Volume 39, Issue 3-4

 Structuralism sought to introduce various kinds of autonomy into the study of language, including the autonomy of that study itself. The basis for this was the insistence on categorial autonomy, whereby categories are identified language-internally (whether in a particular language or in language). In relation to phonology, categorial autonomy is tempered by grounding: the categories correlate (at least prototypically) with substance, phonetic properties. This is manifested in the idea of ‘natural classes’ in generative phonology. Usually, however, in more modern grammars, despite some dissent, no such grounding (in meaning) has been attributed to syntax. This attitude culminates in the thesis of the ‘autonomy of syntax’ which was put forward in transformational-generative grammar. In what follows here it is argued that the consequences of this are very unfortunate. Distribution alone is insufficient to determine the identity of categories; what is relevant is the distribution of the prototypical members of the category, where prototypicality is notionally defined. Prototypical nouns, for instance, denote concrete, discrete, stable entities. Syntax, as well as phonology, is grounded. Groundedness ensures that only the prototypical behaviour of semantically prototypical members of a category determines its basic syntax; and this syntax reflects the semantic properties. Groundedness filters out potential syntactic analyses that are incompatible with this. For instance, given the diverse semantic characters of prototypical nouns and verbs, groundedness predicts that the X-bar theory of syntactic structure, which attributes parallel projections to lexical categories, is false. Consideration of the syntax of nouns and verbs confirms that this is indeed the case. The attribution to syntax of categorial autonomy without grounding should be abandoned.

 The paper introduces a unit of syntax beyond the constituent called the chain . A number of mechanisms are shown to be sensitive to chains, e.g. the formation of predicates and idioms, the ellipses of gapping, pseudogapping and VP-ellipsis, and the elided material of stripping and answer fragments. The presentation is couched in a surface syntax, dependency-based framework, as opposed to a constituency-based one. While the chain can be defined in a manner consistent with constituency, doing so requires that one adopt some controversial assumptions about the nature of constituency structure. The potential of the chain concept is great; it is the tool necessary to address the manner in which semantic compositionality occurs in the syntax.

 1. Introduction This paper argues that a sideward/parallel movement analysis of so-called “Left- Peripheral Ellipsis” (LPE; Sag 1976) and Gapping constructions improves on the traditional deletion analysis in at least two important ways. First, it avoids the need to appeal to the notion of non-constituent deletion. Second, we attempt to show that directionality effects (Ross 1970 and others) can be derived without recourse to the head parameter or directionality stipulated in the application of transformational rules (which would require linear order in the narrow syntax and, further, would need to be parametrized); rather, we show that apparent directionality in LPE and Gapping constructions is a natural consequence of the interaction of Copy, Merge, PF-deletion of copies, and Cyclic Spell-Out— an effect that is derived only under a movement-based approach.

 Die Interaktion zwischen Präpositionalsyntax und der Semantik lokaler Relationen ist sowohl typologisch als auch theoretisch gut studiert. Der Einfluß phonologischer Eigenschaften auf die Syntax präpositionaler Elemente ist hingegen viel weniger untersucht worden. Der vorliegende Artikel konzentriert sich auf diesen Aspekt der Grammatik der Kategorie P. Die Daten dazu stammen aus Chemini Berber, einem Dialekt des Berber aus der Taqbaylit-Familie (Afroasiatisch, Nördliches Algerien). Wir beschreiben eine Reihe von neuen Phänomenen betreffend die Relation zwischen phonologischem Gewicht und syntaktischem Verhalten. Wir versuchen eine umfassende Erklärung, die auf der Analyse in Guerssel (1987), und der Unterscheidung zwischen Wurzeln und Stämmen basiert. Es soll gezeigt werden, dass eine detaillierte morpho- phonologische Analyse signifikante Vereinfachungen in der syntaktischen Struktur der Präpositionen erlaubt.

 This paper aims at showing that the scope of structural phonemics transcends the limits of the ‘foundations of phonology’, contrary to what is tacitly assumed and appears from some textbooks. It will be argued that the classical concept phoneme, defined as a set of distinctive features, presents both obsolete and still relevant properties. One of these properties, linearity , should clearly be abandoned, as follows from acoustic-perceptual evidence as well as from some types of sound change. Thereby, the phoneme in its purest sense can be said to have been superseded by one major trend characterizing post-SPE phonological theory: multilinearity. However, a phoneme-based property of distinctive features, their locality , is still valid, and is empirically supported by cross-linguistic variation. Now, locality and non-linearity are apparently contradictory. It will be shown that this contradiction cannot be resolved, and that both feature properties cannot be captured, unless consonants and vowels are assumed to be universally segregated within phonological representations. This issue leads to several predictions on C/C and V/V interactions, converges with independent processual evidence like vowel-to-vowel assimilation, and addresses the question of the relationship between phonology and morphology.

 Besides their common lexical reading, the Spanish verbs amenazar ‘threaten’ and prometer ‘promise’ can also yield a modal reading. The shift from the former reading to the latter implies a change from a control structure to a raising structure and involves grammaticalization, more specifically auxiliation, which ends up in a layered co-existence of the two (or more) constructions. This paper is mainly concerned with the syntactic and semantic categorization of the grammaticalized quasi-modals amenazar and prometer . Generative accounts have focused on the difference between raising and control, but have not addressed the differences between the two verbs. This paper highlights that modal amenazar faces less constraints than modal prometer . Furthermore, the account presented below will show that epistemic and evidential auxiliaries have scope over the qualification expressed by modal amenazar and prometer , and are therefore not to be called “epistemic”. The paper also deals with the tense restrictions that these verbs undergo. Tenses such as the present perfect are not available in modal constructions with amenazar and prometer because of the activation of agentive patterns inherent to the lexical semantics of the verbs in question. The tense criterion is valid for verbs for which the grammaticalized form is not the most frequent one, but it does not account for true modals in Spanish and other Romance languages. The infinitive criterion turns out to be a good alternative.

 1. Einführung In fast allen europäischen Sprachen gibt es zumindest eine Höflichkeitsdistinktion im Pronominalsystem. Diese Höflichkeitsdistinktion betrifft fast ausschließlich die Referenz auf den Hörer, das heißt die zweite Person. Die Distinktion, von der hier die Rede ist, wird im Deutschen durch die Opposition von du/Sie und im Französischen durch tu/vous ausgedrückt. Durch die Wahl einer Form in diesem Kontrastpaar referiert der Sprecher S nicht nur auf den Hörer H, sondern markiert zugleich einen Aspekt der sozialen Relation, die zwischen S und H besteht, bzw. der Einschätzung dieser Relation auf Seiten des S. Seit der berühmten Studie von Brown & Gilman (1960) über Höflichkeitsdistinktionen in Pronomina nennt man diese Distinktion auch T/V, abgeleitet von Lateinisch 2SG tu versus 2PL vos . T bezeichnet das familiäre Pronomen, V das höfliche Pronomen.