Sunday, July 18, 2010

On being a science warrior

This is my first contribution to IOaT, and I am thrilled to have been invited. Philosophy of science is a new discipline to me, having crossed over from evolutionary biology during the past few years. Indeed, this first post is about how interesting (and somewhat puzzling) it is to be considered too much of a scientist by my new colleagues in philosophy, and too much of a philosopher by my old ones in the sciences. I also feel a bit awkward writing a sort of defense of a review (actually two) of my recent book, Nonsense on Stilts: How to Tell Science from Bunk. Then again, that was the original reason I was invited to write here, so I’ll try to get over my own embarrassment and make what follows not about me but about a general point that I do think deserves discussion.
In a recent review of the book in the Notre Dame Philosophical Reviews, Barry Barnes accuses me of being a “science warrior,” echoing another negative review penned (keyboarded?) by Carlin Romano in the Chronicle of Higher Education, where the term used was “science patriot.” (In case you are wondering, there are in fact positive reviews of the book, for instance here, here, and here.) What both authors vehemently objected to was my apparently uncompromising attitude of defense of science at all costs, and “cocksure hubris” (Romano) in criticizing certain notions as bunk or, as the Bentham-inspired title of the book puts it, nonsense on stilts.
This struck me as rather strange for two reasons. On the one hand, the entire idea of the book is that the famous demarcation problem introduced by Popper — how to tell science from non-science and pseudoscience — is in fact immensely complicated, not the least of which because science itself is a heterogeneous enterprise with its own history of blunders (this is the part that ruffles the feathers of some of my scientistic colleagues, like Richard Dawkins and Jerry Coyne). I should clarify at this point that the book is aimed at the general public, not professional philosophers, and that a major goal of it is to help people appreciate how science and philosophy can work together to provide us with the best understanding of a huge variety of issues that are actually pertinent to our personal lives or to society at large (e.g., the non-existing connection between vaccines and autism, “skepticism” of climate change, and denial of the HIV-AIDS connection, to name a few).
On the other hand, just because the science-non science-pseudoscience landscape is complicated it doesn’t mean that the extremes aren’t easy to spot. Fundamental physics and evolutionary biology are solid sciences, no matter what some misguided philosophers may think of the latter. That is why, to Romano’s horror, I called Bruno Latour to task for his rather silly essay on Einstein’s relativity (where it’s clear that he doesn’t actually understand the theory), or I criticized Paul Feyerabend for cheering creationists at the time of a temporary legal victory in California. By the same token, I classify as bunk (apparently to Barnes’ astonishment) a wide variety of notions ranging from astrology to ufology to intelligent design. In doing all the above, however, I actually provide reasons for the reader to understand why some ideas really do belong to the trash heap of history at the same time that judgment about other ideas needs to be held in check until better understanding and more facts will come along.
What I think is important here is to try to stir a course that keeps us as far as possible from the Scylla of scientism (the uncritical and hubristic promotion of science as the only form of worthwhile knowledge) and the Charybdis of extreme postmodernism (the equally uncritical attitude that all knowledge is socially constructed to the extent that — as Feyerabend put it — society should balance scientists with “magicians, or priests, or astrologers.” And yes, I am aware that Feyerabend actually preceded the postmodern wave and the science wars).
I think philosophers of science have crucial roles to play, both within academia itself and in the public arena. We are generally sympathetic to science and understand its power and importance to society. But we are also well positioned to be critical of science’s excesses because of our understanding of its methodologies and assumptions, as well as appreciation of its nonlinear history. This makes philosophy an ideal bridge between C.P. Snow’s “two cultures,” an invaluable asset not just to scholarship, but to public welfare as well.
Yet, serious philosophy of science also comes with the intellectual integrity to call a spade a spade. Creationism, astrology, ufology and the like really are bunk, and we should make no bones about labeling them as such. Reason and empirical investigations into the nature of things surely have limits, and arguably nobody understands those limits better than philosophers of science. But we cannot pretend that we live in a world where the highest danger is too much reason. Barnes rightly notes (though not in a positive light) that for me “this [the separation of science from pseudoscience] is a moral problem and not simply a technical or aesthetic one: belief in science is conducive to our good, whereas belief in non-science or pseudoscience, of which instances are worryingly abundant, is conducive to harm and has to be opposed.” Not quite what I said. I am not advocating belief in science, but understanding and appreciation of science, of which there is far too little in our society. And yes, belief in pseudoscience is indeed harmful, and sometimes lethal — just look at the millions of people in Africa who don’t believe that HIV causes AIDS, or the millions of American parents who are not going to vaccinate their children on the unfounded fear that they will develop autism.
This isn’t science warfare or patriotism, it simply is taking seriously our intellectual and ethical duties as philosophers and human beings.


  1. Where science fails is where good solid philosophy picks up. Like trying to watch a science-fiction film without analyzing all the scientific flaws. It's more about the idea behind the premise than the actual truth.

    Would you really want to watch Luke Skywalker take the actual scientific time to get to Dagobah?

    That would get boring really fast and you'd be dead before he even got half way.

    Suspending scientific thinking for the philosophical messages once in a while is the only way we can function without becoming so logical and critical that we forget to enjoy life.

  2. The shocking thing about Romano's review is that it was bought and paid for ($15,000 + a trip to Cambridge, England) by an organization dedicated to bribing atheist journalists (Romano has declared himself an atheist in print) into publishing religious-friendly, pseudo-intellectual, accomodationist articles.


  3. Hi Massimo,

    Sympathetic as I am to your project, I wonder whether using the language of the notorious "demarcation problem" might be a hindrance rather than a help.

    I think such language may be what really riles Barnes et al, because it can be taken to imply the existence a timeless, cross-cultural "essence of science" which philosophers can illuminate by means of a priori reflection. This is anathema to sociologists, who portray modern science as a contingent social phenomenon swayed hither and thither by social influences. Naturally, Barnes's reflex is to regard you as a "science warrior" and fire a few potshots.

    In place of science vs. non-science, why not talk of good science vs. bad science?

    I think this terminology still captures the distinction you want to get at. Moreover, it strikes me as preferable in three ways: it straightforwardly allows for borderline cases, it doesn't hide the normative element in the project, and it doesn't risk implying a rigid and impermeable distinction between science and other forms of inquiry.

  4. Jonathan,

    thanks for the sympathetic comment. A couple of things: first, I think philosophers should re-appropriate the demarcation issue regardless of what sociologists say. If the latter have a problem with it, so be it. A recent article by Hansson in the Stanford Encyclopedia ( makes it clear that philosophers interested in the problem certainly do not need to subscribe to a view of science as an unchanging activity through time.

    As for reframing the debate in terms of good vs. bad science, in part that is what I do in the book, but that's not enough. Again see Hansson for a persuasive argument that there are differences among pseudoscience, non-science and bad science, and that the distinctions are worth keeping.

  5. Massimo, thank you for posting these remarks. First, I agree with you that demarcation problem is real and immensely complicated. This is why I welcomed hearing from folk that had read your book, or better yet you.

    Second, about Feyerabend. His defense of non-science was always strategic. By this I mean that what he opposed was Kuhnian paradigms that were hegemonic and univocal [i.e., secularized religious dogma]. (He liked psychology because it was a discipline that welcomed a plurality of theories and remained in touch with its history--this may have been so in the 1970s, but now must seem quaint description of psychology.) in Against Method he has a few footnotes in which he makes clear that in different circumstances he would defend science against other activities. Now I think Feyerabend's position has problems not in the least because it overestimates our possibility to know what kind of times we live in, but you suffer from the same hubris; I quote: "But we cannot pretend that we live in a world where the highest danger is too much reason." I admit that if you think that creationism is the main threat to democratic society or our standard of living you are surely right. But if you think, say, that economists and their overly precise models, or genetic engineering (of whatever sort), nanotechnology (etc, and whatever) is the main danger to our way of life (freedom, democracy, etc) then your position does not seem obvious. I would have thought that since Rousseau (Nietzsche, etc) the value of scientific enlightenment needs to be argued, not assumed. (See my fifth point below.)

    Fourth, I think few philosophers of science dare to be critical of science and scientism.

    Fifth (and most importantly), what struck me about Barnes' review (other than the fact that the gloves were clearly coming off) was the following passage:
    "the most salient chapter is the second, on 'Almost Science', wherein string theory and evolutionary psychology figure among exemplars inhabiting 'a complex . . . intellectual landscape that occupies a transitional zone between science proper and activities that may not be entirely "scientific"' (55). Just how one is supposed to evaluate 'almost science' isn't made entirely clear. Whether out of tact or for some other reason the author pulls his punches somewhat in appraising it; perhaps some of his best friends are almost scientists. But the significance of the chapter here is that it acknowledges just how difficult it is (to say the least) to specify what is distinctive about science, and to identify precisely where the boundary allegedly encompassing it should be drawn, immediately prior to a series of chapters wherein a wide variety of beliefs and convictions credited by many millions of people are rapidly and confidently dismissed as bunk, 'nonsense'."

    I take it that Barnes and you are in agreement here. Now, you (quite reasonably) want to go on and at least 'out' or 'debunk' pseudo-science (or bad science following Jonathan's suggestion--I agree that avoiding Popperian associations may be wise here) because there is a lot of misery that follows from the bunk. Fair enough!
    But Barnes is rightly (I think) hinting that some of the real issues are with institutionally well entrenched almost sciences, where a lot of powerful feathers would be ruffled if you were to take them on. I think he is suggesting (but maybe I am reading into his review) that it would show more courage and credibility if you were to take on almost sciences that are beloved by fellow academics (etc).

  6. Eric,

    you make several good points. However, my beef with Feyerabend is that too often he just comes across as controversial for the sake of controversy. It's not enough to say that creationism too would need to be kept in check, should it actually gain the status and power of science. It's crap, and we ought to be able to say so.

    Good points about economics, genetic engineering and all the rest. But my book is not a defense of science (as evidenced by an entire chapter devoted to science blunders), it's a defense of reason.

    Which brings me to your next point: I do criticize some accepted science. String theory (for not making testable predictions, so far); evolutionary psychology (for exaggerating testability claims); and SETI (for not really having much of a theoretical foundation). However, they still seem to me to be in a whole different ballpark compared to astrology and the like.

    Finally, it is precisely because I wish to go beyond Popper that I distinguish among science, pseudoscience, non-science etc. Popper didn't, for him the demarcation problem was a simple issue of science vs. everything else.

  7. The philosophy of science is one of the biggest deficient in my training as a scientist. I can point out non-science and pseudoscience but I cannot adequately explain what science is. I find this unsettling.

    Pseudoscience is particularly harmful when authored by someone with scientific training who knows how to speak to an educated but non-scientist audience. Unlike non-science, pseudoscience pretends to be something it is not.

    I do not really have anything to add to the discussion other than the fact that an elementary course in the philosophy of science should be a compulsory component for all science degrees!

  8. >" Creationism, astrology, ufology and the like really are bunk, and we should make no bones about labeling them as such."


    >"And yes, belief in pseudoscience is indeed harmful...".

    Is there such a thing as belief in reality? What could make the existence of such a thing even in principle possible (given that we are as science claims, atoms moving about in the void according to mathematical laws)? Why not label belief as bunk as well?

  9. To put Heuristics' question in other terms, is folk psychology science or bunk?

    That's a difficult one! On the one hand, folk psychology makes testable predictions which are often vindicated. On the other hand, it shows little evidence of progress and has no obvious means of making any. There is no institutional setting in which novel folk-psychological hypotheses can be devised and tested.

    This raises a broader question regarding how to classify "folk science" (e.g. folk physics, folk biology) in general. Maybe it deserves a category of its own.

  10. Jonathan: Thank you for your answer.

    One could take my post to refer to folk psychology, but one could also take it to refer to scholastic psychology (or some other type, an odd form of science friendly non-epiphenomenal property dualism perhaps).

    What I wanted to get at was the question of wether science even in principle has the metaphysical toolset required to support such a concept. For scholastic psychology it does not (since it removes 2 of the 4 types of causes).

    Folk psychology might be trickier to say how it fits into science. But even so, would it in principle have the toolset to support it? I would argue that science, in principle, could not support original intentionality and that original intentionality is an integral part of the folk psychological concept of belief (when you have a belief about an apple, you really do have a belief __about__ that apple according to folk psychology).

  11. I'm not sure what "original" intentionality means, but folk science already had a category, namely folk science, that is a non- or para-scientific conception of the world.

    I don't think science needs to "support" folk science, for two reasons: science is a particular type of epistemic activity, which does not and cannot cover everything (that's the scientistic mistake); also, a lot of folk science is, in fact, wrong from a scientific perspective (e.g., much folk physics, and even a lot of folk psychology).

  12. Massimo:
    With original intentionality I was hinting at the problem posed by Franz Brentano which as it appears to me lies at the hear of the problem for science to account for such a concept as belief.

    Now, positing that there is a realm of things that science deals with (properties of quarks etc) and a realm that science does not deal with (beliefs) only has a chance of working at first sight. This is because it was stated in the original post that beliefs are harmful, ie it was claimed that they an influence on the world. This has a direct connection on the field of physics since if beliefs can cause harm, this means that beliefs can modify the movement of atoms through the void (the real of science). Yet, if nothing in science claims that they exist we have a very similar problem as we would have with for example homeopathy.

    The diluted is claimed to cause changes to the world, yet the movement of atoms when measured in the subject are describable by mathematical laws that make no mention of any diluted substance.

    Beliefs are claimed to cause harm in the world, yet the movements of atoms when measured in the brain are describable by mathematical laws that make no mention of any beliefs.

    So, where is the causality? Where are the beliefs?

  13. If folk science is straightforwardly non-scientific, shouldn't we call it something other than "folk science"? "Folk bunk", perhaps?

    I'm not sure it can be dismissed so easily. Let's assume that the "theory theory" is correct: folk psychology makes theoretical claims which can be true or false, and contains theoretical terms which may or may not denote real entities. Doesn't this give folk psychology some minimal claim to a place on the "scientific" side of the divide? It certainly meets Popper's falsifiability criterion!

    I'm not quite sure what to say about this. It looks like one of those awkward borderline cases that seem to plague any attempt to draw a sharp line between scientific and non-scientific theories.

  14. Heuristics,

    I really don't think Brentano is relevant to my points. Science can study belief, and has done so increasingly, first via psychology, now via cognitive science and evolutionary biology. To be sure, there is plenty that we don't understand, but I don't see how belief poses an obstacle in principle.


    I guess I have a more narrow view of science than that. I don't think that any factual claim automatically makes for a scientific investigation. Yes, a necessary component of doing science is that we deal with empirically testable claims, but this is not sufficient. Folk "science" doesn't actually have a body of theory, there is nothing like an experiment, and even observations are casual and certainly not systematic.

  15. My impression of Barnes review was that he was accusing you of something like the following: You debunk some pseudoscience in one part of the book with simple unqualified arguments that can then be deployed against science you defend in other parts of the book (unless you go back and qualify the simple arguments with all the suppressed premises, in which case it is no longer simple...). Without reading the book I can't comment on the fairness of the accusation, but your response here would suggest you would have avoided such errors.

    I would say that such complexities are part and parcel of the demarcation problem. I don't think there is going to be a set of simple criterion demarcating science and non-science (or good science and bad science) definitively across all or most cases, for much the same reason I don't think you can right down some simple scientific method in algorithmic form and apply it like a robot. From what you've said here I don't think your disagreeing with that sort of thing. I agree the complexities do not rule out our ability to make distinctions in the stark cases.

    In fact complexity in science I think part of the explanation of the fact that some pseudo-scientific claims are made by scientists who have widely recognized accomplishments in another area. They apply the same method in a different context and get bad results but fail to see the difference.

    I actual think its evident that there are sociological differences between scientific endeavours and pseudo-scientific ones and so social construction can be part of the demarcation. In my limited experience of the social constructivist literature the opposition between the social construction of science and science's epistemic reliability is unfounded. What sociologists actually show in their case studies bears little relation to the sort of deconstructionist claims about science that some of them then adopt (many would deny being critics of science's reliability, just because they study its social constituents).

    Also, philosophers of science should return to the field on the demarcation question because ignoring it has not made discussion of it go away. As far as I can see, all that has happened is that the old positivist and falsificationist (Popperian) criteria have fossilized, but continued to be used in debate. Since there is presumably a reason philosopher of science are often dissatisfied with these criterion they could at least add the reasons for this dissatisfaction to the debate, I don't think those complications really favour the pseudoscientists. In my limited experience the pseudoscientist often deploys simple demarcation criterion to attack their critics as not being scientific.

  16. I know that this thread is a little cold, but isn't the whole demarcation problem sort of analogous to the problem of defining species in taxonomy? The question of whether or not a "science" is a "true science" or not is rather like the question of whether panda bear is a "true bear" (or an okapi a "true giraffe" or is a Dutchman's Pipe a "true plant"). From a layman's perspective, whether or not a panda is a true bear is not based on genotypic or phenotypic distance but rather on whether a Panda looks like a bear, behaves like a bear and/or has the same uses(food, etc.) as a bear. From a layman's perspective, science is understood functionally as a source of political or epistemic legitimacy, not as historians, scientists or philosophers understand it. Thus the demarcation problem is one of cultural/memetic taxonomy, phylogeny and/or cladistics. If we only had cultural DNA from which to rigorously define true clades for science and religions, it would settle this whole debate. As it is, we are like pre-genomic taxonomists debating about whether echinodermata and cnidarians form a true clade.