| الوصف: |
At least since the 1970s, topic has been widely recognized to reflect an important category in most different approaches to linguistics. However, researchers have never agreed about what exactly a topic is (researchers disagree, for instance, about whether topics express backgrounding or foregrounding) and to what extent topics are elements of syntax or discourse or both. Topics are notoriously difficult to distinguish from a range of related phenomena. Some definitions of topic are suspiciously similar to definitions of definiteness, subject, noun and contrast, so the question arises as to what extent topic is a phenomenon of its own. However, topics are also internally diverse. There is disagreement, for instance, as to whether contrastive and non-contrastive topics should be subsumed under the same notion.This talk tries to approach the category type topic bottom-up by considering cross-linguistic functional diversity in marked topics, semasiologically defined as instances of topics with explicit segmental topic markers. The first part of the talk considers the question as to whether topic markers can be defined as a gram type with one or several prototypical functions that can be studied on the basis of material from parallel texts and from descriptive sources. A tentative set of promising candidates for topic markers from 80 languages from all linguistic continents is compiled. It turns out that topic markers are a challenge for the gram approach, because the candidates do not easily cluster to obligatory use in prototypical contexts, but can at the same time exhibit very high text frequency even though they tend to be astonishingly unstable genealogically (one reason why stratified sampling is not particularly useful). In some languages, topic markers are difficult to distinguish from determiners or demonstratives (which, not unexpectedly, are also possible diachronic sources of topic markers).In an influential paper, Haiman (1978) has argued that conditionals are topics, but it has never been verified in a large-scale typological study how common topic markers really are in conditional clauses cross-linguistically. It has also been argued that the initial clause in correlative constructions often has topical properties. In the second part of this talk I will consider to what extent the candidates for topic markers identified in the first part of the talk occur in conditional clauses and in initial free relative clauses and what we can conclude from the results about the relationship between topics and complex sentences. One result is that topic markers are usually different from conditional markers even if they occur in conditional constructions (with the notable exception of grammaticalization of topic markers from conditional converbs). While topic markers are usually postposed irrespective of other word order typologies, conditional markers differ from them in that they co-vary with other word order typologies to a much larger extent.If topic is a grammatical category type like any other, it can be expected to cumulate with other grammatical categories. The third part of this talk focuses on a few instances of cumulation of topic withgender, number and case. Two languages will be discussed in particular detail. Bolinao (Sambalic, Austronesian; Philippines) has developed a number opposition in topic markers. Nalca (Mek, Trans New Guinea) distinguishes both gender, case and number in topic markers. The two cumulative systems are analyzed both from synchronic and diachronic perspectives in order to explore the language specific ways in which topic markers can be integrated into systems together with other grammatical categories. |
