International Theory of Mind: Wendt, PTJ and Philosophy of Mind

When re-reading some Wendt recently whilst finalising an article for submission, I stumbled on this rather odd claim:

in the end there can only be two possibilities [for types of explanation], materialist and idealist, because there are only two kinds of stuff in the world, material and ideational

– Wendt, Social Theory of International Politics 1999, p. 136

This statement, I think, takes Wendt into some very, very treacherous ontological waters. The idea that matter and mind are the two basic kinds of stuff composing the universe is a highly contentious position within philosophy of mind. Lots of pitfalls lie in the direction Wendt wants us to take, which seems to be some form of substance or at least property dualism. The most basic and obvious problem is that, if material and ideational stuff are distinct from one another, how do they interact?

Patrick Thaddeus Jackson points out another problem for Wendt:

Wendt argues that we can preserve the autonomy of the social and ideational aspects of world politics in our theories by sharply delineating just how much of social life is explained by material factors, so that we can see how much of a difference non-material factors make (Wendt 1999, 135–136), but this strategy is unsustainable if the non-material aspects of world politics are themselves products of material factors!

-PTJ, The Conduct of Inquiry in International Relations 2010, 97

If one takes the, prima facie plausible, view that the brain provides the material substrate of the human mind then it seems difficult to see what room there is for a separate ‘ideational stuff’ to play a role in our explanations. PTJ pushes this argument further:

The problem precisely parallels the challenge confronting philosophers of mind who seek to maintain simultaneously that mental phenomena emerge from physical phenomena and—as the natural sciences seem quite unequivocally to maintain—that physical events have physical causes: there is simply no way that mental states emerging from physical arrangements could ever add anything to an explanation that simply linked antecedent physical arrangements to a physical event directly

-PTJ, The Conduct of Inquiry in International Relations 2010, 97-8

This is a version of Jaegwon Kim’s well known argument about supervenience and the redundancy of mental causations. Alarm bells should be going off at this point though, because PTJ seems to have painted himself into a corner.  I don’t think he wants to accept the conclusion, indeed it is presented almost as a reductio ad absurdum against a critical realist position*. Therefore he must reject one of two propositions: that physical events have physical causes; or that mental phenomena emerge from physical phenomena. Rejecting the former suggests abandonment of the natural sciences, rejecting the latter implies a commitment to a pretty hardcore version of dualism or idealism.

I think we are better off accepting those two propositions (although to say that ‘mental phenomena emerge from physical phenomena’ is a bit vague and doesn’t specify between a range of plausible positions in philosophy of mind) and take issue with his interpretation of the conclusion. PTJ states that:

if one maintained that the world was reducible to purely material objects then there would be little point in studying ideas and beliefs and other mental factors

-PTJ, The Conduct of Inquiry in International Relations 2010, 225

This is misleading, not least because many physicalists in philosophy of mind argue that mind is not reducible to material objects, but rather shares an identity with certain material processes. Furthermore, even if ideas and beliefs really are just pre-scientific labels we use to refer to particular classes of physical processes, these processes remain as plausible candidates as any others to provide explanations for social phenomena. Commitment to physicalism in the philosophy of mind doesn’t necessarily tell us much anything about what sort of processes are causally efficacious in the social world.

It might be worthwhile to try to clarify some of the ontological confusions of this sort in contemporary IR theory, but I’d have to invest much more time than I can spare familiarising myself with the philosophy of mind literature. Maybe I will try to add some more constructivism and critical realist IR theory to my reading list instead.

* Which is given a notably shorter shrift than the three other metholodogical positions PTJ examines in his book.

Advertisements

Posted on January 6, 2013, in Uncategorized. Bookmark the permalink. 8 Comments.

  1. The idea that materialism means psychological explanations must be abandoned for neuroscientific ones is nonsense. Both refer to the same processes and phenomena, just at different levels of abstraction. Accepting that computers are made of electronics doesn’t make computer languages useless in explaining how they work. Neither do microeconomic theories seem to have made macroeconomic ones suddenly redundant. I see no reason why the relationship between social and ideational explanations of political phenomena should be different. I think you’re right to imply that there’s a sort of category error at work here – a presumption that different accounts of the same phenomena must be redundant.

    While accounts of the same process at different levels of abstraction should be reconcilable, and therefore are in principle redundant, I think the fact that these systems are so complex means that predictions across levels of abstraction aren’t possible, and so different models are needed. I think there’s an empirical hierarchy though. When two explanations differ, the least abstract one seems to do better, probably because the relevant evidence base is more tangible, and less messy.

    For an entertaining, if not universally popular armchair read on this theme, try E. O. Wilson’s ‘Consillience’.

  2. My whole point in introducing this example was to suggest that whatever critical realists mean by mind-independence, it can be neither a reductionist form of physicalism nor a “two-substance” kind of dualism, since neither position is tenable. Two-substance dualism raises the problems of interaction that you mention, and unless one is very careful I think physicalism tends towards rendering the mental/psychological epiphenomenal. If the world were purely material, then “immaterial” representations of things in the world would be a) an illusion or b) lacking any autonomy from the material and hence safely able to be ignored. (Steve’s computer language example is misleading, because the electronic system of the computer is constituted precisely so as to facilitate the language we use to control it — not just represent it! — and thus give rise to what appears to be conditional autonomy even though it’s actually just a shorthand or prosthetic manipulation of bits. Unless one is willing to say the same of the material world, which I for one am not, the parallel falls short, and even if the parallel holds, there’s no avoiding the conclusion that the computer language is in the end an epiphenomenal shorthand for the actual electronic processes. Remaining at the level of the computer language might help you do things with the computer, but is it really an “explanation”? Not entirely sure.)

    I would rather say, as I do in the book, that a plausible critical realism depends on a stratified notion of reality that highlights dispositional causal powers. Those dispositions are mind-independent, but since they’re not substances we don’t have two-substance dualism, and since mind isn’t reducible to the dispositions of the objects in the world we don’t have epiphenomenalism.

    • Thanks for the clarification. My example about computers was badly chosen – although I’d still point to behaviour of software as an explanation for the behaviour of a machine, the design element makes it different than other examples I could have chosen. I agree it would be a leap to say the world operates the same way.

      My wider point would be that both immaterial and material accounts of phenomena are themselves only ideas, which depend to some extent on predictive power for their validity.

      You neatly pointed out my agenda though – I do think materialism does lead to an epiphenomenal idea of mind. Where I disagree is that I don’t think that this means mental/psychological descriptions of mind are redundant.

      You’ll have to forgive me though – I’m quite a way from my academic background (lapsed neuroscientist with a passing interest in philosophy of science). I may be talking at cross-purposes to some extent. My training makes materialism and empiricism reflexive positions, rather than ones underpinned by detailed analytical reasoning. That said, I have relatively little sympathy for critical naturalism as a philosophy of science because (unless I’m missing something) it seems to depend on real but unobservable generative mechanisms, which if investigated themselves would only push the problem of causation into an infinite regress of mechanisms. It’s a valid account of how most scientists probably think about their work, but as they say, scientists know about as much about the philosophy of science as fish do about water.

  3. Thanks for the comments. This is tricky territory, but I’ll hazard a couple of comments to PTJ:

    1) I think Steve’s analogy with computation does better than you grant, although I share the same caution as to whether programming languages ‘explain’. Nonetheless, source code in a programming language must run through a compiler before it becomes machine code. Hypothetically, the same machine code and computational function could be programmed for in two distinct programming languages. The rules of the two programming languages may be quite distinct. So there are a set of facts about each programming language that are somewhat autonomous from the machine-code level. Second, computational functions are substrate neutral. The arrangements of bits does not have to be electronic, it could take many other physical forms. So computational functions have multiple realisability – a feature which mental states may also possess and which are often held as making ‘greedy reductionism’ problematic.

    The computational example is a disanalogy with the brain-mind issue, however, in so far as there are special problems created by the fact that it’s ‘like something’ to have a mind – i.e. the tricky issue of how subjective experience and qualia fit into to the world.

    2) I still don’t think that your argument follows. For a start, saying the world is physical rather than material would be more helpful as there are objects in the physical world that aren’t material objects (virtual objects such as holograms for example, energy perhaps depending on definitions). In any case, even if we grant that mental states are epiphenomenal then I don’t see how social scientists have warrant to ignore them. Let’s take a strong version of reductive materialism, according to which mental states have a one-to-one correspondence with identifiable neurological states that can be exhaustively accounted for in terms of the arrangement/motion of fundamental particles. This means that knowledge regarding mental states provides us with reliable knowledge about the physical states and processes at work within a set of physical systems. Those physical systems (which correspond to the epiphenomena we know as ‘beliefs’, ‘desires’, intentions’) may well be causally efficacious. Actual social scientists may still make use interviews, archival sources, close reading of texts etc. to reconstruct the beliefs of agents and thus make inferences about causally-efficacious neurological processes. I don’t see any problems for social science, the objections seem to derive from whether this is a plausible philosophy of mind.

    As I result I don’t think the dilemma you propose has the force you suggest (especially not as stated in your p. 225 footnote), even if reductive physicalism was the only sort of physicalism in philosophy of mind.

    3) I have big problems with the notion of a mind-independent world, but from the physicalist/materialist side of the issue. Instead we might regard the human brain as a physical system embedded the wider physical systems composing the natural world. It’s unfortunate if critical realists have chosen to defend their position in terms of the mind-independence of the world, as it seems to risk relapse into dualism. But, as I said, I need to read some more critical realism and go over the second half of the chapter I quoted from again.

  4. Thanks for the clarification. Nick and I were obviously typing at the same time, so some of what follows will be redundant with his comments. My example about computers was badly chosen – although I’d still point to behaviour of software as an explanation for the behaviour of a machine, the design element makes it different than other examples I could have chosen. I agree it would be a leap to say the world operates the same way. I disagree with Nick’s point about qualia though: would we know if it were ‘like something’ to be a computer? While I wouldn’t join Dennett in ditching qualia, I think the subjective-objective divide makes them impossible to study and, for me at least, difficult to reason about with any level of confidence

    My wider point would be that both immaterial and material accounts of phenomena are themselves only ideas, which depend to some extent on predictive power for their validity.

    You neatly pointed out my agenda though – I do think materialism does lead to an epiphenomenal idea of mind. Where I disagree is that I don’t think that this means mental/psychological descriptions of mind are redundant. Empirically no account is privileged.

    You’ll have to forgive me though – I’m quite a way from my academic background (lapsed neuroscientist with a passing interest in philosophy of science). I may be talking at cross-purposes to some extent. My training makes materialism and empiricism reflexive positions, rather than ones underpinned by detailed analytical reasoning. That said, I have relatively little sympathy for critical naturalism as a philosophy of science because (unless I’m missing something) it seems to depend on real but unobservable generative mechanisms, which if investigated themselves would only push the problem of causation into an infinite regress of mechanisms. It’s a valid account of how most scientists probably think about their work, but as they say, scientists know about as much about the philosophy of science as fish do about water.

  5. It seems to me that, ontologically speaking, we should view a separation between the material and the ideational as false (if we’re scientific realists). This is because every token ideational state (qua ‘ideational stuff’) either reduces to or is caused by some token neural state. However, we can still have good reasons for retaining explanations that employ mental causality. Here are some reasons why:

    -We in-principle accept token-token reduction but we deny type-type reduction[1] due to multiple determination (which I believe is the position of most philosophers of mind currently), therefore making it impossible to reduce psychology to neuroscience, though perhaps possible to reduce history to neuroscience.

    -We accept both token-token and type-type reduction in principle, but only when we achieve ideal completeness of psychology and neuroscience, and until that point we have no capacity to reduce the former to the latter.

    -We believe that explanations succeed on the grounds of practical adequacy, rather than causal sufficiency, so if we are consistently able to resolve enquiries in a satisfying way by retaining mental kinds, we have no reason to reduce.[2]

    PTJ has made two criticisms of the physicalism in Wendt’s position. His first criticism is that Wendt (and other CRs) are unwarranted in denying the causal epiphenomenality (or even believing in the reality!) of mental kinds. His second is that mental kinds are explanatorily useless given physicalism. Given the two reasons I listed above, I believe that PTJ’s first criticism succeeds but that his second fails; every token mental state must reduce to a neural state, and therefore the ideational is epiphenomenal,[3] but we cannot (yet) or should not reduce in our explanations.

    However, I am neither a philosopher of mind nor is my philosophy of science kung-fu as strong as Master PTJ’s – someday, inshallah – so I admit a high probability that I have missed something somewhere.

    [1] To paraphrase Fodor’s summary of reductionism for this particular discussion, we can define it as the claim that all token natural kind predicates in an ideally complete psychological science are translatable into natural kind predicates in an ideally complete neurophysiological science

    [2] This is my position, and I think it is well supportable through various forms of Pragmatism (notably those of Dewey and Laudan), but it would be unpalatable to a critical realist. I think critical realism is wrong, though.

  1. Pingback: Ideas, Causes, and IR Theory debates « Said Simon

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s