Studies in Language. International Journal sponsored by the Foundation “Foundations of Language” - Volume 24, Issue 3, 2000

Binomial coordinate compounds like English give and take are frequent in Khmer. Once the semantic motivation of these is opaque, the ones that survive are predominantly those which manifest some formal symmetry in the structure of their conjoined roots. The result is that Khmer has an enormous number of words like pell mell or zigzag, but, unlike the English examples, these have neither playful nor pejorative connotations. Moreover, the structural basis of their symmetry is neither rhyme, as in pell mell, nor ablaut, as in zigzag, but alliteration. A cursory survey of some other languages in which symmetrical reduplicative compounds exist reveals that alliteration is extremely rare outside of Southeast Asian languages. At the very least, the abundance of compounds of the spic and span type is an areal feature. But it may be that it correlates with an even more restricted typological feature as well. Khmer, like Thai, is an exclusively prefixing language. There is a well-known cognitive basis for preserving parallelism or symmetry in the backgrounded rather than the focussed portions of things that are brought together. It may be that this principle can account for the tendency to mark symmetry in the initial portion of words in a prefixing language.

This paper examines the preferred syntactic structure of the idea-conveying substantive “intonation unit” (IU) in conversational Japanese. It was found that (a) the clause is the syntactic exponent of the Japanese substantive IU; (b) the clauses are overwhelmingly single-IU clauses rather than multi-IU clauses; and (c) the multi-IU clauses exhibit a higher proportion of multiple new NPs than the single-IU counterparts. The results show that despite claims to the contrary, conversational Japanese cannot be regarded as “highly fragmented”, but basically conforms to the “one IU, one clause” strategy. It is argued that the One New NP per IU Constraint is an information-flow factor motivating toward the “marked” production of multi-IU clauses, that is, the breakup of a clause into phrasal IUs.

In Palenquero, an Afro-Iberian creole spoken in Colombia, the -ba marker appears in a wider range of environments than other TMA markers, which are commonly pre-verbal. While most creole scholars have assumed that -ba is a past anterior, I argue that this marker is in fact a past imperfect. The crucial cases for demonstrating this involve examples in which -ba occurs with non-stative verbs. The significance of this finding counters the alleged general uniformity of the prototypical creole TMA system which does not include past imperfect marking.

Verb agreement in Tibeto-Burman languages follows a different principle than in other, especially Indo-European languages. Instead of limiting agreement to the unification of features of the agreement trigger with those of the target, Tibeto-Burman languages also allow NP features to combine with the features marked by the agreement morphology in an appositional (‘as NP’), partitional (‘NP of’) or relational (‘NP with regard to’) structure. This possibility is shown to result from an associative principle in the syntax-semantics interface underlying Tibeto-Burman, and more generally Sino-Tibetan grammar, which is distinct from the more integrative design of Indo-European languages. These differences also manifest themselves in grammatical relations structure, role semantics, and discourse tendencies.

This paper is concerned with the cross-linguistic expression of a universal information-structure category called the ‘thetic’ or ‘sentence-focus’ (SF) category. The SF category differs from the unmarked ‘predicate-focus’ (PF) or ‘categorical’ category by the absence of a topic-comment relation between the subject and the predicate and it differs from the marked ‘argument-focus’ (AF) category by the absence of a focus-presupposition relation between an argument and an open proposition. The theoretical issue explored here is the question of the relationship between the form and the function of SF constructions, i.e. the question of motivation in grammar. I argue that the form of SF constructions is motivated by the need to distinguish them minimally from corresponding PF constructions. The form and interpretation of a given SF sentence is thus determined not only by the syntagmatic relations among its constituents but also by the paradigmatic relation between the SF sentence as a whole and the corresponding PF sentence, i.e. in terms of a systemic opposition. Since the distinctive property of SF sentences is the absence of a topic-comment relation between the subject and the predicate, SF marking entails the marking of the subject as a non-topic. I show that across languages this non-topic marking of SF subjects tends to be done via those morphosyntactic, prosodic, or behavioral features which are normally associated with the focal objects of PF constructions. The analysis confirms the necessity to treat the pragmatic relations topic and focus on a par with the grammatical relations subject and object and the semantic roles agent and patient. In seeking to explain the form-function fit in SF constructions in terms of the structuralist notion of paradigmatic opposition the analysis challenges both functional and formal generative approaches to grammar.