Abstrakt
A lot of research within the usage-based framework has focused on near synonymous constructions above the level of single words, whereas inflectional morphology and constructional homonymy are not as widely researched phenomena, although a usage-based, constructivist approach is readily applicable to word level phenomena (cf. e.g. Booij 2013).
In this paper, I present preliminary results of a study concerned with the interpretation and use of homonymous case forms in Czech. Czech is a morphologically complex language, in which case homonymy is ubiquitous.
While this phenomenon has been widely discussed in theoretical linguistics (e.g. Baerman 2008), I focus on the role of frequency in the interpretation and use of such forms to see whether speakers keep track of frequency asymmetries in the context of two homonymous, word-level constructions.
A sentence elicitation task was created to address this question. In the web-based task, participants (n = 46) were instructed to use a given word form in a sentence.
The word forms were instrumental singular / genitive plural forms of the "soft" feminine paradigms ending in -í, e.g. lahv-í 'bottle-ins.sg/gen.pl' (cf. Cvrček et al. 2010).
Both the instrumental and the genitive are distinct from direct case forms and both may occur with or without a preposition. Word forms that appear in more than 60 % or less than 40 % cases in genitive in a corpus of written Czech (Křen et al. 2015) were selected.
Nine test items matched for lemma and word form frequency were selected from each group. Responses were coded for interpretation (genitive or instrumental) and the percentages of genitive uses were calculated for each item.
Additionally, to test this in a more controlled environment, the same items were used in a maze task in which different participants (n = 46) created genitive- or instrumental-governed prepositional phrases. All items appear in both context, suggesting that both interpretations are indeed possible, acceptable, and accessible.
The elicited sentences mirror almost perfectly the corpus frequencies, and although the results of the maze task are less clear, a similar trend can still be observed. I will discuss the patterns using a follow up corpus analysis, arguing that speakers do to some extent keep track of the frequencies of individual grammatical senses of syncretic forms and of lexically specific higher order constructions containing these word forms, providing additional support for usage-based models of language (e.g.
Taylor 2010).