1. A not insignificant problem of Spanish syntax is that of relating the subdeletion sentence (1) to its comparative-deletion congener (2)

• (1) Juan comió más manzanas que peras había traído.

  Juan ate more apples than he had brought pears.

• (2) Juan comió más manzanas de las que había traído.

  Juan ate more apples than he had brought.

At first glance there appears to be little that is problematic and the solutions that have been proposed for English seem readily applicable; under such an analysis (as we shall see below) both sentences are held to be structurally identical, but subject to a more or less extensive deletion rule in accord with the extent of the identity of lexical elements in the compared constituents of the matrix and subordinate clause.

The aim of this paper is threefold: descriptive, methodological and theoretical. It offers a distinctive collexeme analysis of two pairs of near-synonymous complex prepositions (sous le coup de SN/sous le choc de SN; sous l’emprise de SN/sous l’empire de SN), completed by a qualitative analysis. The analysis of the statistically significant co-occurrents suggests that sous le coup de has a more abstract and general meaning than sous le choc de, while the semantics of the second pair are less differentiated. The analysis touches upon several methodological issues; thus, examination of the problem addressed allows for assessment of the method. It also shows that the collostructional analysis offers an interesting contribution to the debate on the existence of a constructional semantics by identifying repelled collexemes that may yield coercion effects.

Comme le faisait justement remarquer A. M. Kinloch dans une récente notice nécrologique (Kinloch 1980), le nom de W. S. Avis est indissolublement lié—avec celui de mon collègue et ami M. H. Scargill—à l’expansion des études de linguistique canadienne, domaine anglais: “it is no belittlement of the work of C.J. Lovell or of other of his associates to say that it was Avis who put Canadian English on the map.” II est done normal de placer la présente note sous l’égide de Avis, et j’en profite pour rappeler les contacts fructueux qui ont existé entre nous tout au long des années 1950 et 1960. Je voudrais cependant adopter ici une perspective un peu différente de la sienne, celle des canadianismes pancanadiens, si l’on peut dire, en ayant à l’esprit le problème très concret d’un dictionnaire bilingue français-anglais à l’usage des Canadiens (Vinay 1962). Je me propose d’examiner brièvement les différents critères retenus par les auteurs pour cerner le concept de canadianisme; et pour compléter l’image que Avis donne du mot canadianism, je ferai porter l’accent sur l’aspect français de la question.

In this paper I will deal with a type of nominalization which appears in Desano. Emphasis will be placed on arguments justifying a given underlying structure of a class of nominalizations rather than on the types of rules needed to convert the underlying structure into the realized nominalization.

A class of nominalizations will be presented here. I will give evidence to show that these forms are to be derived from relative clauses. Consider the following forms:

Dans cet article nous présentons une nouvelle théorie pour rendre compte du maintien du e [ə] final en ancien français dans l’évolution historique des mots suivants:

Nous allons présenter d’abord les descriptions traditionnelles de ces évolutions, en précisant leurs limites, puis la solution de Fouché (1966), et enfin dans une troisième partie, notre propre solution.

Words which in southern Greek dialects contain [azis], [úsis], and other sequences of stressed vowel + sibilant + [i] + sibilant, appear widely in the north with [áys], [úys] etc., and the phenomenon constitutes a major crux in modern Greek phonology. Although the problem has been widely discussed, no really satisfactory solution based on attested processes has been offered. It is the purpose of this paper to review the data, criticize existing treatments, and offer a new proposal.

The aim of this article is twofold. First, based on passives in Japanese and Korean, it expands the domain of the applicative head (Appl) to include an argument in the structural subject position. These languages provide evidence for a new type of Appl, peripheral Appl, distinct from the well-known high Appl in Bantu: unlike high Appl, peripheral Appl is the highest argument-introducing head under T, and can merge above VoiceP. The recognition of peripheral Appl makes it possible to account for the cross-linguistic positional variation among affectee arguments. Second, the article provides articulated clause structures for passives in Japanese and Korean in terms of Appl. The applicative account proves to be successful in providing a unified account of Japanese possessive and non-possessive passives, and of Korean possessive passives and causatives.

Various proposals have been made in the literature, arguing that bound pronouns are, in some sense, deficient. This article tests this hypothesis with Malagasy pronouns, specifically in the context of Zribi-Hertz and Mbolatianavalona's (1999) claim that Malagasy pronouns may be bound only when they are missing the NumP layer of DP. Zribi-Hertz and Mbolatianavalona show further that other syntactic properties are also attached to the lack of NumP. The variety of Malagasy investigated here (Malagasy2), behaves differently from the one described by Zribi-Hertz and Mbolatianavalona (Malagasy1), and these differences lead to two conclusions. First, there are no syntactically deficient pronouns in Malagasy2, yet these syntactically complete pronouns may, in fact, be bound. Second, Zribi-Hertz and Mbolatianavalona are nevertheless correct that the lack of NumP accounts for a cluster of properties, since none of these distinctions between pronouns that they describe are found in Malagasy2. More broadly, we conclude that pronominal binding does not require syntactic deficiency.

Omission of various components of a sentence is a common phenomenon in first language (LI) acquisition across different languages. Omission can affect arguments, such as subjects and objects, as well as functional elements, such as determiners and inflectional morphology (Borer and Rohrbacher 2002, Pérez-Leroux et al. 2008, among others).