Thursday, April 17, 2014

Playing in sand castles is no game! Funding science

There are rumors that the proposed federal NSF budget will cut some areas (cut not hold steady) by amounts well into double digits (like around 20%).  That's a permanent cut imposed over just one year, I think, on top of the steady-at-best budgets of recent years.  And a new commentary by Bruce Alberts et al. in PNAS bemoans the similarly serious situation in biomedical (NIH-based) research.  These latter authors make many or most of the same points that we have often been making here (and we were not alone by any means): these are not sour grapes rants but are widely perceived as truths about today's circumstances in science.

The points have to do with the poorly justified if not selfish excess production of PhDs, the hypercompetitive funding and publishing environment that eats up too much time while it stifles creativity, the conservative and cumbersome grant system, administrative creep and so on.

How did we get into this situation?

In a way we got into this situation because the idea of an ever-growing economy ran up against the real world (that, ironically, science is supposed to be about understanding).  We could and should have known this, but nonetheless built a sand-castle research/university welfare system too close to the shore, and now the tide of the inevitable is about to wash into or over it.

Sandcastle in Singapore; Wikimedia

We smugly expanded at exponential rates though any idiot (even a scientist!) knows that in the real world, as opposed to Disney perhaps, exponential growth must reach limits.  We behave short-term and totally selfishly, building our programs, training ever more graduate students, asking for ever bigger grants, bloating our administrations, being more hooked on overhead and soft-money salaries than a downtown druggie is addicted to meth.

This was a university 'bubble' that we built, and our society bought into it.  Now we're getting our comeuppance.  It's too bad because the most affected people will be the younger scientists who are innocent of the greedy behavior we elders indulged in during our careers.  It is we who deserve the slap on the backside, but the bruises will fall on our students--is falling on them.  There are not many university jobs and in many fields of scholarship, including hard-core and softer science, as well as the non-STEM subjects, there is a t-choice: a taxi-driving jobs compete with the prospects of a tenure-track job.

Universities are, often cravenly, saving money by denying tenure, hiring nearly unpaid adjunct instructors (but not reducing tuition accordingly, of course), and labs are letting staff off (to go compete for taxi licenses) because even some Dominant Baboon scientists can't get enough grants to feed their mill any more.

Now, we know that nationally, our Wall Street oligarchs treated themselves to a massive recession of which we, not they, were the victims, and they are getting off the hook for their evils.  But even forgetting that, the economy has had its downturn, as economies always do (the cycling tide of exponential growth).  So there is a constriction being laid on top of the overtly exponential-growth behavior of our universities.

In a downturn, there is a legitimate need to sort out priorities, which is less needed when everything is growing like Topsy.  Some areas have to be cut if we are to salvage what's really important. We here have often written critically of the puffed up, incremental rather than creative blowing away of large amounts of funding for various Big Data projects.  We've said that funding cuts might actually be a good thing if they forced people to think about their science rather than just buy more technology.  And both NIH- and NSF-related fields are guilty of devouring logs and spewing out sawdust.

But in a humane society, as ours should be, there should be a phase-out period of areas that are not delivering enough goods.  In our current system, however, there is so much lobbying and jockeying and self-promotion that this is not likely to be a humane process.  This we think is especially so if the cuts are quick, hard, and without much warning.

Either we'll continue with the brutally intense competitive environment, hostile to constructive interaction, in which we are already immersed in many areas of university science, or we'll have to bite some bullets.  We need to train substantially fewer graduate students.  Tenured faculty may need to do more actual teaching ourselves (fewer TAs).  We will have to scale back our labs to have fewer post-docs and technicians, and may need to do more actual science ourselves. We may have to be more selective, and restrictive in what we do or propose to do.  Administrations will have to do with fewer administrators, fewer shiny new buildings or lesser office furniture, and less addiction to overhead. Medical schools may actually have to learn to pay their employees (rather than relying on NIH to do that).

These changes even if they occur won't help those we've already misled into coming into these fields in which it was not hard to see the impending crunch, even years ago: They are the innocent victims.

We think what is needed, if it were possible, is a frank but non-partisan national discussion of what kinds of science and scholarship are most important and to phase in more funds for those and less for areas that, no matter how legitimate, are just less vital these days or less promising of major new discoveries.  We should consider academic employment practices and things like tenure and job security.  If they have to change, it should be in a phased way and not be punitive the way it is becoming now.

Alberts et al. suggest that we train them, our PhDs, for jobs other than in academe.  That's a great point, but if it's just an excuse for us to keep recruiting the same number of graduate students, it's a selfish ruse to preserve our business as usual, because we'd just quickly flood these other job areas if we did that.

The golden days of science (and scholarship--not all the important things in life are STEM things) may not be over, if we can behave properly and leave our six-guns at the coat-check.  But it does not seem likely to be easy or, worse, free of partisan politics unrelated to science itself.

What are you supposed to think, if you're a new graduate student, or a recent PhD?


Manoj Samanta said...

I do not agree with you at all. We got into this situation not because of overgrowing, but because central planners not being able to estimate right places for demand and supply. The solution is to get out of central planning pronto, and then let the chips fall wherever they may. It will be lot less harmful than hearing the ideas of these delusional fools. They may be distinguished scientists, but their track record for central planning is zero to negative.

Ken Weiss said...

This is too big of a question for an exchange of this sort. Some large projects must be centralized (like the human genome sequence for example). NIH, NSF, NASA, and others do have panels of scientists to advise them of where to invest.

I lived and worked through the era when NIH and others became far more democratic in their peer review system, which was mainly based at the time on individual investigator proposals and was not plagued much by central planning.

But central planning did grow and perhaps out of control (Francis Collins himself has recently proposed at least some return to R01's).

But I also lived through the decades in which the university dependance on the drug of grants, and their endless growth ethic. You cannot blame the excess of PhDs or the hyper-competitive grant and publication system on central planner obtuseness, in my view. We are all to blame.

Manoj Samanta said...

"Tenured faculty may need to do more actual teaching ourselves"

In Feynman's book, teaching was described as the most stimulative part of his work. He discussed about how his efforts to make his work easily understandable forced him to revisit many complicated equations. These days, teaching is being presented as a crime - "oh those poor sobs will have to go back to teaching again".

That is another effect of central planners glorifying research and taking over the system.

Ken Weiss said...
This comment has been removed by the author.
Ken Weiss said...

I'll add something else. Big Science is a central planning artifact and certainly endows Big Labs who are then very difficult to pare back. We write about that all the time.

But those big projects do employ people, including graduate students and post-docs and so on. They often involve hundreds of investigators, each with a bit of the pie, and that includes many regular (but usually soft-money) faculty members.

So a return to RO1's would not necessarily solve the problem, because the same people under the same total (or less) grant system would still have to be 'fed'. And they still would be under pressure to train a bevy of their own grad students if they want to keep their jobs (and because med schools don't cover salaries).

So I really don't see that central planning is the main source of the problem despite the fact that it may not be a good thing.

Ken Weiss said...

I basically agree with this point, however! Teaching has a bad name at 'Research1' (pompously self-named) universities. That's why parents are sending students to be taught by poorly paid serfs. Of course this makes less sense in medical schools because there isn't that much teaching done (in non-clinical subjects). PhD faculty in med schools often complain about their teaching 'load' even when it's a few lectures basically meaning powerpoints of their latest meetings talk. So this is a problem.

As to Feynman, he himself often said he was not a good teacher (as I've read in his own writing) but also at least early in his career (but, I think, after his Nobel Prize) he was teaching intro physics to Cal Tech freshman. I wonder how often that sort of thing happens now?

Manoj Samanta said...

I lived under communist government for a large part of my life and understood the system inside out. The education minister used to live next door to us (literally). A number of other ministers lived within stone's throw.

What you have seen in academia is a giant communist system and the dying seems to be in the communist way as well. Central planners in a communist system cannot anticipate growth and so they overproduce things that are not needed. Typically they start by going into the most vibrant sectors, overproduce and then destroy the sectors. That means for a while, you do not see the effects of communism or feel like it is benefiting people, because the terminal effects come later.

> Some large projects must be centralized (like the human genome sequence for example).

Absolutely not. In fact, private companies sequenced the human genome before the government-funded entities.

Ken Weiss said...

There are different stories of the human genome sequencing and the private version used government-funded frameworks, I believe.

But this is neither here nor there. We face an overgrowth problem that many are now writing about, and it isn't clear how it will be worked out.