CPR: The Transcendental Deduction (The “A” Deduction)

Q. What is the special meaning of deduction? Deduction here is the establishment of the right of concepts to apply to objects.

Q. What is the aim of it? To establish the objective validity of the categories. The categories have been established as subjective conditions of thought, but now the question is how they can have objective validity.

I was confused by b117-118, which appears at first glance to be describing the transcendental deduction as being a justification of the application of the categories to non-empirical objects. This is not quite right, or is misleading.

First, remember the big question: how is synthetic a priori knowledge possible? This kind of knowledge is exactly what he’s talking about in this section, and though it is true that the categories apply to objects of experience, they also apply legitimately to a priori objects(?) such as numbers, triangles, etc.** The fact that we must justify the use of the categories in dealing with these objects _ as well as_ the objects of experience means that we cannot use proofs based solely on experience – on an “empirical deduction”.

However (b118), we can use the objects of experience in the deduction as the prime examples of the “occasioning causes” of knowledge or thought.

But remember that numbers and triangles, though conceptual, also depend on a priori intuition, ie., space and time.

Because it is pure intuitions – the forms of intuition – and not the pure concepts of the understanding, which are the conditions under which objects are given in intuition (b122), perhaps objects could appear to us in experience solely through the pure forms of intuition? This makes a justification of the categories necessary.

NOTE that this does mean that the categories do not apply to empirical objects, rather to the manifold in intuition.

Anyway, over and above the “big question”, Kant is showing in what way experience and knowledge itself is possible, not only synthetic a priori knowledge — although it is the latter that concerns him with regard to the possibility of metaphysics. Apart from that special concern, the doctrine established here seems far more radical, because it makes the categories part of the subjective faculty that conditions experience and in fact imposes the laws of nature on to the manifold of appearances given in intuition. Nature as such is generated by the subject — and yet it is objectively valid and true, but this last point is what the deductions are all about.

But we experience things with a regularity that we must put down to concepts, and the synthetic unity of the manifold. The manifold otherwise would be a random jumble. Take cause: Hume showed that we cannot justify our sense of necessity and universality from experience alone. Kant can bring in a priori concepts for this, but this of course means that a “deduction” is required. (b123)

We cannot treat the pure concepts of the understanding as products of intuition, even though they are occasioned by it.

(b125) Experience as such is only possible through the categories, because all objects of experience can be thought, and for some object to be thought requires the forms of thought and therefore the pure concepts of the understanding, which are also a priori because they are prior to objects, ie., objects as thought depend on them.

[the categories] refer to objects of experience necessarily and a priori

The Transition (b124) – come back to it.

The A Deduction (The subjective deduction) (a95)

The following, where concept A is an apriori concept, are together contradictory and impossible: 1. Concept A is (a) not contained in the concept of possible experience, and (b) does not consist of the elements of an experience 2. Concept A refers to an object

This is contradictory because 1 means that it doesn’t correspond to an intuitions, and intuitions exhaust the range of possible experience. If there is no corresponding intuition then there is no corresponding object that the concept can refer to.

If 1 is the case then concept A would be merely the “logical form of a concept, but not the concept itself through which things are thought.”

Therefore, as 1(b) must be true for concept A – because it is a priori – for it to be able to apply to objects 1(a) must be false, ie., a priori concepts which have objective validity must be part of the conditions of possible experience.

Q. BUT am I confusing a priori concepts with pure a priori concepts? Recall that the categories are also called the pure concepts of the understanding, and therefore are the pure a priori concepts. Thus I don’t think I’m making a mistake here.

So, how are pure concepts of the understanding possible?

From above, to answer this question we must answer the following one:

What are the conditions of possible experience? (these are the categories, no? Yes!)

Even illusory derivative concepts regarding stuff we cannot experience are part of the conditions of experience. NKS:

...once I am in possession of pure concepts of understanding, I can think objects which may be impossible.

NOTE: I don’t quite see that he’s shown that all a priori concepts must be part of the conditions of experience. The above argument (a95-96) doesn’t quite seem to establish this.
Q. Has it already been established?

The transcendental deduction will be complete when it is proved that only by means of the categories can an object be thought

Q. But experience does not entail thought. Is Kant depending on a claim that it does?

Knowledge requires receptivity + spontaneity. Spontaneity is “the ground of a threefold synthesis.” Synthesis of…

1. Apprehension in intuition
2. Reproduction in imagination
3. Recognition in the concept

Intuition -> imagination -> concept

(The three subjective sources of knowledge (a115))

He says that the following bits are a bit obscure and will make sense once we get beyond them.

1. Synthesis of Apprehension in Intuition (a99)

This is the synthesis of the manifold of impressions over time which is contained in every intuition. It produces a single representation.

It is also pure or a priori, because it applies to space and time as well as objects in space and time. Q. But how exactly does this follow?

The synthesis of apprehension in intuition is “the transcendental ground of the possibility of all knowledge in general.” (a102)

2. Synthesis of Reproduction in imagination (a100)

Reproduction is the reproduction of appearances, which contain or produce representations that appear together repeatedly, such that only one of those representations will produce the other in the mind. Example: the sound of seagulls immediately conjures up sensations of the seaside in the imagination. (Kant has better examples in a102)

What makes this reproduction of appearances possible is “the a priori foundation of a necessary synthetic unity of them.” What is this foundation?

Isn’t it just the regularity of things, independent of us?

No! Appearances are “the mere play of our representations”, not things-in-themselves.

(a101-102) Kant says he wants to be able to make the following argument sound, which he thinks he can do by showing that (1) is true:

1. Our a priori intuitions give us knowledge IFF they contain a combination of the manifold “that would render a thoroughgoing synthesis of reproduction possible”
2. Experience presupposes the reproduction of appearances

Therefore… – This synthesis of reproduction in imagination is a priori, founded on a priori principles (maybe only the latter — see later regarding Imagination) – There must be a pure transcendental synthesis in imagination that conditions the possibility of experience

Basically then, Kant needs to show that a priori intuitions give us knowledge only when applied to some empirical content.

Also, in eg., drawing a line in thought, I apprehend the beginning of the line and the end of it both as intuitions in the mind (synthesis of apprehension), and I also automatically hold the beginning of the line in the mind if I have the end of the line in mind, so that I can think the whole line (synthesis of reproduction).

The synthesis of apprehension in intuition is inseparably connected to the synthesis of reproduction in imagination. You can’t have one without the other.

Q. these seem very close to me, and hard to distinguish. Why are they distinguished and is this legitimate? (but remember that Kant said that their importance will only become apparent later on)

Now, the synthesis of apprehension in intuition is “the transcendental ground of the possibility of all knowledge in general.” (a102) So, because the synthesis of reproduction is allied to it, as part of the same faculty, it is also transcendental.

3. Synthesis of Recognition in the concept

Finally, we are conscious of the line as a line, and only because we have such a concept. But this consciousness of course depends on the foregoing two syntheses. This consciousness also unites in one representation (the concept, I guess) what has been synthesized in intuition and imagination. Although Kant just says “what has been gradually intuited, and afterwards reproduced”. Perhaps we cannot really talk of a prior synthesis (prior to the uniting consciousness), or rather, maybe the hierarchy goes the other way? I’ve been picturing it all as either (a) a process in time beginning with intuition and making its way into the mind, finally arriving at concepts, or (b) as a structure with concepts, the understanding, as the foundation. These are surely wrong, even for Kant’s scheme.

Q. How am I wrong? (if I am)

What is an object? (a104)

But wait. What do we mean when we talk of an object corresponding to our knowledge? Well, we cannot talk of specific known things outside our knowledge – this is a contradiction – so we must mean some general concept.

Object = X

Knowledge is necessarily referred to an object independent of it. This is what makes knowledge rational, consistent and yielding. And for an item of knowledge to be such, it must have a unity such that it can refer to an object. And this must be a priori. This unity constitutes the concept of an object.

This unity, which knowledge depends on, is the formal unity of consciousness in the synthesis of the manifold of representations.

But this unity requires the three syntheses. Kant says, specifically (a105), that it requires that these syntheses each in turn make the next necessary.

This unity, again, is the unity of apperception

The concept of this unity is the representation of the object X

Concepts are general and serve as “rules”, but they can only do so insofar as they make possible the representation of the intuitive manifold, also as being a certain kind of representation by necessity.

“Necessity is always founded on a transcendental condition.” (a106)

Why? Because Necessity is a priori, and is a condition of experience (eg., in causality), so it is transcendental. This means that there must be a transcendental ground to the aforementioned unity of consciousness, and to general concepts of objects, and thereby to all objects of experience. NOTE: this is slightly ambiguous, isn’t it?

Without this transcendental ground it would be impossible to think objects. But again (see above), Kant seems again to be assuming that all experience involves thinking/the understanding/concepts.

The object = X = that something the concept of which expresses the necessity of synthesis.

This condition = transcendental apperception

• precedes all experience and makes it possible (a107)
• has numerical unity, the transcendental unity of apperception
• this is the a priori ground of all concepts

Apperception = self-consciousness

Transcendental unity of apperception = original and necessary consciousness of one’s own identity = consciousness of one’s own act of synthesising appearances according to concepts. (a108)

The mind could never think its own identity in the manifoldness of its representations, and indeed think this identity a priori, if it did not have before its eyes the identity of its act, by which it subjects all synthesis of apprehension (which is empirical) to a transcendental unity, and thus renders possible their connection according to a priori rules.

Some Notes on The Importance of The Unity of Transcendental Apperception

I’ve used the words in that order intentionally, because it’s specifically the unity I’m going to briefly talk about here. What Kant is doing here is an important development from Hume. Hume argued that the self as a unified object of experience does not exist: we cannot get the concept of the self from experience, because looking inward all we find is, not a single consciousness, but a consciousness of this or that impression (or idea?). All we find is a bundle, rather than an object which can qualify as the unified “I” that we imagine must exist as the subject apprehending these impressions (no doubt owing to its pragmatic utility: another “habit” of mind).

Kant agrees that we cannot reach the concept of a unified subject straight from bare experience, but he is arguing in the deduction that this concept is nevertheless legitimate, and as much more than a mere habit of mind: the unity of apperception is in fact an a priori condition for the possibility of experience in the first place (and therefore transcendental), otherwise there would be no basis for the unity of synthesis of appearances. That a house can appear as a house requires that it can appear as such to aan equally united subject. [Skating over the surface there; much more to be said]

Back to the general object…

The Transcendental Object

Object X = the transcendental object

Q. Is it the thing-in-itself? No, it’s more like the placeholder for objects in general, or the general subjective template of objecthood, allowing for numerical identity, whose source is therefore not the thing in itself.

ESSAY/BLOG: What is the difference between the transcendental object and the thing in itself? In what way can the former not be known.

But wait, that sounds more like the pure concept of the transcendental object, which “is that which alone can provide for all our empirical concepts in general a reference to an object, or objective reality.” (a109)

But the transcendental object, and the pure concept thereof, may be the same thing. Kant’s words here can be interpreted as consistent with that.

Later in the “phenomena and noumena” section, he describes the transcendental object as…

... a correlate of the unity of apperception, for the unity in sensible intuition’s manifold by means of which the understanding unites that manifold in the concept of an object. [A250]

... the concept of an object as such [A251]

The reference to a transcendental object = the objective reality of our empirical knowledge

The connection of appearances to each other, and therefore the experience of objects via appearances – which is how we do know objects – is possible only through the application of a priori rules that apply to a synthetic unity. This application is grounded transcendentally.

that is, they must in experience be subject to conditions of the necessary unity of apperception, just as much as in mere intuition they must be subject to the formal conditions of space and time. (a110)

“The Analytic does for thought what the Aesthetic does for intuition: it uncovers the conceptual components of the structure of experience.” (Gardner p115)

The Possibility of the Categories (a 110)

The form of experience is the synthetic unity of perceptions, which is the synthetic unity of appearances according to concepts.

NOTE: Clearly then, for Kant, experience does necessarily involve thinking. METANOTE: two things: he’s interested in knowledge, remember — a priori knowledge; and Kant’s attitude is probably more that even if some experience is merely intuitive, for it to be experience is for it to be thinkable, or for its objects to be thinkable or cognizable — objects of possible thought.

And remember, these concepts are not empirical concepts, and our empirical concepts themselves must be grounded in transcendental concepts, otherwise appearances would never take the form of experience, that is synthesised and united. Also, our knowledge would not refer to objects if we only had empirical concepts to rely on, because that knowledge wold be merely “thoughtless intuition”, lacking a connection to objects according to “universal and necessary laws”.

The categories are the conditions of thought in a possible experience

Categories -> thought as Space and time -> intuition

Therefore the categories have a priori objective validity.

(a113) Example: the concept of cause is a synthesis according to concepts. Causation is beheld as necessary, and this necessity is something that cannot be supplied empirically (Hume). Experience, as such and no more, cannot teach us that we are able to make an a priori and universal inference to what must follow an appearance (seen as a cause or as part of a chain of causes).

Affinity of the manifold = the association of appearances such that we can talk about causality.

The question is, or was, how does the affinity of the manifold come under universal laws?

The answer now is…

First, knowledge is only possible under a single transcendental apperception. This numerical identity must “enter into” the synthesis of everything manifold in appearances (as shown, I guess), therefore appearances are subject to apriori conditions, to which their synthesis in apprehension must conform. And that to which something must conform is a law, therefore we can say that aappearances are connected according to necessary laws, “and thus stand in a transcendental affinity of which the empirical affinity is a mere consequence. (a113-114)

Of the Relation of the Understanding to Objects in General, and The Possibility of Knowing Them A Priori (a115)

“That which in the preceding section we have discussed individually and separately, we shall now try to treat in a unified and connected way.”

The three subjective sources of knowledge, on which rest the possibility of experience in general, can now be stated slightly differently (now using apperception):

Sense, Imagination, Apperception

The transcendental principle of the unity of the manifold

There’s a principle that holds a priori: the transcendental principle of the unity of the manifold. The principle is that we are conscious of the identity of ourselves in relation to the representations of knowledge, and we are conscious of this fact as a necessary condition of the possibility of representations. We are conscious that our representations, the ones that give us knowledge, must belong to one consciousness.

Pure apperception is what supplies us with this principle.

And as this unity of the manifold is synthetic, we can say that pure apperception supplies us with the principle of the synthetic unity of the manifold in all possible intuition. [A117]

Kant has a footnote here, and it goes something like this…

1. All representations must be capable of being known in an empirical consciousness. That is, all representation “have a necessary reference to a possible empirical consciousness”.
2. All empirical consciousness has a necessary reference to a transcendental consciousness which precedes experience — this is the consciousness of myself as original apperception.

Therefore, it is absolutely necessary that in knowledge, all consciousness belongs to one consciousness — the consciousness of myself (apperception).

This consciousness is a synthetic unity of the manifold. It is known a priori and is the foundation for synthetic a priori propositions concerning pure thought — [once again] “in the very same way as space and time supply the foundation for synthetic a priori propositions concerning the form of mere intiution.

The synthetic proposition: All the variety of empirical consciousness must be combined in one single self-consciousness

This proposition is the “absolutely first and synthetic principle of our thought in general.”

Kant seems to be saying just what was said above about the principle, namely that it is pure apperception that is the source of what is the basic principle of thought, that experience is only possible with the foundation of a single united consciousness. We can know this, or think this, because pure apperception is itself a foundation for propositions about pure thought.

Kant is doing some second-order inference here. Not only is pure apperception a condition of the possibility of experience, but it also allows us to know that it is.

Is there some infinite regress here? Well, maybe there is, and that’s no problem, because we can go on thinking about ourselves, and about our thinking about ourselves, and about our thinking about our thinking about ourselves, and so on — at ever-higher orders until we get bored.

Back to the footnote…

Transcendental consciousness = the representation “I”

Kant says that although this “I” might be obscure at times, it is always present as a faculty that necessarily conditions “the possibility of the logical form of all knowledge”.

Q. Why, in these cases, can he not simply say, eg., a faculty that necessarily conditions “all knowledge”, rather than “the possibility of the logical form of all knowledge.”? What is the difference?

I get the feeling that this question, though it might seem superficial, gets to the heart of what transcendental philosophy is. To ask what conditions our knowledge might simply be to ask what mental faculties we happen to have, e.g., neurologically or psychologically. But Kant has bigger fish to fry. He wants to show what must logically be there so that knowledge, and objects, is even possible, without presuming if or when or how much or empirically how knowledge is gained.

But am I forgetting about the “logical form”? Not only does Kant say that the representation “I” conditions the possibility of all knowledge, but conditions the possibility of the logical form all knowledge.

Well, both work for me, but Kant’s emphasis here is on knowledge with respect to the transcendental object and the transcendental consciousness, i.e., to the basic central form of knowledge.

Summing up: the transcendental unity of apperception is the foundation of the synthetic unity of the manifold (in experience, in intuition).

Back to the main text…

Imagination [A118]

But the transcendental unity of apperception requires synthesis in imagination, not only synthesis in intuition. The pure synthesis of imagination is an a priori condition of the possibility of the unity of the manifold in one knowledge.

But when he last looked at imagination he was talking about the synthesis of reproduction. But that, while it rests on a priori principles, does not itself take place a priori, “because it rests on conditions of experience.” So the imaginative, specifically a priori synthesis, is productive rather than reproductive.

But seemingly out of nowhere we get this: [A118]

Thus the principle of the necessary unity of pure (productive) synthesis of imagination prior to apperception is the ground of the possibility of all knowledge, especially that of experience.

But I thought that the transcendental unity of apperception was the ground of the possibility of all knowledge!

Maybe imagination is key here because it is the central faculty, in that it is the bridge between intuition and (contentful) thought. The synthesis of reproduction is what allows for any other kind of synthesis in the first place. But considered a priori we must call it production, because we cannot make anything before or after part of the equation in this case.

So, we can locate the “ground of the possibility of all knowledge” here because for there to be anything at all depends on this transcendental function of the imagination. There must be something for synthesis of recognition in the concept to recognize in the first place, and this something is not asppearances or intuitions themselves, but rather the basic form of the synthesis in imagination, which is the logical ground for anything to be known at all.

Kant, in trying to show that certain concepts are valid for experience, that is, in trying to show that the categories can properly apply to the synthesis of appearaces, is identifying the source of the possibility of there being any synthesis at all. This is the pure synthesis of the productive imagination.

But: “...the possibility of all knowledge is founded on [the original unity of apperception]”

Taken literally, Kant is saying that both the pure synthesis of the productive imagination and the unity of pure apperception are the ground of the possibility of all knowledge. What he is really getting at is at the end of A118:

As the possibility of all knowledge is founded on this unity of apperception, it follows that the transcendental unity of the synthesis of imagination is the pure form of all possible knowledge; and through it, therefore, all objects of possible experience must be represented a priori. [A118]

The unity of apperception with reference to the synthesis of imagination is the understanding [A119]

The understanding almost takes a back seat here. It appears more passive than the imagination — the latter is productive, and is fairly viewed as the central spontaneous subjective faculty, even though Kant’s primary concern in this section is to establish the legitimacy of the pure concepts. To even get to the stage of justifying these concepts, the pure form of knowledge and its centre of production must be dealt with, and that’s what he’s been doing here.

...and the same unity [of apperception] with reference to the transcendental synthesis of imagination [is called] the pure understanding

Recall…

Synthesis of imagination: TThe faculty of reproducing appearances and combining them together so that they can become knowledge, i.e., so that the understanding can, well, understand them Transcendental synthesis of imagination: The pure a priori ground, the pure form of synthesis in imagination, the principle of production in imagination

An Important Conclusion

Hence in the understanding there exist pure a priori forms of knowledge which contain the necessary unity of the pure synthesis of imagination in regard to all possible appearances. These are the categories...

Looks like a fairly crucial juncture. Let’s see him complete this important conclusion:

The empirical faculty of knowledge of human beings therefore contains necessarily an understanding which refers to all objects of the senses, although only by means of intuition and of its synthesis through imagination; and all appearances, as data for a possible experience, are subject to this understanding.

**As this reference of appearances to possible experience is likewise necessary [...], it follows that the pure understanding, by means of the categories, is a formal and synthetic principle of all experience, and that appearances have thus a necessary reference to the understanding.

Boom!

Taking a different course to show the central importance of the imagination

At the end of A119 he sets out on a different course, to “clarify” how the understanding is connected to appearances necessarily through the categories. To do this he begins at the other end, from the empirical.

What is given to us first is appearance. [A120]

What we call perception is the result of the combination of appearances and consciousness. Perception is appearance regarded as an object of knowledge. But as appearances are appearances at all only when regarded so, there would in fact be no appearances at all – for knowing beings like us, I would probably add – without aa possible consciousness. There woud be no objective validity, no truth, no knowledge — without appearance as perception.

Now, manifold appearances must be combined in a way that they are not in the sensibility alone. This is the active faculty of the synthesis of the manifold, the faculty of imagination.

But the imagination must first apprehend perceptions, or impressions, or the manifld of intuition, before an image can be formed.

Q. Is “image” here representation?

[A121]

But the imagination can’t just apprehend impressions, it must also reproduce them to produce the coherence that characterizes perception. This reproductive faculty of the imagination is “on this account”, “merely empirical”.

Association

But there must be more than this empirical reproduction, because it happens regularly, consistently, according to rules. Certain representations are belong together and not with others. Note that this is still empirical, and Kant calls this reproduction according to rules the association of representations.

Affinity

But Kant wants to say that this coherence of perception is not contingent. Just because appearances appear coherent and consistent, doesn’t mean that they always will or must do — they might just happen to. For this not to be the case, appearances must be apprehended by the imagination only “under the condition of a possible synthetic unity of apprehension.”

In other words, appearances could not appear at all, or become objects of knowledge, if they could not be synthesized and united.

If this were not so, perceptions might exist in the mind yet not belong to a single consciousness. And this is impossible,, because one is only conscious of perceptions at all — one only perceives at all — because one is conscious and perceives. I cannot say that “I” am conscious of my perceptions unless only I am conscious of them, as mine.

There must be an objective ground to all this — necessary universal rules — and this is called the affinity of representations. [A122]

Whence this affinity? — The principle of the unity of apperception. According to this principle, appearances must enter the mind as belonging to one consciousness.

Imagination and Apperception

So…

We have thus seen that the objective unity of all (empirical) consciousness in one consciousness (that of the original apperception) is the necessary condition even of all possible perception; and the affinity of all appearances … is a necessary consequence of a synthesis in imagination which is founded a priori on rules. [A123]

Kant has two conditions which ground possible perception and the affinity of appearances:

1. The objective unity of apperception
2. The a priori synthesis in imagination

Imagination is therefore also a power of an a priori synthesis, which is the reason why we call it productive imagination …

That is, it is “also” a priori, as well as being empirical in reproducing appearances.

This a priori synthesis is the transcendental function of imagination

The affinity, association and reproduction (according to laws) of appearances — and therefore experience itself — is possible only by means of the transcendental function of the imagination.

It is the permanent and unchanging I (of pure apperception) which forms the correlate of all our representations … [A123]

Pure imagination + pure apperception => imagination functioning intellectually [A124]

Without this, the synthesis of imagination is still in the sensible realm, and in the realm of intuitions. But bring this manifold into relation with the unity of apperception and concepts become possible in connection with the appearances of the manifold.

Kant stresses that this happens only by means of this act of the imagination and its relating of the synthesized manifold to the unity of the “I” — the unity of apperception.

We have, therefore, a pure imagination as a fundamental faculty of the human soul, which underlies all knowledge a priori.

By means of pure imagination we bring the manifold of intuition on one side into connection with the condition of the necessary unity of pure apperception on the other.

This is what I was saying above about the imagination acting as a bridge between intuition and the understanding. You could say that it is this transcendental function of the imagination (the function of a priori synthesis in imagination) that pulls everything together, so this is why Kant says that it is so fundamental.

Without this, there would be “no objects of empirical knowledge, and therefore no experience” [A124]

To recap, experience — “actual experience” — is made up of:

• Apprehension of appearances
• Association — reproduction — of appearances
• Recognition of appearances

It’s in the last step, in recognition, that the “last and highest” concepts are (the categories). These “render possible the formal unity of experience and, with it, all objective validity (truth) of empirical knowledge” [A125]

To be slightly more precise, Kant says that the last and highest concepts are the categories insofar as they concern only the form of experience in general. Now why is this important? Well, because the categories are the a priori rules, and as rules they are general and prior to their application. There are concepts employed by the understanding, and in recognition, which could be identified but which are more on the empirical side. The concept of addition, for example, which is a concrete concept (perhaps this is what Kant means by an empirical concept) that partakes of the category of plurality, which is the true of the recognition in the experience of counting apples.

After bigging up the imagination, now it’s the turn for the categories [A125].

On them is founded all formal unity in the synthesis of imagination and, by means of this synthesis, all empirical use of the imagination (in recognition, reproduction, association and apprehension) down to the very appearances;

for it is only by means of those elements [the elements of imagination, I suppose, not the categories] that appearances can belong to knowledge and to our consciousness in general, and so to ourselves.

At first sight Kant looks a bit fickle, because he’s only just described the imagination as “underlying” knowledge [A124]. But we have to try and distinguish his different meanings.

My sense of it is that both are foundational, in that they are transcendental. That is, they are part of the same subjective faculty, or collection of faculties which work together. His “underlying” and “founded” are about what lies beneath knowledge or experience, i.e., the conditions of the possibility of those, namely the transcendental faculties of the human soul.

And because this synthesis for the categories, and with reference to the unity of apperception (I think that’s a legitimate locution), is what accounts for the ability we have to apply concepts, the regularity of appearances, that is, the obedience of our appearances to categorical laws, e.g., causality — this we owe to our own subjective faculties. In other words…

Hence the order and regularity in the appearances which we call nature are carried into them by ourselves… [A125]

This “unity” of nature is necessary, i.e., an priori certain unity of the connection of appearances.

And how should we have arrived a priori at such a synthetic unity if the subjective grounds for such unity were not contained a priori in our mind’s original sources of knowledge, and if these subjective conditions were not at the same time objectively valid, by being the grounds of the possibility of knowing an object in experience at all?

The Understanding

Accordingly, Kant now characterizes the understanding as the faculty of rules.

Sensibility gives us forms (of intuition), but the understanding gives us rules.

And considered transcendentally, with regard to the categories, the understanding particularly gives us laws, which are just objectively valid rules.

NOTE: This is where I’m switching to Pluhar

I noticed — and Pluhar notes this [A125, footnote 168] — that Kant uses nature to refer to the regular, orderly sum of appearances, and also to that regularity and orderliness. Either way, that which we call nature is a product of our creative synthesis.

Kant’s originality is quite neatly indicated by the passage [A125]:

Hence the order and regularity in the appearances that we call nature are brought into them by ourselves; nor indeed could such order and regularity be found in appearances, had not we, or the nature of our mind, put them into appearances originally

For this unity of nature is to be a necessary, i.e., an a priori certain, unity of connection [i.e., synthesis] of appearances.

But how indeed could we have the ability to institute a priori a synthetic unity, if our mind’s original cognitive sources did not a priori contain subjective bases of such unity, and if these subjective conditions were not at the same time valid objectively, viz., by being the bases for the possibility of cognizing an object in experience at all?

What is most startling in all this, I suppose, is the route to objectivity through subjectivity.

Now back to these laws [A126]. They are what make appearances appear as conforming to natural laws, and because experience is experience of a lawful nature, we can say that these laws make experience possible. This, of course, is repeating what has come before, but with a slightly more radical spin: owing to our faculties, experience is only possible as of a lawful complex of appearances. To put it another way, we can perceive nature because we have created it in the first place, because that is just how we must experience.

Hence understanding is not merely a power of making rules for oneself by comparing appearances …

... understanding is itself legislative for nature. I.e., without understanding there would not be any nature at all … [A126/127]

Appearances cannot occur outside us, and nature is the synthetic unity of the manifold of appearances according to rules, therefore without the understanding — which is the source of rules — there would be no nature.

Furthermore [A127], nature is possibile only in the unity of apperception. And this unity of apperception is the transcendental basis of nature existing as one experience.

I don’t understand how his introduction of the unity of apperception functions in the argument here, because the conclusion that follows seems merely to repeat what he’d established just above [A126/127]:

Hence all appearances, insofar as they are possible experiences, lie a priori in the understanding and obtain from it their formal possibility …

Perhaps the transcendental unity of apperception’s role demonstrates that not just nature but its unity (or “formal unity”) is owing to the understanding.

Kant then says that empirical laws, while they do not “in any way” derive their origin from pure understanding, are just particular instantiations (“determinations”) of the pure laws of the understanding. This is a bit like empirical objects: the perception of a tree does not originate in the understanding, but the object which is a tree is — maybe — an instantiation of the transcendental object. (Is that right?)

And with that, Kant reckons he’s done what he set out to do.

Kant’s Summary Presentation [A128-A130]

If objects are considered as things in themselves, then:

1. We cannot get a priori concepts from objects, because they would then be empirical, and we can’t get a priori concepts from experience alone.
2. We cannot get objectively valid concepts from ourselves, because they would apply only to the appearances within us and not to anything outside us, i.e., not to objects.

But if objects are appearances, then:

1. Appearances amount to objects within us
2. The conception that all these appearances (and hence all objects that we can experience) are within me, i.e., are “determinations of my identical self”, consists in their unity in a single apperception
3. The form of the cognition of objects also consists in the unity in apperception
4. So the way in which the manifold in sensible intuition belongs to one consciousness precedes the actual cognition of the object, i.e., it is a formal a priori cognition of objects insofar as they are thought
5. The synthesis of this intuition and the unity of the resulting presentations with respect to original apperception precedes all empirical cognition
6. Our cognition deals with nothing but appearances
7. So pure concepts of the understanding are a priori possible, and for experience necessary C. Certain a priori concepts necessarily precede our empirical cognition of objects, i.e., they make the experience of objects possible and make objects as appearances possible