4 The Coupling-Constitution Fallacy
The coupling-constitution fallacy objection can take slightly different forms. According to Adams and Aizawa:
This is the most common mistake that extended mind theorists make. The fallacious pattern is to draw attention to cases, real or imagined, in which some object or process is coupled in some fashion to some cognitive agent. From this, they slide to the conclusion that the object or process constitutes part of the agent's cognitive apparatus or cognitive processing. (Adams and Aizawa 2001, 408)
Rupert expresses a similar objection, albeit in more cautious terms. Referring, again, to my version of the extended mind, he writes:
Rowlands, however, does not make clear why the use of an internally represented code applied to the contents of an external store implies HEC, rather than what it would seem to imply: HEMC. (Rupert 2004, 410)
HEC is the hypothesis of extended cognition, which Rupert correctly distinguishes from HEMC, the hypothesis of embedded cognition. What reason, Rupert asks, do we have for regarding the external processes as part of cognition rather than simply a form of extraneous scaffolding in which real-internal-cognitive processes can be causally embedded? That is, what reason do we have for thinking that the arguments for the thesis of the extended mind support anything more than the thesis of the embedded mind?
A charge of confusion can be interpreted in two distinct ways. On the one hand, it might mean that proponents of the extended mind have proceeded in their arguments blithely unaware that there is a distinction between cognition and its extraneous causal accompaniments. That is, these proponents are blithely unaware that there is a distinction between the constitution of cognitive processes and the extraneous structures with which such processes might causally interact. Such a charge is, frankly, ridiculous. Far from blithely confusing constitution and causal coupling, the most natural way of understanding the arguments for the extended mind is precisely as arguments for reinterpreting what had traditionally been regarded as extraneous causal accompaniments to cognition as, in fact, part of cognition itself. And, in general, to argue for the identification of X and Y, when X and Y had hitherto been regarded as distinct, if causally related, categories, is not to confuse X and Y.
Thus, consider my arguments for the extended mind, since these occasioned Rupert's charge of confusing extended cognition with embedded cognition. In The Body in Mind, I argued that, in certain cases, the external processes involved in cognition-bodily manipulation and exploitation of information-bearing structures in the cognizing organism's environmentpossess certain abstract, general features of processes commonly regarded as cognitive, while also differing in the sorts of concrete ways required by the integrationist underpinning of the extended mind. Thus, these external processes are employed in order to accomplish cognitive tasks. They involve information processing-the manipulation and transformation of information-bearing structures, and so on. This processing results in the making available to organisms of information that was previously unavailable; and so on. That is, I argued for the cognitive status of external processes of these sorts by trying to show that they satisfy a certain criterion of the cognitive.
The same is true, to take just one more example, of Clark and Chalmers's arguments for the extended mind. In the case of Otto, for example, they argue that the entries in his notebook qualify as a subset of his beliefs because they play a functional role relevantly and sufficiently similar to that played by beliefs when instantiated in an unimpaired brain. Thus, their arguments are an attempt to show that what we normally take to be a mere extraneous causal accompaniment to the domain of cognition should, in fact, be regarded as part of that domain itself.
There is, however, another more charitable way of understanding the charge of confusion. Now, the charge is not that defenders of the extended mind proceed blithely unaware that there is an important distinction between constitution and causation. Rather, it is that they are aware of the distinction but, in their arguments, for one reason or another, flout it anyway. For example, it might be argued that in The Body in Mind, my arguments for extended cognitive processes rested on a criterion of the cognitive that was never rendered sufficiently explicit and/or was importantly inadequate. Against Clark and Chalmers it might be argued that their arguments rest on a criterion of the cognitive as consisting in similarity with what is antecedently accepted to be cognitive, and that this criterion is, for one reason or another, inadequate.
These are reasonable charges, and they deserve to be answered. However, they are double edged. Suppose my arguments, for example, had been developed using an insufficiently explicit or otherwise inadequate criterion of the cognitive. Then, if I were in possession of an explicit, adequate, and properly motivated criterion of the cognitive, and if the sorts of external processes I identified in The Body in Mind were to satisfy this criterion, then I would have made it clear why my view entails the thesis of the extended mind rather than merely that of the embedded mind. In such circumstances, there would be no substance to the charge that I confusein either of the two senses-causation and constitution.
Therefore, more generally, like the differences argument, the couplingconstitution fallacy objection is derivative from the mark of the cognitive objection. If defenders of the extended mind can provide an adequate criterion of the cognitive and can demonstrate that the external processes they regard as cognitive satisfy this criterion, then there is no substance to the charge that they confuse constitution and mere coupling. In the next chapter I shall supply and defend what, I shall argue, is an explicit, adequate, and properly motivated criterion of the cognitive.
5 Cognitive Bloat
The cognitive bloat objection is, in essence, a slippery slope argument usually raised in connection with Clark and Chalmers's discussion of Otto's notebook. To recap, Otto is in the early stages of Alzheimer's and so keeps a notebook with him in which he writes down various facts that can help him with his day-to-day living, facts such as that the Museum of Modern Art is on 53rd Street. At least on one interpretation, Clark and Chalmers argue, notoriously, that the sentences in Otto's book can constitute a subset of his beliefs. The entries in Otto's book are literally beliefs that Otto possesses because they have a functional role in Otto's truncated psychological economy that is sufficiently and relevantly similar to the role played by beliefs in the psychological economy of Otto's unimpaired friend, Inga.
The cognitive bloat objection uses this claim as a starting point. If we are willing to allow that the sentences in Otto's notebook are beliefs, why stop there? Why not also allow that status for the entries in the telephone directory of which Otto makes frequent use? Why cannot these be numbered among Otto's beliefs? Indeed, why stop even there? Why does Otto not believe everything contained on the Internet, given that he is able to use this in a way akin to which he uses his notebook?
Clark and Chalmers try to preclude this problem of bloat by advocating a conscious endorsement criterion on beliefs. The entries in Otto's notebook count as beliefs, whereas the entries in Otto's telephone book do not, because Otto has, at some point, consciously endorsed the former but not the latter. However, this condition is questionable: beliefs can be formed subliminally, as well as through conscious experience, and, presumably, we would not regard a subliminal mode of formation as automatically excluding them from the class of cognitive states.5
In the previous chapter, I rejected Clark and Chalmers's claim that the entries in Otto's notebook form a subset of his beliefs. Instead, I formulated the thesis of the extended mind exclusively in terms of cognitive processes rather than cognitive states. Otto's process of remembering the location of the Museum of Modern Art could be a process that incorporated his manipulation of his notebook. But it does not follow from this that the entries in the book qualify as cognitive states. This latter claim should be rejected. Therefore, since the problem of bloat faced by Clark and Chalmers is one that pertains to cognitive states such as belief, and since I have reje
cted formulation of the thesis of the extended mind in terms of cognitive states, one might think that my version of the thesis is immune to this objection.
Sadly, however, this is not so: a parallel problem of bloat arises for cognitive processes. Suppose I am using a telescope.' Let us assume that the telescope is a reflector, and therefore works by transforming one mirror image into another. Mirror images are information-bearing structurestheir properties are systematically determined, by way of a mapping function that is itself determined by the specific properties of the mirror and by the properties of the visual environment. Therefore, the operation of the telescope is based on the transformation of information-bearing structures. I use these transformations to achieve cognitive tasks I could not have achieved without them-the perception of distant objects. So, the processes occurring inside the telescope are information-processing operations used to accomplish a cognitive task. So why, if the thesis of the extended mind is true, do these processes not count as cognitive? Relevantly similar examples can be easily generated. How can we rule out, for example, processes occurring inside my calculator, or my computer, from counting as cognitive ones? They too are information-processing operations that I employ in the accomplishing of cognitive tasks. If Clark and Chalmers's version of the extended mind faces a problem of bloat for cognitive states, then it seems my version will face a parallel problem for cognitive processes.
I shall argue-although this argument must be postponed until laterthat the cognitive bloat objection can be rebutted by way of an adequate criterion of the cognitive. The key to this rebuttal lies in the role played by the concept of ownership in qualifying a state or process as cognitive. The processes occurring in the telescope, computer, or calculator are not, at least as they figure in the cognitive bloat objection, owned by anyone. And anything that is to count as cognitive, I shall argue, must be owned by someone or some thing. Defense of this claim must be postponed until I have defended the criterion of the cognitive.
6 The Mark of the Cognitive Objection
The mark of the cognitive objection can take two forms. In its more cautious form, it claims that adjudication of the claims of the extended mind requires possession of an adequate criterion of the cognitive, and we possess no such criterion. In its more sanguine form, it claims that we do possess such a criterion, and it precludes the claims of the extended mind. I shall argue that neither the cautious nor the sanguine form of the objection can be sustained. I shall do this by providing what I shall argue is an explicit, adequate, and properly motivated criterion of the cognitive and showing that, far from contradicting the claims of the extended mind, it actually supports those claims.
The provision of an adequate criterion of the cognitive, therefore, would defuse both the mark of the cognitive objection and, if the arguments of this chapter are correct, the three other objections to the extended mind. In the next chapter I shall provide such a criterion.
7 Objections to the Embodied Mind
Now that we have reviewed the major objections to the thesis of the extended mind, it is relatively easy to see that each of these objections, with suitable modification, can also be applied to the thesis of the embodied mind. Note that these objections apply only to, and indeed are specifically tailored toward, the interpretation of the embodied mind thesis I endorsed in the previous chapter. That is, they apply to the embodied mind thesis only as understood as an ontic thesis of (partial) bodily constitution rather than the alternative interpretations-as an ontic thesis of bodily dependence or an epistemic thesis concerning the best way to understand cognitive processes. Indeed, these objections, in their different ways, all pose the question: what reason is there for thinking of the arguments for embodied cognition as establishing the ontic-constitution thesis rather than one of the weaker alternatives?
1. The differences argument There are significant differences between traditional-neural-cognitive processes, and the wider bodily processes invoked by the thesis of the embodied mind. This casts doubt on whether they can both be viewed as cases of a single psychological kind.
2. The coupling-constitution fallacy The thesis of the embodied mind confuses cognition with its extraneous causal accompaniments. That is, it confuses those structures and processes constitutive of cognition with structures and processes in which cognition is (merely) causally embedded.
3. Cognitive bloat Admitting wider bodily processes into the domain of the cognitive places us on a slippery slope. Where will it all end? Our conception of the cognitive will become unduly permissive, and we will be forced to regard all sorts of clearly non-cognitive processes as cognitive.
4. The mark of the cognitive objection The claim that wider bodily processes of the sort invoked by the embodied mind are cognitive violates any plausible mark or criterion of the cognitive.
Once again, and for the same reasons, the mark of the cognitive objection is the crucial one. All the other objections either presuppose the mark of the cognitive objection or can be resolved by the provision of a suitable criterion of the cognitive.
With regard to the differences argument, the thesis of the embodied mind will accept that there are significant differences between cognitive processes as these are traditionally understood and the wider bodily processes that it invokes and claims as cognitive. Given that these wider processes are not neural, significant differences are inevitable. However, if the thesis of the embodied mind can demonstrate that these wider bodily processes, when they are appropriately combined with relevant neural processes, can satisfy an adequate criterion of the cognitive, then these differences would in no way count against the claim that the wider bodily processes are cognitive ones.
According to the modified coupling-constitution fallacy objection, the thesis of the embodied mind confuses cognitive processes with wider structures in which these processes are embedded. In other words, it understands itself as arguing for the constitution of cognitive processes by wider bodily processes, but it really amounts to nothing more than the thesis that cognitive processes are causally embedded in wider bodily structures and processes. Again, however, if the thesis of the embodied mind can show that these wider bodily processes, when conjoined with the relevant sorts of neural processes, can satisfy an adequate criterion of the cognitive, then this charge would have no substance. By the provision of such a criterion, the thesis of the embodied mind would have shown that the wider bodily processes that we might have thought of as merely extraneous casual scaffolding should, instead, be regarded as among the cognitive operations performed by an organism.
Proper discussion of the cognitive bloat objection will have to be postponed until after the provision of what I shall argue is an adequate and properly motivated criterion of the cognitive. I shall provide such a criterion in the next chapter. The basic idea, however, is the same as the one advanced with the extended mind: an adequate criterion of the cognitive will contain an ownership condition: any process that is to count as cognitive must be owned by a cognizing organism or subject. This ownership condition, I shall argue, undermines the charge of cognitive bloat.
All these objections, therefore, ultimately trace back to the mark of the cognitive objection. What we have to show in order to defend the thesis of the embodied mind is that when we combine neural processes with wider bodily ones, then the resulting combination of processes satisfies an adequate and properly motivated criterion of the cognitive. We do not need to show that the wider bodily processes themselves satisfy this criterion. The thesis of the embodied mind is not the thesis that the wider bodily processes on their own, or in themselves, qualify as cognitive. That would be a silly thesis. Rather, the claim is that, when conjoined with the relevant sorts of neural processes, these wider bodily processes can satisfy the mark of the cognitive. Similarly, as we saw earlier, it is no part of the thesis of the extended mind to claim that processes occurring outside the body of the cognizing organism can, by themselves, count as cognitive. What qualifies as cognitive, according to the t
hesis of the extended mind, is the combination of neural processes plus processes of manipulating environmental structures.
In the next chapter, I shall provide and defend a mark or criterion of the cognitive. I shall argue that this mark accurately captures the sense of "cognition" involved in traditional-internalist-examples of cognitive science. And I shall further argue that, far from undermining the claims of the theses of the mind embodied and the mind extended, this criterion actually supports those claims.
8 Reconciling the Mind Embodied and the Mind Extended
The third type of problem facing the new science of the mind concerns not the theses of the mind embodied and the mind extended taken in isolation, but the possibility of combining these in any sort of consistent way. The problem is that each thesis rests on its own set of assumptions, and the assumptions underlying the thesis of the embodied mind and those underlying the thesis of the extended mind are not merely very different; they are, many think, actually incompatible. If this is correct, and if the new science is based on ideas of embodiment and extendedness, the new science appears to be in conceptual tatters. To see why, let us turn to the relevant assumptions made by each thesis, beginning with the extended mind.
Box 4.1
Functionalism
According to functionalism, mental phenomena are identified according to their causal or functional role. Consider an analogy. A carburetor is a physical object located somewhere in the innards of a car's engine (or older cars anyway-fuel injection systems have replaced them in more recent models). What is a carburetor? Or, more precisely, what makes something a carburetor? The answer is that a carburetor is defined by what it does. Roughly, it is something that takes in fuel from the fuel inlet manifold, takes in air from the air inlet manifold, mixes the two in an appropriate ratio, and sends the resulting mixture on to the combustion chamber. It is fulfilling this role that makes something a carburetor, and anything that fulfills this role in a car thereby counts as a carburetor. Most carburetors tend to look pretty similar. But this is at best a contingent fact, because it doesn't matter what a carburetor looks like as long as it fills this role. The details of its physical structure and implementation are of secondary importance compared to the role it fills, for it is filling this role that makes something a carburetor, and not the details of its physical structure or implementation. Of course, not every physical thing is capable of playing the role of a carburetor-a lump of Jell-O, for example, is not. So, the details of the how the functional role is physically implemented are not irrelevant. But as long as you have a suitable physical structure-one that is capable of fulfilling the role of a carburetor-then it doesn't matter what it is as long as it, in fact, fulfills this role.
The New Science of the Mind Page 13