The battle of the languages in national publishing A comparative study of the publishing performance by cnpq (Brazil) and Conicet (Argentina)¹

Abstract

The predominance of English as an academic language in mainstream journals has been extensively studied. In change, it is difficult to gauge the incidence of publication in this language in the Ibero-American world because there are still few studies of regional databases or based in complete academic trajectories of researchers from the countries of the global South. The reasons are simple: a) there are multiple databases of Ibero-American journals, but with large overlaps in between; b) there are curriculum systems in most countries, but they are not openly available or professionally curated. Precisely to collaborate in this direction, this comparative work offers an overview of the complete publications of two scientific communities: researchers from the Conselho Nacional de Pesquisa (CNPq, Brazil) and researchers from the National Council for Scientific and Technical Research (Conicet-Argentina). To focus on the language and country of publication of this scientific production, we used the information uploaded in Sigeva (Argentina) and Lattes (Brazil) curricular system. Its use offers difficulties to bibliometric or citation studies, however, it allows the construction of descriptive statistics of the production of complete production itineraries. A relevant information to offer an overview of multilingualism, bibliodiversity and the weight of the national publication in these communities.

Keywords:
Multilingualism; National publishing; Publishing circuits; Sigeva-Lattes; Conicet; cnpq.

Resumen

El predominio del inglés como lengua académica en las revistas de corriente principal ha sido profusamente estudiado. En cambio, es difícil calibrar la incidencia de la publicación en este idioma en el mundo Iberoamericano porque son todavía escasos los estudios de bases de datos regionales o de trayectorias académicas completas de investigadores de los países del Sur global. Las razones son sencillas: a) hay múltiples bases de datos de revistas Iberoamericanas, pero con grandes solapamientos; y b) hay sistemas de currículo en la mayoría de los países, pero no están disponibles abiertamente o no están curadas profesionalmente. Precisamente para colaborar en esta dirección, este trabajo comparativo ofrece um panorama de las publicaciones completas de dos comunidades científicas: los pesquisadores del Conselho Nacional de Pesquisa (cnpq, Brasil) y los investigadores del Consejo Nacional de Investigaciones Científicas y Técnicas (Conicet-Argentina). Para poner el foco en el idioma y el país de edición de la producción científica, utilizamos la información autocargada en el sistema curricular Sigeva (Argentina) y Lattes (Brasil). Su utilización ofrece dificultades a los estudios bibliométricos o de citaciones, en cambio, permite construir una estadística descriptiva de la producción completa de itinerários completos de producción que es relevante para ofrecer un panorama del multilingüismo, la bibliodiversidad y el peso de la publicación nacional en estas comunidades.

Palabras clave:
Multilingüismo; Revistas nacionales; Circuitos de publicación; Sigeva-Lattes; Conicet; cnpq.

Resumo

A predominância do inglês como língua acadêmica nos principais periódicos tem sido amplamente estudada. Por outro lado, é difícil mensurar a incidência da publicação nesse idioma no mundo ibero-americano, porque ainda são poucos os estudos de bases de dados regionais ou baseados em trajetórias acadêmicas completas de pesquisadores dos países do Sul global. As razões são simples: a) existem várias bases de dados de revistas ibero-americanas, mas com grandes sobreposições entre elas; b) existem sistemas curriculares na maioria dos países, mas eles não estão disponíveis abertamente ou com curadoria profissional. Visando a colaborar nessa direção, este trabalho comparativo oferece um panorama das publicações completas de duas comunidades científicas: pesquisadores do Conselho Nacional de Pesquisa (cnpq, Brasil) e pesquisadores do Conselho Nacional de Pesquisas Científicas e Técnicas (Conicet-Argentina). Para focar o idioma e o país de publicação dessa produção científica, utilizamos as informações disponibilizadas no sistema curricular Sigeva (Argentina) e Lattes (Brasil). Seu uso oferece dificuldades para estudos bibliométricos ou de citação, porém permite a construção de estatísticas descritivas da produção de trajetórias completas de produção. Uma informação relevante para oferecer um panorama do multilinguismo, da bibliodiversidade e do peso da publicação nacional nestas comunidades.

Palavras-chave:
Multilinguismo; Publicação nacional; Circuitos editoriais; Sigeva-Lattes; Conicet; cnpq.

The concern about the growing centrality of English as an academic language is not new. The use of the impact factor in the evaluations and the hierarchy meant for publication in mainstream journals effectively inclined the academic elites of non-hegemonic countries to publish in English (Ortiz, 2009, Gingras, 2016). It even produced linguistically segmented circuits of production and circulation, as documented for the Arab world (Hanafi and Arvanitis, 2014) and a systematic tendency of researchers to publish outside the region (Da Silva Neubert, Schwarz Rodrigues and Mugnaini, 2021). The world report recently published by Unesco (2021) also points out the inequalities produced by the globalization of English as the publication language and observes its effects according to the regions: the hegemony of English seems to have deepened compared to the previous report. For its part, the Organization of Ibero-American States carried out a study that had quite an impact because it was reported that, in 2020, 95% of all articles published in scientific journals were written in English and only 1% in Spanish or Portuguese. Only 13% of the scientists in Spain had published their work in Spanish, 12% of the residents of Mexico, 16% of the Chileans, and around 20% of those from Argentina, Colombia, and Peru. The situation of Portuguese is even more complex because barely 3% of the Portuguese researchers and 12% of the Brazilian researchers analyzed in this report chose their mother language to publish their work, the rest did it in English (Badillo, 2021).

This heteronomous environment is bleak for multilingualism and leaves little room to imagine a transformation of these trends. But this is a landscape that is built through the lenses of the available databases of the oligopolistic commercial publishers (Scopus or wos-Clarivate) whose geographic and linguistic biases have already been widely analyzed (Archambault, Vignola-Gagné, Côté, Larivière and Gingras, 2006; Unesco, 2010). These databases, in fact, obscure a considerable intellectual activity that continues to take place in the world, in many languages, evidencing the value that these have for writers and also for readers. This global reality of multilingualism, which crosses different spheres, not only academic, calls into question the naturalization of English as the privileged language of publication (Curry and Lillis, 2022). In the social and human sciences, on the other hand, not only is writing in native languages, but multilingualism is growing, regardless of the geopolitical position or the size of the academic community (Kulczycki et al., 2020). It is true that the publication in English promotes the citation of literature in English in different ways, but there are studies that show that there is resistance and negotiations on the part of researchers who come from countries with other native languages to preserve their writing styles (Smirnova and Lillis, 2022).

There is sufficient evidence on the co-existence of various circuits, different ways of conceptualizing the notion of “excellence” and different forms of knowledge production, not only in the South but also in the North (Paradeise and Thoenig, 2015; Mbula, Tijssen, Wallace and McLean Eds., 2020). On the other hand, we know that the book continues to develop in the academic world and bibliodiversity practices emerge when the complete trajectories of researchers are observed (Engels et al., 2018; Dacos and Mounier, 2010). In other words, nobody doubts that the hypercentrality of English is a reality, but there is enough empirical material to affirm that it is not the only reality.

However, it is difficult to truly gauge the impact of publishing in English in the Ibero-American world because there are still few studies of academic trajectories or complete universes of published production. The reasons are simple: a) there are multiple databases of journals edited in the region, but non-interoperable and with large overlaps and b) there curriculum national information systems exist in most countries, but are not openly available or professionally curated (Beigel, 2022). However, these databases are being increasingly studied and allow an overview not only of the production in indexed journals but also in other media and formats where local languages predominate (Mugnaini, Damaceno, Digiampietri and MenaChalco, 2019; Beigel and Gallardo, 2021).

This research is part of a comparative study of two scientific communities, researchers from the National Council for Scientific and Technological Development (cnpq, Brazil) and researchers from the National Council for Scientific and Technical Research (Conicet, Argentina). In this article, we focus on the language and country of publication of the scientific production. The origin of the data is similar for the two countries. In the case of Argentina, it consists of information self-loaded by Conicet researchers in its management and evaluation system (Sigeva). In the case of cnpq, the information was extracted from the Lattes system, which centralizes the curricula of Brazilian academics. Both Sigeva and Lattes are curricular information systems, and not bibliometric databases, which is why they do not have a validation process for the records loaded into them. These are filled by the researchers and may contain errors, omissions or repetitions. For this reason, they require more filtering and cleaning work, as well as crossings with complementary databases, such as issn to determine the country of the journals, dois to retrieve unique information from articles, among other procedures. In the absence of integrated Cris systems, these databases allow us to approach entire universes of publications. Its use offers difficulties to bibliometric or citation studies, but, on the other hand, it allows to build a descriptive statistic of the complete production of a specific population of scientific researchers that is adequate to the main objectives of this work: to offer an overview of multilingualism, bibliodiversity and the weight of the national publication in these communities.

Language skills and academic capital asymmetries

The ability to write and publish a production in a language other than the mother language is closely linked to disciplinary diversity and the spatial location of the person’s institution of affiliation. Several studies have already shown that writing in English does not simply arise from the deployment of basic communication skills, but is about a broader set of linguistic abilities (Lillis and Curry, 2010; Chardenet, 2012). What Gerhards (2014) calls “transnational linguistic capital” - whose maximum accumulation is reported by English - is not acquired merely with the typical training of primary socialization. Academic training, the intervention of native directors or collaborators who correct or translate, whose access is defined according to the academic and social capital of the research teams and their international networks (Beigel, 2017) converge for their technical mastery. The accumulation of these resources and the feasibility of acquiring these dispositions to write in English explain the unequal circulation that is registered between academics from the same country and the same discipline, as well as helping to explain gender asymmetries. Thus, a researcher affiliated with a university in the United States or the United Kingdom has a competitive advantage, a greater facility to publish in the majority universe of journals in English included in mainstream databases, given their knowledge of English as their native language and their academic training in that language. On the other hand, for a Chinese, Russian or Colombian researcher, publishing in English implies intense learning, as well as additional revision and translation times, not to mention the need to adapt to certain debates and a literature that does not match the map of your previous readings, which is also in English.

Added to these basic inequalities, in which transnational linguistic capital plays a key role in entering or remaining in a certain academic circuit, there are other structures that define the “room for maneuver” that researchers from non-hegemonic countries have to write in their native languages. We refer to the admission and promotion policies that focus their evaluation indicators on the impact factor of the journals rather than on the assessment of the quality/originality of each published article or chapter. Thus, the rankings of journals, and their weight in university rankings, have had a harmful effect on the multilingualism of scientific production and publishing, twisting publication strategies around journals in English. Another direct effect is related to the devaluation of national journals that frequently lose support from the community and institutional support, with which scientific communication bodies are closed that link universities more directly with their environment and encourage them to develop socially relevant knowledge. It mobilized many journals to change their native language to English.

Comparative studies confirm that researchers from the Southern Cone share the perception that publishing in English is very important to advance in an academic career and, for the most part, they have and wish to have more publications in English. The Survey of Language Capabilities and Internationalization (Ecapin, 2018)² shows that between 92 and 96% of the total number of respondents from the three countries stated that they had published at least once in that language. Many responded that they have a large part of their production in English. Ecapin was developed in selected academic populations of these three countries, belonging to all scientific areas, considering only highly internationalized institutions, since it sought to study the academic elites to investigate the incidence of school capital of origin for publication in English. However, a low incidence was verified, thus demonstrating the weight of other factors which we mentioned above. In the case of Argentina, the survey showed that 95.1% of researchers with very low/low school capital had published in English at some time, and the percentage decreased as the school capital of origin increased, being 93.5% among those with high school capital (Gallardo, 2022). For Brazil, 97.2% of researchers with very low or low school capital had published in English at some time, and among those with high school capital the figure rose to 99.2%. A slightly higher correlation existed in the case of Chile, where 87.5% of researchers with very low/low school capital had published in English at some time, while among those with high school capital the figure rose to 92.3%.

However, when the complete curricula of these researchers were analyzed, despite the great weight of English in self-perceptions, a significant presence of publications in Spanish or Portuguese was observed. This occurred in all scientific areas, and with a significant relationship with publications in journals of the country itself, with more diverse profiles than expected emerged in these academic elites with full integration into mainstream circuits. It is interesting to highlight as a salient feature of Brazil that in this study it was verified that only 8% of the researchers surveyed did not have any publication in Portuguese -something quite striking considering that it was the population of professores of graduate programs with category 7, the most hierarchical sector of that national academic community. This panorama becomes even more interesting when publishing practices are analyzed in broader populations of Brazilian professors, outside of the elites studied in Ecapin. A study by Mugnaini, Damaceno, Digiampietri and Mena-Chalco (2019) analyzes the complete list of publications of 260,663 people in that country and shows that national journals occupy a significant portion of the articles in all areas. Even more striking is that 60% of all journals in which these articles were published correspond to journals not indexed in Scopus, wos or Scielo.

Let us now focus on the findings of this study regarding the incidence of national publication. Figure 1 shows two opposite profiles between Brazil and Argentina: the first with 40% of researchers who have more than half of their publications in the country, and Argentina with just over 10%. The group of Brazilian researchers also has the lowest percentage of researchers without national publications (4%). If we now look at the percentages of researchers who do not have publications in their own country or who have less than 11%, we see that Argentina has 55% of its researchers who only exceptionally publish in national journals. Meanwhile, Brazil draws attention precisely because it presents the opposite pattern, it has practically no researchers without publications in their own country and 58% usually publish regularly in Brazilian journals.

Figure 1
Percentage of national publications, by country

To explain this relevant trend of publication in Brazilian journals, it can be pointed out that many journals in this country have made a transition to English (Beigel et al., 2022), but this does not seem to be the only explanation, as we will see below. In the case of Argentina, it is convenient to clarify that Conicet does not have a salary (or fellowship) system stratified according to productivity or direct incentives for mainstream publication. On the other hand, in this country the national journals indexed in Scopus or wos are exceptional and scarce, the majority being indexed in Latindex. Lastly, this organization has stimulated in its evaluation processes the Latin American circulation in the social sciences and humanities (ssh), valuing the publications in journals indexed in the region. Part of the publications outside the country are found in that circuit, although they are not expressed in Figure 1.

The decisive nature of scientific evaluation policies in the publication style of people and the weight that English has as a mark of prestige in the academic elites of non-hegemonic countries is verified in the comparison between two corpus studied in previous research. The first, carried out on the five “most relevant” productions that Conicet researchers must choose to present themselves for promotion in that body, detects that, of a total of 23,852 productions, 83% is published in English (Beigel, 2017). In contrast, the study of the total productions of these researchers self-loaded in Sigeva shows that only half (54.5%) is in English, a significantly lower percentage (Beigel and Gallardo, 2021). It is a fact that says a lot about the evaluative culture because when researchers have to choose the publications they consider most effective for mobility, they tend to select publications in English, while their complete trajectories show greater diversity. Let us now see in depth how the battle among languages and bibliodiversity is viewed, incorporating the disciplinary dimension.

Linguistic distribution of the published output of the researchers from cnpq (Brazil) and Conicet (Argentina)

To compare two research systems such as those of Brazil and Argentina, it is necessary to establish the main characteristics of two national systems for categorizing researchers that have elements in common and significant differences. Both systems are strongly internationalized, which is evidenced by the weight of the impact indicators of the journals and the publication in English for the promotion of their agents. However, there are important differences between the two countries. On the one hand, at Conicet there is no monetary incentive system for productivity: all researchers in the same category earn practically the same (there are only some small differences due to the geographical area of the professional address). The research career has five hierarchical positions that are achieved through a rigorous promotion process with strict productivity requirements. In Brazil, the productivity fellowship is not a research career with staggered positions, like that of Conicet, but rather it is a subsidy to complete a research project. However, there are five categories, with an ascending order according to the background of the candidate, and has an enabling character in that it grants a professor the status of “productive researcher”. Promotion in its different categories has become increasingly competitive both for the financing it implies and for the status it grants (Beigel et al., 2022).

The evaluation of the publications in both has important differences. In the cnpq, a fairly widespread predominance of the Scopus hierarchy is observed to award the maximum scores. A factor that also affects this field is the existence of a national journal evaluation system. In Brazil, Qualis brings together the classifications made by all the evaluation committees of Capes and cnpq, granting a national instrument that has favored the quantification of the evaluation of trajectories. In Conicet, impact indicators predominate in the “hard sciences”, but there is a special regulation for Social Sciences and Humanities that weights mainstream indexing services and regional ones such as Scielo or Latindex Catalog equally. The Caicyt of Argentina evaluates only the national journals that request it and there is no national instrument that guides the evaluation of international journals. The evaluation boards generally use mainstream criteria to decide on the admission or promotion of researchers, and in the social and human sciences an internationalization oriented to the Latin American circuit is observed. Ultimately, these are two academic populations over-selected and heavily pressured to publish at the highest level based on global criteria. There are significant differences in terms of the gender composition of the two populations: in Argentina, women are the majority (52% of the total) while in Brazil they are a clear minority (35% of the total). For the three major areas of “hard sciences”, it is clearly observed that, in all cases, both for Brazil and Argentina, the average number of publications in English is higher for men than for women. And, on the contrary, the average production in the national language is higher for women. If, as we argue, the publication languages reflect the internationalization of circulation and the positive effects it has on the evaluative culture of these organizations, there is a slight but clear trend towards a masculinization of international prestige and a feminization of local prestige (Beigel et al., 2022).

Let us now see, comparatively, how these two populations of researchers perform when analyzing the publications from the linguistic perspective, one being Portuguese speaking and the other Spanish speaking. For the cnpq researchers in Brazil, we extracted the total production declared in the Lattes system between 2013-2020. For the Conicet researchers in Argentina, we have the total number of productions uploaded to the Sigeva system from the beginning of each person’s career until February 1, 2020, when the extraction was carried out. The total number of individuals included in the study is the universe officially registered as Conicet researchers (10,619) and cnpq researchers from Brazil (14,418).

When we observe only journal articles, the English language clearly predominates, the proportion being higher in the case of researchers from Brazil (79%) than in the case of Argentina, where articles in English represent 64% of the total. As expected, the linguistic distribution changes according to the scientific area. Figure 2 shows the specificity of the ssh, where the articles in English represent a minority portion compared to the other areas. For Argentina it is a portion, around 15%, while in the case of Brazil it represents 25% of the total. It is interesting to note that, in Argentina, in the “hard sciences”, the proportion of publications in Spanish tends to increase in the lower categories of the career, that is, in the younger generations. Meanwhile, in the highest categories, made up of the oldest groups of researchers, the proportion of articles in English, inversely, increases. Among Conicet youth, there is even a growing trend towards publication in Argentina, driven mainly by the ssh and by the regulations that the organization has since 2014, in which the publications of these disciplines are classified, valuing regional indexing as equivalent to hegemonic indexing.

Figure 2
Articles by Conicet and cnpq researchers, by language and scientific area* (cnpq N = 14.784; Conicet N = 10.619)

We will see below that in Brazil the publications in national journals have a big share, but the main difference is that in Argentina they are mostly in Spanish while in Brazil the number of journals published entirely in English has increased significantly. Now, as expected, the distribution of publication languages is closely linked to the disciplinary dynamics. There are significant differences among the four areas selected for comparison³ that are observed with a similar pattern for the two countries. In the former, more than 90% of the articles are in English, while in agricultural sciences they are around 80%, with a greater presence of Portuguese for cnpq researchers than Spanish for researchers in these disciplines at Conicet.

The ssh of Argentina show a greater inclination to publish in Spanish, around 85% of the articles. However, as we said above, among these publications there are many articles in Ibero-American journals. cnpq researchers, on the other hand, may be publishing in English in journals from their own country and have a lower tendency to publish in Latin American journals.

The comparative study on gender asymmetries in this compared production showed that productivity shows a clear gap in favor of men and that when the linguistic variable is introduced, this gap is affirmed in all scientific areas. In the comparison of the average number of articles published in English is a notoriously favorable balance for male production in that language (Beigel et al., 2022). For Argentina, on average for all researchers, men published 23.6 articles while women, 17 (see Figure 3). In Brazil we observe a similar phenomenon: the average for all male researchers is 37.4 articles in English, while for women it reaches 30.1. The distances between the sexes triple in Brazil within the exact and agrarian sciences, while for the biological sciences and the ssh the gap is smaller. This is related to the gender composition of this population of cnpq researchers, which is characterized by significant masculinity, especially in the higher categories. In Argentina, the greatest distance is observed in the exact and natural sciences. The agrarian and social sciences have closer averages between the sexes. It is interesting to observe that in the biological and health sciences, Argentine women accumulate a greater number of articles in English than men, something that corresponds to a high rate of feminization of this group in Conicet.

A look at the complete productions of an academic community allows us to verify the limitations of the hegemonic databases, since not only journal articles are analyzed, but also all communication formats, such as books, conference proceedings, reports, which continue to have a impact on researchers’ publication practices. Let us now see how the idiomatic situation is inverted when we consider the chapters and parts of the book, for all scientific areas. Figure 3 shows that these are mostly published in the national language. In both countries, the ssh have similar proportions between English and the national language, while for the other areas, Conicet shows more book publications in English than Brazil. For Argentina, the book chapters in Spanish represent 61% and analyzed only the books, the values increase to 79%. For cnpq researchers, book chapters in Portuguese represent 74%, while for complete books the proportion in that language rises to 85%. The presence of other languages is minimal in all types of publication: only in ssh it is relevant.

As we see in Figure 3, researchers from both countries, when they publish in book format, do, mostly, in their own language. In contrast, book chapters in English have a significant presence in the “hard sciences”, with a higher incidence in Argentina than in Brazil. A stimulus to the publication of books and book chapters in Portuguese for the biological and health sciences comes from the Scielo Books collection. In total, 1,516 books have been published in this collection since its creation in 2014, with 1,383 published in Portuguese, 125 in Spanish, and only 8 in English.

Figure 3
Chapters and books by Conicet and cnpq researchers by language and scientific area (cnpq N = 14.784; Conicet N = 10.619)

The publishing spatial morphology

Let us now move on to analyze the publishing country of these publications, especially considering the weight of national journals in each universe of researchers. Starting with Brazil, we can see in Figure 4 that the social and human sciences tend to publish most of their productions in Brazilian journals, reaching 73% of the total. Outside of Brazil, only two countries can be noted, the United Kingdom with 5% and the United States with 4% of the total articles, and the rest of the countries are distributed in the remaining small percentage. The disciplines that show the highest percentages of national publication are Linguistics, Social Service and Home Economics. Due to the female predominance of these disciplines, we could think that women have a greater tendency to publish in Brazil, but if we isolate the articles published only by women in this scientific area, the total percentage grows slightly to 75% and the distribution by country is very similar to that of men.

Figure 4
Percentage of articles from social and human sciences (2013-2020) published by Conicet and cnpq, by country of the journal

Considering the publications of Argentine researchers in the social sciences and humanities, we can see that they tend to publish considerably less than their Brazilian colleagues in journals from their own country, reaching 40% of the total (Figure 4). Unlike Brazil, which does not reflect in a relevant way the publication in Portugal, as would be expected due to the linguistic affinity, in the Argentine case a significant portion of articles is observed in Spanish journals (9%). It is worth mentioning that 9% of the articles are published in Brazil, while for Brazilian researchers, Argentine journals do not seem to be a convening medium. In fact, in Figure 4, we see that there are practically no publications in Latin American journals, while a relevant regional publication phenomenon is observed for the Argentine case (journals from Chile and Brazil account for 14% of the total articles). As we anticipated, this is largely due to the fact that Conicet stimulated publication in indexed journals in Latin America, equating its hierarchy with Scopus and Web of Science in Resolution 2249 approved in 2014 (Baranger and Beigel, 2021).

The disciplines that show the highest percentages of national publication in Argentina are History and Geography (with 49% of the total articles in Argentine journals), the rest of the disciplines drop to averages of 43% or less. At Conicet, 58% of social sciences and humanities researchers are women, so it is convenient to isolate articles with female authors to analyze whether there are specific patterns of national publication. However, the distribution is almost identical to that of the group with 49% of articles published in Argentina and a similar share from the rest of the countries.

Figure 5
Percentage of agricultural sciences and engineering articles (2013-2020) published by Conicet and cnpq, by country of the journal

In the other scientific areas, we find a notable tendency to the publication of articles in national journals for Brazil, while, for Argentine researchers, it is something extremely exceptional. Figure 5 shows the area with the greatest presence of national journals is Agrarian Sciences and Engineering, where 36% of the articles by Brazilian researchers are published, while for Argentina the portion of national publications drops to 10%. They are followed by Biological and Health Sciences with 24% of the total for Brazil and only 7% for Argentina (Figure 6).

A phenomenon that should be mentioned and certainly contributes to offering an important communication circuit for Health Sciences and Agricultural Sciences in Brazil is the existence of the Scielo health journals, which constitute almost half of the total number of journals in the collection. Especially noteworthy are 109 journals in the health area, to which is added an important communication tradition that was born with the Bireme library and currently Lilacs, with which the offer of dissemination spaces is varied. Lilacs indexes more than 100 Brazilian journals in the Health area present in the Scielo network. This number is higher than the network journals, in Health, indexed in Scopus (80) or Web of Science (41), reinforcing the importance of this library (Bojo-Canales and Sanz-Valero, 2019). In Argentina there are also many medicine and health sciences journals (122 according to the most recent survey), but only 17 of these are indexed (Beigel, Salatino and Monti, 2022).

Figure 6
Percentage of articles in biological and health sciences (2013-2020) published by Conicet and cnpq, by country of the journal

Figure 7
Percentage of articles on exact and natural sciences (2013-2020) published by Conicet and cnpq, by country of the journal

The Exact and Natural Sciences is the area that exhibits the least participation in national journals, although the proportion for Brazil is still significant (14% of the total articles), while in Argentina it constitutes 7% (Figure 7). Considering gender differences, it is observed that, if we isolate the articles published by women, the percentage of publications in Brazil decreases slightly for Agricultural Sciences and Engineering (34%), increases slightly for Exact and Natural Sciences (19%) and for Biological and Health Sciences (28%). In Argentina, the percentages of national publication for women are practically the same as for the universe in the three areas, without any specific pattern standing out. On the other hand, as we saw before, the linguistic patterns of publication vary significantly in both countries, where there is a significant gender gap in publications in English.

If we now review the countries in which researchers from these countries publish in the “hard sciences” we can point out that the United States is a relevant publishing location, although with a more important role in Argentina. The United Kingdom plays significant role among the journals, on a par with the Netherlands for Argentina. This is largely explained because it is in these countries where most of the journals indexed in Scopus are published, and Argentina has not subscribed to the Web of Science collection for more than a decade, which, in fact, is not available for reading in the National Digital Library. Instead, Brazil has access to these collections and values them in its evaluation system of the productivity. For these three areas, journals published in Argentina represent less than 10% of the total, indicating a strongly internationalized trend, with a predominance of journals published in United Kingdom, Netherlands and United States.

Final considerations

Publishing in a language other than one’s own not only implies a complex process of adaptation and learning, but also involves a series of losses that occur with translation. Spanish and Portuguese are universal languages not only because they are spoken by hundreds of millions of people, but also because they have gathered antagonistic experiences, of conquerors and conquered (Sánchez Cuervo, 2021). Stripping them of their historicity and the richness of their particular path in a forced process of translation into English results in an intercultural impoverishment of science. In turn, it distances scientific research from direct interaction with its social environment, and these communities that could be beneficiaries of the results of science or the public policy makers that could use these inputs to generate informed actions. But, as Curry and Lillis (2014) have shown, academics adopt strategies to align themselves with the demands of publication in English, and at the same time use tactics to support agendas that have conflicting orientations. In some moments they adapt, in others they accommodate, and in others they resist the domination of English and this can be observed especially in their complete publication trajectories.

We were able to see that Argentine researchers publish very little in their own country, but maintain a percentage of their publications in Spanish in all areas. They have quality national journals, but even those who publish there estimate that Conicet will not reward them for their academic careers. A process of re-functionalization of the journals edited by scientific societies is required to improve their indexing and advance multilingual policies. Brazilian researchers regularly publish in journals in their country, but many of these are published exclusively in English. The small number of journals indexed in Portuguese, especially outside of Brazil, has been a strong driver of publishing strategies not only for Brazilians, but also for researchers from different countries that have Portuguese as their official language (Solovova, Santos and Veríssimo, 2018). This country has a quality and professionalized national communication circuit, with international visibility, which could eventually become a multilingual space, if it has the right translation policies and the right incentives from the financing and academic evaluation systems.

The participation of these researchers in the Latin American circuit yielded some conflicting findings, especially in the comparison in the practices of the social sciences and humanities. For Brazil, few publications were registered in Latin American journals, while a relevant phenomenon of regional publication is observed for the Argentine case and even a strong presence of journals from Brazil. This trend of regional internationalization constitutes an adaptive strategy developed in Argentina largely thanks to a recommendation promoted by Conicet that recognizes and values t hese journals in the admission and promotion evaluation process. All of these demonstrates that the incentives of the academic evaluation systems produce changes in the trajectories of researchers circulation.

In relation to the language of publication, we were able to detect levels of productivity differentiated by gender because men publish, on average, more articles in English. Given the valuation of this transnational linguistic capital in evaluation systems and global academic hierarchies, this gender gap can generate significant asymmetries for women when it comes to advancing in academic careers in both countries. For this reason, it is vitally important to address this phenomenon in the current reflections of the national systems for categorizing researchers that exist in both countries, in order to point to more responsible and equitable forms of academic evaluation.

This study has tried to show that when a complete corpus of production is in view and not the one that arises from geographically and linguistically biased databases, the national publication in the case of Brazil and the Latin American publication in the case of Argentina emerge as a significant phenomenon. On the other hand, our national languages continue to struggle and resist. Valuing them and making them visible is essential to combine an international science, at the same time rooted locally, so that it is increasingly relevant to the society that surrounds it.

• 2
  Ecapin Report was developed int the Neies-Mercosur Project No. 3/2015 financed by Capes (Brazil) and spu (Argentina), coordinated by Fernanda Beigel and with the collaboration of Ana María Almeida, Breno Bringel, Denis Baranger, Juan Piovani, Claudio Ramos Zincke and Osvaldo Gallardo. The results of this study can be seen in this same dossier that will be published by the Tempo Social in December 2022.
• 3
  There are four large Conicet areas, while in cnpq there are eight, so for comparison purposes, the Brazilian areas were grouped into the same four Conicet areas.

Referencias

• Archambault, É.; Vignola-Gagné, É. & Côté, G.; Larivière, V. & Gingras, Y. (2006), “Welcome to the linguistic warp zone: Benchmarking scientific output in the social sciences and humanities”. Retrieved December, 18.
• Badillo, A. (2021), El portugués y el español en la ciencia: apuntes para un conocimiento diverso y accesible Organización de Estados Iberoamericanos para la Educación, la Ciencia y la Cultura/ Real Instituto Elcano.
• Baranger, D., & Beigel, F. (2021), “La publication en Ibéro-Amérique en tant que mode d’internationalisation des chercheurs en sciences humaines et sociales du Conicet (Argentine)”. Revue d’Anthropologie des Connaissances, 15 (15-3).
• Beigel, F. (2017), “Científicos periféricos, entre Ariel y Calibán. Saberes institucionales y circuitos de consagración en Argentina: las publicaciones de investigadores del Conicet”. Dados, Revista de Ciências Sociais, Rio de Janeiro, 60 (3): 825-865.
• Beigel, F. (2022), “Multilingüismo y bibliodiversidad en América Latina”. Anuario Glotopolitica, 5.
• Beigel, F.; Almeida, A. Ciriza, A.; Pecheny. M.; Digiampietri, L.; Moschkovich, M. Gallardo, O.; Rossomando, P. & Gomez, M. S. (2022), “A comparative study of gender inequalities in scientific publishing and its impact in career-building in Brazil and Argentina”. Conferencia Chaire Unesco-Dauphine Femmes et Science. Chaire Femmes et Science Rapport Scientifique Unesco. Francia, Paris.
• Beigel, F. & Gallardo, O. (2021), “Productividad, bibliodiversidad y bilingüismo en un corpus completo de producciones científicas”. Revista Iberoamericana de Ciencia, Tecnología y Sociedad - cts, 16 (46): 41-71. Disponível em http://ojs.revistacts.net/index.php/cts/article/view/211
  » http://ojs.revistacts.net/index.php/cts/article/view/211
• Beigel, F.; Salatino, M. & Monti, C. (2022), “Estudio sobre accesibilidad y circulación de las revistas científicas argentinas”. In: Zukerfeld, M. & Terlizzi, S. (Eds.). Políticas de promoción del conocimiento y derechos de propiedad intelectual: experiencias, propuestas y debates para la Argentina Ciudad Autónoma de Buenos Aires, Ciecti. isbn 978-9874193-56-8, pp. 10-46.
• Bojo-Canales, C., & Sanz-Valero, J. (2019), “Las revistas de ciencias de la salud de la red Scielo: un análisis de su visibilidad en el ámbito internacional”. Revista Española De Documentación Científica, 42 (4): e245, doi:10.3989/redc.2019.4.1629.
  » https://doi.org/10.3989/redc.2019.4.1629.
• Chardenet, Patrick. (2012), “Langues et savoirs: perceptions et réalités du capital linguistique dans la circulation des connaissances”. Coloquio Circulación Internacional del Conocimiento Cinvestav-iiesu, México.
• Curry, M. J. & Lillis, T. M. (2010), “Academic research networks: Accessing resources for English-medium publishing”. English for Specific Purposes, 29 (4): 281-295.
• Curry, M. J. & Lillis, T. M. (2022), “Multilingualism in academic writing for publication: Putting English in its place”. Language Teaching, 1-14, doi:10.1017/S0261444822000040.
  » https://doi.org/10.1017/S0261444822000040.
• Curry, M. J. & Lillis, T. M. (2014), “Strategies and tactics in academic knowledge production by multilingual scholars”. Education Policy Analysis Archives, 22 (32). http://dx.doi.org/10.14507/epaa.v22n32.2014
  » http://dx.doi.org/10.14507/epaa.v22n32.2014
• Da Silva Neubert, P.; Schwarz Rodrigues, R. & Mugnaini, R. (March 2021), “Vai para onde? O destino da Ciência Latino-Americana e Caribenha”. Informação & Sociedade: Estudos 30 (4): 1-21. doi:10.22478/ufpb.1809-4783.2020v30n4.57794.
  » https://doi.org/10.22478/ufpb.1809-4783.2020v30n4.57794.
• Dacos, M. & Mounier, P. (2010), “Les carnets de recherche en ligne, espace d’une conversation scientifique décentrée”. Lieux de savoir, t. 2: Gestes et supports du travail savant Paris, Albin Michel.
• Engels, T.; Starcic, A. & Sivertse, G. (2018), “Are book publications disappearing from scholarly communication in the social sciences and humanities?”. Aslib Journal of Information Management, 70 (6): 592-607.
• Hanafi, S. & Arvanitis, R. (2014), “The marginalization of the Arab language in social science: Structural constraints and dependency by choice”. Current Sociology, 62 (5): 723-742.
• Gallardo, O. (2022), “Carrera académica y asimetrías de género en el Conicet, Argentina (2004-2018)”. Temas Sociológicos, 30. En prensa.
• Gerhards, J. (2014), “Transnational linguistic capital: Explaining English proficiency in 27 European countries”. International Sociology, 29 (1): 56-74.
• Gingras, Y. (2016), Bibliometrics and research evaluation: Uses and abuses Cambridge, Massachusetts, Mit Press.
• Mbula, Erika Kraemer; Tijssen, Robert; Wallace, Matthew L. & McLean, Robert (eds.). (2020), Transforming research excellence: New ideas from the Global South Cape Town, South Africa, African Minds.
• Kulczycki, E. et al (2020), “Multilingual publishing in the social sciences and humanities: A seven-country European study”. J. Assoc. Inf. Sci. Technolog, 71: 1371-1385.
• Mugnaini, R.; Damaceno, R. J. P.; Digiampietri, L. A. & Mena-Chalco, J. P. (2019), “Panorama da produção científica do Brasil além da indexação: uma análise exploratória da comunicação em periódicos”. Transinformação, 31.
• Ortiz, R. (2009), La Supremacía del inglés en las ciencias sociales Buenos Aires, Siglo xxi.
• Paradeise, C. & Thoenig, J. C. (2017), In search of academic quality México, fce.
• Sánchez Cuervo, A. (2021), Pensar en español https://www.madrimasd.org/cultura-cientifica/ciencia-cultura/ensayo/pensar-en-espanol
  » https://www.madrimasd.org/cultura-cientifica/ciencia-cultura/ensayo/pensar-en-espanol
• Smirnova, N. & Lillis, T. (2022), “Citation in global academic knowledge making: A paired text history methodology for studying citation practices in English and Russian”. Journal of English for Research Publication Purposes, 3 (1): 78-108.
• Solovova, O.; Santos, J. V. & Veríssimo, J. (2018), “Publish in English or Perish in Portuguese: Struggles and Constraints on the Semiperiphery”. Publications, 6 (2): 25. Disponível em https://doi.org/10.3390/publications6020025
  » https://doi.org/10.3390/publications6020025
• UNESCO (2021), Unesco science report: the race against time for smarter development S. Schneegans, T. Straza and J. Lewis (eds.). Paris, Unesco Publishing. Disponível em https://unesdoc.unesco.org/ark:/48223/pf0000377433/pdf/377433eng.pdf.multi#page=129
  » https://unesdoc.unesco.org/ark:/48223/pf0000377433/pdf/377433eng.pdf.multi#page=129

Publication Dates

• Publication in this collection
  27 Jan 2023
• Date of issue
  2022

History

• Received
  02 Sept 2022
• Accepted
  30 Sept 2022