Yesterday saw the launch of Science for the Future and it certainly did what it was intended to: make a splash!
via sheer lunacy.
Sheer lunacy makes an argument, as well. Two elements in the argument are problematic, however:
1) The move away from investigator driven research to directed research portfolios. No evidence has been supplied by EPSRC to suggest that directed research portfolios produce better science.
This attempts to place the burden of proof on EPSRC. But it is by no means clear that this is where the burden ought to lie. Is there any evidence that a laissez-faire approach produces better science? Moreover, it assumes that EPSRC’s decision ought to be based on something like scientific evidence. That sounds fairly reasonable at first blush, but it ignores the fact that such decisions are political. EPSRC is responding to increasing demands for accountability, and doing so in a way that reserves the majority of the portfolio (60%, I believe) for ‘undirected’ research.
4) The downgrading of peer review in the grant assessment process and the introduction of non-scientific and subjective criteria such as “importance” and “impact” to determine funding. As well as “guidance” being given to panel members as to which proposals have best fit with EPSRC priorities, regardless of scientific excellence.
There are multiple problems here. First, it begs the question, assuming that scientific criteria are the only ones that should matter. It also implies that criteria of scientific excellence are objective, while impact criteria are subjective. We’ve argued against that view at length here.
Second, it implies that guidance given to reviewers is somehow coercive, forcing them to consider impact over scientific excellence. EPSRC’s guidance to reviewers is fairly clear and extensive. It also refers to the even more extensive discussion of impact on the RCUK site. There is also a clear statement that scientific excellence remains the primary criterion for peer review: “The primary criterion for RCUK funding remains excellent research.”
RCUK are clear that this is a quid pro quo arrangement. Researchers are granted significant autonomy in return for engaging the impact agenda. That strikes me as a reasonable deal. Perhaps researchers would prefer simply to be given money on the grounds that they know best how to spend it. But that’s unrealistic, if not irresponsible. It certainly doesn’t seem to fit well with RCUK’s Royal Charters.
Obviously, RCUK (and EPSRC) are engaged in attempting to foster a certain attitude among members of the scientific community — one that is different from that on display at Science for the Future’s mock funeral for science. I suggest that members of the scientific community would be better off engaging EPSRC as co-producers of the impact agenda. Volunteer to serve as a reviewer instead of implying that peer review is being downgraded. Take the political exigency of impact seriously and find ways to own it, rather than simply reacting to it with resistance. EPSRC, too, should reach out and try to engage especially those members of the scientific community most resistant to impact. If scientists and science funders don’t work together, politicians are likely to react by cutting science budgets. That’s an outcome no one in the scientific community (including funding agency officials) wants.
1) The burden of proof should be placed on the EPSRC.
When someone or some organisation introduces a change to a system it is incumbent upon them to provide the evidence to support the changes they are making or suggesting. However, you probably have hit the nail on the head that this is a dogmatic political decision as it certainly doesn’t add up as a rational evidence based one.
4) The whole point is that it should be scientific excellence which drives the distribution of funds. If you start introducing other factors then by definition you are reducing the requirement for scientific excellence. Science is based on fact and verifiable experimental observation, impact or importance are, by their very nature, subjective. As soon as someone starts offering “guidance” then by definition they are seeking to guide you into a particular choice. Once again EPSRC have not provided any evidence to support their new rules so one must again assume this is a political choice. A choice that I happen to believe is bad for science. If you disagree collect the evidence and prove me wrong.
The scientific community have tried to engage EPSRC on many occasions over these issues. We have tried through our professional bodies, universities, campaigns and as individuals, however, the EPSRC are not willing to even discuss the issues. If they were willing to open a dialogue with us then we would be able to reach an amicable compromise. Sadly the EPSRC are set on a strategy where they can’t be seen to compromise and so we find ourselves where we are.
Paul,
Please accept my apologies for not having approved your comment earlier. It must have been inadvertently marked as spam.
It is an interesting question where the burden of proof lies in this case, as well as what standard of proof we ought to require. The first question is more difficult than the latter, it seems to me. We should not require scientific proof of/for policy decisions. It’s a category mistake to assume that science is sufficient for policy decision making (even for decision making about science policy). There are many times when we want scientific input, and even many times when it is essential. But to think there are any times when science alone should simply dictate what we should do in a policy context is a mistake.
So, the standard of proof we should expect here is that of an argument, not that of randomized controlled trials or something similar. I think EPSRC have given an argument. I don’t agree with all of it. For instance, not allowing resubmissions of rejected proposals is a mistake that overvalues efficiency (it’s an ‘easy’ way to reduce the reviewer burden and the rejection rate). I might disagree about their choice to fund this or that big project, too. I even think that there should be more of a bottom-up approach to impact. Where you and I disagree is that I think impact should be a factor. I’d rather it be a factor scientists take ownership of, though.
As for where the burden of proof lies, I don’t think it’s clear cut. EPSRC have given arguments for their moves. If we don’t agree with them, then it is up to us to attack their arguments. Publicity stunts that appeal to hyperbole and authority won’t cut it. You’ve offered an argument above, and I think it’s reasonable. But I think you’re slightly off in arguing about what’s “bad for science.” EPSRC are charged with a dual responsibility — for science and for society. It’s simply not always the case that whatever is good for science is automatically good for society. So, sometimes choices have to be made. Again, we may disagree with the choices. But to argue that the choices are wrong simply because they are bad for science begs the question.