From the philosophical archive for the constructive study of ontological dualism:



Review by Steven Nadler of a book by Gordon Baker and Katherine Morris, and a reply

quoted from: Philosophical Books, 38, 3 (July 1997) pp. 157-169.


All references to Descartes's works are to the Oeuvres de Descartes (12 vols.), edited by Charles Adam and Paul Tannery (J. Vrin, 1974—1983), referred to as 'AT'. The translations are from The Philosophical Writings of Descartes (3 vols.), translated by John Gottingham, Robert Stoothoff, and Dugald Murdoch (Cambridge University Press, 1985), referred to as 'GSM'. 

There is a certain story about the history of early modern philosophy that we have all heard—and maybe even told—at one time or another. According to the traditional textbook narrative, Descartes introduced a dualistic ontological system, a universe of mental substances and corporeal substances, that was fraught with metaphysical inconsistences, epistemological conundrums, and patent philosophical falsehoods. Spinoza's monism, Malebranche's occasionalism, and Leibniz's preestablished harmony are all, we are told, attempts to deal with the problems facing 'Cartesian dualism'. The tale, moved along by Locke, has its stunning climax in the idealism of Berkeley and its sober, logical conclusion in Hume's sceptical minimalism. This story, premised on a specific and common understanding of Descartes's dualism, is a fine example of what a recent Descartes scholar has labelled "shadow history of philosophy".1 Shadow history is "a kind of received view consisting of stories of philosophy that most philosophers accept even though they know that these stories are not really quite precisely right". Such stories usually serve a rhetorical or pedagogical purpose, and they are especially handy when a philosopher wishes to set her own views off from some 'tradition' and show where other thinkers (conveniently dead and, thus, unable to correct the depiction of their views) have gone wrong. Rather than being the misuse of the history of philosophy, it involves "the use of misinterpretations of the history of philosophy".2

1.   Richard A. Watson, 'Shadow History in Philosophy', journal of the History of Philosophy 31 (1993), 95-109.

2.  Watson, 'Shadow History in Philosophy', p. 97.

Gordon Baker and Katherine J. Morris (Descartes' Dualism, Routledge, 1996. xiv + 235) are more interested in legends than in shadows, but the idea is basically the same. Their target is certain "preconceived prejudices" of twentieth-century Anglo-American philosophers; in particular, prejudices about how to conceive of the mind and the body (and their relationship), about how Descartes conceived them, and about the value of how he conceived them. "Our primary goal," they state, "is to arrive at a clearer and deeper understanding of what was Descartes' conception of a person" and to show that, whether or not such a conception is also a "live option" for a late twentieth-century philosopher, it nonetheless should be of current philosophical interest (pp. 3-4). This is legend-bashing, and for contemporary epistemology and metaphysics there is no historical figure more legendary than Descartes.

The Legend that Baker and Morris have in mind is the one according to which Descartes was a Cartesian Dualist. Descartes was no such thing, they insist; nor could he have been. And the belief that he was results from projecting our contemporary idiom onto a seventeenth-century figure.

There are, they explain, four central tenets to 'Cartesian'—as opposed to Descartes's—dualism. First, there is what the authors call the "Two Worlds View": there is a private, inner world of mental objects that parallels the public, outer world of physical things. The inner world is a world of 'ideas': the outer world, a world of bodies. The objects in each world have, by kind, distinctive properties and stand in various relations to each other. Second, the mind is identified with consciousness. Anything that we would now call a 'state of consciousness' or subjective experience—concepts, beliefs, sense perceptions, bodily appetites, pains, pleasures, emotions, etc.—qualifies as a cogitatio and is placed by Descartes in the mind. This is what Baker and Morris call "The Expansion Thesis": Descartes famously (or infamously) extended the range of the mental, of 'thought', to include many things that earlier philosophers attributed to the body. This thesis is necessarily accompanied by the "Contraction Thesis", the claim that the body is thus nothing but an unconscious, insentient machine, "a complicated bit of clockwork". Any being not endowed with a human soul—and this includes all non-human animals—therefore lacks consciousness, even sentience. Third, all of the inner, mental objects—and it is now a rather large domain—are "apprehended by the subject alone through the exercise of a quasi-perceptual faculty called 'introspection", a faculty whose reports are indubitable. This faculty for apprehending inner objects is paralleled by the fallible five senses, which apprehend external objects, including the subject's own body. And fourth, mind and body, while distinct substances, causally—but inexplicably— interact. In sense perception, thoughts are caused by states of the body (as motions in the pineal gland cause sensory images), and in voluntary action mental causes such as volitions produce bodily motions.

The standard philosophical objections to Cartesian dualism are well-known, and Baker and Morris do not waste any time rehearsing them once again. Indeed, their point is that these objections do not really touch Descartes himself, for none of the four tenets of Cartesian dualism represents Descartes's views on the mind, the body or their relations. Their verdict on the attribution of Cartesian dualism to Descartes, "not proved", is accompanied by arguments—sometimes effective, sometimes sketchy—purporting to show that there is no textual foundation for believing that any of those theses is to be found in Descartes or implied by what he said. For example, they demonstrate how Descartes's use of the term sentire (to sense or feel) is ambiguous: sometimes it refers to a bodily process involving sense organs ("sensing1"); at other times it refers to our cognitive apprehension (or "perception" or judgement) that we are undergoing such a bodily process("sensing2"). There is thus a restricted sense of "having a sensory perception" in which it refers to a mental activity, but it does not follow from this that sense perception in its "proper" or "strict" meaning refers to some sheared-off mental object called a "sense datum" (pp. 30-35). This is one of the more crucial points in the authors' case against the Legend's picture of Cartesian consciousness, and it plays a large role in the book's central argument.

The real interest of the work, however, does not lie in the negative side of the authors' project, much of which consists (wisely) not in refuting the traditional account but in showing that undeniably hard-to-interpret passages do not unequivocally support the Legend. Rather, what is most striking and original here is the picture of Descartes's metaphysics that emerges from their argument, particularly when they attempt to reconstruct on the basis of a close reading of his texts what they take to be the real content of his dualism. It is a Descartes that will be very unfamiliar to most readers, especially those who do not keep up with the Cartesian literature.1 As the authors note, their presentation of Descartes's dualism is not just a revision of some of the basic ideas of Cartesian dualism. It is, in fact, a wholesale rejection of the Legend, and its theses are incompatible with many of the notions composing the traditional view.

1.  I should note that the authors do not really take into account any of the revisionist literature that has appeared in the last decade or so, particularly from French scholars. The authors claim at the outset that they are only concerned with the Anglo-American literature on Descartes, but as far as I can tell this is only with respect to the genesis of the Legend they want to attack. It is a mystery why they ignore all of the French literature which might help them in their project, since a lot of it, so different from the Anglo-American work, is sympathetic to their goals.

First, they insist, Descartes held not that there are two (parallel) worlds, the inner and the outer. Rather, he held that there are two (finite) substances, each with its own modifications. The Legend treats the thoughts of the inner world as entities in their own right, as "the logical subjects of singular judgments just as the material objects of the sensible world are" (p. 67), that is, as substances. But only minds or rational souls are immaterially substantial, and only they can be the subjects of singular judgments whose content is mental. Thinking or having a thought is not an object or thing in a mental world, but a mode of a substance. It is, in fact, an activity or operation.2

2.  Here, too, the authors curiously ignore all the recent (Anglo-American!) literature devoted to showing that Cartesian ideas are not objects but operations of an active substance. See, for example, John Yolton's Perceptual Acquaintance (Blackwell, 1984); Nicholas Jolley's The Light of the Soul (Oxford University Press, 1990); and my own Arnauld and the Cartesian Philosophy of Ideas (Princeton University Press,  1989), not to mention articles by Monte Cook, Ronald Arbini, Michael Costa, and others.

Second, the Legend misconstrues the central opposition within Descartes's dualism by setting up a contrast between consciousness and clockwork. Descartes's conception of mental substance or mind is not extensionally equivalent to "consciousness" (although it is extensionally equivalent to conscientia, with its moral dimension). The true dichotomy, according to Baker and Morris, is between rationality and sentience, or the moral/intellectual and the animal. The mind, they argue, is thought; it is intellectus, the rational soul. The activities of the mind, therefore, are all modalities of rational thinking: judgements, in effect, and thus propositional. Any non-cognitive event is a non-mental event. Having a sense-perception (in the restricted, non-bodily sense), therefore, is not to have some qualium or sense-datum hovering before the introspective soul. It just is to have a thought with a particular content, and the content describes a possible state of the body ('my eyes are being stimulated by light', for instance). To feel pain (again, in the restricted sense) is to believe or think or judge that one's body is in a certain condition.

The flip side of all this is that sensing or feeling in the non-restricted sense (sensing1) is an activity of the body. Non-human animate bodies (and, in theory, even the human body without the soul) are sentient, conscious bodies. While they may not be capable of thinking (since they lack soul), they are capable of feeling and consciousness, in sum, of all those processes which do not require rationality. Brutes do not have conscientia (the self-knowledge that rational beings have of their actions) and thus they are not moral agents, but "they do share with human beings many of the things now called 'states of consciousness' " (p. 102). For Descartes, animals do feel pain (in the non-restricted, bodily sense), and they are capable of emotional lives (also in the non-restricted sense). Of course, all of this must be explained in purely mechanistic terms—that is, in terms of matter (extension) and motion alone. But, as Baker and Morris point out, that just is Descartes's project, to extend the scope of mechanistic explanation to cover as many phenomena in the natural world as possible, especially those phenomena that Aristotelian philosophers were able to explain only by inventing vegetative and sentient souls (p. 84). It is clear that what Baker and Morris are aiming for is nothing short of a complete reversal of the Legend: the Contraction Thesis is now a feature of Descartes's conception of the mind, whose activities consist only in cognitive and moral rationality (including volitional activity). The Expansion Thesis, on the other hand, now characterises Descartes's approach to the physical world—a world emptied of the substantial forms of the Scholastics, but whose mechanical processes can account for everything not requiring intellect and will.

Baker and Morris present us with a fascinatingly radical interpretation of Descartes's conception of the mind and of the nature of conscientia. It is also a very compelling one, and the breadth and cogency of their reading is impressive. Many will, no doubt, also find it implausible. The picture of the mind proposed for Descartes is a rather austere one. All those wonderful buzzes, hums, feels, colours, sounds, etc., of the traditional Cartesian mind are, in so far as they are mental (that is, in the "restricted" sense), reduced to propositional attitudes. "The sensation of hunger or pain is the thought or judgment that one feels1 hunger or pain [i.e., that one is in a certain bodily condition]" (p. 129). Descartes's mind may begin to look like the hyper-rational Vulcan mind of Mr. Spock. And Baker and Morris do not help their case by casually dismissing the request to explain what happened to the 'what it's like', to the subjective 'feel' of phenomenological qualities. "How are you to explain to Descartes what exactly it is that this picture leaves out?", they ask (p. 100). This is an important request, one worth answering in greater detail, even though it may be generated only by an overly-long attachment to the legendary picture.

The fourth tenet of the Legend of Descartes concerns the union and interaction of the two substances in a human being. This is also the focus for what has been, historically, the most sustained objection to Cartesian dualism, namely, that given the radical ontological divide at the heart of his anthropology, Descartes cannot consistently claim that mind and body causally interact. But according to Baker and Morris, in what is perhaps the most interesting part of the book, Descartes made no such claim, or at least not in the sense in which it is usually understood. The mind-body relationship is not one of efficient causal interaction. Bodily events are not the real efficient causes of mental events (such as sensations) and mental events (such as volitions) are not the real efficient causes of bodily motions (although events of one ontological kind are efficiently causally related to events of the same kind). What their mutual relationship does consist in is what the authors call "occasionalist interaction" and what I have elsewhere called "occasional causation"1. Motions in the brain "occasion" the soul to have (i.e., to efficiently cause or generate in itself) certain perceptions, while the mind's volitional activities are the occasion for certain bodily movements. What such non-Malebranchian occasionalism, or 'Natural causation', consists in is really just a nomological correlation between mental events (all of which are, for the authors, articulate judgements) and bodily events, a correlation grounded in (in fact, constituting) our Nature but ultimately established by God for our welfare.

1.   See my 'Descartes and Occasional Causation', British Journal of the History of Philosophy vol. ii (1994), pp. 35—54. This is another instance where the authors might have done a little more reading in the Cartesian literature.

Baker and Morris make a persuasive case here for a consistent and attractive reading. There is no doubt that there is something wrong with the Legend's account of Cartesian 'interactionism'. Nonetheless, their own account raises some difficult questions. First, with respect to body>mind relations, is it possible that Descartes's view is really more ambiguous or even unresolved than Baker and Morris allow? At times, Descartes really does seem to think that the body is the efficient cause of sensory ideas. His language in the Meditations suggests as much, where (in Meditation VI) he not only speaks of an active faculty (facultas activa) in bodies for causing such ideas (although Baker and Morris do a good job of explaining how to understand such 'faculty' talk on their reading) but also claims that sensory ideas are "transmitted [emitterentur]" from corporeal things.2 This is still vague language, but it certainly sounds like transient efficient causation.

2.  AT VII, 79-80; CSM II, 55.

Their case for understanding mind>body relations as a species of occasional causation is weaker than their case with respect to body>mind relations. In fact, there are good reasons to believe that Descartes did consistently hold that minds can and do efficiently cause bodily movements. Remember that for Descartes the soul is an active substance, while the body is inert, passive extension. Thus, there is a causal asymmetry here to begin with (one which the authors ignore, as they treat 'powers' or 'activity' in identical terms for both the mind and the body (pp. 174-5)). The body cannot actively generate anything in the soul, and it certainly cannot communicate motions to it, since motion cannot be a mode of a thinking substance. Yet some agent is needed to serve as the "proximate and primary cause [causa proximo, et primaria]" or "proximate and efficient cause [causa proximo {et} effectrix]" bringing about the mental events on the occasion of the bodily motions (the "remote and merely accidental cause [causa remota et accidentaria duntaxat]".)1 That is just how occasional causation works for Descartes: a is the occasional cause of x if and only if a occasions/is the occasion for b to efficiently cause x.2In this case, it is the soul itself that efficiently causes the thoughts to occur.3 On the other hand, if the soul were only the occasional or remote cause of bodily motions, what would be the requisite proximate cause that moves the body on the occasion of the volitional activity? A body cannot put itself in motion. I presume that Baker and Morris would answer that it is simply God's will (as instantiated in our Nature) that has decreed that the body should move in a certain way when a specific mental activity occurs. But I am hesitant to believe that Descartes would agree that God's will is the proximate cause of bodily movements in human voluntary action. Moreover, Baker and Morris basically admit that Descartes never uses the occasioning idiom in the context of discussing human voluntary action, whereas he frequently uses such talk in the body>mind context. This strikes me as evidence that Descartes really did believe the soul to be the efficient cause of bodily motions.4

1.   The terminology comes from Descartes's discussion of the genesis of ideas in the mind on the occasion of bodily motions in the Notae in Programma, AT VIII-B, 360; GSM I, 305.

2.   For an analysis of occasional causation, see 'Descartes and Occasional Causation', pp. 36-44.

3.  Notae in Programma, AT VIII-B, 358-9; GSM I, 304.

4.   This point is argued for at greater length by Daniel Garber, 'Descartes and Occasionalism', in Steven Nadler (ed.), Causation in Early Modern Philosophy (Penn State Press, 1993), pp. 9-26 (see especially pp. 15—19).

Why, then, cannot the causally active soul move the body as an efficient cause? According to Baker and Morris, it is because the correlations between mental states and bodily states fail to satisfy the principle that an efficient causal relation must be an intelligible one. "Causal explanations are sought in order to answer questions of the form 'Why did this happen?'; . . . ascertaining its cause makes a happening intelligible" (p. 145). But the only kind of intelligible relations recognised by Descartes, they argue, occur between things that resemble each other: either logical relations among thoughts or mechanical relations (displacement by impact) among bodies. "Anything which is definitely not a mechanical relation among corporeal things or a logical relation among thoughts cannot be explained; it must be opaque to reason" (p. 147). Since there can be no mechanical relations between bodily motions and thoughts, it follows (according to Baker and Morris) that "correlations between thoughts and states of the body must be unintelligible" (p. 147) and, hence, non-causal. But there seems to be something missing from the authors' account of intelligibility. Descartes believed that intelligibility in the physical realm comes only with laws: he explicitly calls the laws of physics "causes" of the behaviour of bodies.1 What makes a mechanical relation between two bodies intelligible is not just the matter and motion per se, but the laws of physics as well. Without the laws, we could not answer the question 'Why did this happen?'; we could not see why what happens when one body strikes another must happen; perhaps we could not even see that something must happen. And without the laws of logic, we could not see why one thought must follow another thought. No laws, no necessity; no necessity, no intelligibility (as the authors would agree). Those laws might be God-ordained, as Descartes believed, and thus ultimately contingent (as the authors would deny), but they make possible intelligibility in mechanical and logical relations. Now consider the mind-body case. The authors rightly note that, for Descartes, God has decreed that there be certain correlations between mental activities and bodily motions. But isn't this just to say that God has established certain psycho-physical laws, constituting our Nature? And thus isn't it the case that those psycho-physical laws lend a kind of necessity, hence intelligibility to mind-body correlations, just as the laws of physics confer necessity and intelligibility upon the correlations in the motions of bodies?

1.   See Principles of Philosophy, 11.37. For Descartes, the laws of nature are prescriptive for the behaviour of bodies.

It may be relevant here for Baker and Morris's case that the laws of physical nature—at least the most general laws of motion and impact—are for Descartes knowable a priori,2 whereas only experience reveals the laws of mind-body union. But for the teleologically-driven God of Baker and Morris's Descartes, intelligibility is nonetheless a remarkable and undeniable feature of mind-body relations! They go to great lengths to stress the "necessity" inherent in the mind-body correlations constituting our Nature. When establishing the specific correlations between mental states and bodily states, God is motivated, above all, by a concern for our well-being.

2.   Principles of Philosophy, 11.36-52.

The order instituted by God is, in a certain sense, essential or necessary. There are powerful a priori reasons for concluding that our Nature could not have been different from what it actually is. No other arrangement of things . . . could be so conducive to human welfare (and equally simple, uniform, and efficient) . . . The actual arrangement demonstrable [my emphasis] maximizes the benefit and minimizes the risk of error . . . This whole pattern of correlating thoughts with movements (and the derivative natural correlations among thoughts and actions) is necessary in the same sense in which it is necessary that God must have arranged His Creation to maximize human welfare by the simplest and most uniform means (pp. 183-4).

Thus, there are good reasons, relative to human well-being (indeed, to our preservation), why specific states of the body are followed by specific sensations in the mind (pain, for instance) and not others; and why specific mental actions are accompanied by specific bodily motions (removing the foot from the fire, for example) and not others. Now if the particular correlations between mental states and bodily states are demonstrable the most conducive to a known and specifiable end, then I cannot imagine that, even by Baker and Morris's own criteria, anything more could be demanded for their intelligibility: we can see why certain mental states should be followed by specific bodily states, and why certain bodily states should be followed by their corresponding mental states. What more could be required for an intelligible necessity?

Let me be clear about what I am claiming here. I am only suggesting that, given Baker and Morris's reading of Descartes's conception of mind-body union (including human nature and God's role in instituting that nature1), neither body>mind nor mind>body relations lack the minimal explanatory intelligibility that Baker and Morris deem necessary for efficient causality. I am not suggesting that it therefore follows that mind-body relations qualify, for Descartes, as a matter of reciprocal efficient causation. There are independent reasons why Cartesian bodies cannot actively cause mental events, and intelligibility alone is not sufficient for a relationship of efficient causation. Occasional causal relations can still be intelligible, and on Baker and Morris's teleological picture we can see why bodily motions should occasion the soul to cause some sensation rather than another. Perhaps what is further required for efficient causation is activity (in a more robust sense than Baker and Morris are willing to allow), and for Descartes only the soul is active.

1. Many will find the book's characterisation of Descartes's God—acting for the best—too much like Leibniz's God, and the authors explicitly draw the two philosophers rather close in their conceptions of the divine modus operandi (pp. 183—5). On the other hand, and to their credit, Baker and Morris succeed in bringing out the teleology latent in Descartes's metaphysics and natural philosophy that other commentators cautiously shy away from.

It is not often that one can finish yet another book on Descartes and sincerely think that it has added something new to the debates. But, as my discussion should make clear, this is a stimulating work that deserves to be read and discussed not only by Descartes scholars, but by anyone interested in the philosophy of mind and its history (and legends).





Steven Nadler's discussion gives an informative, lucid and balanced account of our book, as well as raising some challenging objections (effectively two complaints and one sustained criticism). We hope that we can respond to these with as much fairness and good grace as he has shown here.

There is one important lacuna in his presentation of our position. He says little about one of our two explicit aims in writing this book: he focuses almost exclusively on our 'first-order' aim of presenting a comprehensive alternative to the standard Anglo-American way of understanding Descartes's dualism (what we labelled 'the Cartesian Legend'), at the expense of our 'second-order' aim of showing why such historical explorations are of philosophical interest. Attention to this philosophical aim might help to alleviate his two complaints and to go part of the way toward meeting the more sustained objection.

(a) One complaint is that we neglect much recent 'revisionist' literature, as well as potentially sympathetic French commentators. This is a fair comment: but such 'neglect' was in a certain sense a matter of policy, deriving in large part from our second-order aim.

First, our primary expository task was to build up a comprehensive and well-integrated account of Descartes's dualism on the basis of close consideration of his texts. We hoped to make our argument sufficiently compelling to command assent and thus to shake the grip of the Legend which is entrenched in Anglo-American thinking about Descartes. Whatever the merits of recent revisionist literature (and they are considerable), there seems to have been as yet little serious progress towards effecting such a Gestalt switch.1 (Nadler effectively acknowledges this point.) It is not clear that noting detailed points of agreement with other commentators would have strengthened the persuasive power of our interpretation, at least for the non-specialist reader (although it might make the revisionists themselves better disposed to our ideas!).

1. 'Revisionism' tends to be piecemeal. This strategy seems unlikely to lead anyone to the realisation that the Legend as a whole is radically flawed, that a new picture of Descartes's thinking must be constructed ex nihilo.

Secondly, our second-order aim in reinterpreting Descartes was to make a contribution to current thinking in philosophy, specifically to the stream of late twentieth-century Anglo-American philosophy of mind in which we are participants. We were not aiming to propose any novel solutions to the problems that now dominate English-language philosophy of mind, but rather to gain some intellectual distance from these problems via a confrontation of its world-view with a radically divergent one. Constructing a new interpretation of Descartes's thinking and showing how to give it support from his texts is an absorbing and worthwhile enterprise. But why should it receive attention from anybody now engaged in philosophy of mind, cognitive science, etc.? We argued that it ought to be of interest to them: coming to a sound appreciation of Descartes's conception of the correlative concepts of 'mind' and 'body' might be a valuable source of self-awareness, bringing to light today's 'prejudices' by getting clearer about his. In contemplating his conception of the rational soul and 'modes of thinking', one might suddenly find it ridiculous to call the mind an entity postulated to explain the complexity of human behaviour; or remarkable that Descartes sketched a picture of human life that has no room for 'phenomenological properties', 'qualia', or the 'what's it like'; or bizarre to call pain and hunger 'mental states'; or noteworthy that he connected the soul with moral agency, treating virtues and vices as modes of thinking. Such transformations could amount to Gestalt switches, so that 'the Mind-Body Problem' might need to be radically reformulated. The revisionist English-language literature on Descartes, excellent though much of it is, does not provide the coherent world-view needed as a point of departure for such a head-on confrontation, while Francophone literature belongs to a quite different tradition of philosophical 'anthropology' which manifests a different framework of thinking. A restricted range of secondary references seemed better to serve our philosophical purposes than a more comprehensive critical apparatus.

(b) These points are also of relevance to Nadler's second complaint, viz. that we, having argued that Descartes's picture of the mind leaves no room for the 'what's it like', "casually dismiss" the request "to explain what happened" to it.

The very idea of the 'what's it like' is supported by a whole tightly-woven and well-entrenched substructure of preconceptions about what counts as mental, what counts as bodily, and how these are to be characterised. Arguing persuasively against the necessity of its inclusion in a philosophical description of human beings requires demonstrating, compellingly and in detail, the possibility of an alternative substructure of ways of looking at these things. We think that Descartes held just such an alternative view, and we attempted in our lengthy fourth chapter to describe it (in detail and, we fondly hope, compellingly). Far from "casually dismissing the request to explain what happened to the 'what's it like' ", it might be said that a large portion of the book was devoted to just this task!

Nadler complains, on behalf of advocates of the 'what's it like', that the picture of the mind we ascribe to Descartes is "implausible" and "austere": it "may begin to look like the hyper-rational Vulcan mind of Mr. Spock".1 Plausibility is in the eye of the beholder (that indeed could be said to be the burden of our second-order aim); but Nadler ought in fairness to stress that the picture of the body is far richer, far less austere, than the 'what's-it-likers' commonly allow. He notes this aspect of our interpretation of Descartes's dualism—the logical correlate of re-contracting the mind to exclude everything bar intellect and volition is re-expanding the body to include sensation and imagination—but he does not bring it to bear at this stage.

1.   Somewhat in the spirit of the developing cottage industry of crying 'Zombie!' at critics of the 'what's it like'?

Moreover his description of the 'sentient body' omits reference to a Cartesian doctrine upon which we laid much stress, viz. the doctrine of the internal senses. Both internal senses have conditions of the body as their objects, and these forms of sentience are shared by humans with many animals. The first internal sense apprehends pain and pleasure, hunger, thirst, and other bodily appetites or needs; the second the emotions such as fear, anger, joy, and wonder.2 Both are essentially concerned with the preservation of animal welfare. Neglect of this doctrine, we argued, was an important part of the explanation of the Legend's ascription to Descartes of the view that pain, hunger, fear, anger, etc. counted as mental properties ('cogitationes').1 By the same token, such neglect partially underlay the Legend's supposition that Descartes's Bete-Machine doctrine implied a denial of sentience to animals. Correlatively (since Descartes held that the functions of the human body were just those which we shared with non-human animals), attention to the doctrine of the internal senses makes the human body a far richer and more interesting object of philosophical study than the Legend supposes: one might almost say that the "wonderful buzzes, hums and feels" whose absence from the mind Nadler regrets belong, not to the mind, but to the body.

2.  Descartes's account of the emotions has complexities which we could not enter into in our book; but it is absolutely clear that his account of the emotions cannot be understood without acknowledging his doctrine of the internal senses.
1. We suggested that the muddle arises, at least in part, from failure to note his distinction between 'pain' (a condition of the body) and 'the sensation of pain' (an articulate sense-based thought as expressed in the assertion 'I feel a pain in my foot'). (Alternatively, the expression 'sensation of pain' is, Descartes held, ambiguous between 'feeling pain (with an internal sense)' and 'thinking one feels pain'.) The sensation of pain, but not pain, is a paradigmatic thought.

(c) Nadler's most sustained set of criticisms focusses on our account of mind-body interaction; we sketch here the broad outlines of a response.

First, he argues that the two cases of body>mind and mind>body interaction must be essentially different, since "the soul is an active substance, while the body is inert, passive extension". This is the ground on which Nadler builds his case that Descartes took the soul to be the efficient cause of bodily movements in human voluntary actions. The axiom that 'the mind alone is active; the body is inert' is a dogma of Descartes scholarship that deserves to be challenged, and we did so. It endorses just that conception of the body (as clockwork) which we argue Descartes did not accept. On the contrary, he saw the human body (and animals) as sentient, i.e. endowed with powers to respond to internal and external conditions of matter (via the internal as well as the external senses), with capacities to anticipate things and to pursue goals ('imagination'), and with the power to initiate and sustain movement (locomotion). Descartes would vigorously dissent from the suggestion that a cat stalking a mouse should be described as "inert and passive"!

Secondly, Nadler's account of our reasons for denying that the soul can be the efficient cause of movements of the body rests on misunderstandings of our position and, in our view, of Descartes's notions of intelligibility and resemblance:

(i) In arguing that "there seems to be something missing from [our] account of intelligibility", Nadler neglects our having distinguished between different kinds (or degrees) of intelligibility. (Hence, by implication, different senses of 'seeing why' something happens.) We drew attention to "the double standard of intelligibility that is essential to occasionalism" (p. 146). We exploited this equivocation in arguing that God's role in establishing "our Nature" makes the correlations of thoughts with bodily movements "intelligibly unintelligible". Against this background, we argued that God-ordained (hence necessary) correlations of thoughts and movements are instances of occasional causation as opposed to efficient causation; these different forms of causation display different kinds of intelligibility. Hence Nadler mistakes our position in arguing that "the minimal explanatory intelligibility that Baker and Morris deem necessary for efficient causality" is to be found in Descartes's argument that 'our Nature' is instituted by God for human welfare.1

1.  By the same token, he misinterprets our use of 'demonstrable' in the claim that our Nature 'demonstrably' promotes human welfare.

(ii) Nadler assigns to Descartes the principle 'No laws, no intelligibility'. (By implication he also assigns the converse principle: regularity implies intelligibility (though not efficient causation).) It should be clear that we disagree. Descartes did not hold that the kind of intelligibility manifested by impact (and logical exclusion) is parasitic on knowledge of what God has ordained. In favourable cases at least, intelligibility of such interactions may be immediate, non-inferential, and evident in single instances (like pushes and pulls). This primitive form of intelligibility is the foundation for rational physics, not solely the product of discovering natural laws. We laid out a case, which Nadler leaves unchallenged, that on Descartes's view regularity is not required for intelligibility. Nor is regularity sufficient for intelligibility. The uniformity of basic mind-body correlations is taken for granted by Descartes. But it calls for explanation, being wholly opaque to reason (not even 'intelligibly unintelligible') unless related to God. (Furthermore, the kind of intelligibility found in physical interactions is independent of consideration of God's benevolence—unlike the intelligibility of otherwise unintelligible mind-body correlations. In this respect, these two forms of intelligibility differ in kind.)

(iii) We do not take lack of (apparent) similarity to be the reason for denying the possibility of causal interaction of mind and body; rather, the two claims are equivalent.2 That there is "no resemblance" between a movement of the blood around the heart and the (soul's) passion of love means (in Descartes-Speak) that this correlation is (absolutely) opaque to human reason; in contrast to the resemblance between cause and effect when one billiard ball displaces another by impact—a case of efficient causation that is transparent to reason in a single instance.3

2.  And the similarity characteristic of efficient causation is apparent not to the senses (as Nadler suggests) but to reason. A cause must resemble its effect in form.
3.  Mechanics exhibits real resemblances even where they are not apparent, e.g. in relating the rubbing together (friction) of dry sticks to the production of fire.

Thirdly, while he and we are in agreement that Descartes's conception of mind-body interaction must be understood against the background of scholastic doctrines of causation, we think that Nadler misrepresents this background and, as a consequence, misconstrues the text central to his argument.4 The issues here are monstrously complex5 and a full discussion of them must await another occasion; but we note here that he takes the terms of the relation 'x causes y' to be events (for example in explaining the 'influx model' of efficient causation and the principle that cause and effect must resemble each other),1 whereas even texts that he cites2 show that Descartes faithfully followed the Aristotelian tradition of focussing on substance-causation. Different understandings (or misunderstandings) of this background are bound to result in different understandings (or misunderstandings) of Descartes's view of mind-body interaction.

4.  AT VIIIB, 360 is his crucial exhibit. He takes an Aristotelian paradigm of an efficient cause (a person who counsels or commands another to produce something: Phys.  194b30) to be offered by Descartes as a paradigm of a cause (an 'occasional cause') to be contrasted with an efficient cause.
5.  E.g., the Port-Royal Logic lists 23  different scholastic distinctions within the category of efficient causes!

1.   See Steven Nadler, pp. 36, 38, 50 of 'Descartes and Occasional Causation', British Journal for the History of Philosophy, vol. ii (1994), pp. 35-54.

2.  And inaccurately paraphrases: whereas Descartes referred to "those who give [the workers] orders to do the work, or promise to pay for it" as the "accidental and remote causes" of the work, Nadler reports him as claiming that the command or the promise of compensation are the occasional cause of the work produced (ibid., pp. 40, 42).

3.  Pace Nadler: "At times, Descartes really does seem to think that the body is the efficient cause of sensory ideas. His language in the Meditations suggests as much. . . .'


Finally, our discussion of occasional causes and mind-body interaction was meant more to establish some general guidelines for further careful investigation than to settle everything once for all. What is most important is to recognise that there have been very different concepts of causation employed in the last four centuries, that the Humean notion is anachronistic when applied to Descartes, that his own use of 'causal locutions' proves nothing,3 that the slogan 'The effect must resemble its cause' is not a naive remark, still less a simple blunder, but a sophisticated scholastic axiom requiring careful interpretation, etc. To the extent that these points are acknowledged and built upon by Descartes scholars, we welcome more detailed dialogue of the sort that Nadler has initiated in his discussion of mind-body interaction.

ST. JOHN'S COLLEGE, OXFORD                                                                GORDON BAKER

MANSFIELD COLLEGE, OXFORD                                                   KATHERINE J. MORRIS