From the point of view of philosophy, the mind is no longer anything but delirium and madness. There is no complete system, synthesis, or cosmology that is not imaginary… The system is a mad delirium. Hume shows in the hypothesis of an independent existence the first step toward this delirium. Subsequently, he studies the manner in which independent existence is formed in ancient and modern philosophy. Ancient philosophy forges the delirium of substances, substantial forms, accidents, and occult qualities – “spectres in the dark.” But the new philosophy has also its ghosts. It thinks it can recuperate reason by distinguishing primary from secondary qualities, but in the end it is no less mad than the other. – Deleuze, Empiricism and Subjectivity, p. 83.
Let epistemic verificationism be the view that no knowable proposition p is knowable except via a specific method M, e.g. the scientific method. This is frequently combined or conflated with either semantic verificationism, according to which the meaningfulness of p is a function (in some sense) of its knowability via M, and ontological verificationism, according to which there is no X such that X (or any proposition p about X) is unknowable via M. By ‘verificationism’ I intend the conjunction of epistemic and ontological verificationism.
As Ladyman et al observe (2007: 61), verificationism “is promoted as a bar against seeking explanation where we have good reason to doubt that it promises anything but temporary psychological satisfaction at the expense of truth.” (Their definition of ‘verificationism’ on p. 29 is a little different to mine, but never mind.) My question in this post will be: where does this “good reason” come from? As they note, verificationism is historically associated with empiricism, and thus really with the quintessential Aristotelean gesture of bringing the Platonists back down to Earth. This is a fascinating gesture, but I have always thought that it was philosophically underdetermined to a certain extent. Be this as it may, these days verificationism is typically read right off of the content of science itself, so we’d better take a look at how this is meant to work. The most common way of doing it, as far as I can see, is in terms of evolutionary theory broadly construed, and/or neuroscience broadly construed. Thus for example witness Lakoff’s and Nuñez’s recent application of cognitive science to mathematics. Somewhat different from this but falling under the same umbrella, witness Krauss’ plea to replace the traditionally philosophical question of nihilism (i.e. why is there something rather than nothing?) with a scientifically tractable surrogate (how does the universe emerge from a primordial quasi-vacuum-like soup?).
Instead of beginning (as Ladyman et al do) with a stark contrast between bad, domesticating metaphysics, and good, science-respecting metaphysics, it is more heuristically apt – because more neutral – to imagine a continuum of strategies each of which attempts to calibrate the heterogenous and incompatible inputs X, Y, Z, etc. into a congruent whole A. This is done either by establishing relations of reductive asymmetry between e.g. X on the one hand and Y and Z on the other; or else by e.g. establishing reductive asymmetry between X and Y whilst eliminating Z altogether. To reiterate, we gently massage the pieces of the puzzle we are given, and then cut off what remains afterward. This involves deciding how to weight the various inputs, so that we know which way to go whenever we reach a fork in the road, as well as whether the fork is merely reductive, or fully eliminative in nature.
Note: Ladyman et al reject the dichotomy of reduction and elimination (p. 253). But they do offer a partially isomorphic distinction between real and non-real patterns, whereby elimination is associated with the latter (either construed as “mere” patterns, as on p. 254, or as not-even-mere-patterns, as on p. 231), and reduction is transmogrified into the “primacy of physics constraint”, which is at least reductive in the sense that sciences other than physics cannot posit real patterns incompatible with physics.
Given this framework, here is the general matter I wish to explore. Typically science-inspired verificationism has two central commitments: to something like the fundamentality or “primacy” of physics (p. 38), and to the critical force of evolutionary biology and neuroscience for buttressing the thesis of verificationism itself and launching periodic raids upon the still succulent villages of luddite philosophy and folk superstition. Thus, specifically, many advocates of scientism and verificationism want to weight the sobering data of cognitive science and evolutionary psychology heavily enough that common sense loses its place at the dinner table altogether. We end up with a sort of error theory of the very concepts of everyday thinking.
However, it is possible to push even further then this, as in the argument between Sheldon and Amy Farrah Fowler (i.e. The Big Bang Theory) about whether physics explains cognitive neuro-biology or vice versa. Again, by focusing on the basic gesture of explanatory asymmetry, which is characteristic of reductionism (as standardly understood) – asymmetry both in the sense that X explains Y but not vice versa, and also in the sense that X subtends (as it were), or underwrites Y, but not vice versa – we are able to tease out an interesting passageway along which critique may venture. Start by distinguishing between the following:
- Arguing over whether science as a whole trumps non-science.
- Arguing over which part of science is wearing the pants.
It is clear that Ladyman et al take their cue from physics, and their answer to A is ostensibly based upon this. However, it is not clear how verificationism might be exclusively supported by the project of unifying the sciences in terms of the fundamental priority of physics. That is why I said that verificationism has two fundamental commitments. To get such support for verificationism, one either needs some sort of philosophically motivated reductionism (which I said above has always struck me as underdetermined), or else reductionism motivated from within a different part of science, e.g. evolutionary biology or neuroscience. That is where B comes in. What is intriguing, then, is the way in which allowing the life and brain sciences to wear the pants can, when pursued consistently and pushed to its logical conclusion, create friction within science.
physical meta-chemistry → chemical meta-biology → biological meta-cognitive science → cognitive meta-physics?
To wit: physicists and mathematicians do not like being told that the content of their disciplines is comprehended by something else that usurps their claim to fundamentality through a combination of reduction and elimination. Indeed it will probably strike any real enthusiast of the former that the latter is deeply anti-scientific (cf. p. 45), i.e. that cognitive science better restrict itself to explaining how e.g. mathematicians are able to do what they take themselves to be doing, and not some other thing. Clearly this is a bridge too far for Ladyman et al – indeed, they are prepared to go quite far in the opposite direction, as is evident when they write (p. 236):
We think there is much attractiveness in the idea that good metaphysics can best ensure its liberation from anthropocentric myths by accepting a requirement for formal articulation, leaving natural languages behind.
This may turn out to be too sanguine. The disconcerting alternative is that we seem to be reaching a point where science is developing its own correlationism, i.e. that it is internalising what was hitherto a largely philosophical debate about the nature of scientific knowledge. That is, we seem to be reaching a point within the development of science akin to the emergence of the meta-critical problem in the years after the publication of Kant’s Critique of Pure Reason.
- Gilles Deleuze, Empiricism and Subjectivity: An Essay on Hume’s Theory of Human Nature, trans. Constantin V. Boundas, (Columbia University Press: 1991).
- James Ladyman et al, Every Thing Must Go: Metaphysics Naturalized, (OUP: 2007).