Revista de Estudos da Linguagem

19-43

Authors: Adriana S. Marusso

This article presents the basic principles of the Acoustic Theory ofSpeech Production, also called Linear Source-filter Theory of SpeechProduction (FANT, 1960; KENT & READ, 1992; PICKETT, 1999).This description is justified since the theory is not available inPortuguese. This theory is useful in understanding articulatoryacousticrelationships. It may be applied to the analysis of a singlelanguage or in cross-linguistic studies. It also provides a foundationfor many procedures for the acoustic analysis of speech.

59-92

Authors: Werner Thielemann

The description of Amerindian languages is closely linked to theevangelization and catechesis of indigenous American peoples byreligious orders (Franciscans, Jesuits and others). Grammaticographyof native languages, since the discovery of America and for a longtime, was being influenced by the translation practice of catecheticalLatin texts and, in this way, from Ars grammaticae principles andterminology of Latin.Though being aware of the particularities of indigenouslanguages, missionaries do not trangress the degree of descriptive hybridity, evidenced by the Brasilian Grammar of Anchieta and alot of Mexican nahuatl grammars (Olmos, Carochi), discussed inthe article.Paradigmatic change arises with Historical ComparativeGrammar in Europe. The description of American languages changestotally, when, whit his knowledge of Basque, Whilhelm vonHumboldt recognizes the propositional value of the nahuatl verband identifies incorporation as the organizing principle. He initiallygives an adequate description and metalinguistic teminology forthe particularities of Amerindian languages. Throughout his life,he is tried to determine the structures of American languages, butthe very results of his endeavour were not published until after hisdeath (1835).

129-154

Authors: Maria José Foltran; Gisele Crisóstimo

In this paper we analyse the occurrences of verbal participial formsas adjuncts within a noun phrase. We also establish constraints thatdo not allow certain monoargumental verbs to occur in thisenvironment. Differently from participles of two argument verbs,we noticed that some participles of one argument verbs cannotwork as adjectives, while others can. This is because of the type ofarguments each verb selects: since the arguments of certain oneargument verbs share the same features of internal arguments, theparticiples of these verbs can occur as adjectives, and therefore wecan classify these as unaccusative verbs. We also investigate thesemantic properties of the arguments which allow a participialform as a modifier.

171-182

Authors: João Miguel Moreira Auto

The aim of this article is to discuss some important questions onthe meaning of the concept of “author”. First, it explains how RolandBarthes criticizes such a notion by stressing the reader’s role in thecomprehension of a text’s meaning. Then, Foucault is mentioned:he notes the important role the reader plays for the unity of thetext’s sense, but he is also aware of the possibility of a readable“author function”; the reader himself is the one who makes usrealize the actual existence of authorship. Also discussed is Possenti’shypothesis that it is possible to look for traces of authorship, forexample, in a simple text. Finally the following definition of theconcept is proposed: authorship is a kind of linguistic and discursiveattitude, an act of inscription by which the agent takes one’s standand articulates foregoing discourses and a later reception, accordingto personal tactics and specific style.

19-36

Authors: Ângela Maria Vieira Pinheiro; César Reis

Experiments of reading aloud isolated words, in which the reactiontime (RT) is measured based on the time between the appearanceof the stimulus on the computer screen and the beginning of theemission of its pronunciation by the subject, were used test thehypothesis that the phonetic nature of the initial phonemes of thewords affects the measurement of the RT, since for words beginningwith stop consonants, for example, the sound itself contains an average silence lasting 100ms. If the phonetic nature of the firstsound influences the measurement of the RT, this means that intasks of reading aloud, this phenomenon should be described andcontrolled. Thus, a list of 58 words varying in the phonetic natureof their first sound – vowel, stop (voiced and unvoiced), fricative(voiced and unvoiced), nasal and lateral – were presented tochildren from 1st to 4th grades and to adult readers, with their reactiontimes recorded. The first results do not confirm the hypothesis thatthe silence of the stop sound is affecting the RT, since no significantdifference was found between the average RT between the stopsand fricatives. The shortest RTs were recorded for the nasalconsonants and the vowels (o, a) for the adult group and for allthe grades (excepting the vowel a for the first grade). The longestRTs were recorded for the fricative sound R (except in the thirdgrade), and u (except in the adult group). The behavior of thesound u was distinctly curious, not following the pattern of theother vowels in the children’s reading.

51-64

Authors: Ricardo Augusto Hoffman Bion

The development of models like Flege’s Speech Learning Model and Best’s Perceptual Assimilation Model directed Interphonology research to the perception of target language contrasts. These model shave triggered the development of perception tests, often using synthesized stimuli. The present paper discusses the ways previous studies have used these stimuli in vowel perception tests, and also presents the results of a pilot test developed in Brazil. While previous tests were concerned with checking the acoustic cues used in the perception of vowel contrasts, the present paper analyzes a test that can access the categorical boundaries in the perception of English front vowels by Brazilian Portuguese speakers and native English speakers. The stimuli for this test were created with the ASLsynthesizer; innovations can be observed in the test developmentand application.

93-109

Authors: António Romano; Lurdes de Castro Moutinho

Cette étude a pour but immédiat celui de mettre en évidence quelqueséléments intonatifs particulièrement typés dans des productionslinguistiques communes du Portugais parlé au Brésil (PB) par rapportaux schémas réalisés le plus communément – dans les mêmesconditions – par des locuteurs du Portugais européen (PE).Le manque d’études dans ce domaine a été une des motivationsde cette recherche.Notre approche privilégie la recherche de configurations mélodiquesassociées à des figures expressives manifestées sur l’échelle des contours d’unité tonale et intonative avec un aperçu qui tienne encompte des variables de structuration plus globales.Notre étude est basée sur un texte lu-interprété par 6 locuteurs (3locutrices portugaises de Aveiro et 3 locuteurs brésiliens).Nos résultats montrent qu’en PB émerge assez systématiquementprofil final de modalité déclarative, présentant des caractéristiquesqui peuvent donner lieu à une sorte de cliché mélodique. Ce contour,ancré à des rapports de hauteur très systématique, se manifeste parune tonie caractérisée par une brusque monté mélodique sur lapréaccentuelle, suivie d’une diminution abrupte de hauteur qui sestabilise ensuite sur la voyelle accentuée, pour se réduire encoresur les voyelles postaccentuelles. Cette tonie est précédée souventpar une longue séquence quasi-monotone qui réalise une protonieplate assez aiguë, généralement inconnue au PE.

143-157

Authors: Larissa Santos Ciríaco; Lorenzo Teixeira Vitral; Cesar Reis

In Brazilian Portuguese reduction occurs in some pronouns, for instance, cê (from você); el and ez (from the pronouns ele and eles);num (from não) and others that have been analyzed as becoming clitics (VITRAL & RAMOS, 2002). In order to prove this hypothesis,we collected a corpus with almost 500 utterances containing the reduced forms above. The reduced forms were acoustically analyzed considering the intensity and duration parameters and then compared with those parameters of real clitics. This paper presents the research results and the conclusions we have achieved in this work.

169-179

Authors: Erica Reviglio Iliovitz

According to Felizatti (1998), dysarthria is a disorder in motor production that affects patterns of movement, coordination and strenght of articulatory organs. It can be caused by brain damage and it is usually characterized by slow speech. The analysis of some prosodic aspects of almost unintelligible dysarthric patients’speech indicates that the strange reaction their speech causes in their listeners is mainly due to the prosodic strategies they use (consciouslly or not) to improve their oral language intelligibility.In other words, despite the fact that dysarthric patients’s brain lesions cause articulatory difficulties, they are able to compensate them through prosody.

193-212

Authors: Regina Célia Fernandes Cruz; Jailma do Socorro Uchôa Bulhões; Léa da Silva Fernandes

This work represents an attempt to facilitate the automatic accessto the spontaneous Portuguese data for studies on experimentalphonetic area and speech synthesis, maintaining qualitative andquantitative information of the data. This corpus constitutes morethan 30 hours of recording with samples of two Portuguese spokendialects in Amazonia Region: the Amazon Portuguese (AM) and theAfro-Brazilian Portuguese (ABP). As a baseline, samples of the StandardBrazilian Portuguese (PB) were also used. The corpus inventory,the digitalization of the speech data and the phonetic segmentationusing the PRAAT software preceded the construction of the oral database.Using language HTML and other resources of computer science adocumentation of the database in a Internet site becomes possible.

247-268

Authors: Thaïs Cristófaro Silva, Daniela Mara Lima Oliveira

This paper investigates cases of sound variation involving sequences of (sibilant + consonant) in Brazilian Portuguese (BP). Emphasis is given to the analysis of alveolar (sibilant + stop) sequences as in the word – festa [ÈfEsta] – and to the alveopalatal (sibilant + affricate)sequences – as in the word triste [ÈtRiStSi]. The cases of variation we will be dealing with involve the loss of the stop or the affricate where only the sibilant is manifested: festa ‘party ‘ [ÈfEsta] à [ÈfEsa]and triste ‘sad’ [ÈtRiStSi] à [ÈtRiSi]. This paper evaluates the role of phonetic properties as potential motivation for the variation:agreement of place of articulation and also agreement of voicing for the consonants in the sequence is considered. The fact that the sequences involved in the variation are related to a complex syllable pattern – that is, (coda-onset) – will also be explored. Some structural aspects which may condition the phenomena will be addressed:stress placement and the quality of the vowel which precedes the(coda-onset) sequence. It will be suggested that the lack of conditioning structural factors leads us to evaluate this case of variation within the Lexical Diffusion model (WANG, 1969). Lexical Diffusion appears to be the most appropriate approach to analyse the phenomena addressed in this paper, since it accounts for the .different behaviour of some structurally similar sequences. Finally,the paper indicates some aspects to be pursued in future research.

289-318

Authors: Luciani Tenani

In order to show the importance of phonological phraseproeminence in Brazilian Portuguese, we make an intonationalanalysis of contexts which characterize vowel merge and voweldeletion processes. We will argue that this phonologicalproeminence blocks vowel sandhi processes because it carries thesyntactic notion of recursivity in Portuguese. The results contributeto syntax-phonology interface discussion.

345-363

Authors: Raquel Santos

The focus of this paper is on the relation between syntax and prosody,by discussing the role of the empty syntactic categories in the application of the stress shift rule. The discussion about the influence of empty categories on phonological processes is not new. Usually, phonology papers argue that the access to syntax is indirect, and there is no access to phonological empty syntactic categories (cf. NESPOR;VOGEL, 1986). However, Santos (2002) argues that phonological empty syntactic categories can intervene on the rhythmic organization of utterances. She defends that pro intervene on stress shift rule, while trace is not taken into account for that rule. In order to show that,she tests utterances with stress clash between verb and adverb.In this presentation, the syntactic structures with verbs which can have a transitive or intransitive reading are tested with pro.The hypothesis is that when there is a transitive reading, the stress clash can not be solved. In the sentences where the intransitive reading is possible, there is no pro between the verb and adverb, and the stress shift rule is allowed. This prediction can be used as an evidence which favors Galves (1989) analysis that null object in Brazilian Portuguese is a pro, and not a trace.

383-404

Authors: Renée Lambert

As propriedades morfológicas e fonológicas da formação diminutivapor reduplicação são estudadas em função das hipóteses deLowenstamm (1996, 1999) sobre a organização universal dossegmentos fonológicos e mais particularmente sobre a sílaba inicialdas palavras de categorias lexicais essenciais. É proposto que aposição inicial seja o lugar da reduplicação diminutiva em francês.Fatos do francês de Quebec bem como do antigo grego sãoapresentados seguidamente para apoiar a hipótese segundo a qualas palavras são representadas universalmente por uma seqüênciaestrita de consoantes e de vogais (CVCV...).

423-437

Authors: Teresinha de Moraes Brenner

The concept of natural class delineated in a geometric figure dates back to the Prague School. The idea was incorporated by multilinear phonology. Inserted on phonological different levels and plurilinear lines, dominated by specific categorial nodes, the phonological features are organized hierarchically. Particular context sallow for the occurrence of phonological processes characterizing linguistic variation.In the Portuguese language of Brazil, certain phenomena are explained by the substitution of features registered in related subgroups belonging to a broader category. Other processes allow for the substitution of elements by others in the hierarchy, relating to the degree of aperture and sonority – phonemic reinforcement orweakening. Finally, assimilation and dissimilation are illustrated and can be explained by segmental transparency or opacity. As aconsequence, the geometry of the features re-covers the linguistic variation.

459-473

Authors: Andreia Schurt Rauber; Barbara Oughton Baptista

Este estudo revisa, verifica e explica os diferentes resultados obtidos por Carlisle (1991, 1992, 1997) e Rebello (1997) sobre a produção de encontros consonantais iniciais /sC(C)/ em inglês por falantes de português e espanhol, complementando os resultados obtidos em pesquisas anteriores na área de interfonologia, já que o mesmo corpus foi usado para comparar a produção de falantes das duas línguas nativas. Os resultados revelam que a adição de uma vogal(epêntese) aos encontros consonantais iniciais é a estratégia mais utilizada para se lidar com as dificuldades relativas às estruturas silábicas mais difíceis, o que pode ser resultado da relação entre a interferência da língua materna e os universais lingüísticos. Os resultados fornecem informações relevantes ao ensino da pronúncia do inglês no Brasil e nos demais países sul-americanos.

491-504

Authors: Magliane De Marco

Language acquisition studies have pointed the first language (L1)influence on this process. In a research with adolescents and adults,Fernandes (2001) and Silveira (2002) show that, in the acquisitionprocess of English as foreign language (L2) by Brazilian Portuguesespeakers, the L1 influences the L2 syllable structure acquisition.However, concerning to children, it seems that there is less infuenceof L1 on the foreign language acquisition. Based on that, the aim ofthis paper is to analyze the word-initial /s/ clusters in English, as wellas the possibility of syllable structure transference from first languageto L2. The data is from a child that is still developing her phonologicalsystem and her age was 3:4 (years/months). She’s acquiring BrazilianPortuguese as a first language and English, as a second.

515-523

Authors: Tanira Castro

The Portuguese language, in the discourse of Brazilians, has specificfeatures that are manifest in a more evident form in its lexis andphonetics, which allows us to speak, therefore, of a BrazilianPortuguese. Leaving aside the discussion of the status of BrazilianPortuguese, we must take into consideration the indisputably broadvariability of its phonetic system, which is due not only to linguisticbut also to sociolinguistic factors, which, in their turn, cannot buthave an influence on the phonetic formation of Brazilian discourseduring, for example, the learning of the Russian language (RL).Besides, the written form of some letters, both in Portuguese andRussian, causes not only phonetic but also graphic-phoneticinterference to arise whether during reading, writing, or oral discourse. The specifics of the native language of Brazilians istransferred to the discourse in the RL, mainly during the initialstage of learning. Research on the phonetic interference of thediscourse of Brazilians who study the RL, is still in the early phase,but, even so, through the experimental results already obtained,we can speak of the specific and universal features of itsmanifestation. A priori, these violations have an orthoepic andorthophonic character, for they have a lot to do with the differentialand integral criteria of the phonemes, since these violations mayoccur because of the specific distributive and phonotactical relationsof the languages in question. The sources of the errors can besubjectively defined by the manner in which the learner assimilaresand perceives the physical and linguistic functions of both languages.Both in written and oral discourse, the action of graphic-phoneticinterference is observed.

39-52

Authors: Glauco F. G. Yared; Jaqueline V. Gonçalves; Plínio A. Barbosa; Luís G. P. Meloni

The main purpose of this work is to analyse the projections ofmaximum and minimum points of two articulatory trajectories duringspeech production over the acoustic signal, namely, the mouthaperture and the jaw opening. In order to do so, we have chosena native French speaker who repeated 5 times the followingsentence: “C’est pas ububuz, c’est pas ybybyz, c’est pas iziziz, c’estpas azhazhaz”. It seems reasonable from the analyses shown toestablish a correspondence between articulatory and acoustic events,which happens to be useful to help determine acoustic boundaries.The articulatory boundaries coincide with the middle of the vowelsand the mouth aperture seems to be a more consistent parameterthan jaw opening for establishing these regularities.

89-113

Authors: Renée Lambert

As propriedades morfológicas e fonológicas da formação diminutivapor reduplicação são estudadas em função das hipóteses deLowenstamm (1996, 1999) sobre a organização universal dossegmentos fonológicos e mais particularmente sobre a presençade um sítio morfológico ao inicial das palavras de categorias lexicaisessenciais. É proposto que o sítio inicial seja o lugar da reduplicaçãodiminutiva em francês. Factos do francês quebequeses bem comodo antigo grego são apresentados sequidamente papa apoiar ahipótese segundo a qual as palavras são representadasuniversalmente por uma sequência estrita de consoantes e de vogais(CVCV...).

133-158

Authors: Ana Cláudia de Souza

This study analyses the simple and compound metaphorical lexicalunits that constitute the keyword list of the economic news genre.The analysed texts are taken from the newspaper Folha de SãoPaulo, and are part of the corpus compiled by NILC – NúcleoInterinstitucional de Lingüística Computacional. The completecorpus is constituted by prose texts from several social domains,among which the journalistic one. Although news is considered asan informational speech genre in which the referential function oflanguage predominates, the use of metaphorical expressions isfound in this kind of text. The theoretical background, consideringspeech genre in the light of Bakhtin’s theory, and concepts ofnews and metaphor, is presented in the first section of the paper.Then, the methodological criteria are described, and the set ofdata is presented and discussed.

175-190

Authors: Christina Abreu Gomes

This paper presents a preliminary study of the emergence ofstructured variation in first language acquisition. We studied thespeech of 5 children aged 2:6 to 5 from the city of Rio de Janeiro.Our aim was to understand how children cope with variable inputin the acquisition of dative alternation in Brazilian Portuguese. Weassume that children’s variability is in part structured and is linkedto variability in adult speech. The results reveal that not all theavailable variants in the speech community are used in children’sperformance and that they tend to use the most frequent ones.

11-15

Authors: Carlos Franchi, Márcia Cançado

Relações semânticas:predicação e papéis temáticos.

83-123

Authors: Carlos Franchi; Márcia Cançado

Teoria generalizadados papéis temáticos.

155-176

Authors: Carlos Franchi

Nesta versão limito-me aqui a aspectos introdutórios eprincipais justificações empíricas (deixando de reproduziros itens sobre adjetivos, advérbios, preposições e gerúndio).Para facilitar a comunicação, evito entrar desde logo natecnicália da lógica combinatória e da gramática categorial1em que desenvolvo minha pesquisa. Isto oculta muitosaspectos do modelo teórico, mas facilita nosso diálogo.

49-69

Authors: Jean- Paul Bronckart

L’article de Jean-Paul Bronckart est centré sur le problèmecontroversé du classement des textes en le situant dans le cadre dela psychologie du langage et dans le cadre théorique del’interactionisme social. La multiplicité et l’hétérogénéité des critèresde classement et de nomenclature ont abouti à des classementshomogènes, intermédiaires et hétérogènes. L’auteur part deconsidérations intermédiaires sur la notion de genre et sur lesproblèmes de classement et présente, sous une forme résumée,quatre thèses lapidaires des fondements du cadre de la psychologieinteractioniste sociale. Un modèle de production textuelle estsuggéré à partir du statut psychologique des textes empiriques etdes opérations psycholinguistiques qui y sont sous-jacents. Unetypologie centrée sur quatre entités basiques relève de ce modèle,nommé “archétypes discursifs”: discours interactif, discoursthéorique, récit et narration. L’auteur tient à montrer que lacompréhension du fonctionnement des textes est essentielle, visà-vis d’une lecture qui envisage une approche pédagogique.

95-118

Authors: Maria Augusta Cavaco Miguel

This paper discusses vowel behavior in relation to syllable structurein European Portuguese. The analysis is based on GovernmentPhonology (Kaye, Lowenstamm and Vergnaud, 1985, 1990).We will argue that syllabic constituents in European Portuguesedetermine, to a great extent, vowel behavior. Branching nucleiand rhymes are the structures which most clearly determine thevalue of vowels in unstressed positions. As we will see throughoutthis paper, vowel reduction is clearly conditioned by syllablestructures.

149-162

Authors: Carmi Ferraz Santos

Les études traditionnelles de l’argumentation soutiennent quel’émergence du discours argumentatif a lieu tardivement chez lesapprentis, c’est-à-dire vers 14-16 ans. Dans une première partie, ennous appuyant sur le concept d’ énonciation pour interpréter etcomprendre les discours produits par des enfants de 9-10 ans enmilieu scolaire, ce travail analyse certaines marques qui indiquentla capacité argumentative déjà présente dans les textes produitspar des enfants de cet âge. Dans une seconde partie, nous faisonsune discussion du rôle de l’ école dans l’enseignement du discoursargumentatif.

173-196

Authors: Maria Cecília Perroni

The aim of this paper is to present some of the studies on theacquisition of Brazilian Portuguese as a first language, which havebeen published in the last few years in our country, making use ofthe generative grammar theory in their analysis of the child´s speechdata. The topics that have been analysed are the systems: D, IP, CP,and complementation/adjunction in complex sentences. Additionalresearch on the acquisition of Brazilian Portuguese as a secondlanguage is also shown. The studies briefly discussed here cancontribute to prove the relevance of the generative theory in thedevelopment of the Language Acquisition area.

223-244

Authors: Kátia Modesto Valério

The aim of this paper is to describe some of the linguistic andstrategic means used for the construction of social relations by theparticipants of face to face interactions. Based on the notion offacework introduced by GOFFMAN (1967), we have analyzed theinteractional moves of 12 members of a social subgroup in aBrazilian speech community when asked to express their views ona few polemical topics in individual and group interview situations.Having in mind that the blend of such interactional moves mightlead to the overriding line of conduct adopted, we have set out todescribe the interactional function of the linguistic andconversational strategies in which these moves seem to be partlyencoded. The strategies focused on here are the ones whose socialmeaning contributes to what GOFFMAN (1967) called “aggressiveuse of facework”, here referred to as Agonism.