I truly believe that learning a second language is the best way for people around the world to share personal and cultural experiences, which can ultimately contribute to mutual understanding and respect. I work to promote the study of multiple languages and support educational projects with bilingual speakers of minority languages.  My current research focuses on the acquisition of a second language and bilingualism in minority contexts. I am also interested in developing and adapting computer-based technology for language instruction.

Current Research

The Wapichana language: description, bilingual acquisition, and language instruction: Wapichana is an Arawak language spoken in the state of Roraima (Brazil) and in Guyana. This research has three equally important goals: (i) describe the language, (ii) study bilingual acquisition by individuals who speak Wapichana, Portuguese and English, and (iii) provide support for local indigenous teachers in the implementation of classroom practices that can foster the linguistic development of bilingual speakers in their communities. My primary research partner (and MA advisee) in this project is Wendy Leandro from Aishalton (Guyana), who is a native speaker of the language. I am also collaborating with colleagues from the Federal University of Roraima.

After describing and testing recursive relative clauses and genitive constructions in Wapichana (see list of publications), I am now working on negation (with Manoel Gomes dos Santos) and on verb morphology (with Marcelo Giovannetti). Wendy Leandro and I are also working on a assessment project to evaluate lexical and grammatical knowledge of bilingual children and teenagers in the Serra da Lua region.

Together with the Wapichana teachers, Wendy Leandro and I are currently writing a Pedagogical Grammar of Wapichana, as part of a documentation project that I am supervising (see below). The grammar should come out in 2016. We have also started working on literacy materials and on establishing a language curriculum at the local schools. The Wapichana teachers from the Serra da Lua region that have been directly involved in the project are Benedita A. da Silva, Thomas Isaac, Wanja Sebastião, Nilzimara de Sousa Silva, Odamir de Oliveira and Atanásio de Souza. This research was initially funded by a Healy grant, and since 2013 it has received support from the Pedagogical Grammars Project described below.

Pedagogical grammars for Brazilian indigenous languages: GM-teams This is the second phase of the documentation project coordinated by the Museu do Índio (Rio de Janeiro) and sponsored by the Brazilian Government and UNESCO. The goal of the project is to produce pedagogical grammars in 5 different native Brazilian languages: Ikpeng, Karajá, Kawaiweté, Paresi and Wapichana. I am the primary consultant responsible for designing the pedagogical grammars, training the teams of linguists and language teachers, and supervising the production of the material. I also prepare workshops for language teachers on how to integrate pedagogical grammars into the language classroom. The project started in July 2013 and will end in August 2015. The grammars will be published and distributed by the Museu do Índio. The project is coordinated by Prof. Bruna Franchetto and Sonia Coqueiro.

Multiple grammars and bilingual acquisition: This research is done in partenship with Tom Roeper. After Amaral and Roeper (2014) presented a first multiple grammars model for L2 representation, I am now working on a descriptive model based on HPSG to explain how different linguistic features can coexist in bilingual grammars affecting what was originally called the "productivity of grammar rules". Andie Faber, one of my current PhD advisees, is testing some hypotheses in a series of experiments with gender agreement in L2 Spanish by native speakers of English and Brazilian Portuguese. Marcus Maia (UFRJ) is Andie's co-adviser and is supervising the processing experiments.

Previous Projects

tagarelaTAGARELA: TAGARELA is an intelligent computer-assisted language learning (ICALL) system that can be viewed as an intelligent electronic workbook that provides  students with opportunities to practice their reading, listening, and writing skills. Because it is a web-based system, it can be used anywhere there is a computer with internet access. TAGARELA uses Natural Language Processing technology to analyze students' input, and detect spelling, morphological, syntactic and semantic errors.
Research Partners: Detmar Meurers and Ramon Ziai (University of Tübingen)

LangBot: LangBot is a project proposed and designed by Scott Payne, a colleague from Amherst College. LangBot is an instant-messenger (IM) computer program that enables students to have queries about various foreign languages answered in real-time, online. The tool that operates as a human IM "buddy" searches various language web sites and a specially developed corpus of language to respond to student queries. The project is funded by the US Department of Education.