Animacy and the notion of semantic gender

1. General The research reported in this paper is part of a project called " Animacy in typology and discourse ". 1 One of the aims of this project is to investigate the role of the notion of animacy in different parts of grammar, and gender is one of the most obvious places where animacy shows up. The aim of this paper, then, is twofold. One is to consider some generalizations about gender systems, based primarily on Corbett (1991), a recent comprehensive survey of gender and gender systems with data from over 200 languages. The other is to discuss, in the light of these typological generalizations, some of the theoretical notions needed to describe gender systems, in particular the concept of semantic gender. For an overview of animacy in general and the problems that interest us in the project, the reader is referred to Dahl & Fraurud (forthcoming). Here, I shall briefly mention some facts that are essential for the understanding of the rest of the paper. Animacy, or the distinction between animate and inanimate entities, is so pervasive in the grammars of human languages that it tends to be taken for granted and become invisible. It is only relatively recently that more systematic studies of animacy have been undertaken in linguistics. Many of these have concentrated on what has been called the animacy hierarchy, its role in various grammatical categories and its interaction with other notions such as definiteness, individuation and degree of referentiality. The animacy hierarchy is supposed to have (at least) the following components: HUMAN>ANIMAL>INANIMATE. Hierarchies of this kind have been assumed by typologists to underlie various implicational universals. The general idea is that grammatical phenomena will " obey " the hierarchy in the sense that certain generalizations will apply to all cases above a certain cutoff point in the hierarchy. For instance, in case marking systems, direct objects are more likely to be marked if they have a high degree of animacy. Below, I will be using the animacy hierarchy quite extensively in making generalizations about gender systems. In fact, grammatical gender is one of the areas where it is easiest to do so, since animacy does not interact with other parameters in the way it does in many other parts of grammar. However, in Dahl & Fraurud (forthcoming), we argue that what has been referred to in the literature as the animacy hierarchy is essentially …