INFLUENCE OF GRAMMATICAL GENDER ON DEDUCTIVE REASONING ABOUT SEX-SPECIFIC PROPERTIES OF ANIMALS

摘要: INFLUENCE OF GRAMMATICAL GENDER ON DEDUCTIVE REASONING ABOUT SEX-SPECIFIC PROPERTIES ANIMALS Mutsumi Imai (imai@sfc.keio.ac.jp) Faculty of Environment and Information Studies, Keio University at Shonan-Fujisawa, Japan Lennart Schalk (schalk@ifv.gess.ethz.ch) Institute for Behavioral Sciences, ETH Zurich, Switzerland Henrik Saalbach (saalbach@ifv.gess.ethz.ch) Hiroyuki Okada (h.okada@eng.tamagawa.ac.jp) Department Engineering, Tamagawa University, Abstract Grammatical gender is independent biological sex the majority animal names (e.g., a male giraffe grammatically treated as feminine). However, there apparent semantic motivation grammatical classes, especially in mapping human terms to classes. This research investigated whether this between grammar affects deductive inference German speakers. We identified two contexts which speakers unconsciously over-generalize grammar-semantics make inappropriate inferences about sex-specific properties. They tended erroneously accept deductions when premise target agreed. The sex-gender agreement affected even was explicitly indicated die [FEM] mannliche (male) Giraffe). Experiment 2 further suggested that these effects occur only gender- marking article accompanied noun. Implications results lintuistic relativity discussed. Keywords: Add your choice indexing or keywords; kindly use semi-colon; each term. Introduction Many languages world have system gender, where nouns are assigned one limited number classes (Corbett & Fraser, 2000). Unlike mark semantically English), with assign all regardless not referents sex. link assignment conceptual properties non-human has widely been said be arbitrary (Aikhenvald, 2000; Fox, 1990), relevant words. For example, German, word feminine elephant masculine, but it case giraffes female elephants male. Nonetheless, ( ) must applied refers noun pronoun sie used an anaphoric reference, referent biologically Here, interesting question what extent able separate from its gender. From perspective speaker language without appears confusing actually Of course, know does directly reflect completely immune influence they draw animal’s properties? It possible few cases correspondence may resulted overgeneralization during process acquisition. In salient such woman, lady, mother feminine, while man, boy, father, masculine (Natural Sex Principle, cf. Zubin Koepcke, 1986). Thus, falsely generalize exceptional class words animated entities general. assumption consistent Vigliocco colleagues’ (Vigliocco, Vinson, Paganelli, Dworzynski, 2005) sex-and-gender- hypothesis, proposes originates speakers’ first noticing corresponding other words, acknowledging some human- specific leads develop general anticipation animals same more similar another than different Most previous asked influences concepts typically feminine/masculine attributes those entities. Konishi (1993) looked how Spanish construe femininity masculinity non-animal objects by having them give gender-related ratings various on potency scale