|
The Distinction Between Mind and Cognition
Tim van Gelder
Appeared in Y-H Houng (ed) in Mind and Cognition: Collected Papers from the 1993 International Symposium on Mind and Cognition; 1-29.
"Excuse me, what do you think of the computer chip shortage?" He scratched his head and replied "Shortage? What's a shortage?" She went to Warsaw, and asked a man in the street: "Excuse me, what do you think of the computer chip shortage?" "Computer? What's a computer?" In Moscow, she asked a man in the street "Excuse me, what do you think of the computer chip shortage?" "Think? What is "think"?" She went to Tel Aviv, and asked "Excuse me, what do you think of the computer chip shortage?" "Excuse me? What is this thing, excuse me?" In Taipei, she asked a man in the street "Excuse me, what do you think of the computer chip shortage?" The man disappeared for five minutes and came back with a brand new computer chip. "How many do you want to buy?" One year there was a shortage of computer memory chips. A reporter was sent out to gauge public opinion on the matter. She went to New York and asked a man in the street: "Excuse me, what do you think of the computer chip shortage?" He scratched his head and replied "Shortage? What's a shortage?" She went to Warsaw, and asked a man in the street: "Excuse me, what do you think of the computer chip shortage?" "Computer? What's a computer?" In Moscow, she asked a man in the street "Excuse me, what do you think of the computer chip shortage?" "Think? What is "think"?" She went to Tel Aviv, and asked "Excuse me, what do you think of the computer chip shortage?" "Excuse me? What is this thing, excuse me?" In Taipei, she asked a man in the street "Excuse me, what do you think of the computer chip shortage?" The man disappeared for five minutes and came back with a brand new computer chip. "How many do you want to buy?" Many people currently believe that the perennial philosophical question what is mind? has recently, in the manner of so many philosophical problems before it, been turned over to science. The mind is now the subject of the exciting new sciences of cognition, cognitive science and cognitive neuroscience. Cognition is that inner realm of structures, states and processes that are causally responsible for our sophisticated behaviors. The universal assumption seems to be that mind and cognition are the same thing; to investigate cognition is to investigate mind. Within the cognitive sciences there is a great deal of confidence that a full understanding of mind is just around the corner. So, as philosophers wanting to understand what mind is, perhaps we should just wait and see what the cognitive sciences come up with - or, if impatient, we should roll up our sleeves and become a cognitive scientist ourselves. A somewhat disconcerting feature of this approach is the threat of eliminativism. Increasingly, cognition as described by the sciences looks very unlike mind as people ordinarily think of it. But if mind is cognition then mind is whatever the cognitive sciences say it is. If cognition turns out to be radically unlike mind as the folk think of it, then so much the worse for the folk mind; it goes the way of fairies and the flat earth. Of course, there are various different research paradigms in the cognitive sciences, and some describe an inner world which, superficially at least, seems more in line with the folk conception than others. The trouble is, that generally speaking, the lines of research that seem most promising are precisely those which describe a phenomenon most radically unlike mind as ordinarily conceived. This gap and its implications have of course already been emphasized by philosophers. Paul Churchland is famous for maintaining that the folk mind is being radically revised and even in large measure eliminated by the relentless and exciting advances of the cognitive sciences (Churchland, 1988). There are at least two ways of responding to the emerging gap between the new science of cognition and the old conception of mind. The brave-hearted will forge ahead with the cognitive sciences and accept the radical revisions they demand. The timid, however, will go back and reconsider the identification which leads to the disconcerting implications. They'll ask, in a purely philosophical vein, whether mind really is the same as cognition. In other words, the threat of eliminativism suggests that we should ask again what mind is, for whether we should radically revise our conception of mind in the light of the cognitive sciences depends on first determining whether mind is what they're really studying. 1. Cartesianism and The Mind as Cognition Doctrine. In identifying mind and cognition, the cognitive sciences are falling in with a long and honorable philosophical tradition, one that extends back at least as far as Descartes and Hobbes. The single unifying idea, of course, is that mind is that inner realm of states and processes that is responsible for our sophisticated behaviors. Important ideas usually come in clusters. The mind-as-cognition doctrine is not so much a single idea as a family of closely related ideas which together form a broad conception of mind. Descartes is, both historically and conceptually, the most important philosopher of this tradition. The broad outlines of Descartes' picture are pretty familiar: mind and body are fundamentally distinct substances, the essence of mind is thought while the essence of body is extension, mental states and events are modifications of mental substance, mind and body interact in perception and action via animal spirits in the pineal gland. For current purposes I want to bring attention to seven major features of this picture. According to Descartes, mind is: 1. Ontologically Homogenous. Roughly, this means that one basic ontological story works for all mental entities. There are two sides to homogeneity. The first is that mind is not made up of a number of fundamentally different kinds of entities. The second is that there is only one account of the nature of any given mental entity. To get a better feel for the homogeneity of the Cartesian mind, consider an example of something that is ontologically heterogeneous. A national economy is a structured totality made up of a wide range of different kinds of things-consumers, financial instruments, products, factories, rules, exchanges-interrelated and interacting in complex ways. The various entities that comprise an economy belong to quite different ontological categories. A financial exchange and a factory are different in ways that any adequate ontological story must bring out. Further, for any given entity, multiple different accounts of the nature of that entity can be given. A chemist and an economist will describe a dollar very differently. The Cartesian ontology of mind is homogenous because, although there's a diversity of particular mental entities, they're all modifications of thinking substance. Moreover, this account, suitably elaborated, is the only story that tells us what mental entities are. 2. Inner. For Descartes, mind was essentially something inner. What does this amount to? For crude materialists such as Hobbes, mind is inner in the straightforward sense that all mental states and events are spatially located inside the skull. This can't be the sense in which mind is inner for Descartes, however, since in his view mind is not in space at all. The property that makes the difference between inner and outer for Descartes is that of introspective accessibility. A mental entity is inside my mind if it is introspectively accessible to me. The boundaries of minds are exclusive and exhaustive; every mental entity is introspectively accessible to exactly one mind. It is this structural feature of the mental domain that makes it possible to say that minds are essentially inner: their contents are always inside a boundary of accessibility that excludes all observers but one. Famously, Descartes thought of mind as distinct from body. This distinctness really has at least three quite separate components: 3. Ontologically Different. Mind and body are fundamentally different kinds. Though they are both substances, mind is a very different kind of substance than body is; for example, mind is not, like body, "in space". 4. Independent. Mind and body are metaphysically independent: neither needs the other in order to be what it is, or to be at all. Mind can exist without the material world, and the material world can exist without mind. Note that difference in kind and metaphysical independence are logically independent of each other. Two pieces of paper are of the same ontological kind but quite independent. A country's legal constitution, on the other hand, is completely dependent on the its people - if there were no people abiding by certain principles, there would be no constitution - yet metaphysically, it is of a different kind. 5. Disembodied. The Cartesian mind is fundamentally disembodied. To say this is not just to repeat the point that it's of an ontologically different kind. A constitution is of a different kind than the people, but the constitution is embodied in the people; it just is the people and their practices considered from a certain point of view. But the Cartesian mind is not made up of body in anything like this way. As a consequence, the Cartesian mind is entirely un-bodily in its character. Mind is not shaped in its nature by any connection with the body. The distinction between disembodiment and the other two aspects of distinctness is subtle but still crucial, for, as we'll see, it's possible to have a conception of mind which rejects ontological difference and independence but is still in a certain way disembodied. 6. Representational. A much-remarked-upon feature of the Cartesian mind is that it is representational. Mental entities present the world as being, or possibly being, a certain way. Insofar as mind has a primary role, function or purpose at all, it is knowledge. 7. The Engine of Behavior. For Descartes, the mind is an essential part of the inner causal mechanisms underlying human behavior. Though producing behavior is not necessarily what mind is for, is it one important thing that it actually does, and our behavior would be impossible without it. Any full causal explanation of human behavior must make reference to mental activity. In short, Descartes thought that the mind is an ontologically homogenous and distinct inner representational realm that is causally responsible for sophisticated human behavior. Since Descartes' account clearly belongs squarely to the mind-as-cognition tradition, I describe that tradition as Cartesian, even though some in that tradition may deny particular aspects of Descartes' picture. In the seventeenth and eighteenth century the ontological distinctness of mind and body came under heavy attack. Distinctness was, most famously, thought to result in insuperable problems in explaining the interaction of mind and body. If you reject the three distinctness theses but keep all the other major features of Descartes' picture, the natural position to take up is a brain-based materialism: as La Mettrie put it (La Mettrie, 1748/1912), the brain is an organ for thinking just as the stomach is an organ for digesting. When brain-based materialism is placed on a scientific footing, it becomes the cognitivism characteristic of mainstream symbolic cognitive science. Jerry Fodor's vision of the mind (Fodor, 1975) just is Descartes' picture with dualism replaced by functionalism and an attempted computational implementation. For Fodor, mind is that inner realm of representational states which, via a computational architecture, is causally responsible for our sophisticated behavior. The only thing Descartes was wrong about was distinctness. From this lofty perspective, even Fodor's arch-rival, the eliminativist Paul Churchland, comes across as a Cartesian, for he too believes that mind is that inner realm of representational states which, via a neuro-computational architecture, is causally responsible for our sophisticated behavior. Descartes, Fodor and Churchland just disagree about the nature of that inner stuff. 1a. Problems for Cartesians The Cartesian conception of mind as the inner engine seems very natural to scientists studying the brain and cognition. It is also rather appealing, since it flatters them by pronouncing that what they are studying is nothing less than the mind. However, it turns out that, for a variety of philosophical reasons it is difficult to maintain a pure Cartesian conception of mind as an inner realm. The problem is containing the mind within its supposed boundary. There is philosophical pressure pushing mind out into the world. In denying Descartes' distinctness thesis, materialists from La Mettrie to Fodor took themselves to be overcoming the Cartesian disembodiment of mind. Mind is not something other than the body; it just is the brain and its operation, considered at a suitable level. Yet one of the most common kinds of criticism of mainstream cognitivism is, precisely, that it is operating with a disembodied conception of mind. Despite its official commitment to brain-based materialism, it is accused of being a residual form of Cartesian dualism. How can this be? The allegation against cognitivists is that their identification of mind with the brain and its operation is a purely formal maneuver, and that they retain a conception of mind which is still insufficiently bodily in nature. There are two basic strategies here. The first is to assume that cognitivists were right in identifying mind with the brain and its operation, but failed to acknowledge the inevitable consequence: that the mental would be fundamentally shaped by this embodiment. Thus one kind of critic argues that, since mind is identical with the brain, mental states and processes must be fundamentally neurobiological in character. These critics are, in other words, emphasizing that mind is embrained. Patricia Churchland is a good example (Churchland & Sejnowski, 1992). Another kind of critic points out that the brain is embedded and arose in a real live human body, and this human embodiment fundamentally shapes the nature of mind; a good example is Mark Johnson in his book The Body in the Mind (Johnson, 1987). Another more radical approach questions whether brain-based materialists made the correct identification in the first place. Mind must be embodied, but why does embodiment mean being identical just with the brain? Where should the spatial boundary around mind be drawn - at the skull, at the skin, or perhaps out in the world at large? The difference between these radical embodiers and their conservative cousins is ontological. The radicals insist not just that embodiment shapes the mental, but that the human body and perhaps even some part of the world is part of or integral to mind. John Haugeland has presented just this kind of radical critique of mainstream cognitivism (Haugeland, 1993). He asked: on what principled basis can a boundary around mind be drawn? He proposed that its appropriate to locate boundaries around systems wherever there are simple or restricted "narrow bandwidth" interfaces: the closer or higher-bandwidth the coupling, the more the complexity is shared across a single system, and the less it's split between two interacting systems. But by this criterion, he thinks, mind cannot be contained within the skull and not even within the body. Bodily skills, public equipment, social organization, even the interstate to San Jose turn all out to be integral to mind. The boundaries of mind are pushed out far into what we standardly think of as the external world. But where if anywhere does this radical embedding stop? As Haugeland asks: "How much "furniture" can we bring back in?" The most extreme possibility is that the boundaries of mind are pushed out to the very horizon. The whole world would then be part of my mind. But there are some obvious problems with this. First, since the whole world is presumably integral to every mind, all minds would be the same and there would be only one mind. Second, you, as part of the world, would be part of my mind and vice versa. But these implications are, surely, absurd. The trouble is that this most extreme form of radical embedding is just Haugeland's radical embedding in the limit, and problems that are unavoidable in the extreme case are present in the more moderate version as well. Thus on Haugeland's account, if both you and I use the interstate to San Jose, then the interstate is integral to both of our minds; our minds overlap. If you are part of the world with which I enjoy high-bandwidth interaction, then you are integral to my mind, and vice versa. Minds seem to have become strangely intertwined. Radical embedding thus complicates the distinction between one mind and another. If the boundary between mind and world is outside the body but not so wide as to include the whole world, then it must presumably lie somewhere in between. But where, exactly? Haugeland suggests that there's no real answer to this question. There's no way of drawing a determinate line between mind and world. But the lack of any fixed or well-defined boundary seems to have an unfortunate consequence: the mental is not a "well-defined realm" at all, and there's no clear mind/world distinction. Thus, the radical embedder has a difficult time supporting crucial distinctions between mind and mind and mind and world. The very idea of a determinate boundary that mind can be inside becomes problematic. These difficulties suggest that there is some ontological confusion in the vicinity. One possibility is that this confusion is particular to the radical embedder, and that it can be avoided by retreating to a more cautious version of embodiment: perhaps mind is really only embodied in the brain or in the body. The trouble is that Haugeland's arguments in favor of the radical embedding position are quite compelling: if you are tempted to insist that the inner engine be properly embodied at all, then you probably should be a radical embedder. Another possibility is that the confusion is really inherent to the whole embodiment enterprise as it is standardly conducted. Notice that embodiers typically retain the Cartesian conception of mind as inner. They either hope to show that those inner entities are fundamentally shaped by embodiment, or go further in arguing that the boundary between mind and world, inner and outer, must be pushed outwards. Perhaps this innerness is the source of the problem. The difficulty for the radical embedder is that it appears that the best way to draw a boundary between mind and world is one that draws so much of the external world into the mind that it is difficult to distinguish mind from world. Perhaps it is a prior mistake to think of mind as behind some kind of boundary, such that the possibility of a mind/world distinction turns on making that boundary determinate. In this way the insistence that the inner engine be properly embodied ends up throwing into question the very idea that mind is a purely inner realm. In doing this, it joins forces with some other major streams of twentieth century philosophical thought. The Cartesian conception of mind as inner has already come under heavy criticism in the light of two very serious epistemological problems. These are the problem of knowledge of the external world (how can I know that the external world is really there, or anything like I think it is?) and the problem of other minds (how can I know that there's really a mind in there behind the facade of your behavior?). From an epistemic point of view, the one problem is how to get out to the world from inside the mind-world boundary; the other is how to get in to the mind from outside that boundary. Now, there are at least two major kinds of response to these problems. The first kind is to retain the basic Cartesian metaphysics of mind as an inner realm, and to argue that somehow the boundary between mind and world can be epistemically crossed. Thus Descartes argued that knowledge of the external world is underwritten by God's guarantees; and, coming from another direction, Paul Churchland has argued that because the folk conception of other minds is a theoretical one, there's no more of an epistemological problem about your mind than there is about any other theoretical entity (Churchland, 1979) The second kind of approach is more radical: it questions the Cartesian metaphysics which gives rise to the epistemological problems. The conception of mind as inner is held to be one of the major culprits. If the mind is "in here" and the world is "out there" then there can be a problem about how I, in here, can really know what is going on outside. Conversely, if your mind is "in there" then it is difficult for me to be sure that you actually have a mind. But if mind is not essentially inner then these problems cannot arise. The most powerful solutions to the epistemological problems are thus ones which reject the conception of mind as inner. For example, Heidegger argues that the problem of knowledge of the external world is eliminated if mind is properly reconceptualized as Dasein which is essentially "worldly" to begin with (Guignon, 1983). Alternatively, Gilbert Ryle has argued that if mind is not a ghost in the machine but rather primarily a matter of what we publicly do, then there can no great difficulty for me in knowing that you have a mind (Ryle, 1949/1984). In short, both attempts to properly embody mind, and attempts to deal with deep epistemological problems, put into question the Cartesian idea of mind as some inner realm. But what's the alternative? Mind as outer? What could that mean? 2. Individualism, Holism and Internality Consider a useful distinction between two very general ways of understanding what makes a thing what it is. Roughly, holists see the fundamental nature of things as depending only upon some larger whole to which they belong, and individualists see the nature of things as depending only on what they're made up of. Suppose we say that some entity is constituted as an A by B if it's being an A depends on or is a function of B. Then metaphysical holists see constitution as solely a matter of context; an individual entity is what it is in virtue of some larger structure or pattern into which it fits. Metaphysical individualists take context to be irrelevant; only facts about the individual entities themselves are relevant to their constitution. Most important is the makeup of an entity - its internal structure and the parts of which it is composed. Corresponding to the holism and individualism of constitution there is the holism and individualism of intelligibility. If an entity is constituted as an A holistically, then to understand that entity as an A you have to understand the relevant larger whole and how As fit into it. To render an A intelligible as an A is to articulate its place in the larger whole. Alternatively, if an entity is constituted as an A individualistically, then to understand that entity as an A you have to understand how As are made up out of their parts. To render that entity intelligible as an A is to break it down and exhibit its characteristic parts and internal structure. An example borrowed from John Haugeland (as, in fact, are these distinctions themselves) will illustrate these distinctions at work. Think of a king in chess, and ask what makes it a king. The individualist will focus on properties of the piece itself: a king is typically tall and often has a cross-like configuration on top. A particular king will have various physical properties: it'll be made of plastic, wood or ivory; it'll have a center of mass placed centrally over its base, and so forth. To understand the king fully would be to understand completely the details of its makeup, right down to the microphysics. To render it intelligible would be to explain these details. The holist, by contrast, will ignore the properties of the piece and focus on the game. What makes something a king is that it plays a certain role which is defined by the rules of chess. Kinghood is determined not by individual makeup-in that sense, a king can be anything from a hunk of ice to a helicopter-but by the place in the game as a whole. To understand the king fully would be to know how to play chess well, since one then understands fully the role that kings play. To render the king intelligible would be to explain the game of chess and the place of a king in a game. Now, the Cartesian conception of mind is, clearly, metaphysically individualist. It makes sense for Descartes to imagine that minds could exist without the external world; thus, minds do not depend on the external world to be what they are, and are intelligible independently of it. A little less obvious, however, is the connection between metaphysical individualism and two of the key features of the Cartesian conception. For the individualist an entity is constituted by what it is made up of. If one demands that mind be rendered intelligible in individualistic terms, one will focus on the internal elements that make it up, which for Descartes of course were particular, introspectively observed mental states. Consequently mental states are inevitably thought of as inner and the mind as a special inner realm. The internality of the mental is in this way part and parcel of an individualist metaphysical orientation. Descartes was, as mentioned, also a natural scientist interested in understanding the human body and behavior. An individualist approach to the explanation of human behavior will seek its causes internally; hence Descartes' deep interest in neuroanatomy and neurophysiology. However, given his conception of the body as a machine, he believed that human behavior was not fully explicable in terms of internal mechanistic causes alone. Hence the need for another kind of internal cause. The mind, conceived individualistically as a domain of inner states, was the perfect candidate. In this way, a conception of the mind as the inner engine room of sophisticated behavior is also a natural concomitant of an individualistic metaphysical orientation. 3. The Heideggerian Conception of Mind as Enacted Social Role If the Cartesian conception of mind as inner follows from metaphysical individualism, then perhaps a conception of mind as outer will follow from a metaphysical holism. In this section I will sketch such a conception. Since it is inspired by one way of reading Heidegger (in particular, that of John Haugeland (1983)), I will give it the label "Heideggerian", even though it is not intended here as an accurate representation of what Heidegger actually thought, but at best has a vaguely Heideggerian flavor. The basic move is to think of minds as analogous to kings in chess. It is fairly clear that the metaphysical holist has the upper hand in understanding what a king is. The individualist offers a certain kind of understanding of the piece, but not of the king as such. Perhaps, in a like fashion, minds will be best understood as minds holistically. Perhaps the descriptions of internal mechanisms offered by the Cartesian individualist, whether dualist or materialist, are about as relevant to understanding minds as they are to understanding kings - which is to say, barely at all. If minds are constituted as minds by some larger whole to which they belong, what might that larger whole be? The broadly Heideggerian answer is: the society. The constitutive role of the society can be understood at two levels: for minds as such, and for individual mental entities. Taking minds first: the claim is, roughly, that to have a mind is to play a socially defined role, and it is the social role that constitutes any behavior as mental. More precisely, to have a mind is to enact a particular, complex, interlocking nexus of social roles. This deserves a little explanation. What is it to be a fireman? We could take an individual fireman into the laboratory and examine him very closely, but of course we wouldn't thereby find out what it is to be a fireman. To understand that, we have to first understand property and its importance to human endeavors, its tendency to burn down, and then the specific social structures whereby certain people are given the special task of dealing with fires. A more-fine grained understanding would come from seeing what specific rules and conventions firemen must abide by, such as the need to remain at the fire-station while on duty, to answer emergency calls, and so forth, right down to codes of dress and etiquette. It is this large and complex institutional and normative structure which both makes it possible for anyone to be a fireman, and constitutes certain individuals as firemen. Thus someone might go through motions that are physically identical to those of a fireman, such as holding a hose to a fire, but unless that person is embedded appropriately in myriad other ways in the social structure, those actions will not count as those of a fireman. So, to be a fireman is to play a social role; to a first approximation, anything that played such a role fully would be a fireman. What is it to be a person? The suggestion isn't that to be a person is just to enact the role of a person, since there is no single coherent structure of norms and practices corresponding to being a person. Person is, as Ryle would have put it, a polymorphous concept, a little like that of dilettante. There are no specific things that one must do to be a dilettante; rather, one has to do a wide variety of things in a characteristically superficial manner. Similarly, to be a person is to enact a wide variety of interlocking roles, each of which constitutes an individual as a more specific social kind. Thus, in my case, to be a person is to simultaneously play the roles of teacher, resident alien, bachelor, cappuccino communist, and squash player, among others. I am also at least expected to enact a number of other roles which are themselves polymorphous and must be played concurrently with all the others: conversationalist, gentleman, law-abiding citizen, friend, and so forth. A person is someone who manages, by behaving appropriately in context, to consistently combine some sufficient number of such roles, and anything that does this counts as a person. Now, on this Heideggerian conception, to have a mind is to be a person. That's to say, personhood, if not identical to mind, is at least necessary and sufficient for it. Perhaps the best argument is simply to look at what, in ordinary practice, we unhesitatingly take to have a mind: all and only those things which are persons in the above sense. There are borderline cases, of course-cats, infants, the severely handicapped-but these cases are borderline both in their ability to function as persons and in whether they have minds. Another way to support the identification of mind with enacted social role is to provide illuminating and persuasive descriptions of particular mental entities from the same perspective. Thus, on this account, to believe that p is to enact a kind of social micro-role. To believe, as I do, that Macintoshes are good computers is to behave in ways that, by virtue of the institutional and normative structures that surround me, count as having that belief. The point here is deeper than just that this belief is, as a matter of practical fact, possible for me, whereas it wasn't possible for my grandmother at the same age, and it is the state of society, rather than anything about our individual bodies, which makes the difference. The point is deeper because the wider social structures don't merely make possible believing-that-p behaviors, they constitute those behaviors as believing-that-p. Had my grandmother, by virtue of some vastly improbable brain malfunction, suddenly started behaving in the kinds of ways that, in me, count as believing that Macintoshes are good computers, she still wouldn't have had that belief. One reason is the Wittgensteinian point that to believe that Macintoshes are good computers requires having the concepts Macintosh and computer , and having these concepts requires the ability to apply them correctly, and correctness of application is something that turns essentially on social agreement - something that wasn't around in her time. Thus a kind of behavior only becomes a believing-that-p in an appropriate social context. The individualist account of believing-that-p, which tells us the internal cognitive structures that lead to a certain kind of behavior, and which presumably could explain my grandmother's bizarre behavior, can thus no more be an account of what it is to believe-that-p than the individualist's descriptions of a particular chess piece are accounts of what it is to be a king. The Heideggerian metaphysical holist, then, denies the Cartesian internality of mind by making mind completely "outer". The social roles themselves are abstract properties of whole communities. All it takes for an individual to have a mind is enact a particular nexus of such roles, and this enactment is simply a matter of appropriate behavior, perfectly external and public. For the Heideggerian, it simply doesn't matter what goes on inside someone's head, any more than it matters what a chess king is made out of. And in rejecting the Cartesian internality of mind, the Heideggerian also rejects another Cartesian principle, that mind is the engine of sophisticated behavior. Since mind is behaving the right way in social context, mind cannot be the internal cause of that behavior. 3a. Problems for Heideggerians. At this point, two familiar philosophical characters enter the stage. The first is Robinson Crusoe - a man shipwrecked on a deserted island, and therefore without the social context upon which mind is supposedly dependent. The second is the zombie, someone who behaves completely appropriately in social context, but has no conscious inner life at all. People usually think that Robinson Crusoe has a mind despite lacking the required socio-behavioral context, and that the zombie lacks a mind despite behaving appropriately. Perhaps, then, the individualist was right after all: having a mind is simply a fact about the individual, and in particular, a matter of the right kind of inner life. Robinson Crusoe and the zombie are intuition pumps, and the intuitions they pump up are almost entirely worthless. To be tempted, by such examples, to suppose that the Heideggerian is wrong about mind just is to be tempted by individualism, and so such temptations can't be used to support individualism against holism. At best, we should take such cases and the intuitions they pump up as pointing, possibly misleadingly, in a certain direction, and then ask whether there are any good arguments that take us there. One might argue, for example, that any object that could play the kinds of roles that allegedly constitute minds would, unlike the chess piece, be enormously sophisticated. It could only generate the right behavior by having a great deal of internal complexity and activity, including, presumably, internal representations and operations on those representations. This internal complexity would give the object a highly complex set of dispositions to behavior, and these dispositions would survive removal of the wider context in which they normally operate. Shouldn't we therefore identify mind with this complex inner representational life? The Heideggerian can cheerfully admit these points about the complexity and dispositional properties of anything that could enact the social roles required for mind. It doesn't follow from these points that we must be individualists about mind. Mind is highly complex, and the internal causal mechanisms that allow us to be persons are surely highly complex, and these parallel considerations tempt us to identify one with the other. But to actually make that identification would be to reassert an individualist metaphysics; it would be to suppose that having a mind is constituted by facts about the individual alone, quite independently of the larger whole into which it fits. But that is precisely what is at issue. Pointing out how complex cognition must be will not convince the Heideggerian that it is not a category mistake to identify it with mind. In short, the argument simply begs the question against the Heideggerian. If there is an argument against holism, it must find a deeper conceptual difficulty with the position. It is, I think, this. It has been pointed out already that there is no single unified structure of institutions and norms which define what it is to have a mind. Rather, each mind is an enactment of a distinctive nexus of roles, where many of those roles are themselves polymorphous. In other words, the larger social context does not fix the character of any particular mind, just as the norms of conversation do not determine the course of any particular conversation. Consequently, each human realizes personhood in a more or less unique way. It follows that individual minds are not wholly intelligible as minds on the basis of the larger whole into which they fit; to fully understand a given mind one must understand not only the social context that constitutes it as a mind but also the distinctive contribution of the individual to the realization of the possibilities afforded by that whole. And failure of holistic intelligibility indicates a failure of holistic constitution; minds are not constituted as minds solely by the larger structures into which they fit. 4. The Rylean Conception of Mind as Activity Up to this point I have suggested that neither the Cartesian conception of mind as inner nor the Heideggerian conception of mind as outer is adequate. Perhaps the problem lies with the inner/outer dichotomy itself; perhaps mind is neither inner nor outer. This possibility is the basis of the Rylean conception of mind. Ryle is famous for rejecting the Cartesian "ghost in the machine". He did not do so, however, by claiming that the mind was outer as opposed to inner; for such a move would be to maintain a problematic boundary between inner and outer. One way to have a conception of mind which is neither inner nor outer is to characterize mind in terms of something which spans both of them, as standardly conceived. Ryle did this by taking the essence of mind to be, very roughly, human activity. As he put it, 'my mind' does not stand for another organ. It signifies my ability and proneness to do certain sorts of things and not some piece of personal apparatus without which I could or would not do them. (p.168) But if mind is activity, it's activity of a very special sort. At a first pass, it's the kind of activity that is described in the everyday mentalistic vocabulary: terms such as 'careful', 'unobservant', 'ingenious', 'vain', 'credulous', 'witty', and hundreds if not thousands of others. According to Ryle, these terms overwhelmingly describe either dispositions to act or active episodes in the light of dispositions. Thus, to describe John as believing that p is to describe him as tending to do any of a wide variety of particular things; or to say that John drove home carefully is to say that he drove home observing the speed limit, watching for pedestrians, etc., and that if a car in front had stopped suddenly he would have applied the brakes in time, etc.. As described so far, it might sound as if Ryle is occupying what is currently the single most unfashionable position in the philosophy of mind, pure philosophical behaviorism. Contemporary philosophers do in fact unanimously describe The Concept of Mind as the magnum opus of that school. The philosophical behaviorist is someone who believes that the mental can be reduced to the behavioral, in the sense that ordinary mental terms can be wholly defined in terms of observable, non-mental behavior or dispositions. Now, it is possible that there once was such an animal; perhaps Carl Hempel for a brief moment. But if one actually reads Ryle it is clear that he was never a philosophical behaviorist in this sense, and since it is basically Ryle's view that I want to contrast with the Cartesian and Heideggerian approaches, it is worth bringing this difference out. There are at least three distinct components of strict philosophical behaviorism: that mental terms are definable in behavioral terms, that this behavior must be non-mental, and that it must be observable. Now, Ryle certainly never asserted any of these particular theses, but neither did he bother to explicitly deny them, since he wasn't thinking about them at all. He did however deny these claims implicitly in numerous places. Thus, he claimed that the dispositions picked out by ordinary mental vocabulary are what he called indefinitely heterogeneous: they can be manifested in an "indefinitely" or even "infinitely" wide variety of ways, depending on the circumstances. This heterogeneity alone means that it is impossible to define the mental in terms of behavior. Ryle also allowed that the activity in which mind is manifested can itself be mental and even internal. Consider his characterization of belief: to believe that the ice is dangerously thin is to be unhesitant in telling oneself and others that it is thin, in acquiescing in other people's assertions to that effect, in objecting to statements to the contrary, in drawing consequences from the original proposition, and so forth. But it is also to be prone to skate warily, to shudder, to dwell imagination on possible disasters and to warn other skaters. It is a propensity not only to make certain theoretical moves but also to make certain executive and imaginative moves, as well as to have certain feelings. But these hang together on a common propositional hook. (p.135) If to believe that p is to have a disposition to act in p-appropriate ways, its manifestations include having other mental episodes, such as imaginings and emotions; and these mental episodes, if they are anywhere at all, take place inside the skull, and are straightforwardly unobservable to another person. The contrast between Ryle and the much-ridiculed philosophical behaviorist could thus not be more stark. In allowing that the mental can be inner as well as outer, Ryle is denying the Cartesian internality of mind without placing mind outside an inner/outer boundary. He does not, like radical embodiers, push the boundary of mind out into the world at large; nor does he, like Heideggerians, insist that mind is essentially out there in a social world. To think of mind as inner, he often said, was to make a category mistake; but so also is thinking of mind as outer as opposed to inner. 4a. Problems for Ryleans The problem with Ryle's position is that he throws out the metaphysical baby with the Cartesian bath water. That is, in rejecting the Cartesian conception of mind as essentially something inner, Ryle characterizes mind in a way that is neither metaphysically individualist nor metaphysically holist; indeed, this was, as far as I can tell, a distinction to which he was oblivious. Consequently Ryle can be criticized for failing to take into proper account the way in which mind can be constituted by both the inner and the outer. Thus, Ryle is very often attacked for not providing an adequate account of the inner phenomenal or cognitive life and its essential contribution to having a mind. While these attacks are often quite misguided, there is a core of truth in them, and Ryle himself acknowledged that the chapters in The Concept of Mind on inner conscious episodes such as sensations and images were inadequate. Further, in seeing all mental episodes, whether inner or outer, as just manifestations of dispositions, he was unable to acknowledge any causal explanatory role for inner mental episodes. It is one thing to deny that the mind is just the engine of behavior; it is quite another to deny that inner mental episodes have anything to do with the generation of behavior. Ryle likewise failed to sufficiently acknowledge the constitutive role of the larger social whole. Thus, to take just one example, the holist's arguments that a community is necessary for any individual to have beliefs in a full-blooded sense work as well against Ryle as they do against the Cartesian. In short, for Ryle, Robinson Crusoe would have a mind even if he were a zombie. But this position, far from overcoming the dichotomy between inner and outer, individualism and holism, is vulnerable to both sides of it. 5. Mind as Both Inner and Outer Thus far I have considered three possible positions: the Cartesian who sees mind as the inner engine, the Heideggerian who sees mind as the outer enacted role, and the Rylean who sees mind as activity, neither essentially inner nor outer. All three positions have deep problems: the Cartesian is pulled outwards, the Heideggerian is pulled inwards, and the Rylean is pulled in both directions. The proper response, I think, is to occupy the only remaining position: to see mind as both inner and outer. We must reject the presumed dichotomy between individualism and holism, not by rejecting both sides of it, but by rejecting it as a dichotomy. Before briefly describing an ontology of mind along these lines, I will briefly discuss an example which, I think, actually forces such an ontology upon us. Think back to the joke at the beginning of this paper. Getting the joke is a complex cognitive activity; it involves, for example, calling up stereotypical knowledge of different nationalities and inferring how a person's response to a question can exemplify that stereotype. But it is much more. To get the joke is to find it funny; and though I'm not sure exactly what that amounts to, it is at least having a pleasant emotion and actually laughing. Notice that the emotion and laughter come from making the inferences: getting the joke means getting pleasure when you correctly make the connection. Notice also that the emotion and the laughter are bodily activities. Emotions manifest themselves in physiological changes throughout the body and laughing is something that the body does. A bodiless computer that could classify jokes as funny even though it never felt the pleasure of humor or laughed-even though, in other words, it never found them funny-would not have a sense of humor, any more than someone who can tell the quality of wines by looking at the price is a wine connoisseur. There are other, more subtle ways in which jokes are bodily affairs. When a joke is told, timing is crucial to whether it is funny or not. Why? because joke-telling must engage both cognitive and bodily reactions at the right pace. Thus telling a joke well means timing the physical delivery, and getting a joke means gauging the timing and reacting appropriately. Someone who sincerely laughs at a badly-delivered joke has a poor sense of humor, as does someone who laughs either too soon or too late. So getting a joke is essentially cognitive, emotional and bodily. But it also depends crucially on the social context. First, telling and getting jokes is essentially interactive. Without somebody to listen, I can't tell jokes, though I can utter the same words. Second, the social context determines what is a joke and what is not. Someone who tells, as jokes, stories that nobody laughs at is not a wit but a bore. People who laugh at things that nobody else finds funny are candidates for the lunatic asylum. And third, what people do find funny depends heavily on social niceties. The same words can be a joke in one context, a faux pas in another. Some subjects, such as national eccentricities, are generally humorous, while others, such as having AIDS, are generally not. Robinson Crusoe cannot have a sense of humor, not only because there's nobody to joke around with, but because, without social conventions of humor, there is nothing to determine what is funny and what is not. He may laugh in response to some things rather than others, but that is just a behavioral quirk. Joking is, I think, quintessentially mental. Yet joking is essentially cognitive, emotional, bodily, skillful, interactive and social. It is both inner and outer, both individualistically and holistically constituted. The various different aspects fit into a complex integrated structure. Now, the ontology of mind must surely be at least as complex as that of joking. An adequate ontology of mind must therefore be neither Cartesian, nor Heideggerian, nor Rylean. Perhaps the best way to sketch such an ontology is to compare it with the Cartesian one along the seven key dimensions. First and foremost, mind on this conception is not ontologically homogenous. A mind is like a national economy; it is a complex, structured totality incorporating entities of many different kinds. Some of those entities are located inside the head and perhaps contribute causally to our behavior. Some are a matter of the behavior itself, which is primarily external. Both categories have numerous important subdivisions. Some elements of mind, such as humor, are themselves complex. There are as many different kinds of answers to the question what is a mental entity? as there are different kinds of entities that make up minds. Since the mind is a structured totality encompassing both inner and outer components, the Cartesian claim that mind is an inner realm must be rejected. To think of the mental as inside the mind or the skull is as much a mistake as thinking of an economy as located inside the walls of factories. To reject a conception of mind as internal is tantamount to rejecting an purely individualist orientation, both metaphysically and as the basis of intelligibility. Mind is not constituted by facts about the individual alone; what it is to have a mind is partly constituted by our place in the world. We cannot fully understand what mind is simply by understanding what is going on inside the head, anymore than we can understand what an economy is by understanding what goes on inside factories. To properly understand mind we must be both individualists and holists, and this means rejecting the Cartesian homogeneity of mind in another sense. If a belief, for example, is basically a disposition to act, two quite different accounts of that disposition can be given, corresponding to the two different accounts of the chess king. We can see it constituted as pure behavior of an individual body with internal causes; or we can understand it as a belief that p by seeing how that behavior fits into a larger totality. For most particular mental entities, there is not one but two equally important accounts of its constitution: one individualist and one holist. One is an account of what makes it up, the other is an account of what makes it what it is. Mind is heterogeneous in the sense that there are at least two importantly different stories to be told about the nature of mental entities. It follows from these considerations that Descartes was right that mind is ontologically completely different than body or the physical world generally. The physical world is complex, but not ontologically heterogeneous; therefore mind and material world belong to different ontological categories. Mind is as different from body as economies are from factories. To say this does not leave us with problems of interaction between ontologically different kinds. Mind does not causally interact with the body, any more than an economy causally interacts with factories, the US Constitution causally interacts with the American people, or a sense of humor causally interacts with laughter. Some components of mind do causally with the physical world, but these components are, in fact, physically constituted. Though ontologically different than body, mind is not independent of it nor is it disembodied. Just as there could be no economy with people, factories, money, etc., there could be no mind without a body to realize it. Minds cannot float free of the physical world; they are one way in which that world is organized. Descartes was mistaken in running together the ontological difference between mind and body with independence and disembodiment. Since mind is not simply inner states and processes, it cannot be inner representational states and processes. There may well be representational states inside my head, and certainly many of the activities in virtue of which I have a mind, such as using language, are representational. Nevertheless, having a mind is not fundamentally a matter of representing. Slapstick humor is thoroughly mental, and while there may be internal representations underlying the ability of the comic to be funny and the audience to see the humor, the emotions, skills, interactions and social conventions that jointly constitute that episode as humor are not representational, nor is the totality which is the humor itself. To think of the mind as essentially representational is to commit an egregious part-whole category mistake. Finally, it follows from all the above that Descartes was wrong in thinking of mind as the inner causal engine that produces sophisticated behavior. If that inner causal engine is cognition, then mind is not cognition. However, we must be careful not to say too much here. Mind is not something other than cognition; rather, cognition is an essential component of mind, just as manufacturing is an essential component of an economy. Rather than thinking of mind as the inner engine of behavior, we should think of cognition as the inner engine of mind. Thus, the cognitive sciences are not, strictly speaking, the sciences of the mind. They are the sciences of just one aspect of mind, just as the study of manufacturing is the study of just one aspect of an economy. Of course, there is nothing wrong in saying, informally, that the cognitive sciences are the sciences of the mind, since to study the part is, in a limited way, to study the whole. The key mistake would be to suppose that the cognitive sciences will give us a full understanding of mind, and in that sense tell us what mind is. It follows that the cognitive sciences pose no general threat of eliminativism. They may of course radically revise our conception of cognition. I think they are in fact doing this, and, like Churchland, I find this a dramatic and appealing development. But to eliminate mind as such, the cognitive sciences would have radically alter not just our conception of what cognition is, but also the social totality that shapes and constitutes our external and internal activities as what they are, and there is no foreseeable possibility of their doing that. Mind is generally safe from the sciences of cognition because mind is only in part a matter of our inner lives. Or, put another way: you can have your mind and e.e.g. it too. References Churchland, P. (1979) Scientific Realism and the Plasticity of Mind. Cambridge: Cambridge University Press. Churchland, P. M. (1988) A Neurocomputational Perspective. Cambridge MA: MIT Press. Churchland, P. S., & Sejnowski, T. J. (1992) The Computational Brain. Cambridge MA: Bradford/MIT Press. Fodor, J. A. (1975) The Language of Thought. Cambridge MA: Harvard University Press. Guignon, C. B. (1983) Heidegger and the Problem of Knowledge. Indianapolis: Hackett. Haugeland, J. (1983) Heidegger on Being a Person. Nous. Haugeland, J. (1993) Mind Embodied and Embedded. Paper presented at Mind and Cognition: An International Symposium, Academia Sinica, Taipei Taiwan, May 27-30 1993. Johnson, M. (1987) The Body in the Mind: The Bodily Basis of Meaning, Imagination, and Reason. Chicago: University of Chicago Press. La Mettrie, J. O. de (1748/1912) Man a Machine. Chicago: Open Court. Ryle, G. (1949/1984) The Concept of Mind. Chicago: University of Chicago Press. - - |
|
This page, its contents and style, are the responsibility of the author and do
not represent the views, policies or opinions of The University of Melbourne.
|