A new international linguistics project funded by the Arts and Humanities Research Council (AHRC) is set to help language learners, teachers and therapists tackle one of the most challenging aspects of learning Czech and Croatian — the variety of word forms used in everyday speech and writing.
The Inflection Points project brings together researchers from the UK, Czechia and Croatia, led by Professor Neil Bermel from the University of Sheffield’s School of Languages, Arts and Societies. The team includes Dr Martin Beneš (Institute for the Czech Language, Czech Academy of Sciences), Dr Tomislava Bošnjak Botica (Institute for the Croatian Language), Dr Gordana Hržica (Faculty of Education and Rehabilitation Sciences, University of Zagreb), and Dr Dominika Kováříková (Institute of the Czech National Corpus, Charles University).
The project builds on work from a previous AHRC-funded study, Feast and Famine (AH/T002859/1), which ran from 2020 to 2024. That research looked at how speakers learn to understand the connections between the many different forms a word can take.
“Two of the languages we examined in the project, Czech and Croatian, have complex inflectional morphology — that’s the different forms a word can take in performing different functions: e.g. speak, speaks, spoke, spoken,” said Bermel. “English has only two to four such forms per word, but in these languages, nouns have 10-12 different forms each, and verbs more than twice as many.”
Speakers of these languages must learn to connect all these possible forms to a single “headword”, creating opportunities for errors and confusion. Mastering these patterns is a key part of becoming fluent, and both language teaching and assessment in these countries focus heavily on correct usage.
Now, the Inflection Points team is working to adapt the digital tools they originally built for academic research into practical resources for teachers, learners, and speech therapists. They are collaborating with education and healthcare professionals to tailor three custom-built apps for real-world use:
- DvojBa, a morphological database of Croatian doublet forms, will be updated to help teachers working with heritage and second-language learners — particularly children returning to Croatian schools after spending time abroad.
- MultiDis, an app for measuring child language development, will be expanded to focus more closely on inflectional forms, supporting speech therapists and other language specialists.
- Teachers of Czech as a second language will help refine GramatiKat, a database of Czech word forms, alongside new educational materials. This will be especially valuable for supporting the 50,000 Ukrainian children now enrolled in Czech schools, whose native language is closely related to Czech.
The project will also carry out user research into the Internet Language Handbook, a popular online resource consulted by millions of Czech speakers each year, to make its guidance even clearer and more consistent.
“We originally created these apps for scholarly use,” Bermel explained. “But we’re excited about the possibilities for transforming them for public consumption. There is a profound equality dimension to this project, as your ability to produce the expected form affects your educational and employment prospects.”