Its a question put out there by Times Correspondent Hannah Devlin on Twitter. I think the answer is a resounding, projects, projects can I say it again projects.
Idealistically it is certainly true you should fund the best people: I think it is pretty safe to say when most of us are hiring someone they will at least try to pick the best person for the job (in the absence of corruption but well that is another matter).
In many ways, the safe bet by the governments/funding bodies is to fund ‘people’ instead of projects – and if you pick those people carefully you are pretty certain they will deliver; they will publish high impact papers etc. In the short-term what is not to love?
But I think in asking these question we have to think of what it means to be ‘successful’. As Jack Stilgoe pointed out on his very nice post about this question the funding people policy is a policy supported by many who are big scientists who define “scientific excellence, as measured by scientists, for scientists.” In the short-term maybe this means you are an FRS and have like I don’t know 30 Nature papers, but time will tell if this is a long-lasting success or just a branch of science that runs into the dust.
Jean-Baptiste Lamarck was a very successful zoologist in his day and gave us a complete classification of invertebrate zoology as well as basically coining the term biology (in the professional sense). However Lamark also believed in what is now called ‘Lamarkism’ – a theory that organisms can pass down acquired traits to thier off-spring – such as the Blacksmith’s arm, or the giraffe’s neck – which doesn’t really work any more, we know this theory pretty much isn’t true. But it was a good theory at the time the only reason we know it isn’t in hindsight is that new ideas (like Darwin) came along out of nowhere and made more sense.
In the long term only funding people disconnected from projects is a bad idea. First of all the people that would like get funded by this mechanism (if we moved entirely to that) probably already get money anyway. And to get more money, as it stands, even the great and the good have to write new grants with a new ideas and yes this is annoying to have to do every few years, but it still means your idea is vetted in some fashion.
Secondly,funding only the elite would cut your science base to shreds, where would the new people come from? Where would the new ideas come from? Where would the diversity be?
I say this because some people are very specialized while others are not, if you are specialized in one technique, one system – while you may do fabulous ground-breaking research, you may not be able to produce a future generation of scientists by which to stimulate new ideas. And your ground-breaking research may actually be the modern equivalent of Lamarkism.
Great post, Sylvia. I couldn’t agree more. I was very disappointed when the Wellcome discontinued project grants, and put most of its clout behind the rising stars who are already, I assume, not strapped for cash.
Yes, the post doesn’t specifically mention the Wellcome Trust, but (at least in the UK) it is the only funding agency that funds people rather than projects. In fact I think this is a good idea, but only for a small number of truly exceptional individuals. For the superstars out there, it makes more sense to provide them with rolling five year grants, subject only to satisfactory quinquennial review. The problem is that the WT have put all their eggs in this one basket. I would have left at least 60-70% of the budget available to open competition, where peer review takes into account not just track record (which is important) but also the project proposal itself. There is no question that by vastly reducing the number of funded researchers, the WT will miss out on some of those unexpected but gilt-edged discoveries that can emerge from almost any lab.
The Inserm has come up with a very frustrating program that rewards, with salary bonuses for five years, people who have already brought home a scientific prize with money attached. There is a tiered system. I will try to find it and post it. Meanwhile, the Inserm is not hiring many young scientists and in particular the needed technicians and engineers – certainly not massively replacing, as promised, the half of the scientific laboratory workforce that is supposedly retiring in the near future. This money wouldn’t pay for that but it would pay for something, and meanwhile is perceived as grossly unfair. For example. why not give the money instead, for research, to the groups that aided the researcher in question to attain such summits?
In order to get 25000 euros tacked on to your salary per year, you’d have had to win things on the order of the Nobel, the Lasker Prize, the Grand Prize of Inserm, the Turing, Albert Lasker, Wolf, Gold or Silver Medal of the CNRS, Japan, or Claude Levi-Strauss prizes/medals.
In order to get 3500-12000 euros per year for five years, the field widens to include things like a number of prizes distributed by the Institut de France, the Emil Jungfleisch award, the EADS computing award, and a bunch of other prizes with money already attached that I hadn’t heard of.
Then, for up to 15000 euros as above, the International Prize for Biology, the EMBO gold medal, the Carlos Finlay prize in microbiology, the 2004 award from the Foundation for FIghting Blindness (WTF?!), the Linnean medal, and others, qualify.
If you are an EMBO Young Investigator, or, as I once was, awardee of the Avenir program (but – within the last five years, otherwise you’re no longer excellent), recipient of the Carus medal from the German national academy of sciences, the Paoletti prize or other slightly more accessible to common mortals award, you can aspire to 3500-6000 euros per year over the five years.
And if you saw the system used to rank people as far as how much, you’d weep. Among other things, you need to enter your best five articles in the last five years in first/last author position, complete with Thomson impact factor… I’ll leave it to your imagination.
Of course, a number of deserving scientists have reportedly refused the principle. But when you’re talking bribes, how long do people hold out on principle?
Couldn’t agree more – it’s a tragedy that the Wellcome have instituted this policy change at a time when the rest of the funding opportunities (i.e. the UK Research Councils) for people who might just be starting out are drying up as well. I’m a post-doc who needs to get a first grant in the next couple of years if I’m going to continue in science, and I honestly don’t know where it’s likely to come from!
@m_wall
By the way, there is an excellent blog post on the above topic of rewarding people rather than projects, but it’s in French.
I’d like to translate it, but I’ll contact the author first perhaps and maybe reconsider for purposes of time. It brings up that in addition, the fact that 19% of the people who got the prime d’excellence at the CNRS were women, whereas 32% of its researchers are women.
Only 1 in 5 of the CNRS’ researchers even applied for it (not clear whether that was self-selection, contestation or what, but probably both) – but in theory, if you’re even recruited (ie. tenured), you’re already excellent.
The author also pointed out that scientists, more than anyone else, are not subject in general to considering themselves excellent, because by the nature of our work we must be modest in the face of our ignorance. The author, blogging for Le Monde, attended a conference by Laurent Schwartz, one of the top male mathematicians of the 20th century, in which he confessed to being terribly subject to imposter syndrome…. of course there are all personality types in science, but do we want to selectively encourage self-aggrandization at the expense of the also-productive majority?
Hm. Must not have closed a bracket there. Sorry – perhaps you could fix?
no worries – just fixed it
Thanks to all for comments,
I should add the caveat that I do think sometimes it is ok to fund people – meaning new folk; I think the idea of ‘start-up’ money such that they have in the states is a good idea – that is when you first put a new academic in post – you give them some money to prove themselves (the institution that is) after which point they can write for money from funding bodies. You have to support people to get started – or they may have the most fantastic ideas but never be able to get off the ground…
Completely agree, but my take on the way things are going is even grimmer. I tweeted this yesterday:
– to which one of my scientific friends (a young-ish tenured lecturer in a biomedical subject) responded:
The Wellcome’s New Grand Idea slightly reminds me of the time (mid 90s) when the MRC famously decided not to have any projects grants, though then they made everyone become ‘networks’, or ‘initiatives’, or something. This was apparently George Radda’s wheeze, for which he is not remembered kindly in the community. I wonder if Mark Walport at the Wellcome will be similarly remembered for this latest rush of blood.
—————————————————————–
Absolutely – see my extended riposte to Sir Paul Nurse et al a year plus back. But it is pretty clear ‘they’ are not listening
I think it’s a shame this debate isn’t gaining a higher profile. Where are the science journalists when we need them?
I entirely share your concern that too much emphasis on ‘people, not projects’ risks freezing a complex, dynamic and highly successful research system into a static, unchanging hierarchy. Where will the leaders of the future come from?
It seems to me that policy makers, funders and the leaders of research organisations tend to underestimate the extent to which the system works as a system precisely because the different parts are interdependent. ‘Excellence’ (which is far too narrowly defined, but that’s another debate) is assumed to be some innate characteristic of specific individuals (or sometimes groups or organisations) rather than an emergent outcome of the way the system as a whole works. We can’t know how much variety we can cut out of the system through concentration of funding before ‘excellence’ is damaged, but I share your concerns that we may be about to find out.
However, I do have a little sympathy with those policy makers, research leaders, etc, because I think they genuinely believe they have no alternative. This is because they exist in a monoculture in which they rarely if ever hear an alternative viewpoint. They are mainly to blame for that themselves, of course, but we all share some blame. It is good to see this debate being stirred online but we need a much more visible debate within and beyond the science community about all these issues.
@Austin – also agree with most of what you say – with the exception of the pointless dig at admin/policy folk.
Sorry, Kieron. It’s a scientific reflex…! I don’t think it in dispute, though, that jobs for scientific administrators (I’ll leave policy out of it this time) have expanded dramatically – and I mean both at the funders and in the Univs.
@Austin – Cheers! I tend to agree with Ben Goldacre on administration – if they are there to support the medic (researcher) to concentrate on their job, then that is surely a good thing. Of course not all administrators are up to the job any more than all academics are, and the same must go for staff in funding agencies, etc.
And we can all identify crazy management information demands that rain down on researchers and academics – but in my experience these come from senior managers (who are, of course, senior academics), frustrated at their practical inability to herd cats when the macho discourse in their management monoculture is all about ‘strategy’ and ‘implementation’, not from the poor old administrators, who are generally just the messenger.
Kieron, science journalists aren’t paid to cover science policy and funding issues. They are paid to cover science. For good or ill, most consumers of science journalism want entertainment. New Scientist (full disclosure: I’m a consultant) is a “leisure read”: most people read about black holes and quantum theory because they are curious to know about this stuff, not because they feel they ought to know about it. The most popular posts on Ed Yong’s excellent Not Exactly Rocket Science seem to be about strange animal sex. It’s not that different to people reading Heat or Grazia because they are curious about the lives of celebrities. Different strokes for different folks (and I don’t pass judgement on either group).
I am genuinely interested in science policy and funding (which is a bitch because I can’t get paid to write about it). But the issues discussed here are so complex and nuanced it is difficult for a science journalist to tell what’s wrong with funding good people, or at what level the (possibly good) idea of funding provably clever, motivated and insightful researchers becomes a problem. If the journalist can’t come to a conclusion, he/she can’t write about it in a way that’s compelling and effective.
Also, framing the debate as people vs projects is a false dichotomy, suggesting that we’ll either get the best people or the best projects. When I’ve met the “best” people, I’ve generally been impressed by their openness to hearing about any good ideas, wherever they come from. I’m not convinced that funding people is going to mean that a whole slew of deserving projects are suddenly going to fail to get funding. Obviously not everyone can be funded, and not everyone at the top will be open-minded. But, on the whole, science won’t grind to a halt.
I also think it’s worth remembering that Walport was a junior scientist once. Nurse spent years doing poorly-funded experiments with barely-functioning apparatus in the wilds of East Anglia. These people were not always the elite. Though it’s far from perfect, science does seem to work fairly well as a meritocracy. I’m not sure that the new system will stop early careers in their tracks: the elite people will still want to hire the best young researchers.
The problem is the pyramid scheme that is science. If you really want to achieve higher profile coverage, angry, frustrated young scientists at the bottom of the pyramid should start throwing eggs at the “elite” that are being talked about here. Or unionise. Demand more accountability. Or find venture capitalists that will support pure science (a la Lazaridis) and opt out of the current system. Hit Walport with a flour bomb (and invite the press along). Until that happens, I’m afraid it will remain an internal issue.
As perhaps it should. I’m doubtful that the general public will side with young, unproven scientists. Given the popularity of The Apprentice and Dragon’s Den, I’d bet they’d see the elite as deserving to be in control of EVERYTHING.
This is an interesting point you make about Nurse not always being elite, no of course he wasn’t, but he seems to have forgotten this.
The real point I think we need to look at is that we should fund ideas not people per se. This elite scheme (at least the way that I read most of the arguments for it) seems to be focused toward funding people detached from ideas themselves. And while the rise of people like Paul Nurse from the ‘wilds’ of East Anglia makes for a great story and is a good story, he was funded eventually for his ideas not because he is Paul Nurse. And yes the great and the good have great and good ideas but they need to keep doing so to be awarded that funding. So the idea of just funding people detached from ideas in this manner is what I think is dangerous.
One thing that puzzles me is, aren’t the “elite” largely in control already? Is there a big difference between asking the elites to sit on peer review panels for grant applications and giving them the money to hire people with impressive ideas and track records?
I would argue that Nurse hasn’t forgotten how hard it is to be at the bottom. I’ve only discussed this with him once (when I interviewed him for the New Statesman), but this seemed to be very much on his mind.
Yes and no – not everyone that sits on a panel is one of the great and the good – there are many academics which participate in reviewing grants; and not all of them are FRS elite researchers.
Yes there is a huge difference! I want money for my own independent research – i don’t want to just be hired as a post-doc for an elite person, killing my own research ideas.
I am not elite, and fortunately I have just gotten funded, but if all of the money out there only went to people who have already ‘made it’ I would never get funding to even try and get off the ground for my own ideas…
This is what i mean about it being about ideas not people
“so complex and nuanced it is difficult for a science journalist to tell what’s wrong”
Er, worse than nuclear physics? Maybe try learning and talking to experts like Kieron (as I know you are quite able and capable to do).
As for the entertainment issue – yeah, I agree with that. But as with every time that argument’s used, this is always just the newsdesk’s idea of what the readers want. Why not try a bit more imaginative writing and see?
I’d say that this particular issue is quite niche though compared to other science policy issues, probably more for the blogs (which could include ones in mainstream media – SWord or Guardian Sci Blog, Eureka covered this very well when it was an active blog).
I’d also like to see more science policy issues covered by political beats.
p.s. the sarcasm in first few lines probably comes over more rudely than it was meant. I don’t want Micheal or any other sci journalist to take offence to that (though the point that they could nerd up on sci policy studies a bit more stands).
Yes, much worse than nuclear physics! There’s a consensus there: it might be complicated, but I can follow the argument and evidence to a definite conclusion.
I would love to spend time talking to Kieron (and you) about sci policy, but it would have to come out of my spare time. I read as much as I can (and participate in debates like this when I feel I might have something useful to offer), but I’m afraid it’s rather self-indulgent: if I write thoughts up for S-word, I don’t get paid – there’s no budget for it.
As for judging what the readers want, it’s clear from reader surveys and feedback that it’s a very small proportion who are interested in science policy. Magazine/website space devoted to it is a reflection of real appetites, I’m afraid.
Well, maybe niche blogs like (and perhaps clever books like yours) this will help spark an appetite…
Great post Sylvia – what an interesting perspective to put a Lamarckian twist on it.
I totally agree with Kieron that this issue deserves wider discussion among the scientific community. Although to be fair, I think these discussions are going on in common rooms and conference halls around the country (it was certainly discussed — along with other funding woes — at the last scientific meeting I attended, earlier this month).
My take on it — from back in Jan — is very much in sympathy with your views. As you and others have said, there is merit in funding people as the WT have opted to do (they are a pretty intelligent organisation). The problem lies in their wholesale switch to a people-oriented model — this approach discounts the complexity of the scientific enterprise in the UK, which relies on much more than just a few key labs. If the rest of the funders in the UK were to follow in the Trust’s footsteps, it would be seriously bad news. That’s why this issue needs as much of an airing as possible.
@Michael – thanks for the response. I appreciate all you say about the focus of most science journalism – and about the complexity of the issues at stake. I still think that if, as @Stephen says above, people are starting to discuss this in common rooms and conference halls across the country, then it is a shame that isn’t currently being reflected in that small part of science journalism which does cover policy issues.
PS just noticed @Alice’s comment. Is there a problem that science policy falls between two stools, and ends up being covered neither by science journalists nor by policy ones?
Kieron, I genuinely think that, rather than there not being journalists available for it, there’s little appetite for it. I get the feeling that the general sentiment is that intra-academia stuff (what’s being discussed in the common rooms and conference halls) is as uninspiring to most people as when journalists discuss (in pubs, rather than common rooms) what makes a really good feature story. People want there to be really good science, really good feature stories, and really good sausages, but very few people can stomach the details of how they are made.
An interesting discussion, and an interesting subchat about science journalism. I take the view that science journalism is far too deferential to science. The dominant mode is as mouthpiece for scientists. This means that asking hard questions of science is rarely done by science journalists, who complain when it is taken out of their hands and made ‘political’. And all that.
But I think we’re being too down on science journalism. I’m constantly impressed at how much science policy is crammed into front half of Nature. I would hope it was there because its readers found it interesting.
There was a time that practically all of the news coverage in the front half of Nature was policy. And very dull it was too. I think the policy news has become more interesting because it’s cut with stories about research – and, in all cases, about people, whether in connection with policy or research. It’s stories about people that other people find interesting.
@Jack – I too had the front part of Nature in mind. And we musn’t forget that Research Fortnight often gives a platform to critical voices – it’s just that I doubt whether many in the science community (as opposed to those in admin and management) read those pieces.
@cromercrox – Ha! I know what you mean but I think that also reflects how we tend to discuss these issues. We have to resist the tendency to restrict it to “This report has come out”, “That scheme has been launched” etc. I think Nature has gotten much better at this over time (especially with the comment pieces) but I guess there is always room for improvement.
“People want there to be really good science, really good feature stories, and really good sausages, but very few people can stomach the details of how they are made.”
I couldn’t agree more. It’s very much how I feel about the morning economics reports on most classical music stations I’d have liked to listen to during my commute.
I’d like to add my support for a policy quite different from that adopted by the Wellcome Trust.
I think that almost all the money should be spent on smallish responsive mode grants. In a fairly extreme version of this, no lab would have more than two postdocs, and their existing funding should be taken into account much more than now. It would also be good if no more than two full papers a year could be published. If these rules were adopted, the PI might actually have time to do some science, rather than spending all their time writing grant applications.
When I put these ideas to Mark Walport, he listened politely but seems adamant for the moment. I did some simple calculations about the dire consequences of a 1 in 8 funding rate, at http://www.dcscience.net/?page_id=4473#030611
Hmm, I’m not sure you should be that limiting, though I like the idea of a greater spread of funding. Different disciplines publish at different rates and need different levels of support, so it might be too much for some and not enough for others… But interesting view
Of course there would have to be some flexibility in the numbers. But it would have two advantages.
First it would end the present situation. Too often. PIs have rather little connection with the lab, not just that they don’t do experiments themselves but sometimes they don’t even have time to contribute theory or ideas
Second, a restriction on the number of papers would do much to prevent people publishing half-finished work (and wasting a lot of time trying to shoe-horn it into Nature). It would reduce the number of mistakes and it would reduce the endless proliferation of “peer-reviewed” journals which will publish just about anything. Fewer. more complete papers would make life easier for reviewers and it would stop largely playing citation games. If we were really lucky, it might put bibliometritions out of business altogether.
In any case, if you want more grants, the only way we’ll get it in the short run would be to reduce the size of each grant.
I don’t think limiting the number of papers out if a grant would improve their quality, largely because I don’t necessarily agree that more time makes a difference to those who are conscientious enough publish quality work anyway. Nor do I think it would make any difference to citations, etc. If you are someone who publishes ‘half-finished work’ I think you would just publish less ‘half-finished work’ –
We can agree on your last sentence anyway.
What worries me is the harm done to science by the publish-or-perish mentality. If you spend three years on a project you’ll get hammered by referees when you apply for your grant for a “”poor publication record”, even if the outcome ends up in Nature. Of course that is a result of lazy referees who glance down a list of publications, doing little more than counting them, and noting where they were published. They should be trying to assess the difficulty of the project and the quality of the outcome but that takes too much time,
From the career point of view, it is very dangerous to spend three years on a project, Most people yield to the pressure to publish short incomplete snippets because they are under pressure to do so both from university administrators and research councils. The more they publish, being careful always to cite themselves, the more citations they get. It’s an iniquitous game that’s harming good science, and anything that can be done to restore honesty and sanity should be done. If you can think of any better method than limiting the number of papers, please tell us.
I have to add that one of the more gratifying experiences of my life was when (some time ago), I was on a selection committee of the Royal Society, I saw a candidate rejected for fellowship on the grounds that they had almost 500 papers. Anybody who has that many is not really thinking. Clearly they have been attaching their name to work they have had little connection with. They are playing the citation game. In this instance, they paid the price, but usually they don’t.
I agree that the pressure for Nature papers etc is awful. Being a young academic I feel this pressure myself and it makes me more high strung than I would like to be. I don’t like having to worry about things like my h- index, etc. And I try not to so much, but I often feel like I have to because if I don’t I will never get funded…
I don’t like this, but I do try to keep my scientific integrity – hopefully I am successful at that at the very least. It is a good thing to remember…
As an “intermediate” career person – slightly too old now to qualify for young investigator things, and that’s just as well since I now have tenure, I would vehemently agree with David on a number of points.
“From the career point of view, it is very dangerous to spend three years on a project.”
Absolutely true from personal experience. And then it’s very hard to get the one painfully produced but excellent paper noticed with all the rest of the production out there, a result of the part-finished work policy bemoaned above. I re-read some papers from the early 90s and am amazed once more at the quantity of high-quality work that went in, a third as many experiments now would get a paper published. The minimum publishable unit has now really been defined.
“They are playing the citation game. In this instance, they paid the price, but usually they don’t.” Absolutely. I know a few of these personally, like them as people, but am saddened by this approach, which is damaging to the scientific enterprise overall, because it reduces the diversity of the intellectual ecosystem. But it’s a selection pressure like any other. Adapt or die – but I regret the creativity of a lot of my colleagues who “die” under these circumstances.
Two full research papers a year in biosciences seems quite adapted to my field. And perhaps a review article on occasion, which is as much work, but it is clearly derivative. If you collaborate on other work, your name would be on other papers, though. How much of that middle-authorship would you find acceptable, David?
@Heather
Any number that one dreams up is entirely arbitrary, so your question has no firm answer, I guess that a lowish-limit like two would apply to first or last authorships, and wouldn’t include reviews. Also it would have to be averaged over several years. One project that I was involved in gave nothing that we thought we wanted to publish for four years. I’m sure a lot of people would have been less sqeamish about publishing half-finished intermediate bits of work, We chose not to, and the work ended up as a Nature article. Nevertheless grant reviewers who can count but not think, are inclined to point to gaps in publication like that as a reason not to fund you. Once having solved the problem, a whole backlog of data became publishable. Avereged over five or six years, it wouldn’t amount to more than two a year, but success being stochastic, the publications can be quite bunched.
Concerning collaborations, I guess it would depend on the nature of the collaboration. It’s not uncommon for someone who gives you a bit of a reagent that they already possess to demand co-authorship, Sometimes co-authors have done next to nothing apart from being around when the work was done.
This sort of “guest authorship” is a another undesirable product of the citation culture, and it’s dishonest. but it isn’t without its risks. When interviewing for jobs, we used to ask people to submit their 3 or 4 best papers. That’s not uncommon. But then I’d read at least one of them quite carefully and ask the candidate questions about the paper in the interview. It was surprising to find that, quite often, people knew very little about a paper that they’d submitted as one of their best. Close questioning would elicit something like “well actually I just provided Figure 7 and I haven’t really read the rest”. Such people didn’t get the job.
Arbitrary though it is, I think that doing something like I suggest would improve the quality and honesty of science. It would also reduce the enormous burden of having to read six half-baked papers in place of one good one. And it would also increase the quality of peer review. It has become hard to find people to review properly the deluge of papers that has resulted from the publish or perish culture. With fewer papers the job could be done much better.