Research Support.

There is a lot going around the science blogosphere right now about how to restructure the NIH, and by extension other large federal funding agencies, like NSF and VA ORD. As usual, the heavy hitters over on twitter are on it. If you’re interested, you should read Michael Eisen, and Kate Clancy and several posts by Drugmonkey. I don’t have a great deal to say on the subject personally, because my position at PECMC is hard money. I won’t be being asked to submit and survive off of grants. But I have opinions of course.

My opinion is that the system, the Complex System that is the research apparatus of the United States, is expressing features of the way it was designed. Unintended features, yes. But these are consequences of how we built the system that are now entrenched. Complex systems can be extraordinarily resistant to global level change. Consider the solar system: it is regularly bombarded by enormous perturbances. Meteors hit planets, cosmic radiation, comets collide with gas giants, and yet, here we all go circling the sun as usual. Consider economies: real estate bubbles and stimulus and taxes and banking crises, and the differences we see are confined to a few percentages change in growth, up or down, acting over years. Individuals may be devastated. The system goes on.

One of the first problems, which arises from a noble ideal, is the federal funding of science. It is indeed a good thing that governments, with their vast wealth, should dedicate some of that to research. No other entity can marshal the resources that a government can. However, it took essentially no time at all for universities to discern that this meant that they got to stop paying for research. Most universities will give startup packages and a few small internal grants. But for the most part, science professors must get external funding, usually government funding, or they lose their jobs.

This means that professors are required to spend a great proportion of their time seeking funding. Professors earn jobs based on how good (or how prestigious) their science has been as a graduate student and a post-doc. They retain their jobs based on how good they are at getting grants. Thus, professors cease, almost immediately, to be scientists. They become managers and grant-writers. Their graduate students and post-docs do most of the science. So one consequence of federal funding is to take some of the greatest scientists in the world, and relieve them of their scientific duties. We built this.

Federal funding is apportioned, many will say, according to the quality of the science in the proposal. But this isn’t true. Or rather, it’s not completely – or even close to completely – true. Grants are scored on many criteria which vary from institution to institution. One of the criteria, for example, is Investigator. Many people use journal quality as a proxy for article quality when evaluating an investigator they don’t know. So, this creates a scrum for the “best” journals. Niche journals that will publish good work in a particular field are often eschewed by authors because it’s far better to get a paper into Cell, or Science, or NEJM, or The Lancet. This leads to all sorts of glamor-publication problems that result in a lot of great science going unpublished. (Ask Michael Eisen what he thinks about this for a long, well-informed lecture.)

Another criterion grants are scored on is Environment. While this is supposed to describe the collaborative environment and quality of the facilities that will be used to conduct the work, what it often comes down to is University. Again, quality of school is used as a proxy of quality of intellectual environment. The same quality investigator with the same collaborators in an identical laboratory, but one is at Washington University and one is at the University of Missouri St. Louis? Guess who’s getting the higher score.

And with federal funding so scarce now, where perhaps 10% of grants which receive scores (already only about a third of those submitted) are funded, many, many fine scientists are leaving science. Either because they become discouraged and aren’t willing to accept low pay and terrible job security, or because they lose their jobs when they are unlucky in the grant review or glamor-publication lotteries. And they are lotteries. Excellent, excellent science is being rejected for nothing but luck. The difference between a score of 5% and a score of 15% is meaningless. It’s random, or based on false signals. Or on old-boys club politics.

The only solution I see is for universities, especially the big private universities which are absolutely swimming in money (In the past 10 years, hundreds of millions of dollars of construction has gone on at Washington University. Opulence that reminds me of a despot’s palaces.), to return to supporting the professors. There need to be more internal grants. There need to be more endowed chairs that can be got at the associate level. There need to be paths to tenure that don’t include federal funding. Or even external funding at all.

Success at the university is driven by one thing above all others, and it isn’t (quite) money. Prestige. And external funding has been exalted as the highest marker of prestige. “Good scientists win federal funding.” “Good scientists publish in the best journals.” “Good scientists graduate the most PhD students.”

It’s all bullshit. Good scientists discover exciting things that change the way we see the world, the way we interact with the world, and the way we live. Good scientists develop actionable hypotheses and then design experiments to test and confirm or refute those hypotheses. Good scientists disseminate those results as clearly and broadly as possible, so that engineers and other developers can implement improvements based on a deep understanding of nature.

The system, as we’ve designed it, is an impediment to science. It was built with noble intentions. It was built on good ideas. But systems are bigger than the intentions and ideas and even the components of which they are constructed. And if we want to change the global behavior of this system, we need to disrupt it entirely. Many good ideas are being floated, like the Public Library of Science and BMJ Open. These efforts have their own problems too, in my opinion, which I’ve discussed before. But they are at least bold, large-scale efforts to make fundamental changes to the dynamics of the system.

Something about university-conducted science is going to change, and soon. Universities, with their aggressive expansion, abandonment of the student-professor dynamic, and total reliance on borrowed money (whether federal or student-loan), are in a bubble that cannot help but pop. There is going to be a disruption in the American higher education system. The question is, will it be a negotiated and guided change, led by competent people with a vision for the future, or will it be a catastrophic failure that results in the extermination of many fine institutions? The way it looks now, I’d bet the latter.