deeper corpus of pre‐Achaemenid records that documents centuries of previously unknown history. The Elamite corpus, however, proved to be much smaller, less diverse, less dense, and less continuous than the Akkadian corpus. It was also much less tractable. While the early study of Old Persian and Akkadian benefited from flourishing comparative and historical research on related Indo‐Iranian and Semitic languages, early attempts to identify and exploit known languages connected to Elamite were vain or misleading. Evidence from the other versions of the multilingual inscriptions was fundamental to the interpretation of Elamite lexicon and grammar.
Almost all Achaemenid Elamite administrative documents come from two groups excavated at Persepolis in 1933–1934 and 1936–1939. This large corpus differs from the royal inscriptions in contents, in rhetorical register and purpose, in syntactic and stylistic complexity, and in grammatical and orthographic variation. From the late twentieth century on, these practical texts have provided a rich context both for internal analysis of Achaemenid Elamite language and for understanding of contact between Elamite and other written and spoken languages around the Achaemenid courts.
Elamite Language
Achaemenid royal inscriptions normally refer not to Elam as a place but to Elamites as a people (Elamite Hatamtip). The indigenous ancient name of the language is not attested. The consensus modern term “Elamite” reflects Sumerian and Akkadian references to the “language of Elam.” Soon after the decipherments, recognizing that the language of Achaemenid inscriptions of “the second kind” also appeared in many older inscriptions from Susa, some scholars called the language “Susian.” Others, contemplating the royal title found in those older texts, “King of Anzan and Susa,” and surmising that the language was original to a region farther from Mesopotamia than Susa was, called it “Anzanite” – even though the location of Anzan long remained a matter of disagreement (Potts 2011: pp. 36–37).
No ancient languages are known to be cognate with Elamite, and no later languages are known to be descended from Elamite. The hypothesis that Elamite is distantly related to Dravidian languages, a matter of disagreement among modern scholars, has had no firm consequences for the elucidation of Elamite grammar or lexicon. In practical terms, Elamite remains a linguistic isolate (Stolper 2004: p. 61 with references; Starostin 2002 with review of proposals and lexical evidence).
Readable Elamite texts are written in versions of the cuneiform script that was developed in Mesopotamia to record Sumerian and Akkadian. The script was also adapted to record other Near Eastern languages between the third and first millennia BCE, including Eblaite, Hurrian, Urartian, and Hittite. The Elamite adaptation of this writing system is economical by comparison with the Mesopotamian applications of it in two senses. First, the number of characters is comparatively small: slightly more than 200 characters are attested in Elamite cuneiform writing from all periods. Only 100–150 are found in any period, about 135 in Achaemenid Elamite (Steve 1992). Second, the incidence of homophony (many characters with the same signification) and polyphony (single characters with many significations) is comparatively small. Late pre‐Achaemenid and Achaemenid Elamite sign inventories include a comparatively high proportion of logograms (characters representing words) because large groups of administrative records come only from these periods.
In Achaemenid multilingual royal inscriptions, the mostly syllabic orthography of the Elamite versions required fewer characters than the Old Persian but more characters than the Babylonian counterparts. For example, in the seal inscription SDa, the phrase “I, Darius, King” requires 12 characters in Old Persian, 10 in Elamite, and even with the title expanded to “Great King,” 9 in Babylonian. In the main Bisutun inscriptions, the first four columns of the Old Persian text cover c. 27.33 m2, the corresponding three columns of the Elamite c. 21.5 m2, and the corresponding Akkadian c. 15.33 m2 (King and Thompson 1907: pp. xxiii–xxiv). As a result, symmetrical displays of versions often required adjustments of character size or spacing and sometimes motivated editorial differences (cf. Henkelman 2003; Kozuh 2003).
Almost all pre‐Achaemenid Elamite cuneiform texts come from sites in modern Khuzestan and Fars. The northern and eastern extent of the area in which Elamite was spoken is a matter of dispute. The oldest documents from the areas where Elamite cuneiform texts are found are written in two as yet undeciphered scripts, called Proto‐Elamite (using about 1000 characters, found on about 1600 clay tablets, most from Susa, a few from sites in Fars, Kerman, and Seistan, dated c. 3100–2900 BCE, see http://cdli.ox.ac.uk/wiki/doku.php?id=proto‐elamite) and Linear Elamite (using about 100 attested characters, found on 21 or more stone, clay, and silver objects, most from Susa, one probably from Fars, one fragmentary and uncertain example from Seistan, another from Margiana, dated c. 2100 BCE; see http://cdli.ox.ac.uk/wiki/doku.php?id=linear‐elamite). It is plausible but not demonstrable that both scripts record texts in Elamite language. If so, the area in which Elamite was once spoken perhaps included easternmost Iran.
The oldest datable Elamite text written in cuneiform script, a treaty with a Mesopotamian king, comes from about 2100 BCE. Other pre‐Achaemenid Elamite texts include about 210 royal inscriptions on bricks, stelae, reliefs, statues, votive objects, and other portable objects; about 400 administrative texts (almost all in two archival clusters, one from Khuzestan and one from Fars); about 20 letters (some excavated at sites in Assyria and Armenia) and legal texts (from Susa); an omen text and a hemerological text (from Susa). A few Elamite passages or glosses appear in Akkadian texts from Mesopotamia. Elamite names, words, and loanwords appear in Akkadian and Sumerian legal and administrative texts from both Mesopotamia and Iran.
Thus, by the time the Achaemenid kings began to keep written records, Elamite was long established as a language of practical literacy in southern and western Iran. It was used often for royal display and commemoration, sometimes for legal and administrative recording, and occasionally even for scholarly transmission. It was the foremost vehicle by which inhabitants of ancient Iran communicated in writing.
Royal Inscriptions
Elamite versions of about 75 Achaemenid royal inscriptions include rupestral inscriptions, architectural inscriptions displayed on building elements, commemorative inscriptions on stone, clay, gold, or silver tablets associated with royal buildings, and inscriptions on portable objects such as ornamental knobs of Egyptian blue, stone vessels, tableware, weights, cylinder seals, and statues. Almost all are versions of texts that were also presented in Old Persian, Akkadian, and occasionally Egyptian, on the same objects or structures as the Elamite versions, or on companion objects.
Most of the inscriptions with Elamite versions were displayed or deposited at sites in the core provinces of the Achaemenid Empire, in Persia, Elam, or Media. A few were displayed elsewhere, including the Suez inscriptions set up by Darius I in Egypt (DZb, DZc; Lloyd 2007: pp. 99–107) and the rock inscription of Xerxes at the citadel of Van in eastern Anatolia (XV). Inscriptions on the statue of Darius I that was made in Egypt and displayed at Susa included Elamite versions (DSab; Yoyotte apud Perrot 2010: pp. 256–299; Vallat apud Perrot 2010: pp. 312–313), as did multilingual inscriptions on stone vessels made in Egypt and carried to many other sites (e.g. DVsc [Westenholz and Stolper 2002], XVa–d).
The evolution of the multilingual Achaemenid inscription that is discernible in the texts on the monument of Darius I at Bisutun (DB, DBa‐l) began with the Elamite versions, as the spatial arrangement of the monument reveals (Schmitt 1991: pp. 18–19 with references). The first text added to the relief was a monolingual Elamite inscription giving Darius' name, ancestry, and claim to royal descent (DBa), placed in the field of the relief, centered above the image of Darius. The second text was a first edition of the fully developed apologia, including the narrative of Darius' rise to power and suppression of his opponents, and his admonitions and exhortations to future rulers and readers, presented in Elamite (DB El. §§1–54), positioned immediately to the right of the relief panel. The agents of Darius first chose Elamite to commemorate his reconquest and restoration of the core territories of the empire. The concept