Volume 53, Issue 4 -

Unlike the traditional approaches that describe and define antonymy on the basis of semantic properties of antonymous pairs in a language system, antonyms are in this paper considered with reference to their roles in actual language use. Antonymy has traditionally been regarded as a paradigmatic relation, but recent studies of antonym co-occurrence in electronic corpora of English and Serbian have identified two major and four minor functions of antonyms in discourse, with its accompanying lexico-syntactic patterns, matching the results of similar analyses in Japanese, Swedish, and Dutch. This paper aims to investigate the performance of three antonymous patterns in retrieving contrastive terms and the possibility to create the contrastive profile of given seed words in Serbian written discourse by means of those patterns. It also examines whether there is a relation between the markedness distinction between members of the antonymous pair and the position of the chosen seed word (i.e., whether it comes in the X or the Y position in the pattern). Results of the study show that it is possible to create the textual profile of antonyms using these lexico-syntactic constructions, and that the unmarked member has a richer contrastive profile in text.

The diachrony of valency patterns is generally an understudied phenomenon. The present article investigates anticausativization from a diachronic perspective, highlighting the parameters determining the morphosyntactic encoding of this type of intransitivization in two early Western Indo-European languages, Latin and Old Norse-Icelandic. It is shown that the structural and lexical aspects of a verb’s meaning and their interplay with the inherent and relational characteristics of verbal arguments affect the synchronic distribution and the diachronic development of the anticausativation strategies in the languages investigated. These features interact, in the course of time, with changes in the encoding of voice and grammatical relations, such as the demise of the synthetic mediopassive and the recasting of the case system.

The antipassive, an object-demoting diathesis commonly associated with ergative languages, has so far largely gone unnoticed in Bantu languages, which are of the accusative type. In this article, comparative evidence is raised to demonstrate that the antipassive is a voice construction to be reckoned with in Bantu. A robust typology of Bantu voice constructions is developed on the basis of the scarce data available in the literature. This evidence is reinterpreted in the light of original data from a number of Bantu languages, such as Cilubà and Kirundi, which were the subject of a more in-depth analysis. It is shown how the antipassive generally developed as a specific reading of the highly polysemous verbal suffix -an -, which is more commonly used as a reciprocal/associative marker. These and other functions can be accounted for by the underlying notion of “plurality of relations”, which is characterized by a low degree of participant/event elaboration. From a syntactic point of view, it is argued that the development of antipassives out of plurality constructions results from the gradual demotion of the second participant of a co-participative event.

It is a common perception that in languages having multiple past tenses with different remoteness specifications, the past tenses cover the entire past without a gap or overlap. This paper demonstrates that this way of looking at multiple-past tense systems is not appropriate for the system in South Baffin Inuktitut (a variety of the Inuit language). The dialect has at least four past tenses: recent, hodiernal, pre-hodiernal, and distant. We argue that the relation between the four tenses cannot be represented by a simple linear scheme for two reasons. First, the pre-hodiernal past has a special status as the “conventionally designated alternative”, which is chosen in cases of remoteness indeterminacy, analogous to, for example, the Russian masculine gender being used in cases of gender indeterminacy. Second, there is overlap in their coverage. The pre-hodiernal and hodiernal past tenses collectively cover the entire past and thus any past situation can be described with one of them. The other two provide means to make more fine-grained and subjective temporal specifications. Comparison will be made between the system in South Baffin Inuktitut and those in some Bantoid languages which have been pointed out in the literature to have a comparable layered system of tenses.

This paper addresses two questions. First, how can we know if a language is topic-prominent (Tp) in the spirit of Li and Thompson (1976, 1981)? Second, is Hong Kong Sign Language (HKSL) topic-prominent? This paper argues that some of the subsequent studies that adopt Li and Thompson’s proposal may have over-interpreted the latter’s initial claims by turning their observed characteristics of topic-prominent languages into “defining criteria” for topic-prominence. With linguistic evidence from Chinese, Lahu, and Lisu, this paper argues that topic-prominence first conceptualized by Li and Thompson is a broad notion that covers different linguistic phenomena that make topics prominent in Tp languages. Suggestions on how to interpret Li and Thompson’s list of topic-prominence characteristics are given. É. Kiss’s theory of discourse configurationality (1995, 1997, 2001) is also used to investigate if HKSL is topic-prominent. It is argued that only “scene-setting” topics, but not “aboutness” topics, are marked non-manually in HKSL. HKSL cannot be considered topic-prominent on a par with Chinese and Lisu, because topicalization in HKSL is infrequent and the non-manual topic markers are not widespread. HKSL is not topic-prominent according to É. Kiss’s theory either because thetic and categorical judgments are not realized in different syntactic structures.

This paper analyzes TAM marking in the nominal domain in Chorote (Mataguayo or Matacoan; Argentina and Paraguay). In this language, some of the verbal TAM markers also occur in the nominal domain, and at least three of them can have scope over only the nominal but not over the predicate of the clause: the distant past pe(j) , the irrealis -a and the conjectural (epistemic or evidential) -t’ey . The phenomenon is analogous to nominal TAM as described for many languages by Nordlinger and Sadler (2004a) but the difference is that Chorote markers are clitics or independent words and not nominal inflectional morphemes. Regarding the distant past marker, we argue that it codifies tense and not aspect or any other category; however, it is syntactically a modifier of the verb or the noun and not a functional category Tense like the one found in the verbal domain of European languages.