The Insularity of Anglophone Philosophy:

Quantitative Analyses

 

Eric Schwitzgebel

Department of Philosophy

University of California at Riverside

Riverside, CA  92521

USA

 

Linus Ta-Lun Huang

Institute of European and American Studies

Academia Sinica

Taipei, 115

Taiwan

 

Andrew Higgins

Department of Philosophy

Campus Box 4540

Illinois State University

Stephenson Hall 412

Normal, IL  61790

 

Ivan Gonzalez-Cabrera

School of Philosophy

Australian National University

Acton ACT 2601

 

 

October 6, 2017

 

The Insularity of Anglophone Philosophy:

Quantitative Analyses

 

 

Abstract: We present evidence that mainstream Anglophone philosophy is insular in the sense that participants in this academic tradition tend mostly to cite or interact with other participants in this academic tradition, while having little academic interaction with philosophers writing in other languages.  Among our evidence: In a sample of articles from elite Anglophone philosophy journals, 97% of citations are citations of work originally written in English; 96% of members of editorial boards of elite Anglophone philosophy journals are housed in majority-Anglophone countries; and only one of the 100 most-cited recent authors in the Stanford Encyclopedia of Philosophy spent most of his career in non-Anglophone countries writing primarily in a language other than English.  In contrast, philosophy articles published in elite Chinese-language and Spanish-language journals cite from a range of linguistic traditions, as do non-English-language articles in a convenience sample of established European-language journals.  We also find evidence that work in English has more influence on work in other languages than vice versa and that when non-Anglophone philosophers cite recent work outside of their own linguistic tradition it tends to be work in English.

 

The Insularity of Anglophone Philosophy:
Quantitative Analyses

 

 

1. Introduction: Insularity, Mainstream Anglophone Philosophy, and Asymmetry of Influence.

Philosophers who write in English tend to read, cite, and discuss mostly other philosophers who write in English.  This is a widely acknowledged fact in informal discussions.  However, to our knowledge no one has published quantitative data demonstrating this fact, nor is there consensus on how severe the insularity of English-language philosophy is.  In this article, we present some quantitative data.[1]

We define insularity as follows: An academic subgroup is insular to the extent that people within that subgroup mostly cite or academically interact with others in that same subgroup.  For example, if specialists in ancient Chinese philosophy almost exclusively cite and academically interact with other specialists in ancient Chinese philosophy, then that subgroup is highly insular.  In contrast, if they cite and interact extensively outside of their subgroup – for example, drawing on resources in 21st-century ethics, or interacting with specialists in ancient Greek philosophy, then they are not very insular.  We claim that mainstream Anglophone philosophy is highly insular.  Scholars who belong to this group mostly cite and interact with other scholars who belong to this group.

Mainstream Anglophone philosophy is vague-boundaried and nebulous.  However, it can be characterized well enough to permit sociological examination.  Participants in this group are philosophers who write primarily in English (regardless of their native language); publish in English-language academic journals that are widely regarded as prestigious by other English-language philosophers, such as Philosophical Review and Ethics; belong to PhD-granting departments that are ranked in the Philosophical Gourmet Report, or have close scholarly ties to people in those departments; and are highly cited in the Stanford Encyclopedia of Philosophy and in prestigious English-language journals.  Individual philosophers differ in how central or peripheral they are to this group, with no sharp boundary of inclusion.  It is this fuzzy-bordered group that is the target of our analysis.[2]

In addition to measuring the insularity of mainstream Anglophone philosophy, we will also measure asymmetry of influence and language dominance.  Group A and Group B have symmetrical scholarly influence if Group A is as influenced by the work of Group B as vice versa – for example, as measured by rates of cross-group citation.  Group A and Group B have asymmetrical scholarly influence to the extent that one group is more influenced by the work of the other than vice versa.  We will argue that, over the past several decades, Anglophone philosophy has had a highly asymmetrical influence on the work of philosophers who write in other languages, for example, in Chinese and Spanish.

The dominance of a language is the influence of work produced in that language, relative to work produced in other languages, in the scholarly community being examined – as measured, for example, by proportion of citations.  For example, if a group of articles in Spanish cited 80% Spanish-language sources, Spanish would be the dominant language among the cited sources.  If it cited 60% Spanish, 30% English, and 10% all other languages, then Spanish would be dominant overall and English would be dominant among foreign-language citations.  Insularity, asymmetry, and dominance are interrelated, but they are not equivalent.  For example, a language could be dominant without being insular: If a group of articles in Spanish cited 80% German-language sources, German would be the dominant language in that group, but no conclusions about the insularity of German would follow. 

Although we believe that insularity, asymmetry, and dominance raise substantial ethical and epistemic issues, we will not address those issues in this article.  Insularity, asymmetry, and dominance might be ethically and/or epistemically justified if, for example, English-language work is much higher quality than work in other languages.  Also, one might celebrate English-language philosophy as a distinctive and valuable cultural enterprise that benefits from insulation, even if there is equally excellent or even superior work available outside of English.  Or one might think it sufficiently valuable that philosophy have a single lingua franca that Anglophone dominance and asymmetry of influence is a price worth paying.  We set these normative questions aside here.

 

2. Study 1: Citation Practices in Elite Anglophone Philosophy Journals.

In elite Anglophone philosophy journals, most of the citations are to works originally published in English.  Although this is easily detected by quick perusal, the magnitude of the phenomenon has never, to our knowledge, been measured.  If non-Anglophone sources are frequently cited in these journals, that is evidence against a high level of insularity and English-language dominance.  In contrast, if non-Anglophone sources are rarely cited, that is evidence of a high level of insularity and dominance.

Method.  We examined a group of twelve elite journals: the top-ranked journals in a 2013 poll of “Top Philosophy Journals, Without Regard to Area” at one of the best-known philosophy blogs (Leiter 2013).  The list has surface plausibility as group of journals regarded as elite in mainstream Anglophone philosophy, with Philosophical Review, Journal of Philosophy, Noûs, and Mind leading the list.[3]  We examined the most recent issue, as of September 8, 2016, of each of these journals, with the exception of Philosophers’ Imprint for which we examined all of 2016 to that date (due to its thin publishing rate and lack of grouping by “issue”).  From each selected issue, we examined only original research articles (not reviews, discussion notes, comments, symposia, etc.).  This generated a target list of 93 articles, most of which cited dozens of sources, for a total of 3556 cited references for analysis – hopefully enough to be a representative sample of citation practices in this group of journals.

A coder with substantial knowledge of both the current state of the field and the history of philosophy then hand-coded the reference sections of each article (or the footnotes when citations were not aggregated in a reference section), noting: the year in which the cited work was first published, the original language of the cited work, and whether the work was translated into English from another language.  Expertise in philosophy was valuable to this task due to uneven citation practices: Not all citations clarify the original language of the target work or its original publication year, or even that it is a translation rather than a work originally published in English.  (For this reason, automated searches of citations in Web of Science or Google Scholar can generate misleading results.)

Sometime after World War Two, English became the common language of most scholarship intended for an international audience, even when the writer’s native language is not English.  With this in mind, we divided the data into four periods by year of original publication: ancient through 1849, 1850-1945, 1946-1999, and 2000-2016.

Results.  Of the 3556 citations included in our analysis, only 90 (3%) were citations of works not originally written in English.  Of the 93 analyzed articles, 68 (73%) cited no works that had not originally been written in English.  Eleven articles (12%) cited exactly one non-Anglophone work, either in its original language or in English translation.  Fourteen articles (15%) cited at least two works originally published in a language other than English.  The only source languages other than English were ancient Greek, Latin, German, French, and Italian.  Thirty-three citations (1%) were of non-Anglophone work in its original source language.

Table 1 shows the breakdown by historical period.  As is evident from the table, there is virtually no citation (< 1%) of post-War work originally written in languages other than English.  Also potentially interesting is that English is the original language of 67% of the citations of work written between 1850 and 1945, despite the fact that the period includes globally influential work by such European philosophers as Marx, Nietzsche, Frege, Heidegger, Wittgenstein, and Sartre.  Only 154/3556 (4%) of citations were to works originally written before 1946. 

 

Table 1: Citation of work originally written in languages other than English, in twelve elite Anglophone philosophy journals, by original year of publication of the cited work.

Original pub year of cited source                   English            non-English     % English

Ancient through 1849                                     12                    51                    19%

1850-1945                                                       61                    30                    67%

1946-1999                                                       1228                8                      99%

2000-2016                                                       2165                1                      100%

 

Discussion.  In a representative selection of articles from elite Anglophone philosophy journals, almost all citations (97%) are to work originally published in English.  Twenty-first century work written in languages other than English was almost invisible: only one instance among 2166 citations.  We interpret these results as strong evidence of high levels of insularity.  As measured by rates of citation, philosophers publishing in elite Anglophone philosophy journals appear not to interact much with work by philosophers writing in languages other than English.

If very little philosophy were being published in languages other than English, that could explain the results.  However, that is not so.  For example, approximately 27% of the journals listed in the PhilPapers journals database are non-Anglophone.[4]  This is a floor estimate, since while the database is likely to contain every major Anglophone philosophy journal, it surely lacks many non-Anglophone journals (for example, it contains only one of the 15 elite Chinese-language journals analyzed in Study 5).  Furthermore, as we will see in Studies 4-6 below, philosophers writing in languages other than English find many recent non-Anglophone sources to be worth citing, both in their own language and in other languages.

These results fit also with evidence from Schwitzgebel (2012) who found that the “big three” philosophy journals (Philosophical Review, Journal of Philosophy, and Mind) each have on average only one article per year that mentions even one of five well-known “Continental” philosophers: Nietzsche, Heidegger, Sartre, Foucault, or Derrida.

Raw data for this study and the other studies are available at http://www.faculty.ucr.edu/~eschwitz/SchwitzAbs/Anglophone.htm.

 

3. Study 2: The Editorial Boards of Elite Anglophone Philosophy Journals.

Serving on the editorial board of an academic journal tends to reflect a high level of prestige and influence in one’s field, as well as (to varying degrees) influence over the scholarly direction of the journal.  We decided to examine the composition of the editorial boards of elite Anglophone philosophy journals.  If those editorial boards are mostly composed of philosophers housed in major Anglophone countries, that suggests a relatively high degree of insularity and/or asymmetry.  In contrast, if editorial boards contain relatively large numbers of philosophers from non-Anglophone countries, that suggests a higher level of influence from and interaction with scholars outside of mainstream Anglophone philosophy.[5]

Method.  We used the same ranking of Anglophone philosophy journals as in Study 1, but extended it to the top 15 instead of the top 12.  This incidentally resulted in a greater representation of journals specializing in philosophy of science.[6]  Some of these journals are “in house” or have a regional focus in the editorial boards.  We did not exclude them on those grounds.  It is potentially relevant to the situation that the two top-ranked journals are edited in-house by faculty at Cornell and Columbia respectively.

We examined the composition of editorial boards based on data on the journals’ websites on April 5, 2017.  We included editors in chief, associate editors, regular editorial board members, consultants, and staff with full-time permanent academic appointments, including emeritus.  We excluded editorial assistants and managers without full-time permanent academic appointments (which are typically graduate students or publishing or secretarial staff).

To determine institutional affiliation, we used the affiliation listed at the journal’s website when that was available.  For systematicity, we did this even in a few cases where the coder had personal knowledge that the information was out of date.  Otherwise, we used personal knowledge or a web search.  For editorial board members with multiple institutional affiliations, we attempted to determine which institution was their primary affiliation.  In a few cases where two institutions appeared to be about equally primary, we used the first-listed institution on the source page.

We did not attempt to track editorial board members’ country of birth or native language.  Our thinking was this: Someone who was born in Italy, for example, and who is now employed full-time in a U.S. academic institution and serving on the editorial board of an elite Anglophone journal, is likely to be interacting primarily with Anglophone philosophers, doing most of their philosophical work in English.  In contrast, someone located in a non-Anglophone country, even if serving on the editorial board of an elite Anglophone journal, is much more likely to regularly interact with and be influenced by philosophers in their non-Anglophone community.

Results.  In all, 562 editorial board members were included in the analysis.  Of these, 538 (96%) had their primary academic affiliation with an institution in an Anglophone country.[7]  Table 2 shows the breakdown by country.  Notably, the journal Synthese showed much more international participation than did any of the other journals, with 13/31 (42%) of its editorial board housed in non-Anglophone countries.  Only 4 (1%) were from non-Anglophone countries outside of Europe.

 

Table 2. Primary academic affiliation of editorial board members at 15 elite Anglophone journals.

Country                                               Number of board members     Percentage of total

Anglophone-majority countries

            USA                                        377                                          67%

            United Kingdom                     117                                          21%                

Australia                                  26                                            5%

Canada                                                13                                            2%

New Zealand                           5                                              1%

Total Anglophone                   538                                          96%

Non-Anglophone-majority countries

            Germany                                 6                                              1%

            Sweden                                   5                                              1%

            Netherlands                             3                                              1%

            China (incl. Hong Kong)        2                                              < 1%

            France                                     2                                              < 1%

            Belgium                                   1                                              < 1%

            Denmark                                 1                                              < 1%

            Finland                                    1                                              < 1%

            Israel                                       1                                              < 1%

            Singapore[8]                               1                                              < 1%

            Spain                                       1                                              < 1%

Total non-Anglophone            24                                            4%      

 

Discussion.  The vast majority (96%) of editorial board members on our list of 15 elite Anglophone journals are housed in universities in Anglophone-majority countries.

There might be excellent reasons for concentrating the editorial board membership in Anglophone countries.  However, it does suggest a degree of insularity.  At least in principle, an Anglophone journal could compose its editorial board in a way that draws its members more broadly from the global academic professoriate.  This would plausibly correlate with more openness to influence from non-Anglophone sources.  The example of Synthese shows that this is possible.

If publication in these journals is professionally valuable for career advancement in philosophy even in non-Anglophone countries – which appears to be the case in at least some countries – then these results also suggest asymmetry of influence: Non-Anglophone philosophers’ careers depend, to some extent, on impressing people in mainstream Anglophone philosophy, while the reverse is not so.

 

4. Study 3: Highly Cited Philosophers in the Stanford Encyclopedia.

The Stanford Encyclopedia of Philosophy (SEP) is widely regarded as the premier resource for up-to-date review articles on major topics in philosophy.  Contributors of articles are typically among the best-regarded researchers on their topics, and citation in the SEP is a plausible measure of prominence in mainstream Anglophone philosophy.

In Study 3, we examined the 100 most-cited 20th-21st century authors in the SEP to see how many of them were raised outside of Anglophone countries and/or wrote primarily in languages other than English.  If mainstream Anglophone philosophy is highly insular, then almost all of the authors most cited in the SEP should live in Anglophone countries and write in English.  If mainstream Anglophone philosophy is relatively less insular, then we might expect a substantial number of non-Anglophone philosophers in the past hundred years who have been influential enough to be highly cited in the Stanford Encyclopedia.

Method.  Citation data from the SEP are difficult to compile, so we rely on the citation data reported in Schwitzgebel (2014), which we believe are recent enough to still be representative of the current situation.  Schwitzgebel’s method was to download the bibliographical sections of every main-page entry in the Stanford Encyclopedia as of July 2014 (excluding notes and appendices), sorting by author.  Every author was counted only once per entry, and only if listed as first author; and authors born before 1900 were excluded.  Authors with common names (e.g., “J. Cohen”) were hand-separated, and prominent authors cited under different names (e.g., “Ruth Barcan” and “Ruth Marcus”) were hand-merged.

In this way, a list of the 267 most-cited authors was produced.  The list has good surface plausibility as a list of the most influential mainstream Anglophone philosophers of the past sixty years or so, with the top five being David Lewis, W.V.O. Quine, Hilary Putnam, Donald Davidson, and John Rawls.  The list also has more surface plausibility as a measure of prominence in mainstream Anglophone philosophy (as we have defined it) than do other widely used bibliometric measures such as Google Scholar and Web of Science.[9]

For this study, we examined the biographies of the top 100 ranked philosophers, using web resources plus personal and professional knowledge.  We determined country of birth, and when that country was not Anglophone-majority, we examined their biographies for countries of residence throughout their lives and in what languages their most influential works were published.  In some cases, where birthplace information was not easily obtainable, we inferred birthplace from location of undergraduate schooling.

Results.  Of the top 100 most-cited authors in the SEP, 87 were born in majority Anglophone countries (51 in the U.S. and 22 in the U.K.).  Among the 13 born in non-Anglophone-majority countries, six earned their undergraduate degrees in Anglophone countries and spent all or virtually all of their academic careers in Anglophone countries (Kim, Nagel, Parfit, Rescher, Van Fraassen, and Williamson), and one (Raz) did the same starting with graduate work.  The remaining six (6%) had substantial philosophical training or careers outside of Anglophone countries.  Four of them – Kurt Gödel, Carl Hempel, Karl Popper, and Alfred Tarski – emigrated to Anglophone countries during the Nazi era, while still less than forty years old, and spent the majority of their careers at Anglophone universities, producing important work in English.  Nonetheless, three of these four (Gödel, Popper, and Tarski) are probably best known for their early-career, non-Anglophone work.  Jaakko Hintikka taught in Finland for most of his career, but split time with various universities in the United States and published primarily in English.  Jürgen Habermas is the only philosopher of the hundred whose scholarly career has been by all measures primarily non-Anglophone.

Notably, despite their fame outside of mainstream Anglophone philosophy, Simone de Beauvoir, Jacques Derrida, and Michel Foucault appear nowhere on the full list of 267.  Jean-Paul Sartre is 133rd most-cited alongside Ruth Millikan, Stephen Schiffer, and Eleonore Stump.

Discussion.  Among the 100 most-cited contemporary authors in the Stanford Encyclopedia of Philosophy, only six had a substantial portion of their academic careers outside of Anglophone countries, and only one worked the majority of his career in a non-Anglophone country, writing primarily in a language other than English.  The period covered is long enough to include globally influential philosophers such as de Beauvoir, Derrida, Foucault, and Sartre, but none of them ranked among the top 100.

We interpret these data as evidence that mainstream Anglophone philosophers do not typically find it necessary to cite work by prominent recent non-Anglophone philosophers to feel that they have adequately reviewed major topics in the field.[10]

 

5. Study 4: Citation Practices in Non-Anglophone Journals in the JSTOR Database.

In the next three studies, we examine citation practices in philosophical journal articles published in languages other than English.  We aim to test three hypotheses.

First, do journal articles in other languages show the same pattern of insularity as do journal articles in English?  That is, do they almost exclusively cite sources from within their own linguistic tradition?  Or do they cite more broadly across linguistic traditions?

Second, is there evidence of asymmetry of linguistic influence?  That is, are recently published works in English more frequently cited in non-English-language journal articles than vice versa?  Or do English and other languages cross-cite each other at about the same rate?

Third, to what extent is English dominant among foreign-language citations in articles not written in English?  That is, when articles not written in English cite sources outside of their own linguistic tradition, are those sources more likely to be English than to be other languages?  Or are sources in English not especially more likely to be cited than sources in Chinese, French, German, Spanish, etc.?

Method.  For our first non-Anglophone study, we used a convenience sample of non-Anglophone journals available in the JSTOR database.  Our intention in using this sample was to find a variety of well established, easily accessible, internationally visible, European-language journals, not confined to a single language tradition.

We included all non-Anglophone European-language journals in that database that met three criteria: (1) they had JSTOR records going back to at least 1999 and extending forward through at least 2010; (2) they publish at least approximately half of their articles in languages other than English; and (3) they are classified as philosophy journals on the PhilPapers journals list.[11]  We then accessed the most recently available issue of each of these journals in the JSTOR archive (as of Aug. 13, 2017) and examined the references of every research article in those issues, excluding reviews, discussion notes, editors’ introductions, etc.  This generated a total of 96 articles for examination, 41 in French, 23 in German, 14 in Italian, 8 in Portuguese, 6 in Spanish, and 4 in Polish.  Although this is not a systematic or proportionate sample of non-English European-language philosophy journal articles, we believe it is broad enough to provide a preliminary test of our hypotheses about insularity, asymmetry, and dominance.  Studies 5 and 6 will examine more systematic samples from Chinese and Spanish respectively.

Five of the journals specialize in history of philosophy.  This is possibly an overrepresentation of history of philosophy journals, and in any case a difference from the elite Anglophone journals none of which specialized in history.  Since history of philosophy journals might be expected to have different language citation practices, these five journals were flagged as such for analysis.

References were hand-coded from reference sections, footnotes, or in-text citations by two expert coders with PhDs in philosophy, each with reading skills in several European languages.  For each citation, we noted the language of the citing article, whether the cited source had originally been published in the same language as the citing source or in a different language, and if it was in a different language whether that language was English.  As in Study 1, sources in translation were coded based on the original language of publication rather than the language into which it had been translated (e.g., a translation of Aristotle into French was coded as ancient Greek rather than as French).  We also noted the original year of publication of the cited source, sorting into one of four categories: ancient to 1849, 1850-1945, 1946-1999, or 2000-2017.

Due to the multilingual nature of the coding, the often unsystematic citation patterns in journals using footnote format, and the often incomplete data about the original year and language of publication of translated works, we were concerned about the accuracy and reliability of coding.  For this reason, after the initial round of coding was complete, five articles by each of the coders (ten total) were randomly selected for independent recoding by the other coder, so that we could check the extent of coder agreement.

Results.  In all, we found 2883 citations across the 96 articles, 258 of which appeared in the ten articles selected for reliability testing.

In the articles selected for reliability testing, the coders agreed on both the original language and year-category of publication in 91% of cases (235/258).  Errors involved missing or double-counting some footnoted citations, typographical error, or mistakes in language or year category, and were corrected based on discussion between the coders.  The errors did not fall into any notable pattern, and in our view are within an acceptable rate given the difficulty of the coding task and the nature of our hypothesis, which is concerned with broad trends rather than exact numbers.

Of the 2883 citations included in our analysis, 44% (1270/2883) were to same-language sources, 30% (864/2883) were to sources originally written in English (some translated into the language of the citing article), and 26% (749/2883) were to all other languages combined.  Only 5% of articles (5/96) cited exclusively same-language sources.  French- and German-language articles showed more same-language citation than did articles in other languages (French 51% [565/1104]; German 71% [489/690]; average of all other languages 20% [216/1089]).  However, we interpret this result cautiously due to the small and possibly unrepresentative sample of articles in each language.

Table 3 shows the breakdown by historical period.  As is evident from the table, citations of very recent work are most likely to be citations of same-language sources, while citations of work published in the period from 1946-1999 are about equally divided between same-language sources and English-language sources.

 

Table 3: Original published language of cited work in recent non-Anglophone European-language journal articles in the JSTOR database, number of citations by year category

 

Original publication year of cited source

Cited source was originally published in same language as citing article

Cited source was originally published in English

Cited source was originally published in some other language

ancient to 1849

68 (14%)

29 (6%)

386 (80%)

1850-1945

98 (43%)

59 (26%)

72 (31%)

1946-1999

544 (43%)

514 (40%)

214 (17%)

2000-2017

560 (62%)

262 (29%)

77 (9%)

 

Perhaps surprisingly, citation of non-English vs English sources differed little if at all between history and non-history journals (27% [165/622] vs. 31% [699/2261], two-proportion z = -2.1, p = .03).  However, history journals were much more likely than non-history journals to cite work in foreign languages other than English: 48% vs. 20% (297/622 vs. 452/2261, two-proportion z = 14.0, p < .001).

Discussion.  In a convenience sample of non-Anglophone articles from established journals in several European languages, the results look very different than they do for the selection of articles from elite Anglophone journals that we examined in Study 1.  Whereas in the Anglophone articles, 97% of citations were to sources originally published in English, in this sample, only 44% of citations were to sources written in the same language as the citing article, suggesting much less linguistic insularity in the latter group.  The asymmetry hypothesis is also supported: Although in our sample of Anglophone articles virtually no 21st century non-Anglophone sources were cited (< 1%), in this sample of non-Anglophone articles, 29% of 21st-century sources are Anglophone.  Anglophone dominance is also supported for recent (post-War) work: Among citations of recent sources written in languages other than the language of the citing article, 73% (776/1067) are to sources originally written in English.

One notable finding is that English-language sources written in the period from 1946-1999 are proportionately more cited than are English-language sources written in the period from 2000-2017.  We see two possible explanations for this pattern.  One possibility is that English-language dominance peaked in the latter half of the 20th century and is now decreasing.  Another possibility – the likelier, we think – is that it’s a recency effect: When citing very recent work, authors are more likely to cite others in their linguistic tradition than they are to cite work in foreign languages, perhaps because it may take a bit longer to become aware of or gain access to foreign-language work or because other philosophers working in the same language may be more likely to be in their immediate academic circles of influence.  The declining-dominance hypothesis and the recency-effect hypothesis, though not incompatible, make different predictions about what citation patterns will look like in the future.  Study 6 will provide some evidence for a recency effect in Spanish-language citations.

 

6. Study 5: Citation Practices in Elite Chinese-Language Journals.

English is the third most commonly spoken native language in the world, after Chinese and Spanish.[12]  In Study 5, we examine citation practices in elite Chinese-language journals.  In Study 6, we examine citation practices in elite Spanish-language journals.

Method.  We examined citation patterns in fifteen elite Chinese-language journals, using a sample of articles in five-year intervals from 1996 to 2016.  The selected journals were Tier I philosophy journals as ranked by the Research Institute for the Humanities and Social Sciences Ministry of Science and Technology, Taiwan, (科技部人文社會科學研究中心, 2016) and the core philosophy journals as ranked in the Chinese Social Sciences Citation Index by the Institute for Chinese Social Sciences Research and Assessment, Nanjing University, China (中國社會科學研究評價中心, 2016).[13]  We sampled original research articles from each journal’s first issue in 1996, 2001, 2006, 2011, and 2016, generating a list of 208 articles for examination.  Three of the selected journals specialize in the history of Chinese philosophy and were flagged as such for analysis.

As in Studies 1 and 4, references were hand-coded from reference sections, footnotes, or in-text citation by an expert coder with a PhD in philosophy and reading knowledge of the target language.  For each citation, we noted the language in which the source had been originally published and whether it was cited in its original language or in translation.

Results.  In all, we found 2952 citations across the 208 articles.  Of these, the original language of publication was discoverable for 2929 (99%).  More citations were discovered in recent articles than in older articles, with articles from 1996 contributing 12% of the citations in our sample, articles from 2016 contributing 30%, and the other years intermediate between these two.

Among the 2929, 1507 (51%) were to Chinese-language sources, 915 (31%) were to English-language sources, and 507 (17%) were to sources in all other languages combined, including ancient Greek, Dutch, French, German, Italian, Japanese, Korean, Latin, Polish, Russian, Sanskrit, Spanish, and Tibetan.  (The third most cited language was German, at 9% of citations.)  Among the English-language sources, 260 (28%) were cited in Chinese translation.  The remainder were cited in their original English.

Journals specializing in the history of Chinese philosophy cited almost exclusively works that had originally been written in Chinese: 98% (860/882).  Excluding those journals from analysis, we found that the plurality of citations, 44%, (907/2047) were to works originally written in English, and only 32% (647/2047) were to works originally written in Chinese (and thus 24% [493/2047] for all other languages combined).  Ninety-two percent (152/168) of these articles cited a source from at least one language other than Chinese.

  We thought it worth seeing whether the English-language citations were mostly of classic historical philosophers like Locke, Hume, and Mill, or whether they instead were mostly of contemporary philosophers.  Although the sources’ unsystematic citation practices made it impractical to code the original year of publication of every cited article, we randomly selected 100 English-language sources for post-hoc examination of original publication date.   In our random sample of 100 cited English-language sources, 68 (68%) were published in the period from 1946-1999 and 19 (19%) were published in the period from 2000-2016.

Finally, we analyzed the results by year of publication of the citing article, excluding the three history journals.  Figure 1 displays the results.  Point-biserial correlation analysis shows a significant increase in rates of citation of English-language sources from 1996 to 2016 (34% to 49%, rpb = .11, p < .001).  Citation of both Chinese and other-language sources may also be decreasing (rpb = -.05, p = .03; rpb = -.08, p = .001), but we would interpret these trends cautiously due to the apparent U-shape of the curves and the possibility of article-level effects that would compromise the statistical independence of the trials (e.g., a single article with many references to ancient Greek sources).

 

Figure 1

Discussion.  Elite Chinese-language philosophy articles appear to cite from a variety of linguistic traditions, with 49% of citations being to sources originally written in languages other than Chinese.  The percentage is similar to what we found for non-English European languages in the JSTOR analysis of Study 5 and very different from what we found in our selection of elite Anglophone journals.  Remarkably, Chinese language journals specifically discussing Chinese history appear to cite Chinese sources at about the same rate (98%) as Anglophone journals cite Anglophone sources when discussing general philosophy (97%).  Among Chinese-language journals not specializing in history of Chinese philosophy, English sources are more commonly cited than Chinese sources, a trend that appears to be increasing over time.[14]

These results constitute strong evidence that articles in elite Chinese-language philosophy journals are not linguistically insular, that they are asymmetrically influenced by recent Anglophone work, and that English is the dominantly cited foreign language by a large margin.

 

7. Study 6: Citation Practices in Elite Spanish-Language Journals.

Method.  For our final study, we looked at citation patterns in twelve elite Spanish-language philosophy journals, examining all original research articles published in Spanish from each journal’s first issue in 1996, 2001, 2006, 2011, and 2016.[15]  We chose the top twelve philosophy journals as ranked by the SCImago Journal Rank, selecting nine among them that matched the suggestions of four faculty members in Spanish-speaking departments of philosophy that are highly regarded in the Spanish-speaking world.  We completed the list by replacing the three journals that had not been recommended by the faculty members with the three that had been most commonly recommended but were not indexed by SCImago (Análisis Filosófico, Contrastes, and Diánoia).  References were hand-coded by an expert coder with a PhD in philosophy and fluency in both Spanish and English.  For each citation, we noted the original language and publication year of the cited source.

Results.  We found 8421 citations across 312 Spanish-language research articles published in these twelve journals in the sampled years from 1996 to 2016.  More citations were discovered in recent articles than in older articles, with articles from 1996 contributing 14% of the citations in our sample, articles from 2016 contributing 28%, and the other years intermediate between these two.

In our sample overall, only 1671 (20%) were citations of work that had originally been written in Spanish.  English-language sources were cited 3712 times (44% of citations), and sources in other languages were cited 3038 times (36% of citations).  Cited languages included Arabic, Catalan, Chinese, Czech, Danish, Dutch, French, German, Greek, Hebrew, Hungarian, Italian, Latin, Pali, Persian, Polish, Portuguese, Russian, and Sanskrit.  (The third most cited language was, as in the Chinese data, German, at 17% of citations.)  Only 7 (2%) of the 312 articles cited only Spanish-language sources.  (Indeed, 70/312 [22%] cited only foreign-language sources.)

Table 4 shows the breakdown by historical period.  As is evident from the table, English is the dominant language of cited sources from the post-War period, drawing a narrow majority of all citations.  Similar to the English-language and JSTOR data, citation of sources that are neither in English nor in the language of the citing article declines sharply by historical period of the cited source.  Also similar to the JSTOR data, the greatest rate of same-language citation is for 21st-century sources.

 

Table 4: Original published language of cited work in elite Spanish-language philosophy journals 1996-2016, number of citations by year category

 

Original publication year of cited source

Cited source was originally published in Spanish

Cited source was originally published in English

Cited source was originally published in some other language

ancient to 1849

45 (5%)

145 (16%)

726 (79%)

1850-1945

130 (14%)

267 (28%)

542 (58%)

1946-1999

948 (20%)

2433 (51%)

1408 (29%)

2000-2016

548 (31%)

867 (49%)

362 (20%)

 

As with our Chinese-language database, we analyzed the data by year of publication of the citing article, to look for temporal trends.  The results are displayed in Figure 2.  Point-biserial correlation analysis shows a significant increase in rates of citation of English-language sources from 1996 to 2016 (36% to 45%, rpb = .06, p < .001), and a corresponding decrease in citation of Spanish language sources (25% to 14%, rpb = -.07, p < .001).  We detected no effect in citation rates of other languages over the time period (rpb = .00, p = .95).  Although we are concerned about article-level violations of statistical independence, we are somewhat more confident in these trends than in the Chinese-language data due to the somewhat larger number of source articles (312).

 

Figure 2.

 

Finally, we conducted a recency analysis to test the hypothesis that same-language citations are generally higher for recent citations than for older citations, regardless of the publication year of the citing article. Citation rates of Spanish-language sources were most common when the source was published less than 15 years previously than when the source was published 15-29 years previously (31% [970/3086] vs. 18% [348/1964], two-proportion z = 10.8, p < .001).  This trend held for all publication years of citing articles (2016: 25% vs. 12%, 2011: 30% vs. 23%; 2006: 37% vs. 24%; 2001: 31% vs. 12%; 1996: 38% vs. 19%).  In other words, citation of recent work was disproportionately same-language compared to citation of older work, in every year studied.

Discussion.  Our sample of recent articles from elite Spanish-language philosophy journals shows very low insularity, with the large majority of citations being to non-Spanish-language sources.  English is the dominant language among recently cited sources, and combining these results with the results of Study 1 supports the hypothesis that recent English-language philosophy has a highly asymmetric influence on recent Spanish-language philosophy.  We also found evidence of a recency effect, with same-language citation rates higher when citing recent sources than when citing sources at least fifteen years old.  Speculatively, such a recency effect might also be explain some of the patterns in the JSTOR and English-language data.

 

8. Conclusion.

Our research supports four main conclusions.

First, mainstream Anglophone philosophy is insular in the sense that mainstream Anglophone philosophers tend to mostly cite or interact with others participating in the same linguistic tradition.  We find this general result unsurprising.  However, we confess to being somewhat surprised by the magnitude of the result.  Fully 97% of all citations in our sample of elite Anglophone journals are to work originally written in English, leaving only 3% of citations for all other linguistic traditions combined.  This 3% includes all citations of Plato, Aristotle, Descartes, Kant, Nietzsche, Frege, Wittgenstein, and Foucault, and all Islamic and “non-Western” work, as well as all recent work by philosophers writing in other languages around the globe.  Similarly, 96% of editorial board members of elite Anglophone journals are housed at universities in majority-Anglophone countries.  And again similarly, among the one hundred most-cited recent philosophers in the Stanford Encyclopedia of Philosophy, only one spent the majority of his career in a non-Anglophone country, writing primarily in a language other than English.

Second, philosophy written in Chinese, Spanish, and a sample of other non-Anglophone European languages is not similarly insular, as measured by citation practices.   In all the non-English languages examined, foreign language citations constituted a substantial portion of citations.

Third, recent Anglophone philosophy has an asymmetric influence on philosophical work done in other languages.  Philosophers writing in Chinese, Spanish, and other languages frequently cite recent work written in English, but philosophers writing in English almost never cite recent work written in other languages.

Fourth, English is the dominant foreign language in foreign-language citations by philosophers writing in Chinese, Spanish, and other languages.  When philosophers writing in languages other than English cite recent (post-1945) work from outside their linguistic tradition, they are about twice as likely to cite a work originally written in English as they are to cite a work written in all other languages combined.

It is widely accepted that English is now the lingua franca of academic philosophy, and the language that one must write in if one seeks a broad international audience.  Our present analyses help quantify the extent to which this is the case.  We take no stand here on whether this is good or bad for global philosophy.  Our results should also help in framing questions about linguistic justice and linguistic issues in the epistemology of philosophy.  If Anglophone philosophy is as insular, asymmetrically influential, and dominant as it appears to be from our analyses, does that create unjust burdens on philosophers for whom English is not their native language?  Rather differently, does philosophy as a discipline suffer epistemically from having become as Anglocentric as it appears to be from our analyses?  We leave these questions for another time.[16]

References:

Chattopadhyay, Subrata, Catherine Myser, and Raymond de Vries (2013).  Bioethics and its gatekeepers: Does institutional racism exist in leading bioethics journals?  Journal of Bioethical Inquiry, 10, 7-9.

Gordin, Michael D. (2015).  Scientific Babel.  Chicago.  University of Chicago.

Healy, Kieran (2013).  Lewis and the women.  Blog post at kieranhealy.org (Jun. 19).  URL: https://kieranhealy.org/blog/archives/2013/06/19/lewis-and-the-women/ (accessed Sep. 1, 2017).

Hurtado, Guillermo (2013).  Filosofía analítica en lengua vernácula.  Crítica, 45, 107-110.

Leiter, Brian (2013).  Top philosophy journals, without regard to area.  Blog post at Leiter Reports (Jul. 3, rev. Jul. 6).  URL: http://leiterreports.typepad.com/blog/2013/07/top-philosophy-journals-without-regard-to-area.html.

Leiter, Brian (2014).  Departments ranked by SEP citations.  Blog post at Leiter Reports (Aug. 13).  URL: http://leiterreports.typepad.com/blog/2014/08/departments-ranked-by-sep-citations.html.

Leiter, Brian (2017).  Best Anglophone philosophers since 1957?  Blog post at Leiter Reports (Mar. 31).  URL: http://leiterreports.typepad.com/blog/2017/03/best-anglophone-philosophers-since-1957.html.

Perez, Diana I. (2013).  The will to communicate.  Crítica, 45, 91-97.

Rodriguez-Pereyra, Gonzalo (2013).  The language of publication of “analytic” philosophy.  Crítica, 45, 83-90.

Ronen, Shahar, Bruno Gonçalves, Kevin Z. Hu, Alessandro Vespignani, Steven Pinker, and César A. Hidalgo (2014).  Links that speak: The global language network and its association with global fame.  PNAS (Dec. 14), E5616-5622.

Ruffino, Marco (2013).  Some remarks about Gonzalo Rodriguez-Pereyra’s advice on the language of philosophy.  Crítica 45, 99-105.

Schwitzgebel, Eric (2010).  Most cited journals in the Stanford Encyclopedia of Philosophy.  Blog post at The Splintered Mind (May 10).  URL: http://schwitzsplinters.blogspot.com/2010/05/most-cited-journals-in-stanford.html.

Schwitzgebel, Eric (2012).  The ghettoization of Nietzsche.  Blog post at The Splintered Mind (Aug. 23).  URL: http://schwitzsplinters.blogspot.com/2012/08/the-ghettoization-of-nietzsche.html.

Schwitzgebel, Eric (2014a).  Citation patterns in the Stanford Encyclopedia of Philosophy, Part I.  Blog post at The Splintered Mind (Aug. 4).  URL: http://schwitzsplinters.blogspot.com/2014/08/citation-patterns-in-stanford.html.

Schwitzgebel, Eric (2014b).  SEP citation patterns: Further analysis and thoughts.  Blog post at The Splintered Mind (Aug. 14).  URL: http://schwitzsplinters.blogspot.com/2014/08/sep-citation-patterns-further-analysis.html.

Schwitzgebel, Eric (2014c). The 267 most-cited contemporary authors in the Stanford Encyclopedia of Philosophy. Blog post at The Splintered Mind (Aug. 7).  URL: http://schwitzsplinters.blogspot.com/2014/08/the-266-most-cited-contemporary-authors.html.

Siegel, Susanna (ed., 2014).  Reflexiones sobre el uso de ingles y el español en filosofía analítica / Reflections on the use of English and Spanish in analytic philosophy.  Informes de Observatorio / Obervatorio’s Reports, 006-12.  ISSN 2372-874X. 

Thomson Reuters’ ISI Web of Science (2007). Most cited authors of books in the humanities, 2007.  Times Higher Education (web news article, Mar. 26).  URL: https://www.timeshighereducation.com/news/most-cited-authors-of-books-in-the-humanities-2007/405956.article.

中國社會科學研究評價中心. (2016). CSSCI（2014-2015）来源期刊收录目录. 中國社會科學研究評價中心. Retrieved from http://cssrac.nju.edu.cn/a/zlxz/20160329/2720.html

科技部人文社會科學研究中心. (2016). 2016年「臺灣人文及社會科學期刊評比暨核心期刊收錄」評比結果暨核心期刊名單. 科技部人文社會科學研究中心. Retrieved from http://www.hss.ntu.edu.tw/model.aspx?no=354

梁春阳 (Liang Chunyang). (1989). 《哲学研究》引文分析. 图书馆理论与实践 (Library Theory and Practice), (03), 11–15+10.

 


[1] Related studies of the centrality of English in science and in cultural influence include Ronen, Gonçalves, Hu, Vespignani, Pinker, and Hidalgo 2014; and Gordin 2015.  On whether “analytic philosophy” ought to be written in English, see Hurtado 2013; Perez 2013; Rodriguez-Pereyra 2013; Ruffino 2013; Siegel (ed.) 2014.

[2] For some evidence that these measures converge, compare Schwitzgebel 2010, 2014a,b,c; Healy 2013; Leiter 2013, 2014; Brogaard and Leiter 2014.

[3] The full list in ranked order: Philosophical Review, Journal of Philosophy, Noûs, Mind, Philosophy & Phenomenological Research, Ethics, Philosophical Studies, Australasian Journal of Philosophy, Philosopher's Imprint, Analysis, Philosophical Quarterly, and Philosophy & Public Affairs.  The top four journals also match the top four journals analyzed in Healy 2013.

[4] The analysis was conducted in 2013, using the PhilPapers database as it existed then.  We believe this is recent enough to still be representative.  We downloaded the full list of journals and searched for language-specific pronouns, prepositions, or conjunctions.  When the journal’s language was unclear from the title, the journals were manually checked by looking at the languages of their most recent publications.  Given that the list runs to more than 1000 titles, a few misclassifications are likely, but note that the method described does not misclassify Noûs, Erkenntnis, etc., as foreign-language journals.

[5] This study was inspired by a similar study of the almost complete absence, from bioethics journals, of editorial board members from low-HDI (Human Development Index) countries: Chattopadhyay, Myser, and De Vries 2013.

[6] Included journals are those listed in note 3 plus Philosophy of Science, British Journal for the Philosophy of Science, and Synthese.

[7] We also noted the Philosophical Gourmet Report ranking of editorial board members’ universities, finding that 40% of editorial board members were housed in the seventeen “top 15”-ranked universities in the Anglophone world.

[8] English is one Singapore’s of four official languages.

[9] For example, a 2007 Thomson-Reuters ISI Web of Science list of “most cited authors of books in the humanities” (Thomson-Reuters 2007) features philosophers most of whose influence has been in other humanities or outside of the mainstream Anglophone tradition, with Foucault, Derrida, Habermas, Butler, and Deleuze topping the list.  A Google Scholar search for profiles in the topic of “Philosophy” has Derrida, Arendt, Rawls, Popper, and Žižek as the top five among 20th-21st century philosophers (accessed August 8, 2017, from Riverside, California).  Some convergent evidence of the SEP list’s surface plausibility comes from a Brian Leiter poll of “Best Anglophone philosophers since 1957?” (Leiter 2017), which has Quine, Kripke, Lewis, Rawls, and Putnam at the top, thus overlapping with our SEP measure in four of the top five positions.

[10] Disclosure: In this regard, the first author of this article appears to be a typical mainstream Anglophone philosopher in his own SEP entries on “Belief” and “Introspection”.

[11] Included journals (H for history journals) were Archives de Philosophie, Archiv für Rechts- und Sozialphilosophie, Crítica: Revista Hispanoamericana de Filosofía, Gregorianum, Jahrbuch für Recht und Ethik, Les Études Philosophiques, Revista Portuguesa de Filosofia, Revue de Métaphysique et de Morale, Revue de Philosophie Ancienne (H), Revue Internationale de Philosophie, Revue Philosophique de la France et de l'Étranger, Rivista di Filosofia Neo-Scolastica, Rivista di Storia della Filosofia (H), Roczniki Filozoficzne, Rue Descartes, Sartre Studies International (H), Studi Kantiani (H), and Studia Leibnitiana (H).  Three journals were excluded for not being on the PhilPapers journal list (https://philpapers.org/journals, accessed Aug. 13, 2017): Bruniana and Campanelliana, Esprit, and Méthexis.  One journal, Philosophische Rundschau, was excluded because it published only reviews.

[12] See https://www.ethnologue.com/statistics/size [accessed September 17, 2017].

[13] Included journals were 臺灣大學哲學論評 (National Taiwan University Philosophical Review), 政治大學哲學學報 (NCCU Philosophical Journal), 東吳哲學學報 (Soochow Journal of Philosophical Studies), and 哲学研究 (Philosophical Researches), 哲学动态 (Philosophical Trends), 自然辩证法研究 (Studies in Dialectics of Nature), 道德与文明 (Morality and Civilization), 世界哲学 (World Philosophy), 自然辩证法通讯 (Journal of Dialectics of Nature), 伦理学研究 (Studies in Ethics), 现代哲学 (Modern Philosophy), 周易研究 (Studies of Zhouyi; history journal), 孔子研究 (Confucius Studies; history journal), 中国哲学史 (History of Chinese Philosophy; history journal), and 科学技术哲学研究 (Studies in Philosophy of Science and Technology).

[14] To put this trend in a larger historical context, we can compare it with an earlier study on the citation pattern of one elite general philosophy journal, 哲学研究 (Philosophical Researches) (梁 (Liang) 1989). According to梁 (Liang) (1989), 90% of the citations in articles published between 1984 and 1988 are to sources written originally in or translated into Chinese. These data are not strictly comparable with our data, because they do not distinguish between the original and translated languages.  However, it does provide a floor number for that journal over that period of time.  At least 10% of citations in it were of sources originally written in foreign languages, and presumably considerably more than 10% once translated works are taken into account.

[15] Included journals were Anales del Seminario de Historia de la Filosofía, Análisis Filosófico, Contrastes, Crítica, Daimon, Diánoia, Ideas y Valores, Isegoría, Pensamiento, Teorema, Theoria, and Tópicos.  Articles published in languages other than Spanish in these journals were excluded from analysis.  In some cases, no Spanish-language articles appeared in a targeted journal issue for a particular year, in which case the journal contributed no data in that year.  For example, Teorema and Theoria had no qualifying articles in 2016.

[16] We would like to thank Wesley Fan, Tsu-Wei Hung, and Qiaoying Lu for their help with the information regarding the elite philosophy journals in China and Taiwan; four anonymous experts who helped with evaluating Spanish-language journals; commenters on our posts on these topics at The Splintered Mind and Eric Schwitzgebel’s public Facebook page; and the editors of this special issue.