Humboldt-Universität zu Berlin - CALLIDUS


Callidus is an interdisciplinary research project that delivers new methods of vocabulary acquisition in Latin. Its main purpose is to show the use of corpus-based methods in real teaching environments. Furthermore the project develops a software which is able to use linguistic research tools for creating corpus-based vocabulary exercises.

Vocabulary Acquisition in Latin using Corpus-based Methods

(Computer-Aided Language Learning: Lexikonerwerb im Lateinunterricht durch korpusgestützte Methoden)


In the CALLIDUS project three different departments are working together: the Computer- and Mediaservices, the Didactics of Latin and the Corpuslinguistics. Our common aim is to prove the usefulness of corpus-based methods for teaching Latin and to develop a digital tool as a prototype of corpus-based vocabulary acquisition.

Fox as Logo of the software Machina Callida

Link icon Machina Callida - Software


The main research questions

  1. Are corpus-based methods really more supportive in vocabulary acquisition than other methods used for teaching foreign languages?
  2. What implications do corpus-based methods have for the development of tasks?
  3. How can we systematically obtain lexical information about (problematic) texts, esp. in order to use it for language acquisition?
  4. How can we reuse available corpora and search tools for use in schools?
  5. What are the criteria of a user-friendly software in a teaching/learning environment?


The status after 2,5 years

On what we worked so far:

  • Linguistic terminology: What is a word? What are collexemes? What are collocations? ...
  • How to analyse a corpus (semi-)automatically? (POS-Tagging, Stemming, Relations, Test methods for collocations and semantic fields)
  • Language acquisition (in general and in Latin)
  • Task-based learning, design of corpus-based tasks
  • research design (DBR), intervention studies
  • AQL / software architecture
  • Embedding of open source educational software (H5P)
  • Research on text complexity and measuring it in Latin texts
  • Research on methods of semantic text analysis


Some of our milestones:

  • Pilot study in May/June 2018 with intermediate learners
  • Prototype of the software (called Machina Callida) (Aug. 2018)
  • Study: Vocabulary acquisition with intermediate learners (Oct. 2018 until Jan 2019)
  • Study: Vocabulary acquisition with beginners (Aug. 2018 until June 2019)
  • International workshop: Digital Approaches to Teaching Historical Languages (DAtTeL) (28.3.-29.3.2019)
  • Software: Machina Callida (fill in the gap and matching exercise, context example, vocabulary comparison, Latin corpus, text complexity, repository, curated vocabulary unit)
  • Study: Corpus-based task vs. rote learning task (ongoing)
  • Publications and talks
  • New layout including documentation of the various contents (Nov. 2019)
  • Software: semantic text analysis (Feb. 2020)
  • New proposal: Daidalos-Projekt (March 2020) [Abstract]
  • Software: technical documentation (April 2020)
  • Software: Exercises can be integrated into Moodle (June 2020)


Posters on concept and software:

poster callidus part1 conceptsposter callidus part2 software