Conceptions and expectations of research collaboration in the European social sciences: Research policies,institutional contexts and the autonomy of the scientific field

Abstract

This paper investigates the interactions between policy drivers and academic practice in international research collaboration. It draws on the case of the Open Research Area (ORA), a funding scheme in the social sciences across four national research agencies, seeking to boost collaboration by supporting “integrated” projects. The paper discusses the scheme’s governance and its place within the European policy space before turning to awarded researchers’ perceptions of its originality and impact on their project’s emergence and development. Drawing on Bourdieu’s field theory, we analyse the scheme’s capacity to challenge researchers’ habitual collaborative practice as well as the hierarchical foundations of the social science field. We relate the discourses of researchers, located in France, Germany, The Netherlands and the United Kingdom, to such structural dimensions of the academic profession as, disciplinary cultures, institutional environments and national performance management of research careers. The paper argues that the ORA introduced novel mechanisms of power sharing and answerability in social sciences research capable of unsettling the autonomy of the scientific field. This analysis offers a new perspective on the often unquestioned superiority of the model of international collaboration induced by schemes such as the ORA.

Introduction

Policies aiming to stimulate research collaboration are not a new phenomenon (e.g. see Sonnenwald, 2007 for a review). In Europe, large scale cooperation programmes have been in existence since the 1950s (e.g. CERN and EURATOM), and in the 1970s several intergovernmental funding schemes were introduced, driven by the ideas of networking and transnational cooperation as part of a “common project” (Lawn and Grek, 2012: 31). However the context of the collapse of communism in Europe, of increased economic competition within and outside Europe and of the rise of the “knowledge-based economy” paradigm in international organisations (Organisation for Economic Co-operation and Development and World Bank) from the early 1990s (Robertson, 2008), contributed to a shift in the implementation of this project. A “soft” approach to governing knowledge production (in education and research) gradually imposed itself in Europe, involving networks, professional associations, public/private partnerships, and other enabling arrangements capable of producing “new strategic geographies” (Lawn and Grek, 2012: 67). The launch of the European Research Area (ERA) strategy by the European Union (EU) in 2000 illustrates this process of strategic construction of European policy spaces. It arguably accelerated the emergence of networks, empowered existing ones (Breschi and Cusmano, 2004) and extended support for international research collaboration beyond resource-intensive disciplines. Such EU-led initiatives have had a defining structuring impact on the collaborative practice of academic researchers in Europe and beyond (Delanghe et al., 2009; Gusmao, 2001).

Yet policy research in Europe and internationally is certainly not the only structuring factor of collaborative practice, particularly in a profession for which collaboration in the form of co-authorship, peer reviewing and academic visits has “always existed as a form of social organisation and inquiry” (Papatsiba, 2013: 436). Incidentally, studies have also shown how most of these EU strategies and policy schemes promoting research co-operation and collaboration were championed by researchers themselves (Nedeva, 2013), and in particular those already enjoying dominant positions at the core of established networks within a fiercely competitive profession. Finally, policy agendas at the European level also come to life in the context of organisations (universities and research centres) whose missions and practice are evolving under multiple forces and steering models (Deem, 2006; Gornitzka and Maassen, 2000; Hazelkorn, 2009), and in which research careers are increasingly driven by quantifiable indicators of esteem and impact (Henkel, 2000; Lucas, 2009; Musselin, 2007). Within this context of converging policy processes and New Public Management reforms, we studied the interactions between policy drivers and academic practice in research collaboration using the case of the Open Research Area (ORA). This multilateral research funding scheme brings together four European national research councils, and seeks to boost collaboration by supporting “integrated” projects in the social sciences. The paper highlights the place of the scheme within the European policy space and the ideological underpinnings of its governance before turning to awarded researchers’ perceptions of its originality and impact on their project’s emergence and development. Drawing on Bourdieu’s field theory (Bourdieu, 1977), we analyse the scheme’s capacity to challenge social science researchers’ habitual collaborative practice as well as the hierarchical foundations of the social science field (Bourdieu, 1999) and its positional autonomy vis-à-vis external forces and cognate fields (Camic, 2011; Maton, 2005).

Field intersections

The autonomy of the scientific field, once established by Bourdieu as a prime consideration in understanding the academic profession, is being discussed here in its intersections with research policy promoting collaborations. Researchers’ strategies to enhance and mobilise their reputational capital – the most important of all in that field according to Bourdieu (2004) – are increasingly constrained if not aligned with institutional strategies themselves in competition to secure funding and reputation (Henkel, 2000). It is therefore likely that renewed conceptions of funded collaboration expressed in schemes such as the ORA, and their expectations regarding the impact of collaboration on research orientation and productivity will further undermine the autonomy of the academic profession in its regulation of research practice and outputs (Musselin, 2007).

The ORA was introduced in 2009 with the aim to promote cross-national research in social sciences. The scheme was developed outside EU supported schemes, building upon bilateral schemes between the United Kingdom, France, Germany and The Netherlands. In this paper, we draw on contextual information, policy documents and on 18 interviews conducted in 2013 with the principal investigators of thirteen funded projects from an early round of the scheme. We sought to understand how conceptions of research collaboration among European social scientists anticipate and accommodate the changes induced in their professional practice by changing policy and institutional environments. We initially carried out a mapping of those projects using publicly available data such as project presentations on the participating research councils’ websites, grantees’ CVs posted on their institutions’ website, and bibliometric information available from Scopus and Google Scholar. Alongside information about co-authorship between or among collaborating researchers, we focused on data reporting their more informal links such as academic invitations, but also joint participations in conferences/symposia, editorial or various advisory boards, as well as connections through networks or large EU projects. Interviews were conducted by Skype or telephone. Our aim was to collect discourses on the meanings associated with collaboration in research, its constraints and opportunities, and to relate these to the experience of their ORA funded project from the bidding stage to its current point, against the backdrop of disciplinary traditions, intuitional and national contexts, as well as individual careers.

Conceptually, our discussion conceives of academic researchers as actors whose practice is influenced “both by strategic calculation (…) and by reference to a familiar set of moral or cognitive templates, each of which may depend on the configuration of existing institutions” (Hall and Taylor, 1996: 955). They are also agents competing for positions in a field governed by “competitive struggle” for the “monopoly of scientific authority” (Bourdieu, 1999: 19). Their ideas of collaboration and expectations of the scheme are therefore related to the dynamics operating at the intersection of policy frameworks and local research spaces (primarily institutional and national) with that scientific field (the social sciences in this case). The focus on a policy scheme allows for transcending the false dichotomy between internalist and externalist perspectives on the academic profession and the higher education and scientific fields (Bourdieu, 2004; Camic, 2011; Maton, 2005). Most researchers interviewed for this study work in universities and are therefore operating at the intersection of fairly autonomous yet largely shared sets of “values and markers of achievement” (Maton, 2005: 689). National higher education sectors, scientific networks and scientific collaborative schemes, constitute the main fields of social practice that drive their everyday interactions. But it is in the scientific field that they have acquired and cultivated the symbolic authority, that is, the “scientific capital” as “product of recognition by competitors” (Bourdieu, 2004: 55), which traditionally has mattered the most to their career and social prestige. It is therefore within this field that we locate our concluding discussion.

The ORA: a funding scheme and a political statement

European research funding and governance at the turn of the 21st century

Although firmly set at the heart of the European research landscape, the ORA is not a programme initiated or financially supported by the EU. As discussed in the introduction its emergence has to be understood as a “non EU-driven” contribution to the “Europe of knowledge” (Elken et al., 2011) by the research councils of four core members of the ERA.

From the 1970s, the EU policy, driven by the principle of subsidiary, had always been to stimulate cooperation between member states, including science cooperation. However, research only became a Community policy in 1986, leading the Commission to take a commanding role in the coordination of national policies from the mid-1990s. By then, the Framework Programme had become an important source of funding for most member states (Guzzetti, 2009).

The formalisation of the ERA is commonly attributed to the 2000 Lisbon Strategy aiming to transform Europe into a world-leading knowledge-based society. The Open Method of Coordination “based on principles of voluntary convergence of States and reciprocal learning process though diffusion of best practice” (Guzzetti, 2009: 74) signalled a cultural change in the mode of collaboration between national agencies involved, as well as in the coordinating role of the EU. To some extent, the flexible steering championed by the ORA scheme (discussed in more detail below) is a reflexion of this new approach to research governance, but takes it further by bypassing the necessarily normative role played by the European Commission as coordinator.

The ORA also emerges in the context of the launching of the European Research Council (ERC) in 2007, which indicates a move in European research away from the principle of juste retour or formal criteria of multinationality (Guzzetti, 2009) and claims instead to “stimulate scientific excellence in Europe by supporting the very best, creative researchers of any nationality and from any scientific domain, including the social sciences and humanities” (ERC, 2010). As limited as it is in terms of volume compared with core EU funding schemes (i.e. the ERC represents 17% of the overall Horizon 2020 budget), the launching of the Council denotes a significant change in the delegation mode of European research funding. Great emphasis is being placed on incentive-type instruments for funding, and on a more “indirect steering” (Luukkonen, 2014; Potì and Reale, 2007) also observed in an increasing number of national contexts at the same time.

The ERC, and to some extent the broader policy reconceptualisation of research collaboration in Europe discussed above, are undoubtedly the result of tensions and lobbying from member states and research organisations within the Union (Luukkonen, 2014; Nedeva, 2013). The active role played by the European Science Foundation (ESF) and the European Union Research Organisations Heads of Research Councils (EUROHORCs) who initiated the ORA scheme, is specifically worth noting here. The ESF in particular believed that an ERC would employ simple and flexible management structures and procedures that would not be “burdensome for the scientific community that it serves” (ESF, 2003: 15).

The ORA within the new European research landscape

At the time of its introduction, the ORA supported social science cross-national research in four European countries, using the procedures for research funding allocation of their respective national research councils. In many respects, this model replicates joint frameworks described as following the “condominium” model (e.g. Schmitter, 1996) of decentralised integration, aimed at national partners working together “on a specific funding scheme, without delegating decisions concerning policies to a supranational body” (Lepori et al., 2014: 393). National agencies were therefore key actors in the launching and management of the ORA. The scheme is primarily about promoting strong collaborative ties amongst excellent researchers, and addresses the social sciences in their greatest diversity without promoting particular research programmes.

The financial commitment of participating countries in the ORA is modest ¹ . However, the initiative carries far more significance when considered in its policy dimensions. Bilateral and small scale multilateral schemes existed before but very few saw the light after the creation of the ERA when the EU became the central orchestrator of public research funding. This is not to say that states became irrelevant, but their role changed as “the European Union focused on initiatives with a higher level of institutionalization in the coordination-integration logics, whereas National States transitioned to lighter initiatives oriented toward collaboration” (Lepori et al., 2014: 398).

Most joint programmes in Europe therefore became either European-led or coordinated (ERA-NET) or benefited from a European financial top-up. According to Lepori et al. (2014), cases of small-scale multilateral schemes emerging in Europe after the creation of the ERA and without EU support have been rare and those who survived normally involve non-ERA partners (bilateral schemes with China or India for examples). From that perspective, the ORA can be seen as an original case of intra-European joint initiative bypassing the EU normative environment.

The ORA also illustrates a new type of set up within national contexts of research governance. States not only delegate the operational control of the scheme to “their” research council, but also the decision regarding the extent of their financial commitment, whereby “research councils are becoming an actor on their own” (Lepori et al., 2014: 399). This agency -led scheme therefore implies a relative similarity of structure of agencies (i.e. the research councils) in participating countries. In the case of the ORA, the scheme became possible for the existing four partners after France had launched its national research agency in 2007.

Yet the ORA brings together well-documented contrasting variants of the academic profession (e.g. see Teichler and Hohle, 2013) as well as different higher education systems, as far as their openness to international and dynamic competition is concerned (Marimon et al., 2011). A scheme that aims to integrate research across such diverse national academic contexts is necessarily one of compromise.

Nevertheless, the ORA sent a political signal to Europe regarding the capacity of national agencies to act as autonomous policy drivers of research collaboration. From its second call launched in 2011, the scheme also attempted to emerge as a broker for research involving collaboration outside Europe (ORA plus). In this respect, the ORA can be classified as a niche programme (Lepori et al., 2014) symbolising a move by research councils towards forms and frameworks of elective collaboration with compatible counterparts in other parts of the world.

Showcase of the EUROHORCs and ESF ‘vision’

As suggested earlier, the scale and financial commitments of the ORA have been limited. In each participating country, the level of funding allocated per project is capped and commensurate with what is made available through standard national open calls. The scheme itself is not subject to agreed budgetary commitments by partner agencies. Thus, the ORA (born out of established bilateral agreements) poses limited budgetary risk to participating countries and requires limited amount of trust among agencies. This suggests that the initiation of this programme was more instrumental in showcasing good practice in research management of national research funding, than attempting to impel social sciences in Europe in new directions. The scarcity of data on the first two calls of the scheme contrasts with the great number of references made to the ORA in national agencies’ communications, and confirms that the novelty of the concept is the key message that agencies seek to convey. Assessing the scheme in 2011, Paul Boyle, Chief Executive of the Economic and Social Research Council (ESRC), the UK research council, said:

The success of the ORA scheme is a testament to the close working relationships which have developed in recent years between the ESRC and its partner agencies in Europe. Together we have demonstrated that national research funding agencies working in partnership can make significant strides towards the establishment of bureaucracy free methods of undertaking international research, without the need for excessive restrictions (ESRC, 2011).

Whatever the actual “success” of the scheme and the significance of those “strides” are the message of the ESRC is a celebration of a selective approach to collaboration among leading European agencies. It is a message of realisation of the “vision” and “Road Map of actions” defined by the councils’ umbrella organisations back in 2008 ² . These programmatic actions were geared towards instituting common approaches to refereeing and evaluating funding schemes, streamlining the collaboration between research organisations, promoting “money follows researchers” principles and “lead agency procedures”, facilitating collaboration with researchers outside Europe (EUROHORCs and ESF, 2008).

The proposed approach to grant selection and management in the Road Map (EUROHORCs and ESF, 2008) is not only appealing to governments and agencies keen to retain more control over these processes within collaborative schemes; researchers themselves, generally dissatisfied with procedures in public research funding (Marimon et al., 2011) are being attracted by discourses of freedom and flexibility. A funding scheme promoting support to the very best researchers without ‘bureaucratic constraints on personal and financial mobility’ (EUROHORCs and ESF, 2008: 3) was always going to appeal to a profession torn between the doxa of collegiality and the fierce individual competition (both characterising the scientific field) (Bourdieu, 1999). Its appeal is further amplified by the individuation induced by performance and career management approaches which have been introduced in research institutions across Europe since the mid-1980s (Bleiklie et al., 2011; Musselin, 2008).

Researchers’ perceptions and responses to the scheme

We now turn to the ORA funded researchers’ perceptions of the scheme and how it meets, accommodates, allows for adjustments or perhaps collides with their conceptions and expectations of research collaboration. By international research collaboration we refer here to those “stronger forms of scientific interaction” involving a “joint research activity with a common aim or shared objective among scientists based at public research institutes in different countries” (Ulnicane, 2015: 434).

According to funding partners, the early ORA calls attracted a lot of interest from the research community (see figures in note 1). This helped in establishing the reputation of the scheme as competitive and of high standards, while further consolidating the largely shared assumption in policy circles that collaborative research is by definition of higher quality (Beaver, 2004; He, 2009). In reality, the number of applications received by the Councils and their success rates were consistent with usual figures for open calls at national levels. Furthermore, our analysis of awarded researchers’ institutional positions and publication records from one of the early calls revealed high concentrations of seniority and esteem in project teams that tended to bring together, in some combination, researchers with previous shared collaborative experiences (joint supervision, co-investigations and co-publications).

A convergence of interests seems to emerge between those highly networked researchers and a funding scheme claiming to offer them a platform for developing genuine integration in collaborative research. Also, the scheme appeared to move away from past policy conceptions of interdisciplinary as a “programmatic value tantamount to innovation” (Weingart, 2000: 26) and allow the unashamed development of niche areas firmly set within disciplinary competences and boundaries ³ . Yet, and perhaps unsurprisingly, researchers’ discourses give insights into a plurality of expectations of research collaboration as well as more nuanced perceptions of the scheme and its ability to ensure greater scientific impact. They reveal in particular how individuals’ views – even among such de-territorialised collaborations – remain deeply anchored in institutional contexts and disciplinary cultures. Relating individual discourses to such structural dimensions of the academic profession as, epistemic identities, institutional environments and national performance management of research careers, offers a new perspective on the often unquestioned heuristic novelty and superiority of the model of international collaboration induced by flexible schemes such as the ORA. This approach also permits revealing multiple interpretations of the “collaborative value” of the scheme by researchers, reflecting both aligned and contradictory logics of action of institutions and actors, rather than homogenous adaptive behaviours of agents to structural environments.

Another scheme, another opportunity: researchers’ agendas first

Most PIs involved in the early calls of ORA were mid-career to well-established academics. Most also had a track record of research grants as principal investigators (PIs) or co-investigators (CoIs) nationally or internationally. Such profiles tend to attract research collaboration requests, and they can rely on wide-ranging networks to monitor and seize funding opportunities in a typical illustration of the ‘Mathew effect’ (Merton, 1968). Networks certainly played a key role in the way most ORA PIs heard about the scheme, which in most cases, happened to map onto existing common research interests and collaborative funding pursuits.

A funding complement

A typical scenario, illustrated below, places the ORA call at the opportunistic end of a chain of joint projects. A Professor of Psychology from The Netherlands had been involved in a series of EU framework projects and bilateral initiatives as PI or CoI and was seeking funds to extend his latest project when he heard of ORA. That project was not necessarily conceived in terms that assumed cross-national collaboration:

Yes, the XXX study was first there, and collaborating with international colleagues and also as a follow up of a private study I did here … we thought it would be interesting to do a more extensive study of basic changes due to treatment …. And then my German colleague discovered this funding possibility and we collaborated on writing the grant application. So that was first things first, but we were just in time to sort of, yeah, set them up in parallel. It was quite complex. But we were lucky to win the grant and to be able to co-ordinate the two studies… (psychologist, The Netherlands).

Elsewhere, an international research team seizes the opportunity of a call that matches their immediate needs and offers the possibility of a speedy submission process:

Really just by chance, it was two weeks before the deadline of the call, we were searching for possibilities for funding [additional case studies for an existing project] because that’s indeed a problem to get funding for a specific bi-national project in Europe and, well, we found the ORA scheme and thought “that fits perfectly” and we submitted this proposal (psychologist, Germany).

In other cases, the search for funding had yet to start in a particular direction, but an operational network was in place at the tail end of collaborative experiences. In such cases, the ORA came as a providence to re-energize the collaboration:

It was only when that European Commission money started to come to an end that both Denny and I wanted to continue this collaboration, said “OK let’s see what’s out there”. And I can’t remember if he or I…I believe DM (UK CoI) found it and said “hey, take a look at this thing” – I think. But there must have been a website, there must have been some kind of information on the ESRC, I don’t remember to be perfectly honest (psychologist, France).

A springboard call

Different is the case where no on-going project was keeping the network active, but colleagues were on the lookout, ready to react at short notice:

we had a meeting in Toulouse in France and were coming from The Netherlands, UK and Germany, and therefore it’s something like a nice coincidence that we then heard about this call for tender and realised this kind of opportunity there. So we had had already some kind of contacts before but were not able to do joint research because we didn’t have the suitable funding for this type of research thing (geographer, Germany).

ORA came, T (PI) told me that V (other PI) had approached him and there was just an idea in a pub “why don’t we do something on XXX”. XXX was my research focus and the fact that I had been involved in a lot of research projects and managing projects I think, made me open to this possibility, and immediately I had ideas of with whom we could network (political scientist, The Netherlands).

In all these scenarios, respondents have either a strong experience of bidding for project funding (and of linking projects to one another as in a chain), international research networks or a relationship built over years of professional collaboration and friendship. The ORA opportunity came in at a time when they were more or less ready to apply for new research funding. The biographical data collected confirm existing links of various types between members of a team: contrary to expectations, co-authorship of research publications was not the main expression of these links, while informal collaboration, joint research projects, training networks or research unit-level formal partnership were frequently mentioned. Therefore, the scheme appears to have had little command over the composition of teams which were actually the offshoots of existing transnational networks. Analyses of evolving networks have already shown how highly connected nodes increase their connectivity faster than their less connected peers (Wagner and Leydesdorff, 2005), and the strong reputational capital of the PIs interviewed here increased both the size of the mobilised network and the speed at which information (on funding opportunities) was being shared within the network. Often this enabled them to submit proposals at short notice, as if they had been on the lookout for the appropriate funding source. The composition of project teams put forward for the ORA often met the requirements and prescriptions of the scheme without challenging the epistemic identities of researchers and their pre-existing collaborative and investigative dispositions of PIs.

“Local orders” and idiosyncrasies: on the weight of institutional and system interferences

As suggested earlier, PIs of the ORA projects tend to operate at the core of their scientific area, as strategic nodes capable of efficiently drawing together partners. Well-networked academics have a great deal of personal control over their associations, yet they are agents operating in different institutions in which the changing modes of knowledge production have created new imperatives and priorities which keep evolving. Our interviews revealed different forms of institutional influence over the ways in which PIs approached collaboration. The prospect of holding a grant bearing both the label of a national research council and an international outlook is undoubtedly appealing to universities and departments in search of resources and visibility, but their influence over the bidding process varies according to national contexts and governance arrangements.

The institutional factor

Most countries around the world are now developing strategies to enhance the international visibility of their most reputed research institutions. A manifestation of this can be seen in the diversification of countries represented in international league tables of universities (Altbach, 2011; Hazelkorn 2011).

In terms of academic career, this is suggesting that the reputation of an institution is a crucial add-on (what Merton refers to as “the institutional version of the Matthew effect”) (Merton, 1968: 7) for academics operating in competitive and stratified higher education systems. In the UK for instance, universities positioned at the top of research league tables contribute to amplify the visibility and appeal of their researchers as collaborative nodes. In other countries involved in the ORA such as France, universities may not secure such reputational capital, and membership of reputed CNRS ⁴ -accredited laboratories or capital city location often provide a better platform for building and fostering international networking capacity. Thus, although most researchers interviewed showed interest in the material and symbolic rewards of their successful ORA application, these did not invariably include institutional recognition. Our interpretation of discourses therefore needs to remain grounded in the idiosyncrasies of national higher education systems and their capacities to affect patterns of work at the individual level (Kogan, 2002; Musselin, 2008). Most comments made about reputation-driven institutional incentives and injunction to collaborate came from the UK. Elsewhere, institutional factors were much more commonly associated with internal politics or with direct pressure on academics to generate income.

Institutional versus individual research reputation

Unsurprisingly, UK-based interviewed PIs work at research intensive universities, themselves well positioned in global institutional rankings. Mid-career respondents were the only ones occasionally referring in interviews to the “institutional factor” as enabling greater personal visibility. They would, in particular, hint at the importance of being part of a well-connected research grouping, centre or department in terms of accessing collaborative opportunities internationally:

It’s essentially evolved through working with XXX, who had already had some networks. We collaborated on a paper and then this colleague said “actually this work aligns with work that I’ve been talking to colleagues in France and The Netherlands about” and it sort of evolved from there. So it was sort of initially a locally- based collaboration, but through kind of connecting with that colleague and his networks it sort of evolved into a European collaboration (geographer, UK).

In the above case, a mid-career researcher located at an internationally reputed unit is benefiting from the PI’s willingness to hand over the leadership of the project. Elsewhere, the unit’s reputation applied more directly: the department of a prestigious university was approached and the message passed on to the researcher by senior colleagues:

And meanwhile my director also came to me and said “hey, I’ve received this e-mail, I think you should participate, it’s interesting” (political scientist, France).

In such cases, the researcher is therefore delegated the responsibility to handle the reputational capital of the department. However, in highly competitive and successful environments, an ORA grant is not necessarily seen as a panacea:

I would have expected bringing in the best part of half a million pounds might have been somewhat more appreciated than it was. There was no buy-out, so it just adds to my workload rather than take away from it …. (political scientist, UK).

Other references were made here and there to the fact that having the University of X or Y on a bid could be perceived by others as a booster but no respondent of more senior rank considered this as influencing their own individual reputation. If anything, senior academics were generally stressing that their own reputation was contributing to the reputation of their institution. They felt they primarily owed their success in the ORA to their longstanding track-record in the field. Younger PIs meanwhile tended to invoke the primacy of their growing personal reputation for their involvement in an ORA team. However, whether this involvement resulted from a first approach to the department or from a personal initiative drawing on the credentials of the place, the common perception of collaborative practice as one building primarily on trust and mutual recognition did not change.

Research reputation as an authority-capital built on social and professional recognition, allows and requires renewed endorsement by peers (or “peer competitors” to use Bourdieu’s expression). Although known to be of great value in consolidating agents’ positions in the field (Bourdieu, 1999), the catalytic effect, so to speak, of institutions rarely surfaces in researchers’ discourses. This resonates with the conclusions of a recent qualitative study of academic reputation by O’Loughlin et al. (2015) which highlights the divergent understandings of research reputation between institutions, drawing on new evaluation and measurement systems, and their academics. At the heart of academics’ subjective perception – variable according to the strength of subject identification and socialisation – is the “recognition of a common or shared research perspective between academics, both of which affect how peers view and rate each other (and their institutions) in terms of reputation” (O’Loughlin et al., 2015: 813); this points to the persistence of norms and evaluation criteria shaped by professionalisation systems that are still operating alongside, or in spite of, institutional strategies. Yet the most senior figures in the most “recognised” areas of research, have more leeway to disfranchise themselves from institutional factors and injunctions to collaborate, while their younger colleagues, who need the support of their institution or centre’s reputation to insert themselves in “strong” international networks, may be – whether they express it openly or not – more exposed to the requirements of institutional strategies.

The collaborative injunction

Institutional strategies playing out in research collaboration were mentioned by PIs in reference to the circumstances of the bidding process. For most, the ORA project came as relief (one more grant, one more box ticked) and as a let-off in environments increasingly defined by “mission definition, prioritization, research concentration and the need to build teams and partnerships” (Henkel, 2007: 96). Yet some distinctive patterns of institutional circumstances are worthy of note here.

More cases of direct institutional injunction to apply for research funding were reported in the UK than anywhere else. The introduction of the full economic costing by the research councils in 2005 (whereby the research time of permanent staff involved is effectively subsidised by the grant) along with pressures related to the quantification of research income for nation-wide research evaluation exercises are said to be largely responsible for this climate (Deem, 2006).

In the case below, the financial contribution of the grant overshadows its symbolic value for a department located at a prestigious university in which similar achievements are part of the standard expectations about academic performance. The announcement of the grant generated an impassive response:

….one senior member of the department said “well done”, the other senior member of the department said “well done, that’s going to be a lot of work”. The School was very positive about the amount of money that I’d brought it, but it doesn’t affect anything. (…) and it doesn’t enhance my position at all as far as I can see (political scientist, UK).

In this case, the grant is just one of many, and is welcome as such, but is not offering much comfort to the individual recipient (mid-career academic), as the institution is not prepared to increase her research time beyond the standard allocation that her position already entailed before the grant success. In this case, it is down to the researcher to fit it in. Yet the university is content to promote this new addition to its awards cabinet:

I was asked yesterday, the ESRC apparently are visiting this month, and asked “could you do a little impact study for us – if you think you’ve got an impact” (political scientist, UK).

Institutional logics operate differently according to the universities’ positioning in the UK market. Here, a Professor of Economics feels the pressure of an institution’s managerial approach to research strategy where activities are more centrally directed and narrowly determined by consideration of income generation and other quantifiable outputs.

I had this big VC or whatever saying “do the project, you must apply for the project” and so on and so forth. OK so I did it. I spent 6 or 7 months doing nothing but preparing this bid. I got the project. And then they summoned me again and said “OK where are your publications, why are you not submitting anything” (economist, UK).

The case above, illustrates the limitations of the protective value of a research reputation in management contexts where institutions’ research income depends directly on the productivity of their staff.

In other participating countries, where academic positions are not so closely tied up to institutions’ research income and where evaluation is used as a performance management tool rather than a resource allocation instrument, the pressure was more likely to be associated with “getting something” in order to get going with one’s research while enhancing the reputation of the institution.

Well I would say it’s informal, there’s no formal incentives like you’re getting more money and so on, but yes informally the feedback and stuff is highly appreciated, and for example the head of my department says that it’s good for our department and the whole institution because international projects are highly regarded. So I would say it’s a bit more appreciated than normal DFG funding so to say, which is already very highly appreciated” (psychologist, Germany).

I don’t feel pressure from my Dean for example to become more collaborative or get more funding, although it has become more or less the implicit rule that you take the opportunities that you see (sociologist, The Netherlands).

In the above situations the grants were welcome by institutions for their symbolic and financial value. International collaboration operates as a capital enhancer for universities, and researchers are expected to show an international profile commensurate with their seniority. Interviewees from France, Germany and The Netherlands were all prompt to stress the value of the ORA for the visibility and reputation of their university, and to acknowledge how getting the funding consolidated their own position within it, without reporting the forms of direct injunction revealed above about the UK context.

Overall, the forms of institutional pressures reported tended to reflect distinct patterns of research governance and incentivisation. Institutions either embraced the integrative conception of collaboration expressed in the ORA call or encouraged their staff to bypass it in favour of the “consortium” types of projects that fit more readily in their organisational culture, and in their national research council’s requirements. In seeking alliances and partnerships for their proposal, researchers preparing ORA bids had to anticipate those institutional expectations, as well as juggle with the differences between national research councils which had been brought together around a loosely defined collaborative objective. Few saw these as obstacles requiring well-planned and clearly articulated strategies. Rather, interviews confirm how highly established researchers draw on the register of experience and implicit knowledge of their environment to explain the success of their collaborative enterprise.

Ownership beyond leadership: trust in collaborative project management

The ORA is somehow disrupting the usual habits of conducting funded research in social sciences because the scheme does not define in any prescriptive ways the type of collaboration expected beyond stating that projects “must involve integrated collaboration between partners” (ORA, 2009), and requesting one principal investigator per country, who in the main reports to their national research agency.

Unusual in social science funding schemes across Europe, the request of multiple PIs in the ORA is close to the multiple-PD/PI model typically applied in Health and Science subjects to “facilitate multidisciplinary and other types of team science projects that are not optimally served by the single-PD/PI model” (National Institutes of Health, 2011). In the case of the ORA though, there is no guidance related to the multiple PI model apart from the fact that each PI is accountable to their national funding agency.

While unanimously expressing their satisfaction with a scheme imposing less interference and steering than many, researchers acknowledged that they had not experienced this model before and that it implied some adjustments to their usual practice of one overall leader and a single accountability channel. Consequently, the ORA leadership model seemed to be challenging expected patterns of authority. In particular the multiple PI model disrupts the conscious act of role-playing implied in large – and often interdisciplinary – collective bids where a PI is identified – who may or may not be the most authoritative voice in the team – to whom, however, all others temporally agree to be subordinated. This is where the question of trust and shared values surfaces, expressed in different ways by researchers, but always linked to this unusual model of governance:

you don’t have to be best friends of course, but still there must be a lot of trust, a lot of confidence, I mean each project partner in the way we have set this up could have ruined a lot for the others. And we trusted each other, we knew what everybody would do (linguist, Germany).

What I really remember was that things were very helpful by, I think all the partners being very clear about the value of the project, I think that was a very clear thing, that we all knew that this was something that we were really keen to do, we had a very kind of strong idea of the proposal (geographer, UK).

In the only four-country interdisciplinary project, bringing together PIs with significant influence in their field, the resulting tensions revealed the high level of instrumentality thrown in the collaborative enterprise. One of the researchers explained how he felt he led the research funding application, seeing himself as natural PI up to a slip of the tongue:

The project is basically a federated project, this means basically it’s a network of national projects, and as a principle investigator I don’t have a lot of influence on the real work that is being done in the other (teams) (sociologist, NL).

Returning to the issue later in the interview:

I think I made maybe a bit of a shortcut in claiming that I was the principle investigator of the project, and so I didn’t mean to play down the role of S, but what I mean is that what I expect from a principle investigator is an attempt to also intellectually make combinations from the different traditions, and in this project this is not so easy because these traditions are really very different (sociologist, NL).

This leads us to the issue of preserving national models of accountability in cross-national research funding schemes aiming to stimulate high levels of interaction and “jointness”.

With no lead agency, and a multiple PI model with separate funding and accountabilities, the ORA poses a challenge to traditional expressions of leadership authority in research collaboration, and in the process, tends to magnify differences, particularly in larger and more interdisciplinary teams. Here the frictions and misunderstandings do not arise from imposing a uniform top down leadership on individuals from different cultural backgrounds and epistemic traditions as often reported in EU-funded “framework” projects; on the contrary, researchers turn to the more or less established, rigid cultural assumptions associated with national types prevailing within epistemic communities:

But then of course there is also the kind of broader intellectual academic culture issue in that people tend to frame different things, you know, French has a very different way of framing a kind of a problem or approaching the framing of a problem than say the more pragmatic English or Dutch framing. I mean the English and the Dutch in some ways are the closest I think on this project in terms of intellectual approach or understanding of what’s involved, and the French are probably the most removed in a way from the English understanding (sociologist, UK).

The reflections collected clearly show how a multiple PI funding scheme can unsettle relationships in teams, particularly where leadership of research grants counts as much as publications in performance indicators and career progression. The model of the single PI in European social sciences has long been established as one of those “crystallising agents” (Luukkonen and Nedeva, 2010) that contribute to the regulation of the scientific field, or more specifically for our focus, to the “forced agreement” (Bourdieu, 1999: 33) deemed necessary to the integration of international interdisciplinary research teams. While aiming to stimulate integrated collaboration (ORA, 2009), the ORA has the potential to disintegrate those “forced agreements” by allowing individual researchers to control equally important dimensions of the research, in the absence of mechanisms of agreed subordination. Yet, despite the tensions and frictions reported above, this disintegration did not occur in the projects we examined. For many of these, the reason is to be found in the socio-epistemic cohesiveness of the networks out of which the bidding teams were constituted. By cohesiveness, we refer to the level of cognitive integration of project teams, as well as the social cohesion of these groups made up of reputed scholars in their field and established academics in their institutions. Even in the most competitive collaborations (ORA also stimulates intra-project competition with this multiple PI approach), collaborators – even the most frustrated by the lack of vertical integration in their project – had no interest in derailing a collective enterprise associating them with internationally reputed colleagues and/or organisations, and thus reinforcing their own scientific authority. In most cases, despite its challenges, the multiple PI system is seen as generating freedom and allowing organisational innovation. In others, the scheme supports alliances of minimum common expectations of utility among individuals competing for scientific authority. As Luukkonen and Nedeva (2010: 677) put it:

The extent to which the entity is cohesive depends on the clarity, communication and level of acceptance of the dominant expectations of utility. Respectively, the extent to which members integrate successfully into the entity depends on whether their expectations of utility and the expectations of utility of the entity are similar or compatible.

General conclusion

The paper sought to shed light on the conception of research collaboration underpinning a particular funding scheme in the social sciences, and on its interpretation by funded researchers in order to ascertain the current permeability of the scientific field to policy incentives and pressures.

After reviewing the context of emergence and broad orientations of the ORA scheme, the paper presented awarded researchers’ reflexions on their collaborative experience within the scheme, and the extent to which they felt it differed from their previous experiences of international collaborative research. We sought to illustrate how collaboration – presented by funders and academics alike as proxy for research excellence and as a core value to the profession – has multiple meanings across a diverse European social science field, as well as being deeply rooted in local models of public governance and institutional configurations (Paradeise and Thoenig, 2013; Thoenig, 2003).

First the scheme itself, as we have shown, can be understood as a political statement on research governance by key European science national self-governing funding agencies in the context of the ERA. For this reason, we suggested that the originality of the scheme lay essentially in its timing, its integrative ambitions (expressed in the 2008 ESF EUROHORCs road map) and in its governance (the “condominium model”, as expression of the increasingly proactive role of institutions and funding agencies in research processes).

While the scheme undoubtedly supports transnational collaborative experiences and networks, the collaboration it induces maintains participating researchers in a state of answerability to “largely localised, mainly nationally bound, research spaces” (Nedeva, 2013: 221), thus allowing space for local social orders within the international nature of the call. For instance, we have shown how the researchers’ comments on their ORA experience revealed contrasting work cultures and institutional loyalties. Universities and research organisations (including national councils) are, in particular, impacting within significant national variability on the scheme’s operationalisation with the introduction of workload allocation and other cost-consciousness measures such as the full economic costing, or by imposing rules on the contractual position of PIs. Variable levels of institutional recognition of research collaboration were also perceptible in the contrasting support granted to researchers’ collaborative efforts. Finally, despite well- documented isomorphic trends (Bleiklie et al., 2011; Shore and Taitz, 2012), national higher education funding models continue to influence the level of incentivisation of research and research collaboration within universities, and to steer research activities more generally through a combination of economic and political controls (Paradeise and Thoenig, 2013). They too influence the formation of research collaboration networks and their level of integration.

It would be tempting to read academics’ responses to these pressures in terms of adaptive strategies but this would hide the richer, more nuanced and subtle experiences and explanations encountered in this research. Interviews revealed that the scheme either funded rather established groups of collaborators or collaborations that partially branched out of these groups, and that teams initially approached the ORA in the way they would consider more familiar funding sources (such as national agencies and EU framework of bilateral schemes). The specificities of the scheme usually unfolded at a later stage of the bidding process. If anything, researchers’ first encounters with and initial reading of the scheme revealed first, the key role of research networks in horizon scanning for funding opportunities and second, the autonomy granted by “reputation” (within the scientific field) in relation to funding strategies. Respondents often referred to the ORA as a funding scheme introducing requirements minimally impacting on their research idea and choice of partners (to do “what we wanted to do”, to complement some existing funding, to work with people they liked, to ease temporary institutional pressure, etc.). In this, they tended to reduce the value of the scheme to a tool at their disposal to help them consolidate collaborations and steer their research in a particular direction.

Yet, accounts of actual collaborative practices, especially from participants in the largest and more interdisciplinary four-country projects, revealed how much the scheme disrupted established patterns of collaboration, and in particular the principle of a negotiated subordination associated with the single PI policy. All PIs admitted to being insecure within the multiple PI framework, whether perceiving it as an obstacle to integration or seeing in it as an opportunity for newly defined terms of what counts as cross-national collaboration.

As limited as it is in terms of budgetary commitment, the ORA funding scheme exemplifies significant trends of policy directions in the European social sciences. Firstly, it offers a concrete illustration of the key role of leading European national funding councils in the steering of international research practice through the construction of (highly visible) collaborative models. Secondly, the ORA appeals to researchers by claiming back the primacy of research excellence over EU-type redistributive policies and bureaucratic interference. In doing so, the scheme does not signal the return of collaborative practice of old, but rather reveals more subtle interdependencies between the fields of science and science policy, where controls and power operate at institutional (performance management), national (funding), and international (partnerships and peer review) levels. As such, this is not novel, as scientific collaboration somehow always occurs “within the larger social context of science, which includes elements such as peer review, reward systems, invisible colleges, scientific paradigms, and national and international science policies, as well as disciplinary and university norms” (Sonnenwald, 2007: 646). However, and most importantly, by reinforcing the networking capacity of established researchers and adopting a much commended non-bureaucratic rhetoric, the scheme is playing a deceiving game: it allows the distribution of power and monopolies operating within the field of social sciences to express itself at the application stage, and later on introduces patterns of cooperation and accountability that challenge it. In doing so it favours the “‘central’ players, the orthodox, the continuers of normal science” (Camic, 2011: 279) but ultimately destabilises them, at least temporarily, with unorthodox management practices that appear to better suit those least endowed with capital (by offering status equivalence and visibility to PIs of each projects) or highly complementary teams in their methodological expertise or access to data. Hence, a perceptible ambivalence in the researchers’ views about a “non-interventionist” scheme that, although ostensibly embracing their collaborative doxa, introduces mechanisms of power sharing and answerability which in turn challenge their capacity to regulate internally the individual roles and relationships at work in research collaboration.