Corpus-based Studies Research Articles

Placement patterns of English and French conjunctive adjuncts of contrast

AbstractIn many languages, conjunctive adjuncts (e.g.however, therefore) are syntactically mobile. Several corpus-based contrastive studies have shown that languages differ in the positions that they tend to prefer for conjunctive adjuncts. However, the studies available have formulated general cross-linguistic differences in placement for languages as wholes, without considering the possibility that such contrasts may be influenced by register. The objective of this paper is to investigate and compare the placement patterns of English and French conjunctive adjuncts of contrast in two written registers (viz. editorials and research articles) in order to measure the impact of register variation on the differences between these two languages. The results suggest that, although register variation plays a significant role on placement within each language system, language is a better predictor of placement than register, since cross-linguistic differences in placement between English and French are stable across communicative situations. In a second stage, the results obtained in the comparable corpus study are complemented with the analysis of translation data, with a view to assessing the translators’ degree of awareness of the inherent word order preferences of the target language. The study is grounded in the framework of Systemic Functional Linguistics and relies on the notions of Theme and Rheme to describe conjunctive adjunct placement.

Introduction

Introduction

Agreement on emotion labels' frequency in eight Spanish linguistic areas

Various traditions have investigated the relationship between emotion and language. For the basic emotions view, emotional prototypes are lexically sedimented in language, evidenced in cultural convergence in emotional recognition and expression tasks. For constructionist theories, conceptual knowledge supported by language is at the core of emotions. Understanding emotion words is embedded in various interrelated constructs such as emotional intelligence, emotion knowledge or emotion differentiation, and is related to, but different from, general vocabulary. A clear advantage of Emotion Vocabulary over most emotion-related constructs is that it can be measured objectively. In two successive corpus-based studies, we tested the predictions of concordance and absolute agreement on the frequency of use of a total of 100 Spanish emotion labels in the eight main Spanish-speaking areas: Spain, Mexico-Central America, River Plate, Continental Caribbean, Andean, Antilles, Chilean, and the United States. In both studies, the intraclass correlation coefficient was statistically different from the null and very large, over .95, as was the Kendall's concordance coefficient, indicating broad consensus among the Spanish linguistic areas. From an applied perspective, our results provide supporting evidence for the similarity in frequency, and therefore cross-cultural generalizability regarding familiarity of the 100 emotion labels as item stems or as experimental stimuli without going through a process of additional adaptation. On a broader scope, these results add evidence on the role of language for emotion theories. In this regard, countries and regions compared here share the same Spanish language, but differ in several aspects in history, culture, and socio-economic structure.

Mandative subjunctive versus should in world Englishes: a new take on an old alternation

This study explores the alternation between the mandative subjunctive and its modal alternative with should across native and non-native Englishes. Methodologically, we try to improve on existing standards by investigating over 3,300 occurrences of the alternation from the Corpus of Web-based Global English and annotated for a range of linguistic factors analysed with a forest of conditional inference trees; also, we are exemplifying a new strategy for the use of random or conditional inference forests in corpus-based alternation studies. We obtain a forest with significant prediction accuracies and a good C-score and discuss the strongest predictors of the subjunctive versus should alternation across Englishes. Contrasting with existing research, our multi-factorial results: ( i) suggest that in British English the mandative subjunctive may not be dying out as much as we thought; and ( ii) individual suasive verbs influence speakers' use of the two variants more than their variety of English.

The trends and developments of L2 Japanese research in the 2010s

AbstractThis article updates the trends and developments of Japanese as a second language (JSL) research since Mori and Mori (2011) by reviewing nearly 200 selected empirical studies published in English or Japanese between 2010 and early 2019. The first section of this review examines the cognitive aspects of second language (L2) Japanese development, focusing on vocabulary and kanji (i.e., Chinese characters transferred into Japanese) learning, syntactic development, and the issues surrounding reading and writing. The second part investigates sociocultural issues in L2 Japanese development and use, including pragmatic development, multilingual/translingual perspectives, and multilingual youth. The third section explores the role of affective variables (primarily learner motivation) in L2 Japanese development. The last section considers various pedagogical issues, including corpus-based studies and the development of materials and resources, instructional designs and classroom studies, technology-enhanced language learning and new instructional models, critical approaches to teaching Japanese and content-based instruction. The review of these themes illuminates trends and emerging areas of interest in post-2010 L2 Japanese research inspired by current developments in applied linguistics and second language acquisition (SLA) research, as well as unique features of Japanese language and sociocultural contexts.

Default sarcastic interpretations of attenuated and intensified similes

This study tests the Defaultness Hypothesis (Giora et al., 2015b, 2018), aiming to show that, “fully abstract phrasal patterns” (Goldberg, 2003, 2006), strongly attenuating (i.e., weakening) highly positive concepts, generates sarcastic interpretations by default – i.e., automatically, without relying on supportive context or internal incongruity. According to the Defaultness Hypothesis, for an interpretation, activated automatically, to be considered as default, it has to be novel, free of internal incongruity, and free of contextual support.The results of our studies support the prediction of the Defaultness Hypothesis, showing that strongly attenuated highly positive concepts (of novel, isolated similes, involving no internal incongruity) generates sarcastic interpretations by default.Additionally, our experimental results diverge from previous corpus-based findings, in which attenuated similes, containing negative internal incongruity, were found to be more sarcastic than their non-attenuated counterparts (Veale, 2013). Given that Veale's corpus-based studies relied heavily on items' internal incongruity, his findings cannot be considered default interpretations. However, our experiments, involving items free of internal incongruity do reveal a significant difference between the attenuated and non-attenuated similes. We conclude that, as predicted by the Defaultness Hypothesis, it is attenuating highly positive concepts that affects sarcasm by default, i.e., even without relying on internal incongruity.

Introducing corpus-based translation studies

Introducing corpus-based translation studies

A corpus-based analysis of Post-Auxiliary Ellipsis voice mismatches in Late Modern English

ABSTRACT This paper analyses Post-Auxiliary Ellipsis voice mismatches between the antecedent clause(s) and the ellipsis site(s) in Late Modern English, using the Penn Parsed Corpus of Modern British English (PPCMBE) (1700–1914). This study focuses on two subtypes of Post-Auxiliary Ellipsis, namely VP ellipsis and Pseudogapping. The results show that voice mismatches were possible in Pseudogapping and VP ellipsis in Late Modern English with low frequencies. This fact serves as counterevidence for the claim about the impossibility of finding voice mismatches in Pseudogapping and confirms corpus-based findings for Present-Day English. As for VP ellipsis, corpus-based studies show that voice mismatches are not attested in Present-Day English. Since they occur in Late Modern English with low frequencies, this contrast may be due to the stylistics or register of the corpora analysed.

Impersonal and reflexive uses of Middle English psych verbs under contact influence with Old French

AbstractThis article discusses the relation between impersonal constructions and the reflexive use of psych verbs of theadmire-type (Levin 1993) copied from Old French to Middle English. The outset of the study is an observation made by van der Gaaf (1904) which was commented on by Fischer (1992) concerning the reflexive use of verbs likeremembren(Old Frenchremembrer) in the course of the development of impersonal constructions to personal constructions. My aim is to answer the following questions: First, was the reflexive use of verbs copied from Old French new to the Middle English system? Second, what is the syntactic and semantic relation between ME impersonal and OF reflexive constructions? I will first examine the reflexive construction in Old English and Old French to answer the first question. I will then deal with the second question taking a closer look at both constructions in Middle English. Supported by lexicon-based and corpus-based small-scale studies of the psych verbremembrenI will provide an analysis of the syntactic and semantic properties of the verb in both constructions and show that what they have in common is an intransitive structure where the subject is anexperiencer, and middle semantics. I will finally address possible effects of language contact through translations, before I conclude.

Translating English Clefts and Pseudoclefts into Arabic A Parallel Corpus-Based Study

Translation practice, theorizing, as well as the long-held preoccupation with equivalence have all recently undergone cutting-edge changes by the development of corpus linguistics by the mid-nineties. Of course, corpus-based studies are not a novelty for translators; translators used to manually explore a parallel corpus (a ST and a TT) for documentation. Before the introduction of electronic data and software, translators had to extensively read through the texts in order to be able to extract any relevant information (O'Keeffe and McCarthy, 2010: 502). It was a tedious and tiring job even with relatively short stretches of texts.

Drawing areal information from a corpus of noisy dialect data

AbstractThis article is an analysis of linguistic survey data representing German dialects in Switzerland in 1933/34 based on the so-called Wenker sentences. The data are impressionistic in terms of applied phonetic transcriptions, which were produced by non-specialists using the Latin alphabet. Due to the lack of pre-defined standardization, the phonetic transcriptions are very heterogeneous. From a technical perspective, this leads to very noisy data, which is why the validity of the Wenker data in general and the Swiss Wenker data in particular has been questioned. Using methods from computational linguistics, we compare, for the first time, Wenker data with linguistic data collected at virtually the same time by linguistics professionals. Direct comparison with a sample from the published atlas of German-speaking Switzerland (SDS) reveals that despite the noisiness of the data, they nevertheless provide reliable information, e.g., in terms of the spatial structuring of Swiss dialects. The study is thus a successful pilot for other corpus-based studies dealing with unstructured Wenker data in other regions.

Onde podemos encontrar a impressão digital da legendista? Um estudo sobre o emprego da colocação pronominal e de marcadores discursivos em legendas de Star Trek: Enterprise

Este artigo apresenta alguns resultados de uma pesquisa de mestrado que objetivou caracterizar o estilo de uma legendista. O foco deu-se na forma que ela empregou a colocação pronominal e marcadores discursivos (MDs) em legendas em Português Brasileiro, feitas para episódios da série de TV Star Trek: Enterprise. A investigação afiliou-se metodologicamente aos Estudos da Tradução Baseados em Corpus, principalmente por meio do auxílio de dois programas do WordSmith Tools©, 7, o WordList e o Concord. O corpus de estudo se constitui por arquivos de legendas disponíveis na Netflix, em português brasileiro, dos episódios 10 e 11 (temporada 1) e 16 e 24 (temporada 2), enquanto o corpus de referência consiste em outros arquivos de legendas das mesmas temporadas da série feitos por outros legendistas. Os resultados mostram que a impressão digital da legendista em estudo pode ser encontrada por sua preferência pela ênclise e por MDs mais facilmente aceitos nos guias normativos de legendagem da Netflix. Sendo assim, nota-se seu uso variado de registros linguísticos, especialmente do registro formal da língua portuguesa.Palavras-chave: Impressão digital da legendista. Registros linguísticos. Colocação pronominal. Marcadores discursivos.

Community and authority in ROAR Magazine

This article responds to a common critique of corpus-based studies as decontextualized exercises in linguistic analysis by illustrating how, in the case of internet-based data, the concordance line can reveal rather than obscure aspects of a textual body’s cultural constitution. The data for the study consists of 100 articles of the online political journal ROAR (Reflections on a Revolution) Magazine, which has reported on global instances of public unrest and dissent since 2011. After sketching the relation between the financial crisis commencing in 2007 and the global protests that followed in its wake, the article investigates textual patterns within ROAR’s varied output. These patterns, ranging from the collocational profile of the keyword democracy to quotation practices, are shown to be constitutive of a virtual sense of community. This process of identity formation is then shown to have a mythopoetic effect, which ultimately impacts the emplotment of the various events covered and considered by the magazine. Additional attention is paid to ROAR as a cross-platform enterprise. In this respect, the fragmentary nature of the Internet is shown to both facilitate and frustrate the creation of a symbolic sense of community.

The Multilingual Student Translation corpus: a resource for translation teaching and research

The Multilingual Student Translation (MUST) corpus is a corpus of translations produced by foreign language learners or trainee translators collected collaboratively by a large number of partner teams internationally. The corpus represents a prime example of community sourcing, as the data are collected and shared by the members of the MUST network. Two key characteristics of the corpus are that it involves a large number of language pairs and that each text is accompanied by a rich set of standardized metadata related to the source texts, the translation tasks and the students. The web interface on which the corpus is stored allows the data to be aligned and annotated with a purpose-built translation annotation system. The resulting corpus data lend themselves to a range of applications (translator training, materials design, pedagogical lexicography) and can also be used to advance empirical research in corpus-based translation studies.

A corpus study of the diachronic development of Chinese patent terminologies

ABSTRACT Corpus-based approaches to the study of language and the compilation of a corpus are increasing at an unprecedented rate in recent years. The majority of the corpus was compiled from newspaper articles or by combining literatures with non-literatures. Corpus in specialized fields is mostly related to legal language, with very few monolingual or parallel corpora in the technical field, not to mention diachronic corpus. The building of a corpus and corpus-based studies facilitates terminologists, translators, researchers, and teachers in specialized English a basis in their analysis and interpretation of specialized terminologies. Diachronic corpus helps us understand the development of language and explores the changes and uses of specialized language in a specific field, in order to provide terminologists and translators a reference when defining specific terms. This study compiles a diachronic corpus of Chinese patent claims and abstracts to explore the linguistic features of the Chinese patent terminologies, the development, and variations of patent terminologies. With the diachronic corpus of Chinese patent claims, the development of patent terminologies can be analyzed. The patent is a strong indicator of the economic power of a country. The authentic material collected in the patent corpus provides representative research data, which, in turn, increases our understanding of the use of terminologies of patents.

Statistical Machine Translation Customization between Turkish and 11 Languages

Statistical Machine Translation (SMT) has been the dominant corpus-based machine translation (MT) approach in the last twenty years. While SMT has been studied in detail among European languages, it has not been studied sufficiently in language pairs including Turkish as source or target language, and its study has been limited mostly to English ↔ Turkish language pair. This study aims to broaden the perspective on Turkish corpus-based MT studies by training MT engines between Turkish and a wide variety of languages with different features. It surveys customized SMT between Turkish and 11 different languages. Twenty-two SMT engines have been trained in KantanMT with open parallel corpora using Turkish as both source and target language. Three automatic evaluation metrics F-Measure, BLEU, and TER have been used for evaluating MT quality. Due to the variations in the corpus quality and size, highly varying results have been achieved. While Turkish ↔ Catalan engines have had the highest automatic evaluation scores, Turkish ↔ Arabic engines have had the lowest automatic scores. While the quality results are highly varying across languages, we obtain baseline scores for a wide variety of languages coupled with Turkish. These results may provide a reference point for evaluating future MT systems including Turkish.

Applying Collocation Analysis to Chinese Discourse: A Case Study of Causal Connectives

AbstractCollocation analysis can be used to extract meaningful linguistic information from large-scale corpus data. This paper reviews the methodological issues one may encounter when performing collocation analysis for discourse studies on Chinese. We propose four crucial aspects to consider in such analyses: (i) the definition of collocates according to various parameters; (ii) the choice of analysis and association measures; (iii) the definition of the search span; and (iv) the selection of corpora for analysis. To illustrate how these aspects can be addressed when applying a Chinese collocation analysis, we conducted a case study of two Chinese causal connectives:yushi‘that is why’ andyin’er‘as a result’. The distinctive collocation analysis shows how these two connectives differ in volitionality, an important dimension of discourse relations. The study also demonstrates that collocation analysis, as an explorative approach based on large-scale data, can provide valuable converging evidence for corpus-based studies that have been conducted with laborious manual analysis on limited datasets.

El corpus paralelo como herramienta para explorar los elementos únicos en el checo

This study makes a contribution to the discussion of one candidate for a translation universal, i.e. the hypothesis concerning «unique items» (Tirkkonen-Condit 2002, 2004). We address one line of criticism of this hypothesis, namely «problems with defining unique items a priori» (Chesterman 2007, 11). We argue that candidates for «unique items» can be revealed through Johansson’s contrastive method of systematically studying «meaning through translation patterns» in a parallel corpus (Johansson 2007a), especially by comparing correspondences of a polyfunctional or vague item in source and target texts. Having previously investigated the correspondences of the Czech polyfunctional particle prý in English (Martinkova and Janebova 2017), we now turn to Spanish. The paper touches upon problems which have to be dealt with in such contrastive parallel corpus-based studies.

More lexically-specific knowledge and individual differences in adult native speakers' processing of the English passive

This paper presents experimental evidence for the role of lexically specific representations in the processing of English be and get passive constructions. Using a self-paced reading task, processing of full and truncated be and get passives was compared on sentences containing verbs strongly associated with these constructions, as determined by supercluster analysis (see e.g., Myachykov et al. 2012). This study complements these corpus-based studies by providing evidence from an on-line processing task that tests whether native speakers are sensitive to the observed distributions. The results support a usage-based functional account of processing and interpreting English be and get passive constructions. Participants' performance was influenced by frequency and lexical specificity. The study also provides evidence of education-related differences in language attainment – the higher educated participants were significantly better at interpreting be and get full passive constructions than the lower educated participants.

Uma análise contrastiva dos bergoglismos em corpus de discursos do Papa Francisco

Em 2013, Jorge Bergoglio, argentino, foi eleito Papa da Igreja Católica de Roma. O interesse por sua personalidade cativante, para além do fato de ser o primeiro pontífice latino-americano, ganhou preponderância, desde então, despertando a curiosidade em torno dos chamados bergoglismos ”“ fala peculiar do Papa Francisco. O objetivo do presente trabalho é analisar marcas de Avaliatividade nos discursos papais oficiais em língua espanhola e nas respectivas traduções para o português. O site oficial do Vaticano disponibiliza, entre outros textos, os discursos papais em diversas línguas. Assim, este trabalho compilou desse site os discursos feitos em países de língua espanhola e suas respectivas traduções para a língua portuguesa. A partir desse corpus, com subsídios da Linguística de Corpus (BERBER SARDINHA, 2004) e as ferramentas do programa WordSmith Tools 6.0 (SCOTT, 2012), foram analisados aspectos de Avaliatividade em torno dos bergoglismos em ambas as línguas. Para isso, a teoria da Avaliatividade (MARTIN; WHITE, 2005) norteou as análises, por meio dos significados dos subsistemas da Atitude (Julgamento, Afeto e Apreciação). Por outro lado, os neologismos foram observados a partir da classificação de Alves (1990). Com esse arcabouço teórico e um quadro metodológico inspirado em Novodvorski (2008; 2013), além das ferramentas da Linguística de Corpus, este trabalho analisou a formação dos neologismos papais e as marcas de Avaliatividade impressas pelo Papa nessas construções neológicas (bergoglismos), comparando-as com as traduções para a língua portuguesa. Para alcançar tais propósitos, além da compilação do corpus de estudo, também se fez oportuna a compilação de um corpus de apoio, com matérias jornalísticas sobre o Papa e, principalmente, sobre os referidos bergoglismos. Espera-se que esta pesquisa contribua para os Estudos da Tradução e da Neologia baseados em corpus, mais especificamente sob a visão da teoria da Avaliatividade e dos recursos da Linguística de Corpus, visto que os procedimentos adotados neste estudo também poderão ser replicados em outros corpora. A combinação teórico-metodológica e a escolha do corpus paralelo espanhol/português mostraram-se produtivas.