Abstract
Science policies and science studies largely share an understanding of scientific knowledge and objects as immutable mobiles. This article shows how the analysis of research assessment in a non-Anglophone country and its effects on social sciences can shed new light on this shared notion. The preference for immutable mobiles in assessment regimes pushes social scientists to publish in specialized, usually Anglophone journals, which can result in the attenuation of local relevance of the knowledge they produce and contribute to the performance of globally converging societies. The author argues that the observed consequences for the social sciences in non-Anglophone countries underscore a larger problem with both the policy ideal and the science and technology studies (STS) idea of immutable mobiles on two counts: the relation of the social and natural sciences to society and the engagement of sciences with the multiplicity of societies as well as natures.
“In funding basic research [we have to] take the citation count as a criterion of success, mainly on the international scale. In funding applied research [we have to] prefer those who are able to convert outcomes into innovations fast (technology transfer). Outcomes of basic research which do not permeate into the ‘global market’ of science are in a similar position as innovations that have already been invented and applied by someone else.” These were remarks made by the then Prime Minister of the Czech Republic (CR) on his government’s funding priorities (Topolánek 2007, 18; original emphasis). It is not only in the CR that politicians and policy makers are intensely focused on global competitiveness, excellence in research and its measurement, and turning knowledge into innovations. While issues of quality and making knowledge travel had been an important feature of scientific activity before they became the focus of policy, contemporary science policies have reshaped them toward new effects.
For the last twenty years, science and technology studies (STS) have critically investigated the rise of entrepreneurialism as the organizing principle of academia (e.g., Law 1994; Slaughter and Leslie 1997) and the introduction and effects of audit and assessment procedures on researchers’ subjectivities, careers, and publication practices (e.g., Velody 1999; Strathern 2000; Brew and Lucas 2009; Felt 2009). Crossing both national boundaries and clearly delimited areas of academic institutions and research, a number of studies situate these changes explicitly in the wider context of neoliberal transformations of policy and society (e.g., two recent special issues of Social studies of science [Lave, Mirowski, and Randalls 2010] and Social anthropology [Wright and Raba 2010]).
Some of these authors see these changes as effects of pressures coming from outside the academy. For example, Lave, Mirowski, and Randalls (2010, 659) call for “a detailed exploration of exactly how the external political–economic forces of neoliberalism are transforming technoscience.” Others, however, argue that treating neoliberalism as an external force risks not only analytical reductionism but also a paralysis of the political imagination and grounds for action as we “bestow an undeserved systematicity upon current trends and relations” (Boyer 2010, 80). Approaches such as these are more sensitive to the complex orderings of academic science itself and argue that science’s own emphasis on individual performance and dedication has further enabled the embedding of the new public management in academia, rather than providing grounds for resistance (Shore 2008). This is the position I take in this article. While aware of wider contexts and the operation of “external forces,” I think it is equally important to investigate how forces and ideas within science—and within science studies—have played their part in these developments. Focusing on STS, I will argue that STS themselves may have contributed, even if unintentionally, to the current climate of audit and entrepreneurialism in science and science policy. I make this point not to criticize STS, but to increase the reflexivity of science studies scholars regarding our own entanglements in and effects on science–society relationships. The starting point for this article, then, is the observation of similarities between science policy, STS, and science’s self-understandings insofar as they all emphasize that in order to be valuable—indeed, in order to be “scientific”—knowledge must be highly mobile and globally effective. These qualities are captured in Latour’s notion of “immutable mobiles” (1987), which has become such a central concept in contemporary science studies that it is now arguably used implicitly more often than it is systematically explained.
I argue here, however, that while providing a description of how scientific facts are produced and recognized, and their networks extended, Latour’s idea simultaneously risks reinforcing an ideal of mobility and unchangeability that is shared with the self-image of scientists and increasingly underwrites the logics of contemporary science policies. In science’s official “common sense” and in the philosophy of science, mobility, and immutability are treated as the necessary results of particular epistemological and methodological procedures. In STS, they have been seen as highly contingent outcomes of painstakingly constructed networks. By emphasizing immutability and mobility, both discourses have privileged the capacity of knowledge claims to travel intact, and neglected the necessity of attending to knowledge claims that maintain strong links with their local contexts of production, and/or which undergo significant changes in the course of their travels.
Mobility and immutability are also taken for granted in science policy, and particularly in research assessment, as hallmarks of scientific quality and value. Aspects of these assumptions have been criticized from within STS. When journal editors working in history of science, technology, and medicine (including those of Science, Technology, & Human Values) objected to the creation of the European Reference Index for the Humanities (ERIH) in 2009, they argued that “[g]reat research may be published anywhere and in any language,” adding that the ERIH “confuses internationality with quality in a way that is particularly prejudicial to specialist and non-English language journals” (Journals under threat 2009). However, there has been relatively little consideration of where these assumptions come from, or reflection on the entanglement of the STS’s own conceptual toolbox with the assumptions of research assessment highlighted in this critique.
In what follows I argue that a shared imagination of globally mobile knowledge is linked to the fact that the instantiation of science for which policies have been developed and on which the vast majority of STS studies have been conducted has been located in Western Europe and the United States (most often in Anglophone contexts), and in laboratory or lab-related natural science. In this article I inquire into science that falls outside these locations. Specifically, I am concerned with research assessment in the CR and focus primarily on its enactments in and repercussions for the social sciences. Looking closely at locations—geopolitical and disciplinary—where knowledge refuses to live up to the ideal of immutable mobility, or where aspiring to its performance involves reconstituting the conditions of epistemic production, reveals the limits of immutable mobility both as an ideal for research evaluation and as a figure for describing the characteristics of truth claims in STS. While I return to Latour’s concept throughout the article, I also argue for the need to critically reevaluate the centrality of immutable mobiles as a point of orientation in science studies. 1
Immutable Mobiles
In this article I talk about scientific “knowledge-objects.” This compound noun and its ambiguity are important for my argument because it permanently destabilizes two distinctions that characterize science policies. First, it troubles the distinction between knowledge and objects of knowledge imagined as preexisting, independent realities out there (Law 2004). Second, it questions the difference between knowledge as textual, discursive, and “on paper,” and the application of knowledge imagined as material and most often technological. These distinctions prevent the study of science as a performative endeavor, and produce a boundary between the natural sciences and technical disciplines on one hand and social sciences and humanities (SSH) on the other, especially regarding “applied” results of research.
What makes scientific knowledge-objects special? In Science in Action (1987), Latour’s influential proposal is that the specificity of science is not a question of its inherent rationality versus the irrationality of other regimes of knowing, but stems from the way in which scientific knowledge-objects are constructed and reconstructed, circulated and used, within a special infrastructure of discursively, technologically, and materially standardized practices. Latour identified two key characteristics of scientific knowledge-objects constructed in this way, mobility and immutability, and speaks about “immutable mobiles.” Mobility and immutability are effects of tight scientific networks that allow for movements of knowledge-objects without their mutation. Changes have to take the form of carefully controlled and traceable “transformations” (i.e., changes in the “form”) so that the chain of reference remains accountable (Latour 1995).
The price paid for immutable mobiles is not only the substantial cost associated with building and maintaining specialized networks but also exclusivity in social terms. One has to be a specialist to operate properly in the networks. Immutability presupposes a framework in which transformations of knowledge-objects are enacted and recognized as legitimate. What happens when knowledge-objects move out of the lab or beyond technical documentation? One possibility is that they perish. They die like lab mice unable to survive when released into nonlab environment (Haraway 1997), and mobility is sacrificed for immutability. The second possibility is the opposite. Scientific knowledge-objects are taken up by other networks and actors, and thus changed to fit their agendas. Immutability is sacrificed for increased and heterogeneous mobility. The third option, nonexclusionary with regard to the former two, is that the world outside the lab is included in or translated into a scientific network, as in Latour’s account of the “pasteurization of France” (1988). Similar arguments, with a more critical flavor, have been recently made in relation to genetically modified (GM) crops: the regulation of their release into the environment in the European Union has, in fact, transformed European agriculture into a massive experimental site (Levidow and Carr 2007).
The three options sketched above take into account possible changes of scientific knowledge-objects on the move out of the lab, but they leave largely intact the idea of science itself being constituted by immutable mobility. There are, however, authors, such as Mol and Law, who argue that the network topology of laboratory science characterized by immutable mobiles has preoccupied the STS imagination too much, and we should look for other types of knowledge-objects and their enactments (Law and Mol 2001). They have developed alternative topological accounts for medicine (Mol 2002; Law and Singleton 2006) and innovation (DeLaet and Mol 2000). I believe that this perspective must be turned back to science as practiced in academic institutions. While laboratory studies have always opposed the view of “pure science,” and studied entanglements of nature cultures and science societies, their approach to networks has been science-centric. In key actor–network accounts, it is scientists who enroll heterogeneous actors in their networks (for the case of Pasteur see Latour 1988), allowing the logic of immutable mobility to dominate.
However, it can be the other way around. knowledge-objects at the “heart” of science can be enrolled by other actors: scientific papers can be enacted as pieces of public relations (PR) for (potential) pharmaceutical products (Sismondo 2009) or they can be enacted as witnesses of scientific productivity for purposes of research assessments (see e.g., discussions about the anthropological monograph in the recent issue of Social Anthropology 2009). It is significant that Latour’s idea of immutable mobiles was developed within studies focused primarily on what happens to knowledge-objects within the extended lab network. But is it possible to understand the life of scientific knowledge-objects beyond the lab in terms other than gradual deconsolidation or descientization? To use Latour’s metaphor picturing scientific facts as trains that do not work “off the rails” (Latour 1983, 155), are knowledge-objects that do not embody immutable mobility necessarily “derailed?” The dilemma—for both STS and science policies—underlying this issue is whether we define science strictly by the network logic of immutable mobility and “hardened facts” (Latour 1987, 208-209) or whether we leave the definition of science more open and look for different topologies involved in it. While Gieryn (2006) argues against the uniqueness of the lab as a truth spot of scientific knowledge production and analyzes the “field site” as a specific and complementary source of credibility, he does not address the relation of credibility and value to the publication process and the life of the knowledge-objects after publication. I believe that this relation has to be addressed, and research outside Anglophone countries and in the social sciences presents a rewarding opportunity to do so.
In the following I first look at the new assessment criteria of Czech science policy and examine how they construct and enforce a narrow vision of immutability and mobility, with a particular emphasis on how the SSH are imagined to fit (or not) these models. I go on to explore the situation within research institutions in the social sciences as they respond to the assessment, revealing convictions about the nature of epistemic production in their disciplines. I argue that some versions of social sciences question the dominant vision of immutable mobility, and should be taken seriously as a challenge to both science policies and science studies.
Get Global, Get Patented: Research Assessment Criteria
I will now focus on research assessment criteria introduced in the CR in 2004 by the governmental Council for Research and Development (recently renamed as the Council for Research, Development and Innovations). Performed annually, the assessment is purely quantitative. Selected types of knowledge-objects are recognized as “research outputs” and assigned a numerical value (see the table below); each annual assessment is based on results achieved in five previous years. The units of evaluation are eleven providers of research funding from the public budget, including grant agencies and ministries. However, the providers tend–and in fact are forced–to apply the same criteria to research institutions and projects they fund; and at least some academic institutions apply the same criteria to research teams and individuals through internal assessment procedures. The logic has thus percolated through academia. All these processes were reinforced and became particularly significant in 2009 when a new law on financial support for research and development from public budgets (Act No. 130/2002, as amended) came into effect establishing a tight link between research assessment results and institutional funding (Table 1 ).
Criteria of Research Evaluation in the CR for Outcomes Registered since 2008 (The Office of the Government of the CR 2009)
Note: aIncludes philosophy and religion studies; history, archaeology, and anthropology; political studies; management; philology; media studies; art; and pedagogy.
bPublication listed in following databases at Web of Science (Thomson Reuters): Science Citation Index Expanded (SCI-EXPANDED)—1945 to present; Social Science Citation Index (SSCI)—1980 to present; Arts& Humanities Citation Index (A&HCI)—1980 to present; Index Chemicus (IC)—1993 to present; Current Chemical Reactions (CCR-EXPANDED)—1986 to present.
cCounted according to the formula Jimp = 10 + 295 × Factor, where: Factor = (1 − N) / (1 + (N / 0.057)), where N expresses the standardized ranking of the journal for a given discipline.
dWorld languages include English, Chinese, French, German, Russian, and Spanish.
eProceedings must be listed in the database Conference Proceedings Citation Index—Science or Social Science& Humanities (earlier ISI Proceedings) of Thomson Reuters;
fCzech or other national patent not yet used or used by the patent holder.
The criteria combine basic research, applied research and development, and are applied across disciplines. Since its introduction in 2004, the types of results and the assigned values have changed from year to year. However, key features of the assessment have remained stable and gradually become more explicit. First, the value of a text depends on its language. Until 2007, the line was drawn between the Czech (and Slovak) language on one hand and the rest on the other; since 2008 the boundary has been between “world languages” (English, Chinese, French, German, Russian, and Spanish) and all other languages including Czech. Texts in Czech are not given any special value, for example, as a means of communicating to national publics or for their capacity to enrich the Czech language with vocabulary and concepts coming from scientific research. The prioritization of “world languages” is done not only explicitly but also through a high value attributed to publications in journals with an impact factor and the steep function that rewards a higher impact factor. Without exception, this characterizes foreign, mostly Anglophone, journals. A similar preference for “international” and “Western” locations is observable in relation to patents: until 2009 patents were assigned the same value regardless of the country of issue; since 2009 there has been a significant difference between international, EU, United States, and Japan patents and “other national patent.”
Second, the results of “applied” research and innovation are imagined mainly as ultimately material (rather than textual) and protected by intellectual property rights such as patents. 2 By failing to recognize them completely, the framework effectively penalizes outputs aimed at the public sphere such as popular specialized journals, media articles, radio and TV appearances, and so on. It also severely limits what is recognized as a result in the form of expertise. It only takes into account “research reports involving confidential information” (i.e., reports for public administration which have to remain undisclosed) and “outcomes realized by a contractor,” that is, results that are verbatim included in legal documents or “demonstrably” used in strategic and conceptual policy documents. This means that the only recognized user of expertise is the state and public administration, not other societal actors.
Third, the stress on impact journals and technological artifacts conforms best to the output patterns of (some of) the natural sciences and technical disciplines. Policy makers are evidently aware of that and have started to differentiate between disciplines: between SSH on one hand and “other disciplines” on the other in 2006 and 2007 and between some of the SSH (the so called “national reference framework of excellence” includes humanities and some of the social sciences except economics, sociology and social geography) and “other disciplines” in 2008. It is, however, not the case that the two groups of disciplines get equal recognition. The criteria for disciplines in the national framework of excellence are adjusted in terms of value assigned to specific results, but not the types of results themselves. This means that the overall framework of what constitutes a legitimate and worthy research output remains tuned to (some of) the natural sciences and technical disciplines. 3
How do the assessment criteria reflect the idea of immutable mobility? At first glance, they strongly insist on and reward immutable mobility through the preference of “world” languages, impact factor publications and international patents, and of IPR protected outcomes and verbatim usage of expertise in legal and policy documents. Traveling and well-packaged results are highly appreciated.
It is, however, important to note that the criteria also impose a strong limitation on what movements and movements in which directions are valued. It is movement into and in “world languages” which is favored, and movement to places which are supposed to “certify” quality of results: by peer review in an impact factor publication and by granting a patent or putting an official stamp on expertise. These are very selective incentives for knowledge-objects to travel across networks—and by means of that to reconnect and reconfigure the networks themselves. There is an emphasis on—indeed, incentive for—academic–industry cooperation in terms of uptake of knowledge-objects and collaborative projects. In other societal arenas, scientific knowledge-objects are meant to enter as black boxes to be passively consumed. As the assessment table shows, no other actors than industry and public administration are recognized as users. There is not only no science–society program beyond the deficit model of educating the public, but there is also no dedicated support for interdisciplinarity in Czech science policy. knowledge-objects are made to travel in a geographical sense but stay in rather homogeneous (disciplinary) networks. In this sense, they travel very little. Moreover, some knowledge is valued because (not in spite!) of a restriction in its movement. Both patents and secret reports derive their value from such restrictions and exclusivity. On the contrary, highly mobile but badly traceable and controllable knowledge-objects such as media articles and TV interventions are entirely disqualified from the evaluation.
While mobility appears to be only a conditional feature (only a specific type of mobility counts), immutability is strongly appreciated in all cases, and explicitly in relation to applied outputs (patents and other forms of protected knowledge-objects; social science input to be incorporated verbatim into legal and policy documents 4 ). While immutability concerns the epistemic quality of a representational chain in the case of basic research, it concerns accountability or attribution of intellectual property rights, authorship, and societal impact in the case of “application.” 5 Because quantitative research assessment in the CR does not allow for more sophisticated modes of accountability (compared to, e.g., the U.K. assessment of the impact of research to be introduced in 2012 (Higher Education Funding Council for England (HEFCE), 2009, 13-18), the system has to rely on a rather mechanical attribution limited to patent or certification holding and authorship. There is no room for giving accounts, only for counting.
The Czech science policy thus combines a preference for (1) mobility from the local to the international, which in practice means from the Czech language and context into English language and publication venues located in the West and (2) immutability that is supposed to be certified by and related to ownership and intellectual property rights. It may be argued that the system of research assessment in the CR is uniquely crude and hence atypical with regard to research assessments in other countries. The Czech academic community has indeed recently started protesting against the system, pejoratively dubbed the “coffee blender,” and it will probably be further modified.
However, I want to argue that in its perhaps unsustainable crudeness the Czech research assessment system makes explicit the fantasies of policy makers (and perhaps wider society) about science and its desired functioning and place in society. It strongly resonates with what Brenneis, Shore, and Wright have termed “‘coercive commensurability’ or the tyranny of numbers: the compulsion to reduce complex social activities to simple numerical scores or ratings so that these, in turn, can be monitored, assessed, displayed and competitively ranked” (2005, 3, emphasis original). The geopolitics of immutable mobility inscribed into coercive commensurability becomes particularly visible in the Czech case, insofar as it is situated out of the Anglophone centers of global science, I will now add one more layer of marginality to the analysis and complicate the issue further by looking into what happens around and with this research assessment in academic institutions in social sciences. I will then come back to the complicity between STS and science policies in the conclusion.
In and Out of the Social Sciences
The scientific status of the social sciences has been disputed since their initiation in the nineteenth century, both within social sciences and in other places. Positivists typically defended a position that the social sciences can and should be (even if they have not yet been) able to produce and circulate knowledge of similar standing as natural sciences, that is, objective, reliable, and independent. They may be wrong, as Latour (2000) argues, in what they imagine to be the basis of objectivity in natural sciences and in their belief in a determinate external reality, but part of their argument is implicitly about being able to produce immutable mobiles: knowledge-objects that do not mutate, or get mutated, after separating from their authors. This is not only a positivist’s dream. In debates about qualitative data archiving similar claims have been made, for example, by Bitrich (2002). With explicit reference to Latour’s notion, Bitrich argues that if data are to be meaningfully archived and used in secondary analysis, it has to become immutably mobile, decoupled from the original researcher. Moreover, he argues, concerns over archiving only epitomize a more general problem with qualitative analysis in the social sciences, which often remain much too linked to a particular researcher and thus unaccountable in scientific terms.
Interpretative streams in the social sciences have formulated a different position. Again, they may be wrong, as Latour (2000) argues, in their imagination of the natural sciences as cold and insensitive toward their subjects of study, but their argument against adjusting social sciences to the model of natural sciences clearly concerns “immutable mobiles.” Interpretative approaches stress localizations of knowledge, embodied knowing, specific relationships between researchers and those researched. They present this as an asset rather than a drawback of social science knowledges. Some of these approaches are also sensitive to broader geopolitical issues of academic knowing, arguing against seemingly universal, unsituated social theory produced in the West and in English being treated as superior and applied to explaining local realities in supposedly peripheral countries (Connell 2009). Burawoy calls in this sense for “provincializing the social sciences,” namely, American sociology (2005, 20-22).
These authors describe the apparently immutable mobiles of (Western/Anglophone) social science as warriors of a neo/colonizing army bringing “democracy” and other amenities to underdeveloped parts of the world. As such they silence local scholarship. At best such imported knowledge-objects miss important features of domestic realities; at worst, they reenact these realities to fit Western theories (of rational choice, individualism, gender relations etc.). However, the agency of these processes does not need to be situated exclusively in the West. Discussing social anthropology in Central and Eastern European countries, Pobłocki argued that references to Western social anthropology are actively used by local scholars when “waging local battles against colleagues” (Pobłocki 2009, 234). According to him, the hegemony of Western theory has nowadays much less to do with neo/colonialism and is much more an effect of local scholars’ strategies. He argues that the epistemic dominance of Western scholarship with its orientation toward “theory” does not simply silence local scholars. It silences some of them in order to make others more prominent.
I believe that science policy operations such as the research evaluation described in the previous section have become a new and increasingly important factor in the formation of global epistemic geopolitics, and map onto existing inequalities and dynamics in multiple ways. To develop my argument I will now discuss the situation in two social science academic institutions in the CR which shows some of the intricacies. Part of the data presented here comes from a research project Knowledge, Institutions and Gender: An East–West comparative study (2006-2008) within which my colleagues and I carried out a qualitative sociological study of two academic institutions, in the area of the social sciences and biosciences. 6 This research took place in the period when the research assessment was gradually coming into effect. There was no significant public debate on the developments in the academic community, but research institutions and researchers were already internally negotiating the new policy framework and integrating them into their institutional policies. Additional data come from a textual analysis and a group interview with influential senior researchers in another social science institution, who have intervened in recent discussions regarding the assessment framework both on the policy level and in the public media.
Both academic institutions are research oriented, consider themselves top in their discipline in the context of the CR, and have international ambitions (foreign students, international recognition). There is a lot of publicly and privately expressed criticism of the current research evaluation system coming from these institutions and researchers. These critical voices mostly agree that the system stimulates the proliferation of “pseudoresults” and prefers quantity over quality, that is, more low-quality results easily compensate for a few high-quality ones. They also refuse assessment rules and criteria applied without considering disciplinary specificities. However, the idea of what constitutes “real” research quality is much more varied and closely connected to the desirability of the production of immutable mobiles as the ultimate goal of academic research endeavor.
The West is the Best
Institution PQR was established at the beginning of the 1990s with the financial support from U.S. foundations. 7 It has a high proportion of foreign students and teachers and most of its Czech members have a degree from U.S. universities. Internal evaluation of researchers at the institution is carried out by an international scientific board which is oriented exclusively to publications in international academic journals. The hierarchy of journals reflects their prestige in U.S. universities and does not strictly follow their impact factor as assigned by Thomson Reuters; the overlap with impact factor is, nevertheless, significant.
What papers can be successfully published in U.S. journals? In the interview, senior researchers agreed that it had been quite easy to publish about the CR in the 1990s as the region of Central and Eastern Europe was trendy in the United States. However, this does not say anything about the quality of the scholarship; on the contrary: “Looked upon historically, total atrocities were published only because the [economic and social] transformation was under way [in the country].” Nowadays, the region is less attractive and it is, according to the researchers, much more difficult to publish in U.S. outlets. Even though the CR is still conceived of as a laboratory where, due to the end of communism and subsequent transformation, there has been a “clean table that generated certain results … without influences producing noise,” “the CR and its problems can be now sold only as comparative studies across more regions or with other systems.” The relevance of a topic and a framework for enactment and recognition of legitimate transformations of knowledge-objects, that is, their immutability in the chain of reference, has come uniquely from the U.S. academic milieu.
The senior researchers also reflected that the Czech academic community in general is not able to write in the preferred language of these high impact journals. “[Czech researchers] have not been taught this in schools, either under socialism or now; (...) people don’t read the articles so the style does not get under their skin.” For these senior researchers, academic quality is unquestionably located in the West, particularly in the United States. The bodily metaphor of a style “getting under one’s skin” is important here, pointing to the incorporation of the hegemony through reading, writing, and literally being there for studies and fellowships.
The dedication to U.S.-based academic journals does not mean that PQR’s researchers do not engage in producing other texts, including working papers, books, expertise, or public media interventions. These are, however, not considered to relate to academic quality and the research career in a strict sense. Working papers, mostly in English, are a vehicle for acquiring feedback from colleagues and ensuring authorship of ideas in a highly competitive and quickly developing academic environment; publishing books is declared to be a “hobby”; expertise is produced for money; and public intervention is considered a personal engagement, way to influence public discourse. The scientific part of their activities is strictly limited to working papers as an intermediate step and ultimately academic journal publications in English, heading westward to the United States.
Importantly, the researchers assumed that the relevant framework constituted by U.S. academic journals is universal not only for the academic world but also in terms of social or societal relevance. They argued that most of their research and publications have in fact an applied character as they deal with effects of policies. Local administrative and policy institutions can and should “pick” practically relevant results from the researchers’ academic publications. If they don’t—as they don’t—it is their failure and a sign of ignorance. The U.S. academic milieu forms an unquestionable framework of reference, relevance, and quality; it is where the general social science theory rests. 8
“Central Europe” Out of the Centre
For seniors at PQR, the current Czech research assessment is too inclusive, not Western enough and thus creates a conflict with their internal evaluation. The situation at the institution RST is different and more complex and ambivalent. The differences come from many factors including the fact that it is an institution with a full education program from BA to postgraduate programs and it is a different social science discipline than at PQR (though neither of them is included into the National framework of excellence and both are evaluated according to the same criteria as the natural sciences).
From a certain perspective, Western-centered geopolitics is also strongly inscribed in this institution in the frame of reference for both academic and social relevance of research. For example, the official research plan of the institution stresses the European dimension of its research topics in terms of disciplines, policy, and social relevance. It reads: “The topics are subject of a dynamic elaboration in the context of European (social sciences) including applied disciplines (…) as it concerns problems that are topical not only for Czech society but also for the whole of Europe.” Expressions such as “the Czech Republic as well as other European societies…” or “the accent will be, in accordance with the current trend in European social science, put on…” is typical of the documents. The dominance of the European framework is strengthened by the declaration that “a delay can be detected in theoretical research in the CR—even though the agenda of the EU will soon concern also the CR and negotiations in the area have already taken place.” The concurrence of the talk about the European and Czech social science on one hand and the European and Czech society on the other is noteworthy. As in the case of the natural sciences and the “common nature” they supposedly share as a subject of their study, it is implicitly assumed here that a common (democratic, capitalist, globalized etc.) society is shared as a subject of social scientific study. Even if in different stages of development, world societies are converging, and sociology should participate in enacting such a convergence. Czech social science as well as policy and society are lagging behind the rest of Europe, which needs—with the help of the avant-garde of social sciences—to catch up with the West. 9 Meriläinen et al. (2008) make a similar argument when they discuss how they were made to rewrite their article by reviewers of a U.K.-based journal in such a way that the Finnish data had to be tackled as deviation from the “standard” of the U.K. data and the data about female managers as a deviation from the males.
The rhetoric of the research plan does not, however, translate in any straightforward way into attitudes toward the research evaluation criteria. The evaluation is on one hand criticized from similar positions as at PQR, that is, that it allows for quantity of low-quality publications to substitute for quality. Additional criticism is that the evaluation makes invisible the “people in the hinterland” (i.e., those who do not publish but teach and contribute organizationally to the running of the department) and, most importantly for my argument, that it pressures for texts lacking local relevance. Let us consider a detailed quote from a senior researcher who used to write, among other things, articles for newspapers and cultural periodicals for a wider public. He reflects on the change after the introduction of research evaluation. My question, which I didn’t pose to myself three four yeas ago, is whether a publication is reviewed and how, and how it will look when published? I have withdrawn from the media altogether and when they call me from a cultural journal, I refuse to talk about it. (…) This became unrecordable in terms of the rules that have been introduced. I have concretely encountered it this January when (the head of the research centre) asked me to remove a publication (from a database) and I knew he was right but at the same time I thought, God, it is a good article. I wrote it for a monthly published by the Jesuits, they asked me as a person from the other side (…) I took up the challenge because I had written about these issues before and discussed them from the perspective of secularized science. I think that I really struck home in the community of strong catholic believers. And I cannot record it. And it is a good text I think. OK, so I will not write it anymore, we standardise ourselves, but it may be a pity because an elite club is being created (...) that publishes only for itself. (…) We publish for our own community, we record it in the system, it gets us our professorships, but all of my actions just simply don't condone this, nor my thinking, this non-engaged, positivist science. I just don't like it, I don't endorse it (…). But the conditions will inevitably push us…
This quote catches several important insights and sentiments. In contrast to PQR, here international (impact factor) publications are not perceived as related to local (societal) relevance. Research evaluation remakes the topologies of scientific texts so that they travel long distances within disciplinary academic networks but very little outside those networks—to the public space or nongovernmental organizations (NGOs)—and they do not have a chance to interfere, clash or merge there with other texts and discourses. This is especially significant due to the strong emphasis on performativity of the assessment criteria. As it is an annual evaluation and records all eligible production, it leaves little space for researchers to dedicate energy to texts that are not recordable. Junior researchers insisted in a similar vein that the institution is not willing to support or even recognize symbolically any outputs beyond the official evaluation criteria. A researcher reflecting on a project she did with her students and an NGO on the mobility of migrant workers summarized the experience in the following way: “I think that for the so-called regional universities it is a way to become centres of intelligence in a broader sense; unfortunately the trend that everybody must have “impact” [English used in original] kills this because [collaboration with NGOs on locally relevant projects] obviously takes time which you cannot record in the database.” Unlike researchers at PQR who consider the quality constituted by U.S. journals to be the best guarantee of local societal relevance, many researchers at RST see these trends as cutting them from societal relevance. This creates tensions in researchers as they feel they need to change their publication practices in order to survive in the academy, and they do it, but they also feel that such a unidirectional reorientation makes their work locally irrelevant. In effect they cannot fully “endorse” their own work.
These dynamics are clearly not specific to the Czech situation. The U.S. anthropologist M. Huber has discussed “many stories about young and not-so-young scholars who have followed their interests in, say, the scholarship of teaching in their field, or applied or public scholarship, and found their careers derailed (Huber 2005, 50). It is notable that Huber uses the metaphor of derailment here. Derailed knowledge-objects can derail careers. The tensions for researchers also resonate with Cris Shore’s analysis of the schizophrenic academic subjects produced by conflicting institutional visions in today’s neoliberalized university (Shore 2010). Shore outlines some fourteen tasks prioritized by the New Zealand university he studied, encompassing everything from “international excellence” to “Maori development,” and analyses the tensions and stresses produced by this conflicting agenda. In the Czech case, I have emphasized the tensions between what is and is not officially recognized as useful and supportable knowledge-objects. As Shore’s account shows, simply adding missions to the existing value framework is no solution.
The empirical material presented above shows different ways of relating to immutable mobiles and their desirability in social sciences. For some, proper social science is constituted by immutable mobility. knowledge-objects without this quality do not belong into science in a strict sense, and research assessment should be further strengthened in the direction of impact factor publications. For others, varied forms of knowledge-objects, those that provide space for links with nonacademic social actors and interferences, should be included in the social sciences and their assessment. If in the first case the boundary between academic and societal relevance is blurred by conceiving of a (global) academic output as a potential source for application to be exploited by (local) practical actors, in the second case, societal relevance framed by local social actors should be let into science and coconstitute knowledge practices. The conflict is within the academy, not only between the academics’ and policy makers’ visions.
Conclusion
The concept of immutable mobiles developed within STS has never bought into an easy dichotomy between universalism and particularity, between the global and the local. It has also never taken immutability for granted, arguing that transportation is always transformation. Immutability is always relative in the sense that it retains the chain of reference and identity in some respects, but not in others (Latour 1995). Immutability, global mobility, and universality are achievements that need to be studied and explained. However, there has been little questioning in STS of the idea that as far as scientific knowledge-objects are concerned, immutability and (global) mobility are ultimate goals of the endeavor. As the study of research assessment and social science institutions in the CR shows, there are strong parallels between this assumption and various managerial approaches to science. I have identified processes of mutual reinforcement between the emphasis on mobility and immutability in Czech research assessment and some aspects of researchers’ understandings of the quality, relevance, and meaning of academic science. However, there are also voices in the research community that question the desirability of orientation toward “global” science. They point to the geopolitical and linguistic orders inscribed in the supposedly neutral and purely quality-driven criteria of globally excellent science. Their critiques direct our attention not only to the limits of contemporary research assessment policies but also to the problem of relying on immutable mobility as a description of scientific knowledge-objects or the outcomes of epistemic practices.
The social sciences (and humanities) face the ideal of an Anglophone version of their disciplines; they also have to negotiate, even if implicitly, the ideal of natural science operating in the logic of immutable mobility, a new version of the old positivist idea of the social sciences following the natural ones. Within those logics, my analysis shows that researchers’ responses have played out within a restrictive binary. They can either argue for more of “the same” and hence increased conformity to the natural science model (as senior researchers from the PQR do); or less of “the same,” arguing for the exceptionality of SSH (as the “national reference framework of excellence” introduced into the research evaluation in 2008 did when it assigned more “points” for some of the outputs in selected SSH disciplines).The first option conceives of non-Western societies as fundamentally the same as Western societies; Western societies constitute the paradigmatic subject of social scientific studies, and non-Western societies are destined to converge with them. The second option treats them as fundamentally different or, as in the case of the “national reference framework of excellence,” it distinguishes between the convergent socioeconomic dimensions of societies, studied by social sciences, and fundamentally different dimensions—such as “culture” or “language”—studied by the humanities. 10 Both of these visions are of course not simply descriptive but performative. In particular, both these visions contribute to the naturalization of the “socioeconomic core” of converging global societies.
It seems to me that there is little to be gained from choosing between the two or attempting to construct a satisfactory middle ground between them. Both are predicated on the problematic assumption or idealization of immutable mobility in relation to knowledge-objects. The alternative is to reexamine the privileging of immutable mobility itself—and here a reflexive science studies can play a pivotal role. Contemporary research assessment policies imply that when knowledge-objects become “derailed,” they lose their value. Alternative topologies to immutable mobility, developed for understanding innovation and medicine (DeLaet and Mol 2000; Mol 2002), can be an inspiration for rethinking science. They invite us to think beyond metaphors of knowledge networks composed of heavy and inflexible tracks, like those of a railway, and consider instead of derailment as a notion of mobility that is lighter and more plural. Knowledge claims that work in this way may not (be seen to) move as far or as fast as those operating on “mainline” tracks, but they may go to interesting places and be usefully transformed en route. Indeed, such an approach would be more faithful to Latour’s own call for acknowledging hybridity in science and on its borders (Latour 1993) and offers to open up new perspectives on some of the unease with contemporary technoscience. These involve (1) the possibility of rethinking science/society relationship beyond immutable mobility for the natural as well as the social sciences and (2) recognizing that “global” scientific networks—including some aspects of STS itself—are involved in enacting converging societies and economies, in the same way that STS has observed that technoscience has been involved in enacting a single nature.
We have seen that some social science researchers responded to research evaluation by insisting that the quality of their knowledge work was related to its capacity to become mobile beyond the “tracks” of formal scientific networks and recognition. Its value was in part constituted by the propensity of knowledge-objects to be changed by and through their engagements with publics not acknowledged in the official criteria. The properties of these kinds of knowledge-objects, which we might consider in terms of relative mutable mobility, seem to be particularly associated with the social sciences (e.g., in action and participatory research that acknowledges research “objects” as epistemic actors; Denzin and Lincoln 2000). But they have analogues in technological innovation, too, as DeLaet and Mol (2000) argued with respect to the “fluidity” that was key to the success of the Zimbabwe bush pump. Acknowledging social engagement in the enactment of knowledge-objects, as well as the hybridity and fluidity of knowledge-objects themselves, might mean some losses in terms of science’s control over immutable mobiles, but there are also gains in terms of developing inclusive relations with social actors outside strictly scientific networks. For STS, studying elements of social scientific knowledge practices, their disorderly mobilities and relative mutability, might open up new kinds of insights into epistemic practices and the nature of truth claims outside big technoscience—and at the same time contribute conceptually to how STS might understand multiplicity and hybridity.
Science studies insist that modern science has enacted, and not simply represented, an external, unified reality (Law 2004; Law and Urry 2004). Laboratories in particular have worked as machineries for performing a unitary nature, and we have seen how STS has used the concept of immutable mobiles to explore this performance and to analyze how “global” technoscientific networks have either ignored local specificity or transposed it into their own reality. 11 The convention in STS of positioning the (Western) lab as the central object of study has enabled this critical description, but has nonetheless kept the spotlight on places and times where Western (natural) science is most powerfully in action, rather than looking carefully at the epistemics and politics of knowledge that is more mutable and less mobile. The absence of the laboratory (at least in its strong form) in the social sciences might, then, be seen as an opportunity for STS to reconsider some of its central tenets, as well as a challenge to some aspects of research evaluation. The social sciences have a chance to take multiculturalism seriously by acknowledging that societies are multiple, drawing on their own traditions of engaging with and enacting such multiplicity (Connell 2007). Though I did not provide a full empirical examination of the relation between epistemic geopolitics and realities performed by social scientists which would require a more fine-grained analysis of research topics, methods and ontological commitments in the two research institutions, my suggestion nonetheless is that the social sciences are well placed to resist or reinvent research assessments that prioritize Anglophone impact journals with the effect of producing a converging reality. This does not mean communication across places or relating to the West should cease. It only stops taking the (Anglophone) West as a default reality. But STS needs to be fully a part of this move, too.
Footnotes
Acknowledgment
I would like to thank Lisa Garforth, Marcela Linková, Katja Mayer and two anonymous reviewers for their comments, criticisms and suggestions that I greatly benefited from, and to Lisa and Marcela for their help with English.
The author(s) declared no conflicts of interest with respect to the authorship and/or publication of this article.
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Work on the paper was supported by grant no. 403/09/P203 awarded by the Czech Science Foundation. Part of the empirical research that informs this paper was conducted within the KNOWING project funded under the European Commission's 6th Framework Programme, Specific Targeted Research Project No. SAS-CT-2005-017617.