The Linear Elamite (or Late Proto-Iranian) script is a writing system currently attested only in the southern half of the Iranian plateau, with 45 known inscriptions as of 2025. First discovered during French excavations in Susa before 1903, it is likely the most recent stage of a distinct Iranian script that originated in the late 4th millennium BCE (ca. 3300–3000 BCE) with the so-called Proto-Elamite (or Early Proto-Iranian) tablets. Based on recent dating, the Linear Elamite (or Late Proto-Iranian) stage likely emerged in the first half of the 3rd millennium BCE and disappeared after 1900 BCE (the most recent inscription currently known, LEI 37, is attributed to the great chancellor Pala-išan, ca. 1900-1880 BCE) due to two simultaneous but independent phenomena: the influence of Mesopotamian culture in southwestern Iran, which led to the adoption of the cuneiform writing system, and a significant urban collapse that brought an end to the scribal tradition in southeastern Iran.
Several scholars attempted to decipher the Linear Elamite/Late Proto-Iranian script during the 20th century, notably Ferdinand Bork, Carl Frank, Walther Hinz, and Piero Meriggi. However, due to the limited number of inscriptions, full decipherment remained out of reach. A breakthrough occurred in 2017–2018 following the publication of a "new" collection of inscriptions on silver vessels, leading to its near-complete decipherment in 2022
So far, among the 2,033 sign occurrences attested in the 45 known inscriptions (Linear Elamite inscriptions/LEI 1 to 45), 1,978 (97.3%) have been successfully deciphered. These correspond to 72 allographs and 69 distinct signs (as k and k2, p and p2, and ru and ru2 are merely graphic variants of the same sign; 72-3=69; allographs and homophonic signs are marked with subscripts). All the 69 deciphered signs are phonographic, with no logograms attested to date. They represent 67 different phonemic values due to three pairs of apparently homophonic signs (different signs with the same phonetic value; h and h2 = /h/; li and li3 = /li/; pi and pi2 = /pi/; 69-3=66) and the polyphony of the u2/w sign (noting both the vowel /u/ and the glide /w/; 66+1=67). The remaining 55 (2033-1978=55) sign occurrences (2.7%) include between 19 and 24 hapaxes, while the other 30 occurrences correspond to five as-yet undeciphered signs.
At the very least, our understanding of this writing system can be consequently considered highly advanced. The script appears to be a purely phonetic (alpha-syllabic) writing system, organized according to a phonetic grid that employs 5 vowel and 12 consonant phonemes, generating 60 syllabic CV values— a structure well-documented in the school exercise LEI 13.
The Linear Elamite/Late Proto-Iranian script has so far been used exclusively to record texts in the now-extinct and isolated Elamite/Hatamtite language. This language is attested in inscriptions from the 3rd millennium BCE until the last royal Achaemenid inscriptions, written shortly before the invasion of Alexander of Macedonia (ca. 330 BCE). Known in later periods as Khuzi, it was likely spoken in southwestern Iran until approximately 1000 CE, when its last speakers probably died out. The language was rediscovered in the 19th century CE, thanks to the decipherment of cuneiform script and the reading of trilingual Achaemenid royal inscriptions (Old Persian/Elamite/Akkadian). The status of the Elamite/Hatamtite language before the 3rd millennium BCE, as well as its origins and linguistic affiliations, remains entirely unknown.
Primarily understood through the trilingual Achaemenid inscriptions, our current knowledge of Elamite/Hatamtite is still quite limited, which sometimes results in approximate translations. As far as is known, it is a positional (Subject-Object-Verb) and agglutinative language, characterized by the use of class markers and three identified verbal modes.
The recent decipherment of the Linear Elamite/Late Proto-Iranian script presents an opportunity to analyze dozens of new texts from the earliest documented stage of this language (Old Elamite/Hatamtite, ca. 2300–1500 BCE). The edition of these texts is currently in preparation and will be published in three volumes:
F. Desset, K. Tabibzadeh, M. Kervran, G.-P. Basello, and G. Marchesi, with the collaboration of L. Colonna d’Istria :
2025
Forthcoming:
The Hatamti-Linear Elamite database was developed between February 2024 and May 2025 at the University of Liège by François Desset and Luc Desert, as part of the ARC project VariGraph, under the supervision of Laurent Colonna d’Istria and Stéphane Polis, with the collaboration of Emil Joubert. The Hatamti-Linear Elamite database can be considered a spin-off and adaptation of the Thot sign list, developed by Stéphane Polis since 2015/2016.
Initially conceived as a training tool for the creation of a database focusing on graphic variation in Proto-Elamite/Early Proto-Iranian script, the Hatamti-Linear Elamite database is expected to expand in the near future to include all (cuneiform) texts written or discovered in Iran between the late 3rd millennium BCE and ca. 1500 BCE, regardless of their language. This expansion aims to facilitate the study of the local adoption and adaptation of the cuneiform script in Iran. Additionally, this next phase of the project will provide an opportunity to further develop the Elamite/Hatamtite linguistic component already integrated into the Hatamti-Linear Elamite database.
The primary aim of the Hatamti-Linear Elamite database is to document the graphic variation of the Linear Elamite/Late Proto-Iranian script in the entire corpus currently known, primarily through geographic, chronological, and material perspectives. This graphic approach is mainly accessible through the phonetic grid, displaying all currently known signs (both deciphered and undeciphered). The grid allows users to visualize the geographic distribution and chronological placement on a timeline of their allographs (graphic variants), and their actual occurrences (tokens) in sources, the meaningful units (frequently propositions) composing each inscription (document).
| Graphic approach | Linguistic approach | |||
| Sign | Allograph | Token | Source | Document |
A linguistic approach is also available, allowing users to search for Elamite/Hatamtite words in transcription and examine their contexts within the documents. The transliteration system for the Linear Elamite/Late Proto-Iranian documents and the transcription of Elamite/Hatamtite words follow the conventions presented in Desset et al. (2022) . Whenever possible, translations based on analytical glosses are provided in English, French, and Persian. Research via translation in any of these three languages is also supported.
Finally, users can explore Linear Elamite/Late Proto-Iranian signs (tokens) or Elamite/Hatamtite words in transcription across each of the 45 currently known documents. For every document, the sources are presented in the following order: signs (the allographic variants), transliteration, transcription, and translation into English, French, or Persian. This section is thus designed to provide immediate access to the content of all known Linear Elamite/Late Proto-Iranian inscriptions. For historical or linguistic commentary and analysis, however, users are invited to consult the first two volumes of the text edition.
The Linear Elamite/Late Proto-Iranian signs and their allographs used in the Hatamti-Linear Elamite database were designed by Sina Fakour as part of the Missing scripts project hosted by the Atelier National de Recherche en Typographie (ANRT) in Nancy. Sina Fakour is currently working on the Unicode encoding of the Linear Elamite/Late Proto-Iranian script, after which his font will be made freely available.
The images of documents no. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 20, and 21 are courtesy of the Louvre Museum (Paris, France).
The images of documents no.17 and 19 are courtesy of the National Museum of Iran (Tehran, Iran).