Kant was very interested in a bizarre author called Swedenborg, and Swedenborg had a certain conception not only of spirits, in the spiritualist sense, but he had a conception of space and time as a function of spiritualism. To answer your question: it seems to me that you aren't posing the problem in Kantian terms. When you say, for example: "I'm thinking of someone", and then this someone comes into the room. You are using "thinking" in an extremely general sense, that is, any activity of any given faculty referable to a so-called thinking subject, whatever the mode of thought. When you say that I am thinking of someone that means that I am imagining someone, or I remember someone, and then by chance, by coincidence, this someone comes into the room. "Thinking" can very well be used in vague and general terms. At the point we are at in our analysis, Kant has substituted a restricted use, in which to think does not mean to imagine or to remember, or to conceive, but in which thinking means solely to produce concepts. To feel means solely: to receive a sensible diversity, to apprehend a sensible diversity. To imagine means: either to produce images, or else to produce the concept's corresponding spatio-temporal determinations.
So grant me that, at the level that we are on, whatever these restricted definitions and their value are, to think, to imagine, to feel, are not treated by Kant as modes of a same type of thought which could be substituted for one another, but as specific faculties. So that when you say "I remember someone", and this someone comes in, there is no activity of thought, there is an act of imagination, there is suddenly the sensible diversity which gives me this someone. That's what Kant would say.
Kant says, in a text of the Critique of Pure Reason: "if cinnabar was sometimes red, sometimes red and sometimes black, sometimes heavy and sometimes light... I would never have the opportunity to associate - i.e. my imagination would never have the occasion to associate - the heavy cinnabar with the colour red..." If nature was not subject to concrete rules, there would be no associations of ideas. In other words, when I have an association of ideas, this implies that things, and no longer ideas, that things are themselves subject to rules analogous to the rules which are associated in us. Which is to say if Pierre did not come to Vincennes, or had not come to Vincennes, I would never have had the opportunity to associate the idea of Vincennes and the idea of Pierre.
I will try to clarify this story of faculties, but you can well see that you can't invoke the example that you just gave as transforming the problem of the thought-imaginary relationship, because in fact it would be a matter of one of the forms of thought. When I think "of Pierre" and then Pierre is there, in fact I haven't thought anything since I haven't formed any concept at all. I imagined or remembered.
There's something very, very curious in Kant. When Kant writes his three great critiques, the Critique of Pure Reason is in 1781, Kant is 57 years old, the Critique of Practical Reason is in 1788, and finally the last very great work by Kant is the Critique of Judgement in 1799, he is 76 years old. I was saying to myself that there aren't that many precocious philosophers. If he had died at the age of 50 he would be a sort of secondary philosopher, a good disciple of Leibniz, a good run-of-the-mill philosopher. There is only one case, the extraordinary case of Hume. With him, he has his whole system, all his concepts, at the age of 22-25, after which he only repeats, improves.
Today, I would like to speak about this extraordinary book that is the Critique of Judgement; if I say that it is an extraordinary book it's because it is a book which founds a discipline, even if the word existed before. There is a particular discipline which will be radically founded by the Critique of Judgement, namely the foundation of all possible aesthetics. Aesthetics came into existence as something different from the history of art with the Critique of Judgement. It's really a very difficult book, don't try to understand each line of it, follow the rhythm. I would like to develop a bit the difference between the Cartesian "I think", such as it appears in Descartes, and the "I think" such as it appears in Kant. We must schematise at the level of a certain labor of thought. Already with Descartes, something appears which, it is said, will be of very great importance in the evolution of philosophy, namely: substance, that certain substances are therein determined as subjects. We can say very schematically that these formulae have been helpful. Not all substances, but a type of substance called thinking substance (?). Thinking substance is determined as subject. It's the discovery which will mark all philosophy said to be modern, from the 17th century onwards, it is the discovery of subjectivity. Why the discovery of subjectivity, why would subjectivity have to be discovered? It's the discovery of a subjectivity which is not the subjectivity of the empirical self, namely you and me. From the point of view of the labor of the concept, if I say: the Cartesian cogito is the assignation of substance as subject: "I think", the Kantian I think is very different. Everything happens as if a further step was taken, namely that the form of subjectivity breaks away from substance. The subject is no longer determinable as a substance. Subjectivity liberates itself from substantiality. Philosophers do not contradict each other, it's like with scholars, there is a whole labor of the concept. I will try to express Descartes' "I think" very concretely. Descartes' point of departure is a famous operation called doubt. He says, in some very beautiful texts, "perhaps this table on which I rap does not exist", and "perhaps my hand which raps on this table does not exist"; everyone knows very well that this is a manner of speaking. There is necessarily a discrepancy between the style and the content. It's not a matter of saying the table doesn't exist. Descartes' problem is something else entirely, it's the ground [fondement] of certainty, which is to say a certainty which would be exempted from all possible doubt. If I say "the table exists", its existence is of no matter to me, I am wondering whether it is a certainty which contains in itself its own ground. No. Certainly the table exists, it's understood, but this certainty does not contain in itself its own ground. Are there certainties which contain their own ground in themselves? At this point I move up a level: we say that we are sure that two and two make four; Dostoyevsky's heroes say: "I don't want two and two to make four". Can one not want two and two to make four? And when he says: I am certain that two and two make four, is that also a certainty which has its own ground in itself? Why would two and two make four? In this case one can demonstrate that two and two make four, which is complicated. On the other hand Descartes thinks that it is the operation of doubt which will give us a certainty which contains in itself its own ground. Namely that there is one thing which I cannot doubt, I can doubt the existence of the table, I can doubt the proposition "two and two make four", I cannot doubt one thing, which is that in so far as I doubt, I think. In other words, the operation of doubt, in so far as doubting is thinking, will provide me with a certainty which contains in itself its own ground: I think! "I think" - it's a funny sort of formula. In certain texts Descartes goes so far as to say that it is a new mode of definition. It's a definition of man. Why is it a definition of man? Before Descartes philosophy proceeded by definitions, scholasticism, definitions were given above all through generic and specific differences. Man is a rational animal. Animal is the genus, rational is the specific difference. Descartes says that when a definition of this type is given we are always referred to something else that we are supposed to know. In order to understand that man is a rational animal, we are supposed to know what an animal is, we must know what rational is. He will substitute a definition of another form entirely: I think. It's very curious, this "I think", because there is no need to know what thinking is. It is given in the act of thinking. There is a kind of implication, which is not at all an explicit relation between concepts, it's an act which is one with the act of thinking.
With doubt, when I doubt, there is one thing which I cannot doubt, which is that as a self who doubts, I think. Self, what is the self? Is it my body, is it not my body? I have no idea since I can doubt my body. The only thing I cannot doubt is that since I doubt, I think. You can see that it is absolutely not a matter of an operation in which doubt would come to bear on ?????, but of an operation which consists in requiring a certainty which contains in itself its own ground as certainty. "I think" is thus an act through which I determine my certainty. The "I think" is a determination. It's an active determination. Not only can I not doubt my thought, but I cannot think without it, which is to say that the same implicit relation which goes from doubting to thinking, goes from thinking to being. In the same way that doubting is thinking, in order to think one must be. You can see the progression of the Cartesian formulae: I doubt, I think, I am. I doubt, I think, I am, I think is the determination, I am is the indeterminate existence, I am what? Well, the determination will determine the indeterminate existence. That the determination determines the indeterminate means: I am a thing that thinks. I am a thinking thing.
Thus it is that what I am is determined by the determination "I think", is determined as the existence of a thinking thing. Descartes is told that that's all very well, but what proves to us that it is not the body which thinks in us? A materialist of the time says this to him. And Descartes replies - as soon as anyone makes an objection to him, he is very rude - he says: you haven't understood anything, I never claimed that it is not the body which thinks in us, he says exactly this: what I am claiming is that the knowledge which I have of my thought cannot depend on things which are not yet known. In other words, it is not a matter of knowing if it is the body or not the body which thinks in us, it is a matter of observing that, within the perspective of the Cartesian method, the consciousness which I have of my thought cannot depend on things which are not yet known, namely the body since doubt [also bears on this?]. Thus this procedure, from a logical point of view, but a new type of logic since it is no longer a logic that operates through genera or differences, it's a logic of implications since Descartes is in the process of... in opposition to classical logic which was a logic of explicit relations between concepts. He launches a new type of logic which is a logic of implicit relations, a logic of implication.
So, he has determined with the "I think", which is a determination, he has determined the existence of what thinks, and the existence of what thinks is determined as the existence of the thinking thing. He thus goes from the determination to the indeterminate, from the determination "I think" to the indeterminate "I am" and to the determined: I am a thing that thinks. He threads along his logic of implications: I doubt, I think, I am, I am a thing that thinks. He has thus discovered the zone where substance was subject.
And Kant appears.
What Descartes affirms is that the soul and the body are really distinct. It's more than an ontological separation. But what is it that he calls a real distinction, in conformity with the whole tradition? Again, words here are as defined as in science. A real distinction is not the distinction between two things, it's the distinction, a mode of distinction, between two things, it's the distinction, a mode of distinction, between two ideas and representations : two things are said to be really distinct when I can form the idea of one of them, which is to say when I can represent to myself the idea of one of them without introducing anything about the other. Representations thus form the criteria for real distinction. Two things being completely distinct is a proposition which, ultimately, has no meaning. We will get to the level of substance, Comptesse, you who know Descartes as well as I, after the fifth meditation. In the second meditation, there is absolutely no way of knowing if it is the body which thinks in me. Descartes says it categorically. The soul and the body, thought and extension are really distinguished - which is not the same thing as really distinct - as two ontologically separate, or separable, substances. He is not able to say this before the end of the meditations. In the second meditation, when he discovers the "cogito", the "I think", he absolutely cannot say it yet, and it's for this reason that among the novelties of Descartes' text, there is something which he very much insists on, and this is the true novelty of the meditations, even if you don't like Descartes very much, namely that it is the first book which introduces time into philosophical discourse.
There is something tremendous in this. What he says in the second meditation, then what he says in the fifth, there is a temporality which has unfolded which meant that he could not say in the second what he will say in the fifth.
This is not true of all philosophies; if I take Aristotle or Plato, there is a succession in the reading, but this succession corresponds to a chronological order and that's all. In Descartes there is the establishment of a temporal order which is constitutive of the metaphysical dimension.
Broadly speaking, during the whole of the middle ages, there was a theory of forms of distinction, each author will create his own forms of distinction, but broadly there were three major types of distinction: real distinction, modal distinction and the distinction of reason. And if you relate these three types of distinction to things themselves, you produce an absurdity, if you give them an ontological bearing, they don't have an ontological bearing yet, they only have a representative bearing, namely: there is a real distinction between A and B when I can think A without thinking B, and B without thinking A. You can see that it is a matter of a criterion of thought, a criterion of representation. For example: two things are really distinct, and not truly distinguished, two things are really distinct when you can form the representation of one without introducing anything of the other, and reciprocally. This lighter is on this book, are they really distinct? Yes, I can represent the lighter to myself without introducing anything of the representation of the book, they are really distinct. It's possible that they are also truly distinguished, it would be enough for me to put the lighter in my pocket. Between the front and back of a piece of paper, there is a real distinction, I can represent to myself one side of the paper without having the least representation of the other. In things, front and back are not separate, but in my representation front and back correspond to two representations. I would say that there is a real distinction between the front and back of the paper. So there can be a real distinction between two things which are not truly distinguished.
Second type of distinction: modal distinction. There is a modal distinction when I can think A, I can represent A to myself without B, but I can't represent to myself B alone. For example: extension and the figure. Let's suppose, broadly, that I can represent to myself extension without figure, I cannot represent to myself a figure without extension. I would say that between extension and figure there is a modal distinction. In relation to this, we must not transport it to the level of ontology too quickly, it does not mean at all that there is an extension without figure in things, perhaps there isn't. You can see it's the same gesture, it's the criteria of representation.
Third distinction: the distinction of reason. When I represent to myself as two, two things which are one in the representation. In other words, the distinction of reason is abstraction. When I distinguish the front and back of the piece of paper, I do not make an abstraction since they are given as two in my representation, since there are two representations, but when you speak of a length without breadth, however small this length, there you make an abstraction. When you can have no possible representation of a length which would have no breadth, however small. Thus between length and breadth there is a distinction of reason.
The way people talk about abstraction is amazing, they have absolutely no idea what it is. Philosophy has a kind of technique and a terminology like mathematics. Generally the word abstract is used for things in which there is no abstraction. The problem of abstraction is how can I make two things out of what only exists as one in my representation. It's not difficult to make a thing into two when I have two representations, but when I say the back of the piece of paper, I am not abstracting at all since the back is given to me in a representation which itself exists. When I say a length without thickness, there I am abstracting because I am separating two things which are necessarily given in each other in my representation.
There is indeed a philosopher who started the theory of distinctions. And then the theologians of the middle ages were not guys concerned with God, that's like saying that the painters of the Renaissance were guys who thought about God, no, they thought about colours, they thought about lines, and they draw out the most bizarre things from Christ's body. What we call theologians are people who are in the process of inventing a logic, a physics, a dynamics, and one of the great things in the theology of the middle ages is the theory of distinctions... ok... up to this point it's completely independent of the question of knowing if things are truly distinguished or confused in themselves, so that in the whole story of the cogito, I doubt, I think, I am, I am a thing that thinks, Descartes can only conclude: the representation that I have of my thought, and the representation that I have of an extended body, are such that I can represent my thought to myself without representing anything to myself of extension; I can represent to myself an extension without representing anything to myself of my thought. This is enough for Descartes to say that thought and extension are really distinct. He cannot add yet that it is not the body which thinks in me...

[interruption of the tape]

So he will have to, in order to draw from the real distinction between representation-substance the ontological separation between substances, he will have to go through a whole analysis of the concept of God in which he says: if the real distinction between representation and substance was such that there was no corresponding true separation in things, an ontological separation in things, then God would be deceitful, God would be lying to us since the world would be double, God would be duplicitous, God would be full of duplicity since he would have made two non-conforming worlds: the world of representations and the world of things. You can see what that implies, philosophically, if God is deceitful... it would imply an entirely new way of posing of the problem of evil. But if I had the power to establish real distinctions between representations without there being a corresponding true separation between things, the world would be double: there would be the world of my representations and the world of things, so God would be always misleading me since he would inspire true ideas in me and these true ideas would correspond to nothing in things.
To reply to Comptesse, I'm just saying that it's true that it's a story of ontological separation, but not so quickly, it will become a matter of ontological separation when Descartes is able to conclude: since I can represent thinking substance as really distinct from extended substance, then thinking substance and extended substance are two substances ontologically, and from that point on it is not the body which thinks in me. But before having gone through [the fifth meditation?], he absolutely cannot say this, he can only say: I conceive thinking substance as really distinct from extended substance, they are really distinct, since, once again, to be really distinct is the same thing as to be conceived as really distinct, two things whose representations are caused without one implying anything of the other are really distinct, he cannot yet affirm that it is not extension which thinks in me, that it is not the body which thinks in me.
The one thing that seems interesting to me is this idea of implicit relations, but Descartes does not call it that, and from this the promotion of an order of time in the writing of philosophy... You are going to tell me that you understand everything.
What does Kant do here? Kant wants to go further. It's inevitable, he wants to go further in relation to a previous philosopher, only this further has no pre-existence, he must create it. One of Kant's most beautiful texts is: "What does it mean: to orient oneself in thinking?" In this very beautiful text he develops a whole geographical conception of thought; he even has a new orientation, we must go further, Descartes did not go far enough: since he determined certain substances as subject, we must go further and break the link between the subject and substance. The subject is not substance. OK. What does that mean? He takes it up again and I will try to mark the stages: he says: "I think", fine. Which is to say that it is an active determination, and it's in this sense that Kant will name the "I think" as the form of spontaneity. It seems strange when he says that "I think" is the form of spontaneity, but everything is clear if you stick closely to the terminology; it means precisely: "I think" is a determination - he takes that from Descartes - and the "I think" accompanies each production of concepts. I cannot think a concept without thereby including the "I think". In other words, the "I" of the "I think" is the subject of all concepts, or, as he will say, it's the unity of the synthesis. Thus on this point, he changes the vocabulary, but he remains in agreement with Descartes. Why does he change vocabulary? It was to be expected, if he changes vocabulary while remaining in agreement with Descartes, it's because he will need this vocabulary for the moment when he will not agree, that's the first point.
Second point: in order to think one must be, in other words, there is a relation of implication between the determination "I think" and the position of an indeterminate existence "I am". Kant says it all the time: the "I think" implies - often the words vary - a feeling of existence (here we can clearly see the lineage, between Descartes and Kant there was Rousseau). Sometimes he says a consciousness of an indeterminate existence; the "I think" implies a pure consciousness of an indeterminate existence. Agreement with Descartes up to this point. From this point on Descartes has no more problems, and it's when a philosopher has no more problems that the next philosopher is about to arrive. Descartes has no more problems because he has a determination, and he has posited an indeterminate existence hence something to be determined, and he will say that the determination determines the indeterminate. The determination: I think, the indeterminate: I am, the determination determines the indeterminate: I am a thing that thinks.
Here Kant says no; it's the birth of German philosophy. I'm thinking of Leibniz. There are objections which are like reproaches. Beneath objections there are always theoretical reproaches. Leibniz already said of Descartes: he is too quick. It's like a judgement of taste. Kant takes on something of this, it's too quickly said. Kant: "I think" is a determination, agreed, determination implies the positing of an indeterminate existence "I am", agreed, but this doesn't tell me under what form this indeterminate existence is determinable, and this Descartes doesn't care about because he hasn't seen the problem. I think, I am, agreed. But what am I? Descartes replied: "I am a thing that thinks" since he applied the determination to the indeterminate. Now what I'm saying is becoming very clear: Descartes carried out an operation whereby he directly applied the determination to the existence to be determined. He directly applied the "I think" to the "I am" in order to get "I am a thing that thinks."
Kant says OK, I think, I am. But what am I, what is it that I am? A thing that thinks? But by what right can he say that? Descartes would have become angry... Kant says to him: but you're stuck, you have posited an indeterminate existence and you claim to determine it with the determination "I think". You have no right to do that. You have a determination, you have posited an indeterminate existence, you can turn it around as much as you like, you will not make any headway. You are stuck there. Why? Because to draw from this the conclusion "I am a thing that thinks", it assumes - and you have no right to assume it - it assumes that the indeterminate existence is determinable as a substance or a thing. Res cogitans, in Latin, the thinking thing.
Kant says, in accordance with all that has come before, which is to say what I tried to say the last time - the extraordinary change in the notion of phenomenon, the phenomenon no longer designating the appearance but the apparition, what appears in space and time - Kant can now say to us that the form under which an existence is determined within the conditions of our knowledge (what happens with angels, we have no idea), well, the form under which an existence is determinable under the conditions of our knowledge is the form of time. Thus the "I think" is the form of spontaneity or the most universal form of determination, but time is the most universal form of the determinable. Descartes' fatal conclusion was to confuse the indeterminate and the determinable, but the determination can only bear on the indeterminate as the mediation of the form of the determinable. In other words, I think, I am, the determination must determine the indeterminate existence "I am", but the indeterminate existence "I am" is only itself determinable under the form of time. It is only under the form of time, as the form of the determinable, that the form of thought will determine the indeterminate existence "I am".
This is how my existence can be determined only as time. But if time is the form of the determinable, under which my indeterminate existence can be determined by the "I think", what form do I receive from the determinable? The form that I receive from the determinable is that of a phenomenon in time, since time is the form of apparition of phenomena. I appear and I appear to myself in time. But what is it to appear and to appear to oneself, to appear in time?
They are the coordinates of a receptive, which is to say passive, being. Namely a being which has a cause, which does not act without also undergoing effects. Ok, we're at the end, and it's here that Kant will name the paradox of inner sense, the paradox of intimate sense: the "I think" is an active determination, it's the same form of the active determination, but the existence which it implies, the "I am", the indeterminate existence that the active determination of the "I think" implies, is only determinable in time, which is to say as the existence of a passive subject which undergoes all its modifications following the order and the course of time. In other words, I cannot - there is one sentence which is splendid, it's the Kantian version of what I was saying last time, namely that I is an other. This is what Kant says in the Critique of Pure Reason: "I cannot determine my existence as that of a spontaneous being, I only represent the spontaneity of my act of thinking". It's exactly "I is an other". I cannot determine my existence as that of an I, but I only represent the I to myself. The spontaneity of my act of thinking. The fact that I represent to myself the spontaneity of my act of thinking means that I represent the active determination of the "I think" to myself as the determination which determines my existence, but which can only determine it as the existence of a being which is not active, but a being on time [tre sur le temps]. This is the line of time which separates the "I think" from the "I am". It's the pure and empty line of time which traverses, which effects this sort of crack in the I, between an "I think" as determination and an "I am" as determinable in time.
Time has become the limit of thought and thought never ceases to have to deal with its own limit. Thought is limited from the inside. There is no longer an extended substance which limits thinking substance from the outside, and which resists thinking substance, but the form of thought is traversed through and through, as if cracked like a plate, it is cracked by the line of time. It makes time the interior limit of thought itself, which is to say the unthinkable in thought.
From Kant onward, philosophy will give itself the task of thinking what is not thinkable, instead of giving itself the task of thinking what is exterior to thought. The true limit traverses and works thought from within.
We rediscover what I tried to say the last time, namely: we find a sort of tension between two forms: the active form of spontaneity, or if you prefer, the "I think" as form of active determination, or form of the concept since "I think" is the formal unity of all concepts, so on the one hand the active form of determination, on the other the intuitive or receptive form of the determinable, time. The two are absolutely heterogeneous to each other, and yet there is a fundamental correlation: the one works in the other. Thought shelters in itself what resists thought.
In what sense is Heidegger Kantian? There are famous phrases such as: "we are not yet thinking"; when he talks about time in relation to thought, it's in this way that he is Kantian. The direct line from Kant to Heidegger is truly the problem of time and its relation to thought. The big problem that Kant discovers is the nature of the relation between the form of determination, or activity, or spontaneity, and on the other hand the form of receptivity, or form of the determinable, time. If I shift slightly, I would no longer say the form of determination and the form of determinable, but: two types of determination which are heterogeneous. You will ask me by what right I can make this shift; passing from the form of determination: I think, form of the determinable: time, the idea that there are two types of determination remains to be seen, but you can sense that it is the outcome of a series of shifts which must be justified, namely the two types of determination, in this case the conceptual determination, as all concepts refer to the "I think", concepts are the acts of the "I think", thus on the one hand a conceptual determination, and on the other hand a spatio-temporal determination. The two are absolutely heterogeneous, irreducible, the conceptual determination and the spatio-temporal determination are absolutely irreducible to each other, and yet they never cease to correspond to each other in such a way that for each concept I can assign the spatio-temporal determinations which correspond to it, just as, the spatio-temporal determinations being given, I can make a concept correspond to them. In what way, this is what remains to be seen.
If you grant me these shifts which we will define in a moment, it amounts to the same thing to say that Kant poses the problem of the relation between the form of determination "I think" and the form of the determinable = time, and in so doing completely upends [bouleverse] the element of philosophy, or to say, on a more precise level: no longer the "I think" but concepts, no longer time but the determinations of space and time, in this case it is a matter of the relation between the conceptual determination and the spatio-temporal determination.

[Break]

Our point of departure is this: how can we explain that conceptual determinations and spatio-temporal determinations correspond with each other when they are not at all of the same nature? What is a spatio-temporal determination? We will see that there are perhaps several kinds. Kant poses the question concerning the relation between the two types of determination on very different levels. One of these levels will be called that of the synthesis, another of these levels he calls that of the schema, and it would be disastrous for a reader of Kant to confuse the synthesis and the schema. I'm saying that the schema and the synthesis are operations which, in a certain way, put a conceptual determination and a spatio-temporal determination into relation, but then it's as if the synthesis will be shattered, pierced, will be overcome by a stupefying adventure which is the experience of the sublime. The experience of the sublime will knock over all the syntheses. But we do not live only on this. We live only on the syntheses and then the experience of the sublime, which is to say the infinity of the starry vault, or else the furious sea... The other case, the schema, is another case where spatio-temporal determinations and conceptual determinations come into correspondence, and there again there are conditions where our schemas shatter, and this will be the astonishing experience of the symbol and of symbolism. But the whole analysis of the sublime, and the whole analysis of the symbol and symbolism, the English had analyzed the sublime before him, but the whole novelty of Kant's analysis is obvious: it will be the Critique of Judgement, in his last book, as if to the extent that he aged, he became aware of the catastrophe. Of the double catastrophe of the crushing of the sublime, the sublime crushes me, and the irruption of the symbol, where our whole ground, the whole ground of our knowledge which we had constructed with syntheses and schemas, starts to shake.
What is the synthesis? It's the synthesis of perception. But don't think that that goes without saying. I'm saying that it's from this level of the analysis of the synthesis of perception that Kant can be considered as the founder of phenomenology. That is, that discipline of philosophy which has as its object the study, not of appearances, but apparitions and the fact of appearing. What is the synthesis of perception? All phenomena are in space and time. There is strictly speaking an indefinite diversity in space and time. Moreover, space and time are themselves diverse: they are not only the forms in which diversity is given, but they also give us a properly spatial and temporal diversity: the diversity of heres and the diversity of nows; any moment in time is a possible now, any point in space is a possible here. Thus not only is there an indefinite diversity in space and time, but also an indefinite diversity of space and time itself. Thus for perception, certainly the diverse must be given to me, but if I had nothing but this given diverse, this receptivity of the diverse, it would never form a perception. When I say "I perceive", I perceive a hat, I perceive a book, for example, this means that I constitute a certain space and a certain time in space and time. Space and time are indefinitely divisible: any portion of space is a space, any portion of time is a time. So it is not space and time themselves which account for the operation by which I determine a space and a time. I perceive a piece of sugar: I perceive a complex of space and time. You will tell me: that works for space, I can see that, there is the form, the grain; but why time? Because it forms part of my perception to wait for the sugar to melt. When I perceive a thing, I perceive a certain temporality of the thing and a certain spatiality of the thing. So there we have, according to Kant, a properly logical order, not at all chronological, he doesn't say that we must start with one.
There are three operations which constitute the synthesis, the synthesis operating on diversity in space and in time, and diversity in space and time at the same time. The synthesis consists in limiting a diversity in space and in time, and a diversity of space and time themselves, in order to say: it begins, it ends, etc.... The first aspect of the synthesis is what Kant calls the successive synthesis of the apprehension of parts, that is: every thing is a multiplicity and has a multiplicity of parts; I perceive parts, my eye runs over the thing. You will tell me that there are things small enough for me to perceive them at once. Yes and no, perhaps not, maybe so; moreover, however small something is, my perception can begin from the right or begin from the left, from the top or the bottom; it doesn't take very much time, it's a very contracted temporality. I carry out a synthesis of successive apprehension of parts.
But by the same stroke things already become complicated, we must distinguish two cases, we have not finished. In any case the apprehension of parts is successive. There are cases where the succession is objective, this already complicates things. I perceive a house, for example: ... the foreground, the background, the perspective, the foreground becoming background etc. ... there is a kind of subjective apprehension. But I begin from the right, or I begin from the left, and I keep going; in both cases my apprehension is successive, but the succession has only a subjective value. I can begin with the top or the bottom, with the right or the left; this will be reversible or retrograde, whether from right to left or from left to right, I can say that it's the wall in front of me. The succession is in my apprehension, it is not in the thing, it is not in the phenomenon. By contrast, you are sitting on ?????, there again you have a succession, a successive apprehension of parts, but the succession is objective. When the succession is objective, you will say: I perceive an event. When the succession is grasped as solely [subjective?], you perceive a thing. We could say that an event is a phenomenon whose successive apprehension of parts is such that the succession therein is objective. By contrast a thing is such that the succession therein is only subjective.
Thus the first aspect of the synthesis which consists in determining the parts of a space and a time is the synthesis of apprehension. Through this I determine the parts of a space. Let's suppose that you have carried out your successive apprehension of parts, suppose that you are in a curious situation, suppose that is that when you have arrived at the following part you have forgotten the previous one, you would not be able to perceive. There must in fact be an operation of contraction such that when you come to the following part, the preceding one is conserved, otherwise if you lose on one side what you gain on the other, you will never manage to determine a space and a time. This second aspect of the synthesis is the synthesis of reproduction. You must reproduce the preceding part when you come to the following part, so not only must you produce successive parts, but you have to reproduce the preceding parts with the following ones. The two aspects of the synthesis refer to the synthesis as the act of what? Not receptivity, receptivity is solely space and time and what appears in space and time is intuition. The concept is something else. The synthesis refers to the imagination, it is the act of the imagination. This act of the imagination is bizarre; see what he means: it's that through the two aspects, the apprehension of parts and the reproduction of parts, I effectively determine a space and a time. But according to Kant, to imagine is not to fabricate images, it is not to think of Pierre who is not there. To imagine is to determine a space and a time in space and time. There is certainly an empirical imagination. Empirical imagination is when Pierre is not there, I think of Pierre, or else I imagine Pierre, I dream. But the imagination which Kant will call transcendental is the act by which the imagination determines a space and a time, and it determines a space and a time through the synthesis of apprehension and the synthesis of reproduction. But something else again is needed. I am no longer in the situation of a diversity in space and in time, or a diversity of space and time itself, I am in the situation of a space and a time determined by the synthesis of the imagination. And yet I cannot yet say that I perceive. In order to perceive we still need for this space and this time, determined by the synthesis, or what comes to the same thing, that which contains this space and this time, must be related to a form, to a form of what? Not to a form of space or time since we have the form of space and time. What other form? You can see the progression. We started from the form of space and time in general, as the form of intuition, then the act of imagination determines a space, a given space and a given time, through the two aspects of the synthesis. In this case it's a form - not the form of space and time - but a spatio-temporal form, the form of a house or the form of a lion for example, but we need yet another form in order for there to be perception. It is necessary for this space and time, or what contains this determined space and time, to be related to the form of an object.
At this point it becomes difficult to understand. What does it mean that I have to relate it to the form of an object? We can imagine a number of sensations where the sensible givens, the diverse, sensible diversity, are not related to the object-form. It's my perception which is constituted in such a way that sensible diversity is related to the form of an object. In other words, I do not perceive an object, it is my perception which presupposes the object-form as one of its conditions, it's not something, it's an empty form. The object-form is precisely the index by which sensible qualities, such as I experience them, are supposed to refer to something. What something? Precisely a something = nothing. Kant will invent the splendid formula: a something = x. You will tell me that it's not a something = x when I say it's a table or it's a lion, it's not nothing, but the any-object-whatever [l'objet quelconque], the object = x, only receives a determination as lion, table or lighter by the diversity that I relate to it. When I relate to the object = x a diversity comprising: long hair in the wind, a roar in the air, a heavy step, a run of antelopes, well, I say it's a lion. And then I say: look a mouse! What I would like you to understand is that in any case there is an any-object-whatever, the object = x is a pure form of perception. I do not perceive objects, and it's my perception which presupposes the object-form. So the object is specified and qualified by myself according to a given diversity, a given space and time that I relate it to; when I relate a given spatio-temporal diversity, when I relate a given spatio-temporal form to the object = x, the object = x is no longer x, I can say that it's a lion or a house. But inversely I could never say that it's a lion or a house if the empty form of the object = x, the any-object-whatever was not available to me, for it is not the sensible diversity and it is nothing in the sensible diversity which accounts for the operation by which the sensible diversity goes beyond itself towards something that I call an object. Thus, apart from the form of space and of time (the form of intuition), apart from the determined spatio-temporal form (the synthesis of the imagination), I also need a third form: the form of the any-object-whatever such as this form is related to the spatio-temporal form in saying "it's this".
Such that the third aspect of the synthesis, after apprehension and reproduction, is what Kant calls recognition. To recognize. I effect a recognition when I say: "it's this". But "it's this" implies an operation whereby I go beyond what is given to me, I go beyond the forms of space and time, I go beyond purely spatio-temporal forms towards the form of an any-object-whatever that the spatio-temporal form will determine as such or such an object. But just as the two first acts of the synthesis, apprehension and reproduction, refer to the imagination, because it consists in determining a space and a time, so recognition is an act of the understanding. Why? You remember the concepts which are the representations of the understanding, they are the predicates of the any-object-whatever, of the object = x. Not every object is a lion, not every object is red, but every object has a cause, every object is one, every object is a multiplicity of parts, etc.... The predicates that you can attribute to any-object-whatever are the categories of the understanding, they are the concepts of the understanding. So recognition, the form of recognition, the form of the any-object-whatever is no longer in this case the synthesis of the imagination but the unity of the synthesis of ????? [understanding?].
It's the three aspects, apprehension, reproduction and recognition which constitute perception under the conditions [of an other of perception?].
A small note in parenthesis: above all never confuse, in the Kantian vocabulary, the object = x and the thing in itself. The thing in itself is opposed to the phenomenon since the phenomenon is the thing as it appears, whereas the object = x is not at all opposed to the phenomenon, it is the referring of all phenomena to the object-form. The thing in itself is situated outside of our possible knowledge, since we only know what appears, the form of the any-object-whatever is on the contrary a condition. The form of the object = x is a condition of our knowledge. We begin again from zero. I have all the elements [ensemble] of the synthesis: apprehension of successive parts, reproduction of preceding parts in the following ones, reference to the form of an any-object-whatever. So I have referred a spatio-temporal form to a conceptual form: the object-form. So Kant says to himself... let's begin again at the beginning. We have tried to analyze an edifice which emerges from the ground: the edifice which emerges from the ground is the synthesis. What is underneath it? I have said: in order to perceive an object I apprehend its successive parts, but how do I choose these parts? It's a funny sort of thing because it varies greatly according to the object. Apprehending successive parts implies, even at the level of perception, it already implies something like a lived evaluation of a unit of measure. But in following the nature of objects there is no constant unit of measure. In reflection, yes; from the point of view of the understanding, yes, I indeed have a constant unit of measure. I can fix a standard and even so, we will see that this is not even true, but we could fix a standard, put it into place for example and say that there are so many meters. But this is obviously not what Kant means by the successive apprehension of parts. It's like a sort of qualitative measure according to the object. What does that mean? When I see a tree, for example, I carry out my apprehension of successive parts, I begin with the top, then I go towards the bottom, or the other way round, and I say that this tree must be as big as ten men... I choose a kind of sensible unit to carry out my successive apprehension of parts. And then, behind the tree, there is a mountain, and I say how big this mountain is, it must be ten trees tall. And then I look at the sun and I wonder how many mountains it is; I never stop changing the unit of measure according to my perceptions. My unit of measure must be in harmony with the thing to be measured; there are some amazing variations.
Kant tells us in the Critique of Judgement, he is very careful not to before, he tells us that the most elementary act of the synthesis of perception presupposes a logical act. This synthesis of perception is in spite of everything a logical synthesis. I say in spite of everything because at the same time he gives "logic" an entirely new meaning. So once again I must choose a unit of measure, and this unit of measure is variable in each case in relation to the thing to be perceived, just as the thing to be perceived depends on the chosen unit. Beneath the successive apprehension of parts, which is a logical synthesis, even though it refers to the imagination, we need an aesthetic comprehension... this is no longer of the same order as measure; the aesthetic comprehension of a unit of measure such as it is supposed by measuring... Kant is in the process of discovering a sort of basis for the synthesis of apprehension, how an aesthetic comprehension of the unit of measure can be carried out because an aesthetic comprehension of the unit of measure is presupposed by the synthesis of the imagination in perception, namely the apprehension of an [evaluation of a rhythm?]. The evaluation of a rhythm will allow me to say: yes, I'll take that as a unit of measure in a given case; and the rhythms are always heterogeneous, we plunge into them in a sort of exploration. Beneath measures and their units, there are rhythms which give me, in each case, the aesthetic comprehension of the unit of measure. Beneath the measure there is the rhythm. But this is the catastrophe. Again we can no longer stop. We had the synthesis, we remained on the ground and the synthesis was established on the ground; we wanted to dig a bit and we discovered the phenomenon of aesthetic comprehension, and we can no longer stop. The rhythm is something which comes out of chaos, and the rhythm is something which can indeed perhaps return to chaos? What could happen? Let's approach this like a story. I look at something and I tell myself that I'm dizzy, or else my imagination wavers. What happens? In the first place I cannot choose a unit of measure. I don't have a unit of measure; it goes beyond my possible unit of measure. I look for an appropriate unit of measure and I don't have one. Each time I find one it is destroyed. So I am pushed as if by a wind at my back to choose bigger and bigger units of measure, and none is adequate. By the same stroke I cannot carry out my synthesis of apprehension. What I see is incommensurable to any unit of measure. Second catastrophe. In my panic I can perhaps distinguish parts, completely heterogeneous parts, but when I come to the next one everything happens as if I was struck by a dizzy spell: I forget the preceding one; I am pushed into going ever further and losing more and more. I can no longer carry out either my synthesis of apprehension or my synthesis of reproduction. Why? Because what I grasped, what struck my senses, was something which goes beyond any possibility of aesthetic apprehension!
We have seen that aesthetic comprehension was - even though Kant does not say it, but it is what he is thinking of - was the grasping of a rhythm as basis of measure and the unit of measure. You can see the whole of the synthesis of perception: I can no longer apprehend the successive parts, I cannot reproduce the preceding parts as the following ones arrive, and finally I can no longer say what it is, I can no longer qualify the any-object-whatever. My whole structure of perception is in the process of exploding. Why? My whole structure of perception is in the process of exploding because we have seen that it was founded - not in the sense of a ground [fondement], but in the sense of a foundation [fondation] - we have seen that this whole perceptive synthesis found its foundation in aesthetic comprehension, which is to say the evaluation of a rhythm.
Here it's as if this aesthetic comprehension, as evaluation of a rhythm which would serve as a foundation of measure, thus the synthesis of perception, is compromised, drowned in a chaos. The sublime.
Two things are said to be sublime. Kant's response: two things are said to be sublime: the "mathematical" sublime (said to be mathematical because it is extensive), and what is called the dynamical sublime (an intensive sublime). Examples: the infinite spectacle of the calm sea is the mathematical sublime; the starry celestial vault when the sky is clear is the mathematical sublime; it inspires a sentiment close to respect within me, it's a dynamical [?] sublime. In this case the infinity of an expanse gives way to the infinity of material forces, the intensive infinity of forces which fill space and time. The dynamical sublime is the tumultuous sea, it's the avalanche. In this case it's terror. Think to what extent Kant is at the centre of a certain conception of German Romanticism. I'll pass over the reasons why the dynamical sublime is more profound than the mathematical sublime. My second question on the sublime is : what effect does it have on me? We can move forward. I can no longer apprehend parts, I can no longer reproduce parts, I can no longer recognize something, and in effect the sublime, as Kant says, is the formless and the deformed. It is the infinite as encompassing all of space, or the infinite as overturning all of space; if my synthesis of perception is suppressed, this is because my aesthetic comprehension is itself compromised, which is to say: instead of a rhythm, I find myself in chaos.
Everything happens as if the imagination (the synthesis of perception) was pushed to its own limit. Great, we are in the process of rediscovering on the level of the faculty of the imagination something which we found on the level of the faculty of thought: it is not only thought which has a consubstantial relation, a fundamental relation, with an interior limit, the imagination is itself traversed by a limit specific to it, and the sublime confronts the imagination with its own limit. The beautiful, according to Kant, is not this at all, the beautiful is a reflection of the form of the object in the imagination. The sublime is when the imagination is in the presence of its own limit, it is alarmed. There was an enormous ambiguity between rhythm and chaos; I refer you to Paul Klee's famous text, how rhythm emerges from chaos, the way in which the grey point jumps over itself and organizes a rhythm in chaos. The grey point having the double function of being both chaos and at the same time a rhythm in so far as it dynamically jumps over itself; it will organize chaos and allow rhythm. Cézanne tells us that we never look at a landscape, it looks at something, and it is absolute chaos, "iridescent chaos". Cézanne says that it's like a landslide, a cave-in.
At this point I am one with the painting - this is Cézanne speaking - we are an iridescent chaos, etc. ... geological strata... translated into Kantian terms, it's really: I go from the synthesis of perception to [aesthetic?] comprehension...
Fortunately we are not caught up in the sublime all the time, this would be terrible, fortunately we hang on to our perception. At the moment that Kant says that in the sublime the imagination is taken to its own limit, and by the same stroke panicked, like a panicked compass, it is in the process of imagining what cannot be imagined; well at that moment, Kant says, in the respect of the mathematical sublime, or in the terror of the dynamic sublime, we suffer [éprouvons].
At the same time that my imagination is crushed by its own limit, it is a limit which is like its founding kernel, it is the bottomless [sans fond]. What is this bottomlessness of the imagination? It's something which makes me discover in myself something like a faculty which is stronger than the imagination, and this is the faculty of ideas.

Question:Can we say that music is the art of the sublime?

Gilles: That wouldn't be difficult. If I think, out of convenience for you, in terms of the history of philosophy, we can distinguish the arts of the beautiful and the arts of the sublime. However, about the arts of the beautiful and the arts of the sublime, you will find a long history with Schopenhauer and Nietzsche. But how do they make the distinction? Broadly, if you like, all art rests on an Idea; but in the arts of the beautiful it's as if the Idea is mediated, which is to say it is represented. There is a representation of the Idea. In the sublime the will appears for itself. Nietzsche, in so far as he is concerned with the origin of tragedy, will remain with this idea of a preeminence of music over all the arts because music makes the Idea appear as such, in opposition to the other arts which are condemned to representation.
You should sense that an Idea is not from the imagination, but neither is it a concept of the understanding, it's something else still. We thus need a very particular status for the Idea since the whole game of the sublime is this: the imagination is vanquished and derailed before its own limit, but the joy which we experience is because an awareness arises in us of a superior faculty, which Kant will call the super-sensible faculty and which is the faculty of the Idea. With Kant we cease to think the problem of evil in terms of exteriority. Very broadly, in the classical tradition, there is a tendency rather to say that evil is matter, evil is the body, it's what opposes, it's what resists. It's with Kant that this very curious idea appears, which obviously comes from Protestantism, of reform, the idea that evil is something spiritual. It is truly within spirit and not matter as exterior. This is precisely what I was trying to say with the notion of limit in Kant: the limit is not something outside, it is something which works from within. Here evil is fundamentally bound to spirituality; it is not at all as it is in Plato, where if there is evil it is because souls fall, and obviously they incarnate themselves in a body. With the reform the devil is taken seriously, only taking the devil seriously can be a philosophical operation. Evil is not the body, evil is truly in thought qua thought.
Question: Can you give the definitions of causality in Kant?
Gilles: There are several. The first definition of causality is: causality is the faculty of making something begin in the order of phenomena. It's a simple definition which implies two causalities: a causality which Kant calls phenomenal, namely that phenomena follow on from each other, and a phenomenon begins something which will be called its effect, and, second causality, the so-called free causality - because phenomenal causality is a determined causality and free causality is the faculty of beginning something in the order of phenomena on the basis of something which is not itself caused.
Second definition of causality, those before were nominal definitions, second definition: it's the relation between phenomena when the succession in their apprehension corresponds to an objective rule. Example: the boat which goes down the flow of the river, there the succession corresponds to an objective rule in opposition to succession in the perception of reason, where there is no causality. I would not say that the right side determines the left side, whereas in the perception of the boat I would say that the preceding state determines the following state.