Immaterial thought and embodied cognition

In a combox remark on my recent post about James Ross’s argument for the immateriality of thought, reader Red raises an important set of issues:

Given embodied cognition, aren't these types of arguments from abstract concepts and Aristotelian metaphysics hugely undermined? In their book Philosophy in the Flesh Lakoff and Johnson argue that abstract concepts are largely metaphorical.

End quote. In fact, none of this undermines Ross’s argument at all, but I imagine other readers have had similar thoughts, and it is worthwhile addressing how these considerations do relate to the picture of the mind defended by Ross and by Aristotelian-Thomistic philosophers generally.

Note first that metaphor doesn’t affect Ross’s point about the determinacy of thought in the slightest. Recall that to say that a thought is “determinate” in the sense Ross, Quine, Kripke, et al. have in mind is to say that there is a fact of the matterabout whether it has one rather than another among a possible range of meanings. To use Quine’s famous example, if I have the thought that gavagai, that thought will be determinate in the relevant sense if there is a fact of the matter about whether it has the content that there is a rabbit over there as opposed to the content that there is an undetached rabbit part over there or the content that there is a temporal stage of a rabbit over there. Now, whether I am using gavagai metaphorically is irrelevant. For example, suppose I am pointing to a human being as I have the thought. If the thought has a determinate content, then there will be a fact of the matter about whether I am describing the person metaphorically as a rabbit, or as an undetached rabbit part, or as a temporal stage of a rabbit. In short, to say that a linguistic utterance or thought is determinate in the relevant sense doesn’t entail that it is not metaphorical.

Second, though there is, accordingly, no need to get into a discussion of Lakoff and Johnson’s claims about metaphor in order to defend Ross, it should be noted briefly that it would be a serious mistake to suppose that anyone who endorses Ross’s claim that a linguistic utterance or thought can have a determinate meaning must be committed to a simplistic account of language that ignores the rich and complex ways that the meanings of words can be extended. On the contrary, the crucial role that the analogical use of language plays in human thought is a longstanding theme in Aristotelian-Thomistic philosophy, and Ross himself wrote an important book on the subject. Moreover, to suppose that we have to regard a concept either as having a straightforward literal meaning or as metaphorical is to assume a false dichotomy. Not all analogy is metaphor, so that a concept can have an analogical but still literal meaning. (See pp. 256-63 of Scholastic Metaphysics for a brief overview of the Thomistic approach to this subject.)

Third, Ross and other Aristotelian-Thomistic philosophers would by no means deny that human cognition is embodied. On the contrary, Aristotelians and Thomists have always insisted that embodiment is natural to us and to our mode of cognition. They do not regard a human being as a res cogitans or the body as something to which the human mind is only contingently attached (as Cartesian dualism implies) much less as a kind of prison (as Platonism holds). Rather, on the Aristotelian-Thomistic view, a human being is a substance to which both incorporeal and corporeal operations are essential, and the incorporeal ones (intellectual and volitional activity) require certain corporeal ones (namely sensation and imagination) as their natural concomitant. That is why the title of the article in which Ross first presented his argument is “Immaterial Aspects of Thought,” and the title of my article developing and defending the argument is “Kripke, Ross, and the Immaterial Aspects of Thought.” And that is also why, though the intellect is incorporeal and thus survives the destruction of the body, it cannot do much on its own. Death is not a liberation, but (as I have put it elsewhere) something like a “full body amputation” that leaves the human being reduced to an incorporeal stub. (See the posts linked to below for further discussion.)

However, there is a special way in which contemporary thinking about embodied cognition might seem at odds with Ross’s argument. The idea of embodied cognition has in recent years been closely connected with the notion of tacit knowledgeexplored by mid twentieth-century thinkers like Wittgenstein, Heidegger, Merleau-Ponty, Ryle, Polanyi, and Hayek, and developed further by more recent writers like John Searle, Hubert Dreyfus, and Charles Taylor. One way to sum up the basic idea is that all conscious and explicit knowing thatsuch-and-such is the case presupposes a background of inexplicit knowing how to cope with the world, where knowing how is a matter of having certain capacities, dispositions, and ways of acting rather than a matter of grasping propositions. These capacities, dispositions, and ways of acting are in turn largely bodily capacities, dispositions, and ways of acting, so that all of our explicit propositional knowledge ultimately presupposes embodiment.

Now, I have long been highly sympathetic to this line of thought, and have discussed Hayek’s application of it in a few places (here, here, and here). In fact it represents, I would argue, a rediscovery of an essentially Aristotelian conception of human nature. (This is a theme I develop in some forthcoming work.) But it might seem hard to square with Ross. One of the central theses of writers on tacit knowledge is that it can never in principle all be made explicit. Even when we make explicit some piece of knowledge that had been inexplicit or tacit, there is always some further body of knowledge that remains inexplicit and exists in the form of dispositions and habits of action rather than propositions. As Polanyi liked to put it, “we know more than we can tell.” Conscious and explicit knowledge is like the tip of an iceberg, and no matter how much of the iceberg you bring up above the surface, there is always more that is left underwater.

Now, if we have knowledge that is never explicit but rather embodied in habits and dispositions, then doesn’t that entail that it is not determinate in the relevant sense, especially insofar as it is embodied and thus sunk in materiality? And doesn’t that conflict with what Ross says?

It does not conflict with it at all. For one thing, the precise sense in which inexplicit knowledge of the sort in question might be said to be indeterminate needs careful spelling out. And it is not clear that it really is indeterminate in the relevant sense. Suppose you have the thought that it is raining, and when it is raining, traffic is bad, and from that thought draw the conclusion that traffic is bad. You have reasoned according to the inference rule modus ponens, but suppose you do not realize this because you have never taken a logic class and it has simply never occurred to you to consider that form of reasoning in the abstract, apart from the concrete examples in which you have deployed it. We might say that your knowledge of modus ponens is in this case tacit or inexplicit, embodied in certain habits or dispositions of thinking and speaking rather than as a proposition or rule you have ever consciously entertained.

Now suppose you take a logic class and explicitly learn the rule. You think “Hmm, I’ve always reasoned that way, though I never before really thought about the fact that that is what I was doing.” You now certainly have a thought with determinate content. But if what you grasp now is something you recognize as a rule you had always applied in the past, it is hard to see how what you knew in the past, inexplicitly or tacitly, was any less determinate in its content in the relevant sense than is the thought you have now. It always was the determinate rule modus ponensthat you were applying, even if you weren’t aware of it. The determinacy of the immaterial intellect arguably seeps down, as it were, into the body, so as to make determinate even tacit knowledge.

But put that aside, because there is a deeper point. Ross never says in the first place that every single thought we ever have is entirely determinate in its content in the relevant sense, and he doesn’t need to say that in order to make his argument. All he needs is the premise that some thought is determinate in its content. So, even if we were to concede that inexplicit or tacit knowledge is indeterminate, that would not affect Ross’s argument, because all he needs is the claim that some of our explicit thought is determinate in its content.

This objection to Ross would also seem once again betray a failure to understand the Aristotelian-Thomistic conception of human nature and how it differs from other conceptions. If a human being were an angel or a Cartesian res cogitans, then perhaps we would be able to say that everything we know we know explicitly and that all of our mental states and processes are entirely determinate in their content in the relevant sense. If a human being were entirely corporeal, as a non-human animal is, then we would be devoid of strictly intellectual activity and thus it could be said that everything we know we know only tacitly or inexplicitly and that none of our mental states and processes (i.e. sub-intellectual exercises in perception and imagination) has any determinate conceptual content in the relevant sense. But neither of these scenarios holds. In fact, human beings straddle the divide between the purely incorporeal and the purely corporeal, having one foot in the angelic realm and one foot in the animal realm. Hence we are mixtures of the determinate and indeterminate, the explicit and the tacit. And that we have at least some determinate mental content is all Ross needs for his argument.

Viral Trends

Immaterial thought and embodied cognition