Historical-Comparative Linguistics and the Decipherment of Ancient Scripts
Any attempt at decipherment of an unreadable ancient script starts with an exhaustive analysis of the corpus of inscriptions/texts, which includes a determination of the type of writing system and the direction of writing, the preparation of a sign inventory or rather catalog of characters and a search for recurring character sequences (cf. e.g. Daniels 1996, Ventris 1953). However, at some point in time the decipherer must proceed to an experimental substitution of concrete phonetic values to particular graphemes, and the sound values assigned to individual signs must result in grammatically correct word forms in a natural human language. At best, this language is already known, and the prospective decipherer has an extensive knowledge of it. But what about cases in which the underlying language has died out in the meantime?
At worst, there are no relatives, and progress in deciphering an ancient script used to write down such an isolated language would exclusively depend on bilingual texts involving familiar languages. A different situation, however, arises in the case of genetically related languages. Even though there may be no direct descendants of the language underlying the unreadable script, the study and comparison of its relatives can eventually result in a successful decipherment. This possibility arises through application of the Historical-Comparative Method (HCM). Historical-comparative linguists routinely use the HCM to reconstruct concrete lexical material of unattested (proto-)languages – i.e., roots, stems, inflected/finite words, whole phrases – and linguistic reconstruction may also help in determining possible word forms of an as yet unknown member of an otherwise well-known language family. If relatives are known, a linguist using the HCM can reconstruct lexical items of the last common ancestor of the relevant languages as well as their morphosyntactic characteristics including the order of syntactic constituents. Reconstructed noun or verb phrases may then be experimentally substituted to particular character sequences previously identified in the course of the distributional analysis.
In my talk, I will demonstrate this procedure with the recent decipherment of the Issyk-Kushan script (Bonmann et al. 2023). In this particular case the noun phrase ‘king of kings’ provided the key to the decipherment. This phrase has a repeated stem, namely ‘king’, whose length and phonological constitution is characteristic for different linguistic phyla. Only a Middle Iranian noun phrase matched a repeated character sequence in two Issyk-Kushan inscriptions hypothesized to write down ‘king of kings’ – remarkably, however, a phrase or rather sequence of phonemes deduced from linguistic reconstruction. Without an application of the HCM the decipherment of the Issyk-Kushan script would have been impossible. Future successful decipherments of other ancient scripts may therefore likewise depend on an application of the HCM by trained historical-comparative linguists.
REFERENCES
Bonmann, Svenja, Jakob Halfmann, Natalie Korobzow & Bobomullo Bobomulloev. 2023. A Partial decipherment of the unknown Kushan script, Transactions of the Philological Society 121 (2), 293–329. DOI: https://doi.org/10.1111/1467-968X.12269.
Daniels, Peter. 1996. Methods of decipherment, in Peter Daniels and William Bright (eds.), The World’s Writing Systems, Oxford: Oxford University Press, 141–159.
Ventris, Michael. 1953. A note on decipherment methods, Antiquity 27 (108), 200–206.
