Denying language privilege in academic publishing

If you’re an academic anywhere in the world, you’re probably under pressure to publish to make progress in your career or just to keep your job. Increasingly, you’re probably also under pressure to publish “internationally”. Thanks to the global dominance of English-speaking academia, “international” is more or less a euphemism for journals published in English. Faced with this requirement, academics from outside English-speaking countries like the US, the UK, Canada, or Australia have commonly reported that writing for publication in English is a source of disadvantage for them. Furthermore, they have been shown to be less successful at having their work published in these journals than those scholars residing in English-speaking countries.

This has led some critics to point out the inherent advantages and disadvantages of such a system, especially that ‘nonnative’ English users tend to be at a grave disadvantage particularly when compared to those who have acquired English from childhood, that is ‘native’ English users.

In a recent article titled “Academic publishing and the myth of linguistic injustice“, Ken Hyland (applied linguist at the University of Hong Kong) tries to argue, as the title makes clear, that the disadvantages facing ‘nonnative’ English users in the domain of academic publishing are overstated or that they are a “myth”.

Hyland’s article has received some media attention, and thus, as an early-career scholar being inculcated into a system that is marked by the contradictions of both expanding participation and persistent inequalities, I am concerned that the assertions made by a well-respected authority on academic discourse will be used as a way to dismiss attempts to work toward more just systems within academia. Thus, I wanted to outline what I think are the shortcomings of Hyland’s argument. Ultimately, I believe that Hyland’s attempt at challenging what he calls an “unchallenged orthodoxy” (p. 66) ignores some important aspects of ideology and political economy pertaining to English as a language of global academic communication that are important to keep in mind when considering whether the academic publishing system is linguistically just.

What is linguistic injustice?

Despite asserting that linguistic injustice in academic publishing is more or less a “myth”, Hyland never thoroughly engages with what linguistic justice or injustice within this context might be. He does discuss one particular aspect of this potential justice: the possibility that reviewers and editors pay undue attention to ‘nonnative’ writers’ non-conformity to the grammatical, lexical, or discursive practices of ‘native’ writing, thus making it much more difficult for ‘nonnative’ writers to publish their research. I take up this claim more thoroughly below.

However, first, it’s worth considering what linguistic justice would entail more broadly. Specifically, I think we need to consider the goals of academic publishing and research in general. I’m going to summarize the goal of academic publications (journals, books, etc.) as disseminating high-quality research that advances or somehow makes an important contribution to the disciplinary conversation.

This conversation is dependent upon human language, but it is not necessarily dependent on any particular language. In other words, academic writing by necessity uses language, but which of the many existing languages is/are chosen should, in theory, be irrelevant, since high-quality academic thought and research can be communicated in any language variety or combination of languages or language varieties. Crucially, determinations about the value of a contribution to the conversation should also not be affected by the question of whether the ideas are expressed in a so-called ‘standard’ variety or not. This, however, is merely a vision of an ideal optimally-inclusive world. The actual world of academic publishing is dominated by English, and more specifically prestige varieties of English (e.g., ‘standard’ US and UK Englishes), and to a lesser extent a few other widely-used languages (e.g., French and Spanish).

The reasons for this linguistic domination are purely political and economic. They are not linguistic. There is nothing special or superior about the grammar of English nor the writing practices of English-speaking academics. Although the use of English within academic disciplines has helped provide the language a rich lexicon of specialized terms, these terms could easily be borrowed into any other language that lacks this specialized vocabulary. Indeed, English writers historically did exactly this with languages of greater prestige like French, Greek, and Latin, and multilingual academics today routinely borrow English terms into other languages. Rather, English has accumulated power not through its linguistic merits but rather through colonialism and economic domination. It was the language of the British Empire, and today it is the language of Harvard and Cambridge.

However, in order to facilitate a global conversation, which academic publishing purports to do, it is unlikely that we could truly be neutral with respect to language choice. Some common medium through which to communicate has to be adopted, and the amount of resources that have been exerted in training the world’s academics to publish in English suggests that English is, for the time at least, the obvious choice.

This ‘selection’ of English as the language of academia inarguably privileges ‘native’ English users, if for no other reason than they, unlike ‘nonnative’ English users, do not need to invest as many resources in the acquisition of an additional language through which they can participate in the global conversation. The presence of such unearned advantage is a form of linguistic injustice that permeates the global academic publishing system.

Of course, even acknowledging the inherent inequality that the use of English creates, academic disciplines, publications, and other institutions can attempt to mitigate linguistic injustice by seeking to govern themselves according to norms that are not fully stipulated by powerful English-speaking countries (especially the US and the UK) but are maximally accepting of variation in the use of English. Global academic communities could engage in just practices of evaluating writing that are tolerant of linguistic variation and do not allow it to detract from perceptions of research quality or importance. To the extent that these communities insist on practices tied to particular linguistic groups, which are either not accessible to or simply not adopted by other groups, they can be said to be linguistically unjust.

Academic writing and the accumulation of linguistic capital

One of the central claims of Hyland’s article is that factors other than language better explain existing inequalities in academic publishing such as the lower acceptance rate of scholars from outside of English-speaking countries. Specifically, he writes that “the disadvantages of physical, scholarly, and financial isolation [experienced by ‘nonnative’ English writers especially in less wealthy regions] may be greater than those of language. Certainly, these factors are frequently associated with poor linguistic skills, at least in non-English speaking periphery countries, but a crude Native vs. non-Native dichotomy fails to capture a far more complex picture” (p. 63).

I do not disagree with Hyland that issues like institutional funding are crucial to academic’s success in publishing or that reducing the issue of inequality merely to a “crude” dichotomy of ‘native’ versus ‘nonnative’ is not the most effective way of dealing with the issue of linguistic inequality.

However, I do not believe that, in making these arguments, Hyland is engaging with the strongest form of the argument. Specifically, despite his acknowledgement that material resources are “frequently associated” with language background, I do not believe that Hyland is considering the degree to which the issue is about how ‘nativeness’ serves as a form of linguistic capital. While the material resources that enable research to be conducted are undoubtedly crucial to academic writers’ success, it is in large part accessed through the acquisition and exchange of linguistic capital. Having English language publications, for example, is an important way by which academics access more prestigious employment, more lucrative grants, and other sources of funding.

It is not hard to see that ‘native’ English speakers’ linguistic resources are highly favored on the academic publishing market. Indeed, the very journal that Hyland has published this paper in advises potential authors to “write your text in good English (American or British usage is accepted, but not a mixture of these)”. Thus, for the Journal of Second Language Writing‘s gatekeepers, “Good English” more or less means “American” or “British” Englishes. Furthermore, I have edited the writing of a ‘nonnative’ English speaker who submitted to this very journal. The author was told to have a ‘native’ speaker edit the manuscript. (As a further illustration of my point about how resources accrue to ‘native’ English speakers through the academic publishing process, I, a ‘native’ English speaker, was paid to do this editing by the ‘nonnative’ author.)

My point then is that funding is not merely one of the causes of academic publishing success, it is, in a very real sense, an effect of publishing success as well. Since resources beget publications beget resources, beginning your academic career as a ‘native’ English speaker is an important source of linguistic capital that allows early career scholars (like me) a competitive advantage in accessing funding and other resources, especially early in their careers. In other words, the ability to more quickly and easily produce academic writing that is deemed “publishable” is an important way in which the norms of the academic publishing community perpetuate inequalities between ‘native’ and ‘nonnative’ English-speaking academics, even if, over the course of their careers, ‘nonnative’ academics also accumulate their own stocks of linguistic capital.

Of course, as Hyland points out, both groups–‘native’ and ‘nonnative’–are also internally heterogeneous, suggesting that membership within either group does not come with a uniform supply of linguistic capital. For example, an African American English speaker clearly does not have the same linguistic capital that a speaker of Received Pronunciation (a high-status British dialect) does. Similarly, due to the fact that not all ‘nonnative’ varieties are perceived equally, a ‘nonnative’ English speaker whose first language is German may more readily secure linguistic capital than one whose first language is Arabic, Swahili, or Vietnamese. This, however, is no reason to reject the distinction between ‘native’ and ‘nonnative’ (although, as my use of scare quotes probably suggest, the terms are somewhat problematic), since analyses that Hyland cites show that despite conceptual flaws it accounts for a great deal of the inequality in academic publishing. Rather, we should view this internal heterogeneity as a call to complicate our understandings of structural disadvantage by exploring how language background intersects with other social structures: class, race, gender, and so on.

In his article, however, Hyland downplays the degree to which ‘native’ English speakers’ linguistic capital is relevant to academic writing. He reviews some of the literature on how ‘native’ English speakers too must learn to navigate the demands of academic writing. Hyland writes that being a ‘native’ speaker of a language “refers more accurately to the acquisition of syntactic and phonological knowledge as a result of early childhood socialization and not competence in writing, which requires prolonged formal education. We don’t learn to write in the same way that we learn to speak, but through years of schooling.” I would agree with Hyland that there are challenges for ‘native’ English speakers who lack experience with academic publishing and can personally attest to this.

Nonetheless, acknowledging that mastering written academic English is not entirely easy or “natural” for ‘native’ English speakers does not directly address the question of whether they enjoy an advantage over ‘nonnative’ English speakers. If, as Hyland clearly states, ‘nativeness’ provides an advantage in syntactic knowledge, then certainly this alone is already a major advantage in writing. For example, being able to trust my own judgment about the grammaticality of a sentence or phrase makes producing and editing written academic English less laborious for me than it is for many ‘nonnative’ writers. It also makes me less susceptible to negative perceptions about my writing or abilities as a scholar, a point I address more fully in the next section.

Indexicality and reviewer bias

One of the aspects of potential linguistic injustice that Hyland more directly addresses is the question of whether reviewers pay undue attention to so-called grammatical (or other linguistic) ‘errors’ in ‘nonnative’ academic writing. Hyland argues that concerns about such biases are unfounded. Specifically, he claims that, while journal editors and reviewers may raise concerns about ‘nonnative’ writers’ language and may do so rather bluntly at times, these concerns are usually not given great weight in decisions about whether a manuscript should be accepted for publication. Rather, looking at research on the content of written reviews, Hyland states that journal reviewers “tend to strongly focus on aspects of the research itself, rather than its presentation” (p. 65), suggesting that their decision-making reflects attention to the quality of the research, not biases against ‘nonnative’ English.

I am unsatisfied with Hyland’s evidence, specifically because it fails to consider an important property of language: indexicality. Indexicality refers to the way that language choices not only convey referential meaning but also convey (intended or unintended) information about the identity of the speaker. For example, the word “y’all” is a second person plural pronoun which is a fancy way of saying that it refers to more than one addressee (or an addressee plus other people). However, when it is used, we do not merely interpret it as a reference to people but also as an indication of the speaker’s identity. If you’re even vaguely familiar with regional dialects in the US, you likely think of “y’all” as expressing the speaker’s Southern-ness, which is itself tied up with other characteristics that are often quite negative: lack of education, lower socioeconomic status, or social conservatism, as well as perhaps some more positive ones, like hospitality. When a speaker uses “y’all”, all of these levels of meaning are potentially conveyed, but we aren’t always aware of how language’s indexical properties are implicated in our perceptions of other people.

This lack of awareness is important because it raises the distinct possibility that reviewer’s perceptions of the quality of research presented in a manuscript are impacted by social information that they glean about the writer from features of the language. There is a great deal of research demonstrating that, when presented with a ‘nonnative’ accent, listeners often perceive both the speech and the speaker negatively, attributing negative personal characteristics to the speaker. For example, this study found that listeners tended to perceive a speaker who spoke with a ‘nonnative’ accent as less credible. ‘Nonnative’ academics may likewise be perceived as less credible by reviewers because of linguistic features of their writing, raising concerns for linguistic justice in academic publishing.

Of course, there’s not nearly as much research on what information readers glean about the characteristics of writers based on their written language. One study presented two versions of the same essay to readers: an original written in English and a version that had been translated into Chinese and then back into English (by Chinese graduate students who used English as a second language). Although the essays were more or less conceptually identical, the authors reported that faculty members, who were asked to rate the essays on criteria like cohesion, clarity, strength of the argument, and organization, showed a bias against the translated essays due to the ‘nonnative’ features that translation introduced into the writing.

In short, I believe it’s likely that reviewers are influenced to some extent by indexical properties of nonnative writing. Reviewers and editors may not comment directly on the language as the source of their perceptions, but the indexical information gleaned from the language may influence their perceptions of more global characteristics of the work, like the strength of the argument or the credibility of the researchers.

Addressing the effects of denying linguistic injustice

Finally, I want to address Hyland’s statements about the apparently divisive effects of arguments pointing to linguistic injustice, such as the one I have made. Specifically, Hyland makes two points at the end of his argument, which I have reproduced here (p. 66):

1. By focusing on language shortcomings it perpetuates a myth of [‘nonnative’] deficit, which discourages [‘nonnative’] authors and tells them to look for prejudice rather than revision.
2. It marginalizes the challenges faced by [‘native’-English speaking authors] by depicting [countries where English is widely spoken] as a kind of safe house where academic publication can be taken for granted.

Hyland’s first accusation is that discussions of injustices within academic publishing that relate to the privileging of ‘native’ English are themselves a source of the ‘deficit’ discourse around ‘nonnative’ English users’ language. I would counter this by pointing out that denial of injustice is a major way in which injustice is perpetuated through two related processes. First, denial protects privilege by stonewalling attempts to address injustice. Second, by asserting that existing inequalities should not be attributed to the presence of disadvantages for some people, denials of injustice subtly imply that systematic inequality (like that seen in the disparate levels of success that ‘native’ and ‘nonnative’ academics enjoy in publishing their work) is the result of personal failings, in this case of language, research, thought, and writing.

Given that, as Hyland himself notes, the percentage of publications produced by ‘nonnative’ English speakers has increased over the past decade or so, I think it is unlikely that the acknowledgment of injustice and bias is seriously hampering their efforts at continuously improving their work.

The second of Hyland’s accusations is that such arguments depict the route to publication for ‘native’ English speakers as somehow effortless, ignoring the very real challenges that academic publication faces for them as well. It is, of course, not easy to consider that your professional accomplishments have not been entirely the result of your own effort and talents. I would prefer to think of my own successes as hard-fought victories won on a level playing field. However, we ‘native’ English speakers will need to see the distinction between claims that characterize our paths to publication as”easier” from those characterizing them as “easy”.

Furthermore, we are, I believe, perfectly capable of approaching issues of academic publication from multiple perspectives: those that address specifically the disadvantages of ‘nonnative’ Englishes in academic publishing and those that address the disadvantages that all inexperienced writers face. There is no need, as Hyland suggests, to collapse the two conversations.