When describing motion events, English encodes Manner of motion in the verb and Path of motion in a satellite (s-framing). Brazilian Portuguese (BP), however, encodes Path in the verb and elaborates Manner adverbially (v-framing). This study investigates at what stages of L2 proficiency L2BP and English learners’ acceptability ratings converge with those of L1 speakers when rating sentences with Manner elaborated in the verb (Manner-verb) or in an adverb: a prepositional phrase (Manner-prep) or subordinate clause (Manner-AdvClause). Participants (n = 176) consisted of L1/L2English and L1/L2BP speakers. L2ers were grouped according to language proficiency (Elementary, Intermediate and Advanced). Results of ordinal logistic regressions show that Intermediate proficiency is associated with Manner-verb (L2BP) and Manner-prep ratings (L2English), and that Advanced proficiency is associated with Manner-verb (L2English) and Manner-AdvClause judgments (L2English and L2BP). These findings contribute to the limited body of work on the acquisition of v-framed L2s and the development of low-proficiency learners.

This paper provides data resources for low-resource hate speech detection. Specifically, we introduce two different data resources: (i) the HateBR 2.0 corpus, which is composed of 7,000 comments extracted from Brazilian politicians’ accounts on Instagram and manually annotated a binary class (offensive versus non-offensive) and hate speech targets. It consists of an updated version of the HateBR corpus, in which highly similar and one-word comments were replaced; and (ii) the multilingual offensive lexicon (MOL), which consists of 1,000 explicit and implicit terms and expressions annotated with context information. The lexicon also comprises native-speaker translations and its cultural adaptations in English, Spanish, French, German, and Turkish. Both corpus and lexicon were annotated by three different experts and achieved high inter-annotator agreement. Lastly, we implemented baseline experiments on the proposed data resources. Results demonstrate the reliability of data outperforming baseline dataset results in Portuguese, besides presenting promising results for hate speech detection in different languages.

This paper presents an audiovisual perceptual analysis of the wh-question and wh-exclamation intonation in Brazilian Portuguese using auditory–visual congruent and incongruent stimuli, to investigate the relative importance of each modality in signaling pragmatic meanings. Ten Brazilian Portuguese speakers (five female) were filmed while producing both speech acts 10 times. Next, artificial stimuli were created: audio and visual cues were either matched (audio and video from the same speech act) or mismatched (audio and video from the different speech acts), resulting in 10 congruent and 10 incongruent stimuli of the wh-questions and the wh-exclamations. The perceptual experiment was taken by 36 Brazilians who identified the stimulus as a question or an exclamation. Results from the logistic regression showed that the factor ‘congruence’ was significant and had a significant interaction with ‘speakers’, which means that the congruent stimuli increased the comprehension of the Brazilian Portuguese wh-questions and wh-exclamations. In contrast, the incongruent stimuli tended to lower listeners’ identification, but to a degree depending on individual speakers’ strategies. Although variation in the accuracy of expressing both speech acts was also found across speakers, this study corroborates that the visual channel impacts the perceptual identification of the pragmatic intonation function of distinguishing sentence mode.

In this paper we present an original approach to analyze the compositionality of indefinite expressions in Romance by investigating the relevance of their syntactic distribution in relation to their meaning. This approach has the advantage of allowing us to explore the question of how syntactic structure can determine the meaning of different forms of indefiniteness. To that end, we postulate a common derivation for bare plurals, bare mass and de phrases, whereby an abstract operator de is adjoined to definite determiners and shifts entities into property-type expressions. Quantificational specificity is proposed to be derived from a syntactic structure in which weak quantifiers select for indefinite de-phrases, no matter whether de is overt at Spell-Out or not; these quantifiers turn properties into generalized quantifiers. The anti-specificity meaning of some indefinites is derived by adjoining in the syntactic structure an abstract operator alg that encodes the speaker’s epistemic state of ignorance to a quantifier encoded for specificity, and it turns a generalized quantifier into a modified generalized quantifier. The paper also brings some general predictions on how indefiniteness is expressed in Romance, as it provides extensive support from five Romance languages: Brazilian Portuguese, Catalan, French, Italian and Spanish.

Sustained interaction between a bilingual's two languages can be a first step toward diachronic language change. We describe two investigations that explore this by examining how bilinguals process innovative syntactic structures in their first language. In the first investigation, a sentence recall/sentence matching task, bilinguals and monolinguals exhibited differences in their tolerance of expressions of induced motion, which vary in acceptability between the two languages (Portuguese and English). In the second investigation, a priming methodology was employed to induce bilinguals to produce in their first language (Spanish) innovative constructions modeled on the second language (English), using materials where the alternation is shared between the two languages (voice, reciprocal) or not (dative). The two investigations provide a window into how languages interact in bilinguals, inducing tolerance of ungrammaticality which, we will argue, could lead to long-term novel representations in the linguistic competence repositories.