Problems in Defining a Prototypical Transitive Sentence Typologically

PROBLEMS IN DEFINING A PROTOTYPICAL TRANSITIVE SENTENCE TYPOLOGICALLY Seppo Kittilei Department of General Linguistics Horttokuja 2 20014 University of Turku FINLAND Email: sepkit@utu.fi ABSTRACT This paper is trying to show that the concept of the prototypical transitive sentence is very useful in the study of transitivity, but is as such not so unproblematic as has often been assumed. This is achieved by presenting examples from different languages that show the weaknesses of the "traditional" prototype. The prototype proposed here remains, however, a tentative one, because there are so many properties that cannot be described typologically because of their language-specific nature. 1. INTRODUCTION Transitivity has been defined in many different ways during the history of linguistics. Traditionally transitivity was understood to be a semantic phenomenon: Transitive sentences were sentences which describe events that involve a transfer of energy from subject to object (e.g. 'he killed the man'). Structurally these are mostly sentences with a grammatical subject and a direct (accusative) object (at least in the languages that were the targets of description). In the golden era of formal grammars transitivity became a purely formal concept: Verbs (or more rarely sentences) with a direct object were classified as transitive and those without one as intransitive without taking their semantics into account (see e.g. Jacobsen 1985:89 [4]). Formal definitions very rarely allow any intermediate forms, i.e. every transitive sentence is equally prototypical (because there are no adequate criteria by means of which the prototypicality could be measured). Therefore, the concept of a prototypical transitive sentence is irrelevant in the formal descriptions. Lately, because of the growing interest in "exotic" languages this purely structural definition (that was mainly based on English or other Indo-European languages) has become insufficient. It is, for example, not capable of describing transitivity in languages in which subject and object cannot be defined so easily as in the Indo-European languages. As a result of this, a new kind of "typological" definition (which is more or less directly based on the traditional semantic definitions) has been developed. This definition is a kind of combination of the two previously presented ones: It defines the prototypical transitive sentence as one which describes a semantically highly transitive event. This definition has the advantage that it does not restrict the structure of the prototypical transitive sentence to some specific structure, which is the case in the formal definitions. It is, as such, however, not so suitable for describing transitivity as one might assume, but it has to be made more accurate. In this paper I present some issues (some of which are very basic in nature (and may even seem naive), but should in any case be explicitly mentioned in a valid definition) that one has to take into account when defining a prototypical transitive sentence. The goal of this paper is to propose a new (specified) prototype that could possibly be more suitable for typological studies. I hope to be able to show that the presented prototype is not so unproblematic as has often …

Content truncated to 500 words