Descartes on Misrepresentation

Paul Hoffman. Journal of the History of Philosophy. Volume 34, Issue 3. July 1996.

Any adequate theory of mental representation must include an account of how sensory experience is capable of misrepresenting the world. Misrepresentation occurs, for example, when we look at a straight stick in a glass of water and see it as bent. In order for our visual experience to misrepresent the stick, it must represent the stick, but it must represent the stick as other than it is, as bent rather than straight. To account for misrepresentation, an adequate theory of representation must therefore explain how our sensory experience can represent an object as other than it is.

Many philosophers have offered accounts of representation according to which sense experience cannot or at least does not misrepresent the world in optimal conditions or even in normal conditions Descartes, however, thinks otherwise. In optimal conditions our visual experiences represent physical objects as colored. Descartes maintains that this too counts as misrepresentation, on the ground that color is not a property of physical objects.

In the Third Meditation Descartes raises the possibility that our ideas of light and colors, sounds, smells, tastes, heat and cold, and the other tactile qualities are materially false (AT VII 43-44). In explanation of this notion, he says that material falsity occurs in ideas when they represent what is not a thing as if it were a thing (non rem tanquam rem repraesentant) (AT VII 43; CSM II 30). He asserts that all ideas are as if of things (nullae idea nisi tanquam rerum esse possunt) (AT VII 44; CSM II 30). So, for example, the idea of cold represents cold to me as something real and positive. But if cold is a privation of heat, then the idea of cold is materially false, since it represents a privation (a non-thing) as a thing (as something real and positive) (AT VII 43-44; CSM II 30).

Material falsity is thus a kind of misrepresentation: to be materially false, at least as it is characterized in the Third Meditation, is to represent a non-thing as if it were a thing. But it is an especially troubling kind of misrepresentation. In her book on Descartes, Margaret Wilson alleges that the concept of material falsity is both a red herring and an embarrassment in the context in which it is presented in the Third Meditation.

In this paper I will examine Descartes’s notion of material falsity. He discusses it not only in the Third Meditation, but also in the Objections and Replies and in the Principles. After a brief introduction to Descartes’s terminology, I will first examine Wilson’s interpretation of the Third Meditation. I will argue that, contrary to her reading, Descartes does not believe that our ideas of light, colors, cold, heat and the like represent what is not a thing as if it were a thing. They are not materially false in that sense. Then I will turn to the Objections and Replies. I will argue that both Arnauld’s objection challenging the very coherence of the notion of material falsity and Descartes’s notoriously obscure reply are based on their acceptance of fundamental elements of the Aristotelian account of perception. Next I will explore how our ideas of light and colors, heat and cold and such might be misrepresentations even if they do not represent what is not a thing as if it were a thing. Finally I will provide a reading of the Principles passages that is consistent with my interpretation and that also vindicates what I take to be the crucial insight of Wilson’s interpretation.

1. Basic Notions

Ideas, Descartes tells us, are thoughts that are as if images of things (AT VII 37; CSM II 25). He says that ideas are as if images of things, rather than that they are images of things, because he thinks that only some of our ideas are images. The imagination, he thinks, can have only corporeal things as objects, and it requires a species that is a real body (AT VII 387; CSM II 265). But he thinks that our ideas represent a variety of things-including God, corporeal and inanimate things, angels, animals, and other men like himself (AT VII 42-43; CSM II 29)-not all of which are corporeal. The idea of God, for example, is not an idea of something corporeal and does not require a species that is a real body. But even if the idea of God is not an image, it is like an image in two respects: (i) there is some thing which I take as the object of that thought, and (ii) that thought is a likeness of the thing (AT VII 37; CSM II 26).

Descartes ties the notion of an idea’s representing a thing to its containing such-and-such objective reality (AT VII 40; CSM II 28). An idea’s objective reality is contrasted with its material or formal reality-that is, with its being a mode of thought or an operation of the intellect (AT VII 8, 40; CSM II 7, 27). The objective reality of an idea, he says, is the “being of the thing which is represented by the idea, in so far as it exists in the idea… For whatever we perceive as if in the objects of ideas, is in the ideas objectively” (AT VII 161; CSM II 113).

Descartes maintains that, strictly speaking, only judgments-that is, affirmations and denials-are true or false. But he also says that ideas can have another kind of falsity, which he calls material falsity (AT VII 36-37, 43; CSM II 25-26, 30). He says that ideas are materially false-they provide material for error-when they represent what is not a thing as if it were a thing (non rem tanquam rem repraesentant) (AT VII 43; CSM II 30).

If an idea represents no thing, Descartes infers that it proceeds from nothing (AT VII 44; CSM II 30). And if it proceeds from nothing, then it can have no objective reality. The reason for this is that an idea can have no more objective reality than its total and efficient cause has formal reality (AT VII 41; CSM II 28-29), and nothing has no formal reality. Thus his account of material falsity commits him to the possibility of ideas that lack objective reality.

Descartes maintains then that all ideas are as if of things, but that not all ideas represent things. What is being distinguished here? One plausible answer is that Descartes is claiming that all ideas seem to represent things, but that not all ideas do represent things. Whether Descartes should or even can draw such a distinction has been thought to be problematic. Let us begin with Wilson’s argument that the introduction of the concept of material falsity in the Third Meditation undermines Descartes’s stated agenda for that meditation.

2. Wilson’s Account of Material Falsity

In the Third Meditation Descartes is attempting to prove that God exists and that our knowledge of God’s existence is prior to our knowledge of the existence of other things outside ourselves, in particular, prior to our knowledge of the existence of bodies. Wilson claims that the concept of material falsity is a red herring in that context because it is ostensibly introduced in order to demonstrate that ideas of sensible real qualities (such as the ideas of cold, of heat, of red) do not by themselves provide sufficient evidence for the existence of anything outside ourselves, but since Descartes also makes use of an independent argument to demonstrate that same conclusion, there is no need for him to introduce the concept of material falsity.

It is an embarrassment, she claims, because it provides the basis for an objection to Descartes’s argument for the existence of God. If there can be ideas that are as if of some positive thing, and thus seem to us to have objective reality, but which in fact are of nothing and thus have no objective reality, then “the objective reality of an idea is not something the idea wears on its face.” If we can be mistaken about how much objective reality an idea has, then perhaps we are mistaken in thinking that the idea of God has so much objective reality (infinite objective reality) that only a being with infinite formal reality could have caused it.

Wilson’s conclusion is that Descartes must have had an ulterior motive for introducing the concept of material falsity. She speculates that he “was determined at all costs to maintain that the ideas of sense, even if they are [as if of things], nevertheless fail to exhibit to us any possibly existent quality in an intelligible manner … [I]n an important (`de re’) sense they are not `of things’.” The implication of her interpretation seems to be, then, that sensory ideas do not have objects, even possible objects; they are merely as if of objects.

In a more recent paper, however, Wilson has modified, or at least clarified, her interpretation of what is involved in an idea’s being materially false. She claims that Descartes thinks that “our sensations are representative in two respects.” First, the idea of cold represents cold to me (even if cold is a privation). Second, the idea of cold presents cold to me in a certain way, as being such and such. In the first sense, she says that the idea of cold referentially represents cold. In the second sense, the idea of cold presentationally represents something real and positive.

Wilson seems to be implying on this new interpretation that Descartes thinks the idea of cold has two objects. One object is what it referentially represents (a privation); the other is what it presentationally represents (something real and positive). An idea presentationally represents what it appears to be of-that is, what it seems to represent referentially. It is still true on this new interpretation, as it was on Wilson’s old interpretation, that the idea of cold does not referentially represent an existent (or even a possible existent); rather, it referentially represents a non-existent.

I am inclined to think that Wilson has got Descartes’s motivations wrong. While it is true that in introducing the concept of material falsity Descartes makes use of our sensory ideas to illustrate the possibility that ideas that appear to have objective reality have no objective reality (because they have no cause), he does not think that as a matter of fact sensory ideas lack objective reality. His argument in the Sixth Meditation for the existence of bodies makes it clear that he thinks that our sensory ideas, however confused and obscure, are caused by bodies or modes of bodies (AT VII 78-80; CSM II 54-55). And his physics rules out the possibility that the idea of cold referentially refers to a privation. The idea of cold is presumably caused either by a particular motion or range of motions of bodies or by the absence of such motions. But according to his physics, the absence of motion is not a privation, it is not a non-thing. Instead, the absence of motion is rest, and rest is no less of a mode, no less of a thing, than motion. Since he is clearly committed to denying that the cause of the idea of cold is a privation, there is nothing motivating him to deny that the idea of cold has objective reality. And he never does deny it. He considers the possibility that cold is a privation only as a way to introduce the concept of material falsity.

Why does Descartes introduce the concept of material falsity in the Third Meditation if he does not believe that sensory ideas lack objective reality? I would argue that the notion of an idea’s having less objective reality than it appears to have is neither a red herring nor an embarrassment. Rather, Descartes is anticipating an objection to his argument for the existence of God:

Nor can it be said that this idea of God is perhaps materially false and so could have come from nothing, which is what I observed just a moment ago in the case of the ideas of heat and cold, and so on. On the contrary, it is utterly clear and distinct, and contains in itself more objective reality than any other idea; hence there is no idea which is in itself truer or less liable to be suspected of falsehood. (AT VII 46; CSM II 31)

Wilson rejects this alternative explanation for Descartes’s introduction of the concept of material falsity:

An alternative explanation, that has been suggested to me, is that Descartes was trying to anticipate what he perceived as a possible response to this theological proof: i.e. that a critic might spontaneously object that the idea of God could, like sensations, represent nothing real. However, it seems that the distinction between the clear and the distinct and the obscure should by itself be adequate basis for an answer to this objection: we don’t need the theory of material falsity.

She seems to be asserting here that we do not need the concept of material falsity to answer the objection that the idea of God could represent nothing. But if we assume that Descartes holds, as surely he does, that the idea of God appears to represent something real, then the objection that the idea of God might not represent something real just is the objection that the idea of God might be materially false. So the concept of material falsity is essential not to answering the objection but to formulating the objection. If her point is that the concept of material falsity itself will not tell us why the idea of God is not materially false, then her comment seems to be true enough, but irrelevant. We should not expect to discover from the concept of material falsity why one idea is materially false and another is not, any more than we should expect the correspondence theory of truth to tell us which propositions correspond to reality and which do not.

This response to Wilson’s objection does not, however, get to the heart of the problem of material falsity. Even if Descartes believes that as a matter of fact ideas of sense are caused by bodies or modes of bodies and hence do have objective reality, he seems committed to the view that it is theoretically possible for a sensory idea to represent a non-thing as if it were a thing. And as consideration of Arnauld’s objection to the concept of material falsity reveals, there are good reasons for denying that there can be a discrepancy between what an idea seems to represent-that is, what it is as if of-and what it represents.

3. Arnauld’s Objection

Arnauld poses the following very powerful dilemma to Descartes. What does the idea of cold exhibit, if cold is a privation? If, on the one hand, the idea of cold exhibits a privation, then it is true. If, on the other hand, that idea exhibits a positive being, then it is not the idea of cold. And if that idea is not the idea of cold, then the idea itself has no falsity-rather, the falsity resides entirely in our judgment that the idea is the idea of cold (AT VII 207; CSM II 145-46).

i. Wilson’s interpretation of Arnauld’s objection

One interpretation of Arnauld’s objection suggested by Wilson is that he is arguing that the only coherent notion of representation is presentational. That is, we can interpret his assertion that if the idea of cold exhibits a positive being it is not the idea of cold as meaning that if the idea of cold presentationally represents a positive being, it does not represent cold (assuming that cold is a privation).

If this is what Arnauld has in mind, then his objection might seem to be easily overcome. We can defend Descartes by defending the plausibility of his other notion of representation, referential representation. So we might argue, for example, that it is entirely plausible to suppose that I can have an idea whose object is a horse, in the sense that it appears to be of a horse, but which in fact is caused by a cow, and thus might be said, in the other sense, to have the cow as its object. Such an idea, by presentationally representing a horse and referentially representing a cow, misrepresents the cow.

Although this strategy helps with the general problem of explaining how an idea which is caused by one thing might still exhibit something else and thereby misrepresent the first thing, it does not help with the specific problem of explaining how the idea of cold could misrepresent cold, supposing cold to be a privation. If cold is understood to be a non-thing and hence ineligible, on Descartes’s view, to be a cause, it is hard to see how an idea could referentially represent cold. Wilson concedes this, claiming that we need to suppose that Descartes’s notion of referential representation is neither causal nor demonstrative; but in the absence of a positive account of referential representation, it seems fair to say on Arnauld’s behalf that this strategy for defending Descartes is empty.

ii. The Aristotelian interpretation of Arnauld’s objection

There is a second, very different interpretation of what underlies Arnauld’s objection. On this second interpretation, Arnauld would grant that there is a distinction like that Wilson draws between presentational and referential representation, but his point would then be that an idea cannot referentially represent something without presentationally representing it., If our starting point is an Aristotelian theory of cognition, then a very interesting argument can be developed along these lines.

According to the Aristotelians, we have cognition of forms, and in order for a subject to have cognition of a form-be it sensory cognition of an accidental form or intellectual cognition of an essential form-that very form must be received in the cognizing subject. The reception of forms also underlies the Aristotelian account of becoming: something becomes brown in virtue of receiving the form of brown, or something becomes a horse in virtue of receiving the form of horse. But since no part of us becomes brown in virtue of our seeing something brown and since we do not become a horse when we come to have knowledge of a horse, Aristotelians distinguish two ways in which forms are received by a subject. Aquinas, for example, distinguishes natural or material reception of forms from spiritual or immaterial reception of forms. When a form is received naturally or materially by a subject, the form is predicated of that subject. When a form is received spiritually or immaterially, the subject has cognition of the form, but the form is not predicated of it. So one and the same form, say the form of a horse, can have two kinds of being: it exists naturally or materially in a horse, but it exists spiritually or immaterially in our soul when we have knowledge of a horse. Or, to take our other example, the accidental form brown exists naturally in a brown thing (say, a horse) but spiritually or immaterially in our eyes when we see a brown horse.

Underlying the Aristotelian view that cognition involves the reception of form are three fundamental theses. The first is that to have cognition of something is to become the same as it is. The second is that things are the same in virtue of sameness of form, even if they receive the form differently. The third is that non-accidental causation involves the production of a form in a patient by an agent which is the same in form as that produced.

Perception of the Aristotelian proper sensibles, such as color, sound, heat, and cold, will therefore involve the production of a form in a patient, the cognizing subject, by an agent which is the same in form. In order for that production of form to result in cognition and not in alteration, the form must come to exist in the cognizing subject spiritually or immaterially. But since the agent is the same in form as the cognizing subject, one can readily understand why the Aristotelians might have thought both that the proper sensibles are really existing external things and that we cannot be mistaken about them.

That Arnauld’s objection to Descartes’s notion of material falsity is based on an account of cognition that is fundamentally Aristotelian is revealed by his own restatement of that objection. He asserts that the idea of cold is coldness itself as it exists objectively in the intellect, so that if cold is a privation, it cannot be objectively in the intellect through an idea whose objective existence is a positive being (AT VII 206; CSM II 145). The crucial claim in this assertion is that the idea of cold is coldness itself as it exists objectively in the intellect. This is just the sort of claim that a seventeenth-century Aristotelian would make, given the terminological shift of referring to what Aquinas called spiritual or immaterial existence in the soul as objective existence.

I do not mean to imply that in making this claim Arnauld is accepting the three fundamental theses identified above as underlying the Aristotelian theory of cognition. But I do take Arnauld to be endorsing what I take to be the most fundamental element of the Aristotelian theory of cognition, namely, the doctrine that we have cognition of things in the world when they come to have another kind of existence in us. In other words, in asserting that the idea of cold must be coldness itself existing objectively in the intellect, Arnauld is committing himself to the view that what an idea referentially represents must itself exist in the intellect objectively. In order to clinch the conclusion that a sensory idea must presentationally represent what it referentially represents, he needs only to claim that an idea presentationally represents (it appears to be of) what exists objectively in the intellect. That is, a sensory idea cannot be caused by something other than what it presentationally represents because (i) it presentationally represents what exists objectively in the soul and (ii) what exists objectively in the soul is the same as what produces it.

It is important to note, however, that several prominent Aristotelians would have rejected this sort of argument precisely because they wanted to allow for the possibility of sensory misrepresentation. As David Clemenson has shown, there were important disputes among the scholastics over whether we could have sensations of non-existents. By the early seventeenth century it was commonly held by Jesuit commentators on De Anima not only that God could miraculously cause us to have direct cognition of non-existents but also that there could be natural causes of the direct cognition of non-existents. One of these commentators, Ruvio, has an especially interesting account of one such natural case, that of the rainbow, which I will discuss below, but first it will be useful to consider the views of Aquinas, who also thinks that the senses can be fallible with regard to the proper sensibles in certain rare instances.

Aquinas holds that the senses have cognition of things insofar as there is a likeness of the things in the senses, but he distinguishes three different ways a likeness of something can be in the senses: (i) as the likenesses of proper sensibles (color, sound, heat, and cold) are in the senses, which he characterizes as being first and essentially (primo et per se), (ii) as the likenesses of common sensibles (movement, rest, number, shape, and size) are in the senses, which he characterizes as being essentially but not first (per se sed non primo), and (iii) as the likeness of a human being is in sight, which is accidentally, because it is in sight not insofar as it is a human being but insofar as it is a colored thing that happens to be a human being. He thinks that even if the senses are functioning properly we can make false judgments when the likeness is in the senses in the second and third ways, because in those cases the sense is referred only indirectly to its object. But in the first case, that of the proper sensibles, we can have a false cognition only if the sense organ is defective and the sensible form is not received properly. So he explains that sweet things seem bitter to sick people because of corruption of the tongue.

For our purposes there are two important features of Aquinas’s account of this limited fallibility of the senses regarding the proper sensibles. First, it seems to be founded on the teleological assumption that a patient is naturally disposed to receive certain forms from an agent and that a patient which is not defective will receive those forms properly. Second, Aquinas has committed himself to the view that there can be a discrepancy between what is received immaterially in the senses and what we seem to be aware of. He thinks that when the sensible form of sweetness is not received properly by the senses, it can happen that we seem to be aware of bitterness.

Ruvio’s analysis of the rainbow is similar to Aquinas’s account of the sick person’s misperception of sweetness in that he also thinks a sensible form (which he refers to as the species) is received improperly and thus appears as something else, but he attributes this improper reception not to a defect in the sense organ but rather to the distance of the object from the sense organ. As he explains:

in the case of the rainbow the eye sees light that appears to it as red, or as some other color-not through a species of color, but through a species of light. Thus the eye sees by an act of vision which is identical to that by which the very same object would be seen if there were no deception. This is proved as follows. The eye sees the apparent color, which is really light, through species that are not species of color; therefore, these species must be species of light. This is evident, for inasmuch as the object is actually light, and [yet] color appears, the species through which it appears must be either of color or of light. The minor premise is proved [in two ways. First: the object] is not seen through a species of color, for the species is produced naturally by an object that really exists; and it is evident that a non-existent object cannot produce a species of itself. But there is no color present by which [a species] could be produced; therefore [the species] can only be produced by light, and it can represent only one thing: light. Secondly, it is proved by the fact that if the light which is seen in the rainbow under the appearance of color were close by, it would doubtless impress species of itself into the eye, through which it would be seen as light; therefore when it is far away it produces the same species in the same eye, however imperfect [those species may be], for every natural agent always produces an effect of the same nature, an effect that is more or less perfect according to whether the agent is closer or farther away. But light is a natural agent with respect to [its] species. Therefore the species it produces at a greater distance is of the same nature as the species it produces when close by, though [that species] is much more imperfect, and represents the light less distinctly-and that is why a vision of light is produced, although under the appearance of color.

Ruvio, like Aquinas, apparently holds that in most circumstances, when the sense organs are functioning properly and the objects are not too far away, the senses are infallible with regard to the proper sensibles. Like Aquinas, he also allows, however, that there can be a kind of falsity even with regard to the proper sensibles, because something received in the senses (light) can appear as other than it is (as color). Indeed, he says explicitly that the species of light both represents light and appears as color.

So Arnauld’s objection to Descartes-that an idea cannot represent something other than what it appears to be of-could equally well be directed against the views of Aquinas and Ruvio. Once again it might be construed in two very different and even opposed ways. First, it might be construed as relying on the view that there is no coherent notion of a sensible species’s (or an idea’s) representing an object besides its presenting that object. So if a species appears as color, that is what it represents, regardless of what that species is or what causes it.

Second, it might be construed as relying on the view that an idea presentationally represents what exists objectively in the soul. On this second interpretation, the crucial issue dividing Arnauld on the one hand and Aquinas and Ruvio on the other is whether something existing objectively in the soul can appear to be other than it is.

One might reasonably anticipate that in responding to Arnauld, Descartes will either (i) agree with Aquinas and Ruvio that something existing objectively in the soul can appear to be other than it is or (ii) reject the Aristotelian theory of cognition according to which things that exist objectively in the soul are the same as the things outside the soul that produce them, differing only in their mode of being.

4. Descartes’s Response to Arnauld

Descartes’s response to Arnauld’s objection is considered to be one of the most obscure and unsatisfactory passages in the entire Cartesian corpus. He agrees emphatically with Arnauld that if cold is a privation and the idea of cold exhibits a positive entity, the idea of cold is not the idea of cold (AT VII 234; CSM II 164).

Of this response Wilson says:

Although Descartes seems to give away the store here, I think he has merely expressed himself ineptly. He does not really intend to retract his position that a particular “positive” sensation counts as the “idea of cold,” even if cold is in fact a privation. Despite apparent verbal indications to the contrary, he is really continuing on his original track: the sensation of cold referentially represents cold (let’s suppose a privation)-but fails to present cold as it is (namely, as a privation). In the latter respect only it is not the idea of cold “but something else, which I wrongly take for this privation.”

I think, however, that Descartes means to say what he says. Under Arnauld’s acute questioning he does intend to retract his apparent position in the Third Meditation that the idea of cold would represent cold even if cold were a privation. He wants to say instead that if cold is a privation, the idea of cold is not the idea of cold. But isn’t that statement contradictory? Isn’t it impossible for the idea of cold not to be the idea of cold?

It is not impossible. Confusion arises because the expression “the idea of cold” makes it sound as if there is an idea that is identified by an objectcold-which it represents. But instead, I would claim, Descartes sometimes uses the expression “the idea of cold” merely as a kind of name for an idea. When he agrees with Arnauld that if cold is a privation, then the idea of cold is not the idea of cold, he is using the expression “the idea of cold” in two distinct senses. What he means is that a particular idea that is called the idea of cold is not of cold in the sense of having cold as its object.

An alternative way Descartes formulates his response is this: If cold is only a privation, the idea of cold is not coldness itself as it exists objectively in the intellect. Instead, the idea of cold is a sensation which has no existence outside the intellect (AT VII 233; CSM II 163). The surprise in his responding this way is that he does not take the opportunity to make a clean break with the Aristotelian account of cognition by denying that it is ever the case that something existing in the soul has another mode of being outside the soul. On the contrary, in his Replies to the First Objections he asserts that “the idea of the sun is the sun itself existing in the intellect-not of course formally, as it does in the heavens, but objectively, that is, in the way in which objects are wont to be in the intellect” (AT VII 102; CSM II 75).

His use of this sort of language, speaking of one and the same thing, the sun itself, having two ways of existing, depending upon whether it is in or out of the soul, implies an endorsement of the most fundamental element of the Aristotelian account of cognition. He is not entirely sure whether things exist objectively in the intellect or in its ideas (as indicated by his saying both-AT VII 233, CSM 163; AT VII 161, CSM II 113-14). But in either case his view is similar to the Aristotelian view in that he thinks cognition occurs when things that exist outside the soul come to have another kind of existence in us.

Thus on my reading of Descartes’s response to Arnauld, it is Descartes’s acceptance of the Aristotelian view that an idea is what it is of, but just in a different mode of being, that underlies his agreement with Arnauld that if the idea that we call the idea of cold is not coldness itself existing in the intellect, then it is not the idea of cold. But in making this concession to Arnauld, Descartes may seem to be conceding to Arnauld that the idea that we call the idea of cold is not a misrepresentation and thus that the falsity resides entirely in our judgment. After all, if the idea that we call the idea of cold does not represent cold, it cannot misrepresent it either.

I would argue, nevertheless, that Descartes has the better of the exchange. He has conceded to Arnauld that it is not because we have an idea of a nonthing that exhibits it as a positive thing that we would be led (if cold were a privation) to make the mistaken judgment that there is a positive being outside the mind that is producing the sensation. Nevertheless, he still can say, and does say, that it is because we have an idea that seems to be as if of a positive being-we have an idea that seems to us to be caused by some positive being existing outside us-that we would be led (if cold were a privation) to make the mistaken judgment that there is something outside the mind producing the sensation.

In other words, Descartes has conceded to Arnauld that if cold is a privation the idea of cold does not misrepresent cold. But it does not follow that the idea of cold does not misrepresent how things are external to us. The idea of cold seems to represent some positive being outside the mind, and therefore it misrepresents how things are if it does not.

This defense might still seem inadequate. In making his concession to Arnauld that if cold were a privation the idea that we call the idea of cold would not misrepresent cold because that idea would not be coldness itself existing in the intellect, Descartes has not retreated from the view expressed in the Third Meditation that such an idea would lack objective reality. Given his further assertion that “whatever we perceive as if in the objects of ideas, is in the ideas objectively” (AT VII 161; CSM II 113), one is led to wonder how an idea lacking objective reality could presentationally represent any thing. How could it be as if of a thing? If we could understand how an idea lacking objective reality might be as if of something, then perhaps we could see how that idea might misrepresent the way things are, even if it fails to misrepresent any particular thing in the world.

There is, moreover, an additional problem. If we are to make any sense of Descartes’s search for the efficient cause of the objective reality of an idea, it must be understood as the search for a causal explanation of the idea’s content.so Accordingly, the reason why an idea that had no objective reality would not need such an efficient cause is that there must be some sense of having content according to which an idea lacking objective reality would not have content. But if such an idea could still have content in the sense of being as if of something, why shouldn’t there be an efficient cause of its content in that sense? If we attribute to Descartes two distinct notions of content, why must there be an efficient cause of an idea’s having content in the sense of representing a thing referentially, but not necessarily of an idea’s having content in the sense of presentationally representing something?

Descartes’s response to Arnauld does suggest a way to reply to these deeper problems. He tells Arnauld that one should not ask, as Arnauld did, what the cause is of the positive objective being in virtue of which the idea is materially false because “I do not claim that an idea’s material falsity results from any positive being; it arises solely from the obscurity of the ideaalthough this does have something positive as its underlying subject, namely the actual sensation involved” (AT VII 234; CSM II 164).

I take this remark to suggest the following two possible responses to the problem of how an idea lacking objective reality can still be as if of something: (i) an idea can be as if of something in virtue of its material or formal reality, or (ii) an idea taken materially or formally can be as if of something in virtue of its obscurity. In either case-and it is not important for our purposes to decide between them-Descartes’s point is that an idea’s being as if of something is not conceptually or logically connected to its having objective reality. An idea taken materially or formally-in other words, an idea considered as an operation of the mind-can also be as if of something.

Two corresponding responses to the problem of finding an efficient cause for the content of ideas lacking objective reality are also suggested: (i) if the idea is taken to have content in virtue of its material or formal reality, then the efficient cause of the content of the idea is the efficient cause of the idea’s material or formal reality, or (ii) if the content of the idea is taken to arise from the idea’s obscurity, then that content need not have an efficient cause.

5. The Sixth Meditation: How Ideas Containing Objective Reality Might Be Misrepresentations

Given that sensations do contain objective reality, does Descartes think they can be misrepresentations? If so, how? Although we have just seen how an idea lacking objective reality might still be as if of something, Descartes seems to think that what we perceive as if in the object of an idea which has objective reality-and as a matter of fact all our ideas do have objective reality-cannot be something that is not contained objectively in the idea. Otherwise there would be no point in his asserting as he does that “whatever we perceive as if in the objects of ideas, is in the ideas objectively.” This assertion thus suggests that he agrees with Arnauld against Aquinas and Ruvio in maintaining that what our ideas seem to represent must exist in those ideas (or in the intellect) objectively. And if, as we have also just seen, Descartes accepts the Aristotelian theory of cognition according to which the objective reality of an idea just is the object of the idea in another mode of being, there would seem to be no possible room for ideas containing objective reality to be misrepresentations.

On the other hand, Descartes’s discussion of the possible causes of our sensations in the Sixth Meditation seems to imply that one can coherently suppose that even though sensations do contain objective reality they are misrepresentations; moreover, his embrace of the new physics seems to commit him to the view that as a matter of fact sensations do misrepresent the properties of bodies.

To see if there is a solution to this puzzle, we need first to examine some difficult concepts that play a prominent role in Descartes’s argument for the existence of bodies in the Sixth Meditation.

Descartes argues that God would be a deceiver if our sensations were caused by something in which their objective reality is contained eminently and not formally, as they would be, he tells us, if they were caused by some creature more noble than bodies. God would be a deceiver because we have a strong propensity to believe that our sensory ideas are caused by bodies and we have no faculty to discover that they are not (AT VII 79-80; CSM II 55).

What does he mean by speaking of the objective reality of an idea being contained formally or eminently in the thing producing that idea? The standard interpretation of that distinction is reflected by the Haldane and Ross translation of Descartes’s definitions of these terms in the Replies to the Second

Objections:

IV. To exist formally is the term applied where the same thing exists in the object of an idea in such a manner that the way in which it exists in the object is exactly like what we know of it when aware of it; it exists eminently when, though not indeed of identical quality, it is yet of such an amount as to be able to fulfil the function of an exact counterpart. (HR II 53)

According to this translation, it would be impossible for an idea whose objective reality is contained formally in the object producing it to be a misrepresentation, since the object is exactly as it is perceived by us.

The Cottingham, Stoothoff, and Murdoch translation is slightly different:

IV. Whatever exists in the objects of our ideas in a way which exactly corresponds to our perception of it is said to exist formally in those objects. Something is said to exist eminently in an object when, although it does not exactly correspond to our perception of it, its greatness is such that it can fill the role of that which does so correspond. (CSM II 114) One could go either way on whether exact correspondence amounts to exact resemblance, but the Haldane and Ross choice of “is like” is a better choice than the Cottingham, Stoothoff, Murdoch choice of “corresponds.” Thus it would appear from this crucial passage that an idea that contains objective reality can misrepresent something in the world only if what exists objectively in that idea is contained eminently in its cause.

Consider the Sixth Meditation counterfactual scenario in which God made the world in such a way that objects more noble than bodies cause our sensations (which Descartes would agree is a possible scenario on the further supposition that God also gave us some means of discovering the true causes of those sensations). In that case there would be a discrepancy between what our sensations seem to represent, namely bodies, and the objects causing them. But what would exist objectively in those sensations? Would it be the objects more noble than bodies causing the sensations that exist objectively in them, in which case the sensations would seem to represent things that do not exist in them objectively? Or would it be bodies that exist objectively in those sensations, in which case the objects causing the sensations would not themselves exist objectively in them?

If we understand an effect to be contained eminently in its cause when it has a nature different from and inferior to that of its cause, then there are grounds for supposing that it would not be the objects more noble than bodies that exist objectively in the understanding. Thus the discrepancy would arise because what exists objectively in the idea is a different object from the object causing the idea.

So it might seem as if there is an easy solution to the problem of explaining how our sensations might be misrepresentations even supposing that they do have objective reality. Descartes’s use of the notion of eminent containment seems to introduce as a theoretical possibility that an idea might represent something that is not contained in it objectively. In other words, his use of the notion of eminent containment might seem to amount to a rejection of the Aristotelian theory of cognition insofar as the Aristotelian theory requires as a necessary condition for cognition that the object of cognition must itself exist in the soul, albeit in a different mode of being. Indeed, in the definition of eminent containment quoted above from the Replies to the Second Objections, Descartes refers to the eminent container not merely, as he does in the Meditations, as the cause of the idea, but as the object of the idea.

But this solution is called into question by Descartes’s reply to Arnauld. Even if it is true that in the Replies to the Second Objections Descartes refers to the cause of an idea that contains only eminently what is in the idea objectively as the object of that idea, he can very plausibly be read as conceding in his subsequent reply to Arnauld that an idea represents something only if that very thing exists in the idea objectively. On this interpretation of his reply to Arnauld, it would follow that if our sensations that are as if of bodies and contain bodies objectively were in fact caused by things more noble than bodies, they would not represent those things. It might still be said that those sensations misrepresent the way the world is without misrepresenting anything in the world, because they represent the world as if bodies exist in it. But since they would not represent anything in the world, they could not misrepresent anything in the world.

A second shortcoming of this solution is that it does not apply to the scenario which is of primary importance. Our primary concern is not the counterfactual scenario in which our sensations are caused by something other than bodies, but the actual scenario in which bodies cause our sensations. That is, our primary concern is cases in which what exists in our ideas is contained formally in their cause. Can such ideas be misrepresentations? On the standard interpretation of formal containment, according to which there must be an exact resemblance between what exists in our ideas objectively and what is contained formally in their object, such ideas could not be misrepresentations.

But the standard interpretation should be rejected because it is based on mistranslations of Descartes’s definitions of those terms in the Replies to the Second Objections (AT VII 161):

IV. Eadem dicuntur esse formaliter in idearum objectis, quando talia sunt in ipsis qualia illa percipimus; & eminenter, quando non quidem talia sunt, sed tanta, ut talium vicem supplere possint. Cottingham, Stoothoff, and Murdoch, following Haldane and Ross, mistakenly add the crucial term “exactly,” even though there is nothing answering to it in the Latin. A more accurate translation implies, first, that in the case of formal containment there is some resemblance, but not necessarily an exact resemblance, between what we perceive as if in the objects of our ideas and what is in those objects, and second, that in the case of eminent containment there need not be any resemblance between what we perceive as if in the objects of our ideas and what is in those objects:

IV. Whatever things are in the ideas themselves objectively are said to exist formally in the objects of the ideas when they are in the objects such as we perceive them. They are said to exist eminently in the objects of ideas when they are not such, but are so great that they can fill the role of such.

This weaker reading is confirmed by the fact that in the Sixth Meditation argument for the existence of bodies Descartes asserts both that what exists objectively in our sensations exists formally in bodies and that bodies may not exist exactly as comprehended by the senses. If formal containment involved exact resemblance, Descartes would be contradicting himself in making these assertions.

Descartes therefore believes that there can be a discrepancy between what we perceive as if in the objects of our sensory ideas and what is contained in those objects, not only when what exists objectively in those ideas is contained eminently in the objects causing them, but also when it is contained formally in them, because, as Descartes explains, “this comprehension of our senses is in many things obscure and confused” (AT VII 80; CSM II 55). In the Third Meditation he suggests even more strongly that our sensation of the sun bears little resemblance to it (AT VII 39; CSM II 27).

Our sensory idea of the sun is an extremely important example. We have already seen that Descartes says in the Replies to the First Objections that our idea of the sun is the sun itself existing objectively in the intellect. And it is clear from the Sixth Meditation argument for the existence of bodies that what exists objectively in our idea of the sun exists formally in the sun. Thus it is very plausible to interpret Descartes as maintaining generally that what exists objectively in a sensory idea is contained formally in its object when that very object exists in the idea objectively. Moreover, since our sensory idea of the sun does not represent the sun as it is in the world, indeed, since it bears little resemblance to it, it follows that on Descartes’s view there need not be much resemblance between an object as it exists objectively and that same object as it exists formally. In this way, then, our sensory idea of the sun can be a misrepresentation of the sun. It is also materially false in the sense that it would lead us to make false judgments, because we are naturally inclined to judge that objects are exactly as we perceive them.

So I think that Descartes’s account of formal containment does explain how our ideas can misrepresent things in the world. But it is not so clear whether it can be employed to explain how our ideas of color, heat, cold and so on can misrepresent bodies or their properties. The problem is this: formal containment requires that there be at least some resemblance between the object of our idea and what is contained in the idea objectively, but given that Descartes seems to think that there is no resemblance between body and the ideas of color, heat, cold and so on, it would follow that what exists objectively in those ideas could not exist formally in bodies.

This problem suggests that there is some plausibility in attributing to Descartes a fine-grained account of formal and eminent containment: that some of the things in our sensory ideas of bodies are contained formally in bodies (for example, shape) and others are contained only eminently (for example, heat and color). I do not know of any clear evidence one way or another whether Descartes would endorse such a fine-grained account of formal and eminent containment. But there would be some obvious puzzle cases on such an account. Consider our idea of the straight stick as bent and our idea of the sun as small. Would Descartes want to say that what is contained in our idea of bentness is contained eminently in the stick or would he want to say that it is contained formally in the stick? Similarly, would he want to say that what is contained in our idea of smallness is contained eminently in the sun or would he want to say that it is contained formally in the sun?

If we reject the fine-grained interpretation of formal and eminent containment, then Descartes’s view would be that whatever exists objectively in our idea of body exists formally in its object. But then what should we say about color and heat? Since they have no resemblance to body or its modes, they could not be contained formally in bodies. Yet Descartes refers to them as being in our idea of body and he asserts, as already noted, that “whatever we perceive as if in the objects of ideas, is in the ideas objectively.” But perhaps Descartes does not mean this in the fine-grained sense. Perhaps he means only that if we firmly believe that a certain body is the object of a given idea and have no way of discovering that that belief is false, then that body exists in the idea objectively. Perhaps he wants to allow-and this would put him in agreement with the views of Aquinas and Ruvio-that an idea might be as if of some aspect of a thing that is not in the idea objectively. So, for example, perhaps Descartes would be willing to deny that colors exist objectively in our sensations and to assert instead that modes of bodies existing objectively in our sensations are sufficiently obscure that they appear as colors.

6. Reconciling the Meditations and Replies with the Principles

In constructing my interpretation of Descartes’s account of misrepresentation, I have focused on the Meditations and Descartes’s response to Arnauld. But Descartes also discusses misrepresentation in the Principles, and my interpretation may well seem inconsistent with these later remarks. There are two crucial passages. The most important is this:

In early childhood our mind was so tightly bound to the body that it had no leisure for any other thoughts, except only those by which it sensed what affected the body: and it did not yet refer these to anything located outside itself, but only sensed pain where something occurred harmful to the body; where something beneficial occurred, it felt pleasure; and where something affected the body without much harm or benefit, for the different parts in which and ways in which the body was affected, it had certain different sensations, namely those which we call the sensations of taste, odor, sound, heat, cold, light, color and the like, which represent nothing located outside thought. At the same time the mind also perceived magnitudes, figures, motions, and the like, which were exhibited not as sensations, but as certain things, or modes of things, existing outside the mind, or at least capable of existing, even though it did not yet recognize this difference among them. (AT VIIIA 35; CSM I 219, emphasis added)

On one very plausible reading of this passage, Descartes is asserting that as a matter of fact, and not merely as a possibility in principle, our sensations of cold and such do not represent anything outside thought. Thus this passage might seem to provide an important piece of evidence for Wilson’s interpretation of Descartes that I have wanted to reject-that Descartes thinks such sensations are not of things. On her interpretation, Descartes thinks that the idea of cold, for example, referentially represents a non-existent, cold. But I have argued that Descartes thinks that as a matter of fact sensations, such as the idea of cold, are caused by bodies or modes of bodies. And one might very plausibly infer from this that he thinks that as a matter of fact sensations referentially represent bodies. This inference would be justified if, for example, one attributes to Descartes the view that the idea of cold is a mode of extension existing in our mind so obscurely that it appears as cold.

Wilson herself at one point wanted to argue that this passage from the Principles constitutes a deep change in Descartes’s thinking from the Meditations, because she read it as implying that sensations are not even as if of (possible) things existing outside our thought. But in her recent paper she has claimed that there is not such a deep change in Descartes’s thinking. She now thinks he does not really mean to deny that sensations are as if of things existing outside our thought, and she cites as evidence for this the other passage. In this second passage Descartes says that sensations do represent things as if existing in bodies: “but if he examines what it might be, which this sensation of color or pain represents, as if existing in the colored body or painful part, he will notice that he is wholly ignorant of it” (AT VIIIA 33; CSM I 217). On Wilson’s current view, Descartes thinks sensations do presentationally represent something, it is just that they do not intelligibly present something.

So how should we read the first passage? When Descartes says that sensations represent nothing outside our thought, I take him to be talking not about what they represent referentially. Instead he is making a point about what sensations presentationally represent-that is, what they are as if of. But his point is not, as Wilson used to think it is, that sensations are not as if of anything. His point is that what sensations seem to represent (colors, sounds, and the like) are not things that exist outside thought. His point, in other words, is that although sensations are as if of things existing outside thought, what they are as if of does not exist outside thought. As Wilson puts the point, “we nevertheless tend to take the presentational content of sense experience to be something real, to refer it to external reality.” So Descartes does mean it when he says that sensations represent nothing outside our thought, but what he means is fully consistent with the views expressed in the Meditations and Objections and Replies. That is, Descartes can consistently maintain both that what sensations represent referentially (namely, bodies or their modes) exists outside thought and that some of what they represent presentationally (colors, heat and the like) does not exist outside our thought.

What about the second passage? In that passage Descartes says that we are wholly ignorant of what the sensation of color represents as if in the colored body. I have been assuming all along that Descartes thinks that the sensation of color represents color as if in the colored body. So for my interpretation to be consistent with the second passage his point would have to be that we are wholly ignorant of color.

In the final analysis, then, I think there is something importantly correct in Wilson’s assertion that Descartes thinks sensations fail to exhibit to us any possibly existent quality. What the sensation of red, for example, is as if of could not be a mode of extension and thus could not be a mode of body. But it does not follow from this, as Wilson has it, that he thinks such a sensation lacks objective reality and so is not of something. On the contrary, it is very plausible to interpret Descartes as holding that sensations are of bodies or modes of bodies. Still, we cannot, without knowing a great deal of physics, tell what bodies or modes of bodies they are of, since what they are as if of cannot exist in bodies.