Can social engineering work in science? The European Commission (EC) last month took steps to answer this question as it advanced its sixth Framework — a four-year, 17.5-billion-euro (US$17.6-billion) plan to distribute funds based in part on researchers' abilities to form pan-European research networks.

The initial signs appear positive. In March, the EC put out a call for research ideas in several broad categories. Those ideas are to be evaluated both on their scientific merit as well as on the number of institutions listed. Last month the EC reported that it had received 15,000 proposals. But some scientists who have submitted proposals say that number may be a bit misleading, as some proposals list 50–60 collaborators and some groups appear on several separate proposals.

These numbers can be interpreted in two ways. Looking at the sheer volume of responses, one could say that the EC accomplished its goal of getting scientists to find out who is doing similar or complementary work in other parts of Europe. But examining the redundancy, one could also say that the plan has led to a duplication in proposals, as scientists strive to increase their odds of being a part of a group that makes its way through the funding process.

As a survey of interest in pan-European research projects, the Framework is a qualified success. But as a mechanism to make researchers who have never worked together cooperate, the result is likely to be less positive. Even if a fraction of the respondents submit formal proposals by the end of the year, it is likely that only a few of those will be funded. And of those, it is unclear how they will be managed and evaluated. Coaxing groups to work together seems to be an admirable goal — but making it happen effectively remains difficult to accomplish.