English academic discourse in translinguistic context: corpus-based study of lexical markers

Размещено на http://www.allbest.ru

Volgograd State University, Volgograd, Russia

English academic discourse in translinguistic context: corpus-based study of lexical markers

Larisa A. Kochetova, Elena Yu. Ilyinova

Abstract

The paper applies the Matrix method to an investigation into translinguistic features of English academic discourse that is serving worldwide a means of cross-cultural exchange between researchers with translingual skills. Based on the corpus approach to the comparison of the two corpora that comprise samples of professional academic writing in various fields of study (Art and Humanity, Natural and Social sciences), the paper seeks to identify both quantitatively and qualitatively correlations in repertoire and frequencies of recurrent linguistic expressions between the native English-language and non-native (Russian) academic discourse performers. The corpora were investigated along with the use of lexical bundles, re-occurring lexical units, which were grouped into noun-based and preposition-based phrases with post-modifier fragments, verb-based phrases with any form of verb components. The data comparison points to a code-mixing trend at the syntagmatic layer, which is a translingual fusion of English words in accord with a mixture of syntagmatic relations typical of English and Russian variations of academic discourse. It was found that non-native writing does not reveal as much lexical flexibility as native writers do and to a large extent relies on formulaic expressions, most of which are not conventional for expert native academic writing. Native Russian writers use excessively noun-based phrases with abstract nouns and underuse noun phrases without prepositions. Verb phrases with that- and to-clauses are mainly characteristic of native professional writing whereas non-native writing employs patterns with active verbs and pas sive constructions. It was found that non-native writing lacks quantifying phrases and hedging expressions that mitigate the proposition.

Key words: academic discourse, corpus, corpus methodology, lexical bundles, English language, translingualism.

Аннотация

АНГЛОЯЗЫЧНЫЙ АКАДЕМИЧЕСКИЙ ДИСКУРС В ТРАНСЛИНГВАЛЬНОЙ СИТУАЦИИ: КОРПУСНО-ОРИЕНТИРОВАННОЕ ИЗУЧЕНИЕ ЛЕКСИЧЕСКИХ МАРКЕРОВ

Лариса Анатольевна Кочетова

Волгоградский государственный университет, г. Волгоград, Россия

Елена Юрьевна Ильинова

Волгоградский государственный университет, г. Волгоград, Россия

В статье представлены результаты, полученные при применении матричного подхода к изучению признаков транслингвальности англоязычного академического дискурса, ставшего каналом кросс- культурного обмена научной информацией между учеными, обладающими трансъязычными средствами коммуникации. Корпусные методики сравнительного анализа двух корпусов научно-ориентированных текстов по различным сферам знаний (гуманитарные, социальные, естественные науки) позволили выявить количественные и функциональные корреляции в лексико-синтагматическом репертуаре единиц дискурса на основе критерия частотности употребления типичных конструкций носителями английского языка и рус- скоговорящими авторами научных статей. В качестве лексических маркеров английского академического дискурса были выбраны лексические коллокации, высокочастотные словосочетания трех синтагматических групп: именные и предложные словосочетания с атрибутивными компонентами, глагольные словосочетания, включающие любые предикативные формы, в том числе to-clause, безличные it-patterns модели. Сравнение корпусных данных указывает на наличие кодовых смешений на синтагматическом уровне, употребление английских слов, являющихся транслингвальными кальками, для выражения синтагматических отношений, характерных для русскоязычного академического дискурса. Показано, что русскоязычные авторы, пишущие на английском языке, часто используют формульные клише, не характерные для стандартного англоязычного академического текста, именные конструкции с абстрактными существительными и беспредложные номинативные конструкции. Глагольные конструкции с придаточными, вводимыми союзом that и частицей to, типичны для профессионального дискурса носителей английского языка, в то время как русскоязычные авторы употребляют преимущественно конструкции с глаголами в действительном залоге и синтагматические клише, содержащие глаголы определенной семантики в страдательном залоге. Установлено, что в академическом дискурсе русскоязычных авторов отсутствуют дискурсивные формулы с семантикой количества и выражения, смягчающие категоричность высказывания.

Ключевые слова: научный дискурс, корпус, корпусная методология, лексические сочетания, английский язык, транслингвизм.

Introduction

Translingualism has become increasingly popular as a theoretical construct that points to a current trend - a dynamic fusion between permeability and convergence of languages and cultures in human brains; it is a characteristic of modern translingual communication in the globalizing world. Translangualism results from an increase in the number of people who apart from their mother tongue have proficiency in one or several foreign languages, and while communicating information they might refer to translingual signs to convey their ideas through language code-switching and messing. It results in translanguaging, a kind of communication process whereby signs and rules of various languages are integratively used in the discursive practice without being merged or submitted.

The phenomenon of translingualism is being studied worldwide in social and psychological perspectives. Linguists focus more on the transcrossing of grammatical and lexical features of various languages in the context of translingual communication, in particular, some theoretical points are set on the bilingual corpus data and by employing a Matrix methodology groups of researchers investigate aspects of codeswitching in written or oral speech [Chauncey, Grainger, Holcomb, 2008; Deuchar, 2006; Fricke, 2016; Gabdreeva, Ageeva, Timirgaleeva, 2013; Gardner-Chloros, 2009; Gevers, 2018; Gullberg, Indefrrey, Muysken, 2009; Maksimova, 2016; Marinova, 2008; Molina, 2011; Moradi, Chen, 2019; Myers-Scotton, 2006; etc.]. While recognizing translanguaging as a hybrid cognitive strategy of communication using mixing language patterns and units from native and second languages, we would like to suggest for consideration another issue on translingualism - discursive aspect of academic writing in English by Russian professionals due to the necessity to exchange scientific facts and discoveries in the global academic discourse. This paper presents an analysis of corpus-based data to support a hypothesis on translinguial code-mixing in scientific articles authored by Russian professional scientists that is manifested in some lexical markers of Russian academic discourse incorporated into papers written in English.

Terminology review

The transcultural and translingual trend was distinguished in speech practices in the late 1990s by specialists in applied linguistics and language teaching [Zamel, 1996; 1997]. While analyzing speech behavior of bilinguals and translinguals, they noted that due to their polylanguage repertoire they could produce translinguistic discursive unities (based on a mix of language units and rules) which pointed to their ability to code-switching and codemixing. Having chosen a syntactic (grammar) basis of one of the languages, they could layer morphemes from various languages with some evident deviations in syntagmatic patterns from the standard grammar and other rules of the basic language [Canagarajah, 2002; 2013; 2016; Pennycook, 2008; Zamel, 1996].

The trend on translanguaging is viewed in modern globalized communities worldwide, especially in prominent economically advanced, and cultural centers where the mixture of languages is an objective reality. In many locations a trend of expansive multilingualism with the English language as a basis of national-wide communicative behavior has been noticed; however, there has established some novel style of globalized communication that is based on English as lingua franca, used by professionally- oriented communities with their national mother tongues to exchange information on science and humanities, technical achievements, social issues.

The trend pronounced contradicts strict requirements in prescriptive grammar to bilingualism that means fluency in two languages (practical skills of efficient communication in a certain linguistic and cultural environment); however, the term translingualism states the fact of dynamics in speech behavior in times oftranscultural globalization [Crystal, 2003; Tlostanova, 2008]. If there exists a trend than it is a task of linguists to start monitoring such type of discourse realization and with the tools of corpus linguistics collect data for further observation on social functions and linguistic patterns of mixed-speech discourse and pave the way to better understanding translanguage abilities represented in cognitive strategies of code-switching and code-mixing.

Thus, the term translingualism specifies some phenomenon in modern communicative practice; it is implemented in translanguage abilities that are used for constructing a kind of translanguage hybrid text through mixing units and grammar rules of various languages. In translingual situations people demonstrate cognitive abilities of code-switching and code-mixing due to which there appears some degree of syntagmatic fusion without one language expansion; by using non-native English the language users preserve their mother tongue identity, which is revealed in some language equivalents and represent some native-like cognitive strategy of discourse production. The mixture of language units and rules of their combinability while the native-like elements are installed into hybrid discourse structures point to the notion of translingualism, the way how nonnative English users communicate in professionally oriented communication in Global English.

Methodological basis

Accepting the fact that translingualism is a dynamic feature of modern globalized communication it is necessary to lay theoretical background and methodological basis for linguistic analysis of translanguaging as a visual representation of the phenomenon under study. A critical review of publication by European, American, Asian, and Russian research groups tells about the corpus- assisted approach to translanguage markers studies which attempt to explain some cognitive strategies of this type of verbal behavior (code-switching and code-mixing, in particular) and offer some methods to single out some language markers.

Translingualism is supposed to present some novel ways of communication in oral and written forms. Translingual situations are specified, firstly, by participants - they are non-native English users with various cultural and linguistic bases; secondly, it is due to their goals of pragmatic efficiency of communication that they refer to hybrid strategies of discourse production; thirdly, in their translanguaging discursive performance, people refer to a fusion of language codes.

The data obtained through corpus analyses point to some translingual strategies that consist in switching and meshing language codes, cross- cultural borrowings, simplifications, and hybridization of syntactic layers of speech utterances [Caragarajah, 2002; Gardner-Chloros, 2009; Lefringhausen, Marchall, 2016; Leonard, Nowacek, 2016; Proshina, 2016; Spencer-Oatey, 2018; etc.]. The methodology of linguistic analysis is based on a discrete approach with language layers distinction and interpretative comparison of real-life (natural) samples of translanguaging to the standard norms of languages that are mixed (as identified languages). To date, linguists have published many reports about dialectal translingual studies (for ex., Welsh-English, Persian-English, Spanish-English, French-English-Cyprian, Russian-English, etc. [Chauncey, Grainger, Holcomb, 2008; Gabdreeva, Ageeva, Timirgaleeva, 2013; Gardner-Chloros, 2009; Deuchar, 2006; Marinova, 2008; Moradi, Chen, 2019; Myers-Scotton, 2006; etc.]). In their works there dominates an opinion about the Matrix language basis for cases of translingualism, with a grammatical (syntactic) structure as some definite core of the utterances (referring to one language standard, usually standard English) and mixed-language lexical nominations inserted (words, phrases, set expressions from different languages). However, their explanations are mainly devoted to social and psycholinguistic aspects of translingualism, enumeration of cases when the transmixture of words are noted as cases of code-switching, identification of cases with deviations from standard norms of syntactic structure [Gardner-Chloros, 2009; Moradi, Chen, 2019; Myers-Scotton, 2006; etc.], the aspect of discourse production by non-native professionals in the translingual environment as the mechanism of code-mixing has not been by now in the focus of linguists' interest.

Considering blurring differences between code-switching and code-mixing, we assume that these mechanisms are described through various methodological approaches with more interest to code-switching, none of them on its own can provide a complete overview. The mechanism of code-switching is described as the process of alternation of units of two or more languages in an individual speech utterance. It is studied with a discrete method, translinguistic cases are described separately as leapfrogging at certain language layers (phonomorphological, grammatical, syntactical), and then multilayered data are used for making a general conclusion about the degree of social integration or acculturation of a person, besides they might formulate grammatical tendencies rather than general syntactic rules of translingual language usage [Gardner-Chloros, 2009; Spencer-Oatey, 2018; Lefringhausen, Marchall, 2016]. However, code-mixing as a strategy of speech production remains practically unclear. Code-mixing can be described as the process of language blending, which presupposes mingled co-occurrence of signs of different languages and norms of their realization in speech practice produced by nonnative users. It is not an individual switch from one language marker to another, it is a kind of switch-mode characteristic of a group of users. The conversational efforts of a certain group of non-natives (let's say in English) should demonstrate some common frequency of other language marker occurrences (mainly, mother tongue cognitive strategies). The veracity (reliability) of such data requires a corpus-oriented matrix methodology grounded on a multilayer search. The multilayered realization of codemixing implies more technical efforts to result in distinguishing translingual cognitive patterns that require references to discourse oriented material for analysis. This research was based on two collections of academic papers written in English by Russian scientists who are thought to be proficient in written academic discourse and by native British scientists; their lexical matrix comparison resulted in distinguishing some syntagmatic and lexical markers of code-mixing.

The English language has undoubtedly become the preferred medium of communication in the academic context on a global scale. Nonnative speakers of English in the academic professional community have been encouraged to use English without constant recourse to the native-speaker ideal. As a result, the English used in globalized professional academic writing lacks uniform standard but displays a variety in the use of language structures that is due to an array of factors among which the most important is a cognitive lingual background that determines how the non-native speakers verbalize their ideas in English. To date, a large number of papers that compare and contrast academic texts produced by native and non-native speakers have been published. They concentrated on the discretelayered method of analysis, outlining lexical and grammatical features of the writings, accentuating the use of linking adverbials, modal verbs, conjunctions, collocations [Altenberg, Granger, 2001; Chen, Baker, 2010; Gao, 2016; Martin, 2003; Siyanova, Schmitt, 2008; Yang, 2018]. These studies, most of which employed the corpus-based approach to counting and comparing the use of linguistic features, have found some ways in which professional academic writing of English natives and nonnatives differ. However, markers of code-mixing in non-native academic writing have not been fully investigated and the lexical layer, which, in particular, is regarded as most prominent in revealing markers of code-mixing, has not received proper attention so far. Recent research in English academic discourse [Biber et al., 1999; Biber, 2006; Biber, Barbieri, 2007; Cortes, 2004; Hyland, 2008; Simpson-Vlach, Ellis, 2010; Wray, 2000] has evidenced the importance of discursive lexical bundles (recurrent sequences of words) as a major component in providing coherent discourse and an essential aspect of the shared knowledge of a professional discourse community on academic discourse genres. The focus of attention has been given to the identification and structural implementation of these discourse bundles as well as their functions and discipline variations. Having discovered some definite markers of academic discourse in native English writings, the researchers continued investigation by analyzing the use of lexical bundles in non-native speakers' writing and stated some deviation in the language means mainly based on implied references to the genre- and discipline-specific recurrent expressions typical of their native academic discourse.

The researches mentioned deal with academic English as lingua franca as it is used by speakers of Arabic, Chinese, German, Finnish, other European and Asian languages. It should be noted that English academic writing by native speakers of Russian is a relatively new field of research, and papers that compare and contrast the academic style writing have been far and few between. One of the most comprehensive comparative research based on the linguistic corpus of paper titles authored by native Russian and English speakers, style-forming properties of academic English [Ryabtseva, 2018a; 2018b; Stebletsova, Torubarova, 2019; Viktorova, 2015] revealed cross-cultural incongruence. Thus, the titles in the academic discourse of Russian language speakers were noted not mapping the standard stylistic patterns characteristic to English academic speech; word-for-word translation techniques and the genre conventions of the Russian academic discourse are identified as the two major factors that determine the frequency and distribution of linguistic features in the academic paper titles offered for publication by Russian researchers. As compared to the English academic style that prefers a two-part title pattern, which, in its turn, uses the and-conjunction, a column, non-finite forms of the verb, prepositions, interrogative constructions, etc., the Russian academic style favors rather extended titles that include multiple abstract nouns, incorporated into a single noun group with repeated genitive constructions [Ryabtseva, 2018b]. Comprehensive contrastive research into two broad classes of discourse markers revealed significant differences in their usage in Russian and English language academic discourse [Viktorova, 2015]. However, the limitations of the studies mentioned are that it does not employ corpus methodology, and conclusions, though they seem to have been plausible, cannot be validated by statistical data.

In our previous studies, we used the corpus approach to contrast and compare the discourse functions of the first person plural pronoun we in the BE2006 (learned) corpus and the corpus of non-native academic writing that comprised papers written in English by Russian authors (EFLAW) which is also used in the present study [Kochetova, Kononova, 2018]. The pronoun we as a generic convention admits exclusive and inclusive references and represents culturally determined discursive strategy of construing communicative categories of solidarity, credibility, politeness. Although the statistical tests did not indicate significant differences in the frequency of the pronoun in the contrasted corpora, qualitative manual analysis of the concordance lines revealed that in the EFLAW corpus the inclusive we are employed to persuade the addressee to share and accept the author's arguments whereas in the BE2006 corpus it is used to construe solidarity based on common background, shared beliefs and opinions [Kochetova, Kononova, 2018].

The present study focuses on the use of recurrent lexical units as markers of code switching in academic papers written in English by Russian scientists. It uses corpus methodology, and its advances rely on statistical measures thus providing reliability in revealing correlations in the data and giving a precise and accurate picture of how language is used across various genres, text types, and registers.

The linguistic features analysis, which is at the center of the present study, uses the Matrix approach; after distinguishing the lexical layers with some core markers of academic discourse organization it allows to state the frequency and distribution of lexical bundles in the academic writing of the two groups, a group of English language native speakers and a group of nonnative English writers, whose mother tongue is Russian. The present study is concerned with the comparative usage of lexical bundles in native and non-native academic discourse performers. The investigation is carried out on a corpus of professional academic writing of the Russian language scholars to find out quantitative data on functional types of lexical bundles typical of expert academic discourse in British and Russian discursive tradition. Lexical bundles as conventionalized expressions are also called formulaic sequences [Wray, 2002], formulaic expressions [Simpson, 2004], fixed expressions [Moon, 1992], lexical phrases [Nattinger, 1988], multiword lexical units [Cowie, 1999], n-grams [Cheng et al., 2009], or lexical bundles [Biber et al., 1999; Cortes, 2004; Hyland, 2008]. While the studies into conventionalized expressions apply different explanations and provide various criteria for the identification of multi-word sequences and thus approach their analysis from altered perspectives, they all conclusively show that conventionalized expressions constitute a large proportion of discourse and therefore indicate competent use of language in a particular context. This is in good agreement with Sinclair's `idiom principle' [Sinclair, 1987] which postulates that speakers and writers do not select the words that they use one at a time, but rather choose units of meaning expressed by pre-constructed phrases.

Research questions and methodology

Based on the corpus methodology to the study of language in use, the research aims to reveal extended fixed collocations in academic discourse produced by the two categories of writers. Specifically, this study will address the following research questions: 1. In what ways does the professional academic writing of native and nonnative speakers of English differ in the use of lexical bundles typical of English academic discourse?

2. How do these most notable differences report on code-switching strategies in the discursive behavior of non-native writers in English?

To achieve the goals a comparative research design was applied, in which frequently-used word combinations (i.e., lexical bundles) were identified by an automated frequency-driven approach, that is corpus linguistics methods employed as the SketchEngine software tool [Kilgariff et al., 2004], then they were compared to the most preferable by the British authors of academic texts functional types of lexical bundles in expert academic discourse.

By definition, lexical bundles are groups of three or more words that are the most frequent recurring fixed sequences in a register [Biber et al., 1999,

p. 990]. Examples of lexical bundles in English academic writing are expressions like on the other hand, as a result of or the end of the, to mention just a few. To conduct the study, a corpus of English academic writing (EFLAW) that comprises papers written in English by Russian language authors was compiled. As a reference corpus of native academic writing, FLOB-J corpus was taken from [Chen, Baker, 2010]. Corpora information is presented in Table 1.

The results for the FLOB-J were taken from [Chen, Baker, 2010], they were obtained in the comparative study of lexical bundles in BAWE-EN and BAWE-CH as compared with native professional academic writing. Lexical bundles retrieved from a corpus of published professional academic texts written in English by native Russians EFLAW and the FLOB-J corpus were investigated both quantitatively and qualitatively.

Table 1 Comparison of FLOB-J and English as a Foreign Language Academic Writing Corpus (EFLAW)

Three broad categories of bundles, namely Noun-based phrases (NP), Preposition-based phrases (PP), and Verb-based phrases (VP) were retrieved. Noun-based bundles include any noun phrases with post-modifier fragments, such as the role of the or the way in which (i.e., Category (1) in Table 3). Preposition-based expressions include those starting with a preposition plus a noun-phrase fragment, such as at the end of or in relation to the (i.e., Category (2) in Table 3). Concerning verb-based bundles, any word combinations with a verb component, such as in order to make or was one of the, is assigned to this category (i.e., Categories (3) to (8) in Table 3).

The proportion of noun-based bundles does not show a great difference in the two corpora. To explore the lexical bundles used in the corpora under study in greater depth, the noun-based combinations were grouped further into two structural subcategories to see if these two corpora could be distinguished from each other. These two subcategories are noun phrase fragments with of (e.g., in the context of) and any other noun phrase fragments without the preposition of (e.g., the way in which). Even though the proportion of NP-based bundles when compared with FLOB-J is relatively equal in the corpora under consideration, the EFLAW writing

is notably different in the subcategory of NP without the preposition of (Table 4). The combinations without the preposition of found in this investigation are all part of relative clauses, such as the extent to which, the fact that this, or the way(s) in which. It is evident that EFLAW writing did not use these types of relative clauses as frequently as native speakers did.

Table 2 The structure of the EFLAW Corpus