Is progressivism the biggest threat to science?

In the latest New Scientist, Alex Berezow and Hank Campbell attempt to redress what they see as an imbalance in perceptions of how political views affect attitudes to science. It is widely held today – in the wake of books like The Republican War on Science, Christian conservative opposition to evolution and well-documented interference into science policy by successive Republican administrations for ideological reasons (not to mention recent Tea Party lunacy with regard to the biology of human fertility) – that ideological opposition to science is a right-wing affliction. Berezow and Campbell (both proud conservatives; Berezow edits RealClearScience, an offshoot of RealClearPolitics, which aims to counter “the bias in media against conservatives, religious conservatives, [and] Christian conservatives”) argue that ideological opposition to science is not restricted to the right.

They are surely right, though the primary example they give seems more to involve petty political point scoring and a failure to conduct proper cost-benefit analyses than ideological opposition to science. Knee-jerk opposition to GMOs is more common on the left than the right for instance (the qualification ‘knee-jerk’ is an important one: there are good reasons to oppose particular uses of GMOs and to be cautious about all its actual applications, though there are no good reasons for thinking that GMOs might not make significant contributions to human and non-human flourishing). In the end, though, Berezow and Campbell exemplifies the very faults they rail against: allowing ideology to trump good sense.

Once upon time, ideological opposition to science seemed more common on the left than on the right, especially in the form of a muddled relativism. Today, to many commentators, it seems that things have reversed and more, or at least more practically significant, ideological opposition to and downright distortion of science is to be found on the right than the left (after all, opposition to the science of global warming is almost exclusively to be found on the right, and there is a strong case to be made to saying that global warming is the single most significant challenge facing humanity). Berezow and Campbell will have none of this. No, for them it is ‘progressives’ that are the major problem:

Of all of today’s political philosophies, progressivism stands as the most pressing problem for science. Progressives, not conservatives, are the ones most likely to replace scientific research with unscientific ideology.

Might they be right? I doubt it very much, but were I to try to make the case, I would look for data. In God We Trust, the Americans like to say, and no one more than the Christian conservatives that RealClearPolitics represents. But everyone else must bring data. Without data, without evidence in the broadest sense of term, you’re simply expressing your political prejudices. And that, whether it comes from the left or the right, is the way in which ideology distorts reason.

  • Facebook
  • Twitter
  • Reddit

8 Responses to Is progressivism the biggest threat to science?

  • Nickolas Schaffer says:

    Their use of the term “progressive” is questionable and rather pointless in some of the examples they offer. Are the anti-vaccine crusaders really identified by most informed people as “progressives”, or just as “nutters”? Are the PETA activists genuinely progressive, or more akin to an irrational religious movement? Most Green politics may position itself on the left, but Green claims and policies are actually very often subject to rational and scientific criticism, when deserving of such. This claim that they get a “free pass” is paranoid Righty nonsense.

    If there is a “bias” against conservatives amongst the intellectual establishment (in both sciences and humanities) it’s simply because conservatives so often make a habit of being stupid and wrong, as Berezow & Campbell demonstrate in this article.

    • Dave Frame says:

      Nickolas wrote: “If there is a “bias” against [Group X] amongst the intellectual establishment (in both sciences and humanities) it’s simply because conservatives so often make a habit of being stupid and wrong, as Berezow & Campbell demonstrate in this article.”

      I find that a ridiculous thing to say. The scientists I know who self-identify as conservative are just as competent as those I know who self-identify as progressive or liberal. Possibly more competent, because their work – when it intersects with issues of policy – is often subject to a higher standard of criticism.

      • Dave Frame says:

        …sorry – I hit reply too fast. The reason I substituted [Group X] for conservatives was to request that you try a few other groups in society in place of “conservatives” and see how sweeping and jarring the claim feels. I’ve posted on this site before about how I don’t think there’s any normative justification for the (observed) progressive bias in universities – to me claims like the one that Nickolas makes are just as offensive* as similar claims made about other groups on campus.

        *Or not. But I don’t think we should think of political diversity as being different from all the other forms of diversity universities are so careful to nurture.

  • Owen Schaefer says:

    Is it even possible to come up with reliable, unbiased data to back up Berezow and Campbell’s claim, or the counterclaim that conservatives are more generally a threat to science? Berezow and Campbell could provide their list of alleged progressive distortions, then progressives respond in turn with their own list of conservative distortions. But two immediate problems emerge. A) the composition of each list is easily prone to bias (e.g., Berezow and Campbell claim that progressives are more anti-vaccine, but a couple of polls indicate the concerns cross party lines: ); and B) it is entirely unclear how to tabulate and weigh up in an unbiased fashion the various infractions, to determine which are more egregious.

    I suppose one could try to use generic polling data. So, one recent analysis found that today US liberals do indeed have much greater confidence in the scientific community, though a few decades ago, conservatives had more such confidence than liberals. ( ) But Berezow and Campbell (or liberals of yesteryear) could simply respond that conservatives’ lowered confidence is entirely justified, and perhaps the biggest threat to science is unquestioning assent to any new scientific findings instead of robust skepticism and demand for more inquiry. So, while data is generally good for claims like these, I suspect they’re going to be generally unverifiable from an empirical standpoint.

  • Neil Levy says:

    You are right, Owen, that’s providing the data would be difficult. One comment on your last point: the comeback you suggest they might make, claiming that lowered confidence in science is justified. Given that science is by far our most successful epistemic enterprise, any such claim must be treated as les reliable than the science it aims to oppose (unless, of course, it is backed up *by* science – that is, data). So they can make the claim if they like; we shouldn’t be impressed.

  • Anny Mouso says:

    To bypass the problems Owen points out, I would suggest that the best way to go about gathering data would be to survey the scientific community itself rather than population at large.

    A survey guaranteeing anonymity which asks working research scientists to outline a research project which they feel would be worthy to undertake on scientific grounds but which they would not dare to put forward for fear of damage to their reputations or careers would, for example, be a good starting point for anyone wanting to study this further.

    Granted there are many problems with what I have just outlined, not least of which the fact that the answers would be as much a sample of the ideological bias of the surveyed researchers themselves as they would be of possible ideological pressure on their professional practice as scientists, but it would have the advantage of providing a databe which could then be interpreted by different researchers working with different definitions of “progressive” or “conservative”. At the very least it might offer an interesting snapshot of the state of scientific research at this point in time. Find out which questions scientists are afraid to ask and you’ll get a fairly good idea of which kind of ideological pressure they are under, as well as from which side of the political spectrum that pressure is coming from.

    On this subject you might find interesting Destructive Trends in Mental Health. The Well-Intentioned Path to Harm, edited by Rogers H. Wright & Nicholas A. Cummings (Routledge 2005)

  • Neil Levy says:

    Anny, that data might be worth collecting but I doubt it will give us the full story. The problem is that it assumes that pressures from the left and the right are symmetrical, and take the same form. A number of scientists have complained that ‘political correctness’ has prevented them from studying various topics (Steven Pinker has said this, so has Simon Baron-Cohen, as well as David Buss and Thornhill and Palmer). They all can be taken as claiming that pressures from the left made them fearful to study something (I take all these claims with a whole salt cellar, since these people have built a public profile on the basis of the very questions they claim under pressure not to investigate, but set that aside). Scientists who complain about distortion from the right don’t claim that they are afraid to study certain topics; rather, they complain about the political spin and misinformation that surrounds public perception of their work (here global warming is the central case). If that’s right, then the data might tell us about ideological pressure from the left, but not about whether there is more pressure from the left or the right.

  • Anny Mouso says:

    Neil, I agree (and generally speaking I’d say no single study ever gives the full story about anything – something the media seems to find particularly hard to accept). On the whole I think pressure from the “left” tends to prevent scientists from carrying out research or puts pressure on them to ignore data that contradicts ideological expectations, whereas pressure from the “right” affects the public understanding of science. But that of course is a sweeping generalisation of a very complex situation, and to some extent limited to an American/European context – I think you might find a very different picture if you look at the situation in other parts of the world.

    In any case, I can certainly not think of any way in which the question of whether there’s more pressure from one side or another can be formulated as a question open to scientific study, and I don’t think the question is relevant in any case. It’s a bit like asking “what is more dangerous, a lion or a shark?” To the man facing a lion, the lion is certainly more dangerous; to the man swimming with sharks the sharks are far worse. The point is to avoid both, because both are bad for you. And I do think there is a great deal of complacency amongst “progressives” that they stand for science whereas in fact their understanding of it is for the most part as poor as that of conservatives, and their resistance to it just as virulent when it contradicts their own prejudices.