The 'Scandal' of Cartesian Interactionism

R. C. Richardson

Mind, New Series, Vol. 91, No. 361. (Jan., 1982), pp. 20-37.

Stable URL:
Archived at

1. Introduction: mind-body interaction

It is standard fare to object to dualism of a Cartesian stripe on the grounds that, because it would make mind and body utterly diverse in nature, it is unintelligible how mind could act on body or body on mind; for the dualist would make the mind non-extended and immaterial, and how such a thing could conceivably act on something as heterogenous as a physical system is something we are incapable of understanding. In describing 'Descartes' Myth' and arguing it is premised on a 'para-mechanical hypothesis', Ryle wrote :

Minds are things, but different sorts of things from bodies; mental processes are causes and effects, but different sorts of causes and effects from body movements.

... there was from the beginning felt to be a major theoretical difficulty in explaining how minds can influence and be influenced by bodies. How can a mental process, such as willing, cause spatial movements like the movements of the tongue: How can a physical change in the optic nerve have among its effects a mind's perception of a flash of light?

'Transactions between minds and bodies', Ryle tells us, 'involve links where no links can be.'1 Bernard Williams dubs this 'the scandal of Cartesian interactionism': 'there was', he tells us, 'something deeply mysterious about the interaction which Descartes' theory required between two items of totally disparate natures, the immaterial soul, and the [pineal] gland or any other part of an extended body.'2

1      The Concept of Mind (New York: Barnes and Noble, 1949), pp. 10 and 66.
2      Descartes (Penguin Books, 1978), p. 287.

This objection to Cartesianism, which I shall dub 'the problem of heterogeneity', was raised by Pierre Gassendi (in May of 1641) while criticizing the Meditations:

[It] still remains to be explained how that union and apparent intermingling [of mind and body] ... can be found in you, if you are incorporeal, unextended and indivisible .... How, at least, can you be united with the brain, or some minute part in it, which (as has been said) must yet have some magnitude or extension, however small it be ? If you are wholly without parts how can you mix or appear to mix with its minute subdivisions ? For there is no mixture unless each of the things to be mixed has parts that can mix with one another (HR II, p. 201).1

1. Elizabeth Haldane and G.R.T. Ross, The Philosophical Works of Descartes (Cambridge: Cambridge University Press, 1931). All references will occur in the text.

Descartes had no appreciation for the objection, responding, in effect, that no such difficulty existed:

At no place do you bring an objection to my arguments; you only set forth the doubts which you think follow from my conclusions, though they arise merely from your wishing to subject to the scrutiny of the imagination matters which, by their own nature, do not fall under it (HR II, p. 232).

Princess Elizabeth raised a similar objection a scant two years later (in May of 1643), and Descartes took up the issue at more length. In his subsequent correspondence with her, Descartes mustered two responses. The first was essentially that deployed against Gassendi; that is, that thinking there is a problem at all is due to an illicit comparison of mind-body interaction to causal interaction in the physical realm. The second was that the union of body and soul was not one which could be known by the intellect, and that it was impossible to comprehend simultaneously the distinctness and the union of body and soul.

Commentators have generally found Descartes' response wanting. The two responses seem, jointly, to be little more than an insistence that the interaction of mind and body be accepted, together with a claim that such interaction is inexplicable. It seems to be the consensus that Descartes has provided no explanation of the interaction of mind and body, and so has not dealt with the problem of heterogeneity.

Descartes'  response1 has,  however,  been  misunderstood  and underestimated. When properly interpreted, the two responses show us that there is no sound reason to think heterogeneity renders mind-body interaction incoherent and effectively explain why his critics would think such interaction to be incoherent. We shall take up the two responses in turn.

2. The interaction of heterogenous substances

In his letter to Princess Elizabeth of May 1643, Descartes drew a distinction between four 'primitive notions which are as it were models on which all our other knowledge is patterned'. These were: first, notions with universal applicability such as being, number, and duration; second, those applicable to all bodies, the most important of which is extension; third, those applicable to the soul, which includes only thought; and, fourth, the notion of the union of body and soul, on which, Descartes says, 'depends our notion of the soul's power to move the body, and the body's power to act on the soul' (K, p. 138).1 Descartes then went on to suggest that knowledge could be gained only if we kept these various domains distinct and did not try to reduce one to another. The error to which Princess Elizabeth is subject has just this source: we go wrong when 'we try to use our imagination to conceive the nature of the soul, or to conceive the way in which the soul moves the body after the manner in which one body is moved by another' (ibid.). This was foreshadowed in Descartes' allegation that Gassendi was illegitimately extending the imagination beyond its proper domain (viz., the spatial; cf. HR I, pp. 185-186) and this is the point of his conclusion against Gassendi: 'when you wish to compare the union of mind and body with the mixture of two bodies, it is enough for me to reply that no such comparison ought to be set up, because the two things are wholly diverse' (HR II, p. 232). Mind-body interaction appears incomprehensible only because Elizabeth, and Gassendi, have illicitly taken mind-body interaction to be similar to causation in the physical realm. They have done this, Descartes thinks, because of an over-extension of the imagination, which is a form of physical representation whose primary function is to represent the geometrical properties of corporeal nature.

1       Anthony Kenny, Philosophical Letters [of Descartes] (Oxford: Clarendon Press, 1970). All references will occur in the text.

In the letter to Elizabeth, he attempted to explain and extend this line using an analogy:

when we suppose that heaviness is a real quality of which all we know is that it has the power to move the body that possesses it towards the center of the earth, we find no difficulty in conceiving how it moves the body nor how it is united to it. We do not suppose that the production of this motion takes place by a real contact between two surfaces (K, p. 139).

Heaviness, so conceived, is an intrinsic motive force which impels bodies toward the earth. Anthony Kenny's response to Descartes' analogy is typical: 'Elizabeth's question was not whether body and soul were united, but how; and the inconceivability of this is not lessened by being compared to another inconceivable.'1 Margaret Wilson remarks, similarly, that 'if Descartes is going to take refuge in the brute fact of interaction, it seems he should not need such apparatus as the "gravity" pseudo-concept in order to do so'.2

The analogy is actually a rich one. Two morals are particularly relevant for Descartes' argument. Descartes is, firstly, explaining what is wrong with 'animistic' theories in natural science.3 He is claiming that the Scholastic account of natural motion has imported concepts proper to mind-body interaction into the physical realm where they do not belong. Descartes tells us he had once held that gravity was 'a real quality of a certain order' (HR II, p. 254). Reflecting on this, Descartes tells us in the Sixth Replies that 'the chief sign that my idea of gravity was derived from that which I had of the mind is that I thought that gravity carried bodies toward the centre of the earth as if it contained some knowledge of this centre within it' (HR II, p. 255). Each sort of concept has its proper domain, and only confusion results from such trans­positions. Gassendi and Elizabeth have implicitly assimilated mind-body interaction to physical interaction just as the Scholastics assimilated physical interaction to mind-body interac­tion. It is no less an error in either case.

1      Anthony Kenny, Descartes: A Study of the Philosophy (New York, Random House, 1968), p. 224.
2      Descartes (London: Routledge and Kegan Paul, 1978), p. 215.
3       I realize this is not Descartes' use of 'animism'. 'Anthropomorphic' would perhaps be more accurate, but also more awkward.

There is yet a second moral to be drawn from Descartes' example. This is simply that non-mechanical 'causation'—causation irreducible to Cartesian 'push'—is not incoherent; for, even transposed illegitimately, the resulting theory of motion makes good sense. Of course, Descartes is quick to urge, such force is unnecessary in explaining gravitational phenomena. But this is irrelevant to the present issue: the resulting account of gravitational phenomena, as Descartes sees it, is incorrect rather than incoherent. Descartes elaborates the point in a letter for Arnauld (29 July 1648):

Most philosophers, who think that the heaviness of a stone is a real quality distinct from the 'stone, think they understand clearly enough how this quality can impel the stone towards the centre of the earth ... in order to represent this heaviness to themselves they are using the idea they have within them of incorporeal substance. So it is no harder for us to understand how the mind moves the body, than it is for them to understand how such heaviness moves a stone downwards (K, p. 236; cf. his letter to Marsenne [20 April 1646], K, p. 191).

We must therefore have a concept of non-mechanical 'causation'. Its proper application, Descartes is pressing, is in the human realm—in understanding mind-body interaction and human action.1

Descartes is quite right to hold that his opponents are thinking of mind-body interaction on a mechanical (or paramechanical) model; moreover, his rejection of this assimilation is not arbitrary in the least. Descartes had, he thought, demonstrated that mind and body were distinct and of different natures. It is, of course, to this dimension of the doctrine of the Meditations that Gassendi and Elizabeth are here objecting; they are doing so by drawing out what they think is an unacceptable consequence of this doctrine. But not only is it an error to think the doctrine of the Meditations would make interaction incoherent, the doctrine itself generates a resolution. Descartes is committed to pressing that mechanical causation—that form of causation proper to the interaction of bodies—is ultimately a function of their extension;2 it follows directly that mind-body interaction cannot be assimilated to mechanical causation. Given Descartes' conviction that he had so clearly demonstrated that mind is not extended, the consequence would have seemed so obvious as to be hardly worth mentioning. It is small wonder that he evinced irritation with Gassendi over the issue.

1      A natural possibility for explicating such non-mechanical causation is in terms of 'agent causation'. See Alan Donagan's 'Determination and Freedom: Sellars and the Reconcilationist Thesis' in H. N. Castaneda, ed., Action, Knowledge, and Reality (Indianapolis: Bobbs-Merrill, 1975); and, also, R. M. Chisholm's Person and Object (La Salle: Open Court, 1976), chapter 2.
2      This is manifested most clearly in the conviction that action at a distance is incoherent.

There is, though, a residual problem which gives the objection of Gassendi and Elizabeth some plausibility. Descartes has no answer to the question of how the mind influences action except in terms of his physiological model, and ultimately this comes down to the action of the mind on the pineal gland; when it is then asked, as it inevitably is, how the mind influences the pineal gland, Descartes has no answer at all. But that this is so is no objection. It is evident that, as a request for an intermediate cause, it would be illegitimate to ask how the mind influences the pineal gland; but the problem is not that Descartes took it that the mind acted directly on the pineal gland.1 Since the distinction between intermediate and proximate causes is one proper to mechanical causation, the only sense to the question would be to explain mind-body interaction in more primitive terms. As we have seen, Descartes held this form of interaction to be one inexplicable in other terms: it is a 'primitive notion'.

1.       Cf. Curt Ducasse, 'In Defense of Dualism' in Sidney Hook, ed., Dimensions of Mind (London: Collier Macmillan, 1960), p. 88.

However much our desire for homology might revolt against this, the position is, logically, impeccable. Perhaps this can be illustrated by shifting to a contemporary context. We can offer an explanation of magnetic force (and electrical attraction) in terms of Maxwell's electrodynamics, and this means, effectively, explaining one sort of interactive force in terms of a more basic one. Recent work has seen the attempt to integrate Maxwell's theory within a more general theory capable of explaining electrodynamics and the so-called 'weak force' of quantum field theory. (One example of weak interaction is beta decay, in which a neutron breaks down into a proton, an electron, and a neutrino with the exchange of a boson.) When asked for an explanation of this force, of gravitation, or of the 'strong force' (which binds together protons and neutrons in the nucleus), the only possible explanation is in terms of a more basic type of force. Should such a theory finally be accepted, we would be able to explain gravitational force in more basic terms common to the other four 'fundamental' forces. We would then be able to explain how gravitation works—something we cannot now do—but then the last postulated and most basic force, usually termed supergravity, would in turn be quite 'mysterious'. To ask how it works would be to ask for a more fundamental force, and there simply is not one. If a unified field theory finally proves unacceptable, then in all likelihood gravitational force will be fundamental, and to ask how it works will be illegitimate. The situation would then be precisely parallel to Descartes'. That some mode of interactive force be simply accepted is unavoidable, and whether there is one or more is of no consequence. So long as Descartes holds, as he must, that there are two fundamental and irreducible forms of causal interaction, there can be no sense given to the question of how mind acts on body. Similarly, if there is one or more than one fundamental type of force, then an explanation of how it, or they, operate can only be rejected as ill-founded. This was the substance of Descartes' reaction.

3. Diversity in nature and the union of mind and body

The second of Descartes' responses to Elizabeth is more difficult to make out and also more important to the Cartesian system. He responded by saying that while those notions applicable to the soul 'can be conceived only by pure intellect', and those applicable to body alone by the pure intellect, but better by the intellect aided by imagination, the final category of notions—those, that is, relevant to the union of body and soul—'can be known only obscurely by pure intellect or by intellect aided by imagination, but ... can be known very clearly by the senses' (K, p. 141). He added in an oft-cited passage:

It does not seem to me that the human mind is capable of conceiving at the same time the distinction and the union between body and soul, because for this it is necessary to conceive them as a single thing and at the same time to conceive them as two things; and this is absurd (K, p. 142).

The response allows of several interpretations.

The first interpretation, defended by Ruth Mattern,1 has Descartes claiming that it is the disparate natures of the mind and body that make the interaction of mind and body impossible to comprehend: so long as we attend to the difference in the natures of mind and body, to comprehend their union is impossible. Such a position is not without credibility. Elizabeth had, after all, written to Descartes that 'it would be easier for me to attribute matter and extension to the soul, than to attribute to an immaterial being the capacity to move and be moved by a body' (K, p. 140). Descartes responded that she should' feel 'free to attribute matter and extension to the soul because that is simply to conceive it as united to the body' (K, p. 143). If, in order to think of mind and body as unified we must think of the soul as possessing matter and extension, then it is the difference in their natures (and in particular, the incorporeality of the soul) that makes the unity difficult to understand.

1. 'Descartes' Correspondence With Elizabeth: Concerning Both the Union and the Distinction of Mind and Body', in M. Hooker, ed., Descartes (Baltimore: Johns Hopkins Press, 1978).

The interpretation will not do, however. The simple fact is, and, as we have already seen, this was the thrust of the first response, Descartes did not think that there was a problem in understanding the interaction of corporeal and incorporeal agents per se. To Cleselier, Descartes had written (in speaking of Gassendi's objection):

I declare that the whole of the perplexity involved in these questions arises entirely from a false supposition that can by no manner of means be proved, viz. that if the soul and the body are two substances of diverse nature, that prevents them from being capable of acting on one another; for, on the contrary, those who admit the existence of real accidents, like heat, weight, and so forth, do not doubt that these accidents have the power of acting on the body, and nevertheless there is more difference between them and it, i.e. between accidents and a substance, than there is between two substances (HR II, p. 132).

If there is no difficulty in the interaction of two substances having different natures, then it cannot be this which makes the unity of mind and body inconceivable together with their distinctness. 

4. Distinctness in number and the unity of body and soul

An alternative interpretation would take Descartes to be claiming that it is the distinctness of mind and body that makes the union of mind and body incomprehensible; so long as we attend to the fact that they are numerically distinct and are capable of separate existence, we will be unable to understand the way in which they are united. The problem, so understood, is not one of grasping the fact that mind and body do interact, but, rather, one of comprehending the nature of their union. The fundamental datum which any interpretation must accommodate is their intimacy:

Nature ... teaches me by these sensations of pain, hunger, thirst, etc., that I am not only lodged in my body as a pilot in a vessel, but that I am very closely united to it, and so to speak so intermingled with it that I seem to compose with it one whole (HR I, p. 192).

Or, as Descartes says in the Discourse:

it is not sufficient that [the rational soul] should be lodged in the human body like a pilot in his ship, unless perhaps for the moving of its members, but ... it is necessary that it should also be joined more closely to the body in order to have sensations similar to our own (HR I, p. 118).

The purport of such passages is far from clear. It may be primarily an insistence that there is a peculiar intimacy of the passions to the soul; that is, we may have no more than an insistence of some sort of epistemic immediacy. There may, though, be more to Descartes' point than this. Merleau-Ponty writes:

if I say that my foot hurts, I do not simply mean that it is a cause of pain in the same way as the nail which is cutting into it, differing only in being nearer to me; I do not mean that it is the last of the objects in the external world, after which a more intimate kind of pain should begin, an unlocalized awareness of pain in itself, related to the foot only by some causal connection ... my body does not present itself as the objects of external impressions do.1

Taken as a gloss on Descartes, the point is not merely one of epistemic immediacy; it is, rather, one which requires us to acknowledge a unity of body and soul—a 'substantial union' constitutive of a person. Whether it is possible to make systematic sense of such a doctrine within a Cartesian framework is a question yet to be resolved. We will find some reason to think it is.

1.       Phenomenology of Perception, translated by Colin Smith (London: Routledge & Kegan-Paul, 1962), pp. 93; cf. pp. 198—199.

(a) Passions as species of cogitatio

It is thought which Descartes holds is the 'principal attribute of the soul' (HR I, p. 240). Sensation, and imagination, are merely 'modes of thought'. Understanding Descartes' position on the union of mind and body depends on this doctrine, and requires resolving an ambivalence in his treatment of the difference between sensations and thought, or, more generally, the passions and actions of the soul. Descartes remarks:

all these sensations of hunger, thirst, pain, etc., are in truth none other than certain confused modes of thought which are produced by the union and apparent intermingling of mind and body (ibid.).

In the Principles, we find Descartes saying that 'thought constitutes the nature [i.e., the principal property] of thinking substance' and that 'everything that we find in mind [including imagination, feeling, will, and sensation] is but so many diverse forms of thinking' (HR I, p. 240). Similarly, Descartes writes: 'By the word thought I understand all that of which we are conscious as operating in us. And that is why not alone understanding, willing, imagining, but also feeling, are here the same thing as thought' (HR I, p. 222).

This is not, of course, to deny any distinction between these various 'modes of thought'. In the Passions of the Soul, after distinguishing between actions and passions of the soul saying that the actions 'proceed directly from the soul, and appear to depend upon it alone' while the passions are received 'from the things which are represented by them' and are such that 'it is often not our soul which makes them what they are' (HR I, p. 340). Descartes offers the following definition:

After having considered in what the passions of the soul differ from all its other thoughts, it seems to me that we may define them generally as the perceptions, feelings, or emotions of the soul which we relate specially to it, and which are caused, maintained, and fortified by some movement of the spirits (HR I, p. 344).

Such passages as these invite us to adopt the following interpretation: there is one attribute proper to the mind, and that is thought; there are, further, several modes of thought, the most basic distinction being that between actions and passions; sensations, and other passions, are but determinate species (or types) of thought; and the passions, no less than the actions, of the soul could therefore exist, like all thoughts, in the absence of the body. On this interpretation, the distinction between actions and passions is a distinction per accidens: actions and passions of the soul are distinguished only 'on the basis of an attribute that is accidental and not essential to them. Sensations are a determinate type of thought that just happen to come about because of 'the union and apparent intermingling of mind and body'.

Such an interpretation is wholly unacceptable. It leaves unexplained both Descartes' claim that experience is requisite to grasp the interaction of mind and body, and his contention that the union of mind and body is incomprehensible while we attend to the distinctness of mind and body. If the passions are simply species of cogitatio, then, like the acts of pure understanding, they should be grasped only (or best) by the pure intellect. Furthermore, the possibility of disembodied sensation is not something Descartes is prepared to embrace. He wrote to More in August 1649: 'the human mind separated from the body does not have sensation strictly so called' (K, p. 256). And, finally, by making sensation and imagination akin to pure understanding, treating the mind as a pilot in a ship—or a ghost in a machine—is quite inevitable. Sensation and imagination would, after all, have only an extrinsic connection with the body. It would be the case that 'when my body is hurt, I, who am merely a thinking thing, should not feel pain, for I should perceive this wound by the understanding only, just as the sailor perceives by sight when something is damaged in his vessel' (HR I, p. 192). The passions are not species of cogitatio.

(b) Passions as attributes of the soul

Interpreting the concept of a mode somewhat differently has marginally different consequences. In the Principles, Descartes suggests we can 'best apprehend the diverse modes of thought' if 'we consider them simply as modes of the things in which they are' (HR I, p. 246). Just as movement and figure are taken as modes of extension without being determinate species of extension, so too, imagination and sensation may be taken as modes of thought without being species of thought. Movement and figure are attributes of things whose nature is extension; sensation and imagination would accordingly be attributes of things whose nature is thought. Margaret Wilson, in her penetrating treatment of Descartes, adopts this—or a closely allied—interpretation. She writes:

Thus, in imagining and in certain kinds of remembering the mind is said to 'utilize' or 'turn to' impressions existing in the brain; and in the experiences of sensations and passion the mind is affected by changes in the body's organs and may even become aware of itself as united or 'intermingled' with the whole body. In these sorts of mental occurrences 'some understanding is comprised ... but because of the dependence of the thoughts on physical states or occurrences they cannot be construed as 'pure understanding'. ... Pure understanding is carried on independently of all physical processes; any physiological study will necessarily be irrelevant to it.1

The distinction between actions and passions will still be accidental. On this, or roughly what Wilson terms the 'Natural Institution' view, the connection between a passion and its proximate cause is contingent. A passion will, to be sure, require some cause, but the specific sort of cause is inessential to the passion per se.

This interpretation will allow us to explain why experience is necessary to grasp the nature of mind-body interaction in so far as the particular cause will remain mysterious without empirical investigation. But the contingency of the connection leaves us no more able to explain the inconceivability of mind-body union (given their distinctness) than did the last interpretation. It still also allows for disembodied sensation and treats the mind as a 'pilot in a ship'. A comprehensive interpretation must take us yet further.3

1      Descartes, p. 181.
2      Cf. ibid. p. 207.
3      Wilson knows quite well that the Natural Institution conception does not accommodate all of Descartes' views. She contends only that Descartes would have done better to stick.with the Natural Institution conception. Cf. ibid. pp. 210—218.

It is possible to amend the treatment of passions as attributes of the soul. What is crucial is the insistence that the defining characteristic of the passions—their causal genesis—is essential to them, as passions, and not accidental at all.4 This allows us to take Descartes' definition of passions as more than merely nominal, and, in proper Cartesian form, to regard the defining characteristic as essential.1 It will, moreover, relieve us of the burden of allowing for disembodied sensation. God, or an evil deceiver, could produce in the mind a mode identical in its intrinsic character with a mode caused by the mind's interaction with a body. But this would not be to produce in it a mode of the same kind as that arising from the union of mind and body. .

4      This interpretation has been defended  in the present context by Wayne Backman (in conversation) and Alan Donagan (in correspondence).

It may even be the case that such a view will mitigate the apparent commitment to treating the mind as a pilot in a ship. It could be held (contrary to my suggestion above) that it is enough to maintain this tenet that the passions be impossible in a disembodied form: damage might afflict a ship without the pilot and the pilot's perceptions could remain, though not veridical; but, save with embodiment, there would be no passions.

What this cannot do is explain the simultaneous incomprehen­sibility of the union and distinctness of mind and body. It is true that this interpretation guarantees we cannot simultaneously think of them as separate and as united. We could, though, simul­taneously think of them as separable and as united. This is tantamount to simultaneously thinking of them as distinct and as united. The interpretation, thus, falls short on a crucial count.

(c) Passions as attributes of a composite

It is tempting to conclude that Descartes' position is afflicted with a fundamental ambivalence. Some commentators—most notably Margaret Wilson—have drawn just this conclusion.2 But we can discern yet another dimension to Descartes' treatment that does make Descartes' response to Elizabeth a cogent and coherent one. Descartes wrote early in Part II of the Principles:

It may be concluded also that a certain body is more closely united to our mind than any other, from the fact that pain and other of our sensations occur without our foreseeing them; and that mind is conscious that these do not arise from itself alone, nor pertain to it in so far as it is a thinking thing, but only in so far as it is united to another thing, extended and mobile, which is called body (HR I, p. 255).

1       It is doubtful that this view can accommodate the demand that knowledge be grounded in clear and distinct perception (HR I, pp. 232, 237). To define the passions in the way now contemplated is to systematically infuse something alien into the intrinsic nature of the passions.
2      Wilson, Descartes, p. 205; cf. D. M. Armstrong, A Materialist Theory of Mind (London: Routledge and Kegan Paul) 1968, pp. 24—26. Also see n. 15 above.

Two points are crucial here: first, that understanding the etiology of our sensations demands positing an interaction with the body; and, second, that sensations do not 'pertain to' the mind save in so far as it is united to the body'. The former point is that emphasized in the Meditations (cf. HR I, p. 191), and is consistent with maintaining the 'active faculty' which produces sensations is in the body and yet only extrinsically connected with the sensations themselves. The latter point is actually more important in understanding the union of mind and body. The first point demands that only an embodied soul can have sensation: a soul no longer in communion with the body would have neither sensations nor any of the passions. This second point, that sensation 'pertains to' the mind only as embodied, goes beyond this simple point.

To grasp how far beyond this point Descartes wants and needs to go, we need again to return to the concept of a mode. In the Notes Against a Programme, Descartes writes: 'the nature of a mode consists in this, that it can by no means be comprehended, except it involve in its own concept that of the thing of which it is a mode' (HR I, p. 440); nonetheless, he tells us, 'we can easily comprehend a substance [in which a mode inheres] apart from a mode [belonging to it]' (HR I, p. 436). We have already acknowledged Descartes' comment that saying, e.g., that sensation or imagination is a 'mode of thought' is saying they are modes of things which think; that is, that they are attributes of things which think, but not the principal attributes of those things. These passages together entail that for sensation or imagination to be a 'mode of thought' is for them to be attributes which cannot be attributed to anything lacking the attribute of thought. It does not entail that the entity to which they are attributed, and which therefore has the attribute of thought, is a simple entity (HR I, p. 437); it thus does not imply that any entity which thinks is, ipso facto, capable of enjoying sensation of imagination.

Descartes says in the Meditations that, although he finds in himself 'faculties employing modes of thinking peculiar to themselves, to wit, the faculties of imagination and feeling, without which I can conceive myself clearly and distinctly as a complete being', nonetheless 'they cannot be so conceived without an intelligent substance in which they reside' (HR I, p. 190). Now, if the entity to which imagination and feeling are attributed as modes is a simple entity, that entity would obviously be the mind. But then why should we grant imagination and feeling, as faculties, are separable from the soul ? It would be true that the passions would never be exercised; that is, there would be no acts of imagination, feeling, or sensation apart from the body. But this does not show, or even suggest, the corresponding faculties are lacking. On the other hand, if the entity to which imagination and feeling are attributed is a composite entity, it is evident that imagination and feeling, as faculties, would not be integral to our conception of the mind: to imagine the mind as separate from body would demand distinguishing thought from sensation, feelings, and imagination. In other words, to distinctly conceive of the mind we would need to press that the faculties of sensation, imagination, and feeling are separable from our conception of the mind only if the existence and not merely the exercise of these faculties depended on the body. These faculties, that is, would be attributes of a complex, and not a simple, entity. Descartes writes:

there are ... certain things which we experience in ourselves and which should be attributed neither to the mind nor body alone, but to the close and intimate union that exists between the body and the mind... . Such are the appetites of hunger, thirst, etc., and also the emotions or passions of the mind which do not subsist in mind or thought alone . . . and finally all the sensations (HR I, p. 238).

The passions are attributes not of the mind, nor yet of the body, but of the embodied mind—the mind-body complex.1 Pure spirit can have neither images nor sensations. Images and sensations we do have. Thus we are not pure spirits.

1. On the first interpretation of subsection (b), the definition of the passions was a definition per accidens: they are those attributes of the soul which happen to be brought about by the interaction of soul and body. On the interpretation now being proposed, like the second in subsection (b), the definition of the passions in now a proper definition

Given that we take Descartes in this way when he claims sensation and imagination are 'modes of thought', we can obtain a clear, coherent, and cogent rendering of his second response to Elizabeth. Sensation and imagination would be taken to be attributes of a composite entity—namely, an embodied soul—and would, therefore, be attributes of something which has the attribute of thought. They would be inconceivable apart from thought; for without that attribute there would be no embodied soul. But if we attend to the fact that mind and body are numerically distinct, we may attribute states to the mind, or we may attribute states to the body, or we may think of their causal interaction; what we do not, and cannot, do is think of them as a composite entity. That is, thinking of them as distinct precludes our thinking of them as forming a single subject, a substantial whole; yet this is what we must do if we are to understand how the passions are a function of the union of mind and body, or to understand properly the nature of that union. As Descartes said to Elizabeth, those who 'have no doubt that the soul moves the body and that the body acts on the soul' do so because they 'regard both of them as a single thing'; and, he continues, 'to conceive the union between two things is to conceive them as one single thing' (K, p. 141). But to conceive of mind and body as a single thing is not to think of them as being of a single nature. Still less is it to think of the mind as, in a literal sense, extended.1 It is, rather, to acknowledge that mind and body are constituents of a single composite entity with a composite nature.

In a sense, this means treating the person as a substance. If a substance is anything to which we can attribute properties2 then the composite consisting of mind and body conjoined is in this sense, a substance. Understanding the nature of the union of mind and body requires understanding the passions, and this demands that we think of the composite as a subject—that is, as if it were a substance. Yet attending to the distinctness of mind and body means treating it no longer as a single unified subject, and therefore no longer as a substance, but as two substances. It is for this reason that this union must be taken as a 'substantial union'—that is, as a union the result of which is thought of as a substance; and it is for this reason Descartes says that in order to understand the union of mind and body 'it is necessary to conceive [of] them as a single thing and at the same time to conceive [of] them as two things' (K, p. 142). This, as he says to Elizabeth, is something we cannot do.

1. We must not neglect Descartes' warning that 'the extension of the matter is of different nature from the extension of the thought' (K, p. 143; cf. his letters to More of 5 February and 15 April 1649 [K, pp. 239, 252]).
2. Descartes means to contrast substance with attribute. The operative concept of a substance, then, is that of the subject of attributes, or that of something which requires nothing else for existence (HR I, p. 223).

5. Conclusion : The interaction of mind and body

The problem of how two such heterogeneous things as mind and body could interact is, as Descartes so clearly saw, no problem at all. The illusion of a problem has two sources: the presumption that psychophysical interaction must be similar in kind to physical interaction, and the conviction that psychophysical dualism must lead to a naive attempt to attain an exhaustive segregation of properties into those that are attributable to the purely physical and those that are attributable to the purely mental.

That psychophysical interaction would be held, by the dualist, to be unique in kind is no surprise; after all, like any dualist, one reason Descartes had for embracing dualism was his conviction  that some human abilities are inexplicable on physical grounds.1 Descartes is very nearly infamous for having argued not only that human linguistic abilities are uniquely human but also that all animal capacities are physically, mechanically, explicable. (See his letter to the Marquis of Newcastle [K, pp. 205-208] and the discussion in the Discourse [HR I, p. 116].) If the basic principles of animal motion are mechanical, and the reason for holding animals are divergent from humans is, e.g., the latter's linguistic competence, then mechanical principles must be incapable of explaining human linguistic abilities. The animals are also not credited with sensation or imagery, though 'all the movements of the spirits and of the gland which excite the passion in us, are none the less in them' (HR I, p. 356). Something more than 'the movements of the nerves and muscles' (ibid.) is needed for the passions no less than for pure understanding. It is impossible to see how a 'para-mechanical hypothesis' would or could be of any avail; the whole goal of dualism is to introduce some sort of explanation for what is thought to be physically (or, for Descartes, mechanically) inexplicable.

1.       See Margaret Wilson, Descartes, pp. 182—184; and her 'Cartesian Dualism' in M. Hooker, ed., Descartes, pp. 197—211.

Thus, lacking an argument, as we surely are, that heterogeneous forms of interaction are unacceptable, this constitutes no argument against dualism. That this is so is something Descartes is presupposing, and not arguing for, in attempting to explain the nature of mind-body interaction. Though more recent critics of dualism have seen that his account of interaction does presuppose this, what they have failed to recognize is that Descartes' goal was to explain the nature of the union between mind and body—not its possibility. This account of a 'substantial union' is a difficult one, and, as we have remarked, Descartes himself was less than definitive.1 The problem, though, of how psychophysical union could be more than simple interaction is clear enough, and the outline of an answer can be discerned in Descartes' writings. The only scandal in this whole matter is the failure of commentators to see the force of Descartes' reply.2



1. This ambivalence is also manifested in Descartes' use of 'thought'. This may be taken to be a general term encompassing all intentional phenomena (see Alan Donagan's 'Descartes' Synthetic Treatment of the Real Distinction Between Mind and Body', in Hooker, ed., Descartes, pp. 186—196). Alternately, it may be taken as the principal attribute of the mind. This is cogitatio proper, and it is in this sense I use the expression.
2. I have had the benefit of discussing this work in two seminars at the University of Cincinnati during 1979—80. I am especially thankful for the help of Wayne Backman, Alan Donagan, Robert Faaborg, Donald Gustafson, and W. E. Morris.

The initial work was supported by a grant from the Taft Faculty Committee at the University of Cincinnati.