Introduction

Concomitant to globalization and the internalization of higher education is the growing prominence of global university ranking exercises (e.g. Hazelkorn 2011). Because an institution’s citation rates per faculty can factor into the influential university ranking lists, academic publishing becomes crucial for institutions of higher education vying to be listed in the top echelons of university rankings (Lillis and Curry 2013). Given the emphasis that national and institutional policies place upon research, publishing has become an important form of “symbolic capital” for scientists (Englander and Uzuner-Smith 2013).

One controversial issue inherent in international academic publishing is the dominance of English for research communication (e.g. Hyland 2015; Lillis and Curry 2010). Heated debates on this trend for Englishization (Phillipson and Skutnabb-Kangas 1999; Muresan and Pérez-Llantada 2014) have raised a range of sociopolitical and linguistic concerns over issues of inequality in knowledge dissemination, a potential risk of global diglossia, and national language domain loss (e.g. Bocanegra-Valle 2014; Ferguson et al. 2011; Flowerdew 2015; Hyland 2015; Lillis and Curry 2010). Research has also focused on individual scholars whose native language is not English and who use English as an additional language in academic publishing (Uzuner 2008). It has been found that these scholars encountered discursive and non-discursive difficulties in academic publishing (e.g. Belcher 2007; Uzuner 2008), sometimes interpreted from the geopolitical distinction between “center” and “periphery” in academic publishing (Canagarajah 2002; Lillis and Curry 2010). Various studies have demonstrated that these scholars’ academic writing practices are mediated by micro-level contextual factors related to disciplinary variation and target audience (Flowerdew and Li 2009; Kuteeva and Airey 2014; McGrath 2014).

Despite these rich findings, very few studies have explored how macro-level sociopolitical factors interact with multilingual scholars’ individual language practices. Lillis and Curry (2010) forcefully argued that “powerful evaluation systems of academic knowledge production based in the Anglophone center are both directly and indirectly supporting the privileging of English as the medium of academic texts for publication” (p. 156), constituting the “centripetal pull” that draws non-Anglophone peripheral scholars towards the dominant practices and ideologies in the Anglophone center (p. 160). It seems that this Anglophone-centered evaluation system may function as an additional mechanism that creates and perpetuates the de facto language policy and practices (Shohamy 2006) of publishing in English.

A few studies have explored academic publishing from the macro perspective of institutional and governmental research policies related to academic publishing (See Curry and Lillis 2013). Englander and Uzuner-Smith (2013) analyzed the discourse of higher education and globalization and revealed the ideology of knowledge as commodity, which promoted a market orientation in higher education. Lee and Lee (2013) discussed how Korean scholars complied with an implicit English-only policy enacted by national and institutional research policies. Feng et al. (2013) revealed the tensions between China’s nationalistic research policies and the centripetal pull towards Anglophone-centered knowledge production practices. Kuteeva and Airey (2014) and McGrath (2014) focused on the parallel language use policy, an explicit higher education language policy in Nordic countries, and pointed out that this “one-size-fits-all” policy failed to account for fundamental disciplinary differences in the Nordic context. These findings point to a promising research direction investigating the interaction between structural forces embodied by research/language policies and individual language practices in varying geolinguistic contexts.

In terms of studies of multilingual scholars’ language practices in academic writing, even fewer have looked at multilingual scholars whose native language is not English and who use non-native languages other than English for research communication. The existing research inadvertently prioritizes English by attending to multilingual scholars’ academic English literacy, but this Anglophone focus risks eclipsing the use of other languages in research communication and neglecting the multilingual nature of research communication. Hence, this paper uses the term “multilingual scholars” to refer to nonnative-English scholars who use a foreign language other than English for research communication purposes. These academics are involved in international academic publishing where “‘English’ and ‘international’ constitute an important indexical cluster used to signify high quality” (Lillis and Curry 2010: 6), but their voices and research may not receive an audience precisely because the linguistic medium they use to disseminate their research is not English. In view of this, the present study aims to focus on this particular group of multilingual scholars’ language practices in academic publishing from the perspective of the linguistic market theory, so as to gain insights into the interplay between individual language practices and the structural forces of the national and institutional research and language policies.

Theoretical framework

Our inquiry into multilingual scholars’ language practices aligns with New Literacy Studies that view literacy/ies as embedded in social practices (Gee 2008; Lea and Street 1998), with academic writing rooted in the cultural traditions, epistemological stance, power relations, and identities underpinning the activities of academic communication (e.g. Canagarajah 2002; Lillis and Curry 2010; Lillis and Scott 2007). We adopted Bourdieu’s (1991) social practice theory and specifically his notion of linguistic markets as the overarching framework guiding the study.

Bourdieu (1991) uses “market” to refer to a structured space of positions in which the positions and their interrelations are determined by the distribution of different kinds of resources or “capital” (Bourdieu 1986). Language varieties in the “linguistic market” receive their value according to particular systems of price formation that are governed by the ideological and the structure of the linguistic market, and thus some languages receive more value than others. Linguistic markets function through the mechanism of “unification”, a common recognition in the laws of price formation.

According to Bourdieu (1986), there are different types of capital that one can possess: economic, cultural, social, and symbolic capital. Linguistic capital is subsumed under symbolic capital and refers to the capacity to produce expressions appropriate for a particular market, so multilingual scholars’ abilities to write and publish in multiple languages can be seen as their multilingual capital in the linguistic market of academic publishing. One form of capital can be converted into another, and thus legitimized linguistic capital in a certain linguistic market can be cashed in for economic rewards, and vice versa. Framed within the academic publishing market where English is the dominant research language, the ability to write and publish in English is valued as a primary type of linguistic capital.

Another key concept in relation to linguistic market is “habitus”—that is, “a system of durable, transposable dispositions … principles which generate and organize practices and representations” (Bourdieu 1990: 53). Linguistic habitus is the sub-set of dispositions acquired in the course of learning to speak in particular contexts, which offers speakers a certain sense of the social value of linguistic utterances and one’s place in the linguistic market (Bourdieu 1991). For example, multilingual scholars’ linguistic habitus in academic publishing is represented by the research expertise acquired and accumulated through years of training in foreign languages. Furthermore, their multilingual habitus confers a certain social status upon them by allowing them to become university professors. Bourdieu (1990, 1991) also tells us that when a linguistic habitus encounters a linguistic market which may or may not be completely congruent with it, language practices are generated. The concept of “practice” hence offers a perspective to appreciate individual “agency” as a counterbalance to the overall structuralist construction of linguistic markets—“a subjective way of attuning oneself practically to the structured linguistic market” (Park and Wee 2012: 36). Applied to the present study, multilingual scholars’ agency can be observed through their critical reflection on their positions in the perceived linguistic market and their conscious efforts to make use of their multilingual capital in the market.

China’s higher education development and foreign language education

To contextualize the present study in the contemporary Chinese context, this section briefly sketches out China’s research policy development and the foreign language education landscape. The Chinese government since 1978 has placed a high priority on research development in higher education. In 2011 the Central Committee of the Communist Party of China (CCCPC) issued the “Decision on Deepening the Reform of the Cultural System and Promoting Cultural Prosperity”,Footnote 1 marking the government’s determination to promote Chinese research to the international academic community. This momentum continued when the government issued the “Coordinated Development of World-class Universities and First-class Disciplines Construction Overall Plan” in November 2015, designed to sharpen the international competitiveness of China’s higher education system.Footnote 2 This master plan, sometimes dubbed the “double first-class” (双一流), showcases China’s determination to have its overall higher education system considered among the world’s best by 2050. A key approach to achieving this goal is to maximize the research outputs of Chinese universities. Many universities began to implement research assessment policies based on key index lists (e.g. Zhang et al. 2013) including the Science Citation Index (SCI), the Social Sciences Citation Index (SSCI), the A&HCI (Arts and Humanities Citation Index), the Chinese Science Citation Index (CSCI), and the Chinese Social Sciences Citation Index (CSSCI).Footnote 3 On average, the number of publications in international journals grew at an annual rate of 14.6% from 1993–2010 (Zhang et al. 2013). According to a report of the Institute of Scientific and Technical Information of China (ISTIC 2016), the number of Chinese science researchers’ publications in SCI-indexed journals now ranks second in the world, and Chinese humanities and social sciences (HSS) publications in SSCI-indexed journals ranks sixth. However, because the major international indexes include mostly English-language journals, encouragement to publish in these indexed journals implicates a de facto language policy of publishing in English (Lee and Lee 2013; Lillis and Curry 2010). A close examination of Chinese HSS scholars’ actual language practice showed that over 95% of their international academic publishing concentrated on English-medium journals, while publishing in other languages was negligible (Zheng and Gao 2016), which suggests that English academic literacy is the primary linguistic capital for Chinese researchers in the linguistic market of academic publishing. The heightened importance of English-medium publications simultaneously implies that publications in languages other than English are less appreciated, which may in turn render the multilingual scholars’ research expertise in other foreign languages a deficit, echoing the orientation of multilingualism as a problem pointed out by Ruiz (1984, 2010).Footnote 4

In addition, multilingual scholars also need to deal with a conspicuous imbalance between the teaching of English and other languages in Chinese foreign language education (Lam 2005). English has replaced Russian as the most important foreign language in China since 1978, and after its initial development during the 1980s, English-language education experienced a boom in the 1990s in conjunction with China’s economic progress; it has now become dominant in the Chinese educational system (Bolton and Botha 2015). A recent survey shows that among the total of 416 million foreign language learners in China, 93.8% learn English exclusively, while 7.1% of the population have learned Russian, 2.5% Japanese, and only 0.3% have reported learning any other foreign language including German, French, Spanish, and Arabic (Wei and Su 2012). The disproportion is sometimes captured by popular terms among the general public, for example, the phrase “big language” (大语种) is used to refer to English and “small languages” (小语种) is used to refer to all other foreign languages.

However, the “Belt and Road Initiative”,Footnote 5 recently launched by China’s central government, constitutes a driving force to expand China’s multilingual education (Wen 2016; Shen 2016). The initiative emphasizes socioeconomic and cultural exchanges with sixty-four countries along the Silk Road Economic Belt and the Twenty-First Century Maritime Silk Road (mostly located in Central Asia and Southeast Asia), which are all non-Anglophone countries speaking more than sixty official and ethnic languages. To fulfill the mission of creating a “people-to-people bond” as explicated in the policy text, English alone is not enough. The Chinese government has realized that, because only around twenty of the languages spoken by the Belt and Road countries are formally taught in China’s higher education institutions, the nation suffers from a perceived lack of multilingual talents, which may impede Chinese organizations from deepening their participation in Belt and Road cooperation (Wen 2016; Shen 2016). It is thus imperative to diversify the nation’s foreign language competencies to meet the multilingual needs of economic internationalization along the Belt and Road. In July 2016 the Chinese Ministry of Education (MOE) issued a plan for “Developing Educational Cooperation along the Belt and Road”.Footnote 6 In April 2017 the MOE approved more than twenty foreign language degree programs, including Hebrew, Persian, and Serbian, in more than thirty universities.Footnote 7 The government’s shifting focus on expanding multilingual education implies an awareness of the need to harness individuals’ multilingual capital, particularly for its extrinsic value in national diplomacy, business, and economic development, reflecting an orientation to treat multilingualism as a resource (Ruiz 1984, 2010).

The backgrounds sketched above reveal the unique position of multilingual scholars in Chinese higher education, as they are simultaneously faculty members subject to institutional research assessment and impacted by the de facto English-only language policy, and foreign language teachers who are key agents in implementing the country’s multilingual education planning. Focusing on multilingual scholars’ individual practices may provide us with a fresh perspective into how the practices are mediated by the structural forces of research and language policies, the challenges and opportunities facing multilingual scholars from a non-English background, as well as the way they may experience multilingualism as a problem and navigate it as a resource in the structured linguistic market. As such, two research questions were formulated to guide the present inquiry:

  1. (1)

    What challenges did the national and institutional research/language policies create for multilingual scholars who specialized in foreign language literature and linguistics? And

  2. (2)

    How did these scholars respond to the challenges through their multilingual publishing practices?

The study

Research context

The site of our research was a top-tier university (anonymized as GHU) located in a southeastern city in China. GHU is a research-oriented university offering seventy undergraduate degree programs across the arts, humanities, and sciences. As of 2016 it enrolls about 30,000 undergraduate and postgraduate students, and employs 2600 faculty members. GHU is well equipped with the material resources essential to academic infrastructure, including library-purchased databases, funding for international collaboration, and overseas exchange. With a good reputation and a solid research foundation, it stands in the front line of China’s strategic plan for Chinese universities and academic disciplines to achieve world-class status.

The specific context within which this study was conducted was GHU’s Faculty of Foreign Language Studies. This faculty offers six foreign language degree programs (English, French, German, Russian, Japanese, and Korean) in linguistics and literature. Apart from the English division which enrolls the largest number of students and recruits 50% of all the faculty members, the other five language divisions have around six hundred undergraduate students and forty faculty members. All faculty members have doctoral degrees earned in either China or in their respective target language countries. All have overseas exchange experience and are well-connected with the target language research communities.

Participants

The study recruited fifteen participants through purposive sampling and a snowball method. To better represent the situations of multilingual scholars across different languages, we decided to recruit three scholars from each of the five foreign language divisions, excluding the English division. We decided to target our recruitment at scholars who were assistant professors or associate professors, since they were supposedly under the greatest pressure for promotion and academic publishing. Among the fifteen participants, six were Associate Professors and nine were Assistant Professors, all in their 30 and 40 s. An effort was also made to balance the linguistic and literature specialists. The first author knew two scholars from the faculty and was introduced to other scholars. To protect the participants’ anonymity, we chose not to specify their genders. When we presented the data, we replaced specific information about the participants’ relatively small academic subfield (e.g. contemporary Russian poetry) that might otherwise reveal their identity, and used generic terminology instead (e.g. Russian literature). Table 1 illustrates the demographics of the fifteen participants.

Table 1 Participant demographics
Full size table

Interviews and analysis

Semi-structured interviews were conducted by the first author. They were carried out in Putonghua, audio-taped on site, and later transcribed and translated into English. The core prompt questions revolved around four aspects: the participants’ language practices in academic research, publishing experiences both at home and abroad, attitudes towards the broad trend of Englishization in academic publishing, and thoughts and feelings about the current research assessment policies.

We were keenly aware of the power relations potentially involved during the interview, as we were English-language researchers and the participants were non-English. The interviewer made efforts to reduce the power relations by sharing with the participants her own struggles in international academic publishing, and encouraging them to talk freely about their feelings surrounding research assessment practices. We maintained a highly reflexive orientation by treating interviews not merely as an instrument for extracting data but as a social practice of knowledge co-construction (Talmy 2010). The interviews therefore served as important ways for both the interviewer/researcher and interviewees/participants to collaboratively achieve an understanding of their social positioning in the nexus between macro-level policy orientation and micro-level language practices. As a matter of fact, one participant (R2) expressed her gratitude for the interview, viewing it as an opportunity for multilingual scholars’ voices to be heard.

We analyzed the data through an iterative process of inductive and deductive reasoning (Merriam 1998) using NVivo 11.0. Three steps were followed. First, the two authors randomly selected three interview transcripts, and independently read the transcripts multiple times for open coding. In order to honor the participants’ voices, we used in vivo coding (Miles et al. 2014)—using words and phrases from the participants’ own language—to capture their lived experiences of multilingual publishing. Then we referred to the key constructs of linguistic market theory, such as “capital”, “unification”, “habitus”, “market”, and “value/devalue”, to move to the second level of pattern coding (Miles et al. 2014). Categories included “devaluation of multilingual capital”, “unified market based on key index lists”, “carving out research niches”, “shifting to English publishing”, and so forth. Next, we compared our categorizations, thoroughly discussed the differences, and revised the categories with reference to the theoretical framework until we agreed on the final coding scheme, which yielded a Kappa’s alpha of .78 for interrater congruence. The first author used the coding scheme to analyze the rest of the data. Finally, with reference to the linguistic market theory, we further abstracted the categories into two themes: “multilingualism as problem” to reveal how individual scholars’ multilingual capital was devalued in the linguistic market and “multilingualism as resource” to examine their agentive practices against the structural constraints.

Findings

This section first presents the challenges that the participants encountered in their academic publishing endeavors. The main policy pressures associated with the index-based research assessment practices are then discussed. Finally, we outline the participants’ new language practices to illustrate the interplay between structure and agency.

Multilingualism as problem

By referring to the structure of the linguistic market and the notion of capital, this part describes the participants’ subjective experience of multilingualism as problem, and attempts to disentangle the issue by revealing the implicit language policy embedded in international and national publishing markets as the hidden structural forces.

Increasing research pressure and devaluation of multilingual capital

All the fifteen participants expressed feelings of being pressurized by the increasingly higher requirements for publishing. J1, a middle-aged scholar specializing in Japanese literature, commented as follows:

Interviewer::

Do you feel that the research pressure has increased over these years?

J1::

Yes, I do. I found it related to the university ranking, the “double first-class”. We associate professors are probably fine, but those assistant professors are facing even higher pressure. […] I don’t think it’s right. The specific circumstances of small languages are totally ignored. (Interview with J1)

J1 was already tenure-tracked, so she realized that the pressure was heavier for her younger colleagues who needed to compete for promotion and tenure positions. The perceived connection between increasing research pressure and the “double first-class” drive shows that participants as university faculty members were affected by the repercussions of the top-down policy enacted by the nation in the context of its master plan for higher education development.

Despite their expertise in “small languages”, all the participants mentioned that their research output in non-English publications was sometimes “not recognized” (不认) and “not counted” (不算) in the university’s research assessment exercises, and they repeatedly expressed their dissatisfaction about “losing out” (吃亏). For example, K3 articulated her frustration when the university began to use the CSSCI list to decide on research bonuses:

From this year on, [the university] began to use CSSCI-indexed articles as the basis to calculate our annual bonus. It’s impossible for us Korean teachers to publish in those CSSCI journals. We can’t publish on the SSCI either. Even English professors find it hard to publish in SSCI journals, not to mention us small-language teachers. […] So you see, the university decides the research bonus on the basis of CSSCI-indexed articles. It means our work is not counted at all, and we are losing out. (Interview with K3)

The above extract illustrates how the newly-implemented research assessment benchmark prevented K3’s linguistic capital in Korean-language academic publishing from being converted to the economic capital of research rewards which generated her vexation of “losing out”. The multilingual scholars seem to have experienced difficulties in legitimizing their multilingual capital (Bourdieu 1986) and in converting it to economic capital in the academic publishing market (Bourdieu 1991).

Publishing in English in the international publishing market

The participants had all attempted to search the SSCI and/or A&HCI list for relevant journals where they could publish internationally in their respective research languages, but to no avail. They realized that their respective target research communities were not enthusiastic about participating in these Anglophone-centered journal ranking exercises. For example, R2 mockingly referred to such exercises as a “game”, and said, “Russians don’t want to play games with the Americans or British” (Interview with R2). As a result, even if they would like to publish in these lists, there were very few journals for them to target. The participants found themselves caught in the dilemma of publishing in the designated lists or publishing good research, because “it was highly likely that the best research published in a well-respected journal in the target-language research community falls out of the key index lists” (Interview with F3).

The predominance of English-medium journals and the non-inclusion of other languages collaboratively form a mechanism in which “English-medium”, “international”, and “high academic quality” become equivalent (Lillis and Curry 2010), whereby scholarship in non-English, non-indexed journals is rendered less worthy. K2 questioned the differential values assigned to English and other languages:

K2::

Does it mean that papers published in English are good research? Those written in Korean or Chinese are not good research?

Interviewer::

Then have you heard of a kind of saying that it’s easy to publish in South Korea or other countries, and so you [multilingual scholars] can publish a lot?

K2::

How can it be easy to use the target language and publish in the target community? Let me ask you one question. Do you find it hard to use English to publish overseas?

Interviewer::

Yes, I find it very hard.

K2::

It’s the same thing. Why are English-language articles counted as more valuable, but when research is published in Korean or other languages, it becomes less worthy? (Interview with K2)

K2 specialized in Korean linguistics and had successfully published several Korean-language articles in South Korea, but she encountered the tacit belief that publishing in Korean-language journals based in South Korea is relatively easy. Her comment made explicit the double standard embedded in the index-based assessment practice. English is as foreign a language to English scholars as the other languages are to multilingual scholars in China, but English-medium research is more likely to be indexed in the key journal lists and recognized as valuable research, whereas research published in other languages and in other localities, which is less likely to be indexed, becomes less valued. As a result, their efforts and success in multilingual publishing are dismissed as “easy” and even less worthy, compared to their counterparts in English studies. The underprivileged position assigned to other languages profoundly reflects the center-periphery ideology in knowledge production, in which the Anglophone knowledge center is a powerful default location, whereby research published in other geological places and in other languages is sidelined and even trivialized (Canagarajah 2005; Lillis and Curry 2010). Framed in terms of the linguistic market theory (Bourdieu 1991), the international academic publishing market is governed by the center-periphery ideology that prioritizes English-language research, and thus other languages such as French and Korean receive less value.

Researching about English in the Chinese foreign language publishing market

One would expect that when the scholars used their native language of Chinese for research communication, inequalities created by the dominance of English in the international index lists would diminish. However, the data revealed quite a different picture. The participants all observed that because their research was not about English, they were again consigned to a peripheral position in the Chinese foreign language publishing market regulated by the CSSCI list, mostly because the CSSCI does not index journals that specialize in other languages. For example, the participants researching in German (G1, G2, G3) and Japanese studies (J1, J2, J3) all mentioned that their discipline-specific journals (i.e. the Japanese Learning and Research Journal and the German Learning and Research Journal) were not CSSCI-indexed, despite these journals’ recognized influence within the local research communities. As a result, their articles published in these journals did not count for anything.

The participants noted a huge discrepancy between the limited number of publications that could appear in CSSCI-indexed journals of foreign language studies and the large number of researchers in these languages:

Among all the foreign language literature journals indexed in the CSSCI, there are only four journals related to Japanese literature. I did a calculation: each one of the journals publishes eight articles at most on Japanese literature per year, which means that the whole Japanese literature research community only gets to publish fewer than forty articles per year. Do you know how many people are doing Japanese literature studies in China? (Interview with J1)

Japanese is currently the second most popular foreign language in China. China had the largest number of Japanese language learners in 2012 (1.046 million out of 3.985 million Japanese language learners worldwide), a 26.5% rise since 2009 (Lv et al. 2017). The number of Japanese departments in Chinese universities increased dramatically from 150 in 1999 to 660 in 2010, and there are sixteen thousand Japanese university teachers nationwide (Tian 2011). Considering the meager amount of research that can be published every year, there is very little room for the professional development of Japanese scholars.

Three other participants (F2, G2, R1) noted that different languages were positioned in a hierarchy in China’s foreign language studies publishing market. For example:

Once a very well-known Chinese scholar of German literature told me that German language and literature rank the lowest [in China]. Aside from English, Russian is at the top of the rankings, followed by French, and lastly Japanese and German studies. (Interview with G2)

Multilingual scholars compete with each other to publish Chinese articles irrespective of their research languages and topics. As page space is limited, resources that could be allocated to the different languages are not evenly distributed. To illustrate, among the ten articles included in one issue of a foreign language literature journal, maybe half of them are on English studies due to the dominance of English scholars, while the other half are distributed among non-English studies. Domestic publishing in foreign language studies thus becomes an uneven playing field for scholars of the “small languages”. Although the linguistic medium of publication is their native language, they suffer from unequal publishing opportunities caused by the multilingual nature of their research topic.

The participants attributed these unequal publishing opportunities to the journals’ obsession with impact factors. As K3 recounted:

Interviewer::

Why do you think CSSCI journals don’t publish Korean language studies?

K3::

Simple answer, impact factor. […] A professor from the XX Foreign Language Studies University told me that several years ago, their university-sponsored CSSCI journal published two articles on Korean-language studies in one year. But just that year, the journal’s impact factor dropped significantly. Since then they have never published any Korean studies. (Interview with K3)

Seven other participants concurred with K3. The first author also contacted several journal editors via personal communication. They did not want to go public with their comments, but they confirmed that publishing research not related to English indeed adversely affected their journals’ rankings the next year.Footnote 8

Impact factor is a bibliometric method devised to measure the influence an article has on the scientific community (Garfield 2006), calculated by a simple ratio of the average number of times articles from a journal published in the past 2 years have been cited in the current year. There have been criticisms that impact factor fails to account for disciplinary variation and is strongly biased towards English in the international publishing market (Curry and Lillis 2018; Lillis and Curry 2013), but nevertheless it continues to be used to identify journals to include in international index lists. In the Chinese publishing market, the CSSCI list basically follows this principle. In order to stay in the CSSCI, journals understandably makes strategic efforts to increase their bibliometric measures, such as maximizing their readership and encouraging more downloads and citations by covering the most relevant topics. Targeting the largest group of readers (i.e. English specialists) evidently attracts more reading and citing, increases impact factors, and secures the journals’ positions in the CSSCI list. Because “the more linguistic capital that speakers possess, the more they are able to exploit the system of differences to their advantage and thereby secure a profit of distinction” (Thompson 1991, p. 18), the indexing process to a large extent perpetuates the divide between the haves (English scholars) and the have-nots (multilingual scholars), which may ultimately accentuate the structural inequality in the national publishing market.

Unified market based on key index lists

The data showed that the most profound impact on the participants’ multilingual publishing practices was exerted by their university’s adoption of three key index lists (i.e. the SSCI, A&HCI, and CSSCI) as the yardsticks to measure research output across the faculties and disciplines. Only articles published in journals indexed in these three lists could be counted as research output for research evaluation or promotion. For the reasons analyzed above, the participants were dismayed by this policy, and felt deeply disadvantaged in academic publishing.

F2’s case is worth some further exploration. F2 was a scholar in French literature, and she pointed out that the practice provided “a relatively fair assessment criterion” despite that her French-language papers published in France had a hard time to be recognized as qualified research. She recognized the diligent work of the research assessment panel and said:

F2::

I believe the panel members are able to make an objective evaluation about us. Overall speaking, it’s fair. But it’s still hard, to us small language researchers. How much they can understand our work …

Interviewer::

They don’t understand?

F2::

For instance, you use French to publish several articles, but they don’t understand what you write about, how can they compare your work with that of other languages?

Interviewer::

So you believe that you should write some English papers?

F2::

Also in Chinese. … Otherwise even if you publish ten or twenty French-language articles in France, people might still question, is that because it’s easier to publish in France? Is it because you have some connections in France? There’s this doubt, and they can’t really evaluate the quality of your research unless your articles are indexed in these key index lists. (Interview with F2, our emphasis)

F2’s comment seems to support the usage of index lists in research evaluation and thus contrasts the negative feelings expressed by her colleagues. However, it needs to be noted that she made the comment after she recounted the story that an article written in French and published in a non-indexed journal in France was not counted as valuable research in her promotion evaluation. This incident triggered her to reflect upon the research evaluation mechanism and made her conclude that the quality of her research might not always be properly evaluated due to a lack of understanding across different languages, “unless [the] articles are indexed in these journal index lists”. In other words, she chose to resort to a set of exonormative, seemingly objective standards to legitimize her research.

However, as has been shown in the previous sections, the apparently objective measurement based on the key index lists innately privileges research published in and about English, which in practice maintains the symbolic power of English researchers, peripheralizes multilingual scholars’ linguistic capital, and entrenches the existing linguistic hierarchy. As the linguistic market functions through the mechanism of “unification” (Bourdieu 1991), we regard the adoption of the key index lists as a mechanism that favors some research areas over others. The multilingual research expertise they have accumulated through years of training—their multilingual habitus (Bourdieu 1990, 1991)—has become incompatible with what is valued in the international and national linguistic markets of academic publishing.

Multilingualism as resource

The participants were keenly aware of their disadvantaged positions and made conscious efforts to navigate their multilingual expertise as language resources (Ruiz 1984, 2010). In other words, their “critical reflection and collective action against such institutional power” (Park and Wee 2012: 32) imply the interaction between structure and agency through shifting language practices. The following section aims to describe the shifting language practices as a result of the encounter of their habitus and the linguistic market.

Carving out research niches

Realizing that the “multilingual habitus” they had acquired was not completely congruent with the demands of the academic publishing market, the participants maneuvered to seek for “a profit of distinction” (Bourdieu 1991: 55) through their language practices. One strategy was to redirect their research into a comparative research paradigm. For example, R3, a Russian linguistic researcher, said:

If I merely work on Russian linguistics, it’s quite hard, because [the Russian scholars] have also produced many studies. […] So I wouldn’t have any advantage in merely writing about Russian linguistics. If I can blend in some Chinese elements [with my research on Russian linguistics], I may gain some advantage or creativeness. (Interview with R3)

R3’s position was echoed by other participants. By blending “Chinese elements”, the participants attempted to harness their Chinese capital to gain some symbolic power to achieve their desired recognition or acceptance by the target-language scholarship. In other words, they exercised their agency to counterbalance the overall structural forces of the linguistic markets (Park and Wee 2012).

In addition to optimizing their Chinese capital, the participants also attempted to mobilize their multilingual expertise to serve as a bridge between different languages and cultures. F3, who specialized in French literature, described her multilingualism as an opportunity:

F3::

In China a lot of scholars are doing reception studies in comparative literature. When you bring in a new batch of French materials, people would find it interesting.

Interviewer::

Is it because there is already too much on English?

F3::

Or maybe, shall we say, there is too little in French.

Interviewer::

So can I say that French gives you a unique voice?

F3::

Yes, yes, yes. (Interview with F3)

Not only does the Chinese academic field need multilingual scholars to bring in new information, the international academia also needs their contribution for enriched knowledge. R1, a specialist in Russian diasporic literature, pointed out:

I think English materials on [my topic] are seriously limited. [The information] was buried in Chinese, so nobody knew about it before. But when they see [my work], they would think, “Oh! We didn’t know that there used to be a Russian diaspora in China.” People only knew about the Russian diasporas in Berlin, Paris or Prague, but never about China. (Interview with R1)

The above extract best illustrates the arguments to sustain multilingual scholars’ publishing efforts because they make unique contributions to the mainstream academic community (Belcher 2007; Flowerdew 2001). By tapping unknown resources and accessing works that the mainstream academia does not know about, multilingual scholars contribute to enriching global scholarship. Furthermore, the multilingual scholars attempted to attune themselves to the requirements of the linguistic market by drawing on their multilingual habitus and thereby rendering their multilingual capital valuable, which implies a scope of individual agency to initiate social change through everyday practices (Park and Wee 2012).

Developing Chinese and English writing skills

The participants agentively generated new language practices as a response to the structural constraints imposed on their multilingual habitus. For one thing, all the seven participants who earned their PhDs from overseas institutions decided to shift to Chinese in their academic writing. This decision was driven by the CSSCI-based research assessment, since they perceived the adoption of the CSSCI list as the authority’s attempt to legitimize Chinese-language articles and reinstate the position of Chinese as an academic language. However, because they were academically trained in a foreign language, their Chinese L1 academic literacy often fell short. For example, J3 described her painstaking efforts to write and publish in Chinese, where the reviewers usually “criticized [my] use of terminologies, and said some expressions were not appropriate [in Chinese]” (Interview with J3). Similarly, G1 mentioned that Chinese academic writing emphasized “a lively writing style”(文笔) which most overseas returnees did not have due to a lack of training in L1 academic writing. What made things worse was that there was no institutional language support for them to develop Chinese academic literacy, and consequently they were all forced to start a painful and slow process of self-training. The adoption of the CSSCI list in research assessment did incline these multilingual scholars towards using Chinese for research communication, which can be considered an implicit language policy that came about through additional devices (Shohamy 2006), but probably due to the lack of language support, the policy turned out to create extra burdens for multilingual scholars.

Furthermore, five participants (F2, G3, J3, K3, R1) had decided or already started to explore academic writing in their third language, English. Two participants (K3 and R1) were motivated to reach out to the international research community. As K3 put it, English represents a gateway to “a brand new world” that otherwise would not be accessible. Two other participants (F2 and G3) pursued English publishing out of a pragmatic concern, because to them English offered a channel to redress their chances vis-à-vis the unequal opportunities to publish in Chinese or the devaluation of their non-English research. A third reason to shift to English was directly linked to assessment pressure. As J3 explained:

Interviewer::

Why do you want to publish in English?

J3::

For survival. Ultimately, it’s for survival.

Interviewer::

But you’ve already survived. You’ve published many Japanese and Chinese articles.

J3::

I hope that I can have some publications that are able to stand all tests.

Interview::

You mean that only English articles can stand all tests?

J3::

Well, others seem to think so. […] Although I can publish in Japanese, I won’t get wide recognition. So I have to publish in Chinese, and also in English. (Interview with J3)

J3 was a successful scholar in Japanese linguistics with around twenty articles published in Japan, but she was still worried about “survival” (生存) and whether her research could “stand all tests” (过硬). This example showcases the pervasive anxiety among multilingual scholars whose multilingual capital was not sufficiently valued. Trying to develop multilingual literacies put huge pressure on their time and energy, as most of them complained that it was “hard to concentrate” (G3), their “energy fell short” (F2, K3, J3) and “time was too limited” (R1).

Using social network resources

The focal university offered ample resources for the participants to engage with international collaboration and exchange. These opportunities proved to be conducive to helping them mobilize their multilingual resources. For example, both R1 and F2 noted that it was through a joint conference between the faculty and a partner university based in an Anglophone country that they began to write in English. With the help of online translation software and English-speaking colleagues, they managed to turn their conference presentations into English-language research papers, and ultimately got them published in English-language journals. Other forms of international exchange include funding international conferences and overseas visiting scholarships. For example, K3 said:

When I attend the English-medium conferences on Korean linguistics or Asian linguistics, I discover new interest areas or new directions. I attended two such conferences, one in Europe and the other in the US. […] I plan, in the future, to apply for a visiting scholarship in the US so that I can learn about English academic writing. (Interview with K3)

International collaboration and exchange opportunities are significant social networks in academic publishing which function as social capital in their own right (Curry and Lillis 2010; Lillis and Curry 2010). In the cases of R1, F2 and K3, participation in these international exchange activities enabled them to convert social capital to the English capital that was favored by the academic publishing marketplace. Nevertheless, it needs to be noted that the focal university is a prestigious institution located in a metropolitan city, so the participants were also privileged in being able to consume social capital and material resources. It is difficult to imagine how multilingual scholars in less prestigious universities or less developed areas might be able to access such social networks. They may be less able to counterbalance the impacts of the power relations linked with English in academic publishing, and may encounter greater challenges in their professional development in the current research-based promotion system.

It is worth noting that individual scholars differed in their abilities to utilize and convert their social capital. An important variable was found to be their linguistic habitus of previous English learning. K1, another scholar in Korean linguistics, described her frustrating experience as follows:

Last year I was an exchange scholar in the US. […] But I still couldn’t use English to communicate. In my time, our English learning was extremely exam-orientated, and I forgot all about it after taking the exam. […] When I was in the US, I borrowed one [English] novel from the library. It was about Korean immigrants in the US. But I even didn’t finish the first chapter. (Interview with K1)

K1 confessed that she would never “dream about” writing and publishing in English despite her strong motives to do so. K1 was almost 10 years older than K3, who as a younger-generation scholar had received her pre-tertiary education in China during the 1990s and benefited from the nationwide improvement of English-language education at that time. By contrast, K1’s linguistic habitus was partially constituted by her English learning in the 1980s, when English-language education in China was just reinstated, and this turned out to restrict her ability to utilize the social network affordances, thereby inhibiting the conversion between social capital and linguistic capital. This finding calls for a more nuanced understanding of individuals’ language practices situated in the macro socio-educational context of their linguistic habitus.

Discussion

In this paper we have explored how Chinese multilingual scholars experienced multilingualism as a problem in their academic career, and navigated multilingualism as a resource to generate new language practices. Bourdieu’s market-based perspective is useful in this inquiry as it helps to reveal how their multilingual capital is devalued in the unified publishing markets that perpetuate the de facto language policy of publishing in and about English. The perspective also sheds light on how multilingualism was leveraged as a resource at the individual level when the scholars’ multilingual habitus encountered the unified market. By exercising agency, they attuned their language practices towards the linguistic market to counterbalance structural forces.

Unification of linguistic markets at different levels

Our findings illustrate the interconnectivity in global knowledge competition and university ranking exercises, the Chinese national drive for higher education development and associated assessment policies, and individual academic publishing practices (Englander and Uzuner-Smith 2013; Feng et al. 2013; Lee and Lee 2013; Lillis and Curry 2010, 2013). The institutional-level research assessment resulting from global-level university ranking exercises enacts a de facto language policy of publishing in English (Curry and Lillis 2018; Feng et al. 2013; Lee and Lee 2013; Lillis and Curry 2013), and the Anglophone-centered textual ideology adversely affects multilingual scholars in non-Anglophone contexts by constraining and illegitimizing their multilingual capital. Aligning with the market orientation towards academic publishing (Englander and Uzuner-Smith 2013), we see the adoption of key index lists in research assessment as a mechanism that unifies the international and national linguistic markets of academic publishing.

A key finding emerging from the analysis is the implicit “researching about English” policy in the Chinese foreign language publishing marketplace. Our findings show that the increasingly supranational evaluation systems based in Anglophone centers and the “centripetal pull” towards Anglophone-center practices (Feng et al. 2013; Lillis and Curry 2010) have affected Chinese scholars even when the medium of publishing is Chinese. Given that a unified linguistic market is “more favorable to the products offered by the holders of the greatest linguistic competence” (Bourdieu 1991: 69), the adoption of the impact-factor-based CSSCI list implicitly favors the symbolic capital possessed by the dominant group of English scholars, and ultimately consigns the smaller groups who do not research “about” English to a peripheral position in the non-Anglophone context. As unified markets at the national level are linked to global-level markets, particularly because of the greater transnational interactions in economic, cultural and political power relations in the globalized world (Park and Wee 2012), our findings demonstrate the top-down influence according to which the market force linked with the use of English is at play at different levels. Global or local, the unified linguistic markets slanted towards English represent a stable constraint on multilingual scholars seeking to maintain or convert different forms of capital.

Capitals, habitus, and agency within the linguistic markets

While previous findings have suggested that English-language scholars are enabled to convert their English capital into monetary rewards by receiving more opportunities to publish in indexed journals (Feng et al. 2013; Flowerdew 2015; Lee and Lee 2013), the present study shows that multilingual scholars experienced difficulty in converting their multilingual capital in academic publishing markets incongruent with their multilingual habitus. In addition, some multilingual scholars possessing social capital embodied in social networks could effectively convert this into the more valuable symbolic capital. This confirms the crucial importance of social networks in the social practice of academic publishing (Belcher 2007; Curry and Lillis 2010; Lillis and Curry 2010).

The linguistic market theory affords some insights into individual subjectivity and agency in academic publishing. In contrast with McGrath’s (2014) finding that Swedish scholars’ language choice was constrained by practical needs, our study shows that fluid language practices were generated from the interaction between one’s linguistic habitus and the structural market (Bourdieu 1990). Linguistic habitus is rooted in subjectivity—“the ensemble of modes of perception, affect, thought, desire, fear, and so forth that animate acting subject” (Ortner 2005: 31). The participants demonstrated heightened self-awareness regarding their specific relationship with the market when they consciously discussed the perceived hierarchy of foreign languages and reflected on their own positions in the social space. Therefore, their subjective reactions of “losing out” and pervasive anxiety can be seen as being generated by their recognition of the mismatch between their linguistic habitus and the demands of the linguistic markets, which trigger their shifting language practices. This finding thus brings to light the fluidity of individuals’ language practices and the significance of agency in navigating multilingualism as a resource against the structural forces. However, we need to bear in mind that individual scholars may possess different abilities to consume the social capital afforded by their context, mostly due to their individual linguistic habitus. Their previous training in English (or lack thereof) to a large extent determines the degree to which they can effectively convert their social capital to the preferred linguistic capital. Put another way, the individuals’ here-and-now literacy practice is grounded in the historical contingencies of the macro-level socio-educational development in the time and locale that they are situated in.

Conclusion

This paper examines the interplay between the structural forces of the linguistic markets at multiple levels and Chinese multilingual scholars’ fluid multilingual practices. We see their challenges and opportunities in academic publishing and professional developments as broadly reflecting the tension between the priority attached to English-language research stemming from the nation’s higher education internalization drive and the needs of expanding multilingual education for the national economic interest. Although this study only involved fifteen participants and were situated in the Chinese sociopolitical context, the findings may resonate with the growing pressure and dilemma of publishing in English or other languages across geolinguistic contexts. Regrettably, the present study did not include multiple perspectives of other stakeholders engaged in research assessment activities, such as faculty-level research panel members. It needs to be acknowledged that we had no intention to judge the decisions made and measures employed in the local context. Instead, we attempted to reveal the multi-level impacts of the implicit language policy and demonstrate the individual agency against structural forces.

Policy makers at the national and institutional level need to remain vigilant to the implicit language policies embedded in research assessment practices, and treat the language issue with sensitivity so as to reconcile differences and legitimize multilingual scholarship. As key agents in implementing the nation’s multilingual education, multilingual scholars need to be given enough room for professional development, particularly under the current research assessment practice. They also need to be empowered to participate more equally in academic publishing, regardless of the languages that they work with. Otherwise, their professional development will stall and the nation’s multilingual education drive will be impeded.

Individual multilingual scholars may consider how to better participate in the local research community, fully utilize social networks, and collaborate with scholars in English studies. Multilingual scholars’ academic publishing practices and professional development are crucial not only to the country’s educational initiative of expanding multilingual education, but also to the enrichment of global scholarship. As English scholars, we are fully aware of the advantage conferred by the structural power linked with the use of English, and thus we find it all the more imperative to unravel the structural forces that perpetuate inequalities in the multilingual academic world. If everything was left to market forces, our world would probably be deprived of the rich legacy of multilingual scholarship.