Compositionality of Complex Graphemes in the Undeciphered Proto-Elamite Script using Image and Text Embedding Models

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 4136–4146 August 1–6, 2021. ©2021 Association for Computational Linguistics 4136 Compositionality of Complex Graphemes in the Undeciphered Proto-Elamite Script using Image and Text Embedding Models Logan Born1 loborn@sfu.ca Kathryn Kelley2 kathryn.kelley@utoronto.ca M. Willis Monroe3 willis.monroe@ubc.ca Anoop Sarkar1 anoop@cs.sfu.ca 1Simon Fraser University School of Computing Science 2University of Toronto Department of Near and Middle Eastern Civilizations 3University of British Columbia Department of Asian Studies Abstract We introduce a language modeling architec- ture which operates over sequences of im- ages, or over multimodal sequences of images with associated labels. We use this architec- ture alongside other embedding models to in- vestigate a category of signs called complex graphemes (CGs) in the undeciphered proto- Elamite script. We argue that CGs have mean- ings which are at least partly compositional, and we discover novel rules governing the con- struction of CGs. We find that a language model over sign images produces more inter- pretable results than a model over text or over sign images and text, which suggests that the names given to signs may be obscuring signals in the corpus. Our results reveal previously un- known regularities in proto-Elamite sign use that can inform future decipherment efforts, and our image-aware language model provides a novel way to abstract away from biases intro- duced by human annotators. 1 Introduction This work sets out to understand a category of signs called complex graphemes (CGs) in the un- deciphered proto-Elamite (PE) script, a writing system from ancient Iran dating to approximately 3300-2900 BC (Dahl et al., 2013).1 PE is partly contemporaneous with the world’s other two ear- liest writing systems, Egyptian hieroglyphs and proto-cuneiform, and is the least deciphered of the three, with the underlying language(s) remaining unknown. PE was used exclusively as an account- ing technology, employing several numerical sys- tems whose bundling principles are known. Al- though written in continuous lines, PE, like proto- cuneiform, is most comparable to an accountant’s spreadsheet; some structures and rules governing 1Our code, data, and trained models are avail- able at https://github.com/sfu-natlang/ pe-compositionality sign use have been identified (Hawkins, 2015; Dahl et al., 2018; Englund, 2004). The corpus consists of approximately 1500 pub- lished clay tablets from excavations in Iran, almost all of which exist in electronic transliteration fol- lowing the conventions of a work-in-progress sign list (Dahl, 2006). As with other decipherments, understanding the nature of signs and the nuances of sign use is as important as identifying the un- derlying language(s). Meaningful information can be recovered and the texts partly “read” even if the language remains unknown. To better understand sign usage in PE, this work proposes an architecture for image-aware language modeling, which permits sharing information be- tween visually similar signs much as sub-word units share information between words. We use sign embeddings to demonstrate patterns which are not readily apparent due to the complexity of the accounting system and the large number of sign shapes found in the script. Our analysis of- fers insights …

Content truncated to 500 words