Tuesday, June 9, 2009

More on chapter 1 of Every Thing Must Go (2)

Let me continue with some critical notes on Ladyman & Ross, Every Thing Must Go, chapter, 1. I will number where I left off.

6. (p. 28): I like that R&L (et al) argue from the specialization/division of labor within the sciences and demand for efficiency to a role for metaphysics as 'critically elucidating consilience networks accross the sciences'; shortly thereafter (30ff), they draw on Kitcher's work on unification to spell out what they mean with this phrase. But what I don't get is why they think this argument allows them to rule out competing tasks for metaphysics.

7. (p. 28): Given that R&L insist that metaphysics must in an important sense be constrained by science, they need a criterion to demarcate science from non-science. Amazingly, they opt for "solely institutional norms" (in terms of "institutional error filters"). But they don't consider that these may be a necessary condition for a body of practices and beliefs to be scientific, but by no means a sufficient condition. They are blind to the fact that, especially after success of Kuhn's Structure, a lot of disciplines dressed themselves up with a lot of scientific institutions and insisted on a sole paradigm (etc), stampinging out dissent. There is plenty of what one may call 'zombie' science to go around (they look like science, but produce junk). R&L et al forego any critical evaluation of the epistemic success of these institutions and show a remarkable lack of curiousity in the body of research that empirically investigates institutional error filters in practice. It is especially ironic that they claim "a naturalistic demarcation principle should be based on reference to criteria that are empirically observed to regulate the practices of science" (33)! Yet, they do not want to hear about these practices at all. Instead they use a proxy, "fundable research" (34). They certainly would not want to hear about funding practices.

8. (29): R&L offer a "non-positivist version of verificationism". It has two components: "First, no hypothesis that the approximately consensual current scientific picture declares to be beyond our capacity to investigate should be taken seriously. Second, any metaphysical hypothesis that is to be taken seriously should have some identifiable bearing on the relationship between at least two relatively specific hypothesis that are either regarded as confirmed by institutionally bona fide current science or are regarded as motivated and in principle confirmable by such science." I am not going to kvetch about wording of these two aspects. (R&L admit that there is plenty to complain about.) I wonder how this leaves enterprises that want to unify two distinct domains of science mathematically, but without offering new empirical content (or empirical content to be found in dimensions unavailable to us). It looks to me that this verificationism rules out much effort, say, to connect GR and QM. Also, why *two* hypotheses? Why not develop implications of one? Or why not demand, a minimum of three?

9. (36): R&L wish to exclude metaphysical projects "that are primarily motivated by anthropocentric (for example, purely engineering driven) ambition, as opposed to ambitions anchored around attempts to determine the objective structures in nature." It's nice to see pure/applied distinction so unabashedly affirmed. Yet, I wonder how they evaluate research in the bio-life sciences, where the *funding* agencies have their eyes firmly on the potential medical/technological spin-offs.

10 (36): according to R&L metaphysics should not be motivated by engineering practices. Tell that to three of the best analytic, naturalistic works in philosophy of science/metaphysics: *On the Origin of Objects* by Brian Cantwell Smith; *Re-Enginering Philosophy for Limited Beings* by Bill Wimsatt; John Haugeland's classic work on *Artificial Intelligence*. Only Haugeland is mentioned (as a reader of Dennett, 199), although R&L are blind to Haugeland's important criticism of Dennett's project. (Disclaimer, I was a student of both Haugeland and Dennett at one point or another.) This is to say, Ross & Ladyman are breathtakingly parochial.


  1. Eric, thanks for these comments. We have a group at SLU that is reading this book this summer as well, at the instigation of Scott Berman and Scott Crothers, a grad student writing a dissertation with Berman and me. I'll alert the others to your blog contributions!

    I wonder about your concerns (under #8) about purely mathematical unifications with no new empirical content. It seems to me that attempts to unify GR and QM typically *are* concerned to show new empirical content, at least in principle, and that much of the criticism (e.g., from Lee Smolin) of some such efforts is directed precisely at a failure to be *sufficiently* concerned about this issue.

    On the other hand, these same efforts point to a potential problem in the implementation of R&L's revised verificationism, which is that "the current scientific picture" does not typically directly address the question of what we have the capacity to investigate (though scientists will have opinions about this), leaving investigatability (is that a word?) itself ambiguous. Hence the very wide range of opinions among physicists about string theory.

  2. Hi Kent! It would be nice if we could use a blog like this to create a wider discussion. Fred Muller is also taking notes on our discussion, and I am egging him on to blog these here. (That would have advantage that some of the technical issues also make their way here--I am too smart [i.e., cowardly] to put formulas here.)
    Anyway, in fairness to R&L I think they are fine with the vagueness of investigatibility [sic] as long as they can be rid of projects that are not at all motivated by current science.
    Yet, so far they do not confront how they think of the confirmability (?) of the dimensions posited by string theory.

  3. I continue to be curious about the demarcation issue in connection with recent philosophy of mind.

    There are many themes in philosophy of mind that are metaphysical, ie. not strictly scientific, in character. Consider the thesis that there are directly referential components of intentional content. I don't see how this question could be tackled in a purely scientific way. It is metaphysical.

    Is the observation that I just made supposed to be governed by "institutional norms"? I thought that my grounds were simply that there is no way to test these theses in a purely experimental way. Why can't I make a distinction between what is experimentally testable and what is not without deferring to an institution?

  4. Mohan, I don't think R&L (et al) claim that the claim you made should be governed by institutional norms. Rather I think they propose that if a funding agency would not see how the question of establishing the 'referential components of intentional content' can be made into a research question (such that the normal filtering mechanisms of science can do the fruitful work), it should be rejected as a worthy project *for metaphysics.* So they would agree with you that you can make a distinction in whatever fashion you wish, but they would want to dismiss the project it is aiding. But perhaps Ladyman and Ross see their own stance toward alternative approaches differently. (They are fun to read in part because they do not tolerate alternative approaches to metaphysics; they are dogmatists.)

    I hope to comment on remainder of Chapters 1-2 some time next week. But struggling with deadlines.

  5. Hi,

    We have just added your latest post "More on chapter 1 of Every Thing Must Go (2)" to our Directory of Science . You can check the inclusion of the post here . We are delighted to invite you to submit all your future posts to the directory and get a huge base of visitors to your website.

    Warm Regards

    Scienz.info Team