Abstract

This paper uses the electromagnetic midsagittal articulometry technique (EMMA) for the investigation of tongue tip and tongue blade activity during the production of Catalan clusters consisting of consonants specified for adjacent places of articulation. C2 of these clusters is /d/ (a dental in Catalan); C1 is either an alveolar (/1/, /n/, /r/, /z/), a post-alveolar (/?/) or an alveopalatal (/?/). Apical and laminal trajectories for these clusters are compared to those for intervocalic /d/ and for intervocalic C 1. Data show the presence of a single dental articulation for /nd/ and /1d/ and of two sequential articulatory places (alveolar for Cl and dental for C2) in the case of the cluster /rd/. The clusters /zd/ and /?d/ are produced with an alveolar place followed by an intermediate place between the dental and the alveolar zone, which indicates progressive co-articulatory effects (or even progressive assimilation) from Cl fricative onto C2=/d/. Finally, the production of the sequence /?d/ involves presumably gestural blending resulting into an intermediate location between that for C1 and that for C2. The notion of «articulatory constraint» is discussed in the light of these results.

Abstract

The benefits of a Complex Segment representation for palatals have been considered in the light of gliding processes which turn these consonants into glides. Gliding processes have sometimes been described as «destridentization» and «delateralization». But this view gives no explanation of why vocalization is restricted to sorne segments, and plain stridents and laterals do not suffer from it. I have examined gliding processes affecting stridents and laterals to show that a solution based on Feature Geometry is required. The crucial observation is that every speech sound must have a major articulator, and that this constraint applies along the phonological derivation. Since Coronal is the unmarked specification for Place, its presence crucially depends on that of its subordinate features, Strident and Lateral. In this way, if a rule removes Strident or Lateral from the representation, Coronal will also be lost. Consequently, the remaining Dorsal articulator is interpreted as Primary, resulting in a glide. My paper has focused on two cases that involve the delinking of a dependent of Coronal in Majorcan Catalan.

Abstract

This paper focuses on the nature of a segment which is always problematical —the palatal nasal. Phonetic results found in the acoustic, perceptive and articulate spheres, showing the different features of this nasal with regards to others, are the point of departure for a brief reflection on the phonological nature of this segment, taking into account the feature geometry and the articulatory phonology. It concludes by considering the palatal nasal /?/ as a complex segment formed by /n+j/ and the imputation, to temporary effects, of a degree of major o minor assimilation and gestural overlapping during effective perception of /?/, of /n+j/ or of /n+i/.

Abstract

We argue that Onsets can be licensed not only by Nuclei (the universal case), but also (parametrically) directly by prosodic domains such as Word or Phonological Phrase. The surfacing or deletion of Coda/Word-final consonants follows automatically if we allow that (1) a Coda contains no more than one consonant, (2) an Onset consists of a consonant or a nan-strident obstruent + a liquid, (3) an Onset is licensed by (a) a Nucleus or (b) the right edge of a Phonological phrase, (4) unlicensed elements are subject to Stray Erasure. Of these principles, (1), (2), (3a) and (4) are uncontroversial in GU; (4) corresponds to a parameter setting. This somewhat abstract syllabification allows a simple statement of the distribution of several phonological processes.

Abstract

In the standard generative model, phonological rules are strictly ordered, potentially giving rise to complicated derivations with a large number of intermediate representations. In such a model, the order in which the rules apply is as important as the correct formulation of the rules. There are obvious conceptual problems with this theoretical framework, and in recent years several proposals which do away with rule ordering have been put forward. In this paper, we examine a number of vocalic alternations in Basque and Valencian (Southern Catalan), which in a standard generative analysis require a complex rule ordering. Adapting Lakoff' s model we show that, in every case, we can have a simple mapping between two levels with all operations applying simultaneously. In these cases at least, rule order can be shown to be an unnecessary complication.

Abstract

This article intends to demonstrate that the final form of the clusters /ts/ and /dz/ in the Valencian dialect depends upon the behaviour of a series of universal principles referring to the syllabic structure. The author assumes that such restrictions are ranked on a language particular basis, according to the postulates of Optimality Theory (Prince & Smolensky 1993). Then, the outputs for the alveolar clusters will be the result of applying the constraint hierarchy to the set of possible candidates consistent with every input. The different results of the clusters /ts/ and /dz/ in Valencian, and this dialect' s divergences from other Catalan varieties are explained by the differences in the ranking of the principles which each variety presents.

Abstract

The internal structure of syllables has proved to be crucial in the definition of the basic lexical stress system. In the assignment of stress to word entries, Catalan behaves as a quantity-sensitive language: syllable moras are the basis in computation. Two major patterns have been observed at the surface level: (a) Vowel-final words have penultimate stress, (b) Consonant-final words have final stress. The analysis has benefited from the recognition that morphemes, not words, are the proper domain of stress algorithms. The final stress of words ending in a vowel, as in camió or català, conforms to the regular pattern mentioned above, since the underlying form of the last morpheme ends in a consonant. Most exceptions have shown to he only apparent, although special devices are required to account for some cases. Optimality theory has been used to provide a coherent explanation of both exceptions and regular patterns.

Abstract

The purpose of this essay is basically informative; we are dealing with studies related to the prosody of spontaneous speech which are still not very well known in the world of Spanish linguistics (which includes Catalan linguistics). We offer a general state of affairs, and do not seek, therefore, to present the conclusions or results of any individual empirical study, but rather to point out some of the latest channels of investigation.The criteria adopted in the exposition of authors and schools are chronological and epistemological, both of which overlap. Knowledge regarding the demarcative function of intonation has been perfected comparably to the development of linguistics; from the study of ideal and isolated units of sentence we go over to consider total units formed by elements of an inferior category (Sentence Functional Perspective), which leads us to the need to justify how enunciative contents are presented (School of Prague, English Systematics ...). Once we have transcended the frame of sentences and of the analysis of macro-units in oral language we are taken from the study of the demarcative mission of intonation inside a contribution, to denote the boundaries between contributions (turns). 368   We therefore find ourselves before discursive or interactive and conversational approaches (Schools of Edimburgh and Birmingham; I. Lehiste ...). Finally, it seems correct to defend the demarcative viewpoint in the hierarchical function of intonation, establishing boundaries between units in their different levels of linguistic expression (Phonological Hierarchy of Linguistic Units, Treach 1974).

Abstract

The present study provides a preliminary description of a wide variety of pitch contours in Central Catalan, and shows that the phonological framework developed by Pierrehumbert (1980) and her colleagues successfully accounts for their relevant intonational distinctions. Compared to English, Central Catalan seems to make use of a wider inventory of pitch accents, namely, H*, L*, !L+H*, L*+H, H+L*, H+!, H*+L, and L+H*+H (a summary of the pitch accent plus boundary tone combinations is given at the end). As the model predicts, the relative alignment of the H and L levels with respect to the accented syllable triggers intonational oppositions in Catalan. For example, the distinction between an echo and a neutral pronunciation of a question like Què volia ahir en Joan? can be expressed with the use of H+L* vs. H*+L pitch accents respectively.

Abstract

The purpose of this paper is to give a description of one of the Catalan reduplication processes: the reduplication of one light syllable at the left edge of the word. This process exhibits some prosodic conditions in the output as well as in the input. The input must be one heavy syllable, and the output cannot exceed two syllables. Both are subject to Minimality: base words contain minimally one foot, reduplicative words contain maximally one foot. The reduplicative template is a light syllable, its segmental melody being mapped from the base. The description of this process provides evidence for a templatic approach within Prosodic Morphology.