La comparaison des enfants qui acquièrent une langue seconde (L2) à des enfants avec trouble spécifique du langage (TSL) a montré des similarités entre les performances morphosyntaxiques des deux groupes. Cette étude vise à mieux comprendre l’acquisition L2 enfantine typique afin de permettre l’identification d’un TSL chez les enfants L2. Une étude longitudinale de taux de propositions enchâssées et d’erreurs ont été a été faite à partir d’échantillons de langage spontané en français de 22 enfants anglophones âgés de 6;9-12;7 et de 19 enfants monolingues francophones avec TSL âgés de 6;5 à 12;11. Les résultats révèlent que les L2 font plus d’enchâssées que les enfants avec SLI, mais les deux groupes ont eu des taux d’erreurs comparables. La discussion porte sur l’implication de ces résultats sur l’identification des TSL chez les enfants L2. / Research on children acquiring a second language (L2 children) and children with SLI has revealed similarities in their morphosyntactic performance. This study seeks to better understand how children acquire an L2 in order to distinguish typical from atypical L2 acquisition. A longitudinal study of clausal embedding and error rates was made based on spontaneous language samples of 22 English-speaking children (aged 6;9-12;7) acquiring French as an L2 and 19 monolingual French-speaking children with SLI (aged 6;5-12;11). The results revealed that L2 children used clausal embedding more often than the children with SLI, but the SLI and L2 had similar error rates. Analyses of the frequency of embedding and errors revealed correlations that were positive in the L2 group and negative in the SLI. The discussion focuses on the implications of these findings for the identification of SLI in L2 children.

Previous accounts of preverbal subjects in Spanish and European Portuguese (EP) in the literature have debated the syntactic position of these elements. According to some analyses, preverbal subjects are canonical arguments appearing in an A-position (e.g. Goodall 2001, 2002; Suñer 2003 for Spanish; Duarte 1997; Costa 2004 for EP). Other analyses propose that preverbal subjects are non-arguments appearing in a left-peripheral - perhaps CLLD - A'-position (e.g. Uribe-Etxebarria 1990, 1995; Ordóñez & Treviño 1999 for Spanish; Barbosa 1996, 2000 for EP). Although Galician is an ideal language for insight on this debate due to linguistic ties with EP and political ties with Spain, Gupton (2006) obtained inconclusive results regarding the status of preverbal subjects in Galician. As the literature on Galician lacks descriptions of preferred word orders according to discourse context, I collected quantitative and qualitative experimental data to describe the syntax-information structure interface in Galician. The vast majority of speakers of this minority language are Spanish-Galician bilinguals with (self-reported) high levels of competency in both languages. This is of relevance because a variety of bilinguals, including heritage speakers, attrited L1 speakers, and those who have been claimed to have incompletely acquired the heritage language have been shown to exhibit instability and optionality at the linguistic interfaces, in particular at the syntax-discourse pragmatics interface (e.g. Hulk & Müller 2000; Sorace 2005 among numerous others), which is the subject of investigation in this dissertation. The data collected indicate a marked preference for SVO in a wide variety of discourse contexts, a preference that differs from those claimed to apply in similar contexts in Spanish (e.g. Ordóñez 1997, Zubizarreta 1998, Casielles 2004). Assuming that the presence of clitics implies the projection of f (Raposo & Uriagereka 2005) and the extension of the preverbal field into the left periphery, the cliticization data gathered for Galician in main clauses, subordinate clauses and recomplementation contexts suggest a number of preverbal positions in which preverbal subjects, affective phrases, and Topic elements may appear, one of which I suggest is Spec, DoubledFceP, following Martín-González (2002), but with proposed modifications. The data also suggest necessary modifications for López's (2009) syntax-information structure interface proposal in Romance, which suggests a reduced, syncretic left-peripheral position (Spec, FinP) in which CLLD Topics, wh- elements, and Fronted Focus elements appear and are assigned [+c] (contrastive) by the Pragmatics module. Within the preverbal architecture I propose, preverbal subjects and other left-peripheral elements coincide, but in a variety of syntactic positions. Therefore, for pragmatic feature assignment to successfully assign [+c], Pragmatics must distinguish between preverbal subjects and other left-peripheral phrases.

This thesis investigates the use of local search methods in solving constraint problems. Such problems are very hard in general and local search offers a useful and successful alternative to existing techniques. The focus of the thesis is to analyze the techniques of invariants used in local search. The use of invariants have recently become the cornerstone of local search technology as they provide a declarative way to specify incremental algorithms. We have produced a series of program libraries in C++ known as the One-Way-Solver. The One-Way-Solver includes the implementation of incremental data structures and is a useful tool for the implementation of local search. The One-Way-Solver is applied to two challenging constraint problems, the Boolean Satisfiability Testing (SAT) and university course timetabling problems.

We introduce a special set of relations on a finite set, called clausal relations. A restricted version of the Galois connection between polymorphisms and invariants, called Pol-CInv, is studied, where the invariant relations are clausal relations. Clones arising from this Galois connection, so-called C-clones, are investigated. Finally, we show that clausal relations meet a sufficient condition that is known to ensure polynomial time solvability of the corresponding CSP.

The present study replicated the research framework of a previous study (Biber, Gray, & Poonpon, 2011) that identifies the grammatical complexity of L1 professional academic prose as strongly favoring a dense use of phrasal nominal modifiers such as prepositional phrases as postmodifiers, attributive adjectives, and nouns as premodifiers which characterize its unique structurally compressed discourse style. The main purpose of the present study was to explore syntactic similarities and differences between L1 professional and L2 student academic writing in terms of their reliance on phrasal/nominal compression features to determine characteristics of the grammatical complexity of advanced ESL academic writing. To this end, the distributional patterns of use for 25 specific grammatical complexity features of structural elaboration and compression were investigated in a corpus of 128 short academic essays collected from 16 advanced ESL learners and 16 L1 university students (as comparison data).The results showed a heavier reliance of both the advanced ESL and L1 student academic writing on phrasal nominal modifiers (attributive adjectives and prepositional phrases as postmodifiers) of structural compression than on clausal elaboration features, which lent empirical support to Biber, Gray, and Poonpon’s (2011) findings. In addition to the phrasal compression features, both the advanced ESL and L1 student academic writing were also characterized by a prominent use of specific colloquial grammatical devices such as adverbs as adverbials. Compared to the advanced ESL writing, the L1 student academic writing showed a significantly more preference for one particular colloquial feature: ZERO relative clauses where relative pronouns replacing relativized objects are omitted. This combined reliance on both phrasal compression devices and colloquial features in both the advanced ESL and L1 student academic writing distinguished their grammatical complexities from that of L1 professional academic prose and signaled a possibility for recognizing them as a transitional developmental stage from more casual to more academic writing.

A restricted version of the Galois connection between polymorphisms and invariants, called Pol−CInv, is studied, where the invariant relations are restricted to so-called clausal relations. In this context, the relationship of maximal C-clones and maximal clones is investigated. It is shown that, with the exception of one special case occurring for Boolean domains, maximal C-clones are never maximal clones. / Wir untersuchen eine eingeschränkte Variante der Galoisverbindung zwischen Polymorphismen und invarianten Relationen, bezeichnet mit Pol−CInv, wobei die invarianten Relationen auf sogenannte klausale Relationen beschränkt werden. In diesem Zusammenhang wird die Beziehung zwischen maximalen C-Klonen und maximalen Klonen betrachtet. Es wird gezeigt, daß, mit Ausnahme eines Spezialfalles für Boolesche Grundmengen, maximale C-Klone niemals maximale Klone sind.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-05-11T20:09:12Z No. of bitstreams: 1 marcelazambolimdemoura.pdf: 1517569 bytes, checksum: c9a6836aa681e51ba066ec04a14198c9 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-05-17T15:23:43Z (GMT) No. of bitstreams: 1 marcelazambolimdemoura.pdf: 1517569 bytes, checksum: c9a6836aa681e51ba066ec04a14198c9 (MD5) / Made available in DSpace on 2017-05-17T15:23:43Z (GMT). No. of bitstreams: 1 marcelazambolimdemoura.pdf: 1517569 bytes, checksum: c9a6836aa681e51ba066ec04a14198c9 (MD5) Previous issue date: 2017-03-20 / Nesta tese, dedicamo-nos a investigar orações matrizes, compostas por verbo ser e predicativo, e suas orações encaixadas, que funcionam sintaticamente como sujeito, sob a luz da abordagem construcional da mudança, proposta por Traugott (2008a, 2008b) e Traugott e Trousdale (2013). Objetivamos compreender como o complexo oracional subjetivo composto por verbo ser e predicativo na matriz se organiza e se desenvolve na língua portuguesa. Nesse cenário, intentamos, mais especificamente: (a) identificar níveis esquemáticos, a saber, macroconstrução, mesoconstrução, microconstrução e construto (TRAUGOTT 2008a, 2008b); (b) descrever, pontualmente, as microconstruções do complexo oracional subjetivo na língua portuguesa; (c) propor uma rede construcional para o complexo oracional subjetivo composto por verbo ser e predicativo na matriz. Trabalhamos com ocorrências do século XIII ao século XXI, em uma abordagem pancrônica (NEVES, 1997), através de um estudo sincrônico com comprovação diacrônica. Os corpora sincrônicos são constituídos por textos da modalidade escrita e oral da língua. A modalidade escrita é composta por textos retirados de blogs e de revistas online de circulação nacional, a saber, "Revista Veja", "Revista Isto é", "Revista Época", "Revista Caras", "Revista Cláudia" e "Revista Ana Maria". A oralidade é composta por entrevistas selecionadas em três corpora distintos: " Projeto Mineirês: a construção de um dialeto", projeto "PEUL – Programa de Estudos sobre o Uso da Língua" e projeto "NURC/RJ – Projeto da Norma Urbana Oral Culta do Rio de Janeiro". Os dados diacrônicos foram coletados do projeto " CIPM – Corpus Informatizado do Português Medieval" e projeto "TychoBrahe". A partir da análise de dados, identificamos 9 padrões microconstrucionais do complexo oracional subjetivo composto por verbo ser e predicativo na matriz, que apresentam características funcionais e formais específicas e, ao mesmo tempo, similaridades que permitem agrupá-las em níveis construcionais intermediários. Além disso, identificamos um nível hierárquico, altamente esquemático, que abarca todas as construções observadas. Nesse sentido, a análise realizada, conduzida por uma metodologia qualitativa e quantitativa, nos permite defender que o complexo oracional subjetivo composto por verbo ser e predicativo na matriz apresenta padrões construcionais que atendem a propósitos comunicativos distintos. / In this thesis, based on the constructional approach to language change, proposed by Traugott (2008a, 2008b) and Traugott and Trousdale (2013), we are dedicated to investigate main clauses, composed of the verb to be and predicative, and its embedded clauses, which work syntactically as a subject. We aim to understand how the subjective clausal complex consisting of verb be and predicate in the matrix is organized and developed in the Portuguese language. In this scenario, we undertake, specifically: (a) to identify schematic levels, namely macroconstruction, mesoconstruction, microconstruction and construct (TRAUGOTT 2008a, 2008b); (b) to describe, accurately, the microconstructions of the subjective sentence complex in the Portuguese language; (c) propose a constructional network for the subjective sentence complex consisting of the verb be and predicative in the matrix in the Portuguese language. We work with data from the thirteenth century to the twenty-first century, in a panchronic approach (NEVES, 1997), by means of a synchronic study with diachronic evidence. The synchronic corpora are constituted by texts of the written and oral modality of the language. The written modality consists of texts taken from blogs and online magazines of national circulation, namely "Revista Veja", "Revista Isto é", "Revista Época", "Revista Caras", "Revista Cláudia" and "Revista Ana Maria". The orality is composed of interviews selected from three distinct corpora: "Projeto Mineirês: a construção de um dialeto", project "PEUL - Programa de Estudos sobre o Uso da Língua" and project "NURC / RJ - Projeto da Norma Urbana Oral Culta do Rio de Janeiro". The diachronic data were collected from the project "CIPM - Corpus Informatizado do Português Medieval" and project "Tycho Brahe". From the data analysis, we identified 9 microconstructional patterns of the subjective sentence complex composed by the verb to be and predicate in the matrix, which present specific functional and formal characteristics and, at the same time, similarities that allow them to be grouped in intermediate constructional levels. In addition, we have identified a hierarchical, highly schematic level that encompasses all observed constructs. In this sense, the accomplished analysis, conducted by a qualitative and quantitative methodology, allows us to defend that the subjective sentence complex composed by the verb to be and predicative in the matrix presents constructional patterns that serve different communicative purposes.

The thesis comprises an investigation of three aspects of sentence structure in Classical Greek (henceforth CG) dramatic poetry: order of the main sentence elements (subject, verb, and object) within the clause, the emphatic position at the start of the clause, and the structure of inter-clausal linking. It is argued that these three features, usually considered separately, are interdependent, and that intra-clausal word order is directly related to the structure of compound and complex sentences. The discussion undertakes a systematic survey of subject, verb, and object order in a corpus of texts, proposes an explanation for the observed order, and develops a model which explains how prominence within the clause is exploited in clause linking to produce the complement structures observed in Homeric and tragic complementation.

We introduce a special set of relations on a finite set, called clausal relations. A restricted version of the Galois connection between polymorphisms and invariants, called Pol-CInv, is studied, where the invariant relations are clausal relations. Clones arising from this Galois connection, so-called C-clones, are investigated. Finally, we show that clausal relations meet a sufficient condition that is known to ensure polynomial time solvability of the corresponding CSP.

A restricted version of the Galois connection between polymorphisms and invariants, called Pol−CInv, is studied, where the invariant relations are restricted to so-called clausal relations. In this context, the relationship of maximal C-clones and maximal clones is investigated. It is shown that, with the exception of one special case occurring for Boolean domains, maximal C-clones are never maximal clones.:1 Introduction 2 Preliminaries 3 Proof of the main theorem 3.1 Principle of proof 3.2 Bounded orders 3.3 Non-trivial congruences 3.4 Selfdual functions 3.5 Quasilinear functions 3.6 Functions preserving central and h-regular relations 4 Concluding remarks References / Wir untersuchen eine eingeschränkte Variante der Galoisverbindung zwischen Polymorphismen und invarianten Relationen, bezeichnet mit Pol−CInv, wobei die invarianten Relationen auf sogenannte klausale Relationen beschränkt werden. In diesem Zusammenhang wird die Beziehung zwischen maximalen C-Klonen und maximalen Klonen betrachtet. Es wird gezeigt, daß, mit Ausnahme eines Spezialfalles für Boolesche Grundmengen, maximale C-Klone niemals maximale Klone sind.:1 Introduction 2 Preliminaries 3 Proof of the main theorem 3.1 Principle of proof 3.2 Bounded orders 3.3 Non-trivial congruences 3.4 Selfdual functions 3.5 Quasilinear functions 3.6 Functions preserving central and h-regular relations 4 Concluding remarks References