Lingua最新文献

This article is concerned with the speech act of offers in educated Nigerian English within a variational pragmatics approach. In particular, this study explores the offer strategies chosen by Nigerian English speakers and the effect the speakers’ and hearers’ social status and social distance, the type of offer and formality of the context might have on them. A total of 325 Nigerian respondents filled in questionnaires containing 13 discourse completion tasks varying in these variables. 782 valid responses were analysed according to the structure of the communicative act, the offer superstrategy and substrategy used as well as the use of multiple languages. Results show that the variables context, social status, social distance and offer type all influence the linguistic form of offers. Nigerian English speakers use little code-switching and differ systematically from speakers of other varieties of English in terms of using more directives.

In many languages, female and male names have different phonotactic characteristics. The name–gender relationship is probabilistic; therefore, it can be captured more adequately using stochastic models than deterministic phonological theories. In this study, a total of 6,000 most commonly used names (3,000 for each gender) in Korean were used to train a deep neural network (DNN), which is an ensemble model of recurrent neural networks and convolution neural networks. The phonotactic learner (PL) was used as the baseline model. The DNN and PL models predicted the gender of 50 test names compiled from low-frequency names. The models’ predictions were compared with human judgments on the gender of the test names. The models’ predicted labels matched the names’ actual labels, with a higher accuracy in the DNN (90%) than in the PL (76%). The predictions also matched the labels assigned by human subjects with a higher accuracy for the DNN (86%) than the PL (72%). The DNN model correlated more closely with human judgments (r² = 0.743) than the PL (r² = 0.312). Considering the similarity of responses between the DNN and humans, these results suggest that neural network models should be incorporated into phonological studies.

This article investigates how selected English formulaic sequences, functioning as situation manipulators (SMs) used for the expression of directive stance and belonging to the category of lexical bundles, are translated into Polish. Using parallel and monolingual reference corpora, the study aims to analyse the Polish equivalents of the specific set of SMs found in European parliamentary debates to verify their regularity of use, native-like character, discourse functions and – in selected cases – also their level of formulaicity in contemporary target-language texts. The results show a high degree of variation in the forms, frequencies and the overall currency of the obtained equivalents in native Polish texts. It was found, however, that the manipulative impact of individual SMs was modified rather than entirely changed in translation. The present study demonstrates that the choice of the specific English-to-Polish translation of multi-word items expressing directive stance is subject to the translator’s own judgment of what is equivalent in the specific context, which does not necessarily correspond to what is typical and regular of target-language use.

Adult heritage speakers (HSs) of Spanish show Subject-Verb (SV)/Verb-Subject (VS) word-order variation in unaccusative sentences marking sentence focus, while monolingual speakers favor VS, as in Llegó Sara ‘Sara arrived’. Yet, few empirical studies have explored the distributional rates and patterns associated with this word-order variation in perception and production among HSs. We examine the variable-rule system underlying HSs’ acceptability judgments and written-narrative production to articulate how their choices between SV and VS are constrained. We begin with an alternative argument to the Unaccusativity Hypothesis, that unaccusatives instantiating sentence focus are instances of locative inversion where the preverbal position is occupied by an explicit or silent spatiotemporal argument (stage topic) licensing VS order. The results of two context-rich, novel experiments revealed two properties that contributed to HSs’ use of VS order: an explicit stage topic and a subject longer than four words. If these were not realized, HSs became increasingly inclined towards SV order. Assuming that interpreting a silent stage topic or a short subject taxes the cognitive resources required to retrieve the relevant discourse-pragmatic information, our findings support recent acquisition theories that attribute HSs’ divergent patterns to processing costs rather than the traditional view based on cross-linguistic influence.

This study explores various issues related to negative PPs, focusing on (i) why some negative PPs trigger negative inversion, whereas some do not, (ii) why negative phrases can have scope beyond their c-domain, and (iii) why the distribution of negative PPs is restricted. This study is divided into two parts. The first part demonstrates that negative phrases exhibit a pattern similar to wh-phrases. They cooccur with a zero morpheme NEG_ø, which carries out the function of extending their restriction. Just as Q extends the restriction of a wh-operator and triggers pied-ping, NEG_ø extends the restriction of a negative operator, enabling it to have scope beyond its c-domain. The second part argues that English employs covert negative concord in that in negative sentences T must be merged with a covert head called Polarity, which has the feature [u(ninterpretable) Neg]. Additionally, it suggests that (i) the [uNeg]-feature on Polarity can be deleted via a mutual c-command relationship with a NegP, and (ii) when the [uNeg]-feature on it cannot be deleted in situ, it triggers head movement as a last resort. This claim provides a principled account for negative inversion and the distribution of negative PPs.

Training of a foreign language speech sound pair may lead to the formation of new memory traces in young adults who are not very experienced in the target language, as well as to the strengthening of existing memory traces in advanced target language students. We used listen-and-repeat training to test whether previously formed memory traces exist approximately a year later. Further, we compared these learning results with those obtained from advanced target language students. Both groups participated in a recording that was identical to the one that the less experienced learners had already completed a year before, but only once and without training. The experiments included EEG recordings measuring mismatch negativity (MMN) and N1, as well as listening tests. The less experienced learners’ MMN responses were similar to those of advanced students, and their N1 amplitude had increased close to that of the advanced students. These results suggest long-lasting speech memory traces resulting from a brief training and the development of increased sensitivity.

This study addresses linguistic and cross-disciplinary issues connected with the study of personality trait words, e.g., shy, selfish, outgoing. Although the personality lexicon has long been seen as a primary concern for the field of psychology, it is argued here that the study of personality words belongs primarily within the domain of linguistics and can be performed by methods of lexical semantics and corpus analysis. To demonstrate this, two English personality adjectives representative of the personality dimension Extraversion (namely, shy and outgoing) are analysed using a corpus-assisted Natural Semantic Metalanguage (NSM) approach. The results highlight the status of the English words as folk concepts. The risks of excessive reliance on such folk concepts in personality research are demonstrated by the brief comparison with Russian, and alternative ways forward are outlined.

In this paper, I explore the syntax of two quirky Albanian subject constructions, both containing a dative experiencer that surfaces in subject position and behaves like a subject even if it does not induce agreement with the verb, along with an object bearing the nominative case, triggering agreement on the verb. The first type of quirky subject construction is characterised by the restriction of verb agreement to the third person, while the second type does not show person restrictions. The partial agreement data discussed here can be accounted for by resorting to the Person-Case Constraint. A split-feature checking analysis is conducted, whereby the person feature is checked separately from the number feature in contexts where a dative and a nominative DP are associated with two different features of a unique single probe (Tense).

Copredication is the phenomenon that two or more predicates can jointly apply to the same argument. In “The book is heavy and informative” the word book seems to refer to both a concrete physical object that can be heavy and an abstract content that can be informative. It has been observed that if the concrete sense of the nominal is triggered first, the copredication statement often sounds better compared to when the abstract sense is triggered first. However, the cognitive underpinnings of this effect are not well understood. In this theoretically oriented paper, we propose a predictive processing model of order effects aimed at advancing our understanding of that phenomenon. We also connect the debate regarding ordering preferences with an existing strand of research on ordering preferences in multi-adjective strings and the information structure of sentences.

The present study describes the nature of the dynamic spectral changes of Spanish vowels /a e o/ produced by 10 native Spanish (L1) speakers and 13 native English speakers learning Spanish as a second language (L2). The investigation seeks to identify variations in formant change through the application of both two- and three-point measures, evaluating their efficacy in capturing the formant dynamics of Spanish vowels. F1 and F2 formant dynamic variations were assessed in a set of acoustic measures, including Vector Length (VL), Trajectory Length (TL) and Formant Velocity (FV). Results show that L1 Spanish vowels are more dynamic than L2 Spanish vowels produced by native English speakers, challenging the traditional view that Spanish vowels are rather static. Moreover, the findings indicate that two- and three-point measures yield similar results for most vowels regarding the amount of formant change, rendering them ineffective in accurately describing Spanish vowels. The main difference between L1 and L2 Spanish vowels lies not in the magnitude of formant change, but in the direction of such change. Consequently, a measure that incorporates the direction of formant movement, such as FV, emerges as a more suitable acoustic characterization for both L1 and L2 Spanish vowels.