Multilingual science: discussing language as a place of encounter in knowledge production and exchange

1 Introduction

There is a recognized controversy around language issues in academic publication that nurtures an accruing “malaise” (Nahabedian and von Stecher 2024, 86). Some studies and debates denounce a “linguistic disadvantage” (Politzer-Ahles and Ghali 2020) and list out lived experiences of the increased challenges imposed by the centralisation of English as the academic lingua franca (Ramírez-Castañeda 2020; Suzina 2020). Meanwhile, others consider the idea of “linguistic injustice” as a myth (Hyland 2016) and challenge the lack of rigorous methods to prove such an issue.

It is true that aspects such as investment in science, structures for research, and opportunities of networking can be more relevant to boost the possibilities of landing a piece of research in a top ranked journal – most probably published in English. Those aspects are essential for the very existence of a dynamic scientific ecosystem. This fact does not erase, however, the level of additional pressure and different kinds of burden imposed on scholars using English for publishing when this is not their first language. These can emerge as a symptom of all those structural fragilities, add up with the lack of academic literacy both by authors and peer-reviewers – independent of whether they work in a well-financed research context or not –, and become a complete frustration if the efforts of internationalization are not embedded in a broader project of knowledge exchange.

This article places the language issues within the debate about the exchange of scientific knowledge between communities (Arenas-Castro et al. 2024) to avoid the sterile engagement with arguments such as those highlighted by Ken Hyland (2016), who defends “academic literacy” as a high value that would be challenging to all, independent of language. This is a debate as vicious as those that suggest the existence of reverse racism or the need of a movement to protect male’s rights (in response to feminism). No one is denying that academic writing is a hard skill to develop and that it requires efforts from native English speakers as well. However, nor the numbers suggest equity in opportunities and obstacles (Amano et al. 2023; Arenas-Castro et al. 2024), nor academic literacy should be reduced to writing abilities. There is interest, therefore, to move the discussion from discrimination to domination issues, as suggested by Juan Javier Nahabedian and Pablo von Stecher (2024]), to distance the debate from individual opportunities or biases and tackle the systemic problem that justifies and even motivates discriminatory practices.

It is interesting to note at this point that presenting the cause of the conflict in terms of “historical domination,” in line with the initial chronology referring to the political, military, and economic processes that led to the current situation, would allow us to avoid naturalizing the phenomenon. (Nahabedian and von Stecher 2024: 80, translation by ChatGPT)

The “rule of the game” and the “battle” analogies identified by Nahabedian and von Stecher (2024) suggest that there is nothing wrong with the academic system and that scholars need to embrace the hardship that is part of the process of becoming a recognized scientist. Individuals are not, nonetheless, lacking motivation or ignoring opportunities to reach milestones that are fairly established. The debate about language justice is one part of a critical and creative movement working for the protection or recovery of the very meaning of knowledge exchange, that is presented here as the support for reaching dignified living conditions.

Moving from the discrimination debate is also important to avoid commodified corporate policies of equality, diversity and inclusion (EDI). Once used to pacify a debate – instead of addressing root causes of exclusion – these policies contain the high risk of allowing a couple of benefits to some scholars while the whole system keeps untouched. Pat Strauss analysed the motivations of a small group of peer reviewers and identified profiles that were named as “ringmaster” and “guardian”, among others (Strauss 2019: 3). Even though conscious of bias and inequalities, as well as of their role as gatekeepers, their posture revealed a patronizing posture and some defensive behaviours that sought rather to protect the imbalance of the system than to restore it.

The focus on domination seems more relevant as it moves the issue out of individual opportunities to focus on the role of academic publishing and how languages can help to reach meaningful impact. This approach contemplates the improvement of science and living conditions altogether and this article proposes three analytical perspectives (Figure 1):

1. the (de)colonial perspective: where language is seen as an expression of a worldview

2. the perspective on disruptive capacity: that explores the tension between teach to fit versus teach to transgress (hooks 1994), discussing the capacity of disruption through knowledge

3. the participatory perspective: where the understanding of language as a place of encounter confronts the idea of language as a performance of power

Figure 1:

Conceptual diagram of a tridimensional approach to academic language (generated by Microsoft Create): decolonial issues are concentrated in the center, the disruptive capacity is represented by flows and arrows, the participation comes from all nodes that compose the system.

These perspectives will be unfolded in the following sections of this article. This systematization is based in literature review only, as well as the set of proposed solutions advanced under the notion of plurilingualism. Some examples of actions put in place are listed with these solutions and are not exhaustive. This analysis endorses the argument defended by Henry Arenas-Castro and colleagues (2024]: 6), for whom the “academic publishing must undergo a cultural change”. There is plenty of literature pointing to the consequences of a centralized and dominant lingua franca in science coming from different fields and regions. They suggest a systemic issue that requires engagement with debates on the geopolitics of knowledge (Mignolo 2007).

2 Language and knowledge from a decolonial perspective

Coloniality in science can be very briefly described as the centralization of mechanisms of validation of what can be considered as knowledge, a role historically played and frequently imposed by Western rich countries. Walter Mignolo expresses it by the illustrative statement that “the Western has disciplines while the Eastern has cultures that are studied through Western disciplines” (Mignolo 2007: 60). Therefore, decolonizing is taken here as every effort and resistance towards restoring a perspective that “is no longer understood as a distant echo of what is thought in Europe” (Fornet-Betancourt 2007: 30) but recognized as an authentic and legitimate thought grounded in a certain geopolitical epistemology.

In this context, the decolonial perspective problematizes the simple justification of English as the academic lingua franca for its instrumental character. There is a common understanding about the benefits of having a common language for exchanging scientific questions, findings and advances. Analysing arguments associated with this debate, Nahabedian and von Stecher (2024) reflect that the so-called democratizing features of the English have been used to normalize power asymmetries that are intrinsic to the choice of any language as the platform for science. This frame is relevant because it points to defining rules of the system that must be applied to English today and to any other language fulfilling this role regionally or globally at any time. In short, there is no neutral vehicular instrument and the more centralized a common language becomes the less democratic it will be.

Problems raised by the decolonial perspective include – and are not limited to – the creation and crystallization of hierarchies between publications and between authors, the naturalization of unjust practices for the protection of English as a safeguard of the very scientific system, and the impact over the definition of the scientific agenda as well as policies and actions potentially deriving from it. These aspects are frequently overlapping.

Research has demonstrated that something between 78 % and 98 % of academic articles are currently published in English (Amano et al. 2023; Céspedes 2021; Steigerwald et al. 2022). In social sciences, by 2005, English-written pieces represented 75 % of outputs with most of other languages below 2 % (Law and Mol 2020: 267). This concentration (Table 1) has a consequence over the health of any concurrent system of publication. The Latin American network of journals, for example, is recognized as an independent circuit because of the large number of countries and academics thriving while writing mainly in Portuguese and Spanish. These journals are currently facing additional challenges to adhere to international indexing systems because scientific funding agencies from Latin America itself have been prioritizing publication in English-written indexed journals as part of their internationalization plans. The established quality of the Latin American journals and their contributions for the development of science end up scrutinized for their conformation to the dominant system. Scholars opting to publish in the regional circuit might as well be seen as less competent and face difficulties to obtain grants and career promotions.

Table  1:

Top languages and countries of publication (Source: Céspedes 2021).

Nahabedian and von Stecher (2024]: 86) observed the naturalization of a practice amidst Spanish scholars; they would select the most important findings of their research for international journals published in English, suggesting that Ibero-American journals would be served with literally the rest. They captured a comment from Golombek affirming that “[n]o scientific researcher in the natural sciences is going to have as a first choice for his good findings one of those journals in his local language, he is going to send it in English” (Nahabedian and von Stecher 2024: 86). Valeria Ramírez-Castañeda (2020) highlighted that 90 % of articles published by Colombian scholars are in English, despite the country being among those with the lowest English proficiency. Such a dynamic can only deepen the hierarchisation between academic outlets, but it has another harming dimension. As it will be discussed further in the participatory perspective, this concentration has serious consequences for the access to high quality research in impoverished countries and regions.

Figure 2 illustrates what John Law and Annemarie Mol describe as “institutional arrangements within which English functions as an academic superpower” (2020]: 265). For the authors, this happens when supposedly international academic conversations become “limited and parochial” (idem: 264) because they lose the richness of inter-linguistic exchanges. The artificially created diagram looks illustrative for how it suggests that the supposedly neutral affordance of a shared language ends up by entangling nodes that might have had epistemological independence beforehand.

Figure 2:

Linguistic domination (generated by Microsoft Create).

Karin Bennett (2013) talks about the use of English as lingua franca in academia, among other aspects, as a form of epistemological domestication. What Hyland (2016) describes as the necessary requirement of “academic literacy”, that must be effortfully embraced by all, ends up falling into a short understanding of “international academic English standards”, as pointed by Stephen Politzer-Ahles, Teresa Girolamo and Samantha Ghali (2020]). Their exploratory study with peer reviewers pointed that these standards are “roughly equivalent to what academic reviewers with no special training or expertise in the use of English as a lingua franca or additional language consider to be ‘good’ and ‘native-like’” (Politzer-Ahles and Ghali 2020: 3). The description fits the findings summarized by Pat Strauss as a “hidden agenda” or an “emotional … defence” to protect some kind of purity supposedly embedded in the English language (Strauss 2019: 7–8) and constitutes one of the several procedures of classification employed by the coloniality to maintain its “locus of enunciation” (Mignolo 2007). Strauss taps into previous studies that indicate that “language use receives more attention than any other features of the text” (Strauss 2019: 3). It also fits one of Hyland’s arguments (2016]: 66), claiming that “it is not the English that is usually the problem at all but the fact that the authors violate the reviewer’s expectations of academic writing”.

Hyland (2016) talks about the issue of the “control of the register” as there would be only one appropriate to fulfil the notion of academic literacy. The latter ends up reduced to writing and format standards that completely lose connection with critical and creative aspects that are so necessary for a vigorous practice of science. Emma Steigerwald and colleagues (2022) suggest that writing exclusively in English limits the ability of describing the relationships between ideas. The Latin-American register, to say so, has been historically depreciated as an essay-like form of writing, as this would mean less of accuracy or academic rigour. There is a challenge for preserving the clarity of ideas, but it should never be mistaken with the loss of complexity.

Caleb Everett studies how the language we speak affects the way that we think and how the contexts affect the way we develop language. Together with Keren Madora, they conducted research about language practices in an indigenous tribe called Pirahã, in the Brazilian Amazon Forest (Everett and Madora 2012). The Pirahã language does not have terms for quantities; they actually do not have numbers in their language at all. The findings confirm Everett’s previous studies pointing that this lack of numeric language influences the whole logic perception of these people. Similar effects were observed in the way language shapes the perception of colour, the memory of events, and the awareness of others (Steigerwald et al. 2022). They support the idea that language is a key component of processes of the mind. In short, the way we speak is deeply associated with the way we think and the way we understand the world. Decolonial translation can therefore be understood as the restitution of the capacity of giving sense to the world through language, as it is constituted by and carries along a worldview.

Finally, the centralization of language influences what are the ideas that turn into research topics and its domestication has effects over the structure of expression of these ideas. Nahabedian and von Stecher (2024) identified questions that associate the emergence of English-speaking publishing monopolies with their potential to set research themes globally. Law and Mol (2020]: 270) observe that the que quality of academic debates in Scandinavian countries and the Netherlands has been “seriously eroded” by the prevalence of writing in English even in local outlets. Cathrin Bengesser (2022) is very clear about the way television studies were influenced by the domination of English. Referring to a review made by Schmerhein and Vermeulen in 2016 about scholarship in Germany, the author summarizes that

they also deplore that US television appears to be a more valued research object in German scholarship than the region’s domestic TV. Because English-speaking scholarship has put words and academic value to – most often English-speaking – television (see: Newman and Levine 2012), it is also what becomes studied elsewhere, based on the concepts exported together with the internationally successful televisual object. (Bengesser 2022: 77)

When all these aspects overlap, the definition of an academic lingua franca demonstrates how “constraining global scientific discussions to a single language can limit who builds, has access to, and communicates scientific knowledge to the broader public” (Steigerwald et al. 2022: 989). The whole process ends up verticalizing the circulation of knowledge in a way that the “practice of a literary elite” (Hyland 2016: 67) is the only one capable of validating the knowledge that might be considered relevant and then, transferred to the rest of the world, including those regions where data was extracted in the first place. Overall, the emerging question is to whom the democratization of science through English as lingua franca is serving.

3 Language as a capacity to produce disruptive knowledge

As previously indicated, this article moves the debate about academic language from discrimination to domination debates, looking for the very reasons why scholars write and publish papers. The idea of disruption is considered here as a relevant feature for advancing scientific knowledge and raised for how it is threatened by a less diverse academic ecosystem where most participants employ the same “register” (Hyland 2016) and a “literary elite” (idem) defines the research agenda.

The notion of disruption and its relationship with knowledge is borrowed from Michael Park, Erin Leahey and Russell J. Funk (2023]). The authors describe two kinds of breakthrough that lead to innovation in science and technology; one consolidates existing knowledge adding confirmation elements to it, and the other renders pieces of knowledge obsolete pushing the frontiers of science (idem: 137–138). The latter is what the authors call disruption. Park and colleagues published a concerning article in Nature analysing how papers and patents are becoming less disruptive over time, which prompted a lively debate (Nature 2023). Considering that knowledge needs to be constantly challenged to advance, the capacity of producing disruption is core. This brings us back to the issues of complexity and performance.

Artificial Intelligence resources constitute a promising solution for the issue of academic literacy understood as “English eloquence” (Suzina 2020). It is already possible to ask an AI to translate an article using academic English prose, although the results are still limited for the lack of machine training in different fields and from different schools of thought. As pointed by a recent study, in general, “[o]verlooking non-English studies can result in large gaps within global databases, which affects policy, management, and decision-making” (Steigerwald et al. 2022, 989). Combined with the obligation of performance imposed by funding agencies and career progression definitions, the technological solution may, therefore, optimise overexploitation or homogenisation instead of promoting critical thinking and creativity.

The missed disruption pointed by Park and colleagues (2023) is not related with a quantitative issue. According to the U.S. National Science Foundation (NSF), more than 2,5 million science and engineering (S&E) articles were published around the world in 2018 (McCarthy 2019). The access to the right technology may reduce some structural inequalities and lead more scholars to adequately apply the acceptable register expected by reviewers and editors in international journals. The disruptive gap is, however, rather associated with the lack of transgression. Among the reasons for the regression, the authors observed that scientists rely on a narrower set of existing knowledge.

There, language domination joins citation issues (Isasi Velasco and Rio Riande 2022; Macgilchrist et al. 2022) as referring to literature coming from diverse linguistic backgrounds may constitute an issue for a good number of international journals. Arenas-Castro and colleagues (2024) examined the 2020 Journal of Citation Reports and selected journals in biological sciences, for their potential coverage of issues related to the global biodiversity crisis and, therefore, requiring global input. Their results massively confirm general linguistic barriers as the limitation to writing in English and less than 1 % – out of 736 journals analysed – indicated in the guidelines that a manuscript would not be rejected exclusively for a perceived lack of English standard. Additionally, “[a]lthough most editors-in-chief indicated that they allow or encourage citing non-English-language literature, only 10 % of the journals explicitly mentioned this in the author guidelines” (Arenas-Castro et al. 2024, 4).

At this point, the relevant question is if the definition of a lingua franca is serving to fit the system or to transgress it. The term “transgress” here makes direct reference to bell hooks’ book Teaching to transgress (1994), to highlight transgression as a contribution to emancipatory thinking and bold contributions to the development of science. Following this logic, the notion of academic literacy must be broadened and include the ability to demonstrate how a piece of research is seeking to push knowledge frontiers. At the same time, considering that “reviewer decision-making helps determine what science will be presented at a scholarly meeting and will ultimately help progress the field” (Politzer-Ahles and Ghali 2020: 3), peer reviewers need to overcome the postures of “ringmaster” and “guardian” (Strauss 2019) of a supposedly untouchable academic standard.

From the literature reviewed for this article, the supposed effort to reach academic standards is likely imposing limitations to the potential contribution of authors, as “[c]onstraining points of view to fit within the structure and vocabulary of a single language impoverishes scholarly discourse and observations of nature” (Steigerwald et al. 2022: 989). It may also be reducing the potential reach of the research results, as it restricts “the dissemination of science to non-English-speaking institutions and communities, which can leave new knowledge inaccessible to the people for whom it is most relevant” (idem) and “complicates applying scientific evidence to decision-making outside academia” (idem). The fitting orientation finally “hampers efforts to achieve diversity and equity in science, technology, engineering and mathematics (STEM)” (Arenas-Castro et al. 2024: 2), as it is probably the case for any other research field. In short, the concentration of academic publication in English-language journals is not contributing to cognitive justice and not even to the expected performance of the knowledge system.

4 Language and knowledge from a participatory perspective

The decolonial and disruptive arguments just exposed open a participatory angle to the debate about language and knowledge. Considering the definition of a lingua franca as a form of democratizing the academic practice approximates it to the notion of “parity of participation” (Fraser 2005), as it will raise questions about who is allowed to take part as peers in the global academic debate and under which conditions. This will be framed here as a tension between understanding language as a performance of power or as a place of encounter.

The language as a performance of power feeds colonial processes and the commodification of science (Albuquerque and Oliveira 2021). Participants of the debate belong to an elite recognized by the mastery of the dominant language or by the access to enough resources to perform as equals. It follows the same flaws observed in the constitution of the public sphere, where the acquisition of a certain ability of speaking became an entry barrier, marginalizing groups who do not share the same code (François and Neveu 1999; Fraser 1992). As summarized by Emma Steigerwald and colleagues, “English proficiency also influences who participates in science at a global scale, which is detrimental to science, because a diversity of perspectives bolsters the construction of robust and innovative scientific knowledge” (2022]: 989). To provide a scale, according to the UNESCO, more than 7,5 thousand languages are spoken in the world but less than 1 % of them are used in academic research (Muller de Oliveira 2024).

As observed in several studies (Amano et al. 2023; Arenas-Castro et al. 2024; Ramírez-Castañeda 2020), conforming to the system is highly resource-consuming and still does not necessarily pay the price, as the balance of participation keeps uneven (Table 1). Researchers whose first language is not English undertake a true obstacle race to thrive in academic global spheres: they spend years preparing and correcting manuscripts – 91 % more time reading and 51 % more time writing –, considerable money in language services – 75 % of them ask someone to edit their work before submission and there are 12.5 times more chances that reviewers and/or editors require a language revision from them –, and may still end up with a rejection – the frequency is 2.6 times higher – or, if successful, behind a paywall that prevents a large reach (Amano et al. 2023). It may contribute to individual careers, but it does not seem to be supporting a fair and productive dialogue.

The language as a performance of power adopts the notion of dominant language and establishes the parameter of the native speaker as a normative aspect – in scientific terms, the international academic English standards (Politzer-Ahles and Ghali 2020). The dominant language is taken as neutral despite the epistemological conditionalities are part of it (Bennett 2013). The institutional and technical solutions confirm its dominance, allowing adaptations to conform to the system without necessarily questioning its shortcomings.

A previous article (Suzina 2020) challenged the idea of a distinctive “research that travels” as the naturalization of a top-down evaluation of certain topics that belong to the global stage and should then be translated into English, while others should keep confined to local or regional circulation in their native languages. As already stated, this leads to practices that harm the publication ecosystem, academic careers, and the broad value of science for societies. Golombek breaks this idea affirming that “[i]f one studies the Pilcomayo River basin, or the Chagas disease, or studies the phenomena of endemic poverty in our country, of course the examples are local, but if you study it well, it is universal” (apud Nahabedian and von Stecher 2024: 79). The academic literacy is, therefore, about connecting local issues to global debates, something that is only possible if there is free flow of knowledge between all scientific communities. If the “research that travels” poses a starting issue of hierarchising the topics from a dominant perspective, publishing them only in English also prevents non-English-speakers to even try and engage with the most updated debate.

Moving forward in the disparities created by this dynamic, Jennifer Isasi Velasco (2024) points to the extractive practice of capturing information in non-English speaking communities and publishing them in English, reducing drastically the possibility of restitution that is already not ideal. Ramírez-Castaneda (2020]) questions how publishing in English limits the access to national peers to research done in their own countries, frequently funded by public grants. The damage goes beyond the research community and jeopardize teaching and the very possibility of a new generation of scientists. Giorguli Saucedo asks “[w]ith what texts are we educating our students if they do not know English?” (apud Nahabedian and von Stecher 2024: 84). The language as performance of power creates a whole cycle that needs to be interrupted systemically to bring about a significative change.

In opposition to it, the participatory perspective understands language as a place of encounter and welcomes the establishment of linguistic platforms as a common ground for exchange, accessible for a variety of manifestations. In this context, language is not anymore seen as a tool and, consequently, cannot be addressed by technical gadgets alone. Language becomes a territory and, as such, it is inhabited by its members in different ways allowing the diversity of thoughts to compose, question, accommodate and disrupt ideas.

The language as a place of encounter is based on the notion of common language, looking for mutual comprehension. It implies an effort from all stakeholders to establish the exchange for the purposes of learning more and sharing more. The cultural and epistemological traits of original languages permeate the common language and are taken as resources because they allow to grasp the dynamics behind the articulation of ideas. The common language respects the plurilingualism and takes it as a resource. In the next section, the notion of plurilingualism and the idea of language as a place of encounter will be further explored, looking for solutions to overcome the barriers of using English as the academic lingua franca.

5 Looking for solutions

Gilvan Müller de Oliveira is the head of the UNESCO Chair on Language Policies for Multilingualism (UCLPM) that has its headquarters at Universidade Federal de Santa Catarina, a public university in the South of Brazil. He offers a challenging but optimist perspective regarding the construction of alternatives for the broader domination of English in the world.

The so-called dominance of English has a real and an ideological component. The ideological component makes us believe that the dominance and expansion of English is unstoppable, necessary and that this is an unavoidable situation, but this is not true. (Müller de Oliveira 2024, translation by Google)

Oliveira describes how the domination of English over internet smoothly narrowed down, accounting that it used to represent 90 % of the content and decreased to 21 % currently (Muller de Oliveira 2024). According to him, the volume of English content continues to increase, and the proportional difference comes from the reduction of obstacles for people posting contents in other languages, once national digital breaches are overcome. From one side, this comparison highlights the importance of supporting strong national research policies that allow the production of relevant scientific content; from the other, it points out to the need of creating an international publication system accessible enough to enable inclusion of the variety of contributions whose flow may be currently blocked by obstructed paths.

Based in the examined literature, three potential solutions are advanced here: the encompassing idea of a multilingual academic publication system in articulation with a collective network, the central role of translation, and the development of a refreshed idea of academic literacy for both authors and peer-reviewers.

6 A multilingual and collective network

Steigerwald and colleagues clearly summarize the task: “transform a monolingual scientific hub into a multilingual scientific network” (Steigerwald et al. 2022: 988). The interest of this proposition is that it encapsulates two required transformations. One refers to the diversity of languages and the other, to a shift from a hub to a network.

Oliveira employs the conceptual framework used by the European Union to explain that “the multilingualism is a social phenomenon that points to the diversity of languages in a defined society” while, according to other references, the plurilingualism is a “state of valorisation of multilingualism” built to take advantage of the variety of languages spoken for the benefit of speakers and their communities (Muller de Oliveira 2024). The definition of multilingualism proposed by Steigerwald and colleagues states that it is “the outcome of a prolonged process of inclusion of languages brought about by improved translation technologies and changes in community norms” (Steigerwald et al. 2022: 995). Looking back at Figure 1 and understanding language as a place of encounter, a multilingual and collective network would increase the flow between the different nodes, allowing them to grow and reduce the attraction to the center.

Steigerwald and colleagues (2022]) briefly describe a dynamic of secondary and tertiary language hubs that would support international networking between scholars from different language backgrounds, while equalizing the importance of these different convergent points. The idea of network, however, invites to explore other possibilities closer to collective dynamics. In a survey of policies of 736 journals in biological sciences, Arenas-Castro and colleagues identified that “ownership by a scientific society was the clearest positive predictor of linguistic inclusivity in scientific publishing” (Arenas-Castro et al. 2024: 6). This is more relevant than having a diverse editorial board or operating open access policies. They argue, therefore, that “[a]s organizations with the capacity to define disciplinary norms and shape culture within academic communities, scientific societies are uniquely positioned to reform academic publishing towards linguistic inclusivity” (idem) and advocate to support community-led initiatives.

The Bulletin of Latin American Research (BLAR) can be taken as an illustrative example.^[1] The journal belongs to the Society for Latin American Studies (SLAS), based in the United Kingdom, and is published by Wiley. Although articles are published only in English, authors can submit manuscripts also in Portuguese and Spanish. All the peer-review process is done in the original languages, which means that the works are evaluated in their integral forms of expression and the authors will only engage in financial expense for translation once the manuscript is accepted. Most of SLAS members and BLAR peer reviewers are native-speakers or have experience with languages spoken in Latin America, making multilingualism even more meaningful. Topic-oriented societies have siblings in different countries and could benefit from collaborative peer-review in order to enable submissions in a variety of languages. This could be an initial step for future collaborations exchanging translations of published articles.

Several journals belonging to what can be called the alternative system and some from Area Studies adopted the practice of publishing English versions of articles written in original languages; the Latin American example will be mentioned below. Understanding that decolonizing opposes dependence, subordination, domination, and control, mainstream English-written journals could adopt a decolonial reciprocity and publish their articles in original languages too. Steigerwald and colleagues’ perspective (2022]) aligns with Oliveira’s argument that the dominance of English is not unstoppable. They consider that multilingual science is also a healthy way to prepare and navigate likely transitions more easily.

The key point here is that common languages will always be important but need to be framed under collective standards that take into account local, national and global contributions and interests. The findings highlighting the positive role played by scientific societies suggest that the idea a multilingual publication system associated with a networked structure of collaboration is likely the best solution for the benefit of the science.

7 The role of translation

The role of translation is also central to this transformation. It refers to increased and facilitated resources for translating research pieces, and to the creation of editorial spaces for translated material and for discussing the practice of academic translation itself. The practice of translation is essential to preserve a meaningful relationship with local networks of knowledge, that include peers at national level and communities with whom data collection was done. Steigerwald and colleagues (2022) consider it necessary

when the work is conducted in a region or country in which the primary language is not English, when the team involves researchers whose primary language is not English, or when the research directly or indirectly affects a group of people whose primary language is not English. (Steigerwald et al. 2022: 993)

For the authors, funding agencies can “encourage budget items to support translating results and outreach that engages with local communities in their local languages” (Steigerwald et al. 2022: 995). Bengesser (2022) recalls the importance of translation also to keep a dialogue between academia and industry.

In a detailed reflection about translation in academic publication, Steigerwald and colleagues (2022) refer to an abundance of measures that could be put in place to foster a multilingual network. They include neural machine translation, support for reciprocal translation through journals or collective academic platforms, copyright weavers from publishers for translating articles, repositories for translated versions of published articles, soft word limits, multilingual scholars on journals boards and committees, mentorship programmes, translations of abstracts and keywords at least and full article translations at best in all journals, valorisation of publication in languages other than English in development career plans, among others.

The discussion about practices of translation is also important to avoid a colonial and commodifying use of the resource. Some remedies can treat the symptom without curing the disease. Increasing the volume of translations can work as an appreciated bandage and can be quickly achieved with the support of artificial intelligence. However, a translation that ritually follow the formats of the dominant system may create a one-way solution, increasing the number of materials published in English without creating the local opportunities above-mentioned.

In March 2020, The Sociological Review published a whole issue to discuss “what it takes and what it adds to think in other terms” (Law and Mol 2020). The editorial invites to a reflection about a series of relevant elements, including the relevance of fostering translations that carry along the original manuscripts’ contexts. Following, a group of authors from different origins discuss how they navigate the challenges of translation to occupy a place in the global dialogue.

The journal Critical Studies in Television published a new section on its first issue of 2022 called In Translation, including two articles and an introductory reflection. The abstract stated that it “addresses how contributions from languages other than English enable adequate understanding of translational TV phenomena through methods and theory developed on a diversity of objects and contexts” (Bengesser 2022: 76). From the journal guidelines and quick research on issues published afterwards, it is not clear if the section progressed, but that was a good example of how journals can identify relevant pieces of research, translate them, and take the opportunity for valuing the effort of intercultural dialogue.

Multilingual publishing is another mechanism highlighted by Steigerwald and colleagues (2022) for actively promoting and valuing contributions in different languages. Ana Suzina and Thomas Tufte invited a series of speakers from different geographic origins to speak on their own languages during a series of talks in celebration of the birth centennial of Paulo Freire in 2021, adding subtitles to pre-recorded videos that were presented to the audiences during seven online meetings.^[2] The content turned into teaching material published both in English (Suzina and Tufte 2022) and Portuguese (Suzina and Tufte 2023). Stefania Milan, Emiliano Trere and Silvia Masiero (2021]: 19) embraced the challenge of publishing a multilingual book to contribute towards “multicultural encounters” and to “make readers experience first-hand what individuals and communities at the margins undergo on a regular basis: discomfort … often forced to read texts in a language that reflects a worldview that is not their own”.

Initiatives like the above feed a necessary discussion about paths to inform and increase efforts towards a multilingual science. These processes also need to be translated and communicated so more governments, institutions and scholars can reproduce and improve them.

8 A broaden understanding of academic literacy

Between the fluid but always high expectations of peer-reviewers regarding international academic English standards (Politzer-Ahles and Ghali 2020) and the lack of clarity or direct barriers regarding language in journal policies (Arenas-Castro et al. 2024), it seems that the notion of academic literacy must be reviewed. From one side, associating academic literacy with the capacity of speaking as Shakespeare^[3] means a hard obstacle to non-native English speakers, as most of literature reviewed indicates. From another side, reaching English native standards can be possible with the support of technologies without necessarily improving the dialogue that is currently obstructed by a monolingual system (see e.g. Steigerwald et al. 2022).

We need to ensure that the language in the journals that we review and edit adequately conveys meaning. We are entitled to ask for clarity and succinctness – we are not entitled to ask that everyone write American and British English. (Strauss 2019: 8)

The idea of a network, exposed before, can only be meaningful if the different linguistic hubs imagined by Steigerwald and colleagues (2022) talk to each other and move from a productive to a collective mindset, where the development of science is more important than metrics. The academic literacy is, therefore, much more related with the idea of disruption (Park et al. 2023) and transgression (hooks 1994) previously unfolded. Nahabedian and von Stecher (2024) suggest that a way to free authors in their decision about where to publish is to recall that there are “other elements that intervene in the weighting of scientists when choosing the medium in which to publish (local scope, language accessibility, tradition, ideology), elements that the ideologeme of English as a lingua franca nullifies or attends to only secondarily” (Nahabedian and von Stecher 2024: 87, translion by Google). This strategic decision is the academic literacy. It is not imposed from the top but must be part of the research design, responding to demands regarding the potential application of the findings, their potential contribution to local and global debates, and the continuation of lines of inquiry or dialogues with precise fieldworks through the training of new generations of scientists in rich and impoverished countries.

Steigerwald and colleagues recall that establishing common languages “is also useful for the dissemination and recognition of research performed by scientists whose primary language is not widely spoken, as well as for facilitating communication between such scientists and the wider scientific community” (2022]: 988). Oliveira points to a crisis in countries as Germany, France and the Scandinavian regarding the decrease in academic publication in national languages, something that led the government of The Netherlands to offer support to their scientists to publish in Dutch (Muller de Oliveira 2024). Bengesser recalls that “it is also important to keep and develop Danish as a language for research dissemination for peers and students” (Bengesser 2022: 78). In a multilingual network, “translations would not only happen from other languages to English but also from English to other languages” (Steigerwald et al. 2022: 992).

The publication practices in Latin America are highlighted by the head of UCLPM as responsible for a permanent increase in the scientific production in Portuguese and Spanish. Several journals in the region publish in original languages with an added translation to English, the great majority of them in Open Access, publicly funded and indexed in regional platforms such SciELO and Latindex. Although ticking many boxes of a multilingual network, these journals do not reach the so-called international academic standards and are, therefore, excluded or poorly evaluated in global indexes. Under a refreshed understanding of academic literacy, many of their practices could serve as exploratory experiences for a multilingual networked publishing system.

If these ideas suggest that authors and scientific institutions must review their practices to cope with a more meaningful academic literacy, it is fundamental to highlight the role of peer-reviewers. Among the profiles identified by Strauss (see above), the “ally” is a peer-reviewer that is “eager to facilitate non-anglophone authors” participation “within the discourse community of published scientists” (2019]: 3). These efforts must include anti-discriminatory postures under the perspective that these are a guarantee for a more relevant, meaningful and ethical science.

9 Concluding remarks

This article reviewed literature about the use of English as lingua franca in academia under three perspectives, one (de)colonial, one observing the disruptive capacity of science; and a participatory one. This framework allows to move the analysis from discrimination to domination issues. This is relevant to place language in academic practice within the debate of geopolitics of knowledge that validate what is relevant, legitimate, and worth considering in social and political design. Language participates in the meaning-making metalevel that direct or indirectly influence interpretations that inform the design-making. These reflections lead to three potential solutions that include the development of a multilingual academic publication system, the central role of translation, and a refreshed idea of academic literacy for authors and peer-reviewers. These solutions are not exhaustive, and their adaptation or formulation of others must take into account the health of the global and the national systems of knowledge. A shared platform is fundamental for scientific development, but this will only be healthy and just if sustained by dynamic local hubs where knowledge circulates and renews itself continuously.