Volume 49, Issue 5

The present article investigates the pattern of vowel epenthesis in word-initial position (prothesis) in the Walliser dialect of German. In that dialect, /a/ is inserted before a word-initial /r/, provided that an unstressed vowel does not precede it. It is argued that Vowel Prothesis is a strategy for avoiding word onsets beginning with highly sonorant nonsyllabic segments and since /r/ is the most sonorous nonsyllabic segment in that position, the process therefore only takes place in that context. Vowel Prothesis is inhibited after an unstressed vowel because Walliser German avoids hiatus sequences consisting of unstressed vowels.

Capitalizing upon the recursive nature of Saxon Genitives in English, this note examines the constituent structure of double Saxon Genitives such as (he took) my sister-in-law's child's life . An analysis of 200 cases from both the spoken and the written modality reveals a distribution of 70% left branching and 30% right branching. This left-branching bias exists in the face of a general right-branching preference in English. It is argued that left branching stems from the recursive structure of the Saxon Genitive. As listeners and readers take in the acoustic signal, they apply probabilistic decoding strategies and expect the more frequent nonrecursive Saxon Genitive to come up. Thus, the first two lexemes are treated as sister constituents. This introduces an advantage for left branching because such a configuration can be easily expanded into a double Saxon Genitive. By contrast, a right-branching structure leads the listener up the garden path. Branching direction in double Saxon Genitives is also found to be influenced by the animacy of their constituents. An animate possessor amplifies the left-branching bias whereas an inanimate possessor increases the incidence of right branching.

This paper discusses a hitherto undescribed usage of the particle so as a dedicated focus marker in contemporary German. I discuss grammatical and pragmatic characteristics of this focus marker, supporting my account with natural linguistic data and with controlled experimental evidence showing that so has a significant influence on speakers' understanding of what the focus expression in a sentence is. Against this background, I sketch a possible pragmaticalization path from referential usages of so via hedging to a semantically bleached focus marker, which, unlike particles such as auch ‘also’/‘too’ or nur ‘only’, does not contribute any additional meaning.

The domain of space presents some interesting properties. On the one hand, spatial understanding by human beings is said to be universal and to share the same biological heritage; one the other hand, the linguistic systems encoding spatial knowledge vary strikingly. This paradox raises fundamental questions, including for language typology. In this context, the present study examines adults' expression of caused motion events in Chinese versus English in the frameworks proposed by Talmy, Toward a cognitive semantics, vol. 2: Typology and process in concept structuring, MIT Press, 2000 (i.e., verb-framed versus satellite-framed) and Slobin, The many ways to search for a frog: linguistic typology and the expression of motion events, Lawrence Erlbaum, 2004 (i.e., equipollently framed) regarding the expression of spatial information across languages. The results of a cartoon-based production task reveal that, although both languages tend to express equally frequently the same set of semantic components for motion (Path, Manner, Cause, etc.), Chinese greatly differs from English in terms of where this information is encoded and how it is distributed across utterances. On the basis of these findings, it is suggested that it is better to understand Chinese on its own merits: this language shows both satellite- and verb-framing properties, thus demonstrating a “parallel” system.

In this article I test Hawkins' (Efficiency and complexity in grammars, Oxford University Press, 2004) theory on wh -questions against Basque data from the 16th to 20th centuries. I focus on wh -fronting. Basque has experienced a noticeable word-order change in wh -questions; namely, a restricting conventionalization (fixation) of relative orders out of a situation with freer word-order possibilities. What has been fixed is the adjacent position of the verb relative to the fronted wh-phrase. The result of the present testing is that the predictions and principles proposed by Hawkins are well supported by the data. Yet, it seems that other principles may also account for some of the factors that play a role in the evolution of historical Basque.

This study claims that, in contrast with previous proposals in the literature, essentially all instances of stop epenthesis in two consonant clusters (e.g., [ml] > [mbl], [ls] > [lts], [wl] > [wgl]) may be attributed to the articulatory requirements and aerodynamic constraints involved in the production of the original cluster. The inserted stop results from the perceptual categorization of a transitional closure event. Several mechanisms may give rise to this momentary stoppage of air, and to an intraoral pressure rise which causes the stop burst to become prominent enough so that the emergent stop can be successfully perceived. Apparently exceptional cases such as [nl] > [ngl] and [sl] > [skl] are accounted for through direct epenthesis assuming that [l] is strongly dark and thus, produced with a back postdorsal constriction. Data on stop deletion in consonant clusters appear to be in support of this production-based explanation of stop insertion.

The article provides an overview of the grammatical realization of focus in four West Chadic languages (Chadic, Afro-Asiatic). The languages discussed exhibit an intriguing crosslinguistic variation in the realization of focus, both among themselves as well as compared to European intonation languages. They also display language-internal variation in the formal realization of focus. The West Chadic languages differ widely in their ways of expressing focus, which range from syntactic over prosodic to morphological devices. In contrast to European intonation languages, the focus marking systems of the West Chadic languages are inconsistent in that focus is often not grammatically expressed, but these inconsistencies are shown to be systematic. Subject foci (contrastive or not) and contrastive nonsubject foci are always grammatically marked, whereas information focus on nonsubjects need not be marked as such. The absence of formal focus marking supports pragmatic theories of focus in terms of contextual resolution. The special status of focused subjects and contrastive foci is derived from the Contrastive Focus Hypothesis, which requires unexpected foci and unexpected focus contents to be marked as such, together with the assumption that canonical subjects in West Chadic receive a default interpretation as topics. Finally, I discuss certain focus ambiguities which are not attested in intonation languages, nor do they follow on standard accounts of focus marking, but which can be accounted for in terms of constraint interaction in the formal expression of focus.