The present diploma thesis focuses on the unconscious code-switching between second and third language. The main objective of this thesis is to deal with principles of unconscious code-switching, describe and categorize actual appearances of unconscious code-switching between second and third language. After introducing issues, the second chapter sums up the way of second and third language acquisition in terms of bilingualism and multilingualism. The third chapter presents a definition and theoretic bases of code-switching. The fourth chapter concerns the analysis of the reason for code- switching in terms of usage-based and psycholinguistic approaches. The fifth chapter presents a categorization of code-switching from syntactic and pragmatic perspectives. The sixth chapter concerns the analysis of cases of code-switching from recorded interviews and the explanation of possible factors causing unconscious code-switching. Key words: Unconscious code-switching, bilingualism, multilingualism, usage-based approach, psycholinguistic approach, function words, acquisition, second language, third language, WIPP

This study tested the unique checking constraint hypothesis and the usage-based account concerning why young children produced tense and agreement morphemes variably via three experiments. Experiment 1 investigated whether subject types influenced the production accuracy of auxiliary 'is' in three-year-olds through an elicited production task. The rate of use of auxiliary 'is' increased as children's tense productivity increased, but the pattern was different for each subject type. The rate of use increased more rapidly with tense productivity for lexical NP subjects than it did for pronominal subjects. Experiment 2 further examined the role of subject types, predicate types, and predicate word frequency on the use of copula 'is' in three-year-olds via an elicited production task. Overall, the production accuracy of copula 'is' was higher with nominal predicates than with permanent- or temporary-adjectival predicates, followed by locative predicates. Children also produced copula 'is' more accurately with low-frequency predicate words than with high-frequency predicate words. Moreover, the effect of subject types on the use of copula 'is' varied with children's tense productivity. For sentences with nominal, permanent-adjectival, or temporary-adjectival predicates, children with lower tense productivity used copula 'is' more accurately with lexical subjects than with pronominal subjects in. In contrast, children with higher tense productivity produced copula 'is' more accurately with pronominal subjects than with lexical subjects. Experiment 3 extended Experiment 1 by exploring the degree of abstractness of representations of auxiliary BE via a structural priming task. The production accuracy of auxiliary 'is' in three-year-olds increased above the baseline when the prime-target pair shared the same structure and subject + auxiliary 'is' combinations, but not when the prime-target pair only shared the same structure. However, the production accuracy of auxiliary 'are' did not change with prime types. These experiments suggest that young children have only lexically-specific representations of auxiliary BE. Frequency, rather than structural properties, of sentence elements influenced the production accuracy of auxiliary and copula 'is' in young children. These findings support the usage-based approach that young children use tense and agreement morphemes variably because they have not yet learned adult-like abstract representations and use highly frequent/ lexically-specific constructions for the production of these morphemes.

The work in this thesis presents and evaluates a number of strategies by which English-learning children might discover the major open-class parts-of-speech in English (nouns, verbs and adjectives) on the basis of purely distributional information. Previous work has shown that parts-of-speech can be readily induced from the distributional patterns in which words occur. The research reported in this thesis extends and improves on this previous work in two major ways, related to the constructional status of the utterance contexts used for distributional analysis, and to the way in which previous studies have dealt with categorial ambiguity. Previous studies that have induced parts-of-speech from word distributions have done so on the basis of fixed “windows” of words that occur before and after the word in focus. These contexts are often not constructions of the language in question, and hence have dubious status as elements of linguistic knowledge. A great deal of recent evidence (e.g. Lieven, Pine & Baldwin, 1997; Tomasello, 1992) has suggested that children’s early language may be organized around a number of lexically-specific constructional frames with slots, such as “a X”, “you X it”, “draw X on X”. The work presented here investigates the possibility that constructions such as these may be a more appropriate domain for the distributional induction of parts-of-speech. This would open up the possibility of a treatment of part-of-speech induction that is more closely integrated with the acquisition of syntax. Three strategies to discover lexically-specific frames in the speech input to children are presented. Two of these strategies are based on the interplay between more and less frequent words in English utterances: the more frequent words, which are typically function words or light verbs, are taken to provide the schematic “backbone” of an utterance. The third strategy is based around pairs of words in which the occurrence of one word is highly predictable from that of the other, but not vice versa; from these basic slot-filler relationships, larger frames are assembled. These techniques were implemented computationally and applied to a corpus of child-directed speech. Each technique yielded a large set of lexically-specific frames, many of which could plausibly be regarded as constructions. In a comparison with a manual analysis of the same corpus by Cameron-Faulkner, Lieven and Tomasello (2003), it is shown that most of the constructional frames identified in the manual analysis were also produced by the automatic techniques. After the identification of potential constructional frames, parts-of-speech were formed from the patterns of co-occurrence of words in particular constructions, by means of hierarchical clustering. The resulting clusters produced are shown to be quite similar to the major English parts-of-speech of nouns, verbs and adjectives. Each individual word token was assigned a part-of-speech on the basis of its constructional context. This categorization was evaluated empirically against the part-of-speech assigned to the word in question in the original corpus. The resulting categorization is shown to be, to a great extent, in agreement with the manual categorization. These strategies deal with the categorial ambiguity of words, by allowing the frame context to determine part-of-speech. However, many of the frames produced were themselves ambiguous cues to part-of-speech. For this reason, strategies are presented to deal with both word and context ambiguity. Three such strategies are proposed. One considers membership of a part-of-speech to be a matter of degree for both word and contextual frame. A second strategy attempts to discretely assign multiple parts-of-speech to words and constructions in a way that imposes internal consistency in the corpus. The third strategy attempts to assign only the minimally-required multiple categories to words and constructions so as to provide a parsimonious description of the data. Each of these techniques was implemented and applied to each of the three frame discovery techniques, thereby providing category information about both the frame and the word. The subsequent assignment of parts-of-speech was done by combining word and frame information, and is shown to be far more accurate than the categorization based on frames alone. This approach can be regarded as addressing certain objections against the distributional method that have been raised by Pinker (1979, 1984, 1987). Lastly, a framework for extending this research is outlined that allows semantic information to be incorporated into the process of category induction.

Submitted by Maike Costa (maiksebas@gmail.com) on 2016-07-19T13:59:09Z No. of bitstreams: 1 arquivo total.pdf: 3655935 bytes, checksum: c32c80be0b5d66b04eb2ca7b1e59f308 (MD5) / Made available in DSpace on 2016-07-19T13:59:09Z (GMT). No. of bitstreams: 1 arquivo total.pdf: 3655935 bytes, checksum: c32c80be0b5d66b04eb2ca7b1e59f308 (MD5) Previous issue date: 2015-12-04 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This dissertation points out how the appearances of the neologisms in a language, by lexicalization, can contribute to enrichment and updating of the lexicon of the same language. Therefore, it looked for: (i) expose the main concepts about lexicon, neologism and lexicalization, based on the Usage-Based Linguistics (UBL), (ii) it presents 13 lexical items selected from the digital corpus and (iii) present the real relevance of the lexicalization for the formation of new words, for to understand how this affects/changes the multi-system. The corpus used was the one of the Project AC/DC: corpo Corpus Brasileiro, which has about one billion words employed in the most varied use contexts. For the fundamentation of the dissertation, some scholars were consulted, among them we highlight: Martelotta (2011), Gonçalves (2011), Contiero and Ferraz (2014), Correia and Almeida (2012), Carvalho (2009a), Biderman (1981), Câmara Jr. (2011), Pontes-Ribeiro (2007), Castilho (2003a; 2003b; 2008), Cunha (2011), Mendes and Seabra (2006), Ferraz (2006; 2007) and Fortunato (2008). The methodology consists in three stages: a) select of lexical elements samples in the corpus, b) extraction of this samples and compilations of them in tables and c) analyses of collected data. The results revealed that some of the 13 lexicalized words/neologisms, possibly, appeared to fulfill an existing space of linguistic signs in the multi-system, others acquired new meanings when used in new contexts of use and many others are in process of disappearance. The frequency of use was determining in the change of meaning. / Esta dissertação aponta como o surgimento dos neologismos em uma língua, pela lexicalização, pode contribuir para o enriquecimento e atualização do léxico desta mesma língua. Deste modo, buscou-se: (i) expor os principais conceitos sobre léxico, neologismo e lexicalização, com base na Linguística Centrada no Uso (LCU), (ii) apresentar 13 itens lexicais selecionados a partir do corpus digital e (iii) discutir a relevância da lexicalização para a formação de novas palavras, para entender como isso afeta/altera o multissistema. O corpus utilizado foi o Projeto AC/DC: corpo Corpus Brasileiro, que contém cerca de um bilhão de palavras empregadas nos mais variados contextos de uso. Para a fundamentação da dissertação, alguns estudiosos foram consultados, entre os quais se destacam: Martelotta (2011), Gonçalves (2011), Contiero e Ferraz (2014), Correia e Almeida (2012), Carvalho (2009a), Biderman (1978; 1981), Câmara Jr. (2011), Pontes-Ribeiro (2007), Castilho (2003a; 2003b; 2008), Cunha (2011), Mendes e Seabra (2006), Ferraz (2006; 2007) e Fortunato (2008). A metodologia consistiu em três etapas: a) coleta de amostras de itens lexicais no corpus, b) extração dessas amostras e compilação em tabelas e c) análise dos dados coletados. Os resultados revelaram que alguns dos 13 neologismos/palavras lexicalizadas, possivelmente, surgiram para preencher um vazio de signos linguísticos no multissistema, outros adquiriram novos sentidos ao serem empregados em novos contextos de uso e outros tantos estão em processo de desaparecimento. A frequência de uso foi determinante para a mudança no sentido.

This article investigates the role of direct input in the code-mixing of three bilingual children aged 2–4 years acquiring English as one language, and either German, Polish, or Finnish as the other. From a usage-based perspective, it is assumed that early children’s utterances are item-based and that they contain many lexically fixed patterns. To account for such patterns, the traceback method has been developed to test the hypothesis that children’s utterances are constructed on the basis of a limited inventory of chunks and frame-and-slot patterns. We apply this method to the code-mixed utterances, suggesting that much of the code-mixing occurs within frame-and-slot patterns, such as Was ist X? as in Was ist breakfast muesli? “What is breakfast muesli?” We further analyzed each code-mixed utterance in terms of priming. Our findings suggest that much of the early code-mixing is based on concrete lexically fixed patterns which are subject to input occurring in immediately prior speech, either the child’s own or that of her caregivers.

This paper offers an inductive, exploratory study on the role of input and individual differences in the early code-mixing of bilingual children. Drawing on data from two German-English bilingual children, aged 2–4, we use the traceback method to check whether their code-mixed utterances can be accounted for with the help of constructional patterns that can be found in their monolingual data and/or in their caregivers’ input. In addition, we apply the tracebackmethod to checkwhether the patterns used by one child can also be found in the input of the other child. Results show that patterns found in the code-mixed utterances could be traced back to the input the children receive, suggesting that children extract lexical knowledge from their environment. Additionally, tracing back patterns within each child was more successful than tracing back to the other child’s corpus, indicating that each child has their own set of patterns which depends verymuch on their individual input. As such, these findings can shed new light on the interplay of the two developing grammars in bilingual children and their individual differences.

This study is part of the IN.MIGRA-2 CM project, which studies the sociolinguistic integration of the immigrant population of Madrid. The present study focuses on the language attitudes of 16 residents of Alcalá de Henares, a community of Madrid. The participants were asked to rate their level of agreement with the following affirmations: (1) The Spanish of Madrid is more correct than the forms of speech of Latin American immigrants; (2) Mastery of the Spanish language is the principal demonstration of the integration of immigrants; (3) Immigrants of Latin American origin are integrated because they speak the same language. The study finds a connection between higher levels of contact with immigrants and lower ratings of agreement with the affirmations. This investigation supports others that show connections between social networks and language attitudes, and it adds to the sparse research on language attitudes in Madrid. Adverbial adjectives modify both a verb and the subject of that verb. Their purpose is to describe a quality that pertains to both the subject and the way the subject is performing the verb. Because they modify both the verb and the noun, adverbial adjectives agree with the noun in number and gender. The generativist approaches to this linguistic phenomenon do not provide a sufficient explanation of verb + adverbial adjective constructions nor do they predict which subjects and predicates that can be used in these constructions. This paper takes a usage-based approach to adverbial adjectives. It explores the token frequencies of use of different verb + adverbial adjective phrases and attempts to categorize the components of these phrases based on these frequencies.

A luz de modelos de língua baseados no uso (Langacker, 2000), que assumem que os princípios fundamentais da estrutura linguística são derivados da experiência com a língua, o objetivo deste estudo é contribuir para a investigação de como se dá o contínuo processo de atualização do nosso conhecimento linguístico, em especial explorando a questão de como algumas inovações são incorporadas na língua. Para responder a essa questão, ainda em aberto na literatura baseada no uso, o fenômeno linguístico investigado foi o infinitivo flexionado, cujo uso gramáticos modernos consideram intrigante. A partir de um recorte de construções com o infinitivo que apresentam variação entre as formas flexionada e não-flexionada, foi feita uma quantificaçãoo da variação no emprego da flexão do infinitivo em um corpus sincrônico de língua escrita culta. Para o levantamento dos dados, foi compilado um corpus de 11.000.000 palavras formado por 180 teses e dissertações de alunos do curso de Letras da Faculdade de Filosofia, Letras e Ciências Sociais da USP, que recebeu o nome de Corpus LLIC-PósLetrasUsp. Como se trata de um grande volume de dados, a extração dos dados feita de modo automático com o software livre R, apontado atualmente como o mais completo concordanceador disponível (Gries, 2009). Os resultados encontrados no estudo de corpus apontam para diferentes tendências de emprego da flexão a depender dos contextos sintáticos em que ocorrem. Apesar de ter sido verificada uma tendência à não-flexão em algumas construções, tais como as perífrases modais e aspectuais, nas orações finais, causais e temporais, por exemplo, a preferência pela forma flexionada do infinitivo é clara. Uma vez que modelos baseados no uso propõem que há uma correlação entre a frequência de ocorrência de estruturas e seu grau de arraigamento cognitivo, a alta frequência de ocorrência de formas flexionadas do infinitivo em variados contextos sintáticos, inclusive em contextos em que o sujeito do infinitivo está claro, demonstra que a flexão do infinitivo está se tornando cada vez mais arraigada na gramática dos falantes. Assim, esta pesquisa vem ressaltar a importância dos estudos de frequência para o mapeamento dos fatos linguísticos e para a descrição gramatical das línguas naturais. Neste estudo, também é feito um esboço de um modelo de língua baseado no uso mais dinâmico e mais social, que incorpora as propostas dos estudos da terceira onda\" (third wave) de investigação sociolinguística e que representa uma tentativa inicial de acomodar a teoria da variação dentro do quadro teórico baseado no uso. / In light of usage-based models (Langacker, 2000), which assume that the fundamental principles of linguistic structure derive from the experience with language, the aim of this study is to contribute to the investigation of the continuous update of linguistic knowledge, especially by exploring the question of how some innovations propagate and are incorporated in speakers\' grammars. To address this open question in the usage-based literature, the in ected innitive, which modern grammarians consider intriguing (Cunha & Cintra, 2008), was investigated. After selecting some innitive constructions which show variation between the in ected and unin ected forms, the variation in the usage of the innitive in these constructions was quantied in a syncronic standard written corpus. To obtain the data, a corpus containing 11.000.000 words and consisting of 180 theses and dissertations written by students of the Faculty of Philosophy, Letters and Social Sciences of University of S~ao Paulo was compiled and named Corpus LLIC-PosLetrasUsp. Given the size of the corpus, the data extraction was achieved automatically through the use of the open software R, considered the most powerful concordancer available at the moment (Gries, 2009). Depending on the constructions in which the innitive occurs, the results point to dierent tendencies in the usage of the in ection. Even though there is no preference for the in ection in constructions such as modal and aspect periphrases, in causal, nal and temporal clauses, for instance, the preference for in ected forms is clear. Because usage-based models assume that there is a correlation between the frequency of occurence of a structure and its degree of entrenchment, the high frequency of occurence of the in ected innitive in dierent syntactic contexts, including contexts in which where the innitive subject is clear, shows that the in ection of the innitive is becoming continuously more entrenched in speakers\' grammars. Thus, this study highlights the importance of frequency studies in the investigation of linguistic phenomena and in the grammatical description of natural languages. This study also presents the sketch of a proposal named A more dynamic and a more social usage-based model, which, by incorporating the view of the \\third wave\"of sociolinguistic investigation (Campbell-Kibler, 2008, 2009; Eckert, 2005, 2008), represents an initial attempt to accommodate variation theory within the usage-based framework.

Submitted by Aline Batista (alinehb.ufpel@gmail.com) on 2017-06-08T15:25:06Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Dissertação - Fernanda Peres Lopes.pdf: 3142979 bytes, checksum: 5720a00dae8803fe8625b52b9c76c001 (MD5) / Approved for entry into archive by Aline Batista (alinehb.ufpel@gmail.com) on 2017-06-09T14:37:22Z (GMT) No. of bitstreams: 2 Dissertação - Fernanda Peres Lopes.pdf: 3142979 bytes, checksum: 5720a00dae8803fe8625b52b9c76c001 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-06-09T14:37:29Z (GMT). No. of bitstreams: 2 Dissertação - Fernanda Peres Lopes.pdf: 3142979 bytes, checksum: 5720a00dae8803fe8625b52b9c76c001 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-02-24 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Este estudo tem por objetivo analisar o processo variável de apagamento das vogais átonas [a, i, u] em posição final na fala de indivíduos da cidade de Pelotas (RS), conforme verificado em tarif[a] ~ tarif∅, equip[e] ~ equip∅, serviço ~ serviç∅, por exemplo. Para isso, partiu-se dos pressupostos da Fonologia de Uso (BYBEE, 2001, 2006, 2010), da Teoria de Exemplares (PIERREHUMBERT, 2001, 2003) e da Sociofonética (THOMAS, 2011; FOULKES; SCOBBIE; WATT, 2010). A amostra sob análise é constituída por oito informantes (quatro homens e quatro mulheres) entre 18 e 50 anos de idade e de dois níveis de escolaridade – sujeitos com até seis anos de escolaridade e sujeitos com, no mínimo, nove anos de escolaridade. A taxa de aplicação do apagamento na amostra analisada foi de 53% (N = 242) para a vogal [i], de 41% (N = 196) para a vogal [u] e de 0,8% (N = 4) para a vogal [a]. Os resultados indicaram que fatores como tipo de vogal, contexto precedente, ordem de produção e frequência lexical favorecem o apagamento, revelando a natureza predominantemente linguística do fenômeno. A única variável extralinguística que teve influência no apagamento foi a variável indivíduo. A análise acústica revelou que as vogais postônicas [i] e [u] produzidas pelos pelotenses tendem a abaixar enquanto a vogal [a] tende a elevar-se. Além disso, percebe-se uma centralização de [u], que ocupa, entre os homens, quase o mesmo espaço acústico que a vogal [a]. [ɐ], [ɪ] e [ʊ] são as vogais que representam o sistema encontrado na amostra. Com relação à duração, comparando-se os valores encontrados com os dados de Quintanilha-Azevedo (2016), percebe-se que tanto homens quanto mulheres produziram vogais mais curtas. Por fim, conclui-se que o apagamento representa o ponto final de uma trajetória que se inicia com a realização plena da vogal, passa pela redução de sua duração e pelo seu desvozeamento. / This thesis aims to analyze the variable process of deletion of the final unstressed vowels [a, i, u] by Brazilian Portuguese native speakers from the city of Pelotas, in the Southern state of Rio Grande do Sul, as noticed in words such as tarif[a] ~ tarif∅, equip[e] ~ equip∅, serviço ~ serviç∅, for instance. In order to achieve said aim, the concepts addressed by Usage-Based Phonology (BYBEE, 2001, 2006, 2010), the Exemplar Theory (PIERREHUMBERT, 2001, 2003) and Sociophonetics (THOMAS, 2011; FOULKES; SCOBBIE; WATT, 2010) were used as this research project's theoretical bases. The sample analyzed in this study consists of 8 informants (4 men and 4 women) from two different educational backgrounds (one group including subjects with up to 6 years of formal learning and the other including subjects with 9+ years of formal learning) with ages ranging from eighteen to fifty. The rate of deletion in the sample was of 53% (N = 242) for the vowel [i], of 41% (N = 196) for the vowel [u], and of 0,8% (N = 4) for the vowel [a]. The results indicate that such factors as type of vowel, preceding context, vowel production order and lexical frequency favor deletion, revealing the predominantly linguistic nature of the phenomenon. The variable "subject" was the only extralinguistic variable to influence deletion. Acoustic analysis showed that the posttonic vowels [i] and [u] produced by the subjects tend to lower as the vowel [a] tends to rise. In addition, the study also revealed a centralization of [u], which occupies among men almost the same acoustic vowel space as does the vowel [a]. Vowels [ɐ], [ɪ] and [ʊ] are representative of the system found in the sample. Regarding duration, when comparing the values included in this research project with those presented by Quintanilha-Azevedo (2016), it can be noticed that both men and women produced shorter vowels. Lastly, it was found that deletion represents the final stage of a process that begins with the production of a full-quality vowel, moves on to its reduction, and ultimately reaches the devoicing of said vowel.

A luz de modelos de língua baseados no uso (Langacker, 2000), que assumem que os princípios fundamentais da estrutura linguística são derivados da experiência com a língua, o objetivo deste estudo é contribuir para a investigação de como se dá o contínuo processo de atualização do nosso conhecimento linguístico, em especial explorando a questão de como algumas inovações são incorporadas na língua. Para responder a essa questão, ainda em aberto na literatura baseada no uso, o fenômeno linguístico investigado foi o infinitivo flexionado, cujo uso gramáticos modernos consideram intrigante. A partir de um recorte de construções com o infinitivo que apresentam variação entre as formas flexionada e não-flexionada, foi feita uma quantificaçãoo da variação no emprego da flexão do infinitivo em um corpus sincrônico de língua escrita culta. Para o levantamento dos dados, foi compilado um corpus de 11.000.000 palavras formado por 180 teses e dissertações de alunos do curso de Letras da Faculdade de Filosofia, Letras e Ciências Sociais da USP, que recebeu o nome de Corpus LLIC-PósLetrasUsp. Como se trata de um grande volume de dados, a extração dos dados feita de modo automático com o software livre R, apontado atualmente como o mais completo concordanceador disponível (Gries, 2009). Os resultados encontrados no estudo de corpus apontam para diferentes tendências de emprego da flexão a depender dos contextos sintáticos em que ocorrem. Apesar de ter sido verificada uma tendência à não-flexão em algumas construções, tais como as perífrases modais e aspectuais, nas orações finais, causais e temporais, por exemplo, a preferência pela forma flexionada do infinitivo é clara. Uma vez que modelos baseados no uso propõem que há uma correlação entre a frequência de ocorrência de estruturas e seu grau de arraigamento cognitivo, a alta frequência de ocorrência de formas flexionadas do infinitivo em variados contextos sintáticos, inclusive em contextos em que o sujeito do infinitivo está claro, demonstra que a flexão do infinitivo está se tornando cada vez mais arraigada na gramática dos falantes. Assim, esta pesquisa vem ressaltar a importância dos estudos de frequência para o mapeamento dos fatos linguísticos e para a descrição gramatical das línguas naturais. Neste estudo, também é feito um esboço de um modelo de língua baseado no uso mais dinâmico e mais social, que incorpora as propostas dos estudos da terceira onda\" (third wave) de investigação sociolinguística e que representa uma tentativa inicial de acomodar a teoria da variação dentro do quadro teórico baseado no uso. / In light of usage-based models (Langacker, 2000), which assume that the fundamental principles of linguistic structure derive from the experience with language, the aim of this study is to contribute to the investigation of the continuous update of linguistic knowledge, especially by exploring the question of how some innovations propagate and are incorporated in speakers\' grammars. To address this open question in the usage-based literature, the in ected innitive, which modern grammarians consider intriguing (Cunha & Cintra, 2008), was investigated. After selecting some innitive constructions which show variation between the in ected and unin ected forms, the variation in the usage of the innitive in these constructions was quantied in a syncronic standard written corpus. To obtain the data, a corpus containing 11.000.000 words and consisting of 180 theses and dissertations written by students of the Faculty of Philosophy, Letters and Social Sciences of University of S~ao Paulo was compiled and named Corpus LLIC-PosLetrasUsp. Given the size of the corpus, the data extraction was achieved automatically through the use of the open software R, considered the most powerful concordancer available at the moment (Gries, 2009). Depending on the constructions in which the innitive occurs, the results point to dierent tendencies in the usage of the in ection. Even though there is no preference for the in ection in constructions such as modal and aspect periphrases, in causal, nal and temporal clauses, for instance, the preference for in ected forms is clear. Because usage-based models assume that there is a correlation between the frequency of occurence of a structure and its degree of entrenchment, the high frequency of occurence of the in ected innitive in dierent syntactic contexts, including contexts in which where the innitive subject is clear, shows that the in ection of the innitive is becoming continuously more entrenched in speakers\' grammars. Thus, this study highlights the importance of frequency studies in the investigation of linguistic phenomena and in the grammatical description of natural languages. This study also presents the sketch of a proposal named A more dynamic and a more social usage-based model, which, by incorporating the view of the \\third wave\"of sociolinguistic investigation (Campbell-Kibler, 2008, 2009; Eckert, 2005, 2008), represents an initial attempt to accommodate variation theory within the usage-based framework.