Wednesday, May 24, 2023

On Mizrahi on Scientism

 Moti Mizrahi has a very poor essay on scientism at It's sometimes been said that a sign that common criticisms of scientism are right is that the philosophical attempts to defend it are so provably bad, and this essay is a very good example of it, since it features in very blatant form a number of the common failures of such defenses. For instance, Mizrahi can't even get out of the introductory portion of the essay without being inconsistent in his characterization of it -- he repeatedly conflates 'best form of knowledge' with 'best way of knowing', despite the fact that these are not even remotely the same things, and then later goes on to conflate (although this is very deliberate and explicit) 'knowledge' and 'research', despite the fact that almost nobody uses these terms as synonyms, and then conflates 'research' and 'what is found in academic publications of an academic discipline' despite the fact that everyone recognizes that some of the latter is actually not research (e.g., fraudulent papers) and that there is much more to research than what ends up in academic publications (e.g., practical skills). So, despite pretending to be talking about what everyone else is talking about when they talk about 'scientism', Mizrahi's 'Weak Scientism' seems to end up being the position that science (by which he seems to mean only 'what ends up in what we call '-"scientific journals"') has the academic journals or the ways of putting together academic journals (we don't know which he means) that are the best at producing things that go (or are supposed to go? -- it isn't very clear what the standard is here) in academic journals. Needless to say, very few people other than Mizrahi would call this mush 'scientism' at all, and it will obviously be at least questionable sometimes -- for instance, it's a well-known view of academic publishing that some scientific journals are very bad for science itself and some are very good, and there is plenty of reason (fraudulent papers being one) to think that the very bad scientific journals are not better for science (or for society, or for humanity, or for any number of other possible standards of bestness) than the very best non-scientific academic journals are. 

On the qualitative side, Mizrahi says, "Scientific knowledge can be said to be qualitatively better than non-scientific knowledge because scientific knowledge is explanatorily, predictively and instrumentally more successful than non-scientific knowledge." But since 'knowledge' can only mean 'research' here, by Mizrahi's own definition, this will obviously just depend on what journals and papers you are looking at. If one wants to argue that Physical Review is a better journal for learning about physics-journal research than Philosophy of Science, this seems like something everyone accepts; if one wants to argue that Physical Review is a better journal for learning about philosophy-of-science research than Philosophy of Science, this seems highly doubtful; if one wants to argue that Physical Review has better editorial practices than Philosophy of Science, OK, maybe, although this seems to be a practical assessment that requires an investigation of what each journal is for; if one wants to argue that Physical Review does more for physics than Philosophy of Science does for philosophy of science, OK, maybe, although this would require some sort of sociological investigation of the impact of each, the results of which are not likely to be particularly interesting. But if you want to argue that anything published in a scientific journal, including the fraudulent papers, is better for either physics or philosophy of science than anything published in Philosophy of Science, there are literally legions of reasons for rejecting this position. On the other hand, since Mizrahi characterizes 'better' on the quantitative side as producing more knowledge (=research) or having more of an impact (=citations, apparently), then we have to consider all the journals together. But there it seems entirely a matter of historical accident whether scientific journals happen overall to publish more research or involve more citations than non-scientific journals -- the first depends heavily on the economics of journal publication and the latter on customs of citation, both of which are different across different fields for any number of reasons.

Mizrahi also shows that he doesn't understand what it means to treat something as wrong by definition; he accuses the psychologist Steve Taylor of doing this despite the fact that Taylor quite clearly is not putting wrongness into the definition of 'scientism' but giving a reason why scientism is a non-starter, namely, that its proponents fail to make proper distinctions between assumptions and facts. Whether or not this is true, it is obviously absurd to claim that this is treating scientism as wrong by definition; a reason, one that explicitly concerns argumentative practice and not definition, is explicitly given, and the claim that scientism is a dogmatic belief system is very clearly presented as a conclusion rather than a premise. Mizrahi's mistake here is immensely embarrassing; whether or not it is right, Taylor's argument is not structurally complicated and is literally the kind of argument I give to Introduction to Philosophy students in order to give them practice in distinguishing premises and conclusions. When a professional philosopher in trying to defend a position is making literal Intro-level errors, this is a red flag that there is something wrong with the position being defended -- people who spend their days studying arguments are unlikely to get an analysis of an elementary argument so confused unless the position they are defending is itself extremely confused.

Mizrahi also shows that he doesn't understand what a 'persuasive definition' is. He defines it as "definitions that are intended to transfer emotive force, such as feelings of approval or disapproval"; but this is false. 'Persuasive definition' is a term of art, due originally to C. L. Stevenson, that means a definition intended to change the meaning of an already emotively charged term (which a very large number of terms are) in order to elicit the same reaction of approval or disapproval toward the new meaning. None of the examples Mizrahi gives are persuasive definitions; most are not definitions at all, being diagnoses rather than definitions, and the few that could plausibly be considered attempts at definitions are not trying to change the meaning of 'scientism'. Mizrahi, on the other hand, is trying to change the meaning of 'scientism', since none of the critics use it in his new sense, but he is also not giving a persuasive definition because he is not trying to keep the emotive meaning the same but reverse it.

The whole article is an embarrassment, and should never have been published, much less in a popular venue where people might be misled into thinking that Mizrahi's eccentric position on these matters is commonly accepted by philosophers of science or that Mizrahi's argument, as presented, is the sort that could seriously be accepted by philosophers without complete reworking. Just utterly irresponsible.