As a follow up to Sergey’s insightful post about the bizarre logic which accompanies the job hunt, I’ve been mulling over what drives all these dynamics. It looks to us that whatever the situation may be, at the end of the day hiring committees really only care about papers – how many you’ve got, and where they were published. Individuals circumstances really don’t seem to matter when there’s a glut of PhDs to choose from.
OK. A biologist’s perspective (the kind interested in ecology and evolution): why not follow this logic through – what happens when you take a population of scientists (say, those studying living systems) and subject them to extreme selection, all of which feedback on each other, and let the thing run for a few years? Here are some of the main selection pressures –
Abusive environments: We are used to hearing that one has to develop thick skin, and of course doing good science requires the ability to move constructively past criticism. Those of us who are too sensitive (perhaps cultivated through a lifetime of other experiences) may not fair as well. But abuse is another thing. Kate Clancy has made this rather shocking disclosure on the horrific things people have to endure in the course of doing fieldwork, including but not limited to sexual abuse – particularly against female graduate students, and propagated by their seniors. Add this the myriad other challenges for women in science, and is it really any wonder that we leave. Turning this on its head though – what incentive is there to stay if you have abilities that would allow you to succeed outside science? In a fascinating reversal of the traditional thinking, there’s some evidence that people with high verbal as well as mathematical ability tend to leave STEM careers. This disproportionately tends to be women. This self-selection no-doubt makes it a very difficult dynamic to reverse. Clearly there is selection against individuals who can’t take abuse, women in particular in subtle and not-so-subtle ways.
Clancy’s observation that the establishment seems to value science over scientists, is astute. This transformation is even evident symbolically. For instance the logo of NSF, which used to have a globe encircled by a ring of people holding hands, has morphed over the years to something more abstract resembling the spokes of a wheel. Which brings me to the next point.
Funding constraints: Of course being able to win grants is an important skill. But what happens when the playing field isn’t level? If it’s ideas we value over scientists, then the amount of money we allocate to those ideas (and endeavors) must reflect how much we value them. The new congressional budget came out last week. NSF got a nice increase of 4.2 % this year. But its $7.1 billion is still just a quarter of the NIH budget, which is $29.9 billion. NIH funds medical research largely with relevance to humans, NSF funds just about everything else to do with our planet (little things like how ecosystems work, how we could better feed ourselves, why other species behave as they do, the ocean, how to create new materials…). Here then is how I picture our priorities:
(Of course – for context, science funding is still tiny relative to how much we spend on shaking sticks at other countries but let’s leave that aside for now).
This asymmetry is probably the basis of some of the greatest scientific inequities. What if we were to throw that as much money at understanding how the world works, as we do at understanding how our guts work? I bet that would yield some insights. There are problems though – studying the world is inherently messy. Looking around it is exceedingly evident to me that biomedical research labs are generally richer than, say, ecology labs. I know people who spend more money on coffee makers for their labs than it takes to run my field site (probably several sites combined) to run for a year. This has now been quantified; as it turns out, there is an ever widening gap between the haves and have nots in science as much as elsewhere in society, where a small percentage of “stars” command a disproportionate amount of resources – and thus, publications.
Since these stars and their progeny are likely to populate departments, and sit on hiring and grant committees, doesn’t it seem likely that both positive and negative feedbacks toward certain fields or entire disciplines will propagate themselves to their eventual domination or extinction? What’s evident is that with funding success rates that hover between 9-12% there is strong selection going on against the type of science that can be done, and that selection is decisively not random. How much variation in ideas and indeed scientific and intellectual linneages, can be maintained when forced to thread the eye of this needle each cycle?
Publications: Top tier journals try to determine in some sense the zeitgeist of science in any given era. Certain fields are represented more than others. This would be fine if those who publish elsewhere could still land a job. But if only those individuals with papers in those select journals are able to obtain interviews let alone job offers, then in effect, the journals (and journal editors) end up determining where science goes and not the practitioners.This is the tail wagging the dog. Of course, the “stars” will have pull with the journals, but this goes back to the problem of who the stars are in the first place, and what they value.
Speed: If pumping out publications is the goal, then it probably pays to study a faster system than a slow one. This works for theory too – you’re probably better off putting out volumes rather than stewing on an idea for a while and trying to get it right. Writing five short papers in the space of one. I think of this as the scientific equivalent of fast food. Taking that analogy further – wouldn’t that then lead to scientific obesity? Given that there are more papers published each year than anyone can possibly stay on top of even within one’s own specialty, why this unyielding push for quantity? Some side effects of this include what’s been called an “avalanche of low-quality research” and gaming the system – inflating one’s apparent contributions at best, dishonest at worst. Now opinions differ on whether there is such a thing on too many publications, whether anything should be done about it, and if so what (great discussion here). But the reason of course is that the number of publications (even if baselines are field-specific) is a quantity that has no stable equilibrium – it will always pay to publish more than the average, and those who work on systems which allow them to do so will be rewarded. By sheer volume alone, if you put out more ideas, you’re also likely to hit on more good ones (or results that appeal to the field). The point here is not whether it’s better or worse to publish more vs. less, but that those who do the former will be favored.
Job: Academic hiring committees are so overwhelmed that they no longer have the time or desire to consider individuals as individuals. Of course it’s easier to skim journal titles, count papers, and tally grant totals on a CV in making the first cut of job applicants – it’s impossible to do otherwise when you’ve got 400 applications for a single position. Regardless of why we’re in this impossible situation and whether it’s ok, I wonder what this does to the scientific enterprise as a whole fundamentally. As in Sergey’s post, I am sure that people apply for these positions from all walks of life, those from affluent/academic backgrounds vs. not, the opportunities provided by state schools vs. private. I understand that it is challenging to evaluate applicants relative to their circumstances vs. their qualities on paper.
Yet, I wonder – don’t college admissions panels do this all the time? And don’t they do it for HUNDREDS of students EACH YEAR? I recall being told that the personal essay was the most important thing in the application packet,that each individual was considered as an individual. The hiring process, by contrast, seems all about having the right connections in a sea of impersonal CVs. Somewhere along the way, it became acceptable to reduce people to their CVs – something analogous to reducing students to their GPAs and SAT scores (which are at least standardized). Academia is probably one of the most rigidly stratified places I have ever seen where movement between tiers is probably limited by the system as well as candidates’ own preferences. If it is only these narrow qualities that are being being assessed, it feels silly to speak of promoting women in science, the importance of diversity, or any other lofty goals aside from what departments really care about: how much money you got, how many papers you have, and where you published them. What is the use of trying to promote diversity at lower levels, if no one cares about anything other than these narrow attributes when it comes time for the ultimate decision? Surely, when it comes to filling a faculty position, even more so than a college admissions slot, it seems only fair to devote more than a cursory glance at a CV before making the initial cut. The uncertainties of the academic job market, and the prime earning years that may be lost while waiting to land a job, it is likely that those with more pressing economic concerns are unlikely to stick around.
In summary, here’s where I think publicly-funded academic science* is headed: It will still largely favor men over women, and status quo over socioeconomic/cultural diversity. Stable systems are probably likely to be favored over unstable ones, lab over field, fast over slow (or careful and meticulous). Perversely, more expensive research will probably be favored over cheaper research. It will generally reward privilege over grit, favoring those who are carefully groomed and can play the system (or learn fast) than those who have pursued their own way. The only way to change these dynamics would be for “stars” and established scientists to start changing their own priorities. This seems unlikely to occur any time soon.
A more accurate title for this post might be “Are we evolving the scientific endeavor”. Selection can occur at so many levels – not just on individuals, but on entire fields, different branches of science. In departments where multiple fields within a single discipline must coexist, it’s difficult to see how those with fewer resources would not eventually be eliminated; likewise departments within Universities etc. This means that certain questions become un-askable. Once that happens, these tendencies will be reinforced. For instance, it is not enough to want to study something simply because we don’t know about it (the kind of grant proposal beginning with “Little is known about X…”). Naive maybe, but I thought that this was very the basis of curiosity-driven research. But this is of course the natural consequence of resource competition. In some ways, this is probably fundamentally changing the system; but in others, it strangely reinforces more of the same.
The disconcerting thing for me about all this is that I have long thought (again, idealistically) of the scientific endeavor as a way of understanding the world. A careful, evidence-based way at arriving at understanding for which just about any inquiry is fair game and anyone could make a valuable contribution. This has long ceased to be the case.
*Naturally, the next question is – are there ways to fund and do science? Of course there are several answers. I’ll leave that for another post.