Accueil > Governance > Reflecting upon supporting High risk research

Reflecting upon supporting High risk research

vendredi 21 janvier 2011, par julien

There have been numerous debates about how to undertake this objective. Should we, once we had recognised enduring challenges, adopt a top-down approach, concentrating funds on these topics and allocating them to those members addressing them ? This was the solution favoured by most policymakers in Brussels (the notion of ‘network of excellence’ having been designed as an alternative to the initial proposed mechanism of ‘centres of excellence’) and quite a few members within the community. The collective choice of those managing the NoE was different. Such a solution would fit with deepening existing paradigms, i.e. developing further existing approaches, a situation which we felt the community was very good at doing both at national and European levels, using the existing support mechanisms (for research, studies and evaluations). The issue was that, with regard to the enduring issues, these were not addressed ‘well’ by existing theories, concepts and methods, and instead we needed to encourage variety and the emergence of new paradigms. Our choice has been to foster bottom-up initiatives and the assembly of topic-based ‘critical size’ teams through a repeated competition process (with 3 calls per year for 4 years), and a selection process geared toward improvement (many initiatives were initially rejected and only finally accepted after revision taking into account the advice given) and risk-taking. We also recognised that such developments required a staged process.

We thus initiated two types of research supports : exploratory research projects on potentially new breakthrough methods and approaches on the one hand, and comparative and aggregative research projects on the other, enabling us to reframe the approaches and issues mostly dealt with previously within a national context. The aim was (and indeed this has been achieved in a large number of cases) that these approaches would, through these mechanisms, have time to demonstrate their value, and attract academic recognition, and would then be fit to enter into existing programmes and their selection mechanisms (from national funding agencies and from FP collaborative programmes).

Taking risks also requires some effort to bound them. Instead of a tedious monitoring process, we tested another approach based on what could be termed a ‘step by step’ funding process. When selecting a project, we only funded a first step (not less than 12 months, but no more than 18 months) and then asked project participants to come back to the competition with the results of the previous step and a redesigned next phase. This proved efficient both in terms of not encouraging projects to ask for a second phase, but also in helping others to concentrate on the central issues. However risk-taking has two aspects : for the few projects that succeed completely, a number will inevitably ‘fail’ (although such failures also provide strong opportunities for learning depending on organisational memory of organised groups). Our ratio of ‘productive’ projects is however quite high, which may raise questions about our concept of a ‘breakthrough’.

After two years of activity, we had an open debate about critical areas that had not been taken up and the need to adapt the approach. This was debated with some annual reviewers and with the ‘Governing Board’. We were thus more pro-active in finding champions to nurture an initial process but kept the open competitive process to select operations. One central point of debate was about the effects of such an approach : were we not de facto fostering vertical silos, based on a number of small sub-networks, each focused on a different topic, rather than pursuing a wider integration process ? This was the central view expressed by annual reviewers. We had anticipated this issue and created an independent ‘characterisation’ group to monitor on-going developments in particular through two surveys, an initial one and a second one three and a half years later. By and large the results demonstrate that this was not a problem.