Studies in Language. International Journal sponsored by the Foundation “Foundations of Language” - Volume 26, Issue 2, 2002

In the face of evidence of considerable word order variation in Mixe languages, this article posits a basic word order of SOV for Isthmus Mixe, analyzing the language as having an inverse voice category that partially explains the observed surface word order variation. Using functional criteria established for voice distinctions by Givón (1994), it is shown that the Object is higher in topicality than the Subject in inverse-transitive clauses, as attested by ellipsis of the Subject and by topicality measures of Referential Distance and Topic Persistence. When inverse-clause word orders are separated out from direct-clause word orders, and when discourse considerations are taken into account, the basic SOV order of the Isthmus Mixe direct clause becomes apparent.

Several researchers have reported the increased use of o and semantic/pragmatic or syntactic motivations for direct object marking based on texts from between the 8th and 11th centuries. However, very few studies documented direct object marking after that period on until Modern Japanese, and the claims regarding semantic/pragmatic or syntactic motivations were not substantiated. By quantitatively examining an 11th century text and two later translations of it, this study documents evidence that o-marking was pragmatically motivated in the 11th century. However, as o-marking gradually increased over the years to be fully grammaticized as a direct object marker by the 20th century, its correlation with the pragmatic factors weakened.

Semantic extensions of spatial deictics and of verbs meaning ‘come’ and ‘go’ have aroused interest in the context of studies of grammaticalisation. Reflexes of the Proto Polynesian directional particles *mai ‘towards speaker’/venitive, *atu ‘away from speaker’/andative, *hake ‘upwards’ and *hifo ‘downwards’ show the development of psychological, expressive and temporal meanings from concrete (in this case spatial) meanings which has been seen as central to many cases of semantic change. This discussion of Tokelauan mai and atu is based on a quantitative analysis of a corpus of spoken Tokelauan. In most types of discourse, the expression of spatial deixis is by no means the most frequent use of the particles. Uses relating to communication, perception, emotion, point of view and speaker’s attitude predominate. Most interestingly, there is also evidence for the development of aspectual functions. I show that the venitive particle mai can function as a marker of continuative aspect and that this development arises from its temporal and evidential uses. The andative particle atu can indicate ingressive aspect, particularly in certain idiomatic collocations and in ‘begin’ constructions. These meanings lack spatial content, but the deictic element is always detectable, though at a more abstract level.

Researchers investigating the Spanish word mismo and its Catalan congener mateix have traditionally focused exclusively on phonological issues, ignoring a number of worthy semantic topics. In this paper I explore the polysemy patterns of these forms in a cross-linguistic perspective. I argue that such examination of polysemy patterns in different languages will help us understand the nature of polysemy. I also claim that the evidence points to a three-way division of the lexicon into lexical, semi-lexical, and grammatical subparts.

The languages of the Yi group shed new light on the Perfect forms of other languages through their perfect particles. Several Yi languages display two perfect particles which unite in one paradigmatic picture the English type and the Mandarin type. I will first present Liangshan Nuosu’s ta33 (stative perfect) and o44 (dynamic perfect) with Klein’s aspect-tense terminology (1992) and will then establish a formal definition of the notion of stative and dynamic perfect. It appears that both perfects have in common their relating the utterance situation to the discourse topic (in Klein’s terminology: TU þ TT). I will claim that the property ‘TU þ TT’ is the essence of perfect and can be viewed as an aspect-tense characterization of ‘current relevance’, a vague term that is employed by an array of authors without precise definition.

This paper elucidates one interacting aspect of two problematic issues in Japanese grammar: the functional differences between the nominative marker GA and the topic marker WA, and the referential identity of ellipted subjects. My analysis shows that the interaction of WA and GA in complex sentences has an analogous property to switch-reference systems which determine the identity of ellipted subjects; GA in subordinate clauses signals Different Subject, and WA in all loci Same Subject. This syntactic finding is further evaluated in the light of non-syntactic features, some of which can override the interpretation created by these particles. Statistically in examining texts, however, over 90% of ellipted subjects in complex sentences are successfully identified solely by WA and GA.

This paper deals with morphosyntactic splits in the coding of actancy in transitive and intransitive sentences. In transitive sentences the splits are motivated by particular actant/person constellations and their interplay with modality, especially with negation. In intransitive sentences the splits are motivated by modality alone. All the different actancy coding procedures and intricate split patterns are explained here in terms of semantic motion, that is, the local increase or decrease of agency/salience properties of the actants involved. Thus motion manifests itself as promotion or demotion; this leads to a dynamization of the language-specific nominal hierarchy. Semantic motion is not bound to a particular type of alignment since it occurs with accusative as well as ergative coding behavior. The grammar of Yimas illustrates a broad variety of motional procedures rich enough to discuss the typologically useful notion of motion on the basis of one single language.