Abstract
Inflectional languages such as Czech are very rich in morphological forms. For instance, a single noun may have up to 14 possible forms. However, not every paradigmatic cell must have an attested exponent; sometimes it appears there are none available for use. This phenomenon is termed defectiveness (Sims, 2007). Alternatively, for one cell we may have two or more forms that are able to fill the same cell in a paradigm. This phenomenon is called overabundance (Thornton, 2012). Examining adult learners' language from the point of view of its overabundance or, on the contrary, its defectiveness provides completely new insight into the acquisition of Czech as a foreign language. It opens up space for the definition of new standards in teaching and makes it possible to find out what kinds of deviations from the norm are present in the writing of native speakers of inflectionally impoverished languages like English. The crucial question is what a learner's 'paradigm' looks like.
The aim of this study is to find out how learners deal with the multitude of morphological forms in the language acquisition process. Three research questions are posed: 1) How much variability is there in the paradigmatic cells in non-native writing where overabundance is expected in native production? 2) How much variability is there in the paradigmatic cells in non-native writing where overabundance is not expected in native production? 3) If learners are unsure of a word form, do they tend to fill the cell anyway, or avoid filling it through circumlocution, substitution, etc.?
The research is based on the analysis of texts collected from 10 English students of Czech as a foreign language. Each student completed four different tasks including: a letter, a description of a place, a reflection (argumentation) on advantages and disadvantages of media, and a story about the world in the year 3000. The data were stored in TEITOK (Janssen, 2016), which enables manual annotation of the words under examination. There were four categories: 1) canonical overabundance, 2) marginal overabundance, 3) examples beyond the margins of overabundance, and 4) defectiveness. The first category consists of obvious examples where learners mostly: a) adopted endings from another conjugation/declension class (e.g. *azylantami instead of azylanty 'asylum seekers'), b) added endings characteristic of a different adjective declension (e.g., stabilní 'stable' vs. nový 'new'), c) declined items according to a different grammatical gender (jeli jsme tramvajem instead of jeli jsme tramvají 'we went by tram'). The latter categories were also subjected to analysis as "language phenomena tend to be scalar rather than categorical" (Janda & Tyers, 2021, p. 114). It revealed interesting tendencies in English learners' texts such as: additional stems resulting from consonant mutations, patterns characteristic of other foreign languages (Slovak and Russian) or vowel length alternations.
Preliminary results show that the most frequent of these three categories in learner data is variability in the paradigmatic cells where overabundance is not expected in native production. However further investigation on a larger corpus is needed to spot more evident examples of defectiveness.