Papers by Mônica Rigo Ayres
Revista Linguística, 2022
Null subjects have been extensively investigated in natural languages. Brazilian Portuguese (BP),... more Null subjects have been extensively investigated in natural languages. Brazilian Portuguese (BP), in particular, raises a lot of interest in the subject, for it is apparently in a process of grammatical change, from a pro-drop language to a partial pro-drop language (cf. BIBERAUER et al., 2010; HOLMBERG et al., 2009; DUARTE; MARINS, 2021; SOARES et al., 2019). In this article, we investigate null subjects in BP assuming that null referential subjects are only used in ‘marked’, specific contexts, whereas overt subjects are the ‘unmarked’, more frequent strategy. We have analyzed a contemporary corpus of spoken BP and found 1,252 occurrences of null subjects. We then present four factors that are traditionally known for playing a role in determining the optimal contexts for null subjects in BP (and several other languages). We argue here that all these four factors are both necessary and sufficient to explain all null subject occurrences in any given corpus of contemporary spoken BP. We analyzed all the 1,252 occurrences of null subjects in the corpus and could explain 99.8% of the data, leading us to a better understanding of the null subject phenomenon in BP, at least when it comes to identifying its licensing contexts.
Revista Linguística Rio, 2021
Haag & Othero
(2003) set out to study the
p r o c e s s i n g o f p r o n o m i n a l
anaphoras, ... more Haag & Othero
(2003) set out to study the
p r o c e s s i n g o f p r o n o m i n a l
anaphoras, especially with regard
to the preference for linking the
anaphoric pronoun to the closest
antecedent or topic of the
sentence. To do so, the authors
used a questionnaire that was
applied to 30 informants. In the
questionnaire applied by Haag and
Othero, in each target sentence
t h e r e w e r e t w o p o s s i b l e
antecedents, a topic and another
one closer to the anaphora.
However, the anaphora was
already filled in by a pronoun, but
it could also happen through an
empty category, the null subject. It
is in this context that the idea of
investigating the same sentences
arose, but without inserting the
anaphoric pronoun, to test the
speakers' preference for completing
the sentence using the pronominal
anaphoric or null form. In this
study, our objectives are (i) to test
the speakers' preference for
completing the sentence using the
pronominal or null form; (ii) check
whether there is a pattern or a
relationship in the preference for
linking this anaphora (pronominal
or null) to one of the possible
antecedents (topical or closest
referent); and, (iii) verifying if
antecedent features influence the
choice for overt or null subject.
Regarding objective (i), we found a
greater number of pronominal
anaphoras; about the objective (ii),
o u r d a t a d i d n o t s h o w a
relationship between the factors;
regarding objective (iii), we present
trends that should be further
investigated.
Cadernos do Instituto de Letras | Estudos Linguísticos, 2021
In this article, we approach the null subject phenomenon in Brazilian Portuguese (BP) through occ... more In this article, we approach the null subject phenomenon in Brazilian Portuguese (BP) through occurrences arising from a written language corpus, composed of journalistic texts. Thus, we seek the hypothesis that the null subject in BP might be explained through a multifactorial analysis that takes into account: (i) verbal morphology, (ii) the semantic feature [+/- semantic gender], (iii) the linear sentential pattern V2 e (iv) the optimal connection discursive factor. This hypothesis is initially proposed by Ayres (2021) when considering a corpus of transcribed spoken language. In our study, we verified that the union of different factors is satisfactory to describe the null subjects of BP found in the corpus of analysis.
A conexão discursiva e a manifestação de sujeito pronominal e nulo em português brasileiro, 2020
Investigamos o fenômeno do sujeito pronominal nulo e preenchido por um viés funcional. Assumimos ... more Investigamos o fenômeno do sujeito pronominal nulo e preenchido por um viés funcional. Assumimos a hipótese funcionalista da conexão discursiva ótima e investigamos os contextos em que sujeitos nulos e expressos de 1ª pessoa são favorecidos ou desfavorecidos. Analisamos 1.270 ocorrências de sujeitos de 1ª pessoa em um corpus de língua falada verificando o contexto discursivo em que ocorreram. Fizemos quatro cruzamentos entre os dados (sujeitos nulos em contextos de conexão ótima; sujeitos nulos em contextos de conexão não ótima; sujeitos expressos em contextos de conexão ótima; e sujeitos expressos em contextos de conexão não ótima) e apresentamos nossos resultados.
Neste trabalho, revisitamos o estudo de Ayres e Othero (2016), sobre o condicionamento de objetos... more Neste trabalho, revisitamos o estudo de Ayres e Othero (2016), sobre o condicionamento de objetos nulos e pronomes em PB na retomada anafórica de objeto direto de 3ª pessoa, na fala infantil, a partir das duas hipóteses consolidadas na literatura, a saber: a hipótese do conjunto de traços de animacidade e especificidade (cf. CYRINO, 1994/1997; CASAGRANDE, 2007, por exemplo), e a hipótese do gênero semântico (cf. CREUS; MENUZZI, 2004; OTHERO et al., 2016, por exemplo). Como em seu estudo Ayres e Othero (2016) não distinguiram elipses de VP de objetos nulos, aqui nós separamos esses dois tipos de categoria vazia e as analisamos. Além disso, verificamos se o condicionamento dos objetos nulos e das elipses de VP ocorre em contextos semelhantes, tendo em vista que essas estruturas possuem formas superficialmente parecidas.
_
In this text, we review the study of Ayres and Othero (2016), on the conditioning of null objects and pronouns in BP, concerning anaphoric direct object of third person in children’s speech. They based their analysis on the two hypotheses consolidated in the literature, namely the semantic gender hypothesis (cf. CREUS; MENUZZI, 2004; OTHERO et al., 2016, and others), the animacy and specificity hypothesis (c. CYRINO, 1994/1997; CASAGRANDE, 2007, for example). Since Ayres and Othero (2016) did not distinguish VP ellipses from null objects, here we separate these two types of empty categories and analyze them. In addition, we verify if the conditioning of the null objects and the VP ellipses occurs in similar contexts, considering that these structures have superficially similar forms.
Desde, pelo menos, o século XIX, o uso do clítico acusativo de terceira pessoa (o, a) em portuguê... more Desde, pelo menos, o século XIX, o uso do clítico acusativo de terceira pessoa (o, a) em português brasileiro cedeu espaço para duas estratégias: (a) o uso do pronome tônico ele, ela ou (b) o uso do chamado objeto direto nulo. Partindo da hipótese básica de Creus & Menuzzi (2004) sobre o traço semântico do referente ter papel central para o condicionamento da retomada anafórica com pronome ou com objeto nulo, defendemos a ideia de que existe uma estratégia não marcada e outra marcada em se tratando da retomada anafórica para objetos diretos em 3ª pessoa. Através de reanálises de testes propostos por Creus & Menuzzi e aplicações de novos testes, procuramos mostrar que a estratégia marcada é a utilização de um pronome e a estratégia não marcada, o uso de categoria vazia na posição de objeto, sendo o traço semântico do referente a ser retomado relevante para cada opção.
Palavras-chave: objeto direto; retomada anafórica; objeto nulo; português brasileiro.
At least, since the XIX century, the use of accusative clitic for third person (o, a) in Brazilian Portuguese has yielded to two strategies, (a) the use of tonic pronoun ele, ela or (b) the use of the called nulled direct object. Assuming the Creus & Menuzzi’s (2004) hypothesis and believing that, for the PB grammar, the relevant feature to use null object is the semantic gender of the referent, in this article we will bring the idea that there is an unmarked strategy and other marked strategy in the case of anaphoric recover for direct objects in 3rd person. Through reanalysis of tests proposed by C&M and applications of new tests, we aim to show that the marked strategy is the use of a pronoun and the unmarked strategy is the use of null object, and the semantic-pragmatic features of the referent are relevant to each option.
Keywords: direct object; anaphor; null object; Brazilian Portuguese.
Apresentamos, neste artigo, nosso trabalho de anotação morfológica automática de trechos de um co... more Apresentamos, neste artigo, nosso trabalho de anotação morfológica automática de trechos de um corpus de língua falada – pertencentes ao projeto Varsul –, utilizando um etiquetador automático morfossintático gratuito, o Aelius, em 20 textos, perfazendo um total de 154.530 palavras. Basicamente, apresentamos a ferramenta de anotação automática, o processo de análise morfossintática automática efetuada pelo anotador, o trabalho de revisão manual da etiquetagem automática e as sugestões de melhorias para tratar especificamente de aspectos da oralidade. A partir dos erros do etiquetador, buscamos depreender certos padrões de anotação para superar limitações de desempenho apresentadas pelo programa, propondo algumas sugestões de implementações para que o Aelius etiquete de maneira ainda mais satisfatória um corpus de língua falada. Tratamos especialmente dos casos de interjeições, aféreses, onomatopeias e marcadores conversacionais.
PALAVRAS-CHAVE: etiquetagem Automática; etiquetagem morfossintática; Linguística de Corpus.
In this paper, we present the results of our work on automatic morphological annotation using the free morphosyntatic Aelius tagger. The excerpts analyzed were taken from a corpus of spoken language – part of the VARSUL project. We present 20 texts containing 154,530 words, which were automatically annotated and manually corrected. This paper also describes the Aelius tagger and our effort of manual reviewing the texts, as well as our suggestions for improving the tool, concerning aspects of oral texts tagging. We assessed the performance of the tagger at the morphosyntactic tagging of a spoken language corpus, an unprecedented challenge for the tool. Based on its errors, we try to infer certain patterns of annotation to overcome the limitations presented by the program. We also make suggestions in order to allow Aelius to tag spoken language corpora in a more effective way, specially when dealing with interjections, apheresis, onomatopoeia and conversational markers.
KEYWORDS: tagger; morphosyntactic tagging; Corpus Linguistics.
O quadro pronominal do português brasileiro (PB) vem passando por modificações ao longo do tempo.... more O quadro pronominal do português brasileiro (PB) vem passando por modificações ao longo do tempo. Desde o século XIX, o clítico acusativo de terceira pessoa (o, a) vem perdendo espaço no conjunto de pronomes. Para retomar elementos anafóricos em posição de objeto direto, a gramática do PB fornece duas estratégias no lugar do clítico: o pronome pleno (ele, ela) ou uma categoria vazia. A escolha por uma estratégia ou outra nãó e aleatória; acontece por influência de traços semânticos (e talvez discursivos) do referente anafórico. De acordo com a literatura sobre o assunto, os traços de animacidade e especifici-dade ou de gênero semântico são os que parecem condicionar o uso de pronomes e objetos nulos em PB, e ´ e isso que investigaremos aqui. Nossa hipótese centraí e que apenas uma dessas características do referente seja de fato aquela que condicione o uso do pronome ou do objeto nulo na retomada anafórica: o traço de gênero semântico. Para corroborar essa hipótese, analisamos aqui a fala de crianças entre as idades de 1 a 9 anos, dos corpora do CEAAL (PUCRS) e PEUL (UFRJ).
Palavras-chave: retomada anafórica, objeto nulo, português brasileiro, linguagem infantil.
The pronominal inventory in Brazilian Portuguese (BP) has been undergoing changes over time. Since the nineteenth century, the accusative third person clitic (o, a) has been falling away. BP grammar provides two new strategies to replace the clitic: the full pronoun (ele, ela) or a null element. The choice of one strategy or the other is not random; it is straightforwardly related to semantic (and perhaps discoursive) features of the antecedent. According to the literature on the subject, features of animacy and specificity or semantic genderaretheonesthatconditiontheuseofpronounsornullobjectsinBP.Thisiswhatwe investigatehere. Ourcentralhypothesisisthatonlyoneofthesefeaturesisactuallytheone that triggers the use of a pronoun or the null object in anaphoric resumption, namely the semanticgender. Tosupportthishypothesis,weanalyzethespeechofchildrenbetweenthe ages of 1 to 9 years, extracted from two corpora: CEAAL (PUCRS) and PEUL (UFRJ).
Keywords: anaphora resolution, null object, Brazilian Portuguese, child language.
Book Reviews by Mônica Rigo Ayres
Resenha do livro Gramáticas na escola (2016), de Roberta Pires de Oliveira e Sandra Quarezemin, 2021
O livro “Pronomes pessoais”, publicado em 1994, engloba o vasto trabalho elaborado por José Lemos... more O livro “Pronomes pessoais”, publicado em 1994, engloba o vasto trabalho elaborado por José Lemos Monteiro durante seu período de doutoramento pela Universidade Federal do Rio de Janeiro (UFRJ). A obra é dividida em duas partes: a primeira abrange todos os conceitos e fundamentos relacionados ao uso de pronomes em português, enquanto a segunda apresenta uma descrição abrangente do quadro pronominal do português brasileiro (PB) a partir de 60 inquéritos gravados do corpus NURC, sendo 45 diálogos entre informante e documentador (DID) e 15 elocuções formais, contemplando 5 capitais do Brasil, a saber, Rio de Janeiro, São Paulo, Porto Alegre, Salvador e Recife.
Teaching Documents by Mônica Rigo Ayres
Aspectos condicionadores do objeto nulo e do pronome pleno em português brasileiro : uma análise da fala infantil, 2016
The usage of Brazilian Portuguese pronouns has undergone changes. According to Cyrino
(1994/1997)... more The usage of Brazilian Portuguese pronouns has undergone changes. According to Cyrino
(1994/1997), Nunes (1993) and Bagno (2011) among others), since 19st century, the use of the
accusative clitic for the third person (o, a) has decreased. To recover an anaphoric element in
a direct object position, speakers use two strategies instead of the clitic: the full pronoun (he
‘ele’, she ‘ela’), or the null direct object.
The choice of using the full pronoun or null object is not random, before, this choice is made
by the influence of semantic features (and maybe discourse features) associated with the
referent for the pronominal anaphor. The literature shows that features of animacy, specificity,
definiteness and semantic gender condition the use of full pronouns and null objects in
Brazilian Portuguese and this is what we investigate in our work. Our central hypothesis is
that only one of these referent characteristics may be the one which triggers the use of the full
pronoun or null object: it would be the semantic gender feature. Another of our hypothesis is
that the use of a null category on anaphoric recapture in object position (a null object) is the
default strategy, unmarked in BP (in contrast, the use of a pronoun in this place is the marked
strategy). To confirm our hypothesis, we have analyzed the speech of children of 1 to 9 years
old, from the corpora CEAAL (PUCRS) and PEUL (UFRJ).
Our hypothesis that the semantic gender feature is enough to explain the condition of
anaphoric resume in BP proved to be inconclusive, but it seems to be promising, as the results
analyzed from the semantic gender of the referents seem to polarize better the recaptures
between null objects and full pronouns. Furthermore, it is better to explain a phenomenon by
only one trace and not by a combination of features. Our hypothesis that the unmarked way to
resume a null object in BP is through an empty category was confirmed after analyzing the
data from our corpus, and we believe that on kids’ speech the use of null objects is
generalized, because the empty category happens in cases in which we hoped to see a
pronoun
Contextos licenciadores de sujeitos nulos em português brasileiro, 2021
The phenomenon of the null subject has already been extensively investigated in natural
languages... more The phenomenon of the null subject has already been extensively investigated in natural
languages. In this regard, Brazilian Portuguese raises a lot of interest for having passed,
from the 19th to the 20th century, through a change in the preference for expressing the
subject or not. With this in mind, this thesis focuses on the phenomenon of the null subject
in BP, a language in which null subjects were once preferred and today are muchacho less
frequent (cf. Duarte, 1993/1995; Duarte, Mourão & Santos, 2012; and, Othero & Spinelli
2019a, b, for example). The literature pounds out elements that would be relevant for this
change, such as the hypotheses of rich verbal morphology (Duarte, 1993/1995); the
semantic features [+/- human, +/- referential] (Cyrino, Duarte & Kato 2000) and [+/-
semantic gender] (Creus & Menuzzi, 2004), the linear prosody V2 (Kato, 2020) and the
context discursive (Paredes Silva, 2003). However, although these assumptions show
strong trends, none of them explains all null subject’s data. So, our goal here was to
investigate the contexts that still allow null subjects in BP - a language that was +prodrop
-, to provide an explanation that would account for the totality of the data found,
and, moreover, that can predict any null subject data that may be found. In this search,
we propose a unified analysis of the phenomenon.
Aelius Falado 1.0, 2014
This project is included in the area of Corpus Linguistics and morphological analysis of Brazilia... more This project is included in the area of Corpus Linguistics and morphological analysis of Brazilian Portuguese, which is about automatic morphological annotation corpus of spoken language. The work aims to contribute to the improvement of the automatic morphosyntactic tagger Aelius, originally developed by prof. Dr. Leonel Alencar, from Universidade Federal do Ceará, coordinator of the project CompLin - Natural Language and Computation. The tags used in Aelius are the same tags of Tycho-Brahe historical Portuguese corpus (electronic corpus already tagged, composed of texts in Portuguese written by authors born between 1380 and 1845). The texts that were used for tagging come from the bank of Varsul project, which studies the linguistic variation in southern Brazil. We worked with automatic tag of 20 texts, containing 154.530 words. This project presents the tool for automatic tagging, the automatic morphosyntactic analysis process executed by the tagger and our auto-tagging manual review work, besides some suggestions of improvement to deal specifically with aspects of orality. We will verify the Aelius’ performance in automatic morphosyntactic tagging of a corpus of spoken language, an unprecedented challenge to this tool. From the errors of the tagger, we seek to infer certain patterns of annotation to overcome limitations presented by the program, proposing some suggestions of implementations so that Aelius can tag even more satisfactorily corpus of spoken language. We will deal especially with cases of interjections, apheresis, onomatopoeia and conversational markers.
Uploads
Papers by Mônica Rigo Ayres
(2003) set out to study the
p r o c e s s i n g o f p r o n o m i n a l
anaphoras, especially with regard
to the preference for linking the
anaphoric pronoun to the closest
antecedent or topic of the
sentence. To do so, the authors
used a questionnaire that was
applied to 30 informants. In the
questionnaire applied by Haag and
Othero, in each target sentence
t h e r e w e r e t w o p o s s i b l e
antecedents, a topic and another
one closer to the anaphora.
However, the anaphora was
already filled in by a pronoun, but
it could also happen through an
empty category, the null subject. It
is in this context that the idea of
investigating the same sentences
arose, but without inserting the
anaphoric pronoun, to test the
speakers' preference for completing
the sentence using the pronominal
anaphoric or null form. In this
study, our objectives are (i) to test
the speakers' preference for
completing the sentence using the
pronominal or null form; (ii) check
whether there is a pattern or a
relationship in the preference for
linking this anaphora (pronominal
or null) to one of the possible
antecedents (topical or closest
referent); and, (iii) verifying if
antecedent features influence the
choice for overt or null subject.
Regarding objective (i), we found a
greater number of pronominal
anaphoras; about the objective (ii),
o u r d a t a d i d n o t s h o w a
relationship between the factors;
regarding objective (iii), we present
trends that should be further
investigated.
_
In this text, we review the study of Ayres and Othero (2016), on the conditioning of null objects and pronouns in BP, concerning anaphoric direct object of third person in children’s speech. They based their analysis on the two hypotheses consolidated in the literature, namely the semantic gender hypothesis (cf. CREUS; MENUZZI, 2004; OTHERO et al., 2016, and others), the animacy and specificity hypothesis (c. CYRINO, 1994/1997; CASAGRANDE, 2007, for example). Since Ayres and Othero (2016) did not distinguish VP ellipses from null objects, here we separate these two types of empty categories and analyze them. In addition, we verify if the conditioning of the null objects and the VP ellipses occurs in similar contexts, considering that these structures have superficially similar forms.
Palavras-chave: objeto direto; retomada anafórica; objeto nulo; português brasileiro.
At least, since the XIX century, the use of accusative clitic for third person (o, a) in Brazilian Portuguese has yielded to two strategies, (a) the use of tonic pronoun ele, ela or (b) the use of the called nulled direct object. Assuming the Creus & Menuzzi’s (2004) hypothesis and believing that, for the PB grammar, the relevant feature to use null object is the semantic gender of the referent, in this article we will bring the idea that there is an unmarked strategy and other marked strategy in the case of anaphoric recover for direct objects in 3rd person. Through reanalysis of tests proposed by C&M and applications of new tests, we aim to show that the marked strategy is the use of a pronoun and the unmarked strategy is the use of null object, and the semantic-pragmatic features of the referent are relevant to each option.
Keywords: direct object; anaphor; null object; Brazilian Portuguese.
PALAVRAS-CHAVE: etiquetagem Automática; etiquetagem morfossintática; Linguística de Corpus.
In this paper, we present the results of our work on automatic morphological annotation using the free morphosyntatic Aelius tagger. The excerpts analyzed were taken from a corpus of spoken language – part of the VARSUL project. We present 20 texts containing 154,530 words, which were automatically annotated and manually corrected. This paper also describes the Aelius tagger and our effort of manual reviewing the texts, as well as our suggestions for improving the tool, concerning aspects of oral texts tagging. We assessed the performance of the tagger at the morphosyntactic tagging of a spoken language corpus, an unprecedented challenge for the tool. Based on its errors, we try to infer certain patterns of annotation to overcome the limitations presented by the program. We also make suggestions in order to allow Aelius to tag spoken language corpora in a more effective way, specially when dealing with interjections, apheresis, onomatopoeia and conversational markers.
KEYWORDS: tagger; morphosyntactic tagging; Corpus Linguistics.
Palavras-chave: retomada anafórica, objeto nulo, português brasileiro, linguagem infantil.
The pronominal inventory in Brazilian Portuguese (BP) has been undergoing changes over time. Since the nineteenth century, the accusative third person clitic (o, a) has been falling away. BP grammar provides two new strategies to replace the clitic: the full pronoun (ele, ela) or a null element. The choice of one strategy or the other is not random; it is straightforwardly related to semantic (and perhaps discoursive) features of the antecedent. According to the literature on the subject, features of animacy and specificity or semantic genderaretheonesthatconditiontheuseofpronounsornullobjectsinBP.Thisiswhatwe investigatehere. Ourcentralhypothesisisthatonlyoneofthesefeaturesisactuallytheone that triggers the use of a pronoun or the null object in anaphoric resumption, namely the semanticgender. Tosupportthishypothesis,weanalyzethespeechofchildrenbetweenthe ages of 1 to 9 years, extracted from two corpora: CEAAL (PUCRS) and PEUL (UFRJ).
Keywords: anaphora resolution, null object, Brazilian Portuguese, child language.
Book Reviews by Mônica Rigo Ayres
Teaching Documents by Mônica Rigo Ayres
(1994/1997), Nunes (1993) and Bagno (2011) among others), since 19st century, the use of the
accusative clitic for the third person (o, a) has decreased. To recover an anaphoric element in
a direct object position, speakers use two strategies instead of the clitic: the full pronoun (he
‘ele’, she ‘ela’), or the null direct object.
The choice of using the full pronoun or null object is not random, before, this choice is made
by the influence of semantic features (and maybe discourse features) associated with the
referent for the pronominal anaphor. The literature shows that features of animacy, specificity,
definiteness and semantic gender condition the use of full pronouns and null objects in
Brazilian Portuguese and this is what we investigate in our work. Our central hypothesis is
that only one of these referent characteristics may be the one which triggers the use of the full
pronoun or null object: it would be the semantic gender feature. Another of our hypothesis is
that the use of a null category on anaphoric recapture in object position (a null object) is the
default strategy, unmarked in BP (in contrast, the use of a pronoun in this place is the marked
strategy). To confirm our hypothesis, we have analyzed the speech of children of 1 to 9 years
old, from the corpora CEAAL (PUCRS) and PEUL (UFRJ).
Our hypothesis that the semantic gender feature is enough to explain the condition of
anaphoric resume in BP proved to be inconclusive, but it seems to be promising, as the results
analyzed from the semantic gender of the referents seem to polarize better the recaptures
between null objects and full pronouns. Furthermore, it is better to explain a phenomenon by
only one trace and not by a combination of features. Our hypothesis that the unmarked way to
resume a null object in BP is through an empty category was confirmed after analyzing the
data from our corpus, and we believe that on kids’ speech the use of null objects is
generalized, because the empty category happens in cases in which we hoped to see a
pronoun
languages. In this regard, Brazilian Portuguese raises a lot of interest for having passed,
from the 19th to the 20th century, through a change in the preference for expressing the
subject or not. With this in mind, this thesis focuses on the phenomenon of the null subject
in BP, a language in which null subjects were once preferred and today are muchacho less
frequent (cf. Duarte, 1993/1995; Duarte, Mourão & Santos, 2012; and, Othero & Spinelli
2019a, b, for example). The literature pounds out elements that would be relevant for this
change, such as the hypotheses of rich verbal morphology (Duarte, 1993/1995); the
semantic features [+/- human, +/- referential] (Cyrino, Duarte & Kato 2000) and [+/-
semantic gender] (Creus & Menuzzi, 2004), the linear prosody V2 (Kato, 2020) and the
context discursive (Paredes Silva, 2003). However, although these assumptions show
strong trends, none of them explains all null subject’s data. So, our goal here was to
investigate the contexts that still allow null subjects in BP - a language that was +prodrop
-, to provide an explanation that would account for the totality of the data found,
and, moreover, that can predict any null subject data that may be found. In this search,
we propose a unified analysis of the phenomenon.
(2003) set out to study the
p r o c e s s i n g o f p r o n o m i n a l
anaphoras, especially with regard
to the preference for linking the
anaphoric pronoun to the closest
antecedent or topic of the
sentence. To do so, the authors
used a questionnaire that was
applied to 30 informants. In the
questionnaire applied by Haag and
Othero, in each target sentence
t h e r e w e r e t w o p o s s i b l e
antecedents, a topic and another
one closer to the anaphora.
However, the anaphora was
already filled in by a pronoun, but
it could also happen through an
empty category, the null subject. It
is in this context that the idea of
investigating the same sentences
arose, but without inserting the
anaphoric pronoun, to test the
speakers' preference for completing
the sentence using the pronominal
anaphoric or null form. In this
study, our objectives are (i) to test
the speakers' preference for
completing the sentence using the
pronominal or null form; (ii) check
whether there is a pattern or a
relationship in the preference for
linking this anaphora (pronominal
or null) to one of the possible
antecedents (topical or closest
referent); and, (iii) verifying if
antecedent features influence the
choice for overt or null subject.
Regarding objective (i), we found a
greater number of pronominal
anaphoras; about the objective (ii),
o u r d a t a d i d n o t s h o w a
relationship between the factors;
regarding objective (iii), we present
trends that should be further
investigated.
_
In this text, we review the study of Ayres and Othero (2016), on the conditioning of null objects and pronouns in BP, concerning anaphoric direct object of third person in children’s speech. They based their analysis on the two hypotheses consolidated in the literature, namely the semantic gender hypothesis (cf. CREUS; MENUZZI, 2004; OTHERO et al., 2016, and others), the animacy and specificity hypothesis (c. CYRINO, 1994/1997; CASAGRANDE, 2007, for example). Since Ayres and Othero (2016) did not distinguish VP ellipses from null objects, here we separate these two types of empty categories and analyze them. In addition, we verify if the conditioning of the null objects and the VP ellipses occurs in similar contexts, considering that these structures have superficially similar forms.
Palavras-chave: objeto direto; retomada anafórica; objeto nulo; português brasileiro.
At least, since the XIX century, the use of accusative clitic for third person (o, a) in Brazilian Portuguese has yielded to two strategies, (a) the use of tonic pronoun ele, ela or (b) the use of the called nulled direct object. Assuming the Creus & Menuzzi’s (2004) hypothesis and believing that, for the PB grammar, the relevant feature to use null object is the semantic gender of the referent, in this article we will bring the idea that there is an unmarked strategy and other marked strategy in the case of anaphoric recover for direct objects in 3rd person. Through reanalysis of tests proposed by C&M and applications of new tests, we aim to show that the marked strategy is the use of a pronoun and the unmarked strategy is the use of null object, and the semantic-pragmatic features of the referent are relevant to each option.
Keywords: direct object; anaphor; null object; Brazilian Portuguese.
PALAVRAS-CHAVE: etiquetagem Automática; etiquetagem morfossintática; Linguística de Corpus.
In this paper, we present the results of our work on automatic morphological annotation using the free morphosyntatic Aelius tagger. The excerpts analyzed were taken from a corpus of spoken language – part of the VARSUL project. We present 20 texts containing 154,530 words, which were automatically annotated and manually corrected. This paper also describes the Aelius tagger and our effort of manual reviewing the texts, as well as our suggestions for improving the tool, concerning aspects of oral texts tagging. We assessed the performance of the tagger at the morphosyntactic tagging of a spoken language corpus, an unprecedented challenge for the tool. Based on its errors, we try to infer certain patterns of annotation to overcome the limitations presented by the program. We also make suggestions in order to allow Aelius to tag spoken language corpora in a more effective way, specially when dealing with interjections, apheresis, onomatopoeia and conversational markers.
KEYWORDS: tagger; morphosyntactic tagging; Corpus Linguistics.
Palavras-chave: retomada anafórica, objeto nulo, português brasileiro, linguagem infantil.
The pronominal inventory in Brazilian Portuguese (BP) has been undergoing changes over time. Since the nineteenth century, the accusative third person clitic (o, a) has been falling away. BP grammar provides two new strategies to replace the clitic: the full pronoun (ele, ela) or a null element. The choice of one strategy or the other is not random; it is straightforwardly related to semantic (and perhaps discoursive) features of the antecedent. According to the literature on the subject, features of animacy and specificity or semantic genderaretheonesthatconditiontheuseofpronounsornullobjectsinBP.Thisiswhatwe investigatehere. Ourcentralhypothesisisthatonlyoneofthesefeaturesisactuallytheone that triggers the use of a pronoun or the null object in anaphoric resumption, namely the semanticgender. Tosupportthishypothesis,weanalyzethespeechofchildrenbetweenthe ages of 1 to 9 years, extracted from two corpora: CEAAL (PUCRS) and PEUL (UFRJ).
Keywords: anaphora resolution, null object, Brazilian Portuguese, child language.
(1994/1997), Nunes (1993) and Bagno (2011) among others), since 19st century, the use of the
accusative clitic for the third person (o, a) has decreased. To recover an anaphoric element in
a direct object position, speakers use two strategies instead of the clitic: the full pronoun (he
‘ele’, she ‘ela’), or the null direct object.
The choice of using the full pronoun or null object is not random, before, this choice is made
by the influence of semantic features (and maybe discourse features) associated with the
referent for the pronominal anaphor. The literature shows that features of animacy, specificity,
definiteness and semantic gender condition the use of full pronouns and null objects in
Brazilian Portuguese and this is what we investigate in our work. Our central hypothesis is
that only one of these referent characteristics may be the one which triggers the use of the full
pronoun or null object: it would be the semantic gender feature. Another of our hypothesis is
that the use of a null category on anaphoric recapture in object position (a null object) is the
default strategy, unmarked in BP (in contrast, the use of a pronoun in this place is the marked
strategy). To confirm our hypothesis, we have analyzed the speech of children of 1 to 9 years
old, from the corpora CEAAL (PUCRS) and PEUL (UFRJ).
Our hypothesis that the semantic gender feature is enough to explain the condition of
anaphoric resume in BP proved to be inconclusive, but it seems to be promising, as the results
analyzed from the semantic gender of the referents seem to polarize better the recaptures
between null objects and full pronouns. Furthermore, it is better to explain a phenomenon by
only one trace and not by a combination of features. Our hypothesis that the unmarked way to
resume a null object in BP is through an empty category was confirmed after analyzing the
data from our corpus, and we believe that on kids’ speech the use of null objects is
generalized, because the empty category happens in cases in which we hoped to see a
pronoun
languages. In this regard, Brazilian Portuguese raises a lot of interest for having passed,
from the 19th to the 20th century, through a change in the preference for expressing the
subject or not. With this in mind, this thesis focuses on the phenomenon of the null subject
in BP, a language in which null subjects were once preferred and today are muchacho less
frequent (cf. Duarte, 1993/1995; Duarte, Mourão & Santos, 2012; and, Othero & Spinelli
2019a, b, for example). The literature pounds out elements that would be relevant for this
change, such as the hypotheses of rich verbal morphology (Duarte, 1993/1995); the
semantic features [+/- human, +/- referential] (Cyrino, Duarte & Kato 2000) and [+/-
semantic gender] (Creus & Menuzzi, 2004), the linear prosody V2 (Kato, 2020) and the
context discursive (Paredes Silva, 2003). However, although these assumptions show
strong trends, none of them explains all null subject’s data. So, our goal here was to
investigate the contexts that still allow null subjects in BP - a language that was +prodrop
-, to provide an explanation that would account for the totality of the data found,
and, moreover, that can predict any null subject data that may be found. In this search,
we propose a unified analysis of the phenomenon.