Answer 2 questions and come up with a question

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=rbeb20

International Journal of Bilingual Education and Bilingualism

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rbeb20

Monolingual ideologies versus spatial repertoires: language beliefs and writing practices of an international STEM scholar

Valeriya Minakova & Suresh Canagarajah

To cite this article: Valeriya Minakova & Suresh Canagarajah (2020): Monolingual ideologies versus spatial repertoires: language beliefs and writing practices of an international STEM scholar, International Journal of Bilingual Education and Bilingualism, DOI: 10.1080/13670050.2020.1768210

To link to this article: https://doi.org/10.1080/13670050.2020.1768210

Published online: 11 Jun 2020.

Submit your article to this journal

Article views: 16

View related articles

View Crossmark data

Monolingual ideologies versus spatial repertoires: language beliefs and writing practices of an international STEM scholar Valeriya Minakova and Suresh Canagarajah

Department of Applied Linguistics, Penn State University, PA, USA

ABSTRACT This paper explores language beliefs and writing practices of an international scholar in Biochemistry working at a large U.S. research university. Although the participant articulated monolingual ideologies and a desire to become like a ‘native speaker,’ he did not consider advanced writing skills in English a prerequisite for publishing in English in his field. Through ‘talk-around-text’ (Lillis 2008) and visits to his laboratory, we examine what resources the participant deemed valuable in the process of producing a scientific article. Adopting a spatial orientation to writing (Canagarajah 2018a), we pay particular attention to the spatial repertoires that shaped his recent first-author publication. We bring out the tensions between his language ideologies and actual communicative practices and discuss the theoretical and pedagogical implications of our research. Ultimately, we argue that a spatial orientation to communication expands the notion of bilingualism by urging us to consider people’s actual creative practices of meaning- making in particular spaces rather than focus on isolated cognitive abilities.

ARTICLE HISTORY Received 8 January 2020 Accepted 4 May 2020

KEYWORDS Language ideologies; bilingualism; bilingualism; spatial repertoires; linguistic justice

Introduction

Previous scholarship has paid considerable attention to writing and publishing practices of bi/multi- lingual scientists who use English as a second or ‘additional language.’ Recognizing their major role in the production of scientific knowledge, on the one hand, and the rise of English as ‘the language of science’ (Englander 2013) on the other, linguists have explored the challenges perceived by multilin- gual writers (e.g. Flowerdew 1999; Cho 2004; Huang 2010) and scrutinized how ‘non-native’ writers’ manuscripts deviate from ‘native-speaker’ norms of language use (e.g. Kaplan and Baldauf 2005; Li and Flowerdew 2007). Although this research is valuable for understanding bi/multilingual writers’ experiences and struggles, it rests on a number of assumptions that have remained unquestioned until recently. For one thing, it tends to focus on individual writer competence and, more specifically, on ‘gaps’ in one’s English language proficiency. In other words, scholarly concern has been with what ‘non-native’ English academic writers lack and what language skills they need to acquire to write appropriately (Lillis and Curry 2016). Furthermore, past research on STEM writing has been primarily ‘logocentric,’ i.e. focused on verbal resources while neglecting visuals and objects.

A poststructuralist orientation to communication, in particular, the notion of spatial repertoires, challenges these long-standing assumptions. Pennycook and Otsuji (2014) define spatial repertoires as ‘linguistic resources at people’s disposal in a given place’ (p.162). The authors contend that the idea of ‘spatial repertoires’ transcends ‘language-to-language relations (bilingualism, code-switching, mul- tilingualism, translanguaging)’ by drawing our attention to actual ‘language practices in relation to space and activity’ (162). The focus thus shifts from what individuals lack to what people do with

© 2020 Informa UK Limited, trading as Taylor & Francis Group

CONTACT Valeriya Minakova [email protected]

INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM https://doi.org/10.1080/13670050.2020.1768210

available resources. While Pennycook and Otsuji still mainly discuss linguistic features, Canagarajah (2018a) broadens their definition to include ‘all possible semioticized resources’ (37) and examines how spatial repertories are constructed and leveraged in academic text production. Within this spatial orientation, the image of a struggling individual researcher with imperfect English is replaced by that of a networked collaborative activity. Linguistic resources do not lose their significance but are considered part of an assemblage, on a par with space, objects, visuals, and bodies. Bi/multilingual writers thus are not viewed as constrained by their linguistic competence; instead, they are seen as bricoleurs, drawing on diverse spatial repertoires to produce meaning.

Despite these theoretical developments, monolingual ideologies still inform much research on international STEM scholars’ writing practices and are often shared by multilingual writers them- selves, especially by those who work outside mainstream Anglo-American academia. Research has consistently shown that multilingual scientists often view their ‘non-nativeness’ in English as a barrier to participating in academic activities (e.g. Huang 2010; Corcoran 2019) due to structural biases in the knowledge production system. To some degree, multilingual scientists are influenced by dominant ideologies of academic and publishing networks. While whether language identities dis- advantage one against academic publishing is being debated (e.g. Hyland 2016; Flowerdew 2019), we wish to contribute another perspective from writing practice. Acknowledging that structural biases might negatively affect multilingual, ‘periphery’ scholars, we would like to shift the focus to the role of diverse semiotic resources in scientific texts. We argue that by focusing solely on language, we risk misunderstanding the processes of knowledge production in STEM fields. That said, we are not disregarding multilingual writers’ language concerns; on the contrary, we view them as of great importance for understanding scholars’ academic communication practices. Yet, we seek to broaden the scope of analysis of these practices by including spatial resources and looking closely at the practices of academic text production.

In this study, we draw on Canagarajah’s (2018a, 2018b) spatial orientation to communication to analyze language beliefs and writing practices of an international scholar in Biochemistry. In particu- lar, we examine his recent first-author publication focusing on the ‘backstage’ of its production and his views on what role his writing abilities in English played in this process. Before introducing our participant, we briefly summarize the research on the use of different semiotic resources in pro- fessional writing practices of STEM scholars.

STEM writing beyond grammar and vocabulary

Although the literature on spatial repertoires and STEM writing is not abundant, several scholars in applied linguistics have recently turned to observing STEM scholars’ actual writing practices, paying close attention to the kinds of resources being used. For example, Hanauer (2014) investigates laboratory notebooks of microbiology students. Observing how the notebooks are created and used over time, he notes that as students grow more familiar with the research procedures, the text gradu- ally recedes in significance, giving space to visuals. In other words, text appears to play a mediating role since the detailed recording of the procedures helps novice researchers carry out and under- stand experimental work. At the stage of presenting results, however, long textual descriptions seem unnecessary since images are sufficient for capturing the outcome of experimental work.

Researching writing practices of faculty members and students at a U.S. engineering school, Curry (2014) also highlights the importance of visuals. One of the findings of her extensive three-year study was that graphics in engineering publications do not merely illustrate points in the text; rather, they ‘act as invention devices that enable engineers to map out arguments to put further in research dis- semination’ (102). Oftentimes, the participants in the study started their writing by considering what figures should be included in the text. Thus, visuals helped them build and articulate their argument. The author concludes that this generative power of graphs should be taken into account by teachers of writing for academic purposes classes, which tend to focus solely on text.

2 V. MINAKOVA AND S. CANAGARAJAH

Canagarajah (2018a, 2018b, 2018c) conducted a series of studies on multilingual STEM scholar’s socialization in the U.S. academia. Like Curry (2014), he points to the crucial role of graphs and images in STEM work. His participants expended considerable time and effort on producing figures, viewing them as persuasive devices. The majority of journal reviewers’ comments also con- cerned graphs rather than textual descriptions, which further highlights the significance of figures in presenting scholars’ arguments.

The qualitative study reported in this article is informed by close attention to materiality in STEM scholars’ work. Adopting a case study approach, we examine what resources international scientists use and consider significant in their work by analyzing the language beliefs and writing practices of a young multilingual scholar (Alexey1) from Russia. In doing so, we contribute to an understanding of tensions between ideologies and communicative practices experienced by international scholars in U.S. academia. Our research was guided by the following questions:

(1). What are Alexey’s attitudes towards his bilingualism in Russian and English? (2). What are his beliefs about the role of linguistic and other semiotic resources in producing aca-

demic publications? (3). What spatial resources shaped Alexey’s recent first-author publication?

Methodology

Focal participant

The data were collected at a Biochemistry Laboratory at a large U.S. research university over a four- month period. Our focal participant, Alexey, is a young scholar in his early 30-s who moved to the U.S. from Russia in 2013. His heritage language is Russian, and he received his doctoral degree from a leading research university in Moscow. He started learning English in the 5th grade. Since English classes were not available at his school, he attended commercial courses and studied with a tutor. In college, Alexey took three years of English and continued learning the language by himself, mainly by watching T.V. shows in English. He described himself as ‘proficient’ and felt comfortable communicating at work. However, he aspired to higher standards of English proficiency, which we will discuss later in the paper. After working as a postdoctoral researcher at the U.S. university for almost five years, Alexey was appointed an Assistant Research Professor, a contract-based position that is reviewed annually for renewal. By the time of the study, he had been holding this position for almost 1,5 years. Since 2008, he has co-authored 13 articles in English and was the lead author on two of them.

Data collection

Adopting an ethnographic perspective, we complemented textual data (i.e. Alexey’s article and its drafts), with regular visits to the lab, during which Alexey gave us a video tour of his department, allowed us to participate in one of his experiments, and gave several interviews. We also video- recorded three research group meetings (RGMs) in which Alexey presented his data. Our goal was to contextualize his academic writing and become familiar with the nature of his work, its physical setting, and his social networks.

In this paper, we mainly focus on ‘talk around texts’ (Lillis 2008), a method that ‘usefully directs the researcher’s attention beyond the written text towards a consideration of some elements of writers’ perspectives about texts’ (355). To explore Alexey’s views about the process of scientific writing, we conducted four semi-structured interviews that were audio-recorded and lasted from 45 to 85 min. The questions focused on his most recent first-author publication and other related documents, such as early versions of ‘Methods’ and ‘Results’ sections that he co-authored, two final drafts of the

INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 3

manuscript (before and after the journal’s revisions), the journal reviewers’ comments, and his response to them.

All the interviews were held at Alexey’s lab, his preferred place for our meetings. Although he shared the same heritage language with the interviewer (i.e. first author), most of the research- related interactions took place in English. Using Russian for greetings and small talk, the interviewer offered Alexey to choose in what language he preferred to be interviewed. He opted for English, and we explain the significance of his choice in the following section.

Data analysis

Alexey’s language beliefs

Before considering the role of spatial resources in Alexey’s work, we explore his beliefs about his bilin- gualism in English and Russian. In doing so, we seek to understand what semiotic resources he finds valuable and how comfortable he feels at work with his current linguistic repertoire. Overall, he aligned with a subtractive model of bilingualism (García 2009), trying to ‘suppress’ his heritage language and shift to English. For example, he claimed that he tried to think in English at home, believing that it would help him express his ideas easily in speaking. He also mentioned that he did not have any Russian-speaking friends in his town and did not seek contact with the Russian com- munity. The only time when he used his heritage language in his daily life was while Skyping with his father, who resided in Russia. As far as his workplace was concerned, he strongly associated it with English, claiming that even if there were other speakers of Russian in the lab, he would still use English with them. Notably, most of our work-related interactions took place in English despite the common mother tongue with the interviewer. He explained that it felt more natural to him and was more convenient since he could not always remember the names of equipment in Russian. It should be noted that the interviewer, who was not yet aware of Alexey’s language ideologies, occasionally switched to Russian (e.g. during greetings) to establish affiliation and build a rapport. Alexey replied in Russian but, reflecting on his language choices towards the end of the study, stated that he did not deem these switches necessary (see Appendix 1 for transcription conventions):

Excerpt 1 (INT 4/23/2019):

(1). A: let’s say that if you didn’t start speaking Russian I would spoke English= (2). INT: =the whole time? (3). A: the whole time (4). INT: and you would never switch? (5). A: no (6). INT: $why not$ (7). A: I don’t know just (1.0) didn’t see a reason to @ ha ha ha [..] I guess I’m not that attached (8). to it I’ll be fine speaking English for the rest of my life without speaking Russian […] I (9). mean I speak Russian better because it’s my (1.5) like my native? and (1.5) I guess I wish I (10). spoke English as I speak Russian rather than speak Russian (11). INT: uhum what does it mean (12). A: like be (1.5) be as fluent in English as a native speaker

As illustrated by the excerpt, Alexey did not view his bilingualism as a resource or a source of emotional attachment. In lines 9-12, he stated that he would rather ‘replace’ his native proficiency in Russian with that in English. In our interviews, he repeatedly expressed the idea that it was impor- tant to him to be like a ‘native speaker,’ which, in his view, meant speaking fluently, effortlessly, cor- rectly, and without a foreign accent. Alexey did not believe that he had achieved this ‘standard’ because he noticed minor errors and accent in his speech and often had to make a conscious effort to formulate his ideas in English. When inquired why it was important for him to be like a

4 V. MINAKOVA AND S. CANAGARAJAH

‘native speaker,’ he explained that he was a ‘perfectionist,’ and being a native speaker was ‘the best way to know a language’ (INT, 4/23/2019). Interestingly, Alexey did not report any communication- related difficulties or experiences of linguistic marginalization, and presented his desire to be like a ‘native’ as a self-imposed requirement rather than a necessity. Video recordings of his RGMs also show that he didn’t experience any problems with his current level of English and successfully com- municated with the members of his lab.

Thus far, we have discussed Alexey’s monolingualist views on oral communication in English. A different picture, however, emerged when our conversation shifted to writing. ‘Nativeness’ was no longer the ultimate measure of language mastery because Alexey did not consider writing a naturally acquired skill:

Excerpt 2 (INT 4/23/2019):

(1). INT: it’s important for you to be like a native speaker only in speaking or in writing as well (2). A: uhm I mean I don’t think that writing can be native? in g(h)ener(h)al (3). INT: ○that’s true @ha○

(4). A: I mean if you-if you are native speaker and you know how to write you’ll be like a native (5). writer too doesn’t make you a good writer @ha ha

In line 3, Alexey rightly challenges the interviewer’s framing of the question and laughingly adds in line 6 that ‘nativeness’ did not equal ‘good’ writing. In his view, ‘non-native’ speakers of a language could become effective writers as long as they had strong writing skills in their first language. He referred to writing as ‘both a skill and gift,’ transferrable to any language that one speaks proficiently, and contended that he was not ‘gifted’ in this respect, being a ‘poor’ writer in both English and Russian (INT 3/22/2019). To illustrate his point, he told us how during his PhD program in Russia, his advisor, a ‘non-native’ speaker of Russian, ‘corrected’ his writing:

Excerpt 3 (INT 03/22/2019):

(1). A: my PhD advisor was from Azerbaijan so his native was Azerbaijanian and he (1.0) (2). he still speaks with an accent and (1.0) his writing is good but when he talks he can like (3). confuse genders and like the ending of the word like something that a person who is a not (4). native Russian speaker would – that kind of mistakes that a person will do but he would still (5). correct my writing

In this narrative, Alexey distinguishes between competence in speaking and writing in relation to the idea of ‘nativeness.’ In lines 3-4, he constructs his advisor’s speech as accented and peppered with minor grammatical errors, once again making an unfavorable comparison with a ‘native’ Russian speaker. At the same time, in lines 4-5, a ‘native’ Azerbaijani is positioned as having more expertise in writing than Alexey for whom Russian is a mother tongue. For the first time in our data, Alexey did not treat the ‘native speaker’ as a model for imitation. On the contrary, he presents his advisor as not constrained in his writing by his ‘nonnative’ status and being able to offer valuable editing help to a ‘native’ Russian. In doing so, Alexey appears to contradict the native speaker fallacy, which he pre- viously demonstrated.

During his work in the U.S., Alexey continued to struggle with writing and relied on the help of his Principal Investigator (PI), this time a ‘native’ speaker of the target language. His P.I., whom we will call Dr.M, has been in the field of Biochemistry for over 40 years and has an extensive record of publi- cation. Alexey claimed that Dr. M had to ‘rewrite’ his texts as he was rarely satisfied with the quality of his junior colleague’s writing. In the following excerpt, Alexey explains what kinds of changes his P.I. made while continuing to present his own writing as ‘horrible’:

Excerpt 4 (INT 3/22/2019):

(1). A: if my boss tells me like remake this figure I can do it the same day

INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 5

(2). and he’ll be like I like it it’s great– but if he will tell me like rewrite (3). this I dunno section it will take me a week (INT: @ha ha) and he’ll tell me like not it’s still (4). horrible do it again @ha ha ha (5). INT: umm horrible like in what way? (6). A: like like the langu- well he usually um corrects the language? because he has his own style (7). and he would take your ideas and he will reword it in something he thinks is better? (8). >and I mean it’s- it is better < he is really good with writing a:nd but sometimes he says maybe (9). it’s like not enough information or I dunno but for me it’s always easier to work (10). with figures with data with numbers rather than with words

In lines 1–4 and 9–10, Alexey juxtaposes writing with data analysis and graph generation, present- ing the former as challenging and more time-consuming. Interestingly, editing a writing piece is com- pared to editing visuals. We will discuss the role of visuals later in the paper; for now, suffice it to say that this kind of work usually escapes the attention of linguists. Alexey’s comparison, however, suggests that figures require attention and polishing not less than does text. In talking about his P.I.’s edits in lines 6-9, Alexey focuses on style, wording, and clarity of ideas rather than language accu- racy. In other words, the revisions did not necessarily have to do with Alexey’s status as a ‘non-native’ user of English, struggling with grammar or word choice. As he suggested in line 9 and explained later in the interview, one of his main problems with writing was an inability to express his ideas clearly and provide sufficient information. These skills, in his view, were key to good writing and of more importance than excellent grammar.

Despite these self-reported difficulties, Alexey seemed reluctant to expend extra effort to enhance his writing:

Excerpt 5 (INT 3/22/2019):

(1). A: I probably could have taken writing courses or go to like seminars or things like that (2). trying to improve my writing by not just (1.5) next time I write a paper but by doing (3). something more than that (INT: uhum) (4). A: so I- I don’t do it knowing that $I can get away with it basically @ ha ha ha (5). INT: so how do you know you can get away with it how do you know it will be good in the end (6). A: well because my PI will be the one who make it good @ha ha ha

Alexey’s explanation seems somewhat surprising given that earlier he presented himself as a ‘per- fectionist,’ striving to achieve a ‘native’ command of English. It appeared that he was much less invested in writing than in oral proficiency and preferred to rely on ‘literacy brokering’ (Lillis and Curry 2006) rather than address the problem more proactively. Alexey’s choice of words in line 4 (‘get away with it’), smiley voice, and laughter in lines 4 and 6 present his strategy as not necessarily legitimate or commendable as he is aware of what he could have done to become a better writer. However, he lacked motivation, knowing that Dr. M would take care of his texts.

There was another reason why Alexey did not strive to achieve mastery in writing. Although it may sound paradoxical, he did not consider this skill crucial for publishing in English. In the next sections, we unpack this idea by examining the relationship between writing and authorship and looking closely at the use of different semiotic resources in Alexey’s recent article.

Authorship as doing

It has been noted that science is not a ‘solo activity’ and that the majority of articles nowadays are published by large teams (Englander 2013, 15). The order of authorship on such papers does not necessarily indicate the ownership of text and ideas and might be determined by other factors (Cana- garajah 2018b). Alexey’s recent publication is a case in point. It was written by the members of two different labs at his university. Out of seven authors, five were directly involved in writing the

6 V. MINAKOVA AND S. CANAGARAJAH

manuscript. The remaining two (the fourth and fifth authors) did not produce any text but generated figures and ran analyses of data. Alexey was assigned the lead authorship because he performed novel optical measurements while the experimental work of his colleagues from a different lab (the second and third authors) was based on previous research. All three co-authored the Methods and Results sections, describing the procedures they performed. These sections were then substantially revised by the PIs, the owners of the two labs, Dr. M and Dr. K. Their names appear last on the paper, although they contributed a lot in terms of writing. Dr. K wrote the intro- duction while Dr. M added the Discussion section and put all the drafts together, adding necessary clarifications. When we inquired why the PIs were the last authors despite their significant textual contributions, Alexey replied that being mentioned last in the sequence of authors signals the P.I.’s position of power, i.e. the fact they own labs and supervise projects executed by their team members. The amount of writing they did had little effect on this convention.

Overall, while discussing how the authorship was determined, Alexey assigned greater importance to doing experimental work and generating figures rather than to actual writing. He stressed that it is not the amount of writing or theorizing that gives a scientist the right to be an author on a manu- script; rather, it is concrete, material work in the lab:

Excerpt 6 (INT 03/22/2019):

(1). A: so let’s say you didn’t contribute data-wise you didn’t do any experiment, (2). you didn’t do any purification, you didn’t generate a single graph (INT: uhum) (3). A: you just ↑talk to pe:ople, you look at the gra:phs and you just wrote it down (4). like- like I dunno a writer hired writer and it doesn’t mean that you contributed (5). enough to be even on the paper

In this excerpt, Alexey prioritizes physical activity, involving bodies, spaces, and objects, over purely cognitive efforts in generating meaning and determining authorship. In line 3, he somewhat belittles the importance of writing by using the limiting modifier ‘just’ before ‘talk’ and ‘wrote.’ There is also a noticeable pitch rise on ‘talk’ and syllable elongation on ‘people’ and ‘graphs.’ These prosodic and lexical features present the described activities as undemanding and of minor significance. In line 4, Alexey goes on to suggest that this work could be done by a ‘hired writer,’ a person who is not directly involved in the knowledge production that takes place in the lab. A ‘hired writer’ might have no expertise in the area, and his work is portrayed as only tangentially relevant to the emer- gence of the manuscript.

In brief, Alexey presents labs as the key site of scientific knowledge and, ultimately, article pro- duction. The heart of the article is its data, generated not in the minds of scientists but in a concrete material environment as a result of concrete physical actions (e.g. preparing data samples, operating machines). By contrast, writing, according to Alexey, plays only a secondary role in the process and can be done by an ‘outsider’ who never entered the lab. In other words, spatial resources, coupled with technical expertise, are prioritized over the textual description of what occurred in the lab. In the final section, we look more closely at the process of obtaining data and generating graphs, attempting to understand what spatial resources Alexey employed in these activities.

Shaping a scientific article: the role of spatial resources

Alexey assigned great importance to the visuals in the paper, claiming that for experts in his field, it might be sufficient to just look at figures to understand the author’s argument. His 13-page article included eight figures, some of which were divided into subfigures, and was accompanied by six pages of supplemental visual materials. Thus, textual descriptions did not appear effective enough to report on the experimental work. ‘Every experimental data that we obtain we have to show [..] and the best way to describe things in our area is to plot a graph’ (INT 3/22/2019), commented Alexey on a large number of visuals. Typically, graphs are produced before the text, and the

INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 7

process of their generation starts from performing measurements in the lab. For example, to build spectrum graphs (Figure 1), Alexey used a spectrometer that provided a text file with the values of the x- and y-axis.

He then analyzed this file in the specialized software and plotted a graph. The process may seem straightforward but in fact, even its first step, i.e. obtaining data from the machine, involved multiple spaces, materials, and some degree of indeterminacy. During our visits to the lab, we observed and participated in Alexey’s experiment involving a laser and an electron paramagnetic resonance machine. While preparing the sample for analysis, we moved between different spaces to obtain certain tools, supplies, and oxygen conditions. Alexey also had to troubleshoot since the machines were ‘acting weird’ while measuring the signal. Commenting on this experience, he explained that the work of the lab equipment depended on a multitude of factors:

Excerpt 7 (INT 4/15/2019):

(1). A: it’s very sensitive to alignment of everything? and it’s not something you can just read a (2). manual you need experience either your own or someone who has it to make it work (3). because some people work in that room and they get really bad signals and other people (4). have really good signal and there are things that you can do to minimize the noise, and (5). make the signal stable so I guess it’s an art as much as a science in terms of making those (6). machines work […] they have their own quirks @ ha ha

This excerpt provides a fine illustration of how ‘human agency is qualified, as achieved in attune- ment with objects and networks’ (Canagarajah 2018b, 285). In line 6, Alexey positions the machines in the lab as agentive and having ‘their own quirks.’ A scientist to some extent depends on these quirks and needs to know ways to work around them. Particular material conditions, e,g., the level of noise, are also crucial in data generation. In other words, there is a complex network of objects, conditions, and agents involved in the very first step of graph production. As Alexey explained, it takes experi- ence and some professional ‘wisdom’ to emplace oneself in this network and facilitate the emergence

Figure 1. Spectrum graph.

8 V. MINAKOVA AND S. CANAGARAJAH

of data. This is not a purely cognitive activity that takes place in one’s brain. It is material, spatially distributed, and networked.

Early versions of the graphs needed editing before they became ‘publishable’ (INT 3/22/2019), and the changes were negotiated between Alexey and his P.I. For example, the following early (Figure 2 (a)) was divided into several subfigures (2b) in the final manuscript:

As Alexey explained, the original graph was hard to follow, so the intersecting lines were separ- ated. The figure legend, originally positioned next to the lines, appeared in the text box below the figure in the published manuscript. Such revisions can take considerable time:

Excerpt 8 (INT 4/15/2019):

(1). A: it took a lo::t of time to make this one [Figure 2(b)] because I would make it I would (2). show it to my boss and then he like oh yeah it’s great and then in a couple of days he like oh you (3). know I thought about it let’s do it differently like that part was edited the last moment […] (4). so it’s a process you go back and forth with your collaborators with your (5). colleagues with your boss and you generate something

Like the production of text, generation of figures is not a ‘one-shot’ activity; it is recursive, colla- borative, and time-consuming. As writers edit and refine their manuscripts, so do scientists who work with non-verbal resources. This activity requires certain technical, mathematical, and designing skills as well as access to technology and a graph plotting software (see Canagarajah 2018b for a more detailed discussion). What it does not require is advanced linguistic competence in English that is often considered central to a multilingual scholar’s success in academic publishing. In other words, knowledge generation in STEM is by no means limited to words. As our data illustrate, Alexey authored the article by skillfully generating data and crafting high-quality visual represen- tations of the results of his experiments. It is this work, rather than the production of coherent and elaborate text that made him the first author of the paper and a valued member of his professional community.

In this case, linguists’ exclusive focus on text in scientific writing seems unjustified. STEM articles should rather be treated as a conventional form of dissemination of experimental work that does not necessarily reflect the actual practice of knowledge generation. Excessive attention to grammatical structure takes a linguist further away from what matters most in a scientific publication: the con- ditions and results of experimental work. Therefore, we contend that the notion of scientific writing should be extended to include the context of its production if we are to understand its meaning and significance.

Discussion

In this paper, we presented the story of Alexey, a young international scholar in Biochemistry working in the U.S. Examining his views on bilingualism, we discovered that his thinking about language was dominated by monolingualist ideologies. Alexey aspired to achieve a ‘native’ command of spoken English, and treated grammatical structure as primary, disregarding spatial repertoires. Our interview and video data suggest that his desire to be ‘native’ did not stem from his professional practice, but reflected social biases. However, further observations are needed to explain his language ideologies.

Seeking excellence in speaking, Alexey was reluctant to work on his ‘poor’ writing skills. His lack of proficiency in writing did not seem to interfere with his current work as he was able to publish in academic journals and appear as the lead author. In explaining how this was possible, Alexey down- played the role of prose in the production of a scientific article and prioritized bench work and data plotting. Huang (2010) reports similar findings in his study with international PhD students in STEM fields. Six out of eleven participants considered experimental work more important and therefore were not motivated to improve their writing. However, the author seems to be skeptical about their claims, relating them to an ‘inferiority complex,’ which might be experienced by non-native

INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 9

Figure 2. (a) Draft graph representing ligh-induced spectra. (b) Final graph. Adapted by permission from Springer Nature.

10 V. MINAKOVA AND S. CANAGARAJAH

speakers of English (NNES) (Liu 2004). Huang (2010) concludes that ‘this sense of inferiority might also account for the phenomenon that NNES PhD students would rather spend most of their time con- ducting experiments than improve their writing skills.’ (41). However, Alexey did not appear to experi- ence a sense of inferiority because of his ‘non-nativeness’ when it came to writing. A more likely explanation for his lack of motivation to improve his writing is that this particular skill is indeed not crucial for his work at this moment. He authored articles by contributing data and visuals. Clearly, authorship policies might vary in different labs. Yet this specific focus on lab work does not seem to be rare. Kassis (2017) researched the question of authorship in STEM departments across the U.S, asking faculty members to rate several criteria in terms of their importance for deter- mining authorship. Among eleven criteria, ‘the time spent performing experiments had the highest overall median importance score’ (4). This finding indicates that prioritizing material over discursive activity is not an idiosyncratic characteristic of Alexey’s lab but a pattern found in other universities.

Consideration of spatial repertoires in STEM fields might add a new ‘twist’ to the linguistic inequal- ity debate. If it is material and visual resources that matter the most, then it might be expected that scientific manuscripts are less likely to suffer for their ‘imperfect’ English. A similar argument is pro- posed by STEM scholars themselves. For example, Pronskikh (2018) argues that STEM discourse is ‘more linguistically inclusive and egalitarian’ than that in social sciences because it relies on ‘the language of signs (numbers, formulae, graphs) or nonlinguistic forms of communication’ (e.g. protons, devices, data) rather than ‘figures of speech’ (82–83; emphasis in the original). This suggests that multilingual STEM scholars, both in Anglo-American academia and ‘at the periphery,’ are not overly dependent on verbal resources and ‘doomed’ to marginalization because of their ‘non-native- ness’ in English. Strategic use of spatial repertoires (i.e. visual and material resources, social networks) can help them become legitimate and valuable members of the English-dominant academic dis- course community.

Given the exceptional role of spatial resources in STEM work, a question arises as to how effective articles are as a means of knowledge dissemination. As this study and Canagarajah’s (2018b) inves- tigation suggest, scientists as well as journal reviewers are more preoccupied with the effectiveness of visuals rather than verbal descriptions. Advancement in technologies makes possible alternative ways of data sharing that reflect this focus on visual knowledge. Some scientific journals now ask for a video to accompany the Methods section (Englander 2013); others produce video articles (https://www.jove.com/journal) based on accepted manuscripts. While in these cases visual materials accompany text or result from it, it is not unreasonable to imagine the possibility of video sub- missions in the future that would not require the use of advanced linguistic resources. This might make the process of knowledge production and dissemination more linguistically inclusive for multi- lingual scholars in both ‘mainstream’ and ‘peripheral’ academia, although access to technology will continue to create inequalities.

It remains unclear why Alexey so zealously tried to achieve ‘nativeness’ in speaking while acknowl- edging that writing extended beyond one’s mind and involved multiple spatial and material resources. Could it be the case that STEM scholars experience more pressure to be like ‘natives’ in spoken communication because enunciation is an individual activity? Or could these concerns be driven by job market pressures? How equitable are workplaces in hiring multilingual scientists, and to what extent does speaking with a ‘non-native’ accent have repercussions on one’s career development? To answer these questions, it is necessary to compare how spatial repertoires function in both written and oral genres of academic communication and conduct more research on hiring practices in STEM fields.

Conclusion and limitations

The idea of spatial repertoires pushes us to reconsider our understanding of bi/multilingualism. Examination of concrete communication practices reveals that bi/multilingualism is not limited by linguistic features that individuals possess or lack. People employ a wide gamut of resources to

INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 11

create meaning. Spatial repertoires matter not only in everyday, routine contexts (e.g. at urban market places studied Pennycook and Otsuji 2015) where the material environment readily lends itself as a means of solving communication problems. As our study illustrates, spatial resources matter in academic contexts, usually associated with a high degree of abstraction and reliance on verbal communication. In the case of STEM writing, verbal resources appear neither sufficient nor of primary importance. It is the materiality of the experiment presented in figures, graphs, and photos that matters most.

This finding offers a more nuanced view on ‘the linguistic disadvantage orthodoxy’ (Hyland 2016). As we demonstrated, knowledge production in STEM fields is not limited to text, and multilingual scholars might be valued and acknowledged for their material and visual contributions. Thus, by focusing solely on language and positioning ‘non-native’ science writers as disadvantaged, we risk misunderstanding how knowledge is produced and shared in their fields. To clarify, we are not arguing that verbal resources don’t matter, but suggest that other semiotic resources might be more valuable in STEM fields.

We also argue that multilingual STEM writers’ language beliefs and ideologies should be critically examined in relation to specific/academic purposes. Educators could encourage international scho- lars to reflect on what spatial resources they use in their work, and help them explore distributed practices in both spoken and written interactions without anxiety from dominant language ideol- ogies. At the same time, we acknowledge that their deficit thinking about their language proficiency might stem from actual experiences of marginalization. Therefore, recognition of the crucial role of spatial repertoires should occur not only on an individual but also on an institutional level. That is, decision-makers should be made aware of the role of spatial repertoires in STEM communication to ensure linguistically non-discriminatory hiring practices.

Although in our study we attempted to contextualize Alexey’s writing, we analyzed the article that had already been published and thus did not have access to its production processes. For example, we could not observe how the experiments were done or how the results were discussed in RGMs. Future research should explore the process of article production in situ, rather than performing a ret- rospective analysis. Finally, although we attempted to depart from the linguistic injustice debate by focusing on spatial repertoires, we discussed language beliefs and practices of a scholar at a presti- gious research university in the U.S. who has a supportive professional network. We acknowledge Alexey’s privileged position and recognize the need to conduct more research on the role of spatial repertoires in STEM outside mainstream academic communities.

Note

1. A pseudonym.

Acknowledgement

We would like to thank Alexey for his time and patience. We are also grateful to two anonymous referees for their valu- able comments and suggestions

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes on contributors

Valeriya Minakova is a PhD candidate in the Department of Applied Linguistics at The Pennsylvania State University. She holds a BA in English Philology from Moscow State University and an MA in TESOL from American University, Washing- ton, DC. Her main research interests include second language learning and teaching, sociocultural theory of learning, translingual practices, heritage language maintenance, and language planning.

12 V. MINAKOVA AND S. CANAGARAJAH

Suresh Canagarajah is Edwin Erle Sparks Professor of Applied Linguistics, English, and Asian Studies at Penn State Uni- versity. He teaches World Englishes, Language Socialization, Interactional Analysis, and Academic Writing. His most recent publication is Transnational Literacy Autobiographies as Translingual Writing (Routledge, 2020).

ORCID

Valeriya Minakova http://orcid.org/0000-0003-1888-0811 Suresh Canagarajah http://orcid.org/0000-0002-1292-2366

References

Canagarajah, S. 2018a. “Translingual Practice as Spatial Repertoires: Expanding the Paradigm Beyond Structuralist Orientations.” Applied Linguistics 39 (1): 31–54.

Canagarajah, S. 2018b. “Materializing ‘Competence’: Perspectives from International STEM Scholars.” The Modern Language Journal 102 (2): 268–291.

Canagarajah, S. 2018c. “English as a Spatial Resource and the Claimed Competence of Chinese STEM Professionals.” World Englishes 37: 34–50.

Cho, S. 2004. “Challenges of Entering Discourse Communities through Publishing in English: Perspectives of Nonnative- Speaking Doctoral Students in the United States of America.” Journal of Language, Identity, and Education 3: 47–72.

Corcoran, J. 2019. “Addressing the “Bias gap”: A Research-Driven Argument for Critical Support of Plurilingual Scientists’ Research Writing.” Written Communication 36 (4): 538–577.

Curry, M. J. 2014. “Graphics and Invention in Academic Engineers’ Writing for Publication.” In Language, Literacy, and Learning in STEM Education: Research Methods and Perspectives from Applied Linguistics, edited by M. J. Curry, and D. Hanauer, 87–106. Amsterdam: John Benjamins.

Englander, K. 2013. Writing and Publishing Science Research Papers in English. New York: Springer. Flowerdew, J. 1999. “Problems in Writing for Scholarly Publication in English: The Case of Hong Kong.” Journal of Second

Language Writing 8: 243–264. Flowerdew, J. 2019. “The Linguistic Disadvantage of Scholars who Write in English as an Additional Language: Myth or

Reality.” Language Teaching 52 (2): 249–260. García, O. 2009. Bilingual Education in the 21st Century: A Global Perspective. Malden, MA: Wiley-Blackwell. Hanauer, D. 2014. “A Genre Analysis of Student Microbiology Laboratory Notebooks.” In Language, Literacy, and Learning

in STEM Education: Research Methods and Perspectives From Applied Linguistics, edited by M. J. Curry, and D. Hanauer, 27–42. Amsterdam: John Benjamins.

Huang, J. C. 2010. “Publishing and Learning Writing for Publication in English: Perspectives of NNES PhD Students in Science.” Journal of English for Academic Purposes 9: 33–44.

Hyland, K. 2016. “Academic Publishing and the Myth of Linguistic Injustice.” Journal of Second Language Writing 31: 58– 69.

Kaplan, R., and R. B. Baldauf Jr. 2005. “Editing Contributed Scholarly Articles from a Language Management Perspective.” Journal of Second Language Writing 14: 47–62.

Kassis, T. 2017. “How do Research Faculty in the Biosciences Evaluate Paper Authorship Criteria?” PLoS ONE 12 (8). https:// journals.plos.org/plosone/article?id=10.1371/journal.pone.0183632#sec001.

Li, Y., and J. Flowerdew. 2007. “Shaping Chinese Novice Scientists’ Manuscripts for Publication.” Journal of Second Language Writing 16: 100–117.

Lillis, T. 2008. “Ethnography as Method, Methodology, and “Deep Theorizing”.” Written Communication 25 (3): 353–388. Lillis, T., and M. J. Curry. 2006. “Professional Academic Writing by Multilingual Scholars.” Written Communication 23 (1): 3–

35. Lillis, T., and M. J. Curry. 2016. “Academic Writing for Publication in a Multilingual World.” In Handbook of Second and

Foreign Language Writing, edited by R. M. Manchón, and P. K. Matsuda, 201–222. Boston: De Gruyter. Liu, J. 2004. “Co-constructing Academic Discourse From the Periphery: Chinese Applied Linguists’ Centripetal

Participation in Scholarly Publication.” Asian Journal of English Language Teaching 14: 1–22. Pennycook, A., and E. Otsuji. 2014. “Metrolingual Multitasking and Spatial Repertoires: “Pizza mo two Minutes Coming”’.”

Journal of Sociolinguistics 18: 161–184. Pennycook, A., and E. Otsuji. 2015. Metrolingualism: Language in the City. Milton Park: Routledge. Pronskikh, V. 2018. “Linguistic Privilege and Justice: What can We Learn From STEM?” Philosophical Papers 47 (1): 71–92.

INTERNATIONAL JOURNAL OF BILINGUAL EDUCATION AND BILINGUALISM 13

Appendix 1: Transcription conventions (adapted from CLAN Manual for the CEAPP project).

Symbol Meaning = Latching − Cut-off ? Rising pitch at utterance end (not necessarily a question) , Slight rise in pitch at the end of an utterance. Continuing intonation ↑ Marked upshift in pitch (1.0) The length of the pause : Elongation. Each: represents 0.2 s [ Overlap Word Stressed word @ Laugher (h) Laughter within a word $word$ Smiley voice ° ° Soft speech > < Fast speech

14 V. MINAKOVA AND S. CANAGARAJAH