Grammatical gender has recently received a lot of attention within psycholinguistic research (Schriefers & Jescheniak, 1999). In a well-known theory of speech production (Levelt, Roelofs & Meyer, 1999), gender is represented as a lexical-syntactic property: the lemma node is connected to an abstract gender node. This implies that gender is lexicalized (as opposed to computed during production) and that gender assignment follows a uniform mechanism.
However, analyses of the German vocabulary by Köpcke and Zubin (1996) have shown that the gender distribution in German follows a variety of phonological, morphological and semantic regularities (i.e. rules that apply in most, but not all cases). These regularities suggest an inhomogeneous basis for gender assignment with both lexicalized and computed gender assignments, and a diversity of gender assignment mechanisms.
Before the influence of gender assignment regularities on speech production can be evaluated, it has to be established that native speakers of German know these regularities. This has been shown for phonological regularities (e.g. Köpcke, 1983). The aim of this study is to show that native speakers of German know semantic gender assignment regularities, in particular direct associations between semantic categories and genders.
The association between semantic categories and genders is experimentally tested in a forced choice category assignment task. On each trial, a semantic category and a pair of pseudowords with different gender markings (determiners) were presented and the participants were required to select the pseudoword that most likely was a category member. We predict that for gender-associated categories the pseudoword marked with the associated gender will be preferred, whereas for control categories, no gender preference will occur.
Six gender-associated categories (e.g. feminine: fruit; masculine: predators) and three non-associated control categories (e.g.: tools) were selected based on an analysis of semantic caregory - gender associations by Zubin and Köpcke (1986). To control the influence of pseudoword wordform on category membership selections, pseudowords were matched for wordform and category membership ratings (obtained in a pretest), and gender marking for each pseudoword was balanced between subjects. In an experiment with 24 native speakers of German we found that for gender-associated categories the pseudo-word marked for the associated gender was selected significantly more often, while for non-associated categories no preference was obtained.
In conclusion, the study seems to indicate that native speakers of German know semantic regularities in gender assignment. It remains an open question if these regularities are used in online language production.
References
Köpcke, K.-M. & Zubin, D. (1983). Die kognitive Organisation der Genuszuweisung zu den einsilbigen Nomen der deutschen Gegenwartssprache. Zeitschrift für germanistische Linguistik, 11, 166-182
Köpcke, K.-M. & Zubin, D. (1996). Prinzipien fuer die Genuszuweisung im Deutschen. In E. Lang, G. Zifonun (Ed.) Deutsch-typologisch. Institut für deutsche Sprache Jahrbuch 1995 (pp 473-491). Berlin: Walter de Gruyter
Levelt, W. J. M, Roelofs, A. & Meyer, A. S. (1999). A theory of lexical access in speech production. Behavioral and Brain Sciences, 22, 1-75
Schriefers, H., & Jescheniak, J. D (1999). Representation and processing of grammatical gender in language production: a review. Journal of Psycholinguistic Research, 28, 575-597
Zubin, D. & Köpcke, K.-M. (1986). Gender and folk taxonomy: The indexical relation between grammatical and lexical categorization. In Craig, C. (Ed). Noun Classes and Categorization. Proceedings of a symposion on categorization and noun classification, Eugene, Oregon, October 1983. Amsterdam: John Benjamins Publishing Company 1986