian scholarly texts from the early Old Babylonian period are likewise written with a fairly limited selection of characters — 112 syllabic and 57 logographic signs in the case of the Old Babylonian omen corpus (Charpin 2008: 53).

It would have been easy to reduce the complexity of the cuneiform writing system even further, but somewhat surprisingly, this did not happen. No systematic attempt was ever made by the scribes to dispose of the hundreds of signs and the thousands of possible readings as­sociated with them that were for all intents superfluous. On the contrary: starting with the later Old Babylonian period, when logographic writings of technical terms in the aforemen­tioned extispicy texts became the rule (see Richardson, this volume) and then for more than a thousand years, from the middle of the second millennium to the end of the first millennium B.C., the repertoire of signs used by the scribes, not so much for letters and documents but for scholarly texts, became progressively more complex. For instance, 84 percent of the signs of a typical first-millennium tablet of the terrestrial omen series Summa alu are logograms (Civil 1973: 26), and while in the Old Babylonian period most syllabic values belonged to the rather

simple CV (consonant — vowel) and VC types (ba, ab, etc.), scribes now employed a much larger number of CVC values (bar, sad, etc.). This development is all the more remarkable if one takes into account that the Aramaic alphabet, which became widely used in Mesopotamia in the first millennium B.C., operated with an extremely limited repertoire of characters.

It seems the main reason why the Babylonian and Assyrian scholars continued to culti­vate this graphemic embarras de richesse, and even added to it in later periods, was that they regarded the overabundance of possible meanings associated with the polysemy of the cunei­form writing system as an inexhaustible source of knowledge and wisdom. The Mesopotamian literati of later times believed that language and writing were intimately connected, and that their basic elements, words and signs, were not arbitrarily chosen conventions, as claimed by Aristotle and Saussure, but representations that denoted their objects by nature.^{6 7 *} Consequently, Sumerian and Akkadian words, however obscure and rare, had to be collected in lexical lists to be never forgotten, and so had the numerous signs used to write them. Giving up any of them, or reducing the complexity of their meanings, would have meant to lose access to some particular truth they conveyed.