TM Formulae was once intended to be an all-encompassing database of formulaic language. It started with an experiment in Named Entity Recognition to extract abbreviations from Latin inscriptions (TM Abbreviations). In a next stage, we thought to bring together work on formulaic language in different languages and various genres, starting with letters, petitions and other epistolary texts. The basis would be the PhD studies of Baetens on Ptolemaic petitions (Greek and Demotic - TOP SS5 - KU Leuven), and Nachtergaele on formulae in private papyrus letters (Greek - TOP SS6 - UGhent), to which would be added Depauw's database on Demotic letters (published as The Demotic Letter in the Gisela Zauzich Verlag, 2006 - KU Leuven).

As we gradually realized, the structures of these datasets are too closely connected with their scholarly analysis to be easily combined into a single database. What one scholar calls a petition is a business letter for another, and some will group specific phrases into a single formula while others will keep them separate. In view of this we have opted to make each of them available as a separate online database. Unlike most other Trismegistos databases, these ones are not kept up to date systematically. They are meant to be stable, which will also make them outdated gradually. Perhaps one day a new study will be necessary as a result, building on what is collected here.

The following datasets are currently available or planned:

TM Abbreviations (based on experiments with Named Entity Extraction from Latin epigraphic texts).

TM Petitions (based on TOP SS 5: G. Baetens, A Survey of Petitions and Related Documents from Ptolemaic Egypt, Leuven 2020).

TM Letters TOP SS6 (based TOP SS 6: D. Nachtergaele, The Formulaic Language of the Greek Private Papyrus Letters, Leuven 2023).

(planned) TM Letters DS 14 (based on Demotische Studien 14: M. Depauw, The Demotic Letter, Sommerhausen 2006).