Three recent studies on the research systems of nations that do not belong to the group of developed countries provide different evidence that helps us understand why we are the way we are.
The first is a dataset produced by Professor Brito Cruz (Elsevier) entitled New Geography of Research; the second is a study on the cartography of open science in the world and the position of countries in the Global South, developed by Professor Maria Fernanda Beigel, from the University of Cuyo (Argentina), and entitled “Cartographies for an inclusive Open Science”; the third is an article written by Professor Luciano Digiampietri and collaborators with the title “Approaching bibliometrics and prosopography".
Between geographies, cartographies and prosopography (in this case, in short, the study of professional trajectories), we discovered evidence about where we are and why we are the way we are. And, above all, we reflected on the future we are building in Brazilian research (and in other countries with similar income and development levels).
In New Geography, we see that global scientific production grew at a rate of 2003% per year (pa) between 2023 and 5, and that the growth rate in the number of authors occurred at an even faster pace: 5% per year in high-income countries and 10% per year in low- and middle-income countries. We also see that, according to the Scopus database, low- and middle-income countries now produce more than 50% of articles and have more than 50% of authors.
It is worth noting that this figure includes China and India, two giants that are likely to become even bigger. Since 2018, China has surpassed the United States in the number of publications, now accounting for more than a quarter of the global total (measured by the Scopus database). India still produces half as much as the United States, but it is already twice as much as the United Kingdom. Both are large, but have very different trajectories that would require separate analysis.
Even leaving aside these two countries, in terms of growth in the number of authors of articles, other countries in the middle or lower-middle income range, such as Brazil, Iran, Malaysia and Pakistan, appear with high annual growth rates of authors and higher than those of central countries.[1]
One question that readers may already be asking is about the increase in the number of citations of publications from low- and middle-income countries. According to data from Nova Geografia, today the numbers would rival those of central countries in the highest strata (articles in the top 1% of citations).
For these reasons, the author projects a new geography of research, since scientific production has been growing at a faster rate than in developed countries.
It is possible that we are still harboring feelings of inferiority and that we still see the global scenario from a third-world perspective. But it is also possible that this new geography is just announcing itself, without changing, in a perceptible way, the status quo of research on a global scale (except in the case of China).
Anyway, the ticket We have been able to participate in the game for quite some time now. I believe that this is true in all areas of knowledge. We produce quality knowledge in increasing quantities. We appear less than we produce and I suspect that we use the knowledge we produce much less than we could.
The reasons? Well, there are many, as with any complex problem. I discuss some of them below.
In the second study, on the mapping of open science, the author presents data that complement those cited above. The study addresses the situation of seven topics related to scientific production in countries of the Global South, with a focus on open science. I will use only two of them here: the prevalence of the Digital Object Identifier (DOI) as a mechanism for automatically identifying publications and citations; and multilingualism in scientific publications.
The widespread use of persistent digital object identifiers, of which DOI is the best known and most widely used for articles and journals, influences the ability to automatically search for publications and citations and, consequently, the metrics that are used to measure the performance of researchers, universities, journals and countries.
In her work on cartography, the author highlights that the use of DOI in countries of the Global South is much lower compared to those of the Global North.[2] On the Directory of Open Access Journals (DOAJ) platform, 54% of journals published in Latin American countries do not have this identifier. In Brazil, 48%. The author estimates that countries with higher income and smaller populations have between two and eight times more journals with DOI.
Considering the total number of DOIs (not just open objects), Latin American countries have much lower numbers than their number of publications. It is quite possible that this contributes to their lower frequency in global indicators. Without identifiers, almost nothing enters the numerology produced by large scientific publication databases.
Multilingualism is a topic that has been highlighted in several forums, especially those that seek to report on the impacts of science in languages other than English. I have always heard my colleagues say that English is the language of science. And it is, indeed.[3] But would it be the only one? As a general rule, publishing in Portuguese, for example, tends to make the text have less reach than if we do it in English. The point that stands out in the current discussion is that English should not be considered the only relevant language of science.
I could give some more or less obvious factual reasons as to why this is the case. But let's look at some data that will help us understand.
In the study on cartography (tables 7 and 8), we see the following data:
- Nearly 95% of publications based on the Web of Science are in English, while Spanish and Portuguese together account for 0,7%;
- In the Scopus database, this number is a little different: 85% in English; Portuguese and Spanish add up to almost 1,6%;
- In the OpenAlex database, 73% and 4% respectively for English and for Portuguese and Spanish together;
- Scielo basically records one third for each of these three languages.
Obviously, the source changes the search result. The graphs presented in the first study, of the new geography, would be different if the search database were OpenAlex or Scielo. Question: is there a way to better account for what is produced? Or are we stuck with a supposed (imposed) golden rule that does not allow us to diverge?
We already know that searching for our articles and citations in the Web of Science and Scopus influences the results. However, we need to study this subject further to better understand how the model works – and find solutions.
The third study I mentioned at the beginning of the text, that of “prosopography” (Digiampietri et al., 2024), analyzed publications by researchers from the National Council for Scientific and Technological Development (CNPq) and showed that 54% and 57% of the articles by these authors registered on the Lattes platform are found in the Web of Science and Scopus databases, respectively. The scholars did the same in the case of the researchers' publications tenured from the Consejo Internacional de Investigaciones Científicas y Técnicas (Conicet) of Argentina, finding 50% of them in those two bases.
Would the articles present in these databases be less scientific?
But wait. These numbers are relatively smoothed out when analyzed according to the major areas of knowledge. For Brazil, the percentages rise: 66% (agriculture), 70% (biology) and 80% (exact sciences) of the articles by those researchers are found in both databases or in at least one of them.
The areas that are most “left out” of the databases are those of social sciences and humanities: 80% of the articles by productivity researchers from Brazil and 72% of those from Argentina are not in the Web of Science or Scopus.
Not coincidentally, most publications in these areas are not in English. The dots come together.
What would happen to assessments if the databases were expanded, and today there are several of them, from the less supervised ones, such as Google Scholar, to others, broader, with curation, such as Dimensions, Scielo and Open Alex. The study by Digiampietri et al. (2024) gives a good idea of what would happen.
There is an elephant in the room. It has been there for a while. As long as it is around, geographies, cartographies and prosopography are not going to change any time soon.
This text does not necessarily reflect the opinion of Unicamp.
[1] In this text we will use different terms to refer to country blocks, precisely because the three texts on which we base ourselves use different approaches. The new geography uses income bands defined by the World Bank: high, middle and low income; the new cartography uses terms such as core countries and Global North and Global South; prosopography uses peripheral centers to talk about Latin America and especially Brazil and Argentina.
[2] There is a base cost of $1 per DOI. This cost, while not high, can be a factor in lower adoption in large quantities.
[3] The current King Charles III, when he was still a prince, said that England's main export product, since the formation of its empire, had been language, which became universal for science and almost everything else in terms of global communication. This is still the case, we just don't know for how long.
