Machina Callida

Machina Callida 

There is hardly any empirical research on lexicon acquisition in the LU (Petersmann 1989; Bösch 2012; Sass 2015), although the vocabulary is more didactic (Steinthal 1971; Untermann and Wülfing 1981; Kubik 1989; Thurow 1981; Waiblinger 2002); Nickel 1999; Utz 2000; Freund und Schröttel 2003; Daum 2016; Kuhlmann 2016) and methodical (Steinhilber 1978; Hermes 1988; Esser 1999; Stirnemann 2009; Doepner und Keip 2014; van de Loo 2016) perspective has been repeatedly addressed for many years. Consequently, the interdisciplinary, DFG-funded research project CALLIDUS is not only unique in its design (time, scope, interdisciplinarity), but above all extremely challenging, because findings from very different sub-areas have to be theory-driven and application-oriented at the same time in order to improve the acquisition of lexicons in Latin teaching. A central question of the Callidus project is whether the approach of data driven language learning (Braun 2007; Gilquin and Granger 2010) represented in second and foreign language research can be transferred to a historical language such as Latin. According to this theory, vocabulary acquisition is supported by the fact that tasks or exercises are generated from a corpus of authentic language utterances in order to enable the learner to learn immersively, i.e. completely immersed in the language, using the example (Farr and Murray 2016). These 'real' linguistic actions - be it literature or a communicative everyday situation - are intended to provide the learner with the necessary context to grasp vocabulary in its variability and complexity. Starting from a sufficiently large corpus that is highly likely to offer different formal and functional aspects of a word or phrase, the learner can be provided with linguistic exercises and sample applications that illustrate the multidimensionality of a word or phrase. This corpus-based vocabulary work thus aims at a reflected, intralingual linguistic action as well as a more complex vocabulary knowledge, as a result of which the vocabulary can be better networked in the so-called mental lexicon (Bruza et al. 2009; Kersten 2010). However, there is little empirical evidence to date on the usefulness of this approach, nor are there any theory-based methodological considerations on how to implement it in a teaching unit, i.e. how to construct the exercise and work instruction, in order to initiate the desired vocabulary learning. Accordingly, the CALLIDUS project is not only entering largely unknown territory from a late didactic perspective, but also from a corpus linguistic perspective.