Wednesday, April 02, 2008

Overcoming Scientism

I take 'Scientism' to be the view that empirical inquiry is the only form of rational inquiry, perhaps coupled with the even stronger claim that only "scientific"/testable claims are meaningful, or candidates for truth or falsity. In other words, it is to dismiss the entire field of philosophy (and arguably logic and mathematics too, though this is less often acknowledged). Indeed, a primary symptom of scientism is that sufferers are incapable of distinguishing philosophical arguments from religious assertions. They claim not to comprehend any non-empirical claim; it is mere 'gobbledygook' to them.

It's worth noting right away that Scientism is self-defeating, for it is not itself an empirically verifiable thesis. Insofar as its proponents have any reasons at all for advancing the view, they are engaging in (bad) philosophical, not scientific, reasoning. This is the familiar point that one cannot assess philosophy (even negatively) without thereby engaging in philosophy oneself. [For a more positive argument in support of the a priori, see my post on conditionalizing out all empirical evidence.]

This bias against philosophy is unfortunate for the obvious reason that there are a lot of interesting and important philosophical truths, which the scientismist would never think to look for. (My original 'scientism' post quoted some ignorant dismissals of Nick Bostrom's very interesting 'Simulation argument'. Not that I think his conclusion is true; but it is eye-opening just to consider.) Moreover, as I once wrote:
All your “common sense” beliefs rest on philosophical assumptions. Most people prefer not to examine them, but that doesn’t mean they aren’t there. It just means that everything you think and do could be completely misguided and you wouldn’t even realize it.

The scientismist will no doubt have many false philosophical beliefs in addition to their scientism. (We all do.) But if they are unaware of the rational tools that allow us to identify and correct such errors, then they will be stuck with them -- not a situation that any dedicated truth-seeker would consider desirable.

I think it's especially unfortunate that most folk seem unaware that reasoned inquiry into normative questions -- e.g. ethics and political philosophy -- is possible. This is at least part of the explanation why public discourse on these matters is so impoverished and sub-rational. So I think it's very important for more people to appreciate that we can go beyond mere instrumental rationality and also assess one's ultimate ends in terms of rational coherence.

Scientism also leads to more mundane mistakes. For example, in a recent 'Overcoming Bias' thread, one commenter defended the common-sense view that different observers experience the same colour qualia (rather than my 'red' being your 'yellow'), on the grounds that the alternative claim is "purely metaphysical with no implications for reality". But that can't be the right reason, because the same could be said of the eminently reasonable -- and presumably true -- view that he was defending. Whether we experience the same qualia or different, either answer is "purely metaphysical" with no scientific implications. So the right justification for the former view must lie elsewhere (e.g. in philosophical principles of parsimony that count against drawing unmotivated ad hoc distinctions).

Fortunately, this bias is easily overcome. Accept no substitutes: Think!

[See also: The Problem with Non-Philosophers.]


  1. Hi Richard,

    I like this argument. One thing I'd question, though, is your acceptance of the claim that inverted spectra cases are "purely metaphysical with no implications for reality". This itself is a metaphysical claim, and one that might turn out to be wrong. One problem with scientism is that it often goes hand in hand with physicalism; and this is, I think, also a problem with phenomenal realism. Why do we think that phenomenal theories have "no implications for reality"? Because given the state of our scientific knowledge today, we don't know how to fit qualia into empirical research. But the state of scientific knowledge today is not unchangeable.

  2. Right, I'm assuming epiphenomenalist property dualism here. (There are principled grounds for thinking that the third-personal methods of science could never get a perfect grip on essentially private, first-personal, subjective experience.)

  3. Yes, I did realize that you were assuming property dualism. I am somewhat skeptical about the principled grounds--not because I don't think there are principled grounds for the scientific/subjective distinction, but because I think the validity of this distinction relies on the current state of the sciences. For all we know, the next paradigm shift might overturn the third-personal focus of scientific inquiry.

  4. It just so happens I am the Overcoming Bias poster in question.

    I initially took some umbrage at the implication that I have not thought about my position, although I understand you did not mean it in that way. My abilities with philosophy-jitsu aren't very strong, but I love the subject. On closer thought, though, I realized that I am falling into the pit you identify in your post.

    I wouldn't consider my point of view to be pure "scientism" as I do see value in philosophical inquiries that lack empirical basis. I have chosen, however, to set aside some of the hard questions of subjective experience because they seem like philosophical whirlpools on the ocean of reason. Traps in which I could spend the remainder of my lifetime paddling for an escape. My reasoning is that I have to, at some point, assign probabilities to idealism or realism and in so doing "take a position."

    By ignoring the more fundamental issues of, say, idealism vs realism and just accepting materialism out of hand (by giving it a higher probability for the sake of establishing ground for further arguments) or because "it seems correct" I'm just practicing a kind of rationalization, instead of meaningful reasoning.

    I still believe that I have to "take a position" (by which I mean assign probabilities to various beliefs) but in reflection I need to make sure I have stronger evidence supporting the probabilities I assign. At worst failing to do so would lead to me only reading sources that I believe will confirm my priors.

  5. Hi Richard,

    There's a lot going on here; here are some thoughts.

    - You seem to think that "empirical inquiry is the only form of rational inquiry" amounts to a dismissal of the entire field of philosophy. I'm not sure about this. All sorts seem to take philosophy to be "continuous" with science. For many, I think, this amounts to taking philosophy to consist entirely in such activities as (a) the explication of key scientific terms and (b) the regimentation of languages or language-fragments that constitute (in part) or are capable of expressing our best scientific theories. This sure sounds like a plausible metaphilosophical view on which philosophy is a part of "empirical inquiry." But then we have a plausible metaphilosophy consistent with scientism. So we probably shouldn't say that scientism, as you state it, amounts to the dismissal of the entire field of philosophy.

    - The proposition, that empirical inquiry is the only of rational inquiry, is empirically verifiable if we take people's verbal behavior to be the basic source of evidence or consideration when we are trying to say just what forms of inquiry are rational, or, better, when we are trying to explicate "rational."

    - Whoever objected to the (epistemic) possibility of spectrum inversion on the basis of the impossibility of detecting spectrum inversion is, I think, mistaken, but not for the reasons you give. I won't go into the details, but I think that thought experiments about futuristic imaging techniques and clever psychophysics can show how to confirm or disconfirm inversion hypotheses on a pretty uncontroversial background theory. Scientismist philosophers, anti-scientismist philosophers, and anti-philosophers alike probably tend to have too narrow a view of what is testable stemming from a too-narrow view of what the right sort of background theory can do to make even the loftiest or most theoretical claims empirically consequential.

    - There is some truth to what you're saying. I suspect that all of your criticisms are damning of some view like what you're calling "scientism," but that you need to state it a little bit differently. Perhaps what you really want to get at is Mach-style positivism (and Mach-style anti-philosophical posturing) - everything is a logical construction from the given, and so on.

  6. > Whether we experience the same qualia or different, either answer is "purely metaphysical" with no scientific implications

    again not sure this is true - it seems to deny various forms of logic that might not be particularly strong but do exist. For example
    1) 'I experience red as red'
    2) my experience of red as red seems to be associated with a set of brain patters because it occurs when those patterns occur and only when they occur.
    3) you have the same brain patterns

    sure the identical pattern could be associated with a different Qualia - maybe it does, but it doesn't mean that there isn't some sort of evidence there and it seems like pretty basic science.

    Similarly people have similar descriptions of the associations of qualia - so you could create a sort of model of part of qualia as 'a sum of certain words' and as long as they coincided for a person say that that person probably has in those regards similar qualia. 9unless we want to say that the entire structure is somehow almost duplicated in another persons mind. the defense against the latter proposition would be something about that being an improbably ordered structure.

    I presume remaking romans point your argument seems to be the mirror image of your issue with scientism - regardless of what principled grounds you might have to hold it (scientists have principled grounds also).

  7. (It's interesting when you read what you take to be an obviously uncontroversial post and then see people disagreeing in comments)

    A few unrelated comments:

    Richard, and everyone in comments,
    "Purely metaphysical with no implications for reality" is an odd phrase, given that metaphysics is the study of the nature of reality.

    To think that everything is either (a) science, or (b) to do with subjective experience is also mistaken. Mathematics and moral philosophy both fall into neither of these.

    Your first bullet point expresses a very implausible view, according to which only science and philosophy concerned with science, are worthwhile endeavours. What happened to moral philosophy, political philosophy, aesthetics, and so on?

    A further argument against scientism:
    1) Science is concerned with explanation.
    2) Normative properties, by definition, are not explanatory entities (they justify, not explain).
    So: 3) Science cannot address normative questions.


  8. I must admit, the more time goes, the more disillusioned I get with philosophy. While I can't really disagree that strictly speaking, you're right 'scientism' as you define it, is self defeating, the areas where philosophy in itself without empirical work seems useful very slim.

    I just looked over your list of philosphical 'truths', and very few of them seem particularly interesting unless you're already invested in the debates. To the outside world a slight clarification over the definition of knowledge is not going to be particularly helpful - they weren't looking for one in the first place. They just don't get close as a list of achievements to say, quantum mechanics, evolution, relativity. whatever.

    This sounds like I'm just slagging off the subject as whole and as worthless, but I'm not - I think it can be fun, enjoyable. But if you were trying to make a prediction of the next 100 years - do you really think that progress in understanding the fundamental nature of the universe is more likely to be made in the physics department or the philosophy department? Are we more likely to learn about how humans think and feel, about our morality and art, from philosophy or from cognitive science?

    I don't think its neccessary that philosophy is always useless - as you say, mathematics has an impressive history of results. Philosophy, as far as I can see, doesn't.

  9. I think the opposition of scientism and philosophy is a false dichotomy based on flawed definitions.

    I picture the relationship between science and philosophy as a concentric venn diagram with science in the center.

    Scientism would be a case of a science-oriented observer saying that some or all the philosophy inside the science circle was tollerable but some part on the outside failed for insufficient reason (or was unempirical bs).

    Philosophism would be a case of a philosophy-oriented observer saying that much or all of the science lacked sufficient epistemic grounds. He would take an extreme view of what "sufficient" meant.

    The reasonable scientist would freely admit to an epistemic "veil" at the frontiers of his empirical observations. He would admit to the value of philosophy in testing, interpreting, and organizing existing knowledge; and he would admit that some portion of the philosophy lying beyond the boundary of science was useful for asking informed questions and making reasoned hypotheses about the unknown. To be polite he would be quietly agnostic about the rest (i.e. take what you can use and leave the rest).

    The reasonable philosopher would have about the same point of view in reverse.

    Of course, there would still be about a 20% - 40% divergence between the reasonable scientist and the reasonable philosopher on where all the veils and boundaries and frontiers were, and that tension would be a healthy and congenial one.

    I hope that settles the argument.

    Poor Richard

    Poor Richard's Almanack 2010


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)