Machines Reading and Deciphering Maya Hieroglyphs: Towards a Digital Epigraphy of Maya Hieroglyphic Writing

Machines Reading and Deciphering Maya Hieroglyphs: Towards a Digital Epigraphy of Maya Hieroglyphic Writing
16 April 2021, 5.00pm - 6.15pm

Christian Prager, University of Bonn

Cristina Vertan, University of Hamburg

Live online at:

All welcome

Maya script is the only readable writing system of the Americas. It consists of 1000 morphograms and phonograms and was used from 300 B.C. to 1500. It is only partially deciphered and the subject of the project “Text Database and Dictionary of Classic Mayan.” To this end, the 10,000 texts are made accessible using digital methods and technologies. The inscriptions are documented in XML/TEI, for which we created a digital sign catalogue. Due to the uncertain state of decipherment it is not possible to directly record hieroglyphic texts in phonemically transliterated values, contrary to epigraphic projects in the Classics. Thus we transliterate the texts numerically using sign codes from the catalog. The workflow for transforming the numerical transliteration into a textual form implies 10 steps, at each step variants are possible. This complex process is scalable only by means of digital tools. As none of the current available tools match our requirements we developed ALMAH: “Annotator for the Linguistic Analysis of Maya Hieroglyphs.” It allows the semi-automatic generation of the phonemic transliteration from the numerical one and enables multi-level annotation and linguistic analysis of the Maya texts. The transcriptions are accepted, rejected or revised in the further steps. Alternative readings can be entered, two or more decipherment proposals can be worked on in parallel, and thus the hieroglyphic texts can be analysed and translated with alternative proposals. ALMAH is based on a graph-data model. It reads the data from the TEI-Repository and saves the annotated texts into an instance of OrientDB. Written in Java, it has a friendly user-interface and will be available as open-source to the research community. We would like to present the data model and the functionality as well as the underlying TEI scheme allowing to encode Maya hieroglyphic texts formally and semantically.


Valerie James
020 7862 8716