Studies in Language. International Journal sponsored by the Foundation “Foundations of Language” - Volume 26, Issue 1, 2002

This article attempts a more complete explanation of the position of verbal agreement affixes which, unlike other inflections, lack a correlation with word order and are prefixes relatively more frequently. An analysis of the factors that determine the position of affixes during morphologization reveals that agreement inflections originate in relatively more mobile sources (unstressed pronouns) and are associated to polymorphemic hosts. This, in combination with a proposed psycholinguistic principle, which during morphologization, prevents fusion of grammatical markers to complex sequences of inflections (for which crosslinguistic statistical evidence is supplied) can be used to explain the position of agreement markers.

While the structure consisting of a topic and a clausal comment, often referred to as the double-topic, double-subject, or double-nominative sentence, has been discussed in the literature, the internal constituent order within the comment has rarely been a target of study of linguistic typology. The purpose of the paper is to compare two constituent order possibilities found in Thai and other Southeast Asian languages to argue specific constructional meanings associated with them. In Thai, the topic with a clausal comment is used when a body-part is involved, and it may take the form of either [ N1 [ N2 V ]] or [ N1 [ V N2 ]]. In this structure, the topic (N1) is a possessor of N2 (= body-part), and the comment is a clause consisting of an intransitive verb (V) and a body-part noun (N2). These two constituent orders are distributed between the external-state and proprioceptive (internal) state. Thus physical sensations (e.g. “I have a headache”) and emotions (e.g. “I am sad”) are expressed with the [VN] structure in the comment, while the description of external-states such as physical appearances (e.g. “My head is big”) is expressed with [NV]. Since proprioceptive states can only be experienced directly by the experiencer himself and contrast with external states which can be observed by anyone, the constituent order reflects fundamentally different cognitive realities experienced by the speaker. I claim that the [VN] order has a structural meaning of “proprioceptively registered experience” in the Thai grammar. This explains further why some other internal events such as “recovering from cold” or “exhausting one’s physical strength” which do not directly involve body parts also take the [VN] order in the comment. I will also argue that the initial noun (N1) in the proprioceptive-state expression is a topic noun added to the more fundamental topicless structure, [VN], whereas the initial noun in the external-state description is an integral part of the basic argument structure. Finally, I will situate the distributional pattern of [VN] and [NV] found in Thai among other East and Southeast Asian languages.

This paper examines island effects in Japanese in four extraction types — relativization, topicalization, postposing, and wh-question formation — in terms of the “aboutness” condition claimed by Haig (1996). The condition is supported for relativization and topicalization; however, further specification is necessary for postposing and wh-question formation. It is also demonstrated that the proposed construction-specific conditions are all related to the same cognitive theory of island phenomena (Deane 1991), which reflects the mechanism of attention and short-term memory. The cognitive description thus provides insights as to why there are island effects in a putative “island-free” language.

Based on a corpus of conversational English, I argue that the standard view of complements as subordinate clauses in a grammatical relation with a complement-taking predicate is not supported by the data. Rather, what has been described under the heading of complementation can be understood in terms of epistemic/evidential/evaluative formulaic fragments expressing speaker stance toward the content of a clause. This analysis, in which CTPs and their subjects are stored and retrieved as formulaic stance markers accounts for the grammatical, pragmatic, prosodic, and phonological data more satisfactorily than a complementation analysis.

Kita (1999) compares Japanese and English Enter/Exit verbs in spatial expressions, and argues that Japanese Enter/Exit verbs lack semantic encoding of motion. He claims that this runs counter to the view which considers motion and location to be primitives in the semantics of spatial expressions; instead, he proposes that discrete change of state should be included in the set of primitives. In this reply,I will first show that Kita’s evidence does not support lack of motion in Japanese Enter/Exit verbs, but that instead these verbs do pattern with motion verbs in the language, where conflation of motion is not disputable. I finally demonstrate that Kita’s claim about change of state may be well taken, but it should be put in a larger context of regular polysemy.