1 A What? A Publication Strategy?

This might seem curious to many of our readers — especially the younger ones — but when I was a PhD student with Michel Crozier and Erhard Friedberg in the mid 1980s, nobody would have used the expression “publication strategy” or speak about it. In order to have a chance to get a position at the CNRS (largest French national research institution), it was advised to have at least one paper published in one of the main French sociological journals (Revue Française de Sociologie or Sociologie du Travail for instance), but there was no “publication strategy” to imagine and to adopt. It was also rather evident that when a research project or a PhD was over, a monograph should be published. On top of it, some papers could also be issued, but a monograph was the rule. Furthermore, writing with a colleague was possible but not so frequent and if you wanted to have a chance to get a position, solo papers or books should also appear in your résumé.

After I got a position at the CNRS, regularly publishing has always been expected for the every-two-year individual evaluation of my records by the CNRS. I tried to write papers or books and I edited some special issues, but I first of all followed my own preferences, or grasped an opportunity to publish, or accepted an invitation from a colleague to write a chapter in a book, with no further consideration. I also regularly published in English but not by pure strategy. My field of study was almost inexistent in France and therefore I have been very early engaged in international networks in order to find colleagues with whom to exchange on my research. Publishing in English was thus a necessity.

I happened to first see a publication strategy “in action” by the beginning of the 2000s, when I led a research on gender bias in the access of French academics to professorships, with a colleague in management studies. When we got the first results of our study, he opened his computer and said: “let’s see in which journal we publish them… I suggest this one. Its impact factor is very high.” It was of course a journal published in English. For him, writing a monograph for a French audience on a research led in France, was not an option. He clearly expressed it: “I do not need it for my CV.” And if he was to write a paper, it could only be for a journal with a high impact factor. He therefore had an access to a ranking of journals recognized in its discipline with their respective impact factors on his computer.

I am still not that far. When I think I have a good idea and decide to publish it I still do not choose a journal because of its impact factor. I know which journals are considered as relevant by my colleagues and I also know what kind of papers they publish by reading some of these papers. Depending on the idea and on the material (qualitative or quantitative data for instance) I have, I will first send it to a journal for which I think it may be of interest but I will also read the recent publications of this journal to check whether it could fit with it and in order to cope with the style of the journal when I start writing. As one can see, the relative lack of concern for a publication strategy that characterized the first part of my career is not possible any longer. Writing a paper is somewhat like hiring: one has to adapt one’s paper (candidature) to the style (the profile) of the journal (the position). Furthermore, because the administrative responsibilities I took over during the last decade left me less time, but also because the requirements for recognition in the world and in France have become more precise, more formalized and more standardized, I have become more selective in my publication choices. I less and less agree publishing chapters unless there is a strong editorial project for the edited book or if the editors are well known scholars in the field or in the discipline, and I refrain from accepting to publish papers in non peer-reviewed journals or in professional journals.

This being said, many choices still have to be made and even if it is not a very calculated and long and well thought out decision, I intentionally try to manage the following alternatives in a quite balance way.

2 Papers or Books? Papers and Books!

Only in books can the complexity and the systemic character of social facts be tackled. It is almost impossible in papers and it is more and more impossible as journal reviewers expect papers to be focused on one idea, one mechanism, one issue. Therefore no paper can replace a book. I always warn PhD candidates to resist the temptation of trying to write a paper summarizing their thesis. It never works and can never be published. Either they transform their PhD into a book or they slice it into different papers. But a PhD is (hopefully) too rich, too complex to be summed up into one paper.

I nevertheless do not consider that books and papers are contradictory. They are complementary. Books tell the whole story, gives room for details and above all, allows for linking mechanisms together. When I for instance wrote the French version of the book translated as The Market for Academics (2009), it was important for me to be able to put in the same piece the analysis of the different phases of academic hiring: decision to (re)open a position, definition of the position’s profile, selection of the candidates and academic judgment, negotiation of the “price” of the selected candidate. And then, to show how these phases are more or less articulated and to insert this specific moment of hiring within a wider understanding of academic labor markets in the three countries under study in the last chapter. But a series of papers on related issues preceded and followed this book on more specific aspects and more focused arguments: academic labor markets as an economy of quality; comparative analysis of European academic labor markets; the paradox of pricing in higher education; a comparison between the market for musicians and the market for academics…

Some could say that I am the only one seeing the coherence and interactions between all these publications as I am the only one who read them all; that only very few people if any, read the whole book and that most readers focused on the chapter(s) they were interested in and did not look at the others, thus losing the systemic character I mentioned above. This is probably true but I think the book and the papers would have been very different if I had not been able to develop this broader perspective. And for that very reason, I am not in favor of article-based PhD. The ones I read never completely convinced me and I mostly found the synthetic introduction and conclusion trying to bring the different papers together rather poor. And it can hardly be different. If you ever transform a chapter of a book into a paper or reciprocally, you probably know that it means quite a lot of work for the chapter to be detached from the logic of the book and to become a piece in itself, or for the paper to be integrated in the logic of the book and to become part of it. Papers stand alone and can not be easily be articulated one with another. Or if they are, it means that they are too close and too similar.

I know this position is not shared by some colleagues of the discipline but I really think that we will lose the richness and the complexity of the content and of social facts if we were just to publish papers. Some years ago, a mainstream economist explained to me that he was not interested in the results of the papers he reads but first of all in the beauty, the robustness and the novelty of the method leading to the results. I really hope that sociology will not be reached by such rationales and that publications in our discipline should first provide new evidence and analysis.

But the capacity to still publish book is also linked to funding issues. Publishing books based on a research program implies that ambitious multi-sided research programs are funded and not, or not only, one-question based research projects. It also implies that one does not run from one funded project to another but has the time to seat, to think and find a way to describe and analyze in a relevant and understandable way the complexity of what one been studied. Conditions that are not always provided nowadays.

3 English or French? English and French!

It is a crucial issue when you work in a non-Anglophone country, even if today almost all of my colleagues read English. My concern is about managing two different goals.

A first goal consists in being read by as many colleagues as possible and the only chance to reach that is to be published in English. And in a journal that is not invisible. We are also pushed in that direction by the objectives set by all higher education institutions nowadays. Some offer specific funding for the translation or for the editing of texts in order to encourage the publication of papers in non-French journals. Having published in a respectful non-French journal has almost become a condition to get a CNRS position nowadays. It was not the case when I applied thirty years ago.

It is revealing to observe that non-francophone journals are called “international” journals. As if, French journals were only read by French people and had no international outreach. As if not being written in French was synonym of quality. As if Anglophone journals were “international” by definition. These three assumptions are taken for granted in my home country but all can be contested. As for the last one, some years ago, I collected data on the institutions of the authors of all the papers published between 2001 and 2010 in the American Journal of Sociology (AJS), American Sociological Review (ASR), American Journal of Political Science (AJPS) et American Political Science Review (APSR). These four journals are considered in France as top international journals and publishing in one of them is highly regarded in CVs. But I observed that the percentage of papers written by authors affiliated to US universities reaches more than 82% in all of them: it is even almost 90% for AJPS (89,5%). The percentage of papers with at least one author affiliated to a US university reaches 92% for AJS and 94% for ASR but almost 95% for AJPS and et over 95% for APSR.

This is not the case in top field journals — generally less prestigious than the former ones — as they are more often entrusted to non-US-based editors in chief. It is also different for journals that have been created in Europe or that are linked to a discipline that is more internationalized (economics or management for instance) than sociology or political science. But the international character and the assumed quality of journals published in English should be more often questioned and at least checked out by evaluators.

But even if we are more and more encouraged to publish in English, I maintain another strategy related to my second goal. I think it is important to maintain the French tradition of writing sociology, i.e. writing a paper that tells a story in order to demonstrate an idea (and sometimes two); writing papers for which each outline is different and deviates from the “state of the art — method — results — discussion” standard of many “international” journals; writing papers with long sentences; writing papers with some paragraphs that are not completely useful for the demonstration but introduce nuances and make detours; and also writing more literary papers where my vocabulary will be richer and more precise because it is my native language.

I generally choose French when I write more empirical papers, based on French data only, or when I think that the argument of the paper would be difficult to “sell” to a non-francophone audience (for instance if it is related to a rather French debate, atypically French institution, or to recent French publications). I also choose French if I have not published in French on a specific research project. In the recent years, I limited the publication of papers in French to what I considered as the top journals sociology or political science in France. It is also what I advise the PhD candidates I supervise and suggested my colleagues of Sciences Po to do, when I was the vice-president for research of this university specialized in humanities and social sciences. “Publishing in French is fine, but it must be in the most respected journals of our disciplines.” I nevertheless never convinced my colleagues of the department of economics to publish in French…

4 Alone or with Colleagues? Alone and with Colleagues

As explained at the beginning of the paper, co-authorship was not always well considered in the 1980s. It is much more encouraged nowadays, especially if one publishes with authors located outside France. The opportunity for co-authorship is also more frequent than it was before. This is linked to the fact that French and European research funding schemes promote networking and collaborative research. Papers with more than two authors have become more frequent. But this kind of collaboration is easier with some disciplines than with others. In a research project led with an economist and a colleague in management sciences, we accepted that the economist publishes alone in economics journals in order for her work to be recognized by her peers. She nevertheless also publishes a paper in a journal of social sciences with my colleague in management sciences but she knew it will not be considered by her peers as a “normal” publication, i.e. a publication to be taken into account.

Furthermore, even if it has become more frequent and if I appreciated and found very productive an rich the collaborations I experienced recently or in the past, I still like developing my own ideas by myself and being confronted with myself and my computer.

Co-authorship also raise concerns about how to proceed to individual evaluation. Because it is difficult to assess the autonomy of a colleague who only publishes with co-authors and to identify his or her individual contribution, co-authorships might lead to relying on the reputation of the journals or reputation of the co-authors rather than on the contribution of the papers themselves. For that very reason, individual publications should still be encouraged and required.

5 Open Science or Private Editors? Not an Issue yet… But

In a final point, I would like to stress that open science issues are becoming more and more crucial. France contributed to the European “Plan S” and the mobile barrier has been reduced to one year for French private editors according to the 2016 law for a digital Republic.

Up to know I never chose to publish here or there in relation to open science policy. I never accepted to pay for my paper to become freely available when it is published by a private editor, but such questions will probably become more and more accurate in a near future. The free access to the data on which the paper relies will also become an issue. This might impact the publication strategy of all of us in a very near future.