A Defense of Moderate Language Pluralism in Scholarly Production in the Humanities

In this paper we argue in favor of a moderate language plurilingualism in scholarly production in the humanities. We start by proposing some desiderata that a good policy regarding language use in the humanities should meet. We then survey institutional/policy options regarding the use of languages in the humanities and argue that different versions of monolingualism fall short of satisfying the desiderata. As a result, we consider different alternatives supporting plurilingualism and we defend that one of the options, that we call Moderate Language Pluralism or Moderate Plurilingualism, does better than the others vis-à-vis the desiderata. Finally, we suggest some policies that would help to implement Moderate Plurilingualism.


I. Five desiderata to assess monolingualism vs plurilingualism in the humanities
We understand scholarly production in the humanities in a broad sense covering intellectual, scientific and academic production in areas traditionally understood as humanities. While we will not offer a definition of "humanities", we will assume that the disciplines that are normally included under the label engage in bona fide production of knowledge about human experience of a type that no other areas of human enquiry can provide. 2 When thinking about the relationship between language of scholarly/intellectual/scientific/academic production and how that influences or affects that production, we need to distinguish between the influence that the choice of language(s) has on the person or groups who are or could be active scholars, and the effect it has on the discipline itself including the audience the work actually reaches. Of course both of these dimensions are connected in several ways, but a detailed analysis of the discussion about monolingualism vs. plurilingualism benefits from that distinction. We propose that a view about this discussion be judged according to how well that view does vis-à-vis five desiderata. 3 The first two are within the domain of the individual researcher, and the last three within the domain of the discipline.

1) Fairness in access:
Scholars in the humanities should be able to have full or at least substantive access to their discipline both for communicating scholarly results and for accessing discussions (not only, but primarily contemporary) about the topics of their scholarly endeavors. This means having a fairly elaborated proficiency in the language or languages in which the discipline is practiced. Authors can have their work translated to the canonic language (the lingua franca) or have access to work translated from the canonic language to their own language, but that still limits access and communication because a vast majority of scholarly production takes many years to be translated to languages other than the original, or just never is. We think that scholarly policies need to be sensitive to access in this sense, so that the widest possible community of scholars is achieved with no limitation extrinsic to the value of the scholarly production.
2) Fairness in expressivity: Scholars in the humanities will be able to have full or at least substantive expressive power in their production only if they have a fairly elaborated proficiency in the language or languages in which the discipline is practiced. As we will argue later, the use of language in the humanities is essential to the scholarship itself in a way that is asymmetrical with other disciplines, because the subtleties of natural language are an essential feature of scholarly production in the humanities. Thus, scholars without a subtle acquaintance with the canonic language(s) of the discipline will be at a disadvantage 4 that goes well beyond the potential stylistic challenges that we would encounter in non-humanities areas, where most of the production involves mathematical and/or technical vocabulary of the discipline. Alternatives are possible but they are suboptimal and very difficult to implement: For instance, an author can work with a bilingual helper who knows the discipline to produce research in the canonic language. 5 Here again, scholarly policies have to be sensitive to expressive power in this sense, so that the widest possible community of scholars is achieved with no limitation extrinsic to the value of the scholarly production.
3) Advancement of a topic/scholarship line in a given discipline: From a discipline standpoint, we assume that the wider the network of scholars who can participate in detailed discussions about issues in different sub-disciplines, the stronger the sub-discipline (and a fortiori the discipline) will be. Thus, restricting the access of specific discussions to speakers of a single language will almost certainly restrict the development of lines of scholarship across a discipline. Note that this desideratum is independent from the point we make regarding whether different languages have peculiar ways of addressing issues or even on whether there are whole issues that may be dependent on certain languages; it is rather a matter of enhancing the probability of advancing knowledge. Thus, scholarly policies, all things being equal, have to favor the likelihood of a wider engagement in specific issues. In fact, this desideratum may be seen as a discipline specific consequence of a more general desideratum, namely, communicability. Whether wide communicability in scholarly enterprises is intrinsically desirable or not, it is certainly valuable insofar as it maximizes the chances of a discipline's advancement. 4) Contribution to novel perspectives or creative contributions to the discipline: Also from a discipline standpoint, we claim that it is at least conceivable that in some areas, the language in which the scholarly work takes place, or alternatively some non-universalizable feature of a language or group of languages, is in itself essential in the process and output of such production. Since the humanities aim at understanding our own human experience, of which languages are a significant part, it follows that narrowing the production in the humanities to a single language (or just a few of them) will result in the loss of richness for the discipline, and therefore in the loss of self-understanding. It seems pretty obvious to us that scholarly policies have to favor the enrichment of the humanities as opposed to standardizing the way humanists reflect on our nature or eliminating perspectives from the discipline. While scholarly exchange can certainly be a productive factor to incorporate novel perspectives and stimulate creative contributions, unifying production in a given language would impoverish the possibilities (textures) of intellectual inquiry in a way that would parallel not allowing tempera, fresco and watercolor in the world of painting. This is, we think, the most important of the desiderata. We argue for its relevance in more detail in section III. 5) Scope of the audience for scholarly production: Finally, it is important to note that production in the humanities is intended to general audiences more frequently than production in the natural sciences, and that is a consequence of the nature of the disciplines. Those general audiences are of course much wider than the relatively small cadre of scholars producing knowledge and, as a rule, are far less likely to include people trained in a given language outside from their national/regional languages. Ideally, the disciplines have to ensure that intellectual production can be accessed by such a larger audience to keep the impact that scholarship in the area is supposed to have.
In the next section we will describe different possible approaches (and, as a result, different possible courses of action when it comes to policies) regarding the idea of unifying the language used in scholarly production in the humanities by means of a lingua franca common to all scholars and disciplines.

II. On the very idea of a lingua franca
Monolingualism is defended today by assuming that English is the lingua franca for many human activities including but not limited to science, business, transportation and diplomacy. 6 But what exactly is involved in the idea of a lingua franca? As we know there are more than 7,000 natural languages spoken by humans on Earth today. 7 Obviously, there is no single language that every human being can speak. Could there be a lingua franca, i.e. a single language that all scholars could use, in order to have a common second language for everyone?
At the end of the 19th century Esperanto was generated as the first constructed international language, offering a flexible and easy lingua franca that could be acquired by everybody as a second language to supply a tool for communication between different cultures and societies. Esperanto is an International Auxiliary Language (IAL), created as a neutral language, i.e. a non-ethnic, culture-neutral language. It was also considered as a prospect for being the lingua franca for science. 8 But, as a matter of fact, the project failed. It is estimated that only 2 million people speak Esperanto today, and there are only 350 native speakers (basically insignificant numbers if we think that the world population is about 7.5 billon people). And most scientific journals do not publish papers written in Esperanto. 9 However, when we hear today that English is the lingua franca for humanities as well as science, it is certainly Latin the first historical example that comes to mind. In fact, it has been argued that English is the Latin of our times (Toribio 2014). Probably there are historical similarities (they are both the "languages of the Empire", as Toribio states), but there are also very significant differences. Latin was the language of the Church in the Middle Ages, as well as the language of "knowledge", and this fact depended upon a series of factors which include a homogeneous academic community shaped by a similar education, a set of shared methodological assumptions, a set of writing genres, and a philosophical tradition based on a restricted set of established models (Gracia 2014). And, probably more importantly, all the members of that community were educated in the same language: Latin.
Thus, having Latin as a lingua franca had, as a necessary condition, shared methodological assumptions, and a restricted set of writing genres and models. Today, a similar scenario would likely be possible only in the case of some scientific disciplines. Also, in medieval times, it had the necessary condition that education itself was provided in the lingua franca. 10 The fact that Latin played the role of a lingua franca had to do with specific historical reasons. However, the idea that today an arbitrarily chosen language (English or another one) is developed as a lingua franca for the highly educated elites that produce knowledge would go beyond that historical context.
Reasonably, it could be argued that it is not necessary to force a universal education in English (the de facto lingua franca of our days) for the "educated" in order to hold that it should be the language in which scientific and intellectual production should be made (i.e., original contributions written in the paper format or in a book format and published by the most prestigious publishing companies). Initial education can be done in whatever language, but then, by making an individual effort, non-English speakers should publish their ideas in English, not in their native languages.
So far, thus, we have reviewed three forms of monolingualism, which can be summarized in these ways: a) Neutral artificial monolingualism: Scholarly/scientific/intellectual production should happen in a universal language that is culturally and linguistically neutral and levels the field for all participants in the scholarly community. b) Adopted elite monolingualism: Scholarly/scientific/intellectual production should happen in a language institutionally chosen to be the education and production language of the elites who have true access to the production of knowledge. In terms of knowledge-related contexts, the language a) takes precedence to vernacular languages in education, and b) fully replaces vernacular languages in production. c) Adopted general monolingualism: Scholarly/scientific/intellectual production should happen in a language institutionally chosen to be the production language of all those who are interested in the production of knowledge. In terms of knowledge-related contexts, the language mostly replaces vernacular languages in production. However, vernacular languages are still the education language in most cases/countries, and training in the lingua franca (outside of the countries where the lingua franca is the spoken language) is an individual and/or national effort. 11 On the other end of the spectrum, we have a view that would reject any attempt towards monolingualism by combating the idea of a lingua franca altogether. This view, which has implicit or explicit defenders in the humanities, promotes that intellectual and scholarly production should use the national/regional languages. So we get a fourth possible view: d) Radical plurilingualism: Scholarly/scientific/intellectual production should happen in the national/regional languages that are the mother tongues of the scholars. Scientific and intellectual exchanges can happen in translation (if at all) but the scholars are rarely engaged in such exchanges. 12 Finally, there is another view that defends the value of the exchanges and allows for scholars to be actively involved in them while restricting the role of the lingua franca. This view recognizes not just the intellectual and scientific importance of pervasive communication and finds it essential in the development of human knowledge, but also acknowledges that the best way of achieving this goal is a shared language. However, it also stresses the importance of production in vernacular languages and maintains that the lingua franca should not have a privileged role when it comes to the value of scholarly production. So we have: e) Moderate plurilingualism: Scholarly/scientific/intellectual production should happen in the national/regional languages that are the mother tongues of the scholars or in a lingua franca.
Faced with this palette of options, we propose that deciding among them should be done by analyzing how well they do when confronted with the desiderata set forth in section I.

III. Arguments against monolingualism
How do the monolingualist options fare vis-à-vis the "fairness" desiderata? For simplicity purposes we do not explicitly discuss neutral monolingualism (exemplified by Esperanto) because of the impractical character of this proposal and its actual failure to gain any traction. It is enough to say that this view shares the defects we will find in the other monolingualist views, adding the fact that it is not the language of any human culture, and so a poorer way to express human experience, which is an entanglement of nature and nurture. There are some clear differences between the elite and general monolingual views regarding the influence that the choice of language has on the person or groups who are or could be active scholars. The elite view restricts the use of the lingua franca to a small number of individuals who are given access to knowledge in that language. For our purposes, consider early education that in the case of the humanities includes literature, philosophy and history, among other areas, done primarily in the chosen language so that the future scholars develop that language as their primary tongue for the "knowledge" part of their lives (whatever they speak in their everyday contexts). This option provides scholars with enough tools to have a high level of expressive power when generating work in their disciplines. Still, potential gains in expressivity fairness will be inversely proportional to how early in the scholars' lives the language is introduced and thus would make it more restrictive in most national contexts. Regarding our first desideratum (fairness), when it comes to monolingualism the "general" or non-elitist variety has to be preferred.
General monolingualism is less restrictive regarding fairness of access because it leaves open that, institutionally or individually, multitudes of scholars who are not originally speakers of the lingua franca can produce knowledge in it. This is, however, very different from a plain field providing fairness in access: Many countries without enough resources may not be able to institutionally support a bilingual effort that produces extensive access. Also, there are countries with regional/native languages that are already forced to bilingualism, so introducing a third language soon enough in the educational process would be a very heavy burden. 13 Beyond institutional training, early individual access will depend on economic status, and in many countries that would leave out a clear majority of potential scholars. A scholar already committed to a discipline can surely be able to be trained in the lingua franca later in life but that makes the scholar's true proficiency more difficult and places an extra burden on scholars.
More generally, and regarding fairness in expressivity, it should be pointed out that neither Latin nor English (nor any specific language for that matter) are truly cases of a lingua franca because none of them is ethnically and culturally neutral. Both are natural languages, belonging to a certain culture, with a specific history, involving specific social values, epistemic contexts, etc. The very fact that there is a difference between native speakers and non-native speakers is the source of the linguistic and epistemic injustice we should add to the points made above.
A monolingual approach, in some sense, fares better regarding the third desideratum. By increasing the chances of communication, and a fortiori the scope of discussion within specific areas of scholarship, a monolingual approach has good chances of advancing the development of lines of investigation in different subdisciplines. However, a moderate pluralist approach of the type we are defending, including the policy suggestions we make, will lose very little if at all compared with the monolingual approach in this respect, as we will argue later.
Regarding the fourth desideratum (novel perspectives or creative contributions to the different disciplines), there are, we think, clear reasons to avoid the monolingualist positions. In the humanities (as most likely in every other kind of academic enterprise) the possibility of exchanging arguments and viewpoints with others enriches our understanding of the topics we are working on. Many times ideas embodied in different languages show subtle differences that induce us to think about a given subject in another way and to explore alternative aspects of the topic. Monolingualism, thus, whatever language we choose as the lingua franca, is a sub-optimal guide for innovative knowledge. 14 Let us present a few examples from philosophy in order to understand why allowing the exchange of ideas in different languages augments the chances of getting multiple viewpoints and novel perspectives on a given field. The examples are from just one area, the role of personal pronouns in philosophy, but can be replicated in several other areas. In the 17 th century, philosophical reflections about subjectivity flourish after Descartes famous dictum "Cogito, ergo sum". 15 Since then, the first-person singular pronoun has been at the center of epistemology and theoretical philosophy. But there are some differences -for example between English and Spanish-concerning the words that can be used in order to think about these issues. In Spanish (as well as in French and Latin) we use the first person singular pronoun ("yo") as a noun in philosophical contexts, while in English we do not usually use "I" (the first person pronoun) as a noun, but "Self", when dealing with this kind of philosophical problems. However, "Self" involves a second-order stance towards the subject, a reflexive level that the first-personal pronoun does not involve. We also use the word "Self" as a prefix generating many other specific philosophical issues like "Self-knowledge", "Self-reference" etc. In Spanish, the translations would be "autoconocimiento" and "autoreferencia," respectively, which are not a complex word including "yo". "Auto" is a prefix that cannot be used as a noun, which indicates the reflexive stance we were mentioning before. It might seem that we are concentrating on just a simple linguistic detail, but the burden of the philosophical reflections embedded in the reification of the first person viewpoint using a noun ("Self") are not reflected in Spanish, where there isn't -outside from technical uses-any noun to talk about subjectivity. We can talk about "El Yo" but this expression still has the flavor of a pronoun. Moreover, there are other languages where there is no single word that has the role of a first person singular pronoun: In Japanese, for example, there are more than fifteen ways to say "I" depending on the relation between the speaker and the person is being addressed (see Arisaka 2001: 199).
There are also other philosophical problems concerning personal pronouns. We think in a language (English or Spanish, in our case) where there are three grammatical persons: First, second and third (singular and plural). Philosophers and psychologists debate about what the source of our knowledge of minds is: Our own mind (the first person, the subject, the Self) projected onto the other, or an objective stance about the world, where all subjects are also studied as objects, as observable physical objects, through their public behavior (the third person stance) (Nagel 1986). And a recent view 13 It should be noted that the difference between becoming bilingual by mastering a national/unifying language in addition to a regional/native language, and becoming bilingual adding an international lingua franca is an important one. On the one hand a national language, as opposed to a lingua franca, is present in everyday life in real interaction contexts and media. On the other hand, countries have abundance of people who can teach their national language but would obviously struggle to recruit instructors to produce that type of generalized bilingual education in a foreign language. (The historical fact that some of those national languages were imposed as a lingua franca should not make the idea of a new imposition more palatable.) 14 Additional arguments can be found in Gobbe and Russo (2020). 15 Descartes is a pioneer using his vernacular language in the Discours de la Méthode as an alternative to the lingua franca in order to have a wider audience for his ideas, even if some of his main contributions were still written in Latin, such as Meditationes de Prima Philosophia, in qua Dei existentia et animae immortalitas demonstratur. He is a good model for a version of the moderate plurilingualism we are advocating in this paper. even claims that intersubjective exchanges (second-personal interactions) are the basic level of analysis in order to give an account of understanding minds (Gomila 2002, Scotto 2002, Reddy 2008. But there are some languages where there are more than three grammatical persons, such as Aymara, Guaraní and Quechua (Mannheim 1982, Bossong 2009). Of course, there will always be a speaker, an addressee and third parties in human conversations, but we cannot take for granted that every human language has three grammatical persons in their lexicon. The way we think about minds, attached to personal pronouns in Anglo-Saxon philosophy research percolates philosophical arguments and theories (See Pérez 2018 for additional arguments). Natural languages are the starting point to philosophize, and hence they guide the enquiry to one place or another: Language is not a neutral vehicle for our thoughts. Taking different languages as a starting point to our enquiry allows us to explore more philosophical options and enriches philosophical discussions. Finally, a monolingual approach, at least in the humanities, where a good part of the production is intended to non-scholarly audiences, 16 would patently fail when it comes to desideratum 5 (scope of the audience for scholarly production). Essays in disciplines such as history, if they were produced in the lingua franca, would not reach enormous segments of the world population, and more importantly the population that it should reach (think about Czech history not being written at all in Czech). Clearly, translations would not be a substantive solution, in particular in the case of languages with a volume small enough that would make translations economically unfeasible.
Hence, at this point, it could be stated that we should reject that very idea of a lingua franca in its monolingual sense.

IV. Arguments in favor of moderate plurilingualism
So far we have established that none of the versions of monlingualism constitutes good policy for scholarly production in the humanities. How should a plurilingual alternative look like? There are at least three alternatives to explore. First, we can consider the view we called radical plurilingualism. This approach would do well regarding the first two desiderata: Fairness in access and fairness in expressivity. Nevertheless, radical plurilingualism is not the right stance when it comes to the impact in the discipline itself. If policies along this view were adopted, there would be the risk of generating a situation where a global community cannot be developed, as everyone shares their ideas only with the members of their own linguistic group. Insularity is the risk of this strategy. 17 Regarding the third desideratum (advancement of topics in a discipline), this view prevents the creation of a wide network of scholars who would participate in detailed discussions in different subdisciplines, derailing progress in those disciplines and in the humanities as a whole. But this stance also works against the fourth desideratum of generating novel perspectives and stimulating creative contributions, since different linguistic communities (in particular the smaller ones) would have very limited exposure to issues and ideas not being developed in those particular communities. Finally, regarding the scope of the audience (desideratum 5), while radical monolingualism would allow scholars and intellectuals to reach their own linguistic communities with less technical production, it would limit the audience a scholar can reach beyond it, again relying on the market-driven publishing companies choices.
Thus, we propose a plurilingualism that is not radical. As a possible version of it, let us consider a view that can be called functional plurilingualism. According to this view, given the contingent historical fact that English is nowadays the language chosen for all kinds of global exchanges, there are pragmatic reasons for choosing it when we want to communicate our scholar achievements to a global audience. But there are also pragmatical/practical reasons (of the kind given above) in order to use the vernacular language when someone wants to communicate research results to a local audience. So, it can be argued that scholars could produce their more "technical" scholarship in the lingua franca and their more general audience scholarship in the vernacular language. In our view this is better than the monolingual options but is still a sub-optimal scenario. First, the distinction between general audience and technical production is not clear-cut in the humanities. In the case of natural sciences, there is a growing division of labor between scientific researchers and those who explain science to the public, for instance, scientific journalists. A good scientist might not be a good science communicator and viceversa. But in the humanities this division, for the most part, does not exist. It is the same scholar who gives a talk to a non-specialist audience and who writes a difficult to understand book about their new insights. Consider for example Foucault's Histoire de la Sexualité or Marx's Das Kapital and the social impact they had, even being quite complex scholar productions. 18 Second, institutional pressure to adopt a lingua franca in countries with a different vernacular language would discourage authors to produce in their native languages altogether. And given the fact that, as we said above, there is no division of labor between researchers and communicators in the humanities, the social impact could turn out to be smaller. Finally, only English-speaking students will have direct access to main contributions in the field, leaving all other students condemned to read manuals, instructional texts or (possibly scarcely available) translations.
All the above arguments bring us to propose a moderate plurilingualism as the preferred non-radical version of plurilingualism, where everyone could produce their scholar contributions in whatever language they choose. But, given the fact that it is also valuable to have the opportunity to exchange our ideas as widely as possible, every scholarly production should be published in two languages, one of them chosen by the author, the other one the lingua franca (or, in the case of English-speaking scholars, a second language that could be the editorial language, the intended audience language, etc.) We will suggest some policies along these lines in the final section.

V. Some institutional actions to encourage moderate plurilingualism in the humanities
As we argued above, our proposal is to adopt a pluralist view about language use for the generation of knowledge, including publication in the humanities. This language pluralism is moderate, since according to it everyone should be encouraged to write both in their native tongue and in the lingua franca so authors would have the opportunity to share their ideas with an international community but also with their own community (in particular the members of their community who do not have access to the lingua franca).
We propose that in order to achieve this goal, all scholarly production should be published in two languages: The scholar's native language and English. 19 With this modest policy change, access to scholars' production will be larger than today, because there will be more papers published in languages other than English. In this way, more people who are non-English speakers will have access to high quality material in their disciplines. 20 Scholars will also be in a better position concerning expressivity, because everyone will be able to publish in their own mother tongue without being penalized. 21 The disciplines themselves will also be benefited because they will generate new publications in a broad number of languages, and therefore there will be more potential readers. Finally, the peculiarities required to express a given content in a given language will also be respected, because one of the languages will always be the one in which the ideas of the scholar can be better expressed. In this ideal world every contribution would be published in two languages, hence the "original" language of publication will be both of them.
To make this idea come true, publishing companies that manage scholarly journals and books should accept contributions in different languages, 22 and when a paper/book written in a language other than English is accepted for publication, the venue should publish it in both languages. 23 In order to be fair with non-English writers and readers, the publishing companies should require English-speakers to publish their work in two languages as well: English and some other language of their choice: The one they are familiar with, the one used by the audience they want to reach, or the one of the country of origin of the publishing company. Probably, English speakers will have trouble in producing papers in other languages, but this is exactly what happens today with non-English speakers, and if everyone is required to publish their work in two languages, it would be at least as difficult for a non-English speaker as for an English speaker to get published. This would democratize access at the same time it builds a stronger scholarly community, where scholars would be actively engaged in being understood and discussed by a wider audience of specialists. 24 Still, these changes most likely will take a long time to be implemented, and certainly we cannot change the past. Therefore, we also propose that the institutions in charge of evaluating scholars, in all evaluative situations, assess all of the scholars' production without bias, no matter in what language it was produced. Scholars and scientific researchers are continuously evaluated: Publications involve peer review, each position -permanent or temporary-is filled after an evaluative process, salary raises are frequently the result of some kind of evaluation of performance/publications, promotions and transitions from probationary/tenure-track to permanent/tenured statuses heavily rely on publication records. 25 We suggest that the considerations made here be applied in all these instances of evaluation. The quality of the paper/book is what matters, not the language in which it is written.
If all the scholarly production is to be evaluated without bias in all languages, the list of potential evaluators should be enlarged: there should be evaluators that are competent readers of written material in different languages. We know that this is not an obstacle, given the fact that in many non-English speaking countries there is a long tradition in different disciplines and there are many well-known journals and publishing companies with high quality standards (independently of whether they are indexed or not), many of them belonging to universities or research institutes in non-English speaking countries. 26 Undoubtedly, implementing these policies would require a comprehensive collective effort; our proposal should be considered as part of a bigger project involving several groups and actors around the globe, 27 generating tools in order to promote trans-lingual communication without losing written works in any language. We think that preserving the production of written works in our fields in different languages would be highly beneficial for scholars, for the audience, and for the discipline itself.