Less Citation, Less Dissemination: The Case of French Psychoanalysis

We contribute to inquiries about the visibility of globalized psychoanalytic research in the digital era (cf. Stepansky 2009) by adopting a comparative perspective on a specific geographic area of historical importance for psychoanalysis: France. The largely digital globalized psychoanalytic research field relies on standard bibliometric measures of journal quality (Impact Factor, SJR, etc.), which depend on the number and type of academic cites received by a journal. Thus, citing shapes academic publishing space by differentially valuing its component journals. Conversely, not to cite practically means not to engage with the field. Hence, we took citedness rate as a proxy for global visibility. By drawing on an original database created by one of us, we determined the global citational visibility of French vs. Anglo- American psychoanalytic productions (respective global outreach); and we related it to a first look at French vs. Anglo-American citation practices (geographic breakdown of article cites). We found that, on a 15-year period, the global outreach (citedness rate) of French articles is ten times smaller than that of Anglo- American articles; and that French articles are cited in Anglo-American journals five times more than Anglo-American articles in French journals – which in turn don’t seem to cite their French peers very often. These specific French citation practices could be explained by the implicit modes of reference at work in clinical settings shaped by rich theoretical and clinical local legacies. We conclude by considering that this situation presents French psychoanalytic research with a formidable opportunity for increased citational visibility.


INTRODUCTION
The future of all publishing is open to question, and this is especially true in the case of psychoanalytic publishing. Stepansky (2009) has explored the future of psychoanalytic publishing with a particular emphasis upon how the digital era has had an impact upon the decline of scholarly publication in the United States. If this trend continues, the survival of contemporary psychoanalytic research will depend upon its capacity to embrace and utilize digital publishing.
Echoing this perspective, we tried to determine whether the seemingly small international visibility of contemporary French psychoanalytic research could be related to its lack of acknowledgment of the impact of digitalization on rules of writing and publishing.
We believe that achieving visibility doesn't chiefly depend on overcoming a language barrier (not such an issue for younger generations), cultural differences, or geographical distance (made irrelevant by the Internet). Rather, our intuition is that French psychoanalytic work would become more visible if it demonstrated familiarity with psychoanalytic work in English, citing it and engaging in dialogue.
Anglo-American psychoanalytic publications follow a specific rule of digitalized, databaseanchored research. This rule is that the bibliometric value of a journal, which largely conditions its academic visibility and value, depends on the algorithmic measure of the number and type of cites that it receives from published articles published. (This is the case, in different ways, with the commonest algorithms: Thomson's Impact Factor and SCImago's Journal and Country Rank.) In other words, the algorithmic measure of this value ultimately depends on whether and how researchers cite a given journal-these citation practices will then, in turn, make it a more desirable publishing venue. The bottom line is that citations have a real effect, as they directly contributing to the space of academic publishing by differentiating journals both numerically and hierarchically.
It is plausible to imagine that an increase in the number of Anglo-American journal citations in French journals would be likely to produce a rise in international visibility for French journals. Of course, this is speculative and cannot (yet) be put to test. Let us begin with a preliminary effort: to draw on an original database in order (a) to determine whether French psychoanalytic productions are visible internationally, and (b) to clarify the citation practices comparatively between the French and Anglo-American contexts. To the best of our knowledge, this issue has not been addressed so far; in providing preliminary data and reflections, our hope is modest: to inspire discussion and debate.
To that end, we first measure the citational visibility of French vs. Anglo-American psychoanalytic journals across all disciplines and languages (their respective global outreach); we then relate this outreach measure to a geographic breakdown of psychoanalytic article citations, in order to make sense of specific geographical differences in citation practices.

METHODS
One of us (RP) put together a dedicated database, which is the first result of this research. This database draws on the data available in Elsevier's Scopus database, which indexes the highest number of Social Science and Humanities journals.
The first step was to establish a corpus of psychoanalytic journals, selected through content analysis. All of these journals refer to psychoanalysis either exclusively or significantly (by publishing psychoanalysis-oriented papers relevant to the fields of psychiatry and/or psychology). They have similar editorial processes and publishing circuits (e.g. non-open access); and their language is that of the publishing body's country (they are not bilingual). Being indexed in Scopus, they all have an SJR metrics. The list of selected journals is the following: RP has collected the whole bibliometric data of this corpus of 33 journals (13 + 20) over the last 15 publication years (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015).
The database variables are, for each journal over this period: SJR per year, number of article citations per year, number of cited articles, number of published citable articles per year, number of non-cited articles per year, country of origin of cites.
For Table 1, we started from the number of documents, the number of citations and the number of non-cited documents, and just calculated the percentage.
For Table 2, starting from the number of articles that cite each journal (with their geographic origin), we added the numbers and then calculated the percentage.

RESULTS
All of the following results are to be understood over a 15-year period (2000-2015).

Global Outreach of French vs. Anglo-American Psychoanalytic Articles
Are French psychoanalytic journals almost absent from the international research scene? A first step toward answering this question is to assess the global outreach of French vs. Anglo-American articles, by determining the ratio of citations per psychoanalytic article depending on its geographic origin (French vs. Anglo-American).
Aiming at a global outreach measure, we didn't differentiate between origin (French vs. Anglo-American vs. other) or disciplinary affiliation of citing journals.
For this Table 2 cites referring to one and the same article in the same document count as 2; and self-cites are included.
The results are telling: across all publication supports and languages, each French psychoanalytic article is cited about only once (1,14 %), while for each Anglo-American article, the ratio is of 10,6% -that is, about 10 times higher.
On the other hand, of the 11370 articles published in French psychoanalytic journals over the last 15 years, 70,52% haven't been cited in any support, while this figure drops to an average of 41,995% for the 12,540 articles published in Anglo-American journals. In other words, looking back today over the last 15 years, the global rate of non-citedness of Anglo-American psychoanalytic articles is almost twice as small as that of French articles.

Geographic Breakdown of Psychoanalytic Articles Cites across All Disciplines
We then sought to establish an overview of the geographic breakdown of psychoanalytic articles citations in French vs. Anglo-American journals of all disciplines.
In the following table, the unit isn't the citation, but the article: two citations of one and the same article count as 1, if the citation is in the same article, and as 2 if in two different articles.
Out of the 6812 total cites referring to articles published in French psychoanalytic journals, two thirds (66,23%) come from French journals (whatever their discipline), and only 5,66% from Anglo-American journals. On the other hand, out of the 28,204 total cites referring to articles published in Anglo-American psychoanalytic journals, 63,08% (17,791) come from Anglo-American journals, and 1,20% (339) from French Journals.

DISCUSSION
These data demonstrate that the global outreach of French psychoanalytic articles is 10 times smaller than that of Anglo-American ones (Table 1) and that, unsurprisingly, such articles are mostly referred to in French journals, wherein references to Anglo-American ones are almost nonexistent ( Table 2).

Limitations of the Current Data
One might argue that, since the considered time range is 2000-2015, the language barrier could partly account for the difference in outreach ratio, and for the virtual absence of Anglo-American citations in French journals-most of which are psychoanalytic. For in France, only recently has the language barrier genuinely ceased to become less of an issue. Nevertheless, the difference in outreach ratio is so striking, and the references to Anglo-American work in French journals so scarce, that it is doubtful whether language barrier could account for it. Also, even though Scopus is the best available database for psychoanalytic journals, it doesn't-yet-index certain French psychoanalytic journals (such as Research in Psychoanalysis or Adolescence) which have quite an important readership amongst clinicians.
Most importantly, both outreach measures and geographic breakdown of citations are not, in the present article, restricted to psychoanalytic journals: they include all research and languages (Table 1). Indeed, when taking into account geographic breakdown ( Table 2), they do not differentiate among the disciplines of the citing journals. Clearly, this would be important to examine in future studies.
We believe that, in light of the quantity of further database work required to reach such detail and accuracy and the absence of papers addressing the current topic, it makes sense to produce a few preliminary figures, in order to stimulate discussion and continuing research. For example, it would be important to look at outreach within different languages or disciplines; and specifically at the geographic breakdown of psychoanalytic citations.
The figures provided in the present paper constitute an intriguing proxy as to both the international outreach of French psychoanalytic research, and the specific citation practices at work in the French psychoanalytic community. French psychoanalytic work has quite a small outreach; and when cited, it is mostly in French journals-most of which are, most likely, psychoanalytic. Therefore, in view of the outreach ratio, it is likely that French psychoanalytic researchers (who arguably constitute most of the readership of French psychoanalytic work) do not cite their peers very often. And since we know ( Table 2) that they almost never cite Anglo-American journals, it is dubious that they are following standard citation practices.In order to prove this, it would have to be studied more systematically. Our figures constitute a tempting first step in this direction.
Finally, drawing a global picture (a long-term goal, in terms of database construction) would require to integrate journals from other places, especially from Germany and South America.

Explanatory Considerations: A License to Cite?
French citation practice can be traced to context. French psychoanalytic thought has always partly taken place outside of the academic world, thereby following rules different from standard academic discourse, for at least two reasons.
First, analysts appointed to full professorships in psychoanalytically-oriented psychology (D. Lagache, J. Laplanche) had already gained prominence in analytic schools. Second, becoming a licensed clinical psychologist in France doesn't require a full course of graduate study: it is a 5-year Masters-level training, which aims at training psychologists for clinical work in non-academic institutions. Thus, psychoanalytic thought taught in such curricula mostly feeds clinicians who are not very concerned with academic research.
This relative academic extraterritoriality of French psychoanalytic thought has an impact on its visibility. A shared legacy inherited by the next generations, it is mostly based upon the elaboration of local clinical practices: mental care institutions, supervision groups in psychoanalytic schools, and the training of licensed psychologists. In these contexts, this legacy is often already shared by clinicians; and when it isn't (as in universities, where students become acquainted with it), it is referred to in a practice-oriented, citation-free manner. Therefore, French psychoanalytic publications often refer to this shared legacy in an implicit manner, at odds with the explicit citation conventions that are standard practice in contemporary academy.
Perhaps, French psychoanalytic researchers have imposed restrictions upon themselves, inhibited about switching from their inherited, implicit mode of referring to an explicit mode of acknowledgment through citation 1 . They would, so to speak, lack the license to cite. This could account for the discrepancy between French and Anglo-American citations in French psychoanalytic journals ( Table 2): the extreme scarcity of Anglo-American journal citations could be understood as the effect of an injunction to remain faithful to a national heritage.

CONCLUSION
The current lack of international outreach should be viewed as an opportunity. French psychoanalysts should not be afraid to lose their singularity, as in the narcissism of minor differences (Freud, 1921(Freud, , 1930Gabbard, 1993): engage with fellow international researchers, which will help to clarify their clinical and theoretical originality.
However, such engagement would entail a precondition concerning research. Articles should be written not for peers who share an implicit network of references, but for an international community with diverse, local legacies, with the ultimate aim of promoting an exchange of perspectives. Aspiring for international exchanges requires that one takes recent work into greater account. While this focus might rely on the short timespan of algorithmic measures of citation rate, and therefore has questionable epistemic justification, it can also stir the field's vitality. Even though citation practices do differ among disciplines (as between, say, biology and psychoanalysis), the rate of citations of recent productions could count as an interesting proxy for field vitality. The experience of one of the authors (EJ) is telling: he is the editor of the psychoanalytic journal that has the highest impact factor over the last 5 years, and has found that the increase in impact factor, has affected both the quality and the quantity of submissions in a positive way 2 .
Finally, both French and English-speaking researchers and journals could benefit from an increase in citational acknowledgment, which could help unite the field around common, trans-national challenges. For example, an interesting-if complex-avenue for psychoanalytic researchers would be to try and expand bibliometric quality parameters. Since psychoanalytic research primarily aims at developing therapeutic theories and practices (a distinctive status within the social science), an interesting task would be to try and devise a social value metrics, reflecting the social value of psychoanalytic research. Such a goal, which might well complement the proliferating research on the assessment of therapeutic change, would certainly require an international collaboration.

AUTHOR CONTRIBUTIONS
RP and OP are equal contributors to this paper (co-first authors). RP has put together the original database on the basis of Scopus, and has calculated the results. He has contributed to writing the paper (structure of the argument); he has co-constructed the interpretation of the results and of their limits, and their discussion. OP has contributed to the structure of the argument, and has written the detailed version of the paper. He has coconstructed the interpretation of the results and of their limits, and their discussion. CD has contributed to the design of the original database, to calculating the results and to their statistical interpretation. EJ has contributed to writing the paper (structure of the argument, references). All authors read and approved the final manuscript.