1 Modern philosophy, and Descartes in particular, are often accused of having accentuated the danger of skepticism (the “loss of the world”) in seeking to prove the existence of the material world; and, to this end, of having assisted in ushering in ideas as a sort of intermediary entity forming a screen, or a veil, in between the mind and things. In his Essays on the Intellectual Powers of Man, Reid writes:

2

Modern philosophers […] have believed that external objects can”t be the immediate objects of our thought; that there must be some image of them in the mind itself, and that the external thing is seen in or by means of its mental image, like seeing something in or by means of a mirror. And the name “idea”, in its philosophical sense, is given to those internal and immediate objects of our thoughts. The external thing is the remote or mediate object; but the idea or image of that object in the mind is the immediate object, without which we could have no perception, no memory, no conception of the mediate object. [1]

3 This critique of modern philosophers is not new and remains current in our times. [2] For the question to what extent one can be a direct realist is still a live debate within philosophy today. In this gigantomachy, the name “Descartes” is used to reduce to ashes a sort of naive representationalism that we find translated in the contemporary vocabulary of cognitivist philosophy. [3]

4 But is Descartes really a partisan of this so fiercely decried representationalism? We certainly find literally in his philosophy the incriminating thesis: that we only know things through the intermediary of our ideas.

5

I am certain that I can have no knowledge of what is outside me except by means of the ideas I have within me; and so I take great care not to relate my judgments immediately to things, and not to attribute to things anything positive which I do not first perceive in the ideas of them. But I also think that whatever is to be found in these ideas is necessarily also in the things themselves. [4]

6 Although the subject can know things only “by means of” his ideas, Descartes does not consider the idea as a mere mental, private, object, since “to have an idea” is nothing other than to have “a perception that responds to the meaning of a word”. [5] He is far from identifying the idea with a copy, a similitude, of things outside us, in which things would be seen “as in a mirror”—ideas are tanquam rerum imagines, “like images of things”; his critique of scholastic species shows that he does not agree that resemblance explains the representative function of ideas. [6]

7 In Descartes”s vocabulary, “perception” designates, in general, the possession of an idea, the mental act or state of representing such and such a thing, and not necessarily what we now call “perception”—or more precisely, “sensible perception”. As we know, Descartes proposes a particular explanation for the latter in Dioptrics IV. After all, his description of the nature of ideas returns to the framework of the general theory of a perception “mediated” by ideas.

8 In the examination of the nature and the status of the Cartesian philosophy of ideas, Austin is useful because, although he takes up the critique of the theory of “indirect” perception, he begins by emphasising that the question of knowing whether we should be direct realists is badly posed:

9

I am not, then—and this is a point to be clear about from the beginning—going to maintain that we ought to be “realists”, to embrace, that is, the doctrine that we do perceive material things (or objects). This doctrine would be no less scholastic and erroneous than its antithesis. [7]

10 In Sense and Sensibilia, Austin gives an analysis of the famous illusion of the stick “bent” by its immersion in water, [8] which goes back prior to Descartes, to a fourteenth-century author, Pierre d”Auriole. [9] Austin demonstrates the ambiguity of expressions such as “looks like” or “seems to be”, as in when we say “the stick seems to be bent”. The argument from illusion would maintain that experience is representational and that, in perception, we have “a visual appearance” that is the immediate object of vision. In other words, it would judge that the stick is bent in so far as it is effectively represented as bent. Austin argues, on the contrary, that this is not a case of an illusion of the senses, but maintains that what we have an experience of, is a stick partly immersed in water, and which seems bent to us. Things look like what they are, no more and no less. [10]

11 The “argument from illusion”—the idea that our senses can be fooled—is thus a pretext for positing the existence of intermediary entities, or what he designates, in the context of analytic philosophy, with the term “sense-perceptions”:

12

These entities, which of course don”t really figure at all in the plain man”s language or among his beliefs, are brought in with the implication that whenever we “perceive” there is an intermediate entity always present and informing us about something else. [11]

13 Austin emphasises that we are not deceived by our senses, but neither do they inform us; sensory perceptions are not incorrigible: the way in which things appear to us is perpetually corrigible. Rereading the Cartesian texts on the basis of Austin”s analyses will allow me to show that treating perception as a mental act does not necessarily imply that representations must be considered as objects that are perceived. It is because we can change our way of seeing things, that we can say that we only know things through the intermediary of our ideas. If things look like what they are, as Austin thinks, it would be regrettable to judge them, as we often do, as mere appearances. Certainly, to see is above all to see the things of the world; but to say that our perceptions are that by which we think, or perceive, is simply to draw our attention to the fact that, in reflecting upon our thoughts, we can change our way of seeing, and that, in so far as philosophy is an activity which consists in making our thoughts clear, it constitutes a desirable change. [12]

Ordinary Language and Perception

14 Sense and Sensibilia critiques a doctrine of perception generally admitted in classical philosophy (by Descartes and Berkeley, for instance) and of which Austin stigmatizes certain contemporary advocates including Ayer, Price and Warnock, who, as he emphasises, are remarkable in furnishing an excellent exposition of its principal arguments:

15

The general doctrine, generally stated, goes like this: we never see or otherwise perceive (or “sense”), or anyhow we never directly perceive or sense, material objects (or material things), but only sense-data (or our own ideas, impressions, sensa, sense-perceptions, percepts, &c). [13]

16 Without using the later vocabulary of “direct” and “indirect” perception, which belongs to Russell, Descartes does indeed think that we know things “by means of” of our ideas, and that the immaterial spirit cannot be physically “touched” by that which we see, but only by ideas of which things are the cause:

17

For example, how is it we know that the sky exists? Because we see it? But this vision does not touch the mind except in so far as it is an idea—ista visio mentem non attingit, nisi quatenus idea est: an idea, I say, inhering in the mind itself, not an image depicted in the corporeal imagination. [14]

18 Perception, from the mechanical point of view, is but the communication of an impression, of a movement that touches the nerves and is transported to the brain. But a movement is not a perception—it may cause a cerebral image, but an idea is not an image. The idea is some content comprehensible to the mind; to its eyes, it is something meaningful.

19 Austin judges this way of conceiving of perception rather severely: it is an erudite theory, a scholastic view, founded on a usage of words that is distant from, or deviates from, ordinary usage. The arguments of the “general doctrine of perception” can be summed up as consisting in three procedures: the invention of an artificial philosophical terminology, the misappropriation of the ordinary use of certain words, and finally the construction of a supposed phenomenon of sensory illusion.

20 He gives an example of the first procedure with the expression “material things”, which is supposed to designate, in the philosophical language of perception, a general category: the “object” of sensible perception. Now, what we perceive is far more diverse—for example, even if it is really seen, the rainbow is not a “material thing”. The question “What sort of thing are we perceiving?” is badly posed since, despite what philosophers would have us believe, there does not exist just one sole “sort” of object for perception. The problem takes on a “metaphysical” allure when the sciences are able to give us the wherewithal to determine these different types of things that we can perceive.

21 The function of the expression “material things”—Descartes would prefer “corporeal things” (res corporeae) [15]—is to reinforce the distinction between appearance and reality, by way of the dichotomy material things/sense-data, so that the mental intermediaries (“ideas”, or “sensible impressions” in the language of modern philosophy) are inscribed in the register of real entities:

22

One of the most important points to grasp is that these two terms, “sense-data” and “material things”, live by taking in each other”s washing—what is spurious is not one term of the pair, but the antithesis itself. [16]

23 The expression “material thing” has no other function than to establish the usage of the term “sense-data”, and to ratify the general theory of perception: we perceive things through the intermediary of ideas, of sensible impressions, or of sense-data. The same goes for the expression “direct perception”:

24

[…] [I]t is essential to realize that here the notion of perceiving directly wears the trousers—”directly” takes whatever sense it has from the contrast with its opposite; while “indirectly” itself (a) has a use only in special cases, and also (b) has different uses in different cases […]. [17]

25 There are indeed cases in which we could say that we perceive indirectly: cases where an object interposes itself between us and that which we see; or where, for example, we change our location to see it in a mirror. But we do not know what “to see indirectly” in itself might mean. This is why the expression “we do not see things directly” seems more telling, whereas in reality it is the other expression that “wears the trousers”. The argument of such an “indirect” conception of perception thus rests upon this type of promotion of a philosophical vocabulary that is far distant from ordinary usage, as are also the expressions “sense-perception” or “deceived by the senses”.

26 Philosophers, secondly, misappropriate the ordinary meaning of words when they conflate them even though their usage shows that they function quite differently, and do not mean the same thing. [18] This deviation from ordinary usage makes them play a “metaphysical” role: as, for example, with the word “real”, which Austin analyses in chapter 7 of Sense and Sensibilia, but which was already the object of an interesting remark in “Other Minds”:

27

The doubt or question “But is it a real one?” must always (must have) a special basis, there must be some “reason for suggesting” that it isn”t real, in the sense of some specific way, or limited number of specific ways, in which it is suggested that this experience or item may be phoney. Sometimes (usually) the context makes it clear what the suggestion is: the goldfinch might be stuffed but there”s no suggestion that it”s a mirage, the oasis might be a mirage but there”s no suggestion it might be stuffed. If the context doesn”t make it clear, then I am entitled to ask “How do you mean? Do you mean it may be stuffed or what? What are you suggesting?” The wile of the metaphysician consists in asking “Is it a real table?” (a kind of object which has no obvious way of being phoney) and not specifying or limiting what may be wrong with it, so that I feel at a loss “how to prove” it is a real one. [19]

28 What characterises the metaphysical usage of “real” is the way in which the metaphysician supposes the existence of a unique meaning of the term (“the real world” or the domain of “material objects”), and it is this procedure that confers upon it a certain density and mystery. Now, the way in which the metaphysician poses the question is stripped of the “specific reason” that ordinarily accompanies the question “Is it real?”. For we only pose this question if there are specific reasons to doubt; we suspect, for example, that there is some trickery afoot (an automaton that perfectly imitates a goldfinch), or a mirage. If I have specific means to respond to anyone who asks me whether the goldfinch is real—by showing him, for example, that there is no machinery involved, or that the bird is not stuffed, I have, on the contrary, no way to show what the table is not, in order to prove to him that it is indeed real. I am lost with such questioning, because I do not have the specifications that would allow me to find the method of proof. Austin utilises a similar argument to that employed by Wittgenstein in On Certainty, [20] where he shows that our language games presuppose that it is not at all possible to doubt certain empirical propositions, propositions that however are not answerable to experience. Thus, if a philosopher intends to prove that one can doubt that this chair is real, according to Austin he comes up against the ordinary meaning of words, and is thrown against the barrier that he constitutes against scepticism:

29

But in fact the plain man would regard doubt in such a case, not as far-fetched or over-refined or somehow unpractical, but as plain nonsense; he would say, quite correctly, “Well, if that”s not seeing a real chair then I don”t know what is.” [21]

30 It is not possible to doubt reality in the absolute sense, without abandoning common sense, which is present in our usage of language, and falling victim to the philosophical manipulation of the ordinary terms of language. And, to tell the truth, what possible profit could one hope to draw from this general loss of confidence?

Austin and the Argument from Illusion

31 Philosophers are the dupes of the expressions and metaphors they create: “deceived by the senses”, for example, is a common metaphor, but in reality our senses tell us nothing, either true or false; our senses “are dumb”.

32

In fact, of course, our senses are dumb—though Descartes and others speak of “the testimony of the senses”, our senses do not tell us anything, true or false. [22]

33 There is no intrinsic correctness to sense-data, no incorrigibility that can be accorded to them a priori and that would make of them “sensory information”. If Locke speaks of sensible knowledge as a degree of certitude to accord to the “evidence of our senses”, [23] Descartes, for his part, does not use this expression, preferring, in the Meditations, the term “teachings of nature”, where nature is present firstly as a poor schoolmistress:

34

When I say that I have been so taught by nature, all I have in mind is that I am driven by a spontaneous impulse to believe this…. [24]

35 So that, although he speaks of the teachings of nature, Descartes is perhaps not to be situated at the origin of the characterisation of perception in terms of “informational content”, conceptual or no conceptual, for he does not seek to rule, from the outset, on the intrinsic correctness or incorrectness of experience. Before concluding that something is taught to me by the senses, I must follow the entire course of the Meditations, which keep in question, until the very last moment, the teachings of nature. Austin”s critique of the phrase “deceived by the senses” might have been rather striking to Descartes, since “talk of deception only makes sense against a background of general non-deception”. [25] But it is also what is proved by Descartes, who would have admitted the justice of this remark, for he unveils the fact that the incorrigibility of the videoed videor, of the Cogito, in the sense of the validity of its form, or of the act “it seems to me that…” independent of its content, is necessarily the ground of the “non-deception” of the generalised deception of the Evil Demon or of the hypothesis of an omnipotent God. But it is true (we shall come back to this) that Austin does not admit intrinsically incorrigible data, and that consciousness does not seem to benefit from any privilege in this regard. [26]

36 Let us examine the role that Austin has the argument from illusion play in his critique of modern philosophers. For him, the aim of this argument is to lead us to think that illusions are deceptive. That is, according to the general doctrine of perception, illusion and deceptive illusion—delusion—are the same thing, when this is not so at all. Austin lays out a list of authentic examples of illusions—optical illusions (Müller-Lyer), theatrical illusions. But deceptive illusions are something else entirely: for example, a delusion of persecution, hallucinatory delusions of grandeur—in short, serious behavioral disorders. And these deceptive illusions have nothing to do with perception. We must distinguish, as the plain man would, cases where the sense organs are disturbed, those where the conditions of perception are abnormal, and finally those where an invalid inference has been drawn. The misunderstanding of the distinction between illusion and deceptive illusion bespeaks the belief that, in illusion, there is truly something that is seen, a non-existing object that we perceive. In making this distinction between illusion and delusion, Austin deprives the “indirect” conception of perception of its principal support that allowed it to posit an entity serving, in Putnam”s phrase, as “the highest common factor” between correct and incorrect perception:

37

The most important differences here are that the term “an illusion” (in a perceptual context) does not suggest that something totally unreal is conjured up—on the contrary, there just is the arrangement of lines and arrows on the page, the woman on the stage with her head in a black bag, the rotating wheels; whereas the term “delusion” does suggest something totally unreal, not really there at all. [27]

38 The Austinian analysis of illusion shows us that to see is to see the things of the world, and that, even in the case of illusion, we must recuse the notion that we see some appearance, and maintain instead that we see the things themselves:

39

Looking at the Müller-Lyer diagram (in which, of two lines of equal length, one looks longer than the other), or at a distant village on a very clear day across a valley, is a very different kettle of fish from seeing a ghost or from having DTs and seeing pink rats. And when the plain man sees on the stage the Headless Woman, what he sees (and this is what he sees, whether he knows it or not) is not something “unreal” or “immaterial”, but a woman against a dark background with her head in a black bag. If the trick is well done, he doesn”t (because it”s deliberately made very difficult for him) properly size up what he sees, or see what it is; but to say this is far from concluding that he sees something else. [28]

40 Austin”s position is extremely radical here. [29] And he analyses the “supposed” illusion of the partly immersed stick with the same radicality. [30] He shows that Ayer makes it a case of deceptive illusion, whereas the stick does not exactly resemble a bent stick (outside of water), but rather a bent stick that is partly immersed in water. We cannot prevent ourselves seeing it as immersed in water, and that is what it is. And given this, it is hard to see what is deceptive about it. We may perhaps be disposed to admit that the stick seems to be bent, but then we see that it is partly immersed, so that it presents itself exactly as we would expect it to. This is what Descartes says:

41

Hence, when it is asserted that a stick in water appears broken on account of refraction, this is the same as saying that how it appears to us is the basis on which a child would judge it to be broken and even the basis on which we make the same judgment. [31]

42 We deceive ourselves as a child might: we form a false judgment. It is not vision that is incorrect, or which fools us—it is we who fool ourselves. In this sense, for Descartes also, the senses are dumb. What interests him above all is to show that judgment, which is the third degree of sense-certainty, is a judgment of the understanding, not an empirical correction of what is seen through touch. Descartes does not have visual appearances intervene; he says that things appear to us thus and that we may deceive ourselves. He speaks several lines later of “errors of sight”, not to indicate the presence of a false appearance, but to emphasise that it is not touch that corrects vision, but understanding, because it is the understanding that “teaches us that in this matter we ought to give more credence to a judgment based on touch than to a judgment elicited from sight.” [32]

43 If it is possible to say that the senses deceive us, as Descartes does in the First Meditation, it is because they are, in the world constituted by sensible things, the proper way in which our understanding is ordinarily exerted. And it is the understanding, and it alone, that commits mistakes, and corrects them. It is thus not a question, in sensible perception, of traversing visual, tactile, etc. appearances in the direction of the thing, but of conceptual capacities exercised at the heart of receptivity, and which guarantee that the activity of representation does refer to an object in the world. There is, doubtless, an opacity of representation, which is a thought, with its own shadow, because what can often be an obstacle to comprehension are old representations. How can one avoid this, without dogmatism?

44 According to Descartes, we cannot say that it is our eyes that see that it is the same wax that remains, nor touch, nor any other organ; in other words, for him the senses are not the subject of perception.

45

But what is this piece of wax which is perceived only by the mind? Surely it is the same piece of wax that I see, touch, and imagine; in short it is the same piece of wax I took it to be from the very beginning. But I need to realize that the perception [perceptio] of the wax is neither a seeing, nor a touching, nor an imagining. Nor has it ever been, even though it previously seemed so; rather it is an inspection on the part of the mind alone [mentis inspectio]. This inspection can be imperfect and confused, as it was before, or clear and distinct, as it is now, depending on how closely I pay attention to the things in which the piece of wax consists. [33]

46 The Cartesian idea of a perception of bodies by the mind, or the understanding, does not at all signify that it is not the bodies themselves that are perceived. There is no reason to think that the understanding splits the real into a sensible thing (the real wax), seen through the intermediary of another, intelligible or immaterial(?), wax. No more reason than there is to consider that the perception of the understanding is an intelligible vision opposed to a vulgar vision of the eyes(?). To say that perception is a judgment, or even an argument, is to speak of an implicit knowledge, of an actualised potential synonymous with apperception, recognition or indexicality, and not of an inference of the presence of the thing on the basis of sensible qualities, or other supposed signs of this presence.

47

For we say that we see the wax itself [ipsammet], if it is present, and not that we judge it to be present from its color or shape. Whence I might conclude straightaway that I know the wax through the vision had by the eye, and not through an inspection on the part of the mind alone. [34]

48 What can we draw from this fragment of ordinary language analysis? It is customary to say that we see the wax itself; the perception of the understanding is the ordinary vision of a human being, and this not because seeing is nothing but judging, but because there is no pure vision, no vision without intelligence as to what one sees, no perception without comprehension, except in the case of a vision that takes place without any consciousness of seeing—that is to say, without a human mind. [35] To see, for man, is to see and understand what is seen, and this understanding envelops an activity of representation, as for example that of seeing the identity of the thing [36] or recognising it, which does not appear to be reducible to a simple linguistic facility, and implies a psychological dimension. The ostensive gaze thus designates the thing, refers back to the identity of the thing. A thesis of this type is to be distinguished from a different thesis: that to perceive, the senses form an intermediary entity—”such and such things” that are seen—a thesis that is, of course, open to criticism since it reifies an obscure and useless mental entity.

49 There are two different problems in relation to the thing: what we could call the “skeptical problem”, the problem of the thing”s existence; and the epistemological problem of its identity, of its recognition by the mind—that is to say, of the causality that comes into play in the sensible perception of that thing. Austin seems to me to put an end to the first problem, but there remains, on the other hand, the problem of the identity of the thing, and of the causality in play in the formation of the idea of this thing, which is not at all governed by the question of specific criteria, or recognised methods of recognition as outlined in “Other Minds”. There are, doubtless, criteria to identify this as a real and not a stuffed goldfinch. But what is it that makes this piece of wax the same piece of wax? What is it that justifies, from an epistemological point of view, the judgment one ordinarily makes: it is the same piece of wax that remains after this change? Such reference requires different faculties of linguistic performance. The analysis of ordinary language permits us to resolve a certain number of questions concerning perception, but does it permit us to resolve them all? Not necessarily; and it is probable that, armed with this demystificatory perspective, we may also need to vary our point of view, and to embrace certain scientific and cognitivist perspectives.

Idea and Object

50 But for all this, does consciousness transform the idea into an object? Descartes emphasises, in the Preface to the Meditations, the ambiguity of the term “idea”, which can mean the act, which he calls the “formal reality”, of the idea, or else the content, its objective reality. Here, as we know, he is indebted to Suárez, in his adaptation of the vocabulary of esse objectivum, [37] derived from Duns Scotus and Aurioli. How can Descartes avoid attributing to objective reality, precisely, a “reality”? In Des vraies et des fausses idées, Arnauld wished to lay out the Cartesian theory of ideas emphasising, against Malebranche, that ideas are not “representative beings, distinct from perceptions”—in other words, that it is not necessary, in order for an external object to be perceived, that a non-physical intermediary representing that object be present to the mind and directly and immediately perceived by it. Explaining what he understands by “to represent”, namely, “to be objectively in the mind”, he maintains that ideas are not the mental objects of perception, but the acts of perception directed immediately toward external objects, for the act of perceiving supports a double relation:

51

I have said that I took for the same thing perception and Idea. It must however be remarked that this thing, albeit unique, has two relations: one to the mind which it modifies, the other to the perceived thing, in so far as it is objectively in the mind; and that the word perception marks more directly the first relation, that of idea the second. Thus the perception of a square marks my mind more directly as perceiving a square: and the idea of a square marks more directly the square in so far as it is objectively in my mind. […] They are not at all two different entities, but one and the same modification of our mind, which comprises essentially these two relations. [38]

52 Mental perception is, in the ordinary sense of the term, a direct and immanent perception. One might nevertheless say of ideas, qua perceptions, that they are the means of perceiving external objects, [39] in the sense that to perceive one must perceive, but not in the sense that there would exist an intermediary milieu between our perceptions and the object. For Arnauld, the ideas are not that which we perceive, but the perceptions through which we perceive physical things. Arnauld takes up Thomas Aquinas”s vocabulary and applies it to ideas: they are id quo intelligitur. [40]

53 In chapter 6, he seeks to reconcile the Cartesian theory of ideas with those “façons de parler” that claim that we do not see things immediately. He examines reflection and distinguishes between direct and reflective perception. Does reflective perception make an object of the idea?

54

I do not at all reject these façons de parler. I believe they are true when properly understood. […]
Our thought or perception is essentially reflective upon itself: or, more elegantly in Latin, est sui conscia. For I cannot think without knowing that I think. I cannot know a square without knowing that I know it. I do not see the sun, or, to put aside all doubt, I do not imagine that I see the sun, without being certain that I imagine that I see it […] Apart from this reflection which one might call virtual, […] there is express reflection, through which we examine our perception by means of another perception […]. Every perception being essentially representative of something, and according to this being called idea, it cannot be essentially reflective of itself, and its immediate object is not this idea, i.e. the objective reality of the thing that my mind is said to perceive. For example, if I think of the sun, the objective reality of the sun, which is present to my mind is the immediate object of this perception; and the possible or existing sun, which is outside of my mind, is its mediate object, so to speak. [41]

55 A careless reading of this text might consist in making of virtual reflection a support for the theory of indirect realism. Now this immediate consciousness is that which accompanies every act of perception as a means of perceiving the external object. In other words, direct perception, the consciousness of perceiving, or again virtual reflection, does not transform the idea into an immediate object of thought. To perceive, is to perceive external things. However—and this is a matter of the determination of the meaning of “seeing”—in a certain sense of “seeing”, when I reflect upon my impressions, or on ideas that I have had previously, the idea can become the immediate object of thought (id quod intelligitur). This does not mean to say that it always is so.

56 Our capacity to reflect upon our representations is one of the forces of the cogitatio, for it is our capacity to change our impressions, to understand certain older representations that constitute a screen in the relation we entertain with things and beings, and in general with the world. In “Other Minds”, Austin compares “I know” with “I promise”, arguing that “If I know, I cannot be deceived”, understanding this to be a grammatical rather than empirical necessity: one cannot say “I know that it is thus, but I may be wrong”. Thus “I know” is not descriptive—It does not describe a state of mind, nor a quality of my cognition. It pertains to my speech: I give my speech to another, I authorise him to say “P is Q”. In a certain sense, “I know” has a performative character.

57 Perhaps it is possible to understand the act of cogitatio in the Cartesian theory of ideas in this way, as an activity, rather than envisaging it uniquely as the motif of the incorrigibility of “It seems to me…”. For Wittgenstein, it is clear that one cannot teach a child to speak the language of things (“This is red”) by teaching her the language of the mode of appearance (“This seems red”). [42] Thus, “It seems…” or “It seems to me…” are expressions learned after we have learnt to describe the objects of the world. They manifest a certain suspicion: I say “This seems red” because I know from experience that under this light an object may appear red when in fact it is mauve. I have already experienced being deceived on this subject, and so I say “It seems to me that…”. But there is, doubtless, something bizarre afoot with something that testifies, as Brandom says, [43] to a “retreat from engagement”, the incorrigible foundation of every enunciation: even if “This is red” is false, nevertheless, “It seems that this is red” is necessarily true.

58 The conclusion to which these analyses lead me, is that to treat thought, or perception, as a mental act, does not necessarily imply that we need make of representations objects that are really perceived. In the case of explicit reflection, my attention may be focussed on an idea that I examine, that becomes an object of thought. But as Arnauld says: “Considered carefully, these words [represent, representative, representation] belong properly and primarily only to the perceptions of the mind, which are the formal representations of their objects, and it is only in relation to our perceptions that other things, like tables, images, words, the letters used in writing, are said to represent, or are called representative.” [44] To see, or to think, does indeed consist in seeing or in thinking the things of the world, but a subjective meaning is grafted onto this objective signification of the word “see” or “think”. Of course what is given in perception, as in thought, are things, or the world, and not the mere mental “sssmatter” of ideas. Moreover, that consciousness must always be able to accompany our representations does not imply, either, that it should be an object, for its only reality lies in its being a form or an act. Ideas are true objects only in so far as we express them, through the work of expression, which is the activity through which we render our thoughts clear.