In lieu of an abstract, here is a brief excerpt of the content:

Determinants of Split Intransitivity in Blackfoot: Evidence from Verbs of Emission Sara JohanSSon and ElizabEth rittEr University of Calgary dEtErminantS of Split intranSitivity According to Perlmutter’s (1978) Unaccusativity Hypothesis, intransitive verbs can be divided into two classes: unaccuSativE verbs, which have an underlying internal argument, such as a theme or patient, and unErgativE verbs, which have an underlying external argument, such as an agent. Van Valin (1990) identifies two types of languages based on the semantic factors which determine whether a particular intransitive verb is unergative or unaccusative.1 In one type of language, classification of intransitive verbs is based exclusively on argument selection. More specifically, all verbs whose single argument is a volitional agent are unergative. We will refer to these as Type I languages. In the other type of language lexical aspect also plays a role. In this type of language, telic verbs are unaccusative, and unergative verbs are agentive and/or atelic, and thus constitute a more heterogeneous set.2 We will refer to these as Type II languages. See Table 1. 171 1. In fact, Perlmutter (1978) proposed that unaccusativity is semantically determined, but syntactically represented. On his analysis, a direct internal argument is an underlying direct object and an external argument is a subject at all levels of representation. However, the question of whether the single argument of an unaccusative verb is an underlying direct object has been the subject of significant debate. (See Levin and Rappaport Hovav 1995 for an overview.) Van Valin (1990) argues that the distinction between unergative and unaccusative verbs can be characterized entirely in semantic terms, without recourse to syntactic movement. In this paper, we eschew the question of syntactic derivation, and focus on the semantic determinants of unaccusativity in Blackfoot. 2. A tElic verb has an incremental theme, i.e., a direct internal argument that measures out the event. For example, the table is an incremental theme in John painted the table. Progression of the painting event in this example correlates with changes in the table, e.g., when half the table is painted, half the event has been completed. See Dowty 1991 and Tenny 1994 for discussion. 172 Sara JohanSSon and ElizabEth rittEr The goal of this paper is to identify the semantic determinants of split intransitivity in Blackfoot. In order to address this question, we develop language-specific tests for unergativity/unaccusativity and apply them to verbs that are both non-agentive and atelic. Such verbs are unaccusative in Type I languages but unergative in Type II. We show that these verbs are unergative in Blackfoot, and thus that both agentivity and (a)telicity are determinants of split intransitivity in this language.3 animacy, agEntivity, and blackfoot vErb claSSification Following Bloomfield (1946), it is standardly assumed in the literature on Algonquian languages that verb stems belong to one of four classes, depending on the number of arguments they require and the gender of one of those arguments. As summarized in Table 2, there are two classes of intransitive verb stems, distinguished by the grammatical gender (animate or inanimate) of the subject, and two classes of transitive verb stems, distinguished by the grammatical gender (animate or inanimate) of the object.4 Verb class is Table 1. A Typology of Split Intransitivity dEtErminantS of typE Split intranSitivity unErgativE unaccuSativE Type I Argument-based split agentive non-agentive Type II Argument & Aspect-based split agentive non-agentive and/or atelic and/or telic 3. We thank Rachel Ermineskin and Noreen Breaker for teaching us about their language. Both are native speakers of Siksika Blackfoot. This research was funded in part by a PURE grant to Sara Johansson and SSHRC grant 410-2008-0892 to Elizabeth Ritter. Additional funding was provided by the Faculty of Social Sciences at the University of Calgary. 4. We use the terms SubJEct and obJEct for ease of exposition. The term object refers to the internal argument of a transitive verb, and corresponds to the traditional Algonquianist term, goal. The object is a proto-patient in the sense of Dowty (1991), or an undergoer in the sense of Van Valen (1990). The term subject refers here to the single argument of an intransitive verb or the external argument of a transitive verb, and corresponds to the traditional Algonquianist term, actor. The subject of a transitive verb is a proto-agent in the sense of Dowty, and an actor in the sense of Van Valen. The subject of an intransitive verb is either a proto-agent...

Share