Scholarly journal articles are the largest category of scholarly publications. The number of journal publications increased steadily except during the two World Wars.

Solid lines show fits to the data which have, in some cases, been shifted vertically for clarity. a.g Scholarly works (with relevant metadata) indexed in the OpenAlex database are categorized into seven groups. Most publications are journal articles. b. Number of journal articles has been doubling approximately every 10 years for over a century. While there were significant drops during World War I (1914 to 1918) and World War II (1939 to 1945), the growth of journal article publications quickly recovered. The solid line shows a linear fit to the logarithm of the dependent variable. c. Number of actively publishing journals has been doubling approximately every 14 years for over a century, but the growth rate has not been constant. The two solid lines show the linear fits to the logarithm of the dependent variable to different time periods. d. Number of countries represented in the affiliations of the authors of annually published journal articles. The data is well-described by a logistic curve centered at 1970. Most of the growth in the number of countries occurred between 1949 and 1991.

Long-standing, high reputation multidisciplinary journals have shifted their disciplinary focus quite dramatically over the last 100 years.

a. Time evolution of the interdisciplinarity indices, Id, of Nature, Science, and PNAS, measured by the entropy of the annual distribution of disciplines in published articles. b-d. Time evolution of the disciplinary partitioning of articles published in PNAS (b), Nature (c), and Science (d). For all three journals, there has been a shift from ‘Other Fields’ to Physics, Chemistry and Biology. In the 1920s, nearly 60% of articles in Nature or Science were in fields other than Physics, Chemistry and Biology. Interestingly, after following different paths, all three journals are now publishing nearly 40% of their articles in fields other than Physics, Chemistry and Biology. In PNAS, there was a noticeable shift towards Biology from 1960 to 1990 and away from Physics and “Other Fields”. After 1990, “Other Fields” grew at the expense of Biology. In Nature, there has been a steady growth of Biology focused articles. For both Nature or Science, the number of Chemistry publications began its ascent in 1940, reaching a pinnacle around 1970, and a slower decline after that.

Long-standing, high reputation multidisciplinary journals have become more international.

a. Time evolution of the internationalization indices, In, of Nature, Science, and PNAS, measured by the entropy of location of the affiliations of the authors of published articles. The index has increased drastically over the last 100 years for all three journals. b-d. Time evolution of the percentage of articles whose authors are affiliated with institutions from Asia, Europe, or North America in PNAS (b), Nature (c), and Science (d). For all three journals, the greatest growth has occurred for authors with affiliations in institutions located in Asia. Surprisingly, PNAS has now the largest fraction of authors from Asian institutions, even though Nature had a higher level in the 1940s. For both Science and PNAS, the fraction of authors from European institutions has increased dramatically since the 1960s. Around the same time, the fraction of authors from European institutions publishing in Nature decreased dramatically as the fraction of authors from North American institutions grew rapidly.

Trends in interdisciplinarity across journals in natural sciences, life sciences, social sciences, and multidisciplinary sciences in 1970, 1990, 2010, and 2020.

Most recent interdisci-plinarity index at a specific year, Id, versus the linear trend of the journal’s interdisciplinarity index over the preceding 5 years. Multidisciplinary journals consistently exhibit a substantially higher in-terdisciplinarity index (Id ≈ 3) compared to all disciplinary journals throughout the observed period (Table S1). Disciplinary journals in natural science (Chemistry and Physics) and life science (Biology and Medicine) appear to fall into two groups based on their interdisciplinarity indices: The first group of journals maintains a narrow focus, publishing research that draws on a single discipline (Id ≈ 0). The second group of journals publishes moderately interdisciplinary research (Id ≈ 1.5). This separation between mono-disciplinary and moderately interdisciplinary journals is visually apparent for journals in natural and life sciences. In contrast, the disciplinary journals in Economics and Political Science we studied seem to be moderately interdisciplinary (Id ≈ 1.5).

Trends in internationalization across journals in natural sciences, life sciences, social sciences, and multidisciplinary sciences in 1970, 1990, 2010, and 2020.

Most recent internationalization index at a specific year, In, versus the linear trend of the journal’s internationalization index over the preceding 5 years. The internationalization of disciplinary journals exhibits a variety of behaviors (see also Table S2). Physics journals have steadily increased their internationalization, which is now higher than for most multidisciplinary journals. comparable to those of multidisciplinary journals during this period. In 2020, the mean In of Physics journals is approximately 4.1. Journals in Biology and Chemistry display similar trends, but with lower mean In, approximately 3.6 for both Chemistry and and Biology journals. In contrast, while a few Medicine and Economics journals have In ≈ 4, most journals in these disciplines and in Political Science have In < 3.

Average interdisciplinarity characteristics of journals from different disciplines.

We report the average Interdisciplinarity Index Id, its most recent 5-year trend, the Spearman’s correlation coefficient of Id and the most recent 5-year trend, and the p-value of the correlation. We highlight statistically significant cases in bold face.

Average internationalization characteristics of journals from different disciplines.

We report the average Internationalization Index In, its most recent 5-year trend, the Spearman’s correlation coefficient of In and the most recent 5-year trend, and the p-value of the correlation. We highlight statistically significant cases in bold face.

Average team size of journal publications in different disciplines.

We report the average team size and its standard deviation for publications across various disciplines. We highlight Physics in bold face since it stands out for having the largest average team size.

Journal article publication trends of the five most prolific countries.

a. Total number of journal publications (1900-2021) of the most five prolific countries. b. Annual number of journal articles published by those five countries. For clarity, we show 5-year moving averages.

Number of journal articles over time across three bibliometric databases.

a. Number of journal articles indexed in OpenAlex over the past decade, categorized by the presence or absence of authorship and affiliation metadata. b. Number of journal articles indexed in Microsofe Academic Graph (MAG) over the past decade, categorized by the presence or absence of authorship and affiliation metadata. c. Number of journal articles indexed in Dimensions over the past decade, categorized by the presence or absence of authorship and affiliation metadata.

Number of journal publications in each Field of Study (FoS).

a. Total number of journal publications (1900-2021) in each L0 FoS. b-f. Annual publication count of each L0 FoS.

Relationship between the 2021 Interdisciplinarity Index and the ratio for disciplinary journals in Chemistry, Physics, Biology, Medicine, Economics, and Political Science.

Two Physics journals — Annals of Physics and General Relativity and Gravitation — and one journal in Medicine, Medical Clinics of North America have ratio values equal to zero, meaning that they do not display in the semi-logarithmic plot. Notice also that, for Political Science, the range for the values of both variables is quite narrow.

The marginal distribution of the Interdisciplinarity Index and the 5-year trend of 16 disciplinary journals compared to seven multidisciplinary journals for 2020.

Relationship between the 2021 Internationalization Index and the ratio for disciplinary journals in Chemistry, Physics, Biology, Medicine, Economics, and Political Science.

Notice that the range of values of the ratio is in many cases quite limited and close to 1, unlike the findings for the interdisciplinarity. Moreover, correlation is not always positive.

The marginal distribution of the Internationalization Index and the 5-year trend of 16 disciplinary journals to seven multidisciplinary journals for 2020.

Team sizes in journal articles are notably large in the fields of Physics, Biology, and Medicine.

The maximum number of authors in a Physics journal article is 4,473, while Biology and Medicine journals have maximum author counts of 676 and 798, respectively. We define large teams as those with 10 or more authors and small teams as those with fewer than 10 authors across all three disciplines. The 5-year moving average of articles with large teams began to increase across all three fields after 1960.

The marginal distribution of the Interdisciplinarity Index (Left) and Internationalization Index (Right) and the 5-year trend of 16 journals in Physics, Biology or Medicine excluding the articles with 10 and more authors compared to seven multidisciplinary journals.