PORTUNHOL E A AVALIAÇÃO ESCRITA DO CELPE-BRAS PORTUÑOL Y LA EVALUACIÓN ESCRITA DEL CELPE-BRAS PORTUNHOL AND THE CELPE-BRAS'S WRITTEN EXAM


Diego Giovanni RODRIGUEZ1 Laura Márcia Luiza FERREIRA2


RESUMO: O portunhol pode ser abordado sob diversas perspectivas. É no contexto de ensino e avaliação de larga escala que queremos discutir suas potencialidades e limitações. Problematizamos as discussões sobre o portunhol a partir de diversas perspectivas teórico- metodológicas para ensino de línguas próximas e dos construtos teóricos do exame Celpe-Bras. Apresentamos uma análise de dados que foram coletados em 2020 durante um curso online preparatório para o Celpe-Bras. Investigamos as notas atribuídas às tarefas escritas e as justificativas dadas pelos professores-avaliadores para selecionar um conjunto de seis textos, sendo três representativos da nota 4 e, três, da nota 5. A partir da classificação de estruturas híbridas dos textos selecionados constatamos que a frequência em que é usado o portunhol pode ter influenciado a variação de notas. Estruturas de natureza morfossintática e ortográfico- fonológica foram mais frequentes nos textos analisados do que as de natureza léxico-semântica.


PALAVRAS-CHAVE: Portunhol. Ensino de línguas. Avaliação de línguas. Celpe-Bras.


RESUMEN: El portuñol puede ser abordado desde diferentes perspectivas. Es en el contexto de enseñanza y evaluación de larga escala que discutiremos sus potencialidades y limitaciones. Problematizamos las discusiones sobre Portuñol, desde diferentes perspectivas teórico- metodológicas para la enseñanza de lenguas próximas y los constructos teóricos del examen Celpe-Bras. Presentamos un análisis de datos que fueron colectados en 2020 durante un curso en línea de preparación para el Celpe-Bras. Investigamos las calificaciones atribuidas a las tareas escritas y las justificaciones dadas por los docentes-evaluadores para seleccionar un conjunto de seis textos, trés representando nota 4 y tres representando nota 5. A partir de la clasificación de estructuras híbridas de los textos descubrimos que la frecuencia en la que se usa el portuñol puede haber influido en la variación de notas. Las estructuras morfosintácticas y ortográfico-fonológicas fueron más frecuentes en los textos analizados que las de naturaleza léxico-semántica.


PALABRAS CLAVE: Portuñol. Enseñanza de idiomas. Evaluación de idiomas. Celpe-Bras



1 Universidade da Integração Latinoamericana (UNILA), Foz do Iguaçu – PR – Brasil. Professor de Espanhol e Português como Línguas Estrangeiras. ORCID: https://orcid.org/0000-0002-5330-0571. E-mail: dgv.rodriguez.2016@aluno.unila.edu.br

2 Universidade da Integração Latinoamericana (UNILA), Foz do Iguaçu – PR – Brasil. Estudante de Espanhol e Português como Línguas Estrangeiras. ORCID: http://orcid.org/0000-0001-7632-0834. E-mail: dgv.rodriguez.2016@aluno.unila.edu.br



ABSTRACT: The portunhol can be approached from different perspectives. Our aim is to discuss its potentialities and its limitations in the large-scale assessment context, such as Celpe- Bras exam. First, we discuss how portunhol can be related to different language teaching theoretical-methodological perspectives and also to Celpe-Bras exam´s theoretical constructs. The data that we analyze was collected in 2020 during an online preparatory course. For the analyses, we selected a set of six texts, which was related to high levels of proficiency. We investigated the scores attributed and the justifications given by the teachers-evaluators. According to the classification of hybrid structures found on our set of texts, we found out that the frequency of portunhol could have influenced the scores variations. Besides, in the analyzed texts morph-syntactic and orthographic-phonological hybrid structures were more frequent than lexical-semantic structures.


KEYWORDS: Portunhol. Language teaching. Language assesment. Celpe-Bras.


Introdução


O portunhol pode ser definido como uma língua híbrida utilizada em diferentes contextos, tais como: o portunhol de fronteira ou fronteiriço, o portunhol literário, portunhol em sala de aula, dentre outros. Faulstich (1997), ao estudar o portunhol no território fronteiriço do Paraguai, do Uruguai e da Argentina com o Brasil, afirma que o conceito de língua possui um papel sociopolítico e linguístico e está relacionado às ideias de estado e de nação. Nessa luta ideológica, que pretende delimitar e identificar os espaços geográficos, encontram-se posicionamentos que enxergam o portunhol de fronteira ora como uma língua confusa e uma afronta às línguas oficiais, ora como uma língua que enaltece a identidade fronteiriça. No contexto de ensino, a tensão também existe, pois o portunhol pode ser visto como uma ameaça ao espaço discursivo das variantes normatizadas do espanhol ou do português.

Por meio do portunhol e da representação de uma escrita próxima à fala, escritores acentuam a complexa e ambígua construção de identidade nos espaços fronteiriços, onde encontram-se e desencontram-se duas línguas com suas próprias representações sociais. O uso do portunhol nas produções literárias reivindica a existência daquelas minorias que o enxergam como uma língua familiar e de herança.

Se por um lado, o portunhol é festejado nas produções literárias e nas conversas cotidianas da fronteira, no sistema escolar o portunhol afronta a manutenção do mito do monolinguismo. O poeta fronteiriço Fabián Severo afirma que o sistema escolar uruguaio rechaça o portunhol, pois é classificado como “fala de pobre” ou “incorreto”. Embora o portunhol faça parte do processo de contatos linguísticos e da construção de subjetividades, nos questionamos sobre até que ponto o uso da língua híbrida pode ou não ser um recurso linguístico suficiente.



O portunhol desafia as noções de língua de muitos professores e avaliadores, dada sua natureza híbrida. Por meio desta pesquisa, nos propomos a debater como a interpretação da nota para certificar a proficiência em português língua adicional pode ser influenciada pelo uso do portunhol. Discutiremos o uso do portunhol nas provas escritas do exame Certificado de Proficiência em Língua Portuguesa para Estrangeiros (Celpe-Bras).

Antes de abordar a avaliação, apresentamos uma discussão de propostas teóricas- metodológicas e seus potenciais contribuições para o ensino e avaliação do portunhol. Ao final, apresentamos uma análise e uma discussão sobre a interpretação do portunhol no contexto da correção de tarefas escritas do exame Celpe-Bras.


Interlíngua, ensino e avaliação do portunhol


Baralo (2004) define o conceito de interlíngua como um estado transitório entre a L1 e a L23 no qual existem interferências da L1 na língua alvo. Os estudos de interlíngua têm como objetivo principal compreender a natureza cognitiva da aquisição de uma segunda língua. A pergunta de fundo dos estudos de interlíngua se refere ao quanto a L1 interfere ou não na aquisição da L2. Tais pesquisas impactaram na forma como se ensina e aprende segundas línguas de forma a reforçar o distanciamento da L1 do processo de aprendizagem de uma L2. A relação entre a L1 e o processo de ensino e aprendizagem se tornou um conflito. O cenário fica ainda mais complexo quando se soma, ao mal-estar da relação entre as línguas maternas e adicionais no processo de ensino, as questões de poder implicadas nas políticas linguísticas e as relações de prestígio entre as línguas.

Embora a proposta teórico-metodológica da interlíngua se volte a comparar dois sistemas linguísticos com um terceiro sistema linguístico intermediário, chamado interlíngua, existem alguns pontos que explicam sua influência no âmbito pedagógico. A ideia de uma língua padrão como língua alvo pressupõe um modelo de referência para explicar e prever interferências e erros. O modelo de referência se soma à predileção para o ensino estrutural, que ainda é, infelizmente, o construto de língua mais difundido pelos sistemas educacionais brasileiros. Além disso, uma proposta de ensino supostamente baseada na interlíngua posicionaria os aprendizes em um nível específico de aprendizagem, baseando-se nos tipos de erros que cometem ao falar ou escrever, para que sejam planejadas ferramentas de ensino e




3Os termos L1 (primeira língua) e L2 (segunda língua) são usados para se referir às línguas materna e estrangeira, respectivamente, no contexto do debate sobre interlíngua. Não discutiremos as diferenças entre esses conceitos dentro do campo da aquisição.

Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, dez. 2021. e-ISSN: 2447-3529



aprendizagem que auxiliem os aprendizes em cada um dos estágios da interlíngua. No contexto da avaliação, a classificação do erro por nível do examinando ainda fundamenta exames de avaliação de línguas, especialmente os que pautam os itens avaliados em interferências ou inadequações de formas linguísticas específicas. No Celpe-Bras, por exemplo, os recursos linguísticos também são avaliados, porém avalia-se seu grau de adequação em contextos específicos de uso da língua.

Reconhecemos a importância dos estudos contrastivos, das análises de erro, especialmente no debate sobre aquisição de segunda língua, ou interpretação sobre a acurácia do uso da língua no contexto de ensino e avaliação, como é o caso deste trabalho. Porém, consideramos um limitador quando o professor de línguas toma para si a tarefa de classificação de erros no processo de ensino, quando deveria estar fomentando o uso da linguagem e a reflexão linguística em contextos diversos. Abaixo discutimos duas perspectivas teóricas que acolhem a presença da L1 no processo de ensino e aprendizagem da segunda língua de forma menos conflituosa, tais como a da translinguagem (CANAGARAJAH, 2011; JASPERS, 2018) e da intercompreensão (ESCUDÉ; OLMO, 2019).


Translinguagem, ensino e avaliação do portunhol


A translinguagem surge como uma abordagem pedagógica para aprendizes multilíngues e parte do pressuposto de que as línguas não devem ser consideradas como sistemas separados. De acordo com Canagarajah (2011), as línguas interagem e formam um repertório linguístico único, de forma que o indivíduo pode escolher os elementos necessários para se comunicar e ser compreendido em cada situação. O autor aponta para a necessidade de estudar as estratégias de comunicação empregadas por falantes multilíngues, de modo que o foco passa a ser o engajamento dos indivíduos em práticas sociais e discursivas que promovam o desenvolvimento do seu repertório linguístico, valorizando as biografias linguísticas dos estudantes. O objetivo do ensino de línguas seria o de negociar significados e usos. Sobre a norma, Canagarajah (2011) advoga contra o ensino de regras gramaticais.

No entanto, afirmar que o repertório linguístico dos estudantes sempre pode ser aproveitado de forma positiva não reflete a complexidade social e cultural da educação. Sobre as limitações da perspectiva da translinguagem, Jasper (2018, p. 18) adverte “que mesmo uma escola baseada na translinguagem, provavelmente exigirá que os alunos renunciem a certas partes de seu repertório linguístico (linguagem sexista e racista, gíria juvenil, jargão de jogo) para expandir novas habilidades (redação acadêmica, por exemplo)”. O autor defende a




politização do debate sobre o quê e por quê aprender e desaprender, uma vez que, para o desenvolvimento social do indivíduo é necessário aprender e desenvolver novos conhecimentos e ferramentas e, ao mesmo tempo, reformular ou esquecer aquilo que não é considerado “adequado” culturalmente. Jasper (2018) defende que o repertório linguístico dos aprendizes deve encontrar um equilíbrio entre a dimensão individual e a dimensão coletiva. Este debate está diretamente relacionado à interpretação do portunhol em contextos de avaliação, uma vez que o portunhol nem sempre será um recurso suficiente.

A partir da perspectiva da translinguagem, podemos afirmar que a presença do portunhol no processo de ensino, aprendizagem e avaliação já não seria uma manifestação negativa de uma interlíngua, mas parte de um repertório linguístico. Porém, se pensarmos no processo de interação em portunhol, cabe a reflexão e o debate sobre seus limites em determinadas situações de uso da língua. Em relação aos processos de avaliação, a translinguagem é uma perspectiva que desafia instrumentos de avaliação linguística baseados em um construto de língua como estrutura, normas e padrões normativos. Cabe ressaltar aqui que o exame Celpe-Bras operacionaliza o construto de língua em suas tarefas de avaliação de maneira a levar em conta diversos aspectos do uso da língua e não apenas o uso adequado ou não de suas formas. Podemos dizer que, prioriza-se o repertório linguístico do examinando ao avaliar o seu desempenho quando se engaja nas propostas de tarefas de avaliação que simulam o uso que se faz da língua no mundo real. Porém, nos perguntamos neste trabalho se há e onde estaria um equilíbrio entre a dimensão individual e a dimensão coletiva do uso do portunhol neste contexto de avaliação escrita.


Intercompreensão, ensino e avaliação do portunhol


Tanto a perspectiva da translinguagem quanto a da intercompreensão partem do pressuposto de que é natural os seres humanos negociarem significados ao interagir em várias línguas. Ambas as perspectivas teóricas se dedicaram a oferecer diretrizes pedagógicas para o ensino de línguas. Embora a translinguagem e a intercompreensão defendam o plurilinguismo nos sistemas escolares, cada perspectiva propõe uma didática que acolhe outras línguas no processo de educação linguística dos estudantes.

Segundo Escudé e Olmo (2019), a intercompreensão refere-se a um fenômeno associado às práticas comunicativas observáveis nas quais há um esforço espontâneo entre falantes de diferentes línguas e variedades para se compreenderem mutuamente. A partir dos estudos históricos comparativos de sistemas linguísticos dentro e fora do grupo de línguas românicas,

Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, dez. 2021. e-ISSN: 2447-3529



a intercompreensão surge como uma abordagem pedagógica baseada na exploração de semelhanças e/ou transparências entre as línguas.

Partindo das dinâmicas de identificação e distanciamento, os autores afirmam que a intercompreensão tem como objetivo pedagógico aumentar a capacidade de compreensão entre as línguas. As formas linguísticas são parte do processo pedagógico do ensino de línguas, diferentemente do que propõe a translinguagem, ao fomentar o cancelamento do ensino com o foco na forma. Na perspectiva da intercompreensão, almeja-se desenvolver a percepção de traços da linguagem, que pertencem a duas ou mais línguas, se assemelham e se distanciam, promovendo a reflexão sobre aspectos da linguagem que são ou não exclusivos de um grupo de falantes e/ou adequados ou não a um contexto de interação.

Embora a intercompreensão não fomente o portunhol ou misturas dos códigos linguísticos no âmbito pedagógico, ao contrário da translinguagem, as noções de intencionalidade e direito a imprecisão são duas grandes contribuições mencionadas por Escudé e Olmo (2019), que podemos aplicar no contexto pedagógico e de avaliação do portunhol. A noção de intencionalidade está relacionada à atitude do interlocutor. Para haver intercompreensão é preciso que os interlocutores tenham intenção de compreender uns aos outros, no nosso caso, estaria relacionado à atitude do(a) avaliador(a) perante o portunhol. No contexto de avaliação, cabe relacionar a intencionalidade com o quanto o sistema de crenças, a concepção de língua, preconceitos e estereótipos de avaliadores interagem com as diretrizes de avaliação de forma a facilitar ou dificultar a interlocução entre avaliandos e avaliadores.

A noção de direito à imprecisão está relacionada aos graus de aceitação do portunhol. O direito à impressão no contexto da avaliação implica na escolha de não descrever o nível mais alto de proficiência de exames, tendo como referência o falante nativo escolarizado. O exame Celpe-Bras, por exemplo, operacionaliza o direito à imprecisão dos examinandos ao aceitar nos níveis mais altos de proficiência uma descrição de performance linguística com inadequações quanto ao uso dos recursos linguísticos, por exemplo.

Tendo feito uma breve revisão teórica de distintas perspectivas pedagógicas que nos auxiliam a pensar o ensino e a avaliação de línguas próximas no par português-espanhol, discorreremos sobre o exame Celpe-Bras.




O Celpe-Bras


O exame Celpe-Bras foi desenvolvido no contexto de negociações do Mercado Comum do Sul (MERCOSUL) na década de 90, juntamente com o exame argentino de espanhol CELU (Certificado de Español: Lengua y Uso). Sobre a relação entre o exame Celpe-Bras e as políticas de promoção da língua falada no Brasil, Dell' Isola e Scaramucci (2003) afirmam que a consolidação de parâmetros de avaliação linguística constitui uma ferramenta política do estado que reivindica e delimita uma língua perante as outras. Embora a delimitação de línguas possa ter um efeito negativo no âmbito escolar, como a tensão sobre o uso do portunhol nas escolas de fronteira, a consolidação de um instrumento de certificação linguística tem um efeito positivo na promoção do Português falado no Brasil pelo estado brasileiro no contexto das políticas educacionais e culturais de soft-power.

Faz-se necessário politizar o debate sobre quando? E por quê? A exigência da certificação é ou não adequada? Consideramos pertinente, por exemplo, que candidatos a pós- graduação em Instituições de Ensino Superior brasileiras, cuja língua materna não seja o português, comprovem condições de frequentar aulas e de escrever trabalhos acadêmicos em língua portuguesa. Em muitos editais de seleção de pós-graduação é mencionada a certificação emitida pelo Celpe-Bras como um possível documento de comprovação da proficiência em língua portuguesa por falantes de outras línguas.

Uma das características do Celpe-Bras é sua natureza comunicativa. Nas palavras de Bachman (2003, p. 84) as habilidades comunicativas da linguagem podem ser definidas como “o conhecimento, ou competência, e capacidade de implementação ou execução dessa competência no uso comunicativo da língua, de modo apropriado e contextualizado''. Scaramucci (2001) afirma que o Celpe-Bras tem como referência um falante de “proficiência plena não nativo”, isto quer dizer, um indivíduo que tenha um controle operacional completo da língua, embora possam ser observados lapsos e imprecisões. A depender do grau de adequação das respostas, o examinando pode ser classificado em quatro faixas de proficiência, a saber: intermediário, intermediário superior, avançado e avançado superior.

No contexto do Celpe-Bras, a interpretação do portunhol implica vários desafios. Se compararmos as habilidades comunicativas dos examinandos hispanofalantes com outros cuja língua materna não seja o espanhol, é provável que o hispanofalante terá mais habilidades para negociar significados, embora possa haver inadequações de recursos linguísticos e mal- entendidos. Sobre as potencialidades e limites do portunhol no contexto do exame, Scaramucci (2001, p. 89):



Será a consideração do critério de comunicação que o exame está incentivando a manifestação do portunhol? E será que o portunhol pode atender às necessidades comunicativas dos candidatos? Por outro lado, como não cair no extremo exigindo um nível de proficiência nativa como parâmetro de proficiência?


A professora resume nas perguntas acima o grande dilema presente em diversas situações de avaliação enfrentadas tanto por avaliadores iniciantes quanto por experientes. Se por um lado, o fato de o português e o espanhol serem línguas próximas facilita a comunicação entre os falantes de ambas as línguas, por outro, até que ponto o portunhol é um recurso adequado? Essa é uma pergunta de fundo que gera muito debate em situações de discussão de notas, pois no contexto da avaliação exige-se dos formuladores do exame e dos avaliadores uma diretriz comum de avaliação a fim de garantir a confiabilidade das notas do exame. No entanto, mesmo havendo uma diretriz de avaliação bem-informada por meio de manuais, treinamento e escalas, diversos especialistas em avaliação de desempenho endossam o pressuposto de que há uma interação entre tarefa, avaliadores e escala (FULCHER, 2003). Por isso, buscamos discutir neste trabalho, a partir da experiência em um curso preparatório para o exame Celpe-Bras, sobre o que é adequado ou não quanto ao uso do portunhol em diversas situações forjadas nas tarefas do exame por meio da análise dos textos.


Recursos linguísticos e o portunhol na avaliação escrita do Celpe-Bras


A prova escrita é composta por quatro tarefas que geram respostas abertas que são avaliadas por dois ou mais avaliadores, a partir de uma grade de avaliação comum. As notas são independentes e há controle de discrepância. Quanto ao desenho da prova escrita, na tarefa 1 o examinando deve escrever um texto após ver um vídeo; na tarefa 2, um áudio; e na tarefa 3 e 4, ler textos impressos. Para exemplificar um comando, analisamos a tarefa 3 que fez parte da prova de 2016. 2.


Você tem um blog sobre alimentação saudável e, depois de ler a notícia sobre o cadastro de hortas em São Paulo, decidiu implementar essa iniciativa em sua cidade. Escreva um texto em seu blog para pedir a participação dos leitores no cadastramento, explicando o funcionamento e a utilidade das informações coletadas (BRASIL, 2016, p. 6).


O comando acima vem acompanhado do texto de leitura intitulado “À procura de hortas em plena São Paulo” publicado na seção estilo de vida do portal de notícias Estadão. O texto versa sobre uma iniciativa chamada Wikiversidade, que consiste em mapear colaborativamente




hortas em áreas urbanas por meio de uma plataforma online. No comando, define-se o gênero textual: postagem; a interlocução: blogueiro escrevendo para leitores; bem como o propósito de escrita: pedir participação dos leitores no cadastramento de hortas em uma cidade definida. É esperado ainda que os examinandos recontextualizem no seu texto informações da reportagem do Estadão no que diz respeito ao funcionamento e a utilidade das informações que serão coletadas.

A definição da situação e contexto de uso da linguagem pelo comando direciona a maneira como os itens que compõem a grade de avaliação de tarefas escritas serão interpretados. Avalia-se a adequação quanto à interlocução no gênero discursivo proposto, à realização da ação solicitada e à recontextualização de informações. A avaliação do uso dos recursos linguísticos mobilizados no texto está em função do cumprimento da tarefa.

A descrição de tais itens avaliados na grade é feita de maneira holística, ou seja, há uma descrição única para cada nota de 0 a 5, que corresponde às faixas de certificação previstas pelo exame. Nos descritores de cada faixa, granulam-se o grau de adequação da resposta às tarefas em relação ao que foi pedido no comando. Quanto à avaliação dos recursos linguísticos, objeto de análise deste trabalho, um texto nota 5 corresponde a “um texto autônomo, claro e coeso, em que os recursos linguísticos acionados são apropriados para configurar a relação de interlocução no gênero solicitado, e possíveis inadequações raramente comprometem a fluidez da leitura” (BRASIL, 2020, p. 39). Na descrição da nota 4, quanto à adequação dos recursos linguísticos, define-se que “os recursos linguísticos acionados são apropriados para configurar a relação de interlocução no gênero proposto, construindo um texto claro e coeso em que possíveis inadequações podem comprometer, em momentos localizados, a fluidez na leitura.” (BRASIL, 2020, p. 39).

Os recursos linguísticos são avaliados tanto em função de um contexto de interlocução forjado no comando das tarefas quanto ao grau de comprometimento da fluidez da leitura por eventuais inadequações linguísticas. Podemos inferir que o portunhol na avaliação escrita do Celpe-Bras, embora não seja objeto de avaliação do teste, está relacionado ao uso dos recursos linguísticos. É a partir da forma como a resposta do examinando consegue estabelecer adequadamente a interlocução, o propósito e oferecer as informações pertinentes que o uso do portunhol será interpretado para fins de atribuição de nota. Ao contrário dos exames focados na forma, Tosatti (2012) reitera que, no âmbito de avaliação das competências comunicativas, os fatores socioculturais e situacionais para a produção do discurso, são tão (ou mais) importantes que as regras gramaticais.





A indefinição sobre o que é ou não considerado, quanto ao uso do portunhol, como uma inadequação que compromete ou não a leitura e compreensão da resposta redigida pelos examinandos faz os avaliadores, em contexto de resolução de discrepância de notas, negociarem graus de tolerância ou leniência na hora de atribuir notas. Para discutir melhor a questão, cito dois exemplos, sendo o primeiro: “Já no final tem um espaço com milhes de sobrenomes gravados numa parede.” e o segundo: “O importante é que a gente este mais saudável e ajudar para que outras pessoas conheçam este site.” No exemplo 01, interpretamos que a inadequação em “milhes” não compromete a compreensão global da ideia, mas no exemplo 02 as inadequações prejudicam compreensão da frase porque no lugar do verbo “esteja” está escrito “este”, que em português é um pronome demonstrativo “este menino”. Além disso, a construção “o importante é que a gente (…) e ajudar...” dificulta a compreensão, provavelmente comprometendo a fluidez da leitura que o(a) avaliador(a) fará do texto. Por isso, cabe-nos discutir a natureza das inadequações relacionadas ao portunhol e como são interpretadas pelos avaliadores para fins de atribuição de nota. Apresentamos abaixo uma análise e discussão para elucidar os graus de tolerância dos avaliadores diante o portunhol no contexto de avaliação escrita do Celpe-Bras.


Metodologia


Os dados foram coletados durante um curso online de preparação para o Celpe-Bras. O curso foi oferecido durante os meses de agosto a outubro de 2020. Todos os vinte estudantes ativos no curso tinham o espanhol como língua materna. A equipe de professores era formada por oito estudantes de licenciatura em Letras. Dos oito, cinco estavam quase se formando e três fizeram menos da metade da graduação. Ou seja, o perfil dos avaliadores deste trabalho é próximo ao perfil dos avaliadores do exame. Fazia parte do planejamento do curso a entrega de quatro tarefas escritas do exame Celpe-Bras. O mesmo texto era corrigido por mais de um(a) professor(a) de forma independente. Em reuniões pedagógicas, resolvia-se a discrepância das notas previamente atribuídas individualmente e uma nota final era negociada entre a equipe e a coordenação. Procurou-se, desta forma, aproximar o contexto da pesquisa com as situações de avaliação real das tarefas escritas do exame Celpe-Bras.

Nesta edição do curso, foram entregues aproximadamente 60 textos pelos cursistas e foram geradas 200 notas. A partir da experiência durante o processo de acompanhamento das reuniões de discrepância e análise das notas, optamos neste trabalho por fazer a discussão da discrepância de notas que variaram entre 4 e 5, que corresponde à faixa avançado e avançado-




superior, respectivamente. A variação de 1 ponto não significa uma discrepância significativa para fins de reavaliação da nota (BRASIL, 2016). A escolha se justifica porque pretendemos levantar informações para discutir as possíveis características de um portunhol que esteja relacionado tanto ao uso de inadequações linguísticas que comprometem raramente a fluidez da leitura, conforme descrição da nota 5, quanto aos textos que, segundo descrição da nota 4, apresentam inadequações que comprometem “em momentos focalizados” a compreensão ou fluidez da leitura que faz o avaliador.

Essa escolha se deve também ao fato de que, segundo a grade de avaliação, os itens avaliados nas faixas 4 e 5 que se referem à adequação discursiva estariam adequados, restando ao avaliador julgar mais como os recursos linguísticos foram usados para o cumprimento da tarefa ou na qualidade do cumprimento da tarefa do que o cumprimento da tarefa em si. Supomos aqui que o examinando, cuja nota variou de 4 a 5, conseguisse estabelecer a interlocução e cumprir o propósito da tarefa, devemos indagar sobre quais seriam as manifestações de portunhol ou inadequações que estariam norteando a interpretação dos avaliadores para fins de atribuição de notas 4 e 5. A seguir, esboçamos uma resposta a esta pergunta a partir dos dados apresentados e analisados.

Como ferramenta de análise de dados, optamos pela análise de erros, baseado nos estudos contrastivos para examinar as interferências linguísticas derivadas da L1. Os erros no contexto deste trabalho se referem às construções híbridas ou inadequações linguísticas. A escolha se justifica pela necessidade de sistematizar e organizar as estruturas híbridas relacionadas ao portunhol presentes no corpus.


Análise e discussão


Analisamos a produção escrita de dois estudantes, Paula e João, em três tarefas diferentes. As tarefas 3 e 1 referem-se às edições do primeiro semestre de 2016 e, a 2, do segundo de 2017. O recorte de análise se justifica pelo padrão das notas. Ao analisar as notas atribuídas independentemente por três avaliadores, pudemos perceber a variação entre as notas 4 e 5. Cabe ressaltar que, na grande maioria das vezes, os avaliadores dos textos de Paula justificavam a nota 4 argumentando que as inadequações linguísticas eram frequentes e que influenciavam na fluidez da leitura do texto. Na tabela 1, abaixo, está organizado o padrão de notas de Paula e João cujas respostas às tarefas serão analisadas em seguida.




Tabela 1 – Notas das tarefas 3, 1 e 2


Cursistas

Paula

João

Avaliadores

Avaliador 1

Avaliador 2

Avaliador 3

Nota Final

Avaliador 1

Avaliador 2

Avaliador 3

Nota Final

Notas atribuídas à tarefa 3


4


3.5


3.4


3


5


5


5


5

Notas atribuídas à Tarefa 1


4


4


4


4


5


5


5


5

Notas atribuídas à Tarefa 2


4


4


4


4


5


5


5


5

Fonte: Acervo dos autores


A tarefa 3 foi a primeira a ser feita pelos cursistas, cabe ressaltar que na primeira tarefa a nota 5 foi unânime entre os avaliadores do texto de João, ao passo que a primeira nota de Paula foi ligeiramente menor que as atribuídas para suas próximas produções. Pela análise das notas das três tarefas atribuídas por três avaliadores independentes, podemos perceber pouca variação da nota 4 atribuída à Paula e da nota 5, aos textos de João. Por este motivo, analisamos o conjunto dos seis textos.

Categorizamos as estruturas híbridas nos textos e sua frequência. Em seguida, comparamos e discutimos os dados. A análise foi feita por um pesquisador, falante de espanhol como primeira língua e português como língua adicional, graduado em Letras, e revisada por uma professora brasileira especializada em análise contrastiva.

Na seguinte tabela, foram classificadas as estruturas híbridas mais frequentemente encontradas. Em geral, as estruturas remetiam à influência da língua materna, no caso o espanhol. Em outros, poderíamos inferir pouco domínio de mecanismos de textualidade. Identificamos e classificamos três categorias, a saber: Morfossintáticos (EM), Ortográfico- fonológicos (EOF) e Léxico-semânticos (EL). Na tabela 2, apresentamos suas definições e os exemplos.





Tabela 2 – Categoria de análise das estruturas híbridas:


Categorias de estruturas híbridas


Definição e exemplos

Morfossintáticas

Transferência de estruturas linguísticas funcionais da L1 na L2 dando como resultado inadequações que podem comprometer ou não a compreensão.

Ortográfica- fonológicas

Dificuldades do aprendiz para identificar fonemas da L2, estabelecendo em alguns casos correspondências gráficas com os fonemas da L1.

Léxico-semânticas

Transferência de vocábulos da L1 na L2 que podem comprometer ou não a compreensão.

Fonte: Acervo dos autores


No gráfico 1 abaixo, apresentamos uma análise qualitativa da classificação das estruturas híbridas dos textos em resposta à tarefa 3, Cadastro de Horta (BRASIL, 2016).


Gráfico 1 – Estruturas híbridas na tarefa 3


Fonte: Acervo dos autores


No texto de Paula, representativo de notas 4, foram encontrados um maior número ocorrência de estruturas híbridas dentro da categoria ortográfica e fonológica, quando comparado ao texto de João. Além da quantidade de inadequações nos dois textos, a diferença entre o texto nota 4 e nota 5 também poderia ser explicada pelas noções de língua dos avaliadores. Como esta foi a primeira tarefa corrigida, muitos avaliadores justificaram morfossintática suas notas com argumentos que apontavam para uma visão de língua/linguagem fortemente normativa. Ao longo do curso, os avaliadores puderam refletir sobre a coerência entre suas interpretações quanto às inadequações linguísticas e o construto de língua




operacionalizado no exame. Inadequações quanto ao cumprimento da tarefa também podem afetar a nota do texto de Paula, porém nos limitaremos às inadequações linguísticas.

Na tarefa 3, as estruturas híbridas quanto aos padrões ortográficos e fonológicos foram as mais frequentes nos textos de ambos os examinandos. Tais estruturas podem remeter a questões de identificação de fonemas ou interferência linguística como nas palavras “convivência”, “quera”, “historia”, “precissa” e “libre”. Além disso, no texto de Paula há uma presença maior de estruturas morfossintáticas claramente influenciadas pelo espanhol, o que justificou a nota 3-4 atribuída ao seu texto, segundo um dos avaliadores. Algumas estruturas híbridas que podemos mencionar dentro desta categoria são frases como: “desde o 2015”, “a cidade a qual facilita” e “ajudar aos”.

Não foram encontradas estruturas híbridas da categoria léxico-semânticas. Seria porque este tipo de estruturas são menos frequentes em níveis avançados, segundo as faixas de proficiência do Celpe-Bras? Outras pesquisas poderiam verificar esta hipótese.

Nas seguintes Tarefas 1 (BRASIL, 2016) e 2 (BRASIL, 2017), Paula obteve nota 4 e João, a nota 5. De acordo com os gráficos 2 e 3 abaixo, podemos notar novamente a grande discrepância entre os textos de João e Paula quanto ao número de inadequações classificadas como morfossintáticas e ortográficas e fonológicas, juntamente com as inadequações léxico- semânticas.


Gráfico 2 – Estruturas híbridas na tarefa 1


Fonte: Acervo dos autores




Gráfico 3 – Estruturas híbridas na tarefa 2


Fonte: Acervo dos autores


Quanto às estruturas híbridas categorizadas como léxico-semânticas, embora tenham sido identificadas nos textos, elas foram menos frequentes e a frequência foi semelhante na resposta à tarefa 2 de ambos os estudantes. Ao fazer uma análise qualitativa desta estrutura híbrida, apontamos que o grau de aceitação deste tipo de estruturas talvez possa depender do quão distante está o empréstimo, por exemplo: “barrios” e “bairros”; “Alemania” e “Alemanha” parecem ser mais toleráveis do que “compartir” e “compartilhar”.

Segundo os dados do gráfico 2, na Tarefa 2, houve um incremento das construções morfossintáticas do texto de João em relação a Tarefa 3, mesmo assim ele manteve a nota 5. Dentro dessas estruturas podemos citar: “a ser um local”, “onde pode-se” e “as vindas pelos imigrantes”. Por outro lado, destacamos que a estrutura híbrida morfossintática mais comum foi o uso inadequado do pronome relativo “a/o qual” no lugar de “que”, como nas seguintes frases: “a turma a qual vai”, “médios os quais demoram” e “o lugar o qual fica perto”.

Finalmente, dentro das Tarefas 1 e 2 foram encontrados estrutura híbridas ortográficas e fonológicas que causaram debates entre os avaliadores, como nas frases: “o museu já tive” ,”Hoje (eu) veio”, que podem ser explicados como o uso inadequado dos tempos verbais ou talvez pela dificuldade de identificar os sons [e] vs [i], no caso do verbo ter. Embora não pareçam ser uma estrutura híbrida, vale a pena indicar este dado, pois tais estruturas são muito frequentes nas produções de hispanofalantes que potencialmente poderiam ser classificados nas faixas avançado e avançado superior do exame.




Discussão dos resultados


De modo geral, podemos afirmar que o texto nota 4 apresenta um maior número de estruturas híbridas em relação ao texto nota 5. No contexto de análise deste trabalho, concluímos que a quantidade de estruturas híbridas pode justificar a variação entre notas 4 e 5. Apontamos, a partir dos dados, que a nota final pode ser influenciada mais pelo acúmulo de estruturas híbridas ou portunhol, do que por uma estrutura específica.

Não obstante, sobre as categorias de estruturas híbridas mencionadas, os resultados apontam que as morfossintáticas e ortográfica-fonológicas são as que mais poderiam influenciar a classificação de um texto na nota 4 ou nota 5. Como foram encontradas uma menor ocorrência de estruturas léxico-semânticas no corpus analisados, inferimos que tais formas possam não estar relacionadas ao desempenho correspondente aos examinandos avançados classificados no Celpe-Bras. Por outro lado, indicamos que as estruturas híbridas morfossintáticas possam ter graus de tolerância variáveis pelos avaliadores, uma vez que traduções literais do espanhol, podem comprometer a fluidez de leitura do texto.

Outro aspecto que poderia influenciar na atribuição de notas é o perfil do avaliador. Avaliadores com uma concepção de língua estruturalista tendem a justificar a nota ressaltando aspectos formais de uso da língua, penalizando com mais frequência o uso do portunhol, sem necessariamente interpretá-lo dentro de um contexto de interlocução proposto na tarefa. Os professores-avaliadores que tinham o espanhol como língua materna também tenderam a atribuir notas mais baixas, justificadas pelo uso do portunhol.


Considerações finais


Neste trabalho, pudemos resenhar variadas perspectivas sobre o papel da língua materna no processo de ensino e aprendizagem de línguas adicionais. Contrapondo-se ao normativismo linguístico, defensores do plurilinguismo declaram que a promoção de um modelo fixo e uniforme como referência para uso e estudo de língua adicional tem se mostrado ineficaz no âmbito da educação, pois ignora-se o potencial multifacetado da linguagem que pressupõe a existência de processos de hibridização, criatividade, identidade, identificação e subjetividade. Na perspectiva da translinguagem, o portunhol poderia ser uma manifestação do repertório linguístico que pode ser aproveitado e valorizado dentro de sala de aula, o que pressupõe aceitar a mistura de códigos linguísticos. Segundo os teóricos da intercompreensão, a língua materna é peça chave para o processo de ensino-aprendizagem. A tensão sobre a presença ou não da língua materna, na forma do que se classifica como portunhol, em respostas às tarefas escritas

Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, dez. 2021. e-ISSN: 2447-3529



de estudantes considerados avançados no sistema de avaliação do Celpe-Bras foi objeto de discussão deste trabalho.

Selecionamos e analisamos as estruturas híbridas de seis textos, sendo três representativos da nota 4 e, três, da nota 5. Pudemos constatar que a frequência em que é usado o portunhol pode influenciar a discrepância entre produções escritas com nota 4 e 5. Além disso, os principais resultados apontam que o acúmulo de determinadas estruturas híbridas, mais que a presença de uma forma híbrida específica, pode influenciar a maneira como o texto é avaliado. Quanto à classificação das estruturas híbridas, as que mais geraram divergências na interpretação da nota foram aquelas de natureza morfossintática e ortográfico-fonológica. As de natureza léxico-semântica, por sua vez, parecem não influenciar a decisão de avaliadores.

O estudo se limitou a discutir as manifestações do portunhol em um conjunto limitado de textos que correspondiam às faixas avançado e avançado-superior. Nesse sentido, vale a pena ressaltar que, por mais que o foco deste trabalho seja a presença do portunhol, o processo de avaliação de textos escritos do Celpe-Bras não se resume a interpretação destes tipos de inadequações linguísticas, pois existem outros parâmetros de avaliação que por limitação de espaço e escopo de análise não foram aprofundados neste trabalho.

Trata-se, portanto, de uma pesquisa exploratória. Outros trabalhos se fazem necessários para compreensão da interpretação do portunhol em textos escritos no contexto de exames. Aspectos como a relação entre a aceitação do portunhol a partir de situações mais ou menos informais de escrita ou quanto ao perfil dos avaliadores, especialmente, carecem de mais análise.


AGRADECIMENTOS: Agradecemos à Fundação Araucária pelo financiamento desta pesquisa.


REFERÊNCIAS


BACHMAN, L. A habilidade comunicativa de linguagem. Linguagem & Ensino, v. 6, n. 1, p. 77-128, 2003.


BARALO, M. La interlengua del hablante no nativo. In: SANCHEZ, J. et al. Vademécum para la formación de profesores. Enseñar español como segunda lengua. Madrid: Sociedad General Española de Librería, 2005. p. 369-390.


BRASIL. Ministério da Educação. Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira. Certificado de Proficiência em Língua Portuguesa para Estrangeiros. Brasília, DF, 2016. Disponível em:





https://download.inep.gov.br/outras_acoes/celpe_bras/legislacao/2016/edital_n1_de28012016

_celpe_Bras_2016.1.pdf. Acesso em: 4 set. 2020


BRASIL. Ministério da Educação. Secretaria de Avaliação da Educação Básica. Documento base do exame Celpe-Bras. Brasília, 2020.


BRASIL. Ministério da Educação. Secretaria de Avaliação da Educação Básica. Certificado de Proficiência em Língua Portuguesa para Estrangeiros: caderno de questões. Brasília, 2016/2017.


CANAGARAJAH, S. Translanguaging in the classroom: Emerging issues for research and pedagogy. Applied Linguistics Review, v. 2, p. 1-28, 2011.


DELLS, R. I.; SCARAMUCCI, M.; SCHLATER, M. A avaliação de proficiência em português língua estrangeira: o exame CELPE-Bras. Revista Brasileira de Linguística Aplicada, Belo Horizonte, v. 3, n. 1, p. 153-184, 2003.


ESCUDÉ, P.; OLMO, F. Intercompreensão a chave para as línguas. 2. ed. São Paulo: Editora Parábola, 2019.


FAULSTICH, E. O portunhol é uma interlíngua? In SEMINÁRIO DE LINGÜÍSTICA APLICADA, 1997, Barcelona. Anais [...]. Barcelona: Universitat Pompeu Fabra (UPF). p. 1- 15.


FULCHER, G. Testing second language speaking. Londres: Routledge, 2003.


JASPER, J. The transformative limits of translanguaging. Language & Communication, Elsevier, v. 58, p. 1-10, 2018.


SCARAMUCCI, M. O Projeto Celpe-Bras no âmbito do Mercosul: contribuições para uma definição de proficiência comunicativa. In: FILHO, J. A. (org.). Português para estrangeiros: interface com o espanhol. 2. ed. Campinas, SP: Pontes, 2001. p. 77-90.


TOSATTI, N. A produção escrita no exame Celpe-Bras. A funcionalidade dos gêneros.

SIELP, Uberlândia, v. 2, n. 1, 2012.


Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, dez. 2021. e-ISSN: 2447-3529



Como referenciar este artigo


RODRIGUEZ, D. G; FERREIRA, L. M. L. Portunhol e a Avaliação Escrita do Celpe-Bras. Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, dez. 2021. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v7iesp.6.15486


Submetido em: 25/08/2021

Revisões requeridas em: 15/09/2021 Aprovado em: 12/11/2021 Publicado em: 28/12/2021




PORTUNHOL AND THE CELPE-BRAS'S WRITTEN EXAM PORTUNHOL E A AVALIAÇÃO ESCRITA DO CELPE-BRAS PORTUÑOL Y LA EVALUACIÓN ESCRITA DEL CELPE-BRAS


Diego Giovanni RODRIGUEZ1 Laura Márcia Luiza FERREIRA2


ABSTRACT: The Portunhol can be approached from different perspectives. We aim to discuss its potentialities and its limitations in the large-scale assessment context, such as Celpe-Bras exam. First, we discuss how Portunhol can be related to different language teaching theoretical- methodological perspectives and also to Celpe-Bras exam´s theoretical constructs. The data that we analyze was collected in 2020 during an online preparatory course. For the analyses, we selected a set of six texts, which was related to high levels of proficiency. We investigated the scores attributed and the justifications given by the teachers-evaluators. According to the classification of hybrid structures found on our set of texts, we found out that the frequency of Portunhol could have influenced the scores variations. Besides, in the analyzed texts morph- syntactic and orthographic-phonological hybrid structures were more frequent than lexical- semantic structures.


KEYWORDS: Portunhol. Language teaching. Language assessment. Celpe-Bras.


RESUMO: O Portunhol pode ser abordado sob diversas perspectivas. É no contexto de ensino e avaliação de larga escala que queremos discutir suas potencialidades e limitações. Problematizamos as discussões sobre o Portunhol a partir de diversas perspectivas teórico- metodológicas para ensino de línguas próximas e dos construtos teóricos do exame Celpe- Bras. Apresentamos uma análise de dados que foram coletados em 2020 durante um curso online preparatório para o Celpe-Bras. Investigamos as notas atribuídas às tarefas escritas e as justificativas dadas pelos professores-avaliadores para selecionar um conjunto de seis textos, sendo três representativos da nota 4 e, três, da nota 5. A partir da classificação de estruturas híbridas dos textos selecionados constatamos que a frequência em que é usado o Portunhol pode ter influenciado a variação de notas. Estruturas de natureza morfossintática e ortográfico-fonológica foram mais frequentes nos textos analisados do que as de natureza léxico-semântica.


PALAVRAS-CHAVE: Portunhol. Ensino de línguas. Avaliação de línguas. Celpe-Bras.



1 University of Latin American Integration (UNILA), Foz do Iguaçu – PR – Brazil. Teacher of Spanish and Portuguese as Foreign Languages. ORCID: https://orcid.org/0000-0002-5330-0571. E-mail: dgv.rodriguez.2016@aluno.unila.edu.br

2 University of Latin American Integration (UNILA), Foz do Iguaçu – PR – Brazil. Student of Spanish and Portuguese as Foreign Languages. ORCID: http://orcid.org/0000-0001-7632-0834. E-mail: dgv.rodriguez.2016@aluno.unila.edu.br

Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, Dec. 2021. e-ISSN: 2447-3529



RESUMEN: El portuñol puede ser abordado desde diferentes perspectivas. Es en el contexto de enseñanza y evaluación de larga escala que discutiremos sus potencialidades y limitaciones. Problematizamos las discusiones sobre Portuñol, desde diferentes perspectivas teórico- metodológicas para la enseñanza de lenguas próximas y los constructos teóricos del examen Celpe-Bras. Presentamos un análisis de datos que fueron colectados en 2020 durante un curso en línea de preparación para el Celpe-Bras. Investigamos las calificaciones atribuidas a las tareas escritas y las justificaciones dadas por los docentes-evaluadores para seleccionar un conjunto de seis textos, trés representando nota 4 y tres representando nota 5. A partir de la clasificación de estructuras híbridas de los textos descubrimos que la frecuencia en la que se usa el portuñol puede haber influido en la variación de notas. Las estructuras morfosintácticas y ortográfico-fonológicas fueron más frecuentes en los textos analizados que las de naturaleza léxico-semántica.


PALABRAS CLAVE: Portuñol. Enseñanza de idiomas. Evaluación de idiomas. Celpe-Bras


Introduction


Portunhol can be defined as a hybrid language used in different contexts, such as border or frontier Portunhol, literary Portunhol, Portunhol in the classroom, among others. Faulstich (1997), when studying Portunhol in the border territory of Paraguay, Uruguay, and Argentina with Brazil, states that the concept of language has a sociopolitical and linguistic role and is related to the ideas of state and nation. In this ideological struggle, which intends to delimit and identify the geographical spaces, some positions see Portuguese border Portunhol sometimes as a confusing language and an affront to official languages, sometimes as a language that praises the border identity. In the context of teaching, tension also exists, because Portunhol can be seen as a threat to the discursive space of the standardized variants of Spanish or Portuguese.

Through Portunhol and the representation of writing close to speech, writers accentuate the complex and ambiguous construction of identity in border spaces, where two languages meet and disagree with their social representations. The use of Portunhol in literary productions claims the existence of those minorities who see it as a family and heritage language.

If, on the one hand, Portunhol is celebrated in literary productions and in everyday conversations on the border, in the school system, Portunhol affronts the maintenance of the myth of monolingualism. Frontier poet Fabián Severo says that the Uruguayan school system rejects Portunhol, as it is classified as “poor speech” or “incorrect”. Although Portunhol is part of the process of linguistic contacts and the construction of subjectivities, we wonder to what extent the use of the hybrid language may or may not be a sufficient linguistic resource.




Portunhol challenges the language notions of many teachers and evaluators, given its hybrid nature. Through this research, we propose to discuss how the interpretation of the grade to certify proficiency in Portuguese as an additional language can be influenced by the use of Portunhol. We will discuss the use of Portunhol in the written tests of the Certificate of Proficiency in Portuguese Language for Foreigners (Celpe-Bras).

Before approaching the evaluation, we present a discussion of theoretical- methodological proposals and their potential contributions to the teaching and evaluation of Portunhol. In the end, we present an analysis and a discussion about the interpretation of Portunhol in the context of the correction of written tasks of the Celpe-Bras exam.


Interlingua, teaching, and assessment of Portunhol


Baralo (2004) define the concept of interlanguage as a transitory state between L1 and L23 in which there are interferences from the L1 in the target language. Interlanguage studies have as main objective to understand the cognitive nature of second language acquisition. The fundamental question of interlanguage studies refers to how much L1 interferes or not with L2 acquisition. Such research had an impact on how second languages are taught and learned to reinforce the distance between the L1 and the L2 learning process. The relationship between L1 and the teaching and learning process has become a conflict. The scenario becomes even more complex when one adds, to the malaise of the relationship between mother tongues and additional languages in the teaching process, the power issues involved in language policies, and the prestige relations between languages.

Although the theoretical-methodological proposal of interlingua goes back to comparing two linguistic systems with a third intermediary linguistic system, called interlingua, some points explain its influence in the pedagogical scope. The idea of a standard language as a target language presupposes a reference model to explain and predict interferences and errors. The reference model adds to the predilection for structural education, which are still, unfortunately, the most widespread language construct in Brazilian educational systems. Furthermore, a teaching proposal supposedly based on interlanguage would position learners at a specific level of learning, based on the types of mistakes they make when speaking or writing so that teaching and learning tools are designed to help learners at each stage of interlingua. In



3The terms L1 (first language) and L2 (second language) are used to refer to the mother and foreign languages, respectively, in the context of the interlanguage debate. We will not discuss the differences between these concepts within the field of acquisition.

Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, Dec. 2021. e-ISSN: 2447-3529



the context of assessment, the classification of error by examinee's level still underpins language assessment exams, especially those that base the assessed items on interferences or inadequacies of specific linguistic forms. At Celpe-Bras, for example, linguistic resources are also evaluated, but their degree of adequacy in specific contexts of language use is evaluated.

We recognize the importance of contrastive studies, of error analysis, especially in the debate on second language acquisition, or interpretation on the accuracy of language use in the context of teaching and assessment, as is the case of this work. However, we consider it a limiting factor when the language teacher takes on the task of classifying errors in the teaching process when he should be encouraging the use of language and linguistic reflection in different contexts. Below we discuss two theoretical perspectives that welcome the presence of the L1 in the process of teaching and learning the second language in a less conflictive way, such as translanguaging (CANAGARAJAH, 2011; JASPERS, 2018) and intercomprehension (ESCUDÉ; OLMO, 2019).


Translingualism, teaching, and assessment of Portunhol


Translanguage emerges as a pedagogical approach for multilingual learners and assumes that languages should not be considered as separate systems. According to Canagarajah (2011), languages interact and form a unique linguistic repertoire, so that the individual can choose the necessary elements to communicate and be understood in each situation. The author points to the need to study the communication strategies employed by multilingual speakers, so that the focus becomes the engagement of individuals in social and discursive practices that promote the development of their linguistic repertoire, valuing the linguistic biographies of students. Language teaching would aim to negotiate meanings and uses. About the norm, Canagarajah (2011) advocates against the teaching of grammatical rules.

However, claiming that students' linguistic repertoire can always be used positively not reflect the social and cultural complexity of education. Regarding the limitations of the translanguage perspective, Jasper (2018, p. 18, our translation) warns “that even a school based on translanguage will likely require students to renounce certain parts of their linguistic repertoire (sexist and racist language, youth slang, jargon of game) to expand new skills (academic writing, for example)”. The author defends the politicization of the debate about what and why to learn and unlearn, since, for the social development of the individual, it is necessary to learn and develop new knowledge and tools and, at the same time, to reformulate or forget what is not considered “appropriate” culturally. Jasper (2018) argues that the linguistic




repertoire of learners must find a balance between the individual dimension and the collective dimension. This debate is directly related to the interpretation of Portunhol in evaluation contexts since Portunhol will not always be a sufficient resource.

From the perspective of translanguaging, we can say that the presence of Portunhol in the teaching, learning, and assessment process would no longer be a negative manifestation of an interlanguage, but part of a linguistic repertoire. However, if we think about the process of interaction in Portunhol, it is worth reflecting and debating its limits in certain situations of language use. About assessment processes, translanguaging is a perspective that challenges linguistic assessment instruments based on a language construct as structure, norms, and normative standards. It is worth mentioning here that the Celpe-Bras exam operationalizes the language construct in its assessment tasks to take into account various aspects of language use and not just the appropriate or inappropriate use of its forms. We can say that priority is given to the examinee's linguistic repertoire when assessing their performance when engaging in assessment tasks that simulate the use of language in the real world. However, we ask ourselves in this work if there is and where would be a balance between the individual dimension and the collective dimension of the use of Portunhol in this context of the written evaluation.


Intercomprehension, teaching, and assessment of Portunhol


Both the translanguaging and intercomprehension perspectives assume that it is natural for human beings to negotiate meanings when interacting in various languages. Both theoretical perspectives were dedicated to offering pedagogical guidelines for language teaching. Although translanguage and intercomprehension defend plurilingualism in school systems, each perspective proposes a didactic that welcomes other languages in the process of language education of students.

According to Escudé and Olmo (2019), intercomprehension refers to a phenomenon associated with observable communicative practices in which there is a spontaneous effort between speakers of different languages and varieties to understand each other. From the comparative historical studies of linguistic systems within and outside the group of Romance languages, intercomprehension emerges as a pedagogical approach based on the exploration of similarities and/or transparencies between languages.

Starting from the dynamics of identification and distancing, the authors claim that intercomprehension has the pedagogical objective of increasing the ability to understand between languages. Linguistic forms are part of the pedagogical process of language teaching,

Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, Dec. 2021. e-ISSN: 2447-3529



unlike what translanguaging proposes, by encouraging the cancellation of teaching with a focus on form. From the perspective of intercomprehension, the aim is to develop the perception of language traits, which belong to two or more languages, are similar and distant, promoting reflection on aspects of language that are or are not exclusive to a group of speakers and/or appropriate or not to an interaction context.

Although intercomprehension does not encourage Portunhol or mixtures of linguistic codes in the pedagogical scope, unlike translanguaging, the notions of intentionality and the right to imprecision are two major contributions mentioned by Escudé and Olmo (2019), which we can apply in the pedagogical and educational context. evaluation of Portunhol. The notion of intentionality is related to the interlocutor's attitude. To have intercomprehension, the interlocutors must intend to understand each other, in our case, it would be related to the evaluator's attitude towards the Portunhol. In the context of evaluation, it is important to relate intentionality to how much the belief system, the language conception, prejudices, and stereotypes of evaluators interact with the evaluation guidelines to facilitate or hinder the dialogue between evaluators and evaluators.

The notion of the right to imprecision is related to the degrees of acceptance of Portunhol. The right to print in the context of the assessment implies the choice not to describe the highest level of proficiency in exams, concerning the educated native speaker. The Celpe- Bras exam, for example, operationalizes the examinee's right to imprecision by accepting at the highest levels of proficiency a description of linguistic performance with inadequacies regarding the use of linguistic resources, for example.

Having made a brief theoretical review of different pedagogical perspectives that help us to think about the teaching and assessment of similar languages in the Portuguese-Spanish pair, we will discuss the Celpe-Bras exam.




Celpe-Bras


The Celpe-Bras exam was developed in the context of negotiations in the Southern Common Market (MERCOSUR) in the 1990s, together with the Argentine Spanish exam CELU (Certificado de Español: Lengua y Uso). On the relationship between the Celpe-Bras exam and the policies to promote the spoken language in Brazil, Dell' Isola and Scaramucci (2003) state that the consolidation of linguistic assessment parameters constitutes a political tool of the state that claims and delimits a language before the others. Although the delimitation of languages can harm the school environment, such as the tension over the use of Portunhol in border schools, the consolidation of a language certification instrument has a positive effect on the promotion of Portuguese spoken in Brazil by the Brazilian state in the context of soft-power educational and cultural policies.

Is it necessary to politicize the debate about when? It's because? Is the certification requirement adequate or not? We consider it pertinent, for example, that postgraduate candidates in Brazilian Higher Education Institutions, whose mother tongue is not Portuguese, prove able to attend classes and write academic papers in Portuguese. In many postgraduate selection notices, the certification issued by Celpe-Bras is mentioned as a possible document proving proficiency in Portuguese by speakers of other languages.

One of the characteristics of Celpe-Bras is its communicative nature. In the words of Bachman (2003, p. 84, our translation) communicative language skills can be defined as “knowledge, or competence, and the ability to implement or execute this competence in the communicative use of language, in an appropriate and contextualized way”. Scaramucci (2001) states that Celpe-Bras refers to a speaker of “full non-native proficiency”, that is, an individual who has complete operational control of the language, although lapses and inaccuracies can be observed. Depending on the degree of adequacy of the answers, the examinee can be classified into four proficiency bands, namely: intermediate, upper-intermediate, advanced, and upper- advanced.

In the context of Celpe-Bras, the interpretation of Portunhol implies several challenges. If we compare the communicative skills of Spanish-speaking examinees with those whose mother tongue is not Spanish, the Spanish-speaker will likely have more skills to negotiate meanings, although there may be inadequacies of linguistic resources and misunderstandings. About the potential and limits of Portunhol in the context of the exam, Scaramucci (2001, p. 89, our translation) asks:


Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, Dec. 2021. e-ISSN: 2447-3529



Is the consideration of the communication criterion that the exam is encouraging the manifestation of Portunhol? And can Portunhol meet the communicative needs of candidates? On the other hand, how not to go to the extreme requiring a native proficiency level as a proficiency parameter?


The teacher summarizes in the questions above the great dilemma present in several evaluation situations faced by both beginners and experienced evaluators. If, on the one hand, the fact that Portuguese and Spanish are closely related languages facilitates communication between speakers of both languages, on the other hand, to what extent is Portunhol an adequate resource? This is a fundamental question that generates a lot of debate in situations of discussion of grades, because, in the context of the assessment, exam formulators and evaluators are required to have a common assessment guideline to guarantee the reliability of the exam scores. However, even though there is a well-informed evaluation guideline through manuals, training, and scales, several performance evaluation specialists endorse the assumption that there is an interaction between task, evaluators, and scale (FULCHER, 2003). Therefore, we seek to discuss in this work, based on the experience in a preparatory course for the Celpe-Bras exam, what is appropriate or not regarding the use of Portunhol in different situations forged in the exam tasks through the analysis of texts.


Linguistic resources and Portunhol in the written assessment of Celpe-Bras


The written test is composed of four tasks that generate open answers that are evaluated by two or more evaluators, based on a common evaluation grid. The notes are independent and there is discrepancy control. As for the design of the written test, in task 1 the examinee must write a text after watching a video; in task 2, audio; and in tasks 3 and 4, read printed texts. To exemplify a command, we analyzed task 3 which was part of the 2016 test. 2.


You have a blog about healthy eating and, after reading the news about the registration of vegetable gardens in São Paulo, you decided to implement this initiative in your city. Write a text on your blog to ask readers to participate in the registration, explaining how the information collected works and how it works (BRASIL, 2016, p. 6, our translation).


The command above is accompanied by the reading text entitled “Looking for gardens in the heart of São Paulo” published in the lifestyle section of the Estadão news portal. The text is about an initiative called Wikiversity, which consists of collaboratively mapping vegetable gardens in urban areas through an online platform. In the command, the textual genre is defined: post; the interlocution: blogger writing for readers; as well as the purpose of writing: to ask

readers to participate in the registration of vegetable gardens in a defined city. It is also expected

Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, Dec. 2021. e-ISSN: 2447-3529



that the examinees recontextualize in their text information from the Estadão report regading the functioning and usefulness of the information that will be collected.

The definition of the situation and context of language use by the command directs how the items that make up the evaluation grid of written tasks will be interpreted. The adequacy regarding the dialogue in the proposed discursive genre, the performance of the requested action and the recontextualization of information are evaluated. The evaluation of the use of linguistic resources mobilized in the text depends on the fulfillment of the task.

The description of such items evaluated in the grid is done holistically, that is, there is a unique description for each grade from 0 to 5, which corresponds to the certification ranges provided for by the exam. In the descriptors of each track, the degree of adequacy of the response to the tasks concerning what was requested in the command is granulated. Regarding the evaluation of linguistic resources, the object of analysis of this work, a text grade 5 corresponds to “an autonomous, clear and cohesive text, in which the linguistic resources activated are appropriate to configure the dialogue relationship in the requested genre, and possible inadequacies rarely compromise the fluidity of reading” (BRASIL, 2020, p. 39, our translation). in the description of note 4, regarding the adequacy of linguistic resources, it is defined that “the linguistic resources activated are appropriate to configure the interlocution relationship in the proposed genre, building a clear and cohesive text in which possible inadequacies can compromise, in localized moments, the fluidity in reading.” (BRAZIL, 2020,

p. 39, our translation).

The linguistic resources are evaluated both in terms of a context of dialogue forged in the command of the tasks and the degree of impairment of reading fluidity due to possible linguistic inadequacies. We can infer that Portunhol in the written assessment of Celpe-Bras, although it is not the object of assessment of the test, is related to the use of linguistic resources. It is from how the examinee's answer can adequately establish the dialogue, the purpose and offer the pertinent information that the use of Portunhol will be interpreted to attribute a grade. Unlike exams focused on form, Tosatti (2012) reiterates that, in the context of assessing communicative skills, sociocultural, and situational factors for the production of speech are as important as grammatical rules.

The lack of definition about what is or is not considered, regarding the use of Portunhol, as an inadequacy that compromises or not the reading and understanding of the answer written by the examinees makes the evaluators, in the context of resolving discrepancies in grades, negotiate degrees of tolerance or leniency when assigning grades. to better discuss the issue, I cite two examples, the first being: “Já no final tem um espaço com milhes de sobrenomes

Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, Dec. 2021. e-ISSN: 2447-3529



gravados numa parede.” and the second: ““O importante é que a gente este mais saudável e ajudar para que outras pessoas conheçam este site.” In example 01, we interpret that the inadequacy in “milhes” does not compromise the global understanding of the idea, but in example 02 the inadequacies impair the understanding of the sentence because in place of the verb “esteja” is written “este”, which in Portuguese is a pronoun demonstrative “this boy”. In addition, the construction “o importante é que a gente (…) e ajudar...” makes understanding difficult, probably compromising the fluidity of the evaluator's reading of the text. That's why we present analysis and discussion below to elucidate the evaluators' degrees of tolerance towards Portunhol in the context of the written evaluation of Celpe-Bras.


Methodology


Data were collected during an online Celpe-Bras preparation course. The course was offered from August to October 2020. All twenty students active in the course had Spanish as their mother tongue. The teaching team consisted of eight undergraduate students in Letters. Of the eight, five were almost graduating and three had completed less than halfway through graduation. That is, the profile of the evaluators of this work is close to the profile of the evaluators of the exam. The delivery of four written assignments for the Celpe-Bras exam was part of the course's planning. The same text was independently corrected by more than one teacher. In pedagogical meetings, the discrepancy of the previously assigned grades was resolved individually and a final grade was negotiated between the team and the coordination. In this way, we tried to bring the context of the research closer to the situations of real assessment of the written tasks of the Celpe-Bras exam.

In this edition of the course, approximately 60 texts were delivered by the course participants and 200 notes were generated. Based on the experience during the process of monitoring the discrepancy meetings and analyzing the grades, we chose in this work to discuss the discrepancy of grades that varied between 4 and 5, which corresponds to the advanced and advanced-higher range, respectively. The variation of 1 point does not mean a significant discrepancy for revaluation of the grade (BRASIL, 2016). The choice is justified because we intend to gather information to discuss the possible characteristics of a Portunhol that is related both to the use of linguistic inadequacies that rarely compromise reading fluidity, as described in note 5, and to texts that, according to the description in note 4, they present inadequacies that compromise “in focused moments” the understanding or fluidity of the evaluator's reading.




This choice is also because, according to the evaluation grid, the items evaluated in tracks 4 and 5 that refer to discursive adequacy would be adequate, leaving the evaluator to judge more how the linguistic resources were used to fulfill the task or as the performance of the task than the performance of the task itself. We assume here that the examinee, whose grade ranged from 4 to 5, managed to establish the dialogue and fulfill the purpose of the task, we must inquire about what would be the manifestations of Portunhol or inadequacies that would be guiding the evaluators' interpretation to attribute grade 4 and 5. Next, we outline an answer to this question based on the data presented and analyzed.

As a data analysis tool, we opted for error analysis, based on contrastive studies to examine the linguistic interferences derived from the L1. Errors in the context of this work refer to hybrid constructions or linguistic inadequacies. The choice is justified by the need to systematize and organize the hybrid structures related to Portunhol present in the corpus.


Analysis and discussion


We analyzed the written production of two students, Paula and João, in three different tasks. Tasks 3 and 1 refer to the editions of the first semester of 2016 and, to tasks 2, of the second semester of 2017. The analysis cut is justified by the standard of the grades. When analyzing the scores independently assigned by three evaluators, we could see the variation between grades 4 and 5. It should be noted that, in most cases, the evaluators of Paula's texts justified grade 4 by arguing that linguistic inadequacies were frequent and that influenced the fluidity of reading the text. In table 1, below, the pattern of grades of Paula and João is organized, whose answers to the tasks will be analyzed next.




Table1 – Notes for tasks 3, 1, and 2


Course participants

Paula

João

Evaluators

Evaluator 1

Evaluator 2

Evaluator 3

Final grade

Evaluator 1

Evaluator two

Evaluator 3

Final grade

Notes attributed to task 3


4


3.5


3.4


3


5


5


5


5

Notes attributed to task 1


4


4


4


4


5


5


5


5

Notes attributed to task 2


4


4


4


4


5


5


5


5

Source: Authors' collection


Task 3 was the first to be done by the course participants, it is worth mentioning that in the first task the grade 5 was unanimous among the evaluators of João's text, while Paula's first grade was slightly lower than those assigned for her next productions. By analyzing the grades of the three tasks assigned by three independent evaluators, we can see little variation from the grade 4 assigned to Paula and from the grade 5, to João's texts. For this reason, we analyzed the set of six texts.

We categorized the hybrid structures in the texts and their frequency. Then we compare and discuss the data. The analysis was performed by a researcher, speaker of Spanish as a first language and Portuguese as an additional language, graduated in Letters, and reviewed by a Brazilian professor specialized in contrastive analysis.

In the following table, the most frequently found hybrid structures have been classified. In general, the structures referred to the influence of the mother tongue, in this case Spanish. In others, we could infer little mastery of textuality mechanisms. We identified and classified three categories, namely: Morphosyntactic (MS), Orthographic-phonological (EOF) and Lexicon- semantics (EL). In table 2, we present their definitions and examples.




Table 2 – Category of analysis of hybrid structures:


Categories of hybrid structures


Definition and examples

Morphosyntactic

Transfer of functional linguistic structures from L1 to L2 resulting in inadequacies that may or may not compromise comprehension.

Orthographic- phonological

Learner's difficulties to identify L2 phonemes, establishing in some cases graphic correspondences with L1 phonemes.

Lexicon-semantics

Transfer of words from L1 to L2 that may or may not compromise understanding.

Source: Authors' collection


In graph 1 below, we present a qualitative analysis of the classification of the hybrid structures of the texts in response to task 3, Cadastro de Horta (BRASIL, 2016).


Graphic1 – Hybrid structures in task 3


Source: Authors' collection


In Paula's text, representative of grades 4, a greater number of hybrid structures were found within the orthographic and phonological category, when compared to João's text. grade 5 could also be explained by the evaluators' notions of language. As this was the first task corrected, many evaluators morphosyntactically justified their scores with arguments that pointed to a strongly normative language/language view. Throughout the course, the evaluators were able to reflect on the consistency between their interpretations of linguistic inadequacies and the language construct operationalized in the exam. Inadequacies regarding the fulfillment of the task can also affect the grade of Paula's text, but we will limit ourselves to the linguistic inadequacies.



In task 3, hybrid structures in terms of orthographic and phonological patterns were the most frequent in the texts of both examinees. Such structures can refer to issues of phoneme identification or linguistic interference as in the words “coexistence”, “quera”, “historia”, “precissa” and “libre”. Furthermore, in Paula's text there is a greater presence of morphosyntactic structures clearly influenced by Spanish, which justified the grade 3-4 attributed to her text, according to one of the evaluators. Some hybrid structures that we can mention within this category are phrases such as: “desde o 2015”, “a cidade a qual facilita”, and “ajudar aos”.

No hybrid structures of the lexical-semantic category were found. Could it be because these types of structures are less frequent at advanced levels, according to the Celpe-Bras proficiency bands? Further research could verify this hypothesis.

In the following Tasks 1 (BRASIL, 2016) and 2 (BRASIL, 2017), Paula got a grade of 4 and João, a grade of 5. According to graphs 2 and 3 below, we can again notice the great discrepancy between João and Paula's texts regarding the number of inadequacies classified as morphosyntactic and orthographic and phonological, together with the lexical-semantic inadequacies.


Graphic 2 – Hybrid structures in task 1


Source: Authors' collection




Graphic 3 – Hybrid structures in task 2


Source: Authors' collection


Concerning hybrid structures categorized as lexical-semantic, although they were identified in the texts, they were less frequent and the frequency was similar in the response to task 2 of both students. By making a qualitative analysis of this hybrid structure, we point out that the degree of acceptance of this type of structure may depend on how far the loan is, for example: “barrios” and “bairros”; “Alemania” and “Alemanha” seem to be more tolerable than “compartir” and “compartilhar”.

According to the data in Graph 2, in Task 2, there was an increase in the morphosyntactic constructions of João's text of Task 3, even so, he kept the grade 5. Within these structures we can mention: “a ser um local”, “onde pode-se” and “as vindas pelos imigrantes”. On the other hand, we emphasize that the most common morphosyntactic hybrid structure was the inappropriate use of the relative pronoun “a/o qual” instead of “que”, as in the following sentences: “a turma a qual vai”, “médicos os quais demoram” and “o lugar o qual fica perto”.

Finally, within Tasks 1 and 2, hybrid orthographic and phonological structures were found that caused debates among the evaluators, as in the phrases: o museu já tive”, “Hoje (eu) veio”, which can be explained as the inappropriate use of the verb tenses or perhaps the difficulty of identifying the sounds [e] vs [i], in the case of the verb have. Although they do not seem to be a hybrid structure, it is worth mentioning this data, as such structures are very frequent in Spanish-speaking productions that could potentially be classified in the advanced and upper advanced ranges of the exam.




Discussion of results


In general, we can say that the text grade 4 presents a greater number of hybrid structures if we make a comparison to the text grade 5. In the context of analysis of this work, we conclude that the amount of hybrid structures can justify the variation between grades 4 and 5. From the data, it is possible to point out that the final grade can be influenced more by the accumulation of hybrid structures or Portunhol, than by a specific structure.

However, on the categories of hybrid structures mentioned, the results indicate that the morphosyntactics and spelling-phonological are the ones that could most influence the classification of a text in grade 4 or grade 5. As a lower occurrence of lexical-semantic structures in the analyzed corpus, we infer that such forms may not be related to the performance corresponding to advanced examinees classified in Celpe-Bras. On the other hand, we indicate that the morphosyntactic hybrid structures may have varying degrees of tolerance by the evaluators, since literal translations from Spanish can compromise the fluidity of reading the text.

Another aspect that could influence the attribution of grades is the evaluator's profile. Evaluators with a structuralist language conception tend to justify the grade by emphasizing formal aspects of language use, penalizing more often the use of Portunhol, without necessarily interpreting it within a context of interlocution proposed in the task. Teacher-evaluators who had Spanish as their mother tongue also tended to assign lower grades, justified by the use of Portunhol.


Final considerations


In this work, we were able to review different perspectives on the role of the mother tongue in the process of teaching and learning additional languages. In opposition to linguistic normativism, defenders of plurilingualism declare that the promotion of a fixed and uniform model as a reference for the use and study of additional language has been shown to be ineffective in the field of education, as it ignores the multifaceted potential of language that presupposes the existence of processes of hybridization, creativity, identity, identification and subjectivity. From the perspective of translanguaging, Portunhol could be a manifestation of the linguistic repertoire that can be used and valued within the classroom, which presupposes accepting the mixture of linguistic codes. According to intercomprehension theorists, the mother tongue is a key part of the teaching-learning process. The tension over the presence or absence of the mother tongue, in the form of what is classified as Portunhol, in responses to the

Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, Dec. 2021. e-ISSN: 2447-3529



written tasks of students considered advanced in the Celpe-Bras assessment system was the object of discussion in this work.

We selected and analyzed the hybrid structures of six texts, three representing grade 4 and three representing grade 5. We could see that the frequency in which Portunhol is used can influence the discrepancy between written productions with grade 4 and 5. The main results indicate that the accumulation of certain hybrid structures, more than the presence of a specific hybrid form, can influence the way the text is evaluated. Regarding the classification of hybrid structures, the ones that most generated divergences in the interpretation of the note were those of a morphosyntactic and orthographic-phonological nature. Those of a lexical-semantic nature, in turn, do not seem to influence the decision of evaluators.

The study was limited to discussing the manifestations of Portunhol in a limited set of texts that corresponded to the advanced and advanced-superior ranges. In this sense, it is worth noting that, although the focus of this work is the presence of Portunhol, the process of evaluating written texts by Celpe-Bras is not limited to the interpretation of these types of linguistic inadequacies, as there are other evaluation parameters. which, due to space limitations and scope of analysis, were not discussed in this work.

It is, therefore, an exploratory research. Other works are necessary to understand the interpretation of Portunhol in texts written in the context of exams. Aspects such as the relationship between the acceptance of Portunhol from more or less informal writing situations or the profile of the evaluators, in particular, need further analysis.


ACKNOWLEDGMENTS: We thank Fundação Araucária for funding this research.


REFERENCES


BACHMAN, L. A habilidade comunicativa de linguagem. Linguagem & Ensino, v. 6, n. 1, p. 77-128, 2003.


BARALO, M. La interlengua del hablante no nativo. In: SANCHEZ, J. et al. Vademécum para la formación de profesores. Enseñar español como segunda lengua. Madrid: Sociedad General Española de Librería, 2005. p. 369-390.


BRASIL. Ministério da Educação. Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira. Certificado de Proficiência em Língua Portuguesa para Estrangeiros. Brasília, DF, 2016. Available: https://download.inep.gov.br/outras_acoes/celpe_bras/legislacao/2016/edital_n1_de28012016

_celpe_Bras_2016.1.pdf. Access: 4 Sep. 2020




BRASIL. Ministério da Educação. Secretaria de Avaliação da Educação Básica. Documento base do exame Celpe-Bras. Brasília, 2020.


BRASIL. Ministério da Educação. Secretaria de Avaliação da Educação Básica. Certificado de Proficiência em Língua Portuguesa para Estrangeiros: caderno de questões. Brasília, 2016/2017.


CANAGARAJAH, S. Translanguaging in the classroom: Emerging issues for research and pedagogy. Applied Linguistics Review, v. 2, p. 1-28, 2011.


DELLS, R. I.; SCARAMUCCI, M.; SCHLATER, M. A avaliação de proficiência em português língua estrangeira: o exame CELPE-Bras. Revista Brasileira de Linguística Aplicada, Belo Horizonte, v. 3, n. 1, p. 153-184, 2003.


ESCUDÉ, P.; OLMO, F. Intercompreensão a chave para as línguas. 2. ed. São Paulo: Editora Parábola, 2019.


FAULSTICH, E. O Portunhol é uma interlíngua? In SEMINÁRIO DE LINGÜÍSTICA APLICADA, 1997, Barcelona. Anais [...]. Barcelona: Universitat Pompeu Fabra (UPF). p. 1- 15.


FULCHER, G. Testing second language speaking. Londres: Routledge, 2003.


JASPER, J. The transformative limits of translanguaging. Language & Communication, Elsevier, v. 58, p. 1-10, 2018.


SCARAMUCCI, M. O Projeto Celpe-Bras no âmbito do Mercosul: contribuições para uma definição de proficiência comunicativa. In: FILHO, J. A. (org.). Português para estrangeiros: interface com o espanhol. 2. ed. Campinas, SP: Pontes, 2001. p. 77-90.


TOSATTI, N. A produção escrita no exame Celpe-Bras. A funcionalidade dos gêneros.

SIELP, Uberlândia, v. 2, n. 1, 2012.


How to reference this article


RODRIGUEZ, D. G; FERREIRA, L. M. L. Portunhol and the Ceple-Bras written exam. Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, Dec. 2021. e-ISSN: 2447-3529 DOI: https://doi.org/10.29051/el.v7iesp.6.15486


Submitted: 25/08/2021 Revisions required: 15/09/2021 Approved: 12/11/2021 Published: 28/12/2021


Rev. EntreLínguas, Araraquara, v. 7, n. esp. 6, e021155, Dec. 2021. e-ISSN: 2447-3529